# Airbridge — Full Content
> Airbridge is a mobile measurement partner (MMP) and attribution platform. This document contains the full text of all published blog articles across all languages.
# English Articles
---
## Airbridge Enables App Campaign Measurement With Amazon Ads Integration
URL: https://www.airbridge.io/en/blog/airbridge-amazon-ads-integration
Language: English
Published: 2026-09-16
Author: Jaehyuk Kim
Airbridge announces support for Amazon Ads as an integrated measurement channel. Marketers running app campaigns with Amazon Ads can now attribute installs and in-app events, send performance signals back to Amazon DSP for optimization, and compare results across their full media mix — all within Airbridge's unified reporting.
## Key Takeaways
- Amazon Ads has expanded beyond retail into a multi-screen advertising offering that includes Prime Video, Fire TV, and Twitch. Its reach is powered by shopping and streaming signals, making it a relevant channel for app marketers in commerce, subscription, and fintech verticals.
- Airbridge handles the setup with no custom development work, so marketers can bring Amazon Ads into their existing attribution stack in just a few steps and start measuring app campaign performance right away.
## What Is Amazon Ads?
Amazon Ads offers placements across:
- **Prime Video and Fire TV:** streaming and connected TV inventory
- **Twitch:** live content and gaming audiences
- **Amazon DSP:** programmatic advertising to help reach audiences across the Amazon store plus thousands of apps and websites
Amazon Ads leverages shopping and streaming signals to help advertisers reach relevant audiences. For mobile app marketers, this can provide access to incremental audiences that complement their existing media mix.
## Why Amazon Ads Is Becoming a Meaningful Channel for App Marketers
Performance marketers are always looking for channels that can reach customers who have not yet been reached by their existing mix. Amazon Ads stands out for two reasons that matter specifically for apps.
- **Shopping and streaming signals.** Amazon reaches audiences with shopping intent. They have browsed, compared, and decided. That context helps Amazon Ads reach relevant customers, making it a productive channel for apps in verticals like shopping, subscription services, and financial products.
- **A growing connected TV footprint.** Amazon Ads offers Streaming TV inventory including Prime Video and Fire TV, giving app marketers access to audiences in living-room environments. For teams looking to diversify beyond mobile-only channels, Airbridge can now measure how this inventory contributes to app growth.
With Airbridge adding Amazon Ads as a supported measurement channel, teams can now bring this channel into their existing attribution setup.
## Airbridge & Amazon Ads Integration: What Marketers Can Do Now
Airbridge connects Amazon Ads to cross-channel attribution, in-app event measurement, and unified performance reporting. Here is what that means in practice.
### 1\. Attribution Across Amazon's Multi-Screen Inventory
Amazon Ads operates across a range of placements: Amazon.com, Prime Video, Fire TV, Twitch, and the Amazon DSP network. When customers click or view an Amazon ad and later install an app, that touchpoint is often invisible to standard attribution setups.
The Airbridge integration connects touchpoints from Amazon Ads campaigns and connects them to downstream app activity, giving marketers a clearer view of how Amazon contributes to installs and conversions across its inventory.
### 2\. In-App Event Postbacks That Feed Campaign Optimization
Measurement alone does not improve campaign performance. What makes the Amazon Ads integration actionable is the ability to send in-app event signals back to Amazon Ads as postbacks. These include:
- **App installs:** the primary acquisition signal
- **Sign-ups and registrations:** early engagement indicators
- **Purchases and subscriptions:** high-value conversion events
- **Other custom in-app events:** mapped to Amazon Ads event types as needed
Amazon DSP uses these signals to refine its bidding and optimization logic, helping improve campaign relevance and performance. This closes the loop between attribution data collected in Airbridge and the optimization decisions made inside Amazon's campaign system.
### 3\. Unified Reporting Across Your Full Media Mix
One of the core challenges with adding a new channel is evaluating it fairly. Platform-reported metrics are measured by each platform's own rules, which makes cross-channel comparisons unreliable.
With Amazon Ads connected in Airbridge, campaign performance sits in the same reporting view as Meta, Google, TikTok, Apple Search Ads, and other channels. Teams can evaluate:
- Which channels are driving installs and activations
- How audiences acquired through Amazon Ads perform on retention and conversion
- Where Amazon Ads fits within the broader media mix, viewed in context rather than in isolation
This gives teams a consistent basis for budget decisions across all channels, rather than relying on platform-native reporting that each network measures differently.
## Bring Amazon Ads Into Your Attribution Setup
Amazon Ads helps app marketers reach incremental audiences across streaming, shopping, and connected TV environments.
With Airbridge supporting Amazon Ads as a measurement channel, connecting this inventory to your existing attribution setup is now straightforward. Campaign performance, event postbacks, and cross-channel reporting are all available with a streamlined setup process.
For step-by-step setup and activation guidance, refer to the [Airbridge Help Center](https://help.airbridge.io/en/guides/amazon-ads-channel-integration).
---
## What Breaks When a Deep Link Is Opened from Inside a Social App Browser
URL: https://www.airbridge.io/en/blog/deep-link-opened-in-social-app-browser-what-breaks
Language: English
Published: 2026-09-04
Author: Luke
A user taps your ad in Instagram. The app is installed. Instead of opening the app, the link loads your website inside Instagram's browser, or worse, opens the App Store for an app they already have. The campaign context is gone, and the click you paid for ends on a web page.
This happens on iOS and Android, on Instagram, TikTok, Facebook, and LinkedIn, and it is not a bug in your app. It is how in-app browsers work. This guide explains exactly what breaks, how the failure looks on each platform, and the three-layer setup that fixes it.
**Key Takeaways**
- **Social in-app browsers keep the tap on the web.** Universal Links (iOS) and App Links (Android) fire when the operating system handles the link tap. Instagram, TikTok, and Facebook handle navigation inside their own browsers and, in practice, do not hand it off.
- **iOS has a second trap for installed users.** The first tap on a tracking link from an app other than Safari can open the App Store even though the app is installed.
- **URI schemes still work inside webviews** because they need an explicit user tap, not OS interception. That makes them the fallback layer, not the primary link.
- **Deferred deep linking is the only layer that survives a fresh install.** A tracking link records the click server-side, so the destination and the parameters carried in the deep link can be restored on first open even when the click happened inside a social app.
- **Test on a physical device from inside the social app.** Simulators and Safari tests pass while Instagram taps fail.
## What Happens When a Deep Link Is Opened from Inside a Social App Browser?
1. The user is inside Instagram, TikTok, Facebook, or LinkedIn and taps your link.
2. The link opens in that app's built-in browser (WKWebView on iOS, WebView on Android), not in Safari or Chrome.
3. The embedded browser loads the URL as a web page. It does not pass the tap to the operating system.
4. Universal Links and App Links fire when the OS handles the tap. The social app's browser handles the navigation itself and, in practice, does not hand it off to the OS.
5. The user sees your web page. If the page redirects to the App Store or Play Store, an installed user is sent to a store listing for an app they already have.
6. Every campaign parameter in the URL stays on the web page. Nothing reaches the app.
Apple's [Universal Links documentation](https://developer.apple.com/documentation/xcode/supporting-associated-domains) describes links opening from Safari and from web views that hand off the tap, and Android's [App Links documentation](https://developer.android.com/training/app-links/verify-applinks) describes the same handoff for verified links. Social in-app browsers handle the navigation themselves and, in practice, keep it on the web. There is no server-side change that overrides that.
## Which Link Types Survive an In-App Browser?
| Link type | Tapped in Safari or Chrome, app installed | Tapped inside Instagram, TikTok, or Facebook, app installed | Tapped with the app not installed |
| --- | --- | --- | --- |
| Universal Link (iOS) | Opens the app at the target screen | Usually stays in the in-app browser and shows the web page | Falls back to the web URL. Destination lost after the App Store |
| App Link (Android) | Opens the app at the target screen | Usually shows the web page inside the webview. Some in-app browsers offer an "open in app" option | Falls back to the web URL. Destination lost after Play Store |
| URI scheme (`myapp://…`) | Opens the app if installed | Opens the app when triggered by an explicit user tap on the page | Error dialog or blank page. No fallback |
| Tracking link with deep linking (Airbridge) | Opens the app at the target screen | Opens the app. On iOS, the first tap from a non-Safari app may open the App Store unless the Stopover Airpage option is on | Redirects to the store. With deferred deep linking, the destination and campaign parameters are restored on first open |
## How the Failure Looks on iOS vs Android
| Platform and situation | What the user sees | Cause | Fix |
| --- | --- | --- | --- |
| iOS, app installed, first tap on a tracking link inside Instagram or Facebook | App Store page for an app they already have | iOS first-click behavior from apps other than Safari, documented in the [Airbridge deep link options guide](https://help.airbridge.io/en/guides/deeplink-option-setting) | Enable the Stopover Airpage option on the tracking link. The user gets a prompt to open the app instead of the store |
| iOS, app installed, second tap on the same tracking link in the same app | The app opens at the deep link destination | Expected behavior after the first tap | None. This is why "it works for me" bug reports are common: the tester already tapped once |
| iOS, app installed, tap on a plain Universal Link inside a webview | Web page inside Instagram's browser | The OS never intercepted the tap | Add a URI scheme fallback button on the landing page, or use a tracking link |
| Android, app installed, tap inside the Facebook or TikTok in-app browser | Web page inside the webview. Sometimes an "open in app" banner | The host app's WebView decides whether to hand off the URL, and usually does not | URI scheme fallback button on the landing page, or a tracking link |
| Either platform, app not installed | Store listing. First open lands on the home screen | Universal Links and App Links carry nothing through the store | Deferred deep linking through a tracking link |
## How to Fix It: The Three-Layer Setup
| Layer | What it handles |
| --- | --- |
| Universal Links | Standard iOS link taps (Safari, Mail, Messages, etc.) |
| App Links | Standard Android link taps (Chrome, Gmail, SMS, etc.) |
| URI scheme fallback, or a tracking link with the Stopover Airpage option | In-app browser contexts on both platforms |
### Option A: URI scheme fallback on your own landing page
1. Route the ad or bio link to a lightweight landing page, not directly to app content.
2. On that page, detect the in-app browser context with the User-Agent string.
3. If an in-app browser is detected, show an "Open in [App Name]" button prominently.
4. Have that button trigger a URI scheme link (for example `myapp://product/123`).
5. URI schemes work inside webviews because they require an explicit user tap rather than automatic OS interception.
The maintenance cost is real: Instagram and TikTok ship new in-app browser User-Agent strings, and the detection breaks silently. Re-test after every major social app update.
### Option B: Tracking link with deferred deep linking and the Stopover Airpage option
A tracking link moves the routing decision to the server. The click is recorded before any redirect, so the destination survives the in-app browser and the store.
1. In the Airbridge dashboard, go to **Tracking Link** > **Link Generation** and choose **App (Deep Link)** as the destination.
2. Set the deep link destination for installed users and the store fallback for new users.
3. In the deep link options, check **Add a Stopover Airpage**. This replaces the iOS first-tap redirect to the App Store from apps other than Safari with a prompt to open the app.
4. Use this link for bio, story, and re-engagement placements aimed at installed users. For install-focused ad campaigns, test whether the extra prompt costs store conversions before turning the Stopover Airpage on.
With the Stopover Airpage on, an iOS user who has the app and taps the link inside App X for the first time sees a prompt asking whether to open your app. Tapping **Open** goes to the deep link destination. Users who tap from Safari go straight to the destination with no prompt. Users without the app are sent to the App Store after a prompt, and deferred deep linking restores the destination on first open when they install and open the app within 1 hour of the click ([help center](https://help.airbridge.io/en/guides/retargeting-with-deep-links)).
## How to Test It
1. Install the app on a physical iPhone and a physical Android device. Simulators skip OS-level link verification.
2. Post the link in an Instagram story or send it in a DM, then tap it from inside Instagram. Do not paste it into Safari.
3. On iOS, tap once and record what happens, then tap again. The first and second taps behave differently without the Stopover Airpage.
4. Delete the app, tap the link again, install from the store, and open the app. Confirm the first screen matches the link destination.
5. Run the [deep linking test](https://help.airbridge.io/en/guides/deep-linking-test) in the Airbridge dashboard to confirm the deep link path and parameters.
## Frequently Asked Questions
### What breaks when a deep link is opened from inside a social app browser?
The link opens in the social app's embedded browser instead of Safari or Chrome, so the operating system never intercepts the tap. Universal Links and App Links do not fire, the user lands on a web page, and any store redirect loses the campaign context. On iOS, an installed user's first tap on a tracking link from a non-Safari app can also open the App Store.
### Why does the link work in Safari but not in Instagram?
Safari hands link taps to iOS, which checks your Apple App Site Association file and opens the app. Instagram's in-app browser handles the navigation itself and keeps the link on the web instead of handing it to iOS, so the same link stays on the web.
### Do URI schemes work inside in-app browsers?
Yes, when the user taps a button that triggers the URI scheme. They fail when the app is not installed, so use them as a fallback layer behind Universal Links, App Links, and deferred deep linking, not as the primary link.
### How does a tracking link fix this?
The tracking link records the click on the server before redirecting, so the destination and campaign parameters survive the in-app browser and the app store. On iOS, the Stopover Airpage option replaces the first-tap App Store redirect with a prompt to open the app. Airbridge Core Plan includes tracking links, deep linking, and deferred deep linking. Airbridge Core Plan starts at $40+/mo after a 30-day free trial, with 500K data points a month included, $0.0001 per additional data point, and no annual contract. Attribution events and tracking link clicks do not count as data points.
## Test From Inside the App, Not From Safari
Most social deep link bugs are found by users, not by QA, because QA tests in Safari. Put the link in a story, tap it on a real device, tap it twice, then delete the app and try again. If any of those four taps ends on a web page or a store listing for an installed app, you have found the leak that is eating your paid social budget.
[**_Get Started Free with Airbridge Core Plan and send every social link through a tracking link that survives the in-app browser._**](https://www.airbridge.io/en/plans/core)
---
## What Is Airbridge Signal Hold? Filter Early Cancellations Before They Reach Meta Ads
URL: https://www.airbridge.io/en/blog/what-is-airbridge-signal-hold
Language: English
Published: 2026-08-21
Author: Chaigun Jung
A trial starts. Five minutes later, it is canceled. A user makes a first payment, then asks for a refund.
If those conversions have already been sent to Meta Ads, Meta may still use them as positive optimization signals.
Airbridge Signal Hold adds a delay on purpose. It holds an eligible conversion postback for a set window, checks for a matching cancellation, and sends the conversion only if no cancellation arrives during that window.
That is signal engineering in product form.
[New to signal engineering? Read it first ->](https://www.airbridge.io/en/blog/what-is-signal-engineering)
## What problem does Signal Hold solve?
Meta sees the conversion event you send. It cannot know in advance that the user will cancel five minutes later.
Say your subscription app sends Start Trial to Meta as soon as a free trial begins. Meta can use that event as an optimization signal right away.
But if the trial is canceled almost immediately, the original success signal has already been sent.
Signal Hold gives that conversion a short verification window. If a matching cancellation arrives in time, the conversion is filtered out. If not, it is sent normally.
## How does Signal Hold work?
[ABR blog_Signal Hold.png]
Signal Hold pairs a conversion event with a cancellation event using the same Transaction ID.
- A conversion event happens.
- Airbridge holds the conversion postback instead of sending it immediately.
- If a cancellation with the same Transaction ID arrives during the hold window, the held conversion is not sent to Meta Ads.
- If no matching cancellation arrives before the window ends, the conversion is sent to Meta Ads.
The important part is the Transaction ID. That is how Airbridge knows which cancellation belongs to which conversion.
## When is Signal Hold a good fit?
Signal Hold works best when reversals happen fast - for example, trial cancellations soon after signup or refunds shortly after the first payment.
The point is not to send fewer conversions for the sake of it. The point is to keep clearly reversed conversions out of the signal Meta uses for optimization.
## How long should you hold a conversion?
Use your actual cancellation timing, not a guess.
If most early cancellations happen within the first hour, you may only need a short window. Signal Hold supports hold periods of up to 24 hours.
If the cancellations you care about usually happen days later, Signal Hold is probably not the right fit for that use case.
The goal is to catch the fast reversals without delaying good conversions longer than necessary.
## What do you need to use Signal Hold?
Your conversion and cancellation events need the same Transaction ID, and the conversion event needs enough recent volume to be selectable in Signal Hold.
Signal Hold currently works with Meta Ads. If you want the exact event-volume thresholds, campaign requirements, and setup steps, use the Airbridge setup guide.
[Airbridge Signal Hold setup guide](https://help.airbridge.io/en/guides/signal-hold)
## How do you know if it is working?
Signal Hold controls one input: which conversion postbacks Meta receives. It does not control your creative, bids, budget, audience, pricing, or product experience.
So do not judge it by Meta conversion volume alone.
For a subscription app, look further down the funnel:
- Trial-to-paid conversion rate
- Cost per paid subscriber
- Cancellation or refund rate
- Renewal rate
- ROAS
- LTV
Signal Hold may send fewer conversion postbacks to Meta. That's expected when cancellations are being filtered out.
The better question is whether the conversions that remain are more closely aligned with paid subscribers and long-term revenue.
## Wait. Verify. Then send.
For subscription apps, a conversion can change meaning quickly. A trial can be canceled. A first payment can be refunded.
Signal Hold gives you a short window to catch those reversals before the conversion reaches Meta Ads.
Simple idea, cleaner signal: wait, check for a matching cancellation, then send.
[Learn how to set up Signal Hold →](https://help.airbridge.io/en/guides/signal-hold)
---
## What Is Signal Engineering? How to Control What Ad Platforms Learn
URL: https://www.airbridge.io/en/blog/what-is-signal-engineering
Language: English
Published: 2026-08-21
Author: Chaigun Jung
Ad platforms optimize around the conversion signals they receive. The catch: a conversion can look good now and turn bad five minutes later.
A user starts a free trial, then cancels. Another subscribes, then asks for a refund. If those conversions have already been sent, the ad platform may still see both as success signals.
When performance slips, marketers usually check creative, targeting, or bids. One more thing is worth checking: the signal itself.
What are you actually telling the ad platform to optimize for?
## What is signal engineering?
Signal engineering is the practice of choosing which conversion signals reach ad platforms, how they are filtered, and when they are sent.
In plain English: don't just send everything. Decide what should count as success.
For a subscription app, the funnel might look like this:
Install -> Sign Up -> Start Trial -> Subscribe -> Renew
Each event gives the platform a different optimization signal. Install points it toward people likely to install. Start Trial points it toward people likely to start a trial. Subscribe points it toward people likely to pay.
Your job is to pick the signal that best matches the users you actually want more of.
## Should you always optimize for the deepest funnel event?
No.
Events closer to revenue are usually more valuable, but they also happen less often.
If Subscribe fires only a handful of times while Start Trial fires much more often, the deeper event may not give the ad platform enough data to learn from.
In that case, an earlier event such as Start Trial or Complete Onboarding can be a more practical optimization signal.
The sweet spot is simple: close enough to business value, frequent enough to learn from.
## What makes a good signal?
### 1\. Is it close to business value?
If your goal is paid subscription growth, a trial start tells you more than an install. A paid subscription tells you more than a trial start. A renewal may tell you even more about long-term value.
The right event depends on your funnel. The closer it is to the outcome you care about, the better—as long as you still have enough volume.
### 2\. Does it happen often enough?
A high-value event is not automatically a useful optimization signal if it barely happens.
For earlier-stage subscription apps, trial starts or strong activation events can be a better balance: more volume than paid subscriptions, but still meaningfully connected to revenue.
### 3\. Does it stay a good outcome after the conversion?
This is where subscription apps get interesting.
A trial that is canceled immediately is not the same as one that converts to paid. A first payment that is refunded is not the same as a subscription that renews for months.
If both land in the same conversion bucket, the ad platform cannot tell them apart.
## Why does this matter so much for subscription apps?
Subscription funnels have a long tail: trials, first payments, renewals, cancellations, refunds, and churn.
The first conversion is only a snapshot. Two users can look identical at trial start and end up with very different LTV.
That is why subscription marketers need to ask a better question than 'Did this user convert?'
Which users actually become valuable subscribers?
[Airbridge for subscription apps](https://www.airbridge.io/en/solutions/subscription)
## How can a subscription app design better signals?
Say Start Trial is your Meta optimization event. Volume is healthy, but you know some trials cancel almost immediately.
You can move deeper to Subscribe if volume supports it. You can use a stronger activation event. Or you can keep Start Trial and filter out conversions that are reversed shortly after they happen.
There is no universal 'best event.' The useful question is:
What behavior do you want Meta to count as success?
## Is more conversion data always better?
Not necessarily.
More events ≠ a better signal.
If your conversion stream is full of immediate cancellations, refunds, low-value actions, or duplicates, you may be adding volume without adding useful information.
Signal engineering is not about sending less data. It is about deciding which data should be used for optimization.
## What about duplicate events?
Modern marketing stacks often send the same event through more than one path: a platform SDK, web pixel, MMP, or server-side API.
Multiple paths are fine. Double-counting isn't.
If the same conversion is sent through more than one path, follow each platform's deduplication requirements so one conversion stays one conversion.
## How is Airbridge applying signal engineering?
Airbridge built Signal Hold as one concrete way to put signal engineering into practice.
Instead of sending every conversion immediately, Signal Hold can briefly hold a conversion, catch a matching cancellation during that window, and keep the canceled conversion out of the Meta postback stream.
That gives marketers more control over which conversions Meta can use for optimization.
Signal engineering comes back to one question: Are you sending the platform the conversions you actually want more of?
[Learn how Signal Hold works →](https://www.airbridge.io/en/product/signal-hold)
---
## 4 Months of Airbridge AI: How 820 Brands Actually Used It
URL: https://www.airbridge.io/en/blog/airbridge-ai-usage-record-data
Language: English
Published: 2026-08-09
Author: Jaehyuk Kim
It's already been four months since [Airbridge AI](https://www.airbridge.io/ko/product/airbridge-ai) launched. When we first announced it, we got a flood of questions about how, exactly, teams could put it to work.
Four months in, we went back and reviewed how customers actually used it. Every team developed its own way of asking. And instead of the one-question, one-answer pattern of a traditional chatbot, follow-up questions became the norm. More teams than we expected asked in languages other than Korean.
Plenty of articles explain what an AI marketing assistant can do. Very few publish the record of what the teams who adopted one actually asked, and how much.
This post lays out the real usage data from Airbridge AI's first four months, five ways your team can put it to work, and the MCP usage patterns that are expanding beyond the dashboard.
📌 **Key takeaways**
- In four months, 820 brands asked 16,610 questions. Airbridge AI became a whole-team tool, not one specialist's tool.
- The most common topics were event taxonomy, SDK, and reporting. Implementation and operations questions vastly outnumbered analysis questions.
- Outside the dashboard, 800+ people connected Airbridge to their own AI via MCP, and those connections pulled data 22,207 times in the last 30 days alone.
## Airbridge AI Fielded 16,610 Questions in Four Months
[Airbridge AI_after 4 month_01.png]
_16,610 total questions, 820 customer accounts, and 1,700 users over four months_
Between March 25 and July 23, 2026, Airbridge AI received **16,610 questions**. That came from roughly 820 brands and about 1,700 individual users.
The monthly trend tells the same story: by day 23, July had already passed all of June. That's a sign Airbridge AI has spread broadly into day-to-day work.
## The 5 Things Customers Asked Airbridge AI About Most
Sort the questions by type and it becomes clear which Airbridge AI capabilities customers reach for. **Event and taxonomy design** drew the most questions by far, followed by SDK installation and integration, then report lookups.
[Airbridge AI_after 4 month_02.png]
_Top 5 question types — event taxonomy design 14.1%, SDK/install/integration 13.3%, reports & dashboards 11.8%, ad channel & media integration 11.2%, deep linking 10%_
| Rank | Question type | Share |
| --- | --- | --- |
| 1 | Event & taxonomy design | 14.1% |
| 2 | SDK, install & integration | 13.3% |
| 3 | Reports, dashboards & data lookups | 11.8% |
| 4 | Ad channel & media integration | 11.2% |
| 5 | Deep linking | 10.0% |
SDK, install and integration was the second most-asked area after event and taxonomy design — a clear signal that most newly onboarding teams hit a wall at least once in this stretch.
[Airbridge AI_after 4 month_03.png]
Look at the distribution and all five of the top types cluster around **implementation and day-to-day operations**. Questions about standing up and running a measurement foundation overwhelmingly outnumbered questions about interpreting data. Add troubleshooting and data discrepancies, ad spend and ROAS, and attribution, and the top 8 types account for 76% of all questions.
In other words, **when the guides don't immediately answer the situation in front of you**, teams turned to Airbridge AI to solve it themselves instead of filing a request with a specialist.
## 5 Real Questions Teams Sent to Airbridge AI
Here's how real teams asked Airbridge AI and what came back. We've left out company names and identified each one by vertical only — see whether any of them look like your situation.
[Airbridge AI_after 4 month_04.png]
_Five real user questions — media mix reallocation, weekly performance and retention, filling an empty taxonomy, native-language SDK guides, complex attribution consultation_
### 1\. Analyze our media mix — and recommend next month's budget
A commerce app team operating outside Korea asked Airbridge AI this:
> "Review this month's paid channel spend and tell me where to concentrate next month."
Airbridge AI came back with **a table of Cost, Cost Share, Installs, CPI, Orders, MAU and ROAS by channel, plus a budget reallocation recommendation layered on top**. Because it applied the account's own currency and time zone, the team could read it immediately without converting anything.
[airbridge-ai-4-months-usage-5-1785995978505.webp]
_Airbridge AI answer example — a paid channel performance table with a budget reallocation recommendation_
### 2\. Roll weekly performance and retention into a single question
> "Compare last week's user acquisition and remarketing performance by country, including retention, against the prior week."
For this one, Airbridge AI **automatically identified custom metrics like CPA, CVR and D1/D7 retention**, then built four Actuals reports at once and returned a period-over-period comparison. Work that used to mean opening reports one by one and lining up metrics by hand collapsed into a single question.
### 3\. Fill an empty event taxonomy using app data
Taxonomy design demands knowledge of the user journey and [Standard Event](https://help.airbridge.io/ko/guides/airbridge-event-types) mapping, which is exactly where new teams with no template to work from tend to get stuck.
> "Design an event taxonomy and Standard Event mapping based on our core user journey."
Ask Airbridge AI that, and **even with an empty taxonomy template it infers your vertical from app data — app store listing, platform, currency** — and maps a Standard Event set for you.
### 4\. Generate SDK guides in each team's own language
One team asked, in Chinese, "How do I integrate the Android SDK?" Airbridge AI returned **a step-by-step guide in Chinese with code blocks inline**, covering everything from adding the repository to initialization.
[airbridge-ai-4-months-usage-6-1785995979620.webp]
_Airbridge AI answer example — a step-by-step Android SDK integration guide written in Chinese_
### 5\. Work through complex attribution questions against official docs
One user asked, in English, how to think about the chicken-and-egg problem between [ATT (Apple App Tracking Transparency)](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork) consent timing and attribution.
For a question that deep, Airbridge AI **searched the official documentation first, then separated the scenarios by whether attribution is deterministic**, and explained how the product's design resolves the problem with the reasoning behind it. This is a problem you can only untangle if you understand the concept and the product design together — so Airbridge AI became a channel for getting immediate help on a question you wouldn't even know who to ask.
## Power Users Trigger Whole Workflows with Slash Commands
Everything above is how most people use Airbridge AI. A small group of power users goes one step further and leans hard on [**AI Skills Slash Commands**](https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch).
Looking at Slash Command usage since launch, the taxonomy design command was used most, with the data analysis command right behind it.
[Airbridge AI_after 4 month_05.png]
_9.1% of users used Slash Commands to run Airbridge AI skills_
| Slash Command | What it does |
| --- | --- |
| /airbridge-event-taxonomy-designer | Auto-design an event taxonomy |
| /airbridge-data-analyst | Request a data analysis |
| /airbridge-monthly-mediamix | Monthly media mix analysis |
| /airbridge-weekly-review | Weekly performance review |
| /airbridge-industry-benchmark | Industry benchmarks |
| /airbridge-campaign-brief | Write a campaign brief |
| /airbridge-kpi-taxonomy | KPI taxonomy design |
| /airbridge-analyze-discrepancy | Data discrepancy analysis |
One command runs a repeatable task, and you can build your own skills to fit how your team works. On volume, though, Slash Commands account for roughly 1.5% of all questions and are used by about 9.1% of users — a subset, not the majority.
[Airbridge AI_after 4 month_06.png]
_800+ people connected Airbridge MCP, 22,207 data calls in the last 30 days, 84% of them reports and data lookups_
So most Airbridge AI users treat it as an implementation-and-operations assistant, while a small group of power users extends it into full analysis workflows.
## Airbridge MCP: 800+ People Connected Airbridge to Claude and Cursor
Everything so far happened inside the Airbridge dashboard. Now let's look at the data on how Airbridge shows up inside the AI tools customers already use.
Alongside the Airbridge pilot, we opened the [**Airbridge MCP (Model Context Protocol) server**](https://help.airbridge.io/ko/guides/mcp), which connects your data to the AI tools you're already in — Claude, Cursor and others. Since launch, **800+ people have connected Airbridge to their own AI.**
Over the last 30 days (June 23 – July 22), those connected AI tools pulled Airbridge data 22,207 times. Connect it once and it becomes infrastructure you use every day.
Reports and data lookups make up about 84% of those calls — MCP is establishing itself as the channel for **performance analysis done through AI**.
## What Four Months of Airbridge AI Data Told Us
Pulling four months of data together confirmed one thing: Airbridge AI isn't a chatbot returning canned answers. It's a tool that reads the data and settings as they are in that moment and builds an answer from them. And it's no longer confined to the dashboard — through MCP, it's a tool you can work with conversationally inside multiple AI tools.
Pick one question you're stuck on right now, put it to Airbridge AI in the dashboard exactly as you'd say it, and connect Airbridge MCP to the Claude or Cursor you're already using. See for yourself whether the answer comes back with tables and sources attached, and whether it reflects your account settings. That's how you'll get a feel for how Airbridge AI fits your team's situation.
---
## How Profitable Subscription Apps Set Up Web-to-App Campaigns on Google Ads
URL: https://www.airbridge.io/en/blog/how-profitable-subscription-apps-set-up-web2app-campaigns-on-google-ads
Language: English
Published: 2026-07-11
Author: Ashley Black
Web-to-app feels like all the rage these days, but many marketers conflate two very different strategies that require two distinct set-ups.
In this article, we’ll go step-by-step on how to set up a direct to app store web-to-app campaign on Google with **Ashley Black, founder of Candid Consulting and previously at Google App Ads.** In the process, we'll be using Airbridge Core Plan, built for subscription apps to accurately attribute your web-to-app installs, trials, and subscriptions.
To clarify the difference between the two commonly referred to 'web-to-app' strategies, here are their definitions:
1. **Web-to-App**: Driving users from a **web-based ad directly to the App Store**. No landing page in between. This is what we’ll be digging into in this guide.
2. **Web-to-App aka Web Funnels**: Driving users from a **web-based ad to a web page** where users complete funnels/purchases on web. This is popular amongst subscription apps as it can help avoid App Store fees. This guide will not cover this set up as it warrants its own playbook
## **Why subscription apps use Web-to-App vs. UAC**
Apple reshaped app marketing in 2021 with ATT and SKAN, dramatically reducing both performance and measurement visibility. Google felt this especially acutely. Unlike Meta, TikTok, and Snap, whose app campaign inventory is largely in-app, Google’s app network skews more heavily toward web traffic, particularly Search, where it cannot collect IDFAs to deterministically link clicks to conversions. Combined with Google’s delayed ATT prompt rollout on YouTube, its largest app channel, this left Google with fewer deterministic conversion signals and a heavier dependence on modeled conversions. Those modeled conversions may hit your targets within Google’s platform, but when marketers go to view performance on other measurement sources, the numbers rarely look as good.
Another significant limitation of Apple’s SKAdNetwork and AdAttributionKit is the delay in receiving post-install conversion data. Important events such as trial starts and subscription purchases can take days (or longer) to be reported back to ad networks due to “privacy-preserving” timers. These delays reduce the speed and accuracy with which Google can learn from user behavior and optimize campaigns. In contrast, a web-to-app strategy is not constrained by SKAN/AAK reporting delays, allowing conversion signals to be shared more quickly and enabling Google’s algorithm to react and optimize faster.
While Google has made a lot of positive improvements on their iOS app product (see ICM, ODM) some marketers still want the same level of accuracy they had pre-2021; this is where web-to-app comes in. Web-to-App is effectively a hidden redirect enabling Airbridge to collect Google identifiers and marketing-level data (campaign, keyword, ad) and attribute it to specific in-app conversions, like trial starts and subscriptions. This in-app conversion data is then shared back with Google to inform it of the clicks’ effectiveness in driving the intended results.
Though Airbridge makes it easy to enable web-to-app campaigns with Google, it’s important to evaluate whether it’s the right strategy for your app and ensure it’s properly set up from the start.
## **Is Web-to-App a good fit for my app?**
While you can run web-to-app on Google’s Display Network or YouTube, the highest quality users are from Search. However, Search is finite, so you first have to determine whether there is enough volume to be worthwhile. To do that, you can use Google’s internal Keyword Planner to research the search volume for your category.
For example, the search term “meditation app” has ~50K monthly searches in the US. This is a viable number to run a search campaign. Combined keywords under 5K will have a hard time driving enough traffic to sustain a campaign.
[01-keyword-planner.png]
## **How to set-up a Web-to-App Campaign on Google**
Setting up a web-to-app campaign requires some back and forth between Airbridge and Google Ads. To help with the flow of that set-up, \[AB\] will be used to indicate the step is done in Airbridge and \[GA\] will be used to indicate the step is done in Google Ads.
### Steps
1. \[AB\] Ensure you have Google Ads set up as an “Ad Channel Integration” in Airbridge.
2. \[AB\] Complete the channel integration and ensure you have added all “In-app Event Postbacks” that you want to send to Google.
3. \[AB\] Make sure the toggle is on to send these events to Google.
[02-inapp-event-postbacks.png]
4. \[GA\] Link Airbridge to your Google Ads account under “Connected Products”. This automatically imports Airbridge conversions into Google Ads (can take a few hours for any new conversions added).
[03-connected-products.png]
5. \[GA\] Navigate to “Goals” and create a “Custom Goal”.
[04-create-custom-goal.png]
6. \[GA\] Select the Airbridge conversions you want to use to optimize your Search campaign.
[05-select-conversions.png]
7. \[GA\] Create a Google Search campaign by selecting “Create campaign without guidance” and select “Search”
[06-create-search-campaign.png]
8. \[GA\] Under “Add Goal”, select the custom goal you created. Proceed without selecting “Website Visits” or “Phone Calls”.
9. \[GA\] Set up your bid strategy with or without a bid target.
10. \[GA\] Structure your campaign settings based on preferences.
11. \[GA/AB\] Within Campaign Settings > Campaign URL Options, copy and paste the Tracking Template from Airbridge under the “Web Campaign Integration” section.
_Note: While Airbridge’s documentation says to copy and paste this at the Account Level, those directions are primarily if you are using a landing page to redirect to the store. For this set up, keep it at the campaign level._
[08-tracking-template-google-paste.png][07-tracking-template-airbridge-copy.png]
12. \[GA\] Enter Final URL which is your App Store URL.
13. \[GA\] Complete Search campaign set-up.
14. \[GA\] Turn on Parallel Tracking and Auto-Tagging within Admin > Account Settings.
[09-parallel-tracking.png]
## **Best Practices for Search Web-to-App campaign**
In a lot of ways, setting up a Search web-to-app campaign is the same as you would if you were running a traditional search campaign to a website. However, there are some features available on this campaign type that won’t translate for a web-to-app campaign.
- **AI Max**: While some features of AI Max could work for web-to-app, the dynamic routing to specific webpages makes this feature a bad fit for this strategy.
- **Sitelinks**: Traditional sitelinks won’t work as they will direct users away from the App Store listing. However, you can run Custom Product Pages in place of traditional sitelinks.
- **Dynamic Images**: These pull from your _landing page_ which may not be the best source of images to be used in a search campaign given the only images available via this set-up are screenshots.
### For newly launched Search web-to-app campaigns, here are some best practices:
- **Start broad (literally)**: Apply keywords with broad and/or phrase match vs. exact. Allow the system to figure out what works and what doesn’t and hone in once you have enough data.
- **Less is more**: You don’t need thousands of keywords. Identify what keywords have the highest volume and go from there. With more data, you can add in additional keywords based on search term volume.
- **Apply a targe**t: Initially you can start with max conversions to gather data, but switch to a target bid within 1-2 weeks to help control performance.
- **Disable “networks”**: Google will auto-opt in the campaign to Search Network and Display Network. Opt out of those two to start so it doesn’t muddy results.
- **Check location settings:** Just like UAC, the default setting targets users who have “Presence or interest in” your targeted location. This may work for physical businesses, but for apps it’s best to keep it just to people who have a “Presence In” your targeted location.
## **Recap**
Running web-to-app campaigns on Google is one of the most effective ways to reach high-intent users while avoiding many of the measurement and attribution challenges associated with iOS acquisition. For subscription apps looking to test this strategy, Airbridge’s Core Plan makes setup simple, reliable, and fast—so you can launch with confidence and trust the data behind your results.
If starting out, be sure to use the promo code: **ASHLEY** to receive **20% off the Core Plan**.
---
## Google I/O 2026: 5 measurement shifts every app marketer needs to revisit
URL: https://www.airbridge.io/en/blog/google-io-2026-app-marketing-data-measurement-stack-changes
Language: English
Published: 2026-06-23
Author: Jaehyuk Kim
Google I/O 2026 and Google Marketing Live 2026 took place in Mountain View, California, on May 19-20. From search, payments, and discovery to ad automation and subscription models, the events touched nearly every domain where marketing meets AI. In the keynote, CEO Sundar Pichai formally announced the agentic Gemini era.
We picked five announcements that app marketers should watch. In each one, AI agents slot themselves between users and your app, changing what your measured signals actually mean. So if Google is building a new AI paradigm, what should app marketers prepare for?
> “We're firmly in our agentic Gemini era. I'm excited to see how it will unlock new ways to accelerate our mission and transform our products to be radically more helpful, for everyone everywhere.”— Sundar Pichai, CEO of Google and Alphabet
🎥 Watch: [Google I/O 2026 keynote in 35 minutes](https://www.youtube.com/watch?v=OMhKgQmeMhI) (YouTube)
**Key takeaways**
- Ask Play and the Gemini app (900 million monthly users) now handle app discovery directly, scattering acquisition paths outside the Play Store.
- AI Max and AI Brief hand campaign execution to AI. What marketers still own: signal quality and brand guidelines.
- Universal Cart and AP2 close transactions inside Google, so first-party app conversion data underreports commerce performance.
- Google Play's payment grace period extended to 60 days, shifting the LTV baseline for subscription apps. Bidding and budget models trained on last year's data need recalibration.
- 88% of AI Mode queries end without an external click. The measurement question shifts from clicks to incremental revenue.
## Ask Play and the Gemini app are recommending apps directly
Google says its Play Store AI Q&A already answers **95%** of user queries. On top of that, Ask Play adds a conversational discovery layer, and the Gemini app, now past **900 million** monthly users, has started recommending apps directly on Android and the web (Android Developers Blog · blog.google, Google I/O 2026, 2026).
> “We're entering the era of Search agents, where you can easily create, customize and manage multiple AI agents for your many tasks, right in Search.” — Google official announcement, Search at Google I/O 2026
[]
🎥 Watch: [Google — Ask Play](https://youtu.be/fwLiTPtPHjw) (YouTube)
👉 More detail: [Android Developers Blog: What's new in Google Play at I/O 2026](https://android-developers.googleblog.com/2026/05/io-2026-whats-new-in-google-play.html).
### 1\. Discovery now flows through four channels, and ASO is shifting to intent
Play Store search is no longer the single entry point. At the same event, Google said Engage SDK reaches **30 million+** users monthly and drove a **45%** YoY increase in app opens (Android Developers Blog, 2026).
The Gemini app, Ask Play, and Engage SDK slots all run on semantic recommendation rather than keyword matching. Add Play Store search itself, and you get four entry points. If you only watch the Play Store referrer, all four flow in as the same "organic" bucket, so you'll need to break them out as a quarterly task.
ASO is moving from keyword coverage to semantic coverage (ASO World, Google Play May 2026 Updates, 2026). Rewrite app descriptions and screenshot copy around use cases, and inventory which scenarios you serve through dynamic metadata slots: In-App Events, Promotional Content, and Custom Store Listings. Layer this on top of your [ASO fundamentals](https://www.airbridge.io/en/blog/aso-strategy).
### 2\. Separate agent traffic from human traffic
Gemini Spark is a personal AI agent that runs 24/7 on Google Cloud virtual machines. It integrates with Gmail, Google Docs, and Slides, and extends to third-party tools like Canva, OpenTable, and Instacart via MCP (Model Context Protocol, the standard for AI agents to call external tools) ([Google Cloud announcement](https://blog.google/intl/ko-kr/company-news/technology/innovations-from-google-io-26-on-google-cloud/), 2026).
The measurement issue: when an agent calls your app on behalf of a user, that session shows up as DAU (daily active users), but it makes no sense to retarget with push or ads. If you don't split automated and human traffic, retention cohorts inflate, and any LTV (lifetime value) model running on top of them operates on a broken baseline.
👉 Full architecture and workflow: [Google Cloud announcement — Innovations from Google I/O 2026](https://blog.google/intl/ko-kr/company-news/technology/innovations-from-google-io-26-on-google-cloud/).
## AI Max for Apps and AI Brief: more automation, less marketer visibility
[]
At Google Marketing Live 2026, Google announced AI agents that can build app deep links in "minutes, not days." Google also unveiled AI Brief: marketers describe brand voice, audience, and guidelines in a single natural-language prompt, and the AI generates ads within those bounds.
The center of gravity in campaign operations is moving from execution to signal design.
🎥 Watch: [Google — AI Max for Apps](https://youtu.be/KUyRq7szZsM) (YouTube)
👉 More detail: [Gemini Omni Flash official announcement — Google blog](https://blog.google/intl/ko-kr/company-news/technology/gemini-omni-kr/).
### 1\. Campaign controls shrink to a single natural-language line
Faster deep link generation means you can produce dozens of A/B variants overnight. The weekly creative meeting goes away, replaced by a review meeting that audits whether auto-generated variants stay inside your brand tone and measurement signals. If you don't predefine how to group 100 creatives for review, the time saved by automation gets eaten back by review overhead.
This isn't just AI Brief. Performance Max and Google App Campaigns are moving in the same direction. Once AI Max for Apps lands, AI will generate the ad creatives directly, and the data quality you control will determine nearly all of your campaign performance.
### 2\. Signal design and brand tone are the last guardrails
Auto-generated creative now extends to images and video. Newly announced [Gemini Omni Flash](https://blog.google/intl/ko-kr/company-news/technology/gemini-omni-kr/) generates video from a single line of natural language, swaps backgrounds, and personalizes digital avatars per viewer.
[Google Pics](https://blog.google/intl/ko-kr/company-news/technology/workspace-update-io-2026-kr/), powered by Nano Banana, handles object segmentation, text editing, and multilingual translation in one pass. Type what you want, and the AI produces text, video, and image variants together.
Creative production time is about to collapse. In that environment, signals decide outcomes. If your definitions for conversion value, interest signals, and high-value users are vague when you hand AI automation broad authority, the AI's idea of a "good user" could land on a low-business-value cohort.
Brand tone follows the same logic. A single sentence in AI Brief shapes ad copy and recommended images, so leaving brand guidelines in marketers' heads makes the input drift every time, and your ad consistency falls apart. This isn't a copywriting problem, it's a system problem.
## Universal Cart: half the commerce app funnel now happens inside Google
Universal Cart is a shopping hub where users add to cart and complete checkout across Search, Gemini, YouTube, and Gmail. Major global retailers, including Nike, Walmart, and Shopify merchants, are launch partners.
[]
Google hasn't announced regional launch timing yet, but Universal Commerce Protocol (UCP) is expanding to Canada, Australia, and the UK, with YouTube, hotel booking, and local food delivery categories on the roadmap.
👉 More detail: [Google Shopping's Universal Cart announcement](https://blog.google/products-and-platforms/products/shopping/google-shopping-cart/).
### 1\. When checkout happens outside your app, audit the SKU feed first
UCP is an open standard that ties merchant, payment, and ad data together. With Agent Payments Protocol (AP2) layered on top, agents can transact on a user's behalf, a "Human Not Present" payment where no human authorizes each transaction directly.
For example: when a user says "Find me a running shoe under $300 and buy it," Gemini selects the product and completes checkout. But when the transaction happens outside your app's SDK, your first-party data alone can't verify whether ad impressions or clicks drove that revenue.
For agents to pick your product, the Merchant Center feed (the Google Shopping product catalog) is table stakes. On top of that, you need Content API to sync price, inventory, and promotions in real time, plus the merchant metadata UCP requires. So you'll need to track quarterly what percentage of your SKUs sit in "Eligible" status. If you don't, no amount of ad spend will fix the mismatch between your data and ad performance.
### 2\. You need a join key between agent transactions and first-party data
When an agent transacts, the signals that show up in your data are limited to order ID, payment amount, and shipping address. If you don't tie those to the ad or content exposure that influenced them, and to the same user's prior in-app behavior, your ad ROI will read lower than reality.
In Airbridge, joining the same user's in-app activity requires a [cross-channel attribution setup](https://www.airbridge.io/en/blog/cross-channel-attribution-guide-conversion-performance). Based on this announcement, agree with your data team on candidate join keys (common identifiers that connect Google order data to your first-party user data) and document them now. A year from now, your attribution gaps will be smaller and your data accuracy higher.
## Google Play's 60-day payment grace period rewrites the subscription LTV baseline
As of December 1, 2025, Google Play extended its default account hold (the grace period that keeps a subscriber's account active when payment fails) from 30 to 60 days (auto-calculated). Plans previously set to 30 days were auto-migrated.
Google reported in the same announcement that, among top developers, involuntary churn dropped up to 18% and total churn fell 9%. Given AI bidding automation and LTV models share the same baseline, a single policy line can throw off the entire measurement system's accuracy.
👉 LTV impact analysis: [RevenueCat: Play Billing Library 9.0 breakdown](https://www.revenuecat.com/blog/engineering/play-billing-v9/).
### 1\. As the retention curve climbs, LTV models underestimate accordingly
Compare Q4 2025 and Q1 2026 retention curves and you'll see the curve rise naturally. But this isn't a marketing or product win, it's the policy change. If you don't separate the two and conclude "retention is up, push UA budget," you'll spend against an inflated efficiency reading.
The subscription [predictive LTV (pLTV) model](https://www.airbridge.io/en/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app) is typically calculated as retention curve × ARPU (average revenue per user). A model trained on the 30-day grace curve will systematically underestimate LTV in the 60-day environment.
For instance: if your LTV reading is 15% below reality, automated bidding under the same ROAS target will set conservative bids, and channel and creative evaluation drift by the same margin. The model won't auto-adjust to policy shifts, so you need to retrain the LTV model yourself.
### 2\. Manage pre- and post-policy cohorts separately
A first step: add a December 1 toggle in your BI dashboard and run v1 and v2 LTV models in parallel. Then document, with reasoning, which model feeds bidding automation each quarter. That way, the next quarterly review doesn't relitigate the same decision.
## AI Mode at 1 billion: the next attribution question is incrementality
AI Mode crossed **1 billion** monthly users one year after launch, and **88%** of AI Mode queries end without an external click ([Google I/O 2026 keynote](https://blog.google/products-and-platforms/products/search/search-io-2026/) · [Digital Applied Zero-Click Search Statistics 2026](https://www.digitalapplied.com/blog/zero-click-search-statistics-2026-complete-data), 2026). At the same event, Google's total AI processing went from 480 trillion tokens monthly last year to **3.2 quadrillion** this year, a **7x** jump in a single year.
AI is no longer a layer bolted onto search. It's the search infrastructure itself.
> “Search has become less about individual queries and feels more like an ongoing conversation, giving you deeper insights and connecting you with the vastness of the web.” — Sundar Pichai, CEO of Google and Alphabet
🎥 Watch: [Sundar Pichai I/O 2026 Keynote — AI Mode segment (4:13~)](https://youtu.be/duHhImuaZGU?t=253) (YouTube)
👉 More detail: [Google blog: Search at Google I/O 2026](https://blog.google/products-and-platforms/products/search/search-io-2026/).
### 1\. In a 60% zero-click environment, the next question is incrementality
**64.82%** of all Google searches now end without a click ([Zero-Click Search Statistics 2026](https://www.digitalapplied.com/blog/zero-click-search-statistics-2026-complete-data), 2026). An Ahrefs field study from the same period found that for searches where AI Overviews (the AI-answer cards at the top of results) appeared, zero-click rates rose from 54% to 72%, and organic clicks dropped 38% (Search Engine Journal, Ahrefs field study, 2026).
Evaluating channels purely on click counts means missing real exposure. The question marketers ask has to shift. Instead of "where did this converting click come from?" the question becomes "how much revenue would I lose if I turned this channel off?" — that is, incrementality.
In line with that, **46.9%** of U.S. marketers said they plan to invest more in marketing mix modeling (MMM) over the next year, and **36.2%** in incrementality ([eMarketer 2026 Measurement Trends Report](https://www.emarketer.com/content/mmm--incrementality--other-measurement-trends-that-will-define-2026), 2026).
The lightest entry point: a holdout test. Turn one channel off in a single region for 1-2 weeks, measure the revenue change, and you'll have a baseline to correct your click attribution against. See our [bias-free incrementality test design](https://www.airbridge.io/en/blog/mobile-game-reengagement-and-retargeting) for retargeting and re-engagement.
### 2\. Measure your brand mentions inside AI answers
Zero-click also means AI answers are a new exposure surface. When AI Mode or Gemini mentions your brand in an answer, awareness happens without a click. The measurement domain for this exposure is GEO (Generative Engine Optimization).
Take 50-100 representative category queries, run them quarterly across AI Mode, Gemini, and ChatGPT, and log your brand's mention frequency, position, and citation type (summary, link, image). Then compare against competitor mentions to calculate share of voice. A quarterly automated script is enough.
## Where does your measurement stack stand in the AI-agent era?
Each of the five shifts forces a re-check on a different part of the measurement stack: discovery (Ask Play, Gemini), campaign signals (AI Max, AI Brief), commerce attribution (Universal Cart, AP2), subscription cohorts (60-day grace), and post-click measurement (AI Mode, incrementality, GEO). Picking one to address won't be enough — the stack as a whole needs review.
### A diagnostic: can your measurement stack answer the five shifts?
- **Acquisition path breakdown.** Can you separate organic installs into Play Store search, Ask Play, Gemini recommendations, Engage SDK, and others? → With only Play Store referrer, all four flow in as the same "organic" bucket.
- **Signal documentation.** Are your campaign's conversion values, high-value user definitions, and brand guidelines documented and updated quarterly? → AI Brief and AI Max are only as good as the input signal quality.
- **SKU eligibility tracking.** Do you check quarterly what percentage of your SKUs sit in "Eligible" status in the Merchant Center feed? → In Universal Cart, AI agents only pick from products properly registered in the feed.
- **Cohort separation.** Are you separating pre- and post-December 1, 2025 cohorts when reading LTV and retention curves? → December 1, 2025 is the day Google Play's grace period extended to 60 days. Mixing the cohorts blurs policy effects with marketing effects.
- **Post-click validation.** In the last 12 months, have you validated marketing impact outside click attribution (holdout tests, conversion lift)? → In a zero-click environment, click counts alone can't reveal real channel impact.
Two or more "no" answers mean your [measurement stack](https://www.airbridge.io/en/blog/what-is-mmp) should be a quarterly priority. One "no": build a remediation plan within the quarter. None? Then in the next quarter, push your priorities into newer domains like GEO measurement.
## Closing thoughts
Each announcement looks different on its own, but pull back a step and they point to the same place. Visibility into what users do before they reach your app is shrinking on the first-party side, and the measurement unit itself is shifting from "clicks" to "intent and incrementality."
Before adopting a new solution, lay out your measurement stack and start from the weakest of the five shifts. That's where your next quarter's OKR begins.
**References**
1. Sundar Pichai. ["Google I/O 2026 keynote"](https://blog.google/innovation-and-ai/sundar-pichai-io-2026/). blog.google, 2026.
2. Android Developers Blog. ["I/O 2026: What's new in Google Play"](https://android-developers.googleblog.com/2026/05/io-2026-whats-new-in-google-play.html). 2026.
3. Google. ["Innovations from Google I/O 2026 on Google Cloud"](https://blog.google/intl/ko-kr/company-news/technology/innovations-from-google-io-26-on-google-cloud/). 2026.
4. Google. ["Gemini Omni Flash announcement"](https://blog.google/intl/ko-kr/company-news/technology/gemini-omni-kr/). 2026.
5. Google. ["Google Workspace I/O 2026 update (incl. Google Pics)"](https://blog.google/intl/ko-kr/company-news/technology/workspace-update-io-2026-kr/). 2026.
6. Google. ["Google Shopping's Universal Cart announcement"](https://blog.google/products-and-platforms/products/shopping/google-shopping-cart/). 2026.
7. Google. ["Search at Google I/O 2026"](https://blog.google/products-and-platforms/products/search/search-io-2026/). 2026.
8. RevenueCat. ["Play Billing Library 9.0 breakdown"](https://www.revenuecat.com/blog/engineering/play-billing-v9/). 2026.
9. Nobori AI. ["Google AI Mode zero-click rate, 100M users"](https://nobori.ai/blog/google-ai-mode-zero-click-rate-100m-users-2026). 2026.
10. Digital Applied. ["Zero-Click Search Statistics 2026 Complete Data"](https://www.digitalapplied.com/blog/zero-click-search-statistics-2026-complete-data). 2026.
11. eMarketer. ["MMM, Incrementality, Other Measurement Trends that Will Define 2026"](https://www.emarketer.com/content/mmm--incrementality--other-measurement-trends-that-will-define-2026). 2026.
---
## Deeplinks for E-Commerce Apps: Cart Recovery and Personalization
URL: https://www.airbridge.io/en/blog/deeplinks-for-e-commerce-apps
Language: English
Published: 2026-06-17
Author: Harper (Trang Nguyen)
Cart abandonment emails are a solved problem on the web. A shopper leaves, you send a reminder, they click, and they land back on the exact cart they left. In your app, that chain breaks. The reminder fires, the user taps, and they get dumped on your home screen with no cart in sight. Most never dig back to find it.
The cost is real. Mobile shoppers abandon carts at roughly **80%**, well above desktop, and across all of e-commerce abandonment hovers around **70%**. The fix is not another discount. It is a routing problem, and deeplinks are how you solve it.
**Key Takeaways**
- **A reminder is only as good as where it lands.** A push or email that drops users on the home screen instead of their cart throws away the click you paid for.
- **Deeplinks route taps to the exact in-app screen.** A deeplink opens the specific cart, product, or offer, not the app's front door.
- **Deferred routing covers the users who matter most.** A deferred link preserves the destination even when the app is not installed yet, sending the shopper through the store and into the right screen after install.
- **Timing beats everything.** A first cart reminder sent within one to two hours can drive far more recovered revenue than one sent the next day.
- **Cart recovery is the entry point, not the ceiling.** The same routing layer powers personalized product, post-purchase, and loyalty journeys.
## Why Cart Recovery Breaks Inside Your App
On the web, every page has a URL. Your abandoned-cart email links to `yourstore.com/cart`, the browser opens it, and the cart is right there. There is no install step, no app-versus-browser decision, no routing logic to get wrong.
Apps do not work that way by default. Tapping a notification or an email link opens the app at whatever screen it feels like, usually the home feed. The shopper who was 30 seconds from checkout now has to search for the product, re-add it, and re-enter their details. Each extra tap is a chance to quit, and on a small screen with easy distractions, most do.
This is why app cart-recovery campaigns underperform their web cousins even when the copy and timing are identical. The message is fine. The landing experience is broken. Baymard Institute estimates that a large share of the roughly **$260 billion** in abandoned-cart value across the US and EU is recoverable through better checkout and re-engagement design. For an app, "better design" starts with sending the tap to the right place.
## Deeplinks 101: Standard vs Deferred
A deeplink is a URL that opens a specific screen inside an installed app instead of the home screen. Tap one that points to a product and the app opens on that product page. It is the in-app equivalent of a web link that resolves to `/cart` rather than the domain root.
The difference that decides whether your campaign works comes down to two types:
- **Standard deeplink.** Opens the right in-app screen, but only when the app is already installed. If it is not, the link fails: it shows an error or falls back to a mobile web page, and the routing context is lost.
- **Deferred deeplink.** Holds onto the destination across the install boundary. If the app is missing, it routes the user to the App Store or Google Play first, then opens the intended screen automatically once the app launches for the first time.
For cart recovery this matters more than it sounds, because a meaningful slice of your "lapsed" shoppers have deleted the app or never installed it. A standard link abandons them. A deferred one carries them all the way to their cart.
[diagram-deeplink-routing.webp]
If you want the deeper technical comparison of how each type behaves across iOS and Android, this breakdown of [deferred vs standard deeplinks](https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link) is a good companion read.
## How to Build a Deeplink Cart-Recovery Flow
You do not need a large team to ship this. If you have ever wondered how to send users back to abandoned cart via app deeplink, the answer is three moving parts: a cart event, a channel, and a link that points at the cart. Here is the sequence.
### 1\. Capture the abandoned-cart event
Define what "abandoned" means for your app: an item added to cart with no purchase within a set window, often 30 to 60 minutes. Fire an event the moment it happens. This event is the trigger for everything downstream, so make sure it carries the cart contents and a stable identifier you can map back to the user.
### 2\. Choose the recovery channel
Each channel has a different reach and conversion profile. Push is fast and cheap but only reaches users who opted in. Email reaches everyone but converts lower. SMS converts hard but costs more per send.
| Channel | Typical open rate | Typical cart conversion | Best for |
| --- | --- | --- | --- |
| Push | 30 to 40% | 8 to 12% | Opted-in active users, fast first touch |
| Email | 15 to 25% | 5 to 8% | Broad reach, richer content |
| SMS | High | 21 to 40% | High-intent, time-sensitive nudges |
Most teams stack them: a push within the hour, an email a few hours later, an SMS for the highest-value carts. Adding SMS on top of email has been shown to lift overall cart recovery toward **38%**.
### 3\. Route the tap with a deeplink
This is the step that separates a recovered sale from a lost one. Every recovery message must carry a deeplink that opens the specific cart, not the home screen. Build the link with the cart identifier from step 1 so the app knows exactly what to render. Vendor case studies report that email-to-app journeys built on proper routing can lift conversions by up to **4x** versus links that dump users on a generic screen.
### 4\. Cover the app-not-installed case
For any user who might not have the app, use a deferred deeplink instead of a standard one. The shopper who deleted your app three weeks ago still gets routed: store, install, cart. Without this, a large and valuable segment of your reminders simply dead-end.
### 5\. Time the sequence
Speed is the highest-leverage variable. A first reminder sent within one to two hours of abandonment can dramatically outperform one sent a day later. Send the immediate push first, then space follow-ups over the next 24 to 48 hours. Sending a conversion-focused message minutes after a user exits without buying is a well-established lifecycle tactic (Law 36, Push Notification Immediately After Exit, in the 48 Laws of Subscription App Success), because intent is never higher than right after they leave.
A simple, reliable cadence looks like this: a push within the first hour, an email at four to six hours for anyone who did not return, and a final SMS the next day reserved for higher-value carts (the next-day follow-up echoes Law 37, Send a Second Offer the Next Day). Each touch carries the same deeplink to the same cart, so no matter which message a shopper acts on, the destination is identical and the path to checkout stays short.
## Beyond the Cart: Personalization with Deeplinks
Once the routing layer exists, cart recovery is just the first use case. The same link that opens a cart can open any screen tied to a specific user and a specific intent. This is where deeplinks shift from a recovery tool to a retention engine.
A few high-value plays:
1. **Personalized product re-engagement.** Browsed but did not add to cart? Send a link straight to that product or a curated set of similar items.
2. **Post-purchase journeys.** Route a buyer to order tracking, a how-to-use screen, or a complementary product. First-time buyers who get personalized post-purchase follow-ups have shown materially higher repeat-purchase rates.
3. **Loyalty and offers.** Drop users directly into a rewards screen or a personalized discount instead of making them hunt for it.
The retention payoff is measurable. Deeplinking strategies have been associated with roughly **12 to 15%** lifts in one-day, one-week, and one-month retention, and personalized deeplink onboarding has been linked to large day-30 retention gains. The mechanism is the same every time: less friction between the message and the moment of value.
## Setting Up Deeplinks the Right Way
Deeplinks rely on platform-level standards, and the setup is where most implementations quietly break. On iOS you configure Universal Links, which require an `apple-app-site-association` file hosted over HTTPS on your domain. On Android you configure App Links, verified with a Digital Asset Links file. You need both, since they are not interchangeable.
A few practical rules that save days of debugging:
1. **Always include a web fallback.** If the link cannot open the app, it should still land the user on the equivalent web page with the same parameters. Never let a tap dead-end.
2. **Design platform-agnostic URLs.** Use query parameters both platforms can parse and safely ignore when unknown, so an iOS-shared link does not break for an Android user.
3. **Budget time for propagation.** iOS caches the association file aggressively. Apple's guidance puts propagation at up to 24 hours (Technical Note TN3155), and CDN refresh can add more, so plan for a day or two. Test on real devices with a fresh install, not the simulator.
If maintaining association files, fallbacks, and per-platform routing across both stores sounds like more than you want to own, this is exactly the layer a deeplinking provider handles for you, which brings us to the last piece.
## Common Mistakes That Break Deeplink Cart Recovery
Even a correct setup leaks revenue when these slip through:
1. **Linking to the home screen "for now."** It is the single biggest conversion killer and it never gets fixed later. Route to the cart from day one.
2. **Using standard links everywhere.** Lapsed and uninstalled users are often your highest-intent recovery segment. Skipping deferred routing abandons them.
3. **Ignoring the web fallback.** A broken tap is worse than no message, because you paid for the click and taught the user your reminders do not work.
4. **Sending late.** A perfectly routed reminder that arrives a day after abandonment competes with a cold lead, not a warm one.
5. **Not knowing which message worked.** If you cannot tell whether the push, the email, or the SMS drove the recovered purchase, you cannot cut the spend that is wasted.
## Turn Abandoned Carts Into Recovered Revenue
The web already taught your shoppers to expect a reminder that lands them right back where they left off. Your app can meet that expectation, but only if every tap routes to the exact cart and survives the case where the app is not installed.
That takes a deeplinking layer that handles standard and deferred routing across both stores, plus a way to see which messages actually drive purchases instead of just opens. If you are running cart-recovery and re-engagement campaigns and want both, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your deeplinks to the purchases they generate. Start free with 15K attributed installs, or [book a demo](https://www.airbridge.io/en/book-a-demo) to see it on your own flows.
---
## Universal Links vs URI Schemes: When to Use Each for iOS Deep Linking
URL: https://www.airbridge.io/en/blog/universal-links-vs-uri-schemes-when-to-use
Language: English
Published: 2026-06-17
Author: Harper (Trang Nguyen)
If you are building deep linking for the first time, the first decision is also the most confusing: should a tapped link use a Universal Link (`https://myapp.com/product/123`) or a custom URI scheme (`myapp://product/123`)? Both can open your app and route the user to a specific screen, so they look like they do the same job.
They do not. The two mechanisms work differently under the hood, fail differently when the app is missing, and carry very different security properties. Pick the wrong one for a given flow and you get broken links, a confusing "open in app?" prompt, or, in the worst case, another app intercepting data meant for yours. This guide breaks down how each works, what Apple recommends today, and exactly when to reach for each one.
**Key Takeaways**
- **Universal Links are Apple's recommended default.** They use ordinary HTTPS URLs tied to a domain you own, so they open your app when it is installed and your website when it is not, with no broken-link dialog.
- **URI schemes cannot prove ownership.** Any app can register `myapp://`, which makes custom schemes hijackable and unsafe for sensitive data like login tokens.
- **Universal Links rely on an AASA file.** A small JSON file hosted on your domain (`apple-app-site-association`) tells iOS which app is authorized to open which paths.
- **URI schemes are not dead.** They remain the right tool for OAuth callbacks, app-to-app communication, and hybrid webview navigation where no public web URL exists.
- **Most apps use both.** Universal Links for user-facing links from email, ads, and the web; URI schemes for internal and inter-app plumbing.
- **The hard part is operating, not choosing.** AASA hosting, edge caching, fallbacks, and deferred routing across the install gap are where first implementations break.
## What Is a URI Scheme?
A URI scheme (also called a custom URL scheme) is the original way iOS apps registered themselves to be opened by a link. You declare a scheme such as `myapp` in your app's `Info.plist`, and from then on the system knows that a link like `myapp://product/123` should launch your app and hand it that URL to parse.
### How a URI Scheme Works
The flow is simple, which is part of its appeal:
1. Your app registers the scheme `myapp` in `Info.plist` under `CFBundleURLTypes`.
2. iOS records that `myapp://` belongs to your app.
3. When any link, button, or other app opens `myapp://product/123`, iOS launches your app.
4. Your app reads the URL and routes the user to the right screen.
There is no server involved and no domain to configure. That makes custom schemes fast to set up and useful for private communication between apps you control, or as a lightweight callback during a login flow when you do not have a backend in place.
### The Catch: No Ownership Verification
Here is the problem that shaped everything Apple did next. **iOS does not verify that you own a URI scheme.** Nothing stops a second app from also registering `myapp://`. When two apps claim the same scheme, the system resolves the conflict by install order rather than by verifying identity, so the behavior becomes unpredictable.
That ambiguity is a real security surface, not a theoretical one. A malicious app can register a scheme already used by a legitimate app and intercept whatever data gets sent to it, which can include login authorization codes or access tokens. Security researchers, including FireEye in 2015, documented this as URL scheme hijacking, and it is the main reason custom schemes are unsafe for anything sensitive.
Two more practical drawbacks:
- **No graceful fallback.** If the app is not installed, tapping `myapp://product/123` does nothing useful. The user sees an error or a dead link, because the scheme is meaningless to a browser.
- **Not a real web link.** A custom scheme URL cannot be opened as a normal web page, shared cleanly, or indexed, because it only means something on a device where the app is installed.
## What Are Universal Links?
Universal Links, introduced by Apple in **iOS 9** at WWDC **2015**, solve the ownership problem by building deep linking on top of standard HTTPS URLs. Instead of inventing a private scheme, you use a real web address you already control, such as `https://myapp.com/product/123`, and prove to Apple that your app is allowed to handle links for that domain.
### How a Universal Link Works
The verification happens through a file called the Apple App Site Association, or AASA. Per [Apple's setup documentation](https://developer.apple.com/documentation/xcode/supporting-universal-links-in-your-app), the steps are:
1. Host a JSON file named `apple-app-site-association` at `https://yourdomain.com/.well-known/apple-app-site-association`, served over HTTPS with a valid JSON body and no file extension (Apple recommends the `application/json` MIME type).
2. List your app's identifier (Team ID plus bundle ID) and the URL paths your app should handle inside that file.
3. Add the Associated Domains capability in Xcode and list your domain with the `applinks:` prefix, following [Apple's associated domains guide](https://developer.apple.com/documentation/xcode/supporting-associated-domains).
4. When the app is installed, iOS fetches and caches the AASA file, confirming that your app is authorized for that domain.
5. When a user taps `https://myapp.com/product/123`, iOS checks the cached file. If the path matches, it opens your app. If not, or if the app is not installed, the same link opens in Safari and the user lands on your website.
### Why Apple Built Them
Because the link is an HTTPS URL bound to a domain you control, another app cannot claim it. Apple verifies the relationship between domain and app through the AASA file, which closes the hijacking hole that custom schemes leave open. Universal Links also degrade gracefully: the same URL works as a normal web link, so a user without the app still reaches your content instead of hitting a dead end. That single property, one link that works whether or not the app is installed, is why Universal Links are [Apple's recommended choice](https://developer.apple.com/documentation/xcode/allowing-apps-and-websites-to-link-to-your-content) for any link a user might tap from email, an ad, a message, or a web page.
## Universal Links vs URI Schemes: Side by Side
| Property | URI Scheme (myapp://) | Universal Link (https://...) |
| --- | --- | --- |
| Link format | Custom scheme | Standard HTTPS URL |
| Ownership verification | None: any app can register it | Verified via AASA file on your domain |
| Hijackable by other apps | Yes | No |
| Fallback when app not installed | Fails or dead link | Opens your website in Safari |
| Server setup required | No | Yes (host AASA file) |
| Works as a normal web link | No | Yes |
| Introduced | Early iOS | iOS 9 (2015) |
| Best for | Internal flows, OAuth callbacks, app-to-app | User-facing links from web, ads, email |
The pattern is clear: Universal Links win on security and user experience, while URI schemes win on simplicity for cases where there is no web URL and no untrusted party involved. The diagram below shows why the install-state fallback is the deciding factor.
[diagram-link-fallback-behavior.webp]
## When to Use Each
You do not have to pick one mechanism for the whole app. You pick per flow. Here is the decision framework.
### Use Universal Links When the Link Is User-Facing
If a human will tap the link from outside your app, default to a Universal Link. That covers the large majority of deep linking:
- Links in marketing emails, push notifications, and SMS
- Links in paid ads and social posts
- "Open in app" links from your own website
- Shared links between users, such as referrals, invites, and shared content
In all of these cases you need the link to work whether or not the app is installed, and you need it safe from interception. Universal Links deliver both.
### Use a URI Scheme When There Is No Web URL
Custom schemes are still the right tool for specific, lower-risk plumbing where a public HTTPS URL would be awkward or unnecessary:
1. **OAuth and OIDC callbacks.** Native apps commonly register a custom scheme as the redirect target so the authorization server can hand control back to the app after login. This is a long-standing, accepted pattern documented in [OAuth 2.0 for Native Apps (RFC 8252)](https://www.rfc-editor.org/rfc/rfc8252.html), which also requires PKCE to close the same interception risk schemes carry.
2. **App-to-app communication.** When your own suite of apps needs to pass data or context to each other, a custom scheme (often following the `x-callback-url` convention) is a lightweight way to do it without a backend.
3. **Hybrid and webview navigation.** Moving between native screens and embedded webviews inside your own app, where the routing never leaves the device.
4. **Legacy support.** Older integrations and OS versions that were built around a scheme and have not been migrated yet.
The common thread: the link stays inside a trusted boundary (your own apps, or an auth flow you started), and there is no expectation that a browser should ever open it.
### Why Most Apps Need Both
In practice, a mature app ships both. Universal Links handle everything a user taps from the outside world. A custom scheme quietly handles OAuth redirects and internal routing. The mistake first-time implementers make is treating it as an either-or decision. It is not. It is a question of matching each mechanism to the job it does best.
## Common Mistakes That Break Deep Linking
Choosing the right mechanism is only half the work. Most broken first implementations fail in operation, not in design. Watch for these:
- **AASA edge caching surprises.** From iOS 14 onward, Apple routes AASA requests through its own CDN rather than fetching directly from each device, so your file can be cached at Apple's edge and lag your latest changes by hours. Do not expect instant updates after you edit the file. During development, append `?mode=developer` to your `applinks:` entry to bypass the CDN cache and fetch the file directly.
- **Wrong file delivery.** The AASA file must be a valid JSON body served over HTTPS, with no extension and no redirects (Apple recommends the `application/json` MIME type, and keeping the file under 128 KB). A 200 response with HTML, or any redirect, silently breaks verification.
- **No fallback page.** A Universal Link that lands a no-app user on a 404 wastes the whole point. The web URL should serve a real page, ideally one that nudges an install.
- **Forgetting** [**deferred deep linking**](https://www.airbridge.io/en/glossary/deferred-deep-link)**.** A standard Universal Link cannot, on its own, route a brand-new user to the right screen after they install from the App Store. Routing across the install gap is a [separate problem](https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link) that needs a deep linking layer to solve.
- **Testing only the happy path.** Always test app-installed and app-not-installed, plus cold start versus already-running, on a real device. A good [link testing and debugging routine](https://www.airbridge.io/en/blog/airbridge-deeplink-testing-debugging-guide) catches most of these before launch.
## Choosing Your Approach
If you are wiring up your first deep links, the rule of thumb is short: **Universal Links for anything a user taps, URI schemes for internal and inter-app plumbing.** That single sentence is the right call for the vast majority of your links, and it lines up with how Apple wants modern apps built.
The deeper challenge shows up once links start driving real growth. As soon as you are sending links from ads, emails, and referral flows, you have to host and maintain AASA files, handle the no-app fallback, route new users correctly after install, and know which of those links and channels actually drive sign-ups and subscriptions. Doing all of that by hand is where small teams lose weeks.
This is the point where a [deep linking platform](https://www.airbridge.io/en/product/deep-linking) earns its place. Airbridge provides app links that handle the Universal Link and URI scheme setup, the website fallback, and deferred routing across the install gap, while also showing you which links and channels turn into paying users. Deep linking is included in both the Core Plan and the Growth Plan, so the linking layer comes built in alongside measurement rather than as a separate tool to wire up.
**If you want app links that route users correctly and show you which channels drive subscriptions,** [**start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core)**: 15K installs included.**
---
## Why Your Universal Links Open in Safari Instead of the App
URL: https://www.airbridge.io/en/blog/your-universal-links-open-in-safari-instead-of-the-app
Language: English
Published: 2026-06-16
Author: Harper (Trang Nguyen)
You tap a link that should drop straight into your app, and Safari opens instead. The app is installed. The link looks correct. Yet iOS quietly treats it as an ordinary web page.
When Universal Links fail this way, the problem is almost never the link itself. It is the **chain of trust** behind it: your apple-app-site-association file, your Associated Domains entitlement, and the cached decision iOS makes before it ever opens your app. Get one detail wrong in that chain and every tap lands in the browser.
**Key Takeaways**
- **Universal Links break silently.** iOS does not show an error when the association fails. It just opens Safari, which is why the cause is so hard to spot.
- **The AASA file is the usual culprit.** A wrong App ID, a syntax error, or a redirect on the file path will sink the whole setup.
- **The entitlement and the file must agree.** Your `applinks:` entitlement and the apple-app-site-association file have to reference the same domain and the same App ID.
- **Apple caches the decision.** Since **iOS 14**, a content delivery network fetches and caches your file, so edits do not take effect instantly.
- **A direct paste never opens the app.** A Universal Link only fires on a real user tap from another app, not from typing a URL into the address bar.
## What a Universal Link Is and Why It Sometimes Opens in Safari
A Universal Link is a standard web URL that iOS can route directly into your app when the app is installed, and fall back to your website when it is not. There is no custom scheme and no confirmation prompt. The same `https://` link works whether the app is present or not.
For that routing to happen, iOS has to trust that your app owns the domain. It checks this through a file your domain serves at `/.well-known/apple-app-site-association`, paired with an Associated Domains entitlement inside your app. **When both sides match, the tap opens the app.** When anything is off, iOS falls back to the safest option: it opens the web page in Safari.
This is the exact moment most teams get stuck. Search engines surface the question "why do universal links open in safari instead of app" more often than any other deep linking issue, precisely because the failure is invisible. There is no crash, no log line, and no warning. The link simply behaves like a normal URL.
[universal-link-flow-diagram.webp]
## The Most Common Reasons Universal Links Aren't Working
Almost every case of a broken Universal Link traces back to a short list of misconfigurations. The table below maps each symptom to its **root cause** and the fix, so you can match what you are seeing to where the break actually is.
| Symptom | Root cause | Fix |
| --- | --- | --- |
| Link always opens Safari | App ID in the file is wrong | Use TEAMID.bundleID, not the App ID prefix |
| File never loads on device | Redirect on the file URL | Serve the file at the exact path with no redirect |
| Worked, then stopped after edit | Apple CDN cached old file | Wait for the cache, or use the development toggle |
| Some paths open, others don't | Path or component mismatch | Fix the paths or components entries |
| Pasted link opens web page | Not a user-initiated tap | Test from a real tap in Messages or Notes |
### 1\. The App ID is in the wrong format
The single most common error is the App ID. It is **always** written as `TEAMID.bundleID`, never the bundle ID alone. Many teams paste the App ID prefix from the developer portal instead of the real Team ID, and the two do not always match. When the App ID in the file does not match the entitlement in the build, iOS rejects the association and routes the tap to Safari.
### 2\. The file has a syntax or format error
The apple-app-site-association file must be **valid JSON** with no trailing commas, no stray brackets, and no comments. Since **iOS 14**, the structure changed: `appID` became `appIDs` as an array, and a `components` field was added alongside the older `paths` syntax. A file that still uses the old single-string format, or that mixes the two incorrectly, can fail to parse. Since **iOS 14**, the file is plain JSON with no CMS signature required. Legacy guides that tell you to sign the payload predate this change, and following them recreates the failure.
### 3\. The file is not served cleanly over HTTPS
The file has to sit at `https://yourdomain.com/.well-known/apple-app-site-association` and return directly over HTTPS with **no redirect**. Even a harmless-looking redirect, such as HTTP to HTTPS or a trailing slash added by your server, will cause iOS to abandon the fetch. The file should be served with a valid certificate and no authentication wall.
### 4\. The entitlement and the file disagree
Your app declares which domains it handles through the `applinks:` entry in its Associated Domains entitlement. If that entitlement names a domain or App ID that the apple-app-site-association file does not match, iOS treats the link as a normal web URL. The entitlement also has to be present in the build that is actually installed, not just in your project settings.
### 5\. The link is not a real user tap
Universal Link routing is deliberately conservative. A link typed or pasted into the Safari address bar will open the web page, **by design**. The same is true for links opened programmatically without user intent. To test correctly, tap the link from another app, such as a note, a message, or an email.
## How to Fix and Validate Your AASA File Step by Step
Once you know the likely cause, the fix is a short, ordered process. Work through it in sequence rather than changing several things at once, so you can tell which step resolved the issue. Apple documents the full debugging flow in its [universal link troubleshooting note](https://developer.apple.com/documentation/technotes/tn3155-debugging-universal-links), and a clear end-to-end walkthrough is laid out in this [universal link setup guide](https://www.airbridge.io/en/blog/universal-links-setup-guide-ios).
### 1\. Confirm the App ID and entitlement match
Open your apple-app-site-association file and your entitlements side by side. Confirm the `appIDs` array uses the real Team ID and bundle ID, and that the `applinks:` entitlement names the same domain. These two have to agree before anything else can work.
### 2\. Validate the file format and hosting
Fetch the file directly in a browser and confirm it returns valid JSON at the exact `/.well-known/` path, over HTTPS, with no redirect and a **200** status. Then verify the association on a device: tap the link with the Xcode Console open, or enable Associated Domains Development mode in iOS Settings to query your domain directly. The Console output, filtered for `swcd`, shows whether iOS accepted or rejected the association payload. Check for trailing commas and confirm the `components` or `paths` entries actually cover the URLs you want to open.
### 3\. Account for the Apple CDN cache
Since **iOS 14**, the device does not fetch your file from your server directly. It asks an Apple-managed content delivery network, which caches the result, as documented in [Apple's technical note](https://developer.apple.com/documentation/technotes/tn3155-debugging-universal-links). An edit you publish may take time to propagate. During development, enable the Associated Domains Development toggle in your device settings to bypass the cache and query your domain directly.
### 4\. Reinstall and test with a real tap
Delete the app, reinstall it from Xcode or TestFlight so the latest entitlement ships, then tap your link from inside another app. A fresh install forces iOS to re-evaluate the association. If the app opens, the chain of trust is intact.
## When a Link Management Platform Makes Sense
Maintaining this yourself is workable for one domain and a handful of paths. It gets fragile fast when you add marketing subdomains, run campaigns across regions, or need links that survive an install and still route to the right screen. Every new domain is another file to host, validate, and keep in sync with your entitlement.
A link management platform handles that infrastructure for you. It hosts and serves the association file, keeps the routing rules consistent, and gives you smart links that open the app when it is installed and the web when it is not. If you want to understand the moving parts first, this [deep linking overview](https://www.airbridge.io/en/glossary/deep-linking) and [mobile deep linking guide](https://www.airbridge.io/en/library/airbridge-handbook-deeplinks) cover the fundamentals.
The bigger payoff is **measurement**. Once your links route correctly, the next question is which links and which channels actually drive paying users. Airbridge connects each link tap to the install and the subscription that follows, so you can see which campaigns produce subscribers rather than just opens. Core Plan focuses on the four major ad channels, Meta, Google, Apple Search Ads, and TikTok, using a set of standard events rather than custom event schemas, which keeps setup decisions to a minimum.
## Get Your Links Working, Then See Which Ones Pay Off
A link that opens Safari instead of your app is a **trust problem, not a typo**. Fix the App ID, serve a clean file, respect the cache, and test with a real tap, and the routing falls into place.
From there, the value is knowing which links earn their keep. If you want to connect your links to installs and the subscriptions they drive, [start free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core), which includes **15K** attributed installs.
---
## How to Set Up Deeplinks for TikTok and Meta App Install Ads
URL: https://www.airbridge.io/en/blog/tiktok-meta-ad-deeplinks-setup
Language: English
Published: 2026-06-15
Author: Luke
Your TikTok ad shows a free trial for your subscription app. A user taps, installs, opens the app, and lands on a blank home screen. The trial offer is nowhere. They leave.
This is not a creative failure. It is a configuration failure. **44.5% of all in-app purchases happen on the day of install** ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)), and **82% of free trials start on that same day** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). If the first screen a new user sees does not match the ad that drove them there, you have burned your highest-conversion window on a dead end. Learning how to set up deeplinks for TikTok and Meta app install ads is the direct fix.
**Key Takeaways**
- **A deeplink routes users to a specific in-app screen instead of the home screen.** The ad promises a product page, free trial, or promotional offer. A deeplink delivers it.
- **For user acquisition campaigns, you need the deferred version.** Standard links only work if the app is already installed. The deferred version handles new installs and still routes to the correct screen post-download.
- **TikTok has a hard iOS 14.5+ restriction.** This feature is not supported on iOS 14.5+ (SKAN) campaigns on TikTok. This affects how you structure your iOS UA campaigns.
- **Meta's setup requires the Facebook SDK and Advertiser ID collection.** Without these, deferred linking on iOS will not function.
- **A fallback URL is non-negotiable.** If routing fails due to a wrong OS, old device, or unsupported format, users need a working backup destination, not a broken redirect.
## Why App Install Ads Drop Users on the Home Screen Without Deeplinks
A deeplink is a URL that routes users directly to a specific screen inside your app, bypassing the default home screen. Without one, every new install from a paid ad starts from zero: no context, no offer, and no continuity with the creative that drove the click.
### 1\. Day 0: The Critical Conversion Window for New App Installs
Among **75,000+** subscription apps, **82%** of free trials start on the day of install. From a separate analysis of **16,000+** apps, **44.5%** of all in-app purchases also happen on that same first day. This is not coincidence. It reflects peak intent. A user who just installed your app after clicking a paid ad is more motivated to act than they will ever be again.
When that window opens on a generic home screen instead of the specific screen your ad promised, intent decays fast. Put another way: a misconfigured routing setup locks you out of the day when 82% of trial conversions and 44.5% of all purchases statistically occur.
### 2\. How a Wrong Landing Screen Costs You ROAS
In-app landing experiences that match ad creative have a direct, measurable impact on ROAS. Across paid campaigns, **ads that route users to specific in-app screens produce conversion rates 2.8 times higher than ads that route users to mobile web** (based on Google's reported benchmarks, [Google Ads Help, April 2025](https://support.google.com/google-ads/answer/16413010)). That gap compounds across your entire campaign spend.
## Direct Deeplinks vs Deferred Deeplinks: Key Differences for App Install Ads
Not all deeplinks work the same way for user acquisition. Using the wrong type for a new-install campaign is one of the most common setup errors.
| Type | App Already Installed? | Primary Use Case | What Happens If App Is Missing |
| --- | --- | --- | --- |
| Direct deeplink | Required | Retargeting, re-engagement | Fails silently or opens app store without routing |
| Deferred deeplink | Not required | User acquisition, new installs | Routes to app store, then to correct screen post-install |
A **direct deeplink** opens a specific screen inside the app, but only if the app is already on the device. Use it for retargeting campaigns reaching existing users.
This link type stores the destination information, redirects new users to the app store if the app is not present, and then delivers them to the correct screen after installation. This is the setup required for app install ad campaigns targeting new users.
The distinction matters because the platform behavior differs. If you attach a standard direct link to a user acquisition campaign and the user does not have the app, they go to the app store, install, and land on the home screen. The routing data is lost.
[Screenshot 2026-06-15 at 15.35.18.png]
## How to Set Up Deeplinks for Meta App Install Ads
Meta's linking system uses [Meta App Links](https://developers.facebook.com/docs/app-ads/deep-linking/), which is metadata embedded in your ad that instructs the platform where to route users after they tap it and install your app.
### 1\. Enable the Facebook SDK and Advertiser ID Collection
Before touching Ads Manager, two technical conditions must be in place.
**Install and initialize the Facebook SDK** in your iOS or Android app. The SDK handles deferred link resolution: it reads the App Links metadata stored from the ad click and passes the destination to the app after install. Without the SDK, the platform has no mechanism to deliver routing instructions post-install.
**Enable Advertiser ID collection** in your iOS app settings. On iOS, this is required for deferred linking to function. Without the advertising identifier, Meta cannot match the install event back to the original ad click.
On Android, the setup is more straightforward. Android's App Links standard handles routing natively once the SDK is initialized.
### 2\. Add Your Deeplink URL in Meta Ads Manager
Once prerequisites are met, the in-platform steps take a few minutes:
1. Open Meta Ads Manager and navigate to your ad creative settings.
2. Under **Text & Links**, locate the **Deep Link** field.
3. Paste your destination URL. This can be a Universal Link (`https://yourdomain.com/offer`) or a custom URI scheme (`yourapp://offer`). Universal Links are preferred on iOS.
4. Meta automatically converts this into App Links metadata and embeds it in the ad.
5. Add a **Fallback URL** (an app store listing or mobile web page) for cases where routing cannot resolve.
When a user taps the ad and opens the app for the first time, Meta reads the stored metadata and routes them to the destination you specified.
### 3\. Configure AEM for iOS Deeplink Signal Accuracy
For iOS campaigns, confirm that your destination URL is included in your Aggregated Event Measurement (AEM) configuration. Without AEM set up correctly, Meta's privacy framework may block the event signal and cause the deferred routing to fail silently on iOS 14.5+ devices.
For Advantage+ App Campaigns (AAC), Meta manages placement and delivery automatically. You can still pass destination URLs through the ad creative settings under the same Text & Links section.
## How to Set Up Deeplinks for TikTok App Install Ads
TikTok's linking system operates by campaign type, and the iOS restriction is where most marketers run into problems before a campaign goes live.
### 1\. TikTok's iOS 14.5+ Deeplink Restriction Explained
TikTok's deferred deeplinks are **not supported on iOS 14.5+ dedicated campaigns**. These are campaigns that use Apple's [SKAdNetwork](https://www.airbridge.io/en/glossary/skadnetwork) (SKAN) framework for attribution. If your iOS campaigns run with the iOS 14.5+ dedicated toggle enabled, this feature will not work.
Deferred linking on TikTok is supported for:
- **Android campaigns** (all eligible objectives)
- **iOS 14 or lower dedicated campaigns** (App Installs, App Event Optimization, Value-Based Optimization)
Standard direct links are supported on iOS 14.5+ campaigns for the Traffic objective on a cost-per-click basis only.
Note: TikTok's deferred linking feature remains in Beta and requires activation approval from your TikTok account representative.
### 2\. Full Deeplink Setup for Android and iOS 14 or Lower Campaigns
For eligible campaigns, the in-platform steps are:
1. In TikTok Ads Manager, open your App Promotion or App Installs campaign.
2. In ad group settings, locate the **Destination URL** field.
3. Paste your destination URL.
4. Add a **Fallback URL**: TikTok generates an error if this field is empty.
5. Test the link using TikTok's built-in link validator before the campaign goes live.
### 3\. TikTok Deeplink URL Formats Compared
TikTok supports three formats ([TikTok Business Help Center](https://ads.tiktok.com/help/article/list-of-supported-deeplink-formats)):
| Format | Example | Recommended? |
| --- | --- | --- |
| Apple Universal Links | https://yourdomain.com/offer | Yes, preferred for iOS |
| Android App Links | https://yourdomain.com/offer | Yes, preferred for Android |
| Custom URI scheme | yourapp://offer | Use as fallback only |
Universal Links and App Links are the recommended formats. Custom URI scheme links open the correct screen on an installed app, but have no built-in fallback. If the app is not present, scheme links produce a broken redirect with no recovery path.
**Note:** Universal Links and App Links are link _formats_. Deferred linking is a _behavior_. The two work together when your SDK (Facebook SDK for Meta, TikTok SDK for TikTok) handles destination routing post-install. Choosing a Universal Link format does not automatically enable deferred routing behavior.
## 4 Common Deeplink Setup Mistakes That Hurt App Install Campaign ROAS
Technical setup is only the first half of the problem. These are the errors that break campaigns after configuration is complete.
**No fallback URL.** Both TikTok and Meta require a fallback destination in case routing cannot resolve. Without one, users on older OS versions, unsupported devices, or misconfigured SDK installations hit a dead end. Set the fallback to your app store listing or a relevant mobile landing page.
**Scheme-only links on iOS.** Custom URI scheme links (`yourapp://`) are fragile on iOS. They have no built-in fallback, and some browsers block them. Universal Links are the correct primary format for iOS campaigns. Scheme links are acceptable as a secondary format or as the fallback itself.
**Skipping end-to-end testing.** Platform link validators only check URL format, not actual routing behavior. Test on a physical device with the app uninstalled, on both iOS and Android, and verify that routing works correctly after a fresh install. On iOS 14.5+, ATT-related signal loss can reduce deferred link match accuracy on certain SDK setups, which makes real-device testing more important than on Android.
**Using deferred routing on TikTok iOS 14.5+ campaigns.** This is TikTok-specific and frequently missed. Adding a deferred linking URL to an iOS 14.5+ dedicated campaign either gets ignored or triggers an error at launch. If you need this routing feature on iOS, restructure into iOS 14 or lower campaigns. SKAN campaigns default to home-screen routing by design.
If your iOS audience is primarily on iOS 14.5+ (which most iOS users are as of 2026), consider these alternatives: use the Traffic objective with direct links on a cost-per-click basis, deploy a mobile web landing page with a smart banner that routes users to the right in-app screen after install, or focus onboarding optimization on cold-start users and accept that SKAN campaigns land on the home screen.
## Final Step: Test Your Deeplink Setup Before Launch
Getting the routing right solves the core problem. Users land where the ad promised. Day 0 intent stays intact.
Testing is the step most teams skip. Run each campaign destination URL on a real device with the app uninstalled, verify routing on both iOS and Android, and check first-session screen data in your analytics. Catching a misconfigured setup before launch costs nothing. Discovering it after a campaign has been running for a week costs real budget.
Once routing is confirmed, the next layer to optimize is the landing screen itself: how quickly value is communicated, and how the trial or paywall offer is presented on arrival.
---
## How to Track Influencer App Installs with Deeplinks (Not Promo Codes)
URL: https://www.airbridge.io/en/blog/influencer-marketing-deeplinks
Language: English
Published: 2026-06-15
Author: Luke
You gave an influencer a promo code. The campaign ran for three weeks. Your dashboard shows **180 redemptions**. During that same window, organic installs climbed by **1,400**.
Without an influencer deeplink on that campaign, you cannot connect a single one of those 1,400 installs to the creator. You do not know if any of them started a trial, subscribed, or renewed. For budget decisions, the campaign produced 180 visible conversions and a spike you cannot explain.
**Key Takeaways**
- **Promo codes only capture users who enter the code.** Most influencer-driven installs come from users who watched the content, searched the app name directly, and installed without touching a code.
- **An influencer deeplink is a trackable URL that survives the App Store redirect.** Unlike a standard link, it preserves attribution data from the tap all the way through to the install.
- **Deferred deep linking handles new users automatically.** The link records the tap, routes the user to the App Store, and matches the eventual install on first app open. This works up to 30 days after the original tap, depending on your attribution platform's configured deferred window.
- **The right attribution window for influencer campaigns is 14-30 days, not 7.** Influencer content drives delayed conversions that standard paid-ad windows miss entirely.
- **Installs are not the conversion metric that drives budget decisions.** Among 75,000+ subscription apps, the gap between median (34.8%) and top-performing (45.7%) trial-to-paid conversion shows that install quality, not volume, determines revenue. ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/))
## Why Promo Codes Fail to Track Most Influencer-Driven App Installs
### 1\. TikTok and Instagram Users Watch, Then Search, Not Click
On TikTok and Instagram, users rarely tap links. The typical pattern: watch content, decide to try the app, close the platform, open the App Store, search the name. The link in bio is never touched. The install happens, and it is logged as organic with no connection to the creator who drove it.
This is not a fringe edge case. Platform design limits outbound clicks. TikTok restricts clickable links in organic posts to the bio (universally available since late 2023). Instagram does not allow clickable links in feed captions or Reels descriptions, though Story link stickers do allow taps and have been available to all accounts since October 2021. For standard feed and Reels content on both platforms, watch-then-search is still the primary conversion path.
### 2\. Most Viewers Skip the Code, Even When They Convert
When a creator says "use code TRAINER20 for 20% off," some viewers enter it. More do not. They forgot the code. They installed two days later. For subscription apps where the trial is free, the incentive to type a code is low.
The redemptions you see represent motivated users who completed an extra manual step. The installs you do not see include users who converted without friction. Those are often your best users: low-friction converters who did not need a discount to act.
### 3\. Promo Codes Never Show Subscription Conversion per Creator
Even if codes captured every install, they would still miss the metric that drives influencer budget decisions: subscription conversion per creator.
Among 75,000+ subscription apps analyzed, the median trial-to-paid conversion rate is **34.8%**. Apps running 17-32 day trials reach **45.7%**. An influencer driving 500 installs with a 5% trial rate and 15% trial-to-paid is less valuable than a micro-influencer driving 150 installs with a 35% trial rate and 42% trial-to-paid. Promo codes give you zero visibility into this comparison. An **influencer deeplink** is what closes that gap.
## What Is an Influencer Deeplink and How Does It Work?
A deeplink is a URL that tells a mobile device exactly where to send a user: which app to open, and which screen or experience to land on. When built through an attribution platform, it also carries attribution data that survives the App Store redirect.
Standard App Store URLs break attribution at the redirect. UTM parameters appended to a link are stripped the moment the user leaves your link domain and enters the App Store. By the time the install happens, the source data is gone.
Apple's [App Tracking Transparency](https://www.airbridge.io/en/glossary/app-tracking-transparency-att) added a constraint: the IDFA-based deterministic matching attribution platforms historically used now requires user opt-in, and most iOS users do not opt in. Probabilistic and clipboard-based matching fill the gap, but with lower precision than pre-ATT IDFA matching.
An influencer deeplink solves both problems through a mechanism called deferred deep linking.
### 1\. How Deferred Deep Linking Carries Attribution Through the App Store
[Screenshot 2026-06-15 at 15.28.04.png]
When a user taps the influencer's deeplink, the attribution platform records the event: which influencer, which campaign, a device fingerprint, and a timestamp. The user is redirected to the App Store and installs. On first open, the app queries the attribution platform. The platform matches the first open to the earlier tap and assigns credit.
**The App Store is not a dead end. It is a handoff.** The delay between tap and first open can span hours or weeks. Deferred deep linking bridges that gap.
### 2\. Side-by-Side: What Deeplinks Track vs. What Promo Codes Track
| Signal | Promo Code | Influencer Deeplink |
| --- | --- | --- |
| Tap registered | No | Yes |
| Install attributed (click-through) | No | Yes |
| Install attributed (search-direct) | Yes (if code entered) | No (needs code fallback) |
| Trial start by influencer | No | Yes |
| Subscription conversion by influencer | No | Yes (with billing integration) |
| Delayed conversion, up to 30 days | Yes | Yes (with extended window) |
Neither method captures everything. The strongest influencer attribution setup uses both: deeplinks for click-through users, promo codes as a fallback for search-direct installs.
## How to Set Up Influencer Deeplinks: A Step-by-Step Guide
### 1\. Create One Unique Link per Creator, Not One per Campaign
In your [mobile attribution platform](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp), create a unique **influencer deeplink** for each creator. Do not reuse campaign-level links across multiple influencers.
Per-creator links let you:
- Compare subscription conversion rates by individual creator
- Identify traffic quality problems (high click count, low install rate signals)
- Scale or pause specific partnerships based on actual revenue data, not install volume
Include in the link parameters: influencer handle, campaign name, content format (video, bio link, story), and campaign date range.
### 2\. Set Your Attribution Window to 14-30 Days for Influencer Traffic
Many attribution platforms default to 7 days post-click for paid ad campaigns, using [attribution windows](https://www.airbridge.io/en/glossary/attribution-window) built for fast-deciding ad audiences. Influencer content works on a different timeline.
A creator posts a 20-minute product walkthrough. Some viewers install the same day. Others return to the video two weeks later. Others forward it to a friend who installs three weeks after the original post. **A 7-day window systematically undercounts influencer-driven installs** by treating all delayed conversions as organic.
Configure 14-30 day windows for influencer-sourced links (most attribution platforms support extended windows; verify your platform's default and adjust before the campaign goes live). Confirm how your platform handles post-window installs: they should be classified as organic, not incorrectly attributed to an older campaign.
### 3\. Add Link Placement Instructions to Your Influencer Brief
Do not assume creators know how attribution links work. Include in the brief:
- The unique deeplink URL (shortened by your platform, not by a third-party shortener that strips parameters)
- Exact placement: bio link for TikTok and Instagram, description link plus QR code for YouTube
- What to avoid: running the link through an additional shortener removes the attribution parameters
For YouTube specifically, a QR code version of the same deeplink is important. Viewers watching on a TV screen use QR codes to navigate to the App Store, not description links.
### 4\. Embed a Promo Code as a Parameter for No-Click Installs
Deeplinks can carry parameters that pre-fill fields in your app after install. If your onboarding includes a promo code step, pass the influencer's code as a link parameter. After install, the app reads the parameter and fills the code field automatically.
The result: click-through users get attributed via the link. Users who searched directly and installed without clicking can still enter the code manually. Both paths produce attribution data, and the user does not need to remember a code they saw in a video three days ago.
## Deeplink vs. Promo Code: Which Attribution Method to Use on Each Platform
| Platform | Primary Method | Fallback |
| --- | --- | --- |
| TikTok | Deeplink in bio | Promo code (verbal mention) |
| Instagram | Deeplink in bio or Story link | Promo code |
| YouTube | Deeplink in description + QR code | Promo code (verbal mention) |
| Podcast | Promo code (verbal) | Deeplink in show notes |
| Newsletter or email | Deeplink (direct click) | Promo code |
On audio platforms, the promo code carries more weight because listeners do not interact with links during playback. On short-form video, the deeplink in bio is primary. On YouTube, both the description link and QR code matter depending on whether the viewer is on mobile or a TV screen.
Alongside both methods, measure **organic lift**: record average daily organic installs for 30 days before the campaign and compare to the campaign period. The delta estimates the share of influencer-driven installs that neither the link nor the code captured.
## How to Measure Subscription Conversion by Influencer, Not Just Installs
### 1\. Why Install Volume Is a Misleading Signal for Influencer Spend
Install count is a reporting metric. Subscription conversion per influencer is the budget metric.
The difference is substantial. Here is the same budget evaluated two ways:
| Metric | Macro-Influencer (600K followers) | Micro-Influencer (45K followers) |
| --- | --- | --- |
| Installs | 1,200 | 200 |
| Trial starts | 24 (2% trial rate) | 70 (35% trial rate) |
| Trial-to-paid conversion | 40% | 40% |
| Subscribers generated | ~10 | ~28 |
| Subscribers per 100 installs | 0.8 | 14 |
Without subscription-level data, the macro campaign looks **6x more productive by install volume**. With it, the micro-influencer shows a **17x higher subscriber rate per install**.
### 2\. The Comparison That Changes Which Influencer You Scale
Trial quality compounds over the subscription lifecycle. A full trial experience boosts first renewal rates by up to 60% across subscription app categories ([Adapty, State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)).
An influencer who drives high trial-start rates is not just generating installs. That influencer is feeding your highest-LTV subscriber cohort.
### 3\. Two Rates to Track per Creator
To evaluate any influencer campaign at the subscription level, track two rates per creator:
1. **Trial start rate** (trial starts / installs): Shows which audience is genuinely interested in the product, not just curious enough to download.
2. **Trial-to-paid conversion rate**: Shows which audience actually commits to a subscription after experiencing the product.
A high trial start rate with a low trial-to-paid rate signals an interest gap: the audience is curious but the product is not delivering enough value in the trial period. A low trial start rate with a high trial-to-paid rate signals an intent gap: the onboarding is not capturing enough users at the paywall.
### 4\. Three Components You Need Connected for Full-Funnel Visibility
When you track influencer app installs with deeplink instead of promo code, you capture the click-through path from tap to install. To see subscription conversion by creator, three components must be aligned:
1. **Unique deeplinks per influencer** generating per-creator install attribution
2. **Post-install event tracking** firing on trial start, subscription, and cancellation
3. **Billing integration** routing subscription events back to the attributed install source
Without all three connected, the install and the subscription remain in separate datasets, and the budget question stays unanswerable.
## Promo Codes and Deeplinks Work Best Together, Not in Competition
Promo codes and deeplinks are both necessary. Codes capture the no-click, search-direct conversions that are invisible to link-based attribution. Deeplinks capture the click-through path and carry it into the subscription funnel.
The useful question is not "which influencer drove more installs." It is "whose audience converts to paying subscribers." That answer requires connecting the tap to the trial, through the App Store, and into the billing system. Deeplinks are the infrastructure that makes the connection possible.
---
## Email Deeplinks: How to Make Email Links Open Your App
URL: https://www.airbridge.io/en/blog/email-deeplinks-open-app-setup-best-practices
Language: English
Published: 2026-06-15
Author: Luke
Your marketing email lands. The subscriber taps a link. A browser window opens. The app you spent months building never appeared. This is the **email deeplink** gap: a configuration problem that routes users into a browser instead of the app where they convert.
**Apps convert at 3 to 4 times the rate of equivalent mobile web sessions** ([Criteo Mobile Commerce Report](https://www.criteo.com/blog/retail-travel-apps-higher-conversions-mobile/) — the canonical benchmark; cross-industry studies consistently show a 2 to 4x conversion advantage for app over browser). Routing users into the browser is not small friction. It is a measurable revenue leak caused by a missing configuration on your domain and in your app. This guide explains how to fix it.
**Key Takeaways**
- **Email links open browsers by default.** Without an OS-level app link configuration, every link in every email routes to the browser, even on a phone with your app already installed.
- **iOS and Android use different standards.** Universal Links (iOS) and App Links (Android) are HTTP-based links the operating system routes directly into your app once the domain-to-app association is verified.
- **Email platforms break deeplinks.** Klaviyo, Mailchimp, HubSpot, and most email platforms wrap every link through a redirect domain for click tracking. That redirect breaks OS domain trust and the link falls back to the browser.
- **Deferred deeplinks preserve context through install.** When the app is not installed, a deferred deeplink carries the destination through the App Store install, so the user lands on the right screen on first open.
- **Three components are required:** a hosted configuration file on your domain, updated app code, and an email-platform-aware routing solution.
## Why Email Links Open in a Browser Even When the App Is Installed
### 1\. Why Your App Does Not Open When Users Tap Email Links
When a user taps a link in an email, the operating system follows a simple rule: find a registered handler for that URL scheme and open it. Standard `https://` links have a default handler: the mobile browser. Unless you explicitly declare that your app owns a specific domain, the browser opens every time.
Custom URL schemes such as `myapp://product/123` do not solve this reliably. Gmail does not support them in email links. Microsoft Outlook wraps all links through its SafeLink system, which breaks custom schemes. Even where they work, they require the app to be installed and provide no fallback if it is not. **Universal Links on iOS and App Links on Android are the only reliable approach for email.**
### 2\. The Conversion Cost of Sending Email Clicks to a Browser
At **3 to 4 times** the conversion rate of mobile web, your app is a meaningfully better destination for every email click. When a user who already has your app installed still lands in a browser, you are paying the full cost of re-engaging a loyal subscriber for an experience that requires them to log in again on a smaller screen.
For lifecycle marketing, the loss compounds. Among subscription apps analyzed across more than **75,000 apps**, nearly **30% of annual subscribers cancel within their first month** (RevenueCat, State of Subscription Apps 2025). The re-engagement window is narrow. A win-back email that routes an existing subscriber into a browser login screen reintroduces exactly the friction the user was hoping to avoid by tapping the link in the first place.
### 3\. The Root of the Problem: Domain Association
The underlying cause is not the link itself. It is the absence of a verified association between your domain and your app binary. Until you host the right configuration files on your domain and declare the matching capability in your app, the OS has no basis for routing links there.
If you are asking how to make email links open the app instead of the browser, the answer requires changes at three levels: your web server, your app code, and your email platform.
## What Is an Email Deeplink? How iOS and Android Route Links to Apps
An email deeplink is a standard `https://` URL that the mobile operating system routes directly into a native app instead of a browser, based on a verified file hosted on the link's domain.
The two platforms handle this differently but share the same underlying logic: **the OS checks a configuration file on your domain at install time, and when the user taps a matching link, it opens the app directly instead of the browser.**
### 1\. Universal Links: Apple's System for Opening Apps From Email
[Apple's Universal Links](https://developer.apple.com/ios/universal-links/) work by hosting an `apple-app-site-association` (AASA) file on your domain. When your app is installed, iOS fetches and caches this file. When the user taps a matching URL, iOS opens the app and passes the full URL for your code to handle. If the app is not installed, iOS falls back to the browser or a web fallback page.
### 2\. App Links: Google's System for Opening Apps From Email
[Android's App Links](https://developer.android.com/training/app-links/verify-android-applinks) work by hosting an `assetlinks.json` file on your domain. When your app installs, Android verifies domain ownership using this file, completing the process in under **20 seconds**. A verified app with a matching `intent-filter` in its manifest opens automatically when the user taps the URL, with no disambiguation dialog.
### 3\. Standard Deeplinks vs Deferred Deeplinks: Which Do You Need?
A standard deeplink routes a user into the app only if the app is already installed. If the app is absent, the link falls back to the browser.
**A deferred deeplink adds a layer that preserves the destination through the install process.** The user taps the link, is directed to the App Store or Google Play, installs the app, and on first open lands on the screen the email was pointing to, not the generic home screen.
For re-engagement and win-back emails, where the recipient may have uninstalled the app, deferred deeplinks are the right choice. See [Airbridge's comparison of standard and deferred deeplinks](https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link) for a full breakdown.
Here is how a link click gets routed:
[Screenshot 2026-06-15 at 15.15.06.png]
## How to Set Up Email Deeplinks on iOS and Android
### 1\. iOS: Setting Up Universal Links
#### 1) Create and host the AASA file
Host your `apple-app-site-association` file at the path below, served over HTTPS with a `Content-Type: application/json` response header and no file extension. The file has a **128 KB size limit** ([Apple Developer documentation](https://developer.apple.com/ios/universal-links/)):
```
https://yourdomain.com/.well-known/apple-app-site-association
```
A minimal AASA configuration:
```json
{
"applinks": {
"apps": [],
"details": [
{
"appID": "TEAMID.com.yourcompany.yourapp",
"paths": ["/app/*", "/offer/*", "/product/*"]
}
]
}
}
```
Replace `TEAMID` with your Apple Developer Team ID and `com.yourcompany.yourapp` with your bundle identifier. The `paths` array defines which URL patterns trigger app routing. Any URL path not listed opens in the browser.
#### 2) Configure Associated Domains in Xcode
In Xcode, open your app target, go to **Signing and Capabilities**, and add the Associated Domains capability. Add an entry in the format `applinks:yourdomain.com`. iOS fetches and caches your AASA file at install time using this entitlement. Without it, Universal Links will not work even if the file is correctly hosted.
#### 3) Handle the incoming URL in app code
Implement `application(_:continue:restorationHandler:)` in your AppDelegate and read `userActivity.webpageURL`. Parse the path components and route the user to the correct in-app screen.
Test only on physical devices using a real email tap. Pasting a Universal Link into the Safari address bar does not trigger app routing by design.
### 2\. Android: Setting Up App Links
#### 1) Create and host assetlinks.json
Host your Digital Asset Links file at the path below, served over HTTPS returning HTTP 200 with no redirects ([Android Developer documentation](https://developer.android.com/training/app-links/verify-android-applinks)):
```
https://yourdomain.com/.well-known/assetlinks.json
```
```json
[{
"relation": ["delegate_permission/common.handle_all_urls"],
"target": {
"namespace": "android_app",
"package_name": "com.yourcompany.yourapp",
"sha256_cert_fingerprints": ["YOUR_SHA256_FINGERPRINT"]
}
}]
```
Get your SHA-256 fingerprint with `keytool -list -v -keystore your.keystore`. Any redirect on this URL silently breaks domain verification.
#### 2) Add the intent filter to AndroidManifest.xml
```xml
```
The `android:autoVerify="true"` attribute tells Android to verify domain ownership at install time, completing verification in under **20 seconds**. Without it, tapping the link shows an "Open with" disambiguation dialog.
#### 3) Verify your configuration
Use the [Google Digital Asset Links testing tool](https://developers.google.com/digital-asset-links/tools/generator) to confirm your `assetlinks.json` is reachable, correctly formatted, and associated with the correct signing certificate before any send.
For a full technical reference covering both platforms, see [Airbridge's guide to understanding deeplinks](https://help.airbridge.io/en/deeplink-guides/understanding-deep-link), including URL scheme fallback options and platform-specific edge cases.
## Why Your Email Platform Breaks Deeplinks and How to Fix It
### 1\. How Click Tracking Wrappers Prevent Deeplinks from Working
Every major email platform wraps outgoing links through its own tracking domain to measure clicks. When you place `https://yourdomain.com/offer/123` in a Klaviyo email, the URL that actually goes out looks like this:
```
https://trk.yoursendingdomain.com/track/redirect?url=https%3A%2F%2Fyourdomain.com%2Foffer%2F123
```
The OS-level deeplink system only works when the URL the user taps matches a domain your app has claimed in its entitlements. **The OS sees the tracking domain, finds no app claiming it, and opens the browser.** Your AASA file and `assetlinks.json` are irrelevant here because they live on your domain, not the tracking domain.
| ESP | Default tracking domain | Deeplink support method | Source |
| --- | --- | --- | --- |
| Klaviyo | Klaviyo subdomain (trk.*) | Upload AASA and assetlinks.json via Klaviyo Settings > Push notifications > Universal & App Links | Klaviyo docs |
| Mailchimp Marketing | list-manage.com | Limited for standard campaigns; Mailchimp Transactional (Mandrill) supports full custom tracking domain | Mailchimp Transactional docs |
| HubSpot | HubSpot-controlled domain | Custom click-tracking domain available in Marketing Hub settings | HubSpot docs |
| SendGrid | SendGrid link branding domain | Branded link domain + host AASA and assetlinks.json on that subdomain; must manually flag links with universal="true" in HTML | SendGrid docs |
### 2\. Three Ways to Solve the ESP Link Wrapping Problem
**Option 1: Use a custom tracking subdomain.** Most platforms let you point their click-tracking wrapper to a subdomain you own, such as `email.yourdomain.com`. Host your AASA and `assetlinks.json` files on that subdomain. The OS recognizes your domain and routes the link to the app. This requires ongoing file maintenance whenever your app's bundle ID or signing certificate changes.
**Option 2: Disable link tracking for specific URLs.** Some campaigns can turn off click wrapping for individual links. You lose click-level analytics for those links, but the deeplink works. This trade-off is acceptable for transactional or re-engagement emails where downstream in-app events provide a more meaningful attribution signal.
**Option 3: Use a dedicated deep link management platform.** A purpose-built tool handles hosting your configuration files on a trusted tracking domain, managing custom subdomain routing, and connecting click data to downstream in-app events. For a detailed platform comparison of Universal Links and App Links behavior, see [Airbridge's cross-platform deep linking guide](https://www.airbridge.io/en/blog/universal-links-vs-app-links-cross-platform-guide).
### 3\. The Microsoft Outlook SafeLink Complication
Outlook adds a second layer: Microsoft Defender's Safe Links feature rewrites all URLs through `safelinks.protection.outlook.com` before delivery. This cannot be disabled from the sender side. Enterprise users on Outlook will not receive working deeplinks regardless of how correctly you have configured your own domain.
Design every deeplink fallback page with a visible "Open in app" button. iOS Smart App Banners surface this automatically for Safari users who land on the web fallback.
## Email Deeplink Best Practices for Re-engagement Campaigns
### 1\. Always Include a Working Web Fallback
Every deeplink URL must have a real destination page on the web. If the app is not installed, if the OS routes incorrectly, or if the email client breaks the URL, the user needs a page that makes sense. A 404 on the fallback means every non-app user hits a dead end.
### 2\. Use Deferred Deeplinks for Win-Back Emails
For subscribers who may have uninstalled your app, standard deeplinks are not enough. The data backs this up: **13.7% of monthly subscribers who lapse do reactivate within 12 months** (RevenueCat, State of Subscription Apps 2025). That is a meaningful revenue recovery pool — but only if the re-engagement experience lands them inside the app, not on a browser login screen.
A deferred deeplink gives lapsed users the best possible first experience after reinstalling: they land on the screen the email pointed to, not a generic home screen. Use deferred deeplinks for win-back campaigns, referral emails to non-app users, and any segment that has not opened the app in **30 or more days**.
### 3\. Test on Real Devices Before Every Send
iOS Simulator and Android Emulator do not replicate email client behavior accurately. Before any campaign send, test the link on physical devices in Apple Mail (iOS and iPadOS), Gmail (iOS and Android), and Outlook. Confirm the app opens on the correct screen and the fallback works when the app is uninstalled.
### 4\. Track Deeplink Performance Separately from Email Click Rate
Your email platform's click rate shows how many taps happened. It does not show how many taps resulted in an in-app session or a downstream conversion. Track app open source, target screen reached, and conversion events as separate funnel steps. Without this, you cannot distinguish between a deeplink configuration that works and one that quietly falls back to the browser on every send.
## Stop Losing Conversions to the Browser
Setting up email deeplinks is a one-time technical investment: two configuration files, an app entitlement, and a resolved tracking domain. Most apps complete this in a few days of engineering time.
The downstream gap most teams discover next is attribution. Your email platform shows click count. Your app shows sessions. Nothing connects those email clicks to subscription renewals. The routing works, but the revenue measurement does not.
[Airbridge's deep linking platform](https://www.airbridge.io/en/deeplink-plan) handles both the routing infrastructure (AASA hosting, custom tracking domains, deferred deeplinks) and the downstream measurement that connects the email click to the subscription event in your app.
---
## Deep Linking for Subscription Apps: Fitness, Health, Edtech
URL: https://www.airbridge.io/en/blog/deep-linking-subscription-apps-fitness-health-edtech
Language: English
Published: 2026-06-15
Author: Luke
Your Meta ad shows a specific HIIT workout. A user taps it, installs your fitness app, and opens it for the first time. They land on the home screen. They scroll around, don't see the workout, and leave. No trial started.
This is a deeplink routing problem. [**82% of free trial starts happen on Day 0**](https://www.revenuecat.com/state-of-subscription-apps-2025/), making the first session the most critical moment in your subscription funnel. Apps that route new users to the exact screen the ad promised convert measurably better than apps that drop everyone on the home screen. This guide covers how to fix that, with specific patterns for fitness, health, and edtech.
**Key Takeaways**
- **Day 0 decides whether the trial starts.** 82% of trial starts happen on first open. Context mismatch between the ad and the landing screen kills conversion before the user ever sees a paywall.
- **App links preserve the promise.** A deferred deeplink carries the intended destination across the install step, so the first app open routes a new user to the exact screen they clicked in the ad.
- **Fitness and health apps have a tight conversion window.** Health and fitness purchases cluster on Day 0 and Day 4-7 with limited activity in between. Getting routing right in the first session is essential.
- **Edtech apps lose subscribers at the content handoff.** A user who lands on the course they saw in the ad converts better than a user who must navigate to find it.
- **The right routing patterns differ by vertical.** Fitness needs workout routing. Edtech needs lesson and course routing. Both need deferred deeplinks for paid acquisition campaigns.
## Why the First Screen After Install Decides Your Trial Conversion Rate
### 1\. The Day 0 Conversion Window
The data on this is consistent across two large datasets.
**From** [**RevenueCat's analysis of75,000+ subscription apps**](https://www.revenuecat.com/state-of-subscription-apps-2025/)**:**
- **82%** of free trial starts happen on Day 0.
- Health and Fitness apps reach a **trial-to-paid median of 39.9%**, compared to the all-category median of 34.8%.
- The install-to-paid conversion rate for Health and Fitness sits at a median of **2.7%**. That rate is largely determined by the first session.
**From** [**Adapty's analysis of16,000+ apps**](https://adapty.io/state-of-in-app-subscriptions/)**:**
- **90%** of trial starts happen on Day 0.
- Health and Fitness purchases follow a bimodal pattern: conversions cluster on Day 0 or Day 4-7, with limited activity in the days between.
Two large datasets reach the same directional conclusion. RevenueCat reports **82%** of trial starts on Day 0 across **75,000+ apps**, while Adapty reports **90%** across **16,000+ apps**. The 8-point spread likely reflects different sample compositions — Adapty's dataset skews more toward subscription-focused apps — but both confirm that the majority of trial decisions happen on first open.
If a user does not engage on the first open, they rarely return to encounter a paywall again before churning.
### 2\. What Happens When the Landing Screen Does Not Match the Ad
When a user taps an ad for a specific yoga flow and opens to a home screen with six categories and no clear starting point, trust breaks at the moment it should be confirmed. The practical consequences:
- The user must spend effort finding what they came for.
- That effort creates doubt about whether the app is what they expected.
- Most users close the app rather than browse.
- The trial start rate from that campaign drops below its potential.
Apps that route new users to the correct content on first open remove this effort entirely. The user's first in-app experience directly confirms the ad's promise.
### 3\. What Deeplink Routing Means for Your App
Deeplink routing is the mapping between an external click (from an ad, email, push notification, or QR code) and a specific in-app destination. Three outcomes depending on your setup:
| Setup | What new users see on first open |
| --- | --- |
| No deeplinks | Home screen, regardless of which ad they clicked |
| Universal or App Links only | Correct screen for existing users; home screen for new installs |
| Deferred deeplinks | Correct screen for all users, including brand new installs |
The deferred deeplink is the pattern that solves the new user problem specifically.
## Three Types of Deeplinks Every Subscription App Team Should Know
Not every deeplink type works for every situation. Using the wrong type for a campaign targeting new users will either fail silently or show an error.
| Link type | App installed | App not installed | Primary use |
| --- | --- | --- | --- |
| URI scheme (myapp://) | Opens correctly | Fails silently or shows error | Internal app-to-app routing only |
| Universal Link / App Link | Opens correctly | Falls back to web | Emails, push, social, returning users |
| Deferred deeplink | Opens correctly | Stores context, routes after install | Paid UA campaigns, influencer links |
### 1\. URI Schemes: Internal Navigation Only
A URI scheme like `myapp://workout/hiit-30` is a custom protocol that works only if the app is already installed.
- **When the app is not installed:** The link fails silently on iOS or shows an error dialog on Android.
- **Never use URI schemes in paid campaigns** or any marketing link where new users may be clicking.
- **When to use it:** Internal app-to-app navigation only, like opening a specific screen from an in-app browser.
### 2\. Universal Links and App Links: The Standard for Marketing Links
Universal Links (iOS) and App Links (Android) are verified links. The operating system checks that you own both the domain and the app before allowing the seamless handoff.
**How it works:**
- A URL like `https://myapp.com/workout/hiit-30` opens in your app if installed.
- If the app is not installed, the URL falls back to the web version of that page.
**What you need to set up:**
- iOS: host `/.well-known/apple-app-site-association` on your domain ([Apple documentation](https://developer.apple.com/documentation/xcode/supporting-universal-links-in-your-app))
- Android: host `/.well-known/assetlinks.json` on your domain ([Android documentation](https://developer.android.com/training/app-links/about))
- Allow **24-48 hours** for OS verification to propagate after updating these files.
**When to use it:** Emails, push notifications, social links, and in-app sharing where users are existing customers or a web fallback is acceptable.
### 3\. Deferred Deeplinks: The Key Tool for New User Campaigns
A deferred deeplink is a link that stores the intended in-app destination and campaign metadata across the app store install step, so the first app open routes the new user to the correct screen.
Without deferred deeplinks, every new user from a paid campaign lands on the home screen regardless of the ad they clicked. With deferred deeplinks, new users land on the exact screen the ad promised. The diagram below shows the full flow.
[Screenshot 2026-06-15 at 15.04.08.png]
## Deeplink Patterns for Fitness and Health Apps
A solid deeplink strategy for fitness subscription app onboarding begins with one principle: every ad creative should map to a specific in-app screen before the campaign goes live.
### 1\. Ad-to-Content Routing: Match Every Creative to a Specific Screen
**How it works:** A Meta, TikTok, or YouTube ad features a specific workout, program, or fitness challenge. The deeplink carries that content ID through the install, so the user's first screen is the exact workout from the ad.
**Example deeplink payload:**
```
https://myapp.com/content?screen=workout&id=hiit-30&campaign=meta_lookalike&trial=true
```
**What happens on first open:**
1. App reads the `screen` and `id` parameters.
2. Navigates directly to the workout.
3. Optionally triggers a trial prompt in context: "Start your free **7-day trial** to unlock all **120 workouts**."
Before your campaign launches, map every creative to a route:
| Ad creative | Target screen | Required payload |
| --- | --- | --- |
| "30-Day HIIT Challenge" | Program intro | screen=program&id=hiit-30day |
| "Morning Yoga Flow" | Specific workout | screen=workout&id=yoga-morning |
| "Free Trial: All Programs" | Trial paywall | screen=paywall&offer=7day |
### 2\. Trial Onboarding: The Step-by-Step Deferred Deeplink Flow
For paid acquisition, a deferred deeplink automates the meaningful first-task delivery: instead of dropping the user on a blank dashboard, they land directly on the specific workout, lesson, or offer the ad promised.
**The complete onboarding trial flow:**
1. User sees an ad for a specific workout or program.
2. User taps the link. The deeplink service stores the destination and campaign metadata.
3. User is redirected to the App Store.
4. User downloads and installs.
5. First app open triggers context restoration from the deeplink service.
6. User lands directly on the workout or program intro screen.
7. Trial prompt appears in context, tied to the content they were shown.
This flow consistently outperforms routing to a generic onboarding questionnaire first, because the trial prompt appears after the user has already seen what they are paying for.
**Optional enhancement:** Some fitness apps pass the user's stated goal from the ad's UTM parameters into the onboarding form. This pre-fills "Your goal: Lose weight" without the user having to select it again, reducing friction before the paywall.
### 3\. Re-engagement via Push and Email: Routing Table
Lapsed trial users and inactive subscribers need links that go somewhere specific. A push notification saying "Your streak is waiting" must open the streak screen or the last workout, not the app root.
| Campaign type | Target screen | Payload example |
| --- | --- | --- |
| Trial expiry (Day 6 nudge) | Trial paywall with offer | screen=paywall&offer=7day&plan=annual |
| Streak recovery | Last completed workout | screen=workout&id={last_workout_id} |
| New program launch | Program intro | screen=program&id={program_id}&source=email |
| Win-back (30+ days lapsed) | Personalized content feed | screen=feed&segment=lapsed&source=push |
| Renewal reminder | Subscription management | screen=account&tab=subscription |
The principle throughout: match the behavioral message in the notification to the behavioral destination in the app.
## Deeplink Patterns for Edtech Apps
Edtech subscription apps have a different routing challenge. The content is a structured curriculum, not a standalone workout. Routing needs to preserve both the content context (which course) and the progression context (how far the user has gotten).
### 1\. Course and Lesson Routing from Paid Campaigns
**The problem:** An ad promising "Learn Python in 30 days" routes to the home screen. The user opens a course catalog, does not immediately see the Python course, and closes the app.
**The fix:** Include the course ID in the deeplink payload. Optionally include the first lesson ID so users can preview content before seeing the paywall.
**Example deeplink payload:**
```
https://myapp.com/course?id=python-30&lesson=1&campaign=google_search_01
```
For professional certifications and skills-based learning, include the credential or track ID. The landing screen should display the certification name and outcome, reinforcing the ad's promise.
### 2\. Live Classes and Cohorts: Time-Sensitive Routing
Live classes, cohort programs, and scheduled sessions need temporal context in the deeplink. A "Join live class at 7pm" email must route to the class entry screen, not the class catalog.
**Example deeplink payload with fallback:**
```
https://myapp.com/class?id=python-live-06&fallback=recording
```
If the session has ended, the app reads the `fallback` parameter and routes to the recording. Without this, users see an error screen or the home page.
Build fallback states for every time-sensitive content type:
- Live class ended: show recording
- Cohort enrollment closed: show waitlist or next cohort date
- Promotional offer expired: show standard paywall, not a broken screen
### 3\. Trial Expiry: Route Back to Progress, Not a Paywall
When a trial ends, the re-engagement email should not route to a generic paywall. It should route to where the user left off.
**The high-converting flow for edtech trial expiry:**
1. User receives "Your trial is ending" email.
2. Link opens the app at lesson 4 (the next step in their course).
3. A paywall overlay appears, anchored to the course the user was engaged with.
4. The annual plan is pre-selected with a message tied to the course: "Complete your Python certification."
Routing to the progression screen before the price appears works because the user sees their incomplete progress before making the payment decision. The decision shifts from "Do I want to subscribe?" to "Do I want to finish what I started?"
For certificate-track courses, pre-select the plan duration that matches the course length. A **6-month Python curriculum** should default to the annual plan, not monthly.
## How to Build Your Deeplink Map Before Running Paid UA
### Step 1: Map Every Target Screen and Its URL Parameters
Before any paid campaign, create a complete routing table. Every ad creative needs to map to a row in this table. If it does not, the campaign defaults to the home screen.
| Screen | Route | Required parameters | Optional parameters |
| --- | --- | --- | --- |
| Specific workout | /workout | id | source, trial |
| Program intro | /program | id | discount_code, trial_days |
| Trial paywall | /paywall | none | offer, plan, source |
| Personalized feed | /feed | none | segment, goal |
| Streak tracker | /streak | none | days |
| Subscription management | /account | none | tab=subscription |
| Course intro | /course | id | lesson, campaign |
| Live class entry | /class | id | fallback |
### Step 2: Set Up Deferred Deeplinks (6-Step Checklist)
1. **Configure verified domain files.** Add `apple-app-site-association` (iOS) and `assetlinks.json` (Android) to your `/.well-known/` path. The [Universal Links and App Links cross-platform guide](https://www.airbridge.io/en/blog/universal-links-vs-app-links-cross-platform-guide) covers this in detail.
2. **Integrate a deeplink management service.** This service handles deferred context storage and retrieves it on the user's first open after install.
3. **Add deeplink handling code in your launch lifecycle.** On first open, your app reads the context parameters and navigates to the correct screen before showing any generic onboarding.
4. **Map incoming parameters to navigation actions.** Each parameter combination corresponds to a specific screen. Build this as a routing config your team can update without shipping a new build.
5. **Define fallback behavior.** Every route needs a fallback for: expired content, missing IDs, unauthenticated users, and deleted content.
6. **Read the** [**deferred deeplink onboarding guide**](https://www.airbridge.io/en/blog/deferred-deeplink-for-onboarding) for step-by-step implementation detail.
### Step 3: Test Your Deeplink Flow Before Launching Any Campaign
- Fresh device with no app installed: click the campaign link, install from the App Store, verify the first screen is correct.
- Device with the app already installed: click the same link, verify it opens the correct screen directly.
- After an app update: re-verify both paths. Parameter handling frequently breaks between versions.
- From the ad platform's click tracker, not just a browser. The platform may add tracking parameters that affect your routing logic.
- With expired content IDs: verify the fallback screen appears correctly and not a blank or error screen.
### Step 4: Five Routing Mistakes That Kill Trial Starts
- **URI schemes in new user campaigns.** `myapp://workout/hiit-30` shows an error on any device without the app installed. Replace with Universal Links or a deeplink management service.
- **Routing to authenticated screens without a gate.** A deeplink pointing to a user's personal workout log fails for unauthenticated new users. Add a sign-up or trial-start gate before the routing destination fires.
- **Expired offer parameters without a fallback.** A deeplink carrying `offer=50off` after the promo ends either shows a broken screen or opens the app root. Add expiry handling in your routing logic.
- **Email platform parameter encoding.** Marketing email platforms (Klaviyo, Braze, Iterable) rewrite URLs for click tracking and often encode parameters incorrectly. Always test deeplinks from an actual email send, not a browser. Encoded characters like `%3D` break routing logic in apps that do not decode parameters before parsing.
- **Missing fallback for deleted or expired content.** A deleted workout or an ended cohort creates a broken destination. Add a default fallback route for any content `id` your server cannot resolve.
## Start Deep Linking Before Your Next Campaign
Trial conversion does not happen at the paywall. It happens in the moment when a new user's first in-app experience confirms what the ad promised. If that first screen is right, the trial starts. If it is the home screen, the moment is gone.
Onboarding setups that combine a trial with a structured onboarding flow achieve a [**1.78% median install-to-paid conversion rate**](https://adapty.io/state-of-in-app-subscriptions/) in Adapty's analysis of **16,000+ apps**, the highest of any onboarding format. Context-appropriate deeplink routing is one component of this pattern: it ensures the trial prompt and the user's actual intent converge on first open.
---
## How App Developers Find New Growth with Mobile OEM Advertising Across 86% of Android Devices
URL: https://www.airbridge.io/en/blog/new-growth-with-mobile-oem-advertising-across-android-devices
Language: English
Published: 2026-06-15
Author: Harper (Trang Nguyen)
Most UA conversations this year have been circling the same drain. Meta CPI is up. Google UAC is fine but flatlining. iOS signal is what it is post-ATT. Spend the same money in the same places, get a slightly worse return. Repeat next quarter.
This piece is for anyone running Android user acquisition. Optimization has done what it can. The bigger problem is that everyone's optimizing against the same Android audience pool, on the same platforms, with the same algorithmic constraints. There's a ceiling, and most teams running app UA at scale have hit it. It's not an argument that Meta and Google are over (they're not). **It's an argument that any growth team relying solely on the duopoly for Android UA is leaving a third channel untouched.**
That channel is mobile OEM advertising. Mobile OEM advertising reaches roughly 1.85 billion daily active users across 86% of global Android shipments. It runs on inventory the duopoly doesn't own (alternative app stores, on-device ad surfaces, and Dynamic Preloads on new devices) and integrates directly with Airbridge for measurement. Here's where it fits in a modern performance mix.
What's clear is that mobile OEMs aren't just a niche. Mobile OEMs are now part and parcel of the mainstream.
## **A quick primer on mobile OEM advertising**
**Mobile OEM advertising** is an Android-only channel. Apple's tightly controlled iOS ecosystem doesn't have an equivalent, which means OEM is one of the few places Android UA gets dedicated, scaled inventory that Meta and Google don't already monetise.
OEM stands for Original Equipment Manufacturer: Samsung, Xiaomi, HONOR, OPPO, vivo, Transsion (which owns Tecno, Infinix and itel), Huawei, and others. They sell the devices, but they also run their own software stacks on top of the hardware. That includes alternative app stores like Galaxy Store and GetApps, pre-installed native apps like App Vault, Mi Browser and Themes, and a lot of advertising surface area on the device itself.
Across the OEMs AVOW works with as official core agency for Xiaomi, vivo and OPPO, and as an early partner for Samsung and HONOR, the combined footprint covers around 86% of global Android shipments and 65% of all mobile devices in use, working out to the 1.85 billion daily active users referenced above.
Inventory comes in four shapes: placements inside the alternative app stores, on-device display ads (lock screen, app drawer, notification feed), Dynamic Preloads via Google Play Auto Install on first device boot, and branding placements like splash screens. The buying side will look familiar to anyone running performance: CPI, oCPC, ROAS. None of this competes with Meta or Google for the same impression.
Learn more about [OEM Advertising here](https://www.airbridge.io/en/glossary/oem-advertising).
## **Three places mobile OEM advertising earns its spot in the mix**
It would be wrong to frame mobile OEM advertising as a Meta or Google replacement. The right framing is complementary. Three specific gaps where OEM ads tend to do real work:
**1\. Android reach beyond Meta and Google.**
This is the use case that gets most large advertisers in the door. If you've been running Meta and Google in your core Android markets long enough, the addressable pool is already saturated. New spend buys overlapping users at higher prices.
OEM inventory sits outside that pool. A user browsing the Galaxy Store, looking through App Vault or unboxing a new device is somewhere social spend can't reach. Tripledot Studios is one example. The London-based gaming studio partnered with AVOW to run display advertising and Dynamic Preloads via Google PAI across Samsung, Xiaomi, OPPO and vivo, scaling first in EMEA before testing additional growth in LATAM and SEA. The campaigns came in 20% above their performance targets in EMEA and delivered 15% higher results than Tripledot's other preload channels, reaching audiences their existing Meta and Google spend couldn't.
**2\. Fraud-clean, performance-priced inventory.**
Two things make OEM unusually clean from a buying perspective. The impression and install both happen on hardware the manufacturer controls, which means there's no third-party network in the middle. MMPs like Airbridge add another layer of verification by checking traffic quality with device-level signals and fraud validation rules, helping advertisers identify suspicious installs, touchpoints, and conversion patterns.
Explore more about [Airbridge’s Fraud Validation Rules here](https://www.airbridge.io/en/product/fraud-protection).
The second is pricing. Most placements are CPI. On-device display can run on oCPC with a daily budget, where the OEM bids in real time on the user's likelihood to convert. Dynamic Preloads via PAI need no SDK integration, and from paperwork to live campaign is usually around two weeks. Channel test cost is genuinely low.
**3\. Geographic expansion outside the duopoly's home turf.**
This is where OEM stops being tactical and gets strategic. Through aggregated OEM partnerships, mobile OEM advertising reaches a substantial share of the Android base in every major region outside North America:
- Africa: 91%
- Southeast Asia: 87%
- India: 84%
- CIS: 75%
- MENA: 70%
- Europe: 57%
- LATAM: 55%
These reach figures cover AVOW's combined OEM partnerships in each region. You can [check OEM reach for your specific markets](https://avow.tech/user-reach-calculator/?utm_source=airbridge&utm_medium=referral&utm_campaign=partners2026) using the AVOW reach calculator for more granular planning.
For app companies expanding into any of these regions, OEM is often a more efficient first-wave channel than pumping more budget into Meta or Google in the same geos. Kredivo, an Indonesian buy-now-pay-later fintech, partnered with AVOW to run user acquisition across Xiaomi, OPPO and vivo with a customized commercial setup for each OEM. The campaigns delivered a 46% install-to-loan-application rate, 2.2x YoY install volume growth, 4.7x YoY growth in ad spend, and a 24% share of new product exploration via Dynamic Preloads within six months. OEM is now Kredivo's top performance channel outside the SRNs.
## **Where Airbridge fits**
The single most common objection on mobile OEM advertising is measurement. UA leads have built their dashboards around their MMP, and a channel that doesn't show up at the same fidelity is a channel they don't trust.
OEM ad inventory plugs straight into Airbridge, flowing installs and post-install events through the same attribution stack you’re already running. In practice, that means:
- **Unified cross-channel reporting.** CPI, ROAS, retention curves, and event funnels for OEM appear next to Meta, Google, and other paid channels in a single view. No separate dashboards to reconcile, no manual exports to run before your weekly review.
- **Device-level fraud verification.** Because both the impression and the install happen on hardware the manufacturer controls, Airbridge verifies OEM inventory at the device level. Reported install fraud is negligible compared to open programmatic, which makes OEM one of the cleaner buys in performance advertising.
- **Incrementality measurement without running experiments.** Airbridge’s observational incrementality lets you measure the real lift OEM delivers on top of your existing channel mix, without holding back spend or designing holdout tests. You see the true additive effect, not just attributed volume.
- **330+ ad network integrations already in place.** OEM surfaces connect through the same integration layer as every other network in your stack. There’s no new SDK implementation required, no new data pipeline to build before you can start measuring.
A couple of things to configure properly from the start:
- **Standardise event definitions across channels.** Cost-per-deposit or cost-per-registration comparisons only hold up if you’re tracking the same event with the same definition everywhere. Set this up before you start comparing OEM performance against social – mismatched event definitions are the most common reason OEM looks artificially cheap or expensive in cross-channel reviews.
- **Don’t rely solely on last-touch attribution.** OEM tends to sit earlier in the user journey than social does – particularly for Dynamic Preloads and app store placements on new devices. Last-touch models will systematically undercount OEM’s contribution. Multi-touch or incrementality-based measurement will give you a more accurate read of what the channel is actually delivering.
## **What's harder about mobile OEM advertising**
Mobile OEM advertising isn't a clean, single-vendor channel like Meta. The biggest practical challenge is fragmentation. Each OEM has its own ad formats, bidding logic and inventory mix. Galaxy Store featuring isn't the same product as Xiaomi App Vault, which isn't the same as vivo Apps-Selected. Running OEM properly at scale means working across thirteen-plus distinct surfaces. Most teams don't build that capability in-house, which is why aggregation partners exist.
The second challenge is that OEM strength varies sharply by region. Samsung leads in Europe and LATAM. Xiaomi leads in CIS. OPPO and vivo lead South and Southeast Asia. Transsion leads in Africa. An OEM strategy can't be a single global plan, it has to break down market by market.
## **The growth that's hiding in plain sight**
The performance ceiling most app marketers are running into isn't really a budget issue. It's a mix issue. After two years of optimising harder against the same Meta and Google inventory, the room left to find new efficiency on those platforms gets smaller every quarter.
The teams that break through aren't the ones running cleverer social campaigns. They're the ones who've added a third channel to their stack, on inventory their competitors haven't tested. Mobile OEM advertising is that channel. The audience is at scale, the buying models are familiar, Airbridge integration makes it measurable from the first install, and the activation lift is light enough that there's not much reason to put off testing it.
By default, mobile OEMs need to be part of every marketer's mix.
For a deeper walkthrough of each OEM placement type, geographic priorities and how to structure a first test, AVOW's [definitive guide to mobile OEM advertising](https://avow.tech/mobile-oem-advertising-guide/?utm_source=airbridge&utm_medium=referral&utm_campaign=partners2026) covers the channel end to end.
---
_About AVOW: AVOW is a global app growth company specialising in mobile OEM advertising. Recognised as Xiaomi's Core Agency of the Year three years running (2022, 2023, 2024), winner of vivo's Best Agency Partner award, OPPO's Strategic Agency Award, Huawei's Best Agency of the Year, and HONOR's Global Best Support Partner for 2025. AVOW works with 150+ advertisers across 13 OEM partnerships globally._
---
## Push Notification Deep Links: Fix the Wrong Screen Problem and Recover CTR
URL: https://www.airbridge.io/en/blog/push-notification-deep-links-fix-wrong-screen-ctr
Language: English
Published: 2026-06-10
Author: Luke
Your push notification gets tapped. The user lands on your home screen instead of the promoted content. They close the app.
**That single failed link costs more than one missed click.** Users who land on the wrong screen learn to distrust your pushes. CTR drops over time, opt-out rates rise, and your highest-intent lifecycle channel slowly degrades. The problem is rarely the copy or the timing. It is where users end up after they tap.
**Key Takeaways**
- **Most push notification CTR problems are deep link problems.** The message gets tapped. The destination breaks. Fix the link before you rewrite the copy.
- **Three deep link types exist, each with different failure modes.** URI schemes, Universal Links (iOS), and App Links (Android) break in different ways and require different fixes.
- **iOS cold-start is the most common cause of home-screen landings.** When your app is fully closed, the system may not pass the deep link URL to your handler before the navigation stack loads.
- **Android 12 changed how unverified links behave.** Links that worked on older Android may now open the browser with no error message and no obvious cause.
- **Fixing the destination is only half the job.** Knowing which push campaigns actually drive subscriptions, not just taps, tells you whether your lifecycle channel is working.
## Why Broken Deep Links Silently Kill Push Notification CTR
A **push notification deep link** is the URL or routing instruction embedded in a notification that sends the user to a specific screen inside your app on tap. When it breaks, the result is a **push notification not opening the correct app screen**. That is the root cause behind falling CTR.
### 1\. The 3-Second Moment You Are Losing to a Home Screen
A push notification has one job: move the user from a notification to a specific piece of content inside the app.
The message, timing, and personalization all build intent. The deep link converts that intent into action. When the deep link fails, you spend the entire cost of building intent for zero conversion:
1. The user taps the notification
2. The app opens to the home screen instead of the target content
3. The user has no context for what they were supposed to do
4. They exit without converting
After this happens two or three times, they stop opening your pushes entirely.
### 2\. What Falling Push Notification CTR Actually Means
Push notification CTR varies widely by category. According to [Pushwoosh's 2025 Benchmark Study](https://www.pushwoosh.com/blog/push-notification-benchmarks/) (600+ apps, Q4 2024 to Q2 2025):
| App Category | iOS CTR | Android CTR |
| --- | --- | --- |
| E-commerce & Retail | 3.05% | 3.78% |
| Fintech | 2.09% | 2.84% |
| Media & Entertainment | 0.83% | 1.69% |
| Hypercasual Games | 0.82% | 1.05% |
| Action Games | 0.46% | 0.82% |
The spread between the best and worst-performing categories is more than **8x**. The gap is not primarily explained by message quality. One consistent separator is reliable deep link destinations. Users quickly learn whether your notifications lead somewhere useful. When they do not, they stop tapping.
### 3\. The Revenue Cost of Every Wrong-Screen Landing
In a subscription app, push notifications serve two specific jobs:
1. **Moving free users toward conversion** (trial-end reminders, upgrade prompts, feature reveals). Sending conversion-oriented pushes right after users exit without converting is among the highest-ROI lifecycle moments for subscription apps (48 Laws of Subscription App Success, Law 36)
2. **Keeping subscribers engaged enough to renew** (content updates, progress nudges, win-back triggers)
Both depend on landing users in the right context at the right moment. A "trial ends in 24 hours" push that drops users on the home screen means they never see the renewal offer. These missed moments compound over weeks into higher churn and lower trial-to-paid conversion.
## The Three Types of Push Notification Deep Links Explained
Not all deep links work the same way. The type your app uses determines which failure modes are possible and which fix applies.
For a complete technical breakdown of each type at the URL routing level, [Airbridge's deep link guide](https://www.airbridge.io/en/blog/what-is-deep-link) covers URI schemes, Universal Links, and App Links in detail.
### 1\. URI Schemes (Quick to Build, Not for Production)
A URI scheme looks like `myapp://screen/product-id`.
**Properties:**
- Simple to configure, requires no domain verification
- Any app can register the same scheme (no ownership protection)
- Produces an error dialog on iOS if the app is not installed
- May trigger an ambiguous app-chooser on Android if another app shares the scheme
**When to use it:** Internal navigation only (widget to main app, app-to-app routing). URI schemes are not suitable for production push notification campaigns.
### 2\. Universal Links for iOS and App Links for Android (The Right Choice)
Both use verified HTTPS URLs, for example `https://myapp.com/offers/summer-sale`.
**Properties:**
- Domain ownership is verified by the OS before routing
- Opens the app directly if installed
- Falls back gracefully to your website if the app is not installed
- Works consistently across email, SMS, social, and push notifications
- Secure: only your app can claim links tied to your verified domain
**When to use it:** Every live app running push notification campaigns should use Universal Links or App Links.
### 3\. Deferred Deep Links (For Re-engagement After Reinstall)
A deferred deep link stores the intended destination before an install happens.
**Properties:**
- User taps a campaign link, installs the app, lands on the correct screen on first open
- Survives the install process without losing context
- Requires attribution infrastructure to store and replay the destination
- Used for re-engagement campaigns targeting lapsed users who may have deleted the app
**When to use it:** Re-install campaigns where you need to restore context for returning users.
[Screenshot 2026-06-10 at 19.48.19.png]
## Why Push Notifications Keep Opening the Wrong Screen
Three root causes account for the vast majority of deep link routing failures. Identifying which one applies to your app determines which fix to apply first.
| Root Cause | Platform | Symptom | Fix |
| --- | --- | --- | --- |
| Cold-start routing | iOS | Home screen on closed app | Handle getInitialURL before navigator loads |
| Unverified App Link | Android 12+ | Link opens browser | Add and verify assetlinks.json |
| Incomplete state handling | iOS + Android | Works sometimes, fails others | Code handlers for all 3 app states |
| URI scheme conflict | Both | Wrong app opens | Migrate to Universal Links / App Links |
### 1\. The iOS Cold-Start Deep Link Bug
**What happens:**
When a user taps a push notification on a fully closed iOS app, the app launches from scratch. The navigation stack has not initialized yet when the deep link URL arrives. If your handler fires before the navigator is ready, the call is lost and the app lands on its default screen.
**The fix:**
Queue the deep link URL on receipt and process it only after your navigation stack is ready:
- **React Native:** Configure `Linking.getInitialURL()` to capture the deep link URL on cold start.
- **React Navigation:** Set up `linking.getInitialURL` in `NavigationContainer`'s `linking` prop to return the notification URL before the navigator renders ([React Navigation, NavigationContainer documentation](https://reactnavigation.org/docs/navigation-container/)).
**For native iOS**, implement `UNUserNotificationCenterDelegate.userNotificationCenter(_:didReceive:withCompletionHandler:)` as the primary handler. For cold-start specifically (where the delegate may not fire before app launch completes), also check `launchOptions[UIApplicationLaunchOptionsRemoteNotificationKey]` in `application(_:didFinishLaunchingWithOptions:)` to capture the payload before any view controller loads.
### 2\. Android 12 App Link Verification: What Changed
**What happened:**
Starting with Android 12, Google changed how unverified HTTPS links behave. A link without a verified App Link configuration now always opens in the user's default browser instead of the app ([Android Developer Documentation](https://developer.android.com/about/versions/12/behavior-changes-12)).
**How to check if this is your issue:**
Run this on a connected device:
```
adb shell pm get-app-links --user cur com.yourpackage
```
If the verification status shows `none` or `legacy_failure`, App Link verification is the cause.
**The fix:**
- Add `android:autoVerify="true"` to the intent filter in `AndroidManifest.xml`
- Host `assetlinks.json` at `https://yourdomain.com/.well-known/assetlinks.json`
- Register your app's package name and SHA-256 signing certificate in the file
### 3\. Notification Handlers That Only Cover One App State
**The problem:**
Both iOS and Android require explicit code to intercept a push tap and route to a specific screen. The push payload carries a URL or custom data field. Your handler must read it, parse the destination, and trigger navigation.
**The three app states you must handle:**
- **Foreground:** App is open and active when the notification arrives, user taps the banner
- **Background:** App is in memory but not active, user taps from the lock screen or notification center
- **Cold start:** App was fully closed, tap launches the app from scratch
Most home-screen landing complaints come from cold-start taps. The app opens, the default screen renders, and the navigation intent from the push is never processed because the handler only covers foreground or background states.
One distinction saves debugging time: a push tap always comes from a user who has the app installed, so a home-screen landing after a push is a configuration problem in your handler. A home-screen landing after a **paid ad click and a fresh install** is a different problem, a missing deferred deep link. That case is covered in [how deep linking fixes the click-to-install conversion gap](https://www.airbridge.io/en/blog/deep-linking-app-store-conversion).
[Screenshot 2026-06-10 at 19.49.04.png]
## How to Fix Push Notification Deep Links: A 5-Step Guide
### Step 1: Move Away from URI Schemes
If your app uses `myapp://` URI schemes for push campaigns, migrate first before fixing anything else.
URI schemes were never designed for production marketing use. The failure modes (no fallback on missing app, no domain verification, browser errors on iOS) will keep breaking campaigns regardless of how well the rest of your setup is configured.
**What to do:**
- Audit every push template for `scheme://` links
- Replace with HTTPS Universal Links (iOS) or App Links (Android)
- Update your push notification tool or CMS to use the new URL format
### Step 2: Set Up Domain Verification Files
**For iOS (Universal Links):**
1. Enable Associated Domains in your app target settings
2. Add `applinks:yourdomain.com` to your entitlements file
3. Host an AASA file at `https://yourdomain.com/.well-known/apple-app-site-association`
4. Serve it as `Content-Type: application/json` with no redirect on that path ([Apple Developer Documentation](https://developer.apple.com/documentation/xcode/supporting-associated-domains))
**Note:** Apple's CDN caches the AASA file aggressively. After deploying changes, allow up to 24 hours before testing on a new device. If testing repeatedly on the same device, clear the iCloud / Apple ID cache between tests.
**For Android (App Links):**
1. Add `android:autoVerify="true"` to your intent filter in `AndroidManifest.xml`
2. Host `assetlinks.json` at `https://yourdomain.com/.well-known/assetlinks.json`
3. Include your app's package name and SHA-256 signing certificate in the file
**Note:** Android 15+ verifies App Links periodically in the background. After deploying changes, allow up to 7 days for full re-verification across devices in the field.
### Step 3: Write Handlers for All Three App States
Your notification handler needs three separate code paths, one for each app state.
**Foreground (app is open):**
Navigate immediately when the tap event fires. Most frameworks provide a direct listener for this state.
**Background (app is in memory, not active):**
Check for a pending notification intent on app resume. The payload is typically available in the intent or notification object when the app comes to the foreground.
**Cold start (app was fully closed):**
For React Native, use `Linking.getInitialURL()` to capture the deep link URL before any screen renders. For React Navigation, configure `linking.getInitialURL` in `NavigationContainer`'s `linking` prop to return the notification URL on launch.
For native iOS, implement `UNUserNotificationCenterDelegate.userNotificationCenter(_:didReceive:withCompletionHandler:)` as the primary handler. For cold-start specifically, also check `launchOptions[UIApplicationLaunchOptionsRemoteNotificationKey]` in `application(_:didFinishLaunchingWithOptions:)` to capture the payload before any view controller loads.
### Step 4: Test on a Physical Device Before Every Campaign
Universal Links and App Links do not behave reliably in simulators or emulators. Domain verification is a system-level OS check. Simulators bypass it.
**Pre-campaign device checklist (run on real iOS and Android):**
1. Kill the app completely from the task switcher
2. Tap the push notification
3. Confirm the correct screen loads (cold-start test)
4. Re-open the notification while the app is backgrounded
5. Confirm the correct screen loads (background test)
### Step 5: Add Fallback Logic for Expired or Removed Destinations
Your handler should never strand a user with a blank or error screen. Destinations expire and content gets removed.
**Build a fallback hierarchy:**
1. Check if the target screen exists
2. If not, try the nearest parent screen (e.g., the product category if the product was removed)
3. If that fails, route to a relevant section (search, home tab, or a related content screen)
A user who lands on a blank state from a push notification associates that experience with your notifications as a whole. A graceful fallback preserves trust even when the specific content is no longer available.
## How to Confirm Your Push Deep Link Fix Is Working
### 1\. Track Click-to-Screen Rate Separately from CTR
CTR measures whether users tap your notification. Click-to-screen rate measures whether users actually reach the intended destination.
**The gap between these two numbers is your deep link failure rate.**
A campaign can show a 4% CTR alongside a 2% destination-reach rate, meaning half of every user who tapped never saw the content you promoted. You will never catch this by watching CTR alone.
**How to instrument it:**
- Fire a screen-load event for the target screen within 5 seconds of a notification tap
- Compare the destination-screen event rate to the notification-tap event rate
- A ratio below 0.85 indicates a routing failure somewhere in the stack
### 2\. Run a Quick Destination Audit Before Each Campaign
Deep links break silently after routine code changes. Common causes:
- A navigation route is renamed in a product update
- A content ID or product slug changes
- A promotional screen is removed after its campaign ends
**Pre-campaign audit (takes 2 minutes):**
1. Copy the deep link URL from your push template
2. Open a browser on your device and navigate to the URL
3. Confirm the correct app screen loads
4. Repeat on both iOS and Android
### 3\. Connect Push Taps to Subscription Revenue
Once your deep links reliably route to the right screen, the next question is whether those screens convert.
A push notification that moves a free user to a trial-end screen, feature gate, or renewal prompt is part of your subscription funnel. Knowing which of those pushes resulted in an actual trial start or subscription renewal, not just a tap, tells you which lifecycle moments your push channel is winning.
Every push tap that converts to a subscription renewal contributes to [subscriber lifetime value (LTV)](https://www.airbridge.io/en/glossary/lifetime-value-ltv). Knowing which pushes are moving that metric separates a lifecycle channel that pays from one that just fills notification slots.
If you are running paid campaigns alongside lifecycle push and want to see which source drives paying subscribers, Airbridge Core Plan connects ad spend, push engagement, and subscription events in one view. It tracks the install-to-trial-to-subscription funnel across paid channels and lifecycle triggers, so you can see whether your push re-engagement or your paid retargeting is driving renewals.
## Frequently Asked Questions
### Why does my push notification open the app but land on the home screen?
In most cases the app was fully closed, and the deep link URL arrived before the navigation stack was ready, so the handler dropped it. Capture the initial URL on cold start and process it after navigation is initialized. On Android 12 and later, also confirm App Link verification, since unverified HTTPS links open the browser instead.
### Is a home-screen landing after a push a configuration problem or a missing capability?
For push notifications it is a configuration problem. The user already has the app installed, so the fix is in your notification handler and link verification. Missing capability applies to paid ad links that lead to a fresh install, where Universal Links and App Links cannot carry the destination through the app store and deferred deep linking is required.
### Do push notifications need deferred deep links?
Only for win-back campaigns that reach users who deleted the app. For everyone else the app is installed, so Universal Links on iOS and App Links on Android, with handlers for the foreground, background, and cold-start states, are enough.
## Fix the Link First, Then Optimize the Message
A push notification is one of the highest-intent moments in your user's relationship with your app. A user who opted in, read your message, and tapped was ready to act. A broken deep link is the only thing standing between that intent and a conversion.
Fix the destination before you rewrite the copy. Test on a physical device before every campaign goes live. Instrument click-to-screen rate so you know your fix holds after every code update.
---
## App Clips vs Full App Install: When to Use Each One
URL: https://www.airbridge.io/en/blog/app-clips-vs-full-app-install
Language: English
Published: 2026-06-10
Author: Luke
**77.9% of smartphone users say they abandoned at least one transaction because they were required to install an unfamiliar app** ([Heady.io Market Study, 2021](https://www.heady.io/blog/market-study-mobile-customer-experience-issues-highlight-use-cases-for-ios-app-clips) — the most-cited benchmark for install friction; no comparable updated study has been published since). Among users aged 18 to 34, that number climbs to over 80%.
Apple's App Clips were designed to solve exactly this problem. They are a different tool for a different job. The real question is not which is better overall. The question is which approach fits your use case, your users, and your growth stage.
**Key Takeaways**
- **App Clips are lightweight iOS-only experiences** (up to **100 MB** on iOS 17+ for digital-only invocations) triggered by QR codes, NFC tags, or links. Users complete one task without a full install.
- **The full app install funnel wins on subscription conversion, retention, and lifecycle marketing.** If your model depends on recurring revenue, the full app is the primary growth engine.
- **App Clips are best for high-intent, one-time physical transactions** such as restaurant ordering, parking payment, or event check-in, where users need to act now but have no prior brand relationship.
- **Attribution breaks down inside App Clips.** IDFA is unavailable and SKAdNetwork signals differ from standard install flows. Plan your measurement approach before launching at scale.
- **A hybrid strategy is the strongest play for most apps.** Use an App Clip to remove the install barrier at first touch, then earn the full install from a position of demonstrated value.
## What Is an App Clip? A Quick Technical Overview
### 1\. How App Clips are triggered and delivered
An App Clip is a lightweight version of your iOS app built for a single, focused task. It is not a standalone app. It lives as a separate build target inside your existing Xcode project, sharing code and assets with the full app.
**Size limits by iOS version and trigger type:**
| Trigger Type | Max Size | iOS Requirement |
| --- | --- | --- |
| Physical: QR code, NFC, App Clip code | 15 MB | iOS 16 or later |
| Digital: Safari, iMessage, Spotlight (digital-only, requires reliable connection) | 100 MB | iOS 17 or later |
Source: [Apple Developer Documentation](https://developer.apple.com/help/app-store-connect/reference/maximum-build-file-sizes/)
> **Important:** App Clips require iOS 14 or later and have no Android equivalent.
**The five ways an App Clip can be triggered:**
1. QR codes or App Clip codes at a physical location
2. NFC tags on products or surfaces
3. Safari Smart App Banners on a website
4. Links shared via iMessage
5. Place cards in Apple Maps
### 2\. What App Clips support (and what they don't)
Not all app functionality is available inside a clip. Understanding the boundaries upfront saves engineering time.
**App Clips can:**
- Process payments via Apple Pay
- Sign users in with Sign in with Apple
- Access device location
- Write data to a shared App Clip group container, which the full app can read after install
**App Clips cannot:**
- Send push notifications without an explicit opt-in during the active clip session
- Run background processes
- Access most Keychain data from the full app prior to install
- Persist on device: iOS automatically deletes App Clips after **30 days** of inactivity
### 3\. How users move from the clip into the full app
When a user finishes a task inside an App Clip, iOS shows a native prompt to install the full app. Data written to the shared App Clip group container carries forward, so the user resumes exactly where they left off without starting from scratch.
Users who complete a clip task are a warmer audience for that install prompt than cold App Store visitors, because they have already experienced the product before being asked to commit to it.
[Screenshot 2026-06-10 at 19.44.18.png]
## App Clips vs Full App Install: Full Comparison
| Factor | App Clip | Full App Install |
| --- | --- | --- |
| Best for | One-time physical transaction, feature preview | Subscription, lifecycle, habit-forming engagement |
| Platform | iOS 14 or later only | iOS and Android |
| Conversion strength | High for immediate task completion | High when onboarding and paywall are optimized |
| Retention | None: auto-deleted after 30 days of inactivity | Full lifecycle marketing, push, re-engagement |
| Engineering scope | Xcode target, 15 to 100 MB ceiling | Standard iOS development |
| Attribution | IDFA unavailable; SKAdNetwork inactive inside clip | Standard iOS attribution flow |
| Lifecycle marketing | Not possible after clip session ends | Full push, email, and in-app messaging available |
## When App Clips Win: 2 Core Use Cases
### 1\. One-time, high-intent physical transactions
App Clips are purpose-built for moments where a user needs to complete one action immediately, with no prior relationship with your app.
**Strong use cases:**
- Paying for parking by scanning a QR code on the meter
- Ordering at a restaurant table without needing a loyalty account
- Checking in at a venue, hotel, or event
- Renting a shared vehicle, bike, or device
At these touchpoints, requiring a full install is a conversion blocker. The user needs to act now, not browse the App Store. An App Clip removes that friction entirely.
A customer who ordered through your App Clip is already a motivated prospect for the loyalty program and repeat engagement that the full app enables.
### 2\. Capturing customers who skip the install step
77.9% of users abandoned a transaction because an install was required (Heady.io Market Study, 2021). Among 18 to 34 year olds, that climbs to over 80%.
Physical retail, hospitality, and packaged goods brands have historically been unable to close this gap:
- **Mobile web** offers no native push notifications, no home-screen presence, and no lifecycle marketing channel, even when Apple Pay is available at checkout
- **App Store redirect** loses the majority of users at the friction point
- **App Clips** deliver a native, Apple Pay-ready experience without requiring an install, with a clear path to a full-app relationship after the transaction
## When Full App Install Outperforms App Clips: 3 Scenarios
### 1\. Subscription apps that need recurring sessions
Subscription revenue depends on retention. Retention depends on the type of engagement only a fully installed app can sustain.
**What a full install enables that an App Clip cannot:**
- Push notifications for re-engagement and habit reinforcement
- Home screen presence as a daily habit trigger
- Widgets and Live Activities for passive visibility
- Session-level personalization that builds over time
Among more than 75,000 subscription apps analyzed, the median 12-month retention rate for annual subscribers is **44.1%** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps/)). That retention is built through repeated, habitual use across dozens of sessions. An App Clip session cannot build that habit.
If your business model requires users to return 20, 30, or 50 times per year, the full install is the only vehicle for that relationship.
### 2\. Apps where paywall conversion happens in the first session
**82% of trial starts happen on Day 0** (RevenueCat State of Subscription Apps 2025). The very first session determines most subscription conversion outcomes.
Top-performing subscription apps drive that conversion through:
- Multi-screen, goal-driven personalization before the paywall
- Micro-commitment steps that build emotional investment early
- Paywall timing, copy, and social proof optimization
- Hard or soft paywall placement at peak intent moments
A **15 MB** App Clip cannot support that depth of experience. The first-session onboarding that produces a trial start requires the full app environment.
### 3\. Apps that rely on lifecycle marketing to retain and win back subscribers
Push notifications, in-app messages, triggered emails, and lifecycle offers all require the app to be installed. An App Clip session that does not lead to a full install leaves no durable marketing channel open.
A concrete data point: among monthly subscribers who cancel, **13.7% reactivate within their first year** (RevenueCat State of Subscription Apps 2025). That reactivation only happens if the app is installed and a re-engagement mechanism exists.
## Should You Build an App Clip? 5 Questions to Answer First
Deciding when to use app clips instead of focusing on full app installs comes down to five checks. Work through each one before committing engineering time.
**Question 1: Is the entry point physical?**
App Clips perform best when triggered by a real-world object: a QR code on a table, an NFC tag on packaging, a place card in Apple Maps.
- If your acquisition runs primarily through paid social, organic search, or email, the App Store is the standard endpoint
- Adding an App Clip to a digital-first funnel adds build complexity without a clear conversion benefit
**Question 2: Is your core value deliverable in one session and within size limits?**
App Clips are designed for single-task flows: pay, scan, check in. If delivering real value requires any of the following, a clip cannot do that job:
- Account history or data from prior sessions
- Multi-step personalization before value is clear
- Heavy assets, frameworks, or media beyond the **15 MB** physical limit
**Question 3: Are your users primarily on iOS?**
App Clips are iOS-only. Before allocating engineering resources to an iOS-exclusive feature, verify that iOS represents the majority of your audience and revenue. Check your own analytics, not industry averages.
**Question 4: Can you absorb the attribution gap?**
Inside an App Clip, IDFA is unavailable and SKAdNetwork calls return empty values or do not fire (Apple Developer Documentation). The downstream full-app install can be attributed through deferred deep linking, but the clip session itself resists standard precision measurement.
If your team makes budget decisions based on channel-level LTV data, build the measurement plan before launching App Clips at scale, not after.
**Question 5: Do you have capacity to maintain two surfaces?**
An App Clip is a separate build target with its own requirements:
- Must stay within size limits (**15 MB** for physical invocations, **100 MB** for digital)
- Requires its own App Store review and approval
- Needs updates whenever relevant main app functionality changes
For teams of **1 to 5 engineers**, that is a real and ongoing maintenance cost. The tradeoff is worth it at high transaction volume. It is rarely worth it for an early-stage app still finding product-market fit.
## App Clip or Full App: Make the Right Call
App Clips solve a specific problem: reducing the friction cost of the install barrier at a moment of high physical intent. They are not a shortcut to a better subscription product, and they are not a replacement for the onboarding funnel that drives trial starts and conversions.
**The decision simplifies to two scenarios:**
- **Build an App Clip if:** you have physical touchpoints, an iOS-first audience, and a single-task flow that benefits from zero-install access
- **Prioritize the full install funnel if:** your app depends on subscription revenue, session-by-session personalization, or lifecycle marketing to retain users
The hybrid approach works best for businesses that operate at both ends. Use the App Clip to capture the first transaction and earn trust. Use the full app to build the recurring revenue relationship that follows.
---
## Deferred Deeplink for Onboarding: How to Close the Install-to-First-Action Gap
URL: https://www.airbridge.io/en/blog/deferred-deeplink-for-onboarding
Language: English
Published: 2026-06-10
Author: Harper (Trang Nguyen)
You spent real money getting someone to tap your ad. The ad promised a specific thing: a workout plan, a discounted annual plan, a personalized result. They tap, they hit the App Store, they install, they open the app. And they land on a generic home screen that has nothing to do with what they were promised.
That blank, context-free first screen is where most paid installs quietly die. The user can't find the thing they came for, momentum evaporates, and within seconds they close the app. You paid for the install. You did not pay for a confused user who never reaches the screen you advertised. A **deferred deeplink** is the mechanism that closes this gap, and getting it right is one of the highest-leverage onboarding fixes available to an early-stage subscription app.
**Key Takeaways**
- **The install step erases context by default.** A normal link can't survive the trip through the App Store or Google Play, so the destination and campaign intent you set up are lost the moment the user installs.
- **A deferred deeplink restores that context after install.** It holds the intended in-app destination and campaign metadata, then delivers the user there on first open instead of dropping them on a home screen.
- **First impressions are decided fast.** More than **82%** of trial starts happen on Day 0 (RevenueCat State of Subscription Apps 2025), so the first session is where your funnel is won or lost.
- **Personalized first sessions retain better.** Apps that land users on the promised content instead of a generic home screen report measurable gains in post-install conversion and retention (AppsFlyer and Branch case studies).
- **You can't optimize what you can't see.** Once the link is in place, the next step is knowing which channels and creatives actually drive users who reach the promised screen and subscribe.
## Why the Install-to-First-Action Gap Happens
Most founders assume that if a user installs, the hard part is over. The opposite is often true. The single riskiest moment in the funnel sits between the ad tap and the first meaningful in-app action, and two things conspire to break it.
### 1\. The app store breaks the link
When a user already has your app, a standard deep link can open it straight to the right screen. But when they don't have the app yet, the flow has to detour through the App Store or Google Play to install first. That detour is where context dies. The store has no idea what screen the user was promised, what offer they clicked, or which campaign sent them. By the time the app opens for the first time, the original link and everything attached to it are gone.
So the user who tapped "Start your **30-day** strength plan" doesn't arrive at a strength plan. They arrive at your sign-up wall or your default dashboard, and they have to go find what they came for. Every extra step they have to take to get there is another chance to drop off between the store and first action.
### 2\. Generic home screens kill momentum
The reason this matters so much is timing.
**The first session is not just one session among many. It is the session.** [More than **82%** of trial starts happen on Day 0, and roughly **44.5%** of purchases land on Day 0](https://www.revenuecat.com/state-of-subscription-apps/) as well. If the first thing a user sees doesn't match the promise that got them to install, you have spent your most valuable attention window on a mismatch.
This also shows up later as churn. The most common Google Play cancellation reason is "not enough usage," accounting for **37.2%** of stated reasons _(RevenueCat State of Subscription Apps 2026)_. Users who never connect their original intent to a concrete in-app action rarely build the habit that justifies a renewal, which drags down the lifetime value of every install you paid for. A generic home screen is the first domino in that chain.
## What a Deferred Deeplink Actually Does
A deferred link solves one specific problem: it carries the destination and context across the install step so the first open lands the user exactly where they expected to be.
### 1\. Standard deep link vs deferred link
A standard deep link assumes the app is already installed. Tap it, and the app opens to the target screen. Useful for re-engaging existing users, useless for new installs because there is no app to open yet.
A deferred link is built for the new-install case. It "defers" the routing until after the app is installed and opened for the first time, then sends the user to the intended screen. The practical difference for a marketer running acquisition ads is simple: the standard link works for people who already have you, and the deferred version works for the people you are paying to acquire.
### 2\. The flow: click, store, install, restore
Under the hood, the deferred flow works in four steps:
1. **Encode the intent.** When you create the link, you attach the target in-app destination plus metadata such as the campaign, the offer, or a referral ID.
2. **Capture on click.** When a user without the app taps the link, the deep linking platform records those parameters before handing the user off to the store.
3. **Install and open.** The user downloads and opens the app for the first time, as normal.
4. **Restore and route.** On that first open, the platform matches the user back to the captured parameters and routes them to the promised screen with the original context intact.
[diagram-deferred-deeplink-onboarding-flow.webp]
The payoff is that the shopper who tapped a "40% off annual plan" ad opens straight onto that discounted plan, not a generic welcome screen. The promise and the experience finally match.
## How to Use Deferred Deeplinks to Optimize Your Onboarding Funnel
Installing the technology is not the same as using it well. Here is a practical sequence for turning deferred deeplink onboarding into measurable funnel gains.
### 1\. Map every ad promise to a specific destination
Start with your ads, not your app. List every campaign and creative you are running, and write down the exact promise each one makes. A weight-loss creative promises a meal plan. A discount creative promises a price. A feature creative promises a tool. Each of those promises should map to a specific in-app destination, not your default home screen.
The rule is direct: **if an ad makes a promise, the link should land the user on the screen that delivers it.** If you can't name the destination screen for a given ad, that ad has nowhere to send people, and the gap is built in from the start.
### 2\. Carry campaign context into the first session
Beyond the destination screen, decide what context to pass through the link. The channel the user came from, the offer they clicked, and any referral information can all travel with a [deferred link](https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link). That context lets you shape the first session instead of treating every new user as a blank slate.
For example, a user from an _"annual plan, 40% off"_ creative can open directly onto a paywall that already shows the discounted annual plan, rather than the standard pricing screen where they have to hunt for the offer they were promised. The fewer decisions you force in the first **60 seconds**, the more users reach a meaningful action.
### 3\. Personalize onboarding with the intent you carried over
Onboarding converts best when it feels like it was built for the specific person in front of it. The carried-over context gives you the raw material to do that without asking the user a single question.
Some practical moves:
1. **Pre-fill the goal.** If the ad was about muscle gain, set the onboarding goal to muscle gain and reuse that exact phrase on later screens and the paywall.
2. **Match the social proof.** Show testimonials from users with the same goal the ad targeted, so the proof feels relevant rather than generic.
3. **Skip the redundant questions.** If the channel already told you what the user wants, don't make them re-enter it. Every redundant step is a drop-off point.
This connects to a broader onboarding principle: small interactive steps and continuous personalization before the paywall increase emotional investment and reduce drop-off. Restored context is what lets that personalization start at second zero instead of after a five-screen questionnaire.
### 4\. Shorten the path to the 'aha moment'
The destination screen should be the shortest possible distance to the action that makes the app valuable. If the promised screen still sits behind a forced account creation wall, a permissions prompt, and three tutorial slides, you have reintroduced the friction you were trying to remove.
Audit the path from first open to first real action. Move account creation to the moment intent is highest rather than the very first screen, offer SSO instead of email-only signup, and make any permission prompt explain a concrete benefit before the system dialog appears. The link gets the user to the right room. Your job is to make sure the room isn't locked.
## Measure What the Link Actually Changed
Here is the part most teams skip. You can land users on the perfect screen, but if you can't see which campaigns and creatives produce users who actually reach that screen and subscribe, you are still optimizing in the dark.
This is where the funnel work and the [measurement work](https://www.airbridge.io/en/blog/deep-linking-app-store-conversion) meet. To know whether your routing strategy is paying off, you need to connect the channel and creative a user came from to what they did after install: did they reach the promised screen, start a trial, and convert to paid? Averages across all installs won't tell you. You need the funnel broken down by where users came from.
Concretely, that means tracking the install-to-first-action and install-to-subscription funnel by channel and creative, so you can pour budget into the ads that send users who convert and cut the ones that send users who bounce. That is also how you measure true [lifetime value](https://www.airbridge.io/en/glossary/lifetime-value-ltv) per channel rather than judging campaigns on installs alone. **If you're running paid acquisition and want to see which channels and creatives actually drive subscriptions, not just installs,** [**start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core)**, which includes 15,000 attributed installs.** It surfaces the subscription funnel by channel so you can tell which deferred deeplink campaigns make money and which just spend it.
## Common Mistakes That Undo the Gains
Even with the right routing in place, a few recurring mistakes quietly cancel out the benefit:
1. **Sending everyone to the same screen anyway.** If every campaign points at your home screen "for simplicity," you have the technology but none of the value. Map distinct destinations per promise.
2. **Forgetting the no-app-yet case in testing.** Teams test deep links on devices that already have the app installed, see it work, and ship. The deferred path only matters for users without the app. Test on a clean device every time.
3. **Reintroducing friction after the landing.** A perfect destination screen behind a forced signup wall defeats the purpose. Keep the path to the first action short.
4. **Not measuring by channel.** Without channel-level visibility into who reaches the promised screen and subscribes, you can't tell which campaigns to scale.
## Close the Gap Between What You Promised and What They See
Every paid install is a promise. A deferred link is how you keep it: the destination and context survive the install, and the user opens onto the exact screen that made them tap in the first place. Pair that with onboarding that uses the carried-over intent, and you stop paying for installs that go nowhere.
The apps that win the first session are the ones where the ad, the install, and the first screen all tell the same story. Start by mapping one ad to one destination, test it on a clean device, and watch what happens to your install-to-first-action rate.
---
## Deferred Deeplink Implementation Guide for iOS & Android
URL: https://www.airbridge.io/en/blog/deferred-deeplink-implementation-guide-for-ios-android
Language: English
Published: 2026-06-09
Author: Harper (Trang Nguyen)
A deferred deeplink should send a brand-new user to a specific in-app screen even when they did not have your app when they tapped the link. The concept is simple, but the implementation is where most teams get stuck: the official docs are split across Apple, Google, your link provider, and your own SDK, and none of them tell you which method to call in what order. If you have been searching how to implement deferred deeplink iOS Android SDK calls and found the docs scattered, this guide puts the pieces in one place.
This walkthrough covers how deferred routing actually works, the exact iOS and Android setup steps, the SDK callbacks you need to handle, the most common mistakes that silently break the flow, and how to test it before you ship. The cost of getting this wrong is concrete: when the August 2025 shutdown of Firebase Dynamic Links forced millions of apps to migrate, the teams that rebuilt their routing correctly kept their post-install conversion intact, and the teams that did not lost every new user to a generic home screen.
**Key Takeaways**
- **Deferred routing has two distinct halves.** A pre-install half that stores the destination, and a post-install half where your SDK reads it back on first launch and routes the user. You must implement both.
- **iOS and Android solve the match problem differently.** Both prefer deterministic matching, but Android can fall back to probabilistic matching when the install referrer is missing, while iOS cannot, because Apple banned fingerprinting in iOS 14.5. That missing safety net is why iOS match rates need extra care.
- **The SDK callback is the single most important line of code.** On both platforms the link arrives through the same callback as a normal deeplink, fired once on first launch. If you do not register it, the destination is silently dropped.
- **The deferred window expires, often after about an hour.** The exact default varies by provider, so users who install and open after it lapses land on your default screen rather than the intended page. Test within the window.
- **Verification is not optional.** Universal Links and App Links fail quietly when the domain association file is misconfigured, so you must test real install-from-store flows on physical devices.
## What Is a Deferred Deeplink?
A deferred deeplink is a link that routes a user to a specific in-app destination after they install the app, rather than to a generic landing screen. It defers the routing until the app exists on the device.
Compare the three link types so the difference is clear:
| Link type | App already installed? | Where the user lands |
| --- | --- | --- |
| Standard deeplink | Yes | The specific in-app screen, immediately |
| Deferred deeplink | No | The app store, then the specific screen after first launch |
| Regular install link | No | The app store, then the default home screen |
The value is entirely in retention and conversion. A user who taps an ad for a specific product, a referral to a specific friend, or a promo for a specific plan expects to see that exact thing when the app opens. Dropping them on a home screen breaks the promise of the link and is one of the most common silent leaks in early-stage paid acquisition.
## How Deferred Routing Works: The End-to-End Flow
Before touching code, you need a mental model of the full round trip. Every implementation, regardless of provider, follows the same five steps.
[diagram-deferred-deeplink-flow.webp]
The hard part is step 4, the match. Because the click happens in a browser and the install happens in a fresh app with no shared identity, the system has to reconnect two separate events. How reliably it can do that is the single biggest factor in deferred deeplink accuracy, and it is exactly where iOS and Android diverge.
## How to Implement Deferred Routing on iOS
iOS rests on Universal Links plus a matching strategy that works around the absence of an install referrer. Universal Links are HTTPS URLs that iOS can route into your app instead of opening Safari. If you need a primer on how they differ from Android's equivalent, see [Universal Links vs App Links](https://www.airbridge.io/en/blog/universal-links-vs-app-links-cross-platform-guide).
### 1\. Set up Universal Links and the AASA file
Universal Links are HTTPS URLs that iOS verifies against an `apple-app-site-association` (AASA) file hosted at the root of your domain. Configure these three things:
1. Host the AASA file at `https://yourdomain.com/.well-known/apple-app-site-association`, served over HTTPS with no redirects and a JSON content type.
2. List your App ID in the AASA file. The App ID is your App ID Prefix plus your Bundle ID.
3. Add the Associated Domains capability in Xcode with an `applinks:yourdomain.com` entry.
When this is correct, a tap on a verified Universal Link opens an installed app directly to the matching screen.
### 2\. Understand iOS matching and the ATT constraint
Here is the core iOS problem: Apple does not provide an install referrer API, so there is no Android-style channel for passing a click identifier through the App Store into a fresh install. iOS matching is deterministic and uses one of two signals:
- **IDFA matching**, which ties the click to the install through the advertising identifier. This only works when the user grants App Tracking Transparency (ATT) consent.
- **Clipboard matching**, which writes a token to the clipboard on click and reads it on first launch. It works without IDFA, but only when the user permits clipboard access.
What you cannot do on iOS is fall back to probabilistic fingerprinting. Apple has prohibited matching users by IP address and device characteristics since iOS **14.5**, regardless of ATT status. That removes the safety net Android still has, which is why iOS deferred match rates drop when ATT opt-in is low and no clipboard token is available. Treat the iOS match rate as a number you monitor, not a setting you configure once.
### 3\. Register the deeplink callback on first launch
On iOS, you register a single deeplink callback. Information about the deferred destination is delivered through that same callback, and it fires only once, on the first launch after install. With Airbridge's iOS SDK, you set this callback during SDK initialization in your `AppDelegate`, and the deferred destination then arrives through it automatically without separate handling. The exact method signature varies by SDK version, so confirm yours against the [iOS SDK setup docs](https://help.airbridge.io/en/developers/ios-sdk-v4).
The practical rule: parse the URL you receive in the callback, then perform your normal in-app navigation to that screen. If you skip registering the callback, the destination is computed and then silently discarded.
## How to Implement Deferred Routing on Android
Android gives you a more reliable foundation because it exposes a deterministic install signal.
### 1\. Set up App Links and the assetlinks.json file
Android App Links are HTTPS URLs verified against a Digital Asset Links file. Configure these:
1. Host `assetlinks.json` at `https://yourdomain.com/.well-known/assetlinks.json`.
2. Include your app's package name and its `sha256_cert_fingerprints` in that file. `sha256_cert_fingerprints` is a JSON array, so list every fingerprint you sign with, including the Play App Signing key, as a separate array element. Do not join them into one comma-separated string, which is invalid JSON and breaks verification.
3. Declare the matching intent filter with `android:autoVerify="true"` in your `AndroidManifest.xml`.
When verified, an App Link opens the app directly with no disambiguation dialog.
### 2\. Use the Install Referrer API for deterministic matching
Unlike iOS, Android provides the Google Play Install Referrer API. On install, the Play Store can pass a referrer string that carries the click token, giving you a definitive one-to-one match between the click and the install. This deterministic match is the primary method and is why Android routing is generally more accurate than iOS. When the referrer is unavailable, Android can fall back to probabilistic matching, where accuracy depends on the lookback window: a tight window (for example, **10 to 15 minutes**) limits matches to users who clicked and installed in rapid succession and reduces false positives, while a wide window (such as **24 hours**) pools unrelated users behind the same IP and creates false matches. Most attribution SDKs, including Airbridge's, read the referrer automatically; your job is to make sure the SDK that reads it is initialized before you query for the deferred destination.
### 3\. Register the deeplink callback
On Android, when the deferred destination is resolved it is converted into a scheme deeplink and passed to your deeplink callback. Deferred destinations are passed to that callback automatically, so no separate handler is required beyond the one you already use for standard deeplinks. In Airbridge's Android SDK this is the deeplink callback you register on the SDK option during initialization; see the [Android SDK setup docs](https://help.airbridge.io/en/developers/android-sdk-v4) for the current method name. As on iOS, read the URL and route the user with your normal navigation logic.
## Which SDK Methods to Call: A Quick Reference
The scattered-docs problem usually comes down to one question: which methods do I actually call? Here is the consolidated answer for a provider-managed setup.
| Step | iOS | Android |
| --- | --- | --- |
| Domain association | AASA file plus Associated Domains entitlement | assetlinks.json plus autoVerify intent filter |
| Initialize SDK | Initialize on app launch, in AppDelegate | Initialize on app launch, in Application |
| Register routing | Set the deeplink callback at init | Set the deeplink callback at init |
| Deferred delivery | Same callback, fires once on first launch | Same callback, passed automatically |
| Match method | Deterministic only: IDFA (with ATT) or clipboard token | Install Referrer (deterministic), with probabilistic fallback |
| Forward the link | Pass Universal Link / scheme into the SDK handler | Pass App Link / scheme into the SDK handler |
The one idea to hold onto: standard deeplinks and deferred links share the same callback. You write the routing logic once, and the SDK decides whether the link arrived immediately or was deferred until first launch. You do not write two separate code paths for the destination handling.
## Common Implementation Mistakes
These are the failures that produce a working demo but a broken production flow.
1. **Forgetting the post-install half.** Teams configure Universal Links and App Links, see standard deeplinks work, and assume deferred works too. It does not. Deferred routing only happens if you read the destination from the SDK callback on first launch.
2. **A misconfigured association file.** A redirect on the AASA path, a wrong content type, a missing Play App Signing fingerprint in `assetlinks.json`, or a typo in the App ID will all cause silent verification failure with no error in your build.
3. **Testing on a device that already has the app.** That tests standard deeplinking, not the deferred path. You must uninstall first and install from the store to exercise the real flow.
4. **Ignoring the deferred time window.** Providers expire the stored destination after a set window. Airbridge's deferred window, for example, is **one hour** after the click. Install and open the app well inside your provider's window when testing, or you will see the default screen and assume the code is broken.
5. **Over-wide probabilistic windows on Android.** Probabilistic matching is an Android-only fallback. Setting a long window to lift match rates instead inflates false positives and sends users to the wrong screen, so keep it tight. On iOS, remember that fingerprinting is not available at all, so do not expect an IP-based fallback to rescue a low ATT opt-in rate.
## How to Test and Verify Your Setup
Deferred links fail quietly, so a deliberate test pass is the only way to trust the implementation.
1. **Validate the association files first.** Confirm both `apple-app-site-association` and `assetlinks.json` are reachable over HTTPS, return JSON, and contain the correct App ID, package name, and fingerprints. Apple and Google both publish validators for these files.
2. **Run a clean install test on a physical device.** Uninstall the app, tap your tracking link, install from the store, then open the app. Confirm you land on the intended screen, not the home screen.
3. **Test inside the time window.** Complete the install-and-open flow within the deferred window, then separately test after it expires to confirm the expected fallback behavior.
4. **Test both platforms and both states.** iOS and Android, app installed and app not installed, are four distinct paths. Each one needs to pass.
5. **Watch your match rate over time.** Especially on iOS, treat the match rate as a metric you monitor, not a checkbox. A sudden drop usually points to an OS change or a broken association file.
## Get Deferred Deeplink Routing Right Without Stitching the Docs Together
This is not hard because the concept is complex. It is hard because the moving parts live in four different places, and a single quiet misconfiguration sends every new user to the wrong screen.
A managed deep linking solution removes most of that surface area. Airbridge includes deep linking in its [Core Plan](https://www.airbridge.io/en/plans/core) and Growth Plan: it supports URL scheme deeplinks, App Links, and Universal Links together and automatically selects the right method based on the OS, browser, and install state, so deferred routing arrives through one callback you register once. It also handles the iOS matching problem that breaks most homegrown setups; you can see how that post-IDFA accuracy works in [this breakdown of iOS match rates](https://www.airbridge.io/en/blog/deferred-deeplink-post-idfa-accuracy). If you are migrating off Firebase Dynamic Links, or fighting iOS match rate drops after App Tracking Transparency, start with the SDK setup docs linked above and verify on real devices before you ship.
---
## Universal Links vs App Links: Setting Up Deep Linking on iOS and Android
URL: https://www.airbridge.io/en/blog/universal-links-vs-app-links-cross-platform-guide
Language: English
Published: 2026-06-09
Author: Luke
Your app runs on both iOS and Android. A user taps a link in your email, and it opens the app correctly on iPhone. The same link on Android opens a browser instead. Or the reverse. Half your users get a broken experience, and you are not sure which setup is responsible.
The cause is structural. **iOS and Android each have their own native linking standard, and they do not cross over.** Universal Links vs App Links is not a choice between two options. They are separate implementations for separate platforms. If you want to know how to handle deep linking across iOS and Android without losing users, the answer is: you need both. This guide explains what each standard does, how to set it up, where both fail, and how to decide what you need.
**Key Takeaways**
- **Universal Links and App Links are platform-specific and non-interchangeable.** A Universal Link configured for iOS will not open an Android app. Cross-platform coverage requires two separate implementations running in parallel.
- **Both require a verified JSON file hosted on your domain.** iOS uses the Apple App Site Association (AASA) file. Android uses `assetlinks.json`. Neither works without domain verification.
- **In-app browsers break both on every platform.** Links tapped inside Instagram, TikTok, Facebook, or LinkedIn bypass OS-level routing on iOS and Android alike. This is the most common failure for apps running paid social campaigns. The full diagnosis is in [what breaks when a deep link is opened inside a social app browser](https://www.airbridge.io/en/blog/deep-link-opened-in-social-app-browser-what-breaks).
- **Neither handles uninstalled-app journeys natively.** When a user taps a link without the app installed, the intended destination is lost after the App Store or Play Store redirect. Preserving it requires a separate mechanism called deferred deep linking.
- **Firebase Dynamic Links shut down on August 25, 2025** ([Firebase Deprecation FAQ](https://firebase.google.com/support/dynamic-links-faq)). Teams relying on it for cross-platform or deferred linking need a replacement.
## What Are Universal Links and App Links?
Universal Links (iOS) and App Links (Android) are OS-level linking standards that open a specific screen inside your mobile app directly from a standard HTTPS URL, without routing through a browser first.
They were created to replace the earlier approach: URI schemes.
**Why URI schemes fall short:**
- Any app can register any custom scheme (e.g., `myapp://`). There is no ownership verification.
- Two apps can register the same scheme, causing silent conflicts.
- When the app is not installed, there is no reliable fallback. Users typically land on a blank screen or error page.
**What Universal Links and App Links add:**
- Domain-verified ownership. Only your app can handle links from your verified domain.
- A graceful fallback to your web URL when the app is not installed, typically a redirect to the App Store or Play Store.
### 1\. How Universal Links Work on iOS
Universal Links are Apple's native standard, available on iOS 9 and later. When a user taps a Universal Link, iOS checks two conditions before routing the link into your app:
1. The app is installed on the device.
2. Your domain is verified for that app.
If both are true, iOS opens the app directly to the intended screen. If not, the link falls back to the web URL you configured.
**Technical requirements:**
- Verification uses a JSON file called the **Apple App Site Association (AASA)** file, hosted at `/.well-known/apple-app-site-association` ([Apple Developer Documentation](https://developer.apple.com/documentation/xcode/supporting-associated-domains)).
- The file must be served over HTTPS with no redirects on that exact path.
- The file must be under 128 KB uncompressed.
- Your Xcode project must declare the domain in the **Associated Domains entitlement** using the `applinks:` prefix.
- From iOS 14 onward, Apple fetches the AASA file through its own CDN rather than directly from your server. Updates can take several hours to propagate to new users.
### 2\. How App Links Work on Android
Android App Links are Google's native standard, available on Android 6.0 (Marshmallow) and later. The mechanism works similarly: Android verifies domain ownership before routing links into your app.
**Technical requirements:**
- Verification uses a JSON file called `assetlinks.json`, hosted at `/.well-known/assetlinks.json` ([Android Developers Documentation](https://developer.android.com/training/app-links/verify-applinks)).
- The file must include your app's **package name** and the **SHA-256 fingerprint** of your release signing certificate.
- Your app manifest must declare `android:autoVerify="true"` on the relevant intent filters.
- Android verifies the domain at install time. Initial verification takes up to 20 seconds.
- From Android 15 onward, the system re-verifies domains periodically in the background. Changes to your `assetlinks.json` can take up to seven days to propagate across all devices.
## Universal Links vs App Links: Setup Requirements Compared
Both standards share the same foundation: an HTTPS domain with a valid SSL certificate and a JSON verification file at a well-known path. Everything else is platform-specific.
| Requirement | Universal Links (iOS) | App Links (Android) |
| --- | --- | --- |
| Verification file | apple-app-site-association | assetlinks.json |
| File location | /.well-known/AASA | /.well-known/assetlinks.json |
| App configuration | Xcode Associated Domains entitlement | android:autoVerify="true" + intent filters |
| Developer account | Apple Developer Program ($99/year) | Google Play Console |
| Certificate requirement | Valid SSL, no redirects on file path | SHA-256 fingerprint of signing cert in JSON |
| Minimum OS | iOS 9+ | Android 6.0+ |
| Update propagation | Hours (Apple CDN, from iOS 14+) | Up to 7 days (Android 15+ background re-verify) |
### 1\. Setting Up iOS Universal Links
Complete these steps in order:
1. **Host the AASA file** at `https://yourdomain.com/.well-known/apple-app-site-association`. Serve it with `Content-Type: application/json`, no redirects, and file size under 128 KB.
2. **Define path patterns** in the AASA JSON. Specify only the URL paths your app needs to handle. Avoid overly broad wildcards like `/*` unless you intend to intercept every path on your domain.
3. **Add the Associated Domains entitlement in Xcode.** Under Signing and Capabilities, add `applinks:yourdomain.com`. List each subdomain you want to support as a separate entry.
4. **Enable developer mode for testing.** In debug builds, set the entitlement to `applinks:yourdomain.com?mode=developer`. This bypasses Apple's CDN and fetches your file directly, making iteration much faster.
5. **Check the cached version.** Visit `https://app-site-association.cdn-apple.com/a/v1/yourdomain.com` to inspect what Apple's CDN is currently serving for your domain.
**Most common mistake:** A server-side redirect on the AASA file path. Apple's CDN crawler does not follow redirects. One redirect causes a silent verification failure with no error message.
### 2\. Setting Up Android App Links
1. **Create and host** `assetlinks.json` at `https://yourdomain.com/.well-known/assetlinks.json`. Serve it with `Content-Type: application/json`. It must be publicly accessible without VPN or authentication.
2. **Get the correct SHA-256 fingerprint.** If you use Play App Signing, use the fingerprint from the Play Console under App integrity. This is different from your local keystore fingerprint.
3. **Update the app manifest.** Add `android:autoVerify="true"` and `android:host="yourdomain.com"` to the relevant ``.
4. **Handle multiple hostnames separately.** If your intent filters declare more than one hostname, each needs its own `assetlinks.json` file. Android verifies each domain independently.
5. **Validate with App Links Assistant.** In Android Studio, go to Tools > App Links Assistant. It validates manifest declarations and lets you simulate link handling locally.
**Most common mistake:** SHA-256 fingerprint mismatch. Teams that migrated to Play App Signing often use their local keystore fingerprint instead of the Play Console value, which are different certificates.
For a detailed debugging walkthrough, the [Android App Links setup guide](https://www.airbridge.io/en/blog/android-app-links-setup-guide) covers fingerprint errors, redirect issues, and testing tools step by step.
## Why Universal Links and App Links Break in Instagram, TikTok, and Facebook
This is the most common production failure for apps running paid social campaigns, and it affects iOS and Android equally. A link tapped inside Instagram, TikTok, Facebook, or LinkedIn opens in that app's built-in browser. The host app controls navigation inside its browser, and Instagram, Facebook, and TikTok keep the tap on the web in practice, so Universal Links and App Links do not fire and the user lands on your web page instead of your app. On iOS there is a second trap: an installed user's first tap on a tracking link from a non-Safari app can go to the App Store.
The fix is a three-layer setup:
| Layer | What it handles |
| --- | --- |
| Universal Links | Standard iOS link taps (Safari, Mail, Messages, etc.) |
| App Links | Standard Android link taps (Chrome, Gmail, SMS, etc.) |
| URI scheme fallback or tracking link with a Stopover Airpage | In-app browser contexts on both platforms |
The symptom-by-symptom diagnosis, the per-platform behavior table, and the step-by-step fix now live in a dedicated guide: [What breaks when a deep link is opened from inside a social app browser](https://www.airbridge.io/en/blog/deep-link-opened-in-social-app-browser-what-breaks).
## What Is Deferred Deep Linking, and Why Native Standards Skip It
**Deferred deep linking** is a mechanism that preserves an intended in-app destination across the app installation process. When a user taps a link and the app is not installed, deferred deep linking stores their destination, redirects them to the App Store or Play Store, and then routes them to the correct screen on their very first app open after install.
Neither Universal Links nor App Links provide this natively. They only handle routing when the app is already on the device. When the app is absent, both fall back to your web URL, and the link destination is gone after the app store redirect.
**Why this matters for paid user acquisition:**
Consider this sequence with and without deferred deep linking:
| Step | Without deferred deep linking | With deferred deep linking |
| --- | --- | --- |
| User taps your ad (free trial offer) | Link goes to App Store | Link stores the destination, goes to App Store |
| User installs the app | App opens normally | App checks for stored destination |
| First open | Default home screen | Trial offer screen |
| Result | Ad-to-activation disconnect | Seamless ad-to-first-session match |
Paid acquisition volume is growing on both platforms. That growth raises the cost of every lost activation: more users arriving through paid campaigns means more first-opens where a missing deferred link drops the user on the wrong screen.
Deferred deep linking requires a third-party SDK or service. Apple and Google do not provide it as part of their native standards. For teams dealing with iOS-specific privacy changes since iOS 17 that affect deferred linking match rates, Airbridge's guide on [deferred deep linking after iOS 17](https://www.airbridge.io/en/blog/deferred-deeplink-post-idfa-accuracy) covers the implications in detail.
## Choosing Your Cross-Platform Setup
| Your situation | Recommended setup |
| --- | --- |
| iOS only, no paid acquisition | Native Universal Links |
| Android only, no paid acquisition | Native App Links |
| Both platforms, organic traffic only | Universal Links + App Links + URI scheme fallback |
| Both platforms, running paid social ads | Universal Links + App Links + URI scheme + deferred linking tool |
| Post-Firebase Dynamic Links migration | Dedicated tool, or native rebuild with custom deferred linking backend |
**When to use a dedicated cross-platform tool:**
Each of the three layers fails differently. The AASA file breaks on server-side redirects. The `assetlinks.json` breaks on SHA-256 fingerprint drift after every Play App Signing rotation. The URI scheme fallback breaks when Instagram or TikTok ships a new in-app browser User-Agent string. Re-testing each layer after every major iOS and Android release, and after each social app update, is the real ongoing cost. A dedicated tool consolidates all three into one link and handles those failure modes automatically.
## Cross-Platform App Linking: The Bottom Line
Universal Links and App Links are the correct native foundation for in-app linking. They are platform-specific, non-interchangeable, and both required if your app runs on iOS and Android.
Getting cross-platform deep linking right involves three pieces:
1. **Native Universal Links and App Links:** domain verification on each platform separately. Both verification files share a single HTTPS domain, so the maintenance is lighter than it sounds.
2. **URI scheme fallback:** for in-app browser contexts across Instagram, TikTok, Facebook, and LinkedIn.
3. **Deferred linking layer:** for paid acquisition campaigns on either platform.
Start with the native setups, verify them end-to-end on both platforms, and add deferred linking once the foundation is solid.
---
## Android App Links Not Verifying? The Complete assetlinks.json Setup Guide
URL: https://www.airbridge.io/en/blog/android-app-links-setup-guide
Language: English
Published: 2026-06-08
Author: Luke
You followed Google's documentation step by step. You uploaded the file. You added `android:autoVerify="true"` to your manifest. But when you tap a link on the test device, it still opens in the browser instead of your app.
This is the most common Android App Links failure pattern, and the root cause is almost never what developers expect. Android App Links are a class of deep link that open directly inside your app once a domain association is properly established through a [Digital Asset Links file](https://developer.android.com/training/app-links/verify-android-applinks). If that verification fails, Android shows no error. The link silently opens in the default browser.
**This guide covers how to verify Android App Links with your assetlinks.json file and fix the five issues responsible for most failed setups.**
**Key Takeaways**
- **The SHA256 fingerprint mismatch is the #1 failure cause.** If you use Play App Signing, Google re-signs your APK with a different key. Your local keystore fingerprint will not match, and verification fails silently.
- **Your assetlinks.json must return HTTP 200 with Content-Type: application/json.** Any redirect (301 or 302) breaks Android's verification crawler, even if your browser handles it fine.
- **Android 12 removed the fallback chooser dialog.** Before Android 12, an unverified app link showed a chooser. On Android 12 and later, it opens directly in the browser with no indication that verification failed.
- **robots.txt can silently block Android's domain verification crawler.** If your `/.well-known/` path is disallowed, verification fails with no visible error.
- **ADB commands let you test verification without waiting for real users.** Run `adb shell pm get-app-links` to see the exact verification state for each declared domain.
## How Android App Links Verification Works (And Why It Fails Silently)
Android App Links open directly inside your app with no chooser dialog, once the domain and app are properly associated via a Digital Asset Links file hosted on your server. This is different from regular deep links, which can still trigger the "Open with" chooser on unverified setups.
When a link click cannot be verified, Android shows no error. The link silently opens in the default browser. This is by design, but it makes debugging difficult.
### What Happens During the Verification Process
According to the [official Android App Links documentation](https://developer.android.com/training/app-links/verify-android-applinks), the verification sequence works as follows:
1. **App installs or updates** on the device.
2. **Android reads your** `AndroidManifest.xml` and collects every hostname declared in intent filters where `android:autoVerify="true"` is set.
3. **Android fetches** `https://[your-domain]/.well-known/assetlinks.json` for each hostname.
4. **Android compares** the `sha256_cert_fingerprints` value in the file against your app's signing certificate.
5. **If they match**, the domain is marked as verified and links to that domain open directly in your app.
6. **If they don't match**, or if the file cannot be fetched, the domain remains unverified.
Android waits up to **20 seconds** after install for this asynchronous process to complete. If anything fails during that window, there is no user-facing error.
### What Changed in Android 12 That Makes Failures Invisible
Before Android 12, an unverified intent filter still triggered a chooser dialog, letting users pick between the browser and your app. That dialog was a visible signal that something was misconfigured.
Starting with [Android 12 (API level 31)](https://developer.android.com/about/versions/12/features#app-links-verification), the behavior changed:
- Unverified App Links open **directly in the default browser**, with no chooser.
- There is no user-facing indication that your app was a candidate to handle the link.
- A broken setup on Android 12+ is **completely invisible** unless you use ADB to check the verification state.
This also means a setup that appeared to work fine on Android 11 can silently break on Android 12 without any code change on your side.
## Step 1: How to Create and Host Your assetlinks.json File
The Digital Asset Links file is a JSON array that tells Android which apps are authorized to handle links for your domain. It must be hosted at a specific path on your web server with specific server settings.
### What the assetlinks.json File Should Contain
Every field in this structure is required. Missing or misspelling any key causes verification to fail:
```json
[
{
"relation": ["delegate_permission/common.handle_all_urls"],
"target": {
"namespace": "android_app",
"package_name": "com.yourcompany.yourapp",
"sha256_cert_fingerprints": [
"AA:BB:CC:DD:EE:FF:00:11:22:33:44:55:66:77:88:99:AA:BB:CC:DD:EE:FF:00:11:22:33:44:55:66:77:88:99"
]
}
}
]
```
Two formatting mistakes that fail silently:
- **Lowercase fingerprint.** The SHA256 fingerprint is case-sensitive. It must be **UPPERCASE** with colons between bytes. `aa:bb:cc` will fail; `AA:BB:CC` will pass.
- **Wrong relation string.** The `relation` field must be exactly `["delegate_permission/common.handle_all_urls"]`. Any variation breaks verification.
### Where to Host assetlinks.json and What Your Server Must Return
The file must be accessible at this exact URL:
```
https://[your-domain]/.well-known/assetlinks.json
```
Your server must meet all four of the following requirements, as specified in the [Google Digital Asset Links statement documentation](https://developers.google.com/digital-asset-links/v1/statements):
| Requirement | What It Means | Common Mistake |
| --- | --- | --- |
| HTTPS only | Serve the file over HTTPS, even if your intent filter uses http:// | Serving from HTTP or a mixed scheme |
| HTTP 200 response | Return exactly HTTP 200. Android does not follow 301 or 302 redirects | CDN or server redirects on /.well-known/ |
| Content-Type: application/json | The response header must specify application/json | Server defaults to text/plain for .json files |
| robots.txt allows crawling | The /.well-known/ path must not be blocked | Disallow: / or Disallow: /.well-known/ in robots.txt |
**Multiple domains:** If your app handles multiple hostnames, you must host a separate `assetlinks.json` file on each domain. A single file on one domain does not cover other domains.
The diagram below shows how these requirements fit into the full verification flow:
[Screenshot 2026-06-08 at 19.20.18.png]
## Step 2: How to Get the Correct SHA256 Certificate Fingerprint
The fingerprint mismatch between what is in your `assetlinks.json` and what is signed on the device is the single most common cause of App Links verification failures. This is almost always caused by Play App Signing.
### Why Play App Signing Changes Your SHA256 Fingerprint
[Play App Signing](https://support.google.com/googleplay/android-developer/answer/9842756) is mandatory for all new apps published to the Google Play Store since August 2021. When you enroll, Google holds and manages your app's release signing key, then re-signs your APK with that Google-managed key before delivering it to users.
This means two separate keys exist for your app:
- **Your upload key:** The key you use to sign the AAB before uploading to Play Console. Google verifies you are the publisher, then strips this signature.
- **Google's app signing key:** The key Google uses to sign the final APK delivered to users' devices.
If you extract a SHA256 fingerprint from your local keystore using `keytool` and put it in your `assetlinks.json`, App Links verification will fail for every user who installed the app from the Play Store.
### How to Find Your Correct SHA256 Fingerprint in Google Play Console
**For apps using Play App Signing (all new apps):**
1. Open [Google Play Console](https://play.google.com/console).
2. Select your app.
3. Go to **Release > Setup > App signing**.
4. Under **"App signing key certificate"**, copy the **SHA-256 certificate fingerprint**.
Play Console also shows a pre-formatted `assetlinks.json` snippet on this page. Using that snippet directly eliminates copy-paste formatting errors.
**For legacy apps with self-managed signing:**
Run the following command against your release keystore:
```bash
keytool -list -v \
-keystore your-release.keystore \
-alias your-key-alias
```
Find the `SHA256:` line in the output. Format it as uppercase hex bytes separated by colons.
### How to Support Multiple Signing Keys in assetlinks.json
If your app is distributed through multiple channels (Play Store, direct APK download, enterprise distribution), include all signing key fingerprints as separate entries in the array:
```json
"sha256_cert_fingerprints": [
"AA:BB:CC:...(Google Play App Signing key)...",
"11:22:33:...(your upload key or legacy key)..."
]
```
Including all keys lets verification succeed regardless of which signing path was used for a given install.
## Step 3: How to Configure AndroidManifest.xml and Test Verification
### Required AndroidManifest.xml Intent-Filter Configuration
Every domain you want to handle as an App Link requires a properly structured ``. Four attributes are required, and missing any one causes Android to skip verification for that domain entirely:
```xml
```
The four required elements are:
- `android:autoVerify="true"` on the `` tag (not the `` tag)
- `android.intent.action.VIEW`
- `android.intent.category.DEFAULT`
- `android.intent.category.BROWSABLE`
**Multiple domains and Android version differences:**
- On **Android 12 and later**, verification is per-domain. A failure on one domain does not affect others. Add `android:autoVerify="true"` to each intent filter individually.
- On **Android 11 and earlier**, if any declared domain fails verification, all domains in the app fail. This makes it especially important that every declared domain has a valid, reachable `assetlinks.json` file before publishing.
### How to Check Verification Status Using ADB
Use this command with your device connected and the app installed:
```bash
adb shell pm get-app-links --user cur com.yourcompany.yourapp
```
The output lists each declared domain with its current verification state:
- `verified`**:** Domain is verified. Links open directly in your app.
- `selected`**:** User manually selected your app for this domain (not auto-verified by the system).
- `none`**:** Verification has not run yet, or the state was reset.
To reset and re-trigger verification without reinstalling the app:
```bash
# Step 1: Reset the current verification state
adb shell pm set-app-links --package com.yourcompany.yourapp 0 all
# Step 2: Trigger re-verification
adb shell pm verify-app-links --re-verify com.yourcompany.yourapp
```
Wait at least **20 seconds**, then run `get-app-links` again to check the updated result.
### How to Test App Link Routing Directly on a Device
Simulate a link tap to confirm your app opens instead of the browser:
```bash
adb shell am start \
-a android.intent.action.VIEW \
-c android.intent.category.BROWSABLE \
-d "https://www.yourapp.com/your-path"
```
- If verification succeeded: your app opens directly.
- If it opens the browser or shows a chooser: verification failed for that domain.
You can also use [Google's Statement List Generator and Tester](https://developers.google.com/digital-asset-links/tools/generator) to check whether your `assetlinks.json` is reachable, valid JSON, and matches your app's package name and fingerprint, without needing a device.
## 5 Mistakes That Silently Break Android App Links Verification
Most App Links failures share one trait: Android produces no error and the link just opens in the browser. The table below covers the five most common silent failure causes and how to fix each one:
| Mistake | Why It Fails | How to Fix |
| --- | --- | --- |
| Using the local keystore fingerprint with Play App Signing | Google re-signs your APK. The installed cert does not match your local key. | Get the SHA256 from Play Console under Release > Setup > App signing. |
| HTTP redirect on the assetlinks.json URL | Android does not follow 301 or 302 redirects when fetching the file. | Serve the file with a direct HTTP 200. Remove CDN or server-level redirects on /.well-known/. |
| robots.txt blocking /.well-known/ | Android's verification crawler is blocked and cannot fetch the file. | Add Allow: /.well-known/assetlinks.json to your robots.txt, or remove the Disallow rule for that path. |
| Wrong Content-Type header | Android expects Content-Type: application/json. Some servers default to text/plain for .json files. | Configure your web server or CDN to serve .json files with Content-Type: application/json. |
| Lowercase fingerprint in assetlinks.json | The SHA256 comparison is case-sensitive. Lowercase bytes fail verification. | Use uppercase hex characters with colons between bytes, exactly as shown in the Play Console snippet. |
If you manage campaign URLs across multiple domains and find yourself updating `assetlinks.json` files manually each time a subdomain changes, consider using a [deep linking tool](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers) that handles asset links file generation automatically.
## How to Use Verified App Links in Paid Marketing Campaigns
Once your Android App Links are verified, every `https` link to your domain opens your app with no browser detour. That matters most in paid acquisition.
**Why this affects your conversion numbers:**
When a user clicks a paid ad pointing to your domain, a verified App Link sends them directly to the right screen inside your app. Without verification, the user lands in the browser, has to find the app manually, and often abandons the flow entirely. That gap shows up in trial start rates, onboarding completion, and subscription conversion across every campaign.
**How deep linking tools use App Links:**
If you use a tool to manage campaign URLs and measure which ad drove which install or trial, App Links verification is a prerequisite for any of that routing to work correctly. [Airbridge's deep linking product](https://www.airbridge.io/en/product/deep-linking) handles the `assetlinks.json` configuration as part of its SDK setup. Once the domain is registered, the asset links file is maintained automatically, and your marketing team can create campaign links without manually editing the JSON file each time a domain changes.
For teams measuring subscription conversions end-to-end, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects the ad click to the install, trial, and subscription in one funnel. It supports Meta, Google, Apple Search Ads, and TikTok, with 15K attributed installs free to start.
---
## Airbridge Core vs Growth: Which Plan Fits Your Team?
URL: https://www.airbridge.io/en/blog/core-vs-growth-plan-comparision
Language: English
Published: 2026-06-07
Author: Harper (Trang Nguyen)
Two Airbridge attribution plans, one decision: Core (free to 15K attributed installs) or Growth (custom). Core covers Meta, Google, Apple Search Ads, and TikTok. Growth covers everything else. The feature table alone doesn't tell you which one fits where your team is today.
Picking the wrong plan has a real cost. Undershoot and you hit Core's limits right when your paid UA is scaling. Overshoot and you're paying for raw data pipelines and 330+ ad network integrations you won't need for 12 months.
**Key Takeaways**
- **Core covers the four paid UA channels most subscription apps actually run.** Meta, Google, Apple Search Ads, and TikTok, plus the full install-to-trial-to-subscription funnel. Free up to 15K attributed installs.
- **Growth adds everything Core does not.** 330+ ad networks, custom events, raw data export, fraud detection, and a dedicated CSM.
- **Most teams starting paid UA belong on Core.** Meta, Google, Apple Search Ads, and TikTok cover the channels where their first paid dollars land. Self-serve, no sales call, starts at $0.
## The Airbridge Core Plan: Subscription Attribution for Focused Paid UA Teams
**Core is the right starting point for most subscription app teams running paid campaigns.** It covers the four channels that drive the majority of paid installs for consumer subscription apps, and it starts at $0.
### 1\. What it includes
- **4 Self-Attributing Networks:** Google Ads, Meta Ads, Apple Search Ads, and TikTok for Business (known as **GMAT**)
- **25 standard subscription events:** Start Trial, Subscribe, Unsubscribe, and 22 more. Predefined schema means no event naming decisions required.
- **SKAN 4.0 support** for iOS attribution
- **Web-to-App attribution** to track installs originating from your mobile website
- **Audience sync** to ad networks for retargeting and exclusion lists
- **Predictive LTV** (180-day modeling window)
- **Up to 2 third-party integrations:** RevenueCat, Adapty, Amplitude, and others on the supported list
- **Funnel, retention, and revenue reports** available from day one
### 2\. Pricing
**15,000 attributed installs included free.** After that: **$0.05 per install.** No annual contract. No sales call required. Self-serve onboarding.
Enterprise attribution platforms typically require minimum contracts of $2,000-$5,000 per month. Core starts at $0 and scales with your actual install volume.
### 3\. Core Plan limitations
Core is intentionally scoped. Confirm none of the following block your use case before choosing it:
- **No custom events.** You are limited to Core's [25 standard events](https://help.airbridge.io/en/guides/airbridge-event-elements). If your funnel includes milestones outside that set, you need Growth.
- **Max 2 third-party integrations.** A third platform connection requires Growth.
- **GMAT channels only.** Campaigns on any network outside Google, Meta, Apple Search Ads, or TikTok will not be attributed in Core.
- **No raw data export.** Reports are available inside the Airbridge dashboard only. No BigQuery, S3, or warehouse export.
- **No Meta Ads CAPI.** Server-side conversion API for Meta is not included.
- **No agency or partner dashboard access.** External media agencies that need their own login require Growth.
### 4\. When Core is the right fit
- You are running (or about to run) paid campaigns on Meta, Google, TikTok, or Apple Search Ads
- Your subscription funnel fits within 25 standard subscription events
- You use RevenueCat or Adapty as your subscription management layer
- Your team is 1-20 people and prefers self-serve onboarding over a managed implementation
- Your monthly install volume is under approximately 50,000 (at 50K installs per month, billable installs are 35K above the free tier: 35,000 x $0.05 = **$1,750/month**. At that scale, request a Growth quote to compare)
## The Airbridge Growth Plan: Advanced Measurement for Scaling Teams
**Growth is the full Airbridge platform.** It adds every capability that Core deliberately excludes.
### 1\. What Growth adds over Core
- **330+ ad network integrations.** DSPs, affiliates, influencer networks, and regional ad platforms are all covered.
- **Custom events.** Design your own event schema for any funnel step Core's standard events do not capture.
- **Raw data export.** Push event-level data to S3, Google Cloud Storage, or your data warehouse.
- **ML-powered fraud detection.** Identifies and filters fraudulent install traffic automatically.
- **Unlimited third-party integrations.** Connect every analytics and engagement platform your stack requires.
- **Multi-touch attribution and advanced reports.** Touchpoint analytics and advanced conversion analysis.
- **Dedicated CSM and migration support.** A named customer success manager handles onboarding and ongoing optimization.
### 2\. Pricing
Custom, based on MAU or install volume. Contact Airbridge for a quote.
### 3\. When Growth is the right fit
- You run campaigns on networks outside the four GMAT channels
- Your funnel includes events Core's standard set does not cover
- Your data team pulls raw event data into a warehouse for analysis
- You are evaluating Airbridge as a replacement for your current attribution platform at scale
- You work with an external media agency that needs dashboard access
## Comparing Core and Growth
### Side-by-side feature comparison
| Feature | Core | Growth |
| --- | --- | --- |
| Ad attribution (paid UA) | Yes | Yes |
| Supported ad channels | GMAT (4) | 330+ |
| Standard subscription events | 25 | 25 + custom |
| Custom events | No | Yes |
| Third-party integrations | Up to 2 | Unlimited |
| Web-to-App attribution | Yes | Yes |
| Raw data export | No | Yes |
| SKAN 4.0 support | Yes | Yes |
| Fraud detection | No | Yes |
| Dedicated CSM | No | Yes |
| Starting price | 15K free, $0.05/install | Custom |
### How to choose in 2 questions
Answer these in order. The first "yes" is your plan.
1. **Does your funnel fit within 25 standard subscription events, 2 integrations, and campaigns on Meta, Google, TikTok, or Apple Search Ads only?** Core is your plan. It starts at $0, requires no sales call, and connects your GMAT ad spend directly to subscription revenue. Check the [standard event list](https://help.airbridge.io/en/guides/airbridge-event-elements) to confirm your funnel is covered.
2. **Do you run campaigns outside the four GMAT channels, need custom events or raw data export, or are you evaluating Airbridge as a replacement for your current attribution platform at scale?** Contact Airbridge for a Growth quote. Pricing is customized to MAU or install volume.
In a recent internal Airbridge ICP study (n=21 subscription app founders), 13 said they would not take a sales call to evaluate a measurement tool. Core is built for that preference. It is the only Airbridge plan that starts without a sales conversation.
## Start on Core, Move to Growth When the Data Says So
Most subscription app teams working through their first $5,000-$10,000 in monthly paid UA will find Core covers everything they need. You see which channels drive paying subscribers, not just installs. You connect RevenueCat or Adapty as your subscription revenue source of truth. You pay nothing until you exceed 15,000 attributed installs.
Core to Growth is a progression, not a platform switch. Core's attribution history carries into Growth, and every integration you built on Core stays in place.
When your install volume, network mix, or data requirements outgrow Core, Growth is there. Until then, the right plan is the one that costs you nothing while your paid UA is still finding its footing.
**Ready to connect your ad spend to subscription revenue?** [Start free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core): RevenueCat and Adapty supported, 15K attributed installs included, no contract required.
---
## How to Set Up Universal Links on iOS (2026 Guide)
URL: https://www.airbridge.io/en/blog/universal-links-setup-guide-ios
Language: English
Published: 2026-06-05
Author: Luke
You added the apple-app-site-association file, toggled on the entitlement, and built to your phone. You tapped the link. Safari opened anyway, and your app sat there untouched while the web page loaded behind it.
This is the most common way iOS app linking goes wrong, and the cause is almost never your app code.
> **Quick answer:** Universal Links are HTTPS URLs that open your app directly to specific content when it is installed, and fall back to the web when it is not. They break when one of two pieces is misconfigured: the JSON file on your domain, or the Associated Domains entitlement in your project. Get both exactly right and links route to your app in milliseconds. Get one character wrong and iOS silently ignores the whole thing.
**Key Takeaways**
- **Two pieces must agree, byte for byte.** The apple-app-site-association file on your server and the `applinks:` entitlement in Xcode have to reference the same domain and the same App ID, or iOS treats the link as a normal web URL.
- **The file format is strict.** No `.json` extension, `Content-Type: application/json`, served over HTTPS with no redirects, and under **128 KB** uncompressed, per [Apple's associated domains documentation](https://developer.apple.com/documentation/xcode/supporting-associated-domains).
- **The App ID format trips up most teams.** The `appID` value is your **Team ID** followed by your bundle identifier, with no `https://` and no trailing slash anywhere in the file.
- **iOS fetches the file through an Apple CDN, not your server.** Since iOS 14, validation runs through Apple's content delivery network, so a freshly edited file can take time to propagate, per [Apple's Technical Note 3155 (TN3155) on debugging](https://developer.apple.com/documentation/technotes/tn3155-debugging-universal-links).
- **Test on a real device with a real tap.** Pasting the URL into Safari's address bar will never open the app. The link has to be tapped from another app like Notes or Messages.
## What Are Universal Links, and Why Do They Fail?
A Universal Link is a standard HTTPS web address that iOS routes to your installed app instead of Safari. The same URL still works as a web page for anyone without the app, so you ship one link that serves both audiences.
The guidance below reflects the file format introduced in iOS 13 and the validation flow that has been stable since iOS 14. As of iOS 26, the current release in 2026 after Apple moved to year-based version names, no breaking changes affect this setup.
When a link does not open your app, it is tempting to start debugging your `AppDelegate`. That is the wrong place to look. iOS decides whether to hand a URL to your app long before your code runs, during a verification handshake between two declarations:
- **Your entitlement** declares which domains your app claims.
- **The apple-app-site-association file** on each of those domains declares which apps may open its links.
Both halves have to match. If the App ID in the file does not match the App ID that Apple computes from your signed build, the claim is rejected and nothing happens.
### The Two Mistakes Behind Most Failures
1. **The protocol is in the domain string.** The entitlement value is `applinks:www.example.com`, never `applinks:https://www.example.com`. Adding the scheme breaks the match, and iOS gives you no error message when it does.
2. **The App ID is incomplete.** The `appID` is your Team ID and your bundle identifier joined by a dot, for example `ABCDE12345.com.example.myapp`. Teams often paste the bundle identifier alone, or copy a Team ID from the wrong membership when they belong to more than one Apple Developer account.
One subtle variant catches teams with more than one provisioning profile. The App ID prefix is usually your Team ID, but apps using a custom App ID prefix can have a different value. If a link works on one build and not another, check for an identifier mismatch between your development and distribution provisioning first.
Here is how iOS resolves a tapped link:
[Screenshot 2026-06-05 at 14.33.15.png]
## How to Set Up Universal Links on iOS in 4 Steps
If you have searched for how to set up universal links on iOS and found conflicting answers, the reason is that the file format changed across iOS versions. The four steps below cover the current approach.
### Step 1: Create the apple-app-site-association File
Create a file named exactly `apple-app-site-association`, with no `.json` extension even though the contents are JSON. The modern format introduced in iOS 13 uses a `components` array, which matches on paths, query items, and URL fragments:
```json
{
"applinks": {
"details": [
{
"appIDs": ["ABCDE12345.com.example.myapp"],
"components": [
{
"/": "/products/*",
"comment": "Opens any product detail page"
}
]
}
]
}
}
```
Two rules govern this file:
- **The** `appIDs` **value** is your Team ID (the 10-character prefix from your Apple Developer account), a dot, then your bundle identifier.
- `components` **overrides** `paths`**.** On iOS 13 and later, if `components` is present the older `paths` key is ignored, so do not mix the two formats expecting both to apply.
- **The** `comment` **field is documentation only.** Apple's parser preserves it but ignores it for routing, so use it to explain a rule to the next engineer, not to change behavior.
If you still support iOS 12 or earlier, keep a `paths` key alongside `components` as a fallback. Modern versions read `components`, older ones read `paths`, and including both is safe as long as the entries stay consistent. By 2026 iOS 12 is at roughly zero percent share, so most teams can drop it.
The `components` array also gives you finer control, as described in [Apple's guide to configuring an associated domain](https://developer.apple.com/documentation/xcode/configuring-an-associated-domain). Each entry can match the path (`/`), the query (`?`), and the fragment (`#`), and an entry flagged `"exclude": true` routes matching URLs to the browser instead of the app. Order matters: iOS uses the first matching entry, so list exclusions and specific rules before broad ones.
For example, to open `/products/123` in the app but keep `/products/preview` on the web, place an excluded `"/products/preview"` rule above your broad `"/products/*"` rule. The preview URL is matched first and falls through to Safari, while every other product path opens the app.
### Step 2: Host the File Over HTTPS
Upload the file so it is reachable over HTTPS at one of these paths:
- `https://yourdomain.com/.well-known/apple-app-site-association` (current recommendation)
- `https://yourdomain.com/apple-app-site-association`
Three server requirements are non-negotiable:
1. **Content type:** the file must be served with the header `Content-Type: application/json`.
2. **No redirects:** it must respond over HTTPS with a valid certificate and no redirects.
3. **Size:** it must be under 128 KB uncompressed.
A redirect or a wrong content type is enough for iOS to discard the file without telling you.
### Step 3: Add the Associated Domains Entitlement
In Xcode, select your target, open **Signing & Capabilities**, and add the **Associated Domains** capability. Add one entry per domain in the form `applinks:yourdomain.com`. The scheme is the literal word `applinks:`, and the value after it is a bare domain with no `https://`.
Use this table to sanity-check the four values teams get wrong most often:
| Configuration item | Correct value | Common wrong value |
| --- | --- | --- |
| Entitlement entry | applinks:example.com | applinks:https://example.com |
| File name | apple-app-site-association | apple-app-site-association.json |
| App identifier | TEAMID.com.example.app | com.example.app |
| Content type | application/json | text/html |
### Step 4: Handle the Incoming Link in Your App Code
Once verification passes, iOS launches your app and calls the continue-user-activity handler. In UIKit this is `application(_:continue:restorationHandler:)`; in SwiftUI it is the `.onContinueUserActivity` modifier, which is the default path for most apps built in 2026. Read the incoming `webpageURL`, parse the path, and route the user to the matching screen. This is the only step that lives in your code, and it rarely breaks. Two details save debugging time:
- **Route from one shared function.** The handler can fire while your app is already in the background, not just on a cold launch, so do not put routing only in your launch path.
- **Always handle unknown paths.** Send the user to a sensible default like your home tab when a path does not match a known screen, rather than showing a blank view.
## How Do You Fix a Universal Link That Won't Open?
When the link still opens Safari, work through the chain in order rather than changing several things at once.
1. **Confirm the file is reachable.** Open the URL in a desktop browser and verify it returns raw JSON, not an HTML error page or a login wall. A frequent cause is a firewall or security rule that blocks reads from arbitrary IP addresses.
2. **Account for the Apple CDN.** Since iOS 14, the device does not request the file from your server directly. It asks an Apple-managed CDN (documented in TN3155 above), so the file must be publicly fetchable for that CDN to cache it. An edit you just published may not take effect immediately.
3. **Bypass the cache during development.** Enable the **Associated Domains Development** toggle in the device's Developer settings to query your domain directly instead of through the CDN.
4. **Watch for HTTP/2 surprises.** Some servers that answer over HTTP/1.1 fail to return the file over HTTP/2, which Apple's CDN negotiates. If a browser can fetch the file but the device cannot, confirm the file serves correctly over HTTP/2.
5. **Test the right way.** A tap from Safari's address bar will never hand off to the app, by design. Apple's documented method, in [Technical Q&A 1916 (QA1916)](https://developer.apple.com/library/archive/qa/qa1916/_index.html), is to email yourself the link, then long press it and confirm an **Open in \[Your App\]** option appears. If it does, the association is working.
Consider a concrete case. A user gets a password-reset email, taps the link, and lands on the mobile web page instead of the in-app reset screen. In the most common version of this story, the file resolves correctly in a browser, and the real fault is a redirect on the email-tracking domain that strips the request before Apple's CDN sees a clean response.
If you want a primer on how URI schemes, App Links, and Universal Links differ, this [breakdown of deep linking methods](https://www.airbridge.io/en/blog/what-is-deep-link) is a useful reference, and the [implementation walkthrough for marketers and developers](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers) covers the moving parts end to end.
## How to Connect Your App Links to Marketing Campaigns
Getting the file and entitlement right solves the engineering problem. It leaves a separate one open. Once your links open the app, you still cannot see which email, ad, or post sent each user there, and you cannot route users who do not have the app yet.
This is where a deep-linking platform earns its place. Airbridge hosts a verified apple-app-site-association file on its own link domains, so you point your entitlement at `applinks:YOUR_APP_NAME.airbridge.io` and `applinks:YOUR_APP_NAME.abr.ge` instead of maintaining the file yourself. It then picks the right method, whether URI scheme, App Link, or Universal Link, based on the device and install state, as described in the [Airbridge Help Center](https://help.airbridge.io/en/guides/understanding-deep-link).
The payoff shows up in two places:
- **Deferred routing for new users.** Attach a destination to a campaign link, and a user without the app is sent to the App Store, then dropped onto the intended screen the first time they open it.
- **Campaign-level visibility.** Each tap is tied to the channel and campaign that produced it.
## Your Checklist for a Working Universal Link
A Universal Link breaks for boring reasons: a stray `https://`, a missing content type, a file behind a redirect. Walk the chain from the file to the entitlement to the device test, and the failure points reveal themselves quickly.
Once the plumbing works, the question shifts from "does the link open" to "which campaign sent this user." That is the more valuable problem to solve.
---
## Which MMP SDK Supports Your Framework? React Native, Expo, Flutter, iOS & Cordova Compared
URL: https://www.airbridge.io/en/blog/mmp-sdk-framework-selection-guide
Language: English
Published: 2026-06-04
Author: Luke
You picked a cross-platform framework and shipped your app. Now you need to connect a mobile measurement partner, and the first question is: does the MMP you want actually have an official SDK plugin for your framework?
Most developers discover this gap after committing to a framework. A Flutter app cannot drop in a React Native npm package. Expo Go blocks native modules entirely. **The right time to check MMP SDK compatibility is before you commit to a framework**, not after your first failed integration attempt.
This guide answers which MMP SDK supports React Native, Expo, Flutter, iOS, and Cordova. For each framework, you will find which major MMPs offer official support, how the integration model works, and what limitations to expect.
**Key Takeaways**
- **All major MMPs support iOS native, Android native, React Native, and Flutter with official SDKs.** Coverage is consistent across AppsFlyer, Adjust, Airbridge, Singular, and Branch for these four frameworks.
- **Expo is the most common source of SDK integration confusion.** Expo Go does not support native modules. MMP SDKs require EAS Build or Expo prebuild to work.
- **React Native leads subscription revenue performance.** Across [75,000+ subscription apps](https://www.revenuecat.com/state-of-subscription-apps-2025/), React Native generates a median Y1 LTV of **$22.80** per payer, versus Flutter's $13.81 and native's $16.15 (RevenueCat, State of Subscription Apps 2025).
- **Cordova plugins typically receive less active maintenance** than native, React Native, or Flutter plugins. Verify the last release date before committing.
- **S2S (server-to-server) integration is a framework-agnostic fallback.** If subscription events live on your backend, S2S bypasses the client SDK entirely.
## Why Your Framework Determines MMP SDK Compatibility
A mobile measurement partner (MMP) SDK is, at its core, a native iOS or Android library. It relies on OS-level APIs for device identification (IDFA on iOS, GAID on Android), app lifecycle events, and network calls. None of these can be made from a pure JavaScript or Dart runtime.
That is why each framework needs a different integration approach, and why not all frameworks behave the same when you try to add an MMP SDK.
| Framework | Language | How It Reaches Native Code | MMP SDK Method |
| --- | --- | --- | --- |
| iOS / Android Native | Swift / Kotlin | Direct OS access | Native library (CocoaPods / Gradle) |
| React Native | JavaScript | JSI bridge | npm plugin |
| Expo | JavaScript | JSI bridge (EAS Build only) | Config plugin + EAS Build |
| Flutter | Dart | Platform channels | pub.dev package |
| Cordova / Ionic | HTML + JS | WebView JS bridge | Cordova plugin |
The key rule: **if your framework compiles to native code** (React Native via EAS, Flutter), the SDK wrapper can call native methods and you get full feature access. **If it runs in a managed sandbox** (Expo Go, some Cordova setups), the native module layer is missing and the SDK will fail silently or throw an error.
## How Each Framework Integrates an MMP SDK
[Screenshot 2026-06-04 at 16.48.13.png]
### 1\. iOS and Android Native: Direct SDK Integration, No Bridging Needed
Native apps connect to MMP SDKs directly through the platform's standard package managers.
- **iOS**: CocoaPods or Swift Package Manager
- **Android**: Gradle dependency
- **Auto-linking**: Not needed; direct library integration
- **SDK access**: Full feature set, earliest access to new SDK releases
This is the baseline all other frameworks abstract on top of.
### 2\. React Native: npm Plugins via the JSI Bridge
React Native communicates with native iOS and Android code through the JavaScript Interface (JSI). MMP providers ship React Native plugins as npm packages that wrap their native SDKs.
- **Install command**: `npm install [plugin]`, then `npx pod-install` (iOS)
- **Auto-linking**: Enabled by default in React Native 0.60+
- **SDK access**: Full feature set on both iOS and Android
- **All 5 major MMPs**: Official npm plugins maintained and versioned
### 3\. Expo: Full SDK Support Requires EAS Build, Not Expo Go
Expo is the officially recommended starting point for new React Native projects. It adds managed build tooling and OTA updates. The critical limitation: **Expo Go is a pre-built runtime that does not include arbitrary native modules.** Any attempt to initialize an MMP SDK in Expo Go results in a "native module not found" error.
To use an MMP SDK in an Expo project, you have two paths:
1. **EAS Build**: Configure the MMP config plugin in your `app.json` file. EAS then compiles a custom development client with the native SDK bundled in.
2. **Expo prebuild**: Run `npx expo prebuild` to generate the native `ios/` and `android/` directories locally, then integrate the MMP SDK as you would in a bare React Native project.
Most major MMP providers have dedicated Expo config plugins. See the provider's [Expo SDK documentation](https://help.airbridge.io/en/developers/expo-sdk-v4) for setup steps.
### 4\. Flutter: Clean pub.dev Integration with a Version Requirement
Flutter compiles Dart to native ARM code. MMP providers publish Flutter plugins on pub.dev that expose native SDK methods through Dart APIs.
- **Install command**: `flutter pub add [plugin_name]`
- **SDK access**: Full feature set on iOS and Android
- **Key requirement**: Most current plugins require **Flutter 1.20+ and Dart 2.12+** (null safety)
- **All 5 major MMPs**: Official pub.dev packages, actively maintained
Legacy Flutter projects running older Dart versions need to migrate before they can use current MMP plugin versions.
### 5\. Cordova and Ionic: Official Plugins with Active Maintenance Caveats
Cordova wraps a web app in a native WebView with JavaScript bridges to device APIs. MMP providers publish plugins installable via `cordova plugin add [plugin]`. All five major providers have official Cordova plugins, but these tend to receive fewer updates than their React Native and Flutter counterparts.
- **SDK access**: Functional, but check version compatibility with current native SDKs
- **Key risk**: Plugins that have not been updated in 12 to 18 months may not support newer iOS attribution APIs such as [SKAdNetwork](https://www.airbridge.io/en/glossary/skadnetwork) 4.0
- **Ionic users**: Verify the plugin's last release date before integrating
## MMP SDK Compatibility Matrix: All Major Providers by Framework
The table below covers the five major [mobile measurement partners (MMPs)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) and is based on each provider's official developer documentation (reviewed 2026).
| Framework | AppsFlyer | Adjust | Airbridge | Singular | Branch |
| --- | --- | --- | --- | --- | --- |
| iOS Native | Official | Official | Official | Official | Official |
| Android Native | Official | Official | Official | Official | Official |
| React Native | Official npm | Official npm | Official npm (v4) | Official npm | Official npm |
| Expo | Via RN + config plugin | EAS Build config plugin | Official Expo SDK (v4) | Via RN + config | Official Expo guide |
| Flutter | Official pub.dev | Official pub.dev | Official pub.dev (v4) | Official pub.dev | Official pub.dev |
| Cordova / Ionic | Official plugin | Official plugin | Official Cordova-Ionic SDK | Official (v1.8+) | Official plugin |
**Note on Expo integration approach:** Airbridge publishes a dedicated `airbridge-expo-sdk` npm package, but it requires `airbridge-react-native-sdk` as a mandatory peer dependency and uses a config plugin architecture: the same mechanism as "Via RN + config plugin." For other providers, "Via RN + config plugin" means they use their React Native package with an Expo config plugin wrapper. The underlying approach and EAS Build / prebuild flow is equivalent across all providers; the difference is packaging convenience, not architecture.
## Common MMP SDK Integration Mistakes to Avoid
### 1\. Initializing an MMP SDK Inside Expo Go
**What you see**: A "native module not found" error at SDK initialization, or silent failure with no events tracked.
**Why it happens**: Expo Go is a sandboxed environment. It does not compile arbitrary native modules. The MMP SDK relies on native code that simply is not there.
**Fix**:
1. Run `npx expo prebuild` to generate native project directories, or set up EAS Build in your project.
2. Configure the MMP config plugin in your `app.json`.
3. Build and install a custom development client that includes the native SDK.
Budget time for this: initial EAS Build setup typically takes 2 to 4 hours for teams unfamiliar with it, covering Apple/Google credentials, EAS profile configuration, and a first successful preview build.
### 2\. Ignoring Flutter Null Safety Version Requirements
**What you see**: Package resolution errors or compilation failures when running `flutter pub get`.
**Why it happens**: Most current Flutter MMP plugins require Dart 2.12+ for sound null safety. Pre-null-safety projects (built before 2021 and not yet migrated) are incompatible with current plugin versions.
**Fix**: Migrate your project to null safety before integrating. If migration is not immediately feasible, ask your MMP provider whether they maintain a legacy-compatible branch.
### 3\. Committing to a Stale Cordova Plugin
**What you see**: Attribution events fail silently on iOS, especially for newer installs that go through SKAdNetwork 4.0.
**Why it happens**: A Cordova plugin that has not been updated in 12 to 18 months may not support newer native attribution APIs, even if it initializes without error.
**Fix**: Before integrating, check the plugin's GitHub repository for three signals: the last commit within the past 6 months, open issues receiving responses within 30 days, and a named maintainer in the README. If the plugin fails any of these, contact your MMP provider to confirm active support or ask whether S2S is the recommended path.
### 4\. When to Use S2S Instead of a Client SDK
S2S (server-to-server) integration sends subscription events directly from your backend to the MMP's API, bypassing the client SDK layer. This is relevant when:
- Subscription state is managed server-side (e.g., via RevenueCat or Adapty)
- Your client framework has a SDK compatibility gap
- You want subscription events to fire regardless of whether the user has the app open
You still need the client SDK for install attribution on the device. But the subscription event pipeline (Start Trial, Subscribe, Unsubscribe) can route through S2S independent of framework.
## Subscription Revenue by Framework: React Native vs Flutter vs Native
SDK compatibility is a developer concern. Framework selection also has a business dimension.
According to the [RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/) report (75,000+ apps analyzed), subscription revenue performance differs measurably across frameworks:
| Framework | D35 Conversion (Median) | D60 Revenue per Install | Y1 LTV per Payer (Median) |
| --- | --- | --- | --- |
| React Native | 2.2% | $0.44 | $22.80 |
| Native iOS/Android | 1.9% | $0.27 | $16.15 |
| Flutter | 1.8% | $0.26 | $13.81 |
| Other (incl. Cordova/Ionic) | 1.4% | $0.29 | $16.38 |
_Note: The native iOS/Android D35 median (1.9%) matches the SOSA 2025 global median because native apps form the majority reference population. The per-framework cut confirms this is a real framework-level result, not a rounding artifact._
React Native leads across every metric. At the 90th percentile, the gap is sharper still: React Native apps reach a D60 revenue per install of **$3.54**, compared to $1.53 for native iOS/Android apps. The "Other" bucket, where most Cordova and Ionic apps land, trails on all three metrics, reinforcing the recommendation to avoid Cordova for new subscription apps in 2026.
One structural factor behind React Native's lead: the combination of Expo and AI-assisted coding tools has driven a sharp increase in React Native adoption since 2025, attracting more product-focused development teams to the ecosystem (RevenueCat, State of Subscription Apps 2025, §7).
These figures reflect the full population on each framework and include differences in product quality and marketing spend, not just framework effects. If you are choosing between React Native and Flutter for a subscription app and the decision is otherwise close, the revenue data gives React Native a further edge.
## How to Choose the Right Framework for Your MMP SDK
### If You Have Not Chosen a Framework Yet
| Your Situation | Best Choice | Reason |
| --- | --- | --- |
| JavaScript/React team, new project | React Native + Expo (EAS Build) | Broadest MMP plugin ecosystem, strongest revenue benchmarks |
| Dart team or pixel-perfect UI needed | Flutter | Official pub.dev support from all 5 major MMPs, clean integration |
| Single-platform, native-experienced team | iOS or Android native | No bridging overhead, earliest access to new SDK features |
| Web-first team, new project in 2026 | Capacitor (not Cordova) | Capacitor, by the Ionic team, is the spiritual successor to Cordova. It uses native iOS/Android projects directly instead of the WebView bridge, giving it better native module access and a more actively maintained MMP plugin ecosystem than Cordova. |
### If You Are Already on a Framework
- **On Expo without EAS Build configured**: Set this up before starting MMP integration. It is the single most common blocker.
- **On React Native**: Check that your MMP provider's npm plugin version is compatible with your React Native version.
- **On Flutter**: Confirm your Dart version is 2.12+ before running `flutter pub add`.
- **On Cordova**: Check the MMP plugin's last GitHub commit and open issues. Ask your provider if S2S is available as a fallback.
- **Already integrated and seeing missing events**: Verify the SDK is initialized after native code is compiled, not in a managed sandbox.
## Start Measuring Subscriptions Across Any Framework
Once your SDK is integrated and subscription events are flowing, the next question is which MMP to use for subscription app revenue measurement specifically.
Airbridge Core Plan has official SDK support for every framework covered in this guide:
| Framework | Airbridge SDK | Version |
| --- | --- | --- |
| iOS Native | Official | v4 |
| Android Native | Official | v4 |
| React Native | Official npm | v4 |
| Expo | Official Expo SDK | v4 (EAS / prebuild) |
| Flutter | Official pub.dev | v4 (Flutter 1.20+, Dart 2.12+) |
| Cordova / Ionic | Official Cordova-Ionic SDK | Latest |
Core Plan includes Airbridge's full library of 25 standard events, covering e-commerce, subscriptions, gaming, and user lifecycle. App installs are automatically tracked by the SDK. Key events relevant to subscription apps include Start Trial, Subscribe, Unsubscribe, and Order Complete.
Attribution covers Meta, Google, Apple Search Ads, and TikTok for Business, the four major [self-attributing networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network) supported by Airbridge Core Plan and the dominant paid UA channels across most app categories.
Core Plan supports a maximum of two third-party integrations and does not include raw data export or advanced reporting features, to reduce setup complexity for early-stage teams.
---
## Deferred Deeplink After iOS 17: How to Fix Your Match Rate
URL: https://www.airbridge.io/en/blog/deferred-deeplink-post-idfa-accuracy
Language: English
Published: 2026-06-04
Author: Luke
Your deferred deeplink match rate dropped and you did not misconfigure anything. The ground shifted under you.
Apple's App Tracking Transparency (ATT) framework, the iOS 17 privacy manifest enforcement (May 2024), and the fundamental architecture of SKAdNetwork have together dismantled the IDFA-based matching that made deferred deep linking near-perfectly accurate. Today, the majority of iOS users open your app without an IDFA available, which means the majority of every campaign's installs fall back to probabilistic methods, with real accuracy costs.
**Key Takeaways**
- **Deferred deep linking works by preserving ad click intent through the install journey.** When IDFA was available, deterministic matching made this near-100% reliable. Post-ATT, the same flow works only for the subset of users who grant consent.
- **Three compounding changes broke the pre-ATT system:** ATT consent requirements (2021), iOS 17 fingerprinting ban (enforced May 2024), and SKAdNetwork's architecture, which carries no deep link parameters at all.
- **Probabilistic matching accuracy ranges from** [**70-90% in controlled conditions**](https://linkrunner.io/blog/how-deferred-deep-linking-can-impact-attribution-accuracy-\(and-best-practices\)) and can fall lower on shared networks, VPNs, or when iCloud Private Relay is active.
- **Your ATT pre-prompt strategy is the single highest-leverage fix available.** How and when you ask for consent directly determines what share of your users receive near-perfect routing.
- **A layered matching stack is the only way to approach pre-ATT coverage.** No single method covers all users, and no single fix restores the pre-2021 baseline.
## What Is Deferred Deep Linking and Why Did IDFA Make It Reliable?
**Deferred deep linking** is the mechanism that routes a new user to a specific in-app destination after they install an app. The "deferred" part means the routing intent survives the gap between clicking an ad and opening the app for the first time, even when the user has to visit the App Store in between.
Before ATT, the flow worked in four deterministic steps:
1. **User clicks an ad.** The click URL captures the IDFA, Apple's device-level identifier, alongside campaign parameters (ad ID, creative, destination URL).
2. **User is redirected to the App Store.** The IDFA and deep link context are stored server-side by the deep linking provider.
3. **User installs the app and opens it.** The SDK fires a first-open event and queries the server for stored context.
4. **Server matches the IDFA.** The provider retrieves the stored context and routes the user to the intended destination: a specific product page, a trial offer, or a campaign-specific onboarding flow.
This was a deterministic match: same device identifier, same user, same session. Near-100% accuracy at any traffic volume.
> **50%+**
>
> Average conversion rate increase from deferred deep linking vs. home screen drops, according to [Moburst](https://www.moburst.com/blog/how-app-deep-linking-helps-you-improve-your-conversions/). That lift degrades directly when match rates fall.
[Screenshot 2026-06-04 at 16.35.28.png]
## Why Your Deferred Deeplink Match Rate Dropped: 3 Root Causes
### 1\. App Tracking Transparency: Why Most iOS Users Don't Share Their IDFA
In April 2021, Apple's App Tracking Transparency (ATT) framework required apps to explicitly request user consent before accessing the IDFA. When shown the native system prompt without context, the majority of iOS users decline.
The impact on your matching stack is immediate:
- **Users who grant consent:** Eligible for deterministic, near-100% accurate matching
- **Users who decline:** Fall back to probabilistic methods, with accuracy of 70-90% in controlled conditions and meaningfully lower on shared networks or with iCloud Private Relay active (Apple's Safari IP-masking feature, which hides device IP addresses from matching systems)
Apps in privacy-sensitive categories (finance, health, productivity) tend to see lower consent rates than gaming or entertainment apps. Users associate data tracking with higher perceived risk when an app handles sensitive information. Your fallback exposure depends directly on where your app falls in that spectrum.
### 2\. iOS 17 Privacy Manifests: Apple Closes the Fingerprinting Loophole
After ATT limited IDFA availability, many deep linking vendors shifted to **device fingerprinting**: combining signals like IP address, screen resolution, OS version, and timezone to probabilistically identify a user across the click-to-install journey.
iOS 17 closed this loophole in two documented steps:
- [**WWDC 2023 announcement**](https://developer.apple.com/videos/play/wwdc2023/10060/)**:** Apple declared that SDKs using APIs capable of fingerprinting must declare their usage in a privacy manifest. Fingerprinting for tracking is explicitly prohibited regardless of ATT consent status.
- **May 1, 2024 enforcement:** Apps submitted without compliant privacy manifests began receiving rejections, confirmed by [Apple's developer news](https://developer.apple.com/news/?id=pvszzano) and [Bitrise's enforcement analysis](https://bitrise.io/blog/post/enforcement-of-apple-privacy-manifest-starting-from-may-1-2024).
The practical consequences for deferred deeplink iOS 17 implementations:
- Vendors relying on undisclosed fingerprinting signals had to narrow their matching scope
- Compliant probabilistic matching now operates within a short time window using only IP address and limited device type
- Accuracy drops sharply on shared IPs (corporate networks, public WiFi) and disappears entirely for iCloud Private Relay users
### 3\. SKAdNetwork Postbacks Carry No Deep Link Parameters
[SKAdNetwork (SKAN)](https://www.airbridge.io/en/glossary/skadnetwork) is frequently misunderstood as a solution to the deep linking problem. It is not. SKAN is Apple's privacy-preserving **attribution framework**. It tells you which campaign drove an install. [Apple's developer documentation](https://developer.apple.com/documentation/storekit/skadnetwork) is explicit: SKAN postbacks are aggregate and parameterless by design. There is no mechanism for passing a destination URL, a product ID, or any session context through SKAN.
| | SKAdNetwork | Deferred Deep Linking |
| --- | --- | --- |
| Purpose | Campaign attribution | Post-install user routing |
| Data passed | Campaign ID, conversion value | Destination URL, session context |
| Privacy model | Aggregate (no user-level data) | Per-session matching |
| Fixes match rate? | No | Yes, when configured correctly |
SKAN 4.0 (iOS 16.1+) added more postback tiers and time windows. The routing architecture is unchanged. Fixing your SKAN configuration has **zero effect** on your deferred deeplink accuracy.
## Deferred Deeplink Match Rate by Method: 2025 Accuracy Breakdown
Here is what the deferred deeplink iOS 17 and later matching landscape looks like across all methods still available on iOS:
| Matching Method | Accuracy | Privacy Compliant | Key Limitation |
| --- | --- | --- | --- |
| Deterministic (IDFA) | Near 100% | Yes, with ATT consent | Only for users who opt in |
| Probabilistic (IP + device) | 70-90% controlled, lower in practice | Yes, within Apple guidelines | Falls on shared networks, VPNs, iCloud Private Relay |
| Clipboard/pasteboard | Near 100% | Yes, entirely on-device | Requires clipboard permission prompt |
| First-party link context | Near 100% | Yes | Limited to owned channels (email, QR, push) |
**The core problem:** No single method covers all users. An app relying only on deterministic matching misses every user who declined ATT. An app relying only on probabilistic matching accepts a 10-30% error rate with no visibility into which users were misrouted.
[Screenshot 2026-06-04 at 16.36.05.png]
## 4 Ways to Recover Deferred Deeplink Accuracy Without IDFA
### 1\. Improve Your ATT Opt-In Rate with Pre-Permission Screens
Every additional user who grants ATT consent is one more install with near-100% routing accuracy. Growing your consenting pool is the highest-leverage fix in the stack.
When shown the native ATT prompt without context, most users decline by default. The solution is a custom pre-permission screen shown before the system dialog, giving you full control over the message, design, and timing.
According to [Purchasely's research on ATT optimization](https://www.purchasely.com/blog/att-opt-in-rates-in-2025-and-how-to-increase-them), the apps improving consent rates in 2025 treat it as a product experience, not a privacy checkbox. What works:
- Show the prompt after users complete a meaningful action, not at cold launch
- Explain a specific, tangible benefit ("We use this to show you the promotion you came for")
- Use a full-screen design rather than a modal overlay
- A/B test prompt copy, timing, and design with the same rigor you apply to your paywall
**The math:** On a campaign delivering 10,000 iOS installs per month, every 10 percentage points of improvement in consent rate moves 1,000 more users from probabilistic fallback to deterministic matching. At a subscription app trial-to-paid median of [34.8%](https://www.revenuecat.com/state-of-subscription-apps-2025/), that shift carries meaningful revenue impact.
### 2\. Shorten Your Probabilistic Lookback Window
For users who decline ATT, probabilistic matching is the primary fallback. Its accuracy depends heavily on the lookback window you configure.
**Why window length matters:**
- A **wide window (24 hours)** means many unrelated users share the same IP during the match period, creating false positives
- A **tight window (10-15 minutes)** limits the pool to users who clicked and installed in rapid succession, dramatically reducing false matches
- Most install-to-open sessions happen within 15-30 minutes of install (varies by app category)
**What to do:**
1. Check your deep linking provider's settings for lookback window configuration
2. Pull your install-to-open latency data from your provider's reporting dashboard
3. Set your probabilistic window to the shortest window that captures 90%+ of your historical install-to-open distribution
4. Monitor match rate and re-adjust quarterly as traffic patterns change
### 3\. Add Clipboard-Based Matching for Non-IDFA Users
Clipboard-based (pasteboard) matching is the most accurate privacy-compliant fallback for users who decline ATT.
**How it works:**
1. Before sending the user to the App Store, the deep link context is **copied to the device clipboard** via the browser, server-side
2. After install, when the app opens for the first time, the SDK **reads the clipboard** and retrieves the stored context
3. The match is made **entirely on-device**, with no cross-app identifier exchange
**Key trade-off:** iOS shows a system prompt asking if the app can read the clipboard. Some users will decline.
**Best use cases:**
- High-value campaign audiences (loyalty programs, referral rewards)
- Limited-time offers where context loss is especially costly
- Campaigns where you can explain the personalization benefit before install
### 4\. Encode Session Context Directly in Owned-Channel Links
For campaigns you fully control, bypass device-level matching entirely by embedding the routing context directly in the link.
**How it works:** Append session parameters (product ID, campaign ID, destination path) to the deep link URL. When the user installs from that link, the referrer data arrives through the link itself, not through any device matching process.
**Works for:**
- Email campaigns linking to a specific product or offer
- QR codes in physical locations (packaging, print, out-of-home)
- Push notifications to re-engage existing web visitors
- Web-to-app flows from your own landing pages
**Limitation:** This does not apply to paid social or search ads. For those channels, use methods 1-3.
## SKAdNetwork vs. Deferred Deep Linking: Two Different Problems
The most common misconfiguration is treating SKAN reporting and deferred deep linking as the same issue, and solving one while ignoring the other.
| | SKAdNetwork (SKAN) | Deferred Deep Linking |
| --- | --- | --- |
| Answers | "Which campaign drove this install?" | "Where should this user land on first open?" |
| Data mechanism | Aggregate postbacks to ad networks | Per-session matching (IDFA, probabilistic, clipboard) |
| Privacy constraint | Parameterless by Apple design | Governed by ATT and fingerprinting rules |
| Configured via | Ad network + attribution settings | Deep linking SDK + provider settings |
| Fixing SKAN helps match rate? | No | N/A |
For a full comparison of [deferred deeplinks vs. standard deep links](https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link) and how the post-ATT environment affects each type, the distinction between routing and attribution is worth understanding fully.
**The practical risk:** A campaign could be perfectly attributed in SKAN while its deferred deeplinks fail for a meaningful share of users, easily 30% or more during high-traffic periods when probabilistic windows are configured too wide. Both problems require separate fixes.
## How a Low Deeplink Match Rate Costs Subscription Revenue
Deferred deeplink accuracy is not a developer metric. It is a paid UA performance metric with a direct line to subscription revenue.
**The failure chain, step by step:**
1. User clicks an ad featuring a specific trial offer
2. Deferred deep link match fails (probabilistic misroute or no match)
3. User lands on the home screen instead of the offer page
4. User does not find the trial offer on their own
5. Trial start rate drops for that campaign cohort
6. Cost per trial rises; ROAS declines
> **34.8%**
>
> Median trial-to-paid conversion rate across 75,000+ subscription apps, per [RevenueCat's 2025 State of Subscription Apps](https://www.revenuecat.com/state-of-subscription-apps-2025/). Every percentage point of match rate recovery works against this baseline directly.
Research from [Braze](https://www.braze.com/resources/articles/deep-linking-can-double-conversion-increase-retention-boost-engagement) confirms that deep linking can double conversion rates on specific campaigns. For subscription apps at the 34.8% trial-to-paid median, closing even a portion of the match rate gap compresses your cost per subscriber meaningfully.
**The layered stack, in order of priority:**
1. **Optimize your ATT pre-prompt** to grow your deterministic pool (highest leverage, directly expands IDFA-eligible installs)
2. **Tighten the probabilistic lookback window** (quick configuration fix, improves accuracy for every non-IDFA install)
3. **Add clipboard matching as a fallback** (best accuracy for non-IDFA users willing to allow clipboard access)
4. **Embed context in all owned-channel links** (covers email, QR, push with deterministic accuracy at no matching cost)
No single method restores the pre-2021 baseline. The combination narrows the gap enough to measure directly in your trial funnel.
---
## How to Test Airbridge Deeplinks: A Step-by-Step Dashboard Debugging Guide
URL: https://www.airbridge.io/en/blog/airbridge-deeplink-testing-debugging-guide
Language: English
Published: 2026-06-04
Author: Luke
You set up an Airbridge tracking link with a deeplink destination, hit "Create," and moved on. But clicking a link once on your own phone is not a test: it is a guess. A broken deeplink silently fails for every user who does not match your exact device and browser combination. The damage shows up later, in campaigns where users install the app and land on the home screen instead of the product page your ad was promoting.
Airbridge's [deep linking platform](https://www.airbridge.io/en/product/deep-linking) includes a dedicated Testing Console that walks you through four structured test scenarios before you spend a dollar on traffic. Each test ends with a direct link to the Actuals Report so you can confirm the Deeplink Open event fired. This guide shows you where the console lives, how to run every test, and what each common failure looks like on the inside.
**Key Takeaways**
- **An Airbridge deeplink is a URL that navigates users to a specific in-app destination.** Airbridge supports three methods: URL Scheme (`appScheme://`), App Links (Android, `https://`), and Universal Links (iOS, `https://`). The platform selects the best method automatically based on OS, browser, and app installation status.
- **The Testing Console lives at Settings > Testing Console > Deep Linking.** It is separate from the tracking link builder and requires two pre-conditions plus a readiness check before any test can run.
- **Up to four tests cover two scenarios:** URL Scheme and Universal Link/App Link (always required for app-installed users), Custom Domain (only if you use a custom domain), and Deferred Deep Link (for users without the app). Most setups run three tests total.
- **The Actuals Report is where success is confirmed.** Look for the Deeplink Open event after completing each scenario. A missing event usually means ATT consent was not given on the test device.
- **iOS has two silent failure modes** that pass the visual test but produce no event: ATT/Extended Privacy Control (EPC) consent not granted, and a missing path segment in the URL scheme deep link.
## What to Register in Airbridge Before Running a Deeplink Test
### 1\. Register Deep Link Info at Tracking Link > Deep Links
Go to **\[Tracking Link\] > \[Deep Links\]** in the Airbridge dashboard. This is where you give Airbridge the technical identifiers it needs to route users correctly on each platform.
For Android: URL scheme, package name, and sha256\_cert\_fingerprints. For iOS: three separate fields in the Airbridge dashboard: iOS URI Scheme, Bundle ID, and Team ID (App ID Prefix). All three are available from the [Apple Developer Dashboard](https://developer.apple.com/account/resources) under App Identifiers.
Airbridge supports [three deep linking methods: URL scheme deep links, App Links (Android), and Universal Links (iOS)](https://help.airbridge.io/en/guides/understanding-deep-link). The platform selects the most appropriate method automatically based on the OS, browser, and app installation status. All three need to be registered to cover every user path. For a detailed walkthrough, see [Apply Deep Links to Tracking Links](https://help.airbridge.io/en/guides/retargeting-with-deep-links).
### 2\. Set Up the Fallback Path at Settings > App Settings
Users who click a deeplink without the app installed need somewhere to go. Under **\[Settings\] > \[App Settings\]**, fill in the "Platform" section with your App Store and Google Play URLs. Without this, deferred deep linking has no redirect destination after the user installs the app.
### 3\. Check the Platform Readiness Status Before Testing
Once you enter the [Testing Console](https://help.airbridge.io/en/guides/deep-linking-test), each platform displays either **"Ready for testing"** or **"Info incomplete."** Click the tooltip next to the platform name to see exactly which fields are missing. If you just completed setup in a separate browser tab, refresh the console tab before reading the status. The console does not auto-update.
This guide assumes your developer has already implemented the `trackDeeplink` SDK call. If not, confirm implementation against the [Airbridge developer guide](https://help.airbridge.io/en/deeplink-developers/sdk-quickstart) first, since the Deeplink Open event will not appear in reports without it.
## How to Access the Airbridge Deeplink Testing Console
### 1\. Navigate to Settings > Testing Console > Deep Linking
Airbridge deeplink testing follows a step-by-step dashboard flow: pre-check setup, run the Testing Console, then verify in the Actuals Report. From the dashboard, go to **\[Settings\] > \[Testing Console\]** and select the **\[Deep Linking\]** tab. This is the only dedicated testing environment for verifying deeplink behavior.
### 2\. Enter and Encode Your URL Scheme Deep Link
Enter the URL scheme deep link you want to test, for example: `myapp://productpage?id=123`. Two requirements:
- The input field accepts **URL scheme deep links only** (not HTTPS Universal Links or App Links). Airbridge uses this to test all three methods indirectly.
- **URL-encode the deep link before entering it.** Special characters, spaces, and non-standard punctuation must be percent-encoded. An unencoded URL breaks the standard format and the deeplink will not fire.
Use the switch icon on the right of the input field to toggle between free-text input and a path/parameter builder if that makes encoding easier.
## Why Your Airbridge Deeplink Test Is Failing: 5 Root Causes and Fixes
Before running the tests, knowing the most common failure points helps you spot and fix issues as they occur. These five causes account for the majority of [Airbridge deeplink](https://www.airbridge.io/en/product/deep-linking) test failures.
### 1\. "Info Incomplete" Status in the Testing Console
At least one required field is missing for that platform. Click the tooltip to identify it. The most commonly missed items: **sha256\_cert\_fingerprints** for Android App Links, and the **iOS App ID** (App ID Prefix + Bundle ID) for Universal Links. Both require developer input. Fix these at Tracking Link > Deep Links, then return to the console and refresh.
### 2\. Deeplink Open Event Not Appearing on iOS
Extended Privacy Control (EPC) is on by default in Airbridge. Without ATT consent on the test device, the Deeplink Open event fails silently on iOS. The link may have opened the correct in-app page: the event collection is what failed, not the deeplink itself. Check both the EPC settings in Airbridge and the ATT status on your test device before concluding the deeplink is broken. See the [iOS deeplink FAQ](https://help.airbridge.io/en/guides/measuring-performance-with-deep-links-faq) for more detail.
### 3\. Deferred Deep Link Not Working on iOS
If you are on the **Airbridge DeepLink Plan**, set `setAutoDetermineTrackingAuthorizationTimeout` to 0 seconds in the ATT prompt configuration of the Airbridge iOS SDK. The DeepLink Plan does not use IDFA for attribution, so the ATT wait window is unnecessary for deferred deep linking.
If you are on **Core Plan or Growth Plan**, do not change this value without consulting your developer. The default 30-second window exists to collect IDFA alongside the install event — setting it to 0 skips that window and can reduce attribution quality.
### 4\. Deep Link URL Missing a Path on iOS
This is the most common silent failure on iOS ad placements, particularly Instagram. The URL scheme deep link registered in Airbridge must include a path segment between the scheme and any query parameters ([iOS deeplink FAQ](https://help.airbridge.io/en/guides/measuring-performance-with-deep-links-faq#the-deep-linking-does-not-work-as-intended-on-ios)).
Broken: `myapp://?type=blog&content=promo`
Working: `myapp://productpage?type=blog&content=promo`
Without the path, iOS deep linking fails on certain browsers and ad environments. This is a requirement enforced by iOS, not Airbridge.
### 5\. Deep Link URL Not Encoded
URL-encode the deep link before entering it into the Testing Console input field. If the link contains spaces, special characters, or parameters with values containing `&` or `=`, those characters must be percent-encoded. An unencoded URL breaks the standard format before Airbridge even processes it.
Use the path/parameter builder mode (toggle via the switch icon on the input field) if you want Airbridge to handle the encoding automatically.
> **Stopover Airpage: a fix for iOS first-click redirects from non-Safari apps.** If iOS users with the app installed are being sent to the App Store when clicking from Instagram, Line, or other third-party apps, enable the **Stopover Airpage** option when creating the tracking link. This is a general deeplink reliability feature available on any tracking link with a deeplink destination — not limited to re-engagement. It is particularly recommended for re-engagement campaigns, but it also benefits UA campaigns where users click from in-app browsers.
>
> Learn more in [Additional Options for Deep Linking](https://help.airbridge.io/en/guides/deeplink-option-setting).
## How to Test Airbridge Deeplinks When the App Is Already Installed
This is the most important scenario for re-engagement campaigns. The user has your app, clicks a link, and should land on a specific in-app page, not the home screen.
**Before each test:** Use a physical mobile device with the app installed and opened at least once. Set the default browser to Chrome or Safari. The testing console only supports those two browsers. Other browsers will produce unreliable results regardless of your configuration.
| Test | What It Checks | When to Run |
| --- | --- | --- |
| URL Scheme Test | URL scheme deeplink opens the correct in-app location | Always |
| Universal Link / App Link Test | HTTPS-based deep linking method works on that OS | Always |
| Custom Domain Test | Custom short link domain (e.g., links.yourbrand.com instead of Airbridge's default) triggers deep linking correctly | Only if using a custom domain |
For each test: click the test name, scan the QR code on your test device (or use the URL shown below the QR code), then tap **Trigger deep linking** on the testing page. The console runs automatically across five items ([Airbridge Testing Guide](https://help.airbridge.io/en/guides/deep-linking-test)). If any item fails, fix instructions appear inline. Run the test again after fixing.
After all items pass, click **"Check event in the report."** A pre-set Actuals Report opens in a new tab. This is where you confirm the deeplink fired correctly.
## How to Test Deferred Deep Linking When the App Is Not Installed
### 1\. What Deferred Deep Linking Does and Why It Matters for UA
Deferred deep linking covers a critical UA scenario: the user clicks your tracking link, does not have the app, installs it from the App Store or Google Play, then opens the app for the first time. Airbridge routes that user to the specific in-app page from the original ad, not the default home screen.
This matters for conversion. If a user sees your ad for a specific product page, installs the app, and lands on the home screen instead, the continuity of the experience breaks. The in-app conversion that your ad budget was optimized to drive does not happen.
Deferred deep linking has a **1-hour time limit**. If the user installs and opens the app more than 1 hour after clicking the tracking link, Airbridge sends them to the home screen instead of the deeplinked page. Plan your UA creative flows and post-install messaging with this window in mind.
### 2\. How to Run the Deferred Deep Link Test
Use a device with the app **not installed** (delete it if needed). Set Chrome or Safari as the default browser. Scan the QR code and click **Trigger deep linking** on the testing page. The console checks three items automatically.
If you are on the Airbridge DeepLink Plan, your developer should configure `setAutoDetermineTrackingAuthorizationTimeout` to 0 seconds in the iOS SDK. Core Plan and Growth Plan users should leave the default (30 seconds) unless advised otherwise, as this affects IDFA collection for attribution.
[Screenshot 2026-06-04 at 16.15.11.png]
_The Airbridge deeplink testing workflow: pre-check, test, verify._
## How to Confirm Airbridge Deeplink Results in the Actuals Report
### 1\. How to Open the Pre-Set Actuals Report
After every test where all items pass, click **"Check event in the report."** Airbridge opens a pre-configured [Actuals Report](https://help.airbridge.io/en/guides/airbridge-report-overview) in a new tab, already filtered to the deeplink data from your test session.
### 2\. What Deeplink Metrics to Look For in the Report
The key event to confirm is **Deeplink Open**. This fires when the SDK detects the app was launched via a deeplink. The pre-set Actuals Report may also show Click and Install as standard metrics, but Deeplink Open is the confirmation that deep linking worked end-to-end.
If Deeplink Open shows 0 after a completed test, two causes cover most cases.
**Extended Privacy Control (EPC) is enabled by default in Airbridge.** On iOS, if the test device has not responded to the ATT (App Tracking Transparency) prompt, or the user opted out, Deeplink Open event collection fails at the OS level. This is not a configuration error: it is privacy protection enforced by iOS. Before testing on iOS, respond affirmatively to the ATT prompt on the test device.
**The Deeplink Open event requires an SDK call.** Specifically, the `trackDeeplink` function must be called in the SDK when the app opens via a deeplink. If your developer has not implemented the deep linking handlers, the link may route users correctly but produce no event in the report. Confirm SDK implementation against the [Airbridge developer guide](https://help.airbridge.io/en/deeplink-developers/sdk-quickstart) before testing.
## Verify Every Airbridge Deeplink Before Your Campaign Goes Live
A deeplink that looks correct in the dashboard is not a verified deeplink. Run all applicable tests in the Testing Console before any campaign goes live (two or three tests for Scenario 1 depending on your domain setup, plus the deferred test for Scenario 2). Check the Actuals Report for the Deeplink Open event. For iOS deferred deep linking, confirm ATT configuration with your developer before launch.
If you have not yet walked through the Testing Console, start with the official [deeplink testing guide](https://help.airbridge.io/en/guides/deep-linking-test) to build a verified baseline for every link you deploy.
---
## How to Set Up the Airbridge SDK: From Signup to Your First Event in 30 Minutes
URL: https://www.airbridge.io/en/blog/how-to-set-up-airbridge-sdk-first-event
Language: English
Published: 2026-06-02
Author: Luke
Signup took **two minutes**. Now the dashboard is open, the docs are open, and a dozen menu items are staring back at you with zero data behind any of them. This is the moment most setups quietly stall, not because the work is hard, but because the path from "I have an account" to "I can see my first event" is rarely laid out in order.
So here is the order. In the next **30 minutes** you will register your app, install the Airbridge SDK, and watch a real event land in your dashboard. One verified event is the whole goal. Everything else the platform can do sits on top of that single confirmed signal, and you cannot trust any of it until that signal arrives.
**Key Takeaways**
- **The work is three steps, not twelve menus.** Register the app, install the SDK, then fire and verify one event. Skip nothing, and do them in that order.
- **One credential unlocks the SDK.** Your app name and SDK token live under Settings then Tokens, and a single mistyped character is the most common reason a first integration sends nothing.
- **Initialize the SDK before anything else runs.** An event that fires before the SDK wakes up is lost, so initialization belongs in your earliest startup code.
- **Verification, not installation, is the finish line.** A silent SDK looks identical to a broken one. You are done when the event appears in the testing console, not when the code compiles.
- **The dashboard rewards patience after the proof.** Testing tools confirm your event in minutes, but full reports need time to fill before the numbers settle.
## Airbridge SDK Setup: What You'll Have in 30 Minutes
Before touching code, fix the target in your head. You are not building a finished marketing stack today. You are proving one thing: that your app can talk to Airbridge and that you can see the conversation happen.
That proof breaks cleanly into three steps. Each one ends with a checkpoint you can verify before moving on, so you never spend 20 minutes debugging step three when the real problem was a typo in step one.
| Step | Minutes | What you do | Checkpoint |
| --- | --- | --- | --- |
| 1. Register | 0 to 10 | Add your app, set time zone and currency, copy your SDK token | Your app exists in the dashboard |
| 2. Install | 10 to 25 | Add the SDK to your project and initialize it on launch | The app builds and runs with the SDK live |
| 3. Verify | 25 to 30 | Log one standard event and confirm it in the testing console | Your first event is confirmed |
The realistic clock assumes an app you already build and run locally. According to [Airbridge's SDK documentation](https://help.airbridge.io/en/developers/airbridge-sdk-overview), an experienced developer needs roughly **10 minutes** of SDK work per platform, so a single-platform first pass fits inside half an hour with room to spare. A second platform adds about another **10 minutes**, not another 30, because the dashboard side is already done.
[Screenshot 2026-06-02 at 18.14.13.png]
## Step 1: Register Your App in the Dashboard
It is tempting to jump straight into the SDK. Resist it. The first step is pure dashboard work, and skipping it is why people later find their events landing under the wrong app or their revenue totaled in the wrong currency. Spend ten minutes here and the next twenty go smoothly.
### 1\. Add your app to Airbridge
From the dashboard, create a new app. You give it a name and select the platforms it ships on: iOS, Android, web, or some combination. Two things matter at this step:
- **The app name becomes part of your SDK credentials.** Pick something stable and recognizable. Renaming later is possible but creates avoidable friction.
- **Select every platform you actually plan to measure.** You can add more later, but setting them now means the dashboard expects the data and you avoid "why is my iOS data missing" confusion on day two.
If any field is unclear, [Airbridge's guide to adding an app](https://help.airbridge.io/en/guides/register-a-new-app) walks through each one.
### 2\. Set your app time zone and currency
Two settings quietly govern every number you will ever read: the app **time zone** and the **standard currency**. Set both before any data flows.
This is not housekeeping you can defer. Both settings decide how events are bucketed by day and how revenue is converted and totaled. Change them after you have history and your old reports and new reports are measured on different rulers, which corrupts every day-over-day and month-over-month comparison you try to make.
- **Time zone:** use the one your team actually reads reports in, not necessarily where your servers live.
- **Currency:** use the one you report revenue and budget in, so dashboard totals match your finance numbers without mental math.
### 3\. Copy your SDK token from Settings
This is the credential the SDK cannot run without. In the dashboard, go to **Settings** then **Tokens**. You will find your **app name** and your **app SDK token** there. Copy both into a scratch file you will reach for in the next step.
**Watch out:** the initialization call fails quietly or throws if either value is off by a single character. A trailing space copied from the dashboard counts as wrong. A mistyped token is, by a wide margin, the most common reason a first integration sends nothing at all, so treat this copy-paste as a precision step, not a formality.
At this point your app exists inside the dashboard, but your codebase still has no idea it exists. The SDK is the bridge between the two, and that is step two.
## Step 2: Install the Airbridge SDK and Initialize It on Launch
Now you connect your code. There are native SDKs for Android and iOS plus wrappers for the major cross-platform stacks, and the shape of the work is identical across all of them: add the package, then initialize it once with your app name and token.
### 1\. Add the SDK to your project
You do not need a special toolchain. Use whatever you already use to add dependencies.
| Platform | How you add it | Language |
| --- | --- | --- |
| Android | Gradle dependency (Maven) | Kotlin or Java |
| iOS | Swift Package Manager or CocoaPods | Swift or Objective-C |
| React Native | npm or yarn | JS over native |
| Flutter | pub | Dart over native |
| Expo | npm with the config plugin | JS over native |
The [Airbridge SDK Quickstart](https://help.airbridge.io/en/developers/sdk-quickstart) lists the exact package name and current version for each platform, so pull the version string from there rather than guessing.
### 2\. Initialize the SDK on app launch
Initialization is a single call. Where you put that call matters more than the call itself: it has to run before any event could fire, which means your earliest startup path.
On Android, that is your `Application.onCreate()`:
```kotlin
// Application.onCreate()
val option = AirbridgeOptionBuilder("YOUR_APP_NAME", "YOUR_APP_SDK_TOKEN")
.build()
Airbridge.initializeSDK(this, option)
```
On iOS, that is `application(_:didFinishLaunchingWithOptions:)`:
```swift
// AppDelegate
let option = AirbridgeOptionBuilder(name: "YOUR_APP_NAME", token: "YOUR_APP_SDK_TOKEN").build()
Airbridge.initializeSDK(option: option)
```
Replace `YOUR_APP_NAME` and `YOUR_APP_SDK_TOKEN` with the values from Settings then Tokens. Build and run once. If the app launches without crashing, the bridge is standing.
**Why the placement is strict:** the SDK can only record an event after it is awake. If `initializeSDK` runs inside a screen that loads later, every event fired before that screen is simply gone, and you will burn time hunting a "missing event" bug that is really a "too-late initialization" bug.
### 3\. Configure the iOS ATT prompt (iOS only)
If you run paid campaigns on iOS, attribution depends on the device advertising identifier, and that identifier requires user permission through Apple's [App Tracking Transparency](https://developer.apple.com/documentation/apptrackingtransparency) prompt. Two pieces are involved:
1. **Show the ATT prompt** so the user can grant or deny tracking.
2. **Set a collection timeout in the SDK** so the identifier is captured once the user responds. [Airbridge's onboarding guide](https://help.airbridge.io/en/guides/getting-started-with-airbridge) recommends a timeout of **30 seconds** or longer, depending on how quickly your onboarding reaches the prompt.
You can skip this for your very first test event. You cannot skip it before launching iOS, because without it your paid attribution runs blind on Apple devices. For the concept and why it matters, see Airbridge's [App Tracking Transparency glossary entry](https://www.airbridge.io/en/glossary/app-tracking-transparency-att).
**Checkpoint before step three:** the app builds, runs, and does not crash on launch with the SDK initialized. If that holds, you are ready to fire an event.
## Step 3: Track Your First Event and Verify It in the Dashboard
An installed SDK that has never sent an event is indistinguishable from one that is broken. Installation is not the finish line. Verification is. The last five minutes are about firing one deliberate event and watching it arrive.
### 1\. Pick the right first event to track
Core Plan ships with [25 standard events](https://help.airbridge.io/en/guides/airbridge-event) that come with predefined names and schemas. Using a standard event means the naming and structure work is already done, so you log immediately instead of designing an event taxonomy on day one. This is exactly how to set up the Airbridge SDK and fire your first event without getting stuck on schema decisions.
For a subscription app, fire the event that maps to the moment that actually matters. Each event answers a distinct business question:
| Funnel stage | Standard event | Event ID | The question it answers |
| --- | --- | --- | --- |
| Acquisition | Sign-up | airbridge.user.signup | How many installs become accounts? |
| Activation | Complete Tutorial | airbridge.completeTutorial | How many reach the "aha" moment? |
| Trial | Start Trial | airbridge.startTrial | How many enter the paid funnel? |
| Revenue | Subscribe | airbridge.subscribe | How many turn into recurring revenue? |
| Churn | Unsubscribe | airbridge.unsubscribe | Where is revenue leaking back out? |
**Subscribe** is the most meaningful first event for a subscription app, but for a smoke test, any standard event proves the pipe. Most first integrations will not have a working subscription flow on day one, so Sign-up or Complete Tutorial is usually faster to trigger and just as valid for confirming that data flows. Pick whichever you can fire right now.
### 2\. Log the event from a real user action
Send the event with a single call. On Android:
```kotlin
// Fire on the actual subscribe action, or a temporary test button
Airbridge.trackEvent("airbridge.subscribe")
```
Wire that call to a real action in your app. The actual subscribe button is ideal because it tests the exact code path you care about. If that flow is not ready, a temporary test button works fine for verification. Run the app and trigger it.
### 3\. Verify the event in the SDK testing console
Open the SDK testing tools in your dashboard and watch for the event to arrive. Real-time testing surfaces the event within a minute or two, which confirms the full path end to end: your code fired it, the SDK transmitted it, and the dashboard received it. [Airbridge's SDK testing guide](https://help.airbridge.io/en/guides/sdk-testing) shows exactly where to look.
When you see your event in the console, you have hit the goal. The connection is confirmed.
**Set the right expectation for reports:** the testing console shows your event almost immediately, but the polished reports, retention, revenue, and funnel, need time to accumulate before the numbers stabilize. Airbridge's onboarding documentation notes that data collection takes roughly **24 to 48 hours** to settle into trustworthy reporting. An empty Revenue report on day one is normal and means nothing. Judge your setup by the event you just verified in testing, not by a report that has not had time to fill.
### 4\. Troubleshoot a missing first event
Most first-event failures come from a short list of repeatable causes. Work down this table before assuming anything deeper is wrong.
| Symptom | Most likely cause | Fix |
| --- | --- | --- |
| Nothing appears in the testing console | Wrong or mistyped token | Re-copy app name and token from Settings then Tokens, character for character |
| SDK initialized but no events log | Initialization runs too late | Move initializeSDK to your earliest startup code |
| Event shows under a different app | Token belongs to another app record | Match the token in code to the app you are watching |
| iOS shows no device or ad identifier | ATT prompt missing or declined | Add the ATT prompt and set the collection timeout |
| Testing works, but reports are empty | Reports lag behind testing | Wait for data to settle, then re-check |
If you have checked all five and the event still will not appear, the issue is usually environment-specific (a proxy, an ad blocker on a test device, or a build that did not pick up the new dependency). Rebuild clean and test on a real device on a normal network before going deeper.
## After Your First Event: Connect Ad Spend to Subscriptions
A verified event proves the plumbing. It is not why you set up tracking in the first place. You did that to answer the question that decides where your money goes: which of your campaigns produced that subscribe, and which ones only produced installs that never paid?
This is where the rest of the setup earns its keep, and the order continues to matter.
### 1\. What to set up next, in order
- **Connect your ad channels.** Core Plan focuses on the four channels most subscription apps actually spend on: **Meta, Google, Apple Search Ads, and TikTok**. Connecting them lets Airbridge tie each subscriber back to the campaign that drove them.
- **Confirm your attribution settings.** Decide the windows and rules that match how your buying cycle actually works, so credit lands where it belongs.
- **Add your subscription funnel events.** You verified one event. Now instrument Sign-up, Start Trial, Subscribe, and Unsubscribe so the full funnel is visible, not just the endpoint. The [subscription event tracking guide](https://www.airbridge.io/en/blog/subscription-app-event-tracking-setup-guide) covers the complete sequence.
- **Let data accumulate, then read the reports.** After a day or two, the funnel, retention, and revenue reports become trustworthy enough to act on.
### 2\. Why connecting spend to revenue is worth it
The gap between an average subscription app and a strong one is large, and it is mostly a measurement problem. According to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/), the median subscription app converts just **1.9%** of downloads to paying subscribers within 35 days, while the top 10% of apps reach **8.5%**. That is a 4.5x gap, and you cannot close it on instinct. You close it by seeing which channels deliver the users who actually subscribe, then moving budget toward them, which is impossible when your ad spend and your subscription data live in separate systems that never talk.
### 3\. What Core Plan does and does not do
Core Plan is deliberately scoped for early-stage subscription apps, and it is honest about its edges rather than promising everything:
- **It gives you 25 standard events, not custom events.** Custom events require the Growth plan. For most early subscription funnels, the standard set is enough.
- **It connects up to two third-party platforms**, such as RevenueCat or Adapty, not an unlimited list.
- **It covers four self-attributing networks**, the channels above, not every niche ad network.
That narrower scope is the point. Fewer decisions during setup is exactly why your first thirty minutes can stay thirty minutes.
| Setup factor | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Pricing entry | Annual contract, sales call | Pay-as-you-go, 15K free attributed installs |
| Billing integrations | Native, often in higher tiers | RevenueCat and Adapty in the base plan |
| Events | Predefined plus custom | 25 subscription-optimized standard events |
| Onboarding | General-purpose | Subscription-app focused |
## Start Seeing Which Channels Drive Paying Subscribers
The hardest part of any new tool is the first proof that it actually works, and you now have it: a real event, fired from your app, confirmed in your dashboard. Everything else, channels, attribution, and reports, compounds on top of the pipe you just built in half an hour.
The next move is to point that pipe at the question that pays for itself. If you are ready to see which channels actually drive paying subscribers and not just installs, [start free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core), which includes 15K attributed installs.
---
## The 5 Best Adjust Alternatives for 2026
URL: https://www.airbridge.io/en/blog/the-5-best-adjust-alternatives-for-2026
Language: English
Published: 2026-06-02
Author: Harper (Trang Nguyen)
Your Adjust contract is up for renewal. Before you auto-renew, it is worth asking whether this is still the right tool for where your app is going.
Adjust is a capable platform. But "capable" is not the same as "right for your stack at your current stage." Pricing requires a sales call to unlock. Features your subscription app actually needs may sit behind a higher tier. And if your team is managing a trial-to-renewal funnel, a general-purpose measurement platform can track installs accurately while telling you nothing about whether those installs generate revenue.
**Key Takeaways**
- **Pricing transparency varies sharply across platforms.** Adjust, AppsFlyer Enterprise, Singular, and Branch all require custom quotes. Kochava and Airbridge publish at least starter pricing publicly.
- **Subscription funnel depth is the real differentiator.** The right tool connects trial starts, paid conversions, renewals, and churn to the specific channels and creatives that drove them.
- **All five alternatives cover the four major ad channels.** Meta, Google, Apple Search Ads, and TikTok are natively supported across every platform listed here.
- **Contract structure affects total cost at least as much as per-unit pricing.** Know whether you are signing annual or pay-as-you-go before entering any negotiation.
- **The best alternative is not universal.** An enterprise app spending $100K/month on paid UA has different requirements than a 5-person team running its first attribution setup.
## What to Evaluate Before Switching from Adjust
Not every Adjust alternative belongs on your shortlist. These four criteria separate the real contenders from the noise.
### 1\. Pricing transparency and contract flexibility
Adjust uses a custom quote model built around monthly tracked users (MTUs). You need a sales conversation to get a number, and annual contracts are standard. Before putting any alternative on your shortlist, confirm whether pricing is publicly listed, whether pay-as-you-go is available, and what the minimum commitment looks like.
A platform that requires a 12-month contract at an undisclosed price solves your renewal problem without solving your transparency problem. If your paid UA spend varies month-to-month or your team is pre-Series A, prioritize platforms that explicitly offer monthly billing or no-minimum pay-as-you-go. Minimum spend thresholds can create effective annual commitments even on ostensibly "flexible" plans.
### 2\. Subscription funnel measurement depth
Install attribution is table stakes. For subscription apps, the measurement that determines whether paid UA is profitable is further down the funnel: did the user from your Meta campaign start a trial? Did they convert to paid? Are they still subscribed 90 days later?
A mobile measurement platform that stops reporting at the install level cannot connect your ad spend to subscription revenue. When evaluating any alternative, ask specifically whether trial start, first subscription, and renewal events are trackable by channel and creative without custom development work. Ask which events are pre-configured vs. which require your engineering team to define schemas from scratch.
### 3\. Channel coverage and integration breadth
The four major Self-Attributing Networks (Meta, Google, Apple Search Ads, and TikTok) account for the majority of paid UA spend for most subscription apps. Every major mobile measurement platform covers these natively. Differences emerge at the edges: support for smaller ad networks, data warehouse connections (BigQuery, Snowflake, Redshift), and billing platform integrations like RevenueCat and Adapty.
If your tech stack includes RevenueCat, confirm native integration exists before signing. If you run non-standard ad networks beyond the four SANs, verify coverage explicitly. A channel your measurement tool does not cover is a blind spot that grows more expensive as spend scales.
### 4\. Migration support and data continuity
Switching measurement platforms requires re-integrating SDKs, reconfiguring postback connections and attribution window settings, and managing a data gap window before the new platform's stream stabilizes. Ask each vendor how long a standard migration takes for a single iOS/Android app, what support is included at no extra cost, and whether historical attribution data carries over.
Vendors with dedicated migration engineers or structured onboarding programs reduce execution risk significantly. Vendors that hand you documentation and leave you to it require your engineering team to absorb the migration overhead. Factor this into your true cost comparison.
[diagram-subscription-funnel-attribution.webp]
## The 5 Best Adjust Alternatives at a Glance
Here is how the five mobile measurement partners compare on the criteria that matter most at renewal time.
| Platform | Best For | Starting Price | Subscription Events | Contract |
| --- | --- | --- | --- | --- |
| AppsFlyer | Enterprise apps, large UA budgets | $0.07/conversion (Growth plan) | Full funnel; creative analytics in ROI360 tier | Annual or PAYG |
| Singular | Unified spend and attribution data | Custom quote | Full funnel; fraud prevention included | Annual |
| Branch | Deep linking, web-to-app flows | Custom quote | Attribution available; linking is primary strength | Annual |
| Kochava | Raw data access, custom analytics | Free up to 10K conversions; from $500/month (Foundation tier) | Full funnel; strong data export | Monthly or annual |
| Airbridge | Subscription apps scaling paid UA | 15K free installs, then $0.05/install | Subscription events pre-configured | Pay-as-you-go |
## In-Depth Review: The 5 Best Adjust Alternatives for 2026
### 1\. AppsFlyer: Best for enterprise apps with large UA budgets
AppsFlyer holds the largest market share among mobile measurement platforms globally, used by more than **15,000 brands** across industries ([AppsFlyer, 2026](https://www.appsflyer.com/solutions/partners/)). Its attribution infrastructure covers every major ad channel and more than **10,000** partner integrations. Fraud detection via Protect360 is among the most thorough available.
The Growth plan charges **$0.07 per attributed conversion** after an initial **12,000**\-conversion Welcome Package. Enterprise plans negotiate lower rates on annual contracts. Creative-level subscription analytics require the ROI360 add-on tier, meaning full subscription funnel visibility comes at a cost premium above the base plan.
**Best for:** Apps with **$50K+/month** in UA spend that need broad ad network coverage, enterprise fraud prevention, and dedicated account support.
**Where it falls short vs. Adjust:** The add-on tier model means teams frequently pay for features they do not use in order to access the features they need. No fully self-serve onboarding path makes it a poor fit for early-stage teams.
### 2\. Singular: Best for unified marketing data
Singular combines attribution with cross-channel cost aggregation in a single dashboard. If your growth team runs campaigns across Meta, Google, TikTok, programmatic networks, and influencer spend, Singular pulls ad spend and attribution data together without manual report reconciliation. Fraud prevention covers more detection methods than most alternatives in this list, and the reporting API is included at no additional charge.
Pricing is not publicly listed. Singular bills by attributed events or ad spend volume, with annual contracts as standard. There is no published free tier.
**Best for:** Growth teams running **10-plus** channel campaigns who need unified spend-to-outcome reporting in one place.
**Where it falls short vs. Adjust:** If subscription funnel depth is your primary need rather than cross-channel spend aggregation, Singular's overhead may not be justified. Getting a price number requires a sales conversation, just like Adjust.
### 3\. Branch: Best for deep linking and web-to-app flows
Branch specializes in deep linking and cross-channel user journeys. If a significant share of new users arrives through email campaigns, social referral links, or web-to-app flows, Branch routes those users into the correct in-app context with attribution intact. Many teams deploy Branch as a specialist linking layer alongside a separate primary measurement tool.
Branch offers attribution for subscription events, though its architecture is built for cross-channel journey continuity rather than subscription billing funnel analysis. Pricing is custom and requires a sales contact. Annual contracts are standard.
**Best for:** Apps where broken web-to-app attribution or deep link failures are the primary pain point, and where a specialist linking layer adds measurable conversion value on top of an existing attribution setup.
### 4\. Kochava: Best for raw data access
Kochava's core differentiator is data control. Row-level attribution data exports directly to BigQuery, Snowflake, and Redshift, giving data engineering teams the flexibility to build custom attribution models alongside Kochava's native reporting. Attribution window configuration, multi-touch models, and fraud rules are more granular here than in most alternatives.
Kochava offers a **free tier up to 10,000 attributed conversions per month**, with paid plans starting at **$500/month** (Foundation tier; Enterprise at **$2,000/month**). Enterprise pricing scales by MTU or conversion volume. The platform carries a steeper learning curve than most alternatives, and onboarding is not self-guided.
**Best for:** Data-science-heavy teams who need granular export capabilities, configurable attribution windows, and direct data warehouse access without a mandatory annual contract.
### 5\. Airbridge: Best for subscription apps scaling paid UA
Airbridge Core Plan is built around the subscription funnel. The **25 standard events** include Start Trial, Subscribe, and Unsubscribe pre-configured. Renewals are tracked via semantic attributes on the Subscribe event, so renewal data flows without additional schema work. Attribution covers the four major [Self-Attributing Networks](https://www.airbridge.io/en/glossary/self-attributing-network): Meta, Google, Apple Search Ads, and TikTok.
Pricing is transparent and public: **15,000 free attributed installs, then $0.05 per install**. No annual contract. No minimum monthly spend. [Lifetime value](https://www.airbridge.io/en/glossary/lifetime-value-ltv) reporting by channel and creative is available in the base plan, as are native integrations with RevenueCat and Adapty, included at no add-on tier. Among the adjust alternatives evaluated here, this is the only plan with fully public per-install pricing.
In a 2026 survey of **21 subscription app founders**, contract flexibility was the #1 reason cited for evaluating adjust alternatives to their existing MMP. Airbridge Core Plan is the only platform in this comparison with no minimum commitment and no annual lock-in.
**Honest limitations:** Core Plan does not support custom events (Growth Plan required). Maximum two third-party integrations. Coverage is GMAT channels only: Meta, Google, Apple Search Ads, and TikTok. No raw data export. No agency or partner portal access. If your stack includes non-standard ad networks, requires custom event definitions, or your team needs row-level data exports, these are real constraints to evaluate before committing.
**Best for:** Subscription apps running paid UA on Meta, Google, ASA, and TikTok, spending **$1K to $30K/month**, where transparent pay-as-you-go pricing and subscription-first measurement matter more than maximum network coverage.
For a deeper head-to-head including Adjust itself, see [AppsFlyer vs. Airbridge vs. Adjust vs. Branch: Best MMP for Subscription Apps in 2026](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026).
## The Right Alternative Starts with the Right Question
The best Adjust alternative is not the one with the longest feature list. It is the one whose pricing model, subscription funnel depth, and channel coverage match where your app is right now and where you need it to be in 12 months.
For subscription apps running paid UA on Meta, Google, Apple Search Ads, and TikTok that need to see which channels are actually driving paying subscribers, not just installs, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) is built for that specific measurement problem. Start free with **15K attributed installs** and see trial-to-subscription data by channel before committing to a contract.
---
## How to Migrate to Airbridge: Channel Integration Checklist
URL: https://www.airbridge.io/en/blog/how-to-migrate-to-airbridge-channel-integration
Language: English
Published: 2026-06-01
Author: Luke
You've made the call to migrate to Airbridge. The contract with your current mobile measurement partner is expiring, the pricing crossed a line, or the subscription analytics gap finally got too wide. The decision is done.
The SDK swap isn't where migrations break. Channel integrations are. Leave one SAN setting pointing to your old MMP, skip converting your tracking links, or miss TikTok's mandatory SAN migration step, and you'll spend weeks diagnosing data gaps that could have been prevented in an afternoon.
This checklist covers the exact failure points, the right order to address them, and what to verify before you restore full campaign spend.
**Key Takeaways**
- **SAN integrations don't auto-switch.** Meta, Google, Apple Search Ads, and TikTok each require manual disconnection from your old MMP and reconnection to Airbridge. Skipping this step causes double-attribution or creates silent data gaps.
- **Dual SDK overlap is intentional, but time-bounded.** Running both SDKs in parallel for at least 14 days protects data for users who haven't updated to the new app version. Running them indefinitely causes event duplication and double charges from **Meta** and **TikTok**.
- **TikTok requires a full SAN migration, not just a link swap.** Apps on legacy integration cannot create new campaigns or share in-app events ([TikTok For Business](https://ads.tiktok.com/help/article/transition-to-san-for-existing-apps)).
- **SKAN postbacks route to one MMP only.** The SKAN data gap starts on your app release date and cannot be recovered retroactively. Plan your release date with this in mind.
- **Start channel cutover from your lowest-spend network.** This limits exposure if something breaks during the link transition window.
## Why Channel Integrations Break When You Switch MMPs
The real failure patterns appear at the channel level, after the new SDK is integrated and tested.
[Screenshot 2026-06-01 at 20.00.03.png]
### Your Old SAN Settings Stay Live After You Switch MMPs
When you first set up Meta, Google Ads, Apple Search Ads, or TikTok inside AppsFlyer, Adjust, or Branch, you configured those channels to route postback events to your old MMP. That setting doesn't expire when your contract does.
If you release the new app with the Airbridge SDK installed but leave the old MMP's SAN configuration active, three things happen at once. Postback events fire to both tools. Ad platform optimization keeps reading signals from the old MMP, not Airbridge. The new MMP shows gaps while the old MMP dashboard looks completely normal.
Your team assumes the migration is working because the old tool looks fine. Attribution is incomplete for weeks.
### Running Both MMP SDKs Without a Cutoff Date Causes Double Attribution
The failure mode is an open-ended overlap. When both SDKs are active and both MMP postback rules are live, the same install event gets attributed by two systems. Ad platforms receive duplicate optimization signals. CPA calculations become unreliable because both cost and conversion data are inflated.
Define the dual-SDK window before release. **Fourteen days** covers the majority of users who update to a new app version. After that window, remove the old SDK.
### Old Tracking Links Keep Sending Data to Your Previous MMP After Launch
For ad networks outside the SAN ecosystem, and any campaigns using custom tracking links generated by your old MMP, those links continue routing attribution data to your old tool after migration. You're paying Airbridge while feeding your old MMP.
This issue is easy to miss. Campaigns keep running. The links don't return errors. Old MMP dashboards continue showing active traffic. The data loss is invisible until you compare attribution totals.
## Pre-Migration Setup: 3 Steps to Complete Before Going Live With Airbridge
Complete these three steps before your developer releases the app update. Steps 1 and 2 can run in parallel. Step 3 (identifier upload) runs immediately before release.
### Step 1: Register Your App and Set Up Your Airbridge Event Taxonomy
Register your app in the Airbridge dashboard. Under Settings > Tokens, you'll find the SDK credentials and API keys. Share these with your developer along with the SDK Quickstart.
Before any SDK code is written, finalize your event taxonomy. This is the event naming structure Airbridge uses to collect behavior. Airbridge's [**29** standard events](https://help.airbridge.io/en/developers/standard-events-and-semantic-attributes) cover the core funnel, including subscription events like Start Trial, Subscribe, and Unsubscribe. They eliminate the custom schema design work that delays migrations from general-purpose setups.
If you're migrating to Core Plan, it covers Google, Meta, Apple Search Ads, and TikTok integrations plus up to **two** additional third-party integrations ([Core Plan](https://www.airbridge.io/en/plans/core)). Confirm your current integration stack fits that limit before setup.
Review the Airbridge attribution model at this stage. Attribution windows and click-priority rules differ across MMPs. Configure your attribution rules before live data flows in. Changes after campaigns are running are harder to interpret cleanly.
### Step 2: Remove Old MMP Postbacks and SAN Settings Before App Release
| Action | Where |
| --- | --- |
| Remove old MMP as attribution partner | Inside each SAN channel dashboard (Meta, Google, ASA, TikTok) |
| Delete postback rules pointing to old MMP | Old MMP's channel integration settings |
| Pause campaigns using old tracking links | Ad platform campaign manager |
If postback settings and SAN integrations aren't disabled before the new app releases, events get counted in both tools. Ad platforms receive conflicting optimization signals during the overlap window.
### Step 3: Upload Device IDs to Prevent False First-Install Spikes in Airbridge
When the new MMP first onboards your app, it has no record of which device IDs or user IDs have already been attributed. This creates a temporary spike in "first events": it counts returning users as new installs, with no record of their previous attribution.
Upload device IDs and user IDs from your old MMP before releasing the new app. These identifiers let it recognize returning users and exclude them from new-install counts, producing a cleaner CPA baseline during the transition window.
This step is strongly recommended when active campaigns are running, and effectively mandatory if significant spend is live.
## Channel Reconnection Checklist: TikTok, Apple, Meta, and Google
Work through channels from highest-risk to lowest. TikTok carries the most migration complexity and should not be last.
| Priority | Channel | Key Risk |
| --- | --- | --- |
| 1 | TikTok for Business | SAN migration mandatory; legacy integration defunct since March 31, 2025 |
| 2 | Apple Search Ads | SKAN postbacks route to one MMP; gap is not retroactively recoverable |
| 3 | Meta Ads | Dual postback creates conflicting optimization signals |
| 4 | Google Ads | Attribution window mismatch skews first-cycle ROAS comparisons |
### Channel 1: TikTok for Business (Mandatory SAN Migration)
TikTok carries the most migration complexity. Handle it first.
TikTok's legacy tracking-link-based MMP integration was formally discontinued on **March 31, 2025**. If your app was on legacy integration with AppsFlyer, Adjust, or Branch, the SAN migration is not optional. Remaining on legacy integration blocks new campaign creation and disables in-app event sharing.
**Steps:**
1. In Airbridge dashboard: Integration > Ad Channel Integration > TikTok For Business
2. Enable the channel integration toggle and register your TikTok App ID
3. Enable In-app Event Postback and configure event mapping (Subscribe, Start Trial, Order Complete)
4. In TikTok Ads Manager: navigate to Tools > Events, find your app, and select "Migrate to SAN"
5. Remove the old MMP as the attribution partner in TikTok Ads Manager
**What can break:** If you plan to use Value-Based Optimization (VBO), check TikTok's current [VBO eligibility criteria](https://ads.tiktok.com/help/article/specifications-for-value-based-optimization-for-app) before launching. VBO requires a sustained volume of purchase-with-value postbacks before it unlocks. App Event Optimization (AEO) activates after a single qualifying event. Budget for a ramp-up window before either feature stabilizes.
[Airbridge dashboard showing TikTok For Business channel integration with In-app Event Postback enabled and event mapping configured]
_Airbridge dashboard: TikTok For Business channel integration with In-app Event Postback and event mapping_
### Channel 2: Apple Search Ads (SKAN Data Gap Risk)
**Steps:**
1. Confirm your Airbridge iOS SDK is version [**1.15.3** or later](https://help.airbridge.io/en/guides/apple-search-ads-channel-integration), required for the AdServices framework
2. Enable the AdServices framework in your iOS app build
3. Connect Apple Search Ads in Airbridge: Integrations > Ad Channel Integration > Apple Search Ads
4. Configure SKAdNetwork (SKAN) authentication and map your conversion value schema to Airbridge's subscription events
**What can break:** SKAN postbacks route to exactly one MMP. Until you release the app version containing the Airbridge SDK, your old MMP receives all SKAN conversion data. That data cannot be retroactively migrated to Airbridge. The gap in SKAN data starts exactly on your app release date and cannot be recovered. Plan release timing with this in mind.
[Airbridge dashboard showing Apple Search Ads channel integration setup]
_Airbridge dashboard: Apple Search Ads channel integration setup_
### Channel 3: Meta Ads (Dual Postback Conflict)
**Steps:**
1. In Airbridge dashboard: Integrations > Ad Channel Integration > Meta Ads
2. Enable the channel integration toggle in the Channel Integration tab
3. Enable In-app Event Postback
4. Verify event mapping: confirm Airbridge's Subscribe and Start Trial events are mapped to the Meta events your campaigns use for optimization
**What can break:** Meta's ad algorithm continues reading signals from your old MMP until you complete the Airbridge SAN setup and remove the old integration. If both are active simultaneously, Meta receives duplicate postback signals, which distorts conversion counts and campaign optimization.
[Airbridge dashboard showing Meta Ads channel integration with Channel Integration toggle enabled and In-app Event Postback active]
_Airbridge dashboard: Meta Ads channel integration with In-app Event Postback enabled_
### Channel 4: Google Ads (Attribution Window Mismatch)
**Steps:**
1. Connect Google Ads in Airbridge under SAN integrations and authorize the account link
2. Map in-app conversion events to Google Ads conversion goals
3. In Google Ads: verify conversion goals reflect Airbridge's event names
**What can break:** Google Ads' default click-through conversion window is [**30 days**](https://support.google.com/google-ads/answer/3123169). A tighter Airbridge window versus Google's 30-day default typically understates first-cycle ROAS. Align these settings before your first reporting cycle closes.
[Airbridge Attribution Rule settings showing Google Ads click lookback window set to 30 days]
_Airbridge Attribution Rule: Google Ads click lookback window default (30 days) versus global setting_
## Post-Migration Verification: How to Confirm Airbridge Is Receiving Clean Data
After releasing the updated app, run these checks before restoring campaign budgets.
### Confirm All Four Channels Are Sending Live Postbacks
Typically within **24-48 hours** of the new app version going live, event data should appear under each channel integration:
- Meta: installs and subscription events appear with correct channel attribution
- Google: conversion data imports typically reflect in Google Ads within **24 hours**
- Apple Search Ads: installs from ASA campaigns appear with correct source attribution
- TikTok: Order Complete and event postbacks appear under TikTok channel reporting
If any channel shows zero attributed events after **48 hours** of active traffic, the SAN setup is likely incomplete or the old MMP is still active as the attribution partner.
### Why Your Airbridge and Ad Platform Numbers Won't Match Exactly
Discrepancies between Airbridge and ad platform dashboards are common and expected ([TikTok For Business](https://ads.tiktok.com/help/article/about-reporting-discrepancies-between-tiktok-ads-manager-and-mmps)). Self-attributing networks apply their own attribution logic. Each platform uses different lookback windows by default. The numbers will not match exactly.
A discrepancy that's unusually large signals a specific problem: a broken postback, an unconverted tracking link, or an old MMP integration that's still active.
### Remove the Old MMP SDK After 14 Days
After **14 days** post-release (rationale in the dual-SDK section above), remove the old SDK, revoke remaining API connections in your old MMP dashboard, confirm no live campaigns are still running on old tracking links, and export any historical data before closing the account.
## Migrate Without the Data Gaps
Channel integrations are where most MMP migrations lose data, not the SDK. The [Airbridge migration guide](https://help.airbridge.io/en/guides/migration-guide) covers the full 7-step process from app registration through performance reporting. The checklist above maps the integration order that minimizes attribution loss: event taxonomy and old MMP cleanup first, identifier upload before release, then SAN channels in order of migration risk.
---
## Should You Offer a Lifetime Subscription? When It Makes Sense
URL: https://www.airbridge.io/en/blog/should-you-offer-a-lifetime-subscription
Language: English
Published: 2026-06-01
Author: Harper (Trang Nguyen)
You spent months building your app. Subscribers are coming in. Some are churning after month two or three. Then a user posts in your community: "Do you offer a lifetime option? I'd pay for that."
The question feels like an opportunity. The math, however, is not simple. **Offer a lifetime subscription app option and you might lock in a year of revenue upfront. But if that user would have renewed for five more years, you have just sold yourself short.**
This is the central tension of lifetime subscriptions: they almost always favor one side. Usually, it is the customer who wins.
---
**Key Takeaways**
- **Lifetime subscriptions are growing but still a minority model.** One-time and lifetime purchases grew from 6.4% to 10.3% of app monetization models between 2023 and 2025 (Adapty 2026), but recurring subscriptions remain dominant across most categories.
- **The biggest risk is cannibalizing your own revenue.** Your most loyal users, the ones most likely to buy a lifetime deal, are also the ones most likely to keep renewing without one.
- **Pricing math is not optional.** Lifetime subscription app pricing ranges from **2x to 12x** the annual price, with most successful premium-tier offers landing at 5x or higher. Price below your cohort LTV and you compress MRR; price too high and nobody converts.
- **Test via owned channels first.** Email and push notifications before the paywall: this keeps targeting precise and avoids disrupting your main conversion flow.
- **Your LTV by cohort, not your average LTV, is the number that matters** when deciding whether a lifetime offer makes financial sense for your app.
## What Is a Lifetime Subscription for Apps (and How Common Is It)?
### 1\. How lifetime subscriptions work technically
A lifetime subscription is not a subscription under App Store rules. It is classified as a **non-consumable in-app purchase**: a one-time payment that grants ongoing access to an app's features.
This distinction has real consequences. Standard auto-renewable free trials are not available for lifetime subscription app purchases. Users typically must pay upfront without a trial period, though some developers have implemented workaround patterns by pairing a free non-consumable purchase with a paid one. And switching from a monthly or annual subscription to a lifetime plan requires the customer to manually cancel their existing subscription before repurchasing. That two-step process regularly produces double charges, refund requests, and support load that you will carry for years.
### 2\. Which app categories use lifetime subscriptions most
For most app categories, a combination of subscription and lifetime access is the second most popular form of monetization. According to RevenueCat's blog analysis (November 2025), **Photo and Video apps lead the way**, often framing the lifetime purchase as a one-off feature unlock rather than "forever access." Education and Travel follow closely: both categories where users see a clear long-term value proposition.
Business apps are the exception. Per-user infrastructure costs are high, products change rapidly, and enterprise buyers expect recurring billing. Lifetime subscriptions rarely fit the B2B model. Across other categories, adoption rates vary meaningfully by niche and audience composition.
### 3\. The hybrid monetization trend
The broader shift is toward mixing revenue models. Analysis of 75,000+ subscription apps shows that **35% of apps now combine subscriptions with consumables, one-time purchases, or lifetime access** (RevenueCat State of Subscription Apps 2025). Gaming leads this adoption at 61.7%, followed by Social and Lifestyle at 39.4% (RevenueCat SOSA 2025).
One-time and lifetime purchase share grew from **6.4% to 10.3% of app monetization models between 2023 and 2025**, based on data from 16,000+ apps representing $3B in subscription revenue (Adapty State of In-App Subscriptions 2026). The trend is real but measured: it reflects niche use cases and seasonal promotions, not a wholesale move away from recurring revenue.
## The Real Benefits of Offering a Lifetime Plan
### 1\. Immediate cashflow and early runway
Lifetime subscriptions front-load revenue. For bootstrapped teams or early-stage apps, a targeted lifetime campaign can fund six months of content production, feature development, or marketing spend.
This works especially well for content-heavy products. If you are investing in new features, partnerships, or media production, a burst of upfront revenue can fund what monthly MRR would take quarters to accumulate.
Seasonal timing amplifies this: **Black Friday, New Year's Day, and product milestone campaigns** are natural windows for lifetime offers. Discount-driven users tend to engage less and churn faster, so a lifetime deal filters for committed buyers while delivering a cash infusion.
### 2\. Locking in users who would otherwise churn
If a subscriber is at risk of canceling, a lifetime offer converts a future loss into a locked-in payment. Instead of losing a **$79/year** subscriber who would have churned in month three, you collect a one-time **$399** and move on.
This logic is strongest when your **early-period retention is weak**. Among subscription apps, the first renewal is the highest-risk moment. Yearly plan first-renewal rate sits at a **median of 66.3%**, meaning roughly one in three annual subscribers cancels before their second year begins. Monthly and weekly first-renewal rates are lower still (55.5% and 51.8% respectively) (RevenueCat State of Subscription Apps 2025). Lifetime deals sidestep this risk entirely for the users who take them. If improving retention is a priority before testing lifetime offers, [App Retention Rates (2026)](https://www.businessofapps.com/data/app-retention-rates/) gives concrete benchmarks to work from.
### 3\. Competitive differentiation and referral leverage
In crowded categories, a lifetime subscription can credibly separate a smaller app from larger players. Most established apps (streaming services, major fitness platforms) do not offer lifetime access. That gap is an opening for apps willing to use it strategically.
Some teams have turned lifetime access into a referral engine. The finance app Origin, for example, grants lifetime membership plus a $100 reward to users who refer three friends. Rather than competing on price alone, the offer becomes a growth mechanic that rewards advocacy.
## The Hidden Risks That Can Hurt Your Revenue
### 1\. Cannibalizing your highest-LTV subscribers
Here is the uncomfortable math. Your most loyal users are the ones most likely to buy a lifetime plan. They are also the ones most likely to renew annually for years without any incentive.
If your high-retention annual subscriber stays for 4 years at $79/year, their realized value to you is around **$316**. Selling them a lifetime deal at **$299** looked like a gain. It was not.
This is why pricing off average [LTV (Lifetime Value)](https://www.airbridge.io/en/glossary/lifetime-value-ltv) is dangerous. **You need LTV by cohort**: specifically, what does your highest-retention, highest-engagement segment actually generate over three, five, or seven years? That number is your floor for lifetime pricing, not your starting point.
### 2\. Upsell limitations and pricing ceilings
Once a user is on a lifetime plan, the ceiling is fixed. You can introduce new premium tiers that are not included, but converting a lifetime subscriber to an upgraded tier is structurally harder than moving a monthly subscriber.
If **higher-tier upsells are a meaningful part of your monetization roadmap**, lifetime subscriptions create a conflict. The user has already bought "everything" from their perspective. Future pricing changes, new feature tiers, or premium content packages become much harder to sell to this cohort.
### 3\. Operational friction and investor perception
Lifetime mechanics generate ongoing support load: double charges from failed subscription cancellations, refund disputes, feature access questions that surface years after purchase. This is manageable, but account for it in support capacity before launching.
From an investor's perspective, lifetime revenue is a one-time event. Recurring subscription revenue compounds, drives retention metrics, and increases valuation multiples. A high share of lifetime revenue in your mix signals that MRR growth is being front-loaded, which can suppress valuation. This is why lifetime subscriptions are rare in B2B SaaS and growth-stage apps approaching institutional funding rounds. For context on how investors read subscription revenue metrics, see [State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps/).
## Lifetime vs. Annual Subscription: Which Wins for Long-Term Profits?
### 1\. Understanding LTV by plan type
Annual subscribers generate the strongest compounding value in most categories. Among 75,000+ subscription apps, **yearly plan 12-month retention sits at a median of 44.1%** versus 17.5% for monthly plans and just 3.4% for weekly plans (RevenueCat State of Subscription Apps 2025). Annual subscribers also stabilize fastest: by the third renewal, renewal rates across plan types approach 80%+ (RevenueCat SOSA 2025). For a deeper look at how subscription plan choice affects revenue, see [Subscription App Pricing by Category: 2026 Benchmarks](https://www.airbridge.io/en/blog/subscription-app-pricing-by-category-2026-benchmark).
| Subscription Plan | 12-Month Retention (Median) | 1st Renewal Rate |
| --- | --- | --- |
| Weekly | 3.4% | 51.8% |
| Monthly | 17.5% | 55.5% |
| Yearly | 44.1% | 66.3% |
Lifetime pricing must account for the revenue stream you are replacing, not just year one.
### 2\. The pricing math: setting your lifetime price
Lifetime subscription app pricing varies widely, from 2x to nearly 12x the annual price. Here is how well-known apps currently structure their offers (RevenueCat blog, November 2025):
| App | Annual Price | Lifetime Price | Multiplier |
| --- | --- | --- | --- |
| Calm | $79 | $399 | 5.0x |
| Jumpspeak | ~$99 | $249 | ~2.5x |
| Moonly | $28.98 | $59.98 | 2.1x |
| Fiit | £119.99 | £375 | 3.1x |
| Waking Up | $129.99 | $1,500 | 11.5x |
To find your range, work through four inputs:
- **Annual price:** Your recurring revenue baseline
- **Annual churn rate:** Higher churn means lower implied LTV, which lowers the price you can justify
- **Gross margin per user:** Low-margin apps (high AI compute or server costs) cannot absorb low lifetime prices
- **Category retention benchmarks:** Use your category's median annual retention as your floor for multi-year LTV modeling
**When uncertain, price higher rather than lower.** Waking Up's $1,500 lifetime offer is strategic aggression: it converts only the most committed buyers and extracts maximum value from a small high-retention segment without cannibalizing recurring revenue from everyone else.
## Should You Offer a Lifetime Subscription App Option? A Decision Framework
Use this checklist before committing. More "yes" answers mean a stronger case for testing a lifetime offer.
1. **Are your ongoing per-user costs low?** High compute, AI inference, or content licensing costs can make each lifetime subscriber a multi-year liability rather than an asset.
2. **Is valuation not your near-term priority?** If you are not approaching a funding round in the next 12 to 18 months, the MRR impact matters less.
3. **Do you need upfront cash to fund development or extend runway?** Lifetime offers function as a funding mechanism for teams that prefer not to raise capital.
4. **Is your early-period retention weak?** If users churn in months one through three, lifetime deals lock in revenue before it walks out the door.
5. **Are users already asking for a lifetime option?** Organic demand is a signal that a committed segment exists and is ready to convert.
6. **Are lifetime subscriptions common in your category?** Photo and Video, Education, and Travel have established lifetime offer norms. Business apps largely do not.
7. **Are higher-tier upsells not central to your model?** If future premium tiers are core to your revenue roadmap, lifetime access creates a ceiling that blocks future monetization.
**Red lights: pause before launching lifetime subscriptions if:**
- Your per-user AI, compute, or content costs are high and likely to increase
- You plan to raise institutional funding in the next 12 months
- Multi-tier upsells are a primary driver of revenue growth
- You do not yet have enough retention data to model multi-year LTV confidently
## How to Launch and Price Your Lifetime Offer
### 1\. Start with a targeted segment, not your entire user base
The worst lifetime subscription rollout is placing the offer on your main paywall for all users. This directly cannibalizes conversions from subscribers who would have paid monthly or annual rates without any incentive.
Start with two segments:
- **Lapsed users:** Win-back campaigns via email to recently churned subscribers. Lifetime access is compelling for someone who already knows the product but did not see enough value in recurring billing.
- **Long-tenure subscribers:** Users who have renewed two or three times already. Some will convert; most will not. But the ones who do tend to have the strongest engagement and referral potential.
### 2\. A/B test the multiplier before setting a permanent price
Run A/B tests across multipliers (3.5x vs 5x vs 7x your annual price) against equivalent segments before committing. Track these metrics:
- **Conversion rate by segment:** Who is actually converting, and what do they have in common?
- **ARPU vs comparable non-lifetime cohorts:** Are lifetime buyers outperforming or underperforming what they would have generated as annual subscribers?
- **Engagement rates post-purchase:** Lifetime subscribers who do not engage are unlikely to refer or advocate for your app.
- **Refund and support volume:** Watch for double-charge complaints and refund spikes in the 14 days following purchase.
- **Revenue mix shift:** If lifetime revenue grows above 20 to 30% of total revenue, your MRR trajectory is likely being compressed by the offer.
### 3\. Use owned channels before your main paywall
Most successful lifetime offers reach users through email, push notifications, or in-app messaging rather than the main subscription paywall. Fiit delivered lifetime offers via email to recently canceled subscribers. Origin built it into a referral loop.
Owned channels let you control who sees the offer, at what price point, and when. You can run clean tests, collect actionable data, and optimize before any broader rollout. If the numbers look strong after two to three months of owned-channel testing, consider a limited paywall placement for a specific user segment only.
## Run the Numbers Before You Commit
Lifetime subscriptions are a specific tool: useful for cashflow injection, churn mitigation in targeted segments, and competitive differentiation where the category supports it. They are not a substitute for strong retention, and they will cost you revenue if offered to the wrong users at the wrong price.
The apps that execute this well treat lifetime access as a premium option for a small, committed segment priced deliberately above expected multi-year LTV. The ones that get hurt sell lifetime access at 2 to 3x annual to users who would have renewed quietly for a decade.
Know your cohort LTV. Model your per-user costs. Test on lapsed users first.
Running that cohort analysis is harder when your ad spend data, subscription revenue events, and user behavior all live in separate tools. [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your paid campaigns across Meta, Google, Apple Search Ads, and TikTok to subscription revenue events (Start Trial, Subscribe, Unsubscribe), so you can see LTV by channel and by creative before making any major pricing decision. If you are evaluating whether a lifetime subscription app option makes sense for a specific user segment, knowing which acquisition channels produced that segment in the first place changes the answer. [**Start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core)**: 15K attributed installs included, no annual contract.**
---
## Hard Paywall vs Freemium in 2026: What 75K Apps Reveal
URL: https://www.airbridge.io/en/blog/hard-paywall-vs-freemium-2026
Language: English
Published: 2026-05-27
Author: Luke
The average subscription app converts **1.9%** of downloads into paying subscribers within the first 35 days. Apps using a hard paywall convert at **12.1%**, which is **5.5 times higher**, per the [RevenueCat State of Subscription Apps 2025 analysis of 75,000+ apps](https://www.revenuecat.com/state-of-subscription-apps-2025/). But hard-gated apps also see refund rates **70%** higher than freemium, and many of those installs never come back.
A hard paywall blocks the core app experience behind a subscription. A freemium model offers ongoing free access with paid upgrades later. Trial-inclusive designs sit between them, gating the product after a short evaluation window.
The hard paywall vs freemium 2026 question is not abstract. It is a category-fit decision that shows up in your D35 conversion, your Y1 LTV, and your refund rate. Hard gates win on per-user economics. Freemium wins on volume and category dynamics.
**Picking the wrong model can compress your conversion rate by 5x or burn your top-of-funnel.** The question is not which is "better." It is which is better for the app you are building.
**Key Takeaways**
- **Hard paywalls convert 5.5x higher than freemium on D35 download-to-paid.** Median rates are **12.1%** vs **2.2%** (RevenueCat SOSA 2025). This is a survivorship effect: most users who do not pay never return.
- **Hard-gated apps generate 2x the Y1 lifetime value (LTV) per payer.** Median **$49.30** vs **$24.24** (RevenueCat SOSA 2025). Higher prices and self-selecting users compound into stronger lifetime value.
- **Stacked together, hard models deliver ~11x revenue per install.** $5.96 vs $0.53 once you multiply conversion by LTV. The economic edge is much larger than the headline 2x suggests.
- **Trial-inclusive paywalls win the most A/B tests.** Trial-format screens beat visual-only layouts in **64.5%** of head-to-head experiments, per the [Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/) report on 16,000+ apps and $3B in revenue.
- **Category dictates the right model.** **86%** of AI apps skip trials (RevenueCat SOSA 2025), Health & Fitness wins with trial-inclusive (RevenueCat), and Productivity often loses LTV when trials are added (Adapty 2026).
- **Hard model refund rates are 70% higher.** **5.8%** vs **3.4%** for freemium (RevenueCat SOSA 2025). The conversion lift is real but partly clawed back.
## 5.5x Conversion Gap: The Hard Paywall Number That Hides a Trap
Start with the headline number. Across 75,000+ subscription apps analyzed by RevenueCat in 2025, median D35 download-to-paid conversion rates split sharply by paywall strategy.
| Strategy | Median D35 Conversion | P90 (Top 10%) |
| --- | --- | --- |
| Hard Paywall | 12.1% | 39.0% |
| Freemium | 2.2% | 7.9% |
| All Categories | 1.9% | 8.5% |
_Source: RevenueCat State of Subscription Apps 2025 (75K+ apps)._
A hard gate pushes more users to a yes/no decision faster. Freemium offers ongoing free access and tries to convert users later. The conversion gap is enormous: **5.5x** at the median, and the top 10% of hard-gated apps reach **39%**.
But that number hides what is happening underneath. Hard gates do not manufacture purchase intent. They surface it earlier and discard everyone else. Users who would not have converted under freemium also do not convert under a hard model. They churn at the door instead of months later.
This only matters if your acquisition can scale with your model. Most apps cannot 10x install volume by switching to freemium. As the category-fit section later shows, the model is partly determined by what your app does, not by what your spreadsheet wants.
## 2x LTV, And Why the Gap Compounds With Conversion
Conversion is the first half of the story. Lifetime value is the second.
| Strategy | Y1 LTV per Payer (Median) |
| --- | --- |
| Hard Paywall | $49.30 |
| Freemium | $24.24 |
_Source: RevenueCat State of Subscription Apps 2025._
Hard-gated payers are worth roughly **2x more over their first year**. Two mechanisms drive the gap.
**1\. Price elasticity favors hard gates.** Apps that force a paid decision upfront can sustain higher pricing because they are filtering for users with explicit purchase intent. Adapty's 2026 analysis of 16,000+ apps shows high-priced apps generate **3x** the LTV of low-priced apps. Hard-gated apps cluster toward the higher-price end of that distribution.
**2\. Self-selection improves retention.** A user who pays before experiencing the product is signaling stronger conviction than a user converted from freemium via aggressive upsells. The same Adapty dataset shows top hard-gated users spend **20 to 33%** more than their freemium counterparts on equivalent products.
The compounding effect is the real story. Stacked on the conversion gap, hard models generate roughly **11x revenue per install**:
- **Hard:** 12.1% × $49.30 = **$5.96 per install**
- **Freemium:** 2.2% × $24.24 = **$0.53 per install**
The LTV gap multiplies the conversion gap. That is the economic argument. The next section is the case against making the decision on revenue per install alone.
## The Hidden Costs of Hard Paywalls: Refunds, Churn, and Top-of-Funnel Loss
Conversion and LTV are the case for hard models. The case against them is quieter but real.
### 1\. Hard Paywall Refund Rates Run 70% Higher
Hard-gated apps see median refund rates of **5.8%** versus **3.4%** for freemium (RevenueCat SOSA 2025). That is a **70% higher refund rate**.
Users who paid before trying are more likely to feel surprise or regret when they realize what they bought. The same dataset shows Education apps already lead refunds at **4.86%** and Health & Fitness at **4.71%**. Layering a hard gate on top pushes those numbers higher.
A 70% lift in refunds does not erase the LTV advantage. It does mean your reported revenue is more volatile, and your customer support load is heavier.
### 2\. Hard Paywalls Collapse the Top-of-Funnel
The 5.5x conversion lift comes with a corresponding install-to-engagement collapse. Hard-gated apps convert more of who they keep. They keep far fewer. For apps that rely on organic growth, network effects, or referral loops, this is a structural problem.
### 3\. Hard Paywalls Limit Onboarding Optimization Gains
The biggest gains in onboarding optimization come from increasing initial engagement before the monetization gate appears. Hard gates limit how much engagement work you can do, which compresses the gains from the highest-leverage optimization area in subscription apps.
RevenueCat data shows **82%** of trials start on Day 0, so placement and first impression are decisive. Hard gates give you less surface area to refine. The [paywall flow setup before scaling](https://www.airbridge.io/en/blog/set-up-a-simple-paywall-flow) covers what to instrument before you experiment.
## When Freemium Wins and Why Trial-Inclusive Paywalls Beat Both
Freemium is not the loser in this matchup. It is the right model for a specific set of conditions.
### 1\. When Freemium Beats Hard Paywall
Freemium works when your app has network effects, when users need time to experience value, when you optimize for the long tail, or when you are already at scale.
- **Network-effect apps.** Communication, marketplace, and social apps need scale to deliver value. Hard gates kneecap that.
- **Value-over-time apps.** Productivity tools that take weeks to integrate into a workflow cannot validate willingness to pay on Day 0.
- **Long-tail monetization.** If 1% of your users pay $100/year and the other 99% generate ad revenue, referrals, or content, freemium might be right.
- **At-scale economics.** Freemium math works at 1M+ monthly actives. It rarely works at 10K.
The 2026 data shows freemium is not going away. One report, based on 2,900+ subscription apps and 1.7B paid installs, shows hybrid models (subscription + ads, subscription + IAP) growing across nearly every category.
- **Gaming.** Subscription + IAP combo moved from **29.5% to 37%**, up **25%**.
- **Lifestyle.** Subscription + ad mix grew roughly **70%** in two years (**5.4% to 9.2%**).
- **Short Drama.** Ad monetization rose from near-zero to **7.4%**, a new freemium-leaning category.
The future is not "pick one." It is freemium as a base layer with subscription monetization stacked on top.
### 2\. Why Trial-Inclusive Paywalls Win Most A/B Tests
The hard-versus-soft framing often misses the option most successful apps actually use: **trial-inclusive paywalls**.
In head-to-head A/B tests, trial-format screens **win 64.5%** of experiments against visual-only layouts (Adapty 2026). Trial-inclusive designs gate the product after a short evaluation window, combining two strengths:
- **The conversion velocity of hard models.** Paid decision happens quickly.
- **The lower refund risk of freemium.** Users have tried before they buy.
For structured experimentation, the [paywall A/B testing setup guide](https://www.airbridge.io/en/blog/ab-test-paywall-setup-duration-results) breaks down sample size and duration considerations.
**Trial length matters more than design.** Apps with 17 to 32 day trials convert at a median **45.7%**, nearly double the **26.8%** rate of 3 to 7 day trials (RevenueCat SOSA 2025). Counter to most copywriting advice, longer trials do not reduce conversion. They let high-intent users self-confirm at a comfortable pace.
**Two categories where trials backfire.** Trial-inclusive is not the right call for every app type:
- **AI apps.** **86%** skip trials and monetize fine because their value is instant and consumable, not gradual (RevenueCat SOSA 2025).
- **Productivity.** Adapty's data shows trials can actually hurt LTV: direct purchasers convert at **$55.45** vs **$40.13** for trial users.
The trial model is the safest default, not the universal answer.
## Best Paywall Strategy by App Category in 2026
Strategy interacts with category dynamics. The four cases below cover the four distinct models most subscription apps end up running.
| Category | Best-Fit Model | Why |
| --- | --- | --- |
| Health & Fitness | Trial-Inclusive | 7.8% trial start rate (highest), Y1 LTV $27.21, trial boosts LTV (RevenueCat SOSA 2025) |
| AI Apps | Hard Paywall or No-Trial | 86% skip trials (RevenueCat SOSA 2025); $1.44 install LTV vs $0.84 average (Adapty 2026) |
| Productivity | Direct Purchase | Trial users convert at lower LTV: $40.13 vs $55.45 direct (Adapty 2026) |
| Gaming | Freemium + IAP Hybrid | Trial conversion 19% (lowest), IAP combo up 25% (Adapty 2026) |
The pattern is simple:
- **Hard paywall** fits when value is instant or commoditized.
- **Trial-inclusive** fits when value is gradual and personal.
- **Freemium** fits when scale and engagement compound.
A Health & Fitness app shipping a hard gate is leaving the **39.9%** trial-to-paid lift on the table. An AI app with a 30-day trial is giving away free compute to users who would have paid Day 0. The [subscription app pricing by category benchmarks](https://www.airbridge.io/en/blog/subscription-app-pricing-by-category-2026-benchmark) help anchor where your prices should land within each vertical.
## How to Choose Between Hard Paywall, Freemium, and Trial in 2026
Three questions narrow the field from two religious camps to two or three testable options. The decision tree below visualizes the flow.
[Screenshot 2026-05-27 at 17.35.54.png]
1. **Does your app deliver value in the first session?** Yes points to hard paywall or no-trial direct purchase. AI apps, instant-utility apps, OTT. No points to trial-inclusive or freemium. Fitness, education, productivity.
2. **Do you need scale to monetize through network effects, ads, or referral loops?** Yes points to freemium or hybrid. Social, dating (lower-end), gaming. No points to hard paywall or trial. Solo-use, premium-positioned apps.
3. **Is your category trial conversion rate above or below 30%?** Above (Education **42%**; Health & Fitness **39.9%**, Travel **48.7%** per RevenueCat SOSA 2025) points to trial paywall. Below (Gaming **19%**) points to skip trials and use hybrid monetization.
The framework will not decide for you. It will narrow your options to a testable set. From there, the right move is a structured paywall experiment.
The 2026 experimentation data is striking:
- **Experimenting teams generate up to 40x more revenue** than non-experimenters (Adapty 2026).
- **Price experiments improve LTV 46% of the time** even when conversion lifts only **28%** of the time (Adapty 2026).
Judge experiments by ARPU, not raw conversion.
## The 3-Step Sequence Most Founders Should Ship in 2026
There is no universal default for subscription app monetization in 2026. There is a category-fit decision, a value-delivery question, and a test plan.
The practical sequence for most founders shipping this year is straightforward:
1. **Default to a 17 to 32 day trial** unless your category data points elsewhere. The data shows this trial length converts at **45.7%** median, nearly double the rate of short trials.
2. **Validate against your category's benchmarks** for D35 conversion, Y1 LTV, and refund rate within 60 days of launch.
3. **Run two to three paywall experiments per quarter** with ARPU as your decision metric, not raw conversion.
The teams that win in concentrated 2026 markets, where the top **10%** of apps now capture **95%** of subscription revenue (up from ~85% in prior Adapty cuts), are the ones who measure faster and iterate harder. They treat their paywall as a product surface that ships weekly.
The 2026 winners are not the apps picking the right model on Day 0. They are the apps shipping the next test this quarter.
---
## Weekly vs Annual Subscription Apps: How the 2026 Revenue Split Is Shifting
URL: https://www.airbridge.io/en/blog/weekly-vs-annual-subscription-app
Language: English
Published: 2026-05-27
Author: Luke
> **The short answer:** Weekly plans now generate **55.5%** of all in-app subscription revenue, up from 43.3% in 2023. Monthly lost nearly half its share. Annual still dominates Health and Fitness. The right plan type for your app depends on category, not the global average.
Two years ago, weekly plans generated **43.3%** of all app subscription revenue. By 2025, that figure had climbed to **55.5%**, according to [Adapty's State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/), which analyzed 16,000 apps and over $3B in subscription revenue. Monthly plans lost nearly half their share in the same window, falling from 21.1% to 11.7%. Annual slipped from 29.2% to 22.5%.
The subscription app market is reorganizing around shorter billing cycles, but the shift is not uniform. Health and Fitness doubled down on annual. Photo and Video abandoned monthly almost entirely. The plan type that works for your app depends on your category, your paywall, and how fast users can find value. **Treating "weekly by default" as a universal rule is the fastest way to leave revenue on the table.**
**Key Takeaways**
- **Weekly plans now generate 55.5% of all in-app subscription revenue**, up from **43.3%** in 2023. Monthly share has collapsed to **11.7%**.
- **Weekly converts 1.7 to 7.4x better than annual** across every price tier, but Day 380 trial-cohort retention sits at just **5.5%** versus **19.9%** for annual.
- **Weekly plus a free trial produces the highest 12-month LTV of any setup at $49.27.** The economics rely on volume, not loyalty.
- **Health and Fitness is the only category where annual dominates**, at **60.6%** of category revenue. Utilities is the opposite, with **73.6%** from weekly.
- **Monthly is in structural decline.** It is the most price-sensitive plan type and rarely the optimal lead in 2026.
- **There is no universal default.** Match plan type to category, validate with tests, and treat any blog-wide recommendation with skepticism.
## Subscription App Revenue Share by Plan Type: 2023 vs 2025
The headline numbers from the 2026 in-app subscription benchmarks are blunt. Across 16,000 apps and over $3B in tracked subscription revenue, the share of dollars flowing through weekly billing cycles has crossed the majority threshold. Two years ago, weekly was the largest of three roughly comparable shares. Today, it is bigger than monthly and annual combined.
| Plan type | 2023 revenue share | 2025 revenue share | Change |
| --- | --- | --- | --- |
| Weekly | 43.3% | 55.5% | +12.2pp |
| Monthly | 21.1% | 11.7% | -9.4pp |
| Annual | 29.2% | 22.5% | -6.7pp |
| Lifetime and one-time | 6.4% | 10.3% | +3.9pp |
_Source:_ [_Adapty State of In-App Subscriptions 2026_](https://adapty.io/blog/weekly-monthly-annual-subscription-plan/)_._
The pattern is hard to read as anything other than a structural shift. Weekly did not just outgrow its peers, it cannibalized them. Monthly took the worst of it, losing nearly half its share in 24 months. Annual gave up ground too, although less.
A second number explains why this is happening. Across all price tiers and most categories, weekly plans convert at **1.7 to 7.4 times** the rate of annual plans on the same paywall. When the same screen sells two products and one converts seven times better, that product wins. Layer in payment accessibility, anchoring effects, and an audience wary of long commitments, and weekly's dominance starts to look like a baseline.
The weekly vs annual subscription app revenue trends emerging in 2026 are clear in direction. The harder question is whether they apply to your subscription app at all.
[Screenshot 2026-05-27 at 17.31.49.png]
## Best Subscription Plan Type by App Category
Industry-wide averages mask the real story. Plan type performance varies sharply by app category. In some verticals, weekly is now the default to beat. In others, it is the wrong answer.
Utilities apps generate **73.6%** of their revenue from weekly plans. Entertainment is similar. These categories share a trait: users want access at low commitment, repeatable in nature, with immediate value. The pitch is "stream this now" or "use this tool to finish this task," not "invest in your year."
Health and Fitness is the visible exception. Annual plans dominate the category, up from 51% revenue share in 2023. Fitness is aspirational, so signing an annual commitment matches the framing users bring to the purchase. The same data set shows Health and Fitness has the highest trial-to-paid conversion of any category at **35%**, paired with the lowest first-renewal retention at **30.3%** (Adapty SOIS 2026). Users commit fast and churn fast. Annual locks in revenue before motivation fades.
| Category | Dominant plan | Notes |
| --- | --- | --- |
| Utilities | Weekly (73.6%) | Highest 12-month trial-cohort LTV at $68.90 |
| Entertainment | Weekly | Heavy weekly share, refund spikes in APAC |
| Photo and Video | Weekly | Declining concentration as AI tools enter |
| Gaming | Weekly | RevenueCat reports gaming weekly share at 77% |
| Health and Fitness | Annual (60.6%) | Only category where annual share increased since 2023 |
| Productivity | Mixed | Direct buyers worth $56.95 vs $49.13 for trial users |
| Lifestyle | Mixed | Trials reduce 12-month LTV by 21.2% |
| Education | Annual or weekly | Annual LTV around $43 is solid, weekly competitive |
_Source: Adapty SOIS 2026, except where noted._
Two patterns are worth pulling out:
1. **Regional skew matters.** [RevenueCat's analysis of weekly plan adoption](https://www.revenuecat.com/blog/growth/weekly-subscriptions/) shows India, Southeast Asia, the Middle East, and parts of Latin America skew heavier toward weekly than North America does. The driver is payment infrastructure and disposable-income patterns more than user preference.
2. **Lifestyle is the rule-breaker.** It is the only category where adding a free trial actively reduces 12-month LTV (down 21.2%). If you are running trial-heavy acquisition for a Lifestyle app, the data argues you are systematically pulling in lower-quality subscribers.
## Subscription App LTV by Plan Type: Why $49.27 Weekly Beats $36.51 Annual
Weekly wins on conversion. Annual wins on retention. The interesting question is which one wins on lifetime revenue.
| Plan type | 12-month LTV (trial-starting cohort) | Day 380 trial-cohort retention | First-renewal rate |
| --- | --- | --- | --- |
| Weekly + trial | $49.27 | 5.5% | ~74 to 81% |
| Monthly + trial | ~$28 to $30 | 14.2% | 48 to 58% |
| Annual + trial | $36.51 | 19.9% | 66.3% |
_Source: Adapty State of In-App Subscriptions 2026 (trial-starting cohorts)._
### Why Weekly Plus Trial Wins on 12-Month LTV
Across all categories combined, weekly plus a free trial produces the highest 12-month LTV at **$49.27** per user, ahead of annual plus trial at **$36.51**. The mechanism is volume. Day 0 LTV on a weekly plan is the lowest of any configuration: you collect $7 to $9 upfront instead of $38 to $54 on annual. But the repeated short renewal cycles compound, and when paired with a trial, the renewal rate at the first weekly checkpoint stays above 70% on average.
### The Retention Catch
Only **5.5%** of weekly trial-starting subscribers are still active at Day 380. Annual trial subscribers retain at **19.9%** at the same checkpoint, monthly at **14.2%**. Weekly LTV depends entirely on a meaningful share of cohorts surviving the first six renewals. If first-week retention slips below the mid-teens, the math collapses fast.
### How Annual Plays a Different Game
Annual offers lower conversion volume, much higher per-user value, and a long retention runway. The churn pattern for annual is also unique: users tend to stay through the full 12-month cycle because they have already paid, then churn in the renewal window. Your retention efforts around **month 10 to 11** matter more than anything you do in the first month.
One more counterintuitive note from the same data set: **weekly plans are the least price-sensitive of any plan type at renewal.** Weekly first-renewal stays in a tight 74 to 81% band across every price tier in the Adapty cut. If you are on weekly today, raising your price is lower risk than your instinct suggests.
[Screenshot 2026-05-27 at 17.32.22.png]
## Why Monthly Subscription Plans Are in Structural Decline
Monthly used to be the safe middle. In 2025, it is the worst seat at the table for most categories. It converts worse than weekly, retains worse than annual, and gets squeezed harder than either when prices move.
### What Changed: Monthly Lost Nearly Half Its Revenue Share
Two years ago, monthly plans collected **21.1%** of all subscription revenue. In 2025, that share is **11.7%**. The drop is steeper than annual's decline in the same window, and it shows no sign of reversing. Per the Adapty SOIS 2026 dataset, the decline is broad-based, hitting most categories rather than a single vertical.
### Why Monthly Is the Most Price-Sensitive Plan Type
The price-sensitivity data is the clearest signal that monthly's middle position is failing. Three points stand out:
- **First-renewal rates on monthly plans drop 8 to 10 percentage points** as price moves from the cheapest tier to the most expensive.
- **Weekly first-renewal stays in a tight 74 to 81% band** across price buckets. Price barely moves the needle.
- **Annual varies by category** but never as steeply as monthly.
Whatever buffer monthly used to have between low-commitment users and high-commitment ones has eroded. Users either want the lowest commitment they can find (weekly) or are willing to commit for the full value cycle (annual).
### Where Monthly Still Wins: Lifestyle Apps
There is one category where monthly still earns its place as a lead plan: **Lifestyle.** What makes Lifestyle different is its trial behavior. It is the only category where adding a free trial reduces 12-month LTV by **21.2%** (direct buyers at **$42.80** versus trial starters at **$33.80**, per Adapty SOIS 2026). The combination of solid monthly economics and trial-hostile dynamics makes "monthly without trial" a defensible default for Lifestyle apps. That pattern does not transfer to other categories.
### European Monthly Plans Carry Extra Price-Sensitivity Risk
European monthly subscribers are unusually price-sensitive at renewal. Low-priced European monthly apps see roughly **40%** higher first-renewal rates than high-priced ones. If you are running monthly in Europe, pricing decisions carry more downside risk than anywhere else in the dataset.
### How to Use Monthly in 2026: As an Anchor, Not a Lead
For everyone outside Lifestyle, monthly works best as an anchor option on a three-plan paywall, not as the primary offer. Two reasons:
1. **Anchoring lifts the surrounding plans.** Showing weekly, monthly, and annual together generally outperforms single-plan paywalls. Annual makes monthly look fair. Monthly makes weekly look accessible.
2. **Leading with monthly starts you in a worse position.** Conversion is below weekly and retention is below annual. The category-level data does not support monthly as a lead choice in 2026.
## How to Choose the Right Subscription Plan Type for Your App in 2026
The temptation when reading data like this is to swap weekly into the default slot and ship it. The data does not actually support that move. What it supports is a category-driven decision, validated with structured tests against the cohorts you already have.
A practical decision framework, distilled from the Adapty SOIS 2026 dataset:
1. **Start from your category, not the global average.** Health and Fitness, Education, and Travel skew annual. Utilities, Entertainment, Gaming, and Photo and Video skew weekly. Productivity and Lifestyle are split.
2. **Match plan duration to value frequency.** If users get value in a single session, weekly is plausible. If value compounds over weeks, annual with a trial usually beats weekly.
3. **Audit the trial-vs-direct LTV gap in your category.** In Productivity ($56.95 direct vs $49.13 trial) and Lifestyle ($42.80 vs $33.80), direct buyers are more valuable than trial users. Adding a trial there may quietly lower subscriber quality.
4. **Price up before changing plan duration.** High-priced apps generate **3x** the LTV of low-priced apps in most categories. In Health and Fitness, expensive annual plans generate **4.5x** the LTV of cheap ones. Pricing is usually the lever, not duration.
5. **Test, do not adopt by default.** Apps that run **50 or more paywall experiments earn 18.7x more revenue** than apps that run just one. The win is in the testing process, not the choice of plan type.
## Subscription App Plan Strategy: What to Watch Next
If you are a Utilities or Entertainment app still leading with monthly, you are leaving the majority of category revenue to weekly-led competitors. If you are a Health and Fitness app defaulting to weekly, the data argues you are burning the annual ceiling.
The next twelve months will tell whether weekly's share crests above 60% or whether the rebalance has overshot. Watch the 2027 Adapty cut for the inflection. In the meantime, pick the plan your category supports, validate with structured tests, and re-check as activation matures.
---
## Does Web to App Improve Profitability? The Real Numbers
URL: https://www.airbridge.io/en/blog/does-web-to-app-improve-profitability
Language: English
Published: 2026-05-27
Author: Harper (Trang Nguyen)
Every subscription app marketer has heard the pitch: skip Apple's 30% cut, use Stripe at 3%, and keep more of your revenue. It sounds like free money. But in RevenueCat's controlled IAP-vs-web test on a subscription audio app, web subscriptions generated just **$0.93 for every $1.00 earned through IAP**, even after accounting for the full App Store fee savings. ([RevenueCat, 2025](https://www.revenuecat.com/blog/growth/iap-vs-web-purchases-conversion-test/))
The economics of web-to-app are real but widely misunderstood. The mistake most founders make is treating it as a billing optimization. It is a distinct growth channel with its own conversion dynamics, audience reach, and operational overhead. Get those right, and web-to-app can meaningfully extend your revenue ceiling.
---
**Key Takeaways**
- **If your annual ARPU is below $60, stay with IAP.** The absolute dollar savings from web checkout are too small to offset a typical 25-35% conversion drop at most low-to-mid price points.
- **Below ~18% web conversion, the channel loses money.** That is roughly the break-even threshold against a 25% IAP conversion rate at standard fee levels. Most web funnels start here, so validate before you build.
- **Validate your audience first: re-routing existing App Store traffic destroys value.** Web funnels work when they reach genuinely new users. If your web campaign targets the same audience as your install campaigns, you are adding friction to a funnel that already works.
- **B2B or invoice billing? IAP cannot serve you.** Web checkout is the only viable path for apps that need team accounts, corporate purchasing, or invoice-based billing.
- **Measure web-to-app like any other growth channel.** Connecting ad spend to downstream subscription events, not just installs, is what separates teams that scale web funnels from those that abandon them.
## What Web to App Actually Is (and Why It Became Popular)
Web-to-app is a funnel where paid advertising drives users to a web landing page first. The user completes payment on the web via Stripe or a similar processor, then downloads the app to access their subscription. Payment is processed outside the App Store, bypassing Apple's and Google's commission structures.
### 1\. How the funnel works
The typical flow: run a paid ad on Meta, Google, or TikTok; user lands on a web landing page explaining your product; user enters payment details; confirmation screen delivers an App Store download link; user downloads the app and logs in to access premium content.
That added friction is the central tension in every web-to-app analysis — and the data-driven evidence for it comes through clearly in conversion rates.
### 2\. Why it became a mainstream strategy
Three forces pushed web-to-app into mainstream app marketing conversation.
**Regulatory pressure on Apple.** Apple is facing growing pressure from regulators and courts around the world to open up its payment and distribution policies. That shift has made it easier for developers to direct users to web checkout, and has put web-to-app strategies on the radar for more app marketers than ever before.
**Rising paid UA costs across the App Store.** Health and Fitness iOS ad spend grew **97%** year-over-year in 2025, with the top five apps capturing **73% of category UA spend**, up from **54%** the year before. ([AppsFlyer State of Subscription Apps 2026](https://www.appsflyer.com/resources/reports/state-of-subscription-apps/)) Web funnels offered access to different audiences and different ad formats outside the install auction.
**The commission math looks obvious on paper.** Apple's standard commission is 30% (15% after the first year of a subscription). Stripe charges 2.9% plus a per-transaction fee. The gap is visible. The catch is what happens to conversion rates and subscriber LTV when you redirect users through a web checkout.
## The Fee Savings Calculation: Where Most Apps Go Wrong
The headline promise of web-to-app, saving 15-30% on every subscription, is technically accurate. The problem is what it ignores.
### 1\. Conversion drops more than expected
In a single-app A/B test run by RevenueCat (Dipsea, a subscription audio app, ~3,100 users per variant), the IAP variant achieved a trial start rate of approximately **27.0%**. The web variant: **18.1%**. That is a **33%** relative decline in the number of users who began a trial. ([RevenueCat, 2025](https://www.revenuecat.com/blog/growth/iap-vs-web-purchases-conversion-test/))
The friction of leaving the app, entering credit card details, and returning is real. It filters out lower-intent users who would have tapped "Start Free Trial" on impulse.
### 2\. The bottom line: web earns $0.93 per $1.00 IAP makes
Trial-to-paid conversion was marginally better for web users: **26.3% vs. 25.0% for IAP**. ([RevenueCat, 2025](https://www.revenuecat.com/blog/growth/iap-vs-web-purchases-conversion-test/)) The higher-friction path filtered out low-commitment trialists. But with 33% fewer trial starts and only a small conversion improvement, web checkout produced **$0.93 in take-home revenue for every $1.00 earned through IAP**. The fee savings were real. The conversion loss was larger.
| Metric | IAP | Web Checkout | Difference |
| --- | --- | --- | --- |
| Trial start rate | ~27.0% | ~18.1% | -33% |
| Trial to paid | ~25.0% | ~26.3% | +1.3pp |
| Take-home per $1.00 IAP | $1.00 | $0.93 | -7% |
_Source: RevenueCat IAP vs. web conversion test._
### 3\. The retention curve reverses over time
Data from 16,000+ apps adds another layer of complexity. Web-acquired subscribers show a striking early retention advantage: Month 1 retention is **84.5% for web vs. 48.2% for in-app**. The users who navigated web checkout were clearly more committed up front.
By month 8, the curve reverses. Web retention falls to around **20%**, while in-app subscribers retain at **30%**. The long-term LTV of web subscribers averages **$10.8**, well below in-app at **$40.1**. This gap persists even after adjusting for the fee differential.
| Period | Web Retention | In-App Retention |
| --- | --- | --- |
| Month 1 | 84.5% | 48.2% |
| Month 8 | 20% | 30% |
| Average LTV | $10.8 | $40.1 |
_Source: Adapty analysis of 16,000+ apps._
The interpretation: web checkout filters out impulse subscribers who would have converted via IAP but churned in the first 30 days. But something about the web subscriber experience drives faster disengagement after the first few months, eroding the initial retention advantage. Notably, even after stripping out Apple's commission from the IAP figures, web subscriber LTV still comes in roughly $4 lower, meaning the gap is not a fee story, it is a retention story.
## When Web to App Improves Economics (and When It Doesn't)
The data above describes the average outcome. Averages hide both the scenarios where web-to-app creates durable value and the scenarios where it destroys it.
If your annual ARPU is below $60 and you cannot identify a web-only audience, web-to-app will likely destroy value. The four conditions below stack — apps seeing durable ROI from web funnels typically hit at least two.
### When it improves your economics
**1\. High-priced annual plans where fee savings are significant in absolute dollars**
The break-even conversion rate is roughly the same regardless of price (around 21-22%), but the absolute dollar difference at stake scales significantly with your subscription price. The table below shows net revenue per user acquired at two realistic web conversion rates — 15% (common in early testing) and 22% (well-optimized funnel).
| Plan | Price | Net per IAP user (25% conv) | Net per Web user (15% conv) | Net per Web user (22% conv) |
| --- | --- | --- | --- | --- |
| Monthly low | $6.99/mo | $1.49 | $1.02 | $1.49 |
| Annual mid | $49.99/yr | $10.62 | $7.27 | $10.66 |
| Annual premium | $99.99/yr | $21.25 | $14.55 | $21.34 |
_Assumes 15% Apple commission (year 2+), 3% Stripe fee. Net = price × (1 - fee) × conversion rate._
At a $6.99/month plan, a web funnel converting at 15% earns $0.47 less per user than IAP at 25% — and that gap widens as the team absorbs landing page, compliance, and entitlement overhead. At $99.99/year, converting at 22% nearly matches IAP economics, and any improvement beyond that produces durable margin. Most web funnels convert at 15-20% in early testing, which means the math becomes viable at annual prices above $60-80.
**2\. Reaching genuinely new audiences through web ad channels**
This is the insight experienced practitioners emphasize most: the real value of web-to-app is audience expansion, not fee avoidance. ([Thomas Petit, Sub Club Podcast, 2024](https://www.revenuecat.com/blog/growth/thomas-petit-sub-club-podcast-2024-web-to-app/))
Web-acquired users share only about **15% overlap** with mobile-acquired users. (Paddle data, via [Business of Apps](https://www.businessofapps.com/insights/winning-web-revenue-in-the-evolving-app-landscape/)) A well-executed web funnel opens a growth channel that your app store page cannot reach: users who respond to long-form web copy, desktop users who research subscription products before downloading, and audiences Meta's algorithm identifies through web behavioral signals that differ from its app-install optimization model.
**3\. B2B use cases that in-app purchases cannot serve**
Apple and Google have not built robust team account or invoice-based billing for enterprise use. If your app serves businesses, IAP creates friction that web checkout eliminates. Users in B2B contexts need to pay personally and expense it, or the company cannot purchase a team subscription at all. For B2B-adjacent apps, web checkout unlocks revenue that IAP structurally cannot capture.
**4\. Pricing flexibility and faster experimentation cycles**
Web checkout lets you A/B test pricing and promotional offers without waiting for App Store review cycles. Combined with email capture at checkout, web-to-app creates a CRM asset that supports lifecycle marketing in ways the App Store funnel cannot.
### When it will hurt your profitability
**You are rerouting existing App Store traffic.** If your web funnel is running ads to the same audience that would have found you through the App Store, you are adding friction to a conversion path that already works. You will lose 25-35% of conversions and net less revenue. Web-to-app creates incremental value when it accesses new audiences. It destroys value when it processes traffic that would have converted natively.
**Your app relies on App Store discovery for growth.** Apps that grow through strong ASO rankings, App Store featuring, or category search have a distribution advantage tied to the platform. Building a parallel web funnel diverts attention and budget from a channel that is already working efficiently.
**You underestimate the ongoing operational cost.** As Thomas Petit, who has guided significant paid UA spend for subscription apps, put it directly: "Many people who move to web-to-app underestimate the amount of work that it takes to maintain it." ([Sub Club Podcast, 2024](https://www.revenuecat.com/blog/growth/thomas-petit-sub-club-podcast-2024-web-to-app/)) A web funnel requires sustained attention to landing pages, payment processor compliance, subscription entitlement sync, and separate [attribution](https://www.airbridge.io/en/glossary/attribution-window) setup. For a two-person team, this overhead compounds quickly.
## Quick Answers: Common Web to App Questions
### Q1. Is web to app legal on iOS in the US in 2026?
Yes. Following court rulings in 2025 that established iOS apps can link to external payment flows, Apple filed for a Supreme Court stay in May 2026, which Justice Kagan denied on May 6. External-payment links remain permitted on iOS in the US for now. In the EU, the Digital Markets Act created similar permissions for EU App Store apps.
### Q2. What do I need to build a basic web to app funnel?
At minimum: a web landing page with a clear value proposition; a web payment processor such as Stripe, Paddle, or Lemon Squeezy (Paddle and Lemon Squeezy handle international tax compliance automatically); a subscription management platform that links web purchases to in-app entitlements; and a post-purchase flow that delivers the App Store download link. Email capture at checkout is strongly recommended, as this is the primary CRM asset that web-to-app creates over IAP.
## Web to App Is a Growth Channel, Not a Billing Hack
The most expensive mistake in web-to-app is applying it as a cost-reduction tactic. The fee savings are real. But they are smaller than they look, and they do not survive a typical conversion drop without the right conditions in place.
**Before you build, run this three-step check:**
- **Audit your ARPU.** If you are below $60/year, the fee savings at realistic web conversion rates will not cover the operational overhead.
- **Define your web-only audience hypothesis.** Which users would respond to your web campaign but would not find you on the App Store? If you cannot answer this, you are rerouting existing traffic, not adding incremental revenue.
- **Test before you scale.** Run $2,000-5,000 in web spend first. Validate that web-acquired users have low overlap with your existing cohorts and that your landing page converts above 20% before committing to full funnel infrastructure.
**Measure it like any other growth channel.**
The apps using web-to-app effectively are reaching new users, building lifecycle marketing lists, and running pricing experiments faster than any App Store cycle allows. But those advantages only compound when you can see what is actually working. If you are running paid campaigns on Meta, Google, or TikTok to a web checkout, you need visibility into what happens after the click: which campaigns drive trial starts, which convert to paid subscribers, and how web-acquired cohorts compare to App Store users. [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects ad spend to subscription revenue events across both web and app funnels, so you can make channel decisions based on subscriber quality, not just install volume. It supports RevenueCat integration and starts free with 15K attributed installs.
**Ready to measure your full subscription funnel, from first ad click to paid subscriber?**
[Start free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core)
---
## How to Ask "How Did You Hear About Us?" in Your App the Right Way
URL: https://www.airbridge.io/en/blog/correctly-ask-how-did-you-hear-about-us
Language: English
Published: 2026-05-26
Author: Luke
You added a "how did you hear about us?" screen to your onboarding. Users are answering. Half say Instagram. You increase your Meta budget. Three months later, your cost per subscription is up and your ROAS is down.
The problem was not the channel. **The problem was the question.**
Self-reported attribution (asking users directly where they first heard about your app) surfaces signal that no ad platform or attribution tool can give you. Done wrong, it floods your dashboard with noise that steers every budget decision in the wrong direction.
This guide covers how to design the question, where to place it, how to read the data, and when to stop relying on it.
**Key Takeaways**
- **Self-reported data fills an iOS attribution gap that probabilistic methods cannot close.** When platforms like Meta and TikTok under-attribute through AEM or ADC, asking users directly is one of the few reliable cross-checks available to app marketing teams.
- **Question design and placement determine data quality.** The same "how did you hear about us?" question placed at different points in onboarding produces significantly different results, with different biases baked in.
- **Survey data reflects perceived influence, not causal attribution.** Users pick the most familiar option, not necessarily the true source. This distinction matters when you act on the data.
- **HDYHAU works best when your media mix is simple.** It becomes misleading with demand-side platforms and channels users cannot easily name by brand.
- **The real value comes from comparing survey data against subscription revenue,** not just against install counts.
## Why Self-Reported Attribution Matters for Mobile App Marketing
### The iOS Attribution Problem
iOS is the primary revenue pipeline for most subscription apps. Among [**75,000+** subscription apps analyzed](https://www.revenuecat.com/state-of-subscription-apps/), **67-76%** earn **80%+** of their revenue from iOS. Across the broader market, [iOS generates **5.6x** more subscription revenue than Android](https://adapty.io/state-of-in-app-subscriptions/) and converts annual plans at **3.6x** the Android rate.
But iOS is also where attribution data is hardest.
Apple's App Tracking Transparency (ATT) limits cross-app data collection. [SKAdNetwork (SKAN)](https://developer.apple.com/documentation/storekit/skadnetwork/) delivers postbacks with **24-72** hour delays and no user-level identifiers. Probabilistic methods like Meta's AEM and TikTok's ADC fill part of the gap, but systematically under-attribute.
### What Under-Attribution Costs You in Practice
An app spending five figures per month on Meta might see **200** subscriptions reported by its attribution tool. The actual number, verified through self-reported data, can be closer to **350** — **3,000** untracked installs per month at a **5%** conversion rate. _(Source:_ [_David Vargas, RevenueCat, 2025_](https://www.revenuecat.com/blog/growth/how-did-you-hear-about-us-surveys/)_)_
Self-reported attribution is not a replacement for your attribution tool. It is a cross-check — telling you which channels are under-attributed and by how much, without depending on probabilistic models or device identifiers. The "how did you hear about us app" question is the simplest form of zero-party data available to app marketers.
[Screenshot 2026-05-26 at 17.42.48.png]
## How to Design an HDYHAU Survey Question That Gets Honest Answers
Most HDYHAU questions fail because they are designed to be convenient for the team, not honest for the user.
The difference between a question that collects real signal and one that collects noise comes down to four factors: timing, answer options, order, and context.
### 1\. When to Show the HDYHAU Question: Best Placement in Your Onboarding Flow
The best moment to ask is right after the user completes a meaningful action: immediately after onboarding, after starting a free trial, or after their first purchase. [Top-performing brands see **45-85% completion rates**](https://fairing.co/resources/guides/hdyhau-attribution-survey-best-practices-guide) when surveys are timed to post-conversion moments, up to 10x higher than the same question delivered by delayed email or SMS.
Avoid asking before users understand what your app does. Pre-onboarding answers are rushed and low-quality.
For subscription apps, the strongest placement is post-paywall, right after a trial start or first purchase. Users who have converted give more intentional answers because their recall is tied to an actual decision.
### 2\. Which Answer Options to Include in Your HDYHAU Survey
Each active paid channel needs its own option. Bucketing them as "social media" destroys the attribution signal you are trying to capture.
A practical option list for a subscription app running standard paid UA channels:
| Option | When to Include |
| --- | --- |
| Instagram / Facebook | Always, if running Meta campaigns |
| TikTok | Always, if running TikTok campaigns |
| Google / YouTube | Always, if running Google Ads |
| Apple Search Ads | If running ASA campaigns |
| Friend or family recommendation | Always (word-of-mouth signal) |
| Podcast / newsletter | If you run any content channels |
| Other (please specify) | Always, as a catch-all with optional text input |
Do not add options for inactive channels — they invite noise. Include "Other" with a text field as a catch-all; it captures unanticipated sources and gives early signal when a new channel starts working.
### 3\. How to Order Survey Options to Reduce Anchoring Bias
[Peer-reviewed research on survey question-order effects](https://academic.oup.com/erae/article/51/4/1129/7879404) confirms that respondents systematically favor options that appear first, regardless of whether those options reflect their true answer.
Randomize your answer order for each user. This distributes selection bias evenly across options and makes your aggregate data more accurate over time. The impact on any individual response is small, but across thousands of responses it removes a systematic skew.
One exception: put "Other" last. Users who select "Other" are actively overriding the offered options. Placing "Other" first changes what it captures.
### 4\. What to Tell Users Before They Answer
A single-sentence explanation ("This helps us know which channels are working so we can improve them") meaningfully improves completion. [Research confirms that stating your purpose](https://academic.oup.com/poq/article/83/S1/185/5520299) builds trust and encourages considered answers. Keep it to one sentence — longer reads as defensive.
## How to Analyze Your "How Did You Hear About Us?" Survey Data
Collecting the data is the easy part. Interpreting it correctly is where most app marketing teams go wrong.
### How to Use HDYHAU Data to Spot Under-Attribution
Compare your HDYHAU responses against what your attribution tool reports by channel. If the tool credits Meta with **200** subscriptions but HDYHAU shows **350** users selecting "Instagram" or "Facebook," you have a **1.5x** correction factor to apply to your CAC calculation.
David Vargas at RevenueCat documented exactly this scenario: the gap drops an official CAC of **$50** to roughly **$29**, and moves the profitability breakeven against LTV from day **90** to day **60**. For a subscription app where cash flow determines scaling speed, that difference is not cosmetic. _(Source:_ [_David Vargas, RevenueCat, 2025_](https://www.revenuecat.com/blog/growth/how-did-you-hear-about-us-surveys/)_)_
### Brand-Familiarity Bias and Where It Breaks (DSPs)
HDYHAU surveys do not capture causal attribution. They capture perceived influence.
When users answer, they are asking themselves: "Where do I remember encountering this app?" Meta, Google, and TikTok benefit from this bias because they are culturally dominant — even users who found your app through an influencer, podcast, or DSP will often select the most familiar platform option.
Demand-side platforms (DSPs) are where this breaks down most visibly. DSPs buy inventory across thousands of apps and placements: games, utility apps, waiting screens. When users encounter your ad in a game and convert, they will not report the channel that served it. They will report Meta or Google. If you spend on DSPs and rely solely on HDYHAU, you will likely credit DSP-driven conversions to your standard social campaigns — making DSP appear to under-perform and social appear to over-perform until DSP spend gets cut.
This is not a flaw in the data as much as a known limitation: it measures brand familiarity, not causality. Knowing the distinction prevents you from making the wrong cut when a new channel fails to appear in your survey results.
### How to Use the "Other" Response as an Early Channel Signal
When you launch a new channel, you will not see it in your HDYHAU options yet — but you will see a rise in "Other" responses. Monitor "Other" as a percentage of total responses after each new channel activation. A sustained rise that coincides with new spend is directional evidence the channel is generating awareness before you can name it explicitly.
## When HDYHAU Surveys Are Reliable (and When They Mislead You)
Its usefulness depends directly on how complex your media mix is.
| Scenario | HDYHAU Reliability | Why |
| --- | --- | --- |
| 1-2 social channels (Meta or TikTok only) | High | Options map cleanly to user experience; minimal confusion |
| 3-4 standard channels (Meta, Google, TikTok, ASA) | Medium | Some cross-channel confusion, but still directional |
| 5+ channels including DSPs | Low | Users cannot distinguish DSP inventory from social; brand familiarity dominates |
| High-spend apps with broad brand presence | Low | Users "see you everywhere" and select the most familiar platform |
_Reliability tiers are heuristic, based on Vargas (RevenueCat, 2025) and practitioner analysis across subscription app campaigns. Your threshold will vary with brand awareness and channel mix._
## What to Use Instead When You Run More Than Three Paid Channels
Once self-reported attribution starts masking channel-level differences, pair it with methods that measure causality, not recall.
The three most practical options: **incrementality tests** (hold a region or segment out of a campaign and measure the conversion-rate difference — most reliable for validating new channels); **Ads ON/OFF tests** (pause one channel while holding others flat and watch whether volume drops); and **budget ramp tests** (scale a new channel from minimal spend while holding others flat and track whether baseline conversion rises). All three are slower than a survey, but they measure whether a channel actually causes subscriptions — the question that matters for budget decisions.
Self-reported data remains useful alongside these methods for qualitative context: which channels are building familiarity, where users think they came from, and which "Other" categories are growing.
## How to Connect HDYHAU Data to Subscription Revenue
### Grouping Users by Survey Response
Group users by their HDYHAU response ("Instagram users," "Google users," "Friend recommendation users") and compare trial-to-paid conversion rates, trial lengths, and 12-month retention. The picture looks very different from installs alone.
Industry data shows that paid users tend to churn at a higher rate than organic ones, making word-of-mouth acquisition more durable than its raw install count suggests. If "Friend or family" starts growing as a share of your HDYHAU responses, treat it as a high-value cohort signal worth tracking separately. _(Source:_ [_RevenueCat, 2026_](https://www.revenuecat.com/blog/growth/subscription-app-trends-benchmarks-2026/)_)_
## Your Next Steps with Self-Reported Attribution
The question only earns its place in onboarding if the data changes a decision. Three steps to make that happen:
- **This week:** Audit your current HDYHAU placement against the four design factors above. Is it post-conversion? Does each active paid channel have its own answer option? Is answer order randomized?
- **This month:** Cross-check your top-channel HDYHAU responses against your attribution-tool numbers to compute a correction factor. If HDYHAU shows **1.5x** the subscriptions your tool credits to a channel, that ratio is your working adjustment for CAC.
- **Next quarter:** If you are running 3 or more paid channels, schedule a single-channel holdout test to measure causal lift. Survey responses tell you where users think they came from; a holdout test tells you whether the channel is actually responsible.
---
## What to Do After a User Rejects Your Paywall: A 4-Layer Recovery Strategy
URL: https://www.airbridge.io/en/blog/what-happens-after-user-rejects-paywall
Language: English
Published: 2026-05-25
Author: Luke
Benchmark data from the [RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/) and [Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/) show that median free-to-paid conversion rates sit at **2.2%** for freemium models and **12.1%** for hard paywalls — meaning more than 9 in 10 users who see a paywall will not convert on the first view. That pattern holds across app categories, pricing models, and onboarding flows.
**The real problem is not the low conversion rate. It is what happens next.** For most subscription apps, the answer is nothing. The user taps the close button, the paywall disappears, and the moment is gone. That silence after a paywall rejection is where a significant share of potential subscription revenue walks out the door.
**Key Takeaways**
- **Most users reject the first paywall view.** A post-paywall recovery strategy is as important as the paywall itself. Most teams have one but not the other.
- **The dismiss is not always a final "no."** Data from the [Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/) shows that **55.5%** of all eventual subscribers did not convert on Day 0, meaning most conversions require more than one touchpoint.
- **Exit offers recover conversions by presenting a lower-commitment alternative before the user leaves the screen.** A [Superwall study across 18 apps and approximately 500,000 users](https://superwall.com/blog/17-revenue-boost-with-transaction-abandon-paywalls-a-case-study/) found that exit offers shown at the moment of abandonment accounted for 17% of total revenue.
- **The best exit offer is a different format, not just a lower price.** Monthly after an annual pitch, or a lifetime option after a subscription, reframes the decision rather than discounting the same one.
- **iOS exit offers carry App Store rejection risk** under Apple Guidelines 5.6 and 3.1.2. Test on Android first, where no similar rejection patterns have been reported.
## Why Paywall Rejection Is a Revenue Problem, Not a Value Problem
### 1\. The Dismiss Is a Timing Signal, Not a Final No
The paywall dismiss has a specific meaning in the data. The [Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/), based on 16,000+ apps representing $3 billion in subscription revenue, shows that **44.5%** of all subscription purchases happen on Day 0. The remaining **55.5%** of purchases occur on later days. That gap represents users who saw the app, felt interest, and chose not to commit immediately.
Some of those users left permanently. But a meaningful share was not saying "I don't want this." They were saying "not right now," distracted, uncertain about value, or not ready to commit to a full-year plan after **90 seconds** with an app.
The paywall dismiss and the final decision to never subscribe are not the same event. Treating them identically forfeits the conversions a different offer or different timing would have captured.
### 2\. The Recovery Window Is Measured in Minutes
The window to recover a dismissed paywall is measured in minutes, not days. A user who just closed your paywall is still in the app, still in the session, still processing the decision. That is as close to conversion intent as you will find outside the paywall moment itself.
Send the first push notification within the first hour of exit. Waiting until the next morning means reaching a user whose decision context has largely dissolved.
### 3\. Most Apps Do Nothing After a Dismiss
The majority of subscription apps have no designed response to a paywall dismiss. No exit offer. No push notification within the session. No follow-up sequence. The user exits the paywall, the app continues in its default state, and the conversion opportunity closes.
Most teams instrument the dismiss event but never wire it to a downstream action. Apps that build a structured post-paywall recovery strategy treat the dismiss as the beginning of a secondary funnel, not the end of the primary one. What to show after a paywall decline is a question worth planning for before it happens.
## The 4-Layer Post-Paywall Recovery Strategy
A complete recovery strategy works in sequence. Each layer targets users at a different point in the recovery window.
[Screenshot 2026-05-25 at 18.07.43.png]
### 1\. Layer 1: Show an Exit Offer at the Moment of Dismiss (0-30 Seconds)
An exit offer fires on dismiss, swapping the closing paywall for an alternative before the user leaves the screen. Throughout this article, "exit offer" refers to any secondary offer shown at the moment of paywall dismissal.
Exit offers are standard in e-commerce to reduce cart abandonment, and they translate directly to subscription apps. [RevenueCat added native exit offer support to their Paywall Builder in December 2025](https://www.revenuecat.com/release/show-exit-offers-when-a-paywall-is-dismissed-2025-12-31), allowing implementation without code changes: select an alternative offering in the dashboard, and the exit offer appears on dismiss automatically.
Effective exit offer formats:
- **Plan format switch**: If your primary paywall leads with annual, the exit offer shows monthly. This captures users who were interested but not ready to commit for a full year.
- **Extended trial**: A user who dismissed a 7-day trial may convert on a 14-day version. More runway to feel value before the billing decision.
- **Introductory price**: A discounted first period reduces perceived risk for price-sensitive users without permanently lowering the full-price offering.
The core principle: present a different decision, not just a lower price for the same decision. A user who dismissed a **$49.99/year** offer is not necessarily waiting for **$39.99/year**. They may need monthly billing, a longer trial, or a one-time option.
\> **Case study: Superwall Transaction Abandon Paywalls** ([August 2024](https://superwall.com/blog/17-revenue-boost-with-transaction-abandon-paywalls-a-case-study/))
> 18 apps, ~500,000 users. Users who started a payment transaction but did not complete it were shown a discounted exit offer at the moment of abandonment.
>
> - Exit offer users drove **17% of total revenue** across the cohort.
>
> - Their refund rate was **3.3%**, compared to **6.8%** for users not shown an exit offer.
>
>
> The lower refund rate rules out accidental conversion. These users converted because the exit offer matched what they were actually willing to pay.
### 2\. Layer 2: Send a Recovery Push Notification Within 24 Hours
When a user exits the app after dismissing the paywall, a recovery push is the next opportunity. Timing is the primary variable.
Send the first push within the first hour of exit. That context — the app recently open, the paywall seen, the decision not yet finalized — decays quickly.
What works in recovery pushes:
- Connect to the user's stated goal, not the subscription mechanics. "Your goal: \[X\]. Here's what's still waiting."
- For trial-first apps: "Your free trial is reserved until \[time\]" creates a specific actionable deadline without being manipulative.
- **Send a second push the next day** if the first goes unanswered. Users who were simply distracted during the first push often convert on the follow-up. The next-day message captures a meaningful segment that the same-day push misses.
For a deeper look at structuring these sequences, see [push notification strategy for subscription apps](https://www.airbridge.io/en/blog/push-notification-strategy-for-subscription-apps).
### 3\. Layer 3: Run a Targeted Re-Trial Between Day 7 and Day 30
Users who continue using the app without converting, returning weekly, completing workflows, engaging with features, represent a specific segment: they have found value but have not crossed the payment threshold.
The mid-lifecycle re-trial is a targeted second trial offer, triggered by behavior rather than a fixed calendar:
- "You've completed **20 sessions**. Try the full version free for **7 days**."
- Triggered after a user engages with a premium-locked feature for the third time.
- Offered to users who started a trial, let it expire, and continued using the free version.
For users who have already spent time in the app, a second trial offer operates differently than the first. [RevenueCat has documented](https://www.revenuecat.com/blog/growth/how-to-increase-trial-conversion-rates-without-adding-new-features/) freemium-to-trial conversion lifting from **0.4% to 4.5%** when trial offers are reintroduced to established free users — a ratio that reflects habit-grounded consideration rather than first-time evaluation.
**Targeting matters.** Re-trials offered to highly engaged users who are close to converting organically can cannibalize full-price revenue. Target medium-engagement users: active, but not power users.
### 4\. Layer 4: Build an Alternative Plan Architecture
Some users will not say yes to your main offering, but they would say yes to something else. If your [paywall flow](https://www.airbridge.io/en/blog/set-up-a-simple-paywall-flow) only offers one plan at one price, you are making a single bet on every user.
| Alternative | What It Solves | Best For |
| --- | --- | --- |
| Monthly after annual lead | Commitment hesitation | Annual-first paywalls |
| Lifetime option | Recurring payment resistance | Users who reject subscriptions on principle |
| Lite tier | Price sensitivity | High-engagement users with low willingness to pay |
| Consumable add-on | "I don't need everything" | Feature-heavy apps where users want one capability |
Plan-count experiments are among the highest-performing paywall tests available. [Adapty's 2026 report](https://adapty.io/blog/high-performing-paywall-2026/) shows a **57.1% LTV win rate** for apps that run plan-count experiments, with three-option paywalls cited as optimal. Weekly plans convert **1.7x-7.4x** better than annual plans at equivalent price points.
\> **Case study: Party Game App: Paywall Simplification** ([RevenueCat / Paywalls.Design, March 2025](https://www.revenuecat.com/blog/growth/paywall-redesigns-case-studies/))
> The app's primary paywall offered three plans: annual, monthly, and weekly. The redesign removed monthly from the main screen, kept it under "View All Plans," and reduced the primary choice to two options.
>
> - Install-to-trial conversions increased by **31%**.
>
> - Revenue lifted by **64%**.
>
>
> Fewer choices, clearly framed, outperformed the original layout across the same user base.
## How to Design a Post-Paywall Second Offer That Converts
### 1\. Match the Exit Offer to the User's Objection, Not to Your Margin
Three users dismiss a paywall for three different reasons: price sensitivity, insufficient value conviction, and timing ("not right now"). A blanket discount treats all three identically and is the right answer only for the price-sensitive group.
A single-question micro-survey inserted between the dismiss tap and the exit routes each user to the appropriate response:
> **"What stopped you from subscribing today?"**
>
> - Price is too high
>
> - Not sure it's right for me yet
>
> - I'll come back later
>
> - Something else
>
Price concern routes to a monthly or discounted exit offer. Uncertainty routes to an extended trial. "Later" routes to a timed push rather than an immediate exit offer. This post-paywall recovery strategy for exit offers converts better than any single blanket approach because it matches the response to the actual hesitation.
### 2\. Use a Lifetime Plan as an Exit Offer
A lifetime subscription as an exit offer exploits a specific dynamic: some users do not resist paying, they resist committing to a recurring charge. A one-time lifetime option reframes the entire decision.
The user is no longer evaluating "should I subscribe?" They are evaluating "is this worth $X one time?" For a specific segment, that framing converts where a subscription offer does not.
**Present the lifetime option as an exit offer, not the primary paywall offering.** Placing it front-and-center trains users to skip the primary offer and wait for the lifetime option. This approach aligns with the broader [paywall conversion structural decisions](https://www.airbridge.io/en/blog/paywall-conversion-structural-decisions) that shape how users process pricing.
### 3\. iOS vs. Android: The Exit Offer Compliance Risk You Need to Know
Exit offers on Android have no known compliance issues. Test freely.
iOS is different. **Apple has rejected apps for showing exit offers under Guideline 5.6 (Developer Code of Conduct) and, in 2026, increasingly under Guideline 3.1.2 (Subscriptions)**, citing "manipulative practices" around in-app purchases. The guidelines do not explicitly forbid exit offers, but enforcement has been inconsistent and has shifted in 2026 toward stricter subscription-specific review, as documented by [Adapty](https://adapty.io/blog/your-toggle-paywall-is-about-to-get-rejected/) and the [RevenueCat Engineering Blog](https://www.revenuecat.com/blog/engineering/exit-offers-in-revenuecat-paywalls/).
The practical path: test exit offers on Android first. Measure the lift in conversion and revenue. If the result is meaningful, assess iOS implementation with accurate performance data in hand.
> One hard rule: do not remotely enable or disable exit offers after App Review approval on iOS. This violates the Apple Developer Program License Agreement directly and risks termination of the developer account.
## How to Measure Post-Paywall Recovery Rate
### 1\. The Three Metrics That Matter
**Recovery rate** is the percentage of users who dismissed the primary paywall and then converted within a defined window (**7 days**, **30 days**). Segment this by acquisition channel. Users from paid campaigns often have meaningfully different recovery rates than organic users. Understanding which [lifecycle stage](https://www.airbridge.io/en/blog/lifecycle-marketing-for-app-timing) each user cohort sits in sharpens this analysis.
**Exit offer conversion rate** measures, among users who saw the exit offer, how many converted. Track this separately from the primary paywall rate to preserve a clean signal on both.
**Push timing performance by bucket** breaks recovery push analysis into windows: **0-60 min**, **1-6 hrs**, **6-24 hrs**, **24-48 hrs**. Recovery push data from experimenting teams consistently shows the earliest bucket outperforming later windows. The gap between a push sent within the first hour and one sent the next day is not subtle.
### 2\. How to Detect Exit Offer Cannibalization
Exit offers can reduce primary paywall conversions if users learn that dismissing always produces a better deal. Track this by comparing full-price conversion rates across cohorts that received exit offers versus those that did not.
If exit offer availability is pulling down primary conversions: introduce a short delay before the exit offer appears, or limit exit offers to users who spent a minimum time on the paywall, indicating genuine consideration. The goal is to recover hesitant users, not to teach all users that the first offer is a decoy.
[Adapty's 2026 report](https://adapty.io/state-of-in-app-subscriptions/) found that apps running structured experiments on monetization earn up to **40x** more revenue than those that do not test. The post-paywall recovery layer is one of the highest-leverage experiments available because the traffic is already engaged, it just needs a better response.
For a primer on the [onboarding steps](https://www.airbridge.io/en/blog/5-steps-app-onboarding-before-the-paywall) that set up the paywall moment correctly in the first place, that context improves how the recovery layer performs downstream.
## Build the Post-Paywall Recovery Layer Before You Optimize the Primary Paywall
Most teams A/B test paywall headlines weekly and never measure recovery rate at all.
The users who dismissed the paywall are not gone. They are users whose specific hesitation has not been addressed yet. The apps recovering meaningful revenue from this group have built the layered sequence above.
Build the recovery layer first. Then optimize the primary paywall. Most teams have that order reversed, and the sequence matters.
---
## How to Spot Users Who Are About to Churn Before They Leave
URL: https://www.airbridge.io/en/blog/how-to-spot-users-who-are-about-to-churn
Language: English
Published: 2026-05-25
Author: Harper (Trang Nguyen)
App churn doesn't announce itself. By the time a subscriber cancels, the decision was made weeks earlier. Their sessions grew shorter, core features went untouched, and eventually they stopped opening the app.
The data to detect this is already in your product. **"Not Enough Usage" is the single leading cause of subscription cancellations, accounting for** [**37.02% of all churn**](https://www.revenuecat.com/blog/growth/subscription-app-churn-reasons-how-to-fix/)**.** That decline in usage happens before the cancellation event. The early churn signals subscription app teams overlook are already visible in your analytics. The challenge is knowing which ones to track and when they become actionable.
**Key Takeaways**
- **Churn is a staged process, not a sudden event.** Users move from Active to Drifting to At-Risk before canceling. Each stage is visible in behavioral data if you track the right metrics.
- **The first renewal window is your highest-risk moment.** Nearly 30% of annual subscribers cancel within the first month of their subscription.
- **Session frequency decline is the earliest reliable signal.** A drop in weekly app opens typically precedes cancellation by 10 to 28 days for most subscription apps.
- **Segment disengaged users before messaging them.** Drifting users (still paying, declining usage) need different messaging than at-risk users who haven't opened the app in 30-plus days.
- **Response timing determines recovery rate.** The window to recover a disengaged subscriber narrows with each passing day. Behavioral triggers fire when intent is still present.
## Why Churn Starts Long Before Cancellation
Most teams measure churn at the point of cancellation. That captures the outcome, not the cause.
### 1\. The drift-to-churn progression
Every user who eventually churns passes through three observable stages before canceling:
1. **Drifting.** Still subscribed, still paying. Session frequency is declining but not zero. Core feature engagement is decreasing. This is the most recoverable stage, and the one most teams miss entirely.
2. **At-Risk.** Session frequency has dropped significantly. Core feature engagement is near zero. The user is one billing cycle away from making a deliberate cancellation decision.
3. **Churned.** Subscription canceled. Recovery now requires a dedicated win-back campaign and, in most cases, a direct incentive.
**Drifting users are recoverable with relevance and timing.** Waiting until the At-Risk stage requires a heavier intervention. Waiting until Churned makes recovery costly and statistically unlikely.
### 2\. Why the first renewal window is the highest-risk moment
[**Nearly 30% of annual subscriptions are canceled in the first month.**](https://www.revenuecat.com/state-of-subscription-apps/) For monthly subscribers, [the median 12-month retention rate is **17.0%**](https://www.revenuecat.com/blog/growth/one-year-retention-rates-insights/). More than 4 in 5 monthly subscribers have churned within a year of their first payment.
The first renewal is when users make a conscious re-evaluation: is this app still worth continuing? If they haven't experienced your core value by that point, the decision is already formed. Detecting disengagement before that evaluation window is where interventions carry the most leverage.
> **Further reading**: [Why Subscription Churn Happens: Top 5 Cancellation Reasons Forecast for Mobile Apps in 2026](https://www.airbridge.io/en/blog/why-subscription-churn-happens-top-5-cancellation-reasons-forecast-for-mobile-apps-in-2026) — a deeper look at the underlying reasons subscribers cancel, with forecasts for 2026 across categories and business models.
[diagram-churn-lifecycle.svg]
## 5 Early Warning Signals of App Churn
Pre-churn signals fall into two categories: **behavioral** (what users do in-app) and **transactional** (how they interact with billing). Both appear before cancellation. Both are trackable today.
### 1\. Session frequency decline
The most reliable early indicator of impending app churn. Many subscription teams observe session counts starting to decline 10 to 28 days before subscribers cancel. That window is large enough to intervene.
Set a session frequency baseline for each user cohort during their first 30 days. When a user's weekly session count drops below **50%** of that baseline, flag them for re-engagement. **Track opens per week, not session duration.** A user can open your app briefly and still be drifting. Frequency is a more forward-looking predictor than time-on-app.
### 2\. Core feature disengagement
Not all features are equal predictors of retention. Identify the one or two features that your highest [lifetime value (LTV)](https://www.airbridge.io/en/glossary/lifetime-value-ltv) subscribers engage with consistently. When a user stops using those features, that is a stronger churn signal than general inactivity.
Retained users have built a habit around a specific workflow in your app. A break in that habit pattern predicts drift more precisely than aggregate engagement scores alone. This includes not completing a workout, not logging a session, or not reviewing a document. This behavior is your activation metric: the single action that most reliably separates churners from retainers.
### 3\. Billing friction and payment failures
**Not all app churn is voluntary.** Payment failures from expired cards or declined transactions drive a meaningful share of involuntary cancellations. [**Apple's iOS billing grace period is opt-in**](https://developer.apple.com/help/app-store-connect/manage-subscriptions/enable-billing-grace-period-for-auto-renewable-subscriptions/) and must be activated in App Store Connect. [Google Play has grace period enabled by default](https://developer.android.com/google/play/billing/lifecycle/subscriptions#grace-period), but you should verify the length (3 days by default) and ensure account hold is also configured. If iOS grace period is not active in your app, you are losing paying subscribers to billing failures who would otherwise have renewed.
Monitor billing error events in real time. A proactive notification sent within hours of a payment failure recovers materially more users than one sent **48 hours** after the event.
### 4\. Negative sentiment signals
Users who are about to churn often express dissatisfaction before they act on it. An uptick in 1 to 2 star app store reviews is a leading indicator of subscriber dissatisfaction. These reviews typically surface before users reach the cancellation screen. Monitor review sentiment on a rolling basis, not just at launch or after major updates.
If you run an in-app NPS survey, segment detractors (scores 0 to 6) immediately. These users represent your highest churn-risk cohort. A personalized outreach within 24 hours of a detractor score gives you a recovery window before they reach the cancellation screen. For a broader view of how [lifecycle marketing sequences](https://www.airbridge.io/en/blog/subscription-app-lifecycle-marketing) map to each engagement stage, that context helps you design the right follow-up for each signal type.
### 5\. Plan downgrade intent
A user who navigates to their subscription settings, views downgrade options, and exits without completing the action is signaling cost sensitivity or value uncertainty. **Track these downgrade intent events in your analytics tool.** They require a different response than passive drift. The underlying issue is usually perceived value, not product failure, and a well-timed feature highlight, plan pause option, or concierge outreach can address it before the user takes action.
| Early Warning Signal | What to Monitor | Typical Window Before Cancellation |
| --- | --- | --- |
| Session frequency drop | Opens per week vs. cohort baseline | 14 to 21 days |
| Core feature abandonment | Activation metric engagement | 10 to 21 days |
| Payment failures | Billing error events | Same day |
| Negative sentiment | App store ratings, NPS detractor scores | Before action is taken |
| Downgrade intent | Plan change intent events | Immediate |
## How to Act on Pre-Churn Signals
Detecting a signal without a response system in place is an incomplete loop.
### 1\. Segment before messaging
**Send different messages to different disengagement stages.** A user who hasn't opened your app in 10 days is in a different state than one who hasn't opened it in 45 days.
Drifting users respond well to personalized re-engagement: where they left off, a feature they haven't tried, a milestone they are close to. At-risk users often need a more direct offer: a pause option, a plan adjustment, or a short-term discount targeted to this segment. Sending a discount to a drifting user who would re-engage without one trains future users to expect discounts before reactivating.
### 2\. Time your response precisely
**The highest-leverage moment is immediately after a disengagement signal fires, not several days later.** A push notification sent within minutes of a cancellation-intent behavior converts at meaningfully higher rates than one sent 48 hours after the event.
Tie your re-engagement cadence to natural product use cases. A wellness app should nudge during morning routines. A productivity app should nudge at the start of the work week. Mistimed notifications train users to ignore them. For teams running paid UA, understanding [how attribution windows](https://www.airbridge.io/en/glossary/attribution-window) affect cohort timing helps you read churn patterns by acquisition source more accurately.
## Stop Guessing. Know Which Campaigns Drive Your Churn.
App churn is measurable when you track the right signals. Session frequency, core feature engagement, billing errors, NPS scores, and downgrade intent all appear in your product analytics before a subscriber cancels. The gap most teams face is not in the data itself — it is in connecting those behavioral signals back to the acquisition sources that created those users in the first place.
Understanding which paid UA channels produce high-churn cohorts, not just high-install numbers, is where the picture becomes complete. [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your Meta, Google, Apple Search Ads, and TikTok campaigns to subscription lifecycle events including Unsubscribe, so you can see which acquisition sources drive subscribers who renew versus those who cancel within 30 days. **Start free with 15K attributed installs.**
---
## How to Read App Ad Campaign Data and Decide When to Scale
URL: https://www.airbridge.io/en/blog/read-app-ad-campaign-data-to-scale
Language: English
Published: 2026-05-22
Author: Luke
You are running two ad campaigns on Meta. Campaign A drove installs at **$4** cost per install. Campaign B drove installs at **$8**. The obvious read is to scale Campaign A and cut Campaign B.
Six weeks later, Campaign A's subscribers have churned at **60%**. Campaign B's subscribers are still paying. **The cheaper install turned out to be the worse campaign.** This is the trap that cost-per-install decisions create, and most subscription app marketers fall into it at least once.
Reading campaign volume is not the same as reading campaign performance. The data you need to decide whether to scale or pull an app ad campaign does not live inside Meta Ads Manager or Google Ads. It lives in the downstream subscription funnel, weeks after the install.
**Key Takeaways**
- **CPI and CTR measure reach, not revenue.** The numbers that drive scale decisions are trial-to-paid conversion, revenue per install, cost per subscription, and first renewal rate.
- **There are three campaign states, not two.** Scale, Optimize, and Pull each have specific criteria. "Not scaling" is not the same as "pull."
- **Cost per Subscription (CPS) collapses the whole funnel into one number.** If CPS exceeds your payback window, no amount of CPI optimization changes the outcome.
- **Adding budget to a broken funnel compounds the loss.** Every dollar added to a campaign with weak downstream conversion increases the total loss, not the total gain.
- **Every ad platform claims its own version of your conversions.** Without a single source of truth for subscription data, scale decisions are based on conflicting numbers.
## What Meta and Google Ads Don't Tell You About Subscription Performance
**App campaign performance analysis** is the practice of evaluating subscription funnel metrics (trial conversion, revenue per install, and renewal rate) to determine whether a campaign should be scaled, optimized, or paused. It starts downstream from the ad platform, not inside it.
### 1\. Why CPI Is a Cost Metric, Not a Subscription Revenue Metric
The cost per paying user (CPPU) runs **4 to 5 times the CPI** across subscription app categories. A campaign showing a $4 CPI likely carries a $16 to $20 CPPU. That number does not appear on any ad platform dashboard ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)).
CPI measures one thing: how much you paid to get someone to tap "download." It says nothing about whether that person started a trial, converted to a paid plan, or renewed. The gap between install cost and subscriber cost is where most scaling decisions go wrong.
### 2\. Actuals Tell You Cash. Cohorts Tell You Profit.
Last month's actuals show what happened in a time window: installs, spend, and revenue. Cohorts show what a specific group of users generates over time: users who installed in May, tracked through June, July, and August.
Actuals tell you your cash position. Cohorts tell you your unit economics. Scaling based on monthly actuals alone is deciding to increase a bet before all the cards are on the table.
### 3\. Why Your Funnel Stage Determines Which Metrics to Read
Most campaigns are failing at one stage while appearing healthy at another. A campaign with strong install volume and a low trial start rate has an onboarding or paywall problem. A campaign with strong trial volume and a low trial-to-paid conversion has a product-trial alignment problem.
Read the funnel in order: install rate, trial start rate, trial-to-paid conversion, and first renewal rate. Strength at the top often hides a break further down.
## The 4 Subscription Metrics That Determine Whether to Scale an App Ad Campaign
These four metrics live downstream from the ad platform. They do not appear on your Meta or Google dashboard. Each one gates the next: a weak trial-to-paid rate makes RPI projections unreliable, and an unexamined renewal rate makes CPS comparisons wrong.
### 1\. Trial-to-Paid Conversion Rate: The First Gate Before Scaling Ad Spend
If your trial-to-paid conversion rate is below **34.8%**, the median across 75,000+ subscription apps, scaling ad spend pushes more users into a funnel that does not convert. Every dollar added amplifies the leak, not the revenue.
The top quartile of apps converts at **51.5% and above**. If you are below the median, the issue is the onboarding sequence or paywall design, not the acquisition campaign. Fix the funnel before scaling the top.
| Benchmark | Trial-to-Paid Conversion |
| --- | --- |
| Median (all categories) | 34.8% |
| Top quartile (Q3+) | 51.5%+ |
| Health and Fitness (median) | 39.9% |
| Travel (median) | 48.7% |
_Source: RevenueCat State of Subscription Apps 2025_
### 2\. D30/D60 Revenue per Install (RPI): Is This Campaign Paying Back?
Revenue per install measures cohort-level revenue divided by install volume over a defined time window, including every trial, subscription, and renewal that follows.
Median D60 RPI across subscription apps is **$0.31**. The top quartile reaches **$0.81** or above. If a campaign is generating D60 RPI below $0.31, it is underperforming the market. If it is above $0.81 and trending upward, it is a scaling candidate.
Health and Fitness apps see a median D60 RPI of **$0.63**. Business apps reach **$0.48**. If your category typically runs above $0.31 and your campaign is still near the floor, the issue is campaign quality, not category difficulty.
The D60 RPI gap between monetization models is significant: apps with hard paywalls reach a D60 RPI of **$3.09** versus **$0.38** for freemium models, an 8x difference that reflects directly in how quickly a campaign pays back ([RevenueCat State of Subscription Apps 2026](https://www.revenuecat.com/state-of-subscription-apps/)).
_Source: RevenueCat State of Subscription Apps 2025 (category medians); RevenueCat State of Subscription Apps 2026 (paywall model comparison)_
### 3\. Cost per Subscription (CPS): The Metric That Collapses the Funnel
CPS is total ad spend divided by total subscriptions generated. It answers the critical question in budget decisions: what does one paying subscriber actually cost me?
The check is direct. If your D60 RPI is **$0.81** and your CPS is **$80**, the campaign does not pay back within 60 days. If your Year 1 [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) per payer is near the median (**$16.27** per RevenueCat SOSA 2025), a CPS of $80 never pays back. Compare CPS against your own payback window. The only relevant benchmark is your unit economics.
### 4\. First Renewal Rate: The LTV Signal That Separates Campaigns Worth Scaling
The first renewal is the most important event after conversion. Across subscription plan types, first renewal rates run at **51.8%** for weekly, **55.5%** for monthly, and **66.3%** for yearly plans. Nearly **30% of annual subscribers cancel in the first month**.
Two campaigns with identical trial-to-paid conversion can have completely different LTV if their first renewal rates differ. Imagine Meta-sourced subscribers renew at **40%** while Google-sourced subscribers renew at **65%**: the Google campaign has higher LTV despite the higher CPI.
Check renewal rates by acquisition channel before projecting LTV. A first renewal rate below the platform baseline for your plan type is a pull signal, not a scale signal.
_Source: RevenueCat State of Subscription Apps 2025_
## Where Each Funnel Stage Tells You to Spend Differently
[Screenshot 2026-05-22 at 19.13.20.png]
_Subscription funnel diagram showing install, trial start, paid conversion, and first renewal stages with scale, optimize, and pull signal labels at each stage_
Before running the three-condition scale check, map where your campaign stands in the funnel. Each stage reveals a different problem and points to a different fix.
### 1\. Top of Funnel: Install Volume and Trial Start Rate
Top-of-funnel data tells you whether targeting and creative are resonating. The benchmark trial start rate is **6.2%** median, with a top 10% of **20.3%**.
If your trial start rate is below 6.2%, there is likely a problem with the onboarding flow or paywall placement. Fix that before scaling spend.
Do not confuse trial volume with trial quality. Lifestyle apps see only **3.9%** trial start rates but achieve **41%** trial-to-paid conversion. Gaming apps see **12.2%** trial starts with only **19%** conversion ([AppsFlyer State of Subscriptions for Marketers 2026](https://www.appsflyer.com/resources/reports/subscription-marketing/)).
The category with 3x lower trial volume is the better scaling candidate when measured at the funnel level. Top-of-funnel data tells you about reach, not money.
### 2\. Mid-Funnel: Trial Quality and Post-Trial Engagement
Mid-funnel is where the first revenue signal appears. Track trial-to-paid conversion rate, trial cancellation rate, and early in-app engagement.
Users who complete a meaningful first-week action — finishing their first workout in a fitness app, completing the first lesson in an education app — convert and renew at significantly higher rates than users who install, start a trial, and go quiet. If post-trial engagement is flat across a campaign's cohort, the acquisition targeting is not the problem.
### 3\. Bottom Funnel: Cost per Subscription and First Renewal Rate Together
A campaign that consistently generates new paying subscribers at a CPS below your payback threshold, with a first renewal rate at or above baseline, has passed all three gates. That is a scaling candidate.
All three conditions must be true. Two out of three is an optimization signal, not a scale signal.
## The Scale, Optimize, and Pull Decision Framework for App Ad Campaigns
[Screenshot 2026-05-22 at 19.14.36.png]
### 1\. When to Scale: Three Conditions That Must All Be True
Scale a campaign when:
1. Trial-to-paid conversion is at or above the median for your category (34.8% general baseline; higher for Health and Fitness)
2. D60 RPI is trending above $0.31 and approaching or above $0.81
3. First renewal rate is at or above baseline for your plan type
When all three are green, increase budget in incremental steps. Meta's standard for exiting the learning phase is 50 optimization events within 7 days; scaling before that threshold is met tends to reset the algorithm and degrade performance ([Meta Business Help Center](https://www.facebook.com/business/help/1579799238701082)).
### 2\. When to Optimize: Reading Mixed Performance Signals
When some conditions are green and one or more are not:
- Low trial-to-paid with normal CPI: the onboarding sequence or paywall needs fixing before adding budget
- Low D60 RPI with normal trial conversion: check plan pricing and trial length, not the campaign targeting
- High CPS with low renewal rate: the campaign is attracting users who subscribe once and leave
Mixed signals mean the problem is downstream of the campaign. Increasing budget on a funnel problem makes the funnel problem larger.
### 3\. When to Pull: Three Stop-Loss Conditions for App Ad Campaigns
Pull the campaign when:
1. CPS exceeds 2x your payback threshold after reaching statistical significance (at least 50 paid subscription events per campaign variant before drawing conclusions)
2. Trial-to-paid conversion drops below 20% and post-trial engagement is also flat
3. First renewal rate falls below 40% for a monthly plan, roughly 75% of the 55.5% monthly median, the threshold at which Year 1 LTV typically cannot cover a sustainable CPS
These are not optimization problems. They are structural signals that the campaign is reaching the wrong audience or driving users into a broken product funnel.
## Why Meta, Google, and Apple All Claim the Same Conversion
**The four metrics above are not visible inside Meta Ads Manager, Google Ads, or Apple Search Ads.** Those dashboards show installs and sometimes trials if you send back the event. They do not show which campaigns generated subscribers who actually renewed.
Every ad platform reports its own version of performance, and each claims credit for the same conversions. A user who sees a Meta ad and clicks a Google ad a week later before subscribing gets counted as a conversion in both systems. This double-counting makes CPS and RPI calculations unreliable when built from platform data alone.
## Run the Three-Condition Check Before Moving Any Campaign Budget
The problem is not that you lack data. The problem is reading CPI when the decision requires CPS, or reading monthly actuals when the decision requires cohort-level renewal rates.
Run the three-condition check before moving any budget. If all three are green, scale incrementally. If one is yellow, find the funnel break first. If two are red, stop the spend and diagnose before adding a single dollar more.
---
## What Makes a Paywall Convert? Six Structural Decisions for Subscription Apps
URL: https://www.airbridge.io/en/blog/paywall-conversion-structural-decisions
Language: English
Published: 2026-05-22
Author: Luke
Most paywalls look reasonable. Clean layout, benefit bullets, a trial offer. Yet the median subscription app converts just **1.9% of downloads to paying subscribers within 35 days**. The top 10% convert **8.5%** ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). That 4.5x gap is not a design problem.
The difference lives in six structural decisions made before anyone opens a design tool: model, placement timing, plan structure, trial length, price level, and experiment discipline. Get these right and design becomes a finishing touch. Leave them to chance and no amount of copy iteration will close the gap.
[Screenshot 2026-05-22 at 19.00.06.png]
_Each stage of the funnel drops significantly._
_Top-performing apps widen the gap at the Trial Start and Trial-to-Paid stages through deliberate placement, model, and trial strategy decisions._
**Key Takeaways**
- **Paywall timing outweighs paywall design.** 82% of trials start on Day 0 (RevenueCat SOSA 2025). What happens in onboarding determines conversion more than anything on the screen itself.
- **Hard paywalls convert 5.5x better than freemium** at Day 35 (12.1% vs. 2.2% median), and they generate roughly double the year-one LTV.
- **Conversion = Motivation x Trust / Friction.** Every element either builds motivation, adds trust, or removes friction. Low conversion means one of these three is broken.
- **Longer trials outperform shorter ones.** Apps with 17 to 32-day trials convert at **45.7% trial-to-paid**, compared to 26.8% for 3 to 7-day trials.
- **Teams that run experiments generate up to 40x more revenue** than teams that do not. Optimization is a process, not a one-time project.
## Why Do Most Paywalls Underperform? Six Decisions That Set Your Ceiling
### The 4.5x Gap Between Median and Top Apps Is Structural, Not Visual
The subscription app market concentrates performance at the top. The top 10% of apps capture **95% of total revenue**. Paywall best practices for subscription apps consistently point to six structural variables as the primary drivers of conversion, not design choices.
The leverage in structural decisions is extreme. Among weekly-plan apps, trial format screens win **64.5%** of head-to-head tests against visual-only layouts, and weekly + trial configurations consistently rank highest for year-one LTV ([Adapty, State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions-report/)). That gap comes from one configuration decision, not a redesign.
### Six Structural Decisions Determine Your Conversion Ceiling
| Decision | What It Controls |
| --- | --- |
| Paywall model | Hard gate vs. soft gate vs. freemium |
| Placement timing | When in the user journey the offer appears |
| Plan structure | Which plans are shown and how they are presented |
| Trial strategy | Whether to offer a trial and for how long |
| Price level | Actual price points and anchoring logic |
| Experiment roadmap | What gets tested and how results are measured |
Design and copy come after these six are deliberate. If the structure is misconfigured, visual iteration produces marginal gains at best.
## When Should Your Paywall Appear? Placement Is the First Conversion Lever
### Day 0 Is the Only Paywall Moment Most Users Give You
**82% of trials begin on the same day a user downloads the app** (RevenueCat SOSA 2025). If the offer is buried behind multiple onboarding screens, motivation has already started to drop by the time users reach it.
Track **paywall visibility rate** (the percentage of new users who ever reach the offer) as a primary metric before optimizing anything on the screen. If fewer than 80% see it, fix that first.
### Onboarding Placements Outperform In-App Gates Across Every Configuration
Adapty's dataset of 16,000+ apps ranks install-to-paid conversion by placement type:
| Placement | Install-to-Paid |
| --- | --- |
| Onboarding + trial | 1.78% |
| In-app + trial | 0.89% |
| Onboarding, no trial | 0.82% |
| In-app, no trial | 0.76% |
Onboarding placements more than double the conversion of in-app gates with no trial. The user has just chosen to install your app and has not yet found reasons to disengage. Fix the onboarding flow first, then move the offer to the screen immediately after the user's first meaningful action.
## What Does a High-Converting Paywall Screen Include?
### Lead with Outcomes, Not Price: Information Order Drives Conversions
Most screens lead with pricing. This anchors users on cost before they have any reason to assign value. High-converting screens follow a different sequence: **outcome, value explanation, reassurance, pricing, CTA**.
This order answers user questions in the sequence they actually ask them: what do I get, how does that help me, is this trustworthy, what does it cost, and what do I do now. Reversing it leaves objections unanswered at the wrong moment.
### Social Proof and Testimonials Can Double Trial Start Rates
Social proof answers the "is this trustworthy?" question before users commit to payment. Effective trust elements include star ratings with a specific review count, outcome-specific testimonials matched to the user's onboarding goal, and recognizable download numbers or media mentions.
One well-documented test showed that **a long-scroll paywall with testimonials, feature explanations, and an FAQ section doubled trial start rates** compared to a minimal single-screen layout (48 Laws of Subscription App Success, Law 16).
### Default Selection and Price Anchoring Shift Plan Choice Without Changing Prices
Two behavioral mechanisms improve revenue without touching price.
**Default bias:** Users tend to accept the pre-selected plan. Pre-selecting your highest-value plan shifts the distribution of choices in your favor. Most apps pre-select the cheapest option by default, which is the opposite of the optimal configuration.
**Anchoring via decoy:** A mid-tier plan placed between the cheapest and best-value options makes the best-value plan appear to be the obvious choice.
## Hard Paywall or Soft Paywall: Which Model Converts Better?
### Hard Paywalls Convert 5.5x Better Than Freemium at Day 35
The performance difference between models is larger than most teams expect (RevenueCat, State of Subscription Apps 2025):
| Model | Median D35 Conversion | Median Y1 LTV |
| --- | --- | --- |
| Hard paywall | 12.1% | $49.30 |
| Freemium | 2.2% | $24.24 |
**Hard paywalls filter for intent.** Users who subscribe without a trial are more committed at the moment of purchase. They churn less, renew at higher rates, and generate more [lifetime value](https://www.airbridge.io/en/glossary/lifetime-value-ltv). The 5.5x conversion advantage compounds into roughly 2x the LTV advantage at year one.
Hard paywalls only work if onboarding earns the ask. Users who hit the gate before understanding the product's value will not convert. Hard paywall Y1 LTV is $49.30, but refund rates are also higher (**5.8%** vs. **3.4%** for freemium).
### Freemium Works When Value Requires Time to Experience
Freemium is not the wrong model. It is an expensive model. Converting **2.2%** of installs instead of 12.1% means you need roughly 5x more installs to generate the same subscriber count.
Freemium makes sense when the value proposition requires extended time to experience, or when growth relies on word-of-mouth from free users. For apps running [paid UA](https://www.airbridge.io/en/blog/subscription-app-free-trial-conversion-benchmarks), the data favors hard paywalls or onboarding-gate trials.
## How Long Should a Free Trial Be? Trial Strategy That Converts
### Longer Trials Win on Habit Formation, Not Time Pressure
The finding from 75,000+ apps is counterintuitive: longer trials convert better. Apps with 17 to 32-day trials achieve a **45.7% trial-to-paid conversion rate**. Apps with 3 to 7-day trials convert at **26.8%**. The longer trial outperforms by nearly 19 percentage points.
The mechanism is habit formation, not generosity. More than **55%** of 3-day trial users cancel on Day 0, versus **31%** for 7-day trials ([RevenueCat](https://www.revenuecat.com/blog/growth/7-day-trial-subscription-app/)). A 3-day trial creates time pressure with no time to form a habit. A 17-day trial gives users enough sessions to build a routine. By the time it ends, they are deciding whether to stop using something already integrated into their workflow. If your trial conversion is below the industry median of **34.8%** and you are running 7-day or shorter trials, test a 14 or 30-day trial before redesigning the screen.
### In High-Intent Categories, Direct Purchase Users Generate More LTV
Trials are not universally positive. The category-level premium from a trial vs. direct purchase varies widely ([Adapty, 2026](https://adapty.io/blog/free-trial-vs-direct-purchase-subscription-apps/)):
| Category | Trial Uplift vs. Direct Purchase |
| --- | --- |
| Utilities | +85.1% |
| Health & Fitness | +63.6% |
| Education | +50.4% |
| Productivity | -13.7% (direct purchase wins) |
For Productivity apps, direct buyer LTV is **$56.95** versus trial user LTV of **$49.13**. A trial introduces hesitation for a user who was already ready to commit. Test a direct purchase option alongside your current offer.
### 90% of Users Who See the First Screen Do Not Convert. Show a Second Offer.
More than **90% of users** who see the first screen do not convert on that view (48 Laws of Subscription App Success, Law 17). Most apps present nothing to that group. High-performing apps show a second offer.
The second offer does not have to be a discount. Effective options include a different plan format (monthly vs. annual), a lifetime purchase option that removes recurring commitment friction, or a short micro-survey that surfaces the hesitation reason. For more on [subscription pricing plan structures](https://www.airbridge.io/en/blog/subscription-app-pricing-weekly-monthly-annual) that work alongside second offers, see our plan type guide.
## How to Build a Paywall Optimization Roadmap
### Segmentation Amplifies Broken Paywalls Instead of Fixing Them
The most common mistake is jumping to segmented offers before fixing the baseline. Segmentation amplifies what already exists. It does not create performance where none exists. Apps that experiment systematically generate up to **40x more revenue** than apps that do not, but the experiments have to be structured correctly.
A practical three-phase roadmap:
- **Phase 1 (Foundation, 0 to 30 days):** Fix visible friction. Confirm the offer loads reliably, the CTA is unambiguous, and the plan selection is logical. Track visibility rate and confirm 80%+ of new users see the offer.
- **Phase 2 (Optimization, 30 to 90 days):** Test structural decisions in order of leverage: trial length, model, plan structure and default selection, price points. One test at a time, measured by ARPU.
- **Phase 3 (Expansion, 90+ days):** Add segmentation and personalization based on onboarding behavior, engagement level, or acquisition source. This is also where connecting paywall performance to acquisition channel data becomes critical. Knowing whether your highest-converting users came from Meta or Apple Search Ads changes how you scale budget.
### Price and Plan Experiments Are the Highest-Leverage Tests. Measure with ARPU.
Adapty's dataset of 16,000+ apps shows the **probability** a given experiment type produces a winning result:
| Experiment Type | LTV Win Rate |
| --- | --- |
| Localization | 62.3% |
| Trial structure | 59.6% |
| Plan duration | 58.7% |
| Number of plans | 57.1% |
| Price changes | 45.5% |
| Visual / copy changes | 34.6% |
Structure and trial experiments win more than half the time. Visual and copy changes are the only lever with a sub-50% win rate, failing nearly **65%** of the time. Price experiments also improve conversion in **28%** of cases (conversion win rate). Start with plan structure and trial length. Move to price testing when you can measure LTV. Save visual and copy testing for last.
**ARPU (average revenue per user)** is the right primary metric. It balances conversion rate, pricing, and plan mix in a single number. Set ARPU as your success metric before running any test, and define ship / iterate / kill criteria before looking at results. See also: [what a good retention rate looks like for subscription apps](https://www.airbridge.io/en/blog/what-is-a-good-retention-rate-subscription-apps).
## Fix the Structure Before You Fix the Design
The gap between your current conversion rate and the top 10% is not a design gap. It is a structural gap. Placement, model selection, trial length, plan structure, pricing, and experiment discipline produce the 8.5% conversion rates. Button colors do not.
Pick the single structural decision you have never deliberately tested. Run one clean experiment this month. Measure ARPU. Then pair that experiment with channel-level attribution to see which acquisition sources are fueling your wins.
---
## How to A/B Test Your Paywall: Setup, Duration, and Results
URL: https://www.airbridge.io/en/blog/ab-test-paywall-setup-duration-results
Language: English
Published: 2026-05-22
Author: Luke
You already know your paywall needs work. Trial starts are lower than you'd like, the plan mix looks off, or the same design has been live for six months without a meaningful change. Knowing how to A/B test your paywall for subscription apps correctly is what separates teams that iterate toward revenue from teams that run tests and see nothing move.
Apps running **50 or more paywall experiments** generate a **median 18.7x revenue premium** over teams running just one ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)). That gap doesn't come from a single breakthrough test. It comes from running the right experiments in the right order, measuring what actually matters, and building on each result.
> **Paywall A/B testing** means running two or more versions of your purchase screen simultaneously, routing a random split of users to each variant, and measuring which version drives more subscription revenue per user.
**Key Takeaways**
- **Test pricing before visuals.** Plan and trial duration changes consistently drive more LTV improvement than copy or layout changes. Most teams get this sequence backwards.
- **ARPU is your north-star metric.** Conversion rate tells you how many people bought. ARPU tells you whether those purchases generate revenue. They do not always move together.
- **Set your sample size before you start.** You need a meaningful number of paid conversions per variant before results are reliable — the exact threshold depends on your baseline conversion rate and the lift you are trying to detect. Calculate this upfront, not after two weeks have passed.
- **Run most tests for a minimum of 2 weeks.** A weekend test does not capture a full weekly cycle of user behavior. Pricing tests often need 4-8 weeks to surface renewal and churn patterns.
- **Define your "done" criteria before looking at results.** Decide ship, iterate, or kill thresholds in advance. If you are using a fixed-horizon test, peeking at early data invalidates p-values. Bayesian tools like Adapty allow ongoing monitoring, but stopping decisions based on early results still introduce bias.
## Common Paywall A/B Testing Mistakes That Kill Results
Bad paywall tests don't fail because of bad luck. They fail because of decisions made before data collection starts.
### Mistake 1: Testing Visuals Before Pricing
The most common mistake is reaching for visual and copy changes first. A new headline, different button color, reshuffled plan layout: these feel manageable and safe to ship. But among subscription apps that run experiments, **plan and trial duration changes drive the most LTV improvement**, consistently outperforming visual and copy tests in both win rate and revenue impact.
The data is specific. When apps run experiments by category, here is how the LTV uplift win rates compare (Adapty State of In-App Subscriptions 2026):
| Experiment Type | LTV Uplift Win Rate |
| --- | --- |
| Locale / language changes | 62.3% |
| Trial duration changes | 59.6% |
| Plan count changes | 57.1% |
| Price point changes | 45.5% |
| Visual-only changes | 34.6% |
Visual-only tests have the lowest win rate of any category — lower even than price changes. When you optimize visual conversion before finding the right price, you're routing more users to a potentially underpriced offer. The price is the foundation. Visual changes are the polish. Polish applied to an unvalidated foundation doesn't hold.
Among apps that run pricing experiments, only **28%** improved conversion rate. But those that succeeded lifted LTV by an average of **~46%** (Adapty State of In-App Subscriptions 2026). Pricing is not the riskiest thing to test. It's the highest-upside lever available, and most teams skip it because it feels more disruptive to touch.
### Mistake 2: Stopping the Test Too Early
User behavior on a Saturday looks nothing like behavior on a Tuesday. Weekend users have different intent patterns, different free time, and different willingness to start trials. A test that runs for four days captures a skewed and unrepresentative sample.
A commonly cited minimum threshold for any paywall test is **2 weeks**, long enough to cover a full weekly behavior cycle at least twice and smooth out day-of-week variation ([Adapty, 2026](https://adapty.io/blog/paywall-ab-testing-mistakes/)). Running a test too short is one of the most expensive mistakes in paywall optimization. You ship a variant that "won" in a small sample, only to find months later that it underperforms the control on renewal metrics.
### Mistake 3: Tracking Conversion Rate Alone
Conversion rate answers one question: how many users purchased. It doesn't tell you whether those users renewed, what plan they chose, or what the average purchase value was.
**ARPU** (Average Revenue Per User) captures all three. Two patterns are common. First, a discounted price lifts trial starts but brings in users who churn faster after renewal: conversion goes up but ARPU goes down. Second, a slightly lower conversion rate at a higher price point generates more total revenue per user cohort: conversion is flat or down but ARPU is up. Track both. Optimize for ARPU.
## How to Build a Hypothesis for Your Paywall A/B Test
The worst version of paywall testing is opening a dashboard, clicking "create experiment," and choosing a variable because it seems interesting. Random testing generates noise, not insight.
### The 3-Part Hypothesis Formula for Paywall Tests
A testable hypothesis has three components: **what you're changing**, **what you expect to happen**, and **why you expect that outcome based on what you already know**. If you cannot articulate all three, the test isn't ready to run yet.
A working hypothesis looks like this: "Changing the default plan from monthly to annual will increase ARPU by 15% or more, because users who encounter annual pricing first are more likely to perceive monthly as expensive by comparison, and anchoring typically shifts plan selection toward annual."
If the "why" section of your hypothesis is empty or vague, that's a signal to do more analysis before testing. A hypothesis without a reason is a guess with extra steps.
### The Testing Sequence That Consistently Generates Returns
The teams that generate compounding results from paywall experiments follow a specific order. Pricing first. Trial structure second. Visual and copy changes third.
[Screenshot 2026-05-22 at 18.48.44.png]
| Round | What to Test | Why This Order |
| --- | --- | --- |
| Round 1 | Pricing (price points, plan mix, anchoring) | Highest LTV impact; all downstream tests interact with pricing |
| Round 2 | Trial duration and structure | Second-largest lever; directly affects trial-to-paid conversion and renewal behavior |
| Round 3 | Visual design, copy, layout | Lowest win rate; only valuable once price and trial are validated |
Testing visuals before pricing means optimizing conversion into an offer that hasn't been validated yet. Each layer affects the next. Add a weekly plan and your annual price perception changes. Raise the annual price and your conversion profile shifts. Get the pricing layer right first, then build on it.
> Benchmarks: apps with [trial periods in the 17-32 day range convert at a median of 45.7%](https://www.airbridge.io/en/blog/subscription-app-free-trial-conversion-rate) trial-to-paid, compared to **26.8%** for short 3-7 day trials ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Trial duration is not a minor variable: it's the second-most-impactful structural lever available, right after pricing.
## How to Set Up a Paywall A/B Test Step by Step
Once you have a hypothesis, the setup is straightforward. The discipline is in what you choose not to do.
### Step 1: Isolate One Variable Per Test
This rule exists for a reason. If you change the headline, price, and trial length simultaneously, you cannot isolate what drove the result. Every experiment should change exactly one thing: one price point, one trial length, one layout element.
If you want to move faster, run multiple experiments on separate user segments. Do not combine variables in a single test. A result from a multi-variable test tells you that some combination of changes worked. It does not tell you which element to keep.
### Step 2: Choose a Paywall Experiment Platform
Paywall A/B testing requires a tool that randomly assigns users to variants and tracks the full [subscription](https://www.airbridge.io/en/blog/subscription-app-pricing) funnel: trial start, paid conversion, and renewal, broken down by variant. Three platforms subscription apps commonly use:
- **RevenueCat Experiments** runs tests based on Offerings (combinations of price points, trial lengths, and plan structures). It supports up to four variants per experiment (A/B/C/D) and surfaces full-funnel analytics including LTV and renewal rates by variant.
- **Adapty** provides a no-code experiment builder with a visual paywall editor. It tracks conversion, LTV, and revenue by variant and allows you to ship paywall changes without a new app release.
- **Superwall** is purpose-built for paywall experimentation, with percentage-based traffic allocation and a visual editor for designing variants. It suits teams running frequent, rapid experiments.
The right tool depends on your existing subscription infrastructure. If subscriptions are managed in RevenueCat, use RevenueCat Experiments. If you use Adapty, use its built-in experiment tools. Keeping subscription data and experiment data in the same platform avoids reconciliation errors.
### Step 3: Calculate Sample Size Before You Launch
Sample size is not a guess. You can calculate it using your baseline conversion rate and the minimum improvement you want to detect. A practical floor: **at least 200 paid conversion events per variant** is enough to detect a large lift of 20% or more. For smaller improvements — which most real pricing changes produce — you will typically need **several thousand visitors per variant** ([Adapty Experiments Playbook](https://adapty.io/blog/paywall-experiments-playbook/)). Use an A/B test sample size calculator before you start, not after.
For an app converting downloads to paid at 2%, detecting a 10% relative improvement requires roughly 30,000 users per variant. For an app at 5%, it's proportionally less. Run this calculation before launching the test. If your traffic volume means reaching significance will take six months, either narrow your test to a higher-converting segment or reconsider the experiment timeline. A test you cannot complete is not worth starting.
## How Long Should You Run a Paywall A/B Test?
Test duration and sample size are related but distinct. Both must be sufficient for results to be trustworthy.
### The Minimum: 2 Weeks and Sufficient Paid Conversions Per Variant
Two weeks is the floor, not the target. The requirement is to satisfy both conditions simultaneously: **at least 2 calendar weeks** and enough paid conversions per variant to detect the lift you care about. The slower condition determines when the test is done.
If your app reaches your conversion threshold in five days, don't stop at five days. Continue to two weeks. If your app takes six weeks to reach the threshold, don't stop at two weeks. Continue to the sample threshold. Both conditions must be met before you trust a result.
### Pricing Tests Need 4 to 8 Weeks to Show Renewal Patterns
A conversion rate result tells you what happened at the moment of purchase. A pricing test is about more than that. It's about whether the users who bought at the new price are renewing and churning at comparable rates to the control group. That behavior only becomes visible after at least one billing cycle has passed.
**For any test that changes a price point, plan for 4-8 weeks of data collection** to capture at least one renewal cycle per variant (Adapty Experiments Playbook). A monthly subscriber who converts in week one hits their first renewal in weeks four or five. If you declare a winner before that data exists, you're working with half the picture.
**Nearly 30%** of annual subscriptions cancel in the first month (RevenueCat State of Subscription Apps 2025). That signal only appears if you wait for it. For more on how [retention rates](https://www.airbridge.io/en/blog/what-is-a-good-retention-rate-for-subscription-apps) behave across plan types, the benchmarks are worth reviewing before setting your evaluation window.
### When to Stop a Test Early (and When to Keep Going)
Stopping early because one variant looks like it's winning is usually a mistake. Early results are often shaped by novelty effects: users encountering a new layout for the first time behave differently than the broader population would over time. The fix is simple: set your duration in advance and do not look at results until both the time and sample thresholds are met.
Two legitimate reasons to stop early: one variant is producing materially lower revenue at statistical significance, which warrants protecting the control; or you have hit both your sample threshold and your time threshold simultaneously, and results are conclusive in both conversion and ARPU.
## How to Read Paywall A/B Test Results Without Being Misled
### Why ARPU Is the North-Star Metric for Paywall Tests
ARPU is the one number that simultaneously reflects conversion rate, price level, and plan mix. A test result showing higher conversion but lower ARPU is not a win. It means more users bought at a lower effective value per user.
| Conversion | ARPU | Decision | Next Step |
| --- | --- | --- | --- |
| ↑ Up | ↑ Up | Ship it | Deploy the variant to all users |
| ↓ Down | ↓ Down | Kill it | Return to control and revise the hypothesis |
| ↑ Up | ↔ Flat | Iterate | Check plan mix and post-trial churn by segment |
| ↑ Up | ↓ Down | Kill or iterate | Revenue is falling; revert unless the discount was intentional |
| ↔ Flat or slightly down | ↑ Up | Likely a win | Verify renewal rates before shipping |
| Mixed by segment | Mixed | Segment first | Analyze by platform and region before deciding |
### Segment by Platform and Region Before Declaring a Winner
A winning variant at the aggregate level can be a losing variant for your highest-value users. Before shipping any result, break down the numbers by at least two dimensions: **platform** (iOS vs Android) and **geography**.
Among subscription apps analyzed across more than 75,000 apps, trial-to-paid conversion rates vary significantly by region: North America converts at approximately **35.5%**, while South and Southeast Asia convert at around **15.9%** (RevenueCat State of Subscription Apps 2025). A test variant that wins in aggregate may be dragged upward by North American users while losing in every other market you care about.
Also segment by acquisition source if you have sufficient volume. Users from [paid acquisition](https://www.airbridge.io/en/blog/how-to-spend-your-first-10k-marketing-a-subscription-app) channels tend to show different price sensitivity than users from organic or referral sources.
### Set Your Decision Criteria Before the Test Starts
Decision criteria set in advance are immune to confirmation bias. Criteria set after looking at early data are not. Before launching any paywall test, write down three explicit thresholds:
- **Ship if:** ARPU improves by X% at 95% or higher statistical confidence over the required test duration.
- **Iterate if:** ARPU is flat but conversion rate changes significantly, indicating a possible offer structure issue worth exploring in a follow-up test.
- **Kill if:** ARPU declines or conversion falls below the control by a meaningful margin at statistical significance.
Once those thresholds are written and agreed before launch, the data tells you what to do. You don't have to decide in the moment.
## What a Paywall Experimentation Roadmap Looks Like
Apps running **50 or more paywall experiments** generate that **18.7x median revenue premium** over apps running just one (Adapty State of In-App Subscriptions 2026). The gap is not explained by any single winning test. It's explained by compounding: each experiment teaches you something that makes the next one more targeted and more likely to produce a meaningful result.
Among subscription apps that actively experiment, the average team runs **14.7 experiments per year**, roughly one test every 3 to 4 weeks (Adapty State of In-App Subscriptions 2026). At that cadence, a practical roadmap looks like this:
In the first three months, focus on establishing the pricing baseline. Test one or two price points against your current control. Once pricing is validated, move to trial duration. After pricing and trial structure are settled, test visual and copy elements. Each round compounds the one before it. Pricing optimized at the wrong trial length will give you different results than pricing optimized at the right one. Sequence matters.
## Your First Paywall A/B Test Builds the Entire System
Here is what the first test actually requires: pick one pricing variable you have never tested, write a full three-part hypothesis before opening any tool, and calculate the sample size your traffic volume can actually reach in a reasonable timeline. Set a fixed duration before the test starts. Write down the ARPU threshold that constitutes a win.
That is the whole system. Every subsequent test builds on what this one teaches you. The teams running **14.7 experiments per year** are not running more sophisticated experiments than you. They are running the same simple loop faster, with tighter hypotheses each time, because they have done it enough that the setup is automatic.
Start with pricing. Measure ARPU. Wait long enough to trust the result. Then build on it.
---
## When to Start Lifecycle Marketing for Your App (and When It's Too Early)
URL: https://www.airbridge.io/en/blog/lifecycle-marketing-for-app-timing
Language: English
Published: 2026-05-22
Author: Luke
Many subscription apps build [lifecycle marketing](https://www.revenuecat.com/blog/growth/lifecycle-marketing-campaigns-optimize-revenue/) campaigns before they are ready. Weeks go into email sequences, push automations, and win-back flows for a user base whose core behavior has not stabilized yet. The result is a polished system pointed at the wrong problem.
The apps that get lifecycle marketing right do not launch the most complete program. They start at the right moment. That moment is more specific than most guides acknowledge.
**Key Takeaways**
- **Lifecycle marketing amplifies what already works.** If your trial conversion or onboarding is broken, campaigns will not fix it. They will surface the problem faster, but the fix still requires product work.
- **Three prerequisites matter before anything else.** A stable conversion funnel, enough users to generate signal, and a churn pattern you can actually name.
- **The highest-ROI work happens in the first 48 hours, not in win-back.** Most subscriptions happen in this window. Users who convert later retain better when they are guided there.
- **Starting too early wastes the effort.** A win-back campaign for 50 churned users returns 1 to 2 reactivations at typical 2 to 5% reactivation rates. The same effort applied to onboarding 500 trial users returns measurable lift.
- **One stage. One problem. One experiment.** A lifecycle program does not need to be a 15-message system on day one.
## What Is Lifecycle Marketing for Apps?
Before deciding when to start, it helps to be precise about what you are committing to.
The same message sent to a brand-new trial user and a three-month subscriber performs very differently, because these users have completely different relationships with your product. Lifecycle campaigns close that gap by reaching each person based on where they actually are.
### 1\. The Four Lifecycle Stages: Potential, Active, Drifting, and Churned
Every subscription app user moves through four stages: **Potential** (installed, has not subscribed), **Active** (paying and engaged), **Drifting** (paying but usage is declining), and **Churned** (cancelled or lapsed). Each stage needs a different goal and a different message.
Treating all four stages as one is the most common segmentation failure. A discount sent to an engaged subscriber trains price-sensitivity. A re-engagement tip sent to a power user signals you are not paying attention. Stage is the variable that determines message, timing, and channel.
[Screenshot 2026-05-22 at 18.37.13.png]
_The four stages every subscription app user moves through, each requiring a different message_
### 2\. The Three Lifecycle Channels: Push, Email, and In-App
Lifecycle campaigns run across push notifications, email, and in-app messages. Most early-stage apps have access to all three from day one. The constraint is not the channels. It is whether you have enough behavioral signal to use them well.
## 3 Signs You Are Starting Lifecycle Marketing Too Early
There is no universal rule for when lifecycle marketing is premature. But three specific signals tell you the foundation is not ready yet.
### 1\. Your Trial-to-Paid Conversion Rate Is Below 34.8%
Across more than 75,000 subscription apps, the median trial-to-paid conversion rate is [**34.8%**](https://www.revenuecat.com/state-of-subscription-apps/) (RevenueCat State of Subscription Apps 2025). If your rate is significantly below that, particularly below **20%**, the problem almost certainly lives in your onboarding or paywall, not in what happens after the trial expires.
Adding a trial-expiry nudge to a 15% converting funnel is treating a symptom. Users who are not seeing enough product value during the trial will not be persuaded to pay by a follow-up email. Lifecycle campaigns can tip marginal decisions, but they cannot manufacture the perceived value the product failed to deliver. **Lifecycle marketing has a ceiling, and that ceiling is set by the conversion foundation underneath it.** Fix the funnel first.
### 2\. You Cannot Name Your Top Churn Reason
Before you build retention campaigns, you need to know what you are retaining against. According to Google Play cancellation data in the RevenueCat State of Subscription Apps 2025 analysis, the most common churn driver is **"not enough usage,"** accounting for **37.2%** of cancellations. The second most common reason is cost-related, at **34.6%**. These two problems require completely different responses.
"Not enough usage" is an engagement and [onboarding](https://www.airbridge.io/en/blog/subscription-app-onboarding-best-practices) problem: users are not forming habits or extracting consistent value. Cost-related churn is a pricing and perceived value problem: users understand the product but do not believe the price is justified. Lifecycle campaigns built without knowing which of these you are fighting will target the wrong audience with the wrong message. Run an exit survey via your subscription platform. App Store Connect exposes cancellation reasons through Subscription Events API; Google Play Console shows them under Cancellation Survey. Identify one pattern before building anything.
### 3\. Your User Base Is Too Small to Segment by Behavior
Segmentation is the mechanism that separates lifecycle campaigns from a mass blast. Sending the same push notification to everyone who installed three days ago, regardless of what they have done in your app, is not lifecycle marketing. It is broadcast messaging with a trigger delay.
Meaningful segmentation requires behavioral data: did the user complete onboarding? Start a trial? Reach the core feature? If your analytics are not tracking those milestones today, building these flows is premature. Build the measurement layer first.
## 3 Signs Your App Is Ready for Lifecycle Marketing
If the warning signs above do not apply, here are the three conditions that indicate genuine readiness.
| Signal | What "Ready" Looks Like |
| --- | --- |
| Conversion funnel stability | Trial-to-paid rate consistent for at least four weeks; you can explain the trajectory |
| Sufficient user volume | Enough monthly trial starts for experiments to return signal within weeks, not months |
| Named churn pattern | One specific, observed pattern you can write a campaign hypothesis around |
All three together create the foundation lifecycle programs need to amplify. If any one is missing, the program will measure the wrong things or move too slowly to generate actionable signal.
## How to Start Lifecycle Marketing: Where to Begin
Assuming you have met the prerequisites, the right starting point is almost never where most apps begin.
[Screenshot 2026-05-22 at 18.38.27.png]
_Start where users are still reachable, not at the end of the funnel_
### 1\. Start With Onboarding Re-engagement Before Win-Back
Win-back campaigns feel urgent because churned users are visible. But the economics do not favor starting there. Win-back reactivation rates typically sit at **2 to 5%** for apps without a strong prior relationship, compared to **30% or more** for nudges reaching users who are still in-trial and evaluating. A churned user who already decided to leave requires significant new evidence to return.
According to RevenueCat's lifecycle marketing research, [**most subscriptions happen within the first 24 to 48 hours**](https://www.revenuecat.com/blog/growth/lifecycle-marketing-campaigns-optimize-revenue/) after a trial starts. More broadly, RevenueCat's 2026 subscription benchmarks show that [80 to 90% of all trials begin on Day 0 of install](https://www.revenuecat.com/blog/growth/subscription-app-trends-benchmarks-2026/), making the early onboarding window the highest-volume, lowest-competition moment in your funnel. A **Day 1 and Day 2** push focused on guiding users to the core value moment is lightweight to build and addresses the most recoverable opportunity. Win-back is the last lifecycle investment, not the first.
### 2\. Find the Biggest Drop in Your Subscription Funnel
Look at your funnel from install to active subscriber and find the step with the sharpest fall-off. If 60% of trial starts happen but only 25% convert, your opportunity is in the trial-to-paid stage. If conversion is healthy but **nearly 30% of new annual subscribers cancel in their first month**, as shown in the [RevenueCat subscription cohort data](https://www.revenuecat.com/state-of-subscription-apps/), the post-subscription gap is your priority.
The largest drop is where a single well-targeted campaign will have the most visible impact. Starting there gives you a clear before-and-after measurement and concrete proof of ROI to justify building the next stage.
### 3\. Run One Lifecycle Experiment at a Time
The apps that build successful lifecycle programs do not launch a 12-email sequence in the first sprint. They run one experiment: one email, one push, one in-app message. They measure it, then build the next one. **Start small: one stage, one problem, one experiment.** The 15-message sequence gets built over six months of iteration, not deployed in week one.
Treating lifecycle as a system to "launch" is what creates the trap of starting too early. Think of it as ongoing experimentation. Each experiment informs the next campaign. The [RevenueCat subscription app growth stack guide](https://www.revenuecat.com/blog/growth/subscription-app-growth-stack/) covers how mature teams build this iteration loop over time.
## Start at the Right Stage, Not the Right Tool
Lifecycle marketing is worth doing. But the return depends heavily on when you start and where you start within it.
Apps that move too early spend months building on an unstable foundation. Apps that delay too long leave recoverable revenue sitting in leaky funnels. Open your analytics. Find the step with the largest drop. Start there.
---
## Push Notification Strategy for Subscription Apps: A Practical Retention Playbook
URL: https://www.airbridge.io/en/blog/push-notification-strategy-for-subscription-apps
Language: English
Published: 2026-05-21
Author: Luke
> **A push notification strategy** for subscription apps works when it is built around user lifecycle stages, not broadcast schedules. The push notifications that reduce churn are behavioral, timed to in-app activity, and specific to where each user is in their journey from install to active subscriber.
You send push notifications, but your open rate is low, and you cannot tell whether any of it is keeping subscribers around.
That is a strategy problem, not a copy problem. Push notifications can increase 90-day app retention by up to 190%, according to [Airship's benchmark study of 63 million users across 1,500 apps](https://www.airship.com/blog/7-mobile-engagement-statistics-that-show-how-push-notifications-boost-roi/). But that lift only materializes when notifications are built around lifecycle stages and behavioral triggers. When they are not, the same channel that could re-engage a drifting subscriber instead pushes 40% of users to permanently disable push permissions from your app.
**Key Takeaways**
- **"Not enough usage" is the leading reason subscribers cancel.** It accounts for 37.2% of cancellations across [75,000+ apps analyzed by RevenueCat](https://www.revenuecat.com/state-of-subscription-apps-2025/). Push notifications that drive usage directly address the root cause of churn.
- **Low open rates signal a targeting problem, not a content problem.** Personalized push notifications generate up to 4x higher open rates than generic broadcast messages, per [Airship's analysis of 50 billion push notifications](https://www.airship.com/blog/push-notification-strategy-customer-retention/).
- **The first 90 days are your highest-leverage window.** Apps that send relevant pushes in the first **90 days** see retention rates nearly **3x** higher than apps that send none, per the [same Airship benchmark](https://www.airship.com/blog/7-mobile-engagement-statistics-that-show-how-push-notifications-boost-roi/).
- **Frequency has a hard ceiling.** Sending 2 to 5 pushes per week causes 40%+ of users to disable push permissions entirely, per [Business of Apps](https://www.businessofapps.com/marketplace/push-notifications/research/push-notifications-statistics/).
- **Lifecycle stage determines what to send.** A trial user on Day 5 and a 4-month subscriber who has not opened in 10 days are different audiences with entirely different needs.
## Why Push Notification Open Rates Stay Low (Even When You Are Already Sending)
The average open rate is **4.6%** on Android and **3.4%** on iOS, per [Business of Apps](https://www.businessofapps.com/marketplace/push-notifications/research/push-notifications-statistics/) citing Airship data. If your numbers are below those benchmarks, the problem is rarely the message. It is the audience, the timing, or the absence of a lifecycle goal.
### 1\. Every user receives the same message
Most subscription apps start with a single push template sent to every user on a fixed schedule. That works when your user base is small and everyone is at a similar lifecycle stage. It breaks down the moment users diverge: some are mid-trial, some are active 6-month subscribers, and some have not opened the app in two weeks.
A notification re-engaging a fitness subscriber who missed five workouts is a different message from one targeting a user who stalled during onboarding. The copy is not the variable. The audience is.
### 2\. Sending too often destroys the channel
One push per week causes **10%** of users to disable push permissions. Two to five per week causes **40%+** to disable them, per Business of Apps.
Once users turn off notifications, that channel is gone. Most teams respond to low engagement by sending more. That accelerates the problem. The correct response is fewer, better-targeted messages triggered by behavior rather than a schedule.
### 3\. No lifecycle stage means no strategy
Every push that moves a retention metric has a specific job at a specific moment in the user journey. A push to a trial user on Day 12 has a fundamentally different goal from one to a subscriber who has gone quiet. Without that distinction built into your system, you are treating unlike users identically.
## The Four Push Types That Actually Reduce Churn
Not all pushes affect retention equally. These four types have a direct, measurable connection to subscriber behavior.
### 1\. Exit-intent pushes (Day 0)
When a new user finishes onboarding and exits without converting, they are at peak intent. Send a push within minutes of that exit, targeted to what they did not complete: reached the paywall but did not convert, finished setup but skipped a key feature, or created an account without adding any content.
Timing aligns with active intent. You are not interrupting their day — you are reaching them at the exact moment the decision is still open.
> _Example: "Your free trial is ready. Tap to start — it takes 30 seconds."_
### 2\. Trial expiry notification sequences
Users who start a trial convert at a median rate of 34.8%, with top-performing apps reaching 51.5% and above, according to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/). The gap between median and top performers is largely driven by how well the trial lifecycle is managed.
A three-touchpoint structure works for most trial windows:
| Trial Length | Push 1 | Push 2 | Push 3 |
| --- | --- | --- | --- |
| 7-day trial | Day 3: feature they have not tried | Day 6: 24-hour warning | Day 7: value recap |
| 14-day trial | Day 7: value reinforcement | Day 12: 48-hour warning | Day 14: last day |
| 30-day trial | Day 15: habit check-in | Day 27: 72-hour warning | Day 30: last day |
Push 1 should surface a premium feature the user has not explored. Push 2 and Push 3 should focus on specific value and conversion, not generic countdown messaging.
> _Day 6 example (7-day trial): "24 hours left on your trial. Here's what you will lose access to tomorrow."_
### 3\. Usage-triggered engagement notifications
The leading churn reason across subscription apps is "not enough usage," accounting for 37.2% of cancellations in [RevenueCat's analysis](https://www.revenuecat.com/state-of-subscription-apps-2025/) of 75,000+ apps. A push tied to a behavioral inactivity threshold directly addresses this.
A fitness app user who has not logged a workout in five days is drifting. A language app user who broke their daily streak is at risk. A meditation app user who has not opened in nine days is close to canceling. These are behavioral signals. Tie your triggers to them, not to a send schedule.
Timing matters as much as the trigger itself. Morning for wellness and fitness apps. Evening for entertainment. Post-work for productivity tools. According to [Pushwoosh](https://www.pushwoosh.com/blog/push-notification-best-practices/), sending at a user's peak activity window delivers up to **50% higher open rates** than off-peak sends.
> _Example: "You are 1 workout away from your 7-day streak. Log it now."_
### 4\. Re-engagement notifications for drifting subscribers
Drifting subscribers are still paying. They have gone quiet, but they have not canceled. They are more valuable to re-engage than churned users because the relationship is still active and no acquisition cost is required.
The window is narrow. After 2-3 weeks of inactivity, re-engagement becomes significantly harder. The push should answer one question for the user: where to pick up. Not "we miss you." Not a generic discount. A specific reference to where they were when they last used the product.
> "You were on Chapter 4 of Atomic Habits. Pick up where you left off?" performs better than "It's been a while. Come back." One is specific and actionable. The other is noise.
## What Good Strategy Looks Like: Two Case Studies
The four push types above are not theoretical. The apps that have built them well have left detailed records of what happened when they did. Here are two documented examples.
### Case Study 1: Duolingo
Duolingo's strategy is one of the most extensively documented examples of behavioral triggering at scale.
The core of their approach was the streak-saver notification: a push sent when a user with an active learning streak was about to lose it because they had not completed a lesson that day. The message was specific, time-sensitive, and tied to the exact behavioral event that mattered to that user.
Notification optimization specifically drove a **21%** increase in current user retention rate (CURR) and a **40%+** reduction in daily churn among their best users, documented by former CPO Jorge Mazal in [Lenny's Newsletter](https://www.lennysnewsletter.com/p/how-duolingo-reignited-user-growth). These gains were one part of a broader product portfolio — alongside leaderboards, streaks, and content investments — that helped grow DAU **4.5x** over four years.
Critically, Duolingo established one foundational rule: **protect the opt-in list**. They tested carefully rather than aggressively, knowing that burning the notification channel through over-sending would cost more than any short-term lift would gain.
### Case Study 2: Blinkist
Blinkist, the book summary subscription app, ran into a common problem: users were starting trials without understanding what they were agreeing to, which led to complaints and early cancellations.
Their solution was to rebuild the opt-in experience around transparency. Before prompting users to allow notifications, Blinkist showed exactly what they would receive: what type of notifications, at what frequency, and what each one was designed to help with.
Push opt-in jumped from **6% to 74%**, a **1,200%** increase, documented in a [Growth Design case study](https://growth.design/case-studies/trial-paywall-challenge) and confirmed via [Purchasely](https://www.purchasely.com/blog/blinkist-paywall-transformation-revolutionizes-app-user-engagement). The broader paywall transparency redesign — which included the new opt-in flow and a pre-trial-end reminder push — drove **+23%** trial sign-ups and **+4%** trial retention. Transparency about the notification experience built enough trust to improve conversion at the paywall itself.
## How to Map Pushes to Your User Lifecycle Stage
Knowing which push types exist is half the problem. The other half is knowing which one to send to which user, and when. That answer comes from lifecycle stage.
[Screenshot 2026-05-21 at 11.31.02.png]
Start with one stage, not all of them. Pick the highest-churn moment in your funnel. For most subscription apps, that is the trial-to-paid transition.
## Segmentation, Timing, and Frequency: The Mechanics of Higher Open Rates
The lifecycle map tells you what to send at each stage. These three mechanics determine how to send it in a way that users actually open and act on.
### 1\. Segment by behavior, not demographics
The most actionable segmentation for pushes uses behavioral signals: session frequency, features used, lifecycle stage, and days since last open.
Apps using behavioral segmentation see up to **10x** higher engagement than those running broadcast campaigns, per [Adapty's research](https://adapty.io/blog/how-to-increase-app-engagement/). Four segments that work across most subscription apps:
- **Highly active** (last open: 1-3 days): Deepen usage, surface underused premium features.
- **Moderately active** (last open: 4-7 days): Habit reinforcement tied to their specific content.
- **At risk** (last open: 8-14 days): Re-engagement with a specific hook back to their last session.
- **Dormant** (last open: 15+ days): Targeted offer, calibrated for users who have genuinely disengaged.
### 2\. Time pushes to user activity patterns, not the calendar
Use time-of-last-open as your starting proxy. A user who consistently opens at 7 AM should receive their push around 7 AM, not 2 PM. Layer your app's natural use-case cadence on top: morning for fitness, evening for entertainment, post-work for productivity.
### 3\. Cap frequency to protect your opt-in list
Set platform-level caps and enforce them:
- Maximum 1 push per day per user, across all push types combined.
- No pushes between 10 PM and 7 AM in the user's local time zone.
- Hard limit: no more than 5 pushes in any 7-day window for any single user.
Build coordination logic into your push setup that checks what a user has already received before triggering the next message. A user should not receive an exit-intent push, a trial expiry warning, and a re-engagement message within the same 24-hour window.
## How to Measure Whether Your Pushes Are Reducing Churn
Open rate tells you about the message. It does not tell you about the subscription.
The test that matters: split your users into those who received and acted on a specific push type versus those who did not, then compare their 30-day subscription retention. If the open cohort retains at a 5+ percentage point higher rate, the push is contributing to retention. If the gap is smaller, the message is generating clicks without changing subscriber behavior.
Three metrics to track alongside open rate:
- **In-app action after push:** High open rate with zero follow-through means the push generated a click but did not re-engage the user with the product.
- **30-day subscription status, pushed versus non-pushed:** This is the actual churn connection test.
- **Push opt-in rate trend:** A declining opt-in rate week-over-week signals frequency or relevance problems before they surface in cancellation data.
One benchmark worth knowing: if your monthly subscriber 30-day retention is below **17.5%**, the median across 75,000+ apps in RevenueCat's analysis, frequency and segmentation are the first levers to audit.
## Start With One Trigger, Then Scale
If you are currently running one push template to all users, the highest-leverage move is not writing a better template. It is building one behavioral trigger.
Pick the highest-churn moment in your funnel. Build a single triggered push for that specific moment. Measure whether the cohort that receives it retains at a higher rate than those who do not. Once you have evidence that one behavioral push works, build the next one.
---
## How to Build a Win-Back Email Sequence for Your Subscription App
URL: https://www.airbridge.io/en/blog/subscription-app-win-back-sequence
Language: English
Published: 2026-05-21
Author: Luke
You spent real money getting users to subscribe. Some cancelled. Right now, most of those cancellations are sitting in a database, untouched.
That is a recoverable problem. Among apps tracked across more than 75,000 subscription products ([State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)), the natural reactivation rate for monthly subscribers reaches **13.7% within 12 months** without any win-back effort. Acquiring a new subscriber costs **5-25x more** than retaining an existing one ([Harvard Business Review](https://hbr.org/2014/10/the-value-of-keeping-the-right-customers)). A structured subscription app win-back sequence captures that opportunity systematically.
**Key Takeaways**
- **Not all churned users are worth pursuing.** Cost-related churners and billing-failure churners (**28%** of Google Play cancellations) respond well to win-back. "Found a better app" churners typically do not. Segment before you send.
- **The first message needs to go out fast.** The most effective first win-back touchpoint goes out within 24 hours of cancellation, while the user is still in the mental frame of your product.
- **Lead with value, arrive at the discount last.** Opening with a coupon trains users to wait for the deal, then churn again at full price once the promotion ends. Non-discounted resubscribers stay **2.6x longer** than discount-acquired ones.
- **Monthly churners are your best starting audience.** They reactivate at **13.7%** within Year 1 and have the shortest commitment cycle. Annual churners reactivate at only **4.6%**.
- **A win-back sequence is a series, not a single email.** A 4-email escalation sequence achieves approximately **14.7%** cumulative reactivation ([Klaviyo, 2025](https://www.klaviyo.com/blog/winback-email-campaign-examples)).
## Why Win-Back Campaigns Outperform New Subscriber Acquisition
Win-back offers can add **10-25%** to your bottom line, making it one of the highest-ROI activities in your lifecycle stack. A subscription app win-back sequence is a timed series of messages sent to churned subscribers, typically spanning 30 to 60 days after cancellation, triggered automatically when a subscriber's status changes to cancelled.
### 1\. The ROI Case for Targeting Churned Users
The baseline reactivation data from 75,000+ subscription apps shows the opportunity by plan type:
| Plan Type | Natural Reactivation Rate (Year 1) | Win-Back Priority |
| --- | --- | --- |
| Monthly | 13.7% | High |
| Weekly | 9.4% | Medium |
| Yearly | 4.6% | Low |
Monthly subscribers reactivate at more than three times the rate of yearly subscribers without any outreach. For an app with 500 monthly churners over a year, that baseline represents 68 users returning with zero effort. A targeted sequence can add meaningfully to that number.
Apps with high-price tiers see **11.9%** reactivation versus **9.4%** for low-price apps. These were users willing to pay a premium once. They are worth contacting again.
### 2\. Which Churned Users Should You Skip?
Segments to deprioritize:
- **"Found a better app" churners** (**9.5%** of Google Play churners). Their decision was comparative. You cannot run a product demo in three email paragraphs against a competitor they are now using daily.
- **Annual subscribers who cancel in month one.** Nearly **30%** of annual subscriptions cancel within the first month. A 4.6% reactivation ceiling means the ROI on a full sequence for early annual churners rarely pencils out.
- **Users with zero sessions in their last 30 days before cancelling.** Without behavioral data to personalize around, win-back messaging turns generic and performs below baseline.
Concentrate your first sequence on monthly subscribers who had some engagement before cancelling. They represent the highest-probability cohort and are the right starting point for learning how to win back churned app users at scale.
## How to Diagnose Churn Before Writing a Single Email
The churn reason determines both the message and the offer. It is the single most important variable in knowing how to win back churned app users effectively.
Data from 75,000+ subscription apps shows that the two most common churn reasons, "not enough usage" at **37.2%** and "cost-related" at **34.6%**, require fundamentally different win-back approaches. Usage-driven churners need value re-education. Price-driven churners need a concrete financial offer.
### 1\. The 5 Churn Types and What They Mean for Win-Back
| Churn Type | Share of Cancellations | Win-Back Suitability | Best Approach |
| --- | --- | --- | --- |
| Not enough usage | 37.2% | Medium | Feature education, re-activation prompt |
| Cost-related | 34.6% | High | Discount or lower-tier plan offer |
| Technical issues | 7.1% | High (if fixed) | Resolution confirmation + reactivation offer |
| Found better app | 9.5% | Low | Differentiation message (low success rate) |
| Other / unspecified | 11.6% | Varies | Requires exit survey data |
"Not enough usage" signals that users never reached the habit loop your product needs to retain subscribers long-term. Win-back for this group is not about price. It is about showing them something specific they missed: a feature they never tried, a workflow that directly solves the problem they came in with, or the outcome they originally signed up to achieve.
Cost-related churn is the most actionable segment. These users valued the product enough to pay. They left because the value-to-price ratio stopped making sense. A time-limited offer, a paused subscription option, or a lower-tier plan can recover them.
Technical churn (7.1%) tends to convert at high rates when the message is timed against the fix announcement. A direct note confirming the fix, paired with an offer to return, is the most targeted type of win-back message you can send.
### 2\. How to Build Your Churned User Segments
If you lack exit survey data, use behavioral proxies to approximate churn type:
- **Recency.** Users who churned within the last 30 days are warmest. Extend outreach to 90 days. Beyond 90 days, reactivation drops to roughly **10-12%**, and beyond 180 days to **2-4%** ([Validity](https://www.validity.com/resource-center/how-successful-are-email-win-back-programs/)).
- **Plan type.** Monthly churners are the starting point. Shorter commitment cycles make them easier to re-engage with a one-cycle offer.
- **Session activity in the final 14 days.** Users who were logging sessions before cancelling are usage-disengaged, not value-disengaged. They are more winnable than users who had been completely inactive.
- [**LTV**](https://www.airbridge.io/en/glossary/lifetime-value-ltv) **history.** Prioritize users in the top quartile. Their resubscription value justifies a stronger offer.
Build at minimum three segments before your first sequence: recent monthly churners with activity, recent monthly churners without activity, and annual churners. Sequence one targets the first group only.
## How to Build Your Win-Back Email Sequence: A Practical Guide
### 1\. When to Send Each Email in Your Win-Back Sequence
The first win-back touchpoint goes out **within 24 hours of cancellation**. Not three days later. Not at the end of the billing period. Within 24 hours.
At that moment, the user is still in the mental frame of your product. They have not fully committed to an alternative. They have not deleted the app. Outreach within the first 30 days is roughly **3x more effective** than later attempts ([Validity](https://www.validity.com/blog/7-email-reactivation-campaign-insights/)).
After the first message, space out the remaining touchpoints:
[Screenshot 2026-05-21 at 11.19.05.png]
_Win-back email sequence timeline: 4 touchpoints over 60 days_
Three emails cover the majority of the reactivation window for monthly churners. A fourth email makes sense only for users who opened at least one prior message.
### 2\. What to Write in Each Win-Back Email
**Email 1** does not ask for anything. It acknowledges the cancellation, references one or two outcomes the user actually experienced in the product, and keeps the return path low-friction.
\> **Subject:** Your progress is still saved.
> "You completed onboarding and unlocked \[Feature\]. It's still here whenever you're ready to come back."
"Your progress is still saved" works better than "We miss you." The first is about the user. The second is about you.
**Email 2** focuses on something the user did not fully explore. This requires behavioral data: which features did they try, which did they skip, what goal did they set during onboarding and never finish.
\> **Subject:** You never tried \[Feature X\].
> "Most users who reach \[outcome\] used \[Feature X\] first. You set it up but never activated it. Here's how it works."
**Email 3** is where the offer arrives. The most effective options for subscription apps:
1. An **extended** [**free trial**](https://www.airbridge.io/en/blog/subscription-app-free-trial-conversion) (**14-30 days**) for users who cancelled early and likely did not hit their aha moment. A second trial reframes the decision without requiring a financial commitment.
2. A **one-cycle discount** (**20-40%** off the next billing period) for cost-driven churners identified through exit surveys or behavioral signals.
3. A **lower-tier plan offer** if your pricing structure supports one. A stripped-down version that retains them is better than no revenue at all.
\> **Subject:** Come back for 30 days at half price.
> "Here's 50% off your next billing cycle. No commitment beyond that."
### 3\. How to Structure Your Win-Back Offer
Discounting immediately is the most common win-back mistake. Users who resubscribe only because of a coupon tend to churn a second time when the promotional period ends, a pattern practitioners call "post-promo churn." Non-discounted resubscribers stay **2.6x longer** than discount-acquired ones ([Phoenix Strategy Group](https://www.phoenixstrategy.group/blog/how-subscription-discounts-impact-churn-rates)).
The three-email structure above addresses this by placing the offer in Email 3, after two rounds of value-first messaging. If a user resubscribes before Email 3 arrives, you did not need to spend the discount.
### iOS Native Win-Back Offers (and Why They Do Not Replace Your Sequence)
Apple introduced native win-back offers for auto-renewable subscriptions at WWDC 2024, with general availability from September 2024 ([Apple Developer](https://developer.apple.com/news/?id=8utnewzk)). These promotional offers appear across the App Store, in iOS Settings (Manage Subscriptions), and inside your app, without any push or email required.
This makes them especially effective for churned subscribers who have opted out of notifications. Treat them as a parallel recovery surface, not a replacement for your sequence. Configure them in [App Store Connect](https://developer.apple.com/help/app-store-connect/manage-subscriptions/set-up-win-back-offers/) under the Subscriptions section. The in-app sheet requires iOS 18 or later; the Manage Subscriptions display works on iOS 14.3 and later.
Google Play offers a comparable mechanism through Subscription Offers and personalized pricing, scoped to Android subscribers.
## The Win-Back Mistake That Undermines Everything
Even a well-structured subscription app win-back sequence underperforms if the cancellation exit experience is broken. If your [churn](https://www.airbridge.io/en/blog/subscription-app-churn-rate) flow is a single button press with no acknowledgment, no recap of what the user built, and no "your data is still here" signal, you are losing users who were still deciding.
A pause option, a brief reminder of saved progress, or a low-friction downgrade path reduces involuntary loss and creates a more receptive audience for win-back messaging later. Fix the exit before optimizing the re-entry.
## How to Launch Your First Win-Back Campaign (Start Small)
Take monthly subscribers who churned in the last **30 days** with at least one session in their final **two weeks**. Build three emails: within 24 hours, at day 7, at day 30. Measure [retention benchmarks](https://www.airbridge.io/en/blog/subscription-app-retention-benchmarks) and revenue recovered per email. Iterate from there.
Win-back is not a one-time campaign. It is an ongoing channel within your [lifecycle marketing](https://www.airbridge.io/en/blog/subscription-app-lifecycle-marketing) stack. Every month that passes without a sequence running is a month of recoverable subscribers who move further out of reach.
---
## How to Build a Simple App Growth Loop for Subscription Apps
URL: https://www.airbridge.io/en/blog/how-to-build-a--subscription-app-growth-loop
Language: English
Published: 2026-05-21
Author: Luke
You open three tabs: RevenueCat shows churn is up. Mixpanel shows activation dropped. Meta Ads Manager shows spend is holding. Three dashboards, three data points, and no answer to the only question that matters: **what do you fix first?**
This is the problem with data-heavy, action-light app growth. Most subscription apps don't lack reports. They lack a loop: a repeatable system that converts a metric into a specific next action. The apps that grow consistently aren't watching more dashboards. They're running a **simple growth loop for subscription apps** that tells them exactly what to test each week.
**Key Takeaways**
- **Reports describe the past. A growth loop drives the future.** A dashboard tells you churn went up. A loop tells you which node is broken and what experiment to run this week.
- **A simple app growth loop has four nodes: Acquire, Activate, Retain, Expand.** Each node has one metric, one action threshold, and one experiment.
- **Activation is where most loops break first.** The top churn reason for subscription apps is "not enough usage" (37.2% of cancellations), not price and not competition.
- **Experimenting teams generate up to 40x more revenue.** The loop is the system that tells them what to test, not intuition and not the metric that looks worst in the weekly report.
- **The loop closes only when acquisition connects to retention.** If you can't see which channels produce subscribers who stay, the Acquire node is running blind.
## Why Your Subscription App Has Data But No Clear Next Action
### Why dashboards answer the wrong question
**Adding a fifth analytics tool doesn't solve this.** The problem isn't missing data. It's missing structure. Data without a decision framework produces more uncertainty, not less. Most early-stage apps already have more than enough data.
A growth loop is a repeatable system that connects each metric to a specific action, cycling continuously so each experiment's output becomes the input for the next cycle. That structure is what converts a reporting habit into a growth habit.
[Screenshot 2026-05-21 at 10.05.37.png]
## The Four-Node App Growth Loop: Acquire, Activate, Retain, Expand
The simplest growth loop for subscription apps has four nodes. You don't need to run all four at once. Start with the one that's leaking most.
| Loop Node | Primary Metric | Action Threshold | Example Experiment |
| --- | --- | --- | --- |
| Acquire | Cost per Subscription (CPS) | CPS exceeds 2x LTV target | Test new creative format on best-performing channel |
| Activate | % users reaching activation event in 48h | Below your category median | Move paywall or key action earlier in onboarding |
| Retain | Monthly plan first renewal rate | Below 55.5% median | Add re-onboarding push at Day 3 for low-usage subscribers |
| Expand | Trial-to-paid conversion | Below 34.8% median | Test 14-day trial vs 7-day trial on paywall |
_All thresholds are all-category medians (RevenueCat State of Subscription Apps 2025). Category benchmarks vary: Health & Fitness monthly renewal runs near 45%, Productivity near 70%._
_If you don't know your category median, the gap between your top cohort and your median cohort is the more useful signal._
### 1\. Acquire: Why cost per subscription (CPS) beats cost per install (CPI)
Cost per Install (CPI) is the most commonly tracked metric in the Acquire node. It is also the least useful for subscription apps.
Imagine a **$1 CPI** from TikTok versus a **$4 CPI** from Meta. If the Meta user has **3x the retention rate**, both channels produce identical subscription revenue. **The metric that matters in the Acquire node is Cost per Subscription (CPS), not Cost per Install.** CPS connects your ad spend directly to the output that generates revenue. CPI does not.
If your only acquisition data is CPI by channel, you're optimizing the wrong number. Refocus the Acquire node on which channel produces subscribers, not which channel produces the cheapest installs.
### 2\. Activate: What the activation sequence looks like in a subscription app
**82% of trials start on Day 0** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Your paywall and your onboarding are not separate decisions. They are part of the same activation sequence. If users are not reaching the paywall in their first session, the Activate node is leaking before it starts.
Build your activation metric around the user behavior that predicts 30-day retention for your specific app. "Completed onboarding" is too generic. The activation event needs to be tied to a specific instance of value delivered: a workout logged, a language lesson completed, a report generated.
Failed activation is what later shows up as "not enough usage" cancellations, the same **37.2%** cited in the Retain node data. When users churn for that reason, the fix belongs here in Activate, not in a win-back campaign.
### 3\. Retain: How to find your dominant churn reason before building a retention system
Before investing in push notification sequences, re-engagement emails, or win-back campaigns, identify which churn reason dominates your app.
**37.2% of subscription app cancellations are driven by "not enough usage"** (RevenueCat State of Subscription Apps 2025). Cost-related churn comes second at **34.6%**. These two reasons require completely different fixes. Usage churn means the product is not habit-forming enough. Lifecycle discounts won't fix it. Cost churn can sometimes be addressed with offer timing and plan restructuring.
Running a discount campaign against a usage churn problem burns budget on the wrong experiment. One cancellation survey question ("Why did you cancel?") tells you which node to fix first. For [subscription app retention benchmarks](https://www.airbridge.io/en/blog/what-is-a-good-retention-rate-for-subscription-apps) by category, see our full data guide.
### 4\. Expand: How subscription apps grow revenue through structured experiments
Expansion means increasing revenue per user: longer subscription periods, better trial-to-paid conversion, or higher plan defaults.
**Apps that actively experiment on pricing and paywall setup earn nearly 40x more revenue than apps that don't** ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)). The average experimenting app runs **14.7 experiments per year** (Adapty State of In-App Subscriptions 2026), roughly one experiment every three weeks. Across 75,000+ apps, the top 10% of subscription apps generate **6x+ the revenue per install** compared to the median (RevenueCat State of Subscription Apps 2025). The difference is not their tech stack. It is their cadence of deliberate improvement.
Apps with longer trial windows (17-32 days) reach a median trial-to-paid conversion of **45.7%**, versus **26.8%** for short-window trials, making trial length one of the highest-leverage experiments in this node. See [trial-to-paid conversion benchmarks by category](https://www.airbridge.io/en/blog/subscription-app-free-trial-conversion-benchmarks).
## How to Run Your First App Growth Loop Cycle in 3 Steps
### 1\. How to find your leakiest growth node
Look at your metrics and find the node with the largest gap between your current number and the category benchmark. If your trial-to-paid conversion is below **34.8%** (the all-category median per RevenueCat State of Subscription Apps 2025), the Expand node is the priority. If your monthly plan first renewal is below **55.5%**, Retain needs attention first.
Once you have identified the leakiest node, drill into the dominant sub-signal within it before writing your hypothesis. If Retain is the priority, run a cancellation survey first: usage churn and cost churn require different experiments. If Activate is the priority, confirm whether users are reaching the paywall in their first session before building any lifecycle sequences.
**Pick the worst one. Everything else waits.** Trying to fix multiple nodes simultaneously splits your team's attention and makes it impossible to learn what worked.
### 2\. How to write a testable growth hypothesis
> _"We are adding a Day 3 push notification because low-usage subscribers are churning before they reach the core value moment, and a personalized reminder should pull them back into the product."_
Run one experiment per cycle. Two simultaneous changes to the same node make it impossible to attribute any improvement to a specific cause.
### 3\. How to feed experiment results back into the loop
After the experiment concludes, the output becomes the input for the next cycle. If the Day 3 push improved monthly plan first renewal from 40% to 46%, ask: which user segment responded? Which platform? Which onboarding cohort?
That specific answer shapes the next experiment. Each cycle produces a more precise question than the last. For more on structuring these experiments, see our [app marketing resource library](https://www.airbridge.io/en/blog).
## Three Mistakes That Stall Your App Growth Loop
- **Optimizing for CPI instead of CPS.** CPI is disconnected from subscription revenue. Scaling spend without knowing which channels produce paying subscribers generates no [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv). Track Cost per Subscription instead. The Acquire node must connect to the Retain node or it is not a loop.
- **Running more than one experiment at a time.** Two simultaneous changes to the same node make it impossible to know what moved the metric. One experiment per cycle is a constraint that produces learning, not a limitation that slows growth.
- **Skipping the Activate node entirely.** Most teams focus on Acquire (more spend) and Retain (reduce churn), and ignore the sequence between install and first subscription. That sequence is where retention is decided. If users never reach the value moment, no lifecycle campaign will recover them.
## From Weekly Reports to Weekly Actions: Make App Growth a Habit
Every subscription app has reports. The ones that compound have a loop. The difference is not the quality of the data. It is the structure you bring to it.
Pick one leaky node this week. Write one hypothesis. Run one experiment. Six months from now, the question will no longer be "what do all these reports mean?" It will be: "which hypothesis are we testing next?"
---
## Why Real-Time Cross-Platform Tracking Improves Ad Performance for Subscription Apps
URL: https://www.airbridge.io/en/blog/real-time-cross-platform-tracking-ad-performance-subscription-apps
Language: English
Published: 2026-05-19
Author: Luke
In 2026, manual ad performance reporting is no longer just inefficient.
A single copy-paste mistake, mismatched timezone, or inconsistent naming convention can distort ROAS and delay the decisions that matter most. The bigger problem is timing. By the time the spreadsheet is cleaned up, normalized, and shared, the best window for same-day optimization is often gone.
That is why real-time cross-platform performance tracking matters now. Let's dig into it.
[diagram-manual-vs-realtime-workflow.svg]
**Key Takeaways**
- **Manual ad reporting is now a growth risk, not just an efficiency problem.** One copy-paste error, broken formula, or misplaced decimal can distort ROAS and lead teams to scale weak campaigns or pause profitable ones.
- **Cross-platform data is too fragmented for spreadsheets to handle reliably at scale.** Different time zones, currencies, attribution windows, and update cycles across Meta, Google, TikTok, Apple Search Ads, and programmatic channels make manual reconciliation slow and fragile.
- **82% of trial decisions happen on the same day as install** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). A report that arrives at 2 PM has already missed the audience it was meant to inform.
- **Teams running structured experiments see up to 40x revenue lift versus teams that don't experiment** ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)). Slow reporting kills the ability to experiment at all.
- **Real-time tracking only matters when it connects spend to revenue.** Marketers need one normalized view that ties channel spend to installs, trials, subscriptions, and revenue, not just faster dashboards.
## Why Traditional Ad Performance Reporting Fails for Subscription Apps
The operational reality in 2026: marketers are still expected to make same-day budget decisions, but many teams are working from reports stitched together hours after spend has already been deployed.
The issue is no longer dashboard convenience. It is whether the team can trust the numbers early enough to act on them. Three factors make the old way of analyzing data obsolete.
### 1\. Human Error in Manual Reporting
Manual reporting works until complexity increases. As more campaigns, creatives, regions, and revenue events are added, the workflow becomes fragile. One copy-paste error, one spreadsheet version conflict, or one late-night mistake can distort the numbers and push the team toward the wrong decision.
The real risk is that these mistakes often look credible. A missed row, a broken formula, or a misplaced decimal can make ROAS appear stronger or weaker than it actually is, leading teams to scale weak campaigns or cut profitable ones.
Common manual-reporting failure points:
- Copy/paste errors across channel exports
- Decimal mistakes that distort ROAS, CAC, or payback calculations
- Spreadsheet version conflicts across team members
- Late-night reporting fatigue that reduces QA discipline
### 2\. Data Fragmentation Across Platforms
Cross-platform reporting is difficult even without manual mistakes because each platform reports data differently. Meta Ads, Google Ads, TikTok Ads, Apple Search Ads, and programmatic channels can all vary by timezone, currency, attribution window, and update speed.
That is where Excel breaks down. The spreadsheet becomes a manual normalization layer, and every extra adjustment adds risk. The usual fragmentation issues include:
- **Time zone mismatch:** one source closes the day earlier or later than another
- **Currency mismatch:** spend and revenue require manual normalization
- **Attribution mismatch:** platforms claim conversions under different windows
- **Update-cycle mismatch:** one dashboard is near-live while another lags behind
### 3\. Optimization Happens Too Late
The bigger issue is timing. Even if the spreadsheet is accurate, it often arrives too late to act on. If reporting is finished by **2 PM**, the morning optimization window is already lost and underperforming campaigns may have already burned the budget.
For subscription apps, that delay is especially costly. RevenueCat's analysis of **75,000+** subscription apps found that **82% of trial decisions happen on the same day as install** ([SOSA 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). When reporting depends on manual cleanup, the audience that matters most for today's spend has already made their decision before the report lands.
That delay also weakens the feedback loop between spend and revenue. Teams react slower, waste more budget, and take longer to find what actually scales.
## How Real-Time Cross-Platform Tracking Works
Real-time cross-platform tracking works by pulling data from ad channels, app events, and revenue signals into one system that updates fast enough for same-day decisions. Instead of reviewing each platform separately, marketers get one view of performance across the full journey, from ad spend to subscription revenue.
### 1\. What the workflow looks like
| Step | What happens | Why it matters |
| --- | --- | --- |
| Data ingestion | Pulls spend and performance data from ad channels | Centralizes fragmented channel data |
| Event tracking | Captures installs, sign-ups, trials, and subscriptions | Connects traffic to real outcomes |
| Normalization | Aligns timezone, currency, and attribution logic | Makes channel comparisons reliable |
| Reporting | Updates dashboards quickly | Helps teams act on today's data |
| Optimization | Budgets and creatives are adjusted faster | Reduces wasted spend |
### 2\. It collects data from multiple sources
The first layer is data collection. A tracking platform pulls in performance data from channels like Meta Ads, Google Ads, TikTok Ads, and Apple Search Ads, then connects that with app and product events such as installs, sign-ups, trial starts, and subscriptions.
That matters because ad platforms can show spend and clicks, but they usually stop short of showing the full revenue outcome. To evaluate ad performance properly, teams need both acquisition data and downstream conversion data in the same place.
### 3\. It normalizes the data
Once the data is collected, it needs to be standardized. Different platforms may use different time zones, currencies, attribution windows, and naming structures. A real-time tracking system cleans that up so teams can compare channels on a like-for-like basis.
Without this layer, cross-platform reporting stays inconsistent. One dashboard may look strong simply because it updates faster or attributes more aggressively, not because it is actually driving better results.
### 4\. It connects ad spend to the funnel
The next step is funnel mapping. The platform links channel-level spend to user actions across the conversion path:
- Install
- Sign-up
- Trial start
- Subscription
- Revenue
This is what turns reporting into decision-making. A campaign may generate cheap installs, but if those users do not start trials or subscribe, the apparent efficiency is misleading. Adapty's 2026 analysis of **16,000+** apps found that **teams running structured experiments achieve up to 40x more revenue than teams that don't experiment** ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)). That advantage is impossible to capture without a feedback loop that connects spend to subscription outcomes in time to act.
### 5\. It updates fast enough for action
The point of real-time tracking is not constant monitoring for its own sake. It is to shorten the time between performance change and optimization. If one creative starts losing efficiency in the morning, the team should be able to catch it before half the day's budget is gone. If one channel starts driving stronger trial-to-paid conversion, budget should be moved while that signal still matters.
## What to Look for in a Real-Time Cross-Platform Tracking Platform
A good platform should reduce reporting work, standardize messy inputs, and help teams move from spend to revenue without guessing.
### 1\. Cross-platform normalization
If a platform cannot align timezone, currency, naming conventions, and attribution logic across channels, it will not solve the actual reporting problem. [Google Ads](https://support.google.com/google-ads/answer/3123169?hl=en), for example, lets advertisers set different conversion windows, which shows why raw platform numbers are not naturally comparable.
### 2\. Full-funnel event tracking
The platform should connect ad spend to the events that matter after the click. For subscription apps, that usually means install, sign-up, trial start, subscription, and revenue. Google [publishes a list of recommended events for measurement](https://support.google.com/analytics/answer/9267735?hl=en) that is a useful starting reference. Event-based measurement is the foundation; without it, teams are still optimizing for shallow metrics like clicks or installs.
For early-stage paid UA budgets, **trial start** is usually the most useful event to send back to ad platforms as the optimization signal. It carries enough volume to train the algorithm, and it sits close enough to the paying event to indicate real intent. Install is too noisy to optimize against, and subscription start is too sparse at small budgets to give the algorithm a stable target.
### 3\. Revenue and retention visibility
A platform should not stop at acquisition. Marketers need to see whether users convert, retain, and generate revenue over time. [Cohort](https://www.airbridge.io/en/product/audience-manager) and [retention](https://help.airbridge.io/en/guides/retention-report) reporting matters because early conversion volume can look strong while long-term value stays weak.
### 4\. Fast enough to support same-day action
Fresh data only matters if the team can act on it. The platform should update quickly enough to catch wasted spend before the day is over, and it should be explicit about gaps that live data cannot close yet. iOS SKAN postbacks can lag **24 to 72 hours**, so a 9 AM dashboard should label which numbers are deterministic versus which are still pending, not paper the difference over to look faster than it is.
### 5\. Export and integration fit
The platform should fit the team's workflow, not force workarounds. That means checking whether it supports the channels you use, whether it connects to the rest of your stack, and whether you can get the data out in a usable form as reporting needs grow.
## How Airbridge Enables Real-Time Cross-Platform Ad Performance Tracking
Airbridge Core Plan is built for early-stage paid UA teams that need to answer one practical question in real time: **are paid users converting into subscriptions, and which channels are driving value?**
If your team is spending across major ad platforms but still relying on manual reporting to connect spend to trials, subscriptions, and revenue, Core Plan is designed to shorten that loop. It supports the **4** major self-attributing networks (Google, Meta, Apple Search Ads, and TikTok) and connects channel spend to subscription-funnel events such as install, sign-up, start trial, subscribe, and order complete inside Funnel, Retention, and Revenue reports.
What that looks like in practice for a paid UA team in week one and beyond:
- **Day 1:** Connect Meta and Google, see channel-level spend in one normalized view
- **Day 3:** Compare same-day cost per trial across channels, not just cost per install
- **Day 14:** Identify which channel's trials retain past Day 7 inside the Retention report
- **Day 30:** Shift budget based on revenue contribution by channel, not install volume
| What Core Plan helps with | What that means for marketers |
| --- | --- |
| Standard event tracking for the subscription funnel | See whether paid traffic moves into trials and subscriptions |
| Coverage across Google, Meta, Apple Search Ads, TikTok | Monitor major paid channels in one measurement setup |
| Funnel, Retention, and Revenue reports | Judge performance beyond installs and clicks |
| Cost data upload and attribution rules | Compare channels with cleaner, more usable data |
## Start with One Channel, Then Tighten the Feedback Loop
The fastest way to find out if your team has a tooling problem is small. Export **one week** of Meta and Google spend into a single sheet. If reconciling timezone, currency, and attribution windows takes more than **90 minutes**, the bottleneck is reporting infrastructure, not headcount.
From there, decide what real-time visibility is worth to the team. If a **4-hour** reporting lag is costing more than one optimization cycle per day, a platform like Airbridge Core Plan is built to compress that loop, taking raw channel data to a subscription-revenue view in time for same-day decisions.
---
## How to Reduce Wasted App Ad Spend Without Losing Subscription Volume
URL: https://www.airbridge.io/en/blog/reduce-wasted-app-ad-spend
Language: English
Published: 2026-05-18
Author: Luke
You're spending $30,000 a month on Meta and Google ads. Installs are coming in. Trials are starting. But subscription revenue isn't moving fast enough, and you have no clear picture of which campaigns are responsible for the users who actually subscribe.
That uncertainty is not a targeting problem. It is a measurement problem. Three forces drive most waste: wrong optimization goals, fatigued creatives, and channels that convert installs but not subscribers. The result is always the same: more spend, same revenue, growing frustration.
**Key Takeaways**
- **Optimizing for installs is optimizing for waste.** Cost per install (CPI) is a vanity metric for subscription apps. The cost per paying user (CPPU) runs **4-5x higher than CPI**, meaning your real acquisition cost is far above what your ad platform reports. ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps/))
- **The gap between median and top performers is enormous.** Across 75,000+ apps, the median D35 subscription conversion rate is **1.9%**, but top-performing apps (P90) reach **8.5%**. A 4x gap driven by optimization decisions, not luck. (RevenueCat State of Subscription Apps 2025)
- **High trial volume can mask low quality.** The apps with the highest install-to-trial rates often have the lowest trial-to-paid conversion. Volume is not a signal of intent.
- **Reallocating beats cutting.** Splitting budget across **70%** proven, **20%** optimization, and **10%** exploration holds subscription volume while removing waste from the bottom.
## 3 Reasons Your Subscription App Ad Spend Gets Wasted
Most wasted budget in subscription app campaigns traces to one of three structural problems. Identifying which applies to your campaigns is the starting point.
### 1\. Optimizing for App Installs Instead of Subscriptions
The default conversion event for most app campaigns is the install. Ad platforms give you exactly what you optimize for, with no regard for whether those users ever subscribe. Across subscription app categories, CPPU (cost per paying user) runs **4-5x higher than CPI** (RevenueCat State of Subscription Apps 2025). If your Meta CPI is $5 on iOS, your realistic cost per subscriber is closer to $20-30 before accounting for trial drop-off.
When your optimization event is an install, the algorithm finds people who install apps, not people who subscribe. Shifting to trial starts or subscription starts initially reduces volume, but it changes the quality of the audience the platform targets. That distinction is where subscription apps start finding actual leverage.
### 2\. Spreading Budget Across Channels Without Channel-Level CPS Data
> "I want to reduce waste, but I'm worried the volume will drop along with it."
This objection stems from a single root cause: no clear picture of which campaigns are profitable at the subscription level.
Without knowing your cost per subscription (CPS) by channel, by audience, and by creative, you cannot cut with precision. You can only cut blindly, which does hurt volume. Running four or five channels without channel-level CPS doesn't run efficient tests — it amplifies losses.
### 3\. Creative Fatigue Drains Budget Even When CPI Looks Fine
Creative fatigue is one of the quieter causes of wasted spend. On TikTok, creative refresh is typically needed weekly as an operational rule of thumb, because audiences cycle through ad content quickly. On Meta, campaigns running against narrow audiences often see frequency spikes that push CPMs up and pull CTRs down within the first two weeks.
When creative fatigues, the algorithm does not pause. It continues spending your budget to reach increasingly poor-fit audiences in order to hit delivery targets. CPI may stay stable or even drop as the algorithm reaches lower-competition audiences. But trial starts and subscription conversions fall.
## 3 Key Metrics to Identify Wasted App Ad Spend
### 1\. Cost per Subscription (CPS): Your True App Acquisition Metric
CPS is your actual acquisition cost. Your ad platform reports CPI because that is what it can measure directly, without additional setup. But CPI tells you nothing about subscription intent or downstream revenue.
The calculation:
> **CPS = Total channel spend / Number of new subscribers from that channel**
Run this separately for each channel, audience segment, and creative. A Meta campaign generating 200 subscribers at $22 CPS against an annual plan LTV of $40 is profitable. The same CPS against a monthly plan with 17.5% 12-month retention is not, because those subscribers will churn before LTV recovers the acquisition cost (RevenueCat State of Subscription Apps 2025).
Context for calibration: CPPU across subscription categories runs roughly 4-5x CPI (RevenueCat State of Subscription Apps 2025). If your iOS Health and Fitness CPI is around **$5.78** (RevenueCat State of Subscription Apps 2025), expect a realistic CPS in the **$23-29 range** before factoring in your specific trial conversion rate.
### 2\. Trial-to-Paid Conversion Rate by Ad Channel
The **median trial-to-paid conversion across 75,000+ subscription apps is 34.8%** (RevenueCat State of Subscription Apps 2025). By category: Travel apps convert at **48.7%** median, Health and Fitness at **39.9%**, Gaming at **19%**, and Utilities around 25%. The channel breakdown within your own campaigns is more actionable than any industry benchmark.
Consider this comparison: a campaign generating 500 trials at 20% conversion produces 100 subscribers. A smaller campaign generating 200 trials at 45% conversion produces 90 subscribers, nearly the same output at 60% of the trial volume. That second campaign carries higher purchase intent and will scale more efficiently.
Trial length is a variable that data consistently shows matters independently of channel performance. Apps offering 17-32 day trials convert at a median of **45.7%**, compared to **26.8%** for shorter trials (RevenueCat State of Subscription Apps 2025). If you're seeing low trial-to-paid rates across channels, trial length and post-trial onboarding may be the culprit before channel targeting is.
### 3\. Revenue per Install (RPI): The Metric That Exposes Ad Channel Waste
RPI normalizes across CPI differences to show actual downstream revenue per install. When you compare RPI by channel alongside CPI by channel, waste becomes visible.
The **median D60 RPI across subscription apps is $0.31** (RevenueCat State of Subscription Apps 2025). Health and Fitness apps median at **$0.63**. North America installs median at **$0.57** D60, while installs from India and Southeast Asia run around **$0.10**. A lower CPI from a geo expansion may still produce worse RPI than a higher-cost North America campaign. The net result is worse subscription revenue per dollar spent.
Top-decile apps (P90) push the D35 download-to-paid rate to **8.5%**, a 4x gap above the 1.9% median driven by optimization decisions, not product category.
| Metric | Median (All Apps) | Q3 Threshold (Top 25%) |
| --- | --- | --- |
| Trial Start Rate | 6.2% | 12.4%+ |
| Trial to Paid | 34.8% | 51.5%+ |
| D35 Download to Paid | 1.9% | 4.3%+ |
| D60 Revenue per Install | $0.31 | $0.81+ |
_Data:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps/)_, analysis of 75,000+ subscription apps._
## How to Audit and Eliminate Your Worst-Performing App Ad Spend
Waste hides in three layers: channel, audience, creative. Audit in that order.
### 1\. Run a Trial Funnel Audit Across Your Ad Channels
Pull 30-60 days of data segmented by channel. For each channel, calculate install volume, trial start rate, trial-to-paid conversion, and CPS. Then lay them side by side:
| Channel | CPI | Trial Start Rate | Trial to Paid | CPS | Decision |
| --- | --- | --- | --- | --- | --- |
| Meta Advantage+ | $5.20 | 8.4% | 31% | $19.90 | Scale |
| Google UAC | $4.10 | 6.1% | 21% | $31.90 | Test or pause |
| TikTok | $2.80 | 4.9% | 18% | $31.70 | Refresh creative |
| Apple Search Ads | $8.50 | 9.2% | 44% | $20.90 | Scale |
_Example figures for illustration. Run the same audit against your own channel data._
The pattern is instructive: TikTok and Google UAC have nearly identical CPS ($31.70 vs $31.90), but the prescriptions differ. Google UAC's weak trial-to-paid rate (21%) signals a channel-level problem — the audience quality is off. TikTok's similar CPS comes from creative fatigue: refresh the assets before cutting the spend.
The industry-level confirmation: gaming apps have the **highest install-to-trial rate (12.2%)** of any subscription category, yet the **lowest trial-to-paid conversion (19%)** ([AppsFlyer State of Subscriptions for Marketers 2026](https://www.appsflyer.com/resources/reports/subscription-marketing/)). High trial volume can conceal low-quality acquisition.
### 2\. Segment Ad Audiences by Subscription Quality, Not Volume
Within each channel, look at audience-level breakdowns. Common waste drivers at the segment level:
**Age segments** with high trial volume but low conversion appear frequently. Younger audiences often trial apps freely but convert at lower rates due to lower spending capacity or intent. If Meta is spending heavily on an 18-24 segment that underperforms your 25-35 cohort by 15 percentage points on trial-to-paid, that is a segmentation correction, not a channel cut.
**Geographic concentration** into lower-CPI regions can look efficient on the install cost side while generating D60 RPI of **$0.10** versus **$0.57** from your North America base (RevenueCat State of Subscription Apps 2025). The aggregate CPS will quietly degrade without a visible signal in install or trial volume numbers.
**Lookalike audiences** built from install events, rather than subscription or annual subscriber events, perpetuate the same optimization problem at the audience layer. Rebuilding lookalikes from your paying subscriber list changes the seed audience the algorithm expands from and tends to produce higher downstream conversion even at a higher CPI.
### 3\. Creative Fatigue Drains Budget Even When CPI Looks Fine
Review frequency, CTR trend, and CPM trend together over 14-21 days per active creative. These are operational thresholds, not platform-published rules: frequency above 3.0 within a two-week window, CTR declining more than 20% week-over-week, or CPM rising more than 25% without a corresponding conversion improvement typically signals a fatigued creative.
When these signals appear, the creative is fatigued, not the campaign structure or the audience. Pausing the creative rather than the campaign preserves the algorithm's learned audience data. Replacing the creative refreshes delivery without resetting the learning phase.
## How to Cut App Ad Spend Without Losing Subscription Volume
With CPS data by channel, the reallocation is mechanical.
### 1\. Apply the 70/20/10 Budget Reallocation Framework
A practical reallocation framework, sometimes called 70/20/10, splits active budget into three buckets:
[Screenshot 2026-05-18 at 20.22.09.png]
When you cut a losing channel, that budget moves to the 70% bucket first.
> **Watch out:** when trimming a losing campaign, pause the creative first, not the whole campaign. Pausing the campaign resets the algorithm's learned audience data and forces a full rebuild from scratch.
### 2\. Scale Winning Campaigns at 20-30% Budget Increments
One of the most consistent budget mistakes is aggressive scaling when a campaign shows strong results. A campaign performing at $100/day gets jumped to $500/day, the algorithm re-enters a learning phase, and efficiency degrades for 7-10 days. By the time performance recovers, the cohort data has been distorted.
The standard agency convention: increase budgets on winning campaigns by **20-30% at a time**, and wait at least five to seven days before the next increment. This preserves the audience signal the algorithm has built while gradually expanding reach.> **Watch out:** don't cut during the learning phase. Meta's learning phase exits at **50 conversions per week per ad set** ([Meta Business Help Center](https://www.facebook.com/business/help/112167992830700)). Other platforms have similar thresholds. If CPS is not improving after three weeks at that rate, then cut. Not before.
### 3\. Feed Ad Platforms Subscription Events, Not Just Install Signals
The highest-leverage change for most subscription apps: stop sending installs to the algorithm, start sending subscription events.
When Meta or Google receives subscription start events, it recalibrates targeting toward users with higher purchase intent. The transition temporarily reduces trial volume. But the users the algorithm finds after recalibrating convert at higher rates, which reduces CPS even as CPI may rise.
A practical middle step: keep trial starts as the primary optimization event but simultaneously send subscription start events and subscription revenue data back to the platform as value signals. Meta Conversions API, Google Enhanced Conversions, and TikTok Events API all support this. This builds a richer audience model without fully switching optimization events and disrupting delivery volume.
## Stop Wasting App Ad Spend: Measure First, Then Cut
The cut-without-shrink play: measure CPS, trial-to-paid, and RPI by channel. Reallocate on a 70/20/10 split. Feed subscription events back to your ad platforms. Volume doesn't drop. It shifts toward users who actually pay.
---
## How to See Which Marketing Channel Brings Paying Subscribers
URL: https://www.airbridge.io/en/blog/how-to-see-which-marketing-channel-brings-paying-subscribers
Language: English
Published: 2026-05-18
Author: Harper (Trang Nguyen)
You're running campaigns on Meta and Google. Installs are coming in. Some users start trials. But when you look at your ad dashboard, you see cost per install (CPI), not paying subscribers. You have no idea which marketing channel is actually making you money.
This is one of the most common data problems in subscription app marketing. **CPI tells you nothing about revenue.** Among subscription apps analyzed across 75,000+ apps, the cost per paying user (CPPU) runs **4 to 5 times** higher than CPI. Most of what you see in your ad dashboard is the wrong number to optimize for.
## Key Takeaways
- **Ad platforms report installs. Subscriptions happen server-side, days or weeks later.** The gap between these two events is where most teams lose visibility.
- **CPI is a vanity metric for subscription apps.** The number that matters is CPPU: how much it costs you to acquire one paying subscriber per channel.
- **Each channel has a different funnel shape.** A channel with cheap installs may produce almost no paying subscribers, while a more expensive channel often converts at a significantly higher rate.
- **Connecting your billing data to your acquisition data is a one-time technical setup** that unlocks permanent channel-level subscription reporting.
- **D30 return on ad spend (ROAS) and channel-level lifetime value (LTV) are the two metrics** that tell you which channels to scale and which to cut.
## Why Your Ad Dashboard Shows Installs but Not Paying Subscribers
Every major ad platform (Meta, Google, Apple Search Ads, TikTok) tracks the moment a user taps your ad, attributes the install to your campaign, and reports it back to you in real time.
The problem is what happens next. When a user starts a free trial, upgrades to a paid subscription, or renews, those events happen inside Apple's or Google's billing infrastructure. They happen completely outside your ad platform's view. A subscription confirmation is a server-side transaction. The ad network cannot see it unless you send that data back explicitly.
[diagram-subscription-funnel.svg]
### 1\. The billing gap between ad platforms and billing platforms
Your ad platform stops tracking the moment the install is attributed. Your subscription platform (RevenueCat, Adapty) starts tracking when billing events occur. These two systems do not talk to each other automatically.
The result: your Meta dashboard shows **1,000 installs**. Your RevenueCat dashboard shows **200** new subscribers. You have no idea which of the 1,000 installs became those 200 subscribers, or which campaign drove them.
### 2\. The source-of-truth conflict across tools
Even when you do connect data sources, Meta, Google, your subscription platform, and your analytics tool will each report different conversion numbers. This is not a bug. It is a structural feature of how mobile attribution works. Here is why each tool produces a different count:
- **Meta** attributes a subscription if it occurred within its attribution window: for example, up to **7 days** after a click.
- **Google** applies its own, separate attribution window, often producing a different count for the same user action.
- **Billing platforms** (RevenueCat, Adapty) use first-party transaction data: the most accurate and complete view of what actually happened.
- **Ad networks** rely on probabilistic or deterministic matching, which can over- or under-count depending on signal quality.
This creates what practitioners call a "source of truth conflict" across ad networks, billing platforms, and app stores. Your job as a marketer is to pick one source of truth for decision-making and build your reporting around it. The common practice is to use your billing platform as the revenue source of truth and your attribution tool as the channel-attribution source of truth.
### 3\. CPI vs CPPU: the metric that determines profitability
CPI measures install cost. CPPU measures the cost to acquire a paying subscriber.
For subscription apps, CPPU is the number that determines whether a channel is profitable. CPPU typically runs **4 to 5 times** higher than CPI. At that 4–5x multiplier, a $4 CPI channel lands at roughly a $20 CPPU. The cheaper channel by installs is often the more expensive channel by paying subscribers.
If your current reporting only shows CPI, you are optimizing for the wrong metric.
## Subscription Funnel Benchmarks by Marketing Channel: Install-to-Paid Conversion Rates
Every acquisition channel has a different conversion profile through the subscription funnel. The install-to-trial rate, the trial-to-paid rate, and revenue per install vary significantly depending on where users came from.
Here is what benchmark data across billions of subscription app installs shows:
| Category | Install to Trial | Trial to Paid | D60 Revenue / Install (Median) |
| --- | --- | --- | --- |
| Health & Fitness | Medium (71% no-trial) | ~39.9% | 0.63 |
| Education | Medium | ~42% | 0.40 |
| Lifestyle | 3.9% (lowest) | 41% | Varies |
| Gaming | 12.2% (highest) | 19% (lowest) | 0.13 |
Gaming is the clearest example of the "cheap install, no subscriber" trap. The install-to-trial rate is the highest of any category at **12.2%**, making campaigns look efficient on an install basis. But only **19%** of those trialists ever convert to paid. Users install, explore, and leave without subscribing.
Education shows the opposite pattern: moderate install-to-trial rates, but once a user starts a trial, **42%** convert to paid. A user who starts a trial in an education app has already made a decision to commit.
### The "cheap install, no subscriber" trap in channel terms
This dynamic plays out at the channel level too. TikTok historically delivers lower CPI than Meta or Apple Search Ads, but top-of-funnel discovery traffic carries lower purchase intent. A user who finds your app through a TikTok video may install out of curiosity. A user who searches for your app on Apple Search Ads already knows what they want.
## How to Connect Billing Data to Acquisition Channels (Step-by-Step)
This is a one-time technical setup.
### 1\. What data you need
You need three data streams connected:
- **Install attribution data**: which user came from which channel, campaign, and ad (from your attribution tool)
- **Subscription event data**: Start Trial, Subscribe, Renew, Cancel (from your billing platform such as RevenueCat or Adapty)
- **A connection between them**: so subscription events carry the original acquisition source for each user
Without all three, you can measure each piece but not connect them.
### 2\. Connect your billing platform via server-to-server
The most reliable way to send subscription events back to your attribution layer is through a [server-to-server (S2S)](https://www.airbridge.io/en/glossary/server-to-server-s2s) connection, not just an in-app SDK.
SDK-based tracking detects subscription state changes when a user opens the app. If a user subscribes after closing the app (for example, when Apple processes a deferred payment), the SDK may not capture that event.
S2S integration sends billing events directly from your backend to your attribution tool in real time, regardless of whether the user has the app open. For subscription apps, this distinction matters: renewal events, failed payments, and resubscriptions often happen while the app is closed. Most major billing platforms have built-in S2S integrations with attribution tools.
### 3\. Define your events consistently
Once the connection is live, you need consistent event naming across all your tools.
Set a single naming convention and apply it everywhere:
- `start_trial`: when a user begins a free trial
- `subscribe`: when a user converts to paid
- `renew`: when a subscription auto-renews
- `cancel`: when a user cancels
---
## 3 Metrics That Reveal Which Marketing Channel Is Driving Paid Subscribers
Once your data is connected, here is what to measure.
### 1\. Cost per paying user (CPPU)
CPPU = total ad spend for a channel divided by the number of paying subscribers acquired from that channel.
Read more about [Cost per Trial and Cost per Subscription](https://www.airbridge.io/en/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics): two companion metrics that break down where exactly your funnel cost is accumulating.
### 2\. D30 ROAS
D30 [ROAS](https://www.airbridge.io/en/glossary/return-on-ad-spend-roas) measures how much subscription revenue a channel generated within **30 days** of an install, relative to what you spent acquiring those users.
For subscription apps targeting a 12-month payback period, D30 ROAS will be well below **100%**. A D30 ROAS of **1.5x** or more is a strong positive signal. A D30 ROAS below **0.5x** often indicates a channel or creative that is not producing users with purchase intent, and may need restructuring before scaling.
### 3\. Channel-level LTV
The most important long-term metric is [lifetime value (LTV)](https://www.airbridge.io/en/glossary/lifetime-value-ltv) per paying subscriber by channel. Health and Fitness apps see a median D60 revenue per install of **63 cents** across all channels. But that varies by acquisition source. Users acquired through Apple Search Ads (high intent, actively searching) consistently show higher LTV and retention rates than users from broad awareness campaigns, reflecting the intent difference between search-based and discovery-based channels.
Tracking LTV by channel tells you not just which channel delivers paying subscribers, but which one delivers paying subscribers who stay.
| Metric | What It Tells You | When to Use |
| --- | --- | --- |
| CPPU | Cost efficiency per paying subscriber | Channel comparison, daily optimization |
| D30 ROAS | Early revenue signal | Weekly campaign review |
| Channel LTV (D60/D90) | Long-term subscriber quality | Monthly budget allocation |
## Meta vs. Google vs. ASA vs. TikTok: Subscription App Performance by Channel
Each of the four major paid channels has a distinct profile for subscription app marketing.
### 1\. Meta (Facebook and Instagram)
Meta is the highest-scale marketing channel available for most subscription app marketers. Its strength is audience targeting: you can reach users by interest, behavior, and lookalike audiences built from your existing subscribers.
Meta campaigns optimized for subscription events (not just installs) tend to find users more likely to convert. Key things to know before running Meta for subscriptions:
- **If your weekly Subscribe volume can't hit ~50 per ad set** (common for early-stage apps), optimize for Start Trial instead and track downstream conversion separately.
- **Start with Advantage+ campaigns and broad targeting** for the first **4 to 6 weeks**. Meta's algorithm needs ~50 optimization events per week per ad set to exit the learning phase.
- **Meta's reporting is self-attributed**: its conversion count will differ from your attribution tool. Pick one source of truth and stick to it.
### 2\. Apple Search Ads (ASA)
Apple Search Ads reaches users who are actively searching for apps like yours in the App Store. This intent-based targeting typically produces higher trial-to-paid conversion rates than discovery channels. What to expect from ASA:
- **Higher purchase intent** than Meta or TikTok. Users are already searching, not just browsing.
- **Smaller audience and higher CPIs** than Meta or Google, but conversion rates often compensate.
- **Works well as an intent-capture layer** on top of Meta's scale for iOS-first subscription apps.
- **Good for establishing LTV baselines**: many teams start with ASA to set benchmarks, then use those benchmarks to guide Meta campaign targets.
### 3\. Google App Campaigns
Google App Campaigns run across Google Search, YouTube, Google Play, and the Display Network. Unlike Meta, you cannot manually segment placements. The algorithm allocates budget automatically. What drives performance on Google:
- **Creative variety matters most**: provide video, static, and copy variations to give the algorithm more to learn from.
- **Optimize for subscription events, not installs**: Google's algorithm is strongest when it has clear revenue signals.
- **No manual placement control**: trust the algorithm to allocate across Search, YouTube, Play, and Display based on conversion data.
### 4\. TikTok
TikTok works best as a top-of-funnel discovery channel for subscription apps. Its CPI tends to be lower than Meta or ASA, but the users it attracts are often in exploration mode rather than purchase mode. What to know before running TikTok:
- **Creative fatigue hits fast**: at typical spend levels, refreshing hooks and opening variations every **7 days** maintains performance.
- **Lower purchase intent than search-based channels**: expect higher install volume but lower trial-to-paid rates versus ASA.
- **Validate on Meta first** before expanding to TikTok. Add a new channel only after your primary channel is producing consistent CPPU results you can explain and repeat.
- **Use TikTok for awareness, not conversion**: pair it with retargeting or ASA to recapture high-intent users further down the funnel.
## The Attribution Setup That Connects Ad Spend to Subscription Revenue
If you are running paid campaigns on Meta, Google, Apple Search Ads, or TikTok and want to see which channels and creatives drive paid subscribers, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your subscription revenue events from RevenueCat or Adapty directly to your acquisition channels. It covers Meta, Google, Apple Search Ads, and TikTok, with channel-level funnel reports and LTV data out of the box. Core Plan supports up to **2** billing platform integrations (RevenueCat and Adapty are both included), with **15,000** free attributed installs and no annual contract.
## Stop Optimizing for Installs. Start Measuring What Pays.
Most subscription app marketers know their CPI. Few know their CPPU. Even fewer can tell you which specific marketing channel produced their highest-LTV subscribers.
The gap is not a lack of effort. It is a missing data connection. Close that connection, watch CPPU and channel LTV instead of CPI, and budget decisions stop being guesses.
**Want to go deeper on subscription app marketing?** [**Explore more guides**](https://www.airbridge.io/en/blog)**.**
---
## Why Your Subscription App Gets Installs but Not Paid Subscribers
URL: https://www.airbridge.io/en/blog/why-subscription-app-get-installs-no-paid-subscribers
Language: English
Published: 2026-05-18
Author: Luke
Your download count is climbing. Ads are running. Users are installing. But your app install growth is not translating into subscription revenue.
This is one of the most common growth problems in subscription apps, and one of the most misdiagnosed. The question of why app users do not convert to paid subscribers has a structural answer: it lives inside the funnel, not in the ad budget. Most teams respond by buying more traffic. The actual problem is structural, living inside the funnel itself. The median subscription app converts just **1.9% of downloads into paying subscribers within 35 days**, and top-performing apps hit **8.5%** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). That 4.5x gap is not explained by ad spend or product quality. It is explained by funnel discipline: knowing exactly where users drop, why they drop, and running focused experiments to fix each stage.
**Key Takeaways**
- Install-to-trial below 6.2%? The problem is your Day 0 onboarding and paywall, not traffic volume.
- Trial-to-paid below 34.8%? Trial length and the core activation action move the number more than paywall copy or pricing.
- **82% of trial conversions happen on Day 0.** Optimizing later in the funnel without fixing the first session produces marginal results.
- **37.2% of cancellations cite "not enough usage."** Fixing trial engagement addresses both the conversion and the retention problem simultaneously.
- The difference between **1.9%** median and **8.5%** P90 install-to-paid is not product quality. It is structural funnel discipline that compounds with every install.
## What Is the Subscription App Conversion Funnel? 3 Stages, 3 Leak Points
Before diagnosing what is broken, it helps to have a shared model of what a healthy funnel looks like.
Every subscription app runs on the same three-stage funnel: install to trial start, trial start to paid subscriber, and install to paid subscriber as the combined output. Each stage leaks differently, responds to different fixes, and requires different data to diagnose.
### 1\. The three-stage model every subscription app shares
**1) Install to trial**
The user downloads the app, goes through onboarding, hits the paywall, and decides whether to start a free trial. This is the first real commitment ask. Most apps lose the majority of their subscription potential here, not during the trial, but before it starts.
**2) Trial to paid**
The user begins a trial and either converts to a paying subscription when the trial ends or cancels. This is where product value needs to show up clearly and fast enough to justify payment.
**3) Install to paid**
The product of both stages above: the percentage of app installs that become paying subscribers. This is your actual acquisition conversion rate, the number that should sit alongside cost per install (CPI) in any channel performance review.
Each stage has its own failure mode. Treating all three as one undifferentiated "conversion problem" is one of the most expensive mistakes subscription app teams make.
### 2\. Subscription app conversion benchmarks: where the real drops happen
| Metric | Median | High Performer |
| --- | --- | --- |
| Install to trial | 6.2% | 20.3% (P90) |
| Trial to paid | 34.8% | 51.5%+ (top quartile) |
| Install to paid (D35) | 1.9% | 8.5% (P90) |
_Sources:_ [_Adapty State of In-App Subscriptions 2026_](https://adapty.io/state-of-in-app-subscriptions/) _(16,000+ apps, $3B subscription revenue) and_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps-2025/) _(75,000+ apps, $10B+ revenue)._
Reading these numbers carefully: at the median, **94% of app installs never start a trial**. Of the 6.2% who do start a trial, 65% still do not subscribe. The funnel leaks at both stages, and fixing only one produces incremental improvement at best.
The P90 benchmark tells you the ceiling. For an app receiving **10,000 installs** per month, the difference between median and P90 performance is **190 paying subscribers vs. 850**, every month. The compounding effect on [lifetime value (LTV)](https://www.airbridge.io/en/glossary/lifetime-value-ltv) is why top apps invest in funnel optimization before scaling paid acquisition.
[Screenshot 2026-05-18 at 15.33.48.png]
## Kill Zone 1: Why 94% of App Installs Never Start a Free Trial
The install-to-trial stage is where most apps lose the most users, and where most teams are not looking.
**82% of all trial starts happen on Day 0** (SOSA 2025), during the same session as the install. By Day 2, the probability that any given user will ever start a trial drops close to zero. You do not have days or weeks to convert an app install into a trial. You have one session.
### 1\. Four reasons 94% of installs never reach a trial
**1) Weak value communication**
Users hit the paywall before they understand what they are paying for. The onboarding showed features rather than outcomes. A fitness app that demonstrates its library before showing a user their personalized goal misses the most persuasive moment in the entire funnel.
**2) Too many screens before the paywall**
Every extra screen between install and paywall that does not build toward the value promise is a drop-off point. Apps that streamline onboarding to the personalization data they actually need consistently outperform those that collect extensive information before delivering value.
**3) Mismatched user intent**
The ad creative promised one thing and the app delivered another. A user who clicked a "lose weight fast" ad and arrived at a general wellness app did not get what they were looking for. This download never had a realistic conversion probability.
**4) Wrong paywall placement**
Some apps place the paywall before users understand value. Others place it so late that users have already decided the app is not for them. Paywall timing is one of the highest-leverage experiments any subscription app can run.
### 2\. How to Fix It: Personalize the Paywall and Shorten the Path to Trial
Apps with the highest trial start rates share a consistent pattern: the onboarding collects the user's specific goal, and the paywall reflects that goal directly. Flo Health, for example, collects health goals during onboarding and surfaces a personalized plan on the paywall, converting a generic subscription screen into a specific health commitment.
**1) Shorten onboarding to the minimum needed to personalize the paywall**
Count the screens between install and paywall. Every screen that does not directly contribute to the value promise on the paywall is a drop-off point. This is not a "fewer screens is always better" rule: Noom uses 100+ onboarding screens to build an exceptionally personalized plan and converts well because of it. The goal is to eliminate screens that do not contribute to the value promise, not to minimize total count.
**2) Replace generic CTA copy with the user's stated goal**
"Start your 7-day free trial" converts worse than "Start your plan to \[specific goal the user just stated\]." Apps that A/B test this approach consistently see a measurable lift in trial starts because it connects the cost to a concrete outcome the user already said they want.
**3) Audit your ad-to-app message match**
The top 5 creatives driving your highest install volume should make a promise that matches the first screen users see after install. Any gap means users arrive with misaligned expectations and a conversion probability close to zero. Fixing message match before scaling spend is one of the highest-ROI actions at this stage.
**4) Test a hard paywall if you currently run freemium**
Hard paywalls convert at a median of **12.1% install-to-paid**, compared to **2.2% for freemium**, a **5.5x** difference driven by purchase intent filtering (SOSA 2025). Hard paywalls work when the value promise is strong and the onboarding has earned the ask. Explore our [subscription app paywall guide](https://www.airbridge.io/en/blog/paywall-optimization) for specific A/B tests to run at this stage.
## Kill Zone 2: Why Two-Thirds of Trial Users Never Become Paid Subscribers
If a user has started a trial, they have passed the hardest gate. But the median trial-to-paid rate is 34.8%. Roughly **two out of three trial starters still do not subscribe**. Most teams try to fix this with push notification sequences. Those help at the margin. The bigger levers are trial length and early activation.
### 1\. Why Two Out of Three Trial Users Still Do Not Subscribe
Two root causes account for most of the gap between trial starts and paid conversions.
**1) Trial length cuts the window short before habits can form**
Apps with 17-32 day trials convert at a median of **45.7%** (Adapty 2026). Apps with 3-7 day trials convert at **26.8%**. A short trial ends before most users have built a routine around the app. They hit the paywall with low usage, no formed habit, and no compelling reason to pay.
**2) Users who do not engage during the trial do not convert after it**
Among Google Play subscribers who cancel, **"not enough usage" is cited by 37.2%**, more than "cost related" at **34.6%** (SOSA 2025). The same pattern applies to non-converters: a user who opens the app twice during a 7-day trial has not experienced the value that justifies a subscription. No paywall copy or pricing change fixes that.
Both problems share the same root: the trial ends before the user has experienced enough of the app to make payment feel like a natural next step.
### 2\. How to Fix It: Extend the Trial and Drive the Core Activation Action
**1) Extend your trial if it is 7 days or shorter**
Test 14 days before changing anything else. The **19-percentage-point** conversion gap between long and short trials exists because longer trials give users time to integrate the app into their behavior. By the time the paywall arrives, they are renewing a habit, not committing to something new. If your [trial-to-paid conversion rate](https://www.airbridge.io/en/blog/subscription-app-onboarding) is below 30% and your trial is short, this is the experiment to run first.
**2) Find the action that predicts conversion**
Analyze what paying subscribers did during their trial that non-converters did not. For a journaling app it might be completing three entries. For a fitness app, finishing the first full workout. For Duolingo, it is a 5-day streak. This is your core activation action, and it is worth more attention than any paywall copy test.
**3) Build Day 0 and Day 1 onboarding to reach that action before Day 3**
Users who complete the activation action on Day 1 convert at materially higher rates than those who do not. Every early session interaction should move users toward that moment, not around it.
**4) Add a behavioral nudge for users who miss it**
If a user has not completed the activation action by Day 2, they are unlikely to convert without intervention. A push notification or in-app prompt that offers to help them complete the action, rather than just counting down the trial, will typically outperform urgency messaging.
## Kill Zone 3: How Channel Mix Silently Kills Subscription Conversion
This is the kill zone that is invisible without the right data, and the one that causes the most wasted ad spend.
Your install-to-subscriber rate is an average. It mixes together high-intent users from Apple Search Ads keyword targeting, broad-audience users from Meta prospecting, and organic users who found the app on their own. That average can look acceptable while hiding a serious structural problem: some channels are producing subscribers at 5-6% conversion and others at under 1%. If you are scaling the low-converting channel because the cost per install is lower, you are spending more money to produce fewer subscribers.
### 1\. Why your cheapest app install source might be your worst subscriber source
For example: a Meta broad-audience campaign delivers **5,000 installs** at $2 CPI. An Apple Search Ads campaign targeting high-intent searches delivers **800 installs** at $7 CPI. On cost per install alone, Meta appears 3.5x more efficient.
But if Meta installs convert to paid subscribers at 1.5% and ASA installs convert at 5.5%, the cost per subscriber from Meta is **$133**. From ASA, it is **$127**. The "cheaper" channel is producing the more expensive result. Scale that signal long enough and you can spend 50% more to produce 30% fewer paying subscribers.
The question that matters is not "which channel gives me the most installs?" It is "which channel gives me subscribers?" These are different questions with different answers, and you cannot answer the second without connecting ad spend data to subscription revenue data at the channel level.
### 2\. How to Fix It: Replace CPI with Cost per Subscriber as Your Channel Metric
Fixing this kill zone requires replacing the metric you are optimizing for.
**1) Connect three data streams**
Install source by channel and campaign, subscription events (trial start, conversion, cancellation) tied to the original install, and cost data from each platform. When these are joined, you can compute install-to-trial rate by channel, trial-to-paid rate by channel, and cost per subscriber by channel.
**2) Rank channels by cost per subscriber, not CPI**
Cost per install is a proxy. Cost per subscriber is ad spend divided by subscribers attributed to that channel. The rankings often flip when you switch metrics. A $2 CPI channel converting at 1.5% to paid has a higher cost per subscriber than a $7 CPI channel converting at 5.5%. The blended average hides this entirely.
**3) Reallocate toward your lowest CPSub channels**
Once you have the ranking, shift spend from high-CPSub channels toward low-CPSub channels, even where the CPI is higher. This reallocation alone can meaningfully reduce blended cost per subscriber without changing a single creative or bid strategy.
## The Fix Starts with Knowing Which Stage Is Leaking
App installs measure interest. Subscribers measure delivered value. If the gap between them is growing, the answer is not more ad spend. It is identifying which of the three kill zones is losing the most revenue and running a focused experiment there.
Measure your install-to-trial rate against the 6.2% median. Measure your trial-to-paid rate against 34.8%. Break your install-to-paid rate out by channel. The largest absolute gap points to the highest-value experiment. Most subscription apps that close the install-to-subscriber gap do not do it by finding a new channel or shipping a new feature. They do it by stopping leaks they did not know existed.
---
## Subscription App Funnel Not Converting? How to Find the Bottleneck Before You Fix Anything
URL: https://www.airbridge.io/en/blog/how-to-fix-subscription-app-funnel-not-converting
Language: English
Published: 2026-05-18
Author: Luke
You updated the creative. Installs improved. Trials barely moved. So you rebuilt the onboarding flow. Trials climbed slightly. Trial-to-paid stayed flat. Three months later, the subscription app funnel looks almost identical to where it started.
**The issue was not any one of those three layers.** The issue was that you optimized before you diagnosed. When you change creative, onboarding, and paywall simultaneously, every experiment adds noise and nothing compounds. Identifying the subscription app funnel bottleneck, and fixing only that layer, is what makes every subsequent experiment count.
**Key Takeaways**
- **Fixing the wrong layer costs more than fixing nothing.** Each simultaneous change resets your baseline and destroys your ability to learn what moved the needle.
- **82% of trial decisions happen on the same day as install** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). If users leave during onboarding, no paywall test will ever reach them.
- **Trial start rate below 6.2% means onboarding is the bottleneck.** That is the median across 75,000+ subscription apps analyzed.
- **Trial-to-paid below 34.8% means the paywall is the bottleneck.** Better creative and stronger onboarding will not fix it.
- **Prioritization is the highest-leverage skill in funnel optimization.** Fewer experiments on the right layer beat more experiments spread across all three.
[Screenshot 2026-05-18 at 15.28.42.png]
## Why Changing Creative, Onboarding, and Paywall at the Same Time Backfires
### 1\. How running parallel tests destroys your ability to learn
Every change in your funnel resets your baseline. When you change creative, onboarding, and the paywall in the same sprint, you cannot determine which change moved which metric. You run more experiments to untangle the noise. You fall further behind while burning through budget and engineering time.
**Good funnel work is sequential.** Fix the loudest leak first. Measure cleanly. Then move to the next layer.
### 2\. Why most subscription app teams diagnose the wrong bottleneck first
Most early-stage teams treat creative as the primary suspect because it is the most visible part of the funnel. Ad spend is measurable. Creative output is tangible.
**But creative is only the bottleneck when it is limiting your install volume.** If you are already getting enough installs to observe downstream behavior, creative is not your problem. The question becomes whether those installs convert into trials, and whether trials convert into paying subscribers. Those are onboarding and paywall problems, and they live in activation data that most teams never look at first.
## How to Measure Your Subscription App Funnel Conversion Rates
**Break your funnel into three rates. Each points to a different fix.** The rates that matter are install-to-trial, trial-to-paid, and the timing of trial starts.
### 1\. The three subscription funnel conversion rates that reveal where users drop
Compare your current numbers against these benchmarks, drawn from analysis of 75,000+ subscription apps ([RevenueCat SOSA 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)):
| Metric | Median | Top 10% (P90) | Upper Quartile |
| --- | --- | --- | --- |
| Trial Start Rate (install to trial) | 6.2% | 20.3% | 12.4%+ |
| Trial-to-Paid Rate | 34.8% | 68.2% | 51.5%+ |
| Download-to-Paid (Day 35) | 1.9% | 8.5% | 4.3%+ |
### 2\. How to read your conversion gaps and decide where to focus
If your trial start rate falls below 6.2%, users are installing and leaving before they reach your paywall. **This is an onboarding problem.** Paywall improvements will not reach these users because they never saw it.
If trial start rate is above 6.2% but trial-to-paid sits below 34.8%, users are hitting your paywall but not converting. **This is a paywall problem.** Creative and onboarding changes will not fix it.
If both rates are within or above benchmark range but total revenue remains flat, the constraint is install volume. **This is where creative work matters.** You need more qualified installs entering the funnel to drive meaningful subscription growth.
## How to Tell If Low Install Volume Is Killing Your App Funnel
Creative is the top-of-funnel input. Its job is to drive installs from users likely to start a trial and eventually pay. But it is also the layer teams optimize first by default, often before knowing whether the installs they already have are converting downstream.
**The risk of fixing creative when onboarding is the real problem:** A documented case study showed a client cutting cost per install by over **80%** through improved targeting and creative. Installs went up. But trial conversion did not move because the onboarding flow was leaking users before they ever reached the paywall. The cheaper traffic flowed in and flowed straight back out ([RevenueCat blog, 2025](https://www.revenuecat.com/blog/growth/fix-onboarding-funnels/)).
### 1\. Three signals that identify creative as your bottleneck
Three signals point clearly to creative as the primary constraint:
1. **Monthly install volume is under 1,000.** A common heuristic: below this threshold you lack sufficient data to detect meaningful conversion patterns.
2. **Trial start rate is at or above benchmark, even with low absolute numbers.**
3. **Installs from different creative sets show similar downstream behavior, regardless of which ad drove them.**
**If you are getting 300 installs a month but an 8% trial start rate, you do not have an onboarding problem.** You have a volume problem. Fix the creative.
### 2\. How to measure creative quality beyond cost per install
Focus tests on the hook (the first three seconds) and problem framing, not production quality. In split tests, how you frame the problem drives trial starts more than polish level. Track trial start rate per creative cohort, not just install cost, or you risk scaling the wrong campaigns.
**The metric that matters from creative is downstream conversion quality, not just install cost.** A campaign with a higher cost per install but a **10%** trial start rate is more valuable than a cheap campaign that drives users who install and leave immediately. Track [cohort LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) per creative from day one.
## Low Trial Start Rate? How to Fix Your App Onboarding First
### 1\. Why onboarding is a Day 0 decision
Across all subscription apps analyzed, **82% of trial starts happen on the same session as install** (RevenueCat SOSA 2025). Users flow through onboarding, reach the paywall, and make a decision. They rarely come back the next day.
This means onboarding is not a warmup act. It is the pitch. Every screen before your paywall either builds intent or drains it. If users exit before reaching the paywall, no optimization further down the funnel will recover them. Fix the funnel before building lifecycle campaigns. Lifecycle marketing has a limited ceiling when the activation rate is broken.
**What top subscription apps do differently:** Flo Health, the period-tracking app with over 420 million users, reaches roughly one in five US women of reproductive age. A key lever is a commitment screen placed before the paywall: a moment where users affirm their intent ("I'm ready") before seeing any pricing. Flo runs hundreds of concurrent A/B tests on onboarding alone, not to add features, but to refine how users feel in those first two minutes. The result is one of the highest conversion rates in the health app category ([RevenueCat blog, 2025](https://www.revenuecat.com/blog/growth/fix-onboarding-funnels/)).
A similar pattern holds at smaller scale. QUITTR, a habit-breaking app, found that longer onboarding increased trial conversion, not reduced it. Founder Alex Slater observed that the more time users invested in onboarding, the more likely they were to convert, because they wanted a return on the time they had already spent. QUITTR reached **$250K in monthly recurring revenue within four months** of launch ([Startup Spells](https://startupspells.com/p/porn-addiction-app-quittr-250k-mrr-4-months)).
### 2\. Three onboarding changes that increase trial start rate
The fastest way to locate your onboarding leak is to map screen-by-screen completion rates. Where does the drop happen? That screen is the problem, not the paywall.
1. **Add a commitment screen before the paywall.** A single screen asking users to affirm intent ("I'm committing to \[goal\]") creates self-identification before the payment ask. QUITTR's onboarding reported a **99% completion rate** (Startup Spells): most users who start the flow, finish it.
2. **Move the paywall to the right moment.** Not before users have experienced the core value promise, and not buried behind 15 steps. The paywall should appear right after users complete a goal-setting action that makes the value feel personal and earned.
3. **Personalize the paywall headline using the user's stated goal.** A headline that reflects what the user just told you, such as "Build your \[goal\] habit in 21 days", outperforms generic feature-benefit copy because it confirms the app understood their specific goal.
## Low Trial-to-Paid Rate? How to Fix Your App Paywall Before Anything Else
### 1\. The trial-to-paid rate benchmark that reveals paywall underperformance
As shown in the benchmark table above, the median trial-to-paid rate is **34.8%**, with upper-quartile apps hitting **51.5%** or better. If your trial start rate is at or above benchmark but trial-to-paid sits below that median, your paywall is the specific constraint.
Data across 16,000+ apps shows that teams running structured [paywall experiments](https://www.airbridge.io/en/blog/paywall-optimization) achieve up to 40x more revenue than teams that do not experiment ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)). And the highest-impact experiment type is not visual design: **changing plan structure, trial duration, or pricing has a higher probability of LTV improvement** than copy or image changes alone (Adapty 2026).
### 2\. Three paywall experiments to run before you redesign
Before testing anything, define a single objective for the paywall: trial start rate, paid conversion rate, or average revenue per user. Without one objective, every test result will be contested.
1. **Test a longer trial duration.** Trials of 17 to 32 days convert at a median of **45.7%**, compared to **26.8%** for trials of 3 to 7 days (RevenueCat SOSA 2025). If you are running a 3-day trial, test a 7-day or 14-day version before touching the design.
2. **Add a decoy plan.** A mid-tier option without a trial, such as a 3-month plan, makes the annual plan appear more attractive by comparison. Adapty's 2026 data shows number-of-plans experiments win on LTV **57.1%** of the time, the highest win rate of any paywall experiment type.
3. **Build a second offer.** Most users do not convert on their first paywall view. A second offer at a lower price point or shorter commitment captures users who were interested but not ready for the primary plan.
## Start With Your Subscription App Funnel Data, Then Fix One Layer
The mistake is not running experiments. The mistake is running experiments without knowing what you are trying to fix.
**Pull your three conversion rates. Compare them to benchmark. Fix the layer furthest below median first.** That single prioritization step compounds more than any combination of creative refreshes, onboarding redesigns, and paywall tests run simultaneously.
---
## What Is an App Activation Metric? Definition, Benchmarks, and How to Define Yours
URL: https://www.airbridge.io/en/blog/what-is-an-app-activation-metric
Language: English
Published: 2026-05-14
Author: Luke
> **An activation metric is the specific event that best predicts whether a user will become a paying subscriber and stay one.** It is not a signup, not an onboarding completion, and not a session count. It is the action that, when a user completes it, statistically separates future subscribers from future churners.
---
You had 10,000 downloads last month. A few hundred started trials. A fraction subscribed.
That gap between installs and revenue does not live in your ad account. It lives in the space between a user opening your app and experiencing something worth paying for. Most early-stage subscription apps are unable to define key metrics between downloads and actual activation, so they optimize the wrong signals and wonder why conversion stays flat.
The fix is not more traffic. It is knowing exactly which event inside your app separates users who will subscribe from users who will leave.
**Key Takeaways**
- **Tracking the wrong event is worse than tracking nothing.** Optimizing for onboarding completion or session count creates false confidence while the actual activation gap widens undetected.
- **First value and core value are not the same thing.** Most apps focus entirely on getting users to their first valuable moment. The habit that follows, which predicts renewal, is left unmeasured.
- **Top-performing apps are 4.5x better at conversion, not 4.5x better at acquisition.** The P90 app reaches 8.5% download-to-paid by Day 35. The median earns 1.9%. The gap is largely explained by what happens in the first few sessions.
- **Category benchmarks diverge sharply.** Travel app trialists convert at 48.7%, while the all-category median sits at 34.8%. Your activation metric must be defined for your category and your users, not copied from another app's playbook.
- **The 2x test separates real metrics from vanity metrics.** If users who complete your activation event do not retain at least twice as well as those who do not, that event is not your activation metric.
## The Download-to-Subscriber Gap Is Mostly an Activation Problem
When conversion is low, the instinct is to buy more installs or test new creatives. More traffic fed into a broken activation funnel produces proportionally more churn.
The real problem is earlier. Across [75,000+ subscription apps](https://www.revenuecat.com/state-of-subscription-apps-2025/), the median download-to-paid conversion rate by Day 35 is **1.9%**. Top-performing apps (P90) reach **8.5%**: a 4.5x gap largely explained by what happens in the first few sessions, not by ad spend.
Timing data confirms how decisive those early sessions are. [**82% of subscription trials start on Day 0**](https://www.revenuecat.com/state-of-subscription-apps-2025/). Users decide almost immediately whether they see enough value to continue. Apps that fail to deliver that value in the first session lose the majority of users before anyone ever reaches the paywall.
This is an activation failure. And it is invisible unless you are measuring the right event.
## What an App Activation Metric Actually Measures
Activation is the moment a user crosses from "I installed this" to "I understand why this is worth paying for."
For a language-learning app, activation might be completing three lessons and forming a first sentence in the target language. For a fitness app, finishing a complete first workout. For a budgeting tool, seeing a personalized insight built from real account data.
The specific event differs by product. The standard does not: **an app activation metric is the user behavior that most reliably predicts conversion to a paid subscription and long-term retention.** Not the step you wish users would complete. The step that, empirically, separates subscribers from churners.
**_Real example: Duolingo._**
**What the data showed:** Users who completed lessons on consecutive days retained at dramatically higher rates than users who completed the same total lesson count in a single sitting. Five lessons in one evening looked like strong engagement in aggregate. Returning on Day 2 and Day 3 was the real activation signal. [Duolingo's engineering team reported](https://blog.duolingo.com/improving-the-streak/) that 7-day streak users were **2.4x more likely** to continue using the app, with streak improvements driving a **21% increase** in current-day user retention and a **3.3%** lift in Day-14 retention.
**What Duolingo did:** The team rebuilt its onboarding and push notification system to drive users toward a first multi-day streak, as [documented in Lenny's Newsletter](https://www.lennysnewsletter.com/p/how-duolingo-reignited-user-growth). The redesign was not about adding features. It was about changing which behavior users completed first.
**The takeaway:** Activation is not the completion of an onboarding checklist. It is the behavior that signals a user has started building a routine.
Consider a common scenario: a fitness app with 90%+ onboarding completion on both iOS and Android still loses most users by Day 2. The onboarding is short and easy, but users never reach the feature that delivers real value. Completion looks healthy. Activation is broken.
## Two Layers of Activation Every Subscription App Needs to Track
Activation is rarely a single moment. It has two distinct layers, and confusing them is where most measurement errors originate.
[Screenshot 2026-05-15 at 14.58.55.png]
### 1\. Time to First Value (TTFV)
TTFV is when a user perceives and experiences the product's core promise for the first time. It is a small moment, but the steepest user drop-off happens right here.
[**44.5% of all subscription purchases happen on Day 0**](https://adapty.io/state-of-in-app-subscriptions/). Users who do not see value fast rarely return. First value does not mean completing a setup checklist. It means encountering the specific thing that makes your product feel useful or worth continuing.
**_Real Example: Meditation Apps_**
**The TTFV signal:** For apps like Headspace, product analysts have noted that completing a first full session is the sharpest TTFV signal available. Users who finish even a three-minute guided session show meaningfully higher trial start rates than those who drop mid-way through the intro setup.
**The key insight:** The session does not need to be long. It needs to be completed. That moment of following through is what separates users who experience the product from those who merely open it.
**Applies across categories:** For a finance tool, TTFV might be seeing a budget category populated with real account data for the first time. The specific action differs by product. The principle is the same: value must be felt before users will pay for it.
If users are not reaching this moment, the rest of the funnel rarely gets a chance to matter. They have already left.
### 2\. Time to Core Value
Core value is when the product stops being a new thing to try and becomes a pattern in a user's routine. This layer is one of the strongest predictors of long-term [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) and first-renewal rates you can measure.
A wellness app might define core value as completing sessions on three separate days in the first week. A productivity tool might define it as completing a project workflow end to end. A language-learning app might define it as reaching the 30-word recognition threshold.
Core value almost never happens in the first session. It requires users to return.
Most apps optimize heavily for TTFV and barely track core value. This is why many apps report solid early engagement but poor first-renewal rates: the product is getting users to first value but not to the habit that actually predicts subscription retention.
## How to Define Your Subscription App Activation Event
### 1\. Start with retention curve comparisons
Pull cohort data for users who performed different actions in their first 7 days. Compare retention curves for users who completed Action A versus those who did not. You are looking for sustained divergence that holds over weeks, not a short-term bump that fades after 14 days.
The action that produces a consistently higher retention curve is your activation signal. Divergence that converges by Week 3 signals an engagement step, not a true activation signal.
### 2\. Apply the 2x retention test
A true activation metric passes a simple test: users who hit the milestone retain at significantly higher rates than those who do not. Research across subscription apps and SaaS products consistently shows the threshold is at least 2x retention among activated users versus non-activated users ([Lenny's Newsletter](https://www.lennysnewsletter.com/p/what-is-a-good-activation-rate)).
If users who complete your candidate event do not retain at 2x+ the rate of non-completers, keep looking. The event may correlate with good users, but it is not causing activation.
### 3\. Validate across segments and channels
Check whether the activation signal holds across iOS and Android, paid and organic, and multiple cohorts. A metric that works for organic users but disappears for paid traffic likely reflects that one channel attracts higher-intent users, not that your activation event is working universally.
When running channel-level validation, the [attribution window](https://www.airbridge.io/en/glossary/attribution-window) you apply will also affect which sessions get credited to which campaigns. Both the event and the measurement window need to be consistent across segments.
Segment out free users when running this analysis. Free-tier engagement routinely inflates activation metrics while paid cohort quality stays poor underneath.
### 4\. Set a threshold, not just a binary
For many apps, how much of the core experience a user completes predicts outcomes better than whether they completed any of it. In one reported case, a fitness app found that users who completed at least two workouts during trial retained at dramatically higher rates than one-workout users. The two-workout threshold was the real activation metric, not "completed a workout."
When you count something matters as much as what you count.
## Activation Benchmarks by App Category
No single activation rate applies across all subscription apps. These figures give a baseline to calibrate your own expectations.
| Category | Trial Start Rate | Trial-to-Paid | What This Tells You |
| --- | --- | --- | --- |
| Travel | N/A | 48.7% | Trialists have strong purchase intent; they are researching, not browsing |
| Media and Entertainment | N/A | 43.8% | Content-led apps convert well once users sample the library |
| Health and Fitness | 7.8% | 39.9% | Strong overall funnel; yearly plans dominate at 68% of subscriptions |
| Business | 8.9% | N/A | Highest trial adoption rate; professional users trial with clear intent |
| All Categories | 6.2% (median) | 34.8% (median) | Broad baseline for calibrating your own targets |
_Source:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps-2025/) _(75,000+ apps, $10B+ revenue)_
A useful signal from the trial-length data: subscription apps with longer trial windows (17-32 days) reach a median trial-to-paid conversion of **45.7%**, compared to **26.8%** for short-window trials. The longer window gives users time to reach core value. That downstream improvement is what well-designed activation infrastructure produces.
Generic SaaS activation rates (often cited at a median around 25-30%, up to 36% for SaaS-specific averages) are not useful benchmarks for mobile subscription apps. Category, paywall model, and trial length all change what a healthy rate looks like for your product.
## Activation Is the Bridge Between Installs and Revenue
When you define an activation metric backed by retention data, the diagnostic question changes. It shifts from "we need more installs" to "users are reaching first value but not core value, and that gap explains why first-renewal rates are soft."
That specificity drives the right intervention: shorten the path to first value, give users enough runway to build the habit that predicts retention, and track both layers with events that fire at the right moments.
Most [self-attributing networks](https://www.airbridge.io/en/glossary/self-attributing-network) (Meta, Google, Apple Search Ads, and TikTok) optimize toward the top of your funnel. Connecting those channels to actual activation events is where the measurement gap lives for most early-stage apps.
> Once you have both activation layers defined, the next step is connecting them to acquisition channels. Most teams run activation analysis in product analytics and channel ROI analysis in ad dashboards as separate workstreams. The gap between those two tools is where budget decisions go wrong: channels that look efficient on cost-per-install are often the same ones producing users who never activate.
---
## Web-to-App Funnels Explained: What They Are, How They Work, and When to Use Them
URL: https://www.airbridge.io/en/blog/web-to-app-funnels-explained
Language: English
Published: 2026-05-14
Author: Luke
Web-to-app is a user acquisition strategy where ad clicks land on a web page before the App Store. That web page qualifies, personalizes, or charges the user first, then redirects them to install the app. It is designed to improve conversion rates, recover ad signal lost after iOS 14, and reduce App Store commission costs.
If you run paid ads for a subscription app and keep hearing that you should "try web-to-app," this guide explains what it actually is, how it works mechanically, what the data says about its trade-offs, and how to decide whether building one is worth it for your situation.
**Key Takeaways**
- **Web-to-app routes ad clicks through a web page before the App Store.** Users land on a quiz, landing page, or paywall, then download the app after engaging on the web.
- **iOS 14's tracking changes made it popular.** Web-to-app uses cookies and UTM parameters to restore ad attribution data that Apple's SKAdNetwork cannot provide.
- **Web payments structurally favor margin.** Every renewal that clears through Stripe instead of Apple IAP keeps up to **~27% more** on that transaction. At volume and high price points, that compounds into a meaningful budget line.
- **Web and in-app subscribers follow opposite retention curves.** Web starts stronger (**84.5%** Month 1 vs **48.2%**) but churns faster. The curves cross around Month 6, and raw LTV ($10.8 vs $40.1) reflects that long-term divergence.
- **Fit matters more than category.** The strongest web-to-app results come when the web page does selling work the App Store listing cannot: turning a quiz into a commitment before the install happens.
- **It adds real engineering complexity.** Web-to-app requires a separate tech stack, a measurement setup, and policy compliance work. It is not plug-and-play.
## What Web-to-App Actually Means
Traditional paid UA looks like this: a user sees your ad on Meta or TikTok, taps it, lands on your App Store product page, installs the app, then goes through in-app onboarding. The entire conversion path runs inside Apple or Google's ecosystem.
Web-to-app inserts a web step before the App Store. The user taps the ad and lands on a mobile web page you control, typically a short quiz, a personalized results page, or a direct paywall, then gets redirected to install. In some implementations, the user subscribes on the web before the install happens at all.
The term covers a wide spectrum. At the simpler end, it is a landing page between the ad click and the App Store link. At the more sophisticated end, it is a full alternative checkout where the user pays on the web and downloads the app as the delivery mechanism for a product they already purchased.
Three things web-to-app is not:
1. It is not a new app store. You still need a live app on iOS or Android.
2. It is not app-to-web, which is a separate strategy where users already inside the app get redirected to a web paywall to purchase.
3. It is not a blanket policy workaround. Web-to-app that complies with platform rules is legal. Using web payments to circumvent IAP in regions or categories where external purchases are not permitted is not.
## Why Web-to-App Became a Standard Tactic
Three structural forces pushed web-to-app from a niche experiment to a mainstream strategy.
### 1\. iOS 14 and the collapse of mobile attribution
When Apple introduced [App Tracking Transparency](https://www.airbridge.io/en/glossary/app-tracking-transparency-att) in 2021, device-level tracking for app ads ended for users who declined consent. [SKAdNetwork](https://www.airbridge.io/en/glossary/skadnetwork), Apple's privacy-safe attribution replacement, gives ad platforms aggregated, delayed, and heavily limited signals, providing far less data than marketers had before.
Web-to-app is a partial workaround. When users click an ad and land on a web page you control, standard web cookies and UTM parameters apply. This restores more granular attribution data before the user enters the App Store environment where tracking degrades. The ad creative, audience segment, and campaign that drove the click all become trackable again at the web layer ([StartApp School](https://www.startapp.school/), Lesson 26: Understanding Attribution).
### 2\. App Store fees on every renewal
Apple charges **15–30%** on subscription revenue processed through IAP. Google charges **15%** after the first year. For a subscription app generating **$20/month** per subscriber, that is **$3–6** per user per month going to the platform on every renewal, indefinitely.
Web payments change that math. Processing subscriptions via Stripe or a comparable provider costs roughly **3–4%** per transaction. The difference is meaningful: up to **~27% more revenue** retained per subscription when comparing web payment fees to Apple IAP ([48 Laws of Subscription App Success](https://www.retention.blog/p/welcome-to-the-48-laws-of-subscription), Law 30).
Following a 2025 contempt finding, Apple updated its App Review Guidelines to allow external payment links in US iOS apps ([TechCrunch, 2026](https://techcrunch.com/2026/04/29/apple-epic-games-app-store-fees-pause-changes-supreme-court/)), giving web-to-app flows clearer legal standing than two years ago. Commission rules remain under active litigation, but the trend suggests web payment flows will retain their current legal standing.
### 3\. Control over onboarding and pre-qualification
App Store product pages are rigid. You can run A/B tests on screenshots and descriptions, but you cannot show a personalized quiz, build a 20-question onboarding flow, or dynamically change messaging based on which ad the user clicked.
A web page has none of those constraints. Marketers build quiz funnels that ask about user goals and surface tailored results before the install. This pre-qualifies users and builds intent before they open the app. For apps with higher-complexity value propositions like habit formation, mental health, and personal finance coaching, the web page does selling work that a static App Store listing cannot.
## How a Web-to-App Funnel Works
The mechanics of a web-to-app funnel are straightforward. Here is what a standard flow looks like:
[Screenshot 2026-05-15 at 14.50.10.png]
### 1\. The ad click lands on your web page, not the App Store
The ad on Meta, TikTok, Google, or Apple Search Ads links to a URL you own, typically a mobile-optimized landing page or quiz. The moment the user lands, standard web analytics fire: cookies, UTM parameters, and browser-side tracking. This is where attribution signal is recovered.
### 2\. The web experience qualifies or converts the user
This step varies by implementation. The simplest version is a short landing page with a single CTA pointing to the App Store. More sophisticated funnels include a multi-step quiz that personalizes the experience based on user goals, or a full web paywall where the user subscribes and enters payment details before installing.
### 3\. The handoff from browser to app
After the web interaction, the user taps a link to the App Store or Google Play. If they already subscribed, the app recognizes them on first launch via an app link or email-based account lookup. If they only expressed intent, the in-app onboarding picks up where the web experience left off, often surfacing the same personalization from the quiz.
Throughout all three steps, web-side events, including page views, quiz completions, and purchase events, are tracked by your web analytics and passed back to your ad platforms as conversion signals. This closes the attribution loop that SKAdNetwork leaves open.
## Web-to-App vs. Direct-to-App: What the Data Shows
The comparison between web-to-app and direct-to-app subscribers looks like this, based on [Adapty's analysis of 16,000+ apps](https://adapty.io/state-of-in-app-subscriptions-report/):
| Metric | Direct-to-App (In-App IAP) | Web-to-App (Web Paywall) |
| --- | --- | --- |
| Conversion rate | 1.65% (download-to-paid) | 1.13% (download-to-paid) |
| Month 1 retention | 48.2% | 84.5% |
| Month 6 retention | 30% | 20% |
| LTV (raw) | $40.1 | $10.8 |
| Processing fee | 15–30% (App Store / Play) | 3–4% (Stripe or equiv.) |
| Attribution quality post-ATT | Degraded (SKAdNetwork) | Higher (cookies + UTM) |
_Web raw LTV ($10.8) is lower than in-app ($40.1) due to faster long-term churn, despite stronger early retention. The ~27% fee savings on web payments are real margin, but they do not fully close the retention-driven gap. Source:_ [_Adapty State of In-App Subscriptions 2026_](https://adapty.io/state-of-in-app-subscriptions-report/)_._
Three things stand out from this data.
**Conversion rates are lower on web.** Users who go through the App Store convert at **1.65%** on average; web paywall users at **1.13%**. The extra step costs conversions for some apps. Web-to-app adds friction, and friction has a price.
**Month 1 retention is genuinely better on web.** Web subscribers hold at **84.5%** in the first month versus **48.2%** for in-app. Deliberate purchase intent is the most plausible explanation: web subscribers completed a quiz, reviewed a paywall, and chose to pay without the frictionless "start free trial" tap that in-app offers. That deliberateness produces stickier early subscribers.
**The long-term** [**LTV**](https://www.airbridge.io/en/glossary/lifetime-value-ltv) **picture requires careful reading.** Web subscribers churn faster over time, so their raw LTV is only **$10.8** versus **$40.1** in-app. That is partly an accounting gap: the ~27% you save by avoiding App Store fees is real margin not reflected in raw LTV. Even so, the retention divergence that begins around Month 6 is the dominant driver, and fee savings alone do not close it.
Web-to-app makes the most sense when you have a hypothesis that your web flow will produce higher-quality users, not simply when you want to reduce platform fees.
## When Web-to-App Works, and When It Doesn't
The following diagram maps the key decision criteria at a glance:
[Screenshot 2026-05-15 at 14.51.28.png]
**When web-to-app tends to work well:**
Web-to-app earns its complexity cost when your app's value proposition requires explanation. Weight loss apps, language learning apps, and financial planning tools benefit from a quiz-driven flow that sets expectations before the install. The web page does selling work that a static App Store listing cannot.
It also makes more sense when your subscription price is high enough that fee savings matter in real dollars. At a **$20/month** price point, retaining **~27%** more per transaction is roughly **$5.40 per user per month**, or **$64.80 per year**, in additional margin. At a **$50/month** subscription, the same percentage becomes a meaningful budget line.
For iOS campaigns running with degraded signal post-[ATT](https://www.airbridge.io/en/glossary/app-tracking-transparency-att), the attribution recovery from web click data alone can justify the investment. If your Meta or TikTok campaigns cannot tell you which creative drives subscribers, the web click layer restores enough signal to optimize.
**When it tends to underperform:**
If your app's value is immediately apparent from the App Store page, like a simple timer, a weather widget, or a casual game, adding a web quiz step introduces friction without adding clarity. Conversion rates will likely drop without a compensating lift in subscriber quality.
Engineering resources matter equally. A broken redirect, a mismatched account login on first open, or a web payment that does not sync with in-app entitlements will produce support tickets and early churn. Web-to-app done poorly costs more than it saves.
## Start Where the Data Takes You
The most useful reframe for teams evaluating web-to-app is to stop asking "should we do this?" and start asking "which user segment is worth a web-qualified path?" The web funnel is not a replacement for your App Store channel; it is a parallel track for users whose purchase decision benefits from more information before they commit.
The data is honest about the trade-offs. The [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) gap between web and in-app subscribers is real and documented. The conversion rate hit is real. The build cost is real. What the aggregate numbers cannot tell you is whether your specific funnel, quiz design, and paywall positioning can outperform the average. That is only knowable through your own data.
> The right approach is to treat web-to-app as a test, not a pivot. Route roughly **10–20%** of one campaign through a simple landing page and measure subscription rates and **90-day retention** against your baseline in-app cohort.
One underappreciated advantage of web funnels: price experimentation is substantially easier without App Store review cycles. Adapty's 2026 data shows price experiments improve LTV by **46%** on average, and web is the fastest environment to run them. If the test shows retention holding and margin improving, scale from there.
---
## Subscription App Pricing by Category: 2026 Benchmarks
URL: https://www.airbridge.io/en/blog/subscription-app-pricing-by-category-2026-benchmark
Language: English
Published: 2026-05-13
Author: Harper (Trang Nguyen)
Subscription app pricing benchmarks can tell you where your app stands relative to the market. They show what Health & Fitness apps charge, how Gaming compares to Education, and what annual vs. monthly pricing looks like across the industry.
But benchmarks alone won't tell you whether your subscription app pricing is actually working once users arrive from paid ads. That question requires a different kind of data.
## Key Takeaways
- **The most common monthly price is $9.99, but the market median is $6.68.** The most common price sits at the top quarter of the market, not the middle.
- **Category benchmarks matter more than overall medians.** Health & Fitness monthly median is $9.70 vs Gaming's $4.99. Using the wrong benchmark mispositions your app pricing by more than 2x.
- **Annual subscribers retain at 44.1% after 12 months vs 17.5% for monthly.** Which billing period your UA channels drive directly determines your LTV trajectory.
- **Higher-priced apps convert at higher rates, not lower.** High-priced apps hit **2.7%** D35 download-to-paid conversion vs **1.5%** for low-priced, because higher prices pre-filter for purchase intent.
- **Teams running structured pricing experiments generate up to 40x more revenue.** Revenue per install at Day 30, not same-day conversion, is the right metric to judge price tests.
## What Subscription Apps Actually Charge in 2026
Across **75,000+ subscription apps** generating over $10B in annual revenue, here is how **subscription app pricing by category** breaks down across plan types in 2026:
| Plan | Bottom 25% | Median | Most Common | Top 25% | P90 |
| --- | --- | --- | --- | --- | --- |
| Weekly | $3.08 | $4.99 | $4.99 | $6.99 | $9.99 |
| Monthly | $3.26 | $6.68 | $9.99 | $11.99 | $19.99 |
| Annual | $17.59 | $29.99 | $29.99 | $54.99 | $89.99 |
_Source:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps/)_, 75,000+ apps, 2024 data_
Monthly app pricing shows a notable split. The median is **$6.68**, but the most common price point is **$9.99**. Many apps anchor at $9.99 because competitors do. But half the market charges under $6.68. Apps pricing at $9.99 are in the top quartile, not the middle. On average, apps offer a **67% discount** on annual vs equivalent monthly pricing.
The overall medians provide a directional baseline. Category-level subscription app pricing is what actually informs decisions:
| Category | Weekly Median | Monthly Median | Annual Median |
| --- | --- | --- | --- |
| Health & Fitness | $4.99 | $9.70 | $39.99 |
| Education | $5.99 | $8.38 | $44.99 |
| Business | $6.15 | $7.58 | $37.50 |
| Gaming | $4.99 | $4.99 | $20.55 |
_Source:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps/)_, 75,000+ apps, 2024 data_
A few patterns stand out:
- Education leads with the highest annual median ($44.99), more than double Gaming's ($20.55).
- Health & Fitness monthly pricing ($9.70) is nearly double Gaming's monthly median ($4.99).
- Business leads in weekly pricing at $6.15, reflecting professional daily use patterns.
- An app priced well below its category median may be leaving revenue on the table, not just competing on price.
## Billing Period by Category: The LTV Multiplier
The plan type your category defaults to is one of the strongest drivers of long-term revenue. With pricing held constant, billing period alone creates a roughly 3x LTV gap between annual and monthly subscribers:
- Annual subscribers retain at **44.1%** after 12 months.
- Monthly subscribers retain at **17.5%** after 12 months.
- Weekly subscribers retain at **3.4%** after 12 months.
[Subscription App Pricing by Category 2026 Benchmarks-1.webp]
**Annual-dominant categories** (annual plan is the primary revenue vehicle):
- Health & Fitness: **67%** annual
- Travel: **65%** annual
- Education: **57%** annual
**Weekly-dominant categories** (first renewal is the critical moment):
- Gaming: **78%** weekly
- Business: **45%** weekly
- Social & Lifestyle: **45%** weekly
In annual-dominant categories, the annual plan is the primary revenue and retention vehicle, not an upsell. In weekly-dominant categories, only **51.8%** of subscribers renew after the first billing cycle, making the first week the highest-risk period in the subscriber lifecycle.
For paid UA, this distinction matters directly. A channel that drives weekly subscribers in a Health & Fitness app generates roughly one-third the LTV of a channel driving annual subscribers at the same cost per install.
## Price, Conversion, and LTV: The Counterintuitive Data
The assumption that lower app pricing drives higher conversion and more revenue does not hold in the data from tens of thousands of subscription apps. Three findings stand out:
- **Higher-priced apps convert at a higher rate.** High-priced apps hit **2.7%** D35 download-to-paid conversion vs **1.5%** for low-priced apps. Higher prices pre-filter installs toward users with stronger purchase intent.
- **High-priced apps generate Y1 LTV 7x greater** than low-priced apps per paying subscriber.
- **Price experiments improve LTV in 46% of tests, but conversion in only 28%.** Judging a price test by same-day conversion leads to the wrong conclusion most of the time. _(_[_Adapty 2026_](https://adapty.io/state-of-in-app-subscriptions-report/)_)_
The paywall model compounds this effect significantly:
| Model | D35 Download-to-Paid | Median Y1 LTV per Payer |
| --- | --- | --- |
| Hard Paywall | 12.1% | $49.30 |
| Freemium | 2.2% | $24.24 |
_Source:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps/)_, 75,000+ apps, 2024 data_
Hard paywalls convert at **5.5x the rate** of freemium and produce **2x the LTV** per subscriber. The trade-off is a **1.7x higher refund rate**. For paid acquisition, hard paywall apps can typically sustain higher CPIs because the downstream LTV justifies the cost. Freemium apps need lower CPIs to maintain positive unit economics, because fewer installs become subscribers and each generates lower LTV.
Teams that run structured pricing experiments generate up to **40x more revenue** than those that don't. _(_[_Adapty 2026_](https://adapty.io/state-of-in-app-subscriptions-report/)_)_ The right metric to judge those experiments is revenue per install at Day 30 and Day 60, not same-day conversion rate.
---
> The benchmarks above tell you what the category does. They give you reference points for pricing, billing period norms, and the LTV levels your vertical typically achieves. That context is genuinely useful: it tells you whether you're priced at, above, or below your category median, and what the gap costs you in retention.
>
> But they can't answer the question that matters most for early-stage apps spending real money on paid UA: _are your paid users actually starting trials, becoming subscribers, and generating enough revenue to justify your acquisition cost?_ That question requires your own data, and most early-stage apps don't have it connected the right way.
## Pricing Benchmarks vs. Paid Acquisition Reality
Knowing where your subscription app pricing stands against the Health & Fitness annual median of **$39.99** is useful context. Knowing whether your Meta campaigns are producing subscribers at that price point, and what their 30-day revenue looks like compared to your acquisition cost, is what drives actual budget decisions.
For early-stage subscription apps, the gap between "we know the category benchmark" and "we know our own numbers" is usually wide. Installs are tracked. Revenue appears in the app store. What's missing is the chain between them: which campaigns drove trial starts, which trials converted to subscribers, what billing period those subscribers are on, and what the revenue per install looks like by channel.
### The Funnel You're Flying Blind Through
Most early-stage apps running paid UA have clear visibility at the top of the funnel: install volume and install cost. The data goes dark from there. Consider what you typically can't see:
- How many of last week's installs actually started a trial?
- Which specific campaign drove those trial starts?
- Were those trial users on the **$9.99** monthly plan or the **$39.99** annual?
- What does their Day 30 revenue look like compared to the CPI you paid?
Without closing this gap, benchmark comparisons become one-sided. You can see that high-priced apps in your category convert at **2.7%**. You cannot see whether your app is at 2.7%, 0.9%, or 4.1%, or which of your campaigns is responsible for the difference.
[Subscription App Pricing by Category 2026 Benchmarks-2.webp]
### The Number That Connects Pricing to UA Spend: CPS vs LTV
**Cost per subscriber (CPS)** is the metric that determines whether your app pricing makes economic sense in a paid acquisition context. It answers one question: for every dollar spent on paid UA, how much does it cost to produce one paying subscriber?
When you set CPS next to your category's LTV benchmark, you have a direct unit economics check. The benchmark data in this article gives you the LTV side. The missing piece is your actual CPS by channel.
| Scenario | CPS (your data) | H&F Y1 LTV Median | Unit Economics |
| --- | --- | --- | --- |
| Scalable | $9 | $27.21 | 3x return. Scale it. |
| Break-even | $25 | $27.21 | Marginal. Optimize first. |
| Unsustainable | $42 | $27.21 | Acquiring at a loss. |
_H&F Y1 LTV median ($27.21) from_ [_RevenueCat SOSA 2025_](https://www.revenuecat.com/state-of-subscription-apps/)_. Hard paywall apps specifically reach $49.30 median. CPS figures are illustrative scenarios._
The answer is almost always different by channel. Your Meta CPS may be $12 while your Apple Search Ads CPS is $9. Your Google campaigns may drive annual subscribers while your TikTok campaigns drive weekly ones. At Health & Fitness category medians, that billing period difference alone represents roughly a 3x gap in LTV. Without breaking down CPS and billing period by channel, you are optimizing for install volume rather than subscription economics.
> **Example:** A Health & Fitness app spending equally across Meta and Apple Search Ads may find that Meta drives more installs at a lower CPI, but ASA drives annual subscribers whose LTV justifies a higher CPI. Rebalancing budget toward ASA increases CPS per channel but reduces total CPS per annual subscriber. At the same total spend, the outcome improves.
### How Airbridge Core Plan Connects the Dots
[Airbridge Core Plan](https://www.airbridge.io/en/plans/core) is built for early-stage subscription apps that want to move beyond install tracking and see whether their paid acquisition is actually working.
It connects four ad channels (Meta, Google, Apple Search Ads, and TikTok) to subscription events, including Start Trial, Subscribe, and Unsubscribe. If you're using RevenueCat or Adapty for subscription management, Core Plan integrates directly. Your subscription revenue data connects to your campaign data without custom backend engineering.
After setup, you can answer questions like:
- Which Meta campaigns produce trial starts, and which drive installs that never convert?
- Are your Google campaigns producing annual subscribers or weekly ones? The LTV difference at Health & Fitness medians is roughly 3x.
- What is your actual CPS by channel, and how does it compare to your category's LTV benchmark?
- Which creative drives the highest Day 60 revenue per install?
---
**If you're spending on Meta, Google, Apple Search Ads, or TikTok and want to see whether those campaigns are producing subscribers at your category's LTV benchmark, Airbridge Core Plan connects the full subscription funnel to your paid campaigns.**
Start free with **15,000 attributed installs**. Pay-as-you-go after that. No annual contract.
## The Gap Between a Benchmark and a Result
Most subscription apps treat pricing as a decision you make once. The teams that outperform their category medians treat it as an ongoing experiment. They track CPS by channel, measure revenue per install at Day 30 and Day 60, and run structured tests against benchmarks they actually understand.
The benchmark tells you where the category sits. Your subscription funnel data tells you whether your app is getting there. Connecting the two is what turns a pricing decision into a pricing strategy.
---
## ROAS for Subscription Apps: How to Read D7, D30, and D60 Numbers
URL: https://www.airbridge.io/en/blog/roas-for-subscription-apps
Language: English
Published: 2026-05-13
Author: Luke
For subscription apps, ROAS is never a single, final figure. It grows every month as subscribers renew, and the time window you choose to measure it (D7, D30, or D60) changes everything.
If you have seen a figure in your Meta or Google dashboard and had no idea whether it was good or bad for your subscription model, this guide explains exactly what you are looking at and how to act on it.
## Key Takeaways
- **Time windows define the number.** D7, D30, and D60 tell different stories about the same campaign. Choosing the wrong window leads to cutting campaigns that are actually working.
- **D30 is the standard decision metric.** Most subscription app marketers use 30-day cohort data as their primary signal, with D7 as an early indicator only.
- **A result below 1.0 is not automatically bad.** If subscribers renew at healthy rates, you may recover your customer acquisition cost by month 3 or 6 and still be profitable.
- **Blended numbers hide the real story.** A Meta campaign at **0.8x** and a TikTok campaign at **0.15x** average to **0.45x**, hiding your best opportunity.
- **ROAS makes most sense alongside LTV.** With a median Year 1 LTV of **$27.21** in Health and Fitness, a D30 result of 0.3x can still be profitable.
## Why ROAS Works Differently in Subscription Apps
When an e-commerce brand tracks this metric, revenue is immediate. A user clicks an ad, buys a **$60** product, and that revenue is counted the same day.
Subscription apps do not work that way.
### 1\. Revenue arrives in monthly installments
A user who installs your app after clicking a Meta ad might start a free trial on Day 0, convert to a paid plan on Day 7, and then pay **$9.99** per month for the next **12** months. From the campaign perspective, that user generates **$119.88** in total revenue, but it arrives over a full year.
On Day 1, the revenue from that install is zero. By Day 30, it might be **$9.99**. By Day 365, it might be **$119.88**, or zero if they churned after month two.
This is why the metric is always a moving number for subscription apps. The question is never "what is my ROAS." It is "what is my **D30** result" or "what is my **D60** result."
### 2\. The time window changes what you are measuring
Two marketers can look at the same campaign and reach opposite conclusions depending on which window they choose.
| Time Window | Revenue Counted | What It Tells You |
| --- | --- | --- |
| D1 | Revenue through Day 1 (install day plus the next day) | Trial starts, intro offers, and annual plans paid upfront. Refunds and cancellations have not been processed yet. Never a decision metric |
| D7 | First 7 days of subscription revenue | Early signal only. Too short to capture trial-to-paid conversion for most apps |
| D30 | First 30 days of subscription revenue | Core decision metric for monthly plan apps |
| D60 | First 60 days of subscription revenue | Captures early renewals. More meaningful for longer trial periods |
| D90+ | First 90+ days of subscription revenue | Reflects true payback for high-LTV categories |
For apps with monthly plans, **D30** is typically the first window that means something actionable. For apps with annual plans, even D30 understates long-term value significantly, since most revenue from an annual subscriber comes from the renewal in month 12, not the first payment.
## Why Does ROAS Look Great on Day 1 and Terrible by Day 30?
This is the most common question subscription teams ask when they move from ad-platform dashboards to cohort reports. The pattern is almost always the same: for example, a Day 1 ROAS above 1.5x in the Meta or Google dashboard, then a D30 cohort ROAS below 0.5x. Neither number is wrong. They measure different things, and only one of them should drive budget.
| Cause | What you see | Why it happens | What to measure instead |
| --- | --- | --- | --- |
| Calendar-window revenue instead of cohort revenue | "Yesterday's ROAS" looks high | Ad platforms and most dashboards divide all revenue booked in the period by spend in the period. Renewals from subscribers you acquired months ago land in today's numerator | Cohort ROAS: only revenue from users who installed in the cohort window, tracked by day since install |
| Refunds, trial cancellations, and failed renewals arrive after Day 1 | Annual plans and intro offers are booked on Day 0 or Day 1, then part of that revenue disappears | App Store and Play refunds, trial cancellations, and involuntary churn are processed days or weeks later. Day 1 counts the gross booking | Day-N cumulative revenue that updates as billing events arrive, and a refund rate you track separately |
| Retargeting and self-attributed renewals | Retargeting campaigns show ROAS far above prospecting, often 2x or more | Self-attributing networks claim renewal revenue from existing subscribers who saw a retargeting ad. That revenue was arriving anyway. The number is inflated by attribution overlap, not by new subscribers | Split prospecting from retargeting. Judge acquisition campaigns on new-subscriber revenue only |
One rule catches most of it: **cumulative cohort ROAS can only fall between Day 1 and Day 30 if refunds or chargebacks are netted out.** If your D30 is lower than your D1 and refunds do not explain the gap, you are not looking at cohort numbers.
### How to read D1, D7, and D30 together
| Metric | Formula | Use it for |
| --- | --- | --- |
| D1 ROAS (cumulative) | Cumulative revenue from the cohort through Day 1 ÷ spend on that cohort | Creative and landing page sanity check. Read cost per trial start instead of revenue here |
| D7 ROAS (cumulative) | Cumulative revenue through Day 7 ÷ spend | Early signal for apps with trials of 7 days or shorter |
| D30 ROAS (cumulative) | Cumulative revenue through Day 30 ÷ spend | Budget decisions for monthly plans |
| D60 and D90 ROAS (cumulative) | Cumulative revenue through Day 60 or 90 ÷ spend | Payback checks for annual plans and long trials |
In the Airbridge [Revenue Report](https://help.airbridge.io/en/guides/understanding-the-revenue-report), set the Start Event to Install, choose ROAS with the Cumulative aggregation, and read Day 1, Day 7, and Day 30 on the same row. The report supports daily granularity from Day 0 through Day 180. Send subscription events through the RevenueCat, Adapty, or Superwall integration so Subscribe and renewal revenue lands in the same cohort. If you need refunds netted out, confirm which billing events your integration sends before treating Day-N revenue as net.
## How to Calculate ROAS for Your Subscription App
> **ROAS** (Return on Ad Spend) = Revenue from ad-attributed subscribers divided by Ad Spend.
If you spent **$2,000** on Meta ads in January, and the subscribers acquired from that campaign generated **$800** in subscription revenue within 30 days, your **D30 ROAS is 0.4** (40%). You spent $2,000 and recovered $800 by day 30.
That sounds low. But if those same subscribers generate **$1,500** by day 60, your **D60 ROAS is 0.75**, and you are approaching breakeven. Whether that pace works depends on your cash runway and predicted LTV.
[Screenshot 2026-05-13 at 17.13.13.png]
_Illustrative example. Revenue figures are hypothetical to demonstrate how D30, D60, and D90 ROAS change as subscribers renew._
### 1\. Use cohort revenue, not total app revenue
One of the most common mistakes is dividing total app revenue by total ad spend. This produces a blended number that mixes organic users, returning subscribers, and newly acquired paid installs.
To get accurate cohort-level data, you need revenue generated specifically by users acquired from a specific campaign, measured over a specific time window. Actuals tell you cash flow. Cohorts tell you marketing performance. For optimizing ad spend, always use cohorts.
### 2\. Count subscription revenue only
If your app includes both subscriptions and one-time purchases, separate them. A subscriber who pays **$9.99/month** looks very different from a user who makes a **$1.99** in-app purchase and never returns. Subscription-only data tells you whether your campaigns are generating paying subscribers. Mixing in consumables distorts the signal.
### 3\. Account for platform commissions
These calculations use gross subscription revenue. Apple takes **30%** in Year 1 (dropping to **15%** from Year 2 onward), so net ROAS is lower. A D30 gross ROAS of **0.4x** is a net ROAS of roughly **0.28x** on iOS. Google Play charges **15%** from Year 1. When comparing iOS and Android ROAS performance, always use the same revenue basis (gross or net) to avoid misleading comparisons.
## ROAS Benchmarks: What the Data Actually Shows
There is no universal target because this metric depends on both revenue per install and cost per install, and CPI varies sharply across categories and platforms.
What you can benchmark directly is revenue per install (RPI) at each time window. Analyzing [75,000+ subscription apps](https://www.revenuecat.com/state-of-subscription-apps-2025/), the data shows:
| Time Window | Median RPI (All Categories) | Health and Fitness Median | Health and Fitness P90 |
| --- | --- | --- | --- |
| Day 14 RPI | $0.20 | — | — |
| Day 60 RPI | $0.31 | $0.63 | $4.19 |
| Year 1 LTV per Payer | $16.27 | $27.21 | $86.35 |
Platform and geography matter significantly. App Store D60 RPI averages **$0.38** globally vs **$0.14** on Google Play, a **2.7x** gap. In North America, D60 RPI reaches **$0.57** at the median, compared to **$0.10** in India and Southeast Asia.
Your result at any window = RPI divided by CPI.
For Health and Fitness apps on iOS globally, the median CPI is **$5.78**. Using median figures, D60 ROAS = **$0.63 / $5.78 = 0.11x**. You have recovered **11 cents** per ad dollar by Day 60. On Android in North America, where the median Health and Fitness CPI is **$3.06**, the same D60 RPI gives **0.21x**.
These numbers show why reading the metric at only one window without knowing your CPI is hard to interpret. An app spending **$0.80** per install with **$0.31** D60 RPI has D60 ROAS of **0.39x**. An app spending **$5** per install with the same RPI has D60 ROAS of **0.06x**. Both might be healthy or struggling, depending on their LTV trajectory.
One more number to anchor your expectations: the cost per paying user (CPPU) runs **4-5x higher** than cost per install. If your CPI is **$5**, plan to spend **$20-25** to acquire one paying subscriber. Your calculations need to account for that full funnel cost, not just the install.
For a deeper look at the metrics that connect to this calculation, see our [subscription app metrics guide](https://www.airbridge.io/en/blog/subscription-app-metrics-explained).
## How to Read ROAS by Channel
Your app-level number is an average. It blends your strongest campaigns with your weakest. The actionable work is breaking the data down by channel, campaign, and creative.
A Meta campaign might show **D30 ROAS of 0.8x** while your Google campaign shows **0.3x**. Same app. Same time period. Opposite conclusions about where to increase budget. [Self-Attributing Networks](https://www.airbridge.io/en/glossary/self-attributing-network) like Meta, Google, and TikTok each report their own attribution, so the number in each platform dashboard does not add up to a unified view of your subscription revenue.
[Screenshot 2026-05-13 at 17.14.05.png]
This pattern is common for Health and Fitness apps running Meta prospecting alongside Google App campaigns. The gap between top and bottom channels often exceeds 2x.
Connecting ad spend data to actual billing events like trial start, subscription, and renewal requires linking your campaign data to your revenue data. Without that connection, you are making budget decisions based on platform-reported numbers that each platform inflates in its own favor.
If you are running paid campaigns on Meta, Google, or TikTok and want to see which channels are actually driving paying subscribers, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your ad spend to subscription revenue by source. You can see the full path from install to renewal, broken down by channel, campaign, and, where the network provides creative-level cost, ad creative. Airbridge Core Plan starts at $40+/mo after a 30-day free trial, with 500K data points a month included, $0.0001 per additional data point, and no annual contract. Attribution events and tracking link clicks do not count as data points.
For more on scaling paid spend once you can read channel-level data, see our guide on [how to improve D30 ROAS for subscription apps](https://www.airbridge.io/en/blog/how-to-improve-d30-roas).
## Frequently Asked Questions
### Our ROAS looks great on day one and terrible by day thirty. What should I be measuring instead?
Measure cumulative cohort ROAS at D7 and D30, net of refunds where your billing integration sends refund events, with prospecting and retargeting campaigns separated. A high day-one number in an ad platform dashboard is usually calendar-window revenue that includes renewals from older cohorts, plus bookings that are later refunded or cancelled. Cohort ROAS by day since install removes both effects.
### What is a good D30 ROAS for a subscription app?
There is no universal target. D30 ROAS equals revenue per install divided by cost per install, and both vary by category, platform, and country. Using RevenueCat's 2025 benchmark data, a Health and Fitness app with median CPI on iOS sits near 0.11x at D60, and it can still be profitable if Year 1 LTV per payer is near the category median of $27.21. Compare your D30 against your own LTV and payback target, not against a single industry number.
### Should I use D1 ROAS at all?
Use Day 1 as a signal for trial starts and cost per trial, not for revenue. Subscription revenue on Day 1 is mostly intro offers and annual plans paid upfront, and refunds have not been processed yet. Budget decisions belong to D30 for monthly plans and D60 or later for annual plans.
### How do I get D-N cohort ROAS when Meta and Google only report their own numbers?
Connect ad spend from each channel and subscription revenue from your billing platform in one cohort report. Airbridge Core Plan pulls cost from Meta, Google, TikTok, and Apple Search Ads, receives subscription events from RevenueCat, Adapty, or Superwall, and shows cumulative ROAS by day since install in the Revenue Report.
## ROAS Is Just the Starting Point
The metric tells you whether an ad campaign is generating subscription revenue relative to what you spent. It does not tell you whether those subscribers will stay for three months or three years. It does not tell you whether the channel you are scaling is actually responsible for the lift, or whether those users would have found you anyway.
That is why the teams who use this data well always read it alongside [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) and cohort retention data. The early number is a signal. Unit economics are the verdict.
---
## 5 App Marketing Shifts After Apple's Privacy Changes (2026)
URL: https://www.airbridge.io/en/blog/5-app-marketing-shifts-after-apples-privacy-changes
Language: English
Published: 2026-05-13
Author: Harper (Trang Nguyen)
Privacy changes didn't just make tracking harder. They changed who wins. [App Tracking Transparency (ATT)](https://www.airbridge.io/glossary/app-tracking-transparency-att), launched with iOS 14.5 in April 2021, cut off automatic access to device-level identifiers. ATT opt-in rates fell to a global low of **13.85%** in mid-2024, down **16%** in Q1 alone and another **12.5%** in Q2. The platforms adapted. The attribution frameworks changed. And the gap between marketers who understand exactly what shifted, and those who are still guessing, keeps widening.
This article covers the five changes that are actually affecting your numbers, what they mean in practice, and what to do differently right now.
**Key Takeaways**
- **ATT didn't kill targeting; it changed who has signal.** Opt-in users now carry far more value per impression, and your bidding strategy should reflect that asymmetry.
- **SKAN replaced user-level attribution with aggregated, delayed postbacks.** Campaign measurement now happens at the cohort level, not the individual level.
- **Creative is doing the targeting work that audience data used to do.** Hook rate and hold rate are the new optimization levers.
- **Android crossed the majority threshold for paid installs.** Android now accounts for more than half of global paid app installs, while North America UA spend fell **\-40% YoY**.
- **Incrementality testing replaced last-click attribution as the standard.** If you are not running holdout tests, you are likely crediting your best channels incorrectly.
Below are the five major app marketing shifts shaping growth strategies after Apple's privacy changes.
## ATT Wiped Out User-Level Targeting
[App Tracking Transparency](https://www.airbridge.io/glossary/app-tracking-transparency-att) requires apps to explicitly ask users for permission to track them across other companies' apps and websites. When users say no, the Identifier for Advertisers (IDFA) is unavailable. Ad networks operate without individual user-level signals.
ATT opt-in rates vary by measurement methodology. Singular's Q2 2024 report placed the global average at **13.85%** in mid-2024. For non-gaming apps, the rate was lower still: **11.92%**, versus **18.58%** for games. The platforms were not designed to function at this consent rate. Meta, Google, and TikTok are now running campaigns on probabilistic and modeled signals rather than deterministic user data.
The platform response was to centralize optimization into their own systems. Advantage+ Campaigns (Meta) and Performance Max (Google) handle targeting internally using first-party data the platform owns. You give up manual control over audience selection in exchange for algorithmic reach. Teams that resist [this shift](https://www.airbridge.io/en/blog/post-att-advertising-strategies), trying to preserve granular campaign structures from 2019, find CPMs rising and performance declining.
The consent rate itself is something you can influence. Apps that show the ATT prompt after a user completes a meaningful action, such as signing up or reaching a paywall moment, consistently see better opt-in results than apps that prompt on first launch ([Purchasely, 2025](https://www.purchasely.com/blog/att-opt-in-rates-in-2025-and-how-to-increase-them)). The timing of the consent request is now a media buying decision, not an afterthought.
[5 App Marketing Shifts After Apple's Privacy Changes (2026).webp]
## SKAN Replaced Real-Time Attribution
[SKAdNetwork (SKAN)](https://www.airbridge.io/glossary/skadnetwork) is Apple's privacy-preserving attribution framework, designed to give advertisers campaign-level data without exposing individual user identities. SKAN 4.0, now the industry standard, delivers up to three postbacks spread over 35 days post-install rather than real-time user-level signals ([Aarki, 2025](https://www.aarki.com/insights/why-apples-privacy-updates-are-a-growth-opportunity-7-winning-moves-for-your-app/)). The data is aggregated and tiered: campaigns that reach volume thresholds receive more granular postback data; those that do not receive suppressed or null values.
| Measurement Model | Data Type | Timing | Granularity |
| --- | --- | --- | --- |
| Pre-ATT (IDFA) | User-level | Real-time | Full funnel |
| SKAN 4.0 | Aggregated | Up to 35 days | Campaign-level |
| Modeled attribution | Probabilistic | Real-time (estimated) | Mixed |
The practical implication: you can no longer optimize campaigns on Day 2 subscription data the way you once did. Mapping your [SKAN conversion schema](https://www.airbridge.io/en/blog/exploring-the-marketing-implications-of-skadnetwork-4) to the early event that best predicts long-term subscription value is the most important technical move for iOS measurement. For most subscription apps, that event is "Start Trial" within the first 24-48 hours.
Apple's next evolution, [AdAttributionKit (AAK)](https://www.airbridge.io/en/blog/skadnetwork-4-must-know), builds on SKAN and expands attribution to additional surfaces including Safari. AAK is not replacing SKAN immediately, but teams who ignore it now will face another migration unprepared.
[Self-Attributing Networks (SANs)](https://www.airbridge.io/glossary/self-attributing-network), including Meta, Google, TikTok, and Apple Search Ads, report their own attribution numbers separately from what any measurement platform sees. Post-ATT, the discrepancies between platform-reported and independently-measured data are larger than before. Focus on directional trends rather than demanding perfect reconciliation between sources.
## Creative Became the Primary Targeting Tool
Before ATT, layering audience targeting on top of average creative still converted. Post-ATT, that safety net is gone. With iOS audience signal mostly unavailable, what self-selects the right audience is the ad itself.
The first second of a video ad now determines whether your campaign gets algorithmic amplification. By the **3-second mark, roughly 50% of viewers have already scrolled past**. Hook rate and hold rate are the metrics that signal to platform algorithms whether to expand delivery or throttle it.
Three implications for subscription app teams:
1. **Build volume.** A library of roughly **100 creative variants** gives you enough test data to identify which angles and hooks resonate. Testing fewer variants means slower learning and higher cost per subscriber.
2. **Format is platform-specific.** Creative that performs on Meta does not automatically work on TikTok. TikTok requires content that matches TikTok culture. Run each platform on its own creative system.
3. **Meta ad group structure changed.** Meta's algorithm prioritizes **1-2 creatives per ad group** regardless of how many you upload. New winners belong in new ad groups, not added to existing ones.
Budget discipline compounds everything. The [10x rule](https://www.startapp.school/courses/paid-user-acquisition-for-apps): set daily budget at a minimum of **10 times your target cost per acquisition (CPA)**. Running on too little budget starves the algorithm of the conversion volume it needs to optimize. With noisier post-ATT signals, this constraint is stricter, not looser.
## Android and Emerging Markets Took the Lead
The geographic and platform picture changed substantially. The data across **1.7 billion paid installs** of **2,900 subscription apps** makes the shift concrete.
**Android crossed the majority threshold.** Android's share of paid app installs rose from **43% to 51%**, crossing the majority line. iOS grew from **30% to 34%**. Android is now the higher-volume platform for paid acquisition and offers more measurement flexibility because ATT does not apply.
**North America is no longer the growth engine.** UA ad spend in North America fell **\-40% YoY**. Indian Subcontinent grew **+423%**, LATAM grew **+77%**, Eastern Europe grew **+96%**, Middle East grew **+197%**. Apps built entirely around US iOS audiences are over-indexed in a declining segment.
**Category concentration is accelerating.** Health and Fitness saw its top 5 apps go from controlling **54% of UA spend to 73%** in one measurement period. In Short Drama and OTT, the top 5 apps hold **90%+** of all UA spend. Incremental ad budget alone cannot break through in saturated categories. Product differentiation has become a prerequisite for paid growth.
---
## Incrementality Replaced Last-Click Attribution
Incrementality testing is the most important methodological shift for post-ATT measurement. Hold out a portion of your audience from seeing ads, then compare conversion rates between the exposed group and the holdout. The difference is your true incremental lift. Without this, you risk crediting campaigns for users who would have converted anyway. For a full explainer on [how mobile attribution works](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution) and where it breaks down in low-signal environments, see our attribution guide.
The simplest version is a geo-based holdout: run ads in one comparable market, pause in another, compare organic conversion. This requires no specialized tooling and gives directional data on whether a campaign is actually driving subscriptions or just capturing existing intent.
- **Modeled LTV replaces D7 ROAS as the optimization target** SKAN's delayed and aggregated signals make early-window revenue metrics unreliable as standalone decisions. Estimating eventual subscription value from early behavioral signals, such as session depth and feature engagement in Days 1-3, gives a more accurate picture of which campaigns are worth scaling.
- **Self-reported attribution adds qualitative signal** that performance tracking routinely misses. Asking users "How did you hear about us?" during onboarding captures word-of-mouth, organic social, and influencer activity that [attribution windows](https://www.airbridge.io/glossary/attribution-window) cannot credit. Correlating those responses with actual [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) reveals which channels your models are underweighting.
## FAQs
**Does ATT affect Android campaigns?**
No. ATT is specific to iOS. Android uses its own evolving privacy framework (Privacy Sandbox for Android), which currently allows more user-level measurement than iOS. This is a significant reason Android paid installs grew **+57% YoY** globally. Android-first strategies are standard practice for subscription apps that need both measurement flexibility and scale.
**Is SKAN 4.0 enough to measure subscription campaigns?**
SKAN 4.0 gives more data than earlier versions, but it has real limits: postback delays up to 35 days, aggregated signals rather than user-level data, and volume thresholds before granular postbacks unlock. Most subscription app teams use SKAN as one input inside a blended measurement model that also includes incrementality testing, modeled cohort LTV, and self-reported attribution. SKAN alone is not sufficient for budget allocation decisions.
**What happened to fingerprinting?**
Apple explicitly prohibits fingerprinting (combining device signals to identify users without consent). Apps using it risk App Store rejection. The industry has moved to compliant methods: SKAN postbacks, modeled attribution, and first-party data collected directly from users during onboarding.
## The Teams Winning Privacy-Era UA Are Already Running Differently
The shift since iOS 14.5 is not a single change you adapt to once. It is a structural reset in how mobile performance marketing works. The teams doing well are testing more creative variants at higher volume, measuring in cohorts and holdouts rather than individual attribution, expanding into Android-first markets, and treating their ATT consent flow as a media buying decision.
The fundamentals of subscription app marketing did not change. Strong onboarding, a paywall that appears at the right moment, and ad spend mapped to [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) rather than installs still define who builds a sustainable business. What changed is how clearly you can see those results, and how quickly you can act on what you see.
Without a [measurement layer](https://www.airbridge.io/en/blog/attribution-tool-mmp) that connects these fragmented signals, you end up comparing dashboards where every network claims credit for the same install. That gap between what each platform reports and what actually drove the subscription is where budget gets wasted.
**For subscription apps that want to connect ad spend to subscription revenue across channels,** [**start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core) **— 15K attributed installs included.**
---
## Social Proof for Apps: When and Where to Place Reviews, Ratings, and Trust Signals
URL: https://www.airbridge.io/en/blog/social-proof-for-apps
Language: English
Published: 2026-05-13
Author: Harper (Trang Nguyen)
Most subscription apps lose users between download and the paywall, not because the product is weak, but because trust hasn't been established yet. The average app converts just **1.9%** of downloads to paying subscribers within **35** days. Apps in the upper quartile hit **4.3%+**. The gap between those two outcomes is rarely pricing or features. It's almost always trust.
Social proof for app conversion closes that gap, but placement matters more than presence. A testimonial on the wrong screen, at the wrong moment in the user journey, adds noise rather than credibility. This guide covers exactly when and where to place reviews, ratings, and trust signals to improve social proof for app conversion at every stage of the onboarding-to-paywall flow.
## Key Takeaways
- **82% of trial decisions happen on Day 0.** The screens between download and your first paywall are your highest-leverage conversion window.
- **Goal-matched testimonials outperform generic ones.** A review that mirrors what the user just said they want ("I lost 18 lbs in 8 weeks") creates far more trust than "I love this app."
- **Trial-inclusive paywalls convert at 64.5%** vs **44.4%** for text-only paywalls. Reviews and user counts embedded in the paywall are direct conversion levers, not just design choices.
- **Products with 5+ reviews are 270% more likely to be purchased.** App Store rating and review volume build trust before users ever open your app.
- **Purchase likelihood peaks at 4.2 to 4.5 stars, not 5.0.** A perfect rating reads as curated. A 4.3 with 10,000+ reviews reads as genuine.
## The Three Trust Moments That Drive App Conversion
Social proof is any signal from other users, including ratings, reviews, testimonials, or user counts, that helps a new user trust the product enough to subscribe. In the context of social proof and app conversion, it is not a single design element but a layered strategy built around three distinct moments in the user journey.
- **Pre-install trust** happens on your App Store listing, before the download. Users evaluate your rating, review count, and screenshots to decide whether the app is worth trying.
- **Onboarding trust** happens in the first two to three minutes after download. The user has committed enough to install but hasn't committed to paying. This is where personalized testimonials and user count signals do their most important work.
- **Paywall trust** happens at the conversion moment itself. The user has seen your product and is deciding whether to start a trial or pay. The social proof on this screen must directly answer the objection standing between them and subscribing.
A key principle from StartApp School's paywall optimization curriculum (Lesson 02) captures the underlying logic: users convert when Motivation, Trust, and Friction all align. "Great monetization doesn't convince people to want something. It catches them at a moment they already do." Timing matters more than the quality of the testimonial copy.
[Social Proof for Apps_1.webp]
## Onboarding: Where to Place Social Proof Before the Paywall
**82% of trial decisions happen on Day 0**, which means the onboarding sequence is your highest-leverage app conversion window, and where social proof does its most important work. The paywall is the destination, but the journey leading to it either builds or destroys the trust that determines whether users convert.
The most common mistake is placing social proof before the user has invested anything in the experience. A splash screen testimonial on the very first screen reads as a brand claim, not peer validation. Moved to the right moment, the same quote becomes a genuine trust signal.
### 1\. Collect a goal first, then show a matching testimonial
The best onboarding flows ask users what they are trying to achieve before displaying any social proof. The moment a user selects their goal, you know exactly which testimonial will resonate.
A fitness app user who picks "lose weight" should see a quote from someone who lost weight, not a generic star rating and not a muscle-building success story. A specific, outcome-matched quote like "Down 22 lbs in 10 weeks, after three failed attempts with other apps" transforms social proof from generic marketing into personalized confirmation. The user reads it as: _other people with my exact goal used this and got results._
### 2\. Use loading screens as a pre-paywall trust layer
Pre-paywall loading screens, such as the "Analyzing your goals" or "Building your plan" animations that appear before the paywall, are the most underused real estate in mobile onboarding. Most teams treat them as neutral wait states. High-converting apps treat them as a final trust layer.
Effective loading screen social proof includes:
- Specific user counts, such as "Used by **247,382** people this month" rather than "Used by thousands"
- Average outcome claims with a time frame, such as "Users report first results in 21 days on average"
- Rating signals placed in the flow, such as "4.8 stars from **41,000+** reviews"
The timing is nearly optimal. Motivation is at its highest because the user just invested effort stating their goal. The loading screen creates no friction. Trust signals here prime the user to evaluate the paywall as a confirmation rather than a cold ask.
### 3\. Match the number format to the claim
Precision builds credibility. "Join **47,392** users" outperforms "Join thousands of users" because the specific number feels verifiable. Vague social signals read as unsubstantiated marketing. Concrete numbers read as data. For teams learning how to add reviews to app onboarding for the first time, this specificity principle is the highest-leverage place to start.
## Paywall Social Proof That Converts
The paywall is where social proof carries its most direct app conversion impact. Trial-inclusive paywalls convert at **64.5%**, compared to **44.4%** for visual or text-only paywalls. The difference is not design quality. It is trust density.
OMENA redesigned their paywall into a scrollable screen with testimonials, a founding story, user photos, and a FAQ. It doubled their trial start rate (Botsi x Retention.Blog, 48 Laws of Subscription App Success, Law 16). The content wasn't new. The placement and density of trust signals was.
The optimal paywall social proof sequence:
| Paywall Element | Recommended Position | What It Communicates |
| --- | --- | --- |
| Star rating + review count | Top, near headline | Scale and credibility |
| Goal-matched testimonial | Middle, below plan options | Outcome proof |
| User count ("Join X members") | Below headline or near CTA | Community validation |
| Trust badge or media mention | Footer | Authority and legitimacy |
### 1\. Lead with the rating, not the price
The first visible element shapes how users interpret everything that follows. If a user sees a price first, they evaluate cost. If they see "4.7 stars from 52,000 reviews" first, they evaluate trustworthiness and then assess the price within that trusted context.
This sequencing change is low-effort and frequently produces measurable lift in paywall optimization experiments. The rating frames the price as reasonable given what users say, rather than expensive at face value.
### 2\. Show the right rating, not a perfect one
Purchase likelihood peaks at around **4.2 to 4.5 stars** ([Spiegel Research Center / PowerReviews, 2017](https://spiegel.medill.northwestern.edu/how-online-reviews-influence-sales/)). A perfect 5.0 activates skepticism, especially when accompanied by a low review count. Consumers recognize that authentic, at-scale products accumulate some critical feedback. A 4.3 with **18,000** reviews signals a real product with a statistically valid sample, and that signal converts.
If your rating is currently 3.8 or below, don't feature it prominently on the paywall. Lead with user counts, outcome statistics, or press mentions instead, and prioritize improving the rating before making it a conversion lever.
### 3\. Answer the purchase objection with the testimonial
Every category has a predictable paywall objection. For fitness apps: "Will this actually work for my specific situation?" For productivity apps: "Is this worth a monthly fee if I get busy?" The most effective paywall testimonial directly answers that objection.
Not "This app changed my life," but "I've tried four other apps. This is the only one I've actually stuck with past week two." Specificity is the trust mechanism. Vague praise creates no resonance. The testimonial that addresses the exact fear the user is feeling at the paywall moment is the one that converts.
[Social Proof for Apps_2.webp]
Read more: [Top 3 paywall builder solutions for subscription apps](https://www.airbridge.io/en/blog/top-3-paywall-builder-solutions-for-subscription-apps)
## App Store Ratings and Pre-Install Trust
App conversion starts before the app is downloaded, which is why your App Store presence is a growth lever, not just a maintenance task. **93%** of consumers say online reviews influence their purchase decisions ([PowerReviews, 2023](https://www.powerreviews.com/power-of-reviews-2023/)). Products with **5** or more reviews are **270%** more likely to be purchased than those with zero ([Northwestern University Spiegel Research Center](https://spiegel.medill.northwestern.edu/)).
For subscription apps, the App Store listing is the first trust gate in the funnel. Three variables determine how much conversion lift it generates:
1. **Rating score.** Aim for **4.2** or higher. Below **4.0** in a competitive category, the rating becomes a liability before users even open the app. Don't surface a low rating prominently. Lead with user count or press mentions instead while you work to improve it.
2. **Review freshness.** **64%** of consumers prefer a smaller number of recent reviews over a large volume of old ones ([PowerReviews](https://www.powerreviews.com/review-volume-and-recency/)). A **4.8** rating built on 18-month-old reviews performs below a **4.4** with reviews from this month. The best trigger for a review prompt is immediately after a user achieves a meaningful outcome, not on a fixed timer or after a set number of app opens.
3. **Review volume.** "4.6 stars (2,300 ratings)" and "4.6 stars (47 ratings)" carry very different trust signals. Review volume is a marketing metric, not just a product metric. For more on how App Store signals connect to [lifetime value](https://www.airbridge.io/en/glossary/lifetime-value-ltv), top-performing apps treat ASO as a top-of-funnel conversion lever that compounds with onboarding and paywall optimization.
## Common Mistakes and Early-Stage Solutions
**Generic testimonials that don't match user goals.** "This app is incredible" says nothing about what the user is trying to accomplish. It registers as a brand claim. The fix is a small library of goal-segmented quotes matched to the goal-selection screen at the start of onboarding. For a fitness app with five goal options, five sets of testimonials is the minimum viable starting point.
**Showing social proof before any user investment.** A testimonial on the first screen, before personalization, reads as marketing. Moved to the post-goal-survey or loading screen, the same testimonial reads as peer validation. The user's emotional state at the moment of exposure determines how the signal is interpreted. Getting the sequence right matters more than improving the testimonial copy.
**High ratings with low review volume.** A 5.0 rating on 11 reviews triggers skepticism, not confidence. As volume grows past **50 to 100** reviews, the rating signal becomes credible enough to feature on the paywall. Until then, deprioritize the rating display and lead with user count or outcome statistics.
For teams just launching, the path forward is to use what is available rather than wait for scale. Here are four trust signals you can build from day one:
1. **Lead with user count, not review count.** If you have 2,000 users but only 20 reviews, "Trusted by 2,000 people" is a more credible signal than surfacing those 20 reviews on your paywall.
2. **Gather targeted testimonials from beta users.** Reach out to 10 to 15 early users who got results. Ask specifically: "What did you set out to accomplish, and what actually happened?" A handful of specific, goal-outcome quotes outperforms hundreds of generic ratings.
3. **Use outcome-based product data.** "63% of users who reach Day 7 are still active at Day 30" is a compelling trust signal derived entirely from your own analytics, with no external reviews needed.
4. **Cite press or partner mentions.** Any publication, newsletter, or app store editorial feature counts. "Featured in \[publication\]" signals external credibility, especially for users who don't find peer reviews sufficient on their own.
For more on building the lifecycle signals that reinforce retention over time, see how leading apps structure [subscription lifecycle messaging](https://www.airbridge.io/en/blog/subscription-app-lifecycle-marketing).
## Trust Is Built Screen by Screen
Social proof that converts is not a single well-placed testimonial. It is a sequence of trust signals, each calibrated to where the user is in the journey: goal-matched quotes after the intake survey, specific user counts on the pre-paywall loading screen, a rating and outcome testimonial at the top of the paywall, and fresh App Store reviews that build confidence before anyone even installs.
**82%** of trial decisions happen on the day of install. The onboarding-to-paywall window is where most of your social proof app conversion opportunity lives, and where most teams leave trust gaps unfilled.
Start with one change: segment your testimonials by the goal the user selects at onboarding. Measure the impact on your trial start rate. Then layer in the loading screen signal, then the paywall sequence. Each layer compounds the one before it.
## See Which Channels Actually Convert Through Your Paywall
Once your social proof placement is optimized, a follow-up question becomes urgent: which acquisition channels are bringing users who actually respond to your onboarding and convert to paid, and which ones aren't?
A user from a Meta fitness ad might convert through your goal-matched [paywall](https://www.airbridge.io/en/glossary/paywall) at **45%**. A user from a broad-audience campaign might convert at **12%**. Same paywall, same social proof, very different users. Without subscription-level data broken down by channel, you're averaging those two numbers together and calling it your "conversion rate." You can't tell whether your social proof app conversion optimization is working for the users it's meant to reach, or just inflating trial starts from users who churn before the first renewal.
The goal isn't just a higher trial start rate. It's more paying subscribers from the channels you're spending on. Those are two different metrics, and only one of them tells you if your social proof is actually working.
This is where connecting ad spend to subscription revenue becomes essential. Tools like [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) link your paid acquisition channels to your subscription events, including trial starts, paid conversions, and renewals. You can see which campaigns drive users who actually subscribe, not just which ones drive installs. It's free to start, with **15,000** attributed installs included in the base plan.
---
## When to Try Web-to-App: A Readiness Framework for Subscription Apps
URL: https://www.airbridge.io/en/blog/when-to-try-web-to-app
Language: English
Published: 2026-05-13
Author: Harper (Trang Nguyen)
Web-to-app funnels are one of the most talked-about growth strategies in subscription apps right now. The promise is real: bypass app store fees, capture better attribution data, and own the entire paywall experience from first impression to subscription.
But compelling strategy and right-for-my-stage are two different things. Most early-stage subscription teams ask some version of this question: "We have heard web-to-app works for bigger apps. Does it work for us, right now?"
**Key Takeaways**
- **Web-to-app is a scaling tool, not a launch tool.** It amplifies a working in-app funnel. It does not replace a broken one.
- **The data shows a real trade-off.** Web subscribers show **84.5%** Month 1 retention vs. **48.2%** for in-app, but that advantage flips by Month 8 (**20%** web vs. **30%** in-app).
- **Fee savings are real, but LTV often closes the gap.** Paying 2-3% to a payment processor beats the app stores' 15-30% cut, but lower web LTV frequently offsets that gain.
- **Small teams face a real capacity constraint.** Maintaining two funnels means two optimization backlogs, two experiment queues, and two performance metrics to monitor in parallel.
- **Attribution gets more complex.** Connecting a web purchase to an in-app subscription event requires UTM parameters, deep links, and a measurement layer that joins both sides of the journey.
## What Is a Web-to-App Funnel?
A web-to-app funnel is a marketing and monetization strategy in which users complete onboarding and payment on the web before downloading the app. Instead of sending users directly to the App Store or Google Play, the subscription transaction happens in the browser first. The app then becomes the delivery mechanism for a product the user has already paid for.
This is the opposite of the traditional direct-to-app model, where users click an ad, land in the app store, install the app, move through onboarding, hit a paywall, and subscribe. In a web-to-app flow, that sequence is reversed: onboarding, decision, and payment happen before the install ever occurs.
**A typical web-to-app journey looks like this:**
1. User sees a paid ad on Meta, TikTok, or Google.
2. User lands on a web page, often a personalized quiz or landing page.
3. User completes web onboarding and views a pricing page.
4. User subscribes through a web payment processor.
5. User receives a deep link or email to download and activate the app.
The model offers three practical advantages. [App stores take 15-30% of every subscription transaction](https://www.revenuecat.com/blog/engineering/small-business-program/); web processors typically charge 2-5%, keeping more revenue in the business. That said, fee savings alone do not guarantee better net revenue. A [RevenueCat head-to-head conversion test](https://www.revenuecat.com/blog/growth/iap-vs-web-purchases-conversion-test/) found web purchases resulted in 6% lower take-home revenue compared to in-app, driven by lower initial conversion rates. The economics work when long-term LTV gains outweigh the conversion gap, not as a simple fee swap. Apple's ATT update in 2021 also degraded signal quality for in-app campaigns, while web flows restore first-party data and give ad algorithms cleaner signals. Finally, users who navigate a web onboarding flow before installing tend to arrive more committed, which supports early retention.
| Advantage | What it means |
| --- | --- |
| Lower platform fees | 2-5% web processing vs. 15-30% app store commission |
| Restored attribution | First-party web data bypasses ATT signal loss |
| Higher-intent users | Longer onboarding filters for committed buyers |
| Monetization freedom | Lifetime deals, bundles, and upsells without store restrictions |
Web-to-app performs best in categories where personalization matters: health and fitness, education, and productivity. Apps like Noom and BetterMe have used this model to build quiz-based onboarding funnels where personalization drives both higher conversion and long-term retention. Categories where the install experience is the product itself, like most mobile games, tend to see weaker results. The term is also used loosely; technically, paying on the web before installing is "web-to-web," but most teams use "web-to-app" for both. The core logic is the same either way: the web handles persuasion and payment, the app delivers the product.
## 3 Signals You're Ready for Web-to-App
Not every subscription app should build this funnel. Here are the signals that suggest the timing makes sense.
### 1\. Your in-app trial-to-paid rate is at or above category median
Web-to-app is not a remedy for a weak in-app funnel. If your current trial-to-paid conversion is underperforming, the root cause is almost always in your product, onboarding, or paywall design, not your payment channel. Adding a web layer to a leaky in-app experience gives you two leaky funnels to manage.
Among more than 75,000 subscription apps analyzed, the category median [trial-to-paid conversion rate is **34.8%**](https://www.revenuecat.com/state-of-subscription-apps-2025/), with top-quartile apps reaching **51.5%**. At or above that median, your product has validated its core value. Web-to-app then becomes a scaling question, not a product question.
### 2\. You're spending $5K–$10K+ per month on paid UA
Web-to-app adds real infrastructure: landing pages, payment processing, deep links, cross-environment attribution, and an ongoing experiment queue for both funnels. That overhead makes sense when you have enough paid UA volume to run meaningful experiments.
Below that spend level, the engineering cost tends to outweigh the payoff. Your time and budget go further optimizing your in-app paywall and running one paid channel consistently. Without sufficient volume, web vs. app cohort comparisons also lack statistical significance, meaning you cannot reliably draw conclusions from the data even if you collect it.
### 3\. You need billing control or serve B2B customers
[App stores take **15-30%** of every subscription transaction](https://www.revenuecat.com/blog/engineering/small-business-program/). Web payment processors typically charge **2-3%**. Note that apps on Apple's Small Business Program already pay a reduced 15%, and commission structures are actively shifting following Epic v. Apple regulatory decisions in 2025. Final fee structures for external payments remain undetermined, so model the economics against your specific tier before assuming full savings.
This especially applies to Business, Productivity, and Utility apps, where enterprise buyers often cannot pay through a personal Apple or Google account. Web billing removes that friction entirely. The data makes the case: subscription apps using web billing in Business categories show a median predicted LTV of **$255**, compared to **$27** for in-app annual subscribers in the same category — a 9.4x multiple. For Productivity apps, web billing P90 LTV reaches **$1,741**.
## 3 Signals to Wait on Web-to-App
### 1\. You haven't reached product-market fit yet
Before product-market fit, the priority is learning fast. Web-to-app introduces significant stack complexity, and that complexity slows the iteration speed you need when you are still validating your core value proposition. A team that builds a web funnel before confirming their paywall converts is taking on technical debt without a proven return.
### 2\. Free trials are your primary conversion mechanism
If a standard free trial is your core offer, web-to-app complicates the model. Web users sometimes enter card details tied to cards with no active balance, which breaks the free trial flow. Many teams work around this with a low-cost intro offer ($0.99) on web instead of a true free trial, though this adds another variable to manage and makes comparing web vs. in-app performance more difficult.
### 3\. You're a small team managing multiple priorities
Getting a web-to-app funnel built from scratch typically takes four to eight weeks. Dedicated tools like [RevenueCat Funnels](https://www.revenuecat.com/feature/funnels) or FunnelFox have shortened this to days, though payment integration, deep linking, and ongoing testing still require sustained effort. As Andrey Shakhmin, CEO of web-to-app platform Web2App, notes: "Unlike in-app where Apple handles payment directly, web requires apps to manage payment infrastructure themselves, creating high barriers to entry."
For a team of one to five people, running both simultaneously spreads focus thin. Most small teams get a better return from picking one funnel, getting it healthy, then expanding.
## What the Data Says: Web vs. In-App
The performance data on web vs. in-app paywalls shows a trade-off that does not resolve neatly in either direction.
| Metric | In-App Paywall | Web Paywall |
| --- | --- | --- |
| Conversion rate | 1.65% | 1.13% |
| Month 1 retention | 48.2% | 84.5% |
| Month 8 retention | 30% | 20% |
| Average LTV | $40.1 | $10.8 |
_Source:_ [_Adapty State of In-App Subscriptions 2026_](https://adapty.io/state-of-in-app-subscriptions/) _— 16,000+ apps, $3B revenue._
The web paywall's Month 1 retention looks strong at **84.5%** vs. **48.2%** for in-app. But that advantage erodes fast. By Month 8, in-app leads at **30%** vs. **20%** for web. According to the [Adapty 2026 benchmark data](https://adapty.io/state-of-in-app-subscriptions/), even after adjusting for the 30% App Store commission savings, web [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) still runs **$4 lower** on average.
This does not mean web-to-app is worse. It means web-to-app serves a different user: someone who needs more context before committing and tends to arrive with higher purchase intent after a longer funnel. The economics depend on your vertical, price point, and how well the web funnel is built.
### Readiness Checklist
Before committing engineering time, run through these five signals.
| Signal | Ready | Not Yet |
| --- | --- | --- |
| In-app trial-to-paid rate | At or above 34.8% median | Below category median |
| Monthly paid UA spend | $5K–$10K+ | Under $5K |
| Team capacity | Can build and maintain two funnels | 1-5 person team, fully allocated |
| Billing model | Annual or high-price plans | Heavy free-trial reliance |
| Attribution setup | Measurement tool already in place | Tracking installs only |
## FAQs
### Does web-to-app work for apps with under 10,000 monthly downloads?
It can work, but the ROI is harder to justify at low volume. The engineering cost of a web funnel is mostly fixed regardless of download count. At low volume, that fixed cost claims a larger share of team capacity. Most small apps see more return from sharpening in-app conversion and mastering one paid channel before layering web infrastructure on top.
### Will running a web-to-app funnel hurt my App Store organic discovery?
Not directly. Web-to-app users bypass the App Store's native install flow but often arrive with more context and higher intent. Strong engagement from these users (deeper sessions, better retention) can improve category rank signals over time. The risk runs the other direction: if your web funnel attracts low-intent traffic, weaker engagement metrics can hurt organic visibility. Funnel quality matters more than channel.
### How do I track whether my web-to-app funnel is working?
Setup is more complex than in-app, but the data you get is more transparent. When a user clicks a paid ad, visits your web page, subscribes on the web, and then opens the app, you need to connect all four touchpoints into one journey. That requires UTM parameters on your web URLs, a [deep link](https://www.airbridge.io/glossary/deep-link) that carries the user from web to app, and a measurement layer that joins the web subscription event with in-app activity. Without it, your web LTV and in-app LTV appear as separate numbers, and you cannot know which ad drove which subscriber.
The upside is that once configured, web tracking provides more signal than in-app. Unlike in-app campaigns constrained by Apple's ATT and SKAdNetwork, web funnels give you full access to cookies and UTM parameters, allowing you to trace the complete customer journey without privacy-driven data gaps. This is one of the strongest structural advantages of the web-to-app model for performance marketers.
If you need to connect paid UA spend to subscription events across web and app, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) tracks the full journey from web purchase to in-app subscription, starting free with 15K installs.
## Build the Foundation Before You Add the Layer
Web-to-app is a legitimate growth lever, but it is one you pull after building the foundation, not before. The teams getting the most from it have already validated their in-app paywall, have enough paid UA volume to run real experiments, and have the capacity to maintain both paths.
If you are still iterating on paywall conversion or running your first paid campaign, the in-app foundation delivers more return. Web-to-app will still be there when your metrics say you are ready.
The uncertainty around platform fees and regulatory changes should not be a reason to delay indefinitely, either. As app growth practitioner Cristian Rotari puts it: "Apple is being ambiguous on purpose. Factor in the risks, go ahead and test." Start with a strong in-app foundation, build the readiness signals, and expand to web when the data supports it.
To explore more web-to-app strategies, see [Web-to-App vs. App-to-Web: Why Hybrid Funnels Win](https://www.airbridge.io/blog/web-to-app-vs-app-to-web).
---
## App Pricing Experiments: How to Test What Users Will Pay
URL: https://www.airbridge.io/en/blog/app-pricing-test
Language: English
Published: 2026-05-12
Author: Luke
You set a price. Maybe you looked at a few competitors, settled on something in the range, and shipped. But **app pricing** is not a one-time decision. It is a hypothesis you test. And **34.6% of subscription cancellations are cost-related** ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)), the second most common churn reason right after not using the app enough. Pricing doesn't just affect who signs up. It affects whether they stay.
This guide walks through how to test willingness to pay for a subscription app at every stage of growth, from before your first install to after you've hit consistent monthly volume. There are three practical methods, and the right one depends on where you are right now.
> **Willingness to pay is the maximum price a user will accept before declining a subscription offer, and it's measurable before you commit to a price.**
## Key Takeaways
- **Higher prices often convert better, not worse.** Apps in the high-price tier show stronger Day 35 conversion because they filter for more committed users, not because they're better built.
- **App pricing experiments deliver outsized returns.** Teams that run structured tests achieve up to 40x more revenue than teams who set a price and never revisit it.
- **There are three methods to test willingness to pay:** survey research before launch, behavioral signals with early users, and live A/B testing at scale.
- **Conversion rate is the wrong primary metric.** ARPU (average revenue per user) balances conversion, price, and plan mix into a single number that actually predicts revenue.
- **Your first price is a hypothesis.** The goal of pricing experiments is to keep learning, not to find one number and stop.
## Your Launch Price Is a Hypothesis
Most founders skip app pricing experiments entirely at launch. They benchmark competitors, pick something near the middle, and move on. The problem is that benchmarking tells you what others are charging, not what your specific users will pay for your specific value. Those are different questions.
Here's the counterintuitive reality: **higher-priced apps often convert better, not worse.** Across more than **75,000** subscription apps analyzed, high-priced apps report a **Day 35 download-to-paid conversion of 2.7%**, compared to **1.5%** for low-priced apps (RevenueCat State of Subscription Apps 2025). The reason is intent. Someone willing to pay more has already decided the problem is serious enough to solve. A lower price attracts a broader audience, but a less committed one.
The gap compounds over time. High-priced apps show a median Year 1 LTV (lifetime value) of **$55.21 per paying subscriber**, approximately **7x** the LTV of low-priced apps (RevenueCat State of Subscription Apps 2025). That's not just because the price itself is higher. It's because the price attracts users who extract real value from the product.
The question to ask isn't "will users pay this?" It's "what price attracts the users most likely to stay?"
## Method 1: Test Before You Launch
Survey-based methods are the most accessible way to test willingness to pay for a subscription app before you have installs to work with. They give you directional signal before you commit.
### 1\. Van Westendorp Price Sensitivity Meter
The [Van Westendorp Price Sensitivity Meter](https://en.wikipedia.org/wiki/Van_Westendorp%27s_Price_Sensitivity_Meter) asks four questions to a representative sample of your target users:
1. At what price would this app be so expensive you wouldn't consider it?
2. At what price would it seem so cheap you'd question the quality?
3. At what price would it start to feel expensive, but you'd still consider it?
4. At what price would it feel like a bargain?
Plot the response curves. The intersections define your acceptable price range. The zone between the "too cheap" crossover and the "getting expensive" crossover is where users consider your price both credible and reasonable. Use this to avoid entering the market outside the range your users will take seriously.
### 2\. Gabor-Granger Pricing Survey
Gabor-Granger asks respondents whether they would subscribe at a specific price, repeated across several price points (for example: **$4.99**, **$7.99**, **$12.99**, **$19.99** per month) with different respondent groups. The result is a demand curve showing how purchase intent drops as price increases.
Stated intent over-estimates real behavior. Use Gabor-Granger for relative ranking — understanding which price threshold causes the sharpest drop in intent — not as an absolute predictor of conversion.
### 3\. User Interviews Focused on Jobs-to-Be-Done
Surveys measure stated willingness to pay. Interviews uncover what's driving it. **Emotional jobs drive higher willingness to pay than functional ones. Urgency and perceived consequence are stronger signals than how often someone uses the app.**
Avoid asking "how much would you pay?" directly. Users anchor to low numbers. Instead, ask:
- "What happens if this problem doesn't get solved?"
- "Have you tried other solutions? What did those cost?"
- "Where does this fit in your monthly spending?"
The answers tell you whether your app is solving a vitamin problem (nice to have) or a painkiller problem (must solve). Painkiller positioning justifies higher prices and retains subscribers longer.
[Screenshot 2026-05-12 at 19.05.53.png]
## Method 2: Test With Behavioral Signals
Surveys tell you what users say they'll do. Behavioral methods measure what they actually do when money is involved. These work when you have some users but not yet enough volume for a statistically valid A/B test.
### 1\. The Fake-Door Price Test
Build two or three landing pages with identical product descriptions but different price points. Run small paid campaigns to each page, targeting **200 to 500** visitors per variant. Measure click-through rate on the subscribe CTA. The relative differences in intent across pages reveal which price anchor is resonating.
### 2\. Pre-Selling Before Launch
Selling before you build, or before you've reached scale, is one of the strongest willingness-to-pay signals available. If users pay upfront for early access at your intended annual price, you've confirmed both the problem and the price in one step. A paid beta with as few as 20 to 50 users produces more actionable information than any survey.
### 3\. Hard Paywall for a Segment
If your app is live but you haven't committed to a price, run a hard paywall (no free trial) for a portion of new users. Hard paywalls report a **median Day 35 download-to-paid conversion of 12.1%**, compared to **2.2%** for freemium apps (RevenueCat State of Subscription Apps 2025). The conversion is higher because only users who genuinely see value pay without trying first.
Hard paywall subscribers also show higher long-term revenue: a median Year 1 LTV of **$49.30**, compared to **$24.24** for freemium users (RevenueCat State of Subscription Apps 2025). The tradeoff is a higher refund rate, **5.8%** versus **3.4%** for freemium. As a short-term pricing signal on a segment of users, the directional data is worth it.
## Method 3: Live App Pricing Experiments at Scale
Once you have consistent install volume (typically 1,000+ per month), live A/B testing is the most reliable method for app pricing experiments. You're measuring actual subscription behavior, not intent, not proxies.
### 1\. What to Test
Start with price point only. Don't change trial length, plan structure, and price simultaneously. If three variables change at once, you can't isolate which one moved the result.
A practical starting range: test your current price against **+25% and -25% variants** ([Purchasely, In-App Subscription Price Testing](https://www.purchasely.com/blog/in-app-subscription-price-testing)). If your monthly plan is **$9.99**, test **$7.49** and **$12.99**.
It's also worth prioritizing plan and trial structure experiments over visual or copy changes. Pricing and plan experiments improve LTV at nearly **2x** the rate of paywall visual or copy changes ([Adapty State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/)), a strong argument for testing price before testing button color.
### 2\. How to Set Up the Test Correctly
**Calculate sample size before you start.** For a baseline trial-to-paid rate around **30%** and a 10% relative lift, you typically need **2,500 to 3,500 users per variant** to reach 95% confidence (use a [sample size calculator](https://www.evanmiller.org/ab-testing/sample-size.html) to get the exact number for your baseline). Run the test to completion even if one variant leads early.
**Test one variable at a time.** Changing price and trial length together makes it impossible to know which change moved the result. This is the most violated rule in pricing experimentation.
**Configure for subscriptions, not trials.** Many A/B testing setups default to trial starts as the success event. Make sure your experiment tracks paid subscriptions, not just the top of the funnel. See [RevenueCat's guide to mobile app price testing](https://www.revenuecat.com/blog/engineering/price-testing-for-mobile-apps/) for a platform-specific walkthrough.
### 3\. Platform Tools
App Store Connect, Google Play Console, RevenueCat Experiments, and Purchasely all support live price tests natively. Note that [Apple limits subscription price increases to once per year](https://support.apple.com/en-us/109501) for existing subscribers without requiring re-consent. Factor this constraint into how aggressively you test upward.
## Reading Results: The Metric That Matters
The key rule for reading app pricing experiments: conversion rate alone is a trap. A lower price almost always converts better. That tells you nothing useful about revenue.
**The correct primary metric is ARPU (average revenue per user).** ARPU accounts for conversion rate, the actual price paid, and plan mix in a single number. If you want to track how ARPU compounds into long-term value, see how [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) is calculated across plan types. If your **$7.49** variant converts at 42% but your **$12.99** variant converts at 28%, ARPU shows which variant generates more revenue per 100 users entering the funnel.
The data supports this: **price experiments improve conversion with only 28% probability, but improve LTV with 46% probability** (Adapty State of In-App Subscriptions 2026). **Teams running app pricing experiments achieve up to 40x more revenue than those who don't** (Adapty, 2026).
Beyond ARPU, track two secondary metrics. **First renewal rate** shows whether the subscribed cohort continues past the first billing cycle, the earliest signal of whether price is aligned with perceived value. **90-day retention** reveals whether the price attracted committed users or just any users. A price test that improves ARPU but collapses 90-day retention is optimizing short-term revenue at the cost of long-term subscribers.
| Metric | What It Measures | When It Misleads |
| --- | --- | --- |
| Conversion rate | How many users start a subscription | Shows lower price as winner even when it generates less revenue |
| ARPU | Revenue per user entering the funnel | Primary metric. Use this to decide. |
| First renewal rate | Price-to-value alignment | Leading indicator for subscriber quality |
| 90-day retention | Long-term subscriber quality | Takes longer but reveals whether you attracted the right users |
[Screenshot 2026-05-12 at 19.06.50.png]
## Common Pricing Experiment Mistakes
**Treating one result as final.** App pricing experiments answer a question about your current product, your current acquisition mix, and your current paywall. Any of those change, and the optimal price may shift. The best teams run pricing experiments continuously, not once at launch.
**Measuring too early.** Nearly **30%** of annual subscriptions cancel within the first month (RevenueCat State of Subscription Apps 2025). If you measure success at Day 7 conversion, you'll miss this early churn entirely. Extend your measurement window to at least 30 days before drawing conclusions from any pricing test.
**Assuming discounts help.** **9 out of 10** subscriptions sell at full price, with only **10%** using discounts of any kind (Adapty State of In-App Subscriptions 2026). Discounting rarely becomes necessary if you've set the right price. It's more often a sign the value framing hasn't caught up with the price point.
**Reading results on one platform only.** iOS captures **85%** of subscription revenue globally, and annual-plan revenue per subscriber runs **7.7x** higher on iOS than Android (Adapty State of In-App Subscriptions 2026). A pricing change that looks flat across platforms can be hiding a significant iOS win or loss. Segment results by platform before making a final decision.
**Changing price without changing value framing.** A price increase not accompanied by a shift in how you communicate value tends to fail. **Pricing stability signals confidence. Constant price changes erode trust.** When you move a price up, move the perceived value framing up with it.
## What This Looks Like in Practice
Here's how a founder moves through all three stages with a fitness app.
**Stage 1 (before launch):** She runs a Van Westendorp survey with **52** potential users recruited from a fitness subreddit. Results show an acceptable price range of **$7.50 to $15.00** per month, with a psychological anchor clustered around **$9.99**. She launches at **$9.99** monthly with a 7-day free trial.
**Stage 2 (early traction):** At around 400 installs per month, volume is too low for a clean A/B test. She runs five user interviews instead, asking about the consequences of not sticking to a workout routine. Three of five respondents mention accountability as the real pain, not the workouts themselves. She updates her paywall copy to center accountability, not features.
**Stage 3 (at scale):** By month four, she's averaging 1,400 installs per month. She runs a 6-week A/B test: **$9.99** versus **$12.99**, one variable only.
| Variant | Trial-to-Paid Conversion | ARPU | 90-Day Retention | First Renewal Rate |
| --- | --- | --- | --- | --- |
| $9.99 / month | 33% | $3.30 | 59% | 52% |
| $12.99 / month | 26% | $3.64 | 71% | 66% |
_These numbers are illustrative. Your results will vary by category and audience._
Reading only conversion rate, the **$9.99** variant wins by 7 percentage points. Reading ARPU, the **$12.99** variant generates **10% more revenue per user**. Reading 90-day retention, the **$12.99** cohort stays **12 percentage points longer.** She moves to **$12.99**, then schedules the next test: **$12.99** versus **$15.99**. That's the loop.
## Pricing Is Where Product Confidence Lives
The anxiety behind "is my price right?" is usually anxiety about whether the product is worth it. Pricing experiments don't just find a better number. They reveal whether users believe your app solves a real problem at real stakes. If your tests repeatedly show that users won't pay above a low threshold, the finding isn't that you need a lower price. It may mean you need a sharper value proposition, a clearer paywall, or a more precisely defined target user.
Run app pricing experiments at every stage, read them by ARPU and retention, and treat each result as input to the next iteration.
---
## How to Research Competitor App Ads for Free
URL: https://www.airbridge.io/en/blog/how-to-research-competitor-app-ads-for-free
Language: English
Published: 2026-05-11
Author: Harper (Trang Nguyen)
Your competitors are spending money on ads right now. The creatives they're running, the hooks they're testing, the platforms they're doubling down on. All of this is public. Most app founders don't realize it's sitting there, one search away, entirely free to access.
The competitive pressure is real. UA ad spend across subscription apps grew **+24%** year-over-year. In categories like Health & Fitness, the top five apps now control **73% of all category UA budget**, up from **54%** just two years ago. If you're not studying what's working in your market, you're not just behind. You're guessing.
## Key Takeaways
- **The biggest ad platforms offer free, public ad libraries.** Meta, TikTok, and Google all let you see exactly what competitor app ads are running. No paid tools required.
- **The hook in the first seconds of a video determines everything.** Meta's own research shows **1 in 3 viewers scroll past** a video that doesn't capture interest within the first **3 seconds**. On TikTok, **60-70% drop off** within the first 1-2 seconds. Study the hook, not the product shots.
- **Copying competitor ads without understanding the mechanism wastes budget.** Your goal is to extract the psychological driver behind each ad, then build your own version of it.
- **You can build a working research workflow in under two hours.** Meta Ads Library, TikTok Creative Center, TikTok Ad Library, and Google Ads Transparency Center cover the majority of subscription app verticals.
- **Competitor research is a starting point, not a strategy.** What works for a competitor with a $500K monthly UA budget may not apply to your stage or audience.
## Why Competitor Ad Research Matters Now
Competitor ad research is the practice of systematically reviewing what paid ads your market competitors are actively running, using free, public ad libraries, to extract creative patterns, offer structures, and platform strategies you can test in your own campaigns. Here's how to research competitor app ads for free using the three biggest ad transparency tools available today.
There are [**31% more apps**](https://adapty.io/state-of-in-app-subscriptions-report/) entering the market each year, but per-app revenue is declining. More competition, fewer dollars per player. In this environment, paid UA is increasingly winner-take-all: a handful of apps capture most of the spend and most of the subscribers.
The fastest way to get smarter about paid advertising is to study what's already working in your market. Not to copy it, but to understand it.
[Law 01](https://www.retention.blog/p/welcome-to-the-48-laws-of-subscription) of subscription app marketing puts this plainly: break down the psychology behind a tactic before adapting it. Ask _why_ it works, not just _what_ it looks like. A competitor running a transformation story is doing it because transformation narratives reduce the core fear: "Will this app actually change anything for me?" A competitor using a countdown timer on a paywall is exploiting urgency. Understanding the mechanism tells you whether it applies to your audience and your app.
Competitor ad research is also the fastest way to avoid expensive creative dead ends. Instead of testing **60** hooks to find the three that convert, you can study which angles the market has already validated. Then run a smaller confirmation test.
## Meta Ads Library: See Every Ad Your Competitors Run
The [Meta Ads Library](https://www.facebook.com/ads/library) is the first place to start for researching competitor app ads. Every active ad running on Facebook, Instagram, Messenger, and the Meta Audience Network is publicly visible, with no account required.
### 1\. Search by brand name or category keyword
Go to the library and search your competitor's brand name. You'll see every ad they're currently running, including video, image, and carousel formats. Filter by country to focus on North America or your primary market.
You can also search by keyword: search "fitness subscription" or "meditation app" to browse a range of players in your vertical, not just named competitors. This reveals who's actually spending in your category, which is often different from who you perceive as competition.
### 2\. What to study in each ad
When you find a competitor's ad, focus on these elements specifically.
- **The first three seconds of any video.** This is the hook. [Meta's own research](https://www.facebook.com/business) shows **1 in 3 viewers scroll past** a video that doesn't capture interest within the first **3 seconds**. On TikTok, analysis of top-performing ads shows **60-70% of viewers drop off within the first 1-2 seconds**. Look for: is the competitor opening with a question, a visual reveal, a relatable scenario, or a bold claim?
- **The pain-outcome structure.** The best-converting ads follow a consistent pattern: show the specific problem, then show how the product resolves it. If a competitor opens with "Still waking up exhausted no matter how much you sleep?", they've identified a precise fear. Identify which pain each ad targets, which tells you which customer segment the competitor is prioritizing.
- **Ad run length.** An ad running for **three months** or more is almost certainly profitable. Ads pulled within **two weeks** usually weren't. Run length is a rough but useful signal for which creative formats are generating positive return.
- **Variation count.** If a competitor has **12** versions of the same creative concept, they're testing hooks or thumbnails at scale. The underlying concept works; they're optimizing the surface. That tells you more than any single ad does.
### 3\. Use comments for qualitative feedback
Meta Ads Library doesn't show performance metrics, but if the ad links to a Facebook post, the comment section often does. Real user reactions, positive and skeptical, reveal what's resonating and where the product creates friction. Ads with high comment volume are often the most revealing.
## TikTok Creative Center and Ad Library
Most app founders know about Meta Ads Library. Far fewer use the [TikTok Creative Center](https://ads.tiktok.com/business/creativecenter/inspiration/topads/pc), which is arguably more useful for understanding creative strategy at the category level.
### 1\. Top Ads: filtered by actual performance
The Creative Center's Top Ads section shows you the best-performing ads across TikTok, filterable by country, industry, objective, time period, and ad format. These ads are ranked based on verified engagement metrics: click-through rate and completion rate inform the curation.
Filter by your vertical (Health & Fitness, Education, Lifestyle) and look at what's performing across the category. This gives you a view of market-level creative patterns, not just one player's experiments.
### 2\. What makes TikTok ad analysis different
TikTok is the most hook-dependent advertising platform available. The **first one second** of a video determines whether a user stops scrolling, a shorter decision window than any other channel. Analysis of top-performing TikTok ads shows that **60-70% of viewers drop off within the first 1-2 seconds** when the hook fails to land.
When studying TikTok ads, you're primarily studying hooks. Look at what visual or audio pattern appears in the first second. A text overlay? A reaction shot? Someone speaking directly to camera? Also study the pacing: TikTok-native content is faster and more abrupt than content made for other platforms. High-production, polished ads consistently underperform native-style content here.
Platform-specific language matters. Top TikTok ads use TikTok's visual grammar and cultural references. Ads that feel like content perform better than ads that feel like ads.
### 3\. TikTok Ad Library (separate tool)
TikTok also maintains a separate [TikTok Ad Library](https://library.tiktok.com/) where you can search specific advertisers by name. Ads remain visible for up to **one year** after their last impression, including paused campaigns. The most detailed transparency data, including audience targeting breakdown by gender, age, and interests, is available for ads targeted to users in the EU, UK, and Switzerland ([TikTok Help Center](https://support.tiktok.com/en/account-and-privacy/personalized-ads-and-data/commercial-content-library)).
The Creative Center's Top Ads feature also includes detailed metrics on selected ads, including an interactive time analysis showing when engagement spikes or drops throughout a video. That granularity makes it the better tool for learning _what's working_, not just _what's running_.
## Google Ads Transparency Center
The [Google Ads Transparency Center](https://adstransparency.google.com/) shows you competitor app ads running across Google Search, YouTube, Gmail, and the Display Network. It covers territory that Meta and TikTok don't touch.
### 1\. Finding competitor Google Ads
Search by advertiser name or website URL. Filter by ad format (video for YouTube, image for Display, text for Search), geography, and time period. The interface is straightforward: search, filter, browse.
### 2\. What to extract from Search and YouTube ads
Google Search ads are particularly useful for understanding competitor positioning. Because Search ads have a tight character limit, competitors are forced to distill their value proposition to its core. If a competitor is running a Search headline that reads "Lose 20 lbs in 8 Weeks," that's their highest-converting promise. They've spent real money testing it.
YouTube ads follow the same hook logic as TikTok and Meta, but the audience intent is different. YouTube pre-roll viewers have passive intent. Users coming through Search have active intent. A competitor may run meaningfully different creative for each, which reveals how they think about their funnel.
### 3\. Auction Insights if you're already running Google Ads
If you're actively running Google Ads, the Auction Insights Report inside your account shows which competitors are bidding on the same keywords, including impression share and overlap rate ([Google Ads Help Center](https://support.google.com/google-ads/answer/2579754?hl=en)). This tells you who your real competitive set is in paid search, which often differs from your perceived market map.
## What to Look For and What Not to Copy
Finding competitor app ads is the easy part. Knowing what to extract from them is the skill.
### 1\. The three signals that matter most
- **Hook type.** Is the ad opening with a relatable pain, a shocking statistic, a before-and-after reveal, or a direct challenge? Each hook type attracts a different intent profile. When a competitor consistently uses pain-first hooks and their ads run for months, their audience responds to empathy over aspiration. That's a validated category insight, not just one brand's experiment.
- **Offer structure.** Is the competitor leading with a free trial, a discount, or a direct purchase ask? How long is the trial? Across the subscription app market, longer trials in the **17-32 day range** convert at significantly higher rates than short trials: roughly **45%** median conversion versus **27%** for short trials. When you see a competitor extending from a 7-day to a 30-day trial, there's testing data behind that decision.
- **Variation count.** A competitor running **15** variants of one core concept is signaling two things: the underlying concept has enough signal to keep testing, and they have creative infrastructure to support it. The right question is what the core concept is, and why it's working.
### 2\. Platform comparison: what each tool shows
The table below covers the five main free tools. LinkedIn Ad Library is particularly useful for apps targeting business audiences, as it shows advertiser-level targeting details and impression ranges not available on other platforms.
| Tool | Platforms Covered | Search by Advertiser | Historical Ads | Performance Data | Cost |
| --- | --- | --- | --- | --- | --- |
| Meta Ads Library | Facebook, Instagram, Messenger | Yes | Active + some archived | None | Free |
| TikTok Creative Center | TikTok | By vertical/category | Recent periods | CTR, completion rate | Free |
| TikTok Ad Library | TikTok | Yes (by advertiser) | Active + 1 year | Limited | Free |
| Google Ads Transparency Center | Search, YouTube, Display, Gmail | Yes | Recent periods | None | Free |
| LinkedIn Ad Library | LinkedIn | Yes | Active + 1 year archive | Impression ranges | Free |
### 3\. What not to copy
- **Don't copy creative formats mechanically:** A competitor using talking-head UGC video doesn't mean UGC is the right answer for your app. It means UGC works for their app, their audience, and their current budget. If you don't have the infrastructure and testing volume, copying the format without the execution fails.
- **Don't copy offers you can't support with data**: A competitor offering a 30-day free trial may have the [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) data to back it up. If you don't know your LTV yet, extending trials to match a competitor could erode early revenue before you've validated your subscription economics.
- **Don't skip the why:** Copying a competitor's "before and after transformation" hook without understanding the mechanism means you're guessing. The transformation hook works because it addresses the core subscriber fear: "Will this app actually change anything for me?" Diagnose the mechanism, then build your version of it.
## FAQs
**Can I see how much competitors spend on their ads?**
No. Ad libraries show creative content but not spend figures. You can see which formats competitors use, how long ads have been running, and on LinkedIn, rough impression ranges. The best proxy for budget is run length: an ad running for 60+ days with multiple variants is almost certainly profitable.
**Are competitor ad libraries updated in real time?**
Meta's Ad Library is close to real time, with new ads typically appearing within hours of launch. Google Ads Transparency Center has a slight delay. TikTok's Commercial Content Library for EU, UK, and Switzerland shows active and paused campaigns within a short window. The global TikTok Ad Library retains ads for up to one year after the last impression.
**Should I use paid tools like SensorTower or Pathmatics instead?**
Free tools cover the majority of what early-stage app founders need. Paid tools add multi-year historical archives, spend estimates, and automated monitoring alerts. Start with free tools to validate your research process and understand what you're looking for, then upgrade when manual research becomes the bottleneck.
## Stop Guessing. Start Measuring.
The most expensive lessons in paid UA are the ones you didn't have to learn yourself. Your competitors have already spent real money figuring out which hooks stop scrolls, which offers convert, and which platforms reach paying subscribers in your category. That information is public. It costs nothing to access.
Once you've identified the mechanisms working in your market, the next step is running your own version. At that point, the question shifts: which of _your_ competitor app ads-inspired tests is actually driving paying subscribers, not just installs?
If you're running paid campaigns on Meta, Google, or TikTok and want to see which channels make you money and which just burn budget, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) starts free with your first 15K installs tracked.
---
## App Campaign Naming Convention: Templates for Meta, Google, TikTok & ASA
URL: https://www.airbridge.io/en/blog/app-campaign-naming-convention
Language: English
Published: 2026-05-11
Author: Luke
You launched **40** app campaigns last quarter. Meta, Google, TikTok, Apple Search Ads. The whole stack. Now you're trying to figure out which ones drove subscriptions. Your campaign list looks like this:
- `Campaign_1`
- `iOS_US_test_v2`
- `Meta_retarget_new_FINAL_copy`
- `TikTok_aug_broad_3`
The data is there. You just can't read it.
This is the **app campaign** naming problem. It's invisible when you're running three campaigns. It becomes expensive once you're managing **50**. Messy app campaign names mean you can't filter by channel, can't compare creative formats, can't track what a specific audience is doing across platforms. You end up scrolling through a spreadsheet trying to remember what "copy\_2\_new" meant six months ago.
## Key Takeaways
- **Campaign names are data infrastructure, not labels.** What you name a campaign determines whether your reporting tools can parse it six months from now.
- **Every app campaign needs 5 core fields:** platform, objective, audience type, launch date, and a unique identifier. That's **5** fields total.
- **Creatives need their own naming system,** separate from campaigns, capturing format, style, and hook concept.
- **UTM naming for app marketing** is the bridge between your ad platforms and your analytics, and it needs to mirror your campaign names, not contradict them.
- **One consistent system beats a perfect one.** The best app campaign naming convention is the one your whole team actually uses.
## Why Campaign Naming Breaks at Scale
Poor data quality costs organizations **$12.9 million per year** on average, according to Gartner. Forrester puts it more directly: **21 cents of every media dollar** is wasted due to data quality issues. For app marketers, one of the most common data quality problems is also the most fixable, campaign naming.
When you're running three campaigns, inconsistent names are a minor inconvenience. When you're running **30**, they become a reporting problem. At **100**, they make decisions impossible.
Here's what typically happens: you launch a campaign, name it something sensible, then duplicate it to test a new audience. The duplicate keeps most of the original name. You do this **20** more times. Three months later, you have **60 campaigns**, half are variations of each other with no clear way to tell what's actually different.
### 1\. Where your app campaign data actually lives
When you run app campaigns, your performance data lives in **3** separate systems: your ad platform (Meta, Google, TikTok), your attribution tool, and your app store. Each defines installs differently, counts subscription events differently, and produces different numbers. Your ad platform counts a click; your attribution tool counts a first open; your app store counts a download. Your campaign names are the one thread connecting all three. If the names don't match across systems, reconciliation becomes a manual job. and a regular source of errors.
## What "Fixed" Looks Like: Before vs. After
The fastest way to understand the system is to see it fix a real mess. Here are **6** common campaign names from an account running **3** months without a convention. Here's what each one looks like after applying the framework.
| # | Before | After |
| --- | --- | --- |
| 1 | Campaign_1 | META_TRIAL_COLD_20260115_V1 |
| 2 | iOS_US_test_v2 | GADS_INSTALL_IOS_US_20260115_V1 |
| 3 | Meta_retarget_new_FINAL_copy | META_TRIAL_RETARGET_20260115_V2 |
| 4 | TikTok_aug_broad_3 | TT_INSTALL_COLD_20260801_V3 |
| 5 | google_app_march | GADS_INSTALL_AND_US_20260301_V1 |
| 6 | ASA_brand_keywords_copy2 | ASA_EXACT_US_20260115_V2 |
Every "after" name answers the same **5** questions in the same order. You can sort by platform, filter by objective, and compare audience types across all **6** campaigns in seconds, without opening a single one.
> _Note: Google campaigns (#2 and #5) add OS and GEO as extra fields. The channel-specific templates section explains this._
## Build Your App Campaign Naming Framework
A solid **app campaign naming convention** is a fixed pattern applied to every campaign name so anyone, and any reporting tool, can understand key details at a glance. It answers five questions from the name alone: where is it running, what is it optimizing for, who is it targeting, when did it launch, and what makes it distinct from similar campaigns.
The formula for any app campaign name:
**Platform + Objective + Audience type + Launch date + Unique identifier**
Put these together using underscores as separators and you get something like:
`META_TRIAL_COLD_20260115_V1`
That name tells you: it's a Meta campaign, optimizing for trial starts, targeting cold audiences, launched January **15**, **2026**, creative variant 1. Anyone on your team can read it without opening the campaign.
[Screenshot 2026-05-11 at 16.15.36.png]
### 1\. Platform (**2–3 letter code**)
Use a short, consistent code for each ad channel. Pick one version and never vary it:
| Channel | Code |
| --- | --- |
| Meta Ads (Facebook + Instagram) | META |
| Google App Campaigns | GADS |
| TikTok Ads | TT |
| Apple Search Ads | ASA |
When you're running the same campaign structure across platforms and comparing performance, having three different names for Meta, `FB`, `META`, `Facebook` creates three separate rows in your analytics tool. They look like three different channels.
### 2\. Objective (what you're optimizing for)
This field encodes what the platform is optimizing toward. For subscription apps, common values are:
- `INSTALL`: app installs
- `TRIAL`: free trial starts
- `SUB`: paid subscriptions
- `ROAS`: return on ad spend
The objective field in your campaign name should always match your actual optimization event. Optimize for installs, you get installs. Optimize for subscriptions, you get subscriptions.
### 3\. Audience type
Keep this high-level. You need to filter and compare, not encode every targeting detail:
- `COLD`: new audiences, no prior interaction
- `LAL1`, `LAL2`: lookalike audiences at **1%**, **2%** size
- `RETARGET`: prior app visitors or engagers
- `REENGAGE`: lapsed users
With audience type in the name, you can compare how the same creative performs on cold vs. warm traffic without opening each campaign individually.
### 4\. Launch date (YYYYMMDD format)
Use `YYYYMMDD`. Always. This format sorts chronologically in any spreadsheet or BI tool, across every country's date convention. `20260115` is unambiguous. `Jan15` or `Q1` is not.
Never use relative dates. "Summer2025" made sense at launch. Six months later, it's noise in your campaign list.
### 5\. Unique identifier
The last field separates otherwise identical campaigns:
- `V1`, `V2`, `V3`: creative variants
- `T1`, `CTRL`: test/control variants
- `PROMO`: limited-time offers
## Name Your Creatives Separately
Campaign-level naming is the foundation of any app campaign naming system. Creative naming is where most teams fall short.
When you're running tests across Meta, TikTok, and Google simultaneously, you can have hundreds of active creatives. Without a separate naming system, you can see that an ad performed well, but you can't find the file, you don't know why it worked, and you can't build on it.
Creative naming answers three questions: **format** (what type of ad), **style** (how it was produced), and **concept** (what hook or idea it uses).
| Dimension | Codes | Example |
| --- | --- | --- |
| Format | VID15, VID30, STATIC, CAR | 15s video, 30s video, image, carousel |
| Style | UGC, ANIM, DEMO, NARR | User-gen, animation, demo, narrated |
| Concept | PAIN, PROOF, OUTCOME, HOWTO | Pain point, social proof, result, tutorial |
A complete creative name follows the same logic: `VID30_UGC_PAIN_V1`. That's a **30**\-second UGC-style video built on a pain point hook, first variant. Enough to filter, compare, and iterate across channels.
One practical rule: platform-assigned creative IDs (the numeric IDs Meta or TikTok generate for each ad) change every time you duplicate or rebuild a campaign. Your internal creative naming does not. Always use your own system as the canonical reference.
TikTok creative fatigue arrives fast. Top-performing creatives typically last **7 to 10 days** before performance drops, far shorter than Meta's 2-4 week cycle. [TikTok's guidance](https://ads.tiktok.com/help/article/creative-best-practices) is to refresh when delivery shows a consistently declining trend. Without creative naming, you can't tell which hooks are burning out vs. which still have runway.
## UTM Naming for App Marketing
Campaign names live inside your ad platform. UTM parameters travel with the click URL and flow into your analytics or attribution tool. Both systems need to work together, and the first step is making sure they speak the same language.
If your campaign names and UTM parameters use different naming schemes, you'll have two systems telling two different stories. Reconciling them manually does not scale.
### 1\. The 5 UTM fields for app campaigns
The goal of UTM naming for app marketing is what StartApp School calls "apples-to-apples" cross-platform comparison, the same campaign structure producing the same data shape in every reporting tool you use.
| UTM Field | What it captures | App example |
| --- | --- | --- |
| utm_source | Ad platform | meta, google, tiktok, asa |
| utm_medium | Channel type | paid_social, paid_search, cpc |
| utm_campaign | Specific campaign | Mirror your campaign name |
| utm_content | Creative variant | Mirror your creative name |
| utm_term | Audience or keyword | cold_lal1, brand_keyword |
**The most common UTM mistake in app marketing** is using inconsistent values across platforms. `utm_source=facebook` on one campaign and `utm_source=fb` on another look like two separate traffic sources in your analytics tool. Your total Meta performance appears split across multiple rows. Your aggregated reporting is wrong, and you won't notice until someone tries to reconcile the numbers.
### 2\. The one rule for UTM naming
Understanding [attribution windows](https://www.airbridge.io/en/glossary/attribution-window) helps you set the right UTM tracking windows across channels. Your `utm_campaign` value should mirror your campaign name structure. If the campaign is `META_TRIAL_COLD_20260115_V1`, the UTM should follow the same logic, not a different scheme that requires mental translation at reporting time.
Keep UTM values lowercase. No spaces (use underscores). No special characters. Google Analytics confirms UTM values are case-sensitive: `utm_source=facebook` and `utm_source=Facebook` are logged as two separate traffic sources ([Google Analytics docs](https://support.google.com/analytics/answer/10917952)).
[Screenshot 2026-05-11 at 16.16.44.png]
## Channel-Specific Naming Templates
Here's how the framework translates to each major app campaign channel.
### 1\. Meta Ads
Meta's three-level structure maps directly to a three-level naming system. In practice, Meta's delivery tends to prioritize **1–2 creatives per ad set**, so naming makes it easy to identify which ads are getting budget and pull winners into new tests.
| Level | Template | Example |
| --- | --- | --- |
| Campaign | META_{OBJ}_{AUD}_{DATE} | META_TRIAL_COLD_20260115 |
| Ad Set | {AUD_DETAIL}_{GEO}_{BID} | LAL1_US_MAXCONV |
| Ad | {FORMAT}_{STYLE}_{CONCEPT}_{VARIANT} | VID30_UGC_PAIN_V1 |
For more on how to structure Meta campaigns for subscription app performance, see [Meta ads for subscription apps](https://www.airbridge.io/en/blog/meta-ads-subscription-apps).
### 2\. Google App Campaigns
Google App Campaigns give you less targeting control than Meta, but naming still matters for filtering and cross-platform reporting.
| Level | Template | Example |
| --- | --- | --- |
| Campaign | GADS_{OBJ}_{OS}_{GEO}_{DATE} | GADS_INSTALL_IOS_US_20260115 |
| Ad Group | {CREATIVE_THEME}_{VARIANT} | OUTCOME_HOOK_V1 |
One important distinction: Google has less user data on iOS than on Android, which limits optimization quality on iOS. If you're running separate iOS and Android campaigns, make the OS explicit in the name: `GADS_INSTALL_IOS_US` vs `GADS_INSTALL_AND_US`. See how [self-attributing networks](https://www.airbridge.io/en/glossary/self-attributing-network) report data differently.
### 3\. TikTok Ads
TikTok mirrors Meta's three-level structure. Creative naming is especially important here because refresh cadence is higher than any other channel. Without clear naming, you can't tell which hooks still have runway and which ones are burning out.
| Level | Template | Example |
| --- | --- | --- |
| Campaign | TT_{OBJ}_{AUD}_{DATE} | TT_INSTALL_COLD_20260115 |
| Ad Group | {GEO}_{AUD_DETAIL}_{BID} | US_BROAD_MAXCONV |
| Ad | {FORMAT}_{STYLE}_{CONCEPT}_{VARIANT} | VID15_UGC_HOOK1_V2 |
The **first second** of a TikTok ad drives most of the performance signal. Your hook concept in the creative name (PAIN, PROOF, OUTCOME, HOWTO) becomes your creative analytics layer, which tells you which categories are working before you burn budget finding out. For creative strategy specifics, see [TikTok ads for subscription apps](https://www.airbridge.io/en/blog/tiktok-ads-subscription-apps).
### 4\. Apple Search Ads (ASA)
ASA only optimizes toward clicks and installs, not purchases. Your naming should reflect that ASA is a high-intent discovery channel, not a direct-subscription driver.
| Level | Template | Example |
| --- | --- | --- |
| Campaign | ASA_{MATCH_TYPE}_{GEO}_{DATE} | ASA_EXACT_US_20260115 |
| Ad Group | {KEYWORD_THEME}_{GEO} | PRODUCTIVITY_APP_US |
For ASA, the keyword theme matters more than audience type, because you're capturing users already searching for a solution. See [Apple Search Ads for subscription apps](https://www.airbridge.io/en/blog/apple-search-ads-subscription-apps) for keyword strategy.
## 3 Naming Mistakes That Tangle Your Data
### 1\. Using special characters or spaces
Spaces, commas, asterisks, and exclamation points break URL encoding in app campaign names, get stripped by reporting tools, and create mismatches between ad platforms and analytics. Use only underscores (`_`) as separators. Never spaces.
This is the most common technical error teams make when figuring out how to name ad campaigns app-side. It's also the easiest to prevent. Establish the rule before you launch and enforce it with a shared naming document.
### 2\. Being too specific or too vague
**Too specific:** `META_TRIAL_US_M25-34_LAL1_PAINPOINT_SUMMERPROMO_V3_20260115_TESTB`
Aside from being impossible to scan, long names create a hard technical problem: AppsFlyer enforces a **100-character campaign name limit**. Names that exceed it are replaced with `c_name_exceeded_max_length` in your reports, making that campaign's data permanently unreadable.
**Too vague:** `Meta_campaign_1`
Useless after the first week.
One practitioner in the StartApp School curriculum described a naming convention with **14 underscores** that caused regular mistakes by their design team. If your designers need a decoder ring to read a campaign name, it's too long.
The sweet spot is **5–7 fields in consistent order.** Enough to filter and compare. Short enough to read at a glance.
### 3\. Not documenting the system
A naming convention only works if everyone uses the same one. That requires documentation: a shared Google Sheet or Notion page listing your approved codes, your date format, and a naming example for each channel.
Without documentation, every new hire invents their own system. The chaos returns.
## FAQ
### Do I need an app campaign naming convention if I'm only running one channel?
Yes, but you can simplify. Drop the platform code (it's redundant when you have one channel) and focus on objective, audience, and date. The habit of consistent naming matters more than complexity. When you add a second channel in three months, you'll be glad you started early.
### What is the difference between a campaign name and a creative name?
Campaign names identify where an ad is running, who it's targeting, and when it launched. Creative names identify what the ad looks like: its format, production style, and hook concept. Both follow the same naming logic: short codes, underscores, consistent order, but they live at different levels of your ad account structure. Campaign names go on the Campaign level. Creative names go on the Ad level.
## Your App Campaign Naming Reference Card
Every code your team will ever need, in one place.
| Field | Format | Approved codes |
| --- | --- | --- |
| Platform | 2–4 letters | META GADS TT ASA |
| Objective | 4–6 letters | INSTALL TRIAL SUB ROAS |
| Audience | Short tag | COLD LAL1 LAL2 RETARGET REENGAGE |
| Launch date | YYYYMMDD | 20260115 |
| Unique ID | Short tag | V1 V2 T1 CTRL PROMO |
| Separator | Character | _ (underscore only. No spaces, no hyphens) |
**Full campaign name formula:** `{PLATFORM}_{OBJECTIVE}_{AUDIENCE}_{DATE}_{ID}`
**Full creative name formula:** `{FORMAT}_{STYLE}_{CONCEPT}_{VARIANT}`
| Creative field | Approved codes |
| --- | --- |
| Format | VID15 VID30 STATIC CAR |
| Style | UGC ANIM DEMO NARR |
| Concept | PAIN PROOF OUTCOME HOWTO |
Keep a copy of this table in a shared doc. When a new code is needed (new channel, new objective), add it here first, then use it everywhere.
## Build the System Before You Scale
App campaign naming conventions are infrastructure. Put them in place before the chaos arrives.
The system here takes about **an hour** to set up, **30 minutes** to document, and zero extra time per campaign once it's a habit. Filter by audience type, compare creatives across channels, and build reports that don't need weekly cleanup.
And when you need to answer the question "which campaigns drove subscriptions last quarter?", the answer is in your data, sorted and filterable, without guesswork.
---
## What Should a Small App Team Track Before Scaling App Ads?
URL: https://www.airbridge.io/en/blog/what-to-track-before-scaling-app-ads
Language: English
Published: 2026-05-11
Author: Luke
Most app teams know they should track something before scaling paid ads. The problem is nobody agrees on what "ready" actually looks like.
There is no official standard, no universal green-light checklist. So teams scale their **app ads** anyway, and discover too late that they were spending on a funnel that was never going to convert.
**Key Takeaways**
- **Trial-to-paid conversion is your most critical pre-scale metric.** The median across 75,000+ subscription apps is **34.8%**, and the top quartile starts at **51.5%**. Know where you stand before committing more budget.
- **Customer means paying user, not installer or trial starter.** Every unit economics calculation starts from this definition. LTV must exceed CPS with enough margin to sustain growth — scaling when that math is negative makes losses larger.
- **Trial design matters as much as traffic volume.** Apps with trials lasting 17-32 days convert at a median of **45.7%**, compared to **26.8%** for shorter trials. Optimize your trial before scaling your ads.
- **Instrument four funnel events before any paid ad goes live.** Without install, trial start, subscription, and renewal events tracked back to your ad platform, you cannot calculate CPS or compare channels.
- **Benchmarks exist.** Industry data gives you clear targets to measure against before committing serious budget to app ads.
## Why Teams Scale App Ads Too Early
Scaling paid ads on an unmeasured funnel is the fastest way to burn budget without learning anything useful.
The math is clear: if your cost per subscription exceeds what a subscriber is worth over 60 days, every dollar added to your app ads budget is a dollar lost. Without knowing your cost per subscription, you are guessing at the number that determines whether your business works.
**56% of new apps earn less than $1,000 in their first year** ([Adapty, 2026](https://adapty.io/state-of-in-app-subscriptions/)). Not because they failed to run ads, but because they scaled spend before proving the funnel. The apps that break through share one trait: they know their unit economics before they scale.
The contrast is visible at the funnel level. Subscription apps using a hard paywall convert **12.1%** of downloads to paying users within 35 days. Apps using a freemium model convert just **2.2%** ([RevenueCat, 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). That is a **5.5x** difference driven entirely by monetization structure, not traffic quality. Scaling the same ad budget into both funnels produces very different cost-per-subscription outcomes.
> The question to answer before scaling is not "do we have enough budget?" The question is "do our numbers justify more spend?"
[Screenshot 2026-05-11 at 13.28.14.png]
## The Four Metrics That Tell You You're Ready
Every calculation starts from the same place: **customer means paying user, not installer, not trial starter.** That distinction matters for every metric below. An install is not a customer. A trial is not a customer. The person who paid is a customer.
There are four numbers every small app team needs before scaling **app ads**. Not fifty. Four.
### 1\. Trial Start Rate
**Trial start rate** is the percentage of installs that begin a free trial. The all-category median is **6.2%**, but top-performing apps (P90) hit **20.3%**, more than three times the median (RevenueCat, 2025).
One insight most small teams overlook: **82% of trials start on Day 0** (RevenueCat, 2025). Paywall placement during the first session is decisive. Users who leave without starting a trial on Day 0 rarely return to start one.
If your trial start rate sits below **3%**, you have an onboarding or paywall problem, not a traffic problem. More app ad spend at that conversion rate amplifies the leak rather than fixing it.
### 2\. Trial to Paid Conversion
This is the number that separates apps ready to scale from apps ready to optimize. The all-category median is **34.8%**. The top quartile (Q3+) starts at **51.5%** (RevenueCat, 2025).
Trial length has an outsized impact on this number. Apps with trials lasting 17-32 days convert at a median of **45.7%**, compared to **26.8%** for shorter trials (RevenueCat, 2025). If your trial is set to 3 or 7 days and conversion is low, lengthening the trial period will often outperform increasing ad spend as a next step.
If your trial-to-paid rate is below **25%**, scaling ads produces more churned trial users. Fix the trial experience first, then scale.
### 3\. Revenue per Install (D60 RPI)
[Lifetime value (LTV)](https://www.airbridge.io/glossary/lifetime-value-ltv), or more practically **Day 60 Revenue per Install (D60 RPI)**, tells you how much each install is worth after 60 days. The all-category median is **$0.31**. The top quartile hits **$0.81** (RevenueCat, 2025).
This number is your effective ceiling for cost per install. If cost per install consistently exceeds D60 RPI, you are underwater from the first impression.
### 4\. Cost per Subscription (CPS) and CPI Reality
**Cost per subscription (CPS)** is what you actually pay to acquire a paying user, not a click, not an install, not a trial start. Because CPPU (cost per paying user) is typically **4-5x the CPI** (RevenueCat, 2025), teams that optimize for cheap installs often spend inefficiently on traffic that never converts.
Concrete CPI benchmarks by category show what you are actually working with (RevenueCat, 2025):
| Category | iOS CPI (Global) | Android CPI (North America) |
| --- | --- | --- |
| Health & Fitness | $5.78 | $3.06 |
| Generative AI | $10.58 | $3.60 |
| Business | $14.08 | $11.25 |
| Gaming | $5.99 | $2.56 |
iOS CPI runs roughly **2-3x Android** across most categories (RevenueCat, 2025). For a Health & Fitness app on iOS, a $5.78 CPI multiplied by the 4-5x CPPU factor puts your cost per subscriber around **$26**. A useful budget rule: keep your daily ad budget at least 10x your target CPA to give the algorithm enough signal to optimize. At a $26 CPA, that means **$260/day minimum**, or roughly **$7,800/month**, before results become reliable.
For teams with budgets under $10,000 per month, starting on Android delivers 2-3x more installs for the same spend. More installs means faster learning and statistically stronger data before you decide to scale.
| Metric | Median | Q3 (Top 25%) | Action if Below Median |
| --- | --- | --- | --- |
| Trial Start Rate | 6.2% | 12.4%+ | Audit onboarding and paywall placement |
| Trial to Paid | 34.8% | 51.5%+ | Improve trial experience and value delivery |
| D60 RPI | $0.31 | $0.81+ | Review pricing or trial length (17-32 days optimal) |
| Y1 LTV per Payer | $16.27 | $33.91+ | Improve retention before adding spend |
_Source: RevenueCat, 2025_
Hitting Q3 on at least two of these metrics is a reasonable "ready to scale" signal.
## What Your Trial Design Reveals Before You Spend
Before scaling app ads, it is worth examining whether your trial is structured to convert the traffic you are buying. Two apps with identical ad budgets can produce very different subscription rates based entirely on how the trial is designed.
**Trial length changes outcomes more than most copy changes.** Subscription apps with trials lasting 17-32 days convert at a median of **45.7%**, compared to **26.8%** for apps with shorter trials (RevenueCat, 2025). The additional window gives users enough time with the product to feel confident paying.
**Paywall design shapes conversion independently of trial length.** Adding a trial offer to your paywall setup consistently outperforms showing pricing alone. Teams that run structured paywall experiments achieve up to **40x more revenue** than those that do not ([Adapty, 2026](https://adapty.io/state-of-in-app-subscriptions/)). That multiplier comes from teams running an average of 14.7 experiments on plan structure, trial duration, and pricing. Those experiments are only possible if you have the funnel events and cohort data the next section covers.
**First renewal is the honest signal of product quality.** It is straightforward to convince a curious user to start a free trial. The harder test is whether they pay, and then whether they stay. Nearly **30% of annual subscriptions cancel in the first month** (RevenueCat, 2025). If your first-year annual renewal rate is below **66.3%** (the industry median), scaling ad spend accelerates churn rather than growth.
**Low usage drives most cancellations.** "Not enough usage" is the leading cancellation reason, cited in **37.2%** of cancellations across subscription apps (RevenueCat, 2025). If users are not returning to the app during their trial, more installs will not fix that.
A practical pre-scale trial checklist:
1. Trial length is at least 7 days (17-32 is the optimal range per RevenueCat, 2025)
2. Paywall includes a trial offer, not just a pricing grid
3. First-year renewal rate is above 50%
4. Post-install engagement shows users returning during the trial period
If all four conditions are met, your trial is ready for scaled traffic.
## The Tracking Setup You Actually Need
Knowing the benchmarks is only useful if you can measure yourself against them. Here is the minimum instrumentation a small team needs before the first app ad goes live.
The **40x revenue difference** between teams that run structured experiments and those that do not (Adapty, 2026) is not the result of bigger budgets. It is the result of better data. The setup below is the foundation that makes systematic testing possible.
### 1\. Four funnel events
You need four events tracked in your analytics and passed back to your ad platform: **Install** (automatic), **Trial Start**, **Subscription Start**, and **Subscription Renewal** (or cancellation). Everything else is optional at this stage.
Without these four events, you cannot calculate trial start rate, trial-to-paid conversion, or CPS. You are measuring traffic volume, not business results.
One addition worth considering for Android teams: payment failure events. On Google Play, billing errors cause **28.2%** of involuntary churn, compared to **15.1%** on the App Store (RevenueCat, 2025). Tracking payment failure events and implementing retry logic can recover a meaningful share of these subscribers before they churn permanently.
### 2\. Channel-level attribution
Every install needs a source tag so you know whether it came from Meta, Google, TikTok, or Apple Search Ads. On iOS, [SKAdNetwork](https://www.airbridge.io/glossary/skadnetwork) handles this through privacy-preserving postbacks. Understanding how postback windows and conversion value schemas work is essential before launching your first paid campaign.
Start with one channel. Prove unit economics. Then add a second channel and compare. Spreading a small budget across three or four channels simultaneously prevents you from reaching the volume needed to draw reliable conclusions from any of them.
### 3\. One source of truth for revenue
[Attribution](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) data, payment data, and ad spend data come from three different systems, and none of them will agree perfectly. Understanding which source is authoritative for which question prevents you from making decisions based on the wrong number.
| Source | Authoritative For | Not Reliable For |
| --- | --- | --- |
| Ad platforms (Meta, Google) | Creative performance: CTR, CPM, spend | Install-to-revenue attribution |
| App stores (Apple, Google Play) | Payment and billing data | Attribution or behavioral insights |
| Attribution tool | Install source tracking | Complete accuracy — discrepancies are normal |
Pick one as your primary decision-making reference before you start, and stick to it. Switching mid-campaign invalidates the comparisons you have already built.
### 4\. Cohort analysis, not just totals
Aggregate metrics mislead. A total conversion rate of 34% might look healthy while hiding the fact that campaigns from last month converted at 12% and campaigns from two months ago converted at 56%. Before scaling, confirm you can segment data by install cohort and by channel. This is the difference between knowing "conversion is 34%" and knowing "Meta converts 41%, Google converts 28%." Only one of those enables a confident scaling decision.
[Screenshot 2026-05-11 at 13.31.09.png]
## Where Do Your Numbers Stand?
The benchmark table in the section above is your reference. If you are below median on more than two metrics, prioritize optimization before adding ad spend. The goal is not to hit P90 on everything before spending. The goal is to understand your funnel well enough that scaling reveals where to invest, not whether you survive the investment.
## Frequently Asked Questions
### Do I need a dedicated tracking tool before my first app ad campaign?
Yes. At minimum, every install from a paid campaign needs a channel source attached to it. Without attribution data, you cannot calculate CPS by channel, which means you cannot compare platform performance. A single-channel setup can start with each platform's native analytics. Once you run two or more channels simultaneously, you need a unified view. Otherwise you are reading separate reports that count the same users differently and drawing wrong conclusions from the discrepancies.
### How many conversions do I need before my data is reliable?
Calculate the sample size you need based on your baseline conversion rate and the minimum lift you want to detect. Do not use a fixed number — the right threshold depends on your current rate and how large a difference matters to your decision. As a rough guide, most subscription apps need several hundred trial starts before cohort-level patterns stabilize. With fewer data points, a handful of outliers can move your observed rate by 10 percentage points or more in either direction.
### Should I optimize app ads for CPI or CPS?
**Always CPS (cost per subscription).** CPI tells you how cheap your traffic is, not how valuable it is. A $2 install from a user who never starts a trial costs more in the long run than a $6 install from a user who subscribes and renews. Most subscription apps discover that the channel with the highest CPI often produces the lowest CPS, because it attracts higher-intent users. Optimize for the metric that represents actual revenue.
### How do I know if my problem is the product or the ads?
Run a simple diagnostic. Take your current organic users and measure their trial-to-paid conversion rate. Then compare it against your paid users' rate. If organic converts at 40% and paid converts at 18%, the ads are attracting the wrong audience. If both convert at similar rates but the rate is low, the product or trial experience needs work before scaling either channel. The gap between organic and paid conversion is the clearest signal available.
## Track First. Then Scale.
The apps that scale efficiently are not the ones with the biggest budgets. They are the ones that know their funnel numbers before they spend.
If you are running app ads on Meta, Google, or TikTok and want to see exactly which campaigns drive paying subscribers (by channel, by creative, by day), [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your ad spend to subscription revenue. It includes 15K free attributed installs, pre-configured subscription funnel events (Trial Start, Subscribe, Renewal), and direct integrations with RevenueCat and Adapty.
---
## How to Test App Ad Creatives Without Burning Budget
URL: https://www.airbridge.io/en/blog/how-to-test-app-ad-creatives-without-burning-budget
Language: English
Published: 2026-05-11
Author: Harper (Trang Nguyen)
You're running paid UA on Meta or TikTok, and your app creative library is growing. But after weeks of testing, you still can't answer the question that actually matters: which creative is making you money, and which one is wasting your budget?
Most teams run ad creative "tests" without a real system. They change the hook, the copy, and the format all at once, then spend days trying to decode why one version outperformed another. **That's not testing. That's guessing with extra steps.** Learning how to test app ad creatives for subscription apps the right way means building a repeatable system that produces a clear, learnable answer every single time you run it.
## Key Takeaways
- **Creative is the primary performance variable.** On Meta and TikTok, the app creative determines whether a user stops, installs, and ultimately subscribes. Algorithms optimize delivery. Creative determines fit.
- **Test one variable at a time.** Changing the hook, format, and copy simultaneously makes results unlearnable. One change, one test, one insight.
- **Use win rate to evaluate results.** A structured test should produce at least a 5% win rate. Below that, the problem is methodology, not creative quality.
- **False positives are common.** A creative that performs well at low spend often collapses when you scale it. Validate every winner in a standalone ad group before promoting it to BAU. Low-spend performance rarely predicts scale performance.
- **CTR is an incomplete signal for subscription apps.** Install-to-subscription rate reveals which creatives attract users who actually pay, not just users who install.
## Why Most Creative Tests Don't Teach You Anything
App creative testing looks simple: make two ads, run them, pick the winner. In practice, most teams end up with data that can't support a decision.
### 1\. Too Many Variables at Once
You change the hook, the background, and the copy at the same time. When one version outperforms the other, you have no idea what caused it.
This problem is most acute in the first few seconds. Most users decide whether to engage with an ad after just one second. By the three-second mark, half the audience has already scrolled past. Testing the hook separately from the copy, and the copy separately from the format, is the only way to know which element is actually doing the work — and the only way to reproduce that performance across future creatives.
Structured creative testing means isolating one element per iteration:
- **Hook vs. hook**: same format, different opening 3 seconds
- **Angle vs. angle**: problem-framing vs. transformation-framing, same format
- **Format vs. format**: UGC-style vs. motion graphic, same concept
Per [Law 44 of the 48 Laws of Subscription App Success](https://www.retention.blog/p/welcome-to-the-48-laws-of-subscription): **calculate your sample size in advance, test one variable at a time, and run tests to completion before deciding.** Changing variables mid-flight wastes money and erases whatever learning was possible.
### 2\. Optimizing for the Wrong Metric
If your Meta campaign optimizes for trial starts, the algorithm delivers your ad to users who start trials, including users who never subscribe.
The gap between median and excellent subscription app performance is almost entirely about traffic quality. Among [75,000+ subscription apps analyzed](https://www.revenuecat.com/state-of-subscription-apps-2025/), the median trial-to-paid conversion rate is **34.8%**. The upper quartile reaches **51.5%**. That gap is not product quality alone. It is the quality of users each creative attracts.
The same principle applies at the creative level. A creative with high trial volume but low subscription conversion is producing the wrong traffic. One with fewer trials but higher subscription rates is the actual winner — visible only if you measure at the right funnel step. Adapty's analysis of 16,000+ apps found that experiments improving conversion only 28% of the time still lift [LTV (lifetime value)](https://www.airbridge.io/glossary/lifetime-value-ltv) by 46%. Fewer trials from higher-intent users produces better revenue per cohort than more trials from users who churn.
## Build Your Creative Testing System
A rigorous testing system starts with production. Before you can identify winners, you need enough creatives to test meaningfully. StartApp School's paid UA curriculum sets a clear production benchmark: **aim for approximately 100 different creatives as your initial testing library.** That number reflects how many concepts you need to cycle through before conversion patterns become statistically meaningful and before the algorithm has enough material to surface genuine signal.
The priority metric table below is ordered from most to least diagnostic for subscription apps:
| Metric | What It Reveals | Priority |
| --- | --- | --- |
| Install-to-subscription rate | Which creatives attract users who pay | High |
| CAC / Cost-per-acquisition | Whether unit economics are sustainable | High |
| Win rate (% of tests producing a winner) | Whether your testing methodology is working | High |
| Hook rate (3-sec view rate) | Whether the first seconds stop the scroll | Medium |
| Hold rate | Whether users stay engaged past the hook | Medium |
| CTR / IPM | Volume signal, useful but incomplete alone | Low |
**Set your baseline from your current best-performing creative.** Every new creative is measured against that baseline, not against zero, and not against other untested concepts.
### 1\. Budget, Volume, and What to Expect
Most creative testing guides cite volume and budget figures without sourcing them. The figures below are drawn from StartApp School's paid UA curriculum:
**Budget allocation.** Allocate 5–10% of your total UA budget specifically for creative testing. Treat this as a learning cost, not a performance budget. Do not hold test campaigns to your standard CAC targets.
**Minimum campaign budget.** Keep your test campaign budget at least 10x your target CPA. Below this threshold, the algorithm cannot exit the learning phase and results are statistically meaningless.
**Ad group size.** Run 5–6 creatives minimum per ad group, up to 10. Meta's algorithm prioritizes only 1–2 creatives for the majority of spend. The remaining creatives must be present for the algorithm to make that determination. Do not over-split.
**Rotation timing.** By week three of any test campaign, refresh your creative lineup actively. Do not wait for explicit performance decline. Creative fatigue develops before it appears in your top-line metrics.
| Daily Budget | Test Structure | Ad Group Size |
| --- | --- | --- |
| Under $500/day | 1 campaign, 1 ad group | 5–10 creatives total |
| $500–$5,000/day | Dedicated test groups + BAU groups | 5–10 per group |
| Over $5,000/day | Multi-campaign, multi-GEO | Isolate false positives per GEO |
At under **$500/day**, focus on one platform (iOS, US) and one optimization event. Splitting ad groups at this budget level fragments data below what the algorithm can learn from.
At **$500–$5,000/day**, you can run dedicated test ad groups alongside BAU (Business as Usual) groups, testing new creatives without putting your proven performers at risk.
## Read Results Without Getting Fooled
Two patterns trip up nearly every team running these tests for the first time.
### 1\. The Win Rate Signal
On Meta and TikTok, the algorithm concentrates spend on whichever creatives in your ad group generate the most optimization events. A creative that earns spend consistently is proving it converts the algorithm's target audience. A creative that receives minimal spend, despite being live in the same ad group, is being deprioritized based on early performance signals.
The minimum threshold for a credible winner: **a 5% win rate across your test cohort.** Below that, review your testing methodology before drawing conclusions from any individual result.
Before scaling a winner, look below the surface. A creative that shows no overall lift might conceal strong lift for a specific segment: users aged 25–34, iOS vs. Android, or users from a specific acquisition source. Segment your results by device, geography, and onboarding behavior before declaring a creative a loser. Some of the strongest performers in a test cohort are hidden inside aggregate results that look mediocre at the account level.
### 2\. Confirming Real Winners vs. False Positives
A false positive is a creative that performs well at low spend, then collapses when you scale it. The algorithm initially found an easy-to-convert micro-segment. That segment is small. When you force more budget onto the creative, performance drops because the algorithm has exhausted the available audience.
**The fix**: move the candidate creative into a dedicated ad group with a meaningful daily budget. Give it **3–5 days** at real spend. If CAC holds, it is a genuine winner. If it collapses, you have avoided the cost of scaling a false positive into BAU.
[How to Test App Ad Creatives Without Burning Budget.webp]
## Why Creatives Convert Differently at the Subscription Level
The install-to-trial rate of a creative tells you whether it attracts curious users. The install-to-subscription rate tells you whether it attracts committed ones. These are not the same thing, and the gap between them is often explained by a single factor: creative promise vs. in-app experience alignment.
Every creative sets a motivation peak — the specific promise that makes a user install. If the paywall appears before the user experiences that promise, they hesitate. If it appears after motivation fades, they leave. StartApp School's Paywall Optimization curriculum is direct on this: timing matters more than copywriting. Contextual paywalls, those appearing after user effort or partial results, outperform paywalls shown immediately after install, not because the paywall design is better, but because the user has already experienced part of what the creative promised.
A fitness app creative that promises "lose 5kg in a month" creates a specific expectation. If the user hits a paywall on the first screen, before experiencing any of the app's capability, the subscription feels premature. The same creative paired with a paywall appearing after the first workout assessment converts at a meaningfully higher rate. The creative did not change. The alignment between promise and paywall timing changed.
This is why two creatives with identical CTR and trial-start rates can produce different subscription outcomes. The creative that attracts users whose motivation peaks earlier in the onboarding sequence converts better at the subscription step, independent of how the paywall looks. Testing creatives in isolation, without accounting for the paywall moment they feed into, produces incomplete results.
For a category-level example of how this plays out in practice, see [Creative Testing for Fitness Apps: Install Metrics vs. Subscription Revenue](https://www.airbridge.io/en/blog/creative-testing-fitness-apps-install-metrics-vs-subscription-revenue).
## Close the Loop at the Subscription Level
Creative testing based on CTR or install-to-trial gives you half the picture. The metric that closes the loop is **install-to-subscription rate by creative**: which ad concept is actually driving paying users, not just curious installers.
To see this, you need your ad spend data connected to subscription events (Start Trial, Subscribe, Renew) in a single view. That connection also surfaces [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) per creative, ROAS at the subscriber level rather than the install level, and retention signals by acquisition source. These are the signals that determine which creatives generate revenue and which generate churn.
Teams that build this measurement infrastructure compound their advantage over time. Among apps tracked across 16,000+ products, those that run structured experiments consistently, with an average of 14.7 experiment runs among active testers, achieve substantially higher revenue than those that test ad hoc. The mechanism is straightforward: every structured test produces one learnable insight, and those insights accumulate.
Connecting [self-attributing network](https://www.airbridge.io/en/glossary/self-attributing-network-san) data (Meta, Google, Apple Search Ads, TikTok) to your subscription funnel, across the right [attribution window](https://www.airbridge.io/glossary/attribution-window), is what makes this possible. If you want to see which creatives are actually driving paying subscribers, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) shows install-to-subscription performance at the creative level across those four channels. Start free with 15K installs.
**Further reading:** See how subscription-level data changes creative decisions in practice: [Creative Testing for Fitness Apps: Install Metrics vs. Subscription Revenue →](https://www.airbridge.io/en/blog/creative-testing-fitness-apps-install-metrics-vs-subscription-revenue)
---
## App Onboarding Before the Paywall: 5 Steps That Convert
URL: https://www.airbridge.io/en/blog/5-steps-app-onboarding-before-the-paywall
Language: English
Published: 2026-05-09
Author: Luke
Most [subscription apps](https://www.airbridge.io/blog) that struggle at the paywall don't have a paywall problem. They have an onboarding problem. By the time the paywall appears, users haven't been given a compelling reason to pay. The screen becomes a wall, not an offer.
The data is unambiguous: **82%** of trial starts happen on Day **0** ([State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Your onboarding sequence (every screen, every question, every small action you ask users to take) sets the ceiling for your trial Conversion - Rate before users ever reach the pricing screen. Most apps waste this window by either rushing to the paywall before motivation peaks or burying value so far into the flow that the opening excitement is already gone.
This guide walks through how to design app onboarding for conversion: specifically the sequence that builds motivation, earns trust, and delivers users to the paywall at exactly the right moment.
## Key Takeaways
- **Timing beats design.** The biggest paywall conversion killer is not copy or layout. It is showing the paywall before user motivation peaks. A simple paywall shown at the right moment outperforms a polished one shown too early.
- **82% of trials start on Day 0.** Your onboarding and first paywall exposure determine most of your trial conversion. There is no recovery mechanism once that session ends without a start.
- **Onboarding + trial is the highest-converting setup.** Apps pairing a structured onboarding flow with a free trial paywall reach **1.78%** install-to-conversion, the top-performing configuration across subscription app categories ([State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions-report/)).
- **Micro-commitments before the paywall increase conversion.** Small interactive steps (goal picks, sliders, short quizzes) build emotional investment and reduce drop-off before the paywall appears.
- **Personalization closes.** Capturing user goals during onboarding and surfacing them on the paywall, even a single string match, outperforms most layout experiments.
- **The paywall should feel inevitable.** When users reach it and think "okay, that makes sense," the flow is working.
## Why Timing Beats Paywall Design
[Screenshot 2026-05-09 at 09.35.55.png]
The moment your paywall appears matters more than how it looks. This is the most underrated principle in subscription app conversion.
Every user who sees your paywall is shaped by three forces: motivation, friction, and trust. Conversion happens when all three align. Motivation is why users want to act. Friction is what makes acting feel effortful. Trust is whether users believe the product will actually deliver what it promises.
### 1\. The motivation curve
Users enter your app at their highest motivation. They just installed it. They believe it might solve something for them. But motivation is not static. It rises when users feel progress and drops when friction accumulates.
Show the paywall before users feel progress, and they are still in evaluation mode. Show it after motivation has faded. Too many setup screens or a slow reveal of value close the window.
### 2\. The three forces behind every conversion
The framework is: conversion = motivation multiplied by trust, divided by friction. Raising motivation by delivering value, building trust by reducing skepticism, and lowering friction by removing unnecessary steps are the three levers.
The most common mistake is treating paywall underperformance as a copy or design problem. Most of the time, it is a sequencing problem. The right message at the wrong moment does not convert. The fix is upstream.
## The 5-Step Onboarding-to-Paywall Flow
[Screenshot 2026-05-09 at 09.36.44.png]
There is no single right onboarding flow, but there is a right sequence. What users experience before the paywall determines how they feel when they see it. These five steps create the conditions for conversion.
### 1\. Capture the job to be done (first **30** seconds)
The first screen should accomplish one thing: understand why the user is here. Ask one goal-oriented question. Not "what features do you want?" but "what are you trying to achieve?"
This matters for two reasons. It personalizes every subsequent screen, which makes the experience feel built for this specific user. It also creates the emotional frame the paywall will reference later: users who have stated a goal are more motivated to take the next step toward achieving it.
Specific question formats that work: "What's your main goal with \[app\]?" or "What brings you here today?" Keep it to one question per screen. The goal is activation, not qualification.
> **Duolingo** applies this before anything else: within the first three screens, users pick why they are learning a language (travel, work, culture) and set a daily time commitment. These two answers shape the entire onboarding that follows, and appear again in the paywall copy.
>
> By putting the user's stated goal first, Duolingo grew its MAU-to-paid rate from roughly **4%** to over **9%** between 2020 and 2025 ([Duolingo Investor Relations](https://investor.duolingo.com/)).
### 2\. Build micro-commitments before showing value
After capturing the job to be done, do not jump immediately to the aha moment. Add **2-3** small interactive steps that increase emotional investment: a slider to set a specific target, a short quiz to shape the experience, or a preference screen to customize their plan.
Small interactive steps before the paywall increase emotional investment and reduce drop-off. The user is no longer a passive observer. Leaving now feels like abandoning work they have already started.
> **Noom** takes this further than almost any subscription app: its pre-paywall onboarding runs to 100+ screens, covering health history, behavioral patterns, and emotional relationship with food. By the time the paywall appears, users have invested 10-15 minutes of real effort.
>
> The drop-off cost has become real. Noom explicitly built this commitment architecture before introducing pricing ([RevenueCat, 2024](https://www.revenuecat.com/blog/growth/web-to-app-onboarding-funnel/)).
The 48 Laws of Subscription App Success documents this effect directly (Law 02): small interactive steps before the paywall increase emotional investment and reduce drop-off. The principle holds whether the quiz has 5 questions or 100.
### 3\. Deliver one clear aha moment before the paywall
Every subscription app has one moment where the user thinks "I need this." Your job is to identify it and guarantee it happens before the paywall. This is not about showing all your features. It is about showing the single most compelling outcome for this specific user.
For a fitness app, this might be a personalized **8-week** plan preview. For a language app, it might be a quick competency assessment result. For a budgeting app, it might be a projection of how much they could save.
The aha moment must feel personal. Because you captured the user's goal in Step 1 and built toward it in Step 2, this preview should reflect their specific situation, not a generic benefit statement.
The aha moment is not a single screen. The 48 Laws of Subscription App Success maps it as a three-part sequence (Law 41): a setup moment where users contribute their data (Steps 1-2), a revelation moment where the app delivers something meaningful (Step 3), and a habit moment where the behavior starts to stick (Step 5). Design for the full sequence, not just the peak.
> **Headspace** anchors its aha moment in a 3-minute guided breathing session delivered during onboarding itself. At the end, the instructor asks users to notice how their internal state has shifted, turning an abstract claim ("meditation helps") into a felt, personal result. The paywall appears immediately after that noticing, not before it ([Behind Login, 2024](https://behindlogin.com/news/headspace-onboarding-a-ux-journey-that-welcomes-and-delights/)).
### 4\. Time the paywall to the motivation peak
The paywall should appear immediately after the aha moment. Not three screens later. Not after account creation. Right after the moment that makes users think "this works."
Before showing any paywall, run a three-check test:
- Has the user expressed intent? (Step 1 completed)
- Has the user experienced or seen value? (Step 3 delivered)
- Does the paywall help them move forward, rather than blocking progress?
If any answer is no, the paywall is appearing too early. The best paywall moments feel inevitable, not surprising. "Users should think: okay, that makes sense."
> **Duolingo** places its paywall after users complete a first lesson, letting them experience value before asking for payment. This post-value sequencing is a core part of its conversion strategy, consistent with StartApp School's paywall placement framework (Paywall Optimization Lesson 05).
### 5\. Personalize the paywall with onboarding data
Surface the user's goal from Step 1 directly in the paywall headline. This is a single string replacement. "Get fit in 30 days" from the goal screen becomes "Start your 30-day fitness plan" on the paywall.
The paywall is no longer a generic pricing screen. It is the logical next step in the user's own stated journey. Users who see their words and their goal reflected back in the paywall copy convert at meaningfully higher rates than those facing a generic benefits list.
> **Noom** is the clearest example: after 100+ onboarding screens worth of questions, the paywall headline reads "Your personalized health plan is ready", referencing the specific goal weight the user typed earlier in the flow. It is not a generic trial prompt; it is the delivery of something the user already asked for.
The 48 Laws of Subscription App Success calls this the one-question rule (Law 29): capture the user's goal in onboarding, then echo it in the paywall headline. The transfer requires no complex logic. One string replacement is enough.
## Onboarding Paywall vs. Contextual Paywall
Two paywall types serve different parts of the subscription funnel. Understanding which to use, and when, prevents an onboarding flow from undercutting its own conversion.
### 1\. The onboarding paywall
StartApp School's lifecycle research (Lesson 04) adds precision to the Day 0 data: the vast majority of initial subscriptions happen within the first 24-48 hours after download. Act on that window with a strong onboarding flow and the numbers follow.
An onboarding paywall appears at the end of the first-run experience, before the user enters the core product. It works because motivation is highest right after install. In well-optimized apps, onboarding accounts for roughly **50%** of trial starts ([RevenueCat, 2024](https://www.revenuecat.com/blog/growth/guide-to-mobile-paywalls-subscription-apps/)). This is the paywall type this entire framework is designed to support.
### 2\. The contextual paywall
A contextual paywall triggers when a user reaches a gated feature or hits a limit in the free version. It appears after the user has already spent time in the product and is actively trying to accomplish something the free tier does not allow.
The dynamics here are different. The user has already invested time and effort. Now the paywall is not blocking them; it is enabling completion. Blocking feels aggressive; enabling feels fair. Contextual paywalls require a free experience strong enough to generate that investment in the first place.
### 3\. Which to use
The same Lesson 04 research provides an important counterpoint: late converters retain better than Day 0 converters. Users who needed more time to decide before subscribing are often more committed once they do. This is part of why contextual paywalls generate lower conversion volume but higher-quality subscribers.
Use the onboarding paywall if your value proposition can be demonstrated compellingly in the first session. Use the contextual paywall if your product needs more time in the user's hands before asking for payment.
Most high-performing subscription apps run both. The onboarding paywall captures users who are ready to convert in the first session, which is the majority. The contextual paywall captures users who needed more time.
## What Your Paywall Screen Must Do
Once users reach the paywall, at the right moment and after the right onboarding, the screen needs to confirm value, reduce risk, and present price clearly. Most paywalls that underperform fail at step one: confirming value.
### 1\. Get the information order right
Outcome first. Value explanation second. Reassurance third. Price fourth. CTA last.
Paywalls that lead with price create a friction spike before the user has finished evaluating the offer. A paywall headline that surfaces the user's goal (from Step 5 personalization) serves as the outcome statement. A value block showing what they specifically get (not a generic feature list) serves as the explanation. Reassurance comes last: social proof, trial terms, and cancellation clarity.
### 2\. Use trust signals that match the user
Trust answers one question: "Will this actually work for me?" Generic testimonials and superlative claims reduce trust. What builds it: social proof that matches user goals specifically, clear trial terms that spell out the length and billing date, and an honest risk reversal such as "cancel anytime, no charge until Day X."
Personalized testimonials matched to onboarding segments (for example, a muscle-gain result shown to users who selected muscle gain as their goal) outperform generic five-star quotes because they answer the user's specific question, not the general one.
### 3\. Use a loading screen before the paywall
A pre-paywall loading screen ("Building your 12-week plan..." or "Analyzing your goals...") creates anticipation and signals that the app has done work based on the user's input. When the paywall appears after this, it feels like the delivery of something personalized, not an interruption.
The transition from "loading your results" to "here is your plan, unlock it with a free trial" is a paywall moment that earns conversion rather than demanding it.
**Noom** executes this with a personalized weight-projection graph that animates while the app "builds your program." The user watches their projected progress curve render in real time before the paywall appears. The loading screen is not filler. It is the paywall's opening argument.
## Conversion Benchmarks by Paywall Setup
| Paywall Setup | Key Metric | Notes |
| --- | --- | --- |
| Onboarding + Trial | 1.78% install-to-conversion | Highest-performing setup overall |
| Hard paywall (no trial) | 12.1% download-to-paid (D35) | Higher LTV per user, lower volume |
| Freemium + contextual | 2.2% download-to-paid (D35) | Higher top-of-funnel volume |
| Trial length 17-32 days | 45.7% trial-to-paid | Highest trial conversion rate |
| Trial length 3-7 days | 26.8% trial-to-paid | Common default; significantly underperforms |
Trial length is a conversion lever that most apps overlook. Trials lasting **17** to **32** days convert at **45.7%**, nearly double the **26.8%** rate of the most common **3-7** day trial (State of Subscription Apps 2025). The trial experience also compounds: users who go through a structured trial boost first renewal rates by up to **60%** (State of In-App Subscriptions 2026) compared to direct purchasers in categories where trials outperform upfront payment. Understanding your [lifetime value (LTV)](https://www.airbridge.io/glossary/lifetime-value-ltv) by trial cohort helps you choose the right trial length for your app.
One caveat worth noting: a free trial is not the right choice for every user. StartApp School (Paywall Optimization Lesson 08) notes that no-trial paywalls can outperform trial paywalls when motivation is already high. For users who arrived from a strong ad with a clear value proposition, adding a trial can introduce friction rather than reduce it. The default should be trial-first, but segment by intent before applying it universally.
## 3 Mistakes That Hurt Paywall Conversion
### 1\. Showing the paywall before intent is established
The most common mistake: moving to the paywall before the user has stated a goal or experienced any value. The screen asks users to pay for a promise they have not yet evaluated. They don't convert. They leave.
The fix is the three-check rule from Step 4. Do not render the paywall until the user has expressed intent, felt value, and reached a natural stopping point. If any check fails, add one more onboarding step before proceeding.
### 2\. Ignoring onboarding data on the paywall
Most apps run personalized onboarding (goal collection, quiz flows, persona selection) and then show the same generic paywall to every user. This wastes the emotional investment built during the first session.
The fix is one mapping rule: take the user's primary goal from onboarding and use it as the paywall headline variable. This single change tends to outperform most layout experiments. The paywall already has the user's attention. Make it feel like it was made for them.
### 3\. Stopping user guidance after conversion
Conversion is not the end of the job. Users who subscribe still need guidance into their first premium action. The period immediately after upgrade, where many apps go silent, is where early churn begins.
A new subscriber who does not reach a meaningful premium aha moment in their first session is a churn risk before the first renewal. After the paywall converts, run a brief premium onboarding: show where premium features live, guide the user to their first premium action, and set the expectation for what they just unlocked. Treat a new subscriber like a new user entering a new product.
> **Headspace** structures this deliberately: new subscribers enter a 10-day "Basics" course immediately after the paywall converts: a guided sequence that walks them through their first premium sessions one at a time. The course exists specifically to prevent the post-subscription drop-off that kills early renewals ([Built for Mars, 2024](https://builtformars.com/case-studies/headspace)).
The 48 Laws of Subscription App Success states this directly (Law 04): onboarding should not stop once someone converts or hits the paywall. Conversion is the beginning of the subscriber relationship, not the end of the growth team's job.
## Build the Flow First, Test the Paywall Second
The sequence matters more than any individual element. An onboarding flow that captures intent, builds commitment, delivers value, and times the paywall correctly will outperform a polished paywall shown at the wrong moment every time.
Start with the foundation: define your one aha moment, map a **3-5** step flow to build toward it, set the paywall to appear immediately after. Once the sequence is live, measure step-level funnel drop-off rather than [conversion rate](https://www.airbridge.io/glossary/lifetime-value-ltv) in isolation. Paywall conversion problems almost always trace back to a broken step upstream, not the paywall screen itself.
Test timing before testing design. When the timing is right, the design matters much less than most teams assume.
---
## How to Track Where Your Paying App Users Actually Come From
URL: https://www.airbridge.io/en/blog/how-to-track-where-paying-app-users-come-from
Language: English
Published: 2026-05-08
Author: Harper (Trang Nguyen)
You're spending money on Meta, Google, and TikTok. Users are installing. Some start free trials. But when a subscriber shows up in your billing dashboard, you have no idea which campaign sent them — or why.
For most early-stage subscription apps, this question about user tracking goes unanswered. And that gap between "someone installed" and "someone actually paid" is where budgets get wasted.
**Key Takeaways**
- **The channel with the most installs is rarely the channel with the most subscribers.** Only subscriber-level tracking reveals which channel drives revenue — not just downloads.
- **Cost per Paying User (CPPU) is the metric that actually tells you where to spend.** It runs **4–5x** your Cost per Install on average, so optimizing for CPI alone points you in the wrong direction.
- **Three things must connect** to track subscribers by source: your ad channels, an attribution SDK in your app, and your subscription platform.
- **Only 1.9% of downloads convert to a paying subscriber within 35 days** at the median — knowing which channel sends those rare subscribers is what separates informed budget decisions from guesses.
- **Setting up subscriber-level tracking takes less than a day** with the right stack.
## Installs Tell You Clicks, Not Conversions
Every ad channel reports how many users it sent. Meta reports installs. Google reports installs. TikTok reports installs.
None of them tell you how many of those users subscribed.
### 1\. The CPI trap
Cost per Install (CPI) is a top-of-funnel metric. It measures what it costs to get a user onto your app — not what it costs to get a user who pays.
The gap is wider than most founders expect. Analysis across thousands of subscription apps shows that **Cost per Paying User (CPPU) runs 4–5x higher than CPI.** If your Meta CPI is $4, your actual cost per subscriber from Meta is likely $16–$20 — assuming that channel even converts to paid users at all.
CPI is "very top funnel" — it doesn't tell you much about who actually subscribes.
### 2\. The subscriber blind spot
Your ad channels only see the install. Your billing platform — RevenueCat, Adapty, or your own backend — sees the subscription. Neither system, by default, knows about the other.
The result: you can tell which channel sent users, and you can tell who subscribed, but you cannot connect those two facts. You are flying with two half-instruments.
**Among subscription apps analyzed, just 1.9% of downloads convert to a paying subscriber within 35 days at the median.** Top-performing apps (P90) reach **8.5%** — a **4.5x gap** driven by knowing which channels send high-intent users and which ones don't.
## What App Attribution Actually Is
App attribution is the system that connects the click on an ad to the moment a user subscribes. It closes the gap between your ad channels and your billing platform.
### 1\. How attribution connects the click to the subscription
Every subscriber follows some version of this path:
1. They see your ad on Meta, Google, TikTok, or Apple Search Ads
2. They click and install your app
3. They start a free trial
4. They subscribe
Without attribution, you can see steps 1 and 4 separately — but not together. Attribution adds the thread that connects them: it records which ad your subscriber clicked before they installed, then follows their journey through to the subscription event.
[Track paying users.webp]
### 2\. Why the trial period creates a tracking gap
For subscription apps, attribution has one extra complication. **82% of free trials start on Day 0** — the same day a user installs. But the subscription payment comes days or weeks later.
If your [attribution window](https://www.airbridge.io/glossary/attribution-window) is shorter than your trial length, the system loses the connection between the ad click and the eventual subscription. That subscriber shows up as "organic" in your reports, even though a paid ad drove them. This is a structural gap — not an app problem.
## The Three Parts of Subscriber Tracking
Closing the subscriber blind spot requires three components working together.
[Track paying user 2.webp]
### 1\. Ad channels
Meta, Google, Apple Search Ads, and TikTok each record the clicks that lead to installs. They are known as [Self-Attributing Networks (SANs)](https://www.airbridge.io/glossary/self-attributing-network) — they determine attribution using their own internal user data rather than external tracking links.
The challenge: each channel only sees its own traffic. They do not communicate with each other, and they cannot see what happens inside your app after the install.
### 2\. Attribution layer (SDK in your app)
This is the missing piece for most early-stage apps. An attribution SDK sits inside your app and performs two jobs.
First, it matches incoming installs to the ad clicks that drove them — using device identifiers and click timestamps. Second, it records the subscription events (trial start, subscribe, renew) and links them back to the original ad.
### 3\. Subscription platform
Your subscription platform — RevenueCat, Adapty, or your own backend — is the record of truth for revenue. It knows the amount paid, the plan type, and the timing of each subscription.
What it does not know by default is which ad drove that subscriber. The attribution SDK sends it that context, closing the loop — so you can see [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) broken down by channel.
## Set Up Subscriber Tracking in 4 Steps
Once you understand what must connect, implementation is straightforward.
### 1\. Install the attribution SDK
Choose a tool that supports your four main ad channels: Meta, Google, Apple Search Ads, and TikTok. Add the SDK to your iOS and Android app. This typically requires a few hours of developer time.
### 2\. Connect your ad channels
In your attribution dashboard, authenticate each ad channel. This gives the tool access to the click data it needs to match ad clicks to app installs.
### 3\. Configure your subscription events
Set up the funnel events that matter: trial start, subscribe (or purchase), and renewal. These are the signals that tell you whether a channel sends users who pay — not just users who explore.
### 4\. Connect your subscription platform
Link your attribution tool to RevenueCat, Adapty, or your billing system. This sends revenue confirmation back to the attribution layer so you can see cost per subscriber broken down by channel — not just installs.
## What to Look at Once Your Data Flows
Once tracking is live, the metrics that matter look different from what your ad dashboards show.
| Metric | What it measures | Why it matters |
| --- | --- | --- |
| Cost per Subscriber (CPS) | Ad spend ÷ paying subscribers from that channel | Replaces CPI as your decision metric |
| Trial-to-Paid Rate by Channel | % of trialists who subscribe, per channel | Reveals which channels send high-intent users |
| LTV by Channel (D30/D60) | Revenue per subscriber per channel over time | Shows long-term value, not just upfront conversion |
| Attribution Rate | % of subscribers with a known channel source | Tells you how complete your tracking setup is |
The trial-to-paid rate is the most revealing early signal. **The median trial-to-paid conversion is 34.8%** across subscription apps — but this varies by channel quality, not just app quality. A channel converting at **45%** is worth roughly twice as much as one converting at **20%**, even if CPI looks the same.
For a deeper look at how channel-level tracking affects budget decisions, see [Which Ad Channel Drives Subscriptions?](https://www.airbridge.io/en/blog/app-subscription-channel-attribution)
## FAQs
### Do I need subscriber tracking before I start running paid ads?
Yes — and the setup should happen before your first paid campaign, not after. If you launch paid UA without attribution in place, your first weeks of data will be unattributed. You will not be able to connect early spend to the subscribers acquired during that period. Setup takes less than a day; there is no reason to delay.
### I'm already using RevenueCat — doesn't that handle this?
RevenueCat tracks payments, renewals, and churn — it does not know which ad sent that user. An attribution layer connects those two datasets. Most attribution tools have native integrations with RevenueCat and Adapty that take minutes to configure.
## Stop Guessing. Start Measuring.
Budget allocation without subscriber-level data is guessing at scale. Every dollar you put into a channel you cannot measure is a dollar that could be better placed somewhere that converts.
The setup cost is low. The return — knowing which channels send subscribers and which ones send churn — compounds with every dollar you spend afterward.
If you're ready to see which channels drive subscriptions — not just installs — [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects your Meta, Google, Apple Search Ads, and TikTok ad spend to subscription revenue. It includes **15K free attributed installs**, subscription events pre-configured, and native integrations with RevenueCat and Adapty — built for early-stage subscription apps, with 4 ad channels and up to 2 third-party integrations on the base plan.
---
## How to Choose App Ad Channels: Meta Ads, Google Ads, TikTok Ads, or Apple Ads
URL: https://www.airbridge.io/en/blog/how-to-choose-app-ad-channels
Language: English
Published: 2026-05-08
Author: Harper (Trang Nguyen)
Most subscription app founders pick their first ad channel based on what a peer recommended, or what feels familiar. Meta Ads because everyone's on it. TikTok Ads because it's trending. Google Ads because it feels safe.
None of these are good reasons. The right channel depends on your specific app, and choosing one that doesn't fit your situation costs more than just a test budget.
**Key Takeaways**
- **The right channel depends on your specific app.** Your category, budget, creative capacity, and platform priority all influence which channel makes sense.
- **Splitting budget across channels too early can reduce performance.** Each channel needs sufficient data to optimize. Concentrating spend on one channel first gives it the best chance to work.
- **Install cost is not the same as subscriber cost.** The actual cost of acquiring a paying user, or Cost per Paying User (CPPU), is higher than the cost of an install. Optimizing for CPI alone can be misleading.
- **Each channel serves a different user intent.** Meta Ads converts. TikTok Ads attracts. Google Ads captures searchers. Apple Ads (formerly Apple Search Ads) captures buyers.
- **Prove one channel before adding a second.** Establish reliable performance on your primary channel before expanding.
## Why Your Channel Choice Changes Everything
Splitting a budget across two or three channels at once is a common instinct. It feels like risk management, but in practice it often backfires. Each channel gets too little data to exit the learning phase, your team is stretched across multiple creative formats and dashboards, and none of the channels perform well. Not because they're wrong, but because none had enough resources to function.
### 1\. The real cost of a subscriber
Your visible cost is Cost per Install (CPI). Your real cost is **Cost per Paying User (CPPU)**. [Analysis across 75,000+ subscription apps shows CPPU runs 4–5x higher than CPI.](https://www.revenuecat.com/state-of-subscription-apps-2025/) A $4 Meta Ads CPI likely means subscribers cost $16–$20, assuming that channel converts to paying users at all. Apps spending under $100K/month consistently outperform split strategies when budget stays concentrated on one channel. It doesn't save money to chase cheaper installs on the wrong channel. It costs more.> **Diagram: The Real Cost of an App Install**
[How to Choose App Ad Channels_1.webp]
### 2\. Why trial-to-paid rate matters more than install cost
Trial-to-paid conversion rates vary significantly across subscription app categories. A channel that drives high install volume doesn't help if those users are unlikely to subscribe, which is why knowing your category's conversion baseline matters before setting CPI targets.
## What Each Major App Ad Channel Is Built For
### 1\. Meta Ads
Meta Ads is the most common starting channel for subscription apps, and usually for good reason. Its algorithm is the most mature in social advertising, and its interest- and behavior-based targeting, including lookalike audiences, lifestyle signals, and demographic profiles, makes it effective for apps with a clear user signal: fitness, wellness, productivity, finance. Advantage+ campaigns have reduced the manual optimization burden significantly, making Meta Ads accessible without dedicated UA expertise.
[How to Choose App Ad Channels_meta.webp]")
### 2\. Google Ads
Google Ads runs across Search, Play Store, YouTube, and Display simultaneously. The key differentiator is intent: users searching "meditation app" or "budget tracker" have already expressed a need. For Android apps specifically, Google Ads is effectively required. It provides direct Play Store placement, and Android CPI runs at roughly a third of iOS equivalents. **Health & Fitness iOS CPI averages $5.78; Business apps hit $14.08.**
One caveat: Google's signal quality on iOS is significantly limited compared to Android. Google performs strongest where it controls the operating system and Play Store data. If your primary revenue comes from iOS subscribers, Meta Ads or Apple Ads will generally outperform Google Ads as a starting channel.
[How to Choose App Ad Channels_gg.webp]")
### 3\. TikTok Ads
TikTok Ads excels at discovery. Its algorithm reaches users who didn't search for your app, making it powerful for awareness and top-of-funnel reach. CPM and CPC tend to run lower than Meta Ads equivalents, and the platform's user base skews toward younger demographics. TikTok Ads' attribution is less mature than other channels, and some conversions it registers may complete later through other touchpoints. It requires a consistent pipeline of native-style, short-form video, and will exhaust your creative library quickly, regardless of budget, if production can't keep pace.
[How to Choose App Ad Channels-tiktok.webp]
### 4\. Apple Ads (formerly Apple Search Ads)
Apple Ads places your app in App Store search results when users search for relevant keywords. It is the highest-intent channel available for iOS. These users are already on Apple's platform, already searching, already in a buying mindset. Apple Ads is also the only major channel unaffected by [App Tracking Transparency (ATT)](https://www.airbridge.io/glossary/app-tracking-transparency-att), meaning clean attribution without opt-in friction. The trade-off is scale: Apple Ads captures existing demand but cannot create it.
One important constraint: unlike Meta Ads and Google Ads, Apple Ads cannot optimize toward trial starts or subscriptions. It optimizes for taps and installs only, which is why it works as a high-intent closer but hits a ceiling when you try to scale beyond branded and competitor keywords.
[How to Choose App Ad Channels-asa.webp]
## The 4 Factors That Determine Channel Fit
### 1\. App category and LTV
Channel economics only make sense relative to your [Lifetime Value (LTV)](https://www.airbridge.io/glossary/lifetime-value-ltv). **Health & Fitness apps average $27.21 Y1 LTV per subscriber; Business apps average $27.24.** A $5.78 iOS CPI is sustainable for Health & Fitness at $27 LTV. For apps with lower subscriber LTV, that same CPI leaves very little room for payback. Know your category's LTV range before setting CPI targets.
### 2\. Budget level
Below **$10,000/month**, pick one channel and go deep. Each channel needs enough conversion events per week to exit its learning phase. Splitting $5,000 across two channels gives each too little to learn from. Above **$50,000/month**, a second channel becomes testable. Above **$100,000/month**, diversification across multiple [self-attributing networks](https://www.airbridge.io/en/glossary/self-attributing-network-san) starts making economic sense.
A practical rule: set your daily budget at minimum 10x your target CPA. If your cost-per-trial target is $10, your minimum daily budget is $100. Below that, the algorithm does not receive enough conversion signals to optimize delivery.
### 3\. Creative capability
Each channel requires a different format and refresh cadence. The table below also shows the minimum starting assets needed — a practical filter before committing to any channel.
| Channel | Primary Format | Refresh Needed | Min. Starting Assets |
| --- | --- | --- | --- |
| Meta Ads | Static + video (Reels, Stories) | Every 2–3 weeks | 5 creatives per ad group |
| Google Ads | Mixed (images, video, copy) | Monthly | 5 headlines, 5 descriptions, up to 20 images/videos |
| TikTok Ads | Short vertical video (UGC-style) | Weekly | 5–6 ads per ad group |
| Apple Ads | App Store screenshots + metadata | Quarterly | Existing App Store assets only |
### 4\. iOS vs. Android priority
**iOS still generates the large majority of global subscription app revenue**, but Android has crossed **51% of total paid install volume**. For iOS-first apps, prioritize Meta Ads and Apple Ads. For Android-first or dual-platform apps, Google Ads is essential for Play Store placement. For iOS subscription apps focused on conversion, note that TikTok Ads' SKAdNetwork attribution on iOS is more constrained due to platform privacy limits, which can affect measurement accuracy.
## When to Add a Second Channel
Add a second channel when your first is working. Adding channels to diagnose why something isn't performing usually spreads the problem thinner rather than solving it. Three questions determine readiness.
**Do you have clear unit economics on your primary channel?** You need CPPU, subscriber LTV, and D30 ROAS before expanding. Without a baseline, you can't tell whether a second channel is better or worse. You're just adding noise.
**Can you produce creative for two formats simultaneously?** Adding TikTok Ads to a Meta Ads setup means a different format, refresh rate, and production pipeline. If your team is already stretched, quality drops on both.
**Do you have budget to fund both without thinning either?** A split at the $5,000–$10,000/month level doesn't produce two winners. It gives you two channels stuck in the learning phase.
A practical approach: adapt your best-performing creative for the new channel's format, allocate **15–20% of total budget** as a 30–45 day test, and measure CPPU, not installs. If results match or beat your primary channel, scale. If not, reallocate.
### What the first four weeks actually look like
Most founders quit a new channel too early because they don't know what the learning phase looks like from the inside. A channel has not been properly evaluated until it completes all four phases:
- **Week 1 — Anxiety:** CPMs spike and can feel alarming. Ads may sit in review. This is normal algorithmic exploration, not failure.
- **Week 2 — Hope:** CPMs drop significantly. First real daily purchases appear. Creative optimization becomes possible for the first time.
- **Week 3 — Confidence:** Volume becomes consistent. The window opens for refreshing or expanding creatives.
- **Week 4 — Stabilization:** Your addressable audience is found. CPA stabilizes. Now you have enough signal to make a scaling or channel-addition decision.
Many subscription apps write off a channel in Week 1 that would have performed well by Week 4. Before adding a second channel, your primary channel should have completed all four phases.
## What to Track Once Your Channel Is Live
Ad dashboards report installs. Your billing platform reports subscribers. Neither connects those facts by default, which is where most early-stage apps lose budget without realizing it. The metrics that matter look different from what ad dashboards show.
| Metric | What it measures | Why it matters |
| --- | --- | --- |
| Cost per Subscriber (CPS) | Ad spend ÷ paying subscribers from that channel | Replaces CPI as your actual decision metric |
| Trial-to-Paid Rate by Channel | % of trialists who subscribe, per channel | Reveals which channels send high-intent users |
| LTV by Channel (D30 / D60) | Revenue per subscriber per channel over time | Shows long-term value, not just upfront conversion |
| Attribution Rate | % of subscribers with a known channel source | Shows how complete your tracking setup is |
You need to know your cost per subscriber by channel, not just CPI. [Airbridge](https://www.airbridge.io/en) shows you which campaigns drive actual subscriptions, starting free with 15,000 installs.
**The apps that scale fastest on paid UA are not the ones that tested every channel simultaneously. They're the ones that mastered one.**
## FAQs
**Should I run Meta Ads and Google Ads at the same time from the start?**
Not if your budget is below $10,000/month. Splitting that across two channels gives each too little data to exit the learning phase. Pick one, run for 30–45 days, and expand only once you have unit economics to compare a second channel against.
**Is TikTok Ads worth testing for a subscription app?**
TikTok Ads can work well for subscription apps, particularly in lifestyle, health, and wellness verticals where short-form video content resonates. It offers broad reach and competitive CPMs. However, it requires a consistent pipeline of native-style video, and performance tends to be more volatile than Meta Ads, making it a harder place to validate product-market fit or stretch an early budget.
For most subscription apps, the practical approach is to establish a primary channel first, then test TikTok Ads with a portion of budget once your core unit economics are stable. For a deeper guide, see [The Essential Guide to TikTok Ads for Subscription Apps](https://www.revenuecat.com/blog/growth/tik-tok-ads-guide-mobile-apps/) by RevenueCat.
**Do I need Apple Ads if I'm already running Meta Ads?**
For iOS subscription apps, Apple Ads is worth adding once your Meta Ads budget is established. It captures high-intent users who are actively searching for apps like yours, a segment that Meta Ads doesn't typically reach. Apple Ads doesn't replace Meta Ads. It closes a different part of the funnel. Use it alongside Meta Ads rather than instead of it.
---
## How to Set Your First 30-Day Paid UA Plan for Subscription Apps
URL: https://www.airbridge.io/en/blog/30-day-paid-ua-plan-for-subscription-apps
Language: English
Published: 2026-05-07
Author: Luke
You have the idea to advertise your subscription app. But when you open Meta or Google Ads, you're faced with endless options: channels, budgets, audiences, creatives, placements. Without an execution plan, most early-stage founders either freeze or spend money inefficiently. The result is burned budget with minimal learning.
The difference between a founder who wastes $10K and one who turns it into a profitable playbook isn't luck — it's structure. This guide gives you a step-by-step plan to spend your first $10K–$50K with intention, learn fast, and build the foundation for sustainable growth.
**Key Takeaways**
- **Unit economics come before budget.** Know your LTV (Lifetime Value) and CAC (Cost to Acquire a Customer) target before you set daily spend. Otherwise you're guessing at a number with no basis.
- **Start with one channel.** Meta, Google, or TikTok. Prove one works before you diversify. Spreading budget thin across all channels kills both reach and learning.
- **Structure your 30 days as three phases.** Week 1 is learning (spend 30% of budget). Weeks 2–3 is testing and scaling (50%). Week 4 is validation (20%).
- **Measure cohorts, not actuals.** Ad networks, your attribution tool, and your app's revenue dashboard all report different numbers. Establish one source of truth before Week 2.
- **Paywall placement matters before ad spend.** Fix your onboarding and paywall timing first, then scale acquisition. This single lever affects your CR (conversion-rate) more than any creative.
---
[Screenshot 2026-05-07 at 15.01.01.png]
---
## Phase 1: Calculate Your Unit Economics & Set Up Your Paywall (Days 1–5)
Before you spend a single dollar on ads, you need to know what that dollar should achieve. This phase also includes paywall validation—the strongest lever you control before acquiring users.
### 1\. Know Your LTV (Lifetime Value)
> **LTV = Average subscription price × Average months a subscriber stays**
**Quick example:** $9.99/month × 8 months = **$80 LTV**
**Adjust upward if:**
- You have upsells (annual plans, add-ons, consumables) per [RevenueCat's Hybrid Monetization Guide](https://www.revenuecat.com/blog/growth/hybrid-monetization-techniques/)
- Targeting high-LTV markets like North America
**Industry benchmarks (12-month, per paying user):**
| Category | Median | P90 |
| --- | --- | --- |
| Health & Fitness | $27 | $86 |
| Business | $27 | $112 |
| Education | $18 | $75 |
| Gaming | $8 | $32 |
**If brand new:** Use **$15–$25** conservatively. Refine after 100–200 paying users.**If you have data:** Pull D7, D30, D60 retention from RevenueCat or Adapty and calculate actual LTV.
### 2\. Reverse-Engineer Your CAC Target
> **A common industry benchmark: LTV should be at least 3–5x your CAC to sustain growth.**
**If LTV = $20, CAC target = $4–$6 per paying subscriber**
**Why 3–5x buffer?**
- Ad platform overhead (~15% variance)
- Not all installs → trials; not all trials → subscriptions
- Costs rise as you scale
**CAC per install vs. CAC per subscriber:** Your actual conversion funnel looks like:
| Stage | Rate | Example |
| --- | --- | --- |
| Installs → Trials | 4–12% | 1,000 installs → 60 trials |
| Trials → Paid | 35–45% | 60 trials → 24 subscriptions |
| Total | 1.4–5.4% | 1,000 installs → 24 subs |
**If you want $5 CAC per subscriber:** $5 ÷ 2.4% = **$208 CAC per install needed**
**Reality check:** iOS CPI is $5–$15. Android is $1–$4. If your CAC per install target is high, that's fine. It just means lower trial-to-paid conversion. Fix it with paywall optimization.
**Paywall impact:** Hard paywalls convert at 12.1% (D35) vs. 2.2% for freemium. That's **5.5x higher.** Choose your model first; it determines your budget size.
### 3\. Define Your First 30-Day Goal
Now set a realistic target for the 30-day period.
**If you're bootstrapped:** Aim for **500–1000 trial starts** or **50–150 paying subscribers.** At $3 CPI, that's $1,500–$3,000 spend.
**If you have seed funding:** Aim for **3,000–5,000 trial starts** or **300–500 paying subscribers.** At $3 CPI, that's $9,000–$15,000 spend.> Your goal is not maximum revenue in 30 days. Your goal is **learning which channels, audiences, and creatives work for your app.** Revenue is secondary.
### 4\. Validate Your Paywall Before Week 1
**82% of free trials start on Day 0.** Paywall placement matters more than any creative. Quick checks:
**Paywall timing:** Does the user see value _before_ hitting the paywall? Hard paywalls see 12.1% D35 conversion versus 2.2% for freemium. That's 5.5x higher.
**Trial length:** Longer trials (17–32 days) convert at 45.7% vs. shorter trials (3–7 days) at 26.8%. If cash flow is tight, optimize paywall messaging instead of extending trial.
> Don't launch ad campaigns until your paywall passes this test: Have 10 friends install your app and count how many hit the paywall without abandoning.
---
## Phase 2: Pick the Platform That Will Win for Your App (Days 5–7)
You have four main options for user acquisition: Meta, Google App Campaigns, TikTok, or Apple Search Ads. Trying all four simultaneously in a first 30-day campaign guarantees confusion.
**Choose one.** Here's how to decide:
### 1\. Meta (Facebook/Instagram) — Maximum Control & Scale
**Best for:** First-time marketers, all app categories.
Meta offers the strongest combination of targeting precision and scale for subscription apps. You can target by interest, behavior, lookalike audiences, and detailed demographics. Costs are moderate: $3–$8 CPI depending on category.
**Setup:** Create Advantage+ Shopping Campaigns (Meta's automated format). Provide 3–5 app install creatives (mix of video and static). Let Meta's algorithm optimize for you.
**Timeline:** Results typically visible in 3–5 days (varies by budget and category).
**Drawback:** Saturation in competitive categories (dating, fitness) means higher costs and faster creative fatigue.
### 2\. Google App Campaigns — Algorithmic, Low-Effort Setup
**Best for:** Apps with clear, differentiated messaging; educators.
Google App Campaigns (UAC) use machine learning to find install-hungry users across Google's entire network (Play Store, YouTube, Admob, Gmail, Google Search). Minimal setup: provide headlines, app descriptions, and a few creatives. Google tests combinations and optimizes.
**Setup:** Simpler than Meta. Set budget, provide assets, let Google learn.
**Timeline:** Results typically visible in 5–7 days (varies by budget and category).
**Drawback:** Less targeting control than Meta. You can't pick specific audiences; you trust Google's algorithm.
### 3\. TikTok Ads — Top-of-Funnel, Younger Audiences
**Best for:** Apps targeting Gen Z (under 35), trendy categories (fitness, lifestyle, entertainment).
TikTok excels at awareness but has fast creative fatigue. Users expect TikTok-native, authentic-feeling ads (not polished brand content). CPI can be competitive with Meta, though actual costs vary by category and geography.
**Setup:** Create 5–10 different creative variants (short-form video). Expect to refresh them weekly because users quickly tune out repeats.
**Timeline:** Results typically visible in 2–3 days (varies by budget and category), but require active management.
**Drawback:** Burnout. If your app isn't inherently TikTok-native (fitness, lifestyle), it's harder to create enough organic-feeling creatives.
### 4\. Apple Search Ads — High-Intent Users Only
**Best for:** Apps with strong brand search volume, productivity, utility categories.
Apple Search Ads only show to users actively searching for your app or competitors on the App Store. This is high-intent traffic. CPI is higher ($5–$15) but conversion is strong because users are already in purchase mindset.
**Setup:** Bid on your brand terms and top 5–10 competitor terms. Set daily budget ($5–$20).
**Timeline:** Results typically visible in 1–2 days (varies by search volume). Easiest to start; hardest to scale.
**Drawback:** Only scales if people are searching for your category. Not suitable for awareness or new categories.
### Decision Tree
Ask yourself:
- Are you optimizing for reach and learning? → Start with **Meta** (biggest audience, best targeting).
- Do you want hands-off, algorithmic optimization? → Start with **Google App Campaigns**.
- Is your target audience Gen Z and is your app trendy? → Start with **TikTok**.
- Do users already search for your app or competitors? → Start with **Apple Ads.**
> Our recommendation for 90% of first-time founders: Meta or Google. Both have strong learning curves and moderate costs.
---
## Phase 3: Your First 30 Days: The 30-50-20 Budget Blueprint (Days 7–9)
Don't spend your entire budget in Week 1. Structure it like a scientific experiment: form a hypothesis, test, measure, refine.
Here's a practical starting framework:
### Budget Structure for a $30K Plan
| Week | Goal | Budget | Daily Spend | Purpose |
| --- | --- | --- | --- | --- |
| Week 1 | Learn | $9,000 (30%) | $1,286 | Test 2–3 audience segments, 3–5 creatives |
| Week 2–3 | Scale | $15,000 (50%) | $1,071/day | Identify winners, increase spend on top performers |
| Week 4 | Validate | $6,000 (20%) | $857/day | Hold winners at scale, test one new variable |
**For a $10K plan:**
- Week 1: $3,000
- Week 2–3: $5,000
- Week 4: $2,000
**For a $50K plan:**
- Week 1: $15,000
- Week 2–3: $25,000
- Week 4: $10,000
The key principle: **Scale winners by 20–30% per week; kill losers immediately.**
Don't say, "Let me wait until Week 3 to decide." By Day 4–5 of Week 1, you'll see which audience/creative combinations are driving trials and subscriptions. Double down on those. Kill the bottom 2–3 performers.
---
## Phase 4: Don't Get Lost in the Data: Track What Matters (Days 9–11)
If you don't organize your campaigns now, you'll spend Week 3–4 drowning in data you can't interpret.
### Naming Template
Use this structure:
```
{Channel}_{Product/Offer}_{Audience}_{Creative_Type}_{Date}
```
Example:
- `META_Trial30d_FitnessEnthusiasts_Video_Jan15`
- `GOOGLE_Premium_YogaInterest_Static_Jan15`
- `TIKTOK_Freemium_Age1824_UGC_Jan15`
Benefits:
- Automated reporting becomes possible
- When you look back in Week 4, you instantly see which audiences/creatives worked
- Cross-channel analysis becomes trivial
### Ad Group Organization
Within each campaign, organize by **variable you want to test:**
Example Meta campaign structure:
- Campaign: `META_Trial30d_General_Jan15`
- Ad Set 1: Audience = Fitness Enthusiasts (age 25–40)
- Ad Set 2: Audience = Wellness Curious (lookalike of free trial starters)
- Ad Set 3: Audience = Health App Users
- Each Ad Set has 2–3 creatives (video, static variant A, static variant B)
This structure lets you measure: **"Fitness Enthusiasts converted best. Wellness Curious had highest CPI. Let's double Fitness and pause Wellness in Week 2."**
### Tracking Setup
Connect your app to a consolidated attribution tool and RevenueCat S2S events before Week 1 launches.
Why? Because ad networks report installs, but your revenue dashboard (RevenueCat, Adapty) reports subscriptions. They'll never match perfectly. Establish one source of truth using your attribution tool, which consolidates data from Meta, Google, Apple Search Ads, TikTok, and your subscription platform.
**Minimum tracking setup:**
- Mobile app → RevenueCat for subscription events
- Mobile app → Your attribution tool for cross-channel attribution
- Verify that "trial start" and "subscription" events fire correctly before launching ads
---
## Phase 5: From Week 1 Launch to Day 30 Victory: Your Action Plan (Days 11–30)
| Timeline | Action | Key Metric |
| --- | --- | --- |
| Days 11–13: Launch | Set campaigns live. Monitor 24–48hrs of data. | Cost per Trial Start |
| Days 14–16: Prune | Kill bottom 2–3 performers. Increase winners by 20%. | Trial-to-Paid %, Audience quality |
| Days 17–24: Test | A/B test one variable (audience, creative, or copy). Test 3–5 creatives only; 500+ installs per creative before deciding. | Cost per Subscription (CPS) |
| Days 25–29: Scale | Double spend on winners. Expect slight cost-per-trial rise as audience saturates. This is normal. | CPS vs. CAC budget |
| Day 30: Analyze | Compare cohorts. Identify winning audience/channel combo. Go/No-Go: Scale (if CPS < CAC budget) or pivot (if CPS > CAC or trial-to-paid < 30%). | Cohort-level metrics |
**Critical rule:** Test one variable at a time. If you change audience, creative, and copy simultaneously, you won't know what drove the improvement.
### Measuring Your Results (Day 30)
Pull cohort-level data, not actuals. A cohort is: "All users who installed Jan 15–21 from Meta, Fitness Enthusiast audience."
**For this cohort, track:**
- Install-to-trial %
- Trial-to-paid %
- Cost per Subscription
- Day 7 retention
**Go/No-Go decision:**
- **Go (Scale):** CPS < CAC budget, trial-to-paid > 35%, clear winner identified.
- **No-Go (Pivot):** CPS > CAC budget, trial-to-paid < 30%, or no clear winner. Diagnose and adjust one variable for Month 2.
---
## Common Mistakes & How to Avoid Them
**Mistake 1: Spreading budget across all channels.** Meta + Google + TikTok = $2.5K per channel. Not enough data to optimize anything. **Fix:** Spend 80% on one channel, 20% on learning.
**Mistake 2: Tracking installs instead of subscriptions.** You got 1,000 installs for $3K (nice $3 CPI), but only 2% converted to paid. Your real cost per subscriber is $150. **Fix:** Calculate Cost per Subscription (CPS) from Day 1, not CPI.
**Mistake 3: No measurement framework.** Facebook reports 300 installs. RevenueCat reports 80 trials. Your analytics say something else. Which is true? **Fix:** Pick one source of truth (RevenueCat cohorts or Airbridge attribution) and stick with it.
---
## Scale Without Losing Visibility: Automate Your Measurement
As you scale beyond 30 days, manually calculating cohort metrics becomes impossible. Consolidated attribution platforms like **Airbridge** eliminate manual reporting and catch scaling problems in real-time.
Airbridge connects Meta, Google, Apple Search Ads, TikTok, and RevenueCat in one dashboard. You see cohort-level Cost per Subscription, ROAS, and retention instantly. That saves 5+ hours of analysis per week.
---
## How to Know If Your Ads Are Working in 2026
URL: https://www.airbridge.io/en/blog/are-your-ads-working-app-attribution
Language: English
Published: 2026-05-07
Author: Luke
App attribution is the process of connecting a specific ad to the user action that followed: an install, a trial start, or a paid subscription. For subscription apps, simple attribution tracks three events: install, trial start, and subscribe. You don't need an enterprise tool to start. You need the right three metrics and a way to tie them back to your ad spend.
**Key Takeaways**
- **CPI alone doesn't tell you what's working.** Cost per install is a top-funnel metric. The number that matters for subscription apps is cost per paying subscriber — which runs **4–5x higher** than CPI on average.
- **82% of trial starts happen on Day 0.** If you're not measuring what happens in the first session, you're missing most of your conversion data.
- **A simple setup tracks four channels and three events.** Meta, Google, Apple Search Ads, and TikTok — plus install, trial start, and subscription.
- **The median trial-to-paid conversion rate is 34.8%** across 75,000+ apps. Knowing where your app stands tells you whether to fix your paywall or your audience targeting.
- **You can start measuring without a complex setup.** A basic attribution layer connecting four ad channels and three subscription events is enough to validate your first campaigns.
## What App Attribution Actually Means
App attribution answers one question: which ad brought this user here?
Every time a user installs your app, something drove that action — a Meta ad, an Apple Search Ads result, a TikTok video, or organic search. Attribution is the mechanism that connects the install back to that source. Without it, you see installs but not _where_ they came from.
For subscription apps, attribution extends beyond the install. A user who installs and never subscribes costs you money. A user who installs, starts a trial, and converts to a paid plan earns you money. The difference between those two users, and which ad brought which, is what app attribution reveals.
One critical challenge: every data source reports a different number. Your Meta Ads dashboard shows one install count. The App Store shows another. Your analytics tool shows a third. This is what practitioners call the "source of truth war." That is the reason you need a single attribution layer to reconcile them.
### Why do numbers never match?
Meta, Google, TikTok, and Apple Search Ads are all [Self-Attributing Networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network-san). Each one claims credit for installs independently, without checking whether another platform already claimed the same user. When your audiences overlap across channels, the same install gets counted two or three times.
This is not fraud. It is how SANs are designed.
There are two ways to fix it:
- **Use a single attribution layer.** One tool that sits across all channels picks the winning touchpoint and prevents double-counting.
- **Run incrementality tests.** Pause a campaign for two weeks and measure what changes. If installs don't drop, the channel was claiming organic users — not driving new ones.
## Why CPI Misleads You — and What to Track Instead
[Cost per install (CPI)](https://www.airbridge.io/en/glossary/cost-per-install-cpi) is the most commonly reported metric in paid user acquisition. It is also the least useful metric for subscription apps.
CPI tells you how much you paid to get someone to download your app. It tells you nothing about whether that person ever opened the app, started a trial, or paid a dollar. Across 75,000+ subscription apps, the median cost per paying user (CPPU) runs **4–5x higher than CPI**. That means for every dollar you think you're spending per user, you're spending four to five dollars to get a paying one. ([RevenueCat SOSA 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/))
The three metrics that actually matter for subscription apps:
[Screenshot 2026-05-07 at 14.37.24.png]
**Trial Start Rate** — what share of installs reach your paywall and start a free trial. Median across all categories: **6.2%**. Health & Fitness apps hit **7.8%** at median and **24.1%** at the top 10%.
**Trial-to-Paid Conversion** — what share of trialists become paying subscribers. Median: **34.8%**. Top-performing apps clear **51.5%**.
**Cost per Subscription (CPS)** — your total ad spend divided by paying subscribers acquired. This is the metric that tells you whether your unit economics work.
[Attribution Window](https://www.airbridge.io/glossary/attribution-window) settings determine how long after an ad interaction you'll credit that ad for a conversion. For subscription apps with longer trial periods, 14–30 day windows are standard. Before launching, confirm all platforms (ad networks, attribution tool, app store) report in the same timezone. Mismatched timezones are the most common cause of daily install count discrepancies.
## Subscription Attribution Benchmarks
Here's what good performance looks like across the install-to-subscriber funnel:
| Metric | Median | Top Quartile (Q3+) | Top 10% (P90) |
| --- | --- | --- | --- |
| Trial Start Rate | 6.2% | 12.4%+ | 20.3% |
| Trial → Paid Conversion | 34.8% | 51.5%+ | 68.2% |
| Download → Paid (D35) | 1.9% | 4.3%+ | 8.5% |
| 1-Year LTV per Payer | $16.27 | $33.91+ | $60.68 |
_Source:_ [_RevenueCat State of Subscription Apps 2025_](https://www.revenuecat.com/state-of-subscription-apps-2025/) _· 75,000+ apps, $10B+ revenue analyzed_
If your trial start rate is below **6.2%**, your paywall is the problem: users aren't reaching it or aren't engaging with it. If your trial-to-paid is below **34.8%**, the issue is usually trial length, onboarding quality, or the wrong audience reaching your trial.
On iOS, [Apple's AdAttributionKit (AAK)](https://developer.apple.com/documentation/adattributionkit) has expanded on [SKAdNetwork (SKAN)](https://www.airbridge.io/en/glossary/skadnetwork-skan), adding support for alternative app marketplaces and re-engagement tracking. For subscription apps, SKAN's postback windows (0–2 days, 3–7 days, 8–35 days) mean iOS install data can be delayed by days or weeks, making early campaign decisions based on iOS data unreliable. [App Tracking Transparency (ATT)](https://www.airbridge.io/glossary/app-tracking-transparency-att) prompts also affect your match rates — lower opt-in rates mean more modeled attribution on iOS.
[Screenshot 2026-05-07 at 14.39.43.png]
## Set Up Simple Attribution in 3 Steps
Most early-stage apps don't need a complex multi-touch attribution model. They need to answer one question: which ad drove a subscriber?
### 1\. Choose your channels - max four
SANs, like Meta, Google, Apple Search Ads, and TikTok, cover the majority of subscription app paid user acquisition spend. Starting with all four creates noise. Most teams do better starting with one or two channels, proving unit economics, then expanding. Tracking more channels than you actively manage creates data you never act on.
### 2\. Track the subscription funnel - not just installs
Configure these events at minimum:
- **Install** (app first open)
- **Trial Start** (paywall presented + trial activated)
- **Subscribe** (first paid transaction)
- **Renewal** (subsequent paid transaction)
- **Cancel / Churn** (subscription ended)
Tracking cancel and renewal, not just the initial subscribe, tells you which ad sources drive subscribers who stick versus those who refund within the first week. You can also add a "How did you hear about us?" field during onboarding. Self-reported data does not replace technical attribution, but it reveals blind spots: if 30% of users say "App Store search" but your attribution tool shows 10% organic, your probabilistic matching may be over-crediting paid channels.
### 3\. Connect ad spend to subscription revenue
Your ad platform reports installs. Your subscription management tool (RevenueCat, Adapty) reports revenue. Attribution is the layer that connects them. When Meta tells you Campaign A drove 200 installs and your subscription tool shows 40 new subscribers, attribution tells you how many of those 40 came from Campaign A specifically, and at what cost per subscriber.
> If you want to connect ad spend to subscription revenue without building a custom data pipeline, consolidated attribution tools like Airbridge let you see cost per subscriber across all four channels in one dashboard.
## Frequently Asked Questions
### Do I need attribution from day one?
If you're running paid ads, yes. The moment you spend money on a campaign, you need to know whether that campaign drove users who convert, not just users who install. Without attribution from the start, you accumulate ad spend with no way to evaluate which campaigns performed. Basic attribution setup (installing an SDK, connecting your ad channels, configuring three events) typically takes a day, not a week.
### What's the difference between installs and paying users?
An install is a download. A paying user is a subscriber. For most subscription apps, only **1.9%** of installs become paying subscribers within 35 days (the D35 median across 75,000+ apps). The top 10% of apps reach **8.5%**. CPI measures how much you paid per install. What you actually need to optimize is cost per paying subscriber — which is **4–5x higher than CPI** on average. These two numbers can look radically different, which is why tracking only installs misleads budget decisions.
### How long before I see meaningful attribution data?
Trial periods commonly range from 7 to 30+ days, and some users take additional days after trial expiry to cancel or convert. Plan for a 30-day observation window before making major budget reallocation decisions. Shorter windows will show install and trial data, but won't give you a full picture of which campaigns drive subscribers who actually stick.
### Should I track what happens after the subscription?
Yes. First renewal is the biggest drop-off point. **55.5%** of monthly subscribers and **66.3%** of annual subscribers churn before their first renewal. An ad source that drives 100 subscribers who all churn in Month 1 is worth less than one that drives 50 subscribers with 80% first-renewal retention. Track at least through the first renewal before judging a campaign's true ROAS:
- **Weekly plans:** check ROAS at D35. (12-month retention median is 1.8%)
- **Monthly plans:** check ROAS at D60. (12-month retention median is 8.9%)
- **Annual plans:** check ROAS at D365. (12-month retention median is 18.5%)
### What attribution window should I use?
Most platforms default to 7-day click, 1-day view. For subscription apps with longer deliberation cycles, 7-day click is standard. Shorter windows (1-day) are stricter and reduce over-attribution but may miss users who click an ad on Monday and install on Wednesday.
## Start Measuring. Stop Guessing.
Running ads without attribution is driving without a dashboard. You feel the speed, but you don't know if you're heading the right direction or running out of fuel.
The setup doesn't have to be complicated. Track the install, the trial start, and the subscription. Connect those three events to your four ad channels. Check your cost per subscriber against the industry median. Adjust from there.
---
## How to Set Up a Simple Paywall Flow Before You Scale
URL: https://www.airbridge.io/en/blog/set-up-a-simple-paywall-flow
Language: English
Published: 2026-05-06
Author: Luke
## How to Set Up a Simple Paywall Flow Before You Scale
You're about to spend your first real money on Meta or Google ads. The creatives are ready. The targeting is set. But before you launch, one question should stop you:
> _If 100 users download your app today, how many will hit your paywall — and how many of those will actually convert?_
Hard paywalls convert at **12.1%** on average. That is more than **5 times** the freemium median of 2.2%, and generates nearly double the first-year LTV: **$49.30 vs. $24.24** ([SOSA 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Among apps that run trials, those offering 17 to 32 days convert at **45.7%** — nearly double the rate of short 3 to 7 day trials. These gaps are not marginal. The paywall decisions you make before your first ad campaign determine the ceiling on every dollar you spend after it.
**Key Takeaways**
- **Paywall timing matters more than paywall design.** When you show the paywall — after users feel value, not at an arbitrary screen — determines conversion more than copy or color scheme.
- **82% of trial starts happen on Day 0.** The window between install and first paywall impression is your highest-intent moment.
- **Hard and soft paywalls solve different problems.** Hard paywalls filter for intent and produce **5.5x** higher conversion than freemium; soft paywalls maximize volume.
- **More than 85% of users skip the first paywall.** A structured second-chance recovery flow is where a meaningful share of revenue gets captured.
- **Establish baseline metrics before scaling ads.** Know your trial start rate, trial-to-paid conversion, and Day 0 conversion before spending.
At the median, subscription apps convert **1.9%** of downloads to paying subscribers within 35 days. Upper-quartile apps hit **4.3%**. Same 1,000 installs. Roughly 2x the subscribers at half the cost per subscriber, without changing a single ad.
The common failure modes are predictable:
- The paywall appears before users feel any value
- There is no second-chance offer for users who skip
- The pricing structure asks for a decision the user is not ready to make
Each of these is fixable before you spend a dollar on acquisition.
---
## Step 1 — Choose the Right Paywall Type
Before designing a single screen, decide what model you are running. The three types perform very differently:
| Paywall Type | D35 Conversion | Y1 LTV | Best For |
| --- | --- | --- | --- |
| Hard Paywall (subscribe or trial before using) | 12.1% | $49.30 | Apps where value is obvious in the first session |
| Soft Paywall (free tier + upgrade prompt) | ~3–5% (estimated) | Not benchmarked | Apps with mixed-intent audiences |
| Pure Freemium (free forever, optional upgrade) | 2.2% | $24.24 | Apps where value needs weeks to show |
### Hard paywall — when to use it
Best for apps that solve a clear, immediate problem. Examples:
- Meditation or sleep apps
- Workout and fitness trackers
- Meal planners and nutrition tools
A hard paywall with a free trial filters out users who were never going to pay. Every conversion signal becomes meaningful: if users are not subscribing, you have a value problem, not a distribution problem.
### Soft paywall — when to use it
Best for apps where users need to experience the product before they understand why they would pay for it. Let users complete one meaningful interaction first, then present the upgrade. At that point, subscribing feels like continuing something, not starting something cold.
### Freemium — the trade-off to know
Freemium requires maintaining two product experiences indefinitely. For most early-stage apps, a trial-based model is simpler to test and faster to optimize. Unless your market clearly demands a free tier, default to a trial-first approach.
> **When a free trial actually hurts LTV**
>
> Trials are not universally better. In **Productivity apps**, users who buy directly are worth **$55.45** vs. **$40.13** for trial users. In **Lifestyle apps**, trial users are **27% less valuable** than direct buyers (Adapty 2026).
>
> A free trial addresses one specific user fear: _"Will this work for me?"_ When users already believe the product works, a trial adds friction and delay without benefit. Consider skipping the trial if your onboarding already demonstrates clear value before the paywall.
---
## Step 2 — Build the Onboarding That Earns the Paywall
**82% of trial starts happen on Day 0.** What happens in the five screens before your paywall determines conversion more than the paywall design itself. Good onboarding does not just explain the product. It builds the emotional case for paying before the user ever sees a price.
[Screenshot 2026-05-06 at 18.48.45.png]
### 1\. Add micro-commitments before the price
Micro-commitments are small interactive steps: picking a goal, answering a quick question, selecting a starting point. They build investment before the user sees a price. The most important rule: surface the user's own answers on the paywall itself.
> **Example — Tiimo:** Tiimo asked one question during onboarding and used the user's answer as the paywall headline. Simple string replacement. No complex personalization engine. The paywall felt personal because it reflected something the user had just said about themselves.
"Your 90-day strength plan is ready" converts better than "Start your free trial" because the first line refers to something the user built, not something being sold to them.
### 2\. Use the loading screen as a trust moment
The 2 to 3 seconds between the last onboarding question and the paywall is high-leverage real estate. Use it to show:
- App store rating and review count
- A personalized message: "We've created your plan based on your answers"
- A subscriber count or a single compelling stat
> **Example — OMENA:** OMENA built a long, scrollable paywall with real user photos, testimonials, a FAQ section, and a founder story. It felt like a landing page, not a gate. That format doubled their trial start rate compared to their previous static paywall.
### 3\. Time the paywall to a value moment
Two common timing mistakes:
- **Too early:** Paywall appears right after sign-up, before the user has done anything
- **Too late:** Paywall is buried so deep that initial motivation has already faded
The right trigger: show the paywall after the user completes one meaningful action. Setting a goal, seeing a personalized result, or finishing a first micro-task. At that point, paying feels like continuing something, not starting something.
---
## Step 3 — Build Your Minimum Viable Paywall
A minimum viable paywall is not a placeholder. It is the leanest version of your best hypothesis about what converts, built to be tested, not finalized.
### What every paywall screen needs
1. **Outcome headline** tied to the user's stated goal from onboarding
2. **3 to 5 benefit points** framed as outcomes, not feature descriptions
3. **Social proof:** a star rating, subscriber count, or one real short testimonial
4. **Plan options:** max 2 plans, with one clearly recommended ("Best Value")
5. **Trial length displayed prominently** so users see the risk is low
6. **One clear CTA button** with no secondary links competing for attention
### Pricing: two plans, one recommendation, one trial length
**Two plans max.** Annual plus weekly, or annual plus monthly. More options create decision paralysis. Three principles from behavioral psychology explain why the defaults you set matter as much as the prices themselves:
- **Default Bias:** The pre-selected plan outsells better-priced alternatives simply because it is selected. Pre-select annual.
- **Anchoring:** The first price the user sees sets their reference point. Show the higher price first so annual feels like a discount.
- **Plan count:** Two plans often outperform three. Sometimes one outperforms two. More options increase abandonment, not conversion.> **Example — Nebula:** Nebula uses a three-option structure with a deliberate decoy: weekly with trial / 3-month at full price (no trial) / annual with trial. The 3-month option exists to make the annual plan look like the obvious value choice. It is not intended to sell.
**On trial length:**
| Trial Length | Median Trial-to-Paid Conversion |
| --- | --- |
| 3–7 days (short) | 26.8% |
| 17–32 days (optimal) | 45.7% |
Annual subscribers retain at roughly **2.5x** the rate of monthly subscribers at 12 months (44.1% vs. 17.5%). That retention gap, compounded over the subscriber base, is the real reason to push toward annual.
One more data point worth knowing before you finalize pricing: high-priced apps earn a median **$55.21** Y1 LTV vs. **$8.09** for low-priced apps — a 6.8x gap. In Health and Fitness specifically, high-priced annual plans produce 4.5x more revenue per user than low-priced ones.
### One objective per paywall
Choose one metric to optimize for:
- **Trial starts:** extend trial length, lower the visible commitment barrier
- **Subscription rate:** tighten onboarding-to-paywall flow, sharpen the value message
- **ARPU:** emphasize annual plan, experiment with price anchoring> **On experimenting:** Teams that run structured paywall experiments earn up to **40x more revenue** than those that don't. The average among experimenting teams: **14.7 experiments**. Experiments that change plan or trial structure outperform visual and copy changes for LTV improvement. Your first paywall is not the destination. It is the starting point.
---
## Step 4 — Capture Users Who Skip the Paywall
More than **85% of users skip the first paywall view**. That is not failure. It is the normal distribution. The question is whether you have a recovery system for those users.
[Screenshot 2026-05-06 at 19.00.57.png]
### 1\. Exit offer — address a different objection, not the same one
When a user taps "No thanks," show a second offer immediately. The rule: the second offer should address a different objection, not just restate the same ask at a lower price.
- First paywall led with annual pricing: second offer introduces the weekly plan as a lower-commitment entry
- First paywall was trial-focused: second offer uses a limited-time promotional rate
- User seems price-sensitive: second offer extends the trial length with the same plan
Some apps offer a lifetime subscription as a second offer for users who resist recurring billing entirely. It can convert a segment that would otherwise never subscribe. Be aware of cannibalization risk if your price is too low relative to annual.
### 2\. Push notification — send within minutes, not hours
A user who just closed your app is still on their phone. Send a push within minutes of exit. Waiting 24 hours loses most of the recovery window. Keep the message specific and low-pressure: "Your fitness plan is ready when you are. Start free, no card needed."
### 3\. Day 1 contextual upgrade — meet them at the feature they want
Users who didn't subscribe on Day 0 but returned on Day 1 are still interested. Do not show them the full paywall again. Show an upgrade prompt at the moment they try to use a locked feature:
- A lock icon on premium content they are trying to access
- A "You've reached your daily limit" message with a one-tap upgrade option
- A contextual banner tied to the specific feature: "Unlock \[feature\] — start your free trial"
> **Example — Mojo:** Mojo placed lock icons not just on the paywall, but throughout the app on individual features and templates. The constant visual reminder of what users couldn't access increased free-to-paid conversion — without changing the paywall itself.
**A note on first renewal:** Nearly **30% of annual subscribers cancel in Month 1**. First renewal survival rates: Weekly 51.8%, Monthly 55.5%, Yearly 66.3% (SOSA 2025). Conversion is not the end of the funnel. The 30-day experience after subscribing determines whether you keep the user long enough to recover your acquisition cost.
---
## Step 5 — Know When You're Ready to Scale
> **Diagnostic: why is your paywall underperforming?**
>
> Before running tests, diagnose the root cause. Three variables drive every paywall conversion:
>
> - **Motivation:** Did the user arrive at the paywall with a reason to subscribe? (Onboarding quality, timing)
>
> - **Trust:** Does the user believe the product will deliver on the promise? (Social proof, testimonials, clarity)
>
> - **Friction:** Is there anything in the paywall flow that makes subscribing feel harder than it should? (Too many plans, confusing copy, no trial)
>
>
> Low motivation is an onboarding problem. Low trust is a paywall design problem. High friction is a UX problem. Each requires a different fix.
You do not need a perfect paywall to start scaling. You need to be past the point of obvious waste. These three numbers tell you where you stand:
| Metric | Below Median — Fix First | Upper Quartile — Ready to Scale |
| --- | --- | --- |
| Trial Start Rate | Below 6.2% | 12.4%+ |
| Trial to Paid Conversion | Below 34.8% | 51.5%+ |
| D35 Download to Paid | Below 1.9% | 4.3%+ |
Two different problems — don't confuse them:
- **Below median on all three:** product problem. Fix onboarding clarity, paywall timing, and offer structure before running A/B tests.
- **At or above median:** testing problem. Iterate on trial length, pricing, and copy to reach the upper quartile.
Once your funnel hits these baselines, the next unlock is knowing which ad channels send users who actually convert through your paywall. The [cost per subscriber](https://www.airbridge.io/en/blog/marketing-dashboard-cost-per-subscriber-by-channel) by channel is what drives efficient scaling, not CPI.
To see which channels send users who actually convert through your paywall, you need to connect ad spend to subscription events. [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) does this for free up to 15K installs — covering trial starts, conversions, and renewals across Meta, Google, and Apple Search Ads.
---
## The Paywall Runs Before the Ads Do
A paywall that converts **4.3%** of downloads at $1,000/month in ad spend becomes a real business at $5,000/month. A paywall at **1.9%** at $5,000/month is just burning money faster.
The sequence that works:
1. Pick the right paywall type for your app's value model
2. Build onboarding that earns the paywall with micro-commitments and personalization
3. Design a minimum viable paywall around one clear outcome and one offer
4. Add a recovery flow for the users who skip the first view
5. Hit your baseline metrics, then scale
---
## Turn Inspiration into Action with Airbridge & Pinterest Ads Integration
URL: https://www.airbridge.io/en/blog/airbridge-pinterest-ads-integration
Language: English
Published: 2026-05-06
Author: Harper (Trang Nguyen)
Pinterest is where people go with intent. Whether they're planning a home makeover, discovering personal style, or curating gift ideas, users are actively searching for inspiration they can act on.
With the Airbridge and Pinterest Ads integration, marketers can now measure ad performance directly in Airbridge, from the first impression to the final conversion, across both mobile app and web environments.
In this post, we'll explain what the integration includes, which events you can track, and how to activate it in your Airbridge dashboard.
## Pinterest Ads: Built for Discovery and Conversion
Pinterest functions as a visual discovery engine. Unlike traditional social platforms, Pinterest users are not just browsing passively. They are actively looking for ideas, solutions, and products that align with their goals.
Pinterest offers a variety of ad formats designed to fit naturally within the platform's search and feed environments:
- **Standard Ads**: Simple image ads that appear in feeds, search results, and related Pins.
- **Video Ads**: Motion-based creatives that play as users scroll, ideal for demos or visual storytelling.
- **Idea Ads**: Multi-page, multimedia formats (formerly Idea Pins) supporting images, video, and text — great for tutorials or step-by-step content.
- **Carousel Ads**: Swipeable image ads that showcase multiple products or features in a single unit.
- **Shopping Ads**: Pull products directly from your catalog, with real-time pricing and availability.
- **Collections Ads**: A hero creative (image or video) paired with product Pins displayed underneath.
- **Lead Ads**: Capture leads on Pinterest with a native, pre-filled form so people can sign up or request info without leaving the experience.
- **Quiz Ads**: Run an interactive quiz that asks a few questions and directs people to tailored products or recommendations.
- **Showcase Ads**: Highlight a curated set of Pins in one ad unit to tell a richer story or feature multiple products together.
- **Premiere Spotlight Ads**: Secure premium, high-impact placement on Pinterest's most visible surfaces to reach intent-driven audiences at scale.
[Pinterest integration.webp]
These formats are built to inspire action. With the Airbridge integration, you can now see which creatives are performing across app and web, directly within your Airbridge dashboard.
## Airbridge & Pinterest Ads Integration Benefits
The integration is built on a complete tracking link and postback setup, enabling marketers to monitor and attribute every conversion event driven by Pinterest Ads.
Here's what you can do with it:
- Run Pinterest campaigns and track performance directly in the Airbridge dashboard.
- Measure top-funnel and mid-to-down funnel actions, including on-site searches, product views, sign-ups or subscriptions, wishlist additions, and completed purchases.
- Create custom events tailored to your app or website for more precise tracking.
- Track user journeys from Pinterest to your website or app.
- Attribute performance to Pinterest across devices and platforms, with support for privacy-safe methods and probabilistic modeling.
You'll be able to identify which Pinterest ads are driving early interest, which are converting, and what is performing best across platforms.
## Connecting Creative with Conversion
Pinterest helps people find what they love. Airbridge helps you track what that love leads to. With this integration, marketers can go beyond surface-level metrics and clearly understand how Pinterest campaigns translate into meaningful growth.
**Ready to get started?**
Visit our [Help Center](https://help.airbridge.io) for setup instructions, or reach out to your Airbridge Team to enable the integration today.
If you have questions about event mapping, campaign performance, or next steps, we're here to help.
---
## How Small App Teams Know Which App Ads Make Money
URL: https://www.airbridge.io/en/blog/how-small-app-ads-make-money
Language: English
Published: 2026-04-30
Author: Jaehyuk Kim
Small app teams know which app ads make money by connecting three data sources. ad platform spend, app store installs, and billing platform revenue. and measuring cost per subscription (CPS) and cohort ROAS by channel.
If you are spending your first $3,000-$5,000 a month on Meta or Google and wondering whether any of it is coming back as subscription revenue, you are asking the right question. Most small teams cannot answer it. because the data they need lives in three systems that do not talk to each other.
**Among 75,000+ subscription apps, the median download-to-paid conversion is just 1.9% within 35 days.** Only 6.2% of downloads even start a trial. The top 5% of subscription apps earn 400x more revenue than the bottom 25%. and that gap is widening. The difference is not bigger budgets. It is knowing which ads produce subscribers and cutting the ones that do not.
If you are new to [app marketing](/blog/what-is-app-marketing) in general, that overview gives useful context before diving into paid channels.
**Key Takeaways**
- **Installs are a vanity metric for subscription apps.** Only 1.9% of downloads become paid subscribers at median. install counts tell you almost nothing about which ads make money.
- **Cohort ROAS is the single most useful measurement for small teams.** Group users by install week and campaign, then track cumulative revenue over 30-90 days.
- **You need three data sources connected.** Ad networks report clicks, app stores report installs, and billing platforms report revenue. none talk to each other by default.
- **Two metrics are enough to start.** CPS and D30 ROAS give a small team enough signal to allocate budget.
- **Benchmarks tell you if you are on track.** Health and Fitness median D60 revenue per install is $0.63. but hard paywall apps earn 8x more than freemium.
## Why Do Installs Tell You Nothing About Which Ads Are Profitable?
Installs are the default metric in every ad dashboard. Meta shows installs. Google shows installs. TikTok shows installs. This creates a dangerous illusion: the channel with the most installs looks like the winner.
But for a subscription app, an install is the beginning of a long funnel. The user has to open the app, hit a paywall or trial prompt, start the trial, survive the trial period, and convert to a paid subscriber. The median trial start rate is just **6.2%**. meaning 93.8% of installs never even begin a trial. Top-performing apps (P90) reach 20.3%, but that is still 4 out of 5 users dropping off.
### 1\. The same installs can produce wildly different revenue
High-tier weekly subscription plans generate **5.2x more revenue per install** than low-tier plans. Two campaigns can drive identical install counts but produce completely different subscription revenue. depending on user quality and what offer those users see.
This is why CPI alone is misleading. The metric that matters is **CPS (cost per subscription)**. total ad spend divided by the number of paid subscribers from that channel. CPS runs **4-5x higher than CPI**. If your CPI is $5.78 for a fitness app on iOS, expect to pay $23-$29 per actual subscriber.
### 2\. Each ad channel attracts fundamentally different users
Meta reaches ~50% of the world's population through social feeds. high volume, but users are browsing, not searching. Apple Search Ads captures users actively looking for an app in the App Store. lower volume, but much higher intent. Google dominates Android and search. strong for intent-based discovery, but iOS campaign optimization is limited to clicks and installs (not purchases).
If you are spending your first $5K on Meta and wonder why trial starts are high but subscriptions are low, the channel might be attracting browsers, not buyers. If you want help choosing, [this guide on picking your first channel](/blog/first-ad-channel-new-app) walks through the tradeoffs.
## What Is Cohort ROAS and Why Does It Change Everything?
Cohort ROAS. return on ad spend measured by user groups over time. is the method that connects ad spend to actual subscription revenue.
### 1\. How it works (with a real example)
Group every user by their install week and the campaign that brought them. Then track cumulative revenue from that group at D7, D30, and D90.
**Worked example:**
| | Meta | Google |
| --- | --- | --- |
| Monthly spend | $3,000 | $2,000 |
| Installs | 650 | 400 |
| Trial starts | 40 | 32 |
| Paid subscribers (D30) | 14 | 11 |
| CPS | $214 | $182 |
| D30 ROAS | 0.047x | 0.055x |
Both channels look unprofitable at D30. But if Google's subscribers retain at 44% through month 12 (yearly plan median), that cohort reaches **0.66x D365 ROAS**. potentially profitable with renewals. Meta's cohort, if retention is lower, might never break even. **You cannot see this difference in an install dashboard.**
### 2\. Why standard ROAS misleads subscription apps
Standard ROAS divides total revenue by total spend over a time period. This mixes users who installed three months ago. now paying subscribers. with users who installed yesterday. It tells you nothing about which campaign drove the paying users.
The subscription ROAS formula should account for **first payment + expected renewals x renewal probability**, divided by spend. A one-time ROAS snapshot misses the bulk of lifetime value because subscription profitability depends on retention over subsequent billing periods.
### 3\. The right measurement cadence
Use **D1-D3** for creative testing. which ad images and copy drive trial starts. Use **D7** for campaign optimization. pause underperformers. Use **D30** for budget allocation. move spend to channels with proven subscriber output.
Your budget should be **at least 10x your target CPA** to give the ad platform enough signal to optimize. A $40 target CPA means a $400 minimum daily budget. anything less and the algorithm cannot learn.
## What Is the Minimum Setup to Know Which Ads Make Money?
If your RevenueCat dashboard says you got 200 trial starts last week but only 12 turned into subscriptions, and you have no idea which campaign those 12 came from. this section is for you.
### 1\. The three-platform data problem
Ad networks know clicks and impressions. App stores know installs. Revenue platforms (RevenueCat, Adapty, or your own billing) know trials and subscriptions. **These three do not connect by default.** That is why you can spend $5,000 and not know if it made money.
The simplest fix: pass a campaign identifier (UTM parameter or custom user attribute) at install time so your billing platform can group revenue by source.
### 2\. Two metrics are enough to start
You do not need 20 dashboards. You need two numbers per channel:
- **CPS**: total ad spend / paid subscribers from that channel
- **D30 ROAS**: cumulative revenue from a cohort at 30 days / ad spend for that cohort
Both are calculable with a billing platform export, ad platform cost export, and a spreadsheet.
[Three-platform data flow for subscription app ad measurement]
## What Benchmarks Should a Small Team Compare Against?
Without benchmarks, you cannot tell whether your numbers are good, bad, or normal.
### 1\. Cost benchmarks by category
| Category | iOS CPI | Android CPI | Notes |
| --- | --- | --- | --- |
| Health and Fitness | $5.78 | $3.06 | Most subscription apps start here |
| Generative AI | $10.58 | $3.60 | Fast-growing, competitive |
| Gaming | $5.99 | $2.56 | High volume, low conversion |
| Business | $14.08 | $11.25 | Highest CPI, highest LTV |
Android paid installs grew **57% YoY** versus iOS at 9%. Android is roughly one-third the iOS cost across categories. a real opportunity for budget-constrained teams.
### 2\. Revenue and conversion benchmarks
| Metric | Median | Top 10% (P90) |
| --- | --- | --- |
| Trial start rate | 6.2% | 20.3% |
| Trial-to-paid | 34.8% | 51.5% |
| D60 RPI (all categories) | $0.31 | $1.95 |
| D60 RPI (Health and Fitness) | $0.63 | $4.19 |
| Hard paywall D60 RPI | $3.09 | . |
Hard paywall apps generate **8x higher D60 RPI** than freemium ($3.09 vs $0.38). High-priced apps produce **$55.21 median Y1 LTV**. 7x more than low-priced apps. If your D60 RPI is below $0.31, fix your [paywall or trial configuration](/blog/freemium-vs-free-trial) before scaling ad spend.
## Why Do Some Ads Look Good in the Dashboard But Lose Money?
If your Meta dashboard shows 200 trial starts but RevenueCat shows 12 subscriptions from last month's campaigns, here is what is likely happening.
### 1\. First-renewal churn erases campaign revenue
More than half of subscribers churn at the first renewal. **51.8% for weekly plans, 55.5% for monthly, and 66.3% for yearly**. After the third renewal, 80%+ of remaining users stabilize. But the damage to cohort ROAS happens in the first billing cycle.
If your ad optimization targets "trial starts," you might be acquiring users who start a trial, cancel on Day 0, and never pay.
### 2\. Every ad platform overcounts conversions
Meta, Google, TikTok, and Apple Search Ads are all [Self-Attributing Networks](https://www.airbridge.io/glossary/self-attributing-network-san). Each claims credit when a user who saw their ad converts. even if the user saw ads on all four platforms. User overlap is massive: only **1% of YouTube users** and **0.2% of Facebook users** are unique to a single platform. The result: your dashboards add up to 3x more conversions than you actually have.
This is why first-party billing data. not ad dashboards. must be your source of truth for revenue.
[Cohort ROAS measurement cadence: D1-D3 creative testing, D7 optimization, D30 budget allocation]
## How Does iOS Privacy Change What You Can Measure?
Post-ATT, iOS lost user-level ad tracking for most users. But measurement is not dead.
SKAdNetwork (SKAN) 4.0 gives you **0-63 conversion values** with a 24-hour+ delay. This is enough to encode trial starts, subscription events, and revenue tiers. if you configure it. The critical step most small teams skip: mapping those 64 values to **revenue events** (trial start, first payment, renewal) instead of just installs.
You cannot say "this specific user came from this specific ad." But you can say "this campaign drove approximately this many subscriptions." For a team running 2-3 campaigns on Meta, Google, Apple Search Ads, and TikTok, that is enough signal to allocate budget. If you need a step-by-step walkthrough, [this SKAN 4.0 setup guide](/blog/skan-4-setup-guide) covers the configuration.
## What Should a Small Team Actually Do This Week?
Here is a Monday morning action plan you can finish in one sitting.
### 1\. Pull two exports (30 minutes)
- **Ad platform cost**: In Meta Ads Manager, go to Campaigns > select last 30 days > export CSV. Do the same in Google Ads. You need one column: **spend by campaign name per week**.
- **Billing platform revenue**: In RevenueCat, go to Charts > select "Initial Conversion" > group by "Install Date" and filter by the custom attribute you pass at install (campaign name or UTM source). Export CSV. If you use Adapty, the equivalent is Analytics > Cohorts > filter by attribution source.
If you are not passing a campaign identifier at install yet, that is step zero. Add a UTM parameter or custom user attribute to your install link so RevenueCat/Adapty can group revenue by source.
### 2\. Build this spreadsheet (30 minutes)
Open a Google Sheet and create one row per channel per install week:
| Install Week | Channel | Spend | Installs | Trial Starts | Paid Subs (D30) | CPS | D30 ROAS |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Apr 7-13 | Meta | $750 | 160 | 10 | 3 | $250 | 0.040x |
| Apr 7-13 | Google | $500 | 95 | 8 | 3 | $167 | 0.060x |
| Apr 14-20 | Meta | $750 | 155 | 11 | 4 | $188 | 0.053x |
| Apr 14-20 | Google | $500 | 100 | 9 | 2 | $250 | 0.040x |
**CPS formula**: `= Spend / Paid Subs`**D30 ROAS formula**: `= (Paid Subs x Monthly Price) / Spend`
### 3\. Read the table and decide (10 minutes)
Compare CPS across channels week by week. In the example above:
- **Google week 1** had the best CPS ($167). worth keeping.
- **Google week 2** jumped to $250. creative fatigue or audience exhaustion. Check if your ad frequency spiked.
- **Meta** is consistent at $188-$250 CPS. stable but expensive. Worth testing new creatives before cutting.
**Rule of thumb**: If a channel's CPS has been above 2x your monthly subscription price for 3+ weeks, reduce spend. If CPS is below 1.5x your monthly price, scale up.
## Connecting the Data Without an Enterprise Stack
The spreadsheet method works for 2-3 channels. But it breaks down when you need real-time trial-to-subscription data by campaign instead of weekly exports.
[Airbridge Core Plan](https://www.airbridge.io/en/plans/core) solves this for small teams. It connects Meta, Google, Apple Search Ads, and TikTok to your billing platform (RevenueCat or Adapty) and shows cohort ROAS, CPS, and trial-to-subscription rates by channel. The plan includes **15K free installs** at **$0.05/install** after that, pay-as-you-go.
Limitations: **4 ad channels**, **2 third-party integrations**, no custom events, no Meta CAPI, no raw data export. For a team spending $5K-$20K/month on 2-3 channels, it replaces the spreadsheet without enterprise complexity. If you need more, you will outgrow it.
## Common Questions About Ad Measurement for Small Teams
### Q1. How long should I wait before deciding if a campaign is profitable?
For apps with **7-day free trials**, wait at least **D30**. Users need time to finish the trial, convert, and make one renewal payment. For annual plans, wait **D60-D90** because the first payment is delayed. Cutting campaigns before D30 risks killing channels that are profitable on a lifetime basis.
### Q2. Should I use the same click-through window across all ad platforms?
Yes. Mismatched windows are the most common cause of data discrepancies. Set all channels to the same click-through window. **7 days** is the standard. and either enable or disable view-through consistently. This does not eliminate discrepancies, but it makes cross-channel comparison valid.
## Your Ads Are Already Telling You the Answer
The data to find your profitable campaigns already exists. sitting in three separate systems, waiting to be connected. You do not need a data team. You need a spreadsheet, two metrics, and 30 minutes a week.
Start with last week's data. Calculate CPS by channel. The answer will be obvious.
---
## How Much Should a Subscription App Spend on UA?
URL: https://www.airbridge.io/en/blog/subscription-app-ua-budget
Language: English
Published: 2026-04-30
Author: Jaehyuk Kim
**There is no universal UA budget percentage for subscription apps.** The right number depends on your lifetime value (LTV), your cost per subscription (CPS), and how fast you need to learn. A useful starting point: back into your budget from a 3:1 LTV-to-CAC ratio, then allocate $500-$1,000 per channel per two-week test window.
The median subscription app converts **1.9% of downloads to paying subscribers** within 35 days. Only 6.2% of downloads even start a trial. The gap between apps that grow and apps that stall maps directly to how teams allocate their first dollars.
The question _how much should a subscription app spend on user acquisition_ has no standard answer. because the right budget is an output of your unit economics, not a benchmark to copy. Most early-stage teams pick a number based on what they can afford, not what the math supports.
They spend $3K/month across three channels, get installs, and have no idea which channel. if any. produces subscribers.
**Key Takeaways**
- **Your UA budget should be derived from LTV, not revenue.** A 3:1 LTV-to-CAC ratio is the baseline for sustainable growth.
- **The minimum useful budget is $500-$1,000 per channel per two-week sprint.** Below that, you are buying noise, not signal.
- **Cost per subscription runs 4-5x higher than cost per install.** Plan for $23-$29 CPS if your median CPI is ~$5.78.
- **Concentrate your first budget on one channel.** Spreading $5K across four channels teaches you nothing.
- **Measure at three cadences: D7, D30, D90.** Each serves a different decision. campaign tweaks, budget allocation, and LTV modeling.
## The Short Answer: Size Your Budget Around LTV, Not Revenue
Percentage-of-revenue budgeting. "spend 20% of revenue on marketing". breaks down for early-stage subscription apps. If your app earns $8K/month in subscription revenue, 20% gives you $1,600. That is not enough to run a statistically meaningful test on even one ad channel.
### 1\. Why percentage-of-revenue rules fail for early-stage apps
Revenue-based rules assume you already have scale. A subscription app doing $8K MRR is still in learning mode. The budget needs to fund experimentation. not mirror what a $2M ARR app does.
### 2\. The LTV-to-CAC formula that replaces arbitrary budget rules
The golden rule: **LTV must exceed customer acquisition cost (CAC) by 3x or more.** If your 12-month LTV per subscriber is $90, your maximum sustainable CPS is $30. Everything flows from that number.
Median annual subscription retention sits at **44%**, with monthly churn running 5-7% after the first renewal. That retention curve defines your LTV ceiling. If you do not know your LTV yet, that is fine. but it means your first UA budget is a learning budget, not a scaling budget.
## What Is the Minimum UA Budget to Generate Real Data?
If you are still figuring out [what paid user acquisition means for your app](/blog/what-is-paid-user-acquisition), the minimum investment question comes first.
### 1\. The minimum spend that produces real signal
A two-week test window on a single channel requires **$500-$1,000** to produce enough installs for meaningful conversion data. At a $5.78 CPI, $1,000 buys roughly 170 installs. At 1.9% download-to-paid, that yields ~3 subscribers. barely enough signal, but enough to validate your CPS range.
### 2\. Why spending below the floor produces noise, not signal
Spend $200 on Meta over two weeks and you get 35 installs. Zero or one subscriber. You cannot tell if the channel works or if you got unlucky. **Below-floor budgets create false negatives**. teams abandon channels that could have worked because they never gave them a fair test.
[UA budget floor: why $500-$1K per channel per sprint is the minimum for meaningful data]
## How to Set Your UA Budget Based on Your App's Unit Economics
If you are new to [app marketing metrics](/blog/app-marketing-metrics), start here.
### 1\. Calculate your current LTV
Pull your D30 or D60 revenue per install (RPI) from your subscription management tool. Hard paywall apps generate **$3.09 D60 RPI** in Health & Fitness. roughly **8x more** than freemium apps at $0.63. Your paywall model directly determines how much you can afford to spend.
If you are pre-launch, use industry medians as placeholders and plan to update monthly.
### 2\. Determine your max sustainable CPS
Divide your projected 12-month LTV by 3. That is your CPS ceiling.
| Scenario | 12-Month LTV | Max CPS (3:1) | Max CPI (at 1.9% conversion) |
| --- | --- | --- | --- |
| Low-tier monthly | $36 | $12 | ~$0.23 |
| Mid-tier annual | $90 | $30 | ~$0.57 |
| High-tier weekly | $187 | $62 | ~$1.18 |
High-tier weekly subscription plans generate **5.2x more revenue per install** than low-tier plans. Your pricing model is a UA budget lever. not just a monetization decision.
Notice that low-tier and mid-tier apps cannot sustain paid UA at median conversion rates. the max CPI is below what any ad channel can deliver at $5.78 market CPI. **This is why improving trial-to-paid conversion or raising your price point comes before scaling ad spend.**
### 3\. Estimate installs needed and set channel budget
Here is the math for a real scenario. A fitness app charging $9.99/month. accounting for ~50% first-renewal churn and 5% monthly churn thereafter. has a 12-month LTV of roughly **$54**. At 3:1, max CPS is $18. With a 1.9% conversion rate and $5.78 CPI, acquiring 100 subscribers requires 5,263 installs at a cost of \*\*$30,400\*\*.
That is the real number. not $3K or $5K. If the budget feels too high, you have two options: improve conversion rate or raise your price point. You do not have the option of pretending the math works at a lower budget.
## How to Split Your UA Budget Across Channels
Once you have a total budget, the next question is allocation. If you are choosing your [first ad channel](/blog/first-ad-channel-new-app), the answer is simpler than you think.
### 1\. For your first $5K: concentrate, do not diversify
Split **60-70%** into one active channel for testing. Reserve **15-20%** for ASO support and **10-15%** for lifecycle and retention. Running four channels at $1,250 each means none gets enough budget to generate a clear signal.
Apple Search Ads delivers the highest-intent users among the four major channels. Meta, Google, TikTok, and ASA. For subscription apps, starting with ASA often produces the cleanest conversion data because users are already searching for what you offer.
### 2\. Channel allocation guide by budget level
| Monthly Budget | Recommended Channels | Split |
| --- | --- | --- |
| $1K-$3K | 1 channel (ASA or Meta) | 100% active testing |
| $3K-$10K | 1 primary + 1 secondary | 70/30 |
| $10K-$30K | 2 primary + ASO + lifecycle | 50/25/15/10 |
| $30K+ | 3 channels + ASO + lifecycle | Per CPS performance |
Do not add a second channel until the first one has either proven profitable or been ruled out with at least two full sprint cycles.
[Channel allocation decision flow by budget level]
## How to Know When Your UA Budget Is Working
Spending is not the hard part. Knowing whether the spend is working. that is where most teams stall. Understanding your [cost per trial and cost per subscription](/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics-fitness-app) per channel is the minimum bar.
### 1\. The three-cadence measurement framework
| Cadence | Decision It Informs | What to Measure |
| --- | --- | --- |
| D7 | Campaign optimization | CPI, trial start rate, creative performance |
| D30 | Budget allocation across channels | CPS, trial-to-paid conversion |
| D90 | LTV modeling and scale decisions | Cohort ROAS, retention curve |
D7 tells you if a creative is working. D30 tells you if a channel is worth your budget. D90 tells you if the business model is viable.
### 2\. The two metrics that tell you if a channel is profitable
**CPS per channel** and **D30 cohort ROAS**. If a channel's CPS exceeds your LTV/3 ceiling, pause it. If D30 cohort ROAS is below 0.3x, the channel likely will not pay back at D90 either.
**Example decision**: Your fitness app has a $54 LTV and $18 max CPS. After 4 weeks, Meta D30 CPS is $15 and Google D30 CPS is $45. Meta is below your ceiling. move 70% of budget there and test scaling. Google is 2.5x over your ceiling. test two new creative sets, and if CPS stays above $30 after a full sprint, cut it. This is the kind of decision D30 data makes possible.
If you are [just beginning to market a subscription app](/blog/how-to-begin-marketing-a-subscription-app), focus on CPS first. ROAS becomes meaningful once you have 60+ days of cohort data. For subscription apps with annual plans, expect breakeven at **D90-D180**. the first payment is delayed, so patience is part of the model.
## Build Your Measurement Stack Before Scaling Spend
The first dollar of UA budget should buy data, not downloads. **Allocate 10-15% of your UA budget to measurement infrastructure** before you scale. Without a measurement loop connecting ad spend to subscription revenue by channel, every dollar above that is a guess.
### 1\. What measurement infrastructure means for a small team
At minimum: a subscription management tool (RevenueCat, Qonversion), an analytics layer that ties installs to subscription events, and a way to see [which ads actually make money](/blog/how-small-app-teams-know-which-app-ads-make-money). You do not need a data warehouse on day one. You need one dashboard that shows CPS by channel.
### 2\. Connecting ad spend to subscription revenue
The gap most teams hit: ad platforms report installs, subscription tools report revenue, and nothing connects the two. You need infrastructure that maps an install from a specific campaign to its downstream subscription event. trial start, conversion, renewal.
If you are at this stage, [Airbridge Core Plan](https://www.airbridge.io/en/plans/core) connects ad spend data to subscription events at no cost for your first 15K installs. It is built for exactly this gap. small teams that need to see CPS by channel before they scale.
## Common UA Budget Questions for Subscription Apps
### Q1. Should I pause all UA during a product overhaul?
Not entirely. Drop to a maintenance budget. **$500-$1,000/month on your best-performing channel**. to keep baseline data flowing. A full pause creates a data gap that makes it harder to distinguish product changes from channel performance changes when you restart.
### Q2. Is it ever too early to spend on paid UA?
Yes. If your onboarding flow is not finalized or your paywall is untested, paid installs will churn at rates that make CPS calculations meaningless. **Get organic or incentivized installs to validate your funnel first**. 200-500 installs is usually enough to establish a baseline conversion rate.
The reason is simple: paid UA amplifies whatever your funnel currently does. If your trial-to-paid conversion is 0.5% because the paywall copy is untested, scaling spend at a $5.78 CPI means you are paying **$1,156 per subscriber**. a number no LTV can support. Fix the funnel with cheap traffic first, then apply paid budget once you know the conversion baseline.
### Q3. How much of my UA budget should go to retargeting?
For subscription apps under $10K/month ad spend, **zero**. Retargeting works at scale. you need a large enough lapsed-user pool to build meaningful segments. Below 50K monthly installs, every retargeting dollar is better spent on prospecting.
The math behind this: retargeting audiences for subscription apps typically consist of trial expirers and lapsed subscribers. At 1,000 monthly installs with a 6.2% trial start rate, your retargetable pool is ~62 users per month. No ad platform can optimize delivery to a 62-person audience. Save retargeting budget until your install volume crosses 50K/month. at that point, your lapsed pool becomes large enough to segment by behavior and bid meaningfully.
### Q4. Should I budget differently for iOS vs Android?
Yes. In Health & Fitness, iOS CPI averages **$5.78** versus **$3.06** on Android. nearly 2x. But iOS subscribers typically have **higher LTV and lower churn**, so the higher CPI often pays for itself. Calculate separate CPS ceilings for each platform. Many early-stage apps start iOS-only because the higher LTV supports the higher acquisition cost.
## Your Budget Is a Hypothesis, Not a Commitment
Every number in this framework is a starting point. The real answer to "how much should I spend" reveals itself after two to three sprint cycles of data. The teams that grow are the ones that treat budget as a testable hypothesis. not a line item they set once and forget.
Start with your LTV. Work backwards to your CPS ceiling. Fund one channel at the minimum floor. Measure at D7, D30, D90. Adjust.
---
## Why Painkiller Apps Convert 9x More: 4 Questions to Diagnose
URL: https://www.airbridge.io/en/blog/painkiller-vs-vitamin-app-pricing
Language: English
Published: 2026-04-29
Author: Jaehyuk Kim
A painkiller app achieved a **37% App Store conversion rate**. A vitamin app in the same category hit 4%. That is a 9x gap — driven entirely by how much users needed the product, not how well it was built. ([Pocketworks, 2024](https://pocketworks.co.uk/blog/vitamins-pain-killers-and-app-store-conversion-rates/))
This distinction — painkiller vs. vitamin — is the strongest predictor of what users will pay for your app. It shapes your app pricing strategy, your paywall model, and whether "too expensive" feedback means you have a pricing problem or a positioning problem.
**A painkiller app solves an urgent, painful problem users will pay to fix immediately. A vitamin app offers a nice-to-have improvement users can live without. This distinction should drive every pricing decision you make.**
**Key Takeaways**
- **Painkiller apps convert at 5-9x the rate of vitamin apps.** The gap comes from motivation, not product quality or design.
- **Willingness to pay is driven by urgency, not usage frequency.** A fertility tracker used once daily has enormous WTP; a weather app used five times daily has almost none.
- **The same app can be a painkiller for one audience and a vitamin for another.** Pricing strategy should target the painkiller segment.
- **Painkiller apps should charge more, not less.** Vitamin apps should give more away, not charge less.
- **"Too expensive" is rarely a pricing problem.** It is usually an activation problem, an audience problem, or a value clarity problem.
## What Is a Painkiller App vs. a Vitamin App?
The framework has three categories:
- **Painkiller**: Solves an urgent, painful problem. Users actively seek a solution and will pay to make the pain stop.
- **Vitamin**: Offers a nice-to-have improvement. Users might enjoy it, but they can live without it.
- **Medicine**: Addresses a regulated or mandated need (compliance apps, tax filing). Users must comply regardless of preference. This article focuses on the painkiller-vitamin distinction — the one most app founders actually face.
The category your app falls into determines four things:
| Dimension | Painkiller | Vitamin |
| --- | --- | --- |
| Motivation | High urgency — users need this now | Aspirational — users want this eventually |
| Activation speed | Immediate action | Delayed, habit-dependent |
| Willingness to pay | High | Low to moderate |
| Retention | Need-driven (sticky) | Habit-dependent (fragile) |
Most apps are not purely one or the other — the same app can be a painkiller for one audience and a vitamin for another. (More on this in the dual-audience section below.)
Understanding painkiller vs vitamin app pricing starts with knowing which category your app falls into — and for which audience. Price for the segment that needs you most.
## How Much Does It Actually Affect Your App Pricing?
The data gaps are not subtle.
**Conversion rates:**
| Metric | Painkiller-type apps | Vitamin-type apps | Gap |
| --- | --- | --- | --- |
| App Store conversion | 37% | 4% | 9x |
| Hard paywall Day-35 conversion | 12.1–21.7% | 2.2–4.5% (freemium) | ~5.5x |
**Lifetime value:**
| Metric | Painkiller-type apps | Vitamin-type apps | Gap |
| --- | --- | --- | --- |
| Hard paywall LTV | $49.30 | $24.24 (freemium) | 2x |
| High-priced apps LTV | $55.21 | $8.00 (low-priced) | 7x |
| Health & Fitness Y1 LTV (median) | $27.21 | — | P90: $86.35 |
Health & Fitness — the category most aligned with painkiller positioning — shows the highest install [lifetime value (LTV)](https://www.airbridge.io/glossary/lifetime-value-ltv) across all app categories. Entertainment — the most vitamin-like category — shows the lowest. The gap between top and bottom categories is 2x. ([Adapty, State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/))
When users need your product, they pay more, stay longer, and forgive more friction. When they do not, no amount of paywall optimization closes the gap.
## How Do You Know Which One Your App Is?
Most founders answer this intuitively and get it wrong. The common mistake is using **usage frequency** as a proxy for willingness to pay. Frequency is a weak predictor. A weather app gets checked five times a day. A fertility tracking app gets checked once. The fertility app has dramatically higher WTP because the stakes are higher.
Use these four questions instead. Score each 1–5.
### 1\. How urgent is the problem?
**1** = "Nice improvement to their day" | **5** = "Urgent daily pain they actively try to solve"
### 2\. What happens if they do not solve it?
**1** = "Nothing meaningful changes" | **5** = "Real consequences — health, financial, emotional"
### 3\. How much emotional relief does solving it provide?
**1** = "Mild satisfaction" | **5** = "Significant relief or transformation"
### 4\. How painful are the alternatives?
**1** = "Easy free alternatives exist" | **5** = "Alternatives are expensive, slow, or frustrating"
**Score interpretation:**
- **16–20**: Painkiller. You have pricing power. Lean into it.
- **10–15**: Vitamin leaning painkiller. You have pricing power with the right audience segment.
- **4–9**: Vitamin. You need a volume strategy or a positioning shift.
The score does not just tell you what to charge. It tells you whether your [value proposition](https://www.airbridge.io/blog/app-value-proposition) can support a paywall at all.
[Painkiller vs Vitamin Diagnostic]
_Use this self-scoring test before setting your price. The right model depends on your product's urgency level, not industry convention._
## What If the Same App Is a Painkiller for Some Users and a Vitamin for Others?
Most apps serve both audiences. A calorie tracking app is a painkiller for someone diagnosed with diabetes — they need it to manage a medical condition. It is a vitamin for someone casually interested in eating healthier. A [fitness app just starting out](https://www.airbridge.io/blog/how-to-begin-marketing-a-subscription-app) can be a painkiller for someone recovering from injury and a vitamin for someone who thinks they should exercise more.
The winning strategy is to **position, market, and price for the painkiller segment** — even if the vitamin segment is larger.
This affects three things:
1. **Ad targeting.** Run ads to the audience with the most urgent problem. "Recover from knee injury at home" converts better than "Get fit this year."
2. **Paywall copy.** Emotional job framing outperforms feature-list framing. "Find calm in 5 minutes" outperformed "Access 500+ meditations" in a real paywall test. Painkiller language focuses on outcome and relief, not feature count.
3. **Price point.** Do not drop to $4.99 to capture the vitamin segment when your painkiller segment pays $9.99. Full-price subscribers retain significantly better than discount cohorts — price acts as a commitment filter.
## How Should Painkiller Apps Price Differently from Vitamin Apps?
Two tracks. Different logic for each.
**Painkiller track:**
- [**Hard paywall**](https://www.airbridge.io/blog/freemium-vs-free-trial) **or short free trial** — high motivation overcomes friction
- **Higher price points** — users with urgent problems tolerate and even expect higher prices
- **Annual plan emphasis** — painkiller users commit because they know they need this long-term
- **Outcome-framed copy** — "Stop waking up at 3am" not "Sleep tracking with AI insights"
**Vitamin track:**
- **Freemium or soft paywall** — low motivation means you need a free tier to build the habit first
- **Lower entry price** — reduce the decision weight since there is no urgency pushing them forward
- **Habit-building free tier** — the free experience must create enough value that upgrading feels natural
- **Feature expansion upsell** — once the habit exists, gate the advanced version
Real examples show these tracks in action:
| App | Category | Pricing | Why it works |
| --- | --- | --- | --- |
| Headspace | Painkiller (anxiety/sleep) | $69.99/year | Users pay for relief from a daily problem |
| Noom | Painkiller (weight loss) | $59/month | High stakes + accountability justifies premium |
| Duolingo | Vitamin (casual learning) | Free + $6.99/month | Massive free tier, habit-first, premium optional |
| MyFitnessPal | Vitamin (casual tracking) | Free + $19.99/year | Broad free tier converts a fraction at low cost |
Here is the counterintuitive part: **when motivation is high, higher prices often convert well. When motivation is low, no price feels right.** This is why painkiller apps can charge premium prices while vitamin apps struggle even at low price points.
When users say "too expensive," it is rarely about the number on the screen. It is one of three things: an **activation problem** (the user has not experienced your core value yet), an **audience problem** (you are targeting the vitamin segment instead of the painkiller segment), or a **value clarity problem** (your paywall does not communicate the outcome clearly enough).
Painkiller users already earmark budget for the problem category. They are not creating new spending — they are replacing an existing solution. Vitamin users must create a new budget line, which is psychologically harder. ([Entrepreneur](https://www.entrepreneur.com/starting-a-business/is-your-product-a-vitamin-or-painkiller/230736))
If you are setting a price for the first time, these ranges are typical starting points:
| App type | Painkiller pricing | Vitamin pricing |
| --- | --- | --- |
| Health & Fitness | $39.99–$79.99/year | Free + $19.99–$29.99/year |
| Productivity | $4.99–$9.99/month | Free + $2.99–$4.99/month |
| AI tools | $9.99–$19.99/month | Free tier + credits |
| Utility | $4.99–$14.99 one-time | Free with ads |
These are starting points, not ceilings. It is easier to lower prices than raise them.
[Pricing Strategy by App Type]
_Painkiller apps optimize for conversion. Vitamin apps optimize for habit formation. The strategy diverges at the model level._
---
## Can You Turn a Vitamin App Into a Painkiller?
You cannot change the product category. But you can change which audience you target and how you frame the problem.
Three levers:
1. **Narrow the audience.** Stop marketing to everyone who "might" use your app. Find the segment where your app solves a real, urgent problem. A generic habit tracker is a vitamin. A habit tracker for people in addiction recovery is a painkiller. Same product, different audience, different pricing power.
2. **Add urgency and stakes.** Streaks, accountability partners, progress tracking with consequences. These do not change the product category, but they raise the emotional cost of not using the app. Duolingo's streak mechanic turns casual language learning into something users feel compelled to maintain.
3. **Reframe the value proposition from features to consequences.** "Track your calories" is a vitamin. "Know exactly what is stalling your weight loss" is a painkiller. The feature is identical. The framing changes the perceived urgency.
**One caution:** if core motivation is low, increasing trust or reducing friction will not compensate. Conversion is a function of **Motivation x (Trust / Friction)** — when motivation is near zero, the other variables cannot make up for it. Fix the audience and the framing before optimizing the funnel.
To track how pricing changes affect your key numbers, start with the [app marketing metrics](https://www.airbridge.io/blog/app-marketing-metrics) that actually matter for subscription apps. For deeper funnel visibility, set up [subscription event tracking](https://www.airbridge.io/blog/subscription-app-event-tracking) from day one.
## Quick Answers to Common Pricing Doubts
### Q1. Is a vitamin app always a bad business?
No, but it requires scale. Vitamin apps need millions of free users to make a 2–3% conversion rate work — Spotify is the textbook example. If your TAM is under 10 million potential users, a vitamin positioning makes profitability very difficult. The alternative is to find the niche audience where the product is actually a painkiller and price for them.
### Q2. Should vitamin apps avoid subscriptions entirely?
No, but lead with a generous free tier that builds the habit first. Once users are engaged, offer an annual plan upsell rather than a monthly gate. Annual plans reduce decision frequency and increase commitment. Earning the right to charge by delivering free value first is the key — not blocking access upfront.
### Q3. How do I test whether my app is a painkiller before launch?
Run a landing page test. Describe the problem your app solves, show a price, and measure "sign up for early access" conversion. If conversion exceeds 5% with a stated price, you likely have painkiller-level motivation. You can also survey your target audience: "If this product disappeared tomorrow, how disappointed would you be?" A rate of 40%+ saying "very disappointed" is a strong painkiller signal.
## Price What You Solve, Not What You Build
The features you ship do not determine what users will pay. The problem you solve does. Every pricing decision — paywall type, price point, trial length, discount strategy — flows from one question: does your user need this, or just want it?
Score your app with the four-question diagnostic. If you are a painkiller, charge what the problem is worth. If you are a vitamin, find the audience segment where you become a painkiller — or build a volume model that does not depend on high willingness to pay.
The founders who get pricing right are not the ones with the best features. They are the ones who understand what they are really selling: relief from a problem, not access to a product.
---
## Subscription or One-Time Purchase? How to Pick the Right Model for Your App
URL: https://www.airbridge.io/en/blog/subscription-vs-one-time-purchase-app
Language: English
Published: 2026-04-29
Author: Jaehyuk Kim
The right app monetization model depends on how your users experience value — not on what the top charts suggest.
Most apps default to subscriptions without asking whether their product actually delivers recurring value. Then they spend months fighting churn without realizing the model itself is the problem.
Choosing the wrong revenue model doesn't just cap your earnings. It shapes your entire product roadmap, your paywall design, and how users perceive your value. Among 75,000+ subscription apps analyzed, **the top 10% capture 94-97% of all subscription revenue**. The gap starts before you build a paywall, before you run your first ad. If you're still figuring out the basics of [how to begin marketing a subscription app](https://www.airbridge.io/blog/how-to-begin-marketing-a-subscription-app), your monetization model is one of the first decisions to lock in.
**Key Takeaways**
- **Not every app should be a subscription.** If your product solves a problem once or gets used sporadically, a one-time purchase may outperform. Match your model to how users experience value, not to industry defaults.
- **Hard paywalls convert at 12.1% vs freemium's 2.2% — but refund rates are 1.7x higher.** Every model has tradeoffs. The numbers should guide your decision, not industry defaults.
- **Yearly plans retain 2.5x better than monthly.** Median 12-month retention is 44.1% for annual vs 17.5% for monthly. Plan structure matters as much as model choice.
- **Pricing experiments raise LTV by 46% on average.** Picking a model is the start, not the finish.
- **Hybrid models are gaining share.** One-time purchases, consumables, and lifetime plans grew from 6.4% to 10.3% of plan type share between 2023 and 2025.
## The Question Most Apps Skip Before Building Their Paywall
There's a reason most monetization advice starts with "just go subscription." Only 4% of mobile apps use subscription models, but they account for 45.4% of total app revenue globally. ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)) The math looks obvious.
But obvious math hides a deeper question: **does your app deliver value continuously, or does it solve a bounded problem?**
A meditation app that users open daily delivers ongoing value. A subscription makes sense. A photo editing app that users open once a week to apply a filter delivers value in bursts. Forcing a subscription on that usage pattern creates friction — users feel they're paying for something they rarely use.
The real diagnostic is not "can I charge monthly?" It's "will users feel they're getting monthly value?" When the answer is no, users churn fast.
This question matters more than paywall design, trial length, or pricing tier. Get the model wrong, and every optimization after it fights an uphill battle.
## What 'Subscription' and 'One-Time Purchase' Actually Mean for Your Revenue
Before choosing, understand what each model optimizes for.
**Subscriptions** create recurring revenue. Users pay weekly, monthly, or annually for continued access. Revenue compounds as your user base grows, and predictable cash flow lets you invest in product improvement. Subscription revenues reached $79.5 billion in 2025, with iOS responsible for 73%. ([Business of Apps, App Revenue Data 2026](https://www.businessofapps.com/data/app-revenues/))
**One-time purchases** generate immediate returns. Users pay once and own the feature or content permanently. Revenue scales linearly with volume — no compounding, but no churn either. Apple supports several models here: non-consumable purchases (permanent unlocks), consumables (credits, tokens), and non-renewing subscriptions. ([Apple Developer: In-App Purchase Types](https://developer.apple.com/help/app-store-connect/reference/in-app-purchases-and-subscriptions/in-app-purchase-types/))
The revenue landscape is more concentrated than most founders expect. **56% of new apps earn less than $1,000 in their first year.** Only 9% break $100K. The market is getting more crowded — 31% more new apps launched in 2025 vs 2024, but median monthly revenue dropped 22% year-over-year from $627 to $492. ([Adapty, State of In-App Subscriptions 2026](https://adapty.io/state-of-in-app-subscriptions/))
Meanwhile, the subscription-or-nothing assumption is weakening. **One-time purchases, consumables, and lifetime subscriptions grew from 6.4% to 10.3% of plan type share between 2023 and 2025.** The market is diversifying.
| Model | Revenue Pattern | Compounding | Best When |
| --- | --- | --- | --- |
| Subscription (monthly/annual) | Recurring, predictable | Yes — grows with retention | Value is continuous, habitual usage |
| One-time purchase | Immediate, one-shot | No — linear with volume | Bounded use case, clear deliverable |
| Consumable (credits/tokens) | Variable, usage-based | Partial — repeat purchases | AI tools, generation-based apps |
| Lifetime subscription | Upfront lump sum | No — but higher per-user | Trust-sensitive users, commitment anxiety |
There's also a platform incentive to consider. **Apple takes a 30% commission on all in-app purchases — but drops to 15% for subscriptions after the first year.** ([Apple Developer: Auto-Renewable Subscriptions](https://developer.apple.com/app-store/subscriptions/)) One-time purchases stay at 30% forever. For apps where users stick around, this 15-point commission difference compounds into a meaningful revenue advantage over time.
## Which Pricing Model Fits Your App? 4 Questions to Decide
Instead of defaulting to what competitors do, answer four questions about your product.
### 1\. How often does the user need this?
Daily or near-daily usage signals subscription fit. If users open your app 4-5 times per week, they're deriving ongoing value. Fitness apps, meditation apps, language learning — these are natural subscription products.
If usage is sporadic or event-driven — a tax calculator, a resume builder, a photo enhancer — subscriptions create friction. Users resent paying in months they don't use the app.
### 2\. How long will they use it?
Subscriptions reward long retention windows. **Yearly plans deliver median 44.1% 12-month retention vs 17.5% for monthly and just 3.4% for weekly.** If your product's natural usage arc is 2-3 months (a workout program, a course), you're fighting churn by design.
One-time purchases sidestep retention entirely. The user pays once, and you don't need to re-earn their commitment every month.
If you're unsure where to start on pricing, here are typical ranges across categories:
| App Type | Common Model | Typical Price Range |
| --- | --- | --- |
| Fitness / Meditation | Annual subscription | $39.99-$79.99/year |
| Productivity / Notes | Monthly or annual | $4.99-$9.99/month |
| Photo / Video editing | One-time or lifetime | $4.99-$29.99 |
| AI tools (GPT wrappers) | Sub + credits hybrid | $9.99/month + $0.01-$0.05/generation |
| Utility (scanner, converter) | One-time purchase | $2.99-$9.99 |
These are starting points, not ceilings. Remember: **it's easier to lower prices than raise them.**
### 3\. Is value continuous or bounded?
This is the most important question. **Continuous value** means the app keeps delivering new outcomes over time — new workouts, new content, new data insights. **Bounded value** means the problem gets solved once — file converted, background removed, document formatted.
Subscriptions work for continuous value. One-time purchases work for bounded value. Mismatching these creates the "subscription fatigue" that a growing share of consumers report experiencing.
### 4\. Does willingness to pay come from urgency or frequency?
**Emotional jobs monetize better than functional jobs.** The strongest predictors of willingness to pay are urgency, consequences of not solving, emotional relief, and pain of alternatives — not frequency of use.
A fertility tracking app used once a day has enormous willingness to pay because the stakes are high. A weather app used five times a day has low willingness to pay because alternatives are free and consequences of switching are minimal.
[Product-Model Fit Diagnostic]
_Use this diagnostic before committing to a model. The right answer depends on your product's value pattern, not industry convention._
## The Numbers Behind Subscription vs. One-Time Purchase Apps
Here's what the data shows about conversion, retention, and lifetime value (LTV) across models. Note: paywall type (hard, soft, [freemium](https://www.airbridge.io/blog/freemium-vs-free-trial)) is a separate decision from revenue model. A hard paywall can gate a subscription or a one-time purchase. The benchmarks below compare paywall approaches, which tend to correlate with model choice.
### 1\. Hard Paywalls Convert at 5.5x the Rate of Freemium
**Hard paywall apps convert at D35 median 12.1% download-to-paid vs freemium's 2.2%** — a 5.5x difference. But hard paywall refund rates (5.8%) run 1.7x higher than freemium (3.4%). Hard paywalls force a fast decision; some of those decisions reverse.
For apps weighing which paywall to pair with their model, see our guide on [hard paywall vs soft paywall](https://www.airbridge.io/blog/hard-paywall-vs-soft-paywall).
### 2\. The 7x LTV Gap Between High-Priced and Low-Priced Apps
| Segment | Y1 LTV per Payer (Median) | Y1 LTV per Payer (P90) |
| --- | --- | --- |
| Health & Fitness overall | $27.21 | $86.35 |
| Hard paywall | $49.30 | — |
| Freemium | $24.24 | — |
| High-priced apps | $55.21 | — |
| Low-priced apps | $8.00 | — |
**Hard paywall delivers $49.30 median Y1 LTV vs freemium's $24.24** — roughly double. High-priced apps earn 7x more than low-priced apps in Y1 LTV. This isn't about charging more for the sake of it. Apps that price higher tend to deliver (and communicate) more value.
### 3\. Why Yearly Plans Retain 2.5x Better Than Monthly
| Plan Duration | 12-Month Retention (Median) | Share of Apps Using |
| --- | --- | --- |
| Yearly | 44.1% | 67% (H&F) |
| Monthly | 17.5% | — |
| Weekly | 3.4% | 55.6% (global) |
Yearly plans retain **2.5x better than monthly** after 12 months. In Health & Fitness, 67% of apps offer yearly plans. Globally, weekly plans dominate at 55.6% — but their retention is extremely low. Weekly works for fast revenue but creates a revolving door.
### 4\. The experimentation multiplier
**Pricing experiments raise LTV by 46%** even though they only improve conversion rate 28% of the time. **Hard paywall delivers 21% higher LTV than soft paywall.** And growth teams that run experiments earn up to **40x more revenue** than those that don't.
The takeaway: your first model choice matters, but your willingness to iterate matters more. For tracking how pricing changes affect your [app marketing metrics](https://www.airbridge.io/blog/app-marketing-metrics), measure conversion and LTV from day one.
### 5\. How Model Choice Affects Your Ad Budget Payback
If you're planning to run paid ads, model choice directly shapes how fast you recover your acquisition cost.
**Subscription example:** You spend $5 CPI on Meta. Your app charges $9.99/month. A subscriber who stays 4 months generates $40 in revenue (minus Apple's 30% cut in year 1 = $28 net). You break even around month 2-3. But if they churn after the free trial, you recover $0.
**One-time purchase example:** Same $5 CPI. Your app charges $4.99 one-time. After Apple's 30% cut, you net $3.49 per paying user — **which doesn't cover the acquisition cost** unless your install-to-purchase conversion rate is high enough to offset unpaid installs.
| | Subscription ($9.99/mo) | One-Time Purchase ($4.99) |
| --- | --- | --- |
| CPI | $5.00 | $5.00 |
| Net revenue per payer | $28.00 (4-month retention) | $3.49 |
| Breakeven | Month 2-3 | Needs 60%+ conversion rate |
| Risk | Churn before breakeven | Low per-user margin |
This is why subscription apps can afford higher CPIs — they recover cost over time. One-time purchase apps need either very high conversion rates or very low acquisition costs to make paid UA work.
---
## When One-Time Purchase Is the Right Answer
Subscriptions get the spotlight, but one-time purchases solve real problems that subscriptions create.
### 1\. When commitment anxiety is your biggest conversion blocker
A significant share of consumers report subscription fatigue. Some users will pay $30 once but will never agree to $5/month. A lifetime purchase or one-time unlock captures revenue that subscriptions leave on the table.
**Lifetime purchases reduce friction by removing commitment anxiety.** The tradeoff is capped upside — you can't compound revenue from a single-payment user. But for early-stage apps that need cash flow now, capturing $30 today beats hoping for $5/month over six months.
### 2\. When your app solves a bounded problem
Photo editors, document converters, audio enhancers, single-purpose utilities — these deliver value in a single interaction. A subscription model asks users to pay perpetually for something they use intermittently.
Apple's guidelines reinforce this: if you change your existing app to a subscription-based model, you cannot remove functionality that existing purchasers already paid for. ([Apple App Store Review Guidelines](https://developer.apple.com/app-store/review/guidelines/)) If your app's core value is a one-time transformation, a one-time price is honest and sustainable.
[Model Comparison Spectrum]
_Revenue models exist on a spectrum. Pure subscription and pure one-time purchase are the extremes. Most successful apps land somewhere in between._
## Using Both Models Without Confusing Users
You don't have to choose one model forever. The most effective monetization strategies layer models thoughtfully.
### 1\. Lifetime as a second offer after decline
Some users hesitate at recurring pricing but are comfortable with a one-time payment. **Offering a lifetime purchase as a second offer — shown only after a user declines the subscription — captures value that would otherwise be lost.** This works without cannibalization when positioned correctly: the lifetime option appears only at the moment of rejection, not alongside the subscription.
### 2\. Subscription plus consumables (the hybrid model)
AI apps pioneered this: a monthly subscription covers base access, and consumable credits handle heavy usage. Users get predictability (flat monthly cost for normal use) plus flexibility (buy more when they need it).
This model works when usage is variable but ongoing. If some sessions require 10x the compute of others, a flat subscription either overcharges light users or undercharges heavy ones. Consumables solve this.
### 3\. The rule of clarity
Whatever combination you use, users should understand their options in under 5 seconds. If your paywall requires a paragraph of explanation, you've added too much complexity. Start with one model, learn what works, then layer. For how to think about this in the context of [growing your subscription app](https://www.airbridge.io/blog/how-to-grow-a-subscription-app), simplicity in early monetization preserves your ability to iterate.
## Common Mistakes When Choosing Your Monetization Model
### 1\. Choosing freemium without the scale to support it
**Freemium requires massive volume.** At a 2.2% conversion rate, you need roughly 4,500 downloads to generate 100 paying users. Hard paywall at 12.1% needs only 825 downloads for the same 100 conversions.
For early-stage apps with limited ad spend, freemium is a slow bleed. Free users also give misleading feedback — their expectations differ from paying users. The features they request may not align with what drives revenue.
### 2\. Interpreting "too expensive" literally
**"Too expensive" usually means unclear value, poor activation, or wrong audience — not wrong pricing.** When users say your app costs too much, the real problem is usually that they haven't experienced enough value to justify the price.
Start slightly higher than you think is right. **Lowering prices is easier than raising them**, especially on the App Store where a price increase above 50% triggers additional review friction. Lowering a price is a promotion. Raising it is a penalty.
### 3\. Offering free trials when they attract the wrong users
Trials aren't universally better. **In Productivity apps, direct purchasers have an LTV of $55.45 vs trial users at $40.13.** In Lifestyle, trial users are 27% less valuable than direct purchasers.
Trials help when value is experiential and trust is a blocker. They hurt when they attract low-intent users who start trials with no intention of paying. If your conversion data shows high trial starts but low trial-to-paid rates, the trial itself may be the problem. For deeper analysis, make sure you're measuring trial quality, not just trial quantity — our guide on [subscription app event tracking](https://www.airbridge.io/blog/subscription-app-event-tracking) covers how.
## Your First Model Is a Hypothesis, Not a Life Sentence
Nobody gets monetization perfect on day one. The founders who win are the ones who pick a model that fits their product's value pattern, ship it, and start learning from real paying users.
**Your first 50-100 paying users will tell you more about your pricing than any benchmark report.** If they churn fast, the model may be wrong. If they ask for a lifetime option, commitment anxiety is real. If they never convert from free, you may need a paywall.
Start with one model. Measure what happens. Then adjust. If you're ready to move from model choice to your [first ad channel](https://www.airbridge.io/blog/first-ad-channel-new-app), that's the next decision.
---
## How to Choose Your First Ad Channel: App Marketing Guide
URL: https://www.airbridge.io/en/blog/how-to-choose-your-first-ad-channel-app-marketing-guide
Language: English
Published: 2026-04-27
Author: Jaehyuk Kim
You have budget set aside for your first app marketing campaign. You open Meta Ads Manager, then Google Ads, then TikTok — and freeze. Each platform promises results, each has different creative requirements, and each has a different minimum spend. **The problem is not which channel is "best" — it is which channel fits your situation right now.**
Every channel can work for subscription apps. The question is which one matches your goals, your existing demand, your budget, and the creative assets you already have. This guide gives you a framework to answer that question.
**Key Takeaways**
- **Your first channel shapes your entire paid strategy.** It determines your creative direction, feedback loop speed, and how fast you learn what converts.
- **Match the channel to your situation, not popular opinion.** If search demand exists for your app category, start with intent channels. If not, start with discovery.
- **Creative readiness narrows your options fast.** No video? That rules out TikTok and limits Meta. Strong App Store screenshots? ASA is ready today.
- **Each channel has a different learning phase and budget floor.** Knowing these upfront prevents wasted spend on a channel you cannot afford to test properly.
- **Measure trials and subscriptions, not just installs.** [Cost per trial](https://www.airbridge.io/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics-fitness-app) tells you more about channel quality than cost per install ever will.
## Four Ad Channels for App Marketing — and Why They Are Not Interchangeable
The four major channels for subscription app marketing are not interchangeable. Each one reaches a different type of user at a different moment in their decision process.
| Channel | How It Reaches Users | Creative Format | Learning Phase | Min Monthly Budget |
| --- | --- | --- | --- | --- |
| Meta Ads | Shows ads to users who match a profile — they are not looking for you | Video (15-30s), image carousels | 7+ days, needs 50+ conversion events per week | $1,000+ |
| Google App Campaigns | Surfaces your app when users search related keywords | Text headlines, images, optional video | 7+ days minimum, budget must be 50x target CPI daily | $1,500+ |
| Apple Search Ads | Places your app at the top of App Store search results | Your existing App Store screenshots | 3-5 days | $500+ |
| TikTok Ads | Puts short-form video in users' feeds — content-native format | Short-form video (UGC style) | 7+ days | $1,000+ |
A few things stand out from this table. [About 65% of App Store downloads happen directly after a search](https://developer.apple.com/app-store/search/), and [ASA ads convert at over 60% in the top position](https://ads.apple.com/app-store) — making it the highest-intent channel. TikTok delivers [lower CPM and CPC than Meta but typically higher cost per acquisition](https://www.trendtrack.io/blog-post/tiktok-vs-meta-cpm) — useful for creative testing, less reliable for direct conversion. CPI ranges vary widely: Meta [$2.00-$5.50](https://www.businessofapps.com/ads/cpi/research/cost-per-install/), Google [$1.50-$4.50](https://www.businessofapps.com/ads/cpi/research/cost-per-install/), ASA [~$4.06 median US](https://www.apptweak.com/en/aso-blog/apple-ads-benchmarks), TikTok [$1.00-$4.50](https://www.businessofapps.com/ads/cpi/research/cost-per-install/).
But CPI alone is misleading. A channel with a high CPI can still be your cheapest source of paying subscribers if its trial start rate is strong. Always evaluate channels on cost per trial and cost per subscription.
## Four Questions to Pick Your First Ad Channel
You do not need a spreadsheet model. Answer four questions and the right starting channel becomes obvious.
[diagram-channel-decision-tree.svg]
### 1\. Does Search Demand Already Exist for Your App Category?
Open the App Store and Google Play. Search the keywords your app targets. If competitors show up with established ratings and reviews, **search demand exists** — users are already looking for what you offer.
**Search demand exists** → Start with Apple Search Ads or Google App Campaigns. You are capturing intent that already exists. This is the fastest path to learning whether your app converts searchers into subscribers.
**No search demand** → Start with Meta or TikTok. You need to create awareness first. These channels show your app to people who match a profile but are not actively looking.
### 2\. How Much Can You Spend on Paid Ads?
Each channel has a budget floor below which the algorithm cannot learn effectively.
With less than $1,500 per month, **Apple Search Ads** is your only viable option — it has the lowest minimum budget and the highest-intent users. With $1,500-$3,000, **Meta** becomes testable if you target tightly. Above $3,000, **Google App Campaigns** open up, since Google needs higher daily spend to exit its learning phase.
### 3\. What Ad Creative Can You Launch With Tomorrow?
This question eliminates options faster than any other.
| You Have | Best Channel |
| --- | --- |
| Strong App Store screenshots only | ASA |
| Text headlines + a few images | Google App Campaigns |
| 15-30 second video ads or carousels | Meta |
| UGC-style short-form video | TikTok |
| Nothing yet | ASA (uses existing store listing) |
If you have no video assets, TikTok is off the table. If your App Store page is weak, ASA will underperform. Your creative readiness often makes the decision for you.
### 4\. Are You Chasing Installs — or Subscribers?
If you are optimizing for installs alone, reconsider. CPI is a [top-of-funnel](https://www.airbridge.io/glossary/funnel) vanity metric that does not reflect business health. The metrics that matter: **cost per trial**, trial-to-paid conversion rate, and cost per subscription. Set your first test goal as: "Find a channel where I can acquire trials at an acceptable cost."
## How to Test Your First Ad Channel Without Wasting Budget
Once you have picked a channel, structure the test so you get a clear answer — not noise.
**Duration**: 7-14 days. Shorter than 7 days does not account for platform learning phases or weekday vs. weekend variation. Google specifically needs [at least 7 days](https://support.google.com/google-ads/answer/13020501) before performance stabilizes.
**Budget**: Enough for the platform to learn. For ASA, $500-$1,000 is sufficient. For Meta, plan for $1,500+. For Google, set daily budget to at least 50x your target CPI.
**What to watch**: After 5-7 days, check three things. Is CPI stabilizing? Are trial starts coming in? If installs are high but trials are low, the issue is your app store page or onboarding — not the channel. Do not scale before day 7 or kill before day 5.
**What to track**: Do not stop at installs. Track trial starts, subscription activations, and [ad spend](https://www.airbridge.io/glossary/ad-spend) per channel. Without post-install data, you cannot compare channels on what actually matters.
## FAQ
### Q1. How long should I test before switching channels?
Minimum **7 days**, ideally **14 days**. This accounts for platform learning phases, weekday vs. weekend variation, and enough post-install events to measure trial and subscription rates.
### Q2. Can I run ads on multiple channels at the same time?
Yes, if you can afford the minimum budget for each channel simultaneously. Below that threshold, you collect noise instead of signal. Most early-stage apps are better off testing one channel at a time.
## Pick One, Learn Fast, Then Decide
The best first ad channel for a new app is the one that matches your current situation — your demand landscape, budget, and creative assets. It is not permanent. Once you have data from one channel, use it to decide whether to scale or test the next.
The harder question comes after: which campaigns drive trials, which trials become paying subscribers, and what does each subscriber actually cost per channel?
**If you are ready to connect your ad spend to subscription events, start free with Airbridge Core Plan — up to 15K installs included, no setup fee.**
---
## What Should You Track Before Spending More on App Marketing
URL: https://www.airbridge.io/en/blog/what-to-track-before-scaling-app-ads-marketing
Language: English
Published: 2026-04-27
Author: Jaehyuk Kim
You launched ads. Installs are coming in. But when someone asks "is it working?" you open a dashboard, stare at a line going up, and realize **you have no idea what "working" actually means**. If you have ever wondered what metrics to track for app ads, you are not alone. Install count tells you people downloaded your app. It does not tell you whether those people will ever pay you a dollar.
Most growth teams start scaling ad spend based on installs and CPI alone. That is like judging a restaurant by how many people walk through the door, ignoring whether anyone orders food. **The gap between downloading and paying is where your budget either compounds or evaporates.** In most subscription apps, fewer than 3% of installs ever become paying subscribers. If you cannot see where in the funnel users drop off, you cannot fix the leak.
If you cannot see those numbers broken down by channel, by campaign, by week, you are spending blind.
**Key Takeaways**
- **Seven metrics in funnel order give you full visibility.** CPI, install-to-trial rate, cost per trial, trial-to-paid rate, cost per subscription, LTV, and ROAS. Track them together, not in isolation.
- **Installs are an exposure metric, not a success metric.** A campaign with 1,000 installs and zero trials is worse than one with 200 installs and 40 trials.
- **No single metric tells the truth alone.** CPI without post-install conversion data is noise. Always compare cost per subscription against lifetime value.
- **Scaling before you have enough conversion data is the most expensive mistake.** You need at least 50 conversion events per campaign before making spend decisions.
## Why Install Count Alone Will Mislead You
### 1\. What Installs Actually Measure
An install tells you one thing: someone tapped "Get" or "Install." It confirms that your ad creative caught attention and your App Store listing did not scare them off. That is it.
Installs do not tell you whether the user opened the app, started a trial, hit your paywall, or subscribed. They are an [upper-funnel metric](https://www.airbridge.io/blog/what-is-app-marketing) that measures distribution, not monetization.
Growth teams optimize for CPI because it is the first number they see. But a $2 CPI campaign that generates zero trials costs you infinitely more per subscriber than a $6 CPI campaign where 15% start a trial.
### 2\. The Real Gap Between Downloading and Paying
Here is what the funnel actually looks like for a typical subscription app:
[Subscription app conversion funnel from install to renewal with benchmark rates at each stage]
The median install-to-trial rate across 75,000+ subscription apps is **6.2%**. Health & fitness apps perform better at **7.8%**, but that still means over 92% of installs never start a trial.
Of those who do start a trial, **34.8%** convert to paid (39.9% in health & fitness). And **82% of trials start on Day 0**. If a user does not start your trial the day they install, they probably never will.
This is why [tracking in-app events](https://www.airbridge.io/blog/subscription-app-event-tracking) matters more than counting installs. The install is the beginning of the story, not the end.
## The 7 App Marketing Metrics That Actually Matter
These seven metrics follow your user from ad click to recurring revenue. Each one answers a different question about your [paid UA](https://www.airbridge.io/blog/what-is-paid-user-acquisition) performance.
### 1\. Cost per Install (CPI)
**What it is**: Total ad spend divided by total installs.
**Formula**: Ad Spend / Installs
**Benchmark**: iOS health & fitness averages **$5.78**. Business apps run higher at $14.08. iOS CPI runs roughly 3x Android across categories.
**What it tells you**: How efficiently you are generating downloads. But CPI alone is a vanity metric. A low CPI with poor post-install conversion is a money pit.
### 2\. Install-to-Trial Rate
**What it is**: The percentage of users who install and then start a free trial.
**Formula**: Trial Starts / Installs
**Benchmark**: Median **6.2%** across all subscription apps. Health & fitness: **7.8%**. The global average across 16,000+ apps is 10.9%, so beating that figure puts you in strong territory.
**What it tells you**: Whether your onboarding and paywall are working. If this number is below 5%, fix your app experience before spending more on ads.
### 3\. Cost per Trial (CPT)
**What it is**: How much you spend to get one user to start a trial.
**Formula**: Ad Spend / Trial Starts (or CPI / Install-to-Trial Rate)
**Benchmark**: A healthy blended CPT for subscription apps is [**$20-$40**](https://www.linkedin.com/pulse/what-good-cost-per-trial-subscription-apps-user-shamanth-m-rao). If you are paying $5.78 CPI with a 7.8% trial rate, your CPT is $74. That is expensive.
**What it tells you**: The real cost of acquiring an engaged user. CPT is the [first meaningful cost metric](https://www.airbridge.io/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics-fitness-app) in your funnel because it filters out dead installs.
### 4\. Trial-to-Paid Rate
**What it is**: The percentage of trial users who convert to paid subscribers.
**Formula**: Paid Conversions / Trial Starts
**Benchmark**: Median **34.8%** across subscription apps. Health & fitness: **39.9%**. Education: **42%**. Longer trial periods (17-32 days) convert at **45.7%** versus 26.8% for short trials of 3-7 days.
**What it tells you**: Whether your trial experience delivers enough value to justify the price. If this is below 30%, your trial length, onboarding, or pricing needs work. Consider reading about [freemium vs. free trial models](https://www.airbridge.io/blog/freemium-vs-free-trial) if you are still deciding on your monetization approach.
### 5\. Cost per Subscription (CPS)
**What it is**: How much you spend to acquire one paying subscriber.
**Formula**: Ad Spend / Paid Subscribers (or CPT / Trial-to-Paid Rate)
**Benchmark**: CPS typically runs **4-5x your CPI**. If your CPI is $5.78, expect CPS in the $23-$29 range under good conditions. Many apps see CPS above $50.
**What it tells you**: The actual cost of acquiring revenue. This is the number you compare against LTV to know if your ads are profitable.
### 6\. Lifetime Value (LTV)
**What it is**: Total revenue a subscriber generates over their lifetime.
**Formula**: ARPU x Average Subscriber Lifespan (or Revenue / Total Subscribers for a simpler version)
**Benchmark**: Median Year 1 [LTV per payer is $16.27](https://www.airbridge.io/glossary/lifetime-value-ltv) across subscription apps (P90: $60.68). Health & fitness: **$27.21** Year 1. High-priced apps generate 3-7x the LTV of low-priced ones, depending on category and pricing tier.
**What it tells you**: The ceiling on what you can afford to spend acquiring a subscriber. If your LTV is $27 and your CPS is $35, you are losing money on every conversion.
### 7\. Return on Ad Spend (ROAS)
**What it is**: Revenue generated per dollar of ad spend.
**Formula**: Revenue from Acquired Users / Ad Spend
**Benchmark**: [Cross-industry medians](https://segwise.ai/blog/roas-benchmarks-industry-standards) show Google Ads averaging around **3.31x**, Meta around **2.19x**, TikTok around **1.41x**. These are blended averages across industries, not subscription-app-specific.
**What it tells you**: Whether your ad spend is generating positive returns. ROAS is the single summary metric that combines everything above. A ROAS below 1.0 means you are losing money.
Here is how all seven metrics compare at a glance:
| Metric | Formula | Benchmark (Subscription Apps) | What to Watch For |
| --- | --- | --- | --- |
| CPI | Spend / Installs | $5.78 iOS H&F | Low CPI with low trial rate = wasted spend |
| Install-to-Trial | Trials / Installs | 6.2% median, 7.8% H&F | Below 5% = fix onboarding before scaling |
| CPT | Spend / Trials | $20-$40 blended | Above $50 = check CPI and trial rate |
| Trial-to-Paid | Paid / Trials | 34.8% median, 39.9% H&F | Below 30% = trial or pricing problem |
| CPS | Spend / Paid Subs | 4-5x CPI | Must be less than LTV |
| LTV | Revenue / Subscribers | $16.27 median Y1 | Compare against CPS by channel |
| ROAS | Revenue / Spend | 2-3x healthy | Below 1.0 = unprofitable |
## When One Metric Looks Good but the Math Does Not Add Up
Individual metrics without context lead to bad decisions. Here is how to connect them.
### 1\. The Golden Rule: LTV Must Exceed CPS
This sounds obvious but most teams violate it without knowing. If your median Year 1 LTV is $27.21 (the health & fitness benchmark) and your blended CPS is $30, you are underwater. And that assumes a full year of retention.
The math gets worse with short payback windows. If you are on a monthly plan and your first-renewal retention is 55.5%, nearly half your subscribers churn before Month 2. **Your effective LTV on Day 30 is far lower than your annualized projection.**
The target: **CPS should be no more than 30-50% of projected Year 1 LTV.** This aligns with the [widely cited 3:1 LTV-to-CAC benchmark](https://www.wallstreetprep.com/knowledge/ltv-cac-ratio/) and gives you margin for churn, seasonality, and measurement error.
### 2\. Why You Should Compare Channels by CPS, Not CPI
A $3 CPI on Meta looks cheaper than a $6 CPI on Apple Search Ads. But if Meta's install-to-trial rate is 4% and ASA's is 12%, the math flips:
- **Meta**: $3 CPI / 4% trial rate = **$75 CPT**
- **ASA**: $6 CPI / 12% trial rate = **$50 CPT**
Now apply trial-to-paid rates. If Meta converts trials at 30% and ASA at 40%:
- **Meta**: $75 / 30% = **$250 CPS**
- **ASA**: $50 / 40% = **$125 CPS**
The "expensive" channel delivers subscribers at half the cost. This is why CPI comparisons across channels are unreliable without funnel data.
### 3\. Use Cohorts to See What Averages Hide
Averages hide the truth. A D30 ROAS of 1.5x looks good until you realize one cohort from a seasonal spike is pulling it up.
Break your data by:
- **Time cohort**: Weekly install groups. Did the March 15 cohort convert better than March 22?
- **Channel cohort**: Same week, different channels. Which one has a higher trial-to-paid rate?
- **Segment cohort**: iOS vs. Android, country, campaign type.
Revenue per install at D60 is **$0.31 median** across subscription apps. If your D60 revenue per install is above that, your funnel is working. If it is below, find out which cohort is dragging the number down.
## 5 Metrics Mistakes That Burn Through Ad Budget
### 1\. Chasing Cheap Installs That Never Convert
This is the most common trap. You find a campaign with a $1.50 CPI and pour budget into it. Installs flood in. Trials do not. Your CPT is $75 and your CPS is $200+.
The fix: set a **CPT ceiling**, not a CPI target. If your LTV is $27, your CPS needs to stay below $14 to hit a 2x LTV-to-CPS ratio. Work backward from there.
### 2\. Treating All Channels as if They Do the Same Job
ASA captures high-intent users already searching for your category. Meta introduces your app to people who were not looking for it. TikTok reaches users in a discovery mindset. Each [channel plays a different role](https://www.airbridge.io/blog/first-ad-channel-new-app) in your funnel, and expecting the same CPI or conversion rate from all of them leads to wrong conclusions about which ones work.
### 3\. Scaling a Campaign With Too Few Conversion Events
Ad platforms need conversion data to optimize. The minimum threshold is roughly **50 conversion events per week** per campaign. If you are optimizing for "trial started" and only getting 8 trials per week, the algorithm is guessing.
**Budget should be at least 10x your target CPA.** If your target CPT is $30, spend at least $300/day on that campaign before evaluating performance. Scaling before this threshold means you are reading noise, not signal.
### 4\. Looking at Averages Instead of Segments
Your blended trial-to-paid rate might be 35%. But when you split by OS, iOS converts at 42% and Android at 18%. When you split by creative, video ads convert at 38% and static at 25%.
Averages kill profitable segments. Break every metric by channel, OS, creative type, and campaign before making budget decisions.
### 5\. Ignoring the Payback Period
A $50 CPS with a $100 LTV looks profitable. But if your average subscriber takes 9 months to pay back that $50, and you are spending $20K/month on ads, you need $180K in working capital before you break even. High LTV means nothing if cash runs out before the payback period closes.
Track monthly revenue per cohort. Know exactly which month each cohort turns profitable.
## How to Wire Up Tracking for All 7 Metrics
Knowing what to track is half the problem. The other half is actually collecting the data.
### 1\. Events You Need to Fire From Your App
Your app needs to send events at each funnel stage. At minimum:
- **app\_open**: confirms the install led to an active session
- **trial\_started**: user initiated a free trial
- **trial\_converted**: trial user became a paid subscriber
- **subscription\_renewed**: paid user renewed (weekly, monthly, or annually)
- **subscription\_cancelled**: user cancelled before next renewal
Without these events, you cannot calculate anything beyond CPI. Most analytics platforms and ad networks accept server-side event streams. Map your [subscription events to your measurement stack](https://www.airbridge.io/blog/subscription-app-event-tracking) early, before you scale spend.
### 2\. Connecting Ad Spend Data to In-App Events
This is where most setups break. You have ad spend data in Meta, Google, and ASA dashboards. You have in-app event data in your analytics tool. Connecting the two requires a measurement layer that links the ad click to downstream events.
If you are [starting your first campaigns](https://www.airbridge.io/blog/how-to-begin-marketing-a-subscription-app) with a small budget, you can do this manually by pulling reports and matching dates. But once you run multiple channels simultaneously, you need an automated connection between ad spend and subscription events.
Airbridge Core Plan connects your ad channels to in-app subscription events in a single dashboard, with 15,000 monthly installs included free. If your current setup cannot show you cost per trial by channel in one screen, it is worth [starting free](https://www.airbridge.io/en/plans/core) before you increase spend.
### 3\. Building Your First Metrics Dashboard
Your dashboard needs three views:
**Daily view**: CPI, CPT, and install volume by channel. This catches sudden changes.
**Weekly view**: Trial-to-paid rate, CPS, and ROAS by channel cohort. This is your decision-making view.
**Monthly view**: LTV progression by cohort, payback period, and blended ROAS. This tells you whether your unit economics are improving.
Start with a spreadsheet if needed. The format does not matter. What matters is looking at these numbers weekly and making budget decisions based on CPS and ROAS, not CPI and install volume.
[Metrics framework showing how the 7 metrics connect in funnel order from ad spend to ROAS]
## FAQ
### What is a good ROAS for a subscription app?
It depends on your payback window. For subscription apps with monthly billing, a **D30 ROAS of 0.5-0.8x** is common because most revenue comes in over months, not days. A **D90 ROAS above 1.0x** is the minimum target. Annual plans front-load revenue, so D30 ROAS can hit 1.5-2.0x. Cross-industry averages ([Google 3.31x, Meta 2.19x](https://segwise.ai/blog/roas-benchmarks-industry-standards)) include e-commerce and lead gen, which monetize faster than subscriptions. Do not benchmark against those numbers.
### How long should I wait before judging a campaign?
At minimum, **14 days after the campaign exits its learning phase**. For most platforms, that means 21+ days total. For subscription apps specifically, add the length of your trial period on top of that because trial-to-paid conversions happen days or weeks after install. A 7-day trial app needs at least 28 days of data. A 14-day trial app needs 35+. Judge based on CPS and ROAS at the end of that window, not CPI on Day 3.
### Do I need different metrics for iOS and Android?
Yes, and the differences are significant. iOS CPI runs roughly **3x higher** than Android. But iOS users convert to paid at higher rates and have higher LTV. Your CPS on iOS might actually be lower than Android despite the higher CPI. 12-month retention also differs: yearly subscribers retain at **44.1%** versus monthly at **17.5%**. Always split your metrics by OS. A blended view hides these differences and leads to bad allocation decisions.
## Start With These 7 Numbers Before You Spend Another Dollar
You do not need 50 metrics. You need seven. And you need to look at them in funnel order, by channel, every week.
The goal is simple: know your CPS by channel and compare it against your LTV. Everything else is a supporting detail. If you can answer "which channel delivers paying subscribers at the lowest cost, and is that cost less than what those subscribers are worth?" you have enough data to make smart scaling decisions.
If you cannot answer that question today, do not increase your ad budget. Set up your event tracking, collect two weeks of conversion data, and then decide where to put your next dollar.
---
## How to Write an App Value Proposition People Understand in 5 Seconds
URL: https://www.airbridge.io/en/blog/how-to-write-app-value-proposition
Language: English
Published: 2026-04-27
Author: Jaehyuk Kim
You built the app. You know what it does. But when someone lands on your App Store listing, **most visitors leave without installing**. Page-view-to-install conversion rates average just 33.7% on iOS and 26.4% on Google Play ([Business of Apps](https://www.businessofapps.com/marketplace/app-store-optimization/research/app-store-optimization-statistics/), based on AppTweak 2020 data). The problem is rarely the product. It is the words you use to describe it.
An **app value proposition** is a single, clear statement that names who your app is for, what problem it solves, and why your approach is different. Written so a stranger can grasp it in seconds. Get it wrong and you lose the install before a user ever sees your onboarding, your paywall, or your product.
56% of new apps earn less than $1,000 in their first year. Only 9% break $100K. The common assumption is that these apps have a product problem or a pricing problem. But most apps approach conversion as a pricing problem when it is actually a belief problem. If users do not understand why your app matters to them, they will never get far enough to see your price.
**Key Takeaways**
- **Your value proposition decides whether someone installs.** It names who your app is for and why it matters to them, before they ever see a feature.
- **Start with one audience and one problem.** A value proposition for everyone speaks to no one. Pick the segment that feels the pain most acutely.
- **Use your users' words, not your product language.** Pull exact phrases from App Store reviews, Reddit threads, and support tickets.
- **Apply the 4-component formula.** Specific audience + job/problem + how you address it + your unique approach. Fill in the blanks and you have a working draft.
- **Test with the 5-second rule.** If someone outside your team cannot repeat your value proposition after reading it for 5 seconds, rewrite it.
## Why Most App Descriptions Fail (And What They Cost You)
Most app descriptions read like a product spec sheet. They list features, mention technical capabilities, and use language that makes sense internally but means nothing to someone scrolling the App Store at 11 PM.
The cost is not abstract. It is lost installs, wasted ad spend, and a [paywall nobody reaches](https://www.airbridge.io/blog/freemium-vs-free-trial).
### 1\. The 7-Second Decision Window
Users spend an average of seven seconds on an App Store listing page before deciding whether to explore further, according to [research by Phiture and AppAgent](https://phiture.com/asostack/why-7-seconds-could-make-or-break-your-mobile-app-f41000fb2a17/) based on a sample of over 25,000 visitors. That means two out of three people who find you walk away.
Your value proposition is what fills those seven seconds. If it is unclear, no amount of screenshots or review stars will compensate.
### 2\. Feature Lists Are Not Value Propositions
Consider two descriptions for the same meal planning app:
| Approach | Description |
| --- | --- |
| Feature list | "AI-Powered Meal Planner. 10,000 recipes. Smart grocery list." |
| Value proposition | "For busy parents who want to feed their family well but don't have time to plan. We turn 'What's for dinner?' into a 30-second decision." |
The first tells you what the app has. The second tells you who it is for and what changes in your life. The difference is the gap between a spec sheet and a reason to care.
A feature list answers "What does this app do?" A value proposition answers "Why should I care?" Users in the App Store are asking the second question.
### 3\. The Belief Gap: Why Unclear Messaging Kills Conversion
Product-market fit has three components: the right product, the right market, and the right messaging. Sometimes an app does a great job solving a problem but struggles to explain it. Positioning and messaging are just as important as the solution itself.
This belief gap is why you see apps with strong retention among existing users but poor install conversion rates. The product works. The explanation of the product does not.
When OMENA, a wellness app, replaced a standard paywall with one that explained the app's value before asking for payment, their trial start rate doubled. The product did not change. The communication did.
## Identify One Audience and One Problem
Before you write a single word of your value proposition, you need two inputs: who this is for and what pain you solve. Skip this step and everything downstream falls apart.
### 1\. Why "Everyone" Is the Wrong Audience
When you target everyone, everything breaks. Messaging becomes generic. Users cannot identify themselves in your description. Feedback from different user types contradicts. Pricing does not map to perceived value.
The test is simple: ask five current users how they would explain your app to someone else. If they struggle or each gives a wildly different answer, your positioning is not specific enough.
A fitness app that says "Get fit" speaks to no one in particular. A fitness app that says "For runners training for their first half-marathon" speaks to a person who immediately thinks, "That is me."
### 2\. Define Users by Context and Pain, Not Demographics
Demographics tell you who people are. Context tells you when they need you.
The Jobs to Be Done framework captures this well: "When I \[situation\], I want to \[motivation\] so I can \[outcome\]." Users do not buy features. They hire apps to make progress in a specific situation.
A budgeting app might define its audience as "millennials aged 25-35." That is a demographic. A sharper definition: "People who just got their first salaried job and have no system for managing money." That is a context. The second version writes your value proposition for you.
Emotional jobs drive retention and willingness to pay far more than functional jobs. A meditation app solves the functional job of "guided breathing exercises." The emotional job is "I want to stop feeling like my brain never shuts off."
### 3\. The Painkiller Test: Is Your Problem Urgent Enough?
Not all problems are equal. Painkiller apps that solve urgent, felt pain monetize faster. They drive quicker activation, higher conversion, and stronger retention.
Here is the critical insight: **the same app can be a painkiller for one audience and a vitamin for another.** A sleep tracking app is a vitamin for someone who is "curious about sleep." It is a painkiller for a new parent running on four hours a night who desperately needs to optimize the little sleep they get.
Your value proposition should target the audience for whom you are a painkiller. That is where urgency lives, and urgency drives installs. If you are still figuring out [who your first audience should be](https://www.airbridge.io/blog/how-to-begin-marketing-a-subscription-app), start with the group that feels the pain most acutely.
[Diagram showing the same app positioned as a vitamin for a casual audience vs. a painkiller for an urgent audience, with conversion rate indicators]
## Translate the Problem Into User Language
You have your audience and their pain. Now you need to describe it in words they actually use, not words your product team uses.
### 1\. Where to Find Your Users' Exact Words
The best value proposition copy is stolen, not written. Go to these sources and pull exact phrases:
- **App Store and Google Play reviews** (yours and competitors'). Look for how users describe the problem your app solves, not how they praise features.
- **Reddit threads and forums** where your target audience discusses their frustrations.
- **Support tickets and chat logs.** Users describe problems in their rawest form when asking for help.
- **Interview transcripts.** If you have done user interviews, search for emotional language.
A wellness app found that users experiencing brain fog described it as: "It feels like I have cotton wool in my brain." They used that exact phrase in an ad. It worked because it felt real, not manufactured.
### 2\. Product Language vs. User Language (With Examples)
| Product Language | User Language |
| --- | --- |
| "AI-powered personalized workout algorithm" | "Workouts that actually fit my schedule" |
| "Comprehensive financial tracking dashboard" | "Finally know where my money goes" |
| "Advanced sleep analysis with biometric integration" | "Figure out why I wake up tired" |
| "Intelligent recipe recommendation engine" | "Tell me what to cook with what I have" |
Product language describes what you built. User language describes what changes in someone's life. Your value proposition uses user language.
### 3\. The Friend Test: Would a Real Person Say This?
Read your value proposition out loud. Would a user say this to a friend? If the answer is no, rewrite it.
"Our app leverages machine learning to optimize your fitness journey" fails the friend test. Nobody says that at dinner. "It builds workouts around your actual schedule, even if you only have 20 minutes" passes. That is something one friend tells another.
This is not about dumbing things down. It is about matching the language your users already think in.
## Write Your Value Proposition Using a Proven Formula
You have your audience, their pain in their own words, and a clear sense of what your app does about it. Now assemble these into a single statement. Here is how to write an app value proposition that a stranger can repeat back to you.
### 1\. The 4-Component Formula
A positioning statement has four components:
1. **Specific audience**: who is this for?
2. **Job or problem**: what are they trying to do or solve?
3. **How your app addresses it**: what does your app do about this problem?
4. **Your unique approach**: what makes your way of solving it different?
The template:
> **For \[specific audience\] who \[job/problem\], \[your app\] helps by \[how you address the pain\] through \[unique approach\].**
Here is how a financial app might fill this in:
> For first-time salary earners who have no system for managing their money, \[App Name\] helps by turning every paycheck into a clear spending plan through automatic categorization and weekly check-ins that take under 2 minutes.
Every word earns its place. The audience is specific (first-time salary earners, not "millennials"). The problem is concrete (no system for managing money). The approach is distinct (automatic + 2-minute check-ins).
### 2\. Two Alternative Frameworks (Moore and Blank)
If the 4-component formula does not click, two other proven templates work well ([rapptrlabs.com](https://rapptrlabs.com/blog/create-value-proposition-app/)). The key difference: Moore's formula adds a competitive frame ("Unlike X"), while Blank's strips everything down to a single sentence.
**Geoffrey Moore's formula** (from _Crossing the Chasm_):
> For \[target user\] who \[need/opportunity\], our app is a \[category\] that \[primary benefit\]. Unlike \[alternative\], we \[key differentiator\].
**Steve Blank's XYZ formula:**
> We help \[target user\] do \[job\] by doing \[primary benefit\].
Real examples of these in practice:
- **Bumble:** "For people who want meaningful connections, Bumble is a dating app where women make the first move." (Moore's framework applied; [rapptrlabs.com](https://rapptrlabs.com/blog/create-value-proposition-app/))
- **Pocket:** "We help busy readers save articles for later by letting them bookmark anything in one tap."
Pick the formula that feels most natural. The important thing is that all three force you to name your audience and their problem before describing your solution.
### 3\. Before-and-After Examples for Subscription Apps
| App Type | Before (Generic) | After (Value Proposition) |
| --- | --- | --- |
| Fitness | "The complete fitness app" | "For people who quit gyms but still want to get strong. Home workouts that adapt to your level, no equipment needed" |
| Meditation | "Mindfulness made easy" | "For overthinkers who can't fall asleep. Guided sessions that quiet your mind in under 10 minutes" |
| Finance | "Smart budgeting tool" | "For freelancers with irregular income. A budget that flexes when your paycheck does" |
| Language | "Learn a new language" | "For travelers who need conversational basics in 30 days. Skip grammar drills, learn the 200 phrases that matter" |
Notice the pattern. Each "after" version names a specific person, states their situation, and explains what changes. The "before" versions could describe any app in the category.
Once you have your value proposition written, the next step is [turning it into a growth plan](https://www.airbridge.io/blog/grow-subscription-app), making sure your messaging reaches the right people through the right channels.
[Diagram showing the 4-component formula as connected blocks: Audience → Problem → Solution → Unique Approach, with an example filled in below each block]
## Test and Validate Your Value Proposition
Writing a value proposition is not a one-time exercise. It is a hypothesis that needs testing.
### 1\. The 5-Second Test
Show your value proposition to five people who are not on your team. Give them five seconds to read it. Then ask: "What does this app do, and who is it for?"
If they cannot answer both questions clearly, your value proposition is not working. This is not a focus group. It is a basic readability check. If a stranger cannot parse your statement in five seconds, a distracted App Store visitor certainly will not.
### 2\. Where Your Value Proposition Lives (Store Listing, Onboarding, Paywall, Ads)
A value proposition is not a one-time artifact. It needs to appear consistently across every touchpoint where users decide whether to continue:
- **App Store subtitle and first screenshot.** The first thing users see. Your value proposition goes here in its shortest form.
- **Onboarding screens.** Reinforce why the user made the right choice to install.
- **Paywall.** The moment of maximum friction. Repeat your value proposition here, not a feature list. Users who see why the app matters to them convert at higher rates than users who see what the app contains.
- **Ad creative.** Your ads are a test of your value proposition at scale. If click-through rates are low, the value proposition is not resonating with that audience.
Consistency matters. If your App Store listing promises one thing and your onboarding says something different, users feel a disconnect and drop off.
### 3\. Iterate as You Learn
Your value proposition will evolve. As you get more users, you learn which audience responds most strongly and which language resonates. Run A/B tests on your App Store subtitle. Change your paywall headline. Test different ad copy angles.
The goal is not to write the perfect value proposition on day one. The goal is to write a clear one, test it, and sharpen it based on real behavior. Every new cohort of users teaches you something about how they describe their own problem.
## 5 Value Proposition Mistakes That Cost You Downloads
These are the five most common anti-patterns. If your value proposition matches any of these, it needs a rewrite.
| Mistake | What It Looks Like | Why It Fails |
| --- | --- | --- |
| Too broad | "The app for everyone who wants to be healthier" | When you speak to everyone, no one identifies with it |
| Feature-obsessed | "500+ exercises, AI tracking, social features" | Lists what you built, not what it does for the user |
| Competitor-focused | "Better than [Competitor X]" | Positions you in their frame, not yours |
| Jargon-heavy | "ML-driven adaptive periodization engine" | Users do not speak this language |
| Aspirationally vague | "Transform your life" | No specifics. Transform how? For whom? By when? |
The underlying problem in all five cases is the same: the value proposition is about the app instead of about the user. Flip the lens. Start with who the user is and what changes in their life, then work backward to what your app does.
## FAQ
### Q1. How is a value proposition different from a tagline?
A tagline is a brand asset, a short, memorable phrase like Nike's, memorable phrase like Nike's "Just Do It." It is designed for recall, not explanation. A value proposition is functional. It explains who you are for, what problem you solve, and how. Your tagline might live on your marketing site. Your value proposition drives your App Store listing, onboarding copy, and ad creative strategy. You need both, but the value proposition comes first because it informs the tagline.
### Q2. Should I write different value propositions for different audiences?
Yes, if you serve meaningfully different user segments. A fitness app might target beginners and experienced lifters. Each group has a different pain point and a different reason to subscribe. Write a core value proposition first, then adapt it per segment. Your App Store listing gets the version that matches your largest or highest-converting audience. Ads targeting specific segments get tailored versions. The underlying formula stays the same. Only the audience and pain change.
### Q3. Where should I put my value proposition first?
Your App Store subtitle and first screenshot. This is where the highest volume of decisions happen. According to [Apple](https://ads.apple.com/app-store), **65% of App Store downloads come directly after a search.** The subtitle (30 characters on iOS) is your value proposition in its most compressed form. If you can only optimize one thing today, start there. Then extend to your paywall headline and onboarding.
## Your App Has 5 Seconds. Make Them Count.
Your app might be great. But if someone lands on your listing and cannot understand why it matters to them in five seconds, they will never find out. They will scroll past, install a competitor, or simply close the App Store.
The fix is not better features. It is clearer words. Pick one audience. Name their problem. Describe what changes. Test it on real people. Rewrite until a stranger can explain your app back to you in one sentence.
---
## How to Choose Between Free Trial and Freemium With Data
URL: https://www.airbridge.io/en/blog/how-to-choose-free-trial-vs-freemium
Language: English
Published: 2026-04-27
Author: Jaehyuk Kim
**56% of new subscription apps earn less than $1,000 in their first year.** Only 9% break $100,000. The gap between the two groups starts with one of the earliest decisions you make: how you structure your first offer.
The wrong monetization model doesn't just cost conversions. It changes how fast you learn what works, how quickly revenue comes in, and whether your onboarding actually drives upgrades.
This guide walks you through the decision, the setup, and the mistakes that kill early subscription revenue. If you're new to [app marketing](/blog/what-is-app-marketing), your subscription offer is one of the first things to get right.
---
**Key Takeaways**
- **Free trials convert 5-10x higher than freemium.** The median trial-to-paid rate is 34.8% across 75,000+ apps, while freemium sits at 2-5%. For most new apps, trials are the faster path to revenue.
- **Trial length matters less than you think.** Trials from 5 to 32 days show roughly equal conversion (44-46% median). Pick based on time-to-value, not competitor defaults.
- **Freemium works when your market is massive and value compounds over time.** If you need scale before monetization makes sense, freemium is the play.
- **Your first offer should be simple.** One model, one or two plans. Learn before you optimize.
- **82% of trial starts happen on Day 0.** Your first-session onboarding determines almost everything about conversion.
## Understand What Each Model Actually Does
Before you pick a model, understand what each one optimizes for. Free trials and freemium look similar from the outside (both start with "free") but they create different conversion dynamics. (For a quick definitions overview, see our [freemium vs free trial FAQ](/blog/freemium-vs-free-trial).)
### 1\. How free trials create urgency
A free trial gives users full access for a fixed window. When time runs out, they pay or lose access.
This creates a built-in decision point. The user knows the clock is ticking, which compresses the evaluation period and forces a yes-or-no choice.
Among 75,000+ subscription apps analyzed, **the median trial-to-paid conversion rate is 34.8%**. Health & Fitness apps hit 39.9%. Travel apps reach 48.7%.
The tradeoff: urgency can also create anxiety. Users worried about forgetting to cancel may avoid starting a trial entirely. That's why trial start rate (not just trial-to-paid rate) is a critical metric.
### 2\. How freemium creates a long tail
Freemium gives users a limited version of your app permanently. No countdown, no deadline. Certain features, content, or capacity are locked behind a paid plan.
The economics are a volume play. **Freemium free-to-paid conversion is approximately 2-5% for most products**, with a median around 2.1%. You need a massive top of funnel to generate meaningful revenue. Categories like social, entertainment, and utility apps tend to see the strongest freemium performance because their user bases are large enough to make low conversion rates viable.
But freemium captures users that trials miss. About **23% of freemium conversions happen six or more weeks after download**. These are users who would never start a timed trial but gradually hit a limit, see the value, and upgrade on their own schedule. The typical upgrade trigger is a capacity constraint: running out of storage, hitting a usage cap, or needing a feature that only appears after sustained engagement.
### 3\. The conversion gap in numbers
| Metric | Free Trial | Freemium |
| --- | --- | --- |
| Median conversion to paid | 34.8% (trial-to-paid) | 2-5% (free-to-paid) |
| Time to revenue signal | Days to weeks | Weeks to months |
| Late conversions (6+ weeks) | Low | 23% of total conversions |
| Best-fit scenario | Defined vertical, fast value delivery | Massive TAM, value compounds over weeks |
The numbers favor trials for speed and efficiency. Freemium wins on patience and scale. Your choice depends on which game you're playing.
## Decide Which Model Fits Your App
### 1\. Answer three questions before you choose
Before touching App Store Connect or Google Play Console, answer these:
**How fast can a new user experience core value?** If your app delivers a meaningful outcome in a single session (a completed workout, a generated meal plan, a language lesson), a free trial works. If value builds over weeks of accumulated data or habits, freemium may fit better.
**How large is your addressable market?** Freemium economics require massive volume. If you're targeting a specific vertical with a defined audience, free trials are more efficient.
**What's the user's primary fear at the paywall?** If it's "Will this work for me?" a trial removes that fear. If it's "I don't want to commit yet" freemium lets them stay without pressure.
### 2\. When free trials win
Free trials are the better starting point for most new subscription apps:
- **Faster revenue signal.** You learn within days whether users convert, not months.
- **Higher conversion efficiency.** Hard paywall apps convert at 12.1% download-to-paid versus freemium at 2.2%.
- **Clearer onboarding feedback.** Trial start rate and trial-to-paid rate tell you exactly where the funnel breaks.
- **Better for defined verticals.** Fitness, health, education, and productivity apps with clear outcomes see strong trial performance.
Data from 16,000+ apps shows that **in Health & Fitness and Education, free trials boost lifetime value (LTV)**. Trial experience alone boosts first renewal by up to 60%.
### 3\. When freemium makes sense
Freemium is not the wrong choice; it's the wrong _default_ choice for most new apps. It makes sense when:
- **Value compounds over time.** Collaboration tools, habit trackers, and data-accumulation apps need time to prove their worth.
- **Your market is massive.** Consumer apps targeting millions of potential users can sustain low conversion rates.
- **Virality is a growth lever.** Free users who invite others create a distribution engine that paid-only models can't match.
- **Your product competes on distribution, not conversion speed.** Think Spotify or Duolingo.
Tettra, a B2B knowledge management tool, **tripled its upgrade rate** and sustained 70%+ monthly retention after switching from a 15-day trial to freemium ([source](https://productled.com/blog/freemium-business-model-example)). This is a B2B example, but the principle applies to mobile apps too: if your product's value builds over weeks of accumulated usage, putting a clock on it can be counterproductive.
[diagram-decision-flowchart.svg]
## Set Up Your Trial or Freemium Offer Step by Step
### 1\. Pick your plan structure
Start simple. **One model, one or two plans.** You can always add complexity later. Launching with three tiers, a free plan, and a trial simultaneously creates noise that makes it impossible to learn what's working.
For your first offer:
- **If trial:** Weekly + trial is the highest-performing combination. Trial-inclusive paywalls convert at 64.5% versus visual or text-only paywalls at 44.4%.
- **If freemium:** Gate one clear feature set. Don't scatter limitations across the app. Users should understand instantly what they get free and what requires payment.
Use a **mid-tier decoy plan** to make your target plan look more attractive. If you're offering monthly and annual, add a quarterly option priced to make annual the obvious choice. Across subscription apps studied, **90%+ of users don't convert on the first paywall view**. A well-structured plan comparison gives them a reason to reconsider.
### 2\. Choose your trial length
The data here is counterintuitive. **Trials from 5 to 32 days show roughly equal median conversion at 44-46%.** The difference is not length but whether users experience core value before the trial ends.
Practical guidance:
- **3-7 days:** Best for apps with immediate value (fitness workouts, meal plans, meditation sessions). Shorter trials also mean faster revenue.
- **14 days:** The safe middle ground. Works for most apps.
- **30 days:** Only if your app requires weeks of data accumulation or habit building before the value is clear.
The trend is toward shorter. **46.5% of trials are now 4 days or less.** Adding a free trial to a weekly subscription raised 12-month LTV from $7.40 to $54.50, showing that short trial + weekly plan is a high-performing combination.
Don't copy your competitor's trial length. Match it to your time-to-value.
### 3\. Decide opt-in vs opt-out
This single decision changes your conversion rate more than almost any other variable.
- **Opt-out trials** (credit card required upfront): **48.8% trial-to-paid conversion**
- **Opt-in trials** (no credit card required): **18.2% trial-to-paid conversion**
Opt-out converts higher because users who enter payment details are more committed. But opt-out also carries a higher refund rate. Hard paywall apps see 5.8% refunds versus 3.4% for freemium.
**For most new apps, start with opt-out.** The conversion advantage outweighs the slightly higher refund rate. You can always test opt-in later if trial start rates are too low.
Both Apple and Google have specific implementation requirements for introductory offers. Review [Apple's subscription offer documentation](https://developer.apple.com/documentation/storekit/in-app_purchase/original_api_for_in-app_purchase/subscriptions_and_offers/implementing_introductory_offers_in_your_app) and [Google Play's subscription setup](https://developer.android.com/google/play/billing/subscriptions) before building.
### 4\. Design your first paywall
Your paywall is where the decision happens.
**Onboarding + trial setup achieves 1.78% install-to-paid conversion**, the highest of any paywall configuration measured across 16,000+ apps.
Paywall design principles:
- **Show the trial prominently.** "Start your 7-day free trial" should be the primary call to action, not buried below plan options.
- **Display price per day or per week**, not just the annual total. $0.27/day feels different from $99/year.
- **Include social proof or outcome data** if you have it. "Join 50,000 users" or "Average user completes 4 workouts per week."
- **Offer a second chance.** Most users skip the first paywall. Trigger a second one after the user experiences a key feature, with a different price or format.
For a deeper comparison of paywall strategies, see our [hard vs soft paywall guide](/blog/hard-vs-soft-paywalls).
## Avoid the Three Mistakes That Kill Early Subscription Revenue
### 1\. Copying competitor trial length without testing
Your competitor's 7-day trial was probably not tested either. They copied someone else.
Trial length should match your app's time-to-value, not your category's default. If users get value in one session, a 3-day trial may outperform a 14-day trial. **Plan and trial duration changes are the most effective experiments for LTV improvement**, more effective than price changes alone.
### 2\. Offering too many plan options at launch
Three tiers, a free plan, and a trial is a recipe for confusion and slow learning. You can't tell what's working when five variables change at once.
**Start with one model and one or two plans.** Learn what converts. Then add complexity.
### 3\. Ignoring Day 0 onboarding
**82% of trial starts happen on Day 0.** If your onboarding doesn't surface the trial within the first session, most users will never see it.
Top-performing apps (P90) achieve a 20.3% trial start rate versus a 6.2% median. The gap is almost entirely explained by first-session onboarding quality. If your trial start rate is low, the problem is not your offer; it's that users never reach the offer.
Build your onboarding to do three things: demonstrate value, build intent, and present the trial before the session ends. For a complete onboarding framework, see [how to begin marketing a subscription app](/blog/how-to-begin-marketing-a-subscription-app).
## Test, Learn, and Iterate
### 1\. What to measure in your first 30 days
Track these four metrics from day one:
| Metric | What it tells you | Benchmark |
| --- | --- | --- |
| Trial start rate | Is your onboarding working? | 6.2% median, 20.3% P90 |
| Trial-to-paid conversion | Is your offer compelling? | H&F 39.9%, Travel 48.7%, All 34.8% |
| Day 0 conversion rate | Is first-session urgency working? | 44.5% of all purchases happen Day 0 |
| Refund rate | Are you attracting the right users? | 3.4% (freemium) to 5.8% (hard paywall) |
If trial start rate is low, fix onboarding. If trial-to-paid is low, reconsider trial length or offer. If Day 0 conversion is low, your paywall timing or design needs work.
### 2\. When to switch models
Switching from trial to freemium or vice versa is a significant change. Do it based on data, not instinct.
**Consider switching to freemium if:** Users consistently engage after the trial ends but don't convert. Your retention post-trial-expiry is high but paid conversion is low. Your market is larger than you initially targeted.
**Consider switching to trial if:** Freemium users engage but never upgrade. Your free-to-paid conversion is below 2%. You need faster revenue signal to fund growth.
### 3\. How experimentation compounds revenue
Most founders treat their first offer as permanent. It shouldn't be.
**Price experiments show a 28% conversion improvement probability but a 46% LTV improvement probability.** Small, systematic changes to plan structure, trial length, and pricing compound over time.
Run one experiment at a time. Measure for at least two full trial cycles before calling a winner. Document what you learn. The apps that break $100K in their first year are not the ones that guessed right on day one; they're the ones that iterated fastest.
## Start With One Offer and Get It Right
For most new subscription apps, start with a free trial. The conversion data is stronger, the revenue signal comes faster, and the feedback loop is cleaner.
If your app relies on habit formation or has a massive addressable market, freemium may be the better bet, but test that assumption with data, not instinct. Pick one model, launch, and let your users tell you what to change.
---
## Paid User Acquisition, Explained: What It Is, What It Costs, and When to Start
URL: https://www.airbridge.io/en/blog/what-is-paid-user-acquisition-costs-when-to-start
Language: English
Published: 2026-04-21
Author: Jaehyuk Kim
**Paid user acquisition (paid UA)** is the practice of paying ad platforms — Meta, Google, TikTok, Apple Search Ads — to show your app to targeted users and drive installs or trial starts. Unlike organic growth, paid UA gives you direct control over who sees your app, how much you spend, and what action you optimize for. For subscription apps, the goal isn't installs — it's paying subscribers.
**Key Takeaways**
- **Paid UA covers four primary channels.** Meta, Google, TikTok, and Apple Search Ads (ASA) drive the vast majority of paid installs — not influencer deals or app store optimization.
- **Don't optimize for CPI alone.** CPI tells you what your ads cost, not whether they're working. What matters is cost per paying user — divide CPI by trial start rate (median: 6.2%) to get cost per trial, then divide by trial-to-paid rate (median: 34.8%) to get cost per paying user.
- **You need three things before spending.** A paywall converting at least 1.9% of installs to paid subscribers, a defined optimization event, and a budget of at least 10x your target CPA.
- **Start where intent is highest.** For iOS, Apple Search Ads is the first channel. For Android, Google App Campaigns. Add TikTok only when your creative engine is ready.
- **The funnel has five checkpoints.** CPM → CPI → cost per trial → cost per paying user → LTV. You need all five to know if paid UA is working.
- **LTV must exceed CAC.** Among Health and Fitness subscription apps, the median year-one LTV per paying user is $27.21 — your cost to acquire that user must stay below it.
## What Does Paid UA Actually Include?
Paid UA is often conflated with every form of marketing that costs money. It isn't.
**Paid UA means running ads on self-serve platforms** that charge you per impression, click, or install. The four primary channels are **Meta** (Facebook, Instagram), **Google** (App Campaigns), **TikTok**, and **Apple Search Ads (ASA)**.
Paid UA does not include influencer deals, podcast sponsorships, brand awareness campaigns, or ASO (App Store Optimization). Those may complement paid UA, but they operate on different mechanics.
Channel reach varies significantly. TikTok reaches roughly 20% of the world's population — substantial scale, but only useful if your creative is built for the format. On iOS, the AppsFlyer Performance Index ranks Apple Search Ads first by install quality, followed by TikTok, Meta, and Google. On Android, Google App Campaigns leads, with Meta second. These are performance rankings — the order you should actually start them is covered in the channel selection section below.
The practical implication: **start with one channel, not four**.
## What Does Paid UA Cost?
The number most people track — **cost per install (CPI)** — is the wrong metric for subscription apps.
Typical CPI ranges for subscription apps:
- iOS: **$2–$6+** (Health & Fitness average: $5.78; category range is wide — Photo & Video averages $14.66)
- Android: **$1–$4+** (Tier-1 markets can exceed $5)
But a $2 CPI on iOS sounds cheap until you do the math. Among subscription apps, the median trial start rate is **6.2%** and the median trial-to-paid conversion is **34.8%**. Run those numbers:> $2 CPI × 100 installs = $200 spend → ~6 trial starts → ~2 paying users → **approximately $100 cost per paying user**
If your CPI is $5 (common on iOS), that same funnel produces approximately a **$250 cost per paying user**.
| Metric | Median | Top Quartile |
| --- | --- | --- |
| Trial start rate | 6.2% | 20.3% |
| Trial-to-paid conversion | 34.8% | 62%+ |
| Download-to-paid by Day 35 (D35) | 1.9% | 8.5% |
| Y1 LTV per payer (H&F) | $27.21 | — |
| D60 Revenue per Install (H&F) | $0.63 | — |
The top-quartile trial start rate of 20.3% is more than 3x the median. That gap is entirely paywall and onboarding — not ad creative.
[Paid UA Funnel diagram]
_A $5 CPI with median conversion rates produces approximately $250 per paying user. At top-quartile trial start rate (20.3%), that drops to ~$71._
## When Should You Start Paid UA?
The single most common mistake: starting paid UA before the conversion funnel works.
You need three things in place first:
**1\. A paywall that converts.** The median download-to-paid rate across subscription apps is **1.9%**. If your paywall is below this, paid UA will amplify a broken funnel — not fix it. Get to 1.9%+ before spending on ads.
**2\. A clear optimization event.** Platforms like Meta and Google need a conversion signal to optimize against. "Install" is too weak. "Trial start" or "purchase" is what you want. Without a defined event, the algorithm cannot find the right users.
**3\. Enough budget to learn.** At 10x CPA as a daily budget minimum, a $25 target CPA requires $250/day — or $7,500/month just to get clean data. Below this, the algorithm doesn't get enough signals to optimize.
The competitive pressure to start is real. Health and Fitness app UA spend on iOS grew **97% year-over-year (2024–2025)**. With over **2.2 million apps** on the App Store (42matters, April 2026), organic discovery alone won't build a business. But spending on a broken funnel is worse than waiting.
The threshold: **launch paid UA when your paywall hits 1.9% download-to-paid and your budget can sustain 10x CPA for at least 30 days**.
## How Do You Pick Your First Paid UA Channel?
The right answer depends on your platform, not your preference.
| Channel | Best For | Strength | When to Add |
| --- | --- | --- | --- |
| Apple Search Ads | iOS first installs | High-intent users, clean attribution, keyword-level targeting | Start here for iOS |
| Google App Campaigns | Android first installs | Android-first advantage, broad reach | Start here for Android |
| Meta (Facebook/Instagram) | Both platforms | Broadest targeting, lookalike audiences | After ASA/Google baseline |
| TikTok | Both platforms | Scale at low CPM | Only when creative engine is ready |
**iOS apps**: Start with Apple Search Ads. Users searching the App Store have high purchase intent — they're actively looking for an app like yours, not passively scrolling. One caveat: ASA optimizes toward clicks and installs, not purchase events, so pair it with downstream conversion tracking to measure trial starts. Once you have a cost-per-trial baseline from ASA, expand to TikTok (if your creative engine is ready) or Meta.
**Android apps**: Google App Campaigns is the natural starting point. Google's ecosystem gives it a structural advantage on Android — Play Store data informs targeting in ways other platforms can't replicate.
**TikTok**: Only start TikTok when you have a creative team (or freelancer) who can produce hook-first video at volume. TikTok's algorithm rewards creative iteration above all else. If you can't regularly test new creatives each week, you're not ready.
**Meta**: Useful for scale and lookalike audiences once you know what a paying user looks like. Start with broad targeting and let the algorithm find your audience — don't over-restrict.
## How Do You Know If Your Paid UA Is Working?
Track five metrics in sequence. If any one breaks down, you've found the problem.
**CPM (cost per 1,000 impressions)**: Measures ad auction competitiveness. If CPM spikes, your creative may have fatigued or competition in your category increased.
**CPI (cost per install)**: The product of CPM and click-through/install rates. High CPI usually means weak creative or wrong audience, not the wrong channel.
**Cost per trial start**: CPI ÷ trial start rate. If this is high, your app store page or onboarding is filtering out the users your ads are reaching.
**Cost per paying user**: Cost per trial ÷ trial-to-paid rate. This is the number that determines whether paid UA is profitable. Among subscription apps, the median trial-to-paid rate is **34.8%** — below that, investigate your paywall.
**ROAS / LTV ratio**: The final check. For Health and Fitness subscription apps, median year-one LTV per paying user is **$27.21** — and the top-quartile threshold is **$33.91**. If your cost per paying user is $40, you're losing money on every subscriber. No amount of creative testing fixes negative unit economics.
The practical rule: **don't optimize for CPI. Optimize for cost per paying user, then back-calculate the CPI that makes the math work**.
## Paid UA Is a Funnel Problem, Not an Ads Problem
Most subscription apps that struggle with paid UA aren't failing at ads — they're failing at conversion. The ad platforms will deliver users. What happens after the install determines whether paid UA is a growth engine or a money drain.
Fix your paywall first. Define your optimization event. Set a budget that gives the algorithm room to learn. Then choose one channel, measure all five funnel metrics, and scale what works.
_Sources)_
- _Analysis based on RevenueCat State of Subscription Apps 2025 (75K+ apps)_
- _Adapty State of In-App Subscriptions 2026 (16K+ apps)_
- _RevenueCat Startup School — Paid User Acquisition (35-lesson course); 42matters App Store data (April 2026)._
---
## Why 56% of New Apps Stay Under $1K: 3 Marketing Mistakes
URL: https://www.airbridge.io/en/blog/how-to-begin-marketing-a-subscription-app
Language: English
Published: 2026-04-21
Author: Jaehyuk Kim
You built the app. You hit publish. Then nothing happens.
No surge in downloads. No word-of-mouth. Not even a trickle of new users. Most founders assume the silence means the product is broken. It usually doesn't. It means the go-to-market hasn't started yet.
This guide gives you the exact sequence from pre-launch readiness through organic **app marketing** tactics to the point where paid UA becomes viable.
**Key Takeaways**
- **56% of new apps earn less than $1,000 in their first year.** The gap is almost never about product quality alone.
- Run a 4-point readiness checklist before spending a dollar on any channel.
- Almost every successful consumer app built its first 1,000 users through **one primary strategy** — not five.
- iOS Health & Fitness CPI averages **$5.78**, with a cost per paying user of **$23-29**. Paid UA only works when your funnel is already converting.
- A single onboarding question — "How did you hear about us?" — is your best zero-cost measurement tool at this stage.
- Lifecycle marketing (email, push) can add ~10% to ARR but cannot rescue a broken funnel.
---
[Marketing Sequence Diagram]
## Why Most New Apps Never Find an Audience (And It's Not the Product)
**App marketing** is the practice of acquiring, activating, and retaining users for your app through deliberate channel strategy — not luck or virality.
**The top 10% of subscription apps capture 95% of all subscription revenue.** The median subscription app earns just $92/month after its first full year. That gap is stark, and it rarely comes down to product quality alone.
The real pattern: most founders ship the app and then wait. They wait for the App Store algorithm to surface them. They wait for word-of-mouth to kick in. They wait for organic traffic that never comes.
The apps in the top tier treated go-to-market as a deliberate discipline, not something that happens automatically after shipping.
There are 2.3 million apps on the App Store and 1.5 million on Google Play. Consumer spending on iOS reached $103 billion in 2024. At **258,000 app downloads per minute globally**, user attention is the scarcest resource in the market. Attention must be earned or purchased. Neither happens by accident.
The three components of product-market fit tell you where things break down: **Product** (the right solution), **Market** (a specific audience), and **Messaging** (packaging that helps people understand why it matters). Getting messaging wrong is as costly as getting the product wrong. Most early-stage apps fail at the third piece, not the first.
## Run the Pre-Marketing Readiness Checklist
Marketing amplifies what's already working. If the core conversion experience is broken, spending on channels makes the problem worse faster.
Before you pick a single marketing channel, verify that these four conditions are true.
### 1\. You can reach your target audience
You need at least one path to the people you're building for. A community you're active in. A network segment. A platform where they congregate. If you have no idea where your target users spend time online or offline, channel selection is premature.
### 2\. Users can actually try the core experience
Onboarding friction is invisible to the builder. Someone who has never used your app should be able to reach the moment of value within the first session. If your onboarding requires too many steps or too much prior knowledge, users leave before they experience the product.
### 3\. The app genuinely helps with the Job to Be Done
**82% of free trial starts happen on Day 0 — the first session.** The decision to try or not is made almost immediately. That means your store listing, screenshots, and onboarding flow are marketing assets as much as the product itself. Weak store creative kills conversion before any channel gets a chance to prove itself.
### 4\. You have some signal on willingness to pay
Before scaling acquisition, you need at least soft evidence that someone would pay for this. Willingness to pay signals: users who return without prompts, users who tell others organically, users who ask about pricing. Vanity metrics — downloads, social following, App Store rankings — are misleading at this stage.
Run through the checklist honestly. If more than one item fails, fix those before investing in marketing.
## Get Your First 100 Users Without Paying for Ads
The principle here: **make a small group genuinely happy before expanding outward.**
Trying to reach everyone at the start dilutes the signal you need to improve. Your first 100 users are a learning cohort, not a revenue target.
Research on how the biggest consumer apps reached their first 1,000 users consistently shows the same pattern: one primary strategy, executed deeply. Buffer wrote 150 guest posts in 9 months and built 100,000 users organically. Loom got 3,000 users from a single Product Hunt launch. Slack's press blitz generated 8,000 invitation requests on day one.
Pick one channel. Go deep before you go wide.
### 1\. Start with your personal network
The fastest path to your first users is direct. Email people you know in your target segment. Be specific about what you're looking for: "I built an app for \[specific problem\], and I think it would help you. Can I send you a link and get 15 minutes of feedback next week?" This is not scale, but it gives you signal fast with zero budget.
### 2\. Go where your audience already exists
Online communities (Reddit, niche Discords, Facebook Groups, Slack communities) are distribution channels if used right. The wrong approach: drop a launch post and leave. The right approach: be an active contributor for weeks before mentioning your app, then share it as something that might help — not as a promotional post.
The most effective first-channel strategy for community-based apps: find the 3-5 forums where your target audience asks questions you can answer with your product, contribute genuinely, and reference your app when it's directly relevant.
### 3\. Optimize your App Store presence before promoting anything
**40% of app users discover apps through App Store search or browsing.** Your store listing is the highest-leverage free marketing asset you have.
The components that move conversion:
| Element | What to prioritize |
| --- | --- |
| Title + Subtitle | Primary keyword in title; secondary keyword in subtitle |
| Screenshots | First screenshot communicates the core value proposition in 3 seconds |
| Preview video | 15-30 seconds, shows actual product experience |
| Ratings and reviews | Prompt at the right moment — post-achievement, not mid-task |
| App quality signals | Crash rates and retention now influence ranking algorithms |
A/B test your store listing assets before spending on paid. This is free and directly improves every channel's conversion rate.
### 4\. Partner with micro-influencers in your niche
Micro-influencers (1,000-50,000 followers) consistently outperform larger accounts for early-stage app launches. Their audiences are more engaged and more niche, which means your conversion rate is higher even at lower reach.
The outreach approach that works: give them early access, let them form their own opinion, give them creative freedom on how they describe it. Require a trackable referral code so you know what's working. Do not write their script for them.
**The Earned/Owned/Paid framework** is the mental model for this phase. Owned = your app, website, store listing. Earned = word-of-mouth, press, organic social. Paid = advertising. In the early stage, you are almost entirely in Owned and Earned territory. Virality is unpredictable — do not plan for it. Build for it, but don't count on it.
## When (and How) to Start Paying for Users
Paid UA is an amplifier — it accelerates what's already converting organically, not a fix for what isn't. Here's what that means in practice.
**iOS CPI for Health & Fitness apps averages $5.78 globally.** Cost per paying user (CPPU) runs approximately 4-5x CPI — so **$23-29 per paying user**. Median Day 60 revenue per install in Health & Fitness is $0.63. At those economics, you need a strong conversion funnel and a clear payback timeline before a paid campaign becomes profitable.
Paid users tend to pay off 6-8 months after the initial install. When your app earns $5/month in subscription revenue but you must pay the ad platform today, cash flow can kill the business before LTV is ever realized.
[Paid UA Economics Diagram]
### 1\. Know which channels match your platform
**On iOS, Apple Search Ads consistently ranks first for install quality.** On Android, Google Ads is the equivalent anchor channel. Meta and TikTok are strong supplementary channels for non-gaming subscription apps in North America. Start with the platform where your audience already has intent, not the one with the lowest CPI.
### 2\. Define your readiness conditions
Three things must be true before you run paid:
- You have at least 50-100 organic users whose behavior you understand
- Your trial start rate shows users find value — even 5-8% is a start, but make sure it's not declining
- You know your payback window and have cash to sustain it
If any of these fail, organic work is not finished.
### 3\. Start with a test budget, not a campaign budget
Run a $500-1,000 test per channel before committing. Look at cost per trial, not cost per install. Cost per install tells you about the channel's efficiency. Cost per trial tells you whether the channel is sending people who actually want what you're building.
## How to Know Whether Your Marketing Is Working
Early-stage measurement does not require complex infrastructure. Most founders over-invest in tooling and under-invest in reading the data they already have.
The metrics that matter at this stage:
| Metric | What it tells you | Target signal |
| --- | --- | --- |
| Activation rate | % of installs that reach core feature | Should improve with each onboarding change |
| Day 1 retention | % who return next day | Does it trend upward as you improve onboarding? |
| Trial start rate | % of installs that start a free trial | 90th percentile: 20%+; top quartile: 12%+; all-category median: ~6% (SOSA 2025) |
| Trial-to-paid conversion | % of trials that convert to paying | All-category median: ~35%; H&F median: ~40% (SOSA 2025) |
| Channel source distribution | Where your actual users come from | Tells you where to double down |
### 1\. Ask "How Did You Hear About Us?" in onboarding
**This single question is the simplest zero-cost channel signal source you have — and it complements, not replaces, any analytics tools you add later.** Add it as one screen during onboarding — before the paywall, after the core experience. Offer 5-8 options (App Store search, friend/family referral, Instagram, TikTok, Google search, podcast, other) and an open text field.
You will immediately see which channels are producing real users, not just installs. This data correlates with downstream LTV: users who came from referrals consistently outperform paid installs on retention and subscription conversion. You do not need any external tools to run this.
### 2\. Track activation, not downloads
Downloads are vanity at this stage. **Activation** — defined as the moment the user completes the core intended action for the first time — is the real leading indicator. Every optimization to onboarding should be measured against its effect on activation rate.
### 3\. Watch early retention as a product signal
**Nearly 80% of users abandon new apps within 3 days.** If your Day 3 retention is below 20%, no amount of marketing will build a sustainable app. This is a product problem, not a marketing problem. Fix it before investing further in acquisition.
## 3 Mistakes That Kill Early-Stage App Marketing
### 1\. Spreading across too many channels at once
The data from the most successful early-stage consumer apps is consistent: one channel, done deeply, outperforms five channels done shallowly. Spreading across channels at this stage means you understand no channel well enough to improve it. You end up with noise, not signal.
The practical consequence: founders who spread across three channels at launch typically can't tell which channel is working, can't allocate budget correctly, and can't iterate fast enough on any single channel to improve it. The data stays noisy indefinitely.
### 2\. Running paid ads before organic validation
Paid UA applied to a leaky funnel produces expensive data about a broken product. If your trial start rate is 5% and your trial-to-paid conversion is 10%, spending on Meta ads gives you a high CPPU and no clarity on what to fix.
Organic users are more forgiving of product imperfections and more likely to give feedback. Use them to fix the funnel. Then pay to fill it.
### 3\. Investing in lifecycle marketing too early
Email sequences and push notifications have a hard ceiling: even done well, they move revenue by around 10%. That's meaningful at $1M ARR. It's irrelevant if your trial start rate or trial-to-paid rate is below median. If your core conversion isn't working, spend zero time on welcome sequences and re-engagement campaigns. Come back to them when you have a subscription base worth retaining.
## Audience Is Built, Not Found
Every app that has an audience today started with zero users. The ones that built something real were deliberate about the sequence: validate the experience, earn the first hundred, understand what's working, then scale.
There is no shortcut that skips the early organic phase. The data on paid UA payback windows makes this concrete: you need a working funnel before you can afford to fill it.
The sequence in this guide is the one that works consistently across verticals:
**Readiness check → one organic channel → store optimization → paid when the funnel converts → measurement from day one.**
Start where you are. Build the audience the way every successful app did — one deliberate user at a time, until the channel compounds on its own.
_Sources)_
- _Lenny's Newsletter, "How the Biggest Consumer Apps Got Their First 1,000 Users"_
- _MobileAction, "Organic App Growth in 2025"_
- _SplitMetrics, App Store Acquisition Channels glossary. Benchmark data from: RevenueCat State of Subscription Apps 2025 (75,000+ apps)_
- _Adapty State of In-App Subscriptions 2026 (16,000+ apps)_
- _RevenueCat StartApp School (Paid UA + App Monetization modules)_
- _48 Laws of Subscription App Success playbook._
---
## What Is Freemium and How Is It Different From a Free Trial?
URL: https://www.airbridge.io/en/blog/freemium-vs-free-trial-difference
Language: English
Published: 2026-04-20
Author: Jaehyuk Kim
Every subscription app founder hits this question early: should I let users in for free forever, or put a clock on it? The wrong choice doesn't just affect sign-ups. It reshapes your entire conversion economics, your payback period, and how much each user is actually worth over time.
Freemium gives users a limited version of your app permanently. A free trial gives full (or near-full) access for a fixed window, then asks them to pay or leave. Both feel like "free," but they create very different user journeys and very different businesses.
Which model you choose will determine how fast revenue comes in, what kind of users you attract, and whether your onboarding actually converts. Here is what the data says.
## Key Takeaways
- **Freemium gives permanent access to a limited product.** Free trials give full access on a countdown clock.
- **Freemium converts 2–5% of users to paid.** Free trials convert 8–25%+ depending on setup, with opt-out trials reaching 48.8%.
- **Free trial users generate up to 64% higher LTV** than direct buyers in categories like Health and Fitness.
- **Adding a free trial to a weekly subscription** can raise 12-month LTV from $7.40 to $54.50, a 636% increase.
- **82% of free trial starts happen on the day of install.** Your first-session onboarding determines almost everything.
- **Freemium works best for massive-TAM consumer products** like Spotify or Duolingo. Free trials work better for most subscription apps.
## What Freemium Actually Means
Freemium is a permanent free tier. Users get real access to your app with no expiration date, but certain features, content, or capacity are locked behind a paid plan.
There are three common ways apps gate the premium experience:
- **Feature-gating**: Core features are free; advanced features require payment (Notion, Slack)
- **Capacity limits**: Basic usage is free; higher volume costs money (Dropbox storage, Spotify skips)
- **Ad-supported**: The product is free, but paying removes ads or unlocks more (Duolingo, Spotify free tier)
The economics of freemium are a volume play. Industry data puts freemium conversion at 2–5% for most products (RevenueCat's aggregate median sits at 2.1%), with exceptional products reaching 7–10%. That means you need a very large top of funnel to generate meaningful paid subscribers.
Freemium captures late converters better than any timed model. About 23% of freemium conversions happen six or more weeks after download. Users who would never start a timed trial may gradually hit a limit, see the value, and upgrade on their own schedule.
There is also a less obvious risk. Free users have different expectations than paying users, and optimizing for their engagement can pull your roadmap away from what actually drives revenue. If usage metrics go up but no one upgrades, you may be building the wrong thing.
## How Free Trials Work
A free trial gives users access to the full (or substantially full) product for a limited time. When the window closes, they either pay or lose access.
The first decision that shapes trial performance is whether you require a credit card upfront.
- **Opt-out trials** (credit card required at signup): 48.8% trial-to-paid conversion
- **Opt-in trials** (no credit card required): 18.2% trial-to-paid conversion
Opt-out trials convert much higher, but the gap isn't explained by friction alone. It comes down to which fear the user brings to the paywall.
Users who are unsure the product will work for them benefit from a trial. It removes the need to believe upfront. But users worried about forgetting to cancel are more anxious with a trial than without one, and that anxiety reduces conversion. And users who simply don't want to commit will postpone the decision rather than make it, which is what drives high trial starts but low trial-to-paid conversion in some categories.
Trial length matters less than most founders assume. RevenueCat research across 115,000+ apps found that trials between 5 and 32 days show roughly equal median conversion (44–46%).
What changes with longer trials is drop-off before conversion: 30-day trials see 51% pre-conversion cancellation, compared to just 26% for 3-day trials. Shorter trials often work just as well, and they bring revenue in faster.
The right trial length is as long as it takes for a user to experience your core value, and no longer. Not what competitors do. Not what feels safe.
Timing matters more than trial length. 82% of trial starts happen on Day 0, the same day as install. Top-performing apps (P90) achieve a 20.3% trial start rate versus a 6.2% median.
That gap is almost entirely determined by first-session onboarding. If users do not start a trial during their first session, most never will.
## Freemium vs Free Trial: The Key Differences
| Dimension | Freemium | Free Trial |
| --- | --- | --- |
| Access level | Limited features or capacity | Full (or near-full) access |
| Time limit | None | Fixed window (3–30 days typical) |
| User psychology | "I can use this forever" | "I need to decide before time runs out" |
| Conversion rate | 2–5% (exceptional: 7–10%) | 8–25%+ depending on setup |
| Best for | Large-TAM consumer products | Most subscription apps |
| Revenue speed | Slow (long tail) | Fast (convert or churn quickly) |
| Examples | Spotify, Duolingo, Notion | Most fitness, productivity, health apps |
The fundamental tradeoff is breadth versus speed. Freemium casts a wide net and waits. Free trials create urgency and filter for users who are genuinely interested.
Free trials suit apps where value is visible quickly and the audience is defined. Freemium suits products that need time, scale, or virality to work.
[Freemium vs Free Trial User Journey]
---
## Which One Converts Better?
When you look at the aggregate numbers, free trials win clearly. Hard paywall apps (trial-first) convert at 10.7% versus 2.1% for freemium, based on RevenueCat data across 115,000+ apps analyzed.
The LTV story is similarly compelling. Adding a free trial to a weekly subscription plan raises 12-month LTV from $7.40 to $54.50, a 636% increase. In Health and Fitness specifically, trial users show a 63.6% LTV premium over direct buyers, with a 35% trial-to-paid conversion rate.
But the category matters. In Productivity apps, direct buyers generate higher LTV ($56.95) than trial users ($49.13).
Trials underperform there because productivity tools have a longer time-to-value. Users who do not see the benefit during the trial do not convert, and those who do are willing to pay without the prompt.
The Tettra case study illustrates this nuance. Tettra, a B2B SaaS knowledge management tool, tripled its upgrade rate and sustained 70%+ monthly retention after switching from a 15-day trial to a freemium model. For a product where value accumulates over weeks of team collaboration, putting a clock on it was counterproductive.
The answer to "which converts better" depends on two things: how quickly a new user can experience real value, and whether your product's value is better felt in breadth or in time.
## How to Pick the Right Model
Before choosing, answer three questions.
**How fast can a new user experience core value?** If a user gets meaningful value in the first 10 minutes, a free trial works. If value builds over weeks of data, habits, or collaboration, freemium may fit better.
**How large is your addressable market?** Freemium needs volume. A 2–5% conversion rate only works if your top of funnel is enormous. If you are targeting a specific vertical or ICP, free trials are more efficient.
**Does your app rely on habit formation over weeks?** Apps where users need to build a routine before seeing results (journaling, meditation, language learning) may find freemium more forgiving. Trials cut off users before the habit sticks.
Use a free trial if:
- Your core value is visible within a single session
- You are in a vertical with clear outcomes (fitness, language learning, health)
- You need faster revenue signal to optimize onboarding
- Your market is specific rather than mass-market
Use freemium if:
- Value compounds over time (collaboration, habit, data accumulation)
- Your addressable market is very large (consumer, horizontal SaaS)
- You want to build product-led growth through viral sharing or team invites
- Your product competes on distribution, not conversion speed
One option worth testing: a hybrid model. Start users with a free trial. If they do not convert when the trial ends, offer a freemium fallback instead of cutting them off entirely. You keep the user in your ecosystem while removing the time pressure. Some apps have seen meaningful late conversions from users who re-engaged weeks after a failed trial.
## Start Simple, Then Test
For most subscription app founders, the right starting point is a free trial. The conversion data is stronger, the revenue signal comes faster, and the onboarding feedback loop is cleaner.
Once you launch, track two numbers: trial start rate (what percentage of installs start a trial) and trial-to-paid conversion. Top-performing apps hit 20%+ trial start rates. If you are well below that, the problem is in your first-session onboarding, not your pricing.
If you find that users consistently fail to convert within your trial window but remain engaged afterward, that is a signal to explore freemium or a longer trial. Let the data tell you which model fits your product.
Want more on subscription app growth? Read our paywall optimization guide at /blog.
---
---
## What App Marketing Actually Means for a Subscription App
URL: https://www.airbridge.io/en/blog/what-app-marketing-means-subscription-app
Language: English
Published: 2026-04-20
Author: Jaehyuk Kim
Most subscription app founders treat marketing as running ads. Spend money, get installs, watch the numbers.
But ads are only one of five things that determine whether your app grows. App marketing covers the full journey, from getting discovered in the App Store to keeping subscribers from canceling. Most of the work happens after the install.
This guide breaks down all five parts, the benchmarks to measure them against, and the order to build them.
**Key Takeaways**
- **App marketing is much broader than ads.** ASO, onboarding, paywall design, and lifecycle campaigns are all part of it.
- **Subscription apps optimize for LTV, not downloads.** A user who installs and churns after one month is a loss. A user who renews annually for two years is the goal.
- **The median trial start rate across 75,000+ apps is 6.2%.** The top 10% hit 20.3%, more than 3x higher. The gap is driven by onboarding and paywall design, not ad spend.
- **Annual subscribers retain at 3x the rate of monthly subscribers.** The 12-month retention median is 44.1% for annual vs. 17.5% for monthly. How you design your subscription plan shapes your marketing economics.
- **Fix your funnel before you run ads.** Pouring budget into a leaking funnel just makes the losses bigger.
## Most Founders Treat Marketing as Just Ads — Here's What They're Missing
What gets missed is everything between the install and the renewal. App marketing spans five stages, and paid ads cover just one:
1. **Discovery** — ASO and organic content get users to find your app
2. **Acquisition** — Paid UA (Meta, Google, TikTok, Apple Search Ads) drives installs
3. **Activation** — Onboarding delivers the first moment of real value
4. **Conversion** — Paywall design turns trial users into paying subscribers
5. **Retention** — Lifecycle marketing prevents cancellation and drives renewals
[The 5-Stage App Marketing Funnel]
Subscription apps also have a fundamentally different revenue structure than games or e-commerce apps. A one-time purchase delivers immediate revenue. A subscription delivers revenue month after month, but only if the user sticks around.
In Health & Fitness, for example, the median CPI (cost per install) on iOS is **$5.78**. But the actual cost to acquire one paying user (the CPPU) runs **$23–$29**, roughly 4–5x higher. That gap exists because only a fraction of installs ever subscribe.
Here's what that means in practice: if it costs $25 to acquire a paying user and that user cancels after one month of a $12 subscription (about $8.40 after the app store's 30% cut), you've lost $16.60 on that customer.
If that same user converts to annual and stays for two years, the math looks completely different. **The goal is LTV optimization, not download count.**
## What App Marketing Actually Includes (Beyond Ads)
Here's what subscription app marketing covers beyond paid ads. The order below follows the user journey, not the sequence in which you should build them. That comes in the last section.
### 1\. ASO: The free installs you're probably leaving on the table
ASO is the work of making your app rank higher when users search the App Store or Google Play. It covers your app title, keyword field, screenshots, ratings, and review responses.
It's the most cost-efficient way to drive organic installs. No ad spend required. And it needs to be in place before you launch paid UA.
Users who click your ads also see your App Store page before installing. Weak screenshots and a poor description waste paid traffic.
### 2\. Paid UA: The amplifier (but only if the funnel is ready)
Paid user acquisition is running ad campaigns to drive installs. Subscription app UA spend has grown **+24% year-over-year**, with Android paid installs up **+57%**.
Each channel has different characteristics. Early-stage apps do better by concentrating budget on one channel first rather than spreading thin.
Apple Search Ads (ASA) targets users who are already searching the App Store for apps like yours. Because intent is already present, conversion rates tend to be higher relative to other channels. Setup is also simpler than Meta or Google, making it the right starting point for teams new to paid UA.
Build baseline data from ASA first, then expand to Meta and Google once your funnel is validated.
### 3\. Onboarding: The part that actually drives trial starts
Onboarding is the path from a user's first app open to the moment they commit to a trial. The mistake most teams make is treating it as a single moment to optimize — a search for the one "aha moment" that makes users pay. High-converting apps engineer a sequence: get the user set up, deliver a real first result, then start building a habit. Each step has a different job.
The median trial start rate is **6.2%** across all subscription apps. The top 10% hit **20.3%**. For Health & Fitness specifically, the median is **7.8%** and the top 10% reach **24.1%**.
That gap is not driven by ad targeting. It's driven by onboarding design.
### 4\. Paywall: A marketing asset, not just a checkout screen
Your paywall is not just a checkout screen. It's a marketing asset that needs to be optimized. The four dimensions to work on: layout design, pricing structure, when and where the paywall appears, and what offers you show.
The median trial-to-paid conversion rate is **34.8%**. Apps offering longer trials (17–32 days) convert at **45.7%**, 70% higher than short-trial apps (26.8%).
**82% of trial users start their trial on Day 0.** The first session determines whether a user starts a trial at all. Paywall timing and first-session design are not separate problems.
### 5\. Lifecycle marketing: The last layer, not the first
Once a user is a paying subscriber, lifecycle marketing works to prevent cancellation and drive renewals. User lifecycle has four stages: Potential → Active → Drifting → Churned. Each stage needs a different message.
Annual subscribers retain at a **44.1%** 12-month median, compared to **17.5%** for monthly and just **3.4%** for weekly.
Lifecycle marketing's revenue contribution ceiling is roughly **10%**, even when done well. Investing in lifecycle campaigns before your onboarding and paywall are working is a common mistake. Sequence matters.
## How to Tell Which Part of Your Funnel Is Broken
You can't know where to focus unless you're tracking the right numbers. For each metric, the question to ask when you're below the median is listed alongside.
| Metric | Median | If you're below median, check... |
| --- | --- | --- |
| Trial Start Rate (install → trial) | 6.2% | Onboarding flow, first session experience |
| Trial → Paid conversion rate | 34.8% | Paywall layout, trial length, pricing |
| 12-month retention (annual subscribers) | 44.1% | Annual plan positioning, upsell timing |
| 12-month retention (monthly subscribers) | 17.5% | Lifecycle messaging, churn intervention |
| CPI (Health & Fitness, iOS) | $5.78 | Ad creative, audience targeting |
| CPPU (cost per paying user) | $23–$29 | Trial start rate, trial→paid conversion |
_Based on analysis of 75,000+ subscription apps_
Category-level patterns reveal something important. Gaming apps have the highest Install→Trial rate at **12.2%**, but the lowest Trial→Paid rate at **19%**.
Lifestyle apps flip that: Install→Trial is just **3.9%**, but Trial→Paid hits **41%**. Education reaches **42%** Trial→Paid.
High trial volume isn't automatically good. It can signal poor ad targeting or a paywall that isn't converting. Quality matters more than quantity.
## The Right Order to Build Your Marketing (Sequence Matters More Than You Think)
The five activities above all matter, but they don't all matter at the same time.
### 1\. Before you spend on ads, check these two numbers
Ask yourself two questions before spending a dollar on ads: "What is the trial start rate for my organic users right now?" and "Of users who start a trial, what percentage convert to paid?"
If those numbers are below the benchmarks above, adding ad spend will just amplify the losses. Fix the funnel first.
### 2\. Get organic installs flowing before you open your wallet
Optimize your App Store title, keyword field, and screenshots. When organic installs start flowing in, you can accumulate funnel data without burning ad budget.
This data becomes the baseline you'll need to evaluate paid channels.
### 3\. When you're ready for paid, start with the simplest channel
Once your organic funnel shows acceptable conversion, start paid UA with ASA. Use a small budget. Measure Trial Start Rate and Trial→Paid conversion.
If those numbers hold up, increase budget and expand to Meta and Google.
### 4\. Retention is the last thing you build, not the first
After onboarding, paywall, and paid UA are stable, layer in lifecycle marketing. Send re-engagement push notifications to Drifting users. Offer annual upsells to monthly subscribers.
The sequence isn't optional. Lifecycle campaigns on top of a leaking funnel produce almost nothing.
## Where to Go From Here
App marketing isn't about running more ads. It's about knowing which of the five stages is underperforming and fixing that first.
If installs are growing but revenue isn't, the answer is in your funnel data, not your ad creative. Check your Trial Start Rate, Trial→Paid conversion, and 12-month retention against the benchmarks above. The number furthest below median tells you exactly where to start.
---
## Subscription App Event Tracking: A 5-Step Guide
URL: https://www.airbridge.io/en/blog/subscription-app-event-tracking-setup-guide
Language: English
Published: 2026-04-02
Author: Jaehyuk Kim
**Subscription app event tracking is the process of defining and recording the specific in-app actions - from Install to Subscribe - that connect ad spend to subscription revenue.** A fitness app tracks dozens of in-app events. Amplitude warns that [too many events](https://amplitude.com/docs/get-started/select-events) "actually obscures insights." The growth team exports a funnel report to find which Meta ad set drives paid subscribers. The report is empty - none of those events map to the subscription funnel.
**The cost of event sprawl is invisible revenue.** Without a clean Install-to-Subscribe funnel, every dollar spent on [paid UA](https://www.airbridge.io/blog/what-is-mobile-app-attribution) goes unattributed.
**Key Takeaways**
- **Five events cover the subscription funnel.** Install, Onboarding Complete, First Workout (activation), Start Trial, and Subscribe - each answers a distinct business question.
- **Naming consistency matters more than event volume.** Inconsistent event names across teams fracture funnel reports and make attribution data unreliable.
- **Predefined standard events eliminate naming decisions.** Small teams without a data engineer can skip taxonomy design entirely.
- **Airbridge Core Plan offers 25 subscription-optimized standard events** with GMAT attribution and RevenueCat integration - 15K free attributed installs.
## Why Most Small Teams Get App Event Tracking Wrong
If you have run paid UA for a subscription app, you have seen funnel reports that do not match ad platform numbers. The root cause is usually how events were set up.
### 1\. The Event Sprawl Problem
Product teams add events for features, engineering for debugging, growth for campaigns. Within months, the catalog has hundreds of entries. **No single funnel connects Install to Subscribe.** As Amplitude's taxonomy guide warns, "[distrust in the data and incorrect conclusions](https://amplitude.com/blog/event-taxonomy)" follow when event names are unclear.
### 2\. The Fitness App Tracking Gap
Generic schemas track Install, Sign-up, and Purchase. But for fitness apps, the gap between onboarding and trial contains the most important signal: **did the user complete their first workout?** Without this activation event, growth teams cannot tell if poor subscription rates come from bad targeting or bad activation.
## The 5 Events That Map Your Subscription Funnel
Start with five events that map directly to business decisions.
[Vertical flow diagram showing the 5 subscription funnel events: Install, Onboarding Complete, First Workout, Start Trial, Subscribe]
| Funnel Event | Question It Answers | Business Decision It Enables |
| --- | --- | --- |
| Install | Which channels drive downloads? | Allocate top-of-funnel budget |
| Onboarding Complete | Are users finishing setup? | Fix onboarding friction or shorten flow |
| First Workout | Are users reaching activation? | Optimize in-app nudges; identify best creative-to-activation paths |
| Start Trial | Which users are considering paying? | Measure cost per trial by channel; refine targeting |
| Subscribe | Which channels drive paying users? | Calculate true ROAS and reallocate spend |
### 1\. Install - The Attribution Anchor
The install event ties every downstream action to its source channel. Without an [MMP](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) to deduplicate, Self-Attributing Networks each claim credit for the same install.
### 2\. Onboarding Complete - The Engagement Gate
Fires when a user finishes initial setup. A sharp drop between Install and Onboarding Complete from one channel signals audience mismatch.
### 3\. First Workout (Activation) - The Conversion Predictor
For fitness apps, this is the most diagnostic event. Users who complete their first workout within the first few sessions convert to trial at significantly higher rates than those who do not. Standard events like "Complete Tutorial" or "Achieve Level" can map to this activation moment.
When this event is attributed to a channel, it reveals something installs alone cannot: **which ad creatives drive users who actually engage with the product, not just download it.** A channel with 1,000 installs and 50 first workouts tells a different story than one with 400 installs and 200 first workouts.
### 4\. Start Trial - The Revenue Signal
Start Trial marks the transition from free user to revenue candidate. Attributed to a channel, it becomes [cost per trial](https://www.airbridge.io/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) - the metric that separates productive ad spend from wasted budget.
If your Meta campaign drives 500 trials at $4 each and your TikTok campaign drives 200 trials at $12 each, the budget decision is clear - but only if Start Trial is tracked and attributed per channel.
### 5\. Subscribe - The Payback Moment
With Subscribe attributed, you can calculate [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv) by channel and determine which ad sets pay back within your target window.
[**_25 subscription-optimized standard events. Skip taxonomy design. Start Free._**](https://www.airbridge.io/en/plans/core)
## How to Name and Organize Events Without a Data Engineer
Event tracking fails because naming conventions break down. When one team calls it "sign\_up" and another calls it "SignUpEvent," you get what Woopra calls "[errors, duplicate events, and messy data](https://www.woopra.com/blog/event-tracking-mistakes)."
[Side-by-side comparison of DIY Custom Schema vs Predefined Standard Events]
### 1\. Predefined vs. Custom Event Schemas: What Startups Actually Need
Most [MMPs](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) offer both predefined standard events and custom events. The question for small teams: which approach ships this week?
| Factor | Custom Schema (DIY) | Predefined Standard Events |
| --- | --- | --- |
| Schema design time | Days to weeks of taxonomy work | Already decided |
| Naming decisions | Team must define every name | Pre-decided (Start Trial, Subscribe) |
| Data engineer needed? | Strongly recommended | Not required |
| Flexibility | Unlimited custom events | Limited to predefined set |
| Risk of inconsistency | High without governance | Low - standardized |
| Best for | Mature teams with analytics resources | Small teams needing fast time-to-value |
Custom schemas require a data engineer to maintain naming governance. Experienced practitioners estimate schema design takes [a week or two](https://mixpanel.com/blog/build-event-tracking-scheme-business-metrics/) - time a small growth team cannot spare.
### 2\. The Naming Convention Trap (and How to Avoid It)
Common mistakes: mixing camelCase with snake\_case, ambiguous names like `event_1`, and duplicates across platforms. Predefined events solve this by removing the decision. The event name is `Start Trial`, not whatever your team decided last Tuesday.
## How Airbridge Core Plan Connects Events to Attribution
For small teams needing attribution data this week, the setup path matters as much as the feature set.
Airbridge Core Plan provides 25 standard events with predefined names - Install, Complete Tutorial, Achieve Level, Start Trial, Subscribe, Unsubscribe, and more. Customers log these events in their app code, but schema design is already done. **Core Plan does not support custom events.** For teams whose priority is tracking Install to Subscribe by channel, 25 standard events cover the funnel without taxonomy overhead.
Attribution runs across GMAT channels - Google, Meta, Apple Search Ads, and TikTok. [RevenueCat](https://www.airbridge.io/blog/abr-revenuecat-integration) and Adapty S2S integrations capture subscription events like renewals and billing failures regardless of app state - closing the gap that SDK-only tracking leaves when the user is not in the app.
Core Plan supports up to 2 third-party integrations (e.g., RevenueCat + Amplitude). Pricing: 15K free attributed installs, then $0.05/install - pay-as-you-go, no annual contract.
## Your Funnel Is Only as Good as the Events That Define It
Which channels drive users who actually pay? The answer lives in five events - not hundreds. Get those right, and every growth decision has data behind it.
[**_Start Free with Airbridge Core Plan - 25 standard events, GMAT attribution, RevenueCat integration, 15K free attributed installs._**](https://www.airbridge.io/en/plans/core)
---
## Influencer Marketing Fraud: How Fake Clicks Drain App Budgets
URL: https://www.airbridge.io/en/blog/influencer-marketing-fraud-fake-clicks-mobile-app
Language: English
Published: 2026-04-02
Author: Jaehyuk Kim
A fitness app spends $8,000 on three influencer partnerships. The campaign dashboard reports 12,000 link clicks, 2,400 installs, and a 4.2% engagement rate. RevenueCat logs 11 new subscribers. Two of the three influencers generated zero revenue.
**Without per-influencer conversion data, most teams have no practical way to know whether those 2,400 installs came from real users or bot traffic -- and they typically discover the gap months too late.** Influencer marketing fraud is not a fringe risk. It is a structural problem in how mobile app teams measure influencer performance, and it drains budgets quietly.
**Key Takeaways**
- **74% of marketers have experienced influencer fraud** ([AMRA & ELMA, 2025](https://www.amraandelma.com/influencer-fraud-statistics/)). The problem spans every vertical, but subscription apps are especially vulnerable because fraud hides behind top-of-funnel vanity metrics.
- **Fake followers, bot clicks, and engagement pods inflate the metrics that most teams use to evaluate influencer ROI.** Each fraud type creates a different distortion in your funnel.
- **Standard analytics tools measure engagement, not subscription revenue.** GA4 and UTM links track clicks and installs but cannot connect those installs to downstream subscription events.
- **The most reliable fraud signal is post-install conversion data.** If 500 installs from one influencer produce $0 in subscription revenue, you have your answer -- no dedicated fraud tool required.
- **Airbridge Core Plan provides per-influencer** [**attribution**](https://www.airbridge.io/blog/what-is-mobile-app-attribution) **from install to subscription revenue** -- giving small teams conversion-level visibility without enterprise pricing.
## The Scale of Influencer Marketing Fraud in Mobile Apps
Influencer marketing fraud is the deliberate inflation of performance metrics -- followers, clicks, installs, or engagement -- to extract payment for results that never occurred. Industry data puts the annual cost between **$1.3 billion** ([AMRA & ELMA](https://www.amraandelma.com/influencer-fraud-statistics/), 2019 baseline) **and $4.6 billion** ([SociaVault, 2025](https://sociavault.com/blog/fake-follower-study-key-findings)) globally. The problem is concentrated among mid-to-large influencer tiers, where the financial incentive to inflate metrics is highest.
For subscription app teams, the risk is amplified because the gap between a fraudulent install and a real subscriber is wide -- but invisible without conversion-level data.
A 2025 fake follower study found that **37.2% of influencer followers are fake** across major platforms. Macro-tier influencers - those with 100K to 500K followers -- showed the highest fraud rate at **48.3%** ([SociaVault](https://sociavault.com/blog/fake-follower-study-key-findings)). For subscription app teams paying per-install or per-click, nearly half the reported reach may be artificial.
### 1\. Fake Followers and Inflated Reach
Purchased followers inflate an influencer's apparent audience size. A fitness influencer with 250K followers and 15K fake accounts looks identical to one with 250K real followers -- until you measure downstream behavior. Fake followers never install, never start trials, never subscribe.
### 2\. Bot-Generated Clicks and Installs
Bot traffic goes beyond fake followers. Sophisticated click farms generate real-looking clicks and even app installs that pass basic analytics filters.
TrafficGuard data indicates **31% of iOS app installs are fraudulent**, and **15-25% of ad spend is lost to invalid traffic annually** ([TrafficGuard](https://www.trafficguard.ai/guides/invalid-traffic-in-the-mobile-app-industry)). These bot installs appear in your dashboard as legitimate acquisition - they just never convert.
### 3\. Engagement Pods and Manufactured Social Proof
Engagement pods are groups of accounts that artificially like, comment, and share each other's content. For app marketers evaluating influencers by engagement rate, pods make fraudulent influencers indistinguishable from legitimate ones.
The common thread across all three fraud types: **top-of-funnel metrics look healthy while downstream conversions collapse.** If your evaluation stops at clicks or engagement rate, fraud remains invisible.
## Why Standard Analytics Miss Influencer Fraud
Most influencer platforms and analytics tools measure what happened on the platform - impressions, clicks, engagement. They do not measure what happened after the install.
[diagram-fraud-funnel.webp]
GA4 and UTM-based tracking can attribute an install to a specific link. But for subscription apps, the critical question is not "did they install?" - it is "did they subscribe?" UTM parameters break at the app store boundary. GA4 cannot connect a web click to an in-app [subscription event](https://www.airbridge.io/blog/subscription-app-attribution) without additional attribution infrastructure.
This creates a blind spot that fraudulent influencers exploit. Without post-install conversion visibility, **a channel that drove 500 real users looks identical to one that drove 500 bots**. Same install count. Completely different revenue outcome.
Influencer platforms report engagement. Ad networks report clicks. Your billing platform reports revenue. **Nothing connects these layers per influencer - unless you build that connection deliberately.**
[**_Per-influencer attribution from install to subscription. 15K free installs. Start Free._**](https://www.airbridge.io/en/plans/core)
## How Per-Influencer Attribution Exposes Fraudulent Traffic
If you are asking how to detect influencer marketing fraud - fake clicks, bot installs - in a mobile app, the answer is simpler than most vendors suggest.
You need **per-influencer conversion data from install to subscription revenue**. When the data is transparent, fraud reveals itself.
### The Attribution Approach: One Tracking Link Per Influencer
The foundation is simple: assign a unique tracking link to each influencer. Not one link for the campaign - one link per partner. This lets you measure every step of the funnel individually:
[diagram-analytics-blind-spot.webp]
- Influencer A: 800 installs, 120 trials, 34 subscriptions
- Influencer B: 1,100 installs, 4 trials, 0 subscriptions
- Influencer C: 500 installs, 95 trials, 28 subscriptions
Influencer B's pattern - high installs, near-zero trials - is a strong fraud signal. No engagement metric or follower audit would surface this. Only post-install conversion data does.
The revenue test is binary: **if hundreds of installs produce $0 in subscription revenue, the traffic is either fraudulent or so low-quality that it has the same effect on your budget.**
### How Airbridge Core Plan Makes This Practical for Small Teams
Dedicated fraud detection tools often cost $500-$2,000+/month and target enterprise buyers. For a small subscription app team, that price point is hard to justify.
Airbridge Core Plan provides the attribution layer that makes influencer fraud visible - not through fraud detection algorithms, but through **conversion-level transparency**.
Here is what the setup looks like:
- **Custom domain tracking links** - create one per influencer. Each link feeds into the same attribution pipeline, so you can compare performance side by side.
- **25 standard subscription events** - including Start Trial, Subscribe, and Unsubscribe. Customers log these events in their app code using predefined event names, reducing schema design work.
- [**RevenueCat**](https://www.airbridge.io/blog/abr-revenuecat-integration) **or Adapty S2S integration** - S2S enables real-time signal transmission independent of app state. Renewal, cancellation, and billing events flow into Airbridge without depending on the user opening the app.
- **Funnel and Revenue reports** - see Install to Start Trial to Subscribe per tracking link. When one influencer's funnel drops to zero at the trial step, you know where the problem is.
A [Mobile Measurement Partner (MMP)](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) connects these layers. Core Plan focuses on the four ad channels that cover 80-90% of typical spend - Google, Meta, Apple Search Ads, and TikTok - plus up to 2 third-party integrations like RevenueCat.
**The pricing removes the barrier.** 15K free attributed installs, then $0.05 per install. Pay-as-you-go, no annual contract. For most early-stage teams, the free tier covers initial influencer testing.
## FAQ: Influencer Marketing Fraud Detection for App Marketers
### What percentage of influencer followers are fake?
**37.2% on average, rising to 48.3% for macro-tier influencers** with 100K-500K followers ([SociaVault](https://sociavault.com/blog/fake-follower-study-key-findings)). Nano and micro influencers (under 50K) tend to have lower fraud rates, which is one reason small-audience partnerships often outperform larger ones on a cost-per-subscriber basis for subscription apps.
### Can I detect influencer fraud without a dedicated fraud tool?
Yes. Dedicated fraud platforms use device fingerprinting and bot-pattern algorithms - useful, but expensive ($500-$2,000+/month). For subscription apps, a simpler proxy works: **compare install-to-trial and trial-to-subscription rates across influencers**. If one partner's conversion rate is 10x lower than the campaign average, the signal is clear enough to act on - drop the partner or demand an explanation before the next payment. The data you need is [Lifetime Value (LTV)](https://www.airbridge.io/glossary/lifetime-value-ltv) per influencer, not a fraud score.
### What is the most reliable signal that an influencer partnership is fraudulent?
**The install-to-trial ratio, measured per influencer - not per campaign.** Legitimate influencer traffic typically converts 5-15% of installs to trial starts. An influencer with a 0.3% trial rate against a campaign average of 12% is a statistical outlier that warrants immediate investigation.
The critical mistake is measuring in aggregate. A campaign with three influencers might show a healthy 10% overall trial rate - but that average can hide one partner at 15%, one at 14%, and one at 0.4%. **Only per-influencer measurement surfaces the outlier.**
## Fake Clicks Cost Money - Conversion Data Saves It
Influencer marketing fraud persists because most teams stop measuring at the install. Engagement looks real. Installs look real. The divergence only appears downstream - at the trial step, at the subscription step, at the revenue line.
The fix is not more tools. It is connecting the data you already have - installs, trials, subscriptions - per influencer, in one view.
[**_Start Free with Airbridge Core Plan - per-influencer tracking, install-to-subscription funnel, 15K free._**](https://www.airbridge.io/en/plans/core)
---
## Airbridge Core Plan Is Here: Attribution That Finally Fits Your Stage
URL: https://www.airbridge.io/en/blog/airbridge-core-plan-introduction
Language: English
Published: 2026-03-27
Author: Jaehyuk Kim
# Airbridge Core Plan Is Here: Attribution That Finally Fits Your Stage
Your team spends $5,000--$10,000 per month on ads, but you still can't answer "**which channel is actually driving paying subscribers.**" Traditional MMPs cost more than your ad budget. GA4 can't connect ad channels to subscription events. Nothing fits your stage.
That's why Airbridge built **Airbridge Core Plan**, a focused attribution solution for **early-stage subscription apps** covering Meta, Google, TikTok, and Apple Search Ads. Signup to live attribution in hours, starting at $0.
**Key Takeaways**
- **Airbridge Core Plan is live.** A new Airbridge product built for subscription apps that need attribution without traditional MMP complexity.
- **Four channels, one funnel.** Tracks the full install-to-trial-to-subscribe journey across Google, Meta, TikTok, and Apple Search Ads.
- **Self-serve SDK, live in hours.** Install the SDK yourself, no sales call or consultant. Start collecting attribution data the same day.
- **$0 to start, $0.05 per install after.** 15,000 free attributed installs, then pay-as-you-go. No annual contracts.
- **RevenueCat, Adapty, and Superwall ready.** Connect your subscription revenue source of truth and see which channels drive real LTV.
## Why Existing Tools Don't Fix This
**Traditional MMPs** solve the attribution problem -- but they're built for companies with dedicated analytics teams, 20+ ad networks, and six-figure annual budgets. For a team of 2--10 people: 1,000+ network integrations you don't use, custom event schemas you don't have the engineering bandwidth to design, and a two-week setup process that delays optimization while your campaigns burn budget without signal.
**GA4** is free but unreliable for subscription revenue attribution. No channel-level [LTV](https://www.airbridge.io/glossary/lifetime-value-ltv). No creative-level analysis. No direct ad network integrations.
The gap is that **nothing is right-sized for your stage**. Airbridge Core Plan fills that gap.
## What Airbridge Core Plan Changes for Your Team
[diagram-product-ladder.webp]
Airbridge Core Plan sits between Airbridge's Base Plan (deep linking) and Growth Plan (full-scale analytics). It's not a stripped-down Growth Plan. It's a separate product, purpose-built for a specific stage: **past organic-only, actively spending on paid UA, not yet needing complex multi-channel measurement**.
### You Stop Guessing Which Channel Drives Subscribers
Airbridge Core Plan integrates with the four self-attributing networks where you actually spend money -- **Google, Meta, Apple Search Ads, and TikTok** (what Airbridge calls **GMAT**). These four channels account for 80--90% of typical mobile ad spend.
[diagram-dashboard-attribution.webp]
Each integration includes cost data, so you don't just see installs by channel. You see **Cost per Subscription,** [**ROAS**](https://www.airbridge.io/glossary/return-on-ad-spend-roas)**, and revenue by channel**. That's the difference between "Meta drove 2,000 installs" and "Meta drove 120 subscribers at $14 CPS, while Google drove 80 subscribers at $22 CPS." One is a vanity metric. The other is a budget decision.
### SDK Setup in Hours, Not Weeks — Fully Self-Serve
[diagram-sdk-onboarding.webp]
**The SDK is self-serve.** Your developer installs it, logs a few predefined events, and you're live in 1--2 hours. No sales call. No implementation ticket. No two-week onboarding blocking your campaigns. The **Quick Start Guide** walks you through each step, and the **Test Console** validates your setup before you go live. GMAT channel integrations and 25 standard subscription events come preconfigured out of the box.
**You start collecting attribution data the same day you decide to.** Your campaigns get signal faster, your budget decisions happen sooner, and you prove time-to-value before competitors finish their SDK setup.
### You See the Full Subscription Funnel, Not Just Installs
Most attribution tools tell you where the install came from. Airbridge Core Plan tracks what happens after.
[diagram-subscription-funnel.webp]
The **25 standard events** cover the full subscription lifecycle -- Install, Sign-up, Start Trial, Subscribe, Unsubscribe, Order Complete, and more. With these events connected to channel data, you can answer questions like:
- Which channel has the highest trial-to-subscription conversion rate?
- Which campaigns produce users who churn after the first billing cycle?
- Where should the next $1,000 go?
### You Get Answers from Day One, Not Day Fourteen
Six reports, each built to answer a specific question about your subscription business:
| Report | The Question It Answers |
| --- | --- |
| Actuals | What are my raw metrics by channel, campaign, or creative? |
| Trend | How are my key metrics changing over time? |
| Active User | What are my DAU, WAU, and MAU trends? |
| Funnel | Where do users drop off between install, trial, and subscription? |
| Retention | Which cohorts retain and which churn? |
| Revenue | Which channels drive actual subscription revenue? |
The **Revenue report** is where it comes together. This is where you see ad spend connected to subscription revenue by channel. That's the data that turns "I think Meta is working" into "Meta's D30 ROAS is 1.8x, Google's is 0.9x, let's reallocate."
### Your Subscription Revenue Data Actually Connects
[diagram-integrations.webp]
Airbridge Core Plan supports up to **2 active third-party integrations**. For most teams, this means connecting the two tools that matter most:
**RevenueCat, Adapty, or Superwall** -- your subscription revenue source of truth. Without this connection, attribution stops at the install. With it, Core Plan can trace revenue back to the ad that started the journey.
**Amplitude, Mixpanel, or Braze** -- your product analytics. Connect attribution data to your existing analytics stack so every user event carries channel context.
### You Also Get
- **Custom domain tracking links** -- branded deep links for campaigns
- **Attribution rules:** configure [attribution windows](https://www.airbridge.io/glossary/attribution-window) and priority
- **SKAdNetwork** **conversion value settings** -- iOS attribution configured from day one
- **Audience creation and export** -- build segments, push them back to ad channels
### Side by Side: What You're Choosing
| | Traditional MMP | Airbridge Core Plan |
| --- | --- | --- |
| Ad networks | 1,000+ | 4 GMAT channels |
| Event setup | Custom schema design | 25 standard subscription events |
| Setup time | 2+ weeks | Hours |
| Contract | Annual commitment | Pay-as-you-go |
| Reports | 15+ report types | 6 subscription-focused reports |
| Third-party integrations | Unlimited | 2 active integrations |
Airbridge Core Plan gives you less. That's the point. Less infrastructure to configure, fewer decisions to make, faster path to the signal you actually need.
## Airbridge AI: Support That Doesn't Make You Wait
Traditional MMPs hand you a dashboard and a help center link. When you're stuck, you open a ticket and wait. Airbridge built three AI tools so you don't have to.
[Airbridge Pilot 2.webp][Airbridge Pilot 3.webp]
### Airbridge Pilot: Instant Answers, Any Time
An AI assistant inside the Airbridge dashboard. Ask about tracking links, reports, attribution rules, SKAdNetwork, and get an answer immediately in your language. **Multi-turn conversations**, no question limits, available on **all plans** including Airbridge Core Plan.
[Onboarding Pilot 2.webp][Onboarding Pilot 3.webp][Onboarding Pilot 4.webp]
### Onboarding Pilot: SDK Setup Without the Engineering Bottleneck
AI-guided SDK installation that walks you through platform selection, step-by-step setup, and **real-time log verification** on iOS, Android, React Native, Flutter, or Expo. Choose MCP-assisted installation (via Cursor or Claude Code) or manual setup. Three activation milestones are checked automatically so you know the moment you're live.
[Airbridge MCP 1.webp][Airbridge MCP 4.webp][Airbridge MCP 5.webp]
### Airbridge MCP: Query Your Data from AI Tools
Connect Claude or Cursor to Airbridge and query attribution data in natural language. Ask "What was my Meta Ads CPS last week vs Google?" and get the answer in seconds. No dashboard navigation required. Strict security: **no data storage, no LLM training, no sharing**.
## What It Costs: $0 to Start
Traditional MMP pricing assumes you know your scale and you're ready to commit. Airbridge Core Plan assumes you're still figuring it out.
**Free tier:**
- **15,000 attributed installs** -- no cost, no credit card
**After free tier:**
- **$0.05 per attributed install** -- billed monthly, based on actual usage
No annual contracts. No minimum commitments. If your install volume drops next month, your bill drops with it.
[diagram-pricing-comparison.webp]
| Monthly Installs | Monthly Cost |
| --- | --- |
| 10,000 | $0 (within free tier) |
| 20,000 | $250 |
| 50,000 | $1,750 |
| 100,000 | $4,250 |
At 20,000 installs per month, Airbridge Core Plan costs **$3,000 per year**. A traditional MMP contract for the same volume starts at **$10,000--$30,000+ per year**. For a team spending $5K--$10K monthly on ads, that difference is another month of campaign budget.
## Who Airbridge Core Plan Is For
- **Stage:** Pre-seed to Series A, team of 1--10 people
- **Business model:** Subscription (free trial to paid, or freemium to paid)
- **App categories:** Health & Fitness, Education, Productivity, Lifestyle, Utility
- **Ad channels:** Primarily Meta and Google, expanding to TikTok and Apple Search Ads
- **Current pain:** "I'm spending on ads but I can't see which channel drives subscribers"
If that's you, Airbridge Core Plan was built for this exact moment in your growth.
**When you'll outgrow Core Plan:** When you need custom events, more than 2 third-party integrations, non-GMAT ad networks, raw data exports, fraud detection, or agency access. At that point, Airbridge Growth Plan is the next step, and the upgrade path is straightforward.
## FAQ with Airbridge Core Plan
### Can I use Core Plan if I only run ads on Meta and Google?
Yes, and that's the most common starting point. Most early-stage subscription apps begin with Meta and Google before expanding to TikTok and Apple Search Ads. Core Plan covers all four, so you won't need to switch tools as you add channels.
### Does Core Plan support iOS attribution?
Yes. Core Plan includes SKAdNetwork conversion value settings for Meta, Google, and TikTok. iOS attribution works from day one. For deterministic attribution, Airbridge's standard SDK handles both iOS and Android.
### What's the difference between Core Plan and GA4?
GA4 provides web and app analytics but is not built for subscription revenue attribution. It can't connect ad spend to subscription events at the channel level. Core Plan does, with direct GMAT integrations and RevenueCat/Adapty/Superwall support.
## Your Ad Budget Deserves Better Than Guesswork
Right now, your campaigns are running. Users are installing. Some are subscribing. But without attribution connecting those dots, you're making budget decisions in the dark -- and every dollar misallocated to the wrong channel is a subscriber you didn't acquire.
Airbridge Core Plan exists so that doesn't have to be your reality anymore.
**Start Growing with Airbridge Core Plan. Your first 15,000 attributed installs are free.**
---
## MMP Pricing Goes Up. Support Quality Goes Down. Something Has to Give.
URL: https://www.airbridge.io/en/blog/mmp-pricing-goes-up-support-quality-goes-down
Language: English
Published: 2026-03-26
Author: Jaehyuk Kim
A fitness app running 50,000 monthly installs across Meta and Google Ads signs up for an enterprise [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) on what looks like a reasonable plan. Six months later, the attribution bill has doubled. Per-conversion fees scaled with volume. Creative analytics required an upgrade. Revenue attribution was locked behind a premium tier.
**For subscription apps spending $5,000 to $20,000 per month on paid acquisition, MMP costs that quietly climb to $3,500 to $7,000 per month represent a direct hit to unit economics.** The MMP that was supposed to help you measure ROI is now actively eroding it.
**Key Takeaways**
- **MMP per-conversion fees scale faster than subscription revenue.** A $0.05 to $0.07 per-install fee compounds quickly when trial-to-paid conversion rates sit at 35 to 40%, meaning most installs never generate revenue.
- **Feature gates force upgrades at the worst time.** Creative analytics, cost aggregation, and revenue attribution are often locked behind premium tiers, so teams discover the real price only after committing.
- **Health and fitness apps face seasonal cost spikes.** January resolution surges can double MMP bills at the exact moment ad budgets are already elevated.
- **Hidden operational costs do not appear on the invoice.** Engineering time lost to poor documentation, missing support, and complex integrations adds weeks of delay.
- **Airbridge Core Plan uses transparent pay-as-you-go pricing.** 15K free attributed installs, $0.05 per install after, with subscription analytics included in the base offering.
## The Real Cost of MMP Pricing for Subscription Apps
MMP pricing models were designed for gaming and e-commerce apps with high average revenue per user (ARPU) and immediate purchase events. Subscription apps operate differently. Revenue comes weeks or months after the install, conversion rates are lower, and the metrics that matter most (trial-to-paid rate, [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv), churn by channel) require analytics that enterprise MMPs often reserve for higher tiers.
This structural mismatch means subscription app teams consistently pay more than they expected, and more than they should.
[Four layers of hidden MMP costs for subscription apps]
### 1\. Per-Conversion Fees That Scale Against You
Enterprise MMPs typically charge $0.05 to $0.07 per attributed conversion. For a subscription app at 50,000 monthly installs, that translates to $30,000 to $42,000 per year in attribution fees alone.
The problem is what happens on the revenue side. According to RevenueCat's State of Subscription Apps report, the median trial-to-paid conversion rate across subscription apps is 34.8% ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). For health and fitness apps specifically, only about 2.7% of downloads convert to paid subscribers within 35 days.
That means the vast majority of installs you pay to attribute never generate subscription revenue. The MMP fee is a fixed cost per install, but the revenue per install is a fraction of that fee for most users.
| Cost Component | 50K Monthly Installs | 100K Monthly Installs |
| --- | --- | --- |
| MMP fees at $0.07/install | $42,000/year | $84,000/year |
| Median D60 revenue per install (H&F) | $0.63 | $0.63 |
| Annual revenue from those installs | ~$378,000 | ~$756,000 |
| MMP cost as % of revenue | ~11% | ~11% |
When your attribution tool costs 10% or more of the revenue it measures, the ROI math stops working for early-stage teams.
### 2\. Feature Gates and Forced Upgrades
The base price is rarely the final price. Enterprise MMPs gate the features subscription app teams need most behind premium tiers or add-on packages:
- **Creative analytics** (which ads drive subscribers, not just installs) often requires an advanced tier (e.g., AppsFlyer locks Creative Optimization behind the ROI360 tier)
- **Cost data aggregation** across channels may require a separate premium product
- **Revenue attribution** by channel requires additional integrations or higher-tier access
- **Customer support and dedicated CSM** are stripped from lower-priced plans
The pattern is consistent: teams sign up for what appears to be an affordable entry plan, then discover that the metrics they actually need for [subscription revenue optimization](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025) require an upgrade that doubles or triples the cost.
This forced upgrade path creates a lock-in effect. By the time a team realizes the true cost, they have already invested engineering time in SDK integration, event configuration, and dashboard setup.
### 3\. The Compounding Cost Problem for Health and Fitness Apps
Health and fitness subscription apps face a unique version of this problem. [Subscription churn in fitness apps](https://www.airbridge.io/en/blog/why-subscription-churn-happens-top-5-cancellation-reasons-forecast-for-mobile-apps-in-2026) follows seasonal patterns tied to New Year resolutions, summer preparation, and back-to-school cycles.
January and February typically bring 2 to 3x install volume spikes as users sign up for fitness goals. Per-conversion MMP pricing means the attribution bill surges at the same time as:
- Ad spend increases to capture seasonal demand
- Trial volume rises, but trial-to-paid conversion rates often drop during high-volume periods
- Engineering and marketing teams are stretched thin
The result is a triple cost burden: higher ad spend + higher MMP fees + lower conversion quality. According to industry data, Health and Fitness apps have a median [Cost per Install (CPI)](https://www.airbridge.io/en/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics) of $5.78 on iOS ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Adding $0.07 per install in MMP fees pushes the effective CPI to $5.85. At a per-install level, 1.2% seems minor, but for a fitness app running a January campaign that generates 100,000 installs, the MMP fee alone for that month is $7,000. For a team already operating on thin margins between [Cost per Subscription and LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv), that margin erosion adds up fast.
## How to Reduce MMP Costs Without Losing Attribution
### 1\. Audit Your Current MMP Spend
Before making any tool switch, quantify what you are actually paying:
- **Calculate your true cost per attributed install.** Include base fees, per-conversion charges, add-on products, and overage penalties. Most teams underestimate by 30 to 50% because they only count the base subscription.
- **Map which features you use versus which features you pay for.** Many subscription apps pay for fraud detection, raw data export, and dozens of ad network integrations they never use.
- **Measure your MMP cost as a percentage of attributed revenue.** If it exceeds 5% of the subscription revenue your MMP helps measure, the ROI case weakens.
- **Track engineering time spent on MMP maintenance.** SDK updates, event schema changes, and debugging attribution discrepancies consume hours that do not appear on the MMP invoice.
### 2\. How Airbridge Core Plan Eliminates MMP Pricing Surprises
Airbridge Core Plan was built specifically for early-stage subscription app teams who need clear [attribution signals](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) without enterprise pricing complexity. The pricing model directly addresses the three cost traps described above.
**Transparent pricing with no feature gates:**
- **15,000 free attributed installs** per month, covering early-stage teams entirely
- **$0.05 per install after the free tier**, with no annual contract
- **All subscription analytics included in the base offering**: funnel reports, retention analysis, revenue attribution, active user tracking
- **No premium tiers for creative data or cost aggregation**
**Subscription-optimized feature set:**
| Capability | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free |
| Subscription funnel tracking | Available (may require add-on) | Built-in (Install, Start Trial, Subscribe) |
| Revenue attribution | Higher-tier feature | Included in base |
| Billing platform integration | Native (included in higher tiers) | Native (RevenueCat, Adapty in base) |
| Creative analytics | Locked behind premium tier | Included in base |
| Ad channel coverage | All networks | GMAT: Meta, Google, Apple Search Ads, TikTok |
| Third-party integrations | Unlimited | Max 2 (e.g., Amplitude + RevenueCat) |
| Custom events | Yes | No (25 standard subscription events by design) |
Core Plan supports 25 standard events optimized for the subscription funnel, including Start Trial, Subscribe, and Unsubscribe. Custom events are intentionally excluded to keep setup focused and fast. For teams that need custom events, additional ad networks, or raw data export, [Airbridge Growth Plan](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026) provides the full analytics stack.
The four Self-Attributing Networks (SANs) supported by Core Plan (Meta, Google, Apple Search Ads, TikTok) typically represent 80 to 90% of early-stage paid acquisition spend. Combined with [RevenueCat integration](https://www.airbridge.io/en/blog/abr-revenuecat-integration), this provides a complete view of which channels drive subscribers, not just installs.
[diagram-cost-comparison.webp]
**How the cost math changes with Core Plan:**
A subscription app running 50,000 monthly installs on Core Plan would pay $1,750 per month, or $21,000 per year. Compare that to $42,000 to $84,000 per year with an enterprise MMP at $0.07 per install with no free tier. The savings cover an additional month of ad spend on Meta or Google, and early-stage teams can validate their paid acquisition strategy with full attribution visibility before spending anything on measurement.
## FAQ: MMP Pricing and Hidden Costs
### How do I calculate the true cost of my current MMP?
Add your base subscription, per-conversion fees, add-on products (fraud detection, cost data, premium support), and estimated engineering hours for setup and maintenance. Most teams find the true annual cost is 1.5 to 2x the quoted base price.
### When should a subscription app switch MMPs?
Consider switching when your MMP cost exceeds 5% of the subscription revenue it helps measure, when you are paying for features designed for gaming or e-commerce that you do not use, or when essential subscription metrics (trial-to-paid by channel, renewal attribution) require a tier upgrade you cannot justify.
### What subscription analytics does Core Plan include that enterprise MMPs charge extra for?
Core Plan includes funnel analysis (install to trial to subscription conversion), retention reporting by acquisition channel, and revenue attribution. These reports are in the base offering. Enterprise MMPs may require premium tiers or add-on products for equivalent subscription analytics.
## Stop Paying More for Attribution Than You Spend on Growing
Every dollar that goes to hidden MMP fees is a dollar not spent on the campaigns that actually drive subscribers. For subscription app teams, attribution should clarify where to invest, not become another line item that erodes margins.
[**Start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core) and see which channels drive your subscriptions, starting with 15K free attributed installs and no annual contract.
---
## Influencer Marketing ROI for Fitness Apps: 400K Views, Zero App Subscribers?
URL: https://www.airbridge.io/en/blog/influencer-marketing-roi-fitness-apps
Language: English
Published: 2026-03-25
Author: Jaehyuk Kim
An influencer posts a 60-second TikTok featuring your fitness app. The video pulls 400,000 views, 12,000 likes, and 3,200 link clicks. Your team celebrates. Then someone asks the question nobody can answer: how many of those clicks turned into paying subscribers?
**For subscription apps, the gap between influencer engagement and actual subscription revenue is where most marketing budgets go to waste.** Measuring influencer marketing ROI remains the most commonly cited challenge in industry surveys. The reason is straightforward: influencer platforms report likes and clicks. Billing platforms report revenue. No system connects the two.
**Key Takeaways**
- **Likes, views, and clicks do not predict subscription revenue.** These vanity metrics measure audience interest, not purchase intent or willingness to pay monthly.
- **The two numbers that matter are Cost per Subscription (CPS) and D30 ROAS per influencer.** These separate revenue-driving creators from engagement-only performers.
- **UTM links and coupon codes break for app subscriptions.** UTMs are stripped at the app store boundary, and coupon codes capture only a fraction of influencer-driven conversions.
- **Influencer ROI is bimodal, not average.** Benchmarks like "$5.20 per $1 spent" hide that some influencers drive strong trial-to-paid conversion while others drive near-zero.
- **Airbridge Core Plan connects influencer tracking links to subscription events**, giving teams per-influencer CPS and ROAS visibility through RevenueCat or Adapty S2S integration.
## The Vanity Metric Trap: Why Likes Tell You Nothing About Influencer Marketing ROI
Growth teams running influencer campaigns for fitness and health apps face a fundamental measurement problem. The metrics that influencer platforms surface describe audience reaction. They do not describe revenue outcomes.
### 1\. What Vanity Metrics Actually Measure, and What They Miss
Influencer platforms are designed to measure content performance, not business performance. Views tell you how many people saw the content. Likes tell you how many reacted positively. Clicks tell you how many tapped a link.
None of these answer the question subscription apps need answered: **did this influencer drive users who converted to a paid subscription?** The disconnect is structural. Influencer platforms have no visibility into what happens after the click. They cannot see app installs, trial starts, or subscription activations.
### 2\. The Engagement-to-Revenue Gap in Subscription Apps
For e-commerce, the path from click to purchase is short: link, product page, buy. Attribution is straightforward. Fitness apps have a fundamentally different conversion path:
1. User sees influencer content
2. User clicks a link or searches the App Store
3. User installs the app
4. User starts a free trial
5. User converts to a paid subscription (often 7-14 days later)
Each step introduces friction and delay. By the time a user subscribes, the connection to the original influencer content has been severed. This is why e-commerce influencer marketing ROI benchmarks like [$5.20-$5.78 per $1 spent](https://www.shopify.com/enterprise/blog/roi-influencer-marketing) do not translate to subscription apps.
### 3\. Why Most Marketers Still Cannot Measure Influencer ROI
Across multiple industry surveys, ROI measurement consistently ranks as the single biggest challenge in influencer marketing. The tools, the budgets, and the creator partnerships are all in place. What is missing is the infrastructure to connect influencer spend to subscription revenue.
The root cause is infrastructure, not effort. Influencer platforms report engagement. App stores report installs. Billing platforms like RevenueCat and Adapty report revenue. **No single system connects all three by default.**
Teams that attempt to bridge this gap manually, typically with spreadsheets and coupon codes, capture only a partial picture.
[__wf_reserved_inherit]
## What Influencer ROI Actually Looks Like for Fitness Apps
If your team evaluates influencers by follower count, engagement rate, or cost per click, you are optimizing for the wrong signal. For fitness apps, influencer marketing ROI must be measured at the revenue layer.
### 1\. CPS and D30 ROAS: The Two Numbers That Decide Influencer Investment
**Cost per Subscription (CPS)** is total influencer spend divided by paid subscriptions attributed to that influencer. **D30 ROAS** is subscription revenue generated within 30 days of install, divided by influencer spend. Without these two numbers, teams default to engagement proxies that tell nothing about who drives paying subscribers.
Consider two influencers promoting the same subscription fitness app. This example illustrates why measuring influencer cost per subscription for a mobile app matters more than any engagement metric:
[__wf_reserved_inherit]
Influencer A wins on every engagement metric. But Influencer B produces more than twice as many paying subscribers at less than a quarter of the cost. **If your measurement stops at the install or trial level, you would scale spend on Influencer A and cut Influencer B.**
### 2\. Why Influencer ROI Is Bimodal, Not Average
E-commerce brands with strong attribution [see 6-10x returns](https://www.dataslayer.ai/blog/influencer-marketing-budgets-surge-in-2025-how-to-track-roi-with-data-automation) on influencer spend. But this average obscures a bimodal distribution. For subscription apps, the pattern is even more pronounced.
A fitness influencer can drive thousands of installs from users who start a 7-day workout trial and cancel before the first billing cycle. Without per-influencer subscription data, high-churn influencers look identical to high-revenue influencers in every dashboard except the one that tracks revenue.
## Why Standard Tracking Methods Break for Fitness App Subscriptions
Standard tracking methods used in e-commerce influencer campaigns fail at the app store boundary.
### 1\. UTM Links Lose the User at the App Store
UTM parameters work by appending tracking data to a URL. When a user clicks a UTM-tagged link, the destination page reads those parameters and records the traffic source. For apps, this chain breaks:
1. User clicks the influencer's UTM-tagged link
2. Link redirects to the App Store or Google Play
3. **The app store strips the UTM parameters**
4. User installs the app
5. The app has no record of the influencer source
The install appears as organic. The influencer receives no attribution credit. If the user later subscribes, that subscription revenue is invisible to the measurement system.
[__wf_reserved_inherit]
### 2\. Coupon Codes Capture Only a Fraction of Conversions
Coupon codes are the most common workaround for influencer attribution. The influencer shares a code ("Use code SARAH20 for 20% off"), and the team counts redemptions per code.
The problem: not every user who converts from influencer content uses the code. Coupon codes capture only a fraction of influencer-driven conversions because many users see the content, search the App Store directly, install, and eventually subscribe without ever entering a code. These users are attributed as organic.
This undercount creates a specific problem for budget allocation. The influencer appears less effective than they actually are, and organic acquisition appears more effective. Teams reduce influencer spend based on incomplete data while "organic" numbers that are actually influencer-driven inflate the baseline.
### 3\. The Install-to-Subscription Gap
Even when a team attributes an install to an influencer, the attribution chain often stops there. The subscription event happens days later in a billing platform like RevenueCat or Adapty that has no record of the influencer source. **The two systems operate independently, and the connection between influencer source and subscription revenue is lost.**
Solving this requires a [mobile measurement partner](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) that connects influencer tracking links to post-install subscription events through server-to-server (S2S) integration with the billing platform.
[**_Already losing attribution on paid ad channels too? See how teams are tracking free trial to paid conversion by channel to fix the same visibility gap._**](https://www.airbridge.io/en/plans/core)
## How to Build an Influencer ROI Measurement Stack for Your Fitness App
Measuring influencer ROI for fitness and health apps requires three components working together.
### 1\. Tracking Links with Deep Link Capability
Each influencer receives a unique tracking link that survives the app store redirect, maintaining attribution through the install flow. Unlike standard UTM links, deep link-enabled tracking links preserve the influencer source even when users land in the App Store or Google Play. This replaces the broken UTM approach with link technology designed for app install flows.
### 2\. S2S Integration with Billing Platforms
Server-to-server integration with RevenueCat or Adapty captures subscription lifecycle events (Start Trial, Subscribe, Unsubscribe) regardless of app state. S2S enables real-time, independent signal transmission that does not rely on the user manually opening the app. This closes the install-to-subscription gap by creating a continuous data connection from influencer click to subscription revenue.
### 3\. Per-Influencer Revenue Reports
With both components in place, the attribution system generates per-influencer funnels (Install to Start Trial to Subscribe), CPS by influencer, and D30 ROAS by influencer. These transform evaluation from "who gets the most likes" to "who drives the most subscription revenue per dollar spent."
Airbridge Core Plan provides all three components. Custom domain tracking links per influencer maintain attribution through the app store redirect. S2S integration with RevenueCat or Adapty captures 25 standard subscription events, including Start Trial, Subscribe, and Unsubscribe.
Standard events use predefined event names. Customers still need to log these events in their app code, but event naming and parameter structure is predefined by Airbridge, reducing schema design work. Most MMPs support RevenueCat and Adapty. Core Plan includes these from day one at no extra tier.
The Funnel report visualizes the full Install to Start Trial to Subscribe path per influencer source. The Revenue report shows subscription revenue attributed to each influencer's tracking link.
Teams see per-influencer CPS and D30 ROAS directly, enabling data-driven decisions about which partnerships to scale or end. If you are already tracking free trial to paid conversion by paid ad channel, adding influencer sources to the same attribution system is the natural next step.
Core Plan is the attribution layer, not an influencer management tool. It does not support custom events, fraud detection, or raw data export. These limitations are intentional.
Pricing: 15,000 free attributed installs, $0.05 per install after, no annual contract.
## The Influencer Budget You Cannot Attribute Is the Influencer Budget You Cannot Scale
Every growth team wants to invest more in influencers that drive revenue and less in influencers that drive only engagement. The barrier is not willingness. It is visibility.
When your measurement stops at likes, views, and installs, you make investment decisions with incomplete data. The influencer who generates 400,000 views and 22 paid subscriptions looks better than the one who generates 28,000 views and 52 paid subscriptions, until you see the subscription data.
[**_Get Started with Airbridge Core Plan_** **_to connect influencer tracking links to subscription revenue and see per-influencer CPS and ROAS in your first week._**](https://www.airbridge.io/en/plans/core)
For teams evaluating the broader impact of influencer marketing on CAC, per-influencer subscription attribution is the foundation. And if your attribution numbers don't match between influencer platforms and internal data, proper attribution infrastructure is the fix.
---
## What Manual Influencer Campaign Management Costs Your Fitness App
URL: https://www.airbridge.io/en/blog/influencer-campaign-management-cost-fitness-app
Language: English
Published: 2026-03-25
Author: Airbridge
A fitness app partners with 15 influencers across Instagram and TikTok. Each influencer gets a unique promo code, a UTM link, and a shared Google Sheet row. By week two, three links are broken, two influencers posted with the wrong code, and the spreadsheet has duplicate entries that no one can reconcile.
The campaign report says 4,200 installs. RevenueCat says 89 paid subscribers. No one can connect the two numbers.
**This gap between install counts and subscription revenue is not a reporting inconvenience. It is the single largest source of wasted influencer spend for subscription apps.** When you cannot attribute a paying subscriber back to the influencer who drove the install, every partnership decision is a guess. Renewing a $5,000/month creator because "the engagement looked good" is how growth budgets disappear.**Key Takeaways**
- **60% of brands still rely on spreadsheets and manual processes for influencer campaign management.** Link creation, data entry, and reporting consume entire workdays at scale. ([InfluenceFlow](https://influenceflow.io/resources/influencer-campaign-software-the-complete-2025-guide-to-finding-and-running-winning-campaigns/))
- **Manual influencer tracking costs $45K to $65K per year in labor alone.** This does not include the cost of misattributed spend or missed optimization windows. ([InfluenceFlow](https://influenceflow.io/resources/analytics-dashboards-for-tracking-influencer-performance-the-complete-2026-guide/))
- **Influencer campaign management tools reduce tracking overhead by 60 to 75%.** But most tools stop at vanity metrics and cannot connect installs to subscription conversions. ([InfluenceFlow](https://influenceflow.io/resources/analytics-dashboards-for-tracking-influencer-performance-the-complete-2026-guide/))
- **Many brands struggle to attribute influencer-driven revenue to specific creators or campaigns.** Install data and subscription revenue live in separate systems with no connection between them.
- **Airbridge Core Plan bridges the attribution gap with custom tracking links, 25 standard subscription events, and RevenueCat integration** starting at 15,000 free attributed installs.
## Why Influencer Campaign Management Drowns Fitness App Teams in Spreadsheets
Influencer campaign management looks simple on paper: find fitness creators, negotiate terms, distribute links, track results. In practice, the tracking step alone consumes more time than every other step combined.
### 1\. The Link-Creation-to-Reporting Loop
Every influencer partnership generates a chain of manual work:
1. Create a unique tracking link or promo code
2. Send the link to the influencer with posting instructions
3. Log the link, influencer name, platform, and posting date in a spreadsheet
4. Monitor for the post going live
5. Pull platform analytics (Instagram Insights, TikTok Analytics) into the spreadsheet
6. Cross-reference with app store install data
7. Attempt to match installs to subscription events in a billing platform
Each step requires a different tool. None of them talk to each other. A team managing 10 influencers repeats this loop 10 times. A team managing 50 influencers either hires someone full-time to manage the spreadsheet or accepts that the data will be incomplete.
### 2\. What Manual Campaign Time Actually Looks Like
Industry data shows that [60% of brands still struggle to manage campaigns efficiently, relying on spreadsheets and manual processes](https://influenceflow.io/resources/influencer-campaign-software-the-complete-2025-guide-to-finding-and-running-winning-campaigns/) for influencer campaign management. That ratio holds across team sizes, but the absolute cost scales dramatically.
A two-person growth team managing 10 partnerships loses one full workday per week to spreadsheet maintenance. A five-person team managing 40 partnerships may lose two to three full-time equivalent days per week.
The work is not difficult. It is relentless. Every hour spent copying UTM parameters into a spreadsheet is an hour not spent analyzing which influencers actually drive paying subscribers.
[__wf_reserved_inherit]
## The Hidden Cost of Manual Influencer Tracking: $45K to $65K per Year
The spreadsheet problem is not just an operational headache. It has a specific dollar cost that most teams never calculate.
### 1\. Where the Money Goes
Estimates place the annual cost of manual influencer tracking labor between [$45,000 and $65,000 per brand](https://influenceflow.io/resources/analytics-dashboards-for-tracking-influencer-performance-the-complete-2026-guide/). This figure covers everything from link creation and distribution to cross-platform data aggregation, weekly report generation, and error correction across broken links and conflicting numbers.
These costs assume a mid-market team running 10 to 30 influencer partnerships per quarter, and they do not account for the opportunity cost of misattributed spend or delayed optimization decisions that compound over each campaign cycle.
- **Link management:** Creating, distributing, and verifying unique tracking links for each influencer and each campaign
- **Data aggregation:** Pulling metrics from Instagram, TikTok, YouTube, and other platforms into a central spreadsheet
- **Report generation:** Building weekly or monthly performance reports that require cross-referencing multiple data sources
- **Error correction:** Fixing broken links, resolving duplicate entries, and reconciling conflicting numbers between platforms
### 2\. Why Spreadsheets Break as Partnerships Scale
Spreadsheets work for a handful of partnerships. They stop working reliably as the roster grows. The failure mode is not dramatic. It is gradual:
- **Naming inconsistencies** creep in. One row says "TikTok\_JaneSmith\_Feb" and another says "jane\_smith\_tiktok\_0215."
- **Stale data** accumulates. Someone forgets to update a row for two weeks, and the spreadsheet silently becomes unreliable.
- **Version conflicts** multiply. Three people edit the same Google Sheet, and a formula breaks without anyone noticing.
The result is a tracking system that feels functional but produces numbers no one fully trusts. When your campaign performance data is unreliable, you cannot make confident decisions about which influencers to renew, which platforms to prioritize, or how to allocate next quarter's budget.
## Tools vs. Spreadsheets: What Influencer Campaign Management Software Actually Solves
Influencer campaign management tools exist precisely to replace the spreadsheet workflow. Platforms like CreatorIQ, Grin, and Aspire handle content approval, contract management, communication, and basic performance tracking in a single interface.
### 1\. Content, Contract, and Communication
The operational layer is where these tools deliver clear value. Instead of juggling email threads, Slack messages, and shared drives, teams can manage influencer relationships, review content drafts, and track deliverables in one place. For teams drowning in coordination overhead, this alone justifies the investment.
### 2\. 60 to 75% Tracking Overhead Reduction
Dedicated influencer campaign management tracking automation tools can [reduce tracking overhead by 60 to 75%](https://influenceflow.io/resources/analytics-dashboards-for-tracking-influencer-performance-the-complete-2026-guide/) compared to manual spreadsheet workflows. Automated link generation, centralized dashboards, and platform API integrations eliminate the most time-consuming parts of the reporting loop.
### 3\. What Tools Still Do Not Solve: Attribution
Here is where the category falls short. Most influencer campaign management tools track engagement metrics: likes, comments, shares, clicks, and installs. Some integrate with app stores to show install counts.
But **install counts are not revenue.** A fitness app needs to know which influencer drove installs that converted to paid subscribers. That requires connecting platform engagement data, [mobile app attribution](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution), and billing platform revenue into a single view.
Influencer management tools handle the first layer. They do not touch the second or third. This creates a gap where teams can see that an influencer "drove 500 installs" but cannot answer whether those installs produced $2,000 or $20,000 in subscription revenue.
## The Deeper Problem: Why Influencer Revenue Goes Unattributed
The attribution gap is not a minor data quality issue. It is a structural problem that affects the majority of brands running influencer programs.
### 1\. Why Install Counts Mislead
Many brands running influencer programs struggle to attribute influencer-driven revenue back to specific creators or campaigns. The installs show up in app store dashboards, and revenue shows up in billing platforms, but the line connecting a specific influencer link to a specific paying subscriber is missing. This disconnect means growth teams are making renewal and budget decisions based on vanity metrics rather than actual revenue contribution.
### 2\. The Attribution Gap Between Platforms and Billing
The gap is architectural, not operational. Install attribution and revenue attribution live in entirely different systems, and no single platform in the typical influencer technology stack connects the click on an influencer link to the subscription payment that happens days or weeks later. The table below illustrates where each system's visibility ends and the data blind spot begins:
[__wf_reserved_inherit]
No amount of spreadsheet work bridges this gap. The data simply does not exist in any single system. Closing this gap requires a [Mobile Measurement Partner (MMP)](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) that can attribute installs to specific links and then connect those installs to downstream subscription events.
[__wf_reserved_inherit]
[**_Airbridge Core Plan includes custom tracking links, subscription attribution, and RevenueCat integration in one stack. Get Started with Airbridge Core Plan →_**](https://www.airbridge.io/en/plans/core)
## How Tracking Links and Attribution Close the Influencer Revenue Gap
The solution to the influencer attribution problem is not a better spreadsheet or a more expensive influencer platform. It is adding an attribution layer that connects influencer-driven installs to subscription revenue.
### 1\. Custom Domain Tracking Links
The foundation is a unique tracking link per influencer. Instead of relying on UTM parameters that break across redirects, custom domain tracking links maintain attribution through the entire install flow from click to app open. Each influencer gets a branded link (e.g., yourbrand.link/jane) that carries attribution data from click to install, persisting even when the user does not have the app installed at the time of the click. This eliminates the manual link-creation-and-spreadsheet-logging loop entirely; links are generated centrally, each one tied to a specific influencer and campaign, and attribution happens automatically when the user installs the app through [deep linking](https://www.airbridge.io/glossary/deep-linking).
### 2\. From Install to Subscription with Standard Events
Tracking the install is only the first step. To connect influencer installs to revenue, you need event-level attribution that follows the user through the subscription funnel: install, sign-up, start trial, subscribe, renew, or unsubscribe.
Standard events use predefined event names (Start Trial, Subscribe, Unsubscribe, and others) that map directly to the subscription lifecycle. Customers still need to log these events in their app code, but the event names and parameters are predefined, reducing schema design work. When paired with a server-to-server (S2S) integration with billing platforms like [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) or Adapty, subscription events are captured regardless of whether the user has the app open.
### 3\. What This Looks Like with Airbridge Core Plan
Airbridge Core Plan provides the tracking link and attribution layer that influencer platforms lack:
- **Custom domain tracking links** for each influencer, with [deferred deep linking](https://www.airbridge.io/en/blog/deferred-deeplink-benefits) that maintains attribution even when the app is not yet installed
- **25 standard subscription events** including Start Trial, Subscribe, and Unsubscribe, covering the full subscription lifecycle
- **RevenueCat and Adapty S2S integration** that captures revenue events in real time, independent of app state
- **Funnel, Retention, and Revenue reports** that show which influencers drive installs that convert to paying subscribers
The pricing model removes the barrier that keeps most subscription apps from adopting proper attribution. Core Plan starts with **15,000 free attributed installs** and charges **$0.05 per install after that, with no annual contract.** For context, once startups outgrow free MMP tiers, paid contracts typically land in the [low-to-mid four figures per month with annual commitments](https://www.vendr.com/marketplace/appsflyer), which prices out most early-stage fitness apps running influencer campaigns.
What Core Plan intentionally does not include: custom events, non-SAN ad network support, raw data export, and fraud detection. These are features designed for mature teams with complex analytics needs. For fitness apps focused on influencer attribution, the 25 standard events and GMAT channel integrations (Google, Meta, Apple Search Ads, TikTok) cover the core measurement requirements.
## Revenue You Cannot Attribute Is Revenue You Cannot Scale
Every fitness app running influencer partnerships faces the same question: which creators actually drive paying subscribers? Spreadsheets cannot answer it. Influencer platforms alone cannot answer it. The answer requires an attribution layer that follows the user from the influencer link through the install, the trial, and the subscription.
[**_Get Started with Airbridge Core Plan — connect influencer installs to subscription revenue with custom tracking links, 25 standard events, and RevenueCat integration. 15,000 free attributed installs, no annual contract._**](https://www.airbridge.io/en/plans/core)
The gap between influencer installs and subscription revenue is not a data problem you solve later. It is the measurement foundation you build now, before scaling the partnerships that depend on it.
---
## Stop Searching for Complex Marketing Data. Just Ask with Airbridge AI
URL: https://www.airbridge.io/en/blog/airbridge-ai-mcp-pilot-launch
Language: English
Published: 2026-03-25
Author: Airbridge
Have you ever opened your dashboard to check campaign performance, only to spend five minutes searching for the right report?
Where do you check ROAS again? Which section shows retention? How did you set up that filter last time? These are questions most marketers ask themselves multiple times a day. And the time it takes to answer them adds up fast, **anywhere from five to fifteen minutes per session**. **More time spent pulling data than actually analyzing it.**
And that's just the data side. When it comes to onboarding, setting up an MMP means installing an SDK, a process that varies by platform, requires jumping between documentation and your development environment, and **often can't move forward without a developer**. It's the kind of friction that drains momentum before the real work even begins.
As AI reshapes the way we work, Airbridge has built three features designed to give marketers that time back. Introducing **Airbridge AI**, including Airbridge MCP, Airbridge Pilot, and Onboarding Pilot.
### **Airbridge AI Key Features**
- **Airbridge Pilot** — Get instant answers to your Airbridge questions right inside the dashboard
- **Onboarding Pilot** — Let AI walk you through SDK installation step by step
- **Airbridge MCP** — Query your Airbridge data in natural language directly from the AI tools you already use
[[Stop Clicking Through Dashboards. Just Ask Airbridge Pilot.]](https://www.youtube.com/watch?v=Hfm_MT0Y270)
## **Airbridge Pilot: An AI Agent That Answers Inside Your Dashboard**
"**No more opening the Help Center every time you have a question. Just ask AI and get your answer on the spot.**"
Questions come up constantly while you work.
_How does this Attribution rule actually work? Where do I configure SKAN settings? Why isn't the deep link open event registering?_
Every time you leave the dashboard to find an answer, opening the Help Center, searching, scanning through documentation, you break your workflow. For anyone new to an MMP, **the time spent navigating support docs can easily exceed the time spent doing actual work.**
**Airbridge Pilot eliminates that friction entirely**, keeping your workflow intact.
[You can use Airbridge Pilot whenever you need]
*You can use Airbridge Pilot whenever you need*
[Just click and ask questions to Airbridge Pilot]
*Just click and ask questions to Airbridge Pilot*
[Then, Airbridge Pilot will answer and show you marketing dashboard you look for!]
*Then, Airbridge Pilot will answer and show you marketing dashboard you look for!*
[If you have any further questions, just ask!]
*If you have any further questions, just ask!*
### **1\. How Airbridge Pilot Differs from a Standard Chatbot**
Airbridge Pilot is not a keyword-matching or menu-driven chatbot. It is a **conversational AI agent**, and you interact with it the same way you would with any AI chat tool, **asking questions in natural language and getting answers immediately**. Whether it is tracking links, reports, Attribution rules, or iOS SKAN, just click Pilot and ask.
It also supports **multi-turn conversation**, meaning it remembers the context of your previous messages and carries the thread forward.
"How do I attach a deep link to a tracking link?" → After receiving an answer → "And what happens on iOS if the app isn't installed?"
### **2\. Airbridge Pilot Key Features**
- **No usage limits** — Ask as many questions as you need, whenever you need to
- **Available on all plans** — Every Airbridge customer has access regardless of plan tier
- **Respond in your language** — Ask in Korean, get an answer in Korean. Ask in English, get an answer in English
- **Help Center-backed responses** — Answers are grounded in official documentation written directly by the Airbridge team
Note: Like any AI, Airbridge Pilot can occasionally produce inaccurate responses. For anything involving critical settings or policies, we recommend cross-checking with the official Airbridge Help Center.
[[Stop Struggling with SDK Setup - Install Airbridge SDK with AI (Onboarding Pilot Demo)]](https://www.youtube.com/watch?v=sTpfv4l5SY0)
## **Onboarding Pilot: SDK Installation Guided by AI**
"**No more fumbling through docs alone.**"
Every app marketing setup starts with SDK installation. And while it sounds straightforward, it rarely is.
Installation steps differ by platform. You have to cross-reference documentation while navigating your development environment. And if you need a developer to handle the technical side, **progress stalls until they are available**. It is one of the most common points where onboarding slows down or stops entirely.
**Onboarding Pilot guides you through the entire process via conversational AI.** If anything is unclear along the way, just ask. Instead of reading through documentation, you follow along with the assistant and by the end of the conversation, installation is complete.
[To start with Onboarding Pilot, you need to enter the email first.]
*To start with Onboarding Pilot, you need to enter the email first.*
[Select installation method. If you want, you can install with Airbridge MCP!(Recommneded)]
*Select installation method. If you want, you can install with Airbridge MCP!(Recommneded)*
[Just follow the instruction with Onboarding Pilot]
*Just follow the instruction with Onboarding Pilot*
[After installation, you can test SDK installation at your simulator right away!]
*After installation, you can test SDK installation at your simulator right away!*
### **How SDK Installation Works with Onboarding Pilot**
1. **Who's installing?** → Install yourself or invite a developer to join
2. **Which installation method?** → MCP-assisted or manual installation
3. **What's your environment?** → Select from Android, iOS, React Native, Flutter, or Expo and receive a tailored guide for your platform
4. **Did it work?** → Real-time log verification automatically confirms whether data is being received correctly
5. **Activation checklist** → Three key milestones are verified automatically:
- App install event received
- Tracking link created
- Deep link open event received
Onboarding Pilot handles guidance and verification, and it does not deploy code directly. **Final code implementation is always carried out by a developer.**
[[How to Use Airbridge MCP in Claude Code | Query Your Data with AI]](https://www.youtube.com/watch?v=REHnl1vIAVE)
[[How to connect your Airbridge MCP with Claude Code | Start using MMP with AI]](https://www.youtube.com/watch?v=4hW2xtHV8OY)
## **Airbridge MCP: Query Your Airbridge Data Directly from the AI Tools You Use**
"**No more opening the dashboard every single time.**"
A marketer's day is built on context-switching. You are writing a brief, you need a campaign metric, you jump to the dashboard, find the number, then navigate back. **The more this happens, the more your context breaks down and the slower your work gets.**
[If you want to try Airbridge MCP, you have to use custom conncetor for now.]
*If you want to try Airbridge MCP, you have to use custom conncetor for now.*
[Once you set up MCP, chat with your AI tools and pick Airbridge apps you desired.]
*Once you set up MCP, chat with your AI tools and pick Airbridge apps you desired.*
[Once you set up MCP, chat with your AI tools and pick Airbridge apps you desired.]
*Once you set up MCP, chat with your AI tools and pick Airbridge apps you desired.*
[Once you confirm it, you don't need to apporve again!]
*Once you confirm it, you don't need to apporve again!*
[Then, you will see the data on your AI tools right away.]
*Then, you will see the data on your AI tools right away.*
**Airbridge MCP** connects Airbridge to AI tools like Claude and ChatGPT, so you can **retrieve data through natural language without ever leaving your workspace**. Just ask, and the data comes to you.
"Show me last week's ROAS by channel." "How has retention trended this month?" "What were yesterday's Active Users for the iOS campaign?"
**No extra tabs. No digging through report menus.** Ask AI and get your answer immediately.
### **1\. What Can You Query with Airbridge MCP?**
The current beta version supports:
- Actuals, Retention, Revenue, and Active Users reports
- App settings
- Tracking link data
- OAuth authentication
- RBAC (Role-Based Access Control) integration
Additional capabilities, including tracking link creation and editing, audience queries, chart visualization, multi-app queries, and workflow automation, are planned for the **full release in April**.
### **2\. How to Connect Airbridge MCP**
On Claude Desktop, go to Settings → Connectors → Custom Connectors and enter the following:
- Name: Airbridge
- URL: https://mcp.airbridge.io/mcp
On Claude Code CLI:
- claude mcp add --transport http airbridge https://mcp.airbridge.io/mcp
For detailed setup instructions, visit the [Airbridge Help Center](https://help.airbridge.io/en/guides/mcp).
### **3\. Airbridge MCP Is Especially Useful If You:**
- Spend more of your day inside AI tools than inside a dashboard
- Regularly field data requests from others on your team
- Need to monitor multiple campaigns simultaneously
## **Enterprise-Grade Security Built Into Airbridge AI**
Data security is likely your first concern when adopting any AI feature. Every piece of data within Airbridge, including everything accessed through Airbridge MCP and Airbridge AI, is governed by **six core principles**:
1. No Storage: Queried data is never stored on our servers
2. No Training: Your data is never used to train AI models
3. No Sharing: Data is never shared with third parties
4. PII Blocking: Personal identifiers, including Device IDs, IPs, and emails, are blocked from reaching AI systems entirely
5. RBAC Inheritance: Your existing Airbridge account permissions apply in full to all AI features
6. Global Compliance: Fully compliant with GDPR, CCPA, and PIPA
## **Get Started with Airbridge AI Now**
We are moving into an era where **data no longer waits for you to find it. It comes to you.**
Query data instantly with Airbridge MCP. Get answers on the spot with Airbridge Pilot. Walk through SDK setup with Onboarding Pilot. You do not need to know the dashboard inside out, and you do not need to be familiar with every feature to get value right away. With Airbridge AI, even the most complex tasks can be handled quickly and efficiently.
**While your competitors are still building last week's report, you are already executing on your next move.** The teams that use data faster move faster. Start accelerating yours with Airbridge AI today.
## **FAQ with Airbridge AI**
### **Q. Which AI tools support Airbridge MCP?**
Any environment that supports MCP. Claude Desktop, Claude Code, Cursor, and Codex CLI are the most common. ChatGPT integration is also on the roadmap.
### **Q. Will my company's data be exposed to AI systems when I use MCP?**
No. Only the response to your query is returned. Data is never stored or used for training. PII access is blocked at a technical level.
### **Q. Which plans include Airbridge Pilot?**
All of them. Every Airbridge customer can use Airbridge Pilot with **no question limits**.
### **Q. Can I install the SDK without a developer using Onboarding Pilot?**
Onboarding Pilot guides the full process, but **final code implementation requires a developer**. Marketers can initiate the flow and bring in a developer when it is time to apply the code.
### **Q. What if Airbridge Pilot gives me an incorrect answer?**
It is rare, but it can happen. For anything involving important settings or policies, we recommend verifying with the official Airbridge Help Center.
---
## Which Ad Channel Drives Subscriptions? How to Attribute App Revenue to the Right Source
URL: https://www.airbridge.io/en/blog/app-subscription-channel-attribution
Language: English
Published: 2026-03-25
Author: Jaehyuk Kim
# Which Ad Channel Drives Subscriptions? How to Attribute App Revenue to the Right Source
Channel A delivers 2,000 installs. Channel B delivers 1,000. **If you are optimizing for installs, Channel A wins — and gets more budget next month.**
But when you trace those installs through to paid subscriptions, [the picture reverses](https://www.airbridge.io/en/blog/which-ad-campaign-drives-subscriptions-mmp-cant-answer). Channel A produced 24 subscribers at ~$83 each. **Channel B produced 65 subscribers at ~$15 each — 2.7x more paying subscribers at one-fifth the cost.**
**This is what happens when install attribution and subscription attribution live in different systems.** Most UA teams do not notice this reversal because their MMP is not optimised towards measuring channel performance based on subscription outcomes.
**Key Takeaways**
- **Install performance and subscription performance often invert by channel.** The channel with the most installs is not always the channel with the most paid subscribers.
- **Attribution windows shorter than your trial period create a structural blind spot.** For a 7-day trial, 100% of conversions happen at or after day 7. If attribution cannot span this gap, every subscription is unattributed.
- **Health & Fitness trial-to-paid median is 39.9%** — but without channel-level data, you cannot tell whether a given channel is closer to the bottom or the [top 10% reaching 68.3%](https://www.revenuecat.com/state-of-subscription-apps-2025/).
- **of Health & Fitness subscriptions are annual.** Not knowing which channel drives annual subscribers means not knowing where your highest-LTV users come from.
- **Airbridge Core Plan connects ad channels to subscription outcomes.** Native RevenueCat/Adapty S2S, Funnel and Revenue reports by channel. $0.05/install, 15K free.
## The Problem: Install Metrics Hide Subscription Performance
Every ad platform reports installs. Meta shows installs. Google shows installs. TikTok shows installs. They also report subscription conversions — but only within their own attribution and postback windows. **Conversions that fall outside those windows, like a subscription from a 7-day trial attributed to a click from 8 days ago, are simply not recorded.**
[__wf_reserved_inherit]
This creates a fundamental mismatch:
- **What you can see:** installs, CPI, click-through rate — by channel, by campaign, by creative
- **What you cannot see:** trial starts, trial-to-paid conversions, subscription revenue — by channel, by campaign, by creative
**The gap is not a reporting limitation. It is a structural one.** Subscription conversions happen server-side, after the attribution window closes, in a billing system that is not connected to your ad channels.
## Why Channel-Level Subscription Data Goes Missing
Three structural gaps prevent ad channels from connecting to subscription outcomes.
[__wf_reserved_inherit]
- **Attribution window < trial period.** For a 7-day free trial, the subscription conversion happens on day 7 or later. [If your attribution lookback window is shorter than your trial, the conversion falls outside the window](https://www.airbridge.io/en/blog/which-ad-campaign-drives-subscriptions-mmp-cant-answer) — and cannot be tied back to the original ad click.
- **Server-side billing.** Subscription events (Subscription, Renewal, Cancellation) are processed by Apple App Store and Google Play — not inside your app. SDK-based attribution captures app opens, but **subscription state changes happen on billing servers regardless of whether the user opens the app.**
- **Data silos.** Your billing platform (RevenueCat, Adapty) knows who subscribed. Your MMP knows which ad drove the install. But unless these two systems exchange data via server-to-server integration, **the connection between "ad click → install → trial → paid subscriber" is broken.**
The result: your MMP only shows events triggered by SDK attributed to channels. Your billing platform shows subscribers. **No single view shows subscribers by channel.**
## What This Costs You
Without channel-level subscription data, every strategic decision is compromised.
- **Budget allocation:** You push more budget to the channel with the most installs — which may be the channel with the fewest subscribers. **The Channel A vs Channel B reversal is invisible.**
- **Creative testing:** You can A/B test creatives for install rate. But you cannot test which creative drives subscribers who convert past the trial. Install-optimized creatives and subscription-optimized creatives are often different.
- **Persona targeting:** You cannot identify which audience segments convert to paid at higher rates — because conversion data is not connected to acquisition data.
- **Annual vs monthly mix:** With 57% of Health & Fitness subscriptions being annual, knowing which channel drives annual subscribers — your highest-LTV users — is critical. **Without channel-level data, this is invisible.**
[Industry benchmarks](https://www.revenuecat.com/state-of-subscription-apps-2025/) put the Health & Fitness trial-to-paid median at 39.9%, with the top 10% reaching 68.3%. **The gap between median and top performers is driven by channel mix, creative strategy, and audience targeting — but you can only optimize these if you can see subscription data by channel.**
## How to Connect Ad Channels to Subscription Outcomes
Closing the gap between install attribution and subscription attribution requires four things.
- **Billing-to-attribution integration.** Your billing platform (RevenueCat, Adapty) must send subscription events — Subscription, Renewal, Cancellation — directly to your MMP via a server-to-server connection. SDK alone cannot reliably capture server-side billing events.
- **Attribution window ≥ trial period.** Your MMP's attribution window must span the full trial-to-paid conversion cycle. For a 7-day trial, a 7-day window is the minimum. Longer trials or delayed conversions need longer windows.
- **Channel-level funnel visibility.** You need the full funnel — install → trial → paid → renewal — broken down by channel, campaign, and creative. Not just installs by channel and subscribers in aggregate.
- **Revenue attribution by channel.** Subscription revenue should map back to the acquiring channel. A channel that drives 10 annual subscribers is worth more than one that drives 20 monthly subscribers — but only if you can see revenue by channel, not just conversion counts.
[**_Connect ad channels to subscription revenue. Native RevenueCat/Adapty S2S. Funnel and Revenue reports by channel. $0.05/install, 15K free._**](https://www.airbridge.io/en/plans/core)
## How Core Plan Connects Channels to Subscriptions
Core Plan includes native S2S integration with [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) in the base offering — subscription events flow into attribution without additional tier upgrades.
With subscription data connected to acquisition data, the reports change:
- **Funnel reports by channel.** See install → trial → paid broken down by Meta, Google, Apple Search Ads, and TikTok. Identify which channel's installs actually convert — not just which channel has the lowest CPI.
- **Revenue reports by channel.** Map subscription revenue back to the acquiring channel and campaign. See whether Channel B's $15/subscriber generates annual or monthly subscribers.
- **Retention reports by cohort.** Track subscriber retention by acquisition channel. **A channel with high trial-to-paid but fast churn after month 3 looks different from one with moderate conversion but strong long-term retention.**
25 predefined subscription-optimized standard events — including Start Trial, Subscribe, and Unsubscribe — are already defined. The events that matter for channel-level subscription analysis do not require custom schema design.
### Core Plan vs Traditional MMP: Channel-Level Subscription Attribution
Seeing which ad channel drives subscriptions requires billing integration, funnel visibility, and revenue attribution. The difference is where these capabilities sit in the pricing structure.
[__wf_reserved_inherit]
## Your Install Data Is Not Your Subscription Data
Channel A gets the installs. Channel B gets the subscribers. **If you cannot see both in the same report, you are optimizing for the wrong metric.**
The structural gap — attribution windows, server-side billing, data silos — is solvable. It requires billing-to-attribution integration, channel-level funnel visibility, and revenue attribution. **The question is whether you need an enterprise contract to access these capabilities, or whether they are included from day one.**
[**_See which channel drives subscribers. Native RevenueCat/Adapty S2S. $0.05/install. 15K free. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## Mobile App Deep Linking: How to Fix the Click-to-Install Conversion Gap That Kills Paid UA ROI
URL: https://www.airbridge.io/en/blog/deep-linking-app-store-conversion
Language: English
Published: 2026-03-24
Author: Jaehyuk Kim
A fitness app runs a TikTok campaign promoting a 7-day free trial for a personalized workout plan. The ad generates 10,000 clicks. The app store listing shows 3,200 installs. But only 1,900 users ever open the app - and fewer than half of those see the trial offer they clicked on. The rest land on a generic home screen, confused about what brought them there.
That gap between the ad click and the first meaningful session is where paid UA (user acquisition) budgets go to die. **Every user who installs but never activates is a subscription you paid to acquire and never collected.** The problem is not the ad creative or the app store listing. It is the link between them - and for most teams, that link is broken.
**Key Takeaways**
- **Standard web links break at the app store boundary.** Users lose all campaign context during the redirect, and a significant share of high-value users drop off before their first session.
- **Deferred deep links preserve context through the install flow.** New users who download through a deferred deep link land on the exact screen the ad promised - not a generic home page.
- **Deep-linked campaigns deliver dramatically higher conversion rates.** Moburst reports a 51% conversion rate increase with deep links compared to standard links.
- **Influencer campaigns without deferred deep links waste the trust influencers build.** Mobile app deep linking for influencer campaigns can increase user acquisition by 40% and improve retention by 25%.
- **Airbridge Core Plan connects deep links to the subscription funnel** - linking the install to trial start, subscription, and renewal via GMAT attribution and RevenueCat server-to-server (S2S) integration.
## Why Regular Links Lose Users Between the Ad Click and the App Store
Every paid UA campaign makes an implicit promise: click this ad, get this experience. Standard web links break that promise the moment a user hits the app store.
### 1\. The 'Ghost Install' Problem: Users Forget Why They Downloaded
A user clicks an Instagram ad for a 50% discount on a meditation app's annual plan. The link opens the App Store. By the time the install completes, the discount context is gone. The app opens to a generic onboarding flow with no mention of the offer.
This is the **ghost install** - the user installed, but the reason they installed has vanished. Without [deep linking](https://www.airbridge.io/glossary/deep-linking), the app cannot know what campaign or offer drove the install.
### 2\. UTM Parameters Vanish at the App Store Redirect
UTM parameters exist in URLs - and the app store does not pass URL parameters through to the installed app. The moment a user transitions from web to app store to app, **every campaign identifier disappears**.
All installs from paid campaigns look identical to organic installs. You cannot distinguish a high-intent campaign install from a random app store browse.
### 3\. The Compounding Cost: Lost Context Means Lost Subscriptions
Ghost installs and vanished UTMs create the same outcome - **the app cannot distinguish a high-intent campaign user from an organic browser**. Everyone gets the same default onboarding, regardless of what drove the install.
For subscription apps, this is where the cost compounds. A user acquired through a paid campaign has a known CPA attached. When that user churns before ever seeing the offer they clicked on, the acquisition cost is spent and the subscription revenue never arrives. Multiply that across every campaign, and the gap between reported installs and actual trial starts becomes the single largest leak in the paid UA funnel.
[__wf_reserved_inherit]
## Three Types of Mobile App Deep Links and When Each One Matters
Not all deep links solve the same problem. Understanding the three types - and when to use each - determines whether your mobile app deep linking strategy actually closes the conversion gap.
### 1\. Standard Deep Links: Routing Users Who Already Have the App
**Standard deep links** route users who already have the app installed directly to a specific in-app screen. If the app is installed, the link opens to the exact page.
The limitation: standard deep links fail for new users. If the app is not installed, the link breaks - redirecting to an error page or the app store listing with no context. They work for re-engagement, push notifications, and email-to-app flows targeting existing users.
### 2\. Deferred Deep Links: Preserving Context Through the Install
**Deferred deep links** solve the new-user problem. When a user without the app clicks a deferred deep link, the system stores the intended destination and campaign context. After the user downloads and opens the app for the first time, the deferred deep link routes them to the correct screen - with all context intact.
This is where the mobile app deep linking app store conversion gap gets fixed. The user who clicked on a 7-day trial offer actually sees the trial offer after install. The user who tapped an influencer's link to a specific feature lands on that feature.
Deferred deep links are the foundation of effective paid UA for any app targeting new users.
### 3\. Contextual Deep Links: Carrying Attribution Data End-to-End
**Contextual deep links** extend deferred deep links by embedding attribution parameters - campaign ID, ad group, creative variant, channel, and referrer. These parameters travel with the user through the entire journey from ad click to subscription.
For growth teams, contextual deep links answer the question standard analytics cannot: which campaign, on which channel, drove this subscriber? Learn more about how [Mobile Measurement Partners (MMPs)](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) use contextual deep links to connect installs to revenue events.
[**_Airbridge Core Plan includes deferred deep linking, custom tracking links, and subscription attribution in one stack. Get Started with Airbridge Core Plan →_**](https://www.airbridge.io/en/plans/core)
## Deep Links Open the App but Land on the Home Screen: Configuration Problem or Missing Capability?
"Our deep links open the app, but users always land on the home screen, so the campaign context is lost every time." This is the most common deep linking support question, and the answer depends on one fact: did the user already have the app installed when they tapped?
| Symptom | Most likely cause | Configuration or capability? | Fix |
| --- | --- | --- | --- |
| Existing user taps the link, the app opens, home screen | The app received the URL but nothing routed it: the deep link callback is not implemented, the route is not registered, or the handler runs before the navigation stack is ready | Configuration | Implement the SDK deep link callback and handle the cold-start case. See [push notification deep links: fix the wrong screen problem](https://www.airbridge.io/en/blog/push-notification-deep-links-fix-wrong-screen-ctr) |
| New user installs from the link, first open lands on the home screen | Universal Links and App Links carry nothing through the App Store or Play Store. No destination was stored | Missing capability. You need deferred deep linking | Send ads through a tracking link with deferred deep linking. Airbridge Core Plan includes it |
| The app opens on the right screen, but campaign, ad group, and creative values are empty | The link opened the app as a plain HTTPS or URI scheme link, not as a tracking link that carries attribution parameters | Configuration | Route every ad, influencer, and email link through a tracking link and verify the parameters with the deep linking test tool in the dashboard |
| Works from Safari or Chrome, but goes to the App Store when tapped inside Instagram or Facebook on iOS even though the app is installed | iOS first-click behavior from apps other than Safari: the first tap on a tracking link inside App X can open the App Store instead of the app | Configuration (a link option) | Turn on the Stopover Airpage option for links aimed at installed users (bio, story, re-engagement). For install-focused ad links, test whether the extra prompt costs store conversions before enabling it. See [what breaks when a deep link is opened inside a social app browser](https://www.airbridge.io/en/blog/deep-link-opened-in-social-app-browser-what-breaks) |
| Works when the app is in the background, fails from a full cold start | The handler only covers the foreground and background states | Configuration | Add a cold-start handler that reads the initial URL before the first screen renders |
| Deferred deep link works when the user installs and opens the app soon after the click, fails for a user who installs hours or days later | Deferred deep linking works for 1 hour after the click ([help center](https://help.airbridge.io/en/guides/retargeting-with-deep-links)). Attribution uses a separate, longer lookback window | Expected behavior, not a bug | Keep the click-to-install path short. Late installs are still attributed inside the attribution lookback window, but the routing context is gone |
**Three tests that tell you which row you are in:**
1. Install the app, tap your tracking link from Safari or Chrome, and confirm the target screen loads. If it does not, the problem is in the app handler (row 1 or row 5).
2. Delete the app, tap the same link, install from the store, and open the app. If you land on the home screen, you do not have deferred deep linking in place (row 2).
3. With the app installed, tap the link from an Instagram story or DM on iOS. If the App Store opens, that is the iOS first-tap behavior (row 4). Enable the Stopover Airpage option on that link and test again.
## How Mobile App Deep Linking Fixes App Store Conversion Drop-Off
The conversion gap between ad click and app activation is measurable - and deep linking closes it at every stage.
### 1\. Deep Links Deliver Higher Click-to-Install Rates
Industry benchmarks consistently show that **deep-linked campaigns achieve significantly higher click-to-install (CTI) rates** than standard links. [Moburst reports](https://www.moburst.com/blog/how-app-deep-linking-helps-you-improve-your-conversions/) a 51% conversion rate increase with deep links - turning the most expensive part of the funnel into the most efficient.
The mechanism is straightforward. Instead of dropping users on a generic app store page, deep links create a continuous experience from ad to app.
### 2\. Post-Install First-Session Experience Matches the Ad Promise
What happens after install matters more for subscription apps - the install is free, but the subscription is where revenue lives.
Deferred deep links ensure the **first session matches the ad promise**. If the ad promoted a free trial, the user lands on that trial signup screen. If a QR code linked to a workout program, the user opens directly to that program. This context match drives higher trial start rates.
### 3\. Conversion Lift Across Email, QR, and Referral Channels
Deep linking benefits extend beyond paid ads. Across owned and earned channels - email, QR codes, referral programs - deep links consistently [drive higher click-to-app conversion rates](https://www.airbridge.io/en/blog/organic-app-growth-with-deep-linking) than standard links. These channels depend on context: a referral link that drops the user on a generic home screen loses the social proof that drove the click.
Web-to-app flows in particular are growing rapidly as more teams adopt [deep linking for organic growth](https://www.airbridge.io/en/blog/organic-app-growth-with-deep-linking), turning passive web visitors into active app users.
## Mobile App Deep Linking for Influencer Campaigns and User Onboarding
Two areas where broken links cost the most: influencer campaigns and new user onboarding. Both rely on trust and context - exactly what standard links destroy.
### 1\. Deferred Deep Links in Influencer Campaigns
An influencer posts a story recommending an app. Followers tap the link. But the link goes to the app store, and every tracking parameter is stripped. The app sees a spike in installs but cannot attribute them to the influencer.
A proper **mobile app deep linking influencer campaign** setup solves this with custom tracking links carrying deferred deep link parameters. Each influencer gets a unique link. When a follower installs and opens the app, the deferred deep link routes them to the promoted content - and attribution data connects that install to the specific creator.
According to [Rocket Farm Studios](https://www.rocketfarmstudios.com/blog/deep-linking-and-deferred-deep-linking-in-mobile-apps/), apps using deferred deep links in influencer campaigns see **up to 40% increase in user acquisition and 25% improvement in retention**.
### 2\. Personalized Onboarding: Why Retention Improves with Context
When the app knows the context that drove the install, it can tailor the first session. A user who installed through a "beginner yoga" campaign sees beginner content first. A user from a "meal prep" influencer link lands on meal planning features. This is **mobile app deep linking user onboarding** done right - the app feels personalized from the first second.
The retention impact compounds. Users with a relevant first session are more likely to complete onboarding, start a trial, and convert to paid.
### 3\. The Firebase Dynamic Links Sunset and What Replaced It
Google shut down Firebase Dynamic Links on August 25, 2025 - after deprecating the service in 2023. The shutdown accelerated a shift toward [MMP-based deep linking](https://www.airbridge.io/glossary/deep-linking) - where the same platform that handles attribution also manages deep link routing.
For teams evaluating solutions post-Firebase, the key question is not "can this tool route users?" but "can it connect the deep link to the subscription event that happens days or weeks later?"
[__wf_reserved_inherit]
## Connecting Deep Links to Subscription Attribution with Airbridge Core Plan
Deep linking solves the front-door problem - getting users from ad to app with context intact. But for subscription apps, the real question is what happens after the install. Did that deep-linked user start a trial? Did they subscribe? Did they renew?
### General Approaches
Most growth teams use separate tools for deep linking, attribution, and subscription analytics. Connecting a specific deep link click to a subscription event requires manual stitching across systems.
The gap: deep linking tools know where users came from, and subscription platforms know who pays - but connecting the two requires an [attribution layer](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution) that spans the full journey.
### How Airbridge Core Plan Connects the Full Journey
Airbridge Core Plan includes deep linking and deferred deep linking as part of the attribution stack. Here is what that means in practice:
- **Custom domain tracking links** - each campaign, influencer, or channel gets a branded link with deferred deep link parameters built in
- **GMAT attribution** - installs from Google, Meta, Apple Search Ads, and TikTok are attributed to the specific campaign and creative that drove them
- **25 standard events including Start Trial, Subscribe, and Unsubscribe** - the subscription funnel is tracked with predefined event names, reducing schema design work
- **RevenueCat and Adapty S2S integration** - subscription events flow directly from the billing platform to Airbridge, capturing renewals and cancellations regardless of whether the user opens the app
- **Funnel report** - visualize the journey from deep link click to install to trial to subscription, broken down by channel and campaign
- **Test Console** - validate deep link behavior across iOS Universal Links and Android App Links before launching a campaign
Airbridge Core Plan starts at $40+/mo after a 30-day free trial, with 500K data points a month included, $0.0001 per additional data point, and no annual contract. Attribution events and tracking link clicks do not count as data points. [Deferred deep linking](https://www.airbridge.io/en/blog/deferred-deeplink-benefits) is included in Core Plan, with no enterprise tier required.
Core Plan supports GMAT channels and a maximum of 2 third-party integrations - enough for a subscription app running paid UA on major channels with RevenueCat or Adapty handling billing. Teams needing custom events, non-SAN networks, or raw data export can upgrade to Growth Plan.
## Frequently Asked Questions
### Our deep links open the app but always land on the home screen. Is that a configuration problem or a missing capability?
If the user already had the app installed, it is a configuration problem: the app is not handling the deep link URL, most often on cold start. If the user installed the app from the link and then landed on the home screen, it is a missing capability. Universal Links and App Links do not carry a destination through the app store, so you need deferred deep linking from a tracking link provider.
### Do Universal Links or App Links carry campaign context through the App Store?
No. Both standards only route users who already have the app installed. When the app is missing they fall back to a web URL, and the standards themselves carry nothing through the store. Deferred deep linking stores the destination and campaign parameters at click time and restores them on the first app open.
### Why do my deep links go to the App Store from Instagram when the app is already installed?
On iOS, the first tap on a tracking link inside an app other than Safari can open the App Store instead of the installed app. Enabling the Stopover Airpage option on the Airbridge tracking link shows a prompt to open the app instead. A second tap on the same link in the same app routes to the deep link destination.
## Every Lost Install Is a Subscription You Paid For and Never Collected
The cost of a broken link is not a lost click. It is a lost subscriber - someone who saw your ad, trusted your promise, and downloaded your app, only to land somewhere that had nothing to do with why they came.
Deep linking is not a technical nice-to-have. It is the connective tissue between your ad spend and your subscription revenue. Every campaign without deferred deep links is a campaign where you pay for installs and hope users figure out the rest.
[**_Get Started with Airbridge Core Plan - connect your deep links to your subscription funnel with attribution that spans the full journey from ad click to paid subscriber._**](https://www.airbridge.io/en/plans/core)
---
## Influencer Attribution Tracking: Promo Codes vs Tracking Links -- Which Actually Works for Mobile Apps?
URL: https://www.airbridge.io/en/blog/influencer-attribution-promo-code-vs-tracking-link
Language: English
Published: 2026-03-24
Author: Jaehyuk Kim
You gave 10 influencers unique promo codes. Three months later, the codes show 200 redemptions. But your organic installs spiked by 2,000 during the same campaigns. You gave another 5 influencers tracking links. The links show 150 clicks - but you know from App Store Connect that installs were 3x higher than what the links captured.
Neither method is telling you the full story. **This is the core challenge of influencer attribution tracking for mobile apps: neither promo codes nor tracking links tell you which influencer's audience converted to paying subscribers.**
**Key Takeaways**
- **Promo codes are deterministic but incomplete.** They confirm a user came from a specific influencer - but only a fraction of influencer-driven users actually redeem a code. The rest search the App Store directly and never enter it.
- **Tracking links provide click-through visibility but break on social platforms.** Users on TikTok and Instagram rarely click links - they watch content, then search for the app. Click-based attribution misses the majority of influencer-driven installs on these platforms.
- **Neither method captures post-install subscription data.** Promo codes tell you who redeemed. Links tell you who clicked. Neither tells you who started a trial, subscribed, or renewed.
- **The real gap is not codes vs links - it is connecting influencer attribution to subscription revenue.** That requires deep links, an [MMP](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp), and server-to-server billing integration.
- **Airbridge Core Plan bridges this gap** with deep links that preserve attribution through the App Store, plus S2S integration with RevenueCat and Adapty to connect installs to subscription events.
## Why Influencer Attribution Breaks on Mobile
Influencer attribution tracking is the process of connecting an influencer's content to downstream user actions - installs, trials, and subscriptions - so growth teams can measure which partnerships produce revenue. [On the web](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution), this is relatively straightforward: user clicks link, lands on page, converts. Cookies and UTM parameters track the journey. Mobile apps break this model at every step.
**Users do not click links.** A fitness trainer posts a workout video on Instagram or TikTok. The audience watches, becomes interested, and searches the app name directly in the App Store. They never tap the link in bio, never click a swipe-up, never touch a trackable URL. **On social platforms, the dominant user behavior is watch-then-search - not click-through.**
**Privacy features block cross-app tracking.** Even when a user does click a link, iOS App Tracking Transparency (ATT) prevents connecting that click to a later install unless the user opts in. Most decline. The click happens in one app, the install in another, and on most iOS devices those two events cannot be connected.
**Delayed conversions fall outside attribution windows.** A user discovers a trainer's "12-week transformation" series. They follow along for weeks before installing the recommended app. Standard attribution windows are 7-28 days. **By the time they install, the window has closed and the install appears organic - even though the influencer clearly drove it.**
The result: most influencer-driven installs show up as organic in your MMP. This is not a tracking bug. It is a structural mismatch between how influencer marketing works - through awareness and trust - and how attribution systems work -- through clicks and cookies.
## Promo Codes vs Tracking Links: What Each Method Actually Captures
[__wf_reserved_inherit]
Both methods have real strengths. Both have structural blind spots. Understanding what each one captures - and what it misses - is the first step toward closing the gap.
### 1\. How Promo Codes Work
An influencer shares a unique code - "TRAINER20" - and the user enters it at checkout or during onboarding. The code connects that conversion to the influencer deterministically. No cookies, no click tracking, no device-switching problems.
**Where promo codes work well:**
- **Podcast and YouTube** - audio/video mentions where clicking a link is not natural
- **Cross-device** - the code works regardless of which device the user installs on
- **Privacy-proof** - no reliance on IDFA, cookies, or cross-app tracking
**Where promo codes fail:**
- **Most users never enter the code.** They hear about the app, search it, install it, and start using it - without remembering or bothering with a code. [Promo codes work as clickless tracking](https://www.tune.com/blog/promo-codes-track-influencer-campaigns-without-click/) but only capture users who actively remember and enter them - a minority of total influencer-driven conversions.
- **No journey data.** A code tells you who redeemed. It does not tell you when they first saw the content, how many times they visited the App Store listing, or what they did before converting.
- **Code sharing and fraud.** Codes leak to coupon aggregator sites. A "TRAINER20" code intended for one influencer's audience ends up on RetailMeNot, and you attribute conversions to the wrong source.
### 2\. How Tracking Links Work
An influencer shares a unique URL - often placed in their link-in-bio or story swipe-up. When a user clicks, the link routes them through the App Store while preserving attribution data. If the link uses deep linking, it can also route the user to a specific screen after install.
**Where tracking links work well:**
- **Full-funnel click visibility** - you see the click, the install, and the in-app events that follow
- **Deep linking** - users land on the relevant content after install, not a generic home screen
- **Scalable attribution** - generate unique links per influencer, per campaign, per content piece
**Where tracking links fail:**
- **Platform restrictions limit link placement.** Instagram does not allow clickable links in feed posts. TikTok limits link placement to bio. When [comparing coupon codes vs tracking links](https://influencermarketinghub.com/coupon-codes-vs-tracking-links/), links consistently underperform on platforms that restrict or strip URLs.
- **UTM parameters do not survive the App Store redirect.** Standard UTM-tagged links lose their parameters when the user is redirected to the App Store. You need deep links with an [MMP](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) to preserve attribution through the install flow.
- **Click-to-install connection breaks.** Even when a user clicks, iOS ATT, VPNs, and ad blockers can prevent the attribution system from connecting the click to the install. The data trail disappears mid-funnel.
### 3\. Side-by-Side Comparison
[__wf_reserved_inherit]
**Neither method alone captures the full picture.** Promo codes work where links fail - and links work where codes are impractical. But both share the same critical blind spot: \*\*neither tells you what happens after the install.
[**_Airbridge Core Plan connects deep links, promo code attribution, and S2S billing integration in one stack. Get Started with Airbridge Core Plan →_**](https://www.airbridge.io/en/plans/core)
## The Hybrid Approach - And What Is Still Missing
The practical solution is to use both methods together.
**Embed promo codes inside deep links.** When a user clicks the influencer's link, the deep link routes them to the App Store and preserves attribution data. After install, the app reads the deep link parameter and pre-fills the promo code - so the user does not need to remember or type it. If the user does not click - they search the App Store directly - the promo code still works as a fallback. This creates dual-layer attribution: link for click-through, code for no-click.
**Use deep links for journey data, codes for reach.** The link captures the users who click. The code captures users who heard about the app through any medium - podcast, live stream, word of mouth - and entered the code without clicking anything.
**But there is a gap both methods leave open.** For subscription apps, the install is not the conversion. The conversion is: trial start → paid subscription → renewal. **Without connecting attribution to billing events, you cannot tell which influencer's audience actually generates subscription revenue.**
An influencer with 500K followers may drive 2,000 installs and 10 subscribers. A micro-influencer with 30K followers may drive 200 installs and 50 subscribers. Without subscription-level attribution, both look the same in your budget review.
Closing this gap requires connecting attribution data to billing events - trial start, subscribe, renew, cancel - through a server-to-server integration with your billing platform.
[__wf_reserved_inherit]
## How to Connect Influencer Attribution to Subscription Revenue
### General Approaches
Regardless of tooling, growth teams can take these steps today:
- **Assign unique deep links AND unique promo codes per influencer.** Double coverage captures both click-through and no-click conversions.
- **Measure organic lift during campaign windows.** Compare baseline organic installs (before campaign) to installs during the campaign period. The difference estimates influencer-driven organic volume.
- **Track subscription events by attributed source.** If your billing platform (RevenueCat, Adapty) supports source tagging, tag subscription events with the influencer attribution data from your MMP.
- **Extend attribution windows for influencer campaigns.** Influencer content has a longer conversion cycle than paid ads. Standard 7-day windows miss delayed conversions - configure 14-30 day windows for influencer sources.
### How Airbridge Core Plan Connects the Full Journey
Airbridge Core Plan connects influencer attribution to subscription revenue in a single stack - without enterprise contracts or custom implementations.
- **Deep links per influencer, per campaign.** Generate unique [tracking links](https://www.airbridge.io/en/blog/abr-revenuecat-integration) that preserve attribution through the App Store redirect. Supports deferred deep linking - users land on the relevant content after install, not a generic home screen.
- **S2S billing integration with RevenueCat and Adapty.** Subscription events - Start Trial, Subscribe, Unsubscribe - flow back to the attributed install source. See which influencer's audience converts to paid, which tier they choose, and whether they renew.
- **Unified reporting across influencer and paid UA.** Influencer deep link data and GMAT channel data (Google, Meta, Apple Search Ads, TikTok) appear in the same Funnel and Revenue reports. Compare influencer CAC to paid CAC on an equal basis - not in separate spreadsheets.
- **25 subscription-optimized standard events.** No custom event schema design needed. Events like Start Trial, Subscribe, and Order Complete are predefined - reducing setup decisions for the subscription funnel.
- **15K free attributed installs, then $0.05/install.** All features included from day one - deep links, S2S billing, unified reporting. No tiered feature gates.
Core Plan's limitation is intentional focus: it supports GMAT channels and a maximum of 2 third-party integrations. If you need non-SAN ad networks or raw data export, that is Growth Plan territory. For teams running influencer campaigns alongside Meta, Google, Apple, and TikTok paid UA, Core Plan covers the full attribution-to-subscription stack.
## Neither Promo Codes Nor Tracking Links Tell You What Matters Most
The debate between promo codes and tracking links misses the real question for subscription apps. **The question is not which method captures more installs - it is which influencer's audience actually becomes paying subscribers.**
Codes and links are both necessary inputs. But the output that drives budget decisions - subscription revenue by influencer - requires connecting attribution to billing events.
Until you can see which influencer drives subscribers, not just installs, you are optimizing the wrong metric.
[**_Get Started with Airbridge Core Plan_** **_— deep links + RevenueCat S2S + unified reporting. Connect influencer attribution to subscription revenue._**](https://www.airbridge.io/en/plans/core)
---
## Why Influencer Marketing Drives Views But Not App Installs for Fitness Apps
URL: https://www.airbridge.io/en/blog/influencer-marketing-app-installs-tracking
Language: English
Published: 2026-03-24
Author: Jaehyuk Kim
A fitness trainer you partnered with posts a 45-second workout video. It goes viral — 500,000 views, 12,000 likes, hundreds of comments asking "what app is this?" You check your dashboard the next morning. **200 new installs.**
You re-check the numbers. The video is real. The engagement is real. But somewhere between "what app is this?" and the App Store download button, **499,800 people vanished.** And your analytics cannot tell you where they went.
This is not a content quality problem. The video performed. **It is a funnel and measurement problem — and most fitness app teams cannot see where the drop-off actually happens.**
**Key Takeaways**
- **The view-to-install gap is structural, not a content failure.** Industry average is [4.27 installs per 1,000 impressions](https://www.businessofapps.com/data/app-conversion-rates/). A 500K-view video producing 200 installs is not underperforming — it is normal. The question is whether you can improve it.
- **Users drop off at every step between video and install.** Leaving the social platform, searching the App Store, reading the listing, deciding to download — each step loses people. [Only ~25% of users who reach an App Store page actually install](https://www.businessofapps.com/data/app-conversion-rates/).
- **Most of the drop-off is invisible to your tracking.** Cross-device behavior, delayed conversions, and privacy features like iOS ATT mean your analytics [cannot connect many touchpoints](https://www.cometly.com/post/marketing-funnel-attribution-challenges) between content view and install.
- **93% of marketers use influencer marketing, but measurement lags behind.** [Only 70% track ROI at all](https://iqfluence.io/public/blog/influencer-marketing-statistics) — and "tracking ROI" often means counting installs, not understanding the funnel.
- **Closing the gap requires reducing steps, capturing delayed installs, and measuring what happens between view and download.**
## The Math Behind View-to-Install
Before diagnosing what is broken, it helps to understand what "normal" looks like.
The global median [IPM (installs per mille) is 4.27](https://www.businessofapps.com/data/app-conversion-rates/) — meaning for every 1,000 ad impressions, roughly 4 people install. Influencer content is not a paid ad, but the conversion physics are similar: **most people who see your app mentioned will not install it.** That is true even when the content is excellent.
The [App Store page conversion rate averages about 25%](https://www.businessofapps.com/data/app-conversion-rates/). Google Play is slightly higher at 27.3%. **This means even among users who actively navigate to your app's store listing, three out of four leave without installing.**
For a 500K-view fitness video, the funnel math works like this:
- 500,000 people see the video
- A fraction engage (like, comment, save)
- A smaller fraction leave the platform to search for your app
- Of those who reach the App Store page, ~25% install
[__wf_reserved_inherit]
**Each step is a multiplier that shrinks the number.** The gap between 500K views and 200 installs is not a mystery — it is compounding drop-off across 4–5 steps that you probably cannot see individually.
## Where Fitness App Users Actually Drop Off
The journey from "watched a trainer's video" to "installed the app" is longer than it looks.
**Step 1: Platform exit.** The user sees the video on Instagram or TikTok. To install your app, they must leave the platform. Most do not. Social platforms are designed to keep users scrolling — **every swipe-up, link-in-bio tap, or App Store search requires the user to actively interrupt their feed behavior.** Most users save the video and move on.
**Step 2: Search friction.** Users who do leave the platform rarely click a link. Instead, they search your app name in the App Store. If your app name is generic ("FitPro," "WorkoutPlus"), they may find competitors first. **If the search does not return your app in the first 2–3 results, you lose them.**
**Step 3: App Store page conversion.** The user finds your listing. Now they evaluate — screenshots, ratings, reviews, app size, subscription price. For fitness apps, a visible subscription price ($9.99/mo or $59.99/yr) on the listing page causes additional hesitation. **Even users who reach this step — the ones who left the platform and searched — convert at only ~25%.**
These three steps are where users actually leave the funnel. But there is a second category of drop-off that is harder to see — not because users leave, but because your tracking cannot follow them.
## Why Your Tracking Cannot See the Rest
[__wf_reserved_inherit]
**Users who convert later are invisible.** The user saw the video on their phone at lunch. They plan to install later. By evening, they have forgotten the app name, cannot find the saved video, or install a competitor's app that appeared in an App Store ad. Workout decisions are often made at different moments than content consumption — **and this delayed, cross-device behavior is where fitness apps lose the most users without knowing it.**
**Click-based attribution misses impression-driven behavior.** Influencer marketing works through awareness, not clicks. A user watches a video, does not click anything, and searches your app three days later. [Most attribution systems are built around clicks](https://www.adjust.com/blog/influencer-marketing-for-mobile-apps/) — if there is no click, there is no data point. **The most valuable influencer-driven installs are the ones your tracking is structurally designed to miss.**
**Privacy features block cross-app tracking.** iOS App Tracking Transparency requires users to opt in to cross-app tracking — [and most decline](https://www.cometly.com/post/marketing-funnel-attribution-challenges). When a user watches a TikTok video and later installs your app from the App Store, those two events cannot be connected on most iOS devices. **The user journey exists. The data trail does not.**
**Delayed conversions fall outside attribution windows.** A user discovers a trainer's "12-week transformation" series, follows along for a month, then installs the app. Standard attribution windows are 7–28 days. **By the time they convert, the window has closed and the install appears organic** — [even though the influencer clearly drove it](https://www.cometly.com/post/marketing-funnel-attribution-challenges).
**You see totals, not the funnel.** Most dashboards show views (from the social platform) and installs (from the MMP). Everything in between — platform exits, App Store page visits, search queries, delayed decisions — is a black box. \*\*You cannot optimize a funnel you cannot see.
[**_Close the view-to-install gap. Deep links, view-through attribution, unified reporting. $0.05/install, 15K free._**](https://www.airbridge.io/en/plans/core)
## How to Close the View-to-Install Gap
The gap between views and installs has two causes: too many steps, and too little visibility. Fixing it means addressing both.
### Reduce the Steps
**Use deep links from content directly instead of link-in-bio CTA.** A deep link can take the user directly to a specific page in your app once they open the app after installing. A trainer promoting a "30-day ab challenge" should link directly to that program — not send users to a generic link-in-bio page. Every tap you eliminate between video and install recovers users who would have dropped off.
**Optimize your App Store listing for influencer traffic.** Users arriving from influencer content have different expectations than paid ad traffic. They saw a specific workout, a specific trainer, a specific result. If your App Store screenshots show generic features instead of the content they just watched, **you lose them at the biggest remaining drop-off point.** Consider creating custom product pages (Apple) or store listing experiments (Google) aligned with your top influencer campaigns.
### Expand What You Can See
**Use view-through attribution, not just click-through.** An [MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) with view-through attribution can connect an ad impression (via influencer whitelisting) to a later install — even if the user never clicked. This captures the users who saw the content, did not click, but installed later.
**Extend attribution windows for influencer campaigns.** Fitness content has a longer conversion cycle than paid ads. A user may follow a trainer for weeks before installing the app they recommend. Standard 7-day windows miss these conversions. **Configure longer windows for influencer sources — 14 to 30 days — to capture the delayed installs that are currently showing as organic.**
**Measure organic lift during campaign windows.** Compare your organic install rate during an influencer campaign to your baseline. If your daily organic installs jump from 50 to 120 during a campaign, the lift is likely influencer-driven. This is not per-user attribution, but it quantifies aggregate impact that click-based tracking misses entirely.
### How Core Plan Bridges the Gap
Airbridge Core Plan connects these measurement layers in a single stack — without requiring enterprise contracts or custom implementations.
- **Deep links at scale.** Generate unique deep links per influencer, per campaign. Each link preserves attribution data and supports deferred deep linking — so users land on the right screen after install, not a generic home screen.
- **View-through attribution built in.** Capture installs driven by influencer impressions (via whitelisting), not just clicks. Users who see the content and install days later are attributed — not lost to organic.
- **Longer attribution windows.** Configure windows up to 30 days for influencer sources to capture the delayed conversions that fitness content naturally produces.
- **Unified reporting.** Influencer and paid campaign data in the same Funnel and Revenue reports — broken down by channel, campaign, and creative. No more spreadsheets vs. dashboards.
- **S2S billing integration.** Native RevenueCat and Adapty integration connects install attribution to subscription outcomes. See which influencer's installs convert to paying subscribers — not just which ones drive downloads.
- **$0.05/install. 15K free attributed installs.** All features included from day one.
## Before You Scale: Five Blind Spots to Check
Before launching or scaling influencer partnerships, test your measurement setup against these:
- **Can you see how many users reached your App Store page from an influencer's content?** If you only see views (from the platform) and installs (from the MMP), the entire middle of the funnel is invisible.
- **Do you know the average delay between content view and install?** If your attribution window is 7 days but most influencer-driven installs happen on day 10–14, you are systematically undercounting.
- **Can you distinguish influencer-driven organic installs from true organic?** If organic installs spike 2x during a campaign but your dashboard labels them all "organic," you cannot evaluate the campaign.
- **Are your deep links routing users to relevant content — or to the home screen?** A trainer's audience expects to land on the specific program they saw. A generic home screen is another drop-off point.
- **Can you compare view-to-install rates across different influencers?** If Influencer A gets 500K views and 300 installs while Influencer B gets 100K views and 250 installs, the second partnership is 8x more efficient. But you can only see this if your tracking connects views to installs per influencer.
If your setup cannot answer these, the framework above shows where to start.
## The View-to-Install Gap Is a Measurement Problem, Not a Content Problem
[93% of marketers already use influencer marketing](https://iqfluence.io/public/blog/influencer-marketing-statistics). [86% of consumers make influencer-inspired purchases](https://iqfluence.io/public/blog/influencer-marketing-statistics). The channel works. **The issue is not that influencer content fails to drive interest — it is that the path from interest to install has too many steps and too little visibility.**
For fitness apps, the gap is especially wide. Your audience discovers content on Instagram and TikTok, but installs happen in the App Store — a completely separate platform with no native connection to social engagement data. **Every user who crosses that gap does so invisibly unless your measurement stack is built to see it.**
The views are real. The interest is real. **The question is whether you can see where 499,800 people went — and bring more of them to the other side.**
[**_See where users drop off between content and install. Deep links + view-through attribution + unified reporting. $0.05/install. 15K free. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## Influencer Marketing Cuts CAC — But Only If You Can Measure It
URL: https://www.airbridge.io/en/blog/influencer-marketing-app-cac-attribution
Language: English
Published: 2026-03-24
Author: Jaehyuk Kim
Your Facebook cost per lead [jumped 21% this year](https://searchengineland.com/facebook-ad-costs-jump-beat-google-461690) to $27.66. Google search CPC [hit $4.66](https://thisisswell.com/google-ads-costs-2025/) — up from $4.22 last year and $4.01 the year before. You are spending more to acquire users who are not subscribing at a higher rate.
Meanwhile, a fitness influencer posts a 60-second workout video mentioning your app. Comments fill with "what app is this?" The App Store ranking ticks up. New installs appear — but in your MMP, they show up as organic. **You cannot tell which influencer drove them, whether those users started trials, or if any of them became paying subscribers.**
This is the influencer marketing attribution gap. The channel that [returns $5.78 for every $1 spent](https://pageoneformula.com/influencer-marketing-cost-cpm-benchmarks-2024-2025/) — nearly 3x paid media's ~$2 return — is the one most teams cannot measure.
**Key Takeaways**
- **Paid ad costs are rising faster than conversion rates.** Facebook CPL up 21%, Google CPC up 16% in two years. More spend, not more subscribers.
- **Influencer marketing produces higher ROI — but only in aggregate.** [ROI of $5.78 per $1](https://pageoneformula.com/influencer-marketing-cost-cpm-benchmarks-2024-2025/) and [CAC reductions of 15–61%](https://mightyscout.com/blog/how-can-you-reduce-cac-with-influencer-based-ads-and-whitelisting-in-2025) are proven. The problem is attributing those results to specific influencers and campaigns.
- **Most influencer-driven installs appear as organic.** Users see a video, search the App Store directly, and install — without clicking a trackable link. No click, no attribution.
- **Without attribution, you cannot connect influencer spend to subscription revenue** — making it impossible to compare influencer CAC to paid CAC on an equal basis.
- **Measuring influencer marketing for fitness apps requires deep links, an MMP, and billing integration** — not just UTM parameters or coupon codes.
## Paid Ad Costs Are Rising — Conversions Are Not
The cost of acquiring users through paid channels is increasing across every major platform.
- **Facebook:** Average cost per lead reached [$27.66 in 2025 — up 21% year-over-year](https://searchengineland.com/facebook-ad-costs-jump-beat-google-461690). Conversion rates dropped from 8.67% to 7.72%. **You are paying more per lead while fewer leads convert.**
- **Google:** Average search CPC [hit $4.66 in 2024](https://thisisswell.com/google-ads-costs-2025/), up from $4.01 in 2022. Some sectors saw 25%+ annual increases. The long-term trend is consistent: CPC grows faster than inflation.
For fitness and health apps, this creates a specific problem. **Your subscription payback window depends on keeping CAC below a threshold — and that threshold is not moving up as fast as ad costs.** When CPL rises this fast but your subscription price stays at $9.99/month, every dollar of CAC inflation compresses your margin.
This is why teams are shifting budget toward influencer marketing.
[__wf_reserved_inherit]
## Why Influencer Marketing Works for Fitness Apps
The numbers support the shift:
- **ROI:** Influencer campaigns [return nearly 3x](https://pageoneformula.com/influencer-marketing-cost-cpm-benchmarks-2024-2025/) what traditional paid media generates per dollar spent.
- **CAC reduction:** Brands using influencer-based ads with whitelisting have achieved [15–61% lower CAC](https://mightyscout.com/blog/how-can-you-reduce-cac-with-influencer-based-ads-and-whitelisting-in-2025) compared to standard paid campaigns.
- **Content efficiency:** One fitness brand obtained [10 videos and 30 photos for $3,500](https://mightyscout.com/blog/how-can-you-reduce-cac-with-influencer-based-ads-and-whitelisting-in-2025) — budget previously allocated to a single professional production.
- **Trust transfer:** 85% of marketers report that influencer marketing improves customer acquisition quality. For fitness apps, a trainer demonstrating your workout feature carries more credibility than a performance ad.
**Fitness apps have a natural advantage here.** The product is visual — workouts, transformations, progress tracking — and the audience already follows fitness creators. The channel-audience fit is stronger than almost any other app category.
But there is a gap between "influencer marketing works" and "we know which influencer drives subscribers."
## The Attribution Blind Spot
Paid ads have a clear attribution path: user clicks ad → installs app → MMP attributes the install to the campaign. Influencer marketing does not follow this path.
**Here is what actually happens:**
1. User watches an influencer's Instagram Reel mentioning your app
2. User does not click the link in bio — instead searches "FitApp" directly in the App Store
3. User installs the app
4. MMP records this as an organic install
**The influencer drove the install. Your attribution system does not know it.** The install is real. The influence is real. But the data connection is missing.
This is not a tracking bug — it is a structural problem. Influencer marketing operates through awareness and trust, not through clicks. And most attribution systems are built entirely around clicks.
The result:
- **Influencer-driven installs are hidden inside your organic numbers.** You see organic installs spike after a campaign, but you cannot isolate which influencer caused it.
- **Install-level data is not enough.** Even when an influencer uses a trackable link and you capture the install, you still cannot see if that user started a trial, subscribed, or renewed — because billing events live in a different system.
- **Coupon codes only capture a fraction.** Users forget codes, skip them, or subscribe after the code expires. Coupon-based attribution captures maybe 20–30% of actual influencer-driven conversions.
- **You cannot tell which influencers produce subscribers vs installers.** An influencer with 500K followers may drive 2,000 installs and 10 subscribers. A micro-influencer with 30K followers may drive 200 installs and 50 subscribers. Without subscription-level attribution, both look the same in your budget review.
[**_Measure which influencers drive subscribers — not just installs. Deep links, S2S billing integration, unified reporting. $0.05/install, 15K free._**](https://www.airbridge.io/en/plans/core)
## How to Measure Influencer Marketing for Fitness Apps
Measuring influencer marketing requires connecting three data layers that are usually disconnected: influencer activity, install attribution, and subscription billing.
[__wf_reserved_inherit]
### Layer 1. Deep Links : Direct-Click Attribution
Every influencer needs a unique deep link — not a generic UTM parameter. Deep links route users through the App Store while preserving attribution data. When a user clicks the link and installs, the MMP can attribute that install to the specific influencer, campaign, and content piece.
**What this solves:** Direct-click installs — users who actually tap the link in bio or swipe-up. This is the measurable baseline, but it only captures the fraction of users who click.
**What this does not solve:** The no-click problem described above. Most influencer-driven installs come from users who never click a link — they search the App Store directly. Deep links cannot capture what was never clicked.
The link should also support deferred deep linking — so a user who installs from a trainer's "30-day ab challenge" link lands on that specific program, not the generic home screen. And if you work with dozens of micro-influencers per campaign, confirm you can generate and manage unique links at scale.
### Layer 2. View-Through Attribution and Organic Lift : The No-Click Solution
**This is the layer that solves the core blind spot.** The Attribution Blind Spot section above identifies that most influencer-driven installs appear as organic because users never click a trackable link. Deep links do not address this — view-through attribution and organic lift analysis do.
An [MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) extends attribution beyond clicks. View-through attribution can capture users who saw an influencer ad (via whitelisting) but did not click. Organic lift analysis compares organic install rates during and outside campaign periods to estimate influencer-driven organic installs.
**What this solves:** The no-click organic install problem. If a trainer posts a workout video on Monday and your installs spike Tuesday through Thursday, organic lift analysis quantifies that lift — not just shows it in a chart.
### Layer 3. Billing Integration : Subscription Attribution
Install attribution alone is not enough. You need to connect billing events — trial start, subscribe, renew, cancel — to the attributed install source. This requires a server-to-server integration between your billing platform (RevenueCat, Adapty) and your MMP.
**What this captures:** Which influencer's installs actually converted to subscribers, and whether those subscribers renewed or churned. **This is the layer that turns influencer marketing from a brand expense into a measurable acquisition channel.** The integration should distinguish subscription tiers — monthly ($9.99) vs annual ($59.99) — by attributed source. An influencer whose users pick annual plans is worth far more than one who drives monthly trials that churn after 30 days.
Without all three layers, you have partial data. Deep links without billing integration tell you which influencer drove installs — but not revenue. Billing data without attribution tells you subscription trends — but not which influencer caused them.
When these three layers work together, the measurement stack also needs to support unified reporting — influencer and paid campaign data in the same dashboard, not separate tools — and predefined subscription events so you are not designing custom event schemas for every campaign.
## Questions Your Measurement Stack Should Answer
If you are running or planning influencer campaigns, test your current setup against these questions:
- **Can you tell which influencer drove which subscribers?** Not just installs — subscribers. If you can only see "Influencer A drove 500 installs," you are missing the part that matters.
- **Can you compare cost-per-subscriber across influencer and paid channels?** If Facebook shows CPS of $40 but your influencer data stops at installs, you have no basis for budget allocation.
- **Do influencer and paid ad performance appear in the same dashboard?** If influencer results live in a spreadsheet while paid performance lives in your MMP, optimization decisions are made on incomplete data.
- **Can you track the full journey — from content view to install to trial to renewal?** Partial visibility means partial decisions. A user who installs but never subscribes is a different outcome than one who subscribes and renews for 12 months.
- **Does your setup scale to dozens of concurrent influencer partnerships?** One deep link for one influencer is easy. Managing unique links, attribution windows, and billing data across 30 micro-influencers per campaign is where most stacks break.
If your current stack cannot answer all five, the three-layer framework above shows where the gaps are.
## Influencer Marketing Is Not a Measurement Problem — It Is a Connection Problem
The ROI of influencer marketing for fitness apps is already proven — the data above confirms it. **The gap is not whether influencer marketing works — it is whether you can see which part of it works.**
As paid acquisition costs continue climbing, the pressure to diversify into influencer marketing only increases. **But diversifying without measuring means replacing one blind spot — rising paid CAC — with another: unattributed influencer spend.**
The connection between influencer content, app installs, and subscription revenue is technically possible. **The question is whether your current measurement stack closes that gap.**
[**_See which influencers drive subscribers. Deep links + RevenueCat S2S + unified reporting. $0.05/install. 15K free. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## Free MMP for Fitness App Startups: 1 Year of Attribution Without the Financial Risk
URL: https://www.airbridge.io/en/blog/free-mmp-fitness-startups
Language: English
Published: 2026-03-22
Author: Jaehyuk Kim
# Free MMP for Fitness App Startups: 1 Year of Attribution Without the Financial Risk
Your fitness app has $5,000/month in UA budget and a CAC around $30. You are spending $60,000 a year to acquire users. **An MMP contract at $15,000–$50,000/year would consume 25–80% of that budget — on measurement, not acquisition.**
So you skip the MMP. You run Meta and Google campaigns, check RevenueCat for subscriber counts, and hope the channel that gets the most installs is also the one producing paying users. **That hope is the most expensive assumption in your budget.**
For a fitness app startup at this stage, the question is not whether you need attribution. It is whether you can access it without the financial risk of an annual contract before you know which channels work.
**Key Takeaways**
- **Traditional MMP contracts ($15K–$50K/year) do not fit startup unit economics.** A fitness app with a $30 CAC and $90 LTV cannot justify spending 25%+ of UA budget on measurement tools.
- **Without channel-level attribution, every budget decision is a guess.** You cannot tell whether Meta or Google is producing subscribers — you only see aggregate installs and aggregate revenue.
- **15K free attributed installs ≈ 1 year of attribution for an early-stage fitness app.** At $5K/month spend and $4 CPI, you generate ~1,250 installs/month. 15,000 ÷ 1,250 = 12 months at $0.
- **Free tiers exist in the MMP market.** The question is whether the free tier includes subscription-specific features — RevenueCat S2S, funnel reports, revenue attribution — or only basic install counting.
- **Pay-as-you-go after the free tier means no commitment before proof.** You see which channels work first. You pay only when scale justifies it.
## The MMP Cost Problem for Startups
The economics of traditional MMP pricing are built for established teams, not early-stage startups.
[__wf_reserved_inherit]
Fitness app unit economics set the constraints:
- **CAC is ~$30** — [projected to drop toward $20 by 2030](https://financialmodelslab.com/blogs/kpi-metrics/personal-fitness-mobile-application), but in 2026 the benchmark holds.
- **LTV must be ≥ $90** (3x CAC) to justify UA spend. With ~80% contribution margins and **Pre-revenue startups have $0–$150/month for tools.** An MMP at $15K–$50K/year is not in this budget. It is not even in the conversation.
**The result:** startups that need attribution the most are the ones least able to afford it. They run campaigns without channel-level data and make budget decisions based on CPI alone — which, as any subscription app team learns, [does not predict subscriber quality](https://www.airbridge.io/en/blog/abr-revenuecat-integration).
## What Happens Without Attribution
Without an MMP, a fitness app startup has two data sources that never connect.
- **Ad platforms** show installs, CPI, and click-through rates — by channel, by campaign, by creative. But they do not show which installs became subscribers.
- **RevenueCat** shows subscribers, revenue, trial-to-paid rates, and churn. But it does not show which ad channel produced each subscriber.
You see installs from Meta. You see subscribers in RevenueCat. **You cannot see subscribers from Meta.** The connection is missing.
This means:
- **Budget allocation is based on install volume, not subscriber quality.** The channel with the cheapest CPI gets more budget — even if its trial-to-paid rate is half that of a more expensive channel.
- **You cannot kill underperforming campaigns early.** Without channel-level subscription data, a campaign that produces installs but no subscribers runs for weeks before you notice — burning budget a startup cannot afford to waste.
- **You cannot prove channel ROI to investors.** "We spent $30,000 on Meta and got 180 subscribers" is a fundable story. "We spent $30,000 on ads and got subscribers" is not.
[**_Connect your fitness app's ad channels to RevenueCat. Native S2S. Funnel and Revenue reports by channel. $0.05/install, 15K free._**](https://www.airbridge.io/en/plans/core)
## The 1-Year Free Math
The fear of MMP cost assumes you pay from day one. **With a 15K free attributed installs tier, the math changes entirely.**
A fitness app startup at $5,000/month UA spend with a $4 CPI generates approximately 1,250 installs per month.
- **Month 1–12:** 1,250 × 12 = 15,000 installs. **All attributed. All free.**
- **Month 13+:** At $0.05/install, you pay $62.50/month — 1.25% of your UA budget.
- **If UA scales to $10K/month:** ~2,500 installs/month = 6 months free, then $125/month.
**In the first year, you get the same channel-level attribution that an enterprise team pays $15K–$50K for — at zero cost.** By the time you start paying, you already know where your paying users come from, which campaigns to kill, and whether your Meta spend outperforms your Google spend.
This is not a trial. There is no feature gate. Every report — funnel, revenue, retention — is available during the free tier. **The only limit is volume, not capability.**
## What a Startup MMP Actually Needs
Enterprise MMPs bundle features that startups do not use — fraud detection, raw data export, agency access, dozens of ad network integrations. **A fitness app startup running Meta, Google, Apple Search Ads, and TikTok needs four things.**
- **Attribution across GMAT channels.** Meta, Google, Apple Search Ads, TikTok — these four cover 80–90% of startup UA spend. Integrations with 50+ ad networks are unnecessary at this stage.
- **RevenueCat or Adapty S2S integration.** Subscription events need to flow from your billing platform to your MMP without SDK-only limitations. S2S ensures Start Trial, Subscribe, and Renew events are captured even when the user does not open the app.
- **Subscription funnel and revenue reports.** Install → trial → paid → renewal, broken down by channel. Not just installs by channel and subscribers in aggregate.
- **No annual contract.** A startup's UA budget changes month to month. Committing to a 12-month contract before knowing which channels work is a financial risk that a pay-as-you-go model eliminates.
## How Core Plan Fits Startup Economics
Core Plan is built for exactly this profile — early-stage subscription apps running paid UA across major channels, with a team that cannot design custom event schemas or justify an enterprise MMP contract.
Native S2S integration with [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) is included in the base offering. 25 predefined subscription-optimized standard events eliminate the event naming decisions that slow down setup. **The features that matter for a fitness app startup — funnel reports, revenue attribution, retention by channel — are not locked behind a higher tier.**
### Core Plan vs Traditional MMP: Startup Fit
The difference is not capability — it is accessibility. Both can attribute installs and connect billing data. The question is whether the pricing and packaging fit a startup's constraints.
[__wf_reserved_inherit]
## Start Measuring Before You Start Paying
A fitness app startup cannot afford to spend $60,000/year on UA without knowing which channels produce subscribers. **But it also cannot afford to spend $15,000/year to find out.**
The 15K free installs tier removes this tradeoff. You get channel-level subscription attribution for up to a year. You see which channels work. You prove ROI. **You pay only when scale makes $0.05/install trivial compared to the budget you are optimizing.**
[**_See which channel drives your fitness app's subscribers. 15K free installs. No contract. $0.05/install after. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## MMP vs GA4 for Fitness Apps: Why GA4 Can't Track Trial-to-Paid by Channel
URL: https://www.airbridge.io/en/blog/mmp-vs-ga4-fitness-app
Language: English
Published: 2026-03-22
Author: Jaehyuk Kim
# MMP vs GA4 for Fitness Apps: Why GA4 Can't Track Trial-to-Paid by Channel
Your fitness app's GA4 dashboard shows 3,000 monthly installs. Google Ads is the top source. But when you open RevenueCat, you see 180 paid subscribers — and **GA4 cannot tell you how many came from Meta, how many from TikTok, how many from Apple Search Ads.**
You know the total. You do not know the breakdown. **The channel that looks best in GA4 may not be the channel that produces subscribers.**
This is not a configuration problem. [GA4 was not designed for mobile attribution — it lacks cross-channel deduplication and cannot maintain user-level tracking across ad networks over time.](https://www.measured.com/faq/what-is-the-impact-of-ga4-google-analytics-4/) For fitness apps running trials that convert days later and renewals that happen months later, the gap is structural.
**Key Takeaways**
- **GA4 shows installs from Google Ads. It does not reliably attribute installs from Meta, TikTok, or Apple Search Ads.** Non-Google channels appear as "Direct" or "Unknown."
- **GA4 cannot connect trial starts to paid conversions by channel.** A 7-day trial converts after the session ends. GA4 loses the thread.
- **Health & Fitness trial-to-paid median is 37.7%.** Without channel-level data, you cannot see whether a channel is at 20% or 50% ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)).
- **57% of Health & Fitness subscriptions are annual.** Not knowing which channel drives annual subscribers means misallocating budget away from your highest-LTV users ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)).
- **An MMP with billing integration connects what GA4 cannot** — ad channels to subscription outcomes via S2S.
## What GA4 Shows — And What It Doesn't
For a fitness app running paid UA across multiple channels, GA4 provides one slice of the picture and hides the rest.
[__wf_reserved_inherit]
### **What GA4 can see:**
- Installs and sessions attributed to Google Ads
- In-app events (if configured with Firebase)
- Aggregate user behavior — screens viewed, features used, retention curves
### **What GA4 cannot see:**
- Which Meta, TikTok, or Apple Search Ads campaign drove a specific install
- Trial starts by acquisition channel
- Trial-to-paid conversion rates by channel
- Subscription revenue by channel
- Renewal and churn rates by channel
**The result:** Your GA4 dashboard shows traffic and engagement. Your RevenueCat dashboard shows subscribers and revenue. **No view connects the two — the subscription metrics that matter for budget decisions are invisible in GA4.**
## Four Reasons GA4 Breaks for Fitness App Subscriptions
GA4's limitations for subscription attribution are not missing features. They are architectural constraints. [Even in a post-ATT environment, subscription apps need persistent user-level attribution that GA4 was never built to provide.](https://www.021newsletter.com/p/do-you-still-need-an-mmp-in-2025)
- **Google-only channel attribution.** GA4 natively attributes installs to Google Ads. For Meta, TikTok, and Apple Search Ads, GA4 relies on UTM parameters or manual campaign tagging — which [do not work for mobile app installs](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it). The result: non-Google installs appear as "Direct" or "(not set)." **If 40–60% of your UA spend is on Meta and TikTok, GA4 is blind to half your budget.**
- **Last-touch attribution only.** GA4 offers Last Touch and Data-Driven Attribution — both of which [cannot track view-through conversions](https://www.measured.com/faq/what-is-the-impact-of-ga4-google-analytics-4/). A user sees your fitness app ad on Instagram, searches for it later, and installs from the App Store. **GA4 credits the App Store. The Meta ad that drove the decision is invisible.**
- **No persistent user-level attribution.** [GA4 breaks once revenue becomes delayed.](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it) A fitness app with a 7-day trial converts the subscriber on day 7 or later. GA4 cannot reliably tie that conversion back to the original ad click. **Trials convert days later. Renewals happen months later. GA4 loses the connection at every step.**
- **No billing platform integration.** RevenueCat and Adapty process subscription events — Start Trial, Subscribe, Renew, Cancel — on Apple and Google billing servers. GA4 has no S2S connection to these platforms. **The subscription data that determines your actual ROAS exists in RevenueCat, not in GA4.**
## What Fitness App Teams Actually Need
Fitness and health apps have specific metrics that drive budget decisions. GA4 cannot surface any of them at the channel level.
- **Trial-to-paid by channel.** The Health & Fitness median is 37.7%, with the top 25% at 51.4%+ ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). The gap between median and top quartile is enormous — but you can only optimize for it if you see which channel is at 25% and which is at 50%.
- **Renewal rate by channel.** Health & Fitness apps have the highest active renewal rate among app categories at 86.4% ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). But this is an aggregate. **A channel driving users who renew at 90%+ is worth more than one driving users who churn after month 1** — and this difference is invisible without channel-level renewal data.
- **Annual vs monthly mix by channel.** 57% of Health & Fitness subscriptions are annual ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). Annual subscribers have the highest LTV. **If Channel A drives mostly monthly subscribers and Channel B drives mostly annual subscribers, the budget allocation changes entirely.**
- **Creative-level subscription data.** Which ad creative drives trial starts? Which drives paid conversions? Install-optimized creatives and subscription-optimized creatives are often different — but only visible when subscription events connect back to acquisition data.
[**Connect your fitness app's ad channels to RevenueCat subscription data. Native S2S. Funnel and Revenue reports by channel. $0.05/install, 15K free.**](https://www.airbridge.io/en/plans/core)
## GA4 vs MMP: What Each Can Do for Fitness App Subscriptions
GA4 is a product analytics tool. An MMP is an attribution tool. For fitness apps, the distinction matters at every step of the subscription funnel.
[__wf_reserved_inherit]
**GA4 and an MMP are not competing tools — they answer different questions.** GA4 tells you what users do inside your app. An MMP tells you which ad channel brought the users who pay.
## How Core Plan Bridges the GA4 Gap
Core Plan includes native S2S integration with [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) in the base offering. Subscription events flow into attribution without requiring a higher-tier upgrade.
For fitness app teams currently using GA4 for everything, the addition is specific: **keep GA4 for product analytics. Use Core Plan for acquisition attribution.** The subscription metrics GA4 misses — trial-to-paid by channel, renewal rates by source, revenue by campaign — become visible once RevenueCat data connects to ad channel data through an MMP.
25 predefined subscription-optimized standard events — including Start Trial, Subscribe, and Unsubscribe — map directly to the events RevenueCat sends. Native integrations with Meta, Google, Apple Search Ads, and TikTok cover the channels GA4 misses.
[__wf_reserved_inherit]
## GA4 Shows Traffic. An MMP Shows Revenue by Source.
Your fitness app's GA4 dashboard is not wrong — it is incomplete. It shows installs, sessions, and engagement. **It does not show which channel's installs become subscribers, which subscribers renew, and which channels are worth more budget.**
That gap is not fixable within GA4. It requires a system that receives data from both sides — ad channels and billing platforms — and connects them. **For a fitness app running paid UA across Meta, Google, Apple Search Ads, and TikTok, that system is an MMP.**
[**_See which channel drives your fitness app's subscribers. Native RevenueCat S2S. $0.05/install. 15K free. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## Why Your RevenueCat Numbers Don't Match Your Ad Platform — And What It Costs You
URL: https://www.airbridge.io/en/blog/revenuecat-attribution-data-mismatch
Language: English
Published: 2026-03-21
Author: Jaehyuk Kim
# Why Your RevenueCat Numbers Don't Match Your Ad Platform — And What It Costs You
Meta Ads Manager says your ROAS is 3.2x. RevenueCat says you added 40 subscribers last month at $9.99/month. You multiply 40 by $9.99, get $400 in subscription revenue, and compare it to your $2,000 ad spend. **That is a 0.2x ROAS — not the 3.2x Meta is reporting.**
The numbers do not reconcile because they were never designed to. Meta counts view-through conversions using its own attribution model. RevenueCat records billing events from Apple and Google servers. **These two systems measure different things, at different times, using different definitions of "conversion."**
[The mismatch is not a configuration error. It is built into how ad platforms, MMPs, and subscription billing systems work.](https://www.airbridge.io/en/blog/mmp-numbers-match-ad-platform-cost-subscription-funnel)
**Key Takeaways**
- **A 10–15% install discrepancy compounds to 40–60% at the subscription level.** Each funnel step — install → trial → paid → renewal — [adds its own gap](https://www.airbridge.io/en/blog/mmp-numbers-match-ad-platform-cost-subscription-funnel).
- **Three structural causes drive the mismatch:** self-attributing networks count their own conversions, attribution windows close before trials end, and billing events happen on servers RevenueCat sees but ad platforms do not.
- **AI app ROAS is especially distorted.** Trial-to-paid may look strong, but 30% faster churn means the revenue your ROAS calculation assumes will not materialize.
- **Health & Fitness active renewal rate is 86.4% — highest among app categories** ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)). But without channel-level data, you cannot see which channels drive renewers vs churners.
- **Connecting RevenueCat to ad channels requires an MMP with S2S billing integration** — not more dashboard tabs.
## A 10% Install Gap Becomes a 60% Revenue Gap
At the install level, the discrepancy between your ad platform and your actual data is often 10–15%. Meta reports 1,200 installs. Your MMP shows 1,020. **Most teams accept this as normal — and at the install level, it is.**
[__wf_reserved_inherit]
But subscriptions are not installs. Every step in the funnel — install → trial start → trial-to-paid → first renewal — introduces its own discrepancy layer:
- **Install → Trial:** Not every install starts a trial. The trial start rate differs by channel, but your ad platform does not report trial starts.
- **Trial → Paid:** Conversion happens after the attribution window closes. The ad platform counts it as its own conversion; the billing system does not.
- **Paid → Renewal:** Renewal happens server-side, weeks or months later. No ad platform tracks this.
[A 10–15% discrepancy at the install level can reach 40–60% by the time you compare ad platform revenue claims to actual subscription revenue in RevenueCat.](https://www.airbridge.io/en/blog/mmp-numbers-match-ad-platform-cost-subscription-funnel) **The numbers diverge more as the funnel deepens — not less.**
## Three Structural Reasons the Numbers Never Match
The mismatch between RevenueCat and your ad platform is not a tracking bug. It comes from three architectural layers that each produce their own version of "the truth."
[__wf_reserved_inherit]
- **Self-attribution vs deduplication.** Meta, Google, and TikTok are self-attributing networks — they report their own conversions, including view-through attribution, without cross-channel deduplication. Each SAN only sees its own slice of the funnel and claims credit accordingly. An MMP has visibility across all channels at once — it sees installs from Meta, Google, and TikTok in a single view and deduplicates. **Both are technically correct within their own definitions. The numbers will never match because the definitions do not match.**
- **Attribution window vs trial period.** A 7-day attribution window cannot capture a subscription conversion from a 14-day free trial. The ad platform counts the install as attributed, but the subscription — which happens on day 14 — falls outside the window. **RevenueCat records the subscription. The ad platform already closed the book.**
- **Billing events live on different servers.** Subscription state changes — Start Trial, Subscribe, Renew, Cancel — are processed by Apple App Store and Google Play billing servers. RevenueCat captures these via its SDK and server-side listeners. **Ad platforms have no access to these billing servers.** The result: RevenueCat knows who subscribed, the ad platform knows who clicked — but no system connects the two without an intermediary.
RevenueCat's own documentation [acknowledges data discrepancies](https://community.revenuecat.com/featured-articles-55/about-data-discrepancies-116) driven by IDFA availability, event delivery timing, and SDK configuration differences. **These are not bugs to fix — they are structural realities to work around.**
## The AI App ROAS Illusion
For AI-powered subscription apps, the mismatch is even more deceptive.
Trial-to-paid conversion may look strong — the novelty of AI features drives initial signups. But the downstream metrics tell a different story:
- **Revenue per payer is 41% higher** than non-AI subscription apps — meaning higher price points and more price-sensitive users ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/))
- **Churn is 30% faster** — users who subscribed out of curiosity leave sooner once the novelty fades ([RevenueCat, State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/))
- **ROAS calculations break** because the revenue your ad platform assumes will persist over 12 months decays significantly faster**The problem is not just that ROAS is wrong — it is that you cannot tell which channel brings "AI curiosity subscribers" who churn after month 2 versus subscribers who stay.** Without channel-level renewal data connected to acquisition source, every channel looks equally productive at the trial-to-paid stage and equally invisible at the renewal stage.
## What the Mismatch Actually Costs
When RevenueCat data and ad platform data live in separate systems, every budget decision is based on incomplete information.
- **Budget misallocation.** Three months of pushing $10,000/month toward a channel with inflated ROAS — while starving the channel that actually produces long-term subscribers — costs $30,000 in misdirected spend. **The cost is not the spend itself. It is the subscribers you did not acquire from the better channel.**
- **Invisible renewal rates by channel.** Health & Fitness apps have the highest active renewal rate among all app categories at [86.4%](https://www.revenuecat.com/state-of-subscription-apps-2025/). But this is an aggregate number. **Some channels may drive users who renew at 90%+. Others may drive users who churn after the first billing cycle.** Without RevenueCat subscription data connected to acquisition channels, you cannot see this difference.****
- **Annual vs monthly mix.** Knowing which channel drives annual subscribers — your highest-LTV users — versus monthly subscribers changes your payback calculation entirely. \*\*This data exists in RevenueCat. It is not connected to your ad channels.
[**_Connect RevenueCat subscription data to your ad channels. Native S2S integration. Funnel and Revenue reports by channel. $0.05/install, 15K free._**](https://www.airbridge.io/en/plans/core)
## How to Connect RevenueCat to Your Ad Channels
RevenueCat does not attribute subscriptions to ad channels — that is not what it is designed to do. An MMP does not process billing events — that is not what it is designed to do. **The connection requires both systems to exchange data.**
- **S2S integration between RevenueCat and your MMP.** RevenueCat sends subscription events to your MMP via a server-to-server connection. This ensures billing events are captured regardless of whether the user opens the app.
- **Attribution window ≥ trial period.** Your MMP's lookback window must span the full trial-to-paid conversion cycle. A 7-day trial needs at minimum a 7-day window. Longer trials or grace periods need longer windows.
- **Channel-level subscription reporting.** Once RevenueCat events flow into your MMP, you need reports that break down install → trial → paid → renewal by channel, campaign, and creative — not just installs by channel and subscribers in aggregate.
## What to Look for in an MMP to Connect RevenueCat to Ad Channels
Both traditional MMPs and newer alternatives support RevenueCat integration. The question is whether the features you need are accessible without enterprise pricing. Confirm these capabilities:
- **RevenueCat S2S integration included in base.** Not gated behind a higher pricing tier. Subscription events should flow into attribution from day one.
- **Subscription funnel by channel in base reports.** Install → trial → paid breakdown by acquisition source — without requiring an advanced analytics add-on.
- **Revenue attribution by channel.** Subscription revenue mapped back to the ad campaign that drove the install, included in standard reporting.
- **Predefined subscription events.** Standard events that map directly to RevenueCat's event schema — Start Trial, Subscribe, Unsubscribe — without custom configuration.
- **Subscription-focused setup guidance.** Onboarding designed for subscription app workflows, not general-purpose mobile app setup.
- **Usage-based pricing with no annual lock-in.** Pay per install, scale down when needed, and start with a free tier that includes full feature access.
## The Numbers Will Never Match — But They Can Connect
Meta will always report different numbers than RevenueCat. Google will always count conversions differently than your billing system. **The goal is not to make the numbers match. It is to connect them so you can see what each channel actually produces in subscription revenue.**
The structural gap — self-attribution, timing mismatches, billing silos — is not fixable inside RevenueCat or inside your ad platform. It requires an intermediary that receives data from both sides. **The question is whether that intermediary requires an enterprise contract, or whether it is accessible from day one.**
[**_See which ad channel drives RevenueCat subscribers. Native S2S integration. $0.05/install. 15K free. Start on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## Install, Trial, Subscribe: Which Conversion Event Should Your Fitness App Send to Each Ad Platform?
URL: https://www.airbridge.io/en/blog/app-event-optimization-subscription-apps
Language: English
Published: 2026-03-20
Author: Jaehyuk Kim
Your fitness app generates 200 trial starts per week. That is enough volume for Meta to optimize. But only 70 of those convert to paid subscribers — spread across 4 ad sets, that is 17 Subscribe events per ad set per week. **Meta needs 50.**
So you optimize for Start Trial instead. The algorithm gets enough data. Trial volume grows. But trial-to-paid keeps falling — because the algorithm is finding users who start trials, not users who subscribe.
**This is the core tension of app event optimization for fitness apps:** deeper funnel events carry better signal, but they often fall below the volume threshold that ad platforms need to learn.**Key Takeaways**
- **The event you optimize for is the user you get.** Send Install events, the algorithm finds installers. Send Subscribe events, it finds subscribers — but only if volume is sufficient.
- **Each platform has different volume thresholds.** Meta: [~50 conversions/ad set/week](https://www.facebook.com/business/help/112167992830700). Google: [volume-dependent learning phase](https://support.google.com/google-ads/answer/13020501) (fewer than 5/day extends learning to 14 days). TikTok: [50 conversions or 7 days](https://ads.tiktok.com/help/article/best-practices-for-app-event-optimization).
- **The sweet spot for most fitness apps is the Qualified Trial** — deep enough for strong signal, high enough volume for algorithm learning.
- **The optimal event differs by platform.** Your Meta event strategy and your Google event strategy should not be identical.
- **Airbridge Core Plan shows Install → Trial → Subscribe conversion rates by channel** — so you can validate whether your event strategy is driving actual subscribers, not just the event you optimized for. Start with 15K free attributed installs.
## The Event You Optimize For Is the User You Get
When you select a conversion event for app event optimization, you are not just choosing a metric to track. **You are telling the algorithm which users to find more of.**
### Volume vs Intent: The Core Tradeoff
Every conversion event sits on a spectrum. Shallow events (Install, App Open) generate high volume but carry weak intent signal. Deep events (Subscribe, Renewal) carry the strongest signal but generate the least volume.
Ad platforms need minimum conversion volume to learn. Below that threshold, the algorithm cannot identify patterns in who converts — delivery becomes erratic, CPAs spike, and the ad set enters what Meta's Ads Manager labels ["Learning Limited"](https://www.facebook.com/business/help/269269737396981) — a status indicating the ad set is not getting enough optimization events to exit the learning phase.
**The tradeoff:** optimize for a shallow event and the algorithm learns fast but learns the wrong lesson. Optimize for a deep event and the algorithm learns the right lesson but may never learn at all.
### What Happens When You Optimize for the Wrong Event
A fitness app optimizing for Install on Google will attract users who download the app — many of whom never open it, never start a trial, and never consider subscribing. **The algorithm delivered exactly what you asked for: installs.** It just never learned what a subscriber looks like.
Conversely, optimizing for Subscribe on Meta with only 12 events per ad set per week means the algorithm never exits the learning phase. It cannot build a reliable profile of who subscribes. **You are sending the right signal, but not enough of it.**
The fix is not choosing between volume and intent. **It is finding the deepest event that still meets the platform's volume threshold** — and that event is different for each platform.
[__wf_reserved_inherit]
## Platform-by-Platform: What Each Algorithm Needs
### Meta: 50 Conversions per Ad Set per Week
Meta's optimization tiers for fitness apps:
- **MAI (Mobile App Install):** Optimizes for install volume. Cheapest CPI, weakest subscriber quality. Not recommended as primary optimization for fitness apps.
- **AEO (App Event Optimization):** Optimizes for a specific in-app event — Start Trial, Subscribe, or a [qualified trial](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets). Best when you have enough event volume at a specific funnel stage.
- **VO (Value Optimization):** Optimizes for conversion value/ROAS. Requires at least 30 attributed purchases with 5+ distinct values over 14 days.**The critical threshold:** [**50 optimization events per ad set over 7 days**](https://www.facebook.com/business/help/112167992830700) **to exit the learning phase.** Below this, the ad set enters ["Learning Limited"](https://www.facebook.com/business/help/269269737396981) status — delivery destabilizes and CPAs spike.**Budget constrains event choice.** The 50-event threshold is per ad set, not per account. A fitness app with 200 trials/week sounds healthy — but spread across 5 ad sets, that is only 40 per ad set.
The entire account stays in Learning Limited. To hit the threshold, either consolidate ad sets or move to a shallower event. This also means your daily budget must support at least 50 × target CPA per week per ad set — low budgets mathematically prevent deep-event optimization.
**For most fitness apps:** Start with AEO on Qualified Trial. Graduate to VO when you consistently hit 30+ purchases/week with 5+ distinct values.
### Google: 10 Events per Day per Campaign
Google App Campaigns work differently from Meta. The main optimization paths:
- **tCPA (target Cost per Action):** Set a target cost per trial start or subscription. Standard for most fitness apps. Budget minimum: 10x your target CPA.
- **tROAS (target Return on Ad Spend):** Optimizes for conversion value. Requires sufficient purchase/revenue data via Firebase. Best for apps with varied subscription tiers.
Google's [learning phase duration scales with volume](https://support.google.com/google-ads/answer/13020501): campaigns receiving fewer than 5 conversions/day can take up to 14 days to exit learning, while those above 100/day exit in 1-3 days. **As a practical guideline, aim for at least 10 in-app events per day per campaign** for stable performance.**For most fitness apps:** Use Trial Started as the core conversion event with tCPA. This generates enough volume for Google's algorithm while correlating to subscription intent. Graduate to tROAS when you have sufficient revenue data flowing through Firebase.
### TikTok: 50 Conversions or 7 Days
[TikTok's AEO](https://ads.tiktok.com/help/article/app-event-optimization?lang=en) finds users who will install AND perform a specific in-app event. The learning phase runs until **50 conversions OR 7 days**, whichever comes first.
- **AEO:** Optimize for Subscribe or Purchase events. Do not adjust bids by more than 10% during the learning phase.
- **VBO (Value-Based Optimization):** Requires 30+ purchase events with value in 7 consecutive days. **Currently Android-only.**
**For most fitness apps:** Use Subscribe or Purchase for AEO. VBO when volume allows — but note the Android-only limitation.
> [**See how each conversion event performs across Meta, Google, and TikTok — from install to subscription. Start with 15K free attributed installs.**](https://www.airbridge.io/en/plans/core)
## The Subscription Event Hierarchy: Where Volume Meets Intent
Not every event in the subscription funnel is suitable for optimization. Here is how the conversion event strategy breaks down:
[__wf_reserved_inherit]
**The rule: optimize for the deepest funnel event that still meets the platform's volume threshold.**
For a fitness app generating 200 trials/week: Qualified Trial (filtering ~30% impulse cancellers) yields ~140 events/week — enough for 2-3 ad sets on Meta. Subscribe at 70/week only works with 1 ad set. **The event choice directly constrains your campaign structure.**
## A Practical Event Map for Fitness Apps
[__wf_reserved_inherit]
[__wf_reserved_inherit]
**Common mistakes to avoid:**
- **Same event on all platforms.** Meta and Google have different volume requirements and optimization mechanics. What works on Meta may starve Google of data, or vice versa.
- **Raw trials instead of qualified trials.** [Impulse cancellers contaminate the signal](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets). A 2-4 hour delay filters them out before the event reaches the ad platform.
- **Optimizing for Install when you have trial volume.** If you generate 50+ trials per ad set per week, there is no reason to optimize for Install. You are leaving signal quality on the table.
## How to Validate Your Event Strategy With Channel-Level Data
Changing your optimization event is a hypothesis. If you switch from Start Trial to Qualified Trial on Meta, you need to know: did CPS improve, or did volume drop too far?
### What to Measure After Changing Your Optimization Event
- **Conversion rate at each funnel stage by channel** — Install → Trial → Subscribe. If trial-to-subscribe improved after switching events, the new signal is working.
- **CPS by channel over time** — the primary metric. Track weekly for 4-8 weeks after the change.
- **Event volume by ad set** — verify that the new event meets the platform's learning threshold. If not, consolidate ad sets or move one level up in the event hierarchy.**This requires an** [**attribution system**](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) **that tracks the full funnel — install to trial to subscription — by channel.**
### How Airbridge Core Plan Connects Events to Subscription Outcomes
Core Plan tracks Install, Start Trial, and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. The Actuals Report shows conversion rates at each funnel stage by channel — **so you can see whether your event strategy is actually driving subscribers, not just the event you optimized for.**
With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, subscription events flow into attribution automatically. **Before changing your optimization event, this data shows which channels have the widest gap between trials and subscriptions. After the change, it validates whether the new event improved CPS.**
### Airbridge Core Plan vs Traditional MMP
[__wf_reserved_inherit]
## The Conversion Event Is an Instruction — Send the Right One
The event you send to Meta, Google, and TikTok is not a tracking configuration. **It is an instruction that tells the algorithm which users to find.** Send installs, get installers. Send raw trials, get impulse cancellers. Send qualified trials, get engaged users.
The right event depends on your platform, your volume, and your budget. The only way to validate your choice is to measure the full funnel — install to trial to subscription — by channel.
> [**See how each event drives actual subscribers across your channels. Start with 15K free attributed installs on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## MMP Time to Value: Why MarTech TTV Is 44 Hours — And What It Costs You
URL: https://www.airbridge.io/en/blog/mmp-time-to-value
Language: English
Published: 2026-03-20
Author: Jaehyuk Kim
Most MMP evaluations start with feature lists and pricing. They should start with one question: **how long until this tool tells me whether my campaigns are working?**
That question has a name — Time to Value. And for MarTech, [the answer is worse than almost any other software category](https://userpilot.com/blog/time-to-value-benchmark-report-2024/). **MarTech products take an average of 44 hours to deliver first value — the second-longest TTV in SaaS, with the lowest onboarding completion rate of any category.**
For UA teams running paid campaigns, every hour between "we signed up" and "we can see which channels convert" is an hour of ad spend without attribution. That is not a setup inconvenience. **It is budget burned without visibility.**
**Key Takeaways**
- **Time to Value — not LTV prediction — is the metric that determines whether your MMP is useful in your first campaign cycles.** LTV models take months to validate. TTV tells you now.
- **MarTech has the second-longest TTV in SaaS** at [44 hours on average](https://userpilot.com/blog/time-to-value-benchmark-report-2024/), with the lowest onboarding completion rate of any category.
- **65.7% of organizations report data integration as their top MarTech challenge.** [34% cite training gaps](https://chiefmartec.com/wp-content/uploads/2024/05/state-of-martech-2024-report.pdf), and 32% underutilize their current stack.
- **MMP TTV is driven by friction outside the SDK** — contract processes, event schema design, and feature complexity add more delay than technical integration.
- **Airbridge Core Plan is designed for fast TTV.** Self-serve signup, 25 predefined subscription events, GMAT-only channel focus. $0.05/install, 15K free.
## TTV Over LTV: The Metric UA Teams Should Be Watching
When evaluating an MMP, teams typically ask about LTV prediction, multi-touch attribution, and reporting depth. These matter — eventually. **But none of them determine whether the tool delivers value in your first weeks.**
Time to Value measures the gap between adopting a tool and getting the first actionable insight from it. For an MMP, that means: **how long from signup to "I can see which channel drove the most trial-to-subscription conversions"?**
LTV models require months of subscription data to validate. You need multiple renewal cycles, churn curves, and cohort maturity before LTV predictions become reliable. By the time your LTV model is trustworthy, you have already spent months of ad budget based on incomplete data.
TTV flips the question. Instead of "how well will this tool predict long-term revenue?", it asks: "**how fast can this tool show me what is working right now?**" For early-stage UA teams — where campaign budgets are limited and every week of blind spend matters — TTV is the more actionable metric.
## MarTech Has the Second-Longest TTV in SaaS
[Userpilot's 2024 benchmark](https://userpilot.com/blog/time-to-value-benchmark-report-2024/) measured Time to Value across SaaS categories. The results for MarTech are not encouraging.
[Bar chart comparing TTV across SaaS categories. SaaS average at 36 hours. MarTech at 44 hours — second-longest, with lowest onboarding completion rate.]
- **SaaS average TTV:** 36 hours (1 day, 12 hours, 23 minutes)
- **MarTech TTV:** 44 hours (1 day, 20 hours, 47 minutes) — **second-longest in SaaS**
- **MarTech onboarding completion rate:** lowest of any category**This is not just a MarTech problem — it is a compounding one.** [ChiefMartec's 2024 State of MarTech report](https://chiefmartec.com/wp-content/uploads/2024/05/state-of-martech-2024-report.pdf) found that:
- **65.7% of organizations** experience data integration difficulties
- **34%** report team training and experience gaps
- **32%** say they underutilize the features they already pay for
The pattern: MarTech tools are complex to set up, hard to learn, and most teams never use the full capability. **For MMPs specifically, this means teams are paying for features they will never configure — while waiting longer to get the one insight they need.**
## What 44 Hours of TTV Actually Costs You
44 hours is not just a setup delay. **It is a strategic cost.**
Campaign cycles do not pause while you onboard. If your UA team launched campaigns on Monday, those campaigns are spending budget and generating installs whether your MMP is ready or not. Every install that arrives before attribution is active is an install you cannot tie to a channel, a creative, or a cost.
What this looks like in practice:
- **First week of a new channel test:** No attribution data. You are making bid adjustments based on platform-reported numbers — which cannot deduplicate, cannot track post-install events, and cannot show subscription conversions.
- **First budget review:** The team cannot answer "which channel drove the most subscribers?" because the MMP is still in setup. The answer defaults to "we don't know yet."
- **First month-end report:** Partial data. The first 1–2 weeks of installs have no MMP attribution. **You are reporting on an incomplete picture.**
The cost of slow TTV is not the setup time itself — **it is the decisions you cannot make while you wait.** [Slow onboarding directly correlates with lower activation rates](https://www.agilegrowthlabs.com/blog/user-activation-rate-benchmarks-2025/), and in the MMP context, activation means the moment you can act on attribution data.
## The MMP TTV Equation: What Adds Up
MMP Time to Value is not one step. It is a sequence — and not all steps contribute equally.
[TTV equation breakdown showing steps: Sales Process and Contract (friction — varies by MMP), SDK Integration and Channel Setup and QA Testing (required — same for all MMPs), Event Schema Design and Feature Navigation (friction — varies by MMP). Friction steps are highlighted in red.]
**Steps every MMP requires (unavoidable):**
- SDK integration into your app
- Channel credential setup (Meta, Google, Apple Search Ads, TikTok)
- QA testing and validation
- First data collection period
**Steps that vary by MMP (friction):**
- **Sales process and contract negotiation:** Demo calls, pricing negotiation, legal review, annual commitment. These can take days to weeks before technical setup even begins.
- **Event schema design:** Deciding which events to track, naming conventions, parameter definitions. The more flexibility the MMP offers, the more decisions your team needs to make upfront.
- **Feature navigation:** Most MarTech platforms ship capabilities far beyond what a subscription app needs. For MMPs, this means navigating features — fraud detection, raw data export, agency access — that your team does not need yet, but that add to the onboarding surface area.**The unavoidable steps — SDK, channels, testing — are roughly the same across all MMPs.** The TTV gap lives in what surrounds them: the purchase process, the schema decisions, and the feature complexity.
> [**Reduce MMP friction. Self-serve signup, predefined events, subscription-focused. $0.05/install, 15K free.**](https://www.airbridge.io/en/plans/core)
## How Core Plan Reduces MMP TTV
Core Plan is built around one principle: **keep the technical steps every MMP requires, strip everything else.** The friction layers identified above — purchase process, schema decisions, feature complexity — are the variables Core Plan targets.
The result is a self-serve MMP where you sign up, integrate the SDK, connect your channels (GMAT — Meta, Google, Apple Search Ads, TikTok), and start collecting data. No demo calls. No contract negotiation. **No decisions about which events to track —** [**25 subscription-optimized standard events**](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) **(Start Trial, Subscribe, Unsubscribe, etc.) are predefined**, reducing schema design work without removing the implementation step.
Features that add onboarding scope without value for early-stage subscription apps — fraud detection, raw data export, agency access — are intentionally excluded. **The platform surface area is smaller because it is designed that way**, not because it is missing capability.
Native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) S2S integration is included in the base — subscription events flow into attribution without additional tier upgrades.
### Core Plan vs Traditional MMP: TTV Factors
For UA teams where every day without attribution data is a day of budget without visibility, the path to first insight matters.
| TTV Factor | Traditional MMP(Typical) | Airbridge Core Plan |
| --- | --- | --- |
| Purchase process | Sales demo,contract negotiation | Self-serve signup,no contract |
| Event configuration | Predefined + custom events | 25 subscription-optimizedstandard events |
| Feature surface area | Full platform (fraud,raw export, agency) | Subscription-focused only —reduced onboarding scope |
| Billing integration | Native (included inhigher tiers) | Native RevenueCat/Adaptyvia S2S, included in base |
| Pricing model | Annual contract,$10K–$50K/year | Pay-as-you-go,$0.05/install |
| Free tier | Limited or none | 15K free attributed installs.All features included |
## Your Campaigns Are Running. Your TTV Clock Is Ticking.
MarTech's 44-hour TTV is an industry average — and for MMPs with annual contracts and enterprise onboarding, the real number is often higher. **Every hour between "we need attribution" and "we have attribution" is an hour of ad spend without visibility.**
TTV is not a setup metric. **It is a business metric.** It determines how fast your team can answer the question that matters: which channels are converting into subscribers?
If your UA team is running campaigns now — or planning to launch soon — the MMP that gets you to first insight fastest is the one that saves you the most budget.
> [**Self-serve MMP attribution. 25 predefined events. $0.05/install. 15K free. Start now on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## AI Fitness App Costs: LLM Inference Drops 10x Per Year — Your MMP Bill Doesn't
URL: https://www.airbridge.io/en/blog/ai-fitness-app-cost-stack
Language: English
Published: 2026-03-20
Author: Jaehyuk Kim
The cost of running a frontier-class LLM dropped from [$20 to $0.40 per million tokens in two years](https://a16z.com/llmflation-llm-inference-cost/). That is a 50x reduction — faster than the PC revolution, faster than dotcom-era bandwidth.
If you are building an AI fitness app, this is the trend powering your product. **LLM inference — the single largest new cost layer for AI apps — is falling at a rate the industry has never seen.**
But not everything in your cost stack is falling. Your cloud bill scales with usage. Your CDN bill scales with video content. And your MMP bill? **It is either a fixed annual contract or a per-conversion fee that scales with success — and it has not dropped at all.**
For AI fitness apps where [the market is projected to reach $46.1B by 2034](https://www.insightaceanalytic.com/report/ai-in-fitness-and-wellness-market/2744), understanding which costs are falling, which are scaling, and which are fixed is the difference between extending your runway and burning through it.
**Key Takeaways**
- **AI fitness app infrastructure has a new cost layer that traditional apps do not: LLM inference.** AI-powered coaching, workout generation, and personalization require per-user API costs of [$100–$800/month](https://www.groovyweb.co/blog/fitness-app-development-cost-2026) — on top of cloud and CDN.
- **Inference costs are dropping at** [**10x per year**](https://a16z.com/llmflation-llm-inference-cost/)**.** Frontier-model performance went from $20 to $0.40 per million tokens. This cost will keep falling.
- **MMP costs are not dropping.** Traditional MMP pricing locks you into fixed commitments ($10K–$50K/year) that do not scale down when your install volume drops — creating a rigid line item in a stack where everything else is variable.
- **AI app subscribers churn 30% faster than non-AI app subscribers** despite generating 41% higher revenue per payer. This shortens the payback window — and fixed MMP costs make it harder to hit.
- **Airbridge Core Plan turns MMP cost from fixed to variable.** $0.05/install, pay-as-you-go, no long-term commitment. 15K free attributed installs.
## The AI Fitness App Cost Stack
Traditional fitness apps have a straightforward infrastructure cost: cloud hosting, maybe a CDN for video content, and third-party services like billing and attribution.
[Three cost categories for AI fitness apps: Declining (LLM inference, dropping 10x/year), Variable (cloud, CDN, billing — scales with usage), and Fixed (MMP annual contract — same bill regardless of performance).]
**AI fitness apps add a cost layer that did not exist before: LLM inference.** Every time a user asks the AI coach for a workout plan, every personalized recommendation, every natural language interaction — it costs money. Not a fixed cost. A per-user, per-session cost that scales with engagement.
[A typical AI fitness app's monthly operating costs](https://www.groovyweb.co/blog/fitness-app-development-cost-2026):
| Cost Category | Monthly Range | Scales With |
| --- | --- | --- |
| Cloud hosting (AWS/GCP) | $300–$3,000 | Users, data |
| AI/LLM API calls | $100–$800 | Sessions, queries |
| Video CDN | $500–$3,000 | Content, views |
| Billing platform (RevenueCat/Adapty) | Usage-based | Subscribers |
| MMP (attribution) | $500–$4,000+ | Installs or fixed annual |
**Every cost in this stack except the MMP scales with actual usage.** Cloud costs go up when users increase, down when they decrease. LLM API costs follow session volume. CDN costs track content delivery. **The MMP — if it is a locked-in commitment — stays the same whether you have 5,000 installs or 50,000.**
## The Good News: Inference Costs Are in Free Fall
The cost line that AI fitness app founders worry about most — LLM inference — is the one falling fastest.
[Line chart showing frontier-model LLM inference cost dropping from $20 to $0.40 per million tokens (50x in 28 months) while MMP annual contract cost stays flat.]
[Andreessen Horowitz's analysis](https://a16z.com/llmflation-llm-inference-cost/) shows that **the cost of frontier-model inference dropped from $20 to $0.40 per million tokens** between March 2023 and July 2025. That is a 50x reduction in 28 months. [Epoch AI's data](https://epoch.ai/data-insights/llm-inference-price-trends) confirms the trend: **prices for equivalent model performance are dropping 10x per year**, with some benchmarks showing up to 900x annual reduction.**What this means for your cost stack:**
- The $100–$800/month AI API line item today will likely be $10–$80 within a year at the same usage level
- Per-user inference cost is approaching a point where it becomes negligible relative to subscription revenue
- **The cost layer that makes AI apps structurally different from traditional apps is rapidly becoming a non-issue**
This is the tailwind. The technology cost of being an AI fitness app is converging toward zero.
## The Problem: MMP Costs Are Not Falling
While inference costs drop an order of magnitude annually, MMP pricing has not changed structurally. **Traditional MMPs charge either a fixed annual contract or a per-conversion fee — and neither model has benefited from the kind of cost deflation that AI infrastructure has seen.**
The standard MMP pricing structure:
- **Annual contract: $10K–$50K/year.** Committed upfront regardless of actual usage. If your app scales faster or slower than projected, the bill stays the same.
- **Per-conversion fee: $0.03–$0.07/install.** Scales with volume, but the rate does not drop as the industry matures.
- **Add-on costs: $10K–$50K/year** for fraud detection, raw data export, and premium support — priced separately from the base.**For an AI fitness app spending $1,500/month on LLM inference, a $20K/year MMP commitment represents more than the entire AI cost stack.** And unlike inference costs, there is no trend line suggesting MMP pricing will halve next year.
## Why This Hits AI Apps Harder: The Payback Window Problem
AI fitness apps have a unique revenue profile. Industry benchmarks show that **AI app subscribers generate 41% higher revenue per payer than non-AI app subscribers — but churn 30% faster.**
[Side-by-side comparison: Traditional app has base revenue with longer retention. AI app has 41% higher revenue but 30% faster churn, creating a tighter payback point — fixed MMP costs eat further into that margin.]
This creates a specific economic constraint: **the payback window is shorter.** You earn more per subscriber, but you have less time to recoup the acquisition cost before they leave.
When the payback window is tight, every fixed cost matters:
- **LLM inference:** falling rapidly — this cost is solving itself
- **Cloud/CDN:** variable, scales with usage — controllable
- **MMP (fixed commitment):** locked in regardless of performance — **not controllable**
**A fixed MMP cost does not adjust when campaigns underperform, when a channel gets paused, or when seasonal dips reduce install volume.** It is a constant burn against a payback window that is already compressed by AI app churn dynamics.
## Controllable Costs vs Fixed Costs
The AI fitness app cost stack has three types of costs:
- **Declining costs:** LLM inference. Falling at historic rates. Will become negligible.
- **Variable costs:** Cloud, CDN, billing platform. Scale with usage. Go up when you grow, down when you contract.
- **Fixed costs:** Locked-in MMP commitments. Same bill regardless of performance. Do not benefit from any market trend.**The strategic move is to convert fixed costs into variable costs wherever possible.** Variable costs preserve cash when campaigns underperform. Fixed costs burn cash regardless.
[You need an MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) — platform dashboards cannot deduplicate conversions or show cost per subscriber by channel. **The question is whether your MMP has to be a fixed cost.**
[**_Turn attribution from a fixed cost into a variable one. $0.05/install, pay-as-you-go, 15K free attributed installs._**](https://www.airbridge.io/en/plans/core)
## How Core Plan Makes Attribution a Variable Cost
Core Plan charges $0.05 per attributed install. **No long-term commitment. No minimum volume. The bill scales with actual installs — just like every other variable cost in your stack.**
**What this means for an AI fitness app:**
- **15K free attributed installs.** The first months of UA testing cost nothing for attribution — you can validate channel performance before attribution adds to your burn rate.
- **Cost scales down when you scale down.** Pause a channel, reduce spend, hit a seasonal dip — your MMP bill drops proportionally. A locked-in commitment does not.
- **Native** [**RevenueCat**](https://www.airbridge.io/en/blog/abr-revenuecat-integration) **and** [**Adapty**](https://www.airbridge.io/en/blog/airbridge-adapty-integration) **S2S integration.** Subscription events flow into attribution automatically — critical for tracking the payback window that AI apps need to monitor closely.
- **Funnel, Retention, and Revenue reports by channel.** See which channels produce subscribers who retain long enough to pay back acquisition cost — the metric that matters most when your payback window is compressed.
### Airbridge Core Plan vs Traditional MMP
For AI fitness apps where every dollar of fixed cost compresses an already-tight payback window, the pricing model matters as much as the feature set.
| Capability | Traditional MMP(Typical) | Airbridge Core Plan |
| --- | --- | --- |
| Pricing model | Annual contract,$10K–$50K/year | Pay-as-you-go,$0.05/install |
| Cost when installs drop | Same (fixed contract) | Drops proportionally |
| Free tier | Limited or none | 15K free attributed installs.All features included |
| Billing integration | Native (included inhigher tiers) | Native RevenueCat/Adaptyvia S2S, included in base |
| Event configuration | Predefined + custom events | 25 subscription-optimizedstandard events |
| Contract requirement | 12–24 month commitment | No annual contract.Cancel anytime |
## Your Inference Costs Are Falling. Your MMP Bill Should Not Be the Cost That Stays.
The AI fitness app cost stack is restructuring itself. Inference costs are in free fall. Cloud and CDN costs scale with usage. **Attribution is the one cost that does not need to be fixed — but has been, because that is how the MMP industry has always priced it.**
If your AI fitness app generates $100–$800/month in LLM costs that are falling every quarter, a $20K/year MMP commitment that never drops is the wrong cost structure. \*\*Turn it into a variable cost. Pay for what you use. Scale down when you need to.
[**_Attribution at $0.05/install. 15K free. No annual contract. Pay-as-you-go on Airbridge Core Plan._**](https://www.airbridge.io/en/plans/core)
---
## The MMP "Success Tax": Why Per-Conversion Fees Kill Fitness App ROI at Scale
URL: https://www.airbridge.io/en/blog/mmp-conversion-fee-roi
Language: English
Published: 2026-03-19
Author: Jaehyuk Kim
Your fitness app hit 50,000 installs last month. Your team scaled Meta and Google campaigns. Trials are converting. MRR is climbing.
Then you open the MMP invoice: **$3,500.**
Next month, installs hit 75,000. The invoice: **$5,250.** The month after that, 100,000 installs. **The invoice increased to $7,000.** Your MMP bill is growing at the same rate as your success — and unlike UA spend, it does not generate any additional revenue. **It is a tax on growth.**
This is the structural problem with per-conversion MMP pricing. The better your campaigns perform, the more you pay for the tool that measures them. And for fitness apps where [average ARPU is $25.78](https://www.businessofapps.com/data/fitness-app-market/), every dollar of attribution cost directly compresses the margin between acquisition cost and subscriber value.
**Key Takeaways**
* **Per-conversion MMP pricing creates a "success tax."** The more installs your campaigns generate, the higher your attribution bill — with no additional value delivered per conversion.
* **At $0.07/conversion, attribution costs reach [$42K–$84K/year](https://checkthat.ai/brands/appsflyer/pricing) at 50K–100K monthly installs** — a significant line item for fitness apps with $25.78 average ARPU.
* **The pricing structure is regressive.** Startups and growth-stage apps pay the highest per-unit rate ($0.07), while enterprises negotiate down to $0.03–$0.05. **The companies with the least budget pay the most per conversion.**
* **Add-on fees compound the base rate.** Fraud detection, audience segmentation tools, and premium support are priced separately — adding $10K–$50K/year on top of per-conversion charges.
* **Airbridge Core Plan charges $0.05/install with no add-ons.** 15K free attributed installs. Same rate at any volume. Attribution cost stays below 0.2% of subscriber LTV.
## How Per-Conversion MMP Pricing Works
Most traditional MMPs charge per attributed conversion — typically per install or per re-engagement. **The rate depends on your scale, but the direction is always the same: more success, higher bill.**
The standard structure:
* **Growth tier: ~$0.07 per conversion.** This is the rate most early-stage and growth-stage apps pay. No volume discount. No ceiling.
* **Enterprise tier: $0.03–$0.05 per conversion.** Available at [1M+ annual conversions](https://checkthat.ai/brands/appsflyer/pricing), typically requiring a 12–24 month commitment.
* **Free tier: limited.** Often capped at 12,000 lifetime conversions — roughly one month of moderate UA activity.
**The regressive dynamic:** a startup generating 20,000 installs/month pays $0.07 each. An enterprise generating 200,000 installs/month negotiates to $0.04. **The company with 10x the volume pays 43% less per unit.** The pricing penalizes the exact stage where companies are most budget-constrained.
And the base rate is only the beginning. Traditional MMPs layer additional costs on top:
* **Fraud detection:** $10K–$30K/year as a premium add-on
* **Audience segmentation tools:** Custom pricing, often a "significant part of an Enterprise contract"
* **Premium support:** Additional fee for dedicated CSM access
**The total cost of attribution at 50K monthly installs is not $42K/year. It is $42K + $10K–$50K in add-ons — potentially $52K–$92K/year** for a fitness app that may be generating $500K–$1M in total annual revenue.
## The Math: What $0.07/Conversion Costs a Fitness App
The [fitness app market generated $3.4B in revenue in 2025
[Annual attribution costs comparison: traditional MMP ($0.07/conversion) vs Core Plan ($0.05/install)]](https://www.businessofapps.com/data/fitness-app-market/), with average ARPU of $25.78. For subscription fitness apps, the unit economics are straightforward — and per-conversion attribution fees take a meaningful bite.
**At $0.07/conversion (standard growth tier):**
| Monthly Installs | Monthly MMP Cost | Annual MMP Cost |
|:-----------------|:-----------------|:----------------|
| 10,000 | $700 | $8,400 |
| 25,000 | $1,750 | $21,000 |
| 50,000 | $3,500 | $42,000 |
| 100,000 | $7,000 | $84,000 |
**At $0.05/conversion (Core Plan):**
| Monthly Installs | Monthly MMP Cost | Annual MMP Cost |
|:-----------------|:-----------------|:----------------|
| 10,000 | $500 | $6,000 |
| 25,000 | $1,250 | $15,000 |
| 50,000 | $2,500 | $30,000 |
| 100,000 | $5,000 | $60,000 |
*At 10% install-to-subscribe and $25.78 ARPU, attribution consumes 3.2% of subscriber revenue at $0.07 — and 2.3% at $0.05. This ratio is constant regardless of volume.*
**The difference at 50K installs: $12,000/year.** At 100K installs: $24,000/year. And these numbers exclude the add-on costs that traditional MMPs layer on top.
## Why This Kills Scaling Confidence
The math problem is clear. But the operational problem is worse: **per-conversion fees make attribution costs unpredictable.**
When you plan a UA budget, you can estimate CPI, set daily spend caps, and control channel allocation. But attribution costs are a function of campaign *success* — which is exactly what you cannot predict in advance.
This creates a specific psychological barrier:
* **Fear of scaling.** If a campaign is working, scaling it means increasing both UA spend *and* MMP costs simultaneously. The attribution bill becomes an uncontrolled variable.
* **ROI calculation breaks.** You can calculate ROAS against UA spend. But when attribution costs scale proportionally, the effective CAC includes a hidden component that changes every month.
* **Budget defense becomes harder.** Explaining to leadership why the measurement tool costs $84K/year — for a fitness app — is a conversation most growth leads would rather avoid.
As a growth marketer at a fitness app shared with our team: *"You can't scale out when your MMP charges on every conversion. The bill grows faster than you can prove the ROI."*
> [Attribution that costs less than 2.5% of subscriber revenue. $0.05/install, 15K free, no annual contract. Start free.](https://www.airbridge.io/en/plans/core)
## What Affordable Attribution Should Cost
**Attribution is infrastructure.** Like hosting or analytics, it should scale with your business without becoming a growth constraint. The question is not whether to pay for attribution — [you need an MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) because platform dashboards cannot deduplicate conversions. **The question is what percentage of revenue attribution should consume.**
A reasonable benchmark:
* **Below 2.5% of attributed subscriber revenue.** If your attribution tool costs more than 2.5% of the revenue it helps you measure, the measurement layer is too expensive relative to the insight it provides.
* **Flat rate regardless of volume.** The same per-unit cost at 10K installs and 100K installs. No regressive pricing that penalizes growth-stage companies.
* **No add-on costs above the base rate.** Reports, integrations, and attribution rules included. The per-unit price should be the total price.
## How Core Plan Keeps Attribution Below 2.5% of Revenue
Core Plan charges $0.05 per attributed install. **This is a per-conversion fee — the same model traditional MMPs use.** The difference is in the rate, the structure, and what is included.
**Why $0.05 works for fitness apps:**
At 10% install-to-subscribe conversion and $25.78 ARPU, every 1,000 installs generates ~100 subscribers producing ~$2,578 in revenue. Attribution cost for those 1,000 installs: **$50 — or 1.9% of subscriber revenue.** At scale, this ratio stays constant because the rate does not change.
**What makes it structurally different:**
* **Same rate at any volume.** No enterprise tier required for a lower rate. A startup at 10K installs and a growth-stage app at 100K installs pay the same $0.05.
* **15K free attributed installs.** This effectively covers the first months of UA testing — before the team has enough data to know whether scaling is justified.
* **Funnel, Retention, and Revenue reports included.** Native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) S2S integration. 25 standard events. Meta, Google, Apple Search Ads, and TikTok pre-integrated. Everything in the base rate.
* **No annual contract.** If the math does not work at your scale, you can leave. No [12-month commitment trapping you with a tool that caps your growth](https://www.airbridge.io/en/blog/is-your-mmp-capping-your-growth-5-features-to-audit-before-your-2026-renewal).
### Airbridge Core Plan vs Traditional MMP
The rate difference is only part of the equation. Here is what changes when the total cost structure is compared side by side.
| Capability | Traditional MMP (Typical) | Airbridge Core Plan |
|:-----------|:--------------------------|:-------------------|
| Per-conversion rate | $0.07/conversion (growth tier) | **$0.05/install** |
| Free tier | 12K lifetime conversions | **15K free attributed installs. All features included** |
| Add-on costs | Fraud, audience segmentation, support priced separately ($10K–$50K/yr) | **No add-ons. All features included** |
| Onboarding | General-purpose onboarding | **Subscription-app focused onboarding, 25 standard events** |
| Contract requirement | 12–24 month commitment for enterprise rate | **No annual contract. Pay-as-you-go** |
| Attribution cost as % of subscriber revenue | ~3.2% (before add-ons) | **~2.3% (all features included)** |
## Your MMP Should Not Be Your Biggest Line Item After UA
Attribution exists to help you spend UA budget more efficiently. **When the measurement tool itself becomes a significant cost center, the economics are inverted.** Your MMP bill should be a rounding error on your revenue — not a line item that competes with your ad spend.
**[Attribution at $0.05/install. 15K free. No add-ons. No annual contract. Start free with Airbridge Core Plan.](https://www.airbridge.io/en/plans/core)**
---
## Why MMP Pricing Lacks Transparency: 3 Pricing Traps to Avoid
URL: https://www.airbridge.io/en/blog/mmp-pricing-transparency
Language: English
Published: 2026-03-19
Author: Jaehyuk Kim
Most [Mobile Measurement Partners (MMPs)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) do not publish pricing on their websites. A startup spending $5,000 to $20,000 per month on paid acquisition has to book a sales call, sit through a demo, and negotiate a custom quote before learning whether the tool fits their budget.
**The cost of this opacity compounds quickly.** Teams delay attribution adoption by weeks or months while evaluating opaque options, running campaigns without measurement in the interim. Every week of blind spending is a week of budget flowing to channels that may not convert trial users into paying subscribers.
**Key Takeaways**
* **Most MMPs hide pricing behind sales calls.** Subscription app teams cannot compare costs or plan budgets without committing to a lengthy evaluation process.
* **Hidden add-ons inflate the real cost beyond the base fee.** Raw data export, fraud detection, and premium support are often priced separately and disclosed after the initial conversation.
* **Annual prepayment is the industry default.** Startups are asked to predict 12 months of usage before they have 3 months of campaign data.
* **Health and fitness apps face seasonal cost risk.** Fixed annual contracts do not accommodate January resolution spikes or summer surges, leading to overpayment during low-volume months.
* **Airbridge Core Plan publishes pricing and uses pay-as-you-go billing.** 15K free attributed installs, $0.05 per install after, with no annual commitment.
## Why MMP Pricing Is Still Opaque in 2026
MMP pricing transparency is one of the biggest evaluation blockers for early-stage subscription app teams. The problem is structural: most enterprise MMPs were built for large advertisers spending millions per year, and their pricing models reflect that origin.
### 1. The "Contact Sales" Default
This evaluation barrier is not just an inconvenience. For a startup with 2 to 5 people, the cycle stretches from days to weeks while [paid acquisition campaigns](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution) run without attribution. A growth marketer spending $10,000 per month on Meta and Google cannot wait 2 to 3 weeks for a pricing quote when the cost of misallocated spend compounds daily.
### 2. Hidden Add-Ons Beyond the Base Fee
Even after the sales call, the initial quote often represents only part of the total cost. Enterprise MMPs typically charge separately for capabilities that early-stage teams may assume are included:
| Add-On | Typical Cost Impact | Do Early-Stage Teams Need It? |
|:-------|:-------------------|:-----------------------------|
| Raw data export (S3, BigQuery) | $500 to $2,000/month | Rarely before Series A |
| Fraud detection | $1,000 to $3,000/month | Rarely at low volume |
| Premium support / CSM | $500 to $1,500/month | Often needed, but priced as upsell |
| Additional ad network integrations | Per-network fees | Only if running beyond Meta/Google/Apple/TikTok |
Platform fees for mid-scale subscription apps can reach $7,000 to $10,000 per month before implementation costs and premium add-ons are factored in ([Business of Apps, 2026](https://www.businessofapps.com/marketplace/app-analytics/mobile-measurement-partners/)). For teams spending $10,000 to $20,000 per month on ads, MMP costs can consume a significant share of the total budget.
The problem is not that these features lack value. Raw data export matters for teams with data warehouses. Fraud detection matters at scale. The problem is that these costs are disclosed incrementally, making it difficult to calculate total cost of ownership before signing.
### 3. Annual Lock-In Before You Have Data
Annual prepayment is the industry default for most enterprise MMPs. A startup launching its first [paid UA campaigns](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) is asked to commit to 12 months of usage before it has 3 months of campaign performance data.
This creates two problems:
* **Cash flow risk.** Locking $24,000 to $120,000 per year in an MMP contract limits ad spend flexibility. For health and fitness apps with seasonal install patterns, fixed annual costs mean overpaying 8 to 9 months of the year while underfunding UA during high-season windows like January and summer.
* **Switching cost lock-in.** Migration to a new provider typically takes 2 to 3 months, during which attribution data has gaps. Teams that signed a mismatched contract often stay not because the product works, but because leaving costs more than staying.
## How to Evaluate MMP Pricing for Subscription Apps
The pricing opacity described above is a real obstacle, but it is not unsolvable. Teams can approach the problem from two angles: building a total cost audit before engaging vendors, and choosing providers whose pricing model matches their stage.
### 1. Build a Total Cost Checklist Before Talking to Sales
Before any vendor conversation, subscription app teams should document their requirements and map them to a cost checklist:
* **Monthly install volume.** Estimate your current and projected attributed installs for the next 6 months. Most MMPs price by installs or events.
* **Required integrations.** List the ad channels you run (Meta, Google, Apple Search Ads, TikTok) and the third-party tools you need (RevenueCat, Adapty, Amplitude). Count how many integrations are included in the base tier versus charged as extras.
* **Data access needs.** Determine whether you need raw data export. Most pre-Series A teams do not, but if you do, factor the additional monthly cost.
* **Contract flexibility.** Ask specifically about monthly versus annual billing, cancellation terms, and overage pricing.
* **Support model.** Clarify whether ticket-based support is included or whether you need to pay for a dedicated CSM.
This checklist turns opaque vendor conversations into structured cost comparisons. The total cost of an MMP for an early-stage subscription app is not just the platform fee. It is the platform fee plus add-ons, implementation time, and the opportunity cost of budget locked in contracts.
[MMP cost evaluation checklist]
If you run paid acquisition for a subscription app and your monthly ad spend is under $20,000, you likely do not need fraud detection, raw data export, or agency access today. Identifying which features to defer saves both cost and implementation time.
### 2. How Airbridge Core Plan Approaches Pricing Differently
Airbridge Core Plan was built for the stage where most MMP pricing falls short: early-stage [subscription apps](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025) running their first paid acquisition campaigns across Meta, Google, Apple Search Ads, and TikTok.
**Core Plan pricing is public and predictable:**
* **15,000 free attributed installs.** No credit card required to start.
* **$0.05 per install after the free tier.** Cost scales linearly with actual usage.
* **Pay-as-you-go billing.** No annual commitment, no minimum contract.
This pricing model directly addresses the three problems above. There is no sales call required to see the price. There are no hidden add-ons because Core Plan intentionally removes features that early-stage teams do not need (fraud detection, raw data export, agency access). And there is no annual lock-in, so teams with seasonal install patterns pay only for what they use.
**What Core Plan includes in the base offering:**
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:--------------|:-------------------|
| Pricing visibility | Requires sales contact | Public: 15K free, $0.05/install |
| Billing model | Annual prepayment | Pay-as-you-go |
| Minimum contract | 12 months, $2K to $10K/month | None |
| Ad channels | All networks (dozens) | GMAT: Meta, Google, Apple Search Ads, TikTok |
| Billing platform integration | Native (included in higher tiers) | Native ([RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration), Adapty in base) |
| Reports | All (with add-on tiers) | 6 built-in: Actuals, Trend, Active User, Funnel, Retention, Revenue |
| Unnecessary features for startups | Fraud, raw export, agency access included | Intentionally removed |
Core Plan supports 25 standard events including Start Trial, Subscribe, and Unsubscribe. These cover the full subscription funnel without requiring custom event schema design. The trade-off is intentional: no custom events, maximum 2 third-party integrations, and GMAT channels only. For teams running $5,000 to $20,000 per month across 2 to 4 major ad platforms, this scope matches the actual need.
The core question Core Plan answers: **"Are paid users converting into subscriptions, and which channels are driving value?"** If your team needs that signal first, before building complex analytics infrastructure, transparent pricing means you can evaluate the cost in minutes rather than weeks.
**A concrete cost example.** A subscription fitness app generating 8,000 attributed installs per month would pay approximately $400 per month on Core Plan. Compare that to enterprise MMP contracts starting at $2,000 to $5,000 per month with annual lock-in. For a team with $15,000 in monthly ad spend, the difference between $400 and $5,000 per month in MMP fees directly impacts how much budget remains for actual [user acquisition](https://www.airbridge.io/en/blog/subscription-app-growth-guide).
[MMP annual cost comparison]
**[See Core Plan pricing and features](https://www.airbridge.io/en/plans/core)**
## FAQ: MMP Pricing for Early-Stage Subscription Apps
### How much does an MMP typically cost for a startup?
MMP pricing for startups varies widely depending on the provider and tier. Enterprise MMPs typically start at $2,000 to $5,000 per month with annual contracts, and total costs can reach $7,000 to $10,000 per month when add-ons like raw data export and premium support are included ([Business of Apps, 2026](https://www.businessofapps.com/marketplace/app-analytics/mobile-measurement-partners/)). See the comparison table above for how Airbridge Core Plan compares on pricing and contract terms.
### When should a subscription app upgrade from Core Plan to Growth Plan?
Core Plan is designed for teams running paid UA on Meta, Google, Apple Search Ads, and TikTok with standard subscription events. Consider upgrading to [Growth Plan](https://www.airbridge.io/en/glossary/lifetime-value-ltv) when your team needs custom event tracking, additional ad network integrations beyond GMAT, raw data exports to warehouses like BigQuery or S3, or agency and media partner access.
### What events does Core Plan track for subscription apps?
Core Plan supports 25 standard events covering the full subscription lifecycle: Install, Sign-up, Start Trial, Subscribe, Unsubscribe, and Order Complete, among others. These predefined event names reduce schema design work compared to custom event implementations. The trade-off: custom events are not supported in Core Plan, so teams with non-standard funnel steps would need [Growth Plan](https://www.airbridge.io/en/blog/3-monetization-flow-best-practices-for-subscription-apps-in-2026).
## Stop Budgeting Blind for Attribution
Every month without clear [attribution signals](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026) is a month where budget flows to channels based on assumptions, not subscription revenue data. Opaque MMP pricing should not be the reason your team delays measurement.
**[Start Free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core)** and see which channels actually convert trial users into paying subscribers, starting with 15K free attributed installs.
---
## Line Yahoo JAPAN Ads Display Ads Integration: Maximize Your Japan Marketing ROI with Airbridge
URL: https://www.airbridge.io/en/blog/airbridge-line-yahoo-japan-ads-integration
Language: English
Published: 2026-03-19
Author: Harper (Trang Nguyen)
Airbridge has integrated with LINE and Yahoo! Display Ads(LY Ads), the Japan's leading unified advertising platform. You can now analyze your LY Ads Display Ads campaign performance more clearly and transparently within Airbridge.
If you're planning to enter the Japanese market or already running online marketing in Japan, you know that LINE and Yahoo! JAPAN are non-negotiable. For most brands, these are top-priority channels with the largest share of the marketing budget.
With this new integration, you can move beyond basic metrics. Airbridge empowers you to measure the true incremental impact of your LY Ads campaigns and build sophisticated retargeting strategies that take your performance marketing in Japan to the next level.
**📌Key Takeaways**
- LY Ads Display Ads is the unified advertising platform combining LINE, Japan's largest messenger app, and Yahoo! JAPAN, the leading search portal. It is Japan's largest advertising platform, capable of reaching over 80% of Japanese internet users.
- Through the Airbridge + LY Ads Display Ads integration, you can precisely measure LY Ads' true contribution via hybrid attribution and integrated SKAN management, while maximizing Japan marketing performance through real-time data postbacks and custom audience transfers for ad optimization and retargeting.
## **What Is LINE and Yahoo! Display Ads?**
[LINE and Yahoo! Display Ads](https://global-marketing.yahoo-net.jp/) is the unified digital advertising platform operated by LY Corporation, Japan's largest IT company. By merging the ad services of LINE and Yahoo! JAPAN into a single interface, it offers a streamlined way to reach the Japanese digital population.
Previously, LINE Ads and Yahoo! JAPAN Ads operated as separate silos. Today, advertisers can manage inventory across both giants under one account, including:
- Yahoo! JAPAN App
- LINE App
- LINE Ad Network
## **LY Ads Display Ads: The Heart of Japanese Digital Marketing**
Japan is a unique market where local platforms often outperform global giants. While Google, Meta, and TikTok are relevant, dominating the local landscape via LINE and Yahoo! JAPAN is essential for success.
- **Massive combined ad inventory:** Access a combined reach of over 90M+ MAU on LINE and 85M+ monthly users on Yahoo! JAPAN. Your ads appear on the apps Japanese consumers use every single day.
- **Broad, nationwide reach:** With over 80% of the users active on LINE across all age groups, it is the most effective channel for broad-scale consumer acquisition.
- **Advanced user segmentation:** By merging Yahoo!’s search intent data with LINE’s behavioral data, LY Ads Display Ads enables hyper-precise interest-based targeting.
Given this level of platform influence, LY Ads is truly at the center of Japan marketing. If you're considering entering Japan, Yahoo! JAPAN Ads Display Ads is not optional – it's a must – and your marketing strategy and results need to be analyzed with greater precision.
## **Airbridge × LY Ads: Ad Optimization and Retargeting, All in One**
With this integration, Airbridge customers running Japan marketing campaigns can measure their performance more accurately and objectively than ever before.
Simply put: you can clearly see which users are most valuable and where they come from – then use that data to run automated ad optimization and retargeting, all in one place.
### **1\. Accurate Attribution with Hybrid Attribution**
LY Ads Display Ads serves ads across diverse placements (news, search, etc.). Airbridge uses a machine learning-based incrementality model and multi-touch attribution to eliminate cross-channel overlap and calculate true, incremental contribution to overall conversions.
### **2\. Campaign and ROAS Optimization via Real-Time Postbacks**
In-app events collected by the Airbridge SDK such as sign-ups, subscriptions, and high-value purchases can be compared across channels (installs, sign-ups, purchases) in a single view, making ROAS assessment much easier.
Campaign, ad group IDs are also mapped in Airbridge reports, so you can see exactly which ad sets are performing best. The ad platform's machine learning then learns from high-value users to automatically optimize bids – **achieving lower CPA and higher ROAS compared to manual management.**
### **3\. SKAN Support for an iOS-Heavy Market**
Japan has one of the highest iPhone market share rates in the world. Airbridge integrates and manages LY Ads Display Ads' SKAdNetwork (SKAN) data natively. SKAN data can be sent directly to Airbridge, preserving performance signals in iOS environments for machine learning optimization – ensuring full performance visibility and data-driven decision-making without technical limitations.
### **4\. Audience Retargeting Based on First-Party Data**
Sync your Airbridge first-party data (e.g., "users who added to cart but didn't buy") directly to LY Ads Display Ads. This allows you to create seamless cross-channel retargeting segments or build Lookalike Audiences to find new Japanese customers who mirror your best existing users.
## **Scale Your Business in Japan Today with Airbridge and LY Ads**
In the Japanese market, visibility is everything. The Airbridge + LY Ads Display Ads integration provides the data engine you need to scale efficiently.
Connect Yahoo! LY Ads Display Ads in your Airbridge dashboard today and see the results of your Japan marketing – proven by data.
For a detailed integration guide, refer to the link below:
👉 [\[Integrate with Airbridge\] LY Ads](https://help.airbridge.io/en/guides/yahoo-japan-ads)
---
## AI Fitness App Attribution: You Built the App in a Weekend — Why Does MMP Setup Still Take Weeks?
URL: https://www.airbridge.io/en/blog/ai-fitness-app-attribution-mmp-setup-still-take-weeks
Language: English
Published: 2026-03-19
Author: Jaehyuk Kim
A fitness app team spends $15,000 across Meta and Google Ads in January. Installs triple. Trial starts surge. But when March arrives and half those users have churned, the team cannot answer a basic question: which campaigns brought subscribers who stayed, and which ones brought resolution-season tourists?
**Without channel-level subscription attribution, every budget increase is a bet placed without data.** For AI fitness apps running trial-to-subscription funnels, this blind spot compounds fast. Health and fitness apps generated over $6 billion in revenue in 2025 ([Business of Apps, 2026](https://www.businessofapps.com/data/fitness-app-market/)), with roughly 80% coming from subscriptions. The teams capturing that revenue are the ones that can trace every subscriber back to the campaign that acquired them.
**Key Takeaways**
* **AI fitness apps face unique attribution challenges.** Seasonal spikes, long trial-to-paid cycles, and high trial volume make it harder to identify which channels drive lasting subscribers.
* **Install counts alone mislead budget decisions.** A channel producing 5,000 installs but only 30 subscribers is outperformed by one producing 1,500 installs and 200 subscribers.
* **Cost per Subscription (CPS) matters more than CPI for fitness apps.** RevenueCat data shows Health and Fitness apps have a median trial-to-paid conversion rate of 39.9% ([RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)), but this rate varies dramatically by acquisition channel.
* **Enterprise MMPs price out most growth-stage fitness teams.** Minimum contracts of $2K-5K/month consume 20-50% of a team spending $10K-20K on ads.
* **Airbridge Core Plan connects ad spend to subscription revenue for AI fitness apps.** With RevenueCat integration, GMAT channel attribution, and pay-as-you-go pricing starting with 15K free installs, it is built for growth-stage subscription apps.
## Why AI Fitness Apps Lose Revenue Without Channel-Level Attribution
AI fitness apps operate in one of the most competitive subscription categories. The global fitness app market reached $12.12 billion in 2025 ([Grand View Research, 2025](https://www.grandviewresearch.com/industry-analysis/fitness-app-market)) and is projected to reach $33.58 billion by 2033. Scaling paid acquisition in this market without attribution is not just inefficient. It is structurally dangerous.
The root cause is a measurement gap between ad platforms and subscription platforms. Meta reports installs. Google reports conversions. RevenueCat reports subscribers. But none of these systems connect to each other by default. A [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) bridges this gap by deduplicating installs across channels and connecting them to downstream subscription events.
### 1. Seasonal Spikes Distort Attribution Windows
Health and fitness apps experience extreme seasonality. January traffic surges 2-3x as users set New Year's resolutions, then drops sharply by March. For teams scaling ad spend during these peaks, the question is not whether installs will increase. They will. The question is whether those installs convert into subscribers who survive past the first renewal.
RevenueCat's 2025 data shows that nearly 30% of annual subscribers cancel within the first month. For fitness apps, this churn concentrates in Q1 when resolution-driven users realize they will not sustain their habits. **A January campaign that produces 2,000 trial starts looks identical to one that produces 2,000 lasting subscribers in install-level data. Only channel-level subscription attribution separates the two.**
Without cohort-level attribution by channel, teams cannot distinguish between a Meta campaign that brought high-retention annual subscribers and a TikTok campaign that brought trial-only users who churned within 14 days. Both campaigns show strong install numbers. Only attribution reveals which one generated lasting revenue.
### 2. Trial-Heavy Funnels Create a Measurement Gap
Most AI fitness apps use free trials as the primary conversion mechanism. RevenueCat data shows 67% of Health and Fitness subscriptions use yearly plans, often preceded by a 7-14 day trial. This creates a structural delay between the install event (tracked by ad platforms) and the subscription event (tracked by RevenueCat or Adapty).
During this delay, the [attribution window](https://www.airbridge.io/en/glossary/attribution-window) may expire. Ad platforms optimize for trial starts, not subscriptions. The result: campaigns optimized for volume, not value.
This matters because AI fitness apps typically have higher trial start rates than average. RevenueCat data shows Health and Fitness apps have a median download-to-trial rate of 7.8%, above the 6.2% category average. High trial volume looks encouraging in platform dashboards. But if 60% of those trials come from one channel with a 15% trial-to-paid rate while another channel converts at 45%, the team is celebrating the wrong metric. **A channel producing high trial volume but low trial-to-paid conversion drains budget without generating subscription revenue.**
[AI Fitness App Subscription Funnel showing where attribution breaks]
### 3. The Hidden Cost: Budget Flows to the Wrong Channel
Consider a fitness app spending $20,000/month across three channels:
| Channel | Installs | Trial Starts | Paid Subscribers | CPS |
|:--------|:---------|:-------------|:-----------------|:----|
| Meta Ads | 8,000 | 2,400 | 280 | $24 |
| Google Ads | 4,000 | 1,000 | 180 | $37 |
| TikTok | 6,000 | 1,800 | 45 | $148 |
Without attribution, this team splits budget evenly across all three channels. With attribution, the data is clear: **Meta produces the highest subscription volume at the lowest CPS, while TikTok drives high trial volume but minimal paid conversions.** A budget reallocation from TikTok to Meta could increase total subscribers by 30-40% at the same spend level.
This is not hypothetical. According to RevenueCat's 2025 benchmark, the gap between top-performing (P90) and median fitness apps in revenue per install is over 6x ($4.19 vs $0.63 at Day 60). Channel-level attribution is what separates these tiers.
[The Fitness App Attribution Gap showing ad platform data vs revenue truth]
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Connect your ad spend to real subscription revenue — setup takes minutes, not weeks.%%
## How to Connect Ad Spend to Subscription Revenue for Fitness Apps
### General Approaches Any Team Can Apply Today
Before selecting any attribution tool, growth teams can take immediate steps to reduce measurement gaps:
* **Align attribution windows across platforms.** Set click-through and view-through windows consistently across Meta, Google, and TikTok. Mismatched windows create artificial discrepancies in channel performance data.
* **Track CPS alongside CPI.** [Cost per install tells only part of the story](https://www.airbridge.io/en/blog/subscription-app-growth-guide). A channel with a $2 CPI but a $200 CPS is underperforming a channel with a $5 CPI and a $30 CPS.
* **Connect your billing platform to your campaign data.** If you use RevenueCat or Adapty, export subscription events and match them to install cohorts manually if no automated integration exists.
* **Set up UTM parameters for every campaign.** This provides a minimal attribution signal for web-to-app flows, even without an MMP.
* **Focus on trial-to-paid conversion rate by channel, not aggregate trial volume.** A 50% conversion rate on 500 trials beats a 10% conversion rate on 2,000 trials for [lifetime value (LTV)](https://www.airbridge.io/en/glossary/lifetime-value-ltv).
### How Airbridge Core Plan Solves This for AI Fitness Apps
The general approaches above work, but they rely on manual data matching that breaks at scale. Airbridge Core Plan automates the connection between ad spend and subscription revenue for fitness apps running paid acquisition on Meta, Google, Apple Search Ads, and TikTok.
**Core Plan answers one question: "Which paid channels are driving subscribers, and at what cost?"**
Here is how it maps to the specific challenges AI fitness apps face:
| Fitness App Challenge | How Core Plan Addresses It |
|:----------------------|:---------------------------|
| Seasonal attribution confusion | Retention and Funnel reports compare cohort performance by channel across time periods |
| Trial-to-subscription gap | Standard events (Start Trial, Subscribe) connect the full funnel to acquisition source |
| Budget misallocation | Revenue report shows subscription revenue by channel, enabling data-driven reallocation |
| Billing platform disconnect | Native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and Adapty integrations connect subscription data to attributed installs |
| Enterprise MMP cost barrier | 15K free attributed installs, $0.05/install after, pay-as-you-go pricing |
**What Core Plan tracks for fitness apps:**
* **25 standard events** including Install, Sign-up, Start Trial, Subscribe, Unsubscribe, and Order Complete. These cover the full subscription funnel without requiring custom event schema design.
* **4 Self-Attributing Networks (GMAT):** Meta Ads, Google Ads, Apple Search Ads, TikTok for Business. These channels typically represent [80-90% of early-stage paid acquisition spend](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025).
* **6 built-in reports:** Actuals, Trend, Active User, Funnel, Retention, and Revenue.
**Intentional design decisions:**
Core Plan does not support custom events. For fitness apps tracking the standard subscription funnel (install, trial, subscribe, renew), standard events cover the full journey. Core Plan supports a maximum of 2 third-party integrations (e.g., RevenueCat + Amplitude), keeping the stack focused. Teams needing custom events, additional ad networks, or raw data exports can [upgrade to Airbridge's Growth Plan](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) as their operations mature.
**Pricing context for growth-stage fitness teams:**
Enterprise MMPs typically require minimum annual contracts of $2K-5K/month. For a fitness app team spending $10K-20K/month on ads, this means the measurement tool alone consumes 10-50% of the marketing budget. Core Plan removes this barrier: 15K free attributed installs cover the first months of paid acquisition, and pay-as-you-go pricing at $0.05/install means measurement cost scales with actual growth.
If you are running paid acquisition for an AI fitness app and your [GA4 data does not match](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it) what RevenueCat reports, the gap is not a reporting error. It is a structural blind spot that an MMP resolves.
## FAQ: AI Fitness App Attribution
### What events should a fitness app track for attribution?
At minimum, track Install, Start Trial, Subscribe, and Unsubscribe. These four events let you calculate CPS by channel and trial-to-paid conversion rate by campaign. Adding Sign-up and Order Complete provides a more complete funnel view. Core Plan supports all of these as standard events, so no custom schema design is required.
### When should a fitness app upgrade from Core Plan to Growth Plan?
Upgrade when your operations outgrow the focused feature set. Common triggers include needing custom events beyond the 25 standard events, requiring more than 2 third-party integrations, wanting raw data exports to a data warehouse like BigQuery, or expanding to ad networks beyond Meta, Google, Apple Search Ads, and TikTok. Most fitness apps reach this point when monthly ad spend exceeds $50K and the team adds a dedicated data analyst.
### How does fitness app seasonality affect attribution accuracy?
January and September are peak acquisition months for fitness apps. During these spikes, multiple ad platforms claim credit for the same installs, inflating reported conversions. An MMP deduplicates these claims and assigns each install to one channel. Without deduplication, teams over-credit every channel simultaneously and cannot identify which one truly drove the subscription. This is especially important for annual subscription plans where the revenue event occurs days or weeks after the install.
## Every Dollar You Spend Without Attribution Is a Dollar You Cannot Optimize
The gap between fitness apps that scale efficiently and those that burn budget is not creativity or channel selection. It is measurement. With the fitness app market projected to grow from $12 billion to over $33 billion by 2033, the teams that connect every subscriber back to the campaign that acquired them will capture disproportionate share of that growth.
Attribution is not a nice-to-have reporting layer. It is the foundation that turns paid acquisition from a cost center into a growth engine. Every month without it is a month of budget decisions made on incomplete data.
**[Start free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core) and see which channels actually drive your fitness app's subscribers, beginning with 15K free attributed installs.**
---
## The First Payment Is Just the Beginning: What Subscription Lifecycle Analytics Actually Measures
URL: https://www.airbridge.io/en/blog/first-payment-beginning-what-subscription-lifecycle-analytics-actually-measures
Language: English
Published: 2026-03-19
Author: Jaehyuk Kim
A [5.3% monthly churn rate](https://marketingltb.com/blog/statistics/subscription-statistics/) sounds manageable. Lose 53 subscribers out of 1,000 each month — you can replace them.
**But churn compounds.** That 5.3% monthly rate means losing 47% of your subscriber base over a year. A fitness app with 10,000 subscribers in January will have 5,300 by December — even if nothing else changes. **And most teams do not see this happening because they stop measuring at the first payment.**
[10,000 subscribers in January declining to 5,300 by December at 5.3% monthly churn — 47% annual loss]
The install got tracked. The trial got tracked. The first subscription got tracked. What happens after — whether that subscriber renews, churns, or fails a payment — is where most measurement systems go dark.
**This is the gap that subscription lifecycle analytics fills.** It is not a single metric. It is the practice of tracking every revenue-relevant event from trial to renewal to churn — and connecting each event back to the channel that acquired the user.
**Key Takeaways**
* **Most fitness app teams stop measuring at first payment.** But [the average subscription app loses 5.3% of subscribers monthly](https://marketingltb.com/blog/statistics/subscription-statistics/) — compounding to 47% annual loss. What happens after the first payment determines whether your UA investment pays back.
* **Involuntary churn is the hidden revenue leak.** [Failed payments account for 20–40% of total churn](https://www.slickerhq.com/blog/2025-failed-payment-benchmarks-b2c-subscription-ecommerce-ai-recovery), and Health & Fitness apps see 15–22% failed payment rates. These are subscribers who wanted to stay but could not pay.
* **Renewal tracking by channel reveals which UA channels produce lasting subscribers** — not just first-month converters.
* **Lifecycle analytics connects four stages to acquisition source:** Trial → First Payment → Renewal → Churn. Without this connection, you cannot tell which channels produce subscribers who stay versus subscribers who leave.
* **Airbridge Core Plan tracks the full lifecycle by channel** — Funnel, Retention, and Revenue reports with RevenueCat/Adapty integration. Start with 15K free attributed installs.
## What Most Teams Track — and Where They Stop
The typical subscription app measurement stack covers three events:
* **Install** — tracked by the ad platform
* **Trial start** — tracked by in-app analytics
* **First subscription** — tracked by the billing platform (RevenueCat, Adapty)
**These three events tell you how much it costs to acquire a subscriber. They tell you nothing about what that subscriber is worth.**
[Four lifecycle stages: Trial, First Payment, Renewal, Churn — most teams track only the first two] A subscriber who cancels after one month has a completely different LTV than one who renews for a year — but at the point of first payment, they look identical.
Subscription lifecycle analytics extends measurement beyond the first payment to capture the events that actually determine LTV:
| Stage | Event | What It Tells You |
|:------|:------|:------------------|
| **Acquisition** | Install → Trial → Subscribe | Cost to acquire a subscriber |
| **Retention** | Renewal (month 2, 3, 6, 12) | Whether the subscriber stays |
| **Involuntary churn** | Failed payment → Recovery attempt | Whether billing failures leak revenue |
| **Voluntary churn** | Cancellation | Why subscribers actively leave |
**The first row is where most teams stop. The last three rows are where LTV is determined.**
## The Four Stages of Subscription Lifecycle Analytics
### Stage 1: Trial to First Payment — The Stage Everyone Measures
This is the most visible conversion in the subscription funnel. [Health & Fitness apps convert trials to paid at 35%](https://www.slickerhq.com/blog/2025-failed-payment-benchmarks-b2c-subscription-ecommerce-ai-recovery) — meaning 65% of trials never generate revenue.
**What to track:** Trial-to-paid conversion rate by channel. If Meta trials convert at 20% and Google trials convert at 45%, the cost per subscriber is radically different even at similar CPI. This stage is well-covered in [our CPI vs CPS analysis](/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics).
### Stage 2: Renewal — The Stage Most Teams Miss
The first payment proves willingness to pay. **The second payment proves willingness to stay.** Month-1-to-month-2 renewal is the single strongest predictor of long-term LTV — and most teams do not track it by channel.
Why this matters: [annual subscribers are 2.4x more profitable than monthly subscribers](https://marketingltb.com/blog/statistics/subscription-statistics/) and churn 51% less. **But if you do not track which channels produce annual vs monthly subscribers, you cannot optimize for retention quality.**
**What to track:** Month-2 renewal rate by acquisition channel. A channel with high first-payment conversion but low renewal produces churners, not subscribers.
### Stage 3: Involuntary Churn — The Silent Revenue Leak
Not every subscriber who leaves chose to leave. [Failed payments account for 20–40% of total churn](https://www.slickerhq.com/blog/2025-failed-payment-benchmarks-b2c-subscription-ecommerce-ai-recovery), and Health & Fitness apps experience **15–22% failed payment rates.**
These are subscribers who intended to keep paying — but their card expired, the bank declined the charge, or the payment method had insufficient funds. **They did not cancel. The billing system lost them.**
[Churnkey's 2025 data](https://churnkey.co/reports/state-of-retention-2025) shows that 70% of involuntary churn can be recovered through retry logic and recovery campaigns. But most teams do not even know how much revenue they are losing to failed payments — let alone which acquisition channels produce subscribers with higher payment failure rates.
**What to track:** Failed payment rate and recovery rate by cohort. If a specific acquisition channel consistently produces subscribers with higher involuntary churn, the true CPS of that channel is higher than it appears.
### Stage 4: Voluntary Churn — Why Subscribers Actively Leave
Voluntary churn accounts for the majority of subscription losses. [Budget limitations drive 33% of cancellations, followed by infrequent usage](https://churnkey.co/reports/state-of-retention-2025). The ratio varies by app category, price point, and user profile — meaning the dominant reason your subscribers leave may look different across users acquired from different channels.
**What to track:** Cancellation rate and cancellation reason by acquisition channel. If subscribers from TikTok cancel for "infrequent usage" at 2x the rate of subscribers from Google, TikTok is delivering lower-intent users — regardless of CPI.
## What You Lose Without Lifecycle Tracking
Without tracking beyond the first payment, budget decisions rely on acquisition metrics alone. Here is what that misses:
* **Channel A** produces 100 subscribers/month at $40 CPS. Month-2 renewal rate: 85%.
* **Channel B** produces 80 subscribers/month at $50 CPS. Month-2 renewal rate: 95%.
**Channel A looks 20% cheaper on CPS.** But that 10-point renewal gap compounds every month. By month 6, Channel B retains more subscribers in absolute terms than Channel A — despite starting with 20% fewer. If each subscriber generates $15/month, Channel B produces significantly more cumulative revenue per cohort despite higher acquisition cost.
**This calculation is impossible without renewal tracking by channel.** And renewal tracking by channel requires connecting billing events (RevenueCat, Adapty) to acquisition source — which is exactly what lifecycle analytics provides.
## Connecting Lifecycle Events to Acquisition Source
Tracking the full lifecycle does not require custom data pipelines. It requires connecting three systems that most teams already use:
* **Ad platform → Attribution:** Which channel drove the install?
* **App → Attribution:** Which installs became trials and subscribers?
* **Billing platform → Attribution:** Which subscribers renewed, churned, or failed a payment?
**Without an [attribution system](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) that integrates with your billing platform, these metrics exist in isolation** — the billing platform sees the churn, but not the channel. The ad platform sees the install, but not the churn.
> [See which channels produce subscribers who stay — and which produce subscribers who leave. Funnel, Retention, and Revenue reports by channel. Start free with 15K attributed installs.](https://www.airbridge.io/en/plans/core)
## What to Look for in an MMP for Subscription Lifecycle Tracking
When evaluating an MMP for subscription lifecycle analytics, confirm it supports:
* **Install → Trial → Subscribe funnel by channel.** Not just installs by source — the full conversion funnel, broken down by acquisition channel, without requiring a premium tier.
* **Cohort retention by acquisition source.** Retention curves by the channel that originally acquired the subscriber, not just aggregate retention.
* **Revenue attribution by channel.** Subscription revenue tied back to the ad campaign and channel that drove the install.
* **Native billing platform integration.** S2S connection to RevenueCat or Adapty — so subscription events are captured even when the app is not open.
* **Predefined subscription events.** Standard events for Start Trial, Subscribe, Renew, and Cancel — without requiring custom event mapping for each use case.
* **Usage-based pricing with no annual lock-in.** Pay for what you use, not a fixed annual commitment that stays the same regardless of volume.
## Stop Measuring at the First Payment
The first subscription tells you one thing: the user paid. **Everything that determines whether that payment was worth acquiring — renewal, retention, churn — happens after.**
Subscription lifecycle analytics is not a new category of tool. It is the practice of connecting billing events to acquisition source — so budget decisions reflect the full value of a subscriber, not just the cost of acquiring one.
**[See which channels produce subscribers who stay. Funnel, Retention, and Revenue reports by channel. Start free with 15K attributed installs on Airbridge Core Plan.](https://www.airbridge.io/en/plans/core)**
---
## Airbridge Is Now Be Eligible Under Türkiye’s Government Incentive Program for Service Exporters
URL: https://www.airbridge.io/en/blog/airbridge-is-now-be-eligible-under-turkiyes-government-incentive-program-for-service-exporters
Language: English
Published: 2026-03-18
Author: Luke
Türkiye’s Service Sector Support Program can help eligible app and digital service companies reduce the cost of growth tools, including software and technology platforms. For Turkish companies expanding globally with Airbridge, that can make performance infrastructure far more affordable.
## What is Türkiye’s Service Sector Support Program?
### **A support program for service exporters**
[Türkiye’s Service Sector Support Program](https://ticaret.gov.tr/destekler/hizmet-sektoru-destekleri/hizmet-sektorlerinin-desteklenmesine-iliskin-genelge-ve-ekleri/markalasma-programi/ortak-destek-unsurlari-basvuru-belgeleri/rapor-ve-veri-tabani-uyelik-destegi) is designed to help Turkish companies in eligible service categories grow internationally by subsidizing selected operational and technology expenses. Supported areas include IT, mobile applications, software development, artificial intelligence, cybersecurity, big data, digital services, digital platforms, consulting, education, fintech, and creative industries.
### **Who may be eligible**
In general, companies should meet the following conditions:
- Operate in an eligible service sector
- Be registered in Türkiye
- Generate service export revenue
- Typically show that at least 51% of revenue comes from the relevant service sector
### **What the support can cover**
Eligible expenses may include:
- Software and technology services
- International marketing and promotion
- Platform and digital infrastructure costs
- Consulting and certification expenses
### **Support rate, duration, and process**
Most incentives provide up to 50% reimbursement, with caps depending on the support category. The policy also notes a typical support period of 3 years, with some categories extending to 4 to 5 years, and applications should be submitted through the Ministry’s DYS system with supporting documents. Programs covering services such as Airbridge may support up to around TRY 2.5 million under certain initiatives.
## How the policy impacts Airbridge users in Türkiye
Many Turkish companies already chose Airbridge long before this incentive entered the conversation because the product already solves the measurement problems that slow growth. It brings Android, iOS, SKAN, web, PC and console, and CTV into one dashboard, with deeper campaign analysis, an all-inclusive model, and advanced capabilities that often cost extra elsewhere.
**Why teams already choose Airbridge:**
- Cross-platform measurement in one dashboard
- Strong web-to-app tracking and deep linking
- Granular reporting for campaigns, creatives, funnels, retention, and ROAS
- Predictive LTV and stronger optimization signals
- Fraud protection and privacy-ready measurement support
- Raw data access and advanced capabilities without a paywall
- Official measurement partnerships with major ad platforms such as Google, Meta, and TikTok
With Türkiye’s Service Sector Support Program, the financial value of Airbridge becomes even stronger. Eligible Turkish companies may receive up to 50% reimbursement on qualified software and technology costs, lowering the effective cost for existing customers and making adoption easier for companies still evaluating Airbridge. For teams that want stronger attribution, better optimization, and clearer visibility across growth channels, this makes the case to move forward even easier.
If your company is unsure whether it qualifies, or if you want to explore what Airbridge can offer at a lower effective cost, [connect with our team](https://www.airbridge.io/book-a-demo) to review your fit, assess possible eligibility, and find the best way to scale with Airbridge.
---
## CPI, Cost per Trial, Cost per Subscription: The Three Metrics That Actually Measure Fitness App UA
URL: https://www.airbridge.io/en/blog/cost-per-trial-cost-per-subscription-subscription-app-ua-metrics
Language: English
Published: 2026-03-17
Author: Jaehyuk Kim
You run two channels for your fitness app. Same monthly budget.
- **Channel A:** CPI $2. Install-to-trial rate 10%. Trial-to-paid rate 20%. **Cost per subscriber: $100.**
- **Channel B:** CPI $5. Install-to-trial rate 40%. Trial-to-paid rate 50%. **Cost per subscriber: $25.**
CPI says Channel A is 2.5x more efficient. Cost per subscription says Channel B delivers subscribers at 4x lower cost. **If you allocate budget based on CPI, you are overspending on the wrong channel.**
This is not a hypothetical. Most fitness app teams report CPI because every ad platform calculates it automatically. But CPI measures the cheapest, shallowest part of the funnel — and tells you nothing about what happens after the install.
**Key Takeaways**
- **CPI is the only UA metric with standardized benchmarks — and the worst predictor of fitness app UA success.** [Facebook CPI $2–$5.50, Google $1.50–$4.50, TikTok $1.75–$4.00](https://www.blog.udonis.co/user-acquisition/cost). None of these numbers tell you whether those installs become subscribers.
- **Cost per trial (CPT) is the middle metric most fitness app teams skip.** [Healthy blended CPT: $20–$40](https://www.linkedin.com/pulse/what-good-cost-per-trial-subscription-apps-user-shamanth-m-rao/) for most subscription apps. But no standardized benchmark exists because CPT requires attribution infrastructure most teams lack.
- **Cost per subscription (CPS) is the metric that actually measures UA success** — but almost no one tracks it by channel because it requires attribution + billing integration.
- **Terminology trap: "CPT" means different things in different contexts.** Apple Search Ads uses CPT for ["cost per tap"](https://ads.apple.com/app-store/help/reporting/0023-reporting-options-and-definitions) (global median [$0.92](https://www.apptweak.com/en/aso-blog/apple-ads-benchmarks)). Subscription app teams use CPT for "cost per trial" ($20–$40 range). Do not confuse the two.
- **Airbridge Core Plan connects CPI to trial to subscription by channel** — so you can calculate CPT and CPS per channel without custom data pipelines. Start with 15K free attributed installs.
## Three Metrics, Three Stories
Every fitness app UA metric answers a different question. The problem is that most teams only measure the first one.
[CPI measures installs. Cost per trial measures intent. Cost per subscription measures revenue.]
### CPI: Cost per Install — What Everyone Measures
**Formula:** Total ad spend ÷ number of installs.
CPI is universal. Every ad platform reports it. Industry benchmarks are [widely published](https://www.blog.udonis.co/user-acquisition/cost):
| Platform | CPI Range |
| --- | --- |
| Facebook Ads | $2.00–$5.50 |
| Google Ads | $1.50–$4.50 |
| TikTok Ads | $1.75–$4.00 |
| Apple Search Ads (US) | $4.06 median |
**CPI is useful for one thing: measuring the cost of getting a user to download your app.** For fitness apps, that is the least important step. An install is not a trial. A trial is not a subscriber. A $2 install from someone who never opens the app costs infinitely more than a $5 install from someone who subscribes to your workout plan.
**The trap:** because CPI benchmarks exist, teams use them to compare channels. But CPI comparisons alone are insufficient for fitness apps — they measure acquisition efficiency without capturing what happens after the install.
### Cost per Trial: The Metric Most Teams Skip
**Formula:** Total ad spend ÷ number of trial starts.
Cost per trial (CPT) tells you how much you pay to get a user to start a free trial — the first real signal that someone is considering your workout plan or meditation program.
[Shamanth Rao's analysis](https://www.linkedin.com/pulse/what-good-cost-per-trial-subscription-apps-user-shamanth-m-rao/) puts healthy blended CPT at **$20–$40** for most subscription apps, with trial-to-paid conversion of 45–50%. A target below $12 is "incredibly aggressive — no app I know of has attained this at meaningful scale." For fitness apps, where [impulse trial starts are common](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets), CPT is especially important — it reveals how much you pay for genuine interest versus motivation-spike downloads.
**Why most teams skip CPT:** ad platforms do not report it. Calculating CPT by channel requires an [attribution system](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) that connects ad spend to in-app trial events. Without this, CPT is invisible — teams see CPI (from the ad platform) and trial volume (from their app analytics) but cannot connect the two by channel.
**Terminology warning:** Apple Search Ads uses "CPT" to mean ["cost per tap"](https://ads.apple.com/app-store/help/reporting/0023-reporting-options-and-definitions) — their click metric, with a global median of [$0.92](https://www.apptweak.com/en/aso-blog/apple-ads-benchmarks). This is **not** cost per trial. When you see "CPT" in an Apple Ads context, it means the cost of a tap on your ad, not the cost of a trial start. In this article, CPT means cost per trial.
### Cost per Subscription: The Metric That Actually Matters
**Formula:** Total ad spend ÷ number of paid subscriptions.
Or equivalently: **CPS = CPI ÷ (install-to-trial rate × trial-to-paid rate).**
CPS is the true unit economics metric for subscription-model fitness apps. It tells you how much you paid to acquire a paying subscriber — the primary driver of recurring revenue.
**Why no benchmarks exist:** CPS depends on two conversion rates that vary enormously. [Health & Fitness apps convert trials to paid at 39.9%](https://www.revenuecat.com/state-of-subscription-apps-2025/), while the global average is 25.6%. Install-to-trial rates vary from 5% to 40%+ depending on onboarding, paywall design, and channel quality. Multiply these variables together, and no universal CPS benchmark is meaningful.
**Why no CPS benchmarks exist — and why CPT benchmarks barely exist either — comes down to infrastructure, not importance:**
- **CPI:** The ad platform calculates it automatically. Every team has this number.
- **CPT:** Requires connecting ad spend to in-app trial events via attribution. The ad platform sees the install; your app sees the trial — without attribution, no one sees both by channel.
- **CPS:** Requires attribution + billing integration. Three systems need to talk: ad platform → attribution → billing (RevenueCat, Adapty).
**Each step deeper requires one more system connection.** The metric gets harder to measure — not less important.
> [See CPI, cost per trial, and cost per subscription by channel — all in one report. Start free with 15K attributed installs.](https://www.airbridge.io/en/plans/core)
## The CPI-to-CPS Calculator: See What Your CPI Actually Costs
The intro showed the principle — cheap CPI can mean expensive CPS. Here is how to calculate it with real benchmark ranges.
**CPS = CPI ÷ (install-to-trial % × trial-to-paid %)**
### Worked Example: Fitness App on Meta vs Google
[Meta CPI $3.50 looks cheaper than Google CPI $4.50. But Meta CPS $58.48 is 30% more expensive than Google CPS $45.00.]
| | Meta | Google |
| --- | --- | --- |
| CPI | $3.50 | $4.50 |
| Install-to-trial rate | 15% | 25% |
| Trial-to-paid rate | 39.9% (H&F benchmark) | 40% |
| CPT (CPI ÷ install-to-trial) | $23.33 | $18.00 |
| CPS (CPI ÷ install-to-trial ÷ trial-to-paid) | $58.48 | $45.00 |
**CPI says Meta is 22% cheaper. CPS says Google delivers subscribers at 23% lower cost.** Allocating more budget to Meta based on CPI alone increases blended CPS — the opposite of what the team intended.
### What Changes When You Optimize for CPS
- **Budget allocation flips.** Channels that look expensive on CPI may be the cheapest on CPS.
- [**Creative testing**](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets) **changes.** The winning creative is not the one with the lowest CPI — it is the one with the lowest CPS.
- **Channel evaluation changes.** A channel with $5 CPI and 50% trial-to-paid rate beats a channel with $2 CPI and 10% trial-to-paid rate — by 2.5x on CPS.
## How to Track CPI, CPT, and CPS by Channel
### What You Need for Each Metric
| Metric | Data Source | Infrastructure Required |
| --- | --- | --- |
| CPI | Ad platform dashboard | None — reported by default |
| CPT | Attribution + trial event | Attribution system with in-app event tracking |
| CPS | Attribution + subscription event | Attribution system + billing platform integration |
**CPI is free. CPT requires attribution. CPS requires attribution + billing.** This is why most teams stop at CPI — not because CPT and CPS are less important, but because they need more infrastructure.
### How Airbridge Core Plan Connects Install to Trial to Subscription
Core Plan tracks Install, Start Trial, and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. The Actuals Report shows conversion rates at each funnel stage by channel — **so you can calculate CPT and CPS per channel without building custom data pipelines.**
With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, subscription events flow into attribution automatically. **CPI from the ad platform, trial events from your app, subscription events from your billing system — all connected by channel in one report.**
### Airbridge Core Plan vs Traditional MMP
| Capability | Traditional MMP (Typical) | Airbridge Core Plan |
| --- | --- | --- |
| CPI by channel | Available by default | Available by default |
| CPT (cost per trial) by channel | Available, but may require premium tier for full funnel | Standard events with full funnel attribution |
| CPS by channel | Available with configuration, often behind add-on pricing | Actuals Report with event breakdown |
| Billing platform integration | Native integrations available (RevenueCat, Adapty supported) | Native RevenueCat/Adapty via S2S |
| Onboarding focus | General-purpose | Subscription-app focused onboarding |
| Pricing model | Annual contract + add-ons | Pay-as-you-go, $0.05/install |
| Free tier | Limited or none | 15K free attributed installs. All features included |
## CPI Measures Attention. CPS Measures Revenue.
CPI tells you the cost of getting someone to download your app. Cost per trial tells you the cost of getting someone to consider paying. Cost per subscription tells you the cost of getting someone to actually pay.
**Most fitness app teams only measure CPI — and make budget decisions on incomplete data.** The fix is not picking one metric over another. It is measuring all three by channel — so you can see where CPI is cheap but CPS is expensive, and reallocate accordingly.
[**See CPI, cost per trial, and cost per subscription by channel. Start free with 15K attributed installs on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## Qualified Trials for Subscription Apps: How to Send Trial Signals That Actually Improve Ad Performance
URL: https://www.airbridge.io/en/blog/qualified-trials-fitness-apps-trial-signals-ad-performance
Language: English
Published: 2026-03-16
Author: Jaehyuk Kim
Your Meta campaign generated 1,200 trial starts last month. Your CPI held steady. Trial volume hit target. Everything in the dashboard looks healthy.
**But your cost per subscriber has been climbing steadily over the same period. The problem is not your targeting, your creative, or your budget. It is the trial event itself.** Every time a user taps "Start Free Trial" and cancels 10 minutes later, that event reaches Meta as a valid conversion. The algorithm learns from it, and it learns the wrong lesson.
**Key Takeaways**
* **Raw trial events include impulse cancellers, and the algorithm treats them as success signals.** A user who cancels in 10 minutes looks identical to a 6-month subscriber in Meta's conversion data.
* **Over 4-8 weeks, dirty trial signals degrade algorithm quality.** CPI stays flat, trial starts look fine, but CPS climbs because the algorithm optimizes for the wrong user profile.
* **A qualified trial delays the conversion signal by 2-4 hours,** firing only after the user demonstrates real engagement and filtering out impulse cancellers before the signal reaches the ad platform.
[Raw trial fires immediately including impulse cancellers. Qualified trial fires after 2-4 hours, filtering out cancellers.]
* **Fitness apps are especially vulnerable.** Health & Fitness apps have a median trial-to-paid conversion rate of 39.9% ([RevenueCat, 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)), but this median masks wide variance between high-intent and impulse trial starters.
* **Airbridge Core Plan shows trial-to-subscription conversion rates by channel,** so you can diagnose which channels have the most dirty trial signals and validate whether qualified trials are improving CPS. Start free with 15K attributed installs.
## Raw Trial Signals Train the Algorithm to Find the Wrong Users
The trial event is the most common conversion signal subscription apps send to ad platforms. It fires within seconds, generates high volume, and for most fitness app teams, it is the only signal they send. **The problem: not all trials are equal, but the algorithm treats them as if they are.**
### 1. The 10-Minute Cancel Problem
A user sees your fitness app ad on Instagram. They tap through, hit "Start Free Trial," see the subscription terms, and immediately cancel. All within 10 minutes. They never opened the app a second time. They never started a workout. They never intended to pay.
But Meta received a Start Trial conversion event. **From the algorithm's perspective, this user is a success.** The lookalike model updates. The next delivery cycle targets more users with this same profile: [impulse-driven, charge-anxious users](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets) who never convert to paid.
These dirty trial signals degrade algorithm quality invisibly:
* **Week 1-2:** Trial starts look normal. No red flags.
* **Week 4:** Trial-to-paid conversion dips slightly.
* **Week 8:** CPS is climbing, but CPI is still flat. The team sees no warning in the metrics they watch.
* **Week 12:** The algorithm has been trained on 8+ weeks of contaminated data. Subscriber growth stalls, but trial numbers look healthy.
[Signal degradation timeline: CPI stays flat while CPS climbs over 12 weeks of dirty trial signals]
RevenueCat data shows that [nearly 30%](https://www.revenuecat.com/state-of-subscription-apps-2025/) of annual subscriptions are cancelled in the first month. For fitness apps, 82% of trials start on Day 0 (install day), and a significant share of those are impulse-driven.
### 2. What Happens When You Cannot See Trial Quality
Consider a concrete scenario. A fitness app growth team runs Meta and TikTok campaigns simultaneously. Both channels generate roughly 600 trial starts per week. CPI is within target on both. The team has no way to see trial-to-paid conversion rates broken down by channel, because their analytics stack only tracks aggregate trial volume.
What they cannot see: Meta's trial-to-paid rate is 18%. TikTok's is 42%. Meta is flooding the funnel with impulse trialists who cancel within minutes. But because trial volume looks identical across channels, the team splits budget 50/50. Worse, they increase Meta spend in Week 6 because Meta's CPI is slightly lower.
By Week 10, overall CPS has climbed 35%. The team cuts creative, tests new audiences, adjusts bids. None of it works, because the problem was never targeting or creative. It was the signal itself. **Without channel-level trial quality data, the team optimized budget toward the channel producing the most waste.** They spent two months and thousands of dollars debugging the wrong problem.
This is not a hypothetical edge case. It is the default outcome for any subscription app team that sends raw trial events and lacks attribution-level trial quality visibility.
### 3. Why Fitness Apps Lose More to Dirty Trial Signals
The gap between trial intent and subscription intent is category-dependent. For a productivity app, a user who starts a trial probably needs the tool. **For a fitness app, a user who starts a trial might just be having a motivation spike,** not someone ready to commit to a $14.99/month workout plan.
RevenueCat's [State of Subscription Apps report](https://www.revenuecat.com/state-of-subscription-apps-2025/) shows Health & Fitness apps have a median trial-to-paid conversion rate of 39.9%. But the variance within a single app's trial cohort is where the real story lives. Users who engage immediately (completing onboarding, browsing workouts, setting goals) convert at far higher rates. Impulse cancellers inflate the trial count without contributing to conversion.
**For fitness apps, "Start Trial" is closer to a bookmark than a commitment.** Users start trials while scrolling in bed, after watching a transformation video, during a motivation burst. Many cancel before the app finishes loading. Every one of these cancellations is a training signal telling the algorithm: find more users like this.
## How Qualified Trials Fix Algorithm Learning
A qualified trial is a conversion event that fires not when the user starts a trial, but after the user demonstrates real engagement. It is the same trial, but the signal is delayed until the user passes a minimum engagement threshold.
### 1. Implement Qualified Trials for Your Fitness App
**Define your qualification window.** The industry standard is 2-4 hours after trial start:
* **Under 1 hour:** Still includes many impulse cancellers. Signal quality improvement is marginal.
* **2-4 hours:** Filters out the majority of impulse cancellers. Long enough for genuine engagement signals: completed onboarding, started a workout, browsed content.
* **Over 6 hours:** Loses signal volume. If weekly qualified trials per ad set drop below 50 events, the algorithm's learning phase stalls.
**Define your qualification criteria:**
* **Minimum viable:** Trial still active after X hours (user has not cancelled)
* **Engagement-based:** User completed onboarding OR started first workout OR browsed 3+ workout categories
* **Hybrid:** Trial active after 3 hours AND at least one in-app action completed
Start with the simplest criteria (trial still active after 3 hours) and iterate. Overly complex qualification logic adds engineering overhead without proportional signal quality improvement.
**Measure trial quality by channel before and after.** You need channel-level trial quality data to validate whether your qualification criteria are improving CPS:
* **Trial-to-paid conversion rate by channel:** The primary indicator of trial signal quality. If Meta's trial-to-paid is 15% while Google's is 40%, Meta has a bigger dirty signal problem.
* **CPS by channel over time:** Track weekly for 4-8 weeks post-change.
* **Trial volume by channel:** Monitor for excessive volume drop. If qualified trial volume falls below the algorithm's learning threshold, widen the qualification window.
**This measurement requires an [attribution system](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) that connects trial events to subscription outcomes at the channel level.**
**[See which channels send you impulse trialists. Start free with 15K attributed installs.](https://www.airbridge.io/en/plans/core)**
### 2. How Airbridge Core Plan Measures Trial Quality by Channel
Implementing qualified trials is an app-level change. Your app decides when to fire the event. **But validating whether it works requires channel-level attribution data that connects trial starts to subscription outcomes.** This is what Airbridge Core Plan provides.
Core Plan tracks Start Trial and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. The Actuals Report breaks down trial-to-subscription conversion rates by channel, so you can see which channels have the widest gap between trial starts and paid subscriptions.
With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, subscription events flow into the attribution system automatically. Before implementing qualified trials, this data diagnoses the problem. After implementation, it validates the fix.
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:---------------|:--------------------|
| Trial-to-subscription by channel | Available (may require premium tier for full breakdown) | **Standard events with channel attribution (included in base)** |
| Billing platform integration | Native integrations available (RevenueCat, Adapty supported) | **Native RevenueCat/Adapty via S2S** |
| Trial quality diagnosis by channel | Available with standard reporting | **Actuals Report with channel breakdown** |
| Event configuration | Predefined + custom events | **25 subscription-optimized standard events** |
| Minimum contract | Annual, $10K+ | **Pay-as-you-go, $0.05/install** |
| Free tier | Limited or none | **15K free attributed installs, all features included** |
## The Algorithm Is Only as Good as the Signal You Send It
The trial event you send to Meta and Google is not just a data point. **It is a training signal that shapes who the algorithm targets next.** For fitness apps, where the gap between "started a trial" and "willing to pay for a workout plan" is wide, raw trial signals systematically train the algorithm to find the wrong users.
Every week you keep sending raw trial events, the algorithm drifts further from the users who actually pay. CPS climbs. Subscriber growth stalls. And the worst part: your dashboard never warns you. Trial volume stays steady, CPI holds flat, and by the time you notice the damage, the algorithm has been training on contaminated data for months. The fix is not complicated, but the cost of waiting is.
Qualified trials fix the input. Channel-level trial quality measurement validates the output. **Start now, before your next campaign cycle compounds the problem further.**
**[See which channels drive trials that convert. Start free with Airbridge Core Plan.](https://www.airbridge.io/en/plans/core)**
---
## Creative Testing for Fitness Apps: Why the Winner by Installs Loses by Revenue
URL: https://www.airbridge.io/en/blog/creative-testing-fitness-apps-install-metrics-vs-subscription-revenue
Language: English
Published: 2026-03-16
Author: Jaehyuk Kim
You tested five creatives last week. One of them crushed it: $2.10 CPI, double the installs of the next best performer. You scaled it. You killed the $4.80 CPI creative that barely hit your install threshold.
**The $2.10 creative had a 1.5% trial-to-paid conversion rate. The $4.80 creative had 9%. You just killed your best revenue driver and scaled your worst.** This is not a hypothetical. It is the default outcome when fitness app teams rank creative winners by installs, because CPI is the only creative-level metric most tools show them.
**Key Takeaways**
- **Your lowest-CPI creative is often your lowest-revenue creative.** Flashy hooks drive curiosity installs, not committed subscribers.
- **No single tool shows creative-to-subscriber performance.** Ad platforms stop at installs. Creative analytics tools do not integrate with MMPs. Most MMP setups lock creative-level data behind premium tiers.
- **Fitness apps are especially vulnerable.** Health & Fitness apps have a median trial-to-paid conversion of 39.9% ([RevenueCat, 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)), but the variance by creative is where budget gets wasted.
- **The fix: replace CPI with Cost per Subscriber (CPS) as your creative ranking metric** and connect creative IDs to subscription events through your MMP.
- **Airbridge Core Plan connects creative performance to subscription revenue** with standard events, native RevenueCat/Adapty integration, and creative-level breakdown in the Actuals Report. Start free with 15K attributed installs.
## The CPI Trap: Why Your Best Creative by Installs Loses by Revenue
Creative testing is the core of paid UA for fitness apps. Teams test 5-20 creatives per week across Meta, Google, and TikTok, and the winner gets the budget. **The problem is how the winner is chosen.**
### 1\. CPI Picks the Wrong Winner
Most teams judge creative performance by CPI because it is the fastest metric available. Within 24-48 hours, you can see which creative drives the cheapest installs. **But CPI measures the cost of getting someone to download your app, not the cost of getting someone to subscribe.**
Consider two fitness app creatives running on Meta:
| Creative | CPI | Installs (1K spend) | Trial Starts | Trial-to-Paid | Subscribers | Cost per Subscriber |
| --- | --- | --- | --- | --- | --- | --- |
| A: "30-Day Transformation" | $2.10 | 476 | 143 (30%) | 1.5% | 2 | $500 |
| B: "Your First Coached Workout" | $4.80 | 208 | 104 (50%) | 9% | 9 | $111 |
Creative A wins by CPI. Creative B wins by revenue by a factor of 4.5x. If you kill Creative B at the 48-hour CPI check, you never see its subscription performance. **The data that matters most arrives after the decision has already been made.**
### 2\. Why Fitness Apps Are Especially Vulnerable
The gap between install intent and subscription intent is category-dependent. For a utility app, a user who installs is likely to use it. **For a fitness app, a user who installs is often someone who was curious, not someone ready to commit to a $14.99/month workout plan.**
[Same creatives ranked by CPI vs Cost per Subscriber — the winner flips depending on the metric]
RevenueCat's [State of Subscription Apps report](https://www.revenuecat.com/state-of-subscription-apps-2025/) shows Health & Fitness apps have a median trial-to-paid conversion of 39.9%. But this is the median across all users. **The variance by creative is where the real story lives.** A transformation before/after creative attracts aspirational browsers. A structured workout preview attracts people who already exercise. Both install the app. Only one type subscribes.
Fitness app creatives that perform best by CPI tend to target the widest emotional trigger, which also attracts the least committed users. This is the CPI trap: optimizing for volume instead of value.
### 3\. No Single Tool Connects Creative to Subscriber
The creative-to-subscriber gap is not a knowledge problem. Most growth marketers know CPI is incomplete. **It is a tooling problem.** No single tool in the typical fitness app stack connects creative IDs to subscription outcomes. Understanding [what an MMP does](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) helps clarify why this gap exists: attribution platforms were built to measure installs, not the revenue events that follow weeks later.
- **Ad platforms** (Google, Meta, TikTok, Apple Search Ads) show creative-level CPI, clicks, impressions, but nothing about subscription revenue. The ad platform knows which creative the user saw. It does not know whether that user subscribed, renewed, or churned. Each platform reports in its own silo. Even if you pull creative reports from all four channels, the best you get is CPI by creative. None of them can tell you which creative drove a $14.99/month subscriber versus a user who uninstalled on Day 2.
- **Creative analytics tools** like Motion were designed for e-commerce workflows. They pull data from ad platforms and tag creatives by visual element, hook type, or format. But they do not integrate with MMPs, which means they cannot access in-app subscription data. You can see which creative has the best CTR or the lowest CPI. You cannot see which creative produces the highest trial-to-paid rate.
- **Enterprise MMPs** have the attribution and the billing data. But creative-level subscription breakdowns are often locked behind premium tiers (e.g., AppsFlyer's ROI360 tier includes Creative Optimization). The data is technically available. It is not accessible at the price point where most subscription apps operate. A fitness app spending $10K-$30K/month on UA often cannot justify a $60K+ annual MMP contract just to unlock creative-level revenue data.
The result is a three-way data silo. The ad platform holds the creative ID. The MMP holds the attribution. The billing platform holds the subscription revenue. **Getting all three into the same row of a spreadsheet requires either expensive enterprise tooling or hours of manual data stitching every week.** Most teams give up and default to CPI.
[Three-way tooling gap: Ad platforms, creative analytics tools, and traditional MMPs each miss a critical piece]
## How to Evaluate Creatives by Subscription Value
### 1\. Replace CPI With CPS and Extend Your Decision Window
**Define CPS as your primary creative ranking metric:**
- **CPS = Total Creative Spend / Subscribers Attributed to That Creative**
- A $5 CPI creative with 8% trial-to-paid will always beat a $2 CPI creative with 1.5% trial-to-paid on CPS
- CPI remains useful for filtering (creatives with extremely high CPI may not get enough volume to test), but the winner should be picked by CPS
**Extend the decision window from 48 hours to 7-14 days:**
- **Day 0-2:** Monitor for delivery issues only. Do not make performance judgments.
- **Day 3-7:** Check trial start rate by creative. Creatives with high installs but low trial starts are already showing intent-quality issues.
- **Day 7-14:** Evaluate CPS. This is the decision point. Creatives that win on CPI but lose on CPS get cut. Creatives with higher CPI but strong CPS get scaled.
- **Day 14+:** For creatives still running, check Day-14 retention by creative. This filters for long-term subscriber quality.
**The key shift: patience.** It prevents killing your best revenue-driving creative based on a metric that does not predict revenue.
[Subscription-aware creative testing timeline: Day 0-2 monitor, Day 3-7 trial starts, Day 7-14 evaluate CPS]
Most teams feel pressure to declare winners fast because ad platform budgets auto-optimize toward lower CPI. The discipline to hold creatives through a full 7-14 day evaluation window is what separates teams that scale revenue from teams that scale installs.
There is a practical challenge here: holding budget on a creative with a $4.80 CPI while a $2.10 CPI creative is available feels wrong. The CPI-based instinct says you are wasting money. But if the $4.80 creative produces subscribers at $111 each while the $2.10 creative produces them at $500 each, every day you run the cheaper-CPI creative is a day you are actively losing money. **The discomfort of paying more per install is the cost of finding subscribers.**
**This measurement requires an** [**attribution system**](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) **that connects creative IDs to subscription events at the channel level.**
[**See which creative actually drives your subscribers. Start free with 15K attributed installs.**](https://www.airbridge.io/en/plans/core)
### 2\. How Airbridge Core Plan Connects Creative to Subscription Revenue
Airbridge Core Plan is an [MMP built for subscription apps](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) that connects ad platform creative IDs to subscription events, closing the gap between creative spend and subscriber value. Core Plan focuses on the four channels that matter most to subscription app growth: **Google, Meta, Apple Search Ads, and TikTok (GMAT)**. Native integrations with all four mean creative IDs flow into attribution automatically.
Core Plan provides subscription-optimized standard events (Start Trial, Subscribe, Unsubscribe, Order Complete, Order Cancel) with native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S. **The Actuals Report breaks down these subscription events by creative dimension,** so you can see which creative drives trial starts, which drives subscriptions, and which drives cancellations across Google, Meta, Apple Search Ads, and TikTok.
The creative ID flows from the ad platform through attribution, and the subscription event flows from the billing platform through S2S. Both meet in the same report. No manual CSV exports. No weekly data stitching. The connection is automatic and continuous.
| Capability | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Creative-level subscription data | Available (e.g., AppsFlyer ROI360 tier includes Creative Optimization) | Included in base with creative dimension built in |
| Billing platform integration | Native integrations available (RevenueCat, Adapty supported) | Native RevenueCat/Adapty via S2S |
| Creative-to-CPS visibility | Available (premium tiers for full cost + revenue data) | Actuals Report with creative breakdown |
| Event configuration | Predefined + custom events | 25 subscription-optimized standard events |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, $0.05/install |
| Free tier | Limited or none | 15K free attributed installs, all features included |
## The Problem Is Not Your Creatives. It Is How You Measure Them.
For fitness apps, the creative that wins by installs and the creative that wins by subscriptions are rarely the same. **If your creative testing loop ranks winners by CPI, you are systematically scaling the wrong creatives and killing the right ones every week.**
Think about what that means over a quarter. You test 10-15 new creatives per month. Each time, CPI picks the winner. Each time, the real revenue driver gets cut at the 48-hour mark. After three months, you have scaled dozens of high-install, low-subscriber creatives and killed dozens of low-install, high-subscriber ones. Your CPI looks great. Your subscriber acquisition cost has quietly doubled. **By the time the revenue numbers catch up to the creative decisions, the damage is already compounded across hundreds of thousands of dollars in ad spend.**
The fix is not more creatives or a bigger budget. It is not a better hook or a new ad format. It is connecting creative performance to the metric that actually matters: subscription revenue. Every week you delay making this connection is another week of scaling the wrong creative and killing the right one.
[**Stop ranking creatives by installs. See which creative drives subscribers. Start free with Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## MMP SDK Setup Takes Longer Than Your First Campaign — Something Is Wrong
URL: https://www.airbridge.io/en/blog/mmp-sdk-setup-takes-longer-than-first-campaign
Language: English
Published: 2026-03-14
Author: Jaehyuk Kim
A three-person fitness app team launches Meta and Google campaigns on Monday. By Friday, installs are climbing. But the [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) SDK is still stuck in QA. The developer is debugging event callbacks. The marketer is waiting for attribution data. And the campaigns that needed measurement the most are running completely blind.
**For growth-stage subscription apps spending $5,000 to $20,000 per month on paid acquisition, every week without attribution is a week of budget allocated by guesswork.** According to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/), the median subscription app earns just $92/month in revenue after one year. At that scale, two weeks of misallocated ad spend can consume an entire month of subscription revenue.
**Key Takeaways**
* **MMP SDK setup typically takes 2 to 3 weeks for small teams.** The delay comes from event schema design, cross-platform testing, and marketer-to-developer handoff gaps.
* **The bottleneck is decisions, not code.** Choosing which events to track, mapping subscription states, and configuring channel postbacks require coordination between marketing and engineering.
* **Health and fitness apps face an added timing penalty.** Seasonal spikes like January resolutions create narrow acquisition windows where setup delays mean missed attribution for the highest-spend period.
* **Campaigns running without attribution produce unreliable data.** Budget decisions based on platform-reported installs rather than actual subscription conversions lead to scaling the wrong channels.
* **Airbridge Core Plan reduces setup decisions with 25 subscription-optimized standard events and guided onboarding.** Growth-stage teams get first attribution signals in hours rather than weeks.
## Why MMP SDK Setup Time Is a Growth Blocker for Subscription Apps
MMP SDK setup time is the elapsed time from starting SDK integration to receiving reliable [attribution](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution) signals in a dashboard. The problem is structural: the setup process requires coordination across marketing and engineering that small teams are not equipped to handle quickly.
### 1. Event Schema Design Stalls the Process Before Code Is Written
Before a developer writes a single SDK call, someone needs to decide what events to track. For subscription apps, this means defining events for installs, sign-ups, trial starts, subscriptions, renewals, cancellations, and billing failures. Each event needs parameters: revenue amount, currency, plan type, trial duration.
Enterprise [MMPs](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) support both predefined and custom events, which gives flexibility but requires schema design upfront. A growth marketer knows they need to track "trial start to subscription" conversion by channel. But translating that into an event schema with correct parameter names, data types, and validation rules requires engineering input.
This handoff between marketing intent and technical implementation is where most setup timelines stretch from days to weeks. The marketer writes a brief: "I need to see which Meta ad sets produce subscribers." The developer responds: "What event name? What parameters? Should revenue be gross or net? Should I include currency conversion?" These questions are valid, but each one stalls progress until both sides agree.
For fitness apps using [billing platforms like RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) or handling [in-app purchases through StoreKit](https://developer.apple.com/documentation/storekit/in-app_purchase), the subscription state machine adds another layer. Events like "billing retry," "grace period started," and "voluntary vs. involuntary churn" all need definitions. Most teams spend 3 to 5 days on schema design alone.
### 2. Cross-Platform Testing Multiplies the Timeline
A subscription app supporting iOS and Android needs separate SDK integrations for each platform. After implementing events, the team must verify that:
* Events fire correctly on both platforms
* Attribution data matches between the MMP dashboard and the billing platform
* [Attribution windows](https://www.airbridge.io/en/glossary/attribution-window) are configured consistently across channels
* Postbacks to Meta, Google, and Apple Search Ads return accurate data
A task that takes a dedicated engineer 2 to 3 days stretches to 1 to 2 weeks when SDK integration is not the only priority. For a three-person team where the developer also handles product work, MMP setup competes with feature releases and bug fixes.
The [GA4 vs MMP comparison](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it) adds another dimension: teams that relied on GA4 for initial analytics often discover that session-based web attribution does not map cleanly to mobile subscription funnels. Migrating from GA4 to an MMP introduces a second round of event mapping, extending the timeline further.
### 3. The Cost of Setup Delay Is Not Theoretical
For health and fitness subscription apps, setup timing is especially critical. January resolution-driven campaigns generate the year's highest install volume. A team that starts MMP setup in early January and finishes in late January has already missed two weeks of its most expensive acquisition period with zero attribution.
The real cost is not the setup itself. It is the budget decisions made without data during the setup window.
| Scenario | Ad Spend During Setup | Attribution Data | Budget Optimization |
|:---------|:---------------------|:----------------|:-------------------|
| 2-week setup delay | $5,000 to $10,000 | None | Guesswork |
| 3-week setup delay | $7,500 to $15,000 | None | Guesswork |
| Attribution from Day 1 | $0 (pre-launch) | Full visibility | Data-driven |
**Growth-stage teams typically spend $5,000 to $20,000 per month on paid UA.** Two to three weeks of blind spending is not a minor inconvenience. It is the difference between identifying your best channel in Week 1 and discovering it in Month 2.
Consider the economics: enterprise MMP contracts often start at $2,000 to $5,000 per month with annual commitments. For a team spending $10,000/month on ads, adding $3,000/month for an MMP means paying for attribution tooling before receiving any attribution data. The financial mismatch compounds the operational delay.
## How to Reduce MMP SDK Setup Time
### General Approaches Any Team Can Take Today
Reducing MMP SDK setup time does not require switching tools. These steps help any growth team move faster:
* **Define events before touching the SDK.** Create a shared document listing every event name, parameters, and expected values. Get marketing and engineering sign-off before implementation starts.
* **Prioritize the subscription funnel first.** Track install, trial start, and subscribe events on Day 1. Add secondary events like "add to cart" or "view product" in a later phase.
* **Use your billing platform as the source of truth.** If you already use [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) or Adapty for subscription management, route subscription events through their server-to-server integrations rather than reimplementing them in the MMP SDK.
* **Test one platform first.** Run the full setup on iOS or Android first, verify it works, then replicate. Parallel testing introduces debugging complexity that slows both platforms.
* **Block dedicated engineering time.** MMP setup that competes with sprint tasks stalls at every context switch. Allocate 2 to 3 focused days to complete integration.
These steps help significantly, but the fundamental constraint remains: setup speed scales inversely with the number of configuration choices the team must resolve.
### How Airbridge Core Plan Reduces Setup Decisions
Airbridge Core Plan was designed around a specific question: "Are paid users converting into subscriptions, and which channels are driving value?" That focus shapes every setup decision by removing it.
**Predefined subscription events eliminate schema design.** Core Plan includes 25 standard events optimized for subscription apps: Install, Sign-up, Start Trial, Subscribe, Unsubscribe, Order Complete, and more. Event names and parameters are predefined by Airbridge, which means the team skips the event naming, parameter mapping, and schema review cycle that typically consumes the first week of setup.
**GMAT channel integrations cover 80 to 90% of early-stage ad spend.** Core Plan supports four Self-Attributing Networks: Meta Ads, Google Ads, Apple Search Ads, and TikTok for Business. These typically account for the majority of growth-stage paid acquisition spend, eliminating the need to configure dozens of additional ad network integrations.
**Built-in billing platform integrations reduce backend work.** Core Plan integrates natively with RevenueCat and Adapty, the two most common subscription management platforms for early-stage apps. Subscription revenue events flow directly into attribution reports without requiring custom server-side implementation.
| Setup Dimension | Enterprise MMP | Airbridge Core Plan |
|:----------------|:--------------|:-------------------|
| Event configuration | Predefined + custom events | 25 subscription-optimized standard events |
| Channel integrations | All networks supported | GMAT (80 to 90% of spend) |
| Billing platform | Native (included in higher tiers) | Native (included in base) |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Setup guidance | General-purpose | Subscription-app focused onboarding |
**Core Plan intentionally limits scope to accelerate time-to-value.** No custom events, maximum two third-party integrations (e.g., RevenueCat + Amplitude), and GMAT channels only. These constraints are deliberate: they match the actual needs of growth-stage subscription apps without the configuration overhead of a full enterprise platform.
The result: growth teams get [subscription revenue attribution](https://www.airbridge.io/en/blog/subscription-app-growth-guide) and funnel analytics running in hours, not weeks. Six built-in reports (Actuals, Trend, Active User, Funnel, Retention, Revenue) surface which channels drive subscribers from the moment data starts flowing.
**Pricing removes the financial barrier during setup.** Core Plan starts with 15,000 free attributed installs and $0.05 per install after that. Pay-as-you-go pricing means teams pay for actual usage, not for attribution capacity they have not set up yet. There is no annual contract, no $2,000/month minimum, and no cost during the days before attribution data starts flowing.
For growth-stage teams evaluating their [first MMP](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025), this pricing structure eliminates the scenario where you pay for an enterprise tool while spending weeks configuring it. Attribution cost scales with your actual install volume, not with a sales team's contract terms.
[MMP SDK Setup Timeline: Enterprise MMP vs Airbridge Core Plan]
## FAQ: MMP SDK Setup Time for Subscription Apps
### How long does MMP SDK setup actually take for a small team?
The time splits roughly into event schema design (3 to 5 days), SDK integration and testing per platform (3 to 5 days each), and channel configuration and verification (2 to 3 days). When the developer also handles product work, each phase stretches further. Teams using predefined standard events rather than custom schemas can compress this to under a week.
### When should a team upgrade from Core Plan to a full MMP?
Growth teams typically upgrade when they need custom event tracking beyond the subscription funnel, additional ad network integrations beyond GMAT, raw data exports to a data warehouse, or advanced capabilities like fraud detection and agency access. Until then, Core Plan's focused attribution covers the install-to-subscription funnel that matters most for growth-stage teams.
## The Campaigns Running Right Now Need Attribution Right Now
Every day of MMP SDK setup is a day of ad spend without measurement. For growth-stage subscription apps, the question is not whether you need attribution. It is whether you can afford to wait weeks for it.
**[Start Free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core) and get subscription attribution running in hours, with 15K free attributed installs.**
---
## Billing Failures, Refunds, Renewals: The Subscription Events Your MMP Never Shows You
URL: https://www.airbridge.io/en/blog/subscription-events-mmp-never-shows-billing-failures-refunds-renewals
Language: English
Published: 2026-03-13
Author: Jay Kim
Your [MMP](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) tracks the install, attributes the trial start, and records the first payment. After that, the subscription lifecycle continues for months. Your MMP does not follow it.
**Renewals, billing failures, grace periods, refunds, cancellations. These events decide whether a subscriber is profitable. But none of them reach your attribution system.** RevenueCat or Adapty shows you aggregate totals. Your MMP shows you acquisition channels. No system connects both, which means your LTV-by-channel calculation is built on incomplete data.
**Key Takeaways**
- **Your MMP stops tracking after the first payment.** Subscription lifecycle events happen server-side and never connect back to the acquisition channel.
- **Billing errors account for 28% of Google Play subscription cancellations.** Most MMPs cannot show which channels have higher billing failure rates ([RevenueCat, 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/)).
- **LTV-by-channel without lifecycle data is a guess.** First-payment revenue multiplied by an average retention curve hides real differences between channels.
- **Full lifecycle tracking requires S2S integration with your billing platform** and channel-level attribution for every post-payment event.
- **Airbridge Core Plan provides subscription-optimized standard events with native RevenueCat/Adapty integration,** connecting billing events to acquisition channels. Start free with 15K attributed installs.
## The Post-Payment Blind Spot That Breaks Your LTV Calculations
If you run paid UA for a subscription app, your MMP's attribution engine answers one question well: which channel drove this install? For apps with 7-14 day trials, it also tracks trial starts and first payments. After that, every event that determines real subscriber value is invisible to your attribution system.
### 1\. What Your MMP Tracks vs What It Misses
| Event | Your MMP | RevenueCat / Adapty |
| --- | --- | --- |
| Install | Attributed to channel | -- |
| Trial Start | Attributed to channel | ✓ |
| First Payment | Attributed to channel | ✓ |
| Renewal (Month 2+) | Not tracked | ✓ |
| Billing Failure | Not tracked | ✓ |
| Grace Period Entry | Not tracked | ✓ |
| Refund | Not tracked | ✓ |
| Cancellation | Not tracked | ✓ |
Everything below the first payment line is invisible to your attribution system. Your billing platform knows exactly what happened. Your MMP has no idea.
[The Post-Payment Tracking Blind Spot: MMP tracks install, trial, and first payment but misses renewal, billing failure, and refund events]
**Why?** Subscription billing events are processed server-side by Apple and Google, not by the app itself. A device-side SDK can only detect subscription state changes when the user opens the app. For events like renewals or billing failures that happen while the app is closed, the signal is unpredictable or absent entirely. Without a server-to-server (S2S) connection between your billing platform and your [MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-tool), these events remain disconnected from the channel that drove the original install.
Consider a subscriber acquired through Meta in January. They renew in February, hit a billing failure in March (expired card), enter a grace period, and recover in April. That entire revenue journey is invisible in your MMP dashboard. The subscriber shows as one first-payment conversion in January and nothing after. Multiply this across thousands of subscribers and dozens of campaigns. The gap between what your billing platform knows and what your MMP reports grows every month.
### 2\. Billing Failures Are a Bigger Problem Than Most Teams Realize
RevenueCat's [State of Subscription Apps report](https://www.revenuecat.com/state-of-subscription-apps-2025/) shows billing errors account for 28% of Google Play subscription cancellations. On iOS, the rate is lower at 15%, but still significant. These are not users who chose to leave. They are users whose payments failed silently.
The same report shows Health & Fitness apps face an additional challenge: a 4.7% refund rate, the second highest across all app categories. Apps using hard paywalls see refund rates of 5.8% compared to 3.4% for freemium models, suggesting that aggressive monetization without sufficient value demonstration creates more post-purchase regret.
**If your MMP cannot track billing failures and refunds by channel, you cannot tell which acquisition sources have higher involuntary churn rates.** A channel that looks efficient by CPS might have a 15% billing failure rate at month 3, but that data point does not exist in your attribution system. You would continue scaling spend on a channel that is quietly leaking subscribers after the first payment.
### 3\. Average Retention Curves Hide Channel Differences
Most teams fill the lifecycle data gap by applying a single average retention curve across all channels. This assumes every channel's subscribers retain at the same rate, which is almost never true.
RevenueCat data shows the first renewal is the biggest churn point. Depending on plan type, 30-50% of subscribers churn at their first renewal. Annual subscriptions retain at 66% past the first renewal, while weekly plans retain only 52%. These rates vary significantly by acquisition channel, but without renewal data attributed to channels, the variance is invisible.
**Consider this hypothetical scenario** for a fitness app running $30K/month across three channels:
| Channel | CPS | Month-2 Renewal | Month-6 Renewal | 6-Month LTV |
| --- | --- | --- | --- | --- |
| Meta Ads | $45 | 70% | 35% | $85 |
| Google Ads | $55 | 85% | 65% | $180 |
| Apple Search Ads | $60 | 90% | 70% | $210 |
Meta has the lowest CPS but the lowest 6-month LTV. Apple Search Ads costs more per subscriber but produces 2.5x more value over 6 months. **A CPS-only budget decision would scale Meta and cut ASA. The lifecycle data recommends the opposite.**
[diagram-ltv-channel-gap.webp]
But without post-payment tracking by channel, this team would never see the difference. They would continue optimizing for the cheapest acquisition, not the most profitable.
## How to Close the Post-Payment Tracking Gap
### 1\. What Full Lifecycle Tracking Requires
Closing the post-payment blind spot requires three capabilities that most MMP setups lack:
- **Standard events beyond first payment.** Tracking the subscription lifecycle means defining events for every major transition: Subscribe (first paid conversion), Unsubscribe (voluntary cancellation), Order Complete (renewal payment processed), and Order Cancel (refund or involuntary cancellation). Without standardized event definitions, teams build custom events with inconsistent naming across channels, making cross-channel comparison unreliable.
- **S2S integration with your billing platform.** Post-payment events happen server-side. The only reliable way to capture them is through a server-to-server connection between your billing platform (RevenueCat or Adapty) and your MMP. This ensures every renewal, billing failure, and refund reaches the attribution system regardless of whether the user has the app open. Apps that implement [billing grace periods](https://www.airbridge.io/en/glossary/attribution-window) can recover subscribers who would otherwise churn to billing failures, but these recovery events are only actionable if they reach your attribution data.
- **Channel-level attribution for every lifecycle event.** Aggregate lifecycle data tells you total renewals increased by 200 this month. Channel-level data tells you which campaigns drove those renewals, which ones drove the billing failures, and which channels produce subscribers who renew past month 6. Every lifecycle event must connect back to the original attributed install to be actionable for budget decisions.
[**See which channels drive subscribers who actually renew. Start free with 15K attributed installs.**](https://www.airbridge.io/en/plans/core)
### 2\. How Airbridge Core Plan Connects Billing Events to Channels
Airbridge Core Plan is an [MMP built for subscription apps](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) running paid UA on Google, Meta, Apple Search Ads, and TikTok. It closes the post-payment blind spot by connecting billing platform data to channel-level attribution.
Core Plan provides 25 subscription-optimized standard events (including Start Trial, Subscribe, Unsubscribe, Order Complete, Order Cancel) with native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration. Lifecycle events flow from your billing platform into the attribution system automatically via S2S, with no custom backend required.
The Actuals Report attributes subscription events to channels and campaigns, so you can see billing failure rates and refund rates by acquisition source. The Retention Report shows which channels retain subscribers beyond the first payment, broken down by cohort. **Together, these reports replace first-payment LTV guesswork with actual lifecycle data by channel.**
| Capability | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Post-payment event tracking | Native integrations available (included in higher tiers) | Standard events via native billing integration (included in base) |
| Billing failure attribution | Available with standard reporting | Channel-level via Actuals Report |
| Renewal tracking by channel | Available (standard reporting) | Automatic via S2S integration |
| Lifecycle LTV by channel | Available (premium tiers) | Retention Report (included in base) |
| Event configuration | Predefined + custom events | 25 subscription-optimized standard events |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, $0.05/install |
| Free tier | Limited or none | 15K free attributed installs, all features included |
## Subscription Revenue You Cannot See Is Revenue You Cannot Optimize
Every month, your billing platform processes renewals, catches billing failures, issues refunds, and logs cancellations. Every month, none of that data reaches your attribution system. The subscriber who renewed six times and the subscriber who churned to a billing failure at month 2 look identical in your MMP: one first-payment conversion.
**That invisible gap is where your budget decisions go wrong.** Without lifecycle event tracking by channel, you cannot tell which channels produce $85 subscribers and which produce $210 subscribers. You optimize for CPS because it is the only post-install metric you can see, and you scale channels that look cheap but churn fast.
[**Connect billing events to acquisition channels. Start free with Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## The MMP Support Trap: Fast Answers During Sales, Silence After the Contract
URL: https://www.airbridge.io/en/blog/mmp-support-trap-fast-answers-sales-silence-after-contract
Language: English
Published: 2026-03-13
Author: Jaehyuk Kim
You submitted a support ticket on Monday about a trial-to-subscription attribution discrepancy. It is now Thursday. The response links to a generic help center article about SDK event configuration that does not address the specific RevenueCat integration issue you described.
**For subscription app teams running paid UA, a 3-5 day support delay is not just frustrating. It is 3-5 days of ad spend allocated based on incomplete or broken attribution data.** When your [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) cannot answer a specific question about subscription funnel tracking, budget decisions default to guesswork.
**Key Takeaways**
- **MMP customer support quality often degrades after contract signing.** The sales-stage responsiveness that convinced you to sign does not predict the post-sale support experience.
- **Canned ticket responses cost more time than silence.** A generic answer to a specific integration question forces a second ticket, a re-explanation, and another 2-3 day wait.
- **Support delays translate directly to misallocated ad spend.** Every day a subscription attribution issue goes unresolved is a day of budget decisions based on incomplete signals.
- **The structural problem is dependency on human support for routine questions.** Most MMP support requests involve report configuration, event verification, or integration setup that should not require a ticket.
- **Airbridge Core Plan reduces support dependency with an opinionated, self-serve setup.** With 25 predefined standard events and subscription-focused onboarding, routine questions are eliminated before they become tickets.
## Why MMP Customer Support Fails Subscription App Teams
The frustration with MMP customer support is not random. It follows a structural pattern that affects subscription app teams more than other verticals.
### 1\. The Sales-to-Support Gap
During the evaluation phase, [MMP vendors](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) are responsive. Sales engineers answer technical questions within hours. Demo environments are configured to your exact use case. Onboarding calls happen on schedule.
Then the contract closes. The sales engineer moves to the next prospect. Your day-to-day questions go to a general ticket queue. According to [G2 reviews comparing major MMPs](https://www.g2.com/compare/adjust-vs-appsflyer), enterprise customers with dedicated account managers report satisfactory support, while smaller teams frequently cite slow responses and difficulty resolving specific issues.
For early-stage subscription app teams, this gap is particularly acute. You are typically on the lowest pricing tier, which means you do not qualify for a dedicated Customer Success Manager (CSM). Your support channel is ticket-based, with response times measured in days, not hours.
ICP interviews with subscription app growth marketers confirm this pattern. Teams consistently report that their MMP's basic plan has no CSM, and that integration documentation is insufficient for subscription-specific setups. The result: teams that need the most hands-on help are the ones least likely to receive it.
### 2\. Why Subscription Apps Suffer More from Support Delays
Generic mobile apps can tolerate a 3-day support delay without major consequences. Subscription apps cannot.
The subscription funnel has a narrow attribution window. If your app offers a 7-day free trial, the events that matter most for [attribution and LTV](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) happen within days of install: trial start, subscription activation, and first renewal. A support delay during this window means an entire cohort's trial-to-paid data may be misattributed or missing entirely.
| Scenario | Support Delay | Attribution Impact |
| --- | --- | --- |
| SDK event not firing for Subscribe | 3 days | 3 days of subscriptions unattributed to any channel |
| RevenueCat integration mismatch | 5 days | Revenue data missing from Funnel and Revenue reports |
| SKAN conversion value misconfigured | 4 days | iOS campaign optimization signals degraded for 4+ days |
**The cost is not frustration. The cost is budget decisions based on incomplete data.** If you cannot see which channels drive subscriptions for an entire week, you are either pausing campaigns and losing momentum, or continuing to spend based on outdated signals.
### 3\. The Hidden Cost of Canned Responses
A slow response is bad. A canned response is worse.
When a support team replies to a specific subscription attribution question with a link to a generic help center article, the real cost is the second cycle. You re-explain the issue, provide additional context, resubmit the ticket, and wait another 2-3 days. Two response cycles at 3 days each means nearly a week of unresolved attribution gaps.
For subscription apps running [paid acquisition across multiple channels](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026), this compounds. Each day of missing channel-level subscription data is a day where budget allocation cannot be optimized.
Consider a health and fitness app running a January campaign during the New Year's resolution spike. Trial sign-ups are 3x the normal rate. A RevenueCat integration issue means subscription events are not reaching the MMP. The support ticket gets a canned response on day 2.
By the time the issue is resolved on day 6, the January spike has passed and six days of the highest-converting cohort of the year has no channel-level subscription attribution. The team cannot determine whether Meta, Google, or Apple Search Ads drove the subscribers who will generate the highest LTV.
This is not a hypothetical edge case. Seasonal conversion spikes in health and fitness apps create narrow windows where attribution accuracy matters most. Support delays during these windows have outsized budget consequences.
[Where MMP support tickets originate]
## How to Fix the MMP Support Problem
### Practical Steps to Reduce Support Dependency
Before evaluating new tools, there are steps any team can take to minimize the damage from slow MMP support:
- **Document every integration detail during onboarding.** Capture SDK version, event naming conventions, and integration configuration while the onboarding team is still responsive. This becomes your self-serve reference when support slows down.
- **Test attribution events before launching campaigns.** Use test consoles and sandbox environments to verify that trial start, subscribe, and revenue events fire correctly. Catching issues before spend begins eliminates the most urgent support tickets.
- **Set up monitoring for critical subscription events.** Build simple daily checks for key event counts (installs, trial starts, subscriptions). If numbers drop suddenly, you catch the issue in hours rather than discovering it days later in a weekly report.
- **Request escalation paths upfront.** During contract negotiation, ask for the specific process to escalate urgent attribution issues. Knowing the escalation path saves days when a critical SDK issue surfaces.
### How Airbridge Core Plan Addresses the Support Problem
The most effective way to reduce [MMP support dependency](https://www.airbridge.io/en/blog/mmp-pricing-goes-up-support-quality-goes-down) is to eliminate the questions that generate tickets in the first place. Airbridge Core Plan is built around this principle.
**Opinionated setup reduces configuration questions.** Core Plan uses 25 predefined standard events optimized for the subscription funnel: Install, Sign-up, Start Trial, Subscribe, Unsubscribe, and Order Complete. Event naming decisions, schema design choices, and parameter configurations that generate support tickets with enterprise MMPs are predefined. The difference versus custom events: event names and parameters are predefined by Airbridge, reducing schema design work.
**Subscription-focused onboarding eliminates integration ambiguity.** Instead of general-purpose MMP onboarding that covers dozens of use cases, Core Plan's onboarding path is built specifically for subscription apps running paid UA on Meta, Google, Apple Search Ads, and TikTok (GMAT channels). This focused scope means fewer configuration options that can go wrong.
**Built-in reports surface subscription signals without custom setup.** Core Plan includes 6 built-in reports (Actuals, Trend, Active User, Funnel, Retention, Revenue) that cover the subscription funnel out of the box. Report configuration issues that generate support tickets with enterprise MMPs are handled by default.
| Support Friction Point | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Event schema design | Custom event design, frequent tickets | 25 subscription-optimized standard events |
| Report configuration | Requires setup guidance | 6 built-in reports, pre-configured |
| Channel integration | Multiple networks, varied setup | GMAT focus (4 SANs, 80-90% of spend) |
| Billing platform integration | Native (included in higher tiers) | Native (RevenueCat, Adapty in base) |
| Pricing model | Annual contract, CSM gated by tier | Pay-as-you-go, 15K free installs |
**Intentional limitations that reduce complexity.** Core Plan does not support custom events, limits third-party integrations to 2, and focuses on GMAT channels only. These are not missing features. They are design decisions that remove the configuration surface area where support tickets originate.
**RevenueCat and Adapty integration included in the base offering.** Subscription revenue platforms are where the most support-intensive integration issues originate. Enterprise MMPs typically include these integrations in higher-tier plans. Core Plan includes RevenueCat and Adapty integration at the base tier, reducing the most common source of subscription-specific support friction.
**Pay-as-you-go pricing with 15K free attributed installs.** For teams spending $5,000-$20,000 per month on paid UA, enterprise MMP contracts of $2,000-$5,000/month are disproportionate. Core Plan's pricing model ($0.05/install after the free tier) aligns cost with actual usage, and pay-as-you-go means no annual lock-in if the tool does not meet expectations.
The question for subscription app teams is whether the support model matches the team's operational reality. If you are a 3-person growth team without a dedicated analytics engineer, a self-serve attribution tool that eliminates routine support tickets is more valuable than a dedicated CSM you never receive.
[MMP support delay impact on subscription attribution]
## FAQ: MMP Customer Support for Subscription Apps
### How long should MMP support take to resolve a subscription attribution issue?
For critical attribution issues affecting live campaigns, resolution within 24 hours is the baseline expectation. For configuration questions, same-day responses keep the team moving. If your MMP consistently takes 3-5 days to respond to specific technical questions, the support model is not built for subscription app velocity.
### What should subscription app teams evaluate about MMP support before signing a contract?
Ask three questions: What is the average response time for technical support tickets (not sales inquiries)? Do teams at your pricing tier receive a dedicated CSM or ticket-based support? What percentage of support requests are resolved in the first response versus requiring escalation? The answers reveal whether the vendor's support model matches your team's needs. [Comparing MMP options](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026) on support quality is as important as comparing features.
### When should a subscription app team switch MMPs because of support quality?
If support delays have caused measurable attribution gaps more than twice in a quarter, or if routine configuration questions consistently require 3+ days for resolution, the support model is structurally misaligned with your team's needs. Evaluate whether a [self-serve MMP](https://www.airbridge.io/en/blog/why-mobile-attribution-numbers-dont-match-ad-platforms-vs-mmp-vs-first-party-data) with predefined subscription events would eliminate the questions generating those tickets. Switching costs are real, but so is the compounding cost of broken attribution data.
## The Attribution Data You Cannot See Is Budget You Cannot Optimize
Every subscription app team running paid acquisition faces a choice: depend on a support queue for answers about your own data, or use a tool that surfaces those answers without a ticket.
The cost of slow MMP customer support is not measured in frustration. It is measured in days of ad spend allocated without complete attribution signals. For [subscription apps scaling paid UA](https://www.airbridge.io/en/blog/subscription-app-growth-guide), those days compound.
[**Start Free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core) and see which channels drive trial-to-subscription conversions, starting with 15K free attributed installs.
---
## Your Ad Algorithm Is Learning the Wrong Lesson — How Dirty Signals Drain Subscription App Budgets
URL: https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets
Language: English
Published: 2026-03-13
Author: Jaehyuk Kim
Every Monday, the same scene plays out across subscription app growth teams. CPI from Meta looks great. Install counts are climbing. But cost-per-subscriber keeps getting worse, and no one on the team can explain why.
**The problem is not your creatives, your targeting, or your budget. The problem is what you are telling Meta and Google to optimize for.** Every conversion signal you send teaches the algorithm what a "good user" looks like. If those signals include impulse cancellers, trial-only users, and unverified installs, the algorithm learns the wrong lesson and spends your budget finding more of the same.
**Key Takeaways**
* **Ad platform algorithms optimize toward whatever signal you send them.** If you optimize for installs, the algorithm finds cheap installs, not paying subscribers.
* **Impulse cancellations contaminate your conversion data.** Users who subscribe and cancel within minutes still register as successful conversions, teaching the algorithm to target the wrong audience.
* **Subscription apps face a structural signal delay.** The event that matters most (paid subscription) happens days or weeks after install, often outside the ad platform's optimization window.
* **Signal engineering is the systematic fix.** Choosing, filtering, and timing conversion events to match actual subscriber value is the practice top growth teams use to improve ad campaign performance.
* **Airbridge Core Plan makes signal engineering accessible.** Standard subscription events, native RevenueCat and Adapty integration, and GMAT channel connections enable cleaner signals starting with 15K free attributed installs.
## Why Ad Campaigns Fail When Your Conversion Signals Are Wrong
Signal engineering is the practice of choosing, filtering, and structuring conversion events to improve how ad platform algorithms learn and optimize ([Sub Club / Thomas Petit](https://subclub.com/episode/signal-engineering-strategic-data-filtering-for-better-ad-performance-thomas-petit-independent-consultant)). The core principle is straightforward: send the ad platform better data, and it produces better results.
For subscription apps, the signals are almost always wrong. Here is why that happens, and what it costs.
### 1. The Algorithm Learns Whatever You Teach It
Most subscription apps optimize for Install events because installs fire immediately and generate enough volume for the algorithm's learning phase. Meta requires [approximately 50 optimization events per ad set per week](https://www.facebook.com/business/help/112167992830700) to exit the learning phase. Subscribe events rarely hit that threshold for early-stage apps, so teams default to installs.
The optimization event you choose determines what the algorithm learns:
| Optimization Event | What the Algorithm Learns | Campaign Outcome |
|:-------------------|:-------------------------|:-----------------|
| Install | Find users who download free apps | Low CPI, high cost-per-subscriber |
| Trial Start | Find users who start trials, including impulse cancellers | More trials, low trial-to-paid conversion |
| **Subscribe** | **Find users who actually pay** | **Higher CPI, lower cost-per-subscriber** |
When you optimize for installs, you get exactly what you asked for: cheap installs from users who were never likely to subscribe. **The algorithm is working perfectly. Your signal is the problem.**
According to a [Lunio study analyzing 2.7 billion ad clicks](https://www.mediapost.com/publications/article/412156/ad-spend-wasted-on-invalid-traffic-reaches-63b.html), 8.51% of all paid ad traffic is invalid, costing advertisers an estimated $63 billion globally. For subscription apps, the waste compounds further because even valid installs from low-intent users degrade the optimization signal over time.
[Signal Engineering Hierarchy: Install at the bottom, Subscribe/Renew at the top]
### 2. Impulse Cancellations Poison Your Lookalike Models
Subscription apps see a specific form of signal contamination that most other app categories do not: impulse cancellations.
A user taps "Start Free Trial," gets charge anxiety, and cancels within 10-15 minutes before opening the app a second time. That Subscribe event still reaches Meta or Google as a valid conversion. **The algorithm treats this user as a success and builds lookalike models around them.**
According to RevenueCat's State of Subscription Apps 2025 report, nearly 30% of annual subscriptions are cancelled within the first month. For weekly plans, the first renewal rate is only 51.8%. A significant portion of those "conversions" are users who never intended to stay.
Each optimization cycle compounds the damage:
* **Week 1-2**: Campaign learns from a mix of real subscribers and impulse cancellers
* **Week 4**: The algorithm's predictive accuracy for high-LTV users decreases as impulse cancel ratio increases
* **Week 8**: Cost-per-subscriber climbs, but CPI looks unchanged. The degradation is invisible in the dashboard
* **Week 12**: Team resets the learning phase, losing another 2-4 weeks of optimization data
[How dirty signals compound vs. clean signals over 12 weeks]
### 3. The Trial Delay Gap Breaks the Optimization Loop
Many subscription apps offer a 7-14 day free trial. If the [attribution window](https://www.airbridge.io/en/glossary/attribution-window) is shorter than the trial period, the subscription event never links back to the ad that drove it.
**For apps using standard 7-day free trials, the revenue event occurs outside the default optimization window of most ad platforms.** These conversions are counted as organic, hiding the true value of paid channels.
Consider a fitness app spending $15K/month across Meta and Google. The app offers a 14-day free trial. With a standard 7-day click-through attribution window, every subscription that converts on day 8 or later is invisible to the ad platform. If 60% of conversions happen after day 7, the algorithm is learning from less than half the actual subscriber data. The campaigns that produce the best long-term subscribers may appear to underperform because their conversions fall outside the window.
On top of the timing gap, server-side payments create an additional delay. When a user subscribes, the payment is processed by Apple or Google, not by the app. Device-side SDKs can detect subscription state changes only when the user opens the app again, creating latency and dependency on user behavior. **For real-time, independent signal transmission, the [MMP](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) needs a server-to-server connection** that most small teams have not built.
Revenue data lives in [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) or [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration). Ad platforms know nothing about subscription revenue. **Without a bridge between billing and attribution, campaigns cannot optimize for what actually drives the business.**
## How to Engineer Signals That Drive Subscribers Instead of Installs
The good news: signal engineering is a well-defined practice, not a mystery. According to a Funnel.io analysis of internal studies by Meta, TikTok, and LinkedIn, improving signal quality can increase conversions by 24% and lower cost per action by 15% ([Funnel.io](https://funnel.io/blog/signal-engineering)).
### Practical Signal Engineering Steps Any Team Can Apply Today
These steps work regardless of your attribution tooling:
* **Move your optimization event closer to revenue.** Switch from Install to Subscribe or Start Trial. If subscription volume is too low for the learning phase (50 events/week), Start Trial is the next best proxy.
* **Filter impulse cancellations.** Define a "qualified trial" window (2-4 hours after trial start). Only fire the conversion event after the threshold passes. Users who cancel before the window closes are excluded.
* **Extend attribution windows to cover the full trial period.** If your trial is 14 days, configure a 21-30 day click-through window. Every subscription outside the window is [counted as organic](https://www.airbridge.io/en/blog/why-mobile-attribution-numbers-dont-match-ad-platforms-vs-mmp-vs-first-party-data), hiding the true value of paid channels.
* **Connect billing data to attribution.** Build a server-to-server integration that matches subscription events to attributed installs and forwards them to ad platforms. Budget at least a week of engineering time for edge cases like billing retries, grace periods, and family sharing.
* **Track Cost per Subscriber (CPS), not just CPI.** CPS equals total ad spend divided by paid subscriptions attributed to that channel. Evaluate CPS alongside [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) to determine true channel profitability. A channel with high CPS but higher LTV may still be your best source.
* **Audit signal accuracy weekly.** Compare the conversion counts your ad platform reports against your [MMP](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) and billing platform. If discrepancies exceed 10-15%, something in the signal pipeline is broken.
### How Airbridge Core Plan Enables Cleaner Signals at Scale
Each of the steps above requires infrastructure that most early-stage subscription app teams do not have. Core Plan was designed to solve this exact problem: getting clean subscription signals into ad platform optimization loops without custom engineering.
**Standard subscription events, not custom event schemas.** Core Plan supports 25 standard events including Start Trial, Subscribe, Unsubscribe, and Order Complete. These events map directly to the signal engineering hierarchy (install < trial < subscribe < renew). Teams can set their optimization event to Subscribe or Start Trial from day one without designing custom event schemas.
**Native billing platform integration.** [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration connects verified subscription status to attribution data. This means the [MMP](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) knows the difference between a genuine subscriber and an impulse canceller. Teams get cleaner signal data without building custom backend pipelines.
**GMAT channel integrations.** Clean signals reach Meta, Google, Apple Search Ads, and TikTok through native [Self-Attributing Network](https://www.airbridge.io/en/glossary/self-attributing-network) integrations. These four channels cover 80-90% of early-stage paid acquisition spend.
**Configurable attribution rules.** Teams can set attribution windows that cover their full trial period. [SKAN Conversion Value Settings](https://www.airbridge.io/en/blog/asa-and-skan-to-optimize-ios-campaign) optimize iOS signal quality within Apple's [post-ATT](https://www.airbridge.io/en/blog/post-att-advertising-strategies) privacy framework.
**Intentional simplicity.** Core Plan does not support custom events or more than 2 third-party integrations (e.g., RevenueCat + Amplitude). This is by design. Standard events cover the full subscription funnel without requiring schema planning, and the focused feature set reduces decisions during setup.
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:--------------|:-------------------|
| Subscription event tracking | Predefined + custom events available | 25 subscription-optimized standard events |
| Billing platform integration | Native (included in higher tiers) | Native (included in base) |
| Signal engineering setup | General-purpose, more configuration | Subscription-app focused, fewer decisions |
| Attribution window config | Configurable | Configurable for trial periods |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Unnecessary features for signal engineering | Fraud detection, raw export, agency access included | Intentionally removed |
**Pricing**: 15K free attributed installs, then $0.05/install. No annual contract, no add-on pricing for subscription events or billing integrations.
**How signal engineering works with Core Plan in practice:**
1. Install Core Plan SDK and configure standard subscription events (Start Trial, Subscribe, Unsubscribe)
2. Connect RevenueCat or Adapty to pipe verified billing data into attribution
3. Set attribution windows to cover your trial period (e.g., 21 days for a 14-day trial)
4. Choose Subscribe or Start Trial as your ad platform optimization event
5. Monitor the Funnel report to track install-to-trial-to-subscription conversion by channel
The entire setup takes hours, not weeks. Once the pipeline is running, clean signals flow to Meta, Google, Apple Search Ads, and TikTok automatically.
## FAQ: Ad Campaign Signal Engineering for Subscription Apps
### What is signal engineering in mobile marketing?
Signal engineering is the practice of choosing, filtering, and timing conversion events sent to ad platforms to improve how their algorithms optimize campaigns. Instead of sending every event (installs, all trial starts), teams send only high-quality signals (qualified trials, verified subscriptions) that represent actual business value. The term was popularized by Thomas Petit and RevenueCat's Sub Club podcast.
### Why can't GA4 handle ad campaign optimization for subscription apps?
GA4 uses a session-based attribution model designed for web analytics. It cannot reliably connect a mobile ad click to an in-app subscription event that occurs days later, across different sessions and devices. For subscription apps running paid acquisition on Meta, Google, and TikTok, an MMP provides the cross-channel, cross-session attribution that GA4 lacks. [Read more about GA4 limitations for subscription apps](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it).
### How many conversion events does Meta require to exit the learning phase?
Meta recommends approximately [50 optimization events per ad set per week](https://www.facebook.com/business/help/112167992830700) to exit the learning phase. If your subscription volume is below this threshold, optimize for a higher-volume proxy event (Start Trial) and filter impulse cancellations before they reach the ad platform.
## Clean Signals Are the Foundation Every Other Ad Campaign Optimization Depends On
Every dollar spent while your signals are dirty is a dollar training Meta and Google's ML models to find users who will never pay. Creative testing, audience targeting, bid optimization: none of these work if the underlying signal tells the algorithm that impulse cancellers are your ideal customers.
**Cleaning your signals is not one optimization tactic among many. It is the foundation that every other ad campaign optimization depends on.**
**[Stop training your algorithm on the wrong users. See which campaigns actually drive subscribers with Airbridge Core Plan, starting with 15K free attributed installs.](https://www.airbridge.io/en/plans/core)**
---
## Your Marketing Dashboard Is Missing the Only Metric That Matters — Cost Per Subscriber by Channel
URL: https://www.airbridge.io/en/blog/marketing-dashboard-cost-per-subscriber-by-channel
Language: English
Published: 2026-03-13
Author: Jaehyuk Kim
A fitness app growth team runs Meta, Google, and Apple Search Ads campaigns. Meta Ads Manager reports 4,200 installs at $2.38 CPI. Google shows 2,800 installs at $3.10 CPI. RevenueCat logs 189 paid subscribers for the month. But no single dashboard connects those 189 subscribers back to the channel that drove them.
The team knows total subscriber count. They know total ad spend. **What they cannot see is cost per subscriber by channel, which means every budget reallocation is based on install volume, not subscription revenue.** For a fitness app spending $15,000 per month on paid acquisition, optimizing for CPI instead of CPS (Cost per Subscription) can mean scaling the channel that produces the cheapest installs while starving the one that produces the most paying subscribers.
**Key Takeaways**
* **Marketing dashboards show installs, not subscribers.** Ad platforms track clicks and installs but have no visibility into subscription events that happen days or weeks later.
* **CPI and CPS often point in opposite directions.** The channel with the lowest cost per install frequently has the highest cost per subscriber, making install-based optimization counterproductive for subscription apps.
* **The gap exists because subscription data lives in a separate system.** RevenueCat or Adapty tracks who subscribed, but that data never connects back to the ad platform natively.
* **Health and fitness apps face compounding seasonal distortion.** January resolution spikes inflate install numbers while March churn erodes subscriber counts, making channel-level CPS even harder to track without attribution.
* **Airbridge Core Plan connects ad spend to subscription revenue in one view.** With native RevenueCat and Adapty integrations, it surfaces cost per subscriber by channel without requiring custom data pipelines.
## Why Marketing Dashboards Hide Cost Per Subscriber
A [marketing dashboard](https://www.airbridge.io/en/blog/marketing-dashboard-cost-per-subscriber-by-channel) that only shows CPI creates a structural blind spot for subscription apps. The problem is not a feature missing from one tool. It is an architectural gap between three systems that were never designed to talk to each other.
### 1. Ad Platforms, Billing Systems, and Analytics Operate in Silos
When a user subscribes, the payment is processed by Apple or Google, not by the ad platform that drove the install. RevenueCat, Adapty, or the App Store tracks who subscribed, renewed, or churned. Meta and Google know nothing about this.
This creates three disconnected data layers:
| System | What It Knows | What It Cannot See |
|:-------|:-------------|:-------------------|
| Ad platforms (Meta, Google, TikTok) | Clicks, installs, CPI | Trial starts, subscriptions, renewals |
| Billing platforms (RevenueCat, Adapty) | Subscribers, revenue, churn | Which ad or channel drove each subscriber |
| Analytics tools (GA4, Amplitude) | In-app behavior, sessions | Channel-level subscription attribution |
Without a [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) bridging these layers, CPS by channel remains invisible.
[Three data silos: ad platforms, billing platforms, and analytics tools]
### 2. The CPI vs. CPS Inversion Problem
The disconnect creates a measurable budget risk. Consider this fitness app scenario:
| Channel | Installs | CPI | Subscribers | CPS | Budget Signal |
|:--------|:---------|:----|:-----------|:----|:-------------|
| Meta Ads | 4,200 | $2.38 | 52 | $192 | Reduce |
| Google Ads | 2,800 | $3.10 | 98 | $89 | **Scale** |
| Apple Search Ads | 1,000 | $5.20 | 39 | $133 | Optimize |
A dashboard showing only CPI would scale Meta ($2.38 CPI) and cut Apple Search Ads ($5.20 CPI). But the subscription data tells the opposite story: Google produces subscribers at nearly half the cost of Meta. **Without CPS visibility, the team systematically scales the most expensive subscriber channel while cutting efficient ones.**
[CPI vs CPS inversion across Meta, Google, and Apple Search Ads]
This is not a hypothetical risk. According to the [RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/) report, the median Health & Fitness app trial-to-paid conversion rate is 39.9%, but this rate varies dramatically by acquisition channel. Teams that cannot measure channel-level conversion are optimizing blind.
### 3. Algorithm Optimization Without CPS Feeds the Wrong Signal
Meta's Advantage+ campaigns, Google's App Campaigns, and TikTok's Smart Performance Campaigns all use machine learning to find users most likely to complete a target event. If that target event is "install," the algorithm optimizes for users who download quickly, not users who subscribe.
This creates a compounding problem. The algorithm learns from install signals, finds more install-prone users, and reports declining CPI as proof of efficiency. Meanwhile, the [trial-to-subscription conversion](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) rate silently declines because the new users were never high-intent subscribers. The marketing dashboard shows improving CPI while actual CPS climbs.
Without feeding subscription events back through an MMP, the growth team has no way to detect this signal degradation. They see a "healthy" dashboard and keep scaling spend into an algorithm that is optimizing for the wrong outcome.
### 4. Seasonal Spikes Amplify the Problem for Fitness Apps
Health and fitness apps face a unique attribution challenge. January resolution campaigns generate massive install spikes, but most subscription conversions happen 7 to 30 days after install, well past the resolution window. By March, churn accelerates and the true subscriber value of each January channel is buried under aggregate numbers.
The [RevenueCat State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/) report shows that Health & Fitness has one of the highest trial-to-paid rates (39.9% median), yet this metric varies sharply by acquisition source. A channel that drives 10,000 resolution installs in January may produce only 200 subscribers by March. A smaller channel with 2,000 installs may produce 300 subscribers who remain active through June. Without CPS by channel, the seasonal noise hides the signal.
This is especially damaging for fitness apps because the [AppsFlyer 2026 State of Subscriptions](https://www.appsflyer.com/resources/reports/subscription-marketing-report/) report found that Health & Fitness UA budget concentration among top apps increased from 54% to 73%. In a market where the biggest spenders dominate, smaller teams cannot afford to waste budget on high-CPI-low-CPS channels.
Growth teams that measure only aggregate subscriber count across all channels lose the ability to make channel-specific budget decisions during the highest-stakes period of the year.
## How to Build a Unified Marketing Dashboard for Subscription Revenue
The CPS visibility gap is solvable. It requires connecting attribution data to billing data so that every subscriber traces back to the channel, campaign, and creative that drove the original install.
### General Approaches to Unifying Dashboard Data
Growth teams can take immediate steps to close the CPS gap, regardless of their current tooling:
* **Align [attribution windows](https://www.airbridge.io/en/glossary/attribution-window) across all ad platforms.** Set Meta, Google, and TikTok to the same click and view-through lookback periods so comparisons are apples-to-apples.
* **Export RevenueCat subscriber data and join it with install-level attribution data in a spreadsheet.** Manual but possible for teams under 5,000 monthly installs.
* **Set up UTM parameters for web-to-app flows** to track which landing pages contribute to subscriptions.
* **Build a weekly CPS-by-channel report** using combined exports from your ad platforms and billing system. Even a manual process forces the team to look at subscriber acquisition cost, not just install cost.
These steps improve CPS visibility but remain manual, error-prone at scale, and disconnected from real-time decisions.
### How Airbridge Core Plan Unifies Cost Per Subscriber Tracking
Airbridge Core Plan was designed to answer one question: **"Are paid users converting into subscriptions, and which channels are driving value?"**
Instead of requiring growth teams to stitch together data from three separate systems, Core Plan connects [attribution](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution) signals directly to subscription billing data. This is how it works for subscription apps:
**1) Attribution across GMAT channels.** Core Plan integrates natively with Meta, Google, Apple Search Ads, and TikTok (the four [Self-Attributing Networks](https://www.airbridge.io/en/glossary/self-attributing-network) that represent 80 to 90% of early-stage paid acquisition spend). Every install is attributed to the channel, campaign, and creative that drove it.
**2) Subscription billing integration.** Native RevenueCat and Adapty integrations bring subscription events (Start Trial, Subscribe, Unsubscribe) directly into the attribution pipeline. No custom backend work required.
**3) Six built-in reports that surface CPS.** The Revenue report attributes subscription revenue by channel. The Funnel report shows where users drop off between install, trial, and subscription. Together, they surface cost per subscriber by channel in one view.
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:--------------|:-------------------|
| Billing integration (RevenueCat, Adapty) | Native (included in higher tiers) | Native (included in base) |
| CPS by channel report | Available (requires configuration) | Built into Revenue report |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Setup guidance | General-purpose | Subscription-app focused |
| Event configuration | Predefined + custom events | 25 subscription-optimized standard events |
| Unnecessary features | Fraud detection, raw export, agency access | Intentionally removed |
Core Plan does not support custom events or non-SAN ad networks. These are intentional design decisions. The 25 standard events cover the subscription funnel (Install, Sign-up, Start Trial, Subscribe, Unsubscribe), and GMAT channels cover the platforms where most subscription apps concentrate spend. Two third-party integrations (such as RevenueCat + Amplitude) keep the stack focused. The result is fewer setup decisions and faster Time-to-Value.
For teams spending $5,000 to $20,000 per month on paid UA, the pricing removes the cost barrier: 15K free attributed installs, then $0.05 per install. No annual contracts. No minimum commitments. Compared to enterprise MMPs that require $10K+ annual agreements before usage costs begin, Core Plan lets fitness app teams start measuring CPS by channel without adding a fixed cost line to their budget.
If you currently export data from Meta Ads Manager, Google Ads, and RevenueCat into separate spreadsheets to calculate CPS manually, Core Plan replaces that workflow with a single [attribution](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it) pipeline that connects the install event to the subscription event automatically.
> If your current marketing dashboard shows CPI but not CPS by channel, see how [Airbridge Core Plan connects installs to subscription revenue](https://www.airbridge.io/en/plans/core).
## FAQ: Marketing Dashboards and Subscription Attribution
### How does CPS differ from CPI for subscription app budget decisions?
CPI (Cost per Install) measures what you pay to acquire a download. CPS (Cost per Subscription) measures what you pay to acquire a paying subscriber. For subscription apps, CPI alone is misleading because a channel with cheap installs may produce very few subscribers. **Budget decisions based on CPI can scale the wrong channels and waste 30 to 50% of ad spend on non-converting traffic.**
### When should a subscription app move from manual CPS tracking to an MMP?
Manual CPS tracking (spreadsheets combining ad platform and RevenueCat exports) works at low volume but breaks above roughly 5,000 monthly installs across 2 or more channels. At that scale, attribution windows, deduplication, and the delay between install and subscription make manual reconciliation unreliable. An MMP automates this connection and provides real-time CPS visibility.
### Can GA4 replace an MMP for measuring cost per subscriber by channel?
GA4 tracks in-app behavior and can report subscription events, but it uses a session-based attribution model that does not deduplicate across [Self-Attributing Networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network). When Meta and Google both claim credit for the same install, GA4 has no mechanism to resolve the conflict. For channel-level CPS, an MMP provides the deterministic attribution that GA4 cannot.
## The Dashboard Metric That Decides Where Your Next Dollar Goes
Every dollar your growth team cannot trace from ad click to paid subscriber is a dollar allocated by assumption, not evidence. For [subscription apps](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025) running paid UA across Meta, Google, and Apple Search Ads, the gap between what your dashboard shows (installs) and what your business needs (subscriber acquisition cost by channel) is not a reporting inconvenience. It is the structural reason budgets flow to the wrong channels.
**[Start Free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core) and see cost per subscriber by channel with 15K free attributed installs.**
---
## MMP Numbers Don't Match Your Ad Platform? Here's What It's Costing Your Subscription Funnel
URL: https://www.airbridge.io/en/blog/mmp-numbers-match-ad-platform-cost-subscription-funnel
Language: English
Published: 2026-03-12
Author: Jaehyuk Kim
Your [MMP (Mobile Measurement Partner)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) says Meta drove 200 paid subscribers last month. RevenueCat says 140. Meta Ads Manager claims 310. Three systems, three numbers, zero confidence in which one should guide your next budget decision.
**This is not a reporting inconvenience. For subscription apps spending $5,000 to $20,000 per month on paid acquisition, misattributed numbers directly distort CAC calculations, corrupt channel-level ROAS, and cause budget to flow toward campaigns that appear profitable but are not.** The cost compounds monthly because the same flawed data drives the same flawed decisions.
**Key Takeaways**
* **Subscription apps face the widest attribution gaps.** Revenue events happen days or weeks after install, often outside the ad platform's default optimization window.
* **Ad platforms routinely over-report conversions compared to actual revenue data.** When each platform claims credit for the same user, the sum of reported conversions exceeds reality, making channel-level budget decisions unreliable without a reconciled source of truth.
* **The financial cost is tangible.** Growth teams spend 4-8 hours per week on manual data stitching, and misallocated budget on channels that look profitable in platform dashboards compounds quarter over quarter.
* **Airbridge Core Plan closes this gap** by connecting install attribution to subscription revenue through native RevenueCat and Adapty integrations, with 25 standard subscription events tracked across Meta, Google, Apple Search Ads, and TikTok.
## Why MMP Numbers Never Match Your Ad Platform Data
The mismatch between your ad platform, your MMP, and your billing system is not a bug. It is the predictable result of three independent systems measuring overlapping but different things with incompatible rules.
For [subscription apps](https://www.airbridge.io/en/blog/subscription-app-growth-guide), this problem is worse than for most app categories. The revenue event that actually matters (a paid subscription) happens days or weeks after the install. By the time a user converts from trial to paid subscriber, the [attribution window](https://www.airbridge.io/en/glossary/attribution-window) may have already closed on one or more platforms.
Understanding what MMP attribution numbers mismatch costs subscription apps starts with knowing why these numbers diverge in the first place.
### 1. Self-Reporting Networks Inflate Conversions by Design
[Self-Attributing Networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network) like Meta, Google, and TikTok each report their own conversion numbers using their own attribution models. Meta may count a view-through conversion within a 1-day window. Google may attribute the same install to a search click within a 30-day window. TikTok may claim that same user through its own click model.
The result: the same user gets counted by multiple platforms. Independent analysis shows that major ad platforms systematically over-report conversions, with Meta over-reporting by a median of 134% and Google by approximately 18% ([Cassandra, 2025](https://cassandra.app/resources/marketing-attribution-software-analysis)). Your MMP applies a single attribution model to deduplicate these overlapping claims, which means the MMP's total will almost always be lower than the combined platform numbers.
Growth teams reviewing both dashboards see the discrepancy and lose trust in both data sources. The real problem is not that the numbers differ. It is that neither set of numbers connects to actual subscription revenue.
### 2. Attribution Windows Create Structural Gaps
Consider this standard timeline for a health and fitness subscription app:
| Event | Timing | Who Records It |
|:------|:-------|:---------------|
| Ad click (Meta) | Day 0 | Meta Ads, MMP |
| App install | Day 0 | MMP |
| Trial start | Day 1 | MMP, Billing platform |
| Trial-to-paid conversion | Day 7-14 | Billing platform |
| First renewal | Day 37-44 | Billing platform only |
Meta's default click attribution window is [7-day click](https://www.facebook.com/business/help/304909959680406). If a user starts a 7-day free trial on Day 1 and converts to paid on Day 8, Meta may not count the subscription as a conversion from that click. Your MMP records the install but may lack the subscription event unless it is integrated with your billing platform. RevenueCat records the subscription but has no channel attribution data.
Each system captures part of the picture. None captures the complete path from ad click to subscription revenue. The gap between "attributed installs" and "actual subscribers" is where budget decisions go wrong.
[Where attribution data breaks for subscription apps]
### 3. The Subscription Timing Problem
For health and fitness apps, this timing gap is especially costly. January resolution spikes drive the highest install and trial volumes of the year. But the churn wave hits in February and March, separating high-intent subscribers from impulse sign-ups.
If attribution data lags by even two weeks, growth teams cannot distinguish which January campaigns drove subscribers who renewed versus those who cancelled after the trial. According to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/) report, 67% of Health and Fitness subscription revenue comes from annual plans, and Health and Fitness apps see a median trial-to-paid conversion rate of 39.9%. Missing the January optimization window means waiting a full year for the next comparable acquisition opportunity.
The cost is not abstract. A team spending $15,000 per month on paid UA that scales the wrong channel for two months because attribution data was stale has wasted $10,000-$15,000 on campaigns that produced trial starts but not paying subscribers.
### 4. The Manual Reconciliation Tax
Without automated attribution-to-billing integration, most growth teams fill the gap with spreadsheet exports. The typical workflow: export MMP install data weekly, export RevenueCat subscription data weekly, manually match user IDs or campaign tags, and flag discrepancies.
Over a year, that is 200-400 hours of growth team time spent on data hygiene instead of campaign optimization. Worse, the manual process introduces its own errors. Campaign naming inconsistencies, timezone mismatches between export files, and lag between systems mean the reconciled numbers are still approximations.
The compounding effect is what makes MMP numbers mismatch so expensive for subscription apps. Misattributed data leads to wrong budget decisions. Wrong budget decisions produce more misleading data in the next cycle. A team that scales the wrong Meta campaign for three months at $10,000 per month has misallocated $30,000 before the reconciliation report even catches the discrepancy.
[The compounding cost of MMP numbers mismatch]
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Stop guessing which numbers are right — unify your attribution data.%%
## How to Reduce Attribution Mismatch and Recover Lost Budget
### 1. Practical Steps to Minimize Data Gaps
Before adopting any new tool, growth teams can take immediate steps to reduce the noise in their attribution data:
* **Align attribution windows across all ad platforms to the same lookback period.** If Meta uses 7-day click and Google uses 30-day click, your cross-channel comparison is inherently skewed. Standardize to the window that matches your typical trial-to-paid conversion timeline.
* **Establish one system as the revenue source of truth.** RevenueCat or Adapty should be the single authority for subscription revenue data. Do not use ad platform revenue reporting for budget allocation decisions.
* **Run weekly reconciliation between MMP install data and billing platform subscription data.** Flag channels where the gap between attributed installs and actual subscriptions exceeds 20%.
* **Tag campaigns with consistent UTM structures** so that manual matching between systems is at least possible when automated attribution breaks down.
* **Prioritize trial-to-paid conversion rate by channel** over install volume as the primary budget signal. A channel with fewer installs but higher subscription conversion is more valuable than one with high install volume and low paid conversion.
These steps reduce noise and improve decision quality. But they do not eliminate the structural gap between install attribution and subscription revenue. Closing that gap requires a system that connects attribution data to billing data natively.
### 2. How Airbridge Core Plan Closes the Subscription Attribution Gap
[Airbridge Core Plan](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026) was built around one question: "Are paid users converting into subscriptions, and which channels are driving value?"
For teams frustrated by MMP numbers that never reconcile with billing data, Core Plan addresses the structural gap directly:
**Native billing platform integration.** Core Plan integrates with [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and Adapty at the base tier. Most enterprise MMPs include these integrations in higher-tier plans. Core Plan includes them from day one at no extra tier, eliminating the manual reconciliation workflow.
**25 standard subscription events.** Core Plan tracks the full subscription lifecycle with predefined events: Install, Start Trial, Subscribe, Unsubscribe, and Order Complete, among others. These standard events use predefined event names and parameters defined by Airbridge, reducing schema design work compared to building custom event taxonomies from scratch.
**Single attribution source across GMAT channels.** Core Plan deduplicates installs and events across Meta, Google, Apple Search Ads, and TikTok, providing one consistent attribution view instead of four conflicting platform reports. These four channels typically represent 80-90% of early-stage paid acquisition spend.
**Six built-in reports for subscription analytics.** Funnel, Retention, Revenue, Actuals, Trend, and Active User reports surface [channel-level subscription performance](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it) without requiring a data warehouse or custom dashboard.
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:--------------|:-------------------|
| Billing integration (RevenueCat, Adapty) | Native (included in higher tiers) | Native (included in base) |
| Subscription events | Predefined + custom events | 25 subscription-optimized standard events |
| Setup guidance | General-purpose | Subscription-app focused |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Unnecessary features | Fraud, raw export, agency access included | Intentionally removed |
**Pricing designed for early-stage budgets.** Core Plan starts with 15K free attributed installs and charges $0.05 per install after that. Compare this to enterprise MMPs that typically require annual contracts with minimum commitments of $10,000 or more. For a team spending $5,000-$20,000 per month on paid UA, the attribution tool should not cost more than a meaningful percentage of the ad budget itself.
**Design trade-offs, stated honestly.** Core Plan does not support custom events. For subscription apps using standard trial and payment flows, the 25 predefined events cover the funnel without schema planning. Core Plan supports a maximum of 2 third-party integrations and GMAT channels only. If your team requires custom event tracking beyond standard subscription events, or integrations with non-SAN ad networks, Airbridge's Growth Plan provides those capabilities.
## FAQ: MMP Numbers Mismatch and Subscription Attribution
### How much budget can attribution mismatches waste?
For a team spending $10,000-$20,000 per month across 3-4 channels, even a 20% misallocation means $2,000-$4,000 per month directed to the wrong campaigns. Over a quarter, that totals $6,000-$12,000 in misallocated spend.
### What causes the biggest discrepancy between MMP and ad platform numbers?
The largest gap comes from Self-Attributing Networks counting the same conversion across platforms. View-through attribution and different lookback windows compound the problem. For subscription apps, the delay between install and paid conversion amplifies the gap because the revenue event often falls outside the ad platform's attribution window. The MMP deduplicates installs but often lacks direct billing data, creating a second gap between attributed installs and actual revenue.
## Every Dollar Misattributed Is a Dollar Misallocated
The gap between what your ad platforms report, what your MMP records, and what your billing system confirms is not just a data problem. It is a structural blind spot that compounds with every month you scale spend without reconciled attribution.
For subscription apps where the revenue event happens days or weeks after the install, this gap is wider than for any other app category. Growth teams that operate without channel-level subscription attribution do not just misallocate budget. They systematically underfund the channels that produce their highest-value subscribers.
**[Start Free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core)** and connect install attribution to subscription revenue, starting with 15K free attributed installs.
---
## Which Ad Campaign Drives Subscriptions? Your MMP Won't Admit
URL: https://www.airbridge.io/en/blog/which-ad-campaign-drives-subscriptions-mmp-cant-answer
Language: English
Published: 2026-03-12
Author: Jay Kim
A Meta campaign reports 4,200 installs. Google Ads claims 2,800. Apple Search Ads logs another 900. RevenueCat shows 310 paid subscribers for the month. But when you try to match those subscribers back to the campaigns that acquired them, the data trail goes cold.
This is not a reporting glitch. **It is a structural gap between install attribution and subscription revenue that costs subscription apps thousands of dollars in misallocated ad spend every month.** When budget decisions rely on install counts instead of channel-level subscription data, growth teams systematically overfund low-converting channels and starve the ones that actually produce paying subscribers.
**Key Takeaways**
- **Subscription app attribution breaks at the install-to-revenue boundary.** Most attribution setups track installs by channel but lose visibility once users enter the subscription funnel.
- **Ad platforms optimize for installs and trials, not paid subscriptions.** Self-Attributing Networks each claim credit for conversions, inflating reported numbers without deduplication.
- **SDK-only setups miss subscription events that happen outside the app.** Renewals, billing failures, and refunds occur server-side, creating blind spots in channel-level LTV data.
- **Misattributed subscription revenue compounds over time.** Every month without accurate channel data adds another layer of budget misallocation that becomes harder to unwind.
- **Airbridge Core Plan connects install attribution to subscription revenue.** With native RevenueCat and Adapty integration, it maps the full funnel from ad click to paid subscriber across Meta, Google, Apple Search Ads, and TikTok.
## Why Subscription App Attribution Breaks Between Install and Revenue
Subscription app attribution is fundamentally different from single-purchase attribution. A user who installs your app today might start a free trial on day 3, convert to a paid subscription on day 10, and renew (or churn) on day 40. The revenue event that matters most happens weeks after the install, long after most [attribution windows](https://www.airbridge.io/en/glossary/attribution-window) have closed.
This creates a structural problem: the tools that track installs are disconnected from the tools that track subscription revenue.
### 1\. Ad Platforms Optimize for Installs, Not Subscriptions
[Self-Attributing Networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network) like Meta, Google, TikTok, and Apple Search Ads each run their own attribution. Each platform claims credit for conversions based on its own rules and lookback windows.
The result is predictable: Meta reports 200 subscriptions, Google claims 150, but your billing platform logs only 180 total. Without a neutral [Mobile Measurement Partner (MMP)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) to deduplicate these claims, you cannot determine which channel actually drove the paying subscriber.
This is especially damaging for subscription apps where the same user often interacts with ads across multiple platforms before installing. A user might see a TikTok ad, later click a Meta retargeting ad, then search for the app on Apple Search Ads. All three platforms claim credit for the eventual subscription. Only a neutral MMP referee can attribute that subscriber to a single originating source.
[SAN Double-Counting vs. MMP Deduplication]
For subscription apps using standard 7-day free trials, the problem deepens. The revenue event (Subscribe) often occurs at the boundary of or beyond most ad platforms' default view-through optimization windows. Ad algorithms optimize for the signal they can see (installs, trial starts) rather than the outcome that matters (paid subscriptions).
### 2\. The SDK-Only Gap: Revenue Events That Never Reach Your MMP
Many subscription apps rely on SDK-based event capture for attribution. The problem: SDK detects subscription state changes only when the user opens the app, creating latency and dependency on user behavior.
Subscription lifecycle events do not follow this pattern:
- **Renewals** happen server-side on a fixed billing cycle, regardless of whether the app is open
- **Billing failures** occur when a payment method expires or gets declined, often without the user knowing
- **Refunds** are processed through the app store, not through the app itself
For events like renewals or billing failures that occur while the app is closed, SDK capture is unpredictable or absent if the user never reopens. This means your attribution data reflects the initial conversion but not the ongoing revenue reality of each channel.
According to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025) report, the median trial-to-paid conversion rate for Health and Fitness apps is 39.9%. That means roughly 60% of trial users churn before paying. Without channel-level data on which trials actually convert, you cannot distinguish a channel that generates curious trialers from one that generates committed subscribers.
### 3\. The Cost of Flying Blind on Channel-Level Subscription Data
Consider a fitness subscription app spending $15,000/month across three channels:
| Channel | Installs | Trial Starts | Paid Subscribers | Cost per Subscriber |
| --- | --- | --- | --- | --- |
| Meta Ads | 4,200 | 840 | 210 | $24 |
| Google Ads | 2,800 | 420 | 168 | $30 |
| Apple Search Ads | 900 | 180 | 18 | $278 |
Without subscription app attribution, this team would see three channels all producing installs and trials. The budget might get split proportionally by install volume. With channel-level subscription data, the picture changes: Meta and Google produce subscribers at $24 and $30, while Apple Search Ads costs $278 per subscriber in this scenario.
**Without channel-level CPS data, the team cannot identify which source to scale or cut, leaving** [**LTV**](https://www.airbridge.io/en/glossary/lifetime-value-ltv) **on the table.**
For a team spending $15,000/month, even a 20% misallocation toward the wrong channel adds up to $36,000 per year in wasted spend. For subscription apps where [D60 revenue per install varies dramatically by channel](https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it), the compounding effect of bad attribution data is one of the largest hidden costs in early-stage paid UA.
[diagram-attribution-gap.webp]
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Connect installs to subscription revenue and find your best-performing campaigns.%%
## How to Fix Subscription App Attribution
### 1\. Practical Steps Any Team Can Take Today
If you are trying to figure out how to track subscription attribution by ad channel for mobile apps, these process-level fixes can reduce the gap immediately:
- **Align attribution windows across all ad platforms.** Set the same lookback window (click and view-through) on Meta, Google, Apple Search Ads, and TikTok. Mismatched windows are the simplest source of discrepancies to fix.
- **Connect your billing platform to your analytics stack.** If you use RevenueCat or Adapty, ensure subscription events (Start Trial, Subscribe, Renewal, Cancellation) flow into the same system that receives install data.
- **Define "subscriber" consistently.** A trial start is not a subscription. Standardize your definition of a paid subscriber across all reporting tools before comparing channels.
- **Tag campaigns with UTM parameters for web-to-app funnels.** If you run web acquisition alongside app campaigns, UTM tracking provides deterministic attribution that bypasses many iOS privacy restrictions.
- **Review SAN-reported conversions against your billing source of truth.** Export subscription data from RevenueCat or the App Store and compare it to what each ad platform claims. The delta reveals your attribution gap.
### 2\. How Airbridge Core Plan Connects Installs to Subscription Revenue
The process fixes above close some gaps, but connecting installs to subscription revenue at the channel level requires an [MMP built for subscription apps](https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026).
Airbridge Core Plan was designed around one question: **"Are paid users converting into subscriptions, and which channels are driving value?"**
Here is how it solves the specific attribution problems described above:
- **Deduplicated attribution across GMAT channels.** Core Plan integrates with Meta, Google, Apple Search Ads, and TikTok (the four SANs that represent 80-90% of early-stage paid acquisition spend) and provides a single, deduplicated source of truth for each install.
- **Native** [**RevenueCat**](https://www.airbridge.io/en/blog/abr-revenuecat-integration) **and Adapty integration.** Subscription events flow directly into Core Plan via S2S connection. This enables real-time, independent signal transmission that does not rely on the user manually opening the app. Renewals, billing failures, and cancellations are captured regardless of app state.
- **25 subscription-optimized standard events.** Core Plan's standard events include Install, Start Trial, Subscribe, Unsubscribe, and Order Complete. These cover the full subscription funnel without requiring custom event schema design.
- **6 built-in reports for subscription metrics.** Actuals, Trend, Active User, Funnel, Retention, and Revenue reports surface channel-level subscription data without additional setup.
| Capability | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Billing platform integration | Native (included in higher tiers) | Native (included in base) |
| Subscription event tracking | Predefined + custom events | 25 subscription-optimized standard events |
| Setup decisions | General-purpose onboarding | Subscription-app-focused onboarding |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Unnecessary features | Fraud, raw export, agency access included | Intentionally removed |
| Ad channel coverage | All networks supported | GMAT channels |
Core Plan does not support custom events or raw data export. These are intentional design decisions: standard events cover the subscription funnel, and pay-as-you-go pricing keeps costs aligned with early-stage budgets. For teams that need custom events, additional ad networks, or raw data warehouse exports, [Airbridge's Growth Plan](https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026) provides the full measurement stack.
If your current MMP shows installs but leaves you guessing about which channels drive subscribers, the problem is not your team. **It is the gap between install attribution and subscription revenue that most tools were not designed to bridge.** Core Plan was built specifically for this gap.
The pricing model reflects this focus: 15K free attributed installs with $0.05/install after the free tier. For a subscription app generating 5,000 installs/month, the first three months cost nothing. Compare this to enterprise MMPs where minimum annual contracts start at $10,000 or more, regardless of how many features the team actually uses.
## FAQ: Subscription App Attribution
### What events should a subscription app track for proper attribution?
At minimum, track the full subscription funnel: Install, Sign-up, Start Trial, Subscribe, Unsubscribe, and Order Complete. These events let you measure [cost per subscription by channel](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution), trial-to-paid conversion rates, and channel-level LTV. Billing events (renewal, billing failure) are also critical for understanding true subscriber retention by acquisition source.
### How does subscription app attribution differ from standard app attribution?
Standard [app attribution](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) focuses on the install event. Subscription app attribution must extend beyond that to track the full lifecycle: trial start, paid conversion, renewal, and churn. The key difference is timing. Revenue events often happen days or weeks after the install, requiring server-to-server connections with billing platforms like RevenueCat or Adapty to maintain the link between acquisition source and subscription outcome.
### When should a subscription app upgrade from Core Plan to a full MMP?
The upgrade signal is usually complexity, not scale. When your team needs custom event tracking beyond standard subscription events, support for ad networks beyond Meta, Google, Apple Search Ads, and TikTok, or raw data exports to a data warehouse, [Growth Plan](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025) provides those capabilities.
## The Revenue You Cannot Attribute Is Revenue You Cannot Scale
Every dollar of ad spend without channel-level subscription attribution is a dollar allocated on instinct, not evidence. Growth teams that cannot see which channels produce paying subscribers do not just waste budget. They systematically underfund their best-performing channels while overinvesting in channels that generate installs but not revenue.
[**Start free with Airbridge Core Plan**](https://www.airbridge.io/en/plans/core) **and see which ads actually drive your subscriptions, starting with 15K free attributed installs.**
---
## Why Deeplinks Matter for User Experience and Marketing in 2026
URL: https://www.airbridge.io/en/blog/why-deeplinks-matter-for-user-experience-and-marketing-in-2026
Language: English
Published: 2026-03-10
Author: Luke
In 2026, every extra click is a direct tax on your revenue. If your ad promises a '20% discount' but dumps the user on a generic homepage, you've already lost the subscription.
That’s why a strategic **deeplink** is no longer just a technical feature. For subscription apps, it directly affects conversion rate, retention, and revenue.
Yet many marketers still ask: _Why do deeplinks matter?_ **Or worse, they assume deeplinks are just “smarter URLs.”**
In reality, a Deeplink determines whether a user lands exactly where they expect – or gets dumped on a generic homepage and abandons. That single moment shapes onboarding success, paid media ROI, and subscription growth.
In 2026, when acquisition costs continue to rise, precision in user journeys isn’t optional. It’s revenue protection.
**📌 Key takeaways**
If you only remember three things from this guide, let it be these strategic pillars for 2026:
- **Deep Links are Revenue Tools, Not Just Navigation:** In a subscription model, every extra click is a "Friction Tax" that directly reduces your Conversion Rate. Deeplinks eliminate the login wall and the manual search, leading users straight to the paywall or premium content they were promised.
- **Deferred Deep Linking is Non-Negotiable:** For new user acquisition, the journey shouldn't break at the App Store. Deferred linking ensures that the "context" of the ad follows the user through the install, creating a personalized first-open experience that boosts Day 1 retention.
- **Reliable Infrastructure Prevents Attribution "Black Holes":** Modern OS updates and in-app browsers (like Instagram's) frequently break standard links. Using a dedicated solution like the Airbridge DeepLink Plan ensures your links work 100% of the time while providing the granular data needed to prove your marketing ROI.
## **What is a Deeplink and what is the role?**
A deeplink is a URL that routes users directly to a specific page or feature inside a mobile app, bypassing the home screen.
It routes users directly to relevant content inside a mobile app.
**Unlike regular web links, deeplinks preserve context.** They remember what the user clicked and deliver them to the exact screen tied to that intent.
In practice, this means:
- Clicking a promo email opens the exact discounted plan page
- Tapping a retargeting ad opens the abandoned subscription screen
- Selecting a push notification opens a specific content episode
Without a deeplink, users land at the app’s default screen. They must search again. Many don’t.
A common mistake marketers make is assuming that installing the app equals conversion. It doesn’t. The _first screen after install_ often determines whether a user subscribes – or churns.
### **Web Links vs. Deeplinks: The Critical Difference**
The difference between a regular link and a deeplink is destination precision.
Web links send users to websites or app stores. Deeplinks hit app content instantly.
**But the impact goes deeper than navigation.**
| | Web Links | Deeplinks |
| --- | --- | --- |
| Destination | Browser or app store | Specific in-app screen/feature |
| User Friction | High (downloads, searches) | Low (direct access) |
| Conversion Impact | Baseline | Up to 2.5x higher retention (Source: Adapty) |
| Best For | Awareness | Onboarding, re-engagement |
| Fallback | None | Web or store (universal links) |
From a tactical standpoint, the biggest difference is **intent preservation**.
Imagine this scenario:
- A user clicks an ad offering 20% off an annual subscription
- They install the app
- Instead of seeing the offer, they land on the homepage
That friction costs revenue.
## **Why Subscription Apps Fail Without Strategic Linking**
The revenue impact goes far beyond user convenience. Research reveals [more than 90% of mobile time is spent within mobile apps rather than mobile web](https://www.luciq.ai/blog/stop-treating-mobile-apps-like-web-apps) and app users consistently demonstrate higher lifetime value across industries.
For subscription-based apps, the "Leaky Bucket" syndrome is the primary threat to growth.
In 2026, the cost of user acquisition (CAC) has reached a point where losing a single potential subscriber due to a broken link is a significant financial hit. If your link sends a user to a generic mobile web homepage instead of an in-app checkout, you aren't just losing a click – you are losing recurring revenue (LTV).
A common mistake marketers make is assuming the "Install" is the final goal. **For a subscription app, the install is just the beginning.**
If a user clicks an ad for "Premium Meditation Tracks" and lands on a generic app store page, the context is lost. They must download, open, and then _manually search_ for that specific track. Most users will simply close the app because of ‘Friction Tax’ and ‘Login Wall’.
- **The Friction Tax:** For subscription apps, this "Friction Tax" is the silent killer of ROI. Every broken link is a friction tax that compounds. Statistics show that [21–72% of users abandon during onboarding when it’s frictiony.](https://www.setgreet.com/blog/what-the-numbers-actually-say-about-mobile-app-onboarding-\(and-what-to-track\)) If your ad costs $5.00 per click and 20% of users bounce due to a broken deep link, you are effectively paying a 25% premium for every successful install. Strategic deep linking isn't a luxury; it's ROI insurance.
- **The High Cost of the "Login Wall":** Subscription apps often require a login. And the 'Web Login Wall' is the #1 killer of mobile web conversions. When a user is directed to the mobile web instead of the app, they are frequently met with a "Please Log In" screen. Since most users do not remember their passwords on a mobile browser, the journey ends there. Strategic linking bypasses the browser entirely, utilizing the app's existing authentication to keep the user "logged in" and ready to convert.
[]
## **How Subscription Apps Use Deeplinks to Maximize Growth**
Successful subscription apps treat every link as a personalized entry point. They move beyond basic navigation to create "contextual" journeys that feel like a concierge service for the user.
### **1) Onboarding Optimization**
**The first 60 seconds of an app experience determine long-term retention.**
Using [**Deferred Deep Linking**](https://www.airbridge.io/glossary/deferred-deep-linking), you can carry data through the install process. If a user clicks a "Start 7-Day Free Trial" link, the app can automatically open to the trial signup screen immediately after the first launch – even if the user didn't have the app installed previously.
Read more about [Subscription App Onboarding: Get 100% of Users to Your Paywall here.](https://www.airbridge.io/en/blog/subscription-app-onboarding)
### **2) Re-engagement Campaigns**
Win-back campaigns are the lifeblood of subscription models.
When emailing a lapsed subscriber with a "50% Off Renew" offer, the link must point directly to the payment confirmation screen within the app.
Forcing a lapsed user to navigate through the app settings to find the "Renew" button is a guaranteed way to fail.
### **3) Cross-platform User Journeys**
In 2026, the user journey is rarely linear.
A user might discover your content on a desktop, see an ad on social media, and eventually subscribe on their phone.
Deeplinks act as the glue for these cross-platform transitions, ensuring that a "Swipe Up" on Instagram leads directly to the featured content inside the app environment, not a clunky mobile web view.
## **The Technical Challenge: Why Most Deeplink Strategies Fail**
### **1) Common Implementation Pitfalls**
**Implementation pitfalls occur when deeplink logic doesn’t consistently deliver users to the intended in-app destination.**
Here are the most frequent issues subscription teams face for deeplinks:
#### **(1) Broken Fallback Logic**
If the app is not installed, what happens?
- Does the link redirect to the App Store?
- Does it preserve campaign data?
- Does the post-install experience route correctly?
Without proper deferred deep linking, users install the app — and land on the homepage. Intent disappears.
#### **(2) Incorrect Routing Logic**
Even when the app opens, routing can fail.
Examples:
- The link opens the app but not the correct screen.
- Promo parameters are dropped.
- Subscription tiers don’t pre-select correctly.
These errors often go unnoticed until revenue metrics decline.
#### **(3) Inconsistent QA Across OS Versions**
iOS and Android handle links differently.
- iOS relies on Universal Links.
- Android uses App Links and intent filters.
If configurations break after OS updates, deep links may revert to browser opens instead of in-app routing.
This is especially risky for subscription apps running continuous paid campaigns.
#### **(4) Fragmented Team Ownership**
Deeplinks sit between marketing and engineering.
- Marketing owns campaigns.
- Engineering owns routing logic.
- Data teams own attribution.
When ownership is unclear, issues linger.
And while teams debate technical details, conversion rates drop quietly in the background.
### **2) Cross-Device Attribution Problems**
**Cross-device attribution problems occur when user journey data is lost between click, install, and in-app action.**
For subscription apps, this directly impacts budget allocation.
If attribution breaks:
- Paid campaigns look unprofitable.
- High-performing creatives get paused.
- CAC appears inflated.
- LTV calculations become unreliable.
Consider this scenario:
1. User clicks an ad on mobile web.
2. Installs the app.
3. Completes subscription inside the app.
4. Attribution fails to connect the purchase to the original campaign.
The marketing team sees installs — but not revenue.
That disconnect leads to underinvestment in profitable channels.
## **Airbridge Core Plan: The Reliable Solution for Seamless Conversions**
The technical complexities of deeplink implementation explain why many subscription apps struggle with broken user journeys despite understanding their importance.
**Airbridge Core Plan** addresses these challenges by providing enterprise-grade linking infrastructure **(but with start-up price)** that works consistently across devices, platforms, and attribution scenarios.
It’s designed not just to create links, but to own the entire deeplinking workflow. Below is why **Airbridge is the Choice for Subscription Apps:**
- **All-in-One Link Authority:** One single URL that handles iOS, Android, and Web routing automatically.
- **Deferred Deep Linking Excellence:** Maintain the user’s context through the App Store. If they click a "Premium Workout" ad, they see that workout the moment they finish installing.
- **Granular Attribution:** Attribute every subscription back to its original touchpoint, even across devices, ensuring your marketing budget is spent on what actually scales.
- **No-Code Flexibility:** Allow marketing teams to create and test deep links without waiting for a developer sprint.
👉 [Meet all features of Airbridge Core Plan first!](https://coreplan.airbridge.io/waitinglist)
## **Streamline Your Subscription Journey Today**
The subscription economy in 2026 leaves no room for "broken" experiences. Every link that leads to a generic home screen or a mobile web login is a lost revenue opportunity. By implementing a sophisticated **deeplink** strategy, you aren't just improving navigation — you are protecting your marketing spend and respecting your user’s time.
The **Airbridge Core Plan** provides the reliable infrastructure needed to bridge the gap between discovery and conversion.
Don't let technical friction stand in the way of your app's growth.
Explore real successful stories here:
[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://www.airbridge.io/en/customers/fizz-migrates-from-firebase-dynamic-links-to-airbridge-deep-link)
[How Dabang cut deep link management time by 90% with Airbridge](https://www.airbridge.io/en/customers/airbridge-dabang-deeplink-optimization)
---
## How MMP Tools Prove Mobile Marketing ROI (And Optimize Ad Spend) in 2026
URL: https://www.airbridge.io/en/blog/how-mmp-tools-prove-mobile-marketing-roi-and-optimize-ad-spend-in-2026
Language: English
Published: 2026-03-10
Mobile marketers don’t struggle with data. They struggle with trust.
Every network claims performance. Analytics tools show engagement. But when it’s time to prove ROI—real revenue tied to real ad spend—the numbers rarely align.
That’s where the right **MMP** makes the difference. Not just as an attribution tool, but as the system that connects campaigns to profitability. Let’s break down how MMP tools for mobile marketing ROI actually prove what’s working—and what’s not.
**📌 Key takeaways:**
- **Installs ≠ ROI:** Volume means nothing if LTV doesn’t exceed CAC. Profitability—not installs—is the real KPI.
- **Ad Networks Aren’t Neutral:** Self-attributing platforms measure inside their own walls. Without independent attribution, ROAS gets inflated.
- **Subscription Revenue Is Delayed:** Revenue often happens weeks after install. Short measurement windows lead to wrong optimization decisions.
- **Privacy Changed the Rules:** With SKAN and signal loss, proving ROI now requires structured, aggregated measurement—not user-level tracking.
- **The Right MMP Connects Spend to Profit:** When cost, cohort revenue, renewals, and retention live in one framework, budget decisions become strategic.
## **Why Mobile Marketers Struggle to Prove ROI**
**Marketers use** [**Mobile Measurement Partner (MMP)**](https://abr.ge/i7ff5r) because it provides a neutral, unified source of truth for installs, events, revenue, and ROI.
MMP collects attribution signals, connects them to in-app events, and standardizes performance data into one dashboard. Without an MMP, each ad network reports its own version of performance. With an MMP, you get an independent measurement layer.
But many marketers have been struggling with fragmented data while proving ROI.
**Fragmented data in mobile marketing** refers to performance metrics being split across multiple ad networks, analytics tools, and internal dashboards—making it difficult to calculate true ROI with confidence.
For performance marketers and C-level leaders, this fragmentation creates one critical problem: You can’t prove which campaigns are actually driving revenue.
### **1\. Why Every Ad Network Claims Credit**
[Meta](https://www.facebook.com/business/help/460276478298895?id=561906377587030&ref=search_new_3) reports Meta conversions. [Google](https://support.google.com/google-ads/answer/1722022) reports Google conversions. TikTok does the same.
[Each platform uses its own attribution logic](https://www.cometly.com/post/multiple-ad-platforms-attribution-confusion), so overlap is inevitable. When you total conversions across networks, the number often exceeds actual installs. That gap weakens trust and leads teams to optimize based on inflated, network-reported ROAS instead of unified data.
### **2\. Why Subscription Apps Can’t Rely on Installs**
**For subscription and AI services, installs are just the starting point.** Profitability depends on:
- Free-to-paid conversion
- Renewal rate
- Retention curves
- LTV/CAC ratio
### **3\. How Data Silos Distort ROI Reporting**
Marketing tracks campaigns. Product tracks in-app behavior. Finance tracks revenue.
When those systems aren’t connected, ROI reporting becomes manual—CSV exports, spreadsheet reconciliation, and conflicting numbers in leadership meetings. That friction delays budget decisions and limits scalability.
### **4\. How SKAN and Privacy Complicate Attribution**
SKAdNetwork introduced delayed and aggregated reporting on iOS. Without structured interpretation, teams face:
- Delayed revenue visibility
- Reduced campaign-level detail
- Lower confidence in scaling spend
## **Why Basic Analytics Tools Can’t Prove Mobile Marketing ROI**
You may have been using basic analytics tools like GA4 and assumed they were enough to measure marketing performance. After all, they provide dashboards showing installs, sessions, engagement, and even revenue events. For many teams, this visibility feels sufficient at first.
**However those basic analytics tools track user behavior inside your app—but they don’t independently verify which marketing source truly drove the revenue.** That limitation makes them unreliable for proving real mobile marketing ROI.
Here's why MMPs win:
**Basic Analytics vs Ad Networks vs MMP (Comparison Table)**
| Capability | Basic Analytics Tool | Ad Network Dashboard | MMP |
|---|---|---|---|
| Cross-Network Attribution | ❌ Limited | ❌ Biased (Self-reported) | ✅ Unified & Neutral |
| Revenue Event Tracking | ✅ Yes | ⚠️ Limited | ✅ Advanced |
| Cost Data Integration | ❌ Manual | ✅ Network-only | ✅ Cross-channel |
| LTV/CAC Calculation | ❌ Not native | ⚠️ Partial | ✅ Built-in |
| Fraud Detection | ❌ No | ⚠️ Limited | ✅ Advanced |
| SKAN Interpretation | ⚠️ Basic | ⚠️ Network-specific | ✅ Aggregated & Standardized |
### **1\. Analytics Tools Don’t Control Attribution**
Tools like Firebase, GA4, or Amplitude are strong at event tracking. They show:
- Session data
- Funnel drop-offs
- Feature usage
- Retention curves
But attribution often relies on network-provided data or limited referrer logic. That means you’re not applying neutral, cross-network attribution rules. You’re inheriting fragmented reporting.
In practice, this leads to conflicting install counts between analytics tools and ad networks. Finance asks which number is correct. Marketing doesn’t have a definitive answer.
### **2\. Self-Attributing Networks Inflate ROAS**
[Self-attributing networks (SANs)](https://abr.ge/nlx3xg3) report conversions based on their internal methodology.
From a performance standpoint, that creates Overlapping attribution windows, Double counting, or even Inflated ROAS. A common mistake marketers make is optimizing budgets based solely on network dashboards.
Independent attribution is critical. Without it, ROI becomes an estimate—not a verified metric.
### **3\. No Reliable LTV-to-CAC Measurement**
Basic analytics tools can track revenue events. Ad platforms can track ad spend.
But most teams struggle to connect:
Without unified cost ingestion and attribution logic, you can’t reliably calculate:
- Payback period
- LTV by campaign
- Revenue by creative
- ROI by geo
### **4\. No Incrementality or Fraud Protection**
Basic analytics platforms don’t evaluate whether a conversion would have happened without paid media.
They also lack robust fraud detection mechanisms.
Without fraud filtering or incrementality analysis, marketers risk paying for users they would have acquired organically.
## **How MMP Tools Actually Prove Mobile Marketing ROI**
**MMP tools for mobile marketing ROI prove value by connecting ad spend, attribution, in-app behavior, and revenue into one unified measurement framework.** They turn installs into financial outcomes you can validate and scale against.
This is where attribution becomes infrastructure—not reporting.
### **1\. Unified Cross-Channel Attribution**
An MMP applies standardized attribution logic across Paid social, Search, DSPs, Affiliate, and Organic sources.
Instead of trusting each network’s self-reported conversions, you get deduplicated attribution across all channels. In practice, this eliminates inflated conversion counts and reveals which campaign actually drove the user.
### **2\. Revenue-Level Tracking Beyond Installs**
Install optimization is easy, yet revenue optimization is harder. An advanced MMP connects these to the original acquisition source:
- Trial start
- Subscription purchase
- Renewal
- Upgrade
- In-app purchase
For subscription apps and AI service providers, this is critical. Revenue often materializes 7, 30, or 90 days after install. Without cohort-level tracking, you optimize blindly.
### **3\. LTV vs CAC Analysis by Campaign**
This is the core ROI equation: LTV (Lifetime Value) ÷ CAC (Customer Acquisition Cost)
As a result, An MMP ingests Cost data from ad networks, Revenue events from your app, and Attribution data from campaign touchpoints, then calculates ROI at campaign, ad group, creative, and geography levels
From a tactical standpoint, this lets you scale profitable campaigns confidently, pause campaigns before they burn budget, or even forecast payback periods.
### **4\. Distinguishing Real Growth from Organic Noise**
In 2026, simply recording a 'last click' isn't enough to prove ROI. You need to know if an install would have happened even without the ad.
Airbridge employs AI-driven incrementality modeling to filter out organic 'baseline' conversions. By focusing on incremental lift, you can stop spending on users who were already going to convert and reallocate that budget to the channels that drive net-new growth.
### **5\. Cohort-Based Revenue Measurement**
Instead of looking at daily installs, you analyze:
- D7 revenue by campaign
- D30 retention by channel
- Renewal rate by creative
This exposes patterns basic dashboards hide. A common mistake marketers make is judging performance too early. Cohort analysis prevents premature scaling or shutdown decisions.
### **6\. Fraud Filtering**
Not all attributed conversions are incremental. That is why MMPs apply Fraud detection filters, Click validation logic, or Suspicious pattern detection. This protects your budget from:
- Click flooding
- Install hijacking
- Fake traffic
## **What to Look for in an MMP If ROI Is Your Priority**
**If ROI is your priority, the right MMP must connect cost, attribution, and lifetime revenue in a way that is transparent, deduplicated, and decision-ready.** Anything less is just another reporting tool.
### **1\. True Cross-Channel Visibility**
Your MMP should measure across: Paid social, Search, Programmatic, Affiliates, Organic, and Web-to-app flows
In practice, incomplete coverage leads to misallocated budgets. You might scale a channel that only _appears_ profitable because other channels aren’t measured properly.
### **2\. Revenue-Centric Measurement (Not Install-Centric)**
Install metrics are top-of-funnel. ROI lives downstream.
Look for an MMP that supports:
- Subscription lifecycle tracking
- Renewal attribution
- Upgrade revenue mapping
- Custom revenue events
- Cohort LTV analysis
From a tactical standpoint, this is critical for subscription apps where profitability depends on retention curves—not CPI.
### **3\. Automated Cost Ingestion & LTV/CAC Reporting**
Manual cost uploads break ROI accuracy.
Your MMP should:
- Automatically ingest cost data from major networks
- Normalize cost across channels
- Calculate LTV/CAC inside the platform
Without automated cost integration, teams revert to spreadsheets. That increases reporting lag and error risk.
### **4\. Advanced SKAN & Privacy Measurement**
iOS measurement is no longer optional complexity.
A strong MMP must:
- Decode SKAN postbacks
- Map conversion values to revenue signals
- Aggregate delayed data accurately
- Support privacy-safe modeling
If your MMP treats SKAN as an afterthought, your iOS ROI reporting will remain partial.
### **5\. Raw Data Access & Transparency**
For experienced performance marketers and developers, black-box reporting is a red flag.
Your MMP should have:
- Raw log-level data access
- Flexible data export options
- API integration capability
- Warehouse-ready pipelines.
## **Why Airbridge Is Built for ROI-Focused Teams**
**Airbridge is an MMP focused on connecting ad spend to real revenue outcomes across web and app environments.** Rather than centering reporting around installs, it emphasizes lifecycle revenue, cross-channel attribution, and data transparency.
- **Unified App and Web Attribution:** Airbridge connects web and app touchpoints into a single user journey view. This reduces misattribution in subscription flows that start on web and convert in-app.
- **Revenue-Level Measurement:** Airbridge ties acquisition sources to downstream revenue events such as: Trial start, Subscription purchase, Renewal, and Upgrade. This enables D7/D30/D90 revenue analysis and LTV/CAC calculation by campaign.
- **Data Transparency and Flexibility:** Airbridge provides raw, log-level data access and warehouse integrations. Teams can validate: Attribution logic, Revenue calculations, and Cohort performance.
- **Privacy-Ready Measurement:** Airbridge supports SKAN postback decoding, conversion value mapping, and aggregated campaign reporting—maintaining ROI visibility under privacy constraints.
- **Built for ROI Accountability:** Airbridge aligns with teams that need cross-network LTV/CAC clarity, reliable cost ingestion, and executive-ready reporting. The focus shifts from install volume to profitable user acquisition.
True ROI is more than just a metric on a dashboard—it is the foundation of your scaling strategy. Don't let your growth plateau due to invisible data leaks or unverified attribution.
**And don't just take our words for it. Let’s see brands that switched to Airbridge from legacy MMPs—and saw measurable gains in performance and ROI:**
👉[How Loyal Scaled Day 7 Retention Performance Across 650+ Apps Using Airbridge](https://abr.ge/d8ggap)
👉[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://abr.ge/mqx3x8q)
👉[Nightly Cuts CPA 18% with Simulated iOS Attribution, Ranks Top 3 in Japan’s App Store](https://abr.ge/4091q4)
---
## Is Your MMP Capping Your Growth? 5 Features to Audit Before Your 2026 Renewal
URL: https://www.airbridge.io/en/blog/is-your-mmp-capping-your-growth-5-features-to-audit-before-your-2026-renewal
Language: English
Published: 2026-03-10
Author: Luke
Your MMP renewal isn't just a procurement task—it's a strategic pivot point. As privacy frameworks like SKAN 5.0 and Privacy Sandbox redefine the rules, the tool you chose two years ago might now be the very thing blinding your attribution.
This guide breaks down the most important **MMP tools features**, common selection challenges, and how to evaluate platforms so you can pick one that supports scalable, data-driven decisions.**📌 Key takeaways:**
- **MMP drives real growth decisions** by linking ad spend to LTV, retention, and revenue — not just installs.
- **Privacy-first attribution is mandatory** as SKAN and consent-based tracking redefine measurement.
- **Real-time, transparent data speeds optimization** and builds trust across teams.
- **Deep linking and fraud prevention protect ROI** by improving conversions and eliminating wasted spend.
- **Airbridge supports scalable growth** with unified attribution, privacy readiness, and flexible integrations, thus attracting many clients switching from legacy MMPs.
## 3 Signs Your MMP Isn’t Ready for the Privacy-First Era
The privacy landscape for mobile marketing has shifted quickly with ATT, SKAdNetwork, and Privacy Sandbox redefining attribution. If your MMP still relies on device identifiers and legacy tracking models, it may already be limiting your ability to make reliable growth decisions.
Here are three signals that your MMP may not be aligned with the modern privacy environment.
### 1\. Your Attribution Still Relies Heavily on Device Identifiers
If your measurement strategy still depends primarily on IDFA or other user-level identifiers, that is a clear warning sign. Since the rollout of ATT, most iOS users have opted out of tracking, dramatically reducing the availability of device identifiers.
As a result, any MMP that relies on deterministic matching through IDs will inevitably experience major attribution blind spots.
A privacy-ready MMP should instead support aggregated measurement frameworks such as SKAdNetwork and consent-aware attribution models. These systems allow marketers to evaluate campaign performance without relying on individual user tracking.
### 2\. Your SKAdNetwork Setup Feels Manual and Opaque
SKAdNetwork was introduced to replace traditional user-level attribution on iOS, yet its implementation is far from simple. It requires careful configuration of conversion values, timer management, postback decoding, and data modeling.
Many MMPs technically support SKAN but offer little automation or actionable guidance, leaving marketers and developers to navigate a complex setup on their own.
If your team spends excessive time configuring conversion values or interpreting SKAN data with limited insight into performance, your MMP may not be providing sufficient support. A modern platform should streamline SKAN implementation, automate configuration where possible, and translate aggregated data into meaningful performance signals that marketers can actually use.
### 3\. Your Data Visibility Shrinks as Privacy Rules Tighten
Another sign of a misaligned MMP is when visibility drops sharply as privacy restrictions increase.
In a privacy-first ecosystem, some data loss is inevitable, yet modern measurement platforms compensate with modeling, aggregated attribution, and advanced reporting techniques.
If your reporting dashboard becomes less informative over time or provides fewer actionable insights after privacy updates, it suggests the platform lacks the infrastructure to adapt. A privacy-resilient MMP should still allow marketers to analyze campaign performance, identify trends, and make optimization decisions even when user-level tracking is limited.
## Top 6 Pain Points in MMP Selection
Understanding MMPs is one thing, choosing a suitable one is another. A good decision affects engineering workload, data governance, executive reporting, and long-term scalability.
Below are the Top 6 issues experienced by subscription apps, AI platforms, and performance teams while considering a multitude of MMPs in the market.
### 1\. Data Discrepancies Across Channels
Data rarely matches between ad networks and the MMP dashboard.
Google claims one number. Meta shows another. Your BI tool shows something else.
The blind spot for most teams is trusting network-reported data when optimizing spend. Networks naturally favor their own attribution logic. **Without an independent MMP, performance becomes subjective, leading to misallocated spend, inflated ROAS, slower optimization cycles.**
### 2\. SKAN Complexity and Limited Visibility
SKAdNetwork is not plug-and-play. It requires:
- Conversion value mapping
- Timer management
- Postback decoding
- Modeling expertise
In practice, many teams underutilize SKAN because configuration feels overwhelming. **Some MMPs provide only surface-level dashboards without actionable modeling insights.**
### 3\. Heavy Developer Burden
Engineering teams often carry the implementation cost. Common friction points are:
- Large SDK size
- Slow release cycles due to SDK updates
- Limited server-to-server flexibility
- Poor documentation
Developers prefer lightweight SDKs, modular integration, and clear API documentation. **When MMP integration becomes complex, product velocity slows.**
### 4\. Lack of Real-Time Reporting
**Delayed reporting blocks optimization.**
Some MMPs process attribution data with significant lag, sometimes several hours, which doesn’t work in performance-driven environments where bids and budgets change daily—or even hourly.
That might have worked years ago. It doesn’t work in performance-driven environments where media buying decisions happen daily—or hourly.
**Without near real-time insights, creative testing could slow down, bid adjustments lag, and scaling windows close.**
### 5\. Limited Cross-Platform Measurement
Many tools still treat app and web as separate ecosystems.
But subscription brands and AI services operate across web onboarding, app engagement, and cross-device journeys
**If your MMP can’t unify web-to-app and app-to-web attribution, you’re missing the full customer journey.** This fragmentation affects Incrementality testing, Funnel optimization, and Revenue attribution accuracy.
### 6\. Pricing Opacity and Vendor Lock-In
Some vendors restrict raw data access, charge heavily for exports, or lock clients into rigid contracts
For growing companies, this creates risk. **Attribution data should be portable and transparent.** Otherwise, switching costs increase and negotiation power declines.
## The 5 Must-Have MMP Tools Features
Now that you know what is challenging for performance marketers, it’s time to look at some important features that determine whether a mobile measurement partner can deliver accurate attribution, actionable insights, and scalable growth support in a privacy-first ecosystem.
The 5 essential MMP features for 2026 are:
1. Privacy-First Attribution (SKAN + Aggregated Measurement)
2. Real-Time Attribution Data
3. Deep Linking and Seamless User Journeys
4. Ad Fraud Detection and Prevention
5. Flexible Integrations and Automation
Let’s dive in deeply.
### 1\. Privacy-First Attribution (SKAN + Aggregated Measurement)
Privacy-first attribution means measuring performance accurately without relying on user-level identifiers.
Modern MMPs must support:
- [SKAdNetwork](https://abr.ge/sneqiw) configuration and automation
- Conversion value modeling
- Aggregated attribution logic
- Consent-aware tracking
- Future-proof measurement frameworks
**Airbridge’s Actionable Tip:** Map conversion values to early indicators of LTV like trial start, paywall view, or AI feature usage instead of installs.
### 2\. Real-Time Attribution Data
[Real-time attribution](https://abr.ge/idqh6j) gives marketers immediate feedback on campaign performance.
Delayed data creates delayed decisions.
Performance teams optimizing paid channels need:
- Near real-time install attribution
- Event streaming
- Immediate cohort visibility
- Fast anomaly detection
**Airbridge’s Actionable Tip:** Set automated alerts for CPI spikes or conversion drops tied to attribution feeds.
### 3\. Deep Linking and Seamless User Journeys
[Deep linking](https://abr.ge/d10k3u) connects ad clicks to specific in-app experiences.
Without strong deep linking:
- Conversion rates drop
- Onboarding friction increases
- Personalization fails
**Subscription and AI apps benefit heavily from contextual onboarding tied to acquisition source.**
**Airbridge’s Actionable Tip:** Use deferred deep linking to route users directly to subscription offers or feature onboarding tied to campaign messaging.
### 4\. Ad Fraud Detection and Prevention
[Ad fraud](https://abr.ge/ujhbgc4) drains budgets silently.
Strong MMP fraud systems should be able to detect Click injection, Install hijacking, Fake installs, and Bot traffic
**Airbridge’s** **Actionable Tip:** Enable automated and customized fraud blocking instead of post-attribution and rigid reporting whenever possible.
### 5\. Flexible Integrations and Automation
An MMP must integrate across the marketing and data stack, and Developers should have both SDK and server-to-server flexibility.
That includes:
- Ad networks
- Analytics tools
- Data warehouses
- CRM platforms
- BI dashboards
**Airbridge’s Actionable Tip:** Choose an MMP with strong APIs and raw data export to avoid vendor lock-in later.
| Feature | Why It Matters | What to Verify |
| --- | --- | --- |
| Privacy-first attribution | Maintains measurement under ATT/SKAN | SKAN automation, modeling accuracy |
| Real-time reporting | Faster optimization cycles | Data latency benchmarks |
| Deep linking | Improves conversion rates | Deferred linking support |
| Fraud prevention | Protects ad spend | Real-time blocking capability |
| Integrations | Enables data-driven growth | Raw data + API access |
**Quick MMP Feature Evaluation Checklist**
## Why Airbridge Stands Out as the Best MMP Choice
A best-in-class MMP delivers accurate attribution, privacy-safe measurement, developer-friendly infrastructure, and actionable insights in one unified platform without forcing trade-offs between speed, transparency, and scalability.
**Airbridge** stands out because it was built for modern growth teams operating across app, web, and privacy-constrained ecosystems.
### 1\. One View of the Entire Customer Journey
Many MMPs still prioritize mobile-only attribution. Growth teams don’t operate that way anymore. [Airbridge](https://abr.ge/idqh6j) connects Web-to-app attribution, App-to-web journeys, Cross-device behavior, and Subscription lifecycle tracking
In practice, most apps and AI services rely heavily on web onboarding funnels before app engagement. Without unified tracking, attribution breaks mid-journey.
### 2\. Accurate Measurement That Still Respects Privacy
Airbridge handles SKAN and privacy frameworks with AI automation instead of manual overhead.
This includes:
- Conversion value optimization
- Aggregated attribution modeling
- Privacy-safe event tracking
- Future-proof measurement readiness
### 3\. Works Smoothly for Both Business Teams and Engineering
Many tools favor one side. Airbridge balances both.
For marketers:
- Real-time reporting
- Cohort insights
- Cross-channel performance visibility
- Automation capabilities
For developers:
- Lightweight SDK
- Strong documentation
- Flexible server-to-server setup
- Reliable APIs
This reduces internal friction during implementation and scaling.
### 4\. Full Visibility and Control Over Your Own Data
Some vendors restrict raw data access or monetize exports.
Airbridge emphasizes Raw data availability, Warehouse integrations, Flexible reporting, and No black-box attribution logic.
## Your MMP Choice Defines Your Growth Ceiling
Based on everything covered above, you now have the framework to evaluate **MMP tools features** and choose a platform that truly fits your company’s growth model.
The decision ultimately depends on your priorities — data transparency, privacy readiness, developer flexibility, and optimization speed.
That said, a growing number of companies are moving away from legacy MMPs toward solutions that better support cross-platform attribution and faster decision-making.
Many teams are switching to **Airbridge** because it helps them scale more efficiently with unified measurement, real-time insights, and fewer operational bottlenecks.
If your next MMP renewal is coming up, now is the best time to audit these five features. [Talk to Airbridge](https://abr.ge/9x0as3) to benchmark your current setup and see what a modern, privacy‑ready MMP could unlock for your team.
See how Airbridge partners grow with us — and how you can too.
👉[How Buddy.ai secured Top 10 App Store success with Airbridge’s cross-channel measurement](https://abr.ge/sn3r9p4)
👉[Airbridge’s unified cross-platform insights help Shmoody scale to over 1M app installs](https://abr.ge/jjlonz)
👉[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://abr.ge/l0tayy)
---
## Why Your App’s Organic Growth Hits a Ceiling at $10K MRR—And How to Scale Beyond It in 2026
URL: https://www.airbridge.io/en/blog/why-your-apps-organic-growth-hits-a-ceiling-at-10k-mrr--and-how-to-scale-beyond-it-in-2026
Language: English
Published: 2026-03-10
Author: Harper (Trang Nguyen)
Every app founder believes they'll be the exception. You build something great, launch it, watch the early downloads roll in, and assume organic growth will carry you to millions of users. The product is solid, early users love it, word-of-mouth is happening—surely this momentum will compound, right?
Here's the reality check: **organic growth has a structural ceiling that kicks in far earlier than most solopreneurs and early-stage founders expect**.
According to [Amplitude's benchmarking data](https://amplitude.com/benchmarks), the median daily new user growth rate across all apps is just 0.3%, while only the top 10% achieve 3.7% growth. Even more telling, research by RevenueCat shows that [most subscription apps never reach $1,000 in monthly recurring revenue](https://www.revenuecat.com/blog/growth/app-portfolio-vs-single-app/) through organic means alone, and those that do often hit a significant plateau at $10K MRR where traditional tactics stop working.
If you're a solo founder or small team building your first app, this might feel discouraging. But understanding these limitations early actually gives you a massive advantage—you can plan for the transition instead of scrambling when your organic growth inevitably plateaus.
**📌 Key takeaways**
- **Organic Growth Has a Structural Ceiling:** Organic growth eventually plateaus due to demand saturation, algorithm constraints, and diminishing optimization returns — even when retention and conversion improve.
- **Virality Is the Exception, Not the Model:** Most apps never achieve a K-factor above 1. Sustainable scale rarely comes from word-of-mouth alone.
- **Retention Improves Efficiency — Not Volume:** Higher LTV and stronger cohorts increase profitability, but they don’t expand top-of-funnel demand.
- **Paid Ads Expand Demand Beyond Search Limits:** Organic captures existing intent. Paid acquisition creates new exposure, unlocks adjacent audiences, and reduces platform dependency.
- **Measurement Determines Whether Paid Scales or Fails:** Before transitioning from organic to paid, apps need full-funnel attribution, cohort-based LTV modeling, and incrementality tracking to avoid misallocated spend.
## What Is the Organic Growth Ceiling?
The organic growth ceiling is **the natural limit of user acquisition you can achieve without paid amplification**, given your market demand, platform algorithms, and brand reach.
It’s not a failure. It’s saturation.
### So, why does Organic Growth Ceiling exist?
From a tactical standpoint, the ceiling forms when three forces converge:
1. **Demand saturation**
2. **Algorithm constraints**
3. **Diminishing marginal returns from optimization**
Let’s break them down.
#### 1\. Demand Saturation: You’ve Captured the Obvious Users
Every product has a pool of high-intent users actively searching for it. For example:
- People searching “AI meeting note app”
- Users browsing productivity tools in the App Store
- Communities already discussing your problem category
Early on, capturing this group feels like explosive growth. But that audience is finite.
Once you rank for primary keywords and dominate your niche category, new organic installs depend on _new demand entering the market_ — not your internal improvements.
**A common mistake marketers make is assuming better retention or UX will automatically increase top-of-funnel volume.** It won’t. It improves efficiency, not demand size.
#### 2\. Platform Algorithm Constraints
**App Store and Google Play visibility isn’t infinite.** Even if you improve ASO, increase ratings, or even boost conversion rate, you are still competing within a fixed ranking structure.
Only a limited number of apps can occupy top 10 keyword rankings, featured placements, or be in the category charts
> **Airbridge’s take: The App Store Discovery Challenge**
>
> What makes ASO harder is the fact that App store discovery favors search, not browsing. Most users look for specific solutions, which means they already know what they want. This makes it challenging for new apps to get discovered organically.
>
> ASO still matters, but it cannot solve the deeper problem: users rarely find apps they were not already searching for.
>
> Featured spots and top charts reward apps with existing momentum, creating a visibility gap that is especially difficult for new founders without marketing resources.
#### 3\. Diminishing Returns from Optimization
At early stages:
- Improving CVR from 10% → 15% is meaningful.
- Raising retention from 20% → 30% transforms LTV.
At later stages:
- 2% CVR improvement barely moves total installs.
- Retention gains increase revenue per user, not new user volume.
In other words, organic optimization becomes efficiency-driven, not scale-driven. You’re refining performance inside a capped acquisition system.
| Stage | What Drives Growth | Typical Outcome |
| --- | --- | --- |
| Early Stage | ASO improvements, product-market fit, word-of-mouth | Rapid install growth |
| Growth Stage | Retention optimization, referrals, brand visibility | Slower but steady growth |
| Ceiling Stage | High efficiency but flat new user volume | Revenue stabilizes, installs plateau |
**Organic Growth vs Organic Growth Ceiling**
If your app shows strong engagement metrics but stagnant acquisition volume, you’re likely in the third row. And that’s where strategic decisions begin.
> **Airbridge’s take: Viral Growth Is The Exception, Not The Rule**
>
> Many first-time founders believe great products could naturally break this ceiling and spread through word-of-mouth. Startup stories and tech media make viral growth seem normal. The data says otherwise.
>
> Most consumer apps never reach a K-factor above 1.0, meaning each user brings in fewer than one additional user. This is not about product quality. Even apps with satisfied users rarely generate word-of-mouth at a scale that sustains a business.
>
> Breakout cases like Wordle or Among Us are statistical outliers shaped by unique conditions. Even they eventually relied on systematic user acquisition to maintain growth.
## Why Improving Product & Retention Isn’t Enough
**As we said, Improving product quality and retention increases efficiency, but it does not expand total market demand.** Once you approach the organic growth ceiling, optimization alone cannot create new user volume at scale.
### 1) The User Retention Harsh Reality
Here’s a hard truth for new app builders: retention is tougher than you think. [Benchmarks](https://amplitude.com/benchmarks) **show many consumer apps lose 75–85% of users within the first month.** That’s not just a product issue. It’s market reality.
Users have endless options and almost no switching costs. They download impulsively, test briefly, and move on, so even strong apps compete for limited attention. This makes organic growth fragile. If you gain 100 users per week but lose 80 within a month, growth depends on constantly increasing acquisition.
And the worst part? Even if you manage to retain a large number of users, it does not generate new awareness at scale. If 10,000 users discover your app organically each month, increasing retention from 25% to 40% improves revenue per cohort — but you still start with 10,000 users, and the size of the funnel hasn’t changed..
### 2) Word-of-Mouth Has Natural Saturation
Strong products spread. Especially in tight communities like AI builders, SaaS founders, Productivity enthusiasts, or Developer ecosystems
But referral loops weaken over time. Why?
- Your most enthusiastic users have already shared.
- New users have lower referral intent.
- Social overlap reduces new reach (friends already exposed).
Growth from word-of-mouth follows an S-curve. It accelerates early, then flattens as audience overlap increases.
### 3) TAM Isn’t Static — But Your Reach Is
[Total Addressable Market (TAM)](https://www.tempo.io/glossary/total-addressable-market) may be large. But organic reach is constrained by:
- Search volume for core keywords
- App Store browsing behavior
- Existing brand awareness
- Community penetration
For example, If monthly search volume for your main keyword is 50,000 globally, even ranking #1 doesn’t give you unlimited installs. You are competing for a capped pool of demand.
> **Airbridge’s take: Signals You’ve Hit the Organic Ceiling**
>
> 1. **Install Volume Is Flat — Despite Higher Conversion Rates:** You improve App Store page CVR, Onboarding completion, and Free-to-trial conversion, yet monthly new users barely move.
>
> 2. **Brand Search Volume Plateaus:** If branded search queries stop growing, awareness has stabilized. You need to check App name search trends, Direct traffic to landing pages, or Branded keyword impressions in app stores.
>
> 3. **Keyword Rankings Stabilize (But Don’t Climb):** You may rank Top 3 for secondary keywords or Top 10 for high-volume category terms, but positions stop improving. This is algorithm equilibrium.
>
> 4. **Retention Improves — Revenue Grows — Installs Don’t:** You will see Higher LTV, Improved renewal rates, Strong cohort monetization, or Slight revenue rises, But net new user count remains static.
>
> 5. **Referral Contribution Declines:** Your core evangelists already shared. New users are less activated. Audience overlap increases.
>
## Why Apps Turn to Paid Ads at This Stage
**Apps turn to paid ads when organic growth reaches structural limits and internal optimization no longer increases acquisition volume.** Paid media becomes a controlled mechanism to expand demand beyond existing visibility constraints.
### 1) Paid Expands the Addressable Audience
Organic acquisition depends on people already searching, browsing, or hearing about your product. That audience is finite.
Paid acquisition changes the equation. Instead of waiting for users to search “AI note-taking app” or discover your category organically, you proactively introduce your product to:
- Adjacent audiences
- New geographic markets
- Lookalike segments
- Users with latent demand
**For subscription apps and AI tools especially, many potential users don’t know they need the product yet.** Paid campaigns allow you to shape awareness instead of waiting for it.
### 2) Paid Reduces Platform Dependency
Organic growth is largely governed by platform algorithms. Rankings shift. Editorial placements rotate. Competitors adjust pricing and metadata.
Even with strong ASO and retention, you remain dependent on systems you don’t control. However, Paid media introduces operational control, where you can decide budget allocation, audience segmentation, creative direction, or scaling pace
### 3) Paid as a Learning Engine, Not Just an Acquisition Channel
One overlooked advantage of paid media is learning velocity.
Organic testing cycles are slow. If you have fixed traffic volume, experiments take months to reach statistical significance, while Paid campaigns accelerate insight generation. You can rapidly test value propositions, creative angles, pricing narratives, or targeting clusters
In practice, advanced growth teams use paid acquisition as a structured experimentation system. Even if early CAC looks high, the data generated informs positioning, onboarding, and product strategy.
### 4) The Organic Lift Effect
There’s a persistent belief that paid traffic cannibalizes organic installs. That happens only when attribution is misconfigured.
Strategically deployed paid campaigns often increase:
- Branded search queries
- Category visibility
- Store ranking signals
- Review volume
**Paid doesn’t replace organic. It can amplify it** — if incrementality is measured correctly.
Without proper attribution, however, teams may mistake organic lift for paid efficiency or vice versa. That’s where many scaling attempts break down.
## What to Prepare Before Transitioning from Organic to Paid
**Before shifting from organic growth to paid acquisition, you must ensure your measurement, economics, and experimentation systems are mature enough to handle scale.** Otherwise, paid spend exposes weaknesses instead of unlocking growth.
### **1\. Attribution Infrastructure That Goes Beyond Installs**
Once you start spending money to acquire users, measurement transforms from "nice-to-have" to "business-critical." You need to understand:
- Which acquisition channels deliver users who actually engage with your app
- How different user segments behave after installation
- Which creatives lead to better CPI, CPA and ROAS
- Where to allocate your limited marketing budget for maximum impac
This creates immediate demand for attribution tracking, deep linking capabilities, and post-install analytics. The solopreneurs who prepare for this transition early—implementing measurement tools while still growing organically—have a significant advantage over those who scramble to add tracking after their growth stalls.
### **2\. A Clear LTV and Payback Model**
Paid growth shifts the constraint from demand to unit economics. You need clarity on:
- Blended LTV
- Segment-level LTV (by geo, device, audience)
- Acceptable CAC threshold
- Target payback period
For example, If your average 6-month LTV is $120 and your acceptable payback window is 3 months, your CAC ceiling might sit around $60–$70 depending on margin structure.
Without this clarity, teams either overspend aggressively, or shut down campaigns too early
### **3\. Creative Testing Framework**
When organic growth slows, messaging stagnation often follows.
Paid media exposes this quickly. Before scaling with paid ads, define:
- Core value propositions to test
- Persona-based messaging variations
- Visual hooks (problem-focused vs outcome-focused)
- Short-form vs long-form narrative angles
Creative fatigue is one of the biggest hidden growth killers in paid acquisition. That’s why a structured testing cycle should include:
- Weekly creative iteration
- Performance segmentation by audience
- Clear winner thresholds
- Rapid budget reallocation
### **4\. Budget for Learning — Not Just Scaling**
The first 30–60 days of paid campaigns are data collection phases. Expect:
- Volatility
- Learning inefficiencies
- Algorithm calibration
If leadership expects immediate profitability at scale, campaigns will be shut down before optimization stabilizes.
## How Leading Apps Combine Organic & Paid Strategically
Understanding the organic growth ceiling doesn't mean abandoning organic strategies—it means planning intelligently for what comes next. Here's how smart solo founders approach this transition:
1. **Start measurement early**: Implement attribution and analytics tools when you’re beyond PMF point. This gives you clean baseline data and time to learn the tools before you need them urgently.
2. **Reserve budget for testing**: Don't spend every dollar on product development. Set aside 10-15% of your available runway for user acquisition experiments, even if you're not ready to scale them yet.
3. **Design for multiple channels**: Consider how your app store assets, onboarding flow, and value proposition will work in paid advertising contexts, not just organic discovery scenarios.
4. **Study your category**: Research how successful apps in your space handle the transition from organic to paid acquisition. Look at their creative strategies, messaging approaches, and user acquisition funnels.
5. **Focus on unit economics**: Since you'll eventually pay for users, understanding and optimizing lifetime value becomes crucial to making paid acquisition profitable.
6. **Build incrementally**: You don't need to launch massive advertising campaigns. Start with small tests on one or two channels like GMAT (Google, Meta, Apple Ads and TikTok) to learn what works before scaling investment.
The goal isn't to rush into paid advertising, but to be prepared when organic growth naturally plateaus. Solo founders who understand this dynamic can make the transition smoothly rather than panicking when their growth rate inevitably slows.
## Conclusion: The Right Timing for Paid Growth
As a solo founder, you probably have a mental timeline: launch, grow organically for 6-12 months, then maybe consider paid advertising if you need to accelerate growth. The reality is often more compressed.
[According to Adapty's analysis](https://adapty.io/blog/unlocking-growth-to-100k-mrr/), many apps hit meaningful growth slowdowns much earlier than founders expect, often around the $1K-10K MRR range. At this point, you've likely exhausted your immediate network, optimized your app store presence, and maybe gotten some press or social media coverage. The accessible organic opportunities are tapped out.
This timing can feel jarring, especially if you're a bootstrap founder who was counting on organic growth to fund your next phase of development. But recognizing this pattern helps you prepare better resource allocation and timeline expectations.
---
## Best Singular Alternatives in 2026: Key Features to Evaluate Before You Switch
URL: https://www.airbridge.io/en/blog/best-singular-alternatives-in-2026-key-features-to-evaluate-before-you-switch
Language: English
Published: 2026-03-09
Author: Luke
Singular is a well-known MMP, but many growth teams start to feel its limits as they scale. Slow query speeds, short data retention, and rigid reporting can turn everyday optimization into friction, especially when user journeys span both web and app.
If you’re evaluating a **Singular alternative for 2026**, this guide breaks down what to look for and why performance-led teams are increasingly choosing **Airbridge** for unified measurement, native deep linking, and faster access to insights.**📌Key Takeways**
- **Singular works, but shows limits at scale.** Slow queries, short data retention, and rigid reports reduce day-to-day efficiency.
- **User journeys are no longer app-only.** Web-to-app and cross-device flows require unified measurement, not stitched-together tools.
- **Deep linking is core infrastructure.** Singular supports the basics. Airbridge handles deferred and contextual routing natively.
- **Most leading MMPs deliver similar attribution accuracy.** Differences emerge in flexibility, operational overhead, and cost efficiency.
- **Performance-led teams are favoring adaptable platforms.** Tools that unify measurement, linking, and data access are gaining momentum as growth stacks mature.
## What Is Singular ?
**Singular is a Mobile Measurement Partner (MMP)** that helps marketers attribute installs, measure campaign performance, and consolidate data across paid media channels.
Teams use Singular to understand where users come from, how campaigns perform, and which channels drive revenue rather than just installs. The platform offers attribution tracking, fraud prevention, and reporting capabilities that have attracted mobile marketers seeking an attribution solution.
### 1) Why are teams looking for Singular alternatives?
However, real-world implementation often reveals performance gaps that impact day-to-day operations. Based on competitive analysis and [user feedback](https://www.trustradius.com/products/singular/reviews?qs=pros-and-cons), reasons why people look for Singular alternatives are below:
- **Slow query performance**: Delays in data processing make it harder for teams to react quickly, especially during high-impact campaign launches or rapid market shifts.
- **Limited historical data depth**: One-year cohort analysis restricts long-term LTV insights and strategic planning for products with longer user lifecycles.
- **Inflexible reporting**: Rigid report structures and fixed definitions make it difficult to customize KPIs, attribution models, or non-standard conversion events.
## What you should consider to choose the best MMP for your teams
Probably, you’ve already started researching alternatives to Singular. But with so many vendors promising better attribution, deeper analytics, cleaner data, and privacy-safe measurement, it’s easy to feel overwhelmed. should prioritize several key capabilities that directly impact campaign success:
### 1) Attribution trust and data consistency
An MMP should provide stable, explainable attribution across paid channels, SKAN, and organic traffic. During migrations, teams often discover gaps between dashboards and raw data exports, and the right MMP reduces this risk through transparent attribution logic, deterministic fallback methods, and clear documentation for edge cases.
### 2) Real-Time Data Processing and Query Speed
Modern marketing relies on fast access to insights. Slow query processing delays optimization and weakens a team’s ability to respond to performance changes. Strong MMPs deliver complex attribution and performance reports in seconds and maintain speed as data volume scales.
### 3) Comprehensive Data Retention and Historical Analysis
Long-term growth decisions require access to deep historical data. An effective MMP should support year-long cohort analysis so teams can track long-term user behavior, calculate accurate lifetime value, and uncover seasonal trends that shape future acquisition and retention strategies.
### 4) Unified Web-App & Cross platforms Measurement
User journeys increasingly span web and app touchpoints, making unified measurement essential. MMPs that fully integrate web-app, and cross platform tracking provide a clearer view of attribution, while platforms that treat web measurement as secondary risk creating blind spots that lead to inefficient budget allocation and missed optimization opportunities.
## Here Are Top Singular Alternatives & Competitors in 2026
Below are the most commonly evaluated Singular competitors among subscription apps, AI service providers, and performance teams managing multi-channel growth.
### 1) AppsFlyer
Appsflyer is often the first name that comes up in MMP comparisons. It offers strong attribution reliability, broad ad network integrations, and deep experience with SKAN. For teams operating at large scale, AppsFlyer delivers consistency and ecosystem maturity.
However, many marketers report similar challenges to Singular around pricing tiers, operational complexity, and limited flexibility when adapting attribution logic to custom growth models.
### 2) Airbridge
[**Airbridge**](https://www.airbridge.io/) is increasingly evaluated as a direct Singular alternative because **it combines enterprise-grade attribution with a more flexible data and linking architecture.**
Airbridge is positioned for teams that need accurate measurement without treating web, app, and deep linking as separate systems. That’s why **Airbridge is the strongest among brands** that have outgrown install-only measurement and now optimize for revenue, retention, and lifecycle performance.
### 3) Adjust
Adjust is another established MMP known for solid attribution and fraud prevention. It performs well for mobile-first businesses that prioritize campaign measurement and compliance.
That said, teams running web-to-app funnels often find Adjust’s web measurement capabilities less intuitive, requiring additional configuration or parallel analytics tools to fully understand cross-platform journeys.
### 4) Branch
It appears frequently in evaluation conversations, particularly when deep linking is a priority. Its strength lies in link management and routing logic rather than full-funnel attribution.
As a result, Branch is often paired with an MMP rather than replacing one entirely. For teams trying to simplify their stack, this creates an additional integration layer rather than a true Singular replacement.
To make these differences clearer, experienced teams often compare tools across a small set of criteria that actually affect day-to-day performance.
| Criteria | Singular | AppsFlyer | Adjust | Airbridge |
| --- | --- | --- | --- | --- |
| Core attribution accuracy | Strong | Strong | Strong | Strong |
| Web-to-app measurement | Limited | Moderate | Limited | Native & unified |
| Deep linking flexibility | Basic | Moderate | Basic | Advanced (deferred & contextual) |
| Operational complexity | High | High | Medium | Low–Medium |
| Cost efficiency at scale | Low–Medium | Medium | Medium | High |
| Best fit | Large enterprises | Large enterprises | Mobile-first apps | Subscription & hybrid growth teams |
## Singular vs Airbridge: Why Performance Marketers Are Switching
Especially, Airbridge solves the same core problem which Singular can—attribution—but [they approach modern growth very differently.](https://abr.ge/xt2w8tx)
### **1) Difference 1: Web-to-app measurement capabilities**
Both platforms can measure installs, conversions, and revenue across major ad networks. However, the gap appears once teams move beyond install-centric optimization.
One of the clearest differences shows up in **web and app unification**. Singular remains app-first, with web data often handled as a separate layer. For businesses running subscription sign-ups on web or onboarding flows that start in a browser, this creates blind spots.**Airbridge treats web and app as one attribution surface**, allowing teams to follow users across environments without stitching together multiple tools or reports.> **Airbridge’s take**
> Galaxy Play used Airbridge’s web-to-app measurement to track revenue generated both inside and outside the app, connecting web purchases and in-app conversions into a single attribution view.
>
> This eliminated blind spots that other MMPs couldn’t cover and helped them to achieve a 22% revenue increase alongside over 50% improvement in MMP cost efficiency.
>
> 👉 Learn [How Galaxy Play achieved a 22% revenue increase powered by Airbridge Web-to-App tracking](https://abr.ge/i8wyia).
### **2) Difference 2: Deep linking tools**
**Deep linking capability is another major divergence.** Singular’s deep links handle standard routing but struggle with advanced scenarios. Deferred deep linking for new users, campaign-based contextual routing, and predictable fallback behavior often require custom logic or external tools.
[**Airbridge’s deep linking engine**](https://abr.ge/5aywvxp) is built to handle these flows natively, which **reduces engineering overhead and eliminates inconsistent user experiences across channels.> Airbridge’s take**
> Fizz successfully migrated from Firebase Dynamic Links to Airbridge API-powered deep links, ensuring all of its referral, sharing, and campaign links continued to work without disruption despite deprecated Firebase’s service.
>
> By using Airbridge’s API deep links to automate link creation at scale, Fizz maintained zero performance loss and 100% attribution continuity across its high-volume content sharing and user acquisition flows.
>
> 👉 Learn [How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://abr.ge/ugaqt1).
### **3) Difference 3: Pricing** **Cost** is often the final catalyst. Singular’s pricing reflects its enterprise positioning, which can feel misaligned for high-growth subscription and AI-driven products.
[**Airbridge’s pricing model**](https://abr.ge/nxzn0ei) is typically easier to map to actual usage and ROI, especially for teams scaling channels without wanting their measurement costs to scale at the same rate.
## Who Should Choose Airbridge Over Singular?
Of course, Airbridge is not designed to replace Singular for every company. The strongest fit appears where **flexibility, web coverage, and operational efficiency** matter as much as raw attribution accuracy.
### 1) Subscription-based apps
These teams optimize for trials, renewals, churn reduction, and lifetime value—not just CPI. When acquisition starts on the web and converts in-app,
### 2) ROI-focused teams
**Airbridge’s unified web and app measurement** removes the need for parallel analytics stacks or manual reconciliation. This makes it easier to evaluate true channel ROI across the entire subscription lifecycle.
### 3) AI service providers and SaaS-like mobile products
Their user journeys are rarely linear. Users may discover the product on the web, test features before installing, or switch devices during onboarding.
Singular can measure parts of this journey, but **Airbridge is designed to follow the user across touchpoints without fragmenting attribution logic or reporting.**
[Illustration of subscription apps’ complex web-to-app user journey. Source: Airbridge]
[Illustration of attribution differences between standard SDKs and the Airbridge Web SDK. Source: Airbridge]
From an execution standpoint, **performance teams managing multiple channels often** prefer Airbridge when speed matters. Launching new campaigns, testing creative variations, or adjusting deep link behavior can be done without repeated engineering involvement.
A common mistake growing teams make is assuming that attribution tools should require constant developer support. Airbridge reduces that dependency by design.
For **developers and data teams**, predictability is a deciding factor. **Airbridge offers clearer SDK behavior, consistent event handling, and easier integration with data warehouses and BI tools.** This matters for organizations that treat attribution data as production data rather than marketing-only reporting.
## FAQ: Singular Alternatives, Cost, and Replacement Considerations
### **1) Can Airbridge fully replace Singular without losing attribution quality?**
Yes. Airbridge provides comparable attribution accuracy while offering better web coverage and simpler operations.
### **2) Which MMP is best for unified web and app attribution?**
Airbridge is frequently selected by teams that need web and app measured as a single user journey rather than separate funnels.
### **3) Does Airbridge support deferred and contextual deep linking?**
Yes. Airbridge supports both deferred deep linking and contextual routing natively, without extra tools or heavy engineering.
### **4) How does Airbridge compare in deep linking capabilities?**
Airbridge provides native deep linking fully integrated with attribution, enabling seamless routing from ads to in-app content across web and mobile. Unlike solutions that require separate tools or complex setup, Airbridge connects deep linking and measurement in a single system, reducing technical overhead and improving user journey tracking accuracy.
### **5) Which MMP is better for startups and mid-market teams?**
Airbridge is often preferred by startups and mid-market companies due to faster setup, clearer pricing, and lower operational overhead.
## Conclusion: Prioritizing Performance Over Brand Recognition
The MMP landscape continues to evolve, with performance capabilities becoming increasingly important differentiators.
While Singular maintains market presence, marketers who prioritize real-time optimization, sophisticated measurement, and flexible reporting are finding superior value in alternatives like Airbridge.
**The key to successful MMP selection lies in evaluating platforms based on actual performance requirements rather than market perception.**
For teams that need fast query processing, comprehensive incrementality testing, and unified cross-platform measurement, the choice becomes clear: performance-focused platforms consistently deliver better results than those relying primarily on brand recognition.
---
## 4 Best AppsFlyer Alternatives for 2026: A Deep Dive into Costs & Attribution Accuracy
URL: https://www.airbridge.io/en/blog/4-best-appsflyer-alternatives-for-2026-a-deep-dive-into-costs-attribution-accuracy
Language: English
Published: 2026-03-09
Author: Luke
Mobile measurement is at a crossroads. Following the shutdown of Firebase Dynamic Links on August 25, 2025, and growing frustration over feature paywalls at traditional MMPs, mobile marketers are being forced to rethink their entire measurement strategy.
**If you're looking for an AppsFlyer alternative, you're not alone—and this disruption might be exactly the opportunity your team needs to upgrade to something better.**
**Key Takeaways**
- Teams look for AppsFlyer alternatives due to high costs, feature paywalls, and operational complexity, especially as event volume and subscriptions scale
- Modern MMP needs go beyond installs to include full-funnel attribution, deep linking, subscription revenue, and LTV, without premium add-ons
- AppsFlyer and Adjust offer enterprise-grade control but often feel too heavy and expensive for lean or fast-moving teams
- Branch excels at deep linking and web-to-app journeys, while Tenjin appeals to cost-sensitive, early-stage teams with simpler needs
- Airbridge positions itself as a high-performance alternative by offering AppsFlyer-level attribution accuracy with transparent pricing and easier usability
- In 2026, the best AppsFlyer alternative is the MMP that delivers accurate attribution, privacy readiness, and fast time-to-value without locking core features behind paywalls
## Why Teams Start Looking for an AppsFlyer Alternative
**A Mobile Measurement Partner (MMP)** like Appsflyer helps app businesses track where users come from, attribute installs and in-app events, and measure marketing performance across channels.
Maybe you have already heard about AppsFlyer and wondered whether it is truly the right fit for your app. You may have compared reviews, explored pricing, and even searched for alternatives to see if there is a solution that better matches your budget, technical setup, and growth goals.
Teams start looking for an **AppsFlyer alternative** when they want the same core attribution accuracy but with **lower cost, simpler setup, or a product better aligned with their growth stage and campaign mix.**
### Operational Complexity for Lean Teams
AppsFlyer works well with dedicated analytics or ops resources. Smaller teams feel the friction more. G2 said typical feedback from performance marketers with 3–5 years of experience:
- Setup requires careful coordination with developers
- Ongoing maintenance adds overhead
- Dashboards feel powerful but heavy for quick decision-making
[Reviews of AppsFlyer. Source: G2]
### The Paywall Problem
Many teams have grown frustrated with traditional MMPs like Appsflyer that "upsell every step" and restrict critical features behind premium tiers, said [Metacto](https://www.metacto.com/blogs/the-complete-guide-to-appsflyer-costs-setup-integration-maintenance). Locked features create artificial barriers, and when attribution, advanced analytics and deep linking capabilities are paywalled, you're forced to choose between budget and accuracy.
This model shifts decisions away from real business needs and toward paying extra for capabilities that should be standard in any comprehensive measurement platform.
### Mismatch With Specific Campaign or Product Needs
AppsFlyer is optimized for broad use cases. [Some teams need more flexibility](https://www.g2.com/products/appsflyer/reviews?qs=pros-and-cons). Examples include:
- Campaign mixes heavy on web-to-app or deep linking
- Products prioritizing subscription lifecycle metrics over raw installs
- Teams that value faster insight over deep customization
When the platform doesn’t map cleanly to how growth actually happens, teams start comparing **AppsFlyer competitors** more seriously.
## What to Look for in an AppsFlyer Alternative
When evaluating alternatives, focus on capabilities that address the core limitations driving your search in the first place. Here's what actually matters:
### Comprehensive Attribution Without Paywalls
Your alternative should provide full-funnel attribution as a standard feature, not a premium add-on. **Look for solutions that track conversions across all platforms and touchpoints without restricting access based on pricing tiers.** The goal is complete marketing intelligence, not fragmented insights based on what you're willing to pay for.
### Robust Deep Linking Capabilities
Deep linking isn't just about sending users to specific app pages—it's about creating seamless user experiences across all marketing channels. Your alternative **needs to handle different deep link types (URI Schemes, Universal Links, App Links) and provide reliable attribution** for users who arrive through these links.
The solution should also handle edge cases like **deferred deep linking** **for users who don't have your app installed yet**, ensuring no conversion opportunity is lost due to technical limitations.
### Major Platform Partnerships & Integration
Seamless integration with advertising platforms is non-negotiable. **Your alternative must maintain official partnerships with Meta, Google Ads, TikTok, and other major platforms you use for user acquisition.** These partnerships ensure data flows smoothly between your MMP and ad platforms without manual workarounds or data discrepancies.
### Migration Support and Documentation
Moving measurement platforms is complex, so look for **alternatives that provide dedicated migration support, comprehensive documentation, and clear timelines for implementation.** The switch should improve your measurement capabilities, not disrupt your marketing operations.
### Transparent Pricing Structure
After dealing with paywall frustrations, pricing transparency becomes crucial. Your alternative should **clearly communicate what's included at each pricing level without hidden fees or surprise upsells for essential features.** You want to budget based on your business needs, not feature access restrictions.
### Privacy Readiness
Modern MMPs must handle platform constraints by default. At a minimum, you should look for:
- Full SKAdNetwork and Privacy Sandbox support
- Consent-based measurement frameworks
- Server-side and modeled attribution options
## Top AppsFlyer Alternatives: Feature & Cost Comparison
Despite the number of MMPs on the market, most evaluations narrow quickly. Based on market data and buyer behavior, teams usually **compare AppsFlyer with Adjust, Branch, Tenjin, and** [**Airbridge**](https://abr.ge/d4zv0x)**.** Here's how the leading options stack up:
| Platform | Core Strengths | Common Limitations | Pricing Model | Best Fit Teams |
| --- | --- | --- | --- | --- |
| AppsFlyer | Enterprise-grade attribution, deep analytics, strong fraud prevention | High cost at scale, feature overkill for many teams, steeper learning curve | Volume- and feature-based tiers | Large-scale apps with dedicated analytics teams |
| Airbridge | Attribution parity, flexible analytics, strong deep linking, transparent pricing | Growing brand awareness compared to legacy players | Scalable, predictable pricing | Subscription apps, AI services, mid-market teams |
| Adjust | Robust attribution, strong global presence, reliable reporting | Similar pricing pressure to AppsFlyer, feature-dense UI | Volume-based enterprise pricing | Global apps with mature UA operations |
| Branch | Best-in-class deep linking, strong web-to-app flows | Attribution depth less flexible for some use cases | Package-based pricing | Teams prioritizing deep linking over full-funnel attribution |
| Tenjin | Simple setup, UA-focused dashboards, cost-conscious | Limited customization, lighter analytics depth | Lower, transparent pricing | Indie studios and early-stage teams |
### How AppsFlyer Cost Compares in Real Scenarios
AppsFlyer cost is often justified early on. The friction appears as scale increases. From a practical standpoint, teams usually feel the pressure when:
- Event volume grows due to subscriptions or frequent product interactions
- Multiple channels run in parallel, increasing attribution load
- Advanced features become required rather than optional
Tired of paying for features that should be standard? [See how Airbridge delivers enterprise attribution without the enterprise paywall](https://abr.ge/757220)
### Usability Is Where Many Teams Change Direction
AppsFlyer and Adjust are designed for maximum control. That control assumes dedicated resources. **Smaller teams and fast-moving AI companies often prefer platforms that:**
- Require less ongoing developer involvement
- Offer usable dashboards out of the box
- Reduce the time between data collection and decision-making
### Why Branch and Tenjin Enter the Conversation
Branch is frequently evaluated when deep linking and web-to-app flows are central to growth. Its strength lies in user journey continuity rather than full-funnel attribution depth.
Tenjin appears in comparisons when cost sensitivity is high and the use case is primarily user acquisition reporting. It works well for simpler setups but can feel limited as measurement needs grow.
### Where Airbridge Positions Differently
[Airbridge](https://abr.ge/757220) tends to surface later in the evaluation cycle—**when teams realize they don’t need to downgrade attribution quality to reduce cost or complexity.**
Instead of competing on “more features,” Airbridge competes on:
- Attribution reliability comparable to AppsFlyer
- Analytics that support everyday decisions
- Pricing that scales with actual business value
## Why Airbridge Is a Strong AppsFlyer Alternative
A growing number of brands move [from AppsFlyer to Airbridge](https://abr.ge/lhs1gv) due to a fundamentally different pricing approach to mobile measurement, addressing the core frustrations that drive teams to seek alternatives in the first place.
### No-Paywall Measurement Philosophy
Unlike traditional MMPs that restrict features behind premium tiers, **Airbridge provides comprehensive attribution, deep linking, and analytics without artificial limitations.** This means you get complete visibility into your marketing performance based on your business needs, not your willingness to pay for feature access.
### Strong Deep Linking and Web-to-App Cross-Platform Measurement
Airbridge delivers advanced **deeplink** functionality designed for complex, fragmented user journeys. With a robust deeplink infrastructure, marketers can seamlessly connect web, mobile web, and app experiences without losing attribution accuracy.
This is especially critical in:
- Web-to-app flows
- Subscription onboarding paths
- AI products with repeated login and engagement cycles
For teams that rely heavily on web acquisition before driving app installs, Airbridge’s dedicated [**Deeplink Plan**](https://abr.ge/vigxykk) ensures reliable routing, deferred deeplink support, and full cross-platform measurement visibility.
### Proven Platform Partnerships & Integrations
[Airbridge maintains official partnerships](https://abr.ge/2excxm) with Meta, Google Ads, and TikTok, ensuring your measurement data integrates seamlessly with your advertising platforms. These partnerships eliminate the manual data reconciliation and workflow disruptions that can occur during MMP migrations.
### Subscription/AI Journey Tracking
Beyond addressing immediate needs like deep linking and attribution, Airbridge provides comprehensive mobile measurement capabilities including user journey analytics, retention insights, 3-layer fraud protection, and campaign optimization tools. This positions your team for long-term success rather than just solving immediate technical requirements.
## Final thoughts: AppsFlyer worked. Now it’s time to work smarter.
There’s no denying it. AppsFlyer has been the default choice for years, and for a long time, it made sense. Many teams built their measurement stack around it, scaled with it, and learned the hard way what “enterprise-grade” really means in practice.
But the market has moved on. Budgets are tighter, teams are leaner, and paying extra just to unlock the basics no longer feels justified.
If you want attribution quality on par with AppsFlyer, without the heavy paywalls, operational drag, or price shock as you scale, Airbridge is built for exactly that moment. Learn how big names are leveraging Airbridge for excellent results:
👉[How Playio Increased D30 Retention to 30% and Cut Global UA CPA by 40% with Airbridge](https://abr.ge/jqz1p7)
👉[Nightly Cuts CPA 18% with Simulated iOS Attribution, Ranks Top 3 in Japan’s App Store](https://abr.ge/ubscdbp)
👉[Airbridge’s unified cross-platform insights help Shmoody scale to over 1M app installs](https://abr.ge/a016utf)
It gives you the accuracy, deep linking, and platform partnerships you expect from a top-tier MMP, with pricing and usability that actually match how modern teams work.
## FAQ about AppsFlyer Alternatives and MMP Comparison
### **Which alternative is best for small startups or early-stage teams?**
Smaller teams often prioritize **ease of setup, predictable pricing, and lower maintenance overhead**. Tools like **Airbridge and Tenjin** frequently surface in user comparisons for simplicity and lower barrier to entry, while still offering credible attribution performance suitable for early growth stages.
### **Which MMP is best for subscription apps?**
For subscription apps, the best MMP is one that tracks installs, conversion events, retention, and revenue without making event volume prohibitively expensive. Many subscription-based teams choose Airbridge because it balances attribution accuracy with predictable pricing and analytics designed around lifecycle performance.
### **Can Airbridge fully replace AppsFlyer?**
Yes, for many use cases. Airbridge supports install attribution, post-install event tracking, deep linking, and privacy-compliant measurement at a level comparable to AppsFlyer. Teams that switch typically do so without sacrificing attribution accuracy, especially for subscription apps and AI services.
### **What’s the main difference between AppsFlyer and its competitors?**
The biggest difference is not attribution logic—it’s **cost structure, usability, and operational complexity**. AppsFlyer and Adjust focus on enterprise flexibility, while tools like Airbridge prioritize faster time-to-value, simpler workflows, and pricing aligned with growth-stage needs.
---
## 5 Types of Teams That Need Airbridge Core Plan for Better App Marketing Attribution in 2026
URL: https://www.airbridge.io/en/blog/5-types-of-teams-need-airbridge-core-plan-for-better-app-marketing-attribution-in-2026
Language: English
Published: 2026-03-09
Choosing the right app marketing tool is one of the most consequential decisions a subscription app team makes early on. **Every week without proper attribution is a week of budget scaling on assumptions, not data.**
Startups and AI subscription apps often begin with small marketing teams, but quickly face a critical question as they start running paid acquisition:
> **_"Will this tool still work when our team grows from three people to a full growth department?"_**
Many platforms are either too complex for lean teams or too limited once paid user acquisition begins scaling. The real challenge is **attribution visibility** — connecting ad spend directly to trial users and paying subscribers, not just installs or clicks.
Below, we'll cover what growing app teams should look for in an attribution tool, how Airbridge Core Plan is designed to fill that gap, and which five team types benefit most from it.
**📌 Key Takeaways**
- **App marketing success depends on attribution visibility:** growth teams must connect paid acquisition with trial and subscription conversions, not just installs.
- **The best app marketing tools connect ad spend directly to revenue:** without attribution across Meta, Google, TikTok, and Apple Search Ads, optimization becomes guesswork.
- **Subscription apps require deeper funnel analytics:** tracking install → trial → subscription → retention is essential for improving LTV and marketing ROI.
- **Early-stage teams need fast, focused analytics:** tools that deliver quick attribution signals help teams validate paid acquisition without complex infrastructure.
- **Airbridge Core Plan is built for subscription apps:** it connects paid acquisition with subscription revenue without requiring enterprise-level data infrastructure.
## **What Growing App Teams Should Look for in an App Marketing Tool**
As app companies move from early traction to paid acquisition, the requirements for a measurement tool change quickly. Early teams don't just need install tracking — they need visibility into **which marketing efforts actually generate paying users.**
For subscription and AI-powered apps, the most important metric isn't installs — it's subscription conversions and retained users. Without attribution that connects UA spend to revenue, teams often scale campaigns blindly.
A performance marketer may see installs increasing from Meta or Google campaigns, but without proper attribution it's difficult to know which channel actually produces trial users that convert into subscriptions. This creates a major growth blocker: budgets increase while revenue efficiency remains unclear.
Growing teams typically prioritize four capabilities when evaluating app marketing tools:
- **Clear attribution for paid acquisition:** understand which channels drive installs, trials, and subscriptions
- **Fast implementation:** small teams cannot spend weeks building complex event schemas or analytics pipelines
- **Coverage of major ad platforms:** most early-stage app marketing spend focuses on a few dominant channels
- **Actionable growth insights:** reports that reveal conversion funnels, retention, and revenue performance
## **What Is Airbridge Core Plan? An Attribution Platform for Subscription App Marketing**
Airbridge Core Plan is designed to answer one question for subscription app teams: "**Which paid acquisition channels are actually generating subscription revenue?**"
It is part of [Airbridge](https://abr.ge/gm60ts)'s scalable product framework and focuses on delivering fast, reliable attribution signals for early-stage paid acquisition teams — particularly those operating subscription-based mobile apps.
**Where Core Plan fits in the Airbridge product ladder:**
| Plan | Primary Purpose | Best For |
| --- | --- | --- |
| Base Plan | Deep linking and user routing | Pre-UA teams validating product |
| Core Plan | Paid acquisition → subscription conversion measurement | Teams launching or scaling paid UA |
| Growth Plan | Advanced measurement and optimization | Mature teams with complex analytics needs |
This structure allows startups to begin with focused attribution capabilities at the Core Plan stage, while maintaining **a clear upgrade path as their marketing organization scales.**
Here's how Core Plan stacks up against the four criteria that matter most for early-stage subscription app teams:
| What to look for in an app marketing tool | Enterprise MMP | Airbridge Core Plan |
| --- | --- | --- |
| Startup-friendly pricing (no large monthly minimums) | Often requires custom pricing or sales contact | ✅ Pay-As-You-Go |
| Fast setup — attribution running in hours, not weeks | Typically requires dedicated implementation | ✅ Hours, not weeks |
| Built-in subscription funnel events (no custom schema) | Requires custom event design | ✅ Standard events included |
| Coverage of major ad channels (Meta, Google, Apple, TikTok) | ✅ All networks supported | ✅ GMAT (80–90% of spend) |
| No data engineer required for initial setup | Often requires analytics expertise | ✅ Designed for small teams |
| Focused on subscription app attribution | Built for enterprise scale | ✅ Purpose-built for startups |
### **1\. Built for Early Paid UA Teams**
The Core Plan is specifically designed for small app marketing teams running their first serious paid acquisition campaigns. These teams typically have limited engineering resources and need fast time-to-value rather than complex custom analytics setups.
Typical characteristics of teams that benefit most:
- **Business model:** subscription apps with a trial → paid flow
- **UA maturity:** early-stage paid acquisition experimentation
- **Technical resources:** limited ability to design complex event schemas
- **Growth need:** quick visibility into which channels drive paid subscribers, not just installs
### **2\. Core Capabilities for App Marketing Teams**
Core Plan includes the essential capabilities needed to measure and optimize app marketing performance:
| Capability | How It Helps Marketing Teams |
| --- | --- |
| Attribution measurement | Connects paid campaigns to installs, trials, and subscription conversions |
| Funnel analysis | Visualizes the full user journey from install → trial → subscription |
| Retention reporting | Identifies which channels bring long-term, high-LTV users |
| Revenue attribution | Measures which campaigns and channels generate actual subscription revenue |
| Active user tracking | Monitors DAU, WAU, and MAU trends by acquisition channel |
These analytics allow growth teams to understand the full conversion journey, rather than relying on surface-level metrics like installs or clicks.
### **3\. Integration With Major Paid Acquisition Channels**
Early-stage app marketing budgets are typically concentrated in a few dominant advertising platforms. Core Plan supports the **four major self-attributing networks (SANs)** that account for 80–90% of paid mobile acquisition spend:
- Meta Ads
- Google Ads
- Apple Search Ads
- TikTok for Business
By focusing on these primary channels, teams can measure performance across the platforms that typically drive the majority of early-stage paid acquisition — without managing dozens of integrations.
### **4\. A Deliberately Focused Feature Set**
Unlike enterprise-level attribution platforms, Core Plan **intentionally prioritizes clarity and simplicity over complexity:**
- **Standard events rather than custom schemas** — simplifying implementation for small teams
- **Maximum two third-party integrations** — e.g., Amplitude + RevenueCat, keeping the stack lightweight
- **Focused attribution reports** — funnel, retention, revenue, and active users
This design ensures that early-stage teams can **quickly access actionable marketing insights without needing a dedicated analytics team.**
## **5 Types of App Marketing Teams That Benefit Most From Airbridge Core Plan**
Airbridge Core Plan is designed for teams that are entering or scaling paid user acquisition, especially those operating subscription-based apps. Below are five common team types where the platform delivers the most immediate value.
### **1\. Small Subscription App Teams Launching Their First Paid Acquisition Campaigns**
If your team is running its first paid acquisition campaigns, you've likely run into a frustrating gap: **installs are increasing, but you have no clear way to know which campaigns are generating actual subscribers.**
At this stage, the marketing team is typically small — often a founder, a growth marketer, and one developer supporting analytics implementation. The challenge isn't launching ads. It's understanding which campaigns generate real subscribers instead of just installs.
**Common pain points:**
- Installs increase but subscription conversion by channel is invisible
- Trial start → subscription conversion cannot be tied to specific campaigns
- Limited engineering resources make complex analytics setups impractical
**How Core Plan helps:**
- **Tracks the full subscription funnel** — install, sign-up, trial start, subscribe, order complete
- **Connects each event to the paid channel that drove it** — no custom event design required
- **Setup in hours, not weeks** — so small teams can start seeing attribution data fast
### **2\. AI SaaS Apps With Trial-to-Subscription Funnels**
AI-powered subscription apps — writing tools, productivity assistants, creative platforms — rely heavily on **trial-to-paid conversion funnels.** For these businesses, installs alone mean very little.
The real risk: without attribution at the trial and subscription level, **teams may be scaling spend on channels that drive high trial volume but low paid conversion.** Every day of delayed signal is a day of wasted budget.**Common pain points:**
- Trial volume looks strong, but which channel actually converts to paid is unclear
- Ad platforms optimize for trial starts, not subscription activations
- Trial-to-paid conversion rate varies by channel but cannot be measured without an MMP
**How Core Plan helps:**
- **Funnel visualization** shows exactly where users drop off between trial and subscription
- **Revenue attribution** connects each subscription to the channel that originally drove the install
- **Retention analysis** reveals which channels bring subscribers who stay — not just those who convert once
### **3\. Performance Marketing Teams Scaling Meta and Google Campaigns**
Once teams begin allocating meaningful budgets to paid acquisition, performance marketers need more than campaign dashboards. **They need cross-channel attribution** — the ability to compare Meta, Google, Apple Search Ads, and TikTok on equal terms.
Without an MMP, each platform reports its own version of performance — and each claims credit for the same conversions. This makes it nearly impossible to reallocate budget with confidence.
**Common pain points:**
- Each ad platform claims credit for the same users — no single source of truth
- Budget allocation decisions are based on platform-reported ROAS, not actual subscription revenue
- High-performing channels stay underfunded because their true revenue contribution is hidden
**How Core Plan helps:**
- **Deduplicates attribution across channels** — one install, one attributed channel
- **Reveals which platforms drive high-value subscribers** — not just high install volume
- **Enables confident budget reallocation** based on actual subscription revenue per channel
### **4\. Product-Led Growth Teams That Need Funnel Visibility**
In many modern app companies, growth is a collaboration between product and marketing. Marketing drives acquisition, while product teams optimize onboarding and retention. **Without shared funnel analytics, it becomes difficult to identify where users drop off — and who's responsible for fixing it.**
Core Plan bridges the gap between paid acquisition and product performance by providing reports that both teams can use to diagnose conversion problems.
| Report Type | Growth Insight | Who Uses It |
| --- | --- | --- |
| Funnel reports | Shows conversion steps from install → trial → subscription | Product + Marketing teams |
| Retention reports | Reveals which channels bring long-term users | Growth teams |
| Revenue reports | Connects campaigns directly to subscription revenue | Marketing + Finance |
| Active user analysis | Tracks DAU, WAU, and MAU trends by channel | Product teams |
This shared visibility allows marketing and product teams to work from the same data — **identifying whether drop-off is a targeting problem (marketing) or an onboarding problem (product).**
### **5\. Startups Preparing to Scale a Dedicated Growth Team**
Many founders worry about choosing tools that will quickly become obsolete as the company grows. **A platform that works for a three-person team today may break when marketing operations expand.**
Airbridge Core Plan is part of a broader product ladder designed to support different stages of marketing maturity. Teams start with Core Plan's focused attribution capabilities, then move to Growth Plan as their analytics needs become more complex.
**The upgrade path looks like this:**
1. **Core Plan:** fast attribution for paid UA → subscription conversion
2. **Growth Plan:** custom events, additional ad network integrations, raw data exports, advanced attribution
This means the decision to start with Core Plan is not a decision to cap your analytics capabilities — **it's a decision to start with what actually matters at your current stage.**
**If your team fits any of the five profiles above, Airbridge Core Plan was built for your stage. Get attribution running in hours, not weeks.**
## **FAQ: Choosing the Right App Marketing Tool for Subscription Apps**
### **1\. Can Airbridge Core Plan track subscription-related events?**
Yes. Core Plan supports a range of standard subscription lifecycle events without requiring custom event schema design:
- Sign-up
- Start Trial
- Subscribe
- Unsubscribe
- Order Complete
These standard events allow teams to track the full subscription funnel immediately after setup — **no analytics engineer required.**
### **2\. When should a company upgrade beyond Core Plan?**
Core Plan is designed for teams at the early-to-mid stage of paid acquisition. As marketing operations mature, teams typically need:
- Custom event tracking beyond standard subscription events
- Support for additional ad networks beyond GMAT
- Raw data exports to data warehouses (e.g., BigQuery, AWS S3)
- Advanced attribution models or fraud detection
At that stage, Airbridge's Growth Plan provides the deeper analytics infrastructure needed to support larger-scale marketing operations — and because it's part of the same product family, the transition is straightforward.
**Connect Your Paid Acquisition to Subscription Revenue**
If your team is launching or scaling paid acquisition for a **subscription app**, Airbridge Core Plan was built for your stage. Get attribution running in hours, understand which channels actually drive subscribers, and scale with confidence.
Visit our waitlist page to see pricing, features, and how to get started.
**👉** [**Join the Airbridge Core Plan**](https://www.airbridge.io/en/plans/core)
---
## Tracking Free Trial to Paid Conversion for Maximizing Subscription LTV
URL: https://www.airbridge.io/en/blog/maximizing-subscription-ltv
Language: English
Published: 2026-03-09
You're spending $5.00 per install. Trials look healthy. But 30 days later – MRR is flat.
For subscription and AI-powered apps, the "Install" has become a vanity metric. If you are spending $5.00 to acquire a user who starts a free trial but never converts to a paid tier, your ROAS isn't just low – it's effectively zero.
The challenge for most growth teams isn't a lack of data; it's the **attribution blind spot** between the ad click and the actual recurring revenue event. When your acquisition data lives in one silo and your App Store revenue lives in another, scaling decisions lack a reliable foundation.
To build a sustainable growth engine, you need to bridge this gap by tracking the entire lifecycle from initial touchpoint to **Lifetime Value (LTV)**.**Key Takeaways**
- **Installs and free trials don't equal revenue.** For subscription and AI apps, the only metric that validates acquisition performance is the free trial → paid subscription conversion.
- **The "Freebie Hunter" problem can distort marketing performance.** High install and trial volumes may hide campaigns that generate users who never convert into paying subscribers.
- **CAC is immediate, but LTV is delayed.** Without proper LTV tracking, marketers are forced to scale campaigns based on incomplete signals like installs or trial starts.
- **Subscription data often lives in silos.** When App Store revenue data is disconnected from attribution data, teams cannot determine which campaigns are actually driving long-term subscribers.
- **LTV tracking connects acquisition with real revenue.** By linking campaign attribution with subscription lifecycle events such as trial start and subscription, marketers can evaluate channels based on true subscriber value.
- **Closing this visibility gap requires an MMP with subscription-specific capabilities** — standard events, native billing platform integration, and attribution across major ad channels.
## **What is Subscription LTV and Why Does Tracking It Matter?**
Lifetime Value (LTV) is the **total revenue a customer generates throughout their relationship with your app** – from the moment they install to their final subscription payment.
For subscription and AI-powered apps, LTV is primarily determined by one critical transition:
**Free Trial → Paid Subscriber**
Everything before that moment is only a signal of intent. Revenue begins only when a user converts into a paying subscriber.
### **How to Calculate LTV**
There are two common approaches mobile marketers use to calculate LTV:
**1) Using CAC**
_LTV = (ARPU x Average User Lifetime) – CAC_
This formula calculates the average revenue a user will generate over the lifetime of their engagement with the app.
**2) Using Churn Rate**
_LTV = ARPU x 1/Churn_
Since 1/Churn approximates average customer lifetime, this formula gives you the same result when churn is your primary known variable.
To evaluate acquisition efficiency, use the **LTV:CAC ratio** separately – a healthy subscription business typically targets 3:1 or higher.
### **Why LTV Tracking is Your Primary Growth Lever**
Consider two acquisition channels:
| Channel | Installs | Trials | Paid Subscribers |
|---------|----------|--------|-----------------|
| A | 10,000 | 3,000 | 120 |
| B | 4,000 | 1,200 | 320 |
[__wf_reserved_inherit]
At first glance, Channel A appears stronger because it generates more installs and trials. But Channel B produces almost three times more paying subscribers.
If marketers optimize based only on installs or trial starts, they risk scaling the wrong channel – increasing spend while reducing long-term revenue. In this example, if Channel B's **LTV:CAC exceeds 3:1**, it may be worth gradually reallocating a portion of Channel A's budget toward Channel B in your next weekly review cycle.
This is why subscription businesses need to measure **LTV at the campaign level**. Tracking the full conversion path allows teams to answer the question that truly matters: _Which acquisition channels generate long-term paying subscribers, not just free trial users?_
## **The "Subscription Silo": Why Many Teams Fail at Scaling**
### **1\. The Disconnect Between App Stores and Attribution**
The fundamental problem lies in the "Black Box" of Apple and Google's payment ecosystems. When a user clicks an ad, installs your app, and eventually starts a trial, the App Store processes that transaction.
However, the App Store does not natively tell your attribution provider which specific ad creative or keyword led to that dollar.
Without a unified bridge, you face a technical gap:
- **The Attribution Gap:** You see 100 installs from "Campaign A."
- **The Revenue Gap:** You see $500 in new subscriptions in your developer console.
- **The Scaling Blind Spot:** You cannot confirm whether that $500 came from "Campaign A" or organic search.
[__wf_reserved_inherit]
This disconnect often leads teams to rely on "**blended ROAS**" – a metric that can obscure channel-level inefficiency and make it harder to identify which campaigns are truly driving subscriber growth.
### **2\. The Cost of the "Freebie Hunter" Problem**
In competitive markets, optimizing campaigns solely for top-of-funnel volume can lead algorithms toward the lowest-cost users. Often, these turn out to be "**Freebie Hunters**" – users who engage with free trials but show low intent to convert to paid subscriptions.**The Revenue Leak Scenario:** Consider a subscription app that splits $10,000 across two channels:
- **Channel A:** 1,000 Installs → 200 Trials → 5 Paid Subs (High Churn)
- **Channel B:** 200 Installs → 50 Trials → 25 Paid Subs (High LTV)
Without **trial-to-paid attribution**, it's easy to over-invest in Channel A based on install volume alone. High install numbers can mask poor conversion quality – and without connecting attribution to the paid conversion event, budget may be directed toward users unlikely to generate meaningful MRR.
## **CAC vs LTV – The Metric Gap That Blocks Growth**
For subscription-based apps, the "**Metric Gap**" is a common growth challenge. This gap exists because your outgoing cash (CAC) is a **real-time certainty**, while your incoming value (LTV) is a **delayed probability**.
### **1\. Real-Time Costs vs. Delayed Value**
Customer Acquisition Cost (CAC) is visible the moment an ad is served. You know exactly what you paid for the click and the install. However, the true Lifetime Value of that user may not materialize for 7, 14, or even 30 days – depending on your trial length and renewal cycle.
For growth teams, this timing gap can create a "**Scaling Paralysis**":
- **The Waiting Cost:** Holding off on scaling until 30-day cohort data is available means delaying decisions that could accelerate growth.
- **The Scaling Risk:** Scaling aggressively based only on "Trial Starts" may direct budget toward users who cancel before the first billing cycle.
To navigate this, teams benefit from **predictive signals** – moving beyond tracking "what happened" toward understanding "what is likely to happen" by connecting early-funnel behavior to down-funnel revenue.
### **2\. Moving Beyond "Standard" Install Tracking**
For many mobile apps, installs are a reasonable success metric. But for subscription-based products, installs are only the first step in a longer revenue journey. A user who installs your app but never subscribes contributes **zero lifetime value**.
This is why subscription growth teams need to track the **entire monetization funnel**, not just acquisition. Once this connection is established, teams can answer the question that determines sustainable scaling: _Which ad channels actually produce long-term subscriber LTV?_
| Aspect | Customer Acquisition Cost (CAC) | Lifetime Value (LTV) |
|--------|--------------------------------|---------------------|
| Definition | Total cost to acquire one new customer | Total revenue expected from one customer over time |
| Calculation | Total acquisition costs ÷ Number of new customers | ARPU × Average Customer Lifetime OR ARPU × (1/Churn Rate) |
| Time Focus | Immediate cost (upfront investment) | Long-term revenue (future returns) |
| Primary Use | Budgeting and acquisition efficiency | Retention strategy and growth planning |
| Optimization Goal | Lower costs while maintaining quality | Extend relationships and increase revenue per customer |
| Typical Benchmark | Varies by industry; often 20–30% of first-year revenue | Ideally 3x higher than CAC for healthy ROI |
### **💡 Pro Tip: Weekly Optimization Checklist**
- **Track "Start Trial" within 7 days post-install** (top-funnel gate – early signal of UA quality)
- **Measure "Subscribe" conversion rate by channel** (true value signal – target 10–20% depending on vertical; below 10% warrants creative or audience review)
- **Calculate channel LTV:CAC weekly using 30-day cohorts** (target ≥3:1; below 2:1 = hidden MRR leak)
- **Monitor Day-30 retention** for churn signals (Day-30 retention below 40% signals Freebie Hunter channel)
- **Cut channels below 2:1 LTV:CAC ratio**; reallocate budget to channels above 3:1> [**See which channels produce subscribers, not just trial users. Standard events, RevenueCat/Adapty S2S. Start with 15K free attributed installs.**](https://coreplan.airbridge.io/waitinglist/a)
[__wf_reserved_inherit]
## **What to Look for in an MMP for Subscription LTV Tracking**
When evaluating an MMP for subscription LTV optimization, confirm it supports:
1. **Full subscription funnel by channel:** Install → Trial → Subscribe breakdown by acquisition source — not just installs by channel.
2. **Native billing platform integration:** S2S connection to RevenueCat or Adapty — so subscription events are captured even when the app is not open.
3. **Predefined subscription events:** Standard events for Start Trial, Subscribe, Renew, and Cancel — without requiring custom event schema design.
4. **Revenue attribution by channel:** Subscription revenue tied back to the ad campaign that drove the install, included in standard reporting.
5. **Cohort retention by acquisition source:** Retention curves by the channel that originally acquired the subscriber, not just aggregate retention.
6. **Usage-based pricing with no annual lock-in:** Pay for what you use, not a fixed annual commitment that stays the same regardless of volume.
## **LTV Tracking Turns Acquisition Into Predictable Revenue**
In the subscription and AI app economy, growth is a function of **unit economics**. Without visibility into the journey from a paid click to a recurring subscription, scaling decisions are based on incomplete data.
By breaking down the "**Subscription Silo**" and connecting attribution data with real-world revenue events, teams gain the clarity needed to make confident, data-backed decisions.
The right MMP offers a fast path to this visibility – providing essential tools to track **standard revenue events**, integrate with billing platforms like RevenueCat or Amplitude, and ultimately maximize LTV.
[**See which channels drive subscribers, not just trial users. Standard events, RevenueCat/Adapty S2S, $0.05/install — 15K free attributed installs on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
**Explore the real successful stories here:**
- [How UNNI Reduced TikTok Retargeting CAC by 51.8% with SKAN and Airbridge](https://www.airbridge.io/en/customers/gangnamunni-skan-titok-retargeting)
- [How Loyal Scaled Day 7 Retention Performance Across 650+ Apps Using Airbridge](https://www.airbridge.io/en/customers/loyal-foundry-airbridge-case-study)
---
## Best App Measurement Tool for Startups: Why Most MMPs Fail
URL: https://www.airbridge.io/en/blog/best-app-marketing-measurement-tool-in-2026-for-startups-why-airbridge-core-plan-works-for-early-growth
Language: English
Published: 2026-03-06
Many startups are spending $10,000–$20,000/month on paid acquisition **without knowing which channel actually generates paying subscribers.**
Installs grow. Trials increase. Campaign dashboards look promising. But without attribution, **CAC calculations become unreliable and budget decisions are based on guesswork — not revenue data.**
In this guide, we'll explore why proper app marketing measurement matters for startups, why most traditional Mobile Measurement Partners (MMPs) aren't built for early-stage teams, and how Airbridge Core Plan helps startups measure installs, trials, and subscriptions without the complexity of enterprise analytics tools.
**Key Takeaways**
- **App marketing without attribution creates growth blind spots:** startups cannot identify which channels actually drive subscriptions and revenue.
- **Scaling paid acquisition without measurement leads to wasted budget:** startups often increase ad spend before knowing which campaigns produce high-value users.
- **Most traditional MMP platforms are designed for enterprise companies:** high base pricing, complex integrations, and feature overload make them difficult for early-stage teams.
- **Startups need simple attribution focused on subscription conversion:** early growth teams primarily need to track the journey from install → trial → subscription.
- **Airbridge Core Plan provides a streamlined measurement layer for startups:** standard events, pre-integrated channels, and native RevenueCat/Adapty S2S — setup in hours, not weeks. Start with 15K free attributed installs.
## **Why App Marketing Without Proper Measurement Fails Startups**
Many startups begin running paid user acquisition before they install a proper measurement layer. The logic is simple: launch ads, acquire users, and hope conversions follow.
But without accurate attribution, **app marketing quickly becomes guesswork.**
You might see installs increasing. You might see trial signups rising. But you won't know which channel, campaign, or ad actually drove the subscription revenue.
### Why Running Paid UA Without Attribution Leads to Guesswork
Imagine a startup launching an AI writing app. The team begins running ads across Meta Ads, Google Ads, and Apple Search Ads.
Within two weeks, they earn 10,000 installs, 2,000 trials, and 350 subscriptions. On the surface, the campaign looks successful. But without [a mobile measurement partner (MMP)](https://abr.ge/iversv), the team cannot answer critical growth questions:
- Which channel generated the highest-quality users?
- Which campaign drove actual subscriptions?
- Which ads created high trial drop-off?
Without attribution, the startup cannot optimize spend. **Budgets shift based on intuition instead of revenue signals** — and this often leads to scaling the wrong channels.
### Revenue Leakage: What Happens When Startups Scale Ads Without Measurement
Consider that same AI writing app team, now ready to spend $20,000/month across Meta, Apple Search, and Google Ads.
Without an MMP, they cannot determine which channel performs better. **The budget gets split evenly — $6,666 to each channel — regardless of actual performance.**
But with attribution in place, the data tells a very different story:
[__wf_reserved_inherit]
[__wf_reserved_inherit]
This insight immediately changes the budget growth strategy:
- **Scale Meta & Google campaigns** — both show strong subscription conversion
- **Reduce Apple Search Ads spend** — 500 installs, only 15 subscriptions
- **Improve Google targeting** — strong trial-to-subscription rate
Without measurement, **that optimization opportunity remains invisible — and budget keeps flowing to underperforming channels.**
## The Startup Problem: Traditional MMP Tools Are Built for Enterprise
Once startups realize they need proper attribution, they usually start researching mobile measurement partners. Typical search queries look like this:
- AppsFlyer vs Adjust for startups
- Free MMP for under 10k MAU
- Best app marketing measurement tool
But very quickly, founders run into a frustrating reality. **Most attribution platforms were built for mature mobile companies running millions in ad spend** — not early-stage teams experimenting with their first user acquisition campaigns.
### Why Most Mobile Measurement Platforms Target Large Apps
Traditional MMP platforms are designed to support:
- Global advertising operations across dozens of networks
- Large analytics teams with dedicated data engineers
- Complex event schemas and custom attribution models
- Raw data exports to enterprise data warehouses
These capabilities are powerful — but they introduce a problem for startups. The tools assume you already have dedicated growth teams, data engineers, and large marketing budgets.
For early-stage apps, that assumption rarely holds. Instead of needing enterprise-level analytics, these teams primarily need one clear signal: "**Are paid users converting into subscriptions — and which channel is driving that revenue?**"
### Three Reasons Startups Struggle with Traditional MMPs
When startups try to implement enterprise measurement platforms, they often encounter three growth blockers.
#### 1\. High base pricing
Enterprise MMPs like AppsFlyer and Adjust often charge **thousands of dollars monthly minimums** before usage costs even begin — plus long-term contracts and overage pricing. For a startup spending $5K–$20K on ads, this pricing structure is often unrealistic.
Airbridge Core Plan uses a **pay-as-you-go model at $0.05/install** — so startups can get full attribution without a large upfront commitment, regardless of current ad spend level.
#### 2\. Implementation complexity
Many MMP platforms require custom event schema design, complex attribution configurations, and advanced analytics setups. For startups without a dedicated analytics engineer, this slows down time-to-value significantly.
What should take hours can take **weeks of integration work.**
[__wf_reserved_inherit]
#### 3\. Feature overload
Enterprise MMP tools include capabilities that startups rarely need early on:
- Multi-touch attribution (MTA)
- Fraud detection layers
- Raw data warehouse exports
- Dozens of ad network integrations
These features are valuable for mature teams. But for early-stage apps, they create more setup work without improving early growth decisions.
### The Startup Attribution Gap
This creates a clear gap in the market. Startups need a measurement solution that is:
- **Fast to implement** — insights in hours, not weeks
- **Focused on subscription conversion** — install → trial → subscribe
- **Compatible with major ad channels** — Meta, Google, Apple, TikTok
- **Affordable for early-stage budgets** — no $2K+ monthly minimums
- **Designed for small growth teams** — no data engineer required> [**Connect paid UA to subscription revenue in hours, not weeks. Standard events, 4 major channels pre-integrated. Start with 15K free attributed installs.**](https://coreplan.airbridge.io/waitinglist/a)
%%CTA|Get Started Free|https://app.airbridge.io/signin?from=blog-coreplan|Start measuring what matters — attribution, deep links, and audiences at no cost.%%
## **Airbridge Core Plan: A Measurement Solution Designed for Startup Growth**
Early-stage apps don't need a full enterprise analytics stack. **What they need first is a clear signal that connects paid acquisition to subscription revenue.**
Airbridge Core Plan was designed around one core question: "**Are paid users converting into subscriptions, and which channels are driving value?**"
Built specifically for subscription and AI apps running their first paid campaigns, Core Plan gives early-stage teams the visibility they need to understand whether user acquisition is actually generating revenue — without the setup complexity of enterprise tools.
### 1\. Core Measurement Capabilities for Subscription Apps
Core Plan provides the essential measurement layer needed to track the subscription funnel.
**Attribution & Tracking**
- Paid UA attribution across major channels
- Install tracking
- Subscription conversion attribution
- Attribution rules configuration
**Event Tracking**
Core Plan supports 30+ standard app events, including the key subscription lifecycle signals your team needs most:
- Install
- Sign-up
- **Start Trial** — Where ad platforms often claim credit
- **Subscribe** — What actually drives MRR
- Unsubscribe
- Order Complete
These standard events allow startups to **track the entire subscription funnel without designing custom event schemas** — one of the biggest time sinks when implementing enterprise MMPs.
### 2\. Native RevenueCat and Adapty Integration
Most subscription apps already use [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) or [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) for billing. Core Plan connects to both via server-to-server integration — subscription events (trial start, subscribe, renewal, cancellation) flow into attribution automatically. **No custom event mapping required.**
### 3\. Built-In Reporting for Early-Stage Growth Teams
Core Plan includes six built-in analytics reports designed for early growth monitoring:
[__wf_reserved_inherit]
### 4\. Coverage Across the Major Ad Channels
Most startups concentrate their early marketing budget on a small set of ad platforms. Core Plan supports the four major self-attributing networks (SANs) that represent 80–90% of early-stage paid acquisition spend:
- **Meta Ads**
- **Google Ads**
- **Apple Search Ads**
- **TikTok For Business**
This allows startups to measure their main marketing activities without additional integrations or complex setup.
### 5\. Why a Simplified Feature Set Helps Startups Move Faster
Core Plan intentionally simplifies features that early-stage teams often don't need yet:
- **No custom events** — 30+ standard events cover the full subscription funnel
- **Maximum 2 third-party integrations** — e.g., Amplitude + RevenueCat
- **No raw data export** — not needed until you have a data warehouse
- **Major SAN channels only** — covering 80–90% of typical startup ad spend
For small teams, this simplified structure **reduces setup time from weeks to hours** and keeps focus on the core goal: understanding which marketing channels generate subscription revenue.
## **What to Look for in an App Marketing Measurement Tool**
Not all MMP tools are built for the same stage of growth. **The right tool for an enterprise company running millions in ad spend is rarely the right tool for an early-stage startup.**
When evaluating attribution platforms, these are the criteria that matter most for early-stage subscription apps:
[__wf_reserved_inherit]
Enterprise MMPs are powerful platforms — but they're optimized for teams that already have data engineers, large budgets, and complex analytics needs. **Core Plan is built for the stage before that:** when a startup needs reliable attribution fast, without the overhead.
## **FAQ: App Marketing Measurement and MMP Tools for Startups**
### 1\. When should a startup upgrade from Core Plan to a full MMP?
Startups typically upgrade when their growth operations become more complex. This usually happens when teams need:
- Custom event tracking beyond standard subscription events
- Multiple ad network integrations beyond GMAT
- Raw data exports to data warehouses (e.g., AWS S3, BigQuery)
- Advanced attribution models (MTA, fraud detection)
- Agency or media partner access
At that stage, a full-scale MMP like Airbridge's Growth Plan provides the deeper analytics infrastructure needed to support larger-scale marketing operations.
### 2\. How quickly can startups start seeing attribution insights with Core Plan?
Once events such as installs, trials, and subscriptions begin flowing into the measurement system, marketers can start analyzing channel performance. **Core Plan is designed for fast time-to-value — setup typically takes hours, not days.**
After tracking is configured and ads are running, teams can answer:
- Which campaigns drive installs
- Which channels produce trials
- Which users actually convert into paying subscribers
## **Start Measuring What Actually Drives Subscription Revenue**
If you're launching paid user acquisition for a subscription app, Airbridge Core Plan was built for your stage. Get attribution set up in hours, connect installs to subscription revenue, and stop scaling channels based on guesswork.
> [**Connect paid UA to subscription revenue in hours. Standard events, RevenueCat/Adapty S2S, $0.05/install — 15K free attributed installs on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## Why Mobile Attribution Numbers Don't Match: Ad Platforms vs MMP vs First-Party Data
URL: https://www.airbridge.io/en/blog/why-mobile-attribution-numbers-dont-match-ad-platforms-vs-mmp-vs-first-party-data
Language: English
Published: 2026-03-06
If you run paid acquisition for a mobile app, you've probably seen this from the same campaign:
- Meta reports 120 conversions
- Google Ads shows 95 installs
- Your MMP attributes 80 users
- Your backend shows 65 actual subscribers
[__wf_reserved_inherit]
That 55-conversion gap isn't just a reporting inconvenience. **It directly distorts your CAC, corrupts your MRR forecasts, and causes budget misallocation across channels.**
When your ad dashboard shows 120 conversions but your billing system shows 65 paying subscribers, you're not just dealing with a data mismatch — you may be scaling spend on channels that aren't actually driving subscription revenue.
In this guide, we'll break down why mobile ad attribution discrepancies happen, how they affect growth decisions, and what performance teams can do to reduce attribution gaps.
**📌 Key Takeaways**
- **Ad platforms, MMPs, and first-party systems measure different things** — which is why attribution numbers rarely match.
- **Self-Reporting Networks (Meta, Google, TikTok) often over-claim conversions** due to view-through attribution and platform-controlled reporting.
- **Deduplication problems** across channels cause the same user to be counted multiple times.
- **SKAdNetwork delays** create timing gaps between real-time dashboards and actual conversion data.
- **Different attribution windows** produce different interpretations of the same user journey.
- **Best practice:** use a measurement hierarchy — first-party data for revenue truth, MMP for attribution, ad platforms for optimization.
## **Why Mobile Ad Attribution Data Discrepancies Happen**
For growth teams responsible for scaling MRR and LTV, data discrepancy creates a dangerous situation: you can't confidently allocate budget if you don't know which channel actually drove the revenue.
Below are the **three primary causes of mobile ad attribution data discrepancy**.
### **1\. Self-Reporting Networks (SRNs) Claim Credit by Design**
Major ad platforms — Meta, Google Ads, Apple Search Ads, and TikTok — operate as Self-Reporting Networks (SRNs), also known as [Self-Attributing Networks (SANs)](https://abr.ge/w65a43).
Unlike traditional ad networks, SRNs do not rely on external tracking links. They determine attribution internally based on their own user data.
This means the platform itself decides whether it contributed to the conversion — which introduces systematic over-attribution.
**Example scenario:**
- A user scrolls Instagram and briefly sees an ad
- Two days later, the same user searches Google for the app
- The user installs and subscribes
**What happens next?**
- Meta claims the conversion (view-through attribution)
- Google claims the conversion (search click attribution)
[__wf_reserved_inherit]
While your real business outcome is 1 subscriber, your ad platform dashboards show 2 conversions. Key mechanics of SRN attribution:
- Platforms use impression-level data unavailable to external tools
- They apply view-through attribution windows
- They prioritize their optimization models over cross-channel accuracy
As a result, platform dashboards are excellent optimization tools — **but unreliable as a source of truth for total performance.**
### **2\. Cross-Channel Deduplication Problems**
Even without SRN behavior, modern user journeys involve multiple marketing touchpoints. Mobile attribution requires **deduplication logic** — a system that decides which channel gets the final credit for a conversion. Without it, every platform counts the same user independently.
Typical multi-touch mobile user journey:
1. User sees a TikTok ad
2. Later clicks a Meta retargeting ad
3. Finally searches the brand on Google
4. Installs the app and starts a trial
If each platform counts independently, each will report 1 conversion — producing **3 reported conversions for 1 actual install and 1 subscription.**
Why deduplication becomes difficult:
- Platforms cannot see each other's data
- Attribution windows differ across networks
- Privacy policies restrict cross-platform tracking
Common signs of a deduplication problem:
- Platform conversions exceed your actual install count
- CAC calculations vary dramatically by source
- ROAS looks profitable on every channel simultaneously
### **3\. SKAdNetwork Creates Attribution Data Delays**
iOS privacy changes introduced another major source of discrepancy: [**SKAdNetwork (SKAN)**](https://developer.apple.com/documentation/storekit/skadnetwork/). Instead of real-time, user-level attribution, [SKAN provides delayed and aggregated signals](https://abr.ge/zxup1l).
Key SKAN characteristics:
- Postback delays (often 24–72 hours)
- Limited conversion values
- Aggregated reporting
- No user-level identifiers
As a result, the same campaign can appear to generate different performance numbers depending on when you check the dashboard. For subscription apps running **trial → paid flows**, the timing mismatch becomes even more complex because subscription events may occur days after install.
Why SKAN widens attribution gaps:
- Delayed reporting changes daily metrics
- Aggregation hides individual user paths
- Conversion value limits reduce event detail
### **4\. Attribution Windows Change Conversion Reporting**
Every measurement system defines a different [time window](https://abr.ge/kf83hg) during which a conversion can be credited to an ad interaction. Because of this, the same user journey can be interpreted differently.
Example:
- User clicks a Meta ad
- Installs the app 6 days later
- Subscribes 7 days later
Possible outcomes:
- Meta may claim the conversion within its attribution window
- Google may not see any attribution
- Your MMP may attribute the install differently depending on configuration
None of these interpretations are technically incorrect. [They simply follow different measurement rules.](https://portent.com/blog/paid-social/breaking-down-social-media-attribution-windows-a-platform-by-platform-comparison.htm)
## **How Attribution Discrepancies Impact Growth and Budget Allocation**
Attribution discrepancies are often treated as a reporting annoyance. In reality, **they are a growth risk that directly affects your ability to scale subscription revenue.**
When acquisition data is inconsistent across platforms, marketing teams lose the ability to answer the most important scaling question: "**Where should we invest the next $100,000 in acquisition budget?**"
### **1\. Budget Misallocation Across Channels**
When ad platforms over-claim conversions, it becomes extremely difficult to evaluate channel performance accurately. Every channel appears profitable — but when deduplicated attribution is applied, the total number of unique conversions is much lower.
Typical outcomes:
- Growth teams scale budgets on channels that appear profitable
- Actual revenue growth lags behind spend
- CAC begins increasing silently
This creates a **hidden revenue leak** where marketing spend grows faster than subscription revenue.
### **2\. Optimization Signal Pollution**
Modern ad platforms rely on machine learning to optimize campaigns. If attribution signals are inconsistent, the optimization loop becomes distorted.
Common signal problems:
- Duplicated conversions across platforms
- Delayed SKAdNetwork signals
- Incomplete event tracking
- Inconsistent attribution windows
Possible campaign outcomes from polluted signals:
- Platforms prioritize low-quality installs
- Retargeting campaigns dominate budgets
- Top-of-funnel discovery campaigns get undervalued
### **3\. MRR Forecasting Errors**
Attribution discrepancies also create problems outside the marketing team. Finance and executive teams rely on accurate attribution signals to forecast growth. But when marketing dashboards and billing data don't align, forecasting becomes unreliable.
Typical symptoms:
- Marketing reports higher conversion numbers
- Finance reports lower subscription revenue
- Leadership questions campaign performance
> [**One source of truth across Meta, Google, Apple Search Ads, and TikTok. Standard events, no custom schema. Start with 15K free attributed installs.**](https://www.airbridge.io/en/plans/core)
## **Ad Platforms vs MMP vs First-Party Data: Which Attribution Data Should You Trust?**
When attribution numbers don't match, teams often ask: "Which dashboard is correct?" The reality is that each system measures something different, so the numbers will rarely align.
### **1\. Ad Platforms (Meta, Google Ads, TikTok)**
Ad platforms measure conversions within their own ecosystems using impressions, logged-in user data, and engagement activity. Because they cannot see external channels, they often claim more conversions than they actually drove. Their data is mainly designed to optimize ad delivery, not provide neutral reporting.
### **2\. MMPs (Mobile Measurement Partners)**
[MMPs](https://abr.ge/66y6g8) try to solve cross-channel attribution by collecting data from ad clicks, installs, and in-app events, then applying rules such as last-touch attribution and deduplication across networks. This creates a more neutral view of performance, though still limited by platform integrations and privacy frameworks like SKAdNetwork.
### **3\. First-Party Data**
First-party systems (billing, backend databases, product analytics) track what actually happens in your product: subscriptions, payments, renewals, and churn. This makes them the most accurate source of revenue truth, though they cannot reliably identify which marketing channel caused the conversion.
### **Why the Numbers Differ**
Each system answers a different question:
- Ad platforms: Which ads influenced users on this platform?
- MMP: Which channel should receive attribution credit?
- First-party data: Did the user actually generate revenue?
Because these perspectives are different, their numbers naturally diverge. Here's how they compare:
[__wf_reserved_inherit]
### **How Growth Teams Use Them Together**
Instead of forcing dashboards to match, advanced teams use a **measurement hierarchy:**
- First-party data → financial truth
- MMP → channel attribution
- Ad platforms → campaign optimization
When used this way, discrepancies stop being a problem and become part of a complete measurement system.
## **How to Reduce Mobile Ad Attribution Discrepancies (Practical Framework)**
Completely eliminating attribution discrepancies is unrealistic. However, modern growth teams can significantly reduce attribution gaps by implementing a structured measurement framework.
[__wf_reserved_inherit]
### **1\. Align Attribution Windows Across Platforms**
One of the easiest discrepancies to fix is attribution window mismatch. If different systems use different windows, conversions will naturally be counted differently. Recommended approaches:
- Align click attribution windows across platforms
- Limit view-through attribution where possible
- Ensure MMP windows match major channel settings
### **2\. Define a Clear Source of Truth**
One of the biggest operational mistakes is allowing multiple dashboards to compete as the primary reporting source. Instead, growth teams should define data hierarchy rules:
- First-party data → financial truth
- MMP attribution → marketing performance comparison
- Ad platform dashboards → campaign optimization
This ensures that MRR and subscription revenue remain the ultimate performance metric.
### **3\. Separate Optimization Data From Reporting Data**
Ad platforms require rich conversion signals to train their algorithms, but those signals are not always suitable for neutral performance analysis. Growth teams often solve this by splitting events into two layers:
- Optimization signals: install, sign-up, trial start
- Revenue evaluation signals: subscription purchase, renewal, revenue value
### **4\. Validate SKAdNetwork Conversion Logic**
For apps with significant iOS traffic, SKAdNetwork configuration is critical. Growth teams should review:
- Conversion value mapping
- Event prioritization
- Postback interpretation logic
- Campaign ID structure
When designing SKAN conversion values: prioritize high-value funnel events, avoid mapping too many low-impact events, and align conversion windows with subscription behavior.
### **5\. Cross-Check Attribution With Subscription Events**
Attribution performance must always be validated against actual revenue events: trial start, payment method added, subscription activation, renewal.
Routinely compare these three datasets:
- Ad Platforms: Reported conversions
- MMP: Attributed installs or trials
- First-party system: Paid subscriptions
If attribution performance looks strong but subscription events remain weak, the issue is usually low-quality traffic, misconfigured conversion tracking, or inflated platform attribution.
## **Understanding the Subscription Attribution Funnel**
For subscription apps, the distinction between optimization events and revenue events is especially important. Many UA teams make the mistake of optimizing for the wrong moment in the funnel.
The subscription funnel typically looks like this:
1. **Install**
2. **Sign-Up**
3. **Trial Start** _← Ad platforms typically optimize here_
4. **Payment Added**
5. **Subscribe** _← Your business cares most about this_
6. **Renewal** _← Long-term MRR health_
When assessing attribution discrepancies for subscription apps, always validate that your MMP is tracking the full funnel — not just the top of it. **The gap between trial starts (what ad platforms see) and actual subscription activations (what your billing system sees) is often where the biggest MRR leakage occurs.**
## **Why Modern Growth Teams Go Beyond Attribution Dashboards**
Attribution dashboards are helpful, but they often show only one layer of the growth picture. As marketing stacks grow more complex, teams typically rely on multiple tools:
- Ad platform dashboards for campaign optimization
- MMP dashboards for cross-channel attribution
- Product analytics for user behavior
- Billing systems for subscription revenue
Each tool answers a different question — and none of them alone connects the entire acquisition-to-revenue journey.
That's why more mature growth teams build a measurement layer that connects marketing signals with product events. **The goal isn't to make all dashboards agree. It's to know which number to trust for which decision.**
## **Connecting Paid UA to Subscription Revenue with Airbridge Core Plan**
Most MMP solutions are built for mature growth teams with dedicated data engineers — complex event schemas, custom integrations, and weeks of setup. But early-stage subscription apps don't need that complexity.
They need one answer, fast:
> **_"Are paid users converting into subscriptions, and which channels are driving those conversions?"_**
**Airbridge Core Plan** was designed specifically for small UA teams running subscription-based mobile apps in the US market. It provides fast, reliable attribution signals without requiring complex infrastructure or event schema design.
Core Plan supports attribution across the four major Self-Reporting Networks used by most subscription apps:
- Meta Ads
- Google Ads
- Apple Search Ads
- TikTok for Business
In addition to installs, Core Plan tracks the standard subscription funnel events that matter most: **sign-up, trial start, subscribe, and order completion** — giving you a direct line of sight from paid acquisition to actual subscription revenue.
[__wf_reserved_inherit]
## **Ready to connect your paid UA spend to actual subscription revenue?**
Airbridge Core Plan is built for US-based subscription apps that need attribution signals without the complexity. Visit our waitlist page to see pricing, feature details, and how to get started.
> [**Stop reconciling dashboards. Attribution across Meta, Google, Apple Search Ads, and TikTok — standard events, $0.05/install, 15K free attributed installs on Airbridge Core Plan.**](https://www.airbridge.io/en/plans/core)
---
## The 5 Best Branch Alternatives for 2026: A Performance Marketer’s Guide
URL: https://www.airbridge.io/en/blog/the-5-best-branch-alternatives-for-2026-a-performance-marketers-guide
Language: English
Published: 2026-03-06
Branch dominates as a mobile measurement partner (MMP), but rising costs and limited raw data access push performance marketers toward alternatives.
If you're tracking deep links and attribution for subscription apps or AI services, this guide compares Branch's strengths against competitors like Airbridge—without sacrificing accuracy.
**📌Key Takeaways**
- **The Shift in Intent:** Most performance teams are migrating toward alternatives to gain better raw data access and transparent multi-touch attribution (MTA) logic.
- **Privacy-First Maturity:** Modern alternatives are built with SKAN 4.0 and Privacy Sandbox as defaults rather than retrofitted add-ons, ensuring higher attribution accuracy in a signal-limited environment.
- **Fraud & Transparency:** The best alternatives offer configurable, rule-based fraud prevention that allows teams to see exactly why a click was flagged as invalid.
- **Cost Efficiency at Scale:** Newer platforms align costs with business value and measurement depth, often reducing total cost of ownership (TCO) by 20–30%.
- **The Airbridge Edge:** For teams that need Branch-level linking reliability but require unrestricted data ownership and deep subscription analytics, Airbridge serves as the high-performance middle ground.
## What Is Branch as an MMP?
Branch is a mobile measurement partner best known for deep linking, deferred deep links, and cross-platform user routing, with attribution capabilities layered on top. Most teams adopt Branch first to solve _linking_ problems, then extend usage into measurement.
## Why are Marketers Switching from Branch to Other MMPs?
As mentioned, Branch is well known for its deeplink capabilities. However, as marketing needs evolve beyond deeplink alone, many teams start questioning whether it fully supports their broader performance and attribution goals.
Branch's pain points center on high costs, opaque pricing, and dashboards that restrict raw data exports. Performance marketers often hit limits on multi-touch attribution and fraud detection compared to its competitors — and begin exploring Branch alternatives.
### 1) Common Pain Points Teams Report
From a tactical standpoint, most friction clusters around **measurement depth**, not linking reliability. Teams evaluating Branch alternatives often cite:
1. **Strong deep linking, weaker attribution**
- Limited support for advanced or customizable attribution models
- Less control over multi-touch or assist logic
2. **Fraud and traffic quality gaps**
- Fewer configurable rules compared to attribution-first MMPs
- Harder to isolate suspicious installs or post-install behavior
3. **Dashboard-led analysis**
- Proprietary reporting layers can restrict raw data exploration
- BI and warehouse workflows require workarounds
4. **Cost vs. value at scale**
- Usage-based pricing tied to links, events, and traffic volume
- Marginal costs rise faster than insight gained
### 2) Market Trends Driving Re-Evaluation
[Market trends](https://swaarm.com/blog/top-trends-mobile-measurement-partners-mmp-2025/) show MMPs shifting toward privacy-first measurement post-ATT, with **subscription apps** demanding cheaper tools that balance deep linking and full analytics.
Branch's proprietary setup locks teams into vendor-specific views, frustrating devs needing SQL-level access
#### (1) Privacy-first attribution is now the baseline
ATT, SKAdNetwork, and Android privacy changes force teams to:
- Model conversions with partial signals
- Validate postbacks across networks
- Reconcile multiple attribution sources
This favors **attribution-first MMPs** designed for probabilistic and aggregate logic, not just deterministic links.
#### (2) Subscription and AI apps need longer measurement windows
Trials don’t convert instantly. Revenue often lands days or weeks later. That reality requires:
- Cohort-based LTV tracking
- Flexible conversion definitions
- Reliable re-attribution and re-engagement logic
Deep linking helps users land correctly. It doesn’t explain _why revenue moved_.
#### (3) Data teams want ownership, not summaries
A growing number of orgs push attribution data into:
- Data warehouses
- Internal BI tools
- Experimentation frameworks
Tools optimized around closed dashboards create friction here.
## Deep Linking vs. Full Attribution: What’s the Difference for ROAS?
But often, when marketers start exploring alternatives, confusion arises around the difference between deep linking and full attribution. Many teams assume they solve the same problem. Before diving deeper into feature comparisons, it is important to clearly understand how these two capabilities differ and how each impacts ROAS.
Deep linking delivers users directly to specific in-app content via universal links, skipping generic app store flows. Full attribution tracks the entire user journey, assigning credit across touchpoints like installs, events, and purchases for accurate ROAS.
Typical deep linking use cases:
- Web → app routing
- Email or QR to in-app content
- Campaign-specific landing experiences
- Deferred deep links for new installs
Attribution systems typically handle:
- Install and re-attribution logic
- Post-install event measurement
- Conversion windows and lookback rules
- Network postbacks and SKAN reporting
- Fraud and invalid traffic detection
### 1) Why Teams Confuse the Two
A common mistake marketers make is buying a deep-link-first product and expecting it to behave like an attribution-first system under pressure.
C-level marketers buying Branch often mix these up—deep linking shines for retention (e.g., subscription renewals), but full attribution reveals true campaign performance amid iOS privacy shifts.
In practice, apps need both: deep links drive 15-25% higher conversions as [Promodo said](https://www.promodo.com/blog/the-relationship-between-user-experience-ux-design-and-conversion-rates), while attribution spots fraud or multi-touch waste. Many brands prioritize deep linking alone, [leaving attribution gaps that inflate CPA by 30%](https://yodelmobile.com/choosing-mobile-attribution-platform-for-your-app/).
| Area | Deep Linking-First Tool | Full Attribution MMP |
| --- | --- | --- |
| Core focus | User routing | Measurement accuracy |
| Install logic | Basic / deterministic | Deterministic + modeled |
| Post-install analysis | Limited | Cohort, LTV, ROAS |
| SKAdNetwork support | Surface-level | Privacy-first by design |
| Fraud protection | Minimal | Rule-based, configurable |
| Data access | Dashboard-centric | Raw data + exports |
| Best for | Navigation & UX | Growth, spend optimization |
*Comparison: Deep Linking Tool vs Full Attribution MMP*
## The Best Branch Alternatives in 2026: AppsFlyer, Adjust, and Airbridge
Top Branch alternatives include AppsFlyer, Adjust, Airbridge, and Kochava—each balancing deep linking, attribution, and cost for performance marketers. AppsFlyer leads in fraud detection but charges premium rates; Adjust offers strong multi-touch at mid-tier pricing; Airbridge prioritizes raw data access without vendor lock-in and reduced data discrepancy.
| Dimension | Branch | AppsFlyer | Adjust | Singular | Airbridge |
| --- | --- | --- | --- | --- | --- |
| Core Strength | Reliable deep & deferred links | Attribution accuracy at scale | Media optimization | Cross-source reporting | Built for the Subscription Economy |
| Deep Linking Quality | Very strong | Strong | Strong | Moderate | Strong |
| Deferred Deep Linking | Native, mature | Supported | Supported | Limited | Native, stable |
| Attribution Model Flexibility | Limited | High (customizable) | High | Medium (depends on partners) | High (customizable) |
| Fraud Detection | Basic | Strong (rule-based) | Basic | Basic | Strong (rule-based) |
| Cohort & LTV Analysis | Limited | Strong | Strong | Strong | Strong |
| Raw Data Export | Restricted | Full | Full | Full | Full |
*Comparison Table: Branch vs Leading Alternatives*
## How Airbridge Solves the Legacy Gaps in Branch Attribution
[Airbridge](https://abr.ge/van8ny) is a mobile measurement partner built to balance deep linking reliability with attribution depth—without forcing teams to choose between the two.
It’s designed for companies that have outgrown routing-only tools but don’t want enterprise complexity.
> **Airbridge in One Sentence**
>
> Airbridge replaces Branch’s deep linking capabilities while upgrading attribution, analytics, and data ownership for privacy-first growth teams.
### 1) Where Airbridge Matches Branch (No Downgrade)
Teams considering a switch often worry about losing what already works. Airbridge covers the same core ground. From a functional standpoint, Airbridge provides:
- **deep linking:** Stable routing for web-to-app and app-to-app flows
- **Link management:** Campaign-level links without fragmentation
- **Cross-platform consistency:** iOS, Android, web journeys handled in one system
_If your baseline requirement is “links must not break,” Airbridge clears that bar._
### 2) Where Airbridge Goes Further
The difference shows up after the click.
#### **(1) Attribution built for privacy reality**
Airbridge is designed with ATT and SKAdNetwork as defaults, not add-ons. That means:
- Native SKAN reporting and validation
- Modeled attribution designed for signal loss
- Consistent logic across paid, owned, and organic channels
#### **(2) Measurement that matches subscription economics**
Subscription and AI apps don’t win on installs. They win on retention and revenue timing. Airbridge supports:
- Cohort-based LTV analysis
- Flexible conversion windows
- Re-attribution and re-engagement tracking tied to revenue events
_This is where deep linking alone stops being enough._
#### **(3) 3. Fraud protection without black boxes**
Instead of opaque “trust scores,” Airbridge emphasizes **configurable, transparent rules**. Teams can:
- Define what invalid traffic looks like for their app
- Validate installs and post-install events
- Reduce wasted spend without hiding logic
#### **(4) Real data ownership**
A frequent Branch complaint is dashboard dependency.
Airbridge takes a different stance:
- Full raw data exports
- Warehouse- and BI-friendly structure
- Dashboards for speed, not lock-in
_When numbers are questioned internally, raw data matters more than charts._
### 3) Cost and Scaling: A Practical Advantage
Many teams begin their search with _“What is Branch cost?”_ The deeper issue is predictability. Airbridge is positioned to:
- Avoid feature paywalls
- Reduce surprise costs as traffic and events scale
- Align pricing with measurement value, not link volume
## Conclusion: Choosing a Branch Alternative That Actually Scales
Branch works well when the core problem is **getting users to the right place in the app**. That strength doesn’t disappear—but for many scaling teams, it stops being sufficient.
As spend grows and privacy limits visibility, teams need more than routing. They need attribution they can trust, analytics that reflect delayed revenue, and access to raw data when numbers are questioned. That’s why teams look for Branch alternatives in the first place.
Airbridge fits when you want:
- Deep linking without regression
- Attribution built for ATT and SKAdNetwork
- Cohort and revenue analysis for subscription or AI products
- Clear data ownership, not dashboard lock-in
If Branch helped you solve early growth, **Airbridge is designed for the next stage—where confidence in measurement matters as much as activation.**
See how big names leveraged Airbridge for better attribution, measurement, and equally excellent deep linking solutions.
### **Airbridge Case Studies**
- [How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://abr.ge/ja2lzd)
- [Glints Scaled to 1.75M+ MAU and 12.6M+ Job Applications With Airbridge DeepLinks](https://abr.ge/2ezx5t)
- [How Dabang cut deep link management time by 90% with Airbridge](https://abr.ge/v0fx0j)
[Let's contact Airbridge now](https://abr.ge/9vqiax). No cost, no commitment, no risk—just a conversation to evaluate whether a better-fit MMP can support your next stage of growth. So why not?
---
## Analyze Apple Search Ads Placement Performance in Airbridge
URL: https://www.airbridge.io/en/blog/apple-search-ads-placement-report
Language: English
Published: 2026-03-03
Running Apple Search Ads but still checking the Apple Ads dashboard to understand placement-level performance?
Until recently, most MMPs didn’t provide placement-level data for Apple Search Ads. Marketers had to log into the Apple Ads dashboard and manually compare performance.
**Not anymore.**
Airbridge is the **first MMP to support performance analysis across all four Apple Search Ads placements**, so you can analyze detailed performance directly in your Airbridge dashboard.
**Key takeaway**
- **Industry-first:** Analyze performance across all four Apple Search Ads placements in an MMP.
- **No more dashboard switching:** View detailed placement performance directly in Airbridge.
## **What changed in Apple Search Ads measurement?**
Apple Search Ads is a core acquisition channel that shows ads when users explore the App Store.
On **January 28, 2026**, Apple introduced a new field that allows MMPs to identify where ads were shown in the App Store.
Airbridge quickly adopted this update and now provides **placement-level Apple Search Ads performance directly in the dashboard.**
## **Analyze Apple Search Ads performance by placement**
[__wf_reserved_inherit]
Previously, Apple Search Ads performance was aggregated into a single channel. Marketers couldn’t tell where conversions actually came from.
Now in Airbridge, you can apply the **\[Sub Sub Publisher 2\]** grouping in reports to analyze the following four placements separately:
- **Search Results (search\_results)
**Ads shown on the results page after a user searches for a keyword.
- **Search Tab (search\_tab)
**Ads shown in the recommended app list before a search query is entered.
- **Today Tab (today\_tab)
**Ads displayed on the App Store’s main landing page.
- **Product Pages (product\_pages)
**Ads shown at the bottom of another app’s product page.
## Why this matters for marketers
### Identify what actually drives performance
Instead of seeing overall Apple Search Ads performance, you can now identify which placements drive results.
For example:
- **Today Tab** may drive brand awareness.
- **Search Results** often deliver stronger conversion rates.
Placement-level data makes it easier to understand performance drivers and optimize campaigns.
### Optimize budget with better visibility
You no longer need to jump between dashboards to compare data.
Inside Airbridge, you can quickly identify underperforming placements and allocate budget to the ones that deliver better results.
**As of February 2026, Airbridge is the only MMP that supports Apple Search Ads placement-level analysis.**
## Start analyzing Apple Search Ads placements in Airbridge
Airbridge quickly adopts updates to Apple Ads data specifications so marketers can access the latest measurement capabilities first.
With Airbridge, you can manage **all ad performance data—including Apple Search Ads placement performance—in one place.**
Check detailed Apple Search Ads placement data in your dashboard today.
---
## Deferred Deeplink vs Standard Deep Link: What’s the Difference and Why Standard Isn't Enough in 2026?
URL: https://www.airbridge.io/en/blog/deferred-deeplink-vs-standard-deep-link
Language: English
Published: 2026-02-26
Performance marketers know the frustration: a high-intent user clicks an ad, but the journey breaks because they don’t have the app installed.
If you’re relying solely on a deeplink strategy that doesn't account for the "install gap," you are likely burning your acquisition budget.
**Understanding the nuance between a deferred deeplink vs standard deeplink** is the difference between a bounced user and a high-value conversion.
**📌Key Takeaway**
- **Standard deep links only work when the app is already installed.** They break the moment an install is required.
- **Deferred deep links preserve user intent even if the app isn’t installed.** The intended destination is restored after install and first open.
- **In real acquisition flows, install sits between click and conversion.** Losing context at this stage leads directly to drop-off.
- **Standard deep links fail in paid ads, web-to-app, referrals, and influencer traffic.** Exactly where most new-user growth comes from.
- **Deferred deep linking is no longer an upgrade, it’s the baseline.** The requirement to turn installs into meaningful actions.
## **What Is a Deeplink?**
**A deeplink is a URL that routes users directly to a specific page or state inside a mobile app**, instead of opening the app’s home screen or a mobile website. It’s used to shorten the path between intent and action.
Unlike web links or app store URLs, deeplinks use custom URL schemes (e.g., myapp://product/123) to trigger in-app navigation. They shine in campaigns – think ads, emails, or push notifications – bring users directly to specific in-app locations like a referral reward or personalized offer.
Learn more about [Deep Linking Basics That Marketers and Developers Need to Know here](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers).
## **Deferred deeplink vs Standard deeplink: Side-by-Side Comparison**
To choose the right tool for your funnel, you must understand how these two technologies handle the user's "state" (whether the app is installed or not).
### **1) Standard Deep Links (The Ideal State)**
**Standard deep linking** (often utilizing Universal Links on iOS or App Links on Android) **works only if the user already has the app on their phone.** When clicked, the link checks for the app; if found, it opens the specific content immediately.
They rely on immediate URI handling – no install step. Pros: Fast, preserves all data.
**But if the user does not have the app, a standard deep link typically fails.** It either shows an error message or, at best, redirects the user to a mobile web version of the page. It cannot "wait" for the user to go to the App Store, download the app, and then find the original content.
### **2) Deferred Deep Links (The Real-World Solution)**
**Deferred deep links capture the user’s intent before install and restore it after the app is opened for the first time.**
The flow looks like this:
1. User clicks a link (ad, web, email)
2. App is not installed
3. User is sent to the app store
4. After install, the app opens **at the original destination**
This is how acquisition actually works in the wild.
[__wf_reserved_inherit]
Deferred deep linking accounts for:
- New users coming from paid ads
- Platform interruptions (app store, privacy prompts)
- Multi-step journeys across web → store → app
A common mistake is treating deferred deep links as a “nice-to-have.” In reality, they’re the difference between _measuring installs_ and _driving meaningful post-install actions_.
| Feature | Standard Deep Link | Deferred Deep Link |
| --- | --- | --- |
| **App installed** | Opens the app to a specific in-app page | Opens the app to the intended in-app page |
| **App not installed** | Fails or redirects to app store without context | Routes user to app store, then restores destination after install |
| **Post-install experience** | No context preserved | Original intent fully retained |
| **Paid acquisition support** | Limited | Designed for paid channels |
| **User journey continuity** | Fragile | Seamless |
| **Conversion impact** | High drop-off risk | Higher install-to-action rate |
## **Why "Standard" Isn't Enough for Performance Marketing in 2026**
In 2026, [the cost of user acquisition has skyrocketed](https://adjoe.io/blog/mobile-app-user-acquisition-strategy/), with CPMs increasing faster than average revenue. Relying solely on standard deep links for new user acquisition is essentially paying for clicks that are designed to fail when acquisition paths include installs, privacy gates, and multiple platforms.
Here is why "Standard" Isn't Enough for Performance Marketing in 2026:
### **1) The Acquisition Funnel Is No Longer Linear**
In practice, a single “click → open app” flow rarely exists.
A typical journey now looks like this:
- Ad click on mobile web
- Redirect through tracking or privacy layers
- App store visit
- Install confirmation
- First app open
- Permission prompts
- Finally, content
Standard deeplinks don’t survive this sequence. They weren’t designed to.
Once the user is sent to the app store, the link context is gone. By the time the app opens, the system has no idea _why_ the user came.
### **2) Install ≠ Conversion**
A common reporting mistake is treating installs as success.
What actually matters:
- Completed signup
- Viewed product page
- Triggered purchase
- Activated key feature
Without deferred deep linking, users land on the default home screen after install. From there, they must re-discover what motivated them in the first place.
Drop-off here is invisible in many dashboards – but very real in revenue.
### **3) Privacy and Platform Changes Make It Worse**
OS-level privacy updates and browser restrictions have made attribution and routing more fragile.
Examples:
- App Tracking Transparency interruptions
- Browser-to-app handoff inconsistencies
- Webview and in-app browser limitations
Standard deeplinks rely heavily on OS behavior staying predictable. That’s no longer a safe bet.
Deferred deep linking adds a persistence layer that absorbs these disruptions instead of breaking under them.
## **When Should You Use Deferred Deep Linking?**
Use deferred deep linking **whenever your campaign targets new installs or mixes installed/non-installed users.** It bridges the install gap, preserving context like campaign IDs or product SKUs for post-install routing.
App publishers lean on it for UA ads, where standard links fail 70% of the time. A common mistake: skipping it for email flows – yet even re-engagement benefits if users reinstall.
### **Checklist: When Deferred Beats Standard**
- Running CPI/CPA ads with install-first flows.
- Cross-platform campaigns (store redirects common).
- Personalized onboarding.
- High churn funnels needing D1 retention lifts.
- Attribution across MMPs amid ATT/WebView limits.
## **Airbridge DeepLink: Maximum Conversion by Seamless User Journey**
[**Airbridge**](https://www.airbridge.io/deeplink-plan) **provides an enterprise-grade but start-up budget deep linking solution** that eliminates the friction between a click and a conversion.
While standard tools provide basic routing, Airbridge ensures that your attribution data and user experience remain intact across the entire "Install Gap".
[**Airbridge DeepLink**](https://www.airbridge.io/deeplink-plan) **supports:**
- **Web-to-app:** Convert web traffic to app users, seamlessly landing them on specific pages within the app with web-to-app CTA buttons like “Open in app”
- **CRM-to-app:** Embed DeepLinks in your email, SMS, and push notifications, and direct users straight to the specific app page.
- **Offline-to-app:** Whether it’s TV ads, brochures, billboard banners, event signage, or product packaging, QR codes with Airbridge deep links will send users directly to targeted content within the app.
- **Social-to-app:** Users are sent directly to in-app content after installation – not the home page.
- **Influencer & Affiliates:** Distribute short links and QR codes to each influencer so that you can analyze which performs the best
- **Referral:** Scale your referral campaigns with API-generated DeepLinks, directing users to targeted signup flows or exclusive in-app offers.
More importantly, it’s built to remain stable despite:
- OS updates
- Privacy restrictions
- Multi-touch attribution models
## **Outro.**
Standard deep links worked in simpler times. But with fragmented funnels, privacy hurdles, and sky-high acquisition costs, they leave too much revenue on the table.
Deferred deep links fix this by preserving intent across the install gap – turning more installs into signups, purchases, and activations. Performance marketers who ignore this baseline risk watching competitors pull ahead.
In practice, teams using deferred deep links see 20-40% lifts in D1 retention and post-install conversions. A common mistake? Sticking to basic tools that crumble under real-world pressure.
Ready to stop losing high-intent users? Airbridge DeepLink makes seamless journeys effortless – handling web-to-app, CRM, offline QR codes, and more, all attribution-ready.
👉Read to perfect every link and scale every campaign?
[Talk to us!](https://www.airbridge.io/book-a-demo)
👉 Explore how real teams are scaling deeplinking with Airbridge:
[How Dabang cut deep link management time by 90% with Airbridge](https://www.airbridge.io/en/customers/airbridge-dabang-deeplink-optimization)
[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://www.airbridge.io/en/customers/fizz-migrates-from-firebase-dynamic-links-to-airbridge-deep-link)
---
## Deferred Deeplink Benefits for App Installs
URL: https://www.airbridge.io/en/blog/deferred-deeplink-benefits
Language: English
Published: 2026-02-26
Author: Harper (Trang Nguyen)
Paid user acquisition is expensive. If you send a user to the App Store and they land on a generic home screen after install, **you’ve likely lost them**.
**Understanding deferred deeplink benefits is the first step toward fixing this broken funnel.** By maintaining context from the initial ad click through the installation process, deferred deep links ensure your users find exactly what they were promised, **significantly reducing churn.**
This article breaks down how deferred deep linking works, why it impacts growth metrics, and how teams use it to improve conversion from paid installs.
**📌Key Takeaways**
- **Deferred deep linking preserves intent through install.** It ensures users land on the exact in-app content promised in the ad, not a generic home screen.
- **It fixes the biggest paid UA leak.** By maintaining context from click → install → first open, deferred deep links reduce early drop-off and “ghost installs.”
- **Faster time-to-value drives better retention.** Routing users directly to high-intent experiences shortens activation paths and improves Day 7 and Day 30 retention.
- **Paid acquisition becomes more efficient.** Campaigns optimize for engagement and conversion quality (CPE), not just installs (CPI), improving ROI.
- **Critical for subscription and AI apps.** From paywalled previews to pre-filled AI workflows, deferred deep links turn installs into immediate, value-driven sessions.
## **What is Deferred Deeplink?**
**Deferred deeplink is a feature that allows mobile apps to direct users to specific content within an app**, even if the user doesn't have the app downloaded.
A deferred deeplink saves the in-app destination data in the server until the user finishes the app download. It differs from standard deep links, which only work on already-installed apps, by "deferring" the action until post-install.
When a user clicks on a deferred deeplink but does not have the app, they will be directed to the App Store to download it. Once the download is complete, and the user launches the app, they will then be sent to the final in-app destination.
[__wf_reserved_inherit]
### **Why is deferred deeplink important?**
App marketers can reach users across more channels than ever: websites, social platforms, SMS, and even offline touchpoints like QR codes. Reach, however, is only half the job. What matters is what happens after the click.
**Every extra step between a campaign and the in-app destination introduces drop-off**. Users get distracted. They hesitate. They leave. small delays or unclear navigation can materially reduce conversion rates.
**User acquisition is even more complex.** Prospective users don’t have the app installed, so a standard deep link breaks the journey. This is where deferred deep linking becomes essential.
Deferred deeplinks route new users through the app store first. After installation, the app opens to the original intended destination, not a generic home screen. **The context is preserved, the experience feels continuous, and the user lands exactly where they expected to be.**
## **How does deferred deeplink work?**
A Deferred deep link saves the in-app destination data in the server until the user finishes the app download. When the download is complete, and the user opens the app, the deferred deep link then matches the data with the user ID to take the user to the app page.
Here's the step-by-step flow:
1. **Click:** A user clicks an ad or promotional link (e.g., ‘10% off discount for monthly subscriptions’).
2. **Intent Capture:** The attribution provider (like Airbridge) captures the user's device parameters or a unique click ID.
3. **App Store Redirect**: Since the app isn't installed, the user is sent to the official store.
4. **The Install & First Open:** App SDK pings server on launch.
5. **Match & Redirect:** Server links install to click, delivers deep link data – straight to product page or offer, skipping home screen.
This whole process seems complicated, but it is quite similar to the logic that is used in an MMP. **With an MMP like** [**Airbridge**](https://www.airbridge.io/product/deep-linking)**, marketers can set step-by-step details of deferred deep linking scenarios with ease.**
## **Deferred Deeplink Benefits That Directly Impact Growth Metrics**
As app ecosystems become more competitive and AI-powered apps raise user expectations, the margin for post-install friction keeps shrinking. Growth teams can no longer rely on installs alone. What matters is how quickly users reach value after install.
Below are the deferred deep linking benefits that performance marketers actively optimize for in 2026.
### **1) Higher Install-to-First-Open Rates**
**A common mistake marketers make is ignoring the "Ghost Install" phenomenon** – where a user downloads an app but forgets why they did so by the time the icon appears on their home screen.
**Deferred deep linking solves this by providing an immediate "Reason to Stay”**.
By serving the exact content promised in the ad, you eliminate the cognitive load that causes users to bounce before the first session even begins.
### **2) Better Post-Install Activation**
Activation happens the moment a user realizes the app's value. [In 2026, activation benchmarks are rising](https://www.businessofapps.com/data/app-activation-rates/), **especially for subscription and AI-powered apps**, where users expect immediate, tangible value right after install.
When a campaign promotes an “AI Photo Filter,” sending users through a generic tutorial creates friction. Instead, the deferred deeplink should bypass onboarding and drop the user directly into the camera interface with that filter pre-selected.
This is where deferred deep linking plays a critical role. By routing new users into pre-configured, high-intent experiences – like saved AI prompts, unlocked previews, or contextual onboarding paths – it shortens the path to the “Aha!” moment.
The faster users experience value, the more likely they are to stay. In practice, **shorter activation paths are strongly correlated with higher Day 7 and Day 30 retention.**
Learn more about how [Playio optimized the user journey toward the ‘aha moment’, achieving a 30% D30 retention rate for ad-acquired users here.](https://www.airbridge.io/en/customers/playio-d30-retention-cpa-optimization)
### **3) Improved Conversion from Paid Installs**
Paid install conversion measures how efficiently ad spend turns into revenue or qualified users.
Here's where deferred deep linking transforms your unit economics: users who click promotional content are already demonstrating high intent and are "likely to be of higher quality" than broad-based acquisition efforts.
By preserving that intent through the install process, you're optimizing for engagement quality, not just download quantity.
This translates directly to better ROI on paid acquisition campaigns. Instead of measuring success purely on cost-per-install (CPI), you can focus on [cost-per-engagement (CPE)](https://www.airbridge.io/glossary/cost-per-engagement-cpe) – a metric that actually correlates with lifetime value and revenue generation.
| Metric | Without Deferred DL | With Deferred DL | Improvement |
| --- | --- | --- | --- |
| Install-to-Open | Baseline | +20-30% | Seamless routing |
| Day-30 Retention | 1x | 2.5x (Source: [Adapty](#)) | Personalized activation |
| Paid Conversion | Baseline | +30-40%+ (Source: [scn.st](#)) | Direct to offers |
## **Common Use Cases for Deferred Deeplinks in Subscription & AI Apps**
### **1) Paywalled Content Previews**
**Use case:** User clicks an ad for premium content but doesn’t have the app installed.
With deferred deep linking:
- User installs the app
- Lands directly on the locked article, feature, or tool
- Sees the value before being asked to subscribe
Without it, users hit a generic home screen and often fail to understand what they installed the app for.
**This approach consistently improves trial starts and subscription intent.**
### **2) Personalized Subscription Offers**
**Use case:** Campaigns promoting discounts, free trials, or plan-specific messaging.
Deferred deep links allow marketers to:
- Route users to a campaign-specific paywall
- Pre-select plans or pricing tiers
- Match messaging from ad to in-app offer
Mismatch between ad copy and paywall content is a silent conversion killer. Deferred deep linking removes that inconsistency.
### **3) AI App Templates and Pre-Filled Workflows**
**Use case:** AI tools promoting specific outputs (e.g., “Generate a resume,” “Summarize a PDF”).
With deferred deep linking:
- Users install the app
- Open directly into a pre-filled prompt or workflow
- See output immediately or after minimal input
This shortens time-to-value dramatically, which is critical for AI apps competing on perceived intelligence and speed.
### **4) Re-Engagement After Install Delays**
**Use case:** Users install but don’t open immediately.
Deferred deeplinks can be reused in:
- Install reminder emails
- Push notifications
- Retargeting ads
Because the destination remains intact, users return to the exact context they originally wanted, not a generic restart point.
[__wf_reserved_inherit]
## **Airbridge DeepLink: Built for 2026 Performance Marketers**
As we move through 2026, the complexity of the "Privacy-First" era means standard attribution often loses the thread.
While deferred deep linking delivers clear benefits, implementation complexity often becomes the bottleneck. Performance marketers need solutions that integrate seamlessly with existing attribution and analytics workflows while providing the granular measurement capabilities required for optimization.
[**Airbridge DeepLink**](https://www.airbridge.io/pricing) addresses these challenges with privacy-first attribution that connects deferred deep link performance to broader campaign analytics.
It’s designed not just to create links, but to own the entire deeplinking workflow.
### **Features of Airbridge deeplink**
- **Unified deeplink management:** Manage deeplinks in one place instead of spreading link logic across campaigns, channels, and tools.
- **Intent-preserving routing across user states:** Ensure users are routed to the correct in-app destination regardless of whether the app is installed or not.
- **Consistent behavior across channels and touchpoints:** Apply the same deeplink logic across paid media, CRM, web-to-app journeys, and offline entry points.
- **Operational scalability without manual overhead:** Support high volumes of campaigns and links without relying on custom engineering work for each use case.
- **Reliable measurement tied to attribution:** Connect deeplink interactions with install, conversion, and post-install events to maintain data continuity.
## **Outro.**
Deferred deep links aren't optional – they're the fix for post-install drop-off. They boost install-to-open rates by 20-30%, lift D30 retention 2.5x, and sharpen paid conversion by 30-40% by landing users on promised content, not home screens.
In practice, subscription and AI apps thrive with them: previews before paywalls, pre-filled workflows, personalized offers. Skip this, and your CPI budget funds ghost installs.
Airbridge DeepLink Plan scales it all – unified management across channels, privacy-safe attribution, no engineering headaches.
Start Your Free Trial of Airbridge DeepLink and turn installs into revenue today.
👉 Explore how real teams are scaling deeplinking with Airbridge:
[How Dabang cut deep link management time by 90% with Airbridge](https://www.airbridge.io/en/customers/airbridge-dabang-deeplink-optimization)
[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://www.airbridge.io/en/customers/fizz-migrates-from-firebase-dynamic-links-to-airbridge-deep-link)
---
## The Airbridge DeepLink Plan: Why You Don’t Need an Enterprise Budget for Deep Linking in 2026
URL: https://www.airbridge.io/en/blog/enterprise-budget-for-deep-linking
Language: English
Published: 2026-02-23
Google's decision to [shut down Firebase Dynamic Links on August 25th, 2025](https://firebase.google.com/support/dynamic-links-faq) has sent thousands of app developers and founders looking for alternatives.
The problem? Most replacements bundle full MMP suites: attribution, fraud detection, analytics – at enterprise prices. **You pay $10K+/month for features you skip. Not every team needs that.** Sometimes, affordable deep linking is enough to drive users straight to in-app content, boosting retention without bloat.
While competitors bundle everything together and charge accordingly, **Airbridge took a different path:** give developers exactly what they need, when they need it, at a price that makes sense.
## **What Is Deep linking and Why Pricing Matters Now?**
**A Deep Link is a specific URL that bypasses a mobile website or app store to send users directly to a designated location inside an app.**
Deeplinks are used to bring users directly to specific in-app locations, saving users the time and energy of seeing an ad and then locating a particular page themselves – significantly improving the user experience by reducing friction.
**But deep linking itself isn’t complex anymore. Pricing is. It even hits harder in 2026.**
With the sunsetting of free tools like Firebase Dynamic Links, many providers are capitalizing on the urgency by forcing developers into expensive, "all-in-one" attribution contracts.
**App installs cost $2-5 each now.** User acquisition budgets shrink as iOS privacy rules tighten. Teams need deep linking to maximize every install. Yet many solutions bundle it into bloated "enterprise" plans and charge thousands dollars monthly for "basic" deep linking. Startups and mid-tier teams get locked out.
[__wf_reserved_inherit]
## **The "All-in-One" Trap Most MMPs Won't Talk About**
**A common mistake growth teams make is assuming bundled pricing will “pay off later.”** What usually happens instead:
- Deeplinks go live, but attribution features remain unused
- Engineering time is spent integrating systems that aren’t critical
- Budget is consumed before performance is validated
Here's what the mobile measurement industry doesn't want you to know: most app developers and founders don't need attribution tracking, fraud detection, and advanced analytics when they're just starting out. **They need deeplinks that work reliably without breaking the bank.**
Some solutions have built their business models around forcing you into comprehensive plans that include features you may never use. R**eality? You buy the full suite, use 20%, pay for 100%.**
Need deep linking for your referral program? That'll be **$500+ per month for the "starter" plan** that includes attribution tracking, audience building, and a dozen other features collecting dust in your dashboard.
**The** [**Airbridge DeepLink Plan**](https://www.airbridge.io/deeplink-plan) **flips this model on its head.** Instead of bundling everything together, we created a **focused solution for developers who know exactly what they need**: reliable deep linking with advanced capabilities.
## **What You Get with the Airbridge DeepLink Plan (Without Enterprise Pricing)**
The [Airbridge DeepLink Plan](https://www.airbridge.io/deeplink-plan) is designed to provide infrastructure for apps that have outgrown basic tools but aren't ready for a full-scale attribution contract. Instead of locking features behind "Enterprise-only" tiers, this plan focuses on the technical essentials required to maintain a seamless user experience. Below is what makes Airbridge's approach different:
### **Free Usage That Matters**
Most "free" deep linking solutions come with catches – limited clicks, restricted features, or aggressive upgrade prompts after 30 days.
Airbridge's deeplink plan is **free below 10,000 Monthly Active Users (MAU)**. That's not a 14-day trial or a limited feature set – it's full access to their entire deep linking platform.
For context, 10K MAU means your app can have 10,000 different users opening it each month before you pay anything. For most early-stage apps and startups, that's months or even years of free usage.
### **Pay-As-You-Grow Pricing That Actually Makes Sense**
Once you exceed 10K MAU, you're not suddenly hit with a massive monthly fee. The pricing scales with your growth, and crucially, you're charged based on actual app usage – **not vanity metrics like link clicks or the number of deep links you create.**
This matters because other platforms often charge per click or per link generated, which can rack up costs quickly if you're running referral campaigns or have an active user base sharing links. With Airbridge, a viral campaign that generates thousands of clicks doesn't suddenly inflate your bill.
### **Enterprise Features Included by Default**
Despite being positioned as an affordable alternative, the **Airbridge deeplink plan** doesn't skimp on advanced capabilities:
- **Deferred Deep Linking**: Users who don't have your app installed can click a link, go through the App Store installation process, and still land on the intended page when they first open your app. This single feature can dramatically improve conversion rates for referral and marketing campaigns.****
- **DeepLink Generation API**: Perfect for building referral systems where users can dynamically generate links to specific content or actions within your app. No need to pre-create links or manage them manually.****
- **Bulk Link Generator**: Use Google Sheets to generate hundreds of deeplinks at once. Ideal for e-commerce apps that need links for every product or content apps creating campaign-specific links.****
- **Real-Time Analytics**: Track clicks, installs, and deeplink opens through your links with detailed reporting that actually helps you optimize campaigns.
## **Migrating from Firebase Dynamic Links? Here’s the Easiest Path**
Migrating from Firebase Dynamic Links doesn’t require re-architecting your app or rebuilding every link from scratch.
Airbridge provides direct replacement functionality for all core Firebase Dynamic Links use cases, including custom URL schemes, Universal Links (iOS), and App Links (Android).
**In practice, the Airbridge SDK automatically selects the optimal deeplink method based on the user’s device and OS.** This means existing users experience the same seamless routing behavior: install, open, or re-engagement – without added friction or edge-case handling.
### **No-code migration, not an engineering project**
One of the biggest migration risks teams underestimate is operational overhead. Recreating hundreds or thousands of Firebase links manually slows teams down and increases the chance of errors.
To avoid this, **Airbridge offers a** **No-Code Deepink Migration toolset** that allows teams to:
- Import existing Firebase Dynamic Links in bulk
- Automatically convert them into Airbridge tracking links
- Preserve routing logic and parameters without manual rewriting
Learn more about [No-Code DeeplLink Migration From Firebase to Airbridge here](https://www.airbridge.io/en/blog/how-to-migrate-deeplink-in-airbridge).
[__wf_reserved_inherit]
## **Join Over 1,000 Apps: Start Deep Linking in Minutes, Not Sales Meetings**
Over 1,000 apps, from indie hits to scale-ups, run Airbridge DeepLink daily. No lengthy demos. Sign up, generate links, deploy. And live in minutes. Here's where Airbridge really shines compared to enterprise solutions:
### **Built for teams that move fast**
From an execution standpoint, **the Airbridge DeepLink Plan removes the usual friction points:**
- No mandatory sales calls to get started
- No enterprise contracts blocking setup
- No feature gating for core deep link functionality
Most teams can create their first deeplink, test routing, and ship to production in minutes.
### **Deep linking without organizational drag**
A common bottleneck in app teams isn’t engineering complexity. It’s internal approval cycles tied to expensive tooling.
By separating deep linking from enterprise MMP bundles, Airbridge allows:
- Growth teams to move independently
- Developers to integrate once, not repeatedly
- Product teams to validate impact before scaling spend
This structure is especially effective for startups, scale-ups, and teams migrating from Firebase Dynamic Links under real deadlines.
### **Designed to scale when you do**
Starting small doesn’t mean hitting limits later.
As traffic grows, deeplink usage scales seamlessly without forcing a platform switch or renegotiation. The same infrastructure that supports early-stage apps also supports high-volume production traffic.
That consistency is what lets teams focus on performance, not tooling decisions.
Learn more about [**5 types of teams that need Airbridge DeepLinks here**](https://www.airbridge.io/en/blog/5-team-who-need-airbridge-deeplink-plan)**.**
[__wf_reserved_inherit]
## **Airbridge's Deeplink Plan FAQs**
### **Q1. How does Airbridge's pricing compare to Firebase Dynamic Links?**
We offer the same core functionality as Firebase Dynamic Links, **plus advanced capabilities.** This includes deferred deep linking for iOS, custom domain settings, a programmatic API for link creation, and web-to-app functionality (with our Web SDK). These powerful features are provided **free under 10K MAU** and then scale affordably based on your app's actual usage, not arbitrary limits.
### **Q2. Can I upgrade to the full MMP later if I need attribution tracking?**
**Yes, seamlessly.** Airbridge's Growth Plan includes all deep linking features plus full attribution capabilities. Since you're already using their platform, upgrading doesn't require new SDK integration or link migration – just talk with our team and we can make upon your signing the up tier.
### **Q3. What happens to my existing Firebase Dynamic Links when I migrate?**
**Airbridge provides migration tools and maintains functionality.** Their platform supports the same link formats and behaviors as Firebase Dynamic Links, so your existing links continue working. They also provide step-by-step migration guides and video tutorials specifically for Firebase users.
### **Q4. Is there a limit on how many deeplinks I can create?**
**No artificial limits on link creation.** Unlike platforms that charge per link or impose creation limits, **Airbridge allows unlimited deeplink generation.** You pay based on Monthly Active Users, not how actively you use the platform's features.
### **Q5. Do I need technical expertise to implement Airbridge's deep linking?**
**Minimal technical setup required.** Airbridge provides SDKs for all major platforms with clear documentation. Most implementations take under an hour, and their self-service model means you can test everything before committing to paid usage. Please check our [SDK documentation](https://help.airbridge.io/en/deeplink-developers/sdk-quickstart) for more details.
### **Q6. Can I use my own domain for branded links?**
**Yes, custom domain support is included.** You can use your own branded domain instead of the default abr.ge links, which improves click-through rates and maintains brand consistency across your campaigns.
### **Q7. What kind of analytics and reporting do I get with the DeepLink Plan?**
**Real-time reporting with actionable insights.** The DeepLink Plan includes Actuals Report and Trend Report, giving you click tracking, conversion data, and user journey analysis to optimize your campaigns without paying for attribution features you don't need.
## **Ready to Start Deep Linking the Smart Way?**
With over 1,000 apps already benefiting from this approach, the question isn't whether Airbridge's focused strategy works – it's whether you're ready to stop overpaying for features you don't need.
Want to see how Airbridge can transform your mobile marketing measurement? [Check pricing](https://www.airbridge.io/pricing) or [request a demo](https://www.airbridge.io/book-a-demo) to see accurate attribution and deep linking in action.
Start with their free tier and experience the difference of a platform that grows with your needs instead of overwhelming them.
👉 Explore how real teams are scaling deep linking with Airbridge:
[How Dabang cut deeplink management time by 90% with Airbridge](https://www.airbridge.io/en/customers/airbridge-dabang-deeplink-optimization)
[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://www.airbridge.io/en/customers/fizz-migrates-from-firebase-dynamic-links-to-airbridge-deep-link)
---
## Why Most Apps Use Deeplinks Wrong: 3 High-Impact Moments
URL: https://www.airbridge.io/en/blog/where-and-when-to-use-deeplinks
Language: English
Published: 2026-02-23
Most marketers treat the "**click**" as the finish line, but in mobile growth, it’s just the starting blocks. If your mobile marketing deeplink strategy consists of simply driving traffic to your app’s home screen, you are effectively throwing away over half of your acquisition budget.
High-performing campaigns rely on continuity. When a user clicks an ad for a specific pair of sneakers, they expect to see those sneakers – not a generic welcome banner. Bridging this "**context gap**" is the difference between a loyal customer and a bounced session.
This blog explores **the critical role of deeplinking** **in reducing friction and increasing conversions**, from understanding basic mechanics to **deploying advanced strategies** across UA, CRM, and retargeting channels.
## **What Is a Deeplink in Mobile Marketing?**
**A deeplink is a specific URL that bypasses a mobile website or app store to send users directly to a designated location inside an app.**
Below is an example of a deeplink in mobile marketing:
[__wf_reserved_inherit]
Unlike standard web links, deeplinks use unique protocols to ensure users land on specific product pages, articles, or promotional offers, significantly reducing journey friction.
In practice, a deeplink acts as a digital bridge. While a standard URL points to a domain (e.g., website.com), **a deeplink points to a specific piece of content within the app's architecture.**
From a tactical standpoint, there are **three primary types of links** that every marketer must distinguish between to avoid technical debt:
- **Basic Deeplinks**: These only work if the app is already installed. If it isn't, the user usually hits a "Page Not Found" error.
- **Deferred Deeplinks:** These are the "gold standard" for UA. If the user doesn't have the app, they are sent to the App Store; once installed, the link "remembers" where they were going and takes them to the specific page.
- **Universal Links (iOS) / App Links (Android):** OS-level standards that provide a seamless transition between web and app without annoying "Open in App?" pop-ups.
Learn more about [how different types of deeplinks work here.](https://www.airbridge.io/en/blog/what-is-deep-link)
**A common mistake marketers make is assuming these links "just work" out of the box.** Without a centralized management system, these links often break during OS updates or browser changes, leading to high drop-off rates seen in unoptimized campaigns.
## **Why Are Deeplinks Critical for App Growth Today?**
From a growth perspective, every extra step between a click and value creates drop-off. Deeplinks remove those steps. Here is why they are non-negotiable for modern growth:
### **Shortened Path from Click to Value**
Every extra screen a user must navigate is an opportunity for them to quit. **In a standard flow, a user journey might be like these 3 steps:**
1. See an ad
2. Go to the App Store, and open the app, and be dumped onto the home screen.
3. Then, they have to search for the product they just saw.
Deeplinks, specifically deferred deeplinks, remove these steps, taking the user from "Install" straight to "Checkout."
[__wf_reserved_inherit]
### **Reduced Friction and Cognitive Load**
A homepage forces users to think. Where should I go next? What was I looking for?
That pause is friction. Multiple sources show [21–72% of users abandon during onboarding when it’s frictiony.](https://www.setgreet.com/blog/what-the-numbers-actually-say-about-mobile-app-onboarding-\(and-what-to-track\))
**Deeplinks eliminate decision-making by placing users exactly where they expect to be.** From a behavioral standpoint, fewer choices mean fewer exits. This is especially critical on mobile, where attention is limited and interruptions are common.
### **Improved User Experience Across Install and Non-Install States**
Modern deeplinks handle both scenarios:
- If the app is installed, users go straight to the content.
- If not, they install first and land on the same destination.
This continuity matters. Breaking the journey during install is one of the fastest ways to lose users before they ever activate. Deeplinks keep the experience intact from first click to first action.
## **How Deeplinks Reduce Drop-Off and Increase Conversions**
Every unnecessary step between intent and fulfillment multiplies your drop-off rate. Traditional app onboarding requires users to:
- land on homepage → navigate to the right category → search for the promised item → remember why they came → complete their action.
That's at least four opportunities for users to get distracted, confused, or simply give up.
Frictionless onboarding through strategic deeplinking eliminates these unnecessary steps entirely. In fact, [reducing the number of onboarding steps by 30% can **increase completion rates by up to 50%**](https://www.getmonetizely.com/articles/understanding-onboarding-completion-rate-a-critical-metric-for-saas-success)**.**
- Result: **2-3x higher conversions** and better retention. To understand the ROI, compare the two typical user journeys:
Generic Web Link
Smart Deeplink
New User Path
Click → Store → Home Screen → **Search**
Click → Store → **Specific Product Page**
Existing User Path
Click → Mobile Web Browser → Login
Click → **App Opens Instantly**
User Friction
High (User must navigate manually)
**Zero (Automated routing)**
Avg. Conversion
Baseline (1x)
**2.5x - 3x Higher**
Attribution
Often lost at the App Store
**Preserved across the install**
The compound effect is significant. **Better initial experiences lead to higher retention rates, increased lifetime value, and more positive app store reviews.**
Users who successfully complete their first intended action are more likely to return, engage deeper, and become brand advocates.
**It's not just about saving one conversion – it's about optimizing the entire user lifecycle.**
## **Deeplink Best Practices: Where and When to Deploy**
### **Where to Use Deeplinks in Mobile Marketing Campaigns**
Deeplinks aren’t something you “set once and forget.” They only work when they’re used in the right places, at the right moments, with a clear job to do.
In mobile marketing, the highest returns come from deploying deeplinks where user intent is already clear and the next step should feel obvious:
- **User Acquisition (UA) & Social Media:** Use Deferred Deeplinks for Instagram, TikTok, or YouTube ads. This ensures that even if a user doesn't have your app, they are routed to the store and then immediately to the specific product they saw in the ad.
- **Retargeting & Re-engagement:** Target "Cart Abandoners" via programmatic display. These links should bypass the home screen and land the user directly back into their shopping bag.
- **CRM Channels (Email, Push, SMS):** This is the most common use case. A push notification about a "Flash Sale" should never just open the app; it should link to the specific sale category.
- **Organic & Owned Channels:** QR codes on physical packaging or "Refer-a-Friend" links shared via WhatsApp should be deeplinked to credit the referral and show the reward immediately upon app open
### **When to Use Deeplinks (And When Not To)**
**Use Deeplinks When:**
- **Action-Oriented:** Campaign success depends on completing a specific in-app action. If your goal is getting users to make a purchase, sign up for a service, or complete onboarding, deeplinks are non-negotiable. Generic app entry points create too much friction between intent and action.
- **Contextual:** You are running a personalized promotion where the landing page content must match the ad copy.
- **Conversion-Focused:** You are optimizing for conversion, not just installs. Vanity metrics like install numbers mean nothing if users don't engage meaningfully with your app. If your KPIs focus on revenue, retention, or engagement depth, deeplinks are essential for hitting those targets.
- **Lifecycle-Driven:** You run retargeting or lifecycle campaigns. These campaigns target users with established interest or previous app experience. They expect personalized, relevant experiences that acknowledge their history with your brand.
**Avoid or Rethink When:**
- **Deeplinks are unmanaged or inconsistent.** Broken deeplinks are worse than no deeplinks. If you can't guarantee that users will land where you promise, default to standard app links and invest in proper implementation before trying again.
- **You lack fallback handling for non-installed users.** Deeplinks to non-installed users should gracefully handle the installation process and then redirect to the intended content. Without proper fallback mechanisms, you'll lose users during the install process.
- **Measurement and attribution are not connected.** If you can't measure the performance difference between deeplinked and non-deeplinked campaigns, you can't optimize effectively. Proper attribution setup is a prerequisite for strategic deeplink deployment.
## **Why Managing Deeplinks at Scale Is Hard Without the Right Tool**
### **The Complexity Crisis of Deeplinks in Today App Marketing**
Deeplinks start simple. One campaign. One destination. One team. Then growth happens.
Suddenly, deeplinks exist across paid ads, CRM, referral programs, QR codes, influencer campaigns, and organic channels.
Each one may point to a different in-app screen, behave differently for installed vs non-installed users, and rely on separate tracking logic. Marketers often underestimate the technical maintenance required for links to stay "healthy."
At **Airbridge**, we see growth teams succumb to **three primary “growth-killers”** when they lack a centralized management system:
- **Broken Redirection Logic:** OS updates (like iOS 14.5+ or the recent Android changes) frequently break traditional URL schemes. If your links don't have a dynamic fallback, users land on a 404 page or a dead-end web view.****
- **The "Walled Garden" Problem:** Platforms like Instagram, Facebook, and WeChat often try to keep users within their own in-app browsers. Standard links fail here. You need specialized "Web-to-App" logic to break out of these browsers and into your app.****
- **Attribution Silos:** If your deeplink tool doesn't talk to your measurement tool, you might know a user arrived in the app, but you won't know which specific ad campaign drove that high-value purchase.
At scale, unmanaged deeplinks don’t just underperform – they create blind spots. Partnering with a specialized provider like [Airbridge](https://www.airbridge.io/deeplink-plan) can take a lot of that weight off your shoulders.
### **A Scalable Approach to Deeplinking: Introducing Airbridge DeepLink Plan**
When deeplinks become mission-critical across campaigns, managing them with spreadsheets and piecemeal tools simply doesn’t scale. **That’s where the** [**Airbridge DeepLink Plan**](https://www.airbridge.io/pricing) **steps in.**
It’s designed not just to create links, but to own the entire deeplinking workflow:
- **Unified deeplink management:** Manage deeplinks in one place instead of spreading link logic across campaigns, channels, and tools.
- **Intent-preserving routing across user states:** Ensure users are routed to the correct in-app destination regardless of whether the app is installed or not.
- **Consistent behavior across channels and touchpoints:** Apply the same deeplink logic across paid media, CRM, web-to-app journeys, and offline entry points.
- **Operational scalability without manual overhead:** Support high volumes of campaigns and links without relying on custom engineering work for each use case.
- **Reliable measurement tied to attribution:** Connect deeplink interactions with install, conversion, and post-install events to maintain data continuity.
Ready to perfect every link and scale every campaign?
👉Talk to us!
👉 Explore how real teams are scaling deeplinking with Airbridge:
[How Dabang cut deep link management time by 90% with Airbridge](https://www.airbridge.io/en/customers/airbridge-dabang-deeplink-optimization)
[How Fizz achieves ZERO performance loss with Airbridge API deeplinks](https://www.airbridge.io/en/customers/fizz-migrates-from-firebase-dynamic-links-to-airbridge-deep-link)
---
## Appsflyer vs Airbridge vs Adjust vs Branch: Best MMP for Subscription app in 2026
URL: https://www.airbridge.io/en/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026
Language: English
Published: 2026-02-23
In 2026, selecting the right MMP for a subscription app is about more than installs and CPI. Growth teams want to know how AppsFlyer, Airbridge, Adjust, and Branch truly differ, which platform best supports user acquisition, retention, and revenue analysis, how pricing scales, and how each handles privacy-first attribution under ATT and SKAdNetwork.
This comparison breaks down attribution accuracy, cost structure, fraud protection, and web-to-app measurement, with a focus on helping teams find a single source of truth for installs, in-app events, and subscription revenue. If you’re evaluating the best MMP for subscription apps in 2026, this guide is built to help you make a confident, data-driven decision.
## Key Takeaways
- Subscription apps need an MMP that connects installs to trials, renewals, retention, and LTV, not just clicks and CPI
- Traditional attribution models break for subscriptions because revenue is delayed, recurring, and cohort-driven
- Appsflyer and Adjust excel at large-scale UA, Branch at web-to-app linking, but they add friction for subscription revenue analysis
- Airbridge is built subscription-first, treating trials, conversions, renewals, and churn as core attribution signals
- Under ATT and SKAdNetwork, the real differentiator is how well an MMP maps privacy-safe signals to revenue outcomes
- Teams are replacing legacy MMPs with Airbridge to gain one source of truth across acquisition, retention, and revenue without dashboard sprawl or feature upsell.
## Why Subscription Apps Need a Suitable MMP (Mobile Measurement Partner)
An **MMP (Mobile Measurement Partner)** **is a system that tracks where app installs and post-install events** come from, then connects them to revenue, retention, and campaign performance across channels.
For subscription apps, this goes beyond installs, clicks, and CPI. A modern MMP should show which channels drive trials, which campaigns convert to paid users over time, and where LTV truly outperforms cost. In this model, installs are just the starting point—retention, renewals, and revenue are the metrics that define growth.
### Why subscription apps break traditional attribution models
Maybe you already heard of some MMPs that worked very well for industries like gaming or e-commerce. But do they actually work for subscription app as well?
The reason the attribution models used for gaming or e-commerce apps don’t work for subscription apps is that:
- Conversion happens **days or weeks after install**
- Revenue is **recurring**, not one-off
- Web-to-app journeys are common (landing page → app → trial)
- Campaign ROI depends on **churn and renewal**, not day-0 revenue
While many legacy MMP setups still optimize around last-click installs, short attribution windows, or event counts without revenue context, that only works for UA volume. It fails for subscription growth. From a tactical standpoint, subscription teams need:
- Flexible attribution windows
- Accurate subscription event mapping
- Cohort-level LTV and retention visibility
### 3 Common mistakes teams make when choosing an MMP
Even being well aware of this, people are still struggling to choose a suitable MMP. After working with scaling subscription apps, 3 patterns and mistakes show up repeatedly:
#### 1\. Choosing based on market dominance, not fit
Big-name MMPs are often selected by default. Teams later discover they’re paying for features built for gaming-scale UA, not subscription analytics.
#### 2\. Optimizing for UA only
Install attribution looks clean, but renewal, churn, and revenue attribution stay fragmented across tools.
#### 3\. Accepting dashboard sprawl as “normal”
Web analytics in one place. SKAN reports in another. Subscription data somewhere else. Decision-making slows down, and confidence drops.
> At this stage, the question shifts from _“Which MMP tracks installs best?”_ to
> “**Which MMP gives you one source of truth for growth?**”
## Appsflyer vs Airbridge vs Adjust vs Branch: Features, Pricing, and Fit Compared
So now you understand why an MMP matters. The next step is choosing one—and that’s where things get confusing. Once you start researching MMP solutions, you quickly run into dozens
**There are Big 4 global MMPs in the US: Appsflyer, Airbridge, Adjust, and Branch all solve attribution.** But you may ask a question like “Which MMP is the best for us?”
The difference lies in **data depth, pricing structure, and how well they support subscription economics.**
Below is a practical comparison based on how teams actually use these tools, not how they’re marketed.
| Category | Appsflyer | Airbridge | Adjust | Branch |
| --- | --- | --- | --- | --- |
| **Core Strength** | Large UA-scale attribution | Subscription-focused attribution & analytics | Enterprise-grade UA tracking | Deep linking & web-to-app |
| **Best Fit** | High-volume install campaigns | Subscription & AI apps | Global enterprise apps | Content-heavy & web-first apps |
| **Subscription Revenue Tracking** | Available, often add-on based | Native & flexible | Supported, complex setup | Limited depth |
| **Web-to-App Attribution** | Supported via add-ons | Built-in, unified | Supported | Strong, but web-centric |
| **SKAdNetwork Reporting** | Mature, but segmented | Clean, actionable, unified | Robust, heavy UI | Partial focus |
| **Fraud Protection** | Advanced (premium tiers) | Included by default | Strong enterprise controls | Basic |
| **Dashboard Complexity** | High | Low to moderate | High | Moderate |
| **Pricing Transparency** | Tiered, event-based | Predictable, usage-aligned | Contract-heavy | Variable |
| **Ideal for Scaling Subscriptions** | Medium | High | Medium | Low–Medium |
Let’s dive in to see how they differ.
### Attribution accuracy & data granularity
All four platforms claim accurate attribution. The real difference shows up **after day 7**.
- **Appsflyer & Adjust
**Strong at install-level attribution and large network coverage. Granularity often comes with added cost or configuration overhead.
- **Branch
**Excellent for linking web, email, and content journeys. Less reliable for deep subscription revenue modeling.
- **Airbridge
**Designed to connect **install → trial → subscription → renewal** without stitching multiple reports together.
From an execution standpoint, Airbridge reduces the gap between **raw attribution data and revenue decisions**.
### Subscription & revenue analytics
**This is where subscription apps feel friction first.**
With most traditional MMPs, revenue technically exists in the system, but it rarely tells a complete story. Subscription events are logged, yet they’re not cleanly tied back to cohorts. Renewal data gets pushed into BI tools because that’s where “real analysis” still happens. Trial-to-paid conversion is often reviewed outside the MMP entirely.
[**Airbridge**](https://www.airbridge.io/solutions/subscription) **takes a different approach. Subscription events are treated as first-class data, not secondary signals.** Revenue, retention, and LTV sit next to UA metrics, so teams can see acquisition quality without exporting or reconciling dashboards. Core questions—what converts, what renews, what churns—are answered inside the MMP.
### Pricing models and hidden costs
Pricing is rarely discussed openly, but it drives replacement decisions.
**Appsflyer / Adjust**
- Event-based or volume-based pricing
- Advanced features often gated behind higher tiers
- Costs scale fast with subscription event volume
**Branch**
- Competitive entry pricing
- Costs increase as attribution complexity grows
**Airbridge**
- More predictable pricing aligned with actual usage
- Fewer “must-upgrade” moments as teams scale
- Lower friction when adding web, SKAN, or subscription events
For CFOs and growth leaders, predictability matters as much as raw capability.
### Integration ecosystem & workflow fit
All four platforms integrate with major ad networks and analytics tools, so coverage looks similar. The difference shows up in how teams are expected to work with the data.
Appsflyer and Adjust are built around the assumption that attribution feeds multiple downstream systems. They work well as data providers, but analysis and decision-making often happen elsewhere. Branch leans heavily into linking and routing journeys, with analytics playing a supporting role.
[**Airbridge**](https://www.airbridge.io/solutions/partner-network) **is designed differently. It aims to function as the central operating layer, not just a source that exports data to other tools.** Attribution, subscription metrics, and performance insights live in one place, which reduces handoffs and reporting gaps.
## SKAdNetwork, ATT, and iOS Attribution: Which MMP Performs Best in 2026
Under ATT and SKAdNetwork (SKAN), an MMP’s role shifts from user-level tracking to **privacy-safe aggregation, modeling, and interpretation**—without breaking decision-making.
ATT removed deterministic user tracking on iOS. SKAN replaced it with aggregated, delayed install data, limited conversion value windows, and no user-level identifiers
In theory, all MMPs support SKAN. However, the experience varies.
### SKAdNetwork reporting depth by platform
**Appsflyer**
- Strong SKAN coverage and documentation
- Data often split across multiple views
- Requires experienced teams to interpret correctly
**Adjust**
- Enterprise-grade SKAN pipelines
- Powerful but UI-heavy
- Best suited for large teams with dedicated analysts
**Branch**
- SKAN support exists
- Less optimized for subscription conversion modeling
- Focus remains on linking rather than revenue insight
**Airbridge**
- SKAN data presented alongside non-SKAN metrics
- Conversion values mapped to subscription events
- Less manual reconciliation between dashboards
### Where data loss still happens — and who minimizes it best
SKAN doesn’t fail evenly. Most data loss happens long before reports are reviewed.
The usual breakdown points are familiar: conversion values are poorly designed, postback windows are inflexible, and subscription events are disconnected from attribution logic. Once these choices are locked in, teams lose visibility where it matters most—after the install.
Legacy setups tend to reinforce the problem. Conversion values stay install-focused. Event priorities remain static. Iteration slows because every change feels expensive or risky.
[**Airbridge approaches SKAN differently.**](https://www.airbridge.io/product/ios-skan) Conversion value schemas can be updated faster. Subscription events are mapped with intent, not as afterthoughts. SKAN signals are aligned to revenue outcomes, not just install confirmation.
### Fraud prevention in a privacy-first world
**Appsflyer & Adjust**
- Mature fraud detection stacks
- Advanced protections often sit behind premium plans
**Branch**
- Basic protections, not fraud-first
**Airbridge**
- Fraud protection included by default
- Focus on invalid traffic and attribution anomalies
- Practical coverage without enterprise overhead
For subscription apps, protecting early trial quality is often more important than blocking raw install fraud.
## Which MMP Is Right for Your Growth Stage? A Practical Decision Guide
The right MMP depends less on feature lists and more on where your subscription business is today—and how fast it needs to adapt. Let’s check this practical decision guideline according to your growth stage.
### Early-stage subscription apps (finding traction)
At the early stage, the goal is speed and clarity, not sophistication. Teams need to **get tracking live quickly, keep costs under control, and understand whether users are moving through the initial funnel.**
Fast setup matters because experiments change weekly. Reasonable pricing matters because volume is still unpredictable. Clear visibility into installs and early conversion events also matters because that’s where product–market fit starts to show.
**Best fit**
- **Branch** works well if your growth is web-led and content-driven.
- **Airbridge** fits teams that want to get subscription tracking right from day one, without re-platforming later.
### Scaling performance teams (UA + retention focus)
At this stage, installing volume stops being the main problem. The focus shifts to quality—who converts, who stays, and who pays over time.
Teams need **clear trial-to-paid attribution, cohort-based LTV they can trust, and SKAN reporting that doesn’t live in a separate workflow.**
Just as important, they need fewer dashboards. When performance, retention, and revenue data are split across tools, optimization slows and confidence drops.
**Best fit**
- **Airbridge** stands out here with unified UA, retention, revenue, and SKAN into one trusted view, making it easier for scaling teams to optimize for long-term LTV instead of just installs.
Appsflyer and Adjust still perform well for pure UA scale, but require more tooling around them to answer subscription questions.
### Mature teams replacing a legacy MMP
For mature teams, the issue isn’t access to data. It’s trust in it.
At this stage, data confidence matters more than feature breadth. Pricing predictability becomes a board-level concern. Attribution has to hold up under privacy constraints without constant rework. Operational efficiency matters because every extra process compounds across teams.
Legacy MMPs often start to feel heavy here—powerful, but expensive to run and harder to adapt. Replacement decisions are usually driven by the need for clarity, stability, and a system that scales without friction.
**Best fit**
- **Airbridge** when teams want a modern attribution layer without enterprise overhead.
- **Adjust** if you have complex global compliance needs and large analyst teams.
## Conclusion: Choosing One MMP as Your Single Source of Truth
Subscription growth breaks when data lives in too many places.
When install attribution sits in one tool, subscription revenue in another, web analytics somewhere else, and SKAN reports in isolation, teams lose speed and confidence. Decisions turn reactive. Experiments slow down.
Appsflyer, Adjust, and Branch remain strong platforms for specific use cases. But many subscription apps outgrow install-first attribution faster than expected. What they need next is clarity—across acquisition, retention, and revenue.
This is where **Airbridge** is gaining ground:
- Subscription events are treated as first-class data
- Web, app, and SKAN journeys connect in one view
- Teams stop reconciling numbers and start optimizing growth
If your MMP feels like a reporting obligation instead of a growth engine, it may be time to rethink the stack. And don't just take our words for it!
👉[How Playio Increased D30 Retention to 30% and Cut Global UA CPA by 40% with Airbridge](https://abr.ge/jqz1p7)
👉[How Rooster Games Secures 15% ROAS Uplift with Airbridge’s Cross-Platform Measurement](https://abr.ge/g1y748)
👉[Nightly Cuts CPA 18% with Simulated iOS Attribution, Ranks Top 3 in Japan’s App Store](https://abr.ge/ubscdbp)
---
## GA4 vs. MMP: Why Subscription & AI Apps Are Losing Revenue with Google in 2026 (And How to Fix It)
URL: https://www.airbridge.io/en/blog/ga4-vs-mmp-why-subscription-ai-apps-are-losing-revenue-with-google-in-2026-and-how-to-fix-it
Language: English
Published: 2026-02-23
## TL:DR
- GA4 is designed for cross-platform analytics and relies on modeled data, not persistent user-level mobile attribution
- When revenue is delayed, GA4 often drops or reassigns trials, renewals, and usage-based upgrades
- MMPs are built to persist attribution across installs, conversions, and renewals at the user level
- Without an MMP, revenue disconnects from acquisition, making profitable channels look inefficient
- Using only GA4 is insufficient. An MMP acts as the system of record for mobile growth, preserving revenue truth across channels, while GA4 should be used only for product and behavior analytics
## Why Attribution Accuracy Matters More for Subscription & AI Apps
Attribution accuracy is the ability to correctly link acquisition sources to downstream revenue events such as trials, renewals, and churn. For subscription and AI apps, inaccurate attribution directly distorts LTV, CAC, and channel ROI.
Subscription and AI apps don’t monetize at install. Revenue appears days or weeks later due to usage-based pricing, delayed upgrades, and feature gating. **When the initial attribution is wrong, every future event inherits that error.**
In practice, this leads teams to optimize for cheap trials instead of high-retention users, misread payback periods, and scale channels that look efficient but quietly churn.
Stop assuming attribution is just a reporting problem. It’s a revenue integrity problem.
## What is the difference between GA4 and an MMP (Mobile Measurement Partner)
GA4 is Google’s event-based measurement system designed primarily for cross-platform analytics. On the other hand, an **MMP(Mobile Measurement Partner)** is a mobile-first attribution system built to deterministically connect installs, post-install events, and revenue to paid media sources.
**GA4** was built to answer _what users do_ across web and app properties. It relies heavily on modeled data, aggregated event streams, and privacy-driven estimation.
**A common mistake marketers make is treating GA4 as a source of truth for mobile acquisition.** It was never designed to own install-level attribution across ad networks, SKAN, and post-install revenue.
**MMP attribution** was built to answer _where users come from and what they’re worth_. MMPs use deterministic identifiers (where allowed), direct ad network integrations, and mobile-specific frameworks like SKAN to attribute installs and revenue at the user level.
From a tactical standpoint, this is what enables accurate ROAS, cohort LTV, and subscription revenue mapping across paid channels—without relying on modeled guesses.
| Dimension | GA4 Attribution | MMP (e.g., Airbridge) |
| --- | --- | --- |
| Core Purpose | Cross-platform analytics | Mobile attribution & measurement |
| Attribution Method | Modeled, aggregated | Deterministic where allowed |
| Mobile Install Attribution | Limited | Native, install-level |
| Ad Network Integrations | Primarily Google ecosystem | All major mobile ad networks |
| SKAdNetwork (SKAN) | Basic, delayed reporting | Native, structured SKAN handling |
| Subscription Revenue Tracking | Event-based, fragmented | User-level revenue mapping |
| Reinstall & Re-attribution | Weak support | Fully supported |
| Fraud Detection | Not available | Built-in fraud prevention |
| Data Ownership | Google-controlled logic | Advertiser-controlled |
*GA4 vs MMP: Core Differences & Limitations*
## Why Subscription & AI Apps Need an MMP Over GA4
**Subscription and AI apps need an MMP because revenue, retention, and lifetime value** depend on persistent, user-level attribution that GA4 cannot reliably maintain across mobile channels.
> **⚠️ Be careful ⚠️
> **GA4 breaks once revenue becomes delayed. Trials convert days later. Renewals happen weeks later. Usage-based upgrades appear after repeated engagement. GA4’s modeled attribution often reassigns or drops these events, which is why revenue rarely reconciles with acquisition data.
However, **MMPs are built to persist attribution across the entire user lifecycle.** They connect install **source → trial → conversion → renewal → expansion** at the user level. This is especially critical for AI apps, where revenue often correlates with usage depth rather than first-session behavior.
### What Actually Breaks Without an MMP
- Trial conversions get credited to “Direct” or “Unknown”
- Renewals lose their original acquisition source
- Re-engagement campaigns overwrite true acquisition data
- Paid channels appear interchangeable, even when retention differs
### Real-World Success: How Galaxy Play achieved a 22% revenue increase
[Galaxy Play](https://galaxyplay.vn/?srsltid=AfmBOorWLgd8cBw7Or83RAlw8-fszswS1Z0e3vHSvl89exj66Stwx2ox), a leading OTT streaming platform in Vietnam, struggled with fragmented web-to-app attribution and unreliable data from its previous measurement setup, making it difficult to understand which campaigns truly drove revenue.
After switching to Airbridge MMP, Galaxy Play unified its web and app user journeys with accurate Web-to-App tracking, gaining a single source of truth for performance and fraud-free attribution. This shift enabled smarter budget allocation and optimization, resulting in a **22% revenue increase and 52% cost reduction** after adopting Airbridge.
👉[How Galaxy Play achieved a 22% revenue increase powered by Airbridge Web-to-App tracking.](https://abr.ge/v2teduj)
## FAQ: GA4 vs MMP Attribution
### Can GA4 replace an MMP for mobile app attribution?
No. GA4 was designed for cross-platform analytics, not mobile install attribution. It lacks deterministic install tracking, full ad network integrations, and reliable revenue persistence. GA4 works best alongside an MMP, not instead of one.
### Why is GA4 attribution inaccurate for subscription apps?
Subscription revenue occurs well after install. GA4 relies on modeled and aggregated attribution, which often drops or reassigns delayed conversions. This breaks LTV and ROAS analysis for trial-based and usage-based products.
### Why does GA4 show installs correctly but revenue incorrectly?
Because GA4 does not persist attribution at the user level across time. Installs are captured close to the acquisition event, but revenue often occurs later. GA4’s modeled attribution reassigns or drops delayed events, especially under privacy constraints, which is why install counts look stable while revenue attribution drifts.
### Do Subscription apps & AI apps need an MMP?
Yes. These apps often monetize through usage depth, credits, or delayed upgrades. Without user-level attribution, high-value users appear disconnected from their acquisition source, leading to underinvestment in profitable channels.
### Is it redundant to use both GA4 and an MMP?
No. They serve different purposes. GA4 answers _how users behave_. MMPs answer _where users came from and what they’re worth_. Mature teams intentionally run both.
### Why do performance teams choose Airbridge over GA4?
Airbridge provides deterministic mobile attribution, subscription revenue mapping, SKAN-native measurement, and advertiser-controlled data logic. GA4 does not.
## Best Practices for App Attribution in 2026
Modern app attribution in 2026 requires combining privacy-first measurement with user-level revenue accuracy across the full subscription lifecycle.
Use this as a practical baseline:
- Use an **MMP as the system of record** for installs, re-attribution, funnels, and paid media ROAS
- Use **GA4 for product analytics** and engagement analysis
- Track trials, renewals, cancellations, and expansions as first-class events in your MMP
- Separate acquisition and re-engagement attribution logic
- Validate SKAN postbacks weekly against internal revenue data
- Optimize channels based on **cohort LTV**, not early ROAS
## The Bottom Line: Why an MMP Is Non-Negotiable for Subscription & AI Apps
For subscription and AI apps, revenue happens long after install. When attribution cannot persist across that delay, revenue disconnects from acquisition. GA4 was not built to maintain user-level attribution over time, which is why LTV and ROAS quietly drift.
An MMP exists to protect that revenue truth. It preserves attribution across the full user lifecycle, connects paid media to real subscription outcomes, and gives teams confidence to optimize based on long-term value rather than short-term signals. For apps with delayed and recurring monetization, an MMP is not optional. It is the foundation for sustainable growth.
See how Airbridge helped leading subscription and AI apps drive measurable growth:
👉[Monkey Achieves 40% Growth in Subscription Revenue, Strengthening Its Leadership in Vietnam’s EdTech Industry](https://abr.ge/6k1e2l)
👉[Nightly Cuts CPA 18% with Simulated iOS Attribution, Ranks Top 3 in Japan’s App Store](https://abr.ge/znr4eg)
👉[Airbridge’s unified cross-platform insights help Shmoody scale to over 1M app installs](https://abr.ge/a016utf)
---
## MMP for Subscription Apps: How to Fix Broken Attribution and LTV in 2026
URL: https://www.airbridge.io/en/blog/mmp-for-subscription-apps-how-to-fix-broken-attribution-and-ltv-in-2026
Language: English
Published: 2026-02-13
Author: Jaehyuk Kim
A Meta campaign delivers 4,200 installs. Google Ads claims 2,800. Your [MMP (Mobile Measurement Partner)](https://www.airbridge.io/en/glossary/mobile-measurement-partner-mmp) attributes 1,100 trial starts and 310 first subscriptions across both channels. So far, so good.
Then month two arrives. RevenueCat shows 58 billing failures, 41 refunds, and a 22% involuntary churn rate. **Your MMP shows none of it at the channel level.** The 310 subscribers who looked like revenue are quietly shrinking, but your attribution system cannot tell you which campaign acquired the users who churned and which acquired the ones who renewed.
**Key Takeaways**
* **Most MMPs stop attributing after the first subscription payment.** Renewals, billing failures, refunds, and cancellations happen server-side and never connect back to the acquisition channel.
* **This blind spot distorts LTV-by-channel calculations.** A channel that looks efficient by cost per subscription (CPS) may carry a 15% billing failure rate by month three, but your attribution data will never reveal it.
* **Half of all subscription churn is involuntary.** Failed payments from expired cards and bank declines account for a large share of lost subscribers, yet most attribution systems cannot surface this at the channel level.
* **Fixing the gap requires server-to-server (S2S) integration between your billing platform and your MMP.** Device-side SDKs depend on app opens to sync subscription state, creating latency and missed events.
* **Airbridge Core Plan connects RevenueCat and Adapty data to channel-level attribution.** This closes the post-payment blind spot for subscription apps running paid UA on Meta, Google, Apple Search Ads, and TikTok.
## The Subscription App Tracking Blind Spot: Where Attribution Breaks
Most MMPs are engineered for install [attribution](https://www.airbridge.io/en/blog/what-is-mobile-app-attribution). They answer one question well: which channel drove this install? For subscription apps with 7-to-14-day free trials, they also capture trial starts and first payments. But after that initial conversion, the subscription lifecycle continues while the MMP goes silent.
### 1. Why MMPs Lose Signal After the First Payment
Subscription billing events are processed server-side by Apple and Google. Renewals, billing failures, grace period entries, and cancellations all happen outside the app. A device-side SDK can only detect subscription state changes when the user opens the app and triggers a sync. If a user's card declines at month three but they never reopen the app, your MMP has no record of that billing failure.
This creates a structural gap. Everything below the first payment line is invisible to your attribution system:
* Renewal confirmations (month 2, 3, 4...)
* Billing failures and grace period entries
* Voluntary cancellations and refund events
* Subscription downgrades or plan changes
The problem is not that your MMP is broken. It is that MMPs were designed for install attribution, and subscription revenue is a post-install lifecycle event that depends on server-side billing infrastructure, not on the device-side SDK.
[Where your MMP loses signal in the subscription funnel]
### 2. How This Compounds for Health and Fitness Apps
Health and fitness subscription apps face a vertical-specific version of this problem. Seasonal acquisition spikes in January (New Year's resolutions) produce high trial volumes, but by March, churn accelerates as motivation fades. Without post-payment attribution, a growth team running January campaigns cannot distinguish between channels that acquired committed subscribers and channels that acquired seasonal users who cancel within 60 days.
According to [RevenueCat's State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/), the median trial-to-paid conversion rate for Health and Fitness apps is 39.9%. But that conversion rate only captures the first payment. The real question is which channels produce subscribers who stay past the first renewal, and most attribution systems cannot answer it.
The same report shows that nearly 30% of annual subscribers cancel within the first month. For fitness apps that acquire heavily in January, this means a large share of "successful" conversions from winter campaigns may churn before the second billing cycle. If your attribution stops at the first payment, your January ROAS looks strong while your actual subscriber retention tells a different story.
### 3. The Cost of Flying Blind on Channel LTV
When your MMP cannot track post-payment events at the channel level, your [LTV](https://www.airbridge.io/en/glossary/lifetime-value-ltv) calculations are built on incomplete data. Billing platforms like RevenueCat show you the totals. Your MMP shows you the channels. But no system shows you both, unless you connect them.
The result: budget decisions based on CPS alone systematically favor channels that produce high first-payment volume over channels that produce subscribers who renew.
Consider two channels, both spending $10,000 per month:
| Metric | Channel A (Meta) | Channel B (Google) |
|:-------|:-----------------|:-------------------|
| **CPS** | $22 | $28 |
| **First subscribers** | 454 | 357 |
| **Renewal rate (month 3)** | 60% | 85% |
| **Active subscribers (month 3)** | 272 | 303 |
| **12-month projected LTV** | $48 | $89 |
Without post-payment attribution, your MMP tells you Channel A wins because it has a lower CPS. The real winner is Channel B, which produces subscribers who stay and renew. But that signal only becomes visible when renewal and billing failure data flows back into your attribution system.
[How the tracking blind spot distorts channel LTV]
If you run paid acquisition for a [subscription app](https://www.airbridge.io/en/blog/subscription-app-growth-guide), this is not a theoretical problem. Every month without channel-level renewal data is a month of budget flowing toward the wrong channels.
## How to Fix Subscription Tracking Gaps
### Practical Steps Any Team Can Take Today
Before investing in new tooling, growth teams can reduce the tracking blind spot with these actions:
* **Align [attribution windows](https://www.airbridge.io/en/glossary/attribution-window) across all ad platforms.** Mismatched windows between Meta (7-day click), Google (30-day click), and your MMP create false discrepancies before the blind spot even begins.
* **Export RevenueCat or Adapty cohort data manually.** Match monthly subscription cohorts against your MMP's install cohorts by date to approximate channel-level renewal rates.
* **Tag campaigns with consistent naming conventions.** When you do connect billing data to attribution, clean naming makes the join possible. Inconsistent names break the match.
* **Track billing failure rates in RevenueCat by install date.** Even without channel attribution, you can spot whether recent cohorts have worse billing failure rates than older ones.
* **Build a simple spreadsheet reconciliation.** Export monthly data from your MMP (installs and trials by channel) and from RevenueCat (subscribers and churned by cohort date). The overlap gives rough channel-level retention visibility.
These steps give partial visibility. But they are manual, lagging, and cannot replace real-time channel-level subscription attribution.
### How Airbridge Core Plan Closes the Post-Payment Blind Spot
[Airbridge Core Plan](https://www.airbridge.io/en/plans/core) is built to answer one question for subscription apps: **"Are paid users converting into subscriptions, and which channels are driving value?"**
Unlike enterprise MMPs that treat subscription tracking as an add-on locked behind higher pricing tiers, Core Plan includes [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integrations in the base offering. This means subscription lifecycle events flow into the same attribution system that tracks installs and trials, closing the gap between acquisition data and revenue data.
**What Core Plan tracks across the subscription funnel:**
* **Install** and **Sign-up** attributed to channel and campaign
* **Start Trial** with channel-level trial-to-paid visibility
* **Subscribe** with revenue attributed to the originating ad
* **Unsubscribe** and lifecycle transitions
Core Plan supports 25 standard events optimized for the subscription funnel. These use predefined event names (Start Trial, Subscribe, Unsubscribe), reducing the schema design work that slows setup with enterprise tools. Customers still need to log these events in their app code, but the predefined names and parameters eliminate event naming decisions.
**Channel coverage:** Core Plan integrates with Meta Ads, Google Ads, Apple Search Ads, and TikTok for Business. These four [Self-Attributing Networks (SANs)](https://www.airbridge.io/en/glossary/self-attributing-network) represent 80-90% of early-stage paid acquisition spend.
| Capability | Enterprise MMP | Airbridge Core Plan |
|:-----------|:---------------|:--------------------|
| Billing platform integration | Native (included in higher tiers) | Native (included in base) |
| Subscription funnel events | Predefined + custom events | 25 subscription-optimized standard events |
| Channel coverage | All networks supported | GMAT (Meta, Google, Apple, TikTok) |
| Minimum contract | Annual, $10K+ | Pay-as-you-go, 15K free installs |
| Setup guidance | General-purpose | Subscription-app focused |
| Third-party integrations | Unlimited | Maximum 2 (e.g., RevenueCat + Amplitude) |
Core Plan does not support custom events or non-SAN ad networks. This is intentional. For subscription apps running paid UA on major channels, standard events cover the full funnel without schema planning. Teams needing custom events, raw data exports, or agency access can upgrade to [Airbridge's Growth Plan](https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025) as their operations mature.
**Reports that surface subscription blind spots:**
Core Plan includes six built-in reports: Actuals, Trend, Active User, Funnel, Retention, and Revenue. The Funnel report visualizes the install-to-trial-to-subscription path by channel. The Retention report shows cohort retention by acquisition source, revealing which channels produce subscribers who stay past the first renewal. The Revenue report attributes subscription revenue to the campaign that drove the original install.
Together, these reports answer the question that the subscription app tracking blind spot hides: which channels produce subscribers who renew, and which channels produce subscribers who churn?
## FAQ: Subscription App Tracking and Attribution
### How do billing failures affect subscription app LTV calculations?
Billing failures cause involuntary churn, where subscribers lose access not because they chose to cancel, but because a payment failed. When your MMP cannot attribute billing failures to specific channels, your LTV-by-channel numbers overcount active subscribers. The result is inflated LTV estimates for channels with higher involuntary churn rates, leading to misallocated budget.
### What is the difference between SDK-based and S2S subscription tracking?
SDK-based tracking requires an app session to sync, which means subscriptions processed while the app is closed are not captured in real time. S2S (server-to-server) integration transmits subscription events directly from the billing platform to the attribution system, regardless of app state. For subscription apps, S2S is the reliable path to post-payment attribution.
### When should a subscription app upgrade from Core Plan to Growth Plan?
Core Plan is designed for teams running paid UA on Meta, Google, Apple Search Ads, and TikTok with a subscription-based revenue model. Teams typically upgrade to Growth Plan when they need custom event tracking, more than two third-party integrations, raw data exports to a data warehouse, or support for additional ad networks beyond GMAT.
## The Revenue You Cannot Attribute Is Revenue You Cannot Scale
Every month without channel-level subscription attribution is a month of budget decisions built on partial data. The channels that produce your highest-LTV subscribers may be underfunded, while the channels that produce high first-payment volume but poor renewal rates absorb your budget increases.
**[Start Free with Airbridge Core Plan](https://www.airbridge.io/en/plans/core)** and connect your RevenueCat or Adapty data to channel-level attribution, starting with 15K free attributed installs.
---
## Subscription App Onboarding: Get 100% of Users to Your Paywall
URL: https://www.airbridge.io/en/blog/subscription-app-onboarding
Language: English
Published: 2026-02-03
Many subscription apps win installs but lose users fast, they churn before ever seeing the paywall. The reason? Value isn’t clear soon enough.
That’s why Onboarding is more than a welcome flow, it’s the make-or-break moment where users decide whether your app is worth keeping. Done right, it transforms first-time curiosity into active engagement, gets users to your trial and paywall, and ultimately drives revenue growth.
In this blog, we break down **why onboarding is among the key drivers of subscription app growth, how it shapes retention and paywall conversion**, and how data-driven strategies turn first-time users into loyal subscribers.
## **Key Takeaways**
- **A subscription paywall only works if users reach it.** In most subscription apps, churn happens before the paywall is ever shown, making early user experience more critical than pricing or offer design.
- **Early drop-off is an onboarding problem, not a paywall problem.** Users leave because they don’t see value fast enough. Optimizing trials or pricing won’t help if users never experience meaningful usage.
- **Onboarding is the first conversion layer.** Before subscriptions or trials, onboarding determines whether users understand the app, engage with it, and continue forward in the funnel.
- **Data turns onboarding into a growth lever.** Without end-to-end measurement, teams can’t prove which onboarding flows drive retention, trials, or subscriptions.
- **Airbridge connects onboarding to revenue.** By unifying user journeys, retention cohorts, and attribution data, teams can identify which onboarding paths actually lead users to the paywall — and beyond.
## **What Is a Subscription Paywall?**
**A paywall is a digital barricade that limits a user’s access to certain features or content until they subscribe or pay for it.** It is a monetization tool used by many subscription-based platforms to offer exclusive benefits and content to users who pay a fee.
Mobile apps use a paywall screen, which is included in the app and serves as a blockade to restrict user access. A paywall screen typically consists of information about subscription offers, different pricing options, and a call-to-action (CTA) button users can click on to make the purchase.
Each app designs and formats its own paywall screen with elements that best represent the subscription features and value of the app, and marketers can also decide when to display the screen within a user’s app session.
[__wf_reserved_inherit]
Learn more about [different types of subscription paywalls here.](https://www.airbridge.io/glossary/paywall)
## **Why Many Users Drop Off Before Reaching the Paywall?**
Here’s the thing about subscription apps: a significant portion of users don’t quit at the paywall. They never even get that far.
That’s the hard pill to swallow. Marketers often spend weeks fine-tuning the paywall design, debating trial lengths, or testing different pricing models.
But the truth is, **none of it matters if users drop off before they ever see it**. The paywall isn’t where most people make the decision to leave. It’s much earlier.
The numbers sting:
- 25% churn in the first 3 minutes
- 77% churn within 3 days
- D30 retention stuck at 5–6%
[__wf_reserved_inherit]
The data shows how quickly users disengage before monetization even becomes an option.
But what drives this early exit? According to [RevenueCat’s State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025/), the biggest cancellation reason is simply **insufficient usage**, making up over 37% of churn.
The high rate suggests that many subscription apps could boost retention and paywall conversion by improving onboarding and helping users see value earlier.
## **How Onboarding Determines Paywall Reach and Subscription Conversion**
Onboarding is no longer a side project for the product team. It’s the first powerful growth lever a subscription app has. And that’s why onboarding truly becomes the make-or-break moment for subscription apps.
### **Onboarding Is the First Conversion Layer**
Subscription churn rarely occurs because an app lacks features. More often, users leave because they never understand how those features fit into their needs. Onboarding is where curiosity turns into usage. Miss that window, and the rest of the funnel never gets a chance to work.
Every interaction during onboarding contributes to a decision: continue exploring or abandon the app.
**Clear guidance, relevant personalization, and reduced friction help users move forward**, while confusing flows and generic messaging accelerate drop-off.
In this sense, onboarding functions as the app’s first conversion event, long before trials or subscriptions are presented.
### **Reducing Time-to-Value to Increase Paywall Reach**
Time-to-value refers to how quickly users experience their first meaningful benefit from the app. In subscription models, this timing is critical. **The longer it takes for users to feel progress or usefulness, the lower the probability they will return or reach the paywall.**
Effective onboarding shortens this window by focusing users on actions that lead to immediate results, like completing a workout, achieving a learning milestone, or finishing a guided session.
These early successes establish perceived value and increase the potential that users remain engaged long enough to encounter the subscription offer.
### **Using Onboarding to Establish Usage Habits**
Subscription apps don’t just sell content or features. **They sell habits:** daily streaks, weekly progress, personalized milestones. Onboarding is where those behaviors take root.
**By encouraging small, repeatable actions early on, onboarding lays the foundation for habits that extend beyond the first session.** Features like progress indicators, reminders, and personalized goals help reinforce continued engagement.
When onboarding succeeds in shaping behavior, the paywall appears after users have already invested time and effort, making the subscription decision more natural.
## **How to Optimize Onboarding to Drive Paywall Reach**
The most successful subscription apps share a singular focus: **helping users experience a "Value Realization" moment before reaching the paywall.**
To maximize paywall reach and long-term retention, onboarding must evolve from a static walkthrough into a dynamic, performance-driven funnel.
- **Prioritize Value over Features:** Users don't buy tools; they buy outcomes. Instead of listing features, guide users toward their first milestone—whether it’s completing a workout, generating a report, or seeing a personalized insight.
- **Implement Progressive Guidance:** Reduce cognitive load by introducing complexity only when necessary. Successful onboarding uses a tiered approach, revealing instructions based on the user's current stage in the journey rather than overwhelming them at the first launch.
- **Deploy Behavioral Personalization:** Abandon the one-size-fits-all path. By adjusting onboarding flows based on real-time engagement signals and user intent, marketers can maintain high relevance and prevent premature drop-offs.
**The Bottom Line:** Measure to Monetize Optimization is impossible without granular visibility. To identify which flows truly drive revenue, growth teams must track user behavior across every session and channel. Using a robust MMP like [**Airbridge**](http://airbridge.io/pricing) allows you to connect onboarding interactions directly to trial starts and subscription events, ensuring your growth strategy is backed by data.
## **Turning Onboarding into Growth with Data-Driven Strategies**
So here’s the challenge: everyone agrees onboarding matters, but very few can prove it. Teams run tests, tweak flows, adjust copy, maybe shorten sign-up steps.
**But without the right data, it all feels like guesswork.** Did that new tutorial really reduce churn? Did personalizing the welcome screen actually keep users around? Hard to tell when the journey looks like a black box.
Moreover, retention isn’t just about keeping people inside the app. It’s about making sure they come back, across channels, after that first session. And that’s messy.
A user might tap an ad, sign up on the web, download the app later, start a trial on mobile, and then finally pay on desktop. Without connecting all those touchpoints, it’s almost impossible to see which onboarding flows actually worked and which ones fell flat.
This is where **Airbridge makes the difference**. By stitching journeys together end-to-end, it turns onboarding into a growth lever you can track and optimize.
Measuring Onboarding Impact & Retention with Airbridge:
- **Unified tracking**: Shows the full journey, proving which onboarding paths lead to sustained engagement.
- **Cohort analysis**: Identifies onboarding variants that keep users active through D7, D30, and beyond.
- **Attribution clarity:** Reveals which acquisition sources bring in users who not only convert but also stay.
Ready to get 100% of users to your app paywall with better onboarding? Then start [onboarding with Airbridge](https://www.airbridge.io/book-a-demo).
See how teams leverage Airbridge to drive growth in real campaigns:
👉 [How Playio Increased D30 Retention to 30% and Cut Global UA CPA by 40% with Airbridge](https://www.airbridge.io/en/customers/playio-d30-retention-cpa-optimization)
---
## Why Subscription Churn Happens: Top 5 Cancellation Reasons Forecast for Mobile Apps in 2026
URL: https://www.airbridge.io/en/blog/why-subscription-churn-happens-top-5-cancellation-reasons-forecast-for-mobile-apps-in-2026
Language: English
Published: 2026-01-29
## **Alarming Retention and Churn Rate for Subscription Apps**
Subscription growth is no longer the strategic challenge — retention is.
By 2026, the subscription economy will have matured. Acquisition costs continue to rise while acquisition rates have declined, pushing teams toward retention-first growth strategies.
[Recurly](https://recurly.com/content/state-of-subscriptions-report/?) reports that **return acquisitions now make up around 20% of new subscribers**, highlighting the growing importance of keeping users longer rather than constantly replacing them.
However, churning remains a structural issue, even for high-performing apps. Industry benchmarks by [MarketingLTB](https://marketingltb.com/blog/statistics/subscription-statistics/) suggest an average monthly churn of 5–7%, with annual churn exceeding 30% in the absence of strong retention strategies.
For mobile subscription apps, the pressure is even higher, with **up to 30% of annual subscribers canceling within the first month after conversion**, according to [RevenueCat](https://www.subscriptioninsider.com/article-type/news/revenuecats-state-of-subscription-apps-2025-report-ais-dominance-retention-challenges-and-the-shift-away-from-pure-subscriptions).
Rather than accepting churn as inevitable, leading teams are breaking it down into actionable root causes tied to product usage, lifecycle timing, pricing, and user behavior.
Below are the **five most common cancellation drivers in subscription apps today**—and what to do about them.
### **5 reasons of subscription apps churn**
1. Post-Conversion Value Decay
2. Pricing Friction and Subscription Fatigue
3. Cognitive Overload from Feature Bloat
4. Lack of Personalization in a Privacy-First Environment
5. Ineffective Cancellation and Win-Back Experiences
## **1\. Post-Conversion Value Decay**
Perhaps the most under-discussed driver is _value decay_ after the initial conversion.
Many apps succeed at getting users past onboarding and even past their first billing cycle. However, subscribers often plateau shortly thereafter. Across subscription services broadly, **37% of cancellations occur because people stop using the product** — a clear signal from this [cancellation survey data](https://www.revenuecat.com/blog/growth/subscription-app-churn-reasons-how-to-fix/) that subscribers aren’t finding _ongoing utility_. [](https://marketingltb.com/blog/statistics/subscription-statistics/?utm_source=chatgpt.com)
This isn’t just a churn metric; it’s a measurement gap. Top performing teams monitor **value milestones well beyond activation** — for example, how deeply users engage with high-value features at 30, 60, and 90 days.
**Airbridge’s take 💡**
Instead of viewing churn as a post-hoc KPI, think of engagement trends as _leading indicators_. If usage flattens before renewal, intervention — such as contextual lessons, personalized nudges, or feature tours — can re-spark momentum before churn happens.
### **How does Duolingo reduce its churn rate?**
This pattern has been widely observed in language learning and productivity apps. For example, [Duolingo](https://www.latterly.org/duolingo-marketing-strategy) continuously uses a combination of gamification features — including streaks, rewards, and social competition — to turn learning into a daily habit and boost long-term engagement.
[Duolingo Streaks, Rewards, and Social Interaction, respectively. Source: Duolingo]
## **2\. Pricing Friction and Subscription Fatigue**
Subscription fatigue is no longer a fad — it’s a macro-consumer behavior.
Consumers are tightening budgets, rationalizing recurring expenses, and reassessing perceived value. In the United States alone, **about** [**one-third of streaming users reported canceling one or more services**](https://markets.financialcontent.com/stocks/article/globeprwire-2025-7-13-global-subscription-services-in-decline-giskaa-survey-report), citing pricing pressure and “too many subscriptions” as leading factors. [](https://markets.financialcontent.com/stocks/article/globeprwire-2025-7-13-global-subscription-services-in-decline-giskaa-survey-report?utm_source=chatgpt.com)
Crucially, churn attributed to pricing concerns isn’t always about cost per se — it’s often a _value perception issue_. According to [cancellation survey data](https://www.revenuecat.com/blog/growth/subscription-app-churn-reasons-how-to-fix/), roughly **35% of churners cite value-related concerns**, making it one of the top explicit reasons subscribers pull the plug.
### **How does Netflix and Disney+ reduce its churn rate?**
Streaming services provide a clear illustration of this behavior. Platforms like Netflix and Disney+ have seen waves of “subscribe–cancel–resubscribe” behavior as users sign up for specific content releases and leave when perceived value drops.
This pattern reflects **a broader shift toward intentional, short-term subscription usage rather than long-term commitment.**
[Streaming trends in the USA. Source: Fabric]
> **Airbridge’s take 💡
> **In practice, modern pricing strategies that mitigate churn include tiered plans aligned to usage intensity, flexible billing cycles, and pause options that let users _step down instead of leaving entirely_. Recurly’s subscription data shows that pause options surged by **68% year-over-year**, with over $200M generated from paused subscribers who later reactivated.
## **3\. Cognitive Overload from Feature Bloat**
More features do not automatically translate to more loyalty.
As subscription apps evolve, they risk overwhelming users with options before those options demonstrate value. When users feel unsure about what matters most, they disengage and ultimately churn.
This phenomenon aligns with broader product behavior research: users often abandon apps when they don’t quickly form habits or derive recurring value signals. That’s why [nearly 40% of cancellations](https://www.revenuecat.com/blog/growth/subscription-app-churn-reasons-how-to-fix/) come from insufficient usage or low engagement levels. [](https://www.revenuecat.com/blog/growth/subscription-app-churn-reasons-how-to-fix/?utm_source=chatgpt.com)
> **Airbridge’s take 💡**
> The solution is _intelligent feature discovery_ and _contextual relevance_. Instead of showing every capability up front, leading subscription apps guide users through progressively higher-value moments based on behavior patterns. This is why cohort-based onboarding, feature staging, and adaptive UX flows are now core retention tactics.
### **How does Notion reduce its churn rate?**
Productivity tools like Notion have spoken openly about the trade-off between flexibility and complexity. While advanced users thrive on customization, newer or lighter users can feel overwhelmed, and multiple UX articles describe [how the platform’s flexibility often translates to cognitive load](https://www.noratemplate.com/post/why-is-notion-so-hard-to-use) for newcomers.
This is why Notion has increasingly invested in templates and guided starting points to reduce early friction.
## **4\. Lack of Personalization in a Privacy-First Environment**
Users expect personalization, but with privacy constraints tightening, delivering it has become more nuanced.
Generic lifecycle messaging feels stale, and irrelevant communication amplifies churn risk. At the same time, app ecosystems are losing access to traditional identifiers, forcing teams to rethink personalization logic.
The good news? Internal, first-party behavioral signals (like session frequency, feature engagement, and event sequences) enable product-centric personalization without compromising privacy.
**Carefully sequenced messaging** — for example, nudges tied to usage thresholds or inactivity patterns — has been shown to _meaningfully decrease cancellations and improve retention_. [](https://marketingltb.com/blog/statistics/subscription-statistics/?utm_source=chatgpt.com)
> **Airbridge’s take 💡
> **In other words, personalization isn’t about deeper targeting, but _better timing_ and _relevance_ using the data you already own.
### **How does Spotify reduce its churn rate?**
[Spotify](https://open.spotify.com/) is often cited as a benchmark for personalization built on first-party behavior rather than external identifiers. Features like Discover Weekly and Daily Mix demonstrate how relevance and timing—based on listening patterns—can drive long-term engagement even as privacy constraints increase.
[Spotify’s Marketing Strategies. Source: Latterly]
## **5\. Ineffective Cancellation and Win-Back Experiences**
Treating cancellation as a dead end is a lost opportunity.
Churn events are rich with signals about pricing pain points, feature gaps, or usage drop-offs. Yet many apps still rely on generic exit surveys or one-size-fits-all retention offers that don’t address the underlying cause.
> **Airbridge’s take 💡**
> Data suggests that over half of subscribers would consider returning if offered a personalized incentive, highlighting the value of strategically designed re-engagement paths. [](https://marketingltb.com/blog/statistics/subscription-statistics/?utm_source=chatgpt.com)Timing is everything. Immediate, untargeted win-back pushes often miss the mark. Instead, leading subscription teams wait for signals of renewed intent—such as an app revisit—before re-engaging users.
### **How does Calm and Headspace reduce its churn rate?**
Wellness apps like Calm and Headspace treat cancellation as a pause, not a failure, focusing on timely, context-driven re-engagement during periods of heightened need rather than pushing instant discounts.
## **Where Airbridge Fits: Turning Churn Signals Into Retention Strategy**
Identifying churn reasons is only useful if teams can _detect them early and act on them accurately_. That’s where measurement plays a quiet but critical role.
Airbridge supports churn analysis by helping subscription apps move from high-level churn rates to **behavioral, cohort-based insights** across the user lifecycle. Instead of treating cancellation as a single event, teams can examine how engagement, feature usage, and revenue signals evolve leading up to churn.
[Airbridge’s Retention Report with behavioral, cohort-based insightsThe data is randomized for reference purposes only]
In practice, **Airbridge helps teams:**
- Track Day N retention and long-term engagement using clean cohort definitions that separate organic retention from re-engagement noise
- Identify drop-off points across onboarding, feature adoption, and subscription milestones through flexible funnel analysis
- Connect retention patterns to revenue and predictive lifetime value to prioritize the right users for intervention
- Maintain reliable churn insights in a privacy-first environment across iOS, Android, web, and SKAN
Rather than replacing strategy, Airbridge provides the measurement foundation that allows teams to make churn a proactive, data-driven decision rather than a reactive metric.
See how Airbridge helped big names reduce churn rate, improve retention and overall campaign effectiveness with the case studies below:
- [Monkey Achieves 40% Growth in Subscription Revenue, Strengthening Its Leadership in Vietnam’s EdTech Industry](https://abr.ge/htqc66)
- [How Rooster Games Secures 15% ROAS Uplift with Airbridge’s Cross-Platform Measurement](https://abr.ge/ineuz2t)
- [GOSU achieves 25% revenue uplift with Airbridge’s Funnel Report](https://abr.ge/c3c9hn2)
---
## CTV Is Measurable Now: The Airbridge + tvScientific Integration
URL: https://www.airbridge.io/en/blog/airbridge-tvscientific-integration
Language: English
Published: 2026-01-27
For years, TV has sat just outside the performance marketing stack. Budgets were large, reach was broad, and impact was often inferred rather than measured. But that reality is changing fast.
As Connected TV (CTV) matures and performance-driven buying becomes the norm, marketers are no longer willing to treat TV as a black box. They want the same level of accountability they expect from mobile, web, and paid social.
This shift is exactly where the **Airbridge + tvScientific integration** comes in. It connects performance-driven TV buying with a unified measurement framework, making Performance TV measurable in a way that actually fits how modern growth teams work.
# Why CTV needs a new measurement standard
## CTV has matured, but measurement has lagged behind
Connected TV is no longer experimental. Performance-oriented teams actively use it to drive site visits, installs, and downstream conversions. Platforms like tvScientific make it possible to plan, buy, measure, and optimize CTV with clear business outcomes in mind.
While some measurement capabilities are built into CTV platforms, others live in separate reports, shaped by channel-specific logic and viewed in isolation. That makes it harder to evaluate CTV alongside other paid channels, even when it’s influencing the same users and the same journeys.
## Why siloed CTV reporting limits insight
When CTV data sits outside the core measurement stack, clarity suffers. Teams struggle to understand post-view impact, compare outcomes across channels, and confidently decide how much budget CTV deserves.
Over time, this creates friction. Optimization slows. Budget conversations become more subjective. And CTV, despite its potential, remains harder to scale than it should be.
# How the Airbridge + tvScientific integration brings CTV into focus
## CTV activation built for outcomes with tvScientific
[tvScientific](https://www.tvscientific.com/) is the first and only CTV advertising platform purpose-built for performance marketers. The tvScientific platform makes TV advertising accessible and measurable for brands and apps of all sizes.
Rather than treating TV as a reach-only medium, the platform empowers advertisers to plan, activate, measure, and optimize CTV campaigns with business outcomes at the center.
At its core, **tvScientific combines integrated media buying, attribution, measurement, and optimization into a single interface** that brings search-like control and capabilities to TV advertising.
Its proprietary technologies allow advertisers to access premium CTV inventory across major streaming services, **apply precise audience targeting via thousands of unique segments, and analyze campaign impact in real time** — all with radical transparency and scalability rarely seen in traditional TV buying.
## CTV measurement brought into one framework with Airbridge MMP
With the [Airbridge](https://www.airbridge.io/) + tvScientific integration, CTV performance data from tvScientific flows directly into Airbridge’s unified attribution and reporting framework. This means CTV outcomes can be evaluated with the same logic, context, and comparability as other acquisition channels, giving teams a single source of truth for cross-channel analysis.
With the integration, teams now can:
- **CTV performance attribution**: Measure tvScientific CTV campaign performance directly in Airbridge.
- **Full-funnel visibility**: Leverage Airbridge reports such as Actual, Trend, and Funnel Reports to understand CTV’s impact across conversion paths with the same level of visibility teams expect from mobile.
- **Unified Reporting:** Analyze CTV outcomes alongside other acquisition channels. Connect CTV, mobile, and web performance now are in one place.
- **Budget Optimization:** Make budget and optimization decisions based on comparable, end-to-end data rather than isolated insights
By bringing CTV into a single measurement framework, Airbridge helps teams evaluate and scale CTV with confidence.
## Getting started
Activating the Airbridge + tvScientific integration takes just a few steps. Once the integration is enabled, tvScientific campaign data can be brought into Airbridge and analyzed alongside other acquisition channels within a unified attribution framework. This allows teams to begin evaluating CTV performance using the same reports, logic, and workflows they already rely on for mobile and web.
- Set up tvScientific campaign tracking links in Airbridge.
- Configure postbacks for real-time data exchange based on your conversion goals.
- Start tracking ads performance directly in the Airbridge dashboard.
For detailed setup instructions and configuration options, please refer to the Airbridge User Guide. If you would like to find out how Airbridge and tvScientific can help power your campaigns, get in touch with us today!
---
## How LLMs Are Rewriting Mobile App Attribution in 2025 (And What Marketers Must Do Next in 2026)
URL: https://www.airbridge.io/en/blog/how-llms-are-rewriting-mobile-app-attribution-in-2025-and-what-marketers-must-do-next-in-2026
Language: English
Published: 2026-01-26
## **The New Era of App Discovery**
For more than a decade, app discovery followed a predictable pattern: users searched, browsed, compared, then clicked. Marketers optimized every step — from keyword rankings to store creatives to campaign funnels.
But in 2025–2026, that model is dissolving. A new ecosystem is rising, driven not by ads or app stores, but by AI assistants.
Instead of sifting through pages of results, users now ask:
- _“What’s the best app for tracking my workouts?”_
- _“Which puzzle game should I try next?”_
- _“Recommend a reading app for my commute.”_
ChatGPT, Gemini, Perplexity, and Claude no longer behave like search engines. They behave like **decision engines**, compressing the journey from _search → browse → click_ into _ask → answer → act_.
Decision-making happens faster. Intent becomes higher.
But the path users take becomes almost invisible. For mobile marketers, this introduces a new tension: **app discovery is evolving faster than attribution models**.
Traditional measurement frameworks were built around ads, impressions, click IDs, and store referrers. They were designed for a world where every install originated from a trackable tap. But when the starting point is a generative AI model, not an ad network, not the App Store, not Google, the old rules fall apart.
## **What Trends LLMs Are Creating for Mobile Apps**
> **_“How do you measure a user journey that starts inside an AI model that does not expose any marketing signals?”_ - Airbridge Marketing Team**
LLMs don’t just add a new channel. They rewrite how users form intent, choose apps, and ultimately install. And because these journeys start inside a closed model, not inside an ad, **attribution loses visibility**.
### New referral paths that never existed before
LLMs output answers, sometimes with links to websites or App Store pages. These aren’t ad clicks, so there’s no metadata. Without intentional tracking, these installs collapse into “organic.”
### Zero-click influence is rising
According to [Bain](https://www.bain.com/insights/goodbye-clicks-hello-ai-zero-click-search-redefines-marketing), up to 60% of users now stop at the AI-generated answer without clicking through to another site. [Ahrefs](https://ahrefs.com/blog/ai-overviews-reduce-clicks/) reports a similar pattern, noting a 34.5% drop in click-through rates as AI results become more prominent.
> In practice, a user might search for “top budgeting apps” on Google, read the AI overview recommendation, and then go directly to the App Store to look up the suggested brands, without ever clicking a website link.
_Traditional last-click attribution fails here because there wasn’t a last click.
The recommendation influenced the install, but the influence is invisible._
### AI reduces measurable impressions
If more users find apps through AI suggestions, fewer will be exposed to paid ads or keyword-based discovery flows. That means marketers will see:
- Fewer impression-based signals
- Lower click volumes
- Higher “organic” install shares
- More unpredictable install spikes
## **Where Traditional Attribution Fails in the LLM Era**
### No standardized tracking from LLM interfaces
AI models don’t behave like ad networks. There are no impression logs, click IDs, redirect URLs, or SKAN postbacks, and the recommendation simply appears. If the user clicks the link, you can track it — but everything leading up to that moment is a complete black box.
### Invisible or indirect influence paths
Many users don’t click inside the AI interface. They read a recommendation, then hours later:
- search the app on Google,
- type its name in the App Store,
- or click one of your ads elsewhere.
Attribution misses the true origin. The AI has shaped intent, but the data shows an “organic” or “paid” install instead.
### Loss of click data breaks last-click logic
Classic attribution depends on a visible “click → install” chain, but LLM discovery often looks like “recommendation → store visit → install.” With no click signal, MMPs classify the install as organic, thus masking an entirely new acquisition layer.
### The AI recommendation touchpoint is untrackable
The most important moment in the journey — the AI suggestion — happens outside any measurable environment. You can’t pixel ChatGPT. You can’t track impressions inside Gemini. You have no way to know how many times your app was recommended, in what order, or against which competitors.
### Attribution windows need to expand
AI-driven intent is often delayed, as users browse recommendations now, but install later. Short attribution windows fail to capture these lagged behaviors, causing undercounting.
### Fragmented metadata from AI answers
LLMs may rewrite your app description, pull outdated screenshots, reorder your key features, or even link to your website instead of your store page. Each variation can cause major swings in funnel performance, and marketers can’t see any of it.
## **What Mobile Marketers Should Do to Get Mentioned in AI Responses**
Before measuring AI-driven traffic, marketers first need to get recommended.
This is where **Generative Engine Optimization (GEO)** comes in — the practice of shaping your app’s digital presence so AI models naturally surface it in answers.
### What is GEO?
GEO is the practice of optimizing your app’s entire digital footprint so AI models can confidently understand, classify, and recommend it.
Instead of ranking for keywords like traditional SEO, GEO focuses on strengthening the signals LLMs rely on, such as metadata consistency, structured content, review patterns, and brand authority, so your app becomes the most logical answer when users ask an AI for suggestions.
LLMs don’t search the web like Google. They synthesize patterns across public data. To appear in that synthesis, your app needs strong, consistent, and structured signals across every surface AI models analyze.
### What influences LLM recommendations?
- **Clean, structured metadata:** LLMs rely heavily on well-defined information. Remember to ensure consistency across: App Store / Play Store descriptions, Feature lists, Category placement, or Keyword themes.
- **Strong public web presence**: AI models pull from Websites, Reddit and Product Hunt threads, News articles, or Third-party comparisons. The more consistent and credible your presence is, the more confidently LLMs include you in responses.
- **Positive, consistent sentiment:** LLMs detect review patterns at scale. High volume, high-quality reviews increase the likelihood of being recommended.
- **Structured content formats:** Schema-marked FAQs, how-tos, and product info give AI models clearer semantic anchors, improving how accurately they describe your app.
- **Brand authority signals**: Thought leadership, credible mentions, and strong social proof help LLMs judge your app as trustworthy.
### Practical GEO steps marketers can take today
To increase your visibility inside AI answers:
- Strengthen semantic consistency across website, store listing, PR, and social.
- Participate in communities where LLMs scrape data (Reddit, Product Hunt, Quora).
- Create Q&A-style content that mirrors common AI prompts (“best budgeting app for students”, “top running app for beginners”).
- Align ASO with intent language commonly used in AI queries.
- Use CPPs to match the themes LLMs associate with your category.
- Maintain unified branding to prevent AI from misclassifying your app.
## **After Getting Mentioned, How Marketers Can Attribute LLM-Driven Traffic?**
Once your app starts appearing in AI answers, the next challenge is knowing how much it actually influences installs. Here’s how to approach attribution in this new environment.
### Use link parameters where AI provides clickable URLs
When an AI platform includes a link to your website or landing page, ensure that link carries:
- **utm\_source=chatgpt**
- **utm\_medium=organic\_ai**
- **utm\_campaign=recommendation**
Even though you can’t track the interaction inside the AI itself, you can track everything that happens after.
### Rely on deeper funnel signals to validate AI influence
LLM-driven users often convert better because they arrive with clearer intent.
You should monitor AI-originated cohorts for sign-ups, trial starts, purchase/subscription rates, or retention curves.
These downstream signals help you confirm that the uplift is coming from AI-led discovery, even when the “click” is missing.
### Look for uplift correlations
If your app appears consistently in AI answers, you’ll notice patterns:
- organic searches for your brand increase
- direct store visits increase
- category ranking improves
- install spikes align with changes in AI responses or model updates
### Optimize your store journey to capture indirect LLM referrals
Because many users jump from AI → App Store directly, or search for your app manually, your store listing becomes part of the attribution layer.
Improve the handoff by:
- aligning metadata with AI intent keywords
- maintaining consistent visuals
- using Custom Product Pages for AI-driven themes
- tagging deep links with UTM context
## **Where Airbridge Comes In: Measurement for the LLM Era**
Airbridge’s strength in the LLM era comes from its **web-to-app attribution** system, which preserves context even when no ad network is involved.
Instead of relying on click IDs, Airbridge captures the user’s journey across web, app store, and in-app events, letting you attribute installs directly to ChatGPT, Gemini, or any AI interface that outputs a link.
Here’s how the flow works.
### 1\. The AI model outputs a link — you embed UTM parameters
Any link that may appear in an AI response should include UTMs such as:
> utm\_source=chatgpt
> utm\_medium=organic\_ai
> utm\_campaign=recommendation
When a user lands on your site from that AI-generated link, Airbridge captures the parameters instantly through its web SDK.
### 2\. Airbridge carries UTMs through the website session
As the user explores your webpage, Airbridge retains the UTM context.
This ensures no information is lost between the AI recommendation and your install CTA.
### 3\. When the user clicks CTA “Install,” the UTMs are passed into the app-open link
Airbridge automatically injects the session metadata into:
- mobile deep links
- App Store redirect URLs
- deferred deep links
The UTM parameters move with the user into the store.
### 4\. When the app opens, Airbridge attributes the install back to the AI source
The SDK reads the deep link context and assigns the install correctly:
> **Source:** chatgpt
> **Medium:** organic\_ai
> **Campaign:** recommendation
## **Future Outlook: The Next 12 Months of AI-Driven Discovery**
LLM-driven discovery is still early, but the trajectory is unmistakable: AI assistants are becoming a primary gateway for how people decide which apps to download next. Over the coming year, this shift will accelerate — and it will reshape how mobile marketers think about visibility, optimization, and measurement.
Here’s what’s coming, possibly.
### 1\. LLMs will become real acquisition channels
AI platforms are already experimenting with embedded links, app suggestions, and sponsored answers. Within 12 months, marketers should expect:
- attribution-ready AI referral surfaces
- paid placement models
- AI-first recommendation ad formats
### 2\. “AI Store Optimization” workflows will emerge
Just as ASO and SEO matured into core marketing functions, GEO (Generative Engine Optimization) will evolve into a structured discipline. Teams will optimize:
- metadata for AI comprehension
- structured content for model parsing
- brand authority for recommendation ranking
### 3\. MMPs will integrate directly with AI platforms
To give marketers better visibility, measurement partners will begin:
- ingesting AI-driven referral metadata
- supporting dedicated AI source types
- introducing attribution windows optimized for AI-driven journeys
Visibility will become the differentiator between modern and legacy measurement stacks.
### 4\. AI-native ad formats will mature
Expect to see:
- conversational CTAs
- personalized app suggestions
- in-chat deep links
- intent-based recommendation units
These formats will outperform traditional banners because they appear inside the user’s decision moment.
---
## Drive Smarter Web Optimization with Airbridge & Moloco Advanced Web Postback Integration
URL: https://www.airbridge.io/en/blog/moloco-enhanced-web-postback-integration
Language: English
Published: 2025-12-23
Airbridge now supports Moloco’s Advanced Web Postback, combining Airbridge’s unified measurement and user-level analytics with Moloco’s high-performance web advertising.
With this integration, Moloco web advertisers can expect more accurate performance measurement and refined optimization across the entire web funnel. This integration offers significant performance benefits, especially for financial and e-commerce web-to-app campaigns, and for PC/console games where the user journey from a web landing page to an install or signup is crucial.
Now, Moloco web advertisers using Airbridge can measure, analyze, and optimize campaigns at the user level, leveraging the third-party cookie IDs passed via postbacks while adhering to the latest privacy standards. Currently, this third-party cookie-based identification is applicable to campaigns running on Android and PC environments.
## Unlock Performance-Focused Web Growth with Moloco
As an AI-powered advertising platform, Moloco helps global advertisers acquire and retain high-value users, while providing transparent data that enables marketers to clearly see where users are coming from and optimize their campaigns with confidence.
Historically, standard web integrations only passed event and conversion data, making optimization via signal transmission difficult.
Now, with Moloco Advanced Web Postback integrated into Airbridge, advertisers gain access to granular, user-level data across the entire funnel. This enables web marketers to:
- Access both event/conversion data and Moloco’s third-party cookie IDs
- Run more precise optimizations backed by user-level signals based on third-party cookies
- Analyze performance holistically across cohorts, channels, and paths
- Power smarter bid strategies and automated adjustments based on real post-conversion behavior
- View performance for both pure web and web-to-app campaigns concurrently
## Enhance Web Postback Attribution with Airbridge
Airbridge’s measurement infrastructure is built to unify fragmented data, verify true performance, and connect every touchpoint—from impression to lifetime value. By pairing Moloco’s ML-driven optimization with Airbridge’s transparency-first measurement, growth teams benefit from:
- Gain full visibility into web performance: Easily track Moloco web traffic and events alongside other ad sources for a streamlined performance overview.
- Enjoy effortless unified tracking: Airbridge tracking links cover all activity on both web and mobile apps. Connect your links with Airbridge Web SDK to analyze deep link activations, web events, web-assisted installs, and web-to-app conversions—all in one place.
- Achieve true web campaign optimization: Seamlessly optimize for web conversions using user-level signals based on third-party cookies.
- Automate UTM tracking: The Airbridge Web SDK automatically reads UTM values within the link to accurately record the campaign source.
## Getting Started
Setting up the Moloco Advanced Web Postback integration is straightforward:
1. Configure [Moloco postbacks](https://help.airbridge.io/en/guides/moloco#postback-configuration) in Airbridge based on your conversion goals.
2. Enable [Moloco cookie ID collection](https://help.airbridge.io/en/developers/web-sdk#moloco-cookie-id-collection) via the Airbridge Web SDK.
3. Start analyzing user-level events and campaign performance in the Airbridge dashboard.
## Power Growth with Intelligence, Transparency, and Optimization
The Airbridge and Moloco Website Postback integration is designed for growth teams looking to combine performance-level transparency with machine-learning powered optimization across web funnels.
This powerful solution is ideal for teams focused on web optimization. By leveraging this integration, advertisers can efficiently manage the entire user journey—from optimizing web landing pages for better traffic inflow to seamlessly tracking how that traffic converts into web-to-app performance and successful retargeting outcomes.
Ready to achieve exceptional results in web user acquisition and retargeting? Start leveraging the combined power of [Airbridge](https://www.airbridge.io/book-a-demo) and Moloco today.
---
## Airbridge & Xsolla: New S2S integration unifies mobile and web performance insights
URL: https://www.airbridge.io/en/blog/xsolla-airbridge-integration
Language: English
Published: 2025-12-17
**Airbridge** has partnered with **Xsolla**, a global video game commerce company, to launch a new server-to-server (S2S) integration that connects mobile campaign data with purchases made through the Xsolla Web Shop.
As more players buy in-game currency, virtual goods, and exclusive offers through web shops, these off-platform transactions now drive a growing share of total revenue. Yet for most developers, it’s been difficult to connect them back to mobile campaigns.
With this integration, web shop purchases are securely sent to Airbridge as S2S events and attributed to the correct campaign or install source. Developers can finally see how mobile ads drive web-based revenue and understand the complete player journey—from ad click to install to purchase.
## What developers can do with the Airbridge–Xsolla integration
This integration brings mobile and web data together, helping studios understand what’s working and where to grow. With Airbridge and Xsolla, developers can:
- **Track web shop purchases alongside in-app events.** Attribute web transactions to the right campaigns and see how players move between app and web.
- **Measure revenue across all campaigns.** Identify which channels and creatives deliver high-value users, including those who convert later on the web.
- **Calculate true player LTV.** Combine in-app and web data to measure total Lifetime Value (LTV) and optimize spend with confidence.
With full-funnel attribution and over 200 customizable metrics, Airbridge helps teams monitor ROAS, retention, revenue, and churn in one place—removing data silos and improving decision speed.
## Bridging mobile and web performance
As studios embrace direct-to-consumer (D2C) models like web shops, they gain more control over pricing, branding, and player relationships. But this also creates new data gaps.
Airbridge fills those gaps by tracking Xsolla Web Shop purchases alongside in-app events, ensuring every transaction counts toward your performance metrics. Developers can clearly see how marketing efforts translate into real revenue—across both mobile and web.
With predictive lifetime value (pLTV) and predicted lifetime (pLT) modeling, Airbridge helps teams identify high-value players early, predict churn, and act before performance declines.
## Smarter cross-platform growth with Airbridge
The Airbridge–Xsolla integration gives studios a unified, data-driven view of player engagement and revenue across platforms.With deeper insight into ROAS, LTV, and player behavior, teams can align UA, product, and data decisions to drive long-term growth.
---
## Connect acquisition to revenue: How the Airbridge × RevenueCat Integration helps subscription apps measure real ROI
URL: https://www.airbridge.io/en/blog/abr-revenuecat-integration
Language: English
Published: 2025-12-15
Most subscription apps don’t fail at acquisition — they fail at understanding which users become profitable in the long-term. To answer that, you need to connect where a user came from with how they behave across their entire subscription lifecycle.
The Airbridge × RevenueCat integration brings these two data sets together, so teams can attribute trials, renewals, churn, and lifetime value back to the exact campaign or channel that drove the user to subscribe.
# From Visibility to Actionable Insights
Having access to subscription data isn’t enough to understand growth. To measure ROI accurately, marketers need to know where paying users came from and how their lifetime value evolves.
RevenueCat provides a reliable foundation for managing in-app subscriptions. It connects directly with app stores to process billing and generate accurate server-side events such as trial start, purchase, renewal, billing issue, and expiration.
Airbridge complements this by connecting those events to their acquisition sources. With both datasets combined, subscription teams can answer practical questions such as:
- Which UA channels drive stronger trial-to-paid conversion?
- Do Meta campaigns deliver better renewal rates than TikTok?
- Which creatives bring in subscribers who retain through Day 180?
Insights like these become even more actionable when paired with industry benchmarks such as the [RevenueCat State of Subscription Apps Report](https://www.revenuecat.com/state-of-subscription-apps-2025/), which provides broader context on trial, renewal, and churn patterns across subscription apps.
# Turning Connected Data into ROI
With lifecycle events and attribution now in one place, teams can compare channels by renewal rate, churn, and lifetime value — gaining a clear view of which marketing efforts sustain long-term revenue. Through the integration, marketers can:
- Attribute recurring revenue to the campaign, channel, or creative that acquired each subscriber.
- Monitor key lifecycle metrics such as renewal rate, churn, and LTV directly on Airbridge’s unified dashboard.
- Leverage cohort analysis to uncover which campaigns and user segments drive stronger retention and long-term revenue.
- Evaluate true marketing ROI by comparing subscription-driven revenue against campaign spend.
- Optimize budgets and creatives using data that reflects subscriber quality, not just conversion volume.
> “Airbridge has built a great attribution platform, and pairing it with RevenueCat’s subscription data unlocks a level of insight that teams simply haven’t had before. We’re excited to partner together to help apps grow with more confidence and clarity.”
> — Matt Berry, Head of Partnerships, RevenueCat
# Getting Started
Activating the Airbridge × RevenueCat integration takes just a few steps.
- **Check prerequisites:** Make sure your RevenueCat SDK is set up and ensure user IDs match between RevenueCat and Airbridge.
- **Get your credentials from Airbridge:** Copy your Airbridge Subdomain and Token from the RevenueCat integration page in the Airbridge dashboard.
- **Provide credentials to RevenueCat:** Paste your Airbridge Subdomain and Airbridge Token on the Airbridge integration page in the RevenueCat dashboard, set Event names and then click Add integration to complete setup.
Once connected, RevenueCat will automatically send lifecycle events to Airbridge via a secure server-to-server connection.
Learn more about integrating RevenueCat with Airbridge [here](https://help.airbridge.io/en/guides/revenuecat).
Ready to unlock deeper subscription insights and measurable ROI? Contact our team!
---
## Meta AEM: View-Through Attribution Now Supported in Airbridge
URL: https://www.airbridge.io/en/blog/meta-aem-view-through-attribution
Language: English
Published: 2025-12-04
Airbridge now supports view-through attribution for [Meta’s Aggregated Event Measurement (AEM)](https://www.facebook.com/business/help/721422165168355?id=1877298665783613) on eligible iOS App Promotion campaigns. This expands reporting beyond existing click-through attribution, helping to give you clearer visibility into how impressions contribute to installs. As a result, your Meta campaign measurement becomes more complete and more accurate.
This update applies to all AEM-eligible iOS campaigns starting October 27, 2025.
## What is Meta’s AEM?
Meta’s Aggregated Event Measurement (AEM) is a protocol for measuring web and app events from devices running iOS 14.5 and later. AEM provides near real-time reporting, enabling faster system optimization and more granular performance insights.
## What do I need to do to enable AEM campaign view-through results?
If your campaigns are already enabled for AEM, no additional setup is required.
If you have not yet enabled AEM, please refer to the [Airbridge integration guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem) for setup instructions.
## Do I need to select AEM when setting up an App Promotion campaign to see view-through data in Airbridge?
No. Airbridge will receive AEM view-through data automatically for eligible campaigns. To confirm whether your app events are AEM-eligible, please refer to the [Meta Business Help Center](https://www.facebook.com/business/help/617021863580582?id=1877298665783613).
## With this update, what will change in Airbridge?
Airbridge will now automatically receive and report view-through touchpoints in addition to existing click-through attribution. You’ll see impression-driven installs reflected across the following reporting surfaces:
- Attribution reports
- Analytics dashboards
- Raw data exports
Note: SKAN data will continue to be collected and attributed separately.This update does not affect SKAN reporting.
## Attribution Windows
- Meta supports up to a 1-day view-through attribution window for installs.
- Airbridge applies a global 6-hour probabilistic view-through window.
To minimize reporting discrepancies, you can adjust Meta Ads attribution rules in your Airbridge dashboard.
For step-by-step setup guidance, refer to the [AEM integration guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem) or reach out to our [support team](https://help.airbridge.io/en/supports/requests) for assistance.
---
## The CPP Revolution in 2025: Personalizing App Store Experiences for Precision Growth
URL: https://www.airbridge.io/en/blog/the-cpp-revolution-in-2025-personalizing-app-store-experiences-for-precision-growth
Language: English
Published: 2025-11-24
In the age of privacy-first marketing, where signal loss has become the norm, growth teams are being forced to rethink what true optimization means. Marketers can no longer rely solely on granular audience targeting or conversion data. Instead, the focus is shifting toward personalizing the user experience itself, from the ad creative all the way to the App Store.
This is where **Custom Product Pages (CPPs)** are rewriting the rules of user acquisition.
Once seen as a niche App Store feature, CPPs have become one of the most powerful levers for improving conversion, creative alignment, and user quality. For marketers navigating the post-ATT landscape, mastering CPPs means reclaiming control, not through more targeting, but through smarter storytelling.
## **What Are Custom Product Pages, and Why They Matter Now**
Introduced in late 2021, Apple’s **Custom Product Pages** allow marketers to create multiple, highly targeted versions of their App Store page, each optimized for a specific audience, campaign, or creative theme. Each CPP can feature unique screenshots, app previews, and promotional text, enabling precise alignment between ad messaging and store experience.
And these 2025 updates at [**Apple’s WWDC 2025**](https://developer.apple.com/videos/play/wwdc2025/328/) are what make CPPs a critical growth lever for marketers today. CPPs are no longer limited to paid traffic, and Apple now allows developers to assign organic keywords to individual CPPs, letting them appear in App Store search results.
In other words, CPPs have officially evolved from performance-driven landing pages into **discoverability assets**, merging the worlds of paid user acquisition and App Store Optimization (ASO).
### **Simple example of CPPs**
A gamer clicking an ad for “PvP battles” lands on a page highlighting combat modes, while another searching organically for “city-building games” sees a CPP emphasizing creativity and relaxation. Both experiences are personalized, intent-driven, and designed to convert.
This alignment of _message and motivation_ is what makes CPPs such a powerful conversion lever in 2025.
## **The Real Impact of CPPs : Bridging Creative and Conversion**
### **1\. Closing the Creative Disconnect**
Before CPPs, marketers faced a frustrating gap between ad creative and store experience. A user might click a high-performing ad only to land on a generic App Store page that failed to deliver on the creative’s promise. That disconnect led to drop-offs, lower conversion rates, and weaker installs.
CPPs eliminate that friction and help marketers maintain narrative consistency, a psychological principle known as **message match.** According to [**Adaptly**](https://adaptly.dev/blog/ppc-message-match-guide), it not only improves conversion rate but also **reduces CPCs by 30% or more** and elevates user quality via attracting audiences aligned with the app’s true value.
### **2\. Turning CPPs into Conversion Labs**
Today’s CPPs aren’t just landing pages, they’re data-rich micro-experiments.
This deeper analytics layer, introduced in 2025, allows growth teams to go beyond superficial conversion data to now measure each CPP’s performance across **100+ new App Store Connect metrics**, eventually tracking how different creative-CPP pairs influence retention, in-app engagement, and monetization.
Example: a meditation app might discover that “stress-relief” CPPs yield higher Day 1 conversions, while “sleep improvement” pages generate stronger long-term retention. CPP insights now feed directly into creative strategy, product messaging, and LTV optimization.
### **3\. Connecting CPPs to the Broader Funnel**
The magic happens when CPPs are connected to **post-install behavior**. With a measurement partner like **Airbridge**, marketers can attribute installs, in-app conversions, and retention metrics back to the CPP link that drove them.
Moreover, [**as Apple said**](https://developer.apple.com/help/app-store-connect/create-custom-product-pages/configure-multiple-product-page-versions), since iOS 18, CPPs can include [**deep links**](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers) that bring users directly to relevant in-app experiences after installation. That shortens the path to value and strengthens retention from the very first session, a critical step for improving D1 and D7 engagement metrics.
## **But... only 31% of top apps use CPPs in 2025**
Despite these capabilities, CPP adoption remains surprisingly low.
According to [**AppTweak’s 2025 ASO Benchmarks Report**](https://www.apptweak.com/en/aso-blog/aso-app-store-trends-benchmarks-report), only 31% of top apps and 26% of top games actively use Custom Product Pages. Yet, those that do report an average conversion lift of 8.6%, proving the untapped potential for brands willing to invest in personalization.
For performance marketers, this gap represents a rare advantage: a feature still under-utilized by competitors but already delivering measurable ROI.
## **How Leading Growth Teams Use CPPs**
### **1\. Aligning Creatives with User Intent**
High-performing apps treat CPPs as narrative extensions of their ads. A mobile RPG might deploy different CPPs per ad theme: “PvP combat,” “hero collection,” or “idle rewards.” A fintech or banking app could segment pages for bank transfer, investing, saving, and credit management use cases.
By mapping each ad creative to a matching CPP, these marketers build a **coherent intent funnel**, where every user lands on a page that reflects exactly what they were promised in the ad.
[The CPP Revolution in 2025: Personalizing App Store Experiences for Precision Growth]
### **2\. Running Hypothesis-Driven Experiments**
CPPs enable scalable A/B testing across value propositions, visuals, and tones.
For example:
- Does showing real user testimonials increase CVR more than product UI screenshots?
- Does “Save Smarter” resonate better than “Invest with Confidence”?
- Are reward-oriented headlines more effective for first-time users?
Over time, CPP data reveals which emotional triggers, visuals, or tones convert best, feeding back into broader campaign strategy and ASO optimization.
### **3\. Launching Campaign-Specific or Seasonal Pages**
Beyond ad alignment, CPPs shine in **short-term campaigns**. Seasonal events, influencer collaborations, and feature launches all benefit from dedicated CPPs that spotlight relevant content.
This flexibility allows brands to stay agile, responding quickly to market trends, holidays, or regional interests without altering the default App Store listing. And with Apple’s 2025 update enabling organic discovery, campaign-specific CPPs can even rank for timely keywords like “Black Friday deals”, “Holiday Shopping” or “new season update.”
[Illustration of holiday-themed CPPs. Source: Newtonco]
## **The Future of CPPs: Contextual, Cohort-Driven Personalization**
The next wave of CPP innovation lies in **dynamic personalization**, where content adapts to user context, behavior, or audience cohort. As contextual and cohort-driven personalization becomes more accessible, CPPs will start behaving less like static pages and more like dynamic storefronts.
> “_Instead of showing the same screenshots to everyone, future CPPs could automatically adapt based on a user’s intent, past behavior, or even real-time signals such as location, device type, or campaign source. This means a user coming from a TikTok creator might see a fun, social-focused CPP, while a returning lapsed user may see a page emphasizing new features and rewards_.” - Airbridge Marketing Team
What it means for marketers:
- **Higher conversion efficiency**: contextual CPPs show each user messaging that feels relevant to their intent, improving install likelihood.
- **Better long-term user quality:** cohort-driven CPPs attract audiences that match your app’s highest-value groups, boosting retention and LTV.
- **Less wasted spend**: marketers avoid paying for mismatched traffic by aligning CPP content with user expectations.
- **Stronger ROAS**: more relevant CPP experiences lead directly to better downstream monetization performance.
- **Smarter scaling:** with clearer cohort signals and higher-quality installs, growth teams can scale campaigns with greater confidence.
## **Leverage with Airbridge: Turning CPP Insights into Growth Intelligence**
While CPPs empower marketers to customize App Store experiences, **Airbridge amplifies their impact by connecting CPP data with full-funnel attribution and user behavior analytics**.
[The CPP Revolution in 2025: Personalizing App Store Experiences for Precision Growth]
Airbridge’s One-click generation of CPP tracking links to measure its effectiveness
By linking each CPP link to install sources, post-install events, and retention metrics, Airbridge enables marketers to understand not just _which CPP converted_, but _which CPP nurtured long-term value_.
> _CPPs tell you who clicked. Airbridge tells you who stayed, paid, and scaled._
**With Airbridge, marketers can:**
- **Attribute installs and post-install events** to the exact CPP that drove them, even in SKAN environments.
- **Compare user quality by CPP variant**, analyzing retention, LTV, and ROAS across creative cohorts.
- **Map the entire funnel**, from ads → CPP → install → in-app purchase, giving a unified view of user journey efficiency.
- **Leverage deep linking** to guide users from CPPs to specific in-app experiences, shortening time-to-value and improving engagement.
---
## Decoding Subscription Monetization: Proven Strategies from 100+ Paywall and Pricing Experiments
URL: https://www.airbridge.io/en/blog/decoding-subscription-monetization-proven-strategies-from-100-paywall-and-pricing-experiments
Language: English
Published: 2025-11-21
The subscription app business has always hinged on recurring revenue, yet for many teams the real challenge now is less about securing the initial sign-up and more about what happens after someone downloads the app. Acquisition pipelines are crowded, paid media is costlier, and free trials alone aren’t driving enough conversions. In this landscape, the biggest leverage lies not in getting users in, but in how you engage them, convert them, and keep them paying over time.
In this latest Jetstream Session, Airbridge CEO Roi Nam spoke with Vahe Baghdasaryan, Founder of Tangent (formerly Paywalls.Design), whose team has helped around 100 subscription apps of all sizes optimize their monetization models within just a year of its launch. Their focus: not quick wins, but sustainable revenue systems built on data, user behavior, and strategic experimentation.
In unveiling his findings, Vahe shows that successful subscription monetization does not rest on a clever price point or a slick screen. It lies in a philosophy. One where every step in the user journey from onboarding to trial to paid to renewal is designed, measured, and iterated.
**🎥 Watch the full interview video**
https://youtu.be/rVNvEdqb9Jg
# The Philosophy Behind Sustainable Monetization
Many teams still tend to treat revenue like a design project — a single paywall to tweak or a pricing table to test. However, Vahe believes that real results come from viewing monetization as a connected system.
A paywall is only one piece of that system. Pricing strategy, offer placement, and lifecycle marketing all influence how users perceive value and decide to subscribe. When these elements are tested in isolation, the impact often fades quickly. When they are connected and informed by shared data, each experiment strengthens the next, creating measurable and repeatable results.
This mindset reframes monetization from a one-off optimization task to a continuous process of learning, testing, and iteration — the same discipline that drives successful product development.
# Paywall Optimization: The First Step to Increasing Subscription Conversions
When it comes to subscription monetization, where should app teams begin? For Vahe it depends on a few fundamentals: the app’s size, its current revenue, and how broken the existing paywall is.
> “You need to understand your numbers before touching anything else — your conversion rate, ARPU, trial-to-paid ratio — the full picture.”
> — Vahe Baghdasaryan, Founder of Tangent.
In most cases, teams begin by testing and optimizing their paywall. This first stage often focuses on improving the onboarding paywall and other in-app paywall placements. Once those foundations are performing well, new opportunities emerge, such as adding paywalls after transaction abandonment, on the home screen, or as win-back prompts for users who cancel their trial.
[Vahe Baghdasaryan, Founder of Tangent.]
## Multi-page Paywalls: Why pairing them with the right offer delivers stronger results
Among recent optimization patterns, one that’s proven especially effective is the multi-page paywall. Instead of hitting users with one paywall that says everything on a single screen, this format spreads the experience across a few lightweight steps helping users understand value before facing cost.
A common multi-page paywall looks like this:
1. **Welcome screen:** A welcome page invites users to “try the app for free”, often paired with a short demo video that highlights key benefits.
2. **Reassurance screen:** The next screen builds trust with a simple message such as “We’ll remind you two days before your trial ends.”
3. **Progress or benefit screen:** A follow-up screen may show progress or reinforce value before moving users to the final paywall.
4. **Final paywall:** The final screen presents the subscription options, typically annual and monthly plans.
However, Vahe emphasizes that it becomes far more effective when coupled with the right offer. One proven example is the paid-trial annual plan. Instead of offering both monthly and annual subscriptions with a free trial, the app presents two annual options: one with a standard free trial and another where users pay a small amount, such as $1.99, for the first month before the full annual billing begins. This paid-trial option often comes with a slight annual discount, making the commitment feel even more compelling.
This simple change reframes the trial as a low-risk commitment rather than a free giveaway. Users who pay even a small fee tend to be more engaged and less likely to churn after the trial.
> “A well-built multi-page paywall can deliver around 10% uplift in conversion, but when paired with a compelling offer, that lift can rise to 15–20%. It’s not about changing layouts or using tricks. It’s about finding the right packaging for your premium product with an offer that truly fits the user.”
> — Vahe Baghdasaryan, Founder of Tangent.
> **Running pricing experiments without channel-level data?** You need to know if a pricing change lifts ARPU on Meta traffic but drops it on Google — before you roll out globally.[\*\*
> See Install → Trial → Subscribe by channel. Start with 15K free attributed installs →\*\*](https://coreplan.airbridge.io/waitinglist/c)
# Pricing Optimization: Avoiding the Biggest Pitfalls
Pricing can be one of the most powerful levers for subscription growth, yet it’s also the easiest to get wrong. Vahe cautions that pricing isn’t something to test halfway. Either run the test the right way, or don’t do it at all.
## Subscription Pricing Tests Require Patience (Looking Beyond D7, D30 Metrics)
Many teams rush to test price changes expecting quick results, but short-term data rarely tells the full story. Vahe cautions that the biggest mistake teams make is drawing conclusions too early. Many apps decide on new pricing after just a week or two of testing, basing the decision on D7 or D30 ARPU. But these short windows rarely correlate with lifetime value.
> “With paywall or placement tests, even if you make a wrong call, you can notice it quickly and revert. But with pricing, it’s super dangerous. You don’t know your refund rate, your renewal rate, or even disputes if you stop the pricing test after a week. It’s so complicated that without enough time, you’ll never know how it truly affects your business.”
> — Vahe Baghdasaryan, Founder of Tangent.
If you’re concerned about the long testing cycle, Vahe suggests a practical workaround: take 20% of your traffic, run the price experiment on that segment, and let it sit untouched for about three months. The remaining 80% of users can continue through other ongoing experiments.
## Localized Pricing Isn’t Always a Shortcut
Localized pricing is usually an effective strategy for subscription apps expanding into diverse markets. However, a common oversight among many teams is failing to consider how pricing changes actually affect conversion behavior. When prices drop, conversion rates need to rise enough to offset the discount, otherwise, total revenue may decrease.
To minimize risk, teams should always run A/B tests and analyze how ARPU changes before adjusting prices across markets.
# Upsell Techniques: Building Revenue Ladders That Go Beyond “Annual”
Upsells can be considered one of the most underrated monetization techniques in the subscription world. Vahe notes that fewer than 10% of top apps on the App Store actively use upsell strategies, leaving a major source of incremental revenue untapped.
Many teams still view the annual plan as the finish line, but lifetime packages or family plans can certainly work as next steps up in the revenue ladder.
> “One of the most interesting strategies we started to test recently is offering a lifetime purchase upgrade right after users make the first purchase. We showed them another offer: ‘If you want, you can upgrade to lifetime. Just pay another $50 and you’ll have the product forever’. Surprisingly, around 10% of users took the offer.”
> — Vahe Baghdasaryan, Founder of Tangent.
Beyond these simple upgrades, apps can move toward more advanced upsells by launching new features or complementary products for their existing base. These offers tend to perform particularly well among older demographics, who typically have higher purchasing power and are more willing to invest in extending the experience. And once that foundation is built, the opportunities for one-time purchases become nearly endless.
The key to building effective upsell packages is to understand user insights and not be afraid to test new offers. Sometimes, upselling can be as simple as asking users whether they would like to upgrade. For teams already using tools such as Superwall, RevenueCat, or other paywall builders, these tests can be set up and launched within a few hours, provided that product IDs are already configured in the App Store or Play Store.
[Roi Nam, Co-founder & CEO of Airbridge andVahe Baghdasaryan, Founder of Tangent.]
# Turning Experiments into Scalable Growth
Subscription monetization isn’t about finding a perfect paywall layout or a magic price point. It’s about building a connected system where every experiment from onboarding flow to offer design, pricing, and upsell — feeds the next through disciplined measurement and iteration. The teams that win are those who test, learn, and optimize with structure, not chance.
That’s where Airbridge helps subscription apps turn monetization experiments into a repeatable, scalable system.
[__wf_reserved_inherit]
## Subscribe to success with Airbridge
Vahe's core message is clear: monetization is a system, not a single paywall test. But a system only works when each experiment is connected to the data that tells you what changed — and where.
Airbridge Core Plan is the measurement layer that connects paywall, pricing, and trial experiments to channel-level outcomes:
- **See which channels drive subscribers who convert at your paywall.** Core Plan attributes Install, Start Trial, and Subscribe by channel — so you know whether a paywall change lifted conversions from Meta, Google, or Apple Search Ads.
- **Evaluate pricing changes with real subscription data.** The Actuals Report shows per-channel conversion rates at each funnel stage. When you change pricing, you can see which channels respond — instead of reading blended averages that hide the signal.
- **Validate trial quality by channel.** Are your TikTok trials converting to paid at the same rate as your Google trials? Core Plan shows trial-to-subscribe rates by channel, so you can distinguish signal from noise.
- **Connect subscription data without custom event work.** Native RevenueCat and Adapty integration via S2S sends subscription events into attribution automatically — no custom event setup, no engineering sprint.> [**Start with 15K free attributed installs. All features included. →**](https://coreplan.airbridge.io/waitinglist/c)
---
## Power Rewarded UA with Airbridge & Gamelight Integration
URL: https://www.airbridge.io/en/blog/airbridge-gamelight-integration
Language: English
Published: 2025-11-20
Airbridge is now integrated with Gamelight, the leading rewarded user acquisition (UA) platform. This integration enables app and game developers to measure, analyze, and optimize rewarded UA performance with full transparency — from first install to long-term user value.
# **Drive User Growth with Gamelight’s Rewarded UA Platform**
[Gamelight](https://www.gamelight.io/) is a leading rewarded mobile marketing platform that helps app publishers and advertisers optimize their UA strategies. With its focus on Rewarded UA, Gamelight allows developers to engage users in a natural and rewarding way by recognising them for their in-game actions and achievements. Using real-time data and machine learning, the platform personalises user experiences to boost engagement and retention.
With Gamelight, you can:
- Access 100% direct traffic for transparent, fraud-free performance.
- Leverage advanced targeting to reach the right users based on age, genre, and competitor titles.
- Launch rewarded campaigns through a self-serve dashboard and start driving results in minutes.
- Focus on retention, LTV, and ROAS optimization, not just installs.
[__wf_reserved_inherit]
# **Amplify Rewarded Campaigns with Airbridge Measurement**
The Airbridge x Gamelight integration bridges rewarded engagement with unified measurement, empowering marketers to analyze performance with event-level granularity. By connecting Gamelight’s rewarded UA capabilities to Airbridge’s unified measurement infrastructure, teams can now capture the full user journey, from ad exposure and install to post-reward retention and monetization. With this integration, marketers can:
- Review KPIs such as CPI, ROAS, and LTV within a unified dashboard to monitor campaign impact instantly.
- Identify which campaigns drive the highest-value users, using event-level tracking to pinpoint performance by cohort or channel.
- Use funnel and cohort analysis to evaluate user engagement and retention, understanding where users engage, convert, or drop off.
- Make data-driven campaign adjustments with real-time reporting.
This joint solution provides a powerful end-to-end solution: from acquisition to attribution to performance optimization, enabling teams to refine their UA strategies and sustain long-term player growth.
> “The future of user acquisition lies in transparency and intelligence. Partnering with Airbridge brings both — uniting scalable rewarded marketing with a new standard of accurate, privacy-first measurement.”
> — Günay Azer, Founder of Gamelight.
## **Getting Started**
Setting up the Gamelight x Airbridge integration is simple:
1. Set up Gamelight campaign tracking links in Airbridge.
2. Configure postbacks for real-time data exchange based on your conversion goals.
3. Start tracking rewarded UA performance directly in the Airbridge dashboard.
Please refer to the Airbridge User Guide for detailed instructions and information. If you would like to find out how Airbridge and Gamelight can help you power your campaigns, get in touch with us today!
---
## 3 Monetization Flow Best Practices for Subscription Apps in 2026
URL: https://www.airbridge.io/en/blog/3-monetization-flow-best-practices-for-subscription-apps-in-2026
Language: English
Published: 2025-11-14
### **Why Monetization Flow is Changing**
The mobile subscription economy is shifting quickly, especially in competitive verticals like health, fitness, and lifestyle. With rising acquisition costs, success depends on turning installs into paying users early in the journey.
Slow conversion cycles no longer work, and apps must capture intent the moment users land. The urgency is clear: 77% of daily active users drop off within three days, and most uninstall within a week, according to [Business of Apps](https://www.businessofapps.com/guide/mobile-app-retention/). In this environment, an effective monetization flow isn’t optional, but essential for survival and growth.
### **What is Monetization Flow and Why is It Important**
Monetization flow refers to the structured journey that guides a user from their first interaction with an app toward becoming a paying customer. It covers everything from onboarding screens and paywall placements to pricing offers and retention tactics. A well-structured subscription funnel doesn’t just aim for quick conversions, it creates a seamless experience that feels natural and valuable to the user while driving sustainable revenue for the business.
[3 Monetization Flow Best Practices for Subscription Apps in 2026]
This matters because users arrive as **high-intent leads**. By the time they hit “install,” they’ve already done the app store research, compared reviews, and chosen your app. That makes the very first interactions inside the product the prime time for monetization. If developers miss this window, they risk losing users before they’ve had the chance to showcase real value.
There’s also a simple truth that supports strong subscription monetization strategies: “**people who pay, pay attention.**” Paid users don’t just generate revenue, they also engage more deeply, retain longer, and contribute higher lifetime value compared to free users. In other words, monetization and engagement go hand in hand.
[3 Monetization Flow Best Practices for Subscription Apps in 2026]
At the same time, optimized subscription monetization flow strikes the right balance between aggressive paywall strategies and building user trust. It ensures that users feel guided, not pressured, while being nudged toward premium plans or paid tiers naturally. Monetization isn’t just about maximizing revenue; it’s also about improving retention, enhancing product stickiness, and ultimately creating a healthier app ecosystem.
### **Best Practices for Monetization Flow**
This is used successfully by successful apps, especially in the Health & Fitness categories. Specifically, according to [Steve](https://www.linkedin.com/in/stevepyoung/), Founder at App Masters, an app called [Flo Period & Pregnancy Tracker has mastered its monetization flow](https://www.youtube.com/watch?v=ODWgul9tXus&t=1s) to the point of generating an astonishing **$6 million in monthly subscription revenue** from around **one million downloads per month**.
Here are some tips that top app developers are leveraging:
### **1\. Onboarding & User Psychology**
- **One-question-at-a-time onboarding** → reduces overload and increases completion rates.
[3 Monetization Flow Best Practices for Subscription Apps in 2026]
- **Personalization cues** (like using the user’s name) build emotional connection.
- **Social proof + empathy statements** (“you’re not alone”) boost trust.
- **Feature integrations before the paywall** → asking for permissions early increases conversions.
### **2\. Paywall Strategy**
- **First paywall during onboarding** → 60–80% of purchases happen at first app use.
- **Two-step paywall with discount** (initial paywall→ discounted annual plan) increases conversions.
- **Free trial toggle on paywall** → creates psychological commitment, reduces cancellations.
- **Paywall before signup** → one client saw a 52% sales increase when swapping order.
- **Multiple paywall views** → more touchpoints = higher revenue (onboarding, signup, locked content, re-open).
[Example of Multiple paywall views. Source: Flo]
- **Second open “micro-offer”** → unlock for $0.99, renewed at $X/year; works because “people who pay, pay attention”.
### **3\. Content Locking & Engagement**
- **Locked content strategy** → show value by previewing features, but lock access to drive upgrades.
- **Multiple locked tabs** (insights, secret chats, messages) remind users what they’re missing.
### **Leverage Monetization Flow with Airbridge**
Designing a strong subscription monetization flow is only half the battle; the rest is measuring and refining it. Most apps still face rising acquisition costs, fragmented user journeys across iOS, Android, and SKAN, and the challenge of knowing which paywall or offer actually works. Without the right measurement, even the best-designed flow risks leaving money on the table.
That’s where Airbridge comes in. As an MMP, here is how Airbridge empowers app teams:
- Airbridge enables marketers to measure paywall performance across platforms, including **Android, iOS, and SKAN**, with Funnel Report, thus ensuring data clarity for A/B testing.
- With **granular funnel and ROAS analysis**, teams can see which monetization flow steps (e.g., free trial toggle, second paywall, micro-offer) deliver the best ROI.
- Cost data integration helps balance **user acquisition costs** with monetization outcomes, ensuring profitability.
- Airbridge provides the **cross-platform, cross-channel** insights needed to replicate the Flo app’s success at scale.
---
## Web-to-App vs App-to-Web: Why Subscription Apps Need Both Funnels — and How to Measure Them
URL: https://www.airbridge.io/en/blog/web-to-app-vs-app-to-web
Language: English
Published: 2025-10-27
Every subscriber you acquire through in-app purchase loses 15–30% of their revenue to platform commissions before it reaches you. On a $14.99/month plan, Apple takes $4.50 in year one. Over 12 months, that is **$54 per subscriber — gone before you count a single cost.**
You could route payments to your website instead. Stripe charges ~3%. On the same $14.99/month plan, that is $5.22/year instead of $54. **You save $48.78 per subscriber per year.**
But here is the problem: **when payment moves to web, attribution breaks.** The ad platform sees the install. Stripe sees the payment. No one sees both. You cannot answer the most important question: which channel produces subscribers who actually pay on the web?
**Key Takeaways**
- **In-app purchases are frictionless but expensive.** Apple takes [30% in year one, 15% after](https://developer.apple.com/app-store/small-business-program/) (or 15% flat for Small Business Program). Google Play takes 15% on subscriptions from day one.
- **Web payments save margin but break measurement.** Routing users to web checkout saves 12–27% in commissions — but without cross-platform attribution, you lose sight of which channel drove the subscription.
- **The most successful subscription apps use both.** Spotify, Netflix, Duolingo, Bumble, and Noom each blend web and app touchpoints — using the app for engagement and the web for monetization.
- **The attribution blind spot is the real cost of hybrid funnels.** Teams that cannot measure both paths either overpay commissions (staying in-app) or fly blind on channel performance (going web).
- **Airbridge Core Plan connects install source to subscription outcome** — whether the user paid in-app or on web. Start with 15K free attributed installs.
## The 30% Tax on Every Subscriber
Before choosing a funnel, understand what each path costs.
[__wf_reserved_inherit]
[__wf_reserved_inherit]
**The difference between Apple year-one and Stripe is $31.61 per subscriber per year.** For an app with 10,000 subscribers, that is $316,100 annually — enough to fund an entire UA team.
This is why the web-to-app vs app-to-web decision is not a UX question. **It is a revenue architecture decision.**
## Two Funnels, Two Economics
### **Web-to-App: Acquire on Web, Engage in App**
Think of a fitness ad on Instagram promising, “7 days to your healthiest self.” Instead of sending you straight to an app store, it leads to a clean landing page. The page loads quickly, explains the value, and ends with a simple call-to-action: “Continue in the app.” With one tap, you download. Inside the app, onboarding collects your preferences, shows you a quick win, and then presents the subscription paywall.
This is **web-to-app** in practice.
Here’s the play:
1. A user discovers the brand through SEO, ads, or content marketing.
2. They land on a mobile-optimized landing page with clear messaging.
3. The page highlights benefits and shows a strong call-to-action.
4. A deep link directs them to the App Store or Google Play.
5. Inside the app, onboarding flows personalize the experience.
6. The subscription paywall is presented at the right moment.
For marketers, **the purpose is clear:** **acquisition**. Users often find apps through search engines, paid ads, or blog content. The web is where curiosity grows, but the app is where habits take hold. Mobile apps offer push notifications, gamified streaks, and personalized onboarding flows that keep people engaged in ways the mobile web cannot match.
- **Advantage:** Frictionless payment. Apple Pay, one-tap purchase, no leaving the app. The conversion moment happens where the user is already engaged.
- **Disadvantage:** Every subscription goes through the app store. You pay 15–30% on every transaction, every month, for the lifetime of the subscriber.
- **Best for:** Apps where conversion rate is the bottleneck. If your trial-to-paid rate is low and every friction point costs you subscribers, the seamless in-app purchase flow matters more than the commission savings.
### **App-to-Web: Install Free, Pay on Web**
Now flip the script. A user downloads a productivity app, enjoys a free trial, and then sees a prompt to upgrade. Instead of handling payment inside the app, the upgrade button sends them to a web checkout page. There, they find flexible plans, more payment methods, and sometimes even lower pricing. Once the purchase is complete, premium access is activated instantly in the app.
This is **app-to-web**.
Here’s the flow:
1. A user downloads the app and starts using free features or a trial.
2. The app prompts them to upgrade after usage or at the end of the trial.
3. A call-to-action routes them to a secure web checkout page.
4. The user chooses between multiple subscription plans.They complete payment using flexible methods such as credit card, PayPal, BNPL, or local wallets.
5. Premium access is unlocked and synced back into the app.
**The purpose here is monetization.** By shifting payments to the web, brands **avoid app store fees of 15 to 30%**. More importantly, they gain freedom to test offers, create bundles, and use payment methods that fit local markets. At scale, the savings and flexibility are significant.
[__wf_reserved_inherit]
**Advantage:** Commission drops from 15–30% to ~3%. On a $14.99/month plan, you keep $14.55 instead of $10.49–$12.74.
**Disadvantage:** Friction. The user must leave the app, open a browser, enter payment details on a web form. **Every step between the app and the checkout page is a conversion risk.**
**Best for:** Apps with strong engagement where users are already committed before hitting the paywall. If your users complete onboarding and engage with content before the payment prompt, the web redirect friction is lower — because intent is already high.
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Measure both web and app funnels in one dashboard.%%
## **How Spotify, Netflix, and Noom Use Hybrid Funnels**
Here’s the kicker: most successful subscription apps don’t choose. They blend.
### **1\. Spotify**
The music streaming giant has famously mastered the hybrid funnel by prioritizing web-based sign-ups to avoid paying high app store commissions on new subscribers.
- **Web-to-app strategy:** Spotify runs ads on search engines and social media that direct users to its website. There, users can sign up for a premium subscription, often with a promotional offer such as “€0 for 1 month then €11.99 per month after”. Once subscribed, they are encouraged to download the app, with deep links sending them straight to the right place.
- **App-to-web strategy:** The Spotify app itself remains a primary driver of engagement and retention. Users who enjoy the service on a free plan might receive an email or see a targeted ad encouraging them to upgrade to a premium subscription via a link to the Spotify website. This moves the billing off the app stores for higher revenue margins.
[__wf_reserved_inherit]
### **2\. Netflix**
Netflix is a classic example of using the web for sign-ups and the app for engagement. Although Netflix recently reintroduced in-app purchases on some platforms, its hybrid strategy remains prominent.
- **Web-to-app strategy**: For many years, and still commonly today, Netflix directs prospective customers to its website to begin a subscription. The website handles the entire payment and account setup process. Once complete, the user is invited to download the app and sign in to their new account. This was done specifically to avoid paying app store commission fees.
- **App-to-web strategy:** The Netflix app is the primary interface for content consumption, leveraging native features for a seamless, TV-like experience. While many billing functions are now accessible in-app, Netflix still directs users to its website for certain account management tasks, reinforcing a web-based ecosystem for important transactions.
### **3\. Duolingo**
The popular language-learning app effectively uses a hybrid approach to maximize conversions and retain revenue, sometimes offering different pricing on the web versus the app stores.
- **Web-to-app strategy:** Duolingo uses web funnels to acquire new users by offering a full, engaging lesson directly on its website. By providing a tangible, enjoyable experience, it builds trust and demonstrates value before asking the user to commit. At the end of the lesson, the user is prompted to download the app to continue their progress, often with a deep link that recognizes their activity.
- **App-to-web strategy:** For users on the free app, Duolingo's retention team might reach out via email with exclusive web offers for its "Super Duolingo" subscription. This not only bypasses store fees but also allows Duolingo to test different pricing models.
### **4\. Bumble**
The dating app uses a hybrid approach to offer promotional pricing and handle billing outside of the traditional app store system for certain offers.
- **Web-to-app strategy:** Bumble uses quizzes and content on its website to acquire new users with better attribution and lower fees. They can use web campaigns to offer discounts on the initial subscription, which wouldn't be possible through an in-app purchase that is regulated by app store policies. Once a user signs up on the web, they are sent to the app to start using the service.
- **App-to-web strategy:** While users can purchase premium features inside the app, Bumble can also use email marketing to existing users to drive renewals or special promotions to the website, capturing the full revenue. For example, a user with an expired subscription might receive an email for a special "welcome back" price, directing them to the web to re-subscribe.
### **5\. Noom**
The weight loss app built its early growth on a web-to-app funnel, using it as a central pillar of its marketing strategy.
- **Web-to-app strategy:** Noom is famous for its web-based onboarding quiz. The quiz collects personal goals and data, building trust before the app download. Once complete, users are nudged into the app to begin their weight-loss journey.
- **App-to-web strategy:** While the app is crucial for daily engagement, the initial web funnel allowed Noom to gather first-party data for better targeting and optimization. This data can be used to re-engage users through email or other web-based channels, directing them to the web for special upgrade offers.
[__wf_reserved_inherit]
Why the two-step dance? Because acquisition and monetization are different battles. You need the app to build sticky user behavior, but you need the web to protect margins and test offers freely.
It’s like running a café: you want people walking through the door (that’s web-to-app), but you also don’t want to pay a middleman every time someone buys a latte (that’s app-to-web).
## The Attribution Blind Spot: Why Hybrid Funnels Break Measurement
Hybrid funnels save money. They also break your ability to measure what is working.
### What Goes Dark When Payment Moves to Web
When a user sees your Meta ad → installs the app → starts a trial → clicks a web checkout link → subscribes on Stripe, the data splits across three systems:
[__wf_reserved_inherit]
- **Meta** sees the install. It does not know the user subscribed.
- **Stripe** sees the payment. It does not know which ad drove the install.
- **Your app analytics** sees the trial start. It does not know which channel the user came from or whether they converted on web.
**No single system sees the full journey.** Without cross-platform attribution, you cannot answer:
- Which channel produces the highest-LTV subscribers?
- Is your Meta spend driving web conversions or just installs?
- Should you shift budget from Google (where users pay in-app) to TikTok (where users convert on web at higher margin)?
### The Hidden Cost: Flying Blind on Channel Performance
Most teams respond to this blind spot in one of two ways — both expensive:
- **Default to in-app purchase** to keep attribution intact. Cost: 15–30% commission on every transaction.
- **Use web payments** and lose channel-level subscription data. Cost: you cannot optimize spend because you do not know which channels produce paying subscribers.
**The real cost of the attribution blind spot is not the missing data — it is the budget decisions you cannot make.** Every dollar allocated without knowing per-channel CPS is a guess.
## How to Measure Hybrid Funnels Without Losing Attribution
### What You Need to Track Across Web and App
[__wf_reserved_inherit]
**The critical connection is the last row.** Whether the user paid in-app or on web, the subscription event must flow back to the attribution system — tied to the original install source. Without this link, the funnel breaks at the point that matters most.
### How Airbridge Core Plan Connects Ad Click to Web Subscription
Core Plan tracks Install, Start Trial, and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. The Actuals Report shows conversion rates at each funnel stage by channel.
With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, **subscription events flow into attribution regardless of where payment occurred** — in-app or web checkout. The billing platform captures the subscription; S2S sends it to Core Plan; attribution connects it to the install source.
**Result: you can see which channels drive subscribers who pay in-app versus on web — and compare CPS across both paths.**
### Airbridge Core Plan vs Traditional MMP
[__wf_reserved_inherit]
## Use the App for Engagement. Use the Web for Revenue. Measure Both.
Web-to-app gets users into the app. App-to-web gets revenue without the 30% tax. **The combination is how the most successful subscription apps operate.**
But hybrid funnels only work if you can measure both paths. Without cross-platform attribution, you are choosing between margin (web payments) and data (in-app payments). **The fix is connecting both to the same attribution system — so you can optimize for CPS, not just CPI, regardless of where the user pays.**
[__wf_reserved_inherit]
> [**See which channels drive subscribers — whether they pay in-app or on web. Start with 15K free attributed installs on Airbridge Core Plan.**](https://coreplan.airbridge.io/waitinglist/c)
---
## Airbridge Supports TikTok iOS Real-Time Conversion Reporting
URL: https://www.airbridge.io/en/blog/tiktok-ios-realtime-conversion-reporting
Language: English
Published: 2025-10-21
Airbridge now supports TikTok’s Real-Time Conversion Reporting for campaigns on iOS 14.5 and later. As a badged TikTok Mobile Measurement Partner (MMP), Airbridge enables marketers to overcome SKAN delays and make faster, more informed decisions.
## **What Is Real-Time Conversion Reporting?**
TikTok’s Real-Time Conversion Reporting uses MMP probabilistic modeling to complement SKAN. Unlike SKAN, which can delay reporting for up to 48 hours, this feature delivers real-time data directly in TikTok Ads Manager. Marketers can see conversions, cost per conversion, and conversion rate almost instantly without waiting for aggregated reports.
It also powers Smart+, TikTok’s AI-driven campaign automation solution. With fresher and more accurate signals, Smart+ can optimize faster and deliver stronger results.
## **Why It Works Best Alongside SKAN**
Airbridge makes it possible to run Real-Time Conversion Reporting and SKAN together. Real-Time Conversion Reporting is not a replacement for SKAN but a complement. By combining both in Airbridge, marketers gain faster visibility into campaign performance while maintaining the reliability of SKAN reporting.
This dual approach provides deeper insights and gives marketers the confidence to adjust budgets, bids, and creatives exactly when needed.
## **How to Enable TikTok Real-Time Reporting with Airbridge**
Enabling TikTok Real-Time Conversion Reporting with Airbridge is simple.
- Disable [Extended Privacy Control(EPC)](https://help.airbridge.io/en/guides/extended-privacy-control) transmission from Airbridge to TikTok.
- This ensures consistency between TikTok’s postback method and probabilistic matching.
Marketers can update this setting directly in the Airbridge dashboard or contact the Airbridge team for support. Once enabled, Airbridge reduces data gaps between TikTok and MMP reporting, especially for conversions, CPA, and ROAS, while strengthening the performance of automated optimization features such as Smart+.
## **FAQs: What You Need to Know**
**Q. Can Real-Time Conversion Reporting replace SKAN?
**No. Real-Time Conversion Reporting is an additional workflow. If you disable SKAN, SKAN reporting will no longer appear in TikTok Ads Manager or your MMP. Running both together provides the most complete view.
**Q. Can this feature be used without an MMP?
**As of now, TikTok’s SDK and Event API do not support Real-Time Conversion Reporting on their own. An MMP integration is required.
**Q. How do I set it up in Airbridge?
**Once TikTok is integrated with Airbridge, simply disable EPC transmission.
**Q. Which iOS versions are supported?
**iOS 14.5 and above.
**Q. Is every app eligible?
**Eligibility depends on having enough data signals. In TikTok Ads Manager, check for the message: _“Your app is eligible for iOS Real-Time Reporting.”_
## **Getting Started**
Airbridge makes enabling TikTok’s iOS Real-Time Conversion Reporting simple and seamless. Marketers can measure performance more accurately, move faster, and give Smart+ the signals it needs to optimize effectively. Ready to take your TikTok campaigns further? Airbridge can help.
---
## How to Grow Your Subscription App: 5 Proven Elements for Scaling Revenue in 2026
URL: https://www.airbridge.io/en/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025
Language: English
Published: 2025-10-01
The subscription app market continues to grow — but the economics have shifted. Customer acquisition costs are rising, users are more skeptical about recurring payments, and free trials alone are not driving enough conversions. Successful apps like Calm, Noom, and Headspace have proven that recurring revenue models work. The question is what separates the apps that scale from those that plateau.
The distinction comes down to five elements. Not five independent optimizations — **a system where each element multiplies the others.** Teams that treat paywalls, pricing, onboarding, ASO, and UA as separate workstreams improve individual metrics. Teams that connect them build compounding growth.
**Key Takeaways**
- **The paywall is your highest-leverage conversion point.** Hard paywalls convert at [10.7%](https://www.revenuecat.com/state-of-subscription-apps-2025/), freemium at 2.1% — but the right model depends on your app and your traffic sources.
- **Pricing is psychology, not math.** The same app can convert dramatically better by changing how the price is presented — not the price itself.
- **Onboarding is a growth function, not a product function.** Users decide to pay in their first or second session. Failed onboarding eliminates conversion opportunities permanently.
- **ASO and ASA multiply each other.** Keywords validated through ASA testing often succeed organically. Running them separately leaves growth on the table.
- **Paid UA should optimize for subscriptions, not installs.** A $2 CPI channel that produces $100 cost per subscriber is worse than a $5 CPI channel that produces $25. The event you optimize for determines the users you get.
## **1\. Paywalls That Actually Convert**
Let’s be honest. Most users never advance past the paywall — this single screen determines whether users churn from free trials or commit to recurring payments.
The data tells a clear story: hard paywalls convert at [10.7%](https://www.revenuecat.com/state-of-subscription-apps-2025/). Freemium converts at 2.1%. That is a 5x difference — but it does not mean hard paywalls are always better. Hard paywalls self-select for high-intent users. Freemium converts a massive audience at a lower rate. Soft paywalls — trial-based models — sit in between. For health and fitness apps, trial-to-paid rates average [35.0%](https://www.businessofapps.com/data/app-subscription-trial-benchmarks/), well above the 25.6% global average.
The mistake many teams make is treating the paywall like an afterthought. They build it once, slap on a price, and move on. But top subscription apps treat it like a living, breathing sales funnel. They run A/B tests constantly. They swap headlines, tweak button copy, test colors, and even rework the order of information.
[__wf_reserved_inherit]
Top-performing subscription apps treat paywalls as active sales funnels requiring continuous A/B testing. Teams regularly test headlines, button copy, colors, and information sequencing.
For example, a simple wording change from _“Unlock lifetime insights”_ to _“Unlock everything for a lifetime”_ lifted conversion rates by **10.4%**. That is proof that small adjustments create measurable revenue gains.
### Key Implementation Strategies:
- **Frequency:** Display paywalls more often than instinct suggests. Ensure 100% of users encounter the paywall — the fastest revenue improvement is eliminating the users who never see it.
- **Triggers:** Deploy different paywalls for first-time users, repeat visitors, or checkout abandoners.
- **Content:** Test longer-form layouts with prominent CTAs and social proof.
- **Hooks:** Leverage time-limited discounts, personalized messages, or testimonials.
**The common mistake:** choosing a paywall model without measuring its performance by channel. A hard paywall might convert well overall — but if 80% of your Meta traffic bounces at the gate, you are paying for installs that never see your product.
**Deep dive:** [Hard Paywall vs Soft Paywall vs Freemium: Which Model Actually Converts?](/blog/hard-vs-soft-paywalls)
## **2\. Pricing That Matches User Segments**
Pricing is the intersection of psychology and revenue mechanics. Success requires understanding what different user segments _perceive_ as valuable — not just what the app costs to build.
[**Hashtag Expert**](https://www.hashtag.expert/)'s founder Zach Shakked discovered this firsthand. Younger users on Snapchat and TikTok rejected his $20/year plan but embraced **$2.99/week** — resulting in higher conversions and overall revenue growth. The product did not change. The framing did. Similarly, reframing "$60 per year" as "$5 per month billed annually" increased conversions by 38% in another app.
### Practical Application:
- **Three tiers:** Always offer three pricing options. Use the decoy effect to highlight your preferred plan.
- **Channel-aware pricing:** Adjust pricing by acquisition channel, geographic tier, or user demographic. A price point that works for ASA traffic may not work for TikTok traffic.
- **Presentation matters:** Weekly, monthly, and annual framing trigger different psychology. Test which framing works for each segment.
- **Retention offers:** Introduce seasonal deals, first-time discounts, and reactivation offers for lapsed users.
**The connection to measurement:** pricing changes affect cost per subscriber (CPS) — but most teams only see the blended result. Without channel-level data, you cannot tell if a pricing change lifted Meta conversions but dropped Google conversions. You need per-channel CPS to evaluate pricing tests.
[__wf_reserved_inherit]
👉 **Pro tip:** Treat pricing like ad creative. Test it often and adapt it to audience behavior.
## **3\. Onboarding That Activates Subscribers Fast**
Here’s a tough truth. Users typically make purchase decisions during their **first or second session**. Failed onboarding eliminates conversion opportunities permanently.
[__wf_reserved_inherit]
This requires close collaboration between marketing and product teams. Slow signup screens, unclear free trial messaging, or missing value propositions at launch severely impact trial-to-paid conversion rates. One subscription app doubled revenue to **$2,000/month** simply by refining onboarding and paywall flows — without increasing downloads or ad spending.
[__wf_reserved_inherit]
### Critical Focus Areas:
- **Short, intuitive signup flows.** Every unnecessary screen is a drop-off point.
- **Emphasize value early.** Progress tracking for fitness apps, matches for dating apps — show users what they get before asking them to pay.
- **Deep links to high-value features.** Direct trial users to the features that drive retention, not the features that are easiest to build.
- **Habit loops before trial expiration.** Daily reminders, streak mechanics, and progress notifications trigger engagement before the payment decision.
**The trial quality connection:** onboarding quality directly affects the signal you send to ad platforms. If your onboarding produces users who start a trial and cancel in 10 minutes, those impulse cancellers become the algorithm's success signal. Over 4–8 weeks, this degrades targeting quality — CPI stays flat, but CPS climbs. Qualified trials — a 2–4 hour delay before sending the trial event — filter out impulse cancellers so the algorithm learns from genuinely engaged users.
> [**Which channels produce subscribers who stay — and which produce trials that cancel? See trial-to-paid rates by channel. Start with 15K free attributed installs.**](https://coreplan.airbridge.io/waitinglist/c)
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|See which growth lever moves your subscription revenue most.%%
## **4\. ASO and ASA That Feed Each Other**
Discovery drives subscription app growth. [App Store Optimization (ASO)](https://www.airbridge.io/glossary/app-store-optimization-aso) provides sustained, organic reach while Apple Search Ads (ASA) accelerates growth. When executed together, ASA improves metrics including tap-through rate and conversion rate, which subsequently enhance ASO rankings — creating a growth multiplier.
For subscription marketers, ASA and ASO should not operate as separate functions. Keywords validated through ASA testing often prove successful in organic ASO strategies.
### The Organic Multiplier:
There is always a segment of users who see your ad, do not click, but search your app name in the App Store. This brand search lift is the "organic multiplier" — and for established apps like Calm and Headway, brand search volume frequently exceeds keyword-driven searches.
### Early-Stage Budget Allocation:
Early-stage apps should prioritize ASA — keyword-driven, intent-rich, and manageable. One case study showed $800 ASA spend generating $3,000 return: $2,000 direct attribution and $1,000 organic lift. Meta Ads follow naturally once economics improve, beginning with broad campaigns and refining through interest clusters.
**The measurement gap:** ASA drives installs with high intent. Meta drives volume. But without [channel-level attribution](/blog/web-to-app-vs-app-to-web), you cannot compare trial-to-paid rates between the two — and you are making budget decisions based on CPI instead of cost per subscriber.
## **5\. Paid UA That Focuses on the Right Events**
Modern subscription app growth moves beyond optimizing for installs alone. High-performing teams target deeper funnel events — trial starts, qualified trials, and subscriptions.
The event you optimize for determines the users you get. Send Install events, the algorithm finds installers. Send Subscribe events, it finds subscribers — but only if you have enough volume. Meta requires [50 conversions per ad set per week](https://roaspig.com/blog/conversion-events-optimize-for-meta). Google needs approximately 10 per day per campaign. TikTok needs 50 events or 7 days of data.
### The Event Hierarchy:
- **Install** — highest volume, weakest signal. The algorithm finds downloaders.
- **Start Trial** — mid-funnel signal. More intent, but includes impulse cancellers.
- **Qualified Trial** — filtered signal. A 2–4 hour delay removes users who cancel immediately.
- **Subscribe** — strongest signal, lowest volume. Only works at scale.
**The common pitfall:** prioritizing low-cost installs that never convert. A $2 CPI that looks efficient by install metrics can cost $100 per subscriber when only 2% of installs convert. Meanwhile, a $5 CPI channel can cost $25 per subscriber if 20% convert. CPI alone reverses judgment on which channel is actually working.
### Practical UA Strategies:
- Build multiple lookalike audiences for experimental validation.
- Rotate creative consistently to prevent audience fatigue. AI tools (Eleven Labs for text-to-speech, Creatify for ad variations) enable rapid creative testing without substantial production costs.
- Segment campaigns by platform — Meta, TikTok, and Google require different creative formats, different event volumes, and different optimization strategies.
- Tie optimization to events correlating with actual revenue — not installs.
**Deep dive:** [Which Conversion Event Should Your App Send to Each Ad Platform?](/blog/app-event-optimization-subscription-apps)
## **The Optimization Multiplier**
Here is the exciting part. You do not need millions in UA spend to see growth. Sharing in the [Jetstream session with Airbridge](https://www.youtube.com/watch?v=9yjiaeqc_-s), **Steve P. Young from App Masters** has tracked apps that went from
- **$250 to $1,000 a month** with just 15 daily downloads
- **$30 to $2,000 a month** with only 50 daily downloads
The gains came from optimizing paywalls and onboarding — not from increasing ad spend. This demonstrates the multiplier effect: each element amplifies the others. Superior onboarding drives trial starts. Better paywalls boost conversions. Refined pricing increases ARPU. Efficient UA sustains the cycle.
**But the multiplier is invisible without measurement.** You cannot know whether a paywall change improved conversions from Meta traffic or Google traffic without channel-level data. You cannot tell whether a pricing test lifted ARPU on ASA users or TikTok users without per-channel attribution. You cannot validate whether qualified trials improved cost per subscriber without connecting install source to subscription outcome.
Channel-level measurement is the layer that makes the other five elements visible. Core Plan tracks Install, Start Trial, and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, subscription events flow into attribution regardless of where payment occurred.
## **Wrapping Up**
Subscription apps scale globally without logistics constraints, deliver recurring revenue, and measure more straightforwardly than e-commerce through CAC and LTV analysis. But winning requires executing the five elements as a connected system:
- Test paywalls continuously — and measure conversion by channel.
- Segment and refine pricing — and track how pricing changes affect CPS by channel.
- Approach onboarding as growth work — and validate trial quality with qualified trial signals.
- Run ASA and ASO in tandem — and compare trial-to-paid rates across discovery channels.
- Optimize UA for subscriptions, not installs — and use the right conversion event for each platform.
These five elements are compounding forces. When executed together with channel-level measurement, each one multiplies the impact of the others. That is how you build a subscription business that grows — not just runs campaigns.
[__wf_reserved_inherit]
## **Subscribe to Success with Airbridge**
Each of the five elements above compounds — but only when you can see how they connect by channel. **Airbridge Core Plan** is the measurement layer built for subscription apps:
- **Full-funnel attribution by channel.** Track Install, Start Trial, and Subscribe as standard events with attribution across Meta, Google, Apple Search Ads, and TikTok — so you can see which channels produce paying subscribers, not just installs.
- **Paywall and pricing validation by traffic source.** The Actuals Report shows conversion rates at each funnel stage by channel. When you A/B test a paywall or change pricing, see which channels respond — instead of reading blended averages.
- **Trial quality measurement.** Compare trial-to-subscribe rates by channel to identify where impulse cancellers are concentrated — and validate whether qualified trial signals are improving algorithm quality.
- **Cost per subscriber by channel.** Move beyond CPI. Core Plan calculates the metric that actually determines channel profitability — cost per subscription — by connecting install source to subscription outcome.
- **Native RevenueCat and Adapty integration.** Subscription events flow into attribution via S2S regardless of where payment occurred — in-app or web checkout. No custom event setup required.
> [**Start with 15K free attributed installs. All features included. →**](https://coreplan.airbridge.io/waitinglist/c)
---
## What is Hybrid Monetization in Subscription Apps 2025?
URL: https://www.airbridge.io/en/blog/what-is-hybrid-monetization-in-subscription-apps-2025
Language: English
Published: 2025-09-29
## **Introduction**
Hybrid monetization has long been the domain of gaming apps, where ads, in-app purchases (IAPs), and subscriptions coexist to maximize revenue. Subscription apps, by contrast, leaned almost entirely on recurring payments.
But the economics of mobile apps are changing. User acquisition is more expensive, privacy rules limit data visibility, and consumers are feeling “subscription fatigue.” For subscription apps, hybrid monetization is no longer a side experiment. It is now a baseline requirement for sustainable growth.
## **Why Are Subscription App Monetization Models Changing?**
For years, subscription-first worked beautifully. Predictable revenue streams reassured investors and churn cohorts were easy enough to model. Yet three industry dynamics are forcing a rethink.
1. **Rising UA costs**: On iOS, acquisition costs spiked after Apple’s ATT rollout cut off user-level identifiers. Campaigns now demand higher LTV just to break even.
2. **Subscription fatigue**: Consumers are juggling too many recurring charges across streaming, fitness, productivity, and wellness apps, which makes them more cautious about new subscriptions.
3. **Platform competition**: Apple One and Google Play Pass changed the game. With bundled access to music, cloud, fitness, and dozens of apps at a flat fee, standalone apps must prove they are worth paying for separately.
Industry trends confirm this shift. [Liftoff](https://info.liftoff.io/2025-non-gaming-ad-monetization-trends) projects non-gaming ad revenue will outpace gaming by 2026, a historic milestone. [RevenueCat’s State of Subscription Apps 2025](https://www.revenuecat.com/state-of-subscription-apps-2025) report highlights hybrid models as one of the top growth levers for subscription businesses.
In short, what gaming figured out years ago is now the playbook for subscription apps.
## **What Are the Benefits of Hybrid Monetization for Subscription Apps?**
Hybrid monetization is about flexibility. Instead of depending on one revenue stream, apps combine paid apps, subscriptions, IAPs, ads, affiliate partnerships, or commerce into a mix that matches user behavior.
[__wf_reserved_inherit]
Not every user is ready to commit to a subscription. Some will tolerate ads. Others may buy one-off content. A few only upgrade after months of engagement. Hybrid creates multiple entry points into the revenue funnel while still keeping subscriptions at the core.
- **Revenue diversification**: Reduce risk when one channel underperforms.
- **User-centric design**: Let free users create value through ads while giving engaged users premium paths.
- **Higher LTV**: With rising UA costs, maximizing monetization from every user segment matters more than ever.
- **Adaptability**: Hybrid allows apps to pivot quickly when one revenue stream underperforms, ads during downturns, and subscriptions during stronger engagement cycles.
- **Competitive edge**: Offering multiple ways to engage and pay helps apps expand reach while still monetizing effectively.
## **Real-World Examples of Hybrid Monetization in Subscription Apps**
Examples of hybrid monetization aren’t theoretical. They already live across categories.
**Education Apps**
- **Duolingo** is a clear example of hybrid monetization. Its free tier is supported by ads, which generate value from casual learners, while the premium subscription, Duolingo Super, removes ads and adds advanced features such as unlimited hearts and offline lessons. This model ensures revenue from both free and paying audiences.
**Wellness & Fitness**
- **Strava** offers premium subscriptions for advanced analytics, route planning, and safety features, while also monetizing through brand-sponsored challenges to expand its revenue opportunities beyond subscription income.
**Dating Apps**
- **Tinder** combines tiered subscriptions like Tinder Plus, Gold, and Platinum with transactional upgrades such as Boosts and Super Likes. This ensures a steady stream of recurring revenue while also capturing incremental one-time spending.
**Music & Entertainment**
- **YouTube** operates on multiple revenue streams. It provides free access with ads, a premium subscription tier, and in-app purchases through features like Super Chat and memberships. This layered approach enables monetization across a wide spectrum of users and use cases.
- **Wattpad** offers Premium subscriptions for ad-free reading and offline access, while also monetizing through coin-based unlocks of paid stories and advertising for free users.
[__wf_reserved_inherit]
Across verticals, the pattern is clear. Multiple monetization levers allow apps to meet users where they are.
## **Best Practices for Hybrid Monetization in Subscription Apps**
Hybrid monetization is not just about adding ads. Execution matters. Teams should focus on:
- **Segmentation**: Heavy users are more likely to subscribe while casual users may prefer ads.
- **Contextual triggers**: Offer upsells after milestones such as completing a course or maintaining a streak.
- **Ad balance**: Rewarded videos and native ads perform better than disruptive banners.
- **A/B testing**: Experiment with bundles, upsell timing, and ad frequency.
- **Transparency**: Clear communication around pricing and ad load builds trust.
- **Partnerships**: Partnerships: Affiliates or brand collaborations can add incremental revenue without friction (for example, Calm partnered with American Airlines to provide in-flight meditation and relaxation content).
Hybrid monetization has become the natural evolution of subscription apps. Rising acquisition costs, stricter privacy rules, and changing consumer expectations demand a model that is both flexible and resilient. By layering ads, subscriptions, in-app purchases, and partnerships, apps can reach wider audiences, capture more value across user segments, and build sustainable growth in a competitive market.
### **Subscribe to success with Airbridge**
- Optimize Apple Search Ads (ASA) at the keyword level using real-time analysis to target the right audience.
- Boost UGC campaigns on TikTok, Meta, and other platforms to drive more installs and conversions.
- Streamline web-to-app campaigns with our robust Web and Mobile SDK to track user journeys seamlessly.
- Monitor funnel conversion rates across key stages like onboarding, sign-up, trial starts, subscriptions, and re-subscriptions to identify areas for improvement.
- Use SKAN to gain insights into iOS performance, especially for trial starts.
- Retarget effectively by building behavioral audiences for dropped-off or churned users.
- Integrate with RevenueCat and Adapty for full-funnel visibility and analysis.
---
## Subscription App Growth Guide: From Concept to Early Success to Scale
URL: https://www.airbridge.io/en/blog/subscription-app-growth-guide
Language: English
Published: 2025-09-25
The subscription app market is more competitive than ever. From fitness and wellness to productivity and dating, users have endless choices while marketers face rising acquisition costs, tighter privacy rules, and pressure to prove sustainable growth.
In this Jetstream Session, [Airbridge](https://www.airbridge.io/) CEO Roi Nam sat down with Steve Young, Founder of [App Masters](https://appmasters.com/) and one of the most trusted voices in subscription app growth. With over a decade of experience in ASO, Apple Search Ads (ASA), pricing, and paywalls, Steve has helped hundreds of apps grow with his practical, battle-tested advice. Here’s what he shared on idea validation, early success signals, scaling metrics, and what he calls “**The Greatest Growth Hack in Da World.**”
**🎥 Watch the full interview video**
## **The Mindset of Successful App Founders**
Before diving into tactics it is worth rewinding to Steve’s own journey. In 2011 he was coding iOS apps on his train ride to work in San Francisco, making just enough to keep the dream alive. By 2013 he launched the App Masters podcast, not as an expert but as a learner eager to understand how others were building successful apps.
That curiosity grew into a community and over time gave him a front-row to the mindset that set successful founders apart. Across hundreds of cases he saw the same three qualities surface again and again: **persistence, focus, and simplicity**.
> “Most people give up too soon. My first year full-time, I earned less than my old salary. But I stuck with it.”
> — Steve Young, Founder of App Masters
Persistence matters because growth rarely looks good right away. Focus matters because chasing every new channel prevents real learning. Simplicity matters because the best apps double down on the basics until they compound.
> “The most successful apps do not pile on complexity. They double down on the basics and let them compound.”
> — Steve Young, Founder of App Masters
[Steve Young, Founder of App Masters]
## **Idea Validation for Subscription Apps**
Everything starts with the right idea, and that means solving a real problem rather than offering a nice-to-have.
> “You can make a meditation app, or you can make an app for anxiety relief. Both have demand, but only one solves a pain. Always build the painkiller, not a vitamin.”
> — Steve Young, Founder of App Masters
High-intent problems attract users who are more willing to start trials, convert to paid, and stick with the product.To uncover these opportunities, Steve uses Google Trends, AppFigures, and Sensor Tower to spot rising keywords before the market gets crowded.
His rule of thumb is straightforward.
> “If you can find just two keywords with a search score of around 30 and win the top ranking, you are already looking at a six-figure opportunity.”
> — Steve Young, Founder of App Masters
Once the idea has been validated, the next question is whether the app can grow beyond a promising concept.****
## **Early Signals of Subscription App Success**
After launch, a handful of signals reveal whether an app has real potential to scale. The benchmarks to watch are:
- **Install-to-trial conversion** between 5% and 10%
- **Trial-to-paid conversion** in the 30% to 40% range
- **Daily organic installs** of 50 to 100 to provide enough data to start optimizing
[Live example: Steve shows how to spot early signals that show an app’s scaling potential]
Steve points out that at this stage there is no need for complex subscription tools, since App Store Connect provides all the visibility you need to track progress.
> “If I’m paying $1 per install and hitting 70% ROAS, I’m thrilled. Even if I’m not profitable yet, it shows traction I can optimize.”
> — Steve Young, Founder of App Masters
When these benchmarks line up, it is a strong indication that the app is ready for growth. Once the signals are clear, the next challenge becomes scaling that growth without losing efficiency.
## **Scaling Subscription Apps with ASO, ASA, and Meta Ads**
Once those early signals look promising, the focus shifts to scaling growth. Steve emphasizes keeping the playbook simple and building momentum step by step. Most apps that succeed follow the same progression. They start with ASO to capture high-intent searches. They then layer in ASA to buy additional discovery. Finally, when the economics allow, they turn to Meta Ads for broader reach.
The mistake many teams make is cutting Meta when results look flat. Steve warns against this, because Meta often drives indirect benefits that are not immediately visible in ROAS.
> “There’s always a segment that sees your ad, doesn’t click, but searches your app in the App Store. That brand search lifts ASO.”
> — Steve Young, Founder of App Masters
This halo effect is what he calls the organic multiplier. Paid campaigns spark awareness, awareness drives brand searches, and brand searches push ASO rankings higher. It is the reason Calm and Headway dominate search results not only for “meditation” but also on their own brand names.
> “For Headway and Calm, brand search volume in ASO is bigger than the keyword ‘meditation'.”
> — Steve Young, Founder of App Masters
> **Testing paywalls without channel-level data?** You're optimizing blind. See which channels drive subscribers who actually convert at your paywall — not just users who install.
> [**Start with 15K free attributed installs on Airbridge Core Plan →**](https://coreplan.airbridge.io/waitinglist)
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|See which channels actually drive subscribers — not just installs.%%
## **The Paywall Checklist for Subscription Apps**
Paywalls remain one of the strongest levers for subscription app growth. Steve points out that the fastest way to boost revenue is ensuring that every single user sees the paywall.
> “The fastest way to make more money is making sure 100% of users see your paywall.”
> — Steve Young, Founder of App Masters
To get this right, he recommends asking a few simple questions:
- Did you explain the value of the app clearly?
- Did you highlight social proof?
- Did you capture some basic information about your user (for example, fitness goals, preferred language, or focus area) to personalize the paywall?
- Did you make sure the paywall shows up for every install
Once those basics are covered, experimentation is where growth happens. A few examples Steve shared include:
- **Pricing Plan structure**: An app struggling with a $50 annual plan and a seven-day trial saw revenue climb after switching to a weekly plan with a three-day trial, plus an annual option with no trial. Surprisingly, more users began choosing annual once the trial was removed.
- **Pricing presentation**: Reframing pricing as “$5 per month billed annually” instead of “$60 per year” lifted conversions by 38 percent.
- **Multi-page paywalls**: Showing monthly and annual options side by side nudged more users to convert.
[Live example: Steve breaks down a high-performing paywall.]
## **User Acquisition for Subscription Apps on a Budget**
For early-stage subscription apps, marketing budgets are often tight. That is why Steve recommends starting with ASA. They are keyword-driven, intent-rich, and relatively simple to manage. In one case, a client spent $800 on ASA and earned $3,000 in return, with $2,000 directly attributable to ASA and another $1,000 in organic lift.
Once ASA is running smoothly, the natural next step is Meta Ads. Steve suggests beginning with broad campaigns, then layering in interest clusters as data comes in. But the real unlock, he emphasizes, is in the creative.
> “One client’s best-performing video was fully AI. Cheaper and just as effective as UGC.”
> — Steve Young, Founder of App Masters
AI has become a secret weapon for subscription apps looking to test creative quickly without overspending. Steve’s team uses Eleven Labs for text-to-speech, Creatify and Arcads for ad variations, and Reel.farm for reaction shots that strengthen hooks. With these tools, even small teams can test dozens of ad concepts at a fraction of the cost of traditional production.
For subscription apps without large budgets, AI is the closest thing to a cheat code, making rapid creative testing both realistic and sustainable.
## **The Greatest Growth Hack in Da World**
When asked about growth hacks, Steve did not hesitate. His boldest advice is what he calls “**The Greatest Growth Hack in Da World**”: lifetime offer campaigns.
The playbook is straightforward. Create a temporary lifetime free or discounted plan, then promote it through discovery platforms like AppAdvice or IndieAppSanta. The results can be explosive.
One client with an AI note-taking app ran this campaign and achieved results that most paid UA strategies could never match. In just two days, the app generated 106,000 installs and 800 new ratings. Monthly revenue jumped from a few hundred dollars to $7,000, and within a year it passed $90,000.
> “Where else can you get that kind of lift without spending on ads.”
> — Steve Young, Founder of App Masters
What makes this hack so powerful is that the benefits extend beyond revenue. Large install spikes drive reviews, improve App Store visibility, and fuel ASO through download velocity. While it may not work every time, when it does, it can change the trajectory of a subscription app almost overnight.
## **Fundamentals of Subscription App Growth**
Looking ahead, Steve is optimistic about subscription apps, but he emphasizes that the fundamentals have not changed since he first entered the industry more than a decade ago.
> “Focus beats distraction. Persistence beats quick wins. Simplicity beats over-complication.”
> — Steve Young, Founder of App Masters
Whether you are a bootstrapped founder or a marketer at a scaling team, the playbook is the same. Validate with ASO, prove traction with ASA, reinvest with Meta, and never stop testing paywalls. Growth rarely comes from chasing every new tactic. It comes from getting the basics right, and letting them compound over time.
For more insights, visit the [App Masters YouTube channel](https://www.youtube.com/appmasters?utm_source=chatgpt.com) or [appmasters.com](https://appmasters.com?utm_source=chatgpt.com). To measure and optimize your subscription app growth, learn more about [Airbridge](https://www.airbridge.io/).
[__wf_reserved_inherit]
### **Subscribe to success with Airbridge**
Steve highlights the growth path from ASO to ASA to Meta — but each stage only works when you can see which channel produces subscribers, not just installs.Airbridge Core Plan connects Install → Start Trial → Subscribe by channel, so you can answer the questions that actually drive growth decisions:
- **Which channels produce paying subscribers — not just trial starts?** Core Plan tracks Install, Start Trial, and Subscribe as standard events with full attribution across Meta, Google, Apple Search Ads, and TikTok.
- **Is your paywall converting differently by channel?** The Actuals Report shows conversion rates at each funnel stage by channel — so paywall tests are informed by where users come from.
- **What is your real cost per subscriber by channel?** CPI tells you the cost of an install. Core Plan shows the cost of a subscriber — the metric that actually determines whether a channel is profitable.
- **Are your trial signals reaching the right quality?** With native RevenueCat and Adapty integration via S2S, subscription events flow into attribution regardless of where payment occurred — in-app or web.
> [**Start with 15K free attributed installs. All features included. →**](https://coreplan.airbridge.io/waitinglist/c)
---
## 2025 AI trend in Gaming: UA-Driven Products, Ad Creatives, Monetization
URL: https://www.airbridge.io/en/blog/2025-ai-trend-in-gaming-ua-driven-products-ad-creatives-monetization
Language: English
Published: 2025-09-16
Mobile marketing has always been shaped by disruption. Apple’s App Tracking Transparency (ATT) reduced visibility. SKAN added delays and limitations. Now AI is changing how teams build, market, and monetize games. The studios growing in 2025 are those building systems designed to thrive in constant flux.
In a recent conversation on Phiture’s _Brave New Digital World_, [Roi Nam](https://www.linkedin.com/in/roinam/), CEO of Airbridge and Airflux, spoke with the host [Andy Carvell](https://www.linkedin.com/in/andycarvell/) about UA as a driver of product development, the new complexity of iOS measurement, and how AI is reshaping creative and monetization.
🎧 Check out the full conversation on Phiture’s Brave New Digital World, available on [Apple Podcasts](https://podcasts.apple.com/us/podcast/brave-new-digital-world/id1814863716) and [Spotify](https://open.spotify.com/show/6PuaiCBX1I0u11ywEBMeJM?si=29fb16c604c14300).
## **From “Build First” to UA-Driven Development**
For years, the formula was simple. Build the game, then market it. Unfortunately, that model no longer works. Development budgets are tighter, competition is fierce, and the cost of backing the wrong concept is too high. In the mobile gaming world, speed and data have rewritten the rules. Teams now launch stripped-down builds to test CPI, D1 retention, playtime conversion and free trial conversion before committing to full development.
Hypercasual studios push this further by releasing multiple prototypes at once and letting the data decide which survives. Subscription apps often start with ASO tests to validate demand before writing code. The logic is straightforward. UA data can reveal within days if players are interested, so there is no reason to spend months on the wrong build.
> _“UA isn’t just about marketing the product. It’s actually becoming part of the product development process itself. You don’t have to debate which concept to pursue. You can build five different concepts and pick the winner._”
> — Roi Nam, CEO of Airbridge & Airflux
AI has made this cycle even faster. Coding, design, and testing can now be automated, shrinking the gap between idea and launch. Small teams can move at a pace that once required full departments.
> _“We’ve been playfully using this word, vibe coding… because you just really code to the vibe. Frameworks like React Native or Flutter, paired with AI systems like Cursor and Windsurf, make it much easier and faster. A seasoned developer can launch an app within five to seven days.”_
> — Roi Nam, CEO of Airbridge & Airflux
For growth teams, the lesson is clear. UA is no longer only about scaling finished products. It has become a discovery engine, guiding studios toward concepts worth building before serious investment begins.
## **AI at Scale: Creative Supply and Smarter Monetization**
AI is changing creative production at every level. Static ads, playables, and short-form UGC-style videos are now machine-generated, often performing as well or better than human-made content. The influencer-style videos flooding TikTok, Reels, and Shorts are frequently not real people but AI-generated personas. With tools like OpenAI’s Sora, the line between synthetic and real is blurring fast.
> “There’s going to be a massive scale of ad creative production that is already underway. And I’m not just talking about the static images. I’m talking about everything, literally AI-generated UGCs, Playables, and videos.”
> — Roi Nam, CEO of Airbridge & Airflux
The same shift is reaching monetization and live ops. Difficulty curves, ad timing, and reward structures are still tuned through slow manual tests, but that approach is on its way out. Personalization will soon happen in real time, tailoring experiences for each player.
> “Right now, elements like stage difficulty, in-game economy, and ad monetization are tested manually in a slow and sporadic way. Soon these systems will be dynamic, personalized, and fully automated by AI, adjusting puzzle difficulty or ad timing in real time for each player.”
> — Roi Nam, CEO of Airbridge & Airflux
This vision is already in motion. [**Airflux**](https://abr.ge/mqmtum), developed by Airbridge, applies AI to optimize monetization dynamically, grouping players by behavior and adjusting ad delivery in real time to maximize engagement and revenue.
## **iOS Measurement: Complexity as the New Normal**
UA-driven development, AI-powered creatives, and smarter monetization only work if attribution is reliable. That remains the foundation of every growth system, and it is where many marketers still struggle, especially on iOS. Since ATT, measurement has been fragmented, with SKAN providing only partial and delayed signals. The studios that continue to scale are the ones layering strategies instead of relying on a single tactic.
[2025 AI trend in Gaming: UA-Driven Products, Ad Creatives, Monetization]
> “There’s no simple answer. You can’t just rely on a single strategy or a single tactic. You are required to have a multi-layered approach, mixing strategies like web-to-app and different attribution models to constantly adapt.”
> — Roi Nam, CEO of Airbridge & Airflux
This is where the MMP becomes critical. A modern MMP does more than track installs. It models conversions, reconciles SKAN data with organics, and connects directly with ad platforms to keep feedback loops alive. Modeled conversions in particular are now essential to bridge the gap between SKAN reports and real performance.
> “Although SKAN may show 100 installs, you know the organic multiplier is 1.5. That’s why we built the modeled conversion framework at Airbridge, to help marketers better understand install distribution in a privacy-first world.”
> — Roi Nam, CEO of Airbridge & Airflux
👉 [Read more](https://www.airbridge.io/en/blog/why-data-is-gamings-new-superpower) about “Why Data Is Gaming’s New Superpower: Everything You Need To Know About Attribution”
## **Preparing for an AI-Native Future**
AI adoption is moving past experimentation and into day-to-day workflows. At Airbridge, this includes **Airflux**, which applies AI to optimize monetization in real time, and a marketing co-pilot in development that will allow teams to query MMP data conversationally rather than relying on dashboards.
> “AI shouldn’t be just a side project or a productivity tool, but part of every single task we do day to day. In the future, you will see yourself typing and talking with the MMP agent, not accessing the dashboard at all.”
> — Roi Nam, CEO of Airbridge & Airflux
For game marketers, the direction is clear. UA, creative, monetization, and measurement are becoming AI-driven functions. Growth will depend less on adopting individual tactics and more on building systems that can adapt quickly as these changes take hold.
👉 See how [**Airbridge**](https://abr.ge/1i8vxa) and [**Airflux**](https://abr.ge/mqmtum) fit into your growth strategy.
---
## Why Data Is Gaming’s New Superpower: Everything You Need To Know About Attribution
URL: https://www.airbridge.io/en/blog/why-data-is-gamings-new-superpower
Language: English
Published: 2025-09-16
Apple’s App Tracking Transparency (ATT) did not just change mobile marketing. It rewrote the entire playbook. Four years later, studios are still grappling with lost visibility and weaker signals. Yet some continue to scale profitably. How? By building systems that embrace imperfect data and lean on smarter measurement.
As Roi Nam, CEO of Airbridge, explained in his recent conversation on the GameMakers podcast with host Joseph Kim, the wins came from teams that stopped chasing perfect data and instead built systems that could thrive in imperfection. A mobile growth manager from a midcore game studio put it bluntly:
> "We stopped waiting for perfect data and started building a system that handles imperfection. We test more. We validate with MMM. We run CAPI even when it is hard."
That mindset, Roi argued, has now become the standard for studios still growing efficiently.
🎧 Check out the full conversation on GameMakers, available on [Apple Podcasts](https://podcasts.apple.com/td/podcast/why-data-is-the-new-superpower-in-gaming/id1541808441?i=1000719648426) and [Spotify](https://open.spotify.com/episode/7ljfun4uAvaMm6hiRxIg34).
## **The Post-ATT Landscape: More Tools, Fewer Guarantees**
ATT reduced access to user-level data, especially on iOS. The fallout? Two major headaches for marketers:
- **Visibility loss:** Marketers can no longer see the full picture, making it harder to measure which campaigns are actually working.
- **Signal loss:** Ad platforms receive less feedback data, making it tougher for algorithms to optimize targeting and performance.
Ask any UA manager what has changed since ATT and you will hear the same thing: it is about seeing less and sending less.
But it is not all bad news. Over the past four years, the industry has scrambled to rebuild, and a new stack of tools has emerged.
- **SKAN 4.0:** Longer windows and up to three postbacks. Still delayed and coarse, but more usable than earlier versions.
- **AEM (Meta) and ICM (Google):** Platform-led alternatives that broaden conversion tracking while staying privacy-compliant.
- **CAPI (Conversions API):** Passes hashed event-level data (emails, phone numbers, click IDs) directly from the server to the platform.
- **Web-to-App and Web-to-Web-to-App:** Attribution via web flows, enabling better tracking without IDFA.
- **MMM (Marketing Mix Modeling) and Lift Studies:** Provide a higher-level view of performance and validate whether marketing inputs are truly incremental.
None of these are silver bullets, Roi emphasized.
> "The smartest teams do not rely on one method. They combine them."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
Even with more tools available, data from Mobile Measurement Partners (MMPs) like **Airbridge** remains the most valuable signal. On iOS, attribution often leans on probabilistic matching, using device-level cues like IP address, OS version, or screen size to link ads to installs. While not perfect, it gives teams a workable view of performance. And since MMPs also clean up SKAN postbacks and conversion values into formats that are easier to interpret, they have become the go-to source for campaign analysis and daily decision-making.
> "After ATT, the most impactful data signals that studios are relying on right now still come from MMP data. MMP combines probabilistic attribution using device-level data points and ad platform touch signals with SKAN data from Apple’s SKAN. Together, this gives studios the clearest picture of user behavior, even in a privacy-restricted ecosystem."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
## **Tactics That Separate the Leaders: What Top Gaming Studios Are Doing Differently**
### **1\. CAPI as Core Infrastructure**
CAPI (Conversions API) has become foundational for studios that want better match rates and richer optimization signals. On mobile, it offsets low ATT consent rates by passing valuable post-install events like in-app purchases. For PC and console titles on Steam, where SDK integrations are blocked and client-side tracking is impossible, it is a breakthrough.
One free-to-play shooter with 1,000 daily purchases saw a 5 percent ROAS lift in three weeks after sending purchase data through CAPI. The gains were small but steady. Roi cautioned_:_
> "It will not double your revenue overnight, but it compounds... CAPI usually helps with small but steady improvements, typically around 3 to 5 percent uplift in ROAS. It is worth doing if you have the development resources, but do not expect dramatic results overnight."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
### **2\. Web-to-App’s Quiet Comeback**
As iOS app install campaigns became more limited under ATT, marketers started returning to the web as the first touchpoint to regain visibility. Two setups stand out.
- **Web-to-App (W2A):** Users click a web campaign and go straight to the App Store with UTM parameters attached.
- **Web-to-Web-to-App (W2W2A):** Users first land on a web page (quiz, shop, landing page), then continue to the App Store via a CTA. With a Web SDK or tracking script, pre-install activity like clicks or interactions can also be captured.
This matters because attribution signals inside the App Store are limited. The web lets marketers measure intent, test creative concepts, and design richer user journeys.
Subscription and content apps have long used these funnels, and gaming studios are now adopting them too. A strong example is embedding HTML5 playables on a landing page so users can try the game before downloading. This improves both tracking and conversion by filtering for engaged players.
### **3\. SKAN: Focus on the First Postback**
SKAdNetwork (SKAN) is Apple’s privacy-first attribution system for iOS app install campaigns. It works without user-level identifiers and runs independently of MMPs, regardless of ATT consent. While SKAN 4.0 allows up to three postbacks, Roi emphasizes that studios should focus on the first.
Why the first SKAN postback matters most:
- Arrives within two to four days, fast enough for timely optimization.
- Includes fine-grained conversion values (up to 64), giving richer insights.
- Remains consistent across SKAN 3 and 4, keeping setups simpler.
Later postbacks are delayed and contain only coarse data, making them less useful for real-time decisions.
Now, just as important as managing postbacks is what happens before SKAN even kicks in, and that starts with **ATT consent**. The more users who opt in, the more ADIDs you capture, which strengthens visibility through your MMP. In practice, SKAN should be treated as Plan B. Plan A is always maximizing opt-ins.
Roi suggested three proven ways to boost ATT opt-in rates:
- **Use pre-prompts:** Explain why you are asking and how data improves the experience. Clear, honest messaging builds trust.
- **Customize your ATT prompt:** Make it sound human. Some smaller studios succeed with lines like, _“We use this data to show fewer, better ads and keep the game free.”
_**Send post-ATT reminders:** If users select “Do Not Allow,” follow up later and guide them back to settings to update permissions.
> "SKAN is Plan B. Plan A is always maximizing opt-ins. The more ADIDs you have, the stronger your MMP data."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
[Joseph Kim, Founder of LILA GAMES & Host of GameMakers PodcastRoi Nam, CEO & Co-Founder of Airbridge & Airflux ]
### **4\. More Studios Are Investing in MMM**
As user-level tracking becomes less reliable, more teams are turning to Marketing Mix Modeling (MMM). Unlike attribution methods tied to user identifiers, MMM is privacy-safe and well-suited for today’s environment. It helps answer questions MMPs or SKAN alone cannot.
- How much revenue is truly coming from paid UA?
- Which channels are driving incremental value, not just credited conversions?
- Where can spend be reduced without slowing growth?
Casual game publishers often see the biggest benefits since their campaigns are usually simpler, with fewer channels and cleaner data. This makes it easier for MMM models to deliver reliable insights.
Studios looking to implement MMM generally have two options: hire a vendor or build a self-serve setup. A typical self-serve workflow looks like this.
1. **Select a model** such as Meta’s _Robyn_ or Google’s _Meridian_.
2. **Choose variables** that combine marketing inputs (cost, impressions, installs) with context like seasonality or app store rankings.
3. **Calibrate the model** using lift studies from Meta or Google to ground results.
4. **Check accuracy** with statistical metrics such as R², MAPE, or MCMC convergence to ensure statistical reliability.
5. **Run the model** on a recurring basis, updating inputs every 1–3 months and recalibrating every 3–6 months.
6. **Review and optimize** budgets gradually, applying insights without overcorrecting.
> "A growth lead from a casual puzzle studio once told me, MMM helped us uncover that influencer campaigns on YouTube outperformed rewarded video spend in certain regions. We would not have caught that using last-touch alone.''
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
[Roi Nam, CEO & Co-Founder of Airbridge & Airflux ]
### **Smarter Monetization with Airflux**
Measurement is only half the story. With UA costs rising, the real question is: how much more value can you earn from the players you already have? That is why the Airbridge team built [**Airflux**](https://www.airflux.ai/), an AI-powered engine that boosts LTV by running smarter ad policies at the segment level.
Studios are already seeing results.
👉 See how [Clegames](https://airflux.ai/blog/clegames-interstitial-monetization-ai) and [Treeplla](https://airflux.ai/blog/treeplla-ltv-increase-experiment-automation) put Airflux to work.
### **UA and Monetization: One Loop, Not Two**
For a long time, UA and monetization ran on separate tracks. UA teams chased lower CPIs and ROAS goals, while monetization teams focused on ad placements and IAP flows. Different KPIs, different dashboards, little overlap.
That split no longer works. With CPIs climbing and signals shrinking, growth and monetization now have to operate as one loop. The most efficient studios already know this.
Modern DSPs are leading the way. AppLovin, for example, feeds MAX mediation bidding data back into its UA engine. That closed loop has been shown to boost performance by up to 4x, not because ads are cheaper but because campaigns are optimized against real monetization outcomes.
Studios can follow the same approach by tailoring campaign strategy to player behavior:
- **IAP-focused users:** Delay ads and encourage early purchases.
- **IAA-focused users:** Serve ads from the first session for ad-engaged cohorts.
- **Blended ROAS:** Combine both approaches for sustainable growth.
The challenge is measurement. MMPs and SKAN are built on last-touch attribution, which shows who gets credit but not what is incremental. That is why more teams validate with Lift Studies and MMM. These layers help separate what truly drives revenue from what only looks good on paper.
> "The principle here is simple. DSPs are getting very good at identifying players with strong IAP or IAA potential. The job for studios is to match the right campaign strategy to the right audience."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
### **What’s Next: Better Signals, Smarter Systems**
The next phase of measurement will not come from a single source of truth. It will come from combining methods, each one adding a different layer of clarity.
**MMPs** like Airbridge provide real-time visibility across campaigns, creatives, channels, countries, and OS-level trends. **MMM** brings the long view, showing how online and offline media contribute to revenue over time. **Lift Studies** from Meta, Google, or third parties validate incrementality so teams know what is truly moving the needle.
The bigger shift, however, is in **signal engineering**. More data no longer means better performance. Leading studios are filtering out spoofed or low-value signals and focusing on verified, high-quality ones such as app store purchase data that improve platform optimization.
At the same time, AI is reshaping how growth teams operate. We are already seeing:
- **Creative automation tools** like Poolday ML, Blay, and Incymo
- **Agentic platforms** such as RetentionX, AppRadar, and Pollen VC for CRM, ASO, and ad ops
- **Ad monetization engines** like Airflux, which uses reinforcement learning to refine strategies at the segment level
Remember, the studios pulling ahead are focusing on three priorities.
- Better signals that strengthen platform feedback
- Smarter measurement through MMM, Lift Studies, and CAPI
- Closer alignment between UA, product, and monetization teams
> "It is not about rebuilding the old playbook. It is about adapting faster than everyone else."
> — Roi Nam, CEO & Co-Founder of Airbridge & Airflux
The future belongs to the studios that adapt, experiment, and rethink how growth is measured and monetized.
👉 See how [Airbridge](https://www.airbridge.io/) and [Airflux](https://www.airflux.ai/) fit into your growth strategy.
---
## Moloco on Driving CTV Ad Efficiency and Mobile App Performance in 2025
URL: https://www.airbridge.io/en/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance
Language: English
Published: 2025-09-04
Streaming has officially taken over the living room, and advertisers are racing to claim their share of attention. Connected TV isn’t just an alternative to mobile anymore, it’s shaping up to be one of the most efficient performance channels of 2025. What strategies are essential to drive meaningful results with CTV?
For the third installment of Airbridge’s CTV Insights series, we sat down with [Moloco](https://www.moloco.com/?r=0), the AI-powered performance advertising platform fueling growth for businesses of all sizes and industries. As CTV rises as a key driver of successful mobile app marketing, Hyejeong Lee, Head of CTV Partnerships at Moloco, shared must-know insights on how to turn CTV into real performance. Let’s dive in.
## How is the CTV ad market evolving in 2025?
With global streaming adoption on the rise and more success stories emerging, the CTV ad market is experiencing steep growth in 2025.
- **AI-driven personalization**: Advances in AI now enable highly tailored CTV ads based on user behavior data.
- **Standardized measurement**: Core metrics such as reach, frequency, and performance are becoming increasingly unified across platforms, allowing performance marketers to clearly evaluate campaign impact.
- **New ad formats:** Interactive and transactional formats are transforming CTV from a pure branding tool into a true performance channel.
## Why should mobile app marketers invest in CTV campaigns? What are the key advantages?
User attention is shifting rapidly to CTV. YouTube’s CTV viewership has already surpassed mobile, underscoring this trend. The main benefits of CTV as a channel include:
- **Alternative to saturated channels**: For apps reaching diminishing returns on mobile and web, CTV provides a differentiated growth lever to acquire high-value users.
- **Promotion powerhouse:** During game launches, pre-registrations, large-scale events, or seasonal campaigns, CTV drives strong engagement that lifts funnel performance.
- **Cross-device impact:** Since CTV is often watched in shared spaces like the living room, it enables retargeting across household mobile devices, completing the full conversion journey from install to sign-up to purchase.
- **Filling the ad gap:** CTV allows marketers to reach users on ad-free platforms like YouTube Premium or Netflix, recapturing audiences missed on mobile.
- **Proven performance:** In verticals with high CAC pressure or narrative-driven engagement needs, such as gaming, sports betting, and lifestyle apps, CTV consistently drives ROAS and retention.
- **Privacy-friendly:** Because it doesn’t rely on personal device identifiers, CTV stands as a stable performance channel well into the future.
## Which app categories see the strongest performance on CTV?
**Gaming apps**, especially mid-core, casino, and sports betting, show particularly strong results. Apps with streaming, short-form video, e-commerce, or web conversion goals (like site visits and pre-registrations) are also gaining traction.
In general, apps that already perform well on mobile programmatic channels tend to adapt quickly and succeed with CTV advertising.
## How does Moloco help marketers run effective CTV campaigns?
Moloco CTV is built on their deep learning engine, enabling optimization beyond exposure and reach toward down-funnel performance metrics such as **installs, in-app actions, purchases, and revenue.** Marketers can also carry over insights and targeting strategies from mobile campaigns directly into CTV, driving greater efficiency across both UA and re-engagement.
## How does Moloco’s CTV bidding algorithm work?
Moloco’s machine learning engine dynamically adjusts bids based on an advertiser’s KPIs to maximize performance.
Currently, they offer an **app install optimization model**, with the ability to optimize funnel KPIs such as in-app actions, purchases, and ROAS. They are also expanding to optimization for purchases and ROAS in the near future.
Thanks to their integration with Airbridge, advertisers can leverage postbacks in CTV, unlocking performance optimization at the same level as mobile DSPs. This makes it possible not only for mobile apps but also for today’s rising PC and console game publishers to run install-based campaigns with true performance goals. Few solutions currently provide this level of deep funnel optimization.
## Which CTV formats perform best, and how can advertisers upload creatives and measure performance?
Most of the strong results come from **in-stream video ads within AVOD, FAST, and live TV content**. These formats are non-skippable, with completion rates as high as 93–96%.
YouTube’s CTV viewership has already surpassed mobile, making CTV a must-consider channel for digital marketers.
Compared to mobile and web, CTV ads deliver higher immersion and lighter ad load. Unlike linear TV, they also support real-time optimization and creative A/B testing. Campaigns can run flexibly with reasonable budgets, without the need for long-term upfront commitments. Advertisers can upload video assets directly on the Moloco platform and track performance through MMP integrations.
## There’s a perception that CTV targeting options are limited. What capabilities does Moloco provide?
Moloco offers a wide range of targeting features, including:
- Performance ML and first-party audience targeting
- Geo-based targeting (by country or state)
- Device and household-level targeting (where available)
- Contextual targeting based on publisher signals
- IP-based suppression and retargeting
As the **Moloco Household Identity Graph** continues to evolve, cross-device and cross-platform optimization will only become more powerful.
## How can mobile app marketers measure CTV campaign performance with Moloco?
Through integration with Airbridge, advertisers can measure CTV-driven installs, in-app actions, and lower-funnel KPIs within a unified attribution framework.
They can also leverage features like **incrementality testing and ghost bidding** to analyze CTV’s contribution to marketing performance with precision.
Together with Airbridge, they are building a high-performance CTV advertising ecosystem.
## Success Stories: How Leading Apps Achieved Growth with Moloco CTV
Many advertisers are building new success stories through Moloco CTV.
For instance, **Krafton** leveraged Moloco CTV to support the launch of “Cookie Run India” in the Indian market. They used CTV for upper- and mid-funnel reach, while mobile DSPs captured lower-funnel conversions, driving significant growth.
**A leading sports betting app** achieved strong results with geo-targeted campaigns and CTV’s clean ad environment. During a major sports season, they achieved strong conversion efficiency, demonstrating how scalable and performance-driven CTV has become.
A leading Chinese game publisher also ran UA campaigns for PC/console titles with Moloco CTV, reaching high-value users at highly efficient costs.
## What advice does Moloco have for mobile app marketers expanding into CTV?
Start by extending the audiences that already perform well on mobile DSPs into CTV. In premium environments like the living room, you’ll gain higher attention and stronger engagement.
Don’t stop at awareness, but rather optimize for lower-funnel outcomes such as **installs, in-app actions, and purchases**
It’s also critical to integrate with an MMP like Airbridge to accurately track CTV’s contribution.
Above all, don’t treat CTV as a “branding-only” channel. See it as a natural extension of your mobile UA strategy.
## How can PC and console game developers leverage CTV advertising?
PC and console platforms often face limited conversion paths and are less familiar with performance-driven UA. Moloco is extending its proven mobile expertise into the PC and console space.
- **Immersive CTV experience:** Full-screen ads on AVOD and FAST platforms can drive actions like site visits, downloads, and pre-registrations.
- **PC/console-specific advantages:** With higher price points and longer purchase journeys, CTV enables storytelling-driven ads that connect brand awareness to eventual conversions.
- **Measurability:** Through integration with an MMP like Airbridge, developers can track web visits, account sign-ups, downloads, and in-game actions. Incrementality testing and advanced attribution are also supported.
While still early, they are already preparing major promotions with partners and look forward to sharing more success stories soon.
At Moloco, their focus is on driving direct response performance through CTV. They are extending their proven machine learning capabilities from mobile DSPs into CTV, helping advertisers achieve measurable results.
Their goal is to build a performance-driven CTV ecosystem where both supply and demand can thrive.
Many mobile marketers still think CTV is too complex or difficult to measure. But with a platform like Moloco, **launching a campaign is as straightforward as activating one on Facebook Ads or Google Ads**, but with higher engagement rates and lower competition.
## CTV insights for mobile growth
Want to hear more insights from the teams leading the CTV ad industry? Click the links below to explore the rest of the Airbridge CTV interview series.
👉 [Motiv Intelligence](https://www.airbridge.io/en/blog/airbridge-ctv-interview-motiv)
👉 [Vibe.co](https://www.airbridge.io/en/blog/airbridge-ctv-interview-vibe)
---
## Vibe.co on Maximizing User Acquisition with AI-Powered CTV Advertising in 2025
URL: https://www.airbridge.io/en/blog/airbridge-ctv-interview-vibe
Language: English
Published: 2025-09-04
CTV is rapidly emerging as a high-performance UA channel for mobile app marketers, yet practical knowledge on how to leverage it remains limited. For the second session of Airbridge’s CTV Insights series, we sat down with [Vibe.co](http://vibe.co), a specialized CTV advertising solution provider, to discuss the future of CTV advertising.
From strategies to strengthen market leadership to insights on maximizing CTV ad performance with Vibe.co’s AI-powered algorithms, let’s discover it all in this Airbridge CTV interview.
## Why should mobile app marketers invest in CTV campaigns?
Competition is intensifying across mobile ad channels, and costs are rising. Marketers urgently need an alternative channel that can deliver true performance, and CTV provides exactly that. Its advantages can be distilled into three key points:
- **Lower customer acquisition cost (CAC):** Unlike the cluttered mobile environment where ads compete for attention on a small screen, CTV offers a clean, streamlined interface that fosters higher engagement.
- **Broader reach:** Beyond OTT streaming services, CTV extends to viewers on TVs and smart TVs, making it easier to reach users that traditional mobile networks often miss.
- **High-value user acquisition:** CTV viewers are more attentive and engaged, leading to more valuable user conversions.
With its immersive full-screen format, CTV enables mobile marketers to acquire users more efficiently compared to traditional in-app display or social ad networks.
## Which app categories see the strongest results from CTV campaigns?
While CTV delivers strong results across a wide range of app categories, some stand out in particular:
- **Gaming apps:** Casual and mid-core gamers who enjoy streaming also respond well to CTV ads.
- **Lifestyle & fitness apps:** Health, wellness, and self-improvement apps benefit from CTV’s immersive format.
- **E-commerce apps:** Retail and marketplace apps use CTV effectively to drive web-to-app conversions.
- **Finance & fintech apps:** Banking, investment, and crypto apps gain an edge by reaching high-intent users, driving installs and account sign-ups.
- **Streaming & entertainment apps**: CTV is ideal for nudging viewers toward free trials and subscriptions while they’re already engaged with content.
- **Utility apps:** Productivity and utility apps connect seamlessly with users actively seeking solutions.
Ultimately, apps that pair high-quality creatives with strong storytelling achieve the best results. They leverage CTV’s immersive experience to its fullest potential.
## How Does Vibe.co Support CTV Campaign Success?
Traditional TV ads required complex negotiations and long lead times, but CTV makes it much faster and easier to get started. With Vibe.co, marketers can:
- Quickly launch campaigns with a simple setup, eliminating the complexity of traditional TV advertising.
- Access premium streaming inventory in brand-safe environments.
- Easily optimize campaigns with advanced technology and dedicated account manager support.
- Track and manage performance in real time, with the same level of granularity as a mobile DSP.
In short, [Vibe.co](http://vibe.co) enables mobile marketers to test and scale CTV campaigns as quickly and efficiently as they would on any UA channel.
## How Does Vibe.co’s AI-Powered Bidding Algorithm Improve User Acquisition?
Vibe.co uses an AI-powered bidding algorithm that automatically identifies users with the highest likelihood of conversion. Here’s how it works:
- **Goal-based optimization:** Marketers set campaign objectives, such as installs, purchases, or revenue, and the system prioritizes high-value users.
- **Machine learning model:** Targeting is continuously refined based on campaign performance data.
- **Postback integration:** By connecting with MMPs like Airbridge, Vibe.co enables data-driven optimization.
With this structured approach, CTV evolves from a pure awareness channel into a performance-driven UA engine.
## How can advertisers upload creatives and measure performance?
With Vibe.co’s intuitive creative management tools, advertisers can easily upload video assets and set up multiple ad versions. Once campaigns are live, marketers can measure performance through an MMP like Airbridge across key metrics:
- **Conversion funnel:** Track how many viewers install the app after seeing the ad.
- **Retention metrics:** Measure how long users stay engaged at D1, D7, and D14.
- **Cohort-based ROAS analysis:** Assess user LTV to optimize budgets by creative.
By continuously iterating on creatives with granular performance data, marketers can drive higher LTV and maximize ROI through their most effective assets.
## There’s a perception that CTV targeting options are limited. What capabilities does Vibe.co offer?
This is a common misconception. In fact, CTV targeting can be just as advanced as mobile DSPs, and sometimes even more precise. Vibe.co provides marketers with detailed targeting options such as:
- **Audience-based targeting:** Reach specific groups like gamers, sports fans, or education-focused users.
- **Proprietary SSP:** Direct access to premium inventory from leading TV manufacturers like Samsung, LG, and Philips, streaming services like Hulu, Peacock, and Tubi, and popular platforms like Roku, DirecTV, and Pluto.
- **Contextual targeting:** Run ads on channels dedicated to sports, entertainment, or news.
- **IP-based targeting:** Upload first-party IP lists to match CTV users and enable precise retargeting.
- **Household-level targeting:** Ensure ads reach the right viewers even in multi-user households.
With these capabilities, CTV delivers audience segmentation and targeting precision on par with leading mobile DSPs.
## How can mobile app marketers measure CTV campaign performance with Vibe.co?
Vibe.co integrates with MMPs like Airbridge to deliver detailed measurement and attribution, including:
- **View-through attribution (VTA):** Track installs that occur later, even without a direct ad click.
- **Cross-device tracking:** Measure how CTV exposures influence mobile installs and revenue.
- **Funnel-based analysis:** Connect CTV impressions to D7/D14 retention, LTV, and ROAS.
By combining these insights with mobile UA strategies, marketers can move beyond simple impressions to truly optimize user growth.
## Success Story: How Bagelcode Scaled with Vibe.co’s CTV Advertising
A great example is mobile game publisher [**Bagelcode**](https://www.bagelcode.com/en/). To scale its Club Vegas campaign, Bagelcode turned to Vibe.co’s CTV ads with the goal of improving ROAS and re-engaging users.
The results of the partnership were remarkable:
- **ROAS tripled** through optimized attribution powered by CTV Ad Vision.
- D7 retention consistently doubled.
- Multiple video formats were tested to maximize engagement.
- Within just one month, Bagelcode more than doubled its CTV budget thanks to these strong results.
By combining smart targeting, real-time optimization, and scalable budget strategies, Bagelcode successfully re-engaged users and maximized campaign profitability. This case clearly demonstrates how effective CTV can be as a channel for mobile app growth.
## What best practices should mobile app marketers adopt to scale their CTV strategy?
To run CTV campaigns successfully and position CTV as a proven performance channel, marketers should follow these UA best practices:
- **Define clear goals:** Set specific KPIs such as CPI, ROAS, or installs.
- **Use high-quality video ads:** Create immersive content optimized for large screens.
- **Add clear CTAs:** Include App Store and Google Play logos to guide users directly to install.
- **Track cross-device behavior:** Measure how CTV exposure leads to mobile actions.
- **Continuously optimize:** Treat CTV like a mobile DSP with ongoing testing and adjustments.
## How can PC and console game developers use CTV to reach target users?
For PC and console game studios, CTV offers the advantage of showcasing high-quality graphics and immersive gameplay on the big screen. Several factors make gaming a strong fit for CTV:
- **Highlighting visuals:** CTV delivers stunning visuals in high definition, making it ideal for showing off impressive game graphics.
- **Targeting gaming fans**: Reach audiences already interested in gaming content, increasing the likelihood of conversion.
- **Cross-platform promotion:** Promote PC games to console players or the other way around.
- **Launch campaigns:** Leverage CTV’s broad reach to announce new releases or major updates.
## What unique advantages does CTV offer PC and console games compared to mobile?
PC and console titles are a natural match with CTV for several reasons:
- **Alignment with gaming environments:** Many PC and console gamers already play on large screens, making CTV ads feel seamless and natural.
- **Showcasing high-end games:** CTV’s premium format is ideal for highlighting advanced graphics and complex gameplay.
- **Fit with long play sessions:** PC and console games typically involve extended playtime, aligning well with the immersive habits of CTV audiences.
## How can CTV advertising contribute to user acquisition for PC and console games?
CTV can significantly enhance UA for PC and console game publishers by enabling:
- **Cross-device tracking:** Understand how CTV exposures lead to purchases or downloads on platforms like Steam or console stores.
- **Brand lift studies:** Measure how campaigns boost awareness and interest in a game.
- **Retargeting opportunities:** Use viewer data to retarget across other devices.
- **Performance metric tracking:** Monitor KPIs such as CPA and ROAS to optimize campaigns.
With MMP solutions like Airbridge, developers can measure results in real time for accurate performance analysis.
Vibe.co’s case studies highlight multiple PC and console campaigns, showing how CTV’s immersive visuals and precise targeting can significantly boost performance.
## CTV insights for mobile growth
Want more insights from the teams leading the CTV ad industry? Click below to explore the full CTV interview series:
👉 [Motiv Intelligence](https://www.airbridge.io/en/blog/airbridge-ctv-interview-motiv)
👉 [Moloco](https://www.airbridge.io/en/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance)
---
## Motiv Intelligence on Turning CTV into a Performance Channel for Mobile Marketers
URL: https://www.airbridge.io/en/blog/airbridge-ctv-interview-motiv
Language: English
Published: 2025-09-04
Motiv Intelligence on Turning CTV into a Performance Channel for Mobile Marketers
User acquisition UA is always top of mind for mobile app marketers. In today’s hyper-competitive market, finding channels that consistently deliver measurable performance isn’t easy.
That’s why CTV advertising is gaining traction as a scalable opportunity for UA. Yet many marketers still lack clarity on the role CTV can play in their acquisition strategy.
To answer those questions, Airbridge launched a dedicated interview series spotlighting the fast-evolving world of CTV advertising. For our first edition, we sat down with Motiv Intelligence, a leading player in user behavior data and cross-media targeting.
Read on for [Motiv Intelligence’s](https://www.motiv-i.com/en/) take on what CTV really means, why it matters, and how marketers can make it work.
## Why should mobile app marketers invest in CTV campaigns?
TV has always been one of the most strategic advertising channels. As the biggest screen in the household, it commands attention and trust through its size and premium audio-visual experience. Beyond its standalone impact, research from MediaCom shows TV also amplifies the effectiveness of other channels, making it a powerful booster in the overall media mix.
That said, traditional TV advertising has long been a tough fit for performance marketers. With high CPMs, rigid buying models, and limited flexibility, TV often fell short compared to digital. And without data-driven targeting, measurement, or optimization, it simply didn’t meet the needs of app marketers who live and breathe attribution.
CTV changes the equation. Because it’s internet-connected, CTV advertising combines the reach and authority of TV with the precision of digital. Marketers can now buy impressions programmatically at CPM, target audiences in real time, and track performance with the same rigor as other digital campaigns. It’s no surprise that app marketers are already embracing CTV as a core UA channel.
The ecosystem is ready. Mobile app marketers can now activate data-driven CTV campaigns, attribute results through an MMP like Airbridge, and continuously optimize toward performance goals. For UA teams exploring new growth channels, CTV deserves a spot at the top of the list.
[__wf_reserved_inherit]
## Which app categories are finding success with CTV advertising?
Because CTV offers targeting and measurement capabilities on par with digital, advertiser adoption is growing fast. We’re seeing strong adoption from verticals such as beverages, automotive, consumer electronics, fashion, and QSR. On the app side, examples range from location-based community apps to mobile games, rewards apps, and luxury shopping apps. In fact, 63% of campaigns we ran in 2024 continued for more than two months, clear proof of high satisfaction among advertisers.
Motiv Intelligence’s CTV solution supports both brand lift and performance outcomes, leveraging MMP integrations for mobile apps, CTS for web tracking, and DMP-driven data collection and analysis. This gives marketers a full picture, from awareness through to conversions.
## How does Motiv Intelligence help mobile marketers run effective CTV campaigns?
While CTV operates in the digital realm, it works differently from mobile advertising. Because users can’t click directly on a TV screen, marketers need expertise in mapping exposure data to in-app actions, a skill set distinct from traditional mobile UA. Effective campaigns also require deeper audience segmentation, often starting with DMP data extraction and analysis at the planning stage.
Motiv Intelligence runs both as a CTV ad platform and a mobile ad platform, enabling true cross-device campaigns. By bridging TV and mobile, they help marketers deliver more holistic and impactful results.
For some app marketers, CTV can feel unfamiliar. That’s why their team offers tailored consulting backed by hands-on experience in cross-device advertising and data-driven optimization, ensuring every campaign aligns with the marketer’s UA objectives and measurement framework.
## How does Motiv Intelligence’s CTV bidding algorithm improve user acquisition performance?
At the core of Motiv Intelligence’s CTV platform is a patented mobile–TV matching technology, one among the very first of its kind. Combined with one of the largest CTV ad inventories, this allows app marketers to run precise audience-level targeting based on mobile behavior histories. Their platform is also integrated with leading MMPs, including Airbridge, which means that even though ads are delivered on TV, they can be tracked like mobile ads. Postbacks enable real-time measurement and optimization, giving marketers full transparency into performance.
Marketers can target users acquired through previous UA campaigns, or re-engage churned users, by delivering ads on TV and measuring results with MMP data.
They also integrate with six different DMPs, one of the broadest ranges available. This gives advertisers the flexibility to build custom audience segments, whether it’s competitive app users during a launch, shoppers with purchase histories, or audiences defined by specific location visits, and activate them directly on TV.
## Success Stories: Driving App Growth with Motiv Intelligence CTV
A location-based community app measured with an MMP saw its in-app purchase conversion rate reach 116% compared to mobile-only campaigns. An outdoor brand using CTS achieved a return on ad spend (ROAS) uplift of roughly 450%. And a coffee beverage brand leveraging DMP data analysis recorded a 315% lift in ad contribution. For more details, marketers can check out their [CrossTarget TV product guide](https://www.motiv-i.com/crosstarget-tv/files?file=introduction) available on their website.
## What advice does Motiv Intelligence have for mobile marketers expanding into CTV?
Mobile and gaming marketers know how central mobile advertising has been to UA. But every stage of the marketing funnel matters, and performance at each stage influences not only the next step but also long-term retention. That’s why marketers need to engage consumers across their entire media journey with precise, well-placed messaging.
Meanwhile, privacy regulations and rising CPMs are squeezing performance efficiency in mobile channels. Marketers are looking at MMM (Media Mix Modeling) as a solution, but most MMM frameworks have been mobile-first, thus failing to capture the full complexity of today’s cross-device media behavior.
With CTV, that gap closes. Data-driven targeting, digital-level measurement, and performance optimization are now possible on TV, making it easier to integrate CTV into MMM models. For marketers focused on consumer-centric strategies, the question becomes: how can you capture and maximize the “watch time” audiences are spending on CTV to drive brand awareness and long-term growth?
## Mobile growth insights through CTV
Want to hear directly from the teams shaping the future of CTV advertising? Check out the full interview series below for insider perspectives and proven strategies.
👉 [Vibe.co](https://www.airbridge.io/en/blog/airbridge-ctv-interview-vibe)
[👉 Moloco](https://www.airbridge.io/en/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance)
---
## Firebase Dynamic Links Shutdown: Which Alternative Fits Your Team Best?
URL: https://www.airbridge.io/en/blog/firebase-dynamic-links-alternatives
Language: English
Published: 2025-08-18
The end of Firebase Dynamic Links is fast approaching. Are you still searching for a deep linking alternative or trying to build your own?
[__wf_reserved_inherit]
On August 25, all existing Dynamic Links will stop working. After this date,migrating your links may become difficult or even impossible. Deep linking is essential for mobile apps, whether it is used for ads, push notifications, CRM campaigns, share events, or offline QR codes. If your app relies on Firebase Dynamic Links, preparing for the shutdown is urgent.
In this blog, we compare three of the top alternatives to Firebase Dynamic Links so you can choose the deep linking solution that best fits your business needs before the deadline.
## **What Are the Alternatives to Firebase Dynamic Links?**
When Firebase Dynamic Links shuts down, you can switch to deep linking solutions such as the **Airbridge DeepLink Plan** and two other leading platforms officially recommended by Firebase.
However, not all deep linking alternatives are created equal. Each solution varies in features, pricing, and usage limits, so it is important to compare them carefully before making the switch. Let’s take a closer look
[__wf_reserved_inherit]
When choosing a solution, pay attention to **link migration support**, **deferred deep linking** (routes new installs to the right page after app installation), **QR code availability** for offline marketing, and **link or API usage limits**.
## **How Do the Prices Compare?**
Pricing is a key factor. All three solutions use a pay-as-you-go model with a free usage tier. Here’s how they compare:
[__wf_reserved_inherit]
## **Airbridge DeepLink Plan: Complete and Affordable**
The Airbridge DeepLink Plan combines competitive, usage-based pricing with the same level of link tracking and measurement you’d expect from a full MMP.
Key advantages:
- **1-Minute Migration:** Export your existing Firebase Dynamic Links as CSV, paste them into the Airbridge migration sheet, and complete migration in under a minute.
- **All Essential Features Included:** Deferred deep linking, QR codes, and link performance analytics with no extra add-on costs.
- **Fair Pricing:**$3 per 1,000 MAU after 10,000 MAU, billed on a pay-as-you-go basis.
Firebase Dynamic Links shuts down on August 25, 2025. Start your 30-day free trial of the Airbridge Deep Link Plan and future-proof your deep linking. Experience the difference today.
---
## Scale Affiliate Wins In SEA With Airbridge & ACCESSTRADE Integration
URL: https://www.airbridge.io/en/blog/accesstrade-integration
Language: English
Published: 2025-08-14
# **Scale Affiliate Wins In SEA With Airbridge & ACCESSTRADE Integration**
Airbridge has partnered with [ACCESSTRADE](https://accesstrade.vn/) to enhance affiliate marketing strategies across Southeast Asia. This integration brings together Airbridge’s advanced attribution and user-level analytics with ACCESSTRADE’s extensive publisher network, helping marketers unlock deeper insights, reduce wasted spend, and scale their affiliate performance faster and more effectively.
## **ACCESSTRADE’s Network for Conversion-Focused Growth**
ACCESSTRADE is Southeast Asia’s leading affiliate marketing platform, connecting over 1,000 advertisers and over 1.5 million publishers across Vietnam. Its network includes online communities, content creators, and cashback platforms that support performance-based campaigns. With a scalable ,win-win model, ACCESSTRADE drive sustainable revenue growth for both advertisers and partners.
## **Enhance Affiliate Attribution with Airbridge**
The integration between Airbridge and ACCESSTRADE makes affiliate performance more measurable and actionable. ACCESSTRADE’s offering with Airbridge is focused on:
### **Instant Conversion Tracking**
Key conversion events like app installs, sign-ups, and in-app purchases are instantly delivered from ACCESSTRADE to Airbridge. Marketers get real-time, accurate insights without manual effort.
### **Granular Attribution and Unified Performance**
Airbridge attributes each conversion from ACCESSTRADE to the exact publisher, placement, and creative that drove it ,giving marketers the granular clarity they need to make confident decisions. With affiliate performance shown alongside paid media, owned channels, and other touchpoints in a single dashboard, it’s easier to compare results, optimize budgets, and understand how every channel contributes to growth—all without switching between platforms.
# **Getting Started**
Setting up the integration between Airbridge and ACCESSTRADE is simple. Just follow these steps to start tracking and optimizing your affiliate campaigns:
- Create tracking links in Airbridge for your ACCESSTRADE campaigns.
- Configure postbacks for real-time data exchange based on your conversion goals.
- Use Airbridge reports to gain insights into user behavior and campaign performance.
- Adjust your strategies based on the data collected.
Learn more about how [Airbridge supports integration with ACCESSTRADE.](https://help.airbridge.io/en/guides/accesstrade)
Ready to scale your affiliate campaigns?
Talk to the [Airbridge](https://www.airbridge.io/) team to learn how our integration with ACCESSTRADE can help you track smarter, optimize faster, and grow stronger.
---
## Vietnam Game Connect 2025: Retention First for Long-Term Value in Casual Mobile Games
URL: https://www.airbridge.io/en/blog/vietnam-game-connect-2025-retention-first-for-long-term-value-in-casual-mobile-games
Language: English
Published: 2025-08-13
What does retention really mean now? How should studios adapt to a rapidly shifting UA landscape? And more importantly, how can product and monetization teams build games that _truly scale_?
These are just some of the questions that [**Huy Nguyen**](https://www.linkedin.com/in/huy-nguyen-36312017b/), Account Manager at [**Airbridge**](https://www.linkedin.com/company/airbridge-hq/), raised at Vietnam Game Connect 2025. The answers offered more than just insight; they served as a clear reminder of where the industry is heading. For game studios focused on sustainable growth, the priority is shifting away from flashy install numbers or ad spend towards precision, clarity, and systems that deliver _real results_.
[]
Here’s a quick recap of the key questions and insights that bring "Retention First" to life and show what it takes to drive long-term value.
## "**How do you predict the future of UA in the next 12 months?**"
For [**Lion Studios**](https://www.linkedin.com/company/lionstudioscc/)’ Director at Studio Management, [**Chansun**](https://www.linkedin.com/in/fisheatbananas/), the biggest change in UA (User Acquisition) is how little manual work it now requires and how much creative strategy it demands instead.
[Vietnam Game Connect 2025: Retention First for Long-Term Value in Casual Mobile Games]
> “What is important these days is that you feed the algorithm the right set of creatives so that your campaign can scale better.”
UA teams no longer need to micromanage campaign structures and settings. Instead, they need to deeply understand platforms like TikTok and Meta, and deliver creatives that help those platforms do what they’re designed to do: find and scale to the right users. This shift is driven by mounting privacy pressures, such as Apple’s ATT, which restricts user-level data and limits targeting accuracy, thus making creative relevance and UGC essential for reaching the right audience and driving conversions with authenticity.
One more thing to keep in mind, UA is not a single solution to the problem. The product itself needs to support long-term retention, higher LTV, low CPI in order for UA campaigns to scale better.
## “**How is user acquisition evolving in today’s mobile gaming market?**”
[**Jay Heo**](https://www.linkedin.com/in/jayheo/), Chief Business Officer at [**Playio**](https://www.linkedin.com/company/gna-company-playio/), emphasized that reward systems are evolving fast, and AI is right at the center of it. Playio’s AI is built to better understand player behaviour and deliver value not just to users, but also to game companies.
[Jay Heo, Chief Business Officer at Playio]
Old models like offering a one-time reward (e.g., $1 for a download) have nearly zero retention value. The new approach is dynamic and behavior-based: rewards are spaced out by level or action, and continuously optimized by AI to maximize engagement and lifetime value.
The ultimate goal? Convert one-time downloaders into long-term players through smarter, deep-funnel engagement.
## “**Do you see any changes in user behavior in response to AI-driven game design?**”
[**AppLovin’s**](https://www.linkedin.com/company/applovin/) Head of Vietnam, [**Tuan Hoang**](https://www.linkedin.com/in/tuanhhoang/), answered this one with a case study. After launching their new algorithm, they saw dramatic improvements across Day 7 ROAS.
[Tuan Hoang, AppLovin’s Head of Vietnam]
What made the difference? Smarter level design.
He emphasized that moments of friction, like a player needing just two more moves to beat a level, became key monetization triggers. Whether through a rewarded ad or a one-time offer, well-placed friction drove both engagement and revenue.
## **“How are game studios shifting their goals when working with ad agencies?**
[**Helen**](https://www.linkedin.com/in/maithanha/), Head of [**Ecomdy Global Gaming**](https://www.linkedin.com/company/ecomdy/), shared a surprising shift: Studios aren’t just using TikTok for UA. They’re using it for retention.
[Helen, Head of Ecomdy Global Gaming]
Rather than running short-term paid campaigns, studios are now building long-term branded spaces on TikTok to foster community, boost organic loyalty, and reduce acquisition costs over time. It’s no longer just about going viral; it’s about staying visible in users’ feeds and keeping players engaged.
> “Studios started to build the community on TikTok so that they can keep the user staying with their game longer and longer.”
## “**In a high-failure environment like game prototyping, how can new studios keep their teams motivated and focused?**”
Chansun shared that out of hundreds of prototypes, only a fraction move forward, and even fewer make it to launch. Around 20% of them advance to the monetization stage, and among those, less than 10% are eventually published. To keep teams motivated, Chansun focuses on two key practices: sharing learnings across teams so every prototype contributes to collective knowledge, and using a bounty system that rewards developers for hitting KPIs, even if the game doesn’t launch. In other words, they reward smart thinking, not just final outcomes.
She also noted a shift in how publishers validate prototypes. While companies like Voodoo, Rollic, and Homa once relied heavily on Facebook for early testing, many are now turning to platforms like Unity and TikTok to better match evolving audience behaviors.
## “**What’s the best way to leverage rewarded ads for long-term scale?**”
Jay Heo returned to dive deeper into the mechanics behind reward-based user paths, reinforcing why structure matters.
[]
> “There should be a multi-layer engagement where, after the users download, they are incentivized to complete tasks several times to get increasingly appealing rewards.”
He explained that rather than relying on one-and-done rewards, players are encouraged to complete levels, repeat key actions, or unlock items over time. Each layer of reward adds stickiness and boosts retention. It’s part motivation system, part behavior prediction engine.
## “**What are the most common mistakes teams make with monetization strategy?**”
It’s true that many teams now focus on maximizing eCPM by adding as many ad networks as possible to their stack to optimize bidding. But Tuan Hoang pointed out a key metric that often gets overlooked: ROAS multipliers. Instead of just looking at Day 0 or Day 7 revenue, teams should track growth between stages.
> “To us, a good game should have the ROAS uplift by 3 times between Day 0 and Day 7,… and by 2.5 times between Day 7 and Day 30.”
This kind of ratio gives a clearer picture of scale potential. A game with strong early ROAS but no growth afterward is likely peaking too soon. On the other hand, steady multipliers signal growth in monetization efficiency, exactly what platforms like AppLovin look for when deciding where to push traffic.
## “**To sum up, what’s the key advice for game studios looking to drive long-term value in the in the next 12 months?**”
[]
Each speaker closed with a takeaway that captured their outlook for the year ahead:
> “Adapt fast or fall behind.”
> “Go TikTok, go viral.”
> “Diversity in your channel… and in your thinking.”
> “Retention, playtime, and viability.”
No two messages were the same, but together they reflected a fast-changing, high-stakes, data-driven industry where agility, experimentation, and systems thinking separate those who thrive from those who fall behind.
## **Supercharge Your Growth with Airbridge**
If you're ready to take your game marketing to the next level, check out [Airbridge](https://www.airbridge.io/). Our platform provides:
- **Advanced Cohort Analysis for LTV:** Track lifetime value with precision and break down performance at a granular level (keyword, creative, country, OS, and more).
- **Ad Mediation Integration:** Seamlessly combine data from multiple ad networks in one place (e.g., Google AdMob, AppLovin MAX, Unity LevelPlay).
- **Predictive LTV (pLTV) Feature:** Forecast future revenue in real-time, allowing for faster, data-driven decisions.
- **Minutely/Hourly Retention Analysis:** Understand player behavior down to the minute or hour for truly actionable insights.
- **Competitive, All-Inclusive Pricing:** Access enterprise-grade analytics without hidden fees or complicated add-ons.
Whether you're an indie studio or an established publisher, Airbridge gives you the critical tools you need to stay ahead in today's fast-paced market. [Contact our experts](https://www.airbridge.io/book-a-demo) to learn more.
---
## Measure Mi Ads In-Feed Ad Campaigns with Airbridge Attribution
URL: https://www.airbridge.io/en/blog/airbridge-x-xiaomi-integration
Language: English
Published: 2025-08-11
Airbridge has officially integrated with Mi Ads, enabling measurement of in-feed ad campaigns based on clicks and impressions. With this integration, advertisers running in-feed campaigns on Mi Ads can now receive install and in-app event postbacks through Airbridge, bringing consistency, clarity, and scale to their OEM marketing efforts.
As one of the world’s largest smartphone manufacturers, Xiaomi offers tremendous value to app marketers. By supporting in-feed ad measurement for Mi Ads, Airbridge bridges the gap between OEM inventory and reliable performance data, giving marketers the insights they need to grow.
## **What Are In-Feed Ads?**
In-feed ads appear natively within Xiaomi’s content streams, such as icon recommendation, system apps, recommended content, and curated news feeds, delivering relevant promotions in a format that blends naturally into the user experience.
Unlike third‑party banners, these placements run directly on Xiaomi's own UI surfaces, such as content hubs or recommendation widgets, capturing high attention without disrupting the user journey, making it an effective entry point for app discovery and engagement.
## **Measuring In-Feed Ad Performance with Airbridge**
In-feed advertising is on the rise, especially among OEMs like Xiaomi. For app marketers, this format offers:
- **High-quality, intent-driven traffic**
- **Device-native visibility** that captures user attention
- **Scalable reach** across Xiaomi’s global footprint
With Airbridge’s Mi Ads integration, marketers can now measure these in-feed campaigns with the same accuracy and depth as any major ad channel.
Airbridge supports both **click-to-app** and **impression-to-app** attribution. When a user taps or views an in-feed ad and installs the app, Airbridge tracks the engagement, attributes the install or event, and sends real-time postbacks to Xiaomi. All touchpoints, from impression to conversion to in-app events, are visible directly in the Airbridge dashboard.
This enables:
- **Unified campaign tracking across ad channels:** Mi Ads campaign data is fully accessible in the Airbridge dashboard alongside other ad sources, ensuring a streamlined performance overview.
- **Granular event tracking:** Marketers can analyze not only install performance, but also custom in-app events (e.g. sign-ups, purchases) tied back to Mi Ads campaigns.
- **Cohort and retention analysis:** With Mi Ads included in multi-touch attribution and cohort analytics, advertisers can assess long-term value, not just last-click performance.
- **Cross-platform alignment:** Airbridge tracks across app and web, so in-feed ads driving mobile web visits or app conversions can be measured within the same flow.
Now marketers can confidently scale with OEM inventory without sacrificing attribution accuracy or performance visibility.
## **What’s next**
This integration marks another step in Airbridge’s commitment to strengthening its OEM partnerships. As more advertisers recognize the value of OEM inventory, we expect to see continued growth in both demand and collaboration.
**Ready to run in-feed campaigns with Mi Ads?
**Reach out to your Customer Success Manager or visit [](https://www.airbridge.io)[Airbridge](https://www.airbridge.io/book-a-demo) to get started.
---
## Global Game Growth 2025: UA, Monetization, and Measurement Strategies That Scale
URL: https://www.airbridge.io/en/blog/global-game-growth-2025-ua-monetization-and-measurement-strategies-that-scale
Language: English
Published: 2025-07-02
In mobile gaming, building a great game is just the beginning. The bigger challenge? Scaling it globally. To grow beyond borders, studios need to master user acquisition, monetization strategies, payment infrastructure, and accurate data tracking.
[Global Game Growth 2025: UA, Monetization, and Measurement Strategies That Scale]
At the “Actionable Growth Passport 2025: Scale Vietnam Game Studios To Global Success” event, four mobile growth leaders—[**Airbridge**](https://www.airbridge.io/)**,** [**Bidroid**](https://www.bidroid.io/)**,** [**Pingpong**](https://vn.pingpongx.com/vn/index)**, and** [**TopOn**](https://www.toponad.net/)—shared clear, actionable strategies for scaling game studios globally. From predictive LTV to ad mediation and localization, the sessions were packed with practical tips for marketers looking to grow worldwide.
Here’s a recap of the key insights you’ll want to keep in your playbook.
## **5 Global Growth Lessons Game Studios**
- Predictive LTV and AI-driven segmentation are critical for long-term monetization.
- Smart market selection and localization increase user acquisition efficiency.
- Global payment solutions must be seamless and fully compliant to support player trust.
- Ad mediation platforms should optimize revenue without hurting UX.
- A/B testing across creatives, targeting, and monetization is essential for sustainable growth.
## **Airbridge: Scale Smarter With pLTV, Segmentation, and A/B Testing**
Airbridge CEO & Cofounder, [Roi Nam](https://www.linkedin.com/in/roinam/), kicked off the session with a bold point. Relying on Day 7 or even D30 retention is no longer enough. The team introduced **Predictive Lifetime Value (pLTV)** as a more effective way to measure and forecast user value over time.
[Roi Nam, CEO & Cofounder of Airbridge & Airflux]
### **Predictive Lifetime Value (pLTV): Forecasting User Value**
In many casual games, payback periods are getting longer, making early scaling risky. That’s why Airbridge encourages studios to look beyond short-term metrics. With pLTV, marketers can forecast user value across 90, 120, or even 180 days helping them make faster, smarter budget decisions.
Airbridge’s PLTV tailors its approach based on game monetization models:
- For IAA-focused casual games, curve-fitting methods are used
- For IAP or hybrid titles, machine learning drives accuracy
Marketers also gain granular insights by country, OS, channel, campaign, and even creative. These breakdowns help teams find their highest-value audiences and double down on what works.
One real-world example is Alarmy. The productivity app Alarmy replaced slow, manual LTV models with Airbridge’s pLTV. With near-instant predictions by channel and creative, they optimized spend faster and reduced budget waste.
👉 [_Read the full Alarmy case study_](https://www.airbridge.io/en/customers/alarmy-roas-measurement)
### **Segment Smart, Test Often: Personalized Monetization Wins**
Airbridge also stressed that one-size-fits-all monetization doesn’t work. Not every user behaves the same, so why use a single monetization strategy?
With ongoing micro-segmentation based on behavior, location, and other attributes, studios can personalize monetization dynamically. When paired with real-time A/B testing, this lets teams fine-tune interstitial ad frequency and timing to match each user segment.
[]
Airbridge showed that fine-tuning interstitial ad frequency can lift incremental IAP without hurting retention. Since the best settings vary by region and install timing, a cookie-cutter approach won’t cut it.
To support this, Airbridge recommends tools like **Firebase, Amplitude, Mixpanel, and ThinkingData** for smarter segmentation and in-product testing.
### **AI-Powered Ad Monetization: Airflux in the Spotlight**
Another highlight was [Airflux](https://www.airflux.ai/), an AI-powered ad monetization optimizer built for mobile games.
[Global Game Growth 2025: UA, Monetization, and Measurement Strategies That Scale]
Airflux doesn’t just predict LTV. It actively learns and adapts. It automates segmentation, tests interstitial ad policies in real time, and updates every 7 to 10 days using live performance data. This continuous learning loop helps studios adjust ad timing and reward values at the user-segment level.
The result? More revenue without compromising user experience. Studios using Airflux reported blended LTV gains between 13% and 55% in just a few weeks. The longer the system runs, the more it compounds—delivering stronger returns over time.
## **Bidroid: A Strategic Reset for Global UA**
Bidroid’s session challenged studios to think differently about global user acquisition. [Santhosh Malleshwara](https://www.linkedin.com/in/santhosh-malleshwara-a584177/), Business Head & Director at Bidroid, laid out a practical framework to help teams sidestep common mistakes and scale with more purpose.
[Global Game Growth 2025: UA, Monetization, and Measurement Strategies That Scale]
### **UA Isn’t Just Marketing. It’s Product Strategy**
One of the biggest missteps? Treating UA like a short-term marketing spend. Bidroid urged studios to see it as part of a broader product validation loop. It’s not just about driving installs. It’s about testing product-market fit, improving retention, and strengthening the game itself.
Going global doesn't mean charging into Tier 1 markets right away. It means starting where your product is likely to succeed. Test first, scale later. That’s how you avoid burning budget on markets you’re not ready for.
### **Pick the Right Markets, Localize With Intention**
Market selection matters, but many get it wrong. Jumping into expensive regions too early or mismatching genres to local tastes can derail even a solid title. Bidroid recommended starting with balanced test markets like Mexico or Southeast Asia to learn quickly and reduce risk.
And localization? It’s more than translation. Creatives should reflect local culture, play habits, and tone. In Bidroid’s case studies, this kind of thoughtful adaptation boosted retention by 30%, lifted ARPU, and cut CPI by over 25%.
### **Test With Sufficient Budgets, Collect Valid Data**
Another common trap is spreading small budgets too thin. Studios often try to cover too many channels at once, only to end up with shallow data and unclear results. Bidroid recommends focusing on 1 to 3 test markets with $3,000–$10,000 each, enough to generate meaningful cohort insights.
[Santhosh Malleshwara, Business Head & Director of Bidroid]
That data powers a true test–learn–iterate cycle. UA becomes continuous, not campaign-based, resulting in smarter spend, better targeting, and faster growth with fewer surprises.
## **Pingpong: Global Payment Infrastructure Built for Game Growth**
Scaling a mobile game globally takes more than great UA and monetization. It also means making sure you’re getting paid—reliably, efficiently, and across borders. Pingpong helps studios simplify cross-border payments, improve success rates, and reduce costs without added complexity. In his session, Bill Liu, APAC Regional Product Operations Director of Pingpong, broke down what makes their platform uniquely effective for game publishers.
[Bill Liu, APAC Regional Product Operations Director of Pingpong]
### **Compliance-Ready, Scalable, and Built for Global Reach**
With licenses across China, Europe, North America, and more, Pingpong provides the regulatory foundation needed for international growth. Backed by 140+ global financial institutions, the platform supports 200+ localized payment methods including credit cards, local gateways, and bank transfers.
Studios benefit from:
- Fast settlements
- Multi-currency withdrawals
- Up to 95% lower transaction fees
### **Real-World Impact: Xiaomi Achieves 90%+ Payment Success**
Pingpong highlighted its work with Xiaomi’s International Internet Division, which faced fragmented payment systems and low top-up success rates.
Pingpong’s API integrated 200+ localized payment options directly into Xiaomi’s in-app top-up flow, boosting payment success above 90% and enabling one-click top-ups. The result? Higher conversion, happier users, and stronger monetization across global markets.
### **TopOn: Hybrid Monetization with IAA and IAP**
TopOn’s Global BD Head [Cooper Pi](https://www.linkedin.com/in/cooper-pi-604b6b5a/) spotlighted a clear trend in the mobile gaming space. More studios are moving from single-format monetization to hybrid models that combine in-app advertising (IAA) with in-app purchases (IAP).
[Cooper Pi, Global BD Head of TopOn]
Between Q2 2023 and Q2 2024:
- Hyper-casual games using both IAA and IAP grew from 19% to 26%
- RPGs jumped from 41% to 51%
- Strategy games climbed from 59% to 66%
This shift reflects the push to balance user experience with revenue growth. Hyper-casual titles still lean heavily on ads, often 90%+ of revenue, while mid-core and hardcore games see the bulk from IAP, with ads contributing less than 25%. Genres like word, arcade, and sports games typically land somewhere in between.
### **Smarter Ad Mediation and Format Strategy**
A key to effective hybrid monetization is smart ad mediation and format strategy. TopOn’s ad mediation platform is designed to help game studios increase revenue per impression through tools like:
- Waterfall layering, header bidding, and hybrid sorting
- Ad format mixing, such as using splash ads in interstitial slots to boost eCPM
- Intelligent request modes that switch between impression-priority and price-priority
### **Monetization Optimization Through A /B Testing**
With built-in A/B testing, studios can fine-tune strategies based on audience behavior and monetization goals. TopOn reports up to 15% eCPM gains and 10% revenue growth in simulation and hyper-casual genres, driven by multi-layer floor pricing, real-time vendor switching, and a unified SDK that simplifies testing.
### **Growth Clinic: Real Studios, Real Questions, Real Solutions**
The event wrapped with a hands-on Growth Clinic, where top studios worked through real challenges across five key areas of mobile game growth. With expert moderators guiding the discussions, attendees exchanged insights and walked away with actionable strategies.
[Global Game Growth 2025: UA, Monetization, and Measurement Strategies That Scale]
Here are some of the most practical takeaways:
- **Expanding UA channels:** Build a checklist for creatives, tech setup, and onboarding. Prioritize ad networks by genre fit and reputation, and keep testing with small, flexible budgets.
- **Boosting ROAS after soft launch:** If Day 1 retention is high but ROAS is low, refine ad scripts, improve UX/UI, and adjust targeting by behavior or region.
- **Growing LTV with steady UA:** Segment by channel, geo, and spend level. Run continuous A/B tests to refine monetization and retention levers.
- **Tracking with confidence:** Think in layers—from installs and engagement to device types and in-game behavior like funnel drop-offs and ad fatigue.
- **Monetizing by user type:** Use Firebase to define segments and personalize offers for IAP whales, ad-heavy viewers, and low spenders.
Here’s the thing. No single tactic will take your game global. Scaling means making smart moves across UA, measurement, monetization, and payments—and making them work together. If you’re a mobile game marketer ready to level up your growth, connect with [Airbridge](https://www.airbridge.io/book-a-demo) today.
---
## The Steam Marketing Playbook for Mid-Sized Game Studios
URL: https://www.airbridge.io/en/blog/game-studio-steam-marketing-playbook
Language: English
Published: 2025-07-02
**Jetstream Sessions with Nikki DePaola, VP of Global Media at Liquid Advertising**
For most mid-sized game studios, launching on Steam isn’t a maybe. It’s the move. The platform opens the door to millions of PC gamers, offers built-in community features, and can even lead to console attention if things go well.
But there’s a catch. Most Steam marketing advice leans one of two ways. It’s either built for indie devs working with near-zero budget, or it’s tailored to AAA giants with global IPs and blockbuster trailers. If you’re somewhere in the middle — not tiny, not enormous — it can feel like no one’s talking to you.
That’s why [Roi Nam](https://www.linkedin.com/in/roinam/), CEO of Airbridge & Airflux, sat down with [Nikki DePaola](https://www.linkedin.com/in/nikki-depaola/), VP of Global Media at Liquid Advertising, to set the record straight. Nikke’s agency has launched everything from Blizzard hits to Nexon, Bethesda, and a wide range of indie and mid-sized AA titles. Nikki knows what works, and what breaks, when studios of all sizes try to make it on Steam.
Here’s what she wants mid-sized studios to know.
**🎥 Watch the full interview video**
**
**
## **Steam wishlists help, but only when they reflect real demand**
Wishlists are important on Steam. They’re one of the few pre-launch metrics you can track. They influence algorithmic visibility and indicate early audience interest. But wishlists alone won’t carry your game.
> “We’ve seen instances where clients are excited about their chart position, but disappointed by the conversion rate because most of the Wishlists were forced.”
Forced or incentivized wishlists might push your numbers up, but they don’t guarantee launch-day traction. Nikki’s advice is to focus on quality over quantity. **Her benchmark? 70% or more of your wishlists should be organic and regionally relevant**.
> “While a total of X wishlists will earn you visibility on the platform, that visibility won’t resonate strongly with your target audience unless there is real word of mouth happening in parallel.”
If your game isn’t getting organic mentions on Discord, Reddit, or YouTube comments, your wishlist chart won’t mean much. Steam rewards momentum, not just math.
[Roi Nam, CEO of Airbridge & Airflux with Nikki DePaola, VP of Global Media at Liquid Advertising]
## **A real Steam launch plan needs community, PR, and media**
A lot of studios launch with just a Steam page, a trailer, and some ad spend. That’s not enough. Players on Steam don’t convert based on one asset. They follow a trail. They want to hear buzz, see opinions, browse reactions, and read about your game from a source they trust.
> “Wishlists are basically a proxy KPI for how excited people are about your game. When you think about it from that perspective, the question isn’t necessarily about how do you hit your Wishlist goal. It becomes how do we get our consumers excited about this release.”
That shift changes everything. Your wishlist count becomes a result, not a goal. And the effort behind it expands. Nikki recommends including six building blocks in every serious Steam launch.
You need:
- A Steam page with sharp visuals, a clear hook, and a strong trailer
- Paid media that reaches players in places where they’re open to discovering games
- Community engagement, especially on Discord and Reddit
- A real PR strategy that gets you editorial coverage from outlets players trust
- A content rollout plan with staggered beats like reveals, demos, and beta launches
- A measurement setup (like **Airbridge**) that helps you understand what’s working with trustworthy data.
If you treat Steam like a one-shot launch moment, you’re doing it wrong. Think of it as a runway, not a rocket.
## **Your Steam D-DAY isn’t the finish line**
Launch day isn’t your grand finale. It’s the start of a very public test. The Steam algorithm is watching. Players are watching. And your strategy needs to stay sharp and ready to respond. Beyond having a fully customized strategy and media approach, there are a few common must-haves to keep in mind.
> “One is having a plan B in place if you are anticipating mixed reviews or worse. How can you make the most out of a suboptimal situation?”
Some games crush Day 1. Others stumble. Either way, you need a system that lets you respond fast. Nikki recommends setting up measurement tools ahead of time so you can spot early trends and pivot quickly.
> “Another tip is to have a measurement platform installed ahead of time, to gauge how Day 1 and Day 3 are different—if you have a team to activate quickly and if the campaign architecture is designed to surface actionable splits for creative, audience, and territory.”
The sooner you know what’s working, and what isn’t, the faster you can course-correct. If your game supports logins or online play, don’t waste that edge. First-party data gives you an advantage in LTV modeling, retargeting, and regional insights.
[]
## **Brand and performance aren’t rivals. You need both**
Until recently, PC and console game marketing was all about brand. All eyes were on cinematic trailers, press buzz, and influencer hype. Performance marketing? That felt too mobile, too transactional. It was all clicks, installs, and ROAS, with not much room for storytelling or world-building.
But that’s starting to change. As more studios adopt live-service models and early access, data-driven strategies are becoming a key part of the marketing mix.
Nikki explains that brand and performance marketing aren’t at odds. In fact, they work best together. The most effective studios are blending the strengths of both. Nikki sees a growing convergence between two types of marketers:
- **Mobile game marketers**, who are highly data-driven and ROI-focused
- **PC or console game marketers**, who understand branding, positioning, and audience psychology
> “Games marketers are adopting behaviors of more mature categories, which is a balance of media types based on timeless principles like reach, frequency, attention, and share of voice.”
This balance is especially important for mid-sized teams. Mobile marketers bring sharper KPIs and tracking methods, while PC marketers often bring creative instincts and a strong sense of tone. When both sides align on full-funnel goals, campaigns move faster and deliver better results.
## **PC·console players want proof, not hype**
PC·console players don’t tap ads between subway stops. Unlike mobile players, they don’t download on impulse, and they rarely convert after just one ad exposure. Instead, they browse, discuss, compare, and decide on their own timeline.
> “Patience is your biggest virtue. Results are less immediate, and audiences require more exposures to convert.”
Also, PC·console players don’t just see your ad. They see everything around it. The placement, the reviews, the trailer, the context, and the community conversation all influence how they perceive your game. That’s why Nikki advises focusing on discovery moments, not just raw impressions.
> “Where can you reach gamers when they are in a mood for discovery? Context matters in ways that it doesn’t on mobile.”
Be thoughtful about where your ads show up. A cheap placement on a low-quality site (what Nikki calls a “made-for-advertising” site) might drive clicks, but it can hurt your brand image. In PC and console, adjacency matters more than ever.
> “Optimize towards attention and impact. Conversion will follow—even if attribution is more challenging on those ad placements.”
## **Don’t sleep on these underused channels**
If you’re already running on Meta, Google, and YouTube—great. That’s the baseline. Now let’s talk about what you aren’t doing that could give you an edge.
Nikki calls out three areas that are still flying under the radar for many studios, especially those in APAC.
- **Community media (eg. Discord, Reddit):** Everyone knows Discord and Reddit are where gamers live. But knowing that and actually showing up in a meaningful way? Two very different things. It’s not just about pushing announcements or reposting trailers. These platforms are where real conversations happen. It’s where players get excited, vent frustrations, trade tips, and share memes that end up shaping how your game is perceived.
And in the PC and console space, where players care deeply about lore, quality, and community, this kind of engagement is what really counts.
- **High-touch programmatic campaigns:** Most teams focus on the bottom of the funnel. Retarget, retarget, retarget. But PC and console buyers don’t just click and buy. They pause. They compare. They think it through. That’s why Nikki stresses the importance of a full-funnel programmatic strategy. One that spans upper, mid, and lower funnel stages. With smart sequencing, frequency, and audience targeting, programmatic can guide players from awareness to conversion in a way that doesn’t feel like advertising. It feels like timing.
- **OTT and CTV advertising (e.g., Netflix, Hulu, and Disney+):** OTT and CTV are still underused, partly due to measurement accuracy or because they’re not dominant in some non-Western markets. But in the U.S. and Europe, CTV is anything but niche. OEMs like Samsung, LG, and Vizio provide data that shows what players are watching, and in some cases, whether they played a game after seeing an ad. Now, with more ad inventory opening up across Netflix, Hulu, Disney+, and other streaming platforms, there’s more opportunity than ever to reach the right players at the right moment while they’re already in discovery mode.
[The Steam Marketing Playbook for Mid-Sized Game Studios]
## **Creative clarity beats creative volume**
PC and console creatives are more about precision than volume. Unlike mobile, where you test as many variations as possible and optimize after the fact, PC campaigns rely on carefully crafted messaging that speaks to the right audience in the right way.
Nikki’s advice for mid-sized studios:
- **Tailor your message to the audience:** Build creative that speaks directly to your core segments. Focus on the game’s standout features, and test across different audience groups to see what actually resonates.
- **Be clear about genre and USP:** Don’t be vague. Communicate what your game is, who it’s for, and what makes it different.
> “Sometimes it’s really difficult to tell what the game is by the assets. Don’t just say ‘shooter.’ Say ‘80s horror co-op survival FPS with roguelike elements.’ You need strong visuals and clear communication on what your game is about.”
- **Use high-quality key visuals to communicate brand and position:** Your visuals should signal quality, polish, and where your game sits in the market. This is especially important if you’re trying to level up from AA to AAA perception.
> “To be truly triple AAA it’s about communicating a vibe.”
## **Cross-platform strategy isn’t one message fits all**
More studios are going multi-platform. Steam, console, mobile, even HTML5 and web. On paper it makes sense. Broader reach, more monetization paths, and the flexibility to meet players where they already are.
And yes, the strategy can work, but only if it’s grounded in how different audiences think about platforms, especially in markets like the US and Western Europe.
In APAC, mobile is mainstream. But in the West, core gamers on mobile are still a niche group. Console and PC ownership is solid, but when Western players think core gaming, they’re picturing shooters with a controller in hand, not a touchscreen.
> “If you’re serving ads and messaging to core console gamers about the game being available in the App Store, their perception may be that it’s a different type of mobile game — not the one they play on their PlayStation with their bros.”
That disconnect can hurt you, even if the game is identical across platforms. The wrong message in the wrong place can make your game feel like it doesn’t belong.
So how do you approach cross-platform marketing without confusing or alienating anyone? There’s no single right answer, but you do have options.
- **Segment your audiences:** Treat console players, PC players, and mobile first users as separate groups with different expectations and language
- **Adjust the messaging:** Lead with platform specific value. Emphasize high-end visuals and controls for console players. Highlight accessibility or convenience for mobile
- **Or keep it neutral:** If you’re worried about splitting the message too much, it’s also valid to stay platform agnostic and let players figure out what works for them without drawing too much attention to where the game lives
## **Full-funnel measurement matters more than ever**
Last touch attribution is where most game marketers begin. And sure, it’s a helpful starting point. But as your campaign scales, so should the way you measure impact.
Players almost never convert after seeing just one ad, especially not on channels like Instagram. It takes repetition, clarity, and the right message at the right time.
> “Full funnel measurement is super important for any game. At the end of the day, very little people are seeing an ad for the first time on Instagram and converting right away. It takes building frequency with a consumer over time, and giving them the right messages to ensure they are building familiarity with the product, and driving them towards the purchase or install.”
If you’re only tracking last touch clicks, you’re missing the bigger picture. Channels like linear TV, out of home (OOH), or console media might not live in the same ecosystem, but they are shaping player decisions all the same.
Cross-platform or not, the idea is the same. Match your message to each stage and use tools that show how it all works together. That’s how you build campaigns that convert and keep converting.
[The Steam Marketing Playbook for Mid-Sized Game Studios]
## **Entering the U.S.? Avoid these common missteps**
Breaking into the U.S. market isn’t just about translating your ads or copying what worked for some other gaming studio. Without a deep understanding of local behavior, budget norms, and cultural touchpoints, studios risk wasting spend and missing momentum.
For APAC studios heading West, Nikki shares three common pitfalls:
- **Copycat behavior:** It’s tempting to mimic a competitor’s marketing plan, especially as APAC studios tend to be more risk averse, but what worked for them might not work for you. Every game has a different budget, audience, and challenge.****
- **Underestimating budget requirements:** Many APAC publishers assume they can compete with half the spend. But in a premium market, without proper investment, or at least strong measurement, it’s more rare to make a dent. ****
- **Overlooking cultural nuance:** The U.S. is fragmented, trend driven, and packed with subcultures. If you don’t have someone local to guide creative and media mix, you’ll miss what actually moves the needle.
> “Your game isn’t that other game. Your game’s challenges are likely not another game’s challenges, and your game’s budget is probably not even the same.”
## **Small-budget indie games can win on their own terms**
We’ve talked a lot about big-budget launches, but that’s not the only path to success. So what does a win look like for a small studio? Nikki points to [_Squirrel with a Gun_](https://store.steampowered.com/app/2067050/Squirrel_with_a_Gun/), an indie title that’s exactly what it sounds like.
> “It’s basically Squirrel Simulator, except you have a gun. You can’t treat that like a AAA title with takeovers and polished advertorials. It just wouldn’t land.”
Instead, Nikki’s team focused on what made the game special.
> “It’s about leaning into the brand identity of the game. At its heart, this game is silly and fun. How can we amplify that? So we ended up doing a lot of organic-feeling, community-driven media activations—like working with Discord servers to create memes, and working with meme accounts on Instagram to promote content.”
The team paired that energy with smart remarketing and lower-funnel support. The result? A campaign that felt true to the game and delivered real momentum. There’s no silver bullet when it comes to Steam success—but with the right message, the right measurement, and a plan that fits your game, mid-sized studios can absolutely make their mark.
## **Key takeaways for mid-sized studios launching on Steam**
- **Steam wishlists only work when they reflect real excitement:** Organic buzz and regional interest beat forced wishlist numbers every time.
- **Steam isn’t just a store, it’s a discovery engine:** Build momentum through PR, content, community, and media—not just a trailer and a page.
- **Your Steam launch day is just the beginning:** Track early signals, prepare for Plan B, and stay nimble for Day 2 and beyond.
- **Marry brand and performance:** The strongest campaigns mix bold creative with smart targeting and full-funnel strategy.
- **Tailor your creative to the player, not just the platform:** One message won’t resonate everywhere. Segment by audience and speak their language.
- **Going West? Know your market:** Avoid copy cat campaigns. What worked for another studio won’t always work for you. Budget accordingly, localize wisely, and hire people who get the culture.
For more insight, visit [liquidadvertising.com](http://liquidadvertising.com) and [airbridg.io.](http://airbridg.io)
### **Take Your Game to the Next Level with Airbridge**
Understand your players across every platform—PC, web, console, and mobile. Airbridge helps you track cross-platform user flows and optimize performance with powerful, real-time insights. Here’s what you get:
- **Advanced Cohort Analysis for LTV**: Track lifetime value with precision and break down performance at a granular level (keyword, creative, country, OS, and more).
- **Ad Mediation Integration:** Seamlessly combine data from multiple ad networks in one place (e.g., Google AdMob, AppLovin MAX, Unity LevelPlay).
- **Predictive LTV (pLTV) Feature:** Forecast future revenue in real-time, allowing for faster, data-driven decisions.
- **Minutely/Hourly Retention Analysis:** Understand player behavior down to the minute or hour for truly actionable insights.
- **Competitive, All-Inclusive Pricing:** Access enterprise-grade analytics without hidden fees or complicated add-ons.
- **Export Audiences and Raw Data with Ease:** Quickly extract target segments and export raw data to CSV or Google Sheets for custom dashboards or deeper analysis.
Whether you're an indie studio or an established publisher, Airbridge gives you the critical tools you need to stay ahead in today's fast-paced market. Visit airbridge.io to learn more.
### **Build Bold Campaigns with Liquid Advertising**
For creative and media strategy, cross-platform media planning that address modern games marketing challenges:
- Connect with [Nikki](https://www.linkedin.com/in/nikki-depaola/) and the Liquid team to craft a strategy tailored to your game.
- Visit [liquidadvertising.com](https://www.liquidadvertising.com) to see how top studios are launching on Steam, console, and beyond.
- Follow [Liquid Advertising on LinkedIn](https://www.linkedin.com/company/liquidadvertising) and [Instagram](https://www.instagram.com/liquidadvertising/?hl=en) for campaign breakdowns, industry insights, and creative inspiration.
(for Japan, follow us on [Note](https://note.com/liquid_ad_agency) and [X](https://x.com/liquid_ad_jp))
(For China, follow us on [Weixin/WeChat](https://mp.weixin.qq.com/s/6AJfTq5KltEtcdJINypAaw))
---
## Meta AMM: Get Row-Level Meta Attribution via Airbridge
URL: https://www.airbridge.io/en/blog/meta-amm-support
Language: English
Published: 2025-06-18
Airbridge now supports Meta’s Advanced Mobile Measurement(AMM), and that’s good news for app advertisers who rely on granular data to optimize campaign performance.
Starting **June 19, 2025**, advertisers who sign Meta’s AMM terms will be able to access non-aggregated last-touch Meta attribution data through Airbridge’s standard reporting infrastructure.
## What is AMM?
Advanced Mobile Measurement(AMM) is reporting through Mobile Measurement Partners(MMPs) like Airbridge, that enables advertisers to access row-level last-touch attribution data tied to Meta campaigns. Unlike aggregated reports, AMM data lets advertisers analyze each attributed conversion at the row level, supporting business intelligence tools and internal models.
This isn’t a new integration but a relaunch. Meta originally paused AMM in 2021. Since then, advertisers have only been able to receive aggregated performance data for Meta campaigns.
Now, that’s changing.
## What’s Changing (and What’s Not)
As an official Meta Mobile Measurement Partner, Airbridge already receives last-touch attribution data for Meta campaigns. But since Meta sunset AMM in 2021, that data has only been available to advertisers in aggregated form.
Starting **June 19, 2025**, advertisers who accept Meta’s AMM terms will be able to access non-aggregated, row-level conversion data.
This includes:
- Conversion-level attribution info for Meta ads
- Row-level data within your existing MMP reporting
- More precise measurement in your BI or analytics tools
Worth noting:
- Airbridge’s integration with Meta remains the same. There is no change to how we receive or process attribution data.
- This update only impacts the granularity of the data advertisers receive from Airbridge. Ad performance reporting itself will not be affected.
## What You Need to Do
To start receiving row-level attribution data from Meta campaigns:
1. **Sign Meta’s AMM Terms** [**Here**](https://www.facebook.com/ads/manage/advanced_mobile_measurement/app_based_tos)
2. Ensure you have the **Administrator** role for the relevant Meta developer app(s) to accept AMM Terms
- [More on app roles](https://developers.facebook.com/docs/development/build-and-test/app-roles/)
- [Help article on managing roles](https://www.facebook.com/business/help/)
Once your AMM terms are accepted, you will begin receiving non-aggregated Meta attribution data directly through the Airbridge dashboard.
## Key Takeaway
- **Meta re-enables AMM**: Meta is re-enabling non-aggregated last-touch reporting for advertisers who sign its AMM terms.
- **Airbridge supports it**: As a Meta MMP, Airbridge will pass this data to you automatically, updated every 24 hours once you opt in.
- **What’s required?** Sign the AMM terms (App Administrator role is required).
- **What’s the benefit?** Greater visibility into Meta ad performance across your MMP and analytics tools.
- **Any impact on current performance?** None. This update does not change how attribution is measured or reported, only how detailed your access is.
If you’re an advertiser looking to improve your modeling, measurement, or LTV analysis, the return of AMM is a meaningful step forward. Reach out to [Airbridge](https://www.airbridge.io/) if you need help getting started.
---
## Top 15 Subscription App Experts and Resources to Watch in 2025
URL: https://www.airbridge.io/en/blog/15-subscription-app-experts-resources-2025
Language: English
Published: 2025-06-11
In the fast-paced world of subscription apps, staying ahead of the curve is crucial for success. To help you navigate the latest trends and strategies, we've compiled a list of 9 essential resources and 6 influential experts in the industry. Make sure to subscribe! These insights will help you refine your app's growth strategy and stay competitive in the market.
## **9 Must-know subscription app resources**
### **The “Subscription Stack" blog series by Phiture**
[]
Phiture, a leading mobile growth consultancy, developed The _"Subscription Stack",_ a framework and comprehensive guide to optimizing subscription-based apps. This blog series covers key topics like subscriber acquisition, conversion rate optimization, user engagement, and churn prevention, offering actionable insights for app developers and marketers. Phiture has developed additional industry-leading app marketing frameworks, including the [Mobile Growth Stack](https://phiture.com/mobilegrowthstack/what-is-the-mobile-growth-stack/) and [ASO Stack](https://phiture.com/asostack/introducing-the-aso-stack-2019-7d88530d64d2/).**✅** [**Download The “Subscription Stack”**](https://phiture.com/subscriptionstack/)**👍 Must-read blogs**
- [Part 1: Subscriber Acquisition](https://phiture.com/mobilegrowthstack/the-subscription-stack-subscriber-acquisiton/)
- [Part 2: Conversion Rate Optimization (CRO)](https://phiture.com/mobilegrowthstack/the-subscription-stack-conversion-rate-optimization/)
- [Part 3: Activation, Engagement, and Churn Prevention](https://phiture.com/mobilegrowthstack/the-subscription-stack-activation-engagement-churn-prevention/)[](https://phiture.com/mobilegrowthstack/the-subscription-stack-subscriber-acquisiton/)
- [Part 4: Tools and Infrastructure](https://phiture.com/mobilegrowthstack/the-subscription-stack-subscriber-acquisiton/)
### **App Masters’ YouTube channel by Steve P. Young**
[]
Steve P. Young, founder of App Masters, hosts a YouTube channel packed with actionable insights on app marketing and growth strategies. You'll find interviews with successful app founders and experts sharing tips on App Store Optimization (ASO), Apple Search Ads (ASA), pricing strategies, paywall optimization, and user acquisition tactics.
**✅** [**Subscribe to App Masters’ Youtube Channel
**](https://www.youtube.com/channel/UCSyXF669KRoj65UFVBC66Xw)**👍 Must-watch videos**
- [How to Scale ANY App to $1M a Month](https://www.youtube.com/live/_6sis7wLAow)
- [App Marketing Strategies That Led to Headway's 27M Downloads](https://youtube.com/live/MLUECsQdjDQ)
- [Copy This $6M App Monetization Strategy](https://www.youtube.com/watch?v=ODWgul9tXus&t=1s)****
### **The "Sub Club" podcast & "State of Subscription Apps" report by RevenueCat**
[]
RevenueCat’s _"Sub Club"_ podcast explores the latest trends in subscription apps through insightful discussions with industry leaders. Their [_"State of Subscription Apps"_](https://www.revenuecat.com/state-of-subscription-apps-2024/) report offers an in-depth analysis of the subscription app landscape, providing benchmarks and data-driven insights into pricing strategies, packaging models, growth trends, and more.
**✅** [**Subscribe to the "Sub Club" Podcast**](https://subclub.com/)**👍 Must-listen episodes**
- [How to Maximize Revenue with Regional Pricing — Dmitry Gurski, Flo](https://subclub.com/episode/how-to-maximize-revenue-with-regional-pricing-dmitry-gurski-flo)
- [Lessons from 121 A/B Tests - Kenneth Schlenker, Opal](https://subclub.com/episode/lessons-from-121-ab-tests-kenneth-schlenker-opal)
- [Growing to $1M MRR with Paywall and Pricing Experiments — Francescu Santoni, Mojo
](https://subclub.com/episode/growing-to-1m-mrr-with-paywall-and-pricing-experiments-francescu-santoni-mojo)
### **The “Intelligent Artifice” newsletter by Shamanth Rao**
[]
_"Intelligent Artifice"_ is a weekly newsletter that helps subscription app marketers win in a generative AI world. Intelligent Artifice is written by Shamanth Rao, founder of the Gen-AI focused growth agency [Rocketship HQ](http://rocketshiphq.com/), who are Meta, Google, and TikTok partners and have managed over $100M in ad spend with leading subscription apps including Bumble, BetterMe, Lingokids, Mojo, and Recime, among others. Shamanth is also the host of the podcast [Mobile User Acquisition Show](https://mobileuseracquisitionshow.com/) and has [written 15 playbooks](https://www.rocketshiphq.com/playbooks/) on different aspects of mobile marketing.
**✅** [**Subscribe to the Intelligent Artifice Newsletter**](https://www.youtube.com/channel/UCSyXF669KRoj65UFVBC66Xw)**
👍 Must-read posts by Shamanth**
- [I deconstructed Headway’s ad creatives(running over 50,000 ads). Here is how I’d improve their creative strategy.](https://www.linkedin.com/posts/shamanthrao_i-deconstructed-headways-ad-creativesrunning-activity-7255527742727708672-7WrQ/)
- [Rise Science is making ~$6mm a year. I analyzed 1822 of their display ads - here is what I learned.](https://www.linkedin.com/posts/shamanthrao_rise-science-is-making-6mm-a-year-i-activity-7281058131646832641-9r7C/)
- [Liven is making ~$500k/mo - with primarily ‘ugly ads.’ I ran semantic analyses of 364 of their videos - here is what I learned.](https://www.linkedin.com/posts/shamanthrao_liven-is-making-500kmo-with-primarily-activity-7288578827595767809-KSOr/)
### **The "Growth Gems" newsletter by Sylvain Gauchet
**
[]
Sylvain Gauchet's _"Growth Gems"_ newsletter curates valuable insights on app growth strategies, marketing trends, and industry news. It features expert opinions from across the subscription app community, offering a diverse range of perspectives and actionable advice for app developers and marketers.
**✅** [**Subscribe to the "Growth Gems" Newsletter**](https://growthgems.co/)**
👍 Must-read Newsletters**
- [Should you have a hard paywall?](https://growthgems.substack.com/p/should-you-have-a-hard-paywall)
- [Should you discount your subscription?](https://growthgems.substack.com/p/should-you-discount-your-subscription)
- [Onboarding, Activation and Monetization](https://growthgems.substack.com/p/growth-gems-120-onboarding-activation)
### **The "State of In-app Subscriptions" report by Adapty**
[]
Adapty’s _“State of In-app Subscriptions”_ report provides deep analysis of in-app subscription trends to help you optimize your monetization models and stay ahead of market dynamics.
**✅** [**Download The "State of In-App Subscriptions" Report**](https://adapty.io/state-of-in-app-subscriptions/)**👍 Must-read blogs**
- [Turn your side project into $1K MRR](https://adapty.io/ebooks/roadmap-to-1k-dollars-mrr/)
- [Your roadmap to $10K MRR starts here](https://adapty.io/ebooks/roadmap-to-10k-dollars-mrr/)
### **The “Subscription League” podcast by Purchasely**
[]
Purchasely’s podcast focuses on subscription optimization and monetization strategies. It provides actionable advice for increasing subscription revenue through expert interviews and case studies.
**✅** [**Subscribe to the "Subscription League" Podcast
**](https://subscriptionleague.com/)**👍 Must-listen episodes**
- [App Store: The Good, the Bad, and the Glitches](https://subscriptionleague.com/episode/app-store-the-good-the-bad-and-the-glitches-q4hs)
- [Subscriptions: The Good, The Bad And The Overused](https://subscriptionleague.com/episode/subscriptions-the-good-the-bad-and-the-overused)
- [Scaling Mojo: How to Build a Subscription App with 40M+ Downloads](https://subscriptionleague.com/episode/scaling-mojo-how-to-build-a-subscription-app-with-40m-downloads)
### **Newform’s YouTube channel by Hamza Alsamraee
**
[]
Be the first to follow NewForm’s brand-new YouTube channel—hosted by Hamza Alsamraee, Co-Founder and CEO of NewForm AI. The channel dives into how today’s top app marketers are using AI to run smarter performance marketing and scale faster. Expect honest, in-the-weeds conversations with founders and CMOs across the app industry—covering growth tactics, bold experiments, and the latest performance marketing trends.
**✅** [**Subscribe to NewForm’s YouTube Channel**](https://www.youtube.com/@NewFormAI/videos)**👍 Must-watch videos**
- [Superwall's Jake Mor](https://www.youtube.com/watch?v=0_WKb05QuRk)
- [Soundmap's Zibo Gao](https://www.youtube.com/watch?v=8fovsU-pNOk)
- [Liner's Alex Yoon
](https://www.youtube.com/watch?v=7RlbFaF7xUg)
### **The subscription app webinars by aix Inc.**
[]
Aix Webinars dive into subscription growth strategies for Korea and Japan—two of the most advanced and competitive markets in the app economy. Led by local experts, each session unpacks proven tactics for onboarding, paywall optimization, churn reduction, and LTV growth. If you're looking to scale in Asia, this is the playbook you need.
**✅** [**Check out aix Inc.’s upcoming webinars
**](https://aixpost.com/webinars/)**👍 Must-watch webinars**
- [Apps & Games: Successful Culturalization in Korea](https://www.youtube.com/@aixincmarketing)
- [Creatives in Japan: Does Japan Really Like Sakura?](https://www.youtube.com/watch?v=4X7rAmK8XLU)
- [Onboarding & Paywall in Japan: Where Little Changes Make Big Returns](https://www.youtube.com/watch?v=S-UYb7EwpmI)
- [Japan Apps & Games: What We Learned in 2024 for 2025
](https://mailchi.mp/aixinc/japan-apps-games-2024-2025)
## **6 Must-follow subscription app experts**
1. [**Marcus Burke**](https://www.linkedin.com/in/marcusburke/) Marcus Burke is a leading expert in mobile growth and user acquisition, known for his strategic insights into app marketing and monetization. His expertise in analyzing Meta campaigns, web-to-app strategies, SKAN, and iOS user acquisition is considered top-notch in the industry.
2. [**Steve P. Young
**](https://www.linkedin.com/in/stevepyoung/)As the founder of App Masters, Steve P. Young is a highly renowned figure in app marketing. He has spent over 15 years helping app founders grow their apps. He's known for providing practical advice on app store optimization and early-stage growth. His quick tips and tactical guidance have been invaluable to many early-stage app founders.
3. [**Hamza Alsamraee
**](https://www.linkedin.com/in/hamza-alsamraee/)Hamza Alsamraee is an experienced mobile growth strategist and CEO of Newform, a tech enabled performance marketing agency. Hamza and Newform focus on scaling apps through short form first user acquisition strategies. His expertise includes user-generated content (UGC), short-form videos, social media marketing, and platforms like TikTok, Reels, and Shorts.
4. [**Nathan Hudson
**](https://www.linkedin.com/in/nathan--hudson/)Nathan Hudson is an expert mobile growth consultant specializing in app store optimization and user acquisition, with a focus on data-driven decision-making. His expertise helps marketers optimize their app visibility and acquire new users effectively.
5. [**Hannah Parvaz**
](https://www.linkedin.com/in/hnpvz/)As the founder of Aperture, Hannah Parvaz is a leading voice in subscription app growth. She specializes in full-funnel strategies—covering user acquisition, onboarding, retention, and monetization—for fast-growing subscription apps. Known for running high-impact growth experiments that deliver sustainable results, Hannah brings tactical insight and a hands-on approach that make her a trusted partner to early-stage founders and product teams alike.
6. [**Minki Lee**
](https://www.linkedin.com/in/minki-lee-8ab024145/)Minki Lee is a subscription app growth expert with deep experience in the Japanese and Korean markets. As CSO and Co-founder of aix Inc., he specializes in AI-powered marketing strategies that drive user engagement, reduce churn, and maximize LTV. His expertise lies in localizing growth tactics—adapting messaging, pricing, and onboarding to fit the unique behaviors of users in each market.
Engaging with the right expert and resources can supercharge your drive for growth and success in the subscription app space. By leveraging the insights from these 15 experts and resources, you can refine your strategies and scale your app's growth across all aspects of your subscription-based app.
Now it’s time to put these strategies into action! With features tailored specifically for subscription app marketing, **Airbridge** enables you to optimize campaigns and achieve global success. Our comprehensive platform enables you to:
- Optimize Apple Search Ads (ASA) at the keyword level using real-time analysis to target the right audience.
- Boost UGC campaigns on TikTok, Meta, and other platforms to drive more installs and conversions.
- Streamline web-to-app campaigns with our robust Web and Mobile SDK to track user journeys seamlessly.
- Monitor funnel conversion rates across key stages like onboarding, sign-up, trial starts, subscriptions, and re-subscriptions to identify areas for improvement.
- Use SKAN to gain insights into iOS performance, especially for trial starts.
- Retarget effectively by building behavioral audiences for dropped-off or churned users.
- Integrate with RevenueCat and Adapty for full-funnel visibility and analysis.
By combining these expert insights with Airbridge's powerful features, you're ready to unlock new opportunities for growth and success in the competitive subscription app market. So, why wait? Subscribe to success with Airbridge today!
---
## App Marketing in Japan vs. Korea: How to Localize Creatives, UA, and Paywalls for Growth
URL: https://www.airbridge.io/en/blog/win-app-marketing-boost-growth
Language: English
Published: 2025-05-22
**In Part 1** of this series, we explored the biggest misconceptions about launching apps in Japan and Korea—two culturally rich yet fundamentally different markets. We debunked the myth that they can be approached with a single go-to-market strategy and emphasized the importance of localized thinking. Now, in Part 2, we shift from mindset to execution.
Drawing from our conversation with Allen Park, Head of Marketing at aix—a consultancy that helps mobile apps grow through deep localization and thorough market strategy in Japan and Korea—we’ll walk through what it takes to localize effectively across creative, UA, onboarding, and monetization touchpoints.
## **Creative Strategy: Why One Size Doesn’t Fit All**
For many global teams, creative localization begins with translated screenshots and ads. But Japan and Korea demand more than language—they require distinct narratives.
> **“Even the same product needs two completely different faces in these two markets. Japan is about assurance. Korea is about urgency.”
> — Allen Park, Head of Marketing at aix**
Japanese users value clarity and completeness. In Japan, app store screenshots are often text-heavy, with banners that explain features in detail. For example, OTT service Unext uses copy like "Watch all episodes" to reassure users of content access. Korean users, on the other hand, move fast. TVing emphasizes trends and exclusivity with punchy phrases like "Only on TVing."
[__wf_reserved_inherit]
In games, Japanese creatives favor cute or familiar characters with soft tones. Korean creatives lean into intense visuals and strong, impactful characters that create a sense of FOMO.
[__wf_reserved_inherit]
Visuals are not decoration—they're your first impression. And if you reuse the same creatives in both markets, you're likely missing the mark in at least one.
## **UA Channels: Global First, Local When It Makes Sense**
As we explored in Part 1, many marketers assume local ad platforms are essential from the start. But in early-stage growth, global channels like Google, Meta, TikTok, and Apple Ads often offer the fastest, most efficient results.
Japan and Korea both support local opportunities—especially in reward-based media. Japan's longstanding Poikatsu culture (earning rewards for small actions) makes offerwalls particularly effective. Channels like Moppy, Point Income, or Ad networks like AMoAd, and Zucks Ad Network will help you gain more growth in Japanese market.
In Korea, reward ads are newer but quickly catching on, particularly via webtoons. Platforms like AdPopcorn and TNK Factory offer strong performance for mid-funnel conversion. Once you've found traction, scaling into Naver and Kakao can amplify results.
> **“Don’t romanticize local channels from day one. Test what works, then layer in local DSPs once you have signal.”
> — Allen Park**
## **Onboarding & Paywalls: Match the Market’s Tempo**
From first launch to first payment, Japan and Korea follow very different rhythms.
Take the example of a hiking app launched in both markets. In Korea, onboarding was fast and minimal—short headings, quick permissions, and immediate access. It fit users’ expectations to dive right in.
In Japan, the same app used a longer flow with detailed explanations and infographics. Rather than causing friction, it built trust. Japanese users prefer to fully understand a product before engaging—especially when it involves real-world activity.
[__wf_reserved_inherit]
The same philosophy applies to paywalls. Japanese users are open to long-form explanations, comparisons, and FAQs. They take longer to decide, but reward clarity with loyalty. Korean users respond to urgency. Clear discounts, review counts, and bold CTAs convert better than feature lists.
[__wf_reserved_inherit]
> **“Japanese users may take the longest time from install to first subscription—but they will reward you with better retention and better engagement.”
> — Allen Park**
## **Real-World Wins**
These aren’t just theories. aix has supported a wide range of apps seeing measurable lift through localization.
- **Tiimo**, a European productivity app, gained 70% more downloads in Japan after localizing its store listing and keyword strategy.
- **TimeFlik**, a Korean watch face app, improved conversion by 60% and cut Apple Search Ads CPI by 46% after tailoring screenshots to Japanese preferences.
> **“Localization isn’t about starting over—it’s about unlocking the growth already within reach.”
> — Allen Park**
## Where to Start: Prioritizing Your Localization Efforts
Allen Park’s advice?
> “**Ideally, ASO, onboarding, and paywalls should all be localized at once. But if you have to prioritize, start with ASO—it’s your first conversion gateway. Then work down the funnel.**”
aix typically begins with 1–2 weeks of market research and competitive benchmarking, followed by 2–4 weeks of UX and creative updates. With localized assets in place, ongoing A/B testing helps teams refine fit and maximize ROI.
Localization isn’t a one-time fix—it’s a continuous loop of listening, adapting, and improving.
## **Key Takeaways: Two markets, two strategies— one localization mindset**
Japan and Korea may be neighbors—but when it comes to user behavior, creative preferences, and conversion psychology, they couldn’t be more different. A successful go-to-market strategy in this region requires two distinct approaches.
Here’s a quick comparison of what makes each market unique across key areas of growth:
[__wf_reserved_inherit]
## **Turn Strategy into Action**
Japan and Korea reward teams that localize deeply—not just in language, but in UX, creative, and timing. If you're ready to move from insight to impact, now’s the time to take the next step.
### **Win Subscription Growth in Asia with aix**
aix helps subscription and gaming apps go beyond translation and into true cultural fit. From keyword research to paywall UX, their team brings decades of localization expertise rooted in Japan and Korea.
> → Build a localization strategy that actually works [— with aix](https://www.aixinc.io/en)
### **Subscribe to Success with Airbridge**
Airbridge gives growth teams full-funnel visibility—so you can measure, optimize, and scale what works.
• Optimize Apple Search Ads (ASA) with real-time keyword insights
• Launch and track UGC campaigns across TikTok, Meta, and more
• Unify web and app flows with our Web & Mobile SDK
• Monitor full-funnel conversion: from install to subscription
• Leverage SKAN data to improve iOS trial performance
• Retarget dropped-off users with behavioral segmentation
• Integrate with RevenueCat and Adapty for LTV analysis
> → Ready to grow smarter? [Talk to us](https://www.airbridge.io)
---
## How to Win App Marketing in Japan and Korea
URL: https://www.airbridge.io/en/blog/win-app-marketing-bust-myths
Language: English
Published: 2025-05-16
Japan and Korea have long been regarded as two of the most attractive—but also most challenging—markets for global app publishers. Both countries boast high smartphone penetration, strong consumer spending on digital services, and thriving gaming and subscription app cultures. Yet many global teams struggle to gain traction in these markets—not because the opportunity isn’t there, but because they misunderstand how fundamentally different the two cultures are.
In this article, we unpack the biggest myths that hold back global apps in Japan and Korea and explore key cultural nuances that shape user behavior, conversion, and growth. The insights come from a conversation with [**Allen Park**,](https://www.linkedin.com/in/allen-park-165a91243/) Head of Marketing at [**aix**](https://www.aixinc.io/en), a consulting firm that helps global subscription and gaming apps succeed in Japan and Korea through deep localization and market-entry strategy.
**🎥 Watch the full interview video**
## **Japan and Korea: High Potential, High Risk**
According to RevenueCast’s 2025 report, average subscription spending in APAC has nearly matched that of North America and Europe. Japan and Korea, in particular, have become hotspots for subscription-based apps and mobile games seeking to expand eastward.
However, Western founders often underestimate the complexity of these markets—and overestimate how similar they are.
> **"Many founders assume a one-size-fits-all strategy will work. But Japan and Korea are culturally, behaviorally, and linguistically distinct. If you don’t address that from the start, you're setting yourself up to fail."
> — Allen Park, Head of Marketing at aix**
## **The Biggest Myths**
### **Myth 1: "Japan and Korea Are Basically the Same"**
Because Japan and Korea are geographically close (just a two-hour flight apart), many assume their markets are similar enough to use a unified strategy. This couldn’t be further from the truth.
While both are APAC powerhouses, they differ not only in language but also in how users make decisions, perceive value, and engage with apps. Japanese consumers tend to be cautious and methodical, requiring detailed information before converting. Korean users, by contrast, are fast-moving, trend-driven, and highly responsive to exclusivity and social buzz.
Did you know?
Google dominates search in Japan (over 80% share), but in Korea, Naver remains the top search engine—showing just how different user habits can be.
### **Myth 2: "Stereotypes Like Anime or K-Pop Always Work"**
Many app publishers assume that slapping on an anime-style character for Japan or using K-pop or palace imagery for Korea will improve conversions. But stereotypes can backfire—especially when they don't align naturally with the app's core service.
> **“If your service has nothing to do with anime or Korean palaces, using them just makes users wonder, _‘Why is this here?’_”
> — Allen Park**
Stereotypes can work—but only when they _make sense_. For instance, a dating sim or a manga-based game can lean into anime visuals effectively. But if you're offering a finance app or a fitness tracker, you’re better off highlighting clarity, functionality, and local relevance.
### **Myth 3: "Local Ad Channels Always Outperform Global Ones"**
It’s common to believe that platforms like Naver, Kakao (Korea) or Yahoo! Japan, Line (Japan) will automatically outperform Meta, Google, or Apple Ads.
The reality? It depends on your stage.
For early-stage market entry, global UA channels still perform very well. Local platforms require more nuanced setup, partnerships, and often yield better ROI only when you're ready to scale or experiment with reward-based models.
> **"We recommend global platforms for initial traction, and consider local DSPs and offerwalls as a second wave of growth."
> — Allen Park**
## **Cultural Clash: Consumer Psychology 101**
Hofstede’s cultural analysis helps explain the deep-rooted differences between the two markets. Both Japan and Korea are high in long-term orientation, meaning users are willing to invest effort today for a better tomorrow. But their motivations diverge in crucial ways:
- Japan: High achievement orientation. Users are meticulous, goal-focused, and seek depth of understanding before taking action.
- Korea: Lower achievement orientation but high competitiveness. The cultural mantra _“Pali Pali”_ (meaning “quickly, quickly”) captures Koreans’ urgency to act fast and stay ahead.
These traits show up everywhere—from onboarding flows to ad creatives. Where Japanese users expect comprehensive, trust-building information, Koreans are more responsive to bold, simple messages that tap into their FOMO.
> **“Japanese users want to make sure their decisions are smart and well-informed. Korean users want to make sure they’re not left behind.”
> — Allen Park**
## **Case in Point: Slopes’ Expansion into Japan**
A real-world example is Slopes, a skiing and snowboarding app that saw huge gains in Japan after localizing properly.
Originally, Slopes entered Japan with minimal localization—mainly machine-translated ASO elements and creatives. Not surprisingly, growth was flat.
After working with aix, Slopes shifted to a deep localization strategy:
- Optimized keywords based on real user search behavior
- Redesigned app screenshots to highlight culturally relevant visuals
- Partnered with trusted local entities to build credibility
The result? A measurable uptick in organic installs, improved retention, and a spot in Japan’s top three markets for the app.
## **What’s Next: Tactics That Drive Results**
So far, we’ve tackled the myths and key cultural insights for Japan and Korea. But how do you put these lessons into practice? In Part 2 of this series, we’ll break down actionable strategies for creative localization, user acquisition, onboarding, and monetization—based on what actually works in-market.
**Before you go, here are a few key takeaways from Part 1:**
- **Japan and Korea are not the same market.** Cultural nuance isn’t a detail—it’s the strategy.
- **Stereotypes only work when they fit.** Use cultural cues like anime or K-pop wisely and in context.
- **Global UA channels still work.** Don’t default to local platforms before testing what already scales.
- **Japanese users seek clarity and trust.** Korean users respond to trends, speed, and exclusivity.
- **Localization is behavioral, not just linguistic.** Understand how users think—not just what language they speak.
In Part 2, we’ll go one step further and show you how top apps are executing these insights—from ASO redesigns to rewarded ad strategies. Stay tuned.
---
## Airbridge now supports Google Integrated Conversion Measurement (ICM)
URL: https://www.airbridge.io/en/blog/airbridge-google-icm
Language: English
Published: 2025-05-14
Author: Team Airbridge
Airbridge is now fully integrated with Google’s [Integrated Conversion Measurement (ICM)](https://support.google.com/google-ads/answer/16203286?hl=en&sjid=16493375900460775132-NC), giving advertisers deeper insights into non-deterministic attribution data. With this integration, you can track probabilistic conversions with more precision and optimize your campaigns based on richer, privacy-compliant data.
Non-deterministic attribution uses observed conversions to predict unobserved conversions without identifying any one individual. This approach enables more accurate conversion measurement and campaign optimization.
## What is Integrated Conversion Measurement (ICM)?
ICM is Google’s privacy-driven solution for filling in conversion tracking gaps caused by stricter tracking limitations.
As third-party cookies and device identifiers disappear, traditional tracking methods struggle to capture a complete picture of ad performance. Google's measurement framework bridges the gap by designing an innovative app measurement solution in a way that is privacy safe.
Your total reported conversions only include non-deterministic attribution data when there is high confidence that your ad led to conversions. And in cases where Google doesn’t have enough data to be able to confidently model, Google doesn't provide non-deterministic attribution data.
This integration is a game-changer because Airbridge now gives marketers direct access to Google’s proprietary non-deterministic attribution data—previously only available inside the Google Ads UI, and outside the cross-network attribution data flow with Airbridge. With this, marketers can eliminate blind spots in performance measurement and gain a clearer view of ad effectiveness.
## Airbridge & ICM
## How does this integration benefit you?
### 1. Enhanced Conversion Tracking
With Airbridge, you can access Google’s machine learning-based non-deterministic attribution data, enabling more precise and in-depth performance analysis.This data is fully available in the Airbridge dashboard, providing deeper insights than what’s accessible in Google Ads. With more detailed non-deterministic attribution data, Airbridge helps marketers accurately measure marketing impact and optimize ad performance more effectively.
### 2\. Streamlined Data Analysis
Stop switching between platforms to piece together ad performance data. With Google’s non-deterministic attribution insights now available in Airbridge, marketers can analyze everything in one place. This integration simplifies campaign optimization, reduces manual data work, and improves decision-making efficiency.
### 3\. Privacy-First Performance Measurement
Unlike traditional cookie-based tracking, non-deterministic attribution estimates conversions without identifying users individually, ensuring compliance with privacy regulations while maintaining measurement accuracy.
Google enforces strict privacy policies, banning fingerprinting, IP tracking, and any other invasive identification methods. It only includes high-confidence estimates in reports, giving marketers reliable data for campaign optimization without compromising user privacy.
## How do you set up Google ICM on Airbridge?
For step-by-step instructions, check out the Airbridge ICM integration guide [here](https://help.airbridge.io/en/developers/ios-sdk-v4#integrate-googles-icm-attribution-data). Feel free to contact us for any support. Have questions about ICM, performance tracking, or how Airbridge can help your app business? Reach out to our team—we’re happy to help!
---
## The Ultimate Guide to Mobile Game Re-engagement and Retargeting
URL: https://www.airbridge.io/en/blog/mobile-game-reengagement-and-retargeting
Language: English
Published: 2025-05-07
[Taoufik Lazrak](https://www.linkedin.com/in/taoufik-l-75640646/) is the Chief Revenue Officer (CRO) of [Adikteev](https://www.adikteev.com/), a leading platform in game retargeting. Adikteev has worked with over 400 gaming titles, including King, Blizzard, Rovio and Playtika, pioneering the concept of incremental value. With over a decade of experience at Adikteev, Taoufik has guided the company through the evolution of game marketing, from its early days to today's sophisticated retargeting strategies.
[Roi Nam](https://www.linkedin.com/in/roinam/), co-founder and CEO of [Airbridge](http://airbridge.io), recently spoke with Taoufik about effective re-engagement and retargeting strategies for gaming apps. Dive into their conversation to learn tried-and-true tactics, bust common myths, and navigate the new landscape of privacy-first re-engagement.
**🎥 Watch the full interview video**
## **Biggest myths on game re-engagement and retargeting**
Let's dive straight into some common misconceptions about re-engagement and retargeting campaigns for gaming.
### **What are the top 3 misconceptions you've had to debunk?**
**Taoufik:** That’s a great question. I’d be happy to debunk some of the most common misconceptions in the re-engagement and retargeting space.
#### **Myth 1: Re-engagement campaigns do not work for casual games.**
You’ve probably heard that re-engagement is a waste of time and money for games with short user lifecycles. But that’s simply not true. With the right targeting and creative strategy, casual and hybrid-casual games can see significant results from re-engagement campaigns.
We’ve seen great success when campaigns are aligned with in-game events, new content drops, or key monetization windows. The key to re-engagement is personalization—casual users are far more likely to respond when the message feels relevant and timely.
#### **Myth 2: The view-through metric doesn’t matter.**
Some marketers dismiss view-throughs as irrelevant, but in today’s complex advertising landscape, every interaction matters. Adopting a mixed attribution model is essential, especially in gaming, where repeated exposure often drives results. Even if a well-crafted ad doesn’t get a click, it can still influence a user’s decision to return or make a purchase. That impact needs to be properly measured and valued.
[__wf_reserved_inherit]
#### **Myth 3: Incrementality is not reliable.**
There’s a misconception that re-engagement doesn’t deliver real, incremental returns. However, as a pioneer in incrementality testing for re-engagement, Adikteev demonstrates daily that incrementality can be accurately measured—provided the testing framework is clean and unbiased.
> “The key to re-engagement is personalization—casual users are far more likely to respond when the message feels relevant and timely.”
## **Driving incremental revenue for casual hybrid casual games**
Incrementality testing involves creating a control group and a test group, which can be challenging. As Taoufik notes, the challenge is indeed ensuring these tests remain bias-free to accurately measure the incremental revenue generated by re-engagement and retargeting campaigns.
### **How do you measure incrementality for a re-engagement campaign for high-value players? How do you guarantee it’s bias-free?**
**Taoufik:** At Adikteev, we use the ITT (Intention-to-Treat) methodology, originally developed in the medical field, to ensure clean and unbiased results. This means measuring the impact of retargeting much like doctors assess the effectiveness of a treatment in clinical trials. We isolate a control group and compare it to a test group exposed to re-engagement ads, then evaluate whether there's a significant lift in revenue and user activity.
> “With clean and unbiased incrementality testing frameworks, even casual and hybrid-casual games can use re-engagement campaigns to drive incremental revenue.”
### **Are re-engagement campaigns truly incremental, or do they just cannibalize existing revenue streams?**
**Taoufik:** Many gaming marketers worry that re-engagement campaigns simply result in paying more for high-spending or active paying users. However, what we consistently observe is that high-value players, including paying users, tend to spend more when exposed to re-engagement campaigns. This isn't about cannibalization, where existing revenue is simply reallocated, but rather about unlocking additional value that wouldn't have been captured otherwise.
### **How do you explain the real incremental value of re-engagement and retargeting campaigns to your clients?**
**Taoufik:** We focus on straightforward, real-world outcomes. For active users, re-engagement prevents them from spending less or disengaging sooner. For lapsed players, it often prompts their return. We back this up with clear metrics: incremental revenue, ARPU uplift, and behavioral differences between exposed and control groups. It's not about surface-level clicks or impressions; it's about delivering net new business impact.
[__wf_reserved_inherit]
Plus, we're fully transparent, allowing clients to replicate our tests and compare results. Our incrementality measurement data is based on event and attribution data from MMPs (Mobile Measurement Partners) like [**Airbridge**](http://airbridge.io). Since clients have access to this data, they can easily replicate our tests and compare results, ensuring everyone is aligned.
[__wf_reserved_inherit]
### **Suppose I’m a game marketer, and I want to test incrementality with you. Could you walk me through the process?**
**Taoufik:** We start with data-driven insights from pre-launch analysis. Once we partner up, we begin receiving event, attribution, and behavioral data from MMPs like **Airbridge**. This data helps us figure out if a game is a good fit for re-engagement campaigns. We typically spend about 1 ~ 2 weeks analyzing user behavior to get a sense of their patterns. Then, we recommend specific audience segments and calculate the optimal budget for each—usually around $500 to $1,000 daily at a global level. The idea is to scale these efforts effectively.
For incrementality testing, we start measuring from day one of the re-engagement campaign launch. This allows clients to see the true value of their campaigns alongside live performance, ensuring transparency and that every dollar spent drives real results.
> “Re-engagement isn't about cannibalization, where existing revenue is simply reallocated, but rather about unlocking additional value that wouldn't have been captured otherwise.”
## **Proven best practices for re-engagement and retargeting campaigns**
Since 2012 Adikteev has advanced way beyond basic retargeting to deliver real, measurable ROI (Return on Investment). Trained on billions of gaming data points, their algorithms drive their dominance in the industry. If anyone knows what works for game re-engagement and retargeting, it's Taoufik Lazrak.
### **What are the best practices that consistently deliver results for gaming clients?**
**Taoufik:** The top 3 proven best practices have been audience segmentation, creative A/B testing and geo-targeting.
First off, it's all about embracing smart, data-driven **audience segmentation**. It's not just about targeting active users; you need to re-engage lapsed users and even convert non-payers. The more granular the segmentation, the better the performance will be.
Next, understanding what resonates with your audience is everything. That's why **creative A/B testing** is so important. You need to constantly experiment with different ad formats, like interstitials, rewarded videos, and banners, along with placements and messaging strategies specifically adapted for retargeting. It’s an ongoing process designed to help you identify which creative elements truly drive engagement and conversions within each target segment.
Lastly, when it comes to optimizing your campaigns, having a global reach with localized insights is key. You start by casting a wide net with broad **geo-targeting**, and then quickly zero in on the regions that show the highest LTV potential.
### **Is broad geo-targeting still a good bet for game marketers, even with the risk of burning through budget in the wrong regions?**
**Taoufik:** Tools like Adikteev use personalized algorithms to predict which users are most likely to convert, and they automatically stop targeting those who aren't. This way, you avoid pouring money into underperforming regions and ensure your ads are seen by the players who matter most.
> “You start by casting a wide net with broad geo-targeting, and then quickly zero in on the regions that show the highest LTV potential.”
## **Hybrid monetization strategies for casual and hyper-casual games**
Game monetization used to be divided into two main approaches: IAP (In-app purchases) and IAA (in-app advertising). Each strategy targets different players and revenue streams. Today, many casual and hyper-casual games are adopting hybrid monetization models, combining IAP and IAA.
[__wf_reserved_inherit]
### **Is transitioning from a single revenue stream to a hybrid model the right move? What metrics should game marketers focus on?**
**Taoufik:** Absolutely consider hybrid monetization, especially if you're in the casual or hyper-casual space. Try blended tROAS (target ROAS) over traditional ROAS as your key metric. Unlike traditional ROAS, which looks at past performance, tROAS is proactive allowing you to optimize as you go. By setting a target ROAS, you also ensure your ad campaigns align with your business objectives—whether that's maximizing revenue or achieving a specific profit margin.
> “Try blended tROAS over traditional ROAS as your key metric. Unlike traditional ROAS, which looks at past performance, tROAS is proactive allowing you to optimize as you go.”
## **Re-engagement and retargeting strategies in a privacy-first world**
Ever since Apple introduced the App Tracking Transparency (ATT) framework with iOS 14.5 in 2021, privacy regulations have become increasingly stringent. With iOS users now having greater control over their data, game marketers must adapt their re-engagement strategies to ensure their campaigns remain effective.
### How can game marketers run better re-engagement campaigns on iOS in the era of enhanced privacy?
**Taoufik:** Before ATT, retargeting users with IDFA was the standard approach. Now, many users will likely choose not to opt-in for your gaming apps, even if they opt-in elsewhere. This is where **probabilistic attribution** comes into play. By using probabilistic methods, we can unify user identities across different platforms. For instance, if a user agrees to share data in one app, we can infer with a high degree of certainty—say, 90%—that they are the same person we're targeting in another context.
Over the past year, we've significantly improved our ability to target iOS users, increasing our reach from 20-30% on some partners. In the U.S., gaming partners have seen natural opt-in rates of 40-45%. With Adikteev's campaigns, this means they can reach 60-70% of their target users on iOS for re-engagement, which can positively impact revenue.
> “With probabilistic attribution, if a user agrees to share data in one app, we can infer with a high degree of certainty—say, 90%—that they are the same person we're targeting in another context.”
## **6 Key takeaways to remember**
Here’s a quick recap of six key takeaways to help you create successful re-engagement and retargeting campaigns for your games.
- **Mythbusting:** Incrementality testing, view-through metrics, and re-engagement campaigns can be highly effective, even for games with shorter lifecycles like casual and hyper-casual titles.
- **Incrementality:** Clean and unbiased testing frameworks enable casual and hyper-casual games to use re-engagement campaigns to drive incremental revenue.
- **Proven best practices:** Use audience segmentation, creative A/B testing, and geo-targeting for effective re-engagement.
- **Hybrid monetization:** For casual and hyper-casual games, try hybrid monetization and focus on blended tROAS as your key metric.
- **Privacy-first retargeting:** Probabilistic attribution can help you reach 20-30% more users on iOS.
### **Supercharge your game growth with Airbridge**
If you're ready to take your game marketing to the next level, check out [**Airbridge**](https://www.airbridge.io/). Our platform provides:
- **Advanced Cohort Analysis for LTV:** Track lifetime value with precision and break down performance at a granular level (keyword, creative, country, OS, and more).
- **Ad Mediation Integration:** Seamlessly combine data from multiple ad networks in one place (e.g., Google AdMob, AppLovin MAX, Unity LevelPlay).
- **Predictive LTV (pLTV) Feature:** Forecast future revenue in real-time, allowing for faster, data-driven decisions.
- **Minutely/Hourly Retention Analysis:** Understand player behavior down to the minute or hour for truly actionable insights.
- **Competitive, All-Inclusive Pricing:** Access enterprise-grade analytics without hidden fees or complicated add-ons.
Whether you're an indie studio or an established publisher, Airbridge gives you the critical tools you need to stay ahead in today's fast-paced market. Visit [**airbridge.io**](https://airbridge.io) to learn more.
### **Win gaming re-engagement with Adikteev**
Stay ahead of the game with Adikteev's expert insights and strategies for winning re-engagement and retargeting. Dive into resources for actionable tips and real-world success stories at [adikteev.com](http://adikteev.com) or learn more by [getting in touch here](https://content.adikteev.com/contact-us).
---
## Introducing Airflux: AI-Powered Ad Optimizer for Mobile Games
URL: https://www.airbridge.io/en/blog/introducing-airflux-mobile-game-ad-solution
Language: English
Published: 2025-04-09
_What if a casual game could pull in half a billion dollars?_
That’s exactly what _Survivor.io_ did, crossing $500 million in lifetime revenue in 2024. For a tap-to-play title, that’s a staggering number—and a clear sign of how massive the casual gaming market has become.
We’re in a high-growth window for mobile casual games, where a single well-made title can deliver outsized returns.
But the flip side? Competition is tougher than ever. **User acquisition (UA) costs are rising, traffic is plateauing, and teams are under growing pressure to drive more revenue from the players they already have.** That’s no easy task.
That’s exactly why the team behind Airbridge built **Airflux**—a platform designed to help casual game studios maximize ad revenue faster, smarter, and more efficiently. It’s also Airbridge’s first standalone product since the company was founded.
To understand how Airflux came to life—and why it matters more than ever—we sat down with Noel Son, Data Scientist, and Roi Nam, CEO and Co-founder of Airbridge.
## 1\. Can you briefly introduce yourselves?
**Roi:** I’m Roi, CEO and Co-founder of Airbridge. I started the company about ten years ago with Airbridge, our MMP solution, which now supports over 500 companies across 10 countries—including Korea, the U.S., China, Vietnam, and Turkey. Most recently, we launched Airflux, and I’m excited to share the story behind it.
**Noel:** I’m Noel. I’ve been leading the Data Science team at Airbridge for the past three years. We pitched the idea for Airflux internally last year, worked hard to bring it to market, and now I’m leading the project as the product owner. It’s been exciting to see it come to life.
[Noel Son (left) and Roi Nam (right)]
## 2\. What is Airflux?
**Noel:** Before I get into Airflux, we need to take a step back and look at how mobile games monetize.
Most mobile games—especially casual ones—make money through two core models: **in-app ads (IAA) and in-app purchases (IAP). The revenue mix depends heavily on the game genre.**
Casual games lean on IAA, while hardcore titles rely more on IAP. Some games stick to one model entirely. For example, a hyper-casual title might monetize 100% through ads, whereas a MMORPG game could be fully IAP-driven.
Midcore games like idle titles often use both, showing a few rewarded videos (RV) to boost engagement, while IAP remains the main revenue driver. But in casual games, purchases are minimal. The gameplay is light, and there aren’t many purchase-worthy moments—so ads become critical. If users aren’t buying or watching ads, the game doesn’t make money.
Here’s where it gets tricky. Most casual games rely on rewarded videos and interstitials—the latter often referred to as “forced ads” because they interrupt gameplay and aren’t user-initiated. Show too many and players churn. Show too few and you leave revenue on the table.
To manage that balance, monetization managers often hardcode the timing and frequency of interstitial ad displays—like showing one after every one or two stage clears, or triggering them 50% of the time after each stage clear.
But here’s the catch: increasing effective CPM (eCPM) by a few cents—say, bumping your bid from 3¢ to 4¢—adds a small 1¢ margin. **Getting a user to watch just two more ads without churning? That has a far bigger impact on revenue.**
**However, tolerance for interstitials varies by region, device, play style, and player progression; essentially, it eventually varies by each individual user.** Despite that, most games still use coarse rules and broad segments for interstitial ad display frequency and timing, leaving a lot of value untapped.
This is where true optimization needs to happen—dynamically adjusting ad frequency and timing so players watch more ads without getting annoyed. But getting to that level of personalization, with automated split testing and smart user segmentation? Very difficult.
That’s where Airflux comes in.
**Airflux automatically personalizes interstitial delivery—adjusting frequency and timing by user segment. It groups players by country, device, behavior, in-game engagement and progress, then dynamically optimizes when and how often ads are shown.** For example, high-IAP potential users may see fewer interstitials, while non-spenders who tolerate ads may see more.
**The result? More ad views for potential non-payers, less ad views for potential payers, better retention, and higher lifetime value (LTV).**
[]
**Roi:** Noel and our data science team have statistically proven that when users see an interstitial—and how often—has a direct impact on LTV.
We validated this causal relationship across multiple models. **Get the timing and frequency right, and you get higher LTV.**
This is especially critical for hyper-casual and hybrid-casual studios, where even small LTV gains can significantly affect profitability. As mobile game development gets easier and competition intensifies, retention will define long-term revenue.
Airflux automates what used to be a manual process in monetization and LiveOps (live operations, including post-launch updates and ongoing tuning)—using personalization and automated A/B testing to replace guesswork with data-driven precision.
**That’s why we say this with confidence: Airflux’s biggest value to game studios is driving revenue and profitability.**
## 3\. There are plenty of industries—why focus on mobile games?
**Noel:** Right now, casual games need this kind of SaaS platform more than any other sector.
Many casual studios are lean. Some generate $20–30 million in annual revenue with fewer than 50 employees. That’s impressive efficiency, but it also means non-dev teams—monetization, LiveOps, data analysis—haven’t scaled much over the past decade.
In our meetings with studios, we often see data analysts maintaining BI (business intelligence) dashboards, but we rarely see dedicated data scientists focused full-time on modeling.
E-commerce, by contrast, is packed with data scientists, data-driven growth marketers and CRM marketers focused on LTV. They use tools like Amplitude and Braze to scale and automate personalization.
In gaming—especially ad monetization—the tooling still lags behind. Most of the infrastructure is built for UA or dev workflows. And honestly, plenty of teams are still running everything out of Excel.
**Roi**: Like Noel said, most studios don’t have the in-house resources to deeply analyze their data. So even when there’s clear potential to grow revenue, it often goes untapped.
Mobile gaming is massive. According to [Statista](https://www.statista.com/outlook/amo/media/games/in-game-advertising/worldwide?utm_source=chatgpt.com), in-game ad revenue is projected to hit $124 billion by 2025. Yet due to the lack of ad optimization, studios are missing out on a big chunk of that.
**What we’re building is a way for studios to use AI and machine learning to boost ad revenue from the traffic they already have. That’s the value we want to deliver—helping teams earn more from the users they already have.**
**Noel:** This is exactly where a SaaS platform becomes the solution.
Building and testing optimization models requires a full data science team—experts in math, stats, and machine learning. But most studios don’t have that luxury.
So we asked ourselves: what if a SaaS platform could do that work instead? If we handle the complexity, studios can boost revenue without needing to scale their teams.
## 4\. Are mobile game studios seeing this as a real problem, too?
**Noel:** Absolutely. We’re currently running Airflux tests with over ten studios, and the early results are promising.
Even before launch, studios told us they wanted to try it as soon as it was ready. All free trial slots filled up quickly, and we’ve already moved on to paid POCs (Proofs of Concept). Despite that, the response has been extremely positive—which has been incredibly motivating for our team.
There aren’t many tools like this out there, so we’re doubling down on building something that can truly help studios grow revenue and improve profitability.
[]
**Roi:** We’ve seen especially strong interest from hyper-casual studios, where tight margins mean they need to drive high volume to stay profitable. These teams are always looking for ways to squeeze more value from their ad inventory.
We’re also seeing more traction from hybrid-casual studios, especially as we expand optimization into balancing payouts for the rewarded videos.
One of the big drivers behind this is that eCPM growth in 2024 has been underwhelming. Most teams have already tried the usual levers—more placements, higher bid floors, longer videos, added competition—but those tactics just don’t move the needle like they used to.
**Ad monetization is getting harder. So more teams are turning to smarter, automated solutions—like Airflux.**
No studio has unlimited resources. Every monetization touchpoint has to count. In the short term, we’re helping optimize interstitials and rewarded videos. Long term, we’re expanding into IAP optimization, all powered by our proprietary machine learning and automation engine.
Because once LTV goes up, teams can reinvest more into UA, bring in more players, and keep the growth loop going non-stop. It’s a classic case of LTV driving CAC (customer acquisition cost).
## 5\. How is Airflux different from running A/B tests manually with tools like Firebase?
**Roi:** With Firebase, studios need to define their own hypotheses, manually set up tests, and configure variables—like whether to show an interstitial after Stage 1 or Stage 2, and at what probability (say, 50 percent vs. 70 percent).
These tests are one-dimensional but still require a lot of internal effort. And in mobile games, where you’re dealing with hundreds of variables, manual A/B testing just doesn’t scale.
You also have to manage test duration, sample size, and post-test analysis. Doing this across 100+ countries and multiple platforms with varying player behaviors? That’s hundreds or even thousands of variants. It’s almost impossible to keep it under control.
**That’s exactly what Airflux solves. It automates testing across granular segments—no dev work, no custom code.** **It’s plug-and-play and built to scale in ways traditional tools can’t.**
Firebase is great for testing major changes, like game design updates or new IAP offers, such as a welcome pack. But when it comes to optimizing ad delivery at scale, Airflux is purpose-built for the job.
**Noel:** Running statistically meaningful A/B tests is more complex and time-consuming than most people think. There are just too many variables to manage manually, especially in mobile games.
Airflux uses machine learning to automate the entire process—from generating hypotheses to running tests, analyzing results, and applying the insights. The AI segments users, runs thousands of tests in parallel, and continuously optimizes for revenue.
For studios, this means no need to hire a data science team or burn hours on manual testing. It saves time and money.
**More importantly, Airflux helps unlock incremental revenue that static setups simply can’t.**
**Roi:** As a quick recap, here’s the core value Airflux brings to game studios:
- Automatically adjusts interstitial ad frequency and timing based on user attributes and behavior
- Increases LTV from your existing traffic, with results proven through statistically reliable incrementality measurement
- Automates LiveOps tasks that typically require hours of manual work
- Replaces manual A/B testing on monetization with automated, scalable experimentation
- Optimizes ad exposure to each user’s tolerance threshold—maximizing revenue without hurting UX or retention
- Delivers clear, measurable, data-backed results
## 6\. Have you seen real-world success with Airflux?
**Noel:** Absolutely. We’re currently working with over ten game studios, and the early results have been really strong.
For one zombie-themed FPS (first-person shooter) with over 10 million global downloads, we saw a 50% increase in user LTV. In another case, a studio with more than 10 million MAU (monthly active users) saw a 14% uplift in just two weeks.
One of our larger partners is currently seeing a 16% uplift, which is incredibly promising.
[]
## 7\. What strategies do mobile game studios need to stay competitive in the market?
**Roi:** In e-commerce, the early focus was all about driving new user acquisition and direct-response sales. But over time, the strategy shifted toward increasing LTV by leveraging CRM, data analytics, hyper personalization, and loyalty programs to maximize value from existing users. We’re now seeing the same shift in mobile gaming. The days of cheap user acquisition and fast scaling are behind us. What matters now is how much value you can generate from each player.
That’s why optimizing your ad monetization system with machine learning will be critical—maximizing revenue and lifting LTV on a per-user level.
And in my opinion, Airflux is the solution best equipped to do exactly that.
**Noel:** I completely agree with Roi. As the barrier to entry drops, mobile gaming is only getting more competitive.
That’s why the focus needs to shift to growing LTV without sacrificing the player experience. At its core, it’s simple. **You’ve already spent big to acquire these users—now treat them like the high-value assets they are. Protect their experience, but don’t miss a single chance to monetize over the long run.**
---
## How Playio Switched to Airbridge and Now Manages Global Marketing From a Single Platform
URL: https://www.airbridge.io/en/blog/playio-mmp-migration-global-marketing
Language: English
Published: 2025-04-01
## Best Value for Gamers — The Reward App for Gamers, Playio
### Q. Could you introduce yourself and the company?
Hello, I am Hayoung Park, Marketing Team Lead at Playio. I have been passionate about gaming since the start of my career—my very first job was at a game company. Currently, at GNA Company, I am working with Playio toward our mission of "Best Value for Gamers" to build a healthy gaming ecosystem.
### Q. Could you tell us about Playio?
Playio is fundamentally a play-time-based reward app. Users earn points based on the time they spend playing games, and they can exchange those points for products. However, Playio aims to be more than a simple reward app—it is an environment where gamers are recognized for their unique value.
For example, rather than just offering rewards, we rank users into levels and tiers based on their actual play time. Playio also serves as a community where diverse gamers come together. For users, it is a space to enjoy games and connect; for game publishers, it is a marketing channel recognized for having highly engaged "core gamers" who drive installs and retention.
[Visit Playio](https://playio.co/?LANG_CD=ENG)
[]
### Q. We understand Playio operates in multiple countries.
That is right. Playio is currently live in Korea, the United States, Japan, and Taiwan. We are also planning to expand into Hong Kong, Macau, and more.
### Q. How is the marketing team structured and what does it do?
Our marketing team is divided into B2B and B2C parts. The B2B team, newly formed in 2025, works closely with our sales team to demonstrate how effective Playio is as a marketing channel. They publish articles, send newsletters, generate leads to acquire new clients, and strengthen relationships with existing partners.
The B2C team focuses on helping users discover and enjoy Playio. Beyond acquiring new users, they run various marketing actions to keep existing users satisfied and engaged.
## Playio's Journey to Airbridge
### Q. We heard Playio is on its third MMP, and chose Airbridge. What was the background?
Last year, we entered the Japanese market and started working with a local publisher who also took on marketing. That meant we needed to restrict the publisher's dashboard access to Japan-only data—they should not see performance from other countries. Our previous MMP said this kind of country-level data segmentation was not possible.
So we started looking for an MMP that could limit data visibility by country. Most MMPs could not offer it. Airbridge was the only one that confirmed they could do it, which is what led us to migrate.
But that was not the only reason. Personally, I had always felt Airbridge had a more intuitive UI/UX compared to other MMPs. Marketing performance is something that should be accessible to everyone in the organization—not just data analysts or performance marketers—and Airbridge's intuitive interface made that possible.
### Q. How long did the migration take, and how was the process?
About one month. As I mentioned, we had a special requirement for country-level data permission separation, and we had many additional requests during the integration process. We held frequent migration meetings, but thanks to the proactive support from the Airbridge team, everything went smoothly.
MMP migrations are not something you do often, and they tend to be difficult and time-consuming. Airbridge provided a systematic process with a thorough checklist, which allowed us to complete it quickly. They also conducted company-wide onboarding training, enabling more colleagues to use Airbridge.
[]
## How Playio Uses Airbridge
### Q. Playio operates across multiple countries. Country-specific strategies and performance measurement must be very important.
Exactly. We run different strategies and media channels by country. Korea launched in 2021, Japan in April last year, and Taiwan in November. In Korea, D30 retention exceeded 30% last year, so user loyalty is very high. We focus not just on acquiring new users but also on content, character-based, and CRM strategies to effectively communicate the brand vision.
Taiwan and Japan are in a rapid growth phase, with a stronger emphasis on new user acquisition. We actively run major channels like Google and Meta along with DSPs, optimizing country-specific strategies.
### Q. What Airbridge features does Playio use most for mobile marketing optimization?
We use Airbridge metrics and data extensively for measuring ad performance by media channel and creative. One feature Playio particularly values is the variety of reports. Other MMPs do not offer this level of report diversity. Airbridge lets you customize reports by any metric you want—that is a huge plus.
[Explore Airbridge Reports](https://help.airbridge.io/en/guides/airbridge-report-overview)
Our marketing team goals are aligned with company-wide KPIs—things like new user growth and retention improvement. We use Airbridge's various report features to create and share charts so the entire company can track these goals together. Thanks to the onboarding training, colleagues outside marketing now build their own reports directly.
Another key feature is tracking links through custom channels. Beyond mobile ads, Playio runs a lot of influencer marketing, content marketing, and external partnership promotions. For these, we use custom channels to measure and optimize performance by campaign and by influencer.
[Explore Airbridge Custom Channels](https://help.airbridge.io/en/guides/custom-channels)
### Q. Can you share examples of marketing results achieved with Airbridge?
In Korea, we have a case where we improved performance by combining Airbridge funnel reports with internal data analysis. Retention improvement was a key goal in Korea. We used internal data to identify user "aha moments," created various action items to guide users to those moments, ran A/B tests, and used Airbridge funnel reports to check and optimize the results. Ultimately, by optimizing the path to the aha moment, domestic D30 retention exceeded 30%.
We also have a country-specific marketing optimization case. In Taiwan, Airbridge Actuals reports revealed that video creatives performed significantly better than image creatives in UA campaigns. Based on this data, we increased the video creative ratio, resulting in over 100% increase in user acquisition and 40% savings in user acquisition costs.
> Want to learn more about Playio's mobile marketing results with Airbridge?
[Read the full case study](https://www.airbridge.io/en/customers/playio-d30-retention-cpa-optimization)
### Q. As a marketer, what are the biggest advantages of Airbridge?
One of the biggest advantages I see is that, even as an MMP, Airbridge's data visualization is exceptionally well-designed. There are many customizable reports, flexible data combinations, and the design is highly intuitive. Having used multiple MMPs, Airbridge is the only one that offers this level of diversity and intuitiveness in its dashboard.
[]
Another major advantage is that non-marketers can also use it with ease. For people outside of marketing or data roles, looking at data can feel daunting. But Airbridge's intuitive UX/UI and easy usability make it comfortable even for colleagues with no marketing background.
## Playio's Vision: The Future of Gaming Entertainment (Reward 2.0)
### Q. Playio is more than a simple reward channel—it recognizes and respects the value of gamers. What is Playio's vision for the future of game reward platforms?
Gamers play for a variety of motivations—from simple fun to achievement, social connection, and recognition. Playio plans to offer even more features, content, and rewards to amplify these diverse motivations.
We believe Playio's vision goes beyond being just a reward platform. We call this "Reward 2.0." Under this vision, we aim to continue recognizing users' value as gamers while becoming a platform that offers game publishers much more targeted advertising.
### Q. What are the marketing team's goals as Playio walks the path of Reward 2.0?
As I mentioned, Playio strives to create a healthy gaming ecosystem beyond simple rewards. The marketing team moves together toward this goal. Every team member is a gamer themselves, and we work to create more experiences that deliver value to gamers in line with the product's direction. That is why Playio does marketing.
[]
### Q. Any final thoughts you'd like to share?
Having been through multiple MMPs, I am happy to have settled on Airbridge. The complex migration was completed quickly, and now the entire company is achieving growth through Airbridge data. With its intuitive dashboard, responsive CSM support, and many other advantages, I cannot imagine going back to another MMP.
Everyone at Playio is working hard to build a better service. New features are constantly being added and the service is evolving rapidly, so please keep watching Playio's journey. We are also continuously hiring talented people to join us on this exciting path.
[View GNA Company Career Openings](https://gna.career.greetinghr.com/team)
Want to optimize your global mobile marketing performance with Airbridge?
---
## 6 Reasons Why Deep Links Drive Organic App Growth in 2025
URL: https://www.airbridge.io/en/blog/organic-app-growth-with-deep-linking
Language: English
Published: 2025-03-28
## **A brief history of deep links**
[Deep linking](https://www.airbridge.io/en/blog/what-is-deep-link) took off in 2013-2014 as the mobile e-commerce boom reshaped user behavior. Apps outperformed mobile websites in driving purchases, but marketers struggled to guide users into their apps from traditional channels like email and web ads.
Deep links solved this problem by seamlessly guiding users from marketing campaigns directly to specific app destinations. Marketers quickly embraced this technology, which evolved from **standard deep links** (for users who already had the app installed) to **deferred deep links** (which prompted app installation before redirecting users to the intended in-app location).
[6 Reasons Why Deep Links Drive Organic App Growth in 2025]
Fast forward a decade to **2025's deep link renaissance**, deep linking has reclaimed the spotlight as an essential tool for marketers, driving strategies around organic growth, from SEO to web-to-app conversions and affiliate marketing.
So why is this decade-old technology making a comeback in 2025? In this post, we'll explore 6 reasons why deep linking is more crucial than ever for organic app growth today.
## **1\. The shift to organic growth**
Mobile-first apps are now prioritizing profitability and "sustainable growth" as slower growth rates and reduced venture capital funding reshape the landscape. To adapt, brands are actively pivoting toward organic growth strategies, which are typically more cost-effective (though they still require significant human resources). Marketers are driving these efforts by integrating deep linking into a variety of organic tactics, making it a cornerstone of their strategies.
- **SEO:** Deep links enhance search rankings by creating a robust internal link structure, directing users to specific app content, and boosting overall discoverability in search results.
- **ASO:** Deep links integrated with Custom Product Pages (CPPs) and Custom Store Listings (CSLs) create targeted user experiences, improving conversion rates by aligning app content with user interests and enhancing app store visibility.
- **Referral Marketing:** Deep links facilitate easy sharing of app content between users, enhancing referral effectiveness by enabling personalized onboarding experiences for referred users.
- **Content Sharing:** Deep linking simplifies user-to-user sharing of app content, boosting viral potential by ensuring shared content opens directly in-app and improving the overall user experience.
- **Web-to-App Conversions:** Deep links transition users from mobile web to equivalent in-app content, improving conversion rates by directing web users to more engaging app experiences.
> "With programmatic content generation marketers can easily create engaging lists like 'top 10 X' or 'best 5 alternatives for Y,' boosting SEO and traffic. Pair these pages with web-to-app banners featuring clear CTAs and seamless deep linking to ensure a smooth transition to in-app content." — [Roi Nam, CEO and Founder](https://www.linkedin.com/in/roinam) of [Airbridge](https://www.airbridge.io/)
## **2\. The revival of CRM & customer engagement**
Budget cuts are pushing marketers to focus on customer engagement, with CRM strategies taking center stage. Once sidelined by aggressive user acquisition, CRM is now critical as companies aim to maximize Average Revenue Per User (ARPU) from existing users. To overcome challenges like connecting messaging platforms with product ecosystems and measuring impact, marketers are embedding deep links into emails, SMS, and messaging apps like WhatsApp, LINE, Kakao, Zalo, Discord, and Telegram. These deep links guide users directly to relevant in-app content, creating a seamless experience that reduces friction and boosts engagement. By integrating deep linking into CRM efforts, marketers are driving retention
## **3\. The marketing budget pivot From UA to retargeting**
More apps are shifting their marketing budgets from user acquisition (UA) to retargeting campaigns. This shift echoes the growing emphasis on CRM strategies and maximizing ARPU from existing customers.
By prioritizing their current user base, brands minimize risks associated with new, untested audiences while optimizing spend. These retargeting efforts build on established user relationships to drive higher engagement and revenue.
Deep links are crucial for retargeting campaigns, as they provide the precision needed to direct users to specific in-app content or actions. Without them, targeted strategies fall short of their potential.
## **4\. The rise of micro-influencer marketing**
Micro-influencers are transforming mobile app marketing strategies. Platforms like TikTok, Youtube and user-generated content (UGC) ads have fueled this trend. Despite often being restricted to a single link in their profile, which can complicate precise attribution, micro-influencers continue to drive app installs and boost virality effectively.
> "Micro-influencer and UGC have proven exceptionally effective for mobile app user acquisition, particularly in the North American market. But without the proper tracking and deep links, these campaigns are bound to fail. Setting up the proper tracking is a must for success." — [Hamza, CEO and Founder](https://www.linkedin.com/in/hamza-alsamraee/) of [NewForm](https://newform.ai)
[6 Reasons Why Deep Links Drive Organic App Growth in 2025]
**💡** [**See how Nexon used cross-platform measurement to attribute UGC and video ad campaigns on TikTok**](https://www.airbridge.io/en/case-studies/nexon-cross-platform-tracking)
## **5\. The surge of web-to-app strategies**
Mobile app marketers are turning to web-to-app strategies, especially in iOS, where performance visibility is limited by fewer Advertising Identifiers (ADIDs) and complex SKAN measurement. Web campaigns offer more precise, more direct insights compared to in-app counterparts. This clarity makes web-to-app strategies attractive for several app categories.
- **Content apps (e.g., webcomics, short dramas)**: Content apps showcase content previews on the web before driving users to the full app experience.
- **Subscription apps**: Web interfaces can simplify the sign-up process for subscription apps before transitioning users to the app for ongoing engagement.
- **Gaming apps featuring HTML5 mini-games or playable demos:** Gaming apps allow users to try the game on the web and then seamlessly continue in the entire app version.
Marketers are using web-to-app strategies to overcome attribution challenges and gain accurate performance data. Deep linking plays a crucial role in connecting these web-to-app conversions, seamlessly guiding users from web content to specific in-app locations. Apps that offer meaningful web previews to create a compelling first impression, showcase app functionality, and guide users smoothly from discovery to installation can leverage this approach to boost conversion.
## **6\. The spike for re-engagement in casual games**
Re-engagement (or reactivation) strategies are becoming increasingly important in the casual and mid-core gaming sectors in 2025. While e-commerce apps have long leveraged re-engagement, gaming apps are now adopting these tactics to reduce churn and extend user LTV.
Re-engagement strategies have been quite common in hardcore games like MMORPGs and competitive MOBAs, which use them to bring players back for new seasons, maps, or character releases. The notable shift is that casual games, traditionally known for shorter user lifecycles and simpler conversion funnels, are now adopting these tactics. Two key factors are driving this change:
- **Hybrid Monetization**: The industry is combining in-app advertising (IAA) and in-app purchases (IAP) to maximize ARPU.
- **Gameplay Design**: Casual games are evolving from hyper-casual to hybrid formats, incorporating more advanced core loops and meta-game elements to improve engagement and retention.
**💡** [**Learn how to blow up your hybrid and hyper casual games with Two and a Half Gamers**](https://www.airbridge.io/en/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives)
Companies like Adikteev, Remerge, Appier, and Moloco are leading this shift by providing advanced retargeting solutions. These platforms enable gaming marketers to roll out effective re-engagement strategies powered by deep links.
- **Event promotions**: Deep links take players directly to specific in-game events or time-limited challenges, increasing participation rates.
- **Re-engagement campaigns**: Push notifications or emails with deep links bring lapsed players back to new content or personalized offers, driving reactivation.
- **Ad campaign optimization**: Deep links in ad campaigns allow developers to track which ads lead to installs and in-game actions, making marketing spend more efficient.
> "We've helped top casual gaming companies maximize LTV and reduce churn through re-engagement campaigns. In 2025, we expect to see more of this, with deep linking playing a vital role in activating and optimizing these campaigns." — [Xavier Mariani, CEO and Founder](https://www.linkedin.com/in/xaviermariani/) of [Adikteev](https://www.adikteev.com/)
Deep linking is the core technology powering key marketing strategies in 2025—from organic growth to web-to-app conversions and affiliate marketing. Embrace the deep linking renaissance to seamlessly direct your app to success.
## **Drive organic app growth with Airbridge's DeepLink Plan**
### **86.13M tracking links created**
Airbridge users generated more than [86 million tracking links in 2024](https://www.airbridge.io/en/blog/airbridge-recap-2024), resulting in a staggering 2.628 trillion clicks throughout the year. That's an average of 20 billion clicks per month—double the volume of another well-known URL shortener.
Our tracking links go beyond simple link generation, offering deep linking capabilities and seamless navigation across all platforms and operating systems.
- If you're already using **Airbridge MMP**, you can access all of our deep linking features at **no extra cost**, including unlimited REST API calls to generate tracking links.
- If you're looking for a **cost-effective**, standalone deep linking solution, our **Airbridge DeepLink Plan** starts at just **$199/month**. (It's the [best alternative to Firebase Dynamic Links](https://www.airbridge.io/en/blog/introduce-deeplink-plan))
Join **over 1,000 apps** worldwide already growing with Airbridge's deep linking and measurement solutions,
Ready to drive organic growth, boost virality, and unify your marketing channels? See how Airbridge deep linking solutions can help you turn web users into loyal app users today.
---
## Driving Cross-platform Gaming Success with PC and Console Attribution
URL: https://www.airbridge.io/en/blog/cross-platform-gaming-measurement
Language: English
Published: 2025-03-21
The gaming industry is moving at top speed, driven by the pursuit of broader audiences and higher revenue streams. As game studios aim to expand their reach from mobile, many are bringing their titles to PC and console platforms to meet the growing demand for seamless gaming experiences across multiple devices.
In 2024, the gaming industry's revenue was $187.7 billion and is projected to grow to $213.3 billion by 2027. PC and console gaming combined accounted for 51% of the revenue in 2024 and are expected to continue showing steady growth.
([source: Newzoo Global Games Market Report 2024](https://newzoo.com/resources/trend-reports/newzoos-global-games-market-report-2024-free-version))
Several key factors are driving this trend, including the demand for cross-platform compatibility, free-to-play monetization, and influencer & eSport streaming.
### **Cross-platform Compatibility**
Players today expect a seamless gaming experience across PC, console, and mobile platforms. This growing demand has fueled trends like cross-platform play, cross-play, and cross-progression, which allow players to engage with their favorite games on any device, anytime.
- **Cross-platform:** Games available on multiple platforms, such as **Minecraft** and **Fortnite**, enable players to access them on iOS, Steam, or Nintendo Switch.
- **Cross-play:** Titles like **Apex Legends** and **Call of Duty: Warzone 2.0** let users play together in real-time across different platforms, such as PC and PlayStation.
- **Cross-progression:** Games like **Genshin Impact** and **Baldur’s Gate 3** allow players to continue their progress seamlessly across devices, such as switching from Android to PC without losing progress.
By embracing these features, game studios can deliver the seamless, flexible experiences players demand, ensuring greater engagement and loyalty across all platforms.
### **Free-to-Play (F2P) Monetization**
F2P model has made games more accessible to wider audiences while opening up new opportunities for revenue growth. By offering flexibility in monetization strategies, developers can tailor in-app purchases to match player behavior and preferences, ensuring sustained engagement.
Games like **League of Legends**, **Apex Legends**, and **Warframe** have successfully adopted the F2P model, using it to maximize player lifetime value (LTV) and grow their user bases. As gaming shifts away from one-time purchases, understanding in-game player behavior across platforms has become essential for optimizing both engagement and revenue.
### **eSports & Influencer Streaming**
Platforms like Twitch and YouTube Gaming have turned gaming into a spectator experience, with influencers and streamers playing a pivotal role in promoting games. Developers recognize the importance of enabling streamers to create high-quality content, as this directly impacts a game’s visibility and popularity.
PC games like **Valorant**, **Overwatch**, and [**The First Descendant**](http://www.airbridge.io/en/case-studies/nexon-cross-platform-tracking) dominate streaming platforms due to their superior visual quality compared to mobile games. By making games more accessible for streamers, developers can attract larger audiences through influencer-driven engagement.
![[The First Descendant’s TikTok UGC & Video Ads]](https://qhzcaeifhsdzzjpnbdqr.supabase.co/storage/v1/object/public/assets/blog/1774881657392-67dd004e9cd252f5ff080432_ad_4nxcijej2wr3mhucuzy85-lufa5uhjng0g6jaa9xznnktghyn9fwmlwggttfwvvxy0hkas6m6-gbv1y6bwehlsujx8gbe2gmr7lxu3g-1vlntdncp5a9cukz28j4pomwzrpss_sdyww.png)
With multi-platform games now the industry standard for success, player journeys have become more complex. This underscores the importance of accurate cross-platform measurement as the foundation for user acquisition (UA), engagement, and monetization strategies.
In this post, we'll explore how to transform PC and console platforms as performance marketing channels, leveraging precise cross-platform attribution to accurately measure marketing impact.
## Why Cross-platform Measurement Matters
PC·console game marketing extends beyond acquiring users; it involves securing high-value players, maintaining their engagement, and driving conversions. To achieve this, marketers need to require end-to-end visibility into the player journey, from discovery to in-game purchases.
Here are two key reasons why PC·console measurement is challenging and why it's essential.
### **Diverse UA Channels**
Players discover games through a variety of UA channels, such as search ads, display ads, social media, streamer and influencer promotions, and email marketing. Accurate measurement is critical for identifying which channels are most effective at reaching and engaging target audiences, enabling marketers to optimize strategies and drive better performance.
### **Cross-platform Player Journeys**
As players actively switch between multiple platforms and devices, they follow diverse paths to install, play, and make in-game purchases, creating complex cross-platform flows that challenge measurement. Now, marketers need to understand player flow by actively tracking cross-platform UA and in-game scenarios.
- **Cross-platform attribution:** For example, if a user installs a PC game after clicking a mobile ad, marketers must attribute the event to the proper ad to measure its effectiveness accurately.
- **In-game metrics:** Marketers must understand in-game metrics like time-to-first-purchase, transaction amounts, and churn rates to create personalized experiences and inform retargeting efforts.
## Cross-platform Measurement with Airbridge
Airbridge offers comprehensive cross-platform measurement, providing a unified view of user journeys across mobile, PC, and console platforms, from UA to in-game purchases. Conversion events from PC, console, and mobile platforms can be collected by Airbridge, allowing for seamless integration of these data points to optimize ad campaigns across all devices. With Airbridge, marketers can measure performance regardless of whether they have a dedicated marketing landing page for their multi-platform games.
- **Games with marketing landing page:** Games like **Fortnite** have dedicated marketing websites where users can explore details and access downloads. In this scenario, when users click an ad, they're redirected to a landing page with the Airbridge Web SDK installed. Airbridge then collects data from this website and the PC·console game to measure ad performance effectively.
[__wf_reserved_inherit]
- **Games without marketing landing page:** Games like **Darkest Dungeon**, rely on platforms like Steam or GOG for game information and purchases. In this scenario, Airbridge collects data using tracking links across platforms to measure the performance of PC·console ads effectively.
[__wf_reserved_inherit]
Airbridge’s cross-platform measurement links ad clicks to in-game actions, giving marketers a clear view of the user journey to optimize campaigns.
### Setting Up Key Metrics
To drive high conversion and engagement, defining and tracking key player interactions is the first critical step in optimizing cross-platform game marketing. To get started, consider setting up the following metrics to understand player progression and monetization potential.
- Install complete – Game installation confirmed
- Tutorial complete – Player finishes onboarding
- First login – Player logs in for the first time
- Level 10 achieved – Milestone progression
- First purchase – Initial in-game transaction
- Subscription purchase – Recurring revenue events
![[The First Descendant’s metric setup in the Airbridge Dashboard]The data used in this dashboard is randomly generated and intended solely for demonstration purposes.](https://qhzcaeifhsdzzjpnbdqr.supabase.co/storage/v1/object/public/assets/blog/1774881658315-66d6a5e921bbc72d0eb2599c_66d6a56c66cc61b146e5ab73_abr_case-2520studies_nexon_dashboard01-en-2520-2-.png)
Nexon utilized Airbridge's PC·console measurement to optimize the launch of [**The First Descendant**](http://www.airbridge.io/en/case-studies/nexon-cross-platform-tracking). By providing actionable lower-funnel, in-game metrics, Airbridge enabled The First Descendant team to move beyond traditional upper-funnel metrics like CPC, CPM, and CTR, which are limited to branding and awareness.
With Airbridge, The First Descendant team gained comprehensive visibility into user journeys from ad clicks to downloads, in-game purchases, and conversions across both PC·console platforms from a unified dashboard. This insight allowed the team to rapidly iterate on new ad products, identify top-performing channels, campaign types, and creatives, and strategically allocate resources to attract high-value players. The First Descendant achieved success through the following strategies:
- **UA performance analysis with down-funnel metrics:** The team used down-funnel metrics to identify the best-performing ad products across platforms like Google, Meta, TikTok, X, Reddit, and Amazon Ads (Twitch), maximizing the impact of each ad creative.
- **Campaign optimization via CAPI Integration:** The team leveraged Conversion APIs (CAPI) for Meta and X to connect PC·console in-game conversions with ad performance, enabling seamless campaign optimization.
- **Targeted marketing based on high-value players:** The team identified players with high lifetime value (LTV) and developed targeted marketing strategies based on in-game behavior.
### Integrating Server-to-Server (S2S) API
Since PC·console games primarily operate in a server-based environment, [Airbridge integrates S2S APIs](https://help.airbridge.io/en/guides/pc-game-tracking-guide#s2s-event-api-configuration) to track events accurately. For instance, when a player completes their first in-game purchase, marketers can send this data to Airbridge via API, enabling precise marketing attribution and campaign optimization.
### Leveraging Probabilistic Attribution
Airbridge uses [probabilistic attribution](https://help.airbridge.io/en/guides/pc-game-tracking-guide), employing statistical modeling to connect ad clicks with conversions across platforms without relying on cookies or device IDs. As ad-blocking and privacy restrictions increase, this method is becoming vital for providing accurate performance insights while maintaining strict privacy compliance.
## Key Takeaways
- Game studios are expanding from mobile to PC·console platforms to reach broader audiences and revenue streams.
- This shift is also driven by the growing demand for seamless cross-platform gaming experiences.
- Cross-platform measurement is challenging due to players switching between platforms and devices through multiple UA channels.
- Airbridge offers a unified view of user journeys across mobile, PC, and console platforms, from user acquisition to in-game purchases.
- Airbridge also leverages S2S APIs and probabilistic attribution to deliver accurate insights while maintaining privacy compliance.
- With Airbridge, game marketers can transform PC and console campaigns into performance marketing channels.
---
## Airbridge Integrates with Naver Ad Audience
URL: https://www.airbridge.io/en/blog/naver-ads-audience-sync
Language: English
Published: 2025-03-14
Airbridge integrates with Naver Ad Audience, enabling marketers to maximize ad performance on South Korea's leading platform. This integration allows marketers to leverage the most up-to-date customer data, creating smarter, sharper, targeted Naver Ads. By combining Airbridge's advanced features with Naver's robust ad ecosystem, marketers can optimize their campaigns to achieve better results in the South Korean market.
## **Naver Ads: Unlock South Korea’s Largest Audience**
Naver Ads provides a comprehensive suite of advertising tools, including Search Ads, Display Ads, and Shopping Ads, on Naver, South Korea's leading online platform and search engine. Naver's targeted advertising capabilities are powered by its extensive database, proprietary AI technology and features such as Naver Data Lab, which offers in-depth insights into Korean search trends.
## **Key Benefits of Airbridge & Naver Ad Audience Integration**
[__wf_reserved_inherit]
Airbridge's integration with Naver Ad Audience allows marketers to create granular customer segments using first-party data and send them seamlessly to the Naver Ad platform. This integration includes two key features:
- Behavior-based Custom Targeting: On Airbridge, marketers can create audience segments based on specific events like app installations, sign-ups, or purchases. These segments can then be sent to the Naver Ad platform to run more refined, user-tailored ads.
- Customer Data Auto-Sync: Airbridge automatically updates the latest customer data to the Naver Ad platform daily, keeping ad targeting optimized with the most current data and providing real-time insights into user behavior.
With this integration, Airbridge now offers comprehensive integration with Naver Ad Audience, Naver Ads, and Naver Ad costs, providing unmatched connectivity with South Korea’s largest ad platforms, including Kakao, Toss, Karrot, and Everytime, outperforming other MMPs. With Airbridge, marketers can set themselves up for success in the Korean market by leveraging our seamless connections to key ad platforms.
Beyond South Korea, Airbridge is fully integrated with leading global ad platforms, such as Meta Ads, Google Ads, TikTok for Business, Criteo, Altibase, Appier, and Remerge. To learn more about sending audiences to ad channels, check out our [detailed guide](https://help.airbridge.io/en/guides/send-audience).
Ready to expand your audience through Naver’s Ad network? Partner with Airbridge today to drive new growth channels in the South Korean market.
---
## Optimize Your Mobile Marketing with Digital Turbine & Airbridge Integration
URL: https://www.airbridge.io/en/blog/optimize-your-mobile-marketing-with-digital-turbine-airbridge-integration
Language: English
Published: 2025-03-13
Airbridge has partnered with [Digital Turbine](https://www.digitalturbine.com/) to enhance your mobile marketing strategies. This integration combines Airbridge's advanced analytics capabilities with Digital Turbine's innovative advertising solutions, enabling marketers to measure, track, and optimize their campaigns effectively.
## From UA to Monetization, Leverage Digital Turbine's Extensive Network and Growth Suite
Digital Turbine provides exceptional reach across various ad spaces and in-app environments, offering diverse ad formats, innovative features, and global partnerships that elevate your mobile marketing and app distribution goals.
With a diverse growth suite that drives your mobile strategy, DT’s offering with AirBridge is focused on:
- **DT DSP:** Digital Turbine's demand-side platform empowers advertisers to effectively reach and engage mobile users through a variety of ad formats, including display ads, video ads, native ads, and more. The premium environment offers global reach at scale.
- **DT Offer Wall:** Digital Turbine's non-intrusive monetization and UA solution presents users with a curated list of rewarded actions, such as app installs, surveys, or in-app purchases, which they can complete to earn in-app currency or rewards. An extension of the app environment with a natural look and feel, Offer Wall has proven to be a strong driver for [monetization](https://www.digitalturbine.com/case-studies/bagelcode-exceeds-d7-roas-goals-by-7x-with-dt-offer-wall) and [UA](https://www.digitalturbine.com/case-studies/dt-growth-drives-funvents-massive-latam-expansion) strategies.
## Effortless performance tracking with Airbridge
Airbridge's integration with Digital Turbine allows for seamless performance tracking of mobile app campaigns, equipping marketers with clear insights for optimization.
Key benefits include:
- Accurate monitoring of performance metrics from Digital Turbine's various ad placements.
- A comprehensive view of marketing initiatives across channels from a unified dashboard.
- Identification of top-performing campaigns and placements through precise attribution.
- Analysis of user behavior and lifetime value via detailed cohort analysis.
- Opportunities for A/B testing to refine ad messaging.
## Getting Started
Setting up the integration between Airbridge and Digital Turbine is easy. Follow these steps to unlock the opportunity:
1. Create tracking links in Airbridge for your Digital Turbine campaigns.
2. Configure postbacks for real-time data exchange based on your conversion goals.
3. Use Airbridge analytics to gain insights into user behavior and campaign performance.
4. Adjust your strategies based on the data collected.
Contact the Airbridge team today to learn how Airbridge and Digital Turbine can enhance your mobile app campaigns.
---
## 2 Ways to Calculate Predictive Lifetime Value (pLTV) for Your Mobile App
URL: https://www.airbridge.io/en/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app
Language: English
Published: 2025-03-06
## Why pLTV Matters More Than Ever
In today’s highly competitive mobile app ecosystem, understanding Predictive Lifetime Value (pLTV) is crucial. As acquisition costs continue to rise, marketers must gain early insights into user value to make data-driven decisions that optimize their budgets and drive sustainable growth.
> 💡 What is pLTV
> Predictive Lifetime Value (pLTV) is a forward-looking metric that estimates the total revenue a user is expected to generate throughout their entire relationship with a mobile app or service.
### Key Reasons Why pLTV Is Crucial
- Extended Payback Periods: Rising user acquisition costs mean it takes longer to recoup initial advertising expenses. Without a clear understanding of a user's long-term value, it's difficult to optimize acquisition strategies effectively.
- Gradual Revenue Realization: Many mobile apps, particularly those relying on subscriptions or in-app advertising, generate revenue slowly over time. As a result, traditional short-term metrics often fail to capture a user's full potential value.
- Real-Time Budget Optimization: Marketers need to make quick, data-driven decisions to allocate budgets effectively. Accurate pLTV predictions help them optimize spending across campaigns and channels to maximize return on investment (ROI).
By leveraging pLTV, app marketers can proactively adjust acquisition and retention strategies, even when short-term revenue does not fully reflect a user's potential lifetime value.
## 2 Ways to Calculate pLTV
Let's look at two main methods for calculating pLTV: Curve Fitting and Machine Learning.
### Curve Fitting Method
This approach uses historical retention data to model a mathematical curve, predicting future user retention and revenue patterns.
> **How it works
> **1**.** Retention Curve Analysis: Select a mathematical function (e.g., exponential, power) that fits the retention trend and integrate it over a set period (e.g., 30, 60, or 90 days) to estimate user longevity.
> 2\. ARPDAU Calculation: Segment users by cohorts (e.g., country, OS, channel) and determine their Average Revenue Per Daily Active User (ARPDAU), incorporating both in-app purchases (IAP) and in-app ad revenue (IAA).
> 3\. Multiplication for pLTV: Multiply the integrated retention curve by ARPDAU to estimate cumulative LTV.
**Strengths:**
- Effective for analyzing broad trends in user behavior.
- Well-suited for apps with strong correlation between user activity (sessions) and revenue.
**Limitations:**
- Less effective for apps with complex monetization models where a small percentage of users generate disproportionate revenue (e.g., IAP-heavy games).
- Limited personalization and struggles with irregular retention patterns.
### Machine Learning Method
Machine learning enables dynamic, granular LTV forecasts by detecting complex patterns in user behavior.
> **How it works
> **1\. Feature Selection: Incorporate factors such as purchase history, session frequency, user demographics, and engagement metrics.
> 2\. Model Training: Utilize machine learning algorithms (e.g., Gradient Boosting, Neural Networks) to predict future LTV based on historical data.
**Strengths:**
- More accurate and personalized predictions.
- Well-suited for apps with irregular revenue and retention behaviors.
**Limitations:**
- Requires substantial datasets and ongoing model tuning.
- Can be resource-intensive and complex to implement.
### The Role of MMPs in pLTV Calculation
Many marketers assume that pLTV can be calculated using only first-party data (e.g., OS, country, device type). However, incorporating attribution data from a Mobile Measurement Partner (MMP) significantly enhances accuracy.
**Why Attribution Data Matters:**
- Different ad sources attract users with varying lifetime values. Understanding where users came from helps predict long-term engagement and revenue potential.
- For instance, users acquired via short-form video ads targeting a younger audience may have different retention patterns compared to those from search ads.
By integrating first-party data with MMP attribution insights, marketers can make better-informed budget allocation decisions, optimizing campaigns based on user quality, not just acquisition volume.
### Airbridge’s pLTV Feature: Precision at Scale
Airbridge uses a Bayesian curve-fitting approach along with its proprietary [Luft Engine](https://www.airbridge.io/en/blog/introducing-luft), a database built for analyzing user behavior, to deliver an accurate and reliable pLTV solution designed specifically for mobile marketers.
[How to Calculate Predictive Lifetime Value (pLTV) for Your Mobile App]
#### **Key Benefits of Airbridge’s pLTV Feature**
- Granular Insights: Analyze pLTV at multiple levels—by country, OS, campaign, and even ad creative.
- Long-Term Predictions with Minimal Data: Requires only 3 days of data to generate reliable pLTV predictions for up to 180 days.
- One-Click Analysis: Compare pLTV, CAC, and ROAS seamlessly within the Airbridge dashboard.
- No Hidden Costs: Included in the standard Airbridge package, ensuring accessibility for all businesses.
With Airbridge, marketers can make faster, data-driven decisions, ensuring efficient budget allocation across campaigns and regions.
### Case Study: DelightRoom (Alarmy) Optimizes Global Marketing with Airbridge’s pLTV Feature
[How to Calculate Predictive Lifetime Value (pLTV) for Your Mobile App]
#### **The Challenge**
Before using Airbridge, Delightroom faced challenges in manually calculating pLTV for global campaigns. They needed detailed insights segmented by channel, campaign, creative, country, and OS—at quarterly, monthly, and weekly intervals.
#### **The Impact**
By adopting Airbridge’s pLTV feature, DelightRoom instantly accessed granular insights, eliminating a manual process that previously took over 3 hours. They now dynamically monitor pLTV and predictive ROAS (pROAS), enabling real-time budget optimization.
**Delightroom’s Solution: Smart Budget Reallocation**
- DelightRoom used pLTV predictions alongside CPI and retention metrics to make more informed decisions.
- For ad creatives with high pLTV but poor CPI or retention, they adjusted budgets accordingly, preventing inefficient spend.
By leveraging Airbridge’s pLTV capabilities, DelightRoom successfully streamlined budget allocation, marketing efficiency, and overall return on investment.
### Unlock Your Growth Potential with Predictive LTV
With faster, more accurate LTV predictions, marketers can confidently allocate budgets, enhance campaign performance, and drive sustainable growth in an increasingly competitive space.
Contact the Airbridge team today to discover how Airbridge’s pLTV feature can help you make smarter marketing decisions and maximize your app’s long-term profitability.
---
## LLM & RAG: AI Technologies Reshaping the Future of Martech
URL: https://www.airbridge.io/en/blog/llm-and-rag-ai-technologies-reshaping-the-future-of-martech
Language: English
Published: 2025-02-24
## **The Evolution of Martech Solutions**
Martech solutions used by marketers, PM/POs, and growth professionals are evolving at an unprecedented pace, offering sophisticated capabilities across ad operations, customer data analytics, and personalized messaging. However, as these solutions grow in complexity, they can become daunting for users unfamiliar with martech and its applications. Even with comprehensive documentation, keeping up with frequent updates can be challenging—even for experts.
To address this, martech solutions are increasingly integrating AI technologies to lower the barrier to adoption and enhance business impact. Many martech platforms have introduced features leveraging Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG), allowing users to ask questions, generate personalized content, and execute unfamiliar commands without constantly referring to guides or waiting for support responses.
### **Understanding LLM & RAG**
> **Large Language Model (LLM):** An AI model capable of understanding natural language input and generating human-like responses.
> **Retrieval-Augmented Generation (RAG):** A technology that enhances LLMs by retrieving relevant real-time information, ensuring more accurate and context-aware responses.
## **How Leading Martech Solutions Are Leveraging LLM & RAG**
### **1\. Ask Airbridge: An LLM & RAG-Powered AI Assistant**
[LLM & RAG: AI Technologies Reshaping the Future of Martech]
[_Ask Airbridge_](https://help.airbridge.io/en/guides/ask-airbridge) is an AI assistant that utilizes LLM and RAG to answer user queries based on documentation from the Airbridge Help Center. This feature makes it easier for users to navigate the platform, even if they are unfamiliar with martech terminology, without needing to sift through documentation for every update.
**With _Ask Airbridge_, users can effortlessly obtain information on:**
- Basic concepts on User permissions, tracking link creation, attribution rules, and iOS SKAN.
- Integration details for Airbridge ad channels and third-party platforms.
- Development tasks, such as SDK installations and server configurations for data transmission between apps and the web…and more!
#### **Sample Prompts for _Ask Airbridge_:**
- "I want to run a SKAN campaign to measure iOS performance. What do I need to prepare in Airbridge and what should I keep in mind when importing SKAN data?"
- "How do I redirect users with the app installed to the app, and users without the app installed to the app market when they click the tracking link? I want to redirect users with an app to a different in-app destination."
### **2. _Ask Amplitude_: AI-Powered Data Analysis & Visualization**
Since acquiring AI analytics company Command AI in October 2024, Amplitude has been aggressively integrating AI into its platform. [_Ask Amplitude_](https://amplitude.com/docs/analytics/ask-amplitude) and [_Data Assistant_](https://amplitude.com/docs/data/use-ai-data-assistant) are two standout features leveraging LLM and RAG, making it easier to create charts and streamline data organization for better decision-making.
[LLM & RAG: AI Technologies Reshaping the Future of Martech]
**_Ask Amplitude_, powered by OpenAI's API, is particularly helpfulfor users with limited experience in product analytics or data taxonomies. In addition to providing how-to answers, it enables users to:**
- Create or modify charts using natural language commands.
- Search for dashboards, charts, and reports within the platform.
- Share insights with teams through threaded discussions.
Furthermore, _Ask Amplitude_ can analyze event and property usage patterns to identify frequently queried combinations, allowing users to generate charts aligned with their business objectives in just a few clicks.
#### **Sample Prompts for _Ask Amplitude_:**
- "Which category has the highest purchase rate for recommended items?"
- "What is the average session length of users who played five or more songs?"
- "How long does it take for companies on free vs paid plans to invite their first three users?"
[LLM & RAG: AI Technologies Reshaping the Future of Martech]
_Data Assistant_, another AI-powered feature, automatically analyzes complex data structures and suggests optimizations. When dealing with excessive events and properties, it identifies key insights, providing actionable recommendations powered by LLM-generated explanations. Users can then accept or reject these suggestions with a single click, simplifying data management and maximizing business impact.
### **3\. _Braze AI_**_™_**: Personalized Copywriting & Liquid Assistant**
[_Braze AI™_](https://www.braze.com/product/brazeai) streamlines content personalization through LLM technology, offering [_AI Copywriting Assistant_](https://www.braze.com/docs/user_guide/brazeai/generative_ai/ai_copywriting) and [_Liquid Assistant_](https://www.braze.com/docs/user_guide/brazeai/generative_ai/ai_liquid#brazeaitm-liquid-assistant)—both powered by OpenAI API—to optimize marketing messaging.
#### **_AI Copywriting Assistant_**
This feature generates high-quality, brand-aligned copy based on user inputs. Users can specify:
- Message intent
- Brand tone and guidelines
- Desired length, style, and language
The AI then produces optimized copy tailored to these parameters. Additionally, when crafting push notifications, marketers can reference past campaign data to generate messages consistent with prior successful content.
[LLM & RAG: AI Technologies Reshaping the Future of Martech]
#### **_Liquid Assistant_**
**_Liquid Assistant_ simplifies the creation and optimization of Liquid code, eliminating the need to master complex syntax. Users can:**
- Generate personalized Liquid templates effortlessly.
- Receive AI-driven suggestions to refine existing Liquid code.
- Learn Liquid syntax through AI-provided explanations.
Beyond content personalization, _Braze AI™_ also includes AI-driven product recommendations, predictive targeting, and automated A/B testing.
[LLM & RAG: AI Technologies Reshaping the Future of Martech]
## **Conclusion**
By incorporating LLM and RAG, martech solutions are evolving to be more user-friendly, allowing marketers and growth teams to focus on strategy and execution rather than troubleshooting complex platforms. Whether you're a seasoned expert or new to martech, AI-powered assistants can make your workflow more efficient and impactful.
Curious about how **_Ask Airbridge_** can save you time and effort?? Connect with Airbridge experts today.
---
## Power CTV Ads with Airbridge & Vibe Integration
URL: https://www.airbridge.io/en/blog/power-ctv-ads-with-airbridge-vibe-integration
Language: English
Published: 2025-02-21
Airbridge is now integrated with leading CTV ad platform, Vibe.co. This powerful partnership combines Airbridge's robust analytics with Vibe's extensive premium inventory across over 500 streaming TV apps and channels.
# Grow UA with Vibe's CTV ad network
Vibe is elevating TV ad performance by turning CTV into a powerful, outcomes-driven channel. With 238 million monthly U.S. viewers, $30 billion in ad spend for 2024, and 82% of CTV users taking action after seeing an ad, CTV has become an advertising must-buy.
[__wf_reserved_inherit]
Vibe makes it easy for marketers to discover and target audiences across streaming TV apps and channels with simplicity and precision. With Vibe, you can access:
- Diverse Ad Space: Display ads on over 500 streaming TV apps and channels, ranging from movies to live sports and local news.
- Broad Audience Reach: Access 120 million households in the U.S. nationwide or focus on local communities around selling locations.
- Granular Targeting: Target viewers based on socio-demographic characteristics, interests, and intent to buy—free from ATT-related signal loss.
[__wf_reserved_inherit]
# Amplify CTV campaigns with Airbridge
The Airbridge and Vibe integration combines advanced analytics and CTV advertising capabilities seamlessly. With this integration, marketers can:
- Review relevant KPIs in one view on Airbridge's unified dashboard.
- Accurately measure key metrics like LTV, CPI, ROAS, and CTV-enabled revenue growth across campaigns.
- Gain detailed insights into the most effective ad formats and campaigns.
- Understand long-term viewer engagement through comprehensive cohort analysis.
- Make data-driven campaign adjustments with real-time reporting.
# Getting Started
Setting up the Vibe x Airbridge integration is easy. Follow these steps:
1. Set up Vibe campaign tracking links in Airbridge.
2. Configure postbacks for real-time data exchange based on your conversion goals.
3. Use Airbridge analytics to gain insights into user behavior and campaign performance.
4. Adjust your strategies based on collected data.
For detailed instructions, refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/vibe). To learn how Airbridge and Vibe can maximize your CTV advertising efforts, contact the Airbridge team today.
---
## What is LLM(Large Language Model) and Why is it Important?
URL: https://www.airbridge.io/en/blog/what-is-llm
Language: English
Published: 2025-02-17
LLMs (Large Language Models) are transforming the way we interact with AI in 2025. These advanced systems generate human-like text, enabling marketers to create more engaging and personalized content. LLMs have evolved from simple technological advancements to become essential assets for innovative marketers looking to stay ahead in the competitive digital space.
## 💡 What is LLM?
A **Large Language Model (LLM)** learns from vast amounts of text data to understand and generate human-like language.
AI tools like ChatGPT use LLM technology to process and create natural conversations. In this post, we’ll explore how LLMs work, their key features, and how marketers can use them effectively.
## Key Features of LLM and How it Works
### 🔹 Key Features of LLM
- **Big Data Learning:** LLMs study billions of words and sentences. This helps them understand complex language and context.
- **Natural Language Processing:** LLMs analyze text, translate languages, and detect emotions in writing.
- **Human-like Content Generation:** LLMs create text that sounds like a real person.
### 🔹 How Does LLM Work?
LLMs employ a deep learning model called Transformer to grasp word relationships and context. This enables them to produce meaningful and coherent text, rather than simply stringing words together randomly.
For example, when you input "Today's weather is," an LLM predicts likely next words such as "sunny" or "rainy," mimicking how a person would complete the sentence.
### How Marketers Use LLM
LLMs make marketing more efficient by automating tasks and improving customer engagement. Marketers leverage LLMs to:
1. **Automate Content**Create content at scale: LLMs generate blog posts, emails, and social media content from keywords, enabling marketers to produce engaging copy quickly and consistently.
2. **Personalize customer experience
**LLMs analyze customer data to craft tailored emails and push notifications, boosting relevance and conversion rates.
3. **Improve Customer Support and Chatbot responses
**LLM-powered chatbots provide natural, human-like responses to customer inquiries, enhancing satisfaction and operational efficiency.
### Master LLM for Marketing in 2025
LLMs are transforming industries, and marketing is no exception. They automate content, personalize customer interactions, and improve real-time support.
Leading Mobile Measurement Partners (MMP) like **Airbridge** harness the power LLMs to drive marketing success. Want to see how? Click below to learn more! 👇
---
## What is an MMP? The secret tool for campaign success
URL: https://www.airbridge.io/en/blog/what-is-mmp
Language: English
Published: 2025-02-13
## Spending on Ads, but not seeing results?
You're not alone. Clicks don’t always mean installs. Installs don’t always mean conversions into sign-ups or revenue. And tracking it all manually across different dashboards? A nightmare.
The problem? Every ad platform claims credit for the same installs. You’re drowning in duplicated, messy attribution data, and guessing where your budget should go. That’s where **Mobile Measurement Partners (MMPs)** come in. They provide visibility into what truly drives growth—so you stop wasting money and start making smarter moves.
> 💡 What is an MMP?
> An **MMP (Mobile Measurement Partner)** is your **independent source of truth** in app marketing. It attributes, measures, and analyzes performance across every channel (with granularity down to cohort, campaign, creative, country, and more) — so you know exactly what’s driving installs, engagement, and revenue.
## **Key roles of MMP**
### **1\. Tracks what matters with every data in one place**
Instead of jumping between dashboards which may result in duplicate attributions, an MMP [consolidates all your ad performance, cost, revenue, and in-app event data into one place.](https://www.airbridge.io/product/marketing-analytics) It shows you what’s driving installs, sign-ups, and purchases—so you know where to double down and where to cut losses.
### **2\. Offers Diverse Attribution Data**
Not all installs or conversions follow the same path. With an MMP, you set the rules—whether it’s lookback windows, web-to-app, retargeting, re-engagement, reactivation, uninstall, engaged view, organic, or view-through attribution.
These diverse types of attribution methods [enable marketers to effectively track and evaluate advertising performance across multiple touchpoints](https://www.airbridge.io/en/blog/what-is-marketing-attribution-model).
### **3\. Fights Ad Fraud Before It Drains Your Budget**
Fake clicks. Fake installs. Wasted ad spend. Fraud is everywhere. An MMP detects and blocks bad traffic before it eats into your ad dollars—so every cent works harder.
[See how we helped eliminate 90% of affiliate marketing fraud for a hotel booking app](https://www.airbridge.io/en/case-studies/go2joy-fraud-prevention-detection)
## **Benefits of MMP**
Mobile marketers often struggle to determine which channel is truly effective when conducting multi-channel campaigns across social media, search ads, display ads and more. MMPs address this challenge by offering several key benefits:
### **1\. Smarter Decisions, Backed by Data**
No more relying on gut feelings. An MMP gives you **real** numbers—showing which campaigns and creatives drive installs, in-app purchases, subscriptions, and more—so you can double down on what works and cut what doesn’t.
MMPs also unlock granular insights to **pinpoint** high-ROAS (Return on Ad Spend) channels, top-spending cohorts with high LTV, finer user segments categorized by ARPU, ARPPU, ARPDAU, and more.
### **2\. Saves Time, Saves Money**
Manually collecting data from multiple platforms? Painfully slow. An MMP **automates the heavy lifting**, so you can focus on strategy, not spreadsheets. And by optimizing spend, it makes sure every dollar works harder.
### **3\. Real-Time Postbacks for Smarter Targeting**
MMPs send in-app event postbacks to ad networks—helping platforms optimize targeting and bidding **in real time**. MMPs can also create custom audiences for you to sync for retargeting or re-engagement campaigns. This means **better audience precision, reduced wasted spending, and higher ROAS** with every campaign.
### **4\. AI-Powered Future Forecasting**
[MMPs harness AI to deliver predictive analytics](https://www.airbridge.io/product/airbridge-ai). They forecast pLTV, predict cohort-based revenue, and employ simulations to bridge user data gaps in iOS attribution. They also integrate robust ad fraud prevention to automatically detect fraud patterns to ensure cleaner performance metrics.
### **Expect more from your MMP.**
An **MMP** is crucial for your app’s success, but not all MMPs give you what you really need. [**Airbridge**](http://airbridge.io) is the only MMP on the market with **ZERO paywalls** and **everything unlocked**—from cross-platform attribution and raw data export to cost and revenue analytics, fraud protection, audience segmentation, CS support, and deep links with link branding.
> "_Airbridge gave us everything we needed from an MMP to grow and scale, without hidden costs.”
> \-Alarmy
> ■No.1 Alarm clock & Sleep App with 75M with 1.7M store reviews. Selected as the 2023 Google Play App of the Year._
### **Finding your next MMP? See how we measure up.
**
**Airbridge vs. AppsFlyer
**If you’re tired of upsells, premium pricing, and endless CS tickets, Airbridge gives you everything you need—no hidden fees, no frustration.
[→learn more](https://www.airbridge.io/compare/airbridge-vs-appsflyer)
**Airbridge vs. Adjust
**If you’re frustrated with paywalls, attribution gaps, and fraud blackboxes, Airbridge closes every gap.
[→learn more](https://www.airbridge.io/compare/airbridge-vs-adjust)
**Airbridge vs. Singular
**If you’re sick of slow queries and limited granularity? Airbridge is real-time, lightning-fast, and fully granular—because waiting isn’t an option.
[→learn more
](https://www.airbridge.io/compare/airbridge-vs-singular)
Ready to elevate your mobile marketing strategy with an MMP? Don’t settle for less. Choose the MMP that does more. Choose Airbridge.
---
## Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025) — And How to Prevent Them
URL: https://www.airbridge.io/en/blog/top-3-mobile-ad-fraud-tactics-in-vietnams-fintech-industry-2025-and-how-to-prevent-them
Language: English
Published: 2025-02-13
## Overview
The financial technology (Fintech) industry in the APAC region is undergoing significant changes, with remarkable growth in key metrics such as app installations, mobile session activity, and revenue from financial applications. In Vietnam, this market is expanding rapidly with the emergence of numerous applications offering payment services, unsecured lending, and personal financial management.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
However, the market boom has also led to a rise in increasingly sophisticated ad fraud activities, significantly impacting the Fintech sector. The high rate of install fraud in Fintech applications stems from several factors, including large marketing budgets; a lack of in-depth understanding of key advertising KPIs, especially among traditional banks or investment firms new to digital advertising; and the highest average cost per install (CPI) in the market, creating strong incentives for fraudulent activities.
Another key reason for the high fraud rate is that the Fintech industry is still **in the early stages of transitioning from desktop to mobile**. Right now, the focus is more on growing the user base rather than optimizing revenue. Additionally, the cost of acquiring a user in the financial sector is typically higher than in other app categories, as users are expected to have higher retention rates and more frequent transactions.
According to recent reports, Vietnam is one of the most heavily affected markets by mobile ad fraud, with an estimated **10-30% of advertising budgets lost** each year. This issue not only leads to wasted resources but also negatively impacts business performance and erodes user trust in Fintech applications.
## Summary
Ad fraud in the Fintech industry is becoming increasingly complex, with tactics like bot fraud, mistargeting, device farms, and attribution fraud leading to budget losses, data distortion, and reduced campaign effectiveness. Airbridge provides powerful solutions, including raw data analysis, fraud validation rules, and rigorous monitoring on the MMP platform. These tools help Fintech businesses quickly detect and prevent fraudulent activities, safeguard their budgets, and optimize advertising performance.
## Common Types of Ad Fraud in the Fintech Industry
Bots are the most dominant type of fraud impacting the Fintech mobile market, accounting for over **50% of all fraudulent activities**. These bots cause significant harm to businesses by launching large-scale attacks within a short period, rapidly distorting data and compromising campaign effectiveness.
Currently, the ad fraud rate in the Fintech sector is around 31.8%, meaning that 1 in every 3 app installs is fraudulent. To help businesses stay protected, let’s explore the most common types of ad fraud in Fintech, how they operate, and how to identify them with Airbridge.
### Fake Touchpoints & Conversions (Device Farms)
**What are Device Farms?**
Device farms use intelligent bots along with a large number of outdated, low-cost devices to generate massive amounts of fake touchpoints or conversions. According to statistics, device farms can leverage thousands of devices to simulate post-install events for up to a month after each installation.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
**How Device Farms Work**
Data from Interceptd indicates that this type of fraud accounts for up to 25% of all fraud cases, and this figure is even higher in the Fintech industry, reaching 31%, according to Business of Apps. As a result, Jupiter Research estimates that Device Farms consume a significant portion of Fintech companies' advertising budgets, ranging from 12% to 20%.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
**“Symptoms” of Device Farms**
This type of fraud is typically identified by abnormally low conversion rates and engagement levels, as most users from these financial apps are fake. The key difference from attribution fraud is the extremely poor user quality, with retention rates showing little to no improvement.
More specifically, retention rates tend to drop sharply from Day 1 and nearly reach zero between Day 5 and Day 7, as fake accounts have no motivation to perform natural actions such as logging in or interacting with the app after being created.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
The lack of motivation to continue using the app is also reflected in the low in-app conversion rates. By analyzing conversion metrics at each step of the eKYC (electronic Know Your Customer) process, businesses can track the customer journey after installation across different channels, allowing them to identify and eliminate channels that exhibit suspicious in-app event patterns.
For example, after installation, regular users tend to complete the eKYC steps at a consistent rate because they have a real need for the app, giving them the motivation to proceed. This pattern can be observed in channels such as “cauly,” “naver,” and “tradingworks” below.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
However, for installs generated by Device Farms, since they are primarily created to register installs, the devices have no motivation to perform further actions within the app, similar to the installs from the “Discord” channel shown below.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
Additionally, device farms often rely on outdated, low-cost devices with older OS versions. In the example below, the operating systems are noticeably outdated, commonly found in older, budget devices that no longer support upgrades to the latest iOS 18.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
### Attribution Fraud
**What is Attribution Fraud?**
Attribution fraud is often identified through the phenomenon of "organic cannibalization." If the number of non-attributed installs (= organic installs) decreases while attributed installs (= paid installs) increase, despite the ad budget remaining stable, it suggests that organic installs are being hijacked and incorrectly recorded as paid installs.
[]
**“Symptoms” of Attribution Fraud**
Organic Cannibalization often occurs when the Click-to-Install Time (CTIT) is too short or when there is an unusual ratio between clicks and installs, such as a high number of clicks but few actual installs. This type of fraud is carried out through click injection or click spamming, both of which are forms of attribution fraud.
#### Click injection
Click injection occurs when a fraudulent intermediary detects the moment an app is being downloaded and inserts a fake touchpoint or interaction just before the MMP records the completed installation. This allows the fraudster to claim credit for the install as the winning touchpoint.
[]
This type of fraud is specific to Android devices, exploiting the operating system's broadcast system. When a new app is installed on an Android device, a signal is sent to other apps. This system was originally designed to enhance connectivity between Fintech-related apps on a user's device, such as supporting deep linking or streamlining the login process.
#### Click Spamming
Click Spamming, also known as click flooding, is a form of ad fraud where fraudsters generate fake clicks and insert them into the journey of real users without their knowledge or consent. This can be carried out through background-running apps, such as memory cleaners or battery-saving tools, which trigger clicks at any time. As a result, data becomes distorted, reducing its reliability and leading to wasted ad spend on ineffective campaigns.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
### SDK Spoofing
**What is SDK Spoofing?**
This type of fraud occurs when fraudsters hack an MMP's SDK (including login credentials, data packets, etc.) to send fake signals. As a result, they generate fraudulent installs that appear legitimate by using real device data. This technique is also known as a man-in-the-middle attack.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
Specifically, fraudsters break SSL encryption between the tracking SDK and its backend servers to generate a series of test installs for the target app. Once they identify the URLs corresponding to specific in-app events, they manipulate the dynamic parts of the URL to create fake installs.
Once enough information is gathered, fraudsters can repeatedly execute this process indefinitely, resulting in significant financial losses for advertisers.
**“Symptoms” of SDK Spoofing**
SDK Spoofing is most noticeable when multiple transaction IDs or user IDs appear in the MMP data but do not exist in the internal systems (CRM, ERP). In the example below, user ID user\_161110 and transaction ID tx\_97776154 may be fraudulent, as they are recorded only in the MMP but do not appear in the original CRM data.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
Moreover, abnormally low conversion rates and unrealistically short time intervals between events—such as app installation, app opening, and the first user action—are strong indicators of SDK Spoofing, as they deviate from real user behavior.
Some impacts of this fraud on businesses:
- **Wasted UA (User Acquisition) budget**: Ad spend is lost on fraudulent installs or fake interactions.
- **Disruption of key business KPIs**: Critical metrics like conversions and retention rates are distorted, leading to misleading data and ineffective advertising strategies.
## How Can Airbridge Help Fintech Businesses in Vietnam?
Identifying these types of fraud is a major challenge for businesses, often requiring significant time, effort, and financial resources to detect and eliminate fraudulent installs. With **Airbridge’s Abnormal Install Report**, powered by AI, businesses can automatically monitor suspicious install volumes across different fraud categories. This makes fraud detection easier, faster, and more cost-effective than ever.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
Beyond fraud detection, Airbridge actively supports Fintech partners in combating ad fraud by working closely with their **Customer Success teams** to implement **Fraud Validation Rules** tailored to their campaign needs.
More specifically, attribution fraud methods like **Click Injection** and **Click Spamming** can be effectively eliminated using the **Lag Time** and **Frequency Capping** features. With **Lag Time**, businesses can configure settings to reject postbacks for installs where the Click-to-Install Time (CTIT) is under 10 seconds, as it is unrealistic for a user to click, download, install, and open an app within such a short time frame.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
With **Frequency Capping**, businesses can prevent fraudulent IP addresses or channels that generate an unusually high number of clicks within a customized time frame, as real users are unlikely to perform multiple clicks in such a short period.
[Top 3 Mobile Ad Fraud Tactics in Vietnam’s Fintech Industry (2025)]
Additionally, to detect **SDK Spoofing**, Airbridge provides a **Raw Data Report**, which not only offers comprehensive measurement metrics to help businesses gain a better overview of their campaigns but also allows them to compare **user\_id** or **transaction\_id** to ensure that MMP data matches internal systems like CRM and ERP. The **SDK Signature** feature then helps quickly and effectively eliminate fraudulent MMP cases.
![[Airbridge SDK Signature Dashboard]](https://qhzcaeifhsdzzjpnbdqr.supabase.co/storage/v1/object/public/assets/blog/1774881677579-67ad8509f8d91c9812e2199b_ad_4nxcq-feexf2s6l44nq8aaneaumyfzty7iorjdfjijem4yiu2y2bhlf_irsgvw-f1tjejf5c6p024fext_1schfz0tgvisguxcdqcauqaufnmmclqwmhfvpg9n6jb_jmfmjvdx9v0.png)
For issues related to **Device Farms**, Fintech advertisers should review sub-publishers and ad placements with abnormally low retention, conversion, and engagement rates. Additionally, they should check for outdated devices or small clusters of repeated IP addresses. All these factors are covered under the **Conversion Traffic** and **Retention** features, which help identify suspicious conversions based on multiple criteria such as IP, country, and operating system. As a result, Fintech businesses can avoid paying for fraudulent installs originating from duplicate IPs, non-target countries, or installs with unusually low retention rates.
![[Airbridge Conversion Traffic Dashboard]](https://qhzcaeifhsdzzjpnbdqr.supabase.co/storage/v1/object/public/assets/blog/1774881678277-67ad850942995c077a3779cf_ad_4nxduljb4douejafxeqlq8g3eot3slll8lkhk3abi6zbmm57jpjoxavhe2k5nqt39husbcpu6u7lujse6hwhuteufco9yrcon0xuuxnj-f9qo43voo7bfm8lmlj-78a5hfao7j2es.png)
---
## How to Blow Up Hybrid & Hyper-Casual Games with UA, Monetization, and AI Creatives in 2025
URL: https://www.airbridge.io/en/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives
Language: English
Published: 2025-02-11
## **No-BS gaming podcast where 4 a.m. ideas meet caffeinated insights.**
That's the tagline of [Two and a Half Gamers](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH), and it's precisely what co-hosts [Matej](https://www.linkedin.com/in/matejlancaric/), [Jakub](https://www.linkedin.com/in/jakubremiar/), and [Felix](https://www.linkedin.com/in/felix-braberg-7a732b51/) bring to the table. Their conversations are unfiltered, raw, and an odd mix of rants and revelations, but above all, [brutally honest](https://lancaric.substack.com/). In an era where authenticity reigns supreme, their unique no-holds-barred style draws the mobile gaming community like moths to flame, hungry for their no-BS advice on gaming UA, monetization, and creative optimization.
With over a decade of experience under their belts, they've seen games of all shapes and sizes, from big-budget blockbusters to scrappy indies, casual to hardcore, and everything in between, all trying to make it big on the global stage. And they've been in the trenches with studios, publishers, and developers, helping them succeed.
[__wf_reserved_inherit]
[Roi](https://www.linkedin.com/in/roinam/), CEO of [Airbridge](http://www.Airbridge.io), sat down with Two and a Half Gamers to discuss strategies for topping the charts in 2025 with hybrid and hyper-casual games. Get ready for a takeaway-packed interview. It just might be the spark your game needs to explode onto the scene.
### 🎥 Watch the full interview with Two and a Half Gamers on Youtube
## Mobile Game Growth: No Sign of Slowdown
Despite the chatter about mobile game growth hitting the brakes, especially after the COVID-fueled boom that supercharged hybrid and hyper-casual games, the industry is far from slowing down. As Two and a Half Gamers points out, mobile game growth is nowhere near saturation, with its continued expansion driven by three key factors: speed, monetization, and globalization.
Gaming studios are cranking out more games than ever by perfecting and reusing winning formulas. It's no longer about creating one fantastic game and crossing your fingers for success. Now, it's all about rapid iteration: how quickly you can develop and launch multiple games. As Jakub notes:
> “If you don't follow this model of templatization, you just won't make it. Habby is the best example of this. They've nailed their template: a roguelike, action RPG metagame that they continually refine with different core game styles. It's all the same metagame but with a different core, and this is something you just can't catch up with. They just change the core, change the visuals—boom, a completely new game. But in the end, it's pretty much a perfected version of their previous game. So it's not about relying on just one hit game, but how fast you can develop the next one.”
Companies like MiHoYo (Genshin Impact, Honkai: Star Rail), Peak (Match Factory) , and DeNA (Pokémon Masters EX) are further setting examples of templatized growth.
While app store charts might suggest the mobile gaming industry is slowing down, a massive pool of money isn't being accounted for. Webshops are emerging as another revenue stream for games, separate from traditional app store earnings. Felix points out:
> "I worked with companies where 50% of payments went through webshops instead of app stores. This unaccounted revenue shows mobile games are still growing double digits."
[__wf_reserved_inherit]
Over the years, we've seen a globalization trend in hybrid and hyper-casual games. They're popping up from Eurasia and Asia, particularly from countries like Turkey, Cyprus, Vietnam, China, and South Korea. These developers are creating games with global audiences in mind, and the speed and impact of their growth are aggressive. As Matej notes:
> "We were quite amazed at how far ahead China is. For example, when Lucky Defense launched globally, half of the US or European industry didn't even know about it. But in China, there were already 10 companies working on clones or iterations, with prototypes, just two weeks after its release. It's insane!"
## UA Diversification: Start Now, Scale as You Go
User acquisition (UA) is more challenging than ever, but when the going gets tough, the tough get creative. In today's competitive mobile gaming market, diversifying UA strategies is no longer a choice_—_it's a must. As Matej explains:
> "You just need to start UA on multiple channels from scratch and from day one, basically, which wasn't the case a few years ago."
A strong starting point is to focus on four key platforms - Facebook, TikTok, AppLovin, and Unity - and gradually expand to others like Mistplay and [adjoe](https://www.airbridge.io/en/blog/adjoe-integration). This way you ensure broader audience reach while reducing reliance on a single channel, which can lead to higher CPIs and lower returns. Matej continues:
> "I ran a survey on my subscribers, and the average number of UA channels was between 8 to 12, and the budget went from $100,000 per month to a few million."
UA Diversification goes beyond just choosing platforms_—_it means experimenting with different campaign types. A balanced approach combines high-value strategies, like in-app purchase ROAS campaigns, with cost-effective methods such as app event-optimized campaigns.
Success in today's UA landscape demands early action, cross-channel testing, and rapid data-driven adjustments. Those who can't keep up with the pace risk falling behind.
## Hybrid Monetization: The Sweet Spot for Revenue
Two and a Half Gamers define hybrid monetization as a revenue profile where approximately 40-60% comes from ads and 40-60% from in-app purchases (IAP). Combined, these two revenue streams create a hybrid monetization model. Simply put, it's a model that aims to secure quick ad earnings alongside steady in-app purchases, ensuring both immediate returns and long-term profitability.
To achieve hybrid monetization success, your game needs a robust economy with sufficient "spend depth"—the total potential spending available for dedicated players to maximize their progress or power. This means, building a strong game economy, designing an engaging progression system, and gradually unfolding the game's features. As players advance through their journey, you strategically introduce various monetization elements such as interstitial ads, rewarded videos, and banner ads. Companies like Voodoo and SayGames are great examples of this strategy. Jakub emphasizes:
> “For a game with a longer lifecycle, say 3-4 years, it's crucial to incorporate an IAP economy. This ensures players have ongoing opportunities to spend within the game, supporting its longevity.”
[__wf_reserved_inherit]
But, its success hinges on personalized strategies and sharp user acquisition. As Felix points out:
> “The most important thing to remember is that user acquisition is the tip of the spear—how it trickles down to your game design and your ad segmentation. Without that, you can't have ad monetization or cheap CPI."
With sophisticated segmentation, gaming apps can tailor experiences based on UA campaign goals. Players from IAP-driven campaigns might avoid ads initially, while those from ad ROAS campaigns see banners and interstitials sooner for faster monetization. Attribution tools like [Airbridge](https://www.airbridge.io/) help mobile games implement real-time segmentation, ensuring the right monetization strategy for each user. As Jakub warns:
> "Don't throw away all your monetization into the drain by giving high-value offers to low spenders—they'll actually be repelled by it and won't want to buy."
## AI Ad Creatives: Scaling with Speed and Volume
UA success isn't just about creating great ads—it's about creating a ton of great ads, fast. And a big part of that involves AI-powered ad creatives.
Playables are widely considered one of the most effective ad formats for hybrid and hyper-casual games. The challenge is that they're tricky to create and deploy at scale despite their effectiveness. As Matej highlights:
> "You need playables for all the SDK networks. If you run Facebook and Google, it doesn't matter, but for AppLovin, you absolutely need them. For networks like Mintegral and Moloco as well, it's critical."
But now, developers can simplify the process by starting with game tutorials and scaling efficiently using tools like [playablemaker.com,](http://playablemaker.com) which helps users to create interactive and engaging playable ads from existing video content.
AI tools aren't just handy—they are a necessity for keeping up with the demand for fresh content. Matej explains:
> "On Facebook and TikTok, you need to refresh creatives often—like every 2 or 3 days. And if you don't use AI tools, it's impossible."
The best approach is to strike a balance between quality and quantity. As tools like Midjourney evolve from static to video, and from 2D to 3D, it's becoming easier to churn out story-driven narratives with AI voice-overs. As Matej describes:
> "The sweet spot is you have your own creative team building quality creatives. They build their own concepts, which are different and unique. And then on the side, you play around with the AI tools—balancing the volume play."
With tools like Poolday AI and Runway ML enabling rapid creative production, developers and even UA Managers can efficiently churn out new content. Matej mentions:
> “I use Poolday AI. You can export 1 video with 10 different AI people in the time it'd take 100 people every 5 minutes. This year, the name of the game is the velocity of creatives. How many ad creatives can you pump out? Because that's what the algorithm's hungry for these days.”
[__wf_reserved_inherit]
## The Role of UA Managers: Wearing Multiple Hats
UA managers are no longer just media buyers. They now play a much broader role. Beyond hitting ROAS targets, UA managers run marketability tests, collaborate with product teams for soft launches, and even take on design tasks to create ads. As Matej notes:
> "As a UA manager, you need to wear multiple hats, obviously, and sit on multiple chairs."
Gone are the days of relying solely on CPI tests. Strategies today include testing multiple onboarding versions designed as mini-games to quickly identify the most effective approach. Jakub highlights this trend:
> "Successful games take the first 60-70 minutes of the game, cut it into versions A, B, C, D, and run tests. It's the same game, but they see which onboarding works best."
## Predictive LTV: Making Decisions in Real-time
Predictive lifetime value (pLTV) models have become essential tools for mobile game developers to make faster, data-driven decisions. Matej emphasizes:
> "You need to have the numbers daily and ideally for the cohorts, like in the first 3 days. If you're waiting 7 days to make a decision, that's already too late."
While pLTV is critical for projecting performance and optimization, many mobile game companies struggle with maintenance. Matej points out a common pitfall:
> "Most companies fail because they build the pLTV model, and that's it. They don't refresh it or look at historical data, and then it becomes a mess. Recalibrating the model every 3 months is a must to ensure it maintains accuracy."
As difficult as maintaining in-house pLTV models is, many mobile game developers turn to solutions like [Airbridge](https://www.airbridge.io/product/airbridge-ai) to provide pLTV, cohort analysis, and minutely/hourly retention analysis.
## Localization: Adapting Games for Global Audiences
Making it big on the global stage goes beyond simple translation—it's about localization, tailoring games to regional preferences. As Jakub notes:
> "Feature sets are just much more robust in Asia—3 times more robust—because people in Asia are just much more feature-hungry."
Asian audiences favor complex games with layered progression and multiple systems. On the contrary, these feature-heavy designs can overwhelm Western players. To succeed in the West, games must be simplified. Jakub continues:
> "When you're trying to move your game to the West, make it more streamlined, more approachable, more easier. For instance, if we take, let's say, Rush Royale by MY.GAMES, it's just a streamlined version of Random Dice."
## Work Smarter, Not Harder
Insights from Two and a Half Gamers highlight a clear path forward in mobile gaming: adapt fast or get left behind. Success in 2025 isn't just about having great games—it's about mastering the trifecta UA diversification, hybrid monetization, and AI-powered creative optimization. As Matej puts it:
> "Success comes from balancing strong UA, innovative monetization strategies, and scalable ad creatives. It's about working smarter, not just harder."
[__wf_reserved_inherit]
For more insights, visit the [Two and a Half Gamers podcast](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH) and [airbridge.io](http://airbridge.io).
### **Supercharge Your Growth with Airbridge**
If you're ready to take your game marketing to the next level, check out [**Airbridge**](https://www.airbridge.io/). Our platform provides:
- **Advanced Cohort Analysis for LTV:** Track lifetime value with precision and break down performance at a granular level (keyword, creative, country, OS, and more).
- **Ad Mediation Integration:** Seamlessly combine data from multiple ad networks in one place (e.g., Google AdMob, AppLovin MAX, Unity LevelPlay).
- **Predictive LTV (pLTV) Feature:** Forecast future revenue in real-time, allowing for faster, data-driven decisions.
- **Minutely/Hourly Retention Analysis:** Understand player behavior down to the minute or hour for truly actionable insights.
- **Competitive, All-Inclusive Pricing:** Access enterprise-grade analytics without hidden fees or complicated add-ons.
Whether you're an indie studio or an established publisher, Airbridge gives you the critical tools you need to stay ahead in today's fast-paced market. Visit [**airbridge.io**](https://airbridge.io) to learn more.
### **Level Up Your Expertise with Two and a Half Gamers**
For ongoing strategies, case studies, and unfiltered discussions about user acquisition, monetization, and creative optimization:
1. **Subscribe** to the [**Two and a Half Gamers YouTube Channel**](https://www.youtube.com/@2.5gamers) for in-depth tutorials, expert interviews, and real-world examples.
2. **Tune in** to the [**Two and a Half Gamers Podcast**](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH) for industry insights you can't afford to miss.
3. **Listen** to the [**Two and a Half Gamers Album**](https://open.spotify.com/artist/2o80XDfSqkDIqmuc5tOZdt?si=67Hn_PzAS_Sw6w-QS-XWRw) for catchy tunes on how to drive game growth.
4. **Connect** with [**Matej**](https://www.linkedin.com/in/matejlancaric/), [**Jazkub**](https://www.linkedin.com/in/jakubremiar/), and [**Felix**](https://www.linkedin.com/in/felix-braberg-7a732b51/) on **LinkedIn** to dive deeper into your specific growth challenges.
Stay informed, adapt quickly, and get ready to blow up your hybrid and hyper-casual games in 2025!
---
## Unlock new growth channels with Airbridge’s integration with Microsoft Advertising
URL: https://www.airbridge.io/en/blog/unlock-new-growth-channels-with-airbridge-integration-with-microsoft-advertising
Language: English
Published: 2025-02-10
Airbridge is now integrated with Microsoft Advertising, combining advanced measurement capabilities with the broad reach of Microsoft Advertising to help you better identify your top-performing campaigns.
## Reach more audiences with Microsoft Advertising
Microsoft Advertising provides broad reach with accurate targeting options. Discover and connect with your ideal audience by placing your ads across several owned, operated, and partner platforms, including Bing, MSN, Microsoft Edge, Outlook.com, AOL, Yahoo, DuckDuckGo, and more.
## Streamline measurement with Airbridge
Airbridge's integration with Microsoft Advertising simplifies measurement for mobile app install campaigns, providing marketers with streamlined tracking and clear performance visibility for better optimization. Through this powerful integration, you can:
- Accurately track installs from app extensions and app install ads as conversions.
- Identify top-performing campaigns, channels, and keywords with precise attribution.
- Pinpoint audiences with higher conversions.
- See all your ad spend and ROAS in one intuitive comparison view.
- Get the full picture of your campaign delivery in real-time across multiple channels from one single dashboard.
## Getting Started
Setting up the integration between Airbridge and Microsoft Advertising is easy. Follow these steps:
1. Create tracking links in Airbridge for your Microsoft Advertising campaigns.
2. Configure postbacks for real-time data exchange based on your conversion goals.
3. Use Airbridge analytics to gain insights into user behavior and campaign performance.
4. Adjust your strategies based on the data collected.
For detailed instructions, refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/microsoft-advertising). Contact the Airbridge team today to learn how Airbridge and Microsoft Advertising can enhance your mobile app install campaigns.
---
## What is Affiliate Marketing? Winning Strategies for 2025
URL: https://www.airbridge.io/en/blog/what-is-affiliate-marketing
Language: English
Published: 2025-02-03
## **What Is Affiliate Marketing and How Is It Reshaping Marketing Trends in 2025?**
Have you ever seen phrases like "Refer a Friend, Get a Discount" on a shopping app? Or come across a link to a product recommended by an influencer on YouTube or Instagram?
These are what ‘affiliate marketing’ is about: selling products through affiliate links. Affiliate marketing is emerging as a key trend in the competitive mobile marketing landscape, gaining recognition as a strategic method to boost ROI (Return on Investment).
In this post, we will explore the basic concepts of affiliate marketing, explain why you must get involved in 2025, and provide tips on how to get started.
## **Understanding Affiliate Marketing**
> 💡 What is Affiliate Marketing?
> Affiliate marketing is a performance-based marketing strategy where businesses compensate third parties (affiliates) for generating traffic, leads, or sales for their products or services.
Affiliates (companies, individuals, and influencers) earn revenue and commissions for successful referrals through unique tracking links.
In turn, advertisers pay partners for results, making affiliate marketing cheaper and more effective compared to traditional ads. Notable examples of affiliate marketing platforms include [Amazon Associates](https://affiliate-program.amazon.com/), [Walmart Affiliate Program](https://creator.walmart.com/?ebebhghdefhbdjldreibbfljnbirebktvvuvighttigg&gad_source=1&gclid=Cj0KCQiAq-u9BhCjARIsANLj-s0MT8Ec7Vze-Ij9f4AjbCnEzffcweOmWvfgghkcQ87xnuu27WtXQD0aAkAdEALw_wcB&gclsrc=aw.ds), and [Rakuten Advertising.](https://rakutenadvertising.com/affiliate/)
### **Examples of Affiliate Marketing**
**Case A) Affiliate earns commission**
A YouTuber posts a video reviewing a product and attaches a link for viewers to purchase them. When viewers buy the product through this link, the YouTuber earns a commission.
**Case B) User earns rewards**
A shopping app is running an “Refer a Friend” campaign. Users get rewards when their friends use the unique referral link to install the app or make a first purchase .
**Benefits of Affiliate Marketing**
Affiliate marketing is rapidly emerging as a powerful trend in mobile marketing, offering several key advantages:
1. Efficient Cost Management
Pay for advertising only when conversions (purchase, app installs, etc.) occur, minimizing budget waste.
2. Diverse Channel Utilization
Implement campaigns across blogs, social media (Youtube, Instagram, Tiktok, etc.), apps, and more.
3. Rapid Scalability
Quickly expand marketing reach by increasing the number of partnered affiliates.
4. Maximized ROI
Manage budgets effectively through accurate performance measurement.
## **So, where do we start?**
1. **Find the ideal partner and develop the optimal model to achieve your goals**
Successful affiliate marketing begins with identifying the right target audience and selecting an appropriate affiliate partner. Look for a partner that aligns with your specific goals, such as driving sales for particular products, increasing category performance, boosting membership numbers, and enhancing subscription conversion rates.
2. **Design a performance-based compensation system**
Establish a clear and fair compensation standard based on the results. You can choose from examples such as CPC (cost per click), CPA (cost per acquisition), or CPS (cost per sale).
3. **Create and distribute affiliate links**
Once you determine the goal, partner, and compensation system, create an affiliate link to help the affiliate partner promote it. Creating a unique tracking link is crucial; if the affiliate link doesn't work properly, you won't be able to measure the conversions generated through it.
4. **Monitor performance using performance tracking tools**
Once you launch your affiliate marketing campaign, track performance closely to ensure accurate compensation. Use performance tracking tools to monitor the number of conversions occurring through partner links and compensate your partners accordingly.
**Start with Airbridge today!
**Streamline your affiliate campaigns with Airbridge. The Airbridge DeepLink Plan helps you optimize affiliate marketing by efficiently creating and managing affiliate links. We provide seamless user experiences through deep linking, ensuring both users and affiliates enjoy a frictionless interaction. With Airbridge DeepLinks, you can track affiliate performance in real-time, ensuring accurate commission distribution and maximizing campaign ROI.
- Auto-generated API-based tracking links
- Universal compatibility across all systems and platforms—iOS, Android, web, and offline.
- Razor-sharp performance tracking and real-time analytics
- Competitive MAU-based pricing (pay as you grow)
---
## What is Deep linking and How Does It Work? URI Scheme, Universal Link, App Link
URL: https://www.airbridge.io/en/blog/what-is-deep-link
Language: English
Published: 2025-01-20
# What are deep links and why are they so important?
> **💡 What are Deep Links?**
> Deep links direct users to specific locations within apps. When a mobile user clicks a link to a website or another app, Deep links redirect them straight to particular pages within the target app. In essence, Deep Links enable users to access desired in-app pages directly, bypassing the usual navigation process.
Deep links power a wide range of app marketing activities, from CRM campaigns to ad initiatives. Marketers use them to enhance push notifications, in-app messages, texts, social sharing, referrals, affiliate programs and more. Through deep links, marketers can improve **user experience** and **boost conversion rates.**
In this post, we’ll explore the different types of deep links and how to power your deep linking strategies with Airbridge.
## **The Different types of Deep Links**
The three most commonly used mobile deep link standards are URI scheme, Universal Link on iOS, and App Link on Android.
### URI Scheme: The most basic deep link
For example, to open the Airbridge app, you can simply call the pre-assigned scheme value: ‘**airbridge://**’.
When using a URI scheme to open an app, you can also specify which page or section of the app should open. This is done by adding a path after the scheme, similar to how website URLs work. The basic structure is: '**scheme://path**'.
For example, if you want to open the Airbridge app and go directly to a specific product in their stock list, the URI scheme might look like this:
➡️ **airbridge://stocklist/8516**
The URI scheme is intuitive and easy to implement, but the drawback is that multiple apps can use the same scheme. This can lead to app open errors(the wrong app might open, or no app might open at all)or security issues. Also, URI scheme links often fail to direct users to the app store when they don't have the app installed.
In brief, the URI scheme:
- 👍 Pros: Easy to implement and quick to set up.
- 👎 Cons: Multiple apps can use the same scheme causing open errors and security issues. Might fail to direct users to app store for users without the app installed.
To make up for such downsides, iOS and Android each came up with their own deep linking implementation methods. These are **iOS's Universal Links** and **Android's App Links**.
### Universal Link & App Link: Deep linking on iOS and Android
To overcome the drawbacks of URI schemes, namely scheme value duplication and security issues, iOS and Android introduced their own deep linking methods. iOS created [Universal Links](https://developer.apple.com/documentation/xcode/allowing-apps-and-websites-to-link-to-your-content/) and Android developed [App Links](https://developer.android.com/studio/write/app-link-indexing?hl=ko). These links support seamless opening on their respective operating systems and began to replace URI schemes.
However, Universal Links and App Links work properly only on their respective operating systems, not interchangeably. For example, an iOS user who receives a push notification with an Android App Link may encounter issues when trying to open the app.
Simply put, Universal Links and App Links:
- Pros 👍: Eliminate duplicate scheme issues. Open smoothly on their native OS.
- Cons👎: Don't open properly on other OS.
Universal Links and App Links can direct users without the app to the app store. However, they lack user-friendly features like deferred deep links, which allow users to access specific content directly after installation.
To solve these limitations and different implementation methods between Universal Links and App Links, many app developers now choose dedicated deep linking solutions. These deep linking solutions provide cross-platform compatibility and support deferred deep links, extending their functionality beyond operating system limitations. [Google's Firebase Dynamic Links](https://firebase.google.com/docs/dynamic-links?hl=ko) and [Airbridge DeepLink Plan](https://www.airbridge.io/ko/deeplink-plan) are examples of this.
[__wf_reserved_inherit]
# **Airbridge Deep Link Plan: Many Deep Links, One Platform**
Deep links act as gateways to apps, and managing them effectively can save you and your users from a lot of headaches. The **Airbridge DeepLink Plan** fully supports URI schemes, Universal Links, and App Links, providing users with a seamless deep linking experience while saving time and effort for developers and marketers.****
### **Why AirBridge DeepLinks?**
1. **Effortless migration**
Using Firebase Dynamic Links or other deep linking solutions? Don’t worry! Migrate active deep links to Airbridge without code or dev support.
2. **Easy link generation & management**
Airbridge DeepLinks boast universal compatibility on both iOS and Android.
3. **Deferred deep linking**
Direct users straight to their [desired in-app destination](https://www.airbridge.io/en/blog/deeplink-101-deferred-deep-linking)—even if they haven't installed your app yet.
4. **Custom QR & link branding**
No need to use other solutions for [QR codes](https://www.airbridge.io/en/blog/airbridge-qr-code), thumbnails or custom domains. The Airbridge DeepLink Plan includes all of the above.
5. **Optimized for referral and affiliate marketing
**Airbridge DeepLinks are tailored for marketing based on link sharing, like referrals and affiliate marketing.
6. **Accurate, real-time data analysis
**From deep link clicks to conversions, analyze your data in real time with powerful filter options based on your specific needs.
7. **Competitive pricing
**With the impending shutdown of Firebase Dynamic Links, demand for alternative deep linking solutions is on the rise. The Airbridge DeepLink Plan is an excellent solution with pricing based on Monthly Active Users (MAU), so you can pay as you grow.
## **Join Airbridge Today!**
2025's competitive mobile market demands better connections for users and streamlined workflows for marketers. Enjoy seamless deep linking and analytics at the best value with Airbridge.
---
## Airbridge Year in Review 2024
URL: https://www.airbridge.io/en/blog/airbridge-recap-2024
Language: English
Published: 2025-01-14
As the digital landscape continues to evolve at breakneck speed, 2024 was a year of unprecedented growth, strategic partnerships, and groundbreaking innovations for Airbridge. We didn't just adapt – we led, pushing the boundaries of mobile measurement and cross-platform analytics.
Let's dive into the numbers and milestones that defined our success.
[__wf_reserved_inherit]
## 4.3x Increase in Gaming Customers
2024 marked Airbridge's rise as a powerhouse in the gaming industry, with our gaming customer base growing an impressive 4.3 times compared to the previous year. From hyper-casual, RPG, racing, and AR music to interactive gamified video platforms across genres, developers and studios chose Airbridge as their trusted MMP partner.
### Tailored Solutions for Game Growth
Our milestone in the gaming sector stemmed from Airbridge's unmatched performance measurement capabilities. Gaming studios often face challenges in measuring and analyzing performance for campaigns that span multiple channels, operating systems, countries, and audiences. We saw the problem, and we delivered. With our suite of features tailored specifically for game growth, Airbridge enabled ultra-granular real-time data analysis, empowering developers and studios to gain actionable insights to optimize their campaigns.
[Webzen](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization) leveraged Airbridge to optimize their short-form ad campaigns at the creative level. The team also tapped into our Predictive Lifetime Value (pLTV) feature to proactively identify channels with the highest expected revenue per user, enabling efficient marketing spend optimization.
With the growing interest in PC and console games, we stepped up our game too. Airbridge provided cross-platform play analytics to measure and analyze performance across all gameplay channels - including apps, PC, and consoles
[Nexon](https://www.airbridge.io/en/case-studies/nexon-cross-platform-tracking) utilized our platform to measure the ad performance for its PC and console game "[The First Descendant](https://www.playstation.com/en-us/games/the-first-descendant/)," obtaining the same reliable data metrics for PC and console as for mobile games.
### Expanding Our Gaming Footprint
2024 saw Airbridge making waves at major gaming conferences worldwide. We showcased our cutting-edge features and solutions at industry-leading events, including G-Star, Vietnam Games Forum, ChinaJoy, GDC (Game Developers Conference), and Gamescom. Our reach extended beyond conferences as we forged strategic partnerships with over 70 media outlets and solution providers.
Our reach extended beyond conferences as we forged strategic partnerships with over 70 media outlets and solution providers. Notable integrations included Mintegral tROAS, AVOW, ayeT, and adjoe. We also shared our expertise on Korea's gaming market by co-publishing the "[Mobile Gamers of Korea Report 2025](https://adjoe.io/mobile-gamers-of-korea/)" with adjoe, drawing insights from a decade of comprehensive industry data.
[__wf_reserved_inherit]
## 90% Boost in Apps Using Airbridge
Last year, we saw a 90% increase in apps tracking performance with Airbridge compared to 2023, demonstrating growing trust in our ability to provide accurate insights and optimize campaigns.
### Enhanced Integrations and Partnerships
We continued to strengthen our ecosystem by deepening integrations with leading martech solutions like Amplitude and Braze and expanding reach through global ad platforms like Microsoft Ads and Digital Turbine, empowering our customers to run sharper, more far-reaching campaigns.
Through the Airbridge-Braze integration, MyRealTrip increased its conversion rate by 12x, and [Greenlabs](https://www.airbridge.io/en/case-studies/greenlabs-crm-data-integration) doubled its reactivation campaign open rates using personalized messaging powered by Airbridge's precise attribution analysis.
With Airbridge's ability to integrate with diverse platforms, we continue to serve as a key component for full-funnel marketing.
### Unmatched Customer Support
Our customer support team continued to drive growth for customers through tailored strategies. Webzen saw a 90% increase in player sign-in rates for "[Mu Monarch](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization)" by combining Airbridge's measurement capabilities with our dedicated support. [Vortex Gravity](https://www.airbridge.io/en/case-studies/vortexgravity-cut-mmp-cost) achieved dual benefits by switching to Airbridge. Not only did they reduce their annual MMP costs by 50%, but they also received strategic and responsive support through their preferred channel, WeChat.
[__wf_reserved_inherit]
## 14.15M Reports Generated
In 2024, our customers generated over 14 million reports using Airbridge's 10 different report types. These reports provided accurate, systematic app growth insights, enabling real-time data analysis and more efficient campaign optimization. [Shmoody](https://www.airbridge.io/en/case-studies/shmoody-cross-platform-tracking) leveraged our customizable dashboards to unify Android, iOS, and SKAN performance data in one view, streamlining their analysis and accelerating their path to 1M downloads. [Buddy.ai](http://buddy.ai) utilized Airbridge's cost integrations and ROAS metrics to secure Top 10 App Store success, demonstrating the power of our comprehensive analytics in driving app performance. With Airbridge's overview dashboard, [Segari](https://www.airbridge.io/en/case-studies/segari-overview-dashboard) cut its data compilation and analysis time by 50%.
## 86.13M Tracking Links Created
Airbridge users generated more than 86 million tracking links in 2024, resulting in a staggering 2.628 trillion clicks throughout the year. That's an average of 20 billion clicks per month, double the volume of free URL shortener [Bitly](https://bitly.com/pages/resources/press/bitly-wraps-2022-surpassing-100m-in-arr-and-over-500k-global-customers).
Our tracking links go beyond simple generation, offering deep linking capabilities and seamless navigation across all platforms and operating systems. The launch of [Airbridge DeepLink Plan](https://www.airbridge.io/deeplink-plan) in Q4 garnered significant attention, especially with Firebase Dynamic Links set to shut down by August 25, 2025.
## Looking Ahead to 2025: AI-Powered Innovation
This year, Airbridge is introducing AI-powered features to deliver more robust measurements. Our enhanced AI capabilities will forecast 180-day customer lifetime value using just three days of data, detect abnormal installs, and simulate iOS attribution through UCA (Unattributed Conversion Analysis), providing smarter, faster, and even more precise insights to drive your app business growth.
Thank you for being part of the Airbridge journey. Here's to creating even greater success stories in the coming year. Upwards and onwards to an even more remarkable 2025!
---
## Mastering Your App Store Optimization (ASO) Strategy
URL: https://www.airbridge.io/en/blog/aso-strategy
Language: English
Published: 2024-12-20
As iOS expands its market share, [App Store Optimization (ASO)](https://www.airbridge.io/en/blog/aso-success-guide) is no longer just about visibility—it’s about survival. With millions of apps competing for attention, ranking high in search results requires more than just an optimized title, description, and visuals.
While the above mentioned ASO tactics are free, what once provided an edge is now the bare minimum.
So, where does the real competitive advantage lie? Let’s break it down.
## **ASO: No Guarantees, Just Smarter Choices**
There’s no perfect formula for App Store Optimization. No magic keywords, no guaranteed rankings. And with AI-driven algorithms constantly reshuffling search results, predicting success is tougher than ever.
But here’s what we do know: the right moves stack the odds in your favor. Marketers have tested, tweaked, and uncovered what works.
- **Keywords matter**—use them strategically in your title and description
- **Categories count**—prioritize the most relevant first and second
- **Visuals sell**—screenshots should be clear and compelling
- **Go local while you go global**—localize titles, descriptions, and app info
## **ASO Checklists**
ASO isn’t a one-time checklist—it’s a constant hustle of trial and error. With Apple’s algorithms constantly shifting, your app’s visibility and performance demand ongoing attention.
Here’s a quick rundown of tactics that can improve your app’s chances:
- **Monitor ratings and reviews** to stay on top of user feedback
- **Track high-converting keywords** by finding out what your audience is searching for and optimize your title, description, and metadata accordingly
- **Leverage Custom Product Pages (CPP)** to tailor your app’s experience for different user segments. You can create specific pages to test different messaging, promotions, or even different features for targeted groups—giving you a direct path to higher engagement and more relevant installs
For even better results, also analyze competitor-favored keywords, incorporate marketing trends, and stay on top of Apple App Store policies.
💡Want a deeper dive? Explore the [Airbridge Handbook 03: ASO](https://www.airbridge.io/library/airbridge-handbook-aso)
While these tactics are largely free and easy to implement, don’t expect miracles. ASO is an ongoing, iterative process.
## **Apple Search Ads: The ASO Supercharger**
ASO can be hit or miss. You optimize, but still struggle with visibility—because the App Store algorithm is a mystery. So, what’s the solution? Apple Search Ads.
Apple Search Ads enhances your app's visibility by securing targeted keywords, which in turn boosts your ASO performance. By appearing at the top of search results for user queries, Apple Search Ads amplifies your ASO efforts.
Let’s break it down with a real example: A U.S. sports broadcaster outperformed its competitor by securing crucial keywords via Apple Search Ads, resulting in more prominent exposures. Notably, the company maintained its ASO efforts in line with Apple Search Ads and even adjusted some of its marketing plans to create synergy between these two tactics—ASO enhances the effectiveness of Apple Search Ads, while Apple Search Ads, in turn, bolster the ASO strategy.
- Synchronized keywords for Apple Search Ads and ASO. Both used the same high-converting keywords, ensuring consistency
- Leveraged CPP to further improve conversion rate for users acquired through strategic keywords
- Maintained ongoing ASO efforts by keeping their app name, description, and visuals optimized constantly
## **Airbridge: Fully Equipped for ASO Excellence**
Boosting your ASO performance with Apple Search Ads isn’t a one-man job. You’ve got to juggle **budget management**, **keyword optimization**, and **ad performance monitoring**, all while staying on top of your ASO strategy. It’s a lot. But with the right tools, it doesn’t have to be hard.
Airbridge, with seamless Apple Search Ads integration, provides all-inclusive reports to help keep tabs on your ad performance and expenses daily. These reports offer instant access to a wide range of key metrics, making it a breeze to manage multiple ad channels all at once.
- [Integration with Apple Search Ads](https://help.airbridge.io/en/guides/apple-search-ads)
- [Get Started with Airbridge Reports](https://help.airbridge.io/en/guides/airbridge-report-overview)
Ready to elevate your Apple Search Ads strategy with an MMP? Don’t settle for less. Choose the MMP that does more. Choose Airbridge.
---
## 5 types of teams that need Airbridge DeepLinks
URL: https://www.airbridge.io/en/blog/5-team-who-need-airbridge-deeplink-plan
Language: English
Published: 2024-11-26
Deep links are powerful tools for app marketing, guiding users to specific destinations within an app. They're crucial for CRM, advertising, referrals, and affiliate marketing. However, building and managing deep links can be quite a challenge.
Google's announcement to [discontinue Firebase Dynamic Links](https://firebase.google.com/support/dynamic-links-faq) has left many apps at a crossroads. Apps that relied on Dynamic Links must now choose between building their own deep links or switching to a new service provider.
**The Airbridge DeepLink Plan is a straightforward alternative to Firebase Dynamic Links.** In this blog, we'll explore 5 types of teams that can benefit most from Airbridge DeepLinks. If your app business falls into one of these categories, it might be time to see how Airbridge can help.
**💡 Want to know more?
**Explore the [Airbridge Deeplink Plan](https://www.airbridge.io/deeplink-plan)
## **1\. 🚚 Migration | Teams using Firebase Dynamic Links**
On August 25th, 2025, Firebase Dynamic Links will shut down. All links served by Firebase Dynamic Links (both hosted on custom domains and page.link subdomains) will stop working and you will no longer be able to create new links or migrate old ones.
If your team relies on Dynamic Links, you must switch to a new deep linking service before the deadline. To make the transition easy and stress-free, it's important to find a solution that simplifies deep link migration.
The Airbridge DeepLink Plan offers no-code migration in just 3 steps. It keeps all deep linking features of Dynamic Links, including deferred deep linking and custom branded links, while adding benefits such as:
- Optimized user experiences across iOS and Android
- Auto-generated QR codes
- Added custom branding features
With Airbridge DeepLinks, you get everything you enjoyed with Dynamic Links — and more.
## **2\. 💻 Development | Teams struggling to deep links build in-house**
Building and maintaining deep links in-house is a complex and resource-intensive process. It requires developers to build the basic linking system, incorporate advanced features like dynamic and deferred deep linking, and continuously update the links to work across different operating systems, browsers, and apps. This ongoing maintenance is particularly challenging, as frequent platform updates require constant attention from development teams to ensure the links function consistently.
Airbridge DeepLinks streamlines this process with ready-to-use expertise:
- Ensuring deep links work seamlessly on all platforms
- Supporting seamless dynamic and deferred deep linking
- Freeing your team to focus on core development tasks
Learn more about deep links in Airbridge's [**Deep Link Handbook**.](https://www.airbridge.io/library/airbridge-handbook-deeplinks)
## **3\. 🔗 User Experience | Teams seeking better deep link performance**
A seamless user experience is crucial for app success. When users tap a link and land in the wrong place or, even worse, the app fails to open; it often leads to broken user journeys and potential app abandonment.
Flawless deep link performance is critical for apps that rely heavily on ads or shared links. Airbridge DeepLinks works smoothly across all platforms and channels, providing users a consistent and seamless experience.
## **4\. 📆 App Launch | Teams launching new apps**
Launching a new app is exciting, but it comes with many decisions, including how to handle deep linking. With Firebase Dynamic Links deprecated, Google no longer recommends it for new apps. Starting with a dedicated deep linking service from the beginning can save you time and complications down the road.
The **Airbridge Deeplink Plan** offers flexible, MAU-based pricing, making it an excellent choice for new apps across various budgets.
## **5\. 💵 Pricing | Teams seeking an affordable deep linking solution**
When choosing a new deep linking service, price matters as much as features. The Airbridge Deeplink Plan offers competitive, MAU-based pricing that adapts to your app's growth. Our flexible pricing ensures you only pay for what you need.
Visit the Airbridge [**Pricing Page**](https://www.airbridge.io/deeplink-plan) to find the perfect plan for your budget and requirements.
—
Whether your app business falls into the above categories or has unique deep linking needs, Airbridge can help. Contact us to see how Airbridge DeepLinks can support your app's success.
---
## Unlock Game Growth with Airbridge + Hive Integration
URL: https://www.airbridge.io/en/blog/hive-platform-integration
Language: English
Published: 2024-11-26
We’re excited to announce the integration of Airbridge with Hive Platform, streamlining marketing measurement and optimization for game developers. This collaboration enables seamless integration of the Airbridge SDK, helping you measure, analyze, and improve campaigns effortlessly across multiple channels, creatives, and regions.
## **What the Integration Brings**
1. **One-Click Airbridge Integration**
- Hive SDK v4.24.4.0+ (Mobile) or v24.4.0+ (PC) now includes Airbridge SDK integration directly from the Hive dashboard.
- Simply update to the latest SDK to activate Airbridge in seconds—no separate installation required.
2. **Coming soon: Two-Way Data Sharing**
- Airbridge attribution data will soon sync back to Hive, unlocking real-time insights for smarter optimization across acquisition, engagement, and monetization strategies.
## **About Hive Platform**
[Hive Platform](https://hiveplatform.ai/) by Com2uS is a trusted, full-cycle game development solution used by industry leaders like Gravity, StoneAge, Mistgames, and Blue Potion. Key features include:
- **Developer SDK**: Rapid game development for Unity and Unreal Engine.
- **LiveOps Portal**: Real-time game operation and management functions, including event promotions and user engagement tools.
- **Game Backend**: Delivers backend services such as authentication, billing, push notifications, and blockchain integration.
- **Ad Monetization**: Enables effective ad integration without disrupting the gaming experience, supporting revenue generation through various ad formats.
- **Data & Analytics**: Offers game-specific big data analysis services, providing insights into player behavior and game performance.
## **Conclusion**
This integration brings together Hive’s robust game development tools with Airbridge’s cutting-edge measurement analytics. Together, we empower developers to measure accurately, streamline operations, boost user acquisition, and maximize engagement and monetization strategies—all in one place.
## **About Airbridge**
****Airbridge is a global Mobile Measurement Partner (MMP) providing real-time, cross-platform insights across mobile, web, PC, and console with over 300 metrics for user acquisition, engagement, and monetization. Trusted by top game publishers like Nexon and Webzen, Airbridge offers essential tools, including cohort analysis, fraud protection, deep linking, unlimited raw data export, and predictive LTV insights—all accessible through unified dashboards and backed by exceptional support.
[👉 Click here to request demo](https://abr.ge/@ab180themodel/blog?ad_group=crm&campaign=ab180-hive-integration-en&content=cta-request-a-demo&og_tag_id=116456123&routing_short_id=hstryc&sub_id=ab180&tracking_template_id=717340950e7258fa1c9c0171ed9f2fe0&ad_type=click)
---
## Boost your game's reach with Airbridge & adjoe integration
URL: https://www.airbridge.io/en/blog/adjoe-integration
Language: English
Published: 2024-10-25
With Airbridge’s integration with [adjoe](https://adjoe.io/), you can accurately measure and optimize your adjoe campaign performance to elevate your game. This partnership combines adjoe's gaming-focused ad formats with Airbridge's robust analytics, equipping you to enhance your game’s UA and monetization strategies.
## Tap into adjoe’s core gaming audiences
Place your ads directly in front of over 200 million users eager to engage with new, exciting mobile games. This integration gives you direct access to adjoe’s highly engaged gaming audience through a variety of ad formats, including
1. [**Playtime Rewarded Ads**](https://adjoe.io/solutions/advertise-your-app/playtime/)**:** Players are continuously rewarded for engaging with your game during their gameplay sessions.
2. **Rewarded Video:** Players can watch video ads in exchange for in-game rewards, integrating ads into the gaming experience.
3. **Interstitial:** Full-screen ads that appear during natural breaks in gameplay, maximizing visibility without disrupting play.
4. **Banner:** Display ads placed at the top or bottom of the screen during gameplay, providing consistent brand exposure.****
5. **MREC:** Ads that blend seamlessly with the game's content, enhancing user experience while promoting other apps.
[__wf_reserved_inherit]
## Power your game promotion with Airbridge
By combining Airbridge's analytics with adjoe's ad solutions, you gain the tools to:
- **Identify the most effective ad formats and campaigns** through detailed attribution data.
- **Understand long-term user behavior** based on ad exposure with cohort analysis.
- **Optimize your ad placement and messaging** using A/B testing.
- **Make quick campaign adjustments** with real-time reporting.
- **Gain a holistic view of your marketing performance** by integrating data across multiple channels.
With Airbridge & adjoe integration it’s easier to boost your game's visibility, increase player acquisition, and drive long-term engagement helping your game thrive in the ever-competitive mobile gaming market.
## Getting started
Implementing the Airbridge & adjoe integration is simple. Follow these steps to get started:
1. Set up tracking links in Airbridge for your adjoe campaigns.
2. Configure postbacks to ensure real-time data flows between the two platforms.
3. Use Airbridge's analytics to gain insights into user behavior and campaign performance.
4. Continuously refine your strategies based on the data you collect.
For detailed instructions, refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/adjoe). To learn how Airbridge and adjoe can enhance your game’s growth, reach out to the [Airbridge team](https://abr.ge/@ab180themodel/blog?ad_group=integration+&campaign=adjoe-integration&content=cta1-book-a-demo&og_tag_id=103200583&routing_short_id=yl4rer&tracking_template_id=2c12f3db8a77dd4d4d9f37d76476ea29&ad_type=click) today!
---
## No-Code Deep Link Migration From Firebase to Airbridge
URL: https://www.airbridge.io/en/blog/how-to-migrate-deeplink-in-airbridge
Language: English
Published: 2024-10-25
Google will shut down Firebase Dynamic Links on August 25, 2025. After this date, your existing links will break, and migration will be impossible. You have to migrate to a new deep linking solution fast before the deadline.
When switching to a new deep linking solution, migrating your existing links is a big part of the process.
With Airbridge, migrating your Firebase Dynamic Links is easy and code-free. You can transfer all your links quickly and efficiently using **Airbridge's Tracking Link Migrator and Generator.
**
Here's how you can migrate your Dynamic Links to Airbridge in just 3 Steps!
## **Step 1: Export Firebases Dynamic Links**
Export your Dynamic links from Firebase using the Firebase Takeout feature, which provides a CSV file of all your active links.
## **Step 2: Map parameters to Airbridge Tracking Links**
Paste the exported links into the **Dynamic Link Input** tab on **Airbridge's Tracking Link Migrator**. Then, copy the converted parameter values from the **Parameter Mapping Results** tab.
## **Step 3: Generate Airbridge DeepLinks**
Paste the converted parameters into the **Link Generation** tab on **Airbridge's Tracking Link Generator**. Click "Generate Tracking Link" to create your new Airbridge DeepLinks. The new links will appear in the sheet and on your Airbridge dashboard.
## **Go live with Airbridge DeepLinks!**
Replace the old Dynamic Links with the new and improved Airbridge DeepLinks, and you're done! Migration complete.
Airbridge makes migrating your links fast and easy, no matter how many you have. Want to see Airbridge in action? Contact Airbridge today for a personalized demo. Our experts will help you ensure a smooth transition.
---
## Boost your omnichannel strategy with Conversion APIs
URL: https://www.airbridge.io/en/blog/capi-for-omni
Language: English
Published: 2024-10-24
Privacy regulations are reshaping digital advertising. [Google's Privacy Sandbox will soon let Chrome users block third-party cookies](https://privacysandbox.com/news/privacy-sandbox-update/), similar to Apple's App Tracking Transparency (ATT). Considering that in 2023, a mere 30% of iOS users opted to share their data, you can anticipate a substantial decrease in cookie data for personalization.
As privacy regulations become increasingly strict, Conversion APIs become more crucial. These APIs give marketers a comprehensive picture of their audience to help them create effective, targeted campaigns across online and offline platforms.
## So, what exactly are Conversion APIs?
Conversion APIs, CAPIs, or server-side APIs send first-party event data directly from your server to third-party advertising platforms. Since they work on the server side, they gather all the data from the internal servers running your website or app. This means you can skip over client-side headaches like web pixels, browser settings, and cookie restrictions.
Conversion APIs can transmit various types of first-party data, including user behavior and conversion events—such as purchases and subscriptions—from web, app, and offline sources. **This makes them especially attractive in light of growing privacy restrictions in the advertising space.** These APIs offer a way to maintain data accuracy and campaign effectiveness despite the limitations imposed by new privacy regulations.
Recognizing this shift, major advertising platforms such as Meta, Snapchat, TikTok, and X have developed and launched their own Conversion API solutions.
[__wf_reserved_inherit]
## Why are Conversion APIs essential for your omnichannel strategy?
Conversion APIs work in tandem with standard tracking methods to provide marketers with real-time, reliable conversion and user data, enabling data-driven decisions and optimized ad campaigns.
With conversion APIs, you can maximize the use of your:
- **First-party data:
**Unlike pixels, which send data through a user's browser and can be blocked, or postbacks, which rely on Advertising IDs (ADID) that are becoming less reliable due to privacy changes, Conversion APIs collect information directly from your servers. They gather valuable data points like click IDs, hashed emails, and hashed phone numbers, offering a broader range of parameters for better data matching and accurate measurement.
- **Omnichannel data:
**Conversion APIs can send data from various sources, including apps, websites, PCs, consoles, connected TVs, and even offline interactions like in-store purchases. This allows you to:
- Optimize campaigns using data from all online and offline environments where your services operate.
- Measure campaign performance more accurately across multiple channels.
- Plan future operations based on a complete picture of user interactions.
For example, Nexon achieved success in PC and console gaming with "The First Descendant" by using Conversion APIs through Airbridge for their user acquisition (UA) campaign. This allowed Nexon's UA Team to directly send in-game conversion events to advertising platforms like Meta and X. By integrating Conversion APIs with web pixels on landing pages, Nexon enhanced their cross-platform campaign, ultimately resulting in an increase in Return on Ad Spend (ROAS).
> 📚 [See how Airbridge powers Nexon to drive performance marketing for PC and console games](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta1-nexon-the-first-descendant&og_tag_id=102968490&routing_short_id=kthfj3&sub_id=abr-en&tracking_template_id=d0c8d0332ea8449d6e592a2c22e1a591&ad_type=click)
## Connect Conversion APIs through Airbridge
Stay updated with the latest Conversion APIs from advertising platforms. Airbridge integrates with major players like Meta, Snapchat, TikTok, X, and soon Google. With Airbridge, you can manage multiple APIs and monitor performance metrics from one single dashboard, simplifying data analysis and campaign optimization.
Contact the [Airbridge Team](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta2-book-a-demo&og_tag_id=102968689&routing_short_id=mhmecs&sub_id=abr-en&tracking_template_id=9b28522cf38b99dabb7bcf9baf66fdfd&ad_type=click) today to discover how you can use Conversion APIs to boost your omnichannel marketing strategy.
---
## Airbridge now provides near real-time attribution from Meta’s Aggregated Event Measurement
URL: https://www.airbridge.io/en/blog/airbridge-x-meta-aem-integration
Language: English
Published: 2024-10-11
We’re excited to announce that Airbridge is now providing enhanced app install reporting for iOS 14.5+ Meta Ads campaigns. We’ve been working closely with Meta to provide a more comprehensive view of performance by leveraging App Promotion Aggregated Events Measurement in our attribution waterfall and reporting surfaces.
## **What is Aggregated Events Measurement?**
[Aggregated Events Measurement](https://www.facebook.com/business/help/721422165168355?id=1877298665783613) (AEM) is Meta's proprietary protocol designed to measure web and app events from iOS 14.5 and later devices while preserving user privacy. AEM provides near real-time reporting, enabling quicker optimization decisions and offering more granular insights into your campaign performance.**
**
# **Airbridge & AEM**
### **How does this integration benefit you?**
With more granular and timely AEM data available through Airbridge, you can optimize your Meta Ads App Promotion campaigns on iOS 14.5+ with greater precision.
**Near real-time reporting
**By leveraging AEM, Airbridge now provides near real-time app install reporting for iOS 14.5+ Meta Ads campaigns, allowing for faster campaign optimization. This quick and detailed delivery from AEM can help you complement your existing performance insights to better understand your iOS campaigns.
**Comprehensive performance view
**Integrating App Promotion AEM data into your daily Airbridge reports and existing dashboards is as simple as switching on the AEM toggle. This lets you view AEM data directly from **Airbridge's Unified Dashboard** alongside performance data from iOS, Android, SKAN, and the web—all in one place. You can also set customizable lookback windows, from immediate up to 30 days, enabling you to evaluate and analyze attribution exactly as you need. See [our guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#how-to-view-meta-ads-campaign-data-in-airbridge) to learn how.
### **How do you enable AEM?**
To enable Aggregated Event Measurement for your App Promotion campaigns with Airbridge, just toggle it on in your Airbridge dashboard settings. For detailed instructions, please refer to our [AEM integration guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem). You can also [get in touch with us](https://www.airbridge.io/book-a-demo); we’ll be happy to help you make the most of this integration and get started with AEM.
> **💡 FAQ:
>
> Q: Will this update impact my SKAN reporting?
> **A: No, this update will not have any impact on your SKAN reporting. **
>
> Q: Do I need to select AEM when creating an App Promotion campaign in order to see the updated reporting in Airbridge?
> **A: Yes, you will need to select AEM under Campaign Attribution Method when creating your iOS App Promotion campaigns in Meta Ads Manager. Please contact Meta Business Help Center for support.
>
> **Q: How does AEM app install reporting factor into Airbridge’s attribution waterfall?
> **A: Please refer to [our guide](https://help.airbridge.io/en/guides/ad-channel-touchpoint-types) on the attribution logic.
---
## Measure short-form video ads with Airbridge & Kwai integration
URL: https://www.airbridge.io/en/blog/kwai-integration
Language: English
Published: 2024-10-02
Airbridge's integration with Kwai for Business now enables you to accurately measure and optimize your Kwai campaign performance. This powerful partnership allows you to tap into Kwai's massive user base of over 1 billion monthly active users (MAU) across more than 30 countries, while leveraging Airbridge's advanced analytics capabilities.
[__wf_reserved_inherit]
## Precise measurement of Kwai campaigns
With this integration, you can now:
- Create and manage tracking links specifically for Kwai campaigns
- Set up tailored attribution rules for Kwai's platform
- Configure postbacks to send real-time data on in-app events and conversions
These features allow you to gain deeper insights into how your ads perform across various Kwai ad formats including:
- In-feed ads
- Short-form video ads
- Influencer collaborations
## Enhanced analytics for Kwai community
Airbridge's platform helps you better understand and engage with Kwai's primary demographic of Gen Z and younger Millennial users. You can now:
- Track user engagement with your content
- Measure the effectiveness of trend-based campaigns
By integrating your Kwai campaigns into Airbridge, you'll have access to comprehensive data that allows you to optimize your targeting, creative strategies, and budget allocation for Kwai’s highly engaged community.
For step-by-step instructions, please refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/kwai-for-business). To learn how Airbridge and Kwai for Business can drive your app growth, reach out to [Airbridge](https://www.airbridge.io/book-a-demo) today.
## About Kwai for Business
Founded in 2021, Kwai for Business is a global platform based in Beijing, China, offering advertisers a complete marketing solution to creatively engage with the Kwai community. With over 1 billion monthly active users across more than 30 countries, you can connect with your target audience authentically and effectively.
Kwai provides a range of social advertising products, including in-feed ads, brand challenges, and user-generated content (UGC) collaborations. Elevate your creative strategy by crafting localized video ads from scratch and tracking their performance—all in one place. Explore trending topics within your audience's community, capitalize on popular trends, and collaborate with top creators using Kwai for Business.
---
## Airbridge x Meta's AEM
URL: https://www.airbridge.io/en/blog/airbridge-x-metas-aem
Language: English
Published: 2024-09-27
We’re excited to announce that Airbridge is now providing enhanced app install reporting for iOS 14.5+ Meta Ads campaigns. We’ve been working closely with Meta to provide a more comprehensive view of performance by leveraging App Promotion Aggregated Events Measurement in our attribution waterfall and reporting surfaces.
## **What is Aggregated Events Measurement?**
[Aggregated Events Measurement](https://www.facebook.com/business/help/721422165168355?id=1877298665783613) (AEM) is Meta's proprietary protocol designed to measure web and app events from iOS 14.5 and later devices while preserving user privacy. AEM provides near real-time reporting, enabling quicker optimization decisions and offering more granular insights into your campaign performance.**
**
# **Airbridge & AEM**
### **How does this integration benefit you?**
With more granular and timely AEM data available through Airbridge, you can optimize your Meta Ads App Promotion campaigns on iOS 14.5+ with greater precision.
**Near real-time reporting
**By leveraging AEM, Airbridge now provides near real-time app install reporting for iOS 14.5+ Meta Ads campaigns, allowing for faster campaign optimization. This quick and detailed delivery from AEM can help you complement your existing performance insights to better understand your iOS campaigns.
**Comprehensive performance view
**Integrating App Promotion AEM data into your daily Airbridge reports and existing dashboards is as simple as switching on the AEM toggle. This lets you view AEM data directly from **Airbridge's Unified Dashboard** alongside performance data from iOS, Android, SKAN, and the web—all in one place. You can also set customizable lookback windows, from immediate up to 30 days, enabling you to evaluate and analyze attribution exactly as you need. See [our guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#how-to-view-meta-ads-campaign-data-in-airbridge) to learn how.
### **How do you enable AEM?**
To enable Aggregated Event Measurement for your App Promotion campaigns with Airbridge, just toggle it on in your Airbridge dashboard settings. For detailed instructions, please refer to our [AEM integration guide](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem). You can also [get in touch with us](https://www.airbridge.io/book-a-demo); we’ll be happy to help you make the most of this integration and get started with AEM.
> **💡 FAQ:
>
> Q: Will this update impact my SKAN reporting?
> **A: No, this update will not have any impact on your SKAN reporting. **
>
> Q: Do I need to select AEM when creating an App Promotion campaign in order to see the updated reporting in Airbridge?
> **A: Yes, you will need to select AEM under Campaign Attribution Method when creating your iOS App Promotion campaigns in Meta Ads Manager. Please contact Meta Business Help Center for support.
>
> **Q: How does AEM app install reporting factor into Airbridge’s attribution waterfall?
> **A: Please refer to [our guide](https://help.airbridge.io/en/guides/ad-channel-touchpoint-types) on the attribution logic.
---
## Unlock iOS optimization with Apple Search Ads and SKAN
URL: https://www.airbridge.io/en/blog/asa-and-skan-to-optimize-ios-campaign
Language: English
Published: 2024-09-23
Apple's App Tracking Transparency (ATT) framework has significantly altered the landscape of iOS marketing. By requiring explicit user consent for access to device-level data, ATT places control firmly in the hands of your users.
So, how many iOS users actually opt-in? The numbers are telling. The global opt-in rate for ATT averages around [25%](https://www.statista.com/statistics/1234634/app-tracking-transparency-opt-in-rate-worldwide/#statisticContainer). This low opt-in percentage presents a considerable challenge in gathering sufficient data to optimize your campaigns effectively.
Despite these challenges, iOS marketing remains crucial. As of 2023, iOS holds a [27%](https://gs.statcounter.com/os-market-share/mobile/worldwide) market share globally; in some key markets like the US, it's even higher at [57%](https://gs.statcounter.com/os-market-share/mobile/united-states-of-america). Younger users often prefer iPhones, with [85%](https://bgr.com/tech/teens-still-overwhelmingly-want-an-iphone-over-android-new-survey-says/) of US teens using iPhones.
So, how can you effectively optimize your iOS marketing efforts in this evolving landscape? Let's explore strategies that can help you succeed.
## Boost your UA growth with ASA & SKAN
To jump-start your user acquisition (UA) growth on iOS, you'll need to leverage Apple Search Ads and SKAdNetwork. ASA allows app developers and marketers to advertise their iOS apps directly to users across the App Store, making it a powerful platform for acquiring new iOS users. The best part? ASA provides marketers with advertising performance data for all users, regardless of their consent to tracking, within the privacy-focused iOS environment.
On top of that, if you're advertising across multiple channels, SKAN is a must. SKAN is Apple's privacy-focused framework for campaign measurement and attribution. It enables ad networks and advertisers to measure their ad activity—such as impressions, clicks, and app installs—using aggregated data without accessing individual user information. Different ad channels may support various versions of SKAN, so be sure to check the relevant guides for details on how each version may impact your campaigns:[](https://help.airbridge.io/en/guides/skadnetwork-3)
- [Understanding SKAN 3](https://help.airbridge.io/en/guides/skadnetwork-3)
- [Understanding SKAN 4](https://help.airbridge.io/en/guides/skadnetwork-4)
⚠️ Note: Apple only provides app install data through SKAN.
## Power iOS optimization with the right MMP
Apple Search Ads and SKAdNetwork certainly help, but marketers still face ongoing challenges for iOS optimization. If you're also running Android campaigns, you must combine your iOS data from ASA with Android, in-app, and web data to get an accurate picture of your marketing performance. Interpreting SKAN data can also be tricky. This is where an easy-to-use Mobile Measurement Partner (MMP) comes in handy.
### 1\. Unified data analysis
Apple Search Ads' device-level data only tells half the story. To truly understand user behavior, you have to combine ASA data with in-app insights. Without this integration, you're in the dark about whether ad clicks lead to conversions.
An MMP allows you to analyze the entire user journey in one place, helping you uncover crucial insights into conversion rates, user retention, and predictive lifetime value (pLTV).
### 2\. Mobile marketing optimization
Optimizing your iOS marketing is just the tip of the iceberg for improving your overall mobile marketing strategy. To accurately compare performance between Android and iOS, you need to analyze data consistently across both platforms. However, there's a catch. The ASA dashboard falls short as it only provides basic metrics and doesn't measure the effectiveness of different ad channels.
A good MMP can deliver consistent analysis through attribution across multiple channels, helping you identify which channels are most effective and which drive the best user retention.
- [Understanding Attribution](https://help.airbridge.io/en/guides/attribution-overview)
### 3\. Easy SKAN Setup
Apple shares attribution data through SKAdNetwork, and you can access this data using an MMP. MMPs are authorized to receive and interpret SKAN data, making them essential for iOS app attribution. However, not all MMPs are created equal.
Setting up SKAN can be complicated, with intricacies that aren't always easy to navigate. That's why it's crucial to choose an MMP that simplifies the process. The right MMP will handle SKAN setup efficiently, automatically interpreting the data and presenting it in a user-friendly format you can act upon.
## Ready to dominate iOS optimization? Get started with Airbridge
Airbridge offers a suite of powerful features to simplify your iOS optimization. You can seamlessly integrate Airbridge with Apple Search Ads in just a few steps. Once connected, Airbridge will collect, organize, analyze, and deliver detailed reports on everything from app installs to retention and pLTV, providing over 200 metrics for the insights you need.
If you use Apple Search Ads Advanced, Airbridge links your ad spend with campaign details like campaign names, ad groups, keywords, and creatives, giving you a granular view of performance. Plus, with updates every 4 hours, you won't have to constantly check into the ASA platform for real-time results.
Setting up SKAdNetwork with Airbridge is straightforward. Once you configure SKAN before launching your iOS campaigns, Airbridge will analyze SKAN data daily and integrate it with other performance metrics. Each ad channel may have different SKAN setup requirements, so be sure to review the detailed guides below:
- [Integrating Apple Search Ads](https://help.airbridge.io/en/guides/apple-search-ads)
- [Preparing for iOS App Campaign Measurement with SKAN](https://help.airbridge.io/en/guides/preparing-for-skan)
Are you looking for an MMP to optimize your iOS marketing? Let us handle the heavy lifting so you can free your focus for strategic thinking.
---
## Firebase Dynamic Links Shutdown. Airbridge is here to help.
URL: https://www.airbridge.io/en/blog/introduce-deeplink-plan
Language: English
Published: 2024-09-05
Cue the countdown to extinction. On August 25th, 2025, Google's [Firebase Dynamic Links](https://firebase.google.com/docs/dynamic-links) will shut down. All links served by Firebase Dynamic Links will stop working, and you won't be able to create new ones. That gives you less than 12 months to find, assess, and complete the [migration to a new deep linking service provider](https://www.airbridge.io/en/blog/firebase-dynamic-links-shutdown).
So, now the question is: What's the best alternative? Meet the **Airbridge DeepLink Plan**.
# **Airbridge DeepLink Plan:
A reliable replacement to Firebase Dynamic Links.**
As a leading measurement and deep linking solution, Airbridge presents the **DeepLink Plan as a straightforward and reliable replacement for Firebase's deprecated Dynamic Links**. In this blog, we'll explore 3 reasons why the Airbridge DeepLink Plan stands out as your best choice.
## **1\. Airbridge DeepLinks go beyond the standard.**
Switching to our DeepLink plan will not only keep your deep links intact but also provide support that goes beyond the scope of Firebase Dynamic Links, helping you deliver exceptional user experiences that continue to drive growth, retention, and revenue. Let’s go over the details:
### Friction-free user experience
At Airbridge, we focus on user experience. Users shouldn't be stuck on intermediary pages when opening deep links. That’s why our Deep Link Plan offers **universal compatibility**, allowing direct app openings across most platforms and operating systems. This includes online and offline access, web and app interfaces, Android and iOS devices, and popular platforms like Instagram, X, YouTube, Facebook Messenger, Slack and more. By minimizing intermediary pages, Airbridge DeepLinks ensure a smooth and efficient experience for your users.
### **Complete iOS & Android support**
While Firebase Dynamic Links has some limitations when it comes to iOS support, Airbridge DeepLinks provides **complete performance measurement for both Android and iOS**—without requiring extra steps to collect iOS performance data.
### **Seamless deferred deeplinks**
Enjoy our robust [deferred deep-linking features](https://help.airbridge.io/en/guides/retargeting-with-deep-links). With just one click, Airbridge DeepLinks can direct users straight to their desired in-app destination—even if they haven't installed your app yet.
### **Bulk tracking link generator**
Need multiple tracking links for different ad creatives and placements? Our DeepLink Plan includes a fast [Tracking Link Generator template and API](https://help.airbridge.io/en/guides/creating-bulk-tracking-links), allowing you to generate tracking links all in one go directly within Google Spreadsheets.
### **Custom branded deeplinks**
Boost your brand with **custom short links**. With our DeepLink Plan, use your own domain to create personalized links and enhance social share previews by customizing the URL's title, description, and image.
## **2\. Airbridge DeepLinks delivers more than the rest.**
Airbridge delivers more features than other alternatives to Firebase Dynamic Links on the market. With our DeepLink Plan, optimizing your deep linking strategy is both easier and more effective. Let’s see how we compare to others:
### **Create 10,000 tracking links per month**
Easily create **up to 10,000** [**tracking links**](https://help.airbridge.io/en/guides/tracking-link-management) in bulk through our user-friendly dashboard—that's far beyond Brand A's limit of 3,000 links. It's practically unlimited._\*Need more than 10,000 links per month?_ [_Contact Airbridge_](https://www.airbridge.io/book-a-demo)_._
### **Manage QR codes with ease**
QR codes are essential for tracking offline marketing activities. Airbridge DeepLinks **automatically generates QR codes for every link you create**. Unlike other deep linking services —such as Brand B, which doesn't offer QR codes at all, or Brand C, which provides only limited options—we make QR code management simple and centralized.
### **Master complex scenarios**
Experience reliable scenario management and technical support with our DeepLink Plan. We ensure your deep links work flawlessly across all platforms—mobile apps, in-app browsers, and even when pasted directly into the address bar.
## **3\. Migration to Airbridge is simple and seamless.**
Migration isn't just about switching providers; it's about ensuring your existing deep links and data continue to work. Mishandle the migration, and you risk broken links for your marketing campaigns—and we want to help you avoid that. With Airbridge, migrating your Firebase Dynamic Links is easy. Upload your Firebase data to our migration sheet, and it will automatically generate new deep links to replace the old ones. Just copy and paste to complete the transition.
# **Curious about our DeepLink Plan?**
Looking for a Firebase Dynamic Links alternative? Click below, share your details, and we'll get in touch. Discover how Airbridge can support your deep linking needs.
---
## TikTok Mobile Measurement Partner, Airbridge, has been badged
URL: https://www.airbridge.io/en/blog/airbridge-x-tiktok-mobile-measurement-partner
Language: English
Published: 2024-05-31
Airbridge is now a badged TikTok Mobile Measurement Partner. Airbridge, as a TikTok Marketing Partner, has a proven record on TikTok and helps marketers get started, grow their strategy, and find repeatable success on TikTok. With this proven record, Airbridge has been badged and strengthened the partnership between TikTok and Airbridge.
This badge enables Airbridge to continue providing precise measurements of TikTok's ads performance in a complicated user journey. With the rapid developments in the ad industry and technologies, users experience ads across multiple platforms. However, the traditional Last-Touch Attribution (LTA) model, which assigns all credit to the last-clicked ad, accompanies the possibility of undervaluing some ad campaigns.
To more easily and accurately measure TikTok's marketing performance, Airbridge has recently completed preparations for supporting TikTok's Advanced SAN Integration. In addition, automated event mapping for SKAdNetwork integration with TikTok and new types of conversion events like Focused View will soon be available in Airbridge.
As a badged TikTok Marketing Partner, Airbridge will continue to enhance marketing optimization and user journey improvements, providing measurement for TikTok ad campaigns. Optimize your ad campaign by simply integrating with TikTok for Business in Airbridge.
## How to integrate Airbridge and TikTok for Business
By integrating Airbridge with TikTok for Business, you can monitor the performance of your TikTok campaigns in the Airbridge dashboard. Find more than 10 different reports and over 200 marketing metrics provided by Airbridge to measure campaign performance through both Click-through and View-through Attribution.
From channel integration to cost integration and SKAdNetwork integration, get a holistic view of the marketing performance of TikTok's short-form content on your Airbridge dashboard. For more details, check it out in the [Airbridge User Guide](https://help.airbridge.io/en/guides/tiktok-for-business).
### TikTok For Business
[TikTok for Business](https://www.tiktok.com/business/en) is where brands and marketers communicate with TikTok community as creative storytellers. You can freely build fun videos with entertaining music for interactive campaigns.
### TikTok Marketing Partners Program
The [TikTok Marketing Partners Program](https://partners.tiktok.com/pc/en) is a community of unique and innovative tech & service companies that enable marketers to get started, grow their strategy, and find repeatable success on TikTok. You can find Airbridge in the Partner Directory.
> Discover success stories of leveraging TikTok and Airbridge together:
>
> 👉[How Fidget Town accelerated user acquisition with TikTok
> ](https://www.airbridge.io/en/case-studies/fidget-town-tiktok-measurement)👉[How MU Monarch increased sign-in rates by over 90% through ad creative optimization](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization)
---
## Boost your app growth with Airbridge and BIGO Ads integration
URL: https://www.airbridge.io/en/blog/airbridge-bigo-ads-integration
Language: English
Published: 2024-05-22
We are delighted to announce that Airbridge is officially integrated with BIGO Ads, a worldwide mobile UA, retargeting, and monetization platform. You can now send attribution data from Airbridge to BIGO Ads and monetize high-quality users.
By enabling the integration with Airbridge and BIGO Ads, you can add depth to your ad performance measurement by:
- **Tracking Link Management:** By leveraging Airbridge's capabilities to create and manage tracking links, you can precisely measure and tailor their BIGO Ads campaigns. This setup allows for the adjustment of specific lookback and attribution windows, enabling a detailed analysis of which ads drive user conversions and how they can optimize ad placements and budget allocations for maximum effectiveness.
- **Insightful Postback:** Postbacks provide a reliable basis for calculating ad spend, ensuring that you only pay for successful conversions. This system enhances campaign efficiency and boosts return on investment (ROI), making it a crucial tool for maximizing advertising efforts.
For comprehensive details and step-by-step instructions, please refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/bigo-ads). If you would like to find out how Airbridge and BIGO Ads can help you unleash your app growth, feel free to [reach out to us](https://www.airbridge.io/book-a-demo) today!
## About BIGO Ads
BIGO Ads, a prominent mobile advertising and marketing platform, serves as a gateway to global market engagement. Partnered with multiple ad aggregation platforms, BIGO Ads offers high-quality traffic from a diverse array of social and short-video applications. With its business span in over 150 countries, particularly strong in regions like the Middle East, South Asia, Southeast Asia, and Russia, BIGO Ads facilitates the creation and management of ad campaigns but also enables detailed monitoring and optimization of performance. Serving nearly 400 million MAU worldwide, BIGO Ads leverages massive big data for comprehensive analysis, utilizes AI-driven algorithms for maximized ROI, and provides customized marketing solutions, making it an essential partner for advertisers aiming to tap into international markets. To GO BIG with BIGO Ads, visit [adsbigo.com](http://adsbigo.com).
---
## Cracking the first 10,000 App Installs using TikTok
URL: https://www.airbridge.io/en/blog/cracking-the-first-10-000-app-installs-using-tiktok
Language: English
Published: 2024-04-24
App marketers know the struggle: growing an app to 10,000 installs feels like an intricate art. Endless experiments drain budgets, leaving you guessing what truly clicks. But what if you could unlock a predictable formula for successful app growth?
Enter Hamza Alsamraee, founder of NewForm.ai, and his secret sauce: TikTok.
Hamza has cracked the TikTok code for success, and he's got the receipts to back it up too: over 1 billion + views, a 48% average CAC decrease, a staggering 5.6x ROAS, and over $100 million in managed ad spend.
In this post, get expert insights on campaign strategy from Hamza and campaign measurement & optimization from Airbridge, one of TikTok’s marketing partners.
## What do most app marketers get wrong about TikTok?
I’ll be honest–there’s no one single “hack,” but there is a process we follow that has been golden. It’s all about good content. Everything else is secondary.
There is no golden content style (there are tips we will cover below), so the key is quick, efficient iteration. Most teams fail at this.
Most agencies and in-house teams source content the wrong way. They contact a bunch of influencers/UGC creators, or worse, they hire an agency to contact a bunch of random UGC creators. It’s nearly impossible to find reliable, high quality, and cost efficient creators this way.
The reality is–and I know because I was a creator once–creators are notoriously slow and rarely hit deadlines. They also don’t scale very well–a single short form video can cost a brand thousands of dollars. On the other hand–there are a lot of affordable “UGC” creators, but finding a good one is like finding a needle in a haystack. Brands cycle through dozens just to find someone decent.
I was on the other side of the equation before–I ran growth at Faves, a subscription app which has reached 1m+ downloads. I witnessed first hand how hard it is to source creators–so I became the creator. I personally made 1000+ videos for Faves.
But I’m assuming most people don’t have the time (or interest) to make thousands of videos. The key is having reliable, analytical creators that can iterate for you. This is the biggest difference between Newform and other agencies. Our creators are all in house–and they all have analytical backgrounds (tech, engineering, etc).
This is why the biggest growth “hack” is having a reliable, in-house creator that cranks out dozens of videos a month methodically (whether through us or as a direct employee). Instead of throwing out creative briefs to random creators with varying levels of quality–now you can control for content quality and focus on positioning.
It is this iteration on positioning that ultimately unlocks profitable, scalable growth for your app. You don’t want to change too many variables and just throw things at the wall to see what sticks. Focus on building content programmatically, testing your hypotheses along the way.
## Thousands of TikTok videos later, what surprising insights for maximum ROI and low CPI have you uncovered?
The word “hate.” I’m not kidding–almost half our top performing videos start the video with “I hate \_\_\_.”
Painkillers sell much better than vitamins–not only at the pharmacy, but in the app world as well. Position your app as a painkiller and see results rapidly improve.
In addition, the first person works really well (I \[the creator\] have this problem). Second person ads don’t work as well in our experience. Instead of “you,” focus on “I.” It makes content more authentic and relatable.
## How do performance on Spark Ads compare to other ad options on TikTok?
For Spark Ads to work well, a piece of content needs to be doing well organically already - meaning there is already existing engagement which increases watch time and lowers CPM (tiktok users love scrolling through comments). You’ll notice the biggest app performance advertisers rarely use them for new ads with low engagement as the effect is nowhere to be seen.
This shouldn’t discourage you from Spark Ads altogether. In fact a lot of apps have had TikToks blow up about them and their effects amplified through Spark Ads. I encourage all app advertisers to search for their app on TikTok and contact creators to get rights to use their videos as Spark Ads.
## Acquisition is key, but retention is the real beast. You've cracked the 10,000 install code, but how do you convert fleeting users into loyal fans and, ultimately, paying customers?
Across all apps–don’t go after trend based content (sounds, challenges, etc). It brings in fleeting users. You want to pitch the app for what it is to attract loyal users that will retain and monetize.
Across all apps–we’ve found “phone in hand” content to work better than purely green screen/screen recording content. Instead of screen recording the app, throw in more live demos with a phone in someone’s hand. Something like [this](https://drive.google.com/drive/u/1/folders/1fXkJ7LnslREtoTmrJmZ2744CxgdZwHo1).
[]
At the end of the day–most of retention and monetization is an onboarding problem. But onboarding begins with your content. We focus heavily on this “content-onboarding” fit at Newform. If the content sets incorrect expectations about the product, it’s bound to fail.
## App marketers often stumble on TikTok. What are the most common mistakes made when marketing on TikTok for growth?
A few things.
Targeting the wrong audience with install campaigns. If you have a subscription app and have a revenue goal–you should avoid audiences under 25. While they may show great top of funnel metrics (i.e. CPI, CPT), they do not convert well down the funnel. However, if you are optimizing towards a payment event, this matters less.
Targeting too narrow: the algorithm knows better than 99% of media buyers. I’ve seen only a couple of very rare cases where restricting the algorithm worked well. For the vast majority of apps, just go with broad targeting.
> **\*Airbridge Tip\* Optimize app subscription with funnel analytics
> **
> Reallocate ad budgets or optimize subscription funnel by taking a granular view of channel/campaign/creative/country/device drop-off rates across Install, Sign-Ups, Paywall, Free Trial, Subscription, 2nd/3rd Payment steps, and more.
[__wf_reserved_inherit]
Spending too much money on Android or T2 countries. The biggest opportunity by far is iOS, US users. Android users have similar acquisition costs to iOS users, but are harder to monetize. T2 countries may show very low CPI, but it will be hard to scale profitably given low monetization potential. We focus ~90% of our spend on iOS users, and most of that spend is on US users. Other profitable countries to target include the UK, Ireland, Canada, Australia, New Zealand, Germany, France, Italy, Spain, Scandinavian countries, Singapore, South Korea, Japan, etc.
> **\*Airbridge Tip\* Uncover retention and monetization insights post-install
> **
> Whether you’re a gaming, subscription, or social app, dive into granular in-app metrics like d1, d3, d7 ROAS/retention, ARPU, CPI, Time to First Purchase, Predictive Lifetime and Predictive Lifetime Value, down to the channel/campaign/device/country level on the Airbridge dashboard.
[__wf_reserved_inherit]
### About NewForm
NewForm is a full stack growth partner for your app. Unlike other agencies who focus on ecommerce, NewForm has been programmatically building content specifically for app growth with profitability at the forefront. Through their short form content strategy, they've scaled dozens of early stage apps and generated 8 figures in revenue.
[Hamza](http://linkedin.com/in/hamza-alsamraee/) (founder and CEO) and team are happy to chat, you can book a time [here](https://calendly.com/newform/discovery-call-website).
### About Airbridge
Global app marketers ditch costly attribution tools for Airbridge’s unified measurement platform. Our ultra-granular, hyper-configurable dashboard provides real-time visibility of cross-platform user journeys across mobile, web, PC, and console from first click to loyal users. 200+ metrics, iOS measurement, deep linking, fraud protection, and data sharing - at an all-in-one price that fuels your marketing budget, instead of draining it. See why Nexon, Alarmy, and LG made the switch.
Get your free demo at [airbridge.io](https://www.airbridge.io/book-a-demo)!
---
## Mastering AI marketing: the key to mobile marketing success
URL: https://www.airbridge.io/en/blog/mastering-ai-marketing
Language: English
Published: 2024-04-11
As artificial intelligence (AI) technology rapidly advances, we have entered the era of AI marketing. Marketers are now keen to stay ahead in the increasingly competitive landscape, which urges them to understand the intricacies of AI and incorporate it into their marketing strategies. With AI marketing evolving beyond a mere trend to become an essential strategic tool, it's crucial for us to understand its essence. Let's delve into how AI is reshaping the mobile marketing strategy and execution, and explore how marketers can embrace and leverage these changes.
## Evolving paradigms in mobile marketing
The significance of AI and AI marketing is on the rise. AI encompasses technologies that enable computer systems to mimic human intelligence, including learning, reasoning, recognition, and language understanding. These capabilities allow machines to autonomously learn and make decisions through data analysis, pattern recognition, and natural language processing. AI marketing leverages these capabilities to maximize customer engagement by analyzing customer data and generating personalized content, opening new possibilities for marketers to develop advanced marketing campaigns and personalization strategies.
AI's integration is redefining the core of mobile marketing, indicating a deeper and more meaningful evolution of the relationship between businesses and their customers. Marketers can now use mobile apps to predict and analyze customer behavior patterns and preferences, allowing for the optimization of marketing messages and campaigns. Moreover, decision-making driven by data analysis has become a key component of marketing strategies. A deep understanding and application of AI in marketing is indispensable for businesses aiming to enhance customer experiences and achieve sustainable growth.
### Revitalized advertising strategies with AI
Leveraging AI in marketing offers marketers new opportunities. Here's how these strategies work and the benefits they bring:
- **Identifying target audiences and optimizing strategies through machine learning:** AI-driven machine learning algorithms analyze extensive data, capturing nuanced user characteristics and behavior patterns. This analysis enables marketers to identify target audiences with high precision. They can then fine-tune ad campaign targeting to engage these specific groups. By targeting more precisely, marketers can deliver messages that are more relevant to potential users. This approach improves click-through and conversion rates. E-commerce and retail companies have extensively integrated AI marketing to boost customer experience and operational efficiency. For example, Amazon employs machine learning for personalized product recommendations and demand forecasting, significantly contributing to an increase in its net sales.
- **Advanced user behavior analysis and prediction for personalized marketing:** For personalized marketing, advanced user behavior analysis and prediction are key. Machine learning algorithms delve into app user behavior data. They uncover user preferences and response patterns. Marketers can identify user groups with specific behavior patterns. This insight allows them to predict future actions and tailor campaigns, enhancing engagement. Through targeted notifications, personalized emails, and customized ads, they build customer loyalty. Entertainment and streaming services, like Netflix, use this approach for their content recommendation systems. This improves user satisfaction and aids in long-term subscription retention by tailoring movie and TV show suggestions to individual viewing habits.
- **Automating and optimizing ad creatives with AI:** AI revolutionizes how ad creatives are automated and optimized. It plays a significant role in generating and refining ad content automatically. Machine learning algorithms suggest the most effective ad messages and designs by analyzing user interests and response data. This method not only saves considerable time and resources but also increases the diversity and creativity of ad content. With AI, there is continuous optimization as more data is analyzed over time. Popular tools for creative content generation include [ChatGPT](https://chat.openai.com/auth/login) and [Midjourney](https://www.midjourney.com/home). These tools enable understanding consumer preferences through conversations and the creation of customized images for impactful advertising campaigns. This showcases a method to swiftly and effortlessly elevate the variety and creativity of marketing content with AI.
### The impact of AI on marketers
The adoption of AI introduces fundamental changes to marketing practices. Automating repetitive and time-consuming tasks allows marketers to allocate more time to developing strategies and producing ad creatives. This enhances work efficiency and enables real-time adjustments to more personalized marketing campaigns.
In the AI era, marketers should:
- **Enhance technical knowledge and proficiency:** The rapid advancement and integration of AI technology require a deep understanding of the technology and the ability to adeptly use new tools and platforms. This flexibility is crucial for adapting to technological changes and maximizing marketing activities' efficiency and effectiveness through AI.
- **Strengthen data interpretation and strategic thinking:** While AI technology can collect and analyze vast amounts of data, deriving meaningful insights and making strategic decisions remain essential skills for practitioners. The ability to interpret data and develop strategies based on these insights is a core competency in the AI era.
- **Develop creative strategies:** As AI expands the scope of tasks it can handle, marketers must generate new value through creativity and unique approaches. This is key to providing tailored customer experiences and developing differentiated marketing strategies, contributing to a company's competitive edge and growth.
## Mobile marketing: impossible without AI, achievable with Airbridge!
AI technology has become an integral part of mobile marketing, bringing indispensable innovations across the sector. Adapting to and harnessing AI is essential for any company aiming to maximize marketing growth in the competitive era. Airbridge is stepping up the game, ready to supercharge marketers' workflows with AI-enhanced features. As your partner in growth, we are here to elevate your presence in the digital landscape.
As the mobile marketing paradigm shifts, making data-driven decisions, personalizing customer experiences, and developing forward-looking strategies have become more critical than ever. A deep understanding and proactive use of AI technology are mandatory, establishing themselves as essential elements of mobile marketing strategies. [Start preparing for the new era of mobile marketing with Airbridge today](https://www.airbridge.io/book-a-demo).
---
## Decoding game monetization: finding your perfect model
URL: https://www.airbridge.io/en/blog/game-monetization
Language: English
Published: 2024-04-09
In the ever-evolving mobile gaming market, monetization strategies have become more than just a way to make money; they are a critical factor in a game's success. To maximize user engagement and ensure sustainable profitability, the gaming industry is exploring various monetization approaches.
In this blog post, we'll take a look at the game monetization model trends in this rapidly changing industry, and how you can create a strategy that maximizes the value of your game.
## What is mobile game monetization?
Game monetization is the process by which game developers generate revenue from the games they create. This process allows developers to fund ongoing game updates and support, which is essential for maintaining and developing their games' sustainability, quality, and competitiveness. Understanding your game's nature, target audience, and market trends is crucial to finding the most suitable monetization model.
## The types of game monetization models
1. In-App Purchases (IAP)
This is exactly what it sounds like: a model that monetizes purchases made within the game. Typically, you distribute your app as a free-to-play game and drive additional in-game purchases. It's a way to generate ongoing revenue as users pay for specific items, events, or content, such as extra lives, coins, weapons, moves, or boosters, as they progress through the game.
2. In-App Advertising (IAA)
A revenue strategy that monetizes the sale of in-game advertising space. Ads can be displayed as interstitials or banner ads, provided they do not interrupt the user's gaming experience. The most common ad formats used in mobile games include:
- **Interstitial ads**: An ad that suddenly appears across the screen during the game. After viewing the ad (e.g., 10 seconds), the user can exit the ad and return to the game, but it can have a negative impact on the game experience.
- **Rewarded ads**: Unlike interstitial ads, users can choose whether or not to view an ad. Users will receive a reward for viewing the ad.
- **Playable ads**: This ad format allows users to experience your game through a one-minute video before installing it. They are effective for retention and keeping users engaged.
- **Banner ads**: An ad format that appears at the top or bottom of the screen during gameplay.
- **Offerwalls**: An in-app ad format that acts like a mini-store in your game and allows you to reward users for engaging with your ads.
3. Subscription
A type of in-app purchase that gives users access to additional content. They come in a variety of formats, including "Pass" subscriptions that provide access to a desired product or additional service for a set period of time, "Remove Ads" subscriptions that remove in-app ads from your game for a fee, and "VIP" subscriptions that provide a series of exclusive benefits over a set period of time. They foster long-term user engagement.
4. Paid games
This model charges users at the time of download, allowing access exclusively to those who purchase it. It's effective for generating initial revenue. However, it's crucial to demonstrate the game's value upfront to ensure buyer satisfaction.
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Track in-app purchases, ads revenue, and LTV across every channel.%%
## Creating an effective game monetization strategy
### Hybrid monetization model strategy
You don't have to rely on one monetization model to monetize your game. Considering a hybrid monetization model allows for the combination of various strategies described previously. Common combinations include in-app ads + in-app purchases, in-app purchases + subscriptions, in-app purchases + in-app ads + subscriptions, and more. Adopting a hybrid approach lets you simultaneously target both paid and free-to-play users, opening up multiple revenue streams.
Implementing a game monetization model doesn't happen overnight. It's common for developers and publishers to delay forming a monetization strategy until after game development is complete. This can lead to a lack of integration between the monetization model and the game itself, which can be jarring for users. To avoid this, having a detailed strategy for your monetization model before or concurrently with the start of game development is crucial. This approach ensures the monetization is implemented as naturally as possible, enhancing rather than disrupting the player's experience.
### Finding the right monetization model mix for your game
Depending on the genre of your game, the monetization model that works best for your game can vary. Games in the RPG or simulation genre are designed to keep users playing through various stages, so in-app purchases to keep them playing are a popular way to keep them engaged. On the other hand, games that are casual and free to play, such as hyper-casual games, are more likely to be monetized with in-app ads, especially rewarded video ads.
Yet, there's no one-size-fits-all solution; the optimal strategy requires a thoughtful analysis of your game's genre, narrative, and the preferred playstyles of your target audience to balance the mix of ads and in-app purchases effectively.
### Data-driven testing is essential
Once you've chosen the right ratio and strategy, you need to make sure that the model is generating the revenue you expect. Particularly if you decide to use in-app advertising as your monetization model, setting key performance indicators (KPIs) for ad performance and accurately measuring them is essential. It's important to run A/B tests with different scenarios to ensure that you're placing ads effectively and making quick decisions based on the data you derive.
Airbridge provides the Retention Report, a statistical report that measures how often users return to your service, and Predictive Lifetime, which allows you to predict when users will return to your service. We also offer the Revenue Report that allows you to view various metrics about the revenue generated by users in your game, including revenue, ROAS (Return on Ad Spend), ARPU (Average Revenue Per User), ARPDAU (Average Revenue Per Daily Active User), and more.
The Revenue Report also provides a [Predictive LTV (pLTV)](https://www.airbridge.io/en/blog/airbridge-predictive-lifetime) metric that predicts the lifetime value of a user based on historical data, giving you an estimate of how much revenue a user will generate in your game over their predicted lifetime. Being able to predict the future value of a user, rather than simply measuring performance based on the new users you acquire through advertising and their spending within a specific period, represents a critical strategy for cost-effective marketing.
Due to the rapidly evolving nature of the gaming industry, monetization strategies are becoming increasingly diverse. If you're looking for more creative and effective monetization strategies to capture the attention of the gaming market and its users, [consider consulting our marketing experts](https://www.airbridge.io/book-a-demo)!
---
## Transcending boundaries: Airbridge unveils PC & Console Tracking for cross-platform mastery
URL: https://www.airbridge.io/en/blog/mobile-pc-crossplatform
Language: English
Published: 2024-04-03
Imagine seamlessly transitioning from playing a game on a console at your friend's to continuing the adventure on your smartphone on your way home and, finally, wrapping up the quest on your computer. Sounds easy and convenient, doesn't it? More and more games are going cross-platform, and consequently, the need for comprehensive cross-platform measurement is increasing.
Yet, cross-platform measurement poses a challenge. Each platform—whether mobile, PC, or console—acts as a fortified castle, creating boundaries for marketers. While users effortlessly move between platforms, marketers must embark on a solitary quest, navigating these boundaries like adventurers, to measure ad performance effectively.
Enter Airbridge, a trusted companion in this epic journey, unveiling its latest innovation: PC & Console Tracking. At its core lies three powerful skills, each essential for conquering cross-platform measurement.
### **Skill #1: Mastering the art of data collection**
To effectively measure cross-platform performance, Airbridge gathers data from disparate sources across platforms, just as adventurers collect treasures scattered across vast landscapes. Given the varying data formats and transfer methods, a thorough approach is a must. That’s why Airbridge employs a combination of tracking links, SDKs, APIs, and more to collect data.
In addition, the more users move between platforms, the harder it becomes to track their journey. After collecting rich data comes connecting scattered pieces of user data—from ad interactions to in-game activities and identifiers. It’s like piecing together a fragmented map, a near-impossible task.
Undeterred, Airbridge forges ahead. With advanced attribution methods and strong data collection capabilities, Airbridge expertly completes the user journey.
### **Skill #2: Leveraging the power of intuitive analytics**
Armed with a trove of data, marketers face the next obstacle: deciphering the data. Airbridge presents all the data in a real-time unified dashboard, which is equipped with a range of charts and tables that can transform cross-platform complexities into simple clarity.
Get started with the Actuals Report to gain a granular view of various user actions at the channel, campaign, and platform level. For instance, you can see how many times your game was installed on mobile and PC, or how many of your mobile users purchased an item. Or use the Active Users Report to track key metrics such as DAUs, WAUs, MAUs, paying users, revenue, ARPU, and ARPPU—all conveniently presented in one place.
Airbridge empowers marketers to extract actionable insights, enabling timely and informed decisions critical for cross-platform success. Easily visualize how your marketing efforts are contributing to your game’s growth.
### **Skill #3: Exploring the new frontiers of internet cafes**
According to [Newzoo](https://newzoo.com/resources/rankings/top-10-countries-by-game-revenues), South Korea boasts the world’s fourth-largest game market, after the US, China, and Japan, despite its smaller population. Internet cafes, also known as cybercafes or PC bangs, play a pivotal role in the country’s gaming landscape. In 2023, more than a quarter of South Korean gamers reported visiting internet cafes nine or more times per month on average.
Given such prevalence, the Airbridge team embarked on an ambitious endeavor—to measure ad performance among users who play games in both their personal spaces and public gaming spaces. How do you complete the user journey if the user views an ad in an internet cafe and installs the game on their personal PC at home, or views an ad on their smartphone and plays the game at an internet cafe?
After months of rigorous testing with leading game developers, Airbridge’s cross-platform measurement proved its effectiveness within the dynamic environment of South Korean internet cafes. This serves as evidence of Airbridge’s readiness to operate in diverse settings. Airbridge's pursuit of the “ultimate" cross-platform measurement extends beyond not only platforms but also the physical locations, presenting marketers with unprecedented opportunities to engage their target audience.
### Let’s transcend the boundaries, together
The Airbridge team doesn’t stop here; we’re looking for more boundaries to transcend so that we can propel marketers toward fresh horizons of success. With Airbridge PC & Console Tracking, marketers can chart new territories and seize opportunities wherever they may arise. Join on this epic quest for cross-platform marketing mastery with Airbridge by your side.
[Book a demo today](https://www.airbridge.io/book-a-demo) and transform your cross-platform marketing.
---
## 10 essential metrics all game developers should look out for
URL: https://www.airbridge.io/en/blog/10-essential-metrics-all-game-developers-should-look-out-for
Language: English
Published: 2024-02-28
Understanding mobile game analytics can be a bit daunting, but it's crucial for game developers looking to thrive in the competitive market. [Analyzing these metrics](https://gameanalytics.com/analytics-tools) gives you valuable insights into player behavior, game performance, and how to make your game more profitable.
When we talk about metrics, we're referring to all sorts of numbers and data points. Some of them are very straightforward, like downloads and sessions, while others, like user churn and average revenue per paying user (ARPPU), can be a bit trickier to wrap your head around. But each of these metrics is like a piece of a puzzle that, when put together, paints a picture of how players interact with your game.
To make things a bit clearer, we've put together a list of ten key metrics that you've probably already heard of but might not fully understand why they're important or how to use them effectively.
## Daily active users (DAU)
Starting with the basics, DAU refers to the number of unique users who engage with a game or app within a single day. It's a key metric used to measure the daily popularity of a game. This metric provides insights into how frequently users interact with the game, helping developers assess user engagement levels, track trends over time, and identify patterns in player behavior.
When using this metric, remember to set it into perspective. Take a game that has 10,000 engaged users who play the game several times each day and actively monetize. Compare that to a news or messaging app with 1,000,000 DAUs but no monetization mechanics. While the other app attracts more players to engage with the game repeatedly, developers do not dispose with revenue, which can restrict developers from further development and improvements. This shows, that the DAU count is merely a snapshot in time, and the surrounding context can be just as important, if not more important, than a large user base.
## Sessions
Every time any user, not just a unique user, opens your app, that counts as a session. Similar to DAUs, the total number of sessions requires some context to be a helpful number. Specifically, focus on the average number of sessions per DAU, as this metric can tell you about how engaged users are with your game.
An app’s genre does indeed have an effect on the Sessions/DAU ratio, as some game styles lend themselves to more frequent sessions. If players return five to ten times each day, it’s safe to assume they enjoy the game.
You should also pay attention to session length. It measures the average duration of a single gaming session, indicating how long players typically play the game in one sitting. Longer session lengths suggest high player engagement and enjoyment, while shorter session lengths may indicate issues with pacing, content depth, or player frustration.
## Engagement metrics
### DAU/MAU ration
The ratio of daily- to monthly active users (MAU) shows how well an app retains users and is often referred to as the stickiness of a game. This metric shows you how frequently users log in to your app. This metric will be more straightforward to discuss with an example.
Let’s say an app has 100,000 MAU and averages 15,000 DAU. Then, the DAU/MAU ratio would be 15 percent. This means the average user logged in on roughly 15 percent of the days that month.
Since this is a ratio, the metric can only be a value between zero and one. Values closer to one mean users are opening the app on a higher percentage of days. Popular apps like Facebook have reported [DAU/MAU ratios as high as 68 percent](https://investor.fb.com/investor-news/press-release-details/2023/Meta-Reports-First-Quarter-2023-Results/default.aspx), while most successful gaming apps have ratios closer to 20 percent.
### Retention
Retention is arguably the most important metric in a free-to-play game. Successful free-to-play games create long-term relationships with users. Users enjoying the experience are willing to pay for a competitive advantage. A game needs to have strong retention to have time to build this relationship.
We suggest, separating your users into cohorts based on the day they download your app to calculate your retention. The day that the download occurs is Day 0. Users who open your app the next day (Day 1) should be marked as retained. If they do not open the app, they are not retained. This calculation is performed for the user cohort each day after downloading the app. Standard days used for retention are 1, 3, 7, and 30.
You can also measure [event-based retention](https://gameanalytics.com/blog/engagement-tracing-retention) or focus on how various triggers influence the retention rate.
### Predictive lifetime value
Another crucial metric for game developers to monitor is predictive lifetime value (PLTV). This forward-looking metric estimates a player's potential revenue over their entire lifespan within a game. Predictive lifetime value utilizes advanced predictive analytics techniques to forecast a player's future spending behavior based on their past interactions and in-game activities.
By leveraging this metric, game developers can gain valuable insights into the long-term revenue potential of their player base, allowing them to tailor their monetization strategies and marketing efforts accordingly. By identifying high-value players early on and implementing targeted retention and engagement tactics, developers can maximize the lifetime value of their player base and drive sustainable revenue growth.
Predictive lifetime value also enables developers to optimize user acquisition campaigns by identifying the most valuable player segments and allocating resources toward acquiring similar high-value players. By focusing on acquiring players with the highest predicted lifetime value, developers can improve the overall return on investment of their user acquisition efforts and drive more profitable growth.
> Explore further details on how Airbridge supports Predictive Lifetime Value (LTV) and Predictive LTV in our reports by following the link below.
>
> 👉[Mastering predictive LTV for swift marketing decisions](https://www.airbridge.io/en/blog/airbridge-predictive-lifetime)
### Churn
Churn is essentially the opposite of retention. This metric portrays those players who stop engaging with the game over a specific period, indicating the number of players who churn or disengage. A high churn rate suggests issues with player retention and overall game satisfaction, while a low churn rate indicates strong player loyalty and satisfaction.
The churn metric makes the most sense in a subscription business model, while some nuances are involved when applying it to free-to-play games.
### Monetization metrics
Moving right along to everyone’s favorite topic: money! The above metrics measure your relationship with your users, how often they return to your game, and their behavior patterns. But arguably, the most important metric for many [indie developers](https://gameanalytics.com/indie-program/) is whether their game is making enough money.
### Conversion rate
The conversion rate measures the percentage of unique users that purchased out of the total number of users during a specific period. You can also measure the conversion rate of ads served in a free-to-play game.
Getting a user to pay real money for a game they can play for free is challenging. But, as with many other industries, repeat purchasers generate most revenue in free-to-play games. You can encourage players to make that first conversion by offering them a virtual item of incredible value.
### Average revenue per daily active user (ARPDAU)
Average revenue per daily active user is a monetization metric that measures the average revenue generated by each unique user who engages with a game or app within a single day. It provides insights into the effectiveness of monetization strategies and the overall revenue potential of the user base. By dividing the total revenue generated daily by the number of daily active users, developers can calculate the average revenue generated per user per day. ARPDAU helps developers assess the performance of in-game monetization features, optimize pricing models, and track revenue trends over time. It's a crucial metric for maximizing revenue and driving sustainable mobile game and app growth.
This is a great metric to track before and during user acquisition campaigns. Before acquiring users, make sure you know the range of your ARPDAU and how it fluctuates normally. During a campaign, segment your new users by source and see which networks or games perform the best in your app.
### Average revenue per paying user (ARPPU)
Average revenue per paying user measures only the subset of users who have completed a purchase in a game. This metric can vary dramatically based on the game genre. Hardcore games tend to have higher monetization metrics like ARPPU, but they also lack the mass appeal of more casual games.
The above metrics are standards that can help you get started in the world of analytics. The most important part of mobile game analytics is to get started and establish benchmarks for your own games. Once you understand how your users behave, you can measure things like the impact of a game update or changes to your user acquisition strategy.
If you want analytics explicitly tailored to the gaming industry, head over to [GameAnalytics](https://gameanalytics.com/analytics-tools) and explore their cost-effective analytics solutions. From an easy-to-use analytics pipeline to industry benchmarks, and raw data access, GameAnalytics allows you to optimize your games while keeping your data safe and sound.
---
## Advertising trends marketers should keep an eye on in 2024
URL: https://www.airbridge.io/en/blog/2024-digital-marketing-trends
Language: English
Published: 2024-02-28
In today’s ever-changing world, marketers must be quick to identify and adapt to emerging trends to seize new opportunities. These trends reflect consumer preferences and provide rapid insights into market dynamics. Although forecasting in the fast-paced digital industry is difficult, analyzing data from the previous year can guide us in identifying advertising trends to watch in 2024.
## **Shorts, CTV, streaming services... It’s all about videos**
'Short-form' was undoubtedly the buzzword that swept through the media industry last year, reflecting the growing preference among busy modern individuals for videos under a minute. This trend coincided with a shift from traditional TV viewing to smartphones, tablets, and PCs.
The emergence of OTT (over-the-top) streaming services and Connected TV (CTV) has also revolutionized the media landscape. Unlike traditional broadcasting, which is constrained by a schedule set by broadcasters, cable TV, or telecom providers, OTT services offer viewers the freedom to access content anytime, anywhere, via an app or website. These services, accessible on any device with an internet connection and a screen—including streaming devices, smart TVs, set-top boxes, and game consoles—have become a staple in our daily lives.
Moreover, the industry is increasingly focusing on Free Ad-Supported Streaming Television (FAST) services offered by TV manufacturers. These platforms deliver streaming services for free, supported by advertisements, eliminating the need for viewers to sign up, subscribe, or purchase additional devices. Particularly popular in North American regions, these services have caught the attention of major players. Recently, Samsung and LG have announced investments in personalized content and services, aiming to strengthen their competitive edge in the FAST market.
The emergence of new services and platforms presents marketers with a valuable opportunity to explore a wide range of advertising and marketing strategies. Unlike traditional television, which broadcasts the same ads to all viewers, OTT advertising enables precise targeting of users and analysis of the resulting data. As a result, OTT advertising is increasingly popular among marketers and the competition for OTT ad space is expected to intensify in 2024.
[__wf_reserved_inherit]
## **#notsponsored User-generated content (UGC)**
Honest reviews with real user photos often resonate more deeply than a brand's polished ad copy. This is why user-generated content like restaurant reviews, movie ratings, and purchase feedback tends to be more effective than any other marketing strategy. Even in 2024, marketing leveraging user-generated text, images, and videos remains highly active.
In one case, Airbridge customer Whoyaho implemented a marketing strategy aimed at encouraging users to create and share their own gameplay videos to drive organic user acquisition. This strategy targeted the alpha generation, who seek active involvement and a platform for their voices rather than mere consumption of what the brand offers. By facilitating the sharing of videos uploaded by their peers, users naturally engage with each other. With a marketing strategy that keeps up with the times, the company has been able to attract new users, reduce marketing costs, and establish itself in the global market.
[👉Check out how Airbridge client Whoyaho incorporated UGC content into their marketing strategy!](https://www.airbridge.io/en/customers/whoyaho-game-data-marketing-strategy)
[__wf_reserved_inherit]
## **Ads made by AI? Evolving trends in the industry**
There has been ongoing discussion surrounding ChatGPT over the last year. Generative AI is also being explored in creative and artistic fields that were once considered irreplaceable by technology. For example, in the Samsung Life brand campaign, Cheil Worldwide successfully used AI to create all the elements of the ad, including design, imagery, and background music. The fact that an insurance company with a somewhat conservative image attempted to advertise using new technology generated a positive buzz and served as a reminder of the advancements in technology.
Furthermore, the industry is witnessing the emergence of services that allow users to enter a product detail page URL and automatically generate various types of ad banners and videos, or modify ad copy to make it more appealing to customers. Advertisers are also transitioning to machine learning-based real-time bidding, replacing the manual bidding processes of the past. AI's influence and impact on the advertising ecosystem are likely to grow stronger this year.
## **Privacy, no longer a choice, but a necessity**
As privacy laws have strengthened worldwide, companies like Google and Apple have introduced privacy frameworks such as Privacy Sandbox and App Tracking Transparency (ATT) to ensure compliance with these regulations. This has limited the use of third-party data, which was extensively utilized in digital marketing. Consequently, there is now a renewed focus on first-party data, collected directly by companies through their platforms. The service, which provides retailers with ad space and infrastructure on their digital platforms like apps and websites, have gained attention as they allow advertisers to target specific audiences with high conversion rates.
Publishers are also working to reduce reliance on third-party data. For instance, Amazon DSPs have launched a service called Signal based Ads, which uses machine learning to bid and serve ads based on a combination of advertisers' first-party data and publishers' own signals. This year, the industry will continue to look for ways to maximize ad effectiveness while protecting user privacy.
👉[If you're curious about the changing privacy trends in 2024, check out our content here.](https://www.airbridge.io/en/blog/2024-privacy-landscape)
While 2024 will undoubtedly bring disruptive changes to the industry, the new platforms, emerging content formats, and aggressive use of AI automation discussed above will be key breakthroughs for marketers. As marketers, our ultimate goal remains the same through all the changes: to earn the trust of our customers and grow our services. Stay ahead of the curve and embrace emerging trends so your strategy can win the hearts and minds of your customers this year.
---
## Personalized marketing: Win your customers’ hearts with data
URL: https://www.airbridge.io/en/blog/personalized-marketing
Language: English
Published: 2024-02-22
When you first sign up for a new service, you'll notice that an increasing number of apps and websites are asking about your interests and tastes now more than ever. Customized information based on our entered likes and interests has become so familiar that we almost take it for granted.
Let's take a look at why personalized marketing is still important after all these years, and how you can better leverage this valuable information from your users.
## Why personalized marketing continues to be important
Personalized marketing allows businesses to identify the interests and tastes of their customers, enabling the delivery of tailored messages.
Personalization is a key strategy for marketers because it offers several benefits. Personalization allows customers to get offers and information that match their preferences, enhancing their trust and loyalty towards the brand. It lets businesses send the right messages to the right people, cutting down on advertising costs. Satisfied customers are more likely to come back, and brands that become viral can gain a competitive advantage in the market.
### Embracing individuality and diverse preferences
There has been a tendency for averages to become less relevant and traditional means of calculating averages becoming obsolete recently. This reflects a shift from following mainstream tastes to embracing individuality and diverse preferences. Instead of simply categorizing users by demographics like gender, age, or location, it's essential to recognize the unique wants and needs within those groups. While analyzing common characteristics remains important, it's now more crucial than ever to tailor approaches to accommodate individual tastes.
### Changes in response to privacy enhancement
In recent years, a significant focus in the digital marketing industry has been on privacy enhancement. Another major change is expected in 2024, as Google plans to discontinue support for third-party cookies. Third-party cookies, which are snippets of user data from various online sources, have raised privacy concerns. Once this change takes place, traditional marketing methods reliant on third-party cookies will no longer be viable.
As an alternative, ‘first-party data’ is gaining prominence. First-party data collection allows companies to gain deeper insights into user behavior directly through interactions and tools like surveys while respecting privacy. You can use it to personalize everything from marketing campaigns like emails and push notifications to customer experiences within your service. Personalizing the experience increases loyalty and conversion rates for your service.
[👉 Learn more about the privacy landscape of 2024 and Google's third-party cookie phase-out](https://www.airbridge.io/en/blog/2024-privacy-landscape)
### From personalization to hyper-personalization
It's time to move beyond personalization to hyper-personalization. This strategy employs artificial intelligence and big data to craft extremely targeted offerings for each customer. Unlike basic personalization, which might simply suggest appealing products, hyper-personalization analyzes user activity and purchasing patterns, forecasts buying moments, and delivers timely messages to meet customer desires.
Hyper-personalized marketing is growing in popularity as it identifies and anticipates the unique needs of individuals, often before they are aware of themselves. This approach leads to more conversions by leveraging detailed behavioral insights. The potential for technology to advance marketing strategies is truly exciting.
## Personalized Marketing: Hands-on tips to improve
### Create audiences to target ad campaigns
How can you refine your personalized marketing efforts? Once you've successfully acquired new users with your initial campaigns, keeping them active and engaged with your app is crucial. To maximize retention and boost customer lifetime value (LTV), consider executing ad campaigns targeted specifically at audiences identified through personalized marketing strategies.
Airbridge simplifies the process of creating and defining target audiences based on comprehensive data collected via its SDK. You have the flexibility to identify audiences based on specific actions, such as app installs, opens, and membership signups. Furthermore, you can customize your audience segments by combining multiple criteria to meet the unique needs of your service. Once set, these audience conditions are automatically saved, allowing for easy re-use and modification.
[__wf_reserved_inherit]
Merge your extracted data with first-party data for in-depth analysis. It's also possible to create custom audiences automatically in sync with specific ad channels and third-party solutions, facilitating efficient user acquisition through retargeting, detargeting, and lookalike audience strategies. Achieve effective personalized marketing quickly and effortlessly, without additional workload on your developers and marketers.
> 👉 Airbridge's Audience Manager uses Luft, a self-developed database that allows you to work 20x faster than other solutions. If you'd like to learn more about [how to create and extract audiences in Airbridge, check out our guide.](https://help.airbridge.io/en/guides/audience-manager)
### Designing personalized user onboarding flows using multiple solutions
Integrate various solutions to create personalized onboarding experiences for new users. For instance, you can target users based on their interactions with specific campaigns or creatives by transferring app install and deep link attribution results from Airbridge to a customer engagement platform like Braze. This targeting is based on the attribution data, enabling the delivery of personalized messages.
Queenit showcases this by integrating Airbridge with Braze to segment users effectively. Marketers can, for instance, offer a 10% discount to users who discovered Braze after engaging with a specific ad campaign. Additionally, this data enables the creation of targeted campaigns to boost repeat visits, distribute coupons, and send push notifications to users with pending cart items. Leveraging these solutions collectively enables the ongoing refinement of personalized CRM flows for enhanced effectiveness.
[__wf_reserved_inherit]
[👉Learn more about how to integrate Airbridge and Braze for personalized marketing campaigns](https://help.airbridge.io/zh/guides/braze)
Marketing strategies are rapidly evolving alongside current trends. If you're looking for ways to stay on top of the trends using a variety of solutions, [talk to Airbridge's full-funnel marketing experts.](https://www.airbridge.io/book-a-demo)
---
## Digital Markets Act (DMA): Shaping the future of digital competition and privacy
URL: https://www.airbridge.io/en/blog/digital-markets-act
Language: English
Published: 2024-02-06
‘Privacy’ remains the biggest buzzword in the digital marketing industry this year. The European Union (EU) has stepped in to establish a fair and open digital marketplace while safeguarding users’ fundamental online rights. Even for services launched outside of the continental Europe, marketers with a global user base should pay attention to this change. Let's explore the Digital Markets Act (DMA) and its impact on our future marketing strategies.
## What is the Digital Markets Act (DMA)?
The Digital Markets Act (DMA) is a regulation proposed by the European Union to control the power of big tech companies and promote fair competition in the digital market. Aimed at dismantling the monopoly of large platforms acting as “Gatekeepers” between users and business owners, the legislation primarily focuses on regulating big tech companies, ensuring fair competition, and safeguarding user data. EU defined a gatekeeper as a company that provides a core platform service to more than 45 million monthly active users established or located in the EU and to more than 10,000 yearly active business users established in the EU during the last three years. The six companies designated as gatekeepers are Alphabet, Amazon, Apple, ByteDance, Microsoft, and Meta and they are obligated to comply with all DMA regulations.
### Key provisions of the Digital Markets Act
Companies designated as gatekeepers are required to meet the following obligations. Below are some highlights of the legislation.
Gatekeepers must:
- allow interoperability with third-party services.
- allow business users to access the data they generate when using the Gatekeeper platform.
- notify the Commission of any future merger or acquisition plans.
- Gatekeepers must not rank their services and products more favorably than competitors on the Gatekeeper platform.
Gatekeepers must not:
- reuse users’ personal information across services.
- track users for targeted advertising purposes without valid consent.
This means that in the mobile ecosystem, European-based users will be able to easily remove pre-installed apps or services, and new app stores other than the Google Play Store or Apple App Store may emerge.
### The significance and expected impact of the DMA
Once the law is in place, it will enable service providers to enter an industry currently dominated by major platform companies. Furthermore, it will enhance regulations concerning privacy and data usage, ensuring greater security for users’ personal information.
## Gatekeeper responses
[__wf_reserved_inherit]
Big tech companies designated as gatekeepers are making new changes and responses to comply with the bill.
Apple has declared modifications to [iOS, Safari, and the App Store](https://www.apple.com/newsroom/2024/01/apple-announces-changes-to-ios-safari-and-the-app-store-in-the-european-union/). Developers releasing apps in the EU can now distribute iOS apps on alternative app marketplaces and utilize third-party Payment Service Providers (PSP) within iOS apps. Along with these new business conditions, a new billing system with distinct app store fees has also been introduced.
Alphabet intends to present banners in some Google products, seeking consent from users based in Europe regarding the sharing of user data across its products and services, particularly for personalized ad exposure purposes. In addition, they have announced plans to expand tests on Google search results, incorporating dedicated spaces for third-party providers such as comparison sites. Furthermore, Alphabet has [implemented changes in various areas](https://blog.google/around-the-globe/google-europe/an-update-on-our-preparations-for-the-dma/), such as simplifying the process of switching the default search engine and browser on Android phones.
> **Airbridge supports the latest Google EU user consent policy
> **Advertisers must collect consent from end users in the European Economic Area (EEA) by displaying a privacy prompt in their apps. Then, advertisers must share consent signals with Google. Airbridge will update its SDK and API in February 2024 to support Google’s latest EU user consent policy. Advertisers who do not run Google Ads campaigns in Europe or who do not have end users in the EEA do not have to take any action.
> 👉[Learn more in the Airbridge user guide.](https://help.airbridge.io/en/guides/dma-google-eu-consent-policy)
## Coping with constant changes
The industry is waiting with bated breath to see if this is the beginning of the end for what has long been regarded as a ‘walled garden’. Numerous additional changes are on the horizon, including those covered above, and the European Commission’s response to these measures is yet to unfold. Make sure you stay attuned to developments and be prepared to welcome challenges with an open mindset.
As your partner in ad performance measurement, Airbridge is constantly updating and introducing features to support changes for app services and marketers. If you’re contemplating how to tailor your marketing strategy for the forthcoming transformations, [consult with the marketing experts at Airbridge.](https://www.airbridge.io/book-a-demo)
---
## Google Privacy Sandbox: How to tackle new challenges
URL: https://www.airbridge.io/en/blog/google-privacy-sandbox
Language: English
Published: 2024-01-24
On January 4, 2024, Google announced its plan to turn off third-party cookies for 1% of Chrome users, as part of a gradual phase-out aiming for full removal by the end of the year. [Google’s Privacy Sandbox project](https://privacysandbox.com/intl/en_us/), first introduced in 2019, entails collaboration among industry players to allow advertising without invading user privacy. Let’s explore the measures and guidelines Google has announced so far and how we can prepare for the upcoming challenges.
## What’s Changing?
Google Sandbox centers on two pillars across the web and apps on Android The most significant change on the web is the deprecation of third-party cookies. At the same time, Android is limiting tracking through advertising ID (Google Advertising ID, GAID) which was used as a user identifier, to build a safe web/ app environment.
The bottom line, however, is that the changes you’ll see in your business right now aren’t as significant as you might think. Especially, when it comes to attribution performance measurement in Airbridge. Installs via Google Play Install Referrer will still be recorded, and deep link and web UTM parameters will still be available. Based on this, the cross-platform identifier (Airbridge ID) that Airbridge uses for performance measurement is still available.
Nevertheless, it is important to note what will be affected and how you can specifically prepare for it.
### What does the deprecation of third-party cookies mean?
Cookies are small pieces of data that are stored when a user visits a website. Third-party cookies are a collection of data stored by a third-party service other than the owner of the website the user visited. The advertising industry has been able to use this information to track a single user across multiple web browsers and show a customized ad to that user. It was these third-party cookies that allowed products you searched for on the website to later pop up as channel ads on other social networks or platforms. However, privacy concerns emerged due to the collection of user behavior on websites without their consent, leading to the decision to discontinue support by 2024.
### GAID deprecation on Android
Android is also planning to cease support for the Google Advertising ID (GAID), a user identifier. Fortunately, you can still leverage Google Play Install Referrer or identifiers that MMPs like Airbridge use to measure performance. However, considering GAIDs have long been used for user tracking in Android apps and have played an important role in assessing performance, it is now time to begin exploring alternative methods for evaluating ad effectiveness.
## What Google suggests…
### Show relevant content and ads - Topics & Protected Audience API
Google plans to provide the following alternative measures in the form of APIs for the digital advertising industry, which is facing the need for new methods due to significant changes.
While unique identifiers have traditionally been used to understand user behavior and infer their interests, **the Topics API** helps you acquire new users through ad targeting based on their recent interest-based history within a browser or app. Among approximately 400 officially designated subjects of interest, or ‘topics’, machine learning assigns appropriate topic categories to users based on their search and visit history. If a user frequently visits travel-related websites, travel will be on the user’s list of topics. Topics are only kept for 3 weeks, after which they are deleted. Of course, categories related to personal information such as gender and race are not collected. When the publisher calls the Topics API, three topics are randomly selected and shared. Google ensures user anonymity by only sharing general interests rather than analyzing granular user behavior patterns. Beyond that, the Topics API gives users maximum control. You can see which topics have been assigned to you, delete topics you’re not interested in, or even disable the Topics API feature altogether. These features help safeguard your privacy while allowing publishers and advertisers to continue personalizing digital ads to new users.
Google also offers a **Protected Audience API** that allows advertisers to serve remarketing using custom audiences. To enhance user privacy, audiences will be generated and stored directly on the device or browser, eliminating the dependence on unique identifiers or third-party cookies currently in use. To delve into how Protected Audience API works, advertisers and publishers (DSPs) will predefine interest groups along with specific behaviors (like add to cart, and purchase) for their campaigns. When a user triggers the action, the Protected Audience API is called, adding that user to the custom audience. When a user visits an advertising site (SSP or the site itself) or a publisher app later, the Protected Audience API is called, and ad bidding is conducted based on the previously created audiences to serve the most relevant ads to the user. This enhances the safety of remarketing since the device or browser, rather than a third party, possesses information about the user's interests. Moreover, ad bidding is executed on the device without any external leakage.
### Measure digital ads - Attribution Reporting API
Traditional Advertising IDs and third-party cookies have also played a key role in measuring ad performance. These identifiers allow you to match users who were exposed to your ads with users who generated conversions within your web/app. It gives you a bird's-eye view of the effectiveness of campaigns on various channels and the number of users they brought in. To help you with this aspect, Google offers an alternative way to measure ad performance with **the** **Attribution Reporting API**.
Here’s how the Attribution Reporting API works: When a user views or clicks on an ad within a web/app, the Attribution Reporting API stores the corresponding touchpoint event. If a conversion occurs, such as when the user installs an app or makes a purchase on the web, the conversion event is also saved. The Attribution Reporting API then associates the last touchpoints that occurred before that conversion to calculate attribution. This looks similar to the traditional last-touch attribution model, but the difference is that the process of storing touchpoints and attributing them **all takes place within the user’s device or browser.**
Once attribution is calculated within the device (or browser), the Attribution Reporting API provides two types of reports: Event-level reports and Summary reports. These reports are delayed and are sent with a certain percentage of noise included in the data to protect user privacy.
1. **Event-level Report**
The event-level report is a report provided for ad optimization purposes by measuring performance for each conversion event. In this report, you can check ‘which conversion event occurred by which campaign in which app’ through the three pieces of information below.
- Destination - Advertiser app or mobile web address where the conversion event occurred
- Attribution Souce ID - ID with campaign information (channel and campaign information)
- Trigger Type - Types of conversion event, such as app install or purchase
[Google Privacy Sandbox: How to tackle new challenges]
2. **Summary Report**
Summary reports aggregate the performance of multiple ads, enabling analysis from various perspectives. Aggregation Keys, which are ad campaign and conversion-related criteria, can be defined in advance on the publisher and MMP sides. Ad performance then can be analyzed based on that information.
There are two main components of Summary Reports.
- Aggregation Key - Information related to Ad interaction, such as campaigns, ad groups, creatives, targeting regions, etc
- Aggregation Value - Aggregated information corresponding to conversion events, such as the number of purchases and purchase amounts
[Google Privacy Sandbox: How to tackle new challenges]
For instance, by configuring Aggregation Key items on the publisher side, such as campaign, ad group, creative, and targeting region, and on the MMP side, items corresponding to conversions like product category, device information, and app version, you can view results like 'A total of 300 purchases were made by A campaign on B app, and the total purchase amount was 5 million won. Half of the purchases were in the Seoul area.'
However, as you can see from the name of the report, you cannot receive a report for each event; instead, reports are generated after events with the same properties accumulate to a certain level. Therefore, in order to create the largest possible parameter, it is recommended to compose the Aggregation key with general items.
## How to prepare for the changes ahead
We have already ceased collecting iOS advertising identifiers through Apple’s App Tracking Transparency (ATT) framework. Adapting to new changes, as we learned from that experience, is never easy but challenges can create new opportunities. In the case of Google’s Privacy Sandbox, alternative suggestions were released beforehand, and industry players are improving the project based on beta test feedback. We anticipate a more stable and smooth transition this time as we have been preparing in advance.
### What should advertisers know beforehand?
If GAID support is discontinued on Android, we will see lower installs and conversions compared to before, as the identifier that links ad clicks and in-app conversions will no longer be available. However, as mentioned above, attributions matched through the Play Store Install Referrer will still be recorded.
If your organization is building internal user data based on the current Advertising ID, the following tasks are recommended.
1. Check the unique identifier of the user data in the internal database
2. If you are identifying users by Advertising ID (GAID), check whether there is an alternative identifier that can be used e (e.g. first-party data such as email address, User ID provided by MMP)
3. Engage in discussions with publishers and MMP to explore potential alternatives.
If Chrome discontinues support for third-party cookies, sharing attribution results for subdomains may not be possible. For example, if you are using multiple subdomains and sharing attribution results from each site, this functionality will no longer be supported. (Attribution results from site a.ab180.co will not be associated with b.ab180.co)
## Prepare for Google’s Privacy Sandbox with Airbridge
As an App Attribution Partner (AAP) with Google, Airbridge is currently beta testing the Attribution Reporting API with the Google Ads team. Additionally, we are internally evaluating various features of the Privacy Sandbox.
Furthermore, we are actively working on implementing new attribution methods supported by major publishers to enhance user privacy, such as Google’s [Google Click Identifier (GCLID) and gBraid](https://support.google.com/analytics/answer/11367152?hl=en&sjid=7903611105448176055-AP), as well as Meta’s [Aggregated Event Measurement (AEM)](https://www.facebook.com/business/help/721422165168355?id=1877298665783613%2F)). Simultaneously, we are exploring diverse strategies to address the industry’s needs as an MMP.
Airbridge is dedicated to delivering the fastest and most personalized service to the industry in alignment with the impending changes. For any updates on Google’s Privacy Sandbox and advice on your future marketing strategy, [our team of marketing experts at Airbridge is ready to assist.](https://www.airbridge.io/book-a-demo) We are happy to provide more information on the upcoming changes and insights on how to tackle new challenges.
---
## Navigating the privacy landscape of 2024 like a pro
URL: https://www.airbridge.io/en/blog/2024-privacy-landscape
Language: English
Published: 2024-01-17
The data landscape in 2024 promises a continued evolution toward a more privacy-centric approach to marketing activities. Accordingly, advertisers are preparing for upcoming changes, with the most remarkable one in 2024 being the phase-out of Google's third-party cookies.
Predicting future privacy trends with certainty is challenging, yet marketers are actively searching for ways to enhance ad efficiency and improve performance measurement. In this dynamic landscape, the key lies in accurately recognizing and rapidly adapting to changes. Explore anticipated changes for 2024, as revealed by Google and Apple, and equip yourself with effective strategies to navigate forthcoming challenges.
> ⚠️ Disclaimer: This blog post was last updated on January 2, 2024. Please note that it may contain information that requires revision in accordance with the latest announcements from Google and Apple.
## The Privacy Sandbox and Google Chrome’s third-party cookie phase-out
In 2020, Google announced a significant development in online privacy with the introduction of the Privacy Sandbox for the web, specifically in Chrome. This initiative aims to phase out third-party cookies and limit covert tracking, prioritizing the protection of individuals' online privacy. Google’s objective is to provide companies and developers with tools to cultivate thriving digital businesses in an environment that safeguards user privacy.
Building on this commitment, in 2022, Google extended the Privacy Sandbox to Android, unveiling new technology that operates without relying on cross-app identifiers, including the Advertising ID.
The primary mission of the Privacy Sandbox is to foster both the establishment of new privacy standards and the development of innovative technology that ensures the confidentiality of user information. By creating solutions that support essential ecosystem needs without relying on tracking user identifiers, the Privacy Sandbox empowers publishers and developers to offer free content and expand their businesses while preserving user privacy. This dual benefit allows users to enjoy enhanced privacy while sustaining the business of the advertising industry.
The deprecation of third-party cookies and the Advertising ID marks a significant turning point for the advertising industry. To support developers and marketers through this transition, Google is providing various stages and beta programs across the development process. Marking the start of its Privacy Sandbox project, Google disabled third-party cookies for 1% of Chrome users globally on January 4th, 2024.
For more detailed information and updates on the Privacy Sandbox, the [Chromium blog](https://blog.chromium.org/search/label/privacy%20sandbox) regularly features progress updates. Additionally, [developer.chrome.com](https://developer.chrome.com/) and [developer.android.com](https://developer.android.com/design-for-safety/privacy-sandbox) provide comprehensive insights and updates for the latest developments.
> 💡 What it means:
>
> \- **Third-party cookies**: A cookie is a small piece of data stored in the browser when a user visits a website. Third-party cookies are stored by a service that operates across multiple sites. They play crucial roles to gather data about a person’s browsing activity and how they respond to ads. Meanwhile, first-party cookies are stored by the website itself.
>
> \- [**The Google Advertising ID(GAID)**](https://www.airbridge.io/glossary/gaid-google-advertising-id): A unique, user-resettable ID for advertising. GAID plays a key role to identify users on cross-app or device to be able to evaluate the effectiveness of the ads.
### Google Chrome’s test to block third-party cookies in Q1 2024
Chrome has officially joined the ranks in deprecating third-party cookies. As part of this transition, [Chrome introduced an experimental phase](https://developers.google.com/privacy-sandbox/blog/cookie-countdown-2023oct?hl=en), wherein third-party cookies will be deprecated for 1% of Chrome users globally during Q1 2024. This initiative aims to provide developers with an experimental label for independent testing and simulation of third-party cookie deprecation. And starting from Q3 2024, Chrome will progressively reduce support for third-party cookies. This phased approach allows developers ample time to prepare for the impending deprecation in accordance with guidelines provided by Google. During this period, developers and marketers can preview and assess how their website’s functionality will operate without third-party cookies, gaining valuable insights to adapt and optimize it accordingly. This deliberate and gradual process ensures a smoother transition, minimizing disruptions to ad performance and user experience.
### Preparing for Google’s deprecation of GAID
The Google Advertising ID (GAID), another crucial component of the Privacy for Android, is also undergoing changes with the adoption of new technology. While Google has not disclosed the precise timing of the transition and its 2024 roadmap, it initiated the first Beta testing phase in February 2023, conducting developer previews and stable API launches throughout the same year.
The impending deprecation of GAID carries significant implications for how mobile ad performance is measured. Marketers currently rely on cross-app or device identifiers, such as the Advertising ID, to assess the impact of their advertising. In contrast to Apple, which provides advertisers with Identity for Advertisers(IDFA) under its App Tracking Transparency (ATT) policy, Google will not provide GAID, regardless of whether data collection is permitted.
However, the end of GAID does not mean measuring mobile marketing performance is impossible. Google compensates by offering additional marketing data through three APIs to support various marketing campaigns. To ensure stable operation, Google maintains collaborations with diverse AdTech companies, Airbridge is actively engaging and communicating with Google as [Google's App Attribution Partner (AAP)](https://www.airbridge.io/en/blog/airbridge-google-aap). Stay tuned for Airbridge’s other post regarding Google Privacy Sandbox.
Here are some of the key APIs provided by Google and their supported marketing activities:
[__wf_reserved_inherit]
## Apple’s SKAdNetwork and Privacy Manifest
Following the iOS 14.5 update, Apple implemented the ATT policy, reshaping the landscape of mobile ads. This policy mandates that IDFA consent be obtained in an opt-in manner during app installs. With restrictions on IDFA collection, Apple introduced SKAdNetwork (SKAN) as a new framework for measuring the performance of app marketing campaigns while prioritizing user privacy. In 2022, Apple unveiled SKAN 4.0 with several updates. To stay current with the latest iOS updates for marketers, do not miss catching up on the annual World Developers Conference (WWDC).
At WWDC 2023, Apple introduced the Privacy Manifest and Signatures for Software Development Kits (SDKs) for 2024. These changes mark a continued commitment to enhancing user privacy and security. In response to SKAN and the key privacy updates released thus far, Airbridge has been proactive in providing marketers with convenient features aligned with these developments. Beyond just features, Airbridge is committed to delivering information and updates at the fastest pace to keep marketers well-informed and adaptable in the evolving landscape.
> **💡What is SKAdNetwork(SKAN)?**
> SKAdNetwork, or SKAN, is a framework for mobile app install measurement and attribution on iOS 14+. It allows mobile marketers to measure the effectiveness of their ad campaigns by providing data on app installs and in-app events while preserving user privacy. Currently, SKAN 4.0 has been released, and Apple continuously updates it for better measurement.**
>
> 💡Find out more about SKAN in the post below:**
> \- [Post-ATT iOS UA Strategies: SKAdNetwork and More
> ](https://www.airbridge.io/en/blog/mgs-2023-post-att-user-acquisition)\- A comprehensive guide to SKAN install double-counting prevention
> \- SKAdNetwork 4.0: Major changes app marketers should know
> \- [Exploring the marketing implications of SKAN 4.0](https://www.airbridge.io/en/blog/exploring-the-marketing-implications-of-skadnetwork-4)
### Maximize your SKAN campaigns - avoid double counting and over-reporting
As of 2023, Airbridge’s analysis reveals that only an average of 30% of iOS users worldwide have granted permission for IDFA collection. Reflecting this trend, there is a growing demand for SKAN campaigns, with over 54% of app services among Airbridge users running SKAN campaigns.
However, understanding SKAN can be challenging, especially in comparison to traditional attribution models like Last-touch Attribution (LTA). With that in mind, Airbridge assists marketers by [providing detailed explanations of SKAN and offering guidance on setting conversion values](https://help.airbridge.io/en/guides/skadnetwork-overview). Additionally, Airbridge introduces the "Attribution Overlap" feature, designed to identify any overlap in SKAN attributions. This feature serves a crucial role in calibrating Cost Per Install (CPI) and Cost Per Action (CPA) by incorporating non-overlapping SKAN events into the existing mobile marketing performance. The outcome is a comprehensive view of ad performance, providing marketers with the insights needed to optimize their campaigns effectively.
### App Store’s latest privacy updates
During WWDC 2023, Apple unveiled new privacy manifests and signatures for SDKs. In December 2023, Apple further announced [privacy updates for App Store submissions](https://developer.apple.com/news/?id=r1henawx). According to the updates, Apple stated that as below:
- Starting in spring 2024, if your new app or app update submission adds a third-party SDK that is commonly used in apps on the App Store, you’ll need to include the privacy manifest for the SDK.
[Privacy Manifests](https://developer.apple.com/documentation/bundleresources/privacy_manifest_files/describing_data_use_in_privacy_manifests) refer to files that outline the privacy practices of the third-party code in an app, in a single standard format. This makes it easier for developers to get an overview of how each SDK they use handles personal information. Additionally, Signatures for SDKs are used to validate that it was signed by the same developer when adopting a new version of a third-party SDK in their app.
As each app incorporating a third-party SDK assumes responsibility for all the code associated with that SDK, this update serves to enhance awareness regarding how third-party SDKs handle data. For more detailed information and updates, [Apple’s News and Updates](https://developer.apple.com/news/) provides additional resources.
## Deliver on privacy without compromising on marketing performance
Staying ahead in the market demands a deep understanding of changes and coming up with proactive strategies to address them. In this post, we learned about the privacy trends anticipated in 2024. The next step is to discover and implement ways to accurately measure your marketing performance while safeguarding customer privacy. Here’s how Airbridge can assist you.
1. Airbridge stands as a reliable partner, committed to delivering cutting-edge technology through continuous collaboration with advertising channels, ensuring marketers are aligned with changes introduced by the Google Privacy Sandbox and SKAN.
2. Airbridge also provides a convenient conversion value setup for [SKAN 3.0 and SKAN 4.0](https://help.airbridge.io/en/guides/skadnetwork-overview). You can check Android and iOS (including SKAN) data at a glance on Airbridge’s single report to increase visibility into overall marketing performance.
3. In response to the privacy-first trend, Airbridge actively proposes innovative approaches like Marketing Mix Modeling (MMM) to better measure marketing performance.
As the privacy landscape evolves, Airbridge remains dedicated to delivering the fastest and most accurate means of measuring marketing performance. We commit to providing ongoing updates and information at an unparalleled pace. For marketers seeking solutions to deliver on privacy without compromising on performance, [our team of marketing experts at Airbridge is ready to assist](https://www.airbridge.io/book-a-demo). Talk to us to stay at the forefront of the ever-changing marketing and privacy landscape.
---
## Your key marketing approach: Unified web and app attribution
URL: https://www.airbridge.io/en/blog/web-app-attribution
Language: English
Published: 2024-01-08
The advent of mobile devices has significantly transformed the marketing landscape, with these compact and user-friendly devices becoming integral tools in our daily lives. As users seamlessly navigate through diverse devices, platforms, and spaces, accurately measuring marketing performance has become a critical task for digital marketers. However, the difference in attribution logic between the web and apps adds complexity, making it challenging for marketers to assess the effectiveness of their campaigns. This post delves into the ever-increasing importance of unified web and app attribution analysis, shedding light on its role in addressing common challenges in today's dynamic internet landscape.
## Current State of Web and App Attribution
In the early days of the Internet, user journeys were confined to the web, primarily on desktop platforms. Unique users were identified using website cookies, HTTP Referers and UTM parameters helped determine the source website and keyword from where the user originated. While cookies play a role in user identification for web attribution, their limitations, such as expiration dates and browser-dependent differences, prevent them from providing a comprehensive identity. Once a user is identified, the HTTP Referer value for Site A's URL enables verification of whether the user visited Site A before reaching your website. However, it’s important to note that the HTTP Referer can be manipulated and may be lost during user redirection. UTM parameters offer a solution to collect extra campaign information, allowing the addition of five custom campaign details to the URL. Leveraging the data from these parameters is crucial in optimizing campaigns and maximizing marketing performance.
Meanwhile, apps were born with the introduction of smartphones, tablets, and other devices. User journeys have become more complex with the mixed use of web and apps across mobile and desktop. The introduction of the app store and the necessity to analyze app installs and conversions ushered in a new phase of attribution. Unlike web attribution, where clicking a link directs users to a webpage, using apps involves redirection to an app store. This redirection creates a tracking gap in monitoring the user's journey from ad click to app open, particularly given that app stores are outside your ownership, unlike websites where direct access to and analysis of log data is possible.
App attribution tools take center stage in bridging this gap. When running an ad campaign with a link provided by an app attribution tool, you gain the ability to link and trace the user's movement from the app store to the app's opening. Identifying users engaging with various apps and web browsers on mobile is made possible through unique advertising IDs like GAID and IDFA, each assigned to a specific device. In addition to identity matching using these advertising IDs, app attribution tools employ both deterministic methods such as Google Play Referrer and Deeplink matching (providing approximately 100% matching) and probabilistic methods like fingerprinting (matching a combination of the user’s web browser information, such as IP, OS, device name, etc., with information collected from the app SDK) to track users in and out of the app.
However, concerns arose about privacy infringement regarding the advertising ID assigned per device. With the implementation of the GDPR (General Data Protection Regulation) in 2018, the industry faced a unique set of challenges in protecting user privacy while enhancing sophisticated performance tracking. In line with this initiative, Apple launched a new attribution framework called SKAdNetwork(SKAN) to replace unique identifiers and measure the performance of cross-channel platforms.
👉 [Learn more about the SKAdNetwork(SKAN) framework](https://help.airbridge.io/en/guides/skadnetwork-3)
## Observe user journeys across the web and app at a glance
### Unified Attribution Measurement for Web and App
Traditionally, marketers measured attribution separately for web and app spaces, leading to siloed attribution practices. For marketers managing both websites and apps, this has led to using multiple analytics tools to measure ad performance in different segments, even for a single campaign. When a user arrives on the web through a web campaign and subsequently installs the app, triggering an in-app purchase event, the task of identifying if this is the same user becomes challenging.
Moreover, if conversions take place from browsers other than the web campaign or from apps, they might be labeled as organic or unattributed, potentially fostering a perception of underperformance. For a comprehensive understanding of your marketing performance, utilizing an analytics tool capable of measuring attribution across both apps and the web is imperative.
> Airbridge provides both Web SDK and App SDK (Software Development Kit) to collect data from each platform. The Aribridge Identity Resolution engine allows you to combine this data and easily view app and web attribution performance based on Airbridge ID in one dashboard.
### Building a Web-to-App Experience and Measuring Performance
[Your key marketing approach: Unified web and app attribution]
Web and app marketing each have their advantages. Website marketing, drawing in more traffic and offering diverse spaces, generally proves to be more cost-effective than app marketing. However, as conversion rates and revenue tend to be higher on mobile apps than on the web, it becomes essential to encourage users to transition to the app. This can be achieved by using call-to-action (CTA) buttons on the web, prompting users to install the app or navigate to a specific page within the app.
Several Airbridge customers are leveraging contextual deep linking, a powerful feature that allows for the inclusion of custom data regarding the specific links users clicked on, the original link sharer, and additional relevant information. This capability enables the redirection of users to in-app pages seamlessly connected to the content they viewed on the web. Moreover, businesses can enhance user engagement by incorporating personalized welcome messages and promotion codes directly on the web page, tailoring the overall user experience.
By building a finely tuned web-to-app experience and integrating attribution analytics, companies can establish a more effective and targeted marketing strategy. This approach not only enhances user engagement but also enables businesses to understand and optimize the user journey across both web and app platforms.
> Airbridge offers a Smart App Banner feature on the mobile web for seamless transitions between platforms, encouraging app installs and in-app conversions. Through smart app banners, you can build an effective web-to-app attribution environment by improving app installs and in-app conversion rates for users acquired from the web
>
> 👉[Learn more about web-to-app marketing guide](https://www.airbridge.io/library/web-to-app-conversion-whitepaper)
## Key marketing approach: Unified web and app attribution
Along with changes in the era and industry, the methods of analyzing performance and marketing campaigns are continuously evolving. We are now in an age where a user seamlessly moves across various devices and platforms, creating a constant demand for solutions capable of comprehensively monitoring these user journeys. Airbridge is taking a proactive approach to tackle these challenges, prioritizing the protection of personal information and addressing the obstacles faced by marketers. Explore a single view of user journeys across web, app, and web-to-app attribution, all without any disruptions in the analysis.
---
## A comprehensive guide to SKAN install double-counting prevention
URL: https://www.airbridge.io/en/blog/skadnetwork-4-attribution-overlap
Language: English
Published: 2024-01-08
If you’re a marketer exploring iOS app marketing, you’ve likely encountered the difficulty of accurate performance measurement. This challenge arises from the changes introduced by Apple in April 2021 with the iOS 14.5 update, particularly the implementation of the App Tracking Transparency (ATT) policy. Since this pivotal shift, collecting Advertising Identifiers (IDFA) has become a more intricate task, introducing new complexities along the way.
Also, numbers show the challenges marketers face. According to Airbridge’s analysis, as of 2023, only an average of 30% of iOS users globally have granted permission for IDFA collection. This makes it challenging to accurately measure ad performance for 7 out of 10 individuals.
Airbridge offers the Attribution Overlap feature to identify any overlap in SKAN attributions. Check out this article if you are a marketer interested in seeking more accurate ad performance insights in SKAN.
> **💡**You can also use Attribution Overlap in SKAN 3.0.
> The Attribution Overlap feature is available when CSM grants access to you to change the SKAN conversion value settings page to a version that supports up to 4.0. You can see more accurate ad performance with the ability to determine double attributions on channels that only support SKAN 3.0.
## Why you need Attribution Overlap
If you utilize the Attribution Overlap feature, you can easily determine whether an event has been attributed via both SKAN and the Airbridge attribution model. The SKAN and Airbridge attribution models vary in how they measure ad performance. The two attribution models differ in terms of who attributed and whether IDFA is used.
[__wf_reserved_inherit]
Therefore, if you measure ad performance using both attribution models concurrently, you may encounter events that meet the attribution criteria for both. The event is counted as an overlap in the ad performance measured by each attribution model. In fact, the event only happened once, but the report counts as if it happened twice.
Eliminating double attributions allows for a more accurate aggregation of ad performance measured by both attribution models.
[__wf_reserved_inherit]
Events attributed with SKAN are treated as attribution overlaps when the user permits IDFA collection. The double-attributed event signifies ad performance measurable within the Airbridge attribution model independently of SKAN. Conversely, solely attributed events indicate ad performances exclusively measurable through SKAN. The count of such events is calculated by subtracting the number of attributed events in the Airbridge attribution model from the total number of events attributed to SKANs.
## Here’s how to use ‘Attribution Overlap’
CPI(Cost Per Install) or CPA(Cost Per Action) is a key metric for marketers. Attribution Overlap helps you calibrate your CPI and CPA by incorporating non-overlapping SKAN events into the existing app marketing performance, providing a holistic view of attribution results. Let's delve into an example illustrating how to assess actual ad performance by categorizing events based on the attribution model and the presence of overlapping attributions.
[__wf_reserved_inherit]
First, let’s examine a sample CPI calculated for app installs that are attributed using SKAN and Airbridge’s attribution models without accounting for double attributions. The ad spend is $1000.
[A comprehensive guide to SKAN install double-counting prevention]
All 500 app installs attributed to SKAN came from iOS users, including users who allowed IDFA collection and those who did not. CPI of $2 is the ad performance calculated based on the number of app installs by iOS users. App installs by Android users are excluded.
[A comprehensive guide to SKAN install double-counting prevention]
The 800 app installs which are attributed by the Airbridge attribution model are the sum of app installs from iOS users who allowed IDFA collection and Android users. The CPI of $1.25 is the ad performance calculated based on the number of app installs from both user groups. App installs from iOS users who did not allow IDFA collection are excluded.
When using the two attribution models independently, it is challenging to identify app installs performed by specific users based on factors such as OS and IDFA collection. Here you can facilitate Airbridge’s Attribution Overlap feature. Attribution Overlap allows you to see all app installs measured by both attribution models without excluding or overlapping events.
Consider another CPI example below, assuming that there are 400 non-overlapped attribution events for app installs attributed to SKAN. The ad spend is the same at $1000, and the app installs measured by both attribution models are combined.
[A comprehensive guide to SKAN install double-counting prevention]
The 400 duplicate unattributed app installs are from iOS users who did not allow IDFA collection. The 800 app installs, attributed to the Airbridge attribution model, are the sum of app installs from iOS users who allowed IDFA collection and Android users.
The CPI of 0.83, calculated by adding the two numbers together, reflects the ad performance based on all app installs generated through app marketing. No app installs were excluded based on the operating system, IDFA collection, or any other criteria. By leveraging Attribution Overlap, the final CPI was at least 30% and up to 70% lower than if SKAN and Airbridge attribution models were used separately.
## Set up Attribution Overlap in Airbridge right now
How do I set up the Attribution Overlap in Airbridge to obtain the correct CPI and CPA? You can set up Attribution Overlap in Airbridge in just 3 steps below.
1. Update to the version that supports up to SKAN 4.0 conversion settings page
2. Select ‘Attribution Overlap’ from Measurement types
3. Add Attribution Overlap related metrics and GroupBy in Actuals Report
To measure event data with SKAN, you need to set up conversion values. When setting up SKAN conversion values in Airbridge, you can map multiple app events and event attributes to each conversion value. Additionally, you can also map double attributions to a conversion value.
To enable mapping, update the Airbridge iOS SDK to version 1.33.1 or later and switch the SKAN conversion value setting page in Airbridge to the version that supports up to SKAN 4.0. You can ask your Airbridge CSM to grant access to the SKAN conversion settings page.
[A comprehensive guide to SKAN install double-counting prevention]
> 💡One thing to note before changing the version
> Once you switch to the SKAN conversion settings page that supports up to version 4.0, it's not possible to revert to version 3.0. However, the conversion values set in version 3.0 can still be utilized in the updated version. This enables you to continue measuring event data from ad channels that only support SKAN 3.0 by using the version that supports up to SKAN 4.0 on the conversion settings page.
On the SKAN conversion value settings page, select ‘Attribution Overlap’ under the Measurement type. The toggle activates when this option is selected.When Attribution Overlap is enabled, any in-app events mapped to a SKAN conversion value can be identified as having overlap attribution through the Airbridge attribution model. App installs are automatically included in the determination when you map a conversion value in Attribution Overlap.If you select Attribution Overlap, you can also identify overlap attribution for conversion values sent from ad channels that do not support SKAN 4.0.
[A comprehensive guide to SKAN install double-counting prevention]
👉 You can find more information about SKAN conversion settings in the [Airbridge user guide.](https://help.airbridge.io/en/guides/skadnetwork-4-settings#attribution-overlap)
After collecting sufficient SKAN data, you can review overlap attribution determinations by choosing a metric and group in the Actuals Report. You'll find SKAN metrics that explicitly differentiate between overlap and deduplicated installs.
- SKAN Overlap Installs (App): Shows the count of app installs attributed to both SKAN and the Airbridge attribution model.
- SKAN Dedup Installs (App): Shows the total count of app installs solely attributed to SKAN, excluding those attributed with the Airbridge attribution model
Additionally, there is a GroupBy option. Enabling the ‘SKAN Is Also Attributed By Airbridge(IDFA)’ GroupBy allows you to examine the overlap attribution of SKAN metrics, providing three possible values: true, false, and blank.
- true: Shows the number of SKAN metrics attributed to both SKAN and the Airbridge attribution model.
- false: Shows the number of SKAN metrics attributed to SKAN and not attributed to the Airbridge attribution model.
- blank: Shows the number of SKAN metrics attributed to SKAN but the “Attribution Overlap” component is not included in the conversion value mapping, or the install is attributed to Meta Business.
[A comprehensive guide to SKAN install double-counting prevention]
👉 You can find more information on how to retrieve SKAN data in the [Airbridge user guide.](https://help.airbridge.io/en/guides/skadnetwork-4-viewing)
## Airbridge is always delivering accurate ad performance
While the overlap attribution of SKAN events may lead to unmeasured iOS app marketing performance, the ability to more accurately assess ad performance in the era of privacy marks another milestone for marketers.
In addition to attribution, Airbridge has completed the development of the version that supports up to SKAN 4.0 functionality for marketers. Marketers can now preset SKAN conversion values within Airbridge. These preconfigured SKAN conversion values will become increasingly significant as more ad channels adopt SKAN. We are dedicated to advancing these features while closely monitoring SKAN-related trends.
Amplify the success and precision of your iOS app marketing performance with SKAN through Airbridge.
---
## Why you need postbacks and how to set them up
URL: https://www.airbridge.io/en/blog/mmp-postback
Language: English
Published: 2023-12-19
In the competitive realm of mobile advertising, advertisers and ad channels share a common objective: to reach the right users at the right moment with their ads. However, despite the shared agenda, advertisers and ad channels often find themselves out of sync - because they are looking at different user sets based on different data points. Advertisers focus on events that occur within the app, while ad channels collect information related to ad creatives and placement.
To solve this discrepancy in data collection, Mobile Measurement Partners (MMPs) play a crucial role. Serving as a bridge between advertisers and ad channels, an MMP consolidates data from various ad channels and seamlessly transfers it to advertisers. Subsequently, marketers can access this data through MMP's reports or as raw data.
However, what about the reverse flow of data? Do ad channels have access to the advertiser's data? This is where 'postback' comes into play.
> ❗️ Postbacks occur at various points on different platforms. This blog is based on Airbridge.
## Why do you need a postback?
A postback is the name given to the process of sending ad performance and user behavior data from the advertiser's app to the ad channel.
Here are three reasons why you need postbacks for mobile advertising:
**1\. Billing**
The first reason you need postbacks is for billing. If you pay for ads based on cost per action (CPA) or cost per install (CPI), you must provide the ad channel with data - this data should include information about events like installs or purchases and whether the channel was attributed - via postbacks.
**2\. Ad Optimization**
The second reason you need postbacks is for ad optimization. Rich data supplied through postbacks can improve your ad performance significantly. Ad channels, armed with postback data, gain insights into user interactions, identifying those who clicked on ads and triggered install or purchase events. This data helps determine the channel's contribution to these actions, creating a more refined target audience. By strategically displaying ads to more responsive users, you can anticipate higher conversion rates, benefiting both the ad channel and the advertiser.
**3\. Retargeting**
The third reason you need postbacks is for retargeting. Postback data serves an integral role in forming audiences for retargeting campaigns. Postback data is especially valuable when running retargeting campaigns toward users who have installed your app or have a purchase history. The data provided by postbacks can define such an audience since you need to know whether an event actually occurred to determine who performed a specific event.
## How should you set up your postback for each purpose?
The data within a postback serves multiple purposes, and it's crucial to send postbacks with specific settings tailored to each purpose rather than a blanket approach. For example, suppose you're sending postbacks to calculate ad spend on a cost-per-install basis and transmit all events, including both attributed and non-attributed ones. In that case, the ad channel is burdened with extra work to filter and process only the attributed install events. Configuring the channel to send attributed install events exclusively can save resources for both parties and ensure accurate calculations.
Conversely, additional considerations come into play with retargeting. Channels running retargeting campaigns towards users with prior purchases should transmit all purchase event data, regardless of the channel's contribution, to guarantee a big enough audience size. If not all events are sent, the retargeting audience may be insufficient in size, potentially resulting in a less-than-ideal campaign performance.
[Why you need postbacks and how to set them up]
## Airbridge offers flexible postback configurations for ad channels
Different ad channels have different methods for sending postbacks. Self-attributing networks (SANs) determine which events to send as postbacks during integration. The SANs leverage the data sent via postbacks to optimize campaigns and enhance the precision of retargeting advertisements.
Airbridge offers several options to send desired information to the non-SAN ad channel through postbacks. Once you set up the data to be sent - including event timing, channel attribution, and device details - you select a delivery rule, and you're done.
Airbridge provides two postback-sending rules to ensure only the intended data is sent via postbacks.
> **💡Airbridge’s postback delivery rule**
>
> \- Attributed to: allows you to decide whether to send only the events attributed to the specific ad channel or send all events regardless of the attribution results.
> \- First only: allows you to decide whether to send only the events that occurred for the first time on a particular device or send all recurring events to the ad channel.
Now, how should you set up a retargeting campaign for users with prior purchases? For retargeting campaigns, it is important to ensure that your ad channel has as much data as possible on users who completed purchase events. Therefore, using a postback delivery rule to send all events, regardless of whether they were attributed or not, will enhance the performance of your retargeting campaign.
## Send postbacks to any ad channels you want
You can send postbacks to any ad channel integrated with Airbridge. If the ad channel you intend to send postbacks to is currently unavailable in Airbridge, please ask the ad channel to complete an integration request.
- [Request form](https://help.airbridge.io/en/supports/integrations)
---
## Tips on choosing the best MMP for mobile games
URL: https://www.airbridge.io/en/blog/mmp-for-mobile-gaming-apps
Language: English
Published: 2023-12-07
No matter how well-made your game is, it is meaningless if players don’t play it. How do you effectively promote your awesome game to players all over the world? What can we do to make sure these users become fans of our game? If you’re gearing up for a game launch, you should definitely consider implementing MMP (Mobile Measurement Partner) to get the most out of your marketing strategy.
## Why we need MMP for our games
In the past, running ads on a handful of major platforms and measuring their performance was enough. However, the marketing landscape has undergone a seismic shift in recent years, and surviving in the global marketplace now demands more than just a few publishers. As running ads across diverse channels, it’s essential to select the most efficient ones within a limited period. Marketers also need to pinpoint the right publishers for their target audience and app among numerous options.
To run ads across multiple ad networks and platforms, you need an App Attribution Tool / MMP (Mobile Measurement Partner). Once you install MMP’s Software Development Kit (SDK), you can manage all ad networks at once without installing each publisher’s SDK. This simplifies the process for marketers to quickly assess and analyze the performance of multiple publishers at a glance through the MMP dashboard.
[Tips on choosing the best MMP for mobile games]
The significant advantage of MMP is its ability to provide fair and accurate data from a third-party perspective, bridging the gap between the publisher’s data and the data marketers see internally. Additionally, publishers known as SAN (Self-Attributing Network), such as Apple Search Ads and Google Ads, use their own algorithms to attribute the conversions, making it challenging to compare with other publishers on the same page. MMP helps you grasp the data insights from multiple sources in one integrated dashboard.
The ideal time to implement an MMP is during the soft launch phases. You can collect all data in one place from the early stages to detect user drop-off points and proactively improve the game before the official launch. Utilize an MMP to quickly and effortlessly run ads across channels and introduce your app to as many users as possible.
> Airbridge SDK enables you to run ads across multiple channels without the need to install each publisher’s SDK separately. The Airbridge SDK's size is less than 1 megabyte(MB), ensuring it has no impact on game performance while allowing marketers to manage ad networks in one place.
## Seamlessly explore diverse data with MMP’s reports
Once you’ve brought users on board through ads, the next step is delving into an in-depth analysis of user actions within the app. This involves looking into metrics like Lifetime Value (LTV), and cohort-based churn rates. Airbridge's funnel reports help you understand various aspects, including determining where users are dropping off, spotting the quickest converting channels from app installation to item purchase, and providing a comprehensive overview of conversion rates for each funnel in the last seven days.
MMP also facilitates publisher-specific ad Return On Investment (ROI) analysis, providing visualizations of performance and cost data such as CPI, CPC, CTR, and CVR. Gather and optimize costs to conversions and revenue data for partners, countries, and platforms in one place. You can identify the specific campaigns and channels that fetched the users who like your game the most and even reduce marketing costs by disabling the less effective publishers.
> In Airbridge’s funnel report, you can explore the user journey from the app installs to events. LUFT, Airbridge’s self-developed database, allows you to quickly generate specific cohorts and analyze behavioral patterns by channel, country, period, and OS with GroupBy.
> 👉 [Learn more about funnel analysis for mobile games](https://www.airbridge.io/en/blog/funnel-analysis-for-gaming-apps)
[Tips on choosing the best MMP for mobile games]
[Tips on choosing the best MMP for mobile games]
It’s crucial to examine how long users are sticking around in your app. A Retention report provided by Airbridge offers a detailed analysis of user engagement in cohorts based on minutes, hours, days, weeks, and months. Furthermore, by segmenting user retention based on channels, campaigns, creatives, countries, etc., you can also investigate pLT (Predicted Lifetime) metrics. This sheds light on when users who launched your game typically churn, as well as patterns related to purchases and events within the app, all through data-driven conversion rate analysis.
Harnessing an MMP not only grants you insights through the dashboard but also provides the flexibility to extract raw data for further analysis. You can easily filter and load specific data into your database to combine internal data and execute retargeting campaigns. Consider creating user groups based on specific cohorts and implementing personalized marketing strategies as a new approach. People who think your brand message speaks directly to them might come back and turn into big fans of the game.
> With Airbridge’s Audience Manager, you can access and export around 220 types of log data directly from the dashboard. Additionally, you can preview the raw data set before extraction using the preview function.
## MMP’s role in the privacy-first era
The movement towards enhancing user privacy has brought instrumental changes to the marketing ecosystem. With Apple introducing SKAdNetwork (SKAN) to ensure app tracking transparency without advertising identifiers, the industry is undergoing a significant transformation for customized marketing in response to evolving privacy frameworks. MMPs adapt to these changes by interpreting SKAN conversion values received from publishers, providing aggregated data for each publisher on the dashboard. You need MMP integration to verify more precise SKAN performance for key publishers by combining GroupBys and diverse SKAN metrics.
> Airbridge supports an intuitive and easy-to-use tracking set method to help you configure conversion values easily. You can set SKAN conversions in preset form with just a few clicks.
When choosing an MMP, it’s vital to check its support for flexible SKAN conversions and the ability to combine various model types. As we expect another challenge with Google’s Privacy Sandbox, Airbridge, as an MMP, is collaborating with major partners to proactively respond to the evolving marketing landscape.
From game launch to ongoing success, MMP plays a pivotal role in ensuring your game’s marketing strategy aligns with the data-driven decision-making era. Consider factors such as the flexibility of reports, meaningful metrics in provided reports (cost data, cohorts, and retention), and overall suitability for your game. Embrace the power of MMP for a successful mobile game journey!
---
## Mastering predictive LTV for swift marketing decisions
URL: https://www.airbridge.io/en/blog/airbridge-predictive-lifetime
Language: English
Published: 2023-12-06
Navigating the complexities of performance marketing involves a constant evaluation of your expenditure against the revenue generated. A straightforward approach is to scrutinize your ad spend in relation to the income it generates. If your revenue surpasses your expenses, you’re on the right track; if it doesn't, it may be time to reconsider your strategy.
Consider this scenario: a campaign with an initial user acquisition cost (CAC) of $100 or less. If the users acquired through this campaign contribute an average spend of over $100 per month, it’s a strong indication that the campaign is worth maintaining. This strategy allows you to not only manage your ad budget efficiently but also ensures the continuation of high-converting campaigns with proven success.
However, a crucial challenge arises— not all users acquired through ads remain engaged with your service throughout the entire month. Moreover, adapting your marketing strategy swiftly is challenging, as you can only evaluate a user’s spending behavior after the fact. To overcome this hurdle, you need to anticipate how frequently ad-acquired users will return to your service and the potential value they can bring. This is precisely where predictive features come into play, it's the reason why Airbridge started supporting Predictive Lifetime and Predictive LTV in the reports.
## Predictive Lifetime and pLTV to identify users ahead of time
Accurately predicting the future may be impossible, but estimating user behavior based on comprehensive data is within reach. Through its Prediction feature, Airbridge offers a swift estimate of how often a user will return to your service and the corresponding value they'll generate, derived from the data we meticulously collect.
Two key metrics are provided by Airbridge through the Predictive feature: Predictive Lifetime and Predictive LTV (pLTV). Predictive Lifetime, available in the Retention Report, serves as an estimate of the number of days a user is likely to return to your service in the future. For instance, when the predictive lifetime of a cohort is 9.5 days, you can expect users in the cohort to return for 9.5 days on average during your specified timeframe.
pLTV, on the other hand, represents the Lifetime Value estimated from the predictive lifetime. If the calculation period is set to Day 30, and pLTV of a cohort is $300, it signifies that a user in the cohort is likely to generate $300 in revenue for 30 days.
[Mastering predictive LTV for swift marketing decisions]
However, no one can precisely predict the future. This leads to the fundamental question: How can you effectively leverage predictive lifetime and pLTV to gain a proactive edge with your users?
## Predictive metrics can give you more insight into your marketing campaigns
On the Airbridge dashboard, various metrics, including GroupBy, Filter, Cohort, and more, allow you to define user groups effectively. Tracking your campaign performance is made even more insightful with metrics such as predictive lifetime and pLTV.
### Case 1
Consider this example with three user groups, each characterized by predictive lifetime and pLTV:
[Mastering predictive LTV for swift marketing decisions]
- Group C is identified as the longest-returning user based on predictive lifetime alone.
- Group A, however, holds the highest expected value when considering pLTV.
For a service emphasizing user engagement, like a community, targeting Group C in a campaign might be beneficial. Conversely, if short-term revenue is a priority, Group A, with the highest pLTV, could be a strategic choice. Depending on your service’s conditions and current state, refining your target audience using predictive lifetime and pLTV becomes a valuable strategy.
### Case 2
Let’s see another case. Running multiple campaigns with pLTV can serve as a quick indicator of their success. Examining three campaigns, for instance:
[Mastering predictive LTV for swift marketing decisions]
- Campaign A, with a pLTV surpassing its CAC, stands out as a solid performer worthy of continuation.
- Campaign C, with a pLTV smaller than its CAC, might warrant consideration for discontinuation.
While these metrics alone may not be sufficient to make a definitive decision on campaign viability, they provide a swift reference for analyzing performance. Airbridge further streamlines this process by presenting pLTV and CAC together in the Revenue Report, offering a quicker and more comprehensive snapshot of your campaigns.
## Airbridge is here to help marketers make faster, more accurate decisions
In an uncertain future, many marketers face challenges in finding the right answers. Airbridge is dedicated to continually enhancing its features and metrics to empower marketers with swift and precise decision-making. The new forecasting feature is a testament to this commitment. Explore the guide below to experience the benefits of Airbridge's predictive lifetime and pLTV.
👉 [Check how to enable Predictive lifetime calculation on the Retention report](https://help.airbridge.io/en/guides/predictive-lifetime)
👉 [Check how to enable pLTV calculation on the Revenue report](https://help.airbridge.io/en/guides/predictive-ltv)
---
## Airbridge is now supporting Meta Install Referrer
URL: https://www.airbridge.io/en/blog/airbridge-meta-install-referrer
Language: English
Published: 2023-11-28
Airbridge’s app install reporting on Meta Android Campaigns is improving. We have been working closely with Meta on the implementation of a new solution called Meta install referrer and are excited to announce the roll out will begin by 23rd November 2023. We will now be able to share install-level reporting on additional measurement use cases by leveraging this tool.
## What is Meta install referrer?
Meta install referrer is an Android-specific measurement solution that is designed to help attribute views and clicks to the correlating app download. This means that it supports same-session click-through attribution as well as use cases that are unsupported by Google Play Install Referrer: most view-through attribution and cross session click-through attribution.
## What is the user flow?
As part of the user’s ad exposure and conversion journey, when a user views or clicks on an advertiser's app install ad on Meta's app (e.g., Facebook or Instagram), Meta's app encrypts and logs the relevant ad campaign metadata in Meta app's local storage layer on-device.
Once the resulting install completes, when the user first opens the app, Airbridge SDK will read Meta install referrer data from the local storage layer on a device and use it for attribution measurement.
## How does this help?
This is good news for advertisers, as you will now see a more complete picture of your Meta Android Campaigns performance. This is because we anticipate that the volume of unattributed installs in your reports will reduce and instead be shared as view-through or non same session click-through installs. This will be available both through Airbridge reporting and through our Raw Data feature.
## How does Meta install referrer work in conjunction with Google Play Install Referrer?
We will replicate the same process we used for Google Play Install Referrer, which involves a decryption key. If you haven’t set up Install Referrer yet, please see our instructions [here](https://help.airbridge.io/en/guides/meta-business-google-play-install-referrer). We only use one of the 3 ways to attribute across Meta install referrer, Google Play Install Referrer and Self Attributing Network claims, so it should be automatically deduplicated. So customers should be reassured that both aggregate and user-level output will not contain duplication.
## How to view the Install Referrer data in Airbridge
To enable Meta Install Referrer with Airbridge, ensure that the latest version of the Airbridge Android SDK is pre-installed in your app. Your developer must properly configure the Airbridge Android SDK by setting the decryption key to collect Meta Install Referrer data. Refer to our SDK Guide on Meta install Referrer settings [here](https://help.airbridge.io/en/developers/android-sdk#meta-install-referrer-collection-setup).
Once the SDK is set, Set up the Install Referrer Decryption Key according to [our User Guide documentation](https://help.airbridge.io/en/guides/meta-business-google-play-install-referrer). Additionally, feel free to reach out to your Airbridge Customer Success Manager (CSM) for personalized assistance. We are dedicated to helping you maximize the benefits of integrating Meta Install Referrer into your app.
> **\[FAQ\]
>
> • Is there support for iOS?**No, iOS does not have an equivalent to the Meta install referrer mechanism.**
>
> • When is Meta install referrer applicable for attribution?** ‣ Install attribution
> - Click-through attribution for all campaigns
> - View-through attribution for all Advantage+ App Campaigns and Manual App Promotion Campaigns with broad targeting (e.g. where age, gender are set to default)
> ‣ Does not include campaigns redirecting to landing pages**
>
> • How does Meta install referrer factor into Airbridge’s attribution waterfall?**Refer to our [User Guide documentation](https://help.airbridge.io/en/guides/meta-business-google-play-install-referrer#attributing-conversions-to-meta-ads) on the attribution logic.
---
## 5 Smart ways to use Mobile Measurement Partners (MMPs) to grow your mobile apps
URL: https://www.airbridge.io/en/blog/how-to-use-a-mobile-app-attribution-tool
Language: English
Published: 2023-11-24
A mobile app attribution tool or measurement partner (MMP) is a must-have if you run mobile app campaigns. It allows you to easily view and optimize the ad campaigns’ performance of multiple channels and networks at once with a single SDK.
In this post, we'll share five ways you can better utilize Mobile Measurement Partners (MMPs) for your app growth.
> **\[ Highlights\]**
>
> \- Curious about your user journey from the web to the app? Get a unified view of your web-to-app marketing performance
> \- Quickly and easily create target audiences that performed specific events to maximize conversions and LTV
> \- Build a convenient and complete data pipeline with raw data
> \- Prepare for the iOS 14.5 update and apply SKANs to tackle privacy concerns
> \- Check actual performance through last-click-based performance measurement as well as next-generation marketing performance analysis such as MTA and MMM
## Check the unified web-to-app user journey on one page
[5 Smart ways to use Mobile Measurement Partners (MMPs) to grow your mobile apps]
Just because you're running mobile app campaigns doesn't mean the web isn't important. Due to the nature of the limited app store product page, it is common to see richer content and incentives on the mobile web before a user lands in the app store. When you access the mobile web, you can easily find smart app banners that encourage you to install the app or take action in the app, such as "Get more from your app."
To understand the entire user journey from web to app, you must use a mobile app attribution tool that offers a web SDK. Most mobile app attribution tools are app-centric, meaning they can only analyze app installs and in-app events, and you'll need to utilize a separate analytics tool for the web. In this case, only in-browser conversions will be attributed to web campaigns, and any other web-originated app installs and in-app events will be analyzed as organic/unattributed, which can lead to an underestimation of the performance of web campaigns. It leads to the fact that you might not be able to see the true, complete user journey marketing performance.
After using unified web-to-app analytics, Airbridge customer [d.code](https://www.airbridge.io/en/case-studies/dcode-web-to-app-attribution), a fashion pre-order platform, found that users who acquired from web campaigns had a higher conversion rate than those from app install campaigns. Marketers could allocate more budget to web campaigns to improve conversion rates.
Leverage app attribution tools with unified analytics from web and app SDKs and see accurate marketing results.
> 👉 For more information on increasing web-to-app conversions and tracking, check out our whitepaper "[Web-to-App Marketing Guide](https://www.airbridge.io/library/web-to-app-conversion-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=web-to-app-whitepaper)."
## Maximize your conversion rate and LTV with the target audience
[5 Smart ways to use Mobile Measurement Partners (MMPs) to grow your mobile apps]
User acquisition (UA) is essential in mobile app marketing. Still, another important goal is to increase retention and maximize customer lifetime value (LTV) by ensuring that acquired users continue to act and engage in the app. One of the main strategies to achieve this goal is to run retargeting ads or remarketing for users who have taken specific actions, such as signing up as a member. You need user data who performed particular events, such as installing and opening an app or signing up. Mobile app attribution tools can help you create and export a user group that satisfies specific criteria with specific properties efficiently.
The app itself already has user behavioral data, but you might need to ask your development team to create and export the target audience. If your development team is on a tight schedule, you may have to process the data yourself. But when you use a mobile app attribution tool, user behavior data is collected and stored through SDKs, allowing you to create and export target audiences easily. You can also automatically upload your audience data with specific ad channels and third-party solutions to create custom audiences. You can run high-performing retargeting or detargeting campaigns with this target audience data and custom audiences.
Leverage first-party data from mobile app attribution tools to create target audiences and maximize user conversion rates and LTV.
## Build a convenient and complete data pipeline with raw data
[5 Smart ways to use Mobile Measurement Partners (MMPs) to grow your mobile apps]
Mobile app attribution tools collect touchpoints and conversion data via tracking links, APIs, referrers, SDKs, and more for attribution. They process and manipulate it to assign credit to the touchpoints that contributed to a conversion based on an attribution model. You can see the results in the attribution tool's dashboard and download the raw data directly for deeper analysis or data internalization.
Airbridge's customer, [Queenit](https://www.airbridge.io/en/case-studies/queenit-data-pipleline-analytics), a fashion e-commerce service, uses Airbridge to export data and build a data pipeline. After collecting data, they exported all the data to automatically load it into storage in a 3rd party platform, making it easy to build its data pipeline.
Suppose you need more in-depth analysis and data pipeline building beyond the dashboards provided by the mobile attribution. You can utilize the app attribution tool’s raw data export and integration in that case.
## Apply SKAN easily to cope with iOS 14.5 updates
[5 Smart ways to use Mobile Measurement Partners (MMPs) to grow your mobile apps]
The iOS 14.5 update introduced the App Tracking Transparency (ATT) policy, making Apple's SKAdNetwork (SKAN) a crucial solution for measuring iOS campaign performance. If the user consents to the collection of the Identifier for Advertiser (IDFA)) - ATT opt-in - you can find the attribution results as before. However, if the user does not consent - ATT opt-out - you can check the campaign’s performance only when the campaign is executed through SKAdNetwork. As of March 2022, only 6% of users have opted-in to ATT policies, making it essential to use SKAN when marketing to iOS users.
App attribution tool can help you run SKAN campaigns more efficiently by preparing for the iOS 14.5 update and SKAN utilization. It provides insights to app marketers unfamiliar with SKAN and offers a function to set the conversion value you need for iOS marketing. The app attribution tool also allows you to analyze your overall marketing performance, including performance on Android and iOS, as well as the performance of SKAN campaigns, all at once.
👉Read more about SKAdNetwork 4.0: Major changes app marketers should know
In addition, Airbridge lets you use SKAN immediately with a simple setup by installing the latest Airbridge iOS SDK. You can check the performance of Android, iOS, and SKAN campaigns at a glance in one report without any additional development. We also provide the function to set up the conversion easily with presets that collect best practices. If you need to prepare for the iOS 14.5 update and SKANs, Airbridge can help you.
## Check actual performance through last-click-based measurement, as well as integrated marketing performance analysis
[]
To analyze ad performance, you need to use an attribution model that provides a baseline of which touchpoints contributed to which conversion events. App attribution tools typically provide performance data based on a last-touch attribution (LTA) model.
It becomes more difficult to measure based on a LTA model in situations where user-level data collection is limited. Now is the time to consider multiple methods to weave a full picture of the user journey in this privacy-centric era.
Attribution tools respond to this trend by offering multi-touch attribution (MTA) as an alternative to last-touch attribution or marketing mix modeling (MMM) to analyze marketing performance properly while protecting privacy.
While the performance measures will vary depending on your business and marketing strategy, the following Unified Measurement Stack can help you analyze your performance from a balanced perspective to grow your app service and optimize your marketing. For example, MMM analysis during the marketing strategy setting stage can help guide trends and plans for overall marketing channels. When you're running ads, MTA and incrementality analysis can help you see the effectiveness of your ads across channels to verify, validate, and monitor efficiency. Lastly, you can use the analytics from the LTA model for detailed creative and campaign-level strategy.
If you were only checking the performance based on LTA in your app attribution tool / MMP and wondering if this is your actual marketing performance, check out our comprehensive marketing performance analysis with multi-touch attribution and marketing mix modeling.
> 👉 For more information on attribution models that determine ad performance, check out [Airbridge Handbook 01 - Attribution](https://www.airbridge.io/library/airbridge-handbook-attribution?utm_source=blog&utm_medium=cta&utm_campaign=airbridge-handbook-attribution).
---
## Introducing the Overview Dashboard: Track all your key metrics in one glance
URL: https://www.airbridge.io/en/blog/overview-dashboard
Language: English
Published: 2023-11-22
Are you looking for an effortless way to monitor the performance of all your app ad campaigns at a glance? Look no further – Airbridge is thrilled to introduce our latest feature, the Overview Dashboard!
Airbridge has just unveiled the Overview Dashboard, a powerful tool that enables you to easily track the key metrics of all your apps in real-time. Presented as intuitive widgets, charts, and tables, this feature allows you to gain quick insights and monitor the overall performance of either multiple apps or a single app simultaneously. With the Airbridge Overview Dashboard, optimizing your marketing performance has never been more accessible. Stay on top of your app’s ad performance and make data-driven decisions to enhance your marketing strategy.
## What is the Airbridge Overview Dashboard?
Airbridge, a mobile measurement partner (MMP), delivers a unified overview of mobile app marketing performance. To get started with Airbridge and measure your ad performance, the first step is registering your app to Airbridge. Whether you’re managing multiple organizations or apps, Airbridge can accommodate your needs. By unifying data from diverse devices, platforms, and channels, Airbridge delivers an integrated marketing performance analysis by offering a range of report types.
👉 [Learn more about Registering an App with Airbridge](https://help.airbridge.io/en/guides/register-a-new-app)
And now, the Overview Dashboard emerges as your customizable go-to interface for your app growth monitoring. It empowers you to easily track the key metrics of all your apps at a glance in real-time.
[]
### Metric Widgets
Create widgets that display the key metrics of the apps registered to the selected organization. You can also compare the ad performance of multiple apps using the same metric or assess the ad performance of a specific app simultaneously with multiple metrics. One of the most attractive features is the ability to gain intuitive insights with the rate of increase or decrease in metric values compared to the previous date range.
### Charts and tables
The Overview Dashboard presents a visual view of ad performance over a specified period through six charts and one table. Configuration options include two charts in a stacked bar or line graph format and four charts in a horizontal bar or donut chart type. All charts and tables support the selection of multiple apps simultaneously, enabling a comprehensive check of your ad performance in one view.
For a more detailed analysis of a specific metric, simply click on ‘Trend’ or ‘Actuals’ on the chart or table. This action will redirect you to either the Trend Report or the Actuals Report in a new tab, maintaining the same configuration as in the original chart or table. Here, you can further enhance your analysis by adding more GroupBys and Filters to refine your insights.
[]
## Why you need a dashboard for quick insights
The Airbridge Overview Dashboard help marketers to compare the key metrics of the app at a glance and get insights when you are running either multiple apps or one app. In particular, you can see the visualized ad performance without having to connect data to external Business Intelligence (BI) tools.
### Comparing the ad performance of multiple apps
[]
If you are managing multiple apps within a single company at the same time, the challenge of analyzing metrics intensifies due to the sheer volume of metrics and diverse situations to track. This is where tools like the Airbridge Overview Dashboard come into play. By efficiently checking the key metrics of multiple apps simultaneously, you can swiftly identify opportunities and gain insights. This allows for quick budget reallocation across campaigns and the seamless adjustment of your marketing strategy to enhance overall performance effectively.
### Monitoring the ad performance of an app intuitively
Whether you’re overseeing a single app or multiple, the Airbridge Overview Dashboard ensures you get swift insights all on one screen. Moreover, Metric Widgets provide a quick overview of an app’s ad performance, showcasing the rate of increase or decrease compared to the previous period. Customize the Metric Widgets with the key metrics you frequently monitor, making it convenient to track daily updates effortlessly.
[]
Discover more about the detailed settings by clicking on the User Guide linked below. Whether you’re keen on monitoring a single app’s ad performance or comparing multiple apps’ ad performance, the Airbridge Overview Dashboard can help you gain valuable insights and maximize your marketing performance!
[]
---
## 2023 Idle RPG Trends: A comprehensive analysis
URL: https://www.airbridge.io/en/blog/idle-rpg-trend-2023
Language: English
Published: 2023-11-08
Did you know that RPGs make the most amount of revenue in the global gaming market? Analyzing the breakdown of global mobile game revenue by genre for the period from January 1 to August 31, 2023, using data from [Sensor Tower Store Intelligence](https://sensortower.com/product/store-intelligence), RPGs emerged as the most dominant category, contributing significantly to revenue with a substantial 25.1% share. Following closely were strategy games at 20.1%, puzzles at 12.9%, and casinos at 11.8%, based on the combined data from the App Store and Play Store.
What’s particularly intriguing is the proportion of RPGs in mobile game revenue within the Korean market. A 57.5% of mobile game revenue in Korea was attributed to RPGs, significantly outpacing Japan’s 47.8%, China’s iOS market at 27%, and the United States’ at 11.3%. This highlights the immense popularity of RPGs in South Korea, making it a key player in the global gaming landscape.
In this post, we explore the latest statistics on the topic and dive into where we believe in how to level up with the insights from the data and trends.
_\*This post is authored by Airbridge with data contributions from Sensor Tower._
## What is Idle RPG?
An “Idle RPG” is a distinct subgenre of role-playing games (RPGs) characterized by its passive or automatic gameplay design. In an idle RPG, player interaction is intentionally minimized, and in some instances, not necessary at all. Instead, the game unfolds autonomously, with the player’s primary role typically revolving around initial setup, strategic decision-making, and occasional interaction with the game’s mechanics.
Idle RPGs are popular on mobile devices, primarily attracting players seeking a more relaxed gaming experience or those who appreciate the ability to advance within the game without the need for constant active participation. These games often give off a casual and soothing atmosphere while retaining captivating elements related to character development and strategic decision-making.
## The global expansion of Idle RPG
Idle RPGs have been experiencing remarkable growth on a global scale, with a particularly noteworthy surge in popularity observed in South Korea. This surge is part of a continuous upward trajectory that idle RPGs have been following over the past four years. By analyzing revenue and download data spanning from January to September for four years since 2020, a consistent pattern of growth has emerged.
_\*Note: Based on annual January-September data for comparison. The overall game market ranking is based on the number of revenue or downloads among the 86 Sensor Tower game taxonomy subgenres. The ranking of the RPG subgenres represents the ranking based on the number of revenue or downloads among the seven sensor tower game RPG subgenres._
[2023 Idle RPG Trends: A comprehensive analysis]
In the mobile gaming landscape, idle RPGs moved up the ranks, climbing from the 69th spot in downloads from January to September 2020 to the 59th position within the same timeframe in 2023. Within the RPG subgenre, idle RPGs made up 7.5% of downloads in the initial period but have since seen substantial growth, accounting for 10.4% in 2023.
[2023 Idle RPG Trends: A comprehensive analysis]
However, it’s in South Korea where the most striking growth has taken place. South Korea, a traditionally strong market for RPGs, has witnessed a significant rise in the popularity of idle RPGs. According to Sensor Tower’s data, these games have ascended from the 42nd spot to the remarkable 2nd position in downloads in the overall gaming market. Within the RPG genre, idle RPGs have seen an even more impressive increase, shifting from 4th place in 2020 to the leading position in 2023, capturing a substantial share of 28.2% in downloads. In terms of revenue, idle RPGs jumped from 16th place in the total domestic game market from January to September 2020 to a remarkable 5th position in the same period in 2023.
## Top idle RPGs of 2023
Let’s explore the most popular idle RPGs of 2023, based on cumulative revenue and downloads during the period from January to September.
### Top idle RPGs by cumulative revenue
[2023 Idle RPG Trends: A comprehensive analysis]
_\*The above revenue data is an estimate of in-app purchases (IAPs) from the App Store and Play Store, which do not include ad revenue and third-party Android markets._
The chart above displays the global ranking of idle RPGs by cumulative revenue in the first three quarters of 2023. Leading the pack is Hero Wars: Alliance and Girls Chronicle: Idle Heroine, which has consistently dominated the global idle RPG revenue charts. Hero Wars: Alliance secured the first position in 2020, the second in 2021 and 2022, and reclaimed the top spot in 2023. Girls Chronicle: Idle Heroine held the second rank in 2020, clinched the first position in 2021, and stood second again in 2022 and 2023.
Hero Wars: Alliance has impressively surpassed $1 billion in cumulative worldwide sales since its launch in 2016. The United States emerged as the largest market, accounting for 38% of cumulative revenue, followed by Japan (14.1%) and Germany (14.1%). Girls Chronicle: Idle Heroine achieved over $1.2 billion in cumulative worldwide revenue since its debut in 2017, with an overwhelming 99.9% of revenue originating from Japan.
### Top idle RPGs by cumulative downloads
[ 2023 Idle RPG Trends: A comprehensive analysis]
_\*The above download data is an estimate of downloads from the App Store and Play Store and does not include pre-downloads, re-downloads, and third-party Android markets._
On a global scale, Hero Wars: Alliance not only secured the top spot in revenue but also in downloads. In second place was Legend of Slime: Idle RPG, a Korean game released in 2022, driven by a significant uptick in downloads in January and April 2023.
This growth and trends observed in the world of idle RPGs offer valuable insights into the trajectory of this evolving genre. It’s expected to continue its ascent in popularity across various forms and platforms, solidifying its position as a formidable player in the gaming industry.
Level up your marketing with Airbridge! 👉 [Learn more about Airbridge’s support for mobile games here](https://www.airbridge.io/solutions/gaming)
> Airbridge, a Mobile Measurement Partner (MMP) for game marketing, offers you easy access to the A to Z of what gaming apps need for marketing measurement, targeting, and optimization. The features Airbridge offers include retention analysis, funnel analysis, fraud detection/prevention, and audience connection to assist gaming companies in acquiring new users and maximizing revenue.
---
## Deep dive into funnel analysis for gaming apps
URL: https://www.airbridge.io/en/blog/funnel-analysis-for-gaming-apps
Language: English
Published: 2023-10-19
In a world overflowing with gaming apps, competing for users’ attention is no easy feat. As new games continuously emerge and vanish in the market, developers keep searching for the secret to success. The answer, however, lies in one word: “users.” To excel in this competitive landscape, the developers must delve into user behavior data within gaming apps and leverage these insights for improvement.
Many game developers and publishers opt for a pivotal checkpoint in their game’s journey – the soft launch – to ensure its performance before the official release. During this crucial phase, you can measure and unveil how your players move through a sequence of predefined events, or steps, from downloading the game to progressing within your intended user flow. This strategic process is known as “funnel analysis.”
Understanding how to leverage funnel analysis for a gaming app’s growth and engaging with players is the key to unlocking success within the mobile gaming industry.
## What is funnel analysis?
Funnel analysis is a valuable tool for comprehending user behavior in your gaming app. It provides a structured approach to visualize and quantify the progression of your players through a series of predefined events or ‘steps.’ This process involves tracking user actions in depth, from a specific stage, such as downloading an app, to a desired outcome, like purchasing an item. This detailed tracking empowers you to measure and optimize conversion rates at each step of the user journey.
## Why funnel analysis is a must-have for mobile gaming apps
Funnel analysis can provide essential insights that can supercharge your marketing strategy, enhance user experience, and boost revenue. Let’s delve into the reasons why funnel analysis is crucial for your mobile game’s success:
### 1\. Enhancing marketing efficiency in the privacy-first era
In the past, audience targeting based on their behavioral data was relatively straightforward. However, with the evolving landscape of privacy regulations and growing user data concerns, access to data has become increasingly challenging. This shift has prompted a transformation of marketing strategies, moving from specific audience targeting to analyzing user data post-acquisition to identify the most effective marketing channels and campaigns.
Funnel analysis plays a pivotal role in this privacy-centric era. It helps you monitor the actions of acquired users for each marketing channel and campaign, evaluate their conversion rates, and optimize your marketing strategies. By allocating your budget to the most effective channels and campaigns, you can maximize your reach to the target audience. In the face of privacy concerns, funnel analysis becomes crucial for building a user acquisition (UA) marketing strategy and enhancing overall marketing efficiency.
### 2\. Understanding user behavior and elevating user experience
Funnel analysis offers a systematic approach to understanding user behavior within gaming apps. It provides a detailed view of how users interact with your game and progress through it step by step. You can identify which stages present challenges, consume more time, or experience higher drop-off rates. Furthermore, it sheds light on whether users progress through the game as intended.
These insights prove particularly valuable in soft launch phases before the official release. For instance, if you notice a significant drop-off rate during the tutorial phase, you can refine the tutorial or provide more precise guidance to reduce the drop-off rate. It’s crucial to remember that improving the First Time User Experience (FTUE) can ripple effect on user retention and Return on Ad Spend (ROAS). In short, funnel analysis empowers you to make data-driven improvements that enhance the user experience of your game, encouraging players to stay engaged for extended periods.
## Setting up each step of funnel analysis
To kickstart your funnel analysis, you must establish a clear purpose and well-defined goals. Start by asking yourself what specific areas you aim to improve or understand. Your analysis objectives could range from increasing item purchase conversion rates to boosting subscription conversion rates. These goals will shape the structure of your funnel to analyze. Once your analysis objectives are set in place, the next step is to identify the key performance indicators (KPIs), data, and events for each funnel stage to label them.
For example, let’s say your goal is to examine the conversion rate of users acquired from paid ads within the past seven days and subsequently made a purchase. To optimize your performance across various marketing channels, you can structure your funnel analysis stages on Airbridge as follows:
[__wf_reserved_inherit]
[__wf_reserved_inherit]
### Points to consider when defining funnel stages
When creating your funnel stages, there are a couple of critical factors to keep in mind:
1\. **One-way progression**
It’s essential to ensure that the behavior of users moves forward through the funnel stages and does not backtrack to a previous step. This one-way progression is vital for accurate analysis. Using the earlier example, a user should only be able to purchase items after successfully signing up, driven by the game’s settings. This prevents accumulating duplicate data from the same user within a single step.
2\. **Consistent and readable names for each step**
The names assigned to each stage in your funnel must be consistent and crystal clear, ensuring every team member can interpret the results within the same context. This practice simplifies the exploration of the reports, making the data immediately clear and sensible at first glance.
Once you’ve meticulously defined your funnel analysis stages, proceed to collect the data as per your definition and embark on the analysis process.
## Analyzing data with Airbridge’s Funnel Report
When it comes to funnel analysis, you have several options at your disposal. While product analytics tools can certainly do the job, a Mobile Measurement Partner (MMP) offers unique advantages. Unlike other MMPs, Airbridge is a notable exception as the sole MMP with funnel analysis.
Airbridge, as an MMP, presents a funnel report that provides detailed analysis across the entire user journey, from ad impressions or clicks to in-app events, using the exclusive data that media channels share with their partners only. In particular, you can quickly create specific cohorts based on Luft, Airbridge’s in-house OLAP database. With the Funnel Report, you can understand when users tend to drop off and how the user journey varies by country or OS platform.
👉 [For more information on the Funnel Report offered by Airbridge, check out our guide.](https://help.airbridge.io/en/guides/funnel-report)
### Benefits of funnel analysis with MMP
The standout advantage of employing an MMP for your funnel analysis lies in its ability to access a holistic view of advertising data, including that from Self-Attributing Networks (SAN). Only MMPs can use the data from SAN, exclusively through partnership, combined with their data to provide a complete picture of the user journey. In practical terms, even when product analytics tools and MMPs are integrated, product analytics tools fall short in analyzing customer journeys, including attribution data from SAN. Airbridge can bridge this gap by providing comprehensive funnel reports as an MMP.
For mobile game marketers, this means an opportunity to craft a more efficient marketing strategy by leveraging segmented analysis conducted through the essential solution MMP.
If you want to improve your marketing strategies on mobile gaming apps with funnel analysis, [get in touch with us today](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=funnel-analysis-for-gaming-apps)!
---
## Airbridge is now a badged Meta partner for Marketing Mix Modeling
URL: https://www.airbridge.io/en/blog/airbridge-marketing-mix-modeling-meta-business-partner
Language: English
Published: 2023-10-12
Airbridge, a unified measurement platform for mobile marketing, today announced it has been named a Meta Business Partner in Measurement for marketing mix modeling. With its third Meta measurement badge, Airbridge demonstrates its expertise and experience spanning marketing mix modeling (MMM), multi-touch attribution (MTA), and mobile measurement.
Earlier in May 2023, Airbridge launched [MMM Studio](https://www.airbridge.io/mmm-studio) to enhance the accessibility of MMM. This machine learning-powered platform enables brands to evaluate the impact of various marketing channels, both online and offline, on their business outcomes. A privacy-centric solution that does not rely on user-level data, MMM Studio also factors in variables such as seasonality, promotions, and competitor activity, offering a holistic view of marketing effectiveness. Automated onboarding, highly configurable models, and an interactive dashboard are some other features that allow for convenient and data-driven marketing optimization.
In addition, by predicting and simulating the performance of each marketing channel, MMM Studio eliminates guesswork and suggests the optimal budget allocation. With actionable insights on hand, brands can swiftly adapt to changing market conditions.
“Navigating the era of privacy, brands are eager for unparalleled measurement insights,” said Roi Nam, CEO and Co-Founder of Airbridge. “We’re pleased to be able to provide our customers with a complete understanding of their marketing performance as Meta’s MMM partner. Airbridge’s innovative, integrated approach to measurement will empower them to achieve the best possible results in the dynamic digital landscape.”
**About Airbridge**
Airbridge is a unified measurement tool that helps leading brands discover and scale their true sources of growth. The Airbridge Unified Measurement Stack integrates Last-Touch Attribution (LTA), Multi-Touch Attribution (MTA), Incrementality measurement, and Marketing Mix Modeling (MMM), enabling mobile marketers to effectively optimize ad performance and drive maximum return. Airbridge also offers solutions for deep linking, marketing analytics, audience management, and fraud protection. For more information, visit [https://www.airbridge.io](https://www.airbridge.io).
---
## Achieve analytical excellence through enhanced Airbridge integration with Unity Ads
URL: https://www.airbridge.io/en/blog/airbridge-unity-ads-integration
Language: English
Published: 2023-09-13
To help you navigate the dynamic landscape of mobile marketing, Airbridge has made powerful development to deliver actionable insights. We’re delighted to announce an upgrade to the integration of Airbridge and [Unity Ads](https://unity.com/products/unity-ads), the leading ad monetization platform that allows apps to attract the right users and maximize revenue.
While it has long been possible to track the performance of Unity Ads campaigns on Airbridge, it required some extra efforts to analyze with costs in consideration. Following the integration, it is no longer necessary to wait for data imports or labor through manual calculations, and you can **take immediate actions based on real-time cost and performance data**. The convenience of accessing all necessary data without leaving the Airbridge platform enables you to fine-tune your strategies on the go, ensuring every adjustment leads to optimal results even under fast-changing circumstances.
Another significant advantage is the ability to **assess the cost effectiveness of your marketing activities**. With both cost and performance at your fingertips, you can easily track your return on ad spend (ROAS) and return on investment (ROI), gauging the true impact of your marketing activities on your bottom line. Consequently, you get to allocate your limited resources into the campaigns or channels that yield the best returns. This additional layer of context also adds **precision, transparency and accountability to decision-making process**, making it easier to prove your contribution to overall business outcomes.
The data pulled from Unity Ads can be analyzed on Airbridge using metrics such as clicks, impressions, and cost. You can then filter the data by campaign, ad group, ad creative, device country, and many more. For detailed information on how to get started, refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/unityads).
This cost data integration is an enhancement that not only streamlines the analytical process but also deepens your performance analysis. If you would like to find out how Airbridge and Unity Ads can help you optimize for success, [get in touch with us today](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=airbridge-unity-ads-integration)!
---
## MGS 2023 Recap: Improving mobile engagement and retention
URL: https://www.airbridge.io/en/blog/mgs-2023-mobile-engagement-and-retention
Language: English
Published: 2023-08-18
This previous month, on July 26, [Airbridge](http://airbridge.io/) took center stage in Seoul as the host of [Modern Growth Stack 2023 (MGS 2023)](https://www.moderngrowthstack.com/), one of Korea’s biggest AdTech and MarTech conferences. This year witnessed the largest scale yet, with a lineup of 36 speaker sessions presented by 49 distinguished industry professionals from across the world, and a total of 2,000+ participants.
Every speaker came full force, discussing the most industry-relevant topics today in the mobile marketing landscape. In his presentation titled, **“Improving Mobile Engagement and Retention”,** Andy Carvell, Co-Founder & Partner at [Phiture](https://phiture.com/), a mobile growth consultancy, also shares his invaluable tips from a mobile marketer’s standpoint and takes us through the step-by-step process for mastering user retention.
## User retention is the backbone of mobile marketing
Engagement represents the level of interaction and interest users show within an app based on the actions they take. It can be measured by tracking multiple metrics like the frequency, length, and number of actions performed during an in-app session for each user. On the other hand, retention refers to how often users come back to do a certain activity in the app, and it is measured in percentages to evaluate the different levels of engagement each cohort displays within a given period of time. These two concepts are highly correlated and work together to bridge the app and its users throughout their lifetime as customers. Today, mobile apps are all about nurturing a lasting relationship to form a loyal user base, and maintaining a steady user retention rate is one of the most fundamental and sustainable ways to lead your app growth in this direction. While the concept sounds fairly simple, carrying out a successful retention solution requires marketers to plan strategically. They should thoroughly segment and analyze user behavior in order to guide the audience through each stage in the user lifecycle.
## Our main task: Segmenting, cohorting, and measuring
Carvell shares that many companies overlook retention problems due to a lack of segmentation and understanding of their users. Because user acquisition can occur at any point in time within a marketing campaign, retention growth cannot be linear. Therefore, segmentation is key to building a dataset that accurately reflects the variables in retention rates, and different measurement tools can be used based on cohort trends.
### Cohort-based measurements
When utilizing the Monthly Active Users (MAU) graph (Figure 1.1) to evaluate user retention, you should break it down into cohorts based on the time that users have been engaged with your app. When grouped by cohorts (Figure 1.2), you can notice that the seemingly increasing graph is actually signaling a decay, with each cohort’s retention trending down to zero. A cohort table (Figure 1.3) is another way of looking at retention, especially for comparing the retention of different groups over time.
[__wf_reserved_inherit]
[__wf_reserved_inherit]
Due to the unpredictable nature of user behavior, a cumulative MAU value alone cannot be an accurate indicator of user retention. Instead, dividing the graph into cohorts and honing in on the details for each variant is vital for sustainable growth.
### Natural usage frequency
Measuring the natural usage frequency of your users is another valuable segmentation tactic, especially for re-engagement. In order to set realistic goals for your campaign, you should first ask yourself, “How many days within a given month will the average user actually want to engage with my product?” Depending on the type of application, usage frequency may inevitably be lower than others, but this doesn’t necessarily mean that it has a lower value. For example, the usage frequency of a music streaming app would be higher than a document scanning app as the latter is only used for particular occasions. Using a histogram to plot the frequency values, you can understand how much time an average user is willing to spend on your app and craft re-engagement strategies accordingly.
### But, correlation ≠ causation
There are a number of other variables you can use to break down your retention chart and micro-analyze user data, including acquisition source, geographic location, and language. Placing these filters in your chart can help you gain a better understanding of the factors that are most highly correlated with retention, which you can refer to when setting targeting parameters.
However, it is also important to keep in mind that correlation does not equal causation. Just because a certain acquisition channel produced high retention doesn’t mean that you can push users from other channels in the same direction and expect it to automatically improve your results. You should consider the preferences and traits of the cohorts and allocate the most viable option for each.
## Walking users down the ‘spin cycle of engagement’
[__wf_reserved_inherit]
Lifecycle marketing, or as Carvell puts it, the spin cycle of engagement, is at the core of user retention and is beneficial for guiding users through a step-by-step process for acclimating to your app. By going full circle and monitoring users from the moment they click on an ad to the moment they churn, marketers can also gain valuable opportunities to gather additional user data. At each stage, marketers should respond to the shifting user’s intent, and Carvell shares his insights on how to strategically map out the journey.
### 1\. Personalizing the user journey
The user journey should start even before a user installs the app. The moment a user clicks on your ad or performs an organic search on the app store, you can grasp their intent and traffic source to offer them the most personalized journey possible. If there is a specific keyword or feature they searched to land on your app, you could tweak the custom product page to serve a landing page design that emphasizes those words. By having multiple variations of the app store landing page design and offering the most relevant content on a user basis, you can bring prospects to relate to your app on a personal level and better address their needs.
### 2\. Finding ways to make onboarding fun
Onboarding is a critical component of the user retention process, and the same message and value proposition you displayed during the acquisition process should be carried over to this stage. Consistency raises credibility and reassures the newcomers that the app’s benefits still align with their intent. It is extremely important to reiterate your value proposition early on and secure your user base in the initial stages, as a majority of the users churn within the first session if the actual in-app experience does not meet their expectations.
Once you are able to get the users to stay, you can offer them in-depth information about the app, such as educational guidance introducing the app’s features. Here, you should find a way to deliver these messages with fun and interactive content that doesn’t bore them out. Furthermore, don’t be afraid to ask more questions about the users. Collecting more information about the users through in-app surveys will let you enhance the rest of their experience by customizing subsequent interactions.
### 3\. Activation: Securing your community of users
Having a community of loyal users that strongly advocates for your app can be a powerful driver of engagement and retention. You should leverage this opportunity by first, putting forth the effort to cultivate a solid user base that you wish to support. Offering them VIP treatment with exclusive benefits such as special gifts or early-access previews will encourage users to maintain their active participation and contribute to your app’s growth. Building a separate platform for users to gather and proactively engage on customer support issues or anything app-related can also raise the positive experience and create a supportive cult that spreads the word about your product.
### 4\. Commitment: Increasing user engagement under sustainable growth
As a mobile marketer, you need to stay on your toes and ensure that users aren’t bored out or becoming inactive. The activation stage involves maintaining engagement and making sure a user’s usage level is ongoing. To do so, push notifications are one of the most effective messaging tools to take advantage of. Sending out personalized messages that align with a user’s in-app progress can accurately trigger users to return to your app. Reminding users about abandoned carts, or giving new recommendations based on their search history are some helpful tactics to use.
Making sure your app is also up-to-date by consistently rolling out new features is also a driver of user retention. If the features on your app are short-lived, this strategy can be useful for keeping your app on top of the user’s mind and preventing them from churning.
## Remarketing lapsed users: How SoundCloud thrived in the reactivation stage
Before Phiture, Carvell had extensive experience working with data-driven marketing at SoundCloud, where he also led successful remarketing campaigns. He introduces the final step of lifecycle marketing by sharing his story of re-engaging inactive SoundCloud users.
Depending on the app, the definition of a “dormant user” may vary. SoundCloud set different tiers of inactivity by identifying the number of days each user stayed dormant. The re-engagement messages were dependent on these inactivity levels to ensure that the right nudging strategy was used. Users that had never played a track were suggested a playlist, while users with a streaming history would receive more personalized push notifications with specific artist names. Implementing these strategies, Carvell was able to bring an average of 20% and 55% uplift in re-engagement for non-activated and lapsed users, respectively.
## Final thoughts
When it comes to mobile marketing, understanding the user’s psychology, needs, and frustrations at each stage of their lifecycle is vital for success, and to do so, engaging with the users on a regular basis is a must. Marketers cannot underestimate the power of organic growth, and building a sustainable lifecycle marketing program that brings out the app’s potential to the fullest is of the essence for long-term user retention.
> 💡 Andy Carvell is the Co-Founder & Partner at [Phiture](https://phiture.com/), a mobile growth consultancy that works with B2C apps to shape innovative mobile growth strategies. He is a seasoned mobile marketer who has over 20 years of experience working at companies like Nokia and SoundCloud, and his expertise lies in consulting brands to build personalized and multi-channel marketing user journeys. For more information, [connect with Andy on LinkedIn!](https://www.linkedin.com/in/andycarvell/)
---
## Post-ATT iOS UA Strategies: SKAdNetwork and More
URL: https://www.airbridge.io/en/blog/mgs-2023-post-att-user-acquisition
Language: English
Published: 2023-08-17
This previous month, on July 26, [Airbridge](http://airbridge.io) took center stage in Seoul as the host of [Modern Growth Stack 2023 (MGS 2023)](https://www.moderngrowthstack.com/), one of Korea’s biggest AdTech and MarTech conferences. This year witnessed the largest scale yet, with a lineup of 36 speaker sessions presented by 49 distinguished industry professionals from across the world, and a total of 2,000+ participants.
Among the number of impending marketing issues and trends that were discussed, Performance Marketing & Growth Consultant **Marcus Burke** and Airbridge’s CEO, **Roi Nam**, shared their tips on how to excel as a mobile marketer amidst the momentous shift into a privacy-centric marketing era.
Below is a recap of the key highlights from their presentation titled, “**Post-ATT iOS UA Strategies: SKAdNetwork and More**”.
## Entering Apple’s privacy-centric ecosystem
Currently, any Apple devices that run on iOS 16.1 are eligible to use the SKAN 4.0 model. While many advertising platforms have yet to migrate to the newest version, most devices are ready. As of May 2023, [81% of iPhone users worldwide have adopted iOS 16](https://developer.apple.com/support/app-store/) on their devices. On the contrary, user-level data is becoming more scarce, as the overall ATT opt-in rate worldwide is at [a mere 25% as of April 2022](https://www.statista.com/statistics/1234634/app-tracking-transparency-opt-in-rate-worldwide/). Double opt-in rates, needed for last-click tracking, are far lower. With such disproportionate metrics and privacy regulations tightening by the minute, many marketers have been left to rethink their operational procedures and adopt new measurement strategies.
## The potential setbacks of SKAN & ATT
Burke claims that mobile marketers are no longer in a position to follow a single source of truth. Until now, they have been putting all their eggs in one basket when it comes to attribution modeling, solely relying on the last-touch attribution technique to define traffic sources. Moving forward, Burke says that marketers need to take a multifaceted approach to data collection and find ways to adopt privacy frameworks like SKAN while working with new measurement techniques like incrementality testing, extrapolation from subsets, and Marketing Mix Modeling (MMM). However, the constant rollout of privacy updates and measurement strategies has been overwhelming for many traditional marketers, and the change seems to be daunting, with multiple obstacles to navigate.
### Lack of fresh data from postback delays
SKAN 4.0, similarly to SKAN 3.0, applies a postback delay which restricts marketers from receiving instant feedback on conversion events. Ranging from 24-72 hours, the delays are randomized and marketers could be receiving user activity data anytime in between this time window. With time being of the essence in the fast-paced mobile environment, the lack of fresh data with postback delays is a hindrance for marketers. Now, there are certain limitations to making real-time changes in strategies and campaigns for user optimization.
### Privacy thresholds (a.k.a. crowd anonymity tiers)
Privacy thresholds, also known as crowd anonymity tiers in SKAN 4.0, require apps to reach a certain number of installs per campaign in order to track data on a granular level. Ranging from tier 0 to 3, each tier value requires apps to meet a set level of performance to access sufficient data.
### Limits to value extrapolation
Before ATT and SKAN, user-level data was an indispensable part of mobile marketing, as it enabled marketers to provide their prospects with the most relevant and personalized content that maximized conversions. However, with the new privacy implementations, data is now siloed, meaning that they are no longer able to put together pieces of user-level data to identify conversion behavior. Instead, marketers need to rely on comparing aggregate data to identify big-picture trends and create gross strategies that may not be customizable on a user basis.
### Less room for trial and error in creatives
Burke describes the pre-ATT creative testing process as “throwing spaghetti at the wall,” since marketers could easily test out multiple creatives until their algorithm determined the most optimal one. The deprecation of user-level data has put complexities to creative testing, with less room for trial and error and the need for a user-centric approach in campaign building.
## Combating challenges, embracing changes
There are ways to overcome these limitations, and Burke offers tangible and actionable solutions that correspond to each area of difficulty that can be implemented in response to the dynamic circumstances.
### Patience is key with postback delays
With SKAN postbacks, it is crucial to assign a reasonable time frame for data collection. Postback delays may tamper with marketers’ patience, but using short-term aggregate data can cause greater challenges as user behavior is extremely volatile. Keep in mind that accurate decisions will require a significant amount of data. Meanwhile, most app platforms like App Store Connect provide detailed trend visualization features that can be accessed to monitor day-to-day changes and a gradual data compilation.
### Consolidate your ad accounts with mindfulness
Consolidating the app’s ad account by focusing on a fewer number of campaigns that produce high amounts of installs may be a strategic way to meet SKAN’s privacy thresholds. While this ensures that marketers gain a wider reach and more installs, they should cut down wisely. Having fewer campaigns often means to broaden the targeting parameters and give ad platforms too much optimization authority. Since ad platform algorithms don’t always know the actual business goal of an advertiser, Burke recommends using manual bidding instead of automated bidding upon consolidating ad accounts. He also emphasizes the importance of defining specific targeting parameters when placing bids and guiding algorithms in the direction of the optimization goal. That way, marketers can allocate resources to the right group of users and maximize efficiency with aggregate data.
### Engage data from multiple sources for precise extrapolations
Data can be collected from multiple sides of an app, with the two most essential ones being SKAN data and data collected within apps. Not only should marketers collect as much information as possible, but they should also try to create as much overlap between the two datasets to get a fuller picture of the traffic and refine user acquisition strategies.
With SKAN data, marketers can discover the activation events that are driving the highest amount of long-term engagement and tie them with trial start events. This increases the chances of first-time users converting into paid users, moving them down the funnel. Product data can be collected during the onboarding process, in which marketers can place in-app surveys like the example below, asking user-related questions such as their demographics, traffic source, and interests.
By mirroring the two, it is easily understandable where the majority of the traffic is coming from and the traits of the users in this cohort. Essentially, marketers can extrapolate information that is equally elaborate and valuable as metrics derived from granular-level data.
### Adopt a user-centric approach for creative experimentation
With fewer opportunities to experiment and receive feedback under SKAN and ATT, marketers need to gravitate towards a user-centric mindset for delivering creatives, Burke says. The absence of granular data requires them to conduct thorough audience research and set the right angles for their creative from the start. Understanding the audience’s perspective and measuring on-platform metrics like clickthrough rates and ad views is one of the most fundamental approaches to optimizing creatives under SKAN. Tactics like (1) creative testing on Android, (2) staging the roll-out of new creatives to observe downfunnel trends, and (3) sending a qualified trial event, can help mitigate risks when exploring new creative concepts. Burke recommends to send the qualified trial event when users have started a trial and triggered a meaningful activation event to distinguish higher and lower intent trials.
## Airbridge’s action plan for post-ATT marketing
[Airbridge Dashboard]
As one of the world’s leading mobile measurement partners (MMP), Airbridge remains updated with the rapidly developing industry as we constantly innovate and improve our tools to reflect the change. In particular, Nam elaborates on the “SKAN Is Null Conversion Value” feature on Airbridge’s dashboard that uses SKAN postbacks to track campaign conversion rates. He explains how this feature assists the account consolidation process by easily determining campaigns marketers should focus on to meet privacy thresholds and access user-level data.
> 👉 If you’d like to learn more about how Airbridge can help you master your app marketing strategies under SKAN 4.0 & ATT, [request a demo now!](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=mgs-2023-post-att-user-acquisition)
## Final thoughts
[__wf_reserved_inherit]
On a finishing note, Nam and Burke reiterate the significance of adapting to these changes and remaining proactive to face the future. SKAN 4.0 is rolling out as we speak, and the mobile marketing industry is progressively shifting gears to become privacy-focused. In the end, SKAN and ATT are pushing marketers in the right direction by removing the technicalities and heavily equation-based marketing procedures. This hands-on approach will eventually lead them to improved results, which is why Burke finally advises that marketers need to “embrace these new functionalities because they are the future, and they can be beneficial.”
> 💡 Marcus Burke is an industry expert based in Berlin, Germany that works as a Performance Marketing & Growth Consultant. He has over 11 years of experience as a full-stack mobile marketer for subscription apps and games. For more information, [connect with Marcus on LinkedIn!](https://www.linkedin.com/in/marcusburke/)
---
## Hard Paywall vs Soft Paywall vs Freemium: Which Converts Best?
URL: https://www.airbridge.io/en/blog/hard-vs-soft-paywalls
Language: English
Published: 2023-08-17
Hard paywalls convert at 10.7%. Freemium converts at 2.1%. Case closed?
**Not quite.** If hard paywalls were universally better, 90% of App Store apps would not still be using freemium. The headline number hides a more important question: which model is right for _your_ app — and how do you know if it is working?
**Key Takeaways**
- **Hard paywalls convert 5x better than freemium (10.7% vs 2.1%)** — but they self-select for high-intent users. The conversion rate reflects who stayed, not who left.
- **Soft paywalls balance reach and revenue** — metered access or feature-gating lets users build habits before paying, which can improve long-term retention.
- **Freemium dominates by volume** — 90% of App Store apps use it. It works when free users generate value beyond subscriptions (network effects, word-of-mouth, adjacent revenue).
- **Fitness apps face a unique challenge: highest trial-to-paid (35.0%) but lowest first-renewal retention (30.3%).** The paywall model directly shapes which users enter the subscription funnel.
- **Airbridge Core Plan shows trial-to-subscription conversion rates by channel** — so you can validate whether your paywall model is converting across all acquisition channels, not just in aggregate. Start with 15K free attributed installs.
## Hard Paywalls Convert 5x Better — So Why Do 90% of Apps Use Freemium?
[RevenueCat's 2026 report](https://www.revenuecat.com/state-of-subscription-apps/) (115,000+ apps, $16B in revenue) puts the gap in stark terms: **hard paywalls convert at 10.7%, freemium at 2.1%.** The top 10% of hard paywall apps reach 38.7%.
### The Data That Seems to Settle the Debate
The numbers go beyond conversion rate. Hard paywall users generate **21% higher 1-year LTV** and **8x higher Revenue Per Install at Day 14**. After one year, retention between the two models is nearly identical — meaning hard paywalls get more revenue without sacrificing long-term retention.
### Why the Headline Number Is Misleading
**Hard paywalls do not convert better — they filter harder.** Users who hit a hard paywall and leave are never counted. The 10.7% conversion rate reflects the behavior of users who chose to stay despite the gate. That is survivorship bias, not proof of superiority.
Freemium tells a different story when you look at the timeline. **23% of freemium conversions happen 6+ weeks after download.** These are users who would never have paid on Day 0 — they needed time with the product before committing. A hard paywall would have lost them entirely.
**The real question is not which model has the higher conversion rate. It is which model captures the right users from your specific acquisition channels.**
[__wf_reserved_inherit]
[__wf_reserved_inherit]
## Three Paywall Models, Three Different Bets
### Hard Paywall: Pay Before You Play
The user must subscribe before accessing any content. No free tier, no trial browsing, no limited access.
**When it works:** When core value is immediately clear. Headspace progressively tested locking content from 20% free down to 100% locked — and saw **double-digit conversion lifts at each stage**. Users encountered more paywall touchpoints, and the concept was familiar enough from Spotify and Netflix that friction was accepted.
**The tradeoff:** You lose every user who needs to experience the product before paying. For acquisition channels that bring discovery-driven users (TikTok, Instagram), hard paywalls can filter out users who would have converted after engagement.
[__wf_reserved_inherit]
### Soft Paywall: Try Some, Pay for More
The user accesses limited content for free. Premium features, advanced content, or extended usage requires a subscription. Variations include metered access (X free sessions/month), feature-gating (basic free, advanced paid), and time-limited trials.
**When it works:** When free usage builds a habit that makes premium compelling. **Strava** offers free GPS tracking and basic activity logging — but locks training plans, route planning, and detailed analytics behind a $11.99/month subscription. Users build a running habit with the free tier, then upgrade when they want deeper insights. Result: 180M registered users, $415M revenue in 2025.
**The tradeoff:** Requires more product complexity — you need to design both a free experience worth using and a premium experience worth paying for. If the free tier is too generous, users never upgrade. If too restrictive, it feels like a hard paywall with extra steps.
[__wf_reserved_inherit]
[__wf_reserved_inherit]
### Freemium: Free Forever, Premium Optional
The core product is free. Premium features exist but are not required for the primary use case.
**When it works:** When free users generate value beyond their own subscription. **Nike Training Club** eliminated its $14.99/month premium tier entirely in 2020 — making all workouts free. The strategy: use the free app to build brand affinity and drive downstream sportswear revenue. **MyFitnessPal** uses freemium with 220M registered users and 30M+ monthly actives — monetizing through both premium subscriptions ($9.99/month) and the massive free user base.
**The tradeoff:** [Conversion rates are low](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) — 2.1% median. You need scale to make the economics work. And the algorithm learns from your largest user segment (free users), which can dilute [ad signal quality](https://www.airbridge.io/en/blog/ad-campaign-dirty-signals-subscription-app-budgets) if conversion events are not carefully managed.
[__wf_reserved_inherit]
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|Test which paywall model converts best — with real channel-level data.%%
## Why Fitness Apps Face a Unique Paywall Decision
### The Fitness Paradox: Highest Trial Conversion, Lowest Renewal Retention
[Adapty's 2026 benchmark data](https://adapty.io/state-of-in-app-subscriptions/) reveals a pattern unique to Health & Fitness:
- **Trial-to-paid conversion: 35.0%** — highest of any app category
- **First-renewal retention: 30.3%** — lowest of any category (Utilities leads at 58.1%)
- **Install LTV: $1.20** — also highest
**Fitness apps are great at getting users to pay. They are terrible at keeping them.** This means the paywall model does not just affect initial conversion — it shapes the quality of the subscription cohort that enters. A hard paywall that converts 10% of high-intent users may produce better 12-month revenue than a soft paywall that converts 20% of mixed-intent users — because the hard paywall cohort renews.
[__wf_reserved_inherit]
### Annual Plans Dominate — But Only If Users Commit Early
Health & Fitness is **the only app category where annual plans dominate revenue** — 68% of H&F revenue comes from annual subscriptions (RevenueCat 2026). This is significant because:
- Annual subscribers retain at **19.9% after Day 380** (vs monthly at 14.2%, weekly at 5.5%)
- [50% of all paid conversions happen on Day 0](https://www.businessofapps.com/data/app-subscription-trial-benchmarks/)
- Users who complete **2+ workouts during trial** are far more likely to convert
**The paywall model determines how quickly users reach the conversion moment.** Hard paywalls force an immediate decision — which works for users arriving from high-intent channels (Search, ASO). Soft paywalls and freemium let users experience workouts first — which may convert more discovery-channel users (TikTok, Instagram) but delays the conversion window.
> [**Which channels convert through your paywall — and which ones don't? See trial-to-subscription rates by channel with 15K free attributed installs.**](https://coreplan.airbridge.io/waitinglist/c)
## How to Choose: A Framework Based on Your App's Value Pattern
There is no universally correct paywall model. But there are three factors that narrow the decision:
**1\. Time-to-value: How quickly does the user experience core value?**
- **Instant** (guided meditation, single workout class) → Hard paywall works. The value proposition is clear before the gate.
- **Progressive** (fitness tracking, habit building, personalized plans) → Soft paywall or freemium. Users need time with the product to understand what they would pay for.
**2\. Network effects: Does user value increase with more users?**
- **Yes** (social features, leaderboards, community challenges) → Freemium builds the network. Strava's free tier fuels the social graph that makes premium valuable.
- **No** (individual content consumption, guided programs) → No need to subsidize free users. Hard or soft paywall captures more revenue per user.
**3\. Revenue model: Is subscription the only revenue source?**
- **Subscription-only** → Hard paywall maximizes revenue per install. Every free user is a cost center.
- **Hybrid** (ads + subscriptions, commerce + subscriptions) → Freemium funds the free tier through adjacent revenue. Nike Training Club and MyFitnessPal both monetize free users.
**Whichever model you choose, the decision is a hypothesis — not a conclusion.** The same paywall performs differently across acquisition channels. A hard paywall might convert well for Google Search users (high intent) but poorly for TikTok users (discovery-driven). Without channel-level measurement, you are testing blind.
## Choosing a Paywall Without Channel Data Is Guessing
### What to Measure When Testing Your Paywall Model
Paywall A/B tests run at the app level. But the results vary by acquisition channel:
- **Trial-to-subscription rate by channel** — the primary indicator. If Meta converts at 8% through your hard paywall while Google converts at 25%, the paywall is not the problem — the channel fit is.
- **Time-to-conversion by channel** — how many days from install to subscription. If TikTok users take 14 days while Search users convert on Day 0, a hard paywall penalizes TikTok disproportionately.
- **CPS by channel over time** — track weekly for 4-8 weeks after a paywall change. Watch for divergence between channels, not just aggregate improvement.
**This measurement requires an** [**attribution system**](https://www.airbridge.io/en/blog/why-mobile-attribution-tool) **that connects paywall events to subscription outcomes at the channel level.**
### How Airbridge Core Plan Connects Paywall Decisions to Channel Performance
Core Plan tracks Start Trial and Subscribe as [standard events](https://www.airbridge.io/en/blog/maximizing-subscription-ltv) with attribution across Meta, Google, Apple Search Ads, and TikTok. The Actuals Report breaks down trial-to-subscription conversion rates by channel — **so you can see whether your paywall is converting differently across channels, not just in aggregate.**
With native [RevenueCat](https://www.airbridge.io/en/blog/abr-revenuecat-integration) and [Adapty](https://www.airbridge.io/en/blog/airbridge-adapty-integration) integration via S2S, subscription events flow into the attribution system automatically. **Before changing your paywall model, this data diagnoses which channels have the widest conversion gap. After the change, it validates whether the new model improved CPS across all channels or just some.**
### Airbridge Core Plan vs Traditional MMP
[__wf_reserved_inherit]
## The Paywall Is a Hypothesis. Channel Data Is the Proof.
Hard, soft, or freemium — each model attracts a different user profile from each acquisition channel. **The paywall that converts best in aggregate may be losing your highest-value channel.** The paywall that looks worst overall may be your best performer on the channel that drives annual subscribers.
The only way to know is to measure the full funnel — install to trial to subscription — by channel.
> [**See which channels convert through your paywall — and which ones don't. Start with 15K free attributed installs on Airbridge Core Plan.**](https://coreplan.airbridge.io/waitinglist/c)
---
## Top 3 paywall builder solutions for subscription apps
URL: https://www.airbridge.io/en/blog/top-3-paywall-builder-solutions-for-subscription-apps
Language: English
Published: 2023-08-02
With the exponential growth of the subscription app market, competition within the market has intensified, prompting subscription apps to focus on user acquisition strategies. However, focusing solely on acquiring new users can be futile if you do not optimize paywalls in your post-install funnel. In fact, studies reveal that a staggering 80% of purchases in subscription apps occur at the initial paywall display, highlighting the importance of paywalls for revenue generation even before letting users use your app.
In our previous blog post, we delved into the intricacies of [paywall design](https://www.airbridge.io/en/blog/perfect-mobile-paywall) and [optimization](https://www.airbridge.io/en/blog/mobile-paywall-optimization-actionable-tips). Today, we shift our focus towards introducing you to a range of software-as-a-service (SaaS) solutions that streamline the process of building and managing paywalls, making it a hassle-free endeavor.
## Building paywalls: Coding it yourself vs. SaaS solution
When it comes to building a paywall for your app, you have two major options: coding it yourself or leveraging a SaaS solution. One of the major challenges in coding it yourself is the burden of expending resources to develop and design. This becomes a significant problem if your team lacks such resources or needs to move fast as an early-stage startup.
Another challenge is A/B testing. Paywalls require heavy and periodic A/B testing for CTA buttons, pricing tables, quotations (e.g. social proof) and more. To add, re-designing, re-coding, and re-deploying these A/B testing variants at a high frequency is no easy feat for small teams with tight resource constraints.
## Leveraging SaaS Solutions: Streamlining Paywall Building and Optimization
Due to the pain points stated above, utilizing SaaS **paywall builder solutions** to minimize technical resources and instead diverting resources to product development is a more prudent practice. These tools do a lot more than just create paywalls:
**1\. No-code Drag & Drop Paywall Builder**
A user-friendly, drag and drop paywall builder is provided. Some even offer paywall templates or libraries, enabling you to design and implement stunning paywalls.
**2\. Easy payment infrastructure implementation**
Connecting and managing payment systems can be a challenging task for app developers. However, these solutions simplify the process by integrating a purchasing infrastructure and implementing in-app purchases through software development kit (SDK).
**3\. A/B testing without app release**
A/B testing is a must for subscription apps, as even small improvements can accrue a lot of revenue upside. Typically, conducting A/B testing on your own requires releasing a new version of your app and obtaining approval from the app market for each test. However, by utilizing these solutions, you gain the ability to control and modify your paywall without the need to release a new version every time.
**4\. Paywall & Subscription data Analytics**
Within these solutions’ dashboard, you can conveniently access and review the outcomes of your A/B testing and delve into subscription-related metrics. This allows you to validate your hypotheses and make data-driven decisions for your next iterations.
**5\. 3rd party integrations**
Similar to other SaaS products, they seamlessly integrate with a diverse range of third-party solutions. By leveraging collaborations with partners such as Product Analytics platforms or Mobile Measurement Partners (MMPs), you can establish robust data pipelines for comprehensive analytics.
## Three prominent paywall builder solutions to consider
In the realm of paywalls, numerous solutions exist. However, for today's discussion, we will focus on three prominent options: Adapty, Purchasely, and Superwall.
### Adapty
[Source: adapty.io]
[Adapty](https://adapty.io/) offers powerful features to analyze and enhance in-app subscriptions for developers and marketers. Adapty's documentation reveals a startling statistic: a mere 5% of users convert to subscribers. By leveraging Adapty, you can tap into the potential of the remaining 95% and unlock your app's maximum revenue potential.
With Adapty's paywall builder, creating paywalls and A/B testing becomes faster and easier. It allows users to configure paywalls, design visually, and copy them for seamless A/B testing. Additionally, the remote config feature allows instant changes to JSON file elements, reducing A/B testing time without app store review.
Adapty offers more than just a no-code paywall builder. It provides comprehensive analysis of subscription-related metrics, enabling you to evaluate crucial factors such as Monthly Recurring Revenue (MRR), Average Revenue Per Paying User (ARPPU), and many other subscription and trial metrics based on various criteria. With Adapty, you can also compare revenue and advertising spend by cohort to refine your app marketing strategies. Additionally, Adapty facilitates optimization of the user journey through funnel reports, Lifetime Value (LTV) reports, and even LTV prediction utilizing the power of machine learning.
[Source: adapty.io]
### Purchasely
[Source: purchasely.com]
[Purchasely](https://www.purchasely.com/) positions itself as “The app monetization tool that is beloved by every team”. Every element of mobile app monetization and revenue growth fit together in its modern console.
Purchasely boasts one of the most robust and extensively adaptable no-code paywall builders, equipped with a diverse range of fully customizable pre-designed templates, a native, real-time preview of your paywall. This enables users to create and iterate visually appealing paywalls in a matter of minutes, regardless of their limitations in development and design resources. The paywall configuration can be optimized for various devices, including tablets and TVs, in addition to mobile devices. Some other unique benefits of Purchasely include the easiness to build and deploy onboarding screens offering a diverse choice of actions other than to purchase, easily define and target user segments for delivering personalized experience, and collect live in-app transactional and user behavior events. Purchasely also offers a built-in A/B and multivariable test feature for paywall design and price experiments. All test elements, such as the design, plans, prices, deployment location, and target audience, can be effortlessly configured without the need for any coding.
### Superwall
[Top 3 paywall builder solutions for subscription apps]
[Superwall](https://superwall.com/) places significant emphasis on the efficiency and simplicity of constructing and evaluating paywalls. It offers a remarkably adaptable no-code paywall editor that enables users to swiftly and effortlessly build their paywalls using a drag-and-drop interface. Users can conveniently configure trigger events to display the paywall and gain insights into various metrics through A/B testing, presented in numerical and graphical formats.
Where Superwall distinguishes itself from other solutions is in its billing structure. Rather than implementing a fixed monthly fee, Superwall charges based on the number of paywall conversions. If cost is a crucial factor in your selection process, you have the option to compare the rates of Superwall with other solutions to determine which offers greater affordability for your app.
## Finishing up
The diverse availability of subscription management and paywall builder solutions is indicative of the significant value placed on constructing, managing, and optimizing paywalls. This underscores the challenging nature of crafting an effective paywall that successfully converts users into subscribers. Therefore, it is essential to approach this task with efficiency and efficacy.
Expedite your success by seeking a solution that will enable you to surpass competitors in the rapidly expanding subscription app market.
---
## Scale your game by integrating Airbridge with GameAnalytics
URL: https://www.airbridge.io/en/blog/airbridge-gameanalytics-integration
Language: English
Published: 2023-06-26
We are thrilled to announce that Airbridge is now integrated with GameAnalytics, the trusted tool that empowers game developers with smart, data-driven decisions.
By combining powerful game analytics capabilities with advanced attribution tracking, this integration provides a holistic view of your game’s performance and marketing efforts. You can gain deeper insights into player behavior, optimize user acquisition campaigns, and drive success in the highly competitive gaming industry.
The setup is effortless. Select the games for which you want the attribution data in GameAnalytics and paste the Game Key in Airbridge. Before long, Airbridge will start sending your install attribution data to GameAnalytics in real time. You will then be able to identify which channel, campaign, or ad group should receive the credit and thus, determine where to invest more to effectively scale your game.
Please refer to the [Airbridge User Guide](https://help.airbridge.io/en/guides/gameanalytics) for detailed instructions and information. If you would like to find out how Airbridge and GameAnalytics can help you unlock the full potential of your game, [get in touch with us today](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=airbridge-gameanalytics-integration)!
## About GameAnalytics
GameAnalytics is one of the leading analytics tools for games, trusted by 100K+ game developers. With powerful and privacy-first features, GameAnalytics helps the world’s leading studios to rapidly refine gameplay, improve retention, track monetization and improve all sorts of KPIs with real-time insights for every major game engine, platform, and operating system.
---
## Mobile paywall optimization: 5 actionable tips to increase subscription revenue
URL: https://www.airbridge.io/en/blog/mobile-paywall-optimization-actionable-tips
Language: English
Published: 2023-06-13
The subscription app market has witnessed remarkable growth in recent years. In 2021, [global in-app subscription revenues](https://www.statista.com/topics/9684/subscription-apps/#topicOverview) surged to an impressive $18.3 billion, up by 41 percent year-on-year. With an immense financial potential underscored by multiple figures, subscription-based monetization has become a top priority among mobile marketers, especially due to the need for multiple revenue streams during the economic downturn.
However, profitable markets attract new entrants. To succeed in this highly competitive landscape and capitalize on a lucrative revenue stream, you need effective paywall optimization strategies. That is why we have got insights from [Steve P. Young](https://www.linkedin.com/in/stevepyoung/) of [App Masters](http://appmasters.com).
Without further ado, let’s dive right into the five tips that will help you skyrocket your subscription app revenue!
## Tip #1: Increase exposure to the paywall
The paywall serves as the gateway for users to access premium content or features within a subscription-based app, meaning that enhancing paywall visibility is crucial to initiate the user’s journey toward subscription conversion. Without users seeing the paywall, they would not have the opportunity to evaluate the subscription offering.
In fact, [French publisher AOC](https://voices.media/four-tips-to-optimise-paywall-conversion/) increased exposure to the paywall by 64% and saw its click-through rates jump by 329%. Ideally, you should aim at a **paywall visibility rate of 100%**, ensuring that all users encounter the paywall at least once during their app usage.
Here are some tactics for higher paywall visibility:
- **Contextual triggers:** Trigger the paywall at appropriate moments within the user journey. This could include when users access gated features or when they reach a point where the value of the premium content or services becomes evident. This approach ensures that users encounter the paywall when they are most likely to consider subscribing.
- **Strategic placements:** Place the paywall in strategic locations where users naturally focus their attention, such as the home screen or settings menu. This provides consistent visibility and serves as a reminder of the subscription offering without being intrusive. However, be cautious not to obstruct essential app functionality or compromise the user experience.
By implementing these proven strategies, you can maximize your chances of being noticed and converting users into paying subscribers.
## Tip #2: Build an effective app onboarding process
The onboarding process plays a pivotal role in shaping users’ initial impressions and setting the stage for their ongoing engagement with your app. Those who end up buying tend to make their minds up even before using your app, as illustrated by a case study in which the majority of users committed to in-app purchases during onboarding.
Hence, it is necessary to pique users’ interest from an early stage by clearly communicating the unique value proposition of your subscription. Highlight the benefits, exclusive features, or premium content users can access only by subscribing.
You can also provide users with guided tutorials or interactive prompts to help them navigate your app’s features and see the subscription as a worthwhile investment. Users may be more inclined to convert if they have already invested time and effort into exploring your app.
A smooth, frictionless sign-up process is another factor that could reduce the risk of drop-offs and motivate users to take the desired action, or the final step of subscribing. Minimize the number of required fields and consider offering social login options to enhance convenience.
An important point is that the **paywall should be placed at the last part** of the onboarding process. Give users enough time to engage with your app, understand the value of the subscription, and build trust before introducing the paywall. This will help users feel more confident in their decision to subscribe.
## Tip #3: Test different calls-to-action on the paywall
Tests allow marketers to make data-driven decisions and optimize for the best results. This is no exception when it comes to building a mobile paywall.
An essential paywall element you can play around with is the CTA button. Its color, size, and placement can all be modified, but the language holds the utmost importance. Try **variations of wording and messaging** that emphasize the benefits, exclusivity, or urgency of your offering and observe how they influence user behavior.
For instance, [an app saw changes in its subscription revenue](https://www.linkedin.com/posts/stevepyoung_appmarketing-appgrowth-activity-7056597728964874240-obLN?utm_source=share&utm_medium=member_desktop) by creating different versions of the CTA. The test, conducted on iOS and Android users in the United States, the United Kingdom, Germany, France, and Australia, saw positive and negative outcomes. Below is the aggregated impact across both platforms and all countries:
[__wf_reserved_inherit]
As Steve P. Young says, sometimes, the smallest changes can make the biggest difference. You need to keep testing to find the CTA button that drives the highest conversion rate.
## Tip #4: Craft the optimal pricing plan
Creating a pricing plan aligned with user preferences and market dynamics is crucial to boosting subscription revenues. By offering different options and monitoring user responses, you can gauge how price impacts users’ willingness to subscribe, upgrade to higher tiers, or leave.
Differentiate the following elements to find the best combination:
- **Billing periods:** weekly, monthly, annual, or lifetime
- **Intro offers:** free trial, pay-as-you-go, or pay-upfront
- **Pricing structure:** flat-rate, tiered, per-unit, etc.
- **Price:** higher, equal, or lower than competitors
> 💡 [Refer to this blog post](https://www.airbridge.io/en/blog/perfect-mobile-paywall?utm_source=blog&utm_medium=link&utm_campaign=mobile-paywall-optimization-actionable-tips) for more information on mobile paywall A/B testing.
Indeed, there was a [productivity app](https://www.linkedin.com/posts/stevepyoung_which-text-increased-yearly-subscriptions-activity-7056106375633932288-nj7k?utm_source=share&utm_medium=member_desktop) that drove conversion by changing the number of plans displayed. The original paywall only showed the yearly plan, whereas the revised paywall showed both the yearly and monthly plans, as in the image below. This allowed users to see where the value was; with the two plans juxtaposed, users could not help but compare the prices and found the latter to be more cost-effective. Eventually, the app saw a 31% increase in yearly subscriptions.
[__wf_reserved_inherit]
In addition, a [photo editing app](https://www.youtube.com/watch?v=2LLokZQyCL8) generated 38 times more sales over a year through a successful pricing plan revamp.
[__wf_reserved_inherit]
The first set of measures the app took was adding more tiers, increasing the price of the monthly offer, and utilizing decoy pricing. Then, as the installs spiked thanks to some supplementary efforts, the app took the next action – bringing back the free tier, which continued to drive users. Yet, a lot of them ended up paying for the yearly offer since the price was close.
When testing pricing plans, be sure to keep track of metrics like conversion rates, average revenue per user, average revenue per plan, and churn rates to determine what works best for your app and increase your chances of converting users.
## Tip #5: Go for a longer paywall
It is commonly assumed that shorter paywalls would outperform longer ones. However, there have been multiple cases where the opposite has been true.
The logic behind this is that users are willing to read all the necessary information when they want to make a purchase decision. Thus, for revenue-boosting purposes, using a longer paywall could be the smarter move because you have to focus on those who will actually convert.
Some of the elements to put on your long paywall are:
- Headline with benefits
- Pricing table(s)
- Features table
- Testimonials & social proof
- FAQs
- CTA button
Just remember to have a floating CTA button; make it remain visible and accessible to users as they scroll down your lengthy paywall.
## TL;DR – Summing things up
Paywall optimization is essential for subscription apps hoping to get to the next level. To stand out from the competition and maximize revenue, stick to these five simple tips:
- Increase exposure to the paywall
- Build an effective app onboarding process
- Test different calls-to-action on the paywall
- Craft the optimal pricing plan
- Go for a longer paywall
Try them out today and see the results for yourself.
> 💡 Steve P. Young is an app marketing expert who has helped numerous apps unlock success. [Watch the full livestream](https://www.youtube.com/watch?v=h_9dyOe3ufA) for more insights about mobile paywall optimization, and check out other mobile marketing-related videos on his [YouTube channel](https://www.youtube.com/@AppMasters).
---
## Firebase Dynamic Links shutting down? Here’s a better option
URL: https://www.airbridge.io/en/blog/firebase-dynamic-links-shutdown
Language: English
Published: 2023-06-01
> ⚠️ Disclaimer: This blog post was last updated in August 2023. Please note that it may contain some information that has to be revised in accordance with [Google’s latest announcements](https://firebase.google.com/support/dynamic-links-faq).
Firebase, the mobile platform by Google, has been widely used by marketers and developers seeking to streamline their growth strategies. Among its array of powerful features, Dynamic Links has long served as a go-to deep linking solution in the industry.
Dynamic Links send users to specific locations within an app rather than just opening the app’s main page. By providing a seamless, consistent, and relevant journey across different platforms with these smart URLs, apps can build exceptional user experiences and win trust.
However, in a surprising turn of events, Google has recently announced the [impending shutdown of Firebase Dynamic Links](https://firebase.google.com/support/dynamic-links-faq). While marketers and developers will have time until August 2025 to migrate, this decision has prompted many to look for other solutions and rethink their app’s future.
## An alternative that surpasses Firebase Dynamic Links
[App Links](https://developer.android.com/training/app-links) and [Universal Links](https://developer.apple.com/ios/universal-links/) are some of the technologies designed to handle deep linking on mobile devices. Yet, Airbridge deep links have stronger advantages over these and others to effectively replace Firebase Dynamic Links.
### Point #1: Does Airbridge offer advanced deep linking features?
Yes, Airbridge has all the necessary tools and infrastructure to effectively incorporate deep linking into your growth strategies.
Here are some examples of what you can do with Airbridge deep links:
- **Boost user satisfaction with personalized experiences.** Airbridge deep links can include a string of parameters that reflect the user’s preferences, location, referral source, device type, and more. This gives you the ability to land users on specific pages and provide with a more bespoke app onboarding experience.
- **Bridge the gap between offline and online, web and app, and even Android and iOS.** The cross-platform compatibility allows you to embed Airbridge deep links anywhere you want, whether it be social media, emails, or in-store QR codes, to create a unified and reliable user experience.
- **Increase conversion rates through deferred deep linking,** which takes users who have not yet installed your app first to the app store and then to the intended in-app location. Such deferment eliminates the user’s hassle of post-install content searching and simplifies the path to purchase. In case the user leaves the app store without installing, Airbridge will redirect the user to the web fallback equivalent to the in-app page.
[__wf_reserved_inherit]
- **Tailor the look and feel of your deep links for brand consistency.** With Airbridge, you can customize the slug, as well as the preview title, description, and image, with just a few clicks. This helps you to familiarize your target audience with your brand and earn their trust.
[__wf_reserved_inherit]
- **Generate, manage, and export deep links in bulk** using API or Google Spreadsheet. Airbridge offers a centralized system to save your time and effort from manually creating individual deep links for each in-app page.
### Point #2: Does Airbridge have robust attribution capabilities?
Again, yes. Airbridge deep links also function as tracking links, enabling you to benefit from both the seamless redirection of deep links and the attribution capabilities of tracking links.
Airbridge, a trusted mobile measurement partner, integrates its deep linking solution with its measurement suite. By capturing data on user journeys created by deep links, spanning from the initial click to the subsequent app install, Airbridge tracks and measures campaign performance, providing you with insights necessary to optimize strategies and achieve growth.
There are even more advanced features to elevate your attribution accuracy. For instance, you can customize attribution windows to suit your needs, ranging from 5 minutes to 30 days. The Airbridge SDK parses tracking links and stores the UTM values as campaign parameters, effortlessly identifying the campaigns driving traffic to your app.
### Point #3: Does Airbridge provide integrated analytics?
Airbridge is a unified platform that combines deep linking, mobile attribution, and marketing analytics to provide a holistic view of user behavior. You have a wide variety of reports and metrics to choose from to analyze your data and gain real-time visibility into your marketing effectiveness. Hence, yes, Airbridge is the comprehensive solution you have been looking for.
We understand how difficult it is to survive in the rapidly-changing mobile industry, and our goal is to take a weight off your shoulders. Setting up Airbridge is the easiest – just integrate a single SDK for all services. You can also run Airbridge alongside Firebase as you plan for migration, and there will be SLA-backed technical support to resolve any problems you encounter in the process.
We got your back. [Request an Airbridge demo today](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=firebase-dynamic-link-shutdown) to navigate through this change and continue your mobile success.
---
## Maximize ROI in a privacy-first world with Airbridge’s MMM Studio
URL: https://www.airbridge.io/en/blog/introducing-mmm-studio
Language: English
Published: 2023-05-25
In this era of economic downturn and privacy concerns, mobile marketers face daunting challenges in measuring campaign performance. That is why Airbridge has created MMM Studio, a self-service marketing mix modeling platform.
This game-changing launch gives you the power to make confident decisions based on a holistic assessment of your campaign effectiveness. Get ready to embrace a new level of success with MMM Studio in the evolving landscape of mobile marketing.
## Accepting challenges, unleashing opportunities
Amidst the ongoing global economic recession, the importance of ROI-driven marketing has reached unprecedented heights. Businesses today must ascertain whether their digital marketing efforts actually propel them toward their revenue objectives.
Privacy regulations have posed additional challenges to mobile marketers. Limited access to data has cast a veil over their view, impeding their ability to gain the user-level granularity necessary for accurate measurement.
But as Stephen Hawking said, “[If you feel you’re in a black hole, don’t give up. There’s a way out.](https://www.theguardian.com/science/2015/aug/25/black-holes-way-out-stephen-hawking)” MMM, or marketing mix modeling, enables marketers to track and optimize while offsetting any loss against the lack of performance signals.
In fact, MMM has already been around us for decades and proved its usefulness. It was the technological complexity and the exorbitant costs involved that held it back from any wider adoption.
Airbridge, a mobile measurement partner, believed we could unlock its potential and develop it as a way to complement current measurement methods highly reliant on user-level data. With a wealth of market data, years of measurement experience, and the latest technology in our hands, we created an MMM solution that is about to turn the tide.
## MMM Studio: the future of marketing mix modeling
[Airbridge’s MMM Studio](https://www.airbridge.io/mmm-studio?utm_source=blog&utm_medium=cta&utm_campaign=introducing-mmm-studio) bridges the gap between data and strategy, allowing marketers to measure and optimize even with the demise of third-party cookies and identifiers. It eliminates guesswork with regained performance visibility and predictive ROI analysis.
Moreover, we made a deliberate choice to provide marketers with complete transparency and control over the entire process, from onboarding and modeling to reporting. Our self-service platform with three main capabilities makes MMM simpler and faster, mobile marketing easier and clearer.
### Model Configuration
[__wf_reserved_inherit]
Get started by creating models perfectly tailored to your marketing goals and business needs. All you have to do is to upload your historical data in CSV file format, and MMM Studio does the heavy lifting.
A custom marketing mix model is built and trained in under 10 minutes based on a comprehensive analysis of your ad spend by medium, ad impressions and clicks, app installs, and revenue over a six-month period. Ingest contextual data about promotion, country, and holidays for extra accuracy, and you’re good to go.
### Performance Tracker
[__wf_reserved_inherit]
Gain the most complete picture of your campaign performance across app and web, and traditionally hard-to-measure ad media such as TV, billboards, and radio. Leveraging aggregate data that remains unaffected by privacy changes, Performance Tracker helps you assess whether your marketing efforts are driving incremental growth.
Airbridge’s MMM Studio stands out in that it accurately tracks the impact of organic traffic. Visualize the trends of your eCPI, eCPA, eROAS, and eROI by channel over time to identify the real-world user acquisition costs. Filtering options at the country and OS levels also allow for targeted analysis that is beyond comparison.
### Budget Optimizer
[__wf_reserved_inherit]
Discover weekly and monthly budget plans with maximum efficiency. By putting your original and optimized ad spend in tandem, the report highlights where adjustments are needed to improve your future performance.
Powered by advanced machine learning, Budget Optimizer uses predictive capabilities to recommend the most effective media mix within your set budget. You can extract actionable insights about resource allocation and ultimately increase ROI.
### Some more advancements to come
Take note of highly anticipated features that are to be rolled out in the coming months.
- **Simulator** will allow you to experiment with multiple budgeting scenarios and view the predictive performance result using your custom models, allowing for strategic planning.
- **Insight Plus** will provide enhanced analytical capabilities, delivering information on the carry-over effect by channel, response curves by channel, and more.
With these cutting-edge additions on the horizon, Airbridge’s MMM Studio is poised to introduce an era of hassle-free, privacy-resilient mobile measurement.
## Future-proof your growth with Airbridge
Since September 2022, Airbridge has undergone an extensive beta testing phase, collaborating with a select group of partners from the gaming and commerce industries. We have worked closely with enterprises and startups alike, ensuring that our MMM solution caters to app businesses of all sizes.
Now, with great excitement, we are proud to present MMM Studio to everyone, opening up new possibilities for marketing optimization. We would like to express our gratitude to our beta test partners, who truly recognized the role of MMM in enhancing their marketing efforts and placed trust in Airbridge’s technical prowess.
Our Chief Product Officer, Hunjae Jung, says, “With a mission to help marketers around the world discover their true sources of growth, we never doubted it was our job to develop a privacy-preserving measurement solution. MMM Studio helps mobile marketers navigate the ever-changing landscape with unparalleled insights, and it is the culmination of our unwavering dedication to customer success.”
Integrate marketing mix modeling into your growth stack. Reap the power of unified measurement. Turn challenges into opportunities.
[Try MMM Studio today.](https://www.airbridge.io/mmm-studio?utm_source=blog&utm_medium=cta&utm_campaign=introducing-mmm-studio)
> **Get your copy of MMM essentials:**
> 👉 [Marketing Mix Modeling: The Privacy-First Mobile Measurement Method
> ](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=introducing-mmm-studio)👉 [Marketing Mix Modeling in Action: Tips from the Pros](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=introducing-mmm-studio)
---
## 3 must-do’s to improve your mobile user acquisition
URL: https://www.airbridge.io/en/blog/improve-mobile-user-acquisition
Language: English
Published: 2023-05-18
User acquisition stands as both the initial and ultimate objective of mobile marketing. Without an audience to showcase the value of your product or service, profitability ends up becoming an elusive goal.
The challenge is that resources are finite, prompting the need for strategic thinking. In an ideal world, you might be able to pour money into a myriad of ad campaigns, but in reality, you need to find a way to engage users within a set budget.
This is why you need to follow the three action items that are to be introduced in this blog post. To boost your marketing return on investment (ROI), optimize user acquisition, and drive growth, let’s start by getting the basics right.
## Action item #1: Measure your performance
One of the most critical things you should to do improve your return is measuring your campaign performance. You can use an attribution and analytics tool like [Airbridge](https://airbridge.io/) to gain insights about each channel, campaign, and creative.
In fact, only by having visibility into your ad effectiveness can you answer questions like the following:
- What were the factors that drove your app installs in the past two weeks?
- Which creatives attract the highest-paying users?
- Which combination of marketing channels is the most effective?
To measure performance, you first need to decide which metrics to track. Depending on the characteristics of your app, you can choose from a plenty such as cost per action (CPA), average revenue per user (ARPU), return on ad spend (ROAS), user lifetime value (LTV), impressions, clicks, click-through rate (CTR), daily active users (DAU), monthly active users (MAU), retention rate, and funnel conversion. Based on the data collected, create a dashboard that provides you with relevant insights and reallocate your budget to maximize marketing ROI.
## Action item #2: Send user behavior data to ad platforms
Sending your users’ behavioral data to ad platforms is another essential step to take to optimize your mobile user acquisition. This is because it becomes easier for ad platforms to show personalized ads to your target audience when they receive rich in-app event data. The more “hints” you provide, the more likely it is for them to find the users who would perform a desired action, such as making a purchase.
Let’s say you have an e-commerce app. In addition to install data, you can share with ad platforms who completed actions like View Product Detail Page, Add to Cart, and Purchase. You can also transfer event properties like transaction ID and product ID as well as identifiers such as advertising ID, user ID, and hashed email address or phone number.
There are three common means of sending user data to ad platforms: an ad platform SDK, a mobile measurement partner (MMP) SDK, and Conversions API (CAPI). Among them, CAPI is a server-to-server signal sending interface that allow you to send user data on sign-ups, purchases, and other actions directly to ad platforms via API calls.
[__wf_reserved_inherit]
Remember, the more granular data you provide, the more accurately and effectively the ad platform can target your audience.
## Action item #3: Prevent and detect fraud
Ad fraud is a significant concern for mobile marketers, as it can lead to wasted ad spend, decreased ROI, and damage to your brand's reputation. The three most widespread types of mobile fraud are click spamming, click injection, and SDK spoofing.
Click spamming involves generating a massive amount of fake clicks and impressions to be part of the user journeys of those who are supposed to install the app. This type of fraud can be detected by looking for abnormal click-to-install-time (CTIT) distribution, massive numbers of clicks or impressions, unusually high CTR, and unusually low conversion rate.
Click injection occurs when a fraudulent media injects a fake touchpoint immediately after identifying the install on the device but right before the first open. The fraudster will try to attribute the install to its platform by taking credit for the user action and earning the payout. To fight against click injection, you can disable postbacks for any install with CTIT less than five to ten seconds, as it is impossible to click, install, and open the app within this short time.
[__wf_reserved_inherit]
SDK spoofing involves generating and sending fake installs and in-app events by hacking the MMP SDK. Fraudsters create a fake app that can be ad-based or legitimate and pass it off as a real one to earn the payout for installs and clicks. To fight against SDK spoofing, one of the ways is to compare user IDs and transaction IDs from internal server data (e.g., ERP, CRM data) and MMP data to detect any inconsistencies.
Through [rigorous efforts to prevent and detect fraud](https://www.airbridge.io/product/fraud-protection-detection), you can protect your app from and ensure that your ad spend is used efficiently. To learn more about the concept of mobile ad fraud, [refer to our glossary](https://www.airbridge.io/glossary/mobile-ad-fraud).
## How should I get started?
Optimizing your mobile user acquisition can be challenging, but the three proven strategies can always help you drive growth. And by working with Airbridge, which has both the technology and expertise, you can jump through hoops without any hassle.
Airbridge is an MMP that can provide you with an all-in-one dashboard that includes web and app analytics, deep linking features, real-time marketing analytics, an easy-to-use audience builder, and customizable fraud protection. We make it easier for you to implement all of the three action items.
Interested? [Get a free demo from our sales team and find out more!](https://www.airbridge.io/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=improve-mobile-user-acquisition)
> The content of this blog post is based on the presentation that [Roi Nam](https://www.linkedin.com/in/roinam/), CEO of Airbridge, delivered on [‘Performance Day 2023’ event](https://omegagroup.vn/performanceday2023/) hosted by OMEGA GROUP and Omega Martech on April 21, 2023.
---
## SKAdNetwork 4.0: Major changes app marketers should know
URL: https://www.airbridge.io/en/blog/skadnetwork-4-must-know
Language: English
Published: 2023-05-04
If there is one thing that all marketers should know about iOS 14+ advertising, it is Apple’s SKAdNetwork. Also known as SKAN, this privacy-centric framework allows you to measure the success of your campaigns without relying on device-level identifiers.
SKAN is a much-needed response to growing concerns around privacy, as it does not reveal any personal information about the user. However, by putting a limit on the amount and granularity of the data shared, it poses several challenges to marketers as well.
One thing after another, the SKAN 4.0 transition is adding to the complexity. Apple rolled out the latest version of its mobile attribution solution in October 2022. SKAN 4.0 protects privacy like SKAN 3.0 but provides marketers with more information and wider options. Not yet supported by major ad networks, SKAN 4.0 will definitely see increased adoption in the near future.
To help you to get started, here are three updates you should know about SKAN 4.0.
## Not one, not two, but three postbacks for more insights
When a user installs an app or performs an in-app event due to a specific campaign, notifications are sent to an ad network in the form of postbacks.
While SKAN 3.0 had only one postback, **SKAN 4.0 offers up to three postbacks for the winning ad attribution** depending on the number of installs per campaign. These are spread out over time, across three different conversion windows. This means that with SKAN 4.0, marketers have potentially three chances to receive information on installs, in-app events, and campaign details.
[__wf_reserved_inherit]
Each of the three postbacks contains different data because they are designed to **measure different periods of time post-install**. The first conversion window spans 48 hours over days 0 to 2, the second for 120 hours over days 3 to 7, and the third for 672 hours over days 8 to 35.
Let’s take a deeper dive into what happens after an app install, starting at hour 0.
From hour 0 to hour 48, the user activity is measured for the first time. As soon as the conversion window ends, the system prepares the postback and sends it after a random delay of 24 to 48 hours. In other words, Postback 1 contains data collected during the first 48 hours post-install and arrives sometime between hour 72 and hour 96. Simultaneously, another measurement begins at hour 48 and the process continues.
## No more waiting; Take advantage of flexibility
You might be thinking that it still takes way too long for Postbacks 2 and 3 to be sent. For instance, you may wish to receive the second postback four days post-install. However, by default, the system will require you to wait at least three additional days until the end of the second conversion window. The wait would be even longer – over a month – for the third conversion window. Luckily, there is a way to work around this situation: **the lockWindow function**.
[__wf_reserved_inherit]
The lockWindow function, as the name suggests, enables you to finalize a conversion value by locking it in, allowing you to receive postbacks sooner. Although the conversion window is yet to end, the system would immediately prepare the postback to contain information based on activity up until a specific time. The postback is then sent after a random delay: 24 to 48 hours for Postback 1 and 24-144 hours for Postbacks 2 and 3. You can choose to lock in any or all conversion windows.
Think of a case where the conversion value is locked on day 5 post-install, during the second conversion window. This reduces the waiting time for the second postback, but at the same time, the marketer will lose the chance to measure or understand user behavior between days 6 and 7. He or she should, therefore, first identify the data they want and decide whether it is the right move to customize the conversion windows.
You should also remember that changes made to one conversion window do not affect another. Let’s go back to the example above. Even though the second conversion window is shortened by two days, the third conversion window would not begin any earlier than the default. Hence, as expected, Postback 3 will measure days 8 to 35.
## Say hello to a whole new type of conversion value
The mechanism of conversion values is an essential part of the SKAN framework. Postbacks include an array of data including conversion values, which offer marketers a certain degree of visibility into their marketing performance. In SKAN 3.0, a conversion value is a number between 0 and 63. On the other hand, in SKAN 4.0, a conversion value is represented as either a number or a word, depending on whether it is fine-grained or coarse-grained.
[__wf_reserved_inherit]
Fine-grained conversion values are the same conversion values we know from SKAN 3.0. They are the 6-bit value with up to 64 variables that can be set up to collect user behavior signals. Coarse-grained conversion values are the new addition that can result in one of the three possibilities: low, medium, or high, according to the type of event completed by the user. Marketers can use these 64 numbers and three words to map in-app events they wish to track and measure for campaign optimization.
In order to decide the number and the type of conversion values to include in the postback, if any, Apple relies on [crowd anonymity](https://developer.apple.com/documentation/storekit/skadnetwork/receiving_postbacks_in_multiple_conversion_windows#4103229) and assigns a postback data tier to each install. This is a four-tier system that determines the level of user privacy based on the install count. The more installs there are driven by a specific campaign, the more likely you will reach a higher tier; the higher the tier, the richer the data sent.
Because more detailed information is expressed in a fine-grained conversion value, it is only included in the first postback when in Tiers 2 or 3. A coarse-grained conversion value, on the other hand, is sent as a first postback in Tier 1. They are also sent as second and third postbacks with Tier 1 or higher.
Even if the postback data tier is higher, if a user takes no action in your app, there is no ad performance data to turn into a conversion value. Therefore, no conversion value is sent to the postback in this case.
## Get ready for the SKAN 4.0 transition with Airbridge
Airbridge, with a mission to help marketers discover their true sources of growth, makes SKAN 4.0 easier, ultimately contributing to your efforts to protect user privacy.
You can measure the type and number of events generated by their campaigns, as well as the revenue generated from those events through fine-grained and coarse-grained conversion values. You can also see if there is any double attribution of the events across SKAN and Airbridge, as well as user retention from the data sent via the second and third postbacks.
You can also view your SKAN campaign data using various metrics and visualization tools available in the Airbridge dashboard. This allows you to analyze your ad performance from multiple perspectives and utilize the results to optimize your upcoming campaigns.
Make your iOS app marketing a success with Airbridge’s SKAN solutions!
---
## Getting started with Google’s Privacy Sandbox: A comprehensive overview
URL: https://www.airbridge.io/en/blog/google-privacy-sandbox-overview
Language: English
Published: 2023-04-26
> ⚠️ Disclaimer: This blog post was last updated in April 2023. Please note that it may contain some information that has to be revised in accordance with Google’s latest announcements.
Privacy is a key factor in today’s mobile marketing world, driving changes and challenges critical to marketers and developers alike. At the forefront of this trend are Apple and Google, two of the leading ad platforms in the market. They are most notably devising ways to measure ad performance without tracking users. In 2018, Apple introduced its approach to privacy-preserving attribution, called SKAdNetwork, and has continued to update it ever since.
Google also began a beta rollout of the Privacy Sandbox earlier this year, signaling yet another huge change from the contemporary marketing measurement methods. Expected to be fully revealed by 2024, this initiative aims to create a new set of advertising standards that will enable mobile growth while protecting user privacy.
The Privacy Sandbox is expected to completely shift digital marketing. What will privacy-safe marketing look like? Will retargeting die without device identifiers? How will mobile apps have to measure ad effectiveness? Answering these questions starts with understanding Privacy Sandbox’s fundamentals.
## What does the Privacy Sandbox mean for marketers?
The gist of the Privacy Sandbox is allowing various marketing activities while deprecating the major user identifiers. While many have long been forecasting the demise of third-party cookies on Chrome, it was a different story for GAID.
Similar to Apple’s Identifier for Advertisers (IDFA), GAID is a unique device identifier that allows mobile apps to anonymously track user activity in the Android ecosystem. In order to measure ad performance, mobile measurement partners (MMPs) like Airbridge rely on these identifiers and track individual users’ in-app activities such as installs and purchases.
However, GAID is fading into history along with the advent of Google’s Privacy Sandbox. This is in fact one of the biggest differences in the privacy policies of Google and Apple. Apple, according to its App Tracking Transparency framework, does allow advertisers to access the IDFAs of users who have opted into app tracking. Google, on the other hand, will no longer provide GAID-based deterministic attribution regardless of user consent. Considering that Android is the [leading mobile operating system with a 71.8% global market share](https://www.statista.com/statistics/272698/global-market-share-held-by-mobile-operating-systems-since-2009/), the initiative is bound to have a huge impact on mobile marketing.
Google has not announced when exactly it will phase out GAID yet, but the company [rolled out the first Beta for the Privacy Sandbox on Android](https://blog.google/products/android/the-privacy-sandbox-beta-is-coming-to-android/) in February 2023. In addition, Google is revealing more and more details about the Privacy Sandbox with time.
## How can the Privacy Sandbox and SKAdNetwork be compared?
Google’s Privacy Sandbox is similar to Apple’s SKAdNetwork in that it allows mobile apps to measure the success of their marketing campaigns while maximizing user privacy. However, their ways of polishing and delivering the end result are vastly different.
In an effort to best preserve each user’s privacy, SKAdNetwork data is delayed, with the earliest SKAdNetwork postback sent approximately 24 hours after install. Similarly, the Privacy Sandbox also ensures individual users cannot be identified. This pushes Android marketers to define their campaign objectives up-front, before the ad is served, and that only such predefined datasets will become available mainly in aggregated forms.
Still, there is a bright side: these aggregated reports with richer, high-fidelity data are sent with a random delay of only one hour or even shorter. This is clearly in contrast to how SKAdNetwork provides data in a much more time-restricted manner. Moreover, to trigger the SKAdNetwork postback, the privacy threshold should be met, meaning that a marketing campaign needs to hit a certain number of daily app installs. The Privacy Sandbox does not impose such restrictions.
Instead, to minimize the risk of identifying a user, the Privacy Sandbox uses noise. A random amount of data that does not substantially affect its analysis results would be added to the aggregated reports. In other words, noise supports private measurement without preventing marketers and developers from making informed decisions.
## Does this signal the end of an era?
Fortunately, it is still possible to measure performance using the tools we are used to. For instance, the functionality provided by the Play Install Referrer will not be impacted by the Privacy Sandbox. A string of numbers will continue to be sent to the Google Play Store whenever an ad click occurs to allow attribution providers to measure the performance of install campaigns. Some other types of identifiers are also used by MMPs, which will employ all available means to keep delivering advanced measurement solutions.
Furthermore, the Privacy Sandbox proposes a number of features to support privacy-centric mobile marketing. As it will become increasingly difficult to collect data and identify target users to serve ads based on real-time trends, Google will provide a set of APIs, each designed to support different marketing activities. Some of the main ones and their objectives are as follows:
.tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-u5bg{border-color:inherit;font-family:Tahoma, Geneva, sans-serif !important;text-align:left;vertical-align:top} .tg .tg-og15{background-color:#EFEFEF;border-color:inherit;font-family:Tahoma, Geneva, sans-serif !important;font-weight:bold; text-align:left;vertical-align:top}
Privacy Sandbox Proposals
Overview
Topics API
Enable interest-based advertising without relying on user-level data
Protected Audience API
Show customized ads to users based on their app interactions without third-party data sharing; previously known as FLEDGE
Attribution Reporting API
Measure ad performance while limiting user-level information sharing through event-level reports
## How Airbridge is already in action
Without a doubt, Google’s Privacy Sandbox is a game-changer; it will be reshaping the mobile marketing landscape. Saying goodbye to GAID indicates the need to reimagine performance measurement. To stay ahead of the game, it is crucial to ponder over how to develop your marketing strategy using the latest tips and tricks from Google.
Airbridge has already set out to work on the Privacy Sandbox. As an App Attribution Partner (AAP) of Google, we are strategizing from multiple angles to improve our current attribution model. We also offer other measurement methods such as [marketing mix modeling (MMM)](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=google-privacy-sandbox-overview), presenting a more holistic view.
There are endless twists and turns, but you don’t have to pass through them by yourself. Airbridge will be at your side to navigate the future of measurement.
---
## Navigating the new normal: Advertising strategies in the post-ATT world
URL: https://www.airbridge.io/en/blog/post-att-advertising-strategies
Language: English
Published: 2023-04-19
Apple's App Tracking Transparency (ATT) policy, introduced in iOS 14, has had a significant impact on the mobile marketing landscape. It requires apps to ask users for permission before tracking their data for advertising purposes, and has led to many users opting out of tracking.
The effects of ATT have been felt across the mobile marketing ecosystem. Under ATT, marketers face two main challenges: visibility loss and signal loss. In order to effectively tackle these challenges, it is important for marketers to approach each challenge separately and develop strategies that specifically address visibility loss and signal loss. In this article, we will explore some of the strategies and technologies that marketers have used to cope with visibility loss and signal loss.
## Coping with the visibility loss
Visibility loss refers to the difficulties that marketers may have in measuring the effectiveness of their advertising channels, campaigns, and/or creatives, as the limitations imposed by ATT reduce the amount of data that can be collected. Here are some strategies to think about:
**1\. Use SKAdNetwork.**
SKAdNetwork (SKAN) is a framework introduced by Apple that provides a way for advertisers to measure the effectiveness of their campaigns without collecting device-level or user-level data.
**2\.** [**Set up and monitor metrics**](https://developers.facebook.com/blog/post/2021/04/21/apple-announces-general-availabilty-iOS-14-5-att-prompt-requirements/) **in a way that is more friendly to SKAN.**
Here are some tips to consider:
- Use ad platform metrics such as click-through rate (CTR) and cost per mille (CPM), which are more reliable and responsive. Monitor these metrics regularly to not only gain insight at the creative level, but also make quick performance assessments.
- Utilize SKAN-based installs and associated metrics such as conversion rate (CVR), cost per install (CPI), and install per mille (IPM), as SKAN-based install counts are more consistent.
- Consolidate marketing channels and campaigns and focus budget on select channels and campaigns to ensure meeting the privacy threshold. Under SKAN 3.0, 10-20 installs within the last 24 hours is considered the empirical privacy threshold. However, keep in mind that the recommended number of installs per campaign within the last 24 hours can vary by advertising platform, and the recommended number of campaigns per advertising channel can also differ.
- Set up SKAN schema to capture early conversions, as there is a SKAN postback delay. Specifically, you can set up SKAN conversion values to capture D0 metrics. Since most direct response campaigns show cumulative conversions or revenue skewed towards Day Zero or D0, capturing D0 metrics will not only allow you to act quickly, but also give you sufficient visibility to optimize your campaigns. Also, this tip will be particularly useful for SKAN 4.0 users, as SKAN 4.0 introduced the lock feature, which will allow advertisers to close the conversion window and immediately send the nth SKAN postback.
- Use the insights gained from Android and iOS interchangeably, as over-performing and under-performing creatives may be consistent across both platforms.
- Avoid over-optimization. Each ad platform has its own optimization and automated budget reallocation algorithms. Due to SKAN delays, you should expect campaign optimization algorithms to also react relatively slower. Therefore, give campaigns more time to learn and adapt and avoid too much agile reallocation of your budget.
- Lastly, understand modeled conversions on each ad platform such as Google and Meta and use them to gain supplementary information. Modeled conversions are designed to fill in the visibility gap under SKAN 3.0 and 4.0, such as unavailable creative-level analysis or missing post-install information due to the privacy threshold. Carefully read each [advertising channel's modeled conversion documentation](https://developers.facebook.com/docs/marketing-api/conversions-api) to understand how it works and use it for supplementary information in addition to metrics from mobile measurement partners (MMPs).
**3\. Consider using marketing mix modeling.**
Marketing mix modeling, or MMM, is a method of attribution that uses statistical modeling to estimate the relative effectiveness of different advertising channels based on aggregated data, even in the absence of device-level or user-level data.
**4\. Try out lift studies.**
Lift studies are experiments that measure the effectiveness of advertising campaigns by comparing the performance of a control group (who are held out by random selection to be not exposed to the ads) with that of a treatment group (who are exposed to the ads). This allows advertisers to see the incremental lift in conversions or other metrics that can be attributed to their ads.
**5\. Don't forget about Apple Search Ads.**
Keep in mind that Apple Search Ads is an ATT-friendly advertising channel. It uses a separate API called AdServices API, which provides device-level attribution information. Therefore, SKAN is not necessary for tracking attribution on Apple Search Ads.
%%CTA|Try It Free|https://app.airbridge.io/signin?from=blog-inline|See how Airbridge recovers attribution signals lost to ATT.%%
## Coping with the signal loss
Signal loss, on the other hand, refers to the decreased integrity of the user behavioral data that is sent to advertising channels, which can compromise the accuracy of ad targeting. Here are some ideas to ponder:
1. Comply with ATT and collect as many advertising IDs as possible by following Apple’s requirements and asking for users' permission to track their data. This can help maintain the accuracy of ad targeting and measurement.
2. Use additional identifiers such as hashed emails or hashed phone numbers in conjunction with Conversion API (CAPI), which allows ad servers to fetch data directly from the advertiser's server and potentially augment attributes or data that the advertiser has the appropriate rights and permissions to share, but may not have been immediately available through pixel or SDK implementations. These additional identifiers such as hashed emails or click IDs can help to increase the matching rate between advertisers and ad platforms.
3. Use re-engagement signal solutions. These methods append identifiers aggregated at the campaign or other aggregated level to the deep link URIs, allowing ad platforms to collect user behavior data after users enter the app through deep linking. Ad platforms can assess re-engagement performance of campaigns or keywords and reallocate budget or deliver ads to the most successful combinations.
## Rethink strategies and get yourself in the lifeboat
Apple’s ATT is a necessary attempt to shift away from indiscriminate collection of user data, giving users more control over their privacy. However, it does make marketers’ life more difficult as they have to rely on users granting permission for tracking.
In this time of compelling changes, marketers can use various tactics and technologies. However, it should be kept in mind that in the end, the key is to approach each issue separately and develop strategies that address the specific challenges of visibility loss and signal loss.
Airbridge is equipped with SKAN and ATT solutions as well as MMM solutions that can put you on a fast track to mobile growth while adhering to Apple’s policies. If you want to talk about your post-ATT concerns, feel free to [reach out to our marketing experts](https://www.airbridge.io/en/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=post-att-strategy)!
---
## The role of calibration in marketing mix modeling (MMM)
URL: https://www.airbridge.io/en/blog/role-of-calibration-in-mmm
Language: English
Published: 2023-04-12
Your friend’s birthday is coming up, and you decided to bake a delicious chocolate cake. Since weighing ingredients is critical to ensuring the best outcome, you have taken out a kitchen scale that, unfortunately, you have not used for years.
To ensure your scale’s accuracy, you first zero your scale and put a penny on it. It should be exactly 2.5 grams, but the weight is off. You bring another one and place it on, and again, the number reads 2.4 grams. To restore your scale’s accuracy, you replace batteries and reset the scale, and finally, the scale registers the correct 2.5 grams for a single penny.
This procedure of testing a scale with a known weight to maintain its accuracy is known as calibration. In fact, calibration also plays an important role in marketing measurement, specifically marketing mix modeling (MMM). In this post, we will look into how you should work with calibration for the best insights.
> Before moving on, check out the below posts to know MMM better:
> [👉 Your All-In-One Guide to MMM](https://www.airbridge.io/en/blog/marketing-mix-modeling)
> [👉 The Evolution of MMM: A Pathway to Privacy-First Measurement](https://www.airbridge.io/en/blog/modern-marketing-mix-modeling)[
> 👉 MMM vs. MTA: Which One Is Right For You?](https://www.airbridge.io/en/blog/mmm-vs-mta)
## How does calibration work on MMM?
MMM uses aggregated observational data, meaning that the model analyzes historical data and assumes that past patterns will repeat in the future. This is why MMM requires regular calibration. The model can reveal correlation but not causation, and the model might have been right with the past but be wrong with the future due to its negligence of changes.
As a measure to accurately reflect real-world relationships between different variables, the model is calibrated by comparing its predictions to actual outcome data from experiments, known as the "ground truth.”
A common way to determine the ground truth for the effectiveness of a marketing campaign is to conduct a conversion lift study or a holdout test. Once you have identified the ground truth, you can compare it to the projections made by the model. If there is a significant discrepancy between the model's predictions and the ground truth, you can adjust the model's parameters or variables accordingly.
## How do you find the ground truth?
There are several tools that can be used to conduct an experiment to retrieve the ground truth. The most reliable technique is the randomized controlled trial (RCT), which involves exposing a randomly selected group of users to an ad and comparing the conversion rate of that group to a control group that was not exposed to the ad.
- [Meta’s Lift Test](https://www.facebook.com/business/help/552097218528551?id=546437386202686)
- [Google’s Conversion Lift](https://support.google.com/google-ads/answer/12003020?hl=en)
However, conducting an RCT at scale takes work because it requires access to a large audience and the ability to randomly select a control group from that audience. This is something of which only the biggest ad platforms in the world are capable. Meta's Lift Study and Google's Conversion Lift are two RCT tools that can measure the incremental effectiveness of channels and campaigns.
Although an RCT is considered the gold standard in the hierarchy of evidence, there is an alternative: geo experimentation, also known as GeoLift, which tests the effectiveness of an ad in a specific geographic location and effectively measures incrementality.
- [Meta’s Open Source GeoLift](https://facebookincubator.github.io/GeoLift/)
Yet, geo experimentation does have some limitations. Most notably, it assumes that the impact of a marketing campaign is consistent across the chosen geographic location, which may not always be the case. Another limitation is that it may be difficult for smaller companies to collect data from dozens of cities or states across the nation and find out the optimum number of test locations or the ideal control and test locations.
## How often should you calibrate?
Calibration should be performed regularly to ensure that the models remain accurate over time. The frequency can vary, but when you choose to calibrate in shorter periods, your model is less likely to operate outside acceptable parameters.
Periodic calibration is essential in a rapidly changing marketing landscape where the relationships between different variables can change quickly. However, conducting actual experiments can take time. Typically, RCTs can take anywhere from two to three weeks to complete. For this reason, it's important to strike a balance between the need for regular calibration and the time and resources required to conduct experiments.
Ultimately, the key is finding a calibration schedule that works for your organization. The idea is to calibrate often enough to ensure that your models remain accurate but not so often that it becomes a burden on your resources and budget.
## How Airbridge uses calibration for the best insights
The Airbridge MMM, powered by advanced machine learning and statistical techniques, provides actionable insights at a lower cost and a faster speed so that marketers around the world can effectively find and scale their true sources of growth.
Airbridge allows its users to use experiment results for calibration, as many other marketing analytics companies recommend. These results can be compared with MMM results to identify how overestimated or underestimated the impact of different marketing activities is. In other words, you can get a more aligned and consistent understanding of your advertising impact.
However, we also know that some app developers, especially those with smaller marketing budgets, cannot afford the time and cost required to conduct experiments. Airbridge helps its users in such cases use multi-touch attribution (MTA) results, found to resemble real-world experiment results, as an alternative. Nonetheless, for the highest possible accuracy, we advise relying on experimentation.
Ready to power up your marketing measurement? [Request a demo](https://www.airbridge.io/en/book-a-demo) to learn more about calibration, MMM, and your mobile app success.
---
## Boost your subscription revenue through Airbridge integration with Adapty
URL: https://www.airbridge.io/en/blog/airbridge-adapty-integration
Language: English
Published: 2023-04-10
We are excited to share that Airbridge is now integrated with Adapty, an SDK-based service that helps mobile developers grow their app revenue through A/B testing and detailed subscription analytics.
This integration allows you to send attribution data from Airbridge to Adapty, and subscription events from Adapty to Airbridge. With such configuration, you can add an extra layer of depth to your subscription app analysis.
As Adapty tracks whether a user has signed up for a free trial or activated, renewed, or canceled a subscription, Airbridge matches these subscription events with the user attribution data. Without any additional setup, you can get a comprehensive view of both subscription and attribution data, effectively identifying the channels that contribute the most to your revenue. You can also monitor billing issues, grace periods, one-time purchases, and more.
Please refer to the [Airbridge User Guide](https://help.airbridge.io/hc/en-us/articles/16488488623897-Adapty?utm_source=Browsing) for integration instructions. If you would like to explore how to grow your subscription app with Airbridge and Adapty, do not hesitate to [contact us](https://www.airbridge.io/book-a-demo)!
## About Adapty
Adapty provides any mobile platform the ability to track in-app purchases through a low-code SDK which helps developers increase their app revenue. Adapty aggregates purchase events calculated on the user level and transforms the raw data into accessible metrics like cohort analysis, funnels, and purchase conversions. The platform also provides the ability to run and configure A/B paywall tests without needing to release another version of the app, which remains one of the main aspects contributing to the revenue increase in mobile apps.
---
## In-app ads: What mobile apps should consider to make money
URL: https://www.airbridge.io/en/blog/in-app-ad-types
Language: English
Published: 2023-04-05
As mobile usage continues to grow, in-app advertising has become an increasingly important way to monetize apps. The global in-app advertising market size, which was already [valued at USD 67.7 billion in 2018](https://www.grandviewresearch.com/industry-analysis/in-app-advertising-market), is expected to [amount to USD 312.5 billion by 2030](https://www.globenewswire.com/news-release/2022/10/07/2530136/0/en/In-App-Advertising-Market-is-Expected-to-Reach-USD-312-48-Billion-at-a-CAGR-of-17-32-by-2030-Report-by-Market-Research-Future-MRFR.html).
Despite being one of the most common ways to monetize an app, integrating ads into an app is not necessarily easy as it requires a delicate balance. Ads bring income, but they can also negatively impact the user experience if they are intrusive, irrelevant, or too frequent.
Then, how can app developers make smart moves regarding their in-app ads? As in any other problem-solving process, the key is to evaluate available options to determine what is desirable and what is not. This blog post is thus dedicated to those looking to explore different types of in-app ads and figure out what's best for them. Without further ado, let's get started!
## What is in-app advertising?
In-app advertising is a form of mobile advertising that targets users while they are using an app on their mobile device. In-app ads come in various formats, which offer different levels of engagement and interactivity with the user.
By displaying ads to users, app developers can earn money from advertisers who are willing to pay for ad space. At the same time, in-app advertising can provide a way for advertisers to reach their target audience in a more effective and engaging way, as users are more likely to engage with ads catering to their interests.
## The various types of in-app ads
Below are some of the most popular formats of in-app advertising and examine the pros and cons of each.
### Banner ads
Banner ads are small rectangular images that are usually embedded at the top or the bottom of the device screen, alongside the app's content. It is a long-standing form of display ad that many mobile users find familiar. Like a digital billboard, banner ads can either be static or dynamic.
Because they are always visible, even when users are interacting with the app, banner ads are considered a cost-effective way of communicating information about the advertiser. You can also place them on any in-app page of your choice as they do not interfere much with the user flow.
However, this implies that banner ads may not be very good at capturing users' attention. Users may become too accustomed to viewing banner ads that they might disregard them without even realizing. Hence, a great banner ad utilizes eye-catching visuals, concise copy, and a strong call to action to generate clicks.
Banner ads work well for apps with a lot of content, such as news apps, where users spend a lot of time scrolling through articles or videos. Theywork for apps that are not highly immersive or interactive as well, such as weather or finance apps, where users check the app for quick updates.
### Interstitial ads
Interstitial ads are full-screen videos or images that appear at natural transition points in the flow of an app, such as when a user completes a level in a game. Upon seeing the ad, the user can either tap on it and proceed to its destination or exit it and return to the app.
Interstitial ads are great for apps with linear user experiences, like gaming apps, fitness apps, and productivity apps. They can also work well for apps that are designed for long sessions, such as music or video apps, as users are more likely to engage with ads when they take a break from the content.
Thanks to their size, interstitial ads have the potential to deliver high impressions and click-through rates. Yet, if interstitial ads appear too frequently, they may hold back users from enjoying your app. One study found that interstitial ad viewers [spent 22% of their time looking for the X button](https://www.prnewswire.com/news-releases/new-neuroscience-study-shows-how-mobile-users-actually-respond-to-in-app-ads-300334644.html#continue-jump/). It is therefore critical to avoid including more than one ad between every two user actions. You should never place them during key moments of user experience.
### Native ads
[__wf_reserved_inherit]
Native ads are designed to match the look and feel of the app's content. Their goal is to blend in seamlessly with the environment in which they are displayed, but still grab the attention of users scrolling down the mobile screen.
For example, you might have seen content that is marked sponsored in your social media feed. It fits naturally alongside other organic posts that sometimes you do not even notice that it is not from your friends. Shopping apps are another case where native ads can be particularly effective. They allow advertisers to showcase their products in a way that feels relevant to and integrated with the shopping experience.
Native ads are known to be less disruptive than banner or interstitial ads as they are interwoven with the in-app page. In fact, a report revealed that [native ads register an 18% higher lift in purchase intent](https://www.sharethrough.com/blog/ad-effectiveness-study-native-ads-vs-banner-ads) than banner ads, meaning that users are more likely to convert after engaging with native ads. In other words, native ads are capable of providing customized experiences that drive installs.
Yet, you should never forget that native ads should promote content, products, or services that align with your app's audience. The placement also needs to be thought through because the ad's content should be related to the flow of specific in-app pages.
### Rewarded video ads
[__wf_reserved_inherit]
A perfect combination between video ads and reward ads, rewarded video ads have a clear value exchange: the users watch an ad, and the app gives them a clear incentive for paying their time and attention.
Without a doubt, rewarded video ads work great for gaming apps because of the tangible benefit expected in return, namely, extra lives, game currency, and power-ups. Entertainment apps that offer users a limited selection of free content can also use rewarded video ads to give users access to more premium content in exchange for watching an ad.
Usually, users have the power to choose whether to watch a rewarded video ad, and this opt-in structure allows these ads to be less likely to affect user experience. Even more, rewarded video ads can significantly increase user retention and engagement, ultimately boosting revenue. This is especially true if you upgrade the reward for watching the ads for multiple days in a row.
However, like any other type of ad, rewarded video ads should be used sparingly. A good rule of thumb is to display one rewarded video ad every two to three sessions or after the completion of a natural break in the app's content, such as after the completion of a level in a game.
### Playable ads
Playable ads are interactive ads that allow users to try out an app before calling them to install it. These ads are effective in acquiring high-value users because they give potential users a chance to gauge their interest before making a commitment.
Playable ads are primarily associated with gaming apps and indeed, the most commonly found type of playable ads is mini games. Still, these ads can also be used for non-gaming apps that are visually driven, interactive, or experiential.
For example, a travel app could use a playable ad that allows users to "explore" a destination or hotel by swiping through a series of interactive images or videos. Similarly, a fitness app could use a playable ad that allows users to "try out" a workout by performing interactive exercises or following along with a virtual trainer.
## Which type of in-app ad should you give a go?
Now that you have a better understanding of the strengths and weaknesses of different in-app ad formats, it's time to decide which one to implement in your app. As an app developer, there are several factors to consider, such as the following.
- **User experience:** One of the most important considerations when choosing an ad type is its impact on the user experience. You want to choose an ad format that doesn't interrupt or detract from the user's experience with your app.
- **App content:** Another factor to consider is the type of content in your app. For example, if your app is heavily focused on video content, you may want to consider using video ads that will feel more seamless within the app.
- **Audience:** Your app's target audience also matters. Certain ad formats may be more effective with certain demographics or user behaviors. For example, if your app is targeted at younger users, interactive ad formats like playable may be more engaging.
- **Revenue goals:** Ultimately, the goal of implementing in-app ads is to generate revenue. You'll want to choose an ad format that will help you achieve your revenue goals without sacrificing user experience or app quality.
- **Ad network support:** When choosing an ad type, make sure that it is supported by the ad network you are working with. Different ad networks may offer different ad formats, so it's worth doing some research to find the right fit.
## Maximizing Revenue Without Compromising User Experience
In-app advertising has become a crucial part of mobile app development as it provides a reliable revenue stream while offering users free access to the app's content. However, you must carefully consider the types of in-app ads that you use, their placement within the app, and their frequency to avoid negatively impacting the user experience.
Keep in mind that you do not have to stick to a single type of in-app ad. By using a combination of ad formats, you can generate revenue from multiple sources while ensuring that the user experience is not negatively impacted. Additionally, developers should continuously test and optimize their ads to ensure that they are effective and enhance the user experience.
Ultimately, the goal is to craft a strategy that enhances the user experience while still generating revenue for your app. When in doubt, come back to this blog post or [Airbridge blog](https://www.airbridge.io/blog) for tips and guidelines to rely on.
---
## Marketing mix modeling vs. attribution: What's the right tool for marketers?
URL: https://www.airbridge.io/en/blog/marketing-mix-modeling-vs-attribution
Language: English
Published: 2023-03-31
As marketers, we often face a multitude of choices when it comes to the tools we use to measure and optimize our campaigns. Two commonly debated methodologies are marketing mix modeling (MMM) and attribution. Each approach has its merits, but they also have distinct differences. In this post, we'll delve into the intricacies of marketing mix modeling and attribution, and discuss the scenarios in which each tool excels.
## Marketing Mix Modeling (MMM)
MMM is a top-down approach that leverages historical data to measure the impact of various marketing activities on sales or conversions. Using statistical techniques, MMM evaluates the relationships between marketing activities (like TV, radio, and digital ads) and sales results.
### A Holistic View
MMM offers a comprehensive perspective of the entire marketing ecosystem, including online and offline channels. This helps marketers understand the impact of their marketing efforts across multiple channels and identify the most effective strategies for driving growth. By capturing the interaction effects between different channels, MMM can uncover synergies and opportunities that may be overlooked when focusing on individual channels.
### Focus on Long-term Effects
MMM emphasizes the long-term effects of marketing activities, capturing their impact over time. This allows marketers to measure how campaigns contribute to growth in the long run, as well as understand the potential for diminishing returns from oversaturated channels. By incorporating the long-term perspective into decision-making, marketers can better allocate resources and optimize their strategies for sustained growth.
### Adaptable to External Factors
MMM can be adjusted to account for external factors like seasonality or changes in the competitive landscape. This adaptability makes it a valuable tool for identifying opportunities and optimizing strategies in dynamic market conditions. By regularly updating MMM models with new data, marketers can stay agile and responsive to changing trends.
### Limited Granularity and Slower Results
However, MMM has its limitations. One significant drawback is the lack of granularity it provides, as it focuses on aggregated data and does not offer insights at the user or campaign level. Furthermore, MMM relies on historical data, which may not provide immediate feedback for campaign adjustments. This can make it challenging to quickly identify and address underperforming campaigns or capitalize on emerging opportunities.
> Wait, there’s an exception! Modern MMM solutions offered by mobile measurement partners, or MMPs, are much more customizable and adaptive to changes. [Check out this post](https://www.airbridge.io/en/blog/mmp-powered-mmm) to find out more.
## Attribution
Attribution is a bottom-up approach that assigns credit for conversions to specific marketing touchpoints, enabling marketers to understand the effectiveness of individual channels or campaigns. By tracking user-level interactions, attribution models can assign varying levels of credit to each touchpoint, based on their influence in driving conversions.
### Granular Insights and Real-time Feedback
Attribution models excel at providing detailed insights at the user and campaign level, allowing for precise optimization of marketing efforts. By understanding which channels and campaigns are driving conversions, marketers can allocate resources more efficiently and maximize the return on their investment. In addition, attribution allows for real-time feedback, enabling marketers to make data-driven decisions on the fly and quickly iterate on their strategies. This agility can be crucial in the fast-paced digital marketing landscape.
### Customizable to Business Goals
Marketers can tailor attribution models to align with their unique business goals, helping to optimize ROI and make the most of their marketing budgets. By adjusting attribution models to better reflect the desired outcomes, marketers can ensure their decision-making is aligned with overall objectives and drive more meaningful results.
### Scope and Data Issues
Despite the strengths, attribution does have its weaknesses. One of them is its primary focus on digital channels, which may not capture the full impact of offline marketing activities. Furthermore, attribution models often overlook the long-term effects of marketing activities, which can lead to suboptimal decision-making that prioritizes short-term gains over sustainable growth.
Lastly, attribution relies on accurate tracking, which can be compromised by factors like cookie restrictions, ad blockers, or unreliable identifiers. These data issues can lead to incomplete or biased results, making it challenging to accurately assess the effectiveness of marketing activities.
## Which Methodology is Right for You?
The choice between MMM and attribution largely depends on your marketing objectives and the channels you use. For businesses with a diverse marketing mix that includes both online and offline activities, MMM offers a more comprehensive view. On the other hand, if you primarily operate in digital channels and require granular insights for quick optimization, attribution may be a better fit.
In an ideal world, marketers would use both methodologies in tandem. This combined approach allows for a holistic understanding of marketing effectiveness, marrying the long-term insights of MMM with the real-time, granular data provided by attribution. By leveraging both tools, marketers can make informed decisions that drive growth, improve ROI, and ultimately deliver better results.
To sum up, the choice between marketing mix modeling and attribution is not mutually exclusive. By understanding the pros and cons of each methodology, marketers can make informed decisions that lead to success in today's ever-evolving landscape. Balancing the comprehensive perspective of MMM with the detailed, real-time insights of attribution enables marketers to create a robust measurement framework that supports both short-term agility and long-term success.
---
## A/B testing for app marketing to boost your ROI
URL: https://www.airbridge.io/en/blog/a-b-testing-for-app-marketing
Language: English
Published: 2023-03-27
As a marketer, you already know that creating a successful campaign requires more than just a great idea. You need to ensure that your campaign is reaching the right audience and providing them with the best possible user experience.
That's where A/B testing comes in. A/B testing can be a powerful tool for improving your marketing performance, and ultimately, your bottom line. And who doesn't love more money?
Read on to learn what A/B testing is and why it is important in app marketing. We'll also go over some use cases and best practices. By the end, you'll have a solid understanding of A/B testing and how to use it to optimize your marketing efforts. Let's get started!
## What is A/B testing?
A/B testing, also known as split testing, is a method of comparing two versions of the same asset to determine which one performs better. By playing with variations, you can identify the creative, message, or user flow that best resonates with your target audience.
To conduct an A/B test, divide your users into two randomized groups and show Version A and Version B of your campaign to each group. Next, measure the performance using relevant metrics such as click-through rate or engagement rate. The version that generates more desired results would be considered more effective and thus adopted in future campaigns.
The process of A/B testing is typically iterative, with marketers continuously refining and improving their campaigns based on the results of each test.
## Why is A/B testing an app marketing essential?
### The importance of experimentation
A/B testing and other forms of experimentation allow marketers to make data-driven decisions about what changes to make, rather than relying on gut feelings or assumptions. They provide a way to isolate and measure the causal impact of different marketing variables on business outcomes.
Revealing causal relationships is the first step to encouraging specific user responses. For example, suppose you notice that users who interact with your ads across two or more channels also tend to show a higher retention rate. While this correlation is interesting, it does not necessarily mean increasing touchpoints would make users stay longer on the app. There may be a third variable that is driving both behaviors. Hence, to avoid misjudgment, draw a causal inference. And to draw a causal inference, test and experiment.
### A/B testing can be your go-to guy
Still, you might be concerned about the time, money, and labor required. The good news is that A/B testing can provide valuable insights at a low cost, within a short timeframe, and with relatively easy implementation.
By focusing on just two variations of a marketing campaign, A/B testing allows marketers to get a sense of what works without investing significant resources. This makes it an ideal testing method for small businesses or startups that are operating on a tight budget.
Moreover, A/B testing is designed to be fast and efficient. In particular, when simply testing the color of a button or the text of a headline, it can take as little as a few minutes to set up the test. This is especially useful for those who need to make decisions quickly in this rapidly changing mobile marketing industry.
Finally, A/B testing is relatively easy to conduct, even for marketers with limited technical expertise. For instance, you can simply group your newsletter subscribers into two and send each of them an email at the same time, on the same topic, but with different subject lines. Then, you can compare the open rates to see which wording is more eye-catching. Many ad platforms also offer built-in A/B testing tools that allow marketers to test different ad variations quickly and easily.
## How can you use A/B testing for app marketing?
A/B testing can be used for a wide range of marketing purposes, such as optimizing landing pages, testing ad copy or imagery, or experimenting with different email subject lines. In this section, we will take the example of MealFriend, an imaginary app for meal planning and grocery shopping.
### Scenario #1: Organic user acquisition
Let's say you are interested in increasing conversion by optimizing the app's install page. You want to test different variations of the app's title and description to see which version generates more installs from organic search results.
Variation A is the original app install page that highlights MealFriend's core features: meal planning, grocery shopping, and personalized nutrition coaching. The page includes the app's original title "MealFriend - Meal Planning & Grocery Shopping" and a description that emphasizes the app's features and benefits.
Variation B is a new app install page that uses a different app title and description to highlight the app's unique value proposition. The new title is "Eat Better with MealFriend" and the description emphasizes the app's ability to help users discover new healthy recipes and simplify their grocery shopping experience.
The A/B test will measure the number of app installs and conversion rate of users who click "install" for each variation. The data will be analyzed to determine which version of the app install page attracts more users. Then, MealFriend can determine the most compelling message to attract and convert potential users.
### Scenario #2: Paid user acquisition
To improve the effectiveness of its paid ad campaigns, MealFriend wants to test two different ad creatives.
Variation A showcases MealFriend's meal planning and personalized nutrition coaching features. The ad includes a catchy headline, a photo of a healthy meal, and a simple call-to-action that encourages users to install the app, saying “Install Now.”
Variation B introduces the same features of the app, but aims to deliver a trustworthy brand image. This time, the ad includes a headline with some stats, a picture of a female dietician in her white coat, and a call-to-action that reads “Get Your Meals Planned by an Expert.”
The A/B test will measure the click-through rate and conversion rate of each variation, and MealFriend will see which one is better at driving installs. The app can also test whether it is more effective to display discount information, include emojis in headlines, etc.
### Scenario #3: User retention
To reduce churn, MealFriend wants to revise its push notification strategy.
The original push notification strategy sends users a daily meal plan, shopping list, and personalized nutrition tips. The push includes an attractive image of a healthy meal, a short description of the meal plan, and a call-to-action that encourages users to open the app and start cooking.
A new push notification strategy sends users a weekly recipe collection based on their dietary preferences and restrictions. The push includes an image that showcases multiple recipes, a short description of the recipe collection, and a call-to-action that encourages users to browse the recipes and add them to their meal plans.
The A/B test will measure the open rate, click-through rate, and conversion rate of each push notification variation. This way, MealFriend can figure out how to increase user satisfaction.
## Finally, here are some pro tips for you…
A/B testing is a marketing technique popular among app marketers across different verticals. Obviously, anyone can conduct an A/B test, but not everyone can do it well. This is why we are wrapping up this blog post with some best practices that you should follow.
**1\. Drill down on what you really want to know**
Before starting an A/B test, it's important to define clear goals and metrics for success. For example, the goal might be to increase app downloads, improve user engagement, or boost conversions.
**2\. Don't be greedy: Test one variable at a time**
To accurately measure the impact of a particular variable, it's important only to test one at a time. For example, if testing the impact of different headlines on app downloads, only change the headline and keep all other elements of the page or ad the same.
**3\. Use a large enough sample size**
To ensure that the results of the A/B test are statistically significant, you need to have a decently large sample size. As a general rule of thumb, a sample size of at least 100 conversions per variation A and B is recommended to achieve statistically significant results with reasonable confidence. If you are not sure, use [A/B test sample size calculator](https://www.optimizely.com/sample-size-calculator) provided by various marketing analytics platforms. This will help reduce the margin of error and increase the accuracy of the results.
**4\. Make A/B testing a regular habit**
A/B testing should be an ongoing process, with marketers testing new variations and refining their campaigns regularly to ensure continued success. Moreover, user preferences and behavior are constantly changing. Be steady with your tests to meet the evolving needs of your audience.
In sum, A/B testing is an essential tool for app marketers who want to optimize their campaigns and achieve success. By comparing two versions of the same asset, A/B testing allows you to isolate and measure the causal impact of different marketing variables on business outcomes.
For more mobile marketing content like this, subscribe to our newsletter by scrolling down a bit. Never miss out on the latest industry trends and best practices!
---
## 5 reasons to use MMP-powered marketing mix modeling
URL: https://www.airbridge.io/en/blog/mmp-powered-mmm
Language: English
Published: 2023-03-23
Did you know that [campaigns integrating four or more channels outperform single- or dual-channel ones by 300%](https://www.gartner.com/en/marketing/insights/multichannel-marketing)? What about the fact that [those who engage with a brand across multiple channels have a 30% higher lifetime value](https://www.thinkwithgoogle.com/intl/en-gb/consumer-insights/consumer-journey/omni-channel-shoppers-an-emerging-retail-reality/) than those who only engage on a single channel?
The concurrent use of different channels is an integral part of mobile marketing. By leveraging the unique strengths of each channel, app businesses can create a seamless user journey for their target audience, reinforce brand identity, and drive conversions.
Yet, it can be challenging to devise a cross-channel strategy that actually works, and this is where marketing mix modeling (MMM) comes in. A statistical technique using historical data and advanced algorithms, MMM helps you determine the ideal distribution of resources across various channels including TV, print, online, and social media to achieve desired business outcomes. It also enables businesses to make data-driven decisions, optimize budget allocation, and increase marketing ROI.
> For a more detailed explanation of MMM, check out the content below:
> 👉 [Marketing Mix Modeling: The Privacy-First Mobile Measurement Method
> ](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper)👉 [Your All-In-One Guide to Marketing Mix Modeling](https://www.airbridge.io/en/blog/marketing-mix-modeling)
There is no doubt that MMM is a powerful tool. However, as everything comes with a price, it involves a costly and complex process that requires expertise in data analytics and the ability to process large amounts of data quickly and efficiently.
This is why many app businesses choose to outsource the work. Some of the more conventional options include consulting firms, market research companies, and data intelligence solutions. But in fact, mobile measurement partners (MMPs) have the potential to deliver exceptional MMM-driven insights.
Hence, in this blog post, we’ll have a look at why you should consider working with an MMP to incorporate MMM to your strategy and get the most accurate understanding of your marketing performance.
## So, what’s so good about MMP-powered MMM?
There are five main benefits to using MMM provided by MMPs.
### Data
MMPs have access to a wealth of data from a variety of sources, including ad networks, demand side platforms, customer relationship management systems, etc. The data is auto-transferred from these third-party tools to MMPs through simple SDK integration. In ordinary circumstances, MMPs would use the information to identify a user’s interaction with different ads across channels and associate it with a conversion.
MMM does not require such granular data; aggregated channel-level data is all that is asked for. Nonetheless, it helps to take into account a wide range of factors that can influence user behavior. By utilizing various data sources, MMPs can build robust models that provide a more complete and accurate representation of the impact of each channel.
### Technology
MMPs are enthusiastic about developing and adopting the latest technology to improve their performance measurement accuracy. They specialize in statistical techniques like regression analysis and machine learning that help collect, process, and analyze large volumes of data.
Despite the differences between attribution modeling and marketing mix modeling, the technology which MMPs have at their fingertips allows them to build marketing mix models even with limited data and on a tight schedule. Thus, MMPs are capable of delivering timely recommendations that enable marketers to forecast returns and optimize ad spend.
### Expertise
MMPs typically have a team of seasoned data scientists who are experts in marketing analytics. They have the knowledge and experience necessary to perform complex statistical analyses and interpret the results in the context of mobile marketing. Relying on them, you can be confident that your marketing mix model is in good hands.
In addition, MMPs have industry-specific insights that can be valuable for app businesses of certain verticals. For example, an MMP that has worked with gaming apps would know the marketing channels and strategies that are most effective in the gaming industry. With this kind of background knowledge, MMPs can craft and execute a customized modeling strategy that maximizes the ROI of your marketing budget.
### Efficiency
MMM used to be a time-consuming and labor-intensive process that app businesses with smaller marketing budgets could not afford. Thankfully, modern day MMPs have combined their comprehensive data, advanced technology, and extensive expertise to make MMM accessible.
Instead of dedicating your internal staff to the building, validating, and updating the model, let an MMP get the job done. This can free up your team’s valuable time and resources to focus on other important marketing initiatives.
### Optimization
MMM is an ongoing process, and optimization is key to achieving the best possible results. For higher accuracy, the model needs to be regularly updated and reflect any recent changes to variables such as ad budget, promotions, macroeconomic conditions, and seasonality.
In the past, the flexibility of MMM was limited; for instance, many consulting firms revised models only semi-annually. On the other hand, MMPs have repackaged MMM into an adaptable SaaS solution. This ensures that marketers stay responsive to changing trends and make the most out of their marketing investment.
## How to get a complete picture of your marketing performance
The cost and benefit of using MMP-powered MMM may depend on the goals of and resources available to each app business. Therefore, we advise you to carefully evaluate them before making decisions.
Another point to keep in mind is that MMM is not your only answer. There is no single perfect solution in the world of mobile marketing, and app marketers should be able to take a holistic approach, making smart use of different measurement methods according to their needs. MMPs allow you to do this.
### Boost your marketing effectiveness with Airbridge
Airbridge believes in the power of a [unified measurement stack](https://www.airbridge.io/product/unified-measurement-stack). As an MMP, we provide the popular last-click attribution for easy and intuitive analysis. We also offer advanced methods like multi-touch attribution and incrementality measurement to add clarity. And of course, we have an AI-powered MMM solution that only requires six months’ worth of aggregated channel-level data to show the incremental contribution of each channel and find the best marketing mix to maximize your future budget potential.
Mobile marketing can be challenging, especially when there are new regulations and unforeseen changes. To achieve success regardless, app marketers need to work with a partner that gives them multiple ways to track performance and fill the gaps of attribution.
Airbridge is here to do it. [Reach out for a quick chat](https://www.airbridge.io/en/book-a-demo?utm_source=blog&utm_medium=link) and take your marketing to the next level!
---
## Ad fraud in mobile: How to safeguard your marketing budget
URL: https://www.airbridge.io/en/blog/ad-fraud-in-mobile
Language: English
Published: 2023-03-22
Where there is money, there is fraud. Ad fraud is not a new phenomenon, but as global mobile ad spend increases, it becomes more tempting for some individuals or organizations to commit fraudulent activities because the potential rewards grow in accordance.
Today, ad fraud is a major challenge faced by advertisers, publishers, and adtech companies alike. In fact, according to [Business of Apps](https://www.businessofapps.com/ads/ad-fraud/research/ad-fraud-statistics/#:~:text=In%20the%20United%20States%20%2462,15%25%20of%20the%20total%20spend.), in the United States, 15% of the total ad spend, or $62 million out of $407 million, was wasted due to ad fraud in 2022. To make matters worse, ad fraud entails not only immediate financial damage but also decreased ROI and loss of trust.
It is self-evident that ad fraud is a must-know concept in the mobile marketing industry. In this blog post, we’ll explore the definition of ad fraud in mobile marketing, types of ad fraud, and the prevention and detection methods. Let’s dive in!
## What is ad fraud?
Ad fraud in mobile marketing refers to the practice of generating fake impressions, clicks, or conversions on mobile ads for financial gain. Through various methods, scammers deceive legitimate advertisers across multiple publishers and make it appear as if their ads, no matter the format, are being viewed, clicked, or engaged with by real users, when in fact they are not.
There are both direct and indirect impacts of ad fraud on an app business’s profitability. On the direct side, ad fraud causes marketers to spend their ad budget on clicks and impressions without real value, or on traffic generated by bots, resulting in a significant waste of money. This also means that their overall return turns out to be lower than expected. On the indirect side, ad fraud does serious harm to the advertiser’s brand reputation and decreases the effectiveness of future ad campaigns. What is more, these are altogether detrimental to the health of the mobile ecosystem.
## Common types of ad fraud
While there are many different forms of ad fraud in the mobile marketing industry, understanding at least some of the most common ones helps with minimizing the risk.
### Click spamming
Also referred to as click flooding, [click spamming](https://www.airbridge.io/glossary/click-spam) is a type of click fraud that involves the use of bots to “click away like crazy.” By generating fake clicks on mobile ads, this fraudulent activity wrongly takes credit for the last click before conversion, leading marketers to reallocate more budget to ad networks that are not creating any actual value. As a consequence, the fraudster generates revenue, drives up ad costs for a particular keyword, or exhausts competitors’ ad budgets.
### Click injection
A sophisticated form of click spamming, [click injection](https://www.airbridge.io/glossary/click-injection) takes place when the broadcast feature of Android is abused to detect an app install. Upon receiving the signal, the fraudster “injects” a click before the installation is complete and steals credit for an install generated by another ad network. Like click spamming, click injection results in an unnecessary payout to the fraudster and skewed mobile analytics data, affecting the advertiser’s targeting strategy and budget distribution.
### SDK spoofing
Another threat that steals marketing budgets is [SDK spoofing](https://www.airbridge.io/glossary/sdk-spoofing). Malware hidden on apps create fake installs using data of real devices, consuming an advertiser's budget without providing any actual value and even tricking the advertiser into paying for thousands of installs that did not actually occur. SDK spoofing is hard to detect because the fraudsters use legitimate SDKs and mimic legitimate user behavior.
### Pixel stuffing
To commit pixel stuffing, scammers display dozens of ads in a single 1x1 pixel frame, making all of them invisible to the human eye but still getting credit for serving those impressions. Even if the ads are delivered, the ad campaigns do not generate any results for advertisers since the users never get to see them.
### Ad stacking
Also a type of impression fraud, [ad stacking](https://www.airbridge.io/glossary/ad-stacking) is a technique where multiple ads are layered on top of each other, so that only the top ad is visible. This can artificially inflate impressions and click-through rates for the top ad, while the other ads in the stack are effectively hidden.
### Device ID reset
At times, fraudsters reset the unique identifier of a mobile device between installs to make it appear as a new device. This way, they can bypass fraud detection systems and use them to make unauthorized clicks, installs, or purchases. These cyclical actions, or [device ID reset](https://www.airbridge.io/glossary/device-id-reset-fraud) marathons, drive seemingly legitimate user interaction that does not have actual value to the app’s growth or the advertiser.
### Duplicate IP
Duplicate IP in ad fraud refers to a tactic used to generate a large number of app installs from the same IP address within a short period of time. Using automated software or bots, scammers create a false representation of an app’s popularity that can mislead mobile users or advertisers.
## How to protect yourself from ad fraud
Detecting and preventing ad fraud requires a multi-layered approach. In essence, you start off by determining a baseline of “normal” user behavior that allows you to identify anything abnormal. Then, based on the absolute minimum, you can monitor your data and blacklist suspicious domains and IPs while whitelisting trusted publishers and ad networks. You can also use a bot management system and anti-malvertising software.
Yet, to effectively boost your protection efforts and safeguard your marketing budget, it is advised to use advanced tools that are specialized in fighting ad fraud.
### No worries at all, we’ve got you covered!
[Airbridge](https://www.airbridge.io/product/fraud-protection-detection) is a mobile marketing partner that is designed to provide marketers with the most accurate view of their ad performance.
As part of the effort, Airbridge helps marketers closely monitor traffic sources and fraudulent activities in real time. It identifies any unusual patterns or discrepancies and tracks down suspicious anomalies such as click injection and click spamming. Airbridge also allows marketers to personalize fraud validation rules to suit their business needs. In other words, it’s a complete package.
As the mobile advertising industry continues to grow, it is crucial for marketers to remain vigilant and stay up-to-date on the latest developments in ad fraud prevention and detection. By working with Airbridge, you’ll be able to protect your ad budget and maximize your marketing potential.
We’ve got your back. [Contact us today](https://www.airbridge.io/en/book-a-demo) to stay one step ahead of fraudsters and continue to drive success.
---
## Last-touch vs. multi-touch attribution: What’s the difference?
URL: https://www.airbridge.io/en/blog/lta-vs-mta
Language: English
Published: 2023-03-16
App marketers should always be on the lookout for ways to optimize campaigns and measure success. A crucial part of the effort is attribution – the process of assigning credit for a conversion to a particular touchpoint. By identifying and analyzing the source of an install or purchase, you can figure out the effectiveness of different marketing actions and decide on the next steps.
There are various types of mobile attribution models, but two of the most widely used are last-touch and multi-touch. Hence, in this blog post, we will delve deeper into these two and their respective pros and cons. Keep on reading to have a better understanding of which attribution model is best suited for your app marketing needs.
## What is LTA, or last-touch attribution?
The last-touch model, as the name suggests, gives all the credit to the very last touchpoint the user interacted with before converting. For instance, if a user had clicked on a TikTok Ads and then installed your app, the last-touch model would identify TikTok as the channel that is responsible for generating the specific install.
Like any other single-touch model, the last-touch model is straightforward. Even if a user had seen seven ads before converting, last-touch attribution will determine that only the seventh one deserves the conversion credit. Such simplicity and clarity make it easy to implement the model and allow app marketers to understand which touchpoints are driving the most conversions. Consequently, the last-touch model is recognized as the industry standard.
## What is MTA, or multi-touch attribution?
The multi-touch model, on the other hand, assumes that all touchpoints play some role in convincing the user to convert. It thus takes various advertisements and ad networks throughout the user journey into consideration.
### Rule-based multi-touch attribution
In rule-based multi-touch attribution, the credit is allocated according to a predetermined set of rules. Some of the most common include the following:
- The **linear model** distributes the credit evenly among all touchpoints. For example, if a user had seen four ads before making a purchase, each ad would be given 25% of the credit for the conversion.
- The **time decay model** gives more credit to touchpoints closer in time to the point of a specific in-app action. If a user had seen four ads before installing the app, the first ad may be given 10% of the credit, the second ad 20%, the third ad 30%, and the fourth ad 40%.
- The **position-based model** assumes that the first and last touchpoints play a more significant role in the decision-making process. Again, if a user saw four ads before converting, both the U-shape and the W-shape models may assign 40% of the credit to the first and last touchpoints. However, the U-shape may give 5% and 15% to the touchpoints in between, whereas the W-shape model would give an equal credit of 10% each.
### Algorithmic multi-touch attribution
Rule-based models are intuitive but subjective, as it is the marketer who decides how to assign the credit. As an alternative, algorithmic attribution analysis uses statistical modeling and machine learning techniques to guarantee higher objectivity. You can apply concepts of Markov chains, the Shapley value, and incrementality to conduct data-driven analysis.
Of course, such complexity implies that building an algorithm-based model requires a greater investment in time, money, and labor. Nonetheless, multi-touch remains in demand as it provides a more complete and accurate picture of the user journey as well as the value of specific sources.
In fact, working with a mobile measurement partner like Airbridge could be a more cost-effective option than going in-house. [Visit our website](https://airbridge.io/) to find out how you can benefit from our technology, experiences, and expertise to achieve the most accurate and unbiased results and drive success.
## Comparing the two approaches
The last-touch and multi-touch attribution models have their own advantages and disadvantages. Here are some key characteristics you should take into account.
.tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-gal1{background-color:#073763;border-color:inherit;color:#FFF;text-align:center;vertical-align:top} .tg .tg-lt9p{background-color:#F3F3F3;text-align:left;vertical-align:top} .tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top} .tg .tg-hdzc{background-color:#F3F3F3;border-color:inherit;text-align:left;vertical-align:top} .tg .tg-0lax{text-align:left;vertical-align:top}
Last-Touch
Multi-Touch
Accuracy
Accompanies the risk of over- or under-valuing certain touchpoints
Can provide a more complete and accurate view of the true impact of each touchpoint on the conversion
Complexity
Uses a simple and straightforward standard, easy to implement
Requires more data and analysis, more difficult to implement
Actionability
Provides a clear answer to the question of which touchpoint led to a conversion, making it easier for businesses to take action and optimize marketing efforts
Provides a more nuanced view of the user journey, which can make it harder to identify specific areas for improvement
Implications in Mobile
Fails to capture the fragmented nature of user journeys, making it difficult to accurately measure performance
Can use a wealth of data across multiple devices, but there may be disagreements over how credit is assigned
## How to find the right model for your mobile growth
Choosing between last-touch attribution and multi-touch attribution depends on a variety of factors, including the nature of your business, your marketing goals, and the channels you use. Because there is no one-size-fits-all solution, app businesses must carefully identify which approach would best suit their needs.
Here are some recommendations for businesses looking to optimize their attribution strategy:
1. Consider the complexity of your user journey and sales funnel. Last-touch attribution may be sufficient if your business has a simple sales funnel and a short user journey. However, if you are using multiple channels and your sales funnel is more complex, multi-touch attribution may be the way to go.
2. Choose an attribution model that is in line with your marketing goals. Different attribution models prioritize different touchpoints, so choosing a model that aligns with your marketing goals and business objectives is essential. If you’re not satisfied with your current method of tracking, you can always try out other options.
3. Be prepared to adapt your attribution strategy as new channels and touchpoints emerge. The mobile environment is constantly evolving, and businesses must be prepared to change accordingly.
4. Use sophisticated tracking and analytics tools to track user behavior across multiple devices and touchpoints accurately. This will enable you to create a more accurate and comprehensive picture of the user journey.
In fact, [Airbridge](https://www.airbridge.io/) is here to provide app marketers with a unified measurement stack encompassing last-touch attribution, multi-touch attribution, and marketing mix modeling. Having mastered various approaches to mobile measurement, we are capable of offering the most holistic and accurate view of your marketing performance.
If you are looking for a mobile marketing partner that will help you create a robust attribution strategy and optimize your marketing efforts, [get started with us](https://www.airbridge.io/en/book-a-demo) today!
---
## Airbridge achieves SOC 2 Type II certification
URL: https://www.airbridge.io/en/blog/airbridge-soc-2-type-ii
Language: English
Published: 2023-03-08
Airbridge is proud to announce that we have received the SOC 2 Type II report, demonstrating our commitment to security, availability, processing integrity, confidentiality, and privacy.
SOC 2 Type II is a rigorous, industry-recognized audit that validates that Airbridge’s security practices and controls meet the American Institute of Certified Public Accountants (AICPA) Trust Service Principles and Criteria for System and Organization Control. The report is thus a testament to Airbridge’s establishment of and adherence to strict policies that safeguard our customers’ data.
In addition to the SOC 2 Type II certification, we earned the ISO/IEC 27001 certification in 2020 and have successfully renewed it annually. We also plan to be ISO/IEC 27017 and ISO/IEC 27018 certified by the end of this year.
## Airbridge, an MMP that puts privacy first
Airbridge processes billions of records of data every day to provide real-time mobile marketing analytics. As a mobile measurement partner (MMP), we are aware that our customers trust us with their valuable data, and we take that responsibility seriously. We securely manage our customers’ data, protecting their apps and, ultimately, the privacy of their users.
With a mission to help marketers around the world discover their true sources of growth, Airbridge strives to maintain the highest standards of security and compliance. Our SOC 2 Type II certification signifies that we have reached another milestone in this journey, and we are grateful to be contributing to the success of our customers.
Furthermore, given that our customers are primarily app developers, protecting their data is equivalent to dedicating ourselves to safety, privacy, and security across the mobile ecosystem. We promise that we will keep up similar efforts to provide privacy-centric software and services.
> 💡 Interested in discovering how to perfect your mobile measurement in this era of privacy? Check out our [Marketing Mix Modeling (MMM) White Paper](https://www.airbridge.io/library/mmm-basics-whitepaper).
---
## Leveraging custom product pages (CPPs) for mobile growth in iOS
URL: https://www.airbridge.io/en/blog/custom-product-pages-ios
Language: English
Published: 2023-02-15
> This post was written in collaboration with Userbase, a mobile growth marketing agency based in the US that specializes in venture-backed, early-to-mid-stage mobile startups across all app categories looking to scale their mobile user growth. Visit their [website](https://agency.userbase.co/) for more information.
It’s been about 1.5 years since Apple released Custom Product Pages (CPPs), but we still don’t see enough mobile growth marketing teams implementing this powerful capability.
Apple’s CPPs represent an immense opportunity in improving the relevance of the user journey, maximizing the alignment of traffic source or audience to the App Store experience. In terms of marketing efforts, CPPs’ power resides in enabling marketers to better contextualize and homogenize the user journey of the target audiences through more testing opportunities.
For user acquisition specifically, this feature bridges the gap between ad creative and app store page, allowing for a more consistent experience from start to end, thus resulting in lower CPAs as well as higher conversion rates and ROAS on paid media campaigns. A must-use feature, right?
[__wf_reserved_inherit]
Before moving any further, it’s important to note that CPPs are not the same as Product Page Optimization, which is a different App Store product page testing feature. CPPs are not accessible to users when they search organically; CPP access is only through the unique URL or ID that can be embedded in specific marketing initiatives such as performance marketing campaigns. For this occasion, we’ll leave Product Page Optimization out of the scope.
## 6 steps to designing Custom Product Pages
Follow the below steps to start your creative design process on CPPs in a strategic way.
### Step 1: Identify different user personas based on different value propositions
The goal here is to internally review your app value propositions linked to your different user personas. Users discover the app through different channels and the jobs-to-be-done tend to vary.
Let’s think about the case of a running app.
Persona 1: Users who runs frequently with a high-performance athletic mindset
- They are the marathon-running type.
- Running and sports are a major aspect of their life.
- They care about advanced running metrics and data to support training.
Persona 2: Users who wants to begin running casually and needs basic support
- They are beginners who take running not too seriously.
- They want to have fun and get interesting insights about their running.
- They think running apps are mainly for distance and time tracking.
In this case, you’ll create one CPP that highlights the advanced app features that high-performance runners care about, with tailored messaging and visual elements that are relevant to this user persona. Then implement the same approach on the casual runner persona.
### Step 2: Predict where the people seeing your CPP are going to come from
In order to implement CPP strategy most effectively, identifying the sources of traffic that will be directed to the CPP is a vital initial step. Based on the type of audiences that engage with your particular sources of traffic, you can build a CPP that best resonates with them.
In the paid media scenario, the audience will expect similar value proposition elements in the app product page to what caught their attention from the ad in the first place.
### Step 3: Pick specific app features, offerings or releases to showcase
The goal of specific app offerings spins around choosing the one or two offerings that might impact best on a particular audience. This is also useful when intending to split tests against the default set of screenshots.
In this case, you could create a CPP on a recent feature you might want to present to your audience.
> 💡 PRO TIP
>
> Let’s take the example of a finance app that has three product offerings: mobile banking, investing and budgeting. Marketers for this app can ask themselves: “If my app did solely budget tracking, how would I design my screenshots and messaging?” They’d answer this question by building and showing a CPP for audiences that resonate with this particular offering, or certain keywords that specifically relate to it in the case of Apple Search Ads.
### Step 4: Add seasonal elements to tap into heightened excitement
Leverage seasonal elements to build a cohesive campaign that backs up a special promotion.
For some apps, Thanksgiving is a very strong seasonality, or seasonal promotions around Christmas could make sense. Maybe you have a dating app that can leverage a Valentines Day CPP with special mid-February promotions or messaging. If it’s an education app we’re talking about, the back to school period of time can dictate specific CPPs that leverage this twice-a-year seasonality.
### Step 5: Gather design elements to create multiple Custom Product Pages
Having prepared user personas, messaging strategy, etc., you should put visual elements and images that you believe will impact your target audience most. Set the tone according to who will be seeing the CPP.
After assembling the CPP, step out of your marketer’s shoes and do a 5-second test with your team in order to simulate how the CPP will be seen by users. Receive feedback and go back to the design board to finalize any last important detail.
### Step 6: Set your Custom Product Pages up
Once you’re done with the design of a particular CPP, set it up on App Store Connect. The good news is that developers do not need to submit a new app to upload their CPPs since set up is independent of app submission. This does not exempt CPPs from Apple’s review, so it’s important to take a couple days of review time into consideration on your CPP timeline.
A step-by-step guide:
1. Go to the Apps section on your App Store Connect
2. Select the app where you’ll be building the CPP
3. Go to the Features menu and look for Custom Product Pages towards the bottom
4. Create a Custom Product Page and add a reference name
5. Follow the process until the full CPP is uploaded and ready for submission.
## Example: health & wellness app
Consider a health & wellness app that has two core offerings: meditation and yoga.
The app’s default product page showcases both offerings and illustrates its benefits as a whole. On its paid media efforts, it would be greatly beneficial to create a specific Custom Product Page for each of the two offerings. This way, the marketers will be able to facilitate a contextualized and homogenous user journey when people interested in meditation click on its meditation-specific ads (and vice versa). Each CPP will exclusively highlight the app’s value offering around either meditation OR yoga, laser focusing on each user profile and providing a more relevant app store experience for high-intent app users.
[__wf_reserved_inherit]
Utilizing each particular CPP on its meditation-specific and yoga-specific marketing efforts will most likely result in more installs, higher CPIs, and possibly higher in-app conversion rates.
Every time you create a CPP and roll it out, a new product page ID parameter is added to the default product page URL to auto-generate a unique URL. All you have to do is to copy and paste it wherever and whenever.
Remember that the unique URL remains consistent throughout all future versions of the page and that there is no way to change or reuse the URL for a different page. In case you need to pause a campaign, disable the URL to make a page unavailable to the public. If someone taps the disabled link, they will be redirected to your default product page.
## Performance measurement for perfect optimization
Performance measurement helps you assess where you currently are and create an action plan to reach your goal. Of course, this principle applies to CPPs as well; only by evaluating how each of your pages is doing can you understand whether using CPPs actually gets you more users.
The easiest option for tracking your CPP effectiveness is App Analytics in App Store Connect. Under the Acquisition tab, you can see page impressions, installs, reinstalls, and conversion rates, along with retention data and average proceeds per paying user. If users acquired from a custom page stick around your app longer and spend more than users from the default page, you can say that your CPP is well attracting high-profile users.
[__wf_reserved_inherit]
However, App Analytics cannot provide any deeper insights about your CPP performance. This is why we recommend integrating with a mobile measurement partner (MMP) like [Airbridge](https://www.airbridge.io/) for advanced tracking and analysis as well as efficient CPP link management.
There are innumerable paid, earned, and owned media channels where you can share the links to your CPP, and an MMP interacts with these channels independently to attribute app installs and monitor post-install events
### How Airbridge can help you with CPP measurement
With Airbridge, an MMP that provides unparalleled measurement insights, you can perform a granular analysis of how users are acquired. This ultimately allows you to drive more qualified traffic and achieve a higher in-app conversion rate.
To elaborate, while App Analytics in App Store Connect only tracks the most basic metrics, Airbridge tells you exactly which search keyword users entered to be shown a particular CPP. This information would give a hint at the user’s intent and how you should modify the page content to be more relevant, boosting both your organic and paid marketing efforts.
Another benefit of using Airbridge in conjunction with your CPPs is one-stop link creation and management. First, go to App Store Connect and create as many CPPs as you need. Then, copy and paste the CPP links to Airbridge to conveniently generate trackable deep links and manage or modify them as an organized list.
The best part about [Airbridge’s deep linking feature](https://www.airbridge.io/product/deep-linking) is that it provides seamless cross-platform redirections that boost conversions. For users who don’t have the app yet, our deep links smoothly connect users to the App Store through deferred deep linking. All of these contribute to driving better user journeys and delivering high and measurable results.
Moreover, you can brand and customize your links to enhance brand awareness and establish credibility. This means that you can utilize unique URLs for a specific campaign, adding the finishing touch of personalization that would bring you the most loyal users to your app.
### Leverage the power of collaboration between MMP and CPP
CPP is in itself a powerful tool. However, when an MMP like Airbridge comes in, it becomes invincible because the collaboration of CPP and MMP enables you to measure not just on the creative level but even on the channel and campaign level.
Track the impact on CPI, IPM, and other key metrics and measure the true value of your CPPs using an MMP. Provide awesome experiences to your potential users. This is how you will be able to drive a high conversion rate at low costs and win the iOS user acquisition battle.
## End-of-the-day takeaways
Taking a stab to experiment with Custom Product Pages enables you to narrow your focus by properly contextualizing the end-to-end user journey. Not embracing this very valuable tool results in opportunity losses in the form of untapped conversion rate lifts for your app and in competitors from your niche taking the CPP advantage to get ahead on their user journey relevance.
You should see CPP opportunities whenever any particular relevant source of traffic allows you to add a custom URL. That’s a chance for a CPP to boost performance, whatever performance means within that context.
Moreover, even though CPPs are not indexed in the App Store, they do have the possibility of improving ASO, organic rankings specifically, since they help increase install velocity and install conversion rates across the overall app listing.
In order to prevent this article from being dreadfully lengthy, we omitted a few pro tips and CPP implementation tactics. Don’t hesitate to ask us questions and we’ll be happy to chat about your iOS app growth. [Get in touch!](https://www.airbridge.io/en/book-a-demo)
---
## Power up your customer engagement with advanced Airbridge-Braze integration
URL: https://www.airbridge.io/en/blog/airbridge-braze-advanced-integration
Language: English
Published: 2023-02-15
We’re happy to share that Airbridge has strengthened its partnership with Braze, a leading comprehensive customer engagement platform, to level up its integration with best-in-class technology!
An official Braze Alloys Partner since 2021, Airbridge has helped marketers build data-driven engagement campaigns by enabling them to send attribution data to Braze through SDK integration. The new server-to-server (S2S) integration allows even more various types of data, faster transfer, and higher user matching and identification accuracy.
Once you configure the integration, you can transfer attribution data of **App Install Event** and **Deep Link Open Event** from Airbridge to Braze. The first is sent by default, whereas the latter is sent if you choose. Braze will map the data to its filters so that you can segment your user base in the Braze dashboard by the **channel, campaign, ad group, and/or ad creative** that generated an app install or a deep link open.
When used together with robust segmentation tools, attribution data can keep your users engaged with your app. For example, let’s say you have an e-commerce app that sells apparel for women, men, and children. By integrating Airbridge and Braze, you can create a user segment of those who clicked on an ad creative containing photos of children’s pants and send them an in-app message offering a 20% discount on children’s apparel. This kind of personalized communication not only drives purchases but also retains users, ultimately contributing to higher ROI.
Please refer to the [Airbridge User Guide](https://help.airbridge.io/hc/en-us/articles/900004368546-Braze#available-data-fields) for detailed instructions on setting up the integration. Note that current Airbridge clients who want to transition from SDK integration to S2S integration are recommended to remove the old code snippet.
## About Braze
Braze is a comprehensive customer engagement platform that powers relevant and memorable experiences between consumers and the brands they love. Context underpins every Braze interaction, helping brands foster human connection with consumers through interactive conversations across channels that deliver value quickly and continuously. To learn more, visit [braze.com](http://braze.com/).
---
## Your all-in-one guide to marketing mix modeling (MMM)
URL: https://www.airbridge.io/en/blog/marketing-mix-modeling
Language: English
Published: 2023-02-01
Necessity is the mother of invention. We all saw how COVID-19 had accelerated digital transformation over the past three years. Daily lives have been reorganized around social distancing measures, and digital processes have replaced more familiar and conventional ways of working.
The mobile marketing world is also going through a huge inevitable change. With privacy and security concerns increasing among users, we are forced to plan and optimize for the next normal. Thankfully, [a new generation of marketing mix modeling (MMM)](https://www.airbridge.io/en/blog/modern-marketing-mix-modeling) is expected to be a problem solver.
In this post, we will take a look at what MMM is, why it has come into the spotlight recently, how you should work with it, and everything else you need to know about it.
> Before moving on, if you are confused about all the different mobile measurement acronyms – MMM, MTA, MMP, and more, check out the blog posts below.
> 👉 [Why an MMP is an App Marketing Must-Have
> ](https://www.airbridge.io/en/blog/why-mobile-attribution-tool)👉 [MMM vs. MTA: Which Is Right For You?
> ](https://www.airbridge.io/en/blog/mmm-vs-mta)👉 [Marketing Mix Modeling: The Privacy-First Mobile Measurement Method](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper)
## What is marketing mix modeling?
MMM, which stands for marketing mix modeling or media mix modeling, is a statistical methodology used to estimate the impact of various marketing inputs on desired outcomes.
Using aggregated channel-level data as well as information about promotions, macroeconomic conditions, seasonality, etc., MMM allows you to answer questions like “What is the effect on sales of a search ads campaign in February compared to a display ads campaign in August?” or “What is the ROI of my print and TV spends?” The results are then adopted to adjust marketing tactics, forecast returns, and optimize future budgets.
Think of MMM as the process of understanding the effects of caffeine on sleep quality. Let’s take a look at the table below.
.tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-2fym{background-color:#011140;border-color:inherit;color:#FFF;text-align:left;vertical-align:top} .tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top} .tg .tg-oz0b{background-color:#EFEFEF;border-color:inherit;font-weight:bold;text-align:left;vertical-align:top}
Monday
Tuesday
Wednesday
Thursday
Performance
(Hours of Sleep)
4
12
8
2
Channel
(Sources of Caffeine)
2 cups of coffee
1 cup of coffee
3 cups of coffee
1 cups of coffee
1 cup of black tea
1.5 cups of black tea
1 cup of black tea
1 cup of black tea
3 bags of M&M’s
1 bag of M&M’s
5 bags of M&M’s
3 bags of M&M’s
As you can see, sleep duration changes as caffeine intake varies, meaning there is a chance that these two are correlated. Likewise, MMM processes how each part of the marketing mix plays a role in driving sales and improving ROI. Some specific takeaways from this analogy are:
- Where the coffee comes from – Tree A, Tree B, and Tree C – does not matter. What is important is how many cups of coffee you have each day. Similarly, MMM requires daily channel-level spending but not unique user IDs.
- The unit, whether it be cups or bags, does not matter as long as it stays the same for each source of caffeine. The performance of television ads can be measured by comparing daily spending and increase sales, whereas paid socials give access to more granular metrics like cost-per-click (CPC). In other words, even though these channels collect touchpoint data differently, MMM is capable of evaluating the contribution of both.
- Caffeine from Monday may stay till Tuesday, the way ads have a carryover effect. Caffeine overdose does more harm than good, and when channel saturation happens, it becomes inefficient to spend on that channel. MMM takes all these possibilities into regard.
- You can easily test out how shorter your sleep gets with each extra cup of coffee, extra cup of black tea, and extra bag of M&M’s. MMM results can also be calibrated with experiments and show the power of each channel more accurately.
- It is alright if it turns out that caffeine is not the only factor; insufficient exposure to natural light keeps you awake as well. MMM considers variables other than daily channel-level spending for higher reliability.
We know, MMM sounds like the ultimate measurement approach, and you might be wondering why you do not recall using it. In fact, MMM is making a comeback after being put on the back burner.
## The history of marketing mix modeling
The term “marketing mix” was coined in 1949 by Neil Borden. A professor of advertising, he stated in his 1964 journal article, “When building a marketing program to fit the needs of his firm, the marketing manager has to weigh the behavioral forces and then juggle marketing elements in his mix with a keen eye on the resources with which he has to work.”
Over time, many other scholars added breadth and depth to the concept and MMM was one of the few tools available for holistic measurement before the digital age. Kraft Foods, an early user of this analysis, monitored whether their sales increased by promoting Jell-O on different television networks and magazines in different regions at different times of the year.
Nonetheless, the use of MMM was rather limited in the 20th century. Only the largest consumer goods companies could implement it because they were the first to have access to relatively accurate data on marketing, sales, and pricing. In addition, marketers were constantly challenged to optimize their marketing mix, but traditional MMM did not provide day-to-day results. One might say it was not surprising to see attribution modeling become the go-to approach for marketing measurement when real-time user-level tracking became available.
It seemed like MMM was falling powerlessly. However, the looming demise of third-party cookies and recent introduction of tighter privacy regulations made a turning point. Today, MMM has risen an essential part of the marketing survival kit for better decision making in this privacy-first era. Many tech companies reinvented MMM as SaaS solutions, opening doors to speed, scalability, cost effectiveness, and most importantly, inclusivity.
## How to build a marketing mix model
As the latest technological advancements have made MMM much more affordable and accessible, we suggest working with martech companies like [Airbridge](https://www.airbridge.io/) and fully enjoy the benefits of using SaaS MMM. However, if you have already decided to get full-suite consulting service or to try building internal MMM capabilities, here are some steps you should know.
### Choosing the right modeling technique
There are many ways to build a model including the Bayesian approach, Shapley Value Regression, and tree-based ensemble machine learning. However, the most popular is multivariate regression due to easier model interpretability.
While regression models in general “fit a line” to the observed data to describe relationships between variables, multivariate regression is used to estimate (i) the strength of the relationship between two or more independent variables and one dependent variable and (ii) the value of the dependent variable at a certain value of the independent variables. The equation can be either linear or non-linear, meaning that some marketing inputs are directly proportional to the increase in sales while others are not.
Another important point is that MMM uses multivariate regression to analyze time series data, which helps with understanding trends and patterns over time. Time series data is collected at regular intervals, such as daily, weekly, or monthly and can be used to forecast future values of a variable based on past values. This is how MMM not only examines the past but also predicts the future.
### Preparing the most relevant data
Any statistical modeling starts with selecting appropriate variables to find a good model from a set of possible models.
Usually in MMM, the dependent variable is **business performance**, such as sales, revenue, market share. It can be further broken down into base outcome, which is the natural demand for the product driven by pricing, seasonality, brand equity and brand reputation, and incremental outcome, which is produced as a result of marketing activities. This data should be collected over the same period of time as the ad spend data.
Common independent variables include the following.
- **Ad spend** represents the amount of money spent on different types of marketing channels, such as television, radio, print, digital, and outdoor. This is an incremental variable and the data should be collected over a period of time that is long enough to capture the impact, ideally 12 months or longer.
- **Promotion** is a key marketing activity aimed at increasing sales in the short term. Free delivery, 0% APR or cashback, and many other offers would drive sales without the support of any advertisement.
- **Price** is the most flexible yet the most important component of marketing mix. MMM shows the price elasticity of demand, which in most cases, price increases lead to sales decreases.
- **Distribution** is a base variable that allows marketers to know how the sales volume will move due to shifts in the number of stores or number of locations the product is available.
- **Macroeconomic** conditions, such as the fluctuation of the Commodity Channel Index (CCI), inflation rate, unemployment rate, and GDP, all impact the sales volume.
- **Seasonality** also plays a role in changing consumers’ purchase decisions. For instance, sales during the Christmas season are generally higher than the average.
[__wf_reserved_inherit]
Note that variable selection can be very complex because you need to sort out and leave only the factors that actually affect your performance. Moreover, in the real business environment, multiple marketing campaigns tend to happen at the same time. This leads to multicollinearity, which occurs when two or more independent variables in a regression model are highly correlated with one another. This is not a problem per se, but it makes it difficult to determine the unique effect of each independent variable on the dependent variable.
Multicollinearity can be detected by using statistical methods such as correlation matrix and Variance Inflation Factor (VIF). To fix multicollinearity, you can remove one or more correlated independent variables from the model, combine correlated independent variables into a single composite variable, or collect more data to reduce the correlation among independent variables.
## Limitations of marketing mix modeling
MMM is a powerful tool, but like any tool, it has its limitations that you should be aware of. First and foremost, MMM does not establish causality. It only demonstrates correlations between variables, meaning that it does not prove that a business outcome is the consequence of a certain marketing activity or ad spend. This also implies that MMM relies on certain assumptions about the data such as linearity and inalterability. If the assumptions are not met, the analysis results may be biased.
Second, MMM does not capture marketing performance in the most precise manner. Using aggregate data can make it difficult to see the impact of specific media tactics or channels. In addition, the aforementioned multicollinearity can pose challenges to determining the unique effect of each independent variable on the dependent variable.
Third, MMM may not be the most agile measurement approach you can find in the market. Preparing months worth of data and training the model with it takes time, and in the worst case scenario, if the data is of poor quality, the analysis may be limited. It should also be noted that the norm is to update the model on a quarterly or semi-annual basis, thus making it difficult for marketers to respond to the rapidly changing market situations.
Of course, there are alternatives. Airbridge MMM, for instance, auto-collects your performance and cost data and provides several types of reports that are updated every day or every week. Nonetheless, we recommend that you use it in conjunction with other measurement approaches like last-touch attribution and multi-touch attribution to draw a comprehensive conclusion.
## What else should you know?
Phew, we say this is enough for now! We touched upon a lot of different concepts, so be proud of yourself for reading till the end.
Key takeaways of the day:
- MMM helps you understand the relationship between your marketing mix and business performance.
- MMM is a measurement approach essential in the privacy-first era.
- MMM gives you the fullest visibility of your marketing success when used together with additional approaches as part of the [Unified Measurement Stack](https://www.airbridge.io/product/unified-measurement-stack).
If you do not want to miss the latest updates about MMM, stay tuned for our upcoming posts. We are more than excited to share insights and experiences we have acquired as a mobile measurement partner with powerful MMM features!
---
## Optimize your customer marketing campaigns with Airbridge and CleverTap
URL: https://www.airbridge.io/en/blog/airbridge-clevertap-partner
Language: English
Published: 2023-01-31
We are excited to announce our official partnership with CleverTap, **the leading global platform for customer engagement and retention!** You can now transfer attribution data from Airbridge to CleverTap and build more engaging experiences for your customers.
By enabling the integration, you can send three types of event data from Airbridge to CleverTap.
- **Attributed Install Events (Inorganic Install Events):** These install events are attributed to some media sources, such as paid ads on third-party applications. It means that a user installed the app as a result of some engagement with an ad. These events are sent to CleverTap as default at all times when the credentials for integration are submitted.
- **Unattributed Install Events (Organic Install Events):** These install events are attributed to app stores, websites, or referrals. It means a user installed the app without engaging with an ad. If you want to send these events to CleverTap, **make sure to enable the tracking of the event data in CleverTap.**
- **In-app Events (Custom Events):** Tracking these events allow a deeper analysis of user behavior within an app. Examples of in-app events might include Viewed Product, Added to Cart, or Video Watched. The In-app Event data sent from Airbridge can be used to build granular user segments on CleverTap.
The attribution data and events sent to CleverTap can be used to build granular user segments to help you easily create effective campaigns for different audiences.
Please find the [Airbridge User Guide](https://help.airbridge.io/hc/en-us/articles/14266267211289-CleverTap) for instructions to set up the integration. If you want to learn more about our partnership with CleverTap, do not hesitate to [contact](mailto:platform@ab180.co) us!
## About CleverTap
CleverTap is the world’s No. 1 retention cloud that empowers digital consumer brands to increase customer retention and lifetime value. CleverTap drives contextual individualization with the help of a unified and deep data layer, AI/ML-powered insights, and automation, enabling brands to offer hyper-personalized and delightful experiences to their customers. More than 1,300 customers in 100 countries and across 10,000 apps — including Gojek, ShopX, Canon, Electronic Arts, TED, English Premier League, TD Bank, Carousell, AirAsia, Papa John’s, Tesco, Kotak Mahindra Bank, SonyLiv, Swiggy, PharmEasy, and Dream11 — trust CleverTap to achieve their retention and engagement goals and grow their long-term revenue. Backed by leading investors such as Sequoia India, Tiger Global, Accel, CDPQ, and Recruit Holdings, the company is headquartered in Mountain View, California, with offices in Mumbai, Singapore, Sofia, São Paulo, Bogota, Amsterdam, Jakarta, and Dubai. For more information, visit [clevertap.com](https://clevertap.com/) or follow on [LinkedIn](https://www.linkedin.com/company/clevertap/) and [Twitter](https://twitter.com/CleverTap).
---
## Airbridge joins Google’s App Attribution Partner program
URL: https://www.airbridge.io/en/blog/airbridge-google-aap
Language: English
Published: 2023-01-17
We are delighted to kick off the new year with exciting news: Airbridge has been named an App Attribution Partner with Google!
A platform with [4.3 billion users worldwide](https://www.semrush.com/blog/google-search-statistics/) and [5.4 billion search queries a day](https://financialpost.com/personal-finance/business-essentials/google-averages-63000-search-queries-per-second-how-does-this-impact-you), Google without doubt plays a pivotal role in acquiring and engaging high-value app users. In addition, Google has led the privacy-first transformation in recent years, introducing new tools to improve ad campaigns while protecting user privacy.
Google awards its App Attribution Partner badge to a select group of market leaders who have met rigorous standards and respect user privacy, ensuring it offers customers reliable measurement and accurate attribution of Google Ads campaign performance.
Airbridge, now a fully certified Google App Attribution Partner, can better help marketers make informed decisions when working with Google Ads. They will be able to run campaigns through easier setup and identify the marketing activities that are most valuable for their business based on rich data. In addition to campaign measurement and optimization, marketers will have early access to new features as they become available from Google.
“We’re pleased to formally expand our partnership with Google and cooperate for the success of our mutual customers,” said Roi Nam, CEO and co-founder of Airbridge. “The App Attribution Partner badge proves how committed we are to providing best-in-class campaign reporting. Our mission to help marketers around the world discover their true sources of growth fits in perfectly with this program, and we look forward to offering advanced mobile attribution insights and a full understanding of the user journey to our customers with an upgraded integration for Google Ads.”
Airbridge is a powerful, user-friendly, and unified measurement solution that helps marketers boost their mobile app installs and user engagement. With unparalleled insights incorporating Last-Touch Attribution (LTA), Multi-Touch Attribution (MTA), Incrementality measurement, and Marketing Mix Modeling (MMM), Airbridge enables marketers to gain accurate measurement insights at different stages of their marketing cycle even in a privacy-first environment.
Not only does Airbridge employ diverse measurement approaches, it also provides a comprehensive view of marketing performance across web and app, Android and iOS, and SKAdNetwork and non-SKAdNetwork, all in a single unified dashboard. Other tools such as deep linking, marketing analytics, audience manager, and fraud protection also contribute to maximizing mobile growth.
If you would like to know more about Airbridge, visit [our website](https://www.airbridge.io/) or [schedule a meeting with our mobile app marketing experts](https://www.airbridge.io/en/book-a-demo) today. You can also check out [our blog](https://www.airbridge.io/blog) for the latest ad-tech trends and insights.
---
## The evolution of marketing mix modeling (MMM): A pathway to privacy-first measurement
URL: https://www.airbridge.io/en/blog/modern-marketing-mix-modeling
Language: English
Published: 2023-01-13
What is old can be new again with technological developments. This is how marketing mix modeling has made a successful comeback.
Marketing mix modeling (MMM), sometimes referred to as media mix modeling, is a statistical analysis technique that assesses the impact of marketing inputs on desired business outcomes such as sales, conversions, and installs. Using aggregated historical data, MMM helps marketers forecast returns, simulate various business scenarios, and optimize future budgets.
MMM has a long history; the concept was first introduced in the 1950s and rose to popularity in the 1980s. For a long time, only the largest companies have adopted this practice on a wide scale because traditional MMM was time and effort-intensive. However, in recent years, more and more companies – regardless of their size – are turning to MMM to drive growth in a privacy-first landscape.
How has the latest technological advancements brought the new golden age of MMM? Why did MMM evolve to be provided as a software solution? Read on to find out.
> 💡 To get a general idea of what MMM is, [check out our white paper](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper).
## Traditional marketing mix modeling methods
Due to the complexity of variable selection and model building, MMM requires a great deal of time, resources, and expertise. Hence, many marketers interested in MMM have either outsourced every step to external parties or handled everything in-house.
### Using full suite consulting service
MMM has long been and is still provided by major consulting firms. As end-to-end suppliers, they deliver the full breadth of an MMM project from receiving data from their clients to designing an appropriate model and analyzing the results. The best part about this approach is that the clients have access to experts for hands-on support along every step of the project.
However, as with most easy solutions, it is far from ideal. A typical MMM project run by consulting firms involves a very time-consuming process and enormous operating costs. Highly experienced econometricians and statisticians manually validate at least a year’s worth of data, tailor-make a model around the client’s business needs, and write detailed analysis reports. Obviously, none of this is free of charge.
Moreover, once finalized and deployed, the model is updated only on a quarterly or semi-annual basis, thus making it difficult for marketers to respond to the rapidly changing market situations. With such high hurdles, the utilization of MMM has basically been exclusive to big-budget advertisers developing long-term marketing plans.
### Building internal MMM capabilities
On financial costs alone, conducting MMM in-house may seem like a more reasonable option. Other than the already existing labor costs for internal staff, which is a sunk cost, the additional costs are minimal. In addition, you can have full control over your data, thus increasing agility, avoiding risks of information leakage and equipping yourself with always-on measurement.
Even though the benefits of building modeling capabilities internally are definitely worth the effort, it’s easier said than done. In-house MMM as well requires a substantial amount of cost, time and resources. For instance, if your organization does not have any data scientists, you can only set out by recruiting the right people. What’s more, they will have to keep developing and updating the MMM model rather than the essence of the business – the product or service itself.
Furthermore, “in-house” could mean an absence of a second or third opinion, which often serves as an antidote to bias. It is also possible that some critical external factors are neglected, and thus the analysis results are associated with lower objectivity.
## Marketing mix modeling: reborn
As explained above, traditional MMM methods could be exhausting in many ways. Hence, fast-growing early-stage startups or small-budget advertisers had limited access to MMM.
For a long time, this was not a major problem because there were other measurement approaches available in the market. Most mobile measurement partners (MMPs), for example, have long relied on device-level data for effective attribution.
However, with the mobile marketing world leaning into data privacy protection, users now have the option of disabling ad tracking and opting out of personalized ads. In other words, data collection has become increasingly difficult and the need for a privacy-friendly method has arisen.
As MMM runs off aggregated data from various sources rather than granular user behavior data, many are paying attention to the approach with the aim of perfecting their performance measurement. Consequently, the demand for a new generation of MMM has grown among both small- and big-budget advertisers from various industries.
### Back in the game with SaaS solutions
Decades have passed since the introduction of MMM and the world has changed over time. The 21st century, in particular, has witnessed unprecedented technological developments that have revolutionized our way of life. Capitalizing on such innovation, MMM was repackaged as an easy-to-use SaaS solution and started to receive more attention.
The most distinguishing feature of SaaS MMM is that it requires less cost, time, and resources than traditional MMM. These days, thanks to advances in cloud computing, companies no longer have to build a server themselves, and cost and performance data can be auto-collected. In other words, modern MMM is much more streamlined and affordable.
SaaS MMM also allows marketers to expedite analysis and adapt to market changes in a timely manner. In order to speed up the marketing flywheel, you need to be able to assess and update your strategy whenever necessary. This means that efficient and effective performance measurement is a precondition for marketing optimization and informed decision making.
### Presenting the Airbridge Marketing Mix Modeling (MMM)
With a mission to help marketers around the world discover their true sources of growth, Airbridge provides MMM as a user-friendly software solution at a reasonable price. [Airbridge](https://www.airbridge.io/) MMM not only enables marketers to focus on the more important tasks but also contributes to the protection of mobile users’ digital privacy.
In fact, Airbridge is the perfect MMM provider because it is a [mobile measurement partner (MMP)](https://www.airbridge.io/en/blog/attribution-tool-mmp) that is already integrated with major ad platforms. By embedding the Airbridge SDK within your apps, you can auto-collect your performance and cost data and track in-app events to further train your MMM model. Even if you import data manually, the modeling and reporting processes are automated, meaning you will get your analysis results in the blink of an eye.
Airbridge MMM, powered by proprietary machine learning algorithms, never stops improving to drive your marketing success. The model is tailored to suit your business needs based on online and offline channel-level data as well as external factors such as macroeconomic trends and seasonality. You also have full access to key information about your MMM model, from analysis target and input channels to training status and last training time.
Furthermore, Airbridge MMM results are presented on a user-friendly dashboard. The Marketing Mix Analysis report, updated daily, shows the incremental contribution of each channel without relying on user-level data. The weekly and monthly Budget Optimization report finds the best marketing mix to maximize your budget potential. The agility of Airbridge MMM leads to actionability, and marketers can use MMM in their everyday decision-making situations.
> Find out more about Airbridge MMM:
> 👉 [How Marketing Mix Modeling Enables Privacy-First Measurement](https://www.airbridge.io/en/blog/airbridge-marketing-mix-modeling)
> 👉 [Airbridge’s Unified Measurement Stack: Marketer-Centric Approach to LTA, MTA, and MMM](https://www.airbridge.io/en/blog/airbridge-unified-measurement-stack)
## Modernizing MMM to protect user privacy
SaaS MMM is bringing about changes to the mobile marketing industry by overcoming the limitations of traditional MMM.
With lower hurdles to MMM, it may be the norm to measure without relying on third-party cookies or ad identifier details in the future. This is a win-win for all because users can have control over their data while advertisers can assess performance and optimize campaigns no matter what.
Privacy-first performance measurement has been made easy with the return of MMM. If you would like to know more about how SaaS MMM can benefit your business, [talk to our team today](https://www.airbridge.io/en/book-a-demo).
---
## Exploring the marketing implications of SKAN 4.0
URL: https://www.airbridge.io/en/blog/exploring-the-marketing-implications-of-skadnetwork-4
Language: English
Published: 2023-01-05
SKAdNetwork (SKAN) is Apple's attribution framework for iOS 14 that aims to provide a privacy-friendly way for app developers to measure the effectiveness of their ad campaigns. SKAN 4.0 is the latest version of the framework 一 [released on October 24, 2022](https://developer.apple.com/news/?id=31g9nllo), for versions equal to or later than iOS 16.1 and iPadOS 16.1, and it introduces several new features and improvements.
With each update, SKAN brings not only a lot of options but also great complexity to mobile attribution. In this post, we will share the business-side or marketing-side implications of the SKAN 4.0 changes you need to know to stay ahead of the game.
> **Four marketing implications of SKAdNetwork 4.0:**
> \- Small-scale testing and operations get easier
> \- Creative-level or country-level attribution can partly come back
> \- Cohort analysis for ARPU and ROAS can partly come back
> \- Web placements now trackable with SKAN
## Small-scale testing and operations get easier
One of the major improvements in SKAN 4.0 is that it makes small-scale testing and operations easier for app developers. Under SKAN 3.0, the privacy threshold made it difficult to run small-scale ad campaigns. This was because the framework required a minimum number of installs per campaign before it would provide any SKAN postback for in-app events, making it difficult to test and optimize campaigns with a limited budget.
In order to protect user privacy and ensure that conversions are recorded only for a statistically significant number of users, SKAN has a privacy threshold that limits the number of conversions that can be recorded for a given app. Determined by the number of installs per campaign ID that have occurred within the past 24 hours, the threshold is typically known to be within a range of 10-20 installs. If a campaign falls below this, the conversion value field in the postback associated with that campaign will be assigned a null value.
SKAN 4.0 introduces the concept of "crowd anonymity," which is achieved through a four-tiered system ranging from tier 0 to tier 3. Depending on the tier that a source is assigned to, advertisers may receive either fine-grained or coarse-grained values to gather information about users' post-install activities. The first is a 6-bit value allowing for up to 64 combinations while the latter is an approximate value that can be represented by low, medium, or high.
.tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-amwm{font-weight:bold;text-align:center;vertical-align:top} .tg .tg-0lax{text-align:left;vertical-align:top} .tg .tg-teb0{background-color:#D9EAD3;text-align:left;vertical-align:top}
SKAN 3.0
SKAN 4.0
Above privacy threshold
In-app events represented in 0~63 conversion values (fine-grained)
Tier 3
In-app events represented in 0~63 conversion values (fine-grained)
Tier 2
In-app events represented in 0~63 conversion values (fine-grained)
Below privacy threshold
Does not receive any information on in-app events
Tier 1
In-app events represented in low, medium, high (coarse-grained)
Tier 0
Does not receive any information on in-app events
If a campaign/source is assigned to tier 0, no information will be provided, which is similar to the situation when the privacy threshold was not met in SKAdNetwork 3.0. However, if a campaign/source is assigned to tier 1, advertisers may still receive coarse-grained values, which is less specific than fine-grained values but still provides some insights into user behavior. Apple states that advertisers can "receive less specific attribution information when the privacy thresholds for conversion values are not met," which suggests that getting into tier 1 may be less difficult than meeting the privacy threshold under SKAN 3.0.
## Creative-level or country-level attribution can partly come back
[Exploring the marketing implications of SKAN 4.0]
In earlier versions of SKAN, only campaign-level analysis was possible. SKAN 4.0 partially restores this functionality with the introduction of a new parameter called “source-identifier,” which replaces the previous “campaign-id.” The source-identifier can be up to four digits in length (0~9,999), up from the previous two digits (0~99), and can be used to represent additional dimensions such as creative, country, and format.
## Cohort analysis for ARPU and ROAS can partly come back
[Exploring the marketing implications of SKAN 4.0]
SKAN 4.0 partially restores the ability to perform cohort analysis such as day-X average revenue per user (ARPU) and return on ad spend (ROAS) through the use of three postbacks, an increase from the previous single postback mechanism. The conversion window has also been extended to a maximum of 35 + 6 (24 ~ 144 of random delay) days, providing deeper insights into post-install user activities. These postbacks are sent consecutively within a designated time window from the date of app installation and can be used to reconstruct the user's journey after installation.
Mobile measurement partners (MMPs) can use these postbacks, attribute them to a specific date of installation, summarize all user journeys and present them in a cohort analysis format. However, it's worth noting that SKAN 4.0 applies a random delay to each postback, making it difficult to accurately attribute it back to the specific date of installation.
## Web placements now trackable with SKAdNetwork
One of the new features in SKAN 4.0 is the ability to track web ads in Safari browser, in addition to mobile placements. This allows ad networks, MMPs, and advertisers to see how app installs and in-app events are attributed back to the ad placements on the web. This can provide valuable insights into the effectiveness of cross-channel ad campaigns. Please note that this feature will be only supported for the Safari browser.
## Looking ahead
Overall, SKAN 4.0 introduces several new features and improvements that make it easier for app developers to measure the effectiveness of their ad campaigns in a privacy-friendly way.
Airbridge is currently working closely with ad networks to support SKAN 4.0, and we are applying changes to our existing conversion value settings page to reflect the SKAN 4.0 specifications more accurately. We want to ensure that our clients are able to take full advantage of the benefits that SKAN 4.0 has to offer. Tune in for more updates!
---
## Marketing mix modeling (MMM) vs. multi-touch attribution (MTA): Which is right for you?
URL: https://www.airbridge.io/en/blog/mmm-vs-mta
Language: English
Published: 2023-01-03
With so many channels available in the modern marketing world, marketers are always in search of powerful and sophisticated tools to pinpoint which mix of channels generate the most conversions and the biggest return. [Marketing mix modeling (MMM)](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper) and multi-touch attribution (MTA) are two different approaches that can be used to understand the impact of marketing efforts on sales and other outcomes.
While both can uncover business insights, they also have some key differences that make each other more useful than the other depending on the circumstances. What are the advantages of the two approaches, and which is the best for maximizing marketing effectiveness? Read on to find out.
## Key differences and similarities
One of the main differences between the two approaches is their objectives. MTA sheds light on the impact of individual touchpoints on conversions and sales. It allows marketers to attribute credit for a sale or conversion to multiple touchpoints in the user journey, and to understand the relative importance of each touchpoint. MMM, on the other hand, is focused on understanding the overall impact of a company's marketing mix on its sales and other outcomes. It typically involves building a statistical model that incorporates data on marketing activities, such as ad spend and promotions, as well as data on sales and other outcomes.
Another key difference is the type of data that each approach uses. MTA relies on granular device-level data, which allows marketers to track individual users and their interactions with different touchpoints. MMM, on the other hand, typically uses aggregated data at the campaign or channel level, which can make it more difficult to see the impact of individual touchpoints on conversions and sales.
Yet, MTA and MMM do share a commonality: both approaches are not actual experiments, but observational models that are based on machine learning or statistical methodologies. To guarantee their reliability and accuracy, the results can be calibrated using lift studies.
## Competitive advantages of each approach
### Marketing Mix Modeling for a privacy-first era
The strongest advantage of MMM is that it is privacy-friendly. As described above, MMM uses aggregated campaign- or channel-level data rather than granular device-level data. This means that it can still provide valuable insights despite limited access to device-level data. For example, an MMM model could still be built using data on ad spend and touchpoints, as well as data on revenue and other conversions, even if there is no device-level data available.
Another advantage of MMM is that it can be used for both prediction and optimization. For prediction, the model can estimate the likely impact of different marketing strategies on revenue and other metrics. This comes in handy when developing future marketing strategies as it allows marketers to make more informed decisions about how to allocate their budgets and where to focus their efforts.
For optimization, the model can identify the most effective marketing channels that need to be scaled up. By running simulations and comparing the results of different marketing scenarios, marketers can use MMM to identify the marketing mix that will most probably lead to the best results. This can be useful for optimizing the performance of existing campaigns, as well as for planning future marketing efforts.
Although MMM has multiple notable advantages, it does have some potential drawbacks. One is that it may require additional data features, such as macroeconomic variables like GDP and market indexes, as well as internal information like scheduled product releases and promotions, to more accurately perform the analysis. This can be a challenge for marketers who may not have access to all of these types of data.
### Multi-Touch Attribution for a deeper dive
The greatest part about MTA is that it is more granular than MMM, allowing marketers to better understand the impact of individual touchpoints on conversions and sales. This makes it a better option for many marketers who are looking to optimize their marketing efforts and improve the effectiveness of their campaigns.
Another advantage of MTA is that it can be combined with incrementality measurement to draw causal inferences. Using techniques such as propensity score matching (PSM), marketers get to calculate a propensity score for each user based on device-level or user-level features and match the appropriate control and treatment groups. This minimizes the selection bias that may be present in basic observational models.
After determining the incrementality of the overall user journey, marketers can distribute the credits to each touchpoint using a weighting algorithm or other method. Combining MTA and incrementality measurement provides a more accurate and comprehensive view of the effectiveness of their marketing campaigns and helps with data-driven decision-making as well as campaign optimization.
.tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-lnh0{background-color:#D9EAD3;border-color:inherit;font-weight:bold;text-align:center;vertical-align:top} .tg .tg-wlkl{background-color:#D9EAD3;font-weight:bold;text-align:center;vertical-align:top} .tg .tg-baqh{text-align:center;vertical-align:top} .tg .tg-cr3v{background-color:#00195A;border-color:inherit;text-align:left;vertical-align:top} .tg .tg-uvoo{background-color:#00195A;border-color:inherit;color:#FFF;font-weight:bold;text-align:center;vertical-align:top} .tg .tg-7btt{border-color:inherit;font-weight:bold;text-align:center;vertical-align:top} .tg .tg-7yig{background-color:#FFF;text-align:center;vertical-align:top}
MTA
MMM
Exemplary algorithms
Markov Chain, Shapley Value, etc.
Frequentist, Bayseian, etc.
Base data used for analysis
Device-level or user-level behavioral data
Aggregated data
Additional data features required to increase accuracy
No
Yes (e.g. GDP, market indexes, scheduled product releases, promotions, etc.)
Minimum required data period
6+ months
12+ months
Refresh cadence
Max daily (generally weekly, little more flexible to shorten the cadence than MMM)
Max daily (generally monthly or longer)
Computation resources
Relatively bigger computing power is required as it handles user-level and journey-level data sets
Relatively smaller computing power is required as it computes based on aggregated data
Level of analysis
Most likely channel-level or above
(little more flexible to go down than MMM)
Most likely channel-level or above
Cross-network analysis
Yes
Yes
Multi-touchpoint
Yes
Partly (each variable is independent, and hence definitely not last-touch based, but not technically reflecting multi-touch like MTA)
Incrementality
Yes (can compute the baseline using causal inference)
Partly yes (can compute the baseline, but not based on causal inference)
Prediction or media optimization
No
Compatible
ATT impact
High
Low (or none)
Nature of the methodology
Observational
Observational
## So which one is better, MTA or MMM?
MMM provides a top-down perspective of every aspect of a business and attempts to understand how each aspect contributes to a business outcome. MTA is a bottom-up approach, attempting to meticulously measure every consumer exposure to media and then attribute each exposure to an outcome.
Going back to our title – which one approach is right for you, the answer is neither. At Airbridge, we recommend using both MTA and MMM to get a full visibility of your marketing performance, rather than relying on just one, because they have competitive advantages at different stages of the marketing cycle. If you need more information to expedite your decision-making process, our [MMM white paper](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper) can definitely be of help — go check it out!
---
## Airbridge Awarded High Performer in G2’s Winter 2023 Reports
URL: https://www.airbridge.io/en/blog/g2-winter-2023
Language: English
Published: 2022-12-22
We are excited to announce that [Airbridge earned its spot as a High Performer](https://www.g2.com/categories/marketing-analytics#grid) in the G2 Winter 2023 Grid® Reports!
G2 is a trusted software marketplace where millions of people read and write reviews about their experiences with over 145,000 software products. Having approved and published nearly 2 million reviews, G2 helps people around the world make smarter purchasing decisions.
The High Performer badge in the [Marketing Analytics category](https://www.g2.com/categories/marketing-analytics) is awarded to products exceeding customer expectations. The winners are determined based on product reviews by G2 users as well as data aggregated from online sources and social media.
“Potential buyers know they can trust these insights when researching and selecting software because they’re rooted in vetted, verified, and authentic reviews,” said Sara Rossio, Chief Product Officer at G2.
## Recent five-star G2 reviews for Airbridge
We would like to express our gratitude to amazing Airbridge users who took time to let G2 know what they think.
For instance, [Sean L.](https://www.g2.com/products/airbridge/reviews/airbridge-review-7401675) said he was very satisfied overall because Airbridge “takes user's feedback seriously” and uses it to “update the product whenever possible”.
We also couldn’t have done it without [Kevin C.](https://www.g2.com/products/airbridge/reviews/airbridge-review-7380926), who called Airbridge a “must-have app marketing solutions for app services” for its intuitive UI/UX and real-time data sharing features.
Of course, we can’t forget [Carl Z.](https://www.g2.com/products/airbridge/reviews/airbridge-review-7279320) who described Airbridge as “the most helpful and valuable tool” for his small business. He added that “one significant upside of using Airbridge is its knowledgeable and responsive support team” who always “respond promptly and with helpful solutions”.
## How Airbridge brings satisfaction to app marketers
Receiving a G2 badge proves all the work the Airbridge team has done to create a powerful, user-friendly, and unified measurement solution. But we’re only getting started. You’ll see more improvements and new features from Airbridge in the coming year.
One of the main objectives of Airbridge as a mobile measurement partner (MMP) is to help marketers around the world discover their true sources of growth. With an average [star rating of 4.7 out of 5.0](https://www.g2.com/products/airbridge/reviews) on G2, this badge speaks for the numerous users who use Airbridge every day to boost their mobile app installs and user engagement.
With unparalleled insights incorporating last-touch attribution (LTA), multi-touch attribution (MTA), incrementality measurement, and marketing mix modeling (MMM), Airbridge enables marketers to thrive at different stages of their marketing cycle even in a privacy-first environment.
Not only does Airbridge employ diverse measurement approaches, it also provides a comprehensive view of marketing performance across web and app, Android and iOS, and SKAdNetwork and non-SKAdNetwork, all in a single unified dashboard. Other tools such as deep linking, marketing analytics, audience manager, and fraud protection also contribute to mobile growth.
## Ready to see why G2 reviewers love Airbridge?
Visit [our website](https://www.airbridge.io/) or [schedule a meeting with our mobile app marketing experts](https://www.airbridge.io/en/book-a-demo) today. You can also check out [our blog](https://www.airbridge.io/blog) for more relevant ad-tech trends and insights.
---
## How to Design a Perfect Paywall for a Mobile App
URL: https://www.airbridge.io/en/blog/perfect-mobile-paywall
Language: English
Published: 2022-12-21
When designing an app, many developers tend to focus on the feature design and the overall UI, sometimes neglecting the paywall – the page where the purchase should happen. 80% of purchases take place after the onboarding when the user sees the paywall for the first time, so it’s important to make a great first impression. Considering the paywall is the place where the user makes their final choice (to buy or not to buy), every aspect must be thought through, including the number of products, pricing, copy, design elements, and more. Our consistent overview will help you understand where to look and what to do when trying to design a perfect paywall for your app.
## Prices and intro offers
Before getting down to the design part, it’s important to refresh on the way mobile apps are priced these days. Most of the apps use in-app subscriptions of various kinds, usually differentiated by the billing period. Here are the most popular ones:
- **Monthly**. It’s the gold standard of mobile subscriptions. With the price being about $5-10, users willingly pay for it if they see the value in the app. It’s pretty affordable and works great for paywalls with 2-3 offers, as well as being the only pricing option.
- **6-months.** This one is usually found on the paywalls with 3 offers, rarely with 2. It serves as a middle ground between the monthly and the annual subscriptions. If you price it right, people might consider it rather beneficial and prefer it to the monthly one.
- **Annual**. If your app is about providing everyday benefits to the user or presenting them with fresh useful content on a regular basis, this type of subscription may work great for you. Once again, if people see the value in your app, they may be willing to pay more. So it’s up to you to price it so that it appears in the most favorable light.
- **Lifetime.** Basically, it’s a one-time purchase that allows the user to use the app for an unlimited period of time. It might be right to use this option at the beginning of your app’s growing process to get some money for future investments, but you should understand that every lifetime user is a missed paying recurrent one.
Some other types of subscription can also be found in apps, such as weekly, 2 or 3 months, etc. They don’t always work for everybody, but they may be useful to experiment with (we’ll talk about this later). As for the pricing itself, the best way is to check the competitors and start with the prices equal or even lower than theirs, but never higher, it’s too risky.
[__wf_reserved_inherit]
What also should be taken into consideration is the intro offers you may want to use. There are 3 of them: **free trial**, **pay as you go**, and **pay upfront.** The first one offers to use the app for free for a limited period of time before purchasing it, while the other two provide introductory prices that make purchasing look more beneficial. With **pay as you go**, you get to pay for the subscription with a discount usually for a couple of months, when **pay upfront** offers a one-time discount. These offers are a great addition to the wisely chosen subscription types and the product structure of the paywall in general, which we’re going to talk about right now.
[__wf_reserved_inherit]
## Number of products
Now that we’re done with the prices and offers, it’s time to decide what structure your paywall should follow, meaning how many products you want to offer. You can try coming at it from a creative angle and choose the one you visually like best, but we’d recommend taking into consideration the pros and cons of each variant. Mobile app paywalls usually offer 1-3 products, depending on various aspects. Here’s what you should know:
- **1 product.** Paywalls with one product are usually pretty straightforward. They clearly state the advantages of the app and prevent the user from the dilemma of which offer to choose. On the other hand, they don’t provide you with much room for price experiments and can’t let you properly understand the paying capacity of your potential subscribers.
- **2 products.** This type of paywall gives you more freedom both in terms of what types of products to use and what prices to place on them. Bear in mind there should be a decent price contrast between the two offers, otherwise the paywall may look confusing.
- **3 products.** Here you get much more room for experiments and get to better understand the preferences of your target audience. Having three different offers on the paywall is a good way to attract the users of any paying segment. But what’s more important, knowing your target audience may help shift them to the product you actually want them to buy more, by making it look more appealing than the other two.
[__wf_reserved_inherit]
## Design elements
When you’re through with the initial pricing options and the product number, it’s time to get creative (but don’t get carried away, there are still guidelines to follow). The visual representation of a paywall may differ drastically from app to app. It’s possible to identify at least [10 most used types of mobile paywalls](https://adapty.io/blog/the-10-types-of-mobile-app-paywalls), but in reality – your paywall is a construction set that you can adjust to your liking. So it’s simply worth knowing what elements a good paywall should consist of and then create the one that would suit your app the best. Let’s go from top to bottom and see what elements are essential.
First of all, to hook the user’s attention, your paywall should include **imagery** that must neatly correlate to the essence of your app. It’s always good to place a stock photo or a specially designed illustration at the top. Some apps even use videos, which can provide even more visual information about the value of your app.
Next comes the **copy.** This is where you must be both creative and straightforward. You can add some artistic touch to the headline but it’s better to be clear and honest when listing the benefits of your app. Your potential users should be able to see the value in purchasing your app, plus, your paywall should contain no misleading information if you want to pass the app store review.
After the benefits are mentioned, it’s time to place the **pricing options**. Usually they are designed as badges with the subscription options info placed inside of them. The most popular way is to make them in the toggle form, so that the user can tap and select the preferred option and then hit the purchase button below. However, sometimes the badges itself bear the purchase button function, which makes the purchase one step quicker.
The **purchase button** is also important, as it basically calls the user to make a certain action. So make sure to find the right word for it without being misleading. For example, writing “Purchase”, “Buy” or “Subscribe” is straightforward and fair, while writing something like “Start your free trial” for a subscription with a free trial may be wrong, as after the trial is over the user will be billed.
Make sure to also add all the necessary **links** down below. Most of the time they are terms of use, privacy policy, and the restore purchase button.
[__wf_reserved_inherit]
## Analytics and A/B testing experiments
Now that you’re done with the creative part, you need to make sure your paywall works the way you intended. To do this, you need to be able to track all the user events (such as page views, button taps, scrolls, purchases, etc.) and view the conversion rates to analyze the performance of your paywall. A convenient way would be to use an analytics system that works great with mobile apps, like [Adapty](https://adapty.io/). It provides precise and accurate analytics and turns all the user actions into user-friendly reports. But this data is not something you can just view and draw conclusions from, it can be directly used to improve your paywall with the help of experiments.
If you look at the analytics of your paywall and see low conversion rates, it’s not a case for disappointment, because you’ve already managed to get this information and now you can use it to improve your paywall. You may build a few hypotheses on what goes wrong, like maybe the price is too high, or the text doesn’t expose the real value of the app. But making blind changes will be too time consuming (making changes in the app, releasing a new version, waiting for another app review) and even dangerous, as you may accidentally drop your conversion rates even lower if the hypothesis was wrong.
The best way to check your hypothesis is to run an A/B test, and Adapty will cover you in this case as well. Running paywall experiments is not that difficult: you create two paywalls, usually with one variation (e.g. different monthly subscription price), split the traffic (usually 50/50, but not necessarily) and after a certain period of time (2-4 weeks) get the results and check if your hypothesis was correct or not. The advantage of this approach is that you don’t risk much even if the hypothesis is not correct, as you split the traffic. But what’s more important – you don’t have to release new versions of the app, as Adapty SDK enables you to make changes on your paywall with the help of remote config, which saves tons of time. Running A/B tests with your paywalls is a sure way to increase your app revenue up to 30% in the first couple of months.
[__wf_reserved_inherit]
## User acquisition and attribution
Split-testing is a great way of finding the most profitable pricing and increasing your conversion rates, but there may be one problem. If your app doesn’t have enough users/downloads/installs, it might be problematic to run A/B tests, as you’d simply not have enough data to get statistically significant results. So it’s also vital to make sure your app is promoted as much as possible. Although there are many [different ways to promote an app](https://adapty.io/blog/how-to-promote-a-mobile-app-7-effective-app-promotion-strategies), counting on the organic traffic only would be rather naive. To get higher UA numbers you may have to invest in advertising, which is usually carried out via store ad managers, ad networks, social networks, or in-app ads. You might want to try all of these variants to find out which works best for you, and [Airbridge](https://www.airbridge.io/) can help you with that. An MMP like this will help you not get confused and properly measure campaign performance across different marketing channels. This way you’ll know which channel to focus on and invest more into. And if your app has a web version, Airbridge supports cross-platform attribution with the ability to view the data of both platforms in one dashboard.
Once you’ve started running A/B experiments and purchasing traffic, you’re sure to stumble upon an idea of making more than one paywall, and it’s always a good thing to try. Different channels may bring different kinds of potential users, so it may be useful to create a paywall for each channel, with unique offers or unique design. And in this case, Adapty is sure to cover your A/B testing needs, when Airbridge will support your attribution analytics. It takes time to find the most profitable source and the most loyal paying customers, but if you regularly experiment and test new hypotheses, your app will grow much faster than if you just made blind decisions.
## Conclusion
We’ve covered all the essential aspects of making a great mobile paywall. To sum up, here are the key points that will make your paywall more appealing to the user and help you pass the store review:
1. **Choose the right prices and intro offers.** Bear in mind what your target audience is and make sure that your offer coincides with their paying capabilities.
2. **Think of the structure.** Choose the number of products on your paywall depending on the value of your app. The more it has to offer, the more subscription types you can try.
3. **Hook the user with imagery.** People perceive visual information much faster and tend to make their choices based on that, so make sure to strike them with the appropriate graphic design.
4. **Make everything clear.** The user should easily understand the value of your app, the price they’re going to pay, the period of time they’re subscribing to, the length of the trial period, etc.
5. **Use analytics and run experiments.** Check the metrics and conversion rates of your paywall to see its strong and weak points. Come up with hypotheses, check them by running paywall A/B tests, and grow your app’s revenue.****
6. **Make sure to attract enough traffic.** Getting organic traffic may not be that easy in the competitive mobile app world. So don’t be ashamed to purchase ad traffic and make sure to take care of your attribution analytics.
---
## Why a Mobile Measurement Partner (MMP) is an App Marketing Must-Have
URL: https://www.airbridge.io/en/blog/why-mobile-attribution-tool
Language: English
Published: 2022-12-16
On average, [Americans check their phones 344 times a day](https://www.reviews.org/mobile/cell-phone-addiction/), or once every 4 minutes. Shocking or relatable? However you feel, this means precisely one thing to mobile app marketers: limitless opportunities. The more often potential users are on their devices, the more likely they are to come across a product and be persuaded to try it out.
In this sense, mobile app marketing may sound like a piece of cake; after all, who could spend a day without using a single app? However, success never comes easy. There is a lot on a mobile app marketer’s plate, from creating engaging content to analyzing target audiences and running paid ad campaigns to reach potential users. And above all, it’s important to always reflect on past performance and develop future strategies.
This is why a mobile measurement partner (MMP) is a must. In this article, we’ll briefly talk about what MMP is and take a closer look at why mobile app marketers need an MMP.
## What is a mobile measurement partner?
An MMP is a third-party attribution tool that collects data to determine which touchpoint takes credit for each conversion. By offering an SDK that can be integrated into an app, an MMP collects data across various channels and numerous devices. Then, it uses its attribution logic to tie campaign engagements (e.g. impressions, clicks) with target events (e.g. installs, sign-ups, purchases). This way, an MMP delivers a unified overview of mobile app marketing performance.
There are different frameworks, or attribution models, for analyzing the value of a touchpoint in a conversion path. The last-touch attribution model, which gives 100% of the credit to the single last touchpoint, has been widely accepted as the industry standard. However, because it overlooks the earlier touchpoints that could have had an incremental impact, the multi-touch attribution model has risen in importance.
## Why do you need a mobile measurement partner?
### Mobile attribution, a ticket to effective marketing
As explained above, an MMP creates a whole picture of app advertising and shows whether marketing efforts are paying off. The ultimate goal of a marketer is to generate revenue and contribute to the profitability of the overall business. For this, two things need to be done: optimizing ad spend and boosting marketing ROI.
Mobile attribution pinpoints the value of specific channels, campaigns, and creatives. Based on this information, marketers can determine where to prioritize and where not to. For instance, if your video ads brought you less users than your display ads with the same budget over the same period, it could be a smarter move to invest more in the latter moving forward. Or else, if you are working with an MMP that provides [incrementality measurement](https://www.airbridge.io/product/mta-incrementality), you would be able to identify which specific element of your marketing has caused a conversion to occur. Then, you can reallocate resources accordingly and enhance budget efficiency.
In short, measuring mobile app marketing performance is at the heart of effective marketing. By understanding which channel, campaign, or creative is contributing most to your performance, you can make a better use of your budget. At the same time, an MMP equips you with data-driven insights that give you a higher chance of optimizing your campaigns. By engaging the right audience through the right channel with the right creative, you’ll be able to improve LTV as well as ROI.
### Handling the intricacies of mobile attribution
We have seen why marketing measurement is indispensable to attaining business success. However, you might still be wondering why it has to be an MMP that does the job.
While there are many marketing analytics tools available in the market, MMPs stand out in that they are specialized in monitoring the mobile environment. Barely a decade ago, a majority of marketers relied on Google Analytics to capture user behaviors and find top sources of traffic to their websites. Cookies, UTM parameters, and referrers came into play for this kind of web tracking.
By definition, mobile attribution is the same as web attribution. However, the methodologies differ because mobile user journeys are fragmented across platforms. When trying to install an app, a mobile user has to move from mobile web or mobile app to an app market. In this process, user identification is lost because app markets essentially operate as closed ecosystems. Marketers do not have the right to access user data, and consequently, it becomes much trickier to track whether the mobile user has actually installed the app after interacting with an ad.
For this reason, many mobile app marketers implement an MMP, a tool developed specifically for mobile attribution. Fortunately, a reliable attribution tool like [Airbridge](https://www.airbridge.io/) allows you to look into campaigns without missing out on any relevant data. To put it another way, using an MMP, you can have a comprehensive and accurate analysis of your marketing performance, which is the prerequisite to mobile growth.
### An impartial arbitrator in the complicated world of advertising
Let’s go back to the example of video ads and display ads. Assume a mobile user saw your video ad but ignored it. The user then came across a display ad the next day, clicked on it, and installed your app. If your chosen attribution model is the last-touch, the channel where your display ad was on should get all the credit and charge you for the install. The problem is, if you’re not using an MMP, there’s a chance that both channels claim credit because ad networks only see their own touches. You might end up paying double the cost you are accountable for.
Between advertisers and media, an MMP serves as a trusted third-party that conducts a fair assessment of marketing performance. It indicates which touchpoint should receive credit for driving a conversion, and thus, how efficiently you are managing your marketing budget. Independent of all other stakeholders in the advertising world, an attribution provider is the single accurate source of unbiased data that adds confidence to your decisions.
Acting as an objective and authoritative judge, an MMP also helps with distinguishing between right and wrong. According to [Interceptd](https://interceptd.com/wp-content/uploads/2019/12/interceptd-q2-2019-mobile-ad-fraud-report.pdf), 25% of Android app installs and 31% of iOS app installs are fraudulent. In addition, ad fraud costs the ad-tech industry billions of dollars annually, and the loss increases as the mobile ad spend grows. Ad fraud is such a pervasive issue that could seriously affect your marketing budget, and you need strong weapons to fight against it.
The good news is that an MMP has the resources and expertise to deal with it. For instance, Airbridge monitors fraudulent activities in real-time to track down suspicious anomalies such as click injection and click spamming. It also provides fraud detection reports so that marketers can evaluate their media sources without bias and without favor. This way, they can pay only for real conversions, avoid double- or triple-charging, and discover true sources of growth.
### Securing time and resources for things that matter
Imagine going back and forth between heaps of spreadsheets, sparing countless resources and many hours chasing and downloading data. If this gives you a headache, you need an MMP.
A mobile attribution tool greatly improves marketing efficiency by providing a unified overview of your performance. It collects all your campaign data and combines it into a single dashboard, making it more convenient to analyze results across channels, platforms, and devices. With this streamlined view of data, you can unlock insights to bring better quality users and generate revenue.
All this is possible thanks to universal SDKs. Each MMP provides its own SDK to measure and attribute for all relevant media sources at once. For example, with the Google Ads SDK, you can only measure performance on Google whereas with the Airbridge SDK, you can track performance on Google, TikTok, Apple, etc. Offering just a single SDK, an MMP connects marketers to the entire mobile ecosystem while reducing room for error through automated data import.
Furthermore, mobile app marketers need the single universal SDK to effectively scale their efforts. The process of adding new SDKs for each ad platform and manually implementing postbacks puts burden on mobile apps with unnecessary codes, incurs cost, and drains technical resources. Without an MMP, it would take much longer to review your marketing performance, eventually resulting in negative impacts on your app’s success.
### Mobile attribution, reinforced with deep linking
Deep linking, the action of redirecting users to a specific in-app page, is an essential element of the growth strategy today. You can improve user experience by allowing users to directly access the content they are interested in within an app, rather than having to navigate through it themselves. This makes it more likely that users will engage with the app.
Deep linking can also win higher conversion rate and retention rate. By sending users to the right destination instead of uniformly redirecting to the app’s home page, marketers can create seamless user journeys and increase the chances that users will complete a desired action, such as making a purchase or signing up for a service. All of these contribute to enhanced LTV and ROI.
Above all, MMPs have deep linking capabilities to allow marketers to accurately track the source of conversions. Deep links not only drive engagement but also provide rich insights into each campaign’s performance which can be used to optimize future strategy. Mobile attribution and deep linking are deeply intertwined, and thus, you should look for a robust mobile app marketing platform that provides both.
### Breaking through iOS 14.5+ and privacy changes
Since the implementation of Apple’s App Tracking Transparency (ATT) framework, [a majority of iOS users have opted out of app tracking](https://www.statista.com/statistics/1281345/apple-att-opt-in-rate-by-app-category/). This means that SKAdNetwork (SKAN) has become an integral part of iOS campaign measurement.
While SKAN has been received with both skepticism and applause within the industry, it is indisputable that the approach poses challenges to marketers who are more familiar with the traditional way of measurement. Without access to the user’s IDFA by default, mobile app marketers have to rethink how to measure and optimize their campaigns.
This is where MMPs come into play. The SKAN postback includes 6-bit binary conversion values containing data about installs, reinstalls, and post-install user activities. Using an MMP, you can tie these different pieces of information with numbers between 0 and 63. Airbridge, in addition, provides the most commonly used schemes as presets, making it easier to choose a conversion value mapping option that best fits your app. What’s more, you can view both SKAN and non-SKAN attribution data on one unified dashboard.
## Any marketer with an app needs an MMP
A marketing tech stack is incomplete without an MMP because it has so much value to deliver that is critical to effective marketing, revenue generation, and ultimately, mobile growth. By collecting data from multiple sources with just one SDK, an MMP greatly saves the time and resources required to understand your mobile app marketing performance. In addition, there are many other core features that help, from deep linking to audience manager and anti-fraud technologies.
If you would like to learn more about what an MMP can do for you specifically for your app business, or if you’ve got concerns about your marketing strategy, feel free to [reach out to our team of marketing experts](https://www.airbridge.io/en/book-a-demo) and have a chat.
---
## How Marketing Mix Modeling (MMM) Enables Privacy-First Measurement
URL: https://www.airbridge.io/en/blog/airbridge-marketing-mix-modeling
Language: English
Published: 2022-12-12
Marketing seems to be a bit more challenging every single day. You try your best to break through fragmented media consumption habits by juggling a dozen different channels at once. Yet, data privacy comes up, and your ability to track and measure performance is tightened. You want to maximize the impact of your marketing efforts, but you’re not sure which activities are driving profit. If this sounds like you, we’re here to help.
Powered by advanced machine learning, [Airbridge](https://www.airbridge.io/) MMM unlocks powerful insights to help you make current and future marketing decisions with confidence. Keep on reading to find out what marketing mix modeling is and why you need to add Airbridge MMM to your measurement stack to stay ahead in the ever-changing marketing world.
## What is marketing mix modeling?
Marketing Mix Modeling (MMM), sometimes referred to as media mix modeling, is a statistical analysis technique that assesses the impact of marketing inputs on desired business outcomes such as sales, conversions, and installs. Taking various factors into account, from channel-specific ad spend and promotions to seasonality and market elasticity, MMM provides a holistic perspective into growth. In essence, it is a top-down big-picture approach that uses aggregated historical data to help marketers forecast returns and optimize future budgets across an assortment of channels.
> 💡 To get a more detailed explanation on the concept of MMM, [check out our white paper](https://www.airbridge.io/library/mmm-basics-whitepaper?utm_source=blog&utm_medium=cta&utm_campaign=mmm-basics-whitepaper).
## Why is marketing mix modeling important?
### The rise and fall of marketing mix modeling
The concept of MMM was first introduced in the 1950s and saw a rise in demand in the 1980s, back when it was hard to have enough insights into user behavior. To be more tactical with their ad spend, marketers worked with econometricians and statisticians from consulting firms and received modeling recommendations.
However, the popularity did not last due to operational challenges. Built on tech stacks and data scientists, one MMM project could cost millions of dollars. It also took a long time to count all the variables, build models tailored for each company, and interpret the results. That wasn’t all – manual data transfer was not only cumbersome but also had a lot of room for error. The hurdles were high and the use of MMM was practically exclusive to large advertisers developing long-term strategies.
### The impact of privacy changes on marketing mix modeling
By this point, you might be assuming MMM is an outdated technique. The thing is, it does have a long history, but it’s on everybody’s lips these days. Since the iOS 14.5 updates, the App Tracking Transparency (ATT) initiative has shifted the digital marketing landscape and paved the way for the return of MMM.
During the past decade, targeted advertising allowed marketers to drive user engagement and build brand awareness. However, since April 2021, all iOS apps have been required to ask user permission to access the Identifier for Advertisers (IDFA) and collect user-level data. In fact, as of March 2022, [more than half of iOS users worldwide had decided to opt out of tracking](https://www.statista.com/statistics/1281345/apple-att-opt-in-rate-by-app-category/). The impending deprecation of third-party cookies on Chrome and regulations like the GDPR and CCPA are adding to the difficulties. Such changes also place limitations on last-touch attribution (LTA), which relies on device-level tracking.
All of the above have pushed marketers today to look for new measurement approaches that meet their objectives but still protect users. After a long search, their eyes fell on MMM, an oldie but goodie.
### Bringing SaaS to marketing mix modeling
MMM is clearly suited for modern day marketing – it uses aggregated channel-level data rather than detailed user-level data. However, as aforementioned, not everyone can afford the time and money that traditional MMM projects require. This is why many next-gen tech companies like Airbridge are providing MMM as a software solution, to make MMM more accessible by lowering the hurdle.
In specific, Airbridge MMM has the following competitive advantages:
- **Usable:** The solution is cost-effective and easy to implement.
- **Adaptive:** Data can be fed to the model as you prefer.
- **True-to-life:** The model can be calibrated to reduce the gap between the model and reality.
- **Practicable:** The solution not only analyzes but also predicts and prescribes.
## Why should you use Airbridge MMM?
### Reason #1: Usable, affordable, and convenient
Since business outcomes usually have annual seasonality and trend, traditional MMM models needed at least one year worth of data. Building a model using such a large amount of data was time-consuming, which meant MMM was not sufficient for day-to-day optimization.
On the other hand, the Airbridge MMM model, driven by Airbridge’s proprietary machine learning algorithms, can be trained with only six months worth of data. This significantly reduces the modeling cost, whether it be time or money. Moreover, several types of reports based on MMM results are each updated on a daily, weekly, and monthly basis. Marketers can draw further insights from these reports to respond to sudden changes and adjust strategies accordingly.
### Reason #2: Adaptive and flexible
Future-proofing measurement with MMM is incredibly easy for those already using Airbridge. As a [mobile measurement partner](https://www.airbridge.io/en/blog/attribution-tool-mmp) (MMP), Airbridge provides an API for full S2S integration and the Airbridge SDK which can be embedded within your apps to auto-collect performance and cost data. Airbridge also tracks all in-app events and uses this additional relevant data to train the MMM model.
Don’t worry even if you’re not working with Airbridge yet. You can run the analysis as long as you have six months worth of data on channel-level spending, daily clicks, daily impressions, etc. Prepare your data with a CSV file and import it to Airbridge, and the remaining modeling and reporting processes are automated. Even if you import data manually, Airbridge MMM will deliver results faster than traditional MMM projects.
No matter how you feed data, the Airbridge MMM model will never stop improving to show you the most up-to-date results. Powered by advanced machine learning, the model will be tailored to suit your business needs and contribute to your success.
### Reason #3: True-to-life, reliable, and accurate
MMM quantifies the correlation between marketing inputs and historical business outcomes. In other words, the model, which is an imperfect representation of reality, can be subject to the correlation vs. causation fallacy. To provide sensible outputs, Airbridge calibrates the model with alternative approaches to understanding the marketing mix.
Calibration generally involves well-constructed experiments such as lift tests. Even though these guarantee the highest reliability and accuracy possible, they often incur financial costs and slow down the entire analysis process. Hence, for those who can’t afford the luxury of running real-world experiments, Airbridge calibrates the MMM model with incrementality measurement results brought out using propensity score matching (PSM), which is a quasi-experimental method.
### Reason #4: Practicable and actionable
Airbridge MMM is much more agile than traditional MMM, and agility leads to workability. Using three different types of reports, marketers can make an effective use of MMM in their everyday decision-making situations.
**Marketing Mix Analysis**
[__wf_reserved_inherit]
The Marketing Mix Analysis report, updated every day, scrutinizes the incremental contribution of each channel without relying on user-level data. Not only digital channels but also non-digital channels such as TV, radio, billboards and magazines can be attributed with MMM. Choose from performance indicators spanning across installs, orders, revenue and eCPI to refine your analysis.
**Budget Optimization**
[__wf_reserved_inherit]
The weekly and monthly Budget Optimization report finds the best marketing mix to maximize your budget potential in the future. As the model already has your historical cost data, all you have to do is to decide what percentage of your current budget to allocate as your future budget. After putting in the number and previewing the selected amount, you can simulate the model and see how much your marketing performance will increase by optimizing according to the prescribed budget plan.
> 👉 Fashion meta-search platform meliz said Airbridge MMM proved to be a reliable privacy-first solution for their cross-channel marketing measurement. The results showed that the team would have seen a 5% increase in installs if they followed the MMM-prescribed budget plan. Find out the details in this [case study](https://www.airbridge.io/en/case-studies/meliz-privacy-first-mmm).
**Model Manager**
[__wf_reserved_inherit]
With the Model Manager report, you can get a quick overview of key information of your MMM models such as analysis target, input channels, training status, and last training time. Click on “View Details” for extra details like input data type and calibration status.
## Quick start your MMM journey with Airbridge
MMM is just one type of marketing measurement approach you can work with, but it’s an important one to consider.
By examining marketing data holistically, MMM allows marketers to break silos in analyses and prove how their efforts help overall business. Moreover, it is a privacy-safe option that all marketers have been looking for in a privacy-first era.
If you’re interested in trying it out, Airbridge MMM is the way to go. [Talk to our measurement experts](https://www.airbridge.io/en/book-a-demo) and set up a data-driven strategy that will get you the most return on your investments.
---
## Airbridge Now Integrated With RevenueCat to Maximize Your Revenue Tracking
URL: https://www.airbridge.io/en/blog/airbridge-revenue-cat-partner
Language: English
Published: 2022-12-05
We are delighted to announce that Airbridge is now integrated with RevenueCat, a leading subscription solution platform.
With this integration, clients can now enjoy an additional layer of depth when analyzing which channel contributed the most to subscription revenue with metrics such as trials, initial purchases, renewals, and cancellations available for analysis on the Airbridge dashboard.
Please find the [Airbridge User Guide](https://help.airbridge.io/hc/en-us/articles/12964722155929-RevenueCat) for instructions to integrate. If you are interested in exploring more details on Airbridge integration with RevenueCat, please do not hesitate to [contact](mailto:platform@ab180.co) us to learn more!
## About RevenueCat
RevenueCat is a subscription solution provider. Companies use RevenueCat to implement in-app purchases, analyze, and grow revenue on iOS, Android, and the web. RevenueCat provides a backend and wrapper around Apple’s StoreKit and Google Play Billing to simplify implementation and manage subscriptions. No server maintenance or backend code is required. Click [here](http://revenuecat.com) to learn more about RevenueCat.
---
## Intro to Incrementality: The Key to Measuring Your Ad Effectiveness
URL: https://www.airbridge.io/en/blog/marketing-incrementality
Language: English
Published: 2022-11-30
Marketers have always wanted to know whether the time, money, and effort they had put into marketing activities are causing users to convert. To find the answer, you need to be able to distinguish the extent of organic traffic from the impact of your paid ad campaigns.
The catch is that traditional attribution alone is not sufficient for causal inference, thus telling a fragmented story, if not inaccurate. Moreover, there can be additional issues related to budget and marketing mix resulting from incomplete marketing performance measurement.
Then how do you prevent such chaos and take a step closer to success? Incrementality can show you the way.
## What is incrementality in marketing?
Incrementality is the measure of supplemental business outcomes driven by specific marketing efforts. By identifying the causal relationship between touchpoints and conversions, marketers are able to determine which events would not have occurred in the absence of a particular interaction. In other words, incrementality demonstrates the true power of each component of your marketing mix that is otherwise difficult to isolate and assess.
While the concept of incrementality is rooted in science, in the marketing world, it is essentially used to give an understanding of ad effectiveness. This allows marketers to deal with the two essential issues: justifying ad spend and increasing budget efficiency.
**1) Validating your marketing budget**
Profit is inarguably the ultimate business objective. To get there, marketers need to make sure that the return on investment is high and that such a result was driven by marketing activities. The problem is that the Last-Touch Attribution (LTA) model, which is the industry standard today, has limitations in capturing causal effects because it allocates 100% of the credit for a conversion to the single last touchpoint. The model oversimplifies the complex user journey and ignores the contribution of many earlier touchpoints as well as users’ latent inclination to convert.
Thankfully, incrementality measurement could be a way to solve the limitations of LTA, highlighting the significance of marketing campaigns in conversion generation. If you can prove that the conversion would not have happened without your marketing efforts, you can explain why your company needs to invest in marketing.
**2) Optimizing your marketing mix**
Incrementality also helps with channel-level analysis. By identifying the higher-performing channels, you can focus on ad platforms that bring tangible results, thus preventing budget waste and improving marketing efficiency.
There is always an overlap between organic traffic and paid conversions, but the separating line is often blurry, leading to less-informed decisions. This is where incrementality measurement comes in, to reveal which of your marketing components are creating actual value. In specifics, it provides answers to the following questions:
- What happens if you stop running ads on a particular channel?
- What role does a certain channel, campaign, or creative play in increasing revenue?
- Which channel, platform, publisher, or campaign has the biggest influence on metrics such as profitability and user lifetime value (LTV)?
- How much should you spend more or less on which channel to maximize performance?
Working with incrementality allows you to get the most accurate, actionable insights into the impact of your marketing activities. Not only can you see the causality between touchpoints and conversions, you can clearly determine the best set of paid ads across various channels.
## How to measure incrementality
There are several ways to measure incrementality, but **experimental** and **observational** are the principal approaches. An experimental study divides users into a test group and a control group to compare the performance of the two groups and measure the lift, whereas an observational study uses historical data to draw causal inferences.
The **experimental method** produces more accurate results, but it involves difficulties as well. For instance, there must be data science experts on your team to create two groups that share similar characteristics. Still and all, no matter how carefully planned, it is almost impossible to make the groups perfectly identical. To handle such a situation, you can get help from Meta and Google’s Randomized Controlled Trials (RCTs) for incrementality testing, which randomly allocate users to two groups.
Yet, experiments inevitably entail another drawback – time, cost, and resources. Turning off ads to the control group during the study implies a lost opportunity for sales. Moreover, it takes time to confirm and analyze the results, meaning that applying yesterday’s data in today’s decision making is not an option. This is why many martech tools like Airbridge have chosen to work with the **observational method**, providing machine learning-driven insights into user behavior.
### 1. Experimental: Meta & Google
Top ad platforms Meta and Google offer lift measurement tools that control the delivery of ads to groups with similar attributes. Based on real-world experiments and data, not only are the results standardized, accurate, and reliable, but they also effectively show the causal effect. Let’s dig deeper into how each ad platform evaluates incrementality.
Meta’s lift tests help you see the true value of your Facebook ads. This experiment splits your chosen audience into groups who did and did not see your Facebook ads to understand their causal impact on specific business objectives such as brand awareness or sales. The control group here would be those who match your audience but are intentionally kept away from your ads.
There are three kinds of lift tests you can run on Meta:
- [Conversion Lift tests](https://www.facebook.com/business/help/221353413010930?id=547299432790676): See the incremental lift of your Facebook ads for conversion goals.
- [Brand Lift tests](https://www.facebook.com/business/help/1693381447650068?id=546437386202686): Figure out the incremental impact of Facebook ads on brand awareness.
- [Experiments](https://www.facebook.com/business/help/1915029282150425?id=2564729006895902): Answer predefined questions about your Facebook ads.
> 👉 Learn more about [Meta's lift tests](https://www.facebook.com/business/help/552097218528551?id=546437386202686)
Google’s lift tests measure the effectiveness of your Google ads, which you can use to adjust and improve your campaigns. Since it is not available for all Google Ads accounts, you need to contact your Google account representative to measure incrementality on Google.
There are two kinds of lift tests you can run on Google:
- [Brand Lift](https://support.google.com/google-ads/answer/9049825): Evaluate the effectiveness of your video ads.
- [Conversion Lift](https://support.google.com/google-ads/answer/12003020): Measure the number of conversions, site visits, and other actions directly driven by your audience viewing your ads.
> 👉 Learn more about [Google's lift tests](https://support.google.com/google-ads/topic/10283810)
### 2\. Observational: Matching & Propensity Score Matching
As explained earlier, the observational method can be an alternative to the experimental method which requires time, cost, and resources. However, when using historical data, it is impossible to randomly allocate the users into treatment and control groups.
To overcome such a challenge, **matching** is widely used in the industry. By finding similar observable characteristics between two user groups and excluding all unmatched users from the causal analysis, matching enables a comparison of outcomes while minimizing selection bias caused by covariates.
**Propensity Score Matching (PSM)** is another popular technique that uses scores to estimate the probability of a user being exposed to a particular ad. After matching users with similar propensity score values from the test and control groups, all unmatched users are discarded and the actual impact of an ad is found by comparing the performance of the two groups.
## Why incrementality measurement is a must
It is every marketer’s goal to get the most truthful and complete insights into marketing performance and drive growth. While LTA has been highly in demand thanks to its convenience, it is neither a perfect nor the only way to measure marketing performance, especially when it comes to evaluating the true power of your marketing activities.
Incrementality measurement gives a holistic perspective for identifying the most effective campaigns and optimizing paid ad strategies. If you would like to find out the real value of your marketing, [contact our team of experts](https://www.airbridge.io/en/book-a-demo) to discuss. You can also check out our [white paper](https://www.airbridge.io/library/incrementality-whitepaper) to take a deep dive into the fundamentals of incrementality measurement.
---
## Introducing Our Updated Retention Report
URL: https://www.airbridge.io/en/blog/airbridge-retention-report-update
Language: English
Published: 2022-11-15
We proudly announce the official rollout of the Retention Report in Airbridge! The official Retention Report is equipped with behavioral cohort analysis and provides more options for Groupby and Filter selection than the previous beta version.
The cohort analysis feature is powered by Luft, our OLAP database, which was introduced in our previous post, '[Introducing Luft - Airbridge's in-house OLAP Database.](https://www.airbridge.io/en/blog/introducing-luft)'
_**The beta version of the Retention Report is no longer available._
The Airbridge Retention Report provides an N-day retention rate for users who have performed a specific event in-service and came back to the service on a specific date and performed a designated event. The Report shows how well your users are engaged with your app.
The report helps you determine which channels and campaigns bring in the most loyal customers to your app. The results can be used for campaign optimization and ad cost settlement.
## Retention Report Update Highlights
1. You can choose Start Event to view the retention rate of users who performed those events.
2. Measurement Option enables you to analyze retention by Return Event from the User Journey perspective.
3. You can define the cohort that you want to analyze the retention and view the retention data by cohorts.
[]
### 1\. Set Start Event
You can set the **Start Event\***, which is the first engagement of the user with the app within the analysis period to see the retention data of user groups by the performed Event type.
- Start Event options: Installs(App), Deeplink Opens(App)
### 2\. Use the Measurement Option for a detailed view
You can customize the retention view using the Measurement Option. With ‘General’ setting, you can view the retention data by unique users, and with ‘Confined’ setting, you can view the retention data by user journeys separated by Start Events. Additional options will be coming up soon!
- General: This setting is recommended when you need to view the overall user engagement within the analysis period. Even if a unique user started multiple User Journeys within the analysis period, Return Events performed during overlapping User Journey periods are considered for retention calculation for all User Journeys.
[]
- Confined: This setting is recommended if you need to view the retention per User Journey. If a unique user started multiple User Journeys within the analysis period, the Return Events performed are counted for retention calculation of individual User Journeys that are separated by Start Events.
[]
### 3\. Set a cohort
[]
You can create a user segment by using the **cohort** settings and view the retention of that particular user group. This allows you to understand what influence difference Events have on retention.
\*Cohort: A cohort is a set of users that can be grouped based on commonalities. In the Retention Report, a cohort is a user segment that performed a certain Event with certain Event Properties.
### Let’s create a user segment that ‘completed at least one order in the past 30 days’ and view the past 30 days retention data of the users whose Start Event was app install.
1. Click the “+ Add” button under \[Cohort\] and create a user segment that completed at least one order in the past 30 days by setting the filters should be set like the image below:
[]
2\. Set the Start Event as “Installs(App)” and the Return Event as “Any Event(App).”
[]
Read more about how to use and benefit from the Retention Report in [the user guide](https://help.airbridge.io/hc/en-us/articles/4405204740249-Retention). Discover the best performing channels that bring in most loyal users to your app with the new Airbridge Retention Report and use the findings to optimize your marketing strategy!
---
## Mobile App Attribution: What It Is and How It Works
URL: https://www.airbridge.io/en/blog/what-is-mobile-app-attribution
Language: English
Published: 2022-11-10
A mind-boggling amount of data – [2.5 quintillion bytes](https://www.domo.com/learn/infographic/data-never-sleeps-5) – floods out of the connected devices we use every day. Savvy marketers have set on “data-driven” marketing to capitalize on such data abundance, but impactful insights are rare. Having access to data does not exactly mean knowing what to do with it.
The struggle is obvious when it comes to measuring marketing performance. A survey conducted by DemandGen revealed that [more than a third of marketers](https://www.demandgenreport.com/resources/research/marketing-measurement-attribution-survey-report-b2b-organizations-double-down-on-tried-and-true-methods-with-a-focus-on-tracking-digital-engagement/) admitted they needed improvement in understanding marketing effectiveness. What is more, the pandemic has accelerated growth in mobile device usage, pushing marketers to master not only web attribution but also the more complicated mobile attribution.
That is why in this post, we’re walking you through what mobile app attribution is and how it works. Read until the end to effectively drive growth. Let’s dive right in.
## What is attribution?
Marketing attribution, in general, is the practice of matching causes such as channels and campaigns with effects such as purchases and revenue. Simply put, it is the process of understanding the impact marketers made on a specific goal.
As Theodore Roosevelt once said, the more you know about the past, the better you are prepared for the future. An accurate analysis of historical data is a prerequisite for business success. Building a winning paid marketing strategy is especially important because it involves cost – companies cannot remain in business without turning a profit. Given that global digital ad spend was around [$378 billion](https://www.statista.com/statistics/237974/online-advertising-spending-worldwide/) in 2020 alone, marketers must identify the best working channels and campaigns to improve return and profitability within a limited budget.
All of this might sound complex and challenging. Thankfully, with marketing attribution, you can clarify the connection between inputs and outputs and unlock insights vital to growth.
## Web attribution vs. mobile attribution
Market share of mobile devices was as high as [58.33%](https://gs.statcounter.com/platform-market-share/desktop-mobile-tablet) in October 2022, but before such domination, marketing measurement was more focused on desktop web. With web attribution using cookies, UTM parameters, and referrers, marketers could capture certain user behaviors and track traffic sources to their websites.
By definition, mobile attribution is the same as web attribution. However, tracking mobile user journeys is much more difficult because user identification is lost as users move from mobile web to app markets. Mobile apps were built without the notion of cookies, and the “walled gardens” in the mobile ecosystem essentially make it impossible to take the approach that works on the web.
For this reason, many app marketers use app attribution tools or Mobile Measurement Partners (MMPs) like [Airbridge](https://www.airbridge.io/). They provide a streamlined view of data from multiple sources, reducing the burden of going back and forth between various ad platforms. Preventing ad fraud and adapting to new frameworks like SKAdNetwork are some other benefits of working with attribution providers.
## Essential ingredients for mobile attribution
App attribution tools collect and process user-level data, offering insights into the power of marketing efforts. Specifically, they look into user journeys which consist of touchpoints and conversions.
[Mobile User Journey]
**Touchpoints** are clicks, impressions, and other external events that lead to app installs or in-app actions. There are three main sources of touchpoint data:
- A **tracking link** is a regular URL with tags added to the end. These tags, also known as parameters, contain information on user behavior. You can create a tracking link using an app attribution tool.
- **Server-to-Server (S2S) API**, also known as server-side tracking, generates a unique ID when there are app or web activities, sending the data from the user to you before it is transmitted elsewhere. Self Attributing Networks (SANs), in particular, use S2S as they attribute their traffic without involving a third party. Unlike regular ad networks, SANs only report the last-touch activity to attribution providers via API.
- A **referrer** is a unique string of characters specific to Android devices. This anonymous ad tracking ID is sent to Google Play when a user clicks on an ad, and if an app install or an in-app event happens consequently, the result is passed back to the app attribution tool.
**Conversions** refer to in-app events such as install, purchase, and add to cart. Most marketers use app attribution tools or MMPs for conversion tracking.
- Mobile apps can directly implement **SDKs** **provided by app attribution tools**. Your app code lets the attribution SDK know when different actions take place inside your app, and the SDK sends the events back to the attribution provider.
- **Server-to-Server (S2S) API** can be used to track not only touchpoints but also conversions.
## How does mobile attribution work?
Mobile attribution is all about associating touchpoints with conversions. This process, called **identity matching** or **identity resolution**, ties and attributes all a user’s behavior into a single profile. In other words, app attribution tools track user behavior scattered across multiple devices and link their interactions with specific ads. There are two primary matching methods:
- **Deterministic methods** use personally identifiable information such as device ID, prioritizing accuracy. The identifier – GAID for Android devices and IDFA for iOS devices – assign an identity to a mobile user to track that person across web and app. Since iOS 14.5, mobile apps have to request user consent to access IDFA.
- **Probabilistic methods** create device relationships by compiling individual pieces of information such as a device’s operating system, IP address, Wi-Fi network, etc. Devices are grouped together based on the data and associated with an identity that is “probably” accurate. These methods are often used when deterministic data is not available.
While identity matching is the foundation of mobile attribution, the story is a bit different when it comes to SANs. As explained earlier, these “super publishers” measure performance on their own. When a SAN reports user activity to an MMP through an API after the conversion has taken place, the MMP asks the SAN the device ID of that event, and the SAN provides relevant details so that the MMP can draw additional insights from the conversion data. Note that with IDFA deprecated in iOS 14.5, device IDs are no longer used in reporting on iOS devices.
StoreKit Ad Network (SKAN) is another type of independent attribution. Developed by Apple to attribute iOS campaigns in a privacy-preserving way, the SKAN API hides all user-level or device-specific data and measures on an aggregated level. When a user converts, the winning ad network receives validation postbacks from Apple, and then sends them to an MMP which is responsible for connecting the dots to make sense of the data.
Holistic attribution allows marketers to optimize marketing strategy, but it is extremely demanding for an individual app to integrate, report, verify, visualize, and analyze on its own. Hence, for mobile attribution, it is recommended to work with an app attribution tool or an MMP.
## The ground rules of mobile attribution
When pairing touchpoints with conversions as described above, you need to decide how much credit to give to each touchpoint in the conversion path. This is where the concept of **attribution model** comes into play. By assigning values to users’ interactions with ads, marketers can determine which channels and campaigns are the most powerful.
In the majority of cases, user journeys consist of multiple touchpoints, meaning that there are various ways to attribute. The two primary approaches are rule-based and data-driven. Popular attribution models such as the single-touch and the multi-touch are all rule-based.
Most app attribution tools are built on the simple and heuristic Last-Touch Attribution (LTA) model, which gives full credit to the single last touchpoint. Nevertheless, keep in mind that each tool works under different priorities, rules, and criteria, and thus the attribution result for the same conversion could slightly vary.
> 👉 [Click here](https://www.airbridge.io/en/blog/what-is-marketing-attribution-model) to read about attribution models you must know as a marketer.
## Drive growth with mobile attribution
Mobile attribution is a retrospective causal analysis that helps you identify your strongest channels and campaigns. While this is in itself meaningful, what is more important is to keep adapting in the ever-changing marketing landscape. There is no one-fits-all approach to marketing success, and only by consistently and proactively modifying your action plan can you seize opportunities.
Airbridge aims to take a marketer-centric approach and build an app attribution tool that actually benefits marketers. As your reliable marketing partner, Airbridge will continue to provide industry-leading solutions to prepare you for the future.
> 👉 [Check out this post](https://www.airbridge.io/en/blog/airbridge-unified-measurement-stack) to find out about Airbridge’s innovation in tackling marketing challenges.
---
## Your Basic Guide to Marketing Attribution Models
URL: https://www.airbridge.io/en/blog/what-is-marketing-attribution-model
Language: English
Published: 2022-10-06
What drove your customers to install that app or make that final purchase? Was it the new Instagram feed or the paid search ad? Should you be spending more on email campaigns or social media ads? The challenge marketers face today lies in distinguishing high-performing campaigns from underperforming ones in a complex world of concurrent marketing campaigns across multiple channels.
This is where attribution models come in. **Attribution models serve as the lynchpin of digital marketing and assist you in making informed marketing decisions that make sense.**
As a winning marketer, you should understand the strengths and weaknesses of different attribution models and determine what works best for your marketing success!
Let's start with the basics.
## What is an attribution model?
An attribution model is used in marketing analytics to give conversion credit to touchpoints across the customer journey. Conversion generally refers to an in-app event such as an app installation or purchase, and touchpoints are all engagement points in the customer journey, such as ad clicks and views.
To gain maximum return from a limited marketing budget, **you need to know which ad campaign was most effective in driving the final conversion** or which channel acquired the most users that led to the most conversions. Attribution models can provide the insights you need for effective media mix planning amid challenges posed by the rising complexity of the customer journey.
[__wf_reserved_inherit]
There is no one-fits-all model that conveniently feeds you with the data you need, and different attribution models have different strengths and weaknesses. Therefore, you need to know which model to leverage in a particular context for your product. Attribution models are largely categorized into **rule-based models** and **data-driven models**.
## Single-touch Attribution Model
The single-touch attribution model, one of the most widely used rule-based models, gives full attribution credit to only one touchpoint, no matter how many touchpoints the customer went through until the conversion. The first-touch and the last-touch attribution models are the two variations of the single-touch attribution model.
### First-touch Attribution Model
The first-touch attribution model gives full credit to the first touchpoint for the conversion made. This model is recommended when your product has a short sales cycle, and you are running only a few concurrent campaigns within a limited number of channels. Using the first-touch attribution model, you can determine the most effective marketing channel to acquire customers. Focusing your ad spend on this channel may increase your product awareness among a wider audience pool.
Using the first-attribution model means focusing on top-of-the-funnel marketing, which aims to raise brand awareness and generate new leads. As straightforward as it is, the model is easy to implement. However, the first-touch attribution model has its downsides as the model ignores all the other touchpoints beyond the first touch. As a result, the data may lead to an under-investment in retargeting or CRM campaigns and a loss of additional conversion opportunities.
In particular, this model is unsuitable for high-involvement products with a long user journey, such as cars, or services with extended intervals between use, such as travel services. The first touchpoint may lie outside the lookback window and gets lost in the data collection process, failing to measure attribution.
> 💡 A **lookback window** or an **attribution window** is a defined period in which touchpoints such as ad clicks and impressions are considered for attribution. The length of the window should vary depending on the product and marketing channel.
### Last-touch Attribution Model
The last-touch attribution model is the most commonly used model, which gives full credit to the last touchpoint for the conversion made. This model is helpful for products that have already established brand equity and need to focus their marketing efforts on the final conversion. The model provides insights into late-stage campaigns that triggered the ultimate purchase decision and can help you invest in high-performing campaigns. In particular, the last-touch attribution model comes in handy for products with a relatively long sales cycle or when implementing the multi-touch attribution model is too complex and requires a simpler way.
However, the last-touch attribution model ignores all other touchpoints except for the final touchpoint that led to the ultimate conversion. It does not consider the incremental impact of multiple touchpoints or the effect of brand equity and oversimplifies the complex customer journey.
## Multi-touch Attribution model
The multi-touch attribution model addresses the limitations of the single-touch attribution model. Today, the customer journey has become more complex than ever, spanning multiple devices and channels. In this ever-competing digital marketing world, the multi-touch attribution model is the key to success in optimizing your marketing mix.
The multi-touch attribution model assigns conversion credit to different touchpoints with weights based on varying logic. The model looks into various touchpoints along the path toward conversion, including paid ad campaigns and owned media, and provides a granular and person-level view. Here is a list of multi-attribution models that [Google Ads](https://ads.google.com/home/) currently provides its users for performance measurement.
### Linear Model
The linear model gives equal credit to all touchpoints across the customer journey. This model is useful when the advertiser values all campaigns evenly and wants to see them on equal footing. However, suppose the customer journey is highly complicated, or a particular channel has a significant influence on the market. In that case, the linear model can not provide the real story behind the conversions.
### Time Decay Model
The time decay model gives more credit to the touchpoint closer to the final conversion. The closer to the conversion, the higher the attribution weight is given to the touchpoint. This model is useful for high-involvement products and products with long sales cycles as it focuses more on touchpoints at the bottom of the marketing funnel. However, the model could mislead to underinvestment in early-stage touchpoints that are effective in acquiring customers, resulting in the loss of conversion opportunities.
### Position-based Model
Google Ads assigns 40% of the credit to the first and last touchpoints each and the remaining 20% to the in-between touchpoints. This is called the U-shaped attribution model. This model is your pick if you value the first touchpoint for brand awareness and the last touchpoint for conversion as the two most important touchpoints within the customer journey. Unfortunately, the model is not suited for products with a relatively short sales cycle. The model is also not recommended for products with user journeys where the impact of respective marketing channels is unclear.
## Data-driven Attribution Model
The data-driven attribution model assigns the attribution credit based on user-level data such as ad views or clicks. Using this model, you can determine which keywords, ads, and campaigns significantly influence conversions. Here are two methods that are commonly used in the data-driven attribution model:
- The **Shapley Value** is a solution concept that compares the conversion outcome across random variations of touchpoint arrangements and measures a touchpoint's average marginal contribution to the conversion. It provides the conversion likelihood of particular touchpoints by comparing the conversion likelihood with or without particular ad engagements. Google Ads uses this concept for its data-driven attribution model and provides insights through Google Analytics on various types of conversions driven by Google Search (including Shopping), YouTube, and other display ads.
- The **Markov Chain** is a stochastic model used for marketing attribution to figure out the transition probabilities, which is the probability of moving between 'states,' which are marketing channels within the customer journey. The impact of a particular marketing channel is determined by calculating the conversion rate measured without that channel.
The data-driven attribution model is based on real-life data and drives a more accurate view of the performance of individual marketing campaigns. For this reason, Google Ads applies the data-driven attribution model for channel attribution. This data-driven approach captures the complexities of today's customer journey and provides granular insights for effective channel planning.
## Pick the Attribution Model that Makes Sense
No one-size-fits-all marketing attribution model is available for every marketing context, and choosing one is down to individual business needs. Therefore, the first step to measuring your actual marketing performance is understanding the pros and cons of various attribution models. You may pick a model from the list above or customize a model to your needs.
You can experiment by comparing the results of different models and find the one that is well-aligned with your marketing strategy. Along the way, you may also feel the need for marketing mix modeling that provides a macro view to set a long-term, large-scale marketing plan leveraging various channels and tactics, including mobile, web, TV, and outdoor advertising.
Remember, marketing optimization starts with measuring marketing performance with the right tools. If you want to know how Airbridge can help your company with accurate performance measurement, [request a demo](https://www.airbridge.io/en/book-a-demo) today!
---
## Airbridge’s Unified Measurement Stack: Marketer-Centric Approach to LTA, MTA and MMM
URL: https://www.airbridge.io/en/blog/airbridge-unified-measurement-stack
Language: English
Published: 2022-09-28
Accurately analyzing marketing effectiveness is essential to improving performance and growing business. However, privacy policies like the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) have posed challenges to the collection of user-level data, and marketers have been pushed to find ways to measure without violating user privacy.
The last-touch attribution (LTA) model, which has always been a trusted and popular choice, also adds confusion. For sure, its simplicity and cost efficiency is unbeatable, but there’s a possibility of the model not telling the whole story. Top 10 Forbes marketer Neil Patel even dared to say that [LTA is telling blatant lies directly to your face](https://neilpatel.com/blog/last-touch-attribution-lies/).
With the marketing and advertising environment rapidly changing, it is crucial for marketers to choose a mobile measurement partner (MMP) that puts itself in the shoes of marketers. Hence, in this blog post, we’ll be discussing the steps Airbridge has taken to address marketers’ concerns and take attribution to the next level.
## Airbridge’s innovation in tackling marketing challenges
[__wf_reserved_inherit]
### MTA and incrementality measurement overcoming limitations of LTA
LTA has been the standard option for measuring marketing performance because it simplifies the complicated process of attribution analysis. The model tracks with great certainty the channel that serves as the last touchpoint, making it easier to optimize campaigns and creatives. This shows why most MMPs such as Airbridge use LTA.
However, LTA has its own limitations in that it gives 100% of the credit for a conversion to the single last touchpoint in a conversion path and does not dig into earlier touchpoints that could have had an impact on the user. In fact, Airbridge data revealed that at least 30% of conversions occur after two or more touchpoints. Moreover, given users’ latent inclination to convert, there is a chance that the impact of the last touchpoint could be overestimated.
Airbridge believes multi-touch attribution (MTA) and incrementality measurement could be the solution. The MTA analysis allows marketers to accurately see how each touchpoint contributes to conversion and the Airbridge incrementality model helps measure the true effectiveness of marketing in influencing user conversion.
Airbridge uses Propensity Score Matching (PSM) as well as other statistical methodologies to determine marketing incrementality. The **Incrementality Report**, updated every day, shows the incremental growth derived from each marketing channel, facilitating media planning and efficient budget allocation. Furthermore, by adopting an observational method rather than an experimental method, Airbridge provides a more flexible view of the incremental effectiveness of marketing campaigns.
### MMM stimulating growth through privacy-first measurement
User-level mobile attribution is becoming increasingly difficult due to global rise in data privacy regulations. In response, Airbridge proposes Marketing Mix Modeling (MMM) as an alternative to measure marketing performance without collecting user-level data.
MMM uses aggregated historical data to explain the relationships between marketing activities and conversions. There are two primary approaches: Frequentist and Bayesian. MMM came on the scene decades ago when companies with big ad spends were looking for ways to determine the right media mix. Back then, MMM solutions were not easily accessible or affordable because they were long-term high-cost consulting projects, requiring at least six months to one year’s worth of data.
However, MMM is making a comeback as the use of traditional marketing analytics has been limited by its reliance on user data. Combined with LTA and MTA, MMM, which is an aggregated data form of modeling, can effectively attribute iOS campaigns even in iOS 14.5+. It can also be used to measure the performance of Connected TV (CTV) and Digital Out-Of-Home (DOOH) ads.
Another reason why MMM is on the rise is that it can make predictions and run simulations. For instance, you could get recommendations on how to reallocate your ad spends to maximize sales moving forward. Mobile attribution, technically, is based on a retrospective approach, but MMM could take it to the next level, prescriptively determining the best course of action for a given situation. For such reasons, there are B2B SaaS companies trying to lower the bar and let more marketers use MMM with ease.
Airbridge, as a proud participant of Meta’ts MMM SaaS Incubator Program, has been developing various MMM features using the Frequentist approach. Powered by advanced machine learning, Airbridge’s proprietary model measures the performance of each marketing channel and suggests ways to maximize budget efficiency. The **Marketing Mix Analytics Report** will show how the model has been trained and the **Budget Optimization Report** will tell how to best allocate within a fixed budget and maximize performance.
Nonetheless, MMM does have limitations because it estimates marketing impact on historical business outcomes and even makes projections based on probability. Airbridge therefore plans to improve accuracy through regular calibration.
### Unified Measurement Stack: an integrated approach
[__wf_reserved_inherit]
In an effort to relieve marketers’ concerns over the limitations of LTA and tectonic shifts in the marketing industry, Airbridge has been proactively investing in research and development. We’re building the **Unified Measurement Stack** which combines LTA, MTA, and MMM to be able to measure incremental effectiveness of marketing activities without solely relying on user-level data. Our ultimate goal is to capitalize on the strengths of each approach and provide a balanced view.
Airbridge aims to take a step further, striving to find what marketers’ worries are and how to solve them. We believe that our job does not end by simply selling our product, but that we have to guide marketers through the world of marketing analytics, contributing to the growth of their businesses.
## Which solution should you choose for your product?
So far, we’ve explored the journey Airbridge has taken to establish Unified Measurement Stack. In fact, marketers are often confused using LTA, MTA and MMM because they are based on different data sets and models, meaning that they will most likely draw different conclusions. Moreover, being able to choose among various approaches could only add confusion about what to use when. Hence, in this section, we’ll discuss how you can best use these three solutions.
What is best for your product depends on its characteristics and the market environment, and you need to build a tailor-made Unified Measurement Stack. However, rather than jumping to the conclusion that there is no single answer, Airbridge would like to take responsibility as a solution provider and tackle challenges together with marketers.
Below is our suggestion on how to use each solution to adopt the **full-cycle marketing strategy** along the user journey. Continue to read to imagine your own Unified Measurement Stack.
[__wf_reserved_inherit]
### MMM for the strategy development stage
MMM stands out in that it uses a time series regression model based on various types of aggregated data. Therefore, it is more suitable for figuring out overall trends or general projections than drawing specific insights. In addition, the aggregated data allows to measure performance of offline channels such as TV ads and magazines, from which it is impossible to collect impressions or click data.
Considering such characteristics, MMM can be best used **for marketing strategy development**. Again, MMM reads the trends and suggests directions by evaluating both digital and non-digital channels that could drive growth. If you’re using Airbridge, you’ll be able to benefit from the MMM-based Budget Optimization Report to find out your most effective budget allocation.
### MTA and incrementality measurement for the monitoring stage
With Airbridge’s incrementality model, you can discover the true incremental effects of each marketing channel you’re using. In other words, this is a great way to attribute multi-channel campaigns than to see through general trends.
This is why we recommend using MTA and incrementality **for measuring and monitoring the results** of your marketing campaigns. You can reveal channels that provide incremental growth without heavily relying on the cost calculated by ad networks. Identifying the most effective channels and allocating more budget to them can help improve marketing performance.
### LTA for the execution and revamp stage
LTA is the most commonly used and the most granular way to measure marketing performance as it is based on raw data. Hence, this solution comes in handy **when you’re adjusting your marketing strategy to optimize campaigns and creatives**.
LTA might not be the ideal option to check the overall efficacy across various marketing channels. However, it could be useful when you are changing the details of your strategy after distinguishing the most effective channels and creatives with MTA. In Airbridge, for instance, you can refer to the Actuals Report to see the performance of each campaign and creative and draw insights for your next campaign.
## Airbridge’s marketer-centric approach
As emphasized earlier, Airbridge hopes to build the **Unified Measurement Stack relevant to marketers’ day-to-day responsibilities**. We try to think from marketers’ perspective and go the extra mile to be a reliable partner through the different stages of full-cycle marketing.
As a marketer-centric solution provider, Airbridge prioritizes the needs and wants of marketers. Thus, our MTA and MMM solutions, two of the three pillars of the Unified Measurement Stack, have been developed with some unique features introduced below.
### Observational model that minimizes cost and time
Incrementality can be measured using either experimental or observational methods. Meta and Google’s lift studies are based on the experimental method. Users are randomly divided into a test group and a control group, and the test group sees the ads whereas the control group does not. While this guarantees accuracy, it takes a relatively long time and incurs financial costs.
On the other hand, Airbridge has adopted the observational method, meaning that we use machine learning and statistical methodologies to calculate incrementality. This requires shorter time and lower cost than conducting actual experiments, and even more, marketers can get a daily update on the incremental effect of their campaigns.
In addition, every three to six months, Airbridge compares measurement values delivered by its observational model with those of a calibration standard of known accuracy. Korean health brand Dano ran studies and found that Airbridge’s MTA study results were significantly higher than the LTA results, which better aligned with its actual business results. Compared to Meta’s lift study, the MTA results were only slightly lower. This gave Dano further confidence in its campaign and the company is still working with Airbridge to gain accurate insights.
> 👉 [Read more about Dano’s case study](https://www.facebook.com/business/success/dano)
### MMM that stands strong in a fast-changing environment
MMM is great to be used at the strategy development stage of full-cycle marketing. However, since it relies on historical data to determine the average of various marketing results, it can reflect recent changes only to a limited degree. The problem is that today, the marketing landscape is always shifting and that marketers have to respond immediately, coming up with effective strategies.
Fortunately, Airbridge’s MMM solution is more responsive and flexible than traditional ones. Our model, powered by advanced machine learning, can be fully trained with only three months worth of data. We provide daily updates so that marketers can better understand how different channels have contributed to their targets and adjust their media mix accordingly. Moreover, going further from retrospective observation, Airbridge’s MMM proactively and prescriptively recommends media mix to maximize performance.
## What’s the next step for MMPs?
Many solutions have been trying to overcome the limitations of LTA and deal with the privacy-first era. With more choices, marketers are asking for a unified approach to measure performance more efficiently and effectively.
Airbridge believes that our role does not end at simply providing solutions, but rather, goes beyond building the Unified Measurement Stack with LTA, MTA and MMM and making sure that it adapts to the shifting expectations of marketers. We’re not only a measurement partner but also a marketing partner, and will continue to share various success cases and profound insights.
---
## Integrate Airbridge with SearchAds.com for a deeper analysis of your Apple Search Ads campaigns
URL: https://www.airbridge.io/en/blog/airbridge-searchads-com-partner
Language: English
Published: 2022-09-26
Airbridge is happy to welcome [SearchAds.com](https://searchads.com/) as an official integration partner! We are excited to announce that we are an official partner with SearchAds.com and our clients now can leverage SeachAds.com to optimize Apple Search Ads campaigns through this partnership. You can also directly send attribution data from Airbridge to SearchAds.com with the API.
As Airbridge notices your custom goal events like Installs or certain In-App Events attributed to Apple Search Ads, the API automatically transmits those conversion data to SearchAds.com. It enables SearchAds.com to optimize Apple Search Ads campaigns based on accurate conversion data.
For simple instructions to set up the integration, please find the Airbridge [User Guide](https://help.airbridge.io/hc/en-us/articles/900005290943-SearchAds-com). In addition, if you are interested in exploring more details on Airbridge integration with SearchAds.com, please don't hesitate to [contact us](mailto:platform@ab180.co).
## About SearchAds.com
SearchAds.com powered by [MobileAction](https://www.mobileaction.co/) is an official [Apple Search Ads partner](https://searchads.apple.com/partners/) that delivers seamless campaign management and automation experiences to make Apple Search Ads campaigns more efficient and profitable.
SearchAds.com’s all-in-one platform enables marketers to find the best keywords, scale campaigns, and maximize app store reach with actionable insights.
SearchAds.com will shrink hours of work into minutes with automation, bulk editing, and reliable keyword data for marketers to optimize ROI. Learn more at [searchads.com.](https://searchads.com/)
---
## Mobile Measurement Partner: Why Should You Use Them and What Are The Best Alternatives?
URL: https://www.airbridge.io/en/blog/attribution-tool-mmp
Language: English
Published: 2022-09-08
Savvy app marketers use various tools to make data-driven decisions. Connecting a mobile attribution software to your app, in particular, helps you efficiently measure and optimize your campaigns. However, you might not be sure whether you’re using such a software due to its many names – MMP, mobile attribution provider, mobile app analytics software, third-party attribution platform, etc.
While all of these refer to the same concept of **measuring marketing performance**, there’s a variation because each name puts emphasis on different aspects of a mobile attribution software. Hence, understanding where these names come from can help you get a grasp of why you need to use one of these softwares for your app marketing success.
Let’s take a deep dive into the names, one by one.
## Mobile attribution software
The name “mobile attribution software” speaks for itself; it is a set of programs that is used to attribute, or to determine which touchpoint takes credit for each conversion. It collects user-level data to tie successful app events such as installs, sign-ups and purchases to the campaign or channel that created the engagement.
App marketers must have insights into how their actions impact revenue in order to improve their return on investment (ROI). They need to understand which activities performed better and why. With mobile attribution, you can measure and optimize marketing performance metrics in real time across different marketing channels, devices and ad networks.
There are different frameworks, or **attribution models**, for analyzing the value of a touchpoint in a conversion path. The last-touch attribution model, which gives 100% of the credit to the single last touchpoint, has been popular. However, as it doesn’t consider the earlier touchpoints that could have had an incremental impact, the multi-touch attribution model has become critically important.
## Mobile Measurement Partner (MMP)
Mobile Measurement Partners (MMPs) are a subgroup of [Meta Business Partners](https://www.facebook.com/business/marketing-partners) that provide third-party measurement of mobile app ads through the installation of a Software Development Kit (SDK). The name has become so widely known that these days, it is no longer specific to Meta but is considered synonymous with a mobile attribution software that has integrated with the bigger networks like Self-Attributing Networks (SANs).
An MMP helps apps collect data on various events in the mobile environment to measure campaign performance and give credits to appropriate touchpoints. It should be noted that technically, only mobile attribution softwares that are in partnership with Meta are actual MMPs.
> 💡 **Airbridge is the world’s first company to have been badged as Meta’s Mobile Measurement Partner as well as Multi-Touch Attribution Partner.** Check out the details of the partnership [here](https://www.airbridge.io/en/blog/airbridge-becomes-facebook-business-partner-for-measurement) and Airbridge’s profile on Meta’s Partner Directory [here](https://www.facebook.com/business/partner-directory/search?solution_type=measurement&ref=fmp_about_solution_card&capabilities=Mobile%20Measurement&id=3963759510351117§ion=overview).
The benefit of using an MMP is that it provides a unified overview of campaign data across multiple channels and networks, making it more convenient to compare results. For instance, with an MMP, you can analyze your ad performance from Facebook, Google, Apple, etc. whereas with the Facebook SDK, you can only measure Facebook ad performance. In addition, you can send in-app event data collected by an MMP to Facebook to use it for retargeting and optimization.
Furthermore, an MMP could lessen the burden if you’re using Apple’s SKAdNetwork (SKAN) to measure iOS 14.5+ app install campaigns. The SKAN postback includes conversion values that provide post-install information, and you can use an MMP to configure the app events to track. For Facebook, you can use the Facebook SDK or App Events API as well to import your event configuration schema to Meta.
> 👉 Find out more about [configuring SKAN in Meta Events Manager
> ](https://www.facebook.com/business/help/188126096313109)👉 Find out more about [setting conversion values in Airbridge](https://help.airbridge.io/hc/en-us/articles/900006402223-SKAN-Conversion-Value)
You might be confused sometimes because your MMP and Meta Business Suite are showing different results. This usually happens if: your MMP and Facebook are using different attribution models; the attribution windows, or the time periods, are set differently; and/or your MMP and Facebook report conversion data at different times.
## Mobile app analytics tool
A mobile app analytics tool, in essence, serves the same purpose as a mobile attribution software. Both of them capture data on the way users interact with your app to help you see the consequences of changes and fine-tune your product at a granular level.
However, they should never be considered identical since mobile app analytics is the broader concept. It allows marketers to see what draws users to the app, how long visitors typically stay in the app, what factors are correlated with in-app events like purchases, and much more. On the other hand, **mobile attribution is specifically focused on measuring where users come from** by collecting touchpoint and conversion data.
As the diagram above shows, the three different terms explained so far should be categorized separately. This indicates that there are mobile app analytics tools that are not necessarily mobile attribution softwares, and mobile attribution softwares that are not MMPs.
[]
Here are some examples of mobile app analytics tools:
- Amplitude
- Countly
- Google Analytics 4
- Mixpanel
Here are some examples of mobile attribution softwares:
- Firebase
- Tapstream
- Tenjin
Finally, here is a full list of Meta-badged Mobile Measurement Partners:
- Adjust
- Airbridge
- AppsFlyer
- Branch
- Kochava
- Singular
## Third-party tracking platform
The name “third-party tracking platform” implies that it is an impartial observer that applies a single unified set of criteria to measure marketing performance across various channels. Although some ad networks provide attribution results themselves, it is important to involve a third-party due to three main reasons.
First, a third-party tracking platform allows you to take a **holistic approach to marketing measurement**. You might have had experiences creating a dashboard on Google Sheets to bring together results of different creatives and channels. While this could be tedious and time-consuming, the worst part is that you cannot be sure of the accuracy of your assessment if your attribution rules and marketing metrics are not standardized. Using a third-party tracking platform could help greatly with getting full performance visibility and identifying the true sources of growth. In fact, Airbridge is the only third-party tracking platform that has mastered unified web and mobile measurement, offering a comprehensive view of marketing performance from iOS, Android and desktop apps.
[Airbridge Attribution Rule settings page]
Second, mediation of a third-party **guarantees objectivity and equitability** across different channels. Let’s assume that an ad network is analyzing the performance of various ad networks including itself. Then there’s a high chance that it would draw conclusions in favor of itself to receive a larger sum from the total budget. However, if a third-party steps in, it becomes possible to calculate and settle ad spending more accurately. In addition, a third-party tracking platform can set [fraud rules](https://www.airbridge.io/product/fraud-protection), effectively preventing and detecting any possible ad fraud.
Lastly, third-party tracking is **suitable for responding to privacy changes**. Apple’s iOS 14.5 update brought one of the most significant and controversial changes the industry has ever seen. Since April 2021, advertisers have been required to request consent before tracking user activity. Fortunately, third-party tracking platforms have come up with measures to deal with such situations and maintain marketing effectiveness. Airbridge, for instance, has introduced features related to [deferred install tracking, simple automatic conversion values and conversion bits](https://www.airbridge.io/product/ios-skan) to adapt to the changes.
## Measure growth with utmost accuracy
Airbridge is a mobile app analytics tool, a mobile attribution software, an MMP and a third-party tracking platform. By understanding what each term refers to, you can enhance your understanding of tools that provide you with actionable marketing insights.
If you are interested in building improvements and optimizing campaigns through web and mobile measurement, incrementality analysis and multi-touch attribution, Airbridge should be your choice of tool. [Request a demo today!](https://www.airbridge.io/en/book-a-demo)
---
## Four Steps to Successful QR Code Marketing
URL: https://www.airbridge.io/en/blog/airbridge-qr-code
Language: English
Published: 2022-08-31
The Big Game has always been popular among advertisers thanks to its large captive audience. In fact, a 30-second ad slot for Super Bowl LVI was as expensive as $6.5 million in 2022. While there are questions about whether these ads really work, at least Coinbase did it right and took advantage. Its $14 million ad, which simply [bounced a QR code around on a black screen](https://www.youtube.com/watch?v=eIUD_NE1BDo), led to a 309% increase in installs week-over-week, followed by another 286% climb the next day.
As proven by the crypto app’s case, QR codes can be a massive punch in your marketing campaign when deployed well. Beyond simply directing users to a web or app page, these black-and-white bar codes can achieve a variety of goals including raising brand awareness and generating leads.
Let’s take a deep dive into **how useful QR codes can be in marketing** and **how Airbridge’s trackable QR codes with deep linking capabilities are even better**.
## The comeback of QR codes
QR codes – or quick response codes – are two-dimensional codes consisting of black cells on a white background. Invented in 1994 to provide easy access to information, QR codes reached a peak of popularity in 2011, saw a decline afterwards, but have come back recently.
[__wf_reserved_inherit]
[__wf_reserved_inherit]
If the first QR code boom came with the spread of smartphones, the “renaissance” was due to the pandemic, as the need for contactless user experience grew. For instance, many restaurants replaced physical menus with QR code-based digital menus and QR code mobile payment was widely adopted. For-sale signs in front of houses also include QR codes to provide more details these days.
## Four steps to a successful QR code marketing campaign
Nowadays, QR codes are found just about anywhere from receipts, product brochures, and T-shirts to billboards and even TV commercials. Their use cases are as diverse, from directing users to a web page to sharing information in various formats. Such omnipresence of QR codes make them an excellent marketing tool when in the right hands.
However, despite the great potential, there is no guarantee that a user will install your app and make a purchase just after scanning the code. Then, how can marketers make a success out of QR codes? Here are the four steps you should take.
### 1\. Follow your target audience to place your QR codes
Understanding your target audience is always important in marketing, even when deciding where to place QR codes. You want as many people as possible to scan your code, and for this to happen, it should be in areas your target audience frequents and can easily take out their phones. Identify who’s engaged with your brand to provide potential users with the right content at the right time.
**For instance, QR codes can be placed at restaurants to provide seamless digital experiences.**
- Visitors will scan a QR code to read menus or product details on their own devices. Then, they will proceed with contactless ordering and payment – everything like before, but safer.
- Alternatively, the QR code may show various content such as a video about brand values or an infographic on nutritional facts, all of which help raise brand trust when in-person communication is limited.
**But they don’t have to be at the restaurant – consider where your audience is and what they are looking for.**
- To target parents who want to get done with dinner preparation while picking up kids, you can hang flyers with a QR code at school gates and direct them to your app where they can place takeaway orders at nearby restaurants.
- Similarly, you can create ad posters with a QR code for food delivery and post them in subway stations to target hungry passengers on their way home.
### 2\. Give the reason to scan your QR codes
Once you have decided where to place your QR code, the next step is to make people actually scan it. The key here is to specify the incentives to do so – what information does the QR code carry and why does it matter? Be persuasive and transparent so that you can give your target audience the reason to pause for a moment, take out their phones and check out what you are presenting.
Here are some of the values and benefits you can propose:
- Offer discounts and promotions to those who scan your QR code.
- Link instructional videos to your QR code and put it on product detail pages.
- Add Google Maps URL to your QR code to provide directions to your store or restaurant.
Using QR codes in marketing helps with user acquisition, which is the first and foremost stage of the AARRR framework. It could also reduce marketing spendings on paid ads by sending users straight to a web or app page. In the same vein, you can minimize competition on paid channels since user engagement via QR codes is not only direct but also instant.
### 3\. Streamline user journey using QR codes with deep linking capabilities
Now that you’ve got people to scan your QR code, you want to minimize their chances of dropping off by creating a seamless user experience. The web or app page linked to your QR code should be mobile-friendly and meet the audience’s expectations. You also need to check if your code works and update it whenever there are updates.
[__wf_reserved_inherit]
With Airbridge, you can **generate QR codes for your Custom Channel tracking links**. This feature allows you to personalize user experience by directing each user to a different destination. In addition, Airbridge’s QR codes have deep linking capabilities to route users to specific in-app content, thus making the user journey even smoother and preventing churn.
> 📌 A Custom Channel refers to an owned media or earned media channel which is not officially integrated with Airbridge, such as the product or service’s website, e-mail, blog, and social media. You can type in the name of the channel of your choice when you create a link for your Custom Channel.
To give you an idea of how to generate QR codes and tracking links in Airbridge:
1. Select channel, campaign, ad group, ad creative, etc. to define Campaign Optimization Parameters.
2. Set the redirection path to one of the three: an app store, the app (via deep link), or the website. Users can be sent to different destinations depending on their device type.
3. Register a thumbnail that acts as a preview of your tracking link when shared or posted online.
4. Hit the Create Link button and check out your tracking link as well as your QR code.
The strength of Airbridge’s QR codes is that you don’t have to generate a new code every time you modify your deep link or redirection path. In other words, you can guarantee a streamlined user experience all the time.
### 4\. Track, collect and analyze user data to maximize marketing effectiveness
**Using trackable QR codes with deep linking capabilities, like those of Airbridge, can be a great way for you to capture first-party data in this cookieless world.** Furthermore, with the collected data and the help of an attribution solution, you can measure the performance of your marketing campaigns. Airbridge’s QR codes in particular allows you to conduct cross-device analysis for various marketing channels, even for the trickier ones like TV ads.
With Airbridge’s QR codes, you can learn more about:
- Channels that are effective in acquiring target audiences
- The impact of a marketing campaign on long-term user retention and engagement
- Changes in post-acquisition user behavior for each channel.
## What’s next?
We’ve explored the four steps to maximizing marketing effectiveness with QR codes. Specifically, we’ve seen that trackable QR codes with deep linking capabilities, when placed at the right place for the right audience, could serve as a powerful tool for user acquisition and behavior analysis.
Don’t miss out on the benefits of using QR codes in your marketing. If you’d like to learn more, [talk to our marketing experts at Airbridge](https://www.airbridge.io/en/book-a-demo) and grow your business!
---
## How to Fix When Safari Cannot Open the Page Because the Address Is Invalid
URL: https://www.airbridge.io/en/blog/deeplink-101-ios-safari-alert
Language: English
Published: 2022-08-25
Deep linking is the key to successful app marketing. When we say a deep link is working, we expect it to take users smoothly to the app, or use fallback behaviors to redirect them to an app store or a web page, minimizing churn. However, the world is not perfect and it is not easy to create deep links that meet app marketers’ high standards.
[Previously](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers), we took a look at the three types of deep links and the difficulties of handling them. **URI schemes, in particular, have no built-in fallback option. This means that users without the app will see an error message or nothing upon clicking the link.** In response, the two biggest mobile operating systems - Android and iOS - each came up with its own solution.
Nevertheless, you might have noticed that in iOS Safari, scheme-based deep links lead to unexpected alert dialogs at times. How can app marketers overcome such trouble and build a seamless user journey?
## URI scheme fallback for Android
When deep linking with URI schemes in Android, you have two ways to redirect users to a fallback page.
### JavaScript
You can use iframes to load the deep link and redirect the user to Google Play with a delayed JavaScript. This popular option involves the process of setting an **iframe src’s attribute**. Below is an example code snippet of which **airbridge://** is the custom scheme. If the app is installed on the user’s device, the app will open and JavaScript won’t run.
[How to Fix When Safari Cannot Open the Page Because the Address Is Invalid]
However, this is only available in Chrome for Android 18 and earlier and in Android browsers. For Chrome 25 and later, the JavaScript solution stopped working.
👉 Learn more from [Chrome Developers documentation](https://developer.chrome.com/docs/multidevice/android/intents/)
### Android Intent
Chrome versions 25 and later don't support JavaScript and URI schemes won’t work even if the user has the app installed. Instead, you should use the “**intent:**” syntax.
When a clicked link or programmatic request invokes a web URI intent, the Android system tries each of the following actions, in sequential order, until the request succeeds:
1. Open the user's preferred app that can handle the URI, if one is designated.
2. Open the only available app that can handle the URI.
3. Allow the user to select an app from a dialog.
Below is an example code. To learn how to add intent filters, check out [Android Developers documentation](https://developer.android.com/training/app-links/deep-linking).
[__wf_reserved_inherit]
## URI scheme fallback for iOS
You may assume that the same smooth process applies to iOS. **However, as of iOS 9.2, URI schemes for deep linking bring users to an abrupt halt when triggered from Safari.** Bypassing the problem with JavaScript was a valid option for iOS 7 and iOS 8, but the good old days are gone and now all users clicking on a scheme-based deep link in Safari are shown an alert dialog as below.
If the app **is not installed**, the error message reads, “Safari cannot open the page because the address is invalid.”
[__wf_reserved_inherit]
If the app **is installed**, the error message reads, “Open this page in ‘{appName}’?”
[__wf_reserved_inherit]
Apple has taken such measures mainly due to security reasons; given that it is impossible to prove ownership of or reserve a particular scheme, multiple apps may register the same scheme and create URI conflicts. In addition, routing users from Safari to an external app by triggering a deep link with JavaScript could bring unintended results. This is why Safari shows an alert dialog when a scheme-based deep link is used.
This causes inconvenience not only to many of the users who use Safari as their default browser, but also to marketers who are trying to benefit from deep linking. Obviously, alert dialogs act as a barrier, and users will leave if they can’t get straight to the in-app content of their interest.
As an alternative to URI schemes in iOS, you can use the verified URLs - **Universal Links**. However, as Universal Links can’t send a user to the App Store, the deep link may not work as the marketer intended if the user doesn’t have the app installed.
Airbridge’s team of experts are putting continuous efforts into solving the problems explained above. If you want to take your app marketing to the next level with deep linking, [contact us today](https://www.airbridge.io/en/book-a-demo).
---
## MGS 2022: Unified Measurement Stack - MMP, MTA, MMM and Lift
URL: https://www.airbridge.io/en/blog/mgs-2022-unified-measurement-stack
Language: English
Published: 2022-08-24
This July, [Airbridge](http://airbridge.io) successfully hosted the **Modern Growth Stack 2022 (MGS 2022)** adtech & martech conference in Seoul. We brought together around 40 industry leaders and 2,000 attendees to share growth strategies and tactics in the era of digital products.
Out of all the insightful sessions, here’s a recap of “**Unified Measurement Stack - MMP, MTA, MMM and Lift**” by our Chief Strategy Officer, Yongchoul Han. With the advertising and marketing market changing like never before, how should marketers measure performance?
## Emerging challenges in marketing measurement
### Privacy changes restricting user data collection
Apple’s iOS 14.5 update included privacy-first features such as the App Tracking Transparency framework. Since then, all iOS apps have been required to ask users for permission to share their data, meaning their access to the Identifier for Advertisers (IDFA) keys has been limited. In the meantime, Google announced plans to phase out third-party cookies in its Chrome browser by 2024. Privacy policies like the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) also pose difficulties for user tracking, targeting, and attribution.
### Limitations of last-touch attribution
The **last-touch attribution** model gives 100% of the credit for a conversion to the single last touchpoint in a conversion path.
It has been one of the most popular attribution models as it is the more simplistic, cost-efficient and easy-to-implement model. However, it doesn’t consider earlier touchpoints that could have had incremental impact, risking giving incomplete conversion insights. The fact that it is a single-touch attribution model also makes it more difficult to draw insights for multi-channel marketing campaigns.
## How Airbridge is responding to the challenges
### Multi-touch attribution (MTA)
With Airbridge, you can measure the true incremental effectiveness of marketing in influencing user conversion. This is possible because our MTA model reveals the essential touchpoints and how they each contribute to the eventual conversion.
Airbridge leverages extensive partnerships, including the [MTA partnership with Meta](https://www.airbridge.io/en/blog/airbridge-becomes-facebook-business-partner-for-measurement), to gather data for not only the last but also the earlier touchpoints. Our Privacy-Enhancing Technologies (PETs) allow us to measure growth while ensuring users privacy.
### Marketing Mix Modeling (MMM)
MMM is a statistical analysis method for estimating return on investments in different marketing channels. It’s a proven tool to optimize marketing budget allocations and boost growth.
Airbridge uses regression or Bayesian models to calculate marketing incrementality. Powered by advanced machine learning and statistical methodologies, our model can predict future conversions based on given inputs in each campaign. It also helps companies remain compliant in the era of digital privacy by using aggregated data instead of individual user data.
### Unified measurement stack: an integrated approach
[Airbridge’s Unified Measurement Stack]
Han recommends building a unified measurement stack which combines MTA, MMM and last-touch attribution model. By flexibly working with all three of them depending on the changing situation and the characteristics of your digital product, you can achieve the most balanced view of measurement.
### Finishing up with observational and experimental analyses
Airbridge’s unified measurement stack - MMP, MTA, and MMM explained above - is based on an observational model. It is a time-efficient method that gives you full control over campaign execution and performance measurement. However, there are also limitations as statistical modeling uses assumptions to make predictions about the real world, unlike that of the experimental model.
While the experimental model is certainly more accurate as it is an actual experiment, it involves a higher cost - designing and takes longer time to conduct. You also do not have access to the results until the experiment is completed.
Hence, the ideal approach is to use the observational model for day-to-day analysis and conduct experiments every three or six months to check how the results are alike. In other words, the two methods complement each other and it is the marketer’s job to be adept at working with both.
> Korean health brand Dano ran Airbridge’s incremental MTA study as well as a Meta conversion lift test. Compared to the lift study, the MTA results were only slightly lower - showing just 2% fewer app installs and 3% fewer purchases - giving Dano further confidence in its campaign, and convincing the team to have greater trust in Airbridge’s MTA model. Check out the full case study [here](https://www.facebook.com/business/success/dano).
## Airbridge’s approach to marketing measurement
Airbridge uses a last-touch attribution model as other mobile attribution solutions do, but also works with MTA and MMM to measure true effectiveness of marketing campaigns.
By the end of his session, Han emphasized that marketers should not solely rely on a single attribution model and that Airbridge’s goal is not just to measure, but to contribute to marketing success through effective and accurate measurement.
If you want to have a deeper discussion with our team of experts on marketing measurement, [contact us today](https://www.airbridge.io/en/book-a-demo).
---
## Why Every Marketer Should Use Branded Links
URL: https://www.airbridge.io/en/blog/airbridge-branded-link
Language: English
Published: 2022-08-19
Branding matters. You may have heard the quote from Steve Forbes, “Your brand is the single most important investment you can make in your business,” and he couldn’t be more accurate. In addition, a NewsCred study revealed that [69% of millennials believe brand recognition is the most important driver of brand loyalty](https://www.businesswire.com/news/home/20150212005205/en/NewsCred-Study-Reveals-62-Millennials-Feel-Online).
As a marketer, you would already know that a strong brand-driven marketing strategy is the key to turning prospects into customers. However, [the branding process takes at least months](https://www.forbes.com/sites/forbesagencycouncil/2019/12/03/why-branding-takes-so-long-and-why-its-worth-the-wait/?sh=3cc717f54028), forcing you to turn to performance marketing for more immediate results. As a result, you just keep going back and forth between setting the longer-term direction and executing the shorter-term actions.
Thankfully, branded links can make your life easier. Read until the end to find out how you can benefit from them, from boosting your branding efforts to measuring performance.
## How do branded links help your branding?
### Why branding is important
Before moving on, let’s have a look at an excerpt from a [Forbes article](https://www.forbes.com/sites/forbesagencycouncil/2021/03/24/the-importance-of-branding-in-business/?sh=153e8e6e67f7) to see the value of **branding** one more time.
> “A business’s branding is more important than you might think. On the outside, your brand may seem like it consists only of elements such as logos and colors, but your brand is actually the entire identity of your business. Your brand gives you personality.
>
> Branding has always been a vital part of business, but it may be more important now than ever before. With social media, consumers get exposed to new brands every day. This can be great for consumers who have plenty of options and are able to do research to find the best one, but it makes it harder for businesses.”Published on March 24, 2021
> by Kristopher Jones, Forbes Councils Member
In fact, we can often see many marketing campaigns that have already incorporated various brand elements these days. However, the importance of branding links which users directly interact with may have been overlooked. **Branding your links might seem like a small change, but it could make a big difference.**
### What is a branded link?
A **link** is, ad litteram, something that connects and provides access to various web pages and apps. It plays an essential role in bridging one part of the digital world to another, allowing web and app pages to be discovered. This is why marketers shorten links or use Open Graph meta tags and customize the title, description, and preview image of their content - to get clicks and increase engagement.
The more fundamental approach, however, is using **branded links** to tailor the look and feel of your links. In the simplest terms, a branded link is a **short custom domain**. It helps you stand out from competition and familiarize users with your brand.
## The all-in-one Airbridge branded link
Creating branded links is made easy with Airbridge. You just need to define your domain, redirection path, and Open Graph meta tag for each of your brands when generating **tracking links for your Custom Channels**.
👉 [Find out how to create tracking links with Airbridge](https://help.airbridge.io/hc/en-us/articles/900001037886-Tracking-Link-Generation)
Airbridge’s branded links help redirect users whilst simultaneously measuring performance. In other words, it is an all-in-one link that strengthens your brand-building efforts, attributes your marketing performance, and creates a seamless user experience with deep linking.
> 📌 A Custom Channel refers to an owned media or earned media channel which is not officially integrated with Airbridge, such as the product or service’s website, e-mail, blog, and social media. You can type in the name of the channel of your choice when you create a link for your Custom Channel.
>
> 📌 An Integrated Channel refers to a channel officially integrated with Airbridge, such as Moloco, Appier and Remerge. Check out the full list [here](http://abit.ly/integrated-channels).
## Five ways to use branded links
Here are some practical tips for you to make the most out of Airbridge’s branded links.
### 1\. Create branded links with Custom Domain feature
Airbridge offers a Custom Domain feature with which you can create a tracking link for your Custom Channel using a branded domain instead of a default domain provided by Airbridge (abr.ge). Rather than working with URL shorteners like bit.ly and abit.ly, you can use Airbridge to build URLs around your brand name and make them simple and catchy.
[__wf_reserved_inherit]
To add custom domains to Airbridge, you can either create a subdomain of your existing domain or purchase a new domain you like. Keep in mind that the subdomain should not be in use at the moment. The first example in the table above is a subdomain whereas the second and third examples are additional domains.
With the brand name in the domain, sharing your links means sharing your brand, which could lead to wider exposure and higher brand awareness. Moreover, users are more likely to click on branded links as they seem to provide quality content in the brand’s name. Furthermore, if the users actually find the content helpful, they will know that your brand is one they can rely on.
### 2\. Further your branding efforts with Link Short ID
Once you have registered a custom domain, you can customize the Link Short ID of your Custom Channel tracking link. If you leave it blank, a random, unmodifiable address will be generated.
A **Link Short ID**, also known as a URL slug, is the last part of a full URL unique to a web or app page. With Airbridge, you may create an ID that matches your campaign or content.
[__wf_reserved_inherit]
A well-designed Link Short ID improves user experience as it allows users to expect what they are about to see before clicking on the link. It also helps with Search Engine Optimization (SEO) because the keywords in your URL signal Google that your content should be included on the results page. In other words, your content gets to be considered more trustworthy and more likely to be discovered by the right users.
Here are some cases where branded links with Link Short IDs come in handy:
- **eLearning companies** can use branded links to promote specific courses or instructors. In order to stand out and build a reputation, they can put the company name in the domain and the course or instructor name in the Link Short ID.
- **eCommerce companies** can use branded links to increase sales of specific products, putting the company name in the domain and the product name in the Link Short ID.
### 3\. Use deep linking for consistent user experience
As mentioned earlier, Airbridge’s branded links serve both as a tracking link and a deep link. You can set the final destination to one of following:
- An app store, if you wish to increase app installs
- An in-app content, if you wish to show specific app page
- A web page, if you wish to set different URLs for Android, iOS and desktop.
Whichever destination you choose, you can be in full control of your users’ redirection path, thus creating a seamless and consistent user experience. This ultimately adds to your branding efforts.
### 4\. Customize Open Graph meta tags for social media branding
Open Graph meta tags are snippets of code that decide how links are displayed when shared on social media. With this metadata, you can provide details including title, description, and preview image of the page.
[__wf_reserved_inherit]
In the Social Media Thumbnail settings page, you can preview and modify how each Airbridge tracking link would appear when shared on social media. Use the Open Graph Editor feature to quickly modify the meta tag, without involving a developer.
Open Graph meta tags are vital because they make it easier for users to learn more about your brand and to check the details of the linked content, therefore increasing click-through rate (CTR). The key content details included in the meta tags also help with SEO and brand trust.
### 5\. Measure and improve marketing performance
After launching the campaign using branded links, you want to measure and analyze the results.
The good news is that Airbridge’s branded link is a **tracking link** that allows you to monitor **behavior of the users who clicked on the link** and the **effectiveness of your campaign**. Airbridge collects click and impression data as well as other relevant data to process them and prepare various types of reports. You can always track data via [Actuals Report](https://help.airbridge.io/hc/en-us/articles/900001042343-Actuals) and [Trend Report](https://help.airbridge.io/hc/en-us/articles/900003516746-Trend).
## The next step to success
As you have seen, branding your links has many advantages. Beyond the obvious aesthetic appeal, you can build a consistent brand identity, win users’ trust, and improve SEO rankings.
If that’s not enough, it’s time for you to check out Airbridge’s all-in-one branded link - a branded link but a tracking link and a deep link as well. You can add branded elements to your links, measure your performance and redirect users to appropriate web or app pages.
Got any questions? [Request a demo](https://www.airbridge.io/en/book-a-demo) to have an in-depth conversation with our team of experts and find your way to success.
---
## Deep Linking Basics: Everything Marketers and Developers Need to Know
URL: https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers
Language: English
Published: 2022-08-17
Let’s say you’re looking for a new smartphone. To check out your options, you would go online, type in some keywords in the search box, scroll up and down, scratch your head a bit, and click on a link that seems most relevant.
If you’re familiar with navigating the internet as above, you would already know that a typical link looks like something below.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
You would also be well aware that clicking a link takes you to a specific web page. Deep links are used to do the same thing in mobile environments.
## What is deep linking?
Mobile users are directly sent to in-app destinations when they click on **deep links** on a web page or another app. This allows the users to access in-app content simply by clicking on an ad. Hence, deep linking is an essential part of mobile app marketing that significantly increases conversion and retention rates, optimizing user experience.
However, with deep linking technology ever-evolving, occasional crashes are inevitable. How can we do our best to avoid such hassle and make the most out of it? To begin with, let’s take a look at **what deep links do** and **what URI schemes, Universal Links and App Links are**.
## Why do you need deep linking?
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
If a user who already has the app installed clicks on a deep link but is routed to an app store, he or she would leave immediately. The case would be the same if a user clicks on an advertisement of a product only to be sent to the app’s home page. In order to increase conversion and retention rates, thus retaining your hard-won users, you need to send them to the right destination. Deep linking creates a seamless user journey by making web-to-app navigation easier.
## Three types of deep links
There are three deep linking technology standards: **URI schemes, Universal Links, and App Links**. They all bring users to specific in-app content but differ in settings. URI schemes were the first generation standard, but had its limitations which led iOS and Android to develop [Universal Links](https://developer.apple.com/ios/universal-links/) and [App Links](https://developer.android.com/training/app-links), respectively.
Given the various types of deep links today, not only app developers but also marketers willing to implement deep links are often confused by the concept. Hence, we’ll begin by discussing the types of deep links and how each of them work.
### 1) URI schemes
URI schemes are the primary method for deep linking. This way, every in-app content is assigned a unique address which can be tapped by users to run apps and explore resources. Similar to URLs on the internet, URIs direct users to a specific page. The difference is that URLs mostly begin with http:// or https:// whereas URIs begin with a “custom scheme” registered by each mobile app.
- [URI](https://developer.mozilla.org/en-US/docs/Glossary/URI) (Uniform Resource Identifier) is a unique string of characters that identifies a resource. URN and [URL](https://developer.mozilla.org/en-US/docs/Glossary/URL) are subsets of URIs.
- URL (Uniform Resource Locator) is also a text string used to specify the location of a resource such as a web page, an image, or a video on the internet.
The following image shows the standard structure of a URI.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
In the example above, **airbridge://** is the scheme or the protocol. A mobile app can register its own custom URI scheme that distinguishes itself from others. That is to say, each app developer decides which scheme to use.
The rest consists of **page/ko**, the path to the resource, **?product=1**, the parameter or the query string and **#content**, the anchor or the bookmark representing a specific part of the resource. ([Learn more about URL syntax on MDN](https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Identifying_resources_on_the_Web))
- A scheme is registered by the developer to distinguish the app from other apps.
- A path specifies a specific page in an app.
### How Universal Links and App Links overcome the limitations of URI schemes
URI schemes were widely used because of their easy implementation. However, problems emerged as it was impossible to prove ownership of or reserve a particular scheme. With the number of apps increasing, multiple apps happened to register the same scheme, resulting in URI conflicts. In addition, any URI scheme defined by an app could be hijacked by another app. If two apps share the same URI-based deep link, an Android user can choose which app to open while an iOS user is automatically directed to the most recently installed app.
As an app’s scheme is not publicly open, there is no way for developers to make sure their scheme of choice is unique. To deal with this problem, Apple and Google have each developed deep linking protocols: Universal Links and App Links.
### 2) Universal Links and App Links
Universal Links and App Links are deep linking protocols used in iOS and Android, respectively. They overcome the limitations of URI schemes by registering the app’s domain names on the OS to help them prove their ownership.
A **domain name** is a distinctive address used to access websites. In **http://www.airbridge.io/ko**, **airbridge.io** is the domain name. Each and every domain name is unique, and can only be managed by the owner of the domain, assuring its ownership.
During the app development process, the developer registers its domain and verifies it by adding a TXT record to the DNS.. When the app is installed, the OS makes sure that the TXT record matches the one added during the app development process. Once the verification is complete, it stays valid unless the app is deleted.
When a user clicks on a deep link that is created by a verified domain,
- If the app is installed, a specific page within the app opens.
- If the app isn’t installed, deep linking fails and the user is sent to the domain’s web page. Various options for fallbacks are available once that happens; the user can be routed to a mobile web page or the app store.
Below is an example of a Universal Link/App Link.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
## How to set up deep links in Android and iOS
Now, let’s take a look at how to set up URI schemes, App Links (Android) and Universal Links (iOS).
### How to enable deep links in Android
In your Androidmanifest.xml file, set up deep links for your in-app activities. Rather than assigning a different URI for each and every single activity, you can add multiple data tags to define the URI that the activity accepts.
The android manifest file contains the following information about your app:
- Attributes: package name, versionCode, and versionName
- Permissions
- Components: activities, services, broadcast receivers, and content providers.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
As shown above, when using URI schemes or App Links, you can add an intent filter to your manifest file to specify the scheme, host, path and query parameters.
- [Airbridge Developer Guide: How to set up a deep link in Android?](https://developers.airbridge.io/v1.1-en/docs/android-sdk#deep-link-setup)
### How to enable deep links in iOS
Deep links in iOS are configured to open a certain page, which in turn redirects to the final destination.
To work with URI schemes on iOS, register your scheme in Xcode from the Info tab of your project settings. Declare “URL Schemes” in the URL Type section, as shown in the following illustration.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
**To set up a Universal Link**, open your Xcode project and enable the Associated Domains section in the Capabilities tab. Add an entry for each domain, as shown below.
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
- [Airbridge Developer Guide: How to set up a deep link in iOS?](https://developers.airbridge.io/v1.1-en/docs/ios-sdk#deep-link-setup)
## What to be aware of when deep linking for app marketing
Even though Universal Links and App Links provide a solution to the ownership issue of URI schemes, they are far from an app marketer’s ideal. **You need to be strategic about which deep link standards to use depending on the browser.**
Universal Links and App Links work fine in the browser of its provider: Apple’s Safari and Google’s Chrome. However, Apple no longer officially supports URI schemes for deep linking whereas other browsers either exclusively support URI schemes or support Universal Links and App Links but only to a limited extent. Such discrepancy occurs as each in-app and mobile web browser has its API developed differently.
The table below summarizes how the three deep link mechanisms - Universal Links, App Links (https://www.airbridge.io) and URI schemes (airbridge://) - work in major in-app and mobile web browsers. **If the user has landed on the web, this means that the deep link has not worked properly.**
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
For Universal Links, you should also note that:
- Universal Links don’t work if the user directly pastes the link to the address bar.
- Universal Links aren’t triggered by JavaScript.
- Universal Links can’t be opened if your app uses openURL.
Since each in-app and mobile web browser supports different types of deep links under different conditions, running tests and changing settings accordingly are inevitable parts of the deep linking process. You may have to use URI schemes - even if you don’t want to - depending on how the browser is set up.
Moreover, URI schemes have no built-in fallback option, meaning that if the user doesn’t have the app installed, he or she will simply see an error page when clicking the link.
👉 Find out more about fallback redirect [here](https://www.airbridge.io/en/blog/deeplink-101-ios-safari-alert)
While user acquisition is an essential part of app marketing, users are likely to leave if they are routed to a web page or if deep linking fails.
## How to make the best use of deep links in app marketing
Given such concerns, how could app marketers benefit from deep links and drive growth?
Start by setting up deep links based on your understanding of the three types of deep links. As it’s the app’s job to direct the user to specific in-app content, deep links have to be set up within the app in advance.
The next step is to use tracking links supported by MMPs or attribution solutions like Airbridge. When creating a **tracking link** for your app to capture touchpoints, configure it as a deep link. This allows you to:
1. Provide a seamless user journey without having to run tests or change settings for every environment a user interacts with a deep link.
2. Set up different fallback options for Android, iOS and desktop on the MMP’s dashboard. With tracking links, URI schemes not having a fallback option is no longer a problem
[Deep Linking Basics: Everything Marketers and Developers Need to Know]
_With Airbridge, you can specify the fallback path when creating a tracking link._
3. Take advantage of **deferred deep links**, which take a user to the intended in-app content after first routing them to the app store to install the app. To create deferred deep links, you need to use an MMP.
Note that the app is responsible for directing the user to specific in-app content, and thus you first need to set up the links in the app. MMPs like Airbridge only follow the app’s deep linking settings.
Interested in perfecting your user journey with deep linking? [Contact us today.](https://www.airbridge.io/en/book-a-demo)
---
## How to optimize Direct Response Marketing with Touchpoints Overlap Report
URL: https://www.airbridge.io/en/blog/direct-response-marketing-with-touchpoints-overlap-report
Language: English
Published: 2022-06-16
Have you ever wanted new users to download your app or make purchases directly from the ad? Do you have experience running or have plans to run Direct Response Marketing?
If so, you would most likely want your ad to reach as many target audiences as possible and drive the highest conversion. What you would not want is for your ad to reach the same audience in different channels because you would not only be wasting your advertising spend but also opportunity costs. Knowing how much the audience in multiple channels overlaps is significant.
Airbridge provides Touchpoints Overlap Report which shows the percentage of overlapping advertisement views among the same audience. Continue reading this post to see how you can optimize the performances of Direct Response Marketing campaigns and reduce inefficient ad spendings by using Airbridge’s Touchpoints Overlap Report.
## What is Direct Response Marketing(DR Marketing)?
Direct Response Marketing (DR Marketing) is a type of marketing strategy that invokes immediate responses or conversions such as app downloads or in-app purchases from users. The objective of DR Marketing is not to raise or enhance brand awareness but to generate instant outcomes and highest ROI.
DR Marketing is often used in hyper-casual or casual games and e-commerce platforms that sell low-involvement products. Due to the nature of the game which is to centralize all focus on user acquisition in the early stage of launching, the ultimate goal of the campaign is to obtain as many users as possible by delivering ads to a wide range of target audiences in various channels. E-commerce platforms also would want to achieve their revenue goals quickly, and thus will try to make users react more immediately to ads by using marketing campaigns like time-sale.
DR Marketing is a very time-sensitive marketing strategy. And in order to achieve maximum effectiveness, knowing how much of your channel audiences overlap is critical.
## Avoid overlapping target audiences in Direct Response Marketing.
We cannot stop emphasizing the importance of avoiding any audience overlap in DR Marketing. In other words, you must minimize the possibility of users being exposed to the same ad across multiple channels. This will also affect your marketing budget as the opportunity to meet new audiences will decrease if the user is exposed to the same ad repeatedly. As ad frequency equals your opportunity cost, we recommend that you adjust the number of overlapping touchpoints to less than three.
## Discover overlapping touchpoints with Touchpoints Overlap Report
How do you know which touchpoints overlap?
Airbridge analyzes touchpoints such as app install and purchase which exist prior to the conversions, and shows the overlapping touchpoints across channels in the Touchpoints Overlap Report. In this report, you may also find if a converted user (from a specific channel) had clicked or viewed the same ads in other channels prior to the conversion.
[]
Example of Touchpoints Overlap Report
It is easy to gain insight and adjust your media mix by using the Touchpoints Overlap Report. Discover how touchpoints from each channel relate to that of other channels and their overlapping rate.
[]
Example of Touchpoints Overlap Report / Data used in the above report is demo data and does not reflect any actual performance.
## What do overlapping touchpoints tell you?
When you look at the above report, you will see that some of the channels’ conversions overlap with that of others by 70% to 80%. What this means is that the ads are repeatedly exposed to the same audience and you must reconsider whether this specific channel is the right channel for you to run ads on. Again, ad frequency is your opportunity cost in DR Marketing. You may think that overlapping channels may lead users to conversions but you must know that this is a rare case unless the channel is a search engine channel.
## Leverage Airbridge Touchpoints Overlap Report for more effective marketing budget allocation
[]
If you are running time-sensitive Direct Response Marketing campaigns, keep your target audience from overlapping with other channels. Find out how Airbridge’s Touchpoints Overlap Reportquickly identifies overlapping audiences and brings your marketing performances to the next level!
---
## Five convenient features for mobile app measurement
URL: https://www.airbridge.io/en/blog/airbridge-5-features-to-watch
Language: English
Published: 2022-06-07
[Airbridge](https://www.airbridge.io) is a unified web & mobile attribution solution that supports every stage of a marketing campaign: from planning, execution, to performance monitoring and optimization.
[]
> Airbridge makes a 360 degree marketing performance analysis thanks to its capability to simultaneously track all user journeys across various platforms, devices, and channels while also measuring the true ad performance through incrementality analysis which overcomes the weaknesses of the last-touch attribution model.
In this post, we would like to introduce five features that can dramatically improve the convenience of Airbridge users.
## 1. View and share real-time reports without dashboard access
Need to share performance data in real-time with your external partners or members without accessing your Airbridge dashboard? Airbridge allows you to share real-time statistical data from the Actuals Report as a CSV file or via Google Sheet without having to access the dashboard.
[Actuals Report in Airbridge]
**Actuals Report** is a performance report that allows you to freely customize and monitor all the data you are tracking. A variety of data fields are provided by Airbridge to set metrics and specific criteria (e.g. group-by, filters) to segment or filter data, etc.
Once you have customized and saved your report, find the “Share” button and switch on the \[Enable Sharing\] toggle in your saved reports, and you’ll be provided with a URL to download the CSV file. **This CSV file is updated in real-time and provides real-time data whenever the link is loaded.**
### Downloading a real-time report
1. Click the \[Saved Reports\] button at the top corner to go to your saved reports.
2. Click the “Share” button located at the top-right side of the stat table of your saved report.
3. Turn “ON” the \[Enable Sharing\] toggle, and copy the URL.
4. Paste the copied URL into the address bar of your browser and download the CSV file to check the real-time statistical data of your shared report.
For security purposes, every time you turn ON/OFF the \[Enable Sharing\] toggle, the URL for your shared report will be modified, and the previous URL will expire.
### Using Google Sheets for a real-time report
You can extract data from URLs in the form of .csv (Comma Separated Values) or .tsv (Tab Separated Values) by using the importdata function provided by Google Sheets.
**\=importdata (“the URL copied from the \[Enable Sharing\] feature of your Airbridge Dashboard”)**
The importdata function regularly loads the URL (approx. every 15min-30min), so you may use it to check the statistical data on Google Sheets near real-time.
You can use Airbridge’s data loaded via Google Sheet to merge with other data from your company or to integrate several Airbridge reports to create a single holistic report.
## 2\. Copy all your settings with a single click. \[Copy Config\]
Did you ever wonder while looking through your Airbridge reports, why your data is not displayed as you would expect or like it to be? In such cases you’ll probably reach out to AB180 Customer Success Manager for help, however, in order for you and the AB180 Customer Success Manager to view the same data, all of the report, metrics, group-bys, and filters settings must be set identically.
With Airbridge, sharing these settings is incredibly simple – **click the \[Copy Config\] and share the copied config file.**
You can also apply the copied details by using the \[Paste Config\] located in the top right corner of your \[Configuration\] page, or you can simply press Ctrl+V while your report page is open.
[]
## 3\. Ctrl+c & Ctrl+v? It’s simpler – copy your charts and tables with a click of a button.
Ctrl+c and Ctrl+v is probably everyone’s textbook way to copy and share things from your desktop or laptop. You might even use a screenshot from time to time. **With Airbridge, however, you can copy your report with just a click of a button.**
1\. **Copy charts**
For Airbridge reports with charts, click the \[Copy Chart\] button located above the right corner of your charts to copy with ease. You can then paste the copied chart as an image.
- Reports with charts: Trend Report, Active Users Report, Touchpoints Analysis Report, Touchpoints Overlap Report, etc.
2. **Copy tables**
Tables in all Airbridge reports can be copied by just pressing the \[Copy Table\] button located above the right corner of your tables. You can then simply paste the copied information from your tables into other tools such as excel or spreadsheets etc. The great thing is – the format of the tables is kept intact doing so.
## 4\. Mass Create tracking links
Airbridge’s tracking link is an all-in-one link. It is a deep link that helps redirect users whilst simultaneously being a link that measures and attributes performance. After creating a link from the dashboard which includes the campaign parameter (UTM parameter), it can take your users who click it on various platforms, channels, and situations to desired destinations.
**Airbridge also allows you to mass create these tracking links - all at once.**
[]
If you find it burdensome to create one tracking link after another or need to create thousands of links each with separate keywords from your dashboard, you can simply use the Tracking Link Generator to mass create tracking links. Tracking Link Generator is a Google Sheet application created by using the Tracking Link creation API via Google AppScript, and is available to all Airbridge users.
1\. Select and copy the template of the channel you operate.
- [Tracking Link Generator Template (for Integrated Channels)](https://docs.google.com/spreadsheets/d/1I3oHv7gmjTTnbo-Ewcf2wTajUwC5qN0PlVGgk2Ktdig/edit#gid=0)
- [Tracking Link Generator Template (for Custom Channels)](https://docs.google.com/spreadsheets/d/1QTryw9E-lDMMXO89zY4ysQdrINhgV6kMhQFMSz8NQNU/edit#gid=0)
Fill in the app\_name and tracking\_link\_api\_token in the “Configs” tab in the copied sheet. You can find the app\_name and api\_token via the following path: Airbridge Dashboard > Settings > Tokens menu, where you can find the app name (Unique ID) and Tracking Link API token.
2\. Install [Airbridge Tracking Link Generator Add On](https://workspace.google.com/marketplace/app/airbridge_tracking_link_generator/9223133159)
3\. Manage your settings for channels, campaign parameters, user path, etc.
4\. Once you are done with your settings, create your link via the following path: Add-Ons > Airbridge Tracking Link Generator > Generate Tracking Links
## 5\. Still downloading your entire raw data? Pick and select the data you want to export with Airbridge!
If you need the raw data, you can use our raw data export feature **to customize and download the data you need.**
[]
You can select the data you need from the \[Raw Data > Data Export\] menu and export them as a CSV file. Raw data is provided in two distinct forms: App Raw Data and Web Raw Data.
**Click the \[Export Raw Data\] button on the top left corner of the App Raw Data or the Web Raw Data page and select configs to be exported, then click \[Request export\].**
You can include or exclude the desired data in each of the \[Select Event\], \[Select Property\], and \[Select Filter\] stages. The following \[Data Preview\] stage then allows you to check all your settings already chosen in the \[Select Event\], \[Select Property\], and \[Select Filter\] stages while also enabling you to preview the raw data from the last 5 days with the applied settings. **Note that this preview only shows data from the last 5 days with the applied settings, and thus may differ from the data exported based on the export period set in the \[Select Date\] stage.**
In the \[Select Period\] stage, you can choose the desired period for your data, after which you can click the \[Request Export\] button to finish your request for Raw Data. \[Request Export\] may take some time to get ready and a notification will be sent to the email of the requester’s account once it is ready for export.
You can check the details and history of requested Raw Data in the “Request History” tab and can download any Raw Data that is ready for download. From this tab, you can also click the \[Export\] button to request raw data with the same template settings as before, or you can easily request similar raw data by modifying some criteria - such as adding an event or changing the date, etc. \[Make Template\] button allows you to save previous request details which can be loaded later to export raw data.
## Airbridge, a unified web & mobile attribution solution.
[Airbridge](https://www.airbridge.io/) is the unified web & mobile attribution solution that allows you to accurately measure and attribute user acquisitions across the web and mobile to the exact campaign, creative, and advertising channel.
Reach out to our attribution experts at Airbridge to find out more!
---
## 7 questions you should ask before choosing a mobile app attribution tool
URL: https://www.airbridge.io/en/blog/attribution-tool-checklist
Language: English
Published: 2022-05-25
Given the wide availability of attribution tools today, choosing the right attribution tool to measure your app marketing campaigns is critical. Make the wrong choice, and you'd face what doesn't appropriately address your measurement needs, the costs of an inappropriate pricing plan, and the technical hassles of converting from one tool to another.
Then what makes a good mobile app attribution tool? While the answer varies depending on your product, objective, and growth stage, here is a checklist of 7 questions to ask beforehand.
## 1\. How flexible is it in exporting raw data?
An attribution tool provides performance visibility into the ads that contributed to conversions such as installs or in-app events by matching touchpoint data and conversion data. It could also function as a data pipeline as it collects, processes, and streamlines a wide range of data from Self-Attributing Networks (SAN), tracking links, and SDKs built inside your app.
With an attribution tool, you could see the performance data on the dashboard or export the raw data for further usage customized to your needs. However, some attribution tools may limit the data types available for export or charge a premium for downloading them.
> With Airbridge, over 220 different types of raw data collected from web and app SDKs, tracking links, and SANs are available for export in real-time with no additional charges.
## 2\. How well-integrated is it with third-party tools?
There is more to an attribution tool than just measurement and performance analytics.
You could integrate the attribution tool with third-party tools or cloud storages for further usage as your business grows. Checking integration readiness and availability for expansion of the attribution tool is a must. We recommend looking up the attribution tool's user guide, developer guide, and API guide.
> Airbridge does not charge or require extra development effort for data transfer to other platforms. You can easily integrate data in Airbridge with cloud storage like AWS S3 or other SaaS tools like Amplitude and Braze with a simple switch of a toggle button or by programmatic integration via query API.
>
> If you are using Amplitude, you should check whether your attribution tool can send both attribution and events data to Amplitude. Even if you integrate your attribution tool with Amplitude via SDK, there may be some data discrepancies due to time lag or API spec differences. With the new HTTP API V2 integration, all data will be properly recorded on Amplitude, and all potential discrepancies will be eliminated.
## 3\. How customizable is the attribution window?
The most commonly used attribution model is the last-touch model. Most advertisers use this model because of its simplicity in attribution analysis. It is also widely used in discussing advertisement costs between advertisers and ad channels.
Attribution tools offer varying customizations for the attribution windows. Here are some questions to consider:
1) Is it possible to set separate attribution windows for installs and in-app events?
2) Is it possible to set different attribution windows for different matching methods - **the Device ID matching method** and **the Probabilistic matching method**? (Below two methods are used in Airbridge)
- Device ID matching method uses Google Play Referrer query string parameter, deeplink, ADID matching, and platform matching.
- Probabilistic matching is a method that estimates the probability that a user who viewed and clicked an ad and a user who first opened the app is the same individual, using a combination of non-unique values, such as device information and IP.
3) Is it possible to set different attribution windows for each ad channel?
In Airbridge:
You can configure attribution window settings globally, but you can also set different attribution windows for each channel. You can set two different attribution windows for before and after the installation. **The Lookback Window** tracks touchpoints (click or impression) that exist prior to the install and detects which touchpoints contributed to an install, while **the Attribution Window** tracks user interactions (in-app events) after the installation. You can also set different attribution windows according to the matching methods (Device ID matching and Probabilistic matching).
Some attribution tools only analyze installs and no further events. Under this condition, the same channel that was accredited for install will get the final credit for all conversion events after the installation, even if other channels have contributed to those conversions.
## 4\. How customizable is the dashboard for performance tracking?
Running ads and measuring performance for optimization are integral to every marketer. A dashboard that holistically reports all aspects of campaigns can significantly improve work efficiency as the relevant data differs according to each campaign.
If you are a marketer running campaigns on both web and app, check **if an attribution tool supports a unified dashboard across multiple platforms (web, Android, or iOS)** as it would be a nightmare to check each platform's performance on respective reports or even on separate tools. As customers are present across diverse platforms and channels, a holistic measurement is crucial.
Airbridge supports app and web SDKs that collect **conversion data from every platform**, even iOS SKAdNetwork. Airbridge then aggregates all data across devices, platforms, and channels, and its dashboard is easily customizable for your reporting needs.
> d.code, an industry-leading pre-order commerce service, is an avid user of Airbridge's customizable dashboard. Find out how d.code greatly simplified their performance reporting practices with Airbridge on the [case study](https://www.airbridge.io/en/case-studies/dcode-web-to-app-attribution).
## 5\. Does it offer different access permissions by user type?
There are cases where brands plan and execute end-to-end marketing campaigns themselves, but brands usually cooperate with advertising agencies and media partners for planning and execution. In such cases, **different levels of access authorizations must be granted to each type of user** to limit access to any sensitive data and further prevent security issues.
Every attribution tool has its own method of defining user access to various data types. In Airbridge, four different types of users are defined: Owner, In-house Marketer, Agency, and Media Partner.
- **Owner** is the first user that registers the app in Airbridge. The Owner has full access to all features, including data export and changing the settings for new users. In addition, only the owner can delete the app in Airbridge.
- **In-house Marketer** has equal authority as the owner thus can add new users except, they cannot delete user accounts or the app in Airbridge.
- **Agency** can only view agency-generated data. The Agency can view or export data of permitted ad channels with permission.
- **Media Partner** can view channel-attributed data in Actuals report and view the list of tracking links, postback settings, and attribution window which the advertiser has set for the channel.
## 6\. How far does it go for consultation and customer support?
While features are important, having the proper knowledge to maximize the use of an attribution tool is equally important.
Long waiting hours and lack of dedicated support or consultation are common complaints about global SaaS tools, but this does not apply to Airbridge. **Airbridge has a dedicated Customer Success team** equipped with an organized system and numerous digital marketing experts. Airbridge’s Customer Success team offers product trainings, guided walkthroughs, support for event taxonomy, and SDK installation from start to end. In addition, the Airbridge Customer Success team provides tailored consultation to derive deeper insights, regular business check-ins to share new features and improvement measures to maximize product usage, and knowledge sharing sessions to share industry trends, the latest use-cases, and best practices.
## 7\. How is the pricing plan structured?
While pricing plans vary across attribution tools, they are roughly divided into two types: cost per install and pricing based on DAU/MAU. A good practice is to get a quote for each tool in advance. For DAU/MAU-based pricing, you can easily predict the expense and not worry about overpaying for potential users who drop out immediately after the app installation. For cost per install pricing, however, costs may increase as you execute more marketing campaigns. For example, if a paid install campaign leads to a surge in installs, you may face a rapid increase in costs.
Additionally, many attribution tools require additional payments for access to specific features. Therefore, make sure the package includes all your necessary features when receiving a quote. In most cases, you’ll be able to find a detailed quote of pricing by connecting to the sales team.
## In Summary
Finding the right mobile app attribution tool that fits your business is not simple so consider a wide range of aspects before making a decision. As switching to a new attribution tool is more complicated than the first-time implementation, we encourage you to ask the 7 questions above before deciding on an attribution tool. By asking the right questions while comparing each tool, you’ll be able to find the most compatible tool for your product and marketing team.
---
## Introducing Luft - Airbridge’s in-house OLAP Database
URL: https://www.airbridge.io/en/blog/introducing-luft
Language: English
Published: 2022-05-19
As a mobile attribution solution, Airbridge processes over 10 billion records of data every month to provide analytics for campaign performance, user behavior, and growth metrics across the web and app. Analyzing user behavior in real-time is crucial for campaign optimizations.
Airbridge developed an OLAP database that focuses on user behavior to provide real-time analytics. Let’s take a closer look at how Luft works to offer cohort analysis and real-time reports.
## How Cohort Analysis Works
Cohort analysis aims to analyze the behavior of a group of users. For example, calculating the retention rate of ‘female users in their 30s who spent more than $100 in the last six months’ is an example of cohort analysis.
[__wf_reserved_inherit]
It is a two-step process: first, define a specific user group, then load the event data of users who belong to the group and analyze their behaviors. However, the time it takes to process the data for these steps is an issue. Results could be sent via email once the analytics are done, but it would make it virtually impossible for data analysts to test the results in real-time with a UX environment. Hence, the need for performance that can process instant analytics on the Airbridge dashboard.
## Limitations of Conventional Architectures
Airbridge has been using Apache Druid to analyze data. Druid is an OLAP data store that can process analytics queries within a few seconds, through pre-aggregating set metrics periodically and roll-up methods that select then calculate specific values upon query.
Most OLAP databases, including Druid, pre-aggregates set metrics that were precalculated. Since it reads readily available data rather than raw data, queries can be quickly processed. However,
- Druid cannot query metrics that are not pre-computed. So in order to perform a different type of analysis, data has to be newly indexed.
- Advanced behavioral analysis, such as funnel analysis, cannot be done without raw data.
While Druid was an optimal solution for various Airbridge reports in the past, limitations became apparent when real-time cohort analysis was being developed. The fact that the necessary queries can't be immediately processed and the low performance (minimum 15 seconds when testing queries) was an issue.
## Shuffling keys with high-cardinality deteriorates performance
Queries analyzing user behavior involve grouping event data by user ID. Generally, [MapReduce](https://en.wikipedia.org/wiki/MapReduce) frameworks go through a [shuffling](https://data-flair.training/blogs/shuffling-and-sorting-in-hadoop/) phase when running Groupby queries. Once data is sorted by key and partitioned, data is sent through the network.
The shuffling phase is a bottleneck because user ID is a key that has very high cardinality. For example, if a client with more than millions of daily users opens a report, millions of user IDs have to be sent through the network - significantly deteriorating performance for most solutions.
## Why we had to develop Luft ourselves
After testing existing solutions, we came to the conclusion that we would have to develop a new in-house solution tailored to our needs.
- Cost Efficiency: Data warehouses such as Snowflake or Redshift could cover our analytics and performance needs, if scaled to large proportions.However, commercial data warehouses are all-purpose databases that are not optimized for OLAP, and we would have to operate clusters much bigger than what we wanted. We believed that there was room for a lot of optimization considering the characteristics of OLAP queries and data for user behavior analysis, and decided to reduce cluster costs by developing an in-house solution.
- Varying Queries: There were limitations when using an SQL database for not only trend reports and retention reports that we were already providing, but also for other varying features such as funnel analysis or identity matching. Data processing engines (e.g. Spark) were used for certain analyses, but these solutions also had performance and cost efficiency issues. The need for a custom in-house solution was apparent.
## Luft: Datastore designed for user behavior analysis
To solve the issues outlined above, Luft was designed as a data store that could quickly process user behavior queries grouped by user ID. **Airbridge is currently performing cohort analysis on tens of terabytes of user data using less than five nodes and within 3-10 seconds on average.**
Unlike most RDBMSs, Luft’s data are immutable. OLAP datastores are mostly used to analyze event data of a user’s behavior. Since the event data does not change once stored, many datastores do not let you edit data that are already ingested. Instead, they can be overwritten by importing new data from the same time period.
Thanks to these **immutable properties**, we were able to overcome many challenges that were met by DBMSs. Instead of using complex cluster structures for various data consistency issues, we could adopt simple clusters like Druid. Also, we could [rely on pagecache for data caching like Kafka does](https://docs.confluent.io/platform/current/kafka/design.html) and achieve higher performance without the need of a page manager.
However, the most significant advantage is that we could design the data storage formats necessary for Airbridge.
## Storage Engine
RDBMS uses a binary tree or skip-list data structure because its data should be modifiable. However, as Luft’s data are immutable, we were able to optimize data structures for user behavior analysis to greatly improve its performance. We chose to use Adroll’s TrailDB as Luft’s storage engine.
TrailDB is designed to store user events with a rowstore. User data is stored after it has been grouped by users and sorted by time, and TrailDB provides various encoding methods to support these data properties.
- _Delta Encoding_: As events are sorted by event timestamps, event timestamp information is stored as time differences between data.__
- _Dictionary Encoding_: Rather than storing complete data values, it catalogs data into a dictionary and stores only their IDs.__
- _Edge Encoding_: As user information like age, e-mail, and device type rarely changes for behavioral data, only changed columns are stored to prevent unnecessary storage actions.
[__wf_reserved_inherit]
Such data structure enabled us to compress our user data by a surprising **97%** while maintaining decoding performance with the same time complexity. In practice, 13GB of sample data exported from Airbridge in CSV format was compressed to 300MB after it was encoded by TrailDB. Smaller data means more data can be cached in RAM.
**Performance deterioration issues that occur during the shuffling phase were also resolved with the data being already partitioned by the user at the storing phase.** It is much easier to shuffle pre-grouped users than shuffling each event separately. Moreover, since the offset of each data is recorded, it is possible to read the necessary user data only. This means that the amount of data that has to be read in order to perform cohort analysis decreases greatly.
## You only scan once
A query that analyzes user behavior usually consists of multiple subqueries. In order to save scan time, Luft was designed to scan all the necessary data in one go.
Airbridge’s user events are temporal data, and every query contains a time period. Therefore depending on the event timestamp, Luft partitions data into temporal blocks. If the time period of subqueries overlap with each other, a redundant scan for the same time period will happen.
[Query optimization example: A query that analyzes the retention between September 1st and September 3rd for certain users between September 2nd noon and September 3rd consists of several subqueries. These subqueries are split into timeframes, and then similar subqueries are merged together to make one query.]
A query is optimized by Luft during the query planning phase so that only a single scan is needed. Various subqueries are optimized by being split along the time axis and merged if the queries are similar. Since the results from the query are merged, queries can be quickly processed through a single scan without redundancy.
## Architecture for Consistency
We also adopted **Lambda Architecture** for real-time queries. Lambda Architecture is a data-processing architecture that uses both batch and stream-processing methods. The data pipeline is separated into **batch layers** and **real-time speed layers**. Data is then processed after the pipeline is merged at the actual serving layer. Lambda architecture assures **eventual consistency** as the integrity and consistency of real-time data is secured by overwriting batch data.
[Luft’s cluster architecture: Master node queries data to historical nodes(batch data) and real-time nodes (Kafka data stream).]
Luft adopted Druid’s structure that utilizes Lambda Architecture, and defined the nodes’ roles as below:
- **Real-time nodes:** Data shards that reads and stores real-time data from data streams such as Kafka
- **Historical nodes:** Data shards that store batch data. It caches batch data and executes queries.
Even with the architecture above, each node has a possibility of data loss or data between nodes might not be consistent. To prevent this, data is stored in S3 by default, and only necessary data is stored in the nodes’ disk or memory.
If the data for a query is not stored in a node, it may take some time to download it from S3. We protect against cache miss by using methods such as storing recent data that are more likely to be queried in the nodes and use [Linux’s user-space page fault handler to download data from S3 quickly](https://tech.nextroll.com/blog/data/2016/11/29/traildb-mmap-s3.html).
## More Insights From Existing Solutions
We gained many insights from existing advanced solutions.
- We referred to Spark and Kafka to develop a MapReduce layer using Golang and gRPC. We were able to execute online queries with outstanding performance using partition scheduling methods and a pull-based stream structure.
- We designed clusters using [etcd](https://etcd.io/). Zookeeper is widely used, but etcd had lower management costs and restrictions, while also being highly proven through projects like Kubernetes.
- We make the most of cloud storage. We were able to cut cluster costs effectively by storing data in S3, using DynamoDB for managing metadata, and managing Kubernetes workload scaling ourselves.
## Conclusion
Our goal is to perform real-time user behavior analysis on terabytes of data with small-to-medium-size clusters. Luft’s short term goal is to perform funnel analysis on tens of terabytes of user data with less than 40 CPU cores. We are focusing on optimizing performance and improving data structure to reach this goal.
We will continue to upgrade Airbridge’s features using Luft.
---
## [Deep Link 101] Using Deferred Deep Linking to Direct Users Without the App to a Specific In-App Page
URL: https://www.airbridge.io/en/blog/deeplink-101-deferred-deep-linking
Language: English
Published: 2022-04-07
[]
Want to direct users to a relevant in-app page even after they click a link and install your app? If so, you need deferred deep linking.
> For more details about deep links, check out the <[Airbridge Handbook - Deep Links](https://www.airbridge.io/en/library/airbridge-handbook-deeplinks)> guide.
## What Is Deferred Deep Linking?
Deferred deep linking is a technology that extends the capabilities of standard deep links. As the name suggests, it means a "delayed deep link." Since deep linking cannot work for users who don't have the app installed, this technology waits until the user clicks a link, installs the app, and opens it — then executes the deep link. Implementing this on your own is not easy because it requires a server that can store and deliver the deep link, as well as match the information from when the user clicked the link with the information of the user who installed and opened the app.
In this post, learn why deferred deep linking technology is necessary and how to implement it.
## Why Is Deferred Deep Linking Necessary?
A deep link is a technology that, when a user in a mobile environment clicks a link, launches the app and navigates to a specific page. In other words, it's a link with a designated path leading into the app. However, if the app is not installed when a deep link is clicked, the user is either redirected to the app store or encounters an error, and the link/path doesn't work properly. This means that even a user who clicks a deep link won't be directed to a specific in-app page.
Specifically, when the app is not installed, the following issues can occur:
- When using URI Scheme deep links, if the app is not installed, it's impossible to navigate to another destination such as the app store, so nothing happens at all.
- In the case of Universal Links and App Links, it's not possible to use URLs that land directly on the App Store or Google Play, making it difficult to even prompt users without the app to install it.
- To address these situations, using only deep links through an MMP's tracking links still doesn't provide a complete user experience.
Due to these functional limitations, if nothing happens after an ad click, or if after the cumbersome process of installing the app only the main page appears, even potential customers who were attracted by the ad may immediately drop off.
**The technology that solves this problem is deferred deep linking.**
Even for users who don't have the app installed, if they see an ad and install the app, it feels natural to be directed to the relevant content within the app after installation. The likelihood of conversion increases as well. This is exactly why deferred deep linking is essential for app marketing.
## How to Implement Deferred Deep Linking
However, implementing deferred deep linking technology on your own can present several challenges. At the Android OS level, it can be implemented using the Google Play Install Referrer API, but with limitations. Moreover, on iOS, there is no OS-level method for implementing deferred deep linking, so a complex setup such as building a separate server is required.
### The Difficulty of Implementing Deferred Deep Linking In-House
First, **on Android**, you can implement deferred deep linking independently using the Google Play Store's Google Play Install Referrer API.
However, this method is complex to implement, making data collection issues highly likely, and it only works for downloads from Google Play.
**On iOS**, a similar approach is not possible, so building a server is an additional step that becomes necessary.
### Easily Implement Deferred Deep Linking with a Mobile Attribution Solution
Given that deferred deep linking is difficult to implement on your own, how can you do it efficiently and completely? The answer is to use a solution provided by a mobile attribution tool.
Deferred deep linking is a technology that delays the deep link until the user clicks a link, installs the app, and opens it. **A server is needed that can store the user's click information and, when the app is launched, match the click data with the user who opened the app and deliver the deep link.** This is what an attribution tool provides.
The process of implementing deferred deep linking through Airbridge:
- User clicks an Airbridge tracking link -> Click information stored on server -> User installs the app -> On first launch of the app with SDK integrated, the server checks for link information -> Server delivers the deep link address to the SDK -> The specific in-app page is opened
SDK Guides:
- Airbridge [Android SDK](https://developers.airbridge.io/docs/android-sdk) guide
- Airbridge [iOS SDK](https://developers.airbridge.io/docs/ios-sdk) guide
### Airbridge Tracking Links Make Deferred Deep Linking Easy to Apply
With Airbridge, when you set up a tracking link as a deep link, the deferred deep linking feature is enabled by default.
[]
## For Smooth App Marketing, It's Best to Use an MMP's Tracking Links
MMP tracking links let you design the user journey you want without complex per-OS implementation.
Deferred deep linking is an essential technology for app marketing. Implement it efficiently using Airbridge tracking links.
> Check out other articles in the Deep Link 101 series:
👉 [Deep Link 101: Getting Started with Deep Links for Marketers and Developers](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers)
👉 [Deep Link 101: Why Does a Warning Message Appear When Deep Linking in iOS Safari?](https://www.airbridge.io/en/blog/deeplink-101-ios-safari-alert)
---
## The A to Z Guide To Mobile App User Acquisition
URL: https://www.airbridge.io/en/blog/ua-marketing-a-to-z
Language: English
Published: 2022-03-30
If a product doesn’t have any active users, the product will die. Users are a must for the success of your product. In this post, we will learn everything there is to know about user acquisition (UA) marketing. UA is an absolute must compete with many other apps to win the hearts and time of users.
The ultimate goal for the true growth of each product is not merely acquiring users but going over and beyond to acquire users, keeping users around as habitual users, and encouraging them to turn into active users. To achieve this, User Acquisition marketing for apps - the starting point of the user journey - requires strategies for both short-term performance and continued growth.
This post will discuss the following.
1) What you should prepare before launching your product 2) Importance of user acquisition at the initial stage of your product 3) Two strategies of user acquisition - Organic & Paid marketing 4) The importance of deep links 5) Accurately measuring UA campaign performance
## App Marketing should begin even before launching your product
Mobile app marketing is something that should begin even before launching your product. It’s a process that includes branding, understanding your customers, and competitor analysis serving as the basis of your overall app marketing.
### Raising awareness of your brand and products
Raising awareness of your brand and product to as many people as possible is one of the key strategies for a marketer before launching a product. This could fall under branding, but it is an essential part of marketing as well since users who are favorable to your products may be converted after actual product launching to become avid users. Having users recognize and identify with the core values and purposes of your brand will have greater effects on marketing in the future.
This is one of the marketing methodologies considered to be important in Growth Hacking. [Ryan Holiday](https://ryanholiday.net/) introduced several methods to approach your first time user group in his book \[Growth Hacker Marketing\]. He proposed introducing your brand and products to websites frequently visited by potential users and posting blog posts on major issues in relevant fields. Holiday actually applied the methodologies introduced in his book in the process of publishing \[Growth Hacker Marketing\]. He published a brief ebook to test the market before publishing the actual book, overwhelming his competitors in the market. He also could improve the content of his book and receive direction for future marketing by feedback from readers. He wrote relevant articles before publishing his book and submitted them to influential websites. He was able to advertise his book easily through a list of email addresses with his Newsletter list.
### Setting a persona through customer and competitor analysis
Before starting your full-fledged app marketing, you need to go through the process of deciding which user group your product targets and understand your customers and the market to decide on an appropriate persona. Understanding your customer will need the utilization of both quantitative and qualitative data, which will serve as an important key to achieving goals in business.
Consumer surveys could be carried out with a wide spectrum from surveying friends, family, and potential customers to analyzing market demands and interests with accurate figures. Here, expectations for the user for your services and products could be identified to add new features or improve UI/UX for user-oriented product development. This is why you should continue to understand and analyze your customers after launching your product.
Here are some tips for customer and competitor analysis from [14 Ways To Determine Need Before Launching A New Product Or Service](https://www.forbes.com/sites/forbesbusinesscouncil/2021/07/06/14-ways-to-determine-need-before-launching-a-new-product-or-service/?sh=757ef8c113f2) published on Forbes.
- Use [Google Trends](https://trends.google.com/trends/?geo=US): Google Trends will help you find top search queries in Google Search across various regions and languages. This could serve as a helpful indicator to identify what your potential customers are searching for and to plan relevant content and features. But one thing you need to be aware of is that there can be a disparity between an informational search’s volume and actual intent to purchase.
- Analyze competing products and services to seize opportunities**:** It is especially a good idea to thoroughly analyze the No.1 products and services of the same industry. There will be some room for improvement in design or function for even the top products. It would be a good idea to interview or survey users actually using the product.
- Try launching a beta version of the product for free or receive pre-orders**:** Try launching the product for free or receive pre-orders on social media to check the responses of potential customers. Make sure to analyze all potential causes for
- reactions and think of ways to put the best foot forward and make up for the shortcomings.
### Establishing concrete app marketing strategies
After analyzing your customers and competitors, you need to establish specific app marketing strategies based on your analysis. If you’re lost on where to start, a suggestion for you would be to start with ‘app marketing funnels’. With app marketing funnels, a strategy for each funnel is established, and the condition to pass the funnel is established clearly. It is important to understand the funnel for your products and services and establish strategies for the different stages to be connected organically rather than considering each stage separately.
Also, this strategy must have the concrete Key Performance Indicator(KPI). Major KPIs to be considered include DAU, MAU, CPI (Cost per Install), CPA (Cost per Acquisition), CTR (Click-Through Rate), conversion rate, and churn rate. Appropriate metrics for each stage of the customer journey must be tracked to ensure that the process from user introduction to conversion and retention is seamlessly in place.
## Organic and Paid marketing strategies for user acquisition
If you’ve finished preparing for user acquisition before the product launch, now is the time to acquire more users for your app. The success and failure of most mobile apps depend on the number of users introduced at the initial stage of your product. The users introduced at your product’s early stage would be able to lead to loyal customers. In fact, many apps and services carry out massive marketing right after their launch, and this is something you wouldn’t want to miss as a marketer. You must raise awareness of your app to as many users as possible, encourage the installation of your app, and onboard them successfully.
The full-fledged mobile app marketing strategies to raise awareness of your app and acquire users can be categorized into organic marketing and paid marketing.
### Organic marketing Strategy
#### 1\. Building a landing webpage to introduce your app
Even for services focusing on a mobile app, a webpage to land customers and introduce the app is a must.
That is because many users tend to search on the web when they have a need or are experiencing difficulties. They also tend to search on the web before downloading an app when learning about specific products and services. If your webpage has good visibility on search engines such as Google, thanks to SEO, potential customers can easily access the webpage. The website can affect the first impression of your app. If the website cannot persuade users enough to download the app, they may bounce as it is annoying to download the app at the app market, which may be a barrier to acquiring users. Above this, some advertisements including search ads (SA) require redirection to the web that a web page is a must. Since a webpage is an owned media, even no additional advertisement cost occurs.
It is important to show what users would be offered after app installation on this webpage through videos and images. The webpage must include links to app markets and an accurate call to action (CTA) to guide users on their next steps. Email addresses could be collected to send out newsletters and keep customers updated before the app launch to form a foundation for communication with users.
#### 2\. Content marketing
Content marketing is an effective method to capture your potential customers. The keys to content marketing are ‘**target audience**’, ‘**business goals**’ to be achieved, and **‘high-quality content’.**
First of all, you need to plan content marketing by analyzing your users and market trends including who your product’s user base is, which websites and social media your target users use, what their interests are, and why users use your app. It would be great if you could decide on specific goals to be achieved by this plan.
If you are done planning, you must create and release the content. Here, ‘content’ refers to not only short-form posts posted on social media but blog postings, playbooks, customer case studies and videos, newsletters, and emails. Here are some things you may pay attention to for each content marketing channel.
- _Social Media_: In 2020, the average daily time spent on social media globally was 2 hours and 25 minutes. This is an increase by at least 150% compared to 1 hour and 30 minutes in 2012.([Statista statistics](https://www.statista.com/statistics/433871/daily-social-media-usage-worldwide/)) Social media which people spend lots of time is an effective means of naturally approaching customers and raising awareness of your products in addition to advertisements. In addition to simply posting ads, social media serves as a channel to form communities or receive customer feedback through comments.
- _Blog_: Compared to social media focused on images and videos with a relatively short text, blogs mainly offer lengthy and informative content. Content with valuable and high-quality insight relevant to the industry and trends will be shared more and wanted by more people. This is also an effective means of securing users by search engine exposure through SEO. The aforementioned book \[Growth Hacker Marketing\] also emphasizes securing potential users and communicating through blogs.
- _Newsletter and email_: If you’ve posted some great content through your blog, you can let your subscribers know about the new post through newsletters. If you make continuous efforts to let your users know about the latest updates and relevant market trends, active users will be engaged even more deeply and dormant users will be reminded of your products. Newsletters and emails are some of the most important means of engaging with your customers outside of the app.
#### 3\. Media exposure and press releases
If the news of the new product launch is exposed through the press with app launching, many people will naturally get to learn about the app. Carrying out PR activities in line with product launching won’t be easy for small startups or those in their nascent stages, but press releases are one of the most important means of marketing to acquire early stage users.
#### 4\. ASO(App Store Optimization)
You could say that ASO is an app marketing equivalent of SEO and LPO with the same principles as SEO (Search Engine Optimization) and LPO (Landing Page Optimization). According to [the App Store Data of all nations and regions using Apple Search Ads in 2020](https://searchads.apple.com/), 70% of App Store visitors search for which app to use next through searches, and 65% of all downloads happen after a search. This goes to show how much exposure in the app market is important. That is why you need to optimize to ensure that users are able to find your app when they search for your app’s name or specific keywords. You should improve the visibility of your app through the process of continuously updating and testing the app introduction, description, and preview images exposed on the app market, encouraging more downloads.
_👉 Find the more detailed ASO strategy_ [in this post.](https://www.airbridge.io/en/blog/aso-success-guide)
### Paid Ads Strategy
Acquiring new users for your app through paid ads after launching the app is natural and necessary. After ad campaigns through different channels, it is necessary to optimize the advertisement expenditure and the media mix to boost ad performance.
[__wf_reserved_inherit]
The above image shows the proportion of advertisements impacting app downloads researched by Google and Ipsos. Since the impacts of search ads, social media advertisements, in-app banner/graphic advertisements, and in-web banner/graphic advertisements for app downloads are all around 40-50%, the importance of paid advertisement cannot be neglected. Let’s learn about the most notable paid advertisement strategies.
#### 1\. Apple Search Ads and Google Ads
You can advertise on iOS App Store and Android Google Play - the most notable two app markets - to encourage app installation and user engagement. As mentioned before, 65% of downloads occur after organic search on the App Store, it would be more effective to advertise on the App Market in addition to ASO.
One choice is promoting your app on the Search tab or at the top of search results via [Apple Search Ads](https://searchads.apple.com). This channel is the one you wouldn’t want to miss when advertising for app installation on iOS.
Also, [Google Ads](https://support.google.com/google-ads/answer/6247380?visit_id=1-636549956449797142-1627836531&rd=2) offers three types of app campaigns on Google Play including App installs (ACi), App engagement (ACe) and app pre-registration (for Android only). You can plan advertisements for both iOS and Android users since you can advertise your app across Google’s properties including Google search, Google Play, YouTube, AdMob, Discover on Google search, and Google Display Network.
#### 2\. App install ad campaigns on social media
Social media such as [Facebook](https://www.facebook.com/business/goals/promote-app), [Twitter](https://business.twitter.com/en/help/campaign-setup/create-an-app-installs-campaign.html), and [TikTok](https://www.tiktok.com/business/en/how-it-works?tt4b_lang_redirect=1) also offer app install ad campaigns to boost the app installs on their platforms. Here, a tip is to use platforms that are frequently visited by your app’s target customers. Images and videos could be used on social media to effectively introduce and encourage installs of your app. Here, content from social media and owned media in an organic strategy could be used for added effects.
#### 3\. Offerwall and rewarded ads
Rewarded advertisements are literally advertisements that reward your customers, such as receiving in-app coins when they install an app or complete in-app engagements. A case in point would be advertisements through ‘offerwall’, which presents app users with a list of rewards. Easily put, offerwall is a ‘free charging station’. Users can receive rewards in real-time for installing an app, following social media accounts, and purchasing products. With their reward points, customers can pay for online content or products in the app.
You can use the offerwall advertising in various apps to encourage installs of your app and advertise your app. Rewarded ads are especially appropriate when you need to boost your ranking in the app market shortly after launching your service. One thing you need to be aware of is that there are cherry pickers who will churn and delete the app right after receiving their reward.
#### 4\. Non-incentivized Cost Per Install Ads (NCPI)
‘Non-incentivized Cost Per Install Ads(NCPI)’ are advertisements that encourage app installation and specific behavior after app installation without incentivizing the users. NCPIs are in the form of banners and video advertisements and are generally charged based on CPM, CPI, and CPC. Unlike rewarded ads, users acquired naturally through NCPIs are likely to be more active on your app and engage deeply. In other words, your app’s retention rates can be improved. However, since there is no direct nudge to encourage app installation, a long time and greater budget may be spent on user acquisition.
You may want to prepare against ad frauds such as install hijacking and SDK spoofing which may occur on some ad channels. While ad channels will have filtering technologies, filtering by advertisers and fraud detection & protection technologies by MMPs like Airbridge can help effectively defend against ad fraud.
## Connect your users to the apps seamlessly with deep linking
If you’ve succeeded in acquiring users through the various strategies of the UA marketing, you should make sure that your users do not drop off. Here, deep linking can be your life-saver.
Mobile app marketing greatly differentiates itself from digital marketing in that users download apps through the app market. Even when customers have been introduced to an app or service through advertisement and content, they may drop-off as soon as they move on to the app market. Or when they are not redirected to the in-app promotion page after app installation when they’ve decided to download the app by being attracted by specific promotion, this will negatively impact user experience and have a low conversion rate. That’s why you should use ‘**deep linking**’, especially ‘**deferred deep linking**’ to connect users to specific in-app content from advertisements or webpages.
[__wf_reserved_inherit]
Users are able to be landed on the event page of the product detail page by deep linking after installing the app. This will also have a positive impact on the conversion rate since users are provided immediately with what they have been expecting.
Also, Airbridge deep linking helps redirection of users with deferred deep linking, which also serves as a tracking link for attribution, tracking user activity, and measuring advertising performance analysis.
## After UA marketing, it's time to measure the marketing performance accurately.
Marketing performance measurement analysis is the last stage of UA marketing journey, which could also serve as another starting point. After marketing, data and performance must be analyzed in real time and reflected to revise the existing strategy. This involves analyzing the metrics of the previous marketing campaigns and analyzing them with the KPIs of the app marketing strategy. It is recommended to revise the content strategy and adjust the media mix based on the gained insights to improve the marketing performance.
Especially, as apps should be downloaded through the app market, mobile attribution solutions for measuring advertising performance such as [Airbridge](https://www.airbridge.io) are a must, unlike web tracking through referral, cookies, and UTM parameters. Such tools can measure campaign performance across different marketing channels and attribute subsequent conversions so that all the tracked data could be used for campaign optimization. This allows you to focus on real marketing without having to check and analyze ad performances for each ad channel.
## Conclusion
In this post, we’ve learned about marketing strategies to acquire mobile app users. The most effective marketing strategy for each product and market will differ. Nevertheless, user acquisition is one of the most important parts of the product’s growth, and UA marketing is the initial contact point with your customers and the starting point for mobile app marketing. We hope that you would be able to learn from the strategies introduced in this post and apply them to your app appropriately for your product’s success.
---
## What is App Store Optimization (ASO)?
URL: https://www.airbridge.io/en/blog/aso-success-guide
Language: English
Published: 2022-03-17
App Store Optimization (ASO) shares the same mechanism as Search Engine Optimization (SEO) and Landing Page Optimization (LPO). You could say that ASO is an app marketing equivalent of SEO and LPO.
As the process of optimization for an app to rank higher in an app store’s search results. Obviously, higher ranking in the app store will let an app access more potential customers.
> The purpose of ASO is to generate greater traffic to the app page in the App Store and to induce more searching users to carry out specific actions - downloads. ([Source](https://neilpatel.com/blog/app-store-optimization))
## Why ASO is So Important
As discussed in the <[The A to Z of Mobile App User Acquisition Marketing](https://www.airbridge.io/en/blog/why-a-mobile-business-needs-an-app)\>, according to [Google’s research](https://www.thinkwithgoogle.com/marketing-strategies/app-and-mobile/mobile-app-marketing-insights/), 40% of smartphone users search for apps through searching in the App Store. Also, according to [national and regional App Store Data from Apple Search Ads 2020](https://searchads.apple.com), more than 70% of visitors search for which app to look for, and 65% of all downloads are made after searches. This shows how much exposure on the app market is important.
Also, the trend of reinforcement of protection of personal information such as Apple’s new ATT framework and soon-to-be-implemented Google’s third-party cookie removal from Chrome makes marketing for organic user acquisition more important than ever. ASO has the advantage of being free unlike paid marketing, and is an effective strategy to increase the exposure of your products to users in the long run.
That’s why optimization to ensure that users are able to find your app after searching for the app’s name on the app market as well as searching for specific keywords is needed. You also need to make sure that the app page they’ve discovered after searching is attractive enough to lead to app download.
## Exploring Essential Strategies for Successful ASO
If you have experience with SEO, you will be familiar with ASO since the two share similar basic strategies. But you have to be aware of the differences arising from the nature of the app market. We will be examining the strategies for successful ASO in the order of insights from 1. Keywords; 2. Encouraging App Download; 3. App Store & Google Play Guidelines.
### 1\. Keyword
Setting the right keyword is a very fundamental part of ASO. Which keywords you use will decide whether your product would be exposed to the search results of potential users. While exposure on the App Store will depend on the quality of the product, number of downloads and reviews in ASO competition with other products sharing the same keywords, it is important to select keywords that are proper for your products at the same time as likely to be searched by potential users.
To be exposed at the top of the list in App Store and Google Play searches, it is important to analyze your potential customers to identify which keywords they might search, and which keywords are used to search for products similar to your app. Since keywords are relatively easy and straightforward to edit and test compared to other significant ASO elements, it is advised to regularly manage your keywords.
### Tips for Selecting Effective Keywords
This keyword could be used for meta data in the App Market including the app name, title, description, and keywords (only for App Store).
While it is ideal to include keywords in the name and title of the App, there is a character limit (App Store - character limit of 30) and there are negative consequences to listing a jumble of keywords that you must take care in choosing your keywords. Here are several tips for choosing effective keywords.
_1) Data-Based Research_
If it is a product with both an app and a web, you can research keywords on [Google Search Console](https://search.google.com) to see how customers discover your service. The search words used on web and app market are not always the same, but the first-party data you have accumulated so far will provide valuable insight.
Or, if your service is a product that has been newly launched or has no website, it is also advisable to use [Google Keyword Planner.](https://support.google.com/google-ads/answer/7337243?hl=ko) ‘Keyword Planner’ is one of the free tools provided by Google Ads, letting you search for new business-related keywords and identify the estimated search volume and advertising costs for specific keywords. If you have trouble coming up with new keyword ideas, refer to the [best practices for finding keywords by Google Ads.](https://support.google.com/google-ads/answer/9247190?hl=en)
- Think like your customers : Describe how your customers think of your product or service category.
- Be specific but not _too_ specific
- Add a website to improve results
- Highlight well-known brands in your offerings
- Use expressions that describe your products or services well : Use adjectives limitedly
- Use multiple keywords or keyword phrases about the same topic
Or you could use [free keyword/ASO search too](https://www.data.ai/en/insights/aso-app-store-optimization/how-to-improve-your-aso-app-store-optimization-by-maximizing-your-keyword-strategy/)l to find effective keywords.
_2) Selecting keywords just right for your business_
After gaining insight through data-based research, you need to strategically select keywords appropriate for your business. You need to use keywords that are relevant to your product as well as being able to be exposed to many users in a short time since there are limitations to the number of keywords you could use and there are varying degrees of competition for different keywords.
It would be a good idea to check the search volume for each keyword by using services such as Google Keyword Planner or App Annie’s Keyword/ASO Search Tool when choosing keywords. Referring to keyword suggestions from the app market for major keywords is also a good idea.
You will face fierce competition if you want to use keywords such as ‘memo’, 'weather’, ‘social media’, and ‘photo’ that not only represent your business but are generally used. On the other hand, competition may not be as fierce if you are using a detailed keyword with compound words and adjectives.
If you are new to ASO, it would be a good idea to use long-tail keywords with lesser degrees of competition rather than using the big popular keywords first. Promoting downloads through keywords with less competition may boost your ranking for popular keywords with fierce competition. For instance, you can consider keywords such as ‘weather today’ and ‘weather tomorrow’ for ‘weather’.
_3) Other tips you could refer to_
- Using brand and product names as keywords would be something you do naturally. Here, it is important to make a combination of keywords in the local language and English. If you are targeting Korean customers, it is a good idea to both include Korean and English.
- Include the core keyword in the title and subsidiary content in the subtitle in detail.
- Keyword optimization is not everything there is to ASO. No matter how much care you put into which keywords to use, products incapable of satisfying customers will never rank high. But with good products, you need to research thoroughly to make sure that you don't miss keywords or opportunities to expose your app to customers.
- After selecting your keywords, you need to continue to monitor exposure of your app through keywords. You also need to continue to search for more effective keywords and make updates.
### 2\. Encouraging Customers to Download Your App After Finding it
If you’ve improved the likelihood of customers finding your app through relevant keywords, you should now build a detailed page (App Store: Product Page / Google Play: Store Listing) attractive enough to encourage your users to download the app as well as promoting intuitive and quick understanding of your app.
[Ex) My Device ID by Airbridge enabling users to check and share device information ]
_1) Icon_
One of the most prominent elements in app market search results and category would be the app icon. Sometimes they are more eye-catching than app titles that users sometimes tap for more information by just seeing the app icon. Not only that, the app icon has a significant impact on getting users to tap the app on the main screen of smartphones. That’s why you should design an icon that embodies the identity of the product at the same time as intuitively expressing the purpose and services served by the app. Try testing multiple options to decide which icon is the most eye-catching and meaningful.
_2) Screenshot and video_
- App Store: App preview video, screenshot
- Google Play: Graphic images, screenshots, promotion videos
The screenshot images and relevant videos of the product play significant roles on not only the app market’s search results but detail page. It is an opportunity to effectively let potential users know the offerings of the app. The detail page must attractively present the values and core functions provided by the product for the user and have main offerings exposed in the first two images.
While the size of screenshots is fixed, it is possible to add two connecting images if you want to display an image in landscape orientation like the screenshot image for My Device ID by Airbridge presented above.
Just like A/B testing for image and video creatives are needed in advertisement campaigns, A/B testing for screenshots and videos in the app market is important. Testing to find creatives with better reactions and updating regularly will help you find creatives optimized for your product.
It would also be a good idea to try screenshot designing tools such as [AppLaunchpad](https://theapplaunchpad.com/), [AppScreens](https://appscreens.com/), or [Previewed](https://previewed.app/) to create customized screenshot images.
_3) Description_
While descriptions are longer compared to app titles and keywords, longer descriptions do not always lead to better performance. The very first one or two sentences would be the most powerful and not many users will click ‘See more’ to explore further. That’s why you need to include compelling sentences at the beginning of the description to encourage downloading the app. A/B testing to decide on appropriate phrases is a viable option.
Especially, sharing new updates in app description will let users be informed of continuous updates as well as promoting new features to potential users naturally. In the case of the App Store, this information could be shared at the top of the promotional text sharing news on discounts, new features, and other latest news about the app.
_4) Rating and review_
[Like how Apps Store mentions that ratings and reviews have the power to impact the app’s ranking](https://developer.apple.com/app-store/product-page/), ratings and reviews also have significant impacts on ASO. They are also important because potential users tend to examine the experience of different users through ratings and reviews when contemplating whether to download an app. It is natural for users to trust apps with higher ratings and more reviews when there are multiple apps serving the same purpose. That’s why you should plan meticulously to request users using the app frequently to leave ratings and reviews in a timely manner to receive as much feedback as you can.
_5) Testing app detail page_
In building an app’s detail page, continuous tests are needed to ensure the detail page’s best performance. It is ideal to make the best of the test functions provided by each app market since they are either planning to provide such functions or are already providing such functions to help apps.
Tests can be carried out for app icons, screenshot images, and descriptions. It is best to test three different versions for a detailed page. Up to three different versions could be tested for the App Store and Google Play recommends preparation of three or more alternatives.
- App Store will provide [product page optimization](https://developer.apple.com/app-store/product-page/) from the second half of the year to let apps test app icons, screenshots and previews to compare the performance to identify what leads to better user response.
- Google Play helps you find the optimal app icon, image, video, and app description through [store listing experiments.](https://developer.android.com/distribute/best-practices/grow/store-listing-experiments) A/B testing can be carried out for store listing changes for specific target users, changing store listing information and boosting the number of downloads. You can check the performance on [Play Console](https://developer.android.com/distribute/console) and decide on the best settings for your store listing by performance.
### 3\. Utilizing the insight gained from the guidelines of top two app markets (App Store & Google Play)
App Store and Google Play, the top two major app markets have their own regulations and standards on their app detail pages and are used by different users. That is why an app needs a coherent ASO strategy for both markets yet needs and also needs to carry out ASO reflecting the different characteristics and standards of each app market.
Also, both app markets are providing guidelines to comply with on the developer’s page as well as tips for your app to gain more exposure and updates. It would be wise to keep an eye out for these details both before and after launching your app.
Here are some pages for developers from each app market and other websites that may be helpful for you.
#### App Store
[The developer page on the App Store](https://developer.apple.com/) offers the latest updates, cautions in listing your app on the App Store and other details. You could also refer to [App Store Connect Help](https://help.apple.com/app-store-connect/?lang=en#/) for listing and managing apps on the App Store. The [App Store icon, app preview, and screenshots overview](https://help.apple.com/app-store-connect/?lang=en#/dev910472ff2) is especially helpful for ASO.
Searches can be made on the App Store by the name of the app, subtitles of the app, keywords and company name. As for the creatives you could use on the App Store, you can use an app icon and post up to three app previews and ten app screenshots.
Especially, [Creating Your Product Page on the App Store](https://developer.apple.com/app-store/product-page/) presents guidelines on the elements comprising product page such as app name, icon, subtitle, app preview, screenshot, promotional text, keywords, in-app purchase, new features, and ratings and reviews and suggestions on creating your product page. This page is a must-read before building your product page on the App Store.
Also, according to [Apple’s announcement](https://developer.apple.com/app-store/whats-new/), devices equipped with a minimum of iOS 15 or iPadOS15 will be able to build more relevant to customers through product page optimization and custom product pages. This is something you would definitely want to try.
#### Google Play
Google Play also offers a [Google Play developers page](https://developer.android.com/distribute). You may especially want to take note of [build a high-quality app or game](https://developer.android.com/distribute/best-practices/launch/store-listing) page.
According to the guide, app recommendations made by Google are made by both human selection and algorithm. The biggest two factors are ‘relevance’ and ‘quality’. One of the most important elements in evaluating the quality of an app is whether the app has an effective store listing page.
You must build trust with users and leave strong first impressions through effective and accurate store listing page since discovery of an app starts from the store listing page. Users must be able to clearly identify the features and contents provided by an app through the app description, screenshots, and other metadata provided on the listing page. Here are some tips for an app to be exposed to more people on Google Play.
- You can boost conversion rate by adding videos to the app’s listing page and including three or more 16:9 ratio screenshots.
- Select the right [category](https://support.google.com/googleplay/android-developer/answer/9859673?hl=en&ref_topic=3450987&visit_id=637834403309179044-1512339074&rd=1) and [content rating](https://support.google.com/googleplay/android-developer/answer/9859655?hl=en&ref_topic=3450987&visit_id=637834403309179044-1512339074&rd=1) to be able to promote your app for the right audience on Google Play.
- Carry out A/B testing for multiple versions of your listing page by including different icons, descriptions, and screenshots for actual Google Play users on [store listing experiments](https://developer.android.com/distribute/best-practices/grow/store-listing-experiments). You must test each element independently and make sure that the length of your experiment extends over at least a week for meaningful results.
Also, statistics on Organic Insight from the User Acquisition tab on [Play Console](https://developer.android.com/distribute/console) lets you see how users discover your app on Google Play, including search words.
## Conclusion
Like SEO, it is important to make continuous observations and gradual enhancements and optimizations for ASO rather than reaching perfection instantaneously. ASO could be said to be the continuous process of testing and measuring contents and resources in the app market to find the best solution for your app to be effectively exposed and downloaded.
We hope that your app will be able to acquire more users and grow by increasing exposure in the app market based on the ASO insights provided in this post.
---
## 4 Reasons why an app is a must for mobile businesses
URL: https://www.airbridge.io/en/blog/why-a-mobile-business-needs-an-app
Language: English
Published: 2022-03-17
Widespread use of personal devices with increased distribution of smartphones has heralded the exponential development of mobile businesses. Your potential users can always be online via mobile devices without constraints of time and space. They spend considerable time on using mobile apps everyday. Especially, the time spent on mobile apps increased with the breakout of the COVID-19 pandemic. [According to App Annie](https://www.data.ai/en/insights/market-data/q2-2021-market-index-regional-rankings/), the time spent on mobile apps during the 2nd quarter of 2021 is still considerably higher than the figure of the 2nd quarter of 2019 before the COVID-19 pandemic hit.
The increase of mobile device use and time spent on the app led to the increase of potential users for mobile business companies and user data. The changing landscape of the mobile market abetted the need for businesses to optimize their websites for mobile and to have their own apps. The number of apps is increasing steadily for the same reason. However, the increasing number of apps led to fierce competition, and it became harder than ever to catch users on your app.
Why should mobile businesses build their own apps to succeed despite the keen competition?
## What’s the difference between the mobile web and app
Before getting to know why making an app is a great idea, it is necessary to understand the difference between the mobile web and the mobile app. Mobile businesses are mainly based on the two platforms - the web and the app - to boost the visibility of their products and to communicate with customers. Since two platforms play significant roles in the customer journey, it is pivotal to form a strategy for the two elements to be seamlessly combined.
**The mobile web** is a platform capable of acquiring potential customers through search engines and other means, enabling contact with the brand in the extensive online network. The web serves as an opportunity for acquiring varied and many customers since they can access the products with the mobile web without downloading apps. If the customer is satisfied with the experience of encountering a brand through the web, that encounter could serve as a starting point in which a random visitor becomes a real customer.
However, it must be noted that there are limitations for the mobile web in providing tailored customized experiences and customers are likely to have different user experiences depending on what browser they are using. Also, an unstable payment system on the mobile web may lead to user churn during the payment process. This will have negative consequences for retention and conversion rates.
**The mobile app** is capable of making up for the shortcomings of the mobile web. Customers can be provided a tailored experience since they are more likely to be continued to be signed in, making their interaction with the brand seamless. Notifications and individualized promotions also enable high-quality personalization. This will lead customers to stick to your products and put you in a better position for more active customer engagement. The mobile app will also resolve uncertainties of user experience from various mobile web browsers they are using.
Nowadays, more companies are choosing to create their own mobile apps, even though it is technically possible to have only a mobile web to run the business, since apps are capable of complementing the shortcomings of the web and provide certain benefits. Let’s dive into the four reasons why mobile businesses must have their own apps as follows.
## 4 reasons why mobile app is a good idea for business
### 1\. Mobile apps are capable of generating high conversion rates and profits
One of the most decisive reasons for the need for mobile apps is that they have a higher conversion rate than the web. Retention, purchase conversion, and LTV can also be improved as well. While it’s no easy job to get customers to install an app, the value they can generate from the customer after installation is far greater. Apps not only enable fast and easy interactions with the customer but also enable customer engagement and provide a personalized user experience based on data.
Here are some research findings proving the high conversion rate of mobile apps.
First of all, according to [Criteo’s research findings](https://www.criteo.com/blog/retail-travel-apps-higher-conversions-mobile/), mobile apps have a conversion rate 3-4 times higher than that of the mobile web in e-commerce and travel. More specifically, while the conversion rate for the mobile web was only 4% in e-commerce advertisements, the conversion rate for the app was 18%. Not only that, the conversion rate for the mobile web was 6% while the conversion rate for the app was 20%. The following image compares the conversion funnel of global retail companies by mobile web brower and app.
[4 Reasons why an app is a must for mobile businesses1]
Also, [2019 Mobile Commerce Report: Holiday Analysis](https://www.usebutton.com/2019-mobile-holiday-analysis) released by Button analyzed the data collected during 2018 holiday shopping season and shared the following insights on the growth of the mobile market.
- The conversion rate of apps was 14% higher compared to mobile web
- Profit per tab on app was 117% higher compared to mobile web
- Order per customer made through the mobile app was 108% higher compared to mobile web.
### 2\. Apps enable faster and easier interactions with customers
Since mobile apps are optimized for mobile devices, they are fast, efficient, and can be accessed conveniently. Because mobile apps are placed as icons on the home screen, they are capable of boosting accessibility since users are able to easily access products without URLs and browsers.
Not only that, the accessibility to the device’s own features, such as cameras, GPS, biometric authentication (i.e. fingerprint recognition and facial recognition such as touch ID), and Bluetooth has been greatly enhanced, allowing improved seamless interactions with users. Since these functions can be provided more stably on mobile apps rather than mobile browsers, they are capable of improving the usability and convenience of products.
Such features can also have the following positive effects.
- **Enhancing Brand through the delivery of consistent customer experience:** Customers would gain reliability and loyalty by receiving fast and reliable experiences from the brand within the product.
- **Mobile app as an 1:1 communication channel between customers and companies:** The app allows customers and brands to communicate continuously. It is as if it serves as ‘somewhere private’ in the customer’s smartphone for the companies to actively communicate with customers.
- **Nurturing a habit of continuing the customer-brand communication:** The core values of products and services could be communicated through the app anytime, anywhere through mobile devices always carried in the pockets of customers. Customers will naturally revisit apps.
- **Easy and Simple payment:** The process of adding items into the cart and paying for them is relatively more stable and simpler compared to mobile browsers. Online payment services such as PayPal also make it easier to checkout.
To enjoy these benefits, it is absolutely essential to understand how to provide an app experience for customers as well as continuous UI and UX improvements are required.
### 3\. Improve your products with the First-party data
Privacy reinforcement including Apple’s new iOS 14.5 update and Google’s stopping support of third-party cookies is making it harder to collect third-party data. That is why the collection of first-party data based on logged in users has become more important than ever. Once users sign in on the app, users are able to enjoy improved convenience and usability while brands can analyze user activities based on data and guide users on the planned user journey with CRM marketing solutions such as Braze. First-party data also enables collection of more accurate user data compared to cookie-based data.
It is effective to analyze the collected data to identify where customers churn and how to improve the retention rate as well as analyzing by [cohort analysis](https://www.chameleon.io/blog/cohort-analysis) through analytic solutions such as Amplitude. Such data analysis could contribute to improvement of UI and UX of products and increased conversion and retention rates through added core functions. The key to product-led growth is the actual behavior data collected from users.
Furthermore, Mobile Attribution Softwares such as Airbridge can help to organize re-engagement marketing campaigns with users who install the app from ads.
### 4\. Finely personalized customer experience
Customer’s data is also valuable for providing personalized experience to customers. The customer information and data collected from each product enable more sophisticated and customized user experience as well as personalized marketing campaigns. According to the research results [provided by Think with Google](https://www.thinkwithgoogle.com/marketing-strategies/app-and-mobile/smartphone-user-personalization-preferences/), 58% of smartphone users feel more favorable towards companies whose mobile sites or apps remember who they are and their past behavior. This puts into perspective how much personalized experience based on customer behavior is important.
[__wf_reserved_inherit]
Push notifications and in-app messages can be used to send out messages to customers at the right time and place. Sending in-app notifications is even free like an added bonus. CRM marketing solutions such as Braze is a considerable option if you would like to carry out more advanced in-app campaigns. According to [Braze’s research findings](https://www.braze.com/resources/reports-and-guides/in-app-messaging) ([KO](https://www.ab180.co/library/braze-in-app-message-guide?utm_source=ab180-blog&utm_medium=cta&utm_campaign=braze-in-app-message-guide))there are research findings that customer engagement increased by 131% after communicating with users through in-app messages. Promotion through such means will lead to immediate user response and visible revenue growth as well.
## Conclusion
To sum up, mobile business requires both mobile web and app. Despite the resources needed for app development and fierce competition, apps can effectively contribute to improving retention and conversion rates and profit growth. Apps also enable continuous and personalized communication with customers as well as providing first-party customer data to gain a foothold in mobile business through product growth and high-quality personalization.
The app is definitely one of the most important means for the success of mobile business.
---
## Announcing Airbridge Incrementality its first-to-market solution leveraging Facebook's MMP, MTA Programs
URL: https://www.airbridge.io/en/blog/airbridge-becomes-facebook-business-partner-for-measurement
Language: English
Published: 2021-10-07
Airbridge, a universal attribution platform for web and mobile, today announced the launch of Airbridge Incrementality to help brands measure the incremental effectiveness of their advertising campaigns, and with this offering, becomes the first attribution service provider to be both a Facebook badged Mobile Measurement Partnership (MMP) and Multi-touch Attribution (MTA) Partner.
Airbridge Incrementality signals a paradigm shift in the attribution industry: Airbridge Incrementality, made possible with Facebook's MTA program, directly answers the industry’s call for a more holistic measurement technique that analyzes the incremental benefits of their advertising on user conversion, while accounting for the multi-touch and cross-platform values in attribution analysis.
“With Airbridge Incrementality, Airbridge is positioning itself as a complete attribution toolkit that helps bridge the current gaps in the attribution and measurement industry: offering marketers a full-suite solution from unified web and mobile attribution, extensive deep linking, and finally a game-changing incrementality measurement suite,” said Roi Nam, CEO and Co-Founder of Airbridge.
With incrementality results refreshed daily and obtainable once the incrementality machine learning engine is properly integrated with the right amount of data sets, marketers using Airbridge Incrementality are able to gain deeper actionable insights into how various metrics such as CPA for ‘Install’ and ‘Order Complete’ events, and ROAS for ‘Revenue’ are performing incrementally across various advertising channels.
Commenting on a change in their marketing measurement practices, Woohyun Park, Head of Marketing Business from GS Retail, one of Korea’s largest home shopping & retail companies, remarked: “We were able to quickly obtain a trustable second opinion through Airbridge Incrementality to compare whether certain advertising channels were overvalued or undervalued by last-touch attribution, followed by studying the incremental benefits of our advertising towards various metrics before optimizing our spend on advertising channels that deliver the highest incremental benefits for user acquisition and conversion.”
Airbridge Incrementality will immediately be available to all new and existing customers.
To learn more about Airbridge, please visit: [https://www.airbridge.io](https://www.airbridge.io/)
● Request a demo: [https://www.airbridge.io/en/book-a-demo](https://www.airbridge.io/en/book-a-demo)
**About Airbridge**
Airbridge is a universal attribution and incrementality measurement platform, helping brands identify the true sources of their digital growth across web and mobile. Leading brands such as GS Retail, Standard Chartered Korea, and Burger King Korea rely on Airbridge’s full-suite measurement toolkit: from unified web and mobile analytics to extensive deep linking, to bootstrap their campaign measurement process from start-to-end.
---
## Airbridge becomes TikTok’s Mobile Measurement Partner (MMP)
URL: https://www.airbridge.io/en/blog/airbridge-x-tiktok
Language: English
Published: 2021-06-24
Airbridge is proud to introduce you to a new partnership with TikTok! It is our great pleasure to announce that Airbridge has partnered with TikTok as their Mobile Measurement Partner (MMP) and is thrilled for exciting collaboration opportunities as TikTok’s Mobile Measurement Partner (MMP). Through this partnership, Airbridge aims to better help our clients create, target, measure and optimize TikTok campaigns.
##### Here are some of the features you can use to accurately measure and optimize your TikTok campaigns.
- In-App Events Postback: send Airbridge-collected in-app events and event properties data to TikTok platform through simple and easy set-up on Airbridge dashboard. (Use “All Postback” setting inside Airbridge dashboard in order to maximize event volume)
- App Event Optimization (AEO): use TikTok AEO feature with the collected data from Airbridge to target users who have greater chance of conducting specific in-app events, which helps advertisers to more efficiently acquire high-value users. Click [here](https://ads.tiktok.com/help/article?aid=14175) for more information about AEO.
- Pangle Performance Measurement: track performance of Pangle, a global monetization and user acquisition platform operated by Bytedance, in Airbridge dashboard as well!
Airbridge accurately tracks and measures behaviors of users who incurred installs and in-app events by using a tracking link for TikTok. By this, you can not only properly measure performance but also utilize the result in the media mix to maximize ROI.
It is our great pleasure to become TikTok’s Mobile Measurement Partner (MMP) and to be able to provide more effective and accurate measurement to clients of TikTok and Airbridge. We will continue working closely together with TikTok to grow and enhance our integration.
## Introducing TikTok for Business
[TikTok for Business](https://www.tiktok.com/business/ko) is where brands and marketers communicate with TikTok community as creative storytellers. You can freely build fun videos with entertaining music for interactive campaigns. Moreover, TikTok for Business advocates creative marketing activities and their business growth for various marketing partners through [TikTok Marketing Partners Program](https://ads.tiktok.com/marketing-partners/pc/en?rid=1afduegrj1k).
# Korean Articles
---
## 에어브릿지와 Amazon Ads 연동으로 앱 캠페인 성과를 측정해요
URL: https://www.airbridge.io/ko/blog/airbridge-amazon-ads-integration
Language: Korean
Published: 2026-09-17
Author: Jaehyuk Kim
에어브릿지가 Amazon Ads를 통합 측정 채널로 지원해요. Amazon Ads에서 앱 캠페인을 운영하는 마케터는 이제 앱 설치와 인앱 이벤트의 기여도를 측정하고 성과 신호를 Amazon DSP로 다시 전송해 캠페인 최적화에 활용할 수 있어요. 전체 미디어 믹스의 성과도 에어브릿지의 통합 리포트에서 비교할 수 있고요.
**핵심 포인트**
- Amazon Ads는 리테일을 넘어 Prime Video, Fire TV, Twitch를 포함하는 멀티스크린 광고 서비스로 확장됐어요. 쇼핑과 스트리밍 신호를 활용하기 때문에 커머스, 구독, 핀테크 앱 마케터가 주목할 만한 채널이에요.
- 에어브릿지에서는 별도의 커스텀 개발 없이 연동을 설정할 수 있어요. 몇 단계만 거치면 기존 어트리뷰션 환경에 Amazon Ads를 연결하고 곧바로 앱 캠페인 성과를 측정할 수 있어요.
## **Amazon Ads란?**
Amazon Ads는 다음과 같은 인벤토리에 광고를 게재해요.
- **Prime Video와 Fire TV:** 스트리밍 및 커넥티드 TV 인벤토리
- **Twitch:** 라이브 콘텐츠와 게임 오디언스
- **Amazon DSP:** Amazon.com과 수천 개의 앱 및 웹사이트에서 오디언스에게 도달할 수 있는 프로그래매틱 광고 플랫폼
Amazon Ads는 쇼핑과 스트리밍 신호를 활용해 광고주가 관련성 높은 오디언스에게 도달하도록 지원해요. 모바일 앱 마케터는 기존 미디어 믹스를 보완하는 추가 오디언스에게 도달할 수 있어요.
## **앱 마케터가 Amazon Ads를 주목해야 하는 이유**
퍼포먼스 마케터는 기존 미디어 믹스로 아직 만나지 못한 고객에게 도달할 수 있는 채널을 늘 찾고 있어요. Amazon Ads가 앱 마케터에게 특히 의미 있는 이유는 두 가지예요.
- **쇼핑과 스트리밍 신호를 활용해요.** Amazon은 상품을 탐색하고 비교한 뒤 구매를 결정하는 사용자의 쇼핑 의도를 파악할 수 있어요. 이러한 맥락을 활용하는 Amazon Ads는 쇼핑, 구독 서비스, 금융 상품 등의 앱이 관련성 높은 고객에게 도달하는 데 효과적인 채널이 될 수 있어요.
- **커넥티드 TV에서의 접점이 늘고 있어요.** Amazon Ads는 Prime Video와 Fire TV를 포함한 스트리밍 TV 인벤토리를 제공해 앱 마케터가 거실 환경의 오디언스에게도 도달할 수 있게 해요. 모바일 중심의 채널에서 벗어나 미디어 믹스를 다각화하려는 팀은 에어브릿지에서 이러한 인벤토리가 앱 성장에 얼마나 기여하는지 측정할 수 있어요.
에어브릿지가 Amazon Ads를 측정 채널로 지원하면서 마케터는 기존 어트리뷰션 환경에 이 채널을 연결할 수 있게 됐어요.
## **에어브릿지와 Amazon Ads 연동으로 할 수 있는 일**
에어브릿지는 Amazon Ads를 크로스 채널 어트리뷰션, 인앱 이벤트 측정, 통합 성과 리포트와 연결해요. 실제로 무엇을 할 수 있는지 살펴볼게요.
### **1\. Amazon Ads의 멀티스크린 인벤토리 전반에서 기여도를 측정해요**
Amazon Ads는 Amazon.com, Prime Video, Fire TV, Twitch와 Amazon DSP 네트워크를 비롯한 여러 인벤토리에서 광고를 운영해요. 고객이 Amazon 광고를 보거나 클릭한 뒤 앱을 설치했을 때 일반적인 어트리뷰션 환경에서는 이러한 접점이 측정되지 않는 경우가 많아요.
에어브릿지 연동은 Amazon Ads 캠페인의 접점과 이후의 앱 활동을 연결해요. 이를 통해 Amazon의 여러 인벤토리가 앱 설치와 전환에 얼마나 기여했는지 더 명확하게 파악할 수 있어요.
### **2\. 캠페인 최적화에 활용할 인앱 이벤트를 포스트백으로 전송해요**
성과를 측정하는 것만으로는 캠페인이 자동으로 개선되지 않아요. Amazon Ads 연동의 실질적인 가치는 에어브릿지가 수집한 인앱 이벤트를 포스트백으로 전송해 캠페인 최적화에 활용할 수 있다는 데 있어요. 다음과 같은 이벤트를 보낼 수 있어요.
- **앱 설치:** 신규 사용자 획득을 나타내는 기본 이벤트
- **회원가입:** 초기 참여 수준을 보여주는 이벤트
- **구매와 구독:** 매출과 연결되는 주요 전환 이벤트
- **기타 인앱 이벤트:** 필요에 따라 Amazon Ads 이벤트 유형과 매핑한 맞춤 이벤트
Amazon DSP는 이러한 신호를 활용해 입찰과 최적화 로직을 정교하게 조정하며 캠페인의 관련성과 성과를 개선해요. 에어브릿지가 수집한 어트리뷰션 데이터와 Amazon의 캠페인 시스템에서 이루어지는 최적화 의사결정이 하나의 흐름으로 이어지는 셈이에요.
### **3\. 전체 미디어 믹스의 성과를 하나의 리포트에서 비교해요**
새로운 채널을 추가할 때 어려운 점 중 하나는 기존 채널과 공정하게 성과를 비교하는 일이에요. 매체 리포트의 지표는 각 플랫폼의 기준에 따라 측정되기 때문에 채널 간 수치를 신뢰성 있게 비교하기 어려워요.
Amazon Ads를 에어브릿지에 연결하면 Meta, Google, TikTok, Apple Search Ads를 비롯한 다른 채널의 성과와 같은 리포트에서 확인할 수 있고, 다음 항목을 함께 평가할 수 있어요.
- 어떤 채널이 앱 설치와 활성화를 만들었는가
- Amazon Ads를 통해 유입된 사용자의 리텐션과 전환 성과는 어떠한가
- 전체 매체 믹스에서 Amazon Ads가 어떤 역할을 하는가
각 광고 네트워크가 서로 다른 기준으로 측정한 매체 리포트에 의존하지 않고 모든 채널의 예산을 결정할 때 일관된 기준을 적용할 수 있어요.
## **Amazon Ads를 어트리뷰션 환경에 연결해 보세요**
Amazon Ads는 앱 마케터가 스트리밍, 쇼핑, 커넥티드 TV 환경에서 새로운 오디언스에게 도달할 수 있도록 도와요.
에어브릿지가 Amazon Ads를 측정 채널로 지원하면서 이제 Amazon의 인벤토리를 기존 어트리뷰션 환경에 간편하게 연결할 수 있어요. 간소화된 설정 과정을 거치면 캠페인 성과 측정, 이벤트 포스트백, 크로스 채널 리포트를 모두 이용할 수 있어요.
자세한 설정 방법은 [**에어브릿지 도움말**](https://help.airbridge.io/en/guides/amazon-ads-channel-integration)에서 단계별로 확인할 수 있어요.
---
## 금융사 보안팀이 알아야 할 좋은 MMP 고르는 방법과 체크리스트 6가지
URL: https://www.airbridge.io/ko/blog/finance-mmp-security-checklist
Language: Korean
Published: 2026-09-02
Author: Roi Nam
금융사가 모바일 측정 파트너(Mobile Measurement Partner, MMP)를 도입할 때는 마케팅 기능만 비교해서 결정하기 어려워요. 도입 전 보안성 심의부터 도입 후 정기 감사까지 고려해, 사업자의 보안 통제와 증적 지원 역량을 함께 확인해야 하기 때문이에요.
주요 검토 대상은 감사 로깅, 접근통제, 인증 범위, 데이터 레지던시, 테넌트 분리, 위수탁 증적인데요. 이 가운데 계정 및 권한 관리와 로깅 및 모니터링은 금융보안원이 공개한 상용 클라우드 보안관리 기준에서도 다루는 영역이에요.
이 글에서는 금융사가 MMP를 검토할 때 확인해야 할 보안 기준을 6가지로 정리하고, 각 기준을 에어브릿지(Airbridge)가 어떻게 지원하는지 살펴볼게요.
### **📌핵심 포인트**
아래 표는 금융사 MMP 보안성 검토에서 확인할 6가지 영역과 추가 질문을 정리한 내용이에요. 사업자의 지원 여부뿐 아니라 지원 범위와 운영 조건까지 같은 기준으로 비교할 수 있어요.
| 검토 영역 | 확인할 기준 | 추가 확인 질문 |
| --- | --- | --- |
| 감사 로깅 | 활동 기록과 조회·보존기간 | 화면 조회기간과 서버 보존기간은 각각 얼마인가요? |
| 접근통제 | 역할 분리와 권한 회수 | 권한을 회수하면 API 토큰도 무효화되나요? |
| 인증·통제 | 인증의 유효기간과 범위 | 최신 인증서에 도입 서비스가 포함되나요? |
| 데이터 레지던시 | 데이터와 백업의 저장 위치 | 백업과 로그도 국내에 저장되나요? |
| 테넌트 분리 | 프로젝트별 데이터·권한 분리 | 여러 SDK 인스턴스에서도 딥링크가 전달되나요? |
| 위수탁 증적 | 증적의 종류와 기준일 | 자료는 언제, 어떤 주기로 갱신되나요? |
## **금융사 MMP 보안성 검토 체크리스트 6가지**
### **1\. 사용자 활동을 감사 로그로 추적할 수 있나요?**
금융사는 적용되는 법령과 내부 보안 기준에 따라 필요한 접속기록 보존기간을 정해야 해요. 그래서 MMP를 검토할 때는 관리 화면에서 조회할 수 있는 기간과 사업자가 서버에 보존하는 기간을 나눠 확인해야 해요.
특히 MMP를 검토할 때 로그는 두 층으로 나눠 봐야 하는데요. 우리 담당자가 대시보드에서 무엇을 했는지 남는 이용자 활동 로그, 그리고 사업자 직원이 우리 데이터에 접근했을 때 남는 사업자 접근 로그예요.
#### **이걸 확인해 보세요.**
- 관리자 활동과 설정 변경에 대한 감사 로그(Audit Log)를 남기는가
- 로그에서 행위자, 대상, 활동 유형, 발생 시각을 식별할 수 있는가
- 활동 유형과 사용자 기준으로 로그를 필터링해 조회할 수 있는가
- 사업자 직원이 고객 데이터에 접근했을 때도 별도로 기록되는가
#### **화면 조회기간과 서버 보존기간은 각각 얼마인가요?**
화면에서 조회할 수 있는 기간과 서버에 보존하는 기간은 다를 수 있어요. 두 기간을 따로 물어보고, 필요한 장기 이력의 제공 가능 여부와 제공 방식도 확인하세요.
#### **Airbridge는 사용자 활동을 계정별로 기록해요**
- 기록 항목: 행위자, 대상 메뉴, 활동 유형, 발생 시각이 함께 남아요. 활동 상세 화면에서는 접속 IP와 활동 경로도 확인할 수 있어요.
- 조회 방식: 생성·수정·요청·삭제 등 활동 유형과 계정을 기준으로 기록을 필터링할 수 있어요.
- 조회 기간: 관리 화면에서는 최근 7일의 기록을 확인할 수 있어요. 그보다 긴 기간의 이력이 필요하다면 제공 가능 여부와 방식을 도입 전에 확인해야 해요.
사업자 접근 로그에 관한 운영 기준은 도입 검토 단계에서 확인할 수 있어요.
👉 [Airbridge 사용 기록 도움말](https://help.airbridge.io/ko/guides/activity-history)
[Airbridge 사용 기록 화면에서 활동 유형과 계정을 기준으로 기록을 필터링하는 모습]으로 목록을 필터링하고 계정 기준으로도 조회할 수 있어요. 출처: Airbridge 공식 도움말")
[Airbridge 사용 기록 상세에서 행위자, 대상, 발생 시각, 접속 IP, 활동 경로를 확인하는 모습]
### **2\. 사용자 권한을 필요한 시점에 회수할 수 있나요?**
MMP에는 사내 마케팅팀뿐 아니라 데이터팀과 대행사 등 여러 사용자가 접근하는데요. 담당자의 역할이 바뀌거나 외부 협업이 끝났을 때 기존 권한을 그대로 두면 불필요한 접근이 남게 돼요. 따라서 조직과 앱 단위로 권한을 나누고 필요한 시점에 변경하거나 회수할 수 있는지 확인해야 해요.
#### **이걸 확인해 보세요.**
- 역할 기반 접근 통제(RBAC, Role-Based Access Control)를 제공하는가
- 조직 권한과 앱 권한을 분리해 최소 권한 원칙과 직무 분리를 적용할 수 있는가
- 접근 권한의 부여·변경·회수를 화면에서 관리할 수 있는가
- 퇴사·이동 시 권한을 회수하고 정기 접근권한 검토에 사용할 접근 목록을 확인할 수 있는가
#### **권한을 회수하면 API 토큰도 함께 무효화되나요?**
화면에서 권한을 회수했을 때 그 사용자가 발급받은 API 토큰과 연동 자격 증명도 함께 무효화되는지 확인하세요. 화면 접근을 막더라도 토큰이 유효하면 데이터에 접근할 수 있는 경로가 남을 수 있어요.
#### **에어브릿지는 조직 역할과 앱 역할을 분리해 관리해요**
조직 사용자를 초대하거나 제거할 수 있고 사용자별 앱 접근 권한과 역할을 부여·변경·회수할 수 있어요. 앱 접근 권한을 회수하면 해당 사용자가 보유한 API 토큰과 트래킹 링크 API 토큰도 더 이상 쓸 수 없어요.
다만 그 사용자가 이전에 만든 트래킹 링크 데이터는 그대로 남아서, 회수가 과거 성과 데이터를 지우지는 않아요.
에어브릿지의 앱 역할은 Owner(소유자), In-house Marketer(사내 마케터), Agency(대행사), Media Partner(매체 파트너) 4가지로 나뉘고, 역할에 따라 해당 앱에서 접근할 수 있는 기능과 데이터가 달라져요.
| 권한 | Owner | In-house Marketer | Agency | Media Partner |
| --- | --- | --- | --- | --- |
| 앱 삭제 | 가능 | 불가 | 불가 | 불가 |
| 원본 데이터 추출 | 가능 | 가능 | 허용된 속성만 | 불가 |
| 실시간 로그 조회 | 가능 | 가능 | 기본 권한에서는 제한 | 불가 |
| 리포트 조회 | 가능 | 가능 | 일부 리포트·같은 대행사 데이터만 | 액츄얼스·기여 채널만 |
| 어트리뷰션 기간 설정 | 가능 | 가능 | 조회만 | 조회만 |
사용자 초대와 권한 변경 범위는 조직 역할이 결정하고, 특정 앱에서 접근할 수 있는 기능과 데이터는 앱 역할이 결정해요.
대행사는 조직 역할과 앱 역할에 모두 존재하므로 두 역할을 구분해서 확인해야 해요. 조직 역할이 대행사인 사용자는 권한을 부여받은 앱에 한해 같은 대행사 사용자에게 접근 권한을 부여할 수 있고, 앱 역할이 대행사인 사용자는 허용된 범위에서 해당 앱의 기능과 데이터에 접근할 수 있어요.
👉 [조직 사용자 관리](https://help.airbridge.io/ko/guides/organization-level-user-management)
👉 [Airbridge 사용자 권한 체계](https://help.airbridge.io/ko/guides/user-management)
### **3\. 필요한 보안 인증과 통제 체계를 갖추고 있나요?**
MMP 내부 보안 체계를 고객사가 직접 모두 점검하기는 어려워요. 보안 인증과 독립 감사 결과는 사업자가 일정한 기준에 따라 정보보호 체계를 운영하는지 확인할 수 있는 근거로 활용할 수 있어요.
따라서 인증 명칭에서 멈추지 말고 유효기간과 인증 범위, 침해사고 대응과 서비스 연속성 체계까지 함께 살펴봐야 해요.
#### **이걸 확인해 보세요.**
- ISMS(정보보호 관리체계) 같은 국내 인증을 유효하게 유지하는가
- ISO/IEC 27001·27017·27018 국제 표준 인증을 보유하는가
- SOC 2 Type 2(미국 AICPA 기준으로 내부통제 운영을 일정 기간 관찰해 검증하는 독립 감사) 검사를 받았는가
- 침해사고 대응과 서비스 연속성(BCP/DR, Business Continuity Plan and Disaster Recovery) 통제를 운영하는가
#### **인증서가 현재 유효하고 도입 서비스까지 포함하나요?**
인증 목록만 확인하지 말고 인증서 자체를 요청하세요. 인증서가 현재 유효한지, 인증 범위에 도입하려는 서비스가 포함되는지, ISO/IEC 27001이 최신 표준인 2022 버전인지 확인해야 해요.
#### **에어브릿지는 보안 인증과 통제 현황을 공개해요**
에어브릿지는 ISMS(인증번호 ISMS-KISA-2023-138)와 ISO/IEC 27001·27017·27018 인증을 취득했고, Deloitte Anjin LLC가 수행한 SOC 2 Type 2 검사를 완료했어요.
인증서 사본과 유효기간은 도입 검토 단계에서 요청해 확인할 수 있어요.
👉 [Airbridge 정보보호 페이지](https://www.airbridge.io/ko/information-security)
### **4\. 데이터 레지던시를 선택할 수 있나요?**
MMP는 광고 식별자와 앱 활동 데이터처럼 보안 검토가 필요한 정보를 수집하고 처리해요. 그래서 서버가 위치한 지역만 확인해서는 데이터가 실제로 어디를 거쳐 처리되는지 알기 어려워요. 그러므로 저장 지역과 함께 전체 데이터 흐름, 하위수탁자의 처리 위치까지 확인해야 해요.
#### **이걸 확인해 보세요.**
- 국내 리전 등 데이터 레지던시(Data Residency) 옵션을 제공하는가
- 수집·처리·저장 경로를 데이터 흐름도로 제시할 수 있는가
- 하위수탁자(Sub-processor) 목록과 처리 위치를 공개하는가
#### **백업과 로그도 국내에 저장되나요?**
국내 리전이 기본값인지 별도 신청 옵션인지 확인하세요. 원본 데이터 외에 백업과 로그도 국내에 남는지, 국내 리전 선택 사항이 서명할 계약서(DPA)에 반영되는지도 함께 물어봐야 해요.
#### **에어브릿지는 한국 데이터 서버를 지원해요**
에어브릿지는 2026년 1월 AWS 서울 리전 기반의 한국 데이터 서버 도입을 발표했어요. 데이터 흐름도와 하위수탁자 목록, 개인정보 처리 계약(DPA) 등 보안성 검토에 필요한 자료의 제공 범위와 요청 절차는 도입 전에 확인할 수 있어요.
👉 [Airbridge 한국 데이터 서버 안내](https://www.airbridge.io/ko/blog/seoul-region-launch-data-compliance)
👉 [Airbridge Security & Privacy](https://www.airbridge.io/ko/solutions/security-privacy)
### **5\. 멀티 테넌트 환경에서 법인과 서비스를 분리할 수 있나요?**
여러 법인이나 브랜드, 앱을 운영하는 금융사는 하나의 MMP를 사용하더라도 데이터와 접근 권한을 서로 분리해야 해요. 여러 고객사가 같은 인프라를 공유하는 멀티 테넌트 구조에서는, 대시보드에서 프로젝트를 나누는 것뿐 아니라 자격 증명과 SDK에서 전송하는 데이터까지 독립적으로 관리할 수 있어야 하죠. 실제 운영 구조에 맞는 분리 단위를 지원하는지 확인해 보세요.
#### **이걸 확인해 보세요.**
- 법인·브랜드·앱 단위로 프로젝트와 데이터 경계를 분리할 수 있는가
- 프로젝트별 자격 증명과 접근 권한이 독립적으로 작동하는가
- 하나의 앱에서 복수 프로젝트로 데이터를 보내야 할 때 SDK 독립 인스턴스를 지원하는가
#### **여러 SDK 인스턴스에서도 딥링크가 정확히 전달되나요?**
독립 인스턴스를 지원한다는 답을 받으면 어느 SDK 버전부터인지 확인하세요. 인스턴스가 여러 개일 때 딥링크와 지연 딥링크(deferred deep link)가 각 인스턴스로 정확히 전달되는지도 살펴봐야 해요.
#### **Airbridge SDK는 프로젝트별 독립 인스턴스를 지원해요**
- iOS·Android SDK: v4.9.0부터 하나의 앱에서 여러 Airbridge 프로젝트를 독립 인스턴스로 운영할 수 있어요.
- Android SDK: v4.10.0부터 여러 인스턴스가 하나의 앱을 공유할 때 지연 딥링크 처리도 지원해요.
- Web SDK: createAirbridge()로 여러 인스턴스를 만들고 각각 초기화할 수 있어요.
👉 [Android SDK 릴리스 노트](https://help.airbridge.io/ko/developers/release-note-android-sdk)
👉 [iOS SDK 릴리스 노트](https://help.airbridge.io/ko/developers/release-note-ios-sdk)
👉 [Web SDK 다중 인스턴스 안내](https://help.airbridge.io/ko/developers/web-sdk)
### **6\. 보안성 심의와 위수탁 점검 증적을 받을 수 있나요?**
보안 통제를 갖췄어도 심의와 감사 과정에서 이를 보여줄 자료가 없다면 금융사의 심의 담당자가 확인하기 어려워요. 데이터 흐름도, 하위수탁자 목록, 계약서와 인증서처럼 통제를 증명할 수 있는 문서가 필요해요. 도입할 때에 더해 이후 정기 검토에서도 최신 자료를 받을 수 있는지 확인해야 해요.
#### **이걸 확인해 보세요.**
- 데이터 흐름도, 하위수탁자 목록, 개인정보 처리 계약(DPA, Data Processing Agreement) 같은 위수탁 증적을 제공하는가
- 인증서와 감사 보고서를 최신 기준일로 제공하는가
- 금융권 위탁 점검에 대응한 자가 점검 자료가 있는가
#### **증적 자료는 언제, 어떤 주기로 갱신되나요?**
증적은 보유 여부와 함께 기준일과 갱신 주기도 확인하세요. 인증 갱신이나 하위수탁자 변경이 생겼을 때 자료를 다시 받을 수 있는지, 그 창구가 누구인지도 계약 전에 정해 두면 돼요.
#### **에어브릿지는 보안성 검토에 필요한 증적을 제공해요**
에어브릿지는 데이터 흐름도, 하위수탁자 목록, DPA 템플릿 등 감사 준비 자료를 사전에 제공해요. 제공 범위는 Security & Privacy 페이지에 안내돼 있어요.
👉 [Airbridge Security & Privacy](https://www.airbridge.io/ko/solutions/security-privacy)
## **좋은 MMP는 보안 통제를 증명할 수 있어야 해요**
금융사가 MMP를 선택할 때는 보안 기능의 보유 여부뿐 아니라 이를 증명할 자료를 지속적으로 제공할 수 있는지도 확인해야 해요. 도입 심의가 끝난 뒤에도 정기 보안 감사와 위수탁 점검은 계속되기 때문이에요.
앞서 정리한 6개 영역의 추가 질문은 RFP나 보안성 검토 질문지에 활용할 수 있어요. 후보 사업자에게 동일한 형식으로 답변과 증적을 요청하면, 각 사업자의 통제 운영 수준과 도입 후 지원 범위를 비교할 수 있어요.
MMP의 마케팅·데이터 운영 관점까지 함께 검토하려면 [금융사 MMP 도입 전 확인해야 할 5가지](https://www.airbridge.io/ko/blog/finance-enterprise-mmp-selection-guide)도 참고해 보세요.
---
## iOS 게임 캠페인 성과가 매체마다 다르게 나오는 이유와 'SKAN' 활용법
URL: https://www.airbridge.io/ko/blog/ios-game-campaign-performance-skan-guide
Language: Korean
Published: 2026-09-01
Author: Team Airbridge
2021년 애플이 ATT(App Tracking Transparency)를 도입하면서 모바일 게임 광고 시장의 성과 측정 방식이 근본적으로 바뀌었어요.
그 변화의 중심에 SKAN(SKAdNetwork)이 있는데요.
iOS 캠페인을 집행하면서 아래와 같은 경험을 한 적 있으신가요?
- 매체 대시보드와 MMP 숫자가 달라서 보고서를 어떤 기준으로 써야 할지 매번 고민해요.
- SKAN 포스트백은 수신되고 있는데, 전환값이 우리 게임에 맞게 설계됐는지 확인한 적이 없어요.
- 소규모 채널 테스트를 돌렸는데 데이터가 나오지 않은 경험이 있어요.
- D1 기준으로 iOS 캠페인 성과를 판단하는데, 유독 판단이 어렵다고 느껴요.
하나라도 해당된다면, 원인이 SKAN 구조에 있을 가능성이 높아요.
이 글에서는 SKAN의 작동 원리부터 게임이 특히 SKAN에서 어려움을 겪는 이유, 수익 모델에 맞는 전환값 설계 전략까지 순서대로 짚어볼게요.
**📌핵심 포인트**
- 애플의 SKAN은 유저 행동 데이터를 모아서 일정 시간이 지난 후 전송해요. D1 기준으로 캠페인 성과를 판단하는 팀이라면, SKAN 데이터가 그 시점에 아직 전송 전일 수 있어요.
- 게임 수익 모델에 맞게 설계하지 않으면 SKAN 데이터로 캠페인 효과에 대한 유의미한 측정이 어려워요. IAP 게임과 IAA 게임은 측정해야 하는 신호 자체가 다르므로, 지금 전환값 설계가 우리 게임의 수익 모델을 반영하고 있는지 점검해 보세요.
- MMP가 SKAN 포스트백 중복을 제거해주면 iOS 캠페인 성과를 하나의 기준으로 볼 수 있어요. ATT에 동의한 유저는 SKAN과 IDFA 기반 어트리뷰션 양쪽에 기여로 집계돼요. MMP가 이 중복을 걸러내기 때문에, 동일한 설치가 두 번 집계되어 성과가 부풀려지는 걸 막을 수 있어요.
## **모바일 게임 광고, 2021년 이전과 이후가 다른 이유**
모바일 게임 광고가 어떻게 작동하는지 이해하려면 먼저 ‘어트리뷰션’을 이해해야 하는데요.
어트리뷰션(Attribution)은 쉽게 설명해 **유저가 어떤 광고를 보고 앱을 설치했는지 추적하는 일련의 과정**이라고 할 수 있어요.
광고주 입장에서는 Meta에 집행한 광고가 실제로 얼마나 많은 설치를 만들어냈는지, Google 광고는 또 얼마나 전환에 기여했는지를 알아야 예산을 효율적으로 배분할 수 있거든요.
2021년 이전까지 이 추적의 중심에는 IDFA(Identifier for Advertisers)가 있었어요. IDFA는 아이폰마다 부여된 광고용 고유 번호로, 유저가 광고를 클릭하고 앱을 설치하면 광고주는 이 번호를 통해 유저가 유입된 광고와 앱 안에서 한 행동 등을 개인 단위로 추적할 수 있었어요.
게임 UA 마케터가 캠페인을 통해 유입된 유저의 D7 리텐션, 평균 결제액 등을 정확히 볼 수 있었던 건 IDFA 덕분이었죠.
### **1\. ATT 도입으로 IDFA 접근이 사실상 막혔어요**
2021년 4월 iOS 14.5가 출시되면서 구조가 바뀌었어요. 애플이 ATT(App Tracking Transparency)라는 정책을 도입한 건데요.
ATT는 유저가 앱을 처음 열 때 ‘이 앱이 다른 회사의 앱과 웹사이트에서 내 활동을 추적하도록 허용하시겠습니까?’라는 팝업을 표시해요. 유저가 직접 허용을 눌러야만 IDFA를 쓸 수 있게 한 정책이죠.
iOS 14.5 출시 직후인 2021년 4~5월, Verizon Media 산하 모바일 분석 플랫폼 [Flurry Analytics](https://www.flurry.com/blog/ios-14-5-opt-in-rate-att-restricted-app-tracking-transparency-worldwide-us-daily-latest-update/)의 데이터에 따르면 ATT 프롬프트를 본 유저 중 추적을 허용한 비율은 미국 기준 약 4%, 글로벌 기준 약 11%에 불과했어요. 팝업을 봤는데도 대부분이 거부를 선택한 거예요.
게임 UA 마케터 입장에서는 **성과를 측정할 수 있는 유저가 갑자기 대폭 줄어든 셈**이죠.
측정이 어려워지면 최적화도 어려워지고, 같은 예산으로 iOS에서 유저를 데려오는 비용도 올라갈 수밖에 없게 되는데요. 애플은 이 공백을 메우기 위해 SKAN(SKAdNetwork)을 내놨어요.
### **2\. 애플이 IDFA 대신 내놓은 측정 방식이 SKAN이에요**
애플은 IDFA를 통한 사용자 추적을 제한하면서 대안을 함께 내놨어요. [SKAdNetwork, 줄여서 SKAN](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)이라고 해요.
IDFA 없이도 **앱 설치로 이어진 광고를 집계 수준에서 측정할 수 있는 애플의 공식 프레임워크**이죠.
**SKAN의** 작동 방식은 기존 어트리뷰션과 완전히 달라요.
기존에는 MMP(Mobile Measurement Partner, 여러 광고 채널의 성과 데이터를 한 곳에서 같은 기준으로 볼 수 있게 해주는 측정 도구)가 IDFA를 기준으로 유저 한 명 한 명을 추적하면서 유저가 유입된 광고를 직접 판단했어요.
SKAN에서는 그 판단을 애플이 해요.
유저가 광고를 보고 앱을 설치하면, **애플 서버가 이 설치는 어떤 광고에서 왔는지 확인하고, 그 결과를 ‘포스트백(postback)’이라는 데이터 패킷으로 광고를 집행한 매체(Meta, Google 등)에 전달**해요.
매체는 이걸 다시 MMP로 보내고, **MMP가 광고주가 읽을 수 있는 성과 데이터로 정리**해줘요.
여기서 MMP가 처음 나오는데, 여러 광고 채널의 성과 데이터를 한 곳에서 같은 기준으로 볼 수 있게 해주는 측정 도구예요. 에어브릿지가 여기에 해당하죠.
SKAN 데이터는 애플이 매체를 통해 MMP로 보내주는 구조이기 때문에, **MMP 없이는 여러 채널의 SKAN 데이터를 한 번에 읽어내기 어려워요.**
[Business of Apps](https://www.businessofapps.com/data/skadnetwork-statistics/)에 따르면, **2024년 기준 전체 iOS 어트리뷰션의 35%가 이미 SKAN으로 측**정되고 있어요. 나머지는 Apple Search Ads, ATT에 동의한 유저의 IDFA 기반 어트리뷰션, 확률적 모델링 등 여러 방식으로 분산돼 있어요.
ATT 이후 새롭게 등장한 측정 수단 중에서는 SKAN이 가장 높은 비중을 차지하면서, 실제로 IDFA를 대체하는 수단으로 자리 잡아 가고 있다는 거예요.
## **SKAN이 게임 데이터 측정을 어렵게 만드는 3가지 이유**
게임의 핵심 측정 리듬과 SKAN의 작동 방식은 여러 지점에서 정면으로 충돌해요.
### **1\. D1 리텐션을 봐야 하는 날, SKAN 포스트백이 아직 수신되지 않았어요**
게임 UA에서 캠페인 퀄리티를 가장 빠르게 판단하는 지표는 **D1 리텐션**이에요. 이는 설치 다음 날 다시 접속한 유저 비율을 뜻해요. 많은 게임 UA팀이 D1 수치를 **캠페인 초기 퀄리티를 확인할 수 있는 가장 빠른 신호 중 하나로 해석**해요.
[GameAnalytics의 2024년 모바일 게임 벤치마크](https://gamedevreports.substack.com/p/gameanalytics-mobile-gaming-benchmarks)에 따르면 iOS 기준 상위 25% 게임의 D1 리텐션은 31~33% 수준인데요.
하지만 SKAN은 이 리듬에 맞게 데이터를 주지 않아요. 유저가 앱을 설치하고 행동을 해도 그 데이터를 바로 전송하지 않기 때문이죠.
SKAN은 설치 이후 유저의 인앱 행동을 측정하는 Activity window를 내부적으로 운영하는데, 이 윈도우가 닫히면 그때 포스트백을 전송하는 시스템이에요.
SKAN 3 기준으로 이 대기 시간이 최소 24시간이고, 여기에 0~24시간의 랜덤 지연이 추가돼요. 포스트백은 빠르면 24시간, 늦으면 48시간 이후에 수신되고요.
즉, 광고주는 **유저가 설치 직후 앱 안에서 어떤 행동을 했는지 실시간으로 확인할 수 없는 거죠.**
Activity window는 유저가 앱 안에서 **이벤트를 발생시킬 때마다 초기화**돼요.
튜토리얼을 완료하거나 레벨을 올리거나 아이템을 구매할 때마다 초기화되며, 특히 게임처럼 설치 초반 24시간 안에 인앱 이벤트가 집중되는 앱일수록, 유저 행동이 활발할수록 포스트백 전송이 더 지연돼요.
바꿔 말하면 캠페인 퀄리티가 높은 유저일수록 포스트백 수신 시점이 늦어져요.
정리하면,
- D1 기준으로 캠페인 성과를 판단해야 하는 시점에 SKAN 포스트백은 아직 수신되지 않음
- 매체 대시보드에서 설치 수는 확인되는데 유입 유저의 퀄리티는 알 수 없음
- MMP 대시보드의 SKAN 기여 수치는 매체 수치와 다름
이로 인해, 3가지 숫자 중 어떤 것을 기준으로 예산을 조정해야 할지 판단하기 어려운 상태가 돼요.
### **2\. 크리에이티브 테스트를 세분화할수록 어트리뷰션 식별이 어려워져요**
게임 UA는 플레이어블 광고, UGC 영상, 게임플레이 하이라이트 등 형식도 다양한데요.
같은 형식 안에서도 후킹 포인트나 엔딩 장면을 바꾸는 크리에이티브 테스트를 상시 운영해요.
그래야 어떤 크리에이티브가 더 높은 퀄리티의 유저를 유입시키는지 확인해야 제작 방향을 설정할 수 있기 때문이죠.
SKAN 3에서는 크리에이티브 단위 어트리뷰션이 구조적으로 불가능했는데요. [애플 WWDC22 공식 영상](https://developer.apple.com/videos/play/wwdc2022/10038/)에 따르면 campaign identifier가 2자리 숫자 필드로 제한됐기 때문이에요.
매체, 캠페인, 광고 세트, 크리에이티브 정보를 100개 조합 안에 모두 담아야 하니 크리에이티브 단위로 성과를 구분하는 건 어려웠어요.
SKAN 4에서는 source-identifier(광고 캠페인 식별자)가 4자리 숫자로 늘어나면서 일부 개선됐어요.
크리에이티브 유형이나 국가 정보를 숫자 조합으로 인코딩할 수 있게 된 거예요.
세분화된 4자리 데이터는 캠페인이 크라우드 익명성(Crowd Anonymity) 티어 2~3을 충족해야 수신할 수 있어요. 티어 조건을 충족하지 못하면 2자리 또는 데이터 없음으로 수신돼요.
크라우드 익명성 티어는 캠페인 ID당 설치 수에 따라 결정돼요. [에어브릿지 SKAN 4 가이드](https://help.airbridge.io/ko/guides/skadnetwork-4)를 참고하면, 앱 설치 수가 많을수록 포스트백 티어가 올라가고, 더 세분화된 데이터를 수신할 수 있어요.
**크리에이티브 테스트를 잘게 쪼갤수록 캠페인당 설치 수가 분산**되기 때문에, 티어 조건을 충족하기 어려워지는 구조라고 볼 수 있어요.
### **3\. 소규모 채널 테스트에서는 conversion value가 수신되지 않아요**
게임 UA팀은 주요 매체 외에 새로운 채널이 우리 게임에 맞는지 확인하기 위해 오퍼월, 리워드 광고, 인플루언서 채널, 신규 DSP 등을 소액으로 테스트하는 경우가 많아요.
그런데 SKAN은 소규모 테스트에 다소 불리한 프레임워크예요.
SKAN에는 Privacy Threshold(개인정보 보호 임계값)라는 구조적 제약이 있는데, 캠페인당 일정 설치 수를 채우지 못하면 conversion value와 source app ID 같은 세부 데이터가 마스킹 되거나 NULL로 수신돼요.
**개인 유저를 식별할 수 없도록 데이터를 집계 처리하는 애플의 개인정보 보호 장치**라고 볼 수 있죠.
테스트 예산이 작을수록 설치 수가 줄고, 설치 수가 Privacy Threshold에 미달하면 conversion value가 NULL로 수신돼요. 채널 퀄리티를 확인하기 위해 필요한 데이터가 구조적으로 수신되지 않는 거예요.
SKAN 4에서 크라우드 익명성 티어 구조가 개선되면서, 이전보다 낮은 설치 볼륨에서도 coarse value(상/중/하) 수준의 최소한의 데이터를 수신할 수 있게 됐어요. 다만 세분화된 데이터를 수신하려면 여전히 일정 수준의 설치 볼륨이 필요해요.
## **게임 수익 모델별 전환값(CV) 설계 전략**
SKAN에서 측정할 수 있는 정보의 양은 처음부터 정해져 있어요.
유저 설치 이후의 행동을 0부터 63까지 숫자 하나로 표현하는 게 전부예요.
이 64개 슬롯에 무엇을 담느냐가 이후 모든 캠페인 최적화의 기반이 되는데, 게임의 수익 모델에 따라 담아야 하는 정보가 달라져요.
### **1\. IAP 중심 게임은 초기 매출 신호를 CV에 인코딩해야 해요**
IAP(In-App Purchase, 인앱 결제) 중심 게임은 유저가 직접 아이템, 캐릭터, 배틀패스 등을 구매하는 방식으로 수익을 내요. UA 캠페인의 목표는 설치 초반에 결제로 전환하는 유저를 유입시키는 거죠.
IAP 게임의 (Conversion Value, 이하 CV) 설계에서 핵심은 첫 결제 여부와 금액 구간을 함께 담는 거예요.
결제 발생 여부만으로는 캠페인 퀄리티를 판단하기에 충분하지 않고, 정확한 금액은 전환값의 비트 제한으로 담을 수 없어요.
이로 인해 금액 구간으로 표현하는 방식을 사용해요. 실제 유저 결제 분포를 기반으로 0원(결제 없음), 1~5달러, 5~20달러, 20달러 이상처럼 구간을 나누고 각 구간을 CV 숫자에 매핑하는 거예요.
SKAN 4에서는 포스트백이 3개로 늘어났어요. 측정 윈도우는 각각 설치 후 0~2일, 3~7일, 8~35일이에요.
**각 윈도우가 닫힌 후 포스트백이 전송되기 때문에, D2 결제 여부는 첫 번째 포스트백에서, D7 시점의 매출 발생 정도(상/중/하)는 두 번째 포스트백의 coarse value로, D35 기준 장기 활동 여부는 세 번째 포스트백의 coarse value로 대략적으로만 확인**할 수 있어요.
코호트 단위의 ARPU(Average Revenue Per User, 유저당 평균 매출)나 ROAS(Return On Ad Spend, 광고 투자 대비 수익률) 추정이 일부 가능해지는 거예요.
CV에 이벤트를 과도하게 등록하면 이벤트가 발생할 때마다 activity window가 초기화되면서 포스트백 전송이 지연돼요.
CV에 담는 이벤트는 첫 번째 측정 윈도우(0~2일) 안에 반드시 확인해야 하는 핵심 신호만으로 제한하는 게 중요해요.
### **2\. IAA 중심 게임은 광고 시청 행동을 비트 플래그로 설계해야 해요**
IAA(In-App Advertising, 인앱 광고) 중심 게임은 유저가 광고를 시청하고 그 수익으로 운영하는 방식이에요. 캐주얼 게임, 하이퍼캐주얼 게임이 여기 해당해요. 단일 결제 없이 유저가 얼마나 자주, 오래 접속하고 광고를 시청하는지가 수익을 결정하죠
IAA 게임의 CV 설계에서 핵심은 광고 시청으로 발생하는 수익 신호를 비트에 담는 거예요. 여기서 비트 플래그란, 행동의 유무를 0 또는 1로 표시하는 방식이에요. 6개의 비트 각각이 하나의 체크 항목이 되는 거죠.
CV는 0부터 63까지 숫자 하나인데, 이 숫자가 내부적으로 6자리 이진수로 이루어져 있어요.
예를 들어 CV가 7이면 이진수로 000111이에요. 각 자리에 행동을 하나씩 할당하고, 유저가 그 행동을 했으면 1, 안 했으면 0으로 표시하는 식이에요.
IAA 게임에서는 광고 시청 횟수나 누적 광고 수익 구간을 단계별로 비트에 할당하는 방식이 유효해요.
예를 들어 비트 0은 광고 1회 이상 시청 여부, 비트 1은 광고 3회 이상 시청 여부, 비트 2는 광고 10회 이상 시청 여부로 설계하면 CV 숫자만 보고도 이 캠페인에서 온 유저들의 광고 수익 기여도를 파악할 수 있어요.
다만 어떤 행동을 몇 번째 비트에 담을지는 게임의 수익 구조와 유저 행동 패턴에 따라 달라지기 때문에, 우리 게임 데이터를 기반으로 설계해야 해요.
첫 번째 측정 윈도우가 설치 후 0~2일이기 때문에, 이 기간 안에 발생 가능한 행동만 CV에 담아야 포스트백에 의미 있는 데이터가 실려요.
D7 리텐션처럼 더 늦게 확인해야 하는 지표는 SKAN 4의 두 번째 측정 윈도우(3~7일)에서 확인하는 구조로 설계해 보세요.
## **데이터 갭을 줄이는 실전 접근: MMP와 퍼스트파티 데이터 조합**
SKAN이 iOS 어트리뷰션의 공식 수단이지만, SKAN 데이터만으로는 게임 UA에 필요한 전체 그림을 그리기 어렵죠. 리타겟팅과 유저 개인 단위 분석도 얻기 어려우니까요.
MMP와 퍼스트파티 데이터를 어떻게 조합하느냐가 실제 운영 차이를 만들기 때문에, 아래 내용을 참고하여 광고 운영에 적용해 보세요.
### **1\. MMP는 SKAN 데이터의 중복을 제거해요**
SKAN 환경에서 MMP의 역할은 기존과 다른데요. 기존에는 MMP가 IDFA를 기준으로 클릭과 설치를 매칭해서 어떤 광고가 기여했는지 확인했어요.
기존에는 MMP가 광고 클릭 시점과 앱 설치 시점의 IDFA를 대조해서, 일치하는 경우 해당 광고에 기여를 부여한 건데요.
앞서 언급했듯 SKAN에서는 어트리뷰션 판단을 애플이 하기 때문에, 이제 **MMP의 역할은 여러 매체에서 들어오는 포스트백을 모아 집계하고, CV 설정값에 따라 해석해서 광고주가 읽을 수 있는 형태로 정리**하는 거예요.
이 과정에서 MMP가 없으면 놓치기 쉬운 문제가 2가지 있어요.
1. 매체마다 포스트백 데이터를 전송하는 타이밍이 달라요. Meta와 Google은 포스트백에 포함된 CV와 발생 개수를 MMP에 전달하는데, 데이터 집계 시점이 달라서 수치 차이가 생길 수 있어요.
2. ATT에 동의한 유저는 SKAN과 IDFA 양쪽에 동시에 잡히기 때문에 중복 집계가 발생해요.
MMP가 이 2가지 문제를 걸러내고 실제 iOS 캠페인 성과를 보여주는 역할을 해요.
👉 [에어브릿지 SKAN 설정 가이드](https://help.airbridge.io/ko/guides/skadnetwork-4-settings)
**에어브릿지는 SKAN과 에어브릿지 어트리뷰션 기준의 중복 기여를 교차 검증 및 식별하는 기능을 제공**하는데요. 이를 통해 어떤 숫자를 기준으로 삼을지 판단할 수 있게 해줘요
SKAN으로 잡힌 설치가 에어브릿지 어트리뷰션으로도 같은 채널에 기여됐는지 확인할 수 있죠.
이 데이터를 통해 3가지 숫자 중 어떤 것을 의사결정 기준으로 삼을지 판단할 수 있어요.
### **2\. SKAN이 못 보는 영역을 퍼스트파티 데이터로 보완해요**
SKAN은 유저 단위 데이터가 없어서 리타겟팅이 불가능하고, 이탈 후 복귀한 유저를 기존 유저로 인식하지 못해요. 고가치 유저와 일반 유저를 개인 단위로 구분하는 것도 어렵죠.
이 공백을 메우는 현실적인 방법은 **게임 서버 데이터와 SKAN 집계 데이터를 코호트 단위로 연결**하는 거예요.
SKAN으로 이 캠페인에서 유입된 유저 집단의 D2 결제율이 몇 %인지 확인할 수 있다면, 게임 서버에서 같은 기간 같은 채널 유입 유저의 상세 행동 로그와 연결해 해석할 수 있어요.
[모비인사이드 2025년 7월 리포트](https://www.mobiinside.co.kr/2025/07/03/att-ios-targeting/)에 따르면 많은 마케터들이 이미 CRM, 퍼스트파티 데이터, 웹 기반 전략으로 SKAN의 공백을 보완하고 있어요.
SKAN이 잘 보여주는 영역과 그렇지 못한 영역을 명확히 구분하고, 각각에 맞는 데이터 소스를 조합하는 게 현실적인 접근이에요.
## **지금 우리 게임의 SKAN 설정, 어디까지 돼 있나요?**
SKAN을 쓰고 있다면 아래 질문으로 현재 설정 상태를 점검해 보세요.
### **Q1. CV에 매핑한 이벤트가 실제로 첫 번째 포스트백 윈도우(48시간) 안에 발생하는 행동인가요?**
48시간 이후에 발생하는 이벤트를 CV에 담아도 첫 번째 포스트백에 실리지 않아요. 실제 유저 행동 로그를 기반으로 설치 후 48시간 안에 발생하는 이벤트만 선별해서 매핑해야 해요.
### **Q2. CV 설계가 우리 게임의 수익 모델(IAP vs IAA)을 반영하고 있나요, 아니면 기본 세팅 그대로인가요?**
기본 세팅은 수익 모델을 반영하지 않아요. IAP 게임이라면 결제 금액 구간을, IAA 게임이라면 리텐션과 광고 시청 행동을 CV에 담아야 캠페인 최적화에 쓸 수 있는 데이터가 나와요.
### **Q3. 소규모 채널 테스트 예산이 Privacy Threshold를 넘기기에 충분한 수준인가요?**
캠페인 ID당 하루 설치 수가 일정 수준 이하면 conversion value가 NULL로 수신돼요. 새 채널을 테스트할 때는 여러 채널에 예산을 나누지 말고, 한 채널에 집중해서 먼저 Threshold를 충족하는 게 맞아요.
### **Q4. SKAN 데이터와 IDFA 기반 어트리뷰션 데이터의 중복 기여를 제거하고 보고 있나요?**
ATT에 동의한 유저는 SKAN과 IDFA 양쪽에 동시에 잡혀요. 중복을 제거하지 않으면 같은 설치가 두 번 집계돼서 실제보다 성과가 부풀려져 보여요. MMP에서 중복 기여 제거 기능을 활성화했는지 확인해야 해요.
### **Q5. MMP 대시보드에서 SKAN 포스트백이 정상적으로 수신되고 있는지 주기적으로 확인하고 있나요?**
기거나 iOS 버전 이슈로 포스트백이 수신되지 않는 경우가 있어요. 포스트백 수신 현황을 정기적으로 확인하지 않으면 데이터 공백이 생긴 걸 뒤늦게 발견하게 돼요.
## **SKAN 캠페인, 설계 이후 운영이 더 중요해요**
CV(Conversion Value, 전환값) 설계를 잘해도 운영 단계에서 놓치는 부분이 있으면 의사결정에 쓸 수 없는 데이터가 나와요. 아래는 설계 이후 실제 캠페인을 돌릴 때 확인해야 할 포인트예요.
### **1\. CV 변경은 캠페인 비수기에 맞춰서 해야 해요**
CV 설정을 변경하면 약 48시간 동안 conversion value 측정이 중단돼요. 이 기간에 유입된 유저의 인앱 행동 데이터는 잡히지 않아요.
시즌 이벤트나 대규모 캠페인이 집행되는 시점에 CV를 바꾸면 그 기간 데이터가 통째로 빠지는 거예요. CV 변경은 반드시 캠페인 집행 공백기나 저예산 기간에 맞춰 계획적으로 해야 해요.
### **2\. 소규모 채널 테스트는 예산을 묶어서 집행해야 해요**
Privacy Threshold를 넘으려면 캠페인 ID당 하루 일정 수준의 설치가 필요해요. 테스트 채널을 여러 개 동시에 소액으로 돌리면 각 채널의 설치 수가 분산돼서 어느 채널도 Threshold를 넘지 못하는 상황이 생겨요.
새 채널을 테스트할 때는 채널 수를 줄이고 예산을 한 곳에 집중해서 먼저 데이터를 확보한 뒤, 다음 채널로 넘어가는 순차적 접근이 현실적이에요.
### **3\. 포스트백 수신 여부를 주기적으로 확인해야 해요**
캠페인을 집행하고 있어도 MMP에서 SKAN 포스트백이 실제로 수신되고 있는지 주기적으로 확인해야 해요. 광고 채널과 MMP 간 SKAN 연동이 끊기거나, iOS 버전 이슈로 SKAN 4가 아닌 구버전으로 데이터가 잡히는 경우가 있거든요.
포스트백 수신 현황을 MMP 대시보드에서 정기적으로 체크하는 게 필요해요.
## **게임 마케터가 SKAN에 대해 가장 많이 묻는 질문들**
### **Q1. SKAN 데이터와 매체 대시보드 숫자가 다른 건 정상인가요?**
네, 정상이에요. SKAN은 애플이 어트리뷰션을 판단하고, 매체 대시보드는 매체 자체 기준으로 성과를 집계해요. 판단 주체와 기준이 다르기 때문에 숫자가 일치하지 않아요.
중요한 건 어느 숫자가 맞는지 찾는 게 아니라, 어떤 숫자를 의사결정 기준으로 쓸지 팀 내에서 합의하는 거예요. MMP가 이 두 데이터를 나란히 비교해서 볼 수 있게 해줘요.
### **Q2. SKAN 4로 업그레이드해야 하나요, SKAN 3도 충분한가요?**
게임이라면 SKAN 4로 가는 게 유리해요. SKAN 3는 포스트백이 하나뿐이고 24시간 윈도우 데이터만 받을 수 있어요.
SKAN 4는 포스트백 3개로 D2, D7, D35 단위 데이터를 분리해서 볼 수 있고, source-identifier가 최대 4자리로 확장돼 크리에이티브 단위까지 구분해 성과를 분석할 수 있는 환경을 제공해요. 다만, 디바이스의 iOS 버전이 16.1 이상이어야 SKAN 4가 작동해요.
다만 디바이스의 iOS 버전이 16.1 이상이어야 SKAN 4가 작동해요.
### **Q3. CV를 너무 자주 바꾸면 어떤 문제가 생기나요?**
CV 설정을 변경하면 약 48시간 동안 conversion value 측정이 중단돼요. 이 기간에 들어온 유저의 인앱 행동 데이터는 잡히지 않아요.
캠페인이 활발하게 집행되는 시점에 CV를 변경하면 데이터 공백이 생기기 때문에, 변경 타이밍을 캠페인 일정과 맞춰서 계획적으로 해야 해요.
### **Q4. 게임 장르별로 CV 설계 방식이 다른가요?**
장르보다는 수익 모델에 따라 달라져요. IAP 중심 게임은 첫 결제 여부와 금액 구간을 CV에 담고, IAA 중심 게임은 튜토리얼 완료, 레벨 도달, 광고 시청 같은 행동 시퀀스를 비트 플래그로 설계해요.
같은 캐주얼 게임이라도 IAP와 IAA를 함께 쓰는 하이브리드 모델이라면 두 방식을 조합해야 해요. 어떤 행동이 수익과 직결되는지를 먼저 정의하고, 그 신호를 64개 슬롯 안에 담는 순서로 접근하는 게 맞아요.
### **Q5. SKAN 데이터만으로 ROAS를 계산할 수 있나요?**
정확한 ROAS 계산은 어려워요. SKAN은 유저 단위 매출 데이터가 없고, CV에 담긴 금액 구간은 범위값이라 정확한 수치가 아니에요.
다만 SKAN 4의 포스트백 3개를 활용하면 D2, D7, D35 기준 코호트 단위 ARPU를 추정하고, 이를 기반으로 대략적인 ROAS 범위를 계산하는 건 가능해요. 정확한 수치보다는 캠페인 간 상대적 비교 지표로 활용하는 게 현실적이에요.
## **에어브릿지로 CV 설계부터 포스트백 해석까지, iOS 캠페인 측정 공백을 채우세요**
매체 대시보드, MMP, SKAN, 모두 같은 캠페인에서 숫자가 다르게 나오는 이유는 각자 다른 방식으로 성과가 집계되기 때문이에요.
그 안에서 iOS 캠페인을 제대로 운영하려면 CV 설계가 먼저예요. 우리 게임에서 유저의 핵심 행동을 64개 슬롯에 얼마나 잘 담느냐에 따라, 포스트백으로 받은 데이터로 실제 캠페인 효과를 판단할 수 있으니까요.
우리 게임에 SKAN이 어떻게 적용될 수 있는지, 지금 바로 데모에서 먼저 확인해보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 카페24 브랜드앱 광고 성과, 이제 웹과 앱을 하나로 측정할 수 있어요
URL: https://www.airbridge.io/ko/blog/airbridge-cafe24-brandapp-integration
Language: Korean
Published: 2026-08-27
Author: Jaehyuk Kim
안다르, 젝시믹스, 육육걸즈 등 브랜드 자사몰에서는 앱을 어떻게 만들고 운영하고 있을까요? 이런 브랜드몰들은 웹 쇼핑몰의 회원이 유지될 수 있도록 네이티브 앱이 아닌 웹빌더앱을 주로 이용하고 있어요. 대표적으로는 카페24의 브랜드앱, 나쵸코드, NHN커머스의 Shopby 등이 있죠.
하지만, 웹빌더앱은 웹뷰 기반의 하이브리드 앱이다보니 앱 관련된 이벤트와 성과만을 별도로 보는게 어려워요. 그러다보니 자사몰 전체 매출에서도 앱의 매출 기여도를 명확하게 볼 수 없었어요.
이번에 에어브릿지가 국내 주요 웹빌더앱과 연동을 시작하면서, **카페24 브랜드앱·나쵸코드(Nachocode)·NHN커머스 샵바이(Shopby) 마이앱에서도 웹과 앱을 분리하여 광고 성과를 측정**할 수 있게 됐어요.
**📌핵심 포인트**
- 카페24 브랜드앱, 나쵸코드, shopby 마이앱에 에어브릿지 앱 SDK 설치가 지원돼요.
- 매출 기여도를 광고 채널별로, 웹/앱별로 나눠서 분석할 수 있어요.
- [포스트백](https://help.airbridge.io/ko/guides/postback-settings-new)으로 구매·회원가입 등 앱 전환 데이터를 광고채널에 전달하여 앱 광고 성과를 고도화할 수 있어요.
## **커머스 브랜드가 앱을 만드는 이유**
자사몰을 운영하는 커머스 브랜드에게 앱은 웹과 다른 무대예요. 웹이 프로모션·이벤트 등으로 유입되는 신규 고객이 지나가는 통로라면, **앱은 재방문·푸시·개인화가 작동하는 리텐션의 무대**죠. D2C 브랜드가 커지면 자사몰 앱 확보가 다음 스텝으로 자연스럽게 이어지는 이유예요.
그런데 네이티브 앱을 직접 개발하려면 iOS·Android를 각각 만들고, 자사몰 웹과 별개로 상품·주문·회원 시스템을 앱에 다시 붙여야 해요. 그만큼 개발 팀에 부담이 크죠.
이 부담을 없애기 위해 등장한 게 **하이브리드 앱 빌더**예요. 운영 중이던 웹기반 몰을 그대로 앱 안에 담고 푸시·홈 화면 아이콘 같은 앱 전용 UX만 얹으면 별도 개발 없이도 앱 스토어에 출시할 수 있어요.
카페24 브랜드앱은 쇼핑몰 운영자가 신청하면 **카페24가 웹 기반 하이브리드 앱을 제작해 구글 플레이·애플 앱스토어에 등록·출시해 주는** 서비스예요. 나쵸코드와 NHN커머스의 Shopby가 같은 방식으로 자사몰 앱 운영을 지원하고 있어요.
## **웹빌더앱의 광고 성과 측정, 왜 어려웠나**
웹빌더앱이 커머스에 자리 잡으면서 마케팅 팀에게는 **광고 성과 측정이라는 새로운 숙제**가 생겼어요.
기존 웹 쇼핑몰은 클릭을 통한 이동이기에 어떤 광고로 어떤 웹페이지로 갔는지 명확히 구분할 수 있었어요. 하지만 앱은 스토어를 경유하게 되면서 단절이 발생하는데요. 이로 인해, 앱 설치까지는 트래킹할 수 있어도 광고 채널별 리텐션이나 LTV, 퍼널 도달률 등을 파악하기는 어려웠어요.
구체적으로 실무에서 자주 마주치는 상황은 크게 3가지예요.
### **1\. 웹과 앱 성과가 분리돼서 자사몰 전체 매출 기여도가 안 보여요**
같은 브랜드가 운영하는 웹과 앱에서 매출이 발생하더라도 성과 측정 도구가 따로 작동하면, 하나의 광고가 자사몰 전체 매출에 얼마나 기여했는지 정확하게 판단하기 어려워요. 웹 분석 도구와 앱 분석 도구, 카페24 브랜드앱 대시보드의 데이터를 담당자가 일일이 스프레드시트에 옮겨 합치는 작업도 반복되고요.
### **2\. 광고 클릭 후 앱 상품 상세페이지 진입이 매끄럽지 않아요**
메타나 구글의 상품 광고를 클릭한 고객이 앱 안의 해당 상품 상세 페이지로 바로 이동하면 구매까지의 경로가 짧아져요. 하지만 딥링크가 제대로 설정되지 않으면 앱 홈으로 이동하거나 앱 대신 브라우저가 열릴 수 있어요. 고객이 광고에서 본 상품을 다시 찾아야 하므로 이 과정에서 이탈할 가능성도 커져요.
### **3\. 광고 채널이 앱 안의 구매·회원가입 신호를 받지 못해요**
메타와 구글은 실제 전환 신호를 학습해 입찰과 광고 노출 대상을 자동으로 조정해요. 앱 안에서 발생한 구매나 회원가입 이벤트가 광고 채널에 전달되지 않으면, 어떤 광고가 실제 전환을 만들었는지 판단할 학습 데이터가 부족해져요. 그러면 광고 채널이 구매보다 클릭이나 설치에 가까운 신호를 중심으로 학습하면서 앱 랜딩 광고의 ROAS를 개선하기 어려워질 수 있어요.
## **카페24 브랜드앱을 비롯한 웹빌더 앱, 에어브릿지 연동으로 3가지가 달라져요**
카페24 브랜드앱처럼 웹사이트를 기반으로 만든 앱은 웹과 앱에서 발생한 고객 행동을 함께 측정해야 광고 성과를 온전히 파악할 수 있어요. 에어브릿지 연동을 지원하는 웹빌더 기반 앱이라면 웹과 앱의 매출 기여도를 연결하고, 딥링크와 포스트백을 활용해 앱 광고 운영에 필요한 데이터를 확보할 수 있어요.
### **1\. 웹과 앱의 매출 기여도를 광고 채널별로 통합 분석해요**
자사몰 웹에서 발생한 매출과 웹빌더 앱에서 발생한 매출을 광고 채널별로 연결해 분석할 수 있어요.
예를 들어 메타 광고를 클릭한 고객이 웹사이트를 방문한 뒤 앱으로 이동해 구매했다면, 웹투앱 여정을 연결해 해당 구매가 어느 광고에서 시작됐는지 확인할 수 있어요.
이전에는 웹과 앱의 성과를 따로 보고 채널별 ROAS를 판단했다면, 연동 후에는 웹과 앱을 아우르는 **자사몰 전체 매출 기여도**를 기준으로 광고 예산을 조정할 수 있어요.
### **2\. 딥링크로 광고에서 상품 상세 페이지까지 매끄럽게 연결해요**
에어브릿지 딥링크를 광고에 적용하면 고객을 앱 홈이 아닌 광고에 노출된 상품의 상세 페이지로 보낼 수 있어요. 앱이 설치된 고객은 광고를 클릭하자마자 해당 상품 페이지로 이동해요.
앱이 설치되지 않은 고객은 앱 마켓에서 앱을 설치한 뒤 첫 실행 시 해당 상품 페이지로 이동할 수 있어요. 이렇게 앱 설치 이후 지정한 화면을 여는 방식을 디퍼드 딥링크(Deferred Deep Linking)라고 해요.
딥링크를 활용하면 고객이 광고에서 본 상품을 앱에서 다시 검색하는 단계를 줄일 수 있어요. 광고 클릭부터 상품 탐색까지의 경로가 짧아지므로 구매 과정에서 발생하는 이탈 가능성도 낮출 수 있고요.
### **3\. 포스트백으로 앱 전환 데이터를 광고 채널에 전달해요**
에어브릿지 포스트백은 앱 안에서 발생한 구매, 회원가입, 장바구니 담기, 상품 조회 등의 이벤트를 연동된 광고 채널로 전달해요. 전송할 이벤트와 조건은 채널별로 설정할 수 있어요.
광고 채널은 전달받은 전환 데이터를 캠페인 최적화와 리타겟팅 등에 활용해요. 앱 설치 이후의 구매나 회원가입 신호까지 전달하면, 클릭이나 설치뿐 아니라 실제 전환에 가까운 행동을 바탕으로 자동화 입찰을 운영할 수 있어요.
지원되는 광고 채널과 캠페인에서는 구매 금액이나 고객 가치에 기반한 ROAS·LTV 최적화 기능도 활용할 수 있어요. 이를 통해 웹빌더 앱도 설치 수를 늘리는 데서 그치지 않고, 구매 가능성이 높은 고객을 확보하는 방향으로 앱 광고를 운영할 수 있어요.
## **브랜드앱 광고 성과가 보이면, 리텐션과 LTV가 달라져요**
쇼핑몰의 앱에서의 광고 성과 측정이 가능해지면 커머스 마케팅의 무게중심도 옮겨가요.
먼저 **신규 획득 광고의 정확도**가 달라져요. 자사몰 웹과 앱 매출을 통합해서 보고 딥링크·포스트백이 붙으면 채널별 실제 기여도를 근거로 예산을 조정할 수 있어요. 감으로 하던 채널 믹스를 데이터로 재구성하게 되죠.
그리고 **리텐션과 LTV까지 확장**돼요. 광고 클릭에서 앱 상품 상세로, 다시 구매까지 이어진 유저는 앱이 설치된 상태로 남는데요. 웹과 달리 앱에서는 푸시 알림·홈 화면 아이콘·개인화 큐레이션이 상시로 작동해요. 첫 구매 유저에게 재구매 알림을 보내고, 관심 카테고리에 맞춰 상품을 추천하고, 프로모션을 앱에서 먼저 열어주는 방식으로 **LTV가 확장**돼요.
에어브릿지가 제공하는 리텐션·매출·퍼널 리포트와 코호트 분석을 활용하면 "어떤 광고 채널에서 온 유저가 앱에서 가장 오래 남는가", "어떤 상품 딥링크가 재구매로 연결되는가"를 데이터로 볼 수 있어요. 광고 성과가 앱 다운로드에서 끝나지 않고, **앱 안에서 반복되는 구매까지 이어지는 신호로 확장**되는 거예요.
여기에 더해, 에어브릿지 파일럿(Airbridge Pilot)과 MCP 등 에어브릿지 AI(Airbridge AI) 기능들을 활용하면 SDK 연동 상태 점검, 채널별 성과 조회, 딥링크 생성, 리텐션 데이터 분석까지 자연어로 요청하고 결과를 받을 수 있어요.
## **카페24 브랜드앱을 쓰는 커머스라면 Airbridge로 성과 측정을 시작해보세요**
**이번 연동은 카페24 외에도 나쵸코드, NHN 커머스 샵바이 브랜드앱에도 함께 지원돼요.** 이제 브랜드앱 광고 성과를 웹과 통합해서 정밀하게 측정할 수 있을 뿐만 아니라, 자사몰 전체 매출 기여도를 광고 채널별로 보고, 딥링크로 상품 상세 진입 경험을 매끄럽게 만들고, 포스트백으로 자동화 입찰을 최적화할 수 있어요.
이미 카페24 브랜드앱, 나쵸코드, NHN커머스 샵바이 마이앱을 운영 중이라면, 처음부터 성과 측정 구조를 함께 설계하는걸 추천드려요. 훨씬 적은 리소스로 에어브릿지를 도입할 수 있어요.
만약 혼자 결정하기 어렵다면 에어브릿지 팀과 함께 이야기 나눠보시는 걸 추천드려요. 에어브릿지 팀에서 고도화된 앱 마케팅 및 성과 측정이 가능할 수 있게 도와드릴게요.
혹시 더 자세한 상담이 필요하다면 에어브릿지 도입문의 페이지에서 상담을 요청해 주세요.
---
## 엔진부터 스팀 출시까지, 게임사만 아는 MMP 고민 정리했어요
URL: https://www.airbridge.io/ko/blog/game-mmp-selection-guide
Language: Korean
Published: 2026-08-27
Author: Team Airbridge
게임 마케팅이나 운영을 담당하고 계시다면, 아마 이런 점이 궁금하셨을 거예요.
- **우리 게임에서 쓰는 엔진, MMP가 제대로 지원하는지 확인이 될까?**
- **인앱 결제랑 광고 수익을 따로 떼서 ROAS(Return On Ad Spend, 광고비 대비 매출)를 볼 수 있을까?**
- **스팀(Steam)으로 출시하면 어트리뷰션은 어떻게 달라질까?**
이는 MMP(Mobile Measurement Partner, 모바일 앱의 유입 채널과 성과를 측정하는 솔루션)를 이미 쓰고 있어도, 이제 막 입문하는 단계여도 마찬가지일 텐데요.
이유는 게임 앱마다 다른 수익 모델과 배포 환경, 개발 엔진까지 한 번에 정리된 자료를 찾기 쉽지 않기 때문이에요..
그래서 이번 글에서는 게임사의 입장에서 우리 게임에 맞는 MMP를 어떤 기준으로 골라야 하는지 정리했어요. MMP 도입을 검토 중이라면 아래 가이드를 참고해 보세요.
**📌핵심 포인트**
- 게임사가 MMP를 검토할 때는 먼저 우리 게임의 엔진·플랫폼 지원 여부를 확인하고, 이어서 IAP·IAA가 섞인 매출을 나눠서 캠페인 성과를 정확히 판단할 수 있는지도 살펴봐야 해요.
- 다른 MMP로 전환했을 때 리텐션 수치가 다르게 나온다면, 데이터가 잘못된 게 아니라 어트리뷰션 윈도우 설정값이 다르기 때문이에요.
- 영수증 검증 비용, ROAS 측정 기간의 유연성처럼 계약 전에는 잘 드러나지 않는 항목이 도입 후 비용 부담으로 이어질 수 있어요. 이 글의 체크리스트로 미리 확인해두면 효율적인 예산 사용이 가능해요.
## **게임사 위한 MMP, 무엇이 달라야 할까?**
MMP의 기본 역할은 게임사와 커머스 앱 모두 같아요. 유입 채널별로 유저 행동을 추적하는 거죠. 다만 게임사는 무엇을 추적 대상으로 삼아야 하는가에 대해선 커머스 앱과 다른 기준이 필요해요.
### **1\. 혼합 수익 구조: IAP와 IAA를 나눠서 봐야 해요**
게임 앱은 한 유저에게서 크게 두 종류의 수익이 동시에 발생해요.
| 구분 | 발생 조건 | 특징 |
| --- | --- | --- |
| IAP (In-App Purchase, 인앱 결제) | 유저가 직접 결제 | 결제 유저 비율은 낮지만 비교적 높은 1인당 수익 |
| IAA (In-App Advertising, 인앱 광고) | 유저가 광고를 시청 | 결제 없이도 발생하며, 유저 수가 많을수록 누적 |
이 두 수익을 합산한 값 하나로 ROAS(Return On Ad Spend, 광고비 대비 매출)를 계산하면, 캠페인별로 결제 유저가 많이 유입됐는지, 광고 시청이 활발한 유저가 많이 유입됐는지 구분할 수 없죠.
캠페인 성과를 정확히 비교하려면 **IAP ROAS와 IAA ROAS를 나눠서 보고, 필요에 따라 합산된 총 ROAS도 별도로 확인**하는 방식이 필요해요.
### **2\. 플랫폼: PC·콘솔 확장 시 어트리뷰션 구조가 달라져요**
게임은 모바일로 시작한 뒤 흥행 여부에 따라 PC나 콘솔로 플랫폼을 넓히기도 하죠.
플랫폼이 늘어나면 어트리뷰션(Attribution, 유저의 유입 경로를 추적해 특정 채널·캠페인의 성과로 귀속하는 것) 방식도 함께 달라지는데요. 대략적으로 아래와 같아요.
| 구분 | 모바일(앱스토어·플레이스토어) | PC(스팀 등) |
| --- | --- | --- |
| 유저 식별 방식 | 광고 식별자(IDFA, GAID) 기반 | 광고 식별자 없음 |
| 어트리뷰션 방식 | 클릭·설치 기반 자동 매칭 | S2S(Server-to-Server) 연동 등 별도 방식 필요 |
| 기존 모바일 설정 재사용 | 가능 | 불가능, 별도 설정 필요 |
즉, 모바일에서 쓰던 어트리뷰션 설정을 PC 출시작에 그대로 옮길 수 없다는 뜻이에요.
이러한 차이가 생기는 이유는 운영 체제 구조에서 찾아볼 수 있어요.
[모바일은 IDFA·GAID로 클릭·설치 자동 매칭, PC·Steam은 광고 식별자가 없어 S2S 연동으로 서버에서 이벤트를 직접 전송]
iOS와 안드로이드는 OS 차원에서 기기마다 IDFA·GAID라는 광고 식별자를 부여하고, MMP는 이 식별자를 기준으로 특정 광고를 클릭한 기기가 언제 앱을 설치 했는지 자동으로 매칭해요.
반면 스팀 같은 PC 플랫폼은 이런 광고 식별자 체계 자체가 없어서, 클릭과 설치를 자동으로 이어 붙이는 매칭이 되지 않는 거죠.
이 차이로 인해 클릭과 설치의 자동 매칭이 안 되니 설치·매출 등 주요 이벤트를 서버에서 직접 전송하는 S2S 연동을 새로 구성하게 되는 거예요.
또한 S2S 연동은 MMP마다 지원 범위와 설정 방식이 달라서, 모바일에서 쓰던 설정을 그대로 옮기는 게 아니라 PC용으로 별도 세팅해야 하고요.
모바일에서 흥행한 게임이 PC로 플랫폼을 확장하는 경우가 늘고 있는 만큼, **지금 쓰는 MMP가 이 S2S 연동을 지원하는지는 플랫폼 확장 계획 유무와 상관없이 미리 확인**해두는 게 안전해요.
### **3\. SDK 지원 여부를 사전에 확인해야 해요**
엔진 지원 여부도 미리 확인해야 할 항목이에요.
MMP는 게임 엔진이 제공하는 개발 환경 안에 SDK(Software Development Kit, 소프트웨어 개발 키트)를 심어서 유저 행동 데이터를 수집해요.
게임 개발에 가장 널리 쓰이는 Unity, Unreal은 주요 MMP들이 공식 SDK로 지원해요.
다만 그 외 엔진을 쓰고 있다면 이야기가 달라질 수 있어요.
MMP가 해당 엔진용 SDK를 별도로 만들어두지 않았다면, 그 엔진으로 만든 게임에서는 데이터 수집 자체가 이루어지지 않아요. 우리 게임이 Unity·Unreal 외의 엔진으로 개발됐다면, MMP 도입 전에 SDK 지원 여부를 반드시 확인해야 해요.
## **사용 중인 MMP, 왜 숫자가 안 맞을까요?**
MMP를 이미 쓰고 있는 게임사라면, 같은 캠페인의 리텐션 수치가 MMP마다 다르게 나온 적이 있을 거예요.
특히 다른 MMP와 병행하거나 전환하는 시점, 새로운 유입 채널을 연동하는 시점에 혼란이 있을 수 있죠. 이때는 아래와 같은 설정 방식을 확인해 보세요.
### **1\. 리텐션 윈도우와 룩백 윈도우부터 살펴보세요**
리텐션(Retention, 특정 시점 이후 유저가 앱에 재접속하는 비율)을 측정할 때는 두 가지 기준값이 함께 작동해요.
- **리텐션 윈도우(Retention Window)**: 설치 후 며칠째 재접속을 리텐션으로 인정할 것인가? (D1, D7, D30 등)
- **룩백 윈도우(Lookback Window)**: 광고 클릭 또는 노출 후 며칠 안에 발생한 설치까지를 해당 캠페인 성과로 인정할 것인가?
이 두 값의 기본 설정은 MMP마다 다르게 되어 있으며, 이 차이는 서로 다른 방식으로 수치에 영향을 줘요.
[룩백 윈도우 기간 설정 비교 - 짧은 MMP는 오가닉으로 분류, 긴 MMP는 캠페인 유입으로 귀속]
**먼저, 룩백 윈도우가 다르면 ‘캠페인의 유저’로 잡히는 모수가 달라지는데요.** 같은 유저 데이터를 두 MMP에 동시에 연동해도 룩백 윈도우가 짧은 MMP는 특정 설치를 캠페인 성과로 인정하지 않고 오가닉으로 분류해요.
반면, 룩백 윈도우가 긴 MMP는 같은 설치를 캠페인 유입으로 귀속시키고요. 캠페인 유저로 잡히는 모수 자체가 다르니, 그 위에서 계산되는 리텐션 수치도 자연히 달라져요.
[리텐션 윈도우 정의 비교 - 정확히 D7 접속만 카운트 vs D1~D7 사이 한 번이라도 접속 시 카운트]
**또한 리텐션 윈도우가 다르면, 같은 모수 안에서도 접속 로그를 세는 기준이 달라져요.** 예를 들어 D7 리텐션을 ‘설치 후 정확히 7일째 접속’으로 계산하는 MMP와 ‘설치 후 1~7일 사이 한 번이라도 재접속’으로 계산하는 MMP는 같은 유저의 같은 접속 로그를 두고도 다른 숫자를 내놓을 수 있어요.
MMP를 전환하거나 병행하는 시점이라면, 수치 자체를 비교하기 전에 양쪽의 룩백 윈도우와 리텐션 윈도우 설정값부터 먼저 맞춰 봐야 해요.
### **2\. 채널별로 지원되는 이벤트 범위를 점검하세요**
리텐션 윈도우 차이가 MMP 자체의 계산 방식에서 비롯된다면, 이번엔 MMP가 아니라 유입 채널 쪽에서 비롯되는 차이가 있어요.
포스트백(Postback, 클릭·설치 등 유저 행동 데이터를 매체사에 전달하는 것)은 광고주가 설정한 이벤트라면 원하는 대로 채널에 전달할 수 있어요.
다만 **채널마다 포스트백 안에 담을 수 있는 파라미터 범위가 달라요.**
예를 들어 같은 광고 노출 이벤트를 보내더라도, 어떤 채널은 통화(currency) 파라미터까지 받아서 노출당 수익을 정확히 계산할 수 있지만, 어떤 채널은 이 파라미터를 지원하지 않아 해당 정보가 빠질 수 있어요.
이런 채널에서는 마케터가 특정 이벤트의 데이터가 왜 반영되지 않는지 파악하지 못해 혼란을 겪을 수 있어요.
새로운 유입 채널을 MMP에 연동할 계획이라면, 그 채널이 어떤 이벤트까지 포스트백을 지원하는지 사전 확인이 필요해요.
## **게임사에 맞는 MMP 기준, 한 번에 정리하기**
지금까지 게임사가 MMP를 검토할 때 확인해야 할 지점들을 살펴봤어요.
정리하면 아래와 같은 기준이 필요했는데요.
| 평가 기준 | 확인할 내용 |
| --- | --- |
| 연동 범위 | 우리 게임의 엔진·플랫폼(모바일/PC)을 지원하는가? |
| 게임 특화 측정 | IAP·IAA를 분리해 ROAS를 볼 수 있는가? |
| 데이터 정확도 | 룩백 윈도우·어트리뷰션 윈도우 등 설정값을 직접 조정할 수 있는가? |
여기에 하나 더, 게임 앱 특성상 놓치기 쉬운 기준이 있어요. 바로 **가격 모델**이에요.
게임 앱은 매출 대부분이 결제에서 나오다 보니, **결제 데이터를 검증하는 기능 하나만으로도 실제 이용 비용이 크게 달라질 수 있거든요.** 매출과 관련된 두 지점을 먼저 확인하세요.
### **1\. 가격 모델: 영수증 검증 비용을 점검하세요**
게임 앱은 매출 대부분이 인앱결제(IAP)에서 나오다 보니, 결제 데이터의 신뢰도를 확보하는 절차가 특히 중요해요.
이를 결제 검증(Purchase Verification, 영수증 검증) 이라고 하는데, 유저 기기가 보낸 결제 정보를 그대로 믿지 않고 서버가 App Store·Google Play API에 직접 조회해 진위를 대조하는 기능이에요. 이 과정이 없으면 위·변조나 환불된 결제가 그대로 매출로 집계되고, 그 데이터 위에서 ROAS와 LTV가 계산돼요.
확인해야 할 지점은 과금 방식이에요. 이 기능은 MMP마다 제공 방식이 달라서, 기본 요금에 포함된 곳도 있고 별도 애드온이나 상위 플랜 전용으로 분리해 둔 곳도 있어요.
특히 환불 반영이나 수수료·세금 차감 후 순매출 계산은 상위 플랜에서만 지원하는 경우가 흔한데, 게임사에게는 이쪽이 사실상 필수예요. 그래서 기본 요금표만 비교하면 실제 청구액과 차이가 생기죠.
**📍견적 단계에서 확인할 것**
- 기본 플랜 포함인지, 별도 애드온인지
- 애드온이면 과금 기준(정액 / 결제 건수 연동)
- 환불·순매출 반영이 기본 기능인지
- 예상 결제 건수 기준 연간 총액
### **2\. ROAS 코호트 기간을 직접 설정할 수 있어야 해요**
게임 유저는 결제 없이 오래 머물다가 특정 시점(신규 콘텐츠 업데이트, 이벤트 등)에 결제로 전환되는 경우가 흔해서, 매출이 짧은 기간 안에 다 잡히지 않아요.
ROAS 측정 기간을 짧게 고정해둔 MMP를 쓰면, 실제로는 장기적으로 수익을 내는 캠페인인데도 초반 ROAS만 보고 저성과로 오해할 수 있어요.
**MMP를 고를 때는 ROAS 측정 기간을 게임사가 원하는 기간(30일, 90일 등)으로 코호트를 직접 설정**할 수 있는지를 확인해야 해요.
### **3\. 게임사 MMP 도입 전 체크리스트를 활용해 보세요**
오늘 알아본 내용을 종합해서 게임사 마케터가 MMP를 검토할 때 실제로 고려해야 하는 항목을 체크리스트로 정리했어요.
MMP 도입이나 새로운 MMP 전환을 고려하고 있다면 참고해 보세요.
| 우리 게임의 엔진(Unity/Unreal 외 포함)과 배포 플랫폼(모바일/PC)을 지원하는가? IAP·IAA를 분리해서 ROAS를 볼 수 있는가? 룩백 윈도우·어트리뷰션 윈도우, 리텐션 윈도우를 직접 설정할 수 있는가? 연동하려는 유입 채널이 게임 특화 커스텀 이벤트까지 포스트백을 지원하는가? 영수증 검증이 기본 기능에 포함되어 있는가, 별도 비용이 발생하는가? ROAS 측정 기간을 게임사가 원하는 기간으로 조정할 수 있는가? 부정 설치·클릭 스팸 등 fraud를 걸러내는 기능을 제공하는가? |
| --- |
## **게임사를 위한 MMP, 에어브릿지로 시작하세요**
인트로에서던진 질문 3가지로 돌아가 볼까요? 우리 게임 엔진에서 MMP가 작동하는지, 인앱 결제와 광고 수익을 나눠서 ROAS를 볼 수 있는지, 스팀으로 출시하면 어트리뷰션이 어떻게 달라지는지. 이제 세 질문 모두에 답할 수 있어요.
대부분의 MMP는 커머스 앱을 기준으로 설계돼서 게임사 상황을 예외로 다뤄왔어요. **에어브릿지(Airbridge)는 이 지점들을 아래와 같이 기본값**으로 다뤄요.
- **엔진**: Unity·Unreal을 공식 SDK로 지원해요
- **수익 구조**: IAP·IAA를 분리해서 캠페인별 실제 성과를 볼 수 있어요
- **플랫폼 확장**: PC/스팀 출시 시에도 S2S 연동으로 매출 데이터를 놓치지 않아요
- **비용 구조**: 영수증 검증을 기본 기능으로 제공해서 추가 비용 부담을 줄여요
MMP를 검토할 때 확인하는 항목들의 자세한 지원 방법이 궁금하다면, 지금 바로 데모에서 직접 확인해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 에이블리 디스플레이 광고 성과, 이제 에어브릿지로 실시간 측정할 수 있어요
URL: https://www.airbridge.io/ko/blog/airbridge-ably-ads-integration
Language: Korean
Published: 2026-08-20
Author: Jaehyuk Kim
요즘 커머스 시장에서 에이블리(ABLY)가 빠르게 성장하면서 광고 매체로도 적극적으로 활용하는 브랜드가 많아지고 있는데요.
이번에 에어브릿지(Airbridge)가 MMP 최초로 에이블리 연동을 시작해요. 에이블리 디스플레이 광고 성과를 다른 매체와 같은 대시보드에서 실시간으로 측정하고 비교할 수 있게 됐어요.
**📌 핵심 포인트**
- 에어브릿지가 MMP 최초로 에이블리 디스플레이 광고와 연동됐어요.
- 에이블리 디스플레이 광고를 구글, 메타, 틱톡, 네이버 같은 다른 매체와 같은 대시보드에서 비교할 수 있어요.
- 노출, 클릭, 설치, 구매, ROAS를 캠페인과 지면 단위로 실시간 확인해요.
- 에어브릿지 파일럿(Airbridge Pilot)과 MCP 등 에어브릿지 AI를 활용하면 에이블리 디스플레이 광고 성과를 자연어로 조회 및 분석할 수 있어요.
## **에이블리 광고, 국내 커머스 매체로 자리 잡았어요**
에이블리를 패션 쇼핑 앱으로만 알고 있다면 매체 규모가 새삼스러울 수 있어요. **2025년에 월 사용자가 약 1,000만 명까지 돌파**했고, 월 총 사용시간과 월 총 실행 횟수 모두 국내 버티컬 커머스 플랫폼 중 손꼽히는 수준이에요.
[에이블리 월간 활성 이용자(MAU)가 약 1,000만 명으로 국내 버티컬 커머스 플랫폼 상위권]가 약 1,000만 명으로 국내 버티컬 커머스 플랫폼 상위권")
[에이블리 유저 분포 - 10~40대 71%, 여성 유저 비율 93%]
특히 세대별로 보면 매체 포지션이 뚜렷한데요. **10대부터 40대까지 전문몰 앱 시장에서 대표 매체**로 자리 잡았어요. 50대 이상 이용자도 꾸준히 늘고 있으며, 이는 국내 패션·뷰티·라이프 카테고리 소비자 대부분에게 도달할 수 있는 매체라는 뜻이에요.
[에이블리는 AI 프로필·운세·웹툰·웹소설·코디 등 콘텐츠 플랫폼으로 진화 중]
최근에는 AI 프로필과 운세, 웹툰·웹소설, 코디 등 흥미롭고 다양한 콘텐츠를 제공하며, 쇼핑앱을 넘어 여성 쇼핑 유저들을 위한 콘텐츠 플랫폼으로 새롭게 진화하고 있어요.
## **에이블리 디스플레이 광고는 미입점 광고주를 위한 서비스예요**
[에이블리 디스플레이 광고 - 에이블리 미입점 광고주 전용 배너 광고, 외부 웹·앱 아웃랜딩 지원][주요 광고 노출 지면 - 스플래시, 프리미엄 브랜딩 전면, 혜택 탭 Top, 배송조회 페이지, 종료 팝업]
에이블리 디스플레이 광고는 에이블리에 입점하지 않은 브랜드가 집행할 수 있는 배너 광고예요. 앱 안의 대표 지면을 통해 자사 페이지나 앱, 스마트스토어, 카카오스토어 선물하기, 토스 스토어로 유저를 유입시킬 수 있어요.
노출 지면은 앱의 유저 여정 전 구간에 걸쳐 있는데요. 앱을 처음 켤 때 보이는 **스플래시와 프리미엄 브랜딩 전면 배너**는 대규모 유저에게 강한 첫 인상을 남기기 좋고, **혜택 탭 Top과 리스트 배너는 프로모션을 찾는 유저에게 노출**돼요.
**배송조회 페이지와 마이 페이지 배너**는 이미 구매 경험이 있는 유저를 다시 만나는 지면이고, **종료 팝업, 검색 페이지, 커뮤니티 리스트 배너**까지 총 9개 지면 중에서 캠페인 목적에 맞게 조합할 수 있어요.
[에이블리 디스플레이 광고가 특별한 이유 - 대규모 여성 유저 트래픽, 높은 유저 퀄리티, 1st Party 구매 데이터 기반 타게팅]
타겟팅은 에이블리의 1st Party 구매·결제 데이터를 기반으로 해요. 인구통계, 구매 카테고리, 결제 수단·금액 조합으로 구매 가능성이 높은 유저에게 정확히 노출할 수 있어요.
과금은 **CPD(Cost Per Day, 구좌제)** 방식이에요. 지면과 일자를 정가 슬롯으로 확보하는 구조라 하루 단위 노출 재고를 예측하기 쉬운 대신, 담당자와 협의해서 사전에 소재를 준비해야 해요.
자사몰 유입을 만들거나 신규 유저에게 강한 노출로 브랜드를 소개하고 싶은 커머스 마케터가 검토하기 좋은 매체가 되고 있어요.
## **에어브릿지 MMP 연동으로 에이블리 디스플레이 광고를 다른 매체와 같은 대시보드에서 볼 수 있어요**
이번 에어브릿지 연동으로 다음 3가지를 확인할 수 있어요.
### **1\. 노출, 클릭, CPM을 다른 매체와 같은 기준으로 비교해요**
에어브릿지 대시보드에서 구글, 메타, 틱톡, 네이버, 카카오, 당근, 토스애즈와 나란히 에이블리 디스플레이 광고 성과를 비교할 수 있어요. 노출수, 클릭수, 클릭률, 광고비를 매체, 캠페인, 지면 단위로 확인하고 매체별 CPC와 CPM도 나란히 확인할 수 있어요.
매체마다 어트리뷰션 모델과 리포트 로직이 달라 다르게 보이던 성과도 같은 룩백 윈도우와 같은 이벤트 기준으로 다시 확인할 수 있게 돼요.
캠페인, 지면 단위 세분화가 유지되니 어느 지면 배너가 에이블리에서는 잘 반응하는데 다른 매체 광고와는 궁합이 어떤지 데이터로 볼 수 있어요.
### **2\. 유저 여정 전체를 앱 설치·구매 이벤트까지 이어서 봐요**
에이블리 디스플레이 광고를 클릭한 유저가 며칠 뒤 다른 매체 리마케팅을 거쳐 앱을 설치하고 결제했다고 해볼게요.
에어브릿지는 광고 클릭→앱 설치→상품 조회→장바구니 담기→구매까지 모든 유저 여정을 어트리뷰션과 함께 이어서 봐요. 여기에는 **에이블리 디스플레이 광고가 발견 단계에서 만든 기여도까지 성과에 포함**돼요.
어떤 상품 광고가 실제 구매까지 이어졌는지도 캠페인 단위로 판단할 수 있는데요. 특히 커머스에서는 광고를 본 유저가 며칠 뒤 다른 매체를 거쳐 구매하는 흐름이 흔해서 이 연결을 놓치면 매체의 실제 기여도를 낮게 평가하게 돼요.
## **에어브릿지로 국내외 광고 성과를 한 번에 측정해보세요**
이번 연동은 에어브릿지 고객사라면 모두 활용할 수 있어요. 에이블리 디스플레이 광고를 이미 집행 중이라면 에어브릿지 계정에서 에이블리 연동을 활성화하고 이벤트 매핑을 정리하는 것부터 시작해보세요.
여기에 [에어브릿지 파일럿(Airbridge Pilot)과 MCP(Model Context Protocol, AI 에이전트가 외부 도구를 호출하는 표준)](https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch) 등 에어브릿지 AI를 활용하면 더 쉽고 빠르게 데이터를 조회하고 분석할 수 있어요.
에어브릿지에서 통합된 데이터를 모니터링하면서 어떤 이벤트를 앱 전환으로 정의할지, 어떤 룩백 윈도우로 매체를 비교할지, 딥링크 구조를 어떻게 만들지 같은 결정을 초기에 정해두세요. 그러면 매체가 추가돼도 같은 기준으로 성과 데이터가 쌓여요.
에어브릿지는 아시아에 기반을 둔 유일한 MMP로서 구글, 메타, 틱톡 같은 글로벌 매체는 물론이고 에이블리, 당근, 토스애즈처럼 국내에서 성장한 매체까지 커버리지를 넓혀왔는데요.
도입 과정에서 결정하기 어려운 부분이 있다면 에어브릿지 도입문의를 요청해 주세요. 에어브릿지가 SDK 설정부터 매체별 이벤트 매핑까지 함께 도와드릴게요.
---
## 시그널 엔지니어링이란? 광고 플랫폼이 무엇을 학습할지 설계하는 법
URL: https://www.airbridge.io/ko/blog/what-is-signal-engineering
Language: Korean
Published: 2026-08-18
Author: Roi Nam
광고 플랫폼은 전달받은 전환이 나중에 취소될지 스스로 판단하지 못해요. 결제 직후 취소된 주문이나 첫 결제 후 바로 해지한 구독도 이미 ‘성공한 전환’으로 전달됐다면 광고 플랫폼의 학습 대상에 포함될 수 있어요.
광고 성과가 정체되면 흔히 소재와 타기팅, 입찰부터 점검해요. 하지만 플랫폼에 어떤 행동을 성공으로 전달하고 있는지도 함께 살펴봐야 해요. 잘못된 전환을 계속 보내면 알고리즘은 우리가 원하지 않는 사용자까지 좋은 고객으로 학습할 수 있기 때문이에요.
이 글에서는 광고 플랫폼에 보낼 전환을 고르는 기준과 업종별 설계 시나리오를 살펴봐요. 전환을 일정 시간 보류해 취소된 전환을 걸러내는 방법도 함께 알아볼게요.
**📌핵심 포인트**
- 시그널 엔지니어링은 더 많은 이벤트를 보내는 일이 아니에요. 광고 플랫폼이 실제 비즈니스 성과에 가까운 행동을 학습하도록 전환 이벤트와 값을 설계하는 일이에요.
- 구매나 유료 구독처럼 퍼널 아래쪽 이벤트는 비즈니스 가치와 가깝지만 발생량이 적을 수 있어요. 최적화 이벤트를 고를 때는 신호의 품질과 학습에 필요한 발생량을 함께 확인해야 해요.
- 취소·환불될 전환이나 여러 경로에서 중복 전송된 이벤트가 섞이면 광고 플랫폼이 잘못된 성공 기준을 학습할 수 있어요. 전환의 제외 조건과 전송 경로, 중복 제거 기준을 함께 설계해야 해요.
- Airbridge Signal Hold는 전환 이벤트를 일정 시간 보류하고 같은 거래 ID를 가진 취소 이벤트가 발생하면 두 이벤트를 포스트백에서 제외해요. 현재는 Meta Ads에 전달하는 전환 신호에 적용할 수 있어요.
## **시그널 엔지니어링은 광고 플랫폼의 학습 목표를 설계해요**
**시그널 엔지니어링(signal engineering)** 은 실제 비즈니스 성과에 가까운 사용자 행동 데이터를 광고 플랫폼에 보내 최적화 기준을 설계하는 일이에요. 광고 소재나 입찰뿐 아니라 플랫폼이 어떤 행동을 좋은 전환으로 학습할지도 마케터가 설계하는 거죠.
광고 플랫폼은 설치, 회원가입, 체험 시작, 구매 같은 이벤트를 바탕으로 전환 가능성이 높은 사용자를 찾아요. 같은 캠페인이라도 어떤 이벤트를 최적화 목표로 전달하느냐에 따라 플랫폼이 찾는 사용자의 특성이 달라질 수 있어요.
그래서 이벤트를 많이 보내는 것보다 비즈니스 가치와 연결된 행동을 일관된 기준으로 보내는 편이 중요해요. 목표 이벤트와 제외 조건, 전송 경로, 후행 지표를 함께 정해야 시그널이 실제 캠페인 의사결정에 도움이 돼요.
[앱에서 발생한 구매 이벤트가 Signal Hold 게이트에서 대기하고 취소와 짝이 맞은 이벤트는 걸러진 뒤 확정 전환만 광고 채널로 전달되는 애니메이션]
## **퍼널 아래쪽 이벤트가 무조건 정답은 아니에요**
설치 이벤트만 최적화 목표로 삼으면 플랫폼은 설치 가능성이 높은 사용자를 찾는 데 집중해요. 이 선택이 단기 CPI를 낮출 수는 있지만 해당 사용자가 구독·구매·재방문까지 이어질지는 별도로 확인해야 해요.
그래서 시그널 품질을 이야기할 때 흔히 “최대한 퍼널 아래쪽 이벤트로 학습시켜라”는 결론이 나와요. 하지만 한 가지 더 고려할 점이 있어요.
구매나 유료 구독처럼 퍼널 아래쪽 이벤트는 비즈니스 성과와 가깝지만 발생량이 적을 수 있어요. 이벤트 수가 적으면 학습에 쓸 데이터가 부족해지고 알고리즘이 안정적인 패턴을 잡기 어려워요. Meta Ads도 충분한 결과를 얻지 못한 광고 세트를 [‘학습 제한(Learning limited)’](https://www.facebook.com/help/messenger-app/650774041651557) 상태로 구분해요.
처음 시작한 앱이나 소프트 런칭 단계에서 구매·유료 구독 전환을 충분히 확보하지 못했다면 그 앞에 있는 장바구니 담기, 회원가입, 체험 시작처럼 발생량이 많은 이벤트를 최적화 목표로 검토해보세요. 이벤트를 바꾼 뒤에는 구매·구독 유지 같은 후행 지표가 함께 좋아지는지 확인해야 해요.
**이렇게 시그널 설계는 ‘높은 순도’와 ’충분한 학습량’ 사이에서 균형점을 찾는 문제예요.** 캠페인의 목표가 장기 매출이나 LTV라면 전환에 가까운 이벤트를 선택하되 플랫폼이 학습할 만큼 충분히 발생하는지 함께 확인해야 하죠.
목표와 가까우면서도 안정적으로 쌓이는 지점이 서비스마다 다르기 때문에 정답을 외우기보다 우리 서비스의 이벤트량과 전환 구조를 놓고 판단해야 해요.
## **좋은 시그널을 설계할 때 고려할 3가지 기준**
좋은 시그널을 설계하는 일은 하나의 ‘정답 이벤트’를 고르는 작업이 아니에요. 비즈니스 가치와 학습량, 취소·환불 같은 예외, 여러 전송 경로의 중복 가능성을 함께 검토해야 해요.
### **1\. 학습량과 퍼널 단계의 균형을 맞춰요**
가치가 높은 이벤트는 발생량이 적고 자주 발생하는 이벤트는 사용자의 실제 가치를 충분히 설명하지 못할 수 있어요. 어느 정도 볼륨이 나오면서도 우리 비즈니스 목표에 근접한 이벤트를 선정해야 해요.
### **2\. 이벤트의 순도를 높이고, 실제 비즈니스 성과를 반영해야 해요**
전환이 발생했다는 사실만 보내면 이후 취소나 환불을 반영하지 못해요. 따라서 취소나 환불이 많은 경우에는 이를 제외하고 시그널을 보내는 것을 고려해야 해요.
혹은 모든 전환을 다 보내게 되면 낮은 가치의 전환도 같은 선상에서 머신러닝에 반영될 여지가 있어요. 이 때는 구매 금액, 특정 카테고리의 구매 등처럼 비즈니스 결과와 직접 연결된 조건을 정하는게 중요해요.
### **3\. 여러 전송 경로에서 이벤트가 중복되지 않게 해요**
광고 플랫폼의 공식 모바일 SDK(소프트웨어 개발 키트)와 웹 픽셀, MMP(모바일 성과 측정 솔루션), 전환 API(Conversions API, CAPI)를 함께 쓰는 것 자체가 문제는 아니에요.
다만 같은 이벤트가 여러 전송 경로를 통해서 여러 번 전달되면 중복이 발생할 수 있어요. 예를 들어 같은 회원가입 이벤트인데 웹 픽셀을 통해서 먼저 수집되고, 이후에 전환 API를 통해서 두 번째로 수집될 수 있어요. 그래서 같은 이벤트라면 이벤트 ID를 보내서 중복을 제거해야해요.
## **업종별 시나리오로 보는 시그널 엔지니어링**
실제로 어떻게 시그널 엔지니어링을 활용할 수 있을까요? 업계별로 자주 보이는 예시를 모아봤는데요. 아래 내용은 실제 고객 사례가 아니라 업종별 전환 구조를 바탕으로 시그널 설계 방법을 설명한 가상 시나리오예요.
### **1\. 배달 앱: 주문 직후 취소를 학습 신호에서 제외해요**
주문 직후 취소 비율이 높은 배달 앱을 가정해볼게요. 메뉴를 잘못 골랐거나 주소를 확인하지 못해 결제 직후 주문을 취소하는 상황이 생길 수 있어요.
문제는 포스트백이 실시간으로 나간다는 점이에요. 곧 취소될 주문까지 ‘주문 완료’라는 성공 시그널로 매체에 전달되고 매체 알고리즘은 주문했다가 바로 취소하는 사용자까지 전환 사용자로 학습해요.
먼저 자사 데이터에서 주문 완료와 취소 사이의 시간 간격을 확인해보세요. 취소가 주로 주문 후 30분~1시간 안에 발생한다면 주문 완료와 주문 취소를 한 쌍으로 정의하고 해당 시간이 지난 뒤 남은 주문만 전송하는 방식을 검토할 수 있어요.
### **2\. 패션 커머스: 결제 직후 취소를 걸러내요**
결제 직후 취소가 운영상 문제가 되는 패션 커머스를 가정해볼게요. 심야에 충동적으로 결제했다가 1시간 후 취소하거나 세일 기간에 여러 옵션을 결제한 뒤 일부만 남기는 상황이 생길 수 있어요.
품절이 걱정돼 중복 주문을 했다가 하나를 취소하기도 하죠. 이런 주문이 모두 전환 시그널로 나가면 매체는 곧 취소할 사용자까지 전환 사용자로 학습해요.
여기서는 구매 완료와 구매 취소를 쌍으로 묶고 취소가 몰리는 시간 간격을 확인한 뒤 대기 기간을 정해요. 예를 들어 결제 후 1~2시간 안에 취소가 집중된다면 그 구간을 대기 기간 후보로 검토할 수 있어요. 대기 기간 안에 구매와 취소가 짝을 이루면 두 이벤트 모두 포스트백에서 제외할 수 있어요.
다만 한 가지 생각해야 할 점은 대기 기간으로 거를 수 있는 것은 결제 후 짧은 시간 안의 취소까지예요. 배송을 받아본 뒤 며칠 지나 일어나는 반품은 이 방식으로 걸러지지 않으니, 반품이 핵심 문제라면 이벤트 값 조건이나 전송 시점 설계를 병행해야 해요.
### **3\. 구독 앱: 시작하자마자 해지하는 충동 구독을 걸러요**
무료 체험을 시작하자마자 해지 예약을 걸거나 첫 결제 직후 환불을 요청하는 사용자가 많은 구독 앱을 가정해볼게요. 구독 시작 이벤트만 매체에 보내면 알고리즘은 가입 직후 떠난 사용자까지 성공으로 학습해요.
구독 시작과 구독 취소를 쌍으로 연결해 짧은 시간 안의 해지를 걸러내면 결제를 유지하는 사용자에 가까운 신호를 보낼 수 있어요. 구독은 구매보다 발생 빈도가 낮을 수 있으므로 학습에 쓰이는 이벤트가 충분한지도 함께 확인해야 해요.
여기까지 세 업종의 공통점은 분명해요. 업종은 달라도 거르는 대상은 같아요. 전환 직후 짧은 시간 안에 일어나는 충동 구매와 충동 취소예요. 이제 조건 축으로 넘어가 볼게요.
### **4\. 커머스: 구매 금액으로 전송할 신호를 구분해요**
시간만 조정할 수 있는 것은 아니에요. 어떤 구매를 학습시킬지도 설계 대상이에요. 소액 구매가 대부분인 커머스에서 모든 구매를 똑같이 시그널로 보내면 매체는 소액 구매 사용자를 중심으로 학습해요.
이럴 때 특정 금액 이상의 구매만 전환 시그널로 전송하는 방식을 검토할 수 있어요. 예를 들어서 결제 금액의 분포를 보고, 내부 기준에 따라서 5만 원 이상의 결제만 결제 이벤트로 보내는 방식이죠. 결제 금액의 편차가 큰 커머스에서 시험해볼 만한 설계예요.
### **5\. 데이팅 앱: 가입보다 실제 매칭에 가까운 행동을 보내요**
가입자 수보다 실제 매칭이나 대화 시작을 중요한 지표로 보는 데이팅 앱이라면 가입 이벤트만으로 사용자 가치를 설명하기 어려워요. 프로필 작성 완료, 상호 매칭, 첫 대화처럼 서비스 가치에 가까운 행동 가운데 충분한 발생량을 확보할 수 있는 이벤트를 후보로 검토해보세요.
## **시그널 홀드(Signal Hold): 확정된 전환만 광고 채널에 전달해요**
[에어브릿지 시그널 홀드(Signal Hold)]")
시그널 엔지니어링을 제품 기능으로 구현한 것이 에어브릿지가 공개한 **Signal Hold(시그널 홀드)**예요.
에어브릿지 대시보드의 시그널 스튜디오(Signal Studio)에서 설정할 수 있는 기능으로, 취소 가능성이 있는 전환 이벤트를 일정 시간 보류했다가 대기 기간 안에 취소되지 않은 전환만 광고 채널에 포스트백으로 전송해요. 현재는 메타 애즈(Meta Ads)에서 사용할 수 있어요.
광고 매체는 MMP가 보낸 전환 포스트백을 바탕으로 어떤 사용자가 전환했는지 학습하고, 이와 비슷한 특성을 가진 사용자를 찾아 광고를 최적화해요. 구매 직후 취소한 사용자의 전환까지 그대로 전달하면 매체는 곧 취소할 사용자도 성공한 전환의 기준으로 학습할 수 있어요. 시그널 홀드는 이런 전환을 일정 시간 보류하고, 취소되지 않은 전환만 보내 매체가 학습하는 시그널의 순도를 높여요.
### **에어브릿지 시그널 홀드는 이렇게 작동해요**
사용 전에는 전환·취소 이벤트에 동일한 거래 ID가 수집되는지 확인하고 Meta Ads 채널 연동과 인앱 이벤트 포스트백을 설정해야 해요. 전환 이벤트는 최근 24시간 동안 거래 ID와 함께 20건 이상 수집돼야 선택할 수 있어요. 특정 캠페인에만 적용하려면 하루에 기여 전환이 20건 이상 발생한 캠페인이 필요해요.
시그널 홀드는 아래 4단계로 동작해요.
1. 전환 이벤트가 발생해요.
2. 전환 이벤트를 바로 보내지 않고 대기해요.
3. 대기 기간 안에 동일한 Transaction ID를 가진 취소 이벤트가 발생하면, 대기 중인 전환 이벤트와 취소 이벤트 모두 포스트백으로 전송하지 않아요.
4. 대기 기간이 끝날 때까지 취소가 없으면 전환 이벤트를 광고 채널로 전송해요.
[시그널 홀드의 이벤트 쌍 동작 방식: 대기 기간 안에 취소 이벤트가 발생하면 전환·취소 이벤트 모두 포스트백에서 제외되고, 취소가 없으면 전환 이벤트가 전송된다]
에어브릿지 대시보드의 이벤트 쌍 구성 화면이에요. 동작 방식은 [시그널 홀드 가이드](https://help.airbridge.io/ko/guides/signal-hold)에서 볼 수 있어요.
시그널 홀드는 이벤트 이름과 상관없이 동일한 거래 ID로 전환·취소 쌍을 매칭해요. 필요한 이벤트와 거래 ID가 이미 수집되고 있다면 대시보드에서 이벤트 쌍과 대기 기간을 설정할 수 있어요. 거래 ID를 수집하지 않는 환경이라면 먼저 이벤트 구조와 데이터 수집 방식을 수정해야 해요.
구매 금액 기준은 [일반 포스트백 설정](https://help.airbridge.io/ko/guides/postback-settings-new)의 이벤트 밸류(Event Value) 조건으로 다룰 수 있어요. 그 밖의 사용자 속성을 기준으로 시그널을 나누려면 이벤트 수집·분류 방식과 광고 채널의 데이터 정책을 별도로 확인해야 해요.
[에어브릿지 대시보드의 시그널 홀드 설정 화면: 이벤트 쌍(Order Complete → Order Cancel), 대기 기간 1시간, 캠페인 필터가 적용되어 있고 포스트백에서 제외된 전환 이벤트 수와 제외 비율이 표시된다], 대기 기간 1시간, 캠페인 필터가 적용되어 있고 포스트백에서 제외된 전환 이벤트 수와 제외 비율이 표시된다")
에어브릿지 데모 환경의 시그널 홀드 설정·리포트 화면이에요.
대기 기간은 자사 서비스의 취소 패턴에 맞춰야 해요. 공식 가이드의 예시처럼 분석 결과 단순 변심 취소의 70%가 구매 후 30분 안에 발생한다면 30분을 후보로 설정할 수 있어요.
취소가 전환 며칠 뒤에 몰리는 서비스라면 최대 24시간인 대기 기간과 맞지 않아요. 기여 이벤트가 적은 캠페인이라면 시그널을 거르기보다 발생량을 늘리는 편이 먼저예요.
적용 후에는 성과 분석 리포트에서 적용 전 8주와 적용 후 8주의 지표, 그리고 포스트백에서 제외된 전환 수를 확인할 수 있어요. 시그널 홀드는 성과를 보장하는 기능이 아니라 알고리즘이 학습하는 시그널의 품질을 높이는 도구예요. 자세한 설정 방법은 [시그널 홀드 설정 가이드](https://help.airbridge.io/ko/guides/signal-hold)에서 확인할 수 있어요.
**💡이건 알아두면 좋아요**
단, 시그널 홀드 자체가 성과 개선을 보장하진 않아요. 마케팅 성과는 여러 변수에 따라 달라지기 때문이죠. 단순히 취소 시그널을 보낼지 말지 마케터가 직접 대시보드에서 설정할 수 있게 해주는 기능이고, 다른 MMP에선 제공하고 있지 않는 기능이에요. 그래서 성과를 올릴 수 있기 보다는 좋은 시그널을 보내는데 포커스를 두어야 해요.
## **지금 사용하는 시그널부터 한 번 점검해보세요**
시그널 엔지니어링은 더 많은 데이터를 보내는 기술이 아니라 광고 플랫폼이 무엇을 성공으로 학습할지 정하는 과정이에요. 광고비가 큰 캠페인 하나를 골라 ‘목표 → 전송 이벤트 → 제외 조건 → 전송 경로(플랫폼 SDK, 플랫폼 픽셀, MMP, 전환 API 등) → 후행 지표(CPA, ROAS 등)’를 정리해보세요.
목표와 이벤트가 어긋나거나 같은 전환이 여러 경로로 전달되는 지점부터 수정하면 돼요. 전환 직후 취소가 자주 발생한다면 에어브릿지 [시그널 홀드 설정 가이드](https://help.airbridge.io/ko/guides/signal-hold)에서 적용 조건을 확인할 수 있어요.
에어브릿지는 앞으로도 더 많은 시그널 엔지니어링 기능을 출시하여 실제적으로 효과성 높은 캠페인을 운용하도록 도울 예정이에요. 매체는 우리가 보내는 데이터를 기반으로만 학습하여 동작한다는 점을 잊지 말고, 시그널 엔지니어링을 통해서 매체 성과를 더욱 개선해 보세요.
---
## 에어브릿지 AI 출시 4개월, 820개 브랜드는 이렇게 썼어요
URL: https://www.airbridge.io/ko/blog/airbridge-ai-usage-record-data
Language: Korean
Published: 2026-08-06
Author: Jaehyuk Kim
[에어브릿지 AI(Airbridge AI)](https://www.airbridge.io/ko/product/airbridge-ai)가 출시된 지 어느덧 4개월이 지났어요. 처음 이 기능이 공개됐을 때 구체적으로 어떻게 사용할 수 있는지 정말 많은 문의를 받았죠.
4개월이 지나고 실제로 고객사가 어떻게 에어브릿지 AI를 사용했는지 기록을 검토해봤어요. 팀마다 각자의 방식으로 물어보기 시작했고, 기존 챗봇처럼 한 번 답을 받고 끝내는 대신 이어서 되묻는 대화가 늘었어요. 한국어가 아닌 언어로 물어보는 팀도 예상보다 많았고요.
AI 마케팅 어시스턴트의 기능을 소개하는 글은 많지만, 실제로 도입한 팀이 무엇을 얼마나 물었는지 그 기록을 공개하는 경우는 드물어요.
이번 글에서는 에어브릿지 AI 출시 첫 4개월의 실제 사용 데이터를 먼저 보여주고, 팀에서 활용할 수 있는 팁 5가지, 그리고 대시보드 밖에서 확장 중인 MCP 사용 패턴까지 모두 정리했어요.
📌 **핵심 포인트**
- 4개월 동안 820개 브랜드가 16,610번을 물었어요. 특정 담당자가 아닌 팀 전체의 도구로 자리잡았죠.
- 가장 많이 물은 주제는 이벤트 택소노미·SDK·리포트. 분석보다 구현·운영 실무 질문이 압도적이었어요.
- 대시보드 밖에서도 800여명이 MCP로 자신의 AI에 에어브릿지를 연결했고, 최근 30일에만 22,207번의 데이터 조회가 발생했어요.
## **에어브릿지 AI, 4개월 만에 16,610건의 질문을 받았어요**
[4개월 동안 누적 질문 16,610건, 고객사 820개, 사용자 1,700명]
2026년 3월 25일부터 7월 23일까지, 에어브릿지 AI에는 **누적 16,610건의 질문**이 들어왔는데요. 고객사 기준으로는 약 820개 브랜드, 물어본 고객은 약 1,700명이었어요.
월별 흐름을 봐도 7월 사용량은 23일 만에 이미 6월 한 달 수준을 넘어섰는데요. 이는 실무에 에어브릿지 AI가 폭넓게 스며들었다는 것을 의미해요.
## **에어브릿지 AI에 고객이 가장 많이 물은 5가지 질문 유형**
질문을 유형별로 분류해보면, 고객사들이 주로 찾는 에어브릿지 AI 기능이 무엇인지 선명하게 드러나요. 가장 질문이 많이 몰린 유형은 **이벤트·택소노미 설계**였고, 그 뒤로 SDK·설치·연동, 리포트 조회 유형이 이어졌어요.
[질문 유형 상위 5개 - 이벤트 택소노미 설계 14.1%, SDK·설치·연동 13.3%, 리포트·대시보드 11.8%, 광고채널·매체 연동 11.2%, 딥링크 10%]
| 순위 | 유형 | 비중 |
| --- | --- | --- |
| 1 | 이벤트·택소노미 설계 | 14.1% |
| 2 | SDK·설치·연동 | 13.3% |
| 3 | 리포트·대시보드·데이터 조회 | 11.8% |
| 4 | 광고채널 연동·매체 | 11.2% |
| 5 | 딥링크 | 10.0% |
특히 SDK·설치·연동 유형은 이벤트·택소노미 설계 다음으로 많은 고객사가 물어본 영역이었는데요. 이를 통해, 새로 도입하는 팀 대부분이 이 구간에서 한 번은 막힌다는 것을 알 수 있어요.
[Airbridge AI_after 4 month_03.webp]
분포를 보면 상위 5개 유형이 모두 **구현·운영 실무**에 몰려 있고, 데이터 해석 관련 질문보다 측정 기반을 세우고 운영하는 질문이 압도적이었어요. 여기에 트러블슈팅·데이터 불일치, 광고비·비용·ROAS, 어트리뷰션·기여까지 더하면 상위 8개 유형이 전체 질문의 76%를 차지해요.
**가이드를 뒤져도 현재 실무 상황에 맞는 답이 바로 안 나올 때**, 담당자에게 부탁할 필요 없이 이를 해결하는 용도로 에어브릿지 AI를 적극적으로 활용했음을 알 수 있어요.
## **에어브릿지 AI에 실제로 던진 5가지 질문**
실제 팀이 에어브릿지 AI에 어떻게 묻고 어떤 답을 받았는지 흐름을 정리했어요. 회사명은 밝히지 않고 업종으로만 소개했으니, 나와 비슷한 케이스가 있는지 확인해 보세요.
[실제 사용자 질문 5가지 - 미디어믹스 재배분, 주간 성과·리텐션, 빈 택소노미 채우기, 모국어 SDK 가이드, 복잡한 어트리뷰션 상담]
### **1\. 미디어믹스를 분석하고 다음 달 예산까지 권고해줘요**
한 해외 커머스 앱 팀은 에어브릿지 AI에 이렇게 물었어요.
"이번 달 유료 채널 지출을 훑고 다음 달 어디에 집중할지 알려줘."
그러자 에어브릿지 AI는 채널별 **Cost, Cost Share, Installs, CPI, Orders, MAU, ROAS를 정리한 표와 그 위에 얹은 예산 재배분 권고**를 바로 답해줬어요. 고객 계정의 통화와 타임존을 그대로 반영했기 때문에 다시 환산할 필요 없이 바로 읽을 수 있어요.
[에어브릿지 AI 답변 예시 - 유료 채널 성과 정리 표와 예산 재배분 권고]
### **2\. 주간 성과를 리텐션까지 한 질문으로 정리해줘요**
"지난주 국가별 신규모객·리마케팅 성과를 리텐션까지, 전주와 비교해줘."
이 질문에 에어브릿지 AI는 **CPA, CVR, D1/D7 리텐션 같은 커스텀 메트릭을 자동으로 식별**한 다음 Actuals 리포트 4개를 동시에 만들어 기간을 비교하는 답을 제공했어요. 사람이 리포트를 하나씩 열어 지표를 맞춰보던 작업을 한 번의 질문으로 묶은 거죠.
### **3\. 빈 이벤트 택소노미를 앱 데이터로 채워줘요**
택소노미 설계는 유저 여정과 [Standard Event](https://help.airbridge.io/ko/guides/airbridge-event-types) 매핑 지식을 요구해서 참고할 템플릿이 없는 신규 팀이 자주 막히는 지점이에요.
"핵심 유저 여정을 기반으로 이벤트 택소노미와 Standard Event 매핑을 설계해줘."
라고 에어브릿지 AI에 물으면, **택소노미 템플릿이 비어 있어도 앱스토어 정보·플랫폼·통화 같은 앱 데이터로 업종을 추론**해 Standard Event 세트를 매핑해줘요.
### **4\. 각자 모국어로 SDK 가이드를 만들어줘요**
한 팀은 중국어로 "안드로이드 SDK 어떻게 통합하나요"라고 물었어요. 그러자 에어브릿지 AI는 저장소 추가부터 초기화까지 **중국어 설명과 코드블록이 함께 붙은 단계별 가이드**를 제공했어요.
[에어브릿지 AI 답변 예시 - 중국어로 안내한 안드로이드 SDK 통합 단계별 가이드]
### **5\. 복잡한 어트리뷰션 문제도 공식 문서로 상담해줘요**
영어로 "[ATT(Apple App Tracking Transparency)](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork) 동의 타이밍과 어트리뷰션의 닭-달걀 문제를 어떻게 봐야 하나"라고 물은 사례도 있었어요.
이런 심층 질문에 에어브릿지 AI는 **먼저 공식 문서를 검색한 뒤 시나리오별로 확정적(deterministic) 여부를 구분**하고, 제품 설계가 그 문제를 어떻게 해소하는지 근거와 함께 설명하며 답했어요. 개념과 제품 설계를 함께 이해해야 풀리는 문제라, 누구에게 물어야 할지조차 막막하던 난제를 바로 상담할 수 있는 창구로 작동한 것이죠.
## **파워유저는 Slash Command로 워크플로우를 불러요**
여기까지가 대부분의 일반적인 에어브릿지 AI 사용 방식인데요. 그런데 소수의 파워유저는 여기서 한 발 더 나아가 [**AI Skills Slash Command**](https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch)를 적극적으로 사용했어요.
출시 이후 Slash Command 사용 데이터를 보면 가장 많이 활용된 건 택소노미 설계 커맨드였고, 데이터 분석 커맨드가 그 뒤를 이었어요.
[유저의 9.1%가 Slash Command로 에어브릿지 AI 스킬을 사용]
| Slash Command | 용도 |
| --- | --- |
| /airbridge-event-taxonomy-designer | 이벤트 택소노미 자동 설계 |
| /airbridge-data-analyst | 데이터 분석 요청 |
| /airbridge-monthly-mediamix | 월간 미디어믹스 분석 |
| /airbridge-weekly-review | 주간 성과 리뷰 |
| /airbridge-industry-benchmark | 업종 벤치마크 |
| /airbridge-campaign-brief | 캠페인 브리프 작성 |
| /airbridge-kpi-taxonomy | KPI 택소노미 설계 |
| /airbridge-analyze-discrepancy | 데이터 불일치 분석 |
커맨드 하나로 반복 작업을 실행하고, 필요하면 우리 팀에 맞는 스킬을 직접 만들 수도 있어요. 다만 규모를 보면, Slash Command는 전체 질문의 약 1.5%, 유저의 약 9.1%에 해당하는 일부 유저만 사용하고 있어요.
[800여명이 에어브릿지 MCP를 연결, 최근 30일간 데이터 호출 22,207회, 리포트·데이터 조회가 84%]
즉, 에어브릿지 AI를 쓰는 대부분의 유저는 구현·운영 실무 도우미로 활용하고 있고, 소수의 파워유저는 분석 워크플로우까지 확장하고 있다는 걸 알 수 있어요.
## **에어브릿지 MCP, 800여명이 Claude와 Cursor에 에어브릿지를 연결했어요**
지금까지가 에어브릿지 대시보드 안에서 벌어진 일이라면, 이제는 고객이 쓰는 AI 툴 안에서도 에어브릿지가 어떻게 실무를 함께 하고 있는지 데이터로 소개해드릴게요.
에어브릿지 파일럿과 함께 고객이 이미 쓰는 AI 툴(Claude·Cursor 등)에 데이터를 연결하는 [**에어브릿지 MCP(Model Context Protocol) 서버**](https://help.airbridge.io/ko/guides/mcp)도 열었는데요. 출시 이후 **800여명이 자신의 AI에 에어브릿지를 연결했어요.**
최근 30일(6/23~7/22) 실사용 데이터를 보면 연결된 AI가 에어브릿지 데이터를 불러온 횟수가 총 22,207회에 달했는데요. 한 번 연결하면 매일 쓰는 인프라형 도구로 자리 잡았다는 뜻이에요.
그중에서도 리포트·데이터 조회 관련 호출이 전체의 약 84%를 차지해요. MCP가 **AI로 하는 성과 분석** 채널로 자리 잡고 있는 셈이에요.
## **에어브릿지 AI 4개월간 데이터가 보여준 것은 무엇이었을까요?**
4개월간의 데이터를 정리하면서 한 가지 확인한 것이 있어요. 에어브릿지 AI는 정해진 답을 되돌려주는 챗봇이 아니라, 그 순간의 데이터와 설정을 읽어 답을 만드는 도구라는 것이죠. 그리고 이제는 대시보드 안에서만 쓰는 도구가 아니라, MCP를 타고 여러 AI 툴 안에서 대화하면서 실무에 활용할 수 있는 도구가 됐어요.
지금 실무에서 막히는 지점에 대한 질문 하나를 골라 대시보드에서 에어브릿지 AI에 그대로 던져보고, 지금 쓰고 있는 Claude·Cursor에 에어브릿지 MCP를 한 번 연결해보세요. 표와 출처가 붙은 답이 돌아오는지, 우리 계정 설정이 반영됐는지 직접 확인해보세요. 그 과정 속에서 에어브릿지 AI를 우리 팀 상황에 맞게 어떻게 잘 활용할 수 있을지 감을 잡을 수 있을 거예요.
---
## 2026 게임 업계 여름 업데이트, 마케터가 챙겨야 할 장기 리텐션 전략
URL: https://www.airbridge.io/ko/blog/game-summer-event-retention-strategy-for-marketer
Language: Korean
Published: 2026-07-16
Author: Team Airbridge
매년 6월 말이 되면 주요 게임사들의 여름 업데이트 이벤트가 하나 둘 시작돼요.
연중 게임 사용량이 높아지는 여름 시즌에 맞춰 신규·복귀 유저를 끌어들이기 위한 다양한 혜택을 내 거는 것인데요.
마케터 입장에서 이 시기는 숫자가 좋아 보이는 구간으로, 신규 설치가 늘고 D7 리텐션도 평소보다 증가하곤 하죠.
그런데 이벤트가 끝나면 D30 리텐션이 평소보다 낮게 나오는 일이 잦아요.
이때 광고 성과의 문제인지, 이벤트 유입 유저의 자연 이탈인지 데이터만으로는 구분이 어렵고요.
이번 글에서는 2026년 주요 게임사들의 여름 업데이트 이벤트를 살펴보고, 이 시기 유입 유저의 리텐션을 어떻게 읽고 관리해야 하는지 정리해 볼게요.
**📌 핵심 포인트**
- 2026년 여름 업데이트 이벤트를 통해 게임사가 선택한 마케팅 포인트를 참고해 보세요. 이벤트 설계 방식에 따라 유입되는 유저 유형에 따라 이벤트 종료 후 리텐션 패턴이 달라져요.
- 이벤트 시작 전 갖춰야 할 데이터 측정 체계부터 이벤트 종료 후 성과 해석까지 단계별로 정리했어요. 이 글을 읽고 나면 이벤트 데이터를 어떻게 설계하고 읽어야 하는지 구체적인 기준을 잡을 수 있어요.
## **2026 여름 업데이트 이벤트, 게임사는 어떤 방식을 선택했을까?**
2026년 여름 이벤트를 보면 게임사마다 유저를 끌어들이는 방식이 달라요. 각 게임사의 이벤트 설계 전략을 참고해 보세요.
### **1\. 진입 환경 새로 설계하기**
MMORPG(대규모 다중접속 역할수행 게임)는 유저가 시간을 쏟을수록 캐릭터가 강해지는 구조인데요. 게임을 오래 쉬었다가 돌아오면 장비, 레벨, 콘텐츠 진행도가 다른 유저에 비해 뒤처져 있어요.
기존 유저와의 격차를 따라잡는 데 드는 시간과 비용이 눈에 보이니 시작 자체를 포기하게 되죠. 이 방식은 그 부담을 해소하기 위해 **별도 서버를 열어 새 캐릭터를 처음부터 빠르게 키울 수 있는 환경**을 만들어요.
넥슨의 대표작인 **메이플스토리**는 올해 6월 18일 [챌린저스 월드 시즌4](https://maplestory.nexon.com/News/Event/Ongoing/1349)를 열었어요.
기존 유저도 새 캐릭터를 만들어서 함께 참여할 수 있고, 여러 가지 이벤트를 진행 중이어서 메이플 초보 유저들이 이용하기 좋죠. 챌린저스 서버에서 신규·복귀 유저부터 기존 유저까지 새 캐릭터를 만들어 처음부터 육성할 수 있어요.
하이퍼 버닝과 아이템 버닝, 보스 격파 미션, 장비 세팅까지 한 시즌 안에 경험할 수 있는 종합 성장 구조예요. 약 4~5개월 운영 후 일반 서버로 이동할 수 있어요.
[메이플스토리 챌린저스 월드 시즌4 여름 이벤트 안내 페이지]
- 운영 기간: 2026년 6월 18일 ~ 9월 17일
- 주요 혜택: 하이퍼 버닝 MAX(레벨업 시 추가 레벨 획득), 하이퍼 블링크(200레벨 이상 구간 경험치 대폭 증가), 신규 직업 레테 육성 지원
- 함께 진행되는 이벤트: [미션 울티마](https://maplestory.nexon.com/news/update/805)(출석, 퀘스트, 던전 참여로 이어지는 일일 루틴 구성)
챌린저스 서버는 매년 시즌제로 반복돼요. 기존 서버의 생태계를 건드리지 않으면서 신규·복귀 유저를 별도 공간으로 유입시킬 수 있어, 게임사 입장에서 기존 유저 이탈 없이 새 유저를 데려오는 안전한 리텐션 설계 방식이에요.
넷마블의 흥행 대작 **리니지2 레볼루션** 사례도 볼까요? 넷마블은 6월 11일 신규 서버 '[파푸리온](https://l2.netmarble.net/fafurion)'을 열었어요. 기존 서버와 달리 혈맹(길드) 활동에 특화된 구조로 설계됐어요.
[리니지2 레볼루션 신규 서버 파푸리온 이벤트 페이지]
- 운영 기간: 2026년 6월 11일부터 진행
- 주요 혜택: 700레벨까지 성장 부스팅, 혈맹 레벨과 콘텐츠 참여도에 따라 최대 80% 레드다이아 페이백, 최대 15만 레드다이아 지급, 아가시온·탑승펫 풀 성장 지원
- 핵심 콘텐츠: 혈맹 던전에서 획득한 '[결속의 주화](https://forum.netmarble.com/lineageII/view/1747020/1126402)'로 성장 재료 교환 가능
메이플스토리가 개인 성장 환경을 새로 설계했다면, **리니지2 레볼루션은 혈맹(길드) 구조를 리텐션 장치로 활용**해요.
혈맹으로 묶인 유저는 혼자 플레이하는 유저보다 이탈 부담이 커요. 길드원과의 관계, 공동 목표, 혈맹 레벨이 쌓일수록 떠나기 어려운 구조가 만들어지거든요.
혈맹 던전은 길드원과 함께 참여해야 보상이 쌓이는 구조라 길드 전체에 영향을 미치고, 공동 목표가 쌓일수록 나만 빠지기 미안하다는 감정도 생기죠.
특히 해당 서버에 적용된 ‘수중 결계’ 기믹은 유저의 위치 선정과 협동 플레이를 요구해 기존 레이드와 차별화된 공략 요소를 제공하는데요.
게임 안에서 유저간 협심이 중요한 사항라, 유입과 리텐션을 동시에 잡을 수 있어요.
### **2\. 콘텐츠 업데이트와 이벤트 연동하기**
콘텐츠 업데이트로 유저를 끌어들이는 방식이에요. 앞서 두 게임이 새 서버로 진입 환경을 바꾸는 방식이었다면, 던전앤파이터는 콘텐츠 자체를 접속 이유로 만드는 거죠.
**던전앤파이터**는 6월 4일에 썸머 시즌을 열었어요. 핵심은 [썸머 부스트 업](https://df.nexon.com/pg/summerboostup26)으로, **신규 전직 출시와 이벤트를 연동한 구조**예요.
[던전앤파이터 썸머 부스트 업 이벤트 페이지]
- **운영 기간:** 2026년 6월 4일 ~ 8월 27일
- **주요 혜택:** 신규 전직 인파이터(여)·브레이커 캐릭터 특화 성장 지원, 기존 캐릭터 1종 추가 성장 혜택
- **함께 진행되는 이벤트:** [여름맞이 7일간의 여정](https://df.nexon.com/pg/summersevengift), [미라클의 해양 탐구 생활](https://df.nexon.com/pg/oceanmiracle)
던전앤파이터에서 ‘전직’은 캐릭터의 직업을 새롭게 선택하는 시스템이에요. **신규 전직이 추가되면 기존 유저는 새 직업을 키워보고 싶어서 다시 접속**하고, 신규 유저는 새 직업으로 게임 활동을 처음 시작하기 좋은 타이밍으로 받아들이죠.
게임 안에선 명성 63,257 미만의 캐릭터를 대상으로 115레벨까지 빠르게 성장시킬 수 있는 캡슐을 지급하며, 이로 인해 유저는 육성 부담을 덜 느끼게 돼요.
특히 ‘썸머 부스트업 캡슐(캐릭터를 빠르게 성장 시켜주는 아이템)’은 신규 전직 외에 기존 캐릭터 1종에도 적용돼요. 단, 일정 명성 미만 캐릭터에만 해당해서 신규·복귀 유저가 주요 대상이에요.
이렇게 **기존 캐릭터에도 이벤트 아이템이 적용 되면서 새 캐릭터를 만들지 않아도 혜택**받을 수 있게 됐어요. 콘텐츠 업데이트 하나로 신규 유입과 기존 유저 재활성화를 동시에 노릴 수 있죠.
### **3\. 스토리로 다음 화가 궁금해지는 이벤트 만들기**
승리의 여신: 니케는 6월 11일 신규 고등급 캐릭터(SSR) '아크레인저 블랙'을 추가하고, 모든 대사에 성우 더빙이 입혀진 스토리 이벤트 '[ARK RANGER](https://www.youtube.com/watch?v=s95vkFvZLlo)'를 선보였어요.
[승리의 여신 니케 ARK RANGER 스토리 이벤트 화면]
- **운영 기간:** 2026년 6월 11일 ~ 7월 2일
- **주요 콘텐츠:** 전용 맵 탐색, 풀 보이스 스토리 애니메이션, 디펜스 미니게임
- **함께 진행되는 모집:** [여름 한정 니케 6종 선택 모집](https://www.gametoc.co.kr/news/articleView.html?idxno=109128)
수집형 RPG는 다양한 캐릭터를 모으는 것이 핵심인 장르예요. **원하는 캐릭터를 뽑는 행동(가챠)로 획득하고, 그 캐릭터로 전투를 진행**해요. 가챠의 동기가 전투력 강화이죠.
니케는 여기에 스토리를 연동했는데요. **이벤트의 주인공이 픽업(Pickup, 특정 기간 획득 확률이 높아지는 한정 캐릭터)인 구조라 스토리를 보면서 캐릭터에 감정 이입**이 생기고, 그 캐릭터를 뽑고 싶다는 동기를 자연스럽게 느끼게 돼요.
니케는 매년 유저 인기투표를 진행하는데, 캐릭터 순위뿐 아니라 '가장 인상 깊은 이벤트 스토리' 부문이 따로 있어요. 스토리가 단순한 배경 설명이 아니라 유저가 직접 반응하고 평가하는 콘텐츠로 자리잡혀 있죠.
이번 ARK RANGER에 이벤트 스토리 전체에 성우 더빙을 입힌 것도 그 연장선이에요.
## **좋아보이는 이벤트 기간 숫자, 그 안을 들여다보면**
여름 이벤트 기간에는 신규·복귀 유저가 대거 유입되고, 이벤트가 설계한 성장 환경 안에서 행동해요.
이 유저들이 평소 오가닉 유저와 같은 코호트에 섞이면, 이벤트 종료 후 리텐션 하락이 UA 효율의 문제인지, 이벤트 유저의 이탈인지 구분이 어려울 수 있죠.
해당 기간의 성과 해석 시 유의해야 할 세 가지 지표를 중심으로 살펴볼게요.
### **1\. UA 효율을 과대평가하게 만드는 DAU**
DAU(Daily Active Users, 일간 활성 유저 수)가 오르는 건 이벤트 기간의 자연스러운 현상이에요.
문제는 이 시기에 UA(User Acquisition, 유저 획득) 캠페인, 즉 광고를 통해 새 유저를 유입시키는 마케팅을 집행하면 CPI(Cost Per Install, 앱 설치당 광고 비용)가 낮게 나오고 D7 리텐션도 평소보다 좋아 보인다는 거예요.
이는 이벤트 환경이 유저의 초기 행동을 바꿔놓기 때문인데요. 이 숫자를 UA 크리에이티브나 타겟팅의 성과로 읽고 예산을 늘리면, 이벤트가 끝난 뒤 D30 리텐션이 기대치를 크게 밑돌 수 있어요.
또한 그 원인이 소재 문제인지, 타겟팅 문제인지, 아니면 이벤트 유저의 자연 이탈인지 데이터만으로는 정확한 구분이 불가능하고요.
### **2\. 평소와 구성이 달라지는 이벤트 유입 코호트**
이벤트 기간 유입 유저로 구성된 코호트(Cohort, 동일 시점에 유입된 유저 집단)는 평소 코호트와 달리, 이벤트가 제공하던 성장 지원이나 보상이 사라지면 접속 패턴이 바뀔 수 있어요.
이 코호트를 전체 코호트에 합산하면 D30 리텐션이 이전 분기 대비 낮게 나와요. 역시 게임 퀄리티 문제인지, 이벤트 유저가 섞인 탓인지 원인을 특정하기 어렵고요.
코호트를 분리해서 각각의 D30을 비교해야 원인을 알 수 있는데, 유입 소스와 코호트가 합산된 상태에서는 숫자가 해석이 안 돼요.
### **3\. LTV가 다른 이벤트 유저와 오가닉 유저**
이벤트 유저와 오가닉유저는 게임을 시작한 이유가 다르죠.
이벤트 유저는 특정 환경이 만들어졌을 때 반응한 유저고, 오가닉 유저는 게임 자체에 관심이 있어서 유입된 유저이기 때문에 두 그룹은 이벤트 종료 후 접속 빈도, 결제 전환율, D30 리텐션에서 다른 패턴을 보여요.
이 둘을 합산해서 LTV(Lifetime Value, 유저 생애 가치)를 계산하면 각 채널의 실제 기여도를 파악할 수 없어 분리가 필요해요.
예를 들어 채널별 LTV(Lifetime Value, 유저 생애 가치)가 합산된 상태에서는 메타에서 데려온 유저와 구글에서 데려온 유저 중 누가 장기적으로 더 높은 가치를 만드는지, 이벤트 기간 유입 유저와 평소 유입 유저 중 실제 매출 기여가 어디서 나오는지 알기 어렵겠죠.
👉[에어브릿지를 활용한 게임사 성공 사례 바로가기](https://www.airbridge.io/ko/case-studies/vertical/gaming)
## **이벤트 유입 유저를 장기 유저로 전환하기**
이벤트 유입 유저를 장기 유저로 전환하려면 유입 소스별 어트리뷰션 설계와 코호트 분리 추적이 사전에 돼 있어야 해요.
이벤트가 끝난 뒤 데이터를 분리하려고 하면 늦기 때문에, **이벤트 시작 전 아래 3가지를 미리 세팅**해 보세요.
| 참고 지표 | 참고 데이터 | 이렇게 봐야 해요 |
| --- | --- | --- |
| DAU·CPI·D7 리텐션 | 유입 소스별 태깅 데이터 채널별 D7·D14·D30 리텐션 | 이벤트 시작일 기준으로 유입 소스별 태깅 분리 → 이벤트 유입 유저와 오가닉 유저의 D7·D14·D30 리텐션을 각각 추적해 이벤트 효과 구간과 이탈 구간 특정 |
| D30 리텐션 | 이벤트 종료 후 2주 구간 DAU 낙폭 코호트별 D30 | 이벤트 유입 코호트와 오가닉 코호트를 유입 시점 기준으로 분리 → 이벤트 종료 후 2주 구간의 DAU 낙폭을 각각 측정해 이벤트 의존도 파악 |
| LTV | 채널별 결제 전환율 재결제율 구매 주기 | 채널별 유입 유저의 결제 전환율·재결제율·구매 주기를 유입 시점별로 교차 분석 → 이벤트 기간 유입 유저의 초기 결제율과 이벤트 종료 후 재결제율 비교 |
세 지표를 분리해서 보려면 **이벤트 시작 전에 유입 소스별 태깅이 설계**돼 있어야 해요.
특히 여러 UA 채널을 동시에 집행하는 이벤트 기간에는 각 매체가 성과를 따로 리포팅하기 때문에 어트리뷰션이 겹치는 문제가 생길 수 있어요.
어떤 채널이 실제로 유저를 데려왔는지, 그 유저가 이벤트 종료 후에도 남아 있는지를 하나의 기준으로 보려면 여름 업데이트 이벤트가 시작되기 전에 측정 체계부터 점검해 보세요.
## **게임 여름 업데이트 이벤트 성과, 에어브릿지로 정확하게 측정하세요**
게임사의 시즌성 이벤트는 연간 UA 전략에서 가장 많은 예산이 집중되는 구간인 만큼, 데이터를 정확하게 측정해 이벤트 효과를 극대화해보세요.
에어브릿지(Airbridge)는 다음과 같은 측정 체계를 제공해요.
- [**멀티 채널 어트리뷰션**](https://help.airbridge.io/ko/guides/airbridge-attribution-model) : 구글, 메타, 애플 서치 애즈 등 주요 매체와 공식 파트너십을 통해 원본 광고 데이터를 직접 수급해요. 중복을 제거하고 실제 기여도를 정리하기 때문에, 이벤트 기간처럼 여러 채널을 동시에 집행할 때 어트리뷰션이 겹치는 문제를 해소해요.
- [**코호트 분석**](https://help.airbridge.io/ko/guides/funnel-report-viewing) : 이벤트 유입 유저와 오가닉 유저를 유입 시점 기준으로 분리해 D7·D14·D30 리텐션을 각각 추적할 수 있어요. 이벤트 종료 후 리텐션 하락의 원인을 특정하는 데 필요한 데이터를 만들어줘요.
- [**LTV·pLTV분석**](https://help.airbridge.io/ko/guides/predictive-lifetime) : 채널별·유입 시점별로 결제 전환율과 재결제율을 교차 분석해 어떤 채널이 실제로 장기 유저를 만드는지 볼 수 있어요. 예측 LTV로 이벤트 종료 후 유저 매출 추정치도 확인할 수 있어요.
[국내 주요 게임사(컴투스, 스마일게이트, NHN 등)가 에어브릿지로 Steam·PC게임 데이터를 측정하고 있음을 보여주는 로고 모음]
넥슨, 컴투스, NHN, 웹젠, 후야호, 드림에이지, 기어세컨드 등을 포함한 국내외 유수 게임사들은 이미 에어브릿지로 이 측정 체계를 갖추고 성과 향상에 활용하고 있어요.
게임 마케팅에 검증된 MMP(Mobile Measurement Partner, 모바일 성과 측정 파트너)인 에어브릿지와 앞으로 다가올 다양한 이벤트도 함께 준비해보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 에어브릿지 AI 실전 활용 A to Z: 파일럿·MCP로 마케팅 워크플로우 자동화하기
URL: https://www.airbridge.io/ko/blog/airbridge-ai-pilot-mcp-webinar-recap
Language: Korean
Published: 2026-07-15
Author: Jaehyuk Kim
에어브릿지 AI(Airbridge AI)가 출시된 지 어느덧 3개월이 지났는데요. 매일 아침 대시보드를 열어 리포트를 정리하던 시간을 에어브릿지 파일럿에게 질문하는 것으로 대신하고, 어렵게 느껴졌던 SDK 설치는 온보딩 파일럿이 도와주고 있어요. ChatGPT와 클로드(Claude)에 MCP를 연결해 쓰는 고객사도 늘어나고 있어요.
고객사가 실제로 에어브릿지 AI를 어떻게 사용하는지 Airbridge AI Deep Dive 웨비나에서 여러 사례를 공유했는데요.
AB180 Customer Success 팀의 최혜린 님과 장소영 님, 이다인 님이 발표한 실전 활용 팁과 고객사 도입 사례를, 행사에 참여하지 못한 마케터를 위해 핵심만 정리해 드릴게요.
**📌 핵심 포인트**
- 에어브릿지 MCP(Model Context Protocol, AI 어시스턴트가 외부 데이터·도구에 자연어로 접속하는 프로토콜)를 도입한 온라인 패션 플랫폼 A는 일간 리포팅 시간이 1시간에서 5-10분으로 줄고, 분석과 기획에 시간을 더 투자할 수 있었어요.
- 에어브릿지 AI는 온보딩 파일럿, 에어브릿지 파일럿, 에어브릿지 MCP 3가지로 구성되며 자유도의 진입 장벽을 자연어로 풀고 도구 간 워크플로우를 자동화해요.
- 가격 비교 쇼핑 앱 B는 데일리 루틴 자동화로 담당자가 없어도 매일 아침 리포트가 공유되고, 매체 최적화 테스트 횟수가 월 1-2회에서 월 3-6회로 늘었어요.
- 커플 캘린더 앱 C는 담당자 퇴사로 인수인계가 끊긴 상황에서 에어브릿지 MCP를 안전망으로 활용해 UA 분석과 의사결정까지 도달했어요.
- AI 시대 마케터의 새 역량은 "어떤 도구를 쓰는가"가 아니라 "맥락 데이터를 어떻게 명문화하고 워크플로우를 어떻게 설계하는가"로 바뀔 거예요.
## **에어브릿지 파일럿: 데이터를 찾던 마케터가 데이터를 받아보게 됐어요**
최혜린 님은 에어브릿지의 특장점을 자유도로 짚었는데요. 가장 많이 활용하는 액츄얼스 리포트(매체별 상세 성과 리포트)만 봐도 여러 개의 메트릭들을 사용자가 자유롭게 설정할 수 있어요.
하지만 자유롭게 원하는 대로 설정할 수 있다는 장점이 있지만 "어디서부터 뭘 봐야 할지 모르겠다"는 피드백도 함께 받아왔다고 해요. 자유도와 막막함의 중간 지점을 메꿔줄 해결책이 필요했죠.
**그리고 그 문제점을 명쾌하게 풀어준 게 바로 에어브릿지 파일럿(Airbridge Pilot)이었다**고 해요. 대시보드에서 자연어로 질문하면 AI 분석 파트너인 파일럿이 데이터를 찾아 마케터에게 바로 보여줬기 때문이죠.
> “에어브릿지 AI는 MMP의 주 사용자인 마케터가 데이터를 찾아 이동하지 않고, 원하는 데이터가 마케터 앞으로 오게 만드는 AI 도구예요. 우리가 데이터 정리에 쓰던 시간을 줄여서 인사이트 도출과 의사결정에 집중할 수 있도록 돕는 거죠. 지금 버전에서는 온보딩 파일럿, 에어브릿지 파일럿, 에어브릿지 MCP 3가지를 제공하고 있어요.” 최혜린, AB180 CSG 게이밍 팀 리더
[에어브릿지 AI 3가지 도구 오버뷰 - 온보딩 파일럿·에어브릿지 파일럿·에어브릿지 MCP]
### **1\. 온보딩 파일럿이 마케터와 개발자 사이 SDK 핑퐁을 줄여요**
에어브릿지 파일럿에는 2가지 기능이 있어요. 첫 번째는 SDK 설치 단계에서 작동하는 온보딩 파일럿(Onboarding Pilot)이에요.
기존에는 마케터와 개발자가 가이드를 오가며 SDK를 설치·검증하느라 "이게 진짜 우리가 원하는 게 맞나?"라는 의문 속에서 며칠씩 핑퐁을 주고받았어요. 온보딩 파일럿은 이 과정을 AI 대화로 대체해요.
개발자가 아닌 팀원도 온보딩 파일럿과 대화하며 SDK를 설치할 수 있고, Android·iOS·React Native·Flutter·Expo 등 프레임워크별 맞춤 가이드와 iOS ATT 프롬프트 같은 특수 설정까지 안내해줘요.
가장 큰 장점은 설치 과정에서 에어브릿지 대시보드에 데이터가 어떻게 찍히는지 함께 확인할 수 있다는 점이에요. 며칠씩 걸리던 검증 사이클을 그 자리에서 끝낼 수 있어요.
[온보딩 파일럿 Before/After - 기존 SDK 설치 과정과 온보딩 파일럿 대화형 설치 비교]
### **2\. 대시보드를 보다가 에어브릿지 파일럿에게 질문하면 바로 차트를 보여줘요**
두 번째는 데이터 분석 도구인 에어브릿지 파일럿(Airbridge Pilot)이에요. 대시보드에서 일일이 필터를 설정하지 않아도 자연어로 묻기만 하면 파일럿이 데이터를 찾아 차트로 답해줘요.
다만 파일럿은 GPT처럼 메모리를 저장하지 않아요. 그래서 질문할 때 **결과물 포맷**(예: 일별 트렌드 증감률을 테이블로)과 **앱 사전 컨텍스트**를 함께 주면 답변 품질이 눈에 띄게 올라가요.
예를 들어 게임 마케터라면 이런 식으로 입력해보세요.
> _“우리 게임 장르는 방치형 RPG, KPI는 티어1 국가 D-Retention 40%. 현재 글로벌 론칭 2주 차고 ROAS 효율 보고 캠페인 국가를 확장하려고 해.”_
커머스라면 주력 SKU 카테고리·핵심 KPI·매체 비중 같은 도메인 지식을 짧게라도 제공하면 돼요.
[에어브릿지 파일럿 3단계 - 질문·분석·인사이트]
### **3\. 초기·인계·집행·회고 4가지 상황별 파일럿 활용 시나리오**
최혜린 님은 마케터가 처한 상황을 4가지로 나누고 각 상황에서 던질 만한 질문을 함께 제시했어요.
1. **MMP를 처음 쓰는 사용자라면** (툴 개념·언제·왜) _“어트리뷰션 윈도우와 룩백 윈도우는 어떤 로직으로 계산되나요?”, “11개 리포트 각각 언제 봐야 하나요?”_
2. **전임자에게 인계받았다면** (요약·설정·연동) _“지난 3개월 매체별 비중 변화 포함, 캠페인 성과 요약해줘”, “이 앱의 룩백 윈도우 설정은?”, “버티컬에서 많이 쓰는데 안 연동된 매체는?”_
3. **캠페인 집행 중이면** (영향 요인·프로드 변화) _“어제까지 잘 나오던 ROAS가 떨어진 이유는? 매체·캠페인·크리에이티브별로 분석해줘”, “신규 설치 중 프로드 의심 매체와 그 이유”_
4. **캠페인 회고와 다음 미디어 믹스** (회고·효율·예산 분배) _“지난달 회고 리포트 초안 만들어줘”, “최근 한 달 효율 좋았던/나빴던 크리에이티브”, “최근 2주 OS별 효율 + 다음 주 예산 분배 시나리오 제안해줘”_
4가지 시나리오의 공통점은 바로 '파일럿에게 데이터만 요청하지 말고 지금 처한 상황을 먼저 알려주라는 것'이에요.
"ROAS가 왜 떨어졌어?"라고 물으면 일반적인 원인만 나열해주지만, "이 앱을 인수인계받았는데 ROAS가 떨어졌어"라고 물으면 매체 비중 변화나 룩백 윈도우 설정처럼 인계 상황에서 놓치기 쉬운 것부터 짚어줘요.
[에어브릿지 파일럿 사전 컨텍스트 예시 - 원하는 결과물 포맷과 앱 상황 입력]
### **4\. 파일럿에 가장 많이 들어오는 문의 2가지: 데이터 0과 딥링크**
이다인 님은 에어브릿지 파일럿이 출시된 지 두 달이 지난 시점에 마케터가 가장 많이 묻는 질문 유형을 정리했는데요. 크게 데이터 수치 이상과 딥링크 동작 이슈 등 2가지였다고 해요.
#### **1) 데이터가 0으로 나올 때**
"CPP나 ROAS가 0으로 나와요"라는 질문이 가장 많이 들어오는데, 이런 지표는 매출·주문 이벤트가 없거나 그룹 바이(Group by) 설정에 따라 0으로 표시돼요.
여기서 핵심은 **맥락 있는 질문이 빠른 답을 만든다**는 점이에요. 사람끼리 대화할 때도 모호한 질문은 추가 핑퐁이 필요한 것과 똑같은 원리예요.
- **막연한 질문**: _“CPP/ROAS가 왜 다 0으로 나와요?”_ → 추가 핑퐁 필요
- **맥락 있는 질문**: _“이 앱에서 5월 1일~25일 동안 A 매체에서 CPP가 0인데 이유가 뭘까요? JSON 쿼리도 함께 첨부할게요.”_ → 한 번에 원인 분석
[데이터 수치 이상 FAQ - 파일럿에게 어떻게 물어볼까 (X vs O)]")[딥링크 이슈 FAQ - 파일럿에게 어떻게 물어볼까 (X vs O)]")
#### **2) 딥링크가 간헐적으로 이상할 때**
딥링크는 디바이스·환경별로 동작이 달라 원인 파악이 까다로워요. "iOS 사파리를 쓰고 메인 홈 배너 클릭 시 일주일 중 30% 정도 빈도로 발생해요" 같은 식으로 질문하면 파일럿이 원인을 찾고 **개발팀 전달용 요약본**까지 정리해줘요. "어떻게 설명해야 할지 모르겠는" 영역에 파일럿이 다리를 놓아주는 거죠.
## **에어브릿지 MCP: 대시보드 대신 클로드와 ChatGPT에서 자연어로 데이터를 조회해요**
장소영 님은 에어브릿지 MCP로 대시보드 밖에서 일어나는 변화를 짚었어요.
MCP 도입 전 마케터들은 매일 아침 대시보드에 접속해 저장된 리포트를 조회하고 전일·전주 데이터를 비교했어요. 특이점이 보이면 그 시점에 변경한 광고 에셋을 검토했죠. 이렇게 한 사이클을 돌리는데 30분에서 1시간이 걸렸어요.
MCP가 도입된 뒤로는 이 흐름이 완전히 달라졌어요. 클로드나 ChatGPT 같은 AI 솔루션에 에어브릿지 MCP를 연결하고 "이번 주 전반적인 퍼포먼스를 요약해줘"라고 질문하면 액츄얼스 리포트 설정부터 데이터 해석, 변경 시점 분석까지 처리해줘요.
> _“내가 갖고 있는 도구를 100% 활용한다는 거예요. 에어브릿지, 앰플리튜드, 슬랙, 노션, 지라처럼 여러 솔루션을 이미 쓰고 계실 텐데, MCP가 이 도구들을 하나의 대화 안에서 연결해주거든요. 따로따로 대시보드에 들어가던 패턴이 자연어 요청 하나로 합쳐지는 거죠.” 장소영, AB180 CSM_
[내 도구를 100% 활용 - 에어브릿지·앰플리튜드·슬랙·지라·노션·구글 시트 연결]
### **1\. 에어브릿지 MCP로 할 수 있는 3가지: 리포트 조회, 자동화, 협업**
장소영 님은 마케터가 에어브릿지 MCP로 할 수 있는 일을 크게 3가지로 정리했어요.
- **자연어로 리포트 조회:** "인스톨과 CPI를 보고 싶어"라고 물으면 MCP가 액츄얼스 리포트를 만들어줘요.
- **리포팅 자동화:** "매일 아침 9시 광고 채널별 성과를 슬랙으로 보내고 인사이트도 분석해줘"라고 설정하면 매일 그 시간에 리포트가 도착해요. 10분 단위 리포팅도 가능해 프로모션처럼 변화가 빠른 캠페인에 유용해요.
- **협업 자료 생성:** 발견한 이벤트 이슈를 지라 티켓으로 정리해 개발팀에 전달하는 흐름까지 한 대화 안에서 가능해요. "내가 본 데이터 + 원인 가설 + 봐야 할 코드 위치"를 묶어 티켓 초안을 만들어줘요.
### **2\. 에어브릿지 & 앰플리튜드 MCP를 함께 쓰면 풀퍼널 인사이트가 나와요**
MCP의 임팩트가 더 커지는 지점은 여러 도구를 연결할 때예요. 장소영 님은 커머스 앱을 가정한 시나리오로 설명했어요.
의류 서비스 플랫폼이 봄 시즌 의류와 신규 뷰티 카테고리 광고를 동시에 운영하는 상황을 가정해볼게요. 두 카테고리로 유입된 유저들의 이후 행동을 비교하고 싶은데, 기존에는 앰플리튜드에서 카테고리별 90일 LTV 차트와 재구매 주기 차트를 따로 그리고 머릿속에서 합쳐야 하죠.
에어브릿지와 앰플리튜드를 MCP로 연결하면 이 흐름이 훨씬 단순해져요.
1. **에어브릿지 MCP:** 광고 성과 (유입 규모, CPI, 회원 가입 전환율) 조회
2. **앰플리튜드 MCP:** 유입 이후 유저 행동 (크로스셀 카테고리 수, 재구매 주기, 90일 LTV) 분석
3. **결론:** 뷰티 쪽은 유입 단가가 조금 비싸지만 LTV가 더 높고 재구매 주기도 짧다는 인사이트
이렇게 두 도구를 오가며 따로 분석하던 일이 하나의 대화 흐름으로 합쳐지면서, 이런 분석을 바탕으로 예산 배정을 빠르게 판단할 수 있어요.
[Full Funnel 사례 - 커머스 앱 캠페인 A(시즈널 봄 의류)와 캠페인 B(신규 뷰티 카테고리) 비교] vs 캠페인 B(신규 뷰티 카테고리)")
### **3\. AI 활용의 시작점은 도구 연결이 아니라 데이터 설계예요**
MCP가 자동으로 리포팅만 해줘도 임팩트가 있지만, 그 임팩트를 더 키우려면 데이터 설계가 전제되어야 해요.
에어브릿지의 액션·라벨 필드는 액츄얼스 리포트를 나눠서 볼 때 쓰는 단위인데, MCP는 자연어 질문을 이 필드로 필터링해서 답을 만들어요.
사내에서 "상품 카테고리와 브랜드 단위로 데이터를 본다"고 정해져 있다면 액션 필드에 상품 카테고리 값, 라벨 필드에 브랜드 값이 담겨 있어야 MCP가 그 관점으로 정리해줘요. 그래서 사내 합의와 데이터 매핑이 없다면 데이터 설계 점검을 첫 번째 스텝으로 가져가야 해요.
[에어브릿지 택소노미 설계 - Event·Action·Label·Value 4축]
> _“AI 시대에서 데이터를 분석한다는 건 내가 갖고 있는 맥락 데이터를 잘 명문화하는 일이에요. 광고 효율 기준치, 매체별 특성, 상품별 KPI처럼 내가 머릿속에 갖고 있던 룰을 AI에게 미리 알려주면 그 룰을 바탕으로 데이터를 해석해줘요. 데이터 분석의 시작점이 도구가 아니라 맥락을 글로 정리하는 일로 옮겨가는 거죠.” 장소영, AB180 CSM_
## **MCP 도입 후, 마케터의 시간이 리포팅에서 분석과 전략으로 옮겨갔어요**
이다인 님은 에어브릿지 MCP를 도입한 세 곳의 마케팅 팀을 직접 인터뷰한 결과를 공유했어요.
### **1\. 온라인 패션 플랫폼 A: 단순 작업이 줄고 데이터 분석과 기획 시간이 늘었어요**
온라인 패션 플랫폼 A의 마케터는 매일 아침 리포트 정리에 약 1시간을 썼어요.
- 매체별 성과 리포트를 액츄얼스에서 수동으로 다운로드
- 매체·캠페인별 양식 맞추기
- 어제 대비 어디가 달라졌는지 체크
매일 반복하는 일이라 1시간이 결코 작은 비용이 아니었죠. 에어브릿지 MCP를 도입한 뒤에는 이 시간이 5-10분으로 줄었어요. 무엇보다 그 시간이 어디로 갔는지가 더 눈에 띄었어요. 단순 작업 시간이 줄어든 만큼 분석과 기획에 더 많은 시간을 투자할 수 있었다고 해요.
- 어떤 페이지에서 유저 유입이 많은지 확인
- 어디서 이탈이 발생하는지 확인 → 이탈 그룹에 보상형 쿠폰 액션
- 성과 좋았던 프로덕트를 앞단 페이지에 반영 → 선순환
이 사례에 대해 이다인 님은 이렇게 정리했어요.
_“로우 데이터를 뽑고 가공하는 일은 어떻게 보면 단순 작업이잖아요. 이 단순 작업을 MCP한테 시키고, 우리는 놓치기 쉬웠던 데이터를 더 확인하거나, 매체 성과를 개선하려면 어떻게 할까·어떤 테스트를 해볼까 같은 전략과 플래닝에 집중할 수 있게 됐어요.” 이다인, AB180 CSM 팀 리더_
### **2\. 가격 비교 쇼핑 앱 B: 담당자 없이도 매일 아침 리포트가 도착해요**
가격 비교 쇼핑 앱 B의 마케터는 리포트 작성에 약 2시간을 썼어요.
- 매체별 가입자 데이터 직접 확인
- 인앱 액션 데이터 직접 확인
- 리포트 정리 및 작성
더 큰 문제는 담당자가 부재 중일 때 다른 팀이 성과를 확인할 방법이 없어 의사결정이 미뤄진다는 점이었어요.
에어브릿지 MCP를 도입한 뒤 리포팅 시간은 절반(1시간)으로 줄었다고 해요. 더 의미 있는 변화는 데일리 루틴 자동화였는데요. 매일 오전 매체 성과 분석과 액션 제안이 자동으로 슬랙에 공유되면서 담당자가 없어도 다른 팀이 그 데이터를 보고 의사결정할 수 있게 됐어요.
확보된 시간은 매체 최적화 테스트로 이어졌어요. 월 1-2회 하던 테스트가 월 3-6회로 늘면서 "데이터 분석 → 가설 → 테스트 → 결과 → 다음 가설" 사이클이 빠르게 돌기 시작했고, 피로도 높은 매체를 방치하지 않고 타겟팅·소재를 계속 바꿔가며 최적화하는 흐름이 만들어졌어요.
### **3\. 커플 캘린더 앱 C: 인수인계 위기에서 MCP가 가이드가 됐어요**
커플 캘린더 앱 C의 사례는 좀 달라요. 당시 신규 담당자가 마주한 난관들이 너무 많았다고 해요.
- 이전 담당자 퇴사
- 인수인계 문서 부족
- 에어브릿지를 처음 접하는 상황
무엇보다 "에어브릿지에 대해 아무것도 모르는데 어떡하지"라는 막막함을 느꼈다고 하는데요. 이 상황에서 에어브릿지 MCP 출시 소식을 듣고 "기능을 몰라도 일단 시작해보자"라며 적극적으로 MCP를 사용하기 시작했고, 자연어로 질문하고 답을 받아가며 UA 분석과 의사결정까지 도달했다고 해요.
이 사례에 대해 이다인 님은 이렇게 짚었어요.
> _"MCP가 옆자리에 앉아 인수인계를 해주는 것처럼 서포트를 해주니 허들이 낮아지고 빠르게 의사결정까지 도달할 수 있었어요. 쓰다 보니 에어브릿지 공부도 자연스럽게 돼서 학습과 실무가 동시에 이뤄지는 일석이조의 효과도 있었고요." 이다인 · AB180 CSM 팀 리더_
담당자가 바뀌는 위기 상황에서 MCP가 안전망이 된 사례예요. 인수인계 문서가 충분하지 않은 조직이라면 MCP가 그 갭을 메워줄 수 있는 가능성을 보여줬어요.
## **이제 마케터가 답해야 할 질문은 '무엇을 결정할 것인가'예요**
> _“AI로 마케터가 줄인 시간을 본질에 집중해야 한다는 말을 많이 들어요. 그런데 그 본질이 뭔지에 대해서는 저도 고민이 많아요. AI가 시간을 만들어준다면, 그 시간으로 마케터는 무엇을 해야 할까요.” 최혜린, AB180 CSG 게이밍 팀 리더_
[웨비나 Summary - 모르겠으면 파일럿에게 물어보고, 반복되면 MCP로 자동화하세요]
이제 **마케터가 매일 해야 할 일은 데이터를 찾는 게 아니라 데이터로 무엇을 결정할지 고민하는 일**이에요. 마케터의 다음 경쟁력은 '무엇을 결정할 것인가'에 답할 수 있는가에서 갈릴 것이라 생각돼요.
여러분도 에어브릿지를 쓰고 있다면 에어브릿지 파일럿에게 "지난주 CPI가 가장 낮은 캠페인은?"과 같은 익숙한 질문 하나 던져보고 MCP를 연결하는 것을 시작해 보는 것은 어떨까요?
---
## 앱 리텐션을 높이는 딥링크 전략, 단계별로 정리했어요
URL: https://www.airbridge.io/ko/blog/deeplink-app-retention-strategy-step-by-step-guideline
Language: Korean
Published: 2026-07-06
Author: Team Airbridge
[**딥링크**](https://help.airbridge.io/ko/guides/understanding-deep-link)는 유저를 앱 내 특정 페이지로 연결하는 URL이에요. 앱이 설치된 유저라면 **특정 광고나 콘텐츠 링크를 클릭했을 때 해당 화면으로 이동**하죠.
만약 앱이 설치되지 않은 신규 유저라면 링크를 클릭할 시 스토어로 이동하게 되고, 설치 후 앱을 처음 열면 홈 화면을 보게 될 텐데요. 이렇게 되면 유저가 딥링크가 연결하려 했던 특정 페이지로는 가지 못하게 돼요.
[**디퍼드 딥링크**](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)는 이 구간을 처리하는 기술로, **스토어를 거치더라도 처음 약속한 화면으로 연결**되는 방식이죠.
UA 예산에 투자하고 있는데도 D1 리텐션이 좀처럼 오르지 않거나, 재방문율이 기대에 못 미치는 현상이 있으신가요?
이 때 광고에서 본 화면과 앱을 열었을 때 보이는 화면이 다른 건 아닌지 딥링크 설정 방식부터 점검해 보세요.
이 글에서는 딥링크와 디퍼드 딥링크가 온보딩부터 리인게이지먼트까지 리텐션의 각 단계에서 어떻게 활용되는지 살펴볼게요.
📌 **핵심 포인트**
- 딥링크를 설정했는데도 리텐션이 오르지 않는다면 유저 상태에 맞는 딥링크 유형을 쓰고 있는지 먼저 확인해야 해요. 신규, 기존, 휴면 유저는 앱 설치 여부가 다르기 때문에 작동하는 딥링크 유형도 달라져야 해요.
- 딥링크를 제대로 활용하면 광고에서 본 화면으로 바로 연결, 쿠폰 코드가 자동 적용, 재설치 후에도 윈백 오퍼 화면이 바로 열리게 할 수 있어요. 아래 단계별 활용법과 솔루션 선택 기준을 확인하고 우리 팀 캠페인에 적용해보세요.
- 딥링크는 설정만큼 측정도 중요해요. 캠페인별로 트래킹 링크를 구분하고 딥링크 클릭 이후 인앱 전환까지 데이터가 연결되는 구조를 갖추면 어떤 설정이 실제 리텐션에 영향을 미치는지 판단할 수 있어요.
## **온보딩부터 윈백까지, 딥링크가 개입하는 3개 구간**
딥링크가 관여하는 구간은 크게 3개로 나눌 수 있어요.
1. 신규 유저가 앱을 처음 여는 시점
2. 기존 유저가 캠페인 메시지를 받고 앱으로 돌아오는 시점
3. 앱을 삭제한 유저가 재설치 후 앱을 처음 여는 시점
각 구간마다 적합한 딥링크 유형이 다르기 때문에, 각각의 작동 방식을 먼저 이해해야 해요.
### **1\. 유입 유저에 활용할 수 있는 딥링크가 달라요**
딥링크는 3가지 유형으로 구분돼요. 모두 유저를 앱 내 특정 화면으로 연결하지만, 앱 설치 여부에 따라 작동 조건이 다르고 전달할 수 있는 정보의 범위도 달라요.
| 유형 | 작동 조건 | 전달 정보 | 주요 활용 |
| --- | --- | --- | --- |
| 다이렉트 딥링크 | 앱이 설치된 유저 | 목적지 URL | 기존 유저 재방문 유도 (푸시, 이메일, SMS) |
| 디퍼드 딥링크 | 앱 미설치 유저 | 목적지 URL + 클릭 시점 정보 서버 저장 | 신규 유저 온보딩, 휴면 유저 윈백 |
| 컨텍스추얼 딥링크 | 설치 여부 무관 | 목적지 URL + 채널·캠페인·추천인 등 파라미터 | 채널별 온보딩 개인화, 추천 프로그램 |
**다이렉트 딥링크(Direct Deeplink)** 는 앱이 설치된 상태에서만 작동해요. 푸시 알림, 이메일, SMS에서 특정 인앱 화면으로 바로 연결할 때 쓰이고, 리인게이지먼트 캠페인에서 가장 흔하게 적용되는 형태예요.
**디퍼드 딥링크(Deferred Deeplink)** 는 앱이 없는 유저를 위한 방식이에요. 유저가 링크를 클릭하는 시점에 목적지 정보를 서버에 저장해두고, 설치 후 앱을 처음 실행할 때 SDK가 해당 정보를 조회해 지정된 화면으로 안내해요. 신규 유저 온보딩뿐 아니라 앱을 삭제한 휴면 유저가 재설치할 때도 동일하게 작동해요.
**컨텍스추얼 딥링크(Contextual Deeplink)** 는 목적지 URL에 추가 파라미터를 얹어 전달해요. 컨텍스추얼 딥링크는 내부적으로 설치 여부에 따라 다이렉트 또는 디퍼드 방식으로 동작하는데요.
유저가 유입된 채널, 클릭한 캠페인, 추천인 코드와 같은 맥락 정보를 앱이 수신하고 활용할 수 있어요.
인플루언서 추천 링크를 클릭한 유저가 첫 실행 시 추천인 코드가 자동 적용된 화면으로 진입하거나, 채널별로 다른 온보딩 화면을 보여주는 방식이 대표적이에요.
### **2\. 유저 상태에 따라 딥링크를 다르게 써보세요**
유저 상태에 따라 앱 설치 여부가 다르기 때문에, 같은 딥링크를 써도 작동 방식이 달라져요.
상황에 맞는 유형을 선택해야 의도한 화면으로 연결할 수 있어요.
[신규·기존·휴면 유저 상태별로 다르게 적용하는 딥링크 유형 매트릭스]
- **신규 유저 (앱 미설치)**: 광고를 클릭하고 스토어를 거쳐 설치하는 과정에서 목적지 정보가 유실돼요. 디퍼드 딥링크를 설정해야 광고에서 약속한 화면으로 연결할 수 있어요. 채널별로 다른 첫 화면을 보여줘야 한다면 컨텍스추얼 딥링크를 함께 적용해요.
- **기존 유저 (앱 설치됨)**: 다이렉트 딥링크로 충분해요. 푸시, 이메일, SMS에 딥링크를 적용하면 메시지에서 언급한 화면으로 바로 연결돼요. 유입 채널 정보나 맞춤 경험을 더하고 싶다면 컨텍스추얼 딥링크를 써요.
- **휴면 유저 (앱 삭제 후 재설치)**: 신규 유저와 같은 상황이에요. 재설치 과정에서 목적지 정보가 유실되기 때문에 디퍼드 딥링크가 필요하고, 재설치 후 첫 실행 시 윈백 캠페인에서 제안한 화면으로 바로 연결할 수 있어요.
## **리텐션 단계별로 딥링크를 적용하는 방법**
딥링크 유형을 파악했다면, 각 리텐션 단계에 실제로 어떻게 녹여낼지 참고해 보세요. 같은 딥링크라도 유저의 온보딩에 쓸 때와 윈백 캠페인에 쓸 때 설계 방식이 달라질 수 있어요.
### **1\. 온보딩: 광고에서 약속한 것을 앱 첫 화면에서 보여줘요**
광고를 클릭한 유저가 앱을 열었을 때 홈 화면이 나오는 건, 레스토랑 예약을 하고 갔는데 자리가 없다는 말을 듣는 것과 비슷해요. 광고가 약속한 것과 앱이 보여주는 것이 다른 거죠.
딥링크를 실제 캠페인에 적용할 때는 에어브릿지 트래킹 링크와 함께 쓰게 돼요.
**트래킹 링크**는 유저가 어떤 채널에서 왔는지 추적하는 URL로, 딥링크와 결합하면 유입 측정과 화면 연결을 동시에 처리할 수 있어요.
| | 딥링크 Deep Link | 트래킹 링크 Tracking Link | 트래킹 링크 + 딥링크 결합 사용 |
| --- | --- | --- | --- |
| 역할 | 앱 내 특정 화면 연결 클릭한 유저를 앱 내 원하는 화면으로 연결 | 유입 경로 추적 유저가 유입된 채널과 광고 측정 | 화면 연결 + 유입 추적 두 가지를 동시에 처리 |
| 전달하는 정보 | 목적지 URL 유저를 보낼 인앱 화면 | 채널·캠페인 파라미터 유저가 유입된 채널과 클릭한 광고 소재 | 목적지 URL + 파라미터 화면 정보와 유입 정보 모두 처리 |
| 주요 활용 | 인앱 특정 화면 이동 상품 페이지, 이벤트 화면, 쿠폰 적용 화면 등 | 광고 성과 측정 설치·구매로 이어진 광고 확인 | UA 캠페인 전체 유입·추적부터 온보딩 화면 연결까지 한 번에 |
| 에어브릿지 설정 | [트래킹 링크] > [딥링크] 딥링크 정보 등록 (SDK 설정 필요) | 트래킹 링크 생성 시 채널·캠페인 파라미터 입력 | 딥링크 등록 후 트래킹 링크 생성 시 최종 목적지 ‘앱(딥링크)’ 선택 |
**1) 일반적인 상황**
이커머스 앱이 메타에서 ‘신규 가입 시 30% 할인’ 광고를 집행 중이에요. 유저가 광고를 클릭하고 앱을 설치한 뒤 처음 열면 홈 화면이 나와요. 할인 코드를 찾아 직접 입력해야 해요.
**2) 딥링크를 적용하면?**
디퍼드 딥링크를 설정해두면 설치 후 첫 실행 시 할인 코드가 자동 적용된 가입 화면으로 바로 연결돼요. 유저가 해야 할 일이 하나 줄어들 수 있죠.
채널이 여러 개라면 컨텍스추얼 딥링크를 함께 써요. 메타 광고로 온 유저와 인플루언서 추천 링크로 온 유저는 기대하는 게 다르기 때문에 각자의 유입 맥락에 맞는 첫 화면을 보여주는 것이 중요해요.
에어브릿지에서는 트래킹 링크를 생성할 때 딥링크를 함께 적용할 수 있어요.
대시보드 \[트래킹 링크\] > \[딥링크\]에서 딥링크 정보를 먼저 등록하고, 트래킹 링크 생성 시 최종 목적지로 '앱 (딥링크)'를 선택하면 돼요. 광고 소재별로 연결할 인앱 화면을 미리 정의해두고 캠페인마다 해당 링크를 적용해 보세요.
### **2\. 리인게이지먼트: 메시지가 의도한 화면으로 바로 연결돼요**
기존 유저에게 보내는 메시지는 대개 개인화되어 있어요. 그런데 그 메시지를 탭했을 때 홈 화면이 나온다면, 개인화한 의미가 절반은 없어지는 거나 마찬가지이죠.
**1) 일반적인 상황**
패션 앱이 ‘찜한 상품 재입고됐어요’라는 푸시를 보냈어요. 유저가 알림을 탭했는데 홈 화면이 나오게 된다면 해당 상품을 다시 찾아야 해서 창을 닫을 확률이 높아요.
**2) 딥링크를 적용하면?**
다이렉트 딥링크를 적용하면 탭 한 번에 해당 상품 페이지로 바로 연결돼요. 장바구니 담기까지의 단계가 줄어드는 거죠. 쿠폰 코드 자동 적용도 컨텍스추얼 딥링크로 구현할 수 있어요.
한 가지 주의할 점이 있어요. 유저가 알림을 탭할 때 앱이 백그라운드에서 실행 중인 경우와 완전히 종료된 경우 동작이 달라질 수 있어요. iOS와 Android 각각에서 콜드 스타트 상황을 별도로 테스트해두는 것이 좋아요.
### **3\. 윈백: 재설치 유저에게도 맥락을 이어줘요**
앱을 삭제했다가 광고를 보고 재설치까지 한 유저는, 어느 정도 다시 쓸 의지가 생긴 상태예요. 이 의지가 행동으로 이어지려면 첫 화면이 중요해요.
> \*_윈백(Winback): 앱을 삭제하거나 일정 기간 이상 사용하지 않은 유저를 다시 앱으로 데려오는 마케팅 전략. 신규 유저를 획득하는 것보다 비용이 적게 들고, 이미 앱을 경험한 유저를 대상으로 하기 때문에 재전환 가능성이 상대적으로 높음._
**1) 일반적인 상**
게임 앱이 ‘복귀 이벤트, 전설 아이템 증정’ 광고를 집행했어요. 유저가 광고를 클릭하고 앱을 재설치했는데 첫 화면은 홈이에요. 이벤트 페이지를 직접 찾아야 하는데, 찾지 못하고 이탈할 수 있어요.
**2) 딥링크를 적용하면?**
디퍼드 딥링크를 설정해두면 재설치 후 첫 실행 시 이벤트 페이지로 바로 연결돼요. 광고가 약속한 혜택을 재설치 직후 바로 확인할 수 있는 거죠.
[에어브릿지 기준으로 디퍼드 딥링킹의 유효 기간은 링크 클릭 후 1시간](https://help.airbridge.io/ko/guides/retargeting-with-deep-links)이에요. 광고를 클릭하고 1시간 안에 앱 설치와 첫 실행이 이뤄져야 작동하죠. 윈백 캠페인을 설계할 때 이 유효 기간을 미리 감안해서, 광고 클릭 이후 설치까지의 흐름을 최대한 짧게 만드는 게 좋아요.
## **딥링크를 시작하기 전에 점검할 것들**
딥링크를 설정했는데 작동하지 않거나, 설정은 됐는데 기대한 효과가 나오지 않는 경우 대부분 초기 설계 단계에서 빠진 게 있어요. 집행 전에 아래 항목을 먼저 확인해보세요.
### **1\. 광고 소재별로 랜딩 화면을 정의했나요?**
딥링크 설정에서 가장 먼저 해야 할 일은 앱 개발이 아니에요. **광고 소재별로 유저를 어느 화면으로 보낼지 정의하는 거예요.** 운영 중인 캠페인마다 이 광고를 클릭한 유저는 어느 화면으로 가야 하는지 목록으로 만들어두는 것이 시작이에요. 이 작업 없이 딥링크를 기술적으로만 설정하면 결국 모든 광고가 같은 홈 화면으로 연결돼요.
### **2\. 앱 미설치 유저 경로를 테스트했나요?**
디퍼드 딥링크는 앱이 없는 상태에서 클릭할 때만 의미가 있어요. 그런데 많은 팀이 이미 앱이 설치된 기기로만 테스트하고 넘어가요. **앱을 삭제한 클린 기기에서 광고 링크를 클릭하고, 설치 후 첫 실행 시 지정한 화면으로 연결되는지 확인**해야 해요.
### **3\. iOS와 Android를 각각 테스트했나요?**
**딥링크는 운영체제마다 작동 방식이 달라요.** iOS는 유니버설 링크, Android는 앱 링크를 기반으로 하고, 브라우저 환경에 따라서도 결과가 달라지죠. 카카오톡이나 인스타그램처럼 인앱 브라우저를 쓰는 환경에서는 별도 검증이 필요해요. OS별, 주요 인앱 브라우저별로 테스트 시나리오를 구분해서 확인하는 게 좋아요.
### **4\. 딥링크가 끊겼을 때 폴백을 설정했나요?**
**딥링크가 작동하지 않는 상황은 언제든 생길 수 있는데요. 주로 앱이 없거나, 링크가 만료됐거나, 브라우저 환경이 지원되지 않을 때**예요. 이때 유저를 어디로 보낼지 폴백을 설정해두지 않으면 링크가 그냥 끊겨요. 앱 스토어 페이지나 모바일 웹 랜딩 페이지를 폴백\*으로 지정해두는 것이 기본이에요.
> \*_폴백(Fallback): 딥링크가 정상적으로 작동하지 않을 때 유저를 대신 보내는 대체 목적지. 앱이 설치되어 있지 않거나 링크가 만료된 경우처럼 딥링크가 실행되지 않는 상황에 대비해 미리 설정해 두는 것._
### **5\. 딥링크 성과를 채널별로 측정할 수 있나요?**
딥링크를 쓰고 나서 **어떤 채널의 유저가 실제로 지정한 화면에 도달했는지, 그 이후 어떤 행동을 했는지 추적할 수 있어야** 해요. 이 데이터가 없으면 딥링크 효과를 검증할 방법이 없죠. 캠페인별로 트래킹 링크를 구분해서 생성하고, 인앱 화면 도달 이후 이벤트까지 연결해서 측정하는 구조를 갖춰두는 게 좋아요.
## **우리 팀에 맞는 딥링크 솔루션 고르는 법**
딥링크를 직접 구현할 수도 있어요. Android는 Google Play Install Referrer API를 활용해 어느 정도 구현이 가능하고, iOS도 자체 서버를 구축하면 불가능하지는 않아요.
다만 OS별로 구현 방식이 다르고, 카카오톡이나 인스타그램 같은 인앱 브라우저 환경까지 대응하려면 엔지니어링 리소스가 상당히 들어요. 그래서 대부분의 팀은 MMP나 딥링크 전용 솔루션을 쓰죠.
솔루션을 고를 때 확인할 기준은 크게 4가지예요.
### **1\. 파라미터 유연성**
[딥링크 솔루션 선택 기준 1: 파라미터 수 제한이 없는 유연성 도식]
컨텍스추얼 딥링크를 제대로 활용하려면 링크에 담을 수 있는 파라미터 수에 제한이 없어야 해요.
일부 솔루션은 파라미터 수를 제한해서, 매물 ID나 위치 정보처럼 세부 데이터를 담으려면 별도 해시 처리를 개발팀에 요청해야 해요. 링크 하나를 만드는 데 개발팀을 거쳐야 한다면 캠페인 실행 속도가 느려질 수밖에 없어요.
### **2\. 채널 커버리지**
[딥링크 솔루션 선택 기준 2: 웹투앱·CRM·오프라인 QR까지 커버하는 채널 범위 도식]
웹투앱, CRM, 오프라인 QR까지 딥링크가 작동해야 하는 채널이 다양해요. 특정 채널에서만 작동하거나, 인앱 브라우저 환경을 지원하지 않는 솔루션은 실무에서 금방 한계에 부딪힐 수 있어요.
운영 중인 채널 전체에서 딥링크가 끊김 없이 작동하는지 먼저 확인해야 해요.
### **3\. 어트리뷰션 연동**
[딥링크 솔루션 선택 기준 3: 딥링크·설치·인앱 이벤트가 하나로 연결된 어트리뷰션 연동 도식]
딥링크 클릭, 설치, 인앱 이벤트가 하나의 데이터로 연결돼야 어떤 캠페인이 실제 성과로 이어졌는지 볼 수 있어요.
딥링크 솔루션과 어트리뷰션이 분리되어 있으면 데이터를 따로 꿰어 맞춰야 하는 번거로움이 생기죠. 딥링크와 어트리뷰션이 통합된 솔루션이라면 이 과정이 자동으로 처리돼요.
### **4\. 실무 운영 편의성**
[딥링크 솔루션 선택 기준 4: 마케터가 대시보드에서 직접 링크를 만들고 수정하는 운영 편의성 화면]
마케터가 개발팀 없이 직접 링크를 만들고 수정할 수 있어야 해요. 캠페인 중에 랜딩 페이지가 바뀌거나 파라미터를 조정해야 할 때, 이미 배포된 링크를 재배포 없이 즉시 변경할 수 있는지도 확인해보세요.
## **광고 클릭부터 재방문까지, 에어브릿지 딥링크로 앱 리텐션을 높여 보세요**
[에어브릿지 딥링크 플랜 대시보드에서 온드·언드 미디어와 커스텀 채널의 딥링크를 관리하는 화면]
[**에어브릿지(Airbridge) 딥링크 플랜**](https://www.airbridge.io/ko/deeplink-plan)**은 이 4가지를 모두 충족**해요. **파라미터 수 제한 없이 링크를 생성할 수 있고, 웹투앱·CRM·오프라인 QR까지 커버**해요.
딥링크와 어트리뷰션이 통합되어 있어서 클릭부터 인앱 전환까지 하나의 리포트에서 볼 수 있고, 마케터가 대시보드에서 직접 링크를 만들고 실시간으로 수정할 수 있죠.
대시 보드에서 온드 미디어와 언드미디어, 에어브릿지와 연동되지 않은 채널에 사용된 딥링크와 트래킹 링크를 확인하고 관리할 수 있고요.
유저 유입부터 재방문, 윈백까지 딥링크를 전략적으로 활용하고 싶다면 지금 바로 데모를 신청해 보세요. 리텐션 개선을 위한 정확한 데이터 측정이 가능해질 거예요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 클릭부터 전환까지, 트래킹 링크 하나로 딥링크 캠페인 설계하기
URL: https://www.airbridge.io/ko/blog/tracking-link-deeplink-campaign-design-click-to-conversion
Language: Korean
Published: 2026-07-06
Author: Team Airbridge
여러 채널에서 동시에 딥링크 캠페인을 운영했는데 캠페인이 끝난 뒤 어떤 채널이 가장 효과적이었는지 비교하기 어려운 경우가 있어요. 인스타그램 광고, 카카오모먼트, 인플루언서 콘텐츠처럼 채널 수가 늘어날수록 이런 일이 자주 생기죠.
[**딥링크**](https://help.airbridge.io/ko/deeplink-guides/tracking-link-structure-and-parameters)가 하는 일은 유저를 앱의 특정 화면으로 보내는 것까지인데요. **특정 유저가 어느 채널을 거쳐 들어왔는지를 기록하는 일은 별도로 필요한데, 이 역할을 트래킹 링크가 맡아요.**
[**트래킹 링크**](https://help.airbridge.io/ko/guides/understanding-tracking-link)를 만들 때는 채널, 캠페인명, 소재명 같은 항목을 입력하는 칸이 따로 있고 이곳에 적어둔 값이 이후 리포트에서 성과를 나눠 보는 기준이 되고요. 링크를 만드는 시점에 해당 입력값을 얼마나 정확히 기재했는지가 추후 정확한 채널별 성과 비교를 가능하게 하죠.
이번 글에서는 **딥링크 캠페인의 클릭과 설치뿐 아니라, 그 이후에 일어나는 가입이나 구매 같은 인앱 행동이 어떤 채널에서 일어나는지 추적하는 방법과 그 사례**를 정리했어요.
**📌 핵심 포인트**
- 트래킹 링크의 채널 정보를 채우는 방식과 그 정보가 가입·구매까지 어떻게 이어지는지 참고해 보세요. 정확한 기입으로 캠페인 효과를 측정할 수 있어요.
- 채널 추적과 화면 개인화를 트래킹 링크 하나로 동시에 처리할 수 있어요. 블릿은 이 방식으로 가입 프로모션 성과를 10배 개선했어요.
- 지금 운영 중인 딥링크 캠페인에 바로 적용해볼 수 있도록 채널 유형별 점검 체크리스트도 함께 담았어요.
## **트래킹 링크를 채우는 방식은 두 가지로 구분해요**
[트래킹 링크가 캠페인 파라미터와 딥링크 목적지 두 가지 정보를 함께 담는 구조 도식]
트래킹 링크 안에는 2가지 정보가 같이 들어 있어요.
1. 하나는 **캠페인 파라미터로, 유저가 어떤 채널·캠페인·소재를 통해 들어왔는지 적어두는 값**이에요.
2. 다른 하나는 **딥링크 목적지로, 그 유저를 앱의 어느 화면으로 보낼지 정하는 주소**에요.
둘은 따로 움직이는데요. 캠페인 파라미터는 나중에 리포트에서 채널별 성과를 나눠 보는 데 쓰이고, 딥링크 목적지는 유저가 실제로 보게 되는 화면을 결정해요.
이 중에서 캠페인 파라미터를 채우는 방식은, 그 채널이 MMP와 연동되어 있는지에 따라 완전히 달라져요. 주로 연동 채널은 시스템이 대신 채워주고, 커스텀 채널은 마케터가 직접 입력해야 해요.
> \*_트래킹 링크는 에어브릿지만의 개념은 아니에요. MMP 마다 부르는 이름이 다르며, 화면이나 설정 방식도 제각각이에요. 이 글에서 설명하는 구체적인 설정 화면과 항목명은 에어브릿지 대시보드를 기준으로 정리했어요._
### **1\. 연동 채널: 매체가 자동으로 정보를 채우는 방식**
메타, 구글처럼 에어브릿지와 미리 연동된 광고 채널은 트래킹 링크의 채널 항목에서 해당 매체를 선택하면, 캠페인명·광고그룹명·소재명 파라미터가 **매체가 지정한 형식으로 자동 적용**돼요.
광고 세트나 소재는 캠페인이 진행되는 동안 계속 추가되고 바뀌는데, 이 변화를 마케터가 일일이 따라가며 손으로 입력하는 건 현실적으로 어려운 일인데요. 이로 인해 매체와 미리 연동해서 그 시점의 정보를 자동으로 받아오는 시스템이에요.
### **2\. 커스텀 채널: 마케터가 직접 입력하는 방식**
반면, 인플루언서 콘텐츠, 이메일, 자체 공유 링크처럼 **에어브릿지와 연동되지 않은 채널은 이 정보를 마케터가 직접 입력**해야 해요.
이때 채널명이나 소재명을 어떻게 적느냐가 그대로 리포트에 반영되기 때문에, 입력값이 정확하고 일관될수록 나중에 채널별 비교를 정확하게 진행할 수 있어요.
👉🏻[에어브릿지에서 트래킹 링크 생성하기](https://help.airbridge.io/ko/guides/creating-tracking-links-on-the-dashboard)
## **전환과 화면 연결, 값에 따라 역할이 달라요**
### **1\. 캠페인 파라미터: 가입과 구매까지 연결**
에어브릿지에서 트래킹링크의 도메인은 다음과 같이 구조화 돼요.
> https://abr.ge/@app\_name/channel\_name?parameter1 이름={parameter1 값}¶meter2 이름={parameter2 값}?
도메인은 abr.ge이며, app\_name은 에어브릿지에 등록한 서비스의 앱 이름, channel\_name은 트래킹 링크가 사용되는 광고 채널의 이름이에요.
유저가 이 링크를 클릭하면 채널 이름과 파라미터 값이 그 유저의 디바이스 정보와 함께 저장돼요. 단순히 클릭이 발생했다는 기록이 아니라, 이 디바이스는 이 채널, 이 캠페인을 통해 들어왔다는 연결이 만들어지는 거예요.
유저가 앱을 설치하고 처음 실행하면, SDK가 그 디바이스를 다시 인식해서 클릭 시점에 저장해둔 정보를 그대로 불러와요. 그래서 설치 이후 가입이나 결제처럼 앱 안에서 일어나는 행동도, 처음 클릭했던 채널과 캠페인에 자동으로 묶이게 돼죠.
이렇게 연결되면 클릭율은 높지만 가입으로 잘 이어지지 않는 채널과, 클릭은 적어도 가입과 결제로 꾸준히 이어지는 채널을 구분해서 볼 수 있어요.
### **2\. 딥링크 목적지: 유저마다 다른 화면으로 연결**
지금까지 다룬 채널·캠페인명·소재명은 모두 유저가 어떤 채널을 통해 들어왔는지를 구분하는 값이었어요.
**트래킹 링크 안에는 그 유저를 어디로 보낼지를 정하는 딥링크 목적지**도 같이 들어있는데, 이 값도 캠페인 파라미터처럼 동적으로 채울 수 있어요. 에어브릿지에서는 이 기능을 [**딥링크 매크로**](https://help.airbridge.io/ko/guides/retargeting-with-deep-links)라고 하며, 원리는 다음과 같아요.
[에어브릿지 딥링크 매크로가 유저마다 다른 화면으로 연결되는 원리 도식]
딥링크 목적지 안에 중괄호로 감싼 매크로 값을 넣어두면, 그 링크를 받는 유저마다 중괄호 자리에 서로 다른 값이 채워져서 서로 다른 화면으로 연결돼요.
매크로는 그 링크를 만들 때 미리 입력해둔 값에 따라, 유저를 어느 화면으로 보낼지 정할 수 있어요.
### **캠페인 파라미터와 딥링크 목적지, 한눈에 비교하기**
| | 캠페인 파라미터 | 딥링크 목적지 |
| --- | --- | --- |
| 담는 정보 | 채널, 캠페인명, 소재명 | 유저를 보낼 앱 화면 주소 |
| 채우는 주체 | 연동 채널은 매체가 자동, 커스텀 채널은 마케터가 직접 | 마케터가 설계 (매크로로 유저마다 다르게 설정 가능) |
| 리포트에서 하는 역할 | 채널·소재별로 성과를 나눠서 비교 | 유저마다 다른 화면으로 개인화된 경험 제공 |
## **블릿은 에어브릿지 트래킹 링크로 가입 프로모션 성과를 10배 개선했어요**
캠페인 파라미터와 딥링크 목적지를 같은 링크 안에서 함께 활용한 실제 사례를 확인해 보세요.
### **1\. 신규 가입 유저 확보가 핵심 과제였던 블릿**
[블릿](https://www.airbridge.io/ko/case-studies/bleet-deep-linking-ua-growth)은 국내 1위 직장인 커뮤니티 블라인드가 운영하는 직장인 데이팅 앱이에요. 2020년 11월 출시 이후 매년 100% 이상 성장했고, 직장 인증을 거친 사람과만 만날 수 있다는 점이 차별점이죠.
데이팅 서비스 특성상 신규 가입 유저 확보가 가장 중요한 과제였는데요. 하지만 채널별로 신규 가입 전환율이 얼마나 되는지 확인하는 데 상당한 시간이 걸렸고, 프로모션·광고·바이럴 캠페인·인앱 공유처럼 여러 경로로 들어오는 유저를 각각 따로 추적해야 했어요.
### **2\. 링크 하나로 초대 코드 자동 입력부터 채널 성과 추적까지**
블릿은 매월 진행했던 가입 프로모션을 **참가자 확인부터 경품 지급, 채널별 성과 측정까지 모두 트래킹 링크 하나로 처리**했어요.
**'초대하기' 링크를 누르면 가입 화면으로 연결되면서 초대 코드**가 자동으로 입력돼요. 동시에 **이 링크를 트래킹 링크로 만들어서, 그 유저가 어떤 채널을 통해 들어왔는지도 같은 링크로 추적**했어요.
프로모션뿐 아니라 디지털 광고, 바이럴 캠페인, 인앱 공유까지 모든 유입 경로의 링크를 같은 방식으로 운영했어요.
### **3\. 가입 프로모션 성과 10배 개선**
링크 하나로 프로모션 운영과 성과 측정을 동시에 처리하면서, 마케팅 전략과 예산을 채널별 성과에 맞춰 조정할 수 있게 됐어요. 그 결과 가입 프로모션 성과가 10배 개선됐어요.
## **Team42는 커스텀 채널용 트래킹 링크로 지하철 옥외광고 성과를 측정했어요**
옥외광고는 얼마나 많은 사람이 봤는지는 가늠할 수 있어도, 실제로 앱 설치까지 이어진 유저가 몇 명인지는 파악하기 어려운 채널이죠. Team42는 에어브릿지 커스텀 채널 트래킹 링크로 이 문제를 해결했어요.
### **1\. 옥외광고를 정기적으로 집행하지만, 성과 측정이 어려웠던 Team42**
[Team42](https://www.airbridge.io/ko/case-studies/airbridge-tracking-link-ooh)는 마피아42를 포함해 다양한 게임을 개발하는 게임사예요. 마피아42는 출시 이후 꾸준히 사랑받아 온 소셜 디덕션 게임으로, 올해 시즌 14를 맞이하며 다양한 직업과 역할을 추가했어요.
Team42는 브랜딩과 신규 유저 유입을 목적으로 지하철 역사 내 옥외광고를 정기적으로 집행하고 있어요. 에어브릿지를 도입하면서 광고를 본 유저가 바로 앱을 설치할 수 있도록 QR코드를 삽입하고, 그 전환을 추적하고자 했어요.
### **2\. 커스텀 채널 트래킹 링크로 만든 QR코드를 광고 소재에 삽입**
Team42는 에어브릿지 커스텀 채널 트래킹 링크를 생성하면서 채널명과 캠페인명을 직접 입력하고, 그 링크로 QR코드를 만들어 홍대역 지하철 광고 소재에 넣었어요. QR코드를 스캔한 유저는 앱 마켓으로 바로 연결됐고, 앱이 이미 설치된 유저는 앱 화면으로 직접 이동할 수 있었어요.
트래킹 링크에 입력한 채널 파라미터 덕분에, QR코드를 통해 들어온 유저는 리포트에서 다른 채널과 구분돼서 잡혔어요. 어떤 역사의 광고에서 설치가 발생했는지 소재 단위까지 나눠서 볼 수 있게 된 거예요.
### **3\. 측정하기 어렵던 옥외광고 성과에 가시성이 생겼어요**
에어브릿지 리포트에서 QR코드를 통해 발생한 앱 설치를 디지털 광고 채널과 같은 기준으로 나란히 비교할 수 있게 됐어요. 마피아42 마케터 최서윤 님은 "일반적으로 측정하기 어려운 옥외광고의 캠페인 효과를 명확하게 측정할 수 있어 마케팅 활동의 가시성이 크게 향상됐다"고 말했어요.
## **지금 운영 중인 딥링크 캠페인, 체크리스트로 점검해 보세요**
지금까지 다룬 내용을 캠페인 상황별로 정리했어요. 운영 중인 채널과 캠페인 유형에 맞는 항목부터 확인해보세요.
### **1\. 메타, 구글 같은 연동 채널 캠페인**
- 트래킹 링크에서 채널만 선택하고, 자동으로 채워지는 캠페인·소재 파라미터는 따로 손대지 않았나요?
- 매체와 미리 협의된 프리셋 파라미터(하위매체 등)를 임의로 바꾸지 않았나요?
- 운영 중인 광고 매체가 연동 채널 목록에 모두 포함되어 있나요?
- 광고 세트나 소재가 새로 추가될 때, 트래킹 링크를 따로 만들지 않아도 자동으로 반영되고 있나요?
### **2\. 인플루언서, 이메일, 자체 공유 링크 같은 커스텀 채널 캠페인**
- 채널명·캠페인명·소재명을 매번 같은 표기로 입력하고 있나요?
- 트래킹 링크를 만들 때 채널명을 빠뜨리지 않고 있나요?
- 공식 홈페이지, 소셜미디어, 이메일, 공유 링크, 친구 초대 링크처럼 직접 운영하는 채널의 유입 링크를 모두 트래킹 링크로 만들고 있나요?
- 팀 내 여러 담당자가 같은 채널의 링크를 만들 때도, 동일한 표기 규칙을 따르고 있나요?
### **3\. 초대 코드, 개인화 콘텐츠처럼 유저마다 다른 화면을 보여줘야 하는 캠페인**
- 딥링크 목적지에 매크로를 활용해서, 채널 추적과 개인화된 연결을 같은 링크로 처리하고 있나요?
- 유저마다 다른 화면으로 보내야 할 때, 링크를 일일이 따로 만들고 있진 않나요?
- 매크로로 개인화한 링크에서도, 채널 정보가 함께 추적되고 있나요?
- 캠페인을 시작하기 전, 매크로 값이 의도한 화면으로 정확히 연결되는지 테스트했나요?
### **4\. 캠페인 공통**
- 가입이나 구매 같은 인앱 이벤트가 트래킹 링크의 채널 정보와 연결되도록 SDK가 설정되어 있나요?
- 클릭과 설치뿐 아니라, 그 이후의 가입·구매 전환까지 채널별로 나눠서 보고 있나요?
- 채널별로 쌓인 데이터를 리포트에서 정기적으로 비교하고 있나요?
## **에어브릿지 트래킹 링크 관련 자주 묻는 질문**
### **Q1. 트래킹 링크의 캠페인 파라미터, 나중에 수정해도 되나요?**
네, 가능해요. 채널·캠페인명·소재명 같은 캠페인 파라미터는 링크를 만든 뒤에도 에어브릿지 대시보드에서 수정할 수 있어요. 다만 수정하기 전에 이미 쌓인 데이터까지 새 값으로 바뀌는지는 명확하게 안내되어 있지 않아요. 가능하면 처음 링크를 만들 때 정확하게 입력하는 게 안전해요.
### **Q2. 짧은 형태의 트래킹 링크(숏 링크)에도 매크로가 자동으로 반영되나요?**
긴 형태의 트래킹 링크(롱 링크)에서는 딥링크 안에 넣은 매크로 값이 트래킹 링크 파라미터에도 자동으로 반영돼요. 하지만 숏 링크에서는 이 연결이 자동으로 이뤄지지 않아서, 매크로 파라미터를 직접 입력해야 해요.
### **Q3. 채널 표기가 갈라진 채로 쌓인 데이터, 나중에 합쳐서 볼 수 있나요?**
네, 가능해요. 'kakao\_invite'와 'Kakao\_Invite'처럼 표기가 다르게 쌓인 데이터도, 에어브릿지의 커스텀 그룹바이 기능으로 하나의 채널명으로 묶어서 볼 수 있어요. 다만 애초에 표기를 통일해서 입력하는 게 더 정확해요.
### **Q4. 연동 채널 트래킹 링크의 자동 파라미터(하위매체 등), 직접 바꿔도 되나요?**
변경할 수는 있어요. 하지만 이 값은 채널과 에어브릿지가 미리 협의해서 정해둔 형식이라, 임의로 바꾸기보다는 그대로 쓰는 걸 권장해요.
### **Q5. 트래킹 링크의 채널 파라미터를 비워두면 어떻게 되나요?**
채널 파라미터 값이 없는 성과는 `$$default$$`라는 이름으로 한꺼번에 집계돼서, 어떤 채널에서 왔는지 구분할 수 없게 돼요. 다만 에어브릿지 대시보드에서 트래킹 링크를 만들면 채널 파라미터가 항상 자동으로 채워지기 때문에, 이런 경우는 보통 API 등으로 직접 트래킹 링크를 만들 때만 발생해요.
## **에어브릿지 트래킹 링크로 캠페인 성과를 개선해 보세요**
딥링크는 유저를 앱의 원하는 화면으로 보내며, 그 유저가 어떤 채널, 어떤 캠페인을 통해 들어왔는지는 트래킹 링크에 담아둔 정보가 결정하죠.
이 정보를 정확히 채워두면, 클릭은 많아도 가입으로 안 이어지는 채널과, 적은 클릭으로도 꾸준히 가입과 구매를 만드는 채널이 구분할 수 있을 거예요.
에어브릿지는 트래킹 링크 생성과 딥링크 기능을 모든 플랜에 기본 포함하여 제공합니다. 지금 데모 신청을 통해 운영 중인 캠페인의 트래킹 링크부터 점검해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 광고 채널은 늘렸는데 성과 안 나온다면? 게임 UA 채널 이렇게 설계해 보세요
URL: https://www.airbridge.io/ko/blog/game-marketing-ua-channel-performance-design-strategy
Language: Korean
Published: 2026-07-03
Author: Team Airbridge
> [**“이제는 처음부터 여러 채널에서 UA를 시작해야 해요. 몇 년 전만 해도 그렇지 않았지만, 지금은 완전히 달라졌어요.”**](https://www.airbridge.io/ko/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives) **\-** Matej Lancaric
모바일 게임 UA 전문 컨설턴트이자 게임 업계 팟캐스트 Two and Half Gamers의 호스트 Matej Lancaric는 최근 **게임 UA 채널 다각화의 필요성**에 대해 이와 같이 짚었어요. 광고 채널 경쟁이 그 어느 때보다 치열해진 지금, 한두 개 채널에만 의존하는 건 이미 리스크가 됐다는 맥락이었죠.
> Matej Lancaric: 8년 이상 글로벌 게임사들의 멀티채널 UA 전략을 컨설팅해 온 모바일 게임 UA 컨설턴트. 약 4,200만 유로 규모의 UA 캠페인을 흑자로 운영한 이력 보유.
실제로 많은 게임 UA 마케터가 Meta와 Google을 기본 채널로 두고 시작해요. 여기에 Tiktok, AppLovin, 리워드 채널까지 더하면 어느새 5개에서 10개 안팎의 채널을 동시에 운영하고 있어요.
채널이 늘어난 만큼 질문도 복잡해졌죠. 고민을 더 구체적으로 살펴보면, 아래와 같을 거예요.
- **설치 비용은 낮은데 실제 매출로 이어지는 유저가 없다. (CPI는 낮은데 ROAS가 안 따라온다.)**
- **특정 소재가 잘 됐는데, 왜 됐는지 모르니 다음 캠페인 기획에서 막힌다.**
- **채널마다 소재 전략을 다르게 가져가야 한다는 건 아는데, 실제로 어떻게 달라야 하는지 모르겠다.**
위와 같은 고민은 실제로 전환에 기여한 채널을 파악하고, 다음 캠페인 예산을 어디에 투자해야 할지 판단하는 문제로 귀결돼죠.
이 글에서는 게임 UA 채널별 특성과 조합 전략부터, 각 채널의 성과를 정확하게 해석하여 다음 기획에 반영하는 방법까지 함께 살펴볼게요.
**📌 핵심 포인트**
- 게임 UA 시장은 이미 볼륨 경쟁에서 효율 경쟁으로 넘어왔어요. D30 기준 100명 중 5명도 남지 않는 구조에서 많이 데려오는 것보다 오래 남는 유저를 데려오는 게 더 중요해요.
- AI가 소재를 만들고 알고리즘이 타겟팅을 대신하는 지금, UA 팀이 직접 통제할 수 있는 변수는 크리에이티브와 측정 두 가지로 좁혀졌어요. 채널 전략은 이 두 가지가 갖춰진 다음에 의미가 있어요.
- 최적의 채널 조합은 게임 장르와 운영 단계에 따라 달라지기 때문에, 이 글에서 제시한 기준을 참고해 본인의 상황에 맞게 설계해 보세요.
## **게임 UA 시장, 지금 무슨 일이 벌어지고 있나요?**
글로벌 모바일 시장 데이터 분석 기관 [Sensor Tower에 따르면 2025년 모바일 게임 다운로드는 전년 대비 7.2% 감소](https://gamedevreports.substack.com/p/sensor-tower-the-state-of-the-mobile)했어요. 같은 기간 글로벌 게임 CPI(Cost Per Install , 설치당 광고 비용)는 30% 올랐죠. 신규 유저를 확보하기는 점점 어려워지는데, 그 비용은 오히려 높아지고 있는 상황이에요.
### **1\. 다운로드는 줄었지만 매출은 올랐어요**
[2025년 모바일 게임 IAP 매출 상위 게임사 집중도 그래프]
흥미로운 건 다운로드가 줄었는데 매출은 늘었다는 거예요. [2025년 모바일 게임 IAP(In-App Purchase, 인앱 결제) 매출은 전년 대비 1.3% 증가한 817억 달러를 기록](https://gamedevreports.substack.com/p/sensor-tower-the-state-of-the-mobile)했어요. 앱 설치 수는 줄어도 기존 유저의 인앱 결제가 늘어나는 구조로 시장이 재편되고 있다는 신호이죠.
실제로 전체 IAP 매출의 92.5%가 상위 1% 게임사(퍼블리셔)에서 나왔는데요. 소수의 게임이 충성 유저를 단단히 붙잡고 매출을 독식하고 있다고 해석할 수 있어요.
### **2\. 볼륨 늘리기 보다 효율이 중요해졌어요**
[2025년 글로벌 게임 UA 광고비와 D30 리텐션 수치 그래프]
[2025년 글로벌 게임 UA 광고비는 250억 달러](https://www.emarketer.com/content/gaming-user-acquisition-spend-hits-25-billion-ai-fuels-creative-boom)에 달했어요. 많은 돈이 유저 확보에 쏟아졌지만, 설치 수는 오히려 줄어, ‘더 많이 쓰면 더 많이 성장한다’는 공식이 더 이상 통하지 않는다는 걸 데이터를 통해 알 수 있어요.
또한 Business of Apps에 따르면 [iOS 기준 모바일 게임의 D30 리텐션율은 평균 5%, 안드로이드는 2.6%에 불과](https://www.businessofapps.com/data/mobile-game-retention-rates/)해요.100명을 데려와도 30일 후에 남는 유저는 채 5명이 안 된다는 얘기죠.
볼륨을 늘리는 것만으로는 사실상 의미가 없어, 지금 게임 UA의 핵심은 효율이라고 할 수 있어요. 광고비 규모보다 유저의 잔존율과 LTV(Lifetime Value, 생애 가치)가 게임 UA의 핵심 지표로 자리 잡고 있어요.
### **3\. AI로 인해 크리에이티브 경쟁은 레드오션이 됐어요**
[2026년 상위 게임 광고주의 분기당 크리에이티브 변형 제작 수 증가 추이]
모두가 AI로 소재를 찍어내는 시대, 광고 피드 안에서 차별화는 점점 어려워지고 있어요.
[상위 게임 광고주들은 분기당 2,400개에서 2,600개의 크리에이티브 변형을 제작](https://www.socialsamosa.com/report/ai-fuels-gaming-ads-creative-output-rises-ad-impressions-up-2025-11022568)하고 있어요. 전년 대비 25~30% 증가한 수치예요.
소재 제작의 진입 장벽이 낮아진 만큼, 광고 피드 안에서 유사한 포맷의 소재가 넘쳐나고 있는데요.
Meta의 알고리즘 업데이트 이후, 예전에는 6~8주 러닝이 가능했던 소재가 지금은 2~3주 안에 피로도가 쌓여요. 소재를 더 빠르게 만드는 것만으로는 이 흐름을 따라가기 어려워요. 경쟁사도 같은 속도로 소재를 쏟아내고 있으니까요.
결국 지금 크리에이티브 경쟁에서 살아남으려면 특정 소재가 시장에서 왜 통했는지, 어떤 맥락에서 유저의 반응을 이끌어냈는지를 이해하고 다음 기획에 연결해야 해요.
## **게임 UA를 위한 채널, 어떻게 쓸 수 있나요?**
채널마다 유저를 확보하는 방식이 달라요. 채널의 작동 원리를 이해하지 못한 채 예산을 배분하면, 동일한 광고비로도 전혀 다른 결과가 나올 수 있어요.
아래 내용은 각 채널의 특성과 강점을 이해하기 위한 기준이에요. 게임의 목표 지표와 상황에 따라 최적의 조합은 달라질 수 있어요.
### **1\. 가장 넓은 유저풀, Meta와 Google**
- **강점**: 가장 넓은 유저풀, 대규모 볼륨 확보에 유리
- **성과 변수**: 소재 품질과 공급 속도
- **적합한 상황**: 신규 론칭 초기 유저 확보, 빠른 스케일업
**Meta**는 장르를 막론하고 게임 광고 집행액 1위 채널을 유지하고 있어요.
2025년 10월 전면 도입된 Andromeda 알고리즘 업데이트 이후 정교한 타겟 설정보다 소재 자체가 알고리즘의 핵심 변수로 자리 잡았는데요. Andromeda는 소재의 시각 요소, 텍스트, 오디오를 분석해 유저 행동 신호를 기반으로 최적 오디언스를 스스로 탐색해요.
[Andromeda 구조에 맞게 캠페인을 운영한 광고주들은 기존 방식 대비 ROAS가 20~35% 높게 나왔고](https://www.1clickreport.com/blog/meta-andromeda-update-2025-guide), [한 광고 세트 안에 25개의 다양한 소재를 운영했을 때 전환율이 17% 높고 비용은 16% 낮았다는 테스트 결과](https://segwise.ai/blog/meta-andromeda-update-creative-strategy-2026)도 있어요.
게임 UA 전문 에이전시 AppAgent에 따르면, 스케일업 단계에서 월 15~30개 이상의 신규 소재를 테스트하지 못하는 팀은 이미 뒤처지고 있다고 짚어요.
**Google**은 YouTube Shorts 중심으로 광고 포맷이 전환되면서 게임 앱 광고의 CPI 효율이 개선되는 흐름이 나타나고 있어요. Meta와 달리 검색 의도와 YouTube 시청 맥락을 결합한 타겟팅이 강점이라, 특정 게임 장르나 키워드에 관심을 보인 유저에게 도달하는 데 유리해요.
Google UAC(Universal App Campaign)는 광범위한 타겟 설정과 다양한 소재 조합을 넣을수록 알고리즘이 더 효과적으로 최적화하는 구조라고 할 수 있죠.
다만, 두 채널 모두 초기 학습 기간인 2~4주 동안 성과가 낮게 보일 수 있어요. 이 구간에서 예산을 섣불리 줄이면 알고리즘 학습이 리셋되고, MMP와의 전환 이벤트 연동이 정확하지 않으면 엉뚱한 유저를 타겟팅하게 될 수 있으니 주의하세요.
소재 다양성이 부족한 경우에도 알고리즘 학습 속도가 느려져 성과가 정체될 수 있다는 점도 인지하세요.
### **2\. 게이머 데이터로 정밀 타겟팅하는 AppLovin**
- **강점**: 게임 행동 데이터 기반 고정밀 타겟팅, iOS UA에서 압도적 점유율
- **성과 변수**: MAX 미디에이션 연동 여부, 전환 이벤트 데이터 밀도
- **적합한 상황**: iOS 중심 캐주얼·미드코어 게임, ROAS 최적화 캠페인
**AppLovin**은 지난 2년간 게임 UA 시장에서 가장 빠르게 영향력을 키운 채널 중 하나로, 2026년 1분기 기준 iOS 광고 수익 점유율 39%로 1위를 기록하고 있어요.
AppLovin의 핵심 강점은 자체 미디에이션 플랫폼인 MAX에서 나와요. MAX는 게임사가 여러 광고 네트워크를 하나의 플랫폼으로 통합 관리하면서 광고 수익을 최적화하는 도구인데, 현재 모바일 미디에이션 시장의 약 60%를 차지하고 있죠.
AppLovin은 이 플랫폼을 통해 게임 내 유저 행동 데이터를 방대하게 축적하고, 소셜 네트워크나 다른 플랫폼이 접근할 수 없는 인앱 행동 신호와 입찰 데이터를 기반으로 고가치 유저를 탐색해요.
중요한 건 AppLovin이 MAX를 수익화 플랫폼으로 연동한 게임사에만 ROAS 최적화 캠페인을 허용한다는 거예요.
MAX 없이 AppLovin UA만 운영하면 플랫폼의 핵심 강점을 활용하지 못하는 셈이죠. AppLovin 공식 가이드라인에 따르면 알고리즘이 제대로 학습하려면 캠페인당 하루 10건 이상의 전환 이벤트가 필요해요.
다만 초기 학습 기간에 성과가 낮게 보인다고 예산을 섣불리 줄이면 알고리즘 학습이 멈출 수 있어요. MAX를 연동하지 않은 상태에서는 ROAS 최적화 캠페인 운영이 제한될 수 있다는 점도 미리 확인해두는 게 좋아요.
### **3\. 소재가 타겟팅을 대신하는 TikTok**
- **강점**: 크리에이티브 기반 자연 확산, 상대적으로 낮은 CPI
- **성과 변수**: 소재 교체 속도, UGC·크리에이터 콘텐츠 활용 여부
- **적합한 상황**: 하이퍼캐주얼·캐주얼 장르, 크리에이티브 실험 속도가 빠른 팀
**TikTok**은 타겟을 정교하게 설정하는 것보다 소재 자체가 유저를 끌어당기는 힘이 성과를 결정하는 채널이에요. 알고리즘이 소재를 읽고 반응할 유저를 스스로 찾아가는 구조라, 크리에이티브가 곧 타겟팅이에요.
[2026년 기준 TikTok의 D7 리텐션 지수는 108로 Meta의 94보다 높고](https://benly.ai/learn/tiktok-ads/tiktok-ads-gaming-mobile), [Meta와 TikTok 대비 CPI가 20% 이상 낮게 나오는 경우도 보고](https://benly.ai/learn/tiktok-ads/tiktok-ads-gaming-mobile)되고 있어요.
UGC 스타일 소재와 크리에이터 콘텐츠 형태의 광고가 특히 높은 전환율을 보이는데, 실제로 브랜드 제작 광고보다 크리에이터 콘텐츠를 활용한 캠페인이 더 높은 참여율을 만드는 경향이 있어요.
다만 소재 피로도가 빠른 플랫폼 특성상 크리에이티브 교체 주기를 짧게 가져가야 해요.
또한 TikTok의 규제 리스크도 인지할 필요가 있어요. 만약 TikTok이 전체 UA 예산의 15% 이상을 차지하고 있다면, 대안 채널을 미리 준비해두는 게 안전할 거예요.
### **4\. 설치가 아닌 플레이로 과금하는 리워드 채널**
- **강점**: 높은 유저 품질, 설치 후 잔존율과 LTV가 상대적으로 높음
- **성과 변수**: 리워드 조건 설계, CPA 이벤트 설정
- **적합한 상황**: 유저 품질을 우선시하는 게임, 리텐션 지표 개선이 필요한 시점
**리워드 채널**은 유저에게 앱 설치, 회원가입, 광고 시청 등 특정 미션을 수행하면 포인트나 쿠폰 같은 보상을 제공하여 게임 플레이를 유도하는 방식이에요.
리워드 채널은 오랫동안 ‘리워드만 챙기고 이탈하는 체리피커 유저를 확보한다는 한계가 지적돼 왔는데요.
이를 보완하기 위해 최근에는 단순 설치 완료에 리워드를 지급하는 방식에서, 레벨 3, 5, 10 달성 등 게임 내 특정 행동 단계마다 리워드를 지급하는 방식의 멀티리워드 캠페인으로 진화하고 있어요.
레벨 달성 조건이 붙은 캠페인 구조에서는 리워드를 받기 위해 게임을 일정 수준 이상 플레이해야 하므로 설치 직후 이탈하는 유저보다 게임을 더 깊게 경험한 유저가 확보돼요.
이로 인해 볼륨보다 유저 품질을 우선시하는 게임사, 혹은 D7·D30 리텐션 지표를 개선해야 하는 시점에 전략적으로 활용하기 좋은 채널이죠.
다만 리워드 조건 설계가 잘못되면 리워드만 챙기고 이탈하는 유저가 늘어날 수 있어서, 게임의 핵심 루프와 연결된 CPA 이벤트 설정이 중요해요.
## **게임 UA 채널, 조합으로 효과를 극대화할 수 있어요**
채널 조합에 정답은 없지만, 게임 장르, 목표 지표, 운영 단계에 따라 다르게 적용해 볼 수 있죠. 정해진 공식보다 각 채널의 역할을 이해하고 상황에 맞게 설계해 보세요.
### **1\. 론칭 초기: 볼륨보다 학습을 우선으로 하세요**
신규 게임 론칭 초기에는 Meta와 Google을 기본 채널로 시작하는 게 일반적이에요. 두 채널 모두 광범위한 유저풀을 가지고 있고, 알고리즘 학습에 필요한 전환 데이터를 빠르게 쌓을 수 있죠.
이 단계에서의 목표는 스케일업이 아니라 어떤 소재가 어떤 유저에게 통하는지 파악하는 거예요. 예산을 크게 집행하기보다 3~5개의 소재 컨셉을 다양하게 테스트하면서 CPI, D1·D7 리텐션, 첫 결제 전환율을 함께 봐야 해요.
- 추천 조합: Meta + Google → 소재 3~5개 컨셉 테스트 → D1·D7 리텐션 확인 후 스케일업 판단
### **2\. 스케일업 단계: 검증된 지표를 바탕으로 채널을 확장하세요**
CPI, D7 리텐션, 첫 결제 전환율이 안정적으로 나오기 시작했다면 채널을 확장할 준비가 된 거예요. 이때 모든 채널을 한꺼번에 붙이기보다 게임의 플랫폼과 상황에 맞게 단계적으로 확장하는 게 중요해요.
iOS 중심 게임이라면 AppLovin을 우선으로 고려해 볼 만해요. 앞서 Meta와 Google에서 쌓은 전환 데이터가 AppLovin 알고리즘 학습의 토대가 되거든요.
TikTok은 크리에이티브 파이프라인이 갖춰진 팀에 적합해요. 주당 2~3개의 신규 소재를 지속적으로 공급할 수 있는 체계가 없다면 TikTok 운영은 오히려 리소스 낭비가 될 수 있어요.
- 추천 조합: Meta + Google 기반 → AppLovin 추가 (iOS 우선) → TikTok은 크리에이티브 파이프라인 갖춰진 후
### **3\. 유저 품질이 중요한 시점: 리워드 채널을 전략적으로 활용하세요**
CPI는 낮은데 설치 이후 게임에 남는 유저가 적다면, 리워드 채널을 보완적으로 추가하는 걸 고려해볼 수 있어요.
리워드 채널은 게임을 직접 플레이한 유저를 데려오는 구조라, 다른 채널로 확보한 유저보다 게임에 더 오래 남는 경향이 있어요. 볼륨 채널로 규모를 확보하면서 리워드 채널로 유저 품질을 보완하는 조합이에요.
- 추천 조합: 기존 채널 + 리워드 채널 병행 → D7·D30 리텐션 개선 여부 확인 후 예산 조정
### **4\. 채널 조합은 도달 범위를 설계하는 관점으로 접근하세요**
채널을 여러 개 운영하는 걸 단순히 리스크 분산으로만 보는 시각이 있어요.
하지만 채널마다 도달하는 유저가 달라요. Meta와 Google이 닿은 유저풀 바깥에 AppLovin, TikTok, 리워드 채널이 닿는 유저가 따로 존재하거든요. 채널을 추가한다는 건 기존 채널이 가져오지 못한 유저를 새로 확보한다는 의미예요.
이로 인해 모든 채널을 동시에 같은 비중으로 운영하는 건 효율적이지 않아요. 핵심 채널에 예산을 집중하고, 보완 채널은 테스트 규모로 병행하면서 성과를 보고 조정하는 방식이 현실적이에요.
## **그럼에도 잊지 말아야 할 게임 UA의 기본**
채널이 늘어나고 AI가 소재를 만들어주는 시대에도 변하지 않는 게 있죠. 광고 플랫폼의 알고리즘이 타겟팅과 입찰 최적화를 대신해주는 지금, UA 팀이 직접 통제할 수 있는 변수는 사실상 두 가지로 좁혀졌어요. 크리에이티브와 측정이에요.
### **1\. 채널보다 게임의 지표를 먼저 확인하세요**
채널을 고르기 전에 게임의 핵심 지표를 먼저 확인해야 해요. D1 리텐션이 장르 평균에 못 미치거나 게임의 주요 플레이 구간에서 이탈이 집중되고 있다면, 어떤 채널을 써도 UA 예산은 새는 구조예요.
실제로 D7 리텐션이 10% 미만인 게임은 UA 예산을 늘릴수록 손실이 커지는 경우가 많아요. 채널 전략은 게임이 유저를 붙잡을 수 있는 상태임을 확인한 이후에 설계해야 해요.
### **2\. 크리에이티브를 채널보다 우선하세요**
2026년 현재 Meta, Google, TikTok, AppLovin 모두 알고리즘이 타겟팅을 대신해요. UA 팀이 직접 개입할 수 있는 가장 강력한 변수는 소재예요.
어떤 채널을 쓰더라도 소재의 품질과 다양성이 성과를 결정해요. 채널을 바꾸기 전에 소재를 먼저 점검해야 하는 이유예요.
### **3\. 성과를 측정하는 기준이 없으면 채널 조합도 의미 없어요**
채널을 잘 고르고 소재를 잘 만드는 것만큼, 그 성과를 정확하게 측정하는 인프라를 갖추는 게 중요해요. 실제로 높은 퍼포먼스를 내는 팀을 보면 더 나은 소재나 캠페인 운영보다 성과 측정 인프라가 압도적으로 잘 갖춰져 있어요.
CPI만 보는 게 아니라 D7·D30 리텐션, LTV, 채널별 ROAS를 함께 추적해야 다음 예산 배분 기준이 생기고, 채널 조합도 의미를 가져요.
채널을 늘릴수록 각 채널이 실제로 어떻게 기여했는지 파악하는 게 더 중요해져요. 그 측정이 정확해야 다음 캠페인 기획이 감이 아닌 데이터에서 시작할 수 있을 거예요.
## **글로벌 스탠다드에 부합하는 MMP로 성과를 측정하세요**
채널을 늘릴수록 각 채널이 실제로 어떻게 기여했는지 파악하는 게 더 중요해지는데요. 측정이 정확해야 다음 캠페인 기획이 데이터에서 시작할 수 있으니까요.
[에어브릿지 대시보드에서 채널별 기여도와 리텐션 지표를 확인하는 화면]
[에어브릿지는 구글, 메타, 틱톡 등과 공식 파트너십을 맺은 전 세계 7개 MMP 중 하나](https://www.airbridge.io/ko)예요. 아시아에서는 유일하죠.
모바일, 웹, PC, CTV까지 통합 측정이 가능해 멀티 채널 환경에서도 **각 채널의 기여도를 하나의 대시보드에서 파악**할 수 있어요.
채널별 ROAS, D7·D30 리텐션, LTV까지 추적하면서 어떤 채널이 진짜 고가치 유저를 데려오는지 확인하고, 그 인사이트를 다음 캠페인 기획에 바로 반영할 수 있죠.
**Day-90 코호트 ROAS, 예측 LTV, 광고 사기 방지까지 추가 결제 없이 기본으로 제공**돼요.
## **유저에게 통하는 게임 UA 캠페인, 에어브릿지로 설계하세요**
게임 UA 채널 전략을 고민하는 팀이라면 에어브릿지와 함께 시작해 보세요.
에어브릿지는 멀티 채널 환경에서 각 채널의 기여도를 데이터로 보여주고, 그 데이터가 다음 캠페인 기획의 출발점이 될 수 있도록 도와줘요. 지금 데모를 신청하고, 게임 UA 성과가 어떻게 달라지는지 직접 확인해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 게임 마케터를 위한 출시 초기 90일 LTV 예측 가이드
URL: https://www.airbridge.io/ko/blog/game-marketing-ua-90-day-ltv-prediction-guide
Language: Korean
Published: 2026-07-03
Author: Team Airbridge
게임 UA 마케터의 하루는 데이터로 시작해요. 어제 집행한 캠페인의 리텐션(Retention, 유저 잔존율)을 확인하고, 채널별 광고비 소진 속도를 보며 그날 예산을 조정하죠. 성과를 측정할 때 한 달을 기다릴 여유는 없어요. 게임사는 출시 후 3일, 1주일, 2주일, 1개월 시점마다 마케팅 지속 여부를 결정해야 하니까요.
문제는 초반 지표만으로는 한계가 있다는 거예요. CPI(Cost Per Install, 설치당 비용)나 초기 리텐션으로 획득 비용과 잔존율은 알 수 있지만, 그 유저가 이후 인앱 구매(IAP)로 이어질지는 알기 어렵거든요.
장르마다 결제가 본격화되는 시점이 다르고, 같은 채널 유저라도 소재나 타겟팅에 따라 결제 패턴이 달라지기 때문인데, 결국 **수익을 만든 채널은 캠페인이 한참 지난 뒤에야 드러나요.**
이 시간 차이를 넘으려면, **초반 데이터를 채널별로 분리해 90일 후 수익으로 이어질 패턴을 미리 읽어야 해요.**
이 글에서는 그 방법으로 채널별 코호트(Cohort) ROAS를 분리해서 보는 방법과, 초반 데이터로 90일 LTV를 예측하는 방법을 소개해 드릴게요.
**📌 핵심 포인트**
- CPI로 채널 효율을 판단하고 있었다면, 같은 시점에 유입된 유저를 채널별로 묶어 시간 흐름에 따라 수익을 추적해보세요. 초반에는 비슷해 보였던 채널들이 D30 이후 전혀 다른 결과를 만들어낼 수 있어요.
- 전체 ROAS만 보고 있었다면, 채널별로 ROAS를 분리해서 확인해 보세요. 잘 되는 채널과 그렇지 않은 채널이 하나의 수치로 합산되면서 실제 성과 차이가 가려지고 있을 수 있어요.
- 캠페인 론칭 후 3개월을 기다리지 않아도 돼요. 초반 리텐션 데이터로 pLTV를 예측하면 캠페인 집행 초반에 어떤 채널이 장기 수익을 만들어낼지 미리 파악하고 예산을 조정할 수 있어요.
## **CPI가 낮은 채널을 다시 봐야 하는 이유**
### **1\. CPI가 낮은 채널이면 원하는 유저도 적을 수 있어요**
[CPI만 낮은 채널이 실제 원하는 유저를 데려오지 못하는 개념 도식]
\*_CPI: 사용자가 광고를 클릭하여 게임을 1회 설치할 때마다 광고주가 지불하는 마케팅 비용._
CPI는 **유저를 데려오는 비용을 채널별로 비교할 때 가장 많이 보는 지표**예요. CPI가 낮을수록 같은 예산으로 더 많은 유저를 확보할 수 있어서 비용 효율 지표로 많이 활용하죠. 하지만 설치 이후 그 유저가 게임 안에서 어떻게 행동하는지는 담고 있지 않아요.
이는 해당 지표가 소재 클릭율, 캠페인 목표 설정, 앱마켓 최적화 수준 등 다양한 변수의 영향을 받기 때문인데요. 같은 채널에서도 캠페인 목표를 설치 최적화로 설정하느냐, 구매 이벤트 최적화로 설정하느냐에 따라 CPI가 크게 달라지기도 하고요.
즉, CPI가 낮다고 해서 게임에 관심 있는 유저를 데려왔다는 신호로 보기는 어려워요.
특히 타겟 설정이 넓어질수록 게임에 관심이 높은 유저와 낮은 유저가 섞여서 유입될 수 있어요. CPI는 낮지만 실제로 게임을 즐기거나 결제로 이어지는 유저의 비율이 낮아질 수도 있고요.
이로 인해 CPI를 낮추는 방향으로 캠페인을 최적화할수록, **원하는 유저가 아닌 쪽으로 타겟이 열릴 수 있다는 걸 운영 경험으로 체감**하게 돼죠.
결국 CPI는 얼마에 유저를 데려왔는지 보는 지표이며, 그 유저가 게임에 얼마나 몰입하는지, 아이템을 구매하거나 정기적으로 결제하는 유저인지 파악하려면 호트(Cohort, 동일 시점에 유입된 유저 집단) 단위로 채널별 성과를 분리해서 봐야 해요.
### **2\. ROAS는 채널별로 분리해서 성과를 봐야 해요**
메타(Meta)와 애플 서치애즈(Apple Search Ads) 두 채널로 UA 캠페인을 동시에 운영하고 있다고 가정해 볼게요. 한 달 후 대시보드에 표시된 전체 D30 ROAS(Return on Ad Spend, 광고 수익률)는 120%예요. 손익분기점을 넘겼으니 캠페인이 잘되고 있는 것처럼 보이죠.
그런데 **채널별로 분리해서 보면 아래와 같은 상황**일 수 있어요.
| 채널 | 광고비 | D30 수익 | D30 ROAS |
| --- | --- | --- | --- |
| 메타 | 5,000만 원 | 4,000만 원 | 80% |
| 애플 서치애즈 | 5,000만 원 | 8,000만 원 | 160% |
| 전체 합산 | 1억 원 | 1억 2,000만 원 | 120% |
전체 ROAS는 120%로 괜찮아 보이지만, 메타는 아직 광고비를 회수하지 못하고 있어요. 광고비 5,000만원을 쓰고 4,000만원의 수익만 만들었으니, 두 채널의 성과가 합산되면 메타의 손실이 가려질 수 있죠.
이 상태에서 전체 수치만 보고 예산 배분을 유지하면, 손실을 내고 있는 채널에 예산을 투자하게 될 수 있어요.
소재 단위로 들어가면 차이는 더 커질 수 있는데요. 같은 메타 채널 안에서도 게임플레이 영상 소재로 유입된 유저와 인게임 재화 증정(소환권, 게임 코인 등)를 증정하는 보상형 소재로 유입된 유저는 초기 리텐션은 비슷해 보여도, D30 이후 결제 빈도와 결제 금액이 다를 수 있어요.
보상형 소재로 들어온 유저는 프로모션에 반응한 유저일 가능성이 높아서, 이후 정가 결제로 이어지는 비율이 낮은 경향이 있어요.
**채널별, 소재별로 유입된 유저를 시간 흐름에 따라 별도로 추적해야 어떤 캠페인이 실제로 수익에 기여하는지 파악**할 수 있어요. 이로 인해 코호트(Cohort, 동일 시점에 유입된 유저 집단) 단위로 성과를 분리해서 봐야 해요.
### **3\. 장르마다 다른 결제 타이밍도 변수에요**
어떤 게임의 ROAS를 봐야 하는 게임의 수익 구조에 따라 달라요. 크게 2가지 유형으로 나눌 수 있어요.
첫 번째는 **인앱 광고(IAA, In-App Advertising) 중심 수익 구조**예요. **하이퍼캐주얼 게임처럼 유저가 게임을 플레이하는 동안 광고를 노출해서 수익을 만드는 구조**이죠.
결제보다는 유저가 광고를 얼마나 많이 보느냐가 수익을 결정하기 때문에, 초반 유저 잔존율과 플레이 시간이 핵심 지표예요. 이 구조에서는 유저 이탈이 빨라서, **D0(당일)과 D1 시점과 같은 초기 지표**를 면밀히 모니터링해야 캠페인 방향을 잡을 수 있어요.
두 번째는 **인앱 구매(IAP, In-App Purchase) 중심 수익 구조**예요. **미드코어 RPG처럼 유저가 게임에 충분히 몰입한 뒤 인게임 재화, 캐릭터 또는 스킨 등을 직접 구매**하는 형태이죠.
유저가 게임의 재미를 충분히 경험하고 나서야 결제가 발생하기 때문에, **게임에 따라 고과금 유저가 드러나는 시점이 D30 이후인 경우**도 있어요.
기본적으로는 IAP 중심 게임에서 **D7 ROAS를 캠페인의 기본 성과를 확인하는 용도**로 활용하되, 채널의 장기 수익 가능성을 판단하려면 D30 이상의 코호트 데이터를 함께 봐야 해요.
반면, IAA 중심 게임은 초반 수일 안에 수익성이 확인되는 구조라 D30까지 기다리면 오히려 최적화 타이밍을 놓칠 수 있어요.
우리 게임이 어떤 수익 구조인지 먼저 확인하고, 그에 맞는 코호트 측정 기간을 정해야 채널별 성과를 정확하게 비교할 수 있어요.
| 수익 구조 | 대표 장르 | 핵심 지표 | 주요 측정 시점 |
| --- | --- | --- | --- |
| IAA 중심 | 하이퍼캐주얼 | 유저 잔존율, 플레이 시간 | D0, D1 |
| IAP 중심 | 미드코어 RPG | 유저 전환율 | D7(기본 확인), D30(최종 판단) |
## **90일을 기다리지 않고 채널 성과를 보는 방법**
### **1\. 코호트 ROAS로 채널별 장기 성과를 추적하세요**
전체 ROAS가 채널별 성과 차이를 가리는 문제를 해결하려면, 동일한 시점에 유입된 유저 집단을 채널별로 묶어서 시간 흐름에 따라 추적해야 해요. 이게 코호트 ROAS예요.
예를 들어 6월 1일에 메타 광고로 유입된 유저 집단과 애플 서치애즈로 유입된 유저 집단을 각각 따로 묶어요. 그리고 D7, D30, D60, D90 시점에 각 집단이 만들어낸 수익을 광고비와 비교해요.
이렇게 하면 초반에는 비슷해 보였던 두 채널이 시간이 지나면서 어떻게 수익이 달라지는지 보여요.
가령 D7 시점에 메타 코호트의 ROAS가 낮더라도, D30 추이를 보면서 이 채널이 장기적으로 수익을 만들 가능성이 있는지를 판단할 수 있죠. 반대로 초반 ROAS가 높은 채널이 D30 이후 급격히 떨어진다면, 단기 반응만 높은 유저를 데려온 채널일 수 있어요.
### **2\. 코호트 초반 데이터에서 장기 수익 신호를 찾아보세요**
코호트 ROAS가 유용한 지표라도, D90 데이터가 쌓이려면 실제로 90일이 지나야 하는데요. 그동안 캠페인은 계속 돌아가고 예산도 계속 쓰이죠.
결제 데이터가 충분히 쌓이기 전에도 채널 성과를 가늠할 수 있는 신호가 있어요. 결제 이전 단계의 인게임 지표를 함께 보는 거예요.
특정 레벨을 달성하거나 튜토리얼을 완료한 유저가 이후 결제로 이어지는 비율이 높다면, 이 행동 지표가 장기 수익의 조기 신호가 될 수 있어요. 즉, 결제 데이터가 충분히 쌓이기 전에도 어떤 채널이 결제 가능성이 높은 유저를 데려오는지를 가늠해볼 수 있죠.
다만 인게임 지표와 결제 사이의 상관관계는 게임마다 달라요. 우리 게임에서 어떤 행동이 결제와 연결되는지를 먼저 파악해야 이 방법이 의미 있어요. 데이터 포인트가 충분히 쌓이기 전까지는 방향성 있는 신호로만 활용하세요.
### **3\. 초반 리텐션 데이터로 장기 LTV를 예측해 보세요**
[초반 리텐션 데이터로 D90 LTV를 예측하는 곡선 개념 도식]
인게임 지표와 함께 활용할 수 있는 방법이 초반 리텐션 데이터를 바탕으로 장기 LTV를 예측하는 거예요.
D1, D3, D7 리텐션 수치가 쌓이면 유저가 시간이 지나면서 얼마나 남아 있을지를 곡선으로 추정할 수 있어요. 이 리텐션 곡선에 일평균 매출(ARPDAU, Average Revenue Per Daily Active User)을 곱하면 D90까지의 누적 LTV를 계산할 수 있어요.
- **LTV(D90) = Σ (D1~D90 일별 리텐션율 × ARPDAU)**
물론 게임 업데이트나 시즌 이벤트처럼 외부 변수가 생기면 예측값과 실제 수치 사이에 차이가 날 수 있어요. 예를 들어 D1 리텐션이 45%, D3가 28%, D7이 18%로 쌓인 시점에 이후 곡선을 추정하고, 일평균 매출이 100원이라면 D90까지 유저 한 명이 만들어낼 누적 수익을 미리 계산할 수 있어요.
## **에어브릿지 코호트로 90일 LTV 예측하기**
이 방법을 직접 구현하려면 코호트를 수동으로 집계하고 리텐션 곡선을 별도로 모델링해야 해서 공수가 꽤 들어요. 에어브릿지 레비뉴 리포트는 이 과정을 대시보드 안에서 한 번에 처리해줘요.
### **1\. 에어브릿지 레비뉴 리포트에서 채널별 코호트 ROAS를 확인하세요**
앞서 살펴본 것처럼 채널별 코호트 ROAS를 보려면 동일 시점에 유입된 유저 집단을 채널 단위로 분리해서 시간 흐름에 따라 추적해야 해요. [에어브릿지 레비뉴 리포트](https://help.airbridge.io/ko/guides/understanding-the-revenue-report)에서는 **별도 작업 없이 대시보드 안에서 바로 확인**할 수 있어요.
주요 활용 시나리오는 아래와 같으며, 조회 기간은 주기별로 설정할 수 있죠.
**주요 활용 시나리오**
- 광고 채널이나 캠페인별로 발생한 매출과 ROAS를 비교
- 유저당 평균 매출액(ARPU)과 결제 유저당 평균 매출액(ARPPU) 추이로 유저 매출을 분석
- 결제 전환율(PUR)로 채널별 광고 효율을 점검
- 예측 LTV(pLTV)로 미래 매출을 추정
**조회 기간**
| 주기 | 구간 단위 | 최대 구간 개수 |
| --- | --- | --- |
| 일별 | 1일 (24시간) | 181개 (Day 0 ~ Day 180) |
| 주별 | 주 시작 요일 기준 7일 | 53개 (Week 0 ~ Week 52) |
| 월별 | 1달 | 37개 (Month 0 ~ Month 36) |
그룹바이(Group By, 데이터를 묶을 기준)를 채널로 설정하면 메타, 애플 서치애즈, 구글 UAC 등 각 채널로 유입된 유저 집단별로 D7, D30, D60, D90 시점의 ROAS를 한 화면에서 비교할 수 있어요.
캠페인, 광고 소재, 국가, OS 단위로도 쪼개서 볼 수 있어서 어떤 조합이 장기 수익에 기여하는지 파악할 수 있죠.
IAP와 IAA를 통합해서 수익을 집계하기 때문에, 하이브리드 수익 구조인 게임에서도 채널별 실제 수익을 정확하게 비교할 수 있어요.
[게임 마케터를 위한 출시 초기 90일 LTV 예측 가이드 3.webp][에어브릿지 레비뉴 리포트에서 D1~180일 예측 LTV를 채널별로 확인하는 대시보드 화면]
### **2\. D3 데이터로 D90을 예측하는 에어브릿지 pLTV를 활용하세요**
앞서 초반 리텐션 데이터로 장기 LTV를 예측할 수 있다고 했는데요. 에어브릿지는 이 [**예측을 레비뉴 리포트 안에서 바로 확인할 수 있는 pLTV 기능으로 제공**](https://help.airbridge.io/ko/guides/predictive-ltv)해요.
별도로 리텐션 곡선을 모델링하거나 수식을 직접 계산하지 않아도 돼요. 조회 기간과 예측일수만 설정하면 채널별 pLTV를 대시보드에서 바로 보여요.
에어브릿지 pLTV는 아래 공식으로 계산해요.
- **pLTV = 예측 잔존일(Predictive Lifetime) × ARPDAU**
예측 잔존일은 에어브릿지가 현재까지 쌓인 리텐션 데이터를 바탕으로 유저가 앞으로 며칠 동안 더 재방문할지를 추정한 수치예요. 여기에 일평균 매출(ARPDAU)을 곱하면 유저 한 명이 앞으로 만들어낼 예상 수익이 나오죠.
레비뉴 리포트에서 예측일수를 1~180일 사이로 설정하면 해당 기간의 pLTV를 채널별, 캠페인별, 소재별로 확인할 수 있어요. 예를 들어 캠페인 집행 후 D3 시점에 예측일수를 90일로 설정하면, 아직 D90 데이터가 쌓이지 않은 시점에도 각 채널이 앞으로 만들어낼 수익을 비교할 수 있어요.
### **3\. pLTV를 참고해 예산을 조정해 보세요**
pLTV가 의미 있으려면 대시보드에서 수치를 확인하는 것에서 그치지 않고, 그 결과를 예산 결정에 실제로 연결해야 해요. 실전에서는 아래 순서로 활용해보세요.
- **D3~D7 시점에 채널별 pLTV를 비교해 보세요.**
캠페인 집행 초반에 pLTV가 높게 나오는 채널은 장기적으로 수익을 만들 가능성이 높은 유저를 데려오고 있다는 신호예요. 해당 채널에 예산을 더 집중해 볼 수 있어요.
- **CPI는 낮지만 pLTV도 낮은 채널은 주의가 필요해요.**
초반 반응만 높은 유저를 데려오고 있을 수 있으니, 소재나 타겟팅 방식을 점검해 보세요.
- **pLTV가 높은 채널에 예산을 집중하고, 낮은 채널은 예산을 줄이거나 소재를 교체해요.**
90일을 기다리지 않고도 초반 데이터를 근거로 캠페인을 조정할 수 있어요.
pLTV는 예측값이라 실제 수치와 차이가 날 수 있어요. 리텐션 패턴이 비교적 안정적인 캠페인에서는 pLTV가 잘 맞아요.
반대로 게임 업데이트, 시즌 이벤트, 신규 소재 교체가 잦은 시기에는 모델이 변화를 따라가지 못할 수 있으니 예측값을 보수적으로 해석하는 게 좋아요.
pLTV 단독으로 판단하기보다는 코호트 ROAS 추이와 함께 보면서 방향을 잡아 보세요.
## **D3 데이터로 3개월 후 수익을 만들 채널, 지금 확인해 보세요**
결국 마케터라면 어떤 채널이 장기 수익을 가져다줄지를 매일 질문할 수밖에 없는데요. 이제는 그 답을 캠페인이 끝나고 오랜 시간이 지날 때까지 기다리지 않아도 돼요.
**에어브릿지 레비뉴 리포트에서 채널별 코호트 ROAS와 pLTV를 살펴보면, 캠페인 초반 D3~D7 시점에 어떤 채널이 90일 후 수익을 만들어낼지 미리 파악**할 수 있어요. 예산을 더 투자해야 하는 광고 채널과 바꿔야 하는 소재를 데이터 기반으로 결정할 수 있죠.
90일을 기다리지 말고, 에어브릿지에서 지금 바로 시작해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 좋은 MMP는 무엇이 다를까? 마케팅 실무자가 도입 전 알아야 할 핵심 포인트
URL: https://www.airbridge.io/ko/blog/what-is-good-mmp-essential-features-for-marketer
Language: Korean
Published: 2026-06-24
Author: Team Airbridge
**MMP 도입하면 쉬워질 줄 알았던 마케팅, 현실은 왜 다를까요?**
[MMP(Mobile Measurement Partner)는 광고 캠페인의 성과를 추적하고 데이터를 분석하는 데 사용되는 마테크(MarTech) 솔루션](https://www.airbridge.io/ko/blog/what-is-mmp)을 말해요.
광고 채널별로 흩어진 데이터를 한곳에서 볼 수 있으며, 유저의 유입과 전환 등을 통합된 대시보드에서 파악할 수 있게 해줘요.
하지만 막상 MMP를 운영하다 보면 다양한 기능과 복잡한 고객 여정으로 인해 운영에 어려움을 느끼기도 해요. 이럴 때 필요한 것이 우리 기업에 맞는 ‘좋은 MMP에 대한 기준’일 텐데요.
좋은 MMP는 단순히 성과를 보여주는 도구를 넘어, 복잡한 환경에서도 **기여도를 일관된 기준으로 해석하고 운영 효율**까지 높일 수 있어야 해요.
이번 가이드에서는 MMP를 검토할 때 실무자가 꼭 알아야 할 핵심 기능을 정리해 드릴게요. 신뢰도 높은 데이터를 기반으로 광고 성과를 정확하게 해석하고, 운영 효율을 개선하고 싶다면 아래 기능을 기준으로 살펴보세요.
[]
**📌 핵심 포인트**
- 광고를 집행했는데도 매체별 성과가 다르게 보인다면 측정 기준이 서로 다른 것은 아닌지 확인해야 해요. 좋은 MMP는 데이터를 충분히 쪼개고 기준을 정리해, 성과를 더 신뢰도 높게 판단할 수 있게 해요.
- iOS 환경에서 광고 성과가 예전처럼 보이지 않는다면 주목하세요. 제한된 데이터 환경에서도 기여도를 일관되게 해석할 수 있어야 해요. 허위 트래픽, 중복, 사기를 구분할 수 있어야 하죠.
- MMP는 유저 입장에서 다양한 수치를 이해하고 다음 액션으로 연결할 수 있게 해야 해요. 아래 핵심 기능과 기술을 확인하고 다음 액션이 떠올려지는지 확인해 보세요.
## **누가 진짜 데려온 고객일까? 마케팅 성과의 기준점, MMP**
모바일 광고를 여러 채널에 집행하면, 마치 각 매체가 전환에 모두 기여한 것처럼 보여요.
그런데 막상 리포트를 들여다보면, A 채널은 클릭 이후의 설치를 성과로 보고, B 채널은 조회 기여까지 포함하고 있죠. 이 상태로 매체 리포트를 그대로 합산하면 같은 전환이 여러 번 잡히거나 특정 채널이 실제보다 높게 평가될 수 있어요.
MMP는 **사전에 정의한 규칙에 따라 기여도를 다시 해석하는 역할**을 수행해야 해요. 즉, 매체별 측정에 의존하는 대신 기업이 정한 기준으로 성과를 바라보게 해주는 공통 측정 기준이 있어야 하죠.
이를 **단일 진실 공급원(Single Source of Truth)** 이라고 하는데요. MMP는 이 기준을 크게 3가지 방식으로 구현해요.
### **1\. 어트리뷰션: 전환의 인과관계를 찾아요**
고객의 전환 여정은 선형적이지 않기 때문에, 단순 클릭 수나 매체별 전환 수만으로는 어떤 접점이 실제 전환에 기여했는지 파악하기 어려울 때가 많아요.
MMP는 광고 클릭이나 조회 시점의 터치포인트와 앱 실행 이후의 신호를 연결해 전환의 기여도를 해석해요. 즉, 어트리뷰션(기여도)은 여러 매체가 동시에 성과를 일으킨 것처럼 보이는 환경에서 **어떤 접점이 전환 기여를 한 것인지 파악하는 기준**이 돼요.
다만 실제 환경에서는 모든 유저가 동일한 방식으로 식별되지 않고, 웹과 앱, 기기와 기기 사이를 오가며 여정이 끊기기도 하는데요. 그래서 이후에는 이 전환 기여를 어떤 기준으로 확정할지, 또 식별이 제한된 환경에서 어떻게 해석을 보완할지가 중요해져요.
**_\*터치포인트:_** _클릭(Click)과 조회(Impression)가 있으며, 에어브릿지에선 광고 채널이 터치포인트 데이터를 전달해요. 터치포인트 집계 기준은 광고 채널별로 다를 수 있어_ [_채널별 집계 기준을 확인_](https://help.airbridge.io/ko/guides/ad-channel-touchpoint-types)_하세요._
### **2\. 중복 제거: 여러 채널의 성과를 하나의 기준으로 정리해요**
정확한 데이터 측정을 위해선, 인과관계를 파악한 후 어트리뷰션의 중복을 제거하는 것이 필요해요.
이로 인해 여러 광고 접점을 거친 사용자의 전환이 발생했을 때, **MMP는 정해진 우선순위 체계에 따라 어떤 터치포인트에 기여를 인정할지 해석**해요.
에어브릿지는 라스트 터치 기반으로 작동하지만, 실제 판단은 단순히 마지막 클릭만 보는 것이 아니라 **터치포인트 수집 환경, 터치포인트 유형, 식별 방식** 같은 우선순위를 함께 고려해요.
만약 유저가 A매체와 B매체 광고를 모두 클릭했다면, MMP는 사전에 정의된 어트리뷰션 규칙과 우선순위에 따라 전환 기여를 하나의 터치 포인트에 확정하는 식이죠. 많은 경우 라스트 터치 기준이 적용되지만, 실제 판단에는 클릭·조회 여부나 식별 방식 같은 조건도 함께 반영되는 방식이에요.
이러한 구조가 중요한 이유는 그래야만 기업이 채널 간 성과를 같은 기준으로 비교할 수 있기 때문이에요. 결국 실무자에게 필요한 MMP는 성과를 더 많이 잡아주는 도구가 아니라, **중복을 제거한 뒤에도 실제 성과를 일으킨 정확한 데이터를 제공**해야 하죠.
### **3\. 기여 기간 설정: 어떤 시점의 전환까지 광고 성과로 볼 것인지 기준점이 돼요**
어트리뷰션 기준을 정했다면, 그다음에는 어느 시점까지의 광고 접점을 성과로 인정할 것인지도 함께 정해야 해요. 같은 유저 행동이라도 광고를 클릭한 뒤 하루 만에 설치한 경우와, 일주일이 지난 뒤 설치한 경우는 마케팅 기여도로 해석하는 방식이 달라질 수 있기 때문이죠.
이 지점에서 중요한 것이 바로 \*\*기여 기간 설정, [어트리뷰션 윈도우(Attribution Window)](https://help.airbridge.io/ko/guides/setting-attribution-window)\*\*에요. 예를 들어 특정 매체가 광고 클릭 후 28일 이내 설치를 성과로 집계하고, 기업은 MMP에서 7일만 기여 기간으로 인정한다고 가정해 볼게요.
이 경우 8일째 발생한 설치는 매체 리포트에는 포함될 수 있지만, MMP에서는 같은 기준으로 인정되지 않아 비기여 또는 오가닉으로 보일 수 있어요. 이런 차이 때문에 매체 대시보드와 MMP 수치가 다르게 보여요.
[]
즉, 기여 기간 설정은 **어떤 전환까지 광고의 영향으로 볼 것인지 정하는 해석 기준**이라고 볼 수 있죠. 이 기준이 세워져 있지 않으면 채널 간 성과 비교도 어려워지고, 내부 보고에 쓰는 숫자 역시 일관성을 갖기 어려워요. 반대로 기업의 구매 주기와 서비스 특성에 맞는 기준을 설정하면, 여러 채널의 성과를 같은 프레임으로 다시 해석할 수 있고 데이터 신뢰도도 높아져요.
## **광고비는 썼는데 오가닉? iOS와 식별 제한 환경에서 봐야 할 측정 구조와 기능**
MMP를 도입한 후 광고 집행 성과가 오가닉으로 잡히거나, 특정 채널의 기여가 제대로 드러나지 않는 경우가 있죠. 그럴 땐 iOS 측정 환경을 먼저 살펴볼 필요가 있어요.
특히 iOS 환경에서는 개인정보 보호 정책과 기술적 제한 때문에, 광고를 통해 유입된 사용자의 전환이 예전처럼 집계되지 않을 수 있어요.
이럴 때 필요한 것은 **식별이 제한된 환경에서도 기여도를 최대한 일관되게 해석할 수 있는 환경**이에요. 그리고 이 지점에서 좋은 MMP와 그렇지 않은 MMP의 차이를 알 수 있어요.
### **1\. 식별 제한 환경에서도 기여도 해석을 보완할 수 있어야 해요**
광고를 클릭했는데도 유저가 명확한 식별자로 남지 않을 때가 있죠.
사용자가 개인정보 제공에 동의하지 않거나, 브라우저나 OS 정책으로 인해 측정에 필요한 신호가 제한되는 경우가 그러한데요. 이로 인해 광고를 집행하는 실무자 입장에선 광고 효과를 알 수 없는 상황이 발생해요.
좋은 MMP는 식별자 하나에만 의존하지 않고, 제한된 환경에서도 가능한 범위 안에서 기여도 해석을 보완해요.
에어브릿지는 iOS 14.5 이상 환경에서 애플 ATT에 동의하지 않은 유저가 발생시킨 성과, 딥링크를 통해 발생시킨 성과 등 [**어트리뷰션 규칙과 머신러닝 기반의 확률적 모델링**](https://help.airbridge.io/ko/guides/enhancements-to-attribution)을 통해 **더 많은 신호를 바탕으로 어트리뷰션을 보완할 수 있도록 지원**해요.
즉, 일부 유입이 단순 오가닉처럼 보이기 전에 가능한 범위에서 광고 기여를 더 정교하게 해석할 수 있도록 돕고 있어요. 유저 개인 정보가 충부히 수집되지 않은 상황에서도 성과를 측정할 수 있는지 확인해 보세요.
### **2\. 복잡한 iOS 측정을 마케터가 직접 운영할 수 있도록 지원해야 해요**
iOS 캠페인을 운영해본 실무자라면, 성과를 측정하는 일이 단순한 리포트 확인이 아니라는 점을 이미 알고 있을 거예요. [애플의 SKAdNetwork는 개인정보를 직접 제공하지 않는 방식으로 광고 성과를 측정](https://help.airbridge.io/ko/guides/skadnetwork-overview)하기 때문에, 마케터는 예전처럼 사용자 단위 데이터를 그대로 보기보다 **전환값과 포스트백을 중심으로 성과를 해석**해야 해요.
문제는 이 구조가 생각보다 기술적이라는 점이에요. 어떤 이벤트를 전환값으로 볼지, 어느 시점까지의 행동을 성과 해석에 반영할지, 어떤 채널에서 특정 포스트백이 오는지를 이해해야 하죠. 이 과정이 모두 개발 리소스에 묶여 있으면, 실무자는 측정 기준을 빠르게 바꾸거나 캠페인 상황에 맞게 조정하는 데 제한이 있는데요.
MMP는 SKAN 측정을 단순 지원하는 수준을 넘어, **마케터가 대시보드에서 더 쉽게 운영할 수 있는 환경**을 제공해야 해요.
에어브릿지는 [대시보드에서 전환값 설정을 관리할 수 있도록 지원](https://help.airbridge.io/ko/guides/skadnetwork-4-settings)하며, 마케터가 개발팀 의존도를 낮춘 채 iOS 성과 측정을 운영할 수 있게 도와줘요.
[]
또한 조건을 충족한 SKAN 4 환경에서는 포스트백을 최대 3회까지 받을 수 있어서, 캠페인 성과를 더 다양한 시점에서 해석하는 데 도움이 돼요.
다만 이 기능은 iOS 버전, SDK 버전, 광고 채널 지원 여부, 포스트백 티어, 전환값 설정 완료 여부 등 여러 조건이 충족되어야 한다는 점까지 함께 이해하고 써야 해요.
## **숫자가 많아 혼란스럽다면? 리포트와 운영 기능을 확인하세요**
MMP를 도입하면 지표는 늘어날 수 있어요. 하지만 지표가 다양해지면 동시에 의사결정이 혼란스러워질 수도 있죠.
클릭, 설치, 리인게이지먼트, 매출, 리텐션 등 다양한 지표가 동시에 쌓이기 시작하면 실무자는 무엇을 기준으로 판단해야 하는지 기준과 해석이 필요해요.
이로 인해 MMP는 **특정 지표에서 왜 이런 수치가 나왔는지 설명이 가능**해야 해요. 이 지점에서 리포트, 프로드 방지, 딥링크 같은 운영 기능의 중요성이 드러나죠.
### **1\. 리포트는 숫자를 나열하는 것이 아니라, 성과 해석을 도울 수 있어야 해요**
매체 대시보드와 MMP의 수치가 다르게 보인다면, 왜 이러한 차이가 발생했는지 설명 가능한 체계가 잡혀 있는지 확인해야 해요.
리포트는 성과 해석을 위한 프레임이는 관점에서 볼게요.
예를 들어 실시간 성과를 확인하는 액츄얼스 리포트, 추세를 보는 트렌드 리포트, 단계별 이탈을 확인하는 퍼널 리포트, 채널 간 접점 관계를 보는 터치포인트 분석 계열 리포트는 서로 다른 질문에 답하는 구조를 제공해요.
[]
[**에어브릿지는 액츄얼스, 트렌드, 리인스톨, 리텐션, 퍼널, 레비뉴, 액티브 유저, 터치포인트 분석, 터치포인트 오버랩, 랙 타임 등 다양한 리포트를 제공**](https://help.airbridge.io/ko/guides/airbridge-report-overview)해요.
실무자가 유저가 우리 서비스를 자주 찾아오고 있는지, 광고 매출이 얼마나 늘었는지 등 특정 기준으로 성과를 이해하고 다음 액션을 정할 수 있게 하죠.
성과와 지표별 기준이 명확한 리포트가 있어야 매체 수치와 내부 판단 사이의 간극을 줄일 수 있어요.
### **2\. 광고사기 방지 기능은 가짜 성과를 걸러내는 최소 조건이에요**
모바일 광고 환경에는 봇 클릭, 클릭 인젝션, 비정상적으로 짧은 클릭-설치 시간인 CTIT처럼 실제 유저 행동으로 보기 어려운 신호가 존재해요. 이런 데이터를 걸러내지 못하면 광고 성과는 좋아 보여도, 실제로는 잘못된 트래픽에 예산을 쓰고 있을 수 있어요.
이로 인해 광고사기 [방지 기능](https://help.airbridge.io/ko/guides/fraud-validation-rules)은 데이터 정제를 위한 최소 조건이라고 할 수 있는데요. **에어브릿지는** [**Click-to-Install Time(CTIT)**](https://help.airbridge.io/ko/guides/lag-time-report)**,** [**인젝션 방지**](https://help.airbridge.io/ko/guides/injection-prevention)**,** [**컨버전 트래픽**](https://help.airbridge.io/ko/guides/conversion-traffic)**,** [**프리컨시 캐핑**](https://help.airbridge.io/ko/guides/frequency-capping)**,** [**IP 블록리스팅**](https://help.airbridge.io/ko/guides/ip-blocklisting)**,** [**SDK 시그니처**](https://help.airbridge.io/ko/guides/sdk-signature)**까지 총 6가지 등 사기 광고 검증 관련 기능을 제공**하고 있어요. 이를 통해 실무자가 보다 정제된 데이터를 기준으로 성과를 해석할 수 있어요.
[] 분포를 시각화한 Lag Time 리포트예요. 비정상적으로 짧은 클릭-설치 시간을 가진 트래픽을 빠르게 식별할 수 있어요.")
### **3\. 딥링크는 측정 기능이면서 동시에 전환 최적화 기능이에요**
[딥링크](https://help.airbridge.io/ko/guides/understanding-deep-link)는 유저를 특정 앱 화면으로 보내는 링크로 역할과 함께, 광고를 클릭하고 앱을 설치한 뒤에도 사용자가 원하는 상품 페이지나 혜택 화면까지 자연스럽게 이어지게 만들어야 하죠.
그렇지 않으면 측정은 됐더라도 실제 전환은 놓칠 수 있기 때문인데요.
특히 **앱이 설치되지 않은 유저에게는** [**디퍼드 딥링크**](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)**가 중요**해요. 사용자가 광고를 클릭한 뒤 스토어 설치를 거쳐 앱을 처음 실행했을 때, 단순히 메인 화면이 아니라 광고에서 보던 페이지나 프로모션 화면으로 바로 연결될 수 있어야 설치 이후 이탈을 줄일 수 있어요.
또한 엔터프라이즈 환경에서는 광고 채널과 소재가 많아질수록 링크 운영 복잡도도 커지기 때문에, 이때 링크를 만들 수 있는가보다, **링크가 실제 환경에서 문제없이 동작하는지 점검**할 수 있어야 해요.
[**에어브릿지는 딥링킹 테스트 기능을 제공**](https://help.airbridge.io/ko/guides/deep-linking-test)하여, 주요 환경에서 링크 동작 여부를 확인하고 운영 리스크를 줄이는 데 도움을 줄 수 있어요. 다만, 자동으로 모든 링크를 상시 감시하는 개념으로 쓰기보다는 테스트와 점검을 지원하는 기능으로 이해하는 것이 정확해요.
[]
## **우리 기업에 맞는 MMP인지, 이 기준으로 확인해 보세요**
MMP는 우리 조직의 운영 방식, 산업 특성, 보안 요구사항, 데이터 활용 방식에 맞는지를 함께 봐야 실제 도입 효과가 발생해요. 특히 금융권이나 대기업처럼 데이터 무결성과 내부 보고 기준이 중요한 조직일수록 더욱 그러하죠.
아래 체크리스트를 기준으로, 지금 검토 중인 MMP가 우리 조직의 실제 운영 환경에 맞는지 점검해 보세요.
### **1\. 측정과 성과 해석**
- 라스트 터치 기반 측정뿐 아니라, 식별 제한 환경을 보완할 수 있는 추가적인 어트리뷰션 규칙이 있나요?
- 확률적 모델링 등 제한된 환경에서 기여도 해석을 보완하는 기능이 있나요?
- 웹 투 앱 흐름까지 끊김 없이 측정할 수 있나요?
### **2\. iOS 운영**
- SKAN 4를 지원하나요?
- 전환값과 포스트백 해석을 마케터가 이해하고 운영할 수 있는 UI가 있나요?
- 대시보드에서 전환값을 설정하고 조정할 수 있나요?
- SKAN 데이터와 기존 어트리뷰션 데이터를 함께 해석할 수 있나요?
### **3\. 유저 경험과 링크 관리**
- 디퍼드 딥링크를 지원하나요?
- 링크 테스트 기능이 있어 주요 환경에서 동작 여부를 점검할 수 있나요?
- 커스텀 도메인을 지원하나요?
### **4\. 데이터 정제와 비용 해석**
- 광고 사기(Fraud) 방지 기능이 있나요?
- 비용 데이터를 연동해 ROAS 해석을 도울 수 있나요?
- 광고가 없었어도 발생했을 ‘자연 유입’과 광고를 통해 얻은 ‘순수 성과’를 분리하여 마케팅 예산의 효용을 증명할 수 있는 기능이 있나요?
### **5\. 거버넌스와 확장성**
- 로우 데이터를 CSV로 추출하거나 클라우드 스토리지로 연동할 수 있나요?
- 지원 범위가 Amazon S3, Google Cloud Storage, Azure Blob Storage 같은 실제 연동 가능한 스토리지 기준에 부합하나요?
- 오디언스를 외부 매체로 전송해 타깃팅에 활용할 수 있나요?
- 타임존과 통화 단위를 운영 환경에 맞게 설정할 수 있나요?
특히 금융권과 대기업은 내부 보고와 데이터 거버넌스 기준까지 함께 고려해야 해요. 이로 인해 MMP는 마케팅팀만 쓰는 도구가 아니라, 여러 조직이 같은 숫자를 기준으로 판단하게 만드는 운영 인프라가 될 수 있어야 하죠.
## **국내 점유율 1위 에어브릿지로 마케팅 성과를 측정하세요**
에어브릿지(Airbridge)는 국내 MMP 시장에서 국내 기업 대상 고객사 수 1위를 기록했는데요.
커머스, 금융, 교통, 라이프스타일, AI 등 에어브릿지는 국내 마케팅 환경에 맞춘 측정 방식과 높은 채널 연동성으로 MMP 도입 시 느끼는 혼란을 최소화하고 성과를 극대화해줘요.
복잡한 모바일 마케팅 환경과 이로인해 더 정교한 측정 기준과 운영의 필요성을 느끼고 있다면, 지금 바로 에어브릿지를 경험해 보세요.
국내 마케팅에 최적화된 신뢰도 높은 데이터와 광고 운영으로 더 큰 비즈니스 임팩트를 만들 수 있을 거예요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 구글이 던진 신호, 앱 마케터가 데이터 측정 환경을 다시 점검해야 할 5가지 이유
URL: https://www.airbridge.io/ko/blog/google-io-2026-app-marketing-data-measurement-stack-changes
Language: Korean
Published: 2026-06-23
Author: Jaehyuk Kim
지난 2026년 5월 19일과 20일, 미국 캘리포니아 마운틴뷰에서 구글의 글로벌 마케팅 컨퍼런스인 Google I/O 2026과 Google Marketing Live 2026이 열렸어요.
검색·결제·디스커버리·광고 자동화·구독 모델까지 그야말로 마케팅과 AI의 모든 영역을 관통하는 행사였는데요. 구글 CEO Sundar Pichai는 [키노트](https://blog.google/innovation-and-ai/sundar-pichai-io-2026/)에서 에이전틱 Gemini 시대를 본격 선언했어요.
**다양한 발표 중에서 앱 마케터가 눈여겨 봐야 할 변화 5가지를 정리했어요.** 모두 AI 에이전트가 사용자와 우리 앱 사이에 한 단계 끼면서, 우리가 측정해온 신호들의 의미를 바꿔놓고 있어요. 구글이 준비하는 새로운 AI 패러다임, 앱 마케터는 무엇을 준비해야 할까요?
> “We're firmly in our agentic Gemini era. I'm excited to see how it will unlock new ways to accelerate our mission and transform our products to be radically more helpful, for everyone everywhere.” — Sundar Pichai, CEO of Google and Alphabet
🎥 영상으로 보기
[Google I/O 2026 keynote in 35 minutes](https://www.youtube.com/watch?v=OMhKgQmeMhI)
**핵심 포인트**
- Ask Play와 Gemini 앱(월 9억 명)이 앱 추천을 직접 담당하면서 발견 경로가 Play Store 밖으로 흩어졌어요.
- AI Max와 AI Brief가 광고 실행을 AI에 넘겨요. 마케터가 직접 챙겨야 할 건 시그널 품질과 브랜드 가이드라인이에요.
- Universal Cart와 AP2가 구글 안에서 결제까지 끝내요. 자사 앱 전환 데이터만 보면 커머스 성과가 실제보다 작게 잡혀요.
- Google Play 결제 유예 기간이 60일로 늘면서 구독 앱의 LTV 베이스라인이 작년과 달라졌어요. 작년 데이터로 학습한 입찰·예산 모델은 보정이 필요해요.
- AI Mode 쿼리의 88%가 클릭 없이 끝나요. 측정 기준을 클릭 수에서 광고로 만들어진 추가 매출로 바꿔야 할 수 있어요.
## **Ask Play와 Gemini 앱이 직접 앱 추천을 해줘요**
**🎥 영상으로 보기**
[What’s new in Google Play](https://youtu.be/fwLiTPtPHjw)
**구글은 Play Store 내 AI Q&A 기능이 이미 사용자 쿼리의 95%에 답하고 있다**고 밝혔어요. 그 위에 Ask Play라는 대화형 발견 레이어가 추가됐고, 같은 시기 Gemini 앱이 월 9억 명을 넘기면서 Android와 웹에서 앱 자체를 추천하기 시작했어요.
> “We're entering the era of Search agents, where you can easily create, customize and manage multiple AI agents for your many tasks, right in Search.” — Google official announcement, Search at Google I/O 2026
_👉 자세한 내용은_ [_Android Developers Blog: I/O 2026 Google Play 업데이트 정리_](https://android-developers.googleblog.com/2026/05/io-2026-whats-new-in-google-play.html)_에서 확인할 수 있어요._
[]
### **1\. 앱 발견 경로가 4곳으로 흩어졌고, ASO는 의미 기반으로 옮겨가요**
이날 구글은 **Engage SDK가 월 3,000만 명 이상 사용자에게 도달하며 앱 오픈을 전년 대비 45% 늘렸다고 발표**했는데요. 이는 Play Store 검색이 더 이상 유일한 진입점이 아니라는 것을 의미해요.
또한 Gemini 앱, Ask Play, Engage SDK 슬롯 모두 키워드 매칭이 아닌 의미 기반 추천으로 작동하고 있고, 여기에 기존 Play Store 검색까지 합치면 진입점이 4곳으로 늘어나는데요. Play Store referrer만 보고 있다면 이 4가지 경로가 같은 오가닉(Organic) 유입으로 섞여 들어오니, 앱 발견 경로를 분기마다 따로 분류해서 봐야 해요.
ASO도 키워드 커버리지에서 의미론적 커버리지로 옮겨가는 흐름이에요. 앱 설명과 스크린샷 카피를 시나리오 중심으로 재작성하고, 동적 메타데이터 슬롯(In-App Events, Promotional Content, Custom Store Listings)에 어떤 시나리오를 노출하는지 인벤토리로 정리해두면 좋아요. [기본 ASO 전략](https://www.airbridge.io/ko/blog/aso-strategy)이 정리된 위에서 새 레이어를 점검하면 돼요.
### **2\. 에이전트 호출 트래픽을 휴먼 트래픽과 분리해야 해요**
Gemini Spark는 Google Cloud 가상 머신에서 24시간 작동하는 개인 AI 에이전트예요. Gmail과 구글 문서, 프레젠테이션과 통합되고, MCP(Model Context Protocol: AI 에이전트가 외부 도구를 호출하는 표준)를 통해 Canva·OpenTable·Instacart 같은 서드파티 도구로도 확장되죠.
문제는 측정이에요. 에이전트가 사용자 대신 앱을 호출하면 일일 활성 사용자(DAU) 지표에는 잡히지만, 그 세션은 푸시 알림이나 광고로 리타깃팅할 의미가 없어요. 자동화 트래픽과 휴먼 트래픽을 구분하지 않으면 리텐션 코호트가 부풀려지고, 그 위에서 사용자 생애 가치(LTV) 모델을 돌리면 베이스라인이 어긋나게 돼요.
👉 Gemini Spark의 전체 구조와 워크플로우는 [Google Cloud 공식 발표 — Google I/O 2026 혁신](https://blog.google/intl/ko-kr/company-news/technology/innovations-from-google-io-26-on-google-cloud/)에서 확인할 수 있어요.
## **AI Max와 AI Brief, 자동화는 커지고 마케터의 가시성은 줄어요**
[]
**Google Marketing Live 2026에서는 AI 에이전트를 통해 앱 딥링크 구현 시간을 "며칠에서 분 단위"로 단축한다고 발표**했어요. 여기서 공개된 AI Brief는 광고주가 자연어 한 문장으로 브랜드 목소리·오디언스·가이드라인을 제공하면 AI가 그 안에서 광고를 생성하는 솔루션이죠.
즉, 캠페인 운영의 핵심이 실행에서 시그널 설계로 옮겨가고 있어요.
🎥 영상으로 보기
[Introducing Gemini Omni: Create Anything from Anything](https://youtu.be/KUyRq7szZsM)
👉 자세한 내용은 [Gemini Omni Flash 공식 발표 — 구글 공식 블로그](https://blog.google/intl/ko-kr/company-news/technology/gemini-omni-kr/)에서 확인할 수 있어요.
### **1\. 광고 캠페인 컨트롤이 자연어 한 줄로 줄어들었어요**
딥링크 구현이 빨라지면 순식간에 수많은 A/B 테스트 소재를 만들 수 있어요. 매주 소재를 만들기 위해 했던 회의가 사라지는 대신, 자동 생성된 변형이 브랜드 톤과 측정 시그널 안에 있는지 나중에 점검하는 회의가 필요해요. 소재 100개를 어떤 그룹으로 묶어 점검할지 미리 정해두지 않으면, 자동화로 속도가 올라간 만큼 소재 검토에 들어가는 리소스만 더 커져요.
AI Brief뿐 아니라 Performance Max와 Google App Campaigns도 같은 방향으로 가고 있어요. 앞으로 AI Max가 자리잡으면 광고 소재도 AI가 직접 만들게 되니, 마케터가 관리하는 데이터 품질이 캠페인 성과를 거의 좌우하게 돼요.
### **2\. 시그널 설계와 브랜드 톤 & 매너 정의가 마지막 가드레일이에요**
광고 크리에이티브 자동 생성 기능은 이제 이미지와 영상까지 확장됐어요. 이번에 공개된 [Gemini Omni Flash](https://blog.google/intl/ko-kr/company-news/technology/gemini-omni-kr/)는 자연어 한 줄로 영상을 만들고, 배경을 바꾸거나 디지털 아바타를 사람마다 다르게 만들 수 있어요.
[Google Pics](https://blog.google/intl/ko-kr/company-news/technology/workspace-update-io-2026-kr/)는 Nano Banana 기반 이미지 자동화로 객체 분할·텍스트 수정·다국어 번역을 한 번에 처리해요. 원하는 결과를 한 줄로 입력하면 AI가 텍스트·영상·이미지를 모두 자동 생성해요.
앞으로는 이렇게 크리에이티브 소재를 만드는 시간도 줄어들 거예요. 이런 환경에선 시그널이 결과를 결정해요. 전환 가치, 관심사 시그널, 고가치 사용자의 정의가 모호한 채로 AI 자동화에 권한을 많이 넘기면 위험해요. AI가 학습한 좋은 사용자가 실제로는 비즈니스 가치가 낮은 코호트일 수 있기 때문이죠.
브랜드 톤 & 매너도 마찬가지예요. AI Brief에 입력한 한 문장이 광고 카피와 추천 이미지까지 영향을 주는데, 톤 & 매너 가이드를 마케터 머릿속에만 두면 입력값이 매번 달라지고, 결국 광고 일관성이 흔들려요. 이건 카피라이팅 문제가 아니라 시스템의 문제예요.
## **Universal Cart로 커머스 앱 퍼널 절반이 구글 안에서 일어나요**
Universal Cart는 구글 검색이나 Gemini, 유튜브, Gmail에서 상품을 장바구니에 추가하고 결제까지 끝낼 수 있는 쇼핑 허브예요. 주요 글로벌 리테일러(Nike, Walmart, Shopify 가맹점 등)가 출시 파트너로 포함돼 있어요.
[]
한국 출시 일정은 발표되지 않았지만, Universal Commerce Protocol(UCP)이 캐나다, 호주, 영국으로 확장되고 있고, 유튜브, 호텔 예약, 로컬 음식 배달 카테고리까지 늘어날 계획이라고 해요.
👉 자세한 내용은 [Google Shopping의 Universal Cart 공식 발표](https://blog.google/products-and-platforms/products/shopping/google-shopping-cart/)에서 확인할 수 있어요.
### **1\. 결제가 자사 앱 밖에서 끝나면 광고 캠페인은 SKU 피드부터 점검해야 해요**
UCP는 가맹점과 결제, 광고 데이터를 묶는 오픈 표준인데요. Agent Payments Protocol(AP2) 기능이 더해지면서, AI 에이전트가 사용자를 대신해 자율 결제하는 'Human Not Present'(사람이 직접 결제하지 않는) 결제가 가능해졌어요.
예를 들어, 사용자가 "30만 원 이하 러닝화 추천해서 사줘"라고 말하면 Gemini가 상품을 고르고 결제까지 끝낼 수 있어요. 하지만 거래가 자사 앱 SDK 범위 밖에서 일어나면, 광고 캠페인의 노출·클릭이 실제 매출로 이어졌는지를 자사 앱 데이터만으로 확인할 수 없어요.
AI 에이전트가 자사 상품을 고르려면 Merchant Center 피드(구글 쇼핑 상품 카탈로그)는 기본으로 갖추고 있어야 하는데요. 그 위에 가격과 재고, 프로모션을 실시간으로 동기화하는 Content API와 UCP가 요구하는 가맹점 메타 데이터가 필요해요.
그래서 자사 SKU 중 몇 퍼센트가 'Eligible(노출 가능)' 상태로 등재돼 있는지 분기마다 트래킹해야 해요. 그렇지 않으면 아무리 광고 예산을 늘려도 자사 데이터와 광고 성과가 맞지 않는 문제는 해결할 수 없어요.
### **2\. AI 에이전트 거래와 자사 데이터를 결합할 키가 필요해요**
AI 에이전트가 결제하면 자사 데이터에 남는 신호는 주문 ID와 결제 금액, 배송지 정도예요. 그 거래가 어떤 광고나 콘텐츠 영향을 받았는지, 동일 사용자의 이전 자사 앱 행동과 어떻게 이어지는지를 묶어서 보지 않으면, 광고 ROI가 실제보다 낮게 잡혀요.
에어브릿지에서 동일 사용자의 앱 내 행동은 [크로스채널 어트리뷰션 환경](https://www.airbridge.io/ko/blog/cross-channel-attribution-guide-conversion-performance)에서 결합해야 볼 수 있는데요. 이번 발표를 보고 데이터 팀과 합의할 결합 키(Google 주문 데이터와 자사 사용자 데이터를 이어줄 공통 식별자) 후보를 미리 정리해두세요. 그러면 1년 뒤 어트리뷰션 공백이 줄어들고 데이터 정확도가 올라갈 거예요.
## **Google Play 결제 유예 기간이 60일로 늘면서 구독 앱 LTV 베이스라인이 바뀌었어요**
**2025년 12월 1일부터 Google Play의 기본 account hold(구독 결제 실패 시 계정을 유예 상태로 두는 기간)가 30일에서 60일(자동 계산)로 확장**됐어요. 기존에 30일로 설정돼 있던 구독 플랜도 모두 자동으로 전환됐어요.
구글은 이날 상위 개발사 기준 비자발적 이탈이 최대 18%, 전체 이탈이 9% 감소했다고 밝혔는데요. AI 자동화 입찰과 LTV 모델이 같은 베이스라인 위에서 작동한다는 점을 생각하면, 정책 한 줄이 자동화 측정 시스템 전체의 정확도를 떨어뜨릴 수 있는 변화예요.
👉 정책 변경의 LTV 영향 분석은 [RevenueCat: Play Billing Library 9.0 분석](https://www.revenuecat.com/blog/engineering/play-billing-v9/)에서 확인할 수 있어요.
### **1\. 리텐션 커브가 위로 올라간 만큼 LTV 모델은 실제보다 낮게 잡혀요**
2025년 4분기와 2026년 1분기 리텐션 커브(retention curve)를 비교하면 곡선이 자연스럽게 위로 올라가요. 그런데 이건 마케팅이나 제품 개선의 결과가 아니라 정책 변경에서 비롯된 효과예요. 이 둘을 분리하지 않고 "리텐션이 좋아졌으니 UA 예산을 더 쓰자"라고 결정하면, 실제 효율을 잘못 추정한 채로 예산을 쓰게 돼요.
구독 앱의 [예측 LTV(pLTV) 모델](https://www.airbridge.io/ko/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app)은 보통 '리텐션 커브 × 평균 구독 금액(ARPU)'으로 계산하는데요. 그래서 30일 유예 시절의 곡선으로 학습한 모델은 60일 유예 환경에서의 LTV를 실제보다 낮게 잡을 수밖에 없어요.
예를 들어 LTV를 실제보다 15% 낮게 잡고 있다면, 자동화 입찰이 같은 ROAS 목표 아래에서 입찰가를 보수적으로 매기고, 채널·크리에이티브 평가도 같은 폭으로 어긋나요. 모델이 정책 변화를 자동으로 반영하진 않으니, LTV 모델을 직접 다시 학습시켜야 해요.
### **2\. 정책 변경 전후 코호트를 따로 관리해야 해요**
BI 대시보드에 12월 1일 전후 토글을 두고 v1·v2 LTV 모델을 함께 운영하는 게 첫 단계예요. 그 다음, 입찰 자동화에 어느 모델을 연결할지는 분기 리뷰 문서에 결정 근거와 함께 남겨두세요. 그래야 다음 분기 회의에서 같은 논의를 반복하지 않아도 돼요.
## **AI Mode 10억 명 시대, 어트리뷰션의 다음 질문은 인크리멘탈리티예요**
이날 발표에서 또 하나 주목을 받은 게 바로 AI Mode예요.
**AI Mode가 출시 1년 만에 월 10억 명을 돌파했고, AI Mode 쿼리의 88%가 외부 클릭 없이 끝나고 있다고 해요.** 구글 전체 AI 처리량도 한 달 기준 작년 480조 토큰에서 올해 3,200조 토큰으로 1년 사이 7배 늘었다고 하죠.
이는 AI Mode가 단순히 검색 엔진 위에 AI가 얹힌 기능이 아니라, **검색 인프라 그 자체가 됐다는 것을 의미**해요.
> “Search has become less about individual queries and feels more like an ongoing conversation, giving you deeper insights and connecting you with the vastness of the web.” — Sundar Pichai, CEO of Google and Alphabet
🎥 영상으로 보기
[Sundar Pichai Opening Remarks | I/O 2026 Keynote](https://youtu.be/duHhImuaZGU?t=253)
👉 자세한 내용은 [Google 공식 블로그: Search at Google I/O 2026](https://blog.google/products-and-platforms/products/search/search-io-2026/)에서 확인할 수 있어요.
### **1\. 제로 클릭 60%대 환경에서 다음 질문은 인크리멘탈리티예요**
현재 [전체 Google 검색의 64.82%가 클릭 없이 끝나](https://www.digitalapplied.com/blog/zero-click-search-statistics-2026-complete-data)고 있는데요. 비슷한 시기 Ahrefs가 진행한 리서치에서도 AI Overviews가 등장한 검색에서는 제로 클릭 비율이 54%에서 72%로 올라갔고, 오가닉 클릭은 38% 감소했다고 해요.
클릭 수만 보고 채널을 평가하고 있었다면 이러한 현상을 놓칠 수 있어요. 그래서 앞으로 마케터가 던지는 질문 자체가 바뀌어야 해요. "이 전환 클릭이 어디서 왔는가"가 아니라 "이 채널을 끄면 매출이 얼마나 줄어들었을까", 즉 인크리멘탈리티의 관점이죠.
실제로 [미국 마케터의 46.9%가 앞으로 1년 내 마케팅 믹스 모델링(MMM)에, 36.2%가 인크리멘탈리티에 더 투자할 계획](https://www.emarketer.com/content/mmm--incrementality--other-measurement-trends-that-will-define-2026)이라고 밝혔어요.
그럼 우리는 무엇을 시작해 볼 수 있을까요? 가장 가볍게 시작할 수 있는 방법은 홀드아웃 테스트(Holdout test)이에요. 특정 지역에서 1~2주 동안 채널을 끄고 매출 변화를 측정해보면, 이후 클릭 어트리뷰션을 보정할 기준선이 생기기 때문이죠.
### **2\. AI 답변 안에서 우리 브랜드 인용을 측정해야 해요**
제로 클릭 현상은 AI 답변 자체가 새로운 노출 표면이 됐다는 것을 의미하기도 해요. AI Mode와 Gemini가 답변에서 우리 브랜드를 언급하면, 클릭 없이도 브랜드 노출 및 인지가 일어나요. 이 노출을 측정하는 영역이 GEO(Generative Engine Optimization)예요.
카테고리 대표 쿼리 50~100개를 AI Mode·Gemini·ChatGPT에 정기적으로 던져서 우리 브랜드의 노출 빈도·위치·인용 형태(요약 인용, 링크, 이미지)를 기록해요.
그런 다음 경쟁사 노출과 비교해 share of voice(브랜드 점유율)를 계산해보세요. 브랜드 점유율 측정에는 **Ahrefs**나 **Amplitude**, **Ubersuggest**가 가장 많이 사용되고 있어요.
## **AI 에이전트 시대의 성과 측정, 우리는 어디까지 준비됐을까요**
오늘 살펴본 5가지 변화들을 보면, 하나만 골라 대응하는 게 아니라 성과를 측정하는 시스템과 베이스라인을 전반적으로 점검할 시점이라는 것을 알 수 있어요.
아래 체크리스트에서 "아니오"가 2개 이상이면 측정 스택 전체 점검을 우선순위로 올려야 해요. 1개라면 분기 안에 보강 계획을 잡으면 되고, 없다면 GEO 측정처럼 새로 등장한 영역들을 우선순위에 올리세요.
### **측정 스택이 5가지 변화에 답할 수 있는지 점검하는 자기진단**
1. **발견 경로별 구분:** 오가닉 설치를 Play Store 검색·Ask Play·Gemini 추천·Engage SDK 등 경로별로 나눠서 볼 수 있나요? → Play Store referrer 하나로는 4개 경로가 같은 "오가닉"으로 섞여 들어와요.
2. **시그널 문서화:** 광고 캠페인의 전환 가치·고가치 사용자 정의·브랜드 가이드라인을 문서로 정리하고 분기마다 업데이트하고 있나요? → AI Brief와 AI Max는 입력 시그널의 품질이 결과를 좌우해요.
3. **SKU 적격성 트래킹:** 자사 SKU 중 몇 퍼센트가 Merchant Center 피드에 "Eligible" 상태인지 분기마다 보고 있나요? → Universal Cart 환경에서는 피드에 정상 등록된 상품만 AI 에이전트가 선택해요.
4. **코호트 분리:** 2025년 12월 1일 전후 코호트를 분리해서 LTV와 리텐션 커브를 따로 보고 있나요? → 2025년 12월 1일은 Google Play 결제 유예 기간이 60일로 늘어난 날이에요. 두 코호트를 합쳐 보면 정책 효과와 마케팅 효과를 구분할 수 없어요.
5. **클릭 밖 검증:** 지난 12개월 안에 클릭 어트리뷰션 밖에서 마케팅 효과를 검증한 적이 있나요? (홀드아웃 테스트, 전환 리프트 포함) → Zero-click 환경에서는 클릭 카운트만으로 진짜 채널 효과를 알 수 없어요.
## **마무리**
이번 발표들을 따로 보면 다양해 보이지만, 한 발 떨어져서 보면 모두 같은 곳을 가리켜요. 사용자가 우리 앱에 닿기 전부터 자사 데이터로는 보이지 않는 영역이 생겼고, 측정의 기준도 클릭에서 의도와 인크리멘탈리티로 바뀌고 있어요.
새로운 솔루션을 도입하기 전에 자사 측정 스택을 펼쳐놓고, 5가지 변화들 중 가장 약한 지점부터 짚어보세요. 거기가 다음 분기 OKR이 시작되는 자리가 될 거예요.
**참고 자료(REFERENCES)**
1. Sundar Pichai. "Google I/O 2026 keynote." blog.google, 2026.
2. Android Developers Blog. "I/O 2026: What's new in Google Play." 2026.
3. Google. "Innovations from Google I/O 2026 on Google Cloud." 2026.
4. Google. "Gemini Omni Flash 공식 발표." 2026.
5. Google. "Google Workspace I/O 2026 업데이트 (Google Pics 포함)." 2026.
6. Google. "Google Shopping의 Universal Cart 공식 발표." 2026.
7. Google. "Search at Google I/O 2026." 2026.
8. RevenueCat. "Play Billing Library 9.0 분석." 2026.
9. Nobori AI. "Google AI Mode zero-click rate, 100M users." 2026.
10. Digital Applied. "Zero-Click Search Statistics 2026 Complete Data." 2026.
11. eMarketer. "MMM, Incrementality, Other Measurement Trends that Will Define 2026." 2026.
---
## 스팀 마케팅으로 PC 게임 시장 공략해볼까? 실무자에게 듣는 노하우 A to Z
URL: https://www.airbridge.io/ko/blog/steam-pc-console-game-marketing-tips
Language: Korean
Published: 2026-06-18
Author: Team Airbridge
지난 6월 4일, 포스코타워 역삼에서 **** 세미나가 열렸어요.
AB180과 Appier, Playio, TikTok이 공동 주최한 이번 행사에는 게임 UA·마케팅에 관심 있는 마케터와 개발자 수백 명이 자리를 채웠어요.
다양한 발표 중 AB180을 대표한 Customer Success Team Lead(Gaming) 최혜린 님은 **‘글로벌 유저를 사로잡는 스팀 마케팅 전략’ 세션**을 맡아 발표를 진행했는데요. 혜린 님은 게임사 마케터 출신으로 직접 과금도 하면서 25년간 다양한 게임을 플레이해온 게이머이기도 해요.
이날 스팀 런칭을 직접 경험한 실무자로서 본인만의 경험과 노하우를 아낌없이 공개했는데요.
스팀 마케팅은 모바일 마케터에게는 여전히 낯선 영역일 수 있죠. 모바일 게임 다운로드가 줄고 PC/콘솔이 성장하는 지금, **스팀 진출을 고민하는 게임 마케터**에게 도움이 될 발표 내용을 정리했어요.
**📌 핵심 포인트**
- 모바일 시장이 기존 유저 리텐션 중심으로 재편되는 사이, PC/콘솔은 매출과 다운로드 모두 성장세를 보이고 있어요. 성장하는 PC 게임 시장을 노리고 있다면, 지금이 스팀 마케팅을 시작할 적기예요.
- 예산보다 전략이 노출을 결정하는 플랫폼을 찾고 있다면 스팀을 활용할 수 있어요. 모바일 UA처럼 대규모 광고비 없이도, 위시리스트 모멘텀과 커뮤니티 신호를 잘 쌓으면 스팀 알고리즘의 선택을 받을 가능성이 높아져요.
- 스팀 마케팅 전략이 실행력을 가지려면 데이터가 뒷받침돼야 해요. 위시리스트 전환을 만드는 채널, 실제 구매로 이어지는 채널은 에어브릿지 데이터로 확인할 수 있어요.
## **게임 시장의 무게 중심이 이동하고 있어요**
[appier Game UA_03.webp] 최혜린 님")
모바일 중심이었던 게임 시장의 흐름이 달라지고 있어요. 매출과 다운로드 순위는 글로벌 퍼블리셔가 가져가고 있고, 유저들의 플랫폼 선호도도 바뀌고 있는데요. 최혜린 님은 변화를 데이터로 짚으며 스팀에 주목해야 하는 이유를 설명했어요.
### **1\. 매출부터 광고 네트워크까지, ‘글로벌 퍼블리셔’가 강세예요**
글로벌 모바일 게임 매출 상위권은 글로벌 퍼블리셔가 주도하고 있어요.
[AppMagic에 따르면 2025년 모바일 게임 매출 상위권은 Tencent, Scopely, Dream Games, FunFly 같은 글로벌 퍼블리셔가 독식](https://www.pocketgamer.biz/the-top-grossing-mobile-games-of-2025/)했어요. 한국은 한국 게임사들의 게임이 상위권을 차지하고 있지만, 광고 단가와 구매력이 높은 주요 국가인 미국과 일본은 글로벌 퍼블리셔가 점령한 상태예요.
주요 마케팅 집행 국가인 영국, 대만, 인도네시아도 사정은 크게 다르지 않고요.
최혜린 님은 “대형 게임사는 머지, 4X 등 다양한 장르에 걸쳐 네트워크를 선점하고 있다"고 언급했어요.
광고 경쟁도 갈수록 치열해지고 있는데요. [2025년 월평균 모바일 게임 광고주 수는 8만 4천 개 이상으로 전년 대비 21.9% 증가](https://mobidictum.com/mobile-gaming-marketing-trends-2026/)했어요.
대규모 자본을 앞세운 글로벌 퍼블리셔들이 주요 광고 네트워크를 선점하면서, 상대적으로 자본력이 약한 게임사가 경쟁하기 더 어려운 환경이 된거죠.
### **2\. 모바일은 성숙기, PC/콘솔은 지금이 성장기예요**
[appier Game UA_05.webp]
그렇다면 글로벌 퍼블리셔와 더불어 유저의 게임 소비 방식은 어떻게 바뀌고 있을까요?
[1.webp]
2025년 모바일 게임 다운로드는 전년 대비 **7.2% 감소**했고 매출 성장은 1%에 그쳤어요. **신규 유저 확보보다 기존 유저의 리텐션과 소비 확대가 매출을 이끄는 구조**로 바뀌고 있죠.
[2.webp]
반면 PC/콘솔은 **매출 +13%**, 다운로드 +6%로 성장하고 있어요. 모바일이 성숙기에 접어든 사이, PC 게임 시장은 여전히 성장 여력이 남아 있는 거예요.
그 중심에 있는 플랫폼이 스팀이에요.
👉🏻[중견 게임 스튜디오를 위한 스팀 마케팅 전략이 궁금하다면?](https://www.airbridge.io/ko/blog/game-studio-steam-marketing-playbook)
[3.webp]
스팀이 공개한 자료에 따르면, **2025년 Steam MAU는 1억 4,700만 명으로 전년 대비 11.4% 늘었고, 동시접속자(CCU)는 2026년 1월 4,204만 명으로 신기록을 경신**했어요. 2020년 약 2,400만 명과 비교하면 6년 만에 두 배가 된 수치예요.
## **스팀 게임 마케팅, 모바일과는 다른 접근이 필요해요**
그렇다면 스팀 게임 마케팅, 어떻게 시작해야 할까요?
최혜린 님은 “스팀 마케팅은 모바일처럼 정형화된 방법론이 없고, 생태계가 다르게 작동한다”며, 모바일 마케터가 스팀을 처음 접할 때 알아야 할 핵심 노하우를 공유했어요.
### **1\. 스팀 기반 론칭은 모바일보다 호흡이 길어요**
[4.webp]
스팀 런칭은 모바일처럼 사전예약 오픈 후 2~3달 안에 끝나기보다는, 게임 공개부터 정식 출시까지 더 긴 호흡으로 운영돼요.
HADES는 2018년 12월 트레일러 공개 후 에픽게임즈 얼리 액세스, 스팀 얼리 액세스를 거쳐 2020년 9월 정식 출시까지 약 **2년**이 걸렸어요.
Dave the Diver도 2022년 6월 스팀 넥스트 페스트 참가 후 얼리 액세스, 정식 출시까지 **1년**이 소요됐죠. 국내 게임인 탄나비는 스팀 데모 오픈부터 정식 출시까지 **2년 8개월**을 투자했어요.
**스팀 알고리즘은 위시리스트 총량이 아니라 증가 속도와 누적 신호를 기반으로 게임을 추천**해요. 이 신호를 만들려면 스토어 페이지 오픈부터 Next Fest 참가, Early Access, 정식 출시까지 단계별로 시간이 필요하죠.
마케팅 플랜도 이 호흡에 맞게 설계해야 해요.
### **2\. 스팀에는 유료 지면이 없어요**
[5.webp]
모바일에서는 UA 예산을 집행해 CPI, 인앱 이벤트, tROAS 등 다양한 목표로 캠페인을 운영할 수 있어요. 예산이 있으면 원하는 만큼 노출을 늘릴 수 있죠.
이와 달리 **스팀은 공개된 유료 광고 지면이 없어요. 유저의 관심사와 게임 데이터를 바탕으로 한 알고리즘이 노출을 결정**하기 때문에, 모바일처럼 광고비를 직접 진행하는 방식이 아닌 위시 리스트 모멘텀과 커뮤니티 신호를 잘 쌓는 방식으로 노출을 늘려가면 돼요.
**스팀 상점에서 중요한 지면은 4가지**예요.
1. 스팀 관계자가 직접 선정해 모든 유저에게 노출하는 **큐레이션**
2. 개인 알고리즘으로 유저 취향에 맞게 자동 구성되는 **특집 및 추천**
3. 총수익 기준으로 배치되는 **최고 인기 제품**
4. 위시리스트가 많이 쌓인 게임이 올라가는 **인기 출시 예정**
이 중 인기 출시 예정 탭은 단순히 위시리스트 수가 많은 게임이라고 해서 올라가는 단순한 구조가 아니예요.
최혜린 님은 "2주 기준 위시리스트 증가 속도가 핵심"이라고 강조했어요. **총량보다 모멘텀**이 중요한 거죠.
[appier Game UA_01.webp]
## **스팀 게임 마케팅, 4가지 노하우만 기억하세요**
최혜린 님은 스팀 **게임 마케팅의 핵심 영역으로 다음 4가지**를 꼽았어요.
1. 위시리스트
2. 공식 프로그램
3. 커뮤니티
4. 인플루언서
각각을 따로따로가 아니라 통합해서 관리해야 한다고 강조하며, 아래와 같은 실무 경험을 공유했어요.
### **1\. 위시리스트는 최대한 많이 쌓으세요**
[6.webp]
위시리스트는 여전히 출시 게임 출시 전 유저 수요와 시장 반응을 가늠할 수 있는 지표이죠.
스팀이 공식 전환율 수치를 제공하지는 않았지만, 개발사와 퍼블리셔들이 공유한 경험치와 온라인에서 통용되는 자료들을 종합한 평균치로 참고해 볼 수 있어요.
- **5,000+**: 시장 검증 및 초기 판매 기반 확보
- **7,000+**: 인기 출시 예정(Popular Upcoming) 탭 진입 가능
- **15,000+**: 의미 있는 초기 판매량 확보
- **3만~5만+**: 스팀 추천 확대, 흥행 가능성 증가
- **10만+**: 대형 히트작 수준의 성과 기대
### **2\. 스팀 공식 프로그램을 전략적으로 활용하세요**
[7.webp]
스팀에서 공식적으로 제공하는 테스트와 프로모션 프로그램을 출시 전략에 적극 활용해 보세요.
테스트 프로그램으로는 Demo(체험판), Prologue(프롤로그), PlayTest(플레이 테스트) 등 3가지가 있어요.
#### **1) Demo(체험판)**
**스팀 최대 행사인 Next Fest에 참가할 수 있는 빌드**예요. Cloud Save를 설정하면 데모 플레이 데이터를 본편으로 이어갈 수 있어서 유저 이탈을 줄일 수 있죠. 다만 콘텐츠 완성도가 낮으면 부정 평가가 쌓일 수 있어서 퀄리티 관리가 필수예요.
#### **2) Prologue(프롤로그)**
본편과 별도 스토어 페이지를 운영하는 방식이에요. **추가 노출 확보에 유리**하며, 무료 체험을 통해 본편에 유입 가능한 것이 강점이에요.
다만, 프롤로그 페이지에 쌓인 부정 리뷰가 본편 구매 의사결정에 영향을 줄 수 있어 운영 리소스 관리가 필요해요.
#### **3) PlayTest(플레이 테스트)**
본 게임 리뷰·평점에 영향을 주지 않고 밸런스와 서버 부하를 검증할 수 있는데요.
여기서부터는 **플레이 데이터 수집과 밸런스, 서버 부하 같은 기술적 완성도를 높이는 데 목적**이 있어서, 위시리스트 확보보다는 출시 전 게임 완성도를 높이는 데 효과적이에요.
[8.webp]
프로모션 프로그램은 2가지가 있어요.
#### **1) Steam Next Fest(넥스트 페스트)**
연 3회 일주일간 열리는 출시 예정작 축제로, 게임당 단 한 번만 참가할 수 있어요. 행사 기간 집중 노출로 위시리스트를 대량 확보할 수 있는 가장 효율적인 기회인 만큼 타이밍 선택이 중요해요.
**특히 첫 48시간 노출이 행사 후반 노출을 좌우**하는데요. 참가 전에 위시리스트가 많을수록 유리하지만, 일반 게임사는 현실적으로 쉽지 않은 만큼 데모 완성도를 높이고 초반 유저 반응을 집중적으로 관리해 보세요.
#### **2) Early Access(앞서 해보기)**
정식 출시 전 수익 창출과 커뮤니티 형성을 동시에 시작할 수 있는 방법이에요.
**장기적인 팬덤 형성에 유리**하지만 몇 가지 고려할 점이 있어요. 타 스토어에 먼저 정식 출시한 게임은 스팀에서 Early Access로 출시할 수 없어요.
또한 Early Access 기간의 평가와 리뷰는 장기간 누적되기 때문에, 어느 정도 완성도를 갖춘 시점에 진입하는 게 좋아요.
### **3\. 커뮤니티는 게임 출시 전부터 만들어 가세요**
[9.webp]
커뮤니티 운영은 출시 이후보다 출시 전에 진행하면서 적극적으로 유저와 소통하는 게 중요해요.
최혜린 님은 특히 해당 시기의 커뮤니티 운영의 중요성을 강조하면서, “개발 과정부터 유저들과 적극적으로 소통하고, 그 과정이 출시 후 입소문으로 이어질 수 있는 기반을 만들어야 한다”고 언급했어요.
또한 “모바일은 커뮤니티 없이 운영되는 게임도 많고, 마켓 리뷰에 매크로성 답변이 달리는 경우도 많지만, 스팀에서는 유저들이 게임을 플레이하며 느낀 점과 건의 사항을 올리면 개발사가 이를 확인하고 반영하고 있다는 걸 직접 보여줘야 한다”고 덧붙였어요.
혜린 님은 가장 좋은 사례로 민트로켓의 Dave the Diver를 꼽으며 **얼리 액세스 8개월 동안 약 25회의 게임 개선을 진행, 영상·글 등 다양한 형태로 12회의 개발자 노트를 발행**한 사실을 공유했어요.
민트로켓은 업데이트마다 한국어·영어로 유튜브 방송을 진행했고, 디스코드(Discord)에서 유저와 직접 소통했어요.
게다가 얼리 액세스 단계부터 리뷰 영상을 올려준 북미 인플루언서를 NPC로 직접 게임에 넣어준 것도 화제가 됐죠. 커뮤니티 소통이 단순한 홍보가 아니라 게임 안으로 들어온 사례로 이 업데이트를 통해 유저의 긍정적인 반응을 확인했다고 해요.
### **4\. 인플루언서는 도달보다 전환 기여도로 선정하세요**
[10.webp]
팔로워 수보다 **게임 장르와 플레이스타일이 맞는 마이크로·미드 티어 인플루언서 중심**으로 접근하는 게 효과적이에요. 테스트 빌드를 제공해 진정성 있는 콘텐츠를 만들고, 스팀 큐레이터도 함께 활용하면 유저 신뢰도를 높일 수 있어요.
콘텐츠 타이밍도 중요해요. **플레이테스트부터 Next Fest, 출시 직전 1~2주에 콘텐츠 발행을 집중해서 위시리스트 증가 속도의 피크 구간**을 만들어야 해요.
크리에이터별 전용 트래킹 링크로 유입과 전환을 측정하고, 전환 기여도가 높은 크리에이터와 협업을 늘려가는 방식이 효과적이에요.
## **스팀 게임 마케팅 성과, 대형 게임사도 선택한 에어브릿지로 측정하세요**
[12.webp]
스팀 마케팅 전략을 잘 실행했다면, 그 성과를 데이터로 확인하는 것도 중요하죠.
현재 **컴투스, 스마일게이트, NHN, 드림에이지, Nimble Neuron, 엠게임** 등 국내 주요 게임사들이 **에어브릿지(Airbridge)** 로 스팀 게임 데이터를 측정하고 있어요.
스팀과 PC 게임은 모바일과 데이터를 쌓는 방식 자체가 달라, 측정 방법도 다르게 접근해야 하는데요.
모바일은 마케터가 측정하고 싶은 이벤트와 SDK 가이드를 클라이언트 개발자에게 전달하는 방식이에요.
PC 트래킹은 Server-to-Server (S2S) Event API 방식으로, 측정하고 싶은 이벤트와 API 규격을 게임 서버 개발자에게 전달하고 게임 서버에서 에어브릿지 서버로 데이터를 전송하는 구조예요.
[11.webp]
에어브릿지는 Web SDK와 Server-to-Server (S2S) Event API를 활용해 터치포인트와 전환 데이터를 수집해요. [**광고 클릭부터 게임 설치, 결제까지 이어지는 유저 여정을 크로스 플랫폼 ID 통합과 기여도 분석으로 추적**](https://www.airbridge.io/ko/solutions/gaming)할 수 있어요.
채널별 기여도도 명확하게 확인할 수 있어서, **모바일에서 익숙한 방식으로 스팀 게임 성과를 측정**할 수 있어요.
데이터는 대시보드 추출, HTTP 웹훅, Batch 데이터 덤프 등 다양한 방식으로 가져갈 수 있어서 기존 데이터 파이프라인에 맞게 연동할 수 있죠.
다만 실제 연동 과정은 마케터가 혼자 진행하기 쉽지 않을 수 있어요. 로직을 잘 알고 있어도 같이 설득해 주는 솔루션사 사람도 필요하고, 우리 게임에 맞는 이벤트를 함께 설계하는 것도 중요하죠.
[**에어브릿지는 이미 20개 이상의 PC 게임 연동 경험을 바탕으로 이벤트 설계부터 연동까지 함께 지원**](https://www.airbridge.io/ko/solutions/pc-console-gaming)하고 있어요.
PC/콘솔 게임 트래킹이 필요하거나 관심 있다면, 에어브릿지 게이밍팀에 문의해 보세요.
[appier Game UA_04.webp]
## **PC 게임 마케팅 전략, 에어브릿지로 극대화하세요**
아무리 좋은 전략을 실행해도, 데이터가 뒷받침되지 않으면 무엇이 효과적인지 알 수 없죠.
광고 클릭부터 게임 설치, 결제까지 이어지는 유저 여정을 측정할 수 있어야 전략을 다듬고 성과를 높일 수 있을 텐데요.
최혜린 님은 마지막으로 “게임사에 있을 때 운영·개발·마케팅·사업 모든 인원이 함께 스팀을 공부하고 마케팅 전략을 설계했던 기억이 있다. 이번 발표를 준비하면서 그 경험이 많이 떠올랐다”며 “오늘 발표한 내용이 스팀 마케팅을 처음 준비하는 분들께 실질적인 도움이 됐으면 한다”고 이야기의 끝을 맺었어요.
PC·콘솔 게임 진출을 준비하고 있다면, 지금 바로 에어브릿지 게이밍팀에 문의해 보세요.
스팀 마케팅을 직접 경험한 팀으로서, PC 게임 성과를 정확하게 측정할 수 있도록 도와드릴게요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 페이월 없는 MMP를 찾는다면? 기본 기능 3가지를 꼭 확인하세요
URL: https://www.airbridge.io/ko/blog/how-to-choose-mmp-three-functions-without-paywall
Language: Korean
Published: 2026-06-17
Author: Team Airbridge
광고비는 쓰고 있지만 어떤 채널이 실제로 전환을 만드는지 모를 때, 마케터는 MMP(모바일 측정 파트너, Mobile Measurement Partner) 도입을 검토하게 돼요.
[MMP는 앱 설치부터 인앱 이벤트, 구매 전환까지 데이터를 수집하고 어떤 광고 채널이 전환에 기여했는지 측정하는 역할을 하는 솔루션](https://www.airbridge.io/ko/blog/what-is-mmp)으로, 모바일 앱을 운영하는 마케터라면 대부분 도입을 검토하거나 이미 사용하고 있는 툴이에요.
하지만 막상 MMP 도입을 검토하다 보면 비용 구조가 예상보다 복잡하다는 걸 알게 돼요. 당장 ROAS(광고비 대비 매출)도 불안한 상황에서 무료 MMP를 찾게 되는데, 그러다 보면 꼭 필요한 기능을 갖추는 데 생각보다 비용이 든다는 걸 알게 되죠.
무료 플랜은 막상 써보면 채널 어트리뷰션은 가능하지만, 인앱 이벤트 추적은 상위 플랜으로 변경해야 할 거예요. 로우 데이터 내보내기는 별도 계약을 해야 하거나, 광고 사기 방지는 애드온 같은 방식으로 비용이 따로 붙죠.
이렇게 필요한 기능을 하나씩 추가하다 보면, 처음 예상했던 바와 다른 비용이 들어요.
MMP 도입을 검토하는 팀이 결국 원하는 건 완전한 무료 플랜이라기보다, **핵심 기능을 쓰기 위해 추가 비용을 내지 않아도 되는 제품, 즉 페이월 없는 MMP**일 거예요.
이 글에서는 MMP를 도입하는 마케터가 알아야 할 **기본 플랜에서 작동해야 하는 필수 기능 3가지**에 대해서 알아볼게요. 각 기능이 왜 추가 비용 없이 제공돼야 하는지 짚어보고, 합리적인 가격에 사용할 수 있는 MMP를 찾는다면 참고해 보세요.
**📌 핵심 포인트**
- MMP 기본 플랜에서 확인해야 할 기능은 인앱 이벤트 추적, 원본 데이터 접근, 광고 사기 방지예요. 이 세 가지가 빠지면 측정은 되지만, 예산 효율을 정확하게 판단하기 어려워요.
- 기능별로 애드온 없이 처음부터 포함된 과금 구조는 캠페인 규모가 커져도 비용 범위가 예상을 크게 벗어나지 않아요.
- 에어브릿지는 이 기능들을 기본 플랜에 포함하고, MAU 기반과 어트리뷰티드 인스톨 기반 중 서비스에 맞는 과금 방식을 직접 선택할 수 있어요. 측정 범위가 넓어져도 과금 구조는 처음과 같아요.
## **MMP 기본 플랜에서 무엇을 볼 수 있어야 하나요?**
### **1\. 설치 수 이상의 전환 흐름이 나오는지 확인하세요**
MMP를 고를 때 가격만큼 중요한 건 기본 플랜에서 실제 어떤 데이터를 볼 수 있는지 확인하는 거예요.
MMP가 보여주는 데이터를 이해하려면, 먼저 핵심 개념인 어트리뷰션을 알아야 하는데요.
[어트리뷰션(Attribution)](https://help.airbridge.io/ko/guides/airbridge-attribution-model)은 유저가 광고를 보고 앱을 설치하거나 구매했을 때, 해당 전환이 이뤄진 특정 채널과 광고를 추적하는 작업을 뜻해요.
이를 통해 같은 날 메타 광고를 클릭하고 앱을 설치한 유저 1,000명과 구글 광고를 통해 들어온 유저 500명이 있을 때 어느 채널에서 온 유저가 실제로 제품이나 서비스를 구매했는지 채널별로 구분해 볼 수 있죠.
어트리뷰션은 이러한 판단을 정확하게 할 수 있도록 해주며, 이는 광고비 최적화와 효율화에 기여해요.
정리하면, MMP는 이 어트리뷰션을 앱과 웹 전반에서 자동으로 측정하고, 채널별 성과 데이터를 한 곳에서 볼 수 있게 해주는 플랫폼이라고 할 수 있어요.
그런데 막상 MMP를 도입한 후 사용해 보면, **설치 수는 채널별로 확인할 수 있지만 그 이후 데이터는 볼륨에 따라 추가 과금이 붙거나 상위 플랜에서만 제공**되는 경우가 있어요.
마케터가 실제로 확인하고 싶은 데이터와 기본 플랜에서 볼 수 있는 데이터가 다른 거죠.
**마케터가 기본 플랜에서 기대하는 것**
- 채널별 설치 수와 인앱 전환율을 함께 비교
- 특정 채널에서 온 유저가 구매까지 이어지는지 확인
- 인앱 이벤트 볼륨에 상관없이 전체 흐름을 추적
**일반적인 기본 플랜에서 제공되는 데이터 범위**
- 채널별 설치 수 확인
- 설치 이후 행동은 별도 설정 또는 상위 플랜 필요
- 이벤트 볼륨에 따라 추가 과금 가능
### **2\. 인앱 이벤트 추적이 없으면 어트리뷰션은 절반만 작동해요**
설치 10,000건을 만든 채널과 설치 3,000건이지만 구매 전환율이 높은 채널, 어디에 예산을 더 써야 하는지는 인앱 이벤트 데이터가 어트리뷰션과 연결될 때 알 수 있어요.
인앱 이벤트는 앱 설치 이후 유저가 앱 안에서 취하는 행동을 뜻해요. 커머스 앱이라면 장바구니 담기나 구매 완료, 구독 앱이라면 구독 시작이나 결제 완료처럼 서비스마다 추적하는 행동이 달라요.
이 데이터가 어트리뷰션과 연결돼야 ‘메타에서 들어온 유저 중 구매까지 이어진 비율’ 같은 지표를 확인할 수 있어요. 그래야 어떤 채널에 예산을 더 쓰거나 축소할지 결정할 수 있고요.
**이벤트 기반으로 과금하는 구조에서는 마케팅을 집행하지 않아도 유저가 앱을 쓰는 한 이벤트가 계속 발생**하기 때문에, 추적하는 이벤트 범위를 넓힐수록 MMP에 지불하는 비용이 커질 수 있어요.
이러한 구조라면 마케터는 필요한 이벤트를 전부 추적하지 못하는 선택을 하게 될 수 있죠.
## **MMP가 측정한 데이터는 직접 추출할 수 있어야 해요**
### **1\. 원본 데이터가 없으면 MMP 대시보드 안에서만 분석이 가능해요**
MMP가 측정한 데이터는 대시보드 안에서 요약된 형태로 보여요.
유저의 설치, 인앱 이벤트, 어트리뷰션 데이터를 수집하고, 대시보드에서는 이를 채널별 ROAS, 리텐션 수치처럼 집계된 형태로 보여주죠.
그런데 그 집계의 원재료인 원본 데이터를 직접 추출하려면 별도 계약이 필요한 경우가 있어요.
원본 데이터는 집계되기 전 상태의 이벤트 데이터인데, 유저별 설치 시점, 인앱 이벤트 발생 시각, 어트리뷰션 원본 기록처럼 대시보드 숫자 뒤에 있는 원재료라고 할 수 있어요.
내부 BI(Business Intelligence, 비즈니스 인텔리전스 도구) 툴에 데이터를 적재하거나, 자체 분석 파이프라인을 구성하거나, 채널별 퍼포먼스를 더 세밀하게 분석하려면 이 원본 데이터가 필요하죠.
일부 MMP에서는 로우 데이터 API 접근을 30일 트라이얼처럼 제한된 기간 동안만 제공하고, 이후에는 상위 플랜이나 별도 애드온 계약을 해야 사용할 수 있어요.
이렇게 되면 대시보드에서 집계된 수치는 볼 수 있어도, 그 수치를 만든 원본 데이터는 별도로 추출하지 못하는 상황이 생겨요.
### **2\. 데이터를 추출할 수 없으면 외부 연동도 어려워요**
원본 데이터를 추출하지 못하는 구조는 분석 범위뿐 아니라 나중에 다른 툴과 연동하려 할 때도 문제가 될 수 있어요.
원본 데이터를 직접 추출할 수 없으면 그동안 MMP 안에 쌓인 이벤트 이력과 어트리뷰션 기록을 외부로 가져오지 못해요. 새 환경에서는 처음부터 데이터를 쌓아야 하고, 이전 캠페인과의 비교·분석에도 장벽이 생기죠. 데이터는 계속 측정하고 있었지만 정작 그 데이터를 온전히 활용하지 못하게 되는 거죠.
MMP를 고를 때는 설치당 단가나 플랜 가격만큼 **원본 데이터를 직접 추출할 수 있는지, 데이터 웨어하우스 연동이 지원**되는지 확인해야 해요. 도입 초기에는 잘 보이지 않다가, 데이터가 쌓이고 분석 환경이 고도화될수록 중요해지는 기준이에요.
## **광고 사기는 유입 전에 필터링돼야 해요**
### **1\. 사기 탐지가 없으면 유저가 없는 곳에 광고비가 소진돼요**
광고를 집행하면 실제 유저가 광고를 보고 클릭해서 앱을 설치하는 게 정상적인 흐름이에요. 그런데 모바일 광고 환경에서는 이 흐름을 가로채는 방식의 사기가 발생하기도 해요.
대표적인 게 클릭 인젝션(Click Injection)이에요.
디바이스에 설치된 악성 앱이 오가닉 유저가 앱을 다운로드하고 실행하는 사이 구간에 가짜 클릭을 발생시키는 방식인데요. 이를 통해 광고와 무관한 설치를 자신의 광고 성과로 가져가요.
또는 SDK 스푸핑(SDK Spoofing)과 같이 어트리뷰션 툴의 SDK를 직접 해킹해 가짜 설치 데이터를 만들어내는 방식을 쓰기도 해요.
두 경우 모두 실제 유저 유입 없이 광고 성과가 발생한 것처럼 데이터가 기록되고, 광고주는 그 성과에 대한 비용을 지불하게 돼요.
이런 사기가 탐지되지 않으면 어트리뷰션 데이터상으로는 캠페인이 정상적으로 성과를 내고 있는 것처럼 보여요. 실제로는 전환 없는 트래픽에 예산이 집행되고 있어 사기 탐지 기능이 없으면 가짜 데이터와 성과를 확인할 수 없어요.
### **2\. 사기 방지는 캠페인 시작 전부터 적용돼야 해요**
이렇게 광고 사기는 캠페인이 시작되는 순간부터 발생해요. 첫 번째 클릭, 첫 번째 설치부터 사기 트래픽이 섞일 수 있어 [**사기 방지 기능**](https://help.airbridge.io/ko/guides/injection-prevention)**을 추후에 추가하면 그 이전에 쌓인 데이터는 검증할 방법이 없죠.**
특히 사기 방지가 애드온 구조라면 도입 시점이 늦어지는 경우가 생겨요.
처음에는 비용 부담 때문에 미루다가, 데이터가 어느 정도 쌓이고 나서야 추가하게 되면서 그사이에 집행된 캠페인 데이터는 사기 트래픽이 걸러지지 않은 채로 남죠.
하지만 어떤 채널에서 사기가 발생했는지, 어느 시점부터 데이터를 신뢰할 수 있는지 소급해서 확인하기 어렵기 때문에 측정은 캠페인 첫날부터 정확해야 해요. 사기 방지가 플랜에 기본적으로 포함되는지 꼭 확인하세요.
## **에어브릿지는 페이월 없이 핵심 MMP 기능을 기본 플랜에 제공해요**
### **1\. 무료 플랜에서도 추가 과금 없이 사용하세요**
앞서 살펴본 세 가지 기능인 인앱 이벤트 추적, 원본 데이터 접근, 광고 사기 방지는 MMP를 도입하는 목적 자체와 직결돼요. 에어브릿지는 이 기능들을 기본 플랜에서 제공하고 있어요.
| 기능 | A사 | B사 | 에어브릿지 |
| --- | --- | --- | --- |
| 인앱 이벤트 추적 | 볼륨에 따라 추가 과금 가능 | 유료 플랜 포함 | 기본 포함 |
| 원본 데이터 내보내기 | 30일 트라이얼 후 애드온 또는 엔터프라이즈 | 전 플랜 포함 | 기본 포함 |
| 광고사기 방지 | Protect360 별도 애드온 | 유료 플랜 포함 | 기본 포함 |
| 딥링크 | 기본 포함 | 기본 포함 | 기본 포함 |
특히 에어브릿지는 주요 3가지 기능을 포함해 측정에 필요한 기능 전체를 기본 플랜에서 사용할 수 있어요.
인앱 이벤트 추적은 스탠다드 이벤트와 커스텀 이벤트 모두 추가 비용 없이 설정할 수 있고, 퍼널 분석, 코호트 분석, ROAS & 리텐션 분석까지 포함돼요.
원본 데이터는 대시보드에서 직접 CSV로 추출하거나 실시간 이벤트 로그로 확인할 수 있어요. [광고 사기 방지](https://www.airbridge.io/ko/product/fraud-protection)는 클릭 인젝션, SDK 스푸핑을 포함한 비정상 설치 탐지가 캠페인 시작 시점부터 작동해요.
그 밖에 [딥링크](https://www.airbridge.io/ko/deeplink-plan)는 광고를 클릭한 유저를 앱 내 특정 페이지로 바로 연결하는 기능이에요.
앱이 설치되지 않은 유저는 스토어로 이동한 뒤 설치 후 해당 페이지에 자동으로 연결돼요. 유저 경험을 끊기지 않게 이어주는 기능으로, 리타겟팅 캠페인이나 프로모션 페이지 연결에 주로 활용해요.
### **2\. 과금 방식은 두 가지 중에서 선택할 수 있어요**
대부분의 MMP는 과금 모델이 하나로 고정돼 있어요. 에어브릿지는 서비스 특성에 따라 두 가지 중에서 선택할 수 있어요.
**MAU(월간 활성 유저) 기반**
- 매월 앱을 한 번 이상 사용한 고유 유저 수 기준으로 과금
- 시즌별 설치가 특정 시기에 몰리는 앱에 유리
- 오가닉, 웹 트래픽, 비활성 유저는 과금 제외
**어트리뷰티드 인스톨 기반**
- 유료 미디어를 통해 유입된 설치 수 기준으로 과금
- 꾸준히 UA 캠페인을 집행하는 앱에 유리
- 월별 또는 연별 버킷으로 선택 가능, 오가닉 설치는 과금 제외
**두 모델 모두 기능별 애드온이 없고, 숨겨진 비용 없이 처음부터 모든 기능이 포함**돼요.
캠페인 규모가 커지거나 측정 범위가 넓어져도 처음 계약할 때 예상한 비용 구조가 그대로 유지돼요. 기능을 추가할 때마다 비용을 재검토할 필요가 없어요.
## **페이월 없이 핵심 기능을 갖춘 MMP를 선택하세요**
MMP 도입을 가격 때문에 망설이고 있었다면, 기본 플랜에서도 충실히 MMP의 기능을 수행하는 에어브릿지를 사용해보세요.
인앱 이벤트 추적, 원본 데이터 접근, 광고 사기 방지가 처음부터 포함된 MMP라면 측정을 시작하는 데 과금에 대한 부담이 줄어들고, 캠페인이 커져도 비용 구조가 크게 바뀌지 않을 거예요.
데이터는 처음부터 온전히 우리 기업 소유로 자산화할 수 있죠.
페이월 없이 합리적인 가격에 핵심 기능이 모두 포함된 MMP의 MMP를 찾고 있다면 지금 에어브릿지 데모를 신청해 우리 서비스에 맞는 플랜을 확인해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 장기 매출을 만드는 채널을 찾는 LTV·pLTV 분석 가이드
URL: https://www.airbridge.io/ko/blog/ltv-pltv-analysis-guide-long-term-sales
Language: Korean
Published: 2026-06-17
Author: Team Airbridge
**“어떤 채널이 진짜 좋은 유저를 데려오고 있을까?”**
여러 광고 채널로 유저를 획득하다 보면, 구매 전환율이 높거나 리텐션이 긴 ‘고가치 유저’를 데려오는 채널을 파악하는 것이 중요해요.
하지만 앱 설치 수나 이번 달 **ROAS(Return on Ad Spend, 광고 수익률)** 만으로는 어느 채널이 장기 수익을 만드는지 파악하기 어렵죠.
이때 필요한 것이 **MMP(Mobile Measurement Partner, 모바일 측정 파트너)** 예요. 광고 클릭부터 앱 설치, 인앱 행동까지 경로를 추적해서 어떤 광고가 실제 수익으로 이어졌는지 기여도를 측정하는 솔루션을 말하는데요.
MMP는 특히 **코호트(Cohort) 분석**과 결합하면 강력해져요. ‘7월 첫째 주에 메타 광고를 통해 설치한 유저들’처럼 동일한 조건으로 상정한 집단을 시간 축 위에서 함께 추적할 수 있거든요.
이를 통해 유저가 **30일, 90일, 180일 뒤에 얼마를 쓰는지 LTV(Lifetime Value, 고객 생애 가치)로 환산해서 채널별로 비교**할 수 있어요.
즉, UA(User Acquisition, 유저 획득) 의사결정을 감이 아닌 데이터로 뒷받침할 수 있게 되는 거죠.
이번 글에서는 **MMP가 LTV를 계산하는 방식과 그 데이터를 UA 의사결정에 연결하는 방법을 짚어볼게요.** 장기 매출을 만들어내는 우리 기업의 광고 채널을 찾고 싶다면, 아래 방법을 참고해 보세요.
**📌 핵심 포인트**
- 모바일 앱에서 LTV는 설치일 기준 코호트로 봐야 해요. 특정 채널에서 온 유저가 30일, 90일, 180일 뒤에 얼마를 쓰는지 누적으로 추적할 수 있고, 이를 채널별로 비교하면 어떤 채널이 장기 매출을 만들어내는 유저를 데려오는지 확인할 수 있어요.
- D180 LTV 데이터가 쌓이길 기다리지 않아도 돼요. 예측 LTV(pLTV, Predictive Lifetime Value)를 활용하면 D7, D14 초기 데이터만으로도 채널 간 장기 수익성을 미리 비교하고 예산 방향을 잡을 수 있어요.
- 알람 앱 알라미를 운영하는 딜라이트룸은 Airbridge pLTV를 도입한 뒤 매체·캠페인 단위 예측 LTV를 대시보드에서 바로 확인할 수 있게 됐어요. 기존에 최대 3시간이 걸리던 데이터 가공 과정이 사라지면서 절약한 시간을 ROAS 개선 등 더 의미 있는 일에 쓰고 있죠.
## **MMP 맥락에서 LTV는 무엇을 의미하나요?**
LTV는 마케팅에서 광범위하게 쓰이는 용어지만, MMP가 다루는 LTV는 채널별·코호트별로 쪼개진 매출 데이터에서 출발해요. UA 의사결정에 쓰려면 정의와 집계 방식부터 다시 짚어야 해요.
### **1\. 모바일 앱 마케팅에서 LTV란?**
**LTV는 유저 한 명이 앱을 사용하는 전체 기간 동안 발생시키는 누적 수익을 뜻해요.** CLV(Customer Lifetime Value) 또는 CLTV라고도 하죠.
LTV가 없으면 CAC(Cost per Acquisition, 고객 획득 비용)를 얼마까지 써도 되는지 판단할 근거가 없어요.
예를 들어 설치당 비용이 5,000원인 채널이 있을 때, 그 채널에서 온 유저가 30일 안에 평균 3,000원을 쓴다면 손실이에요. 그런데 180일 기준으로 해당 유저가 1만 5,000원을 쓴다면 얘기가 달라지죠.
**이렇게 어느 시점까지의 LTV를 기준으로 삼느냐에 따라 채널 효율 판단이 달라져요.** 그리고 이 판단이 채널 단위로 할 수 있어야 비로소 의미가 생겨요. 전체 평균 LTV가 높아도, 수익을 만들어내는 유저가 특정 채널에서만 오고 있다면 마케팅 예산 분배가 달라져야 하니까요.
### **2\. 앱 내부 분석으로는 채널별 LTV가 안 나오는 이유**
그런데 채널별 LTV를 보려면 앱 내부 분석 툴만으로는 한계가 있어요.
앱 내부 분석 툴을 통해 특정 유저가 며칠 만에 구매했는지, 앱에 얼마나 자주 접속하는지 같은 데이터를 볼 수는 있어요. 링크에 태그를 붙이는 UTM 파라미터로 채널 유입 정도 또한 파악할 수 있죠.
하지만 모바일 앱에서는 그대로 적용되지 않아요. 웹과 달리 앱은 광고 클릭과 앱 실행 사이에 앱스토어 설치 여정이 포함되어 있어요. 이 구간에서 어떤 광고 소재를 클릭한 유저가 설치까지 완료했는지, 이후 구매로 이어졌는지를 앱 내부 툴만으로는 연결하기 어렵죠.
iOS ATT(앱 추적 투명성) 정책으로 기기 식별자 접근이 제한된 이후 이 문제는 더 심화됐고요.
MMP는 이 간극을 메울 수 있어요.
**광고 노출부터 클릭, 설치, 인앱 행동까지 경로를 추적하는 어트리뷰션(Attribution, 광고 기여도 측정)을 제공**하며, 여기에 설치일 기준 코호트 분석을 결합하면 **채널별로 유저가 시간이 지남에 따라 얼마를 쓰는지 LTV로 환산해서 비교**할 수 있는 구조가 만들어져요.
### **3\. 같은 기간을 설정했는데 두 매출이 다르다면?**
다만 이 데이터를 제대로 읽으려면 한 가지를 먼저 짚어야 해요. MMP 대시보드에서 동일한 날짜 범위로 매출을 조회해도 코호트 매출액과 총매출액이 다르게 나오는 경우가 있거든요.
이는 두 메트릭이 서로 다른 날짜 기준으로 집계되기 때문인데요.
**총매출액**은 구매, 구독 등 **수익 이벤트가 실제로 발생한 날짜(Event Date)로 집계**해요. 7월에 여러 채널에서 유저 획득을 집중적으로 진행한 뒤 8월에 멈췄다고 가정해 볼게요.
유입된 유저들이 8월에 구매를 완료하면 총매출액 기준에서는 8월 수치가 높게 측정돼요. 반면 8월 신규 앱 설치 수는 적은데 구매 수는 많으니 전환율이 비정상적으로 높게 집계돼요.
즉, 유저 획득 시점과 수익 발생 시점이 다른 달에 걸쳐 있을 때 생기는 왜곡이에요.
**코호트 매출액**은 **앱 설치 시점을 기준**으로 유저를 묶어서 집계해요. 예를 들면 7월에 앱을 설치한 유저들을 하나의 코호트로 보고, 이 집단 이후 발생하는 모든 수익을 7월 코호트 매출로 귀속시키는 방식이죠.
8월에 구매가 일어났더라도 구매한 유저가 7월에 앱을 설치했다면 7월 코호트 매출로 집계돼요. 이로 인해 유저 획득 시점과 수익 발생 시점이 달라도 채널별 성과를 왜곡 없이 추적할 수 있어요.
어느 메트릭을 봐야 하는지는 분석 목적에 따라 달라질 수 있으며, **UA 성과를 채널별로 정확히 보려면 코호트 매출액**을, **특정 기간 전체 매출을 파악할 때는 총매출액을 기준**으로 해석해 보세요.
| 분석 목적 | 권장 메트릭 |
| --- | --- |
| UA 채널별 유저 품질 비교 | 코호트 매출액 (설치일 기준) |
| 특정 기간 전체 매출 집계 | 총매출액 (Event Date 기준) |
| 캠페인 ROAS 측정 | 코호트 ROAS (설치일 기준) |
| 월간 수익 리포팅 | 총매출액 (Event Date 기준) |
## **어떤 LTV 지표를, 언제 봐야 하나요?**
LTV는 단일 숫자로 보는 것이 아닌, 특정 시점까지의 누적 매출을 기준으로 삼느냐에 따라 다른 의사결정을 뒷받침해요. 우리 앱에 맞는 기준 시점을 잡는 방법부터, 데이터가 쌓이기 전에 의사결정을 내려야 할 때 쓰는 예측 LTV까지 짚어볼게요.
### **1\. 수익 발생 시점에 따라 D 기준점 잡기**
MMP의 코호트 리포트는 설치일 기준으로 유저를 그룹화하고, 이 집단이 시간이 지남에 따라 만들어내는 누적 수익을 D7, D30, D90, D180 단위로 보여줘요. 어느 시점을 기준으로 삼아야 하는지는 우리 앱에서 수익이 발생하는 시점을 먼저 파악하는 데서 출발해요.
기준이 되는 건 **첫 번째 주요 전환 이벤트**예요. 구독 앱이라면 트라이얼에서 유료 전환이 일어나는 시점, 게임이라면 첫 과금이 발생하는 시점이죠.
앱마다 수익 발생 패턴이 다르기 때문에 정답은 없으나, 참고할 수 있는 기준을 잡는다면 첫 번째 주요 전환 이벤트가 발생하는 시점을 기준으로 그 이후를 포함하는 D를 선택하는 방식이에요.
| 주요 전환 이벤트 발생 시점 | 참고 LTV 기준 시점 | 필요한 최소 조회 기간 |
| --- | --- | --- |
| 설치 후 7일 이내 | D14 ~ D30 | 30일 이상 |
| 설치 후 2~4주 | D30 ~ D60 | 60일 이상 |
| 설치 후 1~2개월 | D60 ~ D90 | 120일 이상 |
| 설치 후 2개월 이상 | D90 ~ D180 | 210일 이상 |
### **2\. 예측 LTV(pLTV)의 활용 시점은?**
코호트 기반 LTV는 데이터가 실제로 쌓인 뒤에야 볼 수 있어요. 하지만 D180 LTV와 같이 집계 기간이 긴 데이터는 확인하려면 180일을 기다려야 하는데, 실무에서는 그 전에 결정을 내려야 하는 상황도 생기죠.
**예측 LTV(pLTV, Predictive LTV)** 는 이 때 활용할 수 있는 지표예요.
초기 시점, 예를 들어 D7이나 D14의 리텐션과 수익 패턴을 분석해서 이 유저 집단이 D90, D180까지 얼마를 쓸지 수학적으로 추산하거든요. 즉, **실제** **데이터가 충분히 쌓이기 전에 채널 간 수익성을 비교하고 예산을 활용할 방향을 잡는 데 참고**할 수 있죠.
아래와 상황에서 해당 지표를 쓸 수 있어요.
- 캠페인 초기에 예산 확대 여부를 빠르게 결정해야 할 때
- 여러 채널을 동시에 운영하면서 어디에 집중할지 우선순위를 정해야 할 때
- 신규 오디언스나 크리에이티브를 테스트한 직후 초기 성과를 평가해야 할 때
다만 **pLTV를 쓸 때 염두에 둬야 할 점**은 다음과 같아요.
- 초기 데이터가 충분히 쌓이지 않은 상태일 때는 오차가 커질 수 있어요.
- 채널 간 상대적 수익성을 비교해서 예산을 배분할 채널의 우선순위를 정할 때 효과적이에요.
- 실제 코호트 LTV가 쌓였을 때 pLTV와 대조해서 예측 정확도를 점검하는 게 좋아요.
## **LTV 데이터로 채널 예산을 최적화해 보세요**
상황에 따라 참고해야 하는 LTV 지표를 파악했다면, 다음으로는 해당 데이터를 활용해 채널별로 예산을 어떻게 분배할지 결정해야 하죠.
LTV는 역시 단독으로 보는 지표가 아니며, 광고비와 함께 봐야 어떤 채널이 장기 매출을 만들어내는 유저를 데려오는지, 그 채널에 예산을 더 써도 되는지 판단할 수 있어요.
코호트 ROAS와 채널별 LTV 비교를 참고해 보세요.
### **1\. 코호트 ROAS로 채널 손익분기점 잡기**
단기 ROAS는 특정 기간 광고비 대비 발생한 매출을 비교하죠. 캠페인 직후 성과를 빠르게 파악하는 데 참고하지만, 수익이 장기에 걸쳐 발생하는 앱에서는 해당 지표로만 채널 효율을 판단하기 어려워요.
그럴 때 코호트 ROAS를 참고해 보세요. 설치일 기준으로 유입된 유저 집단이 시간이 지남에 따라 쓴 비용을 누적으로 추적하기 때문에, 단기 ROAS에서 보이지 않던 채널의 장기 수익성이 보여요.
코호트 ROAS는 아래처럼 계산해요.
- **D30 코호트 ROAS = D30까지 누적 매출 / 해당 코호트 획득 광고비**
D30 코호트 ROAS가 낮아도 D90에서 높게 나온다면, 수익이 초반보다 시간이 지날수록 쌓이는 채널이라고 해석할 수도 있죠.
### **2\. 채널·크리에이티브별 LTV 비교 방법은?**
코호트 ROAS로 손익분기점을 파악했다면, 다음은 어떤 채널과 소재가 실제로 장기 매출을 만들어내는 유저를 데려오는지 비교해야 하는데요.
이는 같은 예산을 써도 채널과 크리에이티브에 따라 유입되는 유저의 수익 패턴이 달라지기 때문이에요.
예를 들어 채널별로 D30, D90 LTV를 나열해서 보면 이 차이가 수치로 확인돼요.
CPI(Cost Per Install, 설치당 비용)가 높은 채널이라도 D30 LTV가 높다면 CAC 대비 회수되는 수익이 더 클 수 있어요.
반대로 리워드 광고나 이벤트성 캠페인처럼 설치 수를 빠르게 끌어 올리는 채널은 혜택받고 이탈하는 유저가 많아서 D30 LTV가 낮게 나올 수 있죠.
크리에이티브 단위에서도 같은 방식으로 분석할 수 있어요.
동일한 채널에서 집행한 소재라도 어떤 메시지로 유입됐느냐에 따라 유저의 수익 패턴이 달라져요.
게임 플레이 영상으로 유입된 유저와 할인 프로모션 배너로 유입된 유저의 D30 LTV를 비교하면, 어떤 소재가 실제 매출로 이어지는 유저를 데려오는지 확인할 수 있죠.
아래는 이해를 돕기 위한 가상 시나리오예요.
| 채널 | CPI | D30 LTV | D30 코호트 ROAS | 해석 |
| --- | --- | --- | --- | --- |
| 채널 A | 3,000원 | 9,000원 | 3.0 | 효율 우수 |
| 채널 B | 1,000원 | 1,500원 | 1.5 | 설치 수는 많지만 LTV 낮음 |
| 채널 C | 5,000원 | 4,000원 | 0.8 | D90까지 추적 필요 |
채널 A는 CPI가 채널 B보다 3배 높지만, D30 LTV가 6배 높아요. CPI만 보고 채널 B에 예산을 집중하면 실제로는 더 효율적인 채널을 놓치는 셈이죠.
채널 C는 D30 코호트 ROAS가 1 미만이지만 수익이 장기에 걸쳐 발생하는 구조일 수 있어서 D90까지 추적해 봐야 판단할 수 있다고 해석할 수 있어요.
## **에어브릿지 예측 LTV로 성과를 내는 팀은 이렇게 일하고 있어요**
딜라이트룸 알라미 팀은 예측 LTV(pLTV, Predictive LTV) 도입으로 데이터 가공에 쓰던 시간을 마케팅 의사결정에 쓸 수 있게 됐어요. 에어브릿지가 제공하는 LTV·pLTV 데이터를 통해 성과를 내는 방법을 알아보세요.
### **1\. 예측 LTV 계산에 매번 3시간 쓰고 있다면?**
지금까지 예측 LTV 계산에 시간을 쓰고 있던 마케터라면 다음 사례에 주목하세요.
**알람 앱 알라미(Alarmy)를 운영하는 딜라이트룸**은 7,500만 다운로드, 97개국 앱스토어 카테고리 1위를 기록한 글로벌 서비스예요. 알라미 마케팅팀은 국가·OS·매체별 예측 LTV를 직접 계산해 마케팅 전략을 최적화해 왔어요.
문제는 세분화된 단위로 볼수록 데이터를 직접 가공하는 데 드는 시간이 늘어나, 한 번 계산하는 데 최대 3시간이 소요됐어요. 의사결정에 써야 할 시간이 데이터 가공에 쓰이는 상황이었죠.
에어브릿지 pLTV를 도입한 뒤로는 **별도 가공 없이 국가·OS·매체·캠페인 단위 예측 LTV를 레비뉴 리포트 대시보드에서 바로 확인**할 수 있게 됐어요.
절약한 시간은 ROAS를 개선하는 구체적인 방안을 고민하는 등 더 의미 있는 일에 쓰이고 있어요. pLTV 기반 예산 배분으로 마케팅 비용도 절감했죠.
👉 [딜라이트룸, Airbridge pLTV로 마케팅 전략 최적화한 사례 보기](https://cshub.ab180.co/ko/case-studies/airbridge-pltv-delightroom)
### **2\. 에어브릿지 LTV와 pLTV로 마케팅 전략을 최적화 하세요**
에어브릿지는 매출 성과를 설치일 기준 코호트로 추적하는 [레비뉴 리포트(Revenue Report)](https://help.airbridge.io/ko/guides/understanding-the-revenue-report)와 유저 잔존율을 분석하는 [리텐션 리포트(Retention Report)](https://help.airbridge.io/ko/guides/retention-report-viewing)에서 LTV와 pLTV를 함께 제공해요.
캠페인을 집행한 지 3일만 지나도 최대 180일 pLTV를 추산할 수 있어서, 데이터가 쌓이길 기다리기 전에 채널 간 수익성을 비교하고 예산 방향을 잡을 수 있어요.
주요 기능을 정리하면 아래와 같아요.
- **최소 3일치 리텐션 데이터로 최대 180일 pLTV 추산**
초기 캠페인에서도 장기 수익성 방향을 빠르게 확인할 수 있어요.
- **채널·캠페인·크리에이티브·국가·OS 단위 pLTV 인사이트**
어떤 세그먼트에서 장기 매출을 만들어내는 유저가 유입되는지 세분화해서 볼 수 있어요.
- **IAP·IAA 수익원별 코호트 LTV**
인앱 구매(IAP)와 인앱 광고(IAA) 수익을 구분해서 LTV를 추산해요.
- **코호트 조회 기간 최대 184일**
D180 LTV 분석에 필요한 데이터를 한 번에 확인할 수 있어요.
장기 매출을 만들어내는 유저를 데려오는 채널을 파악하고, 그 채널에 예산을 집중할 수 있게 되면 UA 효율이 달라져요. Airbridge LTV·pLTV 기능으로 지금 바로 실험해 보세요.
## **우리 기업의 비즈니스 지속 가능성을 확인하세요**
지금까지 MMP가 LTV를 어떻게 집계하고 분석하는지, 그 데이터를 채널 예산 배분에 연결하는 방법까지 살펴봤어요.
설치 수와 단기 ROAS만으로는 보이지 않던 것들은 LTV 데이터로 확인할 수 있는데요. 이 데이터를 통해 장기 매출을 만들어내는 유저를 데려오는 채널, 광고비가 회수되는 시점, 실제 수익으로 이어지는 소재 등을 점검할 수 있죠.
지금 운영 중인 채널 중 어떤 채널이 6개월 뒤에도 우리 비즈니스에 기여할지 궁금하다면, 지금 바로 에어브릿지에서 직접 검증해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 에어브릿지 벤치마크 리포트 출시, 동종업계는 어떻게 채널 전략을 짜고 있을까요?
URL: https://www.airbridge.io/ko/blog/airbridge-benchmark-report-release
Language: Korean
Published: 2026-06-12
Author: Jaehyuk Kim
> “다른 앱은 어떻게 마케팅 채널 전략을 짜고 있을까?”
마케팅을 하다 보면 가끔 이런 생각이 들 때가 있죠. '우리만 채널 너무 좁게 돌리는 거 아닐까?', '저 회사는 라이징 채널을 어떻게 저렇게 빨리 잡아냈지?' 매번 의문이 생기는데 자료로 답을 찾기는 어려웠을 거예요. 기껏 찾은 최신 자료도 글로벌 데이터라서, 한국 시장에서 써먹기 어려웠죠.
**그래서 이번에 에어브릿지에서 <한국 마케터를 위한 2026 벤치마크 리포트>를 발행했어요.**
한국 데이터만 사용한 업종별 데이터를 확인할 수 있는데요. 우리 업종 평균은 채널 몇 개를 돌리는지, 다른 회사들이 어떤 라이징 채널을 잡고 있는지, 우리 위치는 분포 어디쯤인지를 한 페이지에서 확인할 수 있어요.
**📌 이런 분에게 특히 도움이 돼요**
- 우리 업종 1위 채널이나 놓치고 있는 라이징 채널 발굴하려는 **UA·퍼포먼스 마케터**
- 마케팅 채널 포트폴리오·예산 배분의 객관적 기준을 알고 싶은 **마케팅 리더·의사결정자**
- 업종별 표준 채널 세트로 초기 미디어 플랜을 설계해야 하는 **신규 앱·서비스 론칭팀**
- 재방문(리인게이지먼트) 채널 전략을 설계해야 하는 **그로스 담당자**
## **같은 업종이어도 한국 데이터만 보면 채널 1위가 달라져요**
국내 마케팅 전략을 세울 때 동종업계 자료를 리서치하는 경우가 많은데요. 막상 자료를 찾아도 활용하기 어려운 경우가 많아요. 대부분의 벤치마크 자료들은 한국과 일본, 넓게는 아시아 전체를 한 지역으로 묶어 산출한 자료이기 때문이에요.
그런데 두 시장의 마케팅 채널 구성은 사실 굉장히 달라요. 한국 데이터만 분석하면 같은 업종이어도 마케팅 채널 비중 1위가 달라져요.
한국에는 네이버 검색 광고와 카카오 비즈보드, 버즈스크린(BUZZSCREEN), 모비온(MOBON) 등 한국 시장에 맞는 고유 채널이 있어요. 일본 역시 일본 시장에 맞는 마케팅 채널이 있고요. 두 시장을 하나로 묶다 보니 정작 한국 마케팅 시장에 대한 정확한 인사이트는 얻을 수 없다는 문제가 있었어요.
그래서 에어브릿지는 이번 **〈2026 벤치마크 리포트〉**에서 한국 데이터만 모아서 분석했어요. 2025년 하반기 월 1,000억 건 이상 이벤트와 2억 대 이상 디바이스, 1,000개 이상 브랜드 데이터를 종합해 분석한 후, 커머스와 금융, 게임 국내, 게임 해외, 라이프스타일&엔터, 기타, 글로벌 등 6개 업종별로 채널 분포를 정리했어요.
[2026 벤치마크 리포트 1.webp][2026 벤치마크 리포트 2.webp]
## **〈2026 벤치마크 리포트〉에서 확인할 수 있는 4가지**
### **1\. 우리 업종 인스톨, 딥링크, 사용 광고주 수 Top 12 채널은 어디일까?**
신규 유저를 데려오는 인스톨, 재방문을 일으키는 딥링크, 실제로 많이 도입된 채널 순위를 각각 확인할 수 있어요.
같은 업종이어도 채널의 역할이 신규 유입이냐 재방문이냐에 따라 결과가 완전히 달라지는데요. 인스톨 Top에 없던 채널이 딥링크 1위에 오르는 패턴이 업종별로 자주 나타나요.
### **2\. 새롭게 급성장하는 라이징 채널 Top 6에는 어떤 채널이 있을까?**
동종 업계에서 빠르게 도입되고 있는 라이징 채널 Top 6을 별도 섹션으로 보여줘요. 이 데이터를 바탕으로 다음 분기에 시범 테스트할 채널 후보로 활용할 수 있어요.
라이징 채널은 아래 3가지 성장 지표를 가중치로 종합해서 상위 6개를 선정했어요. 인스톨과 딥링크 성장은 같은 광고주 집합의 볼륨 변화를 비교(매칭 코호트)해 신규·이탈에 의한 착시 성장을 제거하고, 사용 광고주 수 성장은 전체 광고주 기준으로 산출했요.
이른 시점에 라이징 채널을 확인하면 경쟁이 본격화되기 전에 시범 테스트할 수 있어요.
[2026 벤치마크 리포트 3.webp][2026 벤치마크 리포트 4.webp]
### **3\. UA(유저 획득) 채널 운영 분포에서 우리 위치는 어디일까?**
광고주 규모별로 몇 개의 광고 채널을 동시 운영하는지 분포로 정리했어요. 동종업계 브랜드들의 운영 패턴 안에서 우리 브랜드가 어디쯤에 있는지 숫자로 확인할 수 있어요.
예를 들어, P10, P25, P50, P75, P90 등 5단계 분포도로 확인할 수 있는데요. P50은 같은 업종 회사들 중 중간에 위치한 채널 수, P75는 상위 25%에 해당하는 채널 수예요.
우리가 광고비를 너무 적게 쓰는 건지, 아니면 너무 분산한 건지 헷갈리셨다면, 리포트에서 우리 브랜드 분포 위치를 확인해 보세요. 감으로 정하던 채널 전략을 데이터로 정할 수 있어요.
### **4\. 12개월 시즈널리티 트렌드는 어떻게 움직일까?**
업종별 인스톨과 딥링크 활동의 월별 변동 패턴을 12개월 트렌드로 보여줘요.
각 업종 광고주들의 월별 활동 볼륨을 집계해서 성수기와 비수기 패턴을 산출했는데요. 분기 예산을 시즈널리티에 맞춰 배분하면 성수기 노출 기회를 놓치거나 비수기에 비싼 CPI로 예산을 쓰는 일을 줄일 수 있어요. 특히 커머스나 게임처럼 시즌 변동이 큰 업종이라면 이 트렌드만 따로 봐도 분기 예산 회의가 달라져요.
## **URL 1개로 자사 마케팅 전략 진단까지 받을 수 있어요**
리포트 본문 외에 자사 맞춤 진단도 함께 받아 볼 수 있어요. 진단 방식도 간단한데요. 운영하고 있는 앱 URL 1개만 넣으면 6개 업종 리포트와 자사 진단 결과를 모두 확인할 수 있어요. 별도의 SDK 연동이나 데이터 업로드 과정 없이 약 1분이면 AI가 리포트를 바로 생성해줘요.
진단 보고서는 4개 블록으로 구성돼요.
- **현황 진단:** 우리 회사가 운영 중인 채널과 업종 평균을 비교해 어디쯤에 있는지 보여줘요.
- **UA 전략:** 우리 업종의 Top 채널 중 놓치고 있는 채널을 짚어줘요.
- **리인게이지먼트 전략:** 재방문 활성화에 효과적인 채널을 분리해서 제안해요.
- **추천 액션:** 우리 회사가 어디부터 손대야 할지 단계별 순서를 정해줘요.
예를 들어 커머스 앱의 현황을 진단 받는다면, 현재 운영 채널 효율 점검 → 리인게이지먼트 채널 시범 도입 → 채널 다각화 늘리기 3단계로 액션 아이템을 제안받을 수 있어요. 업종마다 분포가 다르고 앱의 현황도 모두 제각각이어서 같은 진단 방식이라도 추천 액션은 다르게 나올 수 있어요.
## **하반기 마케팅 플랜, 직감이 아닌 데이터로 결정하세요**
"우리 업종 평균이 얼마지? 다른 회사는 어떤 마케팅 채널을 돌리고 있을까?"
하반기 마케팅 플랜을 짤 때 객관적인 데이터가 없어서 답답했을 텐데요. 이번 **<2026 벤치마크 리포트>**가 실마리를 작게 나마 실마리를 잡을 수 있게 도와드릴 거예요.
다음 분기 마케팅 플랜을 짜기 1~2주 전이나 채널 예산 검토 1주 전에 리포트와 진단 결과를 받고 데이터를 확인한 후 미팅에 들어가 보세요. 막연하게 정성적이었던 토론이 데이터 기반 토론으로 바뀌고, 보다 근본적인 문제를 해결하는 의사결정을 내릴 수 있을 거예요.
## **자주 묻는 질문**
### **Q. 맞춤형 진단 결과는 어떻게 만들어지나요?**
맞춤형 진단은 국내 브랜드의 앱 데이터를 기반으로 AI가 도출한 결과로, 단순 추정이 아니라 한국 시장에서 실제로 일어나고 있는 채널 분포와 운영 패턴을 반영하여 분석한 데이터예요.
### **Q. 다른 글로벌 벤치마크 자료와는 무엇이 다른가요?**
**<2026 벤치마크 리포트>**는 한국 마케터들이 벤치마크할 수 있도록 한국 앱 광고주의 데이터로만 구성됐어요. 네이버, 카카오, BUZZSCREEN, MOBON 같은 한국 고유 채널의 실제 비중을 객관적으로 확인할 수 있다는 점이 가장 큰 차이예요.
### **Q. 우리 앱 데이터가 에어브릿지 시스템에 들어가나요?**
앱 URL을 입력하는 방식이라 SDK 연동이나 자사 데이터 업로드 과정이 없고, 업종 벤치마크와 비교해서 진단해요. 진단을 신청해도 개인정보나 민감 데이터가 에어브릿지에 저장되지 않아요. 입력하는 앱 URL과 진단 결과의 자세한 저장·활용 정책은 약관에서 확인할 수 있어요.
### **Q. 에어브릿지를 안 쓰는 회사도 진단을 받을 수 있나요?**
받을 수 있어요. 앱 URL 입력만으로 동작하기 때문에 현재 에어브릿지를 도입하지 않은 회사도 6개 업종 리포트와 자사 진단을 함께 열어볼 수 있어요. 우리 회사의 다각화 위치가 업종 분포에서 어디쯤인지 외부에서 확인할 때 활용할 수 있어요.
### **Q. 데이터는 어느 기간을 기준으로 만들어졌나요?**
에어브릿지로 측정된 광고주의 실제 광고 성과 데이터를 기반으로 2025년 하반기(7월~12월) 데이터를 분석해 만들었어요. 해외 진출 리포트에는 한국 기업이 타겟팅한 해외 시장의 데이터가 포함돼요.
### **Q. Top 12 채널 중 우리가 안 쓰는 채널이 보이면 어떻게 검토하면 좋을까요?**
바로 도입을 결정하는 자료라기보다 테스트 후보를 좁히는 자료로 쓰는 게 잘 맞아요. Top 채널이라도 우리 앱 유저 특성, 현재 예산 규모, 기존 채널 포화도에 따라 적합도가 달라져요. 라이징 Top 6 섹션에서 동종 업계가 막 도입 중인 채널을 1~2개 추려 분기 예산을 잡아 보세요.
### **Q. 우리 앱의 업종이 6개 안에 정확히 포함되지 않으면 어떻게 하나요? 예를 들어 헬스케어나 모빌리티는요?**
6개 업종 분류 중 가장 가까운 카테고리를 기준으로 보거나 '기타' 업종 분포를 참고로 활용해 보세요.
예를 들어, 헬스케어 앱은 라이프스타일&엔터의 리인게이지먼트 패턴과 겹치는 경우가 있어요. 모빌리티는 커머스의 트랜잭션 흐름과 비슷한 채널 구성을 보이기도 해요. 다만 1:1 매칭은 아니니 분포 위치는 절대값보다 상대 비교용으로 읽어 보세요.
### **Q. 진단 결과를 사내 툴로 공유할 수 있나요?**
현재 리포트는 웹 화면에서 열람할 수 있고, 결과 화면을 공유 링크로 동료에게 전달할 수 있어요. 만약 CSV나 PDF 내보내기, 사내 BI 연동이 필요하다면 에어브릿지 담당자에게 별도로 문의해 주세요.
---
## 원하는 앱 화면으로 바로 떨어지게 만드는 '디퍼드 딥링크' 사용법
URL: https://www.airbridge.io/ko/blog/how-to-use-deferred-deeplink-for-app-marketing
Language: Korean
Published: 2026-06-08
Author: Team Airbridge
대시보드를 열어 지난주 앱 설치 캠페인 리포트를 확인해 봅니다. 클릭 10만 건, 설치 1만 2천 건이 발생했어요. CPI도 목표 범위 안에 들어왔죠.
그런데 정작 이벤트 참여율은 3%에 그쳤고, 신청 완료율은 이전 캠페인보다 낮았어요. 매체 효율이나 소재 문제를 먼저 떠올릴 수 있지만, **실제 병목은 앱 설치 이후 첫 화면**에서 발생할 수 있어요.
유저는 광고에서 본 혜택이나 상품, 이벤트를 기대하고 앱을 설치해요. 그런데 앱을 열자마자 홈 화면에 도착한다면, 유저는 다시 메뉴를 찾거나 이벤트 배너를 눌러야 하죠. 여러 과정을 거쳐 신청 페이지로 이동하다 보면 관심은 사그라들고, 전환까지 멀어질 수밖에 없어요.
이렇게 앱 설치 광고, 웹투앱 캠페인, QR 캠페인처럼 앱 밖에서 시작되는 여정일수록 설치 이후 첫 화면은 전환율과 성과 해석에 직접적인 영향을 줄 수 있는데요. 이때 필요한 것이 **디퍼드 딥링크**예요. 디퍼드 딥링크를 활용하면 앱을 설치하지 않은 유저도 캠페인 맥락 안에서 움직이게 할 수 있어요.
오늘은 캠페인 목적에 맞게 사용할 수 있는 디퍼드 딥링크와 표준 딥링크에 대해서 알아보고, 설치 이후 유저 이동과 캠페인 성과를 확인할 때 디퍼드 링크를 어떻게 사용해야 하는지도 함께 살펴볼게요.
**📌핵심 포인트**
- 앱 설치 수와 첫 구매·신청 전환율 사이의 갭이 크다면, 소재나 타겟팅보다 설치 직후 첫 화면을 먼저 의심해 보세요. 광고에서 약속한 혜택과 앱 첫 화면이 다르면, 유저는 이미 그 시점에 이탈할 수도 있어요.
- 디퍼드 딥링크가 없으면 앱 설치 광고의 성과의 온전한 측정이 어려워요. 클릭 → 설치까지는 트래킹되지만, 설치 이후 어떤 행동으로 이어졌는지 캠페인 맥락과 연결되지 않을 수 있어요.
- 캠페인 목적지가 자주 바뀐다면 마케터가 직접 링크를 만들 수 있는 운영 구조가 실행 속도를 좌우해요. 다방은 에어브릿지 도입 후 딥링크 제작에 드는 개발 리소스를 약 90% 줄였어요.
## **디퍼드 딥링크 vs 표준 딥링크, 언제 무엇을 써야 할까요?**
두 딥링크를 비교하기 전, 딥링크에 대한 개념을 먼저 알아야 하는데요.
딥링크는 **광고, 푸시, CRM 메시지 등을 클릭한 유저를 앱 안의 특정 화면으로 연결하는 방식**을 뜻해요.
예를 들어 프로모션 광고는 이벤트 페이지로, 장바구니 이탈 메시지는 담아둔 상품 화면으로, 쿠폰 안내 메시지는 쿠폰 페이지로 바로 연결하는 거죠.
디퍼드 딥링크(Deferred Deep Link)는 앱이 설치되어 있지 않은 유저가 링크를 클릭했을 때, 앱스토어로 이동해 앱을 설치한 뒤 첫 실행 시 원래 연결하려던 앱 내 특정 화면으로 이동시키는 기술이에요. '설치 전 클릭 맥락'을 '설치 후 첫 실행'까지 유지한다는 점에서 표준 딥링크와 구분돼요.
딥링크의 기본 개념과 주요 유형은 딥링크 관련 콘텐츠에서 더 자세히 살펴볼 수 있어요.
👉🏻[딥링크란? URI Scheme, Universal Link, App Link 완벽 가이드](https://www.airbridge.io/ko/blog/what-is-deep-link)
## **캠페인 목적에 따라 구분해 사용해 보세요**
이로 인해 딥링크를 선택할 때 가장 먼저 봐야 할 것은 링크 형식이 아니라 **캠페인이 유저에게 기대하는 행동**이에요.
유저가 앱을 이미 설치했는지도 중요하지만, 가장 중요한 기준은 이 캠페인이 유저를 어디까지 데려가야 성공인지 목적을 명확히 하는 것이죠.
예를 들어 **기존 앱 유저에게 시즌 이벤트 참여를 유도하는 것이 목적인 캠페인**이라면, 유저를 이벤트 화면으로 바로 보내는 것만으로도 충분할 수 있어요. 이 경우에는 **표준 딥링크**가 적합해요.
반대로 **앱을 설치하지 않은 유저에게 신규 가입, 상품 신청, 첫 구매, 게임 보상 수령 같은 행동을 유도**해야 한다면 이야기가 달라져요. 유저가 앱스토어를 거쳐 앱을 설치하는 동안 캠페인 맥락이 끊기지 않아야 하기 때문이죠. 이때는 **디퍼드 딥링크**가 필요해요.
표준 딥링크와 디퍼드 링크를 선택할 때 아래 기준으로 판단해 사용해 보세요.
### **1\. 표준 딥링크가 적합한 캠페인**
| 캠페인 상황 | 활용 포인트 |
| --- | --- |
| 기존 앱 유저에게 이벤트 참여 유도 | 앱 설치가 이미 되어 있어 이벤트 화면으로 바로 연결 |
| 장바구니, 찜한 상품, 특정 콘텐츠로 재방문 유도 | 유저의 재탐색 과정 최소화 |
| 푸시·리타겟팅 광고로 구매·신청 유도 | 상품, 혜택, 신청 화면까지 이동 경로 단축 |
앱 안에서 행동하게 만드는 것이 중요하기 때문에, 다음과 같은 상황에서 활용할 수 있어요.
**1) 게임 앱에서 기존 유저에게 신규 캐릭터 출시 소식을 알릴 때**
유저가 푸시 알림이나 리타겟팅 광고를 클릭했을 때 캐릭터 소개 화면이나 이벤트 참여 화면, 상점 화면으로 바로 이동하면 캠페인의 흐름을 단축할 수 있어요. 유저는 홈 화면에서 배너를 찾거나 메뉴를 탐색하지 않아도 되어 편리해요.
**2) 커머스 앱에서 장바구니에 상품을 담아둔 유저에게 할인 알림을 보낼 때**
앱 설치 유저라면 홈 화면보다 장바구니나 해당 상품 상세 화면으로 연결하는 편이 전환을 만들 가능성이 커져요. 유저가 관심을 보였던 상품으로 다시 돌아가기 쉬워지고, 구매까지 이어질 가능성도 높아져요.
**3) 금융 앱에서 이미 앱을 설치한 고객에게 특정 카드 혜택이나 예금 상품을 안내할 때**
해당 상품의 상세 화면으로 바로 연결하면 고객이 상품 목록을 다시 탐색하지 않아도 돼요. 캠페인에서 안내한 혜택과 신청 화면이 자연스럽게 이어지기 때문에, 조회나 신청 같은 다음 행동으로 넘어가기 쉬워져요.
### **2\. 디퍼드 딥링크가 적합한 캠페인**
| 캠페인 상황 | 활용 포인트 |
| --- | --- |
| 앱 설치 광고를 통한 신규 가입 ·첫 구매 유도 | 설치 후 가입 혜택, 쿠폰, 첫 구매 화면으로 연결 |
| 웹 배너 기반 앱 전용 상품 신청 유도 | 웹에서 본 상품·신청 페이지를 앱 설치 이후까지 유지 |
| QR 코드 기반 오프라인 유저 전환 | 앱 설치 여부에 따른 이동 경로 제공 |
| 게임 사전예약·복귀 보상 캠페인 | 설치 직후 보상 수령·이벤트 참여 화면으로 연결 |
캠페인 성공 조건에 앱 설치 이후 행동이 포함된다면 특히 아래와 같은 캠페인에서 유용해요.
- **앱 설치 광고에서 신규 가입 쿠폰을 강조할 때:** 유저는 쿠폰을 기대하고 광고를 클릭하지만, 앱이 설치되어 있지 않다면 앱스토어를 거쳐야 해요. 디퍼드 딥링크를 적용하면 첫 실행 시 쿠폰 발급 화면이나 가입 혜택 페이지로 바로 연결할 수 있어요.
- **금융 상품 신청 캠페인에서 비대면 계좌 개설 혜택을 안내할 때:** 유저가 혜택을 보고 앱을 설치했는데, 홈 화면에서 상품 메뉴와 계좌 개설 화면을 다시 찾아야 한다면 신청까지의 여정이 길어져요. 앱 설치 후 계좌 개설 화면이나 이벤트 안내 화면으로 바로 이동하게 만들어 보세요.
- **게임 캠페인에서 사전예약 보상이나 복귀 유저 혜택을 제공할 때:** 유저는 보상을 기대하고 앱을 설치하기 때문에 첫 화면에서 보상 수령 위치를 찾지 못하면 이벤트 참여 의지가 낮아질 수 있죠. 보상 수령, 이벤트 참여, 캐릭터 생성 화면을 바로 연결하면 설치 직후 행동을 자연스럽게 유도할 수 있어요.
- **유튜브·인스타그램 인플루언서 캠페인에서 앱 신규 다운로드를 유도할 때:** 영상·게시물에서 본 특정 상품, 챌린지 페이지를 클릭한 유저가 앱을 설치한 직후 동일한 화면으로 도착해야 해요. 디퍼드 딥링크가 없으면 인플루언서 콘텐츠와 앱 첫 화면 사이의 연결이 끊기면서 측정도, 전환도 모두 약해져요.
## **에어브릿지 디퍼드 딥링크로 무엇을 할 수 있나요?**
에어브릿지에서는 캠페인 성과 측정에 필요한 터치포인트와 캠페인 데이터를 수집하는 [트래킹 링크](https://help.airbridge.io/ko/guides/understanding-tracking-link)\*에 딥링크를 적용할 수 있어요. 트래킹 링크는 대시보드에서 설정한 최종 목적지로 광고를 클릭한 유저를 보내요.
앱이 설치된 유저는 앱 안의 특정 화면으로 보내고, 앱이 설치되지 않은 유저는 앱마켓을 거쳐 설치 후 첫 실행 시 설정한 화면으로 이동하도록 설계할 수 있죠.
_\*트래킹 랭크: 에어브릿지에서 성과를 측정하기 위해 필요한 터치포인트, 캠페인과 관련된 데이터 등을 수집. 에어브릿지 대시보드에서 설정한 최종 목적지로 광고를 클릭한 유저를 보낼 수 있음._
### **1\. 앱 설치 전 클릭부터 첫 실행까지 하나의 흐름을 볼 수 있어요**
[딥링크 1.webp]
에어브릿지에서는 트래킹 링크를 기준으로 캠페인 유입을 구분해서 딥링크를 적용할 수 있는데요.
이로 인해 앱 설치 광고, 웹 배너, QR 코드, SMS처럼 서로 다른 접점에서 시작된 유저 여정을 링크 단위로 관리할 수 있어요.
예를 들어 동일한 신규 가입 캠페인이라도 인스타그램 광고, 웹 배너, SMS에 각각 다른 트래킹 링크를 사용하면 유입 경로를 나눠볼 수 있어요. 동시에 각 링크에 맞는 딥링크 목적지를 설정하면, 설치 후 첫 실행 화면까지 캠페인 맥락에 맞게 이어갈 수 있죠.
**각 링크가 어떤 캠페인 메시지와 어떤 앱 화면을 연결하는지** 정리한 후, API로 Google Sheets에서 링크를 대량 생성하고, 모든 링크를 한곳에서 관리해 보세요.
### **2\. iOS와 Android 환경에서도 목적지 연결을 안정적으로 설계할 수 있어요**
디퍼드 딥링크는 OS별 딥링크 설정도 함께 고려해야 해요.
같은 캠페인 링크를 클릭하더라도 유저가 iOS인지 Android인지에 따라 앱마켓, 앱 실행 방식, 앱 내 목적지 연결 방식이 달라질 수 있기 때문이에요.
에어브릿지에서는 각 OS에서 딥링크가 정상적으로 작동할 수 있도록 필요한 앱 정보를 등록할 수 있는데요.
예를 들어 iOS에서는 앱을 식별하고 유니버설 링크를 연결하기 위한 정보가 필요하죠. 다만, 마케터가 모든 기술 항목을 직접 이해할 필요는 없어요.
또한 OS 환경에서는 ATT(App Tracking Transparency) 프롬프트 결정 전까지 디퍼드 딥링크가 작동하지 않을 수 있으니, iOS 대상 캠페인을 기획 중이라면 개발팀과 함께 ATT 처리 순서를 미리 확인해 두는 것을 권장해요.
무엇보다 중요한 건, 캠페인 시작 전에 개발팀과 함께 **iOS와 Android 환경에서 링크 클릭 → 앱마켓 이동 또는 앱 실행 → 설치 후 목적지 도착** 흐름이 정상적으로 작동하는지 확인하는 거예요.
운영 전에는 OS별로 아래 흐름을 확인해 보세요.
- iOS 유저가 App Store로 정상 이동하는지
- Android 유저가 Google Play로 정상 이동하는지
- 앱 설치 유저가 스토어를 거치지 않고 앱 내 목적지로 이동하는지
- 앱 미설치 유저가 설치 후 첫 실행 시 설정한 화면으로 이동하는지
### **3\. 4단계로 디퍼드 딥링크를 간단하게 설정할 수 있어요**
에어브릿지에서는 트래킹 링크, 딥링크 정보, SDK 설정을 준비해 디퍼드 딥링크를 적용할 수 있어요. 설정 흐름은 간단하며, 캠페인 목적지와 앱 내 이동 경로가 정확히 연결되는지 확인하면 돼요.
**1) 개발 팀은 딥링크 정보 등록 및 앱 등록·SDK 설정 확인하기**
먼저 에어브릿지 대시보드의 \[트래킹 링크\] > \[딥링크\]에서 딥링크 정보를 등록하세요. 이 정보가 SDK 설정의 사전 준비 단계이기 때문에, 딥링크 정보 등록을 먼저 완료한 뒤 SDK 설치·초기화를 진행해야 해요. 이 단계에서는 앱이 트래킹 링크에서 전달된 딥링크를 받을 수 있도록 [딥링크와 디퍼드 딥링크 동작을 함께 준비](https://help.airbridge.io/ko/developers/sdk-quickstart)하세요.
출시 앱뿐 아니라 출시 예정 앱이나 개발용 앱에서도 이 방식으로 앱(딥링크)을 설정할 수 있어요.
**2) 마케팅 팀은 트래킹 링크 생성 시 최종 목적지 설정하기**
트래킹 링크를 생성할 때 \[트래킹 링크\] > \[트래킹 링크 생성\] > \[리다이렉트 경로\]에서 [최종 목적지로 앱(딥링크)을 선택하고, 유저를 보낼 앱 화면의 딥링크를 입력](https://help.airbridge.io/ko/guides/retargeting-with-deep-links)해요.
예를 들어 가입 혜택 페이지, 상품 상세 페이지, 쿠폰 발급 화면, 게임 보상 수령 화면처럼 캠페인에서 유저가 도착해야 할 화면을 목적지로 지정할 수 있어요.
앱 미설치 유저를 어디로 보낼지도 함께 설정해야 해요.
에어브릿지의 [Fallback 경로](https://help.airbridge.io/ko/guides/creating-tracking-links-on-the-dashboard)에서는 앱을 설치하지 않은 유저가 딥링크가 적용된 트래킹 링크를 클릭했을 때 이동할 최종 목적지를 설정할 수 있어요. OS별로 앱마켓이나 웹 URL을 선택할 수 있어요.
**3) 마케팅과 QA 팀은 앱 미설치 상태에서 테스트하기**
디퍼드 딥링크는 앱 설치 후 첫 실행까지 이어져야 하므로, 반드시 앱이 설치되지 않은 디바이스에서 테스트해야 해요.
실제 캠페인과 같은 조건에서 트래킹 링크 클릭 → 앱마켓 이동 → 앱 설치 → 첫 실행 → 목적지 도착까지 확인하면 돼요.
이때 한 가지 중요한 제약이 있어요\*\*. 에어브릿지 디퍼드 딥링크는 광고 클릭 후 1시간 이내에 앱 설치 및 첫 실행이 이루어져야 작동\*\*해요. 1시간을 초과하면 디퍼드 딥링크가 작동하지 않으니, 테스트 및 실제 캠페인 운영 시 이 점을 꼭 고려해 두세요.
테스트는 iOS와 Android 디바이스에서 모두 진행해야 해요. 같은 링크라도 OS에 따라 앱마켓 이동, 앱 실행, 앱 내 목적지 연결 방식이 달라질 수 있기 때문이에요.
### **‘다방’도 에어브릿지로 딥링크 제작 리소스를 90% 줄였어요**
[딥링크 2.webp]
디퍼드 딥링크를 안정적으로 운영하려면 설정 흐름뿐 아니라, 실제 캠페인마다 링크를 만들고 관리하는 운영 리소스도 함께 고려해야 해요.
부동산 대표 앱 다방은 메타를 비롯한 다양한 광고 채널을 통해 광고 캠페인을 진행하고 있었는데요. **특정 지역과 매물 정보에 대한 디퍼드 딥링킹을 통해 앱을 설치하지 않은 유저 또한 설치 후 광고에서 확인한 매물로 랜딩**하는 것이 목적이었죠.
다방은 기존에 딥링크 생성 시 개발자 리소스가 필요해 시간이 오래 걸렸고, 마케터가 원하는 시점에 빠르게 캠페인을 실행하기 어려웠어요.
**에어브릿지로 전환한 뒤에는 마케터가 직접 딥링크를 생성**할 수 있게 되면서 **딥링크 제작에 들어가는 개발 리소스를 약 90% 줄일 수 있었죠.**
단순히 리소스가 줄어든 것 이상의 변화도 있었어요. 마케터가 매물 정보별로 딥링크를 직접 생성할 수 있게 되면서, 메타 광고 채널에서 다양한 매물 단위 캠페인을 더 빠르게 실행할 수 있게 됐어요.
기존에는 개발팀 요청 후 수일이 걸리던 링크 생성이 즉시 실행 가능한 구조로 바뀐 거예요. 앱 미설치 유저도 광고에서 본 매물 화면으로 직접 도착하면서 첫 화면 도착률도 함께 개선됐어요.
만약 앱 설치 광고, 웹투앱, QR 캠페인처럼 목적지가 자주 달라지는 캠페인을 운영하고 있다면, 매번 개발팀에 링크 제작을 요청하면서 캠페인 성과 추적이나 딥링크 운영 속도가 느려질 수 있어요.
에어브릿지를 활용하면 캠페인 목적지에 맞는 딥링크를 더 빠르게 만들고, 마케터와 개발팀이 각각 맡아야 할 작업을 분리해 운영 부담을 줄일 수 있어요.
## **디퍼드 딥링크 도입 전, 우리 캠페인 흐름부터 점검해 보세요**
지금 운영 중인 캠페인에서 성과가 측정이 되지 않는 구간이 있다면, 링크 설정보다 먼저 점검이 필요해요. 다음 5가지 질문에 답할 수 있는지 확인해 보세요.
1. **앱 설치 광고·웹투앱·QR 캠페인 클릭 후 첫 화면 도착률을 OS별로 측정하고 있나요?**
- iOS와 Android는 앱마켓 이동과 첫 실행 경로가 다르기 때문에, 합산 지표만으로는 이탈 구간이 보이지 않을 수 있어요.
2. **동일 캠페인의 미설치 유저 경로(설치 후 첫 화면)와 설치 유저 경로(앱 내 화면)를 같은 링크 단위로 관리하고 있나요?**
- 두 경로가 별개 도구에서 관리되면 캠페인 단위 성과 비교가 어려워져요.
3. **캠페인 목적지가 바뀔 때 마케터가 직접 링크를 만들 수 있나요?**
- 매번 개발팀에 요청을 넣거나 일정에 맞춰야 한다면 실행 속도가 캠페인 성과의 병목이 돼요.
4. **클릭 → 앱마켓 → 설치 → 첫 실행 → 목적지의 5단계 전환율을 분리해서 보고 있나요?**
- 어디서 가장 많이 빠지는지 모르면 어디를 고쳐야 하는지도 알 수 없어요.
5. **iOS와 Android에서 같은 캠페인 링크가 동일하게 작동하는지 정기 점검 프로세스가 있나요?**
- OS 업데이트, 앱 버전 업데이트, 도메인 인증 만료로 디퍼드 딥링크가 언제 끊겼는지 모른 채 방치될 수 있어요.
위 5개 중 답이 막히는 항목이 2개 이상이면, 디퍼드 딥링크를 새로 적용하기 전에 운영 구조부터 정리해 보세요. 앞서 살펴본 트래킹 링크 단위 관리, OS별 설정, 4단계 셋업 흐름이 점검 기준이 될 수 있을 거예요.
## **디퍼드 딥링크로 전환까지 이어지는 캠페인을 만들어 보세요**
설치 광고를 운영 중인데 전환이 기대만큼 나오지 않는다면, 소재나 타겟을 바꾸기 전에 설치 후 첫 화면부터 확인해 보세요.
유저가 광고에서 기대하고 들어온 화면이 앱을 열었을 때 나타나지 않는다면, 그 지점에서 이미 이탈이 일어날 수 있으니까요.
에어브릿지 딥링크를 활용하면 캠페인별 목적지에 맞는 링크를 더 빠르게 만들고, 앱 설치 여부에 따른 이동 흐름을 하나의 구조로 관리할 수 있죠.
실제 캠페인에서 어떻게 작동하는지 확인하고 싶다면, 데모에서 직접 살펴보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 캠페인 목적에 따라 달라지는 모바일 딥링크 활용법 4가지
URL: https://www.airbridge.io/ko/blog/mobile-deeplink-types-by-campaign
Language: Korean
Published: 2026-05-26
Author: Team Airbridge
**앱으로 바로 연결되는 링크, ‘딥링크’**
광고를 눌렀는데 앱 안의 특정 화면으로 바로 이동한 적이 있으신가요? 이처럼 **사용자를 앱의 원하는 화면으로 연결하는 링크**를 [딥링크](https://help.airbridge.io/ko/guides/understanding-deep-link)라고 해요.
마케터라면 딥링크가 필요하다는 건 알지만, URI 스킴, 유니버설 링크, 앱 링크, 디퍼드 딥링크 중 무엇을 언제 써야 하는지는 헷갈리기 쉬워요. 캠페인 목적과 사용자 맥락에 따라 선택이 달라지고, 금융권이나 대기업처럼 보안 검토가 들어가는 조직이라면 도입 전에 함께 확인할 항목도 있어요.
이 글에서는 4가지 딥링크 방식의 차이와 캠페인별 적합도, 그리고 엔터프라이즈 도입 시 점검해야 할 기준까지 정리했어요. 자체 캠페인에 딥링크가 어떤 방식으로 들어가야 하는지 판단 기준을 잡는 데 도움이 될 거예요.
**📌 핵심 포인트**
- 딥링크는 주로 앱 설치 사용자를 특정 화면으로 바로 보내는 데 쓰이고, 디퍼드 딥링크는 앱이 없는 사용자도 설치 후 원래 의도한 화면으로 이어지게 하는 방식으로 크게 구분하고 있어요.
- URI 스킴, 유니버설 링크, 앱 링크, 디퍼드 딥링크는 쓰임이 달라서 캠페인 목적과 사용자 맥락에 맞게 골라 사용해야 해요. 특히 리타겟팅, CRM, 레퍼럴처럼 클릭 이후 맥락이 중요한 캠페인에서는 딥링크 설정과 운영 방식이 성과에 직접 영향을 줄 수 있어요.
## **왜 모바일 마케팅에서 딥링크가 중요할까요?**
모바일 마케팅에서는 클릭 수보다 **클릭 이후 사용자가 어디에 도착하는지**가 더 중요해요. 사용자를 앱 메인 화면으로 보내면 다시 원하는 페이지를 찾아가야 하지만, 딥링크를 활용하면 캠페인 의도에 맞는 화면으로 바로 연결할 수 있죠. 그만큼 경로는 짧아지고, 이탈 가능성은 낮아지게 하는 것이 딥링크의 역할이에요.
### **1\. 딥링크로 기본 전환 경로 설계하기**
의도적인 경로 설정이 가능하기 때문에, **딥링크는 전환이 일어날 가능성이 높은 지점으로 연결**할 수 있어요. 특히 아래와 같이 도착해야 할 페이지가 명확한 캠페인에서 딥링크 효과가 크죠.
- 프로모션 광고 클릭 → 이벤트 페이지
- 장바구니 이탈 안내 → 담아둔 상품 화면
- CRM 메시지 클릭 → 쿠폰 페이지
이때 핵심은 연결 기능 자체가 아니라 **클릭 이후 흐름을 끊기지 않게 설계하는 것**인데요. 같은 캠페인이라도 딥링크 적용은 **사용자의 앱 설치 여부**에 따라 실제 이동 경로는 달라질 수 있어요.
이렇게 광고, **콘텐츠 등을 클릭한 유저를 상세 페이지로 보내는 동작을 ‘딥링킹(Deel linking)’**이라고 하며, 딥링킹 유저 시나리오는 보통 앱 설치 사용자와 앱 미설치 사용자, 두 가지 관점에서 볼 수 있어요.
[02_deeplinking_scenario_illustrated.svg]
### **2\. 딥링킹 시나리오 한눈에 보기**
| 구분 | 앱 미설치 사용자 | 앱 설치 사용자 | 이동 단계 | 기대 효과 |
| --- | --- | --- | --- | --- |
| 딥링크 적용 | 클릭 → 앱마켓/웹사이트 이동 → 앱 설치 후 첫 실행 → 앱 기본 화면 이동 | 클릭 → 설정된 경로 이동 → 앱 실행 → 앱 기본 화면 이동 | 단계가 상대적으로 길어짐 | 앱 진입 후 다시 탐색 필요 |
| 딥링크 미적용 | 클릭 → 앱마켓 이동 → 앱 설치 후 첫 실행 → 특정 앱 페이지 연결 | 클릭 → 특정 앱 페이지 바로 연결 | 탐색 단계 감소 | 클릭 맥락 유지, 목적 페이지 빠른 도달 |
표를 보면 앱 설치 사용자와 앱 미설치 사용자 사이의 차이를 먼저 발견할 수 있어요.
**앱 설치 사용자**는 딥링크가 적용되면 클릭 후 특정 앱 페이지로 바로 이동할 수 있어, 기본 화면에서 다시 원하는 페이지를 찾는 과정을 줄일 수 있어요.
반면 **앱 미설치 사용자**는 앱 설치 자체는 거쳐야 하지만, 설치 후 첫 실행 시 특정 페이지로 연결되도록 설정할 수 있어 설치 이후 탐색 단계를 줄이는 데 도움이 되죠.
중요한 건 사용자가 현재 앱을 설치한 상태인지, 딥링크가 적용되어 있는지에 따라 클릭 이후의 이동 경로와 사용 경험이 달라질 수 있다는 것인데요.
딥링크는 사용 환경과 운영체제, 연결 방식에 따라 구현 구조가 달라질 수 있기 때문에 실무에서는 어떤 유형의 딥링크를 써야 하는지 함께 이해할 필요가 있어요. 아래는 실무에서 주로 사용하는 딥링크 4가지예요.
## 대표적인 딥링크 4가지, 캠페인별 이렇게 활용하세요
딥링크를 실무에서 사용할 때는 앞서 설명한 시나리오처럼 먼저 사용자가 어떤 방식으로 앱을 열 것인지와 앱이 설치되지 않은 사용자가 어떻게 연결될 수 있는지 나눠서 봐야 해요.
마케터 입장에서는 특히 스킴 딥링크, 유니버셜 링크, 앱 링크, 디퍼드 딥링크를 비슷한 개념처럼 받아들이기 쉬운데요. 먼저 정리하면, **URI 스킴·유니버설 링크·앱 링크는 ‘딥링크를 여는 방식’**, **디퍼드 딥링크는 ‘설치 전후를 이어주는 도입 시나리오’**라고 이해하면 쉬워요.
[01_deeplink_types_illustrated.svg]
### **1\. URL 스킴 딥링크: 빠르게 적용하기 쉬운 연결 방식**
URI 스킴은 앱이 자체적으로 가진 스킴값을 호출해 앱을 여는 방식이에요.
웹에서 URL로 페이지를 찾듯, 앱도 `myapp://` 같은 전용 주소를 통해 호출할 수 있어요. 여기에 추가 경로를 붙이면 앱 실행뿐 아니라 특정 화면으로의 이동까지 함께 지정할 수 있어요.
**장점**
- 링크 구조가 `scheme://path` 형태로 단순해서 이해하기 쉬워요.
- 빠르게 적용을 시작하기에 부담이 적어요.
**주의할 점**
- 동일한 스킴값을 다른 앱도 사용할 수 있어요.
- 실행 환경에 따라 오픈이 일관되지 않을 수 있어요.
- 앱이 설치되지 않은 경우 원하는 흐름으로 자연스럽게 이어지지 않을 수 있어요.
**이런 캠페인에 적합해요**
- **앱이 이미 설치된 기존 사용자 대상 CRM 캠페인**
- 푸시, 문자, 이메일, 카카오 알림톡처럼 **이미 앱을 쓰고 있는 사용자**에게 특정 혜택 화면이나 쿠폰 페이지로 바로 보내야 할 때 적합해요. 사용자의 현재 맥락과 목적지가 분명할수록 인게이지먼트(engagement)와 컨버전(Conversion)이 높아지는 경향이 있어, 이런 캠페인에서는 빠른 인앱 연결의 효과를 기대하기 좋아요.
- **리타겟팅처럼 ‘다시 들어오게 만드는’ 캠페인**
- 장바구니 이탈, 상품 조회 이탈, 휴면 복귀 캠페인처럼 **이전 행동 이력이 있는 사용자**를 다시 불러올 때 유용해요. 이미 관심을 보였던 사용자일수록 인게이지먼트와 수익 이벤트가 더 강하게 나타날 수 있어, 목적 페이지를 바로 여는 구조가 특히 잘 맞습니다.
- **앱 내 특정 행동을 바로 유도해야 하는 프로모션 캠페인**
- 한정 할인, 특정 상품전, 시즌 이벤트처럼 **도착 페이지가 명확한 프로모션**에 적합해요. 사용자가 앱 메인 화면에서 다시 탐색하지 않고 바로 이벤트 상세나 상품 화면으로 들어갈 수 있어, 클릭 후 흐름을 더 짧게 만들 수 있어요.
- **운영 환경이 비교적 통제되는 채널**
- 사내 앱, 멤버십 앱, 파트너십 앱처럼 **사용자군과 유입 경로가 비교적 명확한 환경**에서 쓰기 좋아요. URI 스킴은 구조가 단순해 빠르게 적용하기 좋지만, 채널과 단말 환경이 복잡해질수록 예외 케이스가 늘 수 있어 통제된 운영 환경에서 더 안정적으로 쓰는 것을 추천해요.
### **2\. 유니버셜 딥링크: iOS 사용자를 앱으로 연결하는 방식**
유니버설 링크는 iOS에서 제공하는 딥링크 방식이에요.
`https://`로 시작하는 일반 웹 링크를 사용하지만, 앱이 설치된 경우에는 웹 대신 앱이 열리고, 설치되지 않은 경우에는 해당 웹페이지가 열리는 구조에요. 마케터 입장에서는 **iOS 사용자에게 더 자연스러운 앱 연결 경험을 설계할 때** 자주 보게 되는 방식이라고 이해하면 돼요.
**장점**
- 일반 웹 링크처럼 보여 사용자 입장에서 더 자연스러워요.
- 앱 설치 여부에 따라 연결 흐름을 유연하게 설계하기 좋아요.
- iOS 사용자를 대상으로 한 웹투앱 시나리오에 활용하기 좋아요.
**주의할 점**
- iOS 중심 방식이라 안드로이드와는 별도로 봐야 해요.
- 설정 구조가 URI 스킴보다 복잡할 수 있어요.
- 링크 연결만 볼 것이 아니라 실제 캠페인 흐름까지 함께 점검해야 해요.
**이런 캠페인에 적합해요**
- **iOS 사용자 대상 CRM 캠페인**
- 문자, 이메일, 푸시, 알림톡처럼 메시지를 통해 사용자를 다시 앱으로 불러와야 할 때 적합해요. 특히 이미 앱을 쓰고 있는 사용자에게 특정 혜택 화면, 쿠폰 페이지, 멤버십 화면처럼 목적지가 분명한 페이지로 자연스럽게 연결해야 할 때 활용도가 높아요.
- **웹투앱 전환이 중요한 프로모션 캠페인**
- 이벤트 페이지, 브랜드 소개 페이지, 상품 기획전처럼 웹에서 먼저 정보를 보여준 뒤 앱으로 전환시키고 싶은 캠페인에 잘 맞아요. 사용자가 iOS 환경에서 링크를 눌렀을 때 앱이 설치되어 있다면 앱으로, 설치되어 있지 않다면 웹으로 이어질 수 있어 흐름을 상대적으로 부드럽게 설계하기 좋아요.
- **앱 설치 여부가 섞여 있는 iOS 사용자 대상 캠페인**
- 같은 캠페인 안에서도 누군가는 앱 설치 사용자이고, 누군가는 아직 비설치 사용자일 수 있어요. 이럴 때 유니버설 링크는 하나의 링크를 기준으로 사용자 상태에 따라 다른 연결 흐름을 설계할 수 있어 운영 효율 측면에서도 유리해요.
- **광고, 콘텐츠, 메시지를 하나의 링크 기준으로 운영하고 싶은 경우**
- 여러 채널에서 같은 프로모션이나 같은 랜딩 목적지를 안내해야 할 때 적합해요. 웹 링크 기반으로 운영할 수 있기 때문에, 마케터 입장에서는 채널별 소재를 운영하면서도 연결 구조를 비교적 일관되게 가져갈 수 있어요.
### **3\. 앱링크: 안드로이드 사용자 연결에 적합한 방식**
앱 링크는 안드로이드 환경에서 활용되는 딥링크 방식이에요. 웹 링크를 기반으로 사용자를 앱 안의 특정 화면으로 연결할 수 있어, 안드로이드 사용자 대상 캠페인에서 자주 활용돼요. 실무에서는 **안드로이드 사용자에게 더 안정적인 앱 연결 경험을 제공하기 위한 방식**으로 이해하면 쉬워요.
**장점**
- 웹 링크 기반이라 운영 구조가 상대적으로 덜 복잡해요.
- 안드로이드 사용자 대상 연결 흐름을 설계하기 좋아요.
- 광고, CRM, 리타겟팅 등 다양한 캠페인에 활용할 수 있어요.
**주의할 점**
- 안드로이드 중심 방식이라 iOS와는 별도로 운영해야 해요.
- 설정과 검증 구조를 함께 확인해야 해요.
- 실제 캠페인 적용 전 테스트가 중요해요.
**이런 캠페인에 적합해요**
- **안드로이드 사용자 대상 CRM 캠페인**
- 푸시, 문자, 이메일, 메시지 캠페인처럼 기존 사용자를 다시 앱으로 유입시켜야 할 때 적합해요. 특히 쿠폰함, 특정 상품 페이지, 프로모션 화면처럼 바로 다음 행동이 정해져 있는 경우 앱 링크를 통해 더 짧은 경로를 설계할 수 있어요.
- **안드로이드 중심 리타겟팅 캠페인**
- 장바구니 이탈, 조회 이탈, 휴면 복귀처럼 다시 들어오게 만드는 캠페인에 쓰기 좋아요. 이미 관심을 보였던 사용자를 앱 안의 관련 페이지로 바로 연결할 수 있기 때문에, 클릭 이후 흐름을 짧게 만들고 재참여 가능성을 높이는 데 유리해요.
- **웹 URL을 함께 쓰는 프로모션 캠페인**
- 광고 랜딩, 배너, 모바일 웹 페이지와 함께 앱 전환을 설계해야 할 활용하기 좋아요. 특히 안드로이드 사용자를 대상으로 웹과 앱의 연결을 함께 운영해야 할 때, URL 기반으로 구조를 잡을 수 있다는 점이 실무적으로 편리해요.
- **다양한 안드로이드 유입 경로를 다루는 캠페인**
- 검색, 광고, 콘텐츠, 메시지 등 여러 진입 경로에서 사용자가 들어올 떄도 적합해요. 유입 채널이 다양할수록 연결 경험이 흔들리기 쉬운데, 앱 링크는 안드로이드 환경에서 비교적 일관된 앱 연결 구조를 설계하는 데 도움이 됩니다.
### **4\. 디퍼드 딥링크: 앱 미설치 사용자까지 연결하는 방식**
앞서 살펴본 URL 스킴 딥링크, 유니버설 링크, 앱 링크는 사용자를 어떤 방식으로 앱에 연결할지에 대한 이야기였다면, 디퍼드 딥링크는 조금 다른 관점에서 이해할 필요가 있어요. **앱이 아직 설치되지 않은 사용자도 앱 설치 이후 원래 의도한 화면까지 이어서 보낼 수 있는가**를 다루는 개념이에요.
예를 들어 사용자가 프로모션 광고를 클릭했는데 앱이 없다면 먼저 앱을 설치하게 돼요. 이때 앱 설치 후 첫 실행에서 메인 화면이 아니라, 원래 클릭했던 이벤트 페이지나 상품 화면으로 이어지게 만드는 것이 디퍼드 딥링크예요.
**장점**
- 앱 미설치 사용자도 설치 후 원하는 화면으로 이어 보낼 수 있어요.
- 광고, CRM, 레퍼럴처럼 클릭 당시의 맥락을 설치 이후까지 유지하기 좋아요.
- 앱 설치 이후 다시 원하는 페이지를 찾게 하는 탐색 단계를 줄일 수 있어요.
**한계**
- 단순 링크만으로 되는 것이 아니라, SDK 설정이나 앱 내 처리 구조가 함께 준비되어 있어야 해요.
- 운영체제, 채널, 캠페인 환경에 따라 기대한 대로 동작하지 않을 수 있어요.
- 설치 이후 어떤 값을 넘기고 어떻게 처리할지까지 함께 설계해야 해요.
- 디퍼드 딥링크는 자체 구축도 가능하지만 SDK 처리, 설치 후 파라미터 전달, OS 업데이트 대응까지 직접 관리해야 해서, 대부분의 팀은 MMP에 위임하는 편이에요. 직접 구축과 MMP 활용 중 선택은 운영 리소스와 측정 정확도를 기준으로 판단해야 해요.
**실무에서는 이렇게 활용하세요**
- **앱 설치가 전제되는 신규 유저 유입 캠페인**
앱 설치 광고나 신규 프로모션처럼 앱 미설치 사용자가 많이 유입되는 캠페인에 적합해요.
설치 이후에도 처음 클릭한 프로모션 화면이나 특정 상품 페이지로 자연스럽게 이어질 수 있어, 설치 직후 이탈을 줄이는 데 도움이 돼요.
- **리타겟팅처럼 다시 관심을 이어가야 하는 캠페인**
- 사용자가 광고를 눌렀지만 앱이 없는 상태라면, 설치 이후에도 원래 관심을 보였던 페이지로 연결되는 흐름이 중요해요. 특히 장바구니 이탈, 상품 조회 이탈, 재참여 유도 캠페인에서는 클릭 당시의 맥락을 유지하는 데 유용해요.
- **레퍼럴 캠페인처럼 특정 정보 전달이 필요한 경우**
- 친구 초대, 추천 코드 입력, 초대 보상 지급처럼 설치 후에도 특정 정보가 유지되어야 하는 캠페인과 잘 맞아요. 예를 들어 추천 링크를 클릭한 사용자가 앱 설치 후에도 추천인 코드가 반영된 화면으로 들어가야 한다면, 디퍼드 딥링크 설계가 중요해요.
- **앱 설치 후 첫 경험이 중요한 온보딩 캠페인**
- 앱을 처음 실행한 사용자를 메인 화면으로만 보내면, 방금 클릭했던 메시지나 혜택 맥락이 끊길 수 있어요. 반면 디퍼드 딥링크를 활용하면 웰컴 혜택, 특정 카테고리, 가입 완료 유도 화면처럼 첫 행동이 분명한 지점으로 연결하기 쉬워요.
## **금융권·대기업의 딥링크 도입 전, 무엇을 확인해야 하나요?**
모바일 마케팅에서는 딥링크를 전환 경로를 짧게 만드는 도구로 보지만, 금융권과 대기업에서는 마케팅팀만의 판단으로 도입을 결정하기 어려운 경우가 많아요.
실제로는 보안, 인프라, 플랫폼 조직과 함께 검토가 진행되기 때문에 마케터도 **어떤 지점에서 내부 검토가 필요해지는지**는 미리 알고 있어야 해요.
특히 딥링크나 디퍼드 딥링크가 SDK와 함께 운영될 때는, 연결 기능 자체보다 **데이터 수집 범위**, **저장 구조**, **고객사별 데이터 분리 방식**, **설정 제어 가능 여부** 같은 질문이 함께 나올 수 있어요.
이런 내용은 마케터가 직접 판단할 영역은 아니지만, [딥링크가 포함된 플랜](https://www.airbridge.io/ko/deeplink-plan)의 도입 논의 과정에서 자주 확인되는 포인트인 만큼 미리 정리해 두면 내부 협업과 검토 속도를 높이는 데 도움이 돼요.
[03_enterprise_checklist_illustrated.svg]
### **심의 통과를 위해 미리 정리해두면 좋은 4가지**
1. **데이터 저장 구조:** 수집 데이터가 어디에 저장되는지, 고객사별 데이터가 어떤 방식으로 분리되는지 확인이 필요해요. 특히 금융권은 우리 데이터가 다른 고객사 데이터와 어떻게 분리되는지 상세히 보고 있어요.
2. **자동 수집 범위:** 딥링크 기능만 사용하더라도 SDK가 기본적으로 어떤 항목을 자동 수집하는지 먼저 확인해야 해요. 이 범위가 불명확하면 보안 심의 단계에서 설명이 어려워질 수 있어요.
3. **기능 최소화 가능 여부:** 딥링크/디퍼드 딥링크만 우선 도입하고 싶다면, 추적 기능이나 자동 실행 기능을 어디까지 제한할 수 있는지 검토해야 해요.
4. **보안 심의 대응 가능성:** 기능 설명만으로는 부족하고, 데이터 흐름·저장 구조·수집 항목·설정 옵션을 내부 검토 문서 수준으로 설명할 수 있어야 해요. 엔터프라이즈 조직에서는 이 부분이 실제 도입 속도를 좌우해요.
## **딥링크 도입을 검토하는 마케터가 가장 많이 묻는 질문만 모았어요**
### **Q1. iOS만 우선 적용해도 될까요, 양 OS를 동시에 적용해야 하나요?**
캠페인 타겟 비중에 따라 달라요. 사용자 OS 비중을 먼저 확인하고, 어느 한쪽이 70% 이상이라면 그쪽부터 우선 적용하는 팀도 많아요. 다만 광고, CRM, 레퍼럴처럼 같은 캠페인을 양 OS에 동시에 운영한다면, 처음부터 유니버설 링크와 앱 링크를 함께 설계하는 편이 운영 효율 측면에서 더 좋아요.
### **Q2. 디퍼드 딥링크는 직접 구축하는 게 좋을까요, MMP 솔루션을 쓰는 게 좋을까요?**
직접 구축도 가능하지만 SDK 처리, 설치 후 파라미터 전달, OS 정책 변경 대응까지 직접 관리해야 해요. 대부분의 팀은 측정 정확도와 운영 리소스를 고려해 MMP에 위임하는 편이에요. 자체 데이터 인프라가 충분하고 보안 정책상 외부 SDK 도입이 어려운 조직이라면 자체 구축을 검토하기도 해요.
### **Q3. 딥링크 도입 후 효과는 어떤 지표로 측정해야 하나요?**
가장 먼저 보는 지표는 클릭에서 도착 페이지까지의 도달률과 도착 페이지에서 다음 전환 행동까지의 전환율이에요. 메인 화면 경유 대비 이탈이 얼마나 줄었는지, 목적 페이지에서의 행동 완료율이 얼마나 올랐는지 비교하는 거죠. 리타겟팅이나 레퍼럴 캠페인이라면 클릭 당시 맥락 유지 후 재참여율도 함께 봐야 해요.
### **Q4. 보안 심의 통과를 위해 미리 준비해야 할 자료는 무엇인가요?**
금융권·대기업에서 자주 요청받는 자료는 데이터 수집 항목 명세서, 저장 구조 및 고객사별 분리 방식 설명, SDK 자동 수집 범위와 비활성화 옵션, 데이터 보관·삭제 정책 등이에요. 이 4가지를 미리 정리해두면 보안 심의 단계에서 설명 부담이 크게 줄어요.
### **Q5. 딥링크를 적용했는데도 원하는 화면이 바로 열리지 않는 이유는 무엇인가요?**
딥링크를 적용했다고 해서 모든 사용자에게 항상 같은 방식으로 동작하는 것은 아니에요. 사용자의 앱 설치 여부, 운영체제, 링크 방식, 앱 설정 상태에 따라 실제 연결 경로가 달라질 수 있어요.
예를 들어 앱 설치 사용자는 특정 페이지로 바로 이동할 수 있지만, 앱이 설치되지 않은 사용자는 설치 이후 별도 흐름을 거쳐야 할 수 있어요. 그래서 실무에서는 어떤 조건에서 특정 경로로 연결되는지를 함께 점검하는 것이 중요해요.
## **우리 캠페인에 딥링크 설계가 필요한지 점검해 보세요**
지금 운영 중인 캠페인을 떠올리며 아래 5가지 중 몇 개가 해당하는지 체크해 보세요. 2개 이상 해당한다면 단순 링크보다 딥링크 설계가 필요한 단계예요.
1. **클릭 후 도착해야 할 페이지가 명확한 캠페인이 있나요? →** 이벤트 페이지, 쿠폰, 장바구니처럼 목적지가 분명한 캠페인일수록 메인 화면 경유는 이탈을 만들어요.
2. **iOS·안드로이드 사용자를 하나의 링크로 운영해야 하나요? →** 운영체제별로 따로 링크를 만들고 있다면 유니버설 링크와 앱 링크를 함께 다루는 구조가 필요해요.
3. **앱 미설치 사용자도 같은 흐름으로 데려와야 하나요? →** 설치 광고, 신규 프로모션, 레퍼럴 캠페인이라면 디퍼드 딥링크가 사실상 필수예요.
4. **설치 이후 첫 화면에서 추천 코드나 쿠폰 같은 값이 유지돼야 하나요? →** 레퍼럴, 웰컴 혜택, 초대 보상처럼 클릭 시점의 정보가 설치 이후로 이어져야 한다면 디퍼드 딥링크와 파라미터 설계가 함께 필요해요.
5. **보안·인프라 팀이 함께 검토하는 도입 절차가 있나요? →** 금융권·대기업이라면 SDK 자동 수집 범위와 데이터 분리 구조까지 검토 대상이에요.
위 5가지 중 어디에 해당하는지에 따라 어떤 딥링크 방식을 선택해야 하는지가 달라져요. 앞서 살펴본 4가지 방식과 캠페인별 적합도를 기준으로 우리 팀의 캠페인부터 매핑해보세요.
## **딥링크 도입, 어디부터 시작해야 할지 막막하다면**
딥링크는 광고, CRM, 리타겟팅, 레퍼럴처럼 클릭 이후 맥락이 중요한 캠페인에서 특히 효과를 발휘해요. 다만 어떤 방식을 어떤 캠페인에 적용할지는 기획 단계부터 함께 검토해야 해요.
**에어브릿지 딥링크 플랜은 10K MAU까지 무료로, 무제한 링크와 QR 코드, 디퍼드 딥링크, 브랜드 도메인, 실시간 분석까지 제공**해요. 우리 팀에 딥링크 설계가 필요한지 먼저 확인하고 싶다면, 지금 바로 데모 신청으로 가볍게 시작해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 마케팅 ROI를 더 정확하게 보는 법: 크로스 채널 어트리뷰션 툴 활용 가이드
URL: https://www.airbridge.io/ko/blog/how-to--monitor-marketing-roi-cross-channel-attribution-tool
Language: Korean
Published: 2026-05-21
Author: Team Airbridge
**메타나 틱톡 광고비 늘렸는데, 설치 수는 늘어도 실제 구매나 가입으로 이어지지 않은 적 있나요?**
여러 광고 채널을 함께 운영하면 이런 일이 종종 생겨요. 채널마다 전환에 기여하는 방식이 다른데, 기여도를 측정하는 방식 중 가장 많이 쓰이는 ‘라스트 클릭’ 방식은 그 차이를 반영하기 어렵죠.
라스트 클릭은 전환 직전 마지막 채널에만 성과를 귀속시키는 방식으로 실제 구매까지 관여한 여러 접점은 기여도에 잡히지 않기 때문인데요.
이로 인해 전환을 마무리한 채널은 기여도가 실제보다 높게 측정될 수 있고, 앞서 앱 설치와 유저 관심을 만들었던 채널들은 기여도가 0으로 남을 수 있어요.
채널별 기여가 제대로 측정되지 않으면, 매출을 만든 채널보다 전환을 받아낸 채널에 예산을 분배하는 판단이 이뤄질 수 있죠.
이럴 때 [**크로스 채널 어트리뷰션**](https://www.airbridge.io/ko/blog/cross-channel-attribution-guide-conversion-performance)이 도움이 될 수 있어요. 라스트 클릭이 놓친 채널별 기여도를 경로 전체 기준으로 측정하기 때문이에요. 즉, **최종 전환이 발생한 광고 채널만 보는 데서 그치지 않고, 전환 이전의 접점과 채널별 역할까지 함께 해석**할 수 있어요.
이번 콘텐츠에서는 크로스 채널 어트리뷰션을 기준으로 마케팅 채널별 ROI를 정확하게 읽는 방법과, 광고비 최적화에 필요한 인사이트를 에어브릿지에서 확인하는 방법까지 살펴볼게요.
**📌핵심 포인트**
- 성과가 나는 채널이 아니라, 성과를 ‘만든’ 채널을 찾아야 해요. 전환 직전 채널만 보면 예산이 하단 퍼널에 과도하게 쏠리고, 실제로 수요를 만든 상단·중단 퍼널 기여는 과소평가될 수 있어요.
- 마케팅 ROI 개선의 본질은 집행 확대보다 예산 재배분의 정확도에 있어요. 크로스 채널 어트리뷰션은 광고 예산 투자를 늘릴 채널과 줄일 채널을 판단할 근거를 제공해 같은 예산으로 더 높은 효율을 만들어요.
## **마케팅 ROI, 왜 실제보다 왜곡되어 보일까요?**
### **ROI를 보려면 퍼널 전반의 기여를 함께 봐야 해요**
같은 전환을 두고도 채널별 성과 해석이 다른 이유는, **퍼널 단계마다 채널이 맡는 역할이 상이하기 때문**이에요. 소셜 광고는 브랜드를 처음 알리고, 리타겟팅은 관심을 유지시키는 역할을 할 수 있죠. 검색광고는 전환을 마무리하고요.
ROI를 정확하게 해석하기 위해 먼저 퍼널 기준으로 채널 역할을 나눠봐야 해요.
- **상단 퍼널(Top of Funnel): 인지와 첫 유입을 만드는 단계**
- 소셜 광고, 영상 광고, 배너처럼 노출 중심 채널은 브랜드를 처음 알리거나 관심을 형성하는 역할을 맡아요. 이 구간에서 도달, 주목도, 조회, 참여 같은 상단 퍼널 지표를 함께 봐야 하죠.
- 이 단계의 채널은 직접 전환보다 이후 검색, 재방문, 고려 단계 진입에 더 큰 영향을 줄 수 있어요.
- **중단 퍼널(Middle of Funnel): 비교와 재방문을 움직이는 단계**
- 리타겟팅, 상품 탐색 유도, 콘텐츠 소비 유도 캠페인은 사용자가 관심을 유지하고 다음 행동으로 넘어가도록 만드는 역할을 해요. 퍼널 중간 구간에서도 단순 클릭보다 체류, 상호작용, 탐색 행동처럼 다음 단계로 이어지는 신호를 같이 봐야 해요.
- 이 구간은 최종 구매보다 상세 페이지 조회, 회원가입 시작, 장바구니 진입 같은 중간 이벤트에서 먼저 반응이 나타날 수 있어요.
- **하단 퍼널(Bottom of Funnel): 전환을 마무리하는 단계**
- 검색광고, 구매 의도가 높은 사용자 대상 리마케팅, CRM 기반 재유입 메시지는 이미 관심이 높은 사용자를 다시 데려와 전환을 닫는 역할을 하는 경우가 많아요.
- 하단 퍼널에서 전환, 구매, 리드, 가입 같은 결과 지표를 핵심으로 보지만, 이것만으로 캠페인 전체 가치를 판단하면 퍼널 앞단의 기여가 포함되지 않을 수 있어요. 하단 퍼널 지표는 중요하지만, 단독으로 보면 전체 ROI 해석이 좁아질 수 있어요.
요약하면, \*\*‘비교 가능하고 전체를 포괄하는 측정 방법’\*\*이 중요해요. 크로스 채널 어트리뷰션은 채널 역할에 따른 기여를 한 화면 안에서 비교할 수 있게 해줘요. 전환 경로 전체를 기준으로 각 채널의 기여도를 배분하는 접근 방식이죠.
## **크로스 채널 어트리뷰션 툴은 ROI 해석에 어떤 도움을 주나요?**
채널 어트리뷰션과 크로스 채널 어트리뷰션의 차이를 알면 같은 리포트에서 다른 인사이트를 읽을 수 있어요.
**채널 어트리뷰션**과 **크로스 채널 어트리뷰션**의 차이점을 정리하면 다음과 같아요.
- **채널 어트리뷰션**: 개별 채널 안에서 발생한 클릭이나 전환을 기준으로 성과를 해석하는 데 초점
- **크로스 채널 어트리뷰션**: 여러 채널을 거친 고객 여정을 함께 보고, 각 접점이 최종 전환에 어떤 역할을 했는지 비교해 해석. 채널별 성과를 따로 보는 방식이 아닌, 전환이 만들어지는 전체 경로를 기준으로 기여도를 읽는 접근
### **채널별 성과, 이렇게 해석해 보세요**
| 채널 | 리포트에서 먼저 보이는 지표 | 퍼널에서의 역할 | 해석할 때 주의할 점 |
| --- | --- | --- | --- |
| 검색광고 | 직접 전환, 가입, 구매 | 전환 직전 수요 포착 | 실제 기여보다 높게 평가될 수 있음 |
| 소셜/영상 광고 | 도달, 클릭, 조회, 참여 | 인지 형성, 관심 유도 | 직접 전환만 보면 기여가 작게 보일 수 있음 |
| 배너/디스플레이 | 노출, 클릭, 방문 | 재노출, 리마인드, 보조 기여 | 검색·방문·전환의 선행 접점일 수 있음 |
| 이메일·푸시 | 오픈, 클릭, 재방문 | 재참여, 재유입, 유지 | 전환 마감뿐 아니라 재방문 유도 기여도 함께 봐야 함 |
표에 정리한 내용은 모든 캠페인에 동일하게 적용되는 고정 규칙이라기보다, **채널별 성과를 해석할 때 일반적으로 참고할 수 있는 기준**이에요. 같은 채널도 캠페인 목적과 운영 방식에 따라 퍼널 안에서 역할이 달라질 수 있어요.
**크로스 채널 어트리뷰션 툴은 이런 역할 차이를 한 화면 안에서 비교**할 수 있게 해줘요. 이로 인해 직접 전환 수치만 보고 예산을 늘리거나, 반대로 수치가 낮다는 이유만으로 채널을 줄이는 판단에서 벗어날 수 있죠.
## **마케팅 채널별 ROI 개선 위해 함께 봐야 할 4가지 데이터**
채널별 역할과 실제 기여를 함께 보려면, 아래 4가지 데이터를 같이 확인해 보세요.
### **1\. 초기 ROAS가 낮아도 재구매율이 높은 채널을 먼저 보세요**
최종 전환은 채널별 결과를 확인하는 데 필요한 지표예요. 다만 이 수치를 곧바로 채널의 전체 기여도로 받아들이면 해석에 오차가 생길 수 있어요. 전환 직전에 포착된 채널과, 그 전환 가능성을 높인 채널이 다를 수 있기 때문이에요.
[구글이 소개한 BCG 분석에 따르면, 소비자는 영상 시청, 소셜 피드 탐색, 검색, 구매를 넘나드는 동적인 접점 구조 안에서 브랜드를 탐색](https://business.google.com/en-all/think/consumer-insights/new-consumer-decision-making-process/?)해요.
이런 환경에서는 마지막에 포착된 채널이 실제보다 더 중요하게 해석될 수 있어요. 이로 인해 최종 전환 수치는 성과가 어디서 발생했는지 확인하는 데는 유효하지만, 무엇이 전환을 만들었는지 바로 설명해 주지는 못해요.
#### **✅ 이 지표는 이렇게 해석해 보세요**
**초기 ROAS가 낮더라도 재방문·재구매가 높은 채널은 다시 봐야 해요.** 반대로 첫 구매는 잘 나오지만 7일·30일 리텐션이 낮은 채널은 겉보기 효율에 비해 실제 가치가 낮을 수 있어요.
특히 앱 마케팅에서는 설치 비용만으로 채널 효율을 판단하기 어려워요. 설치 이후 실제 전환과 재방문으로 이어지는 사용자를 얼마나 확보했는지까지 봐야 해요.
### **2\. 직접 전환이 낮은 채널의 예산을 바로 삭감하지 마세요**
직접 전환이 높지 않아도 전환 여정 초반이나 중간에 반복적으로 등장하는 채널이 있어요. 이런 채널은 **마지막 성과를 가져오지는 않더라도, 브랜드를 처음 인지시키거나 다시 떠올리게 만들고, 이후 검색이나 재방문으로 이어질 가능성**을 높일 수 있어요.
라스트 클릭 기준으로 기여가 작아 보이는 채널도 전환 경로 안에서 반복적으로 등장했을 수 있어요.
특히 디스플레이 광고는 마지막 클릭 기준으로는 기여가 작아 보일 수 있지만, 다중 접점 어트리뷰션 채널에서는 구매 가능성이나 이후 행동에 미치는 영향이 다르게 나타날 수 있어요. 즉, 직접 전환이 낮다는 이유만으로 채널 가치를 낮게 평가하여 실제 기여를 놓치지 마세요.
#### **✅ 이 지표는 이렇게 해석해 보세요**
직접 전환이 낮은 채널을 바로 예산 삭감 후보로 두기보다, 전환 전에 얼마나 자주 등장하는지 먼저 보세요. 선행 접점 비중이 높은 채널은 전환을 닫는 역할보다, 전환 가능성을 높이는 역할이 더 클 수 있어요.
특히 브랜딩, 리타겟팅, 콘텐츠형 광고는 최종 ROAS보다 검색량 증가, 재방문 유도, 전환 경로 내 반복 노출 여부까지 함께 봐야 실제 영향력을 더 정확하게 판단할 수 있어요.
### **3\. 전환율이 떨어졌을 때 먼저 이탈 구간을 확인하세요**
전환율 하락의 원인이 항상 매체에 있지는 않죠.
최근 학술 연구들은 [장바구니 이탈이 가격, 배송, 결제 복잡성, 신뢰 문제뿐 아니라 탐색 행동과 장바구니 재방문 여부 같은 전환 과정의 신호와도 연결된다고 설명](https://link.springer.com/article/10.1007/s12525-024-00697-0?)해요. 즉, 유입이 충분해도 퍼널 중간 단계의 마찰이 커지면 광고 효율 저하처럼 보일 수 있어요.
#### **✅ 이 지표는 이렇게 해석해 보세요**
**전환율이 하락했을 때는 어느 단계에서 이탈이 커졌는지 먼저 확인해 보세요.** 특정 채널만 초반 이탈이 높다면 광고와 랜딩 페이지의 메시지 불일치가 원인일 수 있고, 장바구니 재방문이 줄거나 탐색이 길어질수록 이탈이 커진다면 퍼널 경험 자체를 먼저 점검해야 해요.
### **4\. 재방문·리텐션 흐름으로 채널의 실질 가치를 판단하세요**
재방문·리텐션은 **전환 이후의 품질**을 보는 지표예요. 특히 앱, 구독형 서비스, 재구매 비중이 높은 커머스에서는 첫 전환 이후 사용자가 얼마나 잔류하는지가 채널 가치를 더 잘 보여줄 때가 많아요.
첫 구매는 많이 이뤄졌지만 이후 재방문이 거의 없거나, 반대로 첫 전환은 적어도 꾸준히 돌아와 추가 행동으로 이어지는 경우로 나뉘기 때문이에요.
#### **✅ 이 지표는 이렇게 해석해 보세요**
**첫 전환 수보다, 전환한 사용자가 이후에도 다시 들어오는지 먼저 보세요.** 채널별 리텐션 흐름이 다르면 같은 구매 100건도 의미가 달라져요. 특히 앱 마케팅에서는 설치 이후 회원가입, 첫 구매, 재방문까지 이어지는 비율을 같이 봐야 단기 성과가 아니라 반복 가능한 유입인지 판단할 수 있어요.
## **채널 효율을 다시 판단할 임계값과 예산 재배분 순서**
위 4가지를 함께 본다고 해도 언제 예산을 재조정해야 하는지 모호하면 결정이 미뤄질 수 있죠. 절대 기준은 아니지만, 자체 캠페인의 베이스라인과 비교할 때 참고할 만한 신호를 정리하면 다음과 같아요.
| 신호 | 점검 임계값 (참고) | 다음 액션 |
| --- | --- | --- |
| 초기 ROAS는 평균 이하 + D30 재구매율이 상위 25% | 채널 가치 재평가 후보 | 예산 유지 또는 +10% 테스트 |
| 직접 전환은 하위 30% + 선행 접점 비중은 상위 30% | 보조 기여 채널 | 인지·리타겟팅 KPI로 별도 평가 |
| D7 리텐션이 전 분기 대비 10%p 이상 하락 | 유입 품질 저하 신호 | 매체 손대기 전 크리에이티브·타겟팅 진단 |
| 장바구니 단계 이탈이 주간 5%p 이상 증가 | 퍼널 마찰 의심 | 매체 효율로 단정하지 말고 LP·체크아웃 점검 |
임계값 자체보다 중요한 건 자체 베이스라인 대비 변화 폭이에요. 같은 D7 리텐션 30%도 자사 평균이 25%인 채널이라면 강한 신호고, 평균이 35%인 채널이라면 약한 신호예요.
### **예산 재배분은 3단계 순서로 결정하세요**
채널을 손볼 때 직관 대신 동일한 순서를 거치면 의사결정의 편차가 줄어요.
1. **나누기:** 모든 채널을 직접 전환·선행 접점·재방문 기여 세 버킷으로 분류해요.
2. **비교:** 같은 버킷 안에서 CAC(또는 CPI) 대비 LTV를 비교해요. 다른 버킷끼리 비교하면 역할이 다른 채널을 같은 잣대로 평가하게 돼요.
3. **결정:** 같은 버킷 안에서 효율이 낮은 채널부터 손대요. 버킷 간 비중 자체는 한 주 단위가 아니라 분기 단위로만 조정해요.
이렇게 분류하고 비교한 다음 결정하는 순서가 자리잡히면, 직접 전환이 떨어졌으니 예산을 줄이는 단편적인 판단을 피할 수 있어요.
## **에어브릿지에서 채널 기여와 ROI 인사이트를 어떻게 확인하나요?**
### **1\. 전환을 만든 채널과 마무리한 채널을 나눠서 보세요**
[에어브릿지(Airbridge)에서는 액츄얼스 리포트로 채널별 전환 성과를 집계하고, 터치포인트 분석 리포트로 전환 전에 발생한 터치포인트 개수를 확인할 수 있어요.](https://help.airbridge.io/ko/guides/airbridge-report-overview?) 이로 인해 최종 전환이 많이 잡힌 채널과, 전환 경로 안에 반복적으로 등장한 채널을 같은 화면 안에서 나눠서 볼 수 있어요.
특히 **어트리뷰션 윈도우를 7일과 30일로 같이 띄워서 비교**해 보세요. 윈도우를 늘렸을 때 기여도가 30% 이상 증가하는 채널은 선행 접점 역할이 큰 채널일 가능성이 높아요. 짧은 윈도우 한 가지로만 보면 인지·리타겟팅 채널의 기여가 누락된 채로 의사결정을 내리게 돼요.
#### **✅ 이렇게 활용해 보세요**
최종 전환 수치가 높은 채널을 바로 증액 후보로 두지 말고, 터치포인트 분석 리포트를 함께 보세요. 직접 전환은 낮아도 전환 전에 반복적으로 등장하는 채널이라면, 마지막 성과보다 앞단 기여가 더 큰 채널일 수 있어요. 이 구분이 되어야 예산 증액도 최종 전환 중심이 아니라 실제 기여 방식에 맞춰 조정할 수 있어요.
### **2\. 퍼널 리포트로 성과 저하 원인을 찾아보세요**
[에어브릿지 퍼널 리포트](https://help.airbridge.io/ko/guides/funnel-report-viewing)를 보면 유입 이후 어느 단계에서 이탈이 커지는지 진단할 수 있어요. 전환이 줄었을 때 원인을 바로 매체 효율로 단정하지 않고, 랜딩 이후 경험이나 가입·결제 과정까지 함께 점검할 수 있어요.
**채널을 그룹핑한 상태에서 단계별 전환율을 나란히 비교**하면 원인이 더 빨리 보여요. 특정 채널만 첫 단계 이탈률이 평균보다 5%p 이상 높다면 그 채널의 광고 메시지와 랜딩 페이지 정합성을 먼저 점검하세요. 반대로 여러 채널이 같은 단계에서 동시에 이탈이 늘었다면 매체가 아니라 퍼널 자체의 마찰을 의심해야 해요.
#### **✅ 이렇게 활용해 보세요**
성과가 떨어졌을 때는 먼저 퍼널 리포트에서 이탈 구간을 살펴보세요. 특정 채널만 초반 이탈이 높다면 메시지와 랜딩의 문제일 수 있고, 여러 채널에서 같은 단계 이탈이 높다면 전환 과정 자체를 먼저 손보는 편이 더 맞아요.
### **3\. 리텐션과 레비뉴 리포트로 유입의 질까지 확인해 보세요**
에어브릿지에서는 [리텐션 리포트로 재방문 흐름을, 레비뉴 리포트로 코호트별 매출을 함께 확인](https://help.airbridge.io/ko/guides/airbridge-feature-list?)할 수 있어요. 첫 전환 수만으로는 보이지 않던 유입의 질까지 비교해 보세요.
같은 첫 전환 100건이라도 **D7과 D30 코호트로 나눠서 보면** 채널별 잔존율이 2~3배까지 벌어지는 경우가 있어요. 첫 전환 비용이 비슷한 채널끼리는 반드시 코호트 단위로 한 번 더 비교한 뒤 증액 여부를 결정하세요.
#### **✅ 이렇게 활용해 보세요**
첫 구매나 설치 수가 비슷한 채널이라도, 이후 재방문율이나 매출 기여는 다를 수 있어요. 초기 성과가 비슷하다면 리텐션과 레비뉴까지 함께 보고, 실제로 오래 남는 유입이 어디서 들어오는지 알아보세요.
## **어트리뷰션 툴을 도입하기 전, 성과 측정 구조부터 점검해 보세요**
지금 쓰는 측정 체계로 아래 다섯 질문에 답할 수 있는지 먼저 확인해 보세요. 답이 모호한 항목이 많을수록 채널 기여를 잘못 해석하고 있을 가능성이 높아요.
- **직접 전환이 많은 채널과 선행 접점이 많은 채널을 같은 화면에서 비교할 수 있나요?**
두 그룹이 서로 다른 리포트에 분산돼 있다면 예산 결정이 직접 전환 쪽으로 기울 수밖에 없어요.
- **어트리뷰션 윈도우를 7일과 30일로 동시에 돌려본 적이 있나요?**
한쪽만 보면 인지·리타겟팅 채널의 기여가 누락된 채로 의사결정을 내리게 돼요.
- **첫 전환 이후 D7·D30 재방문율을 채널별로 비교하고 있나요?**
CAC 대비 LTV를 채널별로 정렬하지 않으면 효율 낮은 채널에 예산이 계속 흘러가요.
- **전환율이 떨어졌을 때 매체 효율을 의심하기 전에 퍼널 이탈 구간부터 보나요?**
이 순서가 반대면 매체를 손대다가 진짜 원인을 놓칠 수 있어요.
- **메타·구글 같은 매체 자체 어트리뷰션과 MMP 어트리뷰션을 따로 보고 있나요, 합산하고 있나요?**
둘을 합산하면 같은 전환이 두 번 카운트돼서 ROAS가 부풀려져요.
다섯 개 중 두 개 이상에서 ‘잘 모르겠다’는 답변이 나온다면, 도구를 더 늘리기보다 지금 가진 데이터를 보는 방식부터 정비하세요.
## **지금 리포트에서 확인하고, 마케팅 예산을 점검해 보세요**
마케팅 ROI를 끌어올리는 출발점은 광고비를 늘리거나 효율 낮은 채널을 줄이는 것이 아니에요. 같은 예산 안에서 **'성과를 만든 채널'과 '성과를 거둔 채널'을 분리**하고, 동일한 광고비로 더 나은 의사결정을 하는 일이죠.
에어브릿지의 터치포인트 분석 리포트와 퍼널 리포트를 활용하면 라스트 클릭이 놓치는 영역, 즉 선행 접점, 재방문, 퍼널 이탈 구간을 읽고 예산을 효율적으로 활용할수 있어요. 우리 기업에 들어온 숫자를 정확하게 읽고 싶다면 지금 바로 에어브릿지 데모를 신청해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 에어브릿지 x Pinterest Ads 연동, 노출에서 전환까지 측정하세요
URL: https://www.airbridge.io/ko/blog/airbridge-pinterest-ads-integration
Language: Korean
Published: 2026-05-12
Author: Harper (Trang Nguyen)
Pinterest 유저는 그냥 피드를 스크롤하지 않아요. 인테리어를 계획하거나, 자기만의 스타일을 찾거나, 선물 아이디어를 모으는 것처럼 행동으로 이어질 영감을 적극적으로 탐색하는 사람들이죠.
의도를 가지고 탐색하는 유저가 모여 있으니, 광고가 전환으로 이어질 확률도 높아요. 다만 그 성과를 제대로 보려면 노출부터 전환까지 이어지는 어트리뷰션이 필요해요.
이번에 **에어브릿지와 Pinterest Ads 연동이 시작**되면서, 앱과 웹을 넘나드는 유저 여정을 대시보드 하나에서 추적하고 측정할 수 있게 됐어요.
## 탐색에서 전환까지, Pinterest Ads의 구조
**Pinterest는 비주얼 디스커버리 엔진**이에요. 일반적인 소셜 플랫폼에서 유저가 콘텐츠를 수동적으로 소비하는 것과 달리, Pinterest 유저는 자신의 목표에 맞는 아이디어, 솔루션, 제품을 직접 찾고 있어요.
Pinterest 광고 포맷도 이 탐색 흐름에 맞춰 설계돼 있어요.
- **Standard Ads**: 피드, 검색 결과, 관련 핀에 노출되는 기본 이미지 광고예요.
- **Video Ads**: 스크롤 중 자동 재생되는 영상 포맷으로, 제품 데모나 브랜드 스토리에 적합하죠.
- **Idea Ads**: 이미지, 영상, 텍스트를 조합한 멀티페이지 포맷이에요. 튜토리얼이나 단계별 가이드에서 많이 쓰이고요.
- **Carousel Ads**: 여러 장의 이미지를 스와이프하는 구조로, 다양한 제품이나 기능을 한 단위에서 보여주죠.
- **Shopping Ads**: 카탈로그에서 상품을 바로 가져와 실시간 가격과 재고를 표시해요.
- **Collections Ads**: 히어로 크리에이티브(이미지 또는 영상) 아래에 상품 핀이 함께 붙는 구조예요.
- **Lead Ads**: 유저가 Pinterest를 떠나지 않고 바로 가입하거나 정보를 요청할 수 있는 네이티브 폼 광고죠.
- **Quiz Ads**: 몇 가지 질문에 답하면 맞춤 상품이나 추천으로 연결돼요.
- **Showcase Ads**: 여러 핀을 하나의 광고로 큐레이션해서, 브랜드 스토리나 컬렉션을 한눈에 보여주는 포맷이에요.
- **Premiere Spotlight Ads**: Pinterest 내 가장 주목도 높은 지면에 배치되는 프리미엄 포맷으로, 구매 의도가 높은 오디언스에 대규모로 도달하고 싶을 때 쓰여요.
[Pinterest integration.webp]
모두 유저 행동을 유도하도록 설계된 포맷이에요. 에어브릿지 연동이 되면, 어떤 크리에이티브가 앱과 웹에서 성과를 내고 있는지 대시보드에서 바로 확인할 수 있죠.
## 에어브릿지에서 Pinterest Ads 성과를 추적하는 방법
에어브릿지와 Pinterest Ads의 연동은 트래킹 링크와 포스트백 설정을 기반으로 동작해요. 이를 통해, Pinterest Ads에서 발생한 전환 이벤트를 에어브릿지에서 모니터링하고 어트리뷰션할 수 있죠.
구체적으로는 이런 것들이 가능해요.
- 온사이트 검색, 상품 조회, 가입, 위시리스트, 구매 완료까지 퍼널 전체를 측정해요.
- 앱이나 웹에 맞는 커스텀 이벤트를 정의해서 더 세밀하게 추적해요.
- Pinterest에서 시작해 웹사이트나 앱으로 이어지는 유저 여정도 잡아내요.
- 프라이버시 세이프 방식과 확률적 모델링을 지원해서, 디바이스와 플랫폼을 넘나드는 어트리뷰션까지 커버해요.
어떤 광고가 초기 관심을 만들고, 어떤 광고가 전환으로 이어지는지 크리에이티브별로 비교해 볼 수 있어요.
## 크리에이티브 성과, 전환까지 연결해서 보세요
Pinterest에서 유저의 관심을 끌었다면, 그다음은 그 관심이 실제로 어디까지 이어졌는지를 볼 차례예요. 이번 연동으로 노출 수나 클릭 수 같은 지표를 넘어, Pinterest 캠페인이 비즈니스 성장에 어떻게 기여하는지 명확하게 파악할 수 있어요.
바로 시작하고 싶다면 [에어브릿지 헬프센터](https://help.airbridge.io)에서 설정 가이드를 확인하거나, **에어브릿지 팀에 문의해** 주세요. 또한 이벤트 매핑이나 캠페인 성과 분석 등 궁금한 점이 있으시다면 언제든 연락 주세요.
---
## 2026 AI 통합 바우처(클라우드 바우처) MMP 도입 A to Z 가이드
URL: https://www.airbridge.io/ko/blog/ai-cloud-voucher-mmp-guide-insight
Language: Korean
Published: 2026-04-28
Author: Jaehyuk Kim
앱 마케팅을 제대로 하려면 MMP가 필요하다는 건 알지만, 비용이 부담돼서 도입을 미뤄왔던 중소기업이 많아요. 특히 초기 스타트업이나 앱 서비스를 막 시작한 팀이라면, 연간 수백~수천만 원의 MMP 이용료는 쉽게 결정하기 어려운 금액이에요.
2026년 AI 통합 바우처(클라우드 바우처)를 활용하면 이용료의 최대 75%를 정부 지원금으로 충당할 수 있는데요. 바우처 사업이 어떤 구조인지, 비용은 어떻게 나뉘는지, MMP는 어떤 걸 선택할 수 있는지 궁금했던 분들을 위해 이번 가이드를 준비했어요.
그리고 **올해 바우처 공급 서비스 중 MMP로 선정된 건 에어브릿지가 유일**한데요. 오늘은 2026 AI 통합 바우처(클라우드 바우처)의 비용 분담 구조, 지원 범위, 사업 일정부터 에어브릿지가 제공하는 기능까지 모두 정리해 드릴게요.
**핵심 포인트**
- 2026년 AI 통합 바우처로 도입할 수 있는 MMP는 에어브릿지가 유일해요.
- 이용료의 80%를 정부와 공급기업이 부담하기 때문에, 수요기업은 20%만 직접 내면 돼요.
- 기업당 최대 6,910만 원까지 지원받을 수 있고, 서비스 이용료와 심화 컨설팅(최대 2,500만 원)이 모두 포함돼요.
## **AI 통합 바우처(클라우드 바우처), 어떤 사업인가요?**
공식 사업명은 \*\*'중소기업 클라우드 서비스 보급·확산 사업'\*\*이고, AI 통합 바우처(클라우드 바우처)는 이 사업의 통칭이에요.
과학기술정보통신부와 NIPA가 중소기업의 클라우드 및 AI 서비스 도입을 지원하는 사업으로, 정부가 이용료의 상당 부분을 부담해, 기업이 비용 걱정 없이 클라우드 기반 솔루션을 도입할 수 있도록 돕는 구조예요.
쉽게 말하면, 중소기업이 업무에 필요한 SaaS 솔루션을 정부 지원금으로 도입할 수 있는 프로그램이에요. 그래서 마케팅 분석 도구, HR 솔루션, 협업 도구 등 다양한 클라우드 서비스가 지원 대상에 포함돼요.
### **1\. 2026년, 'AI 통합 바우처'로 바뀌면서 달라진 것들**
올해부터 기존 '클라우드 바우처'가 \*\*'AI 통합 바우처'\*\*라는 이름으로 확대 개편됐어요. 이름만 바뀐 게 아니라, 사업의 방향 자체가 달라졌어요.
가장 큰 변화는 **AI 활용도가 평가 기준에 포함**됐다는 점이에요. 수요기업 평가 배점(총 100점)을 보면 그 비중을 알 수 있어요.
| 평가 항목 | 배점 |
| --- | --- |
| 지원 필요성 | 20점 |
| 서비스 이용 계획의 적절성 | 30점 |
| 디지털 전환 및 AI 활용 | 30점 |
| 기대효과 | 20점 |
'디지털 전환 및 AI 활용' 항목이 30점으로 가장 높은 비중을 차지해요. AI 기능이 있는 솔루션을 선택하고, 이를 구체적으로 활용할 계획을 제시한 기업이 선정에서 유리해졌다는 뜻이기도 해요.
### **2\. 지원 규모와 대상: 60개사, 기업당 최대 6,910만 원**
2026년 총 예산은 **41억 4,600만 원**이고, 수요기업 **60개사**가 선정돼요. 기업당 최대 **6,910만 원**까지 지원받을 수 있는데, 이 안에 심화 컨설팅 비용 2,500만 원도 포함되어 있어요.
지원 대상은 [중소기업기본법 제2조](https://www.law.go.kr/%EB%B2%95%EB%A0%B9/%EC%A4%91%EC%86%8C%EA%B8%B0%EC%97%85%EA%B8%B0%EB%B3%B8%EB%B2%95/%EC%A0%9C2%EC%A1%B0)에 해당하는 국내 중소기업이에요(중견기업·대기업은 신청 불가). 2020~2025년 동일 사업으로 지원받은 기업이나, AX바우처·AI바우처 등 유사 사업과 중복 수혜 중인 기업은 제외돼요.
### **3\. 사업 일정: 수요기업은 이미 솔루션 선정 절차 진행 중이에요**
사업 일정을 정리하면 이렇게 돼요.
| 단계 | 시기 |
| --- | --- |
| 수요기업 신청 | 2026년 4월 3일 ~ 4월 21일 |
| 서류 평가 | 4월 4주차 |
| 발표 평가 (1.5배수, 90개사) | 5월 3주차 |
| 최종 선정 (60개사) | 5월 4주차 |
| 서비스 이용 시작 | 6월~ (순위에 따라 6~9월) |
| 서비스 이용 종료 | 2026년 12월 31일 |
현재는 수요기업 신청이 마감된 상태이고, 선정된 기업들이 공급업체(솔루션)를 선택하는 선정심사 단계예요.
## **에어브릿지, MMP 중 유일하게 서비스 공급업체로 선정**
### **1\. AB180 에어브릿지 공급 서비스로 최종 선정**
AB180이 운영하는 에어브릿지(Airbridge)가 2026 AI 통합 바우처 공급 서비스로 최종 선정됐어요. 광고 성과 측정, 유저 분석, 딥링크 등 마케팅 인프라 전반을 바우처로 도입할 수 있어요.
### **2\. 바우처로 도입할 수 있는 MMP는 에어브릿지뿐**
에어브릿지를 활용하면 앱 사용자가 어떤 광고를 보고 설치했는지, 설치 이후에는 어떤 행동을 하는지를 추적하고 분석할 수 있는데요. 앱 마케팅에서 광고비가 실제로 어디에서 성과를 내는지 파악하려면 반드시 필요한 인프라예요.
이번 바우처 사업에서 MMP로 선정된 건 에어브릿지뿐이에요. 바우처를 활용해 MMP를 도입하고 싶다면, 에어브릿지를 선택하시면 돼요.
## **에어브릿지 도입 비용, 정부가 75%를 지원해요**
### **1\. 이용료의 80%를 정부와 공급기업이 부담해요**
AI 통합 바우처 사업의 비용 분담 구조와 비용 분담 비율은 공급기업 규모에 따라 달지는데요. 에어브릿지를 운영하는 AB180은 중소기업이기 때문에, 아래 기준이 적용돼요.
| 부담 주체 | 비율 |
| --- | --- |
| 정부 지원 | 75% |
| 공급기업(에어브릿지) 부담 | 5% |
| 수요기업 부담 | 20% |
💡**참고:** 공급기업이 대기업인 경우 정부 70%, 공급기업 10%로 달라져요.
수요기업 입장에서는 전체 이용료의 20%만 직접 부담하면 돼요. 예를 들어 연간 5,000만 원 규모의 서비스라면, 실제 부담은 1,000만 원 수준이에요.
### **2\. 솔루션 이용료 + 컨설팅까지 한 번에 지원**
클라우드 서비스 이용료뿐 아니라 컨설팅 비용까지 지원 대상에 포함돼요. 에어브릿지 도입 시 초기 세팅, SDK 연동, 이벤트 설계, 활용 전략까지 컨설팅 지원을 받을 수 있어요.
기업당 최대 6,910만 원 한도 안에서 서비스 이용과 컨설팅을 함께 설계할 수 있어요. 지원 기간은 계약일부터 2026년 12월 31일까지예요.
## **웹투앱 측정부터 Airbridge AI까지, 클라우드 바우처로 모두 경험해 보세요**
### **1\. 안드로이드부터 콘솔까지, 사각지대 없는 크로스 플랫폼 측정**
에어브릿지는 안드로이드, iOS, SKAN, 웹, PC, 콘솔까지 모든 환경을 통합 측정해요. 200개 이상의 분석 지표를 제공하기 때문에, 광고 성과를 다양한 각도에서 파악할 수 있어요.
모바일 앱뿐 아니라 웹과 PC 광고까지 하나의 대시보드에서 볼 수 있다는 점이 가장 큰 강점이에요.
실제로 원티드랩은 에어브릿지로 웹투앱 유저 여정을 통합 트래킹한 뒤 CPA를 38% 절감했어요.
👉 [원티드랩, 에어브릿지 이관 후 웹투앱 유저 여정을 트래킹하고 CPA를 38% 절감했어요](https://www.airbridge.io/ko/case-studies/wantedlab-web-to-app-attribution)
### **2\. Pilot, MCP, Predictive LTV. AI가 데이터를 읽어주는 MMP**
올해 AI 통합 바우처로 명칭이 바뀐 만큼, AI 역량이 중요한 평가 기준이었어요. 에어브릿지는 이미 AI 기술 영역에서 앞서 나가는 성능을 갖추고 있어요.
| 기능 | 설명 |
| --- | --- |
| Airbridge Pilot | 대시보드 내 AI 에이전트, 자연어로 데이터 분석 |
| Airbridge MCP | Claude, ChatGPT 등 외부 AI 도구와 데이터 연동 |
| Predictive LTV | 3일치 데이터로 180일 LTV 예측 |
**Airbridge Pilot**은 대시보드 안에서 동작하는 AI 에이전트예요. 자연어로 질문하면 데이터를 분석해 인사이트를 제공해요. 전 플랜에서 무제한 사용 가능하고, 한국어·영어·일본어·중국어를 지원해요.
**Airbridge MCP**는 Claude, Claude Code, ChatGPT, Cursor 같은 외부 AI 도구와 에어브릿지 데이터를 직접 연결해주는 기능이에요. 마케터가 이미 쓰고 있는 AI 워크플로에 에어브릿지 데이터를 바로 가져다 쓸 수 있어요.
**Predictive LTV**는 단 3일치 데이터만으로 180일 후 유저 생애가치(LTV)를 예측해요. 캠페인 초기에 빠르게 성과를 판단하고 예산을 조정할 수 있게 해주는 기능이에요.
👉 [복잡한 마케팅 데이터도 이제 찾지 말고 질문하세요, Airbridge AI](https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch)
### **3\. 딥링크부터 광고사기 방지까지, 추가 비용 없는 올인원 MMP**
에어브릿지는 딥링크, 광고사기 방지, 리포트 공유, 포스트백, 유저 세그먼트, 광고비 연동까지 모든 기능을 하나의 요금제에 포함하고 있어요. 기능별로 추가 비용을 내는 구조가 아니에요.
바우처 예산 안에서 필요한 기능을 모두 사용할 수 있으니, 추가 비용 걱정 없이 도입 계획을 세울 수 있어요.
👉 [알라미, 글로벌 1위 알람 앱이 에어브릿지를 활용해 ROAS를 개선한 방법](https://www.airbridge.io/ko/case-studies/ko-alarmy)
## **클라우드 바우처로 MMP를 도입할 때 알아둘 점**
### **1\. 지원 종료 후 3년간 서비스 유지율 조사에 협조해야 해요**
바우처 지원이 2026년 12월 31일에 종료된 뒤에도, 수요기업은 3년간 클라우드 서비스 유지 여부 조사에 협조해야 해요. 별도의 유료 사용 의무는 없지만, 조사 회신은 필수예요.
### **2\. 공급업체 선택은 클라우드 서비스 지원 포털에서 해야해요**
수요기업으로 선정된 뒤, 공급업체(솔루션)는 [클라우드 서비스 지원 포털(cloudsup.or.kr)](https://www.cloudsup.or.kr)에 등록된 공급 서비스 Pool에서 선택하게 돼요. 에어브릿지는 이 Pool에 등록되어 있어요.
## **자주 묻는 질문**
### **Q1. 바우처 지원 기간이 끝나면 자동으로 유료 전환되나요?**
아니요. 2026년 12월 31일에 바우처 지원이 종료되면, 이후 유료 사용 여부는 수요기업이 자유롭게 결정할 수 있어요. 자동 전환이나 위약금 같은 건 없어요.
### **Q2. 이미 다른 MMP를 쓰고 있는데, 에어브릿지로 전환할 수 있나요?**
가능해요. 에어브릿지는 MMP 이관을 위한 전담 지원 프로세스를 갖추고 있어요. 바우처 컨설팅 예산 안에서 기존 MMP 데이터 마이그레이션과 전환 작업까지 지원받을 수 있어요.
## **지금이 에어브릿지를 시작할 최적의 타이밍이에요**
2026 AI 통합 바우처 수요기업으로 선정되었다면, 지금이 에어브릿지를 도입할 수 있는 가장 좋은 타이밍이에요. 이용료의 80%를 정부와 공급기업이 부담하고, 컨설팅까지 지원받을 수 있는 기회는 흔치 않아요.
MMP 도입이나 새로운 MMP로 교체를 고민하고 있었다면, 바우처를 활용하면 비용 부담을 크게 줄이면서 에어브릿지를 도입할 수 있어요. 바우처 지원이 가능한 지금 도입 과정에서 궁금한 점이 있다면 **에어브릿지 도입 문의**를 남겨주세요.
---
## 게임 마케터는 AI로 어떻게 데이터 분석을 할까?
URL: https://www.airbridge.io/ko/blog/how-game-marketers-use-ai-for-data-analysis
Language: Korean
Published: 2026-04-24
Author: Subin Kim
지금까지 퍼포먼스 마케터가 캠페인 성과를 분석하려면 에어브릿지와 같은 MMP 솔루션의 어트리뷰션 로직과, 대시보드 사용법을 익혀야 했어요. 그러다 보니 도구를 배우는 시간이 실제로 분석하는 시간보다 길어지곤 했죠.
[**Airbridge Pilot과 MCP**](https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch)**는 그 간극을 줄이기 위해 만들어진 Airbridge AI 솔루션**이에요. Airbridge Pilot과 MCP가 출시된 지 이제 딱 한 달이 됐어요. 지금까지 Pilot을 사용해본 고객사와 미팅을 통해 몇 가지 흥미로운 패턴을 발견할 수 있었는데요.
같은 기능을 쓰고 있는데도 어떤 유저는 캠페인 예산 재분배까지 이어지는 분석을 한 번의 대화로 뽑아내고, 어떤 유저들은 Pilot을 사용하며 원하는 답변을 얻기까지 평균 약 3번의 질문을 주고 받았어요. 원하는 답을 얻기까지 매번 조건을 보완하고 다시 묻고 서너 번 핑퐁을 주고 받는 과정이 반복된 것이죠.
반대로 같은 기능을 쓰고 있는데도 어떤 유저는 캠페인 예산 재분배까지 이어지는 분석을 한 번의 대화로 뽑아내기도 했어요. 이렇게 한두 번 만에 깔끔하게 결과물을 얻은 유저들의 질문들에는 공통된 요소를 발견할 수 있었어요.
[image1]
이 글은 그 관찰에서 출발해, **Airbridge Pilot**을 쓸 때 더 좋은 결과를 얻기 위한 팁과 질문 템플릿을 정리해 보았어요. 단순히 Pilot을 활용한 지표 조회 방법이 아니라, 게임 마케터가 AI를 데이터 분석 동반자로 활용할 수 있는 팁을 알려드리니 참고해 주세요.
**핵심 포인트**
- AI에게 질문할 때 **분석 기간, 이벤트명, 분석 의도, 결과물 포맷** 4가지 조건을 포함하면 되묻기 없이 원하는 답을 얻을 수 있어요.
- 대화 시작 시 **사전 맥락(Context) 블록**을 한 번 전달하면, 이후 짧은 질문으로도 우리 팀 기준에 맞는 분석을 받을 수 있어요.
- 데일리 점검, ROAS 딥다이브, 시장 비교, 코호트 LTV, 어트리뷰션 퍼널 진단까지 **5가지 실무 시나리오 템플릿**을 바로 활용할 수 있어요.
- 같은 AI 모델이라도 **질문의 구체성과 맥락 제공 여부**에 따라 결과 퀄리티가 크게 달라져요.
## **AI에게 질문할 때 반드시 들어가야 할 4가지 조건**
혹시 AI에게 데이터 분석을 맡길 때 “데이터 분석 해줘”라고 지시하기만 하시나요? 만약 그랬다면, 분명 AI가 질문의 의도를 파악하기 위해 여러분에게 되묻곤 했을 거예요. 그 과정에서 대화가 꼬이는 경우도 있었을 것이고요.
내가 던진 질문에 Pilot이 되묻지 않고 바로 **내 의도대로 분석 > 해석 > 액션**까지 이어지려면, 질문에 아래 4가지 조건이 포함되어야 합니다.
### **1\. 분석 기간을 언급해 주세요**
기간을 명시하지 않고 "구글의 ROAS 보여줘"라고 물으면 Pilot 입장에서는 유저가 원하는 기간이 어제인지, 지난주인지, 이번 달인지 알 수가 없어요. \*\*'2026년 3월 1일~31일'\*\*처럼 날짜를 지정하거나, \*\*'최근 2주'\*\*처럼 구체적인 버킷을 정해주세요.
### **2\. 활용하는 이벤트명이 무엇인지 설명해 주세요**
설치는 `Install`, 구매는 `Order Complete`와 같은 스탠다드 이벤트가 보편적으로 활용되지만, 게임 앱은 `stage_complete_1`, `join_channel`처럼 스탠다드 이벤트로 정의되지 않은 커스텀 이벤트를 쓰는 경우도 많아요. '튜토리얼 완료한 유저 몇 명이야?'보다 '`Complete Tutorial` 유저 수'라고 써주면 Pilot이 메트릭이 없다고 하거나 잘못된 이벤트명을 선택할 리스크를 줄여줘요..
### **3\. 분석 의도를 알려주세요**
'ROAS 보여줘'와 'ROAS가 떨어진 원인 분석해줘'는 성격이 전혀 다른 질문이에요. Pilot은 두 요청 모두에 답변할 수 있지만, 어느 쪽인지 모르면 보수적으로 숫자만 보여주는 쪽을 택하거나 불필요하게 많은 정보를 나열할 수 있어요. **보고 싶은 지표를 특정하고, 그 지표를 가지고 무엇을 하고 싶은지까지 같이 적어주시면 더 좋아요.**
### **4\. 원하는 결과물의 포맷을 알려주세요**
같은 데이터라도 일별 트렌드인지, 증감율 테이블인지, 코호트 분석인지에 따라 인사이트가 달라져요. '월별 코호트 테이블로 정리해줘', '상위 5개만 추려줘' 같은 형태 조건을 붙이면 후속 정리 작업도 크게 줄어들어요.
**모호하게 질문했을 때:**
- 기간을 명시하지 않아 Pilot이 대시보드 생성 이래로 지금까지의 데이터를 전부 가져와요.
- 질문 의도가 없으니 채널별로 그룹바이하는 등의 상세 분석도 해주지 않아요.
- 답변을 생성하기까지 파일럿이 계속 업데이트를 해요. _(기간을 불러옵니다, 범위를 넓혀봅니다, 더 넓혀봅니다, ...)_
[image2]
**조건을 정확히 입력했을 때**
- 기간, 이벤트, 활용할 그룹바이를 바로 찾아 원하는대로 정렬해줬어요.
- 모수가 적어 해석에 유의해야 할 값들은 이모지로 표시해줬어요.
- 추론 능력을 통해 예산을 증액/유지/감액할 채널을 찾아줬어요.
[image3][image4]
## **사전 맥락(Context) 하나로 답변 퀄리티 높이기**
위 4가지 조건을 잘 넣는 것만으로도 더 뾰족한 결과를 얻을 수 있지만, 단순히 답을 받는 수준을 넘어서 의사결정까지 이어지게 만드는 방법이 하나 더 있어요.
**바로 사전 맥락(Context)를 AI에게 학습시키는 거예요.** 다른 말로 **컨텍스트 엔지니어링(Context Engineering)** 이라고도 하죠. 대화를 시작할 때 우리 팀의 상황과 기준을 사전에 고정값처럼 주입하는 방법이에요. 마치 새로운 팀원에게 “우리는 이런 KPI를 보고, 이런 전환을 활용하고, 지금 이런 이슈가 있어”라고 배경을 공유하는 것과 비슷해요.. 같은 모델을 써도 결과가 달라지는 이유는 대부분 여기서 나와요.
Pilot도 마찬가지예요. 아래처럼 대화 시작 시 컨텍스트 블록을 한 번 전달해 두면, 같은 세션 안에서는 이후에는 짧게 질문해도 Pilot이 우리 팀의 기준을 반영한 후 데이터를 분석할 수 있어요.
### **사전 컨텍스트 블록 예시**
```json
[우리 게임의 배경]
- 장르: 수집형 RPG
- 핵심 KPI: D7 리텐션 ≥ 30%, tROAS 30일 기준 ≥ 120%
- 타겟 국가: 한국 > 일본 > 글로벌
- 주요 전환 이벤트명: Install, Sign-up, join_channel, Order Complete, level_10/20/30
- BM: IAP 100% (Order Completed로 측정함)
- 현재 상황: 일본 시장 소프트 론칭 2주차, 글로벌 대비 CPI가 1.5배 높아 효율 모니터링 중
- 분석 기본 기간: 보통 최근 7일 일별, 월간 트렌드는 30일
- 자주 쓰는 그룹핑: Channel > Campaign > Ad Group > Creative 혹은 Client IP Country
- 주요 코호트 기준: 설치 후 D0, D7, D14, D30, D60, D120, D180
```
이런 블록을 전달해 두면, '어제 성과 어때?'라고만 물어봐도 주요 매체의 핵심 KPI 기준으로 분석해 줄 수 있어요. 'ROAS는 어때?'라고 물었을 때도 단순히 수치를 나열하는 것이 아니라 KPI 대비 상태를 판단하고 원인, 액션까지 한 번에 분석한 결과를 제안받을 수 있죠.
## **데이터 분석 시나리오별 질문 템플릿**
지금까지 AI 질문에 포함하면 좋을 4가지 조건과 사전 맥락 제공 팁에 관해 알아봤는데요.이 두 가지를 결합해서, **게임 마케터가 실무에서 자주 마주하는 상황에 따라 어떻게 AI나 Pilot에게 질문하면 좋은지 템플릿을 정리해봤어요.**
만약 지금 바로 AI를 데이터 분석에 활용하고 싶은데 어떤 질문을 해야할 지 모르겠다면 한번 참고해 보세요.
### **시나리오 1: 데일리 캠페인 점검**
매일 오전 전일 대비 이상 징후를 확인하는 루틴용 질문이에요.
> 어제 vs 2일 전 기준으로 전체 매체의 Install 수, CPI, 구매 전환율, ROAS를 비교 정리해줘. 전일 대비 변동폭이 가장 큰 캠페인 Top 3를 뽑고, 각각에 대해 변동 원인 가설과 추천 액션을 함께 제안해줘.
비교 기준, 지표, 결과물 형태(Top 3, 가설과 액션)를 한 번에 전달하고 있기 때문에 Pilot이 역질문 없이 오버뷰부터 액션아이템까지 한 번에 답할 수 있어요.
### **시나리오 2: ROAS 딥다이브**
특정 매체의 ROAS가 떨어졌거나, 예상 외로 높아졌을 때 원인을 파고드는 질문이에요.
> 지난 30일 간 `google.adwords`의 캠페인별 ROAS(14일 기준)를 보여줘. 우리 KPI에 미달인 캠페인을 리스트업하고, 해당 캠페인들의 D7 ARPU, 구매 전환율, D7 리텐션을 같이 보여줘. 어떤 지표에서 효율이 빠지고 있는지 진단해 주면 좋겠어.
ROAS만 단순히 확인하는 것이 아니라 떨어진 캠페인의 보조 지표를 함께 요청해서, Pilot이 효율이 어디서 새고 있는지를 짚어낼 수 있어요.
### **시나리오 3: 신규 시장 론칭 성과 비교**
새로운 국가에 진입했을 때, 론칭 전후 성과를 기존 시장과 비교하는 질문이에요.
> 일본 소프트론칭 이후 2주(4/14/15)와 직전 2주(3/173/31)를 비교해서, 일본의 Install 수, CPI, D1/D7 리텐션, 누적 매출 추이를 정리해줘. 같은 기간 한국, 동남아의 동일 지표도 같이 보여줘서 시장 간 효율을 비교할 수 있게 해줘.
소프트론칭 전후라는 기간과 비교하고자 하는 국가 기준에 대해 알려줌으로써 Pilot이 단순히 수치만 나열하는 것이 아니라 시장 간의 상대적인 효율까지 코멘트해줄 수 있어요.
### **시나리오 4: 코호트 기반 LTV 분석**
단순히 기간 내 ROAS를 보는 것을 넘어, 특정 설치 코호트가 시간이 지날수록 얼마나 높은 LTV를 만들어내는지 트래킹할 때 활용하는 템플릿이에요. 채널별로 D7, D30, D60 ROAS를 나란히 놓으면 단기 효율이 좋아 보이는 채널과 장기 LTV가 높은 채널을 구분할 수 있어요.
> 2026년 1월 1일~3월 31일 사이에 설치한 유저의 구매이벤트 기준으로, D7 / D14 / D30 / D60 / D90 ROAS를 채널별로 정리해줘. 유료 채널(`channel type` = `integrated`)만 포함하고, 코호트 테이블 형태로 보여줘.
설치 코호트 기간을 고정하고 여러 ROAS 윈도우를 한 번에 요청하는 질문이기 때문에 단기와 장기 성과를 나란히 두고 비교할 수 있어 성과를 한 눈에 확인할 수 있어요.
### **시나리오 5: 어트리뷰션 퍼널 진단**
Install은 잡히는데 특정 캠페인이 기여를 받지 못하거나, 매체 리포트와 에어브릿지 수치가 맞지 않을 때와 같은 문제 상황에 Pilot을 활용할 수도 있어요.
> `google.adwords` 채널의 `A_캠페인명`과 `B_캠페인명` 두 캠페인의 기여 설치 이후의 전환에 대한 문제 상황을 분석해줘. 지금 Install은 기여가 잡히는데 구매 완료에 대한 기여가 안 잡혀. 기여가 어디서 끊기는지 확인하고, 트래킹 설정이나 어트리뷰션 윈도우 관점에서 원인 가설을 제시해줘.
나타난 데이터가 예상한 것과 다르다면 Pilot과 함께 가능한 원인을 파악하고, 해결방안을 찾는 등 트러블슈팅을 시도할 수 있어요.
## **마무리하며**
출시 초기에는 짧고 모호한 질문이 많았지만, 한 달 정도가 흐른 지금은 분석 의도나 조건을 포함한 질문이 눈에 띄게 늘었어요. 질문에 남겨진 여지만큼 모호함도 커지고, 질문자의 의도가 많이 담길수록 Pilot과의 대화 효율이 좋아진다는 것을 확인할 수 있었어요.
Pilot이 활용할 수 있는 데이터와 우리 팀의 맥락 사이에는 공백이 있어요. 그 공백을 채우는 것은 결국 질문하는 사람의 몫이에요. 이 가이드의 템플릿이 그 출발점이 되기를 바랄게요.
혹시 Airbridge Pilot과 MCP에 관해 더 궁금한 점이 있다면 언제든지 문의해 주세요.
---
## 크로스채널 어트리뷰션이란? 채널별 성과를 실제 전환 기여도로 해석하는 방법
URL: https://www.airbridge.io/ko/blog/cross-channel-attribution-guide-conversion-performance
Language: Korean
Published: 2026-04-17
Author: Team Airbridge
**매체 리포트상 ROAS는 개선됐는데, 실제 매출 기여도는 왜 다르게 보일까요?** 이 질문은 광고 채널이 늘어날수록 더 자주 등장해요.
이는 광고 플랫폼마다 전환 집계 기준과 [어트리뷰션(Attribution)](https://www.airbridge.io/ko/blog/marketing-measurement-1-attribution) 로직이 다르기 때문이에요. 이 차이로 인해 동일한 전환이 여러 채널에 중복 반영되거나, 특정 접점의 기여가 실제보다 크게 보일 수 있어요.
문제는 웹, 앱, 여러 디바이스에 걸친 고객 여정에서 더 뚜렷하게 나타나요. 사용자는 한 번의 클릭으로 바로 전환되지 않고, 여러 터치포인트를 거쳐 전환하는 경우가 많아요.
하지만 데이터가 채널별로 분리돼 있으면 전환 경로를 일관된 하나의 기준으로 측정하기 어렵고, 그 결과 마지막 접점 중심으로 성과를 해석하거나 보조 채널을 과소평가할 수 있어요.
이로 인해 먼저 채널 어트리뷰션과 크로스 채널 어트리뷰션의 차이부터 구분할 필요가 있어요. 두 어트리뷰션의 차이를 먼저 짚어보고, 크로스 채널 어트리뷰션의 장점을 살려 우리 기업에 맞게 활용하는 방법까지 알아볼게요.
**📌 핵심 포인트**
- 매체 안에서 보는 성과와, 비즈니스 관점에서 봐야 할 성과는 다를 수 있어요. 채널 어트리뷰션과 크로스 채널 어트리뷰션의 차이를 이해하고, 어떤 상황에서 각각 필요한지 살펴보세요.
- 성과를 제대로 보려면, 마지막 클릭이 아니라 전체 여정을 봐야 해요. 크로스 채널 어트리뷰션에 대해서 알면 여러 채널이 하나의 전환에 어떻게 관여하는지, 그리고 그 기준을 어떻게 정리해야 하는지 알 수 있어요.
## **채널 어트리뷰션과 크로스 채널 어트리뷰션은 무엇이 다를까요?**
채널 어트리뷰션과 크로스 채널 어트리뷰션은 비슷해 보이지만, 실제로는 측정 범위와 활용 목적이 달라요. **채널 어트리뷰션은 개별 광고 매체 안에서 성과를 해석하는 데 초점**을 맞춰요. 예를 들어 어떤 캠페인이 더 높은 전환율을 만들었는지, 더 효율적 특정 소재가 무엇인지 판단할 때 유용해요.
반면 **크로스 채널 어트리뷰션은 여러 채널을 함께 놓고 전환 기여도를 해석**해요. 검색광고, 디스플레이, 리타깃팅, CRM처럼 서로 다른 접점이 하나의 전환에 어떤 방식으로 관여했는지 보는 데 더 적합해요. 그래서 채널별 운영 최적화보다, 실제 예산 배분이나 성과 판단 기준을 정리할 때 더 중요한 역할을 해요.
두 방식의 차이는 아래처럼 정리할 수 있어요.
| 구분 | 채널 어트리뷰션 | 크로스 채널 어트리뷰션 |
| --- | --- | --- |
| 분석 범위 | 개별 매체 중심 | 여러 채널 통합 |
| 주요 목적 | 캠페인 운영 최적화 | 전환 기여도 비교 |
| 장점 | 소재, 타깃, 캠페인 효율 분석 | 고객 여정 기반 성과 해석 |
| 제약 | 채널 간 중복 기여를 보기 어려움 | 데이터 연결과 기준 정리가 중요함 |
| 적합한 질문 | 어떤 캠페인이 효율적인가? | 어떤 채널이 실제 전환에 기여했는가? |
채널 어트리뷰션과 크로스 채널 어트리뷰션은 서로 대체하는 개념은 아니에요. 채널 어트리뷰션은 매체 운영에 필요하고, 크로스 채널 어트리뷰션은 그 성과를 더 넓은 관점에서 해석하는 데 필요해요.
특히 여러 채널을 함께 운영하는 기업일수록 실제 예산 배분과 성과 판단에는 크로스 채널 어트리뷰션이 더 중요해지는데, **에어브릿지는 웹과 앱, 다양한 마케팅 채널 전반의 터치포인트를 통합해서 볼 수 있다는 면에서 강점**을 가져요.
## **크로스 채널 어트리뷰션으로 무엇이 달라질까요?**
크로스 채널 어트리뷰션을 적용하면, 채널별 숫자를 보는 방식보다 성과를 판단하는 기준이 먼저 달라져요.
- **마지막 클릭 중심 평가에서 벗어날 수 있어요.** 전환 직전 채널만 보는 대신, 전환 이전에 관여한 접점까지 함께 해석할 수 있어요.
- **보조 채널의 역할을 더 분명하게 볼 수 있어요.** 검색광고나 리타깃팅뿐 아니라 디스플레이, CRM, 콘텐츠 채널이 만든 기여도 함께 판단할 수 있어요.
- **채널별 리포트와 실제 성과 사이의 차이를 더 잘 설명할 수 있어요.** 같은 전환이 여러 채널에 반영되더라도, 어떤 접점이 실제 기여했는지 더 입체적으로 볼 수 있어요.
- **예산 배분 기준을 더 정교하게 잡을 수 있어요.** 단순 효율 비교를 넘어 어떤 채널을 확대하고, 어떤 채널을 보조 역할로 볼지 판단하기 쉬워져요.
이 변화는 채널 운영 보고보다 예산을 분배하거나 조정할 때 더 크게 체감돼요. 채널별 리포트만 볼 때는 성과가 좋아 보이는 채널에 예산이 계속 몰리기 쉬워요.
하지만 크로스 채널 어트리뷰션을 적용하면 **전환을 마무리한 채널과 전환 가능성을 높인 채널을 구분**해서 볼 수 있어요. 그래서 실제로는 어떤 채널을 확대하고, 어떤 채널은 보조 역할로 평가해야 하는지 더 분명하게 판단할 수 있어요.
## **크로스 채널 어트리뷰션, 3단계로 실무에 활용하세요**
실무에서는 구체적으로 어떻게 활용할 수 있을까요? 아래 가이드를 참고하여 마케팅 성과를 해석할 수 있어요.
### **1\. 중복 성과를 줄여 실제 CPA와 ROAS를 재해석하세요**
채널별 리포트를 그대로 합산하면 실제보다 성과가 더 좋아 보일 수 있어요. 광고 플랫폼마다 전환 집계 기준과 어트리뷰션 윈도우가 다르기 때문이에요. 그래서 첫 단계에서는 같은 전환이 여러 채널에 반영되고 있지는 않은지부터 점검해야 해요.
이때 중요한 건 **우리 서비스의 전환 주기와 구매 특성에 맞는 기준을 적용**하는 일이에요. [**에어브릿지는 어트리뷰션 윈도우와 룩백 윈도우를 설정**할 수 있고, **채널별 어트리뷰션 규칙도 조정**](https://help.airbridge.io/ko/guides/attribution-scenario)할 수 있어요. 이런 기능을 활용하면 매체가 제공하는 기본 수치를 그대로 받아들이는 대신, 실제 비즈니스에 더 가까운 기준으로 CPA와 ROAS를 다시 해석할 수 있어요.
_\*어트리뷰션 윈도우: 타깃 이벤트 이후부터 어떤 시점에 있는 전환까지 타깃 이벤트와 관련 있다고 판단할지 결정하는 기간_
_\*룩백 윈도우: 타깃 이벤트 전부터 어떤 시점에 있는 터치포인트까지 타깃 이벤트와 관련 있다고 판단할지 결정하는 기간_
### **2\. 전환을 만든 채널과 보조 채널의 역할을 구분하세요**
실제 전환은 하나의 채널만으로 만들어지지 않는 경우가 많기 때문에, 이 단계에서는 웹과 앱에 걸친 전환 경로를 함께 보는 게 중요해요. **에어브릿지는 웹·앱 어트리뷰션과 웹투앱 어트리뷰션을 지원**하고, [**웹 SDK 기반 측정**](https://help.airbridge.io/ko/guides/web-to-app-tracking)과 [UTM 파싱도 제공](https://help.airbridge.io/ko/guides/migrating-tracking-links)해요. 이로 인해 광고 클릭 이후 웹 방문, 앱 설치, 앱 내 전환까지 이어지는 흐름을 더 일관되게 해석할 수 있어요. 이런 방식은 전환을 닫힌 채널과 전환 가능성이 높은 채널을 구분하는 데 도움이 돼요.
_\*UTM 파싱: 에어브릿지가 아닌 다른 솔루션에서 사용한 링크에 포함된 UTM 파라미터를 에어브릿지 트래킹 링크의 파라미터로 자동 매핑하는 기능_
### **3\. 성과 데이터를 실제 타깃팅 운영까지 연결하세요**
여러 채널의 기여도를 다시 해석했다면, 그 결과를 실제 운영 판단으로 연결해야 해요. 어떤 채널을 확대할지, 어떤 채널은 보조 역할로 볼지, 어떤 세그먼트를 다시 타깃팅할지 결정하는 단계예요.
이때는 집계 리포트만 보는 것보다 로 데이터까지 활용할 수 있는지가 중요해요. [**에어브릿지는 앱 로 데이터(App Raw Data)와 웹 로 데이터(Web Raw Data)를 내보낼 수 있고**](https://help.airbridge.io/ko/guides/checking-raw-data), [리얼타임 로그에서 SDK나 이벤트 API로 수집된 이벤트를 바로 점검](https://help.airbridge.io/ko/guides/tracking-events-with-real-time-logs)할 수 있어요. 그래서 마케팅팀은 성과를 빠르게 확인할 수 있고, 데이터팀은 내부 BI나 데이터 웨어하우스에서 숫자를 다시 검증할 수 있어요.
여기서 끝나지 않고, 해석한 결과를 실제 타깃팅 운영까지 연결하는 것도 가능해요. [에어브릿지는 오디언스 기능을 활용해 주요 매체로 오디언스를 연동](https://help.airbridge.io/ko/guides/audience-manager)할 수 있어요. 이를 통해 **리타겟팅, 디타겟팅, 유사 타겟 확장처럼 더 정교한 운영**이 가능해져요.
이렇게 성과를 분석하는 것에 머무르지 않고, 예산 조정과 오디언스 운영까지 이어질 때 크로스 채널 어트리뷰션의 활용 가치는 더욱 커져요.
## **우리 기업에 맞는 크로스 채널 최적화 시나리오**
크로스 채널 어트리뷰션은 업종과 서비스 특성에 따라 활용 방식이 달라져요. 우리 서비스의 전환 경로를 먼저 파악하고, 시작 접점과 최종 전환 접점에 맞게 활용해 보세요.
### **1\. 금융권이라면 웹 유입부터 앱 전환까지 함께 보세요**
금융 서비스는 웹에서 상품을 탐색한 뒤 앱 설치, 상담 신청, 카드 발급이나 계좌 개설로 이어지는 경우가 일반적이에요. 이때 마지막 전환 채널만 보면 실제 기여도를 놓치기 쉬워요. 웹과 앱을 하나의 리포트에서 함께 분석하면, 어떤 채널이 초기 관심을 만들고 실제 전환을 마무리했는지 확인할 수 있어요.
특히 에어브릿지는 핀테크와 금융의 프라이버시 우선 마케팅을 바탕으로 [**첫 클릭부터 첫 대출 실행까지 모바일, 웹, 데스크톱의 크로스 플랫폼 유저 여정을 파악**](https://www.airbridge.io/ko/solutions/finance)해요. **200개 이상의 지표를 캠페인, 크리에이티브, 코호트, 네이버·카카오 키워드별로 실시간 분석**하는 것이 강점이에요.
**이렇게 적용해 보세요**
1. 웹 유입 채널과 앱 전환 채널을 함께 확인해요.
2. 상담 신청, 카드 발급, 계좌 개설처럼 핵심 전환 지표를 기준으로 채널 기여도를 다시 확인해요.
3. 마지막 클릭 채널과 초기 관심을 만든 채널의 역할을 구분해 예산을 조정해요.
### **2\. 대기업은 채널별 리포트가 많지만 예산 판단 기준은 분산돼 있을 때 활용해요**
대기업은 광고 채널, 앱, 웹, CRM이 각각 다른 팀과 도구로 운영되는 경우가 많아요. 이때는 보고용 수치는 많아도 실제 기여도를 한 번에 설명하기 어려워요. 크로스 채널 어트리뷰션을 활용하면 채널별 리포트를 단순 비교하는 대신, 여러 접점의 역할을 함께 보고 예산 배분 기준을 정확하게 정리할 수 있어요.
에어브릿지는 타 MMP 대비 [**가장 많은 국내 고객사를 보유하고 있으며, 대기업과 국내 채널 마케팅에 특화**](https://www.airbridge.io/ko/blog/korea-no-1-mmp)되어 있어요. 월간 활성 사용자수(MAU)가 많고, 운영해야 하는 광고 채널이 다수라면 채널별 성과를 하나의 기준으로 해석하는 것이 중요하죠. 크로스 채널 어트리뷰션으로 전사 관점에서 성과를 판단하는 기준을 마련해 보세요.
**이렇게 적용해 보세요**
1. 광고, 웹, 앱, CRM 리포트가 어떻게 나뉘어 있는지 먼저 정리해요.
2. 여러 접점의 기여도를 같은 기준으로 비교해요.
3. 전사 차원의 예산 판단 기준으로 연결해요.
### **3\. 웹투앱 중심 서비스는 초기 유입 채널의 가치를 다시 보고 싶을 때 사용하세요**
웹에서 유입된 사용자가 앱을 설치하고 나중에 전환하는 서비스라면, 앱 설치나 마지막 클릭만 기준으로 성과를 해석하기 쉬워요.
하지만 이런 방식으로는 초기 유입 채널의 역할을 충분히 보기 어려워요. 하지만 웹투앱 유저 여정을 트래킹하면, 어떤 채널이 실제 전환까지 이어지는 흐름을 만들었는지 더 구체적으로 파악할 수 있어요.
에어브릿지에서는 [**웹과 앱을 하나의 대시보드로 통합 특정이 가능**](https://www.airbridge.io/ko/product/web-app-attribution)해요. [네이버·카카오 등 네이티브 연동으로 한국 시장에 최적화된 크로스 플랫폼 어트리뷰션](https://www.airbridge.io/ko/product/web-app-attribution)을 통해 다양한 접점의 기여도를 확인할 수 있어요. 초기 유입 채널의 가치와 최종 전환 채널의 역할을 더 분명하게 나눠서 보고 싶다면, 참고하세요.
**이렇게 적용해 보세요**
1. 웹 유입부터 앱 설치까지 이어지는 경로를 추적해요.
2. 초기 유입 채널과 최종 전환 채널을 구분해요.
3. 저평가된 유입 채널의 가치를 다시 반영해 캠페인을 조정해요.
## **채널별 성과 해석, 이제는 공통된 기준으로 판단하세요**
오늘도 채널별 성과 해석 기준이 제각각이라, 어떤 숫자를 믿고 판단해야 할지 고민하셨나요? 크로스 채널 어트리뷰션은 이렇게 흩어진 성과 데이터를 하나의 기준으로 해석할 수 있게 돕는 기술이에요.
지금 에어브릿지에서 웹과 앱, 다양한 마케팅 채널에 걸친 고객 여정을 연결해 실제 전환 기여도를 더 정확하게 파악하는 제품을 경험해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 금융사 마케터 필독, MMP 도입 전 확인해야 할 5가지
URL: https://www.airbridge.io/ko/blog/finance-enterprise-mmp-selection-guide
Language: Korean
Published: 2026-04-17
Author: Team Airbridge
**왜 지금 금융사의 MMP 선택 기준이 더 중요해졌을까요?**
2026년 금융권 마케팅은 이전과 다른 기준으로 성과를 봐야 해요. 과거에는 앱 설치 수나 유입 규모 같은 단기 지표만으로도 마케팅 성과를 설명할 수 있었지만, 이제는 리텐션, LTV, ROAS처럼 실제 매출과 장기 기여를 함께 고려하는 방향으로 바뀌고 있어요.
특히 금융사는 예산 집행 이후 성과를 더욱 엄격하게 설명해야 하고, 예치·거래·반복 이용 같은 고가치 행동까지 연결해 해석할 필요가 있죠.
동시에 개인정보 보호, 내부 보안 기준, 클라우드 이용 가이드라인 등 일반 업종보다 더 높은 규제 기준을 충족해야 해요. 결국 금융사와 핀테크 기업에 필요한 MMP는 **규제 대응을 전제로 신뢰할 수 있는 성과 기준을 제공하는 솔루션**이어야 해요.
오늘은 실제로 금융사가 MMP 도입을 검토할 때 살펴봐야 할 조건과 기준을 상세하게 정리했어요.
**📌 핵심 포인트**
- 좋은 MMP는 매체별 성과 중복을 제거해 ‘진짜 기여분’을 제공해요. 신뢰할 수 있는 마케팅 성과 기준을 세울 수 있는지가 MMP를 선택하는 첫 번째 기준이 되어야 해요.
- iOS와 식별 제한 환경에서도 끊김 없는 성과 측정이 가능하면, 유실될 뻔만 마케팅 기여도를 정확하게 파악할 수 있어요.
- 에어브릿지는 업계 최초 AWS 서울 리전으로 보안 심의와 운영 효율을 동시에 잡았어요. 금융권의 보안 규제에 최적화된 국내 서버 인프라를 경험해 보세요.
### **1\. 여러 매체 성과를 하나로 통합하는지 확인하세요**
MMP를 도입했음에도 불구하고, 광고 매체마다 수치가 제각각이어서 당황한 적이 있으실 거예요. 이러한 이유는 매체별로 상이한 성과 산정 로직 때문이에요. 메타, 구글 등 주요 광고 매체는 **자사 플랫폼에 유리한 기여 인정 기준(Attribution Window)을 적용**해요.
예를 들어, 한 사용자가 여러 매체의 광고를 거쳐 최종적으로 금융 상품에 가입했을 때 각 매체는 서로 자신이 결정적으로 기여했다고 측정하며 성과를 중복 집계해요.
이러한 데이터의 파편화를 방치할 경우, 실무자는 다음과 같은 두 가지 리스크에 직면하죠.
- **내부 보고 신뢰성 붕괴:** 각 광고 매체 리포트 상의 가입자 합계가 실제 내부 DB의 유입 수치보다 1.3~1.5배 이상 비대해지는 현상이 발생해요. 이는 경영진이나 감사팀에게 마케팅 성과 자체가 부풀려졌다는 불신을 받을 수 있어요.
- **비효율적 자원 배분의 고착화:** 실제 가입에 결정적이었던 ‘라스트 터치(Last Touch)’ 매체를 구별하지 못하면, 성과가 과다 측정된 매체에 예산이 지속적으로 낭비되는 구조가 될 수 있어요.
2026년의 금융권 및 엔터프라이즈 MMP는 단순한 트래킹 도구를 넘어, **모든 매체에 동일하고 객관적인 측정 잣대를 적용해 중복을 제거**해야 해요. 비즈니스 데이터가 단 하나의 정보를 기준으로 의사결정할 수 있는 구조가 세워지면, 예산이 비즈니스 성장에 기여한 순수 기여분을 객관적으로 증명할 수 있어요.
### **2\. 신뢰할 수 있는 데이터 측정 체계가 있는지 검토하세요**
금융권 마케팅 담당자가 마주하는 큰 장벽 중 하나는 개인정보 보호 정책의 강화예요. 과거에는 사용자의 기기 식별값(ADID/IDFA)을 통해 광고 클릭부터 앱 설치까지 경로를 1:1로 매칭하는 것이 가능했어요.
하지만 **애플의 ATT(앱 추적 투명성) 정책과 구글의 프라이버시 샌드 박스가 전면 도입**되면서, 식별 데이터에만 의존하는 방식으로는 전체 성과의 절반도 채 측정하기 어려워졌죠.
이에 대응하여 최신 MMP는 개인을 식별하지 않고도 성과를 입증할 수 있는 고도화된 측정 기술을 선보이고 있어요.
- **개인 식별 없는 성과 측정(Privacy-Preserving Attribution):** 특정 개인을 추적하지 않고도, 광고 집단과 전환 집단 사이의 통계적 연관성을 분석하여 기여도를 산출해요. 이는 규제를 완벽히 준수하면서도 마케팅 예산의 효용성을 증명할 수 있는 핵심 기술이에요.
- **머신러닝 기반의 성과 추론(Probabilistic Modeling):** 직접적인 매칭 데이터가 유실된 구간에도 OS 버전, 디바이스 모델, 네트워크 상태 등 비식별 정보를 조합하여 높은 확률로 성과를 예측해요. 이를 통해 측정되지 않는 깜깜이 영역을 최소화하고 담당자가 놓칠 뻔한 성과를 데이터로 되찾아줘요.
결국 2026년의 MMP는 매체 간 중복을 제거하는 역할과 더불어, 복잡한 규제 속에도 기술력을 통해 마케팅 기여도를 확인하는 역할을 수행해야 해요. 다음으로는 기업이 MMP 도입 시 어떤 요소를 중점적으로 검토해야 할지 살펴볼게요.
## **비용부터 보안까지, 금융사 MMP 도입 시 확인해야 할 5가지**
도입 전 비용 구조부터 주요 보안 심의까지, MMP 도입 시 알아야 할 5가지 기준을 정리했어요. 더불어 실무자가 도입 단계에서 직접 점검해 볼 수 있는 대표 질문까지 확인해 보세요.
### **1\. 기본 요금과 총 소유 비용(TCO)의 차이**
MMP 도입 시 흔히 설치당 과금 단가를 우선 비교하지만, **실제 운영 단계에서는 ‘기능별 추가 비용(Add-on)’이 예산의 변수**요.
대다수 글로벌 MMP는 기본적인 트래킹 왜에 딥링크 생성, 광고 사기(Fraud) 방지, 오디언스 연동, 로 데이터(Raw Data) 접근성 등 성과 측정과 운영에 필수적인 기능을 유료 옵션으로 분류하기 때문이죠.
비즈니스가 성장하여 캠페인 규모와 트래픽이 늘어날수록 이러한 옵션 비용은 증가할 수밖에 없어요. 이로 인해 매년 예산 수립과 집행 소명이 필요한 금융사에는 예측 불가능한 재무적 리스크가 될 수 있어요. 따라서 초기 계약 단가보다는 **우리 팀이 실무에서 상시 사용하는 기능이 포함된 총 소유 비용(TCO) 관점에서 MMP 검토를 선행**하세요.
#### **이렇게 질문해 보세요**
✅ 월간 앱 설치 10만, 웹 방문 50만인 금융사가 딥링크, 광고 사기 방지, 오디언스 기능까지 사용한다면 도입할 MMP의 예상 TCO 범위는 어느 정도일까?
### **2\. 웹과 앱을 아우르 네이티브 통합 역량**
최근 앱 비즈니스는 사용자 여정이 웹가 앱을 수시로 오가며 최종 가입이나 상품 구매를 완료하는 구조로 변하고 있어요. 예를 들면 핀테크나 금융사는 사용자가 웹에서 상품을 탐색하고, 앱에서 최종 가입이나 예치를 완료해요. 이커머스나 여행, 미디어 산업군 역시 **모바일, PC, 콘솔 플랫폼 전반에서 검색과 콘텐츠 소비를 통해 앱 내 결제로 이어지는 ‘크스플랫폼’ 여정이 일반적**이에요.
이처럼 사용자 접점이 파편화된 환경에서는 앱 어트리뷰션만으로 **광고 매체의 성과와 실제 내부 DB의 유입 경로가 일치하는지 확인**이 어려워요. 글로벌 솔루션 중 상당수는 웹 측정을 보조적인 기능으로 취급하거나, 별도의 SDK 연동을 요구하기 때문이죠. 이로 인해 데이터의 파편화가 생길 수 있어요.
특히 웹 트래픽 비중이 유의미한 비즈니스 환경에서는 웹과 앱의 데이터를 통합하지 못하면, 마케팅 기여도를 왜곡할 가능성이 커요. 웹에서의 유입 경로가 앱 내 전환으로 이어지는 과정을 단일 ID로 연결하지 못하면, 어떤 매체가 실질적인 전환에 기여했는지 정확히 판단할 수 없기 때문이죠. 2026년의 MMP는 이러한 플랫폼의 경계 없이 사용자 여정을 하나로 통합하여 객관적인 성과를 증명할 수 있어야 해요.
#### **이렇게 질문해 보세요**
✅ 우리 비즈니스에서 웹 트래픽이 전체 유입에서 차지하는 비율은 얼마일까?
### **3\. 로 데이터(Raw Data) 보존 기관과 활용 자율성**
데이터 분석 전담 조직이 있거나 자체 DW(데이터 웨어하우스)를 운영하는 기업에 **로 데이터(Raw Data) 접근성은 데이터 자산화의 핵심**이에요. 가공되지 않은 원천 데이터는 내부 CRM 시스템과 연동하여 고객 한 명의 생애 가치(LTV)를 분석하고 고도화된 타팅 전략을 세우는 밑바탕이 되기 때문이죠.
그러나 일부 솔루션은 로 데이터 보존 기간을 60~90일 정도로 제한하거나, 대규모 데이터를 외부로 전송할 때마다 추가 과금(Data Tax) 혹은 쿼리 제한을 두어 데이터 활용의 병목을 만들기도 해요.
데이터 추출 방식에 제약이 있거나 활용할수록 비용이 발생하는 구조는 장기적인 관점에서 분석 자율성을 제한할 수 있어요. **데이터에 대한 통제권을 충분히 보장하며 자율적인 데이터 접근이 가능한 MMP 사용 정책**이 있는지 확인해 보세요.
#### **이렇게 질문해 보세요**
✅ 우리 데이터 팀이 로 데이터 자체를 DW에서 적재해서 분석하고 있을까?
✅ 현재 MMP는 데이터의 보존 기간이나 실시간 추출 비용에 대한 기술적·비용적 제약이 있을까?
### **4\. 운영 리스크를 최소화하는 실시간 대응 체계**
제품에 기술적 결함이나 데이터 송수신 오류가 발생하면 즉각적인 해결이 필요해요. MMP 사용 시 긴급한 이슈 상황에서 해외 본사의 답변을 기다리기 위해 영어로 티켓을 생성하고 수십 시간을 대기해야 한다면, 이는 마케팅 기회비용의 손실로 이어질 수 있어요.
특히 국내 금융 보안 가이드라인이나 네이버, 카카오 등 국내 매체 환경에 특화된 연동 환경은 글로벌 표준 매뉴얼만으로 대응하기 어려운 경우가 많죠.
단순히 기술 문서를 전달하는 수준을 넘어, 국내 시장의 맥락을 이해하고 전용 채널을 통해 **실시간으로 밀착 지원**할 수 있는지 확인해 보세요. 현지 CS 전담 조직의 유무는 안정적인 운영을 위한 필수 조건이에요.
#### **이렇게 질문해 보세요**
✅ 장애 발생 시 모국어로 실시간 소통할 수 있는 채널과 현지 전담 조직이 있을까?
### **5\. 보안 심의 패스트 트랙을 위한 ‘서울 리전’ 보유**
금융권 보안 심의의 가장 큰 허들은 **고객 식별 정보가 포함된 데이터의 국외 이전 문제**예요. 금융보안원 가이드라인에 따라 국내 고객의 민감 데이터는 가급적 국내 법망의 통제가 가능한 곳에서 관리되어야 하기 때문이죠.
대부분의 글로벌 MMP는 서버가 해외에 위치해 있어, 도입 시마다 까다로운 소명 절차와 데이터 유출 리스크에 대한 책임을 감수해야 해요.
따라서 [**우리 데이터에 대한 확실한 통제권을 증명할 수 있는 국내 서버(서울 리전) 보유**](https://www.airbridge.io/ko/blog/seoul-region-launch-data-compliance) 여부는 보안팀의 승인을 이끌어내기 위한 최우선 조건이죠. 서버가 국내에 위치하면 불필요한 국외 이전 심의 부담을 줄이고, 규제 대응 과정도 보다 효율적으로 진행할 수 있어요.
이는 마케터가 보안팀과의 협의에 드는 시간을 줄이고, 캠페인 실행 시점을 더 안정적으로 계획하는 데에도 도움이 돼요. **에어브릿지는 업계 최초로 AWS 서울 리전 서비스를 공식 론칭**해 이러한 금융권의 요구에 대응하고 있어요.
### **MMP 도입 시 주요 비교 항목 한눈에 보기**
지금까지 짚어본 기준을 바탕으로 보면, 금융권 MMP를 검토할 때 어떤 항목에서 차이가 나는지 더 분명하게 확인할 수 있어요. 아래 표에서 에어브릿지와 글로벌 MMP의 주요 차이를 한눈에 살펴보세요.
| 항목 | 세부 검토 사항 | Airbridge | 타사 |
| --- | --- | --- | --- |
| 비용 구조 | 기본 포함 기능 vs Add-on | 주요 기능 기본 포함 | 주요 기능 Add-on 구조 공식 문의 필요 |
| 측정 범위 | 웹 및 앱 통합 지원 여부 | 웹 및 앱 네이티브 통합 | 단순 정적 웹투앱 추적만 가능 |
| Raw Data | 보존 기간· 추출 방식 | 데이터 보존 기간 제한 없음 클라우드 스토리지 실시간 전송 | 추가 결제 필요 또는 제한적인 다운로드만 가능 |
| 고객 지원 | 한국어 전담 지원 여부 | 한국어 전담 지원 | 공식 문의 필요 |
| 데이터 주권 | 국내 서버 옵션 | 국내 서버 선택 가능 | 공식 문의 필요 |
## **체크리스트로 보는 우리 조직과 캠페인에 맞는 MMP는?**
금융권에서 MMP를 검토할 때는 하나의 기준만으로 판단하기보다, 실제 운영에 관여하는 조직별로 확인해야 할 사항을 나눠서 보는 것이 좋아요. 외부 마케팅 데이터와 내부 비즈니스 데이터를 어떻게 연결하고 운영할지 결정하는 일이기 때문에, 마케팅팀뿐 아니라 IT·보안팀, 데이터 조직, 전략기획 조직이 함께 검토해야 해요.
성공적인 도입을 위해 팀 단위에서는 다음 3가지 핵심 영역을 점검할 수 있어요.
### **1\. IT/보안팀이 검토해야 할 컴플라이언스 기준**
가장 먼저 넘어야 할 산은 보안 심의와 데이터 통제 구조요. 국내 금융권 규제와 우리 조직의 가이드라인을 기준으로 다음 항목을 만족하는지 점검해 보세요.
| 점검 항목 | 세부 확인 사항 |
| --- | --- |
| 데이터 물리적 위치 | 개인정보 및 민감 데이터가 처리되는 서버 리전이 국내(서울)에 위치하나요? |
| 심의 절차의 효율성 | 데이터 국외 이전 소명 없이 보안 심의를 빠르게 통과할 수 있는 구조인가요? |
| 세부 권한 관리 | 조직 내 직무(오너, 매니저, 대행사 등)에 따른 역할 기반 접근 제어(RBAC)이 가능한가요? |
| 보안 인증 보유 | ISMS-P 등 국내외 공인된 정보보호 관리 체계 인증을 공식적으로 보유했나요? |
### **2\. 마케팅팀이 확인해야 할 데이터 기술력과 무결성**
| 점검 항목 | 세부 확인 사항 |
| --- | --- |
| 어트리뷰션 무결성 | 앱 스위칭이나 망분리 환경에서도 유입 경로 유실 없이 정확한 측정이 가능한가요? |
| 딥링크 안정성 | OS나 브라우저 제약 없이 의도한 앱 페이지로 정확히 랜딩되는 기술력을 갖췄나요? |
| 데이터 자율성 | 로우 데이터 추출 및 외부 시스템 전송에 제약이나 추가 비용이 없나요? |
| 가짜성과 방어 | 국내 매체 특성을 반영하여 가짜 성과(Fraud)를 실시간으로 필터링할 수 있나요? |
### **3\. 전략기획팀이 고려해야 할 운영 효율과 확장성**
| 점검 항목 | 세부 확인 사항 |
| --- | --- |
| AX 활용 역량 | 마케터의 분석 시간을 획기적으로 줄여줄 실무형 AI 에이전트 기능을 제공하나요? |
| 기술 및 CS 지원 체계 | 이슈 발생 시 해외 본사의 답변을 기다리는 대신, 국내 전담 엔지니어와 소통 가능한가요? |
| 비용 예측 가능성 | 캠페인 규모가 확장되어도 비용이 폭증하지 않는 합리적인 요금 구조인가요? |
| 매체 파트너 | 구글, 메타 등 글로벌 주요 매체와 공식 파트너십을 맺어 데이터 정확도를 보장하나요? |
## **우리 조직에 맞는 금융권 특화 MMP를 선택하세요**
결국 중요한 것은 기능이 얼마나 많은지가 아니라, 금융사의 운영 환경에 실제로 맞는 MMP를 선택하는 일이에요. 에어브릿지는 국내 금융 환경에서 요구되는 조건을 반영해 보다 안정적으로 활용할 수 있는 MMP 환경을 제공해요.
성과를 더 정확하게 해석하고, 더 설명 가능하게 운영하는 것, 금융권 마케팅에서 점점 더 중요해지는 과제를 위해, 에어브릿지를 우리 조직의 MMP 선택지로 검토해 보세요.
**_Written by 프리랜서 에디터 강혜라_**
---
## 게임사가 만우절 마케팅에 매년 진심일 수밖에 없는 이유
URL: https://www.airbridge.io/ko/blog/april-fools-game-marketing-trends
Language: Korean
Published: 2026-04-14
Author: Team Airbridge
# **게임사가 만우절 마케팅에 매년 진심일 수밖에 없는 이유**
만우절은 단순히 장난을 주고받는 날을 넘어, 사람들이 먼저 기대하는 시즌 이벤트에 가까워졌어요. 특히 게임업계는 이런 흐름을 누구보다 적극적으로 활용해 왔죠. 만우절이 되면 유저들은 새로운 이벤트나 이색 콘텐츠를 기대하고, 게임사 역시 이를 하나의 마케팅이자 커뮤니케이션 기회로 활용해요.
2026년에도 게임사들은 만우절을 각기 다른 방식으로 활용했어요. 올해 화제 됐던 게임업계 사례를 중심으로 만우절 마케팅이 실제로 어떻게 작동하는지 살펴보고, 마케터가 참고할 수 있는 포인트까지 정리해 볼게요.
**📌 핵심 포인트**
- 만우절 마케팅의 성패는 이용자 행동을 유도하는 방식에 좌우돼요. 직접 플레이하게 만들지, 공식 발표처럼 기대를 자극할지, 팬덤이 알아볼 포인트를 건드릴지에 따라 화제의 결이 달라졌어요.
- 잘 만든 만우절 콘텐츠는 하루짜리 농담으로 끝나지 않아요. 커뮤니티 반응으로 확산되거나, 팬덤 자산으로 축적되기도 하죠. 실제 게임·체험 아이디어로 이어지면 브랜드에 대한 기억을 더 오래 남기기도 해요.
## **게임 속 ‘만우절 모드’로 직접 플레이해요**
게임업계에서 만우절 콘텐츠는 단순한 시즌성 공지보다 **인게임 체험형 이벤트**로 작동할 때 큰 반응을 얻어요. 이용자가 직접 접속해 평소와 다른 규칙이나 연출, 분위기를 경험하는 순간 콘텐츠의 주목도와 확산 가능성이 함께 높아지기 때문이에요. **배틀그라운드**와 **오버워치**가 대표적인 사례예요.
### **1\. 익숙한 게임을 낯설게 바꿔 흥행, 배틀그라운드와 오버워치**
[]
이미지 출처: [크래프톤](https://www.krafton.com/news/press/%EB%B0%B0%ED%8B%80%EA%B7%B8%EB%9D%BC%EC%9A%B4%EB%93%9C-%EB%A7%8C%EC%9A%B0%EC%A0%88-%EA%B8%B0%EB%85%90-%ED%8A%B9%EB%B3%84-%EB%AA%A8%EB%93%9C-%EC%88%A8%EB%B0%94%EA%BC%AD%EC%A7%88/)
**배틀그라운드**는 2026년 3월 31일부터 4월 8일까지 [**만우절 한정 모드로 ‘Prop Hunt’**](https://pubg.com/ko/news/9889)를 선보였어요. 게임 속에서 유저는 사물로 변신해 숨거나, 반대로 이를 찾아내는 역할을 맡게 됐죠. 배틀로얄 장르의 긴장감과 전투 중심 문법에서 잠시 벗어나, ‘숨바꼭질’이라는 직관적인 규칙으로 유저는 게임을 새롭게 경험하는 거예요.
공식 공지에 따르면 이 모드는 12인 구성, 라운드형 진행, 위장 전환과 ‘Fever Time’ 같은 장치를 통해 단순한 이벤트가 아니라 **반복 플레이가 가능한 별도 아케이드 모드로 설계**됐어요. **기존 게임의 핵심 자산은 유지하면서도 전혀 다른 분위기**를 만들어냈다는 점에서, 만우절 이벤트를 체험형 콘텐츠로 확장한 사례라고 볼 수 있죠.
[]
출처: [PUBG: BATTLEGROUNDS](https://www.youtube.com/watch?v=HOgeNXFJFiY) 캡처
특히 숨바꼭질 모드의 게임을 **홈쇼핑 형식의 티저 영상으로 소개**해 이벤트 공지 자체를 화제성 있는 콘텐츠로 확장한 점이 눈에 띄어요. 즉, 새로운 모드를 추가한 데서 그치지 않고, 이를 어떻게 유저에게 알릴지 함께 설계한 사례죠.
여기에 더해 소셜 해시태그와 e스포츠 이벤트 매치까지 연결하면서, 이벤트를 게임 안팎에서 반복 소비할 수 있게 만들었어요. 크래프톤은 만우절 콘텐츠가 유저의 경험을 확장하고 여러 플랫폼에서 이야기되도록 설계했다는 점에서 마케팅 기획력을 보여주는 사례예요.
[][]
이미지 출처: [블리자드 패치노트](https://overwatch.blizzard.com/en-us/news/patch-notes/)
**오버워치** 역시 비슷한 방식으로 만우절 콘텐츠를 운영했어요. 올해 4월 1일 공개된 [**‘Underwatch’ 패치 노트**](https://overwatch.blizzard.com/ko-kr/news/patch-notes/)가 대표적이에요.
오버워치는 출시 이후 매년 만우절마다 영웅의 능력과 게임 규칙을 평소와 다르게 구성해 플레이 경험 자체로 관심을 모았어요. 특히 **영웅의 능력 변화와 게임 규칙을 만우절**에 스타일에 맞게 조정하고 이를 패치노트에 녹여내서 새로운 콘텐츠에 대한 기대감을 만들었죠.
패치노트가 이용자가 변경 사항을 접하는 공식문서이자, 앞으로 어떤 플레이를 하게 될지 미리 가늠하게 되는 채널인 점을 활용한 거예요.
사용자는 장난스러운 설정에 먼저 관심을 갖고, 이후 게임 안에서 평소와 다른 영웅 능력과 플레이 환경을 직접 경험하게 되죠.
이런 이벤트는 유튜버와 스트리머를 통해 롱폼과 숏폼 콘텐츠로 빠르게 확산되며, 그 과정에서 신규 이용자의 관심을 끌고, 쉬고 있던 이용자의 재참여를 유도하는 계기로도 이어져요.
참고로 2025년 오버워치 개발자 글에서 이전 만우절 모드 ‘Totally Normalwatch’가 시즌 15에서 플레이 시간 기준 주요 모드 가운데 하나였다고 밝혔는데요. 이는 만우절 이벤트를 실제 이용과 화제성으로 연결하는 체험형 마케팅 자산으로 꾸준히 활용해 온 사례라고 볼 수 있어요.
## **‘공식 발표’처럼 말할수록 더 바이럴 돼요**
만우절 콘텐츠는 설정 자체보다, 그 설정을 **얼마나 진지하고 완성도 높게 구현했는지**에 따라 반응이 달라져요. 가볍게 소비되는 장난보다, 공식이 예상 밖의 설정을 설득력 있게 구성할 때 이용자는 한 번 더 보게 되죠. **니케**와 **니어** 시리즈는 이런 방식으로 공지 자체를 화제가 되는 콘텐츠로 만든 사례예요.
### **1\. 설정과 장난의 경계를 흐린 니케**
[]
이미지 출처: [니케 유튜브](https://www.youtube.com/watch?v=EOTZHsHYLaQ)
**레벨인피니트**는 모바일 PRG ‘승리의 여신: 니케’에서 4월 1일 만우절 이벤트 [‘FOOL METAL PANIC!’](https://www.youtube.com/watch?v=EOTZHsHYLaQ)을 공개했어요. 이번 이벤트는 평소에 작전 지원 역할에 가까운 캐릭터 시프티를, 갑자기 ‘메카미 시프티’라는 초절정 미소녀 로봇으로 전면에 내세워 관심을 끌었어요.
여기에 세계 평화를 위한 모험, SD 캐릭터 연출, 전용 로비 테마, 라이브 월페이퍼, 한정 상점까지 더하면서 실제 플레이와 보상이 있는 이벤트로 구성했어요. 즉, 익숙한 캐릭터를 과장된 설정으로 비틀되, 콘텐츠 완성도는 가볍게 가져가지 않은 사례예요.
[]
이미지 출처: [니케 유튜브](https://www.youtube.com/watch?v=EOTZHsHYLaQ)
이 방식이 효과적이었던 이유는 **공식 발표처럼 보이나, 의도적으로 부풀린 설정으로 인해 이용자가 한 번 더 확인하도록** **만들었기 때문**이에요. 니케의 만우절은 매년 4개월 이상 사전 준비하는 것으로 알려진 만큼 완성도가 높아요. 이번에도 역대 만우절 전용 캐릭터를 다시 등장시키고 유형석 디렉터가 영상에 ‘발표회’ 콘셉트로 기획 영상에 직접 등장해 팬덤이 반응할 지점을 넓혔죠.
실제 커뮤니티와 기사 반응에서도 “뭐가 진짜고 뭐가 농담인지 헷갈릴 정도”라는 평가가 나왔어요. 이는 기존의 캐릭터를 색다르게 변주한 설정에 실제 참여를 유도하는 보상까지 더해지면서, 이번 공지는 단순한 이벤트 안내를 넘어 그 자체로 화제를 만드는 콘텐츠가 됐어요.
### **2\. ‘신작 발표’처럼 보인 니어: 코즈믹 호러**
**스퀘어 에닉스**는 만우절 당일 니어 공식 계정을 통해 [**‘NieR: COSMIC HORROR’**](https://x.com/NieR_JPN/status/2038994733087830493) 티저를 공개했어요. “허무 속에서 태어나는 것은 희망일까, 아니면 구원이라는 이름의 절망일까”라는 문구와 함께 영상이 공개되자, 팬들 사이에서는 실제 신작 발표인지 만우절 기획인지 판단하기 어렵다는 의견이 나왔어요.
[]
이미지 출처: [레딧 캡처](https://www.reddit.com/r/PS5/comments/1s8zi0q/square_enix_teases_new_nier_project_but_it_could/?tl=ko)
결과적으로 이 프로젝트는 실제 출시작이 아니라 만우절용 기획이었어요. 다만 반향이 컸던 이유는 티저에 실제 신작 발표의 형식과 톤을 거의 그대로 가져왔기 때문이에요.
게다가 올해 2월 공개 영상 말미에 “NieR: Automata to be continued”라는 문구를 남기며 시리즈 후속 전개 가능성을 이미 암시한 바 있어 팬들 입장에서는 이번 티저를 완전히 농담으로만 보기 어려운 맥락도 있었죠.
이후 니어 [공식 계정은 같은 날 “오늘 0시에 발표한 ‘NieR: COSMIC HORROR’는 만우절 기획”이라고 밝히며 “매우 큰 반향”이 있었다고 직접 언급](https://x.com/NieR_JPN/status/2039168554541679088?utm_source=chatgpt.com)했어요.
이처럼 **실제 발표에 가까운 형식과 연출로 구성**되어 있으면 유저의 추측과 반응이 커뮤니티를 통해 퍼지며 콘텐츠화되기도 해요.
## **추억의 밈 꺼내거나, 세계관 뒤집어 반응 얻어요**
오래 회자되는 만우절 콘텐츠에는 팬들에게 익숙한 요소가 많죠. **잘 알려진 캐릭터, 오래 쌓인 밈, 세계관**에 대한 이해가 더해지면 이벤트는 일종의 팬서비스처럼 받아들여져요. 이러한 유형의 만우절 이벤트는 신규 유입보다 기존 팬덤에 더 큰 힘을 발휘해요.
### **1\. 익숙한 밈과 감성을 다시 꺼낸 마비노기**
[]
츨처: [넥슨](https://mabinogi.nexon.com/page/news/notice_view.asp?id=4893451)
**마비노기**는 2026년 4월 1일부터 4월 9일까지 만우절 이벤트 [‘속닥속닥 만우절 파티!’](https://mabinogi.nexon.com/page/news/notice_view.asp?id=4893451)를 진행했어요. 대표 캐릭터를 등록한 이용자가 NPC를 통해 일일 퀘스트를 수행하고, ‘속삭이는 무지개 토끼 귀’, ‘소곤소곤 양손검’ 같은 보상을 받는 방식이었죠.
이벤트 규모는 크지 않지만, **마비노기가 본래 강점을 보여온 생활형 감성과 귀엽고 엉뚱한 연출**을 그대로 살렸어요. 팬들이 이미 좋아하던 분위기 안에서 만우절다운 장난을 더한 덕분에, 이용자 입장에서도 별다른 설명 없이 마비노기다운 감각으로 자연스럽게 받아들일 수 있었어요.
[]
출처: [에피드게임즈 유튜브](https://www.youtube.com/watch?v=Vk9qUpz0tF4)
### **2\. 팬덤 위한 이벤트도 해요**
서브컬처 게임들은 이런 방식이 더 두드러져요. **트릭컬**은 2026년 만우절에 게임 안팎의 분위기를 한꺼번에 바꾸는 방식으로 팬덤 반응을 끌어냈어요. 공식 채널에서는 에르핀의 그림일기 세계관을 내세운 만우절 기획을 소개했고, 게임 안에서는 타이틀 화면과 로비를 낙서풍으로 바꾸는 등 **평소 팬들이 좋아하던 장난기 있는 연출을 전면**에 드러냈어요.
여기에 더해 게임 개발사인 에피드게임즈는 [지정 기간 굿즈 구매 유저 중 118명을 추첨해 대표가 직접 배송에 나서는 이벤트까지 진행](https://www.youtube.com/watch?v=Vk9qUpz0tF4)했어요. 만우절 전날인 3월 31일 ‘초속 118KG’라는 제목의 영상을 게재하고, 영상에는 한정현 에피드게임즈 대표가 직접 출연해 택배를 차에 싣는 모습이 담겼죠. 대표가 직접 선물을 전달하는 이벤트로 이어졌다는 점에서, 만우절 기획을 유저가 실제로 체감할 수 있는 팬서비스로 확장한 셈이에요.
[]
출처: [프로젝트문 유튜브 캡처](https://www.youtube.com/watch?v=HTRQgFYCXHY&t=1s)
**림버스 컴퍼니**는 만우절에 ‘림버스 유치원’을 배경으로 한 특별 스토리를 공개했어요. 평소에는 무겁고 비극적인 서사, 불안정한 캐릭터 관계, 강한 정서적 긴장감이 특징인 게임인 만큼, 익숙한 캐릭터들이 전혀 다른 분위기에서 등장한다는 점만으로도 팬들의 반응을 끌어내기 충분했죠.
실제로 지난해에는 업데이트 준비로 만우절 콘텐츠를 선보이지 못한다며 사과를 전했을 정도로, 이 이벤트를 기다리는 팬층의 기대도 큰 편이에요.
림버스 컴퍼니는 이번 만우절에 설정을 덧붙이기보다, 팬들이 익숙하게 알고 있는 게임의 분위기를 반전시키는 방식으로 화제를 만든 거예요.
## **정식 게임 출시와 오프라인 체험으로 확장되기도 해요**
만우절 마케팅은 대개 짧게 소비되고 끝나지만, **일부 게임사는 아이디어를 확장하거나 실제 게임으로 출시될 기회를 잡기도 해요.** 이 경우 만우절은 브랜드가 새로운 아이디어를 시험하는 출발점이 되기도 해요. **포켓몬 GO**와 **로스트아크**는 그런 가능성을 보여준 사례예요.
### **1\. 실제 게임이 되어버린 포켓몬 GO**
[]
출처: [구글맵스 유튜브](https://www.youtube.com/watch?v=4YMD6xELI_k)
**포켓몬 GO**는 만우절 장난이 실제 게임으로 발전한 . 시작은 2014년 4월 1일, 구글이 만우절 기획으로 선보인 [**‘Google Maps: Pokémon Challenge’**](https://www.youtube.com/watch?v=4YMD6xELI_k)였어요.
당시 이용자는 구글맵 앱 안에서 세계 곳곳에 숨겨진 포켓몬을 찾고 잡는 방식으로 참여했고, 구글은 이를 마치 실제 채용 프로젝트처럼 소개하며 포켓몬 마스터를 찾는다는 설정까지 붙였어요.
특히 현실 공간을 돌아다니며 포켓몬을 포획한다는 아이디어는 유저에게 어린 시절 포켓몬을 직접 잡아보고 싶었던 감각을 떠올리게 하며 향수를 자극했어요.
실제로 이 만우절 [영상은 공개 일주일 만에 1,350만 회 이상 조회됐고, 공유 수도 48만 건을 넘기며 크게 화제](https://martech.org/google-april-fools-pokemon-challenge-video-2nd-shared-video)를 모았어요. 이후 나이언틱과 포켓폰 컴퍼니는 협업을 거쳐 2016년, 실제 위치 기반 모바일 게임 ‘포켓몬 GO’를 출시됐어요. **만우절 이벤트가 실제 서비스 기획**으로 이어진 거죠. 이 사례는 브랜드가 이후 확장할 수 있는 제품 아이디어의 출발점이 되었다는 점에서 지금도 자주 언급돼요.
### **2\. 체험형 콘텐츠로 확장된 로스트아크**
[]
이미지 출처: [스마일게이트 뉴스룸](https://newsroom.smilegate.com/bbs/board.php?bo_table=news&wr_id=2216)
**로스트아크**는 만우절 장난이 별도 게임 출시로 이어진 사례는 아니지만, 일회성 이벤트에 머물지 않고 오프라인 체험과 팬덤 콘텐츠로 확장된 사례예요.
올해는 공식 유튜브에 [**‘LOST ARK FULL 3D EDITION**](https://www.youtube.com/watch?v=ZsvdbhpZh7I)**’** 영상을 공개해, 원래 쿼터뷰 시점의 게임을 자유 시점 액션 게임처럼 연출했어요. 여기에 게임 안에서는 '예스이호테 일보’를 함께 선보이며, 영상 밖에서도 만우절 분위기를 이어갔죠.
이런 흐름은 이전 만우절 콘텐츠와도 맞닿아 있어요. 로스트아크는 2022년 만우절에 [**‘LOST ARK VR’**](https://www.youtube.com/watch?v=j0_teOmkztg) 콘셉트를 공개해 화제를 모았고, 이후 2023년 **지스타에서는 로스트아크 모바일을 활용한 VR 체험형 부스**를 선보였어요. 만우절에 제시한 상상력을 실제 오프라인 경험으로 확장한 셈이에요. 로스트아크 사례는 만우절 콘텐츠가 당일 화제에 그치지 않고, 이후 브랜드가 이어갈 수 있는 체험 아이디어로 발전할 수 있다는 점을 보여줘요.
## **만우절 콘텐츠, 참여 요인과 확산 방식을 참고해 보세요**
게임업계 사례를 보면 만우절 마케팅은 화제 확보를 넘어 플레이를 유도하거나, 발표 형식으로 주목도를 높이거나, 팬덤이 반응할 만한 요소를 활용하는 식으로 다양한 반응을 만들어요. 콘텐츠의 재미도 있지만, 가장 중요한 건 **이용자의 행동으로 이어지게 만드는 기획**이라고 할 수 있죠.
짧은 시즌 이벤트 안에서도 이렇게 다양한 접근이 가능하다는 점은 꽤 흥미로워요. 다음 만우절에는 우리 브랜드에는 어떤 접근이 잘 맞을지, 이번 사례를 바탕으로 한번 떠올려봐도 좋지 않을까요?
**_Written by 프리랜서 에디터 강혜라_**
**참고자료)**
[Director’s Take: Season 16 on the Horizon, Blizzard Entertainment, 2025.4.10](https://overwatch.blizzard.com/ko-kr/news/24187726/director-s-take-season-16-on-the-horizon/?utm_source=chatgpt.com)
[아케이드: 숨바꼭질, PUBG: BATTLEGROUNDS, 2026.3.31](https://pubg.com/ko/news/9889)
[Overwatch Retail Patch Notes, Blizzard, April 1, 2026](https://overwatch.blizzard.com/ko-kr/news/patch-notes/)
[PUBG: Battlegrounds - Official Prop Hunt Limited Time Mode Launch Trailer, PUBG: BATTLEGROUDS Youtube, 2026.3.30](https://www.youtube.com/watch?v=HOgeNXFJFiY)
[진짜로 왔다, 《승리의 여신 니케 : 월드》 발표회 | 승리의 여신: 니케, 승리의 여신: 니케 - 신큐 업데이트 유튜브, 2026.4.1](https://www.youtube.com/watch?v=EOTZHsHYLaQ)
[【NieR 新プロジェクト 制作血定】, NieR\_JPN X, 2026.4.1](https://x.com/NieR_JPN/status/2038994733087830493)
[스퀘어 에닉스가 새로운 니어 "프로젝트"를 살짝 보여줬는데, 이거 혹시 만우절 장난일 수도 있어., reddit, 2026.4.1](https://www.reddit.com/r/PS5/comments/1s8zi0q/square_enix_teases_new_nier_project_but_it_could/?solution=a2b2e6a015588948a2b2e6a015588948&js_challenge=1&token=bbbe4bf1c9a2b5160829c4be34da58611a91b91c35cb59491b15dbb92810ffb9&tl=ko)
[いつもNieRシリーズを支えてくださりありがとうございます, Nier\_JPN, 2026.4.1](https://x.com/NieR_JPN/status/2039168554541679088?utm_source=chatgpt.com)
[2026 만우절 이벤트 속닥속닥 만우절 파티!, 넥슨, 2026.4.1](https://mabinogi.nexon.com/page/news/notice_view.asp?id=4893451)
[\[특별영상\] 초속 118KG, EPID Games Youtube, 2026.3.31](https://www.youtube.com/watch?v=Vk9qUpz0tF4)
[\[ Limbus Company \] Official Promotion Video, ProjectMoon Officia Youtubel, 2026.4.7](https://www.youtube.com/watch?v=HTRQgFYCXHY&t=1s)
[Google Maps: Pokémon Challenge, Google Maps Youtube, 2014.4.1](https://www.youtube.com/watch?v=4YMD6xELI_k)
[April Fools Day “Google Maps: Pokemon Challenge” Video Goes Viral, Amy Gesenhues, MARTECH, 2014.4.8](https://martech.org/google-april-fools-pokemon-challenge-video-2nd-shared-video/)
[스마일게이트, 만우절 맞이 주요 IP 이색 이벤트 진행, 스마일게이트 뉴스룸, 2026.4.1](https://newsroom.smilegate.com/bbs/board.php?bo_table=news&wr_id=2216)
[\[로스트아크\] 4K | LOST ARK, FULL 3D EDITION, 로스트아크 LOST ARK Youtube, 2026.4.1](https://www.youtube.com/watch?v=ZsvdbhpZh7I)
[\[로스트아크\] LOST ARK VR, 로스아크 LOST ARK Youtube, 2022.4.1](https://www.youtube.com/watch?v=j0_teOmkztg)
---
## AI에이전트 시대, 마케터의 다음 업무방식은?
URL: https://www.airbridge.io/ko/blog/airbridge-ai-po-interview
Language: Korean
Published: 2026-03-27
Author: Team Airbridge
> **_“퍼포먼스 마케터의 역할이 근본적으로 바뀔 거예요. 고통스러운 대시보드 보기는 AI에게 맡기고, 마케터는 기획과 의사결정에 집중하는 방향으로 말이죠.”_**
퍼포먼스 마케터의 오전 10시는 대개 비슷하게 흘러갑니다. 수십 개의 매체 대시보드를 오가며 파편화된 데이터를 엑셀에 옮기고, 트래킹 링크를 검증하는 반복성 운영 업무로 하루를 시작하죠. 광고 매체는 늘어나고 데이터는 복잡해지는데, 정작 마케터의 손발이 되어줄 도구는 여전히 수동적인 관리 방식에 머물러 있어요.
에어브릿지는 이 격차를 해소하기 위해 ‘**Airbridge Pilot**’과 ‘**Airbridge MCP**’를 출시했습니다. 올해 1분기에 선보인 두 기능은 **마케터가 일일이** **데이터를 찾아 헤매는 수고를 대체하는 AI 에이전트**예요. 실제 성과로 이어지는 유의미한 분석과 인사이트를 실시간으로 추출해 사용자 앞에 즉시 내놓는 것이 특징이죠. 이번 제품 출시를 이끈 AB180 PO(Product Owner) 이승헌 님을 만나, AI 에이전트가 바꿀 고객의 일상을 짚어봤습니다.
[이승헌]
## **1\. ‘어떤 비즈니스가 살아 남을까?’ 고민이 이끈 선택**
### **Q. 제품에는 결국 만든 사람의 관점과 경험이 녹아들기 마련이죠. 에어브릿지 PO로 합류하시기 전까지 어떤 커리어를 거쳐오셨나요?**
커리어 시작은 LG디스플레이였어요. 2017년부터 2019년까지 모바일 상품 기획팀에서 자동차와 핸드폰에 들어가는 디스플레이를 기획했죠. 특히 자동차 분야는 5년 정도 뒤 미래에 출시될 디스플레이를 미리 그리는 일이라 긴 호흡의 기획력이 필요했어요.
하지만 하드웨어 중심의 기획 업무만으론 다가올 비즈니스의 변화를 담아내기에 한계가 느껴졌어요. 더 역동적인 환경에서 개인의 성장 가능성을 도모하고 싶어 IT 스타트업으로 눈을 돌렸고, 이후 핀테크와 에듀테크 기업에서 PO 직무로 6년 가까이 경력을 쌓았어요. 그 여정 끝에 지금의 회사인 AB180에 합류했죠.
### **Q. 대기업과 스타트업을 두루 거치셨는데요. AB180을 선택한 이유는 무엇이었나요?**
AI가 아무리 커져도 망하지 않을 회사가 필요했어요(웃음). 콘텐츠도, 기술도 AI에 의해서 무너질 수 있는 해자\*라고 생각했거든요. 단순히 기능 위주의 테크 앱만 잘 만드는 곳도 마찬가지고요. 하지만 ‘데이터’를 쥐고 있는 회사는 비즈니스 관점에서 승산이 있을 것 같았어요.
_\*해자: 경쟁사가 쉽게 넘볼 수 없는 독점적 경쟁 우위._
### **Q. 그 ‘해자’를 데이터로 생각하신 건가요?**
데이터를 포함해서 파트너십이 가장 큰 해자라고 봤어요. AB180는 자체 개발한 글로벌 MMP 에어브릿지(Airbridge)를 통해 [구글, 메타, 애플, 틱톡 등과 공식 인증 파트너십을 맺은 전 세계 7개 기업 중 하나](https://www.dt.co.kr/article/12027854?ref=naver)에요. 아시아에선 유일하죠.
이 파트너십이 중요한 이유는 **정보의 속도와 정확성** 때문이에요. **공식 파트너만이 대형 매체의 원본 광고 데이터를 가장 빠르고 오차 없이 받아올 수 있거든요.**
특히 최근 애플의 ATT(앱 추적 투명성)\* 정책처럼 개인정보 보호가 강화되면서, 마케터는 예전만큼 정확한 데이터를 얻기가 어려워졌어요. 수집할 수 있는 정보가 제한되니 데이터의 구멍이 생기기 시작한 거죠.
이런 상황에서 에어브릿지는 공식 파트너만이 제공받는 퍼스트 파티 데이터(First-party Data)를 직접 수급해요. 제3자를 거치지 않은 이 데이터는 오차를 최소화한, 가장 신뢰할 수 있는 ‘판단의 기준점(Source of Truth)’이 되는 거예요.
사실 MMP의 핵심 역량은 여기서부터인데요. 구글, 메타, 틱톡 등 광고 매체가 자기 성과라고 주장하며 숫자가 제각각일 때, 에어브릿지는 수급된 원본 데이터를 바탕으로 **중복을 제거하고 실제 기여도를 명확히 정리**해줘요.
이렇게 정확한 원본 데이터를 바탕으로 마케터는 숫자를 믿고 의사 결정을 내려도 된다는 확신을 얻을 수 있어요. 결국 MMP의 본질은 데이터에 있고, 이 **독보적 데이터 접근**이 가능한 파트너십이 에어브릿지만의 핵심 경쟁력이 돼요.
> \*_ATT(App Tracking Transparency): 앱 추적 투명성. iOS 14.5부터 도입된 개인정보 보호 정책으로, 앱이 광고 목적으로 사용자를 트래킹하거나 광고 식별자에게 접근하기 전 반드시 사용자의 명시적 동의를 얻게 하는 기능._
[이승헌]
## **2\. 마케터에게, “찾지 말고 주문하세요”**
### **Q. 데이터 수집에서 차별성을 확보한 거군요. Pilot과 MCP는 시장의 어떤 요구로 시작됐나요?**
두 에이전트 기능 중 하나인 ‘Pilot’은 회사 내부의 페인 포인트(Pain Point)를 해결하는 과정에서 시작됐어요. 에어브릿지 고객사가 늘고 매출이 전년 대비 40% 이상 가파르게 성장하면서, CS 팀으로 인입되는 문의의 난이도와 빈도가 급격히 높아졌거든요.
에어브릿지 같은 기술 기반 제품은 정확한 가이드가 필수예요. 이에 프로덕트 팀은 CS 팀의 대응 효율을 높이고자 ‘사내 질문 봇’을 먼저 구축해 질의응답에 활용했어요. 실제로 내부 테스트 과정에서 이 기능의 효용성을 확인했죠. 우리 고객도 이 편리함을 직접 누리면 좋겠다는 아이디어가 나왔고요. 이미 내부 검증을 거친 만큼, 제품을 고도화하고 고객용 서비스로 업그레이드해 탄생한 제품이 바로 Pilot이에요.
**_“Pilot을 고객에게 내놓기 전, 팀의 문제를 먼저 해결하는 제품이었던 거죠.”_**
‘MCP’는 백화점의 퍼스널 쇼퍼에 비유하면 쉬울 것 같아요. 보통 백화점에 가게 되면 내가 찾는 브랜드가 어디 있는지 파악하고 직접 매장을 돌아다니며 물건을 탐색하죠. ‘내가 원하는 곳을 향해 이동하는’ 개념이에요.
그런데 MCP는 VIP를 다루는 퍼스널 쇼퍼처럼 ‘원하는 걸 가져다주는’ 개념이에요. MCP의 근간이 되는 AI 에이전트는 사용자를 대신해 자율적으로 작업을 수행하거든요. 이렇게 되면 고객은 백화점 몇 층에 어떤 물건이 있는지 알 필요도 없고, 직접 이동할 필요도 없죠. 원하는 바만 주문하면 되니까요. 비유하면, 퍼스널 쇼퍼인 MCP가 ‘알잘딱깔센’으로 사용자가 원하는 데이터, 지식, 인사이트를 앞에 가져다주는 거예요.
- Airbridge Pilot: Airbridge 플랫폼 안에서 동작하는 AI 에이전트
- Airbridge MCP: 고객이 사용하는 LLM인 클로드, GPT 등에서 Airbridge 데이터, API, 도구와 쉽게 연동되도록 돕는 서비스
### **Q. 에이전트는 어떻게 마케터가 원하는 정보를 찾아내고 전달하나요?**
에이전트가 질문받으면 에어브릿지 안에서 **어떤 데이터를 봐야 할지 스스로 판단하고, 직접 분석한 후 이를 바탕으로 결과와 해결책까지 사용자에게 내놓아요.** 이게 기본적인 구조예요.
예를 들면 퍼포먼스 마케터가 이렇게 질문할 수 있겠죠. ‘에어브릿지 앱의 최근 일주일 광고 유입 현황을 채널별로 정리해서 알려주고, 주요 인사이트를 도출해줘.’
그럼 에이전트가 오늘 데이터뿐만 아니라 전주 대비 성과를 통해 고객 증감을 살펴볼 수 있고요. 좋은 성과를 보이는 채널이나 예산 재분배가 필요한 곳을 짚어내기도 하죠.
예산 재분배가 필요하다면 구체적인 시나리오나 대안을 말해주기도 하고요. 데이터 분석과 인사이트를 간단한 질문만으로 해결할 수 있는 거예요.
### **Q. 고객의 질문이 모호할 수도 있는데요.**
맞아요. 하지만 에이전트는 고객이 모호한 질문을 던지면 역질문을 던져요. 사용자가 궁금한 것이 단순히 ROAS인지, 아니면 캠페인 활성화 전략과 같은 복잡한 작업인지 말이죠. 지표도 채널별인지, 소재별인지 역질문하여 사용자의 진짜 질문을 파악해요.
이를 위해 8,000여 개의 사용자 질문을 분석해 300~400개의 구조화된 유형으로 추려냈어요. 질문의 패턴은 대부분 이 범위 안으로 수렴하거든요. 에이전트는 이 규칙 기반(Rule-based)의 로직을 바탕으로 사용자의 키워드를 분석해, 개별 데이터가 반영된 정확한 답변을 내놓습니다.
답변을 하는 범위도 에이전트가 데이터를 잘못 조회하지 않도록 도메인 가드레일을 설정해 뒀고요. 때로는 드롭다운 형태의 UI를 제공해 최적의 답변 리스트를 사용자가 선택하도록 돕기도 해요. 일반적인 UX로는 처리하기 어려운, 에이전트만이 가능한 정교한 기능이기도 합니다. 또한 지정된 도메인 안에서만 움직이기 때문에 답변의 신뢰도는 더욱 높아져요.
> **_“구조화된 지식 안에서만 움직이도록, 명확한 가드레일을 정해놓는 게 핵심이에요.”_**
### **Q. 가드레일이 일종의 필터가 되는 거군요.**
그렇죠. 저희가 가장 공들여 컨트롤하는 영역이기도 해요. 시스템 프롬프트와 구조화된 지식을 에이전트에 내재화하고, 마케팅 도메인 지식을 추가로 학습시켜 전문성은 높이고 오답의 가능성은 낮추는 거죠. 설계 단계에서 정교한 가이드가 들어갈수록 할루시네이션은 자연스럽게 줄어들어요.
%%CTA|데모 신청하기|/ko/book-a-demo|마케터의 5시간을 아껴주는 Airbridge AI, 직접 경험해 보세요.%%
## **3\. AX 잘하는 기업이 AI 에이전트도 제대로 만든다**
### **Q. 1분기 만에 두 기능을 선보이셨죠. 빠른 실행이 가능했던 팀의 비결은 뭘까요?**
에어브릿지의 AI 제품 라인업을 담당하는 **‘에이전트 플랫폼 스쿼드(Agent Platform Squad Team)’ 구조**에 답이 있다고 생각해요.
저희 팀은 에이전트 엔지니어 김준환 님과 최수범 님, UX 디자이너 이은지 님, 프론트엔드 엔지니어 진준기 님 등 각 분야의 전문가들이 모여 **기획부터 제작, 수정까지 한 호흡**으로 움직이고 있어요.
제품을 만들다 보면 기획된 그대로 가는 게 생각보다 어려워요. 이를 해결하기 위해 기획한 사람이 제작, 수정까지 할 수 있는 환경을 조성했어요.
그 노력 중 하나는 프론트엔드 디자이너가 코드를 직접 만지는 일이에요.
보통 프론트엔드 개발자가 디자인을 구현하다 보면 라운드값이나 컴포넌트 반영 등 미세한 지점에서 기획과 차이가 생기곤 해요. UX 디자이너는 QA 단계에서 이를 발견하지만, 매번 개발자에게 수정을 요청하고 확인하는 과정에서 상당한 커뮤니케이션 비용과 번거로움이 발생합니다.
저희는 이 비효율을 없애기 위해 디자이너가 직접 ‘스타일 코드’\*를 만져요. 프론트 엔지니어가 디자이너에게 GitHub 활용법과 Claude를 이용한 코딩법을 교육하고요. 디자이너의 작업이 기존 코드를 망가뜨리지 않도록 환경도 구축합니다. 개발의 영역을 이해하고 스타일을 결정하는 디자이너가 직접 코드를 만지니 수정 속도가 빨라질 수밖에 없어요.
이건 저희 기업 내부에서 AX를 하려는 노력이기도 해요. 기능을 구축하고 검증하는 과정 전반에 AI 사용을 체화하는 거죠.
> \*_스타일 코드: 디자인 스타일 코드 : 색상 및 폰트, 레이아웃 및 정렬, UI 버튼 등의 UI 스타일 설정하는 코드_
[이승헌]
### **Q. 개발 중에 가장 치열하게 고민했던 부분도 궁금해요.**
PO 입장에선 전체적인 제품 기획과 함께 토큰 비용을 어떻게 관리할지가 가장 중요했어요. 성능과 비용의 균형을 맞추면서도 지속 가능한 효율이 필요하니까요.
답변마다 다르겠지만 사용자의 모든 질문에 직접 사람이 답하는 것보다 에이전트가 답변하는 게 빠를 때가 많아요. UX 측면에서도 메일이나 전화 연결보다 대기 시간도 짧고요. 한 명의 상담원이 한 시간을 할애하는 동안 다른 고객들이 대기해야 하는 한계도 극복할 수 있죠.
저희는 에이전트 답변의 질과 속도를 모두 잡기 위해 아키텍처 최적화에 공을 많이 들였어요. 특히 아키텍처 담당 엔지니어가 큰 힘을 보태고 있는데요. 사용자가 자유롭게 에이전트와 질문을 주고받을 수 있되, 오용이나 남용으로 토큰이 낭비되지 않도록 지속적으로 모니터링하고 있습니다.
약간의 홍보를 보태자면, 이전 버전의 기능인 ‘[Ask Airbridge](https://help.airbridge.io/ko/guides/ask-airbridge)’는 계정별 일일 질문 횟수가 50회로 제한되어 있었지만, **Pilot과 MCP는 질문 횟수 제한이 없어요.** 에어브릿지는 신규 기능을 오픈해도 추가 과금이 발생하는 구조가 아니니, 마케터를 포함한 고객들이 기능을 마음껏 경험해 보시면 좋겠어요.
### **Q. AX가 가속화될수록 보안과 권한 관리가 중요할 것 같아요.**
네, 특히 신경 쓰는 부분이죠. 권한 관리는 ‘역할 기반 접근 제어’ 모델을 채택해 조직 내 권한을 관리하고 있어요. 데이터 접근은 오너, 매니저, 에이전시 등 역할에 따라 세분화되며, 앱 사용자 내에서도 권한이 더욱 세분화 되어 쪼개지는 형태예요. 오직 허가된 계정만이 데이터에 접근할 수 있어요. 특히 데이터 유출 방지를 위해 LLM 단계에서부터 암호화된 데이터 처리 환경을 구축해 보안성을 높이고요.
에어브릿지는 금융, 핀테크, 공공기관 등 규제가 민감한 산업군으로 시장 진입 기반을 이미 확보했고요. 보안과 규제, 개인 정보 보호 요구가 높은 대형 엔터프라이즈 기업도 믿고 사용할 만큼 보안 역량이 뛰어나요. 이번에 출시한 기능에도 이 점이 반영되어 있어요.
## **4\. 마케터의 하루는 달라져야 한다**
### **Q. 기술적 토대가 탄탄하네요. 고객들이 체감하게 될 가장 큰 변화는 무엇일까요?**
데이터를 찾는 수고에서 벗어나 전략을 결정하는 일에만 온전히 집중하게 되는 것이죠. 많은 마케터가 YoY(전년대비), MoM(전월 대비) 같은 기초 데이터를 추출하고 정리하는 반복 업무에 치여요. 이걸 다시 데일리 보고서, 월별 보고서에 녹여 내면서 하루에 보고성 업무에만 4-5시간을 소비하게 되죠.
에이전트가 반복성 업무를 줄여주면 마케터는 채널마다 어떤 소재를 태울지, 시즌별 예산은 얼마로 조정할지 실제 운영에 대해 고민할 시간을 벌 수 있어요.
특히 지표별 데이터 해석에 어려움을 겪던 팀이라도 에이전트를 통해 ‘ROAS가 개선되었으니, 예산을 재분배하라’는 수준의 구체적인 가이던스를 즉시 받을 수 있고요.
### **Q. 이 에이전트를 더 업그레이드할 계획도 있으신가요?**
물론이죠. 이제 시작이에요. 현재는 리포트 조회나 리스트 추출 같은 ‘Read(읽기)’ 기능을 중심으로 개발되었지만, 앞으로는 트래킹 링크를 대량 생성하거나, OG Tag를 수정하는 등 **에이전트가 직접 액션을 수행하는 ‘Write(쓰기)’ 단계**로 진화할 예정이에요. 다음 분기에는 이 생성 작업을 계획하고, 정기적으로 실행할 수 있는 기능을 탑재하는 게 목표고요.
> _“제품 안에서 고객은 ‘무엇을 해야 할지’ 점점 더 명확히 알게 될 거고요. 에이전트와 함께 의사 결정의 방향을 찾게 될 거예요.”_
[이승헌]
## **5\. 기술적 허들 없이 비즈니스 본질에만 몰두할 미래를 위해**
### **Q. 2026년, 에어브릿지가 AI를 통해 이루고자 하는 넥스트 스텝은 무엇인가요?**
현재 전사가 ‘모든 제품의 AI 에이전트화’라는 기조 아래 그 어느 때보다 바쁘게 움직이고 있어요. 최종 목적지는 ‘Zero Friction Conversion’에 도달하는 것입니다.
사용자가 제품에 유입되어 기능을 탐색하고 결제에 이르기까지 느끼는 모든 심리적·기술적 저항을 제거하는 거죠. AI 에이전트가 도구 사용의 장벽을 낮춰주면, 고객은 자연스럽게 더 나은 인사이트를 얻게 돼요. 모든 전환 과정에서 저항이 사라질수록 고객의 비즈니스 성과는 극대화될 수 있어요.
글로벌 시장에서의 포부도 있어요. 엔터프라이즈급 기업뿐만 아니라 소규모 스타트업이나 아주 작은 팀이라도 **에어브릿지에 가입하는 즉시 전문가 수준의 광고 분석을 시작**할 수 있는 환경을 만드는 거예요. 사실 MMP는 SDK 설치부터 설정, 정교한 데이터 매칭까지 고도의 기술적 이해도가 요구되는 고관여 서비스예요.
저희는 AI 에이전트를 통해 복잡한 기술적 단계를 더욱 직관적으로 풀어내려고 해요. 전문가의 상시적인 가이드가 없어도 누구나 스스로 최적의 세팅을 완료하고 데이터 인사이트를 도출하는 ‘완전한 셀프 서브(Self-Serve)’ 환경을 구축하는 거죠. 기술적 이해나 난도에 구애받지 않고 전 세계 모든 규모의 고객이 비즈니스 본질에만 집중해 성과를 극대화할 수 있도록 돕는 것이 저희의 지향점이에요.
### **Q. 앞으로의 행보가 기대되는데요. 마지막으로 에어브릿지의 신규 제품을 만나보게 될 사용자에게 한 마디 부탁드려요.**
AI 서비스를 구축하며 **기술이 해결하는 비즈니스 문제**에 대해 매번 놀라고 있어요. 저희는 하루빨리 실무 현장에서 변화가 체감될 수 있도록 제품의 완성도를 높이는 데 최선을 다하고 있고요.
저뿐만 아니라 에어브릿지 팀원 모두가 책임감 있게 일하고 있어요. 저희가 풀고자 했던 문제가 여러분의 실무 현장에서 실질적인 해결책으로 작동하기를 바랍니다.
**_Interviewed & Written by 프리랜서 에디터 강혜라_**
---
## 복잡한 마케팅 데이터도 이제 찾지 말고 질문하세요, Airbridge AI
URL: https://www.airbridge.io/ko/blog/airbridge-ai-mcp-pilot-launch
Language: Korean
Published: 2026-03-25
Author: Airbridge
캠페인 성과를 확인하려고 대시보드를 열었는데, 5분이 지나도 원하는 리포트를 못 찾은 경험이 있으신가요? ROAS는 어디서 확인해야 하고, 리텐션은 또 어디서 측정해야 하는지. 필터 설정은 어떻게 했는지 매일 몇 번씩 고민하셨죠?
짧으면 5분, 길면 15분. 데이터를 **분석하는 시간**보다 데이터를 **꺼내고 정리하는 시간**이 더 길어서 답답했던 적이 한두 번이 아니었을 거예요.
게다가 더 편하게 마케팅 데이터를 보기 위해 MMP를 도입했는데 SDK 설치하는 과정도 어렵고, 플랫폼마다 문서가 달라서 헬프센터와 개발 환경을 번갈아 가며 확인해야 하고, 마케터가 직접 할 수 없어서 개발자를 기다려야 하는 경우도 많아요. 온보딩을 시작하기도 전에 지치는 구간이죠.
AI 시대가 빠르게 열리면서 업무 방식도 달라지고 있는데요. 이 흐름에 맞춰, 마케터분들이 더욱 편하게 시간을 아껴드리기 위해 에어브릿지가 개발한 **Airbridge AI(Airbridge MCP, Airbridge Pilot, Onboarding Pilot)** 를 소개해드릴게요.
## 💡Airbridge AI 프로젝트를 리드한 Product Owner 이승헌 님이 직접 소개해 드릴게요.
https://youtu.be/_HRxBqkTZC0
### **Airbridge AI 주요 기능**
- **Airbridge Pilot**: 대시보드 안에서 Airbridge 관련 질문에 즉시 답해줘요
- **Onboarding Pilot**: SDK 설치 전 과정을 AI가 단계별로 안내해줘요
- **Airbridge MCP**: 지금 쓰는 AI 툴에서 자연어로 데이터를 바로 조회해요
https://youtu.be/jKNmsiJUxw4
## **Airbridge Pilot: 대시보드 안에서 즉각 답하는 AI 에이전트**
"**이제 헬프센터 페이지를 일일이 안 열어도 돼요. AI에게 물어보고 바로 데이터를 확인하세요.**"
대시보드를 쓰다 보면 갑자기 궁금한 것들이 생겨요.
_“이 Attribution 규칙은 어떻게 작동하지? SKAN 설정은 어디서 하더라? 딥링크 오픈 이벤트가 안 잡히는 이유가 뭐지?”_
이런 순간마다 헬프센터 페이지를 열고, 검색하고, 관련 문서를 찾게 되는데요. 이 과정이 생각보다 흐름을 많이 끊어요. 특히 MMP를 처음 쓰는 팀원이라면, 기능 하나를 쓰기 위해 문서를 오가는 시간이 실제 업무 시간보다 길어지기도 하죠.
**Airbridge Pilot**은 그런 불필요한 과정 자체를 없애고 업무 흐름을 계속 유지하게 도와줘요.
[대시보드를 보다가 궁금한 점이 있으면 바로 Airbridge Pilot을 켜서 질문해 보세요.(Airbridge Pilot 아이콘은 화면 오른쪽 아래에 있어요.)]
_대시보드를 보다가 궁금한 점이 있으면 바로 Airbridge Pilot을 켜서 질문해 보세요.(Airbridge Pilot 아이콘은 화면 오른쪽 아래에 있어요.)_
[AI 챗봇이랑 대화하는 것처럼 자연어로 질문하면 Airbridge Pilot이 질문을 분석하고 데이터를 정리해서 바로 보여줘요.]
_AI 챗봇이랑 대화하는 것처럼 자연어로 질문하면 Airbridge Pilot이 질문을 분석하고 데이터를 정리해서 바로 보여줘요._
### **1\. Airbridge Pilot이 기존 챗봇과 다른 점**
단순히 키워드만 입력하거나 이미 정해진 선택지에서 고르는 챗봇 방식이 아니에요. Airbridge Pilot은 대화형 AI 에이전트로, 여러분이 평소에 AI 채팅과 대화하는 것처럼 편하게 자연어로 질문과 답을 주고 받으면서 즉석에서 데이터를 확인하고 문제를 해결할 수 있어요.
트래킹 링크, 리포트, Attribution 규칙, iOS SKAN 등 에어브릿지를 쓰면서 궁금한 점이 있었다면 Pilot을 클릭하고 편하게 질문하면 돼요.
또한 **멀티턴 대화(Multi-turn) 대화를 지원**해서 아래와 같이 이전 대화의 맥락을 기억하며 채팅을 이어나갈 수 있어요.
_"트래킹 링크에 딥링크를 연결하려면 어떻게 해?" → 답변을 받은 뒤 → "그럼 그 링크가 iOS에서 앱이 없을 때는 어떻게 동작해?"_
### **2\. Airbridge Pilot의 주요 특징**
- **질문 횟수에 제한 없음**: 언제든, 얼마든지 편하게 질문하세요.
- **모든 플랜 고객 사용 가능**: 모든 플랜에서 동일하게 이용 가능해요.
- **질문한 언어로 답변**: 한국어로 물으면 한국어로, 영어로 물으면 영어로 답변해요.
- **Airbridge Help Center 기반 답변**: 에어브릿지 팀에서 직접 작성한 공식 문서를 바탕으로 답변하므로 신뢰도가 매우 높아요.
https://youtu.be/wiIm7XWmN3I
## **Onboarding Pilot: 답답했던 SDK, AI와 대화하면서 빠르게 설치하세요**
"**문서 보면서 혼자 설치하던 시대는 이제 끝이에요.**"
앱 마케팅의 시작은 SDK 설치예요. 그런데 막상 해보면 생각보다 장벽이 높아요.
플랫폼마다 설치 방법이 다르고, 문서를 보면서 개발 환경과 브라우저를 왔다갔다 해야 하고, 개발자에게 부탁해서 설치해야 하는 경우도 있어요. 이 과정에서 온보딩이 지연되거나 중간에 멈추는 경우가 적지 않아요.
**Onboarding Pilot**은 이 모든 과정을 AI 대화 방식으로 안내해주고 모르는 게 있으면 바로 물어보면서 답을 찾을 수 있어요. 문서를 찾을 필요 없이, AI와 대화하며 단계를 따라가다 보면 설치가 완료돼요.
[Onboarding Pilot을 실행하기 전 Airbridge에 앱을 먼저 등록해야 해요.]
_Onboarding Pilot을 실행하기 전 Airbridge에 앱을 먼저 등록해야 해요._
[Onboarding Pilot을 담당할 작업자의 이메일 주소를 넣어주세요.]
_Onboarding Pilot을 담당할 작업자의 이메일 주소를 넣어주세요._
[SDK는 Airbridge MCP와 직접 설치 중에서 선택할 수 있어요.]
_SDK는 Airbridge MCP와 직접 설치 중에서 선택할 수 있어요._
[Airbridge에 등록한 앱을 찾아보세요. 만약 앱이 출시되지 않았다면 가상의 앱으로 테스트해 볼 수 있어요.]
_Airbridge에 등록한 앱을 찾아보세요. 만약 앱이 출시되지 않았다면 가상의 앱으로 테스트해 볼 수 있어요._
[설치 방법별로 Onboadng Pilot이 작업 순서를 설명해줘요.]
_설치 방법별로 Onboadng Pilot이 작업 순서를 설명해줘요._
[Onbaording Pilot 안내에 따라 SDK 설치를 진행하세요.]
_Onbaording Pilot 안내에 따라 SDK 설치를 진행하세요._
[SDK 설치가 완료되면 잘 작동하는지 바로 확인할 수 있어요.]
_SDK 설치가 완료되면 잘 작동하는지 바로 확인할 수 있어요._
### **Onboarding Pilot으로 SDK 설치하는 과정 간단 소개**
1. **누가 설치할까요?** → 마케터가 직접 설치할 수도 있고, 개발자를 초대하여 설치를 진행할 수 있어요.
2. **어떻게 설치할까요?** → MCP 방식과 수동 설치 중에서 선택할 수 있어요.
3. **어떤 환경인가요?** → Android, iOS, React Native, Flutter, Expo에서 설치 가능하며, 선택하는 플랫폼에 맞는 맞춤 가이드를 제공해요.
4. **설치가 잘 됐나요?** → 실시간 로그를 기반으로 정상 수신 여부를 자동으로 확인해요
5. **Activation 완료 단계별 체크** → 다음 3단계를 자동으로 검증해요
6. 앱 설치 이벤트 수신
7. 트래킹 링크 생성
8. 딥링크 오픈 이벤트 수신
Onboarding Pilot의 AI는 직접 코드를 배포하지 않고, 단계별 안내와 완료 검증만 맡아요. 최종 코드 적용은 개발자가 해요.
https://youtu.be/QwC8fzh1Cpo
%%CTA|데모 신청하기|/ko/book-a-demo|Airbridge AI를 직접 체험해 보세요. 복잡한 데이터도 질문 하나로 해결돼요.%%
## **Airbridge MCP: 지금 쓰는 AI 툴에서 바로 에어브릿지 데이터를 조회하세요**
"**더이상 일일이 대시보드 열지 않아도 돼요**"
마케터들은 하루에도 여러 툴을 오가면서 데이터를 확인해요. 기획 문서를 작성하다가 캠페인 수치가 필요해 대시보드로 이동하고, 확인하고 나서 다시 돌아와 작업을 이어가는 식이죠. 이렇게 여러 툴을 이동하는 일이 많을수록 맥락이 끊기고, 업무 속도도 자연스럽게 느려져요.
**Airbridge MCP**는 Claude, ChatGPT 같은 AI 툴에 에어브릿지를 연결해 자연어로 데이터를 바로 불러오는 기능이에요. 지금 작업하는 AI 환경을 벗어나지 않고, 대화하듯 원하는 데이터를 가져올 수 있어요.
"지난주 캠페인 ROAS를 채널별로 보여줘" "이번 달 리텐션 추이는 어때?" "iOS 캠페인 Active Users 어제 수치 알려줘"
별도 탭을 열 필요도, 리포트 메뉴를 뒤질 필요도 없어요. AI에게 질문하면 바로 데이터를 확인할 수 있어요.
[Airbridge MCP를 사용하려면 AI 툴에서 커스텀 커넥터를 추가해야 해요.(클로드 기준)]
_Airbridge MCP를 사용하려면 AI 툴에서 커스텀 커넥터를 추가해야 해요.(클로드 기준)_
[MCP를 연결한 후 채팅창에 확인하고 싶은 데이터를 편하게 자연어로 질문하세요.]
_MCP를 연결한 후 채팅창에 확인하고 싶은 데이터를 편하게 자연어로 질문하세요._
[AI 툴이 질문을 분석한 후 Airbridge에서 맞는 데이터를 찾아 정리한 후 채팅창에서 바로 보여줘요.]
_AI 툴이 질문을 분석한 후 Airbridge에서 맞는 데이터를 찾아 정리한 후 채팅창에서 바로 보여줘요._
### **1\. Airbridge MCP로 무엇을 확인할 수 있나요?**
베타 버전에서는 아래 캠페인 목록을 확인할 수 있어요.
- Actuals
- Retention
- Revenue
- Active Users 리포트 조회
- 앱 설정 조회
- 트래킹 링크 읽기
- OAuth 인증
- RBAC (Role Based Access Control) 연동
4월 정식 출시 이후에는 트래킹 링크 생성·수정, 오디언스 조회, 차트 시각화, Multi-App 쿼리, 자동화 연동 등을 순차적으로 추가할 예정이에요.
### **2\. Airbridge MCP 연결 방법**
**Claude Desktop**이라면 Settings → Connectors → Custom Connectors에 아래 정보를 입력해서 연결할 수 있어요.
- Name: Airbridge
- URL: [https://mcp.airbridge.io/mcp](https://mcp.airbridge.io/mcp)
**Claude Code CLI**라면 아래를 입력해서 연결할 수 있어요.
- claude mcp add --transport http airbridge [https://mcp.airbridge.io/mcp](https://mcp.airbridge.io/mcp)
더 자세한 설치방법은 [에어브릿지 헬프센터 국문 가이드](https://help.airbridge.io/ko/guides/mcp)에서 확인할 수 있어요.
### **3\. Airbridge MCP는 이런 분께 특히 유용해요**
- 대시보드보다 AI 툴 안에서 업무 시간이 더 많은 마케터
- 팀 내 데이터 조회 요청을 반복적으로 처리하는 데이터 담당자
- 여러 캠페인을 동시에 모니터링해야 하는 퍼포먼스 마케터
## **엔터프라이즈 수준의 데이터 보안 기준으로 관리해요**
AI 기능이라고 하면 데이터 보안이 가장 걱정되실 텐데요. 하지만 Airbridge MCP와 Airbridge AI를 포함한 에어브릿지 내 모든 데이터는 아래 6가지 원칙 아래 철저하게 관리되고 있어요.
1. No Storage: 조회 데이터를 서버에 저장하지 않아요.
2. No Training: 사용자 데이터를 AI 학습에 활용하지 않아요.
3. No Sharing: 데이터를 외부에 공유하지 않아요.
4. PII 차단: Device ID, IP, 이메일 등 개인정보는 AI에 전달되지 않고 기술적으로 접근 자체가 차단되어 있어요.
5. RBAC 계승: 기존 Airbridge 계정 권한이 AI 환경에도 그대로 적용돼요.
6. 글로벌 컴플라이언스: GDPR, CCPA, PIPA를 준수해요.
## **Airbridge AI 지금 바로 시작하세요**
이제 "내가 데이터를 찾으러 이동"하는 게 아니라 "내가 원하는 데이터가 내 앞으로 오게" 만드는 시대예요.
MCP로 데이터를 바로 불러오고, Pilot에게 궁금한 걸 물어보고, Onboarding Pilot과 함께 SDK를 빠르게 설치해보세요. 대시보드 구조를 몰라도, 세부 기능이 아직 익숙하지 않아도 괜찮아요. Airbridge AI를 활용하면 아무리 어려운 작업도 편하고 빠르게 처리할 수 있어요.
경쟁사가 리포트를 만드는 동안, 여러분은 이미 다음 액션을 실행하고 있을 거예요. AI와 함께 데이터를 더 빠르게 활용하는 팀이 더 빠르게 성장하는 시대예요. Airbridge AI와 함께 그 속도를 빠르게 높여보세요.
## **FAQ with Airbridge AI**
### **Q. Airbridge MCP는 어떤 AI 툴에서 쓸 수 있나요?**
MCP를 지원하는 모든 환경에서 사용 가능해요. Claude Desktop, Claude Code, Cursor, Codex CLI가 대표적이며, ChatGPT 연동도 순차적으로 지원될 예정이에요.
### **Q. Airbridge MCP를 쓰면 회사 데이터가 AI에 넘어가지 않나요?**
넘어가지 않아요. 조회 요청에 대한 응답만 전달되고, 데이터는 저장되거나 학습에 쓰이지 않아요. PII 접근도 기술적으로 차단되어 있어요.
### **Q. Airbridge Pilot은 어떤 플랜에서 쓸 수 있나요?**
에어브릿지를 이용하는 모든 고객이 사용 가능하고, 질문 횟수에 제한도 없어요.
### **Q. Onboarding Pilot으로 개발자 없이 SDK 설치가 가능한가요?**
AI가 전 과정을 안내하지만, 최종 코드 적용은 개발자가 직접 해요. 마케터가 먼저 흐름을 시작하고, 필요한 시점에 개발자를 초대하는 방식으로 진행돼요.
### **Q. AI 답변이 틀릴 수도 있지 않나요?**
드물지만 부정확한 답변이 나올 수 있어요. 중요한 설정이나 정책은 Airbridge 공식 헬프센터에서 추가로 확인하는 걸 권장해요.
---
## ‘최대 1.5억 지원’ 2026년 게임더하기 지원사업 신청조건과 방법 총정리
URL: https://www.airbridge.io/ko/blog/global-game-service-platform
Language: Korean
Published: 2026-03-21
Author: Team Airbridge
# **‘최대 1.5억 지원’ 2026년 게임더하기 지원사업 신청조건과 방법 총정리**
글로벌 진출을 준비 중인 게임사라면 한 번쯤 ‘게임더하기 지원사업’에 대해 들어보셨을 거예요. 이 지원사업은 국내 게임사가 해외 시장이라는 큰 무대로 나아갈 때 마중물이 되어주는 정부의 대표적인 지원책인데요.
특히 올해는 개인정보 보호 규제가 더욱 까다로워지고 마케팅 환경이 AI로 급변하고 있어, 그 어느 때보다 철저한 데이터 전략과 준비가 필요해요.
올해 사업도 글로벌 문턱을 넘어야 하는 중소 게임사에 지원금, 컨설팅, 마케팅 등 실질적인 도움을 줄 수 있는 기회가 될 수 있어 기대를 모으고 있어요.
**📌 핵심 포인트**
- 게임사라면 최대 1억 5,000만원 상당의 지원금으로 부담 없이 글로벌 진출을 위한 마케팅 인프라를 구축할 수 있어요.
- 넥슨, NHN, 컴투스, 웹젠 등 대형 게임사와 트렌디한 게임사가 선택한 에어브릿지의 기술력을 게임더하기 바우처를 통해 별도 검증 절차 없이 즉시 도입할 수 있어요.
## **게임더하기 지원사업은 무엇인가요?**
### **중소 게임사가 글로벌 시장의 직진출을 돕는 사업이에요**
한국콘텐츠진흥원이 주관하는 ‘게임더하기 지원사업’은 국내 중소 개임사의 자생력 강화를 위한 통합 지원 프로그램이에요. [**퍼블리셔에 의존하지 않고 글로벌 오픈 마켓에 직접 진출**](https://www.bizinfo.go.kr/sii/siia/selectSIIA200Detail.do?pblancId=PBLN_000000000118975)할 수 있도록, 마케팅부터 현지화까지 필요한 모든 서비스를 바우처 형태로 지원해요.
특히 과거처럼 단순히 마케팅 비용만 지원하는 것이 아니라, 게임의 완성도를 높이는 단계부터 실제 해외 유저를 만나는 런칭 단계까지 게임 전체 라이프 사이클을 서포트해요.
이를 통해 국내에서 개발한 게임이 글로벌 시장에도 독자적인 브랜드 파워를 쌓게 하고, 데이터 주도권을 가질 수 있을 실질적인 환경을 만들어주는 것이 목적이에요. [**2026년 게임더하기 사업 정보와 일정은 아래와 같으며, 2월 26일부터 3월 25일까지 접수**](https://indiegame.com/archives/23723)할 수 있어요.
### **2026 게임더하기 지원사업 한눈에 보기**
| 항목 | 핵심 내용 |
| --- | --- |
| 지원 대상 | 국내 중소 게임사 / 해외 직접 진출을 준비 중인 게임 프로젝트 |
| 지원 유형 | 성과창출형(해외 출시 완료) / 초기진입형(해외 출시 이력 없음) 중 선택 |
| 지원 방식 | 게임더하기 플랫폼 포인트 지급 (1포인트 = 1원) |
| 지원 규모 | 기본 5,000 포인트~최대 1억 5,000만원 상당의 포인트 |
| 기업 자부담 | 기본 포인트의 5% 수준으로 약 250만 포인트 |
| 접수 기간 | 2026년 3월 25일까지 |
| 신청 방법 | KOCCA 사업관리시스템 온라인 접수 |
| 지원기간 | 4월 말 선정 후 2026년 12월 31일까지 |
_게임더하기 사업 상세 내용_
특히 올해는 해외 직접진출 경험 유무에 따라 **성과창출형과 초기진입형으로 지원트랙을 세분화하여, 우리 게임의 빌드 상태에 최적화된 지원**을 받을 수 있게 됐죠. 지원 방식에 대한 상세한 사항은 아래 내용을 참고해 보세요.
### **2026 게임더하기 지원사업 신청 상세 가이드**
**1\. 지원 대상**
- **국내 중소게임사에서 개발되어 해외 직접진출을 희망하는 게임 프로젝트**
- **게임 개발 주체(개발사)가 직접 개발하여 개발사명 혹은 임원명으로 런칭 가능한 게임**
2\. **지원유형**
- **성과창출형:** 공고일(2026.2.26) 기준 이미 해외 정식 출시를 완료한 게임이 해당돼요. 기존 출시 권역에서의 성과를 극대화하거나, 새로운 권역으로 추가 진출을 노리는 팀이 대상이에요.
- **초기진입형:** 게임의 해외 출시 이력이 없고 1개국 이상 신규 해외 진출을 희망하는 게임을 대상으로 해요.**3\. 지원규모 및 방식**
- 현금이 아닌 포인트 지급 방식으로 운영되어 1원은 1포인트인 구조에요. **선정되면 기본 5,000만 포인트를 우선 지급**받고, 이후 평가 결과에 따라 **추가 포인트를 배정받으면 최대 1억 5,000만 포인트까지 활용**할 수 있어요. 포인트는 현금과 1:1 가치로 서비스 장터 내에서 자유롭게 결제할 수 있어요.
- 기업은 자부담금 250만 원을 부담해야 하며 지원받는 포인트보다 우선해서 집행해야 해요.
**4\. 지원내용**
- 해외 직접진출 전략 컨설팅 및 교육 지원
- 해외 직접진출 성과 관리 및 확산 지원
- 게임더하기 서비스 자율선택 포인트 관리 및 이용 지원
- 게임더하기 서비스 산출물 및 검수 관리 지원
- 구글 플레이와의 글로벌 파트너십을 통한 서비스 협력 지원
**5\. 신청방법**
- [KOCCA 사업관리시스템](https://pms.kocca.kr/)에서 온라인 접수
**6\. 신청 및 운영 시 주의사항**
- **필수 서류 제출**: 공고문에서 제공하는 양식에 맞춰 작성한 신청서와 사업계획서를 제출하는 것이 기본이에요. 또한 게임의 핵심 콘텐츠와 특징을 보여주는 소개 자료와 실제 플레이 영상, 현재 개발 중이거나 서비스 중인 게임 상태를 확인할 수 있는 빌드 파일 또는 마켓(App Store, Google Play 등) 연결 링크를 등록하세요.
- **기타 증빙 서류**: 기업의 적격성을 확인하는 서류로 사업자등록증, 국세·지방세 완납증명서 등 기업의 적격성을 확인하는 서류를 준비해 주세요.
- **월별 정산 보고:** 사업 기간에는 월간 매출, 다운로드, DAU, 리텐션, 퍼블리싱 계약 등 성과 자료를 정리해서 KOCCA에 매월 제출해야 해요. 에어브릿지에서 제공하는 정교한 리포트를 이용하면 이 월별 보고 업무의 효율을 크게 높일 수 있어요.
### **게임사가 필요한 서비스를 골라 담으세요**
게임더하기 사업의 큰 장점은 공급자가 정해준 서비스를 수동적으로 받을 필요가 없다는 점이에요. 게임사는 **현재 가장 필요로 하는 솔루션을 직접 선택하는 방식**으로 전용 플랫폼 내 서비스 장터에서 마케팅, 현지화, 데이터 인프라 등 다양한 카테고리 중 **우리 팀에 꼭 필요한 서비스를 쇼핑**하듯 고를 수 있어요.
예를 들어, 북미 진출을 앞두고 현지 맞춤형 마케팅과 정교한 성과 분석이 동시에 고민이라면, 관련 서비스를 자유롭게 조합해 우리 게임만의 맞춤형 패키지를 구성할 수 있어요. 이러한 유연한 운영 방식 덕분에 한정된 지원금을 낭비하지 않고 게임의 해외 진출을 위한 전략적 투자가 가능해요.
[이미지 출처: 게임더하기 공식 홈페이지]
_이미지 출처: 게임더하기 공식 홈페이지_
## **우리 게임사가 지원 대상으로 선정되면 어떤 혜택을 받나요?**
그렇다면 지원 대상으로 선정된 우리 게임사가 구체적으로 어떤 혜택을 통해 글로벌 진출의 가속도를 높일 수 있을까요? 실무 현장에서 활용할 수 있는 핵심 혜택 3가지를 정리해 드릴게요.
### **마케팅부터 데이터 분석 인프라까지 ‘우리 팀 맞춤형’으로 구축할 수 있어요**
앞서 소개한 자금 지원만큼 중요한 것이 게임사가 스스로 성장할 수 있는 ‘인프라’를 갖추는 것이죠. 이번 사업은 **모바일 30개사, PC/콘솔 10개사 내외의 프로젝트를 선발**하여 각 플랫폼 특성에 맞는 서비스를 지원해요.
게임 내 유저 행동을 정교하게 분석하고 싶은 팀이라면 **에어브릿지와 같은 데이터 인프라를 구축하여 유입 경로별 성과를 정확히 측정**할 수 있어요. 이는 사업 종료 후에도 게임사가 스스로 데이터를 보고 의사결정을 내릴 수 있는 데이터 체력을 기를 수 있는 기반이 될 거예요
### **글로벌 진출 리스크를 최소화하는 현지화 컨설팅을 받을 수 있어요**
해외 진출 시 가장 큰 걸림돌은 현지 사정을 모른 채 무작정 제품을 런칭하는 것일 텐데요. 지원사업에 선정되면 단순 번역을 넘어 현지화 완성도를 높이는 LQA(Localization Quality Assurance)와 글로벌 운영 전략 컨설팅을 받을 수 있어요.
국가별로 상이한 개인정보 보호 규제 대응이나 스토어 최적화 가이드라인을 얻을 수 있어, 시행착오로 인한 시간과 비용 낭비를 최소화하고 안정적으로 글로벌 런칭을 완수할 수 있죠.
### **글로벌 마켓과의 직접적인 파트너십 기회도 열려 있어요**
역량이 검증된 우수 프로젝트에는 글로벌 플랫폼인 구글 플레이와의 파트너십 혜택이 추가로 주어져요. 선정된 우수 게임은 구글 플레이 내 프로모션 심사 지원은 물론, 글로벌 출시 전략 세미나와 전문 교육을 받을 수 있죠.
특히 크로스 디바이스 개발 선정사의 경우 전용 온보딩 지원까지 제공되어, 전 세계 유저들에게 우리 게임을 더 효과적으로 노출하고 안착시킬 수 있는 강력한 발판이 될 거예요.
%%CTA|데모 신청하기|/ko/book-a-demo|게임더하기 바우처로 에어브릿지를 무료로 시작해 보세요.%%
## **성공적인 런칭만큼 중요한 건 ‘데이터 기반 성장’이에요**
현지화와 컨설팅으로 성공적인 런칭 준비를 마쳤다면, 이제 실질적인 데이터로 성장 궤도를 그려야 할 때예요. 수많은 마케팅 채널 중 어떤 광고가 진짜 결제 유저를 데려왔는지, 우리 게임의 LTV(생애 가치)가 광고비보다 높은지 정확히 파악해야 하죠.
[MMP(Mobile Measurement Partner)](https://www.airbridge.io/ko/blog/what-is-mmp)는 이 모든 유입 경로와 성과를 투명하게 연결해 주는, 글로벌 게임 마케팅의 나침반 같은 역할을 해요.
### **중복 집계 함정을 피하고 광고 효율을 극대화하세요**
글로벌 마케팅을 시작하면 메타, 구글, 틱톡 등 수많은 채널에 동시에 광고를 집행하게 돼요. 이때 각 광고 매체는 자체적인 기준으로 성과를 집계하기 때문에 여러 광고를 접하고 유입된 한 명의 유저가 각기 다른 마체에서 중복으로 성과가 카운트되는 문제가 발생해요.
MMP는 서드파티 입장에서 모든 매체의 터치 포인트를 통합 분석하여, **최종적으로 기여한 단 하나의 매체에만 성과를 할당**해요. 이러한 데이터 정제 과정이 없다면 매체별 리포트 수치의 총합이 실제 유입된 유저 수보다 부풀려지게 되며, 결국 성과가 낮은 매체에 예산을 중복 배당하는 비효율을 초래할 수 있죠.
결국 MMP를 통해 확보한 객관적인 기여도 데이터는 한정된 마케팅 예산을 성과가 검증된 채널에 집중 배치하여 **전체 캠페인의 ROAS(광고비 대비 매출액)을 최적화하는 필수 근거**가 돼요.
### **가짜 트래픽을 차단하여 소중한 지원금을 보호하세요**
글로벌 시장에는 광고 클릭을 인위적으로 만들어내는 봇(Bot)이나 가짜 클릭 같은 '광고 사기(Ad Fraud)' 리스크가 존재해요. 특히 대규모 UA(User Acquisition) 캠페인을 진행할 때 이러한 허수 트래픽을 걸러내지 못하면, 게임더하기 사업을 통해 받은 소중한 지원금이 유령 유저들에게 흘러갈 수 있어요.
고도화된 MMP는 이러한 **이상 패턴을 실시간으로 감지하고 차단하여, 지원금이 실제 우리 게임을 플레이할 '진성 유저'를 확보**하는 데 온전히 쓰이도록 도와줘요.
### **마케팅 성과를 넘어 프로덕트의 개선 방향을 찾아내세요**
진성 유저를 확보했다면, 이제는 그들이 게임 내에서 어떻게 움직이는지 분석할 차례예요. 유저가 특정 구간에서 이탈하지는 않는지, 어떤 인앱 이벤트가 결제로 전환되는지 파악하여 마케팅 채널별 유저의 질을 평가해야 하죠.
MMP가 제공하는 통합 데이터는 단순한 광고 효율 측정을 넘어, 게임 콘텐츠의 업데이트 방향이나 시장 확장 우선순위를 결정하는 정교한 이정표가 되어줘요.
## **에어브릿지는 2026년 게임더하기 지원사업의 성과를 극대화할 수 있는 확실한 선택이에요**
[이미지 출처: 게임더하기 공식 홈페이지]
_이미지 출처: 게임더하기 공식 홈페이지_
### **넥슨, 슈퍼캣 등 대형 게임사가 검증한 분석 성능을 사용해 보세요**
에어브릿지는 이미 [**넥슨**](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)**,** [**웹젠**](https://www.airbridge.io/ko/case-studies/mu-monarch-creative-optimization)**,** [**퍼즐몬스터즈**](https://www.airbridge.io/ko/case-studies/idle-ninja-online-growth-marketing)**,** [**후야호,**](https://www.airbridge.io/ko/case-studies/whoyaho-game-data-marketing-strategy) **NHN, 컴투스 등 국내외 유수의 대형 게임사들과 협업**하며 그 기술력을 탄탄하게 증명해 왔어요. 수백만 명의 유저가 동시에 접속하고 방대한 이벤트 데이터가 발생하는 게임 환경에서도 지연 없는 실시간 데이터 처리를 지원하죠. 단순히 데이터 수집에 그치지 않고, 마케팅 의사결정에 즉시 활용할 수 있는 데이터 무결성과 정교한 리포팅 기능을 갖추었어요.
### **웹과 모바일, PC/콘솔까지 아우르는 통합 어트리뷰션을 제공해요**
[에어브릿지 통합 어트리뷰션]
유저는 모바일 광고를 보고 PC 버전 게임을 설치하거나, 웹에서 사전 예약을 한 뒤 앱에서 플레이를 시작하기도 해요. 에어브릿지는 이러한 복잡한 유저의 여정을 하나로 연결하는 통합 어트리뷰션 기능을 제공해요. 파편화된 데이터를 하나로 모아 분석할 수 있어, 우리 게임의 진정한 성장 엔진이 어디인지 정확하게 파악할 수 있어요.
### **검증 절차가 필요 없는 ‘준비된 공식 협력사’예요**
새로운 솔루션을 도입할 때마다 겪어야 하는 복잡한 보안성 검토나 공급사 검증 절차는 현업 담당자에게 큰 부담이 되곤 해요. 에어브릿지는 2024년부터 지금까지 게임더하기 사업과 함께 한 공식 협력사로 이미 엄격한 심사 과정을 통과한 만큼, 즉시 도입하여 리소스 낭비를 줄이고 오로지 게임 런칭과 마케팅 최적화에만 집중할 수 있는 환경을 만들어 줘요.
## **게임더하기 사업 바우처로 에어브릿지 100% 활용하기**
### **별도 비용 부담 없이 포인트로 ‘프리미엄 기능’을 마음껏 써보세요**
게임더하기 사업을 활용하면 현금 지출 없이 비즈니스 포인트만으로 엔터프라이즈급 기능들을 누릴 수 있어요. 글로벌 마케팅에 필수적인 SKAN 분석, 광고 사기 방지, 실시간 로 데이터(Raw Data) 추출 등 에어브릿지의 강력한 솔루션을 비용 걱정 없이 도입해 보세요.
### **단계별로 밀착 지원하는 전담 매니저와 함께 빠르게 시작하세요**
새로운 분석 툴을 도입할 때 가장 걱정되는 게 바로 연동 과정이죠. 에어브릿지는 게임더하기 사업을 통해 도입하시는 게임사 분들을 위해 전문적인 온보딩 프로세스를 제공해요. 기술 연동 단계부터 실제 데이터 대시보드를 구성하는 방법까지, 전담 매니저가 우리 게임의 특성에 맞춰 가이드해 드려요.
### **투명한 성과 증빙을 돕는 상세 리포트와 로데이터를 확인하세요**
지원사업을 수행하다 보면 나중에 마케팅 예산이 적절하게 쓰였는지, 어떤 성과를 냈는지 증빙해야 하는 과정이 꽤 까다로워요. 에어브릿지는 여러 매체에 흩어진 광고 비용과 설치, 결제 데이터를 하나로 모아 실질적인 광고 효율을 한눈에 보여주는 통합 리포트를 지원해요. 로 데이터(Raw Data)까지 투명하게 추출할 수 있어, 사업 결과 보고에 필요한 객관적인 성과 근거 자료를 확보하는 데 매우 유용해요.
👉[에어브릿지, 한국콘텐츠진흥원 ‘2023 게임더하기’ 사업 협력사 선정](https://www.airbridge.io/ko/blog/ko-gsp-plus-2023)
## **모바일부터 PC·콘솔까지, 에어브릿지로 게임의 LTV를 극대화해보세요**
2026년 게임더하기 사업 선정은 글로벌 시장이라는 더 큰 무대로 나아가는 소중한 기회예요. 단순히 지원금을 사용하는 것에 그치지 않고, 에어브릿지와 함께 모바일과 웹, PC를 넘나드는 정교한 데이터 인프라를 구축해 보세요.
유입부터 결제까지 유저의 생애 가치(LTV)를 정확히 파악할 때, 우리 게임은 비로소 지속 가능한 성장을 이어갈 수 있어요. 글로벌 직진출이라는 거침없는 여정에서 에어브릿지가 가장 객관적이고 든든한 데이터 파트너가 되어 드릴게요.
---
## 라인야후 애즈 연동, 이제 에어브릿지로 일본 마케팅 ROI를 높여보세요
URL: https://www.airbridge.io/ko/blog/airbridge-line-yahoo-japan-ads-integration
Language: Korean
Published: 2026-03-19
Author: Jaehyuk Kim
**에어브릿지와 일본의 대표 통합 광고 플랫폼인 라인야후 애즈가 연동됐어요.** 이제 라인야후 애즈에서의 광고 성과를 에어브릿지에서 더욱 가시적으로 분석할 수 있어요.
일본 시장 진출을 준비 중이시거나, 이미 일본에서 온라인 마케팅을 하고 계시다면 라인과 야후재팬을 통한 광고가 매우 중요하다는 점을 알고 계셨을 거예요. 이미 예산까지 편성해서 최우선 마케팅 채널로 고려하고 계셨을 텐데요.
이번 에어브릿지와의 연동으로, 라인야후 재팬을 통해 집행하는 광고 및 마케팅 성과 분석 뿐만 아니라 보다 정교한 리타겟팅 캠페인 전략까지 세울 수 있어요. 이를 통해, 일본 시장 내 퍼포먼스 마케팅의 퀄리티를 한 단계 더 끌어올릴 수 있을 거예요.
**📌핵심 포인트**
- 라인야후 애즈는 일본 최대 메신저인 라인(LINE)과 대표 검색 포털인 야후 재팬의 통합 광고 플랫폼으로, 일본 인터넷 사용자의 80% 이상에게 도달할 수 있는 일본 최대 규모의 광고 플랫폼이에요.
- 에어브릿지와 라인야후 애즈의 연동으로, 하이브리드 어트리뷰션과 SKAN 통합 관리를 통해 라인야후 애즈의 순수 기여도를 정교하게 측정하며, 실시간 데이터 포스트백과 커스텀 오디언스 전송을 기반으로 일본 시장 내 광고 최적화 및 리타겟팅 효율까지 일본 마케팅 성과를 극대화할 수 있게 됐어요.
## **라인야후 애즈(LY Ads)란?**
라인야후 애즈는 일본 최대의 IT 기업인 라인야후 주식회사(LY Corporation)가 운영하는 통합 광고 플랫폼으로, 일본의 대표 메신저이자 SNS인 라인과 대표 검색 엔진인 야후 재팬의 광고 서비스를 통합한 디지털 광고 플랫폼이에요.
기존에는 라인 애즈와 야후 재팬 애즈가 별도의 광고 시스템으로 각각 운영되고 있었는데요. 라인과 야후 재팬이 하나로 합치면서 두 광고 시스템도 하나의 플랫폼으로 통합되었어요.
그래서 라인야후 애즈에서는 광고주가 하나의 광고 계정으로 라인과 야후 재팬의 광고 인벤토리를 동시에 운영할 수 있는데요. 예를 들어, 다음과 같은 매체에 광고를 노출할 수 있어요.
- LINE 앱
- LINE 광고 네트워크
- Yahoo! JAPAN 앱
## **일본 온라인 마케팅의 중심 = 라인야후 애즈**
**일본은 글로벌 플랫폼만큼이나 로컬 플랫폼의 영향력이 막강한 시장이에요.** 그래서 구글이나 메타, 틱톡도 중요하지만, 일본 대표 로컬 플랫폼인 라인과 야후재팬을 공략하는 게 매우 중요해요.
- **라인과 야후재팬을 통합한 방대한 광고 지면:** 라인야후 애즈는 일본 내 9,900만 명 이상의 MAU를 보유한 라인과 월간 8,500만 명 이상이 사용하는 아후재팬의 지면에 모두 통합하여 광고를 노출할 수 있어요. 일본인의 일상에 깊숙이 침투한 모든 지면에 광고를 노출합니다.
- **전 국민적 고른 도달범위:** 일본 인터넷 사용 인구의 약 80% 이상이 라인을 사용하고 있고, 전 연령이 고르게 사용하고 있어요. 전 일본 소비자를 타겟으로, 가시성을 확보할 수 있는 가장 효과적인 통로예요.
- **고도화된 유저 세그먼트:** 야후의 검색 의도 데이터와 라인의 메시지 기반 사용자 행동 데이터를 결합하여, 유저의 관심사를 매우 정교하게 타겟팅할 수 있어요.
**이렇듯 라인야후 애즈가 가지는 플랫폼 영향력이 강력해서, 일본 마케팅의 중심이라고 할 수 있어요.** 일본 진출을 고려하고 있다면 라인야후 애즈는 선택이 아니라 필수이며, 마케팅 전략과 결과도 훨씬 정교하게 분석해야 해요.
## **에어브릿지와 라인야후 애즈의 연동, 광고 최적화와 리타겟팅까지 한 번에**
이번 에어브릿지와 라인야후 애즈의 연동을 통해, 일본 마케팅을 진행하시는 에어브릿지 고객사들은 이전보다 훨씬 정확하면서 객관적으로 마케팅 성과를 측정할 수 있어요.
좀더 쉽게 말씀드리자면, 어디에서 온 유저가 얼마나 가치 있는지를 정확하게 보고, 그 데이터를 가지고 자동 광고 최적화부터 리타겟팅까지 한 번에 진행할 수 있어요.
### **1\. 하이브리드 어트리뷰션을 통한 정확한 기여도 분석**
라인야후 애즈는 다양한 지면(뉴스, 채팅, 검색 등)에 광고가 노출되는데요. 에어브릿지는 머신러닝 기반의 인크리멘탈리티 모델과 멀티 터치 어트리뷰션을 활용하여, 매체 간 중복을 제거하고 라인야후 애즈가 전체 전환에 기여한 순수한 가치를 산출할 수 있어요.
### **2\. 실시간 포스트백을 통한 캠페인 및 ROAS 최적화**
에어브릿지 SDK가 수집한 앱 내 심층 이벤트(회원가입, 구독, 고액 결제 등)를 채널별 설치·회원가입·구매 같은 앱 이벤트를 한 화면에서 비교할 수 있게 해줘서 ROAS 판단을 훨씬 쉽게 할 수 있어요.
그리고 캠페인·광고그룹 등 매체 파라미터도 에어브릿지 리포트에 매핑되기 때문에 어떤 세트/키워드가 효율이 좋은지까지 볼 수 있어요.
이 과정에서 광고 플랫폼의 머신러닝이 고가치 유저를 학습하여 입찰가를 자동 최적화함으로써, 수동 운영 대비 낮은 CPA와 높은 ROAS를 달성할 수 있어요.
### **3\. iOS 비중이 높은 일본 시장, SKAN 대응의 필수성**
그리고 일본은 아이폰 점유율이 매우 높은 시장인데요. 에어브릿지는 라인야후 애즈의 SKAdNetwork(SKAN) 데이터를 통합 관리해요. SKAN 데이터도 에어브릿지로 바로 보낼 수 있어 iOS 환경에서도 성과 신호를 최대한 보존한 상태로 머신러닝 최적화를 할 수 있어요. 이를 통해 기술적 제약 없이 성과 가시성을 확보하고 데이터에 기반한 의사결정을 내릴 수 있어요.
### **4\. 퍼스트 파티 데이터 기반의 오디언스 리타겟팅**
에어브릿지 자체 데이터(예: "장바구니에 담았지만 구매하지 않은 사용자")를 라인야후 애즈 디스플레이 광고에 직접 동기화할 수 있어요. 이를 통해 원활한 크로스 채널 리타겟팅 세그먼트를 생성하거나, 기존 우수 고객과 유사한 일본 고객을 찾아 신규 유입 효율을 극대화할 수 있어요.
## **에어브릿지 & 라인야후 애즈와 함께 일본 마케팅 성공에 한 발자국 더 다가가세요**
일본 시장에서의 마케팅, 이제 라인야후 애즈는 선택이 아닌 필수예요. 그리고 라인야후 애즈와 에어브릿지의 연동은 여러분 비즈니스의 영향력과 확장세를 뒷받침하는 가장 강력한 데이터 엔진이 될 거예요.
지금 바로 에어브릿지 대시보드에서 라인야후 애즈를 연동하고, 데이터로 증명되는 일본 마케팅의 성과를 직접 확인해 보세요.
더 자세한 라인야후 애즈 연동 가이드는 아래 링크를 참고해 주세요.
_👉_[_\[에어브릿지와 연동하기\] 라인야후 애즈_](https://help.airbridge.io/ko/guides/yahoo-japan-ads)
**에어브릿지와 함께 성장하는 파트너들의 스토리를 직접 만나보세요!**
👉[나이틀리, 미기여 전환 측정으로 CPA 18% 줄이고 일본 앱스토어 Top 3 진입에 성공했어요](https://www.airbridge.io/ko/case-studies/nightly-ios-performance-measurement)
👉[강남언니, SKAN과 에어브릿지를 활용한 틱톡 리타겟팅 캠페인으로 CAC 51.8% 감소했어요](https://www.airbridge.io/ko/case-studies/gangnamunni-skan-titok-retargeting)
---
## 애플 서치 애드, 이제 에어브릿지에서 지면별 성과를 모두 확인하세요
URL: https://www.airbridge.io/ko/blog/apple-search-ads-placement-report
Language: Korean
Published: 2026-03-03
애플 서치 애드(Apple Search Ads)를 운영하고 있는데, MMP에서는 지면별 성과 정보를 제공하지 않아 자세한 성과가 궁금할 때마다 매번 애플 광고 대시보드를 오가고 계신가요?
이전까지 MMP를 사용하는 마케터는 애플 광고 지면별 성과를 확인하기 위해 매번 애플 대시보드에 접속해 데이터를 대조해야 했어요.
이제 에어브릿지에서는 이런 번거로움 없이 4가지 지면별 성과를 바로 확인할 수 있어요. 이 기능은 애플 최신 업데이트를 반영해 에어브릿지가 업계에서 가장 빠르게 지원하는 기능이에요.
**📌 핵심 포인트
**
- MMP 최초로 애플 서치 애드 4가지 지면별 성과 분석이 가능해졌어요.
- 번거롭게 애플 광고 대시보드에 다녀올 필요 없이 에어브릿지에서 세세한 성과 분석이 가능해요.
## 애플 서치 애드 성과 분석, 무엇이 달라졌나요?
애플 서치 애드란 애플 기기를 사용하는 유저가 앱스토어를 탐색할 때 광고를 노출하는 핵심 매체예요.
이번 업데이트를 통한 가장 큰 변화는 애플 광고 성과 데이터 분석의 편의성이에요. 애플은 2026년 1월 28일, MMP에서 광고 지면을 구분할 수 있는 필드를 새롭게 추가했어요.
에어브릿지는 이러한 애플의 변화를 발 빠르게 반영해 광고가 노출된 정확한 위치 정보를 대시보드에서 제공해요.
## 에어브릿지에서 애플 서치 애드 성과를 지면별로 분석하세요
[]
기존에는 모든 성과가 하나의 채널로만 집계되어 검색 결과에서 전환된 것인지, 투데이 탭에서 전환된 것인지 알 수 없었어요. 하지만 이제는 에어브릿지 리포트에서 [Sub Sub Publisher 2] 그룹을 적용해 아래 4가지 지면을 명확히 구분할 수 있어요.
- **검색 결과 (search_results):** 특정 키워드 검색 시 노출되는 페이지예요.
- **검색 탭 (search_tab):** 검색어를 입력하기 전 추천 앱 목록에 노출돼요.
- **투데이 탭 (today_tab):** 앱스토어 접속 시 가장 먼저 보이는 메인 화면이에요.
- **제품 페이지 (product_pages):** 다른 앱의 상세 페이지 하단에 노출돼요.
## 에어브릿지를 사용하는 마케터에게 어떤 점이 좋을까요?
### 애플 서치 애드 성과 분석이 더 정교해져요
단순히 광고 효율이 좋다는 결론을 넘어 구체적인 원인을 파악할 수 있어요. 예를 들어 투데이 탭은 브랜드 인지도를 높이는 데 효과적이고, 검색 결과는 실제 구매 전환율이 높다는 식의 구체적인 인사이트를 얻을 수 있어요.
### 업계 최신 기술 표준을 가장 먼저 사용할 수 있어요
이제 에어브릿지에서는 애플 대시보드를 오가지 않아도 성과가 저조한 지면의 비중은 줄이고, 효율이 높은 지면에 예산을 집중할 수 있어요.
2026년 2월 현재, 에어브릿지에서만 애플 서치 애드 지면별 성과 구분이 가능해요. 에어브릿지를 사용하면 누구보다 빠르게, 더 정교한 데이터를 통한 성과를 최적화할 수 있어요.
## 더 정확한 애플 광고 지면별 성과 분석, 에어브릿지에서 시작해 보세요
에어브릿지는 변화하는 애플 광고의 데이터 스펙을 가장 발 빠르게 반영하는 파트너예요. 에어브릿지 유저라면 애플 서치 애드 지면별 성과를 포함한 모든 광고 성과 데이터를 한 곳에서 관리할 수 있어요. 지금 바로 대시보드에서 애플 서치 애드의 상세 지면 데이터를 확인해 보세요.
---
## 업계 최초 국내 데이터 서버 런칭, 엔터프라이즈 및 금융사의 국내 규제 및 보안 요구를 충족하는 유일한 MMP를 만나보세요
URL: https://www.airbridge.io/ko/blog/seoul-region-launch-data-compliance
Language: Korean
Published: 2026-01-30
엔터프라이즈, 금융, 공공기관 등 규제에 민감한 산업군에서 마케팅 성과 측정 솔루션(MMP)을 선택하는 기준이 달라지고 있어요. 이제는 단순히 기능이나 서비스 완성도를 넘어서, 데이터가 어디에서 처리되는지, 국내 법·제도와 내부 보안 기준을 충족할 수 있는지, 그리고 실제 도입 과정에서 IT·보안·법무 검토를 통과할 수 있는지가 핵심 조건으로 자리 잡았어요.
이러한 흐름 속에서 에어브릿지는 업계 최초로 AWS 서울 리전 서비스를 공식 런칭했어요. 이번 국내 데이터 서버 증설을 통해, 국내 규제와 보안 기준을 전제로 한 데이터 처리 구조를 갖춘 유일한 MMP를 제공할 수 있게 되었어요.
## 데이터 처리 및 보안 기준이 강화되는 국내 환경
국내 금융사를 중심으로 개인정보 보호, 데이터 주권, 내부 보안 통제에 대한 요구 수준은 계속해서 높아지고 있어요. 특히 데이터 처리 위치와 관리 구조는 내부 보안 검토 과정에서 먼저 검토되는 요소 중 하나예요.
이메일이나 광고 식별자(ADID)처럼 보수적으로 해석할 경우 개인식별정보로 간주될 수 있는 데이터는, 기술적으로 문제가 없더라도 규제와 컴플라이언스 관점에서 까다롭게 검토될 수밖에 없어요.
에어브릿지는 이러한 환경 변화를 단순한 규제 리스크로 보는 것을 넘어서, 고객 신뢰를 핵심 경쟁력으로 삼기 위한 전략적 투자로 판단했어요. 해외 리전을 사용하는 구조에서는 내부 보안 및 IT 검토가 길어지고, 그 과정에서 솔루션 도입 자체가 지연되거나 중단되는 사례가 반복되어 왔어요.
에어브릿지는 이러한 현실적인 도입 장벽을 해소하기 위해 국내 규제 환경을 전제로 한 데이터 처리 구조를 다시 설계했고, 그 결과 업계 최초로 AWS 서울 리전 기반 서비스를 론칭했어요.
%%CTA|데모 신청하기|/ko/book-a-demo|국내 데이터 규제에 안심하고 대응할 수 있는 인프라를 확인해 보세요.%%
## 국내 서버 리전 증설로 완성한 안정적인 도입·운영 구조
AWS 서울 리전 구축 이후, 에어브릿지는 데이터 처리 위치와 관리 구조에 대한 고객사의 우려를 근본적으로 줄였어요. 이메일과 ADID 등 민감도가 높은 데이터 처리 영역에도 금융권 수준의 보안 기준을 적용해, 규제 민감 산업군에서도 안심하고 활용할 수 있는 데이터 처리 환경을 마련했어요. 그동안 해외 리전 사용으로 인해 발생하던 내부 보안·IT 검토 부담도 함께 낮아졌어요.
이 변화는 단순한 기술 개선을 넘어, 조직 내부의 의사결정 방식에도 영향을 미쳐요. 특히 보안 및 규제에 민감한 금융, 엔터프라이즈에서의 마케팅 솔루션 검토가 IT나 보안 단계에서 멈추지 않고, 실제 도입과 운영까지 이어질 수 있는 구조가 만들어졌어요. 그 결과, 서비스 도입 일정과 마케팅 실행 속도가 더욱 예측 가능해졌다는 점도 중요한 변화예요.
시장 반응 역시 이를 뒷받침하고 있어요. 공식 론칭 이전부터 금융 고객사를 중심으로 에어브릿지 도입 논의가 확대됐고, 주요 금융 기업들이 잇따라 에어브릿지 도입을 결정했어요. 보안과 규제 검토 기준이 가장 높은 금융사들이 선택했다는 점은, 강화되는 국내 규제 및 보안 환경 속에서 국내 유일 AWS 서울 리전 서비스의 실효성을 보여주는 명확한 신호로 볼 수 있어요.
또한 개인정보 보호와 데이터 처리 기준을 중요하게 보는 리테일 및 이커머스 기업에서도 더욱 안정적인 데이터 운영 환경을 기반으로 에어브릿지를 검토할 수 있는 조건이 마련됐어요. 국내 서버 리전은 다양한 산업군에서 에어브릿지를 선택할 수 있는 범위를 넓히는 계기가 되고 있어요
### 엔터프라이즈 대상 안정적인 이관 및 기술 지원
에어브릿지는 국내 서버 리전 증설과 함께 엔터프라이즈 고객을 위한 운영 지원도 강화하고 있어요. 금융 기업을 포함한 엔터프라이즈 고객을 대상으로 유료 고객 성공 지원 플랜을 제공하며, 전담 고객 성공 팀을 통해 SDK 연동과 마이그레이션 온보딩을 지원해요.
고객사의 기존 보안·운영 환경을 고려해 서비스 중단 없이 안정적인 이관이 가능하도록 가이드와 전문적인 기술 지원을 제공하고 있어요. 또한 본사와 개발팀이 한국에 있어, 실시간 대응은 물론 국내 환경에 맞춘 기능 설계까지 가능하다는 점이 에어브릿지만의 강력한 차별점이에요.
## 규제 환경 변화에도 변함없이 신뢰할 수 있는 파트너를 선택하세요
강화된 보안으로 인해 규제 요구가 매년 바뀌고 강화되는 환경에서는, 마케팅 솔루션 도입 이후에도 장기적으로 리스크 없이 운영할 수 있는지도 중요하게 고려해야 하는 요소 중 하나예요.
에어브릿지 국내 서버 리전 증설은 한국 시장의 법·제도·보안 요구를 전제로 설계돼, 안정적인 도입과 활용을 모두 고려했어요. 에어브릿지는 규제 환경 변화 속에서도 장기적으로 신뢰할 수 있는 파트너가 되고자 지속적으로 서비스를 고도화하고 있어요.
보안과 규제 이슈로 마케팅 성과 측정 솔루션 도입을 고민하고 있다면, 이제는 선택지가 분명해졌어요. 금융, 공공기관, 엔터프라이즈처럼 내부 보안·법무 검토가 필수적인 조직이라면, 에어브릿지가 어떤 대안이 될 수 있는지 전문가와의 미팅을 통해 알아보세요. 지금의 선택이 앞으로의 마케팅 운영 속도와 데이터 기반 성장의 방향을 크게 바꿀거에요.
---
## 네이버 애드부스트 오디언스 출시, 에어브릿지에서 더 정교한 네이버 광고 타겟팅을 경험해 보세요
URL: https://www.airbridge.io/ko/blog/naver-advoost-audience
Language: Korean
Published: 2026-01-05
네이버는 국내 최대 규모의 검색과 쇼핑 데이터를 보유한 독보적인 플랫폼이에요. 마케터들에게 네이버는 놓칠 수 없는 핵심 채널인 만큼, 최근 AI 기술을 접목해 광고 운영을 자동화하고 효율을 높여주는 ‘네이버 애드부스트(ADVoost) 오디언스’ 베타가 새롭게 오픈되면서 큰 주목을 받고 있어요. 에어브릿지 전환 데이터를 활용한 네이버 애드부스트 오디언스로 한 단계 더 정교한 네이버 광고 최적화를 경험해 보세요.
## 네이버 애드부스트 오디언스 알아보기
네이버 애드부스트는 AI가 타겟 설정부터 소재 제작, 운영까지 자동으로 최적화하는 스마트한 광고 솔루션이에요. 그중에서도 이번에 베타 서비스를 오픈한 [애드부스트 오디언스](https://ads.naver.com/help/faq/924)는 광고그룹 설정 단계에서 선택할 수 있는 AI 기반 자동 타게팅 기능이에요. 웹사이트 전환, 인지도 및 트래픽, 쇼핑 프로모션, 앱 전환 캠페인을 활용하시는 광고주님이라면 간단한 설정만으로 애드부스트 기술을 통해 광고주의 성과 목표에 적합한 사용자를 자동으로 찾아 타게팅함으로써 광고 성과를 최적화할 수 있어요.
애드부스트 오디언스를 활용하면서 얻을 수 있는 이점은 다음과 같아요.
- 애드부스트 오디언스는 광고주의 퍼스트파티 데이터와 네이버의 방대한 데이터를 결합하여, 목표 달성에 적합한 사용자를 자동으로 타게팅하는 서비스예요.
- 시스템이 실시간으로 성과를 학습하고 최적화하며, 복잡한 타겟팅 설정 없이도 효율적인 광고 집행이 가능해요.
- 광고주는 애드부스트 오디언스를 통해 다양한 캠페인 목적에 맞춰 네이버 전 영역에서 광고 효과를 극대화하고 운영 효율성을 높일 수 있어요.
## 에어브릿지와 네이버 애드부스트 오디언스
에어브릿지에서 수집된 전환 시그널이 네이버로 전송되면, 이를 기반으로 더욱 정교한 애드부스트 오디언스가 생성돼요. 네이버 애드부스트 오디언스는 에어브릿지가 전송한 데이터를 포함한 다양한 유저 행동 정보를 종합하여, 캠페인 목표에 가장 적합한 사용자를 스스로 찾아내 광고 효율을 극대화할 수 있어요.
구체적인 내용은 다음과 같아요.
- **정교한 오디언스 타겟팅과 최적화:** 앱 설치, 회원가입, 특정 금액 이상의 구매 등 특정 조건을 만족하는 전환 이벤트 세그먼트를 에어브릿지에서 생성해 네이버로 전송할 수 있어요. 이를 통해 리타겟팅, 디타겟팅(제외 타겟팅), 유사 타겟 확장을 자유롭게 실행할 수 있어요.
- **앱 이벤트를 활용한 AI 학습:** 단순 설치 데이터뿐만 아니라 앱 내에서 발생하는 다양한 이벤트를 포스트백으로 전송해 보세요. 네이버 AI가 ‘실제 구매할 가능성이 높은 유저’를 더 정밀하게 학습해 캠페인 효율을 높여줘요.
- **일 단위 실시간 데이터 업데이트:** 에어브릿지에서 전송되는 고객 데이터는 일 단위로 네이버에 실시간 업데이트되어요. 마케터는 항상 최신 유저 상태를 반영한 정교한 타겟팅 광고를 집행할 수 있어요.
- **운영 자동화와 효율성 증대:** 실시간 데이터를 바탕으로 AI가 최적화를 대신 수행하기 때문에, 마케터는 복잡한 수동 설정 대신 전략적인 의사결정에 더 집중할 수 있어요.
## 성공적인 애드부스트 오디언스 운영을 위한 체크리스트
애드부스트 오디언스는 AI가 주도적으로 학습하고 최적화하는 기능인 만큼, 몇 가지 주의 사항을 확인하면 더 좋은 성과를 얻을 수 있어요.
1. **전환 추적 설정은 필수예요**: 애드부스트 오디언스는 전환 데이터를 기반으로 동작하므로, 설정 여부가 AI 학습에 큰 영향을 미쳐요. 목표로 하는 전환 데이터가 에어브릿지를 통해 잘 수집되고 있는지 미리 점검해 주세요.
2. **충분한 시그널을 제공해 주세요**: 성별, 연령 같은 데모그래픽 정보부터 관심사, 잠재고객 세그먼트까지 다양한 시그널을 등록해 주는 것이 좋아요. 시그널이 부족하면 AI가 효율 지점을 찾는 데 더 많은 시간과 예산이 소요될 수 있어요.
3. **소재를 다양하게 준비해 주세요**: 애드부스트 오디언스는 시스템이 유저와 지면에 맞춰 자동으로 최적화하기 때문에, 게재 위치나 소재 선택 방식을 임의로 변경하기 어려워요. 대신 AI가 가장 적절한 소재를 골라 쓸 수 있도록 다양한 형태의 소재를 등록하는 것을 권장해요.
4. **타겟 제외는 꼭 필요한 경우에만**: 무분별한 타겟 제외는 AI 모델의 효과적인 학습을 방해할 수 있어요. 반드시 제외가 필요한 타겟만 신중하게 설정해 주세요.
5. **충분한 학습 기간을 기다려 주세요**: 광고그룹을 신규 등록한 후 AI가 최적화를 완료하기까지는 보통 1주일 이상의 시간이 필요해요. 초반 성과에 일희일비하기보다 충분한 학습 기간을 확보하는 것이 중요해요.
## 마무리하며
에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 카카오, 당근, 토스 등 주요 매체와의 연동을 지원하며 정교한 타겟 마케팅을 돕고 있어요. 여기에 더해 에어브릿지의 전환 데이터를 시그널로 활용해 더욱 정교한 네이버 애드부스트 오디언스를 만나볼 수 있게 되었어요.
퍼스트 파티 데이터의 가치가 그 어느 때보다 중요해진 지금, 에어브릿지의 데이터를 네이버 AI의 학습 시그널로 활용하는 것은 마케터에게 강력한 무기가 될 거예요. 지금 바로 에어브릿지 데이터를 네이버 AI와 결합해, 한층 고도화된 타겟 마케팅 성과를 경험해 보세요.
---
## 에어브릿지로 카카오 전환 API(CAPI) 연동하고 데이터와 전환 성과를 정확하게 측정하세요
URL: https://www.airbridge.io/ko/blog/airbridge-kakao-capi-integration
Language: Korean
Published: 2025-12-24
Author: Jaehyuk Kim
카카오 전환 API(CAPI: Conversion API)는 광고주가 보유한 1st Party 데이터를 광고주 서버에서 카카오 서버로 직접 전송(Server-to-Server)하는 기능으로, 유저의 전환 시그널을 안정적으로 확보하고 정확한 광고 성과 추적을 가능하게 해요.
카카오 전환 API를 사용하려면 내부 개발팀이 직접 Pixel & SDK ID와 인증토큰을 받아서 설정해야 하는데요. 이 과정에서 생각보다 개발 리소스가 많이 요구돼요.
[**에어브릿지**](https://www.airbridge.io/ko/pricing)는 국내 MMP 중 최초로 카카오 전환 API 연동을 지원하기 시작했어요. 이제 에어브릿지 대시보드에서 직접 카카오 전환 API 연동을 설정하고 관리할 수 있어서, 카카오 광고 성과를 더 쉽고 정밀하게 측정하여 광고 운영 효율을 최적화할 수 있어요.
## **카카오 전환 API란 무엇인가요?**
카카오 전환 API란 광고주 서비스에서 발생하는 식별 기반 유저 이벤트, 즉 광고주가 보유한 1st Party 데이터를 브라우저를 거치지 않고 **광고주 서버에서 카카오 서버로 직접 전송**하여 광고에 활용하기 위한 API예요.
[]
[]
> ✅ 전환 API의 주요 장점
> **1\. 정확한 성과 추적:** 클라이언트 환경의 제약을 받지 않기 때문에 쿠키 차단 이슈나 네트워크 오류로 인한 데이터 손실을 획기적으로 줄일 수 있어요.
> **2\. 광고 효율 개선:** 회원 정보를 기반으로 정교한 타게팅 및 효율적인 광고 학습이 가능해요.
> **3\. 안전한 데이터 전송:** 전송 시 인증 토큰을 사용하며, 회원의 개인정보는 SHA256 해시 처리된 상태로 안전하게 전송돼요.
### **Q. 전환 API가 Pixel & SDK와는 어떤 점이 다른가요?**
기존에 전환 API 대신 Pixel이나 SDK를 사용하시는 분들도 계실 텐데요. **전환 API와 Pixel & SDK의 차이점**을 간단하게 짚어드릴게요.
1. **전송 방식**
- 카카오 전환 API: 광고주 서버에서 카카오 서버로 직접 전송
- Pixel & SDK: 유저 브라우저/앱에서 전송
2. **개인정보 수집**
- 카카오 전환 API: 해시화된 이메일/연락처 등 식별 정보 수집 가능(ex. hem, hpn)하며 광고 추적 식별자(adid)나 쿠키(aid)에 관계없이 이벤트 수집 가능.
- Pixel & SDK: 현재는 픽셀&SDK 개인정보 수집이 불가하며 광고 추적 식별자, 카카오 광고 쿠키가 없는 경우 이벤트 수집 불가
3. **개인정보 보호 안정성**
- 카카오 전환 API: 인증 토큰 전송이 필요하며 개인정보 해싱(SHA-256) 전송 필수
- Pixel & SDK: 별도의 인증 없음
## **에어브릿지에서 카카오 전환 API를 빠르게 연동하세요**
**에어브릿지는 국내 MMP 중 최초로 카카오 CAPI 연동을 지원**하고 있어요.
에어브릿지와 카카오모먼트 전환 API 연동을 통해 에어브릿지에서 수집한 이벤트를 카카오모먼트에 전송해요. 광고주는 전송받은 이벤트로 카카오모먼트 캠페인을 최적화할 수 있어요.
### **카카오 전환 API와 에어브릿지를 연동하는 방법**
1. 카카오비즈니스 대시보드에서 Pixel & SDK ID와 인증토큰을 발급받고 연동 정보를 확인하세요.
2. 에어브릿지 대시보드에서 발급받은 Pixel & SDK ID와 인증 토큰을 직접 등록하세요.
3. 에어브릿지 대시보드에서 카카오 모먼트에 전송할 이벤트 매핑을 설정하세요. 에어브릿지 스탠다드 이벤트가 카카오 표준 이벤트에 매핑돼요.
4. Web/App의 adPersonalization 필드 수집이 활성화되어 있는지 확인하세요. 수집이 활성화된 항목에 대해서만 이벤트가 카카오로 전송돼요. 아직 수집 전이라면 해당 값을 수집하도록 담당 개발자에게 요청하세요.
5. 연동 활성화 토글을 켜면 카카오 모먼트로 전환 이벤트 전송이 시작돼요. 연동 현황(Integration Overview) 페이지에서 카카오 CAPI EPC를 확인하며 연동 상태를 모니터링할 수 있어요.
👉 에어브릿지 카카오 전환 API 연동에 관한 더 자세한 가이드는 [에어브릿지 Help Center 카카오모먼트](https://help.airbridge.io/ko/guides/kakao-moment#%EC%A0%84%ED%99%98-api-%EC%97%B0%EB%8F%99%ED%95%98%EA%B8%B0)에서 확인하실 수 있어요.
### **Q. 에어브릿지로 카카오 전환 API를 연동하면 어떤 점이 좋은가요?**
에어브릿지를 활용하여 카카오 전환 API를 연동할 때 얻을 수 있는 이점은 다음과 같아요.
1. **개발 부담을 최소화할 수 있어요.**
- MMP를 통해 연동함으로써, 광고주가 직접 서버 연동을 위한 인프라 구축 및 유지보수 작업을 수행할 필요가 줄어들어요.
2. **대시보드에서 직접 연동을 관리할 수 있어요.**
- 에어브릿지 대시보드에서 연동 정보 등록, 이벤트 매핑, 활성화/비활성화를 직접 제어할 수 있어요. 별도의 요청 없이 바로 설정하고 연동 현황을 실시간으로 확인할 수 있어요.
3. **에어브릿지 팀이 신속하게 대응해 드려요.**
- 에어브릿지 팀은 다양한 국내외 매체와의 연동을 지원하고 있어 대응이 가장 빠르다는 강점이 있어요. 카카오 전환 API와 관련된 변경 사항이나 업데이트가 있으면 신속하게 대응할 수 있어요.
4. **에어브릿지에서 1st Party 데이터를 통합 관리하세요.**
- 에어브릿지를 통해 수집된 앱/웹 이벤트가 카카오 표준 이벤트로 매핑되어 전송되므로, 확보된 1st Party 데이터를 기반으로 카카오 모먼트 광고 타게팅 및 최적화에 활용할 수 있어요. adPersonalization 수집이 활성화된 항목(App/Web)에 대해서만 이벤트가 전송되므로, 개인정보 보호 기준에 맞는 데이터만 안전하게 활용할 수 있어요.
## **카카오 전환 API 연동 전 이건 꼭 챙겨주셔야 해요.**
에어브릿지를 통한 CAPI 연동을 진행하기 위해서는 카카오비즈니스 환경에서 아래와 같은 꼭 사전 준비가 필요해요.
[]
### **1\. 개인정보 위수탁 및 마케팅 활용 동의 확인**
전환 API가 유저의 식별 정보를 전송하므로, 광고주 서비스에서 발생하는 회원의 개인정보를 카카오로 전송한다는 위수탁 계약이 전제되어야 해요. 또한, **회원의 이메일이나 전화번호를 수집**해야 하므로, 개인정보를 광고에 활용하기 위해 반드시 마케팅 활용 동의를 받으셔야 해요. 광고주는 마케팅 활용에 동의한 유저의 이벤트(에어브릿지 SDK상 adPersonalization에 해당)만 카카오로 전송해야 해요.
### **2\. Pixel & SDK ID 발급 및 연동 확인**
API 전송을 위해 Pixel & SDK ID 발급이 필요해요. 정확한 성과 비교 및 광고 학습 효율을 위해 기존에 사용하시던 Pixel & SDK ID를 그대로 사용하는 것을 권장드려요. 단일 Pixel & SDK로 많은 이벤트를 수집할수록 광고 최적화 학습을 위한 시드 규모가 증가하기 때문이에요. 또한, 캠페인 집행 시 전환 추적을 위해 광고계정과 Pixel & SDK ID 연동이 완료되어야 해요.
## **데이터 안정성을 확보하고 광고 최적화 효율을 높이세요**
에어브릿지에서 카카오 전환 API를 연동하면, 전환 API와 Pixel & SDK 데이터 간의 중복 없이 통합한 성과를 확인하실 수 있어 신뢰도 높은 데이터를 기반으로 전략적인 의사 결정을 내릴 수 있어요.
카카오 전환 API와 에어브릿지 연동을 통해 데이터 기반의 정확하고 효율적인 카카오 모먼트 캠페인을 구축해 보세요.
더 자세한 내용이 궁금하시다면 **지금 바로 에어브릿지 데모를 신청**해서 자세한 상담을 받아보세요.
**참고자료)**
[카카오 Conversion API, 카카오비즈니스 가이드](https://kakaobusiness.gitbook.io/main/tool/pixel-sdk/conversion-api)
[카카오모먼트, 에어브릿지 Help Center](https://help.airbridge.io/ko/guides/kakao-moment#%EC%A0%84%ED%99%98-api-%EC%97%B0%EB%8F%99%ED%95%98%EA%B8%B0)
---
## 몰로코 향상된 웹 포스트백 연동, 에어브릿지에서 똑똑한 웹 최적화를 시작하세요
URL: https://www.airbridge.io/ko/blog/moloco-enhanced-web-postback-integration
Language: Korean
Published: 2025-12-23
에어브릿지가 몰로코와 함께 향상된 웹 포스트백(Advanced Web Postback)을 지원해요. 에어브릿지의 통합 측정과 사용자 수준 분석이 몰로코의 강력한 웹 광고 성능과 결합되어, 웹 퍼널 전반에서 더 정확한 성과 측정과 정교한 최적화를 기대할 수 있어요. 특히 금융사 및 커머스의 웹투앱 캠페인이나 PC/콘솔 게임처럼 웹 랜딩페이지에서 설치나 가입으로 이어지는 흐름이 중요한 경우, 이번 연동이 성과 개선에 더 큰 효과를 줄 수 있어요.
이제 에어브릿지를 활용하는 몰로코 웹 광고주들은 포스트백으로 전달되는 서드파티 쿠키 ID를 활용해, 최신 개인정보 보호 기준을 부합하면서도 사용자 단위로 캠페인을 측정하고 분석하며 최적화할 수 있어요. 이번 연동에서의 서드파티 쿠키 기반 식별은 안드로이드와 PC에서만 적용돼요.
향상된 웹 포스트백 연동은 아주 간단해요:
- 에어브릿지에서 원하는 전환 목표에 맞게 [몰로코](https://help.airbridge.io/ko/guides/moloco) 포스트백을 설정해요
- 에어브릿지 [Web SDK](https://help.airbridge.io/ko/developers/web-sdk)에서 몰로코 쿠키 ID 수집 기능을 활성화해요
- 에어브릿지 대시보드에서 사용자 단위 이벤트와 캠페인 성과를 분석하기 시작해요
## 몰로코와 함께 성과 중심의 웹 성장에 속도를 더하세요
몰로코는 AI 기반 광고 플랫폼으로, 전 세계 광고주들이 가치 높은 사용자를 얻고 유지하도록 지원하고 있어요. 또한 투명한 데이터를 제공해서 마케터가 직접 어느 지면을 통해서 유저들이 획득되고 있는지 모니터링하고 캠페인을 최적화 할 수 있도록 지원해요.
기존의 일반적인 웹 연동 방식은 이벤트 정보와 전환 정보만 전달하기 때문에 캠페인 성과에 대한 분석은 가능해도, 시그널 전송을 통한 캠페인 최적화는 어려웠어요. 하지만 이제 몰로코와 함께 향상된 웹 포스트백을 지원하면서, 광고주는 전체 퍼널에서 사용자 단위 데이터에 접근할 수 있게 됐어요.
덕분에 광고주는 다음과 같은 세밀한 작업이 가능해요:
- 이벤트/전환 정보와 몰로코의 서드파티 쿠키 ID를 함께 확인해요
- 서드파티 쿠키 기반의 사용자 단위 시그널에 기반한 더 정밀한 최적화를 실행해요
- 다양한 유저 그룹(cohort), 채널, 경로별로 성과를 전체적으로 분석해요
- 전환 이후 실제 행동을 기반으로 더 스마트한 입찰 전략과 자동 조정을 실행해요
- 순수 웹 & 웹투앱 캠페인의 성과를 동시에 확인해요
## 에어브릿지로 웹 포스트백 측정을 강화하세요
에어브릿지는 쪼개진 데이터를 하나로 합치고, 정확한 성과를 검증하고, 광고 노출부터 LTV까지 모든 터치포인트를 연결하는 데 특화된 크로스플랫폼 전용 MMP예요. 몰로코의 머신러닝 최적화와 에어브릿지의 투명한 측정이 만나면, 마케팅 팀은 다음과 같은 효과를 얻을 수 있어요:
- 웹 성과를 완전히 투명하게 보기: 몰로코 웹 트래픽과 이벤트를 다른 매체 성과와 함께 쉽게 비교할 수 있어요
- 통합된 추적을 쉽게 사용하기: 에어브릿지 트래킹 링크 하나로 웹과 앱의 모든 활동을 추적할 수 있어요
- 웹 SDK와 연결해 딥링크 오픈, 웹 이벤트, 웹 기여 설치(web-assisted install), 웹투앱 전환까지 한 곳에서 분석할 수 있어요
- 웹 캠페인 제대로 최적화하기: 서드파티 쿠키 기반의 사용자 단위 시그널을 통해서 웹 전환에 대한 최적화를 자연스럽게 달성할 수 있어요
- UTM 추적 자동화하기: 웹 SDK가 링크 속 UTM 값을 자동으로 읽어 캠페인 소스를 정확하게 기록할 수 있어요
## 마무리하며
에어브릿지와 몰로코의 향상된 웹 포스트백 연동은 투명한 데이터, 머신러닝 기반 최적화, 강력한 퍼널 분석을 한 번에 원하는 그로스 팀을 위한 솔루션이에요. 특히 웹 랜딩페이지 최적화로 유입 효율을 높이고, 그 결과가 웹투앱 전환과 리타겟팅 성과로 어떻게 이어지는지까지 한 흐름으로 관리하고 싶다면 이번 연동을 활용해 보세요. 웹 유저 확보와 리타겟팅에서 더 높은 성과를 만들고 싶다면, 에어브릿지와 몰로코 연동을 활용해 보세요.
---
## ASO의 기본: 앱 제목 최적화
URL: https://www.airbridge.io/ko/blog/aso-app-title
Language: Korean
Published: 2025-12-23
앱 제목 최적화는 앱 스토어 내 앱의 순위 변동 그리고 가시성 증대에 굉장히 중요한 역할을 합니다. 그렇다면 앱 제목을 어떻게 구성해야 최적화를 잘했다고 할 수 있을까요? 이번 포스트에서는 앱 스토어 최적화(ASO)의 기본 과정 중 하나인 ‘앱 제목 최적화’에 대해서 다뤄보도록 하겠습니다.
우선 앱 스토어 최적화(ASO)란, 앱 스토어 내의 검색 결과에서 더 높은 순위에 노출될 수 있도록 최적화하는 과정으로, 앱 페이지로 더 많은 트래픽을 유도하여 다운로드를 더 많이 수행하는 것을 목표로 합니다.
앱 제목은 앱의 이름과 브랜딩을 보여줄 수 있는 공간이지만, 그 이상의 잠재력을 가지고 있습니다. App Store와 Google Play의 앱 제목은 키워드 인덱싱(keyword indexation) 측면에서 가장 큰 비중을 차지합니다. 따라서 검색 가시성과 키워드 최적화 효과를 극대화하려면 철저한 조사와 분석을 거쳐 앱 제목을 작성해야 합니다.
다만, 많은 앱 개발자는 앱 제목에 브랜드 이름만 추가해야 하는지, 아니면 추가 키워드를 포함시켜 글자 수 (App Store와 Google Play 모두 30자)를 완전히 최적화해야 하는지 궁금해합니다.
이것에 대한 답을 얻기 위해 대한민국 앱 스토어 내 각 카테고리에서 무료 앱 중 상위 100개를 살펴보았습니다. 그리고 실제로 그 결과, 카테고리별 인기 앱이 검색 결과에서 앱 순위와 가시성을 높이기 위해 앱 제목에 앱과 연관성이 높은 일반 키워드를 활용하고 있음을 발견했습니다.
** 브랜드 이름 옆에 일반 키워드가 추가된 앱 제목의 비율을 기호로 구분하여 계산했습니다.*****
## 일반 키워드로 제목을 최적화하는 앱 살펴보기 - 안드로이드
[]
**안드로이드에서는 데이트, 소셜 및 집 & 홈 카테고리의 앱이 주로 앱 제목의 브랜드 이름 옆에 일반 키워드를 추가하는 경향이 있는 것으로 나타났습니다. **
이러한 데이트, 소셜 및 집 & 홈 카테고리의 대부분은 자체적으로 설명할 수 없는 짧은 브랜드 이름을 가지고 있으며 스토어 방문객들이 해당 브랜드에 대해 들어본 적이 없는 경우에는 흥미를 잃거나 혼란스러워할 수 있습니다. 이러한 경우 브랜드 이름 옆에 일반 키워드를 추가하면 사용자가 **앱의 기능을 더 잘 이해하고 필요에 가장 적합한 앱을 선택**할 수 있습니다.
반면 스포츠 및 캐주얼 게임 하위 카테고리(및 이벤트 카테고리)는 제목에 일반 키워드를 추가하는 앱의 점유율이 가장 낮습니다. 여기서 캐주얼 게임의 상위 차트를 살펴보면, 일반적으로 "Royal Match" 및 "메이플스토리M"와 같은 자기 설명적인 아이콘과 브랜드 이름을 가지고 있음을 알 수 있습니다.
스포츠 게임 카테고리와 같은 경우 일반적으로 "야구", "UFC"와 같이 브랜드 이름 자체 내에서 게임의 종류를 명시하거나 "EA"나 "FIFA"와 같이 브랜드가 유명한 스포츠 게임입니다. 결과적으로 이러한 카테고리에서는 앱의 기능을 강조하거나 다운로드 장려를 위해 제목의 일반 키워드의 중요도는 낮아진다는 것을 알 수 있습니다.
## 일반 키워드로 제목을 최적화하는 앱 살펴보기 - iOS
[]
**안드로이드와 유사하게, 그래픽 및 디자인, 사진 및 동영상과 같은 및 App Store의 생산성 카테고리는 제목에 일반 키워드가 추가된 앱의 점유율이 높았습니다. **”포커,” “바둑," “퍼즐” 등 브랜드 옆에 특정 유형의 게임을 언급하는 경우에도 마찬가지입니다.
브랜드 중심의 제목을 가진 앱이 대부분을 차지하는 카테고리는 게임입니다. 안드로이드와 마찬가지로 레이싱, 스포츠, 시뮬레이션 및 캐주얼 게임 제목은 타이틀을 브랜드 이름으로만 채우는 경향이 있었습니다.
그렇다면 앱/게임 카테고리와 일반 키워드로 앱 제목을 최적화하는 경향 사이에 상관관계가 있을까요? 제목에 일반 키워드를 추가할지 여부를 결정할 때 어떤 것들을 검토해야 할까요?
## 앱 제목에 브랜드 이름만 이용했을 때 얻을 수 있는 이점
앱 제목을 선택할 때, 일부 앱 개발자들은 시간과 자원을 절약하기 위해 브랜드 이름만 이용하는 것에 만족합니다. 앱 개발자들의 **앱 제목 최적화 측면에서 가장 중요한 관심사는 전환율입니다.** 일부 앱 마케팅 담당자들은 브랜드 이름 옆에 일반 키워드를 추가하면 유저가 키워드로만 채워진 제목을 낮은 품질의 앱과 연관시킬 수 있기 때문에 **브랜드 이미지와 독점성에 부정적인 영향**을 미칠 수 있다고 주장합니다.
브랜드 이름만을 사용한 앱 제목의 또 다른 이점은 일관성입니다. 유저의 혼란을 줄이고 앱 설치 광고 캠페인에서 앱 마켓으로 이동하는 과정과 같이 앱이 표시되는 모든 마케팅 채널과 연관성을 보이며 일관되고 원활한 과정을 제공할 수 있습니다.
## 앱 제목에 일반 키워드를 추가할 경우의 이점
### 검색 트래픽 및 가시성 증가
앱 스토어 최적화(ASO)의 관점에서 앱 제목에 대용량의 관련 키워드를 추가하면 **검색 트래픽과 도달 범위가 증가**할 수 있습니다. 특히 제목이 두 스토어 모두에서 인덱싱 알고리즘에 가장 큰 비중을 차지하기 때문입니다.
### 앱 설치 전환율 향상
앱의 특징과 기능을 설명하기 위해 키워드를 추가하면 스토어 방문자가 **앱의 가치를 보다 명확하게 이해할 수 있으므로 **앱의 전환율이 향상될 수 있습니다. 특히 Google Play에서 일반 키워드를 검색했을 경우 짧은 설명과 스크린샷이 검색 결과에 표시되지 않으므로 제목과 아이콘이 더 이해하기 쉬워야므로 일반 키워드르 추가하는 것이 더 좋을 수 있습니다.
아래는 Google Play(대한민국)에서 "소개팅"이라는 키워드에 대한 검색 결과의 예시입니다. 일부 사용자는 데이트 카테고리 내 모든 브랜드에 익숙하지 않을 수 있으므로 일반 키워드를 포함하는 것도 전환에 도움이 될 수 있습니다.
[]
### 유저(및 스토어 알고리즘)가 앱의 기능을 더 잘 이해할 수 있도록 지원
앱 제목에 있는 일반 키워드는 잠재적인 유저와 스토어 알고리즘 측면에서 **앱의 전반적인 인식을 개선**할 수 있습니다. 메타데이터를 완전히 최적화하면 앱 스토어가 앱을 더 잘 이해할 수 있으므로 알고리즘이 앱을 더 관련된 검색 및 찾아보기 섹션에서 재분류할 수 있습니다.
## 효과적인 앱 제목을 만들 때 고려해야 할 사항
### 브랜드 인지도
앱 제목을 만들 때 가장 먼저 고려해야 할 요소는 **브랜드 인지도**입니다. 일반 스토어 방문자 상대로 우리 앱 브랜드 이름이 얼마나 인지도가 높은가요? 브랜드 이름이 인지도가 높고 독립적일수록 일반 키워드의 필요성이 줄어듭니다. 이러한 경우 앱 스토어 최적화 (ASO) 도구를 사용하여 현재 앱 이름 및 각 키워드가 앱에 가져오는 다운로드 수를 추정해야 합니다.
### 앱 제목의 공간 활용
두 번째 검토 요소는 앱 제목에 일반 키워드를 추가 할 수 있는 **글자 공간이 충분한지 여부**입니다. 활용할 수 있는 공간이 남을 경우 제목에 추가할 수 있는 잠재적 키워드 목록을 만들고 **해당 메트릭스를 서로 비교**합니다.
잠재력이 높은 키워드의 우선 순위를 지정하는 데 도움이 되는 주요 메트릭은 **볼륨, 가능성 및 관련성 점수**입니다. 추가할 키워드 선택 후 메타데이터 최적화 기능을 사용하여 우리 앱의 예상 트래픽을 계산할 수 있습니다.
### 우리 앱 브랜딩
이 단계에서는 브랜드 마케팅 팀과 브랜드 가이드라인에 대한 논의를 시작합니다. 이를 통해 타겟팅하려는 메시지와 가장 일치하는 키워드와 앱으로 트래픽을 유도할 수 있는 키워드에 대해 합의점을 찾으실 수 있습니다.
새롭게 업데이트된 앱 제목의 버전이 있으면 스토어에서 메타데이터를 테스트하고 업데이트할 시간입니다. 이후에는 작업의 영향을 측정하기 위해 **최소 2주 정도 기다릴 것을 권장**합니다. 다운로드 및 카테고리 순위에 미치는 영향을 보다 명확하게 파악하기 위해 콘솔을 앱트위크에 통합해 보실 수 있습니다.
## 앱 제목에 대한 일반 가이드라인
### App Store에 대한 권장 사항:
App Store에서는 앱 제목에 **최대 30자**를 사용할 수 있습니다. 앱 제목에 브랜드 이름이 포함된 후에는 가능한 많은 관련 키워드를 위해 남은 공간을 모두 사용합니다.
### Google Play에 대한 권장 사항:
App Store와 마찬가지로 Google Play는 앱 제목을 **30자로 제한**합니다. Google Play에서 앱의 제목과 브랜드 이름을 명확하고 정확하게 사용합니다. 업데이트된 Google 메타데이터 정책(2021)에서는 더 이상 앱 제목에 스토어 성능, 순위, 수상 내역 또는 실행 요청을 나타내는 키워드를 사용할 수 없다는 것 또한 참고해야 합니다.
앱 제목은 앱 메타데이터의 중요한 요소입니다. 이번 포스트에서 Apple Store 및 Google Play 상위 차트의 데이터를 사용하여 상위 앱 중 많은 앱이 중요한 키워드를 앱 제목에 활용하여 앱 순위와 검색 결과의 가시성을 높이는 데 도움이 된다는 것을 살펴보았습니다.
이번 포스트에서 얻은 인사이트를 바탕으로 우리 앱의 앱 스토어 최적화를 하고 싶다면, [앱트위크 툴](https://www.apptweak.com/ko?utm_source=airbridge&utm_medium=referral&utm_campaign=optimizing-your-app-title)을 확인해보세요. 앱트위크는 앱 가시성 증대 및 전환율 확보를 위한 앱 스토어 최적화(ASO) 툴을 제공합니다. ASO 툴 그리고 ASO 전문가와 함께 지금 바로 우리 앱의 키워드 최적화부터 평점 및 리뷰 관리까지 하실 수 있습니다.
👉 **앱트위크 툴 무료 체험판 바로 이용하기: **[**링크**](https://www.apptweak.com/ko/pricing?currency=usd&utm_source=airbridge&utm_medium=referral&utm_campaign=optimizing-your-app-title)
---
## 엑솔라 서버 간 연동, 이제 에어브릿지에서 모바일·웹 게임 성과를 통합 분석하세요
URL: https://www.airbridge.io/ko/blog/xsolla-airbridge-integration
Language: Korean
Published: 2025-12-17
**에어브릿지(Airbridge)**가 [**엑솔라(Xsolla)**](https://xsolla.com/)와 파트너십을 체결했어요. 이번 파트너십을 통해 모바일 캠페인 데이터와 엑솔라 웹숍 결제 데이터를 연결하는 새로운 서버 간(S2S: Server to Server) 연동 기능을 선보이게 됐어요.
최근 더 많은 유저가 **웹숍**을 통해 인게임 재화나 아이템, 전용 특가 상품을 구매하면서, 플랫폼 외 결제가 전체 매출에서 차지하는 비중이 급격히 늘고 있어요. 하지만 대다수 개발사에게 웹 결제 데이터를 모바일 캠페인 성과와 연결하는 작업은 그동안 매우 까다로운 과제였어요.
이번 연동으로 유저가 웹숍에서 아이템을 구매하면, 그 정보가 서버를 통해 엑솔라에서 에어브릿지로 즉시 전달돼요. 덕분에 해당 유저가 과거에 어떤 광고를 보고 앱을 설치했는지, 혹은 어떤 캠페인을 통해 웹으로 넘어와 결제했는지 그 기여도를 빈틈없이 찾아낼 수 있어요.
**이제 개발사는 모바일 광고가 웹 매출에 기여하는 바를 정확히 측정하고, 광고 클릭부터 앱 설치, 최종 구매에 이르는 전체 유저 여정을 완벽하게 파악할 수 있어요.**
## **엑솔라(Xsolla)는 어떤 기업인가요?**
엑솔라는 2005년 설립된 글로벌 비디오 게임 커머스 전문 기업으로, 전 세계 게임 개발사와 퍼블리셔가 게임을 성공적으로 출시하고 수익을 극대화할 수 있도록 돕는 강력한 도구와 서비스를 제공하고 있어요. 전 세계적으로 밸브(Valve), 로블록스(Roblox), 에픽게임즈(Epic Games), 크래프톤(KRAFTON) 등 글로벌 대형 게임사들을 파트너로 보유하고 있어요.
### **엑솔라의 주요 비즈니스**
- **D2C 솔루션:** 엑솔라의 핵심 서비스인 웹숍은 개발사가 앱스토어나 플레이스토어와 같은 플랫폼 마켓의 높은 수수료(최대 30%)를 벗어나, 약 5% 수준의 합리적인 수수료로 유저에게 직접 아이템과 재화를 판매할 수 있는 환경을 제공해요.
- **글로벌 페이먼트 솔루션:** 신용카드, 전자지갑 등 전 세계 200개국 이상에서 통용되는 700개 이상의 현지 결제 수단을 지원하여 글로벌 진출을 도와줘요.
- **게임 런처 및 사이트 빌더:** 개발사가 독자적인 PC 게임 런처를 만들거나 게임 홍보 및 판매를 위한 웹사이트를 손쉽게 구축하도록 도구를 제공해요.
- **크로스 플레이 & 크로스 페이:** 모바일 게임을 PC나 웹, 클라우드로 확장하여 사용자가 어떤 기기에서든 동일한 계정과 결제 경험을 유지하도록 지원해요.
- **법적 및 세무 업무 지원:** 엑솔라가 공식 판매자 역할을 수행하며 전 세계의 복잡한 세금 계산, 규정 준수(GDPR 등), 광고 사기 방지 업무를 대신 처리해줘요.
## **엑솔라 연동으로 성과 분석과 전략 수립을 동시에**
이번 연동을 통해, 모바일과 웹 데이터를 하나로 통합하여 성과를 분석하고 성장 전략을 수립할 수 있는데요. 폭발적인 성장을 원하는 게임 스튜디오라면 에어브릿지와 엑솔라를 활용해서 다양한 성과를 달성해 보세요.
1. **웹숍 구매와 인앱 이벤트 통합 트래킹:** 웹에서 발생한 결제 데이터를 적절한 캠페인에 어트리뷰션하고, 유저가 앱과 웹을 어떻게 넘나드는지 확인하세요.
2. **전체 캠페인 매출 측정:** 웹 결제 전환을 일으키는 고가치 유저가 어떤 채널과 소재를 통해 유입되었는지 식별할 수 있어요.
3. **진정한 유저 LTV 계산:** 인앱 및 웹 데이터를 결합해 총 생애 가치(LTV)를 측정하고 광고 집행 비용을 최적화하세요.
에어브릿지는 풀퍼널 어트리뷰션과 200여 개의 맞춤형 지표를 통해 ROAS, 리텐션, 매출, 이탈률을 한곳에서 모니터링할 수 있도록 지원해, 데이터 사일로를 제거하고 의사결정 속도를 높일 수 있어요.
## **웹숍 구매 데이터, 에어브릿지에서 인앱 이벤트와 함께 트래킹하세요**
많은 게임사가 웹숍과 같은 D2C 모델을 도입하며 가격 정책, 브랜딩, 유저 관계에 대한 통제력을 높여가고 있어요. 하지만 이 과정에서 발생하는 가장 큰 문제가 바로 데이터 파편화예요.
에어브릿지는 엑솔라 웹숍 구매 데이터를 인앱 이벤트와 함께 트래킹하기 때문에, 이러한 데이터 공백을 메우는 데 매우 효과적이에요. 모든 결제 데이터가 성과 지표에 누락 없이 반영되며, 게임 개발사는 마케팅 활동이 모바일과 웹 양쪽에서 실제 매출로 어떻게 이어지는지 명확하게 확인할 수 있어요.
또한, [**에어브릿지의 예측 생애 가치(pLTV)**](https://www.airbridge.io/ko/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app)** 및 **[**예측 생애 기간(pLT)**](https://www.airbridge.io/ko/blog/global-ai-app-insight-seminar-airbridge-appgrowing)** 모델링**을 통해 고가치 유저를 조기에 식별하면, 성과 하락 전 이탈 징후를 예측하여 선제적으로 대응할 수 있어요.
## **크로스 플랫폼 성장 전략, 에어브릿지와 함께 하면 어렵지 않아요 **
에어브릿지와 엑솔라의 연동으로, 앞으로 게임 개발사들은 플랫폼을 넘나드는 유저 인게이지먼트와 매출에 대해 통합된 데이터 기반의 인사이트를 제공받을 수 있어요.
ROAS, LTV 및 유저 행동에 대한 인사이트를 바탕으로 UA, 프로덕트, 데이터 팀의 의사결정을 일치시키고 장기적인 성장을 이끌어내 보세요.
### **에어브릿지와 함께 하는 게임사들의 성공 스토리가 궁금하시다면? 케이스 스터디에서 직접 만나보세요!**
👉[메이플스토리M, MMM 솔루션으로 디지털 옥외 광고 성과 가시성을 확보했어요](https://www.airbridge.io/ko/case-studies/nexon-maplestorym-mmm)
👉[퍼스트 디센던트, 이제 PC와 콘솔 게임도 모바일 게임처럼 광고 성과를 측정하고 최적화할 수 있어요](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)
👉[뮤 모나크, 소재 효율 최적화로 로그인율을 90% 이상 개선했어요](https://www.airbridge.io/ko/case-studies/mu-monarch-creative-optimization)
---
## 2026 비게임 앱 트렌드: 데이터로 본 APAC 국가별 앱 성장 전략
URL: https://www.airbridge.io/ko/blog/2026-apac-app-trends
Language: Korean
Published: 2025-12-12
APAC 비게임 앱 시장은 하나의 통일된 흐름으로 설명하기 어려운, 국가별 특성이 뚜렷하게 갈리는 방향으로 성장하고 있어요. 각국의 문화, 소득 수준, 디지털 인프라가 크게 다르다 보니 성장하는 카테고리도, 상위권에 오르는 앱도 전혀 다른 양상을 보이고 있죠. 유저가 앱을 소비하는 방식 또한 각국의 생활 습관과 밀접하게 연결되어 있어, 비게임 앱의 글로벌 성장을 목표로 한다면 국가별 트렌드와 성장 공식을 이해하는 것이 점점 더 중요해지고 있어요.
이번 콘텐츠에서는 한국, 일본, 중국, 베트남, 인도 등 APAC 25개국의 비게임 앱 데이터를 지난 5년간 종합 분석해, 국가별 생활 방식과 디지털 환경이 어떻게 서로 다른 모바일 앱 성장 구조를 만들었는지 정리했어요. 2026년 APAC 비게임 앱 시장을 이끌 주요 트렌드와 새로운 기회를 한눈에 파악해보세요.
\*모바일 및 디지털 경제 인텔리전스 플랫폼 [센서타워(Sensor Tower)](https://sensortower.com/ko)의 앱 IQ 기준으로 분류한 데이터를 바탕으로 에어브릿지에서 작성한 콘텐츠입니다.
## APAC 비게임 앱 다운로드 및 인앱 구매 수익 트렌드
### 글로벌 증가세 속, APAC과 인도가 주도하는 비게임 앱 다운로드 성장
[__wf_reserved_inherit]
글로벌 총 다운로드 & APAC 점유율
2025년 글로벌 비게임 앱 다운로드는 꾸준한 상승세를 이어가고 있어요. 2021년부터 2024년까지 비게임 앱 다운로드 수는 지속적으로 증가했고, 2025년에는 10월 기준 이미 전년 수준에 근접하면서 연말에는 약 935억 건으로 역대 최고치를 기록할 것으로 예상돼요.
[__wf_reserved_decorative]
APAC 국가별 총 다운로드 수
APAC 역시 절대적인 다운로드 규모에서 꾸준한 성장세를 유지하고 있어요. 2025년에는 약 430억 다운로드로 역대 최고 기록을 경신할 전망이며, 이는 글로벌 전체의 약 45% 이상을 차지하는 규모예요. 다만 APAC의 다운로드 비중은 2021년 48.7%에서 2025년 45.3%로 완만하게 감소하고 있어요.
이 변화는 미국이 2025년 8월 기준 월 다운로드 최고치를 달성하고, 브라질이 2023년 11월 이후 중국(iOS)을 제치며 글로벌 다운로드 3위 시장으로 올라선 영향이 크게 작용한 결과예요. 즉, APAC은 여전히 가장 큰 비게임 앱 소비 권역이지만, 글로벌 시장이 점차 다국가 중심 구조로 재편되는 모습이에요.
그중에서도 인도는 글로벌 비게임 앱 다운로드 1위 국가로서의 위상을 유지하고 있어요. 지난 5년간 APAC 비중 내 약 40% 내외 수준을 꾸준히 차지했으며, 2025년 9월에는 최근 5년 중 최고 월 다운로드 기록을 경신했어요. 폭넓은 인구 기반, 디지털 결제·이커머스 생태계 확장 등이 결합되면서 인도는 APAC을 넘어 글로벌 비게임 앱 다운로드의 핵심 동력으로 자리잡고 있어요.
### 강한 성장세 속 완만한 비중 감소, 변화하는 APAC 인앱 구매 시장
[__wf_reserved_decorative]
글로벌 총 인앱 구매 수익 & APAC 점유율
2021년부터 2024년까지 글로벌 비게임 인앱 구매 수익은 약 390억 달러에서 약 700억 달러로 급성장했어요. 2025년에도 이러한 흐름은 이어져 10월 기준 작년 동기 대비 약 20% 높은 수준을 기록했고, 연말에는 약 830억 달러로 역대 최고치를 달성할 전망이에요.
APAC 역시 절대 금액 기준으로 꾸준한 성장세를 보이고 있어요. 2021년 141억 달러에서 2024년 196억 달러로 상승했고, 2025년에는 약 240억 달러로 또 한번 최고치를 경신할 전망이에요. 다만 글로벌 대비 APAC의 점유율은 36.5%에서 28%로 감소하고 있어요. 이는 APAC이 성장세를 유지하고 있음에도 북미, 유럽, 라틴 아메리카의 유료 전환 속도가 더 가파르게 증가한 결과로 분석돼요.
[__wf_reserved_decorative]
APAC 국가별 인앱 구매 수익
국가별로 보면 중국(iOS)과 일본이 여전히 비게임 인앱 구매 수익 1·2위를 견고하게 유지하고 있어요. 특히 한국은 2023년부터 호주를 추월해 3위로 상승하며 아시아 주요 유료 소비 시장으로 확고히 자리잡았어요. APAC 상위 5개 국가의 수익 집중도는 2021년 88%에서 2025년 83%로 완만하게 줄어든 반면, 동남아 지역의 수익 비중은 7%에서 11%로 확대되며 특정 국가 중심이던 수익 구조가 점차 더 많은 국가로 확산되는 흐름을 보이고 있어요.
이 가운데 한국 시장의 성장이 특히 두드러져요. 한국의 인앱 구매 수익은 지난 5년간 꾸준히 상승했고, 2025년 10월 기준 월 수익 최고치를 경신했어요. 이러한 증가세는 전년 동기 대비 10배 이상 성장한 ChatGPT의 폭발적 유료 구독자 전환 증가가 큰 영향을 미친 것으로 나타났어요.
## 비게임 앱 장르 트렌드
### 다운로드 기준: AI 기반 생성형 앱이 주도하는 새로운 성장 축
[__wf_reserved_decorative]
다운로드 기준 카테고리 TOP5
##### 지난 5년간 APAC 비게임 앱 시장에서는 소프트웨어, 미디어 및 엔터테인먼트, 금융 서비스, 쇼핑, 여행 및 관광 카테고리가 꾸준히 다운로드 기준 상위 5위를 차지해왔어요. 이 중에서도 소프트웨어, 미디어 및 엔터테인먼트 카테고리가 가장 높은 비중을 유지하며 APAC 비게임 앱 다운로드 시장을 주도하고 있어요. 일상적인 커뮤니케이션, 콘텐츠 소비, 생산성 활동이 모바일 중심으로 재편되면서 두 카테고리의 영향력은 5년간 변함없이 견고한 상태죠.
카테고리별 대표 앱 변화를 살펴보면, 2021년에는 캡컷(CapCut), 줌(Zoom), 왓츠앱(WhatsApp) 등이 소프트웨어 카테고리의 상위권을 이끌었어요. 하지만 2025년에는 흐름이 뚜렷하게 변화했어요. 챗GPT(ChatGPT), 캡컷, 구글 제미나이(Google Gemini)가 Top 3를 차지하며, AI 기반 생성형 앱의 급부상이 소프트웨어 카테고리 전체의 성장을 견인하고 있어요. 이 변화는 APAC 전역에서 콘텐츠 생성, 검색, 생산성 도구의 중심이 AI 중심 구조로 빠르게 이동하고 있다는 신호로 읽을 수 있어요.
### 인앱 구매 기준: 엔터테인먼트를 추격하는 소프트웨어 카테고리의 폭발적 성장
[__wf_reserved_decorative]
인앱 구매 수익 기준 카테고리 TOP5
지난 5년간 APAC 비게임 인앱 구매 시장에서는 미디어 및 엔터테인먼트, 소프트웨어, 직업 및 교육, 건강 및 웰니스, 라이프스타일 및 서비스가 꾸준히 수익 상위 5개 카테고리를 유지해왔어요. 이들 카테고리는 구독 기반 서비스 확대, 디지털 소비 패턴 고도화, 자기계발·웰빙 트렌드의 확산 등을 배경으로 안정적인 지출 구조를 형성하고 있어요.
미디어 및 엔터테인먼트 카테고리는 APAC 수익의 최상단을 오랜 기간 지켜온 핵심 영역이에요. 2021년에는 틱톡, 카카오 픽코마, WeTV, iQIYI, 라인망가 등이 상위권을 차지했다면, 2025년에는 틱톡, WeTV, 유튜브, QQ뮤직, 카카오 픽코마가 상위에 오르며 시장 구도가 변화했어요. 특히 틱톡의 견고한 수익 유지력과 유튜브의 뚜렷한 성장세가 눈에 띄어요.
소프트웨어 카테고리는 2021년 12%에서 2025년에는 20%를 넘어서는 폭발적인 성장세를 기록하며 인앱 구매 시장의 두 번째 핵심 축으로 자리 잡았어요. 2025년 10월 기준 약 5억 3천만 달러로 사상 최고 수익을 기록했고, 상위 앱 구도 역시 크게 변화했어요. 2021년에는 구글 원(Google One), Baidu Wangpan, UC 브라우저가 상위권을 형성했지만, 2025년에는 구글 원, 챗GPT, 캡컷이 Top3를 차지했어요.
특히 챗GPT의 폭발적 성장세는 소프트웨어 카테고리 지형을 뒤흔든 핵심 요인이에요. 2025년 5월에는 월 수익 기준 처음으로 구글 원을 추월했고, 이후 총 네 차례 1위를 기록하며 APAC 비게임 인앱 구매 시장에서 가장 빠르게 성장하는 앱 중 하나로 부상했어요.
## APAC 비게임 앱 순위
### 다운로드 기준 비게임 앱 TOP10
[__wf_reserved_decorative]
APAC 다운로드 순위 TOP10
ChatGPT는 APAC 전체 다운로드 1위를 기록하며 정보 탐색·학습·생산성 영역에서 핵심 앱으로 자리 잡았어요. 한국, 일본, 인도에서도 Top3 이내에 공통적으로 포함되며, APAC 전역에서 “AI 퍼스트” 소비 행태가 본격화되고 있음을 보여줘요. 일본과 한국에서는 구글 제미나이도 함께 상위권에 진입해, 챗GPT와 양강 구도를 형성하고 있어요.
[__wf_reserved_decorative]
한국, 중국, 일본, 인도, 동남아시아 다운로드 순위 TOP10
#### 한국: AI + 이커머스 + 디지털 공공 서비스 주도
한국에서는 네이버플러스 스토어가 다운로드 3위를 기록하며 올해 이커머스 시장에 새로운 활력을 불어넣었어요. 테무(Temu) 역시 2023년 출시 이후 2025년에도 5위를 유지하며 글로벌 이커머스 성장세를 이어가고 있어요. 또한 디지털 신분증·건강보험증 등 공공 서비스 앱이 상위권에 진입하며, 정부 행정 서비스의 디지털 전환 수요가 크게 증가하고 있음을 보여줘요.
#### 중국: 로컬 AI 앱 돌풍과 숏폼 드라마의 강세
중국에서는 Doubao(豆包)가 1위에 오르며 중국형 챗GPT로 자리 잡았어요. 중국 로컬 대형언어모델(LLM) 기반 챗봇 딥시크(DeepSeek)가 3위를 기록했으며, 이 외에도 夸克(Quark), 腾讯元宝(Tencent Yuanbao) 등 주요 AI 앱이 TOP10에 포함되며 AI 생태계가 빠르게 확산되는 모습이에요.
콘텐츠 소비 측면에서는 红果短剧(Hongguo Short Drama)가 7월 월 다운로드 최고치를 경신하며 중국 내 숏폼 드라마 앱 인기를 주도했어요. 바이트댄스(Bytedance)는 총 4개 앱을 TOP10에 올리며, 2025년 비게임 다운로드 기준 중국 iOS 퍼블리셔 1위를 차지했어요.
#### 일본: AI·결제·리테일 기반의 생활형 앱 주도
일본에서는 ChatGPT와 Google Gemini가 각각 1위와 3위를 기록하며 AI 기반 정보 탐색 및 생산성 앱의 급성장을 보여줬어요.
한편, 楽天ペイ(라쿠텐페이), マツキヨコカラ(마츠키요코카라), ENEOS 등 생활·결제·리테일 중심 앱이 상위권을 형성하며 일본 특유의 옴니채널 생활 서비스 생태계를 반영하고 있어요.
#### 인도: 소셜 강세 속 AI와 숏폼 앱 부상
인도에서는 인스타그램, 페이스북, 왓츠앱, 스탭챗 등 소셜 앱이 상위권을 장악하며 소셜 중심의 디지털 소비 구조를 보여줘요. AI 기반 앱 중에서는 ChatGPT만이 유일하게 상위권에 진입하며 독보적 존재감을 유지했어요. 또한 Kuru TV가 3위에 오르며, 짧고 간편한 콘텐츠 소비 트렌드가 인도 시장에서도 확산되고 있음을 확인할 수 있어요.
#### 동남아: 숏폼 + AI 투톱 구조
동남아에서는 TikTok과 CapCut이 각각 1위와 5위를 기록하며 숏폼 동영상 중심의 콘텐츠 소비가 여전히 강력한 모습이에요. ChatGPT도 2위에 오르며, 동남아에서도 AI 기반 앱 중 독보적인 존재감을 보이고 있어요.
### 인앱 구매 수익 기준 비게임 앱 TOP10
[__wf_reserved_decorative]
APAC 인앱 구매 수익 순위 TOP10
APAC 전체에서는 틱톡이 작년에 이어 올해도 수익 1위를 기록하며, APAC 대표 소셜 엔터테인먼트 플랫폼으로서의 지위를 공고히 했어요. 글로벌 기준으로는 미국이 틱톡 수익의 27.4%로 1위지만, 중국 iOS가 25.9%로 근소한 차이를 보이며 2위를 차지해 APAC 지역 내 강력한 수요를 다시 확인할 수 있어요.
유튜브는 한국·일본 등 주요 국가에서 꾸준히 상위권을 유지하며 APAC 전체 3위를 기록했어요. 또한 챗GPT는 비게임 부문에서 유일하게 APAC 상위권에 진입한 생성형 AI 앱으로, 구독형 AI 서비스의 본격 확산을 상징하는 사례로 자리 잡았어요.
[__wf_reserved_decorative]
한국, 중국, 일본, 인도, 동남아시아 인앱 구매 수익 순위 TOP10
#### 한국: AI 구독 시장의 대형 소비국으로 부상
한국에서는 ChatGPT가 APAC 중 유일하게 수익 1위를 기록했어요. 글로벌 기준으로도 미국에 이어 2번째로 높은 수익 비중을 차지해 한국이 AI 구독형 서비스의 초기 확산 시장임을 보여줘요.
YouTube는 한국 내에서 수익 2위를 기록했으며, 글로벌 수익 비중에서도 미국(46.7%), 일본(8.6%)에 이어 3위(4.4%)를 차지했어요.
또한 네이버 웹툰·카카오페이지·시리즈 등 국산 웹툰·웹소설 플랫폼 3종이 모두 상위권에 진입하며 스토리 IP 콘텐츠의 견고한 수익 기반을 재확인했어요.
#### 중국: 영상·음악 중심의 강력한 엔터테인먼트 수익 구조
중국에서는 WeTV, iQIYI, 优酷(Youku) 등 동영상 스트리밍 앱이 상위권을 차지하며 엔터테인먼트 소비가 수익의 핵심축 역할을 하고 있어요. 또한 QQ音乐, 网易云音乐, 酷狗音乐 등 음악 스트리밍 앱도 높은 수익 기여도를 보였어요.
흥미로운 점은 구직 플랫폼 BOSS直聘이 7위에 오르며 비엔터테인먼트 앱의 유료화 잠재력도 확인된 점이에요.
#### 일본: 한국 IP 기반 스토리 콘텐츠의 저력
일본에서는 카카오 픽코마(Piccoma)와 라인망가(Line Manga)가 각각 1위와 2위를 기록하며 웹툰·웹소설 콘텐츠의 강력한 존재감을 보여줬어요. 두 앱은 APAC 전체 수익 순위에서도 7위, 9위에 올라 한국 IP 기반 스토리텔링 콘텐츠의 글로벌 경쟁력을 입증했어요.
또한 틱톡, 유튜브, U-NEXT, 라인 망가, 라인, 라인 뮤직 등 라인 생태계 및 글로벌 영상 플랫폼이 상위권에 오르며 일본 시장의 디지털 소비 흐름을 안정적으로 지탱하고 있어요.
#### 인도: 클라우드·AI 중심의 유료 소비 확대
인도에서는 구글 원이 1위, 챗GPT가 2위를 기록하며 클라우드 스토리지와 AI 구독형 서비스 중심의 인앱 구매 구조가 형성되고 있어요. 또한 링크드인(3위)이 상위권에 올랐다는 점은 인도에서 B2B·커리어 기반 서비스의 유료화 잠재력이 매우 크다는 것을 시사해요.
아마존 쇼핑(Amazon Shopping)과 프라임 비디오(Prime Video)가 상위권에 오른 것도, 커머스와 콘텐츠를 결합한 프라임 생태계가 인도에서 강력한 영향력을 행사하고 있음을 보여줘요.
#### 동남아: AI·크리에이티브 도구 앱 존재감 확대
동남아에서는 틱톡이 수익 1위를 차지하며 소셜·엔터테인먼트 중심 소비 시장임을 다시 입증했어요. 캡컷, 캔바와 같은 크리에이티브 도구형 앱이 상위권에 오르며, 동남아 지역에서 콘텐츠 소비뿐 아니라 콘텐츠 제작 수요가 확대되고 있음을 확인할 수 있어요. 또한 챗GPT가 3위를 기록하며 AI 기반 구독형 서비스가 동남아에서도 빠르게 확산되고 있어요.
## 모바일 앱의 글로벌 성장, 에어브릿지와 함께하세요
에어브릿지는 더 많은 모바일 앱이 글로벌 시장에서 빠르게 성장할 수 있도록 지원하고 있어요. 에어브릿지와 함께 마케팅 캠페인 성과, 유저 여정, UA·리텐션 지표 등을 한눈에 파악하고, 데이터 기반의 의사결정을 더욱 정교하게 만들어보세요
2026년 글로벌 시장에서의 성장을 고민하고 있다면, 지금 바로 전문가와 데모를 신청해보세요.
---
## 2026년 금융 마케팅 전략, 이제는 효율을 넘어 행동과 AI로 나아가세요
URL: https://www.airbridge.io/ko/blog/financial-marketing-strategy-with-efficiency-action-ai
Language: Korean
Published: 2025-12-08
2026년이 코앞으로 다가왔어요. 하루가 다르게 변하는 마케팅 트렌드 속에서, 특히 규제와 효율 사이를 줄타기해야 하는 금융 마케터분들의 고민은 더욱 깊으실 텐데요.
지난 11월 20일, AB180이 주최한 **Where Finance Meets Growth: Airbridge VIP Dinner** 행사에서는 금융권 마케터분들을 모시고 내년도 금융 비즈니스 성장을 위한 마케팅 트렌드와 전략을 나누는 자리를 가졌어요.
이 자리에 몰로코, 네이버, NNT까지, 업계 최고의 전문가들이 모여, 앞으로 금융 마케터가 어떤 기준으로 의사결정을 해야 하는지, 그리고 어디에 집중해야 하는지에 대한 실질적인 인사이트를 공유했어요.
월요일 저녁임에도 높은 참석률을 기록해 데이터 기반 성장에 대한 금융업계의 높은 관심을 다시 한 번 확인할 수 있었는데요. 그날의 핵심 인사이트, 지금 바로 정리해 드릴게요.
[]
[]
## **AB180: 금융 파트너의 성장을 만드는 에어브릿지 **
첫 번째 순서는 AB180 대표 남성필 님이 오프닝 연사로 나서며 금융권 마케팅 최신 트렌드와 에어브릿지의 한국 서버 신설 소식을 전했어요.
[]
AB180의 자체 데이터 분석 결과, 최근 금융권은 보상형 **광고의 선호 현상**과 **숏폼 광고의 성장**, **DSP**(Demand Side Platform: 온라인 광고를 집행하는 광고주나 광고 대행사가 광고 인벤토리나 캠페인 등 광고 지면을 관라하기 위해 사용하는 소프트웨어 플랫폼)의 확산, **AI 결합 광고 상품의 등장**, **웹투앱 활용 증가** 등의 흐름이 두드러지고 있다고 했어요.
웹과 앱이라는 두 가지 플랫폼을 동시에 운영해야 하는 금융 마케팅의 어려움에 공감하며, 특히 네이버 검색 광고를 통해 들어온 유저가 앱을 설치하고 계좌 개설까지 이어지는 과정을 끊김 없이 추적하는 것이 중요하다고 강조했어요.
> “웹은 따로, 앱은 따로 보는 게 아니라 하나의 리포트에서 통합해서 분석할 수 있어야 합니다. 이것이 크로스 플랫폼 분석의 가장 중요한 선결 요건입니다.”
이러한 흐름 속에서 금융사들은 웹과 앱을 넘나드는 크로스 플랫폼 트렌드와 금융 서비스 간 경쟁 심화, 개인정보보호 및 규제 준수 등 여러가지 과제를 맞닥뜨리고 있어요.
[]
남성필 님은 이를 해결하기 위해 **에어브릿지에서 웹 & 앱 통합 분석 기능과 네이버와 몰로코 광고 시스템 등 타 매체들과의 심화 연동 기능을 제공**하고 있다고 말씀했어요.
그리고 이 자리에서 에어브릿지의 **한국 지역 서버 증설 소식**을 전했어요. 기존에도 ISMS과 ISO27001 인증 등 엔터프라이즈급 보안을 유지하고 있었는데요. 한국 서버를 신설하면서, 앞으로 보안이 무엇보다 중요한 금융사들의 심의와 소비자 보호 등의 논의를 더욱 긴밀하고 빠르게 지원해 드릴 예정이에요.
마지막으로 AI를 활용한 효율적인 광고 성과 방법과 기술에 관해 공유했는데요. 앞으로 에어브릿지가 금융 파트너들의 성장을 위해 준비한 로드맵을 확인할 수 있는 시간이었어요.
[]
## **몰로코: 효율을 넘어 행동으로의 전환이 필요한 시기**
두 번째 세션에서는 [몰로코](https://www.moloco.com/ko) 시니어 매니저 유동권 님이 참석하여 기존 금융 마케팅의 한계를 짚어주셨어요.
몰로코(Moloco)는 AI 기반의 퍼포먼스 마케팅 플랫폼으로, 국내외 주요 금융사들의 마케팅 파트너로 함께하고 있는데요. 현재 대부분의 금융사가 설치 수나 CPC 같은 단기 KPI 관리는 매우 잘하고 있지만, 장기적 성장으로 이어지는 예치·거래·반복 이용 같은 고가치 행동을 만드는 데는 어려움을 겪고 있다고 설명했어요.
여기서 유동권 님은 이제는 ‘누가 설치했는가’보다 **‘누가 계속 쓰는가’를 학습하는 전략이 필요하다**고 강조했어요.
> “효율 중심 캠페인은 단기 성과에는 용이하지만, 많은 고객사 분들이 기대하는 수준의 진성 유저(Active User)를 남기지 못한다는 명확한 한계가 있습니다.. 이제 효율 중심에서 ‘행동 중심의 캠페인’ 설계 전략으로 넘어가야 할 시점입니다. 얼마나 많은 유저들이앱을 설치했는지를 넘어서 어떤 유저들이 앱을 활발하게사용하는지를 기준으로 캠페인을 최적화 해 나가는게 중요합니다”
내부 보안 절차 때문에 예치금이나 거래 액수 데이터를 공유하기 어려우신 경우에도, 행동 시그널(거래 발생, 예치 완료, 반복 거래 등)만 있으면 에어브릿지와 같은 MMP를 활용해서 충분히 진성 유저를 찾을 수 있다고 설명했어요. 비슷한 산업군인 가상자산 업계를 보면, 거래 볼륨이나 활성 트레이더 비율 등 행동 중심 캠페인으로 이미 변화하고 있다고해요.
이를 위해, **몰로코는 AI 타겟팅 기술을 적극 도입**하고 있는데요. AI가 브랜드 세이프티와 맥락, 유저의 장기적 가치를 모두 고려하여 가장 적절하다고 판단되는 진성 유저에게만 광고를 노출해요.
또한 금융 마케팅에서 가장 중요한 요소인 데이터 보안, 안정성, 투명성도 강조했는데요. 구글·애플이 승인한 앱에서만 광고가 집행되며, TAG·IAB Gold 등 국제 인증 기반 브랜드 세이프티, 광고 운영 데이터 제공 등 금융사가 신뢰할 수 있는 데이터 보안 구조를 갖췄다고 강조했어요.
마지막으로 유동권 님은 설치와 가입이라는 1차원적 목표를 넘어, 거래와 예치를 활발히 반복하는 진성 유저를 확보하는 것이 내년도 성장의 핵심임을 다시 한 번 강조했고, 그 여정에 몰로코와 에어브릿지가 든든한 파트너가 되겠다고 다짐하며 세션을 마무리했어요.
[]
## **네이버: AI와 데이터로 확장되는 네이버의 금융 마케팅 전략**
[네이버 광고](https://business.naver.com/service.html) 세일즈 뉴비즈 팀 노제헌 님은 네이버 광고 생태계의 확장을 핵심 키워드로 제시하면서, 외부 오디언스 연동, 외부 미디어 확장, DOOH 진출 등 플랫폼 밖으로 생태계를 넓히는 전략을 공개했어요.
[]
네이버는 에어브릿지와의 오디언스 연동을 통해 타겟팅 정합성을 끌어올리고 있다고 해요. 실제로 금융권 광고주(증권사, 핀테크사 등)에게서 효율이 자연스럽게 개선되고 있다는 평가를 받고 있어요.
금융 업종은 검색 광고 의존도가 매우 높지만 검색량 자체에는 한계가 있다는 점을 짚으며, 노제헌 님은 다음과 같이 조언했어요.
[]
> “검색량 자체가 드라마틱하게 변하지 않는다면 볼륨 확장에 한계가 존재합니다. 디스플레이 광고가 브랜드 인지를 높여 검색량 증가까지 만드는 선순환 구조를 만들어야 합니다.”
특히 내년도 주목해야 할 변화로 **네이버 AI 광고와 신규 솔루션들**을 소개해 주셨어요.
1. **네이버 AI 광고 전면 확대: **ADVoost 쇼핑/검색/크리에이티브/오디언스 등 네이버의 AI 기반 광고 솔루션이 확대돼요.
2. **NEW 카탈로그 광고: **광고주 DB 직접 연동 방식의 카탈로그 광고(네이버다이내믹광고)도 내년부터 업종이 확대돼요.
3. **TVC ✕ 네이버 크로스 캠페인 분석 솔루션: **TVC 집행 전후의 검색량 변화, 광고 성과 업사이드 등을 분석한 인사이트 리포트를 제공해요.
4. **ADVoost Screen: **파편화되어 있던 옥외광고(극장, 교통, 주거 등) 성과를 온라인 광고처럼 타겟팅하고, 예산을 유연하게 설정하여 네이버 광고 플랫폼에서 성과를 측정할 수 있어요.
노제헌 님은 AI를 통해 광고주의 성과와 편의성을 동시에 높이면서, 네이버 광고 생태계를 확대하는 것이 네이버의 지향점이라고 강조했어요.
앞으로 상품과 기능, 그리고 다양한 프로모션과 행사를 통해 새로운 마케터들의 업무 편의성을 높이는 것은 물론, 온오프라인을 아우르는 통합적인 성과 분석을 가능하게 할 것이라며 네이버의 새로운 도약을 예고했어요.
[]
## **NNT: 검색 엔진의 종말과 GEO 시대의 생존 전략**
마지막 세션에서는 디지털 마케팅 에이전시 [NNT](https://www.nnt-consulting.com/) 대표 조경상 님이 요즘 핫한 주제인 **GEO(Generative Engine Optimization: 생성형 AI 검색 엔진 최적화)**에 관한 인사이트를 공유하셨어요.
GEO란 ChatGPT나 퍼플렉시티(Perplexity), 제미나이(Gemini)와 같은 생성형 AI 검색 엔진이 내 브랜드를 **인용**하고 추천할 수 있게 브랜드 아이덴티티와 웹사이트 등 다양한 항목을 최적화하는 일련의 작업들을 말해요.
[]
최근 생성형 AI 검색이 새로운 트렌드로 자리 잡으면서 검색 시장의 판도가 완전히 바뀌고 있다고 강조했는데요. 내년이나 내후년이면 우리가 알던 전통적인 검색 엔진은 사라질 것이라며, 이제는 키워드로 검색하는 시대가 아니라 **AI에게 질문하고 답을 얻는 대화형 검색의 시대**가 올 것이라고 전망했어요.
특히 한국은 챗GPT MAU가 2천만 명을 넘는 매우 특수한 시장이라는 점에서 AI 기반 노출 경쟁이 더 빨리 심화될 것이라고 분석했어요.
> “저는 검색 엔진은 내년, 늦어도 내후년이면 거의 없어지기 시작할 거라 생각합니다. 절대로 기술은 뒤로 돌아가지 않습니다. 이제 구글 AI 오버뷰(Overview)나 챗GPT에서 우리 브랜드가 인용되게 만드는 싸움을 준비해야 합니다.”
[]
이제 검색 결과 내 클릭 경쟁이 아니라 **브랜드 가시성(Visibility)** 경쟁으로 넘어가고 있으며, 마케터들은 상위 노출을 노리는 SEO를 넘어 AI가 우리 브랜드를 답변에 인용하도록 하는 GEO 전략을 준비해야 해요.
AI에게 선택받기 위해서는 무엇보다 좋은 콘텐츠가 필수적인데요. 그럼 좋은 콘텐츠란 무엇일까요? 조경상 님은 **디테일하거나, 혹은 우리만 가진 독창적인 데이터나 인사이트를 담은 콘텐츠**를 좋은 콘텐츠로 정의했어요. 아래 예시로 설명해 드릴게요.
- **기존 SEO 글쓰기 예시:** 포르쉐 911 GT3는 강력한 6기통 엔진과 듀얼 클러치로 파워풀한 주행을 제공합니다.
- **AEO·GEO 글쓰기 예시:** 포르쉐 911 GT3는 최고출력 510마력의 4.0L 자연흡기 수평대향 6기통 엔진을 탑재했고, 7단 PDK 듀얼 클러치 변속기를 통해 서킷에서 뛰어난 랩 타임을 보여줍니다.
단순히 포르쉐가 빠르다고 쓰는 것이 아니라 구체적인 엔진 스펙과 마력을 명시하는 것처럼 압도적인 디테일이나 독자적인 데이터를 담아야 해요. 또한 다양한 브랜드의 GEO 사례를 소개하며 AI가 신뢰할 수 있는 정보 구조를 만들어야만 가시성을 확보할 수 있다고 강조했어요.
## **2026년 새로운 금융 마케팅, 함께 도약할 준비 되셨나요?**
이렇게 이번 행사에서는 규제 준수를 넘어, 데이터 기반 성장 전략으로 금융업계가 어떤 가능성을 만들 수 있는지 실제 사례와 함께 함께 살펴보았어요.
AB180과 몰로코, 네이버, 그리고 NNT가 입을 모아 이야기한 것은 결국 **변화와 본질**이었어요. AI 기술과 검색 환경은 급격하게 변하고 있지만, 그럴수록 진짜 행동하는 유저를 찾고 고객에게 유용한 정보를 전달하는 마케팅의 본질은 더욱 중요해지고 있죠.
다가오는 2026년, 금융 마케팅의 성장은 데이터를 얼마나 통합적으로 보고, AI라는 도구를 얼마나 영리하게 활용하느냐에 달려 있을 거예요.
이번 세미나가 그 고민을 정리하는 작은 실마리가 되었길 바라며, 바쁘신 와중에도 귀한 시간 내어 참석해주신 모든 분들께 다시 한 번 감사드려요. 에어브릿지는 앞으로도 성장 파트너로서 금융 마케팅의 새로운 가능성에 함께할게요.
---
## 메타 AEM, 메타 광고 노출 성과(View-Through Attribution)도 이제 에어브릿지에서 확인하세요
URL: https://www.airbridge.io/ko/blog/meta-aem-view-through-attribution
Language: Korean
Published: 2025-12-04
에어브릿지가 이제 메타 iOS 앱 프로모션 캠페인 중 AEM(Aggregated Event Measurement) 캠페인에 대한 노출 기반 어트리뷰션을 지원하게 됐어요. 이전에는 클릭 기반 전환만 확인할 수 있었지만, 이제는 광고 노출이 설치에 얼마나 영향을 주는지도 함께 볼 수 있죠. 덕분에 메타 캠페인 성과를 더 완전하고 더 정확하게 분석할 수 있어요.
이 업데이트는 2025년 10월 27일부터 시작하는 AEM 적용 가능 iOS 캠페인에 자동으로 반영돼요.
## AEM(Aggregated Event Measurement)이란 무엇인가요?
[AEM](https://www.facebook.com/business/help/721422165168355?id=1877298665783613)은 iOS 14.5 이상 기기에서 발생하는 웹/앱 이벤트를 유저 프라이버시를 보호하는 방식으로 측정할 수 있게 해주는 메타의 프로토콜이에요. 실시간에 가까운 리포팅을 제공해서, 캠페인 최적화를 더 빠르게 진행하고 성과를 더 세밀하게 해석할 수 있도록 도와줘요.
## AEM 캠페인에서 노출 기반 어트리뷰션 데이터를 보려면 무엇을 해야 하나요?
이미 AEM이 활성화된 캠페인을 운영 중이라면 추가로 설정할 건 없어요. 아직 AEM을 활성화하지 않았다면, 에어브릿지 AEM [연동 가이드](https://help.airbridge.io/ko/guides/meta-business)를 참고해서 설정해 주세요.
## 앱 프로모션 캠페인을 만들 때 AEM을 직접 선택해야 하나요?
아니요. 따로 선택하지 않아도 돼요. 에어브릿지는 AEM 적용 대상 캠페인이라면 노출 기반 전환 데이터를 자동으로 수신해요. 내 앱 이벤트가 AEM 적용 대상인지 확인하려면 Meta 비즈니스 지원 센터를 참고해 주세요.
## 이번 업데이트로 에어브릿지에서 무엇이 달라지나요?
이제 에어브릿지는 기존 클릭 기반 어트리뷰션에 더해 노출 기반 터치포인트(view-through touchpoint)까지 자동으로 수신하고 리포팅해요. 즉, 광고를 봤지만 클릭하지 않은 유저의 설치도 아래 영역 전반에서 확인할 수 있어요.
- 어트리뷰션 리포트(Attribution reports)에서 확인할 수 있어요.
- 분석 대시보드(Analytics dashboards)에서도 함께 볼 수 있어요.
- 원본 데이터 내보내기(Raw data exports)에도 노출 기반 터치포인트가 반영돼요.
참고로 SKAN 데이터는 계속 별도로 수집·어트리뷰션되고, 이번 업데이트로 SKAN 리포트에는 변화가 없어요.
## 어트리뷰션 윈도우
메타는 설치에 대해 최대 1일(24시간) 노출 기반 어트리뷰션 윈도우를 지원해요. 에어브릿지는 전 세계 기준으로 6시간의 확률적(probabilistic) 노출 기반 전환 윈도우를 적용해요.
리포트 간 차이를 줄이려면 에어브릿지 대시보드에서 메타 광고의 어트리뷰션(전환) 규칙을 조정할 수 있어요.
단계별 설정 방법은 [AEM 연동 가이드](https://help.airbridge.io/ko/guides/meta-business-channel-integration#ios-%EC%BA%A0%ED%8E%98%EC%9D%B8%EC%97%90-%EC%B7%A8%ED%95%A9%EB%90%9C-%EC%9D%B4%EB%B2%A4%ED%8A%B8-%EC%B8%A1%EC%A0%95-aem-%EC%82%AC%EC%9A%A9%ED%95%98%EA%B8%B0)에서 확인할 수 있어요. 만약 AEM 연동과 관련된 도움이 필요하다면 에어브릿지 지원팀에 문의해 주세요.
---
## 2026년 글로벌 앱 마케팅, 데이터 예측과 크리에이티브로 승부하세요
URL: https://www.airbridge.io/ko/blog/global-ai-app-insight-seminar-airbridge-appgrowing
Language: Korean
Published: 2025-12-04
AI 앱 시장은 어느 순간부터 새로운 카테고리가 아니라, 모바일 산업의 중심축이 되어버렸어요. 텍스트·이미지·영상 생성 기술은 몇년 사이에 놀라울 만큼 발전했고, 광고 생태계는 더 빠르게 바뀌고 있죠.
게다가 AI 기술의 발전과 함께 앱 시장의 국경도 희미해지고 있는데요. 이제 한국의 앱이 미국에서 1위를 하기도 하고, 중국의 숏드라마가 전 세계를 강타하는 시대예요. 이렇게 급변하는 글로벌 트렌드 속에서, 우리 브랜드는 어떤 생존 전략을 준비해야 할까요?
이러한 흐름 속에서 향후 마케팅 전략의 실마리를 찾기 위해 에어브릿지(Airbridge)와 글로벌 광고 전략 분석 플랫폼 [앱그로잉(AppGrowing)](https://appgrowing.net/)이 함께 ‘Global AI App Insights Seminar’를 주최했어요.
지금 AI 시대의 마케팅이 어디로 가고 있는지, 어떤 준비가 필요한지 핵심 내용을 깔끔하게 정리해 드릴게요.
[]
## 에어브릿지가 바라보는 마케팅 환경의 변화
에어브릿지 사업 개발 매니저 이진성 님은 ‘AI 시대의 마케팅 측정, 에어브릿지가 말하는 MMP의 진화’라는 주제로 첫 세션을 진행해주셨는데요.
AI 시대에 빠르게 변화하는 마케팅 환경 속에서 에어브릿지가 어떤 변화를 준비하고 있고, 실무자들이 이 기술을 활용해 어떻게 성과를 낼 수 있는지에 대한 구체적인 해답을 제시했어요.
### 1. AI를 대하는 마케터의 자세: "AI for What?"
마케팅을 낚시에 비유해 보자면, 기존 마케팅은 광고 소재 A/B 테스트와 같이 그물을 잘게 쪼개 여러 곳에 던져보고, 물고기가 잡히면 그곳을 집중 공략하는 방식이었어요.
하지만 현실적으로 예산·시간·실험 횟수 모두 제한되어 있어요. 이런 상황에서는 어디에 큰 물고기 떼, 즉 고가치 유저가 있는지 정확히 알고 던져야 해요.
AI는 바로 이 물고기 떼를 찾는 탐지기 역할을 해야 해요. 이진성 님은 AI는 한정된 예산으로 고가치 유저가 많은 곳을 먼저 예측해서 의사결정 속도와 생산성을 높여주는 도구라고 강조했어요.
### 2. pLTV로 성과와 수익성을 모두 예측해 보세요.
그렇다면 마케터가 잡아야 할 고가치 유저들을 어떻게 찾을 수 있을까요? 이를 위해, 에어브릿지는 pLT(Predictvie Life Time: 잔존일 예측)와 pLTV(Predictive Life Time Value: 유저 생애 가치 예측) 기능을 제공하고 있어요.
[]
에어브릿지 pLTV는 과거 14일간의 데이터를 바탕으로, 향후 180일까지의 미래 성과를 예측하는데요. 단순히 추측하는 것이 아니라, pLT와 ARPDAU(일일 사용자당 평균 매출)을 곱하여 산출된 값으로 실제 LTV와 매우 높은 상관관계를 보인다고 해요.
[]
[]
[]
이를 활용하면 마케터는 CAC(고객 획득 비용) 대비 예측되는 가치인 pROAS(Predictive ROAS: 예측 ROAS)를 계산할 수 있어, 광고가 얼마나 효율이 날지를 미리 알 수 있어요.
또 pLTV를 pLT와 수익 지표로 쪼개 보면, 지금 서비스가 리텐션에 문제가 있는지, 아니면 구매 전환이나 객단가 문제가 있는지를 정확히 구분할 수 있어요.
### 3. 고객 사례: pLTV로 비구독자 대상 마케팅 전략 효율을 높인 알라미
딜라이트룸이 운영하는 글로벌 알람 앱 [알라미(Alarmy)](https://alar.my/)가 pLTV로 마케팅 전략과 액션 플랜 수립 과정을 효율화한 사례를 소개해 주셨어요.
알라미 마케팅 팀은 페이드 마케팅을 통해 획득한 유저들의 LTV를 개선하고자 했어요. 이를 위해, 마케팅팀은 집행한 마케팅 국가별, OS별, 매체별 예측 LTV를 직접 계산하고 이를 기준으로 마케팅 전략을 최적화해왔죠.
하지만 예측 LTV를 직접 계산하는 것은 상당히 번거로운 일이었어요. 국가별/OS별/매체별로 LTV를 직접 예측하려니, 데이터 가공에만 매번 1~3시간이 소요되는 등 리소스 낭비가 상당했어요.
[]
에어브릿지의 pLTV 기능을 도입하고 나서야 이 문제들을 해결할 수 있었어요.
1. 신속한 의사결정: pROAS를 통해 매체별/국가별 효율을 즉각 확인하고 예산을 효율적으로 배분했어요.
2. 소재 최적화: 광고 소재 단위까지 pLTV를 분석하여 고효율 소재를 발굴했어요.
3. 액션 플랜 수립: 리텐션이나 CPI 같은 후행지표까지 함께 보면서 CRM이 필요한 시점인지, UA 확장이 필요한 시점인지를 명확히 판단했어요.
- 리텐션이 낮게 예측되면? → CRM 마케팅 강화
- 유입이 부족하면? → UA 캠페인 강화
👉 알라미가 마케팅 전략을 최적화한 구체적인 방법이 궁금하다면 [AB180 CSHub](https://cshub.ab180.co/ko/case-studies/airbridge-pltv-delightroom)에서 확인하실 수 있어요.
[]
### 4. 에어브릿지가 선보일 새로운 AI 기능 로드맵
에어브릿지는 단순히 성과를 측정하는 것을 넘어, 마케터의 고민을 기술로 해결하는 방향으로 진화하고 있는데요. 앞으로 에어브릿지가 공개할 새로운 AI 기능 로드맵을 소개했어요.
1. AI 기반 비정상 설치 탐지: 디바이스 ID 초기화, 잦은 재설치, 터치포인트 과다, 인앱 이벤트 미발생 등을 기준으로 비정상 설치를 자동 분류하는 기능이에요. 기존에는 고객사가 직접 룰을 만들어 관리하던 Ad Fraud(부정 광고) 영역을 한층 자동화한 기능이죠.
2. 광고 소재 LLM 패널 기반 사전 테스트: 기존 성공 소재와의 유사도, 기존 자산 대비 독창성을 LLM으로 분석하고, 가상의 패널 그룹을 만들어 소재가 실제로 효과를 낼지 성과를 미리 테스트하는 구조예요.
[]
## 앱그로잉: 글로벌 AI 앱·숏드라마 시장은 무엇이 달라지고 있을까?
전 세계적으로 앱 마케팅 시장은 AI 기술의 발전과 짧은 영상 콘텐츠의 폭발적인 성장으로 마케터들에게 새로운 기회이자 도전이 되고 있어요.
앱그로잉 세션에서는 글로벌 비즈니스 총괄 Angela Zheng 님이 ‘글로벌 AI & 숏드라마 앱 시장 트렌드와 성공적인 광고 전략’을 주제로 AI 앱과 숏드라마에 관해 국가별·버티컬별·플랫폼별로 심도 있는 크리에이티브 전략과 마케팅 인사이트를 공유했어요.
중국 시장에서의 방대한 데이터와 경험을 바탕으로, 글로벌 시장에서 성공하기 위해 반드시 알아야 할 크리에이티브 전략과 마케팅 비법을 공개했는데요. 이번 세션의 핵심 내용을 알기 쉽게 정리해 드릴게요.
### 1. AI 앱 시장은 UA에서 수익화 최적화로 이동하는 흐름
Angela 님은 가장 먼저 2025년 상반기 AI 앱 시장의 폭발적인 성장세를 데이터로 보여주셨는데요.
[]
앱그로잉 데이터에 따르면, 2025년 1~6월까지 AI 앱 광고 집행량은 매달 빠르게 급증했는데, 6월 이후부터는 광고 집행량이 소폭 줄기 시작하면서 단순한 물량 공세보다 수익화에 집중하는 흐름으로 바뀌었다다고 해요. 시장이 성숙해지면서 마케터들이 효율을 따지기 시작한 것이죠.
이러한 흐름을 간단히 보면 아래와 같아요.
- 1~6월: UA 경쟁 과열 → 광고 크리에이티브 폭증
- 6월 이후: CPI 경쟁 완화 → 효율 중심 전략 + 광고 수익화 비중 증가
- 전체 AI 앱 수는 여전히 빠르게 증가 중
지역별로는 여전히 북미·서유럽이 핵심 시장이고, 동남아시아·라틴아메리카·중동은 인구 기반의 신규 성장 시장으로 급부상하고 있어요.
[]
또한 앱그로잉이 집계한 Top20 AI 앱 리스트를 보면, 약 90%가 텍스트나 이미지, 영상을 생성하거나 변환하는 생산성·유틸리티 카테고리에 집중돼 있다고 해요.
### 2. AI 앱의 광고 크리에이티브 전략: 감성과 기능
앱그로잉 데이터베이스에 등록된 수천 개의 AI 앱 광고를 분석해보면, 광고 소재의 크리에이티브를 2가지 축으로 정리할 수 있었어요.
[]
1. Emotional (감성형)
- 사용자의 상황·감정에 공감하는 방식
- AI가 인간적인 ‘도움’을 주는 장면 구성
- 앱과 사용자 사이의 심리적 거리 단축
2. Functional (기능형)
- Before/After 비교
- 특정 기능 하나를 극단적으로 강조
- 문제 → 해결 구조를 직관적으로 제시
기본적으로 AI 앱은 기능형 크리에이티브가 많지만, 최근에는 감성형도 빠르게 늘어나고 있다고 해요. 사용자가 AI지만 나와 실제로 대화하는 느낌을 받을 수 있도록 연출하는 방식이 특히 효과적이라고 강조했어요.
채널별 전략도 중요한데, 초기에는 메타(Meta)에 집중하다가 안정기에 접어들면 틱톡이나 구글로 채널을 다변화하며 각 매체 성격에 맞는 소재를 제작해야 한다는 팁도 공유했어요.
[]
[]
#### 프랙티카(Praktika) 케이스 스터디: 5백만 유저, 연 2,000만 달러 AI 언어 학습 앱의 성장 공식
AI 언어 학습 앱 프랙티카가 성공 사례로 소개됐는데요. 개인화 AI 튜터와 유저 취향에 맞는 아바타 기능으로 글로벌 유저 5백만, 연 매출 2,000만 달러(약 280억 원)라는 놀라운 성과를 기록하고 있어요.
여기서 프랙티카가 집행한 이탈리아의 ‘아페리티보(Aperitivo: 저녁 식사 전 가볍게 술을 즐기는 문화)’를 소재로 한 광고를 예시로 들었는데요. 현지 문화를 반영한 편안한 상황에서 AI 튜터와 사투리까지 섞어가며 대화하는 모습을 보여줌으로써 유저들의 감정적인 반응을 이끌어냈다고 해요.
### 3. CLG 전략 = 감이 아니라 프레임 단위로 분석하기
Angela 님은 마케터들이 감으로 광고 소재를 분석하는 게 아니라, 현미경처럼 프레임을 쪼개서 분석하는 집요함이 필요하다고 강조했어요.
특히 중국 시장에서는 이 방식을 CLG(Creative-Led Growth: 소재 주도 성장)라고 부르며, 3초 단위로 영상을 자르고 테스트하는 수준으로 깊게 들어가는 문화가 자리 잡았다고 해요.
> “중국에서는 CLG라는 말이 있을 정도로 소재에 대해 상당히 깊이 있게 들어갑니다. 같은 비디오라도 3초 단위로 분절해서 각 단위마다 최적화를 하고, 어떤 요인이 성과를 냈는지 분석하지 않으면 성장이 불가능하다고 봅니다.”
[]
### 4. 3초마다 터져야 한다! 숏드라마의 흥행 법칙
최근 전 세계를 강타하고 있는 숏드라마 시장에 대한 이야기도 빠질 수 없는데요. 숏드라마 앱 마케팅의 핵심은 결국 오리지널 콘텐츠예요. 어떤 스토리가 유행하는지 파악하는 것이 곧 마케팅 전략이죠. 재미있는 점은 영상 길이에 따른 훅(Hook)의 법칙이었어요. 숏드라마 광고는 영상 길이에 따라 전략이 완전히 달라요.
[]
#### 롱폼(9~15분)
- 전체 에피소드 중 가장 인기 있는 5~10개의 장면을 압축
- 스토리 흐름을 유지하면서 예고편 느낌으로 구성
- 10초마다 훅을 배치하여 이탈 방지 → 다음 내용이 궁금해서 앱을 깔 수밖에 없게 만들기
- 마지막에 궁금한 포인트에서 끊기
#### 숏폼(1분 이하)
- 3초마다 훅(충격/반전/트위스트)
- 훨씬 자극적이고 극단적 감정 장치를 빠르게 투입
- 스토리를 거의 보여주지 않고 감정 자극만 연속적으로 제공
> “숏폼 콘텐츠는 10초가 아니라 3초마다 흥미를 유발해야 합니다. 갑작스러운 임신 소식, 불륜 발각, 반전 같은 자극적인 요소를 3초마다 넣어서 유저를 붙잡아두지 않으면 바로 이탈해 버립니다.”
또한 타겟 유저의 변화도 흥미로웠는데요. 처음에는 4050 여성층을 타깃 하다가 시간이 지나면 20대 여성, 그다음엔 남성층으로 타겟을 계속 바꿔가며 콘텐츠를 공급해야 앱의 수명을 늘릴 수 있다고 조언했어요. 배우의 얼굴이나 국가 조합에 따라서도 성과가 천차만별이라, 정말 디테일한 분석이 필요해요.
[]
[]
그렇다면 어떤 장면에 훅을 넣어야 할 지 어떻게 알 수 있을까요? 앱그로잉에서는 영상 전체를 클립 단위로 자동 분해해서 어떤 장면이 성과나 이탈를 만들었고, 경쟁 콘텐츠는 어떤 클립을 쓰고 있는지 비교할 수 있어요.
숏드라마는 결국 대본과 배우의 조합이 성패를 좌우하기 때문에, 이 두 요소를 데이터로 연결해주는 기능을 앱그로잉이 제공하고 있다는 점을 강조했어요.
[]
### 5. 앱그로잉 플랫폼 데모: 마케터가 원하는 모든 크리에이티브 데이터가 한곳에
마지막으로 Angela 님은 마지막에 실제로 앱그로잉 플랫폼을 시연하면서, 글로벌 광고 소재 전체를 수집한 라이브러리와 앱별·플랫폼별 광고 집행 추이, 경쟁사와 자사 크리에이티브 비교 등 다양한 핵심 기능을 보여줬어요.
앱그로잉 플랫폼에는 전 세계의 다양한 광고 소재가 축적되어 있어서, 경쟁사가 어떤 소재로, 어느 매체에서, 어떤 타겟에게 광고를 하고 있는지 한눈에 볼 수 있는데요. 특히 경쟁사들이 어떤 스크립트와 배우를 쓰는지 확인하고, 이를 벤치마킹하여 실패 확률을 줄일 수 있어요.
여기서 앱그로잉만의 강력한 무기인 AI 에이전트 기능이 소개되었어요. 과거에는 마케팅 팀의 인턴들이 하루 종일 영상을 보며 “이 영상은 3초에 여자가 울었고, 배경은 학교다”라는 식으로 엑셀에 수기 분석을 했다고 해요.
하지만 이제는 앱그로잉 AI 에이전트에게 이 소재가 왜 잘 됐는지 물어보면 “복수극 테마에 반전 요소가 있고, 초반 3초에 강렬한 사투리를 썼기 때문입니다.”라고 즉시 분석해 준다고 해요.
마지막으로 Angela 님은 데이터를 보여주는 것을 넘어, 마케터가 전략에 집중할 수 있도록 돕는 것이 앱그로잉의 목표라고 강조하며 발표를 마쳤어요.
## AI 시대의 마케팅, 데이터를 통한 예측과 살아있는 크리에이티브를 더하세요.
이번 Global AI App Insights Seminar에서 공통적으로 드러난 메시지는 아주 분명했어요. AI 시대의 마케팅은 더 빠르고, 더 복잡하고, 동시에 더 정교해지고 있다는 점이에요. 그리고 이진성 님과 Angela 님의 서로 다른 듯하지만, 결국 성장의 본질이라는 공통점이 있어요.
다가오는 2026년의 글로벌 마케팅은 미래를 예측하는 데이터와 사용자를 움직이는 크리에이티브가 함께 맞물려야 성장할 수 있어요. 그리고 그 둘을 제대로 읽는 능력이야말로, 지금 AI 시대에 가장 중요한 경쟁력이 될 거예요.
이번 세미나가 여러분의 다음 캠페인과 전략에 작은 힌트라도 되었으면 좋겠어요.
---
## 크로스 플랫폼 측정부터 AI 기반 최적화까지, 에어브릿지 & 에어플럭스와 함께하면 강력한 이유
URL: https://www.airbridge.io/ko/blog/gstar-2025-mini-insight-session
Language: Korean
Published: 2025-12-03
국내 최대 게임 행사인 지스타는 게임 유저들뿐만 아니라, 매년 수많은 개발사와 퍼블리셔, 글로벌 파트너들이 한자리에 모여 미래의 게이밍 성장을 이야기하는 자리예요. 특히 올해는 글로벌 시장 확대와 플랫폼 다변화를 고민하는 게임사가 더 많아지면서, 그 어느 때보다 ‘측정’과 ‘데이터 기반 성장 전략’에 대한 관심이 높았어요.
에어브릿지는 몰로코와 함께 지스타 기간 게임사의 성장을 지원하는 밋업 라운지를 준비했어요. 11월 13일과 14일, 더벤티 카페에서 운영된 밋업 라운지는 실무자들이 편하게 들러 글로벌 UA 전략, 크로스 플랫폼 측정, 광고 효율화 같은 현실적인 고민을 이야기할 수 있는 열린 공간으로 운영되었어요.
이곳에서는 글로벌 성장 노하우와 실제 사례를 담은 인사이트 세션도 함께 진행됐어요. 글로벌 시장에서의 성장 전략부터 모바일을 넘어 PC·콘솔·CTV까지 확장되는 마케팅 환경, 그리고 수익화를 끌어올리는 데이터 기반 접근 방식까지. 게임사가 직면한 핵심 과제를 깊이 있게 다뤄 많은 분이 함께해 주셨는데요.
이번 콘텐츠에서는 인사이트 세션에서 소개된 에어브릿지와 [에어플럭스](https://www.airflux.ai/)의 핵심 인사이트, 그리고 앞으로의 성장 방향성을 정리했어요. 크로스 플랫폼 트래킹과 AI 기반 효율화 전략이 궁금하시다면, 오늘의 콘텐츠를 꼭 주목해 보세요.
## 에어브릿지 세션: 크로스플랫폼 환경에서의 캠페인 성과 측정 방법
[__wf_reserved_decorative]
연사자: 김수빈 에어브릿지 고객 성공 매니저
올해 지스타 현장에서 많이 들린 단어 중 하나가 바로 PC와 크로스 플랫폼이었어요. 실제로 에어브릿지를 활용하는 PC 게임 고객사는 전년 대비 크게 늘었고, UA팀과 라이브팀 모두 기존 모바일 중심 환경과는 다른 고민을 안고 계셨어요.
김수빈 님은 “올해만큼 PC와 크로스 플랫폼 측정 문의가 많았던 적은 없었다”고 말하며 세션을 시작했어요. 하지만 많은 고객사가 스팀 애널리틱스 같은 기존 도구만으로는 전환까지의 흐름을 충분히 파악하기 어렵다 보니 측정 자체를 “회색지대”라고 표현할 정도였죠.
모바일과 PC를 모두 운영하는 경우엔 상황이 더 복잡해요. 모바일 성과는 MMP로 확인할 수 있지만, PC 설치 및 전환은 별도로 관리되다 보니 통합적인 성과 흐름을 한 번에 확인하기 어렵다는 이야기가 반복되고 있었어요.
### PC·크로스 플랫폼 측정이 특히 어려운 이유
[__wf_reserved_decorative]
모바일과 비교했을 때 PC·콘솔 환경이 측정 난이도가 높은 이유는 다음과 같아요.
**1\. 유저 여정 자체가 지나치게 복잡해졌어요.**
모바일에서는 광고 → 앱 마켓 → 설치라는 단순한 흐름이 주류였어요. 하지만 PC로 확장되면 CTV, 웹, 모바일, PC까지 유입 경로가 무한히 늘어나요. 모바일에서 유입된 유저가 PC에서 설치하고, PC에서 시작한 유저가 다시 모바일로 넘어오는 흐름까지 더해지면 경우의 수는 실무자가 수기로 감당하기 어려운 수준이 되죠.
**2\. 환경별 식별자가 다르고, 서로 공유되지 않아요.**
광고를 클릭한 유저와 실제 설치 유저를 동일인으로 연결하기가 어려워지고, PC 플레이와 모바일 플레이를 중복 제거한 ‘하나의 유저 단위’로 보기도 쉽지 않아요. 식별자 체계가 달라서 억지로 맞추려 해도 분석의 신뢰도가 떨어져요.
**3\. 캠페인 최적화는 더 어려워요.**
설사 유저 매칭을 어느 정도 해결해도, 그 성과에 가시성을 더하거나 캠페인을 최적화하는 일은 전혀 다른 문제가 돼요. 결과적으로 LTV 기반 의사결정이 막히고, UA 비용 누수가 늘어나는 구조가 만들어져요.
### 크로스 플랫폼 트래킹, 에어브릿지가 강력한 이유
에어브릿지는 이런 복잡한 문제를 풀기 위해 웹·앱을 같은 유저로 인식하는 기술을 다듬어 왔어요. 이 기반 위에서 PC·콘솔 트래킹까지 지원을 확장한 에어브릿지는 다음과 같은 방식으로 문제를 해결하고 있어요.
#### 통합 식별자를 바탕으로 PC-모바일 흐름을 한눈에 파악할 수 있어요
[__wf_reserved_decorative]
에어브릿지는 모바일, PC, 콘솔처럼 서로 다른 환경에서 발생하는 유저 행동을 Airbridge ID라는 통합 식별자로 하나로 연결해요. 플랫폼마다 식별자가 달라도 Airbridge ID 아래에서는 동일한 유저로 인식되기 때문에, 모바일에서 유입된 유저가 PC에서 설치로 이어지거나, 그 반대의 흐름까지도 자연스럽게 추적할 수 있어요.
이렇게 유저가 하나의 기준으로 정리되면 분석 단계에서도 큰 차이가 생겨요. ROAS나 CPI 같은 핵심 지표를 빠르게 확인할 수 있는 대시보드뿐 아니라, 모바일과 PC의 성과를 하나의 대시보드에서 확인할 수 있는 크로스 플랫폼 퍼널 리포트를 통해 모바일↔PC 간 전환 흐름을 한눈에 파악할 수 있죠. 여기에 캠페인·채널·국가별 성과를 동일한 기준에서 비교할 수 있는 분석 기능까지 더해지면서, 플랫폼마다 따로 보이던 데이터들이 하나의 구조로 정리돼 훨씬 정확하고 직관적인 판단이 가능해져요.
#### 시그널을 전송해 강력한 매체 최적화를 지원해요
[__wf_reserved_decorative]
성과 측정만 되는 것에서 끝나지 않고, 그 데이터를 활용해 어떻게 최적화할 것인가도 중요한 부분이죠. 먼저 기본적인 이벤트 기반 포스트백을 통해 캠페인에서 발생한 전환 정보를 매체에 다시 전달하고, 이를 바탕으로 매체의 최적화 알고리즘이 제대로 작동하도록 도와줘요.
또한 구글, 메타 등 매체에서 제공하는 컨버전 API(CAPI)와의 연동도 적극 지원하고 있어요. 이 기능을 활용한 PC 게임 고객사들이 실제로 CPI 절감이나 ROAS 상승 등 눈에 보이는 개선 효과를 경험하고 있어요.
마지막으로 유저 행동을 바탕으로 코호트를 생성해 캠페인 오디언스로 전송할 수 있어요. 휴면 유저나 복귀 유저처럼 운영팀이 중요하게 관리하는 세그먼트를 에어브릿지에서 오디언스로 정의할 수 있고, 만들어진 오디언스는 별도 작업 없이 곧바로 매체 타겟팅에 활용할 수 있어요.
### 크로스 플랫폼 시대, 에어브릿지가 준비하는 2026
[__wf_reserved_decorative]
에어브릿지는 내년에도 크로스 플랫폼 게임사를 위한 기술 투자를 이어갈 것이라고 강조했어요. PC 환경에서 기여도를 더 정확하게 산출할 수 있는 신규 시그널 연구부터, 플랫폼 간 이동 흐름을 더 직관적으로 이해할 수 있는 리포트와 대시보드 고도화까지 다양한 과제가 이미 진행 중이라고 소개했죠. 또 몰로코를 비롯한 주요 파트너와의 연동을 강화해 PC 지면에서 최적화를 극대화하는 사례를 더 넓혀가겠다는 계획도 전했어요.
에어브릿지는 앞으로도 PC와 모바일을 아우르는 복잡한 환경 속에서 게임사가 글로벌로 성장할 수 있도록, 측정부터 최적화까지 이어지는 전 과정을 끊김 없이 지원하겠다는 목표를 다시 한번 확인해준 자리였어요.
## 에어플럭스 세션: AI가 실험을 대신해 준다면 어떤 일이 가능해질까요?
[__wf_reserved_decorative]
연사자: 최혜린 에어브릿지 & 에어플럭스 그로스 매니저
오늘날 게임사는 성장에 대한 두 가지 과제에 직면하고 있어요. 첫 번째는 낮은 CPI(Cost Per Install)로 고가치 유저를 확보하는 마케팅 최적화, 두 번째는 확보한 유저로부터 지속적인 수익을 만들어내는 리텐션·인게이지먼트 강화예요.
두 영역 중 하나만 잘해서는 성장에 속도가 붙기 어려워요. 게임사는 매체 선정부터 크리에이티브 테스트, 리텐션과 인게이지먼트를 강화하는 운영 전략까지 모든 단계에서 끊임없이 실험해야 하는 환경에 놓여있죠. 하지만 실험부터 설계, 집행까지의 과정은 시간이 오래 걸리고, 리소스와 비용 제약 때문에 한 달에 한 번 실험하기도 어려운 경우가 많아요. 실험의 빈도와 깊이가 낮아지면 결국 성장은 한계에 부딪히는 악순환이 지속될 수밖에 없죠.
### 에어플럭스, 강화학습 기반의 A/B 테스트 자동화
[__wf_reserved_decorative]
이 문제를 해결하기 위해 강화학습 기반의 의사결정 시스템을 적용한 [에어플럭스](https://www.airflux.ai/)가 탄생했어요.
에어플럭스의 역할은 단순해요. 유저가 게임을 계속 플레이하게 하고, 그 과정에서 더 많은 수익을 만들어내도록 광고 정책을 AI가 스스로 실험해 최적의 조합을 찾아주는 것이죠. 그래서 에어플럭스는 사람이 일일이 규칙을 정해주는 방식이 아니라, 유저의 성향을 파악해 유저마다 다른 광고 노출 전략을 자동으로 선택해요. 구매율이나 인게이지먼트가 높을 것으로 예상되는 유저에게는 광고를 적게 노출하고, 이탈 가능성이 높은 유저에게는 광고 빈도를 높여 수익을 늘리는 식의 접근이에요. 유저를 하나의 기준으로 묶어 운영하는 기존 방식과 달리, 개별 유저에 맞춘 정책이 실시간으로 적용된다는 점이 가장 큰 차별점이에요.
### 세그먼트가 세분화될수록 정교해지는 광고 정책 개인화
[__wf_reserved_decorative]
캐주얼 게임은 다른 장르보다 성과 변화를 빠르게 확인할 수 있고, 광고 수익(IAA) 비중이 높아 광고 정책의 영향이 더욱 명확하게 드러나는 특성이 있어요. 특히 캐주얼 게임은 출시 후 광고 정책을 변경하지 않는 경우가 많아 처음 적용한 정책이 최선인지 확인할 기회가 적다는 점도 주요 문제로 지적되었어요.
에어플럭스는 약 30개 이상의 변수를 활용해 유저의 행동 패턴을 세밀하게 판단해요. OS나 디바이스 같은 기본 정보부터 설치 시간, 플레이 시간대, 초기 스테이지 클리어 패턴 등 유저의 ‘행동 신호’를 활용해 이탈 가능성이나 인게이지먼트를 예측하고, 그에 맞춰 광고 정책을 실시간으로 조정해요.
에어플럭스 데이터를 분석해보니, 아시아 유저가 광고에 더 관대하고, 유럽 유저는 광고 경험에 더 민감하다는 것을 확인할 수 있었다는데요. 설치 시간대에 따라서도 이탈 패턴이 크게 달라졌다고 해요. AI 실험 자동화를 통해 사람의 직관으로 파악하기 어려운 패턴을 더욱 깊게 확인할 수 있다는 점이 인상적이었어요.
### 에어플럭스 적용 사례
글로벌 DAU 10만 이상을 기록하는 캐주얼 카드게임 솔리테어(Solitaire)는 IAA(인앱 광고) 기반 수익 비중이 높은 타이틀이에요. 솔리테어에 에어플럭스를 적용한 이후, 적용 그룹은 비적용 그룹 대비 평균 5번 이상 광고를 더 시청했을 뿐만 아니라 유저의 플레이 횟수(=max level)도 더 높은 수치를 기록했어요.
가장 인상적이었던 부분은 회수 기간의 변화였어요. 여러 캐주얼 게임에 에어플럭스를 적용해 분석한 결과, 기존에는 설치 후 약 58일이 지나야 성과가 회수되던 구조가 에어플럭스를 적용한 그룹에서는 33일 만에 동일 구간에 도달하는 흐름을 보였어요. 회수 기간이 짧아지면 더 공격적으로 UA에 투자할 수 있고, 신규 유저 확보가 늘면 다시 수익이 늘어나며 성장의 흐름이 가속되는 선순환 구조를 만들어낼 수 있어요.
### “실험이 어려울수록 AI는 더 큰 힘을 발휘합니다”
[__wf_reserved_decorative]
게임 데이터는 깊이 들어갈수록 더 많은 인사이트가 숨어있지만, 그만큼 설계와 분석에 드는 부담도 커지기 마련이에요. 사람이 직접 실험을 기획하고 결과를 검증하는 방식으로는 충분한 빈도를 확보하기 어렵고, 결국 성장의 속도도 한계에 부딪힐 수밖에 없죠. 에어플럭스는 바로 이 지점을 근본적으로 바꾸는 도구예요.
에어플럭스는 단일 지표 개선에 머무르지 않고, 게임의 성장 속도 자체를 끌어올리는 구조를 만들어 더 많은 게임사가 AI 기반의 새로운 성장 방식을 경험할 수 있도록 지원할거에요.
## 2026 게이밍 타이틀의 로켓 성장, 에어브릿지와 함께하세요
지스타에서 만난 여러 게임사의 공통된 고민은 결국 더 정확하게 측정하고, 더 빠르게 성장하는 방법에 있었어요. 에어브릿지의 강력한 크로스 플랫폼 트래킹과 에어플럭스의 AI 기반 효율화 전략은 이런 질문에 가장 실질적인 해답을 제시하는 도구예요. 앞으로도 에어브릿지와 에어플럭스는 게임사가 글로벌 시장에서 더 빠르게 성장할 수 있도록, 측정부터 최적화까지 이어지는 전 과정을 끊김 없이 지원할게요.
에어브릿지와 에어플럭스가 만들어가는 새로운 성장 방식을 지금 바로 경험해 보세요.
---
## 2026 모바일 게임 트렌드: 데이터로 본 APAC 게임 앱의 현재와 미래
URL: https://www.airbridge.io/ko/blog/apac-gaming-app-trend-2026
Language: Korean
Published: 2025-11-05
2025년 글로벌 모바일 게임 시장은 다운로드 감소와 매출 성장이라는 상반된 흐름 속에 놓여 있어요. 전 세계 다운로드 수는 2021년 약 513억 건에서 2025년 약 484억 건으로 감소했지만, 매출은 같은 기간 약 820억 달러에서 853억 달러로 오히려 증가했어요. 양적 성장에서 질적 성장으로 시장의 중심이 움직이면서 더 적은 유저가 더 오래, 더 깊게 게임을 소비하는 흐름이에요.
하지만 아시아 태평양(APAC) 시장을 자세히 들여다보면 이야기가 조금 달라져요. 일부 국가는 다운로드가 가파르게 증가하며 새로운 기회를 만들어내는 한편, 또 다른 국가는 매출 중심의 안정적 성장 국면에 진입하고 있어요. 이처럼 게임 시장 성장 방식은 각 국가마다 다양하게 분화된 모습이에요.
이번 콘텐츠에서는 한국, 일본, 중국, 베트남, 인도 등 APAC 25개국의 최근 5년간 다운로드 및 매출 데이터를 종합 분석해, 국가별로 달라진 시장의 주요 흐름을 정리했어요. APAC 모바일 게임 시장의 변화를 이해하고, 새로운 기회를 선점할 수 있는 인사이트를 얻어보세요.
\*모바일 및 디지털 경제 인텔리전스 플랫폼 [센서타워(Sensor Tower)](https://sensortower.com/ko)의 데이터를 바탕으로 에어브릿지에서 작성한 콘텐츠입니다.
## APAC 모바일 게임 다운로드 및 매출 트렌드
### 글로벌 감소세 속, 남아시아 신흥국이 이끄는 APAC 다운로드 성장세
지난 5년간 글로벌 모바일 게임 시장은 팬데믹 시기 급격한 성장세를 기록한 이후 점진적인 조정 국면에 들어섰어요. 그러나 흥미로운 점은, 전반적인 하락세 속에서도 APAC 시장의 다운로드 비중은 오히려 소폭 상승세를 보였다는 것이에요.
[__wf_reserved_decorative]
게임 앱 총 다운로드 점유율
[__wf_reserved_decorative]
국가별 게임 앱 총 다운로드 수 TOP5
2025년 9월까지 전 세계 모바일 게임 다운로드는 약 353억 건으로 집계되었어요. 이 중 APAC 시장은 약 153억 건으로 전체의 43%를 차지했어요. 게다가 올해 연말 APAC 지역의 모바일 게임 다운로드는 200억 건을 넘어설 것으로 예상되어, 글로벌 시장 내 존재감이 한층 강화되는 모습이에요. 글로벌 다운로드는 팬데믹 이후 감소세를 보여왔지만, 2023년을 기점으로 APAC 시장의 비중이 다시 소폭 상승하며 2025년 현재 최고치를 기록했어요. 특히 인도와 인도네시아는 5년 연속 APAC 내 다운로드 1·2위를 유지하며, 안정적인 시장 리더십을 이어가고 있어요. 파키스탄은 APAC 내 다운로드수 TOP5에 진입하며 다운로드 순위가 가파르게 상승했고, 2025년 처음으로 APAC 내 다운로드 3위권에 진입했어요.
즉, APAC 시장은 전 세계적으로 다운로드 수가 감소하는 가운데에서도 견조한 수요를 유지하며 성장의 균형추 역할을 하고 있어요. 특히 파키스탄 등 남아시아 신흥 시장의 빠른 성장세는 2026년에도 주목할만한 포인트에요.
### 안정세로 돌아선 APAC 매출, 한국 중심의 반등세 주목
팬데믹 이후 조정기를 거친 APAC 모바일 게임 시장 매출은 이제 안정세에 접어들고 있어요.
[__wf_reserved_decorative]
게임 앱 총 매출 점유율
2025년 9월 기준 전 세계 모바일 게임 매출은 약 610억 달러, 이 중 APAC 시장이 275억 달러(45%)를 차지했어요. APAC의 매출 비중은 2021년 51%에서 2025년 45%로 완만하게 하락했지만, 2023년 이후 안정적인 흐름을 이어가며 회복세를 보이고 있어요. 특히 2025년 매출은 팬데믹 이전 수준으로 회복될 것으로 예상되면서, 조정 후 안정화 단계에 들어선 것을 확인할 수 있어요.
[__wf_reserved_decorative]
국가별 게임 앱 총 매출 TOP5
국가별 매출을 살펴보면, 여전히 중국(iOS), 일본, 한국이 APAC 매출의 중심축을 유지하는 가운데 특히 한국 시장의 다른 국가 대비 증가세가 뚜렷하게 나타났어요. 2025년 3분기 한국의 모바일 게임 매출은 약 14억 달러로 APAC 전체의 14.7%, 지난 5년간 최고치를 기록했어요. 이는 대형 IP 기반 신규 타이틀인 ‘세븐나이츠 리버스(Seven Knights Re:BIRTH)’, ‘마비노기 모바일’의 성공과, 기존 상위권 게임 ‘화이트아웃 서바이벌(Whiteout Survival)’, ‘라스트 워: 서바이벌(Last War:Survival)’ 등의 이벤트 중심 매출 상승이 복합적으로 작용한 결과예요.
APAC 시장은 고지출 국가의 완만한 정체 속에서도 안정적인 매출 구조를 유지하고 있어요. 한국과 일본은 여전히 콘텐츠 퀄리티 중심의 시장으로서 높은 구매력을 보유하고 있으며, 이는 글로벌 퍼블리셔에게 중요한 고수익 기반으로 작용하고 있어요. 반면 인도·파키스탄 등 남아시아 신흥국에서는 매출 비중이 낮은 대신 다운로드 비중이 빠르게 확대되고 있죠. APAC 내부에서 국가별로 지출 시장과 유입 시장의 양극화가 진행되는 모습이에요.
이러한 흐름에 발맞춰 퍼블리셔도 두 시장을 구분하는 전략이 필요해요. 고지출 시장에서는 IP 확장과 프리미엄 IAP 모델을 강화하고, 신흥 시장에서는 광고 기반·보상형 플레이 등 가벼운 수익화 모델을 도입하는 이원화된 수익 전략이 필요한 것이죠.
%%CTA|데모 신청하기|/ko/book-a-demo|APAC 시장에서 게임 성과를 데이터로 관리하고 싶다면, 에어브릿지와 시작하세요.%%
## 모바일 게임 장르 트렌드
### 다운로드 기준: 시뮬레이션·퍼즐·아케이드 강세 속 다시 주목받는 슈팅 장르
[__wf_reserved_decorative]
2025년 다운로드 기준 모바일 게임 장르 트렌드
[__wf_reserved_decorative]
2025년 주요 장르별 다운로드 순위
2025년 APAC 모바일 게임 시장에서는 시뮬레이션, 퍼즐, 아케이드 장르가 각각 두 자릿수 비중을 유지하며 다운로드를 주도하고 있어요. 퍼즐 장르에서는 ‘블록 블라스트(Block Blast)’가 2025년 월간 다운로드 최고치를 기록하며 장르 성장을 견인했어요. 아케이드 장르에서는 2018년 출시된 ‘Hole.io’가 올해 들어 다시 성장세를 보이며 눈길을 끌고 있어요.
한편, 액션 장르는 2022년부터 2024년까지 3년 연속 5위를 유지했으나 2025년에는 슈팅 장르가 이를 대체했어요. ‘가레나 프리 파이어(Garena Free Fire)’의 지속적인 인기와 함께, 작년에 출시된 ‘FPS 스타라이크 옵스(Strike Ops)’, ‘델타 포스(Delta Force)’의 급상승이 장르 확장을 이끌며 슈팅 게임의 존재감을 강화했죠.
### 매출 기준: RPG 약세, 전략과 슈팅이 새 성장 동력으로 부상
[__wf_reserved_decorative]
2025년 매출 기준 모바일 게임 장르 트렌드
[__wf_reserved_decorative]
2025년 주요 장르별 매출 순위
2025년 APAC 모바일 게임 매출은 여전히 RPG와 전략이 양대 축을 이루며 시장을 견인하고 있어요. 다만 RPG 비중은 2021년 41%에서 2025년 29%로 점진적으로 하락한 반면, 전략 장르는 최근 5년 중 최고 비중인 28%를 기록하며 일부 월에서는 RPG 매출을 세 차례나 상회했어요.
RPG 장르에서는 여러 타이틀이 꾸준한 성과를 내고 있지만, 전체 장르 내 매출 성장세는 다소 둔화되는 모습이에요. 반면, ‘아너 오브 킹즈(Honor of Kings)’와 ‘화이트아웃 서바이벌’은 각각 1월과 8월에 월매출 신기록을 세우는 등 전략 장르에서는 뚜렷한 매출 상승세가 이어지고 있어요.
또한 슈팅 장르(8%)는 다운로드뿐 아니라 매출에서도 영향력을 확대하고 있어요. ‘화평정영(和平精英)’이 꾸준한 매출을 유지하는 가운데, ‘델타 포스’가 두드러진 상승세를 보이며 슈팅 시장의 활력을 이끌었어요.
## APAC 모바일 게임 순위
### 다운로드 기준 모바일 게임 TOP10
[__wf_reserved_decorative]
2025년 국가별 모바일 게임 다운로드 순위
한국, 일본, 중국은 다운로드 상위권에 자국 브랜드의 대표 타이틀이 포함되는 등 자국 IP 기반의 게임이 여전히 강세를 유지하고 있어요. 이들 시장은 독자적인 게임 제작 생태계를 기반으로 APAC 다운로드 트렌드와는 다른 구조를 보여주고 있죠.
반면 ‘블록 블라스트’는 중국과 인도를 제외한 APAC 전역에서 공통적인 인기를 얻으며 다운로드 상위권에 자리잡았어요. 특히 2025년 5월 인도네시아에서 약 700만 건 이상의 다운로드를 기록하며 현지 출시 이후 최고치를 달성했고, 6월 이후에는 베트남이 지속적으로 APAC 내 다운로드 비중 1위 국가로 올라서며 퍼즐 게임 중심의 콘텐츠 소비가 동남아시아 전반으로 확산되고 있어요.
인도에서는 전통적인 보드게임과 낮은 사양에서도 원활히 구동되는 접근성 있는 모바일 게임이 인기를 얻는 모습이에요. 이러한 특성 덕분에 ‘루도 킹(Ludo King)’이 인도 내 압도적인 유저 기반을 바탕으로 APAC 다운로드 3위를 차지했으며, ‘주피 루도(Zupee Ludo)’ 등 유사 타이틀도 함께 상위권에 오르며 지속적인 인기를 이어가고 있어요.
### 매출 기준 모바일 게임 TOP10
[__wf_reserved_decorative]
2025년 국가별 모바일 게임 매출 순위
‘아너 오브 킹즈(Honor of Kings)’와 ‘화평정영(和平精英)’은 중국 시장을 기반으로 견고한 매출 구조를 유지하며 각각 APAC 매출 1위와 3위를 기록했어요. 한편 ‘화이트아웃 서바이벌’은 올해 중국, 한국, 일본에서 모두 출시 후 최고 월매출을 경신하며 전략 장르의 성장세를 대표하는 사례로 주목받았어요. ‘라스트 워: 서바이벌’ 역시 한국과 일본에서 꾸준한 매출 흐름을 이어가며 APAC 4위를 기록했죠. 두 타이틀의 다지역 성공은 전략 장르가 지역 편중을 벗어나 글로벌 매출 확산 가능성을 입증한 대표 사례로 꼽혀요.
한국 시장에서는 ‘세븐나이츠 리버스’, ‘마비노기 모바일’, ‘RF 온라인 넥스트’, ‘뱀피르’ 등 신규 RPG 타이틀이 안정적인 성과를 보이며 시장 활력에 기여했어요. ‘포켓몬 카드 게임(Pokémon TCG Pocket)’은 2024년 말 일본 출시 이후 올해 꾸준한 인기를 이어가며 일본 내 매출 1위, APAC 전체 6위를 기록한 한편, ‘던전앤파이터 모바일’은 중국 내 매출 5위, APAC 전체 9위를 차지했어요. ‘포켓몬 카드 게임’과 ‘던전앤파이터 모바일’은 공통적으로 IP 인지도와 지역 맞춤형 콘텐츠 조합을 통해 매출 상승을 이끌었어요.
인도에서는 ‘가레나 프리 파이어’와 ‘배틀그라운드 모바일 인도(BGMI)’가 매출을 주도하며 슈팅 장르가 핵심 수익원으로 자리했어요. 또한 ‘eFootball’은 일본, 인도, 동남아 등 다지역에서 흥행하며 스포츠 장르 중 유일하게 APAC 전체 매출 TOP10에 진입했어요.
## 모바일 게임의 글로벌 성장에 함께하세요
에어브릿지는 더 많은 모바일 게임이 글로벌 시장에서 성장할 수 있도록 지원하고 있어요. 크로스 플랫폼 성과 트래킹 솔루션 에어브릿지와 AI 기반 모바일 캐주얼 게임 수익화 솔루션 [에어플럭스](https://www.airflux.ai/)를 만나보세요. 모바일 게임 성장을 위한 더 많은 지원이 궁금하다면, 지금 전문가와의 데모를 신청해보세요.
---
## 틱톡 iOS 실시간 전환 보고 기능 출시, 에어브릿지에서 틱톡 성과를 더 빠르고 정밀하게 분석하세요
URL: https://www.airbridge.io/ko/blog/tiktok-ios-realtime-conversion-reporting
Language: Korean
Published: 2025-10-21
틱톡이 iOS 14.5 이상 캠페인을 위한 새로운 실시간 전환 보고(Real-Time Conversion Reporting) 기능을 출시했어요. 이제 틱톡 iOS 캠페인을 운영하는 마케터는 기존 SKAdNetwork(SKAN) 기반 측정의 한계를 뛰어넘어 데이터를 기반으로 신속한 의사결정을 내릴 수 있어요. 에어브릿지는 틱톡의 공식 모바일 측정 파트너(MMP)로서 이 기능을 손쉽게 활성화해 iOS 캠페인의 성과를 극대화할 수 있도록 지원해요.
## 실시간 전환 보고란 무엇인가요?
틱톡의 iOS 실시간 전환 보고는 SKAN을 보완할 확률적 모델링 기능이에요. 기존 SKAN은 최대 48시간의 데이터 지연이 발생하고, 리포팅에 한계가 있었어요. 한편 틱톡 iOS 실시간 전환 보고를 활용하면 전환 수, 전환당 비용, 전환율 등의 상세 전환 데이터를 거의 실시간에 가깝게 틱톡 광고 매니저 내에서 확인할 수 있어요.
특히 이 기능은 틱톡의 AI 기반 캠페인 자동 운영 상품인 스마트플러스(Smart+)의 핵심 작동 기반이기도 해요. 실시간 전환 보고를 활성화하면 스마트플러스의 최적화 정확도 향상에 긍정적인 효과를 줄 수 있어, iOS 캠페인의 성과를 한층 더 끌어올릴 수 있어요.
### SKAN의 대체가 아닌 보완 관계
실시간 전환 보고는 SKAN을 완전히 대체하는 기능은 아니에요. 하지만 에어브릿지에서 이 기능을 SKAN과 함께 운용하면 보다 빠르게 성과 데이터를 확인할 수 있고, 마케팅 의사결정 속도를 끌어올릴 수 있어요.
또한 SKAN 대비 풍부한 전환 데이터를 제공받을 수 있어 틱톡 광고 매니저 내에서 실시간으로 캠페인을 조정하고 효율적으로 최적화할 수 있어요.
## 실시간 전환 보고 적용 방법
에어브릿지에서 틱톡의 실시간 전환 보고를 활성화하려면, 에어브릿지에서 틱톡으로의 [확장된 프라이버시 제어(EPC)](https://help.airbridge.io/ko/guides/extended-privacy-control) 전송을 비활성화 해야 해요. 이 설정은 틱톡 측 포스트백 방식과 확률적 매칭 방식의 정합성을 높이기 위한 필수 단계이며, 에어브릿지 대시보드 또는 고객지원팀을 통해 손쉽게 설정할 수 있어요.
해당 설정을 완료하면 틱톡과 MMP 간의 데이터 불일치(특히 전환 수, CPA, ROAS 등 지표 차이)를 줄일 수 있으며, 스마트플러스와 같은 자동 최적화 캠페인의 성과 역시 안정적으로 개선할 수 있어요.
### 자주 묻는 질문 (FAQ)
**Q. SKAN을 완전히 대체해도 되나요?**
실시간 전환 보고는 부가적인 측정 방식으로 SKAN을 대체할 수 없어요.
**Q. MMP 없이 이 기능을 쓸 수 있나요?**
현재는 불가능해요. 틱톡은 자사 SDK나 이벤트 API만으로는 iOS 실시간 전환 보고를 지원하지 않아 MMP 연동은 필수예요.
**Q. 이 기능을 사용하려면 어떻게 세팅해야 하나요?**
에어브릿지를 통해 틱톡을 연동한 후, EPC 전송을 비활성화하면 돼요. 별도로 세팅에 필요한 건 없어요.
**Q. 어떤 iOS 버전부터 지원되나요?**
iOS 14.5 이상부터 지원해요.
**Q. 모든 앱이 해당 기능을 활용할 수 있나요?**
앱에 iOS 캠페인의 실시간 전환을 표시할 수 있는 충분한 데이터 신호가 있는 경우에 활용할 수 있어요. 틱톡 광고 매니저에서 iOS 캠페인을 생성할 때 관련 메시지가 노출되는지 확인해 보세요.
## 에어브릿지와 함께 틱톡 iOS 캠페인 성과를 가장 정확하게 측정하세요
에어브릿지를 통해 틱톡의 실시간 전환 보고 기능을 더 쉽게 연동하고, SKAN보다 빠르고 정밀한 iOS 캠페인 성과 측정을 시작해보세요. 정교한 광고 성과 분석 및 최적화에 도움이 필요하다면 언제든 에어브릿지에 문의해 주세요.
---
## 네이버 애드부스트 쇼핑 연동, 에어브릿지에서 AI 기반 쇼핑 광고 효율을 극대화하세요
URL: https://www.airbridge.io/ko/blog/naver-advoost-shopping-integration
Language: Korean
Published: 2025-09-16
에어브릿지가 [네이버 애드부스트(ADVoost) 쇼핑](https://ads.naver.com/gfa?keyword=ad_voost) 연동을 지원해요. 이제 에어브릿지에서 네이버 AI 기반의 쇼핑 광고 캠페인 성과를 더 정밀하게 측정해 운영 효율을 최적화할 수 있어요.
## 네이버 애드부스트 쇼핑 알아보기
네이버 애드부스트는 AI 기술을 활용해 타겟 설정부터 소재 제작, 운영 자동화까지 지원하는 네이버의 광고 솔루션 브랜드예요. 그중 지난 5월 오픈베타 서비스로 선보인 ‘애드부스트 쇼핑’은 AI 기반 자동화 기능을 바탕으로 광고주의 운영 부담을 줄이면서도 쇼핑 캠페인 성과를 극대화할 수 있도록 설계된 솔루션이에요.
## 네이버 애드부스트 쇼핑 연동으로 기대되는 점
에어브릿지에서 발급한 캠페인별 트래킹 링크를 네이버 대시보드에 붙여넣기만 하면 간단히 연동이 완료돼요. 이를 통해 에어브릿지와 함께하는 마케터는 애드부스트 쇼핑 캠페인과 카탈로그 판매 캠페인의 성과를 빠르고 정확하게 확인할 수 있어요.
연동을 통해 얻을 수 있는 이점은 다음과 같아요:
1. 광고를 통해 전환이 얼마나 일어났는지 확인할 수 있어요.
2. 캠페인, 채널, 소재별 효율을 확인할 수 있어요.
3. 전환율이 높은 사용자 그룹을 찾을 수 있어요.
4. 광고 집행 비용과 광고를 통한 매출을 한 눈에 확인할 수 있어요.
5. 여러 채널의 캠페인 성과를 하나의 대시보드에서 확인할 수 있어요.
더 자세한 연동 방법이 궁금하시다면 [유저 가이드](https://help.airbridge.io/ko/guides/ad-channels-list)를 확인하거나, 에어브릿지 팀에 문의해 도움을 받아보세요.
## 마무리하며
[국내 마케팅 연동성 1위](https://www.airbridge.io/ko/blog/korea-no-1-mmp)를 자랑하는 에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 네이버, 카카오, 토스, 블라인드, 당근 등 다양한 매체와의 연동을 누구보다 빠르고 깊이 있게 지원하고 있어요. 이번에 네이버 애드부스트 쇼핑 연동까지 더해지면서, 에어브릿지와 함께하는 마케터는 국내 환경에 최적화된 캠페인을 한층 더 정교하게 운영할 수 있어요.
지금 바로 에어브릿지와 함께 마케팅 캠페인을 최적화하고, 경쟁에서 한 발 앞서가 보세요.
---
## [CTV 인사이트 인터뷰 3] Moloco: 성과로 증명되는 CTV 광고의 효율
URL: https://www.airbridge.io/ko/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance
Language: Korean
Published: 2025-09-04
CTV는 모바일 마케팅 업계에서 새롭게 떠오르고 있는 채널이지만 여전히 CTV를 통한 모바일 마케팅에 대한 인사이트는 부족한 게 사실이에요. CTV를 통해 유의미한 모바일 마케팅 성과를 만들어내려면 어떤 전략이 필요할까요?
에어브릿지가 준비한 CTV 인사이트 인터뷰 세 번째 시간은 AI 기반 퍼포먼스 광고로, 규모와 산업에 관계없이 다양한 비즈니스의 성장을 이끌고 있는 [Moloco](https://www.moloco.com/ko)(이하 몰로코)와 함께 했어요. 성공적인 모바일 앱 마케팅을 위해 새롭게 떠오르는 CTV. 이 CTV를 활용해서 진정한 성과를 만들어 내기 위해 꼭 알아야 하는 인사이트를 몰로코의 CTV 파트너십 총괄 이혜정 님께서 공유해 주셨는데요. 지금부터 확인해 보세요.
## 2025년 CTV 광고 시장은 어떻게 변화하고 있나요?
전 세계적으로 스트리밍 서비스 이용자가 늘고, CTV 시장에서의 성공 사례들도 많이 등장하면서, 2025년 CTV 광고 시장도 더욱 가파르게 성장하고 있어요.
- AI 개인화 기술 발전: 유저 행동 데이터를 기반으로 더 맞춤화된 CTV 광고를 보여줄 수 있게 되었어요.
- 측정 표준화: 도달률, 빈도, 성과 같은 핵심 지표가 플랫폼 전반에서 점점 더 통일되고 있어요. 덕분에 성과 중심 마케터들이 캠페인 효과를 명확히 평가할 수 있죠.
- 새로운 광고 포맷: 인터랙티브·거래형 광고 형식이 등장하면서, 이제 CTV는 단순 브랜딩 도구가 아니라 퍼포먼스 채널로 발전하고 있어요.
## 왜 모바일 앱 마케터들이 CTV 캠페인에 투자해야 할까요? 주요 장점은 무엇인가요?
사용자들의 관심이 점점 CTV로 이동하고 있어요. 유튜브 CTV 시청률이 유튜브 모바일 시청률을 넘어섰다는 사실이 이를 잘 보여주죠. CTV 채널의 주요 장점은 다음과 같아요.
- 포화된 채널의 대안: 이미 기존 모바일·웹 채널에서 성장을 다 채운 앱이라면, CTV는 새로운 고가치 유저를 만날 수 있는 차별화된 성장 기회를 줘요.
- 프로모션에 강함: 게임 출시, 사전 예약, 대규모 이벤트, 시즌성 캠페인 같은 시기에 CTV는 높은 참여를 유도해 전환 퍼널 성과를 끌어올려요.
- 크로스 디바이스 효과: CTV는 주로 거실 같은 공유 공간에서 시청되기 때문에, 같은 가정 내 모바일 기기와 연계해 리타겟팅이 가능해요. 설치, 가입, 구매까지 이어지는 전체 전환 여정을 완성할 수 있죠.
- 광고 사각지대 해소: 유튜브 프리미엄이나 넷플릭스처럼 광고 없는 플랫폼을 이용하는 유저에게 도달할 수 있어, 모바일에서 놓친 오디언스를 다시 확보할 수 있어요.
- 성과 증명: 게임, 스포츠 베팅, 라이프스타일 앱처럼 CAC 압박이 크거나 스토리텔링이 중요한 카테고리에서 높은 ROAS와 리텐션을 보여주고 있어요.
- 프라이버시 친화적: 개인 기기 식별자에 의존하지 않아, 앞으로도 안정적인 성과 채널이 될 수 있어요.
## 어떤 앱 카테고리가 CTV 광고에서 가장 좋은 성과를 보이나요?
게임 앱, 특히 미드코어, 카지노, 스포츠 베팅 분야가 강력한 성과를 보여주고 있어요. 또 스트리밍, 숏폼 비디오, 이커머스, 웹 전환(홈페이지 방문, 사전 예약) 목표를 가진 앱들도 주목받고 있죠.
대체로 모바일 프로그래매틱 채널에서 이미 성과가 좋은 앱들은 CTV 광고에도 빠르게 적응하는 편이에요.
## 몰로코는 어떻게 마케터들이 효과적인 CTV 캠페인을 운영하도록 돕고 있나요?
Moloco CTV는 우리의 딥러닝 엔진을 기반으로 구축돼 있어요. 그래서 노출이나 도달에 그치지 않고, 설치·행동·구매·매출 같은 실제 성과 지표까지 최적화할 수 있어요. 또한 모바일 캠페인에서 얻은 인사이트와 타겟팅을 CTV에 그대로 연결할 수 있기 때문에, UA와 리인게이지먼트 성과 전반에서 더 높은 효율을 만들어 낼 수 있죠.
## 몰로코의 CTV 입찰 알고리즘은 어떻게 작동하나요?
몰로코의 머신러닝 엔진은 광고주의 KPI에 맞춰 입찰가를 동적으로 조정해 성과를 극대화해요.
현재는 앱 설치 최적화 모델을 제공하고, 인앱 행동·구매·ROAS 같은 퍼널 하위 지표도 측정할 수 있어요. 앞으로는 구매·ROAS 최적화까지 확대할 예정이에요.
특히 에어브릿지와의 연동 덕분에 광고주는 CTV에서도 포스트백을 활용해 모바일 DSP 수준의 성과 최적화를 경험할 수 있어요. 이는 모바일 앱뿐 아니라 최근 각광 받고있는 PC·콘솔 게임 개발사들도 설치 기반 캠페인을 성과형으로 운영할 수 있게 해주죠. 아직 이만큼 깊이 있는 퍼널 최적화를 지원하는 솔루션은 많지 않아요.
## 어떤 CTV 포맷이 가장 성과가 좋은가요? 광고주는 어떻게 크리에이티브를 업로드하고 성과를 추적할 수 있나요?
주로 AVOD, FAST, 라이브 TV 콘텐츠의 인스트림 비디오 광고에서 성과가 많이 나와요. 이 포맷은 건너뛸 수 없고, 시청 완료율이 93~96%에 달해요.
특히 미국에서는 유튜브 CTV 시청률이 유튜브 모바일 시청률을 넘어섰어요. 그래서 디지털 마케터들에게 CTV는 꼭 고려해야 할 광고 채널이 되었죠.
CTV 광고는 기존 모바일·웹보다 더 몰입도가 높고, 광고 부하가 적어요. 또 전통 TV와 달리 실시간 최적화나 A/B 테스트도 가능하고, 대규모, 장기간 사전 청약 없이 합리적이고 유동적인 예산으로 디지털 마케팅과 유사하게 진행할 수 있어요. 광고주는 몰로코 플랫폼에서 바로 영상을 업로드할 수 있고, MMP 연동을 통해 캠페인 성과를 추적할 수 있어요.
## CTV 광고의 타겟팅 옵션은 제한적이라는 인식이 있는데, 몰로코는 어떤 기능을 제공하나요?
몰로코는 다양한 타겟팅 기능을 제공해요.
- 퍼포먼스 ML과 퍼스트 파티 오디언스 타겟팅
- 지역 기반 타겟팅 (국가·주 단위)
- 기기·가구 단위 타겟팅 (가능한 경우)
- 퍼블리셔 신호 기반 컨텍스트 타겟팅
- IP 기반 억제·리타겟팅
몰로코 Household Identity Graph 가 발전하면서, 크로스 디바이스·크로스 플랫폼 최적화도 점점 더 강력해질 거예요.
## 모바일 앱 마케터들은 몰로코에서 CTV 캠페인 성과를 어떻게 측정할 수 있나요?
에어브릿지와의 연동을 통해, 광고주는 CTV 기반 설치, 인앱 행동, 퍼널 하위 KPI까지 연동된 어트리뷰션 체계 안에서 측정할 수 있어요.
또 증분성 테스트, 고스트 비딩 같은 기능으로 CTV가 마케팅 성과에 어떤 기여를 했는지 정밀하게 분석할 수 있죠.
에어브릿지와의 파트너십을 통해 고성능 CTV 광고 생태계를 함께 만들어가고 있어요.
## 몰로코를 통해 성공한 모바일 앱 사례를 공유해주실 수 있나요?
여러 광고주들이 몰로코 CTV를 통해 새로운 성공 사례를 만들어가고 있어요.
예를 들어, 크래프톤은 인도 시장에서 ‘쿠키런 인디아’ 출시를 지원하기 위해 몰로코 CTV를 활용했어요. 상위·중간 퍼널 도달을 CTV로, 하위 퍼널 확보는 모바일 DSP로 진행해 큰 성장을 이끌었죠.
또 미국의 한 스포츠 베팅 앱은 주 단위 타겟팅과 깨끗한 환경의 CTV 광고를 활용해 큰 성과를 냈어요. 특히 인기 스포츠 시즌 동안 고전환 퍼포먼스를 달성하면서, CTV가 얼마나 효과적이고 확장 가능한 채널인지 보여줬어요.
중국의 대형 게임사는 PC/Console 타이틀의 UA 캠페인을 몰로코 CTV와 진행하면서, 고가치 유저들을 아주 효과적인 비용으로 도달하는 성과를 만들어 내기도 했어요.
## CTV 광고 확장을 고민하는 모바일 앱 마케터들에게 조언을 해주신다면요?
우선 모바일 DSP에서 잘 성과를 내는 오디언스를 CTV로 확장해 보세요. 거실 같은 프리미엄 환경에서 더 높은 주목도를 얻을 수 있어요.
그리고 단순 인지도뿐만 아니라, 설치·인앱 행동·구매 같은 퍼널 하위 성과까지 최적화하세요.
에어브릿지 같은 MMP와의 연동으로 CTV의 기여도를 정확히 추적하는 것도 중요해요.
무엇보다 CTV를 ‘브랜딩 전용 채널’로 보지 말고, 모바일 UA 전략의 자연스러운 확장판으로 바라보는 게 좋아요.
## PC·콘솔 게임 개발자들은 어떻게 CTV 광고를 활용할 수 있을까요?
PC·콘솔 플랫폼은 전환 경로가 제한적이고, 퍼포먼스 기반 UA에 익숙하지 않은 경우가 많아요. 몰로코는 모바일 게임사에서 쌓은 성과 경험을 PC·콘솔에도 확장하고 있어요.
- CTV의 몰입형 경험: AVOD·FAST 기반 풀스크린 광고는 사이트 방문, 다운로드, 사전 예약 같은 행동을 이끌 수 있어요.
- PC·콘솔 특화 이점: 가격대가 높고 구매 여정이 긴 PC·콘솔 게임에 맞춰, 브랜드 인지에서 전환까지 이어질 수 있도록 스토리텔링 중심 광고를 지원해요.
- 측정 가능성: 에어브릿지 같은 MMP 연동으로 웹 방문, 계정 가입, 다운로드, 인앱 행동까지 연동 추적할 수 있어요. 증분성 테스트와 고급 어트리뷰션도 지원하죠.
아직 초기 단계지만, 우리는 파트너사들과 함께 주요 프로모션을 준비 중이에요. 곧 더 많은 성공 사례를 공유할 수 있을 거예요.
## 끝으로 하고 싶으신 한 마디가 있으시다면?
몰로코는 CTV에서 직접 반응 성과(Direct Response)를 이끌어내는 데 집중하고 있어요. 모바일 DSP에서 검증된 머신러닝 기반 성과 역량을 CTV에 확장해, 광고주가 측정 가능한 성과를 얻을 수 있도록 돕고 있어요.
CTV 생태계 전반에서 공급자와 수요자가 모두 성장할 수 있도록, 우리는 성과 중심의 CTV 환경을 만들어가고자 해요.
많은 모바일 마케터들이 아직도 CTV 광고가 복잡하거나 성과 측정이 어렵다고 생각해요. 하지만 몰로코와 같은 플랫폼을 활용하면 페이스북·구글 캠페인을 시작하는 것만큼 쉽고, 오히려 더 높은 참여율과 낮은 경쟁 환경을 경험해 보세요.
## CTV를 통한 모바일 성장 인사이트
CTV 광고 업계를 선도하는 팀들이 전하는 CTV 광고 관련 인사이트가 궁금하신가요? 지금 아래 링크를 클릭하고 CTV 인터뷰 시리즈를 읽어 보세요.
👉 [Motiv Intelligence: 모바일-TV 매칭을 통한 효율적인 CTV광고 캠페인 집행](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-motiv)
👉 [Vibe.co: 미국 시장 진출. AI 기반 알고리즘으로 CTV 광고 성과 극대화하기](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-vibe)
---
## [CTV 인사이트 인터뷰 2] Vibe.co: 미국 시장 진출. AI 기반 알고리즘으로 CTV 광고 성과 극대화하기
URL: https://www.airbridge.io/ko/blog/airbridge-ctv-interview-vibe
Language: Korean
Published: 2025-09-04
모바일 앱 마케터들에게 새로운 UA 채널로 급부상 중인 CTV. 하지만 아직 그에 대한 정보는 부족하죠. 에어브릿지가 진행한 CTV 인사이트 인터뷰 두 번째 시간으로 CTV 전문 광고 솔루션을 제공하는 [Vibe.co](http://vibe.co)와 함께 CTV 광고에 대한 이야기를 나누었어요.
미국 시장 진출을 위한 CTV 광고 관련 인사이트와 [Vibe.co](http://vibe.co)가 제공하는 AI 기반 알고리즘을 바탕으로 한 CTV 광고 성과 극대화 인사이트까지. 지금 에어브릿지 CTV 인터뷰를 통해 이 모든 내용을 확인해 보세요.
## 모바일 앱 마케터들이 CTV 캠페인에 투자해야 하는 이유는 무엇인가요?
모바일 광고 채널의 경쟁이 점점 치열해지고 비용도 비싸지고 있죠. 이럴 때일수록 진정한 성과를 도출할 수 있는 새로운 대안 채널이 절실한데요. CTV는 이러한 지점에서 마케터들에게 해답을 제시할 수 있어요. CTV 채널의 주요 장점은 세 가지 정도로 압축할 수 있어요.
- **낮은 고객 획득 비용(CAC):** 한 화면 안에서도 광고가 넘쳐나는 모바일 환경과 달리 CTV는 깔끔하고 정제된 인터페이스로 광고에 집중할 수 있는 환경을 제공하고 이는 높은 참여도로 이어져요.
- **넓어진 도달 범위: **OTT와 같은 스트리밍 서비스 뿐만 아니라 TV나 스마트 TV 등의 시청자들에게도 도달되며 기존 모바일 네트워크에서 잘 잡히지 않던 유저에게 수월히 도달할 수 있어요.
- **가치 높은 유저 확보:** CTV 시청자들은 무심코 화면을 넘기는 게 아니라, 더 집중해서 콘텐츠를 보고 적극적으로 반응해요. 이는 자연스럽게 고가치 유저 확보로 이어질 수 있죠.
그리고 CTV는 몰입감 있는 풀스크린 광고 형식을 제공하기 때문에, 모바일 마케터들이 기존 모바일 네트워크보다 더 효율적으로 유저를 확보할 수 있어요.
## 여전히 많은 모바일 마케터들이 CTV를 낯설게 느끼고 있는데요. 어떤 카테고리 앱들이 CTV에서 좋은 성과를 내고 있나요?
CTV는 여러 앱 카테고리에 두루 좋은 성과를 내고 있지만, 특히 두드러진 성과를 내는 앱 카테고리들이 있어요.
- **게임 앱:** 캐주얼이나 미드코어 게이머들이 스트리밍을 즐기면서 CTV 광고에도 잘 반응해요.
- **라이프스타일 & 피트니스 앱:** 건강, 웰빙, 자기계발 앱은 CTV의 몰입감 있는 포맷 덕분에 성과가 좋아요.
- **이커머스 앱:** 리테일이나 마켓플레이스 앱은 CTV를 통해 웹-투-앱 전환을 효과적으로 유도해요.
- **금융 & 핀테크 앱:** 뱅킹, 투자, 암호화폐 앱은 CTV가 의도 높은 유저에게 도달해 설치와 계정 가입을 이끌어내는 데 유리해요.
- **스트리밍 & 엔터테인먼트 앱:** 이미 콘텐츠를 즐기고 있는 시청자에게 무료 체험이나 구독 전환을 촉진할 수 있죠.
- **유틸리티 앱:** 효율성을 높여주는 유틸리티·생산성 앱은 적극적으로 솔루션을 찾는 유저와 연결될 수 있어요.
결국 CTV 마케팅에서 좋은 성과를 내는 경우는 **고품질 크리에이티브와 강력한 스토리텔링이 있는 앱**이라고 정의할 수 있어요. CTV의 몰입형 경험에서 가장 큰 이점을 얻을 수 있죠.
## Vibe.co는 어떻게 모바일 앱 마케터들이 효과적인 CTV 캠페인 운영을 지원하나요?
기존 TV 광고는 복잡한 협상과 긴 준비 과정을 거쳐야 했지만, CTV는 훨씬 빠르고 쉽게 시작할 수 있어요. Vibe.co와 함께라면:
- 간단한 셋업으로 캠페인을 빠르게 시작할 수 있고, 전통적인 TV 광고의 복잡함을 없앨 수 있어요.
- 프리미엄 스트리밍 서비스의 브랜드 세이프 환경에서 광고 인벤토리에 접근할 수 있어요.
- 첨단 기술과 전담 매니저 지원을 통해 캠페인을 손쉽게 최적화할 수 있어요.
- 실시간 리포팅을 활용해 모바일 DSP처럼 성과를 관리할 수 있어요.
정리하자면, Vibe.co는 모바일 UA 채널처럼 CTV 캠페인도 빠르게 테스트하고 확장할 수 있도록 지원해요.
## **Vibe.co의 입찰 알고리즘은 어떻게 작동하나요? **
Vibe.co는 Ai 기반 입찰 알고리즘으로 전환 가능성이 높은 유저를 자동으로 찾아내고 있어요. 입찰 알고리즘의 작동 원리는 다음과 같은데요.
- **목표 기반 최적화:** 마케터가 캠페인 목표(설치, 구매, 매출)를 설정하면 시스템이 가치 높은 유저를 우선적으로 타겟팅해요.
- **머신러닝 모델:** 캠페인 성과를 반영해 타겟팅을 지속적으로 정교하게 개선해요.
- **포스트백 지원:** Vibe.co는 에어브릿지 같은 MMP와 연동해 데이터 기반 최적화를 가능하게 해요.
위와 같은 체계적인 방식 덕분에 CTV는 단순한 브랜드 광고가 아니라, 성과 중심의 퍼포먼스 채널이 될 수 있는 것이죠.
## **광고주는 어떻게 크리에이티브를 업로드하고 성과를 추적할 수 있나요?**
Vibe.co의 직관적이고 쉬운 크리에이티브 관리 툴을 이용하면 광고주는 영상 에셋을 쉽게 업로드하고 여러 광고 버전을 세팅할 수 있어요. 캠페인이 시작되면, 마케터는 에어브릿지 같은 MMP를 통해 다음 지표를 측정하죠.
- **전환 퍼널:** 광고 시청자가 얼마나 앱을 설치했는지 추적해요.
- **리텐션 지표:** D1, D7, D14 시점에서 유저가 얼마나 앱에 머무는지 확인해요.
- **코호트 기반 ROAS 분석:** 유저 LTV를 분석해 크리에이티브별로 예산을 최적화할 수 있어요.
이렇게 성과 데이터를 기반으로 크리에이티브를 계속 다듬으면서, 최적의 크리에이티드를 통한 최대 매출을 만들어낼 수 있어요.
## CTV는 타겟팅 옵션이 제한적이라는 인식이 있는데요. Vibe.co는 어떤 타겟팅 기능을 제공하나요?
흔히들 비슷한 오해를 하고 계신 게 사실이에요. CTV 타겟팅은 모바일 DSP 못지않게, 아니 오히려 더 정교할 때도 있어요. Vibe.co가 디테일한 타겟팅을 위해 마케터 여러분들께 제공하는 기능은 다음과 같아요.
- **오디언스 기반 타겟팅:** 게이머, 스포츠 팬, 교육 관심 유저 등 특정 그룹을 공략할 수 있어요.
- **자체 SSP:** 삼성, LG, Philips 같은 주요 TV 제조사와 Hulu, Peacock, Tubi 등 스트리밍 서비스, 그리고 Roku, DirecTV, Pluto 같은 인기 채널 인벤토리와 직접 연동돼 있어요.
- **컨텍스트 타겟팅:** 스포츠, 엔터테인먼트, 뉴스 관련 채널에 광고를 집행할 수 있어요.
- **IP 기반 타겟팅:** 퍼스트 파티 IP 리스트를 업로드해 CTV 유저와 데이터를 매칭하고 정밀 리타겟팅을 할 수 있어요.
- **가구 단위 타겟팅:** 여러 사용자가 있는 가정에서도 광고가 적절한 시청자에게 도달하게 할 수 있어요.
위와 같이 CTV에서도 세분화된 카테고리를 통해 정교한 타겟팅이 가능해요.
## 모바일 앱 마케터들은 Vibe.co에서 CTV 캠페인 성과를 어떻게 측정할 수 있나요?
Vibe.co는 에어브릿지 같은 MMP와 연동해 상세한 측정과 어트리뷰션을 제공하고 있어요.
- **뷰스루 어트리뷰션(VTA):** 광고를 클릭하지 않아도 나중에 설치로 이어지면 추적할 수 있어요.
- **크로스 디바이스 추적:** CTV 노출이 모바일 설치와 매출에 어떤 영향을 주는지 측정해요.
- **퍼널 기반 분석:** CTV 노출이 D7/D14 리텐션, LTV, ROAS까지 어떻게 이어지는지 연결해요.
이렇게 CTV 인사이트를 모바일 UA 전략과 결합하면, 단순 노출이 아니라 **실질적인 유저 성장**을 최적화할 수 있어요.
## Vibe.co를 효과적으로 활용한 성공 사례를 소개해주실 수 있나요?
좋은 예로 모바일 게임사 [**Bagelcode**](https://www.bagelcode.com/)가 있어요. 이들은 Club Vegas 캠페인을 확장하기 위해 Vibe.co의 CTV 광고를 활용했어요. 목표는 ROAS를 개선하고 유저를 다시 참여시키는 것이었죠.
양사간 협업의 성과는 놀라웠어요.
- CTV Ad Vision 기반 최적화된 어트리뷰션으로 **ROAS가 3배 향상**됐어요.
- D7 지표를 꾸준히 두 배 이상 달성했어요.
- 다양한 영상 포맷을 테스트해 참여도를 극대화했어요.
- 강력한 성과 덕분에 불과 한 달 만에 CTV 예산을 두 배 이상 늘렸어요.
Bagelcode는 스마트 타겟팅, 실시간 최적화, 확장 가능한 예산 전략을 결합해 성공적으로 유저를 재참여시키고 캠페인 수익성을 극대화했어요. 이 사례는 CTV가 모바일 앱 성장에 얼마나 효과적인 채널인지 잘 보여줘요.
## 미국 시장에서 CTV 광고로 확장하려는 모바일 앱 마케터들에게 조언해주신다면요?
미국시장에서 CTV 캠페인을 성공적으로 진행하기 위해서는 다음 사항을 지킨다면 CTV를 단순히 브랜딩 채널이 아닌 강력한 퍼포먼스를 만들어 내는 광고 채널로 활용할 수 있어요.
- **목표를 명확히 정의하기:** CPI, ROAS, 설치 수 등 명확한 KPI 설정.
- **고품질 영상 광고 활용하기:** 대형 화면에 최적화된 몰입도 높은 콘텐츠 제작.
- **명확한 CTA 추가하기:** 앱스토어와 구글 플레이 로고를 넣어 설치로 바로 이어지는 유저 여정 설계.
- **크로스 디바이스 행동 추적하기:** 유저는 CTV에서 모바일로의 크로스 디바이스 행동 추적.
- **지속적으로 최적화하기:** 모바일 DSP처럼 CTV 캠페인도 지속적인 테스트와 조정을 통한 최적화.
## PC·콘솔 게임 개발자들은 어떻게 CTV 광고를 활용해 타겟 유저에게 도달할 수 있을까요?
PC와 콘솔 게임 개발사는 기본적으로 CTV를 통해 대형 화면에서 고품질 그래픽과 몰입도 높은 게임플레이를 보여줄 수 있다는 장점이 있는데요. 다음과 같은 점도 게임이 CTV와 잘 맞는 이유에요.
- **그래픽 강조:** CTV는 고화질로 게임의 인상적인 비주얼을 보여줄 수 있어요.
- **게이밍 팬 타겟팅:** 게임 콘텐츠에 관심 있는 시청자에게 도달해 전환 가능성을 높일 수 있어요.
- **크로스 플랫폼 프로모션:** PC 게임을 콘솔 유저에게, 혹은 반대로도 홍보할 수 있어요.
- **런칭 캠페인:** 신작 출시나 대규모 업데이트를 알리는 데 CTV의 넓은 도달력을 활용할 수 있어요.
## CTV 광고가 모바일 게임과 비교해 PC·콘솔 게임에 주는 독특한 장점은 무엇인가요?
PC·콘솔 게임과 CTV는 특히나 잘 맞는 조합인데요. 그 이유는 아래와 같은 특성 때문이에요.
- **게이밍 환경과의 일치:** 많은 PC·콘솔 게이머들이 이미 대형 화면에서 플레이하기 때문에, CTV 광고가 더 자연스럽게 다가와요.
- **하이엔드 게임 홍보:** CTV의 고품질 포맷은 고사양 그래픽과 복잡한 게임플레이를 보여주기에 최적이에요.
- **긴 몰입 시간과 맞춤:** PC·콘솔 게임은 플레이 시간이 길기 때문에, CTV 시청자의 몰입 습관과 잘 맞아요.
## CTV 광고는 PC·콘솔 게임 개발자들의 유저 확보에 어떻게 기여할 수 있을까요?
CTV는 PC·콘솔 게임의 유저 확보를 크게 향상시킬 수 있어요.
- **크로스 디바이스 추적:** CTV 노출이 스팀이나 콘솔 스토어에서의 구매·다운로드로 어떻게 이어지는지 등 유저 여정 파악.
- **브랜드 리프트 스터디:** CTV 캠페인이 게임 인지도와 관심을 얼마나 높였는지에 대한 판단.
- **리타겟팅 기회:** 시청자 데이터를 활용해 다른 기기에서 리타겟팅.
- **성과 지표 추적:** CPA, ROAS 같은 지표를 추적을 통한 캠페인 최적화.
에어브릿지 같은 MMP 솔루션을 활용하면 실시간으로 성과를 측정할 수 있죠.
## PC·콘솔 게임에서 CTV 광고를 효과적으로 사용한 성공 사례를 공유해주실 수 있을까요?
PC 및 콘솔 게임을 비롯한 다양한 성공 사례를 vibe.co 케이스스터디에서 확인할 수 있어요. 핵심은 PC 및 콘솔 게임도 CTV의 시각적 몰입력과 타겟팅 정밀도를 활용해 성과를 크게 끌어올릴 수 있다는 점이라고 할 수 있어요.
## CTV를 통한 모바일 성장 인사이트
CTV 광고 업계를 선도하는 팀들이 전하는 CTV 광고 관련 인사이트가 궁금하신가요? 지금 아래 링크를 클릭하고 CTV 인터뷰 시리즈를 읽어 보세요.
👉 [Motiv Intelligence: 모바일-TV 매칭을 통한 효율적인 CTV광고 캠페인 집행](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-motiv)
👉 [Moloco: 성과로 증명되는 CTV 광고의 효율](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance)
---
## [CTV 인사이트 인터뷰1] Motiv Intelligence: 모바일-TV 매칭을 통한 효율적인 CTV광고 캠페인 집행
URL: https://www.airbridge.io/ko/blog/airbridge-ctv-interview-motiv
Language: Korean
Published: 2025-09-04
모바일 앱 마케터들은 언제나 유저 확보(UA)에 대한 고민을 달고 살죠. 그 어느 때보다 경쟁이 심해진 시장에서 좋은 성과가 나는 채널을 찾는 일이란 쉽지 않아요.
CTV(커넥티드 TV) 광고는 최근 새로운 UA 채널을 찾는 마케터들에게 새로운 기회로 주목받고 있어요. 하지만 아직 CTV가 UA 전략에서 어떤 역할을 할 수 있는지에 대한 정보는 부족한 게 사실이에요.
에어브릿지에서 새롭게 떠오르는 CTV 광고에 대해 궁금해하시는 앱 마케터 여러분을 위한 인터뷰 시리즈를 기획했어요. CTV 광고 인터뷰 시리즈 그 첫 번째 시간으로 국내 최대 규모의 유저 행동 데이터와 다양한 미디어를 바탕으로 맞춤형 타겟팅을 지원하는 Motiv Intelligence(이하 모티브인텔리전스)를 만났어요.
[모티브인텔리전스](https://www.motiv-i.com/)가 전하는 CTV의 뜻과 중요성, 그리고 활용 방법에 대한 인사이트를 지금 바로 확인해 보세요.
## 2025년, 한국의 CTV 광고 시장은 어떻게 변화하고 있나요?
CTV 광고 시장의 변화에 대해서 이야기하려면 CTV 광고가 노출되는 CTV 디바이스를 둘러싼 환경을 먼저 이해할 필요가 있어요. CTV란 Connected TV를 의미하는데 여기서 Connected는 인터넷에 연결되어 있다는 의미에요. 기존의 전통적인 TV가 채널을 수동적으로 시청하는 방식에 머물렀던 것과 달리, CTV는 인터넷에 연결되어 있다 보니 OTT 서비스 이용, 웹 검색, 게임 등 다양한 인터넷 기반 서비스의 능동적인 사용이 가능하죠.
한국은 글로벌 최고 수준의 인터넷 가입률을 보유하고 있으며, 2023년 기준 스마트TV 보급률이 75%에 이를 정도로 높은 수준을 기록하고 있어, CTV 광고를 활용하기에 최적의 조건을 갖추고 있어요.
특히 최근 들어 TV에서의 유튜브 시청 증가, 넷플릭스/티빙 등의 OTT 서비스의 광고 도입, 삼성전자와 LG전자의 스마트TV 내 광고 기반 무료 스트리밍 TV(FAST: Free Ad-supported Streaming TV) 서비스 강화 등 다양한 변화가 이루어지고 있어요. 이로 인해 시청자들의 TV 시청 패턴이 빠르게 변화하고 있으며, 광고 시장에서도 CTV에 대한 관심이 점점 커지고 있죠.
모티브인텔리전스는 국내 최초로 프로그래매틱 CTV 광고 플랫폼을 런칭하여 다양한 파트너들과 협업을 통해 국내 CTV 광고 시장을 개척하고 있어요. 위에서 설명드린 시장 변화의 영향으로 광고주들의 CTV 광고에 대한 관심이 높아지고 있으며, 당사가 발행한 2025 국내 CTV 광고 집행 현황 리포트에 따르면, 2023년 대비 2024년의 CTV 광고 캠페인 재집행 비율이 1.4배 늘어난 것으로 나타났어요. 이는 광고주들 사이에서 CTV 광고가 성과가 검증된 주요 매체로 빠르게 자리잡고 있다는 점을 보여주고 있어요.
## 모바일 앱 마케터들이 CTV 캠페인에 투자해야 하는 이유는 무엇인가요?
TV 광고는 전통적으로 전략적인 광고 매체로 활용되었으며, 대부분의 광고 전략에서 핵심적인 위치를 차지하고 있어요. 이는 TV가 개인이 보유한 디바이스 중에 가장 큰 디바이스로, 대형 스크린과 고품질 음향을 통해 높은 주목도와 신뢰도를 제공하기 때문이죠. TV광고 자체의 광고 효과도 있지만 글로벌 광고대행사 MediaCom이 조사한 바에 따르면 다른 광고 매체의 효과를 부스팅하는 효과가 크다는 장점도 있어요.
그러나 기존 TV 광고는 디지털 광고에 비해 노출당 비용이 높고, 탄력적인 광고 구매와 운영이 쉽지 않은 광고 유통 구조로 인해 모바일 앱 마케터, 특히 퍼포먼스 마케터들에게는 활용성이 높지 않았던 것이 사실이죠. 또한, 데이터 기반의 타겟팅, 성과 측정 그리고 최적화가 중요한 퍼포먼스 마케터들에게는 디지털 데이터를 활용할 수 없는 TV 광고는 매력적인 선택지가 아니었어요.
하지만 인터넷에 연결된(Connected) TV에 노출되는 CTV 광고는 실시간 데이터를 활용한 타겟팅과 성과 측정이 가능할 뿐 아니라, 프로그래매틱 방식으로 노출당 단가(CPM)로 실시간 구매가 가능해요. 전통 TV 광고의 장점과 디지털 광고의 장점을 모두 갖춘 매체로, 이미 해외 시장의 모바일 앱 마케터들은 CTV 광고를 적극 활용하고 있어요.
즉, 실시간 데이터를 기반으로 원하는 타겟에게 맞춤형 광고를 노출하고, 에어브릿지와 같은 MMP로 성과를 측정하고 캠페인을 최적화할 수 있는 TV 광고 환경이 이미 국내에 준비되어 있기 때문에 다양한 UA의 채널을 발굴하는 데 관심이 있는 모바일 앱 마케터라면 CTV 광고에 대해 관심을 갖는 것을 권장해요.
[]
## 여전히 많은 모바일 마케터들이 CTV를 낯설게 느끼고 있는데요. 어떤 카테고리 앱들이 CTV 광고에서 좋은 성과를 내고 있나요?
TV 광고가 디지털 광고와 유사한 수준의 타겟팅과 성과 측정이 가능하다는 장점 때문에 저희와 협업하는 광고주가 빠르게 증가하고 있는 추세예요. 주요 광고주 업종으로는 주류/음료, 자동차, 가전, 패션, 식품 등이 있고, 앱 광고주로는 지역 기반 커뮤니티 앱을 포함하여 게임, 게임 포인트 앱, 명품 쇼핑 앱 등의 사례가 있었는데, 2024년도 집행한 전체 광고 캠페인 중 63%가 2개월 이상 광고를 집행할 정도로 높은 만족도를 보이고 있죠.
모티브인텔리전스의 CTV 광고는 MMP(모바일앱), CTS(Conversion Tracking Script, 모바일/PC 웹), DMP 데이터 수집 및 분석을 통해 브랜드 리프트와 같은 브랜딩 성과 및 구매 전환과 같은 퍼포먼스 성과 측정을 지원하고 있어요.
## 모티브인텔리전스는 어떻게 모바일 앱 마케터들이 효과적인 CTV 캠페인을 운영할 수 있도록 지원하나요?
CTV 광고는 디지털 방식으로 작동하지만, 직접적인 클릭이 불가능하기 때문에 광고의 노출로부터 실제 사용자의 인앱 액션까지의 데이터의 흐름을 이해하고 분석하는 데는 모바일 광고와는 다른 전문성이 필요해요. 또한, 보다 정교한 오디언스 타겟팅을 위해 캠페인 기획 단계부터 다양한 DMP 데이터를 활용하기 때문에 DMP에 대한 이해와 세그먼트 추출에 대한 전문성도 필요하죠.
모티브인텔리전스는 자체적으로 CTV 광고와 모바일 광고 플랫폼을 모두 운영하고 있기 때문에, 보다 입체적이고 효과적인 광고 성과를 위해 TV-모바일 간의 크로스 디바이스 광고를 지원하고 있어요.
이처럼 CTV 광고 환경은 모바일 앱 마케터들에게 아직 익숙치 않은 요소들이 많지만, 모티브인텔리전스는 크로스 디바이스 플랫폼 운영 경험과 데이터 기반의 광고 운영 경험이 풍부하여 모바일 마케터들의 눈높이와 캠페인 니즈에 맞춘 CTV 광고 컨설팅을 제공해 드리고 있어요.
## 모바일 앱 마케터들에게 모티브인텔리전스의 CTV 광고 입찰 알고리즘을 포함하여 작동 방식을 간단히 설명해 주실 수 있나요?
모티브인텔리전스의 CTV 광고 플랫폼은 국내 최초로 자체 개발하여 특허를 확보한 모바일-TV 간 매칭 기술이 가장 큰 특징이에요. 이러한 모바일-TV 매칭 기술에 더해 국내 최대의 CTV 광고 인벤토리를 확보하고 있어, 모바일 앱 마케터들이 정교한 모바일 행동 이력 기반의 오디언스 레벨 타겟팅 광고를 집행할 수 있도록 지원하죠. 또한, 에어브릿지를 포함한 주요 MMP와 연동되어 있기 때문에 TV 광고이지만 모바일 광고처럼 성과 측정을 위한 포스트백을 제공해 실시간으로 광고 성과를 모니터링하면서 최적화를 진행할 수 있다는 장점이 있어요.
기존 UA를 통해 확보된 사용자나 이탈한(Churn-out) 사용자를 타겟팅하여 TV 광고를 집행하고, MMP 데이터를 활용해 성과를 측정하고 최적화하는 것이 가능해요.
모티브인텔리전스는 국내 최다인 6개의 DMP(Data Management Platform)를 연동하여 사용하기 때문에, 광고주가 원하는 커스텀 타겟 세그먼트를 DMP로부터 추출하여 TV로 타겟팅 광고를 집행할 수 있어요. 예를 들면, 앱 런칭 캠페인 초기 UA 단계에서 경쟁사 또는 유사 앱 사용자를 추출하여 타겟팅하거나, 특정 제품의 구매 이력이 있거나 특정 장소/방문 이력이 있는 세그먼트를 추출하여 TV에 타겟팅 광고를 집행할 수 있죠.
## 모티브인텔리전스를 통해 효과적으로 CTV 광고를 운영한 모바일 앱 성공 사례를 공유해 주실 수 있나요?
MMP를 통해 성과 측정을 진행했던 지역 기반 커뮤니티 앱의 경우, 인앱 구매 전환율이 모바일 광고 대비 116%를 달성했고, CTS를 통해 성과측정을 진행했던 아웃도어 브랜드의 경우, 매출액 ROI 약 450%를 달성했으며, DMP 데이터 수집 및 분석을 통해 퍼포먼스 성과를 측정했던 커피 음료 브랜드의 경우, 광고 기여도 315%를 달성했습니다. 자세한 내용은 홈페이지의 [크로스타겟TV 상품소개서](https://www.motiv-i.com/crosstarget-tv/files?file=introduction)에서 확인하실 수 있어요.
## CTV 광고로 확장을 고려하는 모바일 앱 마케터들에게 주실만한 조언이 있다면 무엇인가요?
모바일 앱 및 게임 마케터들에게 그동안 모바일 광고가 핵심적인 역할을 해왔다는 점을 잘 이해하고 있어요. 하지만 마케팅 퍼널에 있어 모든 단계는 중요하고, 각각의 단계에서의 성과가 다음 단계의 성과뿐만 아니라 그 성과의 유지 기간에도 영향을 미치게 되죠. 따라서, 마케터들은 소비자들과 소통하기 위해 그들의 미디어 소비 전 과정에서 정교한 메시지를 효과적으로 인식시켜야 해요.
그런데 소비자들은 이제 하나의 디바이스가 아닌 다양한 디바이스를 통해 콘텐츠를 소비하고 있죠. 스마트폰, 스마트패드, 스마트TV가 대표적인 스마트 디바이스인데 최근 방송통신위원회가 발표한 방송매체이용행태조사에 따르면, 국내 TV 보급률과 스마트폰 보급률은 2024년 기준 95%로 비슷한 수준으로 나타났고, 한국지능정보사회진흥원의 발표에 따르면, 2023년 기준 국내 TV 중 스마트TV 비중이 75%를 넘어섰어요. 스마트TV는 과거처럼 일방적인 방송을 시청하는 용도로만 사용되지 않고, 넷플릭스와 유튜브를 비롯한 다양한 OTT 서비스와, FAST 등의 다양한 동영상 콘텐츠를 소비하는 핵심 디바이스로 자리 잡고 있죠.
실제로, 방송통신위원회 발표 자료에 의하면, 2024년도 기준 국내 평균 일간 TV 사용 시간은 144분으로 스마트폰 사용시간(126분) 보다 14% 높다고 조사되었어요. 그리고 앞서 살펴본 것처럼 TV는 기본적으로 디바이스의 성능이 뛰어나 광고 주목도가 높은 것으로 알려져 있는데, 2024년 글로벌 광고 대행사 OMG와 야후가 함께 발표한 조사에 의하면, TV 광고는 모바일 광고보다 8배, PC 광고보다 16배 광고 주목 시간이 길다고 해요.
최근 모바일 광고 환경에서는 개인정보 보호 이슈가 강화되고, 광고 수요 증가로 인해 광고 비용이 지속적으로 상승하면서, 많은 마케터들이 ROI를 개선할 수 있는 대안을 모색하고 있어요. 이러한 흐름 속에서 MMM(Media Mix Modeling)이 주목받고 있지만, 기존의 MMM 방법론은 모바일 기반 미디어 중심으로 논의되어 소비자들의 미디어 소비 패턴을 입체적으로 반영하지 못하는 한계가 있어요.
하지만 CTV 광고의 등장으로 인해 TV에서도 데이터 기반의 디지털 광고 집행과 성과 측정이 가능해졌죠. MMM의 방법론을 TV 광고까지 확장하는 것 또한 가능해져, 마케팅 퍼널에서의 소통 전략을 보다 입체적으로 수립할 수 있는 기반이 마련되었죠. 그렇기 때문에 소비자 중심의 마케팅 전략을 고민하는 마케터라면, CTV라는 디바이스에 머무르고 있는 소비자들의 ‘시청 시간’을 어떻게 전략적으로 공략하여 자사의 브랜드 또는 상품을 인지 시킬 것인지를 고민할 필요가 있다고 생각해요.
## CTV를 통한 모바일 성장 인사이트
CTV 광고 업계를 선도하는 팀들이 전하는 CTV 광고 관련 인사이트가 궁금하신가요? 지금 아래 링크를 클릭하고 CTV 인터뷰 시리즈를 읽어 보세요.
👉 [Vibe.co: 미국 시장 진출. AI 기반 알고리즘으로 CTV 광고 성과 극대화하기](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-vibe)
👉 [[CTV 인사이트 인터뷰 2] Vibe.co: 미국 시장 진출. AI 기반 알고리즘으로 CTV 광고 성과 극대화하기](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-vibe)
👉 [[CTV 인사이트 인터뷰 3] Moloco: 성과로 증명되는 CTV 광고의 효율](https://www.airbridge.io/ko/blog/airbridge-ctv-interview-moloco-ad-efficiency-and-mobile-app-performance)
---
## Firebase Dynamic Links 중단, 최고의 대안은?
URL: https://www.airbridge.io/ko/blog/firebase-dynamic-links-alternatives
Language: Korean
Published: 2025-08-18
파이어베이스 다이나믹 링크(Firebase Dynamic Link)의 서비스 중단이 코앞으로 다가왔어요. 여러분들은 다이나믹 링크 서비스 중단에 대응하고 계신가요? 대안을 아직 찾지 못하셨거나 자체 개발에 어려움을 겪고 있으시진 않나요?
[]
8월 25일 이후에는 기존에 사용 중이던 링크가 모두 동작을 멈출 뿐만 아니라, 마이그레이션을 위한 링크 추출이 불가능해질 수도 있어요. 딥링크는 광고, 푸시 메시지를 비롯한 CRM 마케팅, 공유하기 이벤트와 오프라인 QR 코드 등 다양한 이유로 모바일 앱을 운영할 때 꼭 필요한 기능이기 때문에 다이나믹 링크를 사용 중인 앱이라면 무엇보다 먼저 서비스 중단 대비가 꼭 필요한 시기이죠.
이번 블로그에서는 파이어베이스 다이나믹 링크의 대표적인 대안 솔루션인 에어브릿지 딥링크 플랜을 다른 솔루션과 함께 비교해 보고 여러분의 환경에 알맞은 솔루션을 선택하실 수 있도록 도와드릴게요.
## 파이어베이스 다이나믹 링크, 대표적인 대안은 무엇이 있나요?
파이어베이스 다이나믹 링크가 사라진 뒤 사용할 수 있는 대표적인 솔루션으로는 에어브릿지의 딥링크 플랜과 A사, B사가 운영하는 딥링킹 전용 서비스가 있어요. 세 솔루션 모두 [파이어베이스에서 공식적으로 언급한 다이나믹 링크의 대안](https://firebase.google.com/support/dynamic-links-faq?hl=ko#what_should_i_do_if_i_need_full_feature_parity_with_firebase_dynamic_links)이라는 공통점이 있죠.
### 기능은 어떻게 다를까요?
이들은 공식적으로 언급된 대안 솔루션이라는 공통점이 있지만, 세세하게 살펴보면 제공되는 기능 범위부터 가격, 채널 연동 등 다양한 부분에서 차이점이 있는데요. 그렇다면, 이 세 솔루션은 어떻게 다른지 지금부터 확인해 보세요.
우선, 가장 중요한 기능적 차이는 어떻게 될까요?
[]
많은 분들이 궁금해하는 링크 마이그레이션 방법과 앱이 설치되지 않은 유저도 설치 후 원하는 페이지로 이동시켜줄 수 있는 디퍼드 딥링크, 오프라인 마케팅에서 중요한 요소인 QR코드 제공 여부, 링크 생성 갯수와 API 사용 제한 등을 꼼꼼히 둘러보면 세 솔루션은 확연한 차이가 있어요.
### 한국 모바일 환경 최적화는 어떻게 다를까요?
국내 모바일 시장에서 딥링크를 더 잘 활용하기 위해서는 크롬, 사라피 등 브라우저는 물론 메타, 구글, 왓츠앱, 라인 등 글로벌 채널은 물론 네이버, 카카오와 같은 국내 주요 채널에서 링크가 문제없이 오픈되어야 하겠죠.
또한 솔루션 사용 중 문제가 발생했을 때 지원받을 수 있는 한국어 서비스가 있다면 솔루션 이용이 더욱 편할 거에요. 세 솔루션은 한국 모바일 환경에 어떻게 대응하고 있을까요?
[]
## 변동없는 가격, 한국 모바일 환경에 잘 맞는 에어브릿지 딥링크 플랜
에어브릿지 딥링크 플랜은 MMP와 동일한 수준의 딥링킹을 추가 금액 없이 제공하고 있어요. 또한 한국 모바일 환경에 가장 잘 맞는 딥링킹 솔루션이죠.
에어브릿지 딥링크 플랜의 장점을 정리하면 다음과 같아요.
- **1분 마이그레이션**: 기존에 사용하던 파이어베이스 다이나믹 링크를 CSV로 다운 받아 에어브릿지 시트에 붙여 넣으면 1분 안에 모든 링크 마이그레이션 가능.
- **국내 채널 특화**: 네이버·카카오와 같은 국내 주요 매체 및 다양한 OS 업데이트와 정책 변경에 빠르게 대응되는 딥링크.
- **변동 없는 가격**: 연간 계약으로 유동성 없는 가격 제공.
## 에어브릿지 딥링크 전문가와 지금 이야기를 나누어보세요
다가오는 파이어베이스 다이나믹 링크 중단. 더이상 고민만 하고 기다릴 수는 없어요. 이미 국내외 120여 개 앱에서 에어브릿지 딥링크 플랜을 사용하여 파이어베이스 다이나믹 링크를 대체하고 있어요. 지금 에어브릿지 딥링크 플랜 데모를 신청하고 오늘 기존 링크를 새로운 링크로 변경하세요.
---
## 애디슨 오퍼월 오디언스 연동 지원, 정교한 리워드 광고를 가장 먼저 만나보세요
URL: https://www.airbridge.io/ko/blog/adison-offerwall-audience-sync
Language: Korean
Published: 2025-07-17
보상형 광고를 집행하고 있다면 반가운 소식이 있어요. 에어브릿지가 엔비티와 손을 잡고 MMP 최초로 애디슨 오퍼월 오디언스 연동을 지원해요. 이제 광고주는 앱 내 실사용자 데이터를 기반으로 더 정교하고 효과적인 광고를 가장 먼저 운영할 수 있어요.
## 애디슨 오퍼월 알아보기
[엔비티](https://nbtads.oopy.io/)는 오퍼월 네트워크 ‘애디슨 오퍼월’을 운영하는 국내 최대 규모 보상형 광고 플랫폼이에요. 애디슨 오퍼월은 네이버웹툰, 네이버페이, 토스 등 프리미엄 매체와의 단독 제휴는 물론, 국내 500개 이상의 보상형 매체와 연동되어 다양한 리워드 광고를 한 번의 세팅으로 동시에 집행할 수 있어요.
앱 설치 유도형 광고나 첫 화면 미디어에 노출되는 광고를 집행하고 있다면 애디슨 오퍼월은 강력한 마케팅 채널이 될 수 있어요. 높은 도달률과 보상 기반의 높은 참여율을 기반으로, 다양한 업종의 광고주가 성과형 UA(User Acquisition)를 위해 활용하고 있죠.
## 애디슨 오퍼월 오디언스 연동으로 기대되는 점
에어브릿지의 오디언스 연동 기능을 통해, 광고주는 앱 유저 행동 기반의 오디언스 세그먼트를 만들어 애디슨 오퍼월에 실시간으로 전송할 수 있어요.
이번 연동으로 지원되는 주요 기능은 다음과 같아요:
### 1\. 고객 행동 기반 맞춤 타겟팅
에어브릿지에서 앱 설치, 회원가입, 구매, 재방문 등 특정 행동 조건을 만족하는 오디언스를 정의하고, 이 세그먼트를 애디언스오퍼월에 연동해 더욱 정교한 타겟 광고를 집행할 수 있어요.
### 2\. 실시간 데이터 동기화
오디언스 세그먼트는 에어브릿지에서 자동으로 애디슨 오퍼월에 일 단위로 동기화돼요. 이제 타겟팅 광고는 실시간 사용자 행동을 반영한 최신 데이터를 기반으로 운영할 수 있어요.
애디슨 오퍼월은 높은 참여율과 리워드 기반의 강력한 광고 몰입도를 보유한 채널이에요. 여기에 에어브릿지의 오디언스 연동을 더하면 정확한 유저 타겟팅과 광고 예산의 효율적 집행, 더 높은 전환율까지 기대할 수 있어요.
또한 에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 크리테오, 네이버, 카카오, 토스 등 다양한 국내외 매체와의 연동을 폭넓고 발빠르게 지원하고 있어요. 이번 애디슨 오퍼월 연동을 통해 리워드 기반 광고 채널까지 정교한 오디언스 마케팅을 확장할 수 있게 되었어요.
오디언스를 광고 채널에 전송하는 방법은 [이 가이드](https://help.airbridge.io/ko/guides/send-audience#%EC%98%A4%EB%94%94%EC%96%B8%EC%8A%A4%EB%A5%BC-%EC%A0%84%EC%86%A1%ED%95%A0-%EC%88%98-%EC%9E%88%EB%8A%94-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-1)를 참고해 주세요.
## 마무리하며
이제 애디슨 오퍼월에서도 에어브릿지의 오디언스 연동을 활용해, 더 정밀하고 효과적인 광고를 운영할 수 있어요. 특히 보상형 광고를 집행하는 앱 마케터라면, 이번 연동을 통해 성과 중심의 캠페인 운영 전략을 한 단계 끌어올릴 수 있을 거예요.
에어브릿지 데모를 신청하고, 오디언스 연동 기능을 직접 경험해 보세요.
---
## 중견 게임 스튜디오를 위한 스팀(Steam) 게임 마케팅 완전정복
URL: https://www.airbridge.io/ko/blog/game-studio-steam-marketing-playbook
Language: Korean
Published: 2025-07-02
여러분은 어떤 플랫폼을 통해 게임을 플레이하시나요? 아마 PC를 이용하는 많은 분들이 이 질문에 스팀(Steam)이라고 대답하실 텐데요. 가장 대중적인 게임 플랫폼이자 수백만 명의 PC 게이머가 모이는 커뮤니티인 스팀은 중견 게임 스튜디오에게도 중요한 플랫폼이 되어가고 있어요. 게임의 글로벌 성공과 더 많은 플랫폼 진출의 기회를 찾기 위해서는 스팀 출시는 이제 선택이 아닌 필수예요.
하지만 지금까지 알려진 스팀 게임 마케팅 방법론은 글로벌 IP나 블록버스터급 트레일러 및 시네마틱 영상을 갖춘 트리플A급 게임이나 예산이 거의 없는 인디 게임에 초점이 맞춰져 있었어요. 중견 게임 스튜디오를 위한 마케팅 전략은 많이 알려져 있지 않았죠.
에어브릿지 남성필 대표가 Liquid Advertising의 글로벌 미디어 부사장 Nikki DePaola와 함께 중견 게임 스튜디오를 위한 스팀 마케팅 전략에 대해 이야기를 나누었어요.
Liquid Advertising은 지금까지 수많은 PC 및 콘솔 프랜차이즈의 성공적인 출시를 돕고 있는 팀이에요. 블리자드, 라이엇 게임즈, 넥슨과 코나미 등 다양한 파트너와 함께 하고 있죠. 스팀 진출을 시도하는 게임 스튜디오가 어떤 전략으로 성공하고, 어디에서 실수를 일으키는지 등을 누구보다 잘 알고 있어요..
이번 블로그에서는 Nikki와 남성필 대표의 인터뷰 세션의 핵심을 정리해 드릴게요.
**🎥 전체 인터뷰 영상 보기**
## 스팀 페이지 제작과 트레일러 광고는 시작에 불과해요
많은 게임 스튜디오에서 스팀 출시 페이지와 트레일러를 만들고 이를 바탕으로 광고를 집행하는 게 마케팅의 전부라 생각하곤 해요. 하지만 게임 마케팅 경쟁이 심화되고 있는 요즘, 이정도 마케팅 액션만으로 좋은 성과를 기대하긴 어려워요.
스팀 유저들은 대부분 게임만 보고 구매를 결정하지 않고 흐름에 따라 움직이기 때문이에요. 입소문을 통해 게임을 접하고 다른 유저들의 반응을 확인하고, 신뢰할 수 있는 매체에서 언급되는 이야기를 읽으며 구매를 고민하게 되죠.
> “**위시리스트는 사람들이 게임에 얼마나 기대하고 있는지를 보여주는 바로미터에요. 이러한 관점에서 보면 자연스럽게 목표는 단순히 위시리스트 숫자를 채우는 것이 아닌 ‘어떻게 하면 이 게임에 대한 진짜 기대를 갖게 만들까’로 바뀌게 되죠.**”
Nikki는 성공적인 스팀 게임 출시 마케팅에 꼭 포함되어야 할 여섯 가지 요소를 이렇게 정리해요.
- **명확한 후킹 포인트와 임팩트 있는 트레일러가 담긴 스팀 페이지
**
- **유저들이 새로운 게임을 발견하고 도달할 수 있는 유료 광고 캠페인
**
- **특히 Discord와 Reddit을 중심으로 한 커뮤니티 유저들과의 실시간 소통
**
- **게이머들이 신뢰하는 매체를 통해 다뤄질 수 있는 진짜 PR 전략
**
- **티저 공개, 데모 배포, 베타 런칭 등 타이밍을 나눠 전달하는 콘텐츠 롤아웃 계획
**
- **무엇이 효과적인지 신뢰할 수 있는 데이터로 확인할 수 있는 분석 솔루션(에어브릿지와 같은 성과 측정 도구)**
## 위시리스트 수, 진짜 수요를 반영할 때만 의미가 있어요
스팀 위시리스트는 게임 성공에 가장 중요한 지표 중 하나에요. 출시 전 단계에서 확인할 수 있는 몇 안 되는 수치이기 때문이죠. 유저들의 위시리스트에 많이 담긴 게임이라면 스팀의 게임 추천 알고리즘 노출에도 좋은 영향을 받을 수 있어요.
하지만 위시리스트 숫자가 높다고 꼭 성공이 보장되는 건 아니에요.
> “**차트 순위에 올라서 기뻐했는데, 막상 게임이 출시되고 나면 전환율이 실망스러운 경우를 자주 보았어요. 이런 경우는 대부분 비자발적으로 위시리스트에 추가했기 때문이에요.**”
이벤트로 참여를 유도하거나, 보상을 걸고 만든 위시리스트 숫자는 실제 출시일에 큰 힘을 발휘할 수 없죠. Nikki는 위시리스트에 대해 ‘양 보다 질’이 중요하다고 말해요.
> “**위시리스트가 일정 수를 넘으면 플랫폼에서 노출이 늘어나긴 해요. 하지만 그 노출이 진짜 타겟 유저에게 꽂히려면 동시에 입소문이 퍼지고 있어야 하죠. 적어도 위시리스트에 70% 이상은 자발적이고, 지역 타겟과 맞아야 해요.**”
Discord나 Reddit, Youtube 댓글 등에서 유저들 사이에 자연스럽게 언급되고 있지 않다면, 위시리스트 차트에 올랐더라도 큰 의미가 없다는 뜻이에요. 스팀은 단순한 수치보다 ‘모멘텀’을 중요하게 생각하니까요.
[__wf_reserved_decorative]
## 게임 출시부터 진짜 마케팅 여정이 시작돼요
스팀 출시는 단거리 스프린트가 보단 마라톤에 가까워요. 출시일은 이 레이스의 결승선이 아니라 긴 경주가 시작되는 순간일 뿐이죠. 스팀 알고리즘은 여러분의 움직임을 지켜보고 있고, 유저들도 마찬가지예요. 출시 시점부터는 전략을 더욱 날카롭게 유지하고 즉각적으로 반응하며 마케팅을 고도화해야 해요.
개별 타이틀에 맞춘 커스터마이징 전략과 미디어 운영도 중요하지만, 어떤 게임이든 꼭 갖춰야 할 기본적인 대응 원칙이 있어요.
> “**낮은 유저 평가가 예상된다면, 미리 플랜 B를 준비해두는 게 좋아요. 예상보다 상황이 나쁠 때도, 그 안에서 최대한의 성과를 끌어내야 하니까요.**”
어떤 게임의 경우 출시 초반부터 크게 흔들릴 수 있어요. 하지만 팀 내에 결과에 빠르게 반응할 수 있는 시스템이 갖춰져 있다면 충분히 대응해 나갈 수 있죠. Nikki는 이를 위해 런칭 전부터 성과 트래킹을 미리 세팅해 두는 것을 강력하게 추천해요. 그래야 초반 트렌드를 감지하고 필요한 방향으로 전환을 이뤄낼 수 있으니까요.
> “**1일차와 3일차 데이터가 어떻게 다르게 움직이는지를 미리 파악할 수 있도록 성과 트래킹 솔루션을 미리 세팅해 두어야 해요. 크리에이티브, 타겟 오디언스, 국가별 캠페인 전략이 실제로 어떤 차이를 만들어내는지도 조기에 파악할 수 있어야 하죠.**”
무엇이 잘 작동하고 있고, 무엇이 안되고 있는지를 빨리 알아낼수록 전략을 바로잡는 속도도 훨씬 빨라져요. 만약 게임이 로그인 기능이나 온라인 플레이를 지원한다면 그 데이터를 꼭 활용하세요. 퍼스트파티 데이터는 LTV 예측 모델링이나 리타겟팅, 지역별 유저 분석에서 엄청난 이점을 줄 수 있어요.
## 브랜딩과 퍼포먼스 마케팅은 함께가야 해요
얼마 전까지만 해도 PC와 콘솔 게임 마케팅은 철저히 브랜딩 중심이었어요. 시네마틱 트레일러, 언론 PR, 인플루언서 바이럴 등이 핵심이었죠. 이에 비하면 퍼포먼스 마케팅은 비교적 덜 중요하게 여겨지고는 했는데요. 퍼포먼스 마케팅은 클릭, 설치, ROAS 같은 지표 중심의 마케팅이기 때문에 모바일 게임 마케팅에서만 중요하게 생각되곤 했죠. 게임의 스토리텔링, 세계관 구축과는 거리가 있어 보였기 때문이에요.
하지만 이제 흐름이 바뀌고 있어요. 라이브 서비스 모델이나 얼리 액세스 등을 채택하는 게임 스튜디오가 늘어나면서, 데이터 기반 전략이 마케팅 믹스의 핵심 요소로 자리잡고 있어요.
Nikki는 브랜드 마케팅과 퍼포먼스 마케팅은 서로 대립하는 개념이 아니라고 말해요. 오히려 이 둘이 함께 진행될 때 더 큰 시너지를 불러올 수 있다고 말이죠. 실제로 마케팅을 잘하는 게임 스튜디오는 두 가지 방식을 적극적으로 융합하고 있어요.
Nikki는 게임 마케터들의 특징을 다음과 같이 정의해요.
- **모바일 게임 마케터**는 데이터 중심 사고와 ROI 최적화에 능해요.
- **PC/콘솔 게임 마케터**는 브랜딩, 포지셔닝, 유저 심리에 대한 이해도가 높아요.
> “**요즘 게임 마케터들은 더 다양한 카테고리를 통해 인사이트를 얻고 있어요. 도달률, 프리퀀시, 주의 집중, Share of Voice(SOV) 지표처럼 본질적인 원칙을 바탕으로 다양한 미디어를 균형 있게 운영하죠.**”
이런 밸런스는 중견 팀일수록 더 중요해요. 모바일 마케팅팀은 날카로운 KPI와 정교한 트래킹 역량을 PC 마케팅팀은 직관적인 크리에이티브 감각과 브랜드 톤의 일관성을 유지해야 하죠. 이 두 축이 잘 맞물릴 때에 캠페인에서 더 빠르게 명확한 결과를 얻을 수 있어요.
핵심은, 퍼널 전체에 걸쳐 양쪽의 목표를 하나의 방향으로 설정하는 것이에요.
[__wf_reserved_decorative]
## PC·콘솔 유저에게 중요한 건 ‘하입’이 아니라 ‘증명’이에요
PC나 콘솔 유저는 지하철 광고를 보고 바로 게임을 구매하지 않아요. 모바일 게임처럼 충동적으로 설치하거나 한 번의 노출만으로 바로 전환될 수 없는 구조이기 때문이죠. 이들은 스스로 게임을 탐색하고, 커뮤니티 반응을 살피며 다른 게임과 비교하는 등 자기만의 속도로 게임을 구매하고 플레이해요.
> “**인내심이 가장 큰 미덕이에요. 즉각적인 결과를 기대하기 보다는 더 많이 더 진정성있게 게임을 노출하는 방법을 고민해야 하죠.**”
앞서 이야기한 것처럼 PC·콘솔 유저의 구매를 전환시키는 건 하나의 잘 만들어진 광고 크리에이티브가 아니에요. 어떤 환경에서 광고가 노출되었는지, 어떤 리뷰가 달렸는지, 트레일러는 어떤 느낌이었고, 커뮤니티에서는 이 게임에 대한 어떤 평가가 나오는지 등이 복합적으로 작용해 최종 구매로 이어지는 것이죠.
Nikki는 성공적인 홍보를 위해서 단순히 노출 수보다는 ‘발견의 순간’을 어떻게 설계할지를 고민해야 한다고 조언해요. 단지 광고 수익만을 목적으로 만들어진 채널에 광고를 노출하면 브랜드 이미지에는 악영향을 줄 수 있어요. 광고를 집행한다 하더라도 어떤 채널과 크리에이티브이냐가 중요한 것이죠.
> **“유저가 새로운 게임을 찾고 실을 때 어떤 채널을 가장 먼저 탐색할까요? 모바일과 달리, PC·콘솔 환경에서는 문맥이 훨씬 중요해요. 집중도와 임팩트를 먼저 최적화하세요. 전환은 그 다음 문제. 설령 해당 채널에서 어트리뷰션이 조금 어렵다고 하더라도 말이에요._”_**
## 이 채널들은 꼭 활용해야 해요
Meta, Google, YouTube 캠페인은 이제 게임 마케팅의 기본 요소가 되었어요. 이제는 경쟁이 심해진 대표적인 채널들 이외에도 아직 제대로 활용되지 않는 채널에서 새로운 기회를 찾아야 하죠. Nikki는 아시아 지역 스튜디오들이 놓치고 있는 채널을 크게 세 가지로 꼽아요.
### 1\. 커뮤니티 중심 미디어 (예: Discord, Reddit)
Discord와 Reddit이 게이머들의 주 생활 공간이라는 건 모두가 알고 있어요. 하지만 그걸 ‘안다’는 것과, 실제로 의미 있는 방식으로 존재감을 드러내는 것은 완전히 다른 이야기예요.
이 플랫폼은 진짜 대화가 오가는 공간이에요. 유저들은 여기서 기대감을 나누고, 불만을 털어놓고, 팁을 주고받고, 밈을 퍼뜨려요. 결국 그 모든 게 게임에 대한 인식을 만드는 데 큰 영향을 미치죠.
특히 스토리, 품질, 커뮤니티를 중시하는 PC·콘솔 유저층에게는 이런 커뮤니티를 통한 대화가 곧 마케팅이라 할 수 있어요.
### 2\. 하이터치(high-touch) 프로그래매틱 캠페인
많은 팀이 클릭이라는 전방 퍼널에만 집중해요. 하지만 PC나 콘솔 게임 유저는 단순히 클릭하고 바로 게임을 구매하지 않아요. 리뷰를 읽고, 다른 게임과 비교하고, 진지하게 고민하죠.
그래서 Nikki는 전방부터 후방 퍼널까지 연결된 프로그래매틱 전략의 중요성을 강조해요. 노출 순서, 빈도 설정, 오디언스 타겟팅 등이 조화를 이루면, 피로도 없는 유저 여정을 이끌 수 있어요.
### 3\. OTT 및 CTV 광고
OTT나 CTV는 여전히 과소평가되고 있어요. 비서구권에서는 성과 트래킹의 한계나 플랫폼 점유율 등의 문제로 잘 활용되지 않기도 하죠. 하지만 미국과 유럽 시장에서는 CTV가 결코 틈새 시장이 아니에요.
삼성, LG, Vizio 같은 OEM들은 유저가 어떤 콘텐츠를 보는지에 대한 데이터를 제공하고 있어요. 때로는 광고 시청 후 실제로 게임을 플레이했는지도 파악할 수 있죠. 게다가 최근 Netflix, Hulu, Disney+ 같은 스트리밍 플랫폼들이 광고를 점점 개방하고 있어요.
## 크리에이티브는 ‘많이 만드는 것’보다 ‘정확히 전달하는 것’이 더 중요해요
PC나 콘솔 캠페인에서 중요한 건 얼마나 많은 소재를 테스트하느냐가 아니라, 정확하게 후킹하는 메시지를 어떻게 전달하느냐예요.
모바일처럼 수십 개의 크리에이티브를 돌려보고 나중에 잘 되는 걸 찾는 방식이 아니라, PC 캠페인은 처음부터 타겟에게 제대로 먹히는 메시지를 설계하는 게 핵심이에요.
### 1\. 타겟 오디언스에 맞춘 정교한 메시지 세팅
게임 핵심 유저층에게 직접적으로 말 거는 듯한 크리에이티브를 만들어야 해요. 게임의 가장 강력한 특징을 중심으로 메시지를 짜고, 오디언스 세그먼트 별로 실제 반응이 어떻게 다른지를 테스트해 보는 게 중요해요.
### 2\. 장르와 UPS는 애매하지 않고, 명확하게
모호한 설명은 피해야 해요. ‘이 게임이 어떤 장르고, 누굴 위한 것이며, 무엇이 다른지’ 등 게임의 가치를 정확하게 보여줘야 해요.
> “**가끔은 소재만 보고는 게임이 뭔지 감이 안 잡힐 때가 있어요. 그냥 ‘슈터 게임’이라고 하지 말고, ‘80년대 공포 테마의 로그라이크 협동 FPS’처럼 구체적으로 설명해야 해요. 비주얼도 명확하고, 메시지도 확실하게 보여줘야 해요.**”
### 3\. 브랜드 인상과 포지셔닝을 담은 브랜드 키 비주얼
소재 하나하나가 게임의 퀄리티, 완성도, 시장 내 포지션을 말해줘야 해요. 특히 AA급 인식에서 AAA급 퀄리티로 도약하려는 게임일수록, ‘우리가 AAA에 가까운 게임이다’라는 분위기를 비주얼만으로 전달할 수 있어야 해요.
> “**진짜 AAA처럼 보이려면, 단순히 스펙이 아니라 ‘분위기’를 전달해야 해요.**”
[__wf_reserved_decorative]
## 크로스 플랫폼 전략, 하나의 메시지로는 부족해요
많은 게임 스튜디오들이 멀티 플랫폼 출시를 시도하고 있어요. 스팀, 콘솔, 모바일은 물론 HTML5나 웹까지 확장의 범위가 넓어지고 있죠. 멀티 플랫폼으로 게임을 출시하면 도달 범위는 넓어지고, 수익화 경로도 다양해지며, 유저가 있는 플랫폼으로 직접 찾아갈 수 있는 유연성도 생긴다는 장점이 있어요.
하지만 이런 전략이 진짜 효과를 내려면, 플랫폼에 대한 유저의 인식이 시장마다 어떻게 다른지를 정확히 이해하고 있어야 해요. 특히 미국이나 유럽처럼 시장별 성향이 확연한 지역에서는요.
예를 들어, APAC(아시아 태평양)에선 모바일이 메인스트림이에요. 반면, 서구권에서는 모바일 게이머가 여전히 ‘코어’ 유저라고 보긴 어려운 포지션이에요. 콘솔과 PC 보급률은 높지만, 서구권 유저들이 말하는 ‘코어 게이머 경험’은 여전히 컨트롤러를 쥐고 하는 슈팅 게임이에요. 터치스크린 기반의 모바일 게임은 그 이미지와 거리가 있죠.
> “**코어 콘솔 유저에게 App Store 출시 소식을 광고하면, 유저는 ‘아, 이건 내가 친구들이랑 플레이스테이션으로 하던 게임이 아니라 그냥 모바일 게임이구나’라고 느낄 수 있어요.**”
플랫폼은 같아도, 메시지 위치나 맥락이 다르면 게임 자체가 ‘내가 할 게임이 아닌 것처럼’ 느껴질 수 있다는 거예요. 그만큼 잘못된 문맥에서 잘못된 메시지를 던지면, 게임의 가치를 오히려 깎을 수도 있어요.
그렇다면 크로스 플랫폼 마케팅을 어떻게 접근해야 할까요? 정답은 하나가 아니에요. 대신 선택지는 있어요.
### 크로스 플랫폼 전략을 위한 세 가지 접근법
### **1\. 오디언스를 나눠서 보세요**
콘솔, PC, 모바일 유저는 기대하는 것도, 사용하는 언어도 달라요. 각 그룹을 개별 세그먼트로 인식하고 맞춤화된 메시지를 설계해야 해요.
### **2\. 메시지를 조정하세요**
플랫폼의 강점을 전면에 내세우는 전략이 효과적이에요. 콘솔 유저에겐 고사양 그래픽과 컨트롤 감도를 소구하거나 모바일 유저에겐 접근성과 편의성을 강조하는 것처럼 각 플랫폼에 맞는 언어와 가치를 중심에 두는 것, 그것이 핵심이에요.
### **3\. 중립적인 톤을 유지하는 것도 방법이에요**
메시지가 너무 분산될까 걱정된다면, 플랫폼을 굳이 강조하지 않는 방식도 있어요. 유저가 자연스럽게 자신에게 맞는 플랫폼을 선택하도록 여지를 두는 전략이죠.
## 풀퍼널 트래킹이 중요해요
대부분의 게임 마케터가 라스트 터치 어트리뷰션으로 성과 트래킹을 시작해요. 하지만 캠페인 규모가 커질수록, 성과를 트래킹하는 방식도 함께 확장돼야 해요.
대부분의 유저는 광고를 한 번 보고 바로 전환하지 않아요. 특히 Instagram 같은 채널에서는 더더욱 그래요. 반복 노출, 명확한 메시지, 그리고 적절한 타이밍이 모여야 비로소 전환이 일어나요.
> “**풀퍼널 트래킹은 모든 게임에 필수예요. 인스타그램에서 광고를 처음 본 유저가 곧바로 전환하는 경우는 거의 없어요. 시간을 두고 유저에게 반복적으로 메시지를 보여주면서, 제품에 익숙해지도록 유도하고, 그 흐름 속에서 설치나 구매로 이어지게 만들어야 해요.**”
만약 라스트 터치 클릭만 추적하고 있다면, 실제로 유저가 어떤 경로를 통해 구매 결정을 내렸는지 그 전체 그림을 놓치고 있는 거예요.
예를 들어 TV 광고, OOH(옥외 광고), 콘솔 내 미디어는 디지털 캠페인과는 생태계가 다르지만, 사용자의 인지와 판단에는 비슷하게 영향을 주죠.
크로스 플랫폼 전략을 쓰지 않더라도 본질은 같아요. 유저 여정 단계에 맞는 메시지를 설계하고, 각각이 어떻게 맞물려 효과를 내는지를 보여주는 측정 도구를 함께 써야 해요. 그래야 단발성 캠페인이 아니라, 전환을 만들고 유지하는 마케팅 시스템을 구축할 수 있어요.
[__wf_reserved_decorative]
## 미국 시장 진출, 이것만은 하면 안 돼요
미국 시장에 진출하는 건 단순히 광고를 번역하거나, 다른 스튜디오의 성공 사례를 그대로 따라간다고 해결되는 일이 아니에요. 현지 유저 행동, 예산 기대치, 문화적 접점에 대한 이해가 없으면 예산은 낭비되고 타이밍도 놓치기 쉬워요.
APAC 스튜디오가 서구권에 진출할 때 자주 범하는 실수들이 있어요.
### 1\. 복사 붙여넣기식 전략
경쟁 스튜디오의 마케팅 플랜을 그대로 따라 하고 싶어질 수 있어요. 특히 APAC 스튜디오는 상대적으로 리스크 회피 성향이 강한 편이라 이런 유혹이 더 크죠. 하지만 그 전략이 당신의 게임에도 통할 거란 보장은 없어요. 게임마다 예산도, 오디언스도, 과제도 다르니까요.
> “**당신의 게임은 그 게임이 아니에요. 당신의 과제는 그 게임의 과제와 다르고, 예산 구조도 완전히 다를 수 있어요.**”
### 2\. 너무 낮은 예산
많은 APAC 퍼블리셔가 ‘절반 예산으로도 경쟁할 수 있다’고 생각해요. 하지만 미국과 같은 프리미엄 시장에서는, 적절한 투자 없이는 존재감조차 드러내기 어렵고, 최소한 정교한 트래킹 환경이라도 갖추지 않으면 성과를 만들기 힘들어요.
### 3\. 문화적 뉘앙스 간과
미국은 트렌드 중심, 세분화된 취향, 수많은 서브컬처가 공존하는 시장이에요. 이런 맥락을 읽지 못하면, 아무리 크리에이티브나 미디어 믹스를 잘 설계해도 진짜 유저의 반응은 얻기 힘들어요. 현지에 기반한 팀 또는 컨설팅 파트너가 꼭 필요한 이유예요.
## 소규모 게임 스튜디오도 자신만의 방식으로 충분히 성공할 수 있어요
그동안 대규모 예산 캠페인 위주로 설명했지만, 그게 유일한 성공 경로는 아니에요. 중견 게임 스튜디오는 어떤 방식으로 성과를 만들 수 있을까요?
Nikki가 예시로 든 게임은 _Squirrel with a Gun_. 이름 그대로 ‘총을 든 다람쥐 시뮬레이터’예요.
> “**이런 게임은 AAA 타이틀처럼 풀스크린 테이크오버나 정제된 광고 기사로 홍보하면 안 돼요. 그 방식으론 이 게임의 매력이 전달되지 않아요.**”
대신 Nikki의 팀은 게임의 정체성 자체를 브랜딩하는 전략을 채택했어요.
> “**이 게임은 근본적으로 웃기고 황당하고 유쾌해요. 그 느낌을 어떻게 극대화할 수 있을까 고민했죠.**”
그래서 Discord 서버에서 밈을 만들고, Instagram 밈 계정과 협업해 커뮤니티 기반 콘텐츠를 퍼뜨렸어요. 이런 자발적인 분위기에 스마트한 리마케팅 전략과 전방 퍼널 지원을 더한 결과, 게임 개성과 잘 맞는 마케팅 캠페인이 탄생했고, 실제로도 좋은 성과로 이어졌어요.
## 💡 중견 게임 스튜디오의 스팀 성공 전략 핵심 요약
- **진짜 수요가 반영된 위시리스트만 의미 있어요**: 강제로 모은 숫자보다, 자발적인 관심과 지역 타겟팅이 더 중요해요.
- **스팀은 단순한 스토어가 아니라 ‘발견 엔진’이에요**: PR, 콘텐츠, 커뮤니티, 미디어를 통해 유저의 관심 곡선을 설계하세요.
- **출시는 끝이 아니라 시작이에요**: 빠른 데이터 체크, Plan B 준비, 2일차 이후의 민첩한 대응까지 꼭 포함하세요.
- **브랜딩과 퍼포먼스를 통합하세요**: 강한 크리에이티브와 정밀한 타겟팅, 풀퍼널 전략이 결합돼야 캠페인이 오래가요.
- **플랫폼이 아닌 유저에게 맞는 메시지를 쓰세요**: 모든 유저에게 똑같은 메시지는 통하지 않아요. 세그먼트를 나누고, 그 언어로 말하세요.
- **서구권 진출 시 시장을 정확히 이해하세요**: 복사 전략을 피하고, 예산을 제대로 확보하고, 문화를 읽을 줄 아는 현지 파트너와 일하세요.
## Liquid Advertising과 함께 게임 마케팅 전략을 설계하고 싶으신가요?
크리에이티브 전략부터 크로스 플랫폼 미디어 플래닝까지. Liquid Advertising과 함께라면 2025년 현재, 게임 마케팅의 모든 도전과제를 함께 해결할 수 있어요.
- [liquidadvertising.com](https://www.liquidadvertising.com)에서 스팀·콘솔·멀티 플랫폼 캠페인 사례를 확인해보세요.
- LinkedIn·Instagram에서 캠페인 해설, 업계 인사이트, 크리에이티브 아이디어를 받아보세요.
- Liquid 팀과 직접 연락해, 당신의 게임에 꼭 맞는 전략을 설계해보세요.
## PC, 웹, 콘솔, 모바일 크로스 플랫폼 마케팅 성과 측정을 통해 고도화해 나가세요
좋은 마케팅 전략을 세웠다고 해도 성과 측정이 어렵다면 전략을 고도화해 나가는 데에 어려움이 있겠죠. 에어브릿지는 게임 스튜디오가 크로스 플랫폼 환경에서 유저 여정을 정밀하게 트래킹하고 실시간 인사이트를 통해 마케팅 퍼포먼스 최적화를 돕고 있어요.
- **유저 LTV 분석과 코호트 분석**: 키워드, 크리에이티브, 국가, OS 등 다양한 조건별로 유저 LTV를 추적할 수 있어요.
- **광고 매체 연동 통합**: Google AdMob, AppLovin MAX, Unity LevelPlay 등 여러 매체 데이터를 하나로 모아서 분석할 수 있어요.
- **실시간 수익 예측 pLTV 기능**: 미래 수익을 미리 예측하고, 더 빠르고 정확한 의사결정을 할 수 있어요.
- **분 단위 리텐션 분석**: 유저의 플레이 흐름을 시, 분 단위로 이해해 더욱 실질적인 인사이트를 얻을 수 있어요.
- **합리적이고 투명한 요금제**: 숨겨진 비용 없이, 엔터프라이즈급 기능을 모두 사용할 수 있어요.
- **오디언스와 원본 데이터 간편하게 내보내기:** 타깃 세그먼트를 빠르게 추출하고, 원본 데이터를 CSV 또는 Google Sheets로 내보내서 커스텀 대시보드 구성이나 심층 분석에 활용할 수 있어요.
---
## 메타 AMM 지원, 에어브릿지에서 더욱 정교한 메타 어트리뷰션 데이터를 확인하세요
URL: https://www.airbridge.io/ko/blog/meta-amm-support
Language: Korean
Published: 2025-06-18
메타가 Advanced Mobile Measurement(AMM)을 다시 지원해요. 캠페인 성과를 최적화하기 위해 정교한 데이터를 활용하는 앱 광고주들에게는 반가운 소식이에요. 2025년 6월 19일부터 메타의 AMM 약관에 동의한 광고주는 에어브릿지의 표준 리포팅 인프라를 통해 비집계형 메타 라스트 터치 어트리뷰션 데이터를 확인할 수 있어요.
## AMM이란 무엇인가요?
Advanced Mobile Measurement(AMM)은 에어브릿지와 같은 모바일 측정 파트너(MMP)를 통해 제공되는 리포트로, 메타 캠페인과 연동된 라스트 터치 어트리뷰션 데이터를 행 단위(row-level)로 광고주가 확인할 수 있어요. 행 단위 데이터란, 광고 클릭부터 전환까지의 과정을 유저별로 한 건씩 볼 수 있는 데이터예요. 기존에는 전환 수 등 집계된 형태의 정보를 제공했다면, AMM 데이터는 각 전환 건을 행 단위로 분석할 수 있어 BI 도구나 내부 모델링에 유용해요.
메타는 2021년에 AMM을 일시 중단했고, 그 이후로 광고주들은 집계된 형태의 메타 캠페인 성과 데이터만 받을 수 있었죠. 이번 AMM 도입은 새로운 연동이 아닌, 기존 기능을 재도입하는 것에 의미가 있어요.
## 무엇이 바뀌었나요?
에어브릿지는 메타의 공식 모바일 측정 파트너(MMP)로서 이미 메타 캠페인의 라스트 터치 어트리뷰션 데이터를 수신하고 있어요. 하지만 2021년 메타가 AMM을 종료한 이후로 광고주에게 제공되는 데이터는 집계된 형태로 한정되어 있었어요. 하지만 이제 메타의 AMM 약관에 동의한 광고주는 비집계된 전환 단위의 데이터를 열람할 수 있어요. 여기에는 다음과 같은 정보가 포함돼요:
- 메타 광고 전환에 대한 어트리뷰션 정보
- 기존 MMP 리포트에 표시되는 행 단위 데이터
- BI나 분석 도구에서 더 정밀한 측정을 가능하게 해주는 데이터
한편 어트리뷰션 데이터를 수신하고 처리하는 에어브릿지와 메타의 연동 방식은 변하지 않았어요. 이번 업데이트는 광고주가 에어브릿지를 통해 받는 데이터의 세부적인 수준에만 영향을 주며, 광고 성과 리포팅 자체에는 영향을 미치지 않아요.
## AMM 데이터 적용 방법
Meta 캠페인에서 행 단위 어트리뷰션 데이터를 받으려면 다음 단계를 따라주세요:
1. [Meta의 AMM 약관에 동의하기](https://www.facebook.com/ads/manage/advanced_mobile_measurement/app_based_tos)
2. 메타 개발자 앱에서 ‘관리자’ 권한을 보유하고 있어야 약관 수락이 가능해요
1. [앱 역할 관련 안내](https://developers.facebook.com/docs/development/build-and-test/app-roles/)
2. [역할 관리 방법 안내](https://www.facebook.com/business/help/)
AMM 약관을 수락하면, 비집계된 메타 어트리뷰션 데이터를 에어브릿지 대시보드에서 바로 확인할 수 있어요.
## 핵심 요약
- 메타가 AMM을 다시 지원해요: 메타는 AMM 약관에 동의한 광고주에게 비집계형 라스트 터치 리포팅을 다시 제공하기 시작했어요.
- 에어브릿지는 Meta의 MMP 파트너로서 24시간마다 갱신되는 데이터를 자동으로 전달해드릴 거예요.
- 적용 방법: AMM 약관에 동의하세요. (앱 관리자 권한 필요)
- 장점: 에어브릿지와 분석 도구 전반에서 메타 광고 성과에 대한 가시성이 훨씬 높아져요.
이번 메타의 AMM 재도입은 어트리뷰션을 측정하거나 리포트하는 현재의 성과 측정 방식에는 아무런 변화를 주지 않는 반면, 접근할 수 있는 데이터의 세부 수준을 더욱 정교하게 제공해요. 만약 모델링, 성과 측정, LTV 분석을 더 정교하게 하고 싶었다면, 이번 AMM 재도입은 큰 전환점이 될 수 있어요. 정교한 광고 성과 분석을 시작하는 데 도움이 필요하다면 언제든지 [에어브릿지에 문의](https://www.airbridge.io/ko/book-a-demo)해주세요.
---
## 플랫폼의 경계가 무너지는 크로스플랫폼 시대, 게임 UA 전략은 어떻게 바뀌어야 할까?
URL: https://www.airbridge.io/ko/blog/cross-platform-game-ua-strategy
Language: Korean
Published: 2025-06-09
과거에는 각 플랫폼에 특화된 UA(User Acquisition) 전략을 따로 설계하는 것이 일반적이었어요. 모바일은 퍼포먼스 마케팅 중심, 콘솔은 커뮤니티 기반 브랜딩 중심이라는 공식이 있었죠.
그러나 이제는 모바일, PC, 콘솔의 구분이 점점 모호해지고 있습니다. 모바일에서도 콘솔 유저를 유입시키고, 콘솔 게임에서도 모바일식 트래킹 솔루션을 도입해 성과를 측정하는 시대예요. 플랫폼의 경계가 허물어지고 있는 것이죠.
**이러한 변화는 단순한 트렌드가 아니라, UA 전략 전체의 구조적 재편을 요구하는 신호예요.** 그렇다면 이 변화 속에서 실무자들이 주목해야 할 포인트는 무엇일까요? 크로스 플랫폼 게임을 위한 UA 전략은 어떤 지표를 기준으로 설계되어야 하며, 광고 채널과 크리에이티브 운영 방식은 어떻게 달라져야 할까요?
그래서 에어브릿지가 국내 최초로 ‘크로스 플랫폼’을 주제로 하는 모바일 게임 마케팅 컨퍼런스를 준비했어요. 최신 UA 전략과 기존 UA 문법을 확장하거나 뒤흔드는 다양한 사례들과 실질적인 인사이트가 공유됐는데요. 지금 이 시기에 필요한 게임 마케팅의 방향성을 함께 짚어봤어요.
[]
**오늘의 Key Takeaways**
- 다양한 플랫폼 간의 UA 캠페인 성과를 측정하기 어려운 이유를 알 수 있어요.
- 틱톡이 어떻게 주요 글로벌 크로스 게임 마케팅 채널로 자리 잡았는지 알 수 있어요.
- 넥슨의 ‘퍼스트 디센던트’가 스팀에서 인기 1위를 기록한 후기를 알 수 있어요.
- 지금의 게임 마케팅과 운영에 반드시 필요한 게임 백엔드 플랫폼 ‘하이브’에 관해 알 수 있어요.
- 서구 시장의 PC 콘솔 마케팅 전략과 마케터들이 흔히 빠지는 실수와 해결책을 알 수 있어요.
## 1\. 에어브릿지(Airbridge), 크로스 플랫폼 트래킹의 현실적인 해법을 제시하다
[정주홍 님](https://www.linkedin.com/in/juhong-jung-9378b570/)은 원래 게임 개발자 출신으로, 현재는 **마케팅 측정 솔루션(MMP) 에어브릿지의 엔지니어링팀 리드**로 활동 중인데요. 이번 세션에서 모바일과 PC, 콘솔 간 크로스 플랫폼 환경에서의 게임 성과 측정이 왜 어려운지, 그리고 이를 어떻게 기술적으로 해결하고 있는지를 실제 사례 중심으로 소개했어요.
[]
### 왜 크로스 플랫폼 성과 측정은 어려울까요?
PC/콘솔 기반 게임에서는 UA 캠페인 성과 측정이 어렵다는 목소리가 많아요. **특히 크로스 플랫폼 게임의 경우 유저 여정이 웹, 모바일, PC를 넘나들며, 각기 다른 식별자를 사용하기 때문에 전환 흐름을 통합적으로 분석하기 어려워요.**
[]
모바일은 IDFA/GAID를, 웹은 쿠키를, PC는 이메일이나 유저 ID 등을 사용하다 보니 동일 유저임을 식별하기 힘들다는 점이 문제인데요. 다양한 기기와 플랫폼을 넘나드는 유저 여정을 하나의 흐름으로 묶기 위해서는, 서로 다른 식별자를 하나의 고유한 ID로 연결하는 매칭 기술이 필수적이에요. 에어브릿지는 ‘에어브릿지 ID’를 중심으로 **광고 클릭부터 결제까지의 유저 행동 데이터를 통합해, 하나의 퍼널 리포트로 분석할 수 있는 구조를 제공**하고 있어요. 광고 클릭부터 웹사이트 방문, PC 게임 설치, 모바일 앱 결제까지 다양한 유저 행동 데이터를 하나의 ‘에어브릿지 ID’로 통합하여 매칭하고, 이를 기준으로 퍼널 리포트를 생성할 수 있는 기능을 제공하고 있다고 밝혔어요.
[]
또한 플랫폼에 구애받지 않고 전환 데이터를 유기적으로 연결하고, 이를 포스트백으로 미디어에 전달함으로써 UA 캠페인의 최적화까지 가능하게 만든다고 말했는데요. [넥슨의 ‘퍼스트 디센던트’ 사례](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)를 통해, 콘솔 게임의 성과를 틱톡 모바일 광고 클릭으로 측정하고 컨버전 API를 통해 최적화한 경험을 공유했어요. 모바일 중심의 틱톡 광고를 통해 콘솔 유저까지 성과를 추적할 수 있었는데요. 크로스 플랫폼 측정의 가능성을 보여주는 대표적인 사례였어요. 이외에도 ‘메이플스토리 월드’처럼 PC와 모바일을 모두 지원하는 게임에서도, 각각의 전환 데이터를 하나의 통합 리포트로 확인하고 분석할 수 있었어요.
[]
### **UA 분석 사각지대를 파악하기 위한 앞으로의 계획**
에어브릿지는 앞으로 매칭률과 정확도를 높이는 데 집중하고 있으며, 웹 광고 파트너와의 연동 강화, 고가치 유저 세그먼트 기반의 리타겟팅 기능 확대도 준비 중이라고 밝혔어요. 특히 많은 요청이 있었던 PC 및 콘솔용 SDK 개발도 장기 계획에 포함되어 있으며, 향후에는 더 많은 API 연동과 시그널 전달 방식의 확장이 기대된다고 덧붙였어요.
## **2\. 크로스 플랫폼 게임 마케팅, 왜 지금 틱톡이어야 할까?**
**틱톡 게임 팀**의 이지영 님은 이번 세션에서 ‘크로스 플레이 게임의 글로벌 확장’이라는 주제로, 틱톡이 어떻게 주요 게임 마케팅 채널로 자리 잡았는지 실제 캠페인 사례와 데이터를 통해 설명했어요. 현재 틱톡은 150개국, 75개 언어, 월간 사용자 수 10억 명에 달하는 글로벌 사용자 기반을 갖추고 있으며, 미국, 일본, 동남아 등 주요 시장에서 게임 광고 노출 점유율 1위를 기록하고 있다고 해요.
[]
[]
특히 코어 게이머 사이에서도 타 플랫폼 대비 2배 높은 광고 노출 빈도를 보여주며, 틱톡이 단순한 소셜 채널을 넘어 게임 광고 플랫폼으로 진화하고 있다는 점을 강조했어요.
그 배경에는 틱톡 유저들의 콘텐츠 소비 방식이 소셜 게이밍 환경과 유사하다는 특성이 있어요. 사용자 생성 콘텐츠(UGC), 실시간 소통, 트렌드 참여 등 틱톡의 사용성은 게임 유저들이 콘텐츠를 자연스럽게 소비하고 확산시키는 데 최적화되어 있다고 말했는데요. **실제로 틱톡 유저는 타 플랫폼 대비 1.6배 더 공유를 많이 하고, 구매 전환 가능성도 높은 것으로 나타났어요.**
[]
[]
### **브랜디드 캠페인부터 고급 퍼포먼스 마케팅까지**
틱톡은 브랜드 캠페인뿐만 아니라 퍼포먼스 마케팅에서도 존재감을 드러내고 있어요. 라이엇게임즈의 '발로란트 콘솔 론칭 캠페인'에서는 탑뷰 광고와 브랜디드 미션, 사운드 필터 등을 조합해 유저 참여를 유도했고, 브라질 시장에서는 **브랜디드 이펙트 캠페인을 통해 5억 4천만 뷰를 달성한 사례**도 공유되었어요.
[]
[]
무엇보다 주목할 점은 콘솔 게임 광고의 전환 측정이 가능해졌다는 점이에요. **에어브릿지와의 협업으로 틱톡 광고 클릭 ID와 게임 내 유저 행동 데이터를 매칭하고, 이를 기반으로 오디언스 세분화와 캠페인 최적화를 구현해냈는데요.** 그 결과, 넥슨의 ‘퍼스트 디센던트’ 캠페인에서는 모바일 수준의 전환율(81%)이라는 성과를 냈다고 해요.
틱톡은 크리에이티브 지원도 강화 중이에요. 인플루언서 기반 영상 제작 플랫폼 ‘틱톡 원(One)’과 AI 기반 영상 생성 도구 [‘틱톡 심포니(Symphony)’](https://ads.tiktok.com/creative/creativestudio/home/ko)를 통해, 브랜드가 빠르게 글로벌 캠페인을 실행할 수 있도록 돕고 있어요. 이지영 님은 틱톡이 단순 유입 경로를 넘어, 콘텐츠 확산과 전환 측정, 크리에이티브 제작까지 아우르는 통합 게임 마케팅 플랫폼으로 성장하고 있으며, 앞으로 크로스 플랫폼 UA 전략의 핵심 채널이 될 것이라 강조했어요.
## **3\. ‘퍼스트 디센던트’ UA 캠페인, 크로스 플랫폼 마케팅의 새로운 가능성을 열다**
세 번째 순서인 패널 토크에서는 넥슨 김선희 UA 팀장, 틱톡의 최영준 클라이언트 파트너, 에어브릿지 남성필 대표가 **‘퍼스트 디센던트’ 글로벌 캠페인을 중심으로 크로스 플랫폼 마케팅 전략의 실전 사례**를 공유했어요. 퍼스트 디센던트는 넥슨게임즈가 개발하고 넥슨이 퍼블리싱한 3인칭 루트 슈터 장르의 게임으로, 스팀에서 인기 1위를 기록하며 글로벌 유저들의 높은 관심을 받았는데요. PC와 콘솔 게임에서도 모바일 UA 방식의 마케팅이 가능하다는 점을 입증한 캠페인 사례로 주목받았어요.
👉[퍼스트 디센던트, 이제 PC와 콘솔 게임도 모바일 게임처럼 광고 성과를 측정하고 최적화할 수 있어요](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)
[]
[]
### **모바일 방정식을 PC 콘솔에 적용하다**
김선희 팀장은 이번 퍼스트 디센던트 캠페인을 통해 처음으로 콘솔과 PC 마케팅을 통합 분석하려는 시도를 했다고 밝혔어요. 기존에는 GA4나 내부 BI를 통해 웹페이지 방문이나 버튼 클릭 같은 어퍼 퍼널(Upper Funnel) 중심의 데이터만 확인할 수 있었지만, 인게임 전환을 포함한 LTV 기반 최적화를 위해서는 모바일처럼 실시간 트래킹과 매체 간 데이터 비교가 가능한 솔루션이 필요했어요. 이에 따라 **넥슨은 에어브릿지를 도입해 틱톡, 메타, 아마존 등 다양한 매체와 연동하며 PC 콘솔 캠페인에서도 모바일 UA 방식의 퍼포먼스 마케팅을 구현**할 수 있었어요. 특히 틱톡을 통해 전환 효율이 높은 결과를 확인하면서, 모바일과 PC 성과를 구분해 예산을 조정하는 전략을 수립하는 데 큰 도움이 되었다고 전했어요.
[]
### **크리에이티브 실험의 확장, UGC의 힘**
퍼포먼스 캠페인에서 ‘소재’가 얼마나 중요한지는 이미 많은 마케터가 체감하고 있는데요. 김선희 팀장은 **글로벌 유저에게 먹히는 소재는 국내용 시네마틱이나 트레일러 영상과는 다르기 때문에, 다양한 크리에이티브 실험이 필수라고 강조**했어요. 크리에이티브 측면에서는 틱톡 원 솔루션을 활용해 다양한 UGC 기반의 광고 소재를 빠르게 테스트하고, 성과가 높은 소재를 선별해 집중하는 방식으로 캠페인 효율을 극대화했는데요. 틱톡 클라이언트 파트너 최영준 님도 이번 캠페인을 진행하면서, 크로스 플랫폼 타이틀도 모바일 광고 채널에서 강한 성과를 낼 수 있음을 증명한 계기였다고 밝혔어요. 이 캠페인의 성과는 후속 타이틀 ‘카잔’의 북미 캠페인에도 적용되고 있으며, **클릭률이 평균 대비 25% 이상 향상**했다고 해요. 특히 2030 남성 게이머와의 데모 일치, 자연스러운 UGC 광고 콘텐츠, 그리고 API 연동을 통한 인게임 전환 최적화가 시너지로 작용했다고 덧붙였어요.
[]
### **실무자가 전하는 실전 팁**
마지막으로 김선희 팀장은 실무 팁으로 사전 예약 페이지에 SDK 및 픽셀을 사전 삽입해 오디언스를 구축하고, 주말·야간 타겟팅이나 와이파이 설정 등 세부 설정이 매칭률에 큰 영향을 미친다는 점을 공유하며 실질적인 인사이트를 전했어요.
## **4\. ‘하이브’는 왜 필요한가 : 크로스 플랫폼 시대의 게임 운영을 위한 필수 인프라**
[컴투스플랫폼, 하이브의 강범희 수석님](https://www.linkedin.com/in/beomhee-kang-1389a964/)은 **‘하이브(Hive)’라는 게임 백엔드 플랫폼이 왜 지금의 게임 마케팅과 운영에 반드시 필요한가**에 대해 심도 있는 발표를 진행했어요. 하이브는 로그인, 결제, 커뮤니티, 고객센터 등 게임 서비스 전반을 구성하는 핵심 기능을 통합 제공하는 게임 백엔드 플랫폼(GBaaS, Game Backend as a Service)입니다. 이를 축구에 비유해서 설명했는데요. UA(유저 확보) 스페셜리스트를 날카로운 크로스를 올려주는 이강인 같은 플레이메이커로 비유했어요. 타깃 유저를 정확히 겨냥하고, 유입의 흐름을 만들어주는 역할이죠. 반면, 사업 PM은 그 기회를 골로 연결하는 손흥민 같은 스트라이커로 비유했습니다. 시장의 니즈를 간파해 결정적인 한 방을 만들어내죠. 이 두 포지션이 조화를 이룰 때 진정한 ‘골’을 만들어낼 수 있다는 점에서, UA 스페셜리스트와 사업 PM이 공존하며 유기적으로 연결되어야 함을 강조했습니다.
[]
[]
### **크로스 플랫폼 시대, 개발사도 퍼블리셔도 바뀌고 있다**
과거 퍼블리셔 중심 구조에서 벗어나, 개발사가 직접 셀러로 나서는 ‘셀프 퍼블리싱’이 늘고 있는 가운데, 기술적 지원(로그인 시스템, 결제, 커뮤니티, 고객센터, 유저 데이터 분석 등) 의 공백을 메우는 인프라로서 하이브의 가치가 부각되고 있어요. 현재 100개 이상의 외부 게임이 하이브를 연동하여 서비스 중이며, PC, 모바일, 크로스 플랫폼 등 다양한 장르를 지원하며, 하이브만의 범용성을 보여주고 있다고 해요.
[]
### **“기술이 아닌 사업이 중심” : 운영 효율도 높이고, 비용도 줄인다.**
하이브의 모든 기능은 “사업적으로 득이 되느냐”를 기준으로 설계되었다고 강조했는데요. 간편 로그인(국가별 IDP 및 SSO), 글로벌 약관 대응, 앱/웹 결제 시스템(PortOne, MyCard, Xsolla 연동), PC 런처 제공 등 실제 운영 현장에서 바로 쓸 수 있는 무기들을 갖춘 것이 특징이에요. 또한 하이브 콘솔에서는 주요 운영 지표(매출, DAU, 결제 전환율, 리텐션 등)뿐 아니라, 에어브릿지를 통합해 광고 성과까지 실시간 분석할 수 있는데요. 예를들면, 결제 직전 이탈 유저에게 쿠폰을 보내는 등 정교한 마케팅 액션도 할 수 있다고 해요. 강범희 수석은 이를 ‘K-운영’’이라 표현하며, 한국식 고객 대응의 섬세함을 글로벌 시장까지 확장하는 전략도 소개했어요. 하이브 커뮤니티 기능은 로그인 정보를 연동해 유저 히스토리를 추적하고, 대응 우선순위를 설정할 수 있으며, 기본 내장된 고객센터 시스템을 통해 소규모 개발사도 글로벌 고객 케어가 가능하다고 해요. 그리고 마지막으로 “**게임사는 본질에 집중하세요. 나머지는 하이브가 맡겠습니다.**”라는 말로 발표를 마무리하며, 하이브가 단순 백엔드 툴이 아닌 사업 성장의 파트너임을 강조했어요.
## **5\. "효율만으로는 부족하다" 서구 시장의 PC 콘솔 마케팅 트렌드**
[Liquid Advertising의 니키(Nikki DePaola) 이사](https://www.linkedin.com/in/nikki-depaola/)는 이번 발표에서 **서구 시장의 PC 콘솔 마케팅 전략을 중심으로, 마케터들이 흔히 빠지는 실수와 그 해결책**에 대해 소개했는데요. 주요 메시지는 단순한 퍼포먼스 마케팅만으로는 부족하다는 것이었어요.
[]
[]
### **‘풀퍼널’ 시대에는 성공을 정의하는 방식부터 달라야 한다**
니키는 “PC 콘솔 게임 마케팅에서 퍼포먼스 마케팅만으로는 부족하다”고 강조했어요. 지금은 다양한 접점에서 게이머와 만나는 풀퍼널 전략이 필수라는 것인데요. 단일 지표나 라스트 터치 방식으로는 실제 효과를 정확히 측정하기 어렵고, 퍼포먼스 중심의 단기 캠페인만으로는 장기적인 효과를 기대하기 어렵다고 강조했어요. **특히 광고 효율만을 쫓다 보면 ROAS가 정체되고, 반복적인 캠페인에 의존하는 악순환에 빠질 수 있다고 지적했어요.모바일과 달리, PC/콘솔 게임은 스팀 구매, 유튜브 시청, 입소문 등 유입 경로가 복잡한데요. 성공을 평가할 때도 다양한 시각이 필요하며, 아래와 같은 여러 핵심 지표들**을 추가할 것을 제안했어요.
1. 리뷰 점수의 영향도
2. 커뮤니티 반응
3. 소셜 미디어 언급량
이 외에도 PR, 인플루언서 반응, 레딧 참여도 등 다양한 서브 지표도 함께 고려해야 한다고 강조했어요.
[]
### **퍼포먼스 + 브랜드 전략의 결합이 해답**
리키는 퍼포먼스 마케팅 자체를 부정하지 않으면서도, **브랜드 전략과 결합되지 않으면 지속성이 떨어진다는 점을 강조**했습니다. 단일 어트리뷰션 모델의 한계를 넘어 MTA, MMM, 브랜드 리프트 조사 등 다각적인 분석이 필요하다고 조언했어요. 특히 파이널 판타지 XIV의 리런치 사례를 소개하며, 단순한 검색 광고에서 브랜드 콜라보 및 풀퍼널 전략으로 전환한 결과, 이전보다 더 많은 신규 유저들을 확보하는 데 성공했다고 해요.
### **아시아 마케터에게 전하는 조언 3가지**
서구 시장을 겨냥한 캠페인을 준비하는 아시아 마케터들에게 니키는 몇 가지 조언을 전했어요. 먼저, 타깃 오디언스가 훨씬 파편화되어 있기 때문에, 단일한 마케팅 메시지나 접근법으로는 성과를 내기 어려워요. 그리고 PC 콘솔 게임 마케팅은 성과가 느리게 누적되기 때문에, 단기적인 반응에 일희일비하지 말고 장기적인 전략을 기반으로 한 인내가 필요하다고 강조했어요.
## **게임 유저 확보를 위한 넥스트 스탭, 여러분은 어떻게 준비하고 계시나요?**
**현장에서 느낀 이번 컨퍼런스의 키워드는 “디테일”과 “전환”이었어요.**
브랜딩과 퍼포먼스를 병행하고, 국가별로 다른 유저 여정에 맞춘 메시지를 만들며, 데이터를 수집하고 바로 액션으로 옮길 수 있는 구조까지 알 수 있었는데요. 성공적인 게임 마케팅은 결국 섬세한 실행력과 유저 중심의 시선에서 시작된다는 걸 다시 한번 느꼈어요. 앞으로의 게임 마케팅은 단순히 ‘광고 잘 돌리는 법’이 아닌, **전체 퍼널을 설계하고 의미 있는 유저 경험을 만드는 능력이 핵심**이 될 거예요. 오늘 소개해 드린 인사이트와 전략들이 좋은 영감을 줄 수 있길 바라며, 2025년에도 계속해서 변화하는 게임 시장에 맞춰 더 나은 전략을 고민해 보세요. **에어브릿지**의 새로운 소식과 크로스 플랫폼 측정에 대한 더 많은 인사이트를 얻고 싶으시다면 **아래 버튼을 눌러 전문가와 이야기**를 나누어 보세요.
---
## 2025 앱 마케팅, 같은 예산으로 성과 2배 내는 6가지 실전 전략
URL: https://www.airbridge.io/ko/blog/6-marketing-efficiency
Language: Korean
Published: 2025-06-09
2025년, 모든 모바일 앱 마케터들의 고민은 비슷해요. 광고비는 그대로인데 성과는 반드시 만들어야 하죠. CPI는 높아지고 광고 효율은 떨어지며, 플랫폼 자동화는 점점 정교해지고 있어요.
이제 중요한 건 예산을 더 쓰는 게 아니라, 현재 예산 안에서 성과를 극대화할 수 있는 구조를 어떻게 만들 것인가예요. 에어브릿지는 전 세계 다양한 앱 마케터들과 협업하며 이 질문에 대한 실전적 해답을 쌓아왔어요. 그중 가장 효과적이고 당장 실행 가능한 여섯 가지 전략을 소개할게요.
## 1\. 웹 유입, 그냥 흘려보내면 안 되는 이유
### 웹에 머물던 유저를 앱으로 데려오는 전략
검색, 콘텐츠, 블로그, 커뮤니티 등 다양한 채널을 통해 유입되는 웹 트래픽은 많지만, 앱 설치로 이어지지 않으면 성장은 정체될 수밖에 없어요. 유입만 잘 연결해도 광고비를 늘리지 않고 성과를 끌어올릴 수 있어요. 핵심은 이 유입을 얼마나 자연스럽게 앱 설치로 전환시키느냐예요.
**삼쩜삼**은 에어브릿지 솔루션을 활용해 웹에서 앱으로의 전환 성과를 극대화한 대표 사례예요. 앱 설치와 특정 전환 지표 간의 연결고리를 정확히 추적하고, 유입 지점별 전환율을 상세히 분석함으로써 웹에서 유입된 유저가 설치 후 어디까지 행동했는지를 정밀하게 파악할 수 있었죠. 이를 기반으로 삼쩜삼은 특정 페이지의 CTA 구성을 개선하고, 유입 채널에 맞는 연결 전략을 적용해 앱 설치와 전환 효율을 동시에 개선할 수 있었어요.
또 다른 사례로 **WAUG**는 웹 방문자에게 Braze의 In-App Message로 ‘앱 전용 혜택 보기’를 띄우고, Airbridge 딥링크를 통해 클릭 즉시 설치 후 원하는 화면으로 연결되도록 설계했어요. 그 결과, 상세페이지 전환율은 20%p, 회원가입 전환율은 50% 가까이 개선되었죠. 이처럼 웹 유입 직후의 퍼널 구조를 얼마나 매끄럽게 설계하느냐에 따라, 유입이 곧바로 성과로 이어질 수 있어요.
[__wf_reserved_decorative]
웹투앱은 단순한 보조 수단이 아니라, 이미 확보된 웹 트래픽을 매출로 전환하는 중요한 수익화 전략이에요. CTA 버튼의 위치, 디자인, 문구 표현, 메시지 타이밍과 맥락에 따라 클릭률과 전환율은 크게 달라질 수 있어요.
[👉](https://cshub.ab180.co/ko/case-studies/airbridge-biteme-organic-referrer-attribution) [앱 다운로드 778만 달성, 삼쩜삼은 어떻게 웹에서 앱으로 유저를 데려왔을까요?](https://www.airbridge.io/ko/case-studies/3o3-web-to-app-conversion-tracking)
[👉 웹투앱 전환율 20%, 회원가입 전환율 50%, 와그는 어떻게 만들어냈을까요?](https://cshub.ab180.co/ko/case-studies/airbridge-braze-webtoapp-waug)
## 2\. 클릭은 했는데 설치가 안된다면, 앱스토어가 문제예요
### 유저 맞춤 앱스토어 화면으로 전환을 높이는 전략
광고 타겟팅은 정교하게 설정했는데, 도달하는 앱스토어 화면이 하나뿐이라면 전환이 이뤄지기 어려워요. Apple의 Custom Product Page(CPP), Google의 Custom Store Listing(CSL)를 활용하면 광고 클릭 이후 도달하는 앱스토어 페이지도 개인화할 수 있어요.
예를 들어 **Expedia**는 항공권을 찾는 유저에게는 항공 중심의 스크린샷과 메시지를, 호텔이나 렌터카를 검색한 유저에게는 각각 다른 CPP를 구성해 보여줬어요. **CardSnack**s는 어버이날, 졸업 시즌 등 이벤트 시기에 맞춰 CPP 콘텐츠와 디자인을 바꿔가며 시즌성 수요에 대응했어요. MobileAction에 따르면 CPP를 적용한 앱은 평균 20~35% 더 높은 설치 전환율을 보였다고 해요.
[__wf_reserved_decorative]
MobileAction, CPP Intelligence, 2023 에서 다룬 Expedia 의 사례
[__wf_reserved_decorative]
MobileAction, CPP Intelligence, 2023 에서 다룬 Cardsnacks 의 사례
또한 CPP는 Apple Search Ads(ASA)와 연동돼 특정 키워드 검색 시 자동 노출이 가능하고, Google의 CSL도 세그먼트 기반 광고와 함께 활용할 수 있어요. 에어브릿지 트래킹 링크를 함께 쓰면 설치부터 LTV까지 전환 성과를 정밀하게 측정할 수 있어요. CPP와 CSL은 광고 클릭 이후 앱 설치로 이어지는 여정을 매끄럽게 만들어주는 전략적 퍼널 최적화 도구예요.
## 3\. 콘텐츠, 사람이 쓰지 않아도 돼요
### 데이터로 유입을 만드는 자동 콘텐츠 전략
예산 없이도 유입을 늘릴 수 있는 방법이 있어요. 앱이나 웹사이트의 데이터를 활용해 자동으로 검색 노출 콘텐츠를 생성하는 ‘프로그래매틱 SEO’ 전략이에요. 예를 들어 커머스 앱이라면 “TOP 10 베스트셀러” 페이지, 금융 앱이라면 “지난 주 거래량 상위 20개 주식 종목” 페이지를 자동으로 만드는 식이에요. 특히 상품 피드나 금융 데이터처럼 주기적으로 업데이트되는 정보를 가진 앱이라면 빠르게 적용할 수 있고, 유입도 자연스럽게 자동으로 늘어나는 구조라서 아주 효율적이에요.
하지만 검색 유입만으로는 부족해요. 중요한 건 유입을 앱 설치와 구매로 자연스럽게 이어주는 흐름이에요. 반려동물 라이프스타일 브랜드 **바잇미**는 오가닉 검색 유입 데이터를 광고 전략에 반영하고, 에어브릿지의 딥링크와 리퍼러 분석 기능을 활용해 웹 유입부터 설치, 구매, 재방문까지 전체 퍼널을 정밀하게 추적했어요. 그 결과, CPC를 20% 이상 절감하고 내부 검색 전환율도 높일 수 있었어요. 이런 자동화 전략을 더 정교하게 운영하고 싶다면, [**NNT**](https://www.nnt-consulting.com/)와 같은 마케팅 자동화 컨설팅 전문 기업의 지원도 고려해볼 만해요. NNT는 SEO부터 CRM까지 데이터 기반 마케팅을 자동화하고, 실시간 성과 모니터링과 전환 분석을 통해 검색 유입 이후의 흐름까지 체계적으로 관리할 수 있도록 도와줘요.
한 번의 설계만으로 수백 개의 콘텐츠 페이지가 자동으로 만들어지고, 그 유입이 자연스럽게 설치와 매출로 이어진다면 마케팅은 더 이상 수동이 아니라 구조적으로 자동화되는 시스템이 될 수 있어요.
[👉 CPC 20% 절감, 바잇미는 검색 유입을 어떻게 매출로 바꿨을까요?](https://cshub.ab180.co/ko/case-studies/airbridge-biteme-organic-referrer-attribution)
## 4\. 광고 없이 유저를 데려오는 구조, 아직도 안 쓰세요?
### 딥링크와 추천으로 만드는 자발적 유입 전략
광고 없이도 유저가 유저를 데려오는 구조, 바로 이것이 제휴 마케팅이 다시 주목받는 이유예요. 특정 링크를 통해 유입된 유저가 앱 설치나 구매 같은 전환을 일으킬 때만 비용이 발생하니까, 예산을 낭비하지 않고 퍼포먼스를 극대화할 수 있어요. 전환당 1~4% 수준의 보상이 주어져 참여 유인이 충분하고, 짠테크·앱테크처럼 수익을 나누는 흐름과도 잘 맞아 자발적인 확산을 유도하기 좋아요. 전환이 일어나면 보상은 자동 정산되고, 운영도 간편해요.
과거엔 웹 중심 구조 때문에 앱 설치나 인앱 전환으로 이어지기 어렵고, 성과 측정도 제한적이었어요. 하지만 이제는 에어브릿지 **딥링크** 기술 덕분에 유입부터 전환까지 여정을 세세하게 추적할 수 있어요. 특히 Airbridge **트래킹 링크 API**를 활용하면 파트너별로 고유 링크를 실시간 발급해 유입 경로를 정밀하게 구분할 수 있고, 클릭부터 전환까지의 데이터를 자동 수집해 Amazon S3로 전송, 자체 DB로 분석도 바로 가능해요.
성과가 발생한 채널에만 보상이 자동 분배되고, 인플루언서·리워드 앱·커뮤니티 등 다양한 파트너십도 효율적으로 운영할 수 있어요. 성과 기반으로 예산을 쓰고, 보상은 자동화되는 이 구조는 단순한 비용 절감을 넘어 브랜드 성장을 이끄는 전략이 될 거예요.
## 5\. '누가 왔는지'보다 '왜 왔는지'가 더 중요해요
### 어트리뷰션 데이터를 CRM에 연결하는 리인게이지먼트 전략
앱 설치는 시작일 뿐, 진짜 성과는 유저가 앱을 쓰고, 다시 돌아오며 반복 전환을 만들어낼 때 나와요. 이때 중요한 건 단순한 자동 메시지가 아니라 유입 맥락에 맞춘 정밀한 리인게이지 전략이에요. 그 출발점은 어트리뷰션 데이터를 CRM에 연동하는 거예요. 유저가 어떤 채널, 캠페인, 키워드를 통해 들어왔는지만 알아도 관심사와 목적을 예측할 수 있어요. 예를 들어 ‘여성 패션’ 광고를 클릭한 유저에겐 해당 카테고리 중심 콘텐츠를 보여주는 식이에요.
실제로 **마이리얼트립**은 에어브릿지 어트리뷰션 데이터를 Braze와 연동해 광고 유입 유저를 세그먼트화하고 맞춤 메시지를 발송했어요. 그 결과 CTR은 5%p 상승, 리타겟팅 광고 클릭률은 21%p 개선, 전환율은 12배까지 증가했어요. 단순히 ‘누가 왔는가’가 아니라, ‘왜 왔는가’를 이해하고 대응한 것이 성과의 핵심이었어요.
결국 지금 중요한 건 ‘어떻게 유입시킬까’가 아니라, ‘어떻게 남기고 반복시키느냐’에 대한 고민이에요. 어트리뷰션 기반 리인게이지먼트는 이 질문에 실질적인 해답을 줄 수 있는 전략이에요.
[👉](https://cshub.ab180.co/ko/case-studies/airbridge-biteme-organic-referrer-attribution) [같은 예산으로 전환율 12배, 마이리얼트립은 어떻게 해냈을까요?](https://cshub.ab180.co/ko/case-studies/braze-segmentation-myrealtrip)
## 6\. 자동화의 핵심은 데이터, 진짜 전환 시그널을 보내세요
### 광고 효율을 높이는 시그널 설계와 전환 API (CAPI) 연동 전략
요즘 광고 캠페인은 대부분 자동화되어 있지만, 이 자동화가 어떤 시그널을 기준으로 작동할지는 마케터가 정해야 해요. 단순한 설치나 회원가입보다 ‘환불 없는 구매’, ‘3만 원 이상 결제’처럼 실제 가치가 높은 행동을 시그널로 설계해야 광고 플랫폼이 제대로 학습할 수 있어요. 다만 너무 정교하면 학습량이 부족하고, 너무 넓으면 품질이 떨어지기 때문에, 양과 질의 균형이 중요해요.
여기에 전환 API(Conversion API, CAPI)를 함께 활용하면 SDK로는 수집하기 어려운 클릭 ID, 해시된 이메일, 전화번호, 사용자 행동 데이터를 서버에서 직접 광고 플랫폼으로 보낼 수 있어요. 이 덕분에 플랫폼이 더 정확하게 사용자 행동을 학습하고 타겟팅할 수 있게 되죠.
에어브릿지는 Meta, Google, TikTok 등 주요 매체와의 CAPI 연동을 지원해요. 이를 통해 앱 내 고가치 전환 데이터를 픽셀이나 일반 포스트백보다 안정적으로 전달할 수 있고, 예를 들어 Meta와 연동하면 페이스북·인스타그램에서 발생한 성과를 통합적으로 확인하고, 수집된 이벤트를 [Meta Ads](https://help.airbridge.io/ko/guides/meta-business-capi-integration)로 전송해 캠페인 최적화에도 활용할 수 있어요. 결국 자동화 캠페인의 성패는 전환 데이터를 얼마나 정확하게 수집하고, 그 정보를 얼마나 끊김 없이 매체에 전달하느냐에 달려 있어요.
[👉 전환 API가 어떻게 마케팅 성과를 극대화 하는지 궁금하다면?](https://www.airbridge.io/ko/blog/capi-for-omni)
## 예산보다 구조, 앱 마케팅의 대전환을 주도하세요
2025년 마케팅은 “얼마를 쓰느냐”보다 “어떻게 설계하느냐”의 싸움이에요.
- 웹 유입을 앱 설치로 연결하고
- 앱스토어를 유저 맞춤으로 구성하고
- 자동 콘텐츠로 오가닉 유입을 확보하고
- 딥링크 기반 제휴로 추천을 확산하고
- 유입 맥락 기반 리인게이지먼트를 설계하고
- 전환 시그널을 정교하게 전달하면
예산을 더 들이지 않고도 충분히 2배 이상의 성과를 만들 수 있어요. 지금 필요한 건 더 많은 광고비가 아니라, 더 잘 설계된 마케팅 구조예요. 에어브릿지는 위에서 소개한 다양한 전략을 조합해, 우리 브랜드에 가장 효과적인 성과 구조를 함께 만들어갈 수 있어요. 우리 마케팅에 어떤 변화가 가능할지 궁금하다면, 지금 바로 전문가와 상담해보세요.
---
## 2025년 게임 유저 확보 전략은 어떻게 달라져야 할까요? (The Future of Game UA 현장 스케치)
URL: https://www.airbridge.io/ko/blog/ua-seminar-sketch
Language: Korean
Published: 2025-05-26
[__wf_reserved_inherit]
모바일 게임 마케팅의 판이 빠르게 바뀌고 있어요. 급변하는 게임 시장에서 새로운 유저를 확보하기 위해서 많은 노력을 기울이고 계실 텐데요. 유저 확보 비용은 지속해서 상승하는데 그만큼 수익이 나지 않아 답답하셨을 거예요. 이제 단순히 좋은 게임을 만드는 것만으로는 부족하고, 유저의 취향에 맞으면서 데이터 기반 유저 확보 및 수익 최적화가 그 어느 때보다 중요한 과제가 되고 있어요.
이번 '**The Future of Game UA : 글로벌 트렌드와 데이터 기반 전략**' 세미나에서는 업계 최전선에서 활동하는 전문가들이 모여, 실전 경험과 데이터를 바탕으로 모바일 게임 UA 전략의 미래를 준비하는 방법을 공유했는데요. **AI 기반 타겟팅, 글로벌 퍼포먼스 최적화, 앱스토어 최적화(ASO)까지 실전에서만 알 수 있는 심도 있는 인사이트**가 쏟아졌어요. 그 생생한 현장을 함께 살펴보세요!
**오늘의 Key Takeaways**
- 중국 게임의 한국 진출 전략이 어떻게 변화했는지 알 수 있어요.
- AI를 활용해 유저별 광고 최적화로 수익과 리텐션을 높이는 방법을 알 수 있어요.
- 고가치 유저를 예측하고 이탈을 방지하는 실전 전략을 알 수 있어요.
- 앱스토어 최적화를 통해 광고비 없이 다운로드를 늘리는 방법을 알 수 있어요.
- 로열티 프로그램을 활용해 유저 리텐션과 과금율을 높이는 방법을 알 수 있어요.
- 앞으로 UA 마케팅에서 데이터 기반 세그먼트 전략이 왜 중요한지 알 수 있어요.
## **1\. ENP게임즈 : 중국 게임의 한국 진출 전략 & 한국 게임의 중국 진출 현실**
**이엔피게임즈(ENP Games)**의 방호선 님은 모바일 게임의 한국 시장 진출 및 마케팅을 담당했던 전문가로, 2017년 크게 성공했던 게임 ‘반지’를 포함해 여러 중국 게임의 한국 서비스를 진행했다고 해요. 이번 세미나에서는 **중국 게임의 한국 시장 진출 현황과 전략, 그리고 최근의 시장 변화에 대한 경험과 인사이트**를 공유해주셨어요.
[__wf_reserved_decorative]
### **변화하는 중국 게임사의 한국 진출 전략**
2019년 이후 중국 게임의 한국 진출 방식이 크게 변했다고 해요. 이전에는 한국 내 자회사가 대규모 브랜드 마케팅과 모델들을 활용하면서 직접 마케팅과 광고를 진행했는데요. 최근에는 중국 게임 인기가 높아지면서 **중국 본사가 직접 한국 시장을 관리하고 마케팅을 수행하고 있어요.** 마케팅 예산은 이전보다 다소 축소되고, 광고 대행사와의 협력 범위도 제한적인 편이라고 해요. 특히 AI를 활용한 대규모 광고 소재 제작과 최적화 테스트가 활발히 이루어지고 있다고 설명했어요.
### **중국 시장 공략 시 반드시 알아야 할 두 가지 이슈: 판호와 마케팅 트렌드**
중국 현지 시장에서 매출 창출에 중요한 이슈는 크게 두 가지예요. 하나는 '**판호**', 다른 하나는 ‘**마케팅 플랫폼 트렌드**’이죠.
[__wf_reserved_decorative]
판호란 중국 내에서 게임을 서비스할 수 있도록 허가받는 라이선스인데, 2018년부터 중국 정부의 강력한 규제로 발급이 중단되었다가, 최근 넷마블의 ‘세븐나이츠 키우기’ 사례와 같이 다시 완화되어 발급 시간이 최소 10개월에서 1년 정도로 줄어드는 추세예요. 하지만 경쟁 심화와 경기 침체 등 다양한 이유로 중국에 자회사가 있는 대형 게임사도 판호 획득에 어려움을 겪고 있어서, **판호를 받으려면 ICP 라이센스(중국 내 인터넷 콘텐츠 서비스 자격)를 먼저 확보해야 한다고 해요.**
또한 중국 경기가 어려워지다 보니 ‘**스트리머 기반 플랫폼**’과 ‘**대규모 할인 플랫폼**’이라는 새로운 마케팅 플랫폼이 등장하고 있어요. **스트리머 플랫폼**은 라이브 스트리머와 그 스트리머들이 소속된 길드를 활용한 일종의 인플루언서 마케팅이예요. **대규모 길드를 활용하기 때문에 도달과 매출을 크게 높일 수 있지만 그만큼 수익성은 낮죠.** 누적 매출이 100억 원이어도 실제 수익률은 10% 남짓이기도 해요. **대규모 할인 플랫폼은 말 그대로 엄청난 할인율을 바탕으로 매출을 극대화**해요. 예를 들어, ‘링디엔이저(중국어로 0.1)’라는 플랫폼에서는 게임 내 인앱 상품들을 모두 99% 할인하고 있다고 해요. 할인율이 엄청 나니 단기적으로 매출을 폭발적으로 높일 수 있어요. 하지만 그만큼 **콘텐츠 소모가 빨라서, 게임 내 콘텐츠를 추가로 개발하거나 레벨 상한을 풀어도 게임 수명이 6개월~1년 정도로 짧다는 단점**이 있다고 해요.
끝으로 중국 시장 진출은 매력적이지만 정책과 트렌드의 변화가 빠르고 변수가 많아 확실한 현지 파트너와의 협력이 매우 중요하다는 것을 강조하셨어요.
## **2\. 에어브릿지(Airbridge) : AI를 활용한 인앱 구매와 수익 최적화**
**에어브릿지(Airbridge)**의 함상빈 님은 모바일 게임 수익 최적화를 위한 AI 활용 방안과 실제 적용 사례를 공유했어요.
### **모바일 게임 수익 최적화를 위한 AI 활용의 필요성**
모바일 게임 시장은 연평균 5% 성장률을 기록하며 확장 중이지만, 경쟁은 심화되고 있어요. **특히 캐주얼 및 하이퍼 캐주얼 게임은 최근 5년간 수익이 180% 이상 증가**했지만, 비슷한 게임들이 빠르게 출시되면서 오리지널리티가 침식되고, 콘텐츠 업데이트나 유저 획득 전략에도 한계가 뚜렷하다고 강조했어요.
[__wf_reserved_decorative]
**이런 상황에서 광고 수익화, 특히 인터스티셜 광고 최적화의 중요성을 강조했어요.** 하지만 광고 빈도가 지나치면 유저 이탈이 증가하고, 적으면 수익이 감소하는 딜레마가 존재하죠. 이에 따라 **에어브릿지는 AI를 활용해 유저별 광고 빈도와 타이밍을 최적화하여 LTV를 극대화하는 방법을 제시**했어요.
구체적으로 글로벌 게임사 A사는 지역별 광고 최적화로 LTV를 12% 증가시켰고, 유저 리텐션도 31% 개선했어요. 국내 게임사 B사는 세그먼트별 광고 전략을 적용해 LTV를 45.9% 증가시켰어요.
### **에어플럭스: 광고 최적화를 자동화하는 AI 플랫폼**
하지만 이러한 최적화 작업을 수동으로 관리하는 것은 현실적으로 어렵죠. **그래서 에어브릿지는 광고 최적화를 자동화하는 AI 플랫폼 '에어플럭스'를 개발했어요.** 에어플럭스는 유저 행동 데이터를 실시간 수집·분석하여 최적의 광고 전략을 지속 추천하는데요.
[__wf_reserved_decorative]
실제 에어플럭스의 성과도 긍정적이에요. **2천만 다운로드 게임은 4주 만에 LTV가 54% 상승했고, 다른 캐주얼 게임사들도 13~18% 수준의 LTV 향상을 기록**했어요.
현재 에어플럭스는 별도의 광고 없이 글로벌 게임사들의 높은 관심을 받고 있어요. 성빈님은 향후 에어플럭스를 IP 전략까지 최적화하는 통합 수익화 플랫폼으로 발전시킬 계획이라고 설명했어요. "AI를 통해 유저 경험과 수익화를 함께 향상시킬 수 있다"는 메시지와 함께 발표를 마무리되었어요.
## **3\. 애피어 : 생성형 AI 및 예측적 인사이트와 함께 하는 새로운 시대의 UA**
**애피어(Appier)**의 이보혁 님은 이번 세미나에서 모바일 게임 퍼포먼스 마케팅에서 고가치 유저 예측과 생성형 AI를 활용한 광고 소재 제작 방안을 소개했어요.
[__wf_reserved_decorative]
### **고가치 유저 확보를 위한 AI 기반 예측 전략**
과거 카카오톡 게임하기, 구글 플레이, 페이스북 중심이었던 게임 마케팅 환경은 프로그래매틱 광고, AI 기반 DSP, 리워드 광고 시장의 확대로 진화했어요. 이에 따라 **정확한 타겟 오디언스 예측, 트래픽 관리, 광고 소재 최적화가 LTV 극대화의 핵심 과제**가 되었는데요.
[__wf_reserved_decorative]
에피어는 **매일 비드 요청 데이터를 수집해, 퍼스트 파티 데이터와 결합하여 고가치 유저를 예측**하고 있다고 해요. 북미 캐주얼 앱 데이터를 분석한 결과, 결제 유저들은 다양한 앱을 사용하고, 플레이어블 광고 소재가 고가치 유저 확보에 가장 효과적이었으며, 반면 LTV가 낮은 유저는 배너 광고가 효과적이었다고 해요. 또한 XGBoost와 LightGBM을 활용해 이탈 유저를 조기에 예측하는 모델도 소개했어요. 이를 통해 3~14일 내 복귀 유도를 목표로 리타겟팅 캠페인을 최적화했으며, 실제 한 해외 슈팅 게임 개발사는 이 모델로 KPI를 성공적으로 달성했다고 해요.
### **생성형 AI를 활용한 광고 소재 제작**
**생성형 AI를 활용한 플레이어블 광고 제작 사례도 공개했어요.** 게임 플레이 영상과 앱스토어 정보를 기반으로 10~15분 내에 광고를 제작할 수 있으며, 매체별 전환 예측 스코어링 기능도 지원한다고 해요.
마지막으로 에피어는 고가치 유저 확보, 이탈 방지, 생성형 AI 기반 소재 혁신을 모두 지원하는 통합 마테크 솔루션 기업으로 성장 중이며, 앞으로도 AI 기반 개인화 전략을 더욱 고도화해 마케팅 효율을 높이겠다고 밝혔어요.
## **4\. 앱트윅 : 앱스토어 마케팅 주요 KPI 및 성과 측정 방법 이해하기**
**앱트윅(AppTweak)**의 윤가비 님은 이번 세미나에서 앱스토어 최적화(ASO)와 성과 측정 전략을 주제로 발표했어요.
[__wf_reserved_decorative]
### **키워드 최적화와 크리에이티브 최적화의 중요성**
광고비 부담이 커지는 상황에서, 앱스토어 내 다운로드 수와 품질을 효율적으로 높이는 방법으로 '**키워드 최적화**'와 '**크리에이티브 최적화**'의 중요성을 강조했는데요. 기본 공식은 '**트래픽(노출) × 전환율 = 다운로드**'로, 관련성 높은 키워드를 확보하고 높은 순위로 노출시키는 것이 핵심이라고 말했어요.
[__wf_reserved_decorative]
이를 위해 **제목, 부제목, 키워드 필드 등을 지속적으로 최적화하고 모니터링**해야 한다고 말했는데요. ‘본디’나 ‘알라미’ 앱처럼 장기적으로 키워드 최적화를 진행할 경우 다양한 검색 키워드에서도 성과를 낼 수 있다고 설명했어요. 또한 ‘콜 오브 듀티’ 사례를 통해 실제 가시성과 순위 개선 효과도 소개했어요.
**크리에이티브 최적화에 대해서는 앱 아이콘, 스크린샷, 영상 같은 에셋이 전환율에 큰 영향을 미친다고 했어요**. 반복적인 A/B 테스트를 통해 최적의 에셋을 찾아야 하며, 국가별·플랫폼별 특성도 고려해 업데이트 효과를 세밀하게 분석할 필요가 있다고 조언했어요.
### **앱스토어 성과 측정과 함께 지속적으로 개선해야…**
마지막으로 성과 측정을 위해 임프레션, 페이지뷰, 다운로드 전환율 같은 주요 지표를 설정하고, 앱스토어 콘솔과 앱트윅 같은 전문 도구를 활용해 지속적인 데이터 분석과 개선을 이어가야 한다고 강조했어요.
ASO는 복잡해 보이지만, 체계적인 데이터 관리와 최적화를 통해 다운로드를 효과적으로 증가시킬 수 있다는 점을 다시 한 번 강조하며 발표를 마쳤어요.
## **5\. 플레이오 : 글로벌 트렌드, 로열티(리워드) 프로그램의 급부상**
**플레이오(Playio)**의 허정휘 님은 이번 세미나에서 로열티 앱을 활용해 게임 유저의 LTV(생애가치)를 높이는 전략을 소개했어요.
[__wf_reserved_decorative]
### **퍼포먼스 마케팅 한계를 넘어, 로열티 앱이 주목받는 이유**
**로열티의 개념을 항공사 멤버십 프로그램에 비유하며, 꾸준한 혜택이 유저 충성도를 높인다고 설명했는데요.** 미국의 쇼핑 영수증 리워드 앱 '패치(Fetch)' 사례를 통해 로열티 앱의 성장 가능성도 강조했어요.
게임 분야에서도 로열티 앱의 중요성이 커지고 있어요. 유저 획득 비용 상승과 개인정보 보호 강화로 퍼포먼스 마케팅이 어려워지는 가운데, **특히 보상형 콘텐츠에 익숙한 MZ세대에게 적합한 전략**으로 떠오르고 있다고 말했어요.
### **플레이오 사례로 보는 로열티 앱의 실질적 효과**
플레이오는 게임 플레이에 따라 포인트를 제공하고, 개인화된 게임 추천과 특정 행동 완료 보상으로 유저 이용 빈도와 과금율을 높이는데요. 주요 서비스는 CBT 모집, 사전예약, 플레이 타임 리워드, 리타겟팅, 퀘스트 제공 등이 있으며, 유저 리텐션을 적극적으로 강화한다고 해요.
특히 ‘던전’ 콘텐츠를 통한 협력 이벤트가 유저 대량 유입에 효과적이었다고 밝혔고, ‘블럭냥’과 ‘버서커 키우기’ 같은 실제 사례에서 높은 ROAS와 리텐션 수치를 기록해 로열티 앱의 효과를 입증했다고 해요.
[__wf_reserved_decorative]
마지막으로 허정휘 님은 로열티 앱이 단순 보상을 넘어 유저 데이터를 바탕으로 지속적인 게임 이용과 높은 LTV 창출을 이끄는 전략임을 강조하며 발표를 마무리했어요.
## **6\. 게임 UA 실무자가 전하는 글로벌 성장 전략**
마지막으로 게임사의 UA 전문가들이 모여 **2025년 UA 마케팅 트렌드와 전략**을 논의하고 참여자들의 질문에 답하는 좌담회가 열렸어요.
**넥슨, 카카오게임즈, 그램퍼스, Team 42** 등 다양한 게임사에서 활동하는 UA 실무자들이 참여해, AI 활용, 글로벌 캠페인 운영, 복귀 유저 전략 등 다양한 주제에 대한 인사이트를 공유했어요.
[__wf_reserved_decorative]
### **2025년 UA 마케팅 트렌드, 무엇이 달라질까?**
제일 먼저 **2025년 UA 마케팅의 주요 트렌드와 핵심 과제**라는 질문에 대해 4가지를 꼽았어요.
1. AI의 적극적 활용(넥슨)
2. 개인정보 보호 강화와 플랫폼 다변화(카카오게임즈)
3. 크리에이티브 전략 강화와 자사 생태계 구축(그램퍼스)이 꼽혔습니다.
4. 브랜딩과 퍼포먼스를 결합하는 과정에서의 AI 활용(Team 42)
다음으로 글로벌 캠페인 운영에 관한 질문이 있었는데요. 국가별, 연령별 선호 채널을 고려한 맞춤형 전략이 강조됐고, 특히 일본 시장은 장기적인 캠페인 운영과 현지화된 소재가 성공에 중요하다고 분석했어요.
복귀 유저 캠페인에 관한 질문에는, 넥슨과 카카오게임즈 모두 하드코어 RPG 장르 특성상 복귀 유저의 높은 LTV를 중요하게 관리하고 있으며, 복귀 예측 모델을 활용해 효율을 극대화하고 있다고 밝혔어요.
그리고 요즘 핫한 AI에 관한 질문도 빠지지 않았는데요. AI 기반 광고 소재 활용에 대해서는 빠른 테스트가 가능하지만, 장기 ROAS는 아직 기존 소재에 비해 낮은 편이라는 평가가 나왔고, 특히 여성 타겟 게임에서는 맥락을 고려한 세심한 기획이 필요하다고 조언했어요.
또한 앱스토어 최적화(ASO) 전략의 중요성도 언급됐는데요. 'BTS 쿠킹온' 사례를 통해 자연스러운 유저 유입에 효과적임을 설명했어요.
[__wf_reserved_decorative]
### **글로벌 UA 실무자들이 전하는 현실 조언**
끝으로, 글로벌 UA를 처음 시작하는 게임사들에게는 제한된 범위 내에서 작게 시작하고 점진적으로 확장할 것, 그리고 의사결정권자와의 원활한 소통 및 지속적인 테스트와 학습이 필수라고 조언하며 좌담회를 마무리했어요.
## **변화하는 UA 시장, 앞으로 무엇을 준비해야 할까?**
이번 세미나는 단순한 이론이 아니라, 실제 글로벌 게임 시장에서 부딪히며 얻은 생생한 데이터와 실전 전략을 들을 수 있는 귀중한 자리였어요.
특히 AI를 활용한 고가치 유저 예측, 개인화된 광고 최적화, 새로운 마케팅 플랫폼 트렌드, 앱스토어 최적화 전략까지 귀중한 경험과 인사이트를 배울 수 있었는데요. **앞으로의 UA 마케팅은 훨씬 더 정교하고 데이터 중심으로 변화하고 있다는 것을 실감할 수 있었어요.**
글로벌 시장 공략을 준비 중이거나, 게임 마케팅의 새로운 돌파구를 고민하고 있다면 오늘 다룬 다양한 전략과 사례를 꼭 참고해보세요.
---
## 일본 앱 마케팅, 실전에서 통하는 전략은 따로 있어요
URL: https://www.airbridge.io/ko/blog/win-app-marketing-boost-growth
Language: Korean
Published: 2025-05-22
지난 글에서는 일본 시장에 대한 오해와, 한국과는 전혀 다른 문화적인 차이들을 짚어봤어요. 이번엔 그 인사이트를 실제 마케팅 전략에 어떻게 적용할 수 있을지, 즉 실전 단계로 넘어가 보려고 해요.aix의 마케팅 총괄 Allen Park님과 나눈 대화를 바탕으로, 일본 시장에서 효과적으로 현지화하려면 어떤 방식으로 크**리에이티브, UA, 온보딩, 과금 전략**을 준비해야 하는지 정리해봤어요.
## 크리에이티브 현지화: 제품이 같아도, 스토리텔링은 달라야 해요
많은 글로벌 팀이 크리에이티브 현지화를 단순히 ‘텍스트만 일본어로 바꾸면 된다’거나, ‘일본식 이미지 몇 장 넣으면 충분하다’고 생각해요. 하지만 일본 유저는 단어나 비주얼 그 자체보다, ‘**그 안에 담긴 스토리와 맥락**’을 훨씬 중요하게 봐요.
> “같은 제품이라도 일본과 한국에서는 전혀 다른 얼굴을 가져야 해요. 일본은 신뢰가, 한국은 속도가 중요하거든요.”
— Allen Park
예를 들어 OTT 서비스의 광고를 살펴볼게요. 일본의 U-NEXT는 ‘전편 시청 가능’, ‘광고 없는 감상’처럼 안정성과 신뢰감을 전달하는 문구를 앞세워요. 반면 TVing은 ‘TVing에서만’, ‘지금 화제의 콘텐츠’ 같은 트렌드성과 한정성을 강조하죠.
[]
게임 광고도 마찬가지예요. 일본은 귀엽고 친숙한 캐릭터, 부드러운 색감으로 ‘정서적 안정감’을 주는 데 집중하고, 한국은 강렬한 전투 장면이나 눈에 띄는 캐릭터로 빠른 주목을 유도해요.
[]
결국 크리에이티브란, 유저가 앱을 처음 마주하는 진입로예요. 하나의 글로벌 크리에이티브로 두 시장을 동시에 공략하려 하면, 최소한 한쪽에서는 반응이 약해질 수밖에 없어요.
## UA 채널 전략: 시작은 글로벌, 확장은 로컬
Part 1에서 봤던 것처럼, 많은 마케터들이 일본 시장에선 처음부터 로컬 광고 채널을 써야 한다고 생각해요. 하지만 초반 성장을 원한다면 Google, Meta, TikTok, Apple Ads 같은 글로벌 채널이 훨씬 빠르게 반응을 가져올 수 있어요.
물론 일본에는 로컬 보상형 광고 문화가 잘 자리 잡혀 있어서, 어느 시점에서는 로컬 플랫폼도 꼭 필요해요. 일본 유저는 ‘작은 행동으로 보상을 받는’ 포이카츠 문화에 익숙해서 Moppy, Point Income, AMoAd, Zucks Ad Network 같은 채널이 매우 효과적이에요.
한국도 웹툰 기반 보상형 광고가 빠르게 성장 중이에요. AdPopcorn이나 TNK Factory 같은 플랫폼은 전환 퍼널 중간 구간에서 좋은 성과를 보여줘요. 일정한 유저 반응이 확보되면, Naver와 Kakao 같은 DSP 채널로 확장해보는 것도 좋아요.
> “로컬 채널에 처음부터 집착하지 말고, 먼저 잘 작동하는 글로벌 채널을 테스트해보고 나서 로컬 DSP를 추가하세요.”
— Allen Park
## 온보딩 & 결제 유도 화면: 시장의 속도에 맞춰야 해요
앱을 처음 실행해서 결제까지 이어지는 흐름도 일본과 한국은 완전히 달라요. 예를 들어 한국에서의 하이킹 관련 앱은 온보딩을 아주 간단하게 구성했어요. 짧은 문장, 빠른 권한 요청, 즉시 콘텐츠 접근이 가능한 구조였죠. ‘속도’ 에 민감한 한국 유저는 이런 구조에 익숙해서 잘 반응했어요.
그런데 일본에선 유사한 앱이 온보딩 플로우를 더 길게 가져갔어요. 설명과 인포그래픽을 풍부하게 넣었고, 그게 신뢰를 쌓는 데 도움이 됐어요. 일본 유저는 기능과 혜택을 충분히 이해해야 앱을 사용할 마음이 생기거든요.
[]
결제 유도 화면도 마찬가지예요. 일본 유저는 비교표, 자주 묻는 질문, 장문의 설명이 있는 페이지를 선호해요. 결정을 내리는 데 시간이 걸리지만, 한 번 결제하면 더 오래 쓰고 더 자주 들어와요. 반면 한국 유저는 ‘지금 결제하면 할인’, ‘이 앱은 000명이 사용 중’, ‘강력한 CTA’ 같은 자극적인 요소에 빠르게 반응해요.
[]
> “일본 유저는 설치부터 구독과 결제까지 시간이 오래 걸리지만, 충성도가 높고 리텐션도 좋아요.” — Allen Park
## 실전 사례: 현지화가 실제 성과로 이어진 앱들
aix는 다양한 글로벌 앱의 일본 진출을 지원하며, 현지화 전략이 실제 수치로 어떻게 전환에 영향을 주는지 꾸준히 검증해왔어요.
- **Tiimo**: 유럽 생산성 앱 Tiimo는 일본 스토어 설명과 키워드를 현지화한 뒤 다운로드 수가 70% 증가했어요.
- **TimeFlik**: 한국 시계 앱 TimeFlik은 일본 유저 눈높이에 맞춰 스크린샷을 바꾸고 전환율이 60% 증가했어요. Apple Search Ads의 CPI도 46% 낮아졌어요.
> “현지화는 완전히 새로 시작하는 게 아니라, 원래 잠재되어 있던 성장을 꺼내는 작업이에요.”
— Allen Park
## 일본 현지화 마케팅, 어디서부터 시작하면 좋을까요?
Allen Park님의 조언은 이래요.
> “ASO, 온보딩, 결제 유도 화면을 한 번에 다 현지화하면 좋지만, 우선순위를 둬야 한다면 무조건 ASO부터 시작하세요. 그게 첫 번째 전환 지점이에요.”
aix는 보통 12주의 시장 조사와 경쟁사 분석으로 시작해서, 24주간 UX와 크리에이티브를 업데이트해요. 그런 다음 A/B 테스트로 점진적으로 개선해 나가요. 현지화는 한 번으로 끝나는 프로젝트가 아니라, 유저의 반응을 듣고 계속 개선해 나가는 순환 구조예요.
## 정리: 두 개의 시장, 두 가지 전략, 하나의 마인드셋
일본과 한국은 지리적으로는 가깝지만, 유저 행동, 콘텐츠 취향, 전환 심리는 완전히 달라요. 일본 시장에서 성과를 내려면, 일본 시장에 맞는 전략이 따로 필요해요. **‘언어’가 아니라 ‘사고방식’을 현지화해야 해요. **일본과 한국의 앱 마케팅 전략을 비교하면 다음과 같이 정리할 수 있어요.
[]
## 성과로 이어지는 다음 스텝, aix와 에어브릿지
### 일본·한국 유저에게 진짜 통하는 현지화, aix와 시작하세요
aix는 단순한 번역을 넘어, 진짜 '문화적 적합성'에 집중하는 현지화 파트너입니다.
키워드 리서치부터 과금 UX까지, 일본과 한국 시장에 뿌리를 둔 현지화 전문팀이 앱의 성장을 돕습니다.
> → 일본 시장에 맞는 로컬라이징, [aix와 함께 준비하세요](https://www.aixinc.io/kr)
### 설치부터 구독까지, 데이터로 연결된 성장을 경험하세요
에어브릿지는 퍼널 전반을 연결해주는 통합 마케팅 측정 솔루션입니다. 설치부터 결제 그리고 구독까지, 어떤 전략이 효과가 있는지 명확하게 측정하고 최적화하세요.
• Apple Search Ads 키워드 실시간 인사이트로 성과 극대화
• TikTok, Meta 등 UGC 캠페인 실행 및 성과 측정
• 웹과 앱 전환 흐름을 하나로 연결하는 Web & Mobile SDK
• 설치부터 구독까지 전환 퍼널 전체 측정
• SKAN 데이터 기반 iOS 트라이얼 성과 향상
• 행동 기반 세그먼트로 이탈 유저 리타게팅
• RevenueCat, Adapty 연동을 통한 LTV 분석
> → 데이터로 더 스마트하게 성장하고 싶다면, [지금 문의해보세요](https://www.airbridge.io/ko/book-a-demo)
---
## AI 마케팅 시대, 지금 도입해야 할 핵심 전략 가이드
URL: https://www.airbridge.io/ko/blog/ai-marketing-airflux
Language: Korean
Published: 2025-05-16
최근 Bain & Company는 "[AI가 소비자의 구매 결정을 대신하는 시대가 오고 있다](https://www.bain.com/insights/marketings-new-middleman-ai-agents/)"고 강조했어요. 이제 브랜드는 소비자뿐 아니라, 소비자의 선택을 대신하는 AI 시스템에도 선택받아야 하는 시대를 맞았어요.
AI는 지금, 거의 모든 산업과 사회 전반을 뒤흔들고 있어요. 마케팅은 물론, 제조업, 금융, 헬스케어, 유통, 예술과 창작 분야까지 — AI가 닿지 않은 곳을 찾기 어려울 정도예요. 인간의 시간과 노동, 비용 제약 때문에 불가능하거나 비효율적이었던 일들이 AI를 통해 가능해지면서 전 산업의 패러다임이 빠르게 바뀌고 있어요.
생산성과 효율성을 폭발적으로 높이고, 완전히 새로운 가치를 만들어내는 AI. 그렇다면 우리 마케팅에는 어떻게 접목해야 시장을 선도할 수 있을까요? 지금부터 꼭 알아야 할 필수 전략을 소개할게요.
## AI로 광고 소재에 날개 달기
### 하루 1,000개 광고 소재 제작, 효율이 달라진다
캠페인 세팅과 타겟팅은 점점 자동화되고 있는데, 광고 소재 제작만 여전히 수작업에 머물러 있다면 효율성은 한계에 부딪칠 수밖에 없어요.최신 AI 기반 광고 소재 제작 툴을 활용하면 대량의 소재를 빠르게 만들고, 더 자주 테스트하며, 더 많은 데이터를 기반으로 최적화할 수 있어요.
브랜드가 보유한 제품 카탈로그는 광고 소재로 전환할 수 있는 강력한 원천이에요.AI를 활용하면 카탈로그 내 이미지를 다양한 배경, 텍스트와 조합해 빠르게 가공할 수 있어요.큰 공수 없이 제품을 홍보할 수 있고, 영상 콘텐츠는 플랫폼 알고리즘에도 더 잘 노출돼 사용자 몰입도와 클릭률(CTR), 전환율 향상까지 기대할 수 있어요.
또한 최신 AI 기술을 활용하면 가상의 인물을 통해 실제 소비자가 제품을 리뷰하는 듯한 콘텐츠를 제작할 수 있어요. 목소리, 제스처, 말투까지 조정 가능한 가상 UGC(User Generated Content)는 브랜드가 일관된 메시지를 유지하면서도 소비자 친화적인 콘텐츠를 대량 제작할 수 있도록 도와줘요.
[__wf_reserved_decorative]
AdCreative.ai 가 제품 카탈로그를 광고로 변환하는 예시
## AI로 광고 성과의 판 바꾸기
### 실시간 성과 최적화의 열쇠, LTV 와 리텐션 잡기
광고 소재를 빠르게 제작하고 다양한 버전을 테스트했다면, 그다음에는 캠페인 성과를 정밀하게 측정하고, 더 나아가 미래까지 예측하는 단계가 필요해요.
에어브릿지의 리텐션 리포트는 광고 소재별 리텐션을 분석해, 어떤 광고 소재가 사용자의 재방문 빈도에 긍정적인 영향을 주는지 인사이트를 제공해요. 특히 초반 이탈이 많은 앱이라면, 분 단위, 시간 단위로 리텐션을 드릴다운해 보다 세밀한 최적화가 가능해요. 광고 소재 단위로 ARPU, ROAS도 분석해서 고가치 유저를 유치하는 전략적 투자 결정을 내릴 수 있어요.
> 👉 효율적인 소재 최적화로 로그인율 90%, ROAS 300% 개선에 성공한 [웹젠 뮤 모나크 성공 사례](https://www.airbridge.io/ko/case-studies/mu-monarch-creative-optimization)를 살펴보세요!
단순히 현재의 성과만 보는 데 그치지 않고, 미래를 예측하는 것도 중요해요. 에어브릿지의 **Predictive LTV(pLTV)** 기능을 활용하면, 광고 집행 후 단 3~14일 만에, 최대 180일 후까지의 리텐션과 매출을 예측할 수 있어요. CAC(유저 획득 비용) 대비 미래 수익을 미리 파악해, 광고 예산을 더 빠르고 정교하게 재배분할 수 있어요.
> 👉 딜라이트룸(알라미)은 pLTV 기능을 활용해 글로벌 마케팅 전략을 한층 더 정교하게 다듬을 수 있었어요. 자세한 내용은 [이 블로그](https://www.airbridge.io/ko/blog/pltv-optimization-for-apps)에서 확인할 수 있어요.
## AI로 데이터 시그널 살리기
### 전환 API(CAPI) 로 광고 품질 높이는 비밀
광고 매체의 AI 알고리즘의 핵심 동력은 ‘시그널 데이터’에서 나와요. 시그널 데이터란 사용자 행동 데이터의 다른 말인데요, 광고주의 앱에서 수집되어 픽셀, SDK, API를 통해 광고 매체로 전송되는 사용자 행동 데이터를 말해요. 광고 매체는 특정 전환을 달성할 가능성이 가장 높은 사용자 식별을 위해 머신러닝과 AI를 사용해 이 시그널 데이터를 분석해요.
하지만 문제는 최근 개인정보 보호 움직임에 따른 변화로 기존 시그널 데이터에서 매칭 키로 활용되던 ADID 와 써드 파티 쿠키의 감소 및 질적 하락으로 매칭률이 하락하면서 광고 효율성도 감소하고 있는 부분이예요. 이에 대한 대안으로 전환 API (Conversion API) 가 주목받고 있어요. 추가적인 ID (클릭 ID, 해시화된 이메일)를 보냄으로써 시그널 데이터의 매칭률 향상을 가능하게 하는 방식이에요.
[__wf_reserved_decorative]
에어브릿지는 더욱 풍부한 시그널을 광고 플랫폼에 전송할 수 있도록 전환 API 연동을 메타, 틱톡, X(트위터), 스냅 등에 대해 지원하고 있어요. 이를 통해 매체 성과를 추가 최적화할 수 있어요.
## AI로 성과 측정과 최적화까지 완성하기
### 놓치고 있던 전환과 사기까지 한 번에 잡기
광고 성과를 제대로 측정하지 못하면 최적화는 한계에 부딪칠 수밖에 없어요. 특히 개인정보 보호 강화로 ADID 수집이 제한되면서, 광고로 유입된 유저임에도 '미기여 전환'으로 분류되는 사례가 많아졌어요. 광고 성과는 과소평가되고, 최적화 방향도 흔들릴 수 있죠. 에어브릿지의 **UCA(Unattributed Conversion Analysis)** 기능은 최신 ML/AI 기술을 활용해 과대 추정된 미기여 전환을 찾아내고, 실제 기여한 채널로 정확히 분배할 수 있도록 도와줘요. 덕분에 놓치고 있던 광고 성과를 복원하고, 더 신뢰할 수 있는 데이터로 최적화 전략을 세울 수 있어요.
또한 광고 사기로 인한 성과 왜곡도 무시할 수 없어요. 에어브릿지의 **비정상 설치 분석 기능**은 앱 설치 과정에서 발생할 수 있는 다양한 유형의 광고 사기를 자동으로 탐지해 빠르게 대응할 수 있게 해줘요. 필요하면 원데이터(raw data)로 상세 분석까지 가능해, 예산 낭비를 줄이고 광고 품질을 높이는 데 도움이 돼요.
[__wf_reserved_decorative]
### AI로 광고 성과를 한층 더 끌어올리기: Airflux에 주목하세요
최근 에어브릿지는 **AI 기반의 인앱 광고 수익 최적화 솔루션, Airflux**를 선보였어요. Airflux는 사용자의 행동 데이터를 기반으로 **강제 광고(Interstitial Ads)의 노출 빈도와 타이밍을 유저별로 최적화해요.** 이를 통해 사용자 경험을 해치지 않으면서도 광고 수익을 극대화할 수 있어요.
특히 빠르게 변하는 게임 유저의 행동 패턴을 실시간으로 분석해, **가장 적절한 순간에 광고를 노출**함으로써 이탈을 최소화하고 수익을 높일 수 있도록 설계됐어요. **인앱 광고 운영에 있어 가장 까다로운 '빈도와 타이밍' 최적화**에 특화된 솔루션으로, 글로벌 게임 시장에서의 수익 극대화를 돕고 있어요.
> 👉 게임 내 광고 수익 최적화가 과제라면, 지금 바로 [Airflux를 경험](https://www.airflux.ai/)해보세요.
## AI 마케팅, 지금 시작해야 앞서갑니다
AI는 이제 단순한 도구를 넘어, 마케팅의 게임 체인저로 자리잡고 있어요. 광고 소재 제작부터 성과 최적화, 데이터 기반 전략 수립까지, AI를 얼마나 빠르고 정확하게 도입하느냐가 브랜드의 성장 속도를 결정짓게 됐어요.
앞으로 1~2년 안에는, AI가 단순한 제작 보조를 넘어 캠페인 세팅, 타겟팅, 예산 운용, 퍼포먼스 최적화까지 마케팅 전 과정을 실시간으로 관리하는 파트너가 될 거예요. 마케터는 전략적인 의사결정에 집중하고, AI는 실행과 최적화를 담당하는 새로운 협력 구조가 본격화될 것으로 보여요.
변화는 이미 시작됐어요. 지금 AI 마케팅을 도입하고 체계적으로 활용하는 브랜드만이, 다가올 새로운 시장 질서 속에서 빠르게 도약할 수 있을 거예요. 에어브릿지와 함께라면, AI를 활용한 광고 성과 최적화부터 데이터 기반 미래 예측까지 한층 더 빠르게 완성할 수 있어요.
에어브릿지가 우리 마케팅을 어떻게 도와줄 수 있을지 궁금하다면, 지금 바로 확인해 보세요!
---
## 일본 시장, 왜 어려울까? 앱 마케터가 빠지기 쉬운 3가지 함정
URL: https://www.airbridge.io/ko/blog/win-app-marketing-bust-myths
Language: Korean
Published: 2025-05-16
일본은 전 세계 앱 퍼블리셔에게 가장 매력적이면서도 가장 도전적인 시장 중 하나예요. 스마트폰 보급률이 높고, 유료 앱이나 구독형 서비스에도 지갑을 여는 데에 큰 저항이 없죠. 특히 모바일 게임이나 구독 앱은 글로벌 확장의 핵심 거점으로 일본을 꼽을 만큼 잠재력이 큽니다. 그런데도 많은 글로벌 팀이 일본에서 성과를 내지 못하는 이유는, ‘기회가 없어서’가 아니라 시장 자체를 오해하고 있기 때문이에요.
이번 글은 일본·한국 시장 전문 컨설팅 회사 [aix](https://www.aixinc.io/kr)의 마케팅 총괄 [Allen Park](https://www.linkedin.com/in/allen-park-165a91243/)님과의 인터뷰를 바탕으로, 일본 앱 마케팅에서 자주 보이는 대표적인 오해들을 짚어보고, 실제 유저 행동과 전환, 성장을 좌우하는 문화적 차이를 정리해봤어요.
**🎥 전체 인터뷰 영상 보러 가기 (영문)**
## 일본: 잠재력은 크지만, 리스크도 커요
RevenueCast 2025 보고서에 따르면, APAC 지역의 평균 구독 앱 수익이 북미와 유럽 수준에 거의 근접했어요. 특히 일본은 시장 크기나 유저의 구매력을 고려하면 구독형 앱이나 모바일 게임이 글로벌하게 확장할 때 반드시 짚고 넘어가야 할 핵심 시장이에요. 그런데 많은 창업자들은 일본 시장의 복잡함은 과소평가하고, 다른 아시아 국가들과 비슷하다고 오해하는 경우가 많아요.
> “많은 창업자들이 한국과 일본 시장이 하나의 전략이면 다 될 거라고 생각해요. 하지만 일본과 한국은 문화도 다르고, 행동 방식도 다르고, 언어도 완전히 달라요. 이런 걸 처음부터 고려하지 않으면 실패하기 쉬워요.” — Allen Park, aix 마케팅 총괄
## 일본 앱 마케팅에서 자주 나오는 오해들
### 오해 1. “일본이랑 한국은 비슷하니까 전략도 비슷하게 가면 돼요”
두 나라는 지리적으로는 가까워도, 유저가 어떻게 생각하고 행동하는지는 완전히 달라요. 일본 유저는 꼼꼼하고 신중하게 정보를 따져보는 편이고, 상세한 설명을 확인한 후에야 전환해요. 반면 한국 유저는 빠르게 트렌드를 따라가고, ‘지금 아니면 안 돼’ 같은 메시지에 금방 반응하죠.
예를 들어 일본과 한국 유저들은 사용하는 검색 엔진부터 달라요. 일본은 시장에서 Google 점유율이 80%가 넘는 반면, 한국은 Naver가 여전히 1등이에요. 이런 디지털 행동 패턴도 서로 다르다는 걸 보여줘요.
### 오해 2. “일본에서는 애니메이션을 활용한 마케팅이 무조건 통해요”
많은 회사들이 일본 마케팅을 할 때, 애니메이션 캐릭터나 일본식 디자인만 활용하면 전환율이 오를 거라고 기대해요. 하지만 서비스와 연관성이 없다면 오히려 역효과가 날 수 있어요. 마치 한국에서 마케팅한다고 해서 무조건 K-pop 아이돌 사진을 쓰는 게 정답은 아닌 것과 같은 이치예요.
> “서비스가 애니메이션이랑 전혀 관련이 없는데 굳이 그걸 쓰면 유저가 ‘이게 왜 있지?’라고 생각해요.”
> — Allen Park
물론 연애 시뮬레이션 게임처럼 애니메이션 스타일이 자연스럽게 어울리는 경우엔 예외예요. 하지만 금융 앱이나 운동 앱이라면, 애니메이션보단 기능성과 신뢰감, 그리고 현지 문화에 맞춘 메시지가 훨씬 중요해요.
### 오해 3. “일본은 무조건 로컬 광고 플랫폼이 성과가 좋아요”
Yahoo! Japan이나 LINE 같은 일본 플랫폼이 Meta나 Google Ads보다 항상 더 좋을 거라고 생각하는 것도 오해예요. 초기 시장 진입 단계에선 글로벌 채널이 빠르고 효율적인 성과를 내는 경우가 많아요. 로컬 플랫폼은 세팅이 까다롭고 초기 파트너십 구축도 필요하기 때문에, 보상형 광고나 확장 단계에서 더 효과적이에요.
> “초기엔 글로벌 플랫폼으로 빠르게 테스트하고, 이후에 로컬 DSP나 오퍼월을 써보는 걸 추천해요.”
> — Allen Park
## 일본 유저가 앱을 고르는 기준
Hofstede가 분석한 내용에 따르면, 일본 유저는 불확실성을 기피하고 목적 지향적인 성향이 강해 정보를 충분히 확인한 다음에야 결정을 내려요. 그래서 신뢰를 줄 수 있는 설명, 자세하고 명확한 기능 소개가 중요해요.
이런 특징은 온보딩부터 광고 크리에이티브까지 모든 접점에서 나타나요. 일본 유저는 감정보다 이성이 먼저예요. 직관적인 흐름, 깔끔한 번역, 친숙한 스타일이 중요해요.
> “일본 유저는 ‘내가 이 앱을 선택한 게 현명한 결정인가’에 집중해요.”
> — Allen Park
## 오해를 극복한 사례: Slopes는 어떻게 일본 시장에서 반등했을까?
스키·스노보드 앱 Slopes도 처음에는 많은 글로벌 앱과 마찬가지로, 기계 번역에 의존해 일본 시장에 진출했어요. 하지만 명확한 반응을 얻지 못한 채 어려움을 겪었죠.
이후 aix와 협업하면서 전략을 완전히 재설계했고, 아래와 같은 방식으로 ‘‘일본 유저의 기대와 행동에 맞춘 현지화’를 실행했어요:
- 실제 일본 유저의 검색 패턴을 반영해 키워드를 최적화하고
- 일본 문화와 시각적 기대에 맞춰 앱 스크린샷을 재구성했으며
- 신뢰할 수 있는 현지 파트너와 협업해 브랜드 신뢰도를 높였어요
그 결과 오가닉 유입이 눈에 띄게 증가했고, 리텐션 역시 크게 개선됐어요. 지금은 Slopes의 3대 핵심 시장 중 하나로 일본이 자리 잡았죠.
## 이제, 성과를 만든 전략들을 살펴볼게요
이번 글에서는 일본 시장에 대한 대표적인 오해와, 한국과는 전혀 다른 문화적 차이를 짚어봤어요. 이번 글을 요약하자면:
- 일본과 한국은 같은 전략으로 접근할 수 없는, 문화적으로 완전히 다른 시장이에요.
- 애니메이션 같은 문화 요소는 맥락에 맞을 때만 효과를 발휘해요.
- 글로벌 UA 채널도 일본 시장에서 충분히 효과적일 수 있어요.
- 일본 유저는 신뢰와 명확성을, 한국 유저는 속도와 트렌드, 희소성을 중시해요.
- 로컬라이징은 언어 번역이 아니라, 유저의 사고방식과 행동 방식을 이해하는 일이에요.
다음 편에서는 이렇게 정리한 문화적 차이들이 실제 마케팅 전략에 어떻게 반영됐는지를 살펴볼 거예요. 단순히 시장을 이해하는 데서 끝나는 게 아니라, 실제 팀들이 어떤 방식으로 광고 크리에이티브를 만들고, 어떤 채널을 선택하고 운영했는지, 또 일본 유저에게 맞는 메시지를 어떻게 현지화했는지까지 하나하나 짚어볼 예정이에요. 이런 전략적 선택들이 어떻게 유의미한 성과로 이어졌는지 구체적인 사례를 통해 보여드릴게요.
---
## 에어브릿지가 구글 통합 전환 측정(ICM)을 지원해요
URL: https://www.airbridge.io/ko/blog/airbridge-google-icm
Language: Korean
Published: 2025-05-14
에어브릿지가 구글의 [통합 전환 측정(Integrated Conversion Measurement, ICM)](https://support.google.com/google-ads/answer/16203286)을 지원해요. 이제 에어브릿지를 통해 구글의 비확정적 매칭 데이터에 대한 더 깊이 있는 인사이트를 확보할 수 있어요. 이번 지원을 통해 에어브릿지 대시보드에서 확률적 전환을 더 정밀하게 추적하고, 개인정보 보호가 준수된 풍부한 데이터를 활용해 캠페인을 최적화할 수 있어요.
비확정적 매칭 데이터란 개별 유저를 식별하지 않으면서, 관찰된 전환 데이터를 활용해 관찰되지 않은 전환을 추정하는 방식이에요. 이를 통해 보다 정밀하게 측정된 전환을 바탕으로 캠페인을 최적화할 수 있어요.
## 통합 전환 측정(ICM)이란 무엇인가요?
ICM이란 점점 더 강화되는 추적 제한으로 인해 발생하는 전환 추적 공백을 메우기 위해 구글이 선보이는 개인정보 보호 기반 솔루션이에요. 최근 개인정보 보호 규제가 강화되면서 제3자 쿠키나 기기 ID를 기반으로 사용자를 구분하는 기존 추적 방식은 광고 성과를 완전히 파악하는데 어려움이 있어요. ICM의 프레임워크는 개인정보 보호를 준수하면서도 이러한 간극을 좁히기 위해 설계되었어요. ICM은 광고가 전환으로 이어졌다는 높은 신뢰가 확보되었을 때에만 비확정적 매칭 데이터를 전환에 포함하기 때문에, 유저의 개인정보를 보호하면서도 정밀한 전환 데이터를 얻을 수 있어요.
이번 지원을 통해 구글 애즈 UI에서만 확인할 수 있었던 구글의 비확정적 매칭 데이터를 에어브릿지에서 직접 활용할 수 있게 되었어요. 이제 에어브릿지 대시보드에서 광고 성과 측정의 사각지대를 최소화하고, 이전보다 더욱 다양하고 정밀한 전환 추적 방식을 경험해 보세요.
## 에어브릿지에서 ICM을 활용하면 어떤 장점이 있나요?
**1. 향상된 전환 추적이 가능해요**
구글의 비확정적 매칭 데이터를 에어브릿지에서 직접 활용할 수 있어 더 정밀하고 심층적인 성과 분석이 가능해요. 매칭 데이터는 에어브릿지 대시보드에 온전히 제공되는 것은 물론, 구글 애즈에서 확인할 수 있는 수준보다 더 깊은 인사이트를 제공해요. 더 자세한 비확정적 매칭 데이터를 에어브릿지 대시보드에서 활용해 마케팅 효과를 더욱 정확하게 측정하고, 효과적인 광고 최적화를 경험해 보세요.
**2. 데이터 분석이 더 쉬워져요**
더 이상 데이터 수집을 위해 번거롭게 여러 플랫폼을 오가지 마세요. 이제 에어브릿지 대시보드 하나로 구글에서 제공하는 비확정적 매칭 데이터 정보까지 직접 확인해 통합적으로 분석할 수 있어요. 이를 통해 캠페인 최적화가 더욱 수월해지고, 데이터 분석 과정이 간소화되어 의사결정의 효율성을 높일 수 있어요.
**3. 개인 정보 보호를 준수하면서도 성과를 높일 수 있어요**
비확정적 매칭은 기존 쿠키 기반 추적 방식과 달리 머신러닝을 통해 가능성을 예측하는 방식이기 때문에 개인정보 보호 규정을 준수하면서도 광고 성과를 효과적으로 측정할 수 있어요. 또한 구글은 IP 기반 추적(핑거프린팅)이나 디지털 지문 수집 등 모든 침해적인 식별 방식을 금지하는 것은 물론, 높은 신뢰도의 추정치만 전환 데이터에 포함하도록 엄격히 설계했어요. 마케터 입장에서는 개인 정보 보호를 준수하면서도 신뢰할 수 있는 지표를 기반으로 캠페인을 최적화할 수 있어요.
## 에어브릿지에서 구글 ICM 설정하기
에어브릿지에서 구글 ICM을 설정하는 방법은 [에어브릿지 가이드](https://help.airbridge.io/ko/developers/ios-sdk-v4#icm-%EC%96%B4%ED%8A%B8%EB%A6%AC%EB%B7%B0%EC%85%98-%EB%8D%B0%EC%9D%B4%ED%84%B0-%ED%99%95%EC%9D%B8%ED%95%98%EA%B8%B0)를 확인해 주세요. ICM 지원 이후 에어브릿지를 활용한 캠페인 최적화 전략 등 궁금하신 점이 있다면 문의해 주세요.
---
## 모바일 게임 리인게이지먼트 & 리타겟팅 마케팅 완전 정복 가이드
URL: https://www.airbridge.io/ko/blog/mobile-game-reengagement-and-retargeting
Language: Korean
Published: 2025-05-07
[](https://www.linkedin.com/in/taoufik-l-75640646/)모바일 게임 시장의 성장과 함께 경쟁이 치열해진 요즘, 단순한 유저 확보(UA, User Acquisition)만으로는 성과를 유지하기 어려워지고 있어요. 이제는 기존 사용자 이탈을 막고 이탈된 사용자를 다시 게임으로 돌아오게 하는 리인게이지먼트와 리타겟팅 전략이 그 어느 때보다 중요해지고 있죠.
[Adikteev의](https://www.adikteev.com/) CRO(Chief Revenue Officer)인 [Taoufik Lazrak](https://www.linkedin.com/in/taoufik-l-75640646/)은 게임 리타겟팅 업계 선두주자로서 King, Blizzard, Rovio, Playtika 등 400개 이상의 게임 타이틀과 협업하면서 ‘인크리멘탈 밸류(Incremental Value)’라는 개념을 개척해 왔죠. Taoufik은 지난 10년 넘게 Adikteev에서 일하면서 게임 마케팅의 진화를 직접 이끌어왔고, 지금은 고도화된 리인게이지먼트 및 리타겟팅 전략을 선도하고 있어요.
[에어브릿지](http://www.airbridge.io/ko) 공동 창업자이자 CEO인 [남성필 대표](https://www.linkedin.com/in/roinam/)가 Taoufik과 함께 게임 앱을 위한 리인게이지먼트와 리타겟팅 전략에 대해 이야기했어요. 실전에서 검증된 전략부터 흔히 오해되는 점들, 그리고 개인정보 보호가 강화된 시대에 대응하는 방법까지, 그 대화를 지금 만나보세요.
**🎥 전체 인터뷰 영상 보기**
## **게임 리인게이지먼트와 리타겟팅 오해와 진실**
게임 업계에서 자주 들을 수 있는 리인게이지먼트와 리타겟팅 마케팅에 대한 대표적인 오해와 진실은 무엇이 있을까요? 이번 인터뷰 첫 주제는 게임 리인게이지먼트와 리타겟팅 오해와 진실이에요.
### **Q. 게임 마케팅을 하며 자주 듣는 오해 3가지만 꼽는다면 무엇이 있나요?**
**A.** 정말 좋은 질문이에요. 자주 듣는 오해들이 꽤 있는데요, 그중 대표적인 세 가지를 말씀드릴게요.
#### **오해 1: 캐주얼 게임에는 리인게이지먼트가 효과 없어요**
캐주얼 게임의 경우 유저 생애 주기(이하 LTV)가 짧아 리인게이지먼트는 의미가 없다고 말하곤 해요. 하지만 그건 사실이 아니에요. 뾰족한 타겟과 그에 맞는 크리에이티브 전략을 세운다면 캐주얼, 하이브리드 캐주얼 게임에서도 확실한 효과를 볼 수 있어요.
게임 내 이벤트 혹은 신규 콘텐츠가 출시되었을 때같이 수익을 이끌 수 있는 주요한 모멘텀에 맞춰 캠페인을 진행한다면 성과를 극대화할 수 있어요. 이때 핵심은 ‘개인화’라고 할 수 있어요. 개인에게 소구되는 메시지를 찾고 적절한 타이밍에 맞춰 마케팅을 진행할 수 있다면 유저 반응은 확실히 달라져요.
#### **오해 2: 뷰스루(View-through) 지표는 중요하지 않아요**
흔히 광고를 봤지만 CTA를 클릭하지 않은 유저는 의미가 없다고 생각하는 경우가 많은데요. 광고 접점이 다양해지는 모바일 마케팅 환경에서는 뷰스루, 즉 광고를 확인 후 이탈한 유저도 꼭 챙겨야 해요. 특히 게임은 반복적인 노출이 성과로 이어지는 업계에요. 단순 클릭 외에도 유저가 광고를 봤다는 사실 자체가 영향을 줄 수 있어요. 그래서 다양한 채널을 오가는 어트리뷰션 모델을 도입하는 일이 중요한 것이죠.
[__wf_reserved_decorative]
#### **오해 3: 인크리멘탈리티는 신뢰할 수 없어요**
리인게이지먼트 캠페인은 ‘기존 유저를 다시 데려오는 거라 실질적인 성과가 없다’는 오해도 많아요. 하지만 저희는 [인크리멘탈리티](https://help.airbridge.io/ko/guides/incrementality-report-understanding)(증분) 테스트를 기반으로 리인게이지먼트 캠페인의 성과를 증명하고 있어요.좋은 가설을 세워 테스트를 진행한다면 충분히 신뢰할 수 있는 결과를 이끌어낼 수 있어요.
## **캐주얼 게임도 리인게이지먼트로 매출을 올릴 수 있어요**
인크리멘탈리티 테스트는 테스트 그룹과 컨트롤 그룹을 나누는 작업부터 시작돼요. 쉬운 작업은 아니지만, 유의미한 결과를 도출하기 위해서는 테스트가 편향되지 않도록 두 그룹을 설정하는 일이 중요하죠. Adikteev는 어떤 방식으로 그룹을 나누어 테스트를 진행할까요?
### **Q. 고가치 유저를 대상으로 리인게이지먼트 캠페인을 할 때, 어떻게 인크리멘탈리티를 측정하고 편향을 방지하세요?**
**A.** 저희는 원래 의학 분야에서 즐겨 쓰이던 ITT(Intention-to-Treat) 방법론을 활용해요. 마치 임상실험처럼, 광고에 노출된 유저 그룹과 그렇지 않은 그룹을 나눠서 실제 수익 변화나 활동성 차이를 분석하는 방식이에요. ITT 방법론을 활용하면 캠페인 성과가 진짜 캠페인 덕분에 발생한 성과인지 그렇지 않은지를 보다 명확하게 구분할 수 있어요.
> “테스트 프레임이 깔끔하고 편향 없이 잘 짜여 있으면, 캐주얼이나 하이브리드 캐주얼 게임도 리인게이지먼트 캠페인으로 신규 매출을 만들 수 있어요.”
### **Q. 리인게이지먼트 캠페인은 실제로 새로운 매출에 기여하나요?**
**A.** 많은 게임 마케터들이 리인게이지먼트 캠페인이 이미 돈을 쓰는 유저에게 또 돈을 쓰는 비효율적인 마케팅이라고 생각해서 걱정해요. 하지만 실제로는 그렇지 않아요. 데이터를 보면 리인게이지먼트 광고를 본 고가치 유저는 더 많은 소비를 발생시켜요. 기존 매출을 재분배하는 것이 아니라 해당 유저가 없었다면 발생하지 않았을 매출을 새롭게 만드는 것이죠.
### **Q. 리인게이지먼트와 리타겟팅 캠페인이 만들어내는 '진짜 인크리멘탈 가치'를 클라이언트에게 어떻게 설명하시나요?**
**A.** Adikteev는 캠페인이 실제 비즈니스에 어떤 효과가 있었는지 예로 들어 설명해요. 활성 유저의 경우, 리인게이지먼트를 통해 이탈하거나 지출을 줄이는 걸 방지할 수 있고 이탈 유저는 다시 게임으로 돌아오게 만들 수 있어요.
위와 같은 효과는 명확한 수치로 입증돼요. 인크리멘탈 수익, ARPU 상승, 그리고 광고 노출 그룹과 컨트롤 그룹간 행동 차이를 통해서 확인할 수 있죠. 클릭 수, 노출 수 같은 표면적인 지표가 아니라 실제 새로운 매출이 만들었는지를 확인하는 거예요.
[__wf_reserved_decorative]
그리고 전 과정을 투명하게 공개해요. 클라이언트가 테스트를 그대로 복사해서 직접 결과를 비교해 볼 수 있게 하죠. 인크리멘탈리티 측정은 에어브릿지와 같은 MMP(Mobile Measurement Partner)에서 받는 이벤트와 어트리뷰션 데이터를 기반으로 해요. 이 데이터는 클라이언트도 접근할 수 있기 때문에 누구나 같은 기준에서 확인할 수 있어요.
### **Q. 만약 제가 게임 마케터이고 인크리멘탈리티를 테스트해 보고 싶다면 어떻게 시작하면 좋을까요?**
**A.** 먼저 사전 분석 단계에서 데이터를 바탕으로 캠페인 적합 여부를 파악해요. 파트너십이 시작되면, 에어브릿지와 같은 MMP에서 이벤트 데이터, 어트리뷰션 데이터와 유저 행동 데이터 등을 받아와요. 이 데이터를 기반으로 해당 게임이 리인게이지먼트 캠페인에 적합한지를 판단하죠.
분석은 보통 12주 정도 걸리고 이 과정을 통해 어떤 유저 세그먼트를 타겟할지 예산은 어느정도가 적절할지 등을 제안하게 돼요. 금액의 경우 하루 500~1,000$ 수준으로 글로벌 캠페인을 설정하는 게 일반적이에요. 이렇게 시작한 캠페인을 점진적으로 확장할 수 있도록 설계해요.
인크리멘탈리티 테스트는 캠페인이 시작된 첫날부터 바로 측정에 들어가요. 이렇게 하면 실시간 성과와 함께 캠페인 성과를 투명하게 확인할 수 있을 뿐만 아니라 투입된 비용이 실제 성과를 얼마나 이어졌는지도 바로 파악할 수 있어요.
> “리인게이지먼트는 단순히 기존 매출을 나눠 쓰는 게 아니라, 추가 가치를 끌어내는 과정이에요.”
[__wf_reserved_decorative]
## **지금 바로 게임 마케터가 실행할 수 있는 리인게이지먼트 전략은 무엇일까요**
Adikteev는 2012년부터 리타겟팅을 너머, 실제로 측정 가능한 ROI(광고 수익률)를 만들어내는 전략을 꾸준히 진화시키고 있어요. 수십억 건에 달하는 게임 데이터를 기반으로 학습된 알고리즘으로 업계에서 독보적인 리타겟팅 기술력을 갖추게 됐죠. Taoufik Lazrak과 Adikteev 팀은 게임 리인게이지먼트와 리타겟팅 전략에 대한 다년간 쌓은 인사이트를 바탕으로 게임 마케팅 성공을 돕고 있어요.
### **Q. 게임 업계에서 꾸준하게 좋은 성과를 내는 전략이 있다면 무엇인가요?**
**A.** 가장 검증된 세 가지 전략은 바로 오디언스 세그멘테이션, 크리에이티브 A/B 테스트, 그리고 지오 타켓팅(위치기반 타겟 마케팅)이에요.
먼저, 오디언스 세그멘테이션은 데이터 기반으로 정교하게 오디언스를 세분화하는 게 핵심이에요. 단순히 현재 활동 중인 유저만 타겟팅 해서는 안 돼요. 이탈한 유저도 다시 불러오고, 아직 결제를 한 번도 안 한 유저도 전환시켜야 하죠. 이렇게 유저 세그멘트를 세밀하게 만들면 만들수록 캠페인 성과는 더 좋아져요.
두 번째 크리에이티브 A/B 테스트는 어떤 메시지가 유저에게 반응을 일으키는지를 알게 되는 데에 모든 목적이 있다고 해도 과언이 아니에요. 그래서 전면 광고, 리워드 영상, 배너 등 다양한 포맷을 실험해 보고, 리타겟팅에 맞춘 광고 위치와 메시지를 지속적으로 테스트해야 해요. 한 번에 끝나는 작업이 아닌 타겟 세그먼트 별로 실제 반응을 끌어내는 크리에이티브 요소를 찾기 위해 끊임없이 반복해야 하죠.
마지막으로, 캠페인을 최적화하려면 지역별 정밀한 접근이 중요해요. 처음에는 넓게 지오 타겟팅 설정한 뒤, 실제 LTV가 높은 지역을 빠르게 좁혀 나가는 방식으로 운영하는 게 핵심이에요.
### **Q. 예산이 잘못된 지역에 낭비될 위험이 있는데도, 게임 마케터에게 광범위한 지오 타겟팅이 여전히 효과적인 전략일까요?**
**A.** Adikteev는 개인화된 알고리즘을 활용해서 전환 가능성이 높은 유저를 예측해요. 그리고 전환 가능성이 낮은 유저는 자동으로 타겟팅 대상에서 제외하죠. 덕분에 성과가 낮은 지역에 예산을 쏟지 않아도 되고, 광고가 진짜 반응할만한 유저에게만 도달하게 돼요.
> “처음에는 넓게 지오 타겟팅 설정한 뒤, 실제 LTV가 높은 지역을 빠르게 좁혀 나가는 방식으로 운영하는 게 핵심이에요.”
## **하이브리드 수익 모델, 이제는 필수예요**
이전에는 게임 수익화를 크게 두 가지 방식으로 나눴어요. 바로 IAP(인앱 결제)와 IAA(인앱 광고)죠. 각 전략은 서로 다른 유저 그룹과 수익 구조를 겨냥하고 있었어요. 하지만 요즘은 많은 캐주얼 및 하이퍼 캐주얼 게임들은 두 방식을 결합한 하이브리드 수익화 모델을 도입하고 있어요. IAP와 IAA를 함께 활용해서 더 다양한 유저로부터 안정적인 수익을 확보하려는 것이 그 이유이죠.
[__wf_reserved_decorative]
### **Q. 단일 수익 모델에서 하이브리드 모델로 전환하는 게 정말 좋은 선택일까요? 게임 마케터는 이런 환경에서 어떤 지표에 집중해야 할까요?**
**A.** 특히 캐주얼이나 하이퍼 캐주얼 게임을 운영하고 있다면, 하이브리드 수익 모델은 꼭 고려해야 해요. 그리고 주요 지표로는 기존 ROAS(광고 수익률)보다 blended tROAS(목표 기반 tROAS)를 활용하는 걸 추천드려요.
기존 ROAS는 과거의 성과만을 기준으로 삼지만, tROAS는 미래 수익을 예측하고 실시간으로 최적화할 수 있도록 도와줘요. 목표 ROAS를 설정해두면 캠페인의 운영 목표, 예를 들어 수익 극대화나 일정 수익률 달성 등에 맞춰 광고를 자동으로 최적화할 수 있죠.
> “기존 ROAS 대신 blended tROAS를 핵심 지표로 써보세요. ROAS는 과거를 보는 지표지만, tROAS는 앞으로의 최적화를 위한 실시간 전략이에요.”
## **프라이버시 중심 환경에도 리인게이지먼트 & 리타겟팅 성과를 만들 수 있어요**
2021년 iOS 14.5부터 애플이 ATT 정책을 도입하면서, 개인정보 보호 규제가 훨씬 더 강화됐어요. 이제 iOS 유저는 자신의 데이터 사용 여부를 스스로 선택할 수 있게 됐고, 게임 마케터들은 이런 변화에 맞춰 리인게이지먼트 전략을 새롭게 설계해야 해요.
### **Q. 프라이버시가 강화된 시대에, iOS에서 효과적인 리인게이지먼트 캠페인을 운영하려면 어떻게 해야 하나요?**
**A.** ATT 이전에는 IDFA(Identifier for Advertisers, Apple iOS 기기의 고유한 식별자)를 기반으로 유저를 리타겟팅하는 게 기본 방식이었어요. 하지만 지금은 대부분의 유저가 게임 앱에서는 개인 정보 제공에 동의하지 않을 가능성이 높아요.
이럴 때 필요한 게 바로 확률적 어트리뷰션(probabilistic attribution)이에요. 이 방식은 다양한 플랫폼의 유저 데이터를 통합해, 동일 유저일 가능성을 계산하는 모델이에요. 예를 들어, 한 앱에서 유저가 데이터 공유에 동의했다면, 우리가 타겟팅하려는 다른 앱에서도 90% 이상 동일 인물일 확률이 있다고 보는 거죠. 실제로 지난 1년 동안 저희는 iOS 유저 타겟팅 기술을 크게 개선했어요. 일부 파트너 기준으로 리인게이지먼트 캠페인 도달률이 20%에서 70%까지 올라갔어요.
미국 기준으로는 게임 파트너의 유저 개인 정보 동의 동의율은 40~45% 수준인데, Adikteev 캠페인을 활용하면 확률적 어트리뷰션을 통해 타겟 유저의 60~70%까지 도달할 수 있어요. 이렇게 더 넓은 유저에게 닿을 수 있다면 매출에 긍정적인 영향이 되는 건 당연한 수순이죠.
> “확률적 어트리뷰션을 활용하면, 한 앱에서 데이터 공유에 동의한 유저가 다른 상황에서도 동일 인물일 확률을 90% 이상으로 추정할 수 있어요.”
## **꼭 기억해야 할 6가지 핵심**
Adikteev의 Taoufik Lazrak 님과 함께 진행한 이번 인터뷰에서 게임 리인게이지먼트 및 리타겟팅 캠페인을 성공적으로 운영하기 위해 기억해두면 좋을 여섯 가지 핵심 포인트를 정리해 봤어요.
**1\. 오해에서 벗어나기
**인크리멘탈리티 테스트, 뷰스루 지표, 리인게이지먼트 캠페인은 캐주얼이나 하이퍼 캐주얼처럼 유저 생애 주기가 길지 않은 게임에서도 충분히 효과를 낼 수 있어요.
**2\. 인크리멘탈리티
**좋은 가설과 편향 없는 테스트 환경을 갖춰진다면, 리인게이지먼트 캠페인은 기존 매출을 나눠 갖는 게 아닌 실제 추가 매출을 만들어낼 수 있는 중요한 마케팅 액션이에요.
**3\. 대표적인 전략 실행하기
**오디언스 세그멘테이션, 크리에이티브 A/B 테스트, 그리고 지오 타겟팅(위치 기반 타겟팅)을 전략적으로 활용해 보세요.
**4\. 하이브리드 수익화**
캐주얼 및 하이퍼 캐주얼 게임이라면, 하이브리드 수익화 모델로 전환하는 걸 고려해 보세요. 핵심 지표는 blended tROAS예요.
**5\. 확률 기반 어트리뷰션을 활용한 리타겟팅
**개인정보 보호 정책 변화에도 확률 기반 어트리뷰션을 활용하면 iOS에서도 최대 20~30% 더 많은 실제 유저에게 도달할 수 있어요.
## **에어브릿지와 함께 게임 성장을 가속화해보세요**
게임 마케팅을 한 단계 끌어올리고 더 높은 성장을 이뤄내고 싶다면 MMP 도입은 선택이 아닌 필수에요. 에어브릿지에서는 게임 마케팅 전반을 분석하고 빠른 성장을 실현할 수 있어요.
- **고급 Cohort LTV 분석**: 키워드, 크리에이티브, 국가, OS 등 다양한 기준으로 유저의 LTV를 정밀하게 분석해요.****
- **광고 채널 연동:** Google AdMob, AppLovin MAX, Unity LevelPlay 등 여러 광고 채널의 데이터를 통합 분석할 수 있어요.
- **예측 LTV(pLTV)**: 미래 매출을 실시간으로 예측해 더 빠르고 데이터 중심의 의사결정을 할 수 있어요. ****
- **분 단위/시간 단위 리텐션 분석**: 유저 행동을 ‘분 단위’로 분석해 실시간 인사이트를 확보할 수 있어요.****
- **투명하고 합리적인 가격 정책**: 계약 규모와 상관없이 엔터프라이즈 수준의 기능을 애드온(추가 과금)없이 제공해요.
인디 게임 스튜디오부터 글로벌 퍼블리셔까지 에어브릿지 게임 마케팅 시장에서 앞서 나가기 위한 핵심 솔루션이에요. 지금 바로 [airbridge.io/ko](https://airbridge.io)를 방문해 보다 자세히 솔루션을 자세히 알아보세요.
## **Adikteev와 함께 성공적인 리인게이지먼트 마케팅을 실현하세요**
게임 리인게이지먼트와 리타겟팅 성공으로 이끄는 전략과 인사이트가 궁금하다면, [adikteev.com](https://www.adikteev.com)을 방문하면 실무에 바로 적용할 수 있는 팁과 실제 성공 사례들을 확인할 수 있어요.
---
## 브랜치(Branch)만으로 아쉽다면? 퍼포먼스 마케터를 위한 MMP 5가지 비교 가이드
URL: https://www.airbridge.io/ko/blog/the-5-best-branch-alternatives-for-2026-a-performance-marketers-guide
Language: Korean
Published: 2025-05-01
Author: Jaehyuk Kim
**브랜치(Branch)**는 딥링크 중심 서비스로 MMP 시장에서 강력한 입지를 확보해온 브랜드예요. 다만 최근 비용이 상승하고 로데이터 접근이 제한되면서, 퍼포먼스 마케터들 사이에서 대안을 검토하려는 움직임이 커지고 있어요. 아마 여러분도 브랜치 외에 다른 솔루션이 있는지 고민해보신 적 있으실 텐데요.
딥링크 기반으로 어트리뷰션을 운영하고 있었는데 대안 솔루션을 찾고 있다면, 오늘 드리는 가이드가 실질적인 의사결정에 도움이 될 거예요. 브랜치의 강점은 어디까지인지, 그리고 **에어브릿지(Airbridge)**를 포함한 다른 솔루션과 비교했을 때 어떤 선택지가 현실적인지 정리해봤어요.
**📌핵심 요약**
- **왜 전환을 고민할까요? :** 많은 퍼포먼스 팀이 로데이터에 직접 접근하고, 멀티터치 분석 구조를 더 투명하게 운영하기 위해 다른 대안을 찾고 있어요.
- **프라이버시 환경에서도 제대로 측정할 수 있을까요? :** 최근 솔루션들은 SKAN 4.0이나 Privacy Sandbox를 전제로 설계돼 있어서, 데이터가 제한된 환경에서도 비교적 안정적으로 성과를 측정할 수 있어요.
- **광고사기는 얼마나 투명하게 관리할 수 있을까요? :** 좋은 솔루션은 “왜 이 클릭이 무효 처리됐는지”를 설명할 수 있어요. 단순 차단이 아니라, 기준과 로직이 보이는 구조예요.
- **비용은 합리적인가요? :** 요즘 플랫폼들은 측정 깊이나 활용 범위에 맞춰 비용을 조정할 수 있어서 비용 절감 효과를 보는 기업들도 있어요.
- **에어브릿지는 어떤 솔루션인가요? :** 딥링크 안정성은 유지하면서도, 데이터 활용과 분석까지 필요한 팀에게는 균형 잡힌 솔루션이에요. 딥링크는 이미 잘 쓰고 있는데 어트리뷰션이나 데이터 활용이 아쉬운 팀, 라우팅만으로는 부족하지만 엔터프라이즈급 복잡성까지는 필요 없는 기업에 잘 맞아요.
## 브랜치는 실제로 어떤 역할을 하는 MMP 솔루션인가요?
**브랜치는 원래 딥링크, 디퍼드 딥링크, 그리고 크로스 플랫폼 유저 연결에 강점을 가진 MMP예요.** 어트리뷰션은 이 기능 위에 덧붙여진 형태에 더 가까워요. 그래서 실제 도입 흐름을 보면, 처음에는 딥링크 문제를 해결하려고 도입했다가 이후에 점점 측정이나 분석까지 확장해서 쓰는 경우가 많아요.
## 마케터들이 Branch에서 다른 MMP로 전환하는 이유
브랜치 자체의 문제가 있다기보다, 마케팅 현장에서의 니즈가 바뀌면서 생기는 간극에 가까워요. 예전에는 **유저를 정확한 화면으로 보내는 것이 중요**했다면, 지금은 **그 유저가 실제로 매출에 얼마나 기여했는지**까지 봐야 해요.
이 과정에서 아래 같은 고민이 자연스럽게 나오게 돼요.
- 비용 대비 얻는 데이터가 충분한가?
- 로데이터를 자유롭게 활용할 수 있는가?
- 어트리뷰션 로직을 우리가 원하는 방식으로 조정할 수 있는가?
이 질문들에 대한 답이 부족하다고 느끼는 순간, 마케터들은 대안을 찾기 시작하게 돼요.
### 1) 마케팅 실무에서 자주 나오는 문제가 뭘까요?
마케팅 실무에서 나오는 문제의 중심은 딥링크가 아니라 바로 “측정”이에요. 현장에서 자주 나오는 이야기를 정리해 볼게요.
**(1) 딥링크는 강한데, 어트리뷰션은 아쉬워요**
- 고급 어트리뷰션 모델을 적용하거나 커스터마이즈하기가 쉽지 않아요
- 멀티터치나 어시스트 기여도를 세밀하게 조정하기 어려워요
**(2) 광고사기를 적극적으로 관리하기 어려워요**
- 설정 가능한 규칙이 제한적이에요
- 의심스러운 설치나 행동을 걸러내는 데 한계가 있어요
**(3) 데이터를 깊게 보기가 어려워요**
- 대시보드 중심이라 로데이터 접근이 제한돼요
- BI나 데이터 웨어하우스로 연결하려면 추가 작업이 필요해요
**(4) 비용이 생각보다 빠르게 올라가요**
- 사용량 기반 과금 구조라서 데이터 활용 범위보다 비용이 더 빨리 증가하는 경우가 있어요
### 2) 시장 트렌드 자체가 바뀌고 있어요
[요즘 시장 트렌드](https://swaarm.com/blog/top-trends-mobile-measurement-partners-mmp-2025/)를 보면, 이건 특정 솔루션의 문제가 아니라, 환경 자체가 완전히 달라진 영향도 있어요. ATT 이후로 MMP 시장은 명확하게 **프라이버시 중심 측정 방향으로 이동**하고 있어요. 특히 구독 서비스나 AI 앱을 운영하는 팀은 단순 딥링크가 아니라 아래와 같은 솔루션을 찾고 있어요.
- 장기적인 매출 흐름을 볼 수 있어야 하고
- 데이터 활용까지 가능한 구조여야 하고
- 동시에 비용도 관리 가능해야 해요
또 하나 중요한 변화가 있어요. 이제 데이터 팀은 **툴 안에서 데이터 보는 것에 만족하지 않아요.** SQL로 직접 분석하고, 웨어하우스에 적재하고, 실험까지 연결하는 게 기본인데요. 이런 흐름이다 보니 폐쇄적인 구조로 설계된 솔루션들에 점점 부담을 느끼고 있어요.
#### (1) 이제는 프라이버시 대응이 기본이에요
ATT, SKAdNetwork, Android 정책 변화 이후에는 프라이버시 보호 기능이 필수가 됐어요.
데이터가 일부만 있어도 전환을 추정할 수 있어야 하고, 네트워크별 데이터를 검증할 수 있어야 하고, 여러 소스를 하나로 맞출 수 있어야 하죠.
즉, 단순히 링크 잘 연결되는 것만으로는 부족하고 애초에 어트리뷰션 중심으로 설계된 구조가 필요해졌어요.
#### (2) 구독 서비스는 리드 타임이 길어요
**구독이나 AI 서비스는 설치 → 결제가 바로 이어지지 않아요.** 며칠 뒤, 혹은 몇 주 뒤에 매출이 발생하는 경우가 많아요. 그래서 코호트 기반으로 LTV를 추적하고, 전환 기준을 유연하게 설정하고, 리인게이지먼트까지 안정적으로 측정하는 게 중요해졌어요.
딥링크는 어디로 보낼지는 잘 해결해주지만, 왜 매출이 발생했는지까지는 설명해주지 못해요. 이게 딥링크의 한계예요.
#### (3) 데이터는 이제 기업의 내부 자산이에요
점점 더 많은 조직이 어트리뷰션 데이터를 중요한 내부 자산으로 관리하려고 하고 있어요. 데이터 웨어하우스에 넣고, BI에서 분석하고, 실험과 연결하는 흐름을 구축하고 있죠. 이러한 기준에서 보면 대시보드 중심 구조는 점점 한계가 명확하다고 할 수 있어요.
## 딥링크 vs 풀퍼널 어트리뷰션 솔루션: ROAS에 미치는 차이는 얼마나 될까?
브랜치를 대신할 솔루션을 검토하다 보면 딥링크와 풀퍼널 어트리뷰션을 같은 개념처럼 이해하는 경우가 많아요. 하지만 자세히 들여다 보면 두 기능은 역할과 목적이 명확하게 달라요.
둘의 차이를 쉽게 설명하면, 딥링크는 사용자를 원하는 위치로 정확하게 보내는 기술이고, **풀퍼널 어트리뷰션 솔루션은 그 사용자가 어떤 경로로 들어와서, 어떤 행동을 했고, 실제로 매출에 얼마나 기여했는지를 계산하는 솔루션이에요.** 설치, 이벤트, 구매 등 전체 사용자 여정을 기준으로 각 터치포인트의 기여도를 분석하고 ROAS를 계산하죠.
**딥링크가 필요한 상황**
- 웹에서 앱으로 자연스럽게 이동시키고 싶을 때
- 이메일이나 QR에서 특정 화면으로 보내고 싶을 때
- 캠페인별로 다른 랜딩 경험을 만들고 싶을 때
- 신규 유저 유입을 유도할 때
**풀퍼널 어트리뷰션 툴이 필요한 상황**
- 어떤 채널이 설치를 만들었는지 알고 싶을 때
- 설치 이후 행동까지 추적하고 싶을 때
- 전환 기준과 기간을 직접 설정하고 싶을 때
- SKAN이나 네트워크 데이터를 함께 보고 싶을 때
- 이상 트래픽을 걸러내고 싶을 때
### 1) 왜 둘을 헷갈릴까요?
생각보다 많은 분이 딥링크 중심 솔루션을 도입해놓고, 어트리뷰션까지 완벽하게 분석해주기를 기대해요. 특히 C레벨 의사결정 단계에서 브랜치를 고를 때 이런 상황이 자주 발생해요.
딥링크는 분명 리텐션이나 사용자 경험에는 강점이 있어요. **하지만 iOS 프라이버시 환경에서 캠페인 성과를 정확하게 측정하는 역할은 결국 어트리뷰션 솔루션이 더 정확해요.**
물론 실제 실무에서는 두 가지 모두 다 필요해요. [Promodo에 따르면](https://www.promodo.com/blog/the-relationship-between-user-experience-ux-design-and-conversion-rates) 딥링크는 전환율을 15~25% 높이고, 어트리뷰션 솔루션은 광고사기나 멀티터치 낭비를 잘 잡아내요.
문제는 둘 중 하나에만 집중했을 때예요. 딥링크만 잘 되어 있고 어트리뷰션이 비어 있으면, 실제보다 성과가 좋아 보이거나 반대로 CPA가 불필요하게 높아지는 일이 생겨요. 딥링크에만 집중한 브랜드가 [어트리뷰션 설정이 제대로 되어 있지 않아 CPA가 30%까지 부풀려진 사례](https://yodelmobile.com/choosing-mobile-attribution-platform-for-your-app/)도 적지 않아요.
### 딥링크 솔루션 vs 풀퍼널 어트리뷰션 MMP
| 영역 | 딥링크 중심 솔루션 | 풀퍼널 어트리뷰션 MMP |
| --- | --- | --- |
| 핵심 포인트 | 유저 라우팅 | 측정 정확도 |
| 설치 로직 | 기본, 결정적 방식 | 결정적 + 모델링 |
| 설치 후 분석 | 제한적 | 코호트, LTV, ROAS 분석 |
| SKAdNetwork 대응 | 표면적 수준 | 프라이버시 중심 설계 |
| 광고사기 방지 | 최소 수준 | 규칙 기반, 설정 가능 |
| 데이터 접근 | 대시보드 중심 | 로데이터 + 내보내기 |
| 주요 활동 | 네비게이션·UX | 그로스, 성과 분석, 비용 최적화 |
## 2026년 기준, 현실적으로 실무자들이 많이 비교하는 브랜치 대안 솔루션
브랜치 대안 솔루션들을 리서치하고 검토하다 보면 자연스럽게 같이 보게 되는 솔루션들이 있는데요. 대표적으로 **앱스플라이어(AppsFlyer), 애드저스트(Adjust), 에어브릿지(Airbridge), 싱귤러(Singular)** 등의 제품들이 있어요.
각 솔루션은 딥링크, 어트리뷰션, 비용 구조에서 조금씩 다른 강점을 가지고 있어요.
- **앱스플라이어**는 광고사기 탐지와 글로벌 스케일에서 강점이 있지만, 비용이 높은 편이에요
- **애드저스트**는 비교적 균형 잡힌 가격대에서 멀티터치 어트리뷰션이 강점이에요
- **에어브릿지**는 로데이터 접근성과 데이터 일관성 측면에서 강점을 보여요
- **싱귤러**는 여러 데이터 소스를 통합해서 보는 데 강점이 있어요
### 브랜치 vs 주요 대안 솔루션 비교
| 항목 | Branch | AppsFlyer | Adjust | Singular | Airbridge |
| --- | --- | --- | --- | --- | --- |
| 핵심 강점 | 안정적인 딥링크 중심 기능 | 대규모 어트리뷰션 및 글로벌 표준 | 미디어 최적화 중심 기능 | 다양한 데이터 소스 통합 리포팅 | 딥링크와 데이터 활용을 함께 고려한 구조 |
| 딥링크 품질 | 매우 우수한 편 | 우수한 편 | 우수한 편 | 지원은 가능하나 핵심 기능은 아님 | 매우 우수한 수준 |
| 디퍼드 딥링크 | 네이티브로 안정적으로 지원 | 지원 | 지원 | 일부 제한적으로 지원 | 네이티브로 안정적으로 지원 |
| 어트리뷰션 유연성 | 일부 제한적인 편 | 높은 수준의 설정 가능 | 높은 수준의 설정 가능 | 중간 수준 (환경에 따라 상이) | 높은 수준의 설정 및 확장 가능 |
| 광고사기(프로드) 대응 | 기본적인 기능 제공 | 고도화된 규칙 기반 대응 | 옵션 기반 기능 제공 | 기본적인 기능 제공 | 규칙 기반 설정 및 투명한 운영 가능 |
| 코호트 / LTV 분석 | 일부 제한적인 편 | 고도화된 분석 가능 | 고도화된 분석 가능 | 분석 기능 제공 | 코호트 및 매출 중심 분석에 최적화 |
| 로데이터 접근 방식 | 대시보드 중심 구조 | 로데이터 제공 (Export 기반) | 로데이터 제공 | 로데이터 제공 | 로데이터 기반 직접 활용 구조 |
| 데이터 활용 확장성 | 제한적인 편 | 외부 시스템 연동 가능 | 외부 시스템 연동 가능 | 데이터 통합 중심 활용 | BI·웨어하우스 중심 확장에 최적화 |
## 에어브릿지는 브랜치와 어떤 차별점이 있나요?
[에어브릿지](https://abr.ge/van8ny)는 딥링크 안정성과 어트리뷰션 깊이를 동시에 잡도록 설계된 MMP예요. 딥링크와 어트리뷰션을 따로 보지 않고, 둘을 동시에 충족시키는 방향으로 설계됐죠.
그래서 에어브릿지를 도입하면 둘 중 하나를 포기해야 하는 상황이 생기지 않아요. 특히 **딥링크는 이미 잘 쓰고 있는데 어트리뷰션이나 데이터 활용이 아쉬운 팀, 라우팅만으로는 부족하지만 엔터프라이즈급 복잡성까지는 필요 없는 기업**에 잘 맞아요.
### 1) 리스크 없이 에어브릿지로 전환할 수 있어요
솔루션을 바꿀 때 가장 걱정되는 건 지금 잘 돌아가는 시스템이 깨지는 것일 텐데요. 에어브릿지는 브랜치와 동일한 기능들이 많아 이러한 불안 요소를 상당 부분 줄여줘요.
- 웹 → 앱, 앱 → 앱 딥링크를 안정적으로 처리해요
- 캠페인 단위 링크를 끊김 없이 관리할 수 있어요
- iOS, Android, 웹 흐름을 하나의 구조로 연결해요
즉, 브랜치의 핵심 기능인 링크 안정성을 그대로 유지하면서 전환할 수 있어요.
### 2) 실제 차이는 ‘클릭 이후’에서 발생해요
겉으로는 비슷해 보이지만 에어브릿지와 브랜치의 차이는 클릭 이후 데이터에서 확실하게 드러나요.
#### (1) 프라이버시 환경을 전제로 설계된 어트리뷰션
에어브릿지는 처음부터 ATT나 SKAdNetwork를 고려하여 설계된 솔루션이에요.
- SKAN 리포팅과 검증이 기본으로 제공돼요
- 시그널이 부족한 상황에서도 동작하는 모델링 구조를 갖고 있어요
- 유료, 오가닉, 온드 채널을 하나의 로직으로 연결해요
#### (2) 구독 시스템에 맞는 측정 방식
구독이나 AI 서비스는 설치보다 “그 이후의 유저 행동”이 더 중요해요. 그리고 에어브릿지는 이 흐름을 기준으로 유저 행동을 파악하도록 설계돼 있어요. 딥링크만으로는 설명이 안 되는 영역을 여기서 보완하는거죠.
- 코호트 기반 LTV 분석이 가능해요
- 전환 기준을 유연하게 설정할 수 있어요
- 매출 이벤트 기준으로 리어트리뷰션과 리인게이지먼트를 추적해요
#### (3) 투명한 광고사기 방지 기능
에어브릿지는 불투명하고 측정할 수 없는 신뢰 점수가 아니라, **규칙을 직접 설정하고 결과를 투명하게 확인**할 수 있어요
- 우리 서비스에 맞게 무효 트래픽 기준을 직접 설정할 수 있어요
- 설치와 이후 이벤트까지 검증할 수 있어요
- 로직이 공개된 상태에서 광고비를 줄일 수 있어요
#### (4) 자유롭고 주도적으로 활용할 수 있는 데이터 소유권
브랜치에서 자주 나오는 불만이 “데이터가 대시보드 안에 갇혀 있는 것 같다”는 점이에요. **즉, 대시보드에 대한 의존성이 너무 크다는 것이죠.**
반면 에어브릿지는 데이터에 대한 접근 방식이 완전히 달라요.
- 로데이터를 그대로 내보낼 수 있어요
- 웨어하우스나 BI와 자연스럽게 연결돼요
- 대시보드는 확인용이고, 분석은 밖에서 하도록 설계돼 있어요
실제로 내부에서 숫자를 검증할 때는 차트보다 로데이터가 훨씬 중요해요. 이런 점에서 에어브릿지가 브랜치보다 훨씬 더 자유롭고 주도적으로 데이터를 활용할 수 있다고 할 수 있어요.
### 3) 비용 구조에서 체감 차이가 나요
많은 팀이 처음 솔루션을 도입할 때 비용이 얼마나 되는지를 보지만, \*\*실제로 중요히 봐야하는 것은 “비용이 얼마까지 늘어나느냐”\*\*예요. 그래서 많은 분이 브랜치 가격이나 비용을 검색하기도 하죠.
에어브릿지는 이 부분에서 예측 가능성이 높은 구조예요.
- 기능별로 추가 과금이 붙지 않아요
- 트래픽이나 이벤트가 늘어도 비용이 급격히 튀지 않아요
- 링크 수가 아니라 “측정 가치” 기준으로 가격이 설정돼요
그래서 스케일이 커질수록 비용 차이가 더 크게 느껴져요.
## 이제는 ‘링크를 잘 보내는 것’만으로는 부족해요
브랜치는 여전히 좋은 딥링크 솔루션이에요. 특히 유저를 정확한 위치로 보내는 기능이 필요할 때는 충분히 역할을 해요. 하지만 규모가 커질수록 상황이 달라져요. 광고비는 계속 늘어나고, 프라이버시에 대한 규제는 마케팅 성과의 가시성을 제한하고, 데이터는 점점 덜 보이게 돼요.
이때 필요한 건 단순한 라우팅이 아니라 아래 3가지예요.
1. 신뢰할 수 있는 어트리뷰션
2. 시간 차를 반영한 매출 분석
3. 필요할 때 직접 검증할 수 있는 로데이터
많은 마케팅 팀이 브랜치의 대안 솔루션을 찾는 이유가 여기에 있어요.
**만약 여러분이 에어브릿지에 관심이 있다면, 아래에 해당하는 팀인지 한번 살펴보세요.**
- 딥링크는 그대로 유지하고 싶고
- ATT, SKAN 환경에서도 측정을 이어가야 하고
- 구독/AI 기반 매출 구조를 분석해야 하고
- 데이터는 내부에서 직접 활용하고 싶은 팀
이 조건에 해당한다면, 에어브릿지 도입을 적극 검토해 보세요. 브랜치로 초기 성장을 만들었다면, 이제 에어브릿지로 그 다음 단계로 나아갈 차례예요.
더 나은 어트리뷰션과 데이터 분석, 그리고 딥링크를 에어브릿지로 구현한 사례들을 확인해 보세요.
### 👉 **에어브릿지 케이스 스터디**
- [원티드랩, 에어브릿지 이관 후 웹투앱 유저 여정을 트래킹하고 CPA를 38% 절감했어요](https://www.airbridge.io/ko/case-studies/wantedlab-web-to-app-attribution)
- [알라미, 글로벌 1위 알람 앱이 에어브릿지를 활용해 ROAS를 개선했어요](https://www.airbridge.io/ko/case-studies/alarmy-roas-measurement)
- [다방이 Airbridge로 딥링크 관리 시간을 90% 단축한 방법](https://abr.ge/v0fx0j)
---
## 국내 마케팅 연동 압도적 1위: 에어브릿지와 시작하세요
URL: https://www.airbridge.io/ko/blog/korea-no-1-mmp
Language: Korean
Published: 2025-04-23
Author: Team Airbridge
네이버, 토스, 당근, 카카오, 에브리타임, 그리고 하이브까지. 이들 플랫폼의 월간 활성 사용자 수(MAU)를 모두 합치면 1억 명을 훌쩍 넘어요. 중복 사용자를 감안하더라도, 국내 디지털 사용자 대부분이 이 여섯 채널 안에 있다고 해도 과언이 아니에요.
에어브릿지는 **국내 매체의 구조와 정책 변화에 민감하게 반응하며, 누구보다 빠르게 대응해왔어요.** 국내에서 가장 많은 광고주와 함께하다 보니 새로운 기능이나 매체가 나오면 피드백을 가장 먼저 받고, 실제 연동까지도 자연스럽게 빠르게 이어져요. 또한 본사와 개발팀이 한국에 있어, **실시간 대응은 물론 국내 환경에 맞춘 기능 설계까지 가능하다는 점**이 에어브릿지만의 강력한 차별점이에요.
국내 마케팅을 제대로 하려면 국내 채널의 특성과 환경을 깊이 이해하고, 정교하게 연동되어 있는 솔루션이 필요해요. 지금부터, 왜 에어브릿지가 국내 마케팅에 가장 잘 맞는 솔루션인지 함께 살펴볼게요.
## 네이버, 당근, 카카오에 오디언스를 자동으로 보내고 싶다면
개인정보 보호 규제가 강화되면서, 마케터는 더 이상 서드파티 데이터에만 의존할 수 없게 되었어요. 광고 타겟팅의 정밀도와 효율을 유지하려면, **직접 수집한 퍼스트 파티 데이터를 얼마나 잘 활용하느냐**가 관건이에요. 하지만 아직도 리타게팅 오디언스를 매일 추출하고, 수동으로 정리해 채널마다 따로 업로드하고 계신가요? AI가 광고 소재까지 자동으로 만드는 시대에, 이런 반복적인 수작업은 마케팅 속도를 늦추는 원인이 될 수 있어요.
에어브릿지는 이 과정을 완전히 자동화해요. **에어브릿지 SDK 하나만으로, 네이버·당근·카카오에 오디언스를 실시간으로 연동할 수 있어요.** 앱 설치, 회원가입, 장바구니 담기, 구매 완료 등 원하는 조건을 설정하면, 그에 맞는 세그먼트가 자동 생성되고 바로 매체에 전송돼요.
또한 에어브릿지는 **‘네이버’와 ‘카카오’의 검색 광고(SA)뿐 아니라, 디스플레이 광고(DA)까지 비용 연동을 모두 지원하는 국내 유일의 MMP예요.** 다른 솔루션들이 검색 광고에만 한정된 연동을 제공하는 반면, 에어브릿지는 노출부터 클릭, 광고비까지 모든 지표를 채널 유형에 관계없이 하나의 화면에서 통합 관리할 수 있어요.
이제는 더 빠르고, 더 정확하게 타겟팅할 수 있어요. 국내 주요 매체의 방대한 광고 인프라와 에어브릿지의 실시간 오디언스 싱크를 결합하면, 단순한 자동화를 넘어 전환율과 운영 효율성까지 함께 높일 수 있어요.
> 네이버 오디언스를 자동으로 연동하고 싶다면 → [네이버 오디언스 싱크 연동 방식 자세히 알아보기
> ](https://www.airbridge.io/ko/blog/naver-ads-audience-sync)당근 세그먼트를 수동으로 추출하여 업로드 하고 있다면 → [당근 오디언스 자동화 방식 확인하기
> ](https://www.airbridge.io/ko/blog/karrot-audience-sync)카카오 리타게팅 오디언스를 보다 정밀하게 운영하고 싶다면 → [카카오 오디언스 연동 방식 자세히 보기](https://www.airbridge.io/ko/blog/kakao-ads-audience-integration)
%%CTA|에어브릿지 전문가와 상담하기|/ko/book-a-demo|네이버, 당근, 카카오 등 국내 주요 매체의 퍼포먼스 마케팅 사이클을 제대로 가동하고 싶다면, 에어브릿지 팀에 문의해 보세요.%%
## 토스애즈, 에브리타임 광고 캠페인을 더 정밀하게 운영하고 싶다면
**월평균 290만 명의 대학생이 사용하는 에브리타임은, 대학생 타겟 캠페인에 가장 효과적인 채널 중 하나예요.** 에어브릿지는 에브리타임과 공식 연동된 유일한 MMP로, 클릭부터 설치, 회원가입, 구매까지 전체 퍼널을 정밀하게 추적할 수 있어요. Z세대를 대상으로 한 브랜드 캠페인이라면 특히 더 큰 차이를 만들 수 있어요.
**토스애즈는 광고 슬롯부터 유저 액션까지 고도화된 타겟팅이 가능한 플랫폼이에요.**
광고 배너, 머니 알림, 행운 퀴즈, 라이브 쇼핑 보기 등 다양한 접점을 통해 유저와 연결되며, 광범위한 사용자 기반 데이터를 바탕으로 정교한 데모그래픽 타겟팅도 가능해요.
에어브릿지와 연동하면 이 모든 광고 캠페인을 **더욱 빠르게 최적화할 수 있고, 성과 기반 운영과 ROAS 개선**까지 기대할 수 있어요.
> 다양한 광고 슬롯에서 유저를 만나는 토스, 제대로 측정하고 있는지 궁금하다면 → [토스애즈 연동 방식 자세히 보기](https://www.airbridge.io/ko/blog/airbridge-toss-ads-integration)
## 하이브 플랫폼에서 게임 마케팅을 더 효율적으로 관리하고 싶다면
**컴투스플랫폼의 하이브는 게임 개발과 운영에 특화된 플랫폼으로, 퍼포먼스 마케팅이 핵심인 환경이에요.** 유저 확보와 전환, 수익화까지 전 과정이 수치로 바로 드러나는 만큼, 실시간 데이터와 연동의 정밀도가 매우 중요해요.
에어브릿지는 하이브 SDK v4.24.4.0 이상(모바일) 또는 v24.4.0 이상(PC) 환경에서 간단히 연동할 수 있어요. 이전 버전을 사용 중이더라도 최신 SDK로 업데이트만 하면, 별도 설치 없이 바로 활성화할 수 있어요.
특히 하이브 애널리틱스와 에어브릿지는 **어트리뷰션 데이터를 실시간으로 동기화해요.** 이를 통해 유저 획득, 리텐션, 수익화까지 마케팅 퍼널 전반을 더 정교하게 최적화할 수 있어요.
> 게임 유저 유입부터 수익화까지 한눈에 보고 싶다면 → [하이브 플랫폼 연동 구조 자세히 보기](https://www.airbridge.io/ko/blog/hive-platform-integration)
## 국내 마케팅을 제대로 하고 싶다면
많은 MMP가 “연동된다”고 말해요. 하지만 마케터는 잘 알죠. 단순히 연동되는 것과, 실무에서 ‘쓸 수 있게’ 잘 연동되는 건 다르다는 걸요. 에어브릿지는 가장 많은 국내 고객사를 보유하고 있기에 한국 광고주의 니즈를 누구보다 잘 이해하고 있어요. 광고 성과를 제대로 보기 위해 필요한 연동이 무엇인지, 어떻게 하면 실무자가 더 빠르고 정확하게 사용할 수 있을지를 먼저 고민해요.
**국내 채널 마케팅, 이제는 진짜 ‘잘 되는’ 연동으로 운영해보세요.** 지금 쓰는 MMP와 무엇이 다른지 직접 확인해보셔도 좋아요. 지금 AI 마케팅을 도입하고 체계적으로 활용하는 브랜드만이, 다가올 새로운 시장 질서 속에서 빠르게 도약할 수 있을 거예요. 에어브릿지와 함께라면, AI를 활용한 광고 성과 최적화부터 데이터 기반 미래 예측까지 한층 더 빠르게 완성할 수 있어요. 에어브릿지가 우리 마케팅을 어떻게 도와줄 수 있을지 궁금하다면, 지금 바로 확인해 보세요!
---
## 마케터를 위한 딥링크 완벽 가이드: 개념부터 실전 활용까지
URL: https://www.airbridge.io/ko/blog/deeplink-guide-for-marketers
Language: Korean
Published: 2025-04-22
앱을 설치한 뒤, 메인 화면이 아닌 특정 이벤트 페이지로 바로 이동한 경험 있으신가요? 혹은 메신저로 받은 할인 쿠폰을 적용하려고 앱을 열었더니, 별다른 입력 없이 쿠폰이 자동으로 적용된 적은요? 이런 매끄러운 사용자 경험 뒤에는 ‘딥링크’라는 기술이 숨어 있어요.
이제 딥링크는 선택이 아닌, 대부분의 앱 마케터가 당연히 사용하는 기본 전략이 되었어요. 하지만 다들 쓴다고는 하는데, **정작 왜 중요한지, 어떻게 활용해야 실무에서 진짜 효과를 볼 수 있는지**는 여전히 어렵게 느껴질 수 있어요. 이번 글에서는 딥링크의 개념부터 실전 활용 방법까지, 마케터의 입장에서 꼭 필요한 내용을 알기 쉽게 정리해 드릴게요.
## 딥링크(Deep Link)란?
딥링크란 유저가 웹페이지 또는 다른 앱에서 링크를 클릭하면 앱을 실행하고 특정 페이지로 보내주는 기술이에요. 만일 유저에게 아직 앱이 없다면, 앱스토어로 이동시킨 뒤 다운로드 이후에 특정 페이지로 보내주는 ‘디퍼드 딥링크’ 기능도 있어요. 많이들 이야기하는 유니버설 링크(iOS)와 앱 링크(Android)도 딥링크의 한 종류예요.
딥링크가 무엇인지 더욱 자세히 알고 싶다면 이 [블로그](https://www.airbridge.io/ko/blog/what-is-deep-link)에서 확인해 보세요!
## 딥링크가 왜 중요할까요?
딥링크가 왜 중요한지, 실제 상황을 하나 예로 들어 설명해 볼게요.
### 👟 유저 입장: 중간에 끊긴 경험
인스타그램을 보다가 마음에 쏙 드는 운동화 광고를 발견했어요. 할인 쿠폰도 준다고 해서 광고를 클릭하고 앱을 설치했죠. 그런데 설치가 끝난 후 앱을 열어보니 처음 봤던 그 운동화는 보이지 않고 메인 화면이 나왔어요. 겨우 검색해서 제품을 찾아 장바구니에 담았지만, 이번엔 쿠폰이 어디 갔는지 찾을 수 없었죠. 결국 사용자는 구매를 포기했어요.
이런 상황, 생각보다 자주 일어나요. 유저 입장에서는 기대했던 경험이 중간에 끊겨버린 거예요. 앱을 설치하긴 했지만, 그 이후 과정이 너무 번거롭고 복잡했기 때문에 이탈해버린 거죠.
### 📉 마케터 입장: 연결되지 못한 성과
이번엔 같은 상황을 마케터 입장에서 볼게요. 광고 클릭이라는 목표는 달성했지만, 정작 구매까지 이어지지는 않았어요. 그 과정에서 쿠폰도 제대로 전달되지 않았고, 어디서 유저가 이탈했는지도 알 수 없었죠. 결국 광고비는 들였지만 얼마나 효과가 있었는지 확인하기 어렵고, 데이터도 불완전하게 남게 되었어요.
유입은 있었지만 전환이 일어나지 않았고, 그 원인을 명확하게 파악하기도 힘든 상황. 마케터로서는 가장 안타까운 결과예요.
### 🎯 딥링크가 해결하는 것들
딥링크는 이런 문제를 자연스럽게 해결해 줘요. 앱이 없는 유저라도 광고 클릭 한 번이면 앱 설치까지 진행할 수 있고, 설치가 끝난 뒤에도 메인 화면이 아니라 광고에서 봤던 캠페인 페이지나 상품 상세 페이지로 곧장 이동할 수 있어요. 쿠폰도 자동으로 적용할 수 있고요.
덕분에 유저는 처음 기대했던 그대로의 경험을 끊김 없이 이어갈 수 있고, 마케터는 유저 여정을 정확히 추적하면서 전환 데이터를 세밀하게 분석할 수 있어요. 전환율과 리텐션도 자연스럽게 좋아지는 것은 당연하죠.
## 실전에서 딥링크는 이렇게 쓰여요.
이렇게 중요한 딥링크, 실전에서는 어떻게 하면 잘 활용할 수 있을까요? 요즘 마케터들이 가장 많이 사용하는 시나리오를 캠페인별로 소개할게요.
### 1\. 인플루언서 마케팅: 콘텐츠에서 구매까지의 거리 줄이기
패션 인플루언서가 Instagram에 티셔츠를 소개하는 콘텐츠를 올렸다고 가정해 볼게요. 영상 설명란에 딥링크를 삽입해두면 사용자가 해당 링크를 클릭하는 순간 앱이 실행되고 상품 상세 페이지로 바로 이동해요.
‘AB180://product/t-shirt’처럼 커스텀된 딥링크를 활용하면 콘텐츠에서 구매까지의 여정을 크게 줄일 수 있어요. 전환율을 높이는 데 도움이 되고, 캠페인별 성과 분석도 훨씬 쉬워져요.
[__wf_reserved_decorative]
### 2\. 제휴 마케팅: 브릿지 페이지 + 딥링크로 수익 연결
아마존 어필리에이트 활동을 하는 유튜버의 경우, 영상 설명란에 웹 브릿지 페이지(예: ab180.com)를 넣고 그 안에 딥링크를 연동할 수 있어요.
‘Shop All Amazon Products’ 링크를 클릭하면 아마존 앱이 실행되고, 해당 유튜버의 아마존 스토어프런트로 바로 연결돼요. 앱이 설치돼 있지 않더라도 앱스토어를 거쳐 설치한 뒤 동일한 페이지로 연결돼요.
이런 방식은 제휴 수익 구조를 유지하면서도 유입 경로를 정교하게 추적하고 최적화하는 데 효과적이에요.
[__wf_reserved_decorative]
### 3\. 리퍼럴 마케팅: 친구 초대도 매끄럽게
카셰어링 서비스를 운영하는 앱에서 ‘친구 초대하기’ 버튼을 누르면 쿠폰과 딥링크가 함께 생성돼요. 이 링크를 메신저로 공유하면, 친구는 해당 링크를 클릭했을 때 앱이 실행되면서 예약 가능한 차량 목록 페이지로 바로 이동해요.
이처럼 리퍼럴 경로에도 딥링크를 활용하면 유입 유저의 초기 경험을 간단하게 만들고 전환율도 높일 수 있어요. 단순한 링크 하나가 공유 기반 성장을 빠르게 만들어주는 셈이에요.
[__wf_reserved_decorative]
### 4\. 옴니채널 마케팅: 오프라인 광고부터 온라인 경험까지
지하철역 디지털 사이니지에 ‘앱 설치 시 첫 주문 무료’ 광고를 노출하면서 QR 코드를 함께 제공했다고 해볼게요. 유저가 QR을 스캔해 앱을 설치하면, 앱 오픈 시 쿠폰이 자동으로 적용되고 이벤트 페이지로 곧장 연결돼요.
이처럼 오프라인과 온라인을 연결하는 옴니채널 캠페인에서도 딥링크는 중요한 역할을 해요. 단순히 설치만 유도하는 데서 그치지 않고, 설치 이후 어떤 경험을 줄지도 함께 설계할 수 있기 때문이에요.
[__wf_reserved_decorative]
### 5\. CRM 마케팅: 개인화된 메시지가 직접 행동 유도
메신저나 문자 메시지로 “지금 바로 앱에서 결제하면 30% 할인!” 같은 메시지를 보낼 때, CTA 버튼에 딥링크를 연결하면 사용자는 앱을 실행하자마자 할인 적용된 체크아웃 화면으로 도달해요.
딥링크 덕분에 메시지 → 앱 실행 → 전환이라는 흐름이 끊기지 않고 이어지고, CRM 채널의 효율도 자연스럽게 높아져요.
[__wf_reserved_decorative]
### 마케터에게 딥링크란, 연결 그 이상
딥링크는 단순히 앱 내 특정 화면으로 유저를 보내주는 기술이 아니에요.
캠페인과 유저를, 온라인과 오프라인을, 콘텐츠와 행동을 자연스럽게 이어주는 마케팅 전략의 핵심 도구예요. 유저가 어떤 경로로 들어와서, 어떤 경험을 하고, 어디서 이탈하거나 전환하는지 — 이 모든 여정을 정밀하게 설계하고 측정할 수 있어야 하죠.
안정적인 연결은 물론, 합리적인 가격으로 마케팅 성과를 체계적으로 측정하고 싶다면 지금 바로 Airbridge 팀과 이야기 나누어 보세요.
---
## 유저에 맞게 모바일 게임 광고를 자동으로 설정해 주는 ‘에어플럭스(Airflux)’의 탄생
URL: https://www.airbridge.io/ko/blog/introducing-airflux-mobile-game-ad-solution
Language: Korean
Published: 2025-04-09
모바일 게임 ‘탕탕특공대’의 누적 매출이 2024년을 기준으로 무려 5억 달러(약 7,000억원)를 돌파했다는 사실을 알고 계시나요?
스마트폰으로 간단하게 조작하는 캐주얼 게임 하나로 이렇게나 많은 돈을 번다는 사실에 깜짝 놀라셨을 거예요.
이처럼 지금은 그야말로 모바일 캐주얼 게임 전성시대라고 할 수 있습니다. 잘 만든 게임 하나로 엄청난 수익을 창출할 수 있죠.
그만큼 캐주얼 게임 시장도 경쟁이 치열해졌는데요. 그 어느 때보다 게이머들의 행동을 분석해서 수익을 끌어올리는 게 중요해지고 있습니다. **그런데 막상 트래픽은 그대로인데 수익을 극대화하려니 어려움이 많으시죠?**
그래서 AB180에서는 모바일 캐주얼 게임사가 가장 쉽게 광고 수익을 극대화할 수 있도록 돕는 솔루션 ‘**에어플럭스(Airflux)**’를 출시했습니다. 에어플럭스는 모바일 광고 성과 측정 (MMP) 솔루션인 에어브릿지(Airbridge)가 탄생한지 약 8년 만에 처음으로 출시하는 독립 솔루션인데요.
오늘은 AB180의 데이터 사이언티스트 손상혁(Noel)님과 CEO 남성필(Roi)님에게 에어플럭스 프로젝트가 개발된 스토리에 관한 이야기를 들어봤습니다.
## **1\. 안녕하세요, 자기소개 부탁드립니다.**
**남성필**
저는 AB180 & Airbridge의 대표 남성필이라고 합니다. 10년 전에 AB180을 창업을 했고, 에어브릿지(Airbridge)라는 MMP를 중심으로 현재 한국, 미국, 중국, 베트남, 터키 등 약 10여 개국 500여 개 기업들에 서비스하고 있습니다. 최근에 에어플럭스(Airflux)라는 새로운 프로젝트를 런칭을 하게 되어서 소개해 드리려고 합니다.
**손상혁**
저는 손상혁이라고 하고요. 3년째 AB180 데이터 사이언스 팀에서 일하고 있습니다. 에어플럭스(Airflux) 프로젝트를 처음으로 발제하였고 현재 책임자로서 리드하고 있습니다.
[__wf_reserved_decorative]
## **2\. 에어플럭스(Airflux)는 어떤 서비스인가요?**
**손상혁**
에어플럭스가 어떤 제품인지 설명해 드리기 전에 모바일 게임이 어떻게 수익을 창출하는 지 말씀드려야 할 것 같아요.
모바일 게임, 그중에서 특히 캐주얼 게임은 돈을 방법이 두 종류가 있어요. **첫 번째는 게임을 하다 보면 나오는 ‘인앱 광고(In-App Advertising, IAA)’, 두 번째는 게임을 하면서 유저가 뭔가를 구매하는 ‘인앱 구매(In-App Purchase, IAP)’라고 하죠.** 게임 종류마다 둘의 비율은 게임 종류마다 좀 다른데요.
캐주얼 게임으로 갈수록 인앱 광고 비중이 커지고, 하드코어 게임으로 갈수록 인앱 구매 비중이 커집니다. 예를 들어, 인앱 광고와 인앱 구매 비중이 캐주얼 게임은 100% : 0%, 하드코어 게임은 0% : 100%인 것이죠. 그래서 두 가지를 아예 섞지 않고 하나만 사용하는 경우도 있어요.
그리고 미드코어로 분류되는 방치형 게임은 인앱 광고를 보상형 비디오(Rewarded Video, RV) 정도만 적당히 보여주고 인앱 결제로 과금을 유도합니다. 반대로 캐주얼 게임 같은 경우는 구매가 훨씬 적게 일어나요. 게임성이 가벼워서 유저에게 과금할 요소도 별로 없기 때문이죠.
**그래서 보통 캐주얼 모바일 게임은 인앱 광고로 수익을 창출합니다.** 왜냐하면 구매도 거의 안하는데 광고도 안보면 들어와서 게임을 해도 수익을 창출할 방법이 없으니까요.
그런데 여기서 문제가 하나 생깁니다. 모바일 게임사들이 많이 트는 비디오 광고는 보통 강제 광고라고도 불리는 인터스티셜(Interstitial)과 보상형 비디오(Rewarded Video)의 두 가지 종류로 나눠지는데요.
인터스티셜 광고의 경우 게임을 플레이하는 도중에 나오기 때문에 유저 입장에서는 게임의 경험을 크게 방해받게 됩니다.
그래서 강제 광고를 많이 보게 된다면 유저가 이탈할 수도 있고, 유저가 이탈해버리면 이후에 그 유저가 벌어다줬을 잠재적인 인앱 광고 또는 인앱 결제 수익도 모두 잃는거죠. 그렇다고 너무 강제 광고를 적게 보여주면 게임사의 수익이 줄어듭니다.
현업에서는 강제 광고 수익을 최적화하기 위해 주로 ‘수익화 매니저’들이 직접 작업을 하는데요. 이때 **언제 광고를 보여줄 것인지는 아예 고정 변수로 설정해두거나, 엔지니어들과 함께 그때 그때 수동으로 설정합니다.**
예를 들어, 1, 2 스테이지가 끝나고 무조건 한 번 광고를 보여주게 하거나, 두 스테이지마다 한 번 씩 인앱광고가 뜨게 설정하는 작업을 하죠. 그 후에 광고가 나와야 하는 자리에서 유저에게 최대한 광고 단가가 높은 광고를 보여주기 위한 최적화를 하곤 합니다.
그런데 수익을 높이려면 유저가 어떤 광고든 일단 광고를 봐야하는 거잖아요. 입찰가 45원을 50원으로 높여서 5원 더 버는 것도 중요하지만, 유저가 광고를 한번 더 보게 하면 그게 더 이득이겠죠.
그런데 유저들의 강제 광고에 대한 저항성은 국가, 지역, 디바이스, 유저 성향, 게임 진행도에 따라 모두 다릅니다. 그러나 대부분의 게임사들은 이미 설정해둔 고정 변수 등에 따라서 많은 유저들을 같은 그룹으로 묶어서 광고를 내보냅니다.
그래서 **제가 생각하는 진짜 최적화가 완성되려면 유저들이 거부감을 느끼지 않는 선에서 광고를 더 볼 수 있게 조정하며 수익을 극대화하는 작업**이 반드시 필요하다 생각해요.
하지만 유저마다 광고에 대한 거부감의 정도가 달라서 이것을 일일이 조정하는 게 쉽지 않은데요. **에어플럭스(Airflux)는 이것을 자동으로 조정해주는 툴입니다.**
**에어플럭스(Airflux)는 국가별, 지역별, 디바이스, 유저 성향, 게임 진행도 등에 따라 촘촘하게 그룹을 나누고, 그룹마다 서로 다른 빈도와 타이밍으로 광고를 보여줍니다.** 예컨대 구매를 많이 할 것 같은 그룹이면 광고를 좀 덜 보여주고 그러죠. 이러한 광고의 타이밍이랑 빈도를 에어플럭스가 자동으로 조정합니다.
[__wf_reserved_decorative]
**남성필**
상혁님과 저희 데이터 사이언스 팀은 유저가 게임을 할 때 언제 강제 광고를 보느냐에 대한 타이밍, 그리고 얼마나 강제 광고를 보는지에 대한 빈도수가 유저의 LTV에 막대한 영향을 끼친다는 것을 통계학적으로 증명해냈습니다.
그리고 이 인과관계를 다양한 모델을 통해서 설명하게 되었고요. 실제로 이 인과관계의 핵심이 되는 타이밍과 빈도수를 잘 조절한다면 저희는 LTV가 높아진다는 것을 확인할 수 있었습니다.
많은 모바일 게임사, 특히 하이퍼 및 하이브리드 캐주얼 게임사는 작은 LTV의 차이만으로도 수익성에 큰 영향을 받는데요. 지금과 같이 모바일 게임 개발 난이도가 낮아지면서 경쟁이 심화하고 있는 상황에서는 확보된 유저들을 얼마나 우리 게임에 머물게 만드느냐가 수익 수준을 결정지을 거라 예상하고 있어요.
에어플럭스는 인간이 진행하는 수익화와 라이브옵스(LiveOps : 게임을 출시한 이후 업데이트나 기능 향상을 통해 게임을 지속해서 보강 및 개선하는 작업)의 일부분을 개인화와 자동화된 A/B테스트를 통해서 이를 해결해 줍니다.
그래서 저희는 자신있게 말씀드릴 수 있습니다. 결과적으로 에어플럭스가 게임사에게 제공할 수 있는 가장 큰 효용은 바로 매출과 수익성임을 말이죠.
**에디터**
그럼 정리해 보자면, 이전까지는 A/B테스트와 광고 노출 모두 고정 변수로 설정해두거나, 수동으로 그 때 그 때 조정 했었는데, **에어플럭스로 유저가 사용하는 기기랑 국가, 이용 행태에 따라 광고의 유형과 타이밍과 빈도수까지 모두 자동으로 조정해 주는 자동화 시스템**이라고 할 수도 있겠네요.
**남성필**
네. 그렇습니다. 그렇게 되었을 때 LTV를 상승시킬 수 있다는 것이 핵심입니다.
## **3\. 다른 분야도 있었을 텐데 왜 하필 모바일 게임 쪽이었나요?**
**손상혁**
지금 다른 산업보다 캐주얼 게임 분야에 이런 Saas 솔루션이 가장 필요하다고 느꼈어요.
캐주얼 게임사들이 매출 규모에 비해 조직 규모가 크지 않다는 걸 아시나요? 매출이 500억 원이나 600억 원씩 하는 게임사들 중에 직원 수가 50명 남짓인 곳 들도 많아요. 어떻게 보면 굉장히 효율적인 거죠.
**그러나 반대로 얘기하면, 지난 10년간 그만큼 게임 개발 외적인 일, 특히 수익화나 라이브옵스, 그리고 데이터를 분석하는 인력들이 그렇게 많아지지 않았다는 것을 의미하기도 해요.** 실제로 저희가 여러 게임사와 미팅을 해봤지만, 데이터 대시보드를 만들어주시는 데이터 애널리스트 분들은 있어도 데이터 사이언티스트는 많이 본 적이 없어요.
이커머스 업계는 수많은 데이터 사이언티스트와 CRM 마케터를 고용해서 고객 1명에게서 얼마나 많은 수익을 창출할까 고민하고 있어요. 그리고 그런 업무를 보다 쉽고 효율적으로 하기 위해서 앰플리튜드(Amplitude)나 브레이즈(Braze) 같은 데이터 분석 툴을 쓰는 거죠.
**그런데 게임 분야는 적어도 수익화, 특히 광고 수익화와 데이터 분석과 관련해서는 현재 그러한 시스템이 많이 없어요.** 대신 신규 유저를 유입시키는 마케팅이나 개발 관련된 시스템 위주로 돌아가고 있죠. 엑셀로 모든 것을 관리하는 게임사들도 많습니다.
**남성필**
상혁님 말씀처럼, 게임사 내부에서는 데이터를 정밀하게 분석할 수 있는 인력 자체가 많지 않다 보니까, 수익을 더 낼 수 있는 기회가 많음에도 불구하고 내부에서 이를 극대화하지 못하고 있어요.
그리고 모바일 게임 시장이 생각보다 큽니다. **전 세계에서 이제 인게임 광고 시장 크기가 2025년에만 1,24B억 달러 정도라고 해요. (출처:** [**In-game Advertising - Worldwide, Statista**](https://www.statista.com/outlook/amo/media/games/in-game-advertising/worldwide?utm_source=chatgpt.com)**) 원화로 대략 184조 원이 넘는 시장을 가지고 있는 거죠.** 광고를 최적화할 수만 있다면 지금 놓치고 있는 매출을 더 끌어올릴 수 있는 거예요.
또한 상혁님이 말씀하신 것처럼 이커머스 업계에서 머신러닝과 AI를 통한 예측과 추천은 매우 보편화되었습니다.
그러나 게임 업계에서는 아직 탑 레벨의 극소수의 퍼블리셔, 스튜디오들만 하고 있는 것이기도 하고요. 저희는 광고 수익화 분야에서 게임사들이 같은 레벨의 트래픽으로 수익을 극대화할 수 있는 머신러닝과 AI 모델들을 계속 만들어내고자 합니다.
**손상혁**
저는 이럴 때 Saas의 가치가 빛난다고 생각해요. 이런 데이터들을 정밀하게 분석하고 실험하면서 다양한 최적화 모델을 만들어내려면 수학, 통계, 모델링에 숙련된 데이터 사이언티스트를 여러 명 고용해야 합니다. 이 부분을 대신해 줄 수 있는 Saas를 만들면 게임사들의 수익을 더 높여줄 수 있을 거라 판단했습니다.
## **4\. 그럼 지금 모바일 게임 개발사들도 이러한 문제에 공감하고 있나요?**
**손상혁**
물론이죠. 현재 저희는 10개 이상의 게임 고객사들과 Airflux 테스트를 진행 중에 있고, 좋은 성과를 거두고 있습니다. 출시 이전부터 많은 게임사 관계자 분들께서 만들어서 출시하게 되면 꼭 테스트해보고 싶다고도 말씀해주셨어요.
이미 무료 테스트는 슬롯이 다 차서 유료 POC(Proof of Concept)만 진행하고 있는데도 긍정적인 반응이 많습니다. 그래서 더욱 힘이 나는 것 같아요.
이런 프로덕트가 많지 않았기 때문에 저희도 더 추진력을 갖고 열심히 만들어서 게임사들의 매출 및 수익 증대에 기여하고자 합니다.
**남성필**
기본적으로 상대적으로 적은 마진을 가지고 볼륨 게임을 해야 하는 하이퍼캐주얼 회사들이 특히 관심이 많으신 것 같습니다.
그리고 저희가 인터스티셜을 넘어서 보상형 비디오 광고도 일부 실험을 하고 있어서 하이브리드 캐주얼 회사들도 많이 관심을 보여주시고요.
특히 2024년에 eCPM((Effective Cost Per Mille : 광고 노출 1,000회당 벌어들인 광고 수익)증가폭이 생각보다 적었는데요.
광고 지면을 늘리거나, 최소 입찰가(Bid floor)를 설정하거나, 비디오 광고의 길이를 늘리거나, 우리 지면에 입찰을 열심히 하는 매체들의 수를 늘려서 경쟁 강도를 높이는 등 게임사가 자체적으로 할 수 있는 것은 다 해보고 있지만 아무래도 아이디어가 많이 부족해지고 있는 것 같습니다.
이렇게 갈수록 난이도가 높아지는 매체 최적화의 상황이다보니 새로운 방안으로서도 주목해주시는 것 같고요.
게임사들은 한정된 인력과 리소스를 통해서 수익화를 최적화해야 하기 때문에, 저희는 단기적으로는 강제 광고와 보상형 비디오 광고, 그리고 장기적으로는 인앱 구매(IAP) 등까지 매출과 수익의 영역만큼은 철저히 저희 자체 머신러닝과 AI, 그리고 자동화 시스템을 통해서 극대화할 수 있도록 도와드리고자 합니다.
이렇게 LTV가 높아지면 자연스럽게 유저 확보에 돈을 더 투자해서 더 많은 유저를 확보하게 될 것이고, 여기서 얻은 수익으로 또 마케팅에 투자하는 선순환이 완성되면서 게임이 성장할 수 있기 때문이죠.
[__wf_reserved_decorative]
## **5\. 파이어베이스(Firebase)와 같은 A/B 테스트솔루션을 이용하여 직접 실험을 하는 것과 어떻게 다른가요?**
**남성필**
파이어베이스로 할 수 있는 A/B테스트는 게임사에서 직접 가설을 세우고 설계를 해야 합니다. 예를 들어 스테이지 1, 스테이지 2가 끝나면 50% 확률로 인터스티셜을 틀 것인가 혹은 70% 확률로 인터스티셜을 틀 것인가 변수를 세팅해서 테스트를 할 수 있습니다.
이런 형태의 실험은 단편적인데 비해 많은 리소스가 소모되는 문제가 있습니다. 모바일 게임처럼 변수가 수백 개인데 그 모두를 A/B테스트하기에는 적합하지 않습니다.
특히 실험을 설계하고 실행해야 하고, 그 결과를 보고 피드백해야 하는데 그게 생각보다 시간이 너무 오래 걸려요. 변수도 너무 많고요. 무엇보다도 제대로된, 통계적으로 의미있는 A/B테스트를 한다는 것은 생각보다 굉장히 어렵고 복잡한 일 입니다.
예를 들어, 모바일 게임사가 100여 개 국가에서, 디바이스(안드로이드/iOS), 그리고 유저들의 반응도를 고려해서 각각 A/B테스트를 한다고 하면 수백 수천 개의 변수(Variant)가 필요해요.
**그걸 게임사들이 직접 코드 몇 줄만으로 즉각적으로 A/B테스트를 시행해서 효과를 볼 수 있게끔 완전하게 편하게 만든 솔루션이 바로 에어플럭스인 거죠.**
그러므로 모바일 게임 광고 수익화와 관련된 부분은 에어플럭스와 같은 솔루션에 맡기는 것이 보다 효율적일 것이라고 확신합니다.
기존의 A/B테스트 솔루션은 크고 굵직한 게임 디자인이나 웰컴 패키지와 같은 강력한 BM과 관련된 테스트를 하는데 사용하는 게 더 좋고요.
**손상혁**
제대로 된, 통계적으로 의미있는 A/B테스트를 한다는 것은 생각보다 굉장히 어렵고 복잡한 일 입니다. 생각보다 시간이 너무 오래 걸리고, 변수도 너무 많죠.
**에어플럭스는 머신러닝으로 가설 수립, 실험, 결과 해석 및 적용을 자동화합니다. AI 알고리즘이 스스로 유저 그룹을 나누고, 수 많은 A/B테스트를 자동으로 진행하며 최적의 수익에 다가갑니다.**
따라서 에어플럭스를 사용하는 게임사들은 데이터사이언티스트들을 직접 고용할 필요가 없고, 지속적으로 엔지니어와 수익화 매니저의 시간을 사용하며 광고를 테스트 할 필요가 없어진다는 점에서 많은 시간과 비용이 절약될 겁니다. 그리고 에어플럭스를 사용함으로써 기존에 발생하지 않던 매출이 발생된다는 점이 가장 중요할 것 같네요.
**남성필**
그래서 한 번 더 요약해 보자면,
1. 유저의 특성에 맞게 강제 광고의 빈도 수, 타이밍을 조절해 준다.
2. 에어플럭스를 통해 동일한 트래픽으로 유저 LTV를 높여준다.
3. 광고 수익화 변수 설정 과정에서 일일이 수동으로 처리하던 작업들을 자동화해 준다.
4. 광고 수익화에 대한 A/B테스트를 하고 있었다면, A/B테스트 자체를 직접 할 필요 없이 에어플럭스가 대체 해준다.
5. 유저가 감당할 수 있는 임계점 수준까지의 광고를 최대한 보여줌으로써, 유저 경험을 헤치지 않고 이탈은 줄이면서 수익을 극대화한다
6. 자체 순증가분(Incrementality) 분석을 통해 그 효과를 엄밀하게 측정할 수 있다.
이렇게 정리할 수 있을 것 같습니다.
## **6\. 혹시 실제로 효과를 본 사례가 있나요?**
**손상혁**
위에서 말씀드린 바와 같이 저희는 현재 10개 이상의 게임 고객사들과 에어플럭스(Airflux) 테스트를 진행 중에 있고, 좋은 성과를 거두고 있습니다.
글로벌 1,000만 다운로드 이상의 좀비 FPS 게임에서 약 50%의 유저 LTV 증가 효과를 본 바 있고, 글로벌 1,000만 MAU 게임사와 시행했을 시에도 2주만에 14%의 LTV 증가 효과를 거둔 바 있습니다.
한편, 또 다른 대형 게임사 1곳도 현재 16%에 달하는 개선을 보이고 있어서 고무적입니다.
[__wf_reserved_decorative]
## **7\. 앞으로 모바일 게임 개발사들이 글로벌 시장에서 살아남기 위해 어떤 전략이 필요할까요?**
**남성필**
이커머스도 처음에는 상품을 잘 판매하는 것에만 집중을 하다가, CRM이나 데이터 분석, 로열티 프로그램 등을 활용해서 기존 유저들의 LTV를 높이는 쪽으로 트렌드가 바뀌었습니다.
특히 경제 불황 이후에는 한 명의 유저를 싸게 데려와서 빠르게 확장할 수 있는 시대가 지났고, 지금 현재 모바일 게임도 이러한 흐름을 이어가고 있습니다.
앞으로는 머신러닝으로 광고 시스템을 최적화하여, 한 명의 게이머에서 수익화할 수 있는 LTV를 높이고 광고 수익을 극대화 하는 게 중요해질 것이라 생각합니다.
그리고 이 모든 것을 가장 효과적으로 처리해 줄 수 있는 솔루션이 바로 에어플럭스라는 점을 말씀드리고 싶습니다.
**손상혁**
저도 성필님이랑 같은 의견이고, 갈수록 모바일 게임 시장의 진입장벽을 낮아지고 경쟁은 치열해질 겁니다.
당연히 흐름은 게이머의 유저 경험을 헤치지 않으면서 LTV를 높이는 방향으로 갈 수밖에 없습니다. 더 간단히 하면, **돈 써서 모셔온 유저를 소중히 모시고, 동시에 수익화도 극대화 하는 것이죠.** 여기에 가장 적합한 솔루션이 바로 에어플럭스라는 점을 말씀드리고 싶습니다.
감사합니다.
---
## 플레이오, 에어브릿지로 MMP를 바꾸고 전 세계 마케팅을 함께 관리하고 있어요
URL: https://www.airbridge.io/ko/blog/playio-mmp-migration-global-marketing
Language: Korean
Published: 2025-04-01
## Best Value for Gamers, 게이머를 위한 리워드 앱 플레이오
### Q. 간단한 자기소개 및 회사소개 부탁드려요.
안녕하세요, 플레이오 마케팅 팀에서 리드를 맡고 있는 박하영입니다. 저는 첫 회사 생활도 게임사에서 시작할 정도로 게임에 진심인 사람이에요. 현재는 지엔에이컴퍼니에서 플레이오와 함께 Best Value for Gamers라는 목표 아래 건강한 게임 생태계를 만들어 가기 위해 노력하고 있습니다.
### Q. 플레이오 서비스에 대한 소개 부탁드려요.
플레이오는 기본적으로 플레이 타임 기반 리워드 앱 서비스예요. 유저들이 게임을 플레이한 시간만큼 포인트를 적립 받고, 그 포인트를 상품과 교환할 수 있는 구조를 가지고 있어요. 하지만 플레이오가 지향하는 바는 단순한 리워드 앱이 아닌 게이머가 고유의 가치를 인정받는 환경이에요. 게이머들이 정당한 보상을 받고 가치를 인정받을 수 있도록 여러 기능을 추가하며 나아가고 있어요.
예를 들어 단순한 리워드 제공이 아닌 실제 플레이 시간을 바탕으로 유저 레벨과 티어를 나누어 보상해 주고 다양한 게이머가 모여 있는 커뮤니티 역할도 수행하고 있죠. 유저들에게는 다양한 게이머가 모여있어 함께 게임을 즐기고 공감대를 형성할 수 있는 공간으로, 게임사들 입장에서는 소위'찐 게이머'가 많이 모여있는 플랫폼으로 유저 유입과 리텐션을 높일 수 있는 마케팅 채널로 주목받고 있습니다.
[플레이오 서비스 바로가기](https://playio.co/?LANG_CD=KOR)
[]
### Q. 국내 뿐만 아니라 해외에서도 서비스 중인 것으로 알고 있어요.
맞아요. 플레이오는 현재 한국, 미국, 일본, 대만 지금 서비스 중에 있고 홍콩, 마카오 등으로 서비스 확장을 계획하고 있어요.
### Q. 플레이오의 마케팅을 이끌어가는 마케팅 팀은 어떻게 구성되어 있고 어떤 일을 하고 계신가요?
저희 마케팅 팀은 크게 B2B와 B2C 파트로 나누어져 있어요. B2B 파트는 2025년부터 새롭게 구성된 팀으로 사내 세일즈 팀과 협력하면서 플레이오가 얼마나 효과적인 마케팅 매체인지 알리기 위해 노력하고 있어요. 주로 아티클을 발행하거나 뉴스레터를 보내는 등의 활동이 이에 포함되죠. B2B 파트에서는 리드를 수집해 새로운 고객사를 유치하고 기존 고객사들과도 관계를 다지기 위한 활동을 진행합니다.
B2C 파트에서는 유저들이 플레이오를 더 잘 발견하고 즐길 수 있도록 노력하고 있어요. 신규 유저 유입은 물론이고 기존 유저들이 계속 만족하며 플레이오 안에서 좋은 경험을 받을 수 있도록 다양한 마케팅 액션을 이어가죠.
## 플레이오의 에어브릿지 도입기
### Q. 플레이오는 세 번째 MMP로 에어브릿지를 사용하고 있는 것으로 알고 있어요. 플레이오가 에어브릿지로 MMP를 옮겨오게 된 배경이 궁금해요.
작년에 일본 시장 진출을 시작하면서 현지 퍼블리셔와 협업을 시작했어요. 협업 과정에서 마케팅도 퍼블리셔에서 맡아 진행하게 되었는데요. 그러다 보니 일본 외 다른 국가 마케팅 성과를 퍼블리셔에게 공개하지 못하도록 국가별 대시보드 분기가 필요했어요. 당시 사용 중이던 MMP에서는 해당 기능이 어렵다고 했죠.
이에 따라 국가별 데이터 공개 수준 제한이 가능한 MMP를 찾게 되었죠. 하지만 대부분의 MMP들에서 지원이 어려운 기능이었어요. 에어브릿지에서만 유일하게 가능하다는 답변을 받아 이관을 결정했어요.
물론 이것만이 유일한 이유는 아니에요. 개인적으로는 이전부터 에어브릿지가 다른 MMP 대비 직관적인 UI/UX를 가지고 있다고 생각했어요. 여기에 더불어 팀에서 필요한 기능도 가능하다 하니 이관을 결정하지 않을 이유가 없었던 것 같아요. 마케팅 성과라는 게 단순히 데이터 분석가나 퍼포먼스 마케터를 제외하고도 많은 구성원들이 함께 보고 이야기 나눌 수 있어야 한다고 생각하는데 에어브릿지의 직관적인 UI/UX를 통해서라면 가능할 것이라 생각하기도 했어요.
### Q. 에어브릿지로 이관은 어느정도 소요되었고, 과정은 어떠셨나요?
한 달 정도 소요되었어요. 아까도 말씀드렸던 것처럼 국가별 데이터 권한 분리라는 특수한 요청 사항이 있기도 했고 연동 과정에서도 추가적으로 요청드린 사항이 많아서 이관 미팅을 자주 가지게 되었는데 에어브릿지 팀의 적극적 지원 덕분에 순조롭게 마무리할 수 있었어요.
사실 솔루션 이관이라는 게 자주 있는 일은 아니잖아요. 그래서 더욱 어렵고 오래 걸리기 일쑤인데 에어브릿지에서 꼼꼼한 체크리스트를 기반으로 체계적인 프로세스를 제공해 주셔서 빠르게 마무리될 수 있었어요. 특히 온보딩 기간 전사 교육도 함께 진행해 주셔서 사내 더 많은 동료 분들이 에어브릿지를 사용할 수 있게 되었죠.
[]
## 플레이오의 에어브릿지 활용기
### Q. 플레이오는 한국 뿐만 아닌 다양한 국가에서 서비스를 전개하고 있으십니다.
국가별로 전략도 매체도 다르고, 이에 따른 성과 측정 효율화가 중요할 것 같아요.
맞아요. 말씀하신 것처럼 국가별로 전략과 매체 운영을 다르게 가져가고 있어요. 한국은 2021년부터 서비스를 시작했고 일본은 작년 4월, 대만은 11월 런칭해 운영 중이죠. 한국 같은 경우에는 작년을 기점으로 D30 리텐션이 30%를 넘기면서 유저 애착도가 많이 높아진 상태예요. 그래서 새로운 유저 확보뿐만 아니라 브랜드의 비전을 효과적으로 알릴 수 있는 콘텐츠, 캐릭터, CRM 전략에도 집중하고 있어요.
대만과 일본의 경우 빠르게 성장하고 있는 단계로 신규 유저 확보에 조금 더 중점이 맞춰져 있고요. 기본적으로 구글과 메타처럼 대형 채널들과 DSP 매체를 적극적으로 운영하며 국가별 전략을 최적화하고 있어요.
### Q. 모바일 마케팅 최적화를 위해 에어브릿지에서 가장 잘 사용하고 계신 기능이 궁금해요.
매체별, 소재별 광고 성과 측정에서 에어브릿지 지표와 자료들을 잘 사용하고 있어요. 플레이오에서 특히 잘 사용하는 기능 중 하나는 다양한 리포트에요. 다른 MMP에서는 이렇게까지 다양한 리포트를 제공하지 않는데 에어브릿지는 원하는 지표별로 커스터마이징 가능한 리포트가 많아서 좋아요.
[에어브릿지 리포트 기능 확인하기](https://help.airbridge.io/ko/guides/airbridge-report-overview)
저희 마케팅 팀 목표는 전사 KPI와도 얼라인 되어 있어요. 예를 들어 신규 유저 증가, 리텐션 상승과 같이 말이죠. 이런 목표를 전사가 함께 보기 위해 에어브릿지에서 제공하는 다양한 리포트 기능을 활용해 차트를 만들고 공유하고 있어요. 특히 앞서 이야기한 온보딩 교육을 통해 전사가 에어브릿지를 이용할 수 있게 되어 꼭 마케팅 팀을 통하지 않고도 동료분들이 직접 리포트를 만들기도 해요.
또 다른 기능은 커스텀 채널을 통한 트래킹 링크에요. 플레이오는 모바일 광고 이외에도 인플루언서 마케팅을 포함한 콘텐츠 마케팅, 외부 협업 등을 통한 프로모션을 많이 진행하고 있는데요. 이럴 때면 성과 측정을 위해 커스텀 채널을 활용하고 있어요. 각 캠페인별, 인플루언서별 커스텀 채널을 생성해서 성과를 측정하고 고도화해요.[](https://help.airbridge.io/ko/guides/custom-channels)
[에어브릿지 커스텀 채널 기능 확인하기](https://help.airbridge.io/ko/guides/custom-channels)
### Q. 에어브릿지 기능을 활용해 마케팅 성과를 이뤄낸 사례가 있다면 소개해 주세요.
한국에서는 에어브릿지 퍼널 리포트와 내부 데이터 분석 자료를 연계해서 성과를 개선한 사례가 있어요. 한국에서는 리텐션 개선이 중요한 목표 중 하나였어요. 그래서 내부 데이터를 통해 유저 아하 모먼트를 찾고 이 데이터를 바탕으로 아하 모먼트로 유저를 유도하기 위한 다양한 액션 아이템을 만들어 AB테스트를 진행하고 에어브릿지 퍼널 리포트로 결과를 확인하며 최적화 했죠. 결과적으로 아하 모먼트로 경로를 최적화하며 국내 유저 D30 리텐션이 30%를 넘길 수 있게 되었어요.
또한 국가별 마케팅 최적화 사례도 있어요. 대만의 경우 에어브릿지 액츄얼스 리포트를 통해 UA 캠페인에서 이미지 소재에 비해 영상 소재 효율이 크게 높다는 점을 발견할 수 있었어요. 이 데이터를 바탕으로 UA 캠페인에서 영상 비중을 늘렸고 유저 유입을 100% 이상 상승과 유저 획득 비용 40% 절약의 성과를 이뤄낼 수 있었죠.
> 플레이오가 에어브릿지를 활용해 이뤄낸 모바일 마케팅 성과의 자세한 내용이 궁금하다면?
[케이스스터디에서 확인하기 👉](https://www.airbridge.io/ko/case-studies/playio-d30-retention-cpa-optimization)
### Q. 마케터로서 느끼시는 에어브릿지의 장점이 있다면 소개해 주시면 감사하겠습니다.
제가 느끼는 에어브릿지의 가장 큰 장점 중 하나는 MMP임에도 데이터 시각화가 정말 잘 구성되어 있다는 것이에요. 다양하게 구성할 수 있는 리포트도 많고, 원하는 대로 데이터 조합이 가능할 뿐만 아니라 디자인도 직관적이죠. 앞서 이야기한 것처럼 플레이오가 다양한 MMP를 거쳐왔지만 이렇게까지 다양하고 직관적인 대시보드를 제공해 주는 곳은 에어브릿지가 유일했어요.
[]
또한 마케터가 아닌 다른 직무의 직원들도 함께 잘 사용할 수 있을 만큼 사용성이 좋다는 것도 큰 장점으로 느껴요. 마케터나 데이터 담당자가 아니라면 데이터를 확인하는 일이 부담스러울 수도 있는데 에어브릿지는 직관적인 UX/UI와 손쉬운 사용성이 강점이라 마케터가 아닌 구성원들도 부담 없이 사용할 수 있어요.
## 플레이오가 그리는 게임 엔터테인먼트 매체의 미래 (리워드 2.0)
### Q. 플레이오는 단순히 리워드 매체가 아닌 게이머의 가치를 인정하고 존중하고 있는 매체인데요.
플레이오가 그리는 미래는 게임 리워드 매체의 무엇인가요?
게이머는 단순한 재미부터 성취감, 유대감, 인정욕구 등 다양한 동기를 기반으로 게임을 플레이해요. 플레이오는 이러한 게이머들의 다양한 동기를 더 극대화할 수 있도록 앞으로 더 다양한 기능과 콘텐츠, 리워드를 제공할 예정이에요.
저희는 플레이오의 비전이 단순히 리워드 매체에 있지만은 않다고 생각해요. 그리고 이걸 '리워드 2.0'이라고 명명하고 있는데요. 앞으로도 리워드 2.0이라는 비전 아래 유저들에게는 게이머로서의 가치를 인정해 주고, 게임사에게는 횔씬 뚜렷한 타겟 광고가 가능한 매체가 되도록 나아갈 예정이에요.
### Q. 리워드 2.0의 길을 걸어가는 플레이오 마케팅 팀의 목표는 무엇인가요?
앞서도 이야기했던 것처럼 플레이오는 단순한 리워드 제공을 넘어 건강한 게임 생태계를 만들려 노력하고 있어요. 마케팅 팀도 이 목표 아래에서 함께 열심히 움직이고 있는데요. 팀원들 모두가 한 명의 게이머로서 제품의 방향성에 맞춰 게이머에게 더 많은 가치를 제공하는 경험을 더 많이 만들 수 있도록 노력 중이에요. 그것이 플레이오가 마케팅을 하는 이유에요.
[]
### Q. 끝으로 하고 싶은 이야기가 있다면, 자유롭게 부탁드립니다.
여러 MMP를 거쳐왔지만, 에어브릿지에 정착하게 되어 기뻐요. 복잡한 이관 과정이 빠르게 처리되고 전사가 함께 에어브릿지 데이터를 통해 성장을 이뤄내고 있어요. 직관적인 대시보드와 빠른 CSM의 대응 등 장점이 많아 이제는 다른 MMP를 사용하기 어려울 것 같아요.
플레이오 구성원 모두가 더 좋은 서비스를 만들기 위해 노력하고 있어요. 새로운 기능도 계속 추가되고 서비스도 빠르게 변하고 있으니 플레이오의 행보를 계속 잘 지켜봐 주시면 좋겠어요. 빠르게 성장 중인 플레이오와 함께 하실 분들도 계속해서 모시고 있어요. 관심 가지고 지켜봐 주시면 좋겠습니다.
[지엔에이컴퍼니 채용공고 바로가기](https://gna.career.greetinghr.com/team)
에어브릿지와 함께 글로벌 모바일 마케팅 성과를 최적화하고 싶으신가요?
---
## 딥링크 없는 앱 마케팅이 손해인 6가지 이유
URL: https://www.airbridge.io/ko/blog/organic-app-growth-with-deep-linking
Language: Korean
Published: 2025-03-28
## 딥링크의 시작과 현재
[딥링크(deep linking)](https://www.airbridge.io/ko/blog/what-is-deep-link)는 2013년 이후 커머스가 폭발적으로 성장하면서 유저 행동 변화에 대응하고자 본격적으로 도입되었어요. 당시 모바일 웹에 비해 앱을 통한 구매 전환율은 높았지만, 유저들을 기존 웹 채널에서 앱으로 유도하는데 어려움을 겪었는데요. 이 문제의 해결책이 바로 딥링크였어요.
딥링크란 마케팅 캠페인에서 유저를 앱 내 특정 화면으로 매끄럽게 안내해 주는 기술이에요. 이미 앱을 설치한 유저를 특정 위치로 이동시키는 표준 딥링크뿐만 아니라, 앱을 설치하지 않은 유저를 앱 설치 이후 원하는 특정 위치로 연결하는 디퍼드 딥링크가 발전하면서 마케팅 업계에 빠르게 확산되었어요.
[__wf_reserved_decorative]
10년이 지난 오늘날의 딥링크는 SEO, 웹투앱 전환, 제휴 마케팅 등 오가닉 성장 전략의 핵심 도구로 재조명되고 있어요. 그렇다면 왜 10년 전 기술이 2025년에 다시금 부상하고 있을까요? 앱 성장을 위한 마케팅 전략에 딥링크가 필수인 이유 6가지를 소개할게요.
## 1\. 오가닉 성장 중심의 앱 마케팅 변화
최근 모바일 중심 앱들은 산업 성장률 둔화 및 투자금 감소 흐름에 발맞춰 수익성을 고려하는 ‘지속가능한 성장’을 우선시해요. 이에 따라 마케팅의 방향성도 상대적으로 비용 효율이 높은 오가닉 성장 전략으로 적극 전환중이에요. 여기에 마케터는 딥링크를 다음과 같은 전략에 접목해 핵심 마케팅 전략으로 활용하고 있어요.
- 검색 엔진 최적화(SEO): 딥링크는 검색 결과에서 앱으로 매끄러운 링크 연결을 지원해 검색 순위를 향상시켜요. 또한, 유저를 특정 앱 콘텐츠로 안내해 검색 결과에서 전반적인 검색 가능성을 높여요.
- 앱 스토어 최적화(ASO): 딥링크를 맞춤형 제품 페이지(CPP) 또는 맞춤 스토어 등록정보(CSL)와 결합하면 앱 콘텐츠를 유저의 관심사와 일치시킬 수 있어요. 앱 스토어 가시성은 전환율 향상으로 이어져요.
- 리퍼럴 마케팅: 딥링크는 유저들이 앱 콘텐츠를 손쉽게 공유할 수 있도록 지원해요. 추천으로 가입한 유저를 위한 맞춤형 온보딩 경험을 구현해 그 효과를 더욱 높여요.
- 콘텐츠 공유: 딥링크는 유저 간 앱 콘텐츠 공유를 단순화해요. 공유된 콘텐츠가 앱 내에서 바로 열리도록 구현해 콘텐츠 확산에 기여하고 전반적인 유저 경험을 향상시켜요.
- 웹투앱 전환: 딥링크는 모바일 웹 유저를 앱 내 동일한 콘텐츠로 전환해요. 궁극적으로 웹 사용자를 보다 직관적인 앱 경험으로 유도해 전환율을 높여요.
> “이제 마케터들은 콘텐츠 생성을 자동화해 'A Top 10' 또는 'B의 대안 베스트 5' 등 이목을 끄는 페이지들을 손쉽게 구성하고 SEO와 트래픽을 높일 수 있어요. 또 각 페이지 웹투앱 배너에 명확한 CTA(Call-to-Action)와 매끄러운 딥링크를 입히면 끊김 없이 자연스러운 앱 콘텐츠 전환이 이루어집니다.”
> — [남성필](https://www.linkedin.com/in/roinam), [에어브릿지](https://www.airbridge.io/ko) CEO
## 2\. CRM과 고객 참여의 재조명
예산 삭감으로 마케터들은 고객 참여 전략에 집중하면서 자연스레 CRM이 재부상하고 있어요. 한동안 CRM은 공격적인 유저 확보 전략으로 여겨져 다소 등한시되었는데요. 현재 CRM은 기업의 유저 당 평균 매출 (Average Revenue Per User, ARPU)을 극대화하는 핵심으로 주목받고 있어요.
[__wf_reserved_decorative]
성공적인 CRM을 위해선 제품 생태계를 각종 메시징 플랫폼과 매끄럽게 연결하고 각 플랫폼의 영향력을 정확히 측정해야 해요. 이를 해결하고자 마케터들은 이메일, 문자, 카카오톡, 라인 등의 메시징 앱에 딥링크를 삽입해요. 여기서 딥링크는 유저를 앱 내 관련 콘텐츠로 바로 안내해 마찰과 이탈을 줄이고, 고객 참여로 이어지는 원활한 유저 경험을 만들어 주죠. 이처럼 딥링크를 CRM에 접목해 유저들의 리텐션을 높이는 마케팅 전략이 주목받고 있어요.
## 3\. 리타겟팅으로의 마케팅 예산 재배치
많은 앱들이 신규 유저 확보(User Acquisition)에서 유저 리타겟팅(retargeting)으로 마케팅 예산을 재배치하고 있어요. 앞서 설명한 것처럼, 고객 ARPU 극대화를 위한 CRM 전략의 중요성이 날로 커지고 있음을 보여주는 변화에요. 기업들 또한 지출 최적화를 위해 검증되지 않은 새로운 고객을 끌어오는 UA 캠페인의 비중을 줄이고 있어요. 대신 기존 유저에 대한 리타겟팅을 통해 유저 인게이지먼트를 보다 강화하고 높은 참여도와 수익을 창출해요.
이때 딥링크는 리타겟팅 캠페인에 필수적인 기술이에요. 유저를 앱 내 특정 위치로 정확하게 이어 줄 딥링크 없이는 아무리 잘 짜여진 맞춤형 전략도 진가를 다 발휘하지 못하거든요.
## 4\. 마이크로 인플루언서의 등장
마이크로 인플루언서가 앱 마케팅에 지각변동을 일으키고 있어요. 특히 이들이 주로 활동하는 틱톡, 유튜브 등 플랫폼과 더불어 사용자 생성 콘텐츠(user-generated content, UGC) 광고가 마이크로 인플루언서의 성장세를 가속화하고 있어요. 개인 소셜 미디어 프로필에 게재된 단일 URL을 통해서만 유입 및 전환이 가능해 정확한 어트리뷰션을 파악하기 힘든 경우가 많지만, 확실한 것은 마이크로 인플루언서들이 앱 설치 유도 및 콘텐츠 확산에 아주 효과적이라는 점이에요.
> “마이크로 인플루언서와 UGC는 특히 북미 시장에서 모바일 앱 사용자 확보에 엄청난 효과를 거두고 있어요. 하지만 이런 캠페인은 적절한 추적 및 딥링크 설정 없이는 실패할 가능성이 크기 때문에 반드시 끊김 없는 트래킹을 확보해야 해요.”
> [— 함자 알삼라이(Hamza Alsamraee)](https://www.linkedin.com/in/hamza-alsamraee/), [NewForm](https://www.newform.com/) CE
[__wf_reserved_decorative]
💡 [틱톡 UGC 및 영상 광고의 어트리뷰션을 신뢰할 수 있는 데이터로 활용한 넥슨의 사례를 만나보세요](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)
## 5\. 웹투앱 전환의 급부상
앱 마케터 사이에서 웹투앱 전환이 주목받고 있어요. 모바일 앱, 특히 iOS 환경은 축소된 광고 식별자(ADID)와 복잡한 SKAN 측정 방식 때문에 성과 가시성이 제한적인데 반해 웹 캠페인은 보다 정확하고 직관적인 인사이트를 제공하죠. 때문에 다음과 같은 앱 카테고리에서 웹투앱 전략을 적극 도입하는 추세에요.
- 콘텐츠 앱: 웹툰, 숏드라마 등 콘텐츠 앱의 경우, 웹페이지를 통해 예고편이나 에피소드 일부를 선공개하는 방식으로 유저를 앱 경험으로 유도해요.
- 구독 기반 앱: 웹 인터페이스는 구독 앱의 가입 절차를 보다 간편하게 만들어 주는 만큼, 웹을 통한 가입 후 유저를 앱으로 전환해 지속적인 참여를 유도할 수 있어요.
- HTML5 미니 게임 또는 플레이 가능한 데모가 포함된 게임 앱: 웹투앱 전환을 활용해 유저가 웹에서 데모 게임을 체험한 후, 전체 앱 버전에서 원활하게 게임을 이어가도록 지원해요.
웹투앱 전환을 매끄럽게 연결하는 딥링크는 유저를 웹 콘텐츠에서 앱 내 특정 페이지로 원활하게 안내해요. 특히 웹 미리보기를 활용해 유저에게 강렬한 첫인상을 남기고, 검색에서 설치까지 유저를 끊김 없이 안내하고 싶다면, 딥링크를 활용해 웹투앱 전환을 끌어 올려 보세요.
## 6\. 캐주얼 게임에서도 주목하는 리인게이지먼트 전략
2025년에 들어 캐주얼 및 미드코어 게임에서도 리인게이지먼트(Re-engagement, RE) 캠페인이 점차 중요해지고 있어요. 리인게이지먼트란 앱을 이미 설치한 사용자를 대상으로 앱을 다시 사용하도록 유도하는 마케팅 전략이에요. 이커머스 앱에서 주로 활용되어 왔던 리인게이지먼트 캠페인은 이제 게임 앱에서도 도입되어 유저 이탈을 줄이고 LTV(Lifetime Value)를 늘리고 있어요.
그동안 게임 업계에서 리인게이지먼트 전략은 주로 MMORPG나 경쟁 MOBA 같이 새로운 시즌, 맵 또는 캐릭터 출시를 위해 플레이어를 다시 끌어들이는 하드코어 게임에 쓰였어요. 전통적으로 유저 라이프사이클이 짧고 전환 퍼널이 단순한 것으로 알려진 캐주얼 게임에서도 리인게이지먼트 전략을 적극 채택하고 있다는 것이 특이점인데요. 이런 변화를 이끄는 핵심 요소는 크게 두 가지로 정리할 수 있어요.
- 하이브리드 수익화 모델: 이제 게임 업계는 인앱 광고(IAA)와 인앱 구매(IAP)를 결합해 ARPU를 극대화하고 있어요.
- 게임플레이 디자인 변화: 캐주얼 게임은 하이퍼 캐주얼에서 하이브리드 형식으로 진화하고 있어요. 더욱 발전한 핵심 루프와 메타 게임 요소를 통합해 참여도와 리텐션을 향상시키고 있어요.
게임 업계의 이러한 변화는 대표적으로 [애딕티비(Adikteev)](https://www.adikteev.com/ko-kr/), [리머지(Remerge)](https://www.remerge.io/ko/), [애피어(Appier)](https://www.appier.com/ko-kr/), [몰로코(Moloco)](https://www.moloco.com/ko?utm_term=%EB%AA%B0%EB%A1%9C%EC%BD%94&utm_campaign=KR+-+Brand+-+KOR&utm_source=adwords&utm_medium=ppc&utm_geography=kr&utm_product=brand&hsa_acc=6399580300&hsa_cam=20024536299&hsa_grp=147989426083&hsa_ad=699454143028&hsa_src=g&hsa_tgt=kwd-1395917019720&hsa_kw=%EB%AA%B0%EB%A1%9C%EC%BD%94&hsa_mt=b&hsa_net=adwords&hsa_ver=3&gad_source=1&gclid=CjwKCAjwktO_BhBrEiwAV70jXsPTR8BEvyBOtuJbrnvhq2qNv5vFXTvI8aezr6DYMnIzYa_tOzvMmRoCoPoQAvD_BwE) 등이 제공하는 최첨단 리타게팅 솔루션이 이끌고 있어요. 게임 마케터들은 위 솔루션들을 활용해 딥링크를 기반으로 효과적인 리인게이지먼트 전략을 펼치고 있어요.
- 이벤트 프로모션: 딥링크로 플레이어를 특정 게임 내 이벤트나 시간 제한 챌린지로 바로 연결해 참여율을 높여요.
- 리인게이지먼트 캠페인: 딥링크가 포함된 푸시 알림 또는 이메일로 휴면 플레이어에게 신규 콘텐츠 또는 맞춤화된 혜택을 제공해 재참여를 유도해요.
- 광고 캠페인 최적화: 개발자가 부여한 딥링크를 통해 어떤 광고가 설치 및 게임 내 행동으로 이어지는지 추적할 수 있어 마케팅 비용을 보다 효율적으로 집행할 수 있어요.
> “(애딕티비는) 리인게이지먼트 캠페인을 통해 주요 캐주얼 게임 회사들이 LTV를 극대화하고 이탈률을 줄일 수 있도록 지원해왔습니다. 2025년에도 딥링크가 이러한 캠페인에서 핵심 역할을 할 것으로 기대됩니다.”
> — [자비에르 마리아니(Xavier Mariani)](https://www.linkedin.com/in/xaviermariani/), [애딕티비](https://www.adikteev.com/ko-kr/) CEO
💡 [Two and a Half Gamers 게임 팟캐스트를 통해 하이브리드 & 하이퍼 캐주얼 게임의 필승 성장 전략을 알아 보세요](https://www.airbridge.io/ko/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives)
## 에어브릿지 딥링크 플랜으로 오가닉 앱 성장을 견인하세요
딥링크는 오가닉 성장, 웹투앱 전환, 제휴 마케팅을 두루 관통하는 2025년 마케팅 전략의 핵심 기술이에요. 다시금 급부상하는 딥링크를 에어브릿지와 함께해 앱 마케팅을 성공 궤도에 올리세요.
### 2조 5천억 번의 클릭으로 이어진 매끄러운 연결성
에어브릿지 유저는 [2024년 한 해 동안 8천6백만 개 이상의 트래킹 링크](https://www.airbridge.io/ko/blog/airbridge-recap-2024)를 생성했어요. 생성된 에어브릿지 트래킹 링크는 매달 평균 200억 번, 1년간 총 2조 6천억 번 이상의 클릭을 기록하며 놀라운 활용성을 보여주었는데요. 글로벌 URL 단축 서비스를 제공하는 타 업체가 발표한 월평균 100억 번의 링크 클릭 수보다 2배 이상 높은 수치예요.
에어브릿지 트래킹 링크는 단순 링크 생성을 넘어 모든 플랫폼과 운영 체제에서 딥링크 기능과 더불어 끊김 없이 매끄러운 탐색 기능을 제공해요.
- 에어브릿지 MMP를 이미 사용하고 있다면, 별도의 추가 비용 없이 에어브릿지 딥링크 플랜에 포함된 모든 기능을 활용할 수 있어요. 에어브릿지 REST API를 통해 무제한의 트래킹 링크를 자동으로 생성해 보세요.
- 합리적인 가격대의 독립형 딥링크 솔루션을 찾고 계신가요? 에어브릿지 딥링크 플랜을 사용해 보세요. [에어브릿지 딥링크 플랜, Firebase 다이나믹 링크를 대체할 가장 완벽한 해답이에요](https://www.airbridge.io/ko/blog/introduce-deeplink-plan).
이미 전 세계 1,000개 이상의 앱이 에어브릿지 딥링크 플랜 및 측정 솔루션을 활용해 성장하고 있어요. 오가닉 성장 촉진, 바이럴 강화, 그리고 마케팅 채널 통합으로 도약할 준비가 되셨나요? 딥링크 솔루션을 통해 웹 유저를 충성도 높은 앱 유저로 전환하고 싶으신가요? 지금 바로 에어브릿지와 이야기 나누어 보세요.
---
## GA4 vs 에어브릿지: 앱 성과 측정에 MMP가 반드시 필요한 이유
URL: https://www.airbridge.io/ko/blog/google-analytics-vs-airbridge
Language: Korean
Published: 2025-03-24
2023년 7월, 기존 구글 애널리틱스(Universal Analytics) 지원이 종료되면서 많은 기업이 GA4로 마케팅 성과 측정 및 분석을 하고있어요. 하지만 성공적인 모바일 앱 마케팅을 위해 GA4만으로는 부족해요. 앱 마케팅 성과를 정확히 측정하고 최적화하기 위해 에어브릿지와 같은 MMP(Mobile Measurement Partner) 사용이 필수예요.
GA4는 웹 중심 분석 툴이기 때문에 앱 설치, 전환, 리인게이지먼트 캠페인을 정밀하게 측정하는 데 한계가 있어요. 반면 에어브릿지는 앱과 웹 통합 광고 성과를 측정하는 전문 솔루션이에요. 성과 분석 대시보드를 지원하는 것은 물론, 광고 사기 방지 등 맞춤형 기능으로 정확하게 앱 퍼포먼스를 측정해 마케팅 캠페인 극대화를 위한 맞춤 분석 및 최적화를 지원해요.
웹 캠페인만을 운영한다면 GA4만으로도 충분할 수 있지만, 앱 마케팅을 함께 진행한다면 성과 분석과 최적화에 부족함을 느낄 거예요. 성공적인 앱 마케팅을 위해 에어브릿지를 반드시 사용해야 하는 이유를 하나씩 살펴볼게요.
## 머신러닝 최적화에 필수적인 포스트백 기능
> ❌ **GA4: 포스트백 기능을 지원하지 않아요.
> ✅ 에어브릿지: 포스트백 기능을 지원해 메타, 구글 애즈, X 등 주요 광고 채널의 머신러닝 최적화를 실행해요.**
메타, 구글 애즈, 틱톡과 같은 주요 광고 채널은 MMP로부터 받은 포스트백 데이터를 활용해 머신러닝 최적화를 수행해요. 포스트백은 앱에서 발생한 광고 성과와 유저 행동 관련 데이터를 담고 있어요. 특히 메타, 구글 애즈, X(구 트위터), 스냅챗과 같은 SAN(Self-Attributing Network)이 아닌 매체에서는 에어브릿지가 필수예요. 이들 매체는 포스트백으로 데이터를 받아야만 광고 최적화가 가능하거든요.
반면 GA4에는 포스트백 기능이 없어요. 광고 매체가 앱에서 발생한 광고 성과나 유저 행동 데이터를 수집하지 못해 최적화가 불가능해요. 에어브릿지와 같은 MMP 없이 포스트백을 전송하려면 광고 채널이 제공하는 SDK를 별도로 설치해야 해요. 여러 개의 매체를 운영한다면 유지보수와 개발 리소스가 부담이 될 수 있어요.
커플 다이어리 서비스 ‘썸원(Sumone)’을 운영하는 모니모니는 에어브릿지의 포스트백 기능을 활용해 운영 리소스를 대폭 줄이고 효율적인 마케팅 성과 분석 워크플로우를 구축했어요. 덕분에 [기존에 MMP를 사용하지 않을 때에 비해 CPI 성과를 10배 이상 개선했어요.](https://cshub.ab180.co/ko/case-studies/airbridge-monymony-report#q-%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80-%EB%8F%84%EC%9E%85-%ED%9B%84-%EA%B4%91%EA%B3%A0-%EC%84%B1%EA%B3%BC%EA%B0%80-10%EB%B0%B0-%EC%9D%B4%EC%83%81-%EA%B0%9C%EC%84%A0%EB%90%98%EC%97%88%EB%8B%A4%EA%B3%A0-%EB%A7%90%EC%94%80%EC%A3%BC%EC%85%A8%EB%8A%94%EB%8D%B0-%EC%9D%B4%EB%A5%BC-%EA%B0%80%EB%8A%A5%ED%95%98%EA%B2%8C-%ED%95%9C-%EC%A3%BC%EC%9A%94-%EC%9B%90%EC%9D%B8%EC%9D%B4-%EB%AC%B4%EC%97%87%EC%9D%B8%EA%B0%80%EC%9A%94)
## 앱 캠페인 어트리뷰션
[__wf_reserved_decorative]
웹 투 앱 어트리뷰션
> **❌GA4: 앱 설치와 전환 데이터를 제대로 반영할 수 없어 성과를 정확히 측정하기 어려워요.
> ✅에어브릿지: 커스텀 어트리뷰션 윈도우 기능과 딥링킹 기술로 캠페인 성과를 정확하게 측정할 수 있어요.**
앱 설치 및 전환 성과를 정확하게 측정하는 것은 성공적인 앱 마케팅의 핵심이에요. 서비스와 마케팅 환경에 최적화된 어트리뷰션을 위해 에어브릿지 대시보드에서 [어트리뷰션 윈도우를 설정](https://help.airbridge.io/ko/guides/setting-attribution-window)할 수 있어요. 이를 통해 특정 시간 내 발생한 전환에 어떤 광고 채널이 기여했는지 명확하게 파악할 수 있어요.
뿐만 아니라 에어브릿지는 웹투앱(Web-to-App) 전환을 효과적으로 측정할 수 있는 딥링킹(Deep Linking) 기술을 지원해요. 예를 들어 사용자가 브랜드의 웹사이트에서 앱 설치 페이지로 유입되는 경우, 웹과 앱 사이의 연결을 끊김 없이 이어주는 것이 중요해요. 특히 커머스, 금융, 헬스케어 업계에서는 웹과 앱을 함께 활용하는 마케팅 전략이 필수이기 때문에 정확한 트래킹이 가능한 에어브릿지가 더 적합해요.
반면 GA4는 웹 기반의 분석 도구라서 앱 설치와 전환 데이터를 제대로 반영하기 힘들어요. 특히 광고 채널별 어트리뷰션 비교나 딥링킹을 활용한 전환 경로 분석에는 한계가 있어요.
비대면 진료 앱 서비스를 제공하는 닥터나우는 에어브릿지를 활용해 정밀한 웹투앱 분석을 구현했어요. 에어브릿지의 딥링크와 Protected Attribution Window 기능을 통해 유입되는 유저를 상세히 구분해서 트래킹했고, 그 결과 [개발 리소스를 효율화하고 광고 전략을 최적화](https://cshub.ab180.co/ko/case-studies/airbridge-setdeeplinks-doctornow)할 수 있었어요.
## 광고 사기(Ad Fraud) 검증
> **❌ GA4: 광고 사기를 감지하는 기능이 없어요.
> ✅ 에어브릿지: 광고 사기를 감지하고 방지할 수 있어요. 불필요한 지출을 줄이고 실제 성과를 기반으로 캠페인을 최적화할 수 있어요.**
모바일 광고 시장에서는 클릭 스팸, SDK 스푸핑과 같은 광고 사기(Ad Fraud)가 빈번하게 발생해요. 광고사기를 방지하지 않으면 실제 성과보다 부풀려서 측정된 데이터로 인해 불필요한 비용이 지출될 수 있어요.
GA4는 광고 사기를 감지하는 기능이 없어서 광고 성과를 분석할 때 정확하지 않은 데이터가 포함될 가능성이 있어요. 반면 에어브릿지는 광고 사기 검출 및 방지 기능을 제공해요. 에어브릿지에서 광고 사기 검증 규칙(Fraud Validation Rules)을 설정하면, 특정 기준에 맞지 않는 데이터를 자동으로 필터링할 수 있어요. 설정한 규칙을 따르지 않는 데이터는 프리벤션 레벨에 따라 광고사기로 표기되거나 성과 측정에서 제외돼요.
베트남 1위 숙박 앱을 운영하는 고투조이(Go2Joy)는 에어브릿지를 도입하고 [최대 90%의 허위설치 건수를 제거하는 동시에 실제 설치 건수를 10% 증가시켰어요](https://www.airbridge.io/ko/case-studies/go2joy-fraud-prevention-detection). AI를 활용해 도출한 유저 레벨 원본 데이터를 제공하는 에어브릿지의 비정상 설치 리포트를 자사 서비스와 조합해 비정상적인 활동을 신속하게 감지하고 처리할 수 있었어요.
에어브릿지를 활용하면 불필요한 광고비 지출을 줄이고, 실제 성과를 기반으로 캠페인을 최적화할 수 있어요. [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/fraud-validation-rules)에서 어떻게 광고 사기를 감지하고 정확한 성과를 측정할 수 있는지 더 자세히 알아보세요.
## 리어트리뷰션 및 리인게이지먼트
> **❌ GA4: 리어트리뷰션 트래킹이 제한적이에요.
> ✅ 에어브릿지: 유저 페르소나 설정과 오디언스 연동을 통해 효과적인 리어트리뷰션, 리인게이지먼트 캠페인을 운영할 수 있어요.**
앱 마케팅에서 이탈한 유저를 다시 앱으로 돌아오게 하는 것은 신규 유저를 유치하는 것만큼 중요해요. 에어브릿지는 리어트리뷰션(Reattribution)과 리인게이지먼트(Re-engagement) 캠페인을 효과적으로 운영할 수 있도록 지원해요. 리어트리뷰션 트래킹이 제한적인 GA4에 비해 효과적으로 이탈 유저를 다시 유입시키거나 리타겟팅 캠페인 성과를 분석할 수 있어요.
- 비활성 유저 페르소나 설정: 특정 기간 동안 앱을 사용하지 않은 유저를 정의하고, 해당 유저군에 최적화된 마케팅 전략을 실행할 수 있어요. 예를 들면, 지난 30일간 앱에 접속하지 않은 유저나 지난 60일간 특정 이벤트(구매, 결제 등)를 수행하지 않은 유저 등을 타겟팅할 수 있어요.
- 오디언스 연동 및 타겟팅: 에어브릿지는 카카오, 네이버, 구글, 메타 등 주요 광고 매체와 오디언스를 연동할 수 있어요. 에어브릿지에서 생성한 오디언스를 광고 채널로 전송해 리타겟팅(재참여 유도) 또는 디타겟팅(광고 제외) 캠페인을 운영할 수 있어요.
[에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/audience-manager)에서 새로운 오디언스를 만들고 광고 채널로 전송해 타겟 광고를 최적화하는 방법에 대해 더 자세히 알아보세요.
## 멀티 터치 어트리뷰션(MTA)과 예측 LTV(pLTV)를 비롯한 심화 분석
[__wf_reserved_decorative]
에어브릿지 터치포인트 오버랩 리포트
> **❌ GA4: 여러 광고 채널에서 발생하는 터치포인트 데이터를 통합 분석할 수 없어서 멀티 터치 어트리뷰션이나 pLTV 분석을 활용하기 어려워요.
> ✅ 에어브릿지: 멀티 터치 어트리뷰션, pLTV와 같은 심화 분석을 지원해 더 정밀한 분석이 가능해요.**
유저가 전환에 이르는 전체 여정을 파악하면 더욱 효율적인 앱 마케팅 전략을 세울 수 있어요. 에어브릿지는 멀티 터치 어트리뷰션(MTA, Multi-Touch Attribution)을 지원해 유저가 전환되기까지 어떤 광고 채널이 기여했는지 더 깊이 분석할 수 있어요.
에어브릿지의 터치포인트 오버랩 리포트(Touchpoints Overlap Report)는 유저가 전환되기 전까지 어떤 채널을 거쳤는지 분석해요. 이를 통해 광고 채널 간 중복 노출을 파악하고, 효율적인 미디어 믹스를 구성하는 데 도움을 줄 수 있어요. 또한 Day-N Retention, Day-N ROAS, pLTV와 같은 심화 분석 리포트를 제공해요. 이를 활용하면 광고 성과를 보다 정밀하게 분석하고, ROI(투자 대비 수익)를 극대화할 수 있어요. 이외에 CPA, ROAS, tROAS 기반 캠페인을 진행할 때도 에어브릿지의 정밀한 데이터 분석을 활용하면 최적화된 광고 전략을 수립할 수 있어요.
반면 GA4는 사용자 접점 분석과 앱 내 성과 분석 기능이 제한적이에요. 특히 여러 광고 채널에서 발생하는 터치포인트 데이터를 통합적으로 분석하기 어렵기 때문에, 멀티 터치 어트리뷰션이나 pLTV 분석을 활용하기 어려워요.
글로벌 알람 앱 알라미를 서비스하는 딜라이트룸은 [에어브릿지의 pLTV 기능을 활용해 마케팅 전략을 최적화하고 비용을 절감했어요.](https://cshub.ab180.co/ko/case-studies/airbridge-pltv-delightroom) 딜라이트룸 사례 전문을 통해 앱 마케팅에 최적화된 에어브릿지의 pLTV 기능에 대해 자세히 알아보세요.
## GA4 vs 에어브릿지 한 눈에 비교하기
[__wf_reserved_decorative]
에어브릿지 vs 구글 애널리틱스 4 비교
## 모바일 앱 마케팅, 편하고 강력한 MMP 에어브릿지로 시작하세요
GA4는 웹 중심의 사용자 행동 분석 도구이다 보니 앱 마케팅을 위한 기능에는 한계가 있어요. 모바일 앱 마케팅을 운영하면서 성과를 정확하게 측정하고 전략을 최적화하려면, 앱 마케팅 에 최적화된 솔루션 활용이 필수에요.
앱 마케팅에 필요한 모든 기능을 갖춘 에어브릿지로 더 많은 가능성을 열어보세요. 가장 편하고 강력한 MMP 에어브릿지가 앱 성공을 도울게요.
---
## PC와 콘솔 어트리뷰션으로 크로스 플랫폼 게임 성공 이끌기
URL: https://www.airbridge.io/ko/blog/cross-platform-gaming-measurement
Language: Korean
Published: 2025-03-21
게임 업계의 전체 매출은 2024년 1,877억 달러에서 2027년 2,133억 달러까지 성장할 것으로 예상돼요. 이 중에서 PC와 콘솔 게임이 전체 시장의 51%를 차지해 앞으로도 지속적인 성장세를 이어갈 전망이에요. (출처: [Newzoo Global Games Market Report 2024](https://newzoo.com/resources/trend-reports/newzoos-global-games-market-report-2024-free-version))
게임 업계는 더 많은 유저를 확보하고 수익 구조를 다각화하기 위해 빠르게 변하고 있어요. 특히 게임 스튜디오는 모바일 유저층을 확대하는 동시에, PC와 콘솔 플랫폼에서 타이틀을 출시하며 크로스 플랫폼 게임 수요 증가에 대응하고 있어요.
왜 멀티 플랫폼 게임이 업계에서 새로운 표준으로 주목받고 있을까요? 핵심 배경을 하나씩 살펴 볼게요.
## 크로스 플랫폼 게임이 주목받는 이유
### 향상된 게임 경험에 대한 수요 증가
요즘 게이머들은 PC, 콘솔, 모바일을 넘나드는 원활한 게임 경험을 기대해요. 이러한 수요가 증가하면서 크로스 플랫폼 플레이(cross-platform play), 크로스 플레이(cross-play), 크로스 프로그레션(cross-progression) 같은 기능이 중요한 트렌드로 자리 잡았어요.
- **크로스 플랫폼**: 마인크래프트와 포트나이트처럼 여러 플랫폼에서 제공되는 게임들은 iOS, Steam, 닌텐도 스위치 등 다양한 디바이스에서 플레이할 수 있어요.
- **크로스 플레이**: 에이펙스 레전드와 콜 오브 듀티_:_ 워존 시즌 02같은 게임은 PC와 플레이스테이션 등 서로 다른 플랫폼의 유저들이 실시간으로 함께 플레이할 수 있도록 지원해요.
- **크로스 프로그레션**: 원신과 발더스 게이트 3처럼 안드로이드에서 PC로 전환해도 진행 데이터가 그대로 유지되는 기능을 제공해요.
이러한 기능을 적극적으로 도입하면 유저들에게 보다 유연하고 끊김 없는 게임 경험을 제공할 수 있어요. 플레이어의 몰입도와 유저 충성도 또한 높일 수 있는 전략이에요.
### 무료 플레이(Free-to-play, F2P) 모델의 확대
무료 플레이 모델은 더 많은 플레이어가 게임을 쉽게 접할 수 있도록 하는 동시에 새로운 수익 창출 기회를 만들어내고 있어요. 유저의 행동 패턴과 선호도를 반영해 인앱 구매를 최적화하면 인게이지먼트를 효과적으로 유지하면서 장기적인 수익을 확보할 수 있어요.
리그 오브 레전드, 에이펙스 레전드, 워프레임 같은 게임들은 F2P 모델을 성공적으로 도입해 유저 생애 가치(LTV)를 극대화하고, 유저 기반을 꾸준히 확장해 왔어요. 이제 게임 시장은 단순 패키지 판매를 넘어 크로스 플랫폼 환경에서 유저의 인게임 행동을 정확히 분석하는 것이 필수 요소가 됐어요. 이를 통해 게임의 수익성을 극대화하고, 유저 몰입도를 최적화하는 전략이 더욱 중요해지고 있어요.
### e스포츠 및 인플루언서 스트리밍 시장의 성장
트위치와 YouTube Gaming 같은 플랫폼은 게임을 단순히 플레이하는 것을 넘어 관전하는 문화로 확장시켰어요. 이제 인플루언서와 스트리머는 게임 홍보에 핵심적인 역할을 하면서 게임의 인지도와 인기에 직접적인 영향을 미치고 있어요. 개발사들도 스트리머가 고품질 콘텐츠를 제작할 수 있도록 지원하는 것을 중요한 마케팅 전략 중 하나로 고려해요.
발로란트, 오버워치, [퍼스트 디센던트](https://tfd.nexon.com/ko/main)와 같은 PC 게임은 모바일 게임보다 뛰어난 비주얼 퀄리티를 바탕으로 스트리밍 플랫폼에서 강한 존재감을 보이고 있어요. 개발사가 스트리머들이 게임을 더욱 쉽게 접할 수 있도록 크로스 플랫폼 플레이를 지원하면 인플루언서 중심의 커뮤니티 참여를 활성화할 수 있어요. 이를 통해 더 많은 시청자를 유입하고, 게임의 도달 범위를 더욱 확장할 기회를 얻을 수 있어요.
[__wf_reserved_decorative]
멀티 플랫폼 게임이 확대되면서 플레이어의 여정도 점점 더 복잡해지고 있어요. 유저 확보(UA), 인게이지먼트, 수익화 전략의 핵심이 되는 정확한 크로스 플랫폼 성과 측정이 중요해지는 이유에요.
## 정확한 크로스 플랫폼 트래킹이 어려운 이유
PC·콘솔 게임 마케팅은 단순히 유저를 확보하는 것에 그치지 않아요. 고가치 플레이어를 유입하고, 이들의 참여를 유지하며, 전환을 유도하는 과정까지 포함해요. 이를 효과적으로 실행하려면 유저가 게임을 발견하는 순간부터 인게임 구매에 이르기까지의 전체 여정을 정밀하게 추적할 수 있어야 해요.
하지만 PC·콘솔 게임의 성과를 정확하게 측정하는 것은 쉽지 않아요. 그 이유를 두 가지 말씀드릴게요.
### 다양한 유저 유입 경로
플레이어들은 검색 광고, 디스플레이 광고, 소셜 미디어, 스트리머 및 인플루언서 프로모션, 이메일 마케팅 등 다양한 경로를 통해 게임을 발견해요. 어느 경로에서 유저가 유입되고 있는지 성과를 정확하게 측정해야 어떤 채널이 타겟 오디언스에 효과적으로 도달하고, 참여를 유도하는지 파악할 수 있어요. 이를 기반으로 마케터는 전략을 최적화하고, 더 높은 마케팅 성과를 달성할 수 있어요.
### 복잡한 크로스 플랫폼 환경
플레이어들은 PC, 콘솔, 모바일 등 여러 플랫폼과 디바이스를 오가며 게임을 설치하고, 플레이하며, 인게임을 구매해요. 이 과정에서 발생하는 복잡한 크로스 플랫폼 흐름을 정확히 측정하는 것은 쉽지 않아요. 이제 마케터는 크로스 플랫폼 UA와 게임 내 시나리오를 적극적으로 추적해야 해요. 예를 들어, 플레이어가 모바일 광고를 클릭한 후 PC 게임을 설치하는 경우, 마케터는 이 이벤트의 효과를 정확히 측정하기 위해 해당 광고에 대한 어트리뷰션을 수행해야 해요. 또한, 첫 구매까지 걸리는 시간, 거래 금액, 이탈률 등의 주요 지표를 분석하면, 유저 맞춤형 경험을 제공하고 리타겟팅 전략을 최적화하는 데 필요한 인사이트를 얻을 수 있어요.
## 에어브릿지로 크로스 플랫폼 성과 측정하기
에어브릿지는 모바일, PC, 콘솔 플랫폼 전반에서 사용자 여정을 통합적으로 측정할 수 있는 크로스 플랫폼 트래킹 기능을 제공해요. UA부터 게임 내 구매까지의 전환 이벤트를 에어브릿지를 통해 수집하고, 이를 원활하게 통합하여 모든 디바이스에서 광고 캠페인을 최적화할 수 있어요. 또한, 멀티 플랫폼 게임에 전용 마케팅 랜딩 페이지가 있든 없든, 마케터가 성과를 정확하게 측정할 수 있도록 지원해요.
- **마케팅 랜딩 페이지가 있는 게임**: 포트나이트는 사용자가 게임 정보를 탐색하고 다운로드할 수 있는 전용 마케팅 웹사이트를 운영해요. 이 경우 사용자가 광고를 클릭하면 에어브릿지 웹 SDK가 설치된 랜딩 페이지로 리디렉션돼요. 에어브릿지에서는 이 웹사이트와 PC·콘솔 게임에서 발생하는 데이터를 수집하여, 광고 성과를 효과적으로 측정할 수 있어요.
[__wf_reserved_decorative]
- **마케팅 랜딩 페이지가 없는 게임**: 다키스트 던전은 스팀이나 GOG 같은 플랫폼에서 게임 정보를 제공하고 판매를 진행해요. 이런 경우 에어브릿지는 [트래킹 링크](https://help.airbridge.io/ko/guides/creating-a-new-tracking-link)를 활용해 여러 플랫폼에서 데이터를 수집하고, PC·콘솔 광고의 성과를 정밀하게 측정할 수 있도록 지원해요.
[__wf_reserved_decorative]
에어브릿지의 크로스 플랫폼 측정 기능을 활용하면 광고 클릭과 게임 내 행동을 효과적으로 연결해 사용자 여정을 명확하게 파악하고 캠페인 최적화를 극대화할 수 있어요.
### 핵심 지표 설정하기
높은 전환율과 유저 참여를 유도하려면 먼저 유저의 주요 행동을 정의하고 이를 정밀하게 추적할 수 있어야 해요. 특히 에어브릿지를 활용해 크로스 플랫폼 게임 마케팅을 최적화하려면 이 과정은 꼭 필요해요.
시작 단계에서는 아래와 같은 지표를 설정해두는 것이 좋아요. 이를 통해 유저의 플레이 진행 상황과 수익화 가능성을 보다 명확하게 파악할 수 있어요.
- **설치 완료(Install complete)**: 게임 설치가 완료된 시점
- **튜토리얼 완료(Tutorial complete)**: 유저가 온보딩 과정을 마친 시점
- **첫 로그인(First login)**: 유저가 처음으로 게임에 로그인한 시점
- **10레벨 달성(Level 10 achieved)**: 특정 마일스톤에 도달한 시점
- **첫 결제(First purchase)**: 유저가 최초로 인게임 결제를 완료한 시점
- **구독 결제(Subscription purchase)**: 반복 결제가 발생하는 구독형 상품 구매 시점
이러한 지표들을 체계적으로 추적하면 유저의 전환 흐름을 분석하고 퍼널 단계별로 성과를 개선할 수 있는 기반이 마련돼요.
### 서버 간 API(S2S) 연동하기
PC·콘솔 게임은 대부분 서버 기반 환경에서 운영되기 때문에 정확한 이벤트 트래킹을 위해 에어브릿지는 S2S(Server-to-Server) API 연동을 지원해요. 예를 들어, 플레이어가 첫 인게임 결제를 완료한 시점의 데이터를 API를 통해 에어브릿지로 전송하면, 마케터는 해당 전환이 어떤 광고에서 비롯된 것인지 명확히 분석할 수 있어요. 이를 통해 정확한 마케팅 어트리뷰션과 캠페인 최적화가 가능해요.
### 확률 기반 어트리뷰션 활용하기
에어브릿지는 확률 기반 어트리뷰션(Probabilistic Attribution) 기법을 사용해요. 이는 쿠키나 디바이스 ID 없이도 광고 클릭과 전환을 통계적으로 연결해주는 방식이에요. 광고 차단 기능과 개인정보 보호 규제가 강화되는 환경에서, 이 방식은 정확한 성과 분석을 제공하면서도 개인정보 보호 기준을 충족하는 데 매우 효과적이에요.
[에어브릿지의 크로스 플랫폼 측정 가이드](https://help.airbridge.io/ko/guides/pc-game-tracking-guide)에서 자세한 내용을 확인해 보세요.
## 에어브릿지 X 넥슨 크로스 플랫폼 트래킹 사례
[__wf_reserved_decorative]
넥슨은 [퍼스트 디센던트](https://www.nexongames.co.kr/game/first_descendant.php)의 글로벌 론칭 과정에서 에어브릿지를 활용해 PC·콘솔 유저 여정을 하나의 대시보드에서 통합적으로 추적했어요. 이 데이터를 바탕으로 새로운 광고 상품을 빠르게 실험하고, 가장 효과적인 채널과 캠페인 유형, 크리에이티브를 식별해 고가치 유저 확보를 위한 리소스를 전략적으로 분배했어요.
퍼스트 디센던트 팀이 성과를 낼 수 있었던 주요 전략은 다음과 같아요:
- **하단 퍼널 지표 기반 UA 성과 분석**: Google, Meta, TikTok, X, Reddit, Amazon Ads(Twitch) 등 다양한 플랫폼의 광고 성과를 하단 퍼널 지표를 기준으로 비교 분석해 각 광고 크리에이티브의 효율을 극대화했어요.
- **CAPI 연동을 통한 캠페인 최적화**: Meta와 X의 Conversion API(CAPI)를 활용해 PC·콘솔 인게임 전환 데이터와 광고 성과를 연결하고, 이를 기반으로 매끄럽게 캠페인을 최적화했어요.
- **고가치 유저 기반 타겟 마케팅**: 유저 생애 가치(LTV)가 높은 플레이어를 식별하고, 인게임 행동 데이터를 기반으로 맞춤형 타겟 마케팅 전략을 수립했어요.
넥슨의 정밀한 크로스 플랫폼 어트리뷰션 실행 전략이 궁금하다면 [케이스 스터디 전문](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)을 확인해 보세요.
## 👀 주요 내용 한눈에 보기
- 게임 스튜디오들은 더 넓은 유저층과 다양한 수익 모델을 확보하기 위해 모바일에서 PC·콘솔 플랫폼으로 확장하고 있어요.
- 이러한 변화는 크로스 플랫폼 게임 경험에 대한 수요 증가에 의해 더욱 가속화되고 있어요.
- 하지만 플레이어가 다양한 UA 채널을 통해 여러 디바이스와 플랫폼을 넘나들기 때문에, 정확한 크로스 플랫폼 측정은 여전히 큰 과제예요.
- 에어브릿지는 모바일, PC, 콘솔 플랫폼 전반의 유저 여정을 통합적으로 시각화할 수 있는 크로스 플랫폼 트래킹을 제공해요. UA부터 인게임 구매까지 모든 여정을 한눈에 확인할 수 있어요.
- 또한, S2S API와 확률 기반 어트리뷰션을 통해 개인정보를 보호하면서도 정확한 마케팅 성과 분석이 가능해요.
이제 에어브릿지에서 정밀한 크로스 플랫폼 트래킹을 경험하고 게임 마케팅을 한 단계 더 성장시켜 보세요!
---
## AI 검색엔진의 종류와 ChatGPT, DeepSeek 트래픽 유도 및 측정 방법
URL: https://www.airbridge.io/ko/blog/ai-search-traffic-strategy
Language: Korean
Published: 2025-03-20
생성형 AI의 등장과 함께 온라인 정보 검색 트렌드가 큰 변화를 겪고 있어요. 2024년부터 [대규모 언어 모델(LLM)](https://www.airbridge.io/ko/blog/what-is-llm)을 기반으로 한 생성형 AI가 주목받기 시작했는데요. 이러한 흐름에 힘입어 다양한 AI 검색 엔진 서비스가 등장하고 있어요. AI 검색엔진은 빠르게 기존 검색엔진의 자리를 차지하고 있죠.
Perplexity AI, OpenAI의 ChatGPT와 SearchGPT, 그리고 2025년 시작을 뜨겁게 달군 DeepSeek까지. 다양한 서비스를 바탕으로 빠르게 보편화되고 있는 AI 검색엔진은 자연스럽게 많은 앱들의 새로운 유저 유입구가 되고 있어요.
AI 검색엔진에는 어떤 서비스들이 있고 마케터는 AI 검색을 통한 새로운 유입에 어떻게 대비해야 할까요? 이번 블로그 포스트를 통해 이 모든 내용을 확인해 보세요. 👇
## AI 검색엔진 종류에는 무엇이 있을까요?
앞서 이야기한 것처럼 생성형 AI 검색엔진은 이제 새로운 트래픽 유입 창구 중 하나에요. 최근 온라인 마케터들 사이에서는 검색엔진 최적화인 SEO를 넘어 생성형 검색 최적화(GEO)라는 신조어가 떠오르고 있을 만큼 AI 검색엔진의 점유율은 빠르게 증가하고 있죠. 2025년 현재, 주요 AI 검색엔진은 무엇이 있고 각 서비스들은 어떠한 특징을 가지고 있을까요?
### **Perplexity AI**
AI 검색 시대 포문을 연 [Perplexity AI](https://www.perplexity.ai/)는 2022년 12월 처음 등장했어요. 검색 키워드에 맞는 다양한 웹 페이지를 나열해 주는 링크 기반 검색엔진이 아닌 출처 표기를 기반으로 한 인용을 통해 보고서 형태로 검색 결과를 제공하며 AI 검색엔진의 원형을 만들었죠.
### **ChatGPT**
대화형 AI로 제작된 [ChatGPT](https://chatgpt.com/)는 [OpenAI](https://openai.com/)의 생성형 AI 서비스로 사람과 비슷한 답변을 빠르게 제공하며 가장 대중적인 AI 모델로 평가받고 있어요. ChatGPT는 시간이 지나면서 정보 검색, 고객 지원 등 다양한 분야로도 활용되고 있어요.
#### SearchGPT 프로토타입
초기 ChatGPT는 제한된 학습 데이터를 활용해 답변을 제공해 왔기 때문에 정보 탐색의 용도로는 적절하지 않았어요. OpenAI는 이러한 단점을 극복하기 위해 검색 기술과 GPT의 요약 기능을 결합한 새로운 프로토타입 [SearchGPT](https://openai.com/index/searchgpt-prototype/)를 공개했어요. Search GPT는 외부 소스 링크를 포함하여 더 구체적이고 직접적인 검색 결과를 제공해요.
### Liner AI
[Liner AI](https://getliner.com/ko)는 2015년 하이라이팅 및 콘텐츠 요약 도구로 시작되었으며, 이후 AI 기반 검색 기능을 강화하면서 AI 검색엔진으로 발전했어요. 기존 키워드 기반 검색을 넘어 사용자의 검색 의도를 분석하고, 웹에서 신뢰할 수 있는 정보를 선별하여 요약 제공하는 기능을 갖추고 있어 많은 사랑을 받고 있죠.
### **구글 서치 랩스(Search Labs), SGE(생성형 AI 검색)**
전통적인 검색엔진 중 하나인 구글에서도 AI 검색 시대에 빠르게 대응하고 있어요. 이제 구글 검색 시에도 [서치랩스(Search Labs)](https://support.google.com/websearch/answer/13572013?hl=ko&co=GENIE.Platform%3DDesktop)와 [생성형 AI 검색(SGE)](https://blog.google/intl/ko-kr/products/explore-get-answers/google-search-generative-ai-korea-expansion-kr/) 기술을 통해 검색 결과에서 단순한 페이지 나열이 아닌 AI가 직접 탐색하고 정리한 내용을 확인할 수 있죠. 실제로 구글에서는 [점차 더 많은 지역에서 더 많은 언어로 더 많은 사용자에게 AI를 통한 검색 결과 개요를 제공할 예정](https://support.google.com/websearch/answer/14901683?hl=ko)이라고 밝히고 있어요.
### **DeepSeek**
AI 검색엔진 시장에 새로운 서비스로 떠오르고 있는 [DeepSeek](https://www.deepseek.com/)는 2025년 1월 ‘DeeSeek-R1’을 선보이며 AI 검색엔진 시장에 본격적으로 등장했어요. 경쟁사보다 저렴한 비용으로도 고급 AI 모델을 도입하며 많은 사용자에게 AI 검색 서비스를 제공하고 있죠. 기존 서비스들에 비해 저렴한 비용으로 운영되면서도, 뛰어난 언어 이해력과 논리적 추론 능력을 갖추고 있다는 게 DeekSeek의 가장 큰 특징이에요.
## AI 검색엔진 트래픽을 높이기 위한 콘텐츠 전략은 무엇일까요?
AI 검색엔진이 등장하며 AI 검색엔진을 통한 오가닉 유입 전략도 함께 진화하고 있어요. AI 검색엔진을 통해 발생하는 트래픽을 늘리기 위해서 마케터는 콘텐츠 기획과 제작 시 어떤 점들을 중요하게 생각해야 할까요?
### **질문에 집중하기**
AI 검색엔진을 사용할 때 유저는 "3월 서울에서 아이들과 함께 주말에 방문할 만한 곳들은 어디가 있나요?"와 같이 키워드가 아닌 직접적인 문장으로 질문을 입력하는 경우가 많아요. 이에 따라 콘텐츠를 제작할 때에도 사용자 질문에 답이 될 수 있는 Q&A 형태가 보다 중요해졌죠.
물론 모든 콘텐츠가 꼭 질문과 답 형태로만 구성되어야 한다는 뜻은 아니에요. 직접적인 Q&A 형태가 아니더라도 유저가 AI에게 질문할 수 있는 내용을 바탕으로 한 콘텐츠 기획이 필요한 것이죠.
### **전문성과 신뢰도 높이기**
[자연어 처리(NLP)](https://cloud.google.com/learn/what-is-natural-language-processing?hl=ko) 기반 AI 검색엔진에게 콘텐츠가 더 잘 설득되게 하기 위해서는 브랜드와 콘텐츠의 전문성과 신뢰도를 높이는 일이 중요해요.
기존 검색엔진 최적화는 키워드를 기반으로 한 작업이었다면 AI 검색엔진은 글 전체가 얼마나 유기적으로 작성되었고, 신뢰할 수 있는 정보를 제공하는지를 중요하게 보기 때문이죠. 이에 따라 믿을 수 있는 출처를 추가하거나 더 높은 전문성을 활용해 콘텐츠를 작성하는 것이 중요해요.
### **멀티 모달 콘텐츠 활용하기**
AI 검색엔진은 단순히 텍스트로 제작된 콘텐츠만 검색하지 않아요. [멀티 모달 기술](https://www.ibm.com/kr-ko/think/topics/multimodal-ai)을 통해 영상, 이미지, 음성 등 다양한 종류의 콘텐츠를 깊이 있게 탐색할 수 있죠.
이에 따라 다양한 콘텐츠 포맷 제작이 중요해지고 있어요. 같은 내용이더라도 글 한 편이 아닌 영상, 팟캐스트 등 다채로운 포맷으로 제작할 수 있다면 AI 검색엔진의 신뢰를 보다 빠르게 얻을 수 있어요.
## 에어브릿지와 함께 AI 검색엔진 트래픽 추적하기
지금까지 AI 검색엔진의 선택을 받기 위한 콘텐츠 제작 방법을 살펴보았다면 이제는 우리 앱으로 얼마나 많은 사람들이 유입되고 있는지를 알아야겠죠.
에어브릿지는 오가닉 리퍼러 어트리뷰션을 통해 AI 검색 유입 측정을 제공하고 있어요. 또한 단순히 트래픽만이 아닌 AI 검색을 통해 방문한 유저들이 어떻게 전환되고 있는지 등 유입부터 전환까지 전체 유저 퍼널을 함께 측정할 수 있죠.
### 에어브릿지 오가닉 리퍼러 어트리뷰션이란?
[오가닉 리퍼러 어트리뷰션](https://help.airbridge.io/ko/guides/organic-referrer-attribution) 기능은 식별 가능한 광고 캠페인이 아닌 오가닉 유입 유저 리퍼러 정보 또한 확인할 수 있게 도와줘요. 에어브릿지 오가닉 유입 리퍼러 어트리뷰션에서는 크게 아래와 같은 오가닉 유입 정보를 제공해요.
- **검색엔진 트래픽**: 검색엔진(Google, Bing 등)에서 유입되는 유저.
- **AI 검색엔진 트래픽**: AI 서비스에서 웹사이트 링크를 통해 유입되는 유저.
- **기타 백링크 트래픽**: 다양한 출처(리스트 기사, 커뮤니티 언급 등)에서 유입되는 유저.
예를 들어, ChatGPT를 통한 유입의 경우 리퍼러 소스가 [https://chatgpt.com](https://chatgpt.com)로 표시되며 해당 트래픽의 유입 출처를 확인할 수 있는 것이죠.
> 💡 바잇미가 에어브릿지 오가닉 리퍼러를 활용해 키워드와 매체를 최적화한 사례가 궁금하다면?
> [👉 CS HUB 사례 확인하기](https://cshub.ab180.co/ko/case-studies/airbridge-biteme-organic-referrer-attribution)
## 에어브릿지와 함께 AI 검색엔진 유입 유저 퍼널을 추적하세요.
AI 검색엔진을 통한 웹 방문 이외에도 에어브릿지는 [퍼널 리포트](https://help.airbridge.io/ko/guides/funnel-report-viewing)를 통해 웹에서 앱으로 전환하는 유저를 추적하여 앱 설치와 이후의 앱 내 이벤트 전환을 확인할 수 있어요. 마케터는 AI 검색엔진을 통해 유저가 어떤 콘텐츠로 유입되는지 뿐만 아닌 해당 유저들이 어떻게 행동하며 앱 설치 및 구매로 전환되고 있는지를 데이터로 확인할 수 있는 것이죠.
다양한 AI 검색엔진이 등장하며 광고 이외에도 유저 유입 경로가 빠르게 늘어나고 있어요. 에어브릿지와 함께 AI 검색엔진 시대에 대비해 보세요.
---
## 광고 수익화와 유저 획득의 균형: pLTV를 활용한 앱 비즈니스 최적화 전략
URL: https://www.airbridge.io/ko/blog/pltv-optimization-for-apps
Language: Korean
Published: 2025-03-20
_본 글은 세계 1위 알람 앱 '알라미'를 운영하는 딜라이트룸의 앱 광고 수익화 전문가 그룹 DARO에서 기고한 글입니다._
## **앱 비즈니스의 영원한 딜레마: 수익과 사용성**
모바일 앱 비즈니스에서 가장 큰 고민 중 하나는 광고 수익과 사용자 경험 사이의 균형입니다. 새로운 광고 지면을 추가하면 수익은 늘어날 수 있지만, 사용자 경험이 저하되어 리텐션이 떨어질 수 있습니다. 이런 고민은 특히 무료 앱을 운영하는 많은 개발사들에게 끊임없는 도전 과제입니다.
세계 1위 알람 앱 '알라미'를 운영하며 8000만 명 이상의 사용자에게 서비스를 제공해온 딜라이트룸도 이러한 고민에서 자유롭지 않았는데요. 저희는 이런 고민 끝에 **리텐션과 LTV(Lifetime Value)가 광고 수익화의 핵심**이라는 것을 알게 되었습니다.
## **왜 광고 LTV가 중요한가?**
일반적으로 IAP(In App Purchase, 인앱 구매) 모델이나 이커머스와 같이 실제 구매 행동에 대한 가치가 즉각적으로 드러나고 횟수가 제한적인 수익화 방식에서는 ARPU, ARPPU로 대표되는 객단가 지표가 중요합니다. 이 객단가를 가지고, 유저를 유입시키는 비용과 비교해서 유닛 이코노믹스를 고려한 그로스가 가능하기 때문이죠.
그러나, 광고 수익화의 경우 이것이 꽤나 어렵습니다. 왜냐면 광고 수익은 개별 금액이 크지 않고, 유저의 꾸준한 접속과 광고 노출을 통해 발생하기 때문에 객단가를 즉시 산출하기 어렵습니다. 이로 인해 LTV라는 개념을 사용할 수 밖에 없는데요.
LTV는 사용자가 앱을 사용하는 전체 기간 동안 창출하는 가치를 의미합니다. 광고 수익화 관점에서 LTV는 다음과 같이 계산할 수 있습니다.
> **광고 LTV = Lifetime(일) × 일별 광고 수익(AD ARPDAU)**
이 공식은 단순해 보이지만, 광고 수익화와 사용자 경험의 관계를 한 지표로 통합하여 보여줍니다. 일별 광고 수익이 아무리 높아도 Lifetime이 길지 않다면 LTV는 낮을 수밖에 없죠. 이 Lifetime과 일별 광고 수익은 어느정도 상충관계에 있을 수 밖에 없습니다. 광고 수익을 높이기 위해 광고를 많이 보여줄 수록 유저의 경험은 부정적으로 바뀌게 되기 때문입니다.
따라서, LTV를 계산할 수 있어야, 수익과 Lifetime, 즉 리텐션의 균형을 측정할 수 있습니다. 광고 실험을 통해 수익이 10% 증가했지만 7일 리텐션이 3% 감소했다면 총 수익은 어떻게 변화했을까요? LTV를 통해 이 상충관계를 정량적으로 평가할 수 있고 균형을 맞추면서도 총 수익을 극대화하는 방법을 찾을 수 있습니다.
## **예측 LTV(pLTV)의 중요성**
하지만 광고 LTV 계산에는 한 가지 큰 문제가 있습니다. Lifetime(생애주기)을 정확히 측정하는 것이 어렵다는 점입니다. 사용자의 전체 생애주기를 측정하려면 몇 개월, 때로는 몇 년을 기다려야 할 수도 있습니다. 이런 사후적인 데이터로는 즉각적인 액션을 취할 수가 없기 때문에, 유저 획득 캠페인의 지표로 사용할 수 없습니다.
이런 문제를 해결하기 위해서는 예측 LTV 모델링이 필수적입니다. 예측 LTV는 초기 리텐션 데이터를 기반으로 향후 리텐션 패턴을 예측하여 전체 생애주기를 추정하는 것을 말합니다. 이렇게 예측 LTV를 계산하기 위해서는 현재 리텐션 데이터가 어떤 함수나 모델과 유사한지, 그리고 그것을 얼마나 잘 예측해줄 수 있는지 찾아보는 것이 중요한데요.
## **Airbridge의 pLTV 기능: 빠르고 간편한 예측으로 의사결정 최적화**
이런 상황에서 Airbridge의 pLTV(predictive LTV, 예측 LTV) 기능이 특히 가치를 발휘합니다. Airbridge는 이미 SDK를 통해 수집된 앱 데이터와 광고 수익 데이터를 결합하여 빠르게 리텐션 예측 모델을 구축합니다.
Airbridge의 pLTV 모델링은 다음과 같은 강점이 있습니다.
1. **다양한 곡선 피팅 알고리즘**: 앱마다 리텐션 패턴이 다르기 때문에, Airbridge는 여러 수학적 모델을 사용하여 다양한 곡선 피팅 알고리즘을 갖고 있으며 빠르고 정확한 예측을 제공합니다.
2. **초기 데이터로 장기 예측**: 최근 13일 간 데이터만으로도 D180까지의 리텐션을 간편하게 예측할 수 있어, 빠른 의사결정이 가능합니다.
3. **실시간 데이터 통합**: 일별 리텐션 및 ARPDAU 데이터를 실시간으로 통합하여 항상 최신 LTV 예측을 제공합니다.
## **성공 사례: 알라미의 광고 수익화 균형 찾기**
[__wf_reserved_decorative]
알라미 x 에어브릿지 대시보드
딜라이트룸은 알라미에 대한 광고 수익화를 하면서, 유저 경험과 광고 수익화의 균형을 찾기 위해 계속해서 노력해왔습니다. 시간에 따라 변화하는 LTV를 계산하기 위해 지속적으로 데이터 분석 리소스를 활용해왔는데, Airbridge에서 pLTV 기능을 제공하여 빠르고 간편하게 광고 수익화와 UA에서의 균형을 찾았습니다.
- 유저 획득 비용(CAC) 대비 LTV 비율(LTV:CAC)을 기준으로 마케팅 채널을 최적화
- 국가별 pLTV 데이터를 분석하여 고수익 시장에 광고 예산을 집중함으로써 ROAS 개선
- 새로운 광고 전략의 장기적 영향을 빠르게 예측하여 수익과 사용자 경험의 최적 균형점 발견
## **Airbridge와 DARO를 함께 사용해 광고 LTV를 극대화하세요**
딜라이트룸은 알라미를 서비스하면서 광고 수익화와 유저 경험 사이의 균형점을 찾기 위해 수많은 실험과 데이터 분석을 진행해왔습니다. 이 과정에서 얻은 인사이트들은 비단 알라미뿐만 아니라 다양한 앱들에도 적용할 수 있는 값진 자산이 되었습니다. 이와 같이 저희 딜라이트룸은 알라미를 비롯하여, 10년이 넘게 쌓인 광고 수익화 노하우를 바탕으로 DARO라는 앱 광고 수익화 솔루션을 런칭하여 다양한 앱 개발사의 광고 수익화를 돕고 있습니다.
[__wf_reserved_decorative]
DARO 광고 수익화 대시보드
애드몹과 같은 미디에이션 플랫폼을 별도로 활용하면서 다양한 네트워크 SDK를 붙일 필요 없이, DARO SDK 하나로 광고 수익이 극대화되기 때문에 프로덕트의 그로스 자체에 더 집중할 수 있는 환경을 제공하며, Airbridge의 pLTV 데이터를 결합하면 더욱 강력한 시너지를 발휘합니다.
광고 수익화는 DARO에 맡기면서 Airbridge를 통해 pLTV를 계산하고, 유저의 리텐션 개선에 집중한다면 광고 매출을 극대화하는 지름길이 될 수 있습니다.
특히, 앱 비즈니스에서 광고 수익화와 사용자 경험의 균형을 맞추는 것은 어려운 과제입니다. 그러나 Airbridge의 pLTV와 DARO의 광고 수익화 솔루션을 결합하면, 이 도전을 데이터 기반의 전략적인 의사결정으로 전환할 수 있습니다.
> DARO(다로)는 세계 1위 알람 앱 '알라미'를 운영하는 딜라이트룸의 앱 광고 수익화 솔루션입니다. 수억 명의 사용자를 대상으로 한 광고 수익화 노하우를 바탕으로, 앱 개발사들이 사용자 경험을 해치지 않으면서도 최적의 광고 수익을 창출할 수 있도록 지원합니다.
>
> 자세한 내용은 [DARO 웹사이트](https://daro.so/)를 방문하세요.
---
## 에어브릿지, 네이버 광고 오디언스 연동 시작
URL: https://www.airbridge.io/ko/blog/naver-ads-audience-sync
Language: Korean
Published: 2025-03-14
국내 대표 광고 플랫폼 네이버에서 고객 데이터를 보다 효과적으로 활용할 방법을 찾고 계셨다면, 주목할 소식이 있어요. 에어브릿지가 네이버 광고 오디언스 연동을 지원해요. 에어브릿지와 함께하는 마케터는 최신 고객 데이터로 더욱 정밀한 네이버 광고를 운영해 광고 성과를 극대화할 수 있어요.
## 네이버 광고 알아보기
네이버 광고는 국내 최대 포털 네이버의 다양한 플랫폼에서 광고를 운영할 수 있는 강력한 마케팅 도구예요. 검색광고(Search Ads), 디스플레이 광고(Display Ads), 쇼핑 광고 등 다양한 광고 상품을 제공하며, 네이버 데이터랩 및 AI 기반의 정교한 타겟팅 기술이 강점이에요.
특히, 네이버는 광고주의 다양한 니즈를 충족할 수 있는 맞춤형 광고 솔루션을 제공하고 있어요. 사용자의 검색 패턴과 소비 성향을 기반으로 한 정밀한 타겟팅이 가능하고, 네이버의 방대한 데이터베이스를 활용해 광고 효율을 극대화할 수 있죠.
이제 여기에 에어브릿지가 오디언스 연동을 지원하면서 더욱 강력한 광고 운영이 가능해졌어요.
## 네이버 광고 오디언스 연동으로 기대되는 점
[]
에어브릿지의 네이버 광고 오디언스 싱크는 퍼스트 파티 데이터(First-Party Data)를 활용해 정교한 고객 세그먼트를 생성하고, 이를 네이버 광고 플랫폼에 직접 전송할 수 있는 기능이에요.
에어브릿지가 지원하는 네이버 광고 오디언스 연동의 핵심 기능은 다음과 같아요
- 고객 행동 기반 맞춤 타겟팅: 에어브릿지에서 앱 설치, 회원가입, 구매 등 특정 조건을 만족하는 오디언스 세그먼트를 생성하고, 이를 네이버 광고 플랫폼에서 활용할 수 있어요. 이를 통해 더욱 정교한 맞춤형 광고가 가능해요.
- 실시간 데이터 동기화: 에어브릿지의 최신 데이터를 네이버 광고 플랫폼에 일일 단위로 자동 업데이트할 수 있어요. 이를 통해 광고 타겟팅이 최신 데이터에 기반해 최적화되며, 실시간으로 유저 데이터를 반영한 타겟팅이 가능해요.
이번 오디언스 연동으로 네이버의 강력한 광고 생태계를 활용해 더욱 정교한 타겟팅 광고를 운영할 수 있어요. 이를 통해 광고 예산을 더욱 효율적으로 운영하면서 유저 맞춤형 광고로 고객 경험을 개선할 수 있어요.
여기에 에어브릿지는[ 네이버 광고 캠페인·비용 연동](https://help.airbridge.io/ko/guides/naver-sa#%EC%BA%A0%ED%8E%98%EC%9D%B8-%EB%B9%84%EC%9A%A9-%EC%97%B0%EB%8F%99%ED%95%98%EA%B8%B0)을 지원하고 있어요. 네이버 광고에서 운영 중인 광고 캠페인 데이터, 지출한 광고 비용 데이터, 그리고 터치포인트 데이터까지 에어브릿지 대시보드에서 확인할 수 있어요.
에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 크리테오, 알티비하우스, 애피어, 리머지 등 다양한 매체와의 연동을 지원하고 있어요. 특히 이번 네이버 광고 연동과 함께 카카오, 토스, 당근, 에브리타임 등 국내 주요 광고 매체와의 연동을 지원해 독보적인 국내 매체 연동성을 갖춘 솔루션으로 자리 잡았어요.
광고 채널로 오디언스를 전송하는 자세한 방법은 [여기](https://help.airbridge.io/ko/guides/send-audience#%EC%98%A4%EB%94%94%EC%96%B8%EC%8A%A4%EB%A5%BC-%EC%A0%84%EC%86%A1%ED%95%A0-%EC%88%98-%EC%9E%88%EB%8A%94-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-1)를 참고해 주세요.
## 마무리하며
에어브릿지가 네이버 광고 오디언스 연동을 지원하면서 광고주는 더욱 정교하고 효과적인 캠페인 운영이 가능해졌어요. 특히 네이버의 방대한 데이터와 에어브릿지의 실시간 오디언스 싱크를 결합하면 광고 성과를 극대화하는 것은 물론, 운영 효율성까지 높일 수 있어요.
강력한 네이버 광고 운영을 누구보다 먼저 경험하고 싶다면, 지금 바로 에어브릿지의 오디언스 연동 기능을 활용해 보세요. 에어브릿지 데모를 신청하고, 한 단계 발전된 마케팅 전략을 직접 실행해 보세요.
---
## 모바일 앱 예측 LTV(pLTV) 계산하는 방법
URL: https://www.airbridge.io/ko/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app
Language: Korean
Published: 2025-03-06
경쟁이 치열해지는 모바일 앱 시장에서 예측 LTV(Predictive LTV, 이하 pLTV)는 예산 최적화와 장기적인 수익화를 위해 필수적으로 이해해야 할 요소에요. 유저 획득 비용이 계속 증가하는 상황에서 pLTV를 이해하고 마케팅에 활용하면 유저의 가치를 미리 파악해 지속 가능한 성장을 이끌 수 있어요.
> 💡 pLTV란?
> pLTV(Predictive Life Time Value, 고객 생애 가치 예측)는 유저가 모바일 앱이나 서비스를 이용하는 동안 예상되는 총 수익을 예측하는 지표예요.
## pLTV, 왜 중요할까?
- **투자 회수 기간 장기화**: 유저 획득 비용이 증가하면서 초기 광고 비용을 회수하는 데 더 오랜 시간이 걸려요. 유저의 장기적인 가치를 정확하게 예측하지 못하면 획득 전략을 최적화하기가 어려워요.
- **점진적인 수익 실현**: 많은 모바일 앱, 특히 비게임 앱은 인앱 광고나 구독을 통해 시간이 지나면서 수익을 창출해요. 기존의 단기 지표만으로는 유저의 진정한 가치를 파악하기 어려울 수 있어요.
- **실시간 예산 최적화**: 마케터는 예산을 효율적으로 배분하고, 데이터 기반의 신속한 의사 결정을 내려야 해요. 정확한 pLTV 예측을 통해 캠페인과 채널 전반에서 지출을 최적화하면서 투자 수익을 극대화할 수 있어요.
pLTV를 활용하면 단기 수익이 고객의 잠재적인 생애 가치를 충분히 반영하지 못하는 경우에도 선제적으로 유저 확보 및 유지 전략을 조정할 수 있어요.
## pLTV 계산 방법
pLTV를 계산하는 대표적인 방법으로 **커브 피팅(Curve Fitting)**과 **머신 러닝(Machine Learning)**이 있어요.
### 커브 피팅
커브 피팅 방식은 과거 리텐션 데이터를 활용해 수학적 곡선을 모델링하고, 이를 통해 향후 유저 리텐션과 수익 패턴을 예측하는 방식이에요.
> 작동 방식:
> 1\. 리텐션 곡선 분석: 리텐션 추세에 맞는 수학적 함수(예: 지수 함수, 거듭제곱 함수)를 선택하고, 특정 기간(예: 30일, 60일, 90일)에 걸쳐 이를 통합해 유저의 수명을 추정해요.
> 2\. ARPDAU 계산: 유저를 코호트(예: 국가, OS, 채널)별로 분류하고, 인앱 구매(IAP) 및 인앱 광고(IAA) 수익을 포함한 일일 활성 유저당 평균 수익(ARPDAU)을 산출해요.
> 3\. pLTV 계산: 통합된 리텐션 곡선과 ARPDAU를 곱해 누적 LTV를 추정해요.
장점:
- 유저 행동의 전반적인 추세를 분석하는 데 효과적이에요.
- 유저 활동(세션)과 매출 간의 상관관계가 높은 앱에 적합해요.
한계:
- 일부 유저가 높은 수익을 창출하는 구조(IAP 중심 게임 등)에서는 적절하지 않을 수 있어요.
- 개인화 수준이 낮고, 불규칙한 리텐션 패턴이 있는 경우 정확도가 떨어질 수 있어요.
### 머신 러닝
머신 러닝을 활용하면 유저 행동의 복잡한 패턴을 분석해 보다 정밀한 LTV 예측이 가능해요.
> 작동 방식:
> 1\. 특징 선택: 구매 내역, 세션 빈도, 유저 인구 통계, 참여 지표 등 다양한 요소를 분석해요.
> 2\. 모델 학습: 머신 러닝 알고리즘(예: 그라디언트 부스팅, 신경망)을 활용해 과거 데이터를 학습하고 미래의 LTV를 예측해요.
장점:
- 보다 정확하고 개인화된 예측이 가능해요.
- 수익 및 리텐션 패턴이 불규칙한 앱에도 효과적이에요.
한계:
- 방대한 데이터와 지속적인 모델 튜닝이 필요해요.
- 리소스가 많이 필요하고 구현이 복잡할 수 있어요.
## pLTV 계산에서 MMP의 중요성
OS, 국가, 디바이스 유형 같은 퍼스트 파티 데이터만으로 pLTV를 계산할 수 있다고 생각할 수 있어요. 그러나 모바일 측정 파트너(MMP)가 제공하는 어트리뷰션 데이터를 함께 활용하면 pLTV의 정확도를 더욱 높일 수 있어요.
### 어트리뷰션 데이터가 정확도에 미치는 영향
- 채널, 캠페인, 광고 크리에이티브 등 각 광고 소스는 유저 품질과 인게이지먼트를 결정하는 중요한 요소예요.
- 어트리뷰션 데이터를 활용하면 유저가 누구인지뿐만 아니라 어디에서 유입되었는지도 파악할 수 있어요, 이는 pLTV에 큰 영향을 줄 수 있어요.
- 특히 커브 피팅 방법을 사용할 때 어트리뷰션 데이터의 영향력이 더욱 커지는데, 각 코호트별 리텐션 곡선을 계산해야 하고, 코호트마다 ARPDAU가 크게 다를 수 있기 때문이에요.
### 어트리뷰션 데이터 활용 예시
- 예를 들어, 특정 크리에이티브를 활용해 1020 여성 중심의 오디언스를 타겟팅한 숏폼 동영상 채널에서 확보한 유저는 특정한 행동 패턴과 리텐션 성향을 보일 가능성이 높아요.
- 어트리뷰션 데이터는 이러한 유저 행동 차이를 분석하고, 각 광고 소스가 서로 다른 유형의 유저를 어떻게 확보하는지 예측하는 데 도움을 줄 수 있어요.
MMP는 퍼스트 파티 데이터와 어트리뷰션 데이터를 결합해 마케터에게 각 유저 확보 소스가 장기적으로 유의미한지 심층적인 인사이트를 제공해요. 이를 통해 단순한 인구 통계가 아닌 유저 행동 데이터를 기반으로 보다 전략적인 예산 배분과 캠페인 최적화가 가능해요.
## 쉽고 강력한 에어브릿지 pLTV
에어브릿지는 강력한 데이터베이스 엔진인 [루프트(Luft)](https://www.airbridge.io/ko/blog/introducing-luft)와 결합된 베이지안 커브 피팅 접근 방식을 활용해 모바일 마케터에게 최적화된 강력한 pLTV 솔루션을 제공해요.
[__wf_reserved_decorative]
### 에어브릿지 pLTV 기능이 강력한 이유
- 세분화된 인사이트: 에어브릿지는 광고 크리에이티브뿐만 아니라 국가, OS, 채널, 캠페인별로도 pLTV 인사이트를 제공하는 심층 분석 기능을 갖추고 있어요. 이를 통해 마케터는 특정 차원에서 pLTV의 영향을 파악하고, 지역, 광고 소재, 유저 세그먼트에 따라 캠페인을 최적화할 수 있어요.
- 최소한의 데이터로 장기 예측: 에어브릿지 pLTV 기능은 [최근 3일간의 리텐션 데이터만으로도 최대 180일 동안 신뢰할 수 있는 예측을 생성할 수 있어요](https://help.airbridge.io/ko/guides/predictive-ltv#%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80%EC%9D%98-pltv). 덕분에 마케터는 방대한 과거 데이터 없이도 빠르게 장기적인 인사이트를 확보하고, 데이터 기반으로 예산 배분 및 전략을 결정할 수 있어요.
- 원클릭 분석 지원: 마케터는 에어브릿지 대시보드에서 단 몇 번의 클릭만으로 pLTV, 고객 획득 비용(CAC), 광고 지출 대비 수익률(ROAS)을 동시에 비교할 수 있어요. 여러 도구나 플랫폼을 따로 탐색할 필요 없이, 한눈에 빠르고 실행 가능한 인사이트를 얻을 수 있어요.
- 추가 비용 없는 올 인클루시브: 에어브릿지는 추가 비용 없이 기본 패키지에 pLTV 기능을 포함하고 있어요. 덕분에 모든 규모의 비즈니스가 비용 부담 없이 고급 분석 기능을 활용할 수 있어요.
마케터는 에어브릿지와 함께하면 신속한 데이터 기반 의사 결정을 내릴 수 있고, 캠페인 및 지역별로 더욱 효율적인 예산 배분이 가능해요.
## 고객 성공 사례: 에어브릿지 pLTV로 글로벌 마케팅을 최적화한 딜라이트룸(알라미)
[__wf_reserved_decorative]
170개국 이상에서 서비스되는 글로벌 알람 앱, 알라미를 개발한 딜라이트룸은 에어브릿지의 pLTV 기능을 활용해 pLTV 분석 작업에 드는 리소스를 획기적으로 줄이고 마케팅 최적화 전략에 더 힘을 쏟을 수 있게 되었어요.
### 초기 과제
에어브릿지 pLTV 기능을 사용하기 전, 딜라이트룸 마케팅 팀은 자체 알고리즘을 사용해 데이터를 처리하고 매번 pLTV를 계산하고 있었어요. 하지만 이 과정은 자원 소모가 크고 복잡한 작업이었어요. 글로벌 마케팅을 진행하기 위해서는 채널, 캠페인, 광고 크리에이티브, 국가, OS별로 pLTV 지표를 분기, 월, 주 단위로 세분화해 분석할 수 있는 기능이 필요했어요. 이런 높은 수준의 세분화는 각 지역과 플랫폼별로 빠르게 예산을 배분하기 위해 필수적이었어요.
### 에어브릿지 pLTV 성과
에어브릿지 pLTV 기능을 활용하면서 기존에 직접 계산하면 최대 3시간까지도 걸리던 과정을 대시보드에서 즉각적으로 확인할 수 있었어요. 절약한 시간은 마케팅 최적화 전략을 고도화하는 데에 투자할 수 있게 됐죠. 더불어 채널, 캠페인, 광고 크리에이티브, 국가, OS별로 pLTV와 예측 ROAS(pROAS)를 원하는 시간 단위로 실시간 모니터링하면서 예산을 더 효율적으로 운용하게 되었어요.
[AB180 CS Hub](https://cshub.ab180.co/ko/case-studies/airbridge-pltv-delightroom)에서 딜라이트룸이 에어브릿지에서 어떻게 pLTV 기능을 세팅하고 개선 포인트를 찾아갔는지 더 자세히 알아보세요.
## 에어브릿지 pLTV로 성장 가능성을 극대화하세요
빠르고 정확한 의사 결정이 중요한 모바일 시장에서 pLTV 분석은 마케팅 전략의 판도를 바꿀 수 있는 요소예요. 마케터는 pLTV를 활용해 예산을 조정하고, 유저의 장기적인 가치를 이해해 서비스를 장기적으로 성장시킬 수 있어요.
특히 에어브릿지와 같은 MMP를 통해 pLTV를 분석하면 어트리뷰션 데이터처럼 유저 특성에 맞는 추가적인 데이터를 활용할 수 있어요. 지금 에어브릿지 팀과 함께 pLTV 기능을 활용해 더 스마트한 마케팅 전략을 설계해 보세요.
---
## LLM(대규모 언어 모델)이란? 중요한 이유와 마케팅 활용법
URL: https://www.airbridge.io/ko/blog/what-is-llm
Language: Korean
Published: 2025-02-17
## LLM은 무엇이고 왜 중요할까요?
최근 인공지능(AI) 기술이 발전하면서 LLM이라는 용어가 자주 입에 오르 내리죠. 대규모 언어 모델이라는 뜻을 가진 LLM(Large Language Model)은 2025년 마케터들에게도 중요한 기술이자 키워드로 떠오르고 있어요.
> 💡LLM이란?
LLM은 "Large Language Model"의 약자로, 대량의 텍스트 데이터를 학습해 사람이 쓰는 자연어를 이해하고 생성할 수 있는 AI 모델이에요.
[ChatGPT](https://chatgpt.com/) 같은 AI 도구가 LLM 기술을 기반으로 만들어졌다는 사실, 알고 계셨나요? 이번 글에서는 LLM의 주요 특징과 작동 원리, 그리고 이를 마케팅에 어떻게 활용할 수 있는지 쉽게 설명해 드릴게요.
## LLM의 주요 특징과 작동 원리
### LLM의 주요 특징
- 대규모 데이터 학습: 수십억 개의 단어, 문장을 포함한 방대한 데이터를 학습해요. 덕분에 복잡한 문맥도 이해할 수 있죠.
- 자연어 처리(NLP) 능력: 텍스트 분석, 번역, 감정 분석 등 다양한 자연어 작업이 가능해요.
- 사람과 비슷한 언어 생성: 마치 사람이 대화하는 것처럼 자연스러운 텍스트를 만들어낼 수 있어요.
### LLM은 어떻게 작동할까요?
LLM은 Transformer라는 딥러닝 아키텍처를 활용해 작동해요. Transformer의 핵심은 문맥을 이해하기 위해 텍스트 내 단어 간의 관계를 계산하는 것이에요. 이를 통해 단순히 단어를 나열하는 것이 아니라 의미와 문맥에 맞는 텍스트를 생성할 수 있어요.
예를 들어, “오늘 날씨가”라는 입력을 받으면, LLM은 다음에 나올 가능성이 높은 단어를 예측해 “좋아요”와 같은 문장을 완성하죠.
## LLM, 마케팅에 어떻게 활용할 수 있을까요?
LLM은 마케팅 업무에 혁신적인 변화를 가져다주고 있어요. 다음은 LLM을 마케팅에서 활용하는 대표적인 방법들이에요.
### 1. 콘텐츠 제작 자동화
블로그 글, 이메일, 소셜 미디어 콘텐츠를 자동으로 생성할 수 있어요. LLM은 키워드를 기반으로 독창적이고 매력적인 글을 작성할 수 있는 능력이 있죠.
### 2. 개인화된 고객 경험 제공
LLM을 활용하면 고객의 과거 행동 데이터를 바탕으로 개인화된 메시지를 생성할 수 있어요. 예를 들어, 고객 이름과 구매 이력을 바탕으로 맞춤형 이메일 혹은 CRM 푸시 메시지를 제작하고 전송할 수 잇죠. 보낼 수 있죠.
### 3. 채팅봇과 고객 지원
LLM은 고객 문의를 빠르고 정확하게 처리할 수 있는 AI 채팅봇의 핵심 기술이에요. 자연스러운 대화로 고객 만족도를 높이고, 운영 효율성을 향상할 수 있답니다.
## 2025년, LLM을 이해하고 활용하는 마케터가 되세요!
LLM은 단순히 기술적인 개념을 넘어, 다양한 산업에서 실질적으로 활용되고 있는 강력한 도구예요. 특히 마케팅 분야에서는 콘텐츠 자동화, 고객 맞춤형 경험 제공, 실시간 고객 지원 등 무궁무진한 가능성을 열어주죠.
마케팅 솔루션에서도 LLM을 활용한 AI 모델을 통해 마케팅 성공을 돕고 있어요. 에어브릿지, 앰플리튜드, 브레이즈가 어떤 방식으로 LLM 기술을 활용하고 있는지 궁금하시다면 지금 바로 아래 버튼을 통해 확인해 보세요. 👇
---
## MMP(모바일 측정 파트너)란? 모바일 마케팅의 단짝친구
URL: https://www.airbridge.io/ko/blog/what-is-mmp
Language: Korean
Published: 2025-02-13
## 광고비는 쓰고 있는데, 성과는 어디로 갔을까요?
여러 채널에 광고비를 투자했는데, 어떤 광고가 실제로 효과를 내고 있는지 확신하기 어렵다면 답답할 수밖에 없죠. 클릭 수는 높은데 설치로 이어지지 않거나, 앱 설치는 많지만 정작 중요한 구매나 가입으로 연결되지 않는 상황도 종종 발생해요.
광고 채널 별로 데이터를 일일이 확인할 수 있지만 이는 각 채널 데이터를 단순히 합치는 작업이기 때문에 정확성이 떨어질 수 있을 뿐만 아니라 마케터의 많은 시간과 노력을 필요로 하죠.
측정의 불투명성은 많은 모바일 마케터가 겪는 큰 고민 중 하나예요. 이런 문제를 해결하기 위해 필요한 것이 바로 MMP(Mobile Measurement Partner, 모바일 측정 파트너)예요. 모바일 마케팅에서 점점 더 많은 데이터가 생성되고 있는 요즘, MMP는 마케팅 캠페인의 성공을 좌우하는 중요한 역할을 하고 있죠. 이번 포스트에서는 MMP의 주요 기능과 장점에 대해 이야기할 예정이에요.
> 💡 MMP란 무엇인가요?
MMP는 모바일 측정 파트너(Mobile Measurement Partner)의 약자로, 광고 캠페인의 성과를 추적하고 데이터를 분석하는 데 사용되는 마테크(MarTech) 솔루션이에요.
## MMP의 주요 기능
### 1. 모바일 광고 성과 측정
MMP는 여러 [광고 채널의 데이터를 통합하고, 이를 하나의 대시보드에서 확인](https://www.airbridge.io/ko/product/marketing-analytics)할 수 있게 해줘요. 또한 광고 캠페인에서 발생하는 주요 전환 이벤트(앱 설치, 회원 가입, 구매 등)를 정확히 추적해요. 이를 통해 각 채널이 얼마나 효과적인지, 어떤 광고가 가장 많은 전환을 일으켰는지 한눈에 파악할 수 있어요.
### 2. 어트리뷰션 데이터
유저가 어떤 광고를 보고 유입되었는지를 아는 건 중요하죠. 모바일 환경의 특성상 유저는 광고 하나만으로 유입되지 않았을 가능성이 커요. 유저 여정 전체를 분석하기 위해 MMP는 여러 [광고의 점집을 추적할 수 있도록 다양한 형태의 어트리뷰션 데이터를 제공](https://www.airbridge.io/ko/blog/what-is-marketing-attribution-model)해요.
### 3. 광고 사기 방지
모바일 광고 환경에서는 허위 클릭이나 설치 같은 광고 사기가 빈번하게 발생해요. [MMP는 광고 사기를 방지하는 기술을 활용해 데이터의 정확성을 보장](https://www.airbridge.io/ko/product/fraud-protection)하고, 광고비 낭비를 줄여줘요. 이를 통해 마케터는 안심하고 캠페인을 운영할 수 있어요.
%%CTA|데모 신청하기|/ko/book-a-demo|MMP가 처음이라면, 에어브릿지로 쉽게 시작해 보세요.%%
## MMP의 장점은 무엇인가요?
모바일 마케팅은 다양한 경로에서 사용자를 유입시키는 복잡한 과정이에요. 소셜 미디어, 검색 광고, 디스플레이 광고 등 여러 채널이 섞이다 보면, 어떤 채널이 효과적이고 어떤 채널이 비효율적인지 정확히 알기 어려워요. MMP를 활용하면 다음과 같은 장점이 있어요.
### 1. 데이터 기반 의사 결정 지원
광고 캠페인의 성공 여부는 데이터를 기반으로 한 의사 결정에 달려 있어요. MMP는 정확한 데이터를 제공해 광고 전략을 최적화할 수 있도록 도와줘요.
### 2. 모바일 광고 성과 개선
MMP에서 제공되는 세분화된 광고 성과 데이터 분석을 통해 모바일 광고 성과를 개선할 수 있어요. ROAS가 높은 채널과 크리에이티브가 무엇인지 어떤 유저 코호트가 생애 가치(LTV)가 높은지 등 다양한 데이터를 통해 광고 성과를 개선할 수 있도록 도와요.
### 3. 시간과 비용 절약
MMP를 사용하면 여러 광고 네트워크에서 데이터를 수집하고 분석하는 시간을 대폭 줄일 수 있어요. 이렇게 절약된 시간은 더 창의적이고 전략적인 업무에 투자할 수 있죠. 또한 분석한 지표를 바탕으로 광고 집행 비용을 최적화할 수 있어요.
### 4. 맞춤형 전략 설계
MMP의 데이터는 특정 사용자 그룹을 타겟팅 하거나 맞춤형 광고 메시지를 만드는 데도 활용돼요. 이런 세분화된 접근법은 전환율을 더욱 높이는 데 기여해요.
### 5. AI 분석을 통한 미래 인사이트 확보
고도화되는 AI 기술에 발맞춰 MMP 또한 [AI를 활용한 인사이트를 제공](https://www.airbridge.io/ko/product/airbridge-ai)하고 있어요. 유저 생애 가치를 예측하고(pLTV) 코호트 별 수익을 예상할 수 있고 시뮬레이션을 통해 유저 정보의 빈틈을 메울 수 있죠. 또한 광고 사기 방지 기능을 활용해 사기가 의심되는 설치 정보를 찾아내 더 정확한 성과 지표를 만들 수 있어요.
💡 요약하자면 MMP는 이런 문제를 해결해 주어요:
- 광고 성과 분석과 AI를 통한 의사 결정 지원: 설치 수, 인앱 구매, 가입 같은 주요 지표를 각 광고별로 나눠서 보여주고 AI를 통해 미래를 예측해 마케터의 의사 결정을 지원해요.
- 광고 성과 개선: ROAS가 높은 소재와 채널, LTV가 높은 유저 코호트를 분석해요.
- 시간 및 비용 절약: 광고 성과 데이터 분석이 쓰는 시간과 광고 집행 비용을 절약할 수 있어요.
- 맞춤형 전략 설계: 유저 코호트 별 광고 전략 등 맞춤형 광고 전략 설계를 도와요.
### 모바일 마케팅 성과 측정 에어브릿지와 함께하세요!
MMP는 모바일 마케팅 성공을 위해 필수적인 도구예요. 에어브릿지는 국내 최대 MMP 솔루션으로, 국내외 다양한 채널 광고비 데이터 집계, 광고 성과 측정, 어트리뷰션 분석,심리스한 딥링크, AI를 통한 LTV 예측, 광고 사기 방지, 웹 투 앱 성과 측정, PC/Console 게임 성과 측정 등을 제공하며 많은 기업들의 모바일 마케팅 성공을 돕고 있어요.
모바일 마케팅을 한 단계 더 발전시키고 싶다면 지금 바로 에어브릿지와 이야기 나누어보세요.
---
## 2025년 하이브리드 & 하이퍼캐주얼 게임 UA, 수익화, AI 크리에이티브 전략
URL: https://www.airbridge.io/ko/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives
Language: Korean
Published: 2025-02-11
## **새벽 4시 감성 대화와 카페인 가득한 인사이트가 만나는 헛소리 제로 게임 팟캐스트**
게임 팟캐스트 [Two and a Half Gamers](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH)의 슬로건이자, 세 호스트 [Matej](https://www.linkedin.com/in/matejlancaric/), [Jakub](https://www.linkedin.com/in/jakubremiar/), [Felix](https://www.linkedin.com/in/felix-braberg-7a732b51/)가 이 팟캐스트를 운영하는 이유이기도 해요. 이들의 대화는 가식이 없고, 때로는 ‘잔인할 정도로 솔직하지만’([brutally honest](https://lancaric.substack.com/)) 그만큼 현실적이고 값진 인사이트가 가득해요. Two and a Half Gamers가 UA, 수익화, 크리에이티브 최적화와 같은 주제에 대한 날카로운 조언을 찾는 모바일 게이밍 업계 사람들의 사랑을 받는 이유에요. 세 호스트는 10년 이상의 업계 경험을 바탕으로 대형 게임부터 인디 게임, 캐주얼부터 하드코어 장르까지 다양한 게임이 글로벌 시장에서 성공하는 과정을 지켜봤어요. 실제로 스튜디오, 퍼블리셔, 개발자와 함께 성공을 위해 직접 일하기도 했죠.
[__wf_reserved_decorative]
이번 인터뷰에서는 [남성필](https://www.linkedin.com/in/roinam/) [AB180](https://www.ab180.co/) | [Airbridge](https://www.airbridge.io/ko) 대표와 세 호스트가 만나 2025년 모바일 게임의 성공을 위한 전략을 깊이 있게 다뤘어요. 실전에 바로 활용할 수 있는 인사이트를 몇 가지 정리해 볼게요.
### 🎥 Two and a Half Gamers와의 전체 인터뷰를 유튜브에서 확인해 보세요.
## 2025 모바일 게임 시장의 핵심 3가지 - 속도, 수익화, 글로벌화
코로나19로 하이브리드 및 하이퍼캐주얼 게임이 급성장한 이후, 모바일 게임 시장이 포화 상태에 이르렀다는 이야기가 많아요. 하지만 Two and a Half Gamers는 여전히 모바일 게임 산업이 빠르게 성장하고 있으며, 속도, 수익화, 그리고 글로벌화라는 세 가지 핵심 요소가 지속적인 확장을 이끌고 있다고 지적했어요. 현재 게임 스튜디오는 검증된 성공 공식을 완성하고 이를 반복 적용해 그 어느 때보다 많은 게임을 출시하고 있어요. 이제 단 하나의 멋진 게임을 만들어서 성공하기를 기대하는 것은 더 이상 불가능해요.
> "이제 중요한 건 얼마나 빠르게 여러 게임을 개발하고 출시할 수 있는지예요. 즉, 반복적인 출시 속도가 핵심이죠." - Jakub
> "이 템플릿화된 모델을 따르지 않으면 성공하기 어려워요. Habby가 대표적인 사례예요. Habby는 로그라이크 액션 RPG 메타게임이라는 템플릿을 완성했고, 이를 지속적으로 개선하며 다양한 핵심 게임 스타일을 적용하고 있어요. 모두 같은 메타게임 구조지만, 핵심만 바꾸고 비주얼만 조정하면 완전히 새로운 게임처럼 보이죠. 결국 기존 게임의 완성된 버전을 반복적으로 발전시키는 방식이에요. 히트작 하나에만 의존하는 것이 아니라, 얼마나 빠르게 다음 게임을 개발할 수 있는지가 관건이에요." - Jakub
MiHoYo (Genshin Impact, Honkai: Star Rail), Peak (Match Factory), DeNA (Pokémon Masters EX) 같은 회사들이 이 템플릿화된 성장 전략을 활용하며 글로벌 시장에서 성공을 거두고 있어요. 앱스토어 차트만 보면 모바일 게임 산업이 둔화된 것처럼 보일 수도 있지만 실제로는 전혀 그렇지 않아요. 오히려 웹샵(Webshop)이 새로운 수익 모델로 떠오르고 있죠.
> "앱스토어가 아닌 웹샵을 통해 결제되는 비중이 50%에 달하는 회사를 본 적이 있어요. 이처럼 미계상 매출을 고려하면, 모바일 게임 시장은 여전히 두 자릿수 성장세를 유지하고 있다고 볼 수 있어요." - Felix
[__wf_reserved_decorative]
지난 몇 년 동안 하이브리드 및 하이퍼캐주얼 게임이 빠르게 글로벌화되고 있어요. 특히 튀르키예, 키프로스, 베트남, 중국, 한국 같은 유라시아 및 아시아 국가에서 이러한 게임들이 많이 등장하고 있어요. 이들 개발사는 처음부터 글로벌 시장을 겨냥해 게임을 개발하며, 그 성장 속도와 영향력은 매우 공격적이에요.
> "중국이 얼마나 앞서 있는지 보고 정말 놀랐어요. 예를 들어, Lucky Defense가 글로벌 출시됐을 때 미국이나 유럽 업계의 절반은 이 게임에 대해 알지도 못했어요. 그런데 중국에서는 출시 2주 만에 이미 10개 이상의 회사가 프로토타입을 만들고, 복제품을 개발하거나 개선 작업을 진행하고 있었죠. 정말 대단한 일이에요!" - Matej
## 이제는 필수가 된 UA 전략 다각화
유저 확보(UA)는 그 어느 때보다 어려운 과제가 되었어요. 하지만 시장이 어려울수록 창의적인 전략이 필요해요. 경쟁이 치열한 모바일 게임 시장에서 UA 전략을 다각화하는 것은 이제 선택이 아니라 필수예요.
> "이제는 처음부터 여러 채널에서 UA를 시작해야 해요. 몇 년 전만 해도 그렇지 않았지만, 지금은 완전히 달라졌어요." - Matej
가장 효과적인 UA 전략은 Facebook, TikTok, AppLovin, Unity 같은 4개의 주요 플랫폼을 활용하는 것으로 시작해 점차 Mistplay, adjoe 같은 다른 플랫폼으로 확장하는 것이에요. 이렇게 하면 특정 채널에 의존하는 위험을 줄이면서 더 많은 유저에게 도달할 수 있어요.
> "구독자들을 대상으로 설문조사를 진행했어요. 평균적으로 UA 채널을 8개에서 12개 정도 운영하고 있었고, 예산은 월 10만 달러에서 수백만 달러까지 다양했어요." - Matej
단순히 여러 플랫폼을 활용하는 것이 아니라 캠페인 유형을 적절히 조합하는 것도 중요해요. 예를 들어 인앱 구매(ROAS) 캠페인 같은 고수익 전략과 앱 이벤트 최적화 캠페인 같은 비용 효율적인 방법을 함께 운영해야 해요. 지금의 UA 환경에서 성공하려면 초기부터 빠르게 실행하고, 여러 채널을 테스트하고, 데이터를 기반으로 신속하게 조정하는 것이 필수예요. 이런 속도를 유지하지 못하면 경쟁에서 뒤처질 수밖에 없어요.
## 하이브리드 수익화 성공 법칙: 정교한 경제 시스템과 맞춤형 UA 전략
Two and a Half Gamers는 하이브리드 수익화를 광고 수익 40-60%, 인앱 구매(In-App Purchase, IAP) 수익 40-60%로 이루어진 모델이라고 정의해요. 즉 광고를 통해 빠르게 수익을 확보하면서도 인앱 구매를 통해 지속적인 매출을 창출하는 것이 핵심이에요.
하이브리드 수익화를 성공적으로 운영하려면 게임 내에서 유저가 충분히 소비할 수 있도록 탄탄한 경제 시스템과 ‘지출 규모’를 확보해야 해요. 이를 위해서는 매력적인 성장 구조를 설계하고, 게임의 기능을 점진적으로 확장하며, 중간 광고, 보상형 동영상, 배너 광고 같은 다양한 수익화 요소를 전략적으로 배치하는 것이 중요해요. Voodoo와 SayGames 같은 회사들이 이러한 모델을 잘 활용하고 있어요.
Jakub은 장기적인 수익 창출을 위해 IAP 시스템이가 필수적이라고 강조했어요.
> "라이프사이클이 3~4년 이상인 게임이라면 인앱 구매 시스템을 반드시 도입해야 해요. 그래야 플레이어가 게임 내에서 지속적으로 소비할 기회를 얻고, 게임의 수명을 연장할 수 있어요." - Jakub
[__wf_reserved_decorative]
하지만 단순히 광고와 인앱 구매를 병행한다고 해서 성공하는 건 아니에요. 핵심은 맞춤형 전략과 정교한 유저 확보(UA) 전략을 결합하는 것이에요.
> "가장 중요한 건 UA가 곧 게임 디자인과 광고 세그멘테이션의 핵심이라는 점이에요. UA 없이 광고 수익화나 낮은 CPI를 달성하는 건 불가능해요." - Felix
게임의 UA 캠페인 목표에 따라 유저 경험을 맞춤 조정하는 것도 필수예요. 예를 들어 IAP 중심 캠페인에서 유입된 플레이어는 초반 광고를 거의 보지 않지만, 광고 ROAS 캠페인으로 유입된 유저는 배너 및 인터스티셜 광고를 더 빠르게 노출시켜 수익 창출을 극대화할 수 있어요. 이를 위해서는 실시간 데이터 분석이 필수예요. [에어브릿지](https://www.airbridge.io/ko)와 같은 어트리뷰션 툴을 활용하면 유저 행동을 실시간으로 분석하고, 각 유저에게 최적화된 수익화 전략을 적용할 수 있어요.
마지막으로 Jakub은 수익화 모델을 설계할 때 신중해야 한다고 경고했어요.
> "소액 결제 유저에게 고가의 혜택을 제공하는 실수를 하면 안 돼요. 이런 유저들은 오히려 부담을 느끼고 구매를 꺼릴 가능성이 커요." - Jakub
결국 하이브리드 수익화의 성공 여부는 정확한 유저 세그멘테이션, 최적화된 광고 전략, 그리고 유저가 자연스럽게 소비할 수 있도록 설계된 게임 경제 시스템에 달려 있어요.
## AI 기반 광고 크리에이티브로 광고의 양과 질 모두 잡기
성공적인 UA는 단순히 좋은 광고를 만드는 것이 아니라, 훌륭한 광고를 빠르게, 그리고 대량으로 제작해야 가능해요. 그리고 여기에서 AI가 핵심 역할을 해요. 특히 플레이어블 광고는 하이브리드 및 하이퍼캐주얼 게임에서 가장 효과적인 광고 포맷 중 하나로 잘 알려져 있어요. 하지만 그 효과에도 불구하고 대규모 제작과 배포가 쉽지 않다는 점이 가장 큰 문제예요.
> "모든 SDK 네트워크에서 플레이어블 광고가 필요해요. Facebook과 Google을 운영하는 경우라면 상관없지만, AppLovin을 활용한다면 반드시 필요해요. Mintegral이나 Moloco 같은 네트워크에서도 플레이어블 광고는 매우 중요해요." - Matej
이제는 플레이어블 광고를 직접 제작하는 과정도 훨씬 간소화됐어요. 기존에는 높은 개발 리소스가 필요했지만, 이제는 [playableMaker](http://playablemaker.com) 같은 도구를 활용하면 게임 튜토리얼을 기반으로 손쉽게 인터랙티브 광고를 만들고 확장할 수 있어요. AI 크리에이티브 도구는 단순한 보조 수단이 아니에요. 빠르게 변화하는 광고 트렌드와 알고리즘에 적응하기 위해 반드시 필요해요.
> "Facebook과 TikTok에서는 크리에이티브를 이틀에서 사흘마다 새로 바꿔야 해요. AI 도구 없이는 이런 속도를 따라잡는 게 불가능해요." - Matej
결국, 광고 크리에이티브의 핵심은 퀄리티와 스피드를 동시에 잡는 것이에요. AI 도구를 활용해 반복적인 크리에이티브 제작을 자동화하면서 보다 창의적인 아이디어와 전략에 집중할 수 있도록 해야 해요. AI 크리에이티브가 빠르게 발전하면서 정적인 이미지에서 영상으로, 2D에서 3D로 전환하는 것이 훨씬 쉬워졌어요. Midjourney 같은 도구가 발전하면서 AI 음성을 활용한 스토리 중심의 내러티브 크리에이티브를 제작하는 것도 점점 간편해지고 있어요.
> "가장 좋은 점은 고품질 크리에이티브를 제작할 수 있는 자체 크리에이티브 팀이 있다는 거예요. 이 팀은 독창적인 콘셉트를 직접 개발하죠. 그리고 그 옆에서 AI 도구를 활용해 볼륨 플레이의 균형을 맞추는 것이 중요해요." - Matej
AI 도구를 활용하면 크리에이티브 툴에 익숙하지 않은 UA 매니저나 개발자도 빠르게 크리에이티브를 제작할 수 있어요. 결국 빠르고 효율적인 AI 활용 능력이 모바일 게임 광고에서는 핵심 요소에요
> "저는 Poolday AI를 사용해요. 10개의 다른 AI 모델을 활용해 1개의 영상을 만들 수 있어요. 원래 100명이 해야 할 작업을 5분 만에 끝낼 수 있는 거죠. 올해 가장 성공할 게임 타이틀이 될지 아닐지는 크리에이티브의 속도에 달려 있어요. 얼마나 많은 광고 크리에이티브를 만들어낼 수 있느냐가 관건이에요. 지금 알고리즘이 원하는 게 바로 그것이니까요." - Matej
[__wf_reserved_decorative]
## 더욱 확장된 UA 매니저의 역할
UA 매니저는 더 이상 단순히 광고를 운영하는 미디어 바이어가 아니에요. 이제는 마케팅 테스트, 소프트 론칭을 위한 프로덕트 팀과의 협업, 광고 크리에이티브 제작까지 폭넓은 역할을 수행해야 해요.
> "UA 매니저라면 당연히 여러 역할을 맡고, 여러 자리를 차지해야 해요." - Matej
특히, 예전처럼 단순히 CPI(설치당 비용) 테스트에만 의존하는 방식은 더 이상 효과적이지 않아요. 최근 주목받는 전략 중 하나는 온보딩 과정에서 가장 효과적인 접근 방식을 찾기 위해 미니게임 형태의 여러 버전을 테스트하는 것이에요.
> "성공적인 게임들은 첫 60~70분을 A, B, C, D 버전으로 나눠 테스트해요. 같은 게임이라도 온보딩 방식에 따라 유저 반응이 달라지거든요." - Jakub
## 예측 LTV 분석, 솔루션으로 더 쉽고 정교하게
예측 생애 가치(pLTV) 모델은 모바일 게임 개발사가 더 빠르고 정확한 데이터 기반 의사결정을 내리는 데 필수적인 도구예요.
> "코호트별로 첫 3일 동안의 데이터를 매일 확인해야 해요. 7일을 기다려서 의사결정을 한다면 이미 늦은 거예요." - Matej
pLTV는 성과 예측과 최적화에 필수적이지만 많은 게임 회사가 이를 제대로 활용하지 못하고 있어요. Matej는 흔히 저지르는 실수를 이렇게 설명했어요.
> "대부분의 회사들은 pLTV 모델을 한 번 구축한 뒤 방치해요. 모델을 업데이트하거나 과거 데이터를 검토하지 않으면 결국 엉망이 될 수밖에 없어요. 정확도를 유지하려면 최소 3개월마다 모델을 재조정해야 해요." - Matej
pLTV 모델을 사내에서 직접 유지하는 것이 어렵기 때문에 많은 모바일 게임 개발사들은 [에어브릿지](https://www.airbridge.io/ko)와 같은 솔루션을 활용해 pLTV, 코호트 분석, 그리고 분/시간 단위 리텐션 분석을 진행하고 있어요. 이를 통해 실시간 데이터 기반으로 더욱 정교한 의사결정을 내릴 수 있어요.
## 현지화로 글로벌 시장에 맞는 게임 경험 만들기
글로벌 시장에서 성공하려면 단순한 번역을 넘어서 각 지역의 문화와 플레이 스타일에 맞게 게임을 최적화하는 현지화가 필요해요. 지역별 유저 성향을 이해하고, 그에 맞춰 게임 디자인을 조정하는 것이 핵심이에요.
> "아시아의 기능 세트는 서구보다 3배 더 복잡해요. 아시아 유저들은 더 많은 기능을 원하기 때문이죠." - Jakub
아시아 유저들은 단계별 성장 구조와 여러 시스템이 복잡하게 얽혀 있는 게임을 선호하는 경향이 있어요. 반면, 이런 다층적인 게임 디자인은 서구권 유저들에게는 너무 어렵거나 부담스럽게 느껴질 수 있죠. Jakub은 서구 시장에 맞춘 접근 방식에 대해 이렇게 말했어요.
> "게임을 서구 시장으로 가져갈 때는 더 단순화하고, 더 접근하기 쉽게 만들어야 해요. 예를 들어, MY.GAMES의 Rush Royale을 보면, Random Dice의 기본 구조를 가져오면서도 훨씬 직관적으로 바꾼 사례라고 할 수 있죠." - Jakub
즉 아시아, 유럽, 라틴 아메리카 등 문화권별 유저들이 선호하는 게임 구조를 정확히 분석하고 현지화해야 해요. 게임 메커니즘을 단순화하고 지역별 특화 디자인을 만들면 다양한 문화적 맥락에서도 공감을 얻을 수 있어요.
## 현명한 전략으로 민첩하게 시장에 반응하기
Two and a Half Gamers는 공통적으로 민첩하게 시장에 반응해야 한다고 짚었어요. 단순히 좋은 게임을 개발하는 것만으로는 부족해요. UA 다각화, 하이브리드 수익화, AI 크리에이티브, 유저 데이터와 같은 요소를 어떻게 활용해서 게임을 성공으로 이끌고 더 넓은 시장으로 나아갈지가 핵심이에요.
> "성공은 강력한 UA, 혁신적인 수익화 전략, 그리고 확장 가능한 광고 크리에이티브의 균형에서 나와요. **더 열심히 하는 것이 아니라, 더 현명하게 일하는 것이 중요해요.**"- Matej
[__wf_reserved_decorative]
[Two and a Half Gamers 팟캐스트](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH)와 [](https://airbridge.io)[에어브릿지 홈페이지](https://www.airbridge.io/ko)를 방문해 더 많은 인사이트를 얻어보세요.
## Two and a Half Gamers와 함께 성공 전략을 세워보세요
유저 확보, 수익화, 크리에이티브 최적화에 대한 최신 전략, 성공 사례, 그리고 거침없는 토론을 만나보세요.
- YouTube에서 [Two and a Half Gamers](https://www.youtube.com/@2.5gamers) 채널을 구독하세요. 심층적인 튜토리얼, 전문가 인터뷰, 그리고 실제 사례 분석을 통해 업계의 핵심 인사이트를 얻을 수 있어요.
- [Two and a Half Gamers 팟캐스트](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH)에서 놓칠 수 없는 업계 트렌드와 깊이 있는 분석을 만나볼 수 있어요.
- LinkedIn에서 [Matej](https://www.linkedin.com/in/matejlancaric/), [Jakub](https://www.linkedin.com/in/jakubremiar/), [Felix](https://www.linkedin.com/in/felix-braberg-7a732b51/)와 연결하세요. UA, 수익화, 크리에이티브 최적화 등 실제 성장 과제에 대한 맞춤형 조언과 인사이트를 얻을 수 있어요.
_Two and a Half Gamers_와 함께 업계 최신 정보를 빠르게 습득하고 변화에 맞춰 2025년 모바일 게임 시장을 선점하세요.
## 에어브릿지와 함께 더 빠르게 성장하세요
게임 마케팅을 한 단계 업그레이드할 준비가 되었다면 에어브릿지로 성공을 가속화할 차례에요. 에어브릿지는 [게임 비즈니스에 특화된 맞춤형 기능](https://www.airbridge.io/ko/solutions/gaming)을 제공해요.
- 고급 코호트 분석: LTV(유저 평생 가치)를 정밀하게 추적하고, 키워드, 크리에이티브, 국가, OS 등 세부적인 데이터 분석이 가능해요.
- 광고 미디에이션(Ad Mediation) 플랫폼 통합: 여러 광고 네트워크 데이터를 한 곳에서 간편하게 연동하고 최적화할 수 있어요. (예: Google AdMob, AppLovin MAX, Unity LevelPlay)
- 예측 LTV(pLTV) 기능: 실시간으로 미래 수익을 예측해 더 빠르고 정확한 데이터 기반 의사 결정을 내릴 수 있어요.
- 분/시간 단위 리텐션 분석: 유저 행동 데이터를 분 또는 시간 단위로 분석해 더욱 정교한 인사이트를 확보할 수 있어요.
- 투명하고 합리적인 가격 정책: 숨겨진 비용 없이 엔터프라이즈급 분석 기능을 부담 없이 사용할 수 있어요.
넥슨과 웹젠, 후야호 등 주요 게임 퍼블리셔들이 신뢰하는 에어브릿지는 게임사가 빠르게 변화하는 시장에서 앞서 나갈 수 있도록 지원해요. 아래에서 데모를 신청해 더 자세한 이야기를 만나보세요.
---
## 에어브릿지, Microsoft Advertising 연동으로 새로운 성장 채널 열기
URL: https://www.airbridge.io/ko/blog/unlock-new-growth-channels-with-airbridge-integration-with-microsoft-advertising
Language: Korean
Published: 2025-02-10
에어브릿지가 마이크로소프트 애드버타이징(Microsoft Advertising)과 연동을 지원해요. 이제 에어브릿지에서 마이크로소프트의 폭넓은 광고 채널을 활용하여 모바일 광고 캠페인을 더 고도화할 수 있어요.
## 마이크로소프트 애드버타이징 알아보기
[마이크로소프트 애드버타이징(Microsoft Advertising)](https://ads.microsoft.com/)은 우리에게 Windows와 MS Office 등으로 유명한 글로벌 IT 회사 마이크로소프트에서 운영하는 광고 서비스에요. Bing 및 Yahoo 등 다양한 자체 플랫폼 광고를 운영하고 있죠.
## 마이크로소프트 애드버타이징 연동 상세 내용
에어브릿지와 마이크로소프트 애드버타이징 연동으로 마이크로소프트 플랫폼 내 모바일 앱 설치 캠페인 성과 측정이 간편해졌어요. 이를 통해 마케터는 광고 성과를 더 빠르게 확인하고 캠페인 고도화를 진행할 수 있어요.
에어브릿지에서 마이크로소프트 애드버타이징을 연결하면 👇
- 광고를 통해 앱 설치가 얼마나 일어났는지 확인할 수 있어요.
- 캠페인, 채널, 소재별 효율을 확인할 수 있어요.
- 전환율이 높은 사용자 그룹을 찾을 수 있어요.
- 광고 집행 비용과 광고를 통한 매출을 한 눈에 확인할 수 있어요.
- 여러 채널에서 캠페인이 어떻게 진행되고 있는지 하나의 대시보드로 바로 볼 수 있어요.
## 시작하기
에어브릿지에서 마이크로소프트 애드버타이징 광고 성과를 측정하고 활용하는 방법은 다음과 같아요.
1. 에어브릿지에서 마이크로소트프 애드버타이징 광고 캠페인을 추적할 트래킹 링크를 만드세요.
2. 앱 설치 목표를 정의하고 실시간 데이터를 확인할 수 있도록 설정하세요.
3. 에어브릿지 분석 도구로 사용자 행동과 캠페인 결과를 확인하세요.
4. 광고 집행 인사이트를 바탕으로 전략을 고도화하세요.
더 자세한 사용 방법이 궁금하시다면 [유저 가이드](https://help.airbridge.io/ko/guides/microsoft-advertising)를 확인하거나, 에어브릿지 팀에 문의해 도움을 받아보세요.
---
## 어필리에이트 마케팅이란? 2025년 성공 전략 가이드
URL: https://www.airbridge.io/ko/blog/what-is-affiliate-marketing
Language: Korean
Published: 2025-02-03
## 2025년 새롭게 주목 받는 제휴 마케팅
커머스 앱에서 “친구 초대 시 할인 쿠폰 제공”과 같은 문구를 본 적 있으신가요? 유튜브나 인스타그램 등 SNS 채널에서 인플루언서의 추천 상품 링크를 보신 적이 있나요?
위와 같이 제휴 링크를 통해 제품을 판매하는 마케팅을 ‘제휴 마케팅(Affiliate Marketing)’이라고 해요. 제휴 마케팅은 경쟁이 심화되고 있는 모바일 마케팅 시장의 새로운 트렌드이자 ROI(투자 대비 수익률)을 높이기 위한 새로운 전략으로 주목받고 있어요.
이번 포스트에서는 2025년 더욱 주목받고 있는 제휴 마케팅의 기본 개념을 이해하고 효율적인 시작 방법에 대해 이야기할 예정이에요.
## 제휴 마케팅(Affiliate Marketing)이란 무엇인가요?
> 제휴 마케팅의 뜻 💡
마케팅은 광고주가 기업 혹은 개인(인플루언서)과 협력하여 상품을 홍보하고 제휴 링크를 통해 발생한 수익에 대한 보상을 지급하는 마케팅이에요.
제휴 마케팅은 일종의 성과 기반 마케팅 모델이라 할 수 있어요. 제휴를 맺은 기업과 개인, 인플루언서는 제휴 링크를 통해 발생되는 판매가 늘어날수록 더 많은 수익을 얻을 수 있어요. 단순히 수익금 지급 이외에도 유저가 자발적으로 참여할 수 있도록 친구 초대 시 5,000원 할인 쿠폰 제공과 같은 보상을 유도하는 것도 제휴 마케팅의 일종이죠.
광고주는 성과에 맞춰 파트너에게 보상을 지급하면 되기 때문에 제휴 마케팅은 일반적인 광고에 비해 비용을 절감하고 높은 효율을 얻을 수 있어요.
국내에서는 [쿠팡 파트너스](https://partners.coupang.com/)와 [링크프라이스](https://www.linkprice.com/), 해외에서는 아마존[ Amazon Associates](https://affiliate-program.amazon.com/) 등이 대표적인 제휴 마케팅 운영 플랫폼이에요.
### 제휴 마케팅 예시
1. 한 유튜버가 화장품 리뷰 동영상을 올리고, 구매 링크를 첨부했어요. 시청자가 이 링크를 통해 화장품을 구매하면 유튜버는 일정 금액의 수수료를 받게 돼요.
2. 커머스 앱에서 "친구 초대" 프로그램을 운영해, 초대한 친구가 앱을 설치하거나 첫 구매를 하면 초대한 사용자에게 리워드를 제공해요.
### 제휴 마케팅의 장점
제휴 마케팅은 다음과 같은 장점 때문에 새로운 모바일 마케팅 트렌드로 주목받고 있어요.
1. 효율적인 비용 집행
전환(구매, 앱 설치 등)이 이루어졌을 때만 비용이 발생하기 때문에 광고비 낭비가 적어요.
2. 다양한 채널 활용 가능
블로그, 유튜브, 소셜 미디어, 앱 등 다양한 채널에서 제휴 마케팅을 실행할 수 있어요.
3. 빠른 확장성
제휴사를 늘리는 방식으로 손쉽게 마케팅 활동을 확장할 수 있어요.
4. ROI 극대화
성과를 정밀하게 측정할 수 있기 때문에 효율적인 예산 집행이 가능해요.
## 제휴 마케팅, 어떻게 시작해야 할까요?
1. 목표에 맞는 적합한 제휴 파트너와 모델 설계하기
성공적인 제휴 마케팅은 정확한 목표와 제휴 파트너를 찾는 데에서 시작해요. 특정 상품, 카테고리 판매량, 회원 가입 수, 구독 전환율 등 다양한 목표에 맞춰 제휴 마케팅을 함께할 파트너를 찾아야 해요.
1. 성과 기반 보상 체계 설계하기
성과에 따라 명확하고 공정한 보상 기준을 설정하세요. 예를 들어, 클릭당 보상(CPC), 전환당 보상(CPA), 매출 비율 보상(CPS) 등 다양한 모델이 있어요.
1. 제휴 링크 생성 및 배포하기
목표와 파트너, 보상 체계가 확정되었다면 제휴 링크를 생성해 제휴 파트너가 홍보할 수 있도록 도와야 해요. 이때 제휴 링크가 제대로 동작하지 않는다면 제휴를 통해 일어난 전환을 측정할 수 없기 때문에 링크 생성이 중요해요.
1. 성과 추적 도구를 활용해 성과 추적하기
제휴 마케팅이 시작되면 정확한 보상 제공을 위해 성과 추적이 필수에요. 성과 추적 도구를 활용해 파트너 링크를 통해 얼마나 많은 전환이 발생했는지를 추적하여 파트너에게 보상을 지급하세요.
## 제휴 마케팅, 에어브릿지에서 시작하세요.
제휴 마케팅을 통해 앱 성과를 극대화 하고 싶으신가요? 그렇다면 원하는 페이지로 유저를 정확하게 랜딩 시키는 딥링킹과 성과 측정을 통한 정확한 보상 지급이 중요하겠죠. 에어브릿지는 모바일 측정 파트너(MMP)로서 트래킹 링크를 통해 딥링킹과 성과 측정을 지원하고 있어요. 에어브릿지 트래킹 링크 기능에 집중한 서비스 딥링크 플랜과 함께 제휴 마케팅 성과를 더욱 극대화해 보세요. 에어브릿지 딥링크 플랜에서는
1. API 기반으로 트래킹 링크를 자동 생성할 수 있어요.
2. 다양한 국내 채널에서 심리스한 오픈을 지원해요.
3. 제휴 링크의 성과를 측정할 수 있어요.
4. MAU 기반의 합리적인 가격을 제공해요.
---
## 딥링크란? URI Scheme, Universal Link, App Link 완벽 가이드
URL: https://www.airbridge.io/ko/blog/what-is-deep-link
Language: Korean
Published: 2025-01-20
## 딥링크란 무엇이고, 왜 중요한가요?
> **딥링크의 뜻 💡
**딥링크(Deep Link)란 앱 안 특정 위치에 부여된 링크를 뜻해요. 모바일 환경에서 유저가 웹사이트 혹은 다른 앱에서 링크를 클릭했을 때, 앱을 실행시키고 특정 페이지로 이동하도록 돕는 기술이죠. 유저는 광고를 클릭하면 딥링크를 통해 앱 내 특정 페이지로 바로 도달할 수 있어요.
딥링크는 앱 푸시, SMS, 카카오 알림톡과 같은 CRM 마케팅, 광고, 공유, 리퍼럴, 제휴 마케팅 등 다양한 앱 마케팅 활동을 효과적으로 진행할 수 있도록 도와주는 핵심 기능 중 하나에요. **사용자 경험을 개선**하고 구매와 구독 등을 통한 **전환율을 높이기 위해** 꼭 필요하죠.
이번 포스트에서는 딥링크 구현 방식 별 차이를 이해하고 에어브릿지를 통해 어떻게 딥링크를 효율적으로 활용할 수 있는지 살펴볼 예정이에요.
## 딥링크의 종류에는 무엇이 있나요?
딥링크에는 대표적으로 세 가지 구현 방식이 있어요. **URI 스킴(URI Scheme), iOS의 유니버셜 링크(Universal Link), 안드로이드의 앱 링크(App Link)**가 그것이에요.
### URI 스킴(URI Scheme): 가장 기본적인 딥링크 방식
**URI 스킴**은 앱이 고유한 스킴값(scheme)을 가지고 있어서 특정 스킴값을 호출하면 앱을 실행하도록 하는 방식이에요. 예를 들어 에어브릿지 앱을 열기 위해 미리 부여한 스킴값인 airbridge://를 호출하면 되는 것이죠.
스킴값이 앱 호출에 활용된다면, 앱 내 특정 페이지는 패스(path) 값으로 설정하게 돼요. 이는 웹사이트에 붙는 path 값과 동일한 역할을 해요. 즉, URI 스킴은 scheme://path 형태로 구성되어 있어요.
- **예시: **에어브릿지 앱에서 특정 상품을 확인하고 싶다. ➡️ airbridge://stocklist/8516
URI 스킴은 직관적이고 구현이 쉽고 간단하다는 장점이 있지만, 스킴값을 중복으로 사용할 수 있어 같은 스킴을 사용하는 다른 앱이 존재할 수 있어요. 이에 따라 오픈 오류 또는 보안 문제에 직면할 수 있다는 단점이 있어요.
또한 상황에 따라 앱이 설치되어 있지 않은 유저가 URI 스킴 링크를 클릭했을 때 앱 스토어로 바로 연결되지 않을 수도 있어요.
**정리하자면 URI 스킴은 **
- **장점**: 구현이 쉽고 빠르게 설정할 수 있어요.
- **단점**: 스킴값을 중복으로 설정할 수 있어, 다른 앱이 오픈되는 등 보안상의 문제가 생길 수 있어요. 앱이 설치되지 않은 경우 자동으로 앱 스토어로 연결되지 않을 수 있어요.
이러한 URI 스킴의 단점을 보완하고자 iOS와 안드로이드에서는 각자의 딥링크 구현 방식을 만들었어요. 이것이 바로 iOS의 유니버셜 링크(Universal Link)와 안드로이드의 앱 링크(App Link)에요.
### 유니버셜 링크(Universal Link)와 앱 링크 (App Link): iOS와 안드로이드의 딥링크
스킴값 중복, 보안 문제 등 URI 스킴이 가진 단점을 보안하기 위해 iOS에서는 [유니버셜 링크(Universal Link)](https://developer.apple.com/documentation/xcode/allowing-apps-and-websites-to-link-to-your-content/), 그리고 안드로이드(Android)에서는 [앱 링크(App Link)](https://developer.android.com/studio/write/app-link-indexing?hl=ko)라는 이름으로 자체 딥링크를 만들게 되었어요. 이 링크들은 각각의 운영 체제에서 막힘없는 오픈을 지원하며 URI 스킴을 대체하기 시작했죠.
하지만 이 두 링크는 각자 운영체제에서만 정상적으로 오픈되고 다른 운영체제에서는 그렇지 않는다는 단점이 있어요. 예를 들어 A앱을 사용하는 iOS 유저가 안드로이드 앱 링크로 랜딩 되는 푸시 메시지를 받았다면 앱이 정상적으로 오픈되지 못하는 것이죠.
**유니버셜 링크와 앱 링크**
- **장점:** 스킴값 중복 문제 해결, 각 운영체제에서 문제없이 오픈돼요.
- **단점:** 타 운영체제에서 정상적으로 오픈되지 않아요.
또한 유니버셜 링크와 앱 링크는 앱을 설치하지 않은 유저들을 앱스토어와 같은 앱 마켓으로 이동시켜 주지만 앱 설치 이후 클릭한 광고나 추천 링크 페이지를 바로 불러오는 디퍼드 딥링크와 같은 사용자 친화적 기능을 포함하고 있지는 않아요.
이러한 유니버셜 링크와 앱링스의 구동 방식 차이와 한계를 극복하기 위해 많은 앱들에서는 딥링크 전용 솔루션을 활용하고 있어요. 딥링크 솔루션을 활용하면 OS 구분없는 딥링크 오픈과 디퍼드 딥링크같은 부가 기능을 사용할 수 있죠. [구글 파이어베이스 다이나믹 링크](https://firebase.google.com/docs/dynamic-links?hl=ko)와 [에어브릿지 딥링크 플랜](https://www.airbridge.io/ko/deeplink-plan)과 같은 서비스가 바로 그 예시이죠.
[]
%%CTA|데모 신청하기|/ko/book-a-demo|딥링크 설정부터 성과 분석까지, 에어브릿지 하나로 관리하세요.%%
## 에어브릿지 딥링크 플랜: 모든 딥링크를 하나로 관리하기
딥링크는 다양한 채널에서 앱 안으로 유저를 초대하는 문인만큼 사용자 편의를 위해 효과적으로 관리하는 것이 중요해요. 에어브릿지 딥링크 플랜은 **URI 스킴, 유니버셜 링크, 앱 링크**를 모두 지원하며, 실무자들의 **개발 리소스를 최소화하면서도** 유저에겐 **최적의 딥링킹 경험을 **제공할 수 있도록 돕고 있어요.
### 에어브릿지 딥링크 플랜 주요 기능
1. **손 쉬운 마이그레이션**
- 구글 파이어베이스 다이나믹 링크 혹은 다른 딥링크 솔루션을 사용하고 있어도 문제 없어요. 에어브릿지는 별도의 개발 없이도 [쉽고 정확한 딥링크 마이그레이션](https://www.airbridge.io/ko/blog/how-to-migrate-deeplink-in-airbridge)을 지원해요.
2. **딥링크 생성 및 관리**
- 하나의 트래킹 링크(에어브릿지 딥링크)로 iOS와 안드로이드 모두 대응할 수 있어요.
3. **디퍼드 딥링크(Deferred Deep Link)**
- 앱을 이미 설치한 유저는 물론, 앱 미설치 유저가 설치를 완료한 후에도 [마케터가 원하는 페이지로 유저를 바로 연결](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)해 주어요.
4. **QR & 브랜딩 링크**
- [QR코드](https://www.airbridge.io/ko/blog/airbridge-qr-code)를 만들거나 썸네일과 커스텀 도메인 제작을 위해 다른 솔루션을 사용할 필요가 없어요. 이 모든 기능이 에어브릿지 딥링크 플랜안에 포함되어 있기 때문이에요.
5. **리퍼럴, 제휴 마케팅 최적화**
- 리퍼럴과 제휴 마케팅 등 링크 공유를 바탕으로 한 마케팅에 최적화 되어 있어요.
6. **정확한 데이터 트래킹**
- 딥링크 클릭부터 전환까지 데이터를 원하는 조건에 맞춰 실시간으로 분석할 수 있어요.
7. **합리적인 가격**
- 구글 파이어베이스 다이나믹 링크 서비스 중단과 함께 유료 딥링크 서비스에 대한 수요도 상승하고 있어요. 에어브릿지는 월간 활성 유저(MAU)을 기반으로 합리적인 가격을 제공해요.
### 이미 많은 팀들이 에어브릿지를 사용하고 있어요
2025년, 가속화되는 모바일 마케팅 경쟁에서 살아남기 위해서는 **정확한 사용자 연결**과 **경험 개선**이 필수에요. **에어브릿지**가 제공하는 다이나믹 딥링크인 트래킹 링크는 이미 국내외 많은 기업에서 사용하고 있어요. 에어브릿지 딥링크 플랜을 통해 트래킹 링크의 장점을 모두 활용해 앱 마케팅을 시작해 보세요.
---
## 숫자로 보는 에어브릿지 2024 리포트
URL: https://www.airbridge.io/ko/blog/airbridge-recap-2024
Language: Korean
Published: 2025-01-14
새로운 시작을 준비하는 연초에는 자연스레 지난 한 해를 돌아보게 되죠. 바쁘게 달려왔던 에어브릿지의 2024년을 되돌아보며 그동안 무엇에 열중했고, 어떤 성과를 이루었으며, 또 어디에서 가능성을 발견했는지를 정리해 봤어요. 1년간 에어브릿지가 만들어온 수치적인 성과들과 그 속에 담긴 의미 있는 이정표들을 함께 살펴볼게요.
[__wf_reserved_decorative]
## 게임 고객사 4.3배 증가했어요
2024년 에어브릿지를 특히 주목한 산업군은 단연 게임이었어요. 게임 고객사의 수는 전년 대비 4.3배 증가하며 괄목할 만한 성과를 기록했는데요. 게임 업계를 선도하는 여러 기업들이 에어브릿지를 선택해주시며 에어브릿지의 기술력과 신뢰도를 다시 한번 확인하는 뿌듯한 성과를 이뤘어요.
이러한 성과의 배경에는 에어브릿지의 차별화된 성과 측정 기능이 있었어요. 게임 업계는 다양한 국가, OS, 타겟층과 다채로운 채널, 캠페인, 소재를 넘나드는 입체적인 글로벌 캠페인 운영이 필수적이에요. 에어브릿지 솔루션은 이를 실시간으로 분석하고, 최적화를 위한 액션아이템을 도출하는 데 강점이 있어요. 또 한국에서 자체 개발한 MMP 솔루션으로서 에어브릿지는 각 게임사의 요청에 빠르게 대응해 왔는데요. 원활한 소통으로 고객사의 요청에 신속하게 응답하며 기술 개선을 이뤄낸 점이 큰 호응을 얻었어요.
2024년은 게임 산업에서의 접점을 넓히기 위해 에어브릿지가 바쁘게 움직인 해이기도 했어요. 지스타와 베트남 게임즈포럼, 차이나조이, GDC, 게임스컴 등 국내외 게임 컨퍼런스를 통해 고객과 파트너를 만나 에어브릿지의 경쟁력을 알렸는데요. [하이브 플랫폼 연동](https://www.airbridge.io/ko/blog/hive-platform-integration), [애드조 연동](https://www.airbridge.io/ko/blog/adjoe-integration), 애드엑스 연동 등 70여 곳의 매체 및 솔루션과 신규 파트너십을 진행한 것은 물론, 애드조와 함께 [한국 게이머 리포트 2025](https://adjoe.io/ko/mobile-gamers-of-korea/)를 발행하는 등 게임산업에서 영향력을 확장해나갔답니다.
에어브릿지는 [게임 성장에 특화된 맞춤형 기능](https://www.airbridge.io/ko/solutions/gaming)을 제공하고 있어요. 다양한 채널에서 진행되는 캠페인의 성과와 비용을 하나의 대시보드에서 통합 관리할 수 있고, 자체 개발한 사용자 행동 분석 특화 데이터베이스 LUFT엔진으로 실시간에 가까운 코호트 분석을 지원해요. 예를 들어 퍼널 리포트와 분 단위 리텐션 리포트, 소재별 성과 분석 등은 에어브릿지만의 특화 기능이에요. [웹젠](https://www.airbridge.io/ko/case-studies/mu-monarch-creative-optimization)은 영상 소재 최적화를 위해 에어브릿지의 액츄얼스와 트렌드 리포트로 성과를 확인한 것은 물론, 광고 채널 연동을 통해 리타게팅 및 캠페인 최적화를 진행했어요.
또한 유저별 기대 매출이 높은 채널을 선제적으로 식별하여 마케팅 비용을 최적화할 수 있도록 pLTV(Predictive Lifetime Value) 기능을 통해 유저가 앞으로 발생시킬 매출에 대한 예측값을 계산해 주고 있어요. 여기에 PC와 콘솔 게임에 관한 관심과 출시가 높아지고 있는 게임 업계의 트렌드를 반영해 앱, 웹, 콘솔까지 모든 게임 플레이 채널의 성과를 측정하고 분석하는 크로스 플랫폼 플레이 분석 기능도 제공해요. [넥슨](https://www.airbridge.io/ko/case-studies/nexon-cross-platform-tracking)은 에어브릿지를 활용해 [퍼스트 디센던트(The First Descendant)](https://tfd.nexon.com/ko/main)의 PC와 콘솔 광고 성과를 모바일에서와 같이 신뢰할 수 있는 데이터로 측정하며 캠페인을 고도화하고 있어요.
🎮 게임사를 위한 에어브릿지 쉽게 사용하기 가이드는 [여기](https://cshub.ab180.co/ko/docs/knowledges/knowledge-abr-for-gaming-1)에서 확인해 보세요.
[__wf_reserved_decorative]
## 에어브릿지를 활용하는 앱 수 90% 증가했어요
지난 한 해 동안 에어브릿지를 통해 성과를 트래킹하는 앱의 수가 2023년 전체보다 90% 증가했어요. 더 많은 앱이 에어브릿지를 통해 성과를 확인하고, 캠페인을 최적화하고 있어요.
2024년에는 국내 주요 광고 매체와의 연동성을 한 단계 더 강화해 주목을 받았어요. [카카오](https://www.airbridge.io/ko/blog/kakao-ads-audience-integration)와 [당근](https://www.airbridge.io/ko/blog/karrot-audience-sync) 오디언스 연동, [토스](https://www.airbridge.io/ko/blog/airbridge-toss-ads-integration)와 에브리타임과의 파트너십을 새롭게 채결했는데요. 국내 주요 광고 매체와 뛰어난 연동성으로 캠페인 성과를 에어브릿지에서 더 쉽게 확인하고, 더 정교한 광고 타겟팅을 운영할 수 있어요. [스페이스브이](https://cshub.ab180.co/ko/case-studies/airbridge-audience-spacev)는 에어브릿지 오디언스 기능을 활용해 타겟팅 광고 성과를 25% 개선하는 성과를 얻었어요.
한편, 앰플리튜드, 브레이즈와 같은 마테크 솔루션들과의 연동성도 빛을 발한 한 해였는데요. [마이리얼트립](https://cshub.ab180.co/ko/case-studies/myrealtrip-segmentation)과 [그린랩스](https://cshub.ab180.co/ko/case-studies/airbridge-braze-greenlabs-crm-retargeting)는 에어브릿지의 기여도 분석 결과를 브레이즈로 보내 개인화된 캠페인을 전개함으로써 리타겟팅을 통한 구매 전환율 12배 상승 등 가시적인 성과를 거두었어요. 앞으로도 에어브릿지는 국내외 다양한 마테크 솔루션들과의 연동성을 보다 강화하여 풀 퍼널 마케팅을 지원할 예정이에요.
에어브릿지는 각 고객사의 필요에 맞는 기능과 전문적인 지원으로 지속적인 호응을 받아왔어요. 전담 CS 매니저의 전문성은 곧 고객 성공으로 직결되는데요. [모니모니](https://cshub.ab180.co/ko/case-studies/airbridge-monymony-report)는 MMP 사용 경험 없는 상황에서 체계적인 CS 매니저의 지원으로 에어브릿지를 성공적으로 도입해 광고 성과 10배 이상 개선했어요.
엄격한 데이터 보안에도 빈틈이 없도록 힘쓴 한 해였어요. 에어브릿지는 고객사와 회원 데이터를 다중으로 보호하기 위해 글로벌 레벨의 데이터 보안을 준수하며 고객과의 신뢰를 지키기 위해 최선을 다하고 있어요. 지난 12월에는 엄격한 보안 기준 아래 정보보호 관리 체계를 모범적으로 운영한 공로를 인정받아 [과학기술통신부로부터 표창](https://www.linkedin.com/posts/ab180_%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80%EC%9D%98-%EB%8D%B0%EC%9D%B4%ED%84%B0-%EB%B3%B4%EC%95%88%EC%9D%80-%ED%83%84%ED%83%84%ED%95%A9%EB%8B%88%EB%8B%A4-%EC%A0%95%EB%B3%B4%EB%B3%B4%ED%98%B8-%EA%B4%80%EB%A6%AC%EC%B2%B4%EA%B3%84%EB%A5%BC-%EB%AA%A8%EB%B2%94%EC%A0%81%EC%9C%BC%EB%A1%9C-activity-7278270860489015296-ucqR/)을 받기도 했어요.
[__wf_reserved_decorative]
## 숫자로 보는 에어브릿지와 앱 성장
### 리포트 생성 수 1,415만
에어브릿지 리포트에서는 광고 채널, 트래킹 링크 등을 통해 수집한 데이터를 다양한 지표로 분석할 수 있어요. 예측 LTV를 확인할 수 있는 레비뉴 리포트, 유저의 잔존율 분석을 위한 리텐션 리포트 등 총 10가지의 리포트를 제공해 정확하고 체계적인 앱 성장 인사이트를 전하고 있어요. 지난해 에어브릿지와 함께한 고객사는 리포트를 총 1천만 건 이상 생성해 앱 성장을 보다 정확히 측정했어요.
지난해 에어브릿지와 함께한 [삼쩜삼](https://www.airbridge.io/ko/case-studies/3o3-web-to-app-conversion-tracking)은 액츄얼스, 리텐션, 퍼널 등 다양한 리포트를 활용해 마케터들이 실시간으로 데이터에 접근하고 분석할 수 있는 환경을 조성했어요. 그 결과 데이터를 직접 활용해 빠른 의사결정을 내리고, 실시간 피드백을 바탕으로 캠페인 최적화를 실행해 관련 업무에 소요하는 시간을 크게 단축하는 성과를 얻었어요.
생성한 리포트는 여러 앱과 성과 지표를 한번에 볼 수 있는 오버뷰 대시보드에 추가할 수 있어요. 오버뷰 대시보드에서는 사용자 목적에 맞춰 위젯, 차트, 테이블을 커스터마이징할 수 있어 하나의 화면에서 캠페인 성과를 빠르게 모니터링하고, 마케팅 전략을 최적화하며 더 효율적인 의사 결정을 내릴 수 있어요. 인도네시아 이커머스 플랫폼 [Segari](https://cshub.ab180.co/ko/case-studies/airbridge-overview-dashboard-segari)는 에어브릿지 오버뷰 대시보드로 마케팅 예산을 효율적으로 운용해 더 나은 ROI를 달성하고, 리포트 분석에 드는 반복 업무 시간을 50% 줄일 수 있었어요.
### 트래킹 링크 생성 수 8천613만, 클릭 수 2조6,280억
지난해 에어브릿지를 쓰는 앱은 8천만 개 이상의 트래킹 링크를 생성했어요. 생성된 에어브릿지 트래킹 링크는 매달 평균 200억 번, 1년간 총 2조 번 이상의 클릭을 기록하며 놀라운 활용성을 보여주었는데요. 글로벌 URL 단축 서비스를 제공하는 비틀리가 발표한 [월평균 100억 번](https://bitly.com/pages/resources/press/bitly-wraps-2022-surpassing-100m-in-arr-and-over-500k-global-customers)의 링크 클릭 수보다 2배 이상 높은 수치예요.
에어브릿지 트래킹 링크는 단순한 링크 생성 기능을 넘어 캠페인 데이터를 수집하고 딥링크 기능을 제공하고 있어요. 앱과 웹 등 온라인은 물론, QR코드를 활용해 오프라인 환경에서도 원하는 곳으로 끊김 없이 유저를 전송해요. 특히 카카오톡이나 네이버 카페 등 특히 국내 환경에 최적화된 끊김없는 딥링킹으로 주목을 받고 있어요. 트래킹 링크 성과는 실시간으로 에어브릿지에 전송되어 손쉬운 캠페인 성과 측정도 가능해요.
지난해 에어브릿지는 국내외 300여 개 이상의 기업이 사용하는 MMP의 기술과 노하우를 담은 딥링크 플랜을 새롭게 출시했어요. 구글 파이어베이스 다이나믹 링크 서비스 종료가 예정되면서 국내 환경에 가장 최적화된 딥링킹으로 에어브릿지 딥링크 플랜이 주목을 받았어요.
## 마무리하며
2024년은 에어브릿지가 숨 가쁘게 달리며 눈부신 성과를 만들어낸 한 해였어요. 게이밍에서의 눈에 띄는 성과와 국내외 파트너십 강화, 그리고 딥링크 플랜 출시까지. 에어브릿지는 성장과 혁신을 동시에 이뤄내고 있는데요. 이러한 여정은 보내주시는 지속적인 관심과 따뜻한 응원 덕분에 가능했어요. 2025년에도 에어브릿지는 마음을 다해 고객사와 함께 더 나은 성장을 만들어 나갈 것을 약속드려요!
에어브릿지에 대해 더 궁금한 점이 있다면 언제든 아래 버튼을 눌러 전문가와의 대화를 신청해 보세요.
---
## 앱스토어 최적화(ASO) 전략: 성과를 내기 위해 ‘이것’에 집중해 보세요
URL: https://www.airbridge.io/ko/blog/aso-strategy
Language: Korean
Published: 2024-12-20
최근 iOS 점유율이 늘면서 [앱스토어 최적화(ASO)](https://www.airbridge.io/ko/blog/aso-success-guide)가 주목받고 있어요. ASO는 애플 앱스토어에 있는 앱 이름, 설명, 이미지 등을 관리해 앱스토어 검색 결과 상단에 해당 앱을 노출하는 작업을 말해요. 검색엔진 최적화(SEO)와 유사해요.
ASO의 목표는 앱 다운로드라서 SEO보다 광고 성과에 더 직접적으로 영향을 주는 편이에요. 여기에 추가 비용이 필요 없는 작업이 많아서 하지 않을 이유가 없어요. 하지만 모두가 ASO에 필요한 작업에 투자하면서 앱 이름, 설명, 이미지 관리만으로 눈에 띄는 성과 개선으로 이어지지 않고 있어요.
그렇다면 어떻게 ASO를 해야 성과 개선으로 이어질까요? 먼저 ASO부터 살펴봐요.
## 앱스토어 최적화에 정답은 없어요
앱스토어 최적화(ASO)를 시작하기 전에 반드시 알아야 하는 점이 있어요. ASO에 정답은 없어요. 다른 최적화 작업처럼 ASO로 변하는 결과를 미리 예측하는 것은 어렵기 때문이에요. 애플 앱스토어, 구글플레이 스토어, 검색 엔진 등 다양한 플랫폼이 AI 기반 알고리즘을 도입하면서 예측은 더 어려워졌어요.
ASO, 정답은 없지만 다행히 많은 마케터의 경험에서 나온 정도는 있어요. 기본적인 권장사항은 아래와 같아요. 대부분의 마케터가 이미 알고 있는 내용이기도 해요.
- 앱 이름, 설명에 특정 키워드 사용하기
- 앱 카테고리를 가장 적합한 1순위, 2순위로 설정하기
- 앱 설명에 스크린샷 등 유저에게 도움이 되는 이미지 포함하기
- 글로벌 앱은 앱 이름, 설명 등을 현지화하기
## 앱스토어 최적화를 위한 체크리스트
여기서 조금 더! ASO는 권장사항을 한다고 바로 결과가 좋아지는 작업이 아니에요. 애플 앱스토어 알고리즘도 계속해서 업데이트되기 때문에 여러 가지를 꾸준히 관리해야 해요. 애플 앱스토어에서 효과적이라고 알려진 관리 항목은 아래와 같아요.
- 앱 리뷰 및 별점 관리하기
- 유저가 유입되는 키워드 파악하기
- TTR(Tap-through Rate)과 CVR(Conversion Rate) 관리하기
- 맞춤형 제품 페이지(CCP) 활용하기
이외에 경쟁사가 유리한 키워드 확인하기, 마케팅 트렌드 반영하기, 애플 앱스토어 정책 확인하기 같은 작업도 함께하면 더 좋아요. ASO에 대한 자세한 내용은 에어브릿지 핸드북에서 확인할 수 있어요.
- [에어브릿지 핸드북 03 - ASO](https://www.airbridge.io/ko/library/airbridge-handbook-aso)
대부분의 작업이 추가 비용이 필요 없기 때문에 마케터가 조금만 노력하면 성과 개선으로 이어질 수 있어요. 하지만 항상 좋은 결과로 이어지지 않아요. 모든 항목을 작업하고 관리해도 눈에 띄는 변화가 발생하지 않을 수 있어요.
## 애플 서치 애드가 ASO 성과를 도와줄 수 있어요
앱스토어 최적화(ASO)가 성과 개선으로 이어지지 않는 이유는 무엇일까요? 애플 앱스토어의 검색 결과 알고리즘이 어떤 방식으로 작동하는지 정확하게 알 수 없기 때문에 이 질문에 답하기 어려워요.
준비한 키워드를 경쟁사에 뺏겼을 수도 있고, 유저가 자주 사용하는 키워드가 아닐 수 있어요. 전혀 다른 이유일 수도 있어요. 이처럼 정확한 원인을 파악하기 쉽지 않아서 많은 마케터가 문제 해결에 어려움을 겪고 있어요.
여기서 알고리즘 때문이라고 포기하지 마세요. 마케팅 성공을 위한 더 나은 방법을 실행하세요. 바로 애플 서치 애드에요.
애플 서치 애드로 원하는 키워드를 확보하면 ASO 성과가 함께 개선되는 경향이 있어요. 애플 서치 애드로 유저가 검색하는 키워드를 확보해서 검색 결과 상단에 노출되면 이 결과가 ASO에 좋은 영향을 주는 방식이에요.
예를 들어 미국 스포츠 중계 서비스를 제공하는 현지 기업 A사는 경쟁사 B사보다 앞서 애플 서치 애드로 주요 키워드 확보해 B사보다 더 높은 우선 순위로 노출되는 데 성공했어요. 여기서 중요한 점은 광고를 시작했다고 기존에 하던 ASO를 멈추지 않았다는 점이에요.
오히려 A사는 애플 서치 애드 광고와 ASO 전략이 시너지 나도록 여러 전략을 조율했어요. ASO 전략이 애플 서치 애드 광고 성과 극대화에 도움이 되고, 애플 서치 애드 광고가 ASO 전략을 개선하기 때문이에요.
- 애플 서치 애드 광고 키워드와 ASO 키워드를 통일했어요
- CCP를 활용해 키워드로 유입된 유저의 전환율을 증가시켰어요
- 그외에 ASO에 필요한 항목을 꾸준히 관리했어요
## ASO, 모든 것이 준비된 에어브릿지와 함께 하세요
앱스토어 최적화(ASO) 성과 개선을 위해 애플 서치 애드를 시작하면 챙겨야 할 것이 많아요. 애플 서치 애드 성과를 주기적으로 확인하면서 예산과 키워드를 관리해야 하고, 기존 ASO 전략도 유지해야 해요. 그래서 광고 성과를 쉽게 확인할 수 있는 MMP와 애플 서치 애드 정책을 잘 이해하고 있는 마케팅 전문가 필요해요.
에어브릿지는 애플 서치 애드 연동을 통해 매일 광고 성과와 비용을 쉽게 확인할 수 있는 리포트로 다양한 분석과 함께 제공해요. 광고 채널을 여러 개 운영하고 있다면 다양한 지표를 바로 확인할 수 있는 에어브릿지 리포트는 필수에요.
- [애플 서치 애드 연동하기](https://help.airbridge.io/ko/guides/apple-search-ads)
- [에어브릿지 리포트 시작하기](https://help.airbridge.io/ko/guides/airbridge-report-overview)
여기에 애플 서치 애드 정책을 한국에서 가장 잘 이해하고 있는 전문가가 에어브릿지를 개발하고 있어요. 모든 것이 준비된 에어브릿지에게 번거로운 일은 맡기고 더 중요한 것에 집중하세요. 가장 편한 MMP 에어브릿지가 마케터의 성공을 도울게요.
---
## 당근 오디언스 연동, 꼭 필요한 유저에게만 광고 집행하세요
URL: https://www.airbridge.io/ko/blog/karrot-audience-sync
Language: Korean
Published: 2024-12-16
당근은 ‘당신 근처의 지역 생활 커뮤니티’라는 슬로건으로 중고거래부터 동네 생활 커뮤니티까지 다양한 지역 기반 활동이 가능한 슈퍼앱이죠. 이러한 특징 덕분에 위치정보 기반 광고에 특히나 강점이 있는 채널이기도 해요. 에어브릿지가 당근과 오디언스를 연동했어요. 이를 바탕으로 앱 마케터가 더 높은 정확도의 타겟 마케팅이 가능하도록 도울 수 있게 되었어요.
## 당근비즈니스 알아보기
[당근비즈니스](https://business.daangn.com/)에서는 4,000만 명 이상의 유저 데이터를 활용해서 당근에 퍼포먼스 광고를 집행할 수 있어요. 이외에도 지역 기반, 인구통계학, 관심사 등 정교한 조건으로 원하는 타겟을 설정할 수 있죠. 당근과 에어브릿지의 오디언스 연동으로 에어브릿지를 사용 중인 앱에서 당근에 더 높은 정확도의 광고를 집행할 수 있게 되었어요.
## 당근 오디언스 연동으로 기대되는 점
[__wf_reserved_decorative]
에어브릿지 당근 오디언스 싱크는 우리 앱에서 수집된 고객의 퍼스트 파티 데이터(First-Party Data)를 활용해 마케터가 원하는 세그먼트를 생성하고 이를 당근 전문가모드 광고 플랫폼에 직접 전송하는 기능이에요.
구체적인 내용은 다음과 같아요.
- 앱 설치, 회원가입, 구매 여부 등 특정 조건을 만족하는 오디언스 세그먼트를 에어브릿지에서 생성하고, 당근 전문가모드 광고 플랫폼으로 전송하여 정교한 타겟팅 광고를 집행할 수 있어요.
- 에어브릿지에서 전송되는 고객 데이터는 일 단위로 당근에도 실시간 업데이트되어요.
당근 오디언스 싱크 기능을 활용하면 앱에서 수집된 실시간 고객 데이터와 당근 데이터를 기반으로 정밀한 타겟팅을 실행할 수 있어요. 물론 필요에 따라서는 디타겟팅과 유사 타겟팅도 가능해요.
광고 채널로 오디언스를 전송하는 자세한 방법은 [여기](https://help.airbridge.io/ko/guides/send-audience?_gl=1*un9k8r*_gcl_au*OTc2NzY4NjIwLjE3MzM5OTIzNDQ.#오디언스를-전송할-수-있는-광고-채널-1)를 참고해 주세요.
## 마무리하며
에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 크리테오, 알티비하우스, 애피어, 리머지, 카울리, 퍼스널리, 카카오 등 다양한 매체와 오디언스를 연동하고 있어요. 이번 당근 연동에 힘입어 국내 지역 기반 마케팅이 더욱 정교해질 거에요.
당근 오디언스 연동은 퍼스트파티 데이터의 중요성이 강조되고 있는 지금 더 의미 있는 소식이에요. 당근 광고를 운영하고 있다면 에어브릿지의 새로운 기능을 적극 활용해 보세요.
지금 바로 에어브릿지 데모를 신청해 당근 오디언스 연동 기능을 만나보세요.
👉 [에어브릿지 무료로 사용해 보기](https://www.airbridge.io/ko/book-a-demo)
---
## 어트리뷰션 데이터로 CRM 마케팅 성과 극대화하기: 전략, 성공 사례
URL: https://www.airbridge.io/ko/blog/maximize-crm-performance-with-attribution
Language: Korean
Published: 2024-12-12
개인화는 성공적인 CRM 캠페인을 위한 핵심 요소 중 하나에요. 하지만 개인화된 메시지를 위해서는 사용자 속성이나 이벤트 속성 등이 필요하죠. 많은 기업이 이러한 데이터 포인트를 충분히 수집하지 못해 개인화 메시지를 활용하지 못하고 있어요. 이런 상황에서 어트리뷰션 데이터는 큰 도움을 줄 수 있어요.
어트리뷰션 데이터란 사용자가 광고를 보고 앱을 설치하거나 구매하는 등의 행동을 할 때, 어떤 광고나 마케팅 채널이 그 행동에 영향을 미쳤는지 추적하고 분석하는 정보를 말해요.어트리뷰션 데이터는 사용자 획득(UA)했을 때 확보한 채널, 캠페인, 소재, 키워드와 같은 맥락 정보를 CRM 캠페인에 제공해, 사용자 반응률과 전환율을 효과적으로 높이는 데 기여해요.
어트리뷰션 데이터를 활용하면 앱 설치부터 온보딩 단계, 리타겟팅 캠페인에 이르기까지 극대화된 마케팅 성과를 기대할 수 있는데요. 유저 데이터 기반의 맞춤형 메시지로 높은 전환율과 사용자 참여를 유도할 수 있기 때문이에요. 어트리뷰션 데이터를 활용한 CRM 캠페인의 힘, 구체적인 사례를 통해 알아볼게요.
## 어트리뷰션 데이터를 활용한 온보딩 성과 개선
사용자가 앱을 설치하고 온보딩을 시작하는 초기 단계에서는 사용자의 선호도나 행동, 관심사에 대한 정보가 부족해 개인화된 경험을 제공하기 어려워요. 이때 어트리뷰션 데이터가 핵심 역할을 해요. 어트리뷰션 데이터는 사용자가 어떤 캠페인, 채널, 소재, 또는 키워드를 통해 앱을 설치했는지 알려줘요. 데이터를 통해 사용자의 의도와 관심사를 파악하고, 개인화된 온보딩 경험을 제공할 수 있어요.
예를 들어 한 유저가 여성 패션 광고를 보고 앱을 설치했다면, 온보딩 과정에서 여성 패션 콘텐츠와 프로모션을 우선으로 보여줄 수 있어요. 이런 방식으로 개인화된 경험을 제공하면 사용자 참여도를 크게 높일 수 있어요. 흔히 “소재는 곧 타겟팅”이라는 말처럼, 어떤 광고 소재를 통해 유입되었는지가 유저의 의도를 드러내는 중요한 힌트가 돼요.
### 마이리얼트립 개인화 CRM 마케팅 사례
[__wf_reserved_decorative]
마이리얼트립 사례
국내 여행 스타트업 마이리얼트립(MyRealTrip)은 DA광고의 성과를 에어브릿지로 측정하고, 획득한 어트리뷰션 데이터를 브레이즈로 전달했어요. 이 데이터를 기반으로 유저 세그먼트를 생성해 개인화된 메시지를 보낸 결과, 기존 캠페인 대비 CTR이 5%p 이상 상승하는 성과와 함께, DA 리타겟팅 소재 4개 모두 CTR이 최소 2%p에서 최대 21%p까지 증가했어요. 결과적으로 구매를 망설이던 고객들에게 최적화된 푸시 메시지를 전달해 전환율이 12배 이상 상승하는 놀라운 성과를 이뤄냈어요.
[👉 마이리얼트립의 성공 사례를 자세히 만나보세요](https://cshub.ab180.co/ko/case-studies/braze-segmentation-myrealtrip)
## 어트리뷰션 데이터를 활용한 리타겟팅 캠페인 성과 개선
CRM에서도 리타겟팅이 가능하다는 사실, 알고 계셨나요? 메시지에 대한 반응은 유저의 관심도를 보여주는 중요한 지표예요. 메시지에 반응한 유저들만 타겟팅하는 메시지를 보내면 최종 전환율을 높이는 동시에 유저 피로도를 줄일 수 있어요.
브레이즈에서는 푸시, 이메일, 인앱 메시지 등의 클릭 데이터를 기반으로 세그먼트를 생성하거나 유저 여정의 분기를 나눌 수 있어요. 하지만 카카오 친구톡, 알림톡, SMS 링크 클릭 정보는 Braze에서 수집이 어렵죠. 여기서 에어브릿지가 유용하게 활용돼요.
### 그린랩스 CRM 리타겟팅 캠페인 사례
[__wf_reserved_decorative]
그린랩스 사례
그린랩스(GreenLabs)는 에어브릿지를 활용해 친구톡의 CTA 버튼 클릭시 딥링크 오픈 이벤트로 수집하고, 이를 브레이즈에 전송해 리타겟팅 캠페인을 진행했어요. 그 결과 비교 모수 대비 오픈율이 약 2배 상승했어요. 또한, 유저에게 더욱 관련성 높은 메시지를 제공해 광고에 대한 피로도를 줄이는 효과도 거두었어요.
[👉 그린랩스의 성공 사례를 자세히 만나보세요](https://cshub.ab180.co/ko/case-studies/airbridge-braze-crm-anlaytics)
## 에어브릿지와 브레이즈 연결의 강력함
에어브릿지는 브레이즈와의 서버 투 서버(S2S) 연동을 지원해 고객이 쉽게 두 플랫폼을 연결할 수 있어요. 간단한 데이터 입력과 클릭 몇번 만으로 실시간 어트리뷰션 데이터를 브레이즈에 보낼 수 있고, 이를 기반으로 개인화된 마케팅 캠페인을 즉시 실행할 수 있어요. 어트리뷰션 데이터는 채널, 캠페인, 광고 그룹, 소재 정보뿐만 아니라 딥링크 오픈 데이터까지 포함하기 때문에 온보딩과 리타겟팅 시 모두 유용하게 활용할 수 있어요.
## 마무리하며
어트리뷰션 데이터는 CRM 캠페인의 성과를 극대화할 수 있는 강력한 도구예요. 이를 통해 온보딩 흐름을 개인화하고, 관련성 높은 리타겟팅 메시지를 설계해 사용자 참여와 전환율을 모두 높일 수 있어요.
에어브릿지와 브레이즈 연동으로 한층 더 최적화된 CRM 캠페인을 경험해보세요!
---
## 카카오 오디언스 연동으로 광고 타겟팅이 정교해졌어요
URL: https://www.airbridge.io/ko/blog/kakao-ads-audience-integration
Language: Korean
Published: 2024-12-02
카카오톡을 실행할 때 가장 먼저 보이는 배너 광고, 다들 익숙하실 텐데요. 카카오 디스플레이 광고를 운영하고 있거나, 운영을 고려하는 분들을 위한 좋은 소식이 있어요. 바로 에어브릿지의 카카오 오디언스 연동 기능 출시 소식이에요. 이제 에어브릿지와 함께라면 카카오 광고 타겟팅 운영이 더욱 간편해져요.
### 카카오 광고 알아보기
[카카오 광고](https://business.kakao.com/)는 카카오톡, 다음, 카카오스토리 등 국내 대표 모바일 플랫폼에서 국민 대다수에게 노출하는 광고를 운영하는 플랫폼이에요. 텍스트, 이미지, 동영상 등 다양한 형태의 광고를 니즈에 맞게 운영할 수 있고, 카카오의 빅데이터를 활용한 정교한 타겟팅으로 효과적인 광고 집행이 가능해요.
이러한 강점을 가진 카카오 광고와 에어브릿지가 만나 더욱 효과적인 캠페인 운영이 가능해졌어요.
### 카카오 오디언스 연동으로 기대되는 점
[]
에어브릿지 ‘카카오 오디언스 싱크’는 퍼스트 파티 데이터(First-Party Data)를 활용해 정교한 고객 세그먼트를 생성하고, 이를 카카오 광고 플랫폼에 직접 전송하는 기능이에요.
구체적인 내용은 다음과 같아요
- 앱 설치, 회원가입, 구매 등 특정 조건을 만족하는 오디언스 세그먼트를 에어브릿지에서 생성하고, 카카오 광고 플랫폼에서 정교한 타겟팅 광고를 집행할 수 있어요.
- 에어브릿지의 최신 데이터를 일일 단위로 카카오에 자동 업데이트해요.
카카오 오디언스 싱크를 활용하면 광고주는 실시간 고객 데이터를 기반으로 정밀한 타겟팅을 실행할 수 있어요. 이는 곧 에어브릿지를 사용하는 기업은 물론, 소비자에게도 긍정적인 영향을 기대할 수 있어요. 유저 맞춤형 광고는 곧 고객 경험 개선으로 이어지기 때문이에요.
에어브릿지는 메타 애즈, 구글 애즈, 틱톡 포 비즈니스, 크리테오, 알티비하우스, 애피어, 리머지, 카울리, 퍼스널리 등 다양한 매체와의 오디언스 연동을 지원하고 있어요. 이번 카카오 연동에 힘입어 더욱 정교해질 에어브릿지를 기대해 주세요.
광고 채널로 오디언스를 전송하는 자세한 방법은 [여기](https://help.airbridge.io/ko/guides/send-audience#%EC%98%A4%EB%94%94%EC%96%B8%EC%8A%A4%EB%A5%BC-%EC%A0%84%EC%86%A1%ED%95%A0-%EC%88%98-%EC%9E%88%EB%8A%94-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-1)를 참고해 주세요.
### 마무리하며
이번 에어브릿지와 카카오의 오디언스 연동 지원으로 더욱 정교하고 효과적인 캠페인 전략을 수립할 수 있게 되었어요. 특히 퍼스트 파티 데이터의 중요성이 강조되고 있어 더 의미 있는 소식인데요. 카카오 광고를 운영하고 있다면 에어브릿지의 새로운 기능을 적극적으로 활용해 보세요.
지금 바로 에어브릿지 데모를 신청해 카카오 오디언스 연동 기능을 만나보세요.
[👉 지금 에어브릿지 무료 체험하기](https://abr.ge/@ab180themodel/blog?campaign=kakao-ads-audience-integration&content=cta-demorequest&og_tag_id=113111609&routing_short_id=hh2uhj&sub_id=abr-ko&tracking_template_id=d2cb3ebd7229558114d2af97a50d4438&ad_type=click)
---
## Airbridge 딥링크가 필요한 5가지 팀 유형
URL: https://www.airbridge.io/ko/blog/5-team-who-need-airbridge-deeplink-plan
Language: Korean
Published: 2024-11-26
앱 내 특정 위치로 유저를 보내주는 딥링크(Deeplink)는 CRM부터 광고, 리퍼럴, 오프라인 이벤트 등 성공적인 앱 마케팅 캠페인을 위해 꼭 필요한 기능이죠. 하지만 딥링크를 직접 개발하고 운영하는 데에는 많은 시간과 리소스가 필요해요.
이전까지 많은 앱에서는 무료로 제공되던 구글 파이어베이스 다이나믹 링크(Dynamic Link, 동적 링크)를 통해 딥링크를 만들고 사용했어요. 하지만 얼마 전 구글은 다이나믹 링크 서비스 중단을 발표했죠. 이에 따라 다이나믹 링크를 사용 중인 앱에서는 자체 딥링크를 개발하거나 새로운 딥링킹 서비스를 찾고 있어요.
에어브릿지 또한 파이어베이스 다이나믹 링크를 대체할 딥링킹 전용 서비스 [딥링크 플랜(Deeplink Plan)](https://www.airbridge.io/ko/deeplink-plan)을 출시했어요. 이번 포스트에서는 에어브릿지 딥링크 플랜을 어떤 팀에게 추천하는지 5가지 유형으로 정리해 보았어요. 여러분의 앱이 다음 다섯 가지 상황 중 하나에 속해 있다면 에어브릿지와 딥링크 플랜에 대해 이야기 나누어보세요.
> 💡 들어가기 전에! 딥링크 플랜이 궁금하다면?[
> 에어브릿지 딥링크 플랜 자세히 알아보기](https://abr.ge/@ab180themodel/blog?ad_group=product&campaign=5-team-who-need-airbridge-deeplink-plan&content=cta1-dlp&og_tag_id=111212702&routing_short_id=joji3k&sub_id=abr-ko&tracking_template_id=11b7ed0dc1db61b0b0ca639d2f892925&ad_type=click)
### 1\. 🚚 이관 | 파이어베이스 다이나믹 링크를 사용 중인 팀
2025년 8월 25일, 구글 파이어베이스 다이나믹 링크가 중단되면 새로운 링크를 생성할 수 없을 뿐만 아니라 기존 링크 마이그레이션도 불가능해질 예정이에요. 다이나믹 링크를 사용 중인 팀이라면 서비스가 중단되기 전에 새로운 서비스를 찾아야겠죠? 이관 서비스를 찾을 때에는 이미 사용 중인 링크 마이그레이션 방법이 얼마나 간편한지도 중요할 거예요.
에어브릿지 딥링크 플랜은 별도의 개발 없이도 5분 이내로 사용 중인 모든 다이나믹 링크를 마이그레이션할 수 있는 [마이그레이터](https://www.airbridge.io/ko/blog/how-to-migrate-deeplink-in-airbridge)를 제공해요. 또한 국내 지면에서 최적화된 유저 경험, 자동 QR코드 제작, 디퍼드 딥링크 및 커스텀 링크를 통한 브랜딩 강화 등 다이나믹 링크에서 사용하던 기능은 그대로 다양한 추가 기능들까지 한 번에 이용할 수 있죠.
### 2\. 💻 개발 | 딥링크 개발에 어려움을 겪고 있는 팀
팀에서 직접 딥링크를 개발하고 있으신가요? 딥링크는 단순 개발 이외에도 유지와 보수에 많은 리소스가 필요해요. URI 스킴(URI Scheme), 유니버셜 링크(Universal Link)와 앱 링크(App Link) 개발 등 기초적인 딥링크 개발 이외에도 OS, 모바일 브라우저 및 인 앱 브라우저별로 원활한 오픈을 지원하려면 지속적인 모니터링이 필요해요.
또, 하나의 링크로 iOS와 안드로이드 모두에서 자유롭게 오픈되는 다이나믹 딥링킹, 홈 화면을 거치지 않고 유저가 원하는 페이지로 바로 이동시켜주는 디퍼드 딥링크(Deferred Deeplink) 등 고려해야 할 사항들이 많아요.
현재 딥링크 개발에 어려움을 겪고 계시나요? 에어브릿지는 이미 다양한 OS와 지면에 대응하는 딥링크를 지원하고 있어요. 딥링크 개발은 에어브릿지에게 맡기고 더 중요한 일에 집중하세요. 딥링크 개발과 예외사항에 대해 보다 자세히 알고 싶으시다면 [에어브릿지의 딥링크 핸드북](https://www.airbridge.io/ko/library/airbridge-handbook-deeplinks)을 확인해 보세요.
### 3\. 🔗 사용자 경험 | 딥링크 원활한 오픈이 안되는 팀
앱 서비스가 딥링크를 사용해야 하는 가장 큰 이유 중 하나는 원활한 사용자 경험이에요. 만약 유저가 A상품 광고를 통해 유입되었는데 앱을 열었을 때 원하는 페이지가 아닌 다른 페이지가 오픈되어 다시 상품을 찾아야 하거나, 앱이 오픈되지 않는다면 유저들은 좋지 못한 경험을 느끼고 앱을 이탈하게 될 거예요. 그만큼 링크가 자주 공유되거나 광고가 집행되고 있는 채널에서 원활한 딥링크 오픈이 중요하죠.
딥링킹 서비스 중에는 국내 지면에서 원활한 오픈과 디퍼드 딥링킹을 지원하지 않는 경우가 종종 있어요. 여러분이 사용 중인 서비스가 카카오톡이나 네이버 등 국내 주요 지면에서 제대로 오픈되고 있지 않은가요? 앱 대신 모바일 브라우저가 열리거나, 원하는 페이지로 안정적인 디퍼드 딥링킹이 되지 못하고 있나요? 에어브릿지는 국내 마케팅 시작에 최적화된 딥링킹을 제공해요.
### 4\. 📆 앱 런칭 | 신규 앱 출시를 앞두고 있는 팀
새로운 앱 오픈을 앞두고 계시나요? 그렇다면 딥링크를 자체 개발할지 딥링킹 서비스를 활용할지에 대한 의사 결정이 필요하죠. 구글에서는 [신규 출시 예정인 앱에 파이어베이스 다이나믹 링크를 사용하는 것을 권장하지 않고](https://firebase.google.com/support/dynamic-links-faq?hl=ko) 있어요. 만약 지금 당장 편의를 위해 다이나믹 링크를 사용한다 하더라도 곧 중단 시점이 다가오기 때문에 결과적으로 출시 시점부터 딥링킹 서비스를 사용하는 것이 더 유리할 거예요.
에어브릿지는 신규 앱에서도 딥링크 플랜을 부담 없이 사용할 수 있도록 MAU 단위의 가격제를 제공하고 있어요.
### 5\. 💵 가격 | 합리적 가격의 딥링킹 솔루션을 찾고 있는 팀
이미 새로운 딥링킹 서비스를 찾고 있다면 기능만큼이나 가격이 얼마인지도 의사결정에 중요한 요소겠죠. 에어브릿지 딥링크 플랜은 여러분이 운영 중인 앱 MAU에 맞는 합리적인 가격을 제공하고 있어요. 자세한 가격은 아래 표를 통해 확인해 보세요.
[__wf_reserved_decorative]
> 여러분의 앱이 위 다섯 가지 유형에 속하거나 혹은 그렇지 않더라도 에어브릿지 딥링크 플랜 도입이 궁금하시다면 아래 버튼을 통해 에어브릿지에 연락 주세요. 에어브릿지 딥링크 전문가들이 여러분의 앱 상황에 맞춰 친절히 안내해 드릴게요.
---
## Airbridge + Hive 연동으로 게임 성장 극대화하기
URL: https://www.airbridge.io/ko/blog/hive-platform-integration
Language: Korean
Published: 2024-11-26
에어브릿지가 게임 개발 및 운영 전문 플랫폼인 하이브와의 연동을 지원해요. 이번 연동은 게임 개발자분들이 마케팅을 더 쉽고 효율적으로 관리할 수 있도록 도와드리기 위해 준비되었어요. 이제 에어브릿지 SDK를 하이브 플랫폼과 간단하게 연동하고 다양한 채널과 크리에이티브, 지역에서 캠페인을 손쉽게 측정하고 분석하며 개선할 수 있어요.
### 하이브 플랫폼 연동 상세 내용
- 하이브 SDK v4.24.4.0 이상(모바일) 또는 v24.4.0 이상(PC)부터 에어브릿지 SDK를 하이브 대시보드에서 바로 연동할 수 있어요.
- 최신 SDK로 업데이트만 하면 별도 설치 없이 에어브릿지를 몇 초 만에 활성화할 수 있어요.
- 에어브릿지 어트리뷰션 데이터와 하이브의 동기화가 지원 돼요. 실시간 어트리뷰션 인사이트를 바탕으로 유저 획득, 참여도, 수익화를 더 효과적으로 최적화할 수 있어요.
### 하이브 플랫폼 소개
[하이브 플랫폼](https://hiveplatform.ai/)은 컴투스플랫폼이 만든 게임 개발 및 운영 솔루션으로 그라비티, 플레이위드, 블루포션게임즈, 썸에이지, 하이브로 등 한국과 글로벌 게임 개발사들이 신뢰하는 플랫폼이에요. 주요 기능은 다음과 같아요.
- 개발자 SDK: 유니티와 언리얼 엔진과 연동을 통해 게임을 빠르게 개발할 수 있어요.
- LiveOps 포털: 이벤트 프로모션과 유저 참여 관리 등 실시간 게임 운영 도구를 제공해요.
- 게임 백엔드: 인증, 결제, 푸시 알림, 블록체인 통합 등 다양한 백엔드 서비스를 지원해요.
- 광고 수익화: 게임 플레이를 방해하지 않으면서도 효과적인 광고 수익화를 지원해요.
- 데이터 & 분석: 플레이어 행동과 유저 퍼포먼스를 이해할 수 있는 게임 특화 데이터 분석 서비스를 제공해요.
### 게임 성장과 에어브릿지
에어브릿지는 모바일, 웹, PC, 콘솔에서 실시간 크로스 플랫폼 인사이트를 제공하는 글로벌 모바일 측정 파트너(MMP)입니다. 넥슨과 웹젠, 후야호 등 주요 게임 퍼블리셔들이 신뢰하는 에어브릿지는 코호트 분석, 사기 방지, 딥링크, 원본 데이터 추출, LTV 예측 등 다양한 기능을 제공하고 있어요. 특히 이 모든 지표를 통합 대시보드에서 한눈에 확인할 수 있어요. 게임 비즈니스에 특화된 에어브릿지 솔루션에 대해 자세히 알고 싶다면, [여기](https://www.airbridge.io/ko/solutions/gaming)에서 확인해 보세요.
### 마무리하며
이번 통합은 하이브의 강력한 게임 개발 도구와 에어브릿지의 고도화된 분석 기술을 하나로 모아 탄생했어요. 게임 개발자분들이 더 쉽게 성과를 측정하고 운영을 최적화하며, 유저 획득과 참여도, 수익화를 극대화할 수 있어요.
에어브릿지와 하이브 플랫폼이 함께 만드는 게임 성장을 경험해보고 싶으시다면, 지금 바로 데모를 신청해 보세요!
[👉 지금 에어브릿지 데모 요청하기](https://www.airbridge.io/ko/book-a-demo)
---
## Airbridge와 adjoe 연동으로 게임 도달 범위 확대하기
URL: https://www.airbridge.io/ko/blog/adjoe-integration
Language: Korean
Published: 2024-10-25
에어브릿지와 [adjoe](https://adjoe.io/ko/)가 연동 파트너십을 체결했어요. 이제 에어브릿지에서 adjoe 연동으로 게임의 도달 범위를 확대하고 성과를 최적화할 수 있어요. adjoe의 게임 특화 광고 형식과 에어브릿지의 강력한 분석 기능을 결합한 이번 파트너십은 사용자를 확보하고 수익화 전략을 강화하는 데 도움이 돼요.
## adjoe의 게임 핵심 플레이어 활용하기
[]
새로운 모바일 게임에 관심이 많은 전 세계 2억 명 이상의 유저에게 도달해 성과를 극대화해보세요. 이번 연동을 통해 높은 참여도를 보이는 플레이어에게 노출할 수 있는 adjoe의 다양한 광고 형식은 다음과 같아요.
1. 플레이타임 보상 광고: 플레이어가 게임을 진행하는 동안 지속적으로 보상을 받아요.
2. 보상형 비디오: 플레이어가 인게임 보상을 받기 위한 비디오 광고를 시청해요.
3. 전면 광고: 게임 플레이의 자연스러운 휴식 시간에 전체 화면 광고가 나타나요.
4. 배너 광고: 게임 플레이 중 화면 상단이나 하단에 배치되어 지속적인 브랜드 노출을 제공해요.
5. MREC 광고: 게임 콘텐츠와 자연스럽게 어우러져 사용자 경험을 향상시키면서 다른 앱을 홍보해요.
## 에어브릿지와 게임 프로모션 강화하기
에어브릿지의 분석 기능과 adjoe의 광고 솔루션을 연동하면 다음과 같은 이점을 얻을 수 있어요:
- 상세한 어트리뷰션 데이터를 통해 가장 효과적인 광고 형식과 캠페인을 식별할 수 있어요.
- 코호트 분석을 통해 광고 노출에 따른 장기적인 사용자 행동을 이해할 수 있어요.
- A/B 테스트를 활용하여 광고 배치와 메시지를 최적화할 수 있어요.
- 실시간 보고서를 통해 빠르게 캠페인 조정을 할 수 있어요.
- 여러 채널의 데이터를 통합하여 마케팅 성과를 전체적으로 파악할 수 있어요.
에어브릿지와 adjoe 연동을 통해 게임의 가시성을 높이고, 더 많은 플레이어를 확보하며, 장기적인 참여를 유도해 경쟁이 치열한 모바일 게임 시장에서 성공할 수 있어요.
## 시작하기
에어브릿지와 adjoe 연동은 간단해요. 다음 단계를 따라 시작해보세요:
1. 에어브릿지에서 adjoe 캠페인을 위한 트래킹 링크를 설정하세요.
2. 두 플랫폼 간의 실시간 데이터 흐름을 보장하기 위해 포스트백을 구성하세요.
3. 에어브릿지의 분석 기능을 활용하여 사용자 행동과 캠페인 성과에 대한 인사이트를 얻으세요.
자세한 지침은 [에어브릿지 유저 가이드](https://help.airbridge.io/en/guides/adjoe)를 참고하세요. 에어브릿지와 adjoe가 게임의 성장을 어떻게 부스트할지 궁금하다면, 지금 바로 [에어브릿지 팀에 문의](https://abr.ge/@ab180themodel/blog?ad_group=crm&campaign=ab180-adjoe-integration-ko&content=cta-request-a-demo&og_tag_id=116115612&routing_short_id=swwq6c&sub_id=ab180&tracking_template_id=2085aace4a68597a877ff133c8778bd4&ad_type=click)하세요!
👉 [지금 에어브릿지 데모 신청하기](https://abr.ge/@ab180themodel/blog?ad_group=crm&campaign=ab180-adjoe-integration-ko&content=cta-request-a-demo&og_tag_id=116115612&routing_short_id=swwq6c&sub_id=ab180&tracking_template_id=2085aace4a68597a877ff133c8778bd4&ad_type=click)
---
## Firebase에서 Airbridge로 딥링크 노코드 마이그레이션하기
URL: https://www.airbridge.io/ko/blog/how-to-migrate-deeplink-in-airbridge
Language: Korean
Published: 2024-10-25
파이어베이스 다이나믹 링크 종료가 다가오며 구글에서는 [새로운 다이나믹 링크 생성 중단과 빠른 마이그레이션](https://firebase.google.com/support/dynamic-links-faq?hl=ko#how_much_time_will_i_have_to_migrate)을 권장하고 있어요. 2025년 8월 25일 파이어베이스 다이나믹 링크 서비스가 종료되면 기존에 만들어둔 딥링크가 작동되지 않고 마이그레이션도 어려워지기 때문이에요.
다이나믹 링크에서 새로운 딥링킹 서비스로 옮겨갈 때 이미 사용 중이던 링크에 대한 마이그레이션도 중요하겠죠.
에어브릿지 딥링크 플랜에서는 에어브릿지가 제공하는 마이그레이터와 제너레이터를 통해 별도의 개발 없이 사용 중이던 모든 링크를 마이그레이션 할 수 있어요. 이번 블로그 포스트에서는 에어브릿지를 통한 다이나믹 링크 마이그레이션이 얼마나 간편한지 시연 영상과 함께 살펴보겠습니다.
### STEP ❶ | 링크 추출하기
다이나믹 링크를 마이그레이션 하려면 우선 파이어베이스에서 기존에 사용 중이던 링크를 추출해야 해요. 파이어베이스 테이크아웃 기능으로 기존에 사용 중인 링크를 CSV로 다운로드 받을 수 있어요.
### STEP ❷ | 기존링크 파라미터 추출하기
다이나믹 링크를 모두 다운로드했다면, 다운로드한 링크들을 에어브릿지 트래킹 링크 마이그레이터 속 **다이나믹 링크 입력** 시트에 붙여 넣으세요.
붙여 넣은 다이나믹 링크 파라미터 값은 **파라미터 매핑 결과** 시트에서 확인 후 복사할 수 있어요.
### STEP ❸ | 복사한 파라미터 마이그레션하기
파라미터 값을 복사했다면 에어브릿지 트래킹 링크 제너레이터 속 **링크 생성** 시트에 붙여 넣으세요.
**링크 생성**시트 상단바 **확장 프로그램**에서 **Airbridge Tracking Link Generator**를 클릭하고 **Generate Tracking Link**를 실행하세요.
작업이 완료되면 시트 우측에 마이그레이션 된 에어브릿지 트래킹 링크를 확인할 수 있습니다. 해당 링크들은 에어브릿지 대시보드 속 **트래킹 링크 관리**에서도 동일하게 확인 가능해요.
### STEP ❹ | 우리 서비스에 적용하기
마이그레이션이 끝났나요? 이제 남은 건 마이그레이션된 링크를 서비스에 적용하는 일! 기존 다이나믹 링크들을 마이그레이션된 링크로 교체해 주면 마이그레이션 작업이 완료됩니다.
> 사용 중이던 링크 개수에 상관없이 한 번에 마이그레이션 가능한 에어브릿지 딥링크 플랜에 대해 궁금하시다면 지금 에어브릿지와 이야기 나누어 보세요. 데모를 신청하면 에어브릿지 전문가가 연락해 여러분의 딥링크를 점검해 드릴게요.
#### [데모 신청하기 ➡️](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=how-to-migrate-deeplink-in-airbridge&content=cta1-241025-deeplink-plan&og_tag_id=103166889&routing_short_id=wuqjs0&sub_id=abr-ko&tracking_template_id=ae321d2a133e29af6be8c7d5d0006bc2&ad_type=click)
---
## 전환 API: 옴니채널 데이터 활용으로 마케팅 성과를 극대화해 보세요
URL: https://www.airbridge.io/ko/blog/capi-for-omni
Language: Korean
Published: 2024-10-24
디지털 광고 세계에서 개인정보 보호 규제가 강화되는 흐름은 이제 더 이상 부정할 수 없는 사실이에요. 구글이 개인정보보호 강화를 목표로 발표한 구글 프라이버시 샌드박스는 장기적으로 광고 식별자인 GAID 운영과 서드파티 쿠키 운영의 중단을 예고했죠. [2024년 7월 서드파티 쿠키 지원 중단 계획은 철회되었지만, 크롬 브라우저에 유저가 서드파티 쿠키 차단 여부를 선택할 수 있도록 하는 업데이트를 도입할 예정이라고 밝혔어요.](https://blog.google/intl/ko-kr/company-news/technology/privacysandboxupdates/) 애플이 지난 2021년부터 ATT(App Tracking Transparency) 정책을 시행하면서 유저가 데이터 제공 동의 여부를 선택하도록 묻는 프롬프트를 도입한 것과 비슷한 흐름이에요. 2023년 약 30%의 iOS 유저만이 데이터 제공에 동의한 에어브릿지의 분석 결과로 미루어 보아, 쿠키 차단 여부를 유저가 선택하게 된다면 활용할 수 있는 쿠키 데이터 또한 줄어들 것으로 예상돼요.
특히 앱이나 웹 이외의 다양한 서비스 환경을 통합하여 일관된 경험을 제공하는 옴니채널 마케팅이 더욱 강조되고 있어요. 그러나 모바일이나 웹이 아닌 PC, 콘솔 인게임에서의 전환 데이터는 기존의 모바일 포스트백으로 광고 채널 서버에 전송할 수 없어요. 모바일 ADID나 쿠키를 활용하기 어려운 환경에서 발생한 전환 데이터를 수집할 수 없다면 캠페인 성과 데이터에 간극이 생길 수밖에 없을 거예요.
이러한 흐름에 발맞춰 주요 광고 채널들은 보다 정확한 성과 측정과 캠페인 최적화를 위한 전환 API를 출시하고 있어요. 전환 API는 무엇이고, 전환 API를 활용하면 어떤 점을 기대해볼 수 있을까요?
## 전환 API가 무엇인가요?
전환 API는 광고 채널이 제공하는 별도의 API를 통해 광고 채널 서버로 데이터를 전송하는 방식이에요. 앱, 웹, 또는 오프라인에서 발생한 사용자 행동과 전환 데이터가 포함돼요. 현재 메타, 스냅챗, 틱톡, X(트위터) 등의 광고 채널에서 캠페인 최적화를 위해 전환 API를 지원하고 있어요.
전환 API는 광고주 서버에서 수집한 구매, 구독 등의 전환 데이터를 직접 광고 채널 서버로 전송해요. 다양한 퍼스트 파티 데이터를 안정적으로 전송할 수 있어 기존의 데이터 전송 방식인 픽셀 또는 포스트백에 더해 전환 API를 활용한다면 캠페인 성과를 보다 정확하게 측정하고 최적화할 수 있어요.
[__wf_reserved_decorative]
## 전환 API를 활용하면 어떤 점이 좋은가요?
픽셀과 포스트백 이외에 전환 API를 추가적으로 활용하면 캠페인을 최적화하고 성과를 극대화하는 데에 다양한 장점을 기대해볼 수 있어요.
첫째로 기존의 트래킹 방식으로는 활용하기 어려운 퍼스트 파티 데이터를 측정과 분석에 활용할 수 있어요. 전환 API는 광고주 서비스에서 직접 수집한 click id, 해시 처리된 이메일, 해시 처리된 전화번호와 같은 데이터를 파라미터로 활용해요. 유저 브라우저를 통해 데이터를 전송하는 픽셀이나 디바이스 ADID를 기반으로 유저 행동 데이터를 전송하는 포스트백보다 더 다양한 파라미터를 활용하여 추가적인 매칭 효과를 기대할 수 있어요.
또한 전환 API는 기존의 트래킹 방식에 비해 보다 안정적인 데이터 전송 환경을 제공해요. 유저 측 브라우저 또는 디바이스 설정이나 네트워크 환경 등에 따라 데이터가 전송되지 않거나 누락될 수 있는 픽셀과 달리, 전환 API는 광고주 서비스의 서버 측에서 데이터를 처리하고 전송하기 때문에 유저 측 브라우저나 네트워크 환경에 영향을 받지 않아요. 따라서 안정적으로 전달된 데이터를 통해 보다 정확한 매칭과 측정이 가능하고, 보다 효과적인 캠페인 운영이 가능해요.
마지막으로 전환 API를 활용하면 앱, 웹 뿐만 아니라 PC, 콘솔, CTV(커넥티드 TV), 그리고 오프라인에서 발생하는 전환 데이터까지 전송할 수 있어요. 예를 들어 오프라인 매장에서 구매가 발생하면 전환 API를 통해 해당 데이터를 광고 플랫폼에 전송해요. 이를 통해 광고 채널에서는 서비스가 운영되는 모든 환경에서의 데이터를 활용하여 캠페인을 최적화할 수 있고, 광고주는 캠페인 성과를 보다 정확하게 측정하여 이후의 운영 계획을 세울 수 있게 되는 것이죠.
## 마케터가 전환 API를 눈여겨봐야 하는 이유
활용할 수 있는 식별자가 줄어들고, 점점 더 다양한 환경에서 유저 행동 데이터를 수집해야 하는 상황에서 전환 API는 새로운 열쇠라고 할 수 있어요. 특히 앱 또는 웹 이외의 환경에서 발생하는 사용자 행동과 전환 데이터가 중요하다면, 전환 API는 선택이 아닌 필수가 될 거예요.
앱, 웹 뿐만 아니라 PC, 콘솔, CTV, 오프라인 등의 다양한 환경을 통합하여 일관된 경험을 제공하는 옴니채널 마케팅이 강조되고 있죠. 성공적인 캠페인을 위해 전환 API 활용이 얼마나 중요해질지 알 수 있어요. 에어브릿지를 통해 PC·콘솔 게임의 모바일 광고 캠페인 성과를 크로스 플랫폼으로 측정한 넥슨의 사례에 따르면 픽셀 이외에 전환 API 연동을 통해 캠페인을 최적화하고, 높은 ROAS를 달성할 수 있었다고 해요.
> [📚 넥슨의 퍼스트 디센던트 마케팅 성공 사례 보러가기](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta1-nexon-the-first-descendant&og_tag_id=102968594&routing_short_id=ohoaqg&sub_id=abr-ko&tracking_template_id=2f52808da1d5154a41346405ee63a954&ad_type=click)
## 에어브릿지와 함께라면 전환 API 관리도 간편하게
점점 더 다양한 광고 채널이 전환 API를 제공하고 있는 만큼 각 채널의 전환 API를 각각 연동해서 관리하기는 쉽지 않을 거예요. 에어브릿지는 현재 메타, 스냅챗, 틱톡, X(트위터)등 주요 광고 채널과 전환 API 연동을 지원하고 있고,, 앞으로 구글과의 연동도 예정되어 있어요.. MMP를 통해 각 광고 채널의 전환 API를 연결하면 하나의 대시보드에서 전환 API를 관리하고, 성과를 비교해 볼 수 있어요.
마케터에게 전환 API는 더 이상 선택이 아닌 필수예요. 지금 [에어브릿지](http://https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta2-book-a-demo&og_tag_id=102968742&routing_short_id=2p5383&sub_id=abr-ko&tracking_template_id=be545554ff161e99b67c2d30bf60a756&ad_type=click)에서 전환 API를 연동하고 광고 성과를 극대화해 보세요.
---
## 에어브릿지 AEM 연동, iOS 메타 광고 성과를 더 정교하게 측정하세요
URL: https://www.airbridge.io/ko/blog/airbridge-x-meta-aem-integration
Language: Korean
Published: 2024-10-11
에어브릿지가 메타의 AEM(Aggregate Event Measurement)과 연동 되었어요. 이번 AEM 연동을 통해 iOS 14.5 이상 기기에서 메타애즈(Meta Ads) 캠페인에 대한 분석이 한층 더 강화 되었습니다. 에어브릿지는 메타와 긴밀한 협업을 통해 마케터 여러분이 iOS 캠페인 성과를 더 면밀히 파악할 수 있도록 AEM을 기반으로 한 데이터 통합 리포트를 제공해요.
## AEM (Aggregated Events Measurement)이란 무엇인가요?
[AEM](https://www.facebook.com/business/help/721422165168355?id=1877298665783613)은 iOS 14.5 이상 장치에서 발생하는 웹/앱 이벤트를 유저 프라이버시를 보호하며 측정해 주는 메타가 개발한 독자적 프로토콜이에요. AEM을 통해 실시간에 가까운 리포트를 제공하여 마케터가 캠페인 성과를 더 빠르게 최적화하고 세밀한 인사이트를 얻을 수 있도록 돕습니다.
# 에어브릿지와 AEM
### AEM을 사용하면 어떤 점이 좋나요?
AEM을 활용하면 iOS 14.5 버전 이상의 기기들에서 메타 광고 캠페인 성과를 더욱 정교하게 측정할 수 있어요. 이를 바탕으로 여러분의 웹/앱 캠페인을 최적화하고 광고 효과를 극대화 할 수 있죠.
실시간에 가까운 리포트 제공AEM 연동을 바탕으로 에어브릿지는 메타 광고 캠페인 성과를 실시간에 가깝게 제공해요. 에어브릿지와 AEM이 제공하는 리포트로 마케터는 iOS 유저들에 대해 더 자세히 알고 광고 캠페인도 최적화 할 수 있죠.
통합 성과 측정 대시보드
에어브릿지 대시보드에서 AEM 스위치를 켜보세요. iOS, 안드로이드는 물론 SKAN, 웹 광고 성과와 함께 AEM의 데이터도 대시보드로 확인할 수 있어요. 통합 대시보드를 통해 다양한 운영체제와 프로토콜을 통해 수집된 광고 성과를 한번에 분석할 수 있는 것이죠. 최대 30일까지 기간을 직접 설정하여 원하는대로 성과를 분석하고 평가하세요.
### AEM 연결하는 방법
에어브릿지와 AEM은 대시보드에 있는 스위치를 켜거나 끄는 방법으로 쉽게 연결하고 연결을 해제할 수 있어요. 자세한 방법은[ 에어브릿지 가이드](https://help.airbridge.io/ko/guides/meta-business-channel-integration#%EC%B7%A8%ED%95%A9%EB%90%9C-%EC%9D%B4%EB%B2%A4%ED%8A%B8-%EC%B8%A1%EC%A0%95-aem-%EC%8B%9C%EC%9E%91%ED%95%98%EA%B8%B0-ios)에서 확인할 수 있습니다. 이외에도 AEM 연동과 성과 측정, 에어브릿지 도입 등 궁금하신 점이 있다면 [에어브릿지에 연락](https://www.airbridge.io/ko/book-a-demo) 주세요.
> 💡 FAQ:
Q: AEM 연동이 SKAN 성과 측정에 영향을 미치나요?
A: AEM과 SKAN은 별개의 측정 방식이에요. AEM 연동이 SKAN에 영향을 주지 않습니다.
Q: 에어브릿지에서 AEM 리포트를 확인하려면, 메타 광고 캠페인 제작 시 AEM을 선택해야 하나요?
A: 맞아요. 메타 광고관리자에서 iOS 앱 광고 캠페인을 만들 때 캠페인 성과 측정 방법으로 AEM를 선택해야 해요. 자세한 내용은 메타 비즈니스 지원 센터에 문의하세요.
Q: AEM 앱 설치 리포트가 에어브릿지 성과측정 방식에 어떻게 반영되나요?
A: 보다 자세한 성과 측정 방법은 [에어브릿지 가이드](https://help.airbridge.io/ko/guides/ad-channel-touchpoint-types)를 통해 확인할 수 있어요.
---
## 이제 에브리타임 광고 성과도 에어브릿지에서 측정할 수 있어요
URL: https://www.airbridge.io/ko/blog/airbridge-x-everytime-mobile-measurement-partner
Language: Korean
Published: 2024-09-27
에어브릿지가 국내 최대 대학생활 플랫폼인 [에브리타임](https://everytime.kr/)과 MMP 최초로 연동을 체결했어요. 이제 에브리타임 광고 성과도 에어브릿지에서 확인할 수 있어요.
마케팅 예산을 쓰고 광고를 집행하는 일만큼 성과를 측정하는 일 또한 중요하죠. 광고 캠페인을 진행할 때 각 매체에서 기대한 효과 대비 실제 성과를 분석하면 최적화를 통해 ROAS와 같은 지표를 개선할 수 있어요. 에어브릿지는 모바일 측정 파트너(MMP)로서 다양한 매체를 통해 운영되는 광고 성과를 한눈에 확인할 수 있도록 돕고 있습니다.
이번 연동을 통해 에어브릿지는 MMP 최초로 에브리타임과 파트너가 되었어요. 특히 에브리타임은 국내 최대 대학생활 플랫폼인만큼 에어브릿지를 통해 20대 · 대학생 타겟 광고 캠페인을 더 정교하게 진행할 수 있게 된 것이죠.
## **에어브릿지에서 에브리타임 광고 성과 확인하는 방법**
에어브릿지에서 에브리타임을 연동하는 방법은 간단해요.
[]
연동된 광고 채널 목록인 \[광고 채널 연동\]에서 ‘에브리타임'을 검색해 주세요. 검색 후 에브리타임을 클릭하면 ‘트래킹 링크'를 생성할 수 있는 화면이 나와요.
이후 트래킹 링크를 생성하여 광고 성과를 측정하면 돼요. 포스트백 설정이 필요한 경우 에브리타임에서 인증키를 받아 입력하여 설정할 수 있습니다.
**💡 포스트백 설정 권장**
> 에어브릿지는 포스트백을 설정한 이후에 트래킹 링크를 사용하는 것을 권장합니다.
> 수집한 데이터를 빠짐없이 광고 채널에 전송하기 위해서는 데이터를 수집하기 전에 포스트백이 설정되어 있어야 합니다. 포스트백을 설정하지 않았다면 트래킹 링크로 수집한 데이터가 광고 채널로 전송되지 않습니다.
에어브릿지는 효과적이고 심층적인 마케팅 성과 분석 서비스를 제공하기 위해 여러 플랫폼과의 연동을 확대해 나가고 있어요. 에브리타임을 포함한 모바일 광고 성과 분석이 필요하시다면 [여기 링크](https://www.airbridge.io/ko/book-a-demo)에서 문의를 남겨주세요.
아래 링크에서 에어브릿지와 연동된 광고 채널 목록을 확인할 수 있습니다. 원하는 광고 채널을 찾지 못했다면 광고 채널에 연동 신청서 작성을 요청해 주세요.
- [에어브릿지와 연동된 광고 채널 확인하기](https://help.airbridge.io/ko/guides/ad-channels-list#%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80%EC%99%80-%EC%97%B0%EB%8F%99%EB%90%9C-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-%EB%AA%A9%EB%A1%9D)
- [광고 채널 연동 신청서](https://help.airbridge.io/ko/supports/integrations)
---
## Apple Search Ads와 SKAN으로 iOS 캠페인 최적화하기
URL: https://www.airbridge.io/ko/blog/asa-and-skan-to-optimize-ios-campaign
Language: Korean
Published: 2024-09-23
많은 마케터가 iOS 마케팅 최적화에 어려움을 겪고 있어요. 애플이 지난 2021년부터 디바이스를 식별할 수 있는 데이터를 제한적으로 제공하는 ATT(App Tracking Transparency) 정책을 시행했기 때문이에요. ATT 정책은 데이터 제공에 동의한 유저에 한해서만 를 앱에서 수집할 수 있는 내용이에요.
iOS 유저는 데이터 제공에 얼마나 동의할까요? 에어브릿지 분석에 따르면 2023년 iOS 유저 10명 중 3명만 데이터 제공에 동의했어요. 마케팅 최적화를 하기에 충분한 데이터라고 보기 어려워요. 이에 많은 마케터가 iOS 마케팅 자체를 포기하거나 최적화를 하지 않아요. [한국 성인의 69%가 안드로이드 기반 스마트폰을 사용](https://www.hani.co.kr/arti/economy/economy_general/1148604.html)하고 있기도 하니까요.
하지만 연령대별로 살펴보면 iOS 마케팅 최적화를 포기하기 어려운 통계를 확인할 수 있어요. 20대 스마트폰 사용자의 64%는 아이폰을 사용해요. 특히 아이폰을 사용하는 20대 여성은 72%에 달합니다. 젊은 유저가 타겟인 패션, 게임 같이 젊은 유저가 필요한 앱은 iOS 마케팅을 포기하거나 최적화하지 않기 어려워요. 그렇다면 iOS 마케팅, 어떻게 최적화할까요?
## iOS 유저 확보를 위한 첫 걸음, 애플 서치 애드와 SKAN
시작은 애플 서치 애드(Apple Search Ads, ASA)와 SKAN(SKAdNetwork)에요. 애플 서치 애드는 애플이 운영하는 광고 채널이에요. 애플 서치 애드 광고는 애플 앱스토어에 게재되기 때문에 iOS 유저 확보에 효과적이에요. 그리고 무엇보다 애플은 애플 서치 애드에 한해서 동의 여부에 관계없이 모든 유저에 대한 광고 성과를 마케터에게 제공해요.
여기에 여러 광고 채널에서 광고를 게재하고 있다면 애플이 마케터를 돕기 위해 제공하는 SKAN을 활용해야 해요. SKAN은 추적하기 어려운 형태로 유저에 대한 광고 데이터를 제공하는 애플의 프레임워크에요. 애플은 SKAN을 통해 모든 유저에 대한 데이터를 제공해요. 광고 채널별로 지원하는 SKAN 버전이 달라요. 아래 가이드에서 더 자세한 내용을 확인할 수 있어요.
- [SKAN 3 이해하기](https://help.airbridge.io/en/guides/skadnetwork-3)
- [SKAN 4 이해하기](https://help.airbridge.io/en/guides/skadnetwork-4)
⚠️주의해주세요! 애플은 SKAN을 통해 앱 설치에 대한 성과만 제공해요.
## iOS 최적화를 위한 필수 요소: 편하고 좋은 MMP
애플 서치 애드와 SKAN을 활용해도 아직 어려움은 남아 있어요. 안드로이드에서도 마케팅을 진행하고 있다면 애플 서치 애드에서 확보한 iOS 데이터를 안드로이드 데이터, 인앱 데이터, 인웹 데이터와 함께 활용할 수 있어야 마케팅 성과를 정확하게 분석할 수 있어요. 여기에 SKAN으로 확보한 데이터 해석이 어려울 수 있어요. 그래서 어려움을 덜어주는 편한 MMP가 반드시 필요해요.
### 1\. 애플 서치 애드 데이터의 통합 분석
애플 서치 애드로부터 받은 디바이스를 식별할 수 있는 데이터는 반쪽짜리에요. 광고로 유입된 유저가 앱에서 어떤 행동했는지 확인하기 위해서는 애플 서치 애드의 데이터와 앱에서 확보한 유저 데이터를 통합해서 분석해야 해요. 데이터를 통합하지 않으면 광고로 유입된 유저가 제품을 구매했는지 등을 분석할 수 없어요.
광고부터 앱까지 이어지는 유저의 여정을 통합적으로 분석하기 위해서는 MMP가 필요해요. MMP는 유저가 같은 유저인지 식별한 후에 분석한 결과를 제공해요. 이런 결과를 바탕으로 전환율, 리텐션, pLTV 같은 지표를 더 확인할 수 있어요.
### 2\. 모바일 마케팅 최적화
iOS 마케팅 최적화는 모바일 마케팅 최적화를 위한 하나의 단계일 뿐이에요. 안드로이드 마케팅 성과와 iOS 마케팅 성과를 확인하기 위해서는 같은 기준으로 데이터를 분석해야 해요. 하지만 애플 서치 애드 대시보드에서 확인할 수 있는 여러 수치는 여러 광고 채널의 효과를 고려하지 않은 숫자에요.
여러 광고 채널을 운영하고 있다면 반드시 MMP를 활용하세요. MMP는 어트리뷰션이라는 과정으로 데이터를 같은 기준으로 분석해요. 어떤 광고 채널이 효과적인지, 어떤 광고 채널의 리텐션이 좋은지 정확하게 분석하기 위해서는 MMP가 필수적이에요.
- [어트리뷰션 이해하기](https://help.airbridge.io/en/guides/attribution-overview)
### 3\. 간단한 SKAN 설정
애플은 SKAN을 통해서 제공하는 데이터를 모두 MMP로 제공해요. 마케터가 SKAN을 통해서 데이터를 확인하기 위해서는 반드시 MMP를 사용해야 해요. 그렇다고 아무 MMP나 사용하면 안돼요.
SKAN 설정을 위해 이해해야 하는 내용은 복잡해요. 마케터가 정확하게 SKAN에 관한 내용을 이해하기 쉽지 않아요. 그래서 SKAN을 간단하게 설정할 수 있는 MMP가 필요해요. SKAN 설정을 마치면 MMP가 매일 알아서 SKAN 데이터를 쉽게 이해할 수 있는 내용으로 해석해요.
## 가장 편한 MMP 에어브릿지와 함께 iOS 최적화를 시작하세요
에어브릿지는 iOS 최적화를 위한 다양한 기능을 제공하고 있어요. 간단한 작업으로 에어브릿지와 애플 서치 애드를 연동하면 에어브릿지가 애플 서치 애드 데이터를 정리하고 분석해서 다양한 리포트로 제공해요. 앱 설치 수부터 리텐션, pLTV까지 에어브릿지는 200개가 넘는 지표를 통해서 마케터에게 필요한 인사이트를 제공하고 있어요.
애플 서치 애드 어드밴스드(Advanced) 솔루션을 활용하고 있다면 비용 연동으로 광고 비용과 함께 캠페인 이름, 광고그룹 이름, 키워드, 광고소재 이름을 에어브릿지에서 확인할 수 있기 때문에 마케팅 성과를 더 세분화해서 분석할 수 있어요. 또 4시간 마다 최신 데이터로 업데이트되기 때문에 애플 서치 애드를 찾지 않아도 광고 성과를 에어브릿지에 확인할 수 있어요.
여기에 SKAN 설정도 직관적이에요. iOS 캠페인을 진행하기 전에 SKAN을 설정하면 에어브릿지에 매일 SKAN이 제공한 데이터를 분석해 다른 마케팅 성과와 함께 제공해요. 광고 채널마다 준비해야 하는 점이 달라요. 아래 가이드에서 더 자세한 내용을 확인해 보세요.
- [애플 서치 애드 연동하기](https://help.airbridge.io/ko/guides/apple-search-ads)
- [SKAN으로 iOS 캠페인 준비하기](https://help.airbridge.io/ko/guides/preparing-for-skan)
iOS 마케팅 최적화에 필요한 MMP를 찾고 있나요? 또는 iOS 마케팅 최적화에 어려움을 겪고 있나요? 아래 마케터가 더 중요한 것에 집중할 수 있도록 에어브릿지가 도울게요.
---
## Firebase 다이나믹 링크, 한국에 최적화된 에어브릿지로 갈아타세요
URL: https://www.airbridge.io/ko/blog/introduce-deeplink-plan
Language: Korean
Published: 2024-09-05
구글 파이어베이스(Firebase) 딥링크 서비스[ 다이나믹 링크(FIrebase Dynamic Link)](https://firebase.google.com/docs/dynamic-links?hl=ko)가 다가올[ 2025년 8월 25일을 끝으로 종료될 예정](https://www.airbridge.io/ko/blog/firebase-dynamic-links-shutdown)이에요. 다이나믹 링크를 활용해 딥링크를 만들고 관리할 수 있는 시간이 1년여 밖에 남지 않은 것이죠.
2025년 8월 25일이 지나면 파이어베이스 다이나믹 링크 기반의 새로운 딥링크 생성이 불가능할 뿐만 아니라, 기존에 만들어 두었던 링크도 모두 작동되지 않을 예정이에요. 새로운 딥링크 서비스 도입과 적응, 데이터 마이그레이션을 통한 기존 링크 교체 등을 고려하면 남은 1년여의 시간이 그리 길지만은 않을텐데요. 다이나믹 링크 서비스 종료가 다가올수록 마케터들의 마음도 분주해지고 있습니다.
파이어베이스 다이나믹 링크를 대체할 수 있는 가장 좋은 딥링킹 서비스는 무엇일까요? 새로운 딥링크 서비스를 찾고 있는 여러분의 고민을 해결해 줄 에어브릿지 [**딥링크 플랜**](https://www.airbridge.io/ko/deeplink-plan?utm_source=blog&utm_medium=CTA01&utm_campaign=deeplinkplan-release)을 소개합니다.
## **다이나믹 링크의 대안, 에어브릿지 딥링크 플랜을 소개해요**
국내 최적화 MMP 솔루션 에어브릿지에서 고도화된 딥링킹 전용 서비스 **딥링크 플랜**을 오픈했어요. 에어브릿지는 국내외 300여 개 이상의 기업이 사용하는 아시아 최대 MMP(모바일 측정 파트너)인데요. 에어브릿지의 기술과 노하우가 담긴 딥링크 플랜은 파이어베이스 다이나믹 링크의 가장 적절한 대체 서비스에요.
이번 글을 통해 에어브릿지 딥링크 플랜이 파이어베이스 다이나믹 링크를 대체할 수 있는 이유를 **네 가지 장점**과 함께 소개해 드릴게요.
### **1. 파이어베이스 보다 더 정교하고 유용한 딥링킹을 제공해요**
에어브릿지 딥링크 플랜은 파이어베이스 다이나믹 링크보다 **더 정교하고 유용한** 딥링킹을 제공해요. 다이나믹 링크를 사용하고 계신 마케터 분들이라면 딥링크 플랜에서도 불편함 없이 딥링크를 만들고 이용할 수 있죠.
#### **🦺 심리스한 딥링킹**
에어브릿지는 높은 사용자 경험을 중요하게 생각해요. 딥링크 오픈 시 유저가 중개 페이지에 오래 머문다면 사용자 경험이 좋다고 말할 수 없겠죠. 에어브릿지 딥링크 플랜은 중개 페이지를 최소화하는 딥링크 사용이 가능하도록 온/오프라인, 웹과 앱, 안드로이드와 iOS 등 다양한 환경과 운영체제는 물론 카카오톡, 네이버, 인스타그램, X(트위터), 유튜브 등 **다양한 지면에서 다이렉트 앱 오픈을 지원**해요.
#### **🌐 iOS, 안드로이드 성과 측정 지원**
iOS 성과 측정이 어려웠던 다이나믹 링크와 달리 딥링크 플랜은 **안드로이드와 iOS 운영체제 모두**를 한번에 연결하는 딥링크 생성과 성과 측정을 지원해요. iOS 성과 측정을 위해 별도로 데이터를 수집하지 않아도 되죠.
#### **🔖 디퍼드 딥링크(Deferred Deep Link, 지연된 딥링크)**
딥링크 플랜에서도 [**디퍼드 딥링크(Deferred Deep Link, 지연된 딥링크)**](https://help.airbridge.io/ko/guides/retargeting-with-deep-links#%EB%94%94%ED%8D%BC%EB%93%9C-%EB%94%A5%EB%A7%81%ED%81%AC-%EC%84%A4%EC%A0%95%ED%95%98%EA%B8%B0) 기능을 사용할 수 있어요. 디퍼드 딥링크는 원활한 사용자 경험을 위해 꼭 필요한 기능 중 하나인데요. 디퍼드 딥링크를 통해 신규 사용자는 앱을 설치한 뒤 원하는 랜딩 페이지로 바로 진입할 수 있어요.
#### **🛒 트래킹 링크 대량 생성 지원**
하나의 캠페인에서도 소재와 지면에 따라 트래킹 링크를 여러 개 생성하고 싶으신가요? 딥링크 플랜은 에어브릿지 기본 기능인 [**트래킹 링크 제너레이터**](https://help.airbridge.io/ko/guides/deep-link-plan-creating-bulk-tracking-links)와 링크 생성 API를 함께 제공하기 때문에 손쉽게 대량 트래킹 링크를 생성할 수 있어요.
#### **💈 딥링크를 통한 브랜딩 가능**
또한 딥링크 플랜에서는 **커스텀 도메인 지원**과 **서비스 이름이 포함된 숏 링크(Short Link)** 제작이 가능해요. 소셜 쉐어 프리뷰 (Social Share Preview)를 활용해서 주소, 제목, 썸네일을 모두 브랜드에 맞게 변경할 수도 있죠. 숏 링크와 같은 기본적인 기능도 에어브릿지와 함께라면 더 디테일하게 활용할 수 있습니다.
### **2. 다른 딥링킹 서비스보다 많은 기능이 있어요**
에어브릿지 딥링크 플랜은 파이어베이스 다이나믹 링크를 제외한 다른 딥링킹 서비스와 비교해도 더욱 풍성한 기능을 제공하고 있어요. 에어브릿지가 가지고 있는 차별화된 장점을 활용하면 딥링크와 함께 마케팅 캠페인도 고도화할 수 있죠.
#### **🧺 트래킹 링크 월 최대 10,000개 생성**
**월 최대 10,000개**까지 생성할 수 있는[** 트래킹 링크**](https://help.airbridge.io/ko/guides/deep-link-plan-tracking-link-management)는 대시보드를 통해 손쉽게 대량 생산이 가능할 뿐만 아니라 월 3,000개까지만 제작 가능한 A사 대비 압도적으로 높은 수를 자랑해요. 사실상 트래킹 링크 생성에 제한이 없는 셈이죠.
****월 1만 개 이상의 트래킹 링크 생성이 필요하다면 ***[***에어브릿지에 문의***](https://www.airbridge.io/ko/book-a-demo-deeplink?utm_source=blog&utm_medium=cta&utm_campaign=deeplinkplan-01)***해 주세요.***
#### **🔗 자유로운 QR코드 제작**
오프라인 마케팅의 필수 요소 중 하나인 QR코드. **딥링크 플랜을 이용하면 QR 코드 제작을 위해 다른 서비스를 사용할 필요가 없어요.** 링크가 생성되면 에어브릿지에서 자동으로 QR코드도 함께 만들어주기 때문이에요. 딥링크 플랜에서는 QR코드 제작이 불가능한 B사, 제한적으로만 QR 코드를 만들 수 있는 C사 등 다른 서비스들보다 편하게 QR코드를 생성하고 관리할 수 있습니다.
#### **📄 다양한 시나리오 대응**
에어브릿지 딥링크 플랜은 딥링크가 여러 플랫폼에서 원활하게 작동할 수 있도록 시나리오를 관리하고 기술적인 지원을 하고 있어요. 국내외 앱 서비스와 인앱 브라우저 대응은 물론 주소 창에 복사/붙여넣기를 통한 진입까지 다양한 시나리오에 대응 가능하죠.
### **3. 누구보다 국내 환경을 제일 잘 알아요**
한국에서 개발된 MMP인 에어브릿지는 이미 **국내 웹/모바일 환경에 최적화되어 있어요.** 카카오톡, 네이버, X(트위터), 유튜브와 같은 국내외 대표적인 웹/모바일 서비스를 비롯해 다양한 국내 웹/앱 환경을 모니터링하고 관리하죠.
에어브릿지를 사용하면 국내 주요 앱 인앱 브라우저에서도 딥링크 오픈이 가능해요. 카카오스타일, 마이리얼트립, KFC와 같은 국내 주요 기업이 사용하는 MMP인 만큼 딥링크 플랜에서도 다양한 국내 환경과 이슈, 수많은 예외 케이스에 대해 능동적으로 대응할 수 있어요.
### **4. 손쉬운 마이그레이션으로 데이터를 유지할 수 있어요**
파이어베이스 다이나믹 링크 서비스 종료는 단순히 서비스 이전만으로 해결되는 문제는 아니에요. 데이터 마이그레이션을 통해 이전에 제작한 딥링크를 교체하고 정상적으로 작동할 수 있도록 유지해야 하죠. 만약 마이그레이션이 늦어진다면 여러분의 마케팅 캠페인 링크가 오픈되지 않을 수 있어요.
에어브릿지는** 다이나믹 링크로 제작한 딥링크들에 대한 쉽고 빠른 마이그레이션을 지원**해요. 에어브릿지가 제공하는 마이그레이션 시트에 파이어베이스에서 사용 중인 링크를 업로드하면 자동으로 딥링크가 생성되어 간단한 복사/붙여넣기로 이전 링크들을 대체할 수 있죠.
## **에어브릿지 딥링크 플랜이 궁금하신가요?**
새로운 딥링킹 서비스를 찾고 계신 마케터 여러분들이라면 에어브릿지와 딥링크 플랜에 대해 이야기 나누어 보세요. 아래 버튼을 누르고 간단히 정보를 입력해 주시면 에어브릿지가 순차적으로 연락드릴 예정이에요!
[]
---
## TikTok 배지 모바일 측정 파트너, Airbridge
URL: https://www.airbridge.io/ko/blog/airbridge-x-tiktok-mobile-measurement-partner
Language: Korean
Published: 2024-05-31
에어브릿지가 글로벌 숏폼 동영상 플랫폼 틱톡(TikTok)의 배지 마케팅 측정 파트너(Badged TikTok Measurement Partner)가 됐어요. 틱톡은 높은 표준을 만족하고 지속적으로 함께 발전한 소수의 파트너에게 배지를 제공해요. 틱톡의 모바일 측정 파트너(MMP) 에어브릿지는 배지를 받으며 틱톡과의 파트너십이 더 강화됐어요.
이번 파트너십 강화로 에어브릿지는 복잡한 유저 여정에 숨어 있는 틱톡의 광고 성과를 더 정확하게 제공하기 위한 노력을 계속해서 이어갑니다. 최근 광고 시장과 기술이 빠르게 발전하면서 많은 유저가 모바일을 포함해 여러 플랫폼에서 광고를 경험하는데, LTA 모델은 광고 성과 바로 직전에 클릭한 광고에 모든 성과를 주기 때문에 일부 광고 채널 성과가 실제 효과보다 낮게 평가됩니다. 틱톡에 따르면 전통적인 라스트 터치 어트리뷰션(Last Touch Attribution, LTA) 모델은 틱톡이 기여한 성과(전환)를 무려 최대 73% 낮게 측정해요.
이런 틱톡의 성과를 더 쉽고 정확하게 측정하기 위해 에어브릿지는 최근 틱톡의 고급 SAN 연동(Advanced SAN Integration) 지원하기 위한 준비를 마쳤고, 틱톡 SKAdNetwork 연동 매핑 자동화, 틱톡의 신규 전환 이벤트인 고관여 조회(Focused View)도 제공하고 있어요. 또한, 틱톡의 ADC(Advanced dedicated campaign) 및 Smart+ 캠페인까지 모두 지원해 더욱 강력한 틱톡 광고 성과 분석이 가능해졌어요.
앞으로도 에어브릿지는 공식 틱톡 마케팅 파트너로서 틱톡 광고 캠페인 성과를 제공하고 마케팅 최적화와 고객 여정 개선에 기여해 갈게요. 에어브릿지에서 틱톡 포 비즈니스(TikTok for Business)와 연동하고 틱톡 광고 캠페인을 최적화하세요.
## 에어브릿지와 틱톡 포 비즈니스 연동하기
에어브릿지와 틱톡 포 비즈니스를 연동해서 틱톡의 광고 성과를 에어브릿지에서 확인할 수 있어요. 클릭스루 어트리뷰션(Click-through Attribution)과 뷰스루 어트리뷰션(View-through Attribution)으로 측정된 틱톡 캠페인의 성과를 에어브릿지의 10가지 이상의 리포트와 200가지 이상의 마케팅 지표로 확인해 보세요.
채널 연동부터 비용 연동, SKAdNetwork 연동을 진행하면 틱톡의 숏폼 콘텐츠의 마케팅 성과를 에어브릿지 대시보드에서 확인할 수 있어요. 자세한 내용은 에어브릿지 가이드에서 확인하세요.
[]
### 틱톡 포 비즈니스
[틱톡 포 비즈니스](https://www.tiktok.com/business/ko)는 브랜드 및 마케터가 창의적인 스토리텔러로 틱톡 커뮤니티와 연결하고 소통할 수 있는 솔루션이에요. 틱톡을 통해 창의적인 영상을 즐길 수 있는 브랜드 환경을 구축하고 일방향 소통이 아닌 참여형 캠페인을 진행할 수 있어요.
### 틱톡 마케팅 파트너 프로그램
틱톡 포 비즈니스는 [틱톡 마케팅 파트너 프로그램](https://partners.tiktok.com/pc/en)을 통해 다양한 마케팅 파트너와 함께 마케터의 창의적인 마케팅 활동과 비즈니스 성장을 지원하고 있어요. 파트너 디렉토리의 측정 부문에서 에어브릿지를 찾아보세요.
> 틱톡과 에어브릿지를 함께 활용하고 있는 성공 사례를 알아보세요.
👉 [말랑이 온라인의 글로벌 시장 진출 전략 알아보기](https://www.airbridge.io/ko/case-studies/fidget-town-tiktok-measurement)
👉 [뮤 모나크의 숏폼 콘텐츠 소재 최적화 전략 알아보기](https://www.airbridge.io/ko/case-studies/mu-monarch-creative-optimization)
---
## 놓치면 안 되는 10가지 앱 마케팅 지표
URL: https://www.airbridge.io/ko/blog/app-marketing-metrics
Language: Korean
Published: 2024-05-10
모바일 광고를 시작한 이후에 결과를 가만히 기다리면 소중한 예산을 낭비할 수 있어요. 초기부터 광고를 최적화하는 작업을 통해 더 효율적으로 광고를 운영해야 해요. 이를 위해 중요한 것은 지표(Metric)이에요. 마케터는 지표를 통해 광고 현황을 파악할 수 있어요.
그렇다면 어떤 지표를 봐야할까요? 광고를 보고 앱을 설치한 유저 수, 유저의 기대 성과 LTV 등 확인할 수 있는 지표는 수백 개가 되지만, 마케터가 모든 지표를 확인하기에는 비효율적이에요. 광고 진행 시기에 따라 어떤 지표를 확인하면 좋은지, 그리고 지표로 확인한 정보를 어떻게 활용하면 좋은지 함께 살펴볼게요.
## 광고를 곧 시작한다면
모바일에서 광고를 시작하기 전에 미리 확인하고 싶은 지표를 준비하는 것이 좋아요. 서비스에 따라 중요한 지표가 다를 수 있지만, 광고 집행 초기에 중요한 지표는 일반적으로 아래와 같아요.
**CTR (Click-Through Rate)**
CTR은 광고가 노출된 횟수 대비 클릭 횟수의 비율이에요. 단순히 클릭률이라고 말하는 경우도 많아요. CTR이 높으면 타겟 유저가 광고를 더 자주 클릭한다는 의미이기 때문에 광고 집행 초기에 긍정적인 신호로 파악할 수 있어요.
**CPC (Cost Per Click)**
CPC는 유저의 광고 클릭에 들어간 비용을 의미해요. 광고 예산의 효율성을 판단하기 위해 자주 확인하는 지표에요. 특히 유저가 많이 유입되는 광고 집행 초기에는 CPC에 따라 광고 채널의 예산을 조절해 최적화해요. 예를 들어 예를 들어 CPC가 높은 캠페인의 소재를 변경하거나 오디언스를 변경하여 테스트하며 CPC를 낮출 수 있는 방법을 찾으면 좋아요.
**품질 점수**
품질 점수는 광고, 키워드, 랜딩 페이지의 품질을 평가해 점수를 매기는 지표에요. 품질 점수가 높다면 타겟 유저에게 효과적이라고 기대되는 광고에요. 다만, 일부 광고 채널에서만 제공하는 지표에요. 또한 품질 점수를 평가하는 방식은 광고 채널마다 다르기 때문에 특정 광고 채널에서 품질 점수가 좋아도, 다른 광고 채널에서는 좋지 않을 수 있어요.
### 에어브릿지와 함께 광고를 준비하세요
에어브릿지의 액츄얼스 리포트는 CTR, CPC를 바로 확인할 수 있는 메트릭을 제공해요. 메트릭에서 CTR, CPC를 선택하면 바로 광고 현황을 파악할 수 있어요. 광고 채널을 포함해 전체 마케팅 비용을 고려하거나 온드 채널의 지표가 궁금하다면 셀프 서브 메트릭 또는 칼큘레이티드 메트릭을 정의해서 사용할 수도 있어요.
특히, 칼큘레이티드 메트릭은 기존 메트릭과 사칙연산을 통해 새롭게 메트릭을 정의하는 기능이기 때문에 광고를 시작하기 전에 준비해두면 편하게 광고 현황을 파악할 수 있어요.
[__wf_reserved_decorative]
> 에어브릿지 대시보드에서 빠르게 성과를 확인하고 성과를 개선한 사례가 있어요.
>
> 에어브릿지 고객사 후야호는 광고 소재 별 CTR과 CPC를 확인해서 최종 지표인 CPI를 낮추고 있어요. 에어브릿지 트렌드 리포트의 차트를 통해 날짜, 시간대별 유입량을 확인해서 캠페인을 on/off하고 예산 편성 변경을 참고하고 있죠. 자세한 내용은 후야호의 성공 사례에서 확인하세요.
>
> 👉 [후야호, 데이터 기반 마케팅 전략으로 출시 일주일만에 인기 차트 1위에 올랐어요](https://www.airbridge.io/ko/case-studies/whoyaho-game-data-marketing-strategy)
## 광고를 하고 있다면
광고를 하면서 다양한 데이터가 쌓이면 더 많은 지표를 볼 수 있어요. 다양한 방식으로 캠페인을 분석할 수 있고, 장기적인 마케팅 방향성을 위해 심층적인 분석이 필요해요.
**CVR (Conversion Rate)**
광고를 클릭한 유저가 구매, 회원 가입 같이 의미있는 행동으로 전환(Conversion)된 비율이에요. 전환율이라고 말하기도 해요. 클릭률과 다르기에 실제로 발생한 전환을 기준으로 하기 때문에 광고의 실제 효과를 확인하기 위해 필요한 지표에요. 어떤 행동을 전환으로 정의하냐에 따라서 성과가 다르게 평가될 수 있어요. 서비스 특징을 고려해서 어떤 유저 행동을 전환으로 정의할지 결정하세요.
**CAC(Customer Aquisition Cost)**
직역하면 유저 1명을 획득하는데 필요한 비용이에요. 클릭 당 비용이라고 하기도 해요. CAC가 너무 높다면 유저 1명에 들어가는 광고 비용이 높다는 의미이고, 효율적인 광고가 아닐 수도 있어요. 하지만 서비스와 광고 채널, 광고 키워드, 시장 상황 등에 따라 적절한 CAC가 다를 수 있어요.
**DAU (Daily Active User) / MAU (Monthly Active User)**
DAU, MAU는 하루 또는 한달 동안 서비스를 찾은 액티브 유저(Active User)에요. 유저가 앱을 다시 방문하는 빈도를 측정하는 지표에요. 광고 성과와 관련있는 지표이기도 하면서, 서비스가 잘 운영되고 있는지 확인할 수 있는 지표에요. 특히, 타겟을 잘 설정했다면 광고로 찾은 유저가 꾸준히 서비스를 이용하면서 DAU, MAU가 광고 이전보다 올라요.
**ARPU (Average Revenue Per User)/ ARPPU (Average Revenue Per Paying User)**
ARPU는 유저 1명이 발생시키는 수익이에요. ARPPU는 유료 유저 1명이 발생시키는 수익이고요. 특정 기간 동안 발생한 수익을 유저 수로 나눠서 확인할 수 있는 지표에요. 또한 광고로 목표하는 매출이 있으면 ARPU, ARPPU를 반드시 확인하면서 광고 전략을 준비해야 해요. 특히 ARPPU는 돈을 쓴 유저가 평균적으로 쓰는 금액을 확인할 수 있는 지표라서 비즈니스 모델이 효과적인지 확인할 수 있어요.
### 에어브릿지에서 바로 확인하세요
에어브릿지는 목적에 따라 특화된 리포트로 다양한 지표를 제공하고 있어요. CAC, ARPU, ARPPU는 레비뉴 리포트에서 바로 확인할 수 있어요. 특히, 특정 행동을 한 유저의 성과만 따로 확인할 수 있기 때문에 단순한 CAC가 아니라 CPA(Cost per Acquisition)나 다름없어요.
DAU, MAU도 액티브 유저 리포트에서 바로 확인할 수 있어요. 에어브릿지는 고유 유저(Unique User)를 기준으로 하기 때문에 더 정확한 성과를 파악할 수 있습니다. 또한 마케터가 전환으로 생각하는 유저의 행동을 결정한 후에 CVR을 칼큘레이티드 메트릭으로 정의하면 액츄얼스 리포트에서 언제나 확인할 수 있어요.
[__wf_reserved_decorative]
## 광고를 마쳤다면
광고를 마쳤거나 마무리되는 단계라면 전체적인 성과를 분석해야 해요. 그래야 다음 광고를 더 효과적으로 준비할 수 있어요. 부족한 것은 개선하고, 잘한 것은 유지해야 하니까요.
**ROAS (Return on Ad Spend)**
광고에 쓴 금액 대비 얻은 수익이에요. ROAS는 광고의 전체적인 성과를 평가하는 대표적인 지표에요. ROAS가 높다면 상대적으로 광고로 얻은 수익이 많다는 의미이기 때문에 성공했다고 볼 수 있어요. 다만, 항상 ROAS를 목표로 하는 것은 아니에요. 유저 획득을 목표로 하는 캠페인은 ROAS 보다 DAU, MAU 또는 리텐션을 더 중요한 지표로 생각할 수 있어요.
**생애 가치(LTV, Lifetime Value)**
유저가 서비스를 이용하는 동안에 발생시킬 것으로 예상되는 수익이에요. LTV는 장기적인 전략과 유저의 가치를 이해하는데 도움을 줘요. LTV는 특별한 지표라서 서비스를 운영하는 회사와 지표를 제공하는 회사마다 계산하는 방법이 달라요. 또한 지금까지 수집된 데이터를 기반으로 예측하는 지표이기 때문에 서비스에 적절하게 활용할 수 있는지 먼저 확인해야 해요.
**잔존율(Retention rate)**
잔존율은 유저가 일정 기간 동안 서비스를 이용하는 비율이에요. 리텐션이라고도 해요. 광고 이후에 리텐션이 높으면 광고로 유입된 유저가 계속해서 시비스를 찾는다는 의미에요. 광고 성과에 따라 잔존율이 중요할 수 있고, 반대로 이탈율이 중요할 수도 있어요. 예를 들어 잔존율이 높다면 다음 광고는 이탈한 유저만을 대상으로 타겟 광고를 고려할 수 있어요. 반대로 이탈율이 높다면 이탈한 지점을 찾아서 서비스를 먼저 개선하는 것이 좋아요.
### 에어브릿지로 분석하세요
에어브릿지가 제공하는 지표로 광고 성과를 다각도로 분석할 수 있어요. 레비뉴 리포트에서 메트릭을 선택하면 ROAS를 선택한 주기별로 확인할 수 있어요. 또한 예측 LTV(Predictive LTV)를 활성화하면 바로 LTV를 ROAS와 함께 조회할 수 있습니다.
잔존율은 리텐션 리포트에서 확인하세요. 코호트를 설정해 특정 유저의 리텐션을 분석할 수도 있고, 특정 행동을 한 유저의 리텐션만 주기별로 조회할 수도 있습니다. 특히, 광고 채널별로 리텐션을 분석하면 어떤 광고 채널로 유입된 유저가 서비스를 오랫동안 이용하는지 확인할 수 있어요.
잔존율이 낮고, 이탈률이 너무 높다면 퍼널 리포트에서 유저가 이탈하는 단계를 파악하세요. 광고 채널부터 인앱 이벤트가 이어지는 유저 여정을 퍼널 리포트에서 분석할 수 있어요. 퍼널 리포트는 각 단계의 전환율과 전환에 걸리는 시간을 제공해요.
[__wf_reserved_decorative]
## 하지만 지표가 항상 정답은 아니에요
마케터가 확인할 수 있는 지표는 다양하지만, 모든 지표가 필요하지는 않아요. 서비스 특징과 단계에 따라 의미있는 지표를 잘 선택해야 해요. 여기에 시장 상황에 따라 광고 성과는 변동될 수 있으니 특정 지표만 항상 신뢰해서는 안 돼요. 지표로 확인한 수치가 어떤 의미가 있는지 전문가와 충분히 검토하고 논의한 후에 결론을 내려야 해요. 지표는 정답이 아니라 현재 상황을 판단하는 기준이에요.
어떤 지표를 확인하는 게 좋은지 고민되시나요? 서비스에 맞는 지표가 궁금하신가요? 또는 지표로 확인한 광고 성과가 어떤 의미인지 파악하는데 어려움을 겪고 있나요? 그렇다면 에어브릿지가 마케터의 고민을 덜어 드릴게요. [에어브릿지의 마케팅 전문가와 이야기를 나눠보세요](https://www.airbridge.io/ko/book-a-demo).
---
## AI 마케팅 마스터하기: 모바일 마케팅 성공의 핵심
URL: https://www.airbridge.io/ko/blog/mastering-ai-marketing
Language: Korean
Published: 2024-04-11
인공지능(Artificial Intelligence, AI) 기술이 급속도로 발전하면서 AI 마케팅의 시대가 도래했어요. AI 마케팅이 단순한 트렌드를 넘어 필수 전략적 도구로 자리매김한 지금, 마케터들은 더욱 치열해지는 경쟁에서 앞서 나가기 위해 AI를 이해하고 마케팅 전략에 적용해야 해요. AI가 모바일 마케팅 전략과 실행 방식에 어떠한 변화를 가져오고 있는지 알아보고, 마케터들이 이러한 변화를 어떻게 받아들이고 활용할 수 있는지에 대해 소개할게요.
## 변화하는 모바일 마케팅 패러다임
AI와 AI 마케팅의 중요성은 갈수록 커지고 있어요. 먼저 정의를 살펴보면, AI는 컴퓨터 시스템이 학습, 추론, 인식, 언어 이해 등 인간의 지능을 모방하는 기술을 의미합니다. 이 기술은 데이터 분석, 패턴 인식, 자연어 처리 등 다양한 방법을 통해 기계가 스스로 학습하고 결정을 내릴 수 있게 해요. 이를 활용한 AI 마케팅은 고객 데이터 분석과 개인 맞춤형 콘텐츠 제작을 통해 고객 참여를 극대화하며, 마케터에게 고객과 더욱 밀접한 마케팅 캠페인과 개인화 전략을 구현할 수 있는 새로운 가능성을 열어주죠.
AI의 도입은 단순히 새로운 경험을 제공하는 것을 넘어서 모바일 마케팅 전략을 변화시키고 있어요. 마케팅 전략을 더욱더 데이터 중심적이고 개인화된 방향으로 진화시켜 기업과 고객 간의 관계가 더욱 깊고 유의미하게 발전할 수 있게 되었어요. 마케터는 모바일 앱을 이용하는 고객 행동 패턴과 선호도를 예측하고 분석하여 마케팅 메시지와 캠페인을 최적화할 수 있고 , 데이터 분석을 통한 의사 결정은 마케팅 전략의 핵심 요소로 자리매김했죠. AI 마케팅의 이해와 활용은 더욱 효과적인 고객 경험을 제공하고 지속 가능한 성장을 도모하는 데 있어 반드시 필요해요.
### AI로 재탄생한 광고 전략
머신러닝을 활용한 광고 전략은 마케터에게 새로운 기회의 문을 열어주고 있어요. 마케터가 어떻게 전략을 적용하고 어떤 이점을 얻을 수 있는지 예시와 함께 살펴볼게요.
- 머신러닝을 활용한 UA 전략 - 최적의 타겟 오디언스 발견하기: AI 기반의 머신러닝 알고리즘은 방대한 양의 데이터를 처리해서 유저의 세분화된 특성과 행동 패턴을 분석합니다. 전환 가능성이 높은 타겟 오디언스를 식별해서 광고 캠페인의 타겟팅을 세밀하게 조정해 주죠. 데이터 기반의 타겟팅은 잠재 고객에게 더 관련성 높은 메시지를 전달하고, 광고의 클릭률과 전환율을 상승시키는 데 도움을 줄 수 있어요. 예를 들어, 국내 커피전문점 체인들은 고객의 구매 이력과 위치 데이터를 분석하여 개인별 맞춤형 마케팅 메시지를 전송하면서 매장 방문율과 구매 전환율을 높이고 있어요.
- 리텐션 전략을 위한 유저 행동 분석 및 예측 -개인화 마케팅 캠페인 실행하기: 앱 내 유저의 행동 데이터를 머신러닝으로 심층 분석하면 유저의 선호도와 반응 패턴을 발견할 수 있어요. 마케터는 특정 행동 패턴을 보이는 유저 그룹을 식별해 이들의 다음 행동을 예측하고, 맞춤 캠페인을 구성할 수 있어요. 타겟팅 된 푸시 알림, 개인 맞춤형 이메일 및 리타겟팅 광고 등 다양한 형태로 구현될 수 있으며, 유저의 참여도를 높이고 고객 충성도를 구축하는 데 중요한 역할을 하죠. 개인화된 리텐션 전략은 유저 경험을 향상시키고 기업 및 브랜드에 대한 긍정적인 인식에 기여해요. 우리가 자주 시청하는 OTT를 예로 들 수 있어요. OTT 플랫폼들은 유저의 시청 이력과 선호도를 분석하여 개인화된 콘텐츠를 추천해요. 유저의 만족도를 높이고 장기적인 구독 유지를 도와주죠.
- AI로 광고 크리에이티브 자동 생성하고 최적화 하기: AI 기술은 광고 크리에이티브의 자동 생성 및 최적화에도 중요한 역할을 해요. 머신러닝 알고리즘이 분석하는 유저 관심사와 반응 데이터로 가장 효과적인 광고 메시지와 디자인을 만들 수 있어요. AI가 생성한 광고 크리에이티브는 유저의 선호도와 행동 패턴에 기반하여 개인화되어 광고의 관련성과 참여도를 높여주죠. 이 과정을 통해 마케터는 시간과 리소스를 대폭 절약할 수 있고 광고 콘텐츠의 다양성과 창의성을 높일 수 있어요. AI가 데이터를 지속적으로 수집하고 분석하면서 광고 크리에이티브가 점점 더 최적화되는 효과까지 기대할 수 있어요. 요즘 마케터들이 가장 많이 활용하는 크리에이티브 도구로는 [챗지피티(ChatGPT)](https://chat.openai.com/auth/login)와 [미드저니(Midjourney)](https://www.midjourney.com/home)가 있어요. 챗지피티로 선호도를 파악하고, 이 정보를 바탕으로 미드저니가 생성한 맞춤형 이미지를 사용해 효과적인 광고 캠페인을 진행할 수 있죠. 콘텐츠의 다양성과 창의성을 더 빠르고 쉽게 실현하는 데에 AI 마케팅이 도움을 주고 있어요.
### AI 도입이 바꾸는 업무 방식
AI 마케팅은 업무 방식에까지 변화를 가져오고 있어요. 데이터 분석, 리드 스코어링, 그리고 소셜 미디어 포스팅과 같이 반복적이고 시간이 많이 소요되는 작업들이 자동화되고 간소화되면서 마케터가 창의적인 전략 구축과 콘텐츠 제작에 더 많은 시간을 할애할 수 있게 됐어요. 이는 업무 효율성을 향상시키고, 더 정밀하고 개인화된 마케팅 캠페인을 실시간으로 조정할 수 있게 해주고 있어요.
이러한 변화에 대비하기 위해 마케터는 다음을 준비하면 좋아요:
- 기술 지식 및 숙련도 향상: AI 기술의 빠른 발전과 도입은 마케터로 하여금 기술에 대한 깊은 이해와 더불어 새로운 도구와 플랫폼을 능숙하게 사용할 수 있는 능력을 필요로 해요. 마케터가 기술 변화에 유연하게 대응하고 AI를 활용하면 마케팅 활동의 효율과 효과를 높일 수 있기 때문이에요.
- 데이터 해석 및 전략적 사고 강화: AI 기술은 방대한 양의 데이터를 수집하고 분석할 수 있는 능력을 제공하지만 이 데이터를 통해 의미 있는 인사이트를 도출하고 전략적 결정을 내리는 것은 여전히 마케터의 몫이에요. 마케터에게 데이터를 해석하고 이를 기반으로 전략을 수립할 수 있는 능력은 AI 시대의 핵심 역량이라고 할 수 있어요.
- 창의적인 전략 수립: AI가 처리할 수 있는 작업 범위가 확대됨에 따라 마케터는 보다 창의적이고 독창적인 접근을 통해 새로운 가치를 만들어내야 해요. 개인화된 고객 경험을 제공하고 차별화된 마케팅 전략 개발을 하는 데에 창의적인 전략은 필수에요.
## AI 마케팅, 에어브릿지가 함께 할게요!
AI 기술은 모바일 마케팅의 전반에 걸쳐 피할 수 없는 혁신을 가져왔어요. 그리고 이에 적응하고 활용하는 능력은 모든 기업이 앞으로 경쟁에서 살아남기 위해 반드시 갖춰야 할 핵심 요소가 됐죠. 데이터를 기반으로 한 의사결정, 개인화된 고객 경험, 그리고 미래를 내다보는 전략 개발은 어느 때보다 중요해졌어요.
에어브릿지도 마케터에게 필요한 AI 마케팅 기능을 지속적으로 제공하기 위해 이미 준비하고 있어요. 디지털 마케팅 환경에서 성장을 이루고 경쟁 우위를 확보하는 데 필요한 파트너로서 에어브릿지가 여러분과 함께 할게요. AI의 무한한 가능성을 활용하여 모바일 마케팅의 새로운 미래를 열 준비를 [에어브릿지와 함께 지금 바로 시작하세요](https://www.airbridge.io/ko/book-a-demo)!
---
## 게임 수익화 모델 완벽 가이드: 나에게 맞는 모델 찾기
URL: https://www.airbridge.io/ko/blog/game-monetization
Language: Korean
Published: 2024-04-09
끊임없이 진화하는 모바일 게임 시장에서의 수익화 전략은 단순한 수익 창출을 넘어 게임의 성공을 좌우하는 결정적 요소로 자리잡았습니다. 유저의 몰입과 만족을 극대화하는 동시에 지속 가능한 수익성을 확보하기 위해서 게임 업계에서는 다양한 방법을 모색하고 있습니다.
이번 블로그 포스트에서는 급변하는 게임 산업에서 주목하고 있는 수익화 모델 트렌드를 확인하고 어떻게 하면 게임의 가치를 극대화하는 전략을 세울 수 있을지 알아봅니다.
## 모바일 게임 수익화란?
게임 수익화는 게임 개발사가 창작한 게임을 통해 수익을 창출하는 일련의 과정을 말합니다. 이 과정을 통해 개발사는 지속적인 게임 업데이트와 지원을 위한 자금 확보를 할 수 있어 게임의 지속 가능성, 품질, 경쟁력을 유지하고 발전시키는데 필수적인 역할을 합니다. 이렇게 중요한 게임 수익화 전략을 구축하기 위해서는 게임의 특성, 타겟 오디언스, 그리고 시장 동향을 파악하여 가장 적합한 모델을 찾아야 합니다.
## 게임 수익화 모델은 어떤 것들이 있을까요?
1. 인앱 구매 (In-App Purchases, IAP)
말 그대로 게임 내에서 이루어지는 구매를 통해 수익을 내는 모델입니다. 일반적으로 무료 게임으로 앱을 배포하고 게임 내에서 추가 구매가 이루어지도록 유도합니다. 게임 진행 중에 유저가 추가 생명, 코인, 무기, 동작, 부스터와 같은 특정 아이템이나 이벤트, 콘텐츠를 유료로 구매함으로서 지속적으로 수익을 창출할 수 있는 방법 중 하나입니다.
1. 인앱 광고 (In-App Advertising, IAA)
게임 내의 광고 지면을 판매하여 수익을 내는 매출 전략입니다. 유저의 게임 경험을 방해하지 않는 선에서 광고를 보여주거나 혹은 배너 광고 형태로 표시될 수 있습니다. 모바일 게임에서 주로 사용되는 광고 형식은 다음과 같습니다.
- 전면 광고: 게임 도중 갑자기 화면 전체에 표시되는 광고 형식. 광고를 본 후 (예:10초) 유저는 광고를 종료하고 게임으로 돌아올 수 있으나 게임 경험에 안좋은 영향을 끼칠 수 있습니다.
- 보상형 광고: 전면 광고와 다르게 유저가 광고 시청 여부를 선택할 수 있습니다. 광고를 시청함으로서 유저는 보상을 받게됩니다.
- 플레이어블 광고: 유저들이 1분 내외의 비디오로 게임을 먼저 경험해본 후 설치를 유도하는 광고 형식입니다. 유저의 리텐션 및 인게이지먼트 유지에 효과적입니다.
- 배너 광고: 게임 플레이 중 화면 상단 혹은 하단에 표시되는 광고 형식입니다.
- 오퍼월: 게임 속 미니 상점과 같은 존재로 유저의 광고 참여 활동에 대한 보상을 지급하도록 지원하는 인앱 광고 형식입니다.
1. 구독 (Subscription)
일종의 인앱 구매로 유저에게 추가 콘텐츠에 대한 액세스 권한을 제공하는 유형입니다. 일정 기간 원하는 상품이나 추가 서비스 혜택을 주는 ‘패스' 구독, 수수료를 지불하고 게임 내에서의 인앱 광고를 제거하는 ‘광고 제거' 구독, 일정 기간 동안 일련의 독점 혜택을 제공하는 ‘VIP’ 구독 등 다양한 형식으로 제공됩니다. 장기적인 유저 참여를 유도할 수 있어 지속적인 수익 창출이 가능합니다.
1. 유료 게임 (Paid)
게임을 구매한 유저만 플레이가 가능하도록 다운로드 시에 과금을 진행하는 방식입니다. 빠른 초기 수익 창출이 가능하지만 유저가 게임을 구매하기 전에 충분한 가치를 느끼도록 만드는 것이 중요합니다.
## 효과적인 게임 수익화 전략 세우기
### 하이브리드 수익화 모델 전략
게임 수익화를 위해 하나의 수익 창출 모델에 의존할 필요는 없습니다. 위에서 설명한 여러 모델을 결합하여 하이브리드 수익 창출 모델을 고려할 수 있습니다. 일반적으로 인앱 광고 + 인앱 구매, 인앱 구매 + 구독, 인앱 구매 + 인앱 광고 + 구독 등의 조합이 있습니다. 하이브리드 수익화 모델 전략을 채택하여 유료로 게임을 사용하는 유저 그룹과 무료 유저 그룹을 동시에 공략할 수 있고 여러 수익 창출 경로를 모색할 수 있습니다.
게임 수익화 모델 구현은 하루 아침에 이루어지지 않습니다. 보통 게임 개발을 완료한 후에 수익화 전략을 세우는 개발사 및 퍼블리셔가 많습니다. 이렇게 되면 수익화 모델과 게임 자체가 완벽하게 통합되지 않아 게임을 플레이하는 유저에게 위화감을 줄 수 있습니다. 이를 방지하기 위해서는 게임 개발 시작 전, 혹은 개발과 동시에 어떠한 수익화 모델을 사용할 지에 대한 전략을 상세하게 세울 필요가 있습니다. 유저의 게임 여정을 방해하지 않으면서도 최대한 자연스럽게 수익화 모델을 구현할 수 있을지에 대한 고민이 필요합니다.
### 게임에 맞는 수익화 모델 비율 찾기
게임 장르에 따라 어떠한 수익화 모델이 우리 게임에 가장 적합할지 다를 수 있습니다. RPG나 시뮬레이션 장르의 게임에서는 유저들이 다양한 스테이지를 거듭하여 계속 게임을 진행하도록 설계되었기 때문에 게임을 지속하게 하기 위한 인앱 구매가 활발하게 이루어집니다. 반면에 하이퍼 캐주얼 게임과 같이 가볍게 무료로 즐기는 게임들에서는 인앱 광고, 특히 보상형 동영상 광고의 수익화 비중이 높을 수 밖에 없습니다.
다만 앞서 말한 것과 같이 장르에 한정되어 하나의 수익화 모델만을 사용할 필요는 없습니다. 게임 장르 및 스토리, 유저들의 플레이 성향 등을 다양하게 고려하여 광고와 인앱 구매의 비율을 적절하게 설정하는 것이 중요합니다.
### 데이터 기반의 테스트는 필수
적절한 비율과 전략을 선택했다면 해당 모델이 기대하는 것과 같이 수익을 창출하고 있는지 확인해야 합니다. 특히 인앱 광고를 수익화 모델에 사용하기로 결정했다면 광고 성과에 대한 핵심 성과 지표 (KPI) 설정과 그에 대한 정확한 측정은 필수입니다. 다양한 시나리오로 A/B 테스트를 수행하여 효과적으로 광고를 배치하고 도출된 데이터를 기반으로 빠른 의사결정을 진행하는 것이 중요합니다.
에어브릿지에서는 유저가 얼마나 자주 서비스로 돌아왔는지를 측정할 수 있는 통계 리포트인 리텐션 리포트를 제공하며, 유저가 언제 다시 우리 서비스로 돌아올지 예측할 수 있는 Predictive Lifetime을 제공합니다. 그리고 매출액, ROAS(Return on Ad Spend), ARPU(Averege Revenue Per User), ARPDAU(Average Revenue Per Daily Active User) 등 유저가 게임 내에서 발생시킨 매출에 관한 다양한 지표를 확인할 수 있는 레비뉴 리포트를 제공하고 있습니다.
레비뉴 리포트에서는 과거 데이터를 기반으로 유저의 생애 가치를 예측하는 [pLTV(Predictive LTV)](https://help.airbridge.io/ko/guides/predictive-ltv) 지표를 제공하여 예측 잔존일 동안 게임 내에서 유저가 매출을 얼마나 발생시킬지에 대해 예상할 수 있습니다. 단순히 광고를 통해 모집한 신규 유저와 해당 유저들이 기간 내 지출한 금액만을 토대로 성과를 측정하는 것이 아니라 해당 유저의 미래 가치를 예측할 수 있어야 비용면에서 효율적인 마케팅을 집행할 수 있습니다.
빠르게 진화하는 게임 산업 특성에 맞춰 수익화 전략 역시 더욱 다양해지고 있습니다. 게임 시장과 유저들의 마음을 사로잡기 위해 보다 창의적이고 효과적인 수익화 전략을 고민하고 있다면 [에어브릿지의 마케팅 전문가와 이야기 나눠보세요](https://www.airbridge.io/ko/book-a-demo)!
---
## 모바일을 넘어 PC·콘솔까지: 에어브릿지 크로스 플랫폼 트래킹
URL: https://www.airbridge.io/ko/blog/mobile-pc-crossplatform
Language: Korean
Published: 2024-04-03
모바일을 넘어 PC와 콘솔로, 다양한 플랫폼을 지원하는 게임이 더 늘고 있습니다. 플랫폼 지원으로 게임의 접근성을 높이고 유저와의 접점을 확대해서 비즈니스의 성장을 이끌어내기 위해서입니다. 여러 플랫폼에서 게임을 즐기는 유저도 계속해서 늘고 있습니다. 이런 흐름에 따라 게임 마케팅의 키워드도 ‘크로스 채널’에서 ‘크로스 플랫폼’으로 발전하고 있습니다.
하지만 ‘크로스 플랫폼’이라는 퀘스트는 난이도가 높습니다. 플랫폼은 각각 높은 벽으로 둘러싸인 성과 비슷해 서로 완전히 구분되어 있습니다. 여러 플랫폼을 자유롭게 돌아다니는 게임 유저를 대상으로 광고 성과를 측정하려면 마케터가 여러 플랫폼을 한번에 볼 수 있어야 하지만, 플랫폼의 벽을 뛰어넘는 일은 쉬운 일이 아닙니다.
에어브릿지는 마케터가 벽을 뛰어넘고 선을 넘을 수 있도록 크로스 플랫폼 측정을 지원합니다. 에어브릿지와 함께라면 모바일, PC, 콘솔 어디에서든 광고 성과를 측정하고 분석할 수 있습니다. 더 많은 유저와 더 다양한 데이터를 하나의 에어브릿지 대시보드에서 확인하세요.
## 에어브릿지의 ‘선 넘는’ 광고 성과 측정
플랫폼이 구분한 경계선을 넘기 위해서는 특별한 스킬이 필요합니다. 에어브릿지는 크로스 플랫폼 측정과 분석을 위해 3가지 핵심 스킬을 준비했습니다.
### 핵심 스킬 1: 빈틈없는 데이터 수집
광고 성과 측정의 기본은 빠르고 정확한 데이터 수집입니다. 하지만 크로스 플랫폼 데이터 수집은 까다롭습니다. 플랫폼별로 데이터 형식도, 데이터 전송 방식도 다르기 때문입니다. 이렇게 변수가 많은 환경에서 이기려면 방법은 단 하나, 바로 자원과 수단을 총동원하는 것입니다. 에어브릿지가 트래킹 링크, SDK, API 등을 모두 활용하여 철저하게 데이터를 수집하는 이유입니다.
어려운 점은 이뿐만이 아닙니다. 게임 유저는 모바일에서 광고를 본 다음 PC에서 게임을 설치하기도, 웹에서 광고를 클릭한 다음 앱에서 게임을 실행하기도 합니다. 이 외에도 유저가 플랫폼을 넘나드는 경우의 수는 무궁무진합니다.
문제는 유저가 자유로워질수록 유저가 게임을 찾아온 과정을 파악하기가 어려워진다는 점입니다. 크로스 플랫폼 마케팅의 성과를 측정하려면 유저가 광고와 상호작용한 데이터, 유저가 게임과 상호작용한 데이터부터 유저를 식별할 수 있는 데이터까지 흩어진 유저 데이터를 모으고 연결해야 합니다. 불가능에 가까운 작업입니다.
그래서 더 에어브릿지 팀은 집요하게 고민했고, 그 결과 더 많은 유저 데이터를 수집하고 더 고도화된 유저 식별 방법을 사용함으로써 더 완전한 유저 여정을 완성하고 있습니다.
### 핵심 스킬 2: 직관적인 분석
에어브릿지가 수집한 데이터는 플랫폼이나 운영 체제와 상관 없이 일관적인 규칙으로 집계되어 실시간으로 제공됩니다. 그리고 복잡한 데이터를 쉽게 이해하는 데 도움이 되는 분석 도구도 다양하게 준비되어 있습니다.
에어브릿지가 제공하는 다양한 리포트 중에서 우선 액츄얼스 리포트로 시작해 보세요. 각종 이벤트의 발생 횟수 등을 채널, 캠페인, 플랫폼 수준에서 세밀하게 살펴볼 수 있습니다. 예를 들어 게임이 모바일과 PC에서 각각 몇 번 설치되었는지, 모바일 유저 중 아이템을 한 번 이상 구매한 유저는 몇 명인지 알 수 있습니다. DAU, WAU, MAU부터 결제 유저, 매출액, ARPU, ARPPU까지 한 곳에서 확인할 수 있는 액티브 유저 리포트도 유용한 리포트 입니다.
마케팅 활동이 게임의 성장에 얼마나 기여하고 있는지 하나의 대시보드에서 시각적으로 직관적인 차트와 테이블로 간편하게 확인하세요. 그리고 지금 당장 활용할 수 있는 인사이트를 도출하세요.
### 핵심 스킬 3: PC방 광고 성과 측정
[한국콘텐츠진흥원의 2023년 조사](https://news.mt.co.kr/mtview.php?no=2023101908045776422)에 따르면 게임 유저의 56.8%가 PC방을 이용한 적이 있고, 월평균 9회 이상 PC방을 이용하는 게임 유저의 비율은 28.0%에 달했습니다. PC방 이용 시간도 꾸준한 증가세를 보이고 있습니다. 국내에서 게임 비즈니스의 성장에서 PC방을 빼고 이야기할 수 없다는 뜻입니다.
하지만 집과 PC방을 자유롭게 오가며 게임을 즐기는 유저를 대상으로 마케팅을 진행하고 성과를 측정하기는 일은 어렵습니다. 모바일, PC, 콘솔이라는 플랫폼뿐만 아니라 물리적 장소까지 고려해서 유저를 식별해야 하기 때문입니다. PC방에서 광고를 본 다음 집에 가서 게임을 설치했거나 버스에서 모바일 광고를 본 다음 PC방에서 게임을 실행했다면 어떻게 유저 여정을 완성할 수 있을까요?
에어브릿지는 실마리를 찾고자 수개월에 걸쳐 국내 유수 게임사와 베타 테스트를 진행했습니다. 그 과정에서 에어브릿지만의 크로스 플랫폼 트래킹을 꾸준히 발전시켰고, PC방에서도 광고 성과가 측정되는 것을 확인했습니다.
## 에어브릿지의 선 넘기는 계속됩니다
에어브릿지는 마케터를 위해 항상 한발 앞서 고민합니다. 그렇기 때문에 에어브릿지의 선 넘기는 모바일, PC, 콘솔에서 멈추지 않습니다. 유저가 있는 곳이라면 어디서든 광고 성과를 측정하고, 유저와의 모든 접점이 기회가 되도록 계속해서 지평을 넓히고 있습니다.
지금 바로 [에어브릿지 가이드](https://help.airbridge.io/ko/guides/cross-platform-measurement)와 [데모 신청](https://www.airbridge.io/ko/book-a-demo)을 통해 에어브릿지와 함께 여러분의 게임 마케팅 경쟁력을 강화할 수 있는 방법을 알아보세요.
---
## 마케팅 성과 다각도로 분석하는 방법, 어트리뷰션 알아보기
URL: https://www.airbridge.io/ko/blog/marketing-measurement-1-attribution
Language: Korean
Published: 2024-03-27
디지털 마케팅이 활발해지고 경쟁이 더 치열해질수록, 데이터를 통해 마케팅 성과를 수치화하여 측정하고 분석하고자 하는 욕구도 커지기 마련입니다. 또한, 자원은 한정되어 있기 때문에 이왕이면 같은 예산이라도 더 큰 성과를 낼 수 있는 방법을 고민해야하죠. 즉, ROI(Return on Investment)를 높여야 합니다.
이를 위해서는, 우선 집행한 광고들의 전환(앱 설치, 결제 등)에 대한 기여도를 분석해야 합니다. 과거의 성과가 높게 혹은 낮게 나왔던 이유를 분석해서 그 요인을 명확하게 파악하고, 이를 반영한 미래의 계획을 세워야 합니다. 현재 마케팅을 집행 중인 채널들에서 지금까지 광고의 성과가 좋았다고 하더라도, 앞으로도 계속해서 같은 성과가 나올 것이라는 보장은 없기 때문입니다.
다시 말해, _ROAS(Return on Ad Spend)를 높이기 위해서는 어트리뷰션(Attribution, 기여도) 분석을 통해 각 채널에서 캠페인의 성과를 측정, 분석한 후에 미래 마케팅 기획에 반영해야 합니다._
사실, 모든 데이터를 완벽하게 보여주면서 지속적으로 인사이트를 제공하는 한 가지의 측정 및 분석 방식은 없습니다. 그렇기 때문에 **마케팅 성과 분석을 제대로 하기 위해서는 다각도의 접근 방식이 필요합니다.**
## 마케팅 성과를 다각도로 분석하기 위한 3가지 모델
[__wf_reserved_decorative]
이번 포스트는 마케팅 성과를 분석하기 위한 **다양한 어트리뷰션 모델을 소개**하고, AB180이 개발한 웹&앱 통합 광고 성과 어트리뷰션 솔루션 **에어브릿지(Airbridge)에서 제공하는 멀티 터치 어트리뷰션 분석 기능**을 통해 마케팅의 성과를 다각도로 분석할 수 있는 방법을 알아보겠습니다.
## 어트리뷰션 모델이란?
유저가 전환 이벤트(앱 설치, 구매 등)를 발생시키기까지는 다양한 광고를 접하게 됩니다. ‘**어트리뷰션(Attribution, 기여도)**’은 유저 여정에서 어떤 터치포인트들(광고 클릭, 조회 등)이 있었는지 확인해서 어떤 터치포인트가 전환에 기여했는지 성과를 분배하는 방법론을 의미합니다.
어트리뷰션은 마케팅 성과 측정의 기본이라고 할 수 있습니다. 여기서, **발생한 성과에 대한 기여도를 어떤 터치포인트에 얼마나 잘 분배하는가**가 어트리뷰션의 핵심입니다. 그리고 이 질문에 대해 답하기 위한 다양한 어트리뷰션 모델이 존재합니다. 그 중에서도 대표적으로 **라스트 터치 어트리뷰션 모델**과 **멀티 터치 어트리뷰션 모델**을 다뤄보겠습니다.
> 👉 광고 성과를 결정하는 규칙인 어트리뷰션 모델에 대한 보다 자세한 내용은 [에어브릿지 핸드북 01 - 어트리뷰션](https://www.airbridge.io/ko/library/airbridge-handbook-attribution)에서 확인해보세요.
## 라스트 터치 어트리뷰션(Last-Touch Attribution)
라스트 터치 어트리뷰션은, **싱글 터치 어트리뷰션(Single-Touch Attribution)\*** 모델 중 하나로, 전환 직전 마지막 단 1개의 마케팅 채널에 모든 성과를 인정해주는 모델입니다. 고객의 유입 경로가 몇 개인지와 상관없이, 마지막 터치포인트가 발생한 매체에 100%의 기여도를 부여합니다.
> \***싱글 터치 어트리뷰션 모델**은 단 1개의 마케팅 채널에 모든 성과를 인정해주는 모델로, 퍼스트 터치 어트리뷰션과 라스트 터치 어트리뷰션 두 가지 모델이 있습니다.
라스트 터치 어트리뷰션 모델은 ‘하나의 터치포인트’라는 불충분한 정보를 토대로 광고 성과를 추론하기 때문에 휴리스틱이라고도 볼 수 있습니다. ([휴리스틱 정의](https://terms.naver.com/entry.naver?docId=2070217&cid=41991&categoryId=41991)) 복잡한 어트리뷰션 분석을 단순화하여 빠르게 진행할 수 있기 때문에, 매체와 광고비 정산 기준을 협의할 때 많은 광고주들이 라스트 터치 어트리뷰션 모델을 사용합니다. 한 유저의 구매 여정에 다양한 광고가 영향을 미친다는 사실을 마케터들도 인지하고 있지만, 각 광고별 효과를 정확히 증명하는 것은 매우 어려운 일이기 때문에 전환 시점과 가장 가까운 터치포인트에 성과를 기여시키는 것으로 타협점을 찾게 된 것입니다.
또한 과거에는 크로스 채널에서의 개별 유저에 대한 분석 역량이 부족했기 때문에, 광고주 입장에서는 어느 유저가 어떤 광고에 노출되었는지 파악하기 어려웠습니다. 라스트 터치 어트리뷰션 모델은 전환을 발생시킨 채널을 확실하게 트래킹할 수 있다는 점에서 이러한 페인 포인트를 해소해 주었고, 현재까지도 주류의 마케팅 성과 분석 모델로 활용되고 있습니다.
### 하지만, 라스트 터치 어트리뷰션 모델은 한계를 가지고 있습니다.
1. 원래 구매의도가 있었던 사람이라도, 페이드 광고에 대한 터치 포인트가 하나라도 있었다면 해당 유저의 전환에 대한 기여도의 100%를 마지막 터치포인트가 가져갑니다.
2. 오직 마지막 터치포인트만 기여를 인정받고, 모든 중간 터치포인트는 무시됩니다.
그렇기 때문에, 유저 전환에 영향을 미친 다른 터치포인트들의 증분(incrementality), 즉 진짜 성과를 정확하게 측정하기 어렵습니다. **이 한계를 극복하기 위해서는, 멀티 터치 어트리뷰션 모델을 적용하여 성과를 분석해야 합니다.**
## 멀티 터치 어트리뷰션 (Multi-Touch Attribution)
싱글 터치 어트리뷰션과는 달리, 멀티 터치 어트리뷰션 모델은 특정 기준을 가지고 2개 이상의 매체에 성과를 배분합니다.
대부분의 전환은 여러 번의 터치 이후에 발생합니다. 온라인 상에서 하나의 상품을 구매하기 전까지 여러 번의 광고에 노출되는 경우를 생각해볼 수 있습니다. 혹은 이미 해당 상품의 브랜드 가치로 인해 광고를 보지 않고도 구매를 결정할 수 있죠. 그렇기 때문에 **진짜 마케팅 성과를 내는 매체를 파악하기 위해서는 멀티 터치 어트리뷰션 모델을 통해 분석할 필요가 있습니다.**
### 멀티 터치 어트리뷰션은 라스트 터치 어트리뷰션 모델의 한계를 보완합니다.
멀티 터치 어트리뷰션은 유저 여정에서 발생한 모든 터치포인트를 고려하기 때문에, **마지막 터치포인트는 물론 이전에 발생한 다른 터치포인트들의 가치를 제대로 이해할 수 있게 됩니다.** 마케터들이 유저들의 전환 여정은 물론 앱 설치 유인을 정확하게 분석할 수 있게 되는 것이죠. 유저가 앱을 다운로드하기 전까지 어떤 광고 채널에 노출되었는지 확인하여 광고 채널들의 효과를 파악할 수 있으며, 이를 라스트 터치 어트리뷰션 결과와 비교해서 **과소평가되거나 과대평가된 채널을 파악할 수 있습니다**.
## Touchpoint Analysis 리포트를 통한 Multi Touchpoint 분석하기
그렇다면, 멀티 터치 어트리뷰션 분석을 위한 터치포인트들은 어떻게하면 한 눈에 확인할 수 있을까요?
에어브릿지에서는 앱에서 발생한 전환과 터치포인트에 대한 분석 리포트인 Touchpoint Analysis 리포트를 제공합니다.
[__wf_reserved_decorative]
에어브릿지에서는 서비스 또는 각 광고 채널별로 어트리뷰션 기간을 커스텀하게 설정할 수 있습니다. 먼저 서비스 또는 채널별로 룩백 윈도우 기간을 설정하고, Touchpoint Analysis 리포트를 통해 유저 전환에 기여한 Single Touchpoint, Multiple Touchpoint 또는 No Touchpoint(Organic) 각각의 수 및 비율을 확인할 수 있습니다. 이를 통해 라스트 터치포인트 어트리뷰션 모델과 멀티 터치 터치포인트 어트리뷰션 모델의 성과를 비교해볼 수 있습니다.
> 에어브릿지의 [기여 모델](https://help.airbridge.io/hc/ko/articles/900001955286) 및 [기여 기간 설정](https://help.airbridge.io/hc/ko/articles/900001057386)에 대한 자세한 설명은 각각 연결된 링크에서 확인하실 수 있습니다.
위의 리포트는 데모 데이터를 바탕으로 구성되었지만, 실제 터치포인트 발생 양상도 이와 유사합니다. 리포트를 상세히 살펴보면, 하나의 터치포인트도 가지지 않은 오가닉 전환의 수는 싱글 터치포인트의 경우와 거의 유사합니다. 5개 혹은 그 이상의 터치포인트를 가진 전환의 수가 그 뒤를 잇습니다.
만약 이와 같은 현상이 지속된다면, 멀티 터치포인트들을 제대로 분석할 필요가 있다는 인사이트를 얻을 수 있습니다. 특히 라스트 터치포인트 어트리뷰션 모델 하에서 성과를 낸 것으로 분석된 각 매체들이 진정한 증분 효과를 가져다준 것인지 생각해 볼 수 있습니다.
## Touchpoint Overlap Analysis 리포트를 통한 광고 채널들의 성과 분석하기
각 광고 채널들이 발생시킨 전환의 진정한 성과를 확인하기 위해서는 각 채널에서 발생한 터치포인트들 간의 중복 비율을 확인해 보는 것도 중요합니다. 에어브릿지는 Touchpoint Overlap Analysis 리포트를 통해 채널 간 중첩되는 터치포인트들의 수와 비율을 제공합니다.
[__wf_reserved_decorative]
더 많은 유저를 획득하기 위해서는 새로운 오디언스들에게 광고를 노출해야 합니다. 하지만 위의 리포트와 같이, 일부 광고 채널들이 다른 광고 채널의 터치포인트들과 굉장히 높은 비율로 중복되고 있음을 알 수 있습니다. 즉, 해당 채널들은 같은 유저들에게 반복적으로 광고를 노출하고 있는 것입니다.
Touchpoint Overlap Analysis 리포트를 통해 비효율적인 성과를 보이는 광고 채널을 파악할 수 있으며, 해당 채널들에 대한 예산 조정을 통해 마케팅 비용을 절약할 수 있습니다.
## Conclusion
이번 포스트에서는 마케팅 성과를 다각도로 측정하기 위한 어트리뷰션 분석 모델 두 가지와, 어트리뷰션 분석 결과를 확인할 수 있는 에어브릿지의 터치포인트 리포트들을 알아보았습니다. 모든 마케터가 바라는 것처럼 마케팅 성과를 높이기 위해서는 **과거 성과에 대한 측정과 분석**이 선행되어야 합니다.
[다음 포스트](https://www.airbridge.io/ko/blog/marketing-mix-modeling)에서는 마케팅 성과를 다각도로 분석하기 위한 세 가지 모델 중 남은 한 가지, 두 어트리뷰션 모델을 보완하고 거시적인 마케팅 성과와 전략 분석을 도와주는 또다른 모델인 ‘**마케팅 믹스 모델링(Marketing Mix Modeling)**’을 알아보겠습니다.
---
## 원하는 타겟을 쉽고 빠르게 추출할 수 있는 에어브릿지 오디언스 매니저를 소개합니다
URL: https://www.airbridge.io/ko/blog/airbridge-audience-manager
Language: Korean
Published: 2024-03-27
Beta 기능으로 제공되고 있던 **에어브릿지 오디언스 매니저(Audience Manager)**가 이제 대시보드에서 정식으로 제공됩니다!
디지털 마케팅에서 캠페인의 효과를 높이기 위해서는 우리 프로덕트에 맞는 적절한 ‘타겟’을 선정하는 것이 중요합니다. 게다가 개인정보보호 트렌드 속에서 서드파티 데이터가 아닌 퍼스트파티 데이터를 활용한 마케팅의 중요성이 보다 높아지고 있습니다. 이때 퍼스트파티 데이터를 통해 우리 프로덕트(앱)에 관심을 갖고 있고, 앞으로 관심을 가질 타겟을 대상으로 한 맞춤 마케팅 캠페인을 진행하는 것은 전환율을 높일 수 있는 한 가지 방법이 될 수 있죠. 즉, 리타겟팅 혹은 리마케팅 캠페인을 진행해 보는 겁니다. 이를 위해서는, 우리 프로덕트(앱)를 설치하거나, 회원 가입하거나, 앱을 오픈하는 등의 특정 이벤트를 수행한 유저 데이터를 기반으로 한 오디언스를 생성해야 합니다.
**오디언스 매니저**는 원하는 조건의 ‘타겟 오디언스(Target Audience)’를 빠르게 추출해 주는 에어브릿지의 주요 기능 중 하나입니다. 에어브릿지 자체 개발 엔진인 **Luft**를 기반으로, 설정한 조건에 따른 Airbridge Device ID 기반의 오디언스를 타사 대비 20배 더 빠르게 생성할 수 있습니다. 게다가 여러 개의 조건을 설정하여 오디언스를 생성할 수 있으며, 한번 생성한 오디언스 조건은 자동으로 저장되어 간편하게 재추출하거나 재정의하는 것도 가능합니다.
[👉Luft 더 자세히 알아보기](https://blog.ab180.co/posts/introducing-luft)
## 에어브릿지 오디언스 매니저란?
에어브릿지에서 오디언스 매니저를 활용하면, **특정 조건을 만족하는 사용자 그룹인 오디언스를 생성해 캠페인 최적화를 할 수 있습니다.** 선택한 기간 내에 이벤트를 발생시킨 혹은 발생시키지 않은 유저, 그리고 특정 프로퍼티를 가진 활성 사용자 등까지 세분화 필터링하여 오디언스를 생성할 수 있습니다. 그리고 생성한 오디언스는 CSV 파일로 다운로드 받아서 Meta(Facebook) 및 Ad Network 등에 업로드하여 활용할 수 있습니다.
### 오디언스 생성 방법
[__wf_reserved_decorative]
오디언스 생성을 하기 위해서는 우선 메인 화면에서 \[+ 오디언스 생성\] 버튼을 클릭합니다.
[__wf_reserved_decorative]
오디언스 생성 화면이 나타나면, 유저가 발생시킨 이벤트(performed Event) 혹은 발생시키지 않은 이벤트(didn’t perform Event)를 선택하여 원하는 조건에 따른 오디언스를 생성할 수 있습니다. 이때, 앱 이벤트만 선택할 수 있으며, 이벤트 프로퍼티(Device Type, Country 등) 조건에 맞는 사용자를 포함(and also) 하거나 제외(and not who) 할 수 있습니다.
선택한 조건에 따라 우측의 \[예상 사용자 수\] 부분에 해당 조건에 따른 오디언스를 샘플링해서 계산한 예산 사용자 수가 표시됩니다. 예상 오디언스를 계산하는 기준은 Airbridge Device ID입니다.
원하는 조건을 모두 설정하였다면, 하단의 \[다음>\] 버튼을 클릭하여 오디언스 생성 옵션으로 넘어갑니다.
[__wf_reserved_decorative]
오디언스 생성 옵션에서 오디언스명을 입력하여 저장하거나, 오디언스 생성이 완료되는 즉시 CSV 파일 다운로드를 요청할 수 있습니다.
오디언스를 저장한 후에도 \[추출\] 버튼을 클릭하여 쉽게 추출을 요청할 수 있으며, 요청한 기록은 \[요청 기록\] 탭에서 확인할 수 있습니다. 또한 추출이 완료되면 가입한 계정의 이메일로도 알림을 받을 수 있습니다.
👉[에어브릿지 오디언스 매니저 이용가이드 바로가기](https://help.airbridge.io/hc/ko/articles/900004850303-Audience-Manager)
## 오디언스 매니저, 이렇게 활용해보세요!
**1\. 앱을 설치만 하고 주문은 하지 않은 유저 대상으로 리마케팅 캠페인 집행하기**
[__wf_reserved_decorative]
위의 예시는 지난 30일간 앱을 설치했지만, 주문은 완료하지 않은 고객을 추출하고 싶을 때 설정할 수 있는 조건입니다.
앱을 설치했다는 것은 즉, 우리 프로덕트에 상당한 관심을 가진 잠재 고객이라는 것을 의미합니다. 앱 마켓이라는 큰 허들을 넘을 만큼 구매 가능성이 높은 잠재 고객을 대상으로, 첫 주문 할인 등의 맞춤 프로모션 등의 캠페인을 진행해보는 것은 어떨까요?
이 예시와 비슷하게 활용할 수 있는 리마케팅 캠페인으로는, 장바구니에만 담아놓고 주문은 하지 않은 고객을 대상으로 한 할인 쿠폰 발행 캠페인 등이 있을 수 있습니다.
**2\. ‘특정’ 키워드로 검색해서 해당 상품을 보거나, 혹은 장바구니에 넣었지만 구매하지 않은 유저 대상으로 리타겟팅 혹은 리마케팅 캠페인 집행하기**
[__wf_reserved_inherit]
위의 예시는, 지난 30일간 특정 키워드로 검색해서 해당 프로덕트를 보거나 혹은 장바구니에 담았지만, 구매까지는 이어지지 않은 유저를 추출하고 싶을 때 설정할 수 있는 화면입니다.
이렇게 생성하여 추출한 오디언스를 바탕으로, 특정 상품에 대한 리타겟팅 광고를 집행하여 유저에게 자연스럽게 해당 상품을 여러 번 혹은 같은 상품군 중 다른 추천 상품을 노출시키는 캠페인을 집행할 수 있습니다. 혹은 이메일 혹은 문자 메시지 광고를 통해 해당 상품에 대한 세일 혹은 추가 정보를 보내는 리마케팅 캠페인을 집행해 볼 수도 있습니다.
## 마무리하며,
이제 대시보드에서 정식으로 제공되는 에어브릿지 오디언스 매니저를 통해 빠른 오디언스 생성과 추출을 경험해 보세요! 이렇게 생성한 오디언스를 활용하여, 보다 효과적인 성과를 내는 마케팅 캠페인을 진행해 볼 수 있습니다.
👉[에어브릿지 오디언스 매니저 이용가이드 바로가기](https://help.airbridge.io/hc/ko/articles/900004850303-Audience-Manager)
---
## 에어브릿지, MMP 최초로 카카오모먼트 비용 데이터를 연동하였습니다
URL: https://www.airbridge.io/ko/blog/airbridge-x-kakaomoment-cost
Language: Korean
Published: 2024-03-27
이제 에어브릿지(Airbridge)와 [카카오모먼트](https://business.kakao.com/info/displayad/) 디스플레이 광고 비용을 연동할 수 있습니다. 에어브릿지 대시보드에서 카카오모먼트의 캠페인 성과를 더 편리하게 한눈에 확인하고, 앱 마케팅 성과를 개선해 보세요!
**현재 카카오모먼트 비용 연동은 MMP 중 에어브릿지에서만 가능합니다.**
## 카카오모먼트 비용 연동 방법
카카오모먼트 비용을 연동을 위해서는 카카오 디벨로퍼스 대시보드와 에어브릿지 대시보드에서 설정하는 과정이 모두 필요합니다. 각각의 대시보드에서 설정을 위한 과정은 다음과 같습니다.
### 카카오 디벨로퍼스 대시보드에서 해야할 일
- 새 애플리케이션 등록: [카카오 디벨로퍼스](https://developers.kakao.com/console/app)에서 현재 광고를 운영 중인 카카오 계정으로 로그인 한 후, 새 애플리케이션을 추가합니다.
- 비즈앱 정보 등록: 애플리케이션 추가에, 앱 아이콘 이미지를 반드시 넣고, 비즈앱 정보(사업자 정보)를 등록합니다.
- 플랫폼 등록: 왼쪽 메뉴바에서 \[플랫폼\]을 선택한 후, \[Web 플랫폼 등록\]을 클릭하여 내용을 입력합니다.
- Redirect URI 등록: 에어브릿지 대시보드에 있는 Redirect URI를 복사하여, \[카카오 로그인\] > Redirect URI 등록 클릭 후 나타나는 팝업창에 붙여 넣습니다.
**이때 카카오모먼트 담당자에게 카카오 OpenAPI 사용 권한을 신청하고, 승인을 받는 과정이 반드시 필요합니다.** API 사용 권한 신청은 마케터가 카카오모먼트 담당자에 직접 신청해야합니다.
### 에어브릿지 대시보드에서 해야할 일
- REST API 키 등록: 카카오 디벨로퍼스에 있는 REST API 키를 복사하여, 에어브릿지 대시보드에서 비용 연동\] 탭 내의 REST API를 입력창에 붙여 넣습니다.
- 카카오 DA 비용 연동 토글을 켜고, 카카오 로그인 버튼을 클릭합니다.
- 비용을 연동할 광고 계정과 연결되어 있는 카카오 계정으로 로그인합니다. 로그인 후 에어브릿지에서 비용을 확인할 광고 계정을 선택합니다.
[__wf_reserved_decorative]
연동 완료 직후에는 \[Waiting\] 상태로 표시되며, 데이터 연동 스케줄에 맞춰 데이터가 동기화되면 \[Active\] 상태로 표시됩니다.
이때, 비용 데이터는 연동일 포함 7일 전 비용 데이터부터 연동되며, 4시간에 한 번씩 가져와서 업데이트 됩니다. 따라서 비용 데이터 조회 시점에 따라서 매체 측 데이터와 약간의 오차가 발생할 수 있습니다.
Actuals 리포트에서는 데이터를 가져오는 작업이 완료된 후 30분 이내로 데이터를 확인할 수 있습니다. 데이터 인덱싱 작업은 하루 한 번 이뤄지므로 기존 캠페인 파라미터와 연결은 1~2일 소요됩니다.
### 에어브릿지에서 데이터 조회하기
매체 연동을 통해 수집한 **노출(Impression), 클릭(Clicks), 지출 금액 (캠페인, 광고 그룹, 소재) 데이터**는, 에어브릿지 **Actuals 리포트**에서 아래의 메트릭을 추가하여 확인할 수 있습니다.
- Touchpoint: Impressions (Channel), Clicks (Channel)
- Cost: Cost (Channel)
[__wf_reserved_decorative]
아래 가이드 문서 링크를 통해 더 자세한 에어브릿지와 카카오모먼트의 비용 연동 방법을 확인해 보세요!
[__wf_reserved_decorative]
이제 에어브릿지 대시보드에서 카카오모먼트 디스플레이 광고의 캠페인 성과를 더 편리하게 한눈에 확인하고, 앱 마케팅 성과를 개선해 보세요!
## 카카오모먼트
카카오의 광고 플랫폼은 카카오모먼트, 카카오 키워드광고, 카카오 브랜드 검색광고로 구성되어 있습니다. 그 중 에어브릿지와 비용이 연동된 [카카오모먼트](https://business.kakao.com/info/displayad/) 플랫폼에서는 카카오비즈보드, 디스플레이 광고, 동영상 광고, 메시지 광고 등을 누구나 직접 집행할 수 있습니다.
- **카카오비즈보드**: 카카오톡의 채팅목록탭 최상단 영역 및 카카오의 주요 핵심 서비스, 주요 파트너 서비스를 중심으로 가장 주목도 높은 광고 영역에 광고가 노출됩니다. 프리미엄 확장 형태로 확장 요소를 등록하여 리치한 정보를 제공할 수 있습니다. 혹은 다양한 형태의 랜딩페이지로 연결시킬 수도 있습니다.
- **디스플레이 광고**: 카카오톡과 같은 카카오의 핵심 서비스는 물론 주요 파트너 서비스를 중심으로 다양한 지면에서 광고할 수 있습니다. 노출 가능한 지면은 카카오톡, 카카오스토리, 다음, 카카오페이지 및 프리미엄 네트워크 서비스 지면이며, 모바일부터 PC까지 모두 게재 가능합니다. 이때, 네이티브, 카탈로그 2가지 광고 유형을 활용할 수 있어서, 두 가지 유형의 크리에이티브만으로도 광범위한 카카오 광고 지면 곳곳에 광고를 노출할 수 있습니다. 이외에도, 다양한 광고 목표별 최적화 지원과 정교하고 다양한 오디언스 설정이 가능하기 때문에, 원하는 목표와 타겟에 따른 광고를 집행할 수 있다는 장점이 있습니다.
- **동영상 광고**: 카카오 서비스에서 제공하는 다양한 지면에 인스트림/아웃스트림 동영상 광고를 노출합니다. 카카오의 전국민 대상 유저풀을 기반으로 최적의 오디언스에게 동영상 광고를 노출할 수 있습니다. 동영상 조회 당 비용이 과금되는 방식(CPV)으로 효율적으로 광고를 노출할 수 있습니다.
- **스폰서드 보드**: 카카오 뷰에 내 브랜드 또는 채널 콘텐츠를 담은 광고 보드를 노출합니다.정교하고 다양한 오디언스를 설정하여 적합한 이용자에게만 나의 보드를 소개하고, 채널 친구 추가까지 유도할 수 있습니다. 노출당 비용(CPM)으로 과금되며, 입찰 방식이므로 효율적인 광고 운영이 가능합니다.
---
## 매체별 연동 상태를 한눈에 보여주는 Integration Overview를 소개합니다
URL: https://www.airbridge.io/ko/blog/airbridge-integration-overview
Language: Korean
Published: 2024-03-27
연동되어 있는 광고 채널과 포스트백 전송 채널 현황을 직관적으로 파악할 수 있는 **Integration Overview** 기능이 정식 배포되었습니다! 각 광고 채널 데이터가 에어브릿지와 잘 연결되어 있는지 실시간으로 확인할 수 있습니다.
[__wf_reserved_decorative]
디지털 마케팅을 진행하고 있는 브랜드라면, 유저를 유입시키기 위해 페이드 미디어를 통해 광고를 집행하게 됩니다. 이 때 보다 넓은 범위의 오디언스에게 다가가기 위해 다수의 광고 채널에 브랜드와 상품을 노출시킵니다. 마케터들이 브랜드의 제품과 서비스를 알리기 위해 다양한 광고 채널을 운영할수록, 어떤 채널을 통해 유저가 유입되었는지 분석하는 어트리뷰션의 중요성은 커집니다. 에어브릿지와 같은 MMP를 활용하고 계시다면, 유의미한 어트리뷰션 분석을 위해 MMP에 **광고 채널별 데이터가 정확하게 잘 들어오고 있는지, 그리고 포스트백은 잘 전송되고 있는지**제대로 관리하는 것이 중요합니다.
Integration Overview에서는, 페이스북(메타) 비즈니스, 구글 애즈, 애플 서치 애드 등의 **SAN(Self-Attributing Network)**과 에어브릿지의 연동 상태를 실시간으로 제공하여 고객사가 빠르게 이슈를 발견하고 신속한 대응이 가능하도록 지원합니다. 또한 에어브릿지에서 포스트백을 전송하고 있는 채널들의 목록을 확인하실 수 있습니다.
## 에어브릿지 Integration Overview에서 얻을 수 있는 정보
### Source (매체 연동)
좌측의 Source에서, API 연동을 통해 터치포인트 데이터를 수집하는 SAN들과의 연동 상태를 확인할 수 있습니다. 현재 메타(페이스북), 구글 애즈, 애플 서치 애드 채널을 지원하며, 추후 더 많은 광고 채널들이 추가될 예정입니다.
[__wf_reserved_decorative]
채널별 연동 상태는 아래의 4가지 경우로 구분해 보여줍니다.
- Not Configured: 연동 설정 미완료 상태
- Waiting: 연동 설정 완료 후 데이터 수집 대기 중
- Success: 데이터 정상 수집 중
- Error: 데이터를 정상적으로 수집하지 못함
항목별로 좌측 토글을 열면 연동의 세부적인 단계도 확인할 수 있기 때문에, Error 상태로 표시되는 항목에 대해 이슈가 발생한 지점을 빠르게 파악할 수 있습니다. 해당 광고 채널의 자세한 설정 내용을 확인 혹은 변경하고 싶다면, ‘상세 보기’ 버튼을 클릭하면 해당 광고 채널의 대시보드 설정 메뉴로 바로 접근할 수 있습니다.
[__wf_reserved_decorative]
Integration Overview에서는 터치포인트 데이터 연동 상황 뿐 아니라, 채널별 광고 집행 비용 데이터 연동 상태, 혹은 채널에 따라 SKAdNetwork 데이터 연동 정보도 제공합니다. 각 채널별 구체적인 연동 항목 및 방법에 대해서는 아래 링크를 참고해주세요.
👉 [광고 채널별 제공 정보 바로가기](https://help.airbridge.io/hc/ko/articles/5032424798489#%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-%EB%B3%84-%EC%A0%9C%EA%B3%B5-%EC%A0%95%EB%B3%B4-%EC%9A%94%EC%95%BD)
### Postback (포스트백)
우측의 Postback에서 에어브릿지가 포스트백을 전송하는 채널 목록을 제공합니다. 포스트백을 전송하는 채널의 총 개수도 함께 확인할 수 있습니다. Integration Overview에서는 포스트백 ‘전송 시작’이 적용된 채널만 노출되며, 에어브릿지와 연동 가능한 모든 광고 채널 목록은 Integrated Ad Channels 메뉴에서 별도로 확인할 수 있습니다.
[__wf_reserved_decorative]
에어브릿지는 앱 마케팅 데이터의 허브이자 효과적인 광고 성과 분석 솔루션으로서 고객사의 편의와 효율을 위해 지속적으로 다양한 기능을 개선하고 있습니다. 새로 배포된 Integration Overview 기능을 통해 광고 채널별 데이터 연동 상황을 보다 직관적으로 파악해보세요! 더 자세한 기능 안내는 아래 이용 가이드에서 확인하실 수 있습니다.
👉[에어브릿지 Integration Overview 이용 가이드 보러가기](https://help.airbridge.io/hc/ko/articles/5032424798489)
---
## 릴리즈 노트: 2022년 3~4월
URL: https://www.airbridge.io/ko/blog/release-notes-mar-apr-2022
Language: Korean
Published: 2024-03-27
2022년 3월과 4월, 에어브릿지 대시보드를 한층 업그레이드하였습니다. 연동되어 있는 광고 채널과 포스트백 전송 채널 현황을 직관적으로 파악할 수 있는 Integration Overview를 정식으로 공개했습니다. 더 다양하게 맞춤 분석할 수 있도록 리포트에 새로운 필드와 메트릭 및 그룹바이를 제공하기 시작했으며, 이를 설정하는 Config box의 사용성도 개선했습니다. 이외에도 추가적인 매체 연동 및 비용 데이터 연동으로 더 많은 데이터를 에어브릿지에서 확인할 수 있게 되었습니다. 이 모든 에어브릿지의 최신 프로덕트 업데이트 소식을 한 번에 확인해보세요.
## 🖇 매체별 연동 상태를 한눈에 보여주는 Integration Overview
[]
연동되어 있는 광고 채널과 포스트백 전송 채널 현황을 직관적으로 파악할 수 있는 Integration Overview 기능이 정식 배포되었습니다! 이제 각 광고 채널 데이터가 에어브릿지와 잘 연결되어 있는지 실시간으로 확인할 수 있습니다.
디지털 마케팅을 진행한다면 다양한 광고 채널에서 Paid 광고를 집행하곤 합니다. 그렇기 때문에 에어브릿지와 같은 MMP를 통해 어떤 채널을 통해 유저가 유입되었는지 분석하는 어트리뷰션도 필요하죠. 에어브릿지에서는 광고 채널별 데이터가 잘 들어오고 있는지, 그리고 포스트백은 잘 전송되고 있는지 한 눈에 확인할 수 있도록 Integration Overview 화면을 제공합니다.
위의 화면 예시와 같이 Integration Overview 좌측에는, 페이스북(메타) 비즈니스, 구글 애즈, 애플 서치 애드, Naver Search AD 및 카카오모먼트 등의 매체와 에어브릿지의 연동 상태를 실시간으로 제공하여 고객사가 빠르게 이슈를 발견하고 신속한 대응이 가능하도록 지원합니다. 또한 우측에는 에어브릿지에서 포스트백을 전송하고 있는 채널들의 목록을 확인할 수 있습니다.
이번 Integration Overview 기능을 통해 광고 채널별 데이터 연동 상황을 보다 직관적으로 확인해보세요!
## 📊 에어브릿지 리포트에서 보다 많은 데이터를 편리하게 확인해보세요.
### Revenue 리포트에 ROAS 메트릭이 추가되었습니다.
[]
Paid 마케팅을 집행한다면 광고로 인해 발생한 Revenue를 반드시 확인해야 합니다. 특정 채널에서 캠페인을 집행한 후 사용 비용 대비 성과를 확인하지 않으면 눈을 감고 돈을 쓰는 것과 같죠. 그렇기 때문에 에어브릿지에서는 Revenue 리포트를 통해 ‘분석 기간 내에 Start Event를 수행한 후 Return Event를 수행한 유저 코호트’가 ‘N 번째 날’에 발생시킨 ‘Revenue’와 ‘ARPU(Average Revenue Per User)’ 데이터를 제공하고 있으며, 이와 더불어 이제는 **‘ROAS(Return on Ad Spend)’**도 확인할 수 있게 되었습니다.
여기서 ROAS는, 사용한 비용에 대한 매출액 비중을 의미하며, *Revenue / Cost *100 (%)*으로 계산합니다. 이외에도 Revenue 관련 추가 메트릭들이 추후 더 지원될 예정입니다.
### 새로운 필드로 ‘공헌이익 매출액’이 추가되었습니다.
[]
[]
유저의 구매액 중 실제 회사의 수익으로 이어지는 금액은 얼마나 될까요? 이제 Actuals와 Trend 리포트에서 ‘공헌이익 매출액’을 확인할 수 있습니다. 공헌이익 매출액은 구매액에서 변동비를 제외한 금액으로, 회사에서 수익성을 판단하기 위한 지표로 사용할 수 있습니다. 특히 실제 유저들의 ‘구매’하는 서비스인 커머스에서 활용하기 좋은 지표라고 할 수 있습니다.
이때 변동비에 대해 계산하는 방식이 회사마다 다르기 때문에, 고객사에서 공헌이익을 계산해서 직접 입력한 값을 수집하여 반영합니다.
👉 공헌이익 매출액 계산을 원하신다면, 담당 CSM에게 문의해주세요.
### Config. Box 더 편리하게 사용하기
에어브릿지 리포트에서 확인할 수 있는 데이터가 추가된 것 외에도 리포트 화면에서 메트릭, 그룹바이, 필터를 선택할 수 있는 Config. Box를 더 편리하게 사용할 수 있는 기능이 업데이트 되었습니다. 아래 영상과 같이 드래그하여 이미 선택한 그룹바이의 Property를 필터에서 바로 사용하거나, 필터에서 사용한 Property를 그룹바이에 바로 추가할 수 있습니다.
[]
## 🔗 추가 매체 연동
에어브릿지는 지속적으로 파트너사를 확대해가고 있으며, 추가적인 매체 연동을 지속하고 있습니다. 에어브릿지에 사전에 연동된 채널은 Integrated Channels로 분류되며, 사전에 정의된 구분값을 가지고 있습니다. [Integrations - Integrated Ad channels] 탭에서 각 매체와 연동할 수 있으며, 트래킹 링크를 생성할 때도 Integrated Channels는 Custom Channels와 구분하여 생성하게 됩니다.
- 3월과 4월에는, 패들웨이버(PaddleWaver),비드이즈(bidease), 애드부스트(Adboost)가 추가로 연동되었습니다.
에어브릿지와 연동된 전체 Integrated Channels 리스트는 [여기 링크](http://abit.ly/integrated-channels)에서 확인해보세요!
## 💰 추가 비용 데이터 연동
에어브릿지는 MMP 최초로 카카오모먼트 비용 데이터를 연동한 데에 이어, 추가적인 비용 데이터 연동 작업을 이어가고 있습니다.
- 국내에서 마케팅을 한다면 반드시 필요한 **Naver Search Ad(네이버 검색광고)**의 비용 연동을 진행하였습니다. 현재는 수동으로 연동을 지원합니다.
- 앱 수익화 및 앱 마케팅을 위한 플랫폼 [ironSource](https://www.is.com/ko/)의 User Acquisition에 대한 비용 연동을 지원합니다.
- 에어브릿지는 글로벌 숏폼 모바일 비디오 플랫폼 틱톡(TikTok)의 Marketing Partner Program 공식 마케팅 파트너입니다. 이와 더불어 이제 비용 연동 설정 기능을 대시보드에서 제공합니다. 자세한 비용 연동 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/hc/ko/articles/900003991066-TikTok-For-Business-%EA%B4%91%EA%B3%A0)에서 확인해보세요.
---
## 카카오 키워드 광고 연동으로, 에어브릿지가 MMP 최초로 국내의 모든 SA 비용 데이터를 제공합니다
URL: https://www.airbridge.io/ko/blog/airbridge-x-kakaokeyword-cost
Language: Korean
Published: 2024-03-27
국내에서 마케팅 캠페인을 집행한다면 빼놓을 수 없는 SA(Search Ads, 검색 광고) 매체인 구글 애즈, 네이버 검색 광고, 카카오 키워드 광고. 이번 카카오 키워드 광고와의 비용 데이터 연동을 통해, 이제 에어브릿지에서 이 모든 검색 광고 성과를 한 번에 확인할 수 있습니다.
**에어브릿지가 MMP 최초로 한국에서 운영하는 모든 검색 광고 매체의 비용 데이터를 연동하였습니다.**
> 👉[구글 애즈 비용 데이터 연동 가이드 확인하러 가기](https://help.airbridge.io/hc/ko/articles/900001039466)
> 👉[네이버 검색 광고 연동 가이드 확인하러 가기](https://help.airbridge.io/hc/ko/articles/900001057943)
## 카카오 키워드 광고 전환 성과 측정하기
카카오 키워드 광고는 ‘추적 URL’ 기능을 통해 개별 광고 클릭에 대한 정보를 파라미터로 제공하고 있습니다. 에어브릿지는 이 기능과 에어브릿지 Web SDK를 활용하여 전환 성과를 측정하여 제공합니다. 그렇기 때문에, 키워드 광고의 랜딩 페이지에 Airbridge Web SDK가 반드시 설치되어 있어야 합니다.
### 미리 설정해 두어야 할 사항
1\. 카카오 키워드 광고의 '추적 URL 기능' 설정
[카카오 키워드 광고 대시보드](https://keywordad.kakao.com/)에서 캠페인의 '추적 URL' 기능을 ‘설정’으로 선택합니다. 추적 URL을 설정하게 되면 키워드 광고 클릭 후 연결되는 최종 랜딩 페이지의 URL에 카카오 키워드 광고에서 제공하는 모든 추적 파라미터가 자동으로 추가됩니다.
2\. Airbridge Web SDK 설치 및 urlQueryMapping 기능 설정
카카오 키워드 광고로 연결되는 최종 랜딩 페이지에 Airbridge Web SDK를 설치합니다. 그리고, Airbridge Web SDK의 urlQueryMapping 기능을 활용하기 위해 추적 파라미터 값을 에어브릿지 캠페인 파라미터와 매핑하는 코드를 추가합니다.
3\. Airbridge Web SDK의 setDownloads 또는 setDeeplinks 설정
웹사이트로 유입된 유저들의 웹투앱 전환 성과를 측정하기 위해서는 Airbridge Web SDK의 setDownloads 또는 setDeeplinks 설정이 필요합니다. 이는 웹사이트 내에서 앱 설치 혹은 딥링크를 유도하는 버튼과 Airbridge Web SDK를 연결하는 작업으로, 버튼 클릭 시 리다이렉트 및 성과 추적을 위한 정보를 Airbridge Web SDK에서 처리하게 됩니다.
다만, 카카오 키워드 광고의 ‘추적 URL’ 기능의 유의 사항으로 인해 활용이 어려운 경우, 추적 URL 기능을 미설정하고, Airbridge Web SDK에서 제공하는 ‘utmParsing’ 기능을 사용하여 직접 키워드 레벨에서 추가한 UTM 파라미터 값을 기준으로 성과 측정도 가능합니다.
## 에어브릿지에서 비용 데이터 확인하기
에어브릿지 대시보드에서 카카오 검색 광고 계정 로그인 한 번으로, 카카오 키워드 광고의 비용 데이터를 연동하여 전환 성과와 함께 조회할 수 있습니다.
캠페인을 운영 중인 계정으로 로그인하여 비용을 연동하면, 아래와 같이 에어브릿지 대시보드에서 _노출수(Impressions (Channel)), 클릭수(Clicks (Channel)), 광고비용(Cost (Channel)) 데이터_를 확인할 수 있습니다.
[__wf_reserved_decorative]
### 에어브릿지에서 비용 연동 설정하기
[__wf_reserved_decorative]
1. \[ Integrated Ad Channels > 카카오 키워드 광고 > 비용 연동 \] 메뉴에서 카카오 키워드 광고 비용 연동 토글을 켭니다.
2. 현재 카카오 키워드 광고 캠페인을 운영하고 있는 카카오 계정으로 로그인하고 접근 권한을 허용합니다.
3. 카카오 로그인을 성공적으로 마치면 에어브릿지 대시보드에 돌아와 연동할 카카오 키워드 광고 계정을 선택하여 연동을 완료합니다.
4. 연동이 완료되면 하단의 이미지와 같이 연동된 광고 계정이 리스트에 추가되며, 연동 정보를 확인할 수 있습니다.
[__wf_reserved_decorative]
위 이미지와 같이 '카카오 키워드 광고 비용 연동' 토글이 ON인 상태에서 광고 계정 등록이 완료되면 비용 데이터 수집이 시작됩니다. 연동 직후에는 ‘Waiting’상태로 표시되며, 이후 데이터 연동 스케줄에 맞춰 데이터가 동기화되면 ’Active’ 상태로 표시됩니다.
비용 데이터의 업데이트는 매일 4시간 간격으로(0시, 4시, 8시, 12시, 16시, 20시) 진행되며, 오늘 포함 최근 7일치의 비용 데이터가 업데이트됩니다.
아래 가이드 문서 링크를 통해 더 자세한 에어브릿지와 카카오 키워드 광고의 전환 성과 측정 및 비용 연동 방법을 확인해보세요.
[__wf_reserved_decorative]
## 카카오 키워드 광고란?
카카오 광고 플랫폼 중 하나로, [카카오 키워드 광고](https://business.kakao.com/info/keywordad)는, 유저가 나의 비즈니스와 관련된 키워드를 검색하는 순간, PC부터 모바일까지 시의적절하게 광고를 보여줍니다. 카카오의 수많은 검색 유저를 만날 수 있는 광고입니다. 카카오 계정 기반으로 편리하게 광고를 운영할 수 있으며, 클릭당 과금 방식입니다. 또한, 주요 포털의 통합검색 영역 최상단에 동시 노출하고, 다양한 형태로 추가 정보를 전달하는 확장 소재를 적용할 수 있습니다.
## 에어브릿지에서는 국내에서 운영하는 모든 검색 광고 매체의 비용 데이터를 한 번에 확인할 수 있습니다.
에어브릿지는 국내에서 운영하는 모든 SA 매체의 비용 데이터를 한 번에 확인할 수 있는 유일한 어트리뷰션 솔루션입니다. 이제 각 SA 매체 별 대시보드에 들어가서, 각각의 기준이 적용된 전환 성과 및 비용을 확인할 필요가 없습니다. 에어브릿지를 통해 한국에서 운영하는 모든 검색 광고 매체 비용 데이터를 한 번에 확인해보세요!
> 👉[구글 애즈 비용 데이터 연동 가이드 확인하러 가기](https://help.airbridge.io/hc/ko/articles/900001039466)
> 👉[네이버 검색 광고 연동 가이드 확인하러 가기](https://help.airbridge.io/hc/ko/articles/900001057943)
---
## iOS SKAdNetwork(SKAN) 기본 개념과 측정 방식 이해하기
URL: https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork
Language: Korean
Published: 2024-03-27
Apple의 iOS 14.5 업데이트 이후, SKAdNetwork(SKAN)가 새롭게 대두되면서 모바일 앱 광고 업계에 상당한 영향을 미치고 있습니다. SKAdNetwork(Store Kit AdNetwork)는 iOS 환경에서 개인정보를 보호하며 사용자 또는 기기에 대한 정보 없이 앱 설치 캠페인의 성과를 측정할 수 있는 프레임워크입니다. 이번 포스트를 통해 SKAdNetwork의 개념을 이해하고 SKAdNetwork로 캠페인 성과를 측정하기 어려운 이유와 SKAdNetwork 성과 데이터가 갖는 제약 사항을 함께 살펴보겠습니다.
> [주요 내용]
- SKAdNetwork(SKAN)의 기본 개념을 이해합니다.
- SKAdNetwork 개념의 핵심이 되는 Conversion Value(전환값)과 Postback 전송방식에 대해 살펴봅니다.
- SKAdNetwork의 성과 데이터의 특징을 파악합니다.
## SKAdNetwork(SKAN) 이해하기
### 등장 배경
iOS 14.5 업데이트로 인해, [앱 추적 투명성](https://developer.apple.com/app-store/user-privacy-and-data-use/)(App Tracking Transparency, 이하 ATT) 정책이 도입되며 앱 설치 시 광고 식별자(IDentity For Advertisers, 이하 IDFA) 정보에 대한 수집 동의 여부를 Opt-in 방식으로 수집하도록 의무화 되었습니다.
*IDFA*는 Apple에서 유저의 기기에 할당한 임의의 기기 식별자로, IDFA를 통해 개인 정보에 대한 노출 없이 맞춤형 광고를 제공하거나 유저를 식별할 수 있습니다. iOS 14.5 업데이트 이전에는 IDFA 수집을 거부한 기기를 제외하고 자동적으로 IDFA를 수집하여 [디바이스 매칭 기법](https://help.airbridge.io/hc/ko/articles/900003300526-%EB%8F%99%EC%9D%BC-%EC%82%AC%EC%9A%A9%EC%9E%90-%EC%8B%9D%EB%B3%84-%EA%B8%B0%EC%88%A0#%EB%94%94%EB%B0%94%EC%9D%B4%EC%8A%A4-%EB%A7%A4%EC%B9%AD-%ED%99%95%EC%A0%95%EC%A0%81-%EB%B0%A9%EB%B2%95)을 통해 광고 캠페인의 성과를 측정할 수 있었죠.
여기서 *Opt-in 방식*이란 Option-in의 약자로, 개인정보 주체에게 정보 이용에 대한 동의를 먼저 받은 후에 개인 정보를 수집 및 활용할 수 있는 방식을 말합니다. 만약, 정보 이용에 대한 동의를 받지 않고 먼저 개인 정보를 수집 및 활용한 후, 추후에 동의 거부 의사를 밝힌 주체에 대한 정보만 이용을 중지하기 원한다면 Opt-out 방식을 사용하면 됩니다.
이와 같이 IDFA 수집이 제한이 따르는 상황에서 개인정보 보호를 준수하며 앱 마케팅 캠페인 성과를 측정할 수 있는 새로운 프레임워크로 도입된 것이 SKAdNetwork 입니다.
### 기본 개념
SKAdNetwork는 기존의 [어트리뷰션 모델](https://help.airbridge.io/hc/ko/articles/900001955286)과 달리 개인정보 없는 마케팅 캠페인 성과 측정 방식을 제시하는 프레임워크입니다.
1) iOS 앱 설치 캠페인을 통해 발생한 설치 이벤트에 대해 애플이 직접 선정한 기여 매체에게 유저 혹은 기기 정보를 제외한 **기여 정보**를 **전환 성과**와 함께 제공하고, 2) 매체가 이를 다시 MMP에 전송한 뒤, 3) MMP가 기여 정보를 해석해 광고주에게 성과 데이터를 제공합니다. 따라서 기존 어트리뷰션 모델에서 어트리뷰션을 분석하는 역할을 하는 MMP는 SKAdNetwork 환경에서 매체별 광고 성과를 쉽게 확인할 수 있도록 집계된 데이터를 제공(Aggregator)하는 역할을 하게 됩니다. 즉, MMP를 사용해야만 앱 설치 캠페인에 대한 성과를 쉽게 확인할 수 있습니다.
[]
- SKAdNetwork는 iOS에서 개인정보를 보호하며 마케팅 성과를 측정할 수 있도록 하는 프레임 워크로 유저 또는 디바이스 관련 정보가 생략된 캠페인 성과 데이터를 애플이 직접 매체에게 제공합니다.
- SKAdNetwork 캠페인의 성과 측정은 Conversion Value 라고 하는 숫자 값을 통해, Install 이후의 인앱 이벤트(In-app Event) 정보를 제공합니다. 이때, 해당 이벤트를 발생시킨 유저를 추정할 수 없게 만들기 위해서 랜덤화한 지연 발송 방식을 사용합니다.
- SKAdNetwork 환경에서 MMP는 어트리뷰션 분석이 아닌 매체로 부터 제공받은 SKAdNetwork의 포스트백을 사용자가 입력한 Converison Value 설정값에 따라 해석하여 제공합니다.
## SKAdNetwork 이해를 위한 핵심 개념
### Conversion Value 란?
SKAdNetwork 환경에서는 캠페인의 전환 성과를 Conversion Value이라는 숫자 값으로 측정합니다. Conversion Value는 2진수 6bit 숫자로 iOS 앱 설치 캠페인으로 발생한 앱 설치 및 전환 성과(후속 인앱이벤트)를 추적해 기록한 것입니다. 예를 들어, Conversion Value의 1을 [Sign-up(회원가입)]으로 설정해두었다고 가정해봅시다. 한 유저가 SKAdNetwork 환경에서 앱 설치 캠페인으로 유입되어 앱 설치 후 ‘회원가입’을 했다면, Conversion Value 1로 기록이 남게 됩니다. 따라서 iOS 앱 설치 캠페인의 전환 성과로 측정하고자 하는 인앱 이벤트가 있다면, 사전에 해당 인앱이벤트를 Conversion Value에 설정해야 합니다.
- Conversion Value는 앱 설치 이후 첫번째 앱 실행(First Open) 시점부터 활성화되어 인앱이벤트 발생 여부에 따라 업데이트 됩니다. 이 때 업데이트는 높은 숫자로만(Upper) 진행됩니다.
- Conversion Value에 해당하는 인앱이벤트가 발생할 경우, 해당 Conversion Value가 업데이트되고, 기존 숫자보다 낮은 숫자에 해당하는 이벤트는 이벤트가 발생해도 Conversion Value가 업데이트 되지 않습니다.
- 예를 들어 Conversion Value의 5에 [회원가입] 이벤트를 설정하고 Conversion Value의 3에 [구매] 이벤트를 설정한다면, 앱 설치 이후 [회원 가입]을 진행하고 [구매]로 이벤트가 이어질 때 회원 가입 이후의 구매 이벤트는 측정되지 않습니다. [구매]에 해당하는 Conversion Value 값이 [회원가입] 보다 작은 수이기 때문입니다.
- Conversion Value에는 인앱 이벤트를 측정하기 위해서 일정 시간 동안 포스트백을 전송하지 않고 대기하는 **타이머**가 작동합니다. 인앱 이벤트가 발생해 Conversion Value가 업데이트되면, 타이머는 초기화되고 다음 인앱 이벤트가 발생할 때까지 포스트백을 전송하지 않고 대기합니다. 24시간 이내에 새로운 인앱 이벤트가 발생하지 않으면 Conversion Value는 인앱 이벤트 측정을 종료해 더이상의 업데이트를 하지 않고 Postback 전송을 준비합니다.
### SKAdNetwork의 포스트백 전송 방식
Conversion Value가 업데이트 완료되면, Conversion Value 값을 포함한 Apple에서 측정한 기여 정보는 아래와 같은 포스트백 형태로 매체에 전달됩니다. 이때 기여 정보에는 캠페인 정보, 광고 지면 앱(app-id) 정보, 신규 설치한 광고주 앱(source-app-id) 정보 등이 포함됩니다.
포스트백 전송은 실시간이 아니며 앱 설치 이후의 후속 인앱 이벤트 측정을 위해 일정 시간(에어브릿지 기준 24시간)을 대기한 뒤 0~24시간 이내에 랜덤하게 사용자의 기기에서 기여된 매체(ad network)로 전송됩니다. 전송 시점이 일정하지 않은 이유는 전송 시점에 따른 사용자 혹은 기기 수준의 정보를 유추하는 것을 방지해 개인정보를 보호하기 위한 것으로 볼 수 있습니다.
**포스트백 데이터 예시**
[]
👉 [SKAdNetwork의 포스트백 파라미터에 대해 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/4412679790873#h_01G1365MKXXP0Y5BQKZTMY1X7B)
## SKAdNetwork로 측정된 성과 데이터의 특징
위와 같은 측정 방식으로 인해 SKAdNetwork로 측정된 성과 데이터는 다음과 같은 특징을 가지고 있습니다.
**기여 정보에서 유저 또는 기기 정보를 확인할 수 없으며 따라서 리타게팅 혹은 리인게이지먼트 캠페인의 성과 데이터는 지원하지 않습니다.*
SKAdNetwork는 개인정보를 보호하기 위해 유저 혹은 기기 수준의 정보를 제외한 기여 정보를 제공합니다. 따라서 사용자 혹은 기기 정보를 통해 어트리뷰션을 해야 하는 리타게팅 캠페인 혹은 리인게이지먼트 캠페인의 성과를 측정할 수 없습니다.
**전환 성과의 발생 시간을 알 수 없으며 실시간 성과 측정이 불가능합니다.*
SKAdNetwork 프레임워크는 앱 설치 후 일정 대기 시간동안 Conversion Value로 설정된 인앱이벤트를 측정한 뒤, 일정 대기 시간(Conversion Value 측정 타이머, 에어브릿지 기준 24시간)이 종료한 시점으로부터 0~24시간 이내의 랜덤한 시간에 포스트백을 전송하기 때문입니다.
**SKAdNetwork의 성과 데이터는 기존 어트리뷰션 모델 측정 방식의 데이터와 중복 측정될 수 있습니다. *
SKAdNetwork 프레임워크로 측정되는 인앱이벤트는 기본적으로 앱 내에서 발생한 전환 이벤트이기 때문에 에어브릿지 SDK를 통한 기존 트래킹 방식으로도 측정됩니다. 따라서 SKAdNetwork 방식으로 측정된 전환 성과 데이터를 분석할 때 이 부분을 고려하는 것이 필요합니다.
## SKAdNetwork, 어떻게 대응해야 하는가?
지금까지 SKAdNetwork의 등장 배경과 함께 기본 개념을 이해하고, SKAdNetwork의 핵심이 되는 Conversion Value와 Postback 전송 방식에 대해 살펴보았습니다. SKAdNetwork는 유저의 개인정보를 보호하기 위해 Conversion Value 라는 2진수 6비트 전환값을 통해 앱 설치 이후의 전환 성과를 측정하고, Postback 전송을 지연시키는 방식을 활용합니다. 이러한 특징 때문에 SKAdNetwork는 기존의 어트리뷰션 모델과 달리 실시간 성과 측정이 불가능하고 유저를 타겟팅하는 캠페인을 측정할 수 없다는 제한점이 있습니다.
다음 포스트에서는 이러한 SKAdNetwork의 제약 사항을 에어브릿지와 함께 극복하고 캠페인 성과를 효과적으로 측정하기 위해 해야하는 작업들에 대해 다뤄보도록 하겠습니다.
---
## 릴리즈 노트: 2022년 5~6월
URL: https://www.airbridge.io/ko/blog/release-notes-may-jun-2022
Language: Korean
Published: 2024-03-27
2022년 5월과 6월, 리포트 및 대시보드 업데이트와 신규 연동 매체 추가로 에어브릿지의 사용성이 개선되었습니다. 특히 Beta로 제공되었던 업데이트된 Retention 리포트가 정식으로 릴리즈되었으며, 기존에 Actuals 리포트에서만 지원하던 메트릭과 그룹바이가 Trend 리포트에 추가되어 같은 기준으로 데이터를 살펴볼 수 있습니다. 또한, App Store Connect의 새로운 기능인 Custom Product Page(맞춤형 제품 페이지)로의 트래킹 링크 설정 및 랜딩을 할 수 있게 되었습니다. 그리고 Google Play Install Referrer를 통한 페이스북 로우 데이터를 활용할 수 있는 기능이 배포되었습니다. 이외에도 MMP 최초의 카카오 키워드 광고 비용 연동부터, SDK 업데이트 및 신규 매체 연동까지 에어브릿지의 릴리즈 소식을 소개합니다. 이제 에어브릿지에서 보다 정확하고 다양한 데이터를 확인해 보세요.
## 📈 에어브릿지 리포트가 업데이트되었습니다.
### 업데이트된 에어브릿지 Retention 리포트가 정식 릴리즈되었습니다.
[__wf_reserved_decorative]
Beta로 제공되었던 업데이트된 Retention 리포트가 정식으로 릴리즈되었습니다. Retention 리포트는 앱 내에서 특정 이벤트를 수행한 유저들 중 얼마나 많은 유저들이 특정 날짜에 다시 앱으로 돌아와 특정 이벤트를 수행했는지를 N Day 리텐션(잔존율) 방식으로 제공합니다.
Retention 리포트를 활용하면 데이터를 통해 ‘제품 및 서비스에 대한 충성도가 높은 유저’ 유입에 가장 효과적인 채널과 캠페인, 광고 그룹이 무엇인지 확인할 수 있습니다. 또한 리포트의 성과를 기준으로, 캠페인을 최적화하거나 유료 광고 정산에 활용할 수도 있습니다.
이번 리포트 업데이트는 [에어브릿지가 개발한 데이터베이스 Luft](https://www.airbridge.io/ko/blog/introducing-luft)를 기반으로 이루어졌습니다. 업데이트를 통해 다양한 그룹바이와 필터 설정을 제공하게 되었으며, 특정한 조건을 만족하는 사용자 그룹인 코호트를 통해 더 세분화된 분석을 지원하게 되었습니다.
👉 [에어브릿지 Retention 리포트 업데이트 소식 더 자세히 확인하기](https://www.airbridge.io/ko/blog/airbridge-retention-report-update)
### 에어브릿지 Trend 리포트에 새로운 메트릭과 그룹바이가 추가되었습니다.
[__wf_reserved_decorative]
에어브릿지의 Trend 리포트가 업데이트되었습니다. Trend 리포트는 Actuals 리포트의 데이터를 Trend Line으로 분석할 수 있는 리포트로, 선 그래프를 활용한 시각화를 통해 통계 데이터의 변화 추이를 보다 직관적으로 확인할 수 있습니다. 상단에서는 선 그래프로, 하단에서는 테이블 형태로 시간별 추이 데이터를 확인할 수 있습니다.
기존에 Actuals 리포트에서는 지원하고 Trend 리포트에서는 지원하지 않던 SKAdNetwork 등의 메트릭과 그룹바이가 추가되어, Actuals 리포트와 같은 기준으로 데이터를 확인할 수 있습니다. 이제 Actuals 리포트와 통일된 기준의 데이터를 Trend 리포트에서 선 그래프로 보다 직관적으로 확인하고 마케팅 성과를 개선해 보세요.
👉 [에어브릿지 Trend 리포트 설정 방법 더 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/900003516746-Trend)
## 🔗 Custom Product Page(맞춤형 제품 페이지)로의 트래킹 링크 설정 및 랜딩
Apple의 iOS 15 업데이트 이후 Custom Product Page 기능이 새롭게 추가되었습니다. 이제 발맞춰 에어브릿지에서도 트래킹 링크의 최종 랜딩을 Custom Product Page로 설정할 수 있도록 트래킹 링크 생성 페이지가 업데이트되었습니다.
### App Store Connect의 Custom Product Page(맞춤형 제품 페이지)란?
iOS 15 및 iPadOS 15 이후 버전에서 제공되는 기능으로, App Store의 제품 페이지를 기본 외에도 최대 35개까지 맞춤 페이지를 만들어서 각 페이지마다 스크린샷, 홍보 문구 및 앱 미리 보기를 다르게 구성할 수 있습니다. 그리고 설정한 세그먼트에 따라 각각 다른 제품 페이지로 랜딩 할 수 있습니다.
👉 [Custom Product Page (맞춤형 제품 페이지) 더 자세히 알아보기](https://developer.apple.com/kr/app-store/custom-product-pages/)
### 에어브릿지에서 Custom Product Page로의 트래킹 링크 설정 및 랜딩 하기
에어브릿지에서 트래킹 링크를 생성할 때, 트래킹 링크를 통해 연결되는 랜딩 페이지 설정 시 iOS의 최종 목적지로 ‘애플 앱 스토어: 기본 제품 페이지’ 외에도 ‘애플 앱 스토어: 맞춤형 제품 페이지’가 추가되었습니다. 이때 App Store Connect에서 생성한 맞춤형 제품 페이지 링크를 입력하면 앱 스토어 랜딩 시 해당 맞춤형 제품 페이지로 랜딩됩니다.
👉 [에어브릿지 트래킹 링크 생성 방법 더 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/900001037886)
## 💰 카카오 키워드 광고 비용 연동을 통해, 에어브릿지가 MMP 최초로 한국에서 운영하는 모든 검색 광고 매체의 비용 데이터를 연동하였습니다.
[__wf_reserved_decorative]
에어브릿지가 카카오 키워드 광고 연동을 진행하면서, MMP 최초로 한국에서 운영하는 모든 검색 광고 매체의 비용 데이터를 연동하였습니다.
국내에서 마케팅 캠페인을 집행한다면 빼놓을 수 없는 SA(Search Ads, 검색 광고) 매체인 구글 애즈, 네이버 검색 광고, 카카오 키워드 광고. 이제 각 SA 매체 별 대시보드에 들어가서, 각각의 기준이 적용된 전환 성과 및 비용을 확인할 필요가 없습니다. 국내에서 운영하는 모든 SA 매체의 비용 데이터를 한 번에 확인할 수 있는 유일한 어트리뷰션 솔루션 에어브릿지에서 모든 검색 광고 매체 비용 데이터를 한 번에 확인해 보세요!
👉 [에어브릿지의 카카오 키워드 광고 연동 방법 및 SA 비용 데이터 확인 방법 더 자세히 알아보기](https://www.airbridge.io/ko/blog/airbridge-x-kakaokeyword-cost)
## 🖇 추가 매체 연동
에어브릿지는 지속적으로 파트너사를 확대해가고 있으며, 추가적인 매체 연동을 진행하고 있습니다. 에어브릿지에 연동된 채널인 Integrated Channels는 사전에 정의된 구분 값을 가지고 있으며, 대시보드의 \[Integrations - Integrated Ad channels\] 탭에서 각 매체와 편리하게 연동할 수 있습니다.
**5월과 6월에는, MecrossPro, 블루스택(Bluestacks), 플라이모비(Flymobi), 녹스(Noxmobi)가 추가로 연동되었습니다.**
에어브릿지와 연동된 전체 Integrated Channels 리스트는 [여기 링크](http://abit.ly/integrated-channels)에서 확인해 보세요!
## ⛏ SDK 릴리즈
에어브릿지는 앱(Android & iOS) 및 웹 SDK를 기반으로 웹, 앱 데이터를 통합적으로 분석합니다. 또한 지속적인 업데이트를 통해 사용성을 개선하고 있습니다.
**5월과 6월에는 Airbridge Android SDK 2.18.0, Airbridge Web SDK 1.8.142, Airbridge Web SDK 1.8.150, Airbridge Web SDK 1.8.154가 릴리즈 되었으며, 신규 SDK인 Airbridge Expo SDK 1.0.0가 릴리즈 되었습니다.**
---
## ROI 개선을 위한 모바일 마케팅 전략: 재구매 전환율과 주기 계산하기
URL: https://www.airbridge.io/ko/blog/how-to-increase-repeat-purchases
Language: Korean
Published: 2024-03-27
많은 비용을 들여 유입시킨 신규 유저가 한 번만 구입하고 휴면 유저가 되어버린다면 전체적인 마케팅 투자 대비 수익, 즉 ROI 측면에서 좋은 성과를 거두기 어렵습니다. 마케팅 예산에 여유가 있다면 첫 결제 할인 쿠폰 등을 통해 많은 유저를 한 번에 유입시킴으로서 재구매 가능성을 가진 유저의 모수 자체를 늘릴 수도 있습니다. 하지만 이어지는 경기 침체로 인한 마케팅 예산 감축과 고객들의 제한된 구매 능력 안에서 서비스 간 경쟁도 심화되고 있습니다. 게다가 체리피커로 일컬어지는 첫 구매 혜택 직후 이탈 고객도 상당합니다.
이러한 상황에서 어떤 유저가 재구매 가능성이 높을지 미리 파악할 수 있다면 예산을 보다 효과적으로 사용할 수 있습니다. 특히 채널별/캠페인별 성과를 다각도로 분석해 보면 가장 높은 ROI를 보이는 유저를 데려올 수 있는 채널과 캠페인을 발견할 수 있습니다. 이번 포스트는 ‘재구매’ 분석에 집중합니다. 특정 채널 혹은 캠페인으로 유입된 유저의 재구매 전환율과 주기를 확인하고 개선할 수 있는 방법을 알아봅니다.
## 재구매 개선을 위해 확인해야 하는 지표
재구매를 개선하기 위해서는 얼마나 많은 유저가 첫 구매 이후 재구매를 발생시키는지, 그리고 어떤 주기로 재방문하여 구매를 일으키는지 알아차리는 것이 먼저입니다. 즉, ‘재구매 전환율’과 ‘재구매 주기’를 파악해야 합니다.
### 재구매 전환율이란?
재구매 전환율은 최초 구매 전환 후 재구매를 하는 고객의 비율을 측정하는 지표입니다. 주로 이커머스나 구독 기반의 비즈니스에서 첫 구매 이후 반복 구매를 일으키는 고객의 비율을 구하기 위해 사용합니다. 직접 계산할 때는 아래와 같은 계산식을 활용할 수 있습니다.
[__wf_reserved_decorative]
### 재구매 주기란?
재구매 주기는 고객이 첫 구매를 일으킨 후 구매를 반복하는 평균 시간을 나타냅니다. 특히 이커머스 비즈니스가 고객의 재방문 시점을 파악하는 데에 도움을 줍니다. 고객의 행동을 이해하고 마케팅 전략을 구상하기 위한 주요 메트릭 중 하나입니다. 직접 계산할 때는 아래와 같은 계산식을 활용할 수 있습니다.
[__wf_reserved_decorative]
## 퍼널 리포트에서 재구매 전환율과 재구매 주기 계산하기
재구매 전환율과 주기를 계산하기 위해 앞서 소개한 것과 같은 간단한 계산식을 사용할 수도 있고, 프로덕트 분석 툴에서 제공하는 기능을 활용할 수도 있습니다. 다만, 마케터라면 어떤 채널이 혹은 어떤 캠페인이 가장 높은 ROI를 발생시키는지 확인하고 싶겠지만 프로덕트 분석 툴에서는 유저들이 어떤 광고를 통해 유입되었는지 분석할 수 없습니다. 구글 등과 같은 SAN(Self Attributing Network) 매체는 엄격한 데이터 규정을 바탕으로 파트너십을 맺은 MMP에게만 광고 데이터를 제공하기 때문입니다. 그렇다면 각 채널/캠페인 별 광고로 유입된 유저의 앱 내 행동은 MMP에서만 확인할 수 있고, 특정 채널/캠페인으로 획득한 유저의 재구매 전환율과 주기도 MMP에서만 확인할 수 있다는 것을 의미합니다.
에어브릿지는 MMP 중 유일하게 퍼널 리포트를 제공해서 광고로 유입된 유저가 어떤 이벤트를 수행하는지 확인할 수 있도록 제공합니다. 그리고 이 리포트를 통해 재구매 전환율과 주기를 간단하게 계산할 수 있습니다.
> 에어브릿지에서 프로덕트 내의 유저가 수행한 특정 이벤트의 반복 주기, 우리 서비스의 구매 주기 등에 대한 정보를 바로 확인할 수 있는 별도의 차트 또는 기능을 제공하는 것은 아닙니다. 그러나 에어브릿지 퍼널 리포트를 활용해서 일정 조건 안에서의 유저의 행동 데이터를 분석하여 특정 이벤트 반복 주기 / 구매 주기를 확인할 수 있습니다.
>
> 👉 [가이드에서 자세한 퍼널 리포트 사용 방법 확인하기](https://help.airbridge.io/ko/guides/funnel-report)
### 에어브릿지에서 재구매 전환율과 주기 계산하기
에어브릿지 퍼널 리포트에서 퍼널은 유저가 완료하는 일련의 이벤트로 정의됩니다. 퍼널 분석을 통해 유저가 어떤 이벤트에서 이탈하거나 완료하는지 파악할 수 있습니다. 그래서 앱 설치 후 구매 완료 이벤트의 수행 여부를 분석함으로써 재구매 전환율과 주기를 간단하게 계산할 수 있습니다.
구체적인 방법은 아래의 대시보드 예시 화면처럼 원하는 코호트와 확인하고 싶은 기간을 설정한 후 스텝에서 ‘Install(App) -> 구매 완료(App) -> 구매 완료(App) -> 구매 완료(App) -> 구매 완료(App)’ 와 같이 설정합니다. 그리고 그룹바이를 원하는 대로 설정합니다. 총 세 개의 그룹바이까지 설정할 수 있어서 원하는 수준까지 쪼개어 데이터를 확인할 수 있습니다. 예시에서는 그룹바이를 ‘채널(Channel)’로만 설정하였습니다.
이렇게 설정을 완료하고 나면, 아래 이미지와 같이 **각 채널별 첫 구매 이후 전환율**, 그리고 각 스텝 사이의 **평균 전환 시간**과 **전환 시간의 중간값**을 제공하여 **재구매 주기**를 대략적으로 파악할 수 있습니다.
[__wf_reserved_decorative]
## 마케팅 퍼널 단계별 재구매 개선 방법
이와 같이 계산한 결과를 바탕으로 마케팅 퍼널 중 유저 획득과 리텐션 & 인게이지먼트 단계에서 활용할 수 있는 재구매 개선 전략을 알아보겠습니다.
### 유저 획득 단계 - 재구매 전환율이 유의미한 채널로 예산 분배
유저 획득(User Acquisition, UA) 캠페인의 성과를 확인하는 지표는 주로 CPI(Cost per install)와 ROAS(Return on ad spend)입니다. 이때 ‘재구매 전환율’을 같이 확인하면 CPI가 낮더라도 재구매 전환율이 유의미한 채널에 예산을 늘리는 최적화를 통해 ROAS를 개선할 수 있습니다.
예를 들어, 신규 설치 이후 첫 구매까지의 전환율을 확인했을 때 좋은 성과를 보이는 채널이더라도 재구매 전환율은 낮을 수 있습니다. 단지 첫 구매 전환율까지만 확인하기보다는 재구매 전환율까지 확인하면 우리 앱 서비스의 성장에 진짜 기여하는 채널과 캠페인을 확인할 수 있습니다.
### 리텐션 & 인게이지먼트 단계 - 재구매 주기 기반의 CRM 캠페인 집행
리텐션과 인게이지먼트를 개선하기 위해서는 ‘재구매 주기’를 활용할 수 있습니다.
**1) 평균 재구매 시간 이후 구매 일으키지 않는 유저 대상 캠페인**
예를 들어, 1회 구매 후 평균 재구매 시간이 지났는데도 구매를 일으키지 않는다면 해당 유저에게 푸시 혹은 이메일을 발송하는 CRM 캠페인을 운영할 수 있습니다. 특히 이때 개인화된 캠페인을 통해 장바구니에 담긴 제품에 대한 할인 혹은 상세 페이지를 살펴본 후 구매하지 않은 상품과 비슷한 상품 추천 등 적절한 메시지를 보낸다면 보다 효과를 높일 수 있습니다.
**2) 재구매 주기에 맞춘 캠페인 운영**
우리 유저의 재구매 주기를 통해 마케팅 전략에 대한 인사이트를 얻을 수 있습니다. 예를 들어, 마케팅 캠페인을 일반적인 재구매 주기에 맞추어 조정하는 것도 효과적입니다. 구체적으로는 예상되는 재구매 주기 직전에 알림 혹은 프로모션을 발송함으로써 재구매를 유도할 수 있습니다. 혹은 재구매 전 이탈하는 특정 지점이 있다면, 해당 부분에 대해 프로덕트 팀에 개선 의견을 낼 수도 있습니다.
ROI를 개선하기 위해 확인해야 할 중요한 지표 중 하나인 ‘재구매’. 이번 포스트를 통해 재구매 전환율과 주기를 간단하게 확인하는 방법과 이를 통해 재구매를 개선하는 방법을 알아보았습니다. 각 서비스 별로 이상적인 재구매 전환율과 주기는 다를 수 있습니다. 현재 우리 서비스의 상황을 지표를 통해 확인하고 여기서 얻은 인사이트를 기반으로 성과를 개선해 보세요.
모바일 마케팅 전략에 대한 고민을 해결하고 성장을 극대화하고 싶다면 [지금 모바일 마케팅 전문가들과 이야기 나눠 보세요](https://www.airbridge.io/ko/book-a-demo).
---
## 토스애즈 광고 성과를 에어브릿지에서 확인해 보세요
URL: https://www.airbridge.io/ko/blog/airbridge-toss-ads-integration
Language: Korean
Published: 2024-03-27
Author: Team Airbridge
에어브릿지(Airbridge)와 토스의 광고 플랫폼 [토스애즈](https://tossads.toss.im/)가 파트너로 함께하게 되었습니다. 이제 토스애즈에서 발생한 광고 성과를 에어브릿지에서 확인할 수 있습니다.
마케팅 비용을 들여 광고를 집행했다면 반드시 성과를 측정하여 기대한 효과를 얻고 있는지 분석해야 합니다. 이 과정을 통해 광고 캠페인을 최적화하고 ROAS를 개선할 수 있습니다. 물론 각 광고 플랫폼에서도 집행 비용 및 전환 성과를 제공하고 있지만, 다양한 매체를 통해 운영하고 있는 광고 성과를 한 눈에 확인할 수 있다면 보다 효율적으로 성과를 분석할 수 있습니다. 그래서 에어브릿지는 다양한 광고 플랫폼과의 연동 및 파트너십을 통해 성과 데이터를 제공합니다.
이번 토스애즈와의 연동도 에어브릿지 유저의 효율성과 편의성 개선에 기여할 것으로 기대됩니다. 토스애즈는 토스 앱 혜택탭 최상단의 배너 광고, 머니 알림, 행운퀴즈, 라이브 쇼핑 보기 등 다양한 광고 상품을 제공하고 있습니다. 광범위한 데이터 기반의 데모그래피, 관심사, 소비 업종 등을 고려한 정교한 타겟팅도 가능하다는 점이 장점입니다.
## 에어브릿지에서 토스애즈 광고 성과 확인하는 방법
에어브릿지와 토스애즈의 연동 방법은 간단합니다. 연동 정보 입력이나 추가적인 설정 과정이 필요하지 않습니다.
[__wf_reserved_decorative]
우선 에어브릿지의 에어브릿지와 연동된 광고 채널 목록인 \[Ad Channel Integration\] 에서 ‘토스애즈(toss)’를 검색해 주세요. 클릭하면 ‘트래킹 링크’를 생성하고 확인할 수 있는 화면이 나옵니다.
그리고 광고 성과를 측정하는 데 필요한 트래킹 링크를 생성하여 광고 성과를 측정합니다. 트래킹 링크를 활용하기 전에 기여 기간과 포스트백을 목적에 맞게 설정하여 사용하는 것이 좋습니다.
구체적인 연동 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/toss-ads)에서 확인하세요.
에어브릿지는 효과적이고 심층적인 마케팅 성과 분석 서비스를 제공하기 위해 여러 플랫폼과의 연동을 확대해 나가고 있습니다. 토스애즈를 포함한 광고 성과 분석이 필요하시다면 [여기 링크](https://www.airbridge.io/ko/book-a-demo)에서 문의를 남겨주세요.
> 아래 링크에서 에어브릿지와 연동된 광고 채널 목록을 확인할 수 있습니다. 원하는 광고 채널을 찾지 못했다면 광고 채널에 연동 신청서 작성을 요청해 주세요.[
>
> \- 에어브릿지와 연동된 광고 채널 확인하기](https://help.airbridge.io/ko/guides/ad-channels-list#%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80%EC%99%80-%EC%97%B0%EB%8F%99%EB%90%9C-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-%EB%AA%A9%EB%A1%9D)[
> \- 광고 채널 연동 신청서](https://help.airbridge.io/ko/supports/integrations)
---
## 2024년 마케터가 주목해야 할 디지털 광고 트렌드
URL: https://www.airbridge.io/ko/blog/2024-digital-marketing-trends
Language: Korean
Published: 2024-02-28
오늘날 마케터들은 지속적인 변화 속에서 민감하게 트렌드에 대응하고 새로운 기회를 포착해야 합니다. 트렌드 속에서 지금의 소비자의 생생한 목소리를 들을 수 있고 시장의 반응을 빠르게 파악할 수 있기 때문입니다. 급변하는 디지털 업계 특성상 어떤 방향으로 흘러갈지 예측하기는 쉽지 않지만 작년의 동향을 바탕으로 2024년에 지속해서 주목할 만한 광고 트렌드를 살펴봅니다.
## 쇼츠, CTV, 스트리밍 서비스… 영상으로 소통하는 시대
작년 미디어 업계의 한 해를 휩쓴 키워드가 있다면 단연 ‘쇼츠’입니다. 분초를 다투며 살아가는 바쁜 현대인들이 재생시간이 길어야 1분 이내인 **숏폼(short-form)** 영상을 선호하기 시작함과 동시에 영상 콘텐츠를 시청하는 주요 플랫폼이 TV에서 스마트폰, 태블릿PC 등으로 변화하였기 때문입니다.
쇼츠와 함께 **OTT (over-the-top) 스트리밍 서비스**와 **Connected TV (CTV)**의 출현도 미디어 업계의 판도를 바꿔놓았습니다. OTT 서비스와 기존 방송과의 가장 큰 차이점은 방송국이나 케이블 TV, 통신 사업자가 지정한 시간에 방송이 송출되는 것이 아닌 사용자가 직접 OTT 사업자의 앱이나 웹사이트에 접속해 언제든지 원하는 콘텐츠를 자유롭게 소비할 수 있는 부분입니다. 이러한 OTT 서비스는 인터넷이 연결되고 스크린이 있는 모든 기기에서 지원되며 스트리밍 기기, 스마트 TV, 셋톱박스, 게임 콘솔 등의 Connected TV (CTV)로서 우리 생활에 함께하고 있습니다.
또한 TV 제조사들의 자사 광고형 무료 OTT 서비스인 **FAST (Free-Ad-supported Streaming TV)** 역시 업계의 주목을 받고 있습니다. 단어의 의미대로 광고를 편성하는 대신 시청자에게 가입, 구독, 추가 기기 구매 없이 무료로 스트리밍 서비스를 제공하는 플랫폼으로 특히 북미 지역에서 각광받고 있습니다. 최근 삼성과 LG에서도 FAST 시장에서의 경쟁력 강화를 위하여 맞춤형 콘텐츠 및 서비스 분야에 대한 투자를 진행할 것으로 발표하였습니다.
새로운 서비스와 플랫폼의 등장은 마케터들에게 있어 더욱 다양한 형식의 광고 및 마케팅을 시도할 수 있는 좋은 기회입니다. 모든 시청자에게 일괄적으로 광고를 송출하던 지상파 방송과는 다르게 세밀하게 사용자를 타게팅 할 수 있고 그 결과를 데이터로 분석할 수 있는 OTT 광고가 마케터들 사이 큰 인기를 끌고 있으며 2024년에도 OTT 지면에 대한 광고 경쟁은 더 치열해질 것으로 예상합니다.
[__wf_reserved_decorative]
## #내돈내산 User-generated content (UGC)
브랜드의 휘황찬란한 광고 카피보다 사용자의 실제 사진이 담긴 솔직한 후기가 더 설득력 있게 다가오기 마련입니다. 맛집 리뷰, 영화 평점, 구매 후기와 같은 **사용자가 ‘직접’ 생성에 기여한 콘텐츠**가 그 어떤 마케팅 전략보다 효과적인 이유입니다. 2024년에도 여전히 사용자가 직접 제작한 글과 이미지, 동영상 등을 활용한 마케팅은 여전히 활발하게 이루어질 것으로 보입니다.
하나의 사례를 살펴보면, 에어브릿지 고객사 후야호는 사용자들이 직접 게임 플레이 영상을 제작하고 공유하여 오가닉 유입을 유도하는 마케팅 전략을 시도했습니다. 단순히 브랜드에서 제공하는 서비스를 체험하기보다 직접 참여하여 목소리를 내고 싶어 하는 알파 세대를 공략한 전략이었습니다. 또래 사용자들이 업로드한 영상들을 공유하여 다른 사용자의 참여를 자연스럽게 유도함과 동시에 사용자가 직접 업로드한 영상을 기반으로 광고 소재를 만들어 더 많은 유저들에게 노출되게 하였습니다. 시대에 발맞춘 마케팅 전략으로 많은 신규 유저 유입과 마케팅 비용 절감의 효과를 가져올 수 있었고 글로벌 시장에서도 당당히 자리매김할 수 있었습니다.
👉 [에어브릿지 고객사 후야호가 어떻게 UGC 콘텐츠를 마케팅 전략에 접목했는지 확인해 보세요!](https://www.airbridge.io/ko/case-studies/whoyaho-game-data-marketing-strategy)
[__wf_reserved_decorative]
## AI가 만드는 광고? 업계에 미치는 변화들
작년 한 해 ChatGPT에 대한 이야기가 끊이지 않았습니다. 기술 대체 불가 영역으로 생각되었던 창작, 예술 분야에서도 생성형 AI를 통한 다양한 시도가 펼쳐지고 있습니다. 일례로 제일기획에서는 삼성생명 브랜드 캠페인의 시안부터 이미지 그리고 배경음악의 생성까지 광고의 모든 요소를 AI로 제작하는 데 성공하였습니다. 다소 보수적인 이미지의 보험사가 새로운 기술을 사용해 광고를 시도했다는 점 자체로 긍정적인 이슈를 만들었고 다시금 기술의 발전을 실감하게 했습니다.
그 외에도 업계에서는 상품 상세 페이지 URL만 입력하면 자동으로 다양한 형태의 광고 배너 및 영상을 제작해 주거나 고객이 더 좋아할 것 같은 광고 문구로 수정하여 제공하는 서비스 역시 출시되고 있습니다. 광고 매체사에서도 과거 수동으로 진행하던 입찰을 머신러닝 기술에 기반하여 실시간으로 입찰을 하고 있습니다. 광고 생태계 전반에 미치고 있는 AI의 영향력과 그 입김은 올해도 더욱 거세질 것 같습니다.
## 개인정보 보호, 더 이상 선택 아닌 필수
전 세계적으로 개인정보 보호 관련 법 강화가 이루어지면서 구글, 애플에서도 해당 법령 준수를 위해 Privacy Sandbox, App Tracking Transparency (ATT) 와 같은 개인정보 보호 프레임워크를 출시하였습니다. 이러한 움직임으로 인해 디지털 마케팅에서 활발히 사용되었던 서드 파티 데이터 (Third-party data: 제3자의 데이터 수집 업체가 고객과의 접점 없이 여러 사이트를 통해 수집한 고객의 데이터)의 사용이 제한되었고 그 대안으로 퍼스트 파티 데이터(First-party data: 기업이 자사의 플랫폼을 통해 직접 수집한 고객 데이터)가 사용되는 **리테일 미디어 네트워크 (Retail Media Network)**가 다시 새롭게 주목을 받고 있습니다. 소매 업체가 온라인 쇼핑몰과 같은 자사 앱, 웹사이트 등의 디지털 플랫폼 상의 광고 지면 및 인프라를 제공하는 이 서비스는 타깃 고객층에 집중적으로 광고를 게시할 수 있고 구매 전환율이 매우 높다는 장점이 있어 광고주들의 이목을 끌고 있습니다.
매체사들 역시 더 이상 서드 파티 데이터에 의존하지 않기 위해 노력 중입니다. 대표적으로 Amazon DSP는 Signal based Ads라는 이름으로 광고주의 퍼스트 파티 데이터와 매체사 자체 시그널을 머신러닝으로 조합하여 광고를 입찰 및 송출하는 서비스를 출시하였습니다. 유저의 개인정보를 보호하면서도 어떻게 하면 광고 효과를 극대화할 수 있을지 올해도 업계에서는 다방면으로 대안을 모색할 것으로 보입니다.
👉[2024년 변화하는 개인정보 보호 트렌드가 궁금하다면 여기 콘텐츠에서 확인해보세요.](https://www.airbridge.io/ko/blog/2024-privacy-landscape)
2024년에도 여지없이 업계에는 새로운 바람이 불 테지만 앞에서 다룬 신규 플랫폼들과 새로운 콘텐츠 형식 그리고 AI 자동화의 적극적인 활용 등이 마케터들에게 돌파구가 되어줄 것입니다. 다양한 변화 속에서도 변하지 않는 마케터들의 궁극적인 목표는 고객의 신뢰를 얻고 서비스의 성장을 이루는 일입니다. 새로운 트렌드를 한 발 앞서 예측하고 섭렵하여 올 한 해도 고객들의 마음을 사로잡을 수 있는 전략을 펼쳐보세요.
---
## 개인화 마케팅: 데이터로 고객의 마음을 사로잡는 방법
URL: https://www.airbridge.io/ko/blog/personalized-marketing
Language: Korean
Published: 2024-02-22
새로운 서비스를 처음 시작하거나 가입할 때, 나의 관심사나 취향에 대해 물어보는 앱과 웹 사이트가 유난히 많아짐을 느끼는 요즘입니다. 내가 뭘 좋아하는지, 어떤 것에 관심이 있는지를 입력하면 나를 위한 맞춤형 정보들이 제공되는 편리함에 익숙해지면서 당연하게 느껴지기까지 합니다.
몇 년 전부터 유행처럼 번진 개인화 마케팅이 여전히 중요한 이유와 유저가 남겨준 이 소중한 정보들을 어떻게 하면 마케팅에 더 잘 활용할 수 있을지 알아보는 시간을 가져봅니다.
## 개인화 마케팅이 여전히 중요한 이유
개인화 마케팅이란 기업이 고객의 관심사나 취향을 파악하고 해당 정보에 기반하여 각각에 알맞은 메세지를 전달하는 마케팅 전략입니다. 기존의 고객 층을 보다 세분화하여 맞춤광고 타겟팅을 진행합니다.
개인화 마케팅이 이 시대 마케터의 필수 전략이 된 데에는 여러 가지 이유가 있습니다. 고객은 자신의 선호도와 관심사에 맞는 맞춤형 정보와 제안을 받으며 브랜드에 대한 신뢰와 충성도를 쌓을 수 있고, 기업에서는 특정 고객에게만 관련성이 높은 정보를 제공하면서 전략적으로 광고를 집행할 수 있어 마케팅 비용 절감 효과까지 기대할 수 있습니다. 만족스러운 경험을 한 고객은 해당 서비스를 재방문할 가능성이 높아지며 입소문을 탄 브랜드는 시장에서 경쟁 우위를 확보할 수 있습니다.
### 개인의 취향이 각기 달라진 N극화 현상
‘평균 실종'이라는 말을 들어본 적이 있으신가요? 평균이라는 개념이 사라지고 있어 평균을 내는 게 의미가 없어진 현상을 뜻합니다. 이러한 평균이 사라지고 있는 데에는 다양한 원인이 있지만 그중 하나로 ‘N극화' 현상이 있습니다. 과거에는 대중이 특정한 유행을 따라가고 다수의 공통된 취향을 따라가려 했다면, 오늘날에는 각자의 취향이 다양하게 갈리고 개성을 중시하는 사람들이 늘어나며 생긴 현상입니다. 단순히 성별, 나이, 사는 지역 등으로 그룹화하여 유저를 타겟팅하기 보다 그 그룹 내의 고객들이 각각 원하는 것이 다를 수 있음을 이해해야 합니다. 그룹의 보편적인 특성을 분석하는 것도 중요하지만 고객 개개인의 취향을 존중한 타겟팅이 그 어느 때보다 필요한 이유입니다.
### 개인정보 보호 강화 움직임에 따른 변화
디지털 마케팅 업계에서 지난 몇 년 간 가장 큰 화두는 개인정보 보호 강화에 대한 움직임이었습니다. 2024년에는 구글(Google)의 서드 파티 쿠키(Third-party cookie)의 지원이 종료될 예정으로 또 하나의 거대한 변화가 있을 것으로 예상됩니다. 해당 지원이 종료되면 서드 파티 쿠키를 기반으로 유저를 타겟팅하던 기존의 마케팅 방식은 더 이상 사용할 수 없게 됩니다.
서드 파티 쿠키의 대체제로 ‘퍼스트 파티 데이터(First-party data)’ 가 주목을 받고 있습니다. 서드 파티 쿠키와 달리 서비스 및 기업에서 직접 수집하는 데이터로 유저 행동 패턴은 물론 설문조사와 같은 형식을 통해 보다 다양한 정보를 수집할 수 있습니다. 개인정보를 보호하면서도 정교한 유저 타겟팅이 가능하기 때문에 업계에서는 이미 주목 하고 있습니다. 퍼스트 파티 데이터를 사용하여 이메일, 푸시 알림과 같은 마케팅 캠페인, 그리고 서비스 내에서의 고객 경험까지 모든 것을 개인화할 수 있습니다. 고객 맞춤 경험이 늘어날 수록 서비스의 충성도와 전환율은 자연스레 높아집니다.
[👉 구글의 서드 파티 쿠키 지원 종료 및 2024 개인정보 보호 변화에 대한 자세한 내용을 확인해보세요.](https://www.airbridge.io/ko/blog/2024-privacy-landscape)
### 개인화를 넘어 초개인화시대로?
이제 개인화를 넘어 ‘초개인화' 의 시대입니다. AI 인공지능과 같이 고도화된 기술과 빅데이터를 활용하여 고객별로 매우 세분화된 제품, 서비스 또는 콘텐츠를 만드는 마케팅의 형태를 말합니다. 개인화 마케팅에서는 고객이 좋아할 만한 제품을 추천해 주는 것까지 가능하다면 초개인화 마케팅의 경우 유저 방문 및 구매 패턴을 수집한 뒤 구매 시기를 예측하여 적절한 시간에 유저가 원하는 것은 무엇인지 캐치하여 메세지를 보낼 수 있습니다.
초개인화 마케팅은 개개인의 차별화된 특성을 파악하고 고객이 자신도 인지하지 못한 향후 니즈까지 예측이 가능하기 때문에 다양한 행동 패턴을 기반으로 더 많은 구매 전환을 이끌어내어 큰 인기를 끌고 있습니다. 기술 발전의 도움으로 앞으로의 마케팅 전략이 어디까지 발전할 수 있을지 기대가 되는 부분입니다.
## 개인화 마케팅을 더 잘하기 위한 실전 활용법
### 광고 캠페인 타겟팅을 위한 오디언스 생성
이렇게 중요한 개인화 마케팅을 어떻게 하면 더욱 효과적으로 집행할 수 있을까요? 신규 유저 캠페인을 진행하여 새로운 유저를 획득했다면 해당 유저들이 지속적으로 앱에서 행동하고 소통하도록 만드는 것이 중요합니다. 리텐션과 고객 생애 가치(LTV)를 최대화하기 위해서 타겟 오디언스를 추출한 뒤 해당 오디언스에게만 광고 캠페인을 집행해 보세요.
에어브릿지에서는 SDK를 통해 축적된 데이터를 기반으로 쉽고 편리하게 타겟 오디언스를 생성하고 추출하는 기능을 제공합니다. 앱 설치 및 오픈, 회원 가입 등 특정 이벤트를 수행한 타겟을 설정할 수도 있고 여러 개의 조건을 조합하여 서비스 맞춤 오디언스 조건을 지정할 수도 있습니다. 한번 생성한 오디언스 조건은 자동으로 저장되어 간편하게 재추출 및 재정의할 수 있습니다.
[__wf_reserved_decorative]
그 외에도 추출한 데이터를 사내 자체 데이터와 결합하여 심층 분석을 진행할 수도 있고 특정 광고 채널 및 서드파티 솔루션과 자동으로 연동하여 맞춤 타겟으로 생성할 수도 있습니다. 이를 통해서 리타겟팅과 디타겟팅을 진행하거나 맞춤 타겟을 바탕으로 유사 타겟을 생성하여 효율적으로 유저를 획득할 수 있습니다. 개발팀과 마케터의 추가적인 공수 없이 빠르고 쉽게 원하는 타겟 오디언스를 생성하고 추출하여 효과적으로 개인화 마케팅을 진행해 보세요.
👉 에어브릿지의 오디언스 매니저는 자체 개발 엔진인 Luft를 사용하여 타사 대비 20배 빠른 속도로 작업이 가능합니다. 에어브릿지에서 오디언스를 생성하여 추출하는 방법에 대해서 자세히 알고 싶다면 [가이드 문서를 확인해 보세요.](https://help.airbridge.io/ko/guides/create-audience)
### 개인화된 유저 온보딩 플로우 설계 (멀티 솔루션 활용)
여러 솔루션을 함께 활용하여 보다 효과적으로 개인화된 유입된 유저의 온보딩 플로우를 설계할 수도 있습니다. 에어브릿지의 앱 설치 및 딥링크 어트리뷰션 결과를 브레이즈(Braze)와 같은 고객 인게이지먼트 솔루션으로 전송하면 해당 플랫폼에서는 어트리뷰션 데이터를 기반으로 특정 캠페인/ 광고 소재와 상호작용한 유저를 타겟팅하여 개인화된 메세지를 전송할 수 있습니다.
퀸잇은 위와 같이 에어브릿지와 브레이즈를 통합 활용하여 개인화 마케팅을 진행하고 있습니다. 에어브릿지와 브레이즈를 연동하여 에어브릿지의 어트리뷰션 데이터를 브레이즈로 보내고, 이 데이터를 활용해 브레이즈에서 세그먼트를 생성합니다. 예를 들어 특정 옷 브랜드의 광고를 본 후에 퀸잇으로 유입된 유저에게 브레이즈로 10% 할인 쿠폰을 발송하는 방식입니다. 그 외에도 어트리뷰션 데이터를 활용하여 재방문 유도, 쿠폰 발송, 장바구니에 상품을 추가한 고객에게 푸시 메세지를 보내는 일련의 캠페인을 진행할 수도 있습니다. 이렇게 여러 솔루션을 함께 활용하면 보다 효과적으로 개인화 CRM 플로우를 지속적으로 고도화할 수 있습니다.
[👉 퀸잇이 에어브릿지와 브레이즈를 활용하여 어떻게 개인화 마케팅을 진행하고 있는지 자세한 내용을 블로그 포스트에서 확인해 보세요.](https://blog.ab180.co/posts/rapportlabs-interview)
[__wf_reserved_decorative]
시대의 흐름과 함께 마케팅 전략 역시 빠르게 변화하고 있습니다. 다양한 솔루션을 활용하여 트렌드를 섭렵한 마케팅 전략을 세우는 방법이 궁금하시다면 [에어브릿지의 풀퍼널 마케팅 전문가들과 함께 이야기 나눠 보세요.](https://www.airbridge.io/ko/book-a-demo)
---
## Digital Markets Act(DMA): 디지털 경쟁과 개인정보 보호의 미래
URL: https://www.airbridge.io/ko/blog/digital-markets-act
Language: Korean
Published: 2024-02-06
올해도 여전히 디지털 마케팅 업계에서의 가장 큰 화두는 ‘개인정보 보호'입니다. 유저의 온라인 기본 권리를 보호하면서 공정하고 개방적인 디지털 시장을 만들기 위해 유럽 연합에서 규제에 나섰습니다. 국내에서 출시된 서비스 일지라도 전세계에 걸쳐 유저가 있는 서비스를 운영하고 있는 마케터라면 이번 변화에 주목해야 합니다. 이번 블로그 포스트에서는 아직은 막연하게만 느껴지는 디지털 시장법(Digital Markets Act, DMA)에 대해 알기 쉽게 풀어봅니다.
## Digital Markets Act(DMA)이란?
[디지털 시장법](https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/digital-markets-act-ensuring-fair-and-open-digital-markets_en)은 유럽 연합에서 제정한 디지털 시장 규제 법안으로, 대형 디지털 플랫폼의 힘과 영향력을 조절하고 공정한 경쟁을 촉진하기 위해 발의되었습니다. 유저와 판매자를 중개하는 ‘문지기(Gatekeeper)’ 역할을 하는 대형 플랫폼 기업의 독과점을 완화하기 위해 시작된 법안은 주로 빅테크 기업에 대한 규제, 공정한 경쟁 조건의 유지, 소비자 보호 등을 중점적으로 다루고 있습니다. EU는 지난 3년동안 월 4,500만명의 활성 사용자, 그리고 연 10,000명 이상의 활성 사용자에게 플랫폼 서비스를 제공하는 회사를 게이트키퍼로 정의했습니다. 게이트키퍼로 지정된 6개의 기업은 알파벳(Alphabet), 아마존(Amazon), 애플(Apple), 바이트댄스(ByteDance), 메타(Meta), 마이크로소프트(Microsoft)로 이들은 DMA의 모든 규정들을 준수해야 하며 2024년 3월 본격 시행을 앞두고 있습니다.
### DMA의 주요 내용과 규정
게이트키퍼로 지정된 기업들은 다음과 같은 의무를 지켜야 합니다. 아래 내용은 법안 중 일부에 해당하는 내용입니다.
- 타사 서비스와의 상호 운용성을 허용합니다.
- 경쟁사 및 유저가 게이트키퍼의 플랫폼 사용 시 생성된 데이터에 접근할 수 있도록 허용합니다.
- 향후 합병 및 인수 계획을 위원회에 통보해야 합니다.
- 광고 목적으로 유저를 추적하기 전에 명시적인 동의를 얻어야 합니다.
- 유저의 개인정보를 재사용할 수 없습니다.
- 자사 제품이나 서비스를 경쟁사에 비해 더 유리한 방식으로 순위를 매기는 것을 금지합니다.
조금 더 자세히 풀이하자면 유럽에 기반한 유저는 모바일 생태계에서 사전 설치된 앱이나 서비스를 쉽게 제거할 수 있게 되며, 구글 플레이 스토어(Google Play Store)나 애플 앱 스토어(Apple App Store) 이외에 새로운 앱 스토어가 등장할 수 있습니다.
### DMA가 가지는 의의와 불러올 효과
법안이 시행되면 대형 플랫폼 기업이 이끌고 있는 산업에 다양한 서비스 제공 업체가 여러 형태로 시장에 진입할 수 있게 됩니다. 또한 개인정보 보호 및 데이터 사용에 대한 규제가 강화되어 유저의 개인정보가 더욱 안전하게 보호될 것입니다.
## 게이트키퍼가 제시하는 대응책
[출처: 유럽연합 집행위원회]
게이트키퍼로 지정된 빅 테크 기업들은 법안 준수를 위해 새로운 변경 사항들과 함께 대응책을 제시하고 있습니다.
애플(Apple)은 [iOS, Safari, App Store에서의 변경 사항](https://www.apple.com/newsroom/2024/01/apple-announces-changes-to-ios-safari-and-the-app-store-in-the-european-union/)을 발표하였습니다. 유럽 연합에 앱을 출시하는 개발자의 경우 이제 iOS 앱을 다른 앱 마켓 플레이스에 배포할 수 있으며, iOS 앱 내에서 타사 결제 서비스 제공업체(Payment Service Provider, PSP)를 사용할 수 있습니다. 이러한 새로운 비즈니스 조건에 따라 앱 스토어 수수료를 다르게 지불하는 새로운 과금 체계도 발표되었습니다.
알파벳 (Alphabet)은 일부 구글 프로덕트에 유럽 사용자에게 구글의 프로덕트 및 서비스 간 유저 데이터 공유 동의 여부를 묻는 배너를 표시할 예정입니다. 기존에는 개인화된 광고 노출 등과 같은 목적을 위해 구글의 프로덕트 및 서비스 간 유저 데이터가 공유되었지만, 이제 DMA 준수 시점에 맞춰 EU 유저 동의 정책을 강화합니다. 또한 유럽 내 구글 검색 결과 페이지에 대한 테스트를 확대하여 비교 사이트와 같은 타사 공급업체를 위한 전용 공간을 포함하여 보다 자세한 결과를 표시할 예정이라고 발표했습니다. 그 외에도 안드로이드 휴대폰을 사용할 때 기본 검색 엔진 및 브라우저를 쉽게 전환할 수 있도록 제공하는 등 [다양한 분야에서의 변경 사항을 출시](https://blog.google/around-the-globe/google-europe/an-update-on-our-preparations-for-the-dma/)하였습니다.
> **구글의 EU 유저 동의 정책 업데이트에 대한 에어브릿지의 지원 방안**
> 광고주는 정보를 활용하기 전에 EEA(유럽 경제 지역) 유저에게 정보 활용에 대한 동의 여부에 대한 응답을 앱에서 프롬프트로 수집해야 합니다. 그리고 광고주는 앱에서 수집한 유저 응답을 구글에 전달해야 합니다. 2024년 2월 중으로 에어브릿지를 사용하는 고객사에게 구글의 신규 EU 유저 동의 정책을 지원하는 SDK와 API를 추가로 안내할 예정입니다. 유럽 지역에 광고하지 않거나 유럽 지역에 있는 유저가 사용하지 않는 앱의 광고주가 해야 하는 일은 없습니다.
> 👉[자세한 내용은 유저 가이드에서 확인해보세요.](https://help.airbridge.io/ko/guides/dma-google-eu-consent-policy)
## 끊임없는 변화에 대처하는 법
‘담장이 쳐진 정원(Walled Garden)’로 치부되던 영역에 빗장이 열리게 되는 것은 아닐지 업계에서는 촉각을 곤두세우고 앞으로의 방향성을 지켜보고 있습니다. 위에서 다룬 변화 외에도 다양한 분야에서의 변경 사항이 발표될 예정이며 이러한 대응책에 대한 유럽 연합 위원회의 반응 역시 아직은 알 수 없습니다. 지금은 새로운 소식에 귀 기울이며 다양한 변화를 열린 마음으로 받아들일 준비를 해야 할 때입니다.
에어브릿지는 광고 성과 측정을 돕는 파트너로서, 앱 서비스 및 마케터를 위해 계속해서 새로운 소식을 업데이트하고 변화를 지원하기 위한 기능을 선보이고자 준비하고 있습니다. 앞으로의 변화에 대비해 마케팅 전략을 어떻게 세워야 할지 고민되신다면 [에어브릿지의 마케팅 전문가들과 이야기 나눠보세요.](https://www.airbridge.io/ko/book-a-demo)
---
## Google Privacy Sandbox: 새로운 도전에 대응하는 방법
URL: https://www.airbridge.io/ko/blog/google-privacy-sandbox
Language: Korean
Published: 2024-01-24
2024년 1월 4일, 구글(Google)이 전체 1% 유저의 트래킹을 제한하며 서드 파티 쿠키 지원 중단의 신호탄을 알렸습니다. 2019년 처음 발표된 구글의 [Privacy Sandbox 프로젝트](https://privacysandbox.com/intl/ko_kr/)는 많은 업계 종사자가 함께 참여하여 개인정보를 보호하면서도 온라인상에서의 유저 경험을 향상시키기 위해 다방면으로 노력하고 있습니다. 머지않아 찾아올 다양한 변화를 예측하며 현재까지 구글이 발표한 그에 대한 대응책과 함께 우리는 어떻게 대비할 수 있을지에 대하여 알아보겠습니다.
## 어떤 것이 달라지나요?
구글의 Privacy Sandbox는 웹/ 안드로이드 두 가지 축으로 나뉘어 있습니다. 웹에서의 가장 큰 변화는 서드 파티 쿠키(Third-party cookie)의 지원 중단으로 은밀한 추적이 불가능해지며, 안드로이드 역시 유저 식별자로 사용되던 광고 ID (Google Advertising ID, GAID)를 지원 중단하며 보다 안전한 웹/ 앱 환경을 구축하게 됩니다.
다만 결론적으로 지금 당장은 각 비즈니스에서 체감할 변화가 크지 않을 겁니다. 특히 에어브릿지에서의 어트리뷰션 성과 측정에는 생각보다 큰 영향을 받지 않습니다. 광고 ID를 통한 기여 측정은 어려워졌지만 구글 플레이 인스톨 리퍼러 (Google Play Install Referrer)를 통한 인스톨은 여전히 기록되며, 딥링크 및 웹 UTM 파라미터 역시 사용이 가능합니다. 이를 기반으로 에어브릿지가 성과 측정을 위해 사용하는 cross-platform 자체 식별자 (Airbridge ID) 역시 여전히 사용이 가능합니다.
그렇다고 하더라도 어떤 부분이 영향을 받게 되는지, 어떻게 구체적으로 더 잘 대응해갈 수 있을지 아는 것은 중요합니다.
### 서드 파티 쿠키의 지원 중단은 어떤 변화를 의미할까요?
쿠키란 유저가 웹사이트를 방문할 때 저장하는 작은 데이터 조각으로, 서드 파티 쿠키의 경우 사용자가 방문한 웹사이트 소유자가 아닌 제3자 서비스에 의해 저장되는 데이터 모음을 의미합니다. 그동안 광고 업계에서는 해당 정보를 통해 여러 웹 브라우저를 걸쳐 활동하는 한 명의 유저를 추적하여 해당 유저에 맞춤광고를 보여줄 수 있었습니다. 웹사이트에서 한 번 검색한 제품이 다른 SNS나 플랫폼에서 채널 광고로 등장할 수 있었던 것은 이 서드 파티 쿠키가 있었기 때문입니다. 하지만 유저의 동의 없이 웹사이트 내에서의 사용자 행동이 수집되는 것에 대한 개인정보 침해 우려가 제기되었고 2024년 내로 지원이 중단될 예정입니다.
### 안드로이드의 광고 ID 지원 중단
안드로이드에서도 위와 유사한 형태로 유저 식별자인 광고 ID 지원을 중단할 계획입니다. 다만 앞서 말한 것과 같이 다행히 구글 플레이 인스톨 리퍼러나, 에어브릿지와 같은 MMP가 성과 측정을 위해 사용하는 자체 식별자도 여전히 사용할 수 있습니다. 그럼에도 불구하고 광고 ID는 오랜 시간 안드로이드 앱에서의 유저 추적에 사용되며 광고 성과를 확인하는 데에 중요한 역할을 하였기 때문에 이제는 다른 광고 성과 측정 방법에 대해 고민해야 할 때입니다.
## 구글이 제공하는 대안으로는…
[]
### 관련성 높은 콘텐츠 및 광고 표시 - Topics & Protected Audience API
기존에 개인 식별자를 통해 유저 행동을 파악하고 그 관심사를 유추할 수 있었다면, **Topics API**는 브라우저 혹은 앱 내의 최근 관심 기반 히스토리를 기반으로 광고 타겟팅을 통한 신규 유저 확보를 지원합니다. 공식적으로 지정된 약 400 여개의 관심사 주제, 즉 ‘토픽’ 중에서 머신러닝은 해당 사용자의 검색 및 방문 이력을 기반으로 유저에게 알맞은 토픽 카테고리를 부여합니다. 만약 유저가 여행과 관련된 사이트를 자주 방문했다면 유저의 토픽 목록에는 여행이 포함될 것입니다. 개인의 토픽은 3주 동안만 보관되어 그보다 오래된 토픽은 삭제됩니다. 물론 성별, 인종 등 개인 정보와 관련된 범주는 수집 대상에서 제외됩니다. 매체가 Topics API를 호출하면 타켓팅하고자 하는 유저에게 종속된 3가지 토픽이 공유됩니다. 특정 유저의 세분화된 행동 패턴을 분석하는 것이 아닌 일반적인 관심사만 공유되고, 5%의 확률로 허구의 토픽이 제공되어 유저의 익명성을 보장합니다. 그 외에도 Topics API는 유저에게 최대한의 제어권을 제공합니다. 즉 나에게 할당된 주제를 확인할 수 있으며 관심이 없는 주제에 대해서는 삭제하거나 아예 Topics API 기능을 비활성화할 수도 있습니다. 이러한 기능들을 통해 사용자의 개인정보를 보호하면서도 매체, 광고주가 계속해서 신규 유저 기반 디지털 광고 개인화 마케팅을 진행할 수 있게 도와줍니다.
또한 구글은 광고주들이 커스텀 오디언스를 활용하여 지속해서 리마케팅을 진행할 수 있도록 **Protected Audience API**를 제공합니다. 다만 개인정보 보호를 위해 지금과 같은 개인 식별자 혹은 서드 파티 쿠키가 사용되는 것이 아닌 디바이스 혹은 브라우저 자체에서 생성되어 보관되는 오디언스를 활용하게 됩니다. 광고주와 광고 매체(DSP)에서 캠페인에 대한 관심 그룹과 미리 일련의 동작 (장바구니에 담기 등)을 정의하면 유저가 해당 동작을 트리거 할 경우 Protected Audience API가 호출되어 해당 유저는 커스텀 오디언스에 포함됩니다. 향후 유저가 광고 사이트 (SSP 혹은 사이트 자체)나 퍼블리셔 앱을 방문할 경우 Protected Audience API가 호출되고 기존에 생성된 오디언스에 기반하여 광고 입찰을 실행하여 유저에게 가장 적절한 광고가 송출됩니다. 광고주나 애드 테크 플랫폼과 같은 제 3자가 아닌 디바이스 혹은 브라우저가 유저에 대한 관심분야에 대한 정보를 가지고 있고 외부 유출 없이 기기 내에서 광고 입찰이 진행되기 때문에 보다 안전하게 리마케팅을 진행할 수 있습니다.
### 디지털 광고 측정 - Attribution Reporting API
기존 광고 ID와 서드 파티 쿠키는 광고 성과 측정 측면에서도 중요한 역할을 해왔습니다. 해당 식별자를 통해 광고에 노출된 유저와 앱/ 웹 내에서 전환 이벤트를 발생시킨 유저를 매칭하여 어느 채널의 어떤 캠페인이 효과가 있었고 얼마만큼의 유저를 데리고 왔는지를 한눈에 확인할 수 있었기 때문입니다. 이러한 부분을 지원하기 위해 구글에서는 **Attribution Reporting API**로 광고 성과를 측정할 수 있도록 대안을 제시합니다.
Attribution Reporting API 가 작동하는 원리는 다음과 같습니다. 유저가 웹/ 앱 내의 광고를 보거나 클릭할 경우 Attribution Reporting API가 해당 터치포인트 이벤트를 저장합니다. 그 후 해당 유저가 앱을 설치하거나 웹 내에서 구매를 일으키는 등의 전환(Conversion)이 발생할 경우 해당 전환 이벤트 역시 저장합니다. 그리고 해당 전환 이전에 발생한 여러 개의 터치포인트 중 가장 마지막 터치포인트를 전환과 연결하여 기여를 산정합니다. 기존 기여 방식과 유사하게 보이지만 다른 점은 터치포인트 저장 및 기여까지의 프로세스가 모두 **유저의 디바이스 혹은 브라우저 내**에서 집행된다는 점입니다.
디바이스 (혹은 브라우저) 내에서 기여가 산정되면 Attribution Reporting API는 이벤트 레벨 리포트, 그리고 집계 리포트의 2가지 형태의 리포트를 제공합니다. 해당 리포트들은 개인정보 보호를 위하여 지연이 발생하며 데이터에 일정 비율의 노이즈가 포함되어 발송됩니다.
1. **이벤트 레벨 리포트**
이벤트 레벨 리포트는 각 전환 이벤트가 발생할 때마다 광고 최적화를 목적으로 제공되는 리포트입니다. 해당 리포트에서는 하기 세 가지의 정보를 통해 ‘어떤 앱에서 어느 캠페인에 의해 어떤 전환 이벤트가 발생하였는지’ 를 확인할 수 있습니다.
- Destination - 전환 이벤트가 발생한 광고주 앱, 혹은 모바일 웹 주소
- Attribution Source ID - 광고 채널 등 캠페인 정보를 지닌 ID (채널 및 캠페인 정보)
- Trigger Type - 앱 설치, 구매와 같은 전환 이벤트 유형
[]
1. **집계 리포트**
집계 리포트는 이벤트 레벨 리포트에 비해 더 다양한 정보를 받아볼 수 있어 여러 관점에서의 분석을 가능하게 합니다. 집계 키(Aggregation Key) 라고 불리는 광고 캠페인 및 전환 관련 기준을 매체 그리고 MMP 측에서 사전에 정의할 수 있으며 해당 기준을 토대로 광고 성과 분석이 가능합니다.
집계 리포트를 통해 확인할 수 있는 데이터의 구성은 크게 2가지 입니다.
- Aggregation Key - 캠페인, 광고 그룹, 소재, 타게팅 지역 등과 같은 광고에 대한 정보
- Aggregation Value - 구매 수, 구매액과 같은 전환 이벤트에 해당하는 집계된 정보
[]
예를 들어 매체 측에서 광고가 노출되거나 클릭 발생 시점에 측정 가능한 캠페인, 광고 그룹, 소재, 타게팅 지역과 같은 집계 키 항목을 설정하고, MMP 측에서는 전환이 발생하는 시점에 수집할 수 있는 상품 카테고리, 디바이스 정보, 앱 버전과 같은 전환 관련 항목들을 설정하였다고 가정해 봅니다. 그 경우 집계 리포트에서는 ‘어떤 앱에서 어느 캠페인에 의해 구매가 총 15건이 발생하였고, 전체 구매 금액은 250만 원이다. 그중 절반의 구매는 서울지역에서 발생했다' 와 같은 형태로 성과를 확인할 수 있습니다.
다만 리포트 이름에서도 알 수 있는 것과 같이 이벤트 별로 리포트를 받을 수 있는 것은 아니며 동일한 속성을 가진 이벤트가 일정 수준 이상 쌓인 후에 리포트로 받아볼 수 있습니다. 따라서 최대한 큰 모수를 만들기 위해 Aggregation Key 설정값을 최대한 일반적인 항목들로 구성하는 것을 권장합니다.
## 마케팅, 이제 어떻게 준비할 수 있을까요?
우리는 이미 애플의 ATT (App Tracking Transparency) 프레임워크를 통해 iOS 광고 식별자의 수집을 중단한 적이 있습니다. 해당 경험으로 새로운 변화에 적응하는 것은 결코 쉽지 않지만 도전을 통해 또 다른 기회가 창출될 수 있음을 배웠습니다. 구글 프라이버시 샌드박스의 경우, 정책이 실제로 적용되기 전에 이를 대안할 수 있는 제안 사항들이 공개되었고 업계의 각축에서 베타 테스트 피드백과 함께 프로젝트를 개선해 나가고 있습니다. 사전에 준비할 수 있는 시간이 확보되어 보다 안정적인 변화의 흐름을 만들 수 있을 것으로 예상됩니다.
### 광고주 입장에서 사전에 숙지해야 하는 사항들은?
안드로이드에서 광고 ID 지원이 중단될 경우 광고 클릭과 앱 내 전환을 매칭하던 식별자가 사라져 기존 대비 인스톨 및 전환 수치가 적게 잡힐 수 있습니다. 다만 앞서 말한 것과 같이 Playstore Install Referrer를 통해 매칭된 어트리뷰션은 여전히 기록됩니다.
만일 고객사에서 현재 광고 ID 기준으로 내부 유저 데이터를 구축하고 있었다면 다음과 같은 작업이 필요합니다.
1. 현재 내부 데이터베이스에서 쌓고 있는 유저 데이터의 식별자가 어떤 것인지 확인합니다
2. 만일 광고 ID (GAID) 기준으로 유저를 식별하고 있다면 해당 식별자를 대체할 수 있는 다른 식별자가 있을지 확인합니다 (예: 이메일 주소와 같은 퍼스트 파티 데이터, MMP가 제공하는 ID)
3. 매체, MMP와 대체 방법을 상의하여 적용합니다
크롬에서 서드 파티 쿠키 지원이 종료된다면 서브 도메인 어트리뷰션 결과가 공유되지 않을 수 있습니다. 예를 들어 고객사에서 여러 서브 도메인을 사용하고 있고, 각 사이트에서의 기여 결과를 공유하고 있었다면 이는 더 이상 지원되지 않습니다. ( a.ab180.co 사이트에서의 기여 결과가 발생할 경우 b.ab180.co 와 연결되지 않음)
## 에어브릿지는 이렇게 준비하고 있습니다
에어브릿지는 현재 구글의 AAP (App Atttribution Partner) 파트너로서 Google Ads 팀과 함께 Attribution Reporting API 관련 베타 테스트를 진행 중에 있습니다. 해당 테스트와 더불어 Privacy Sandbox의 다양한 기능들을 내부적으로 확인하고 있습니다.
그 외에도 유저의 개인정보 보호를 위하여 대형 매체들에서 새롭게 지원하는 (예: 구글 - [구글 클릭 식별자 (GCLID) 와 gBraid](https://support.google.com/analytics/answer/11367152?hl=en&sjid=7903611105448176055-AP), 메타 - [Aggregated Event Measurement (AEM)](https://www.facebook.com/business/help/721422165168355?id=1877298665783613%2F)) Attribution 방식을 지원하기 위해 작업 중에 있으며 MMP로서 업계에 어떤 도움을 줄 수 있을지에 대해 다방면으로 고심하고 있습니다.
에어브릿지는 다가오는 변화에 대비하여 업계 전반에 가장 빠르게 맞춤화된 서비스를 지원하기 위하여 노력하고 있습니다. 구글 프라이버시 샌드박스의 최신 소식과 앞으로의 마케팅 전략에 관한 조언이 필요하시다면 [에어브릿지의 마케팅 전문가들과 이야기 나눠 보세요.](https://www.airbridge.io/ko/book-a-demo)
---
## 2024년 개인정보 보호 변화, 마케터가 알아야 할 모든 것
URL: https://www.airbridge.io/ko/blog/2024-privacy-landscape
Language: Korean
Published: 2024-01-17
최근 몇 년 간 강화되고 있는 개인정보 보호 트렌드는 2024년에도 지속될 것이며 이에 따라 광고 업계에도 계속된 변화가 예상됩니다. 대표적으로는 2023년이 애플(Apple)의 SKAdNetwork를 이해하고 적응하는 시기였다면, 2024년에 마주할 가장 큰 변화는 구글(Google)의 Chrome 서드 파티 쿠키 지원 종료입니다.
개인정보 보호 흐름을 예측하긴 어렵지만 마케터들은 어떻게 광고의 효율을 높이고 성과를 잘 측정할 수 있을지 고민하고 있습니다. 이 과정에서 변화를 정확하게 인지하는 것만으로도 고민 해결의 실마리를 얻을 수 있습니다. 이번 포스트를 통해 구글 및 애플에서 공개한 2024년 개인정보 보호 관련 변화를 알아보고 앞으로 마주하게 될 변화를 효과적으로 대응해 보세요.
> ❗ 주의: 본 블로그 포스트는 2024년 1월 2일에 업데이트된 내용입니다. 구글 및 애플의 최신 발표에 따라 수정해야 하는 일부 정보가 포함되어 있을 수 있습니다.
## 구글 프라이버시 샌드박스와 서드 파티 쿠키 지원 종료
구글은 2020년, 웹 프라이버시 강화를 위한 새로운 노력의 일환으로 유저들에게 안전하고 개인정보 보호에 중점을 둔 환경을 제공하고자[ 프라이버시 샌드박스(Privacy Sandbox)](https://privacysandbox.com/intl/ko_kr/)와 Chrome의 서드 파티 쿠키 지원 종료를 선언했습니다. 그리고 2022년, Android에도 프라이버시 샌드박스를 적용하며 구글 광고 ID(Google Advertising ID, 이하 GAID) 제공을 중단한다고 발표했습니다.
구글의 프라이버시 샌드박스는 유저의 개인정보를 보호하면서도 온라인 광고 산업을 지원하기 위한 노력의 일환이라고 할 수 있습니다. 기존에 마케팅 성과 측정을 위해 사용되어 온 쿠키와 GAID 대신 광고주와 퍼블리셔 간의 광고 효과를 측정하고 개선하기 위한 새로운 방식을 제안합니다. 이를 통해 유저는 더 나은 개인정보 보호를 누릴 수 있으며 동시에 광고 산업도 효율적으로 작동할 수 있습니다. 다만 서드 파티 쿠키 및 GAID 제공 중단은 광고 업계에 큰 영향을 미치기 때문에 원활하게 새로운 방식을 적용할 수 있도록 상당한 유예 기간을 제공하고 있습니다. 그리고 **2024년 1분기에 무작위로 선정된 유저 1%에게 트래킹 중단 기능을 제공하여 서드 파티 쿠키의 종료를 위한 실제 테스트를 시작합니다.**
이외에도 최신 업데이트 소식은 프라이버시 샌드박스 사이트 및 [Chrome for Developers](https://developer.chrome.com/blog?hl=ko), [Android Developers](https://developer.android.com/?hl=ko) 블로그에서 확인할 수 있습니다.
> 💡 용어 더 자세히 알아보기
- **서드 파티 쿠키**는 광고주가 유저가 어떤 웹사이트를 방문했는지 트래킹하고 유저의 행동을 기록하기 위해 사용되는 작은 텍스트 파일입니다. 광고 타겟팅과 광고 성과 측정에 사용됩니다.
- **구글 광고 ID(GAID)**는 모바일 앱에서 광고 성과를 측정하기 위해 유저를 식별하기 위해 사용되는 고유 식별자입니다. 유저의 광고 타겟팅 및 성과 측정을 위해 사용됩니다.
### 2024년 1분기, 서드 파티 쿠키 종료를 위한 테스트 시작
이미 대부분의 웹 브라우저들의 경우 서드 파티 쿠키를 차단하고 있었으며, 이제 Chrome도 그 대열에 본격적으로 합류합니다. 얼마 전 [프라이버시 샌드박스의 공식 발표에 따르면](https://developers.google.com/privacy-sandbox/blog/cookie-countdown-2023oct?hl=ko), 서드 파티 쿠키를 사용하는 사이트들에 대해 원활한 테스트를 제공하고자 2024년 1분기부터 무작위로 선정된 Chrome 유저 1%에 대해 트래킹 중단(Tracking Protection) 기능을 제공합니다. 그리고 2024년 3분기부터는 100%까지 확대할 예정입니다. 이 기간 동안 구글에서 안내하는 대로 단계적으로 서드 파티 쿠키의 지원 중단에 대비할 수 있으며, 관련 Web API로 마이그레이션 등을 테스트해 볼 수 있습니다.
### GAID 제공 중단으로 인한 변화를 준비
또 다른 프라이버시 샌드박스의 큰 축인 Andriod의 GAID도 제공 중단을 앞두고 있습니다. 정확한 지원 중단 시점 및 2024년 로드맵을 밝히진 않았지만, 2023년 2월부터 Android 유저를 대상으로 테스트를 시작했으며, 2023년 한 해 동안 개발자 프리뷰 및 안정적인 API 출시를 반복했습니다.
GAID가 사라지게 되면 지금 사용하는 모바일 광고 성과 측정 방법에 중대한 영향을 미칩니다. 모바일 광고 성과 측정을 위해 광고로 유입된 유저를 식별할 수 있는 식별자가 사라지기 때문입니다. 애플은 앱 추적 투명성(ATT) 정책에 따라 개인정보 수집을 허용한 광고 식별자(IDentity For Advertisers, 이하 IDFA)를 광고주에게 제공합니다. 반면, 구글은 개인정보 수집 허용 여부와 관계없이 GAID를 제공하지 않는다는 점이 다릅니다.
다만, GAID 제공 중단이 마케팅 성과 측정 자체가 불가능해짐을 의미하는 것은 아닙니다. 구글은 다양한 마케팅 캠페인 운영을 지원하기 위해 추가적인 마케팅 정보를 3개의 API(Application Programming Interface)로 제공합니다. 그리고 안정적인 운영을 위해 다양한 애드테크(AdTech) 기업들과의 협업을 이어가고 있습니다. 에어브릿지도 [구글의 앱 어트리뷰션 파트너(AAP, App Attribution Partner)로서](https://www.airbridge.io/ko/blog/airbridge-google-aap) 지속적인 협업을 이어가고 있습니다. 또한, 국내에서 가장 빠르게 관련 기능에 대한 테스팅 및 정확한 정보를 제공할 수 있습니다. 구글이 제공하는 주요 API와 지원하는 마케팅 활동은 아래와 같습니다.
[]
## 애플의 SKAdNetwork와 Privacy Manifest
애플은 iOS 14.5 업데이트 이후 [앱 추적 투명성](https://developer.apple.com/app-store/user-privacy-and-data-use/)(App Tracking Transparency, 이하 ATT) 정책을 도입했습니다. 앱 설치 시 IDFA 수집 동의 여부를 Opt-in 방식으로 수집하도록 의무화하며 모바일 광고 시장에 엄청난 변화를 일으켰습니다. 그리고 IDFA 수집이 제한이 따르는 상황에서 개인정보 보호를 준수하며 앱 마케팅 캠페인 성과를 측정할 수 있는 새로운 프레임워크로 SKAdNetwork(이하 SKAN)을 도입했습니다. 2022년 말 SKAN 4.0을 공개하였으며, 매년 개최되는 세계개발자대회(WWDC, Worldwide Developers Conference)를 통해 주요 업데이트를 발표하고 있습니다.
2024년에 iOS에서 마주할 주요한 변화는 지난 WWDC 2023에서 공개된 SDK를 대상으로 한 새로운 **개인정보 보호 매니페스트 및 SDK 서명(SDK Signature)**입니다. 현재까지 공개된 SKAN 및 주요 개인정보 보호 관련 업데이트에 대해 에어브릿지는 마케터들에게 편리한 관련 기능들을 제공하고 있으며, 앞으로도 가장 빠르게 정보와 기능을 전해갈 예정입니다.
> 💡 SKAdNetwork(SKAN)이란?
SKAN은 iOS 기기에서 유저 혹은 디바이스 정보 없이도 iOS 앱 설치 캠페인 성과를 측정할 수 있는 프레임워크입니다. 광고주가 개별 유저를 식별하는 데 사용하는 정보를 최소화하며, 대신 익명화된 정보만을 전달합니다. 즉 유저의 개인정보는 보호하면서도 광고주에게 광고 캠페인의 효과를 측정할 수 있는 방법을 제공합니다. 현재 SKAN 4.0까지 공개되었으며, 계속된 업데이트를 통해 마케터들이 주목해야 할 기능들을 발표하고 있습니다.
💡 SKAN에 대해 더 자세히알아보기
- [앱 마케터가 SKAdNetwork 4.0 (SKAN 4.0)에 주목해야 하는 이유
](https://www.airbridge.io/ko/blog/exploring-the-marketing-implications-of-skadnetwork-4)- 애플 SKAN 4.0을 제대로 활용하고 싶은 마케터가 꼭 알아야 하는 3가지
- SKAN 중복 기여 판별 기능으로 더 정확한 광고 성과를 측정하세요
### SKAN 중복 기여 판별 기능으로 보다 정확한 마케팅 성과 측정하기
에어브릿지 분석에 따르면 2023년 기준 국내외 iOS 유저 중 평균적으로 30%만 IDFA 수집을 허용하였습니다. 이로 인해 SKAN 캠페인 집행에 대한 필요성이 점점 확대되고 있으며, 실제로 에어브릿지를 사용하고 있는 앱 서비스 중 54% 이상이 SKAN 캠페인을 집행하고 데이터를 확인하고 있습니다.
다만, SKAN은 LTA(Last Touch Attribution) 등과 같은 기존의 광고 성과 측정 모델 보다 낯설게 느껴지기도 하며 데이터 처리에 어려움을 겪을 수 있습니다. 이에 에어브릿지는 SKAN에 대한 상세한 설명과 전환값 설정 방법에 대한 안내를 제공하고 있습니다. 또한, 발생한 앱 설치 성과에 대해 SKAN과 LTA 모델에서 중복으로 집계된 앱 설치 성과를 확인하는 ‘SKAN 중복 기여 판별’ 기능을 제공합니다. 이 기능을 활용하면 마케팅이 기여한 앱 설치를 더 많이 측정하고 CPI 비용은 낮추며 보다 정확한 광고 성과를 확인할 수 있습니다.
### 최신 앱 스토어 개인정보 보호 업데이트
지난 WWDC 2023에서 애플은 SDK를 대상으로 한 새로운 개인정보 보호 매니페스트 및 SDK 서명(SDK Signature)을 소개했습니다. 그리고 2023년 12월에는[ App Store 앱 제출을 위한 개인정보 보호 관련 업데이트](https://developer.apple.com/kr/news/?id=r1henawx)를 발표했습니다. “2024년 봄부터 제출하는 앱 또는 앱 업데이트에 App Store에 게시된 앱에 흔히 사용되는 타사 SDK가 추가되는 경우 SDK에 대한 개인정보 보호 매니페스트를 함께 제출해야 합니다.”라고 명시했습니다.
여기서 '[개인정보 보호 매니페스트](https://developer.apple.com/documentation/bundleresources/privacy_manifest_files/describing_data_use_in_privacy_manifests)'는 타사 코드가 앱에서 개인정보를 어떻게 처리하는지 하나의 표준 형식으로 요약한 파일을 의미합니다. 이를 통해 개발자들은 사용하는 모든 SDK들이 각각 개인정보를 어떻게 취급하는지 개요를 보다 쉽게 만들 수 있습니다. 그리고 ‘SDK용 서명’은 새로운 버전의 타사 SDK를 사용할 때 해당 버전이 동일한 개발자가 서명한 버전의 SDK인지 확인하기 위해 사용됩니다.
각 앱들이 타사 SDK를 사용하는 경우 해당 SDK로 인해 앱에 포함되는 모든 코드에 대한 책임이 앱 개발자에게 있기 때문에, 이번 업데이트는 각 SDK의 데이터 수집 및 사용 사례를 파악할 수 있도록 돕습니다. 보다 자세한 내용 및 업데이트되는 내용은[ Apple의 뉴스 및 업데이트](https://developer.apple.com/kr/news/)에서 확인할 수 있습니다.
## 개인정보를 보호하는 정확한 마케팅 측정은 에어브릿지와 함께
시장에서 앞서 나가기 위해서는 변화를 이해하고 이에 대비한 적절한 전략을 세워야 합니다. 이번 포스트를 통해 주요 모바일 앱 플랫폼을 중심으로 변화할 2024년의 개인정보 보호 흐름에 대해 알아보았습니다. 다음 스텝으로는 개인정보를 보호하며 정확한 마케팅 성과를 측정하는 방법을 찾아 활용해야 합니다.
1. 에어브릿지는 구글 프라이버시 샌드박스 및 SKAN 업데이트에 따른 변화에 대해 각 광고 채널들과의 지속적인 협업을 통해 최신의 기술을 제공하고자 노력하고 있습니다.
2. 또한,[ SKAN 3.0과 SKAN 4.0 측정 성과](https://help.airbridge.io/ko/guides/skadnetwork-overview)를 확인하기 위한 편리한 설정 방법과 대시보드를 제공합니다. 하나의 에어브릿지 리포트 화면에서 Android와 iOS(SKAN 포함) 데이터를 한눈에 확인하여 전반적인 마케팅 성과에 대한 가시성을 높여줄 수 있습니다.
3. 개인정보 보호 흐름에 대응하는 마케팅 믹스 모델링(Marketing Mix Modeling, MMM)과 같은 마케팅 성과 측정 방법을 계속해서 고민하고 제안하고 있습니다.
에어브릿지는 개인정보 보호 흐름에 대응하는 마케팅 성과 측정 방법을 가장 빠르게 제안하기 위해 나아가고 있습니다. 앞으로도 마케터 여러분들을 위해 가장 빠르고 정확하게 정보를 전하고 업데이트해가겠습니다. 개인정보를 보호하는 정확한 마케팅 측정 방법을 찾고 있다면[ 에어브릿지의 마케팅 전문가들과 이야기 나눠 보세요.](https://www.airbridge.io/ko/book-a-demo)
---
## 웹과 앱 통합 어트리뷰션: 마케팅 성과 측정의 핵심
URL: https://www.airbridge.io/ko/blog/web-app-attribution
Language: Korean
Published: 2024-01-08
모바일 기기의 등장으로 마케팅 시장에도 혁신적인 변화가 생겼습니다. 데스크톱에서만 이루어지던 모든 활동이 모바일 기기에서도 가능해졌고, 가볍고 편리한 모바일 기기는 사람들의 필수 생활용품으로 자리 잡아 시장을 지배한 지 오래입니다. 새로운 기기의 등장과 함께 유저들은 플랫폼과 웹, 앱 지면을 넘나들며 활동합니다. 이 복잡한 시대를 맞이한 디지털 마케터의 가장 중요한 업무 중 하나는 정확한 성과 측정입니다. 집행한 광고가 유의미했는지, 어떤 채널이 유저를 가장 많이 유입시켰는지에 대한 판단이 필요한 가운데 앱과 웹 지면에서는 서로 다른 방법으로 성과를 측정하기 때문에 정확한 측정에 대한 마케터들의 고민이 더욱 가중될 수 밖에 없습니다. 이러한 어려움을 해결해 줄 수 있는 앱+웹 통합 어트리뷰션 분석이 지금 왜 더욱 필요한지에 대해 이야기합니다.
## 웹 그리고 앱 어트리뷰션의 현재
마케팅에서의 어트리뷰션이란, 마케팅 캠페인이라는 ‘원인’ 과 앱 설치나 인앱 전환과 같은 성과의 ‘결과’를 매칭하여 ‘인과 관계’를 분석하는 것을 의미합니다. 앱이 나오기 전 인터넷 세상은 이 인과 관계를 비교적 간단하게 분석할 수 있었습니다. 유저 여정의 시작과 끝은 웹이었고 플랫폼도 데스크톱 하나였습니다. 웹사이트의 쿠키를 고유 식별자로 사용하고, 브라우저 내에서 링크를 통해 이동 시 기록되는 HTTP Referer 그리고 커스텀 하게 캠페인 정보를 추가할 수 있는 UTM 파라미터를 사용하면 어느 웹사이트에서 어떤 키워드를 통해 유저가 들어왔는지 손쉽게 파악하여 캠페인을 최적화 할 수 있었습니다.
웹 어트리뷰션에서 쿠키는 브라우저마다 다르게 발급되고 만료 기간도 있어 온전한 아이덴티티는 아니지만 하나의 유저로 식별하는 데 사용됩니다. 유저가 식별되면 HTTP Referer 값에 남아 있는 A 사이트 URL 을 확인하여 이 유저가 우리 웹사이트에 오기 전에 해당 A 사이트에 방문했는지 알 수 있고, 어디에서 우리 웹사이트가 홍보가 되고 있는지도 파악할 수 있습니다. 다만 이 HTTP Referer는 조작이 가능하고 유저가 리다이렉트 되는 과정에서 유실이 될 수 있습니다. 또한 마케터가 임의로 추가적인 정보를 태그할 수도 없습니다. 이때 UTM 파라미터를 사용하면 HTTP Referer 의 이러한 제약 사항을 보완할 수 있으며 URL 뒤에 5개의 추가 정보를 커스텀하게 포함할 수 있습니다. 해당 5개의 파라미터를 통해 받은 데이터를 사용하여 캠페인 최적화를 진행해 성과를 극대화할 수 있습니다.
그러던 와중 스마트폰, 태블릿 등 다양한 기기가 출시되기 시작하며 앱이라는 존재가 등장하게 되었습니다. 데스크톱 웹과 함께 모바일에서도 웹 브라우저가 사용되어 유저 여정이 한층 더 복잡한 양상을 띄게 되었고, 앱스토어라는 개념과 더불어 앱 설치 및 전환에 대한 분석 니즈가 생기면서 어트리뷰션 측면 역시 새로운 국면을 맞이하게 되었습니다. 링크만 누르면 바로 웹사이트로 유저가 이동하여 명확하게 경로를 알 수 있었던 웹 어트리뷰션과 달리 앱의 경우, 링크 클릭 시에 유저는 앱스토어로 넘어가며 그 안에서도 앱 설치 버튼을 클릭해야 비로소 앱이 설치되기 때문입니다. 물론 유저가 설치된 앱을 실행하기 전까지 유저를 획득했다고 할 수도 없습니다. 직접 로그 데이터를 확인할 수 있는 웹사이트와 다르게 앱스토어는 우리의 소유가 아니기에 유저의 광고 클릭부터 앱 오픈까지의 과정을 추적하는데 단절이 발생하게 되었습니다.
[__wf_reserved_decorative]
위와 같은 구간 단절 해소를 위해 앱 어트리뷰션 툴이 업계에 도입되었습니다. 앱 어트리뷰션 툴이 제공하는 링크로 광고를 집행하면 해당 링크를 클릭한 유저의 앱스토어 이동 및 앱을 오픈하는 시점을 연결할 수 있습니다. 앱 어트리뷰션의 경우 광고 식별자 (GAID, IDFA)가 기기 별로 제공되며 다양한 앱을 실행하고 웹 브라우저를 오픈하여도 하나의 유저로 인식합니다. 이 광고 식별자를 사용한 ID 매칭과 더불어 Google Play Referrer, Deeplink 매칭 등의 확정적 방법(약 100% 매칭)과 핑거 프린팅(사용자의 웹 브라우저 정보, 예: IP, OS, 기기명 등을 활용한 조합과 앱 SDK에서 수집된 정보를 매칭 시키는 방법)의 확률적 방법을 활용하여 앱 내외에서의 유저 여정을 추적할 수 있게 되었습니다.
다만 디바이스 당 부여되는 광고 식별자와 관련하여 유저의 개인정보 침해 우려에 대한 목소리가 있었고 2018년 GDPR(유럽 연합 일반 데이터 보호 규칙)이 시행되면서 업계에서는 어떻게 하면 개인정보를 보호하면서 보다 정교한 추적이 가능할 것인지에 대한 고민을 하기 시작했습니다. 이러한 움직임의 시작으로 Apple에서는 광고 식별자 대신 SKAdNetwork(SKAN)이라고 하는 새로운 어트리뷰션 프레임워크를 출시하였으며 유저의 개인정보를 보호하면서도 앱과 웹 지면의 크로스 채널 광고 캠페인의 성과 역시 측정할 수 있게 되었습니다.
👉 [SKAdNetwork(SKAN) 프레임워크에 대한 자세한 내용을 가이드에서 확인해보세요.](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)
## 웹과 앱을 오가는 유저의 여정 한눈에 확인하기
### 웹과 앱의 어트리뷰션 통합 측정하기
이렇게 웹과 앱의 어트리뷰션 로직이 상이하고 지면이 다르다 보니 각각의 환경에서 어트리뷰션을 측정하는 것이 일반적이었습니다. 웹사이트와 앱 모두를 관리하는 마케터의 경우 하나의 캠페인을 집행하여도 여러 개의 분석 툴을 사용하여 구간 별 광고 성과를 측정하기에 이르렀습니다. 이 경우, 웹 광고 캠페인을 통해 웹으로 랜딩 된 유저가 나중에 앱을 설치하여 앱 내 구매 이벤트를 일으켰을 때 동일 유저임을 판별할 수 없을 뿐만 아니라 캠페인 웹 지면이 아닌 다른 브라우저나, 앱에서의 전환이 발생하였을 경우 오가닉 혹은 기여되지 않음으로 처리될 수 있어 캠페인 성과가 저조하다고 느낄 수도 있습니다. 마케팅 성과를 완전하게 파악하고 싶다면 앱과 웹을 통합하여 어트리뷰션을 측정할 수 있는 분석 툴을 사용해야 합니다.
> 에어브릿지에서는 Web SDK와 App SDK (Software Development Kit)를 모두 제공하여 각 플랫폼의 데이터를 수집합니다. Airbridge Identity Resolution 엔진을 통해 이 데이터를 결합하여 Airbridge ID를 기반으로 하나의 대시보드에서 앱과 웹의 어트리뷰션 성과를 한눈에 확인할 수 있습니다.
### 웹투앱 경험 구축하고 성과도 측정하기
[__wf_reserved_decorative]
웹과 앱 마케팅에는 각각의 장점이 있습니다. 웹사이트 마케팅은 트래픽이 더 많고 지면도 다양한데다가 앱 마케팅보다 비용도 저렴합니다. 앱스토어 화면의 제한된 지면을 대신해 웹 플랫폼을 활용한 풍부한 콘텐츠 및 경험을 전달하는 마케팅 전략은 신규 유저 획득을 위해 필수적입니다. 다만 궁극적으로 모바일 앱에서의 구매 전환율과 매출이 웹에 비해 높기 때문에 앱 사용을 권장하여 진성 유저로 만들기 위한 노력 역시 필요합니다. 이를 위해 웹에서 CTA (Call-to-Action) 버튼을 통해 유저들에게 앱 설치 혹은 특정 앱 내 페이지 오픈을 유도합니다.
일부 에어브릿지 고객사에서는 유저가 어떤 링크를 클릭했고, 누가 처음 링크를 공유했는지 등 수많은 커스텀 데이터를 가지는 컨텍스트 딥링크 기능을 적극적으로 활용하고 있습니다. 유저가 웹에서 보던 콘텐츠와 연결되는 앱 내 페이지로 유저를 이동시키거나, 개인화된 환영 메세지 그리고 자동 프로모션 코드 등을 웹 페이지에 더하여 유저에게 맞춤 경험을 제공할 수 있습니다. 세밀한 웹투앱 경험과 어트리뷰션 분석 환경을 구축하여 보다 효과적인 마케팅 전략을 세워보세요.
> 에어브릿지는 원활하고 심리스한 플랫폼 간 이동을 위해 모바일 웹 내에서 앱 설치 및 앱 내 페이지 전환을 유도하는 스마트 앱 배너(Smart App Banner) 기능을 제공합니다. 스마트 앱 배너를 통해 웹에서 획득한 유저의 앱 설치와 앱 내 전환율을 개선하여 효과적인 웹투앱 어트리뷰션 환경을 구축할 수 있습니다.
> 👉 [웹투앱 전략 백서에서 더 자세히 알아보기](https://www.airbridge.io/ko/library/web-to-app-conversion-whitepaper)
## 이 시대의 필수 마케팅 전략: 웹+앱 통합 어트리뷰션
시대와 업계의 변화에 따라 마케팅 전략 및 캠페인의 성과 분석 방법 역시 지속해서 변화하고 있습니다. 한 명의 사용자가 여러 기기 및 플랫폼을 넘나들며 사용하는 시대가 도래하면서 이러한 유저 여정을 하나의 솔루션에서 확인할 수 있는 방법은 없을지에 대한 니즈가 끊임없이 생겨났고 에어브릿지는 이러한 문제에 대해 개인 정보를 보호하며 마케터의 고민을 해결하기 위해 노력해왔습니다. 웹과 앱 그리고 웹투앱 어트리뷰션까지 여러 형태의 유저 여정을 한눈에 분석할 수 있는 환경에서 구간 단절 없는 정확한 마케팅 성과를 측정해 보세요.
---
## SKAN 설치 중복 카운팅 방지 완벽 가이드
URL: https://www.airbridge.io/ko/blog/skadnetwork-4-attribution-overlap
Language: Korean
Published: 2024-01-08
iOS 앱 마케팅을 집행하고 있는 마케터라면 정확한 성과 측정에 대한 고민을 한번쯤 해봤을 것입니다. 애플이 iOS 14.5 업데이트로 앱 추적 투명성(App Tracking Transparency, ATT) 정책을 도입한 2021년 4월 이후, 광고 식별자(IDFA) 수집이 어려워졌기 때문입니다.
마케터의 어려움은 수치에서도 드러납니다. 에어브릿지 분석에 따르면 2023년 기준 국내외 iOS 유저 중 평균적으로 30%만 IDFA 수집을 허용했습니다. 10명 중에 7명의 광고 성과를 정확하게 측정하기 어려운 셈입니다. 마케터가 노력으로 피운 소중한 광고 성과, 에어브릿지는 iOS에서 놓치지 않도록 중복 기여 판별(Attribution Overlap) 기능을 제공하고 있습니다.
중복 기여 판별은 여러 어트리뷰션 모델로 광고 성과가 중복해서 기여됐는지 판별하는 기능입니다. 에어브릿지는 SKAdNetwork(SKAN) 4.0 대시보드부터 이 기능을 지원하고 있습니다. SKAN 4.0에서 광고 성과를 더 정확하게 측정하고 싶은 마케터라면 중복 기여 판별에 관한 내용을 확인해 보세요.
> **💡SKAN 3.0에서도 중복 기여 판별을 사용할 수 있어요
> **
> 중복 기여 판별은 에어브릿지에서 SKAN 전환값 설정 페이지를 4.0 버전으로 변경해야 사용할 수 있는 기능입니다. SKAN 3.0만 지원하는 광고 채널에서도 중복 기여 판별로 더 정확한 광고 성과를 확인할 수 있습니다.
## 중복 기여 판별이 필요한 이유
중복 기여 판별을 활용하면 SKAN으로 기여된 이벤트가 에어브릿지 어트리뷰션 모델로 중복 기여되었는지를 판별할 수 있습니다. 중복 기여 여부를 판별하는 기준인 SKAN과 에어브릿지 어트리뷰션 모델은 광고 성과를 측정하는 방식이 다릅니다. 두 어트리뷰션 모델은 기여 주체와 IDFA 사용 여부 등에 차이가 있습니다.
[__wf_reserved_decorative]
따라서 두 어트리뷰션 모델을 동시에 활용해서 광고 성과를 측정하면 양쪽의 기여 기준을 모두 충족하는 이벤트가 발생할 수 있습니다. 해당 이벤트는 각 어트리뷰션 모델로 측정한 광고 성과에 중복으로 집계됩니다. 실제로 이벤트는 1번만 발생했지만, 리포트에는 이벤트가 2번 발생한 것처럼 집계됩니다.
이런 과도한 광고 성과 집계를 방지하기 위해서는 중복해서 집계된 이벤트를 제거해야 합니다. 중복을 제거하면 두 어트리뷰션 모델로 측정한 광고 성과를 더 정확하게 집계할 수 있습니다.
[__wf_reserved_decorative]
IDFA 수집 여부 등 에어브릿지 어트리뷰션 모델의 여러 기여 기준에 부합하면 SKAN으로 기여된 이벤트는 중복 기여로 처리됩니다. 중복 기여된 이벤트는 SKAN을 활용하지 않아도 에어브릿지 어트리뷰션 모델로 측정할 수 있는 광고 성과입니다. 중복 기여되지 않은 이벤트는 오직 SKAN으로 측정할 수 있는 광고 성과입니다. 해당 이벤트 수는 SKAN으로 기여된 전체 이벤트 수에서 에어브릿지 어트리뷰션 모델로 중복 기여된 이벤트 수를 제외해서 계산합니다.
## 중복 기여 판별, 이렇게 활용하세요
CPI(Cost per Install) 또는 CPA(Cost per Action)는 마케터에게 가장 중요한 지표 중 하나입니다. 중복 기여 판별을 활용하면 CPI, CPA를 보정할 수 있습니다. 기존 앱 마케팅 성과에 중복 기여되지 않은 SKAN 이벤트 수를 더해서 앱 마케팅으로 발생한 최종적인 이벤트 수를 중복 없이 늘릴 수 있기 때문입니다. 아래 이미지처럼 어트리뷰션 모델, 중복 기여 여부에 따라 이벤트를 구분해서 실제 광고 성과를 더 정확하게 측정하는 방법을 예시와 함께 살펴보겠습니다.
[__wf_reserved_decorative]
먼저 중복 기여 여부를 판별하지 않고 SKAN과 에어브릿지 어트리뷰션 모델로 각각 기여된 앱 설치 수를 바탕으로 계산한 CPI 예시를 살펴보겠습니다. 지출한 광고 비용은 100만 원입니다.
[__wf_reserved_decorative]
SKAN으로 기여된 앱 설치 500개는 모두 iOS 유저가 실행한 앱 설치입니다. 이중에는 IDFA 수집을 허용한 iOS 유저와 허용하지 않은 iOS 유저가 섞여 있습니다. CPI 2,000원은 iOS 유저가 실행한 앱 설치 수를 기준으로 계산한 광고 성과입니다. 안드로이드 유저가 실행한 앱 설치는 제외됩니다.
[__wf_reserved_decorative]
에어브릿지 어트리뷰션 모델로 기여된 앱 설치 800개는 IDFA 수집을 허용한 iOS 유저와 안드로이드 유저가 실행한 앱 설치의 합계입니다. CPI 1,250원은 두 유저 그룹이 실행한 앱 설치 수를 기준으로 계산한 광고 성과입니다. IDFA 수집을 허용하지 않은 iOS 유저가 실행한 앱 설치는 제외됩니다.
이처럼 두 어트리뷰션 모델을 따로 활용하면 운영 체제, IDFA 수집 여부에 따라 일부 유저가 실행한 앱 설치를 알 수 없습니다. 중복 기여 판별을 활용하면 제외되거나 중복되는 이벤트 없이 두 어트리뷰션 모델로 측정한 모든 앱 설치 수를 알 수 있습니다.
아래 SKAN으로 기여된 앱 설치 수 중 중복 기여되지 않은 이벤트를 400개로 가정한 CPI 예시를 살펴보겠습니다. 지출한 광고 비용은 동일하게 100만 원이고 두 어트리뷰션 모델로 측정한 앱 설치 수를 더했습니다.
[__wf_reserved_decorative]
중복 기여되지 않은 앱 설치 400개는 IDFA 수집을 허용하지 않은 iOS 유저가 실행한 앱 설치입니다. 에어브릿지 어트리뷰션 모델로 기여된 앱 설치 800개는 IDFA 수집을 허용한 iOS 유저와 안드로이드 유저가 실행한 앱 설치의 합계입니다.
두 수치를 더해서 계산한 CPI 833원은 앱 마케팅으로 발생한 모든 앱 설치 수를 기준으로 계산한 광고 성과입니다. 운영 체제, IDFA 수집 여부에 따라 제외된 앱 설치는 없습니다. 중복 기여 판별을 활용한 결과, 최종적인 CPI는 SKAN과 에어브릿지 어트리뷰션 모델을 따로 따로 활용했을 때보다 최소 30% ~ 최대 70% 감소했습니다.
## 에어브릿지에서 중복 기여 판별을 지금 바로 설정하세요
CPI, CPA를 보정할 수 있는 중복 기여 판별은 에어브릿지에서 어떻게 설정할 수 있을까요? 에어브릿지에서 아래 3단계만으로 중복 기여 판별을 설정할 수 있습니다.
1. SKAN 전환값 설정 페이지 4.0 버전으로 변경하기
2. 측정 타입 중 중복 기여 판별 선택하기
3. 액츄얼스 리포트에서 중복 기여 판별 관련 메트릭, 그룹바이 추가하기
SKAN으로 이벤트 데이터를 측정하기 위해서는 전환값을 설정해야 합니다. 에어브릿지에서 SKAN 전환값을 설정하면 여러 앱 이벤트와 이벤트 속성을 각 전환값에 매핑할 수 있습니다. 중복 기여 판별도 전환값에 매핑할 수 있는데요.
매핑하고 싶다면 에어브릿지 iOS SDK를 1.33.1 이상으로 업데이트하고 에어브릿지의 SKAN 전환값 설정 페이지를 SKAN 4.0 버전으로 변경해야 합니다. 에어브릿지 CSM에게 SKAN 4.0 전환값 설정 페이지를 열어달라고 요청하면 SKAN 4.0 전환값 설정 페이지를 만날 수 있습니다.
[__wf_reserved_decorative]
> **💡 버전을 변경하기 전에 알아야 하는 1가지**
>
> SKAN 전환값 설정 페이지를 4.0 버전으로 변경하면 3.0 버전으로 돌아갈 수 없습니다. 하지만 3.0 버전에서 설정한 전환값을 4.0 버전에서 그대로 활용할 수 있습니다. 4.0 버전의 SKAN 전환값 설정 페이지로 SKAN 3.0만 지원하는 광고 채널의 이벤트 데이터를 계속 측정할 수 있습니다.
SKAN 4.0 전환값 설정 페이지에서 측정 타입 중 중복 기여 판별을 선택해 주세요. 중복 기여 판별을 선택하면 토글이 자동으로 활성화됩니다. 중복 기여 판별이 전환값에 매핑되면 SKAN 4.0 전환값에 매핑한 모든 인앱이벤트가 에어브릿지 어트리뷰션 모델로 중복 기여되었는지를 판별할 수 있습니다. 앱 설치는 중복 기여 판별을 전환값에 매핑하면 판별 대상에 자동으로 포함됩니다. 중복 기여 판별을 선택하면 SKAN 4.0을 지원하지 않는 광고 채널에서 전송된 전환값에 대한 중복 기여 여부도 판별할 수 있습니다.
[__wf_reserved_decorative]
👉 더 자세한 SKAN 4.0 전환값 설정 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/skadnetwork-4-settings)에서 확인할 수 있습니다.
SKAN 데이터를 충분히 수집했다면 액츄얼스 리포트(Actuals Report)에서 메트릭과 그룹바이를 선택해서 중복 기여 판별을 확인할 수 있습니다. 앱 설치에 관한 SKAN 메트릭 중에는 중복 기여 여부를 판별해서 수치를 제공하는 SKAN 메트릭이 있습니다.
- SKAN Overlap Installs (App): SKAN으로 기여되고 에어브릿지 어트리뷰션 모델로 중복 기여된 앱 설치 수를 표시합니다.
- SKAN Dedup Installs (App): SKAN으로 기여된 전체 앱 설치 수에서 에어브릿지 어트리뷰션 모델로 중복 기여된 앱 설치 수를 제외한 수치를 표시합니다.
그룹바이도 있습니다. SKAN Is Also Attributed By Airbridge(IDFA) 그룹바이를 추가하면 SKAN 메트릭의 중복 기여 여부를 확인할 수 있습니다. 표시되는 값은 true, false, 공백 총 3가지입니다.
- true: SKAN으로 기여되고 에어브릿지 어트리뷰션 모델로 중복 기여된 SKAN 메트릭 수치를 표시합니다.
- false: SKAN으로 기여되고 에어브릿지 어트리뷰션 모델로 중복 기여되지 않은 SKAN 메트릭 수치를 표시합니다.
- 공백: SKAN으로 기여되었으나 중복 기여 여부를 판별하지 않은 SKAN 메트릭 수치를 표시합니다.
[__wf_reserved_decorative]
👉 더 자세한 SKAN 4.0 데이터 조회 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/skadnetwork-4-viewing)에서 확인할 수 있습니다.
## 에어브릿지는 항상 더 정확한 광고 성과를 제공하고 있습니다
중복 기여 판별로 SKAN 이벤트의 중복 기여 여부를 판별하더라도 측정하지 못한 iOS 앱 마케팅 성과는 존재할 것입니다. 하지만 개인정보보호 시대에 더 정확하게 광고 성과를 측정할 수 있다는 점에서 중복 기여 판별은 마케터에게 또 다른 돌파구가 되어줄 것입니다.
중복 기여 판별 외에도 에어브릿지는 마케터를 위한 SKAN 4.0 기능 개발을 완료했습니다. 마케터는 에어브릿지에서 SKAN 4.0 전환값을 미리 설정할 수 있습니다. 미리 설정한 SKAN 4.0 전환값은 SKAN 4.0을 지원하는 광고 채널이 많아질수록 활용도가 올라갈 것입니다. 앞으로도 에어브릿지는 SKAN 4.0에 관한 동향을 살피며 기능을 꾸준히 개발해 나갈 예정입니다.
에어브릿지와 함께 SKAN 4.0으로 iOS 앱 마케팅 성과를 더 성공적으로, 그리고 더 정확하게 측정하세요.
---
## MMP에서 포스트백을 설정해야 하는 이유
URL: https://www.airbridge.io/ko/blog/mmp-postback
Language: Korean
Published: 2023-12-19
모바일 광고 세계에서 광고주와 광고 채널은 ‘더 많은 유저에게 광고를 좋은 타이밍에 보여주자’라는 하나의 목표를 위해 노력합니다. 하지만 아쉽게도 광고주와 광고 채널은 서로 다른 유저를 보고 있습니다. 수집하는 데이터가 다르기 때문입니다. 광고주가 수집할 수 있는 데이터는 유저가 앱에서 활동한 이벤트에 한정되어 있습니다. 광고 채널은 앱에서 발생한 이벤트를 수집할 없지만, 대신 광고 노출 위치, 광고 소재 같은 광고주가 수집할 수 없는 데이터를 수집할 수 있습니다.
여기서 MMP(Mobile Measurement Partner)는 광고주와 광고 채널 사이의 가교 역할을 합니다. 기본적으로 MMP는 여러 광고 채널이 제공한 데이터를 수집해 광고주에게 전달합니다. 전달한 결과는 에어브릿지 리포트나 로우 데이터로 확인할 수 있습니다.
반대 상황은 어떨까요? 광고 채널은 광고주의 데이터를 확인할 수 없을까요? 그렇지 않습니다. 바로 포스트백이 있기 때문입니다.
> ❗ 실제 포스트백은 다양한 곳에서 다양하게 발생합니다. 이 블로그에서는 에어브릿지를 기준으로 설명합니다
## 포스트백은 왜 필요한가요?
포스트백은 광고주 앱에서 발생한 광고 성과와 유저 행동 데이터를 광고 채널에 전송하는 기능입니다. 포스트백이 필요한 첫 번째 이유는 광고비 정산입니다. 액션당 비용(CPA, Cost Per Action)이나 설치당 비용(CPI, Cost Per Install)을 기준으로 광고비를 정산하는 경우에는 설치 또는 구매와 같은 이벤트 실행 여부와 채널의 기여 여부까지 포함된 데이터를 광고 채널에 포스트백을 통해 제공해야 합니다.
두 번째는 광고 최적화입니다. 포스트백으로 제공되는 풍부한 데이터를 통해 광고 성과를 끌어올릴 수 있습니다. 광고 채널은 포스트백 데이터로 어떤 유저가 광고를 클릭한 이후 설치나 구매 이벤트를 발생시켰는지, 해당 이벤트에 채널이 기여했는지까지도 알 수 있습니다. 결과적으로 보다 정교한 타겟 유저를 설정할 수 있게 됩니다. 광고에 더 반응하는 유저에게 광고를 더 자주 보여줄 수 있기 때문에 자연스럽게 더 높은 전환율을 기대할 수 있습니다. 이는 광고 채널과 광고주 모두에게 유용합니다.
마지막으로 리타겟팅 캠페인을 위한 오디언스 생성에 활용할 수 있습니다. 특히, 앱을 설치한 유저나 구매 이력이 있는 유저 등을 대상으로 하는 리타겟팅 캠페인을 운영하는 경우에도 포스트백 데이터가 필요합니다. 특정 앱 이벤트를 수행한 유저를 정의하기 위해서는 이벤트 실행 여부를 알아야 하기 때문입니다. 여기서 포스트백으로 제공된 테이터를 활용하면 리타겟팅을 위한 오디언스를 정의할 수 있습니다.
## 포스트백은 전송만 하면 되나요?
포스트백으로 전송하는 데이터는 다양한 목적으로 활용됩니다. 단순히 전송하기 보다는 목적에 알맞은 설정으로 포스트백을 전송해야 합니다. 예를 들어 설치당 비용으로 광고비를 정산하기 위해 포스트백을 전송할 때 기여 여부에 관계없이 모든 이벤트를 전송하게 되면 광고 채널은 기여한 설치 이벤트만 확인하기 위한 추가 작업이 필요합니다. 채널이 기여한 설치 이벤트만 전송하도록 설정하면 광고주와 채널 모두 리소스를 절약하고 광고비를 정확하게 정산할 수 있습니다.
리타겟팅 캠페인은 상황이 다릅니다. 구매 이력이 있는 유저를 대상으로 리타겟팅 캠페인을 집행하는 채널에는 오디언스 생성을 위해 기여한 채널에 관계없이 모든 구매 이벤트 데이터를 전송해야 충분한 데이터를 확보할 수 있습니다. 모든 이벤트를 전송할 수 없다면 리타겟팅 오디언스의 규모가 충분하지 않아 낮은 캠페인 성과로 이어질 수 있습니다.
[]
## 에어브릿지는 다양한 포스트백 전송 옵션을 제공합니다
광고 채널에 따라 포스트백 전송 방법이 다릅니다. 대부분의 SAN(Self Attributing Network, 셀프 어트리뷰팅 네트워크)은 사전에 협의한 이벤트를 포스트백으로 전송합니다. 앱 설치와 같은 일부 이벤트는 항상 전송해야 하지만, 일부 이벤트는 연동 설정 단계에서 전송 여부를 선택할 수 있습니다. SAN은 주로 포스트백으로 전송 받은 데이터를 활용해 캠페인을 최적화하거나 리타겟팅 광고를 정교하게 진행합니다.
SAN이 아닌 광고 채널에 원하는 정보를 포스트백으로 전송하기 위해 에어브릿지는 다양한 옵션을 제공하고 있습니다. 이벤트 발생 일시, 채널 기여 여부, 디바이스 정보 등 전송하는 데이터을 설정한 후에 전송 규칙을 선택하면 포스트백 설정이 완료됩니다. 에어브릿지는 2가지 포스트백 전송 규칙을 통해 원하는 데이터만 포스트백으로 전송할 수 있는 환경을 제공합니다.
> 💡 **에어브릿지의 포스트백 전송 규칙**
- Attributed to: 광고 채널에 해당 광고 채널이 기여한 이벤트만 전송하거나 기여 여부와 상관 없이 모든 이벤트를 전송할 수 있습니다.
- First only: 광고 채널에 신규 이벤트만 전송하거나 발생한 모든 이벤트를 전송합니다.
구매 이력이 있는 유저를 대상으로 리타겟팅 캠페인를 진행한다면 어떻게 설정해야 할까요? 리타겟팅 캠페인은 광고 채널이 구매 이벤트를 수행한 유저 데이터를 최대한 많이 확보하는 것이 중요합니다. 따라서 포스트백 전송 규칙으로 기여 여부와 상관 없이 모든 이벤트를 전송하고, 발생한 모든 이벤트를 전송하면 효과적인 리타겟팅 캠페인을 진행할 수 있습니다.
## 원하는 광고 채널로 포스트백을 전송하세요
에어브릿지와 연동된 모든 광고 채널에 포스트백을 전송할 수 있습니다. 아래 가이드에 있는 광고 채널 목록에서 에어브릿지와 연동된 광고 채널을 확인할 수 있습니다. 포스트백을 전송하고 싶은 광고 채널을 목록에서 찾을 수 없다면 광고 채널에 연동 신청서 작성을 요청해 주세요.
- [에어브릿지와 연동된 광고 채널 확인하기](https://help.airbridge.io/ko/guides/ad-channels-list#%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80%EC%99%80-%EC%97%B0%EB%8F%99%EB%90%9C-%EA%B4%91%EA%B3%A0-%EC%B1%84%EB%84%90-%EB%AA%A9%EB%A1%9D)
- [광고 채널 연동 신청서](https://help.airbridge.io/ko/supports/integrations)
---
## 모바일 게임 흥행을 위한 MMP 선택하기
URL: https://www.airbridge.io/ko/blog/mmp-for-mobile-gaming-apps
Language: Korean
Published: 2023-12-07
아무리 잘 만들어진 게임이라도 세상에 널리 알려 플레이어가 모이지 않으면 의미가 없습니다. 공들여 만든 게임을 어떻게 하면 효과적으로 전 세계 유저들에게 알릴 수 있을까요? 그 유저들을 놓치지 않고 내 게임의 팬으로 만드는 방법은 또 어떤 게 있을까요? 게임 출시를 앞두고 있다면 마케팅 집행과 흥행을 책임질 MMP 도입을 필수적으로 고려해야 합니다.
## MMP가 우리 게임에 필요한 이유
몇 년 전만 해도 몇 개 주요 매체에서만 광고를 돌리고 그 성과만 측정해도 충분한 시대였습니다. 다만 시대는 빠르게 변했고 더 이상 몇 개의 매체만 사용해서는 글로벌 시장에서 살아남을 수 없습니다. 최대한 다양한 매체에서 여러 광고를 집행한 뒤 그중 단시간 내에 효율을 내는 매체를 선정해야 합니다. 또한 수많은 매체 중 우리 타겟 유저와 앱에 적합한 매체를 찾는 것도 마케터의 중요한 일 중 하나입니다.
여러 애드 네트워크 및 매체와 광고를 집행하기 위해서는 앱 어트리뷰션 툴 / MMP (Mobile Measurement Partner) 가 필요합니다. MMP 가 제공하는 소프트웨어 개발 키트 (Software Development Kit, SDK)를 앱에 설치한 뒤 광고를 집행하면 MMP의 대시보드 상에서 한눈에 여러 매체의 성과를 비교 분석할 수 있게 됩니다. 각 매체의 SDK를 일일이 수동으로 설치할 필요 없이 하나의 SDK로 모든 네트워크를 관리할 수 있으며 향후 애드 네트워크의 연결 해지 역시 마케터가 쉽게 진행할 수 있습니다.
[]
매체사 데이터와 실제 마케터가 확인하는 앱 내부 데이터에는 차이가 있을 수 있습니다. MMP의 가장 큰 장점은 매체도, 광고주도 아닌 제3자의 입장에서 공정하고 정확한 데이터를 확인할 수 있는 부분입니다. 또한 애플 서치 애즈, 구글 애즈 등과 같이 SAN (Self-Attributing Network)이라 불리는 매체들은 자체 알고리즘을 사용하여 어트리뷰션 결과를 보여주기 때문에 다른 매체들과의 일괄 비교 분석이 어려워집니다. MMP에서는 다양한 출처의 데이터를 하나의 통합된 대시보드에서 볼 수 있습니다.
MMP 도입의 가장 이상적인 시기는 론칭 테스트/ 베타 테스트 기간입니다. 출시 초기부터 모든 데이터를 한곳에 모아 수집할 수 있고, 본격 출시 전 유저가 앱을 이탈하는 지점을 파악하여 미리 개선이 가능하기 때문입니다. MMP 와 함께 빠르고 손쉽게 다양한 매체와 광고를 집행하여 최대한 많은 유저에게 내 앱을 소개해 보세요!
> 각 매체사의 SDK를 설치할 필요 없이 에어브릿지 SDK 만 설치하면 여러 매체의 광고를 집행할 수 있습니다. 에어브릿지의 SDK 크기는 1메가 바이트(MB) 이하로 게임 성능에 영향을 주지 않으며 마케터는 쉽고 빠르게 애드 네트워크를 한곳에서 관리할 수 있습니다.
## MMP가 제공하는 리포트로 편리하게 다양한 데이터 확인하기
광고를 집행해 유저를 획득했다면 내 게임을 인스톨 한 유저들이 그 이후에 앱 내에서 어떤 액션을 일으키는지, LTV (Lifetime Value) 와 이탈률은 어떻게 되는지 코호트 단위로 세밀한 분석이 필요합니다. MMP에서 제공하는 퍼널 리포트를 사용하면 코호트 단위로 유저 여정을 분석할 수 있습니다. 이 리포트에서는 앱을 설치한 유저들이 어느 단계에서 이탈을 했는지, 앱 설치부터 아이템 결제까지 걸리는 시간이 가장 짧은 채널은 어떤 것인지, 지난 7일 동안 퍼널 별 전환율은 어떻게 되는지 등의 다양한 데이터를 한눈에 볼 수 있습니다.
그뿐만 아니라 매체 별 광고 ROI (Return On Investment) 분석을 통해 이 매체가 정말 성과를 내고 있는지에 대한 파악이 필요합니다. MMP에서는 자동으로 네트워크에 대한 비용 관련 정보가 연동되어 성과는 물론 CPI, CPC, CTR, CVR 등의 비용 정보 역시 시각화하여 확인할 수 있습니다. 다양한 파트너, 국가 및 플랫폼의 비용을 한곳에 모아 채널, 캠페인, 소재별로 비용부터 전환, 매출 데이터를 최적화해보세요. 내 게임을 좋아하는 유저가 어떤 매체의 어느 캠페인을 통해서 유입되고 있는지 세분화하여 확인할 수 있으며 성과가 나오지 않는 매체는 비활성화 시켜 마케팅 비용을 절감할 수 있습니다.
> 에어브릿지의 퍼널 리포트에서는 광고부터 이어지는 앱 내 유저 여정을 분석할 수 있습니다. 특히 에어브릿지가 자체 개발한 데이터베이스 LUFT 덕분에 특정 코호트를 빠르게 생성할 수 있고 그룹 바이 기능을 통해 채널, 국가, 기간, OS 별 행동 패턴도 빠르게 분석이 가능합니다.
👉 [모바일 게임 앱에서 퍼널 분석하는 방법 자세히 알아보기](https://www.airbridge.io/ko/blog/funnel-analysis-for-gaming-apps)
[]
[]
여러 각도의 분석을 통해 유저들의 성향을 파악한 뒤에는 그 유저들이 내 앱에 얼마나 머무르고 있는지에 대한 잔존율을 확인할 필요가 있습니다. MMP 가 제공하는 리텐션 리포트를 통해 유저의 게임 몰입도를 코호트 기반으로 분, 시간, 일, 주, 월 단위로 분석할 수 있습니다. 또한 유저 리텐션과 함께 채널, 캠페인, 소재, 국가 등으로 구분하여 pLT (예측 잔존일)를 확인할 수도 있습니다. 내 게임을 실행한 유저들이 어느 시점에서 주로 이탈하는지, 혹은 구매 및 다른 앱 내 이벤트를 발생시키는 패턴은 어떤지 데이터로 전환율을 확인할 수 있습니다.
MMP를 활용하면 대시보드 상에서의 수치뿐만이 아닌 로우 데이터 (Raw Data) 추출 역시 가능합니다. 원하는 데이터만 필터링하여 내부 시스템에 적재할 수 있으며 추출한 유저 데이터를 가지고 리타게팅 캠페인을 집행할 수도 있습니다. 특정 코호트 별로 유저 그룹을 만들어 차별화된 전략의 맞춤 개인화 마케팅을 진행해 보세요. 나에 대해서 잘 알고 있다고 느낀 유저는 게임의 충성 팬이 되어 돌아올지도 모릅니다.
> 에어브릿지가 제공하는 오디언스 매니저를 통해 약 220여 종류의 로그 데이터를 대시보드 상에서 손쉽게 추출할 수 있습니다. 또한 미리보기 기능을 제공하여 추출될 로우 데이터를 사전에 확인할 수 있습니다.
## 개인정보보호 시대의 마케팅, MMP의 역할은?
유저 개인정보 보호 강화에 대한 움직임은 마케팅 생태계에도 큰 변화를 가져왔습니다. 애플이 앱 트래킹 투명성을 위해 광고 식별자를 사용하지 않는 SKAdNetwork (SKAN) 을 도입하면서 업계에서는 어떻게 하면 유저의 개인정보를 보호하면서 개인화 마케팅을 잘할 수 있을지에 대해 고민하고 있습니다. 이 변화에 대응하고자 MMP에서는 매체로부터 전송받은 SKAN 전환값 (Conversion Value) 을 해석하고 데이터를 집계하여 한눈에 매체 별로 확인할 수 있도록 대시보드를 제공합니다. 또한 MMP에서 제공하는 GroupBy 와 다양한 메트릭을 사용하여 보다 정교하게 SKAN 성과를 측정할 수 있습니다.
> 에어브릿지에서는 직관적이고 손쉽게 전환값을 설정할 수 있도록 트래킹 세트 방식을 지원합니다. 프리셋 형태로 제공되어 측정하고자 하는 전환 성과의 인앱 이벤트를 간편하게 설정할 수 있습니다.
MMP를 선택할 때 어떤 전환값이 지원되는지, 여러 모델 유형을 결합할 수 있는지 등 SKAN 전환값 유연성을 어떻게 제공하는지 확인이 필요합니다. 애플뿐만 아니라 구글에서도 유저의 개인정보 보호를 위해 Privacy Sandbox 출시를 앞두고 있어 또 다른 거대한 변화가 예상됩니다. 에어브릿지와 같은 MMP는 애플, 구글은 물론 주요 매체들과 긴밀한 협업을 통해 개인정보보호 시대에 따른 마케팅의 변화에 적극적으로 대응해가고 있습니다.
게임 출시부터 그 이후까지 MMP가 게임 흥행에 어떻게 도움을 주는지에 대해 살펴보았습니다. 데이터 기반 의사결정이 대두되는 이 시대의 모든 마케터들에게 MMP 도입은 더 이상 선택이 아닌 필수 사항입니다. 얼마나 유연한 리포트를 제공하는지, 제공된 리포트에서 내가 어떤 지표를 유의미하게 사용할 수 있을지 (비용 데이터, 코호트 및 리텐션 등) 종합적으로 여러 요소를 고려하여 우리 게임에 가장 적합한 MMP를 선택해 보세요!
---
## 예측 기능으로 더 빠른 마케팅 의사 결정을 내려보세요
URL: https://www.airbridge.io/ko/blog/airbridge-predictive-lifetime
Language: Korean
Published: 2023-12-06
퍼포먼스 마케팅에서 생기는 가장 큰 고민 중 하나는 광고비를 적절한 수준으로 사용하고 있는지 판단하는 것입니다. 직관적인 방법은 사용한 광고비와 발생한 매출을 비교하는 방법입니다. 광고비보다 매출이 많다면 좋은 캠페인이고, 반대로 매출보다 광고비가 많다면 계속 집행하지 않아도 되는 캠페인으로 판단할 수 있습니다. 예를 들어 초기 유저 획득 비용(Cost Acquisition Cost, CAC)이 10,000원 이하로 확인된 캠페인을 통해 유입된 유저가 평균적으로 1달 동안 10,000원 넘게 사용하고 있다면 계속 운영해야 하는 캠페인으로 여길 수 있습니다. 이를 통해 효율적인 광고 예산 관리와 높은 전환율의 캠페인 운영이라는 2가지 토끼를 모두 잡을 수 있습니다.
하지만 여기에는 어려운 문제가 있습니다. 광고로 유입된 모든 유저는 1달 내내 서비스에 방문하지 않습니다. 또한 유저가 1달 동안 사용하는 금액은 1달 뒤에 확인할 수 있기 때문에 발빠르게 마케팅 전략을 바꾸기 어렵습니다. 이 문제를 해결하기 위해서는 광고로 유입된 유저가 얼마나 자주 서비스에 재방문하는지, 그리고 어느만큼의 가치를 발생시킬 수 있는지 미리 알아야 합니다. 그래서 에어브릿지는 예측 기능을 지원하기 시작했습니다.
## 유저를 미리 파악하는 예측 잔존일과 pLTV
미래를 정확하게 예측하는 일은 불가능합니다. 하지만 풍부한 데이터를 기반으로 유저행동을 추정할 수 있습니다. 에어브릿지는 수집한 데이터를 바탕으로 유저가 며칠 동안 서비스에 재방문하는지, 그리고 유저가 만드는 가치가 어느 정도인지 빠르게 추정한 수치를 예측 기능을 통해 제공합니다.
에어브릿지가 예측 기능을 통해 제공하는 지표는 예측 잔존일(Predictive Lifetime )과 예측 LTV(Predictive LTV, pLTV) 2가지입니다. 리텐션 리포트에서 확인할 수 있는 예측 잔존일은 유저가 미래에 서비스를 며칠 동안 재방문하는지 추정한 수치를 확인할 수 있는 지표입니다. 예측 잔존일이 9.5일이라면 유저가 설정한 기간 중에 9.5일 동안 재방문한다고 추정할 수 있습니다.
pLTV는 예측 잔존일을 통해 추정한 LTV입니다. 유저가 서비스에서 매출을 얼마나 발생시키는지 추정한 수치를 확인할 수 있습니다. pLTV가 30,000원이라면 예측 잔존일 동안 서비스를 재방문한 유저가 30,000원만큼 매출 발생시킬 것이라고 추정할 수 있습니다.
[]
하지만 누구도 미래를 정확하게 예측할 수 없습니다. 그렇다면 유저를 미리 파악할 수 있는 예측 잔존일과 pLTV를 어떻게 활용할 수 있을까요?
## 예측 지표로 더 다양한 성과를 확인할 수 있어요
그룹바이, 필터, 코호트 등 에어브릿지에서는 다양한 방식으로 유저를 그룹으로 정의할 수 있습니다. 또한 예측 잔존일과 pLTV를 포함한 다양한 메트릭을 통해 캠페인에서 발생한 성과를 확인할 수 있습니다. 예시를 확인해 봅시다.
### Case 1
먼저, 아래처럼 예측 잔존일과 pLTV가 확인된 3개의 유저 그룹이 있습니다.
[]
예측 잔존일만 보면 C 그룹이 가장 오래 서비스를 재방문하는 것으로 추정됐습니다. 하지만 pLTV 기준으로는 A 그룹이 가장 큰 기대값을 가지고 있습니다. 커뮤니티 등 많은 유저가 참여하는 것이 필요한 서비스라면 C 그룹을 대상으로 캠페인을 운영하는 것이 좋을 수 있습니다. 단기간 매출이 필요하다면 pLTV가 가장 큰 A 그룹이 좋은 선택일 수 있습니다. 서비스 특성과 현황에 맞춰 예측 잔존일과 pLTV로 타겟 유저를 더 세밀하게 겨냥할 수 있습니다.
### Case 2
pLTV로 여러 캠페인 중에 계속해서 운영하면 좋은 캠페인도 빠르게 파악할 수 있습니다. 아래 캠페인 3가지가 있습니다.
[]
이 중에서 A 캠페인은 pLTV가 CAC보다 크기 때문에 계속 운영하면 좋은 캠페인으로 볼 수 있습니다. 반면 C 캠페인은 CAC보다 pLTV가 작기 때문에 캠페인 중단을 고려할 수 있습니다. 실제로는 2가지 지표만으로 캠페인 운영 여부를 결정할 수 없겠지만 캠페인 현황을 빠르게 참고할 수 있는 지표로 활용할 수 있습니다. 에어브릿지는 더 빠르게 현황을 파악할 수 있게 레비뉴 리포트에서 pLTV와 CAC를 함께 제공하고 있습니다.
## 에어브릿지는 마케터의 더 빠른, 더 정확한 의사 결정을 돕기 위해 노력해요
많은 마케터가 불확실한 미래에서 정답을 찾기 위해 고군분투하고 있습니다. 에어브릿지는 항상 마케터가 더 빠르고 정확하게 의사 결정할 수 있는 다양한 기능과 지표를 제공하기 위해 노력하고 있습니다. 이번 예측 기능도 노력의 일환입니다. 에어브릿지가 제공하는 예측 잔존일과 pLTV를 아래 가이드를 참고해 한 번 사용해 보세요.
- [리텐션 리포트에서 예측 잔존일 확인하기](https://help.airbridge.io/ko/guides/predictive-lifetime)
- [레비뉴 리포트에서 pLTV 확인하기](https://help.airbridge.io/ko/guides/predictive-ltv)
---
## 에어브릿지, Meta Install Referrer 지원 시작
URL: https://www.airbridge.io/ko/blog/airbridge-meta-install-referrer
Language: Korean
Published: 2023-11-28
에어브릿지가 메타 안드로이드 캠페인에서 발생한 앱 설치 성과를 더 개선된 형태로 제공합니다. 에어브릿지는 메타와 긴밀한 협업을 통해 새로운 광고 성과 측정 방법인 메타 인스톨 리퍼러(Meta Install Referrer)를 2023년 11월 23일부터 지원합니다. 메타 인스톨 리퍼러를 통해 메타 안드로이드 캠페인에서 발생한 앱 설치 성과를 에어브릿지에서 더 풍부한 데이터로 확인할 수 있습니다.
## 메타 인스톨 리퍼러는 무엇인가요?
메타 인스톨 리퍼러는 안드로이드 캠페인 전용 성과 측정 방법으로 앱 설치에 기여한 광고 노출, 광고 클릭에 대한 정보를 확인할 수 있습니다. 특히, 메타 인스톨 리퍼러는 싱글 세션 클릭 스루 어트리뷰션을 포함해 기존 구글 플레이 인스톨 리퍼러(Google Play Install Referrer)가 지원하지 않는 대부분의 뷰스루 어트리뷰션과 멀티 세션 클릭 스루 어트리뷰션도 지원합니다.
## 어떻게 진행되나요?
유저가 페이스북이나 인스타그램과 같은 메타 앱에서 광고주의 앱 설치 광고를 보거나 클릭하면 메타 앱은 해당 광고 캠페인의 메타 데이터를 암호화하여 기기 내 메타 앱의 로컬 스토리지에 저장합니다.
이후 유저가 앱을 설치한 후에 처음 앱을 실행하면 에어브릿지 SDK가 기기의 로컬 스토리지에서 메타 인스톨 리퍼러 데이터를 가져옵니다. 에어브릿지는 이 메타 인스톨 리퍼러를 복호화한 후에 어트리뷰션에 활용합니다.
## 메타 인스톨 리퍼러가 어떻게 도움이 될까요?
메타 안드로이드 캠페인 성과를 더 정확하게 파악할 수 있습니다. 인스톨 리퍼러 데이터를 기반으로 미기여된 설치(unattributed installs)의 수가 줄어들고 대신 뷰스루 혹은 멀티 세션의 클릭스루 설치로 기여되기 때문입니다. 메타 인스톨 리퍼러가 제공하는 정보는 에어브릿지 리포트와 로우 데이터로 확인할 수 있습니다.
## 메타 인스톨 리퍼러는 구글 플레이 인스톨 리퍼러와 함께 어떻게 작동하나요?
메타 인스톨 리퍼러는 구글 플레이 인스톨 리퍼러와 동일한 암호 해독 키를 사용합니다. 인스톨 리퍼러를 설정하지 않았다면 [에어브릿지 가이드](https://help.airbridge.io/ko/guides/meta-business-google-play-install-referrer)를 참고해 설정해 주세요.
에어브릿지는 메타 안드로이드 캠페인에서 발생한 앱 설치 전환에 대한 기여를 판단하기 위해서는 메타 인스톨 리퍼러, 구글 플레이 인스톨 리퍼러, 채널 연동으로 제공된 3가지 터치포인트 데이터 중 하나만 사용합니다. 데이터에서 중복 집계는 발생하지 않지만, 더 많은 데이터를 고려하기 때문에 더 정확한 데이터 정합성을 기대할 수 있습니다.
## 에어브릿지에서 인스톨 리퍼러 데이터를 어떻게 조회하나요?
에어브릿지에서 메타 인스톨 리퍼러를 수집하고 복호화하기 위해서는 먼저 최신 버전의 에어브릿지 안드로이드 SDK를 설치한 후에 [수집 설정](https://help.airbridge.io/ko/developers/android-sdk#%EB%A9%94%ED%83%80-%EC%9D%B8%EC%8A%A4%ED%86%A8-%EB%A6%AC%ED%8D%BC%EB%9F%AC-%EC%88%98%EC%A7%91-%EC%84%A4%EC%A0%95)을 추가해 주세요. 그리고 메타에서 제공하는 암호 해독 키(Decryption Key)를 에어브릿지에 입력하면 복호화된 메타 인스톨 리퍼러를 리포트와 로우 데이터로 확인할 수 있습니다. 이 과정은 SDK 설정을 제외하면 구글 플레이 인스톨 리퍼러를 복호화하는 과정과 비슷합니다.
보다 자세한 설정 방법은 [유저 가이드](https://help.airbridge.io/ko/guides/meta-business-google-play-install-referrer)에서 확인할 수 있습니다. 또한 전담 에어브릿지 CSM을 통해 설정과 관련한 자세한 지원을 받아 보세요.
> [FAQ]
iOS에 대해서도 지원하나요?
지원하지 않습니다. iOS는 메타 인스톨 리퍼러 메커니즘에 상응하는 구조가 없습니다.
메타 인스톨 리퍼러를 어떤 메타 캠페인의 어트리뷰션 분석에 활용하나요?
• 설치 어트리뷰션
- 모든 캠페인에 대한 클릭스루 어트리뷰션
- 어드밴티지+ 앱 캠페인 및 넓은 타겟의 수동 앱 광고 캠페인(예: 연령, 성별 등이 기본으로 설정된 경우)의 뷰스루 어트리뷰션
• 랜딩 페이지로 리다이렉트하는 캠페인의 경우 포함되지 않습니다.
메타 인스톨 리퍼러는 에어브릿지 어트리뷰션 우선 순위에서 어떻게 반영 되나요?
터치포인트 데이터 간의 어트리뷰션 우선 순위는 [유저 가이드](https://help.airbridge.io/ko/guides/meta-business-google-play-install-referrer#%EB%A9%94%ED%83%80-%EC%9D%B8%EC%8A%A4%ED%86%A8-%EB%A6%AC%ED%8D%BC%EB%9F%AC%EC%99%80-%EA%B5%AC%EA%B8%80-%ED%94%8C%EB%A0%88%EC%9D%B4-%EC%9D%B8%EC%8A%A4%ED%86%A8-%EB%A6%AC%ED%8D%BC%EB%9F%AC)를 참고해 주세요.
---
## 모바일 앱 성장을 위해 모바일 어트리뷰션 툴 똑똑하게 활용하는 방법 5가지
URL: https://www.airbridge.io/ko/blog/how-to-use-a-mobile-app-attribution-tool
Language: Korean
Published: 2023-11-24
모바일 앱 마케팅을 하고 있다면 모바일 앱 어트리뷰션 툴, MMP는 반드시 필요합니다. 하나의 SDK로 간편하게 여러 광고 매체의 성과를 한 번에 확인하고 최적화할 수 있기 때문입니다.
더 나아가서 모바일 앱 어트리뷰션 툴은 마케팅 최적화 및 모바일 트렌드에 앞장선 중요한 마케팅 스택 중 하나로 자리 잡았습니다. 변화하는 모바일 업계 트렌드에 대응할 수 있으면서도 앱을 성장시킬 수 있는 기능들을 계속해서 발전시켜가고 있습니다. 이번 포스트를 통해 앱 어트리뷰션 툴을 보다 잘 활용하여 앱을 성장시킬 수 있는 5가지 방법을 확인해보세요.
> **[주요 내용 미리 보기]
**
- 웹에서 앱으로 유입된 유저의 여정이 궁금하시나요? 웹투앱 마케팅 성과 통합 분석하기
- 특정 이벤트를 수행한 타겟 오디언스를 쉽고 빠르게 생성하여 전환율과 LTV를 극대화
- 로우 데이터로 편리하고 완벽한 데이터 파이프라인 구축
- 개인정보보호 강화 트렌드 속 iOS 14.5 업데이트 대응과 SKAN 적용도 간편하게
- 라스트 클릭 기반의 성과 측정은 물론, MTA와 MMM과 같은 차세대 마케팅 성과 분석으로 진정한 성과 확인하기
## 웹에서 앱으로 유입된 유저의 여정이 궁금하시나요? 웹투앱 마케팅 성과 통합 분석하기
[]
앱 마케팅을 한다고 해서 웹이 중요하지 않은 것은 아닙니다. 앱 마켓의 제한적인 지면에서는 유저들을 소구하여 전환시키기에 한계가 있을 수밖에 없기에, 유저의 앱 마켓 랜딩 전 모바일 웹에서 보다 풍부한 콘텐츠와 인센티브를 제공하는 경우가 많아졌습니다. 그리고 모바일 웹에 접속한 경우, ‘앱에서 더 편리하게 이용하기’ 등의 스마트 앱 배너(Smart app banner)를 통해 앱 설치 혹은 앱에서의 행동을 유도하는 캠페인도 어렵지 않게 찾아볼 수 있습니다.
이렇게 웹에서 앱으로 유입된 유저 여정에 대해 완전하게 파악하려면, 웹 SDK를 제공하고 있는 모바일 앱 어트리뷰션 툴을 사용해야만 합니다. 대부분의 모바일 앱 어트리뷰션 툴은 앱 중심이기 때문에 앱 설치 및 인앱 이벤트에 대해서만 분석할 수 있으며, 웹에 대해서는 별도의 분석 툴을 활용해야 합니다. 이러한 경우 브라우저 내에서 발생한 전환만이 웹 캠페인으로 기여되며, 이외에 웹에서 기인한 앱 설치 및 인앱 이벤트의 경우 오가닉/미기여로 분석되어 웹 캠페인에 대한 성과가 과소평가될 수 있습니다. 즉, 진짜 완전한 유저 여정의 마케팅 성과를 확인할 수 없습니다.
실제로 에어브릿지 고객사[ 디코드(d.code)](https://www.airbridge.io/ko/case-studies/dcode-web-to-app-attribution)는 웹투앱 통합 분석을 활용한 후 웹 캠페인으로 유입된 유저의 구매 전환율이 앱 설치 캠페인으로 유입된 유저에 비해 높게 나타난다는 사실을 발견하였고, 웹 캠페인에 보다 많은 예산을 분배하여 구매 전환율을 개선할 수 있었습니다.
웹 및 앱 SDK의 통합 분석 기능을 갖춘 앱 어트리뷰션 툴을 활용하여 웹에서 앱으로 어트리뷰션 정보를 전달하고, 진짜 마케팅 성과를 확인해 보세요.
> 👉 웹투앱 전환을 높이는 방법과 트래킹에 대한 보다 자세한 내용은 [‘전환율을 높이는 웹투앱 경험 구축하기’ 화이트페이퍼](https://www.airbridge.io/ko/library/web-to-app-conversion-whitepaper)에서 확인해보세요.
## 특정 이벤트를 수행한 타겟 오디언스를 쉽고 빠르게 생성하여 전환율과 LTV를 극대화
[]
모바일 앱 마케팅에서 유저 획득(UA, User Acquisition)도 중요하지만, 획득한 유저가 지속적으로 앱에서 행동하고 소통하도록 만들어서 리텐션을 높이고 고객 생애 가치(LTV)를 최대화하는 것도 중요한 목표 중 하나입니다. 이 목표를 달성하기 위한 주요한 전략 중 하나로, 회원 가입 등 특정 행동을 한 유저 대상의 리타겟팅 광고 혹은 리마케팅을 집행하는 방법이 있습니다. 이를 위해서는 앱 설치 및 오픈 혹은 회원 가입 등 특정 이벤트를 수행한 오디언스 데이터가 필요하며, 이때 모바일 앱 어트리뷰션 툴의 오디언스 생성 및 추출 기능을 활용하면 효율적입니다.
유저의 행동 데이터는 앱 내부에 이미 누적되어 있겠지만, 이를 기반으로 타겟 오디언스를 생성하기 위해서는 개발팀의 도움이 필요하거나 마케터가 직접 데이터를 가공하는 과정을 거쳐야 합니다. 한편 이 데이터는 모바일 앱 어트리뷰션 툴도 SDK 등을 통해 수집하여 저장하고 있기 때문에, 대부분의 모바일 앱 어트리뷰션 툴은 이 데이터를 기반으로 타겟 오디언스를 생성하고 추출할 수 있는 기능을 제공합니다. 게다가 생성된 오디언스 데이터를 특정 광고 채널 및 서드파티 솔루션과 자동으로 연동하여 맞춤 타겟으로 생성할 수도 있습니다. 이를 기반으로 리타겟팅과 디타겟팅 혹은 맞춤 타겟을 기반으로 한 유사 타겟 생성으로 효과적인 성과를 내는 캠페인을 집행할 수 있습니다.
모바일 앱 어트리뷰션 툴이 가지고 있는 퍼스트 파티 데이터를 활용하여 타겟 오디언스를 생성하고 유저 전환율과 LTV를 최대화해보세요.
## 로우 데이터로 편리하고 완벽한 데이터 파이프라인 구축
[]
모바일 앱 어트리뷰션 툴은 어트리뷰션을 하기 위해 트래킹 링크, API, 리퍼러(Referrer), SDK 등을 통해 터치포인트와 전환 데이터를 모두 수집합니다. 그리고 이를 처리하고 가공하여 어트리뷰션 모델을 기반으로 전환에 기여한 터치포인트에 기여도를 부여합니다. 이 결과를 어트리뷰션 툴의 대시보드에서 확인할 수 있는 것이죠. 이때 어트리뷰션 툴로 수집한 모든 로우 데이터는 직접 다운로드해서 심화적인 분석 혹은 데이터 내재화에도 활용할 수 있습니다.
실제로 에어브릿지의 고객사[ 퀸잇](https://www.airbridge.io/ko/case-studies/queenit-data-pipleline-analytics)은 에어브릿지를 데이터 추출과 이를 통해 데이터 파이프라인을 구축하는 용도로 사용하고 있습니다. 데이터를 수집한 후 데이터 내보내기(Export) 기능을 통해 외부 클라우드의 스토리지에 자동으로 적재하여 데이터 파이프라인을 수월하게 구축하고 있습니다.
모바일 어트리뷰션 툴에서 제공하는 대시보드 외에 보다 심화적인 분석과 데이터 파이프라인의 구축이 필요하다면 앱 어트리뷰션 툴의 로우 데이터 추출 및 연동 기능을 활용해 보세요.
## iOS 14.5 업데이트 대응과 SKAN 적용도 간편하게
[]
iOS 14.5 업데이트 이후 앱 추적 투명성(ATT) 정책이 도입되었으며. Apple의 SKAdNetwork(SKAN)가 iOS 캠페인 성과 측정에 필수적인 솔루션이 되었습니다. 유저가 광고 식별자(IDFA, Identifier for advertisers) 수집에 동의했다면(ATT opt-in) 광고 식별자를 확인하여 기존과 같은 어트리뷰션이 가능하지만, 유저가 동의하지 않았다면(ATT opt-out) SKAdNetwork를 통해서 캠페인을 집행했을 경우에만 유저 정보 없는 어트리뷰션 성과를 확인할 수 있습니다. [2022년 3월 기준으로 46%의 유저만이 앱 추적 투명성 정책에 동의](https://www.statista.com/statistics/1281345/apple-att-opt-in-rate-by-app-category/)한 만큼, iOS 유저를 대상으로 마케팅을 한다면 SKAdNetwork의 활용은 반드시 필요한 상황이 되었습니다.
이때 앱 어트리뷰션 툴들이 제공하는 iOS 14.5 업데이트 및 SKAdNetwork 활용에 대한 대응책을 이용하면 보다 효율적으로 SKAdNetwork 캠페인을 집행할 수 있습니다. 없었던 정책이 생겨난 만큼 SKAN의 개념과 대응 방법에 대해 낯설어하는 앱 마케터들에게 다양한 인사이트를 제공하는 것은 물론, SKAdNetwork 활용 시 반드시 필요한 컨버전 밸류 설정을 편리하게 할 수 있는 기능도 제공합니다. 그리고 앱 어트리뷰션 툴에서는 Android 및 iOS에서 발생한 성과와 SKAdNetwork 캠페인의 성과를 포함한 전체적인 마케팅 성과를 한 번에 분석할 수 있습니다.
이외에도 에어브릿지의 경우 별도의 개발 없이도 최신 Airbridge iOS SDK만 설치하면 간단한 대시보드 설정으로 바로 SKAN을 사용할 수 있으며, Android 및 iOS 그리고 SKAdNetwork 캠페인의 성과를 하나의 리포트에서 한눈에 확인할 수 있습니다. 또한 컨버전 밸류 설정에 대한 성공 사례를 모은 프리셋으로 간편하게 컨버전 밸류를 설정할 수 있는 기능을 제공하기도 합니다.
iOS 14.5 업데이트 및 SKAN에 대한 대응이 필요하다면, 앱 어트리뷰션 툴의 도움을 받아보세요.
## 라스트 클릭 기반의 성과 측정은 물론, 통합 마케팅 성과 분석으로 균형잡힌 시각에서 진정한 성과 확인하기
[]
광고 성과 분석을 위해서는 어떤 터치포인트가 어떤 전환 이벤트에 기여했는지 기준을 제공하는 어트리뷰션 모델을 활용해야 합니다. 일반적으로 앱 어트리뷰션 툴들은 라스트 터치 어트리뷰션(LTA, Last Touch Attribution) 모델을 기준으로 성과 데이터를 제공합니다.
하지만 최근 모바일 업계는 유저 레벨 데이터 수집이 어려워지는 개인정보보호 강화 트렌드 속 또 다른 성과 분석 방법을 모색하고 있으며, 마지막 터치 포인트에 모든 성과를 부여하는 라스트 터치 어트리뷰션 모델의 한계에 대한 대안을 찾고 있습니다. 이에 어트리뷰션 툴들은 이러한 트렌드에 앞장서서, 라스트 터치 어트리뷰션에 대한 대안으로 멀티 터치 어트리뷰션(MTA, Multi Touch Attribution)을 제공하거나 개인정보를 보호하면서도 제대로 마케팅 성과를 분석하는 방법으로 마케팅 믹스 모델링(MMM, Marketing Mix Modeling)을 제안하고 있습니다.
각 비즈니스가 놓인 상황이나 기업별 마케팅 전략에 따라 필요한 성과 측정 방식은 달라질 수 있지만, 다음과 같은 [통합 측정 스택(Unified Measurement Stack)](https://www.airbridge.io/ko/blog/airbridge-unified-measurement-stack)이 균형잡힌 시각에서 성과를 분석하여 앱 서비스의 진정한 성장 및 마케팅 최적화를 도와줄 수 있습니다. 예를 들어, 마케팅 전략 설정 단계에서는 마케팅 믹스 모델링 분석을 통해 전체 마케팅 채널의 트렌드 및 미래 전략의 방향성을 확인할 수 있습니다. 그리고 실제 광고를 집행할 때는 멀티 터치 어트리뷰션 및 순증가분(Incrementality) 분석으로 채널의 광고 효과를 실제로 확인하여 효율을 확인하고 검증 및 모니터링할 수 있습니다. 마지막으로 라스트 터치 어트리뷰션 모델의 분석 결과를 소재 및 캠페인 레벨의 세부 전략 수립에 활용할 수 있습니다.
앱 어트리뷰션 툴 / MMP에서 라스트 터치 어트리뷰션 성과만 확인하면서 이 성과가 진짜 우리 마케팅 성과인지 궁금하셨다면, 멀티 터치 어트리뷰션과 마케팅 믹스 모델링을 통해 통합적인 마케팅 성과 분석 결과를 확인해 보세요.
> 👉 광고 성과를 결정하는 규칙인 어트리뷰션 모델에 대한 보다 자세한 내용은 [에어브릿지 핸드북 01 - 어트리뷰션 ](https://www.airbridge.io/ko/library/airbridge-handbook-attribution)에서 확인해보세요.
---
## 주요 마케팅 KPI를 오버뷰 대시보드에서 한눈에 확인해 보세요
URL: https://www.airbridge.io/ko/blog/overview-dashboard
Language: Korean
Published: 2023-11-22
에어브릿지에 등록한 앱들의 마케팅 KPI(핵심 성과 지표)를 다양한 차트와 테이블로 한눈에 확인할 수 있는 오버뷰 대시보드(Overview Dashboard)를 공개합니다. 이제 여러 앱의 데이터를 포괄한 전체 성과를 하나의 화면에서 쉽게 모니터링 할 수 있습니다. 오버뷰 대시보드에서 우리 앱의 데이터로부터 얻을 수 있는 인사이트를 빠르게 획득하고 마케팅 성과를 최적화해보세요.
## 에어브릿지 오버뷰 대시보드란,
에어브릿지와 같은 앱 어트리뷰션 툴 / MMP는 파편화 되어 있는 앱 마케팅 성과를 통합적으로 분석하기 위해 반드시 필요합니다. 에어브릿지 사용을 시작하기 위해 가장 먼저 해야할 일 중 하나는 에어브릿지 대시보드에 광고 성과를 확인하고 싶은 서비스를 등록하는 것입니다. 에어브릿지는 등록된 앱에 대한 마케팅 성과를 다양한 리포트를 통해 제공합니다.
👉 [에어브릿지 시작을 위해 새로운 앱 등록하기](https://help.airbridge.io/ko/guides/register-a-new-app)
그리고 **등록한 앱들로부터 수집한 데이터를 ‘오버뷰 대시보드’에서 메트릭 위젯, 차트, 테이블로 더 쉽게 광고 성과를 확인할 수 있습니다.**
[]
### 메트릭 위젯
최상단에 위치한 메트릭 위젯에서는 선택한 앱의 KPI(핵심 성과 지표)를 다양한 메트릭으로 확인할 수 있습니다. 여러 앱의 광고 성과를 같은 메트릭으로 확인해 비교하거나 특정 앱의 광고 성과를 여러 메트릭 설정으로 한 번에 확인하는 것도 가능합니다. 특히 증감률을 표시하기 때문에 직관적으로 데이터를 확인하여 즉각적인 인사이트를 얻을 수 있습니다.
### 차트와 테이블
오버뷰 대시보드에서는 다양한 차트와 테이블로 설정한 기간 동안의 광고 성과를 시각적으로 확인할 수 있습니다. 총 6개의 시계열 차트와 비시계열 차트를 동시에 제공하며 1개의 테이블도 제공합니다. 모든 차트와 테이블은 여러 앱을 동시에 선택해서 광고 성과를 한 번에 확인할 수 있습니다.
차트와 테이블의 데이터를 보다 자세하게 확인하고 싶다면 우측 상단의 리포트 명을 클릭하여 해당 리포트에서도 같은 설정으로 성과를 확인할 수 있습니다.
[]
## 한눈에 지표를 확인할 수 있는 대시보드가 필요한 이유
에어브릿지 오버뷰 대시보드를 활용하면 여러 앱 혹은 하나의 앱을 운영하고 있는 경우 모두 앱의 주요 지표 데이터를 한눈에 확인하고 인사이트를 얻을 수 있다는 장점이 있습니다. 특히 외부 BI 툴로 데이터를 연결시켜서 확인하지 않아도 직관적으로 마케팅 성과를 확인할 수 있습니다.
### 여러 앱의 광고 성과 한눈에 확인하기
[]
여러 앱을 동시에 운영하고 있다면 확인해야 하는 지표와 상황이 늘어나기 때문에 지표 분석의 난이도가 높아집니다. 그렇기 때문에 에어브릿지 오버뷰 대시보드와 같이 여러 앱의 주요 지표를 한 번에 확인하게 되면 그렇지 않은 경우에 비해 우위에 오를 수 있습니다. 어느 앱이 보다 큰 성장의 가능성을 보이는지 빠르게 비교 및 파악하면 이에 따라 마케팅 전략 및 예산을 빠르게 수정 반영하여 성과를 효과적으로 개선할 수 있습니다.
### 단일 앱의 광고 성과 빠르게 직관적으로 확인하기
여러 앱은 물론 하나의 앱에 대한 마케팅을 운영하고 있는 경우에도 한 화면에서 다양한 데이터를 빠르게 확인하고 인사이트를 얻을 수 있다는 점에서 유리합니다.
게다가 메트릭 위젯을 활용하면 특정 앱의 광고 성과를 이전 기간과 비교하여 증감률로 빠르게 확인할 수 있습니다. 메트릭 위젯에 자주 보는 KPI를 메트릭으로 설정해두고 매일 업데이트되는 성과를 편리하게 확인해 보세요.
[]
더 자세한 오버뷰 대시보드 설정 및 활용 방법은 아래 링크와 연결된 유저 가이드 문서에서 확인할 수 있습니다. 에어브릿지 오버뷰 대시보드를 통해 우리 앱들의 데이터를 포괄하여 빠르게 성과를 모니터링하고 인사이트를 얻어 보세요.
[]
---
## 2023 방치형 RPG 트렌드: 다운로드 및 매출 순위
URL: https://www.airbridge.io/ko/blog/idle-rpg-trend-2023
Language: Korean
Published: 2023-11-08
한국의 RPG에 대한 많은 관심과 사랑은 이미 잘 알려져 있습니다. [센서타워(Sensor Tower)](https://sensortower.com/ko)의 데이터에 따르면 2023년 1월부터 8월까지 전체 모바일 RPG 매출 중 한국에서 세 번째로 많은 비중을 차지합니다. 한국의 전체 모바일 게임 매출 중 57.5%가 RPG에서 발생하는데, 미국, 중국, 일본에서는 RPG가 각각 11.3%, 27%, 47.8%를 차지하는 것에 비해 상당히 높은 수치를 보여줍니다. 전통적으로 한국은 특히 MMORPG가 많은 인기를 선도하는 것으로 알려져 있습니다. 이와 더불어 최근 RPG의 하위 장르 중 스쿼드 및 방치형 RPG가 점점 더 많은 인기를 끌며 RPG에 대한 전반적인 관심도 및 매출도 증가하고 있습니다. (출처: [센서타워 블로그](https://sensortower.com/ko/blog/Korea-stands-out-as-the-leading-country-with-RPG-revenue-constituting-approximately-60-percent-of-total-mobile-game-revenue))
이번 블로그 포스트는 급격한 성장세를 보이는 방치형 RPG의 성장 트렌드를 소개합니다. 데이터로 드러나는 방치형 RPG 트렌드에서 인사이트를 얻고 성장의 흐름에 합류해보세요.
**모바일 앱 마케팅 분석 및 인텔리전스 플랫폼 센서타워(Sensor Tower)의 데이터를 바탕으로 에어브릿지에서 작성한 콘텐츠입니다.*
## 방치형 RPG란?
RPG는 롤플레잉 게임(Role-playing games)으로, 하위 장르로 MMORPG(대규모 멀티 플레이어 온라인 롤플레잉 게임, Massive Multi-player Online Role Playing Game), 팀 배틀, 방치형 RPG 등으로 세분화합니다.
이 중에서도 방치형 RPG는 플레이어가 게임을 직접 활발하게 플레이하지 않아도 게임이 계속 진행되는 장르를 의미합니다. 유저는 캐릭터의 레벨업, 장비 획득, 임무 완수 등을 통해 게임 세계를 탐험하고 강화시키며 게임을 플레이합니다. 이때 자동 플레이 시스템에 기반하여 유저가 능동적으로 참여하거나 상호작용하지 않더라도 게임 캐릭터나 시스템이 자동으로 작업을 수행하거나 진행할 수 있습니다. 가만히 두어도 계속해서 게임이 진행된다는 점에서, 일상과 더불어 게임을 쉽게 진행할 수 있기 때문에 모바일에서 주로 플레이되고 있습니다.
## 한국에서 두드러지는 방치형 RPG의 성장세
방치형 RPG 게임은 최근 4년 간 지속적으로 성장하고 있습니다. 2020년 1월부터 9월까지, 그리고 2023년 1월부터 9월까지 4년 간 동기간의 매출 및 다운로드 데이터를 확인해보았을 때 지속적인 성장 추세를 발견할 수 있었습니다.
**참고: 연도별 비교를 위해 매년 1월부터 9월까지 데이터를 기준으로 했습니다. 전체 게임 시장의 순위는 86개의 센서타워 게임 택소노미 하위 장르 중 매출 또는 다운로드 수 기준 순위를 나타냅니다. RPG 하위장르의 순위는 7개의 센서타워 게임 택소노미 RPG 하위 장르 중 매출 또는 다운로드 수 기준 순위를 나타냅니다.*
[]
전체 게임 시장에서 방치형 RPG의 다운로드 수에 따른 순위는 2020년 1월부터 9월까지 69위였으나 2023년 동기간 동안 59위로 상승했습니다. RPG 하위장르에서도 방치형 RPG의 다운로드 비중이 2020년 1월부터 9월까지 7.5%였으나, 2023년 동기간에 10.4%로 비중이 늘어나기도 했습니다.
[]
이와 더불어 특히 국내에서는 보다 급격하게 성장하는 추세를 보이고 있습니다. 전체 게임 시장에서는 물론 RPG 하위 장르에서도 방치형 RPG의 성장세가 두드러집니다. 센서타워의 데이터에 따르면 전체 게임 시장에서 다운로드 순위가 42위에서 2위로 올랐습니다. RPG 장르 내에서 방치형 RPG의 경우 다운로드 순위가 2020년 4위에서 2023년 1위로 오르며 7.3%에서 28.2%의 비중으로 급격한 증가세를 보입니다. 매출 측면에서도 국내 전체 게임 시장에서 방치형 RPG는 2020년 1월부터 9월까지 16위였으나 2023년 동기간에는 5위로 오릅니다.
## 2023년 가장 인기 있는 방치형 RPG
그렇다면 2023년 올해 1월부터 9월까지 가장 인기 있는 방치형 RPG에는 어떤 게임들이 있는지 누적 매출과 다운로드 기준의 순위로 살펴보겠습니다.
### 누적 매출 기준 방치형 RPG 인기 순위
[]
**위 매출 데이터는 App Store와 Play Store에서 발생한 인앱 구매(IAP)에 대한 추정치로 광고 수익 및 제3자 안드로이드 마켓은 포함하지 않습니다. *
위의 이미지는 2023년 1월부터 9월까지 누적 매출 기준의 글로벌 및 국내 방치형 RPG의 인기 순위입니다.
먼저 글로벌 순위에서 주요 게임들에 대해 살펴보면 1위 ‘Hero Wars: Alliance’와 2위 ‘Girls Chronicle: Idle Heroine’은 지난 몇 년간 전 세계 방치형 RPG 매출 상위권을 차지하고 있습니다. ‘Hero Wars: Alliance’는 2020년 1위, 2021년과 2022년에는 2위, 그리고 2023년에는 다시 1위 자리에 올랐습니다. ‘Girls Chronicle: Idle Heroine’ 또한 2020년 2위, 2021년 1위, 2022년과 2023년에는 2위에 랭크되었습니다. 한국에서 출시한 게임 ‘레전드 오브 슬라임’, ‘아처키우기’의 글로벌 시장에서의 성과도 눈에 띕니다.
여기서 특히 ‘Hero Wars: Alliance’는 지난 2016년 출시 후 전 세계 누적 매출이 10억 달러를 넘어섰습니다. 미국에서의 누적 매출이 38%로 가장 큰 비중을 차지하고 있으며 일본(14.1%), 독일(14.1%) 등이 그 뒤를 잇습니다. ‘Girls Chronicle: Idle Heroine’은 지난 2017년 출시 이후 전 세계 누적 매출이 12억 달러를 넘어섰는데 99.9%가 일본에서 창출되었습니다.
한편 국내에서는 ‘픽셀 히어로’, ‘세븐나이츠 키우기’와 같은 올해 새롭게 출시된 게임이 누적 매출 기준 인기 순위 상위권에 올랐습니다. ‘픽셀 히어로’는 국내 방치형 RPG 매출 1위에 올랐을 뿐만 아니라 같은 기간 국내 전체 모바일 게임 순위에서도 25위에 랭크되었습니다. 2위에 오른 넷마블의 ‘세븐나이츠 키우기’는 1위 ‘픽셀 히어로’보다 약 3개월 늦게 출시되었다는 점을 감안하면 괄목할 만한 성과라고 할 수 있습니다.
### 누적 다운로드 수 기준 방치형 RPG 인기 순위
[]
**위 다운로드 데이터는 App Store와 Play Store에서 발생한 다운로드 추정치로 사전 다운로드, 재다운로드 및 제 3자 안드로이드 마켓은 포함하지 않습니다.*
글로벌에서는 ‘Hero Wars: Alliance’가 매출 순위에 이어 다운로드 순위에서도 1위에 올랐습니다. 2022년에 출시된 한국산 게임 ‘레전드 오브 슬라임: 방치형 RPG’가 2위를 차지했는데, 2023년 1월과 4월에 발생한 다운로드 수 급증이 이러한 성과를 견인했습니다.
한편 국내에서는 유조이 게임즈의 ‘픽셀 히어로’는 한국 시장 방치형 RPG 매출, 다운로드 순위 모두에서 1위에 올랐습니다. 그리고 Joy Net Games의 ‘개판오분전’이 2위에 오르면서 1, 2위 모두 해외 퍼블리셔가 퍼블리싱하는 방치형 RPG가 차지했습니다.
한국 퍼블리셔 게임베리 스튜디오(‘헌터 키우기: 방치형 RPG’, ‘조선 좀비전’)와 모비릭스(‘임모탈 키우기’, ‘무림소녀 키우기’)는 각각 자사가 서비스하는 두 개의 게임을 상위권에 올려놓았습니다.
## 방치형 RPG가 유저들의 주목을 받는 이유
방치형 RPG가 인기를 끄는 이유는 주요 인기 게임들의 유저 리뷰를 기반으로 이해해 볼 수 있습니다.
센서타워 리뷰 분석에 따르면 2023년 한국 매출 및 다운로드 순위권을 차지한 ‘픽셀 히어로’, ‘세븐나이츠 키우기’, ‘레전드 오브 슬라임’, ‘개판 오분전’의 평점 4-5점 리뷰에서 가장 많이 언급된 주요 키워드는 아래와 같습니다.
[]
여기서 네 게임을 공통으로 특징짓는 키워드 중 하나는 바로 ‘귀여움’입니다. 이들 게임은 매력적인 귀여운 캐릭터와 배경을 제공함으로써 트렌드를 반영하고 있습니다. 귀엽고 친근한 캐릭터들과 상황은 게임에 대한 접근 장벽을 낮추는 효과도 있습니다.
게임 개발자들은 이러한 ‘귀여움’을 적극적으로 활용하고 있으며, 플레이어들이 게임 세계에 더 많이 몰입하고 더 긴 시간 동안 게임을 즐길 수 있도록 노력하고 있습니다. 이러한 인사이트로부터 얻을 수 있는 것은, 게임이 유저들에게 사랑받으며 성장하기 위해서는 귀여움을 효과적으로 풀어내야 한다는 것입니다.
[]
이번 포스트에서 소개한 방치형 RPG의 성장세와 트렌드를 바탕으로 성장하는 게임들에 대한 인사이트를 확인해볼 수 있었습니다. 콘텐츠 소비 시간이 짧아진 ‘스낵 컬처’의 유행과 더불어 방치형 RPG는 앞으로도 다양한 형태 및 플랫폼 속에서 더 많은 플레이어들에게 사랑 받을 것으로 예상됩니다.
에어브릿지는 방치형 RPG는 물론 모든 모바일 게임의 성장에 함께하고 있습니다. 에어브릿지의 모바일 게임을 위한 지원이 궁금하다면 여기에서 확인해보세요.
👉 [에어브릿지의 모바일 게임을 위한 지원 더 자세히 알아보기](https://www.airbridge.io/ko/solutions/gaming)
> 에어브릿지(Airbridge)는 게임 마케팅 성과 분석 툴(MMP)입니다. CAC, LTV와 같은 수익성 지표를 하나의 대시보드에서 손쉽게 확인할 수 있으며, 리텐션 분석, 퍼널 분석, 프로드 방지, 오디언스 커넥션과 같은 다양한 기능을 제공하여 게임사가 신규 사용자 확보(UA)와 수익 극대화를 할 수 있도록 도움을 줍니다.
---
## 모바일 게임 앱에서 퍼널 분석하기
URL: https://www.airbridge.io/ko/blog/funnel-analysis-for-gaming-apps
Language: Korean
Published: 2023-10-19
수많은 게임 앱이 유저의 주목을 끌기 위해 치열하게 경쟁합니다. 지금도 계속해서 시장에 새로운 게임이 등장하고 또 사라지고 있습니다. 이러한 경쟁 속 수익을 남기며 이름을 알리는 게임이 되기 위해 다양한 전략이 있겠지만 결국 정답은 ‘유저’에게 있습니다. 즉 유저의 게임 내 행동을 분석하고 이를 기반으로 인사이트를 얻고 개선해가야 합니다.
많은 퍼블리싱사에서 게임 정식 론칭 전에 게임성을 검증하기 위해서 오픈 베타 테스트 혹은 소프트 런칭이라고 불리는 테스트를 진행합니다. 이때 게임 다운로드 및 실행 전부터 후까지 유저의 여정을 세분화하여 분석하는 속에 어떤 단계가 허들이 되는지 파악할 수 있습니다. 유저의 행동 속에서 개선 지점에 대한 인사이트를 얻을 수 있습니다. 이 과정이 바로 ‘퍼널 분석(Funnel Analysis)’입니다.
이번 포스트에서는 모바일 게임 앱에 퍼널 분석이 왜 중요한지, 그리고 어떻게 활용할 수 있을지 알아봅니다. 퍼널 분석을 통해 게임을 성장시키고 유저들의 만족도를 높이는 방법을 이해하는 것은 모바일 게임 경쟁에서 돌파구가 될 수 있습니다.
## 퍼널 분석이란?
퍼널 분석은 앱 내에서 유저 행동을 추적해서 특정한 목표까지의 여정을 시각화하고 분석하는 과정을 의미합니다. 시작점(예: 앱 다운로드)에서 목표 달성 지점(예: 아이템 구매)까지의 유저 행동을 단계별로 추적하여 각 단계의 전환율을 측정하고 분석할 수 있습니다. 이를 통해 어디에서 유저가 이탈하거나 문제가 발생하는지를 이해하여 유저와 인게이지하고 전환율을 개선하기 위한 인사이트를 얻을 수 있습니다.
## 모바일 게임에 퍼널 분석이 필요한 이유
모바일 게임 앱에서 퍼널 분석은 유저 행동에 대한 이해를 기반으로 마케팅 전략 최적화와 유저 경험 개선, 그리고 수익을 높이는 데에 효과적입니다.
### 1. 프라이버시 강화 시대에 마케팅 효율 개선
과거에는 행동 데이터를 기반으로 한 유저 타겟팅이 비교적 쉽게 가능했습니다. 하지만 현재는 프라이버시 강화와 관련된 규제 및 유저의 프라이버시 의식이 높아지면서 트래킹이 어려워지고 있습니다. 이로 인해 대부분의 앱 마케팅 전략은 전환할 가능성이 높을 것 같은 특정 타겟을 유입시키기 보다는 유저를 유입시킨 후의 행동을 분석하여 효과적인 마케팅 채널과 캠페인을 식별하는 방향으로 변화하고 있습니다.
이때 퍼널 분석을 통해 각 마케팅 채널과 캠페인 별로 유입된 유저들의 행동을 추적하고 전환율을 파악하여 마케팅 전략을 최적화할 수 있습니다. 효과적인 채널과 캠페인에 예산을 더 투입하여, 마케팅 예산을 효율적으로 분배하고 타겟 유저에 대한 도달률을 높일 수 있습니다. 프라이버시 강화 시대 퍼널 분석이 UA(User Acquisition) 마케팅 전략 수립과 효율 개선에 도움을 줄 수 있습니다.
### 2. 유저의 행동을 이해하고 유저 경험 개선
퍼널 분석은 모바일 게임 앱에서 유저의 행동을 체계적으로 파악하기에 효과적입니다. 게임 내에서 유저가 어떻게 상호작용하며 게임을 진행하는지를 단계별로 추적할 수 있습니다. 이를 통해 어떤 단계에서 특히 어려움을 겪거나 시간을 많이 소비하고 이탈하는지를 파악할 수 있으며 유저들의 의도한 대로 게임을 진행하고 있는지도 확인할 수 있습니다. 여기서 얻은 인사이트를 바탕으로 개선할 지점을 파악하여 게임을 보다 유저 친화적으로 만들고 더 오래 즐기도록 유도할 수 있습니다.
특히 오픈 베타 테스트 혹은 소프트 런칭 단계에서 활용하면 가장 효과적입니다. 앱의 본격적인 출시 전 유저가 앱을 이탈하는 지점을 파악하고 개선 사항을 도출하여 실제 마케팅 집행에 필요한 인사이트를 얻을 수 있기 때문입니다. 예를 들어, 게임 튜토리얼 단계에서 많은 유저가 이탈한다면, 튜토리얼을 개선하거나 가이드를 보다 명확하게 제공함으로써 이탈율을 줄일 수 있습니다. 특히 ‘초기 이탈 구간’을 개선하게 되면 리텐션과 ROAS에도 긍정적인 영향으로 이어질 수 있기 때문에 반드시 염두해야 합니다.
## 퍼널 분석을 위한 각 단계 설정하기
퍼널 분석을 진행하기 위해서는 먼저 어떤 유저 행동을 개선하거나 이해하고 싶은지 등을 질문하며 분석 목적과 목표를 명확하게 설정해야 합니다. 예를 들어 분석 목표가 마케팅 채널별 혹은 국가별 아이템 구매 전환율 개선, 구독 전환율 개선 등인지에 따라 설정해야하는 퍼널 단계와 코호트 기준이 달라질 수 있습니다. 분석 목적과 목표를 설정하고 나면 이를 해결하기 위해 각 퍼널 단계 별로 적합한 KPI, 데이터, 이벤트 등을 정의합니다.
### 목표에 따른 퍼널 단계 설정 예시
예를 들어, 지난 7일 간 광고를 클릭하여 유입된 유저의 아이템 구매 전환율을 확인하고 채널별 성과를 최적화하고 싶은 경우 에어브릿지에서는 아래와 같이 단계를 설정하여 분석할 수 있습니다.
[]
[]
### 퍼널 단계 정의할 때 주의해야할 점
퍼널 단계를 정의할 때는 다음의 두 가지를 주의해야 합니다. 먼저, 유저들의 행동이 퍼널 상 다음 단계로만 나아가고 이전 단계로는 돌아가지 않아야 합니다. 즉, 일방향이어야 합니다. 위의 예시와 같이 단계를 설정하여 분석하기 위해서는 게임의 설정 상 회원가입 후에 아이템을 구매할 수 있어야 합니다. 또한, 각 단계의 명칭은 결과 해석 시 모두가 같은 맥락에서 결과를 이해할 수 있도록 통일성을 갖추고 분명하게 명명해야 합니다.
퍼널 분석 단계 설정이 완료되었다면 정의한 단계에 따라 필요한 데이터를 수집하고 분석을 시작하면 됩니다.
## 에어브릿지의 퍼널 분석 기능으로 데이터 분석하기
이 퍼널 분석 기능은 프로덕트 분석 툴 외에도 광고부터 이어지는 앱 내 여정을 분석하기 위해 MMP/어트리뷰션 툴을 사용할 수도 있습니다. 에어브릿지가 제공하는 퍼널 리포트 및 기능은 대부분의 MMP에서는 제공하지 않으며, 에어브릿지가 유일합니다.
에어브릿지는 MMP로서 매체가 제공한 어트리뷰션 정보 및 에어브릿지가 수집한 데이터를 바탕으로 광고부터 앱 내까지 이어지는 유저 여정을 분석할 수 있는 퍼널 리포트를 제공합니다. 특히 에어브릿지가 자체 개발한 데이터베이스 LUFT 덕분에 특정 코호트를 빠르게 생성할 수 있습니다. 그룹바이 기능을 통해 특정 기준으로 나뉜 코호트 간의 채널별, 국가별, 기간별, OS 별 행동 패턴도 분석할 수 있습니다.
👉 [에어브릿지가 제공하는 퍼널 분석 기능에 대한 자세한 내용은 가이드에서 확인해보세요.](https://help.airbridge.io/ko/guides/funnel-report)
### MMP로 퍼널을 분석할 수 있는 것의 장점
MMP로 퍼널을 분석하게 되면 프로덕트 분석 툴에서는 확인할 수 없는 SAN(Self Attributing Network) 매체까지 포함한 모든 광고 데이터까지 확인할 수 있다는 것이 가장 큰 장점입니다. 일반적으로 SAN 매체는 많은 유저의 데이터를 관리하기 때문에 자체적으로 엄격한 데이터 규정을 가지고 있습니다. 그래서 파트너십을 통해 데이터를 관리할 수 있는 MMP에게만 데이터를 제공하고 있으며, 파트너십이 맺어지지 않은 써드파티에는 해당 데이터를 보낼 수 없습니다. 즉 프로덕트 애널리틱스 툴과 MMP를 연동했다고 하더라도 프로덕트 애널리틱스 툴에서는 SAN 매체의 어트리뷰션 데이터를 포함한 고객 여정은 분석할 수 없습니다. 그러나 MMP로서 퍼널 리포트를 제공하는 에어브릿지는 이러한 한계를 극복할 수 있습니다.
모바일 게임 마케팅을 한다면 반드시 필요한 MMP를 통해 퍼널 분석까지 진행하여 보다 효율적인 마케팅 전략을 준비해보세요.
📌 [에어브릿지의 모바일 게임 마케팅 전문가와 상담하기](https://www.airbridge.io/ko/book-a-demo)
---
## 에어브릿지, 마케팅 믹스 모델링 분야의 메타 비즈니스 파트너 선정
URL: https://www.airbridge.io/ko/blog/airbridge-marketing-mix-modeling-meta-business-partner
Language: Korean
Published: 2023-10-12
앱 어트리뷰션 툴 / MMP 에어브릿지가 마케팅 믹스 모델링(Marketing Mix Modeling) 분야의 메타 비즈니스 파트너(Meta Business Partner)가 되었습니다. 지난 2021년 [모바일 측정(MMP)과 멀티터치 기여도분석(MTA) 분야의 비즈니스 파트너로 동시에 선정](https://www.airbridge.io/ko/blog/airbridge-becomes-facebook-business-partner-for-measurement)된 데 이어 세 번째 비즈니스 파트너로 선정되었습니다. 이로써 에어브릿지는 모바일 성과 측정 분야에서의 전문성과 역량을 다시 한번 입증하였습니다.
2023년 5월, 에어브릿지는 마케팅 믹스 모델링의 접근성을 높이고자 [MMM 스튜디오](https://www.airbridge.io/ko/mmm-studio)를 출시했습니다. MMM 스튜디오는 머신러닝 기반의 플랫폼으로 온·오프라인의 다양한 마케팅 채널이 비즈니스 성과에 미치는 정도를 수치화하여 보다 정확한 성과 측정을 돕습니다. 유저 레벨 데이터가 아닌 집계 데이터를 기반으로 하기 때문에 개인정보 보호에 특화된 솔루션이며, 계절성, 프로모션, 경쟁사 활동 등의 변수를 고려해 마케팅 효과에 대한 통합적인 분석과 인사이트를 제공합니다. 자동화된 온보딩, 고도로 구현 가능한 모델, 그리고 쉽게 접근할 수 있는 대시보드 등 편리하고 효율적인 마케팅 믹스 모델링 사용으로 데이터 기반 마케팅 최적화를 돕습니다.
게다가 MMM 스튜디오는 마케팅 채널별 성과를 예측하고 시뮬레이션함으로써 단순한 추측이 아닌 데이터 기반의 최적의 예산안 분배를 제안합니다. 이러한 실행 가능한 인사이트를 바탕으로 변화하는 시장 상황에 빠르게 적응할 수 있습니다.
에어브릿지의 CEO 이자 공동 창업자인 남성필 대표는 “개인정보 보호 시대가 도래하며 각 기업들은 이를 위한 최적의 마케팅 성과 측정 솔루션을 필요로 하고 있다.”라며 “메타의 마케팅 믹스 모델링 파트너로서 고객사에 마케팅 성과에 대한 완전한 이해를 제공할 수 있게 되어 기쁘다. 급변하는 디지털 환경에서 최상의 결과를 제공하는 혁신적이고 통합적인 솔루션으로 거듭나겠다”라고 밝혔습니다.
- 에어브릿지 더 알아보기: [https://www.airbridge.io/ko](https://www.airbridge.io/ko)
- 에어브릿지 MMM 스튜디오 경험하기: [https://www.airbridge.io/ko/mmm-studio](https://www.airbridge.io/ko/mmm-studio)
## 에어브릿지란?
에어브릿지는 유저의 행동 및 광고의 기여도를 측정하고 분석하는 모바일 성과 분석 플랫폼입니다. 라스트 터치 어트리뷰션, 멀티 터치 어트리뷰션 및 마케팅 믹스 모델링으로 다각도로 완전한 마케팅 성과 분석을 제공합니다. 또한, 앱과 웹사이트의 마케팅 성과를 하나의 대시보드에서 제공하며 딥링크, 오디언스 생성, 프로드 방지, 데이터 추출 등 모바일 성장에 필요한 모든 기능의 지원으로 모바일 앱 성장을 극대화합니다.
---
## 강화된 에어브릿지-유니티 애즈 연동으로 분석에 깊이를 더하세요
URL: https://www.airbridge.io/ko/blog/airbridge-unity-ads-integration
Language: Korean
Published: 2023-09-13
시시각각 변화하는 모바일 마케팅 환경에서는 '속도'가 관건입니다. 이에 에어브릿지는 고객사가 빠르고 시의적절하게 데이터에 기반하여 의사 결정할 수 있도록 글로벌 앱 수익화 및 광고 플랫폼 [유니티 애즈(Unity Ads)](https://unity.com/kr/products/unity-ads)와의 연동을 강화했습니다.
기존에도 에어브릿지에서 유니티 애즈를 통해 진행한 마케팅 캠페인의 성과를 확인할 수 있었습니다. 하지만 최근 두 플랫폼 간 연동이 강화됨에 따라 에어브릿지에서 유니티 애즈의 성과뿐 아니라 비용까지 통합적으로 분석할 수 있게 되었습니다. 더이상 수동으로 데이터를 다운받아 엑셀이나 스프레드시트에서 관리할 필요가 없기 때문에 분석 시간이 눈에 띄게 단축됩니다. 또한, 에어브릿지에서 실시간 데이터를 확인할 수 있다 보니 흐름이 끊기는 일 없이 효율적으로 업무를 진행할 수 있습니다. 무엇보다 유저의 행동 변화를 즉각 반영해 캠페인 방향성을 조정하고 최적의 성과를 낼 수 있습니다.
성과 데이터와 비용 데이터로 마케팅 활동의 비용 효율성도 측정할 수 있습니다. 마케팅 성과는 그 자체로도 의미 있을 수 있지만, 한정된 자원을 현명하게 활용하고 있는지 판단하려면 투자 비용 대비 성과를 추적해야 합니다. 이는 [ROAS](https://www.airbridge.io/ko/blog/measuring-roas)와 ROI가 마케팅에서 핵심 지표로 여겨지는 이유이기도 합니다. 이렇게 비즈니스 목표에 실질적으로 기여하는 마케팅 채널과 캠페인을 파악하고 있으면 의사 결정 과정에서 정확성과 신뢰성, 투명성을 확보할 수 있습니다.
유니티 애즈 연동을 통해 가져온 데이터는 에어브릿지 리포트에서 Clicks, Impressions, Cost등의 메트릭을 적용하여 조회할 수 있습니다. 또한, Campaign, Ad Group, Ad Creative, Device Country을 비롯한 기준으로 데이터를 분류하여 볼 수 있습니다.
> 👉 구체적인 연동 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/unityads)에서 확인하세요.
에어브릿지는 효과적이고 심층적인 마케팅 성과 분석 서비스를 제공하기 위해 여러 플랫폼과의 연동을 확대해 나가고 있습니다. 에어브릿지와 함께 [모바일 앱 성장을 극대화하세요](https://www.airbridge.io/ko/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=airbridge-unity-ads-integration)!
---
## 오가닉 웹&앱 마케팅 성과를 에어브릿지 오가닉 리퍼러 어트리뷰션으로 확인해 보세요
URL: https://www.airbridge.io/ko/blog/organic-referrer-attribution
Language: Korean
Published: 2023-09-07
모바일 앱 UA 캠페인을 한다면, 앱 설치와 앱 내 전환이 궁극적인 목표입니다. 하지만 에어브릿지 내부 리서치에서 모바일 앱 UA 캠페인 시 유저를 웹으로 먼저 보내는 경우가 60%라는 사실을 발견했습니다. 앱 설치와 앱 내 전환을 목표로 하더라도 전체 캠페인의 절반 이상에서 유저를 웹으로 먼저 유도하는 것이죠. 앱 마켓의 지면은 한정적이기 때문에 모바일 웹에서 보다 풍부한 콘텐츠와 인센티브를 제공하고자 웹 캠페인을 집행하는 경우가 많아졌습니다. 위와 같은 이유로 인해 웹투앱 캠페인이 각광받고 있으며 웹 캠페인으로 발생한 앱 설치 및 앱 내 전환과 같은 앱 성과를 측정하는 것이 보다 중요해졌습니다.
👉 [전환율을 높이는 웹투앱 경험 구축하기](https://www.airbridge.io/ko/library/web-to-app-conversion-whitepaper)에서 웹투앱 캠페인 더 자세히 알아보기
먼저 모바일 웹으로 유저를 유입시키기 위해서는 크게 두 가지 전략을 활용할 수 있는데 우선 검색 광고(Search Ads)와 같은 페이드(Paid) 마케팅을 집행하거나, SEO(Search Engine Optimization, 검색엔진 최적화)를 통해 자연적으로 웹 방문 유저를 늘릴 수 있습니다. 페이드 마케팅의 경우 각 매체에서 제공하는 대시보드를 활용하거나[ 광고 비용 데이터를 연동한 MMP](https://www.airbridge.io/ko/blog/airbridge-x-kakaokeyword-cost)에서 데이터를 확인할 수 있습니다. 다만 오가닉 유저에 대해서는 그 여정을 파악하기가 쉽지 않았습니다. 이러한 페인 포인트 해결을 위해 이제 에어브릿지에서 웹사이트를 통해 자연적으로 유입된 유저 여정까지 파악할 수 있도록 **오가닉 리퍼러 어트리뷰션(Organic Referrer Attribution)** 기능을 제공합니다.
> 💡 **주의**
오가닉 리퍼러 어트리뷰션 기능은 사용을 원하는 고객사에만 제공하고 있습니다. 사용을 원하는 경우 에어브릿지에 문의해 주세요.
## 오가닉 리퍼러 어트리뷰션이란?
에어브릿지는 오가닉 리퍼러 어트리뷰션 기능을 통해 포털 사이트의 검색 결과나 온드(Owned) 미디어에서 우리 웹사이트 혹은 앱으로의 유저 획득과 전환을 올바르게 기여하고 그 데이터를 대시보드에서 제공합니다.
기존에는 페이드 마케팅 외의 경로로 발생한 전환은 대시보드에서 *미기여(Unattributed)*로 확인할 수 있을 뿐 어느 웹사이트에서 유입되었는지 파악할 수 없었습니다. 하지만 에어브릿지의 오가닉 리퍼러 어트리뷰션 기능을 활용하면 리퍼러가 제공되는 모든 웹사이트에서 성과를 측정할 수 있습니다. 리퍼러의 기본적인 구조는 웹 주소와 비슷하나 상세 구조는 웹에 따라 다릅니다. 기본적인 구조는 아래와 같습니다.
- 리퍼러의 기본적인 구조 = 호스트, 경로, 쿼리 파라미터, 해시
에어브릿지는 Web SDK로 리퍼러 데이터를 수집하여 오가닉 리퍼러 어트리뷰션을 제공합니다.
검색 엔진에서 웹사이트에 접속하게 되면, 리퍼러의 각 구성 요소에 해당 사이트에 대한 정보가 담깁니다. 에어브릿지는 이 리퍼러의 정보를 집계해 SEO 및 온드 미디어의 성과를 측정하여 대시보드에서 제공합니다. 에어브릿지의 경우 APP SDK 외에도 Web SDK가 있기 때문에 유저가 웹사이트에 랜딩할 때 리퍼러 값을 수집합니다. 즉 유저가 네이버, 다음, 구글 등 국내외 모든 검색 엔진에서 웹사이트를 검색하여 클릭하거나, 다른 웹사이트들에 게재되거나 공유된 링크를 통해 웹사이트에 접속한 경우 해당 리퍼러 값을 통해 트래킹할 수 있습니다. 이를 통해 어떤 채널과 키워드로 오가닉 유저가 많이 유입되는지 파악할 수 있으며 이 데이터를 마케팅 캠페인 예산 분배에 활용할 수 있습니다.
### 오가닉 리퍼러 어트리뷰션 사용 사례
예를 들어 아래와 같은 경우에 대해 어트리뷰션이 가능하며 **앱과 웹사이트가 모두 존재하는 A 프로덕트**의 마케터는 **어떤 채널이 얼마만큼 기여하고 있는지** 에어브릿지에서 확인할 수 있습니다.
> 💡 오가닉 리퍼러 어트리뷰션을 사용하기 위해서는 웹사이트에 Web SDK가 설치되어 있어야 합니다.
[]
- 유저가 B 포털 사이트에서 ‘탄산수’를 검색 -> 광고 영역 외에 검색 결과로 A 웹사이트가 노출 -> A 웹사이트 링크를 클릭 -> A 웹사이트 내에서 발생하는 성과(구매, 앱 설치 등)를 모두 B 포털 사이트 성과로 기여
[]
- C 프로덕트의 마케터가 공식 블로그에 C 웹사이트 링크를 포함한 포스트를 게재 -> 유저가 블로그에서 C 웹사이트 링크를 클릭 -> C 웹사이트 내에서 발생하는 성과(구매, 앱 설치 등)를 모두 공식 블로그의 성과로 기여
## 이제 에어브릿지에서 페이드 마케팅 성과뿐만 아니라 오가닉 마케팅 성과도 확인해 보세요.
[]
에어브릿지의 오가닉 리퍼러 어트리뷰션 기능은 다음의 두 가지 측면에서 앱 마케팅의 효율화에 기여할 수 있습니다. 우선, 웹에서 유입되어 미기여(unattributed)로 기록된 트래픽을 올바른 웹사이트로 어트리뷰션할 수 있기 때문에 그동안 놓쳐졌던 웹에서 앱으로 유입되는 유저의 여정을 보다 상세하게 확인할 수 있습니다.** **이를 통해 전체적인 UA 캠페인에 대한 가시성을 높일 수 있습니다.
또한, 유입 이후의 유저의 행동 그리고 앱 내 성과를 올바르게 기여하여 어떤 웹 채널이 가장 효과적인지 확인할 수 있습니다. 이 데이터를 기반으로 웹을 포함한 모든 채널에 대한 효과적인 미디어 믹스 예산 분배를 통해 마케팅 최적화를 할 수 있습니다.
> 아직 에어브릿지를 사용하고 있지 않다면? 👉 [에어브릿지의 어트리뷰션 전문가와 이야기 나눠보세요.](https://www.airbridge.io/ko/book-a-demo)
에어브릿지를 사용하고 계신 고객사라면? 👉 담당 CSM에게 자세한 사용 방법을 문의해주세요.
---
## 앱 삭제, 지금 보고 계신가요?
URL: https://www.airbridge.io/ko/blog/mobile-app-uninstall-tracking
Language: Korean
Published: 2023-09-04
앱을 설치한 유저는 언제든 기기에서 앱을 제거할 수 있습니다. 설치한 앱을 제거/삭제하는 ‘앱 삭제’는 앱 개발자 및 마케터 입장에서 치명적일 수 있지만 끝은 아닙니다. 앱을 삭제한 유저 중 많은 유저가 복귀합니다. 또는 앱을 다시 설치한 유저가 많지 않더라도 앱을 개선할 수 있는 힌트를 찾을 수 있습니다. 유저에게 좋은 경험을 제공하는 것도 중요하지만 앱을 떠나게 하는 요소를 빠르게 찾고 제거하거나 개선해야 결과적으로 앱이 지속 가능한 성장을 이룰 수 있습니다.
대부분의 마케터는 앱과 웹 모두에서 다양한 이벤트를 수집하고 있습니다. 수집하는 이벤트 목록에 앱 설치, 구매, 구독처럼 광고 성과에 관계된 이벤트는 항상 포함되지만, 앱 삭제는 그렇지 않습니다. 하지만 앱 삭제에는 다양한 힌트가 숨어 있습니다. 앱 삭제를 트래킹하면 앱을 설치한 유저 중 얼마나 많은 유저가 앱을 삭제했는지 조회할 수 있습니다. 앱 삭제를 트래킹하고 이 힌트를 바탕으로 유저가 겪는 불편을 빠르게 해결해 보세요.
## 유저는 앱을 삭제했지만, 유저 경험은 끝나지 않습니다
앱 시장의 경쟁은 계속해서 치열해지고 있습니다. 새로운 카테고리의 앱도 계속해서 등장하고 있고 인기 있는 카테고리에는 비슷한 앱 수십 수백 개가 있습니다. 하루에도 수많은 앱이 삭제됩니다. 마케터가 유저의 앱 설치 수를 액면 그대로 받아 들이면 안 되는 이유입니다.
다행인 점은 유저의 앱 삭제는 ‘영원한 삭제’를 의미하지는 않습니다. 크리테오의 2020년 글로벌 앱 설문 조사에 따르면 한국 유저의 38%는 3개월 안에 삭제한 앱을 다시 설치했다고 응답했습니다. 1번이라도 앱을 설치한 유저 10명 중 4명은 앱을 다시 설치하는데 마음이 열려있다는 의미입니다. 유저 경험은 앱을 삭제해도 계속해서 이어집니다.
하지만 다시 앱을 설치한 유저가 앱을 삭제하기 전에 느낀 아쉬움이나 겪은 불편함이 반복된다면? 또는 더 뛰어난 다른 앱이 있다면? 유저는 우리 앱을 찾는 횟수가 줄어들고 결국 다시 앱을 삭제하게 됩니다. 따라서 유저가 앱을 삭제한 이유를 파악해 개선하지 않는다면 유저는 삭제한 앱으로 돌아오지 않을 수 있습니다. 이를 최소화하기 위해 마케터는 앱 삭제를 트래킹하고 분석해야 합니다.
## 앱 삭제 분석을 위한 첫 걸음
앱 삭제를 분석하기 위해 가장 먼저 해야할 일은 무엇일까요? 설치된 앱 중에서 삭제된 앱의 수를 트래킹하는 일입니다. 그리고 앱 삭제율이 높다고 판단할 기준 수치를 정해야 합니다. 기준을 정하지 않고 앱 삭제만 트래킹하면 같은 숫자를 두고 다양한 판단이 나올 수 있습니다. 빠른 대응을 위해서는 기준이 필요합니다.
앱 삭제 현황을 파악했다면 유저가 앱을 삭제하는 원인을 파악하고 문제를 해결해야 합니다. 유저가 겪고 있는 불편함을 해결하고 전체적인 유저 생애 주기를 고려해 앱을 개선하세요.
### 먼저 앱 삭제를 트래킹하세요
앱 설치 수에 비해 리텐션이 낮거나 매출이 기대한 만큼 나오지 않는다면 앱 삭제를 반드시 트래킹해야 합니다. 에어브릿지에서 [앱 삭제 트래킹(Uninstall Tracking)을 설정](https://help.airbridge.io/ko/guides/uninstall-event-tracking)해 앱을 설치한 유저 중 얼마나 많은 유저가 앱을 삭제했는지 알 수 있습니다.
에어브릿지는 최근 6개월 동안 앱에서 이벤트를 1번이라도 일으킨 유저의 앱을 대상으로 유저는 확인할 수 없는 사일런트 푸시(Silent Push)를 매일 보냅니다. 앱에서 사일런트 푸시를 정상적으로 받았다는 응답이 오지 않으면 에어브릿지는 앱 삭제를 기록합니다.
[]
### 유저가 겪는 불편함을 빠르게 해결하세요
앱 삭제 트래킹을 시작했다면 이제는 유저가 앱을 삭제한 원인을 파악해야 합니다. 앱을 삭제하는 이유는 유저마다 다양할 수 있습니다. 앱에서 생긴 불편함일 수 있고, 서비스에 대한 아쉬움일 수 있습니다. 정확한 원인을 파악하기 어렵거나 뚜렷한 원인이 없다면 앱을 떠나는 단계를 찾아 보세요.
퍼널을 통해 유저가 어떤 단계에서 앱을 떠나고 삭제하는지 확인할 수 있습니다. 마케터는 파악한 유저 이탈 단계를 내부에 공유해 앱 삭제를 유발한 원인을 제거하거나, 유저를 잡아두기 위한 방안을 마련할 수 있습니다.
예를 들어 마케터가 설정한 퍼널을 기반으로 앱을 삭제한 유저의 비율을 아래처럼 분석했습니다.
- 회원가입 단계에서 앱을 삭제한 유저: 전체 유저의 60%
- 장바구니 담기 단계에서 앱을 삭제한 유저: 전체 유저의 30%
- 나머지 단계에서 앱을 삭제한 유저: 전체 유저의 1%
유저가 불편함을 겪는 주요 단계는 회원가입과 장바구니 담기입니다. 특히 회원 가입 단계에서 많은 유저가 앱을 삭제하고 있습니다. 회원 가입 과정을 단순하게 만들거나 회원 가입을 권유하는 단계를 조정할 수 있습니다. 회원 가입한 유저의 대부분도 구매까지 이어지지 않고 있습니다. 장바구니에 있는 상품이나 관심 상품에 대한 변동사항을 알리거나 할인 쿠폰을 제공해 구매를 유도할 수 있습니다.
유저가 겪는 불편함을 빠르게 파악하고 해결할 수 있는 2가지 방법을 아래에서 확인하세요.
**✔️ 첫 인상이 중요합니다. 유저의 온보딩 과정을 계속해서 개선하세요**
앱을 설치한 유저가 앱을 실행한 후에 가장 먼저 해야만 하는 일은 주로 회원가입 또는 앱에 필요한 설정을 마치는 것 입니다. 이런 과정에서 불편함과 거부감을 느낀 유저 또는 단순히 회원으로 가입하는 일이 내키지 않은 유저는 앱을 삭제할 수 있습니다. 온보딩 과정을 간편하게 만들거나 회원 가입 과정을 과감하게 삭제하는 것도 좋은 방법입니다.
**✔️ 유저 경험을 고려해서 퍼널을 분석하세요**
유저 경험을 고려하지 않고 퍼널을 분석하면 불편함을 놓치기 쉽습니다. 퍼널을 기반으로 장바구니 담기에서 이탈한 유저가 많다고 분석하면 유저가 장바구니에서 상품을 삭제하기 전에 알림을 넣거나 결제 과정을 더 간단히 만들 수 있습니다. 또는 장바구니 담기에 있는 버튼의 위치나 상품을 선택하는 기능이 불편할 수 있습니다. 정확한 원인을 파악하기 위해서는 유저 경험을 잘 파악하고 있어야 합니다.
### 유저 생애 주기에 따른 유저 경험을 개선하세요
우선 유저 생애 주기(User Life Cycle)에 앱 삭제를 포함하세요. 유저가 여행, 개인 인증, 데이팅, 취업, 이직 등 특수한 목적으로 앱을 사용한다면 유저 생애 주기에 앱 삭제가 있어야 합니다. 해당 앱을 사용하는 유저는 주로 목적을 달성한 후에 앱을 삭제했다가 필요하면 다시 앱을 설치하기 때문입니다. 대표적인 경우가 여행 앱입니다. 유저는 휴가 기간에 여행 앱을 설치하고 모든 여행이 끝나면 앱을 삭제합니다. 그리고 다음 여행을 계획하게 되면 앱을 재설치합니다.
그리고 마케터는 다음 앱 생애 주기에서 유저 유형에 따라 어떻게 접근할지 살펴보세요.예를 들어 예약을 하지 않은 유저가 여행 앱을 삭제했다면 예약 과정에 존재하는 불편함을 찾아야 합니다. 반대로 예약을 마친 유저가 여행 기간 후에 앱을 삭제했다면 다음 앱 생애 주기에서 이 유저의 재설치 가능성을 높게 판단할 수 있습니다.
유저가 앱을 삭제하겠다는 결심을 막거나 재설치를 유도할을 수 있는 방법은 많습니다. 유저에게 이메일, 푸시, 메신저 알림을 보내거나 앱 용량을 적절하게 관리할 수 있습니다.
유저 생애 주기에 따른 유저 경험을 개선하기 위한 3가지 방법을 아래에서 확인해 보세요.
**✔️ 앱을 사용하는 유저와 적극적으로 대화 하세요**
유저가 앱에 머무를 요소를 제공해 활성 유저를 확보할 수 있습니다. 유저가 앱에 머무는 시간을 늘리기 위해 인앱메시지를 전송하거나 커뮤니티를 구축해야 합니다. 이 외에도 게이미피케이션(Gamification)을 추가하는 것을 고려해 보세요. 단, 새로운 요소를 추가하기 전에 반드시 앱 특성을 고려해 주세요.
**✔️유저가 다시 방문하고 싶은 앱으로 성장하세요**
앱이 삭제된 이후에 재설치가 발생하지 않는다면 앱의 본질적인 기능을 돌아봐야 합니다. 사용 목적이 특수한 앱일수록 연관 기능을 추가해 평소에도 유저가 앱을 사용할 이유를 제공해야 합니다. 예를 들어 여행 앱은 여행 정보를 제공할 수 있습니다. 데이팅 앱은 커뮤니티를 형성해 추가적인 만남 기회를 제공할 수 있습니다.
**✔️유저 행동을 다양한 방식으로 분석하세요**
서드파티 플랫폼을 활용해 입체적인 관점으로 유저의 앱 설치와 삭제를 분석해 보세요. 예를 들어 에어브릿지는 액츄얼스 리포트, 리텐션 리포트, 리인스톨 리포트로 앱 설치와 삭제를 광고 채널별로 분석할 수 있습니다. 또한 유저 행동을 분석하는 여러 다른 솔루션과의 연동도 지원하고 있습니다.
좋은 앱을 제공하기 위한 실마리는 현재 앱을 사용하는 유저뿐만 아니라 앱을 삭제한 유저에게도 있습니다. 지금까지 앱이 얼마나 설치됐고 몇 명의 유저가 앱에 남아 있는지에만 집중하지 않으셨나요? 앱 삭제를 트래킹해서 실마리를 찾고 유저를 위한 앱으로 발전해 보세요.
---
## 에어브릿지, 한국콘텐츠진흥원 ‘2023 게임더하기’ 사업 협력사 선정
URL: https://www.airbridge.io/ko/blog/gsp-plus-2023
Language: Korean
Published: 2023-08-08
에어브릿지가 한국콘텐츠진흥원이 진행하는 ‘2023 게임더하기’ 사업의 인프라 협력사로 선정됐습니다! AB180에서 자체 개발한 모바일 측정 파트너(MMP) 에어브릿지는 정확하고 균형 잡힌 광고 성과 분석으로 국내 게임 개발사의 글로벌 마케팅을 적극 지원할 예정입니다.
## 게임더하기 사업이란
게임더하기 사업은 국내 중소 게임 개발사의 글로벌 시장 진출 및 게임 산업 경쟁력 강화를 위해 2020년 시작됐습니다. 2022년에는 총 37개 게임을 지원했으며 글로벌 다운로드 1,390만 건, 글로벌 매출액 498억 원을 기록했습니다.
사업을 주관하는 한국콘텐츠진흥원 및 사업별 전문가로 구성된 위원회는 국내 게임 개발사의 성공적인 해외 진출을 지원하기 위해 철저한 검증을 통해 협력사를 선정합니다. 그리고 사업에 참여하는 게임 개발사는 에어브릿지를 비롯한 협력사의 서비스를 자유롭게 선택해 이용할 수 있습니다.
## 모바일 게임 마케팅에 에어브릿지를 활용해야 하는 이유
에어브릿지는 ARPU 및 ROAS 메트릭부터 인앱광고(IAA) 및 인앱구매(IAP) 관련 분석까지 폭넓은 데이터 서비스를 하나의 대시보드에서 제공합니다. LTV가 높은 오디언스 세그먼트를 기반으로 매체에서 유사 타겟을 생성하여 유저 타겟팅, 획득, 리타겟팅 전략을 최적화하고 ROI를 극대화할 수도 있습니다. 다양한 애드 네트워크, 미디에이션 플랫폼과도 연동되어 있어 데이터를 관리하기에도 편리합니다.
또한, 에어브릿지는 다양한 어트리뷰션 모델을 통해 더욱 정확한 광고 성과 분석을 지원합니다. 빠르고 직관적인 의사결정을 돕는 라스트 터치 어트리뷰션(LTA), 한층 더 정교한 인사이트를 제공하는 멀티 터치 어트리뷰션(MTA)과 순증가분(Incrementality) 분석, 그리고 개인 정보 없이도 거시적인 성과를 측정하고 최적화된 예산안을 제시하는 마케팅 믹스 모델링(MMM) 등을 한 곳에서 제공합니다.
에어브릿지는 무엇보다 한국 회사로서 직접 글로벌 시장에 진출하고 마케팅을 진행하고 있습니다. 2022년부터 글로벌 사업을 본격화한지 1년 남짓 된 현재 40여개 국 고객사를 위해 200여개 국의 마케팅 캠페인을 분석하고 있습니다. 이러한 성과를 이루기까지 쉽지만은 않은 과정을 거쳤기 때문에 국내 기업이 해외에서 마주하는 어려움을 누구보다 잘 이해하고 있습니다. 더 나아가, 한국에 본사를 둔 유일한 MMP로서 언제나 고객사를 전방위로 지원하며, 빠르고 적극적으로 피드백을 수용합니다. 이러한 이유로 에어브릿지는 게임더하기 사업을 위한 최적의 협력사라는 평가를 받고 있습니다.
## 에어브릿지는 모바일 게임 앱의 든든한 파트너가 되겠습니다
마케팅 환경은 전례없이 빠르게 변하고 있습니다. 이러한 가운데에도 모바일 게임 앱이 광고 성과에 대한 가시성을 잃지 않을 수 있도록 에어브릿지는 계속해서 기술을 고도화하고 전문성을 확장하고 있습니다. 특히 2023년 중 독자 개발 ‘광고 LTV 예측 모델’을 선보일 계획으로, 해당 모델을 활용하면 채널, 국가, OS별로 가장 높은 LTV를 가진 유저에 맞추어 광고 전략을 최적화하고 수익을 극대화할 수 있습니다.
에어브릿지는 2023 게임더하기 사업을 통해 국내 게임 개발사의 글로벌 파트너로서 함께 할 수 있게 되어 기쁘며, 더 넓은 시장에서 고객사와 함께 도약하기를 기대하고 있습니다.
앱 마케팅의 변화, [에어브릿지](https://www.airbridge.io/ko/book-a-demo?utm_source=blog&utm_medium=cta&utm_campaign=gsp-plus-2023)와 함께 효과적으로 대비해 보세요.
---
## 지속 가능한 유저 획득을 위한 세 가지 방법
URL: https://www.airbridge.io/ko/blog/three-tips-for-sustainable-user-acquisition
Language: Korean
Published: 2023-07-10
경제적인 불황으로 인해 마케팅 예산이 점차 줄어들고 있고, 개인정보 보호 트렌드가 강화되면서 광고 캠페인의 효율도 점차 낮아지고 있습니다.
어려운 상황 속에서도 프로덕트가 꾸준하게 성장하기 위해서는 지속 가능한 유저 획득이 필요합니다. 어떤 광고 캠페인이 성과를 만들었는지 다각도로 파악하고, 이를 바탕으로 마케팅 캠페인을 최적화해야 합니다. 그리고 유저가 정말 필요로 하는 것을 보여주어야 유저의 시간과 마음을 붙잡을 수 있습니다.
이번 포스트에서는 광고 캠페인의 효율 향상을 위해 광고의 성과 측정을 개선하는 방법 두 가지와 유저를 획득하는 과정부터 개인화하여 유저가 우리 앱을 더 잘 이해하고 신뢰할 수 있도록 만드는 방법을 소개합니다.
> 이 포스트는 AB180 오프라인 행사에서 진행한 AB180 남성필 대표의 세션 내용을 참고하여 작성했습니다.
## 양질의 시그널 데이터 전송하기
시그널 데이터(Signal Data)란 유저 행동 데이터를 말하며 광고주의 앱에서 수집되어 픽셀, SDK, API를 통해 광고 매체로 전송됩니다. 광고 매체로 전송된 시그널 데이터를 통해 광고를 보거나 클릭한 유저의 ID와 대조하여 광고의 전환 성과를 측정할 수 있습니다. 그리고 측정한 성과를 바탕으로 전환 가능성이 높다고 식별한 유저를 타겟으로 하여 최적화된 마케팅 캠페인을 집행할 수 있기 때문에, **시그널 데이터는 앱 광고 효율 향상에 매우 핵심적인 역할을 합니다.**
하지만 개인정보 보호 트렌드의 강화로 인해 시그널 데이터의 매칭 키로 활용되는 유저 식별자를 수집하기 어려워지면서 캠페인의 효율이 감소하고 있습니다. 애플(Apple)의 ATT(App Tracking Transparency) 정책 시행 이후 유저가 ATT를 동의하지 않으면 유저 데이터를 수집할 수 없어지면서 매칭률이 낮아진 것입니다.
이러한 이유로 광고 캠페인의 성과를 제대로 측정하기 어렵기 때문에 전환 API(Conversion API) 활용하기, TAT(Tracking Authorized Timeout) 값 설정하기 등을 통해 시그널 데이터의 매칭률 향상이 필요합니다.
### 전환 API(Conversion API)를 활용한 이벤트 데이터 전송
[]
웹사이트에서 유저의 전환 행동을 트래킹하기 위해 그동안 많이 활용되어 왔던 것은 픽셀(Pixel)입니다. 픽셀은 광고 매체에서 제공하는 코드 뭉치로, 광고주의 웹사이트에 삽입되어 유저의 행동 이벤트를 수집합니다. 수집된 데이터는 유저의 쿠키 정보와 함께 **브라우저**를 통해 광고 매체로 전달됩니다. 픽셀을 이용한 이벤트 데이터 전송은 이처럼 서드파티 쿠키에 의존하기 때문에 애드 블록이나 브라우저의 개인정보 보호 정책에 따라 정확한 트래킹이 어려워지고 있습니다. 따라서 픽셀과 다르게 데이터 누락이 발생하지 않고, 더 정확하게 트래킹이 가능한 전환 API(Conversion API)를 활용하여 이벤트 데이터를 전송하는 것이 필요합니다.
전환 API는 브라우저를 거치는 픽셀과 달리 광고주의 서버에 수집된 이벤트 데이터를 매체 서버로 **직접** 전송하는 방식을 말합니다. 전환 API 방식에서는 매체 서버에서 광고주의 서버에 데이터를 요청하고 광고주의 서버는 요청받은 데이터를 매체 서버로 직접 전달합니다. 전환 API로 유저 행동 이벤트와 유저를 식별할 수 있는 ID(클릭 ID, 암호화된 이메일)를 광고 매체 서버로 직접 보냄으로써 **시그널 데이터의 매칭률을 향상시킬 수 있습니다.**
전환 API에 관한 더 자세한 정보를 알고 싶다면, [픽셀 대신 Conversion API를 활용해 이벤트 측정 개선하기](https://blog.ab180.co/posts/facebook-conversion-api-1) 포스트를 확인해 보세요.
### 매칭률 향상을 위한 TAT(Tracking Authorized Timeout) 값 설정
[]
MMP/앱 어트리뷰션 툴은 터치 포인트와 설치 이벤트의 유저 식별자를 대조하여 광고의 앱 설치 기여를 측정합니다. 하지만 애플의 ATT 정책이 시행되고 유저가 ATT에 동의하지 않으면 유저 데이터의 수집 및 활용이 불가능해지면서 광고의 성과 측정에 문제가 생겼습니다. 유저에게 ATT 동의 화면이 노출되기 전, 즉 유저가 데이터 수집을 동의하기 전에 설치 이벤트가 전송되기 때문에 설치 이벤트가 유저 식별자를 가질 수 없게된 것입니다. 따라서 이 문제를 해결할 수 있는 TAT 값을 설정하는 것이 필요합니다.
TAT 값 설정은 설치 이벤트를 MMP로 전송하는 시간을 지연시킨다는 것을 의미합니다. 30~60초 정도의 TAT 값 설정을 통해 유저가 ATT 허용을 한 이후에 설치 이벤트가 전송될 수 있도록 함으로써 설치 이벤트에 유저 식별자가 포함되도록 만드는 것입니다. **TAT 값 설정을 함으로써 매칭률을 높여 광고 성과를 더 정확하게 측정할 수 있습니다.**
## 효과적인 iOS 캠페인 집행을 위한 SKAdNetwork(SKAN) 활용 방법
SKAdNetwork(SKAN)이란 무엇이며, 왜 중요할까요? SKAN은 애플이 소개한 개인정보를 보호하는 어트리뷰션 프레임워크으로, 개인정보 보호를 위해 유저/디바이스 레벨 데이터 대신 집계된 데이터를 제공합니다. iOS 캠페인을 측정하고 최적화하기 위해서는 SKAN을 제대로 활용할 수 있어야 합니다. SKAN의 효과적인 활용을 도와주는 **SKAN 임계값 충족하기**와 **초기 이벤트 설정하기** 두 가지 방법을 소개합니다.
### 최적의 캠페인에 예산을 집중하여 SKAN 임계값(Privacy Threshold) 충족하기
SKAN은 유저의 앱 설치 수와 캠페인 ID 수가 일정 기준을 충족해야 전환 성과를 측정할 수 있는데 이러한 기준을 ‘SKAN 임계값’이라고 합니다. 임계값이 충족되지 않으면 SKAN 전환값, 즉 성과를 확인할 수 없기 때문에 캠페인의 성과 측정과 최적화에 영향을 미치게 됩니다. 임계값을 충족하기 위한 최소 앱 설치 이벤트 수가 공식적으로는 알려지지 않았지만, 대표적인 광고 매체에서 임계값을 충족할 수 있도록 가이드라인을 제공하고 있습니다.
> **대표적인 매체에서 제공하는 가이드라인 확인하기**[
Meta](https://www.facebook.com/business/help/595608511524017)[
Google](https://support.google.com/google-ads/answer/10384955?hl=ko)[
Unity](https://resources.unity.com/ios-14-resources/skad-postback-null-values)[
TikTok](https://ads.tiktok.com/help/article/measurement-considerations-ios14?lang=en)
광고 매체에서 권장하는 가이드라인을 확인하고 SKAN 광고 캠페인을 최적화해 보세요.
SKAN 캠페인 전환값을 확인하기 위해서는 임계값을 충족시킬 수 있도록 일부 캠페인에 예산을 선택적으로 집중하여 가이드라인에서 제안하는 최소 설치 수 이상을 확보하는 것이 좋습니다. 또한 캠페인이 타겟하고 있는 오디언스를 지나치게 세분화하기보단 일정 수준 이상의 규모를 유지하며 캠페인 ID 수가 임계값 기준에 충족할 수 있도록 관리하는 것이 중요합니다.
### SKAN 설정에 초기 전환 이벤트를 포함하기
SKAN 3.0은 앱 설치 후 24시간 동안 유저의 인앱 이벤트를 측정하는데, 이 기간 내에 전환 이벤트가 발생하면 타이머가 초기화되어 추가로 24시간 동안 인앱 이벤트를 측정합니다. 이때 최대 64개의 후속 인앱 이벤트를 설정할 수 있어, 이론적으로 최대 65일 후에 전환 성과를 확인하는 상황이 생길 수 있습니다. 게다가 SKAN은 유저 식별을 방지하기 위해 포스트백을 최대 72시간 지연시키기 때문에, 실제 이벤트 발생보다 뒤늦은 시점에 전환 데이터를 확인하게 됩니다.
에어브릿지(Airbridge)를 포함한 대부분의 MMP는 캠페인 성과 확인이 과도하게 지연되는 경우를 방지하기 위해 앱 설치 후 24시간 동안의 인앱 이벤트만 SKAN에 전송합니다. 따라서 **이 24시간 이내에 발생할 확률이 높은 초기 이벤트를 SKAN으로 전송할 인앱 이벤트로 설정하는 것이 좋습니다.** 대표적인 초기 이벤트로는 온보딩 완료, 구독 화면 노출, 회원가입, 무료 체험 시작 등이 있습니다.
[]
SKAN 3.0은 1번의 포스트백만을 제공했기 때문에 인앱 이벤트 설정에 주의해야했지만, SKAN 4.0부터는 최대 3번의 포스트백을 지원하여 초기 이벤트만 설정하지 않아도 좋습니다. SKAN 4.0은 0~2일, 3~7일, 8~35일에 대한 전환을 관찰하고 유저 행동에 따라 최대 3번 포스트백을 수신할 수 있습니다. 게다가 락 윈도우(Lock Window) 기능을 활용하면 유저가 전환 이벤트를 수행한 시점에 따라 기간을 단축하여 맞춤화할 수 있어 유용하게 활용할 수 있습니다.
> SKAN 4.0에 관한 더 자세한 정보를 알고 싶다면 아래 포스트를 확인해 보세요.
👉 [앱 마케터가 SKAdNetwork 4.0 (SKAN 4.0)에 주목해야 하는 이유
](https://www.airbridge.io/ko/blog/exploring-the-marketing-implications-of-skadnetwork-4)👉 애플 SKAN 4.0을 제대로 활용하고 싶은 마케터가 꼭 알아야 하는 3가지
### 유저/디바이스 레벨 데이터로 성과를 측정할 수 있는 애플 서치 애드(Apple Search Ads)
애플 서치 애드(Apple Search Ads)는 유저가 앱스토어에서 검색할 때 손쉽게 여러분의 앱 서비스를 발견할 수 있도록 애플에서 제공하는 광고 플랫폼입니다.
앞서 언급한 것처럼 애플의 iOS 14.5 업데이트 이후 유저가 ATT 허용을 하지 않으면 전환 성과를 파악할 수 없게 되었습니다. 그러나 애플 서치 애드를 사용하면 애드 서비스 API(Ad Serivce API)를 통해 ATT 상태와 관계없이 실제 캠페인 이름 대신 캠페인 ID로 확인할 수 있는 제한된 형태의 어트리뷰션 데이터를 확인할 수 있습니다.
또한 애플 서치 애드 어드밴스드(Advanced)에만 제공되는 **캠페인 매니지먼트 API(Campaign Management API)**를 활용하면 ATT 정책 시행 이전처럼 캠페인 성과를 캠페인 ID가 아닌 캠페인의 실제 이름으로 확인할 수 있습니다.
## CPP/CSL로 유저 획득 플로우 개인화하기
애플 앱스토어의 맞춤형 제품 페이지(Custom Product Pages, CPP), 구글 플레이스토어의 맞춤 스토어 등록정보(Custom Store Listings, CSL)는 모바일 앱의 설치 페이지를 오디언스에 따라 개인화할 수 있는 기능입니다. 예를 들어 호텔이나 렌트카를 예약할 수 있는 여행 앱을 서비스한다면, 검색 광고에서 ‘호텔’이라는 키워드로 유입된 유저에게는 호텔 관련 콘텐츠를 보여주고, ‘렌트카 예약’을 홍보하는 디스플레이 광고를 클릭해 유입된 유저에게는 자동차 렌탈 기능을 강조하는 페이지를 보여줄 수 있습니다. CPP/CSL을 활용하면 이렇게 개별 유저에게 관련성이 높은 콘텐츠나 제품의 기능을 강조할 수 있습니다.
[]
### CPP/CSL로 전환율 높이기
CPP/CSL은 유저의 트래픽 소스를 기반으로 개인화된 페이지를 제공하기 때문에 리타겟팅과 같은 유저 데이터 기반 마케팅의 효율이 낮아지고 있는 현재, 더 강력한 마케팅 도구가 될 수 있습니다. CPP/CSL은 유저 여정의 가장 첫 단계인 앱 설치부터 유저 경험을 일관성 있게 구성하여 유저가 여러분의 서비스를 훨씬 쉽게 이해하고 신뢰할 수 있게 만듭니다. 이에 따라 **전환율을 높여 광고 캠페인의 ROAS를 향상시킬 수 있습니다.**
### CPP/CSL 최적화를 위한 성과 분석하기
CPP/CSL도 다른 마케팅 캠페인과 마찬가지로 성과를 측정하는 것이 중요합니다. 각 맞춤 페이지가 실제로 유저 획득과 전환에 기여하는지 파악하고, 이를 바탕으로 최적화하는 것이 필요하죠. 맞춤 페이지의 성과를 측정하고 효율적으로 관리하기 위해서는 에어브릿지와 같은 MMP와 연동하여 사용하는 것이 좋습니다.
맞춤 페이지를 생성한 후 MMP를 통해 간단하게 트래킹 링크를 만들고 관리할 수 있습니다. 또한 딥링크(Deep Link) 기능을 사용하면 유저 경험을 획기적으로 향상시킬 수 있습니다. 딥링크란, 링크를 클릭한 유저를 웹사이트나 스토어가 아닌 앱의 특정 페이지로 이동시키는 기능을 말합니다. 만일 앱이 설치되어 있지 않다면 스토어로 이동하게 되는데, 이 때 CPP/CSL이 나타나도록 설정하고, 앱 설치 후에는 딥링크가 기존에 안내하려던 앱 내 페이지로 유저를 이동시킬 수 있는 디퍼드 딥링크(Deferred Deep Link) 기능도 활용할 수 있습니다. CPP/CSL을 이러한 딥링크 기능과 함께 활용하면 설치부터 전환 이벤트 수행까지 유저 여정의 시간과 노력을 줄여줌으로써 **유저 경험을 크게 향상시킬 수 있습니다.**
## 어려운 마케팅 환경에서 효율 극대화를 위한 접근
지속 가능한 유저 획득을 위해 다각도로 광고 성과를 측정하고, 유저 여정의 첫 단계부터 우리 앱을 신뢰할 수 있도록 만드는 개인화 방법에 대해 알아보았습니다.
이번 포스트에서 소개한 지속 가능한 유저 획득을 위한 세 가지 방법을 여러분의 앱의 상황에 따라 우선순위를 정하고 해나가다 보면 마케팅 캠페인의 효율을 높여 꾸준한 성장을 이룰 수 있을 것입니다.
에어브릿지는 대기업부터 스타트업까지 국내외 다양한 기업들과 함께 마케팅 성과를 측정하고 캠페인을 최적화할 수 있도록 지원하고 있습니다. 마케팅 캠페인의 효율을 최적화하고 싶은 앱 마케터라면 지금 바로 에어브릿지와 함께해 보세요.
---
## 하이퍼 캐쥬얼 게임에 광고 미디에이션이 필요한 이유
URL: https://www.airbridge.io/ko/blog/hyper-casual-game-ad-mediation
Language: Korean
Published: 2023-07-06
하이퍼 캐주얼 게임을 즐기다 보면 자연스럽게 광고를 접하게 됩니다. 게임 화면 위에서 확인할 수 있는 광고, 게임을 하기 전에 잠시 나오는 전면 광고 그리고 보상이 주어지는 리워드 광고 등 하이퍼 캐주얼 게임을 통해 다양한 광고가 여러 형식으로 유저에게 제공됩니다.
이처럼 하이퍼 캐주얼 게임에 수많은 광고가 제공될 수 있는 배경에는 모바일 광고 미디에이션 플랫폼이 있습니다. 미디에이션(Mediation) 플랫폼은 앱의 광고 지면에 나오는 광고를 관리할 수 있는 기능을 제공합니다. 또 고가의 미술품을 경매하는 것처럼 모바일 광고 미디에이션 플랫폼은 비딩을 통해 광고 지면을 제공하는 앱의 수익을 늘리는 데 도움이 됩니다.
## 하이퍼 캐쥬얼 게임의 수익을 책임지는 모바일 광고 미디에이션 플랫폼
하이퍼 캐쥬얼 게임은 조작 방법과 규칙 모두 단순하기 때문에 많은 유저가 즐기고 있습니다. 하지만 접근 장벽이 낮은 만큼 게임 수익화가 어렵습니다. 하이퍼 캐쥬얼 게임을 즐기는 유저 중 다수가 제품을 구매하거나 구독하지 않습니다. 또 오랫동안 즐기는 유저도 많지 않습니다.
대신 하이퍼 캐쥬얼 게임에는 수많은 광고 지면이 있습니다. 배너 광고, 전면 광고, 보상형 광고, 네이티브 광고 등 다양한 광고를 유저 경험을 헤치지 않는다면 자유롭게 제공할 수 있습니다. 남은 문제는 광고 수급입니다. 많은 유저에게 광고가 동시에 송출되더라도 끊임 없이 광고 지면에 제공되어야 합니다. 대체로 모바일 광고 시장에서 광고 네트워크가 광고를 제공하는 역할을 맡고 있습니다.
하지만 여러 광고 네트워크를 사용하려면 각 광고 네트워크의 SDK를 매번 추가로 설치해야 합니다. 게임 개발사 입장에서 번거롭고 여러 광고 네트워크를 관리하기 쉽지 않습니다. 모바일 광고 미디에이션 플랫폼이 등장한 배경입니다. 모바일 광고 미디에이션 플랫폼은 여러 광고 네트워크를 간편하게 연동할 수 있는 환경을 제공하고 광고 수익을 지원합니다.
## 모바일 광고 미디에이션 플랫폼이 필요한 이유
모바일 광고 생태계는 SSP(Supply Side Platform, 공급자 중심 플랫폼)와 DSP(Demand Side Platform, 수요자 중심 플랫폼)로 나눌 수 있습니다. SSP는 채널의 광고 지면 판매를 돕고, DSP는 마케터의 광고 지면 구매를 돕습니다.
광고 네트워크는 이러한 SSP, DSP를 모두 포괄하는 플랫폼입니다. 광고 지면을 사고 파는 곳입니다. 모바일 광고 생태계는 광고 네트워크 여러 개로 구성되어 있습니다. 게임 앱 개발사 입장에서는 여러 광고 네트워크를 활용하면 광고 지면에 광고를 계속해서 게재할 수 있습니다. 하지만 모든 광고 네트워크와 연동하는 것은 비용과 운영 모두에서 부담이 발생합니다. 특히 연동하는 모든 광고 네트워크의 SDK를 앱에 설치해야 합니다.
이런 부담을 해결한 것이 모바일 광고 미디에이션 플랫폼입니다. 모바일 광고 미디에이션 플랫폼은 광고 네트워크의 일종으로 앱에 모바일 광고 미디에이션 플랫폼의 SDK 하나만 설치하면 여러 광고 네트워크를 활용할 수 있습니다.
무엇보다 이 플랫폼이 고가의 미술품을 경매하는 경매사처럼 광고 지면을 비딩 등을 통해 판매하기 때문에 광고 수익화 면에서도 유리합니다. 또 모바일 광고 미디에이션 플랫폼은 Fill Rate(광고 지면에 노출한 광고 비율), eCPM(광고 1,000회 노출당 발생한 수익) 등의 광고 성과 지표로 광고 지면에서 발생하는 수익 향상을 돕습니다.
## 하이퍼 캐쥬얼 게임에 필요한 최적의 광고 수익화
광고 지면 판매는 아래와 같은 과정을 빠르게 반복하면서 진행됩니다.
1. 유저가 광고 지면에 접근
2. SSP에서 광고 지면에 나오는 광고를 요청
3. DSP에서 광고를 제공
4. SSP는 유저에게 광고를 노출
광고 지면 판매를 돕는 SSP는 고민이 생깁니다. 광고 지면은 한정되어 있는데, DSP가 여러 광고를 제공하고 있기 때문입니다. 모바일 광고 미디에이션 플랫폼은 광고 지면을 최대한 높은 가격에 판매하기 위해 비딩을 통해 거래합니다.
모바일 광고 미디에이션 플랫폼의 대표적인 비딩 방식은 2가지입니다. 워터폴 비딩(Waterfall Bidding)은 광고 네트워크마다 우선 순위를 지정해 광고 송출 기회를 부여하는 방식입니다. 광고 네트워크의 성과, 특성 등 과거 데이터를 기반으로 우선 순위가 정해집니다. 우선 순위가 낮은 광고 네트워크는 더 높은 가격을 내고 광고 지면을 사용하고 싶어도 어렵습니다.
이런 점을 극복하기 위해 등장한 방식이 인앱 헤더 비딩(In-app Header Bidding)입니다. 인앱 헤더 비딩은 광고 네트워크가 광고 지면에 대해 실시간으로 가격 경쟁을 진행하는 방식입니다. 실시간 경쟁이기 때문에 광고 지면을 가장 높은 가격으로 판매할 수 있습니다. 마케터 입장에서도 광고 가격 등으로 광고 지면의 주인이 결정되기 때문에 투명성이 보장됩니다.
다만 인앱 헤더 비딩은 서버에 무리가 발생할 수 있습니다. 다수의 모바일 광고 미디에이션 플랫폼이 워터폴 비딩과 인앱 헤더 비딩을 섞어 활용하고 있는 이유입니다.
### 모바일 광고 미디에이션 플랫폼의 비딩 이해하기
아래는 최소 비딩 가격이 3달러인 워터폴 비딩과 인앱 헤더 비딩의 예시입니다. 워터폴 비딩의 우선 순위는 광고 네트워크 A가 가장 높고, 이어서 광고 네트워크 B~D 순입니다.
비딩 가격($)워터폴
비딩 결과인앱 헤더
비딩 결과광고 네트워크 A2.5패배패배광고 네트워크 B3.4승리패배광고 네트워크 C3.6호출되지 않음승리광고 네트워크 D3.2호출되지 않음패배
워터폴 비딩은 우선순위에 따라 광고 네트워크에게 광고를 요청합니다. 먼저 광고 네트워크 A를 호출했지만, 최소 비딩 가격이 3달러 이기 때문에 광고 네트워크 A의 광고는 노출되지 않습니다. 우선순위가 낮은 광고 네트워크 C, D는 광고를 요청받지도 않습니다.
반면, 인앱 헤더 비딩은 모든 광고 네트워크의 비딩 가격을 보고 노출할 광고를 결정합니다. 광고 네트워크 A는 최소 비딩 가격을 넘기지 못해 인앱 헤더 비딩에서도 탈락합니다. 광고 네트워크 B, C, D 중에서 가장 비싼 가격을 제안한 광고 네트워크 C의 광고가 광고 지면에 나오게 됩니다.
## 에어브릿지는 모바일 게임 앱의 성장을 위해 모든 것을 지원합니다
모바일 게임 앱이 지속 가능하기 위해서는 유저 획득과 전환에 효과적인 광고 채널을 찾아 최적화해야 합니다. 애드 미디에이션 플랫폼이 광고 수익과 직결되기는 하나, 이외에도 게임 내 아이템 구매 등으로도 수익을 만들 수 있습니다. 이렇게 다양하게 발생하는 수익과 전환은 하나의 플랫폼에서 확인하고 성과를 분석하는 것이 효율적입니다.
에어브릿지는 게임 앱과 관련된 비용 및 수익 데이터를 통합하여 하나의 대시보드에서 제공합니다. 게임 앱을 위한 주요 모바일 광고 미디에이션 플랫폼과 연동되어 있으며, 연동한 데이터를 활용할 수 있는 다양한 기능도 개발하고 있습니다.
에어브릿지와 함께 성공적인 광고 수익화를 통한 앱의 성장을 만들어보세요.
---
## 프라이버시 시대 앱 마케터를 위한 마케팅 믹스 모델
URL: https://www.airbridge.io/ko/blog/marketing-mix-modeling-for-app-marketer
Language: Korean
Published: 2023-06-29
> 이 글은 마케팅 믹스 모델링에 관한 2편의 백서를 요약한 포스트입니다. 마케팅 믹스 모델링에 대한 더 자세한 내용이 궁금하시다면 아래 링크를 통해 다운로드해 주세요.
- [머신러닝에 기반한 앱 마케팅: 마케팅 믹스 모델링으로 성과 측정부터 예측까지
](https://www.airbridge.io/ko/library/mmm-basics-whitepaper?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=airbridge-blog-mmm-basics-whitepaper&utm_content=ko)- [프라이버시 시대의 앱 마케터를 위한 마케팅 믹스 모델링 실전 활용 가이드](https://www.airbridge.io/ko/library/mmm-tips-whitepaper?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=airbridge-blog-mmm-tips-whitepaper)
## 사라지는 데이터, 프라이버시 시대에서 광고 성과 측정이 어려워지고 있습니다.
오늘날 모바일 광고 성과를 측정하는 표준 모델인 라스트 터치 어트리뷰션(Last-touch Attibution, LTA)는 프라이버시 시대로 인해 위기를 맞이했습니다. LTA 모델은 반드시 유저 레벨 데이터가 필요하기 때문입니다.
주요 브라우저에서 유저를 식별할 수 있는 서드파티 쿠키(Third-Party Cookie)를 제공하지 않고 애플과 구글은 유저 데이터 대신 개인을 식별할 수 없는 집계 데이터를 제공하기 시작했습니다. 사회적 인식의 변화로 유럽과 미국에서는 개인정보보호 법률을 강화했습니다.
## 유저 데이터 없이 광고 성과를 측정하는 방법
개인정보를 수집하기 어려워지면서 마케팅 믹스 모델링은 광고 성과 측정 모델로 다시 주목받고 있습니다. 마케팅 믹스 모델링은 유저의 ADID나 개인정보는 필요로 하지 않습니다. 마케팅 전략과 광고 환경의 영향을 고려해 광고 매출과 성과를 머신러닝으로 측정하는 분석법입니다.
마케팅 믹스 모델은 기본적으로 광고 지출과 광고 성과의 관계를 나타낸 수식의 형태입니다. 이 수식은 다양한 광고 특성이 반영되어 있고 머신러닝 과정을 거쳐 최적의 수식을 찾습니다.
> 전체 성과 = 채널 A의 광고 성과 + 채널 B의 광고 성과 + 기타 성과
이 수식을 구하는 것으로 끝이 아닙니다. 마케팅 믹스 모델은 검증과 보정 과정을 통해 계속해서 발전합니다. 광고가 나오는 환경, 앱 특성 등을 고려해 나만의 마케팅 믹스 모델이 구축됩니다.
## 데이터에서 시작하는 마케팅 믹스 모델
마케팅 믹스 모델은 데이터에 따라 다르게 구축됩니다. 따라서 마케터는 마케팅 목표를 고려해 광고 지출, 매출액, 앱 설치 수, 클릭 수와 같은 데이터를 준비해야 합니다.
[]
마케팅 믹스 모델은 복잡한 변수와 수많은 데이터 속에 숨어있는 패턴을 표현한 수식입니다. 예를 들어 채널 A에서 사용된 광고 지출과 신규 앱 설치 수에 관한 데이터 속에는 무수히 많은 변수가 있을 것입니다. 머신러닝을 통한 회귀 분석을 통해 입력값과 결과값 사이의 패턴을 가장 잘 설명하는 선형 함수를 찾는 것을 목표로 합니다.
[]
예를 들어 그림 1처럼 사용된 광고 지출과 신규 앱 설치 수에 관한 데이터가 있다고 했을 때, 명확한 패턴을 찾기 어렵습니다. 그림 2에 나타난 다양한 직선 중에 어떤 직선이 오차가 적을까요? 여기서 그림3처럼 우리는 실제 데이터와 가장 오차가 작은 직선을 찾습니다. 그 직선이 우리가 찾는 마케팅 믹스 모델입니다.
## 마케터가 고려해야 하는 것
수영복은 여름과 겨울 중 언제 더 팔릴까요? 월요일 아침 출근길 광고에서 본 비타민 영양제를 구매하는 회사원은 몇 명이나 될까요? 팬데믹과 같은 특별한 사건이 광고에 어떤 영향을 줄까요?
머신러닝으로 패턴만 찾는 것으로는 마케팅 믹스 모델을 구축하기 어렵습니다. 광고는 복잡한 상황이 자주 연출되기 때문입니다. 따라서 마케터는 마케팅 믹스 모델을 구하는 과정에서 아래 4가지를 반드시 고려해야 합니다.
[]
**이월 효과**
이월 효과를 고려하여 이월 적용 기간을 정합니다. 예를 들어 오늘이 1이라면, 어제는 0.5, 그저께는 0.25로 성과를 측정하는 것입니다. 마케팅 믹스 모델에서는 더 적절한 값을 찾기 위해 지수 감소(Geometric Decay), 와이블 감소(Weibull Decay)와 같은 머신러닝 모델을 적용하고 있습니다.
**포화 효과**
광고 지출을 늘려도 구매하지 않기로 결정한 사람에게도 광고가 노출될 수 있기 때문에 비용의 효율이 점점 줄어듭니다. 이 현상을 포화 효과라고 말하며 지수 함수(Exponential Function)나 힐 함수(Hill Function)을 적용하여 정확한 값을 찾아냅니다.
**계절성**
성수기 숙박 앱의 매출은 따로 광고하지 않아도 늘어납니다. 이처럼 광고가 노출되는 시점에 따라 성과가 변할 수 있습니다. 계절과 같은 주기적인 변화에 대응하기 위해 사인 파동, 스퀘어 파동을 마케팅 믹스 모델에 적용하여 정확한 함수를 찾을 수 있습니다.
**맥락 변수**
앱 마켓에서의 랭킹 상승 또는 팬데믹, 경제 상황 같은 다양한 사건으로 광고 성과가 바뀔 수도 있으며, 이를 맥락 변수라고 합니다. 맥락 변수를 설정해야 정확한 광고 성과를 추정할 수 있습니다.
또한 한번의 회귀 분석으로 구한 선형 함수를 정확한 수식으로 보기 어렵습니다. 우리가 구한 선형 함수가 마케터가 준비한 데이터에 숨겨진 패턴을 잘 설명하는지 확인하는 과정이 필요합니다. 이러한 검증에 자주 사용되는 지표는 다음과 같습니다.
[]
## 실전 활용 예시
마케팅 믹스 모델링에 대해 이해하셨다면, 실제 광고 성과를 측정하는 데 어떻게 활용되는지 궁금하실 것입니다. 아래 예시를 통해 마케팅 믹스 모델링의 실전 활용에 대해 소개하겠습니다.
### 게임사의 ‘진짜 데이터' 찾기
한 게임사가 새로운 모바일 게임 출시와 인플루언서 마케팅을 진행했습니다. 여러 인플루언서와 계약을 맺고 소셜 미디어, 동영상 플랫폼에 광고 콘텐츠를 업로드했습니다. 인플루언서 마케팅 성과는 어떻게 확인하면 좋을까요?
[]
인플루언서 마케팅은 일반적으로 구독자가 많을수록 광고비가 많습니다. 하지만 오히려 구독자가 적은 인플루언서가 많은 설치 수를 발생시킬 수 있습니다. 광고 비용보다 광고 콘텐츠를 접한 유저의 반응이 더 중요한 이유입니다.
따라서 광고 지출 대신 광고 성과를 만든 ‘진짜 데이터’를 정하여 광고 성과를 더 정확하게 측정할 수 있습니다.
### 핀테크의 적절한 분석 기간 설정
핀테크 스타트업은 간편 결제 서비스를 통해 지난 3년 동안 빠르게 성장했습니다. 핀테크 스타트업의 빠른 성장으로 기존 금융권도 공격적인 마케팅을 펼치기 시작했고 핀테크의 마케터는 최소 비용으로 최대 성과를 발생시키기 위해 마케팅 믹스 모델을 사용하기로 했습니다. 어떻게 하면 정확한 광고 성과를 측정할 수 있을까요?
[]
소비 트렌드가 바뀌는 주기는 6개월보다 깁니다. 매년 계절마다, 특정일마다 소비가 늘어나거나 줄어들죠. 마케팅 믹스 모델은 준비한 데이터만으로 구축되기 때문에 1~6월의 데이터로 7~12월 데이터를 예측하기는 어렵습니다.
따라서 장기간 트렌드를 분석하여 마케팅 믹스 모델에 적용하였습니다. 간편 결제 서비스를 출시한 이후 3년 치의 데이터를 통해 결제량이 연말과 연초에 증가하고, 4~10월에 저조했다는 것을 알 수 있었습니다. 장기적인 트렌드 반영으로 더 신뢰할 수 있는 광고 성과를 확인했습니다.
위에서 소개해 드린 활용과 함께 오프라인 광고, OTT 플랫폼, 농산물 광고 등에 적용되는 다양한 예시를 [프라이버시 시대의 앱 마케터를 위한 마케팅 믹스 모델링 실전 활용 가이드](https://www.airbridge.io/ko/library/mmm-tips-whitepaper?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=airbridge-blog-mmm-tips-whitepaper)를 통해서 확인하실 수 있습니다.
## 마케팅 믹스 모델링을 사용할 수 있는 방법, 에어브릿지에서 확인해 보세요.
에어브릿지의 마케팅 믹스 모델링은 머신러닝과 통계적 기법을 통해 순수한 마케팅 성과를 측정합니다. 광고 ID나 유저의 개인정보는 필요하지 않습니다. 디지털 채널에서 발생한 광고 지출, 콘텐츠 조회수, 앱 설치 수부터 TV 광고, 옥외 광고 데이터까지 6개월치만 준비하면 나만의 마케팅 믹스 모델을 만날 수 있습니다.
나만의 마케팅 믹스 모델이 측정한 마케팅 성과를 마케팅 믹스 분석 리포트에서 한눈에 조회해 보세요. 예산 최적화 리포트에서 확인할 수 있는 마케팅 믹스 모델이 추천한 최적의 예산안도 놓치지 마세요. 마케팅 믹스 모델을 바꾸고 싶다면 언제든 모델 매니저에서 상황에 맞는 모델을 선택하세요.
에어브릿지는 데이터에 숨어 있는 작은 힌트도 놓치지 않습니다. 에어브릿지가 지금 새로운 광고 성과 측정을 제시하겠습니다.
👉 [에어브릿지의 마케팅 믹스 모델링에 대해 더 자세히 알아보고 싶다면 마케팅 전문가와 직접 이야기 나눠보세요.](https://www.airbridge.io/ko/book-a-demo?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=marketing-mix-modeling-for-app-marketer)
> 이 글은 마케팅 믹스 모델링에 관한 2편의 백서를 요약한 포스트입니다. 마케팅 믹스 모델링에 대한 더욱 자세한 내용이 궁금하시다면 아래 링크를 통해 다운로드해 주세요.[
머신러닝에 기반한 앱 마케팅: 마케팅 믹스 모델링으로 성과 측정부터 예측까지](https://www.airbridge.io/ko/library/mmm-basics-whitepaper?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=airbridge-blog-mmm-basics-whitepaper&utm_content=ko)[
프라이버시 시대의 앱 마케터를 위한 마케팅 믹스 모델링 실전 활용 가이드](https://www.airbridge.io/ko/library/mmm-tips-whitepaper?utm_source=airbridge_blog&utm_medium=cta&utm_campaign=airbridge-blog-mmm-tips-whitepaper)
---
## 에어브릿지의 광고 기여 데이터를 GA4에서 확인하세요
URL: https://www.airbridge.io/ko/blog/airbridge-google-analytics-4
Language: Korean
Published: 2023-06-28
> *GA4에서 에어브릿지 광고 기여 데이터를 보고 싶다면 담당 에어브릿지 CSM과 확인 후 데이터 연동 작업을 시작해주세요.
최근 구글 애널리틱스에 접속했다면 구글 애널리틱스 4(Google Analytics 4, 이하 GA4) 이전을 유도하는 팝업창을 확인하셨을 텐데요. 오는 7월 1일부터 기존의 구글 애널리틱스인 유니버설 애널리틱스(Universal Analytics) 지원이 중단됩니다. 에어브릿지는 유니버설 애널리틱스가 지원 종료되는 시점에 맞춰 앱 설치 및 딥링크 오픈 이벤트를 GA4에 전송할 수 있도록 지원합니다.
GA4는 유저 행동을 분석할 수 있는 구글(Google)의 웹·앱 프로덕트 분석 솔루션입니다. GA4에 에어브릿지의 광고 기여 데이터를 전송하면 마케터는 유저가 어떤 광고로 프로덕트에 유입되었느냐에 따라 각 유저의 행동 패턴을 분석할 수 있습니다. 이 광고 기여 데이터에는 앱 설치 및 딥링크 오픈 이벤트에 기여한 채널, 캠페인, 광고 소재, 키워드 등이 포함됩니다.
GA4 도입은 이제 디지털 마케팅에서 거부할 수 없는 흐름입니다. 새로운 흐름에 따라 프로덕트를 분석하기 위해서는 누구든 GA4에 적응해야 합니다. 나아가 광고 기여 데이터를 활용해서 GA4로 프로덕트를 분석하고 싶다면 에어브릿지에서 앱 설치 및 딥링크 오픈 이벤트를 전송해야 합니다.
## 에어브릿지에서 GA4로 광고 기여 데이터 전송하기
그렇다면 에어브릿지에서 GA4로 광고 기여 데이터를 전송하기 위해서는 어떤 작업을 해야 할까요? 에어브릿지에서 아래 세 단계만 마치면 전송을 설정할 수 있습니다.
1. 에어브릿지에 구글 디바이스 ID 가져오기
2. GA4에서 확인한 인증 정보 에어브릿지에 입력하기
3. 에어브릿지에서 GA4로 전송할 광고 기여 데이터 선택하기
구글이 디바이스에 자체 부여하는 ID(App Instance ID)를 활용하면 에어브릿지에서 GA4로 광고 기여 데이터를 전송할 수 있습니다. 이 ID를 에어브릿지로 가져오면 광고로 발생한 앱 설치 및 딥링크 오픈 이벤트의 광고 기여 데이터를 담아서 GA4에 전송할 수 있습니다.
에어브릿지에 입력하기 위해 GA4에서 확인해야 하는 인증 정보는 2가지입니다. FIREBASE 앱 ID와 측정 프로토콜 API 비밀번호를 에어브릿지에 입력하면 에어브릿지에서 GA4에 등록한 앱으로 광고 기여 데이터를 전송할 수 있습니다.
에어브릿지가 전송하는 앱 설치, 딥링크 오픈 이벤트는 GA4에서 커스텀 이벤트로 기록됩니다. 두 이벤트는 GA4에서 앱 설치, 딥링크 오픈 이벤트를 집계하는 스탠다드 이벤트(first_open, session_start)와 다르지만, 광고 기여 데이터는 동일하게 기록됩니다. 마케터는 에어브릿지에서 두 이벤트 중에서 원하는 이벤트를 선택해서 GA4로 전송할 수 있습니다.
> ***주의하세요!**
에어브릿지의 앱 설치, 딥링크 오픈 이벤트는 커스텀 이벤트로, GA4에서 확인하는 스탠다드 이벤트(first_open, session_start)와 별개의 이벤트입니다.
👉 더 자세한 전송 설정 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/google-analytics-4)에서 확인할 수 있습니다.
## GA4에서 에어브릿지 광고 기여 데이터 활용하기
전송 설정을 모두 마쳤다면 에어브릿지가 전송한 광고 기여 데이터를 GA4에서 다양한 방법으로 활용할 수 있습니다. 광고 기여 데이터를 맞춤 측정기준으로 만드는 방법은 가장 유용한 방법 중 하나입니다. 에어브릿지에서 광고 기여 데이터를 전송하면 마케터는 GA4에서 앱 설치 및 딥링크 오픈 이벤트의 이벤트 속성을 맞춤 측정기준으로 만들 수 있습니다. 맞춤 측정기준으로 만든 이벤트 속성은 GA4의 모든 리포트에서 비교 기준으로 자유롭게 활용할 수 있습니다.
예를 들어 기여한 채널을 맞춤 측정기준으로 만들어서 어떤 채널로 유입된 유저가 프로덕트를 더 자주 사용하는지를 GA4로 알 수 있습니다.
[]
그외에도 아래 3가지 방법으로 에어브릿지에서 GA4로 전송된 광고 기여 데이터를 확인할 수 있습니다.
- 실시간 데이터: [실시간] 메뉴에서 ‘잠재고객 별 새 사용자 수’로 에어브릿지가 전송한 이벤트가 GA4에 실시간으로 잘 들어오는지를 유저 단위로 확인할 수 있습니다.
- 이벤트·유저 수: [이벤트] 메뉴에서 에어브릿지가 전송한 이벤트 수와 이벤트를 실행한 유저 수를 확인할 수 있습니다.
- 기여한 캠페인·채널: [보고서]>[참여도]>[전환수] 메뉴에서 ‘트래픽 소스’를 추가하면 에어브릿지가 전송한 이벤트에 기여한 채널과 캠페인 정보를 확인할 수 있습니다.
👉 더 자세한 광고 기여 데이터 활용 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/ko/guides/google-analytics-4#ga4%EC%97%90%EC%84%9C-%EC%97%90%EC%96%B4%EB%B8%8C%EB%A6%BF%EC%A7%80-%EB%8D%B0%EC%9D%B4%ED%84%B0-%EC%A1%B0%ED%9A%8C%ED%95%98%EA%B8%B0)에서 확인할 수 있습니다.
이렇듯 마케터는 에어브릿지가 전송한 광고 기여 데이터로 GA4에서 유저를 입체적으로 분석할 수 있습니다. 광고 기여 데이터를 활용하면 어떤 광고를 보고 프로덕트로 유입되었는지에 따라 각 유저의 행동 패턴을 마케팅 관점에서 해석할 수 있기 때문입니다. 지금 바로 에어브릿지와 GA4로 광고 기여 데이터와 프로덕트 내 유저 행동 데이터를 결합해서 더 나은 데이터 분석 환경을 만들어 보세요.
---
## 파이어베이스 다이나믹 링크(Firebase Dynamic Links) 서비스 종료 이후의 딥링크 활용
URL: https://www.airbridge.io/ko/blog/firebase-dynamic-links-shutdown
Language: Korean
Published: 2023-06-01
[]
[파이어베이스(Firebase)](https://firebase.google.com/?hl=ko)는 구글(Google)의 앱 개발 플랫폼으로 전 세계 수백만 개 앱과 게임의 빌드 및 성장에 기여하고 있습니다. 특히 파이어베이스의 동적 링크 혹은 다이나믹 링크(Firebase Dynamic Links)는 딥링크 솔루션으로 널리 사용되고 있습니다. 다만 최근 구글에서 2025년 8월 25일에 파이어베이스 다이나믹 링크 서비스를 종료할 예정이라고 밝혔습니다. 2025년 8월 25일이 되면 파이어베이스 다이나믹 링크 기반의 모든 링크가 작동을 멈추며 새로운 링크 생성도 불가능합니다. 그렇기 때문에 종료일 전에 모든 링크에 대한 마이그레이션 작업도 마무리되어야 합니다. 이에, 기존에 파이어베이스 다이나믹 링크를 통해 딥링크를 사용하고 있었다면 지금부터 대안을 마련해야 합니다.
파이어베이스 다이나믹 링크 서비스 종료와 함께 대체할 수 있는 딥링크 솔루션을 찾고 있다면, 이번 포스트를 확인하고 에어브릿지 딥링크(트래킹 링크)를 활용해 보세요.
> ⚠️ 주의: 본 블로그 포스트는 2023년 8월에 업데이트된 내용입니다. 구글의 최신 발표에 따라 수정해야 하는 일부 정보가 포함되어 있을 수 있습니다. 파이어베이스의 동적 링크 지원 중단 FAQ([링크](https://firebase.google.com/support/dynamic-links-faq))에서 계속해서 업데이트 되는 내용을 확인하세요.
## 파이어베이스 다이나믹 링크, 에어브릿지 딥링크로 대체하기
딥링크는 모바일 환경에 놓인 유저가 웹사이트 혹은 다른 앱에서 URL을 클릭했을 때, 앱을 실행시키고 특정 페이지로 이동하도록 돕는 기술입니다. 유저를 원하는 페이지로 이동시킬 수 있다는 점에서 앱 마케팅에서 매우 중요한 역할을 합니다. 하지만 기술적인 구현이 복잡하며 모든 상황에서 원하는 대로 작동시키기 어렵습니다.
이때 에어브릿지와 같은 어트리뷰션 툴/MMP에서 제공하는 트래킹 링크를 활용하면 손쉽게 딥링크를 설정하고 모든 상황에서 원하는 대로 작동시킬 수 있습니다. 물론, 파이어베이스 다이나믹 링크의 대안으로도 에어브릿지의 딥링크를 활용할 수 있습니다.
### 에어브릿지의 고도화된 딥링크 기술
에어브릿지는 딥링크를 효과적으로 활용할 수 있는 기능과 인프라를 제공하며, 어떠한 상황에도 문제없이 대응할 수 있습니다. 게다가 에어브릿지 딥링크를 활용하면 아래와 같은 이점도 추가로 누릴 수 있습니다.
- **유저에게 보다 개인화된 경험을 제공: **에어브릿지 딥링크에는 유저의 선호도, 지역, 리퍼러, 디바이스 종류 등을 반영한 파라미터(쿼리 스트링)를 추가하여 유저가 도달하는 페이지를 개인화할 수 있습니다. 이를 통해 유저에게 보다 개인화된 경험을 제공하여 유저 인게이지먼트와 만족도를 높일 수 있습니다.
- **오프라인과 온라인, 웹과 앱, 그리고 Android와 iOS 모두를 연결: **에어브릿지 딥링크는 크로스 플랫폼에 대해 완벽한 호환을 제공하여 소셜 미디어, 이메일, QR 코드 등 다양한 소스에서 유입된 유저를 모두 원하는 페이지로 안내할 수 있습니다. 이를 통해 유저에게 통합되고 신뢰할 수 있는 경험을 제공할 수 있습니다.
- **디퍼드 딥링크(Deferred Deep Linking, 지연된 딥링크) 기능 활용:** 앱을 설치하지 않은 유저도 광고 클릭으로 앱을 설치한 후 의도한 인앱 페이지로 바로 연결할 수 있습니다. 앱 설치 후 바로 클릭했던 광고와 연결된 페이지로 이동함으로써 구매 행동이 용이하도록 유도하여 전환율을 높일 수 있습니다. 게다가 에어브릿지의 [리메인 페이지](https://help.airbridge.io/ko/guides/redirecting-users-to-website) 기능을 통해 앱을 설치하지 않고 ‘뒤로 가기’를 누르더라도 관련 페이지로 리다이렉트 시킬 수 있습니다.
- **브랜디드 링크로 브랜딩 강화: **에어브릿지의 딥링크는 도메인은 물론, OG 태그의 미리 보기를 손쉽게 변경하여 브랜드를 자연스럽게 강화하고 신뢰를 높일 수 있습니다. Abit.ly 혹은 bitly 등을 통해 URL을 축약하는 대신, link.brand-name.com과 같이 우리 브랜드명을 담은 깔끔하고 보기 좋은 URL로 생성할 수 있습니다. 👉[브랜디드 링크 설정 방법 알아보기](https://www.airbridge.io/ko/blog/airbridge-branded-link)
- **대량 딥링크 생성 및 관리**: 에어브릿지는 API와 구글 스프레드시트를 통해 대량으로 간편하게 딥링크를 생성하고 효율적으로 관리할 수 있도록 템플릿을 제공합니다. 👉[트래킹 링크 제너레이터 템플릿에서 트래킹 링크 대량으로 만들기](https://help.airbridge.io/ko/guides/creating-bulk-tracking-links)
## 에어브릿지에서 딥링크부터 어트리뷰션과 마케팅 성과 분석까지
이처럼 에어브릿지의 고도화된 딥링크 기술을 활용하면 원하는 대로 편리하게 유저를 원하는 페이지로 이동시킬 수 있으며 브랜딩과 인게이지먼트 개선에도 기여할 수 있습니다. 이와 더불어 에어브릿지 딥링크는 트래킹 링크로써 유저로부터 발생한 행동에 어떤 채널이 기여됐는지 분석할 수 있는 어트리뷰션을 기반으로 마케팅 성과 분석까지도 한 번에 가능합니다. 유저 여정에서 최초로 발생하는 링크의 클릭부터, 이어지는 앱 설치 데이터 수집, 그리고 앱 내 특정 페이지로 연결하는 등 하나의 링크를 다양한 광고 캠페인에서 활용할 수 있습니다. 이를 통해 캠페인의 성과를 추적하고 측정하여 앱 마케팅 전략을 최적화하고 성장을 강화하는 데에 필요한 인사이트를 얻을 수 있습니다.
파이어베이스 다이나믹 링크에서 에어브릿지 딥링크로의 이관이 필요하다면 에어브릿지 팀이 도와드릴 수 있습니다. 지금 바로 [에어브릿지 데모](https://www.airbridge.io/ko/book-a-demo)를 신청하고 에어브릿지의 모바일 마케팅 전문가와 상담해 보세요.
> 딥링크에 대한 가장 쉬운 가이드, <[에어브릿지 핸드북 - 딥링크](https://www.airbridge.io/ko/library/airbridge-handbook-deeplinks)>를 다운로드하고 확인해보세요.
> **딥링크에 대해 더 자세히 이해하고 싶다면? 👉 딥링크101 시리즈 확인하기**
- [[딥링크101] 마케터와 개발자를 위한 딥링크 시작하기](https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers)
- [[딥링크101] iOS Safari에서 딥링크 동작 시 경고 문구가 뜨는 이유가 무엇일까요?
](https://www.airbridge.io/ko/blog/deeplink-101-ios-safari-alert)- [[딥링크101] 디퍼드 딥링크로 앱을 설치하지 않은 유저도 앱 내 특정 페이지로 연결하기
](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)**에어브릿지 트래킹 링크 설정 방법이 궁금하다면?**
👉 [에어브릿지 트래킹 링크로 심리스한 유저 경험 제공하기](https://www.airbridge.io/ko/blog/airbridge-tracking-link)
---
## 애플 SKAN 4.0을 제대로 활용하고 싶은 마케터가 꼭 알아야 하는 3가지
URL: https://www.airbridge.io/ko/blog/skadnetwork-4-must-know
Language: Korean
Published: 2023-05-04
iOS 앱 마케팅을 위해 마케터가 반드시 알아야 하는 것이 있습니다. 바로 애플의 어트리뷰션 프레임워크 SKAdNetwork(SKAN)입니다. SKAN은 광고 식별자를 사용하지 않는 광고 성과 측정법입니다. 유저의 개인정보를 보호하면서 마케터에게 마케팅 활동에 사용할 수 있는 정보를 제공합니다.
하지만 많은 마케터가 복잡한 SKAN 규칙으로 어려움을 겪고 있습니다. 광고 식별자로 광고 성과를 측정할 수 있지만, 마케터가 원하는 정보를 정확하게 얻기 힘들기 때문입니다. 이러한 상황에 처한 마케터를 돕기 위해 에어브릿지는 SKAN의 기본 개념과 SKAN의 주요 특징을 여러 포스트와 가이드로 제공하고 있습니다.
- [iOS의 SKAdNetwork(SKAN) 기본 개념과 측정방식 이해하기](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)
- [앱 마케터가 SKAdNetwork 4.0 (SKAN 4.0)에 주목해야 하는 이유](https://www.airbridge.io/ko/blog/exploring-the-marketing-implications-of-skadnetwork-4)
- [SKAN 포스트백 주의사항](https://help.airbridge.io/hc/ko/articles/13957607002905)
그리고 애플의 SKAN은 또다른 변화를 앞두고 있습니다. 애플은 지난 2022년 10월 SKAN 4.0을 공개했습니다. 공개한 정보에 따르면 SKAN 4.0은 SKAN 3.0처럼 개인정보를 보호하지만, 마케터에게 SKAN 3.0보다 더 많은 정보를 제공합니다. 아직 SKAN 4.0은 주요 광고 채널에서 지원하지 않지만, 곧 많은 광고 채널에서 지원할 예정입니다. 이에 앞서 마케터가 SKAN 4.0을 제대로 활용하기 위해 알아야할 3가지를 준비했습니다.
> 📌 **주요 내용 미리 보기
**
1. 광고 성과 데이터를 최대 3번 받을 수 있습니다.
2. 정해진 시간보다 더 빠르게 광고 성과 데이터를 받을 수 있습니다.
3. 애플에서 제공하는 숫자 또는 문자에 광고 성과 데이터를 지정할 수 있습니다.
## 광고 성과를 최대 3번 받을 수 있습니다
SKAN 4.0으로 측정된 광고 성과는 포스트백이라는 과정으로 제공됩니다. 마케터는 포스트백을 통해 제공된 광고 성과 데이터를 에어브릿지에서 확인할 수 있습니다. SKAN 4.0은 앱 설치 수 등에 따라 포스트백을 최대 3번 제공하며 제공되는 시점도 다릅니다. 따라서 마케터는 SKAN 4.0으로 광고 성과 측정에 사용할 수 있는 데이터를 최대 3번 받을 수 있습니다. 이 데이터에서 캠페인으로 발생한 인앱이벤트, 캠페인을 진행한 채널 관련 추가 정보 등 다양한 내용을 더 확인할 수 있습니다.
포스트백에 포함된 광고 성과 데이터는 포스트백 순서에 따라 다릅니다. 광고 성과 측정 기간이 다르기 때문입니다. 첫 측정 기간은 2일(앱 설치 기준 0~48시간)이고, 다음 측정기간은 5일(48~168시간), 28일(168시간~840시간)입니다. 포스트백에 포함된 광고 성과 데이터도 측정 기간에 따라 달라지게 됩니다.
[]
예를 들어 처음 성과를 측정하는 기간(첫 번째 컨버전 윈도우)은 앱 설치부터 48시간입니다. 첫 번째 포스트백의 전송 기간은 첫 번째 컨버전 윈도우가 끝난지 하루(24시간) 뒤에 시작됩니다. 포스트백 전송 기간은 24시간(앱 설치 기준 72~96시간)이며, 이 기간에서 임의의 시간에 전송됩니다. 첫 번째 포스트백에는 0~48시간에 측정된 데이터가 포함되어 있습니다. 두 번째 컨버전 윈도우는 첫 번째 포스트백 전송 기간과 관계없이 앱 설치 기준 3일차(앱 설치 기준 48시간)부터 광고 성과를 측정합니다.
## 광고 성과를 더 빨리 확인할 수 있습니다
두 번째 포스트백과 세 번째 포스트백을 너무 늦게 확인할 수 있다고 생각할 수 있습니다. 예를 들어 앱 설치 후 4일차에 측정된 데이터를 확인하고 싶어도 마케터는 두 번째 컨버전 윈도우가 끝나는 최소 7일차까지 기다려야 합니다. 세 번째 컨버전 윈도우가 진행 중이라면 1달 넘게 기다려야 합니다. 이런 상황을 해결할 수 있는 기능이 있습니다. 바로 락 윈도우(Lock Window)입니다.
[]
락 윈도우는 마케터가 미리 정한 유저 행동이 발생하면 컨버전 윈도우를 중지하고 24시간 후에 포스트백 전송 기간을 미리 시작하는 기능입니다. 모든 컨버전 윈도우에 적용할 수 있습니다. 예를 들어 두 번째 컨버전 윈도우에 락 윈도우를 설정하면 두 번째 포스트백을 더 빨리 확인할 수 있습니다. 미리 포스트백을 보내는 기능이기 때문에 원래 주는 포스트백은 제공되지 않습니다. 따라서 마케터는 이 락 윈도우로 SKAN 4.0에서 정한 원래 시간보다 더 빠르게 광고 성과 데이터를 받을지 안 받을지 원하는 데이터에 따라 결정할 수 있습니다.
주의사항이 한 가지 더 있습니다. 이전 컨버전 윈도우에 락 윈도우를 설정해도 다음 컨버전 윈도우가 앞당겨지지는 않는다는 것입니다. 예를 들어 5일차에 락 윈도우가 작동해서 두 번째 컨버전 윈도우가 2일 일찍 끝나도 세 번째 컨버전 윈도우는 2일 빠르게 시작하지 않습니다. 다음 광고 성과 데이터는 세 번째 컨버전 윈도우 기간 그대로 8 ~ 35일 동안 측정됩니다.
## 광고 성과를 애플에서 제공하는 숫자와 문자로 확인할 수 있습니다.
SKAN 4.0으로 받을 수 있는 광고 성과 데이터 중 가장 중요한 데이터는 전환값(Conversion Value)입니다. 전환값은 애플에서 광고 성과 데이터를 숫자와 문자로 바꿔서 제공하는 값입니다. 기존 SKAN 3.0은 전환값으로 0~63을 제공했습니다. SKAN 4.0은 상세전환값(Fine Conversion Value)과 단순전환값(Coarse Conversion Value)을 제공합니다.
상세전환값은 0 ~ 63까지 최대 64개로 제공되는 전환값입니다. 단순전환값은 하(Low), 중(Medium), 상(High) 총 3개로 제공되는 전환값입니다. 마케터는 64개의 숫자로 구성된 상세전환값과 3개의 문자로 구성된 단순전환값에 측정하고 싶은 광고 성과 데이터를 지정할 수 있습니다.
[]
상세전환값과 단순전환값은 포스트백 티어(Postback Tier)에 따라 각 포스트백으로 전송 여부가 결정됩니다. 포스트백 티어는 [군중 익명성(Crowd Anonymity)](https://developer.apple.com/documentation/storekit/skadnetwork/receiving_postbacks_in_multiple_conversion_windows#4103229)을 기반으로 결정됩니다. 군중 익명성은 특정 유저를 식별할 수 없는 정도를 나타냅니다. 이는 애플에서 포스트백 티어를 결정하기 위해 사용하는 개념입니다. 포스트백 티어는 0 ~ 3까지 총 4가지입니다. iOS 광고 캠페인으로 발생한 앱 설치 수가 많을수록 포스트백 티어는 높아집니다.
마케터는 포스트백 티어가 높을수록 자세한 전환값을 확인할 수 있습니다. 상세전환값은 단순전환값보다 자세한 정보를 담습니다. 상세전환값은 포스트백 티어가 2이상인 디바이스에서 오직 첫 번째 포스트백으로 채널에 전송됩니다. 단순 전환값은 포스트백 티어가 1인 디바이스에서 첫 번째 포스트백으로 채널에 전송됩니다. 또한 포스트백 티어가 1 이상인 디바이스에서 두 번째, 세 번째 포스트백으로 채널에 전송됩니다.
포스트백 티어가 0이면 어떤 전환값도 채널에 전송되지 않습니다. 포스트백 티어가 높더라도 유저가 앱에서 실행한 행동이 없다면 전환값으로 바꿀 광고 성과 데이터가 발생하지 않습니다. 따라서 이 경우에는 전환값이 포스트백으로 전송되지 않습니다.
## 에어브릿지는 마케터가 SKAN 4.0에 대비할 수 있도록 준비하고 있습니다.
에어브릿지는 마케터가 SKAN 4.0을 쉽게 활용할 수 있도록 준비하고 있습니다. 마케터는 에어브릿지에서 iOS 광고 캠페인으로 발생한 이벤트 종류와 횟수, 이벤트로 발생한 매출을 상세전환값과 단순전환값을 통해 측정할 수 있습니다. SKAN과 에어브릿지 기준 중복 기여 여부와 두 번째, 세 번째 포스트백으로 전송되는 광고 성과 데이터에서는 유저의 리텐션까지 확인할 수 있습니다.
마케터는 이처럼 SKAN 4.0으로 수집한 광고 성과 데이터를 에어브릿지 대시보드에서 다양한 기준으로 조회할 수 있고, 보다 입체적인 관점으로 광고 성과 데이터를 확인해 다음 iOS 광고 캠페인에서 효과적으로 활용할 수 있습니다.
에어브릿지와 함께 SKAN 4.0으로 성공적인 iOS 앱 마케팅을 진행하세요.
---
## 지금까지 알려진 구글 프라이버시 샌드박스의 거의 모든 것
URL: https://www.airbridge.io/ko/blog/google-privacy-sandbox-overview
Language: Korean
Published: 2023-04-26
> ⚠️ 주의: 본 블로그 포스트는 2023년 4월에 업데이트된 내용입니다. 구글(Google)의 최신 발표에 따라 수정해야 하는 일부 정보가 포함되어 있을 수 있습니다.
‘개인정보보호’는 아무리 강조해도 모자라지 않는 모바일 앱 마케팅의 지금 가장 중요한 키워드입니다. 이 키워드를 이끌고 있는 애플(Apple)과 구글(Google)등 대표적인 모바일 앱 플랫폼은 유저를 추적하지 않아도 광고 캠페인 성과를 측정하는 방법을 제시하고 있습니다. 이미 애플은 개인정보가 필요 없는 어트리뷰션 프레임워크 SKAdNetwork(SKAN)을 제공하고 있습니다.
그리고 곧 구글도 개인정보를 보호하면서 광고 캠페인 성과를 측정하는 방법을 제공하기 시작합니다. 바로 **구글 프라이버시 샌드박스(Google Privacy Sandbox)**입니다. 구글 프라이버시 샌드박스의 목표는 마케터가 비즈니스를 성장시키면서 유저의 개인정보를 보호하는 환경을 제공하는 것에 있습니다.
구글 프라이버시 샌드박스 도입은 모바일 앱 마케팅에 더 큰 변화를 불러올 것으로 예상됩니다. 개인정보를 보호하는 마케팅 환경은 어떤 환경일까요? 이제 리타겟팅은 어려워 질까요? 광고 캠페인 성과를 완전히 새로운 방법으로 측정해야 할까요? 이 질문의 답은 구글 프라이버시 샌드박스를 아는 데서 시작합니다.
## 마케터에게 구글 프라이버시 샌드박스는 어떤 의미일까요?
구글 프라이버시 샌드박스로 발생하는 가장 큰 변화는 서드파티 쿠키와 구글 광고 ID(GAID) 제공 중단입니다. 서드파티 쿠키는 크롬을 제외한 대부분 브라우저에서 지원을 중단하고 있어 예상됐지만, GAID는 다릅니다.
GAID는 유저를 기기 단위로 식별할 수 있는 유저 식별자입니다. Apple의 IDFA와 유사합니다. 에어브릿지와 같은 MMP를 통한 성과 측정을 위해서는 광고로 유입된 유저를 이 식별자로 구분하거나 식별하고, 식별한 유저의 앱 설치, 인앱 구매처럼 앱에서 발생하는 활동을 트래킹하기 때문에 필요합니다.
하지만 구글 프라이버시 샌드박스 시대의 시작과 함께 이 GAID가 역사 속으로 사라집니다. 이는 구글과 애플이 이끄는 개인정보보호 정책에서 가장 큰 차이 중 하나입니다. 애플은 앱 추적 투명성(ATT) 정책에 따라 개인정보 수집을 허용한 유저의 IDFA를 광고주에게 제공합니다. 반면, 구글은 개인정보 수집 허용 여부와 관계없이 GAID를 제공하지 않습니다. 구글 안드로이드의 시장 점유율이 세계적으로 70%를 넘기는 점을 고려하면 한국뿐만 아니라 글로벌 마케팅 전략에 큰 영향이 있을 수밖에 없습니다.
아직까지 구글은 GAID 지원을 중단하는 정확한 시기를 밝히지 않았지만, [올해 2월부터 안드로이드 유저를 대상으로 테스트를 시작했습니다](https://blog.google/products/android/the-privacy-sandbox-beta-is-coming-to-android/). 이와 더불어 구글 프라이버시 샌드박스의 또다른 특징도 하나씩 소개하고 있습니다.
## 구글 프라이버시 샌드박스, 애플의 SKAN과 무엇이 비슷하고 무엇이 다른가요?
구글 프라이버시 샌드박스는 개인정보를 보호하면서 광고 캠페인 성과를 측정할 수 있다는 측면에서 애플의 SKAdNetwork(SKAN)과 많은 부분에서 비슷합니다. 하지만 광고 캠페인 성과를 제공하는 방식이 다릅니다.
SKAN은 마케터는 개인정보 수집을 동의하지 않은 유저가 발생시킨 광고 성과도 SKAN을 통해서 확인할 수 있습니다. SKAN은 광고 성과를 집계된 형태로 지연해서 제공합니다. 구글 프라이버시 샌드박스도 개인 식별이 불가능한 환경을 제공합니다. 온라인 환경에서의 마케팅 활동을 통해 이루려는 성과를 미리 정의하고, 정의한 성과에 관해서만 집계된 형태로 지연해서 확인할 수 있습니다.
대신 구글 프라이버시 샌드박스는 항상 광고 캠페인 성과를 제공합니다. 현재 애플의 SKAN은 광고 캠페인 성과를 제한된 정보가 담긴 포스트백으로 제공합니다. 또 하루 동안 발생한 앱 설치 수도 특정 수치를 넘겨야 합니다. SKAN은 수치를 넘기지 못하면 포스트백 자체를 제공하지 않습니다. 구글 프라이버시 샌드박스에는 이런 제한이 없습니다.
구글 프라이버시 샌드박스는 항상 결과를 제공하지만, 유저 식별을 막기 위해서 결과의 추세를 바꾸지 않는 선에서 통계적으로 소음이 될 수 있는 가짜 데이터를 함께 집계해서 제공합니다. 이 가짜 데이터를 노이즈 데이터(noised data)라고 합니다. 노이즈 데이터는 수학적으로 마케팅 성과나 추세 자체를 유지하면서 원천적으로 개별 유저를 트래킹 할 수 없게 만드는 역할을 합니다.
## 지금 사용하는 광고 성과 측정 방법, 이제는 사용할 수 없나요?
구글 프라이버시 샌드박스 도입은 지금 사용하는 광고 성과 측정 방법에 큰 영향을 미칠 수밖에 없습니다. GAID가 삭제되기 때문입니다. GAID가 없으면 유저 활동을 트래킹하기 쉽지 않습니다.
하지만 다행히 일부 식별자는 계속해서 사용할 수 있습니다. 구글은 구글 플레이 인스톨 리퍼러(GIR)를 기존처럼 제공합니다. GIR은 유저가 광고를 클릭한 이후에 구글 스토어로 넘어가면 부여되는 식별자입니다. 또 에어브릿지와 같은 MMP가 성과 측정을 위해 사용하는 대부분의 자체 식별자도 여전히 사용할 수 있으며 MMP는 자체 식별자 등을 기반으로 고도화된 성과 측정을 계속해서 지원할 예정입니다.
구글도 다양한 마케팅을 지원하기 위해 추가적인 다양한 마케팅 정보를 API(Application Programming Interface)로 제공합니다. 지금까지 마케터가 데이터를 수집해서 타겟 유저를 확인하고 광고를 집행했다면, 구글 프라이버시 샌드박스 시대에서는 데이터 수집 및 실시간 동향 파악이 어려워집니다. 대신 구글은 자동화된 캠페인 운영을 API로 지원합니다. 구글이 제공하는 주요 API와 지원하는 마케팅 활동은 아래와 같습니다.
구글이 제공하는 API지원하는 마케팅 활동Topics API유저 개인의 행동 데이터 없이 관심사 기반 타겟팅 광고Protected Audience API (구 FLEDGE)서드파티에 데이터를 전송하지 않고,
기기에서 유저 개인의 행동 데이터를 기반으로 광고 입찰 및 송출Attribution Reporting API개인 정보를 사용하지 않고 기기에서 발생한 이벤트로 광고 성과를 측정
## 에어브릿지는 이미 준비하고 있습니다
마케팅 시장은 구글 프라이버시 샌드박스 도입으로 거대한 변화가 발생할 것으로 예상됩니다. GAID를 지원하지 않기 때문에 지금까지와 다른 광고 성과 측정 방법을 고민해야 합니다. 구글이 제공하는 정보를 마케팅에 어떻게 활용할 것인지 미리 고민이 필요합니다.
에어브릿지는 개인정보보호에 맞춰 이미 준비하고 있습니다. 에어브릿지는 구글 앱 어트리뷰션 파트너(AAP, App Attribution Partner)로 기존 광고 성과 모델을 고도화하고 있으며, 마케팅 믹스 모델링(MMM)과 같은 다양한 광고 성과 측정 방법도 지원하고 있습니다.
앱 마케팅의 변화, [에어브릿지](https://www.airbridge.io/ko/product/unified-measurement-stack)와 함께 효과적으로 대비해 보세요.
---
## 에어브릿지, SOC 2 유형 II 인증 취득으로 보안 역량 검증
URL: https://www.airbridge.io/ko/blog/airbridge-soc-2-type-ii
Language: Korean
Published: 2023-03-08
에어브릿지가 SOC 2 유형 II 인증을 획득했습니다. 이를 통해 에어브릿지의 보안 및 데이터 관리 역량이 글로벌 수준을 상회한다는 점이 객관적으로 검증됐습니다.
SOC 인증은 미국공인회계사협회(AICPA) 및 국제감사인증기준위원회(IAASB)가 제정한 기준에 따라 기업의 내부 통제 수준을 검증하는 제도입니다. 평가 항목과 감사 기간에 따라 세분화되며, 이번에 에어브릿지가 획득한 SOC 2 유형 II 인증은 ‘서비스 보안성’이 ‘일정 기간 이상’ 기준을 만족한 경우에만 주어집니다.
이번 인증을 위해 에어브릿지는 2022년 10월부터 12월까지 3개월에 걸쳐 딜로이트 안진회계법인의 감사를 받았습니다. 최종 제출된 감사 보고서에 따르면, 에어브릿지는 ▲보안성 ▲가용성 ▲개인정보보호 ▲처리무결성 ▲기밀성 등 5개 원칙에 기반한 인증 요건을 모두 충족했습니다.
에어브릿지는 높은 수준의 보안 통제 수준을 인정 받아 SOC 2 유형 II 인증을 획득한 소수의 국내 기업 중 하나가 되었습니다. 이는 누구도 적절한 절차 없이 에어브릿지 고객사의 민감 정보에 접근할 수 없다는 점을 보증합니다.
에어브릿지는 이 외에도 2020년 정보보호 분야 최고 권위 국제 인증인 ISO/IEC 27001을 최초 취득한 후 매년 갱신해 왔습니다. 또한, 올해 중으로 ISMS 정보보호 인증과 ISO/IEC 27017, ISO/IEC 27018 클라우드 서비스 정보보호 인증까지 취득할 계획입니다.
## 개인정보 보호 시대를 선도하는 에어브릿지
에어브릿지는 월 400억 건에 달하는 데이터를 실시간으로 수집하고 분석하는 모바일 어트리뷰션 툴/MMP입니다. 때문에 보안과 프라이버시를 최우선 가치로 두고 전문 인력 확보, 정기 외부 감사, 인증 범위 지속 확대 등의 노력을 기울이고 있습니다.
즉, 에어브릿지는 고객사가 보유한 귀중한 디지털 자산을 철저하고 엄격하게 관리하기 위한 인적·기술적 투자를 아끼지 않고 있습니다. 이를 통해 솔루션의 신뢰성과 안정성을 보장하고 안전한 비즈니스 환경을 제공함으로써 궁극적으로는 고객사의 성공에 보탬이 되고자 합니다.
이는 최근 모바일 생태계 내 개인정보 보호를 강화하고자 하는 움직임과도 일맥상통합니다. 에어브릿지의 고객사는 모바일 앱 개발·운영사이기 때문에, 에어브릿지 고객사의 데이터를 안전하게 관리하는 것은 곧 국내외 모바일 유저의 개인정보를 보호하는 것이라 볼 수 있습니다. 전 세계적으로 프라이버시가 중요한 화두로 떠오른 만큼, 에어브릿지는 앞으로도 책임감을 갖고 정보 보호에 힘쓰겠습니다.
> 💡 개인정보 보호 시대, 마케팅 성과 측정으로 고민이신가요? 에어브릿지가 발행한 [마케팅 믹스 모델링 관련 백서](https://www.airbridge.io/ko/library/mmm-basics-whitepaper)에서 해답을 찾아보세요.
---
## Airbridge-Braze 고급 연동으로 고객 인게이지먼트 강화하기
URL: https://www.airbridge.io/ko/blog/airbridge-braze-advanced-integration
Language: Korean
Published: 2023-02-15
에어브릿지(Airbridge)가 뛰어난 기술력을 바탕으로 고객 인게이지먼트 플랫폼 브레이즈(Braze)와의 연동을 강화했습니다.
에어브릿지는 지난 2021년부터 브레이즈의 기술 파트너(Technology Partner)로서 SDK 연동을 지원하여 어트리뷰션 데이터에 기반한 인게이지먼트 캠페인이 가능하도록 했습니다. 이번 연동은 서버 투 서버(S2S) 방식을 활용하여 보다 다양한 종류의 데이터, 한층 빨라진 데이터 송수신, 그리고 높은 유저 데이터 정합성을 자랑합니다.
연동을 완료하면 에어브릿지에 쌓이는 설치 이벤트 및 딥링크 오픈 이벤트의 어트리뷰션 데이터를 브레이즈로 전송할 수 있습니다. 이 중 설치 이벤트는 연동 완료 시 무조건 전송되고, 딥링크 오픈 이벤트는 선택 시에만 전송됩니다. 각 이벤트에 대한 어트리뷰션 데이터 세부 정보로는 Channel, Campaign, Ad Group, Ad Creative가 있으며, 각각을 브레이즈에서 세그먼트를 생성할 때 필터로 활용할 수 있습니다.
여성복, 남성복, 아동복을 모두 판매하는 이커머스 앱을 마케팅한다고 가정해 보겠습니다. 여러 광고 중 아동용 하의 사진이 포함된 광고를 클릭하여 유입된 유저를 에어브릿지로 파악한 다음, 브레이즈에서 해당 유저에 대한 세그먼트를 만들 수 있습니다. 그리고 그들을 대상으로 아동복 20% 할인 쿠폰이 포함된 인앱 메시지를 발송할 수 있습니다. 이와 같이 정밀하게 개인화된 마케팅 캠페인은 구매를 촉진할 뿐 아니라 유저를 잔존시킵니다. 즉, 어트리뷰션 툴과 세그멘테이션 툴을 함께 사용하면 유저와 모바일 앱 간 상호 작용을 효과적으로 증대시키고 ROI를 높이는 데 도움이 됩니다.
구체적인 연동 방법은 [에어브릿지 유저 가이드](https://help.airbridge.io/hc/ko/articles/900004368546)에서 확인하실 수 있습니다. 에어브릿지와 브레이즈를 SDK 레벨에서 연동한 적이 있다면 S2S 연동에 앞서 이전에 삽입한 [브레이즈 SDK 연동 관련 코드](https://developers.airbridge.io/docs/braze#%EC%97%B0%EB%8F%99)를 삭제하시길 권장합니다.
## 브레이즈 소개
브레이즈는 소비자에게 적절하고 기억에 남을 만한 브랜드 경험을 선사하는 통합 고객 인게이지먼트 플랫폼입니다. 브레이즈는 브랜드들로 하여금 상황에 맞게 여러 채널에서 소비자와 상호 작용하고 연결될 수 있도록 도우며, 이를 통해 빠르고 지속적으로 가치를 전달할 수 있도록 지원합니다. 자세한 내용은 [브레이즈 홈페이지](http://braze.com/)에서 확인해 보세요.
---
## 마케팅 믹스 모델링으로 마케팅 성과 다각도로 분석하기
URL: https://www.airbridge.io/ko/blog/marketing-mix-modeling
Language: Korean
Published: 2023-02-01
많은 마케팅 예산이 투입된 전략은 어떻게 분석하고 기획해야 최고의 성과를 낼 수 있을까요? 특히, 여러 채널에서 대규모 캠페인을 진행하거나 장기간의 마케팅 전략을 기획하는 경우엔, 어트리뷰션 분석 결과 혹은 트렌드나 감에 의존하기엔 위험이 큽니다. 이때는 지금까지 수행해온 전체적인 마케팅 활동(온라인 광고, 프로모션, TV 광고 등)이 매출에 어떤 영향을 미쳤는지 이해하는 것에서 시작해야 합니다. 즉, ‘마케팅 믹스 모델링’이 필요합니다.
[이전 포스트](https://www.airbridge.io/ko/blog/marketing-measurement-1-attribution)에서 마케팅 성과를 다각적으로 분석할 수 있는 방법으로 소개한 ‘어트리뷰션’ 모델들에 이어, 이번에는 거시적이고 장기적인 마케팅 전략과 미디어 믹스 결정을 위한 ‘마케팅 믹스 모델링’을 알아보겠습니다.
## 마케팅 믹스 모델링이란?
마케팅 믹스 모델링(Marketing Mix Modeling)이란, 마케팅 활동으로 인해 발생한 매출 및 ROI(투자 수익률) 증가를 수량화하여 산출하는 거시적인 데이터 기반 통계 분석 모형입니다. 온라인과 오프라인에 걸친 마케팅 성과를 종합적으로 측정 할 수 있는 검증된 솔루션으로, MMM 또는 미디어 믹스 모델링(Media Mix Modeling)이라고도 불립니다. 주로 분기 혹은 연 단위 등 장기간의 전략을 기획하는 데에 사용하며, 혹은 여러 채널에서 대규모 캠페인을 진행하는 경우에도 사용할 수 있습니다.
> 👉 [머신러닝에 기반한 앱 마케팅: 마케팅 믹스 모델링으로 성과 측정부터 예측까지](https://www.airbridge.io/ko/library/mmm-basics-whitepaper)에서 광고 성과 측정 및 캠페인 예산 최적화를 위한 마케팅 믹스 모델링 활용 전략을 확인해보세요
### 마케팅 믹스 모델링 분석 방법
마케팅 믹스 모델링은 ‘마케팅 전략’과 ‘비용’ 등을 변수로 놓고, 다양한 변수 사이의 상관관계를 분석하는 **회귀 분석(regression analysis)\***을 통해, 각각의 마케팅 활동이 매출에 미치는 영향을 알아내서 비즈니스 결과를 예측합니다. 즉, 과거의 데이터를 분석하여 각각의 변수가 지난 매출에 미친 영향을 파악하고, 향후의 마케팅 활동으로 인해 어떤 결과가 발생할지 예측할 수 있습니다. 이 분석 방법에는 기초적인 모델부터 복잡한 모델링까지 다양하게 존재하지만, 그 기반은 모두 회귀 분석\*입니다.
> **회귀 분석(Regression Analysis)\*** : 회귀 분석은 매개변수 모델(parametric model)을 이용하여 통계적으로 변수들 사이의 관계를 추정하는 분석방법이다. 주로 독립 변수(independent variable)가 종속 변수(dependent variable)에 미치는 영향을 확인하고자 사용하는 분석 방법이다. 회귀 분석은 다른 독립 변수들을 고정시키고 한 가지 독립 변수만을 변화시킬 때 종속 변수가 어떻게 변화하는지를 확인한다. 종속 변수와 관련이 있는 독립 변수를 찾을 때, 또 독립 변수들 간의 관계를 이해하고자 할 때 사용한다. _(출처:_ [_네이버 지식백과)_](https://terms.naver.com/entry.naver?docId=5141772&cid=60266&categoryId=60266)
마케팅 믹스 모델링은 **영향을 미칠 만한 변수를 모두 식별하는 것**에서부터 시작합니다. 이러한 변수 중 일부는 마케팅에 관련이 있을 것이고, 일부는 관련이 없을 수도 있습니다.
**마케팅 믹스 모델 분석에 주로 사용되는 변수들은 다음과 같습니다.**
- 기본 볼륨(Base volume): 마케팅 활동 없이도 발생했을 기본 매출
- 증분 볼륨(Incremental volume): 마케팅 활동으로 발생한 매출의 순수 증가분
- 매체 및 광고(Media and advertising): TV 광고, 잡지, 온라인 광고(유튜브 영상, 키워드 광고) 등
- 판촉 활동(Trade promotions)
- 금액(Pricing)
- 유통(Distribution)
- 출시(Launches): 새로운 프로덕트의 출시
- 경쟁(Competition)
## 마케팅 믹스 모델링 공식
마케팅 믹스 모델은 변수와 KPI 간의 통계적 상관관계를 보여주는 수학적 공식으로 표현할 수 있습니다. 대표적으로 페이스북에서는 마케팅 믹스 모델링 공식을 아래와 같이 설계합니다.
**KPIt = ß0 + ß1 시즌별 특성t + ß2 브랜드 인지도t + ß3 TVRt +ß4 Facebook 노출t + ß5 가격 + ...기타 모든 요인 + 오류t**
- KPIt : 모델링 할 시점의 핵심 성과 지표
- β0: 기준 성과, 또는 다른 모든 요인이 최소일 때(미디어, 프로모션, 시즌별 특성 또는 가격 관련 요인이 없을 때)의 성과
- β: 계수, 또는 변수의 변화가 KPI에 미치는 영향
마케팅 믹스 모델을 사용하면, 모델에 포함된 독립 변수를 기반으로 제곱 오차를 최소화하는 계수(β)를 찾음으로서 KPI를 가장 정확하게 예측할 수 있습니다.
## 마케팅 믹스 모델링을 사용하는 주요 목적
1. **마케팅 활동이 판매에 미치는 영향을 이해할 수 있습니다.**
마케팅 믹스 모델링을 사용하면, 다양한 마케팅 활동들이 비즈니스 메트릭에 얼마나 영향을 미쳤는지(**과거 성과 분석**), 그리고 어떻게 영향을 미칠 것인지(**미래 예측**) 이해할 수 있습니다. 통계적 모델을 사용하여 KPI(핵심 성과 지표)에 마케팅 활동 및 이외의 외적 요인(가격, 프로모션, 시즌 특성 등) 각각이 미치는 영향을 구분하여 살펴보는 것입니다. 이를 통해 온라인과 오프라인에 걸쳐 진행한 마케팅 캠페인이 매출 증가에 미치는 영향을 정량화할 수 있습니다. 이러한 정보가 있으면 ROI(Return on Investment)와 같은 핵심 지표를 계산하는 데에 도움이 됩니다.
2. **합리적인 미디어 믹스 구성을 결정할 수 있습니다.**
다양한 마케팅 이니셔티브들의 효과를 측정함으로써, 앞으로의 미디어 믹스를 최적화하고 광고 캠페인을 개선하기 위한 인사이트를 얻을 수 있습니다. 특정한 채널의 예산을 조정할 경우, 어떻게 성과가 변화할 것인지 예측할 수 있기 때문입니다.
3. **전략적 의사결정을 위한 정보를 제공합니다.**
마케팅 믹스 모델을 통해 얻은 인사이트로 수익성을 따져서 투자의 우선순위를 결정하는 데에도 활용할 수 있습니다. 시나리오 계획과 예측을 통해, 마케팅 외적인 부분에 대한 의사결정 도구로도 사용할 수 있습니다.
## 오늘날 마케팅 믹스 모델링이 필요한 이유
마케팅 믹스 모델링은 거시적인 지표들이 포함되어 있기 때문에 **각 유저 레벨에서 트래킹이 불가능하더라도 마케팅 성과를 분석하고 효율을 개선하는 데에 사용할 수 있습니다**. 즉, 터치포인트가 존재하지 않거나 수집할 수 없는 상황에서도 마케팅 성과를 측정할 수 있습니다. 애플의 iOS 14.5와 새로운 앱 추적 투명성(App Tracking Transparency, ATT) 정책 등 광고 시장에서 유저의 개인 정보 보호가 더욱 중요해지는 트렌드 속에서, 터치포인트로 마케팅 성과를 분석하는 어트리뷰션 모델을 보완할 수 있다는 점에서 중요합니다.
또한, **수치화할 수 없는 다양한 변수들도 고려해서 분석할 수 있다는 것도 장점입니다**. 최근의 코로나 팬데믹의 영향과 같은 거시적인 경제 영향력, 경쟁사의 상황, 계절성, 프로모션 행사 등도 함께 포함시킬 수 있습니다. 이를 통해 전체적인 미디어 포트폴리오에 걸친 고차원적인 분석을 할 수 있게 되어, 장기적인 전략 수립을 위한 인사이트를 도출할 수 있습니다. 그리고 이 인사이트를 기반으로 효율적인 마케팅 예산 분배도 가능해집니다.
> 👉 왜 지금 모바일 업계에서 마케팅 믹스 모델링이 주목 받는지에 대해 더 자세히 알아보고 싶다면, [개인정보보호 시대 제대로 마케팅 성과를 측정하는 방법, 에어브릿지 마케팅 믹스 모델링](https://www.airbridge.io/ko/blog/airbridge-marketing-mix-modeling) 포스트에서 확인해보세요.
## 마케팅 믹스 모델링의 한계
한편, 마케팅 믹스 모델링도 한계는 존재합니다. 단기적인 전략 혹은 진행 중인 마케팅 활동에 대한 인사이트를 얻기에는 적합하지 않은 방법입니다. 마케팅 결과에 대한 평균을 결정하기 위해 과거의 데이터에 의존한 모델이기 때문에, 최근에 발생한 채널의 변화나 비즈니스 변화에 대해서는 확인하기 어렵습니다.
또한, 과거 비즈니스 결과에 대한 마케팅 영향력을 '확률'에 기반에서 예측하는 것이기 때문에, 상관관계 vs 인과관계 딜레마에 빠질 수도 있습니다. 비 마케팅적인 요소들에 대해 많은 가정에 의존하게 될 수 있으며, 채널의 종류나 비용 등이 과거와 미래가 다르지 않을 것이라는 가정도 필요합니다.
즉, 빠르게 변화하는 현실 속에서 마케팅 믹스 모델링으로 실시간 데이터를 분석하여 날렵하고 최적화된(애자일한) 의사 결정을 내리기는 어렵습니다.
## 진짜 마케팅 성과 분석에 더 가까워지기 위한 다각적인 분석
마지막으로, 중요한 점은 지금까지 살펴본 다양한 마케팅 성과 분석 모델(라스트 터치 어트리뷰션, 멀티 터치 어트리뷰션, 마케팅 믹스 모델링) 중 **모든 제품과 서비스에 적용되는 단 하나의 완벽한 정답은 없다**는 겁니다. 제품과 서비스의 특성에 따라 사용해야 하는 모델이 달라질 수 있으며, 채널별 마케팅 효과를 종합적으로 측정하기 위해서는 다각적으로 접근하는 것이 좋습니다. 이때 아래와 같이 세 가지 모델들의 장점을 살려서 활용할 수 있습니다.
[__wf_reserved_decorative]
- **라스트 터치 어트리뷰션 모델**을 통해 매일의 마케팅 성과를 확인하며 최적화합니다.
- **멀티 터치 어트리뷰션 모델(MTA)**을 통해 라스트 터치 어트리뷰션 모델의 단점을 주기적으로 보완합니다. 마지막 터치포인트는 물론 이전에 발생한 다른 터치포인트들의 가치를 제대로 이해해서 각 채널들의 성과를 제대로 분석합니다.
- 분기나 연 단위의 전략이나, 개인 정보 보호 강화로 데이터 수집 및 측정이 불가능한 영역에서의 미디어 믹스 전략은 **마케팅 믹스 모델링(MMM)**을 통해 측정하고 확인합니다.
다각도의 마케팅 분석을 통해 현재 마케팅 분석 측면에서 고려해야 하는 요소들을 모두 다룰 수 있으면서도 진짜 마케팅 성과 분석에 더 가까워질 수 있습니다. 이를 통해 마케터들은 고객들의 구매 여정 및 구매에 영향을 미치는 요인을 정확하게 분석하고, 투입한 비용을 극대화를 위한 (즉, ROAS의 상승을 위해) 캠페인을 최적화할 수 있게 될 겁니다.
“마케팅 성과를 다각도로 분석하기 위한 3가지 모델” 1,2편을 통해 각 모델들의 장단점을 이해하고 비즈니스의 성장을 위한 다각도의 인사이트를 얻는 데에 도움이 되기 바랍니다.
- [마케팅 성과를 다각도로 분석하기 위한 3가지 모델 1편 - 어트리뷰션(LTA, MTA)](https://www.airbridge.io/ko/blog/marketing-measurement-1-attribution)
---
## 광고로 유입된 유저의 단계별 여정을 에어브릿지 퍼널 리포트로 분석해 보세요
URL: https://www.airbridge.io/ko/blog/airbridge-funnel-report
Language: Korean
Published: 2023-01-31
광고로 유입된 유저의 여정을 여러 이벤트로 자유롭게 선택하여 분석할 수 있는 퍼널 리포트(Funnel report) - Beta가 공개되었습니다. 우리 앱 서비스로 유입된 유저의 서비스 진입부터 전환까지의 고객 여정을 단계별로 나누어 살펴볼 수 있으며, 어디에서 유저 이탈이 주로 발생하는지 혹은 어디에서 다음 단계 전환까지 오래 시간이 소요되는지 확인해볼 수 있습니다. 퍼널 리포트를 통해 유저 여정을 시각적으로 분석하고 마케팅 성과를 높일 수 있는 부분에 대한 인사이트를 확인해보세요.
## 에어브릿지 퍼널 리포트
### 퍼널(Funnel)이란?
모든 유저는 구매, 인앱 결제 등 각 비즈니스가 정의하는 ‘전환’에 이르기 전까지 반드시 몇 가지 단계(Funnel, 깔대기)를 거쳐가게 됩니다. 이를 마케팅에서는 [퍼널(Funnel)](https://blog.ab180.co/posts/app-marketing-funnel) 이론으로 설명합니다. 이 퍼널은 유저가 완료하는 일련의 이벤트(앱 설치, 회원가입, 로그인, 상품 검색, 장바구니 담기 등)로 정의될 수 있습니다. 그리고 이 퍼널을 분석하게 되면 각 비즈니스가 의도하는 대로 유저가 행동하는지 확인할 수 있습니다.
### 에어브릿지 퍼널 리포트는,
에어브릿지는 앱 어트리뷰션 툴 / MMP로서 매체가 제공한 어트리뷰션 정보 그리고 에어브릿지가 수집한 데이터를 바탕으로 광고부터 앱 내까지 이어지는 유저 여정을 전반적으로 분석할 수 있는 퍼널 리포트를 제공합니다. 그리고 에어브릿지가 자체 개발한 데이터베이스인 Luft 기반의 코호트 기능을 활용하여 분석하고 싶은 특정 유저군을 설정하고 해당 유저군의 행동을 분석할 수 있습니다.
특히, 에어브릿지의 퍼널 리포트는 SAN(Self Attributing Network)에 대해서도 유입 채널부터 전환에 이르기까지 고객 여정을 분석할 수 있다는 강점이 있습니다. 일반적으로 SAN은 굉장히 많은 유저의 데이터를 가지고 있기 때문에, 자체적으로 엄격한 데이터 관리 규정을 가지고 있습니다. 그래서 파트너십을 통해 자사 유저의 데이터를 안전하게 관리할 수 있는 MMP에게만 데이터를 제공하고 있으며 파트너십이 맺어지지 않은 써드파티을 연동할 때는 해당 데이터를 보낼 수 없습니다. 그렇기 때문에 프로덕트 애널리틱스 툴과 MMP를 연동했다고 하더라도 SAN 어트리뷰션 데이터를 포함한 고객 여정은 분석할 수 없다는 한계가 존재했습니다. 그러나 MMP인 에어브릿지의 퍼널 리포트는 이것을 가능케 합니다.
에어브릿지 퍼널 리포트는 유저의 행동, 즉 이벤트를 ‘스텝(step)’으로 정의합니다. 그리고 앱 설치 및 앱 오픈 등과 같은 ‘유저 유입’부터 구매 등 목표로 하는 ‘전환’까지의 유저 행동을 줄세워서 서비스 진입부터 전환까지의 고객 여정을 단계별로 나누어 살펴볼 수 있습니다. 이를 통해 유저 행동의 트리거가 되는 유입 이벤트부터 최종 전환 이벤트 사이에 발생하는 유저의 행동을 유저 여정 전반에 걸쳐 분석할 수 있습니다.
[]
### 우리 비즈니스에 에어브릿지 퍼널 리포트가 필요한 이유
에어브릿지 퍼널 리포트의 가장 큰 장점은, 유저 행동의 트리거가 되는 유입 이벤트부터 최종 전환 이벤트 사이에 발생하는 유저들의 행동을 차트와 테이블로 세세하게 분석할 수 있다는 점입니다.
그래서 최근 앱을 설치한 유저가 특정 서비스를 사용하고 있는지, 유입된 유저가 결제를 완료하는 데 얼마나 시간이 걸리는지, 광고로 유입된 유저가 어떤 단계에서 이탈하는지, 최근 일주일의 유저 전환율 등 마케팅 전략을 개선할 수 있는 다양한 인사이트를 얻을 수 있습니다. 그리고 마케팅 캠페인 이후 고객 여정을 분석함으로써 다음 캠페인의 성과를 증대시킬 수 있는 마케팅 전략을 준비할 수 있습니다.
## 퍼널 리포트 설정 방법
[]
### 코호트
코호트를 통해 퍼널 리포트의 분석 대상이 되는 유저를 설정합니다. 각 서비스에서 타겟하는 유저 정의를 적용하여 코호트로 구성할 수 있습니다. 예를 들어, 게임 서비스에서 ‘특정 레벨을 달성한 후 2주간 앱에 로그인 하지 않은 유저’ 혹은 ‘지난 한 달간 앱을 설치한 유저’로 코호트를 설정하여 캠페인을 진행했다고 가정하면, 이 조건에 충족하는 유저를 코호트로 설정하고 이 유저들의 행동을 퍼널 리포트로 분석할 수 있습니다.
이때 코호트 조건을 설정하면 해당 조건을 만족하는 유저를 대상으로 리포트 조회 결과를 집계하며, 미설정 시에는 전체 유저를 대상으로 분석합니다.
[]
### 스텝
스텝은 유저 이벤트 중 퍼널 리포트 분석에 사용하기 위해 선택하는 이벤트입니다. 코호트로 설정된 유저 여정을 고려하여 선택해야 하며, 2개 이상의 스텝 선택이 반드시 필요합니다.
스텝으로는 모든 에어브릿지 이벤트를 선택할 수 있으며, Install, Open, 회원가입, 로그인, 상품리스트 조회, 광고 노출, 광고 클릭 등이 있습니다.
또한, 선택한 스텝의 순서도 중요한 설정입니다. 에어브릿지 퍼널 리포트는 스텝에 해당하는 유저 중 다음 스텝으로 얼마나 많이 전환됐는지 집계하기 때문에 스텝 순서를 바꾸면 리포트 결과가 바뀔 수 있다는 점을 유의해야 합니다.
### 기간 설정 - 퍼널 진입 기간과 컨버전 윈도우
퍼널 리포트를 조회하기 위해서는 퍼널 진입 기간과 컨버전 윈도우 두 가지의 기간을 반드시 설정해야 합니다.
‘퍼널 진입 기간’은 코호트로 설정한 유저가 첫 스텝에 진입하는 기간입니다. 퍼널 리포트는 설정한 코호트에서 퍼널 진입 기간 동안 첫 스텝을 발생시킨 유저만 집계합니다.
‘컨버전 윈도우’는 첫 스텝 이후 발생한 후속 스텝을 집계하는 기간입니다. 퍼널 리포트는 첫 스텝 이후 컨버전 윈도우 동안 발생한 후속 스텝에서 유저가 얼마나 전환됐는지 집계하며, 컨버전 윈도우 이후 발생한 후속 스텝은 집계되지 않습니다.
### 그룹바이
퍼널 리포트의 그룹바이는 유저를 나누는 기준입니다. 퍼널 분석을 통해 마케팅 성과를 높일 수 있는 부분을 찾기 위해서는 유저 그룹을 나누어 비교하거나 채널, 캠페인 등의 조건으로 나누어 비교하는 과정이 필요합니다. 이를 위해 그룹바이를 활용할 수 있습니다.
퍼널 리포트의 그룹바이는 첫 스텝을 기준으로 유저를 나누고 집계합니다. 예를 들어 Channel을 그룹바이로 적용하면 첫 스텝이 기여된 채널에 따라 유저가 각각 집계됩니다. 어떤 유저가 첫 스텝에서 채널 A로 집계되었다면, 해당 유저의 두번째 스텝이 어느 채널에 기여되었는지는 고려하지 않습니다.
[]
더 자세한 에어브릿지 퍼널 리포트 이용 방법은 아래 링크와 연결된 유저 가이드 문서에서 확인할 수 있습니다. 에어브릿지 퍼널 리포트를 통해 유입 이후 유저의 행동을 단계별로 분석하고, 다음 캠페인의 성과를 보다 개선할 수 있는 인사이트를 얻어보세요.
[]
---
## 에어브릿지, 구글 앱 어트리뷰션 파트너(AAP) 선정
URL: https://www.airbridge.io/ko/blog/airbridge-google-aap
Language: Korean
Published: 2023-01-17
앱 어트리뷰션 툴 / MMP 에어브릿지(Airbridge)가 [구글 앱 어트리뷰션 파트너(AAP, App Attribution Partner)](https://ads.google.com/intl/en_us/lp/appcampaigns/attribution/)로 당당히 이름을 올렸습니다.
구글은 [전 세계 43억 명에 달하는 유저](https://www.semrush.com/blog/google-search-statistics/)를 기반으로 온라인 광고 시장 시장 점유율 부동의 1위를 차지하고 있어, 고가치 유저를 빠르게 확보하고 비즈니스 성장을 도모하려는 앱이라면 반드시 주목하고 활용해야 하는 플랫폼입니다. 2024년 안드로이드 프라이버시 샌드박스 도입 등을 통해 개인정보 보호 강화에 앞장서고 있기도 합니다.
구글 AAP 프로그램은 마케터들이 구글의 모바일 광고 상품을 효과적으로 활용하고 정확한 데이터에 기반하여 합리적인 결정을 내릴 수 있도록 하는 데 그 의의가 있습니다. 때문에 구글은 엄격한 심사를 통과한 일부 기업에만 AAP 자격을 부여하며, 구글 애즈 캠페인에 관한 정확성과 신뢰성을 갖춘 솔루션의 기준을 제시합니다.
에어브릿지는 그간 최고 수준의 마케팅 성과 측정 및 분석, 그리고 모바일 사용자에 대한 깊은 통찰을 제공하기 위해 다방면에서 노력해 왔습니다. 그리고 이번 파트너십 확대를 계기로 구글과의 연동을 한층 강화하여 국내외 고객사에 더욱 큰 가치를 전달할 수 있을 것으로 기대하고 있습니다. 구체적으로는 쉽고 간단한 캠페인 설정, 풍부한 데이터 및 면밀한 분석 제공, 구글 애즈 최신 기능의 선제적 도입을 지원할 예정이며, 이를 통해 빠르고 효율적인 마케팅 의사결정에 기여하고자 합니다. 또한, 구글과의 긴밀하고 지속적인 협력을 통해 고객사가 빠르게 변화하는 디지털 마케팅 환경에서도 성공적으로 캠페인을 기획하고 실행할 수 있도록 도울 계획입니다.
## 에어브릿지에 대해 더 잘 알고 싶으신가요?
에어브릿지는 LTA(Last-touch Attribution) + MTA(Multi-touch Attribution) + MMM(Marketing Mix Modeling)을 모두 제공하는 Unified Measurement Stack으로, 머신러닝과 데이터를 활용하여 마케팅 성과에 대한 정교하고 균형 잡힌 분석을 제공합니다. 또한, 단순 성과 측정에서 그치기보다, 측정을 통한 마케팅 최적화 및 성장에 기여하고자 합니다.
이외에도 에어브릿지는 MMP의 편리한 활용을 돕고자 Android, iOS 앱 및 웹에서 발생한 성과를 하나의 대시보드에서 모두 한눈에 볼 수 있도록 지원하고 있습니다. 딥링크, 오디언스 생성, 프로드 방지 등 모바일 성장에 필요한 기능도 올인원으로 지원합니다.
앞으로도 에어브릿지는 마케터에게 진정 필요한 솔루션으로서, 다양한 성공 사례와 인사이트를 공유하고 최고의 서비스를 제공하겠습니다.
에어브릿지가 궁금하시다면, [웹사이트](https://www.airbridge.io/ko)를 방문하거나 [앱 마케팅 전문가와 직접 이야기](https://www.airbridge.io/ko/book-a-demo) 나눠보세요. 혹은 [블로그](https://www.airbridge.io/ko/blog)에서 최신 애드테크 & 마테크 트렌드와 인사이트를 확인해 보세요.
---
## 새롭게 떠오른 마케팅 믹스 모델링, 무엇이 달라졌을까요?
URL: https://www.airbridge.io/ko/blog/modern-marketing-mix-modeling
Language: Korean
Published: 2023-01-13
과거부터 존재한 것이라도 새로운 기술의 등장과 함께 새롭게 떠오를 수 있습니다. 새로운 기술이 새로운 가능성을 만들어내죠. 마케팅 믹스 모델링(MMM, Marketing Mix Modeling)이 지금 전성기를 맞이한 이유입니다.
마케팅 믹스 모델링은 수치화해서 측정하기 어려웠던 TV 광고나 오프라인 광고를 포함한 모든 마케팅 비용과 계절성(seasonality) 및 거시 경제 상황이 판매에 미치는 효과를 측정하기 위해 사용되었습니다. 구체적으로 정의하자면 비용 및 매출 등의 다양한 마케팅 변수들과 성과 사이의 상관관계를 기반으로 광고의 기여도를 추정하는 통계학적 회귀 모델입니다. 모델이 분석한 결과를 바탕으로 마케팅 전략을 조정하고 최적화하여 향후 마케팅 활동의 다양한 시나리오를 시뮬레이션하며 판매를 예측할 수 있다는 장점으로 인해 활용되었습니다.
이 마케팅 믹스 모델링은 1960년대부터 존재했던 기법이지만, 과거에는 긴 시간과 많은 리소스가 투입된다는 제약 조건들로 인해 일부 기업들에서만 활용할 수 있었습니다. 하지만 개인정보보호 강화 흐름에 대응하는 마케팅 성과 측정 방법으로 다시 주목받고 있으며, 기술의 발달로 인해 기존의 제약 조건을 극복할 수 있는 환경이 만들어지며 보다 많은 기업과 마케터들이 마케팅 믹스 모델링을 활용할 수 있게 되었습니다. 이렇게 새로운 전성기를 맞이한 마케팅 믹스 모델링은 과거와 무엇이 달라졌을까요? 특히 어떤 변화 과정을 거쳐 SaaS(Software as a Service) 형태의 솔루션에 이르렀는지 알아봅니다.
## 과거부터 존재해 온 두 가지 마케팅 믹스 모델링 구현 방법
마케팅 믹스 모델링은 변수 생성과 모델링 과정이 매우 복잡하고 까다롭기 때문에 전문가의 도움을 필요로 하며, 이를 도와줄 수 있는 다양한 기관과 컨설턴트가 존재해왔습니다. 마케팅 믹스 모델링을 활용하는 데 사용되었던 주요 방법은 1) 컨설팅 서비스와 2) 내부에 직접 솔루션을 구축하는 방법(in-house) 두 가지입니다.
1) 컨설팅 서비스로 제공받는 마케팅 믹스 모델링
컨설팅 서비스는 과거부터 주로 큰 규모의 기업들이 활용했던 방법으로, 컨설팅사에 데이터를 보내서 모델을 구축하고 결과를 확인했습니다. 이 방법은 복잡하고 까다로운 준비와 분석 과정에서 전문가의 도움을 받을 수 있고, 데이터를 전달하면 모델링과 분석까지 진행하는 엔드 투 엔드(end-to-end) 서비스라는 장점이 있습니다.
하지만 그만큼 비용이 많이 투입되고 긴 시간이 소요됩니다. 주로 1년 이상의 데이터를 필요로 하며 전문 컨설턴트가 수동으로 데이터 확인, 모델 구축 및 확인, 보고서 생성 등의 과정을 진행합니다. 이 과정에서 맞춤 모델을 구축하여 분석을 진행하기 때문에 결과를 얻기까지 긴 시간이 소요되는 경우가 많습니다. 그리고 한 번 모델을 구축하여 결과를 도출한 이후에는 분기 혹은 반기로 업데이트를 제공하기 때문에 빠른 시장의 변화에 대응하기 어렵습니다. 이러한 한계로 인해 주로 마케팅 예산이 큰 기업들이 장기적인 마케팅 전략을 수립할 때 컨설팅사를 통해 마케팅 믹스 모델링을 활용해왔습니다.
2) 사내에 마케팅 믹스 모델링 구축(in-house)
또 다른 마케팅 믹스 모델링 활용 방법으로는 사내에 마케팅 믹스 모델링을 직접 내재화하여 구축하는 것입니다. 사내에서 모든 데이터를 관리하고 대응하고 싶다면 이 방법을 통해 외부에 데이터를 반출하지 않고 통제할 수 있으며, 내부 인력에 대한 비용 외에 추가적인 비용이 들지 않을 수 있습니다.
다만, 이 과정 또한 컨설팅 서비스와 마찬가지로 상당한 시간과 리소스가 필요합니다. 예를 들어 사내에 데이터 분석 및 데이터 사이언스 전문 인력이 없다면, 적합한 인력을 찾고 고용하는 데에 비용이 들어갑니다. 만약 이미 있다고 하더라도 다른 업무 외에 마케팅 분석만을 위한 리소스를 많이 투입하게 될 것이며, 처음 구축한 이후 유지 보수에 들어가는 리소스도 염두 해야 합니다. 또한 내부에서 자체적으로 목표 설정 및 모델 실행과 분석을 진행하는 만큼, 외부 변수의 영향력에 대한 고려가 충분하지 않을 수 있고 외부 데이터 활용을 위한 추가적인 리소스가 필요할 수 있습니다.
### 과거의 마케팅 믹스 모델링 방법의 한계
과거의 마케팅 믹스 모델링 제공사를 이용하기 위해서는 앞서 언급한 것처럼 인적 물적 자원이 모두 많이 소모되기 때문에 서비스의 접근성에 대한 진입 장벽이 높았습니다.
마케팅 믹스 모델링의 활용을 위해 컨설팅 서비스를 받거나 사내에 직접 모델을 구축하는 경우, 대부분 과정에 사람이 직접 개입하며 수동으로 진행해야 했습니다. 예를 들어 컨설팅 서비스를 받기 위해서는 1년 이상의 데이터를 컨설팅사에서/모델링 전문가가 요구하는 형태로 가공하여 전달해야 했고, 모델 구축이나 업데이트 과정이 모두 수동으로 진행되었습니다. 그래서 긴 시간이 소요되었고, 결과도 보고서 방식으로 제공되어 유저 친화적이지 않았습니다.
결과적으로, 빠르게 성장하는 초기 단계의 기업이나 작은 규모의 마케팅 비용을 운용하는 경우 마케팅 믹스 모델링을 활용하기 어려웠으며, 리소스에 대한 부담을 감당할 수 있는 기업들만 이용할 수 있는 구조였습니다.
## 마케팅 믹스 모델링이 새롭게 떠오르고 있습니다.
앞서 언급한 한계로 인해 소수의 기업들에서만 이용했던 것이 큰 문제로 여겨지지는 않았습니다. 그렇지만 iOS 14.5 업데이트 및 쿠키 관련 정책의 변화 등 개인정보보호가 강화되는 트렌드에 따라 마케팅 믹스 모델링에 대한 필요성이 증대되며 새로운 활용 방법에 대한 니즈가 증가했습니다.
기존에 대부분의 광고주들은 마케팅 성과를 측정하고 전략을 수립하는 데에 어트리뷰션 툴 / MMP를 주로 활용해왔습니다. 하지만 MMP는 어트리뷰션을 위해 유저 레벨 데이터에 의존하고 있기 때문에 필요한 데이터를 제대로 수집하지 못하는 현재의 환경에 제대로 대응하기 어렵습니다.
한편, 마케팅 믹스 모델링의 경우 집계 데이터(aggregate data)만을 필요로 하기 때문에 유저 레벨 데이터가 필요 없습니다. 이러한 특징 때문에 MMP만 사용하던 기업들에게도 보완재로서 마케팅 믹스 모델링에 대한 필요성이 빠르게 증가하고 있습니다. 즉, 과거 마케팅 믹스 모델링의 높은 리소스를 감당할 수 없었던 그리고 감당할 만큼의 필요성을 느끼지 못했던 다양한 기업들이 마케팅 믹스 모델링을 찾게 되었습니다. 그리고 이러한 한계를 보완한 새로운 마케팅 믹스 모델링 방법에 대한 니즈로 이어질 수밖에 없었습니다.
### 마케팅 믹스 모델링의 새롭운 전성기를 끄는 SaaS
이와 같은 시장의 니즈를 충족시키기 위해 마케팅 믹스 모델링은 기존의 접근성을 낮추던 허들을 극복하는 방향으로 진화했습니다. 게다가 비용 및 성과 데이터의 실시간 자동 연동을 통해 데이터 수집이 용이해지고 클라우딩 컴퓨팅의 발달로 인해 광고주가 직접 서버를 구축하여 데이터를 처리할 필요가 없어졌다는 기술적인 진보가 기존 제약 조건을 극복 가능하도록 도왔습니다.
이러한 진화와 기술적 진보로 인해 탄생한 것이 SaaS(Software as a Service)형 마케팅 믹스 모델링입니다. 마케팅 믹스 모델링을 보다 널리 보급하고 새롭게 떠오르는 데에 기여하고 있습니다. SaaS형 마케팅 믹스 모델링은 과거의 마케팅 믹스 모델링 활용에 있어 가장 높은 진입 장벽이었던 리소스의 활용을 줄여주는 것이 가장 주요한 특징입니다. 데이터의 수집과 분석을 자동화해주고, 빠르고 쉽게 진행할 수 있으며, 합리적인 비용으로 사용할 수 있도록 제공합니다.
계속해서 빨라지는 시장의 변화 속도에 대응하기 위해서는 성과 분석과 결과 확인까지의 시간을 단축해야 합니다. 마케팅 성과 개선을 위한 플라이휠에 가속도를 높이기 위해서는 ‘분석 결과 확인 -> 마케팅 전략 수정 및 적용 -> 수정한 전략 기반의 결과 확인 -> 마케팅 전략의 재수정 및 재적용’의 과정을 빠르게 진행해야 하죠. 마케팅 성과 분석을 위한 준비와 진행 과정에 많은 리소스를 투입하기보다는, 진짜 중요한 의사 결정 과정에 집중하여 궁극적인 목표를 달성하는 데에 리소스를 사용해야 합니다.
이와 같은 마케팅 성과 분석을 위한 접근성과 효율성 개선에 SaaS형 마케팅 믹스 모델링이 기여하고 있습니다.
### 에어브릿지가 제공하는 SaaS형 마케팅 믹스 모델링
이러한 흐름에 앞장서서 [에어브릿지(Airbridge)](http://www.airbridge.io/ko)가 SaaS형 마케팅 믹스 모델링을 제공하고 있습니다.
에어브릿지는 마케팅 믹스 모델링을 빠르고 쉽게 그리고 합리적인 비용으로 사용할 수 있도록 제공하여 리소스의 절약을 도와줍니다. 긴 시간이 걸리고 많은 비용이 들어 접근하기 어려웠던 마케팅 믹스 모델링의 진입 장벽을 낮춰서 마케팅 믹스 모델링을 보다 널리 보급하는 데에 기여하고자 합니다.
에어브릿지는 MMP로서 이미 주요 매체들이 연동되어 있어서 광고 비용 및 성과 데이터를 수집하고 있다는 점에서 마케팅 믹스 모델링을 제공하기에 유리한 환경을 갖추고 있었습니다. 이를 통해 데이터를 자동으로 수집하고 고객사의 앱 안에서 발생하는 이벤트를 모두 추적하여 모델 고도화를 위한 데이터도 쉽게 확보할 수 있습니다. 게다가 수동 업로드 기능도 제공함과 동시에 업로드부터 모델링과 리포트 확인까지의 과정이 자동화되어 있어서 기존의 마케팅 믹스 모델링 방법처럼 수동으로 데이터를 업로드하더라도 보다 빠르게 분석 결과를 확인할 수 있습니다.
그리고 에어브릿지는 자체 개발한 머신러닝 모델을 기반으로 마케팅 모델링 솔루션을 제공합니다. 자동 및 수동으로 업로드한 광고 비용 및 성과, 오프라인 매체의 비용 및 성과, 외부 요인 등 고객사별 핵심 데이터를 기반으로 서비스/프로덕트 맞춤 모델 기반의 고도화된 마케팅 믹스 모델링까지도 제공 가능합니다. 또한, 다양한 설정값 및 학습 현황에 대한 정보를 투명하게 언제든 확인할 수 있도록 제공하여 모델에 대한 접근성도 개선하였습니다.
유저 친화적인 인터페이스를 갖춘 대시보드를 통해 분석 결과에 대한 접근성을 개선하기도 했습니다. 기존의 보고서 형식의 결과가 아닌, 매일 및 매주 업데이트되는 대시보드의 리포트에서 결과를 언제든지 쉽게 확인할 수 있습니다. 게다가 다양한 마케팅 시나리오를 자유롭게 테스트하여 성과의 효율을 개선할 수 있도록 최적의 예산 분배 안을 확인할 수 있는 리포트도 제공하고 있습니다. 단순한 분석을 넘어 즉시 실행 가능한 액션 아이템을 제안하며 마케팅 믹스 모델링의 성과를 매일의 의사 결정에 실용적으로 사용할 수 있도록 제안합니다.
> 에어브릿지 마케팅 믹스 모델링에 대한 더 자세한 내용이 궁금하시다면 아래 글을 확인해보세요.
- [개인정보보호 시대 제대로 마케팅 성과를 측정하는 방법, 에어브릿지 마케팅 믹스 모델링을 확인해보세요.
](https://www.airbridge.io/ko/blog/airbridge-marketing-mix-modeling)- [에어브릿지는 LTA + MTA + MMM 통합 측정 스택 제공과 함께 Marketer-centric으로 나아갑니다.](https://www.airbridge.io/ko/blog/airbridge-unified-measurement-stack)
## SaaS형 마케팅 믹스 모델링은 다양한 기업들의 보편적인 활용에 기여합니다.
결론적으로, SaaS형 마케팅 믹스 모델링은 기존 마케팅 믹스 모델링의 제약 조건을 완화시켜서 접근성을 높여주었다는 것이 가장 주요한 차별점입니다.
서드파티 쿠키 및 광고 ID 수집 제한 등 개인정보보호 강화라는 거스를 수 없는 트렌드에서도 마케팅 성과를 확인하고 개선해야 하는 것은 모든 기업들의 필수적인 과제입니다. 이러한 흐름에 가장 효과적으로 대응할 수 있는 방법은 유저 레벨 데이터 없이도 성과 측정을 할 수 있는 마케팅 믹스 모델링입니다.
SaaS형 마케팅 믹스 모델링은 특히 개인정보보호 강화 환경에서 벗어나지 않으면서도 광고 성과의 가시성 확보 및 최적화에 도움이 되고자 합니다. 현재 가지고 있는 데이터만으로도 최대의 효과를 낼 수 있는 기술을 제공하여 보다 보편적인 활용에 기여합니다. 그동안 접근하기 어려웠던 마케팅 믹스 모델링을 SaaS 솔루션으로 활용하여 효율적으로 데이터를 확인하고 진짜 중요한 의사결정에 집중해 보세요.
---
## 에어브릿지가 MMP 이관을 지원하는 방법
URL: https://www.airbridge.io/ko/blog/how-you-can-migrate-your-mmp-to-airbridge
Language: Korean
Published: 2023-01-12
최근 에어브릿지(Airbridge)를 도입하는 고객 중 타 어트리뷰션 툴 / MMP에서 이관하는 사례가 증가하고 있습니다. 앱 서비스가 성장하고 마케팅을 적극적으로 집행하다보면 사용하고 있는 마케팅 솔루션 비용도 함께 증가합니다. 솔루션 이용자의 입장에서 같은 기능과 서비스를 보다 효과적이고 잘 지원하면서도 더 낮은 비용으로 공급하는 솔루션이 있다면 이관을 고려하게 될 것은 당연합니다. 게다가 프라이버시 보호 강화에 따른 마케팅 성과 측정의 불확실성 증가에 대한 방향성과 해결책을 빠르게 제시해줄 수 있는 MMP 니즈도 증가하고 있습니다.
이러한 마케터들의 고민과 니즈에 대해 에어브릿지는 모든 기능을 올인원으로 합리적인 과금 체계로 공급하며 트렌드 변화에 빠르고 적극적인 대응책을 제공하고 있습니다. 또한, MMP 이관으로 발생할 수 있는 불편과 데이터 단절을 최소화하기 위해 유저 친화적인 도입 과정을 제공합니다. 이는 즉, 에어브릿지로의 이관을 검토하고 실행하는 앱 서비스가 증가하는 이유입니다.
그렇다면 실질적으로 에어브릿지는 어떤 절차로 이관을 지원하고 있는지 알아보겠습니다. 에어브릿지는 기존 MMP에서 겪던 비용과 트렌드 대응에 대한 어려움을 모두 해결해드리고 있습니다. [지금 바로 에어브릿지의 모바일 마케팅 전문가와 이야기 나눠 보세요.](https://www.airbridge.io/ko/book-a-demo)
## 에어브릿지 도입 과정은 전문적이고 체계적인 고객 지원과 함께합니다.
에어브릿지를 선택한 많은 기업들이 공통적으로 언급한 도입 이유 중 하나는 전문적인 고객 성공 매니저의 체계적인 지원입니다. 에어브릿지는 도입 과정에서 전문적이고 체계적인 고객 지원을 제공하여 프로덕트 및 서비스의 성장을 시작부터 완벽하게 지원합니다. 우선 에어브릿지 도입이 결정되면 풍부한 마케팅 전략 설계 및 지원 경험이 있는 담당 고객 성공 매니저가 지정되어 온보딩 과정을 시작합니다. 이 과정은 체계적으로 설계되어 있으며, 각 비즈니스의 목표 및 방향성에 맞춰서 진행됩니다. 온보딩을 완료한 후에도 질의응답은 물론 워크샵과 컨퍼런스 등을 통해 비즈니스의 성장으로 이어지는 지원과 교육을 계속해서 제공합니다.
> “(에어브릿지 고객 성공 지원 서비스만의 강점은) 온보딩 서비스를 상세하게 제공한다는 점을 꼽을 수 있습니다. 셋팅을 위해 공식 개발 가이드 문서 하나만 제공하는 솔루션사들이 많다고 들었어요. 하지만 저희는 고객사 케이스에 맞게끔 다양한 방식으로 상세한 온보딩 지원을 드리는 것이 기본적인 업무 프로세스에 포함되어 있습니다.”
“좋은 솔루션이어도 적절한 셋팅이 이루어지지 않으면 제대로 활용하기 어렵기 때문에, 온보딩에 가장 많은 리소스를 투입합니다. 첫 단추를 잘 끼워서, 최대한 시행착오를 줄일 수 있게끔 가이드해 드리는 역할을 하고 있습니다.”
- [에이비일팔공 CSM은 어떻게 고객 성공을 지원하나요?](https://blog.ab180.co/posts/csm-interview) 포스트에서 발췌
## 에어브릿지가 이관을 위해 지원하는 4가지 절차
에어브릿지는 도입 후 빠르게 온보딩을 진행하여 원하는 시기에 실제 활용이 가능하도록 지원합니다. 최고의 고객 성공 팀과 프로덕트 팀이 모든 과정에 함께하며 어떤 어려움이라도 해결될 수 있도록 적극적으로 대응하고 있습니다. 구체적인 절차를 살펴보자면 다음과 같습니다.
#### 1. MMP 교체를 통한 기대 효과 확인
MMP를 교체하는 것이 쉬운 결정이 아니라는 점을 알고 있습니다. 그만큼 MMP 이관을 통해 얻고자 하는 효과가 명확할 것입니다. 이에, 기존 MMP에서 사용하고 있는 주요 기능과 목적부터 명확하게 파악하고 새로운 MMP에 기대하는 기능과 달성하고 싶은 지표를 확인하여 기대 효과 달성을 지원합니다.
또한 보다 구체적으로는 마케팅 계획 상 MMP 교체가 완료되어야 하는 타임라인, 페이드 마케팅을 집행하는 주요 매체 리스트, 마케팅 KPI 지표, 로우 데이터(Raw data) 연동으로 활용하는 대시보드의 메트릭 등을 확인하여 각 고객에 대한 이해도를 높이기 위해 노력합니다.
위 사항은 MMP 이관을 고려하는 시기에 세일즈 담당자와의 컨설팅을 통해 구체적으로 함께 의논할 수 있으며, 도입이 확정된 후에는 고객 성공 팀에서 그대로 맥락을 이어 받아 MMP 변경 후 마케팅 성과의 단절 없이 실무에서 곧바로 사용할 수 있도록 지원합니다.
#### 2. 계정 생성 및 에어브릿지 사용을 위한 기본 준비
담당 고객 성공 매니저와 함께 비즈니스 맞춤 온보딩을 본격적으로 시작하게 되면 가장 먼저 에어브릿지 대시보드 회원가입 후 에어브릿지를 적용할 앱 서비스를 등록합니다. 그리고 이벤트 텍소노미를 통해 마케팅 성과로 트래킹하고자 하는 이벤트 및 이벤트 프로퍼티를 정의합니다. 또한, 광고를 집행할 매체를 확인하고 연동 가능한 혹은 추가 연동을 원하는 매체를 확인합니다. 이 과정에서 에어브릿지 대시보드의 메뉴별 기능과 사용 방법 교육이 진행되며, [AB180 Academy](https://academy.ab180.co/)에서도 에어브릿지 관련 강좌를 확인할 수 있습니다. 이 모든 과정을 담당 고객 성공 매니저와 함께 진행하기 때문에 문제 없이 진행할 수 있습니다.
#### 3. 과거 데이터 이관으로 데이터 유실 방지
MMP를 교체할 때 가장 우려되는 부분은 기존의 데이터가 유실되어 기존 캠페인을 유지하거나 과거 데이터를 확인할 수 없다는 점입니다. 이에 대해 에어브릿지는 다음의 두 가지 해결 방법을 제공하여 이와 같은 우려를 덜어드리고 있습니다.
1) ADID 혹은 User ID 리스트를 CSV로 업로드
MMP는 SDK 설치 이후 데이터를 수집하며, 이벤트를 수집하는 시점으로부터 신규 이벤트 및 재 이벤트를 구분합니다. 그렇기 때문에 기존 유저가 새로운 SDK 설치 후 이벤트를 수행하게 되면 신규 이벤트로 식별하게 됩니다. 이러한 문제를 방지하고자 기존 유저의 ADID 혹은 User ID 리스트를 CSV로 전달 받아 업로드하여 기존 유저의 이벤트를 재 이벤트로 처리하는 과정을 지원합니다.
2) 집계된 데이터(Aggregated data) 수준의 Historical Data를 업로드
에어브릿지는 기본으로 제공하지 않는 지표라도 사용자가 대시보드에 추가해서 보고 싶다면 직접 업로드할 수 있는 기능을 제공합니다. API를 통해 신규 메트릭 데이터를 직접 업로드하고 업로드 상태를 조회할 수 있습니다. 그리고 업로드한 데이터를 에어브릿지가 측정한 전환 데이터 및 인앱 이벤트 데이터와 함께 조회하여 마케팅 성과를 한 번에 확인할 수 있습니다.
👉 자세한 설정은 개발자 가이드의[ Self-serve 데이터 업로드 요청 및 상태 조회하기](https://developers.airbridge.io/reference/self-serve-%EB%8D%B0%EC%9D%B4%ED%84%B0-%EC%97%85%EB%A1%9C%EB%93%9C-%EC%9A%94%EC%B2%AD-%EB%B0%8F-%EC%83%81%ED%83%9C-%EC%A1%B0%ED%9A%8C%ED%95%98%EA%B8%B0)에서 확인해 보세요.
#### 4. 캠페인 및 트래킹 링크의 설정과 매체 연동
현재 진행중인 캠페인의 지속적인 집행과 매체 연동 설정 또한 어렵지 않으며 에어브릿지 팀이 함께 지원합니다.
우선 기존에 활용 중인 트래킹 링크의 활용 케이스를 취합하여 에어브릿지 트래킹 링크 구조와 매핑 후 트래킹 링크 제너레이터를 통해 대량의 트래킹 링크를 생성하여 한 번에 교체합니다.
그리고 운영 중인 매체 리스트에 대해 연동 정보 등록 및 포스트백 전송을 설정합니다. 이외에도 데이터를 외부 클라우드 스토리지에 저장하고 싶다면 자동으로 데이터를 적재할 수 있도록 지원합니다. 그리고 AB180에서 함께 지원하고 있는 앰플리튜드(Amplitude)와 브레이즈(Braze)와의 연동 및 활용도 같은 고객 성공 매니저를 통해 한 번에 문의 및 지원이 가능합니다.
## 에어브릿지의 모바일 마케팅 전문가와 이야기 나누고 직접 경험해보세요.
위 내용을 바탕으로 MMP 이관에 대한 확신이 생겼다면, [데모를 신청하고](https://www.airbridge.io/ko/book-a-demo) 에어브릿지의 모바일 마케팅 전문가와 이야기 나눠 보세요.
에어브릿지의 유저 친화적인 대시보드 사용성과 더불어 전문적인 고객 성공 팀과 프로덕트 팀의 빠르고 정확한 지원을 직접 경험해보세요. 에어브릿지는 합리적인 비용의 올인원 모바일 마케팅 패키지로 앱 서비스 성장에 기여합니다.
👉 에어브릿지를 선택한 고객들의 이야기를 확인해보세요.
---
## 릴리즈 노트: 2022년 11~12월
URL: https://www.airbridge.io/ko/blog/release-notes-nov-dec-2022
Language: Korean
Published: 2023-01-11
2022년 마지막 릴리즈 노트에서 11월과 12월의 에어브릿지 신규 업데이트 소식을 한 번에 확인해 보세요. 에어브릿지는 Calculated Metric을 추가하고 레비뉴 리포트(Revenue report)에 차트를 추가하여 시각화를 강화하는 등 대시보드를 지속적으로 개선해가고 있으며 신규 매체 연동 등을 통해 보다 사용성을 높여가고 있습니다.
> **[주요 업데이트 미리 보기]
**- Calculated Metric 추가
- 레비뉴 리포트에 차트 추가 및 조회 가능한 데이터 범위 확대
- 리텐션 리포트의 조회 가능한 데이터 범위 변경 및 차트 디자인 개선
- 리포트의 필터 사용성 개선
- 추가 매체 연동: RevenueCat 등
- SDK 릴리즈
## Calculated Metric 추가
에어브릿지의 액츄얼스 리포트(Actuals report) 및 트렌드 리포트(Trend report)에서 메트릭, 사칙 연산, 실수를 조합한 새로운 메트릭 Calculated Metric을 확인할 수 있습니다.
- Cost Per Install - CPI (APP) : Cost / Install 으로 계산
- Cost Per Purchase - CPP (App) : Cost / Complete order (App) 으로 계산
- Cost Per Purchase - CPP (Web) : Cost / Complete order (Web) 으로 계산
- Cost Per Purchase - CPP (App+Web) : Cost / Complete order (App+Web) 으로 계산
- Click-through Rate - CTR : Click / Impressions * 100 으로 계산
위의 메트릭들을 활용하여 액츄얼스 리포트에서 해당 메트릭을 포함한 원하는 형태의 테이블을 만들어 확인할 수 있습니다. 또한, 트렌드 리포트에서는 데이터의 변화 추이를 그래프로 한눈에 확인할 수 있습니다.
이외에도 다양한 커스텀 Calculated Metric을 활용하고 싶은 경우 담당 CSM에 문의해주시기 바랍니다.
[]
## 레비뉴 리포트(Revenue report) 업데이트: 시계열 차트 추가 및 조회 가능한 데이터
[]
매출에 관한 다양한 지표를 확인할 수 있는 통계 리포트인 레비뉴 리포트(Revenue report)에 차트가 추가되고 조회 가능한 데이터 범위가 변경되며 한층 더 업그레이드되었습니다.
### 시계열 차트 추가 및 이중축 기능 도입
레비뉴 리포트의 Config Box(설정)와 테이블 사이에 시계열 차트가 추가되었습니다. 이제 레비뉴 리포트의 메트릭인 매출액, 유저 수, 이벤트 수, ROAS, ARPU, ARPPU에 대해서 Day N 별 추이를 차트로 쉽게 확인할 수 있습니다. 테이블 좌측의 체크박스를 통해 8개의 로우를 선택하여 원하는 데이터를 차트로 분석할 수 있습니다.
[]
또한, 특정 항목의 값이 많은 비중을 차지하는 경우를 대비한 ‘이중축’ 기능이 추가되었습니다. 특정 항목 값이 매우 큰 경우 같은 축을 기준으로 설정하게 되면 나머지 값들이 상대적으로 작아 겹쳐져 표현될 수 있습니다. 이럴 경우 상단 이미지의 예시와 같이 큰 값을 가진 항목(unattributed)은 오른쪽 축을 기준으로 설정하고, 나머지 채널(Channel A~G)은 왼쪽 축을 기준으로 설정하면 각각의 스케일에 따른 추이를 한 눈에 확인할 수 있습니다.
### 조회 가능한 Day N 범위 변경
또한, 레비뉴 리포트에서 조회 가능한 Day N 범위가 변경되었습니다. 기존에는 캘린더에서 설정한 기간 내의 Day N 값만 조회할 수 있었지만, 이번 업데이트를 통해 리포트 조회 시점을 기준으로 과거 일자에 대해서는 설정 기간을 초과한 일자에 대해서도 Day N 수치를 확인할 수 있습니다.
예) **2022.12.30에** *2022.12.19~2022.12.21로* 캘린더 기간을 설정한 경우
[]
## 리텐션 리포트(Retention Report) 업데이트: 조회 가능한 데이터 범위 변경, 차트 디자인 개선
앱 내에서 특정 이벤트를 수행한 유저들 중 얼마나 많은 유저들이 다시 앱으로 돌아와 특정 이벤트를 수행했는지를 N Day 리텐션 방식으로 제공하는 리텐션 리포트(Retention report)도 조회 가능한 데이터 범위가 변경되고 차트 디자인이 개선되며 보다 유저 친화적으로 거듭났습니다.
### 조회 가능한 Day N 범위 변경
앞서 소개한 레비뉴 리포트의 조회 가능한 Day N 범위가 변경된 업데이트는, 리텐션 리포트에도 적용되었습니다. 기존에는 캘린더에서 설정한 기간 안에서만 Day N 값을 조회할 수 있었지만, 이번 업데이트를 통해 리포트 조회 시점을 기준으로 ‘과거 일자’에 대해서는 설정 기간을 초과한 일자에 대해서도 Day N 수치를 확인할 수 있습니다.
예) **2022.11.21에** *2022.11.06~2022.11.08*로 캘린더 기간을 설정한 경우
[]
Last 7 days와 같이 아직 Day N들의 모든 셀들이 업데이트 되지 않은 기간으로 캘린더를 설정한 경우에는 기존과 같이 빈 셀이 노출됩니다.
예) **2022.11.21에** *2022.11.19~2022.11.21*로 캘린더 기간을 설정한 경우
[]
### 차트 디자인 변경
[]
뿐만 아니라, 리텐션 리포트의 색상값이 계산된 잔존율을 기준으로 상대적인 색상값으로 선정되는 방식으로 변경되었습니다. 그리고 각 셀 간 구분선이 추가되어 수치를 좀 더 명확하게 확인할 수 있게 되었습니다. 변경된 차트 디자인을 통해 데이터의 변화를 놓치지 않고 보다 빠르게 확인해 보세요.
## 리포트의 사용성 개선: 필터에 콤마를 구분자로 하여 여러 개의 값을 한번에 선택
[]
에어브릿지 대시보드에서 필터 값을 선택할 때 콤마(,)를 구분자로 하여 선택값을 한번에 선택할 수 있는 옵션을 추가로 제공합니다.
이전에는 필터 값을 선택하기 위해서는 하나하나 선택해야 했지만 이제 콤마로 구분된 값을 붙여넣기하거나 키보드로 입력하면 한번에 선택할 수 있습니다. 그리고 콤마가 포함된 값을 만들고 싶을 때는 큰 따옴표("")로 감싸주면 됩니다.
## 추가 매체 연동
### 글로벌 앱 구독 솔루션 레비뉴캣(RevenueCat)과 연동
에어브릿지와 글로벌 앱 구독 솔루션 [레비뉴캣이 연동](https://www.airbridge.io/ko/blog/airbridge-revenue-cat-partner)되었습니다. 이제 레비뉴캣의 웹훅 기능을 통해 구독 관련 이벤트들을 에어브릿지에도 기록할 수 있게 되었습니다. 웹훅 연동을 하게 되면, RevenueCat에서 일어나는 구독 관련 이벤트가 각 이벤트에 따라 에어브릿지의 스탠다드 이벤트(구독, 구독 취소, 체험판 시작 등)와 커스텀 이벤트로 기록됩니다.
👉 자세한 연동 가이드는 [여기](https://help.airbridge.io/hc/ko/articles/12964722155929-Revenue-Cat)에서 확인 가능합니다.
이외에도 Appa Worldwide, 레브엑스(Rev X), 쇼핑머니백(smoneyback), 씨드앱(SeedApp)과 연동이 완료되었습니다. 앞으로도 에어브릿지는 지속적으로 파트너사를 확대하며 추가적인 매체 연동을 진행할 예정입니다.
## SDK 릴리즈
에어브릿지는 앱(Android & iOS) 및 웹 SDK를 기반으로 웹과 앱 데이터를 통합적으로 분석합니다. 또한 지속적인 업데이트를 통해 사용성을 개선하고 있습니다.
11월과 12월에는 아래의 SDK가 릴리즈되었습니다. 연결된 링크에서 자세한 Release Note를 확인할 수 있습니다.
- [Airbridge Android SDK](https://developers.airbridge.io/docs/android-sdk-release-note) - Airbridge Android SDK 2.19.3, Airbridge Android SDK 2.19.4, Airbridge Android SDK 2.19.5
- [Airbridge iOS SDK](https://developers.airbridge.io/docs/ios-sdk-release-note) - Airbridge iOS SDK 1.27.0, Airbridge iOS SDK 1.28.0
- [Airbridge Unity SDK](https://developers.airbridge.io/docs/unity-sdk-release-note) - Airbridge Unity SDK Release 1.8.0, Airbridge Unity SDK Release 1.8.1
- [React Native SDK](https://developers.airbridge.io/docs/react-native-sdk-release-note) - React Native SDK Release 1.5.0
---
## 앱 마케터가 SKAdNetwork 4.0(SKAN 4.0)에 주목해야 하는 이유
URL: https://www.airbridge.io/ko/blog/exploring-the-marketing-implications-of-skadnetwork-4
Language: Korean
Published: 2023-01-05
이제 모바일 앱 마케팅에서 유저의 개인정보를 보호하는 것은 중요하고도 당연한 트렌드로 자리잡았습니다. 이 트렌드로의 변화를 이끌고 있는 것 중 하나가 바로 iOS 14를 위한 Apple의 어트리뷰션 프레임워크 SKAdNetwork(SKAN)입니다. [지난 포스트](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)에서도 소개했던 것처럼, SKAN은 iOS 환경에서 개인정보를 보호하며 유저에 대한 정보 없이도 광고 캠페인의 성과를 측정할 수 있는 방법을 제공하고 있습니다. 그리고 지난 2022년 10월 24일, 최신 버전의 프레임워크인 SKAN 4.0이 릴리즈되었습니다. iOS 16.1 및 iPadOS 16.1이상의 버전을 위한 프레임워크로 여러 새로운 기능과 개선사항을 도입하였습니다.
이번 포스트는 SKAN 4.0의 상세한 기술적인 내용을 다루기보다는, 비즈니스 및 마케팅 측면에서 이 새로운 기술이 미칠 영향력에 대해 보다 집중적으로 알아보려 합니다. SKAN 4.0 업데이트가 마케팅에 미치는 영향력에 대해 이해하고 마주하는 변화에 보다 앞서서 대비해보세요.
> <주요 내용 미리보기>
>
> 1\. 작은 규모의 테스트와 운영이 더 쉬워졌습니다.
> 2\. 크리에이티브 레벨 혹은 국가 레벨의 어트리뷰션이 일부 돌아왔습니다.
> 3\. ARPU 및 ROAS를 위한 코호트 분석이 일부 가능해졌습니다.
> 4\. 이제 웹에서도 SKAN을 트래킹할 수 있습니다.
## 작은 규모의 테스트와 운영이 더 쉬워졌습니다.
SKAN 3.0 이하에서는 임계값(Privacy threshold)이 있어서 작은 규모의 광고 캠페인을 운영하기 어려웠습니다. 왜냐하면 이 프레임워크에서는 인앱 이벤트에 대한 SKAN 포스트백을 발송하기 전에 캠페인 당 발생해야하는 최소한의 설치 수를 넘어야 했기 때문입니다. 그리고 이로 인해 최소한의 설치를 발생시키기 어려운 작은 규모의 예산만으로는 광고 캠페인을 테스트하고 최적화하기 어려웠습니다.
임계값(Privacy threshold)은 과거 24시간 내에 발생한 캠페인 ID 당 설치 수에 기반하여 측정합니다. 이 수치는 일반적으로 10개에서 20개 사이의 설치로 알려져 있습니다. 만약 캠페인이 이 임계값 이하로 떨어진다면, 캠페인과 연관된 포스트백의 전환 값(Conversion value)이 null 값으로 부여되어 값을 확인할 수 없습니다.
하지만 이번 SKAN 4.0 업데이트로 인해 작은 규모의 광고 캠페인 테스트 및 운영이 더 쉬워졌습니다. 가장 중요한 변화 중 하나로 SKAN 4.0은 ‘군중의 익명성(Crowd anonymity)’ 개념을 도입했습니다. 이는 4개의 티어 시스템으로, 티어 0부터 3까지의 범위에 따라 전환 값(Conversion value)의 표기 방법이 달라집니다. 지정된 티어에 따라 광고주들은 보다 상세 전환 값(기존의 64개 전환 값)을 받거나 단순 전환 값(Coarse value)을 받을 수 있습니다. 이 단순 전환 값은 인앱 이벤트를 상(High), 중(Medium), 하(Low)로 표기하며, 유저의 설치 후 행동에 대한 대략적인 정보를 모아서 보여줍니다. SKAN 3.0과 4.0의 인앱 이벤트의 전환 값 표기 방법을 비교하여 정리하자면 아래의 표와 같습니다.
**SKAN 3.0SKAN 4.0**임계값(Privacy threshold) 초과인앱 이벤트가 0~63의
전환 값(Conversion value)으로 표기 됨티어3(Tier 3)인앱 이벤트가 0~63의
전환 값으로 표기 됨
(상세 전환 값, fine-grained value)티어(Tier 2)인앱 이벤트가 0~63의
전환 값으로 표기 됨
(상세 전환 값, fine-grained value)임계값 미만인앱 이벤트에 대한
어떠한 정보도 받을 수 없음티어(Tier 1)인앱 이벤트가 상, 중 하로 표기됨 (단순 전환 값, coarse value)티어(Tier 0)인앱 이벤트에 대한
어떠한 정보도 받을 수 없음
만약 어떤 캠페인/소스가 ‘티어 0’으로 지정됐다면, SKAN 3.0에서 임계값을 넘지 못한 상황과 비슷하게 어떠한 정보도 제공되지 않습니다. 하지만, 만약 어떤 캠페인/소스가 ‘티어 1’로 지정됐다면, 이전과 같이 자세하게 나눠진 값은 아니지만 그래도 유저 행동에 대한 어느 정도의 인사이트를 얻을 수 있는 상, 중, 하로 표기된 단순 전환 값을 받을 수 있습니다. Apple은 광고주들이 “전환 값이 임계값을 넘지 못한 경우라고 하더라도 대략적인 어트리뷰션 정보를 받을 수 있다”고 언급하며, SKAN 4.0에서 ‘티어1’의 기준을 맞추는 것이 SKAN 3.0에서 임계값의 기준을 맞추는 것보다 덜 어려울 것이라고 말합니다.
## 크리에이티브 레벨 혹은 국가 레벨의 어트리뷰션이 일부 돌아왔습니다.
[__wf_reserved_decorative]
이전 SKAN 버전에서는 크리에이티브 레벨의 어트리뷰션이 불가능했으며 캠페인 레벨에서 어트리뷰션한 결과만을 확인할 수 있었습니다. SKAN 4.0에서는 기존의 ‘campaign-id’ 파라미터를 대체하는 ‘source-identifier’라는 새로운 파라미터를 도입함으로써 이 크리에이티브 레벨의 어트리뷰션 기능이 일부 가능해졌습니다. ‘source-identifier’는 0부터 9,999까지의 4자리 숫자로 이루어질 수 있으며, 이는 이전의 ‘campaign-id’가 0부터 99까지의 2자리 숫자로 이루어졌던 것에 비해 늘어났습니다. 그리고 이 숫자들은 크리에이티브, 국가, 형식과 같은 추가적인 형태를 나타낼 수 있습니다.
## ARPU 및 ROAS를 위한 코호트 분석이 일부 가능해졌습니다.
[__wf_reserved_decorative]
이전에는 단 1개의 포스트백만을 전송했지만 SKAN 4.0은 3개의 포스트백을 받을 수 있습니다. 이로 인해 특정 일자에 대한 ARPU(Average Revenue per User, 유저 당 평균 수익)나 ROAS(Return on Ad Spend, 광고비 대비 매출액)와 같은 코호트 분석이 일부 가능해졌습니다. 전환 기간(conversion window)도 최대 35일 + 6일(24~144 시간의 랜덤 지연)까지로 연장되어 설치 이후 유저의 행동에 대한 보다 깊은 인사이트를 제공해줍니다. 이러한 포스트백들은 앱 설치 일로부터 지정된 기간 내에 연속적으로 보내지며 설치 이후의 유저 여정을 재구성하는 데에 사용할 수 있습니다.
앱 어트리뷰션 툴 / MMP는 이 포스트백들을 활용하여 특정 설치일로 어트리뷰션할 수 있으며 모든 유저 여정을 간추려서 코호트 분석 형태로 만들 수 있습니다. 하지만 SKAN 4.0은 각 포스트백에 랜덤한 지연을 적용하기 때문에 정확한 특정 설치일로의 어트리뷰션은 어렵다는 점을 염두해야 합니다.
## 이제 웹에서도 SKAN을 트래킹할 수 있습니다.
SKAN 4.0의 새로운 기능 중 하나는 사파리(Safari) 브라우저에서 SKAN으로 웹 광고를 트래킹할 수 있게 되었다는 점입니다. 이전에는 모바일 지면에서만 트래킹이 가능했었지만, 이번 기능을 통해 애드 네크워크, MMP, 광고주들이 웹 지면의 광고로부터 발생한 앱 설치와 인앱 이벤트를 어트리뷰션할 수 있게 되었습니다. 이는 크로스 채널 광고 캠페인의 효율성에 대한 가치 있는 인사이트를 제공해줄 수 있습니다. 다만 이 기능은 사파리 브라우저에서만 가능하지만, 모바일 iOS 기기 유저 중 사파리 활용 비율이 높은 만큼 영향력이 있을 것으로 기대가 됩니다.
## 개인정보를 보호하며 광고 캠페인 성과 측정하기, SKAN 4.0을 에어브릿지와 함께 활용해보세요.
결론적으로, SKAN 4.0은 새로운 기능과 개선 사항을 도입하여 모바일 앱 마케팅 시 개인정보를 보호하면서도 쉽게 광고 캠페인의 성과를 측정하도록 만들어주었습니다.
에어브릿지는 현재 SKAN 4.0을 지원하기 위해 애드 네트워크들과 긴밀하게 협업하고 있으며, 기존의 전환 값 설정 페이지도 SKAN 4.0의 사양을 반영하여 업데이트할 예정입니다. 이를 통해 에어브릿지를 사용하는 데에 있어 SKAN 4.0이 제공하는 모든 장점들을 활용하실 수 있도록 지원하고자 합니다. 곧 다가오는 업데이트를 기대해주세요!
---
## 고객 생애 가치(LTV)의 의미와 계산 방법
URL: https://www.airbridge.io/ko/blog/what-is-ltv
Language: Korean
Published: 2023-01-02
우리 비즈니스가 얼마나 지속 가능하게 성장하고 있는지 확인하고 싶으신가요? 그렇다면 획득한 유저가 우리 비즈니스에 얼마만큼의 수익을 가져다주는지를 보여주는 **고객 생애 가치(LTV, Lifetime Value)**를 확인해보세요.
LTV는 리텐션(retention)을 돈으로 환산한 지표와도 같습니다. 마케팅 의사 결정에 기반이 되며 비즈니스의 지속가능성을 보여줍니다. 다만 이전에는 계산이 까다로우며 비교적 추상적이고 두루뭉술한 개념으로 여겨지기도 했습니다. 아직 LTV가 낯설게 여겨지는 분들을 위해 이번 포스트를 통해 LTV가 무엇인지 그리고 다양한 계산법을 소개합니다. LTV를 앱 마케팅에 적용하여 지속 가능한 비즈니스의 성장을 이루어보세요.
## 고객 생애 가치(Lifetime Value, LTV)란?
소비자 생애 가치(Customer Lifetime Value / CLV 혹은 CLTV)로 불리기도 하는 지표로, 한 명의 고객이 우리 비즈니스에서 발생시킬 것으로 예측할 수 있는 전체 수익을 의미합니다. 모바일 앱 마케팅의 발전과 함께 수집하고 확인할 수 있는 데이터가 늘어나면서 보다 널리 활용될 수 있는 환경이 만들어졌습니다.
LTV는 간단하게 말하자면 고객 한 명이 비즈니스에서 평생 발생시킬 매출이자, 고객 한 명의 누적 결제 금액을 의미합니다. 그래서 단순하게는 유저가 평생 발생시킬 수익을 전부 더한 값에서 유저를 획득하기 위해 투자한 비용(CAC, Customer Acquisition Cost)을 빼서 값을 얻을 수 있습니다. 물론 LTV를 계산하는 것이 이렇게 단순하지만은 않습니다. 다양한 LTV 계산 방법은 아래에서 보다 자세히 다뤄보겠습니다.
## LTV를 계산하고 확인하는 것이 왜 중요한가?
LTV를 계산하면 마케팅 결과로 획득한 고객이 비즈니스에 기여하는 가치를 수치화하여 고객과 우리 비즈니스 간에 얼마나 관계가 지속되고 있는지 파악할 수 있습니다. 단지 유저 획득으로만 마케팅 성과를 판단하는 것이 아닌, 유저가 얼마나 지속적으로 우리 비즈니스를 찾는지를 통해 비즈니스 사이클 전반에 걸친 마케팅 성과를 판단할 수 있는 효과적인 지표입니다. 이와 더불어 LTV의 중요성에 대해 보다 구체적으로 살펴보자면 다음의 세 가지 이유와 같습니다.
**1. 비즈니스의 수익성과 지속가능성 파악**
LTV를 계산하면 유저가 앱 내에서 발생시킨 매출에 대한 추정치를 확인할 수 있습니다. 이를 기반으로 우리 비즈니스가 얼마나 수익성이 있고 지속 가능한지 살펴볼 수 있습니다. 앞서 언급한 것처럼 LTV는 비즈니스의 성장을 위한 또 다른 중요한 지표인 리텐션을 돈으로 환산한 지표와도 같다고 여겨지기도 합니다. LTV와 리텐션은 고객과 우리 비즈니스의 상호작용과 성장 정도를 파악하는데 빼놓을 수 없는 중요한 두 지표입니다. 고객의 리텐션 비율이 높다고 하더라도 고객의 결제 금액이 적어서는 성장에 한계가 있으며, 높은 지출을 보이는 유저가 있다고 하더라도 고객이 다시 우리 비즈니스를 방문해 주지 않는다면 그 비즈니스는 지속 가능하지 않습니다. 리텐션과 LTV를 함께 확인하고 이를 개선하려는 노력을 통해 지속 가능한 비즈니스로 성장해갈 수 있습니다.
**2. 신규 고객 획득 비용의 최대치 계산 및 효율 개선**
LTV는 신규 고객을 확보하기 위해 얼마만큼 까지의 비용을 지출할 수 있는지에 대한 기준을 정하는 데에도 좋은 지표입니다. LTV를 계산하여 고객들이 우리 비즈니스에서 어느 정도의 매출을 발생시키는지 파악했다면, 신규 유저 획득 비용은 그보다 낮아야 합니다.
게다가 신규 고객을 획득하여 구매를 촉진하는 것보다 기존 고객을 유지시키고 재구매를 유도하는 것이 훨씬 효율적입니다. 일반적으로 신규 고객 획득 비용이 기존 고객 유지 비용에 비해 5배 더 높다고 합니다. 이는 즉 LTV가 높은 고객군을 파악하고 관리하는 것의 중요성과도 이어지게 됩니다. 가치 있는 고객에게 보다 좋은 혜택을 제공하여 상호작용을 증진하게 되면 수익의 개선과 LTV의 증가 그리고 신규 고객 획득 비용의 절감으로 이어질 수 있습니다.
**3. 효율적인 마케팅 채널 파악 및 예산 분배의 효율성 증진**
유입 채널별 유저들을 코호트로 나누어서 LTV를 계산해 보면, 가장 높은 가치를 보이는 유저군을 획득한 채널을 찾을 수 있습니다. 이를 통해 가장 효율적인 마케팅 채널을 파악하여 예산 분배를 조절하면 고가치 유저군을 획득할 수 있는 가능성을 높일 수 있습니다. 즉, 우리 비즈니스에서 가장 많은 수익을 발생시키는 고객군 획득하는 방향으로 UA 전략을 발전시켜서 마케팅 예산의 ROI를 개선해갈 수 있습니다.
위의 세 이유를 종합하여 살펴봤을 때, LTV를 제대로 계산하면 의사 결정을 내릴 수 있는 효과적인 근거를 획득할 수 있으며 비용 효율을 최대화할 수 있기 때문에 중요하다고 할 수 있습니다.
다만, 이때 주의해야 할 점은 모든 업계에서 LTV가 중요한 지표는 아니라는 점입니다. 부동산, 자동차 등 자주 구매할 수 없는 고관여 상품 혹은 서비스를 판매하는 경우 LTV를 계산하거나 유의미하게 활용하기 어렵습니다. 구매를 결정하기까지 시간이 오래 소요되기도 하며 재구매가 발생하지 않을 수도 있기 때문에 LTV를 주요 지표로 두어서는 효율적인 의사 결정을 내릴 수 없습니다.
%%CTA|데모 신청하기|/ko/book-a-demo|우리 앱의 LTV를 채널별로 비교하고 마케팅 ROI를 높여 보세요.%%
## LTV를 계산하는 방법
앞서 LTV가 중요한 이유를 통해 우리 비즈니스에 필요한 지표임을 확인했다면, 다음으로는 LTV를 계산하는 방법을 알아보겠습니다.
모바일 기술의 발전으로 인해 광고 클릭 및 노출, 구매 등 유저의 행동 데이터를 실시간으로 확보할 수 있게 되면서 LTV를 세밀하게 계산할 수 있는 환경이 구축되었습니다. 하지만 다양한 계산 방법이 존재하며 고려해야 할 요소가 많기 때문에 계산이 쉽지는 않습니다. 이번 포스트를 통해 소개하는 다양한 방법을 확인하고, 각 프로덕트 및 서비스의 상황에 따라 가장 효과적인 방법을 선택하는 데에 도움이 되기를 바랍니다.
**1. 수익 x 재구매율 - 고객획득비용**
지난 MGS 2022에서 진행된 세션 중 하나인[ MMP 데이터를 활용한 리텐션 예측과 LTV 추산](https://blog.ab180.co/posts/mgs-2022-ltv-calculation)에서도 소개했듯이, 전통적으로 LTV를 구할 때에는 아래 공식을 사용해 왔습니다.
[]
여기서 t는 ‘특정 기간’, p는 ‘유저가 서비스에 지불한 값’, c는 ‘유저에게 서비스를 제공하는데 필요한 비용’입니다. 그래서 (p-c)는 ‘수익’을 의미합니다. 여기에 해당 유저의 재구매율(r)을 곱하고, 이를 서비스의 현재 가치로 환산한 다음, 고객 획득 비용(CAC)을 뺍니다. 고객 획득 비용은 앞서 언급한 것처럼 유저를 획득하기 위해 투자한 비용을 의미합니다.
다만 이 방식은 개별 유저 혹은 특정 매체를 통해 유입된 유저에 대한 LTV를 구할 수 없다는 단점이 존재합니다.
**2. (평균 구매 금액 x 총 마진 × 구매 빈도 × 고객 수명) – 고객 획득 비용**
평균 구매 금액에 평균 총 마진, 구매 빈도, 고객 수명을 모두 곱하고, 고객 획득 비용을 빼는 방법을 활용할 수도 있습니다.
여기서 평균 구매 금액은 ‘고객의 거래당 평균 구매 금액’을 의미하며, 같은 기간 동안 *총 수익 ÷ 총 주문 수* 로 계산할 수 있습니다. 총 마진은 *(총 매출 - 매출 원가) ÷ 총 수익* 으로 계산합니다. 또한, 구매 빈도는 ‘주어진 기간 동안 고객 1명의 평균 거래 수’를 의미하며, 같은 기간 동안 *총 구매 수 ÷ 고객 수* 로 계산합니다. 고객 수명은 ‘고객이 우리 비즈니스에서 구매를 유지하는 평균 기간’, 즉 고객 관계가 유지되는 기간을 의미하며 *전체* *고객 수명 ÷ 고객 수* 로 계산합니다.
**3. ARPU x 재구매율**
재구매율 및 LTV 예측을 위한 확정적(deterministic) 모형 중 하나로, 과거 데이터에 기반하여 아래와 같은 공식을 통해 LTV를 계산할 수 있습니다.
[]
여기서 ARPU(Average Revenue Per User)는 ‘유저당 평균 수익’을 의미하며, 유저가 1년 동안 잔존한다는 가정 아래 ARPU에 각 시점의 재구매율(r)을 각각 곱한 값을 모두 더하면 LTV를 추산할 수 있습니다.
단순하게는, [모바일 광고 네트워크 Tapdaq에서 만든 공식](https://www.slideshare.net/Tapdaq/ltv-slideshare)인 *LTV = ARPU x *1/Churn 을 활용할 수도 있습니다. Churn은 이탈한 고객의 비율을 의미하며, 이의 역수 1/Churn는 복귀율을 의미합니다. 즉 LTV를 *유저당 평균 수익 x 복귀율*로 계산할 수 있습니다.
## 정확한 LTV를 계산하고 비즈니스의 성장을 위한 인사이트를 얻어보세요.
LTV는 어떤 고객이 우리 비즈니스에 가장 많은 수익을 가져다주고 가장 높은 충성도를 보여주는지 파악하는 데에 효과적인 지표입니다. 의사 결정을 위한 중요한 지표이자, 비즈니스의 성장을 위한 인사이트를 얻을 수 있습니다. 그렇기 때문에 정확하게 계산해야 합니다.
정확한 LTV 계산을 위해서는 1) 정확한 데이터의 수집 2) 적절한 LTV 계산법이 필요합니다. 특히 데이터를 수집하기 위한 기반으로 에어브릿지와 같은 앱 어트리뷰션 툴 / MMP의 도움을 받을 수 있습니다. 고객 획득 비용 확인을 위한 매체별 광고비는 물론, 앱 내에서 발생한 구매 등을 기반으로 한 ARPU 등까지 한 번에 빠르게 확인할 수 있습니다. 그리고 이 데이터를 기반으로 앞서 소개한 LTV 계산 방법 중 우리 비즈니스 상황에 따른 가장 적절한 방법을 선택하여 계산하면 됩니다.
정확한 LTV 계산을 통해 비즈니스 성장을 위한 인사이트를 얻고 지속 가능한 성장을 이루어보세요.
---
## 에어브릿지, 글로벌 리뷰 사이트 G2의 마케팅 애널리틱스 High Performer로 선정
URL: https://www.airbridge.io/ko/blog/g2-winter-2023
Language: Korean
Published: 2022-12-22
앱 어트리뷰션 툴 / MMP 에어브릿지(Airbridge)가 G2의 2023 겨울 그리드 리포트(Grid® Reports)에서 [마케팅 애널리틱스(Marketing Analytics) 분야의 High Performer로 선정](https://www.g2.com/categories/marketing-analytics#grid)되었습니다.
[G2](https://www.g2.com/)는 테크 업계에서 가장 신뢰받고 있는 세계 최대 소프트웨어 및 서비스 리뷰 사이트로, 수백만 명의 사람들이 14만 5천 개 이상의 소프트웨어 프로덕트에 대한 각자의 진짜 경험을 공유하고 공감하고 있습니다. G2는 약 200만 개의 리뷰를 검토 및 승인 후 게재하여 전 세계 사람들이 보다 의미 있는 소프트웨어 구매 결정을 내릴 수 있도록 돕고 있습니다. 그리고 이 리뷰를 바탕으로 매년 최고의 소프트웨어 리스트를 발표하고 있습니다.
이번에 에어브릿지가 받은 High Performer 배지는 [마케팅 애널리틱스 카테고리](https://www.g2.com/categories/marketing-analytics)에서 고객의 기대를 뛰어넘은 프로덕트에게 주어지는 것으로, 실제 유저의 프로덕트 리뷰는 물론 온라인과 소셜 미디어의 집계 데이터를 통해 선정됩니다.
> “소프트웨어 구매를 고려하는 사람이라면 리서치 및 선택 과정에서 G2를 신뢰해도 된다는 점을 잘 알고 있습니다. 실제 유저들이 리뷰를 작성했다는 것이 검증되었으며 신뢰에 기반하기 때문입니다.” - Sara Rossio, G2 CPO
## 에어브릿지가 앱 마케팅에 도움이 될 수 있는 부분
에어브릿지가 마케팅 애널리틱스 분야에서 High Performer로 선정될 수 있었던 것은, 앱 어트리뷰션 툴 / MMP로서 고객사의 마케팅 성과 분석에 기여해가고자 하는 다양한 노력 덕분이었습니다.
에어브릿지는 LTA(Last-touch Attribution) + MTA(Multi-touch Attribution) + MMM(Marketing Mix Modeling)을 모두 제공하는 Unified Measurement Stack으로, 균형 잡힌 시각으로 마케팅 성과를 분석하도록 솔루션을 제공합니다. 특히 머신러닝 기반의 마케팅 성과 분석 기능으로 멀티 터치 어트리뷰션 및 순증가분(Incrementality) 분석과 마케팅 믹스 모델링을 통해 더 정교하고 다양한 시각에서 광고 성과 분석이 가능합니다.
이를 통해 에어브릿지는 개인정보보호 강화 등 변화하는 모바일 환경에서도 균형 잡힌 인사이트를 제공하는 통합 분석을 제공하며 고객의 진정한 성장에 기여해가고자 합니다. 궁극적으로는 단지 측정만을 위한 것이 아닌 측정을 통해 마케팅의 성공 과정에 도움이 되고자 합니다.
이외에도 에어브릿지는 MMP의 편리한 활용을 돕고자 Android, iOS 앱 및 웹에서 발생한 성과를 하나의 대시보드에서 모두 한눈에 볼 수 있도록 지원하고 있습니다. 이외에도 딥링크, 오디언스 생성, 프로드 방지 등 모바일 성장에 필요한 기능을 올인원으로 제공합니다.
앞으로도 에어브릿지는 마케터에게 진정 필요한 솔루션으로서, 마케터 관점에서 유용할 수 있는 다양한 사용 사례와 인사이트를 제공해가겠습니다.
> 에어브릿지에 대해 더 궁금하시다면, [웹사이트](https://www.airbridge.io/ko)를 방문하거나 [앱 마케팅 전문가와 직접 이야기](https://www.airbridge.io/ko/book-a-demo)나눠보세요. 혹은 [에어브릿지 블로그](https://www.airbridge.io/ko/blog)에서 최신 애드 테크 & 마테크 트렌드와 인사이트를 확인해 보세요.
---
## 모바일 앱 마케팅에 어트리뷰션 툴을 사용해야 하는 이유
URL: https://www.airbridge.io/ko/blog/why-mobile-attribution-tool
Language: Korean
Published: 2022-12-16
최근 디지털 마케팅의 흐름은 퍼포먼스 마케팅, 즉 데이터 분석을 통한 ROI 최대화로 나아가고 있습니다.
물론 마케팅에서 브랜딩을 통한 소비자의 인식 제고도 무시할 수 없는 부분입니다. 하지만 그 결과에 대해 정확한 측정이 어렵다는 문제가 있습니다. 퍼포먼스 마케팅은 투입 비용 대비 결과를 정확히 측정하여 마케팅 효율을 직접적으로 높일 수 있는 방법으로 각광받고 있습니다. 실제로 많은 기업들이 퍼포먼스 마케터를 뽑아 마케팅 성과를 분석하고 최적화하고 있습니다.
퍼포먼스 마케팅을 위해 마케터가 가장 먼저 준비해야 할 것은 당연히 데이터입니다. 마케팅에 투입한 비용과 결과에 대한 데이터가 있어야 그 성과를 분석할 수 있기 때문입니다. 하지만 한정된 개발 자원으로 데이터를 모으고 분석하는 것은 일반적인 기업에서는 굉장히 힘든 일입니다. 그래서 대부분의 기업에서는 SaaS 툴을 활용하여 이러한 자원 부족 문제를 해결하고 있습니다.
기존의 전통적인 웹 마케팅에서 가장 많이 활용하는 툴은 단연 **구글 애널리틱스**입니다. 하지만 웹 서비스 시장과 달리 모바일 앱 시장에서는 구글 애널리틱스만으로는 마케팅을 진행할 수 없습니다. 앱 트래킹에서의 부정확성 때문인데요. iOS에 대한 트래킹 한계, 세션 기준의 트래킹, 전수 데이터가 아닌 샘플링 데이터를 통한 분석, 그리고 실시간 데이터를 제공하지 않는 이유 때문입니다. 앱 시장에서 구글 애널리틱스를 사용하지 않는 보다 자세한 이유는 [이 포스트](http://blog.ab180.co/why-google-analytics-is-weak-in-mobile/)에서 확인할 수 있습니다.
이제 앱 마케팅에 어트리뷰션 툴을 사용해야 하는 이유에 대해 자세히 말씀드리고자 합니다.
## 앱 설치 성과 분석
광고를 클릭하면 웹사이트로 바로 랜딩시키는 웹 마케팅과 달리, 앱 마케팅은 마케팅을 하는 대상이 앱이라는 특수성을 가지기 때문에 광고 클릭을 통해 유저가 도달할 수 있는 곳이 앱스토어로 한정됩니다. 앱 마케팅에 대한 데이터 분석을 하는데 있어 가장 큰 문제점은 앱스토어의 소유주가 구글과 애플이라는 것입니다.
정확한 데이터 분석을 위해서는 데이터를 볼 수 없는 음영 구역이 존재해서는 안됩니다. 하지만 앱스토어는 플랫폼사(구글, 애플)의 소유이기 때문에 데이터를 모두 볼 수 없습니다. 결국 광고를 통해 앱스토어에 랜딩시킬 수는 있으나 그 사람이 실제 앱을 설치했는지는 알 수 없는 상황이 발생하게 됩니다.
[]
어트리뷰션 툴은 기본적으로 이와 같은 음영 구역을 없애, 마케터로 하여금 앱 마케팅의 성과를 분석하게 하는 도구로 개발되었습니다. 어트리뷰션(Attribution)이라는 단어가 ‘어떤 행동에 대한 원인을 찾는 과정’이라는 뜻을 갖고 있듯이, 앱 설치 또는 구매와 같은 인앱이벤트에 대한 원인이 되는 광고 클릭을 찾는 것이 바로 어트리뷰션 툴의 목적이라고 할 수 있습니다. 이 때문에 어트리뷰션 툴을 우리나라 말로 바꾸면 광고 기여도 분석 툴이라고도 할 수 있죠.
어트리뷰션 툴을 사용하면 어트리뷰션 툴에서 수집한 광고 클릭 데이터와 SDK를 통한 앱 설치 데이터가 연결되어 성과를 낸 광고를 특정할 수 있게 됩니다. 즉, 광고의 성과를 분석할 수 있게 되는 것이죠. 이는 광고 성과를 분석하여 최적화를 해야 하는 마케터에게 가장 필요한 것이라고 할 수 있습니다.
## 하나의 SDK로 간편하게 여러 광고매체의 성과 최적화
어트리뷰션 툴은 단순 성과 측정 뿐만 아니라 다양한 방법으로 마케팅 최적화를 도와줍니다. 그 중 하나가 바로 **포스트백(Postback)** 지원입니다. 포스트백은 어트리뷰션 툴에서 측정된 광고 성과를 광고매체사로 전달하는 것을 의미합니다.
일반적으로 광고를 집행하면 어트리뷰션 툴이 그 광고의 성과를 분석해주고 광고주는 그 성과를 바탕으로 광고를 최적화합니다. 그러나 광고주만 이 성과를 통해 광고를 최적화하는 것은 아닙니다. 광고매체 또한 다양한 퍼블리셔 간의 성과를 추적하고 최적화 해야하기 때문에 이 광고 성과를 필요로 합니다. 뿐만 아니라, 그들은 성과에 따라 서브 퍼블리셔들에게 광고비를 정산해야 하는 의무도 가지고 있습니다.
따라서, 광고매체는 필수적으로 광고에 대한 성과를 측정해야 합니다. 그러나 앱에서 광고 성과를 측정하기 위해서는 SDK를 설치해야 하죠. 문제는 광고매체마다 각자 자신만의 SDK를 가지고 있어서, 광고주가 여러 매체를 통해 광고를 집행하는 경우 각각의 매체 SDK를 앱에 따로 설치해야 하는 수고가 발생한다는 것입니다.
[]
포스트백은 이런 문제를 해결하고, 어트리뷰션 툴의 통합 SDK를 통해 각 매체사로 데이터를 보낼 수 있게 하는 기능이라고 할 수 있습니다. 어트리뷰션 툴을 사용하게 되면 매체는 SDK 설치 없이도 포스트백을 통해 각자의 광고 성과 분석과 최적화를 할 수 있으며, 광고주는 하나의 SDK로도 다양한 매체의 광고 집행 및 분석이 가능해집니다.
## 사기 설치 방어
어트리뷰션 툴의 사용에는 Fraud(사기 설치) 방어를 위한 목적도 있습니다. 최근 몇 년간 마케팅 시장을 이끌어온 모바일 앱 광고시장은 nCPI 매체가 지배했다고 해도 과언이 아닐 정도로 nCPI 매체의 양적 성장이 두드러졌습니다. 이는 반드시 설치를 해야 실행이 되는 앱의 특성에 따라, 단순 클릭보다는 성과 측정에 유리한 앱 설치를 기준으로 광고시장이 성장했기 때문입니다.
그러나 모든 것엔 명암이 있듯이, nCPI 매체가 성장하는 동안 이를 부정적으로 이용하려는 세력이 등장했습니다. 이들이 광고를 통해 부정 이득을 취하려는 행위를 Fraud(부정 광고 또는 사기 설치)라고 부릅니다.
최근까지 Fraud는 모바일 앱 광고시장을 어지럽히는 주범으로 꼽혀왔습니다. 모바일 앱 시장이 커지면서 광고주들의 설치 수는 증가했지만, 사실 이는 광고 성과가 아니라 Fraud라는 것을 알게 된 것이죠. 광고주들은 고객을 끌어들이기 위한 광고비를 더 이상 Fraud로 허비하고 싶어하지 않습니다.
[]
어트리뷰션 툴은 이런 광고주의 고민을 해결하기 위해 Fraud 방어 기능을 제공하고 있으며, 광고주들로 하여금 Fraud를 명확히 인지하고 대처할 수 있게 합니다. 거짓 성과에 돈을 쓰지 않을 수 있게 된 것이죠.
사실 어트리뷰션 툴은 위 이유 뿐만 아니라 딥링크 기능 활용, 리텐션 확인, 로우 데이터 분석, Re-engagement 캠페인을 위한 세그먼트 추출, 데이터 CDP로서의 역할 등 다양한 기능을 활용하기 위해 사용됩니다. 이런 다양한 기능을 잘 활용할수록 앱의 성장을 이끌어낼 수 있겠죠.
그러나 오히려 너무 많은 기능들 때문에 어트리뷰션 툴에 대해 기술적인 어려움을 느끼는 마케터들도 많은 것 같습니다. 앞으로 블로그 글을 통해 어트리뷰션 툴을 활용한 모바일 앱 마케팅에 대해서 조금 더 쉽게 풀어볼 계획입니다.
---
## 개인정보보호 시대 제대로 마케팅 성과를 측정하는 방법, 에어브릿지 마케팅 믹스 모델링을 확인해보세요
URL: https://www.airbridge.io/ko/blog/airbridge-marketing-mix-modeling
Language: Korean
Published: 2022-12-12
오늘날 모바일 광고의 가장 중요한 문제는 강화되고 있는 개인정보보호 정책으로 인하여 유저 단위 추적에 기반한 기존 광고 성과 측정 방법이 더 이상 정확하지 않다는 점입니다. iOS 14.5 업데이트 이후 앱 추적 투명성(App Tracking Transparency, ATT) 정책의 도입, 브라우저 상에서 쿠키 관련 정책 변화, 그리고 개인정보보호를 위한 다양한 정부의 정책은 유저 단위 데이터 수집을 어렵거나 불가능하게 만듭니다. 기존에 널리 사용되던 라스트 클릭 어트리뷰션(LTA, Last Click Attribution) 모델은 광고를 본 유저와 전환한 유저를 대응시킬 수 있다는 전제에 의존하고 있기에, 유저 단위 추적이 제한된 상황에서는 대단히 왜곡된 결과를 산출할 수밖에 없습니다. 이에 유저 레벨 데이터 없이, 즉 개인정보를 보호하면서도 마케팅 성과를 측정할 수 있는 **마케팅 믹스 모델링(MMM, Marketing Mix Modeling)** 기법이 전 세계적인 주목을 받고 있습니다.
이러한 흐름에 앞장서서 에어브릿지가 차세대 마케팅 성과 분석 솔루션, [에어브릿지](https://www.airbridge.io/ko) 마케팅 믹스 모델링을 베타로 공개하였습니다. 이번 포스트를 통해 마케팅 믹스 모델링이란 무엇인지 그리고 왜 지금 주목받고 있는지 알아보며 에어브릿지 마케팅 믹스 모델링에 대해 소개합니다.
## 마케팅 믹스 모델링이란?
마케팅 믹스 모델링(MMM)이란 ‘각 마케팅 활동’이 매출이나 앱 설치와 같은 ‘비즈니스 성과’에 미치는 영향력을 파악하는 거시적인 데이터 기반 통계 분석 모형입니다. 이 분석 모형을 위해 채널별 마케팅 집행 비용이나 매출과 같은 마케팅 관련 모든 데이터는 물론, 계절성, 경쟁사의 매출 현황, 거시 경제 트렌드 등과 같은 다양한 외부 데이터를 활용할 수 있습니다. 즉, 과거의 데이터를 분석하여 각각의 마케팅 활동이 비즈니스 성과에 미치는 영향을 알아내어 미래의 성과를 예측하는 겁니다. 특히 마케팅 믹스 모델링은 터치포인트나 광고 ID와 같은 유저 레벨의 데이터 없이도 채널별 마케팅 비용과 성과와 같은 집계 데이터(Aggregate data)만 있어도 모델 구축과 분석이 가능합니다.
## 왜 지금 모바일 업계에서 마케팅 믹스 모델링이 주목받을까요?
2021년 이후 모바일 업계를 강타한 화두는 iOS 14.5 업데이트와 ATT 정책의 도입을 비롯한 개인정보보호의 중요성입니다. 특히, ATT 정책의 도입으로 유저의 광고 ID(ADID)를 수집 및 활용하기 위해서는 유저의 명시적인 동의가 필요하게 되었습니다. 여기서 ADID는 광고주가 특정 기기(유저)를 추적할 때 사용하는 식별자로 여러 데이터베이스에서 수집된 터치포인트와 전환 데이터를 매칭하여 어트리뷰션 하는 데에 필요한 데이터입니다.
실제로 [2022년 3월 기준 iOS 유저 중 단 46%만이 ATT에 옵트인(opt-in)](https://www.statista.com/statistics/1281345/apple-att-opt-in-rate-by-app-category/) 하고 있습니다. 즉, 모바일 앱 유저 2명 중 1명에 대해서는 어떤 광고를 클릭해서 앱을 설치했거나 구매를 발생시켰는지 알아낼 수 없는 상황으로 기여가 이루어지지 않은(unattributed) 유저의 비중이 크게 증가하였습니다.
이외에도 브라우저 상에서 쿠키 관련 정책의 변화 및 GDPR과 CPRA 등 정부 정책 변화에 따른 유저 레벨 데이터 수집의 제약도 존재합니다. 또한, 기존 광고 성과 분석의 주류인 라스트 클릭 어트리뷰션 모델이 개인 단위 데이터에 의존하고 있으며 모든 터치포인트의 기여도를 완전히 파악할 수 없다는 점에서 새로운 분석 방법의 필요성도 계속해서 증가하고 있었습니다.
이러한 개인정보보보호 트렌드는 거스를 수 없으며 당연히 존중해야 마땅합니다. 다만, 광고주는 마케팅 비용을 집행한 만큼 얼마만큼의 효과가 있었는지 그리고 어떤 채널에서 가장 효과적이었는지 그 성과를 알고 싶어 합니다. 이에, [마케팅 믹스 모델링이 포스트 프라이버시 시대에 마케팅 집행 비용과 실제로 발생한 성과 간 사일로 현상을 해결해 주는 성과 측정 방법으로 각광받고](https://digiday.com/marketing/wtf-is-marketing-mix-modeling/) 있습니다.
### 과거의 기성 마케팅 믹스 모델링 제공사
마케팅 믹스 모델링은 수십 년 전부터 계량경제학에서 존재했던 기법으로, 주로 포장 소비재 제품(CPG, Consumer Packaged Goods) 회사들이 컨설팅에 기반하여 사용해왔습니다. 그렇다면 마케팅 믹스 모델링이 널리 보급되지 못했던 이유는 무엇일까요? 기성 마케팅 믹스 모델링 솔루션 제공사의 높은 비용과 긴 소요 시간으로 인해 접근성에 한계가 있었기 때문입니다. 과거에는 주로 컨설팅 방식으로 개별 고객사의 상황에 맞춰 다양한 변수들을 조합해 보면서 모델의 최적화를 진행하였습니다. 많은 데이터를 수동으로 전달하는 과정이 번거로웠으며, 모델 구축이나 업데이트 그리고 결과 도출에 시간이 많이 소요되었고, 서비스 비용도 높았습니다. 이로 인해 마케팅 믹스 모델링에 대한 진입장벽이 높았으며 주로 많은 마케팅 예산을 투자하는 기업에서 장기적인 마케팅 전략을 수립하는 데에 활용하곤 했습니다.
### 차세대 마케팅 믹스 모델링 솔루션 제공사
하지만 최근 들어 증가하는 마케팅 믹스 모델링에 대한 관심을 기반으로 새로운 형태의 마케팅 믹스 모델링 솔루션이 등장하기 시작했습니다. 이 차세대 마케팅 믹스 모델링 솔루션 제공사들은 SaaS 형태의 더 빠르고 쉬운 마케팅 믹스 모델링 솔루션을 합리적인 비용으로 제공합니다. 에어브릿지가 제공하는 마케팅 믹스 모델링 솔루션도 SaaS 형태로 각 서비스/프로덕트 맞춤 모델링을 빠르고 효율적으로 진행하여 성과 분석과 최적화를 돕습니다. 에어브릿지 마케팅 믹스 모델링 솔루션은 다음과 같은 특장점을 가지고 있습니다.
- 비용 효율적이고 빠르고 쉽게 활용 가능합니다. 특히 매일 업데이트되는 모델로 결과를 확인하여 빠른 변화에 대응 가능합니다.
- 데이터의 자동 수집 및 수동 업로드 기능을 모두 지원합니다. 그리고 고객사별 핵심 데이터를 기반으로 분석 모델을 고도화할 수 있습니다.
- 모델의 성능을 개선하기 위해 필요한 정기적인 보정(Calibration) 작업에 실제 실험(예: Lift Study) 데이터뿐만 아니라 에어브릿지의 Incrementality & MTA 결과도 활용 가능합니다.
- 단순한 분석을 넘어 예산 최적화와 같이 즉시 실행 가능한 액션 아이템을 대시보드 상에서 제안합니다.
## 1\. 합리적인 비용으로 빠르게 활용 가능한 에어브릿지 마케팅 믹스 모델링 솔루션
앞서 언급한 것처럼 전통적인 마케팅 믹스 모델링 제공사를 통한 분석은 장기간에 걸쳐 상당한 비용을 들이는 컨설팅 프로젝트 형태이기 때문에 보편적으로 접근하기에는 한계가 있었습니다. 하지만 에어브릿지와 같은 SaaS 형태의 마케팅 믹스 모델링 솔루션을 활용하면 매일 및 매주 업데이트되는 결과를 확인할 수 있습니다. 게다가 컨설팅 서비스에 비해 비용도 합리적입니다.
과거의 마케팅 믹스 모델링 서비스는 1년 이상의 데이터를 기반으로 모델을 구축 및 분석하기에 시간이 많이 소요되어 마케팅 전략에 빠르게 적용하기 어려웠습니다. 이에 비해 에어브릿지는 자체 개발한 머신 러닝 엔진이 6개월 치의 데이터만으로도 학습하여 분석 결과를 빠르게 확인할 수 있습니다. 그리고 매일 및 매주 업데이트되는 리포트의 데이터를 기반으로 변화에 대응하고 마케팅 전략을 최적화할 수 있습니다. 구체적으로는 매일 업데이트되는 Marketing Mix Analysis 리포트를 통해 빠르게 채널별 성과를 확인할 수 있으며, 매주 업데이트되는 Budget Optimization 리포트가 제안하는 마케팅 예산 분배로 제한된 예산 내에서 수익을 극대화할 수 있습니다.
## 2\. 데이터 자동 수집 및 수동 업로드 기능을 모두 지원하며, 고객사별 핵심 데이터 기반으로 분석 모델을 고도화할 수 있습니다.
에어브릿지는 [앱 어트리뷰션 툴](https://www.airbridge.io/ko/blog/what-is-mobile-app-attribution)/MMP이기 때문에 API를 통한 매체 연동과 자체 SDK를 통한 광고 성과 및 비용 데이터 자동 수집이 가능합니다. 따라서 이미 에어브릿지를 이미 이용 중인 고객사는 번거로운 데이터 업로드 과정을 최소화해 보다 편리한 마케팅 믹스 모델링 서비스 이용이 가능합니다. 또한, 고객사의 앱 안에서 발생하는 이벤트를 모두 추적할 수 있기 때문에 모델 고도화에 필요한 부가적인 데이터도 쉽게 확보할 수 있다는 장점이 있습니다.
다만 그렇다고 해서 반드시 에어브릿지를 사용해야만 마케팅 믹스 모델링 분석을 할 수 있는 것은 아닙니다. 최소 6개월치의 채널별 집행비용, 클릭 및 조회 수 등의 일간 데이터를 csv로 업로드하여 분석 결과를 확인할 수 있습니다. 또한, csv 업로드 → 데이터 확인 → 모델링 → 모델 확인 → 리포트 확인의 프로세스가 자동화되어 있어서, 기존 마케팅 믹스 모델링 제공사처럼 수동으로 데이터를 업로드하더라도 보다 빠르게 분석 결과를 확인할 수 있습니다.
게다가 에어브릿지 마케팅 믹스 모델링 솔루션은 이렇게 수집한 고객사별 핵심 데이터를 기반으로 모델링을 진행하여 모델의 고도화가 가능합니다. 서비스/프로덕트 맞춤 모델 기반의 고도화된 마케팅 믹스 모델링 분석을 통해 보다 정확한 분석 결과를 확인하고 성과를 극대화할 수 있습니다.
## 3\. 보정(Calibration) 작업으로 모델과 현실의 경향성 맞추기
마케팅 믹스 모델링은 과거의 데이터 상에서 마케팅 채널과 비즈니스 성과 간의 상관관계(correlation)를 바탕으로 각 채널의 성과를 측정합니다. 그러나 상관관계가 반드시 인과관계(causation)를 의미하지는 않기 때문에, 마케팅 믹스 모델링의 결과도 때로 불완전할 수 있습니다. 광고 채널의 성과를 보다 정확하게 측정한 결과가 있다면 그 측정 결과를 모델에 반영하여 보정할 수 있으며, 이 작업을 **보정(calibration) 작업**이라고 합니다.
이 보정 작업을 진행하기 위해서는 Lift Study와 같은 실험적(Experimental) 방법을 통한 실제 실험 데이터가 필요합니다. 하지만 실험적 방법을 진행하기 위해서는 비용이 소모되며 실험이 끝날 때까지 성과를 확인할 수 없다는 단점이 존재합니다. 실제 실험 데이터 기반의 보정이 가장 효과적이겠지만, 실험을 진행할 여력이 없는 경우에는 실제 실험 결과와 유사한 에어브릿지의 순증가분(Incrementality) 분석 결과를 활용하는 것도 가능합니다.
## 4\. 단순한 분석을 넘어 즉시 실행 가능한 액션 아이템 제안
과거의 마케팅 믹스 모델링 제공사는 오랜 시일이 경과한 후 보고서 형태로 결과를 제공하였다면, 에어브릿지 마케팅 믹스 모델링 솔루션을 사용하면 매일 및 매주 업데이트되는 리포트를 통해 결과를 언제든지 쉽게 확인할 수 있습니다. 편리한 인터페이스를 통해 다양한 마케팅 시나리오를 자유롭게 테스트하고, 리포트로 확인한 성과를 매일의 의사 결정에 실용적으로 사용할 수 있습니다.
에어브릿지가 제공하는 마케팅 믹스 모델링 분석 결과를 확인할 수 있는 리포트는 아래의 Marketing Mix Analysis 리포트 및 Budget Optimization 리포트 두 가지입니다.
- Marketing Mix Analysis 리포트
[__wf_reserved_decorative]
Marketing Mix Analysis 리포트에서는 유저 레벨 데이터 없이도 개별 채널의 순증가 기여도를 일자별로 확인할 수 있습니다. 이때 데이터를 비교적 수월하게 수집할 수 있는 디지털 매체뿐만 아니라 지출한 비용 정보만 있는 TV, 잡지 광고 등과 같은 오프라인 매체에 대한 기여도도 분석할 수 있습니다. Installs(App), Order Complete(App), Revenue(App), Ad Revenue(App)를 기본 메트릭으로 지원하며, 각 메트릭별 참고하면 좋은 관련 비용 메트릭으로 Count 및 eCPI도 함께 지원합니다.
- Budget Optimization 리포트
[__wf_reserved_decorative]
Budget Optimization 리포트는 설정한 예산 한도 내에서 최대의 성과를 달성할 수 있는 최적의 채널별 예산 분배를 제안합니다. 주간으로 업데이트되는 리포트를 확인하여 제한된 예산 내에서 예산 분배를 최적화하고 수익을 극대화하는 데에 도움이 될 수 있습니다.
현재의 마케팅 비용 지출 데이터는 분석을 위해 이미 수집되어 있기 때문에, 본 리포트에서는 미래 예산에 해당하는 금액을 기존 마케팅 지출 비용 대비 %로 입력하면 됩니다. 그러면 예산 프리뷰 기능을 통해 계산된 미래 예산 금액을 바로 확인할 수 있습니다. 이 설정된 예산을 기반으로 기존의 채널 믹스와 모델이 제안하는 비용 및 예상 성과를 비교할 수 있습니다.
이외에도 Model Manager를 통해 마케팅 믹스 모델링 모델과 관련된 다양한 설정값 및 학습 현황에 대한 정보를 투명하게 공개하며 언제든 빠르게 확인할 수 있습니다.
> 👉 구체적인 마케팅 믹스 모델링 활용 사례가 궁금하다면 [멜리즈와 에어브릿지가 함께 진행한 마케팅 믹스 모델링 케이스 스터디](https://www.airbridge.io/ko/case-studies/meliz-privacy-first-mmm)를 확인해 보세요.
>
> 에어브릿지의 고객사 멜리즈는 에어브릿지가 자체 개발한 마케팅 믹스 모델링 모델을 활용하여 채널별 마케팅 성과를 분석하고, 마케팅 예산 재분배를 통해 동일한 예산으로 5%의 앱 설치 증가를 추가적으로 달성할 수 있다는 결과를 확인하였습니다.
## 개인정보보호 강화 트렌드 속 떠오르는 차세대 마케팅 성과 분석 솔루션 에어브릿지 마케팅 믹스 모델링
글로벌 마케팅 업계에서 가장 화두인 개인정보보호 트렌드 속에서 에어브릿지 마케팅 믹스 모델링은 차세대 마케팅 성과 분석 솔루션으로서 마케팅은 물론 모바일 앱 성장에 도움이 되어드릴 수 있습니다. 에어브릿지 팀이 자체 개발한 머신 러닝 엔진으로 마케팅 채널의 효과 분석과 예측, 그리고 예산 최적화를 위한 시뮬레이션까지 모두 한 번에 실행 가능합니다. 서드 파티 쿠키 및 광고 ID 수집 제한 등 외부 환경 변화에 따른 변화 속에서도 마케팅 믹스 모델링으로 제대로 마케팅 성과를 분석하고 성과를 최적화해보세요.
> 베타로 공개된 에어브릿지 마케팅 믹스 모델링 솔루션을 직접 활용해 보고 싶다면 담당 고객 성공 매니저 혹은 [에어브릿지의 모바일 마케팅 전문가와 이야기 나눠 보세요.](https://www.airbridge.io/ko/book-a-demo)
---
## Airbridge X RevenueCat 파트너십 체결
URL: https://www.airbridge.io/ko/blog/airbridge-revenue-cat-partner
Language: Korean
Published: 2022-12-05
에어브릿지(Airbridge)가 글로벌 앱 구독 솔루션 레비뉴캣(RevenueCat)과 파트너로 함께 하게되었습니다.
이 파트너십을 통해 구독 관련 이벤트가 레비뉴캣에서 에어브릿지로 전송이 되어, 에어브릿지에서 보다 정확한 광고 채널별 수익을 확인할 수 있습니다.
에어브릿지 대시보드에서 구독, 갱신, 구독 취소와 같은 구독 관련 이벤트 확인은 물론, 어떤 채널이 가장 많은 구독 수익을 창출하고 있는지 더 잘 이해하고 그에 맞게 마케팅 캠페인을 최적화할 수 있습니다.
글로벌 최대 앱 구독 솔루션 기업인 레비뉴캣의 파트너로써, 고객들에게 보다 정확한 채널별 구독과 수익의 기여도를 제공할 수 있게 되어 기쁩니다. 앞으로도 에어브릿지는 레비뉴캣과과의 지속적인 커뮤니케이션 및 공동 작업을 통해 연동을 보다 강화할 갈 예정입니다.
아래 가이드 문서 링크를 통해 에어브릿지와와 레비뉴캣의의 더 자세한 연동 방법을 확인해 보세요.
[]
## 레비뉴캣(RevenueCat) 소개
레비뉴캣(RevenueCat)은 웹과 앱에서 발생하는 구독 관련 이벤트를 추적합니다. 레비뉴캣을 이용하면 구독을 관리와 설정이 간단해집니다. 인앱 결제를 설정은 물론, 웹과 앱의 수익을 분석하고 개선할 수 있습니다. 글로벌 앱 구독 솔루션 레비뉴캣에 대해 더 자세한 내용은 [여기](http://revenuecat.com/)에서 확인해보세요.
---
## 모바일 광고의 실제 성과를 분석할 수 있는 방법, 순증가분(Incrementality) 이해하기
URL: https://www.airbridge.io/ko/blog/marketing-incrementality
Language: Korean
Published: 2022-11-30
모바일 광고 성과 측정, 실제 효과까지 측정하고 계시나요?
모바일 광고 성과 측정에서 가장 어려운 점은 페이드(paid) 광고로 인해 발생한 설치나 앱 내 전환에 섞여 있는 오가닉 트래픽을 구별해낼 수 없다는 점입니다. 이로 인해 오가닉으로도 충분히 발생할 수 있었던 전환에 대해 마케팅 비용을 투자하며 비효율적인 마케팅 예산을 집행하고 있었을지 모릅니다. 혹은 오가닉과 페이드 마케팅의 성과가 섞여 있는 분석 결과로 인해 잘못된 판단에 기반한 미디어 믹스를 구성하고 있었을지도 모릅니다.
이러한 상황 속에서 실제 광고의 효과는 어떻게 파악할 수 있을까요? 광고 없이도 구매까지 할 수 있는 유저에 대한 광고비는 어떻게 절약할 수 있을까요? 이를 파악하기 위해서는 광고의 실제 효과인 ‘인크리멘탈리티(Incrementality)’를 측정해야 합니다.
## 마케팅에서 ‘인크리멘탈리티’의 의미 및 역할
인크리멘탈리티는 사전적 정의로 ‘증분’, ‘순증가분’을 의미합니다. 그리고 이를 마케팅에서 사용한다면 ‘광고로 인한 전환의 순증가분’이라고 정의할 수 있습니다. 인과관계를 측정하여 광고 시청과 같은 특정한 상호 작용 없이도 발생할 수 있었던 전환에 대해 분석할 수 있습니다. 즉, 광고와 전환 간의 실제로 발생한 ‘증분 효과’를 파악한다는 것을 의미하며, 광고를 했을 때와 하지 않았을 때의 성과를 비교하여 진짜 광고 성과를 알려줄 수 있습니다. 마케팅 활동이 얼마나 효과적이었는지 측정하여 마케팅 활동 없이는 발생하지 않았을 실제 가치가 얼마만큼 발생했는지 확인할 수 있습니다.
마케팅에서 인크리멘탈리티를 분석하는 목적은 다음과 같이 두 가지로 정리하여 이야기할 수 있습니다. 먼저, 광고로 인한 전환을 수치로 증명하여 마케팅 비용 사용에 대한 가치를 정당화할 수 있습니다. 두 번째로는, 어떤 채널이 가장 증분 효과가 있었는지 확인하여 마케팅 비용에 대한 효율성을 높여줄 수 있습니다.
### 1) 마케팅 비용 사용에 대한 가치 정당화
모든 브랜드들이 마케팅으로 달성하고 싶은 궁극적인 목표는 ‘수익 증대’이기 때문에, 마케팅으로 사용한 비용에 비해 그 이상의 증분 결과가 나와야 합니다. 투자한 만큼 나아가 그 이상의 전환과 매출이 발생해야 합니다. 그리고 이 성과가 마케팅으로 발생하였다는 것을 증명하는 과정도 필요합니다. 하지만 오늘날 표준으로 자리 잡은 라스트 터치 어트리뷰션(LTA, Last-Touch Attribution) 모델은 최종 광고 채널에 전환에 대한 모든 기여를 분석하기 때문에, 측정된 성과가 실제 광고로 인해 발생했다는 것을 증명하기에 한계가 있습니다. LTA 모델은 마지막 광고가 무조건 100% 전환 성과를 이끌어냈다고 보지만, 유저가 원래부터 보유한 전환에 대한 잠재적인 경향성이나 앞선 터치포인트들을 고려하면 그렇지 않기 때문입니다. 이때 인크리멘탈리티를 측정하면, LTA 모델이 가진 한계를 보완할 수 있으면서도 마케팅 비용 사용에 대한 가치, 즉 투자를 정당화할 수 있는 이유를 발견할 수 있습니다. 전환의 성과가 광고 활동이 아니었다면 일어나지 않았을 결과라는 것을 증명해 줄 수 있습니다.
### 2) 가장 효과적인 광고 채널 파악으로 마케팅 효율 개선
그리고 더 나아가 인크리멘탈리티 분석을 통해 효율적인 광고 매체 운영을 위해 가장 많은 증분 효과를 발생시킨 채널을 파악할 수 있습니다. 그리고 이를 예산 분배에 즉각 적용하여 광고비의 낭비를 막고 마케팅 효율을 높일 수 있습니다.
오가닉 성과와 마케팅으로 발생한 성과 간 겹치는 부분은 구별하거나 정확하게 측정하기 어렵기 때문에 부정확한 의사 결정이나 예산 설정으로 이어질 수 있습니다. 하지만 인크리멘탈리티 측정을 통해서 각 채널의 기여분 및 영향력을 파악하고 분석할 수 있습니다. 구체적으로는 아래와 같은 질문에 답변하고 보다 성과를 개선할 수 있습니다.
- 특정 매체에서 광고 집행을 중단하면 성과에 어떤 영향을 미칠까?
- 특정 매체, 캠페인, 광고 크리에이티브가 매출 증대에 실질적으로 얼마만큼의 영향을 미치는가?
- 어떤 채널, 플랫폼, 캠페인, 퍼블리셔 등이 성과 달성에 가장 큰 영향을 미치는가? ex. 수익 증대, LTV 개선
- 어떤 채널에서 얼마만큼의 예산을 증대 혹은 감소하면 가장 두드러지는 성과를 내는 광고 집행이 가능할까?
이와 같이 인크리멘탈리티는 마케팅 활동에 대한 진정한 영향력을 보다 정확하게 파악하여 인과 관계를 이해하도록 도와줄 수 있습니다. 인크리멘탈리티 분석을 통해 보다 정확한 데이터 및 데이터 기반의 의사 결정을 내릴 수 있는 실행 가능한 인사이트를 얻을 수 있습니다. 나아가 어떤 오디언스에게 어떤 광고를 어떤 플랫폼에서 제공해야 하는지 보다 가시적으로 파악하는 데에 도움이 될 수 있습니다.
## 인크리멘탈리티의 두 가지 측정 방법 - 실험형과 관측형
그렇다면 인크리멘탈리티는 어떻게 측정할 수 있을까요? 광고주가 활용할 수 있는 인크리멘탈리티 측정 방법에는 몇 가지가 존재하지만, 크게는 실험적(Experimental) 방법과 관측적(Observational) 방법으로 나눌 수 있습니다. 간단하게 설명하자면, ‘실험’은 오디언스를 테스트 집단과 컨트롤 집단으로 나누고 두 집단 간의 전환율 간의 차이로 각 채널에 대한 인크리멘탈리티를 측정하는 방식이며, ‘관측’은 과거의 데이터를 활용해 인과 추론을 시도하는 방식입니다.
이 두 방법 중 보다 정확한 방법을 꼽자면 ‘실험’이라고 할 수 있습니다. 이 실험적 방법을 실행하기 위해서는 두 집단을 완전히 동질적으로 나눠야 하는데, 이를 위해서는 전문적인 데이터 사이언티스트의 도움이 필요하며 아무리 신중하게 설계된다고 하더라도 두 집단의 구성이 완전하게 동일하기는 어렵습니다. 이때 주요 광고 매체인 메타(Meta)와 구글(Google)에서 제공하는 RCT(Randomized Controlled Trials) 방법을 통해 테스트 집단과 컨트롤 집단을 랜덤하게 선정하여 실험을 통해 인크리멘탈리티를 측정할 수 있습니다.
다만, 실험을 수행하기 위해서는 상당한 시간과 비용이 소모될 수밖에 없습니다. 실험 기간 동안 컨트롤 집단에 광고가 노출되지 않음으로써 같은 기간 동안 더 많은 매출이 발생할 수 있는 기회를 잃어버릴 수 있으며, 결과 분석에도 시간이 소요되어 어제까지의 분석 결과를 당장 오늘 적용하기 어려울 수도 있습니다. 이러한 한계를 극복하는 방법으로는 에어브릿지와 같은 SaaS 솔루션을 통해 과거 유저의 행동 데이터에 머신러닝과 통계학적 기법을 접목한 관측 방법을 활용할 수 있습니다.
### 1. 메타 및 구글의 실험적 방법
주요 광고 매체인 메타와 구글에서 제공하는 Lift Study를 통해 평균적으로 비슷하게 선정된 테스트 집단과 컨트롤 집단을 대상으로 광고 노출도를 통제함으로써 실제 ‘실험’을 진행할 수 있습니다. 이 방법은 실제 실험 결과를 토대로 하기 때문에 가장 믿을 수 있는 표준 지표이자, 가장 진정한 인과 효과를 보여주는 값을 계산할 수 있는 방식이라고 할 수 있습니다. 또한 실제 데이터를 활용하기 때문에 정확도가 높습니다. 아래는 각 매체에서 제공하고 소개하고 있는 인크리멘탈리티를 측정 방법입니다.
- [메타의 리프트 테스트](https://www.facebook.com/business/help/552097218528551?id=546437386202686)
메타에서 제공하는 리프트 테스트는 페이스북(Facebook)의 광고에 대한 증분 효과를 이해할 수 있는 효과적인 방법입니다. 브랜드 인지(Brand recognition) 혹은 전환(Conversion)과 같은 특정 비즈니스 목표에 대한 인과적인 영향을 이해하기 위해, 광고주의 페이스북 광고를 봤던 그리고 보지 못했던 두 그룹으로 나누어서 비교하는 일종의 실험입니다. 광고주의 오디언스와 일치하지만 의도적으로 광고를 보지 못하도록 만들어서 ‘통제 그룹’으로 만듭니다. 구체적으로는 전환 목표에 대한 페이스북 광고의 증분 효과를 확인할 수 있는 ‘[Conversion Lift tests](https://www.facebook.com/micro_site/url/?click_from_context_menu=true&country=KR&destination=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2F221353413010930&event_type=click&last_nav_impression_id=2pk0uTl2i5zc3YLAd&max_percent_page_viewed=81&max_viewport_height_px=1194&max_viewport_width_px=1268&orig_request_uri=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2F552097218528551%3Fid%3D546437386202686&primary_cmsid=552097218528551&primary_content_locale=en_US®ion=apac&scrolled=true&session_id=1BKd26NmVGzqGimqL&site=fb4b&extra_data%5Bview_type%5D=v3_initial_view&extra_data%5Bsite_section%5D=help&extra_data%5Bplacement%5D=%2Fbusiness%2Fhelp%2F552097218528551)’, 브랜드 인지 목표를 위한 페이스북 광고의 증분을 확인할 수 있는 ‘[Brand Lift tests](https://www.facebook.com/micro_site/url/?click_from_context_menu=true&country=KR&destination=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2F1693381447650068&event_type=click&last_nav_impression_id=2pk0uTl2i5zc3YLAd&max_percent_page_viewed=81&max_viewport_height_px=1194&max_viewport_width_px=1268&orig_request_uri=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2F552097218528551%3Fid%3D546437386202686&primary_cmsid=552097218528551&primary_content_locale=en_US®ion=apac&scrolled=true&session_id=1BKd26NmVGzqGimqL&site=fb4b&extra_data%5Bview_type%5D=v3_initial_view&extra_data%5Bsite_section%5D=help&extra_data%5Bplacement%5D=%2Fbusiness%2Fhelp%2F552097218528551)’, 그리고 ‘[Experiements](https://www.facebook.com/micro_site/url/?click_from_context_menu=true&country=KR&destination=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2Ftestandlearn&event_type=click&last_nav_impression_id=2pk0uTl2i5zc3YLAd&max_percent_page_viewed=81&max_viewport_height_px=1194&max_viewport_width_px=1268&orig_request_uri=https%3A%2F%2Fwww.facebook.com%2Fbusiness%2Fhelp%2F552097218528551%3Fid%3D546437386202686&primary_cmsid=552097218528551&primary_content_locale=en_US®ion=apac&scrolled=true&session_id=1BKd26NmVGzqGimqL&site=fb4b&extra_data%5Bview_type%5D=v3_initial_view&extra_data%5Bsite_section%5D=help&extra_data%5Bplacement%5D=%2Fbusiness%2Fhelp%2F552097218528551)’ 세 가지 방법이 있습니다.
- [구글의 리프트 측정](https://support.google.com/google-ads/topic/10283810)
구글의 리프트 측정 툴은 광고의 효과를 측정할 수 있습니다. 그리고 이를 통해 구글 광고 캠페인을 조정하거나 개선하는 데에 활용할 수 있습니다. 구글에서는 동영상 광고의 효과를 측정하여 캠페인을 조정하거나 개선하는 데에 활용할 수 있는 [Brand Lift](https://support.google.com/google-ads/answer/9049825?hl=en), 그리고 광고를 보는 오디언스가 직접 유도하는 전환 수나 사이트 방문 등의 행동을 측정할 수 있는 [Conversion Lift](https://support.google.com/google-ads/answer/12003020?hl=en)의 두 가지 방법을 제공하고 있습니다. 구글의 리프트 측정 툴은 일부의 구글 애즈 광고주만 활용 가능하기 때문에, 계정 담당자에게 문의해야 합니다.
### 2. 관측형(Observational) 방법 - Matching, PSM
앞서 언급한 것처럼 실험형 방법의 실험 수행에 대한 시간 및 비용 소모라는 제약과 단점을 보완하는 방법으로 관측형 모델을 활용하여 인크리멘탈리티를 측정할 수 있습니다. 하지만 이때 주의해야 하는 점은, 과거 데이터 속에 처치 군(treatment group)과 컨트롤 군(control group)이 이미 정해져 있기 때문에 이를 비교 가능하게 만들어야 합니다.
구체적으로는 Matching 방법을 활용할 수 있는데 우선 두 집단을 유사하게 만들 기준을 잡고, 해당 기준이 서로 가장 유사한 사람들끼리는 짝을 지어 남기고 나머지는 인과 추론에 사용하지 않는 방식으로 진행됩니다. 혹은 Propensity Score Matching(PSM) 분석 방법을 활용할 수 있는데, PSM은 개별 고객이 광고에 노출될 수 있는 확률로 산출된 유저별 경향성을 뜻하며, 여기에 통계적 기법을 적용해 실제로 광고가 순수하게 미치는 증분 효과를 파악할 수 있습니다.
## 광고의 진짜 효과가 궁금하다면, 인크리멘탈리티를 확인해야 합니다.
지금까지 마케팅 성과 측정에 왜 인크리멘탈리티 분석이 필요한지, 그리고 두 가지의 인크리멘탈리티 분석 방법을 소개했습니다. 마케팅 비용을 집행하면 그 성과를 측정해야 하는 것은 당연합니다. 보다 정확하고 가시적인 성과 측정 방식을 활용하고 싶다는 것은 모든 마케터와 브랜드의 바람일 겁니다.
마케팅 성과 측정 모델의 표준인 LTA 모델은 오랫동안 발전되어온 훌륭한 방법론이지만, 진짜 성과 측정을 위한 유일한 정답일 수는 없습니다. 광고의 진짜 효과가 궁금하다면, 인크리멘탈리티를 확인해 보세요.
> 👉 광고의 순증가분(Incrementality)을 알아야 하는 이유와 통계학이 제시한 여러 방법을 통한 순증가분 분석 방법이 궁금하다면, 에어브릿지 화이트페이퍼 [광고가 만든 '진짜 성과'를 찾아라](https://www.airbridge.io/ko/library/incrementality-whitepaper) 에서 확인해보세요.
---
## 릴리즈 노트: 2022년 9~10월
URL: https://www.airbridge.io/ko/blog/release-notes-sep-oct-2022
Language: Korean
Published: 2022-11-16
2022년 9월과 10월 에어브릿지가 대시보드의 활용성 개선 및 추가 데이터 제공, 신규 매체 연동, SDK 릴리즈로 한층 더 유저 친화적으로 거듭났습니다. 보다 정확하고 빠른 데이터와 함께 모바일 앱의 성장에 기여할 에어브릿지의 신규 기능 릴리즈 소식을 한번에 확인해보세요.
> **[주요 업데이트 미리 보기]
**
- 저장하지 않은 Actuals 리포트 공유 기능
- Revenue 리포트에서 Paying Users, ARPPU, User Count 확인하기
- Active Users 리포트에서 Web 메트릭 확인하기
- 추가 매체 연동
- SDK 릴리즈
## 🤝 저장하지 않은 Actuals 리포트 공유 기능
에어브릿지에서 트래킹하고 있는 데이터를 자유롭게 확인할 수 있는 통계 리포트인 Actuals 리포트의 활용성이 개선되었습니다. 기존에는 저장한 Actuals 리포트에 활성화되는 ‘공유하기’ 기능 혹은 [설정값 복사]를 통해 공유했다면, 이제는 저장하지 않은 리포트도 **url을 복사하고 붙여넣기** 해서 리포트를 빠르고 쉽게 공유할 수 있습니다.
[]
상단의 이미지와 같이 확인하고 싶은 데이터를 설정하고 적용하게 되면 url의 마지막에 config id가 추가됩니다. 예를 들어 /{app name}/reports/actual**/c/nqoqgea **과 같이 임의의 id가 추가됩니다. 그리고 해당 url을 복사해서 공유하게 되면 설정한 리포트를 그대로 확인할 수 있습니다. 또한, Actuals 리포트를 저장하지 않은 상태에서 메트릭, 그룹바이, 필터 설정을 바꿔가며 데이터를 조회하다가 이전 페이지로 되돌아가게 되면 이전에 조회한 설정의 url로 돌아갈 수 있습니다.
단, 이 url 공유하기 기능은 보안을 위해 같은 앱의 사용자에게만 공유할 수 있습니다. 그리고 같은 앱에 소속된 사용자라고 하더라도 다른 데이터 권한을 가진 사내 마케터와 대행사 간 url은 공유할 수 있지만 각 권한에 따른 데이터만 조회할 수 있습니다.
## 📊 에어브릿지 리포트에서 더 많은 데이터 확인하기
### Revenue 리포트에서 ARPPU, 유저수, 이벤트수, Paying Users 확인하기
매출에 관한 다양한 지표를 확인할 수 있는 통계 리포트인 Revenue 리포트에서 더 많은 데이터를 확인할 수 있게 되었으며 사용성이 개선되었습니다.
[]
1. 신규 메트릭 추가
Revenue 리포트에서 스타트 이벤트와 레비뉴 이벤트를 실행한 유저 그룹(코호트)을 대상으로 총 6가지 메트릭을 일자별로 확인할 수 있게 되었습니다. 기존에 제공되었던 매출액, ROAS, ARPU 외에도 유저수, 이벤트수, ARPPU가 추가되었습니다.
- 유저수: 스타트 이벤트 및 레비뉴 이벤트를 수행한 고유한 유저의 수
- 이벤트수: 스타트 이벤트 이후로 발생한 레비뉴 이벤트의 수
- ARPPU(Average Revenue per Paying User): 스타트 이벤트와 레비뉴 이벤트를 수행한 유저의 평균 매출액 (계산 방법: Day N까지의 누적 매출액 / Paying Users)
2. 서브 메트릭 조정 가능 기능 및 메트릭 추가 제공
Revenue 리포트는 유저가 선택한 메트릭에 대한 수치를 확인할 때 참고하면 좋을 비교 지표로 서브 메트릭을 제공하고 있습니다. 메트릭을 선택하면 3~4개의 서브 메트릭이 자동으로 선택되고 하단의 표에 고정 열로 추가됩니다. 이번 업데이트를 통해 서브 메트릭에 ‘Paying Users’(선택 기간 내 스타트 이벤트와 레비뉴 이벤트를 실행한 코호트의 유저수)가 추가되었으며, 서브 메트릭을 선택/삭제/순서 조정 기능을 제공하게 되었습니다.
👉 [유저 가이드에서 Revenue 리포트 설정 방법 더 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/4409764368409-Revenue)
### Active Users 리포트에서 Web 메트릭 확인하기
[]
Active Users 리포트는 선택한 기간 동안의 Active Users, Paying Users, Revenue, ARPU, ARPPU을 확인할 수 있는 리포트로, 기간 변경을 통해 일(DAU), 주(WAU), 월(MAU) 분석이 가능합니다.
Active Users 리포트에서 기존에는 앱의 데이터만 확인할 수 있었다면, 이제는 **Web 메트릭이 추가되어 웹 서비스에 대해서도 DAU, WAU, MAU를 조회할 수 있게 되었습니다.** 앱 또는 웹 메트릭을 선택하여 각각의 데이터를 확인할 수 있으며, 앱+웹 메트릭의 동시 조회는 불가능합니다.
👉 [유저 가이드에서 Active Users 리포트 설정 방법 더 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/900001042403-Active-Users)
## 🖇 추가 매체 연동
에어브릿지는 지속적으로 파트너사를 확대해가고 있으며, 추가적인 매체 연동을 진행하고 있습니다. 에어브릿지에 연동된 채널인 Integrated Channels는 사전에 정의된 구분 값을 가지고 있으며, 대시보드의 [Integrations - Integrated Ad Channels] 탭에서 각 매체와 편리하게 연동할 수 있습니다.
**9월과 10월에는, 맨플러스_CPI, VirtueAds, MelodonG, SK Planet이 추가로 연동되었습니다.**
에어브릿지와 연동된 전체 Integrated Channels 리스트는 [여기 링크](http://abit.ly/integrated-channels)에서 확인해 보세요!
## ⛏ SDK 릴리즈
에어브릿지는 앱(Android & iOS) 및 웹 SDK를 기반으로 웹과 앱 데이터를 통합적으로 분석합니다. 또한 지속적인 업데이트를 통해 사용성을 개선하고 있습니다.
9월과 10월에는 아래의 SDK가 릴리즈 되었습니다. 연결된 링크에서 자세한 Release Note를 확인할 수 있습니다.
- [Airbridge Android SDK](https://developers.airbridge.io/docs/android-sdk-release-note) - Airbridge Android SDK 2.19.2
- [Airbridge iOS SDK](https://developers.airbridge.io/docs/ios-sdk-release-note) - Airbridge iOS SDK 1.26.0, Airbridge iOS SDK 1.27.0
---
## 업데이트된 에어브릿지 Retention 리포트를 소개해요
URL: https://www.airbridge.io/ko/blog/airbridge-retention-report-update
Language: Korean
Published: 2022-11-15
에어브릿지(Airbridge)에서 Beta로 제공되었던 업데이트된 **Retention 리포트**가 정식으로 릴리즈되었습니다! 기존의 Retention 리포트보다 더 다양한 그룹바이와 필터 설정을 제공하며, 특정한 조건을 만족하는 사용자 그룹인 코호트를 통해 보다 세분화된 분석을 지원합니다.
특히 이번 업데이트에는 [Airbridge가 직접 개발한 OLAP DB, Luft를 소개합니다.](https://blog.ab180.co/posts/introducing-luft) 포스트에서 다룬 Luft 데이터베이스를 기반으로 한 코호트 분석 기능이 포함됩니다.
**** 기존 Retention 리포트는 지원이 종료됩니다.**
## 업데이트된 에어브릿지 Retention 리포트는,
이번에 에어브릿지에서 업데이트하여 제공하는 **리텐션 리포트(Retention Report)**는 제품을 사용하는 유저들의 어트리뷰션 결과별로 서비스 내에서 **N Day 리텐션*(잔존율)**을 확인할 수 있는 리포트입니다. 리텐션 리포트를 통해 ‘제품 및 서비스에 대한 충성도가 높은 유저’ 유입에 가장 효과적인 채널과 캠페인, 광고 그룹이 무엇인지 확인할 수 있습니다. 이 리포트의 성과를 기준으로, 캠페인을 최적화하거나 유료 광고 정산에 활용할 수 있습니다.
> ***리텐션**: 시간이 지날수록 얼마나 많은 유저가 제품으로 다시 돌아오는지를 측정한 것. (=잔존율)
## Retention 리포트 기능 업데이트의 하이라이트
이번 Retention 리포트에서 업데이트된 주요 기능은 세 가지입니다.
1. 스타트 이벤트(Start Event)를 지정해 특정 이벤트를 수행한 사용자 중 우리 앱으로 다시 돌아온 유저 수를 확인할 수 있습니다.
2. Measurement Option으로 User Journey 상의 리턴 이벤트를 구분하여 리텐션을 분석할 수 있는 기능을 제공합니다.
3. 리텐션 분석의 대상이 될 유저 그룹을 설정하여 ‘코호트’별 리텐션 데이터를 조회할 수 있습니다.
[]
### 1. 스타트 이벤트 지정
앱에 유저가 유입되었다고 판단하는 이벤트인 *스타트 이벤트(Start Event)**를 설정하여 특정 이벤트를 수행한 사용자의 리텐션을 확인할 수 있습니다.
[]
- 스타트 이벤트로 설정 가능한 이벤트: Installs(App), Deeplink Opens(App)
### 2. Measurement Option으로 User Journey 상의 리턴 이벤트 구분하기
에어브릿지의 Retention 리포트는 두 가지 Measurement Option을 제공합니다. (추후 추가 Option 제공 예정) 옵션을 선택함으로써, 기여에 상관 없이 전체 유저 여정에 대한 리텐션을 측정하거나*(General)*, 혹은 스타트 이벤트로 유저 여정을 구분하여 각각의 스타트 이벤트에 대한 리텐션을 측정할*(Confined)* 수 있습니다.
- **General**: 기여가 다른 스타트 이벤트가 다시 발생하더라도, 기존 User Journey가 그대로 유지되기 때문에, *분석 기간 전체에 대한 유저 인게이지먼트를 확인하는 용도*로 사용 가능.
[]
- **Confined**: 리턴 이벤트에 대한 기여를 새로운 스타트 이벤트가 발생하기 전까지로 한정함. 여러 개의 스타트 이벤트가 발생했다면 *각 스타트 이벤트별 유저 인게이지먼트를 ‘General’보다 더 엄밀하게 확인하는 용도로 사용 가능.*
[]
### 3. 특정 코호트(Cohort)의 설정
[]
리텐션 데이터를 볼 사용자 모수로 ‘**코호트(Cohort)**’를 설정할 수 있는 기능이 추가되었습니다. 코호트는 ‘특정한 조건을 만족하는 유저 그룹’으로, 코호트 설정을 통해 리텐션 분석의 대상이 될 유저 그룹을 정의할 수 있습니다. 코호트 분석 기능은 특정 이벤트가 리텐션에 어떤 영향을 미치는지 파악하는 데에 도움이 됩니다.
***코호트**: 동질 집단. 같은 성격을 갖는 유저 집단. 특정 이벤트를 발생시켰거나, 특정 이벤트 프로퍼티를 가진 이벤트를 발생시킨 특정 유저 그룹.
### 예를 들어, '지난 30일간 1번 이상 구매 이벤트를 수행한 사용자'를 대상으로 지난 30일간 Install을 통해 서비스 유입 후 리텐션을 조회해 보겠습니다.
1. **Add Cohort**로 지난 30일간 1번 이상 구매 이벤트 수행 완료한 유저 코호트를 설정합니다.
- performed “Order Complete (App)”, with “more than or equals” “1 time”, any time “during last” “30 days”으로 설정
[]
1. 스타트 이벤트를 Installs(App), 리턴 이벤트를 Any Event(App)으로 설정합니다.
[]
더 자세한 에어브릿지 Retention 리포트 이용 방법은 아래 링크와 연결된 이용 가이드 문서에서 확인할 수 있습니다. 에어브릿지 Retention 리포트를 통해 어떤 채널에서 유입된 유저가 우리 앱과 더 인게이지하고 충성도가 높은지 분석해 보세요!
[]
---
## 모바일 앱 어트리뷰션이란? 개념과 작동 원리 알아보기
URL: https://www.airbridge.io/ko/blog/what-is-mobile-app-attribution
Language: Korean
Published: 2022-11-10
마케팅 전략을 기획하고 수행하는 데에는 반드시 ‘데이터’가 필요합니다. 트렌드에 민감하게 반응하며 상황에 따라 빠르게 변화하는 것도 필요하지만, 이때 데이터가 아닌 단지 감에 의존해서는 대내외적인 변화에 잘 대응하며 성장하고 있는 것인지 확신할 수 없습니다. 우리 서비스 및 프로덕트에 대한 유저의 반응 및 행동 데이터를 기반으로 인사이트를 얻고, 이를 통해 미래의 마케팅 전략을 기획해간다면 보다 지속가능한 성장으로 이어질 수 있습니다.
특히, 모바일 앱 마케팅에서 가장 기본적으로 살펴보아야 하는 데이터는 ‘모바일 앱 어트리뷰션’ 데이터입니다. 이미 집행한 마케팅의 성과 데이터를 기반으로 인사이트를 얻고 이후 어떤 캠페인 혹은 채널에 더 많은 투자를 진행해야 할지 결정할 수 있기 때문입니다. 모바일 앱 마케터라면 앱 어트리뷰션 데이터는 이미 일상적으로 확인하는 데이터 중 하나일 겁니다. 이번 포스트는 모바일 앱 어트리뷰션을 이해하기 위한 시작점으로 어트리뷰션의 의미와 필요성 그리고 모바일 앱 어트리뷰션을 위해 반드시 필요한 앱 어트리뷰션 툴의 데이터 수집 및 처리 과정에 대한 기본 내용을 담았습니다. 모바일 앱 어트리뷰션 데이터에 대한 기본 이해를 기반으로 마케팅 캠페인을 최적화하고 프로덕트의 성장에 기여해보세요.
## 어트리뷰션의 의미
어트리뷰션이란, 마케팅 캠페인이라는 ‘원인’과 앱 설치나 인앱 전환과 같은 성과의 ‘결과’를 매칭하여 ‘기여도’, 즉 ‘인과 관계’를 분석하는 것을 의미합니다. 간단하게 말하자면, 마케팅 성과에 대해 어떤 요소가 영향을 미쳤는지 파악하는 과정입니다.
이렇게 어트리뷰션을 하는 목적은 과거의 데이터를 통해 채널 및 캠페인의 효과에 대해 분석하여 미래 마케팅 계획에 반영하기 위함입니다. 특히, 페이드(Paid) 마케팅의 경우 비용(자원)을 투자했기 때문에 수익이라는 결과로 이어져야 합니다. ROI(Return on Investment)를 높이기 위해서는 한정된 자원 속 최대의 효과를 낼 수 있는 채널과 캠페인을 발견해야 합니다. 어트리뷰션 데이터가 이를 위한 인사이트를 얻는 데에 도움이 될 수 있습니다.
> 👉 광고 성과를 결정하는 규칙인 어트리뷰션 모델에 대한 보다 자세한 내용은 [에어브릿지 핸드북 01 - 어트리뷰션](https://www.airbridge.io/ko/library/airbridge-handbook-attribution)에서 확인해보세요.
## 모바일 앱 어트리뷰션을 위해서는 앱 어트리뷰션 툴이 필요합니다.
모바일 어트리뷰션 이전부터 존재하고 있던 웹 어트리뷰션은 쿠키(cookie), UTM 파라미터, 리퍼러(Referrer) 등을 통해 우리 웹사이트를 방문하기 이전에 유저가 어느 사이트 및 링크를 통해 유입되었는지 파악할 수 있었습니다. 하지만 모바일 앱의 경우 앱 마켓을 거치는 과정에서 이러한 데이터가 사라지고 연결이 끊어지게 됩니다. 이때 [에어브릿지](http://www.airbridge.io/ko)와 같은 모바일 앱 어트리뷰션 툴 / MMP와 같은 서드파티 트래킹 툴을 활용하여 데이터를 수집하고 한번에 통일된 기준으로 확인할 수 있습니다.
> 👉 모바일 앱 어트리뷰션 툴을 지칭하는 여러 가지 이름과 설명은,[ 어트리뷰션 툴을 가리키는 여러 가지 용어들 - MMP, 서드파티 트래커, 앱 트래커](https://www.airbridge.io/ko/blog/attribution-tool-mmp) 포스트에서 확인해보세요.
모바일 앱 어트리뷰션 툴을 활용하게 되면, 한정된 개발 리소스 속에서 직접 데이터를 모으고 분석하지 않아도 하나의 SDK로 데이터를 수집하여 분석할 수 있습니다. 게다가 광고 프로드를 방지하거나 SKAdNetwork와 같은 새롭게 등장하는 프레임워크 등에 효과적으로 대응할 수 있다는 장점이 있습니다.
## 모바일 앱 어트리뷰션을 위해 필요한 데이터
이 모바일 앱 어트리뷰션 툴이 어트리뷰션을 하기 위해서는 **터치포인트**와 **전환** 데이터가 필요합니다. 유저의 여정 속 터치포인트와 전환 데이터를 수집, 처리, 가공하는 과정을 거쳐 광고주/마케터에게 어트리뷰션 완료된 데이터를 제공합니다.
[]
### 터치포인트의 의미와 데이터 수집 경로
터치포인트(Touchpoint)란 광고가 게재된 곳에서 발생한 클릭(Click), 조회(Impression) 등을 의미합니다. 이 터치포인트 데이터를 수집하는 방법은 다음과 같습니다.
- **트래킹 링크(Tracking Link)** - 트래킹 링크는 앱 어트리뷰션 툴에서 제공하는 링크입니다. 이 링크를 통해 앱 어트리뷰션 툴이 대신 터치포인트를 수집할 수 있습니다.
- **서버 API(Server API)** - 일부 광고주 고객사 및 매체는 서버를 통해 터치포인트 및 전환 데이터를 전달합니다. 특히 매체 중, SAN(Self-Attributing Network)의 경우, 자체 트래픽에 대해 앱 설치 및 인앱 이벤트의 어트리뷰션을 독립적으로 수행합니다. 그리고 직접 수집한 수많은 터치포인트 중 전환을 일으킨 기기(또는 유저)의 마지막 터치포인트만 연동된 측정 파트너에게 전해줍니다.
- **리퍼러(Referrer)** - 구글 플레이(Google Play)에서만 지원하는 기능으로, 별도 셋팅이 필요 없이 광고 클릭 시 고유 클릭 ID가 생성됩니다. 이 ID를 통해 해당 유저에 대한 정보 및 광고 클릭 후 발생한 앱 설치 및 인앱 이벤트 등에 대한 데이터를 확인할 수 있습니다.
### 전환의 의미와 데이터 수집 방법
전환(Conversion)은 보통 앱 어트리뷰션 툴이 수집합니다. 앱 설치(Install), 구매(Purchase), 장바구니에 담기(Add to Cart) 등으로 광고주들의 서비스 내에서 발생한 이벤트들입니다.
- **앱 어트리뷰션 툴이 제공하는 SDK(Software Development Kit)** - 소프트웨어 개발자가 특정 운영체제용 응용프로그램을 만들 수 있게 해주는 소스(Source)와 도구 패키지를 의미합니다. 앱 어트리뷰션 툴들은 SDK를 제공하며, 각 모바일 앱에 해당 SDK를 설치함으로써 전환 데이터를 수집합니다.
- **API** - 터치포인트를 수집하는 것과 같이, 일부 광고주 고객사 및 매체는 서버를 통해 전환 데이터를 전달하는 경우가 있습니다.
## 수집한 터치포인트 및 전환 데이터를 어트리뷰션하는 방법
앞서 설명한 터치포인트와 전환 데이터는 무작위로 흩어져서 여러 데이터베이스에서 수집하고 있습니다. 이를 처리하고 가공하는 과정을 거쳐야 어트리뷰션이 완료됩니다.
### 어트리뷰션을 위한 데이터의 처리
이를 분석하기 위해서는 ‘식별자 매칭(Identity Matching)’ 과정이 필요합니다. 이 매칭에는 확률적 방법과 확정적 방법 두 가지가 존재합니다.
**확정적 방법**은 구글과 애플이 디바이스 단위로 제공하는 고유의 광고 식별값을 통해 매칭하는 방법을 의미합니다. 이때, **광고 식별값(ADID)**은 광고주가 특정 기기(유저)를 추적할 때 사용하는 ID(식별자)를 의미하며, 구글은 GAID, 애플은 IDFA라고 지칭합니다. 최근 애플의 14.5 업데이트 이후에는 각 유저에게 ATT 동의를 받은 경우에 한해서만 IDFA 수집이 가능합니다.
**확률적 방법**은 OS 및 IP 주소 등 수집한 데이터의 여러 값을 싱크해서 고유값을 만들고, 이에 대한 키값을 가지는 유저들을 같은 값이라고 간주함으로써 매칭하는 방법을 의미합니다.
### 어트리뷰션 성과 분석을 위해 필요한 기준, 어트리뷰션 모델
위와 같이 데이터를 처리하고 가공하는 과정에서 전환이 일어나기 전 발생한 여러 터치포인트들에 각각 어느 정도의 기여를 분배할 것인지 결정해야 합니다. 즉, ‘어트리뷰션 모델’을 기반으로 어떤 터치포인트가 어떤 전환 이벤트에 기여했는지 확인해야 합니다. 이 어트리뷰션 모델에는 규칙 기반(Rule-based) 모델인 싱글 터치 어트리뷰션 모델과 멀티 터치 어트리뷰션 모델, 그리고 데이터 기반(Data-driven) 모델 등이 존재합니다.
> 👉 마케팅 어트리뷰션 모델에 대해서는 [마케팅 어트리뷰션 모델(Marketing Attribution Model) 알아보기](https://www.airbridge.io/ko/blog/what-is-marketing-attribution-model) 포스트에서 확인해보세요.
어트리뷰션을 완료하여 데이터를 제공하는 앱 어트리뷰션 툴들은 일반적으로 라스트 터치 어트리뷰션(LTA, Last Touch Attribution) 모델을 기준으로 어트리뷰션을 합니다. LTA 모델은 전환 직전 마지막 단 1개의 터치포인트에 모든 성과를 부여하는 모델로, 복잡할 수 있는 기여도 분배 과정을 휴리스틱하게(단순하고 빠르게) 해결할 수 있다는 장점이 있습니다. 이 LTA 모델을 기반으로 어트리뷰션을 완료한 데이터를 앱 어트리뷰션 툴의 대시보드에서 확인할 수 있습니다. 다만, 각 앱 어트리뷰션 툴마다 세부적인 조건과 우선 순위, 규칙은 다를 수 있기 때문에 같은 매체에서 캠페인을 집행하여 발생한 전환이라고 하더라도 그 성과는 조금씩 다를 수 있다는 것을 기억해두면 좋습니다.
## 모바일 어트리뷰션으로 단지 측정 결과를 확인하는 것을 넘어, 측정을 통한 마케팅 성장으로 이어질 수 있어야 합니다.
이번 포스트에서는 모바일 어트리뷰션의 의미부터 필요성, 그리고 앱 어트리뷰션 툴이 데이터를 수집, 처리, 가공하는 과정에 대한 기본적인 내용을 소개하였습니다. 이렇게 모바일 앱 어트리뷰션이 완료된 결과를 통해 확인할 수 있는 것은 과거의 데이터 및 성과입니다. 지난 성과를 분석하고 그 요인을 명확하게 파악해야 다음 전략에도 반영할 수 있습니다.
이때, 성과가 잘 나온 전략이라고 해서 미래에도 계속해서 좋은 성과로 이어진다는 보장은 없다는 점을 염두해야 합니다. 동일한 채널에서 동일한 캠페인을 지속할 경우, 같은 오디언스와 인게이지하는 속에 새로운 타겟군에게 다가갈 수 있는 기회를 놓칠 수도 있기 때문입니다.
보다 지속적인 마케팅 성장으로 이어지기 위해서는 과거 성과 데이터를 확인하는 것을 넘어서 다음 액션 아이템을 선정하고 실행으로 이어질 수 있어야 합니다. 이러한 마케터의 고민에 앞장서서 에어브릿지와 같은 모바일 앱 어트리뷰션 툴은 다양한 마케팅 성과 측정 방법을 모색하고 새로운 솔루션을 제공하고 있습니다.
이제 모바일 앱 어트리뷰션 툴도 단지 데이터를 분석한 측정 결과를 제공하는 것에서 머물러서는 안됩니다. 다양한 마케팅 성과 측정 방법을 모색하고, 트렌드 및 업계 변화의 흐름 속 적합한 솔루션을 제공할 수 있어야 합니다. 에어브릿지도 측정을 통한 마케팅의 성장으로 이어질 수 있도록 마케터의 고민에 앞장서서 마케팅 측정의 비전과 구체적인 방향성을 제안해가겠습니다.
> 👉 에어브릿지가 바라보는 모바일 앱 어트리뷰션의 미래가 궁금하다면, [에어브릿지는 LTA + MTA + MMM 통합 측정 스택 제공과 함께 Marketer-centric으로 나아갑니다.](https://www.airbridge.io/ko/blog/airbridge-unified-measurement-stack)포스트에서 자세한 내용을 확인해보세요.
---
## 마케팅 어트리뷰션 모델이란? 종류별 특징 알아보기
URL: https://www.airbridge.io/ko/blog/what-is-marketing-attribution-model
Language: Korean
Published: 2022-10-06
지금 우리 앱을 다운로드 한 유저, 혹은 지금 그 상품을 구매한 유저가 어떻게 유입되어 전환되었는지 궁금하시나요? 혹은 어떤 채널에서 광고를 집행해야 전환율을 가장 효과적으로 개선할 수 있을까요?
이에 대한 답을 찾아가는 데에 있어서 가장 중요한 것은 명확한 기준을 바탕으로 각 매체가 전환에 얼마만큼의 기여를 했는지 분석하고 가장 효과적인 매체를 찾는 겁니다. 이때 그 기준으로서의 역할을 하는 것이 바로 ‘어트리뷰션 모델(Attribution Model, 기여 모델)’입니다.
이번 포스트를 통해 마케팅을 한다면 반드시 필요한 성과 측정 및 분배를 위한 ‘어트리뷰션 모델’에 대해 이해하고, 우리 프로덕트에 적합한 어트리뷰션 모델을 채택해보세요.
## 어트리뷰션 모델(Attribution model, 기여 모델)이란?
‘어트리뷰션(attribution)’은 한국어로 번역하면 ‘기여’를 의미합니다. 어트리뷰션 모델은 유저의 전환(Conversion) 전, 어떤 터치포인트가 기여했는지 확인하는 방법론입니다. 여기서 ‘전환’은 앱 설치 그리고 결제 등과 같은 인앱 이벤트를 의미하며, ‘터치포인트’는 유저 여정(User Journey)에서 소셜 미디어 광고 캠페인 및 인앱 광고 등의 광고를 클릭하거나 조회하는 행동을 의미합니다. 다시 말해, 어떤 광고가 유저의 전환에 영향을 미쳤는지 파악하기 위해 어트리뷰션 모델을 활용하는 겁니다.
[]
특히 Paid 마케팅을 집행했다면 어떤 캠페인이 가장 효과적이었고 목표 성과 달성에 기여했는지, 어떤 매체에서 유입된 유저가 가장 전환율이 높은지 등을 분석하고 마케팅 효율을 개선해야 합니다. 마케팅 예산은 한정적이고, 한정된 예산 속에서 최대의 효과를 발생시켜야하기 때문이죠. 게다가 유저의 구매까지의 여정이 복잡해진 만큼 다양해진 마케팅 채널과 매체에서 가장 효과적인 미디어 믹스(media mix)를 구성하는 것이 중요합니다.
이때 마케팅 성과를 제대로 측정하기 위해서는 적절한 ‘어트리뷰션 모델(Attribution Model, 기여 모델)’을 적용하여, 전환이 일어나기 전 발생한 터치포인트들에 각각 어느 정도의 기여를 분배할 것인지 결정해야 합니다. 하지만 모든 프로덕트에 공통적으로 적용되는 정답과 같은 어트리뷰션 모델은 존재하지 않기 때문에, 각 서비스 및 프로덕트 별로 적합한 어트리뷰션 모델을 적용해야 합니다.
이를 위해 우선적으로 필요한 것은 다양한 어트리뷰션 모델을 이해하는 겁니다. 주요 어트리뷰션 모델은 크게 규칙 기반(Rule-Based) 기여 모델인 싱글 터치 어트리뷰션 모델(Single-touch Attribution Model)과 멀티 터치 어트리뷰션 모델(Multi-touch Attribution Model), 그리고 데이터 기반(Data-Driven) 기여 모델로 나눌 수 있습니다.
## 싱글 터치 어트리뷰션 모델(Single-touch Attribution Model)
‘싱글 터치 어트리뷰션 모델’은 단 1개의 마케팅 채널에 모든 성과를 인정해주는 모델로, 퍼스트 터치 어트리뷰션(First-touch attribution)과 라스트 터치 어트리뷰션(Last-touch attribution) 두 가지 모델이 있습니다. 전환까지의 여정에 얼마나 많은 터치포인트가 있더라도 상관 없이, 하나의 터치포인트 기여만 인정한다는 것을 의미합니다. 가장 처음 발생한 터치포인트에만 모든 기여를 인정하고 싶다면 ‘퍼스트 터치 어트리뷰션 모델’을 적용하고, 반대로, 가장 마지막에 발생한 터치포인트에만 모든 기여를 인정하고 싶다면 ‘라스트 터치 어트리뷰션 모델’을 적용하면 됩니다.
### 퍼스트 터치 어트리뷰션 (First-touch attribution)
퍼스트 터치 어트리뷰션 모델은, 가장 먼저 발생한 터치 포인트에 모든 기여를 부여하는 모델입니다. 이 모델은 특히, 우리 제품 및 서비스를 우선 많은 유저에게 알리는 것이 중요한 때에 활용하면 좋습니다. 앱 설치 및 구매로 이어지기 위해서는 우리 제품 및 서비스를 알고 있어야만 하기 때문에 첫 인지가 중요합니다. 이 모델을 활용하면 잠재 고객에게 도달하는 데에 효과적인 마케팅 채널을 찾는 데에 도움이 될 수 있습니다. 가장 먼저 우리 제품 및 서비스를 노출하여 조회 혹은 클릭 하도록 만든 매체에게 모든 기여를 부여하기 때문에, 가장 많은 기여를 획득한 채널에 더 많은 캠페인을 집중하면 더 많은 잠재 고객에게 도달할 수 있습니다.
이 어트리뷰션 모델을 적용하는 경우, 주로 TOFU(Top of Funnel) 마케팅 활동에 집중하게 될 가능성이 높습니다. 브랜드 인지도를 높이고, 더 많은 잠재 고객에게 다가감으로써 구매를 더 많이 발생시키고자 노력할 겁니다. 이후의 유저 여정을 트래킹할 필요 없이, 첫 번째 터치포인트만 파악하면 되니 비교적 성과를 분석하고 추적하기에 수월할 수 있습니다.
그렇지만, 단순히 첫 번째 터치포인트에만 모든 기여를 집중하는 것에는 한계가 있을 수밖에 없습니다. 전체 유저 여정을 고려하지 않는 모델이며, TOFU 마케팅 활동을 지나치게 강조하기 때문입니다. 단지 TOFU에만 집중하게 되면, 이후에 푸시 알림이나 이메일 등의 CRM 혹은 리타겟팅 캠페인 등을 통해 전환될 가능성이 높은 유저들을 그냥 놓치게 되어버릴 수 있죠. 아무리 많은 잠재 고객을 불러모았다고 하더라도, 이후의 추가적인 마케팅 활동이 없다면 유저를 전환하게 만들 가능성이 줄어들게 됩니다. 또한, 자동차 등 고관여 제품이라서 유저의 여정이 긴 경우나 해외 여행 등 서비스 사용 텀이 길 수밖에 없는 서비스의 경우, 설정한 기여 기간을 벗어나 첫 번째 터치포인트 데이터를 잃어버려서 성과를 측정할 수 없는 상황이 발생할 수도 있습니다.
> 💡 기여 기간: 기여 모델을 적용시키는 기간으로, 룩백 윈도우(Lookback Window) 및 어트리뷰션 윈도우(Attribution Window)라고 표현합니다. 서비스 또는 채널별 적합한 기여 기간을 설정해야 합니다.
### 라스트 터치 어트리뷰션 (Last-touch attribution)
라스트 터치 어트리뷰션 모델은 가장 일반적으로 활용되는 모델로, 전환 직전 마지막 단 1개의 터치포인트에 모든 성과를 부여합니다. 이미 브랜드 인지도 측면은 갖추어져 있으나, 잠재고객의 ‘전환’에 초점을 맞추고 있는 제품 및 서비스에서 활용하면 좋습니다. 또한 비교적 긴 판매 사이클을 가지고 있거나 멀티 터치 어트리뷰션 모델을 사용하기 어려운 경우에는 라스트 터치 포인트 모델로 단순화하여 성과를 측정할 수 있습니다. 복잡한 어트리뷰션 모델을 단순화하여 진행할 수 있기 때문에 적은 리소스로도 꽤 합리적인 성과를 측정할 수 있습니다. 매체와의 광고비 정산 기준을 협의할 때 주로 활용되는 모델인 만큼, 이 모델을 활용하기만 해도 높은 전환율을 보이는 매체를 파악하여 마케팅 예산을 조정하도록 도와줄 수 있습니다.
하지만, 라스트 터치 어트리뷰션 모델은 오직 마지막 터치포인트에만 모든 기여가 집중이 되기 때문에, 이전에 발생한 터치포인트들은 무시된다는 한계가 발생합니다. 원래 구매할 의도를 가지고 있었거나, 브랜드 인지도 등으로 인한 영향력은 분석할 수 없게 됩니다. Paid 광고에 대한 터치 포인트가 하나라도 있었다면 해당 유저의 전환에 대한 기여도의 100%를 마지막 터치 포인트가 가져가게 되죠. 그렇기 때문에, 유저 전환에 영향을 미친 다른 터치 포인트들의 성과에 대해 제대로 분석할 수 없습니다.
## 멀티 터치 어트리뷰션 모델(Multi-touch attribution model)
한편, 싱글 터치 어트리뷰션 모델의 한계를 보완할 수 있는 모델이 ‘멀티 터치 어트리뷰션 모델’입니다.
앞서 언급한 것처럼, 유저의 여정은 다양하고 복잡해졌습니다. 개인이 사용하는 기기도 늘어났고, 마주하는 플랫폼도 다양해졌습니다. 첫 터치 포인트부터 마지막 전환까지 이어지는 여정에서 다양한 터치 포인트들이 발생할 수밖에 없습니다. 이러한 상황에 대해 마케터들은 어떤 터치 포인트가 가장 영향력있는지 여러 터치 포인트들에 대해 성과를 분석하여 마케팅 믹스를 최적화해야 합니다. 즉, ‘멀티 터치 어트리뷰션’ 모델이 필요합니다.
‘멀티 터치 어트리뷰션 모델’이란, 유저 여정에서 발생한 각각의 터치포인트들에 대해 가치를 부여하는 모델입니다. 즉, 싱글 터치 어트리뷰션 모델처럼 한 개의 터치 포인트에만 모든 기여를 부여하는 것이 아닌, 최종 결과에 영향을 미친 많은 다양한 터치 포인트들에게 기여를 부여하는 겁니다. 이 터치 포인트들에는 Paid 광고 캠페인은 물론, 블로그 포스트 등의 Owned 매체까지도 포함될 수 있습니다. 멀티 터치 어트리뷰션은 발생한 여러 터치 포인트들을 놓치지 않고 분석하여, 고객 전환에 기여한 진짜 마케팅 성과를 측정하도록 도와줄 수 있습니다.
그렇다면 어떻게하면 여러 터치 포인트에 제대로 기여를 분배할 수 있을까요? 대표적인 온라인 광고 플랫폼 중 하나인 [Google Ads](https://ads.google.com/home/)에서 성과 분석을 위해 제공하고 있는 대표적인 멀티 터치 어트리뷰션 모델들은 다음과 같습니다.
### 선형 모델 (Linear)
해당 여정에서 발생한 모든 광고 매체에 동일한 기여도를 부여하는 모델입니다. 몇 개의 터치 포인트(광고 매체)가 존재하든지, 모든 터치 포인트데 동일한 가중치를 부여해서 마케팅 기여도를 평가합니다. 이는 광고주가 모든 마케팅 경로를 중요하게 여겨서 동일한 가중치를 부여하고 싶을 때 유용한 방식입니다. 다만, 각 매체별로 완전히 동일한 가중치를 부여하기 때문에, 특정 마케팅 캠페인의 경로가 상대적으로 길거나 몇몇 채널이 갖는 영향력이 누가 봐도 클 수 밖에 없는 경우에는 다소 비현실적인 어트리뷰션 방법이라고 할 수 있습니다.
### 시간 가치 모델 (Time Decay)
전환 시점을 기준으로 했을 때, 가장 최근에 유입된 채널에 가장 많은 기여도를 부여합니다. 전환 시점에 가까운 채널일수록 어트리뷰션 가중치가 증가합니다. 고관여 제품이거나 구매 주기가 긴 제품이나 서비스인 경우에 유용한 방식일 수 있습니다. 다만, 이 방식은 유입 경로의 앞 부분에 위치한 매체가 더 많은 유저를 획득할 수 있는 채널이라면 향후 더 많은 고객을 확보할 수 있는 기회를 상실할 수도 있습니다.
### 시점 기반 (Position-based)
Google Ads에서는 첫 번째와 마지막 터치포인트에 각각 40%의 기여도를 부여하고, 나머지 터치포인트에 대해 나머지 20% 기여도를 분배하는 방식입니다. U-Shape 모델이라고도 할 수 있습니다. 경로 상의 처음과 끝, 즉 특정 제품에 대한 첫 인지와 마지막 전환에 직접적으로 기여한 채널을 높게 평가하면서도, 중간 지점에 위치하는 매체들에게도 어느 정도의 기여도를 분배해줄 수 있습니다. 다만, 마케팅 경로가 짧은 경우에는 크게 효과적이진 않으며, 다양한 마케팅 채널들의 특징을 판단하기 어려운 경우엔 사용하기 쉽지 않은 모델입니다.
## 데이터 기반 어트리뷰션 모델 (Data-driven Attribution Model)
데이터 기반 어트리뷰션 모델은 전환이 발생하기 전, 유저의 광고 조회 혹은 클릭 등의 데이터를 기준으로 전환 기여도를 분배하는 방식입니다. 전환 경로에서 각 터치포인트들의 실제 기여도를 계산하기 위해 실제 데이터를 사용한다는 점이 규칙 기반 모델들과의 차별점입니다. 이 분석을 통해 계정의 ‘데이터’를 사용하여 비즈니스 목표에 가장 큰 영향을 미치는 키워드, 광고, 캠페인을 찾을 수 있습니다. 데이터 기반 기여도 분석에는 크게 다음의 두 가지의 방법이 사용됩니다.
- 섀플리 밸류(Shapley Value)는 발생 가능한 모든 조합을 통해, 전환 확률 모델을 만들어서 기여도를 분석합니다. 즉, 랜덤으로 다양한 조합을 만들어서 발생할 수 있는 전환 성과를 비교해보는 겁니다. 전환된 유저와 전환되지 않은 유저의 데이터를 모두 포함한 데이터를 기반으로, 특정 마케팅 터치포인트의 존재와 시기가 유저의 전환 가능성에 어떤 영향을 미칠 수 있는지 분석합니다. 광고에 노출된 유저의 전환 확률과 노출되지 않은 유사한 유저의 전환 확률을 비교하여 유저가 특정 광고에 노출될 경우, 전환 경로의 특정 지점에서 전환할 가능성을 분석할 수 있습니다. Google Ads의 데이터 기반 기여 모델의 기초 이론으로 채택된 방법론으로, 특히 Google Ads의 데이터 기반 기여도 분석은 운영하고 있는 다양한 지면인 Google 검색(쇼핑 포함), YouTube, 디스플레이 광고를 통해 발생한 웹사이트, 매장 방문, Google 애널리틱스 전환을 확인할 수 있습니다.
- 마르코프 체인(Markov chain)은 마르코프 성질을 가진 이산 확률 과정으로, 다시 말해 특정 상태에서 다른 상태로 전이하는 확률을 다루는 이론입니다. 이를 마케팅 기여도 분석에 적용하면, 채널 간 전환 확률을 구하여 모델링을 진행하는 방식을 의미합니다. 구매 전환에 이르기까지는 특정 상품에 대한 관심부터 시작하여 여러 마케팅 채널을 거치게 됩니다. 이 과정을 모두 어떤 ‘상태’로 간주하여, 각 마케팅 채널이 구매 전환까지 이어지는 영향력을 측정하는 겁니다. 구체적인 방법으로는 전환 여정에서 특정 채널이 전환에 얼마나 기여했는지를 파악하기 위해, 특정 채널이 여정 상에 없을 때의 전환 확률을 계산하여 확인합니다.
데이터 기반 어트리뷰션 모델은 기존에 널리 사용되고 있는 규칙 기반 어트리뷰션 모델과는 다르게, 실제 수치 데이터를 바탕으로 성과를 확인할 수 있기 때문에, 규칙 기반 어트리뷰션 모델의 한계를 보완할 수 있다는 장점이 있습니다. 그렇기 때문에 Google Ads에서도 대부분 전환에 대해 기본 기여 분석 모델로, 데이터 기반 어트리뷰션 분석을 사용하고 있습니다. 데이터 기반 접근 방식을 사용하면 추측이 아닌 실제 데이터를 바탕으로 적합한 모델을 선택할 수 있으며, 이를 통해 매체 운영을 위한 계획 수립에 도움을 받을 수 있습니다.
## 우리 프로덕트 및 서비스에 적합한 마케팅 어트리뷰션 모델을 채택하고 진짜 마케팅 성과를 분석해보세요.
모든 프로덕트 및 서비스에 정답으로 적용할 수 있는 마케팅 어트리뷰션 모델은 존재하지 않습니다. 진짜 마케팅 성과 분석을 위해서는 우리 프로덕트 및 서비스에 맞는 어트리뷰션 모델을 찾아 선택하는 것부터 시작해야 합니다. 이를 위해서는 먼저, 어트리뷰션 모델들에 대해 알아보고 특성을 살펴보고 이해합니다. 그리고 우리 프로덕트 및 서비스가 가진 특징 및 놓인 상황에 따라 적절한 어트리뷰션 모델을 적용합니다.
이때, 반드시 앞서 소개한 어트리뷰션 모델 중 하나로만 선택해야 하는 것은 아닙니다. 커스텀하게 조정할 수도 있고, 사용하고 있는 MMP의 어트리뷰션 모델을 하나의 기준으로 삼을 수도 있죠. 또는 여러 어트리뷰션 모델을 기반으로 마케팅 성과를 측정하고 분석해보며 최적의 모델을 찾아가는 것도 좋습니다. 더 나아가 모바일과 웹 뿐만 아니라 TV, 옥외 광고 등 여러 채널에 걸친 대규모 마케팅 기획 및 장기간 전략을 수립하기 위해 [‘마케팅 믹스 모델링(Marketing Mix Modeling)’](https://www.airbridge.io/ko/blog/marketing-mix-modeling)과 같은 보다 거시적인 모델이 필요할 수도 있습니다.
> 👉 어트리뷰션 모델에 대한 보다 자세한 내용은 [에어브릿지 핸드북 01 - 어트리뷰션](https://www.airbridge.io/ko/library/airbridge-handbook-attribution)에서 확인해보세요.
마케팅 성과 최적화는 진짜 성과 분석에서부터 시작해야 합니다. 진짜 마케팅 성과를 분석하기 위한 방법, 에어브릿지의 어트리뷰션 전문가들과 함께 알아보세요.
---
## 에어브릿지는 LTA + MTA + MMM 통합 측정 스택 제공과 함께 Marketer-centric으로 나아갑니다
URL: https://www.airbridge.io/ko/blog/airbridge-unified-measurement-stack
Language: Korean
Published: 2022-09-28
포브스가 선정한 탑 10 마케터 중 한 사람인 닐 파텔(Neil Patel)은 [라스트 터치 어트리뷰션(Last Touch Attribution)은 거짓말을 하고 있다](https://neilpatel.com/blog/last-touch-attribution-lies/)고 이야기합니다. 대부분의 마케터들이 알고 있듯이, 라스트 터치 어트리뷰션 모델은 정확한 성과 측정 방법이라고 할 수 없지만 적은 리소스로 합리적인 성과 분석을 위한 타협점으로서 주류 마케팅 성과 분석 모델로 활용되고 있습니다. 하지만 정확한 마케팅 성과 분석에 대한 니즈는 지속적으로 존재했습니다. 이와 더불어, Apple의 iOS 14.5 업데이트와 쿠키 수집의 제한, GDPR과 CCPA 등으로 대표되는 개인정보보호 강화로 인해 계속해서 유저 레벨의 데이터 수집을 제한받고 있습니다. 마케팅 전략을 수립하여 마케팅 비용을 집행했다면 그에 따른 효율과 성과를 분석하고, 이를 보다 나은 성과를 도출하기 위한 전략에 반영하는 것은 당연한 수순입니다. 그렇기 때문에 마케터들은 유저의 개인 정보를 보호하면서도 보다 정확한 마케팅 성과를 측정할 수 있는 방법을 찾아 헤매며 동시에 계속해서 다양한 챌린지를 마주하고 있습니다.
이번 포스트는 빠른 변화의 흐름 속에 챌린지를 마주한 마케터들을 위해 에어브릿지가 나아가고 있는 방향성을 공유합니다.
## 에어브릿지는 마케터들의 고민 해결을 위해 앞서가고 있습니다.
[]
### LTA의 한계와 MTA&Incrementality 제공
**LTA(Last Touch Attribution) 모델**은 대부분의 MMP/어트리뷰션 툴이 기반으로 하고 있는 마케팅 성과 측정 모델입니다. 복잡한 어트리뷰션 분석을 단순화하여 빠르게 진행할 수 있기 때문에 일반적으로 통용되고 있습니다. 전환을 발생시킨 채널을 확실하게 트래킹할 수 있으며 개별 캠페인이나 광고 소재의 성과를 빠르게 파악하여 최적화할 수 있다는 장점을 가지고 있습니다. 에어브릿지도 당연히 MMP로서 LTA에 기반한 마케팅 성과 분석 결과를 제공하고 있습니다. 다만, LTA는 가장 최근에 발생한 광고에 전환 이벤트에 모든 기여도를 인정하기 때문에 앞서서 유저에게 영향을 미친 다른 광고들의 성과와 기여도가 반영되지 않아 표면적인 분석 결과만 보여준다는 한계를 가지고 있습니다. 에어브릿지의 내부 분석 결과에 따르면 최소 30%의 유저가 2개 이상의 광고를 통해 전환 이벤트를 발생시킨 것으로 파악됩니다. 또한, 유저가 원래부터 가지고 있던 전환 이벤트에 대한 잠재적인 경향성을 고려하면, 마지막 광고가 100% 기여를 가져가기엔 마지막 광고에 대한 과대 성과 측정으로 이어질 수 있습니다.
이에 에어브릿지는 MTA(Multi Touch Attribution) & Incrementality 기능을 통해 이 한계를 보완할 수 있는 구체적인 솔루션을 제시합니다. MTA 분석으로 유저가 거친 여러 광고의 기여도를 보다 정확하게 이해할 수 있으며, 자체 Incrementality 모델을 통해 실제 광고로 인해 발생한 전환 성과, 즉 순수한 광고의 효과를 측정할 수 있게 도와줍니다. 구체적으로는 Propensity Score Matching (PSM) 분석 방법을 통해 광고와 전환 성과 간 실제로 발생한 증분 효과를 측정하며, 통계적 기법을 통해 실제로 광고가 순수하게 미치는 증분 효과를 파악합니다. 그리고 매일 업데이트되는 에어브릿지 대시보드의 Incrementality 리포트를 통해 매체별 증분 효과를 파악하고 효율적인 예산 분배를 위해 미디어 믹스에 바로 적용할 수 있습니다. 또한, 관측형(Observational) 방법론을 활용하기 때문에 실험형(Experimental) 방법에 비해 더 빠르고 유연하게 광고 효과의 Incrementatlity를 확인할 수 있다는 장점도 있습니다.
### 유저 레벨 어트리뷰션의 한계로 인한 개인 정보 보호에 앞장선 MMM 기능 제공
한편 개인정보보호 강화 트렌드 속에서 유저 레벨의 어트리뷰션이 어려워지는 환경이 만들어지고 있습니다. 이에, 에어브릿지는 유저 레벨의 데이터 없이도 마케팅 성과를 측정할 수 있도록 MMM(Marketing Mix Modeling)을 대안으로 제시하고자 합니다.
**MMM**은 거시적인 데이터 기반의 마케팅 활동과 영업 성과 사이의 관계를 파악하는 분석 방법으로, 크게 빈도주의(Frequentist) 또는 Bayesian 접근법으로 구분됩니다. 이러한 MMM은 최근에 새롭게 등장한 분석 방법은 아닙니다. 과거부터 대규모 마케팅 예산을 집행하던 비교적 큰 기업들에서 미디어 믹스를 결정하는 데에 사용하곤 했습니다. 다만, 전통적인 MMM 서비스의 경우 적어도 6개월에서 1년 치의 내부 데이터를 기반으로 장기간에 걸쳐 상당한 비용을 들이는 컨설팅 프로젝트의 형태로 MMM 분석이 진행되어 왔기 때문에 제한적인 접근성을 가지고 있었습니다.
하지만 개인 단위 데이터에 의존하던 기존의 마케팅 성과 분석 방법론들이 프라이버시 이슈로 인해 힘을 잃게 되면서 최근 다시 주목을 받고 있습니다. MMM은 유저 레벨의 데이터가 아닌 집계된(aggregated) 데이터만으로도 모델링 기반의 분석이 가능하기 때문입니다. 이로 인해 LTA 및 MTA 결과와 결합하여 iOS 14.5 업데이트 이후로 성과 측정이 어려워진 iOS에 대해서도 성과 추정에 활용할 수 있으며, CTV나 DOOH와 같은 비디지털 채널들의 성과 측정에도 활용할 수 있습니다. 또한 MMM 모델의 다른 장점은 예측과 시뮬레이션이 가능하다는 점입니다. 예를 들어서 다음 1개월의 매체별 성과를 예측하거나, 현재 예산 안에서 어떻게 재분배를 했을 시 가장 매출이 극대화될 수 있을지 등을 처방할 수 있습니다. 이러한 MMM의 기능은 성과 측정의 관점을 회고적 관점(retrospective)에서 실질적인 최적화 액션 아이템을 제시하는 처방적 관점(prescriptive)으로 발전시키는 것에 기여할 수 있습니다. 이러한 트렌드와 더불어, 에어브릿지와 같이 보다 좋은 접근성을 기반으로 마케터들이 실질적으로 MMM을 활용할 수 있도록 SaaS(Software as a Service) 프로덕트 형태의 현대적인 MMM 솔루션 제공사들도 생겨나고 있습니다.
특히 에어브릿지는 Meta의 MMM SaaS Incubator Program에 참여하여 Frequentist 모델을 활용한 다양한 기능 개발을 진행하고 있습니다. 에어브릿지의 MMM 솔루션은 자체 머신러닝 엔진을 통해 기존에 수집 중인 데이터를 활용하여 채널별 집행된 광고의 성과를 측정하고 향후 예산 효율을 극대화할 수 있는 방법을 제안합니다. 구체적으로는 *Marketing Mix Analytics 리포트*를 통해 학습된 에어브릿지 MMM 모델이 학습한 결과를 유저에게 리포트 형태로 제공할 예정입니다. 또한, *Budget Optimization 리포트*를 통해 학습된 머신러닝 모델에 의해 정해진 미래 예산으로 최대 성과가 달성 가능한 예산안을 제안하는 기능을 제공할 예정입니다.
다만 MMM도 마찬가지로 완벽한 마케팅 성과 측정 모델은 아닙니다. MMM은 과거 비즈니스 결과 데이터에 대한 마케팅 영향력을 ‘확률’에 기반해서 미래 성과를 예측하는 것이기 때문에 수치의 정확도에 대한 의문이 존재합니다. 에어브릿지는 이에 대한 문제를 주기적인 Calibration 작업을 통해 풀어나가려고 합니다.
### LTA + MTA+ MMM으로 완성되는 Unified Measurement Stack
[]
에어브릿지는 이처럼 LTA 모델의 한계를 극복하고자 하는 광고주/마케터의 니즈, 그리고 플랫폼과 가이드라인의 변화로 인한 환경 변화 속에서 마케터의 고민을 해결해 가기 위해 앞서서 연구하고 기능을 제공하고자 노력하고 있습니다. MMP로서 LTA 기반의 광고 성과 분석을 기본으로 제공하며, MTA 솔루션을 통해 보다 정확한 그리고 순수한 광고 효과를 측정하도록 돕고, 나아가 MMM으로 유저 레벨 분석이 어려운 환경에서 마케팅 성과 측정이 가능한 **통합된 측정 스택(Unified Measurement Stack)**을 제안합니다. 에어브릿지의 다양한 솔루션을 활용하여 개별 솔루션의 단점은 보완하고 장점은 취하는 완전한 마케팅 성과 분석을 제공하고자 합니다.
그리고 단지 다양한 솔루션을 제공하는 것에서 더 나아가 마케터의 다음 걱정을 함께 고민하고 있습니다. 각각의 솔루션이 분석하여 제공하는 성과를 어떻게 하면 제대로 활용할 수 있을지, 그리고 이용자의 관점에서 업무에 유연하게 적용할 수 있는 진정한 통합 측정 스택으로 어떻게 거듭날 것인지를 고민하고 있습니다.
## 우리 서비스 및 프로덕트에 가장 적합한 솔루션을 적용해야 진정한 마케팅 성과로 이어질 수 있습니다.
지금까지는 마케터의 고민을 해결하기 위해 에어브릿지가 Unified Measurement Stack을 갖춰가는 과정에 대해 알아보았습니다. 사실 이 세 가지 솔루션은 각기 다른 데이터 및 모델을 기반으로 성과를 분석하기 때문에 수치상 완전히 똑같은 결과를 도출하는 것은 불가능에 가깝습니다. 또한, 솔루션이 다양화되고 선택지가 늘어나게 되면서 각 솔루션을 언제 어떻게 활용하면 좋을지에 대한 혼란이 발생하기도 합니다. 이에 이번에는 다양한 솔루션을 우리 서비스 및 프로덕트에 어떻게 적용해야 할지 살펴보겠습니다.
어떤 때에 어떤 솔루션을 활용해야 한다는 정답이 있는 것은 아닙니다. 각 서비스 및 프로덕트가 놓인 상황이 다르기 때문입니다. 비즈니스의 규모나 업계 상황 등을 고려하여 가장 필요한 솔루션을 선택하고 자신의 서비스 및 프로덕트에 맞는 하나의 통합된 측정 스택을 찾아가야 합니다. 다만 그럼에도 불구하고, 솔루션을 공급자로서 단순하게 정답은 없다고 결론지어 버리는 것은 마케터에게 다시 고민을 넘기는 무책임한 행위라고 생각합니다.
에어브릿지는 마케터의 고민에 앞장서서 해결 방안을 찾아가고 있는 만큼, 마케터의 Full-cycle 여정 즉 마케팅 전략을 수립하고 광고 효율을 모니터링하고 고도화된 전략의 실행까지의 흐름에 따른 각 솔루션의 활용을 아래와 같이 제안합니다. 아래 내용을 참고하여 여러분의 서비스 및 프로덕트에 맞는 진정한 의미의 Unified Measurement Stack을 구성해 보는 데에 도움이 되기 바랍니다. 에어브릿지가 정답이 되어드릴 수 있습니다.
[]
### 마케팅 전략 설정 단계에 효과적인 MMM(Marketing Mix Modeling)
MMM은 다양한 종류의 집계된 데이터를 활용하는 시계열의 회귀 분석 방법이라는 것이 가장 큰 특징입니다. 그렇기 때문에 분석 결과를 미시적으로 확인하는 것은 어려우나 전반적인 추세나 앞으로의 방향성을 파악하는 데에는 매우 용이합니다. 또한, 다양한 종류의 집계된 데이터에는 터치포인트 데이터(광고 클릭이나 조회) 수집이 불가능한 TV 광고와 잡지 지면 등의 오프라인 매체에 대해서도 성과를 파악할 수 있다는 장점도 가지고 있습니다.
위와 같은 MMM의 장점과 특징을 살펴보았을 때, **MMM은 마케팅 전략 설정 단계에 활용**하는 것이 좋습니다. MMM은 거시적인 맥락에서 non-digital 채널을 포함한 전체 마케팅 채널의 트렌드 및 미래 전략 방향성을 도출해 줄 수 있기 때문입니다. 이때, 에어브릿지에서 제공할 기능인 Budget Optimization 리포트를 활용하면 미래 최적의 예산 분배 안을 확인할 수 있습니다.
### 광고 효율 모니터링 단계에 효과적인 MTA(Multi Touch Attribution)&Incrementality
Incrementality 솔루션을 통해서는 멀티 채널에 대한 성과 분배 및 실제 광고 효과에 해당하는 ‘증분(incrementality)’ 결과를 확인할 수 있습니다. 전체적인 트렌드를 보기엔 어렵지만, 개별 매체의 실질적인 효율을 확인하기에 가장 적합한 방법입니다.
그렇기 때문에 MTA&Incrementality 솔루션은 **광고의 성과와 효율을 확인하고 검증하는 모니터링 단계에 사용**하는 것이 좋습니다. 매체와의 정산과 무관하게 실제로 각 서비스 및 프로덕트의 마케팅에 진짜 도움이 되는 매체를 파악할 수 있는 분석 방법이기 때문입니다. 이를 통해 디지털 매체 간 효율 비교를 통해 가장 좋은 효율이 좋은 매체를 파악하고, 해당 매체에 더 많은 예산을 분배하는 식으로 성과를 개선해갈 수 있습니다.
### 마케팅 전략의 고도화 및 실행 단계에 효과적인 LTA(Last Touch Attribution)
LTA의 경우 가장 보편적으로 활용되고 있는 성과 측정 방법이자, Raw Data를 활용하는 만큼 가장 세밀한 성과 측정이 가능합니다. 그렇기 때문에 **마케팅 전략을 고도화하고 세부 캠페인과 소재를 최적화하는 단계에 활용**하는 것이 좋습니다.
LTA는 MTA에 비해 광고 매체들에 대한 전반적인 효과를 확인하기엔 한계를 가지고 있습니다. 다만, MTA 솔루션을 통해 확인한 광고 효과가 높은 채널 내에서 실제로 어떤 소재나 캠페인의 성과가 높은지에 대한 세부적인 전략을 수립할 때 활용하기 좋습니다. 예를 들어, 에어브릿지의 Actuals 리포트를 통해 소재별 및 캠페인별 광고 성과를 확인하여 다음 캠페인 기획을 위한 인사이트를 얻을 수 있습니다.
## Marketer-centric, 에어브릿지는 오늘날 마케터가 실제 일하는 환경에 맞는/필요한 솔루션 제공을 지향합니다.
앞서 언급한 것처럼 에어브릿지는 이용자 즉 마케터의 관점에서 진정한 통합 측정 스택으로 어떻게 거듭날 것인지를 고민하고 있습니다. 마케팅 전략의 수립부터 실행 단계까지의 Full-Cycle에 함께할 수 있는 솔루션을 제공하는 것만으로 마케터가 바라는 Unified Measurement Stack이라고 결론짓지 않습니다. 여기서 더 나아가 실제 활용을 하는, **마케터의 실제 업무 환경과 통합된 측정 솔루션**을 지향하고 있습니다. 즉, ‘Marketer-centric’으로, 오늘날 마케터가 실제 일하는 환경에 맞고 필요한 유연한 솔루션으로 개발하는 것을 항상 중요시하고 있습니다. 그렇기 때문에 에어브릿지의 Unified Measurement Stack을 구성하는 주요 솔루션인 MMM과 MTA 기능은 다음과 같은 차별점을 가지고 실제 마케터에게 필요한 기능을 제공해가고 있습니다.
### 비용과 시간 투자 없이도 Incrementality를 산출할 수 있는 관측형 모델
Incrementality는 실험(Experimental)과 관측(Observational)이라는 두 가지 대표적인 방법론이 존재합니다. Incrementality 성과를 제대로 측정하기 위해서는 Meta 또는 Google에서 운영 중인 Lift Study와 같은 실험 방법을 활용하는 것이 좋습니다. 이 실험 방법은 랜덤하게 선정된 test와 control 그룹을 대상으로 광고 노출도를 통제하는 방식으로 실제 광고의 차이를 파악하는 방식으로 진행되기 때문에 상대적으로 긴 시간이 소요되며 실제 비용이 발생하게 됩니다. 반면, 에어브릿지 Incrementality는 직접 실험을 진행하지 않는 대신 과거 유저의 행동 데이터에 머신러닝과 통계학적 기법을 접목시켜 ‘관측형 모델’을 기반으로 Incrementality를 산출합니다. 즉, 비용과 긴 시간을 투자하지 않고 운영하며 매일 실험과 유사한 광고 효과의 Incrementality를 확인할 수 있습니다.
여기에 분기 혹은 반기에 따라서 관측형 모델의 결과가 실제 실험과 유사한지 검증 및 조정(Calibration)하는 과정을 거치면 보다 정확한 결과를 파악할 수 있습니다. 실제로 에어브릿지 고객사인 다노는 LTA, 에어브릿지의 Incrementality, Meta의 Lift Study 결과 간 비교 분석을 통해 Incrementality가 실제와 유사한 분석 결과를 제공한다는 사실을 파악한 사례가 있었습니다. 이를 통해 다노는 진행 중인 캠페인 전략에 확신을 가질 수 있었으며, 에어브릿지의 MTA 분석을 통해 빠르고 정확하게 성과를 분석하고 있습니다.
[👉다노의 케이스 스터디 더 자세히 알아보기](https://www.facebook.com/business/success/dano)
### 빠르게 변화하는 환경에서도 성과 분석과 다음 단계를 제안하는 MMM
MMM은 Full-Cycle 중 마케팅 전략 수립 단계에 활용하는 것이 좋지만, 마케팅 결과에 대한 평균을 결정하기 위해 과거의 데이터에 의존한 모델이기 때문에 최근에 발생한 채널의 변화나 비즈니스 변화에 대해서는 확인하기 어렵다는 한계를 가지고 있습니다. 하지만 오늘날의 마케팅 시장은 빠르게 변화하고 있으며 마케터들은 이 변화 속에서 빠르게 대응하여 효과적인 마케팅 전략을 기획해야 합니다.
전통적인 MMM 서비스 제공사와는 다르게, 에어브릿지는 신속하고 유연한 MMM 솔루션을 제공합니다. 에어브릿지의 MMM 솔루션을 활용해 마케터들은 각 채널별로 집행된 광고가 목표 성과에 얼마나 기여했는지 더 빠르고 손쉽게 이해할 수 있습니다. 특히 에어브릿지 MMM은 최소 3개월 치 데이터만으로도 머신 러닝 기반의 자체 MMM 모델 학습이 가능하며, Daily로 업데이트된 결과를 제공하여 빠른 변화의 대응과 미디어 믹스 조정이 가능합니다. 또한, 이미 집행된 마케팅 전략에 대한 효율을 파악하는 것에서 나아가 다음 단계인 미디어 믹스 조정에 가장 효과적인 성과를 낼 수 있는 방법을 제안해 준다는 점에서 조금 더 적극적으로 마케터에게 필요한 솔루션을 제공하는 방향으로 나아가고 있습니다.
## Next MMP는 단순히 솔루션 공급자가 아닌, 마케팅 여정에서 진정한 솔루션을 제공해야 합니다.
LTA의 한계와 개인정보보호 트렌드 속에서 다양한 마케팅 성과 측정을 위한 방법론과 솔루션들이 등장하고 있습니다. 다양해진 솔루션에 대한 마케터들의 선택의 폭이 넓어진 만큼 보다 효율적인 성과 측정을 위한 통합 측정 솔루션 스택에 대한 니즈도 증가하고 있습니다. 이에 대해 에어브릿지는 단순히 마케팅 성과 측정 솔루션을 제공하는 공급자로서, LTA + MTA + MMM의 솔루션을 모두 함께 제공하는 Unified Measurement Stack에 머물러있지는 않을 겁니다. 마케터의 완전한 마케팅 여정을 모두 함께하고 실제 업무 환경에 잘 녹아들 수 있는 유연한 구조로 서비스를 제공해갈 예정입니다. 또한, 단지 측정만을 위한 것이 아닌 측정을 통해 마케팅이 성공하는 과정에 일조해가고자 합니다. 앞으로도 에어브릿지는 마케터에게 진정 필요한 솔루션으로서, 마케터 관점에서 유용할 수 있는 다양한 사용 사례와 인사이트를 제공해가겠습니다.
---
## 릴리즈 노트: 2022년 7~8월
URL: https://www.airbridge.io/ko/blog/release-notes-jul-aug-2022
Language: Korean
Published: 2022-09-22
2022년 7월과 8월, 이번 여름 에어브릿지는 대시보드가 개선되고 빠른 검색과 실행 기능을 제공하며 한층 더 유저 친화적인 솔루션으로 거듭났습니다. 에어브릿지 대시보드 및 유저 가이드에서 빠른 검색과 실행 기능을 제공하기 시작했으며, 에어브릿지 리포트 테이블의 사용성이 개선되었습니다. Retention 리포트의 다운로드 포맷 옵션 추가 제공, 새로운 Standard Event의 추가, 그리고 국가별 지역 정보 수집 기능 업데이트로 인해 리포트에서 더 많은 데이터를 효율적으로 확인할 수 있게 되었습니다. 또한, MMP 최초로 비용 데이터 연동을 제공하는 카카오 모먼트의 비용 연동 방법이 더 쉬워졌습니다. 점점 더 발전하고 있는 에어브릿지의 신규 기능 릴리즈 소식을 한 번에 확인해보세요.
## 🔎 에어브릿지 대시보드의 빠른 검색과 실행 기능 제공
[]
에어브릿지 대시보드 내 메뉴 접근과 기능을 빠르게 실행할 수 있는 빠른 검색 기능이 추가되었습니다. 대시보드 상단의 [검색] 버튼 혹은 단축키(Mac: cmd+K Window: ctrl+K)를 통해 검색창을 빠르게 오픈할 수 있습니다. 또한, 에어브릿지의 내부 데이터 상 페이지 뷰 및 체류 시간 상위 5개 메뉴인 ‘새로운 트래킹 링크 생성’, ‘Actuals 보기’, ‘App Raw Data 추출하기’, ‘Integrated Ad Channels’, ‘새로운 사용자 초대’를 Shortcut으로 제공합니다. 이와 더불어 곧 대시보드 내에서 이용 가이드 및 개발자 가이드 검색도 가능해질 예정입니다.
### 🆕유저 가이드도 개선되었습니다.
[]
더불어, 유저 가이드도 보다 깔끔한 디자인과 함께 빠른 검색이 가능하도록 개선되었습니다. 검색어를 입력하면, 유저 가이드의 문서는 물론 에어브릿지 블로그 및 개발자 가이드 검색 결과도 함께 제공됩니다. 유저 가이드 내에서 검색창을 열기 위해서는 대시보드와 동일한 단축키(Mac: cmd+K Window: ctrl+K)로 빠르게 검색창을 활용할 수 있습니다.
👉 [지금 바로 유저 가이드 보러 가기](https://help.airbridge.io/hc/ko)
## ⚡️에어브릿지 리포트의 테이블 개선 - 스크롤바 위치 변경 및 컬럼 리사이징
[]
에어브릿지 리포트 테이블의 사용성이 개선되었습니다. 우선 상단의 이미지 상 빨간색 원형으로 표시된 부분은 스크롤바로, 기존에는 테이블 안쪽에 위치하고 있어서 셀의 내용을 가리는 경우가 있었지만 이제 더이상 스크롤바가 셀의 내용을 가리지 않습니다. 이번 업데이트는 스크롤바가 등장하는 모든 리포트에 적용되었습니다.
[]
또한, 테이블의 컬럼 리사이징을 지원합니다. 파란색 원으로 표시된 부분을 드래그하여 원하는 대로 컬럼의 사이즈를 조정할 수 있습니다. 혹은 빠르게 두 번 클릭하는 경우 빠르게 크기를 줄였다 늘렸다할 수 있습니다. 다만 Touchpoints Analysis 및 Touchpoints Overlap 리포트에는 테이블의 특수성으로 인해 본 업데이트가 반영되지 않았습니다.
이와 더불어 전체 리포트에서 테이블의 내용 표시 속도도 개선하였습니다.
## 📊 리포트에서 더 많은 데이터를 효율적으로 확인할 수 있도록 업데이트되었습니다.
### 리텐션(Retention) 리포트의 다운로드 포맷 옵션 추가 제공
[]
이제 Retention 리포트에서 Count 형태뿐만 아니라 % 형태에 대해서도 데이터를 추출할 수 있습니다. 지난 6월 [업데이트된 Retention 리포트가 정식으로 릴리즈](https://www.airbridge.io/ko/blog/airbridge-retention-report-update)되면서 기존 Retention 리포트의 지원이 종료되었지만, 기존에 제공하던 % 형태의 다운로드에 대한 고객사의 요청을 바탕으로 재지원하게 되었습니다. 이제 Retention 리포트 다운로드 시, 상단의 이미지와 같이 CSV (Count), CSV (%), 구글 스프레드시트 (Count), 구글 스프레드시트 (%)의 4가지 옵션을 확인할 수 있습니다. 기존 Retention 리포트와 동일한 형태의 데이터 확인이 필요하셨던 고객사분들의 불편 해소에 도움이 되었기를 바랍니다.
### 새로운 Standard Event 추가
에어브릿지에서 미리 지정해둔 이벤트 카테고리로 이벤트를 전송하게 되면 매체 포스트백 및 서드파티 솔루션에 이벤트 전송 시 이벤트 카테고리가 자동으로 매핑됩니다. 이와 같이 미리 지정해둔 이벤트 카테고리를 Standard Event Category라고 하며, Standard Event Category를 사용하여 전송한 이벤트를 **Standard Event**라고 부릅니다. 한편, 유저가 지정한 Event Category로 전송하는 이벤트는 Custom Event라고 부릅니다. 에어브릿지를 사용할 때 Standard Event를 적절히 활용하면 Actuals 리포트에서 이벤트 카테고리별 지표 제공 및 이벤트 카테고리 자동 매핑 등 에어브릿지를 보다 효과적으로 사용할 수 있다는 장점이 있습니다.
👉[개발자 가이드에서 Standard Event와 Custom Event에 대해 더 자세히 알아보기](https://developers.airbridge.io/docs/standard-event)
더 많은 데이터를 대시보드에서 효과적으로 확인할 수 있도록 에어브릿지에 14개의 Standard Event가 새롭게 추가되었습니다. 새로운 Standard Event는 아래와 같습니다.
- 결제수단등록, 위시리스트추가, 결제시작, 체험판시작, 구독, 구독취소, 광고클릭, 튜토리얼완료, 레벨달성, 과제달성, 평가, 공유, 일정예약, 크레딧 사용
- Add Payment Info, Add to Wishlist, Start Trial, Initiate Checkout, Subscribe, Unsubscribe, Ad Click, Complete Tutorial, Achieve Level, Unlock Achievement, Rate, Share, Schedule, Spend Credits
추가된 Standard Event와 함께 에어브릿지 대시보드에서 보다 효과적으로 데이터를 확인해보세요.
### 국가별 지역 정보 수집 및 수집된 데이터를 바탕으로 메트릭 확인
에어브릿지 리포트에서 국가별 지역 정보를 확인하고 싶은 경우를 위해, clientIP를 기반으로 City와 Subdivision을 알 수 있도록 데이터 필드가 추가되었습니다. 그리고 리포트 설정 시 그룹바이에서 Client IP City, Client IP Subdivision을 선택하여 수집된 국가별 지역 정보를 바탕으로 메트릭을 확인할 수 있습니다.
### Actuals 및 Trend 리포트에서 매출액 데이터의 Currency 확인 가능
여러 데이터 소스로 매출액 데이터의 다른 화폐 단위(KRW, USD 등)에 대해 Currency로 그룹바이를 걸어서 데이터를 확인할 수 있습니다. 예를 들어, Apple Search Ads로 기여된 매출액의 Currency는 USD이고 카카오 채널로 기여된 매출이 KRW 일 때 같은 테이블에서 단위를 확인할 수 없는 경우 혼란이 발생할 수 있습니다. 이를 방지하고자 에어브릿지 Actuals 및 Trend 리포트에서 매출액/광고매출액 관련 메트릭(e.g. Revenue (App), Ad Revenue (App))에 Currency로 Groupby를 걸어서 데이터를 확인할 수 있도록 업데이트 되었습니다.
## 🔗 카카오 모먼트의 비용 연동이 더쉬워졌습니다.
지난 2월 [에어브릿지는 MMP 최초로 카카오모먼트 비용 데이터를 연동하였습니다.](https://www.airbridge.io/ko/blog/airbridge-x-kakaomoment-cost) 다만, 카카오 디벨로퍼스에서 직접 앱을 생성하고 카카오 OpenAPI 사용 권한을 신청 및 승인을 받는 과정으로 인해 약간의 번거로움이 존재하였습니다. 하지만 이제 에어브릿지에서 카카오 로그인만으로 비용 데이터 연동이 가능해지며, 보다 간편하게 카카오 모먼트와의 비용 연동을 할 수 있게 되었습니다.
👉[카카오 모먼트 비용 연동 방법 더 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/4408910960281#%EB%B9%84%EC%9A%A9-%EC%97%B0%EB%8F%99)
## ⛏ SDK 릴리즈
에어브릿지는 앱(Android & iOS) 및 웹 SDK를 기반으로 웹과 앱 데이터를 통합적으로 분석합니다. 또한 지속적인 업데이트를 통해 사용성을 개선하고 있습니다.
7월과 8월에는 아래의 SDK가 릴리즈 되었습니다. 연결된 링크에서 자세한 Release Note를 확인할 수 있습니다.
- [Airbridge Web SDK](https://developers.airbridge.io/docs/web-sdk-release-note) - Airbridge Web SDK 1.8.158, Airbridge Web SDK 1.8.163, Airbridge Web SDK 1.8.168, Airbridge Web SDK 1.8.169
- [Airbridge Android SDK](https://developers.airbridge.io/docs/android-sdk-release-note) - Airbridge Android SDK v2.18.1, Airbridge Android SDK 2.19.0, Airbridge Android SDK 2.19.1[****](https://developers.airbridge.io/docs/ios-sdk-release-note)
- [Airbridge iOS SDK](https://developers.airbridge.io/docs/ios-sdk-release-note) - Airbridge iOS SDK 1.24.4, Airbridge iOS SDK 1.25.0
---
## 어트리뷰션 툴을 가리키는 여러 가지 용어들
URL: https://www.airbridge.io/ko/blog/attribution-tool-mmp
Language: Korean
Published: 2022-09-08
앱 마케터라면, **어트리뷰션 툴, MMP, 서드파티 트래킹 툴, 앱 트래커, 앱 마케팅 툴** 등과 같은 이름들을 들어보셨을텐요. 이 용어들은 에어브릿지(Airbridge)와 같은 모바일 마케팅 성과 분석 및 어트리뷰션 기능을 제공하는 SaaS 솔루션을 가리킵니다. 각각 의미하는 바가 다른 용어들처럼 보이지만, 결국에 하는 역할 및 지칭하는 솔루션은 모두 동일한 ‘어트리뷰션 툴’입니다. 이 어트리뷰션 툴을 MMP, 서드파티 트래킹 툴, 앱 트래커라는 여러 버전의 용어들로 혼용하여 사용하곤 합니다.
## 왜 이렇게 부르고 있을까요?
다양하게 지칭되고 있는 모바일 마케팅 성과 분석 및 어트리뷰션 솔루션, 왜 이러한 명칭을 갖게 되었는지 각 용어가 내포한 의미를 통해 살펴보겠습니다. 각 용어의 의미 및 배경을 알면, 어트리뷰션 툴을 왜 사용해야 하는지 그리고 이 솔루션이 마케팅에서 어떤 역할을 하는지 이해하는 데에도 도움이 될 수 있습니다.
### 모바일 어트리뷰션 솔루션 / 툴 (Mobile Attribution Solution / Tool)
‘모바일 어트리뷰션 솔루션/툴’이라는 용어는, 기능 그 자체를 이름으로 부르고 있는 것과 같습니다. 명칭 그대로, 모바일 상에서 어트리뷰션(attribution)을 분석하는 것을 도와주는 솔루션이라는 의미입니다. 혹은 ‘어트리뷰션’을 한국어로 번역한 ‘기여’를 넣어서, ‘모바일 광고 기여도 분석 툴’이라고 부르기도 합니다.
모바일 앱 마케팅에서 ROI를 높이기 위해서는 집행한 마케팅 캠페인의 전환에 대한 ‘어트리뷰션(기여도)’을 분석하는 것이 반드시 필요합니다. 과거의 성과를 제대로 분석하고, 그 원인을 파악하여 이후의 캠페인에 반영해야 하죠. 현재의 마케팅 캠페인을 계속 유지한다고 해서, 그 성과도 그대로 유지될 것이라는 보장이 없기 때문입니다. 이 어트리뷰션의 핵심은 발생한 마케팅 성과의 기여도를 어떤 터치포인트에 잘 분배하는가입니다. 이를 확인하는 방법론을 ‘어트리뷰션 모델(기여 모델)’이라고 부르며, 대표적으로 ‘라스트 터치 어트리뷰션(Last-Touch Attribution) 모델’ 및 ‘멀티 터치 어트리뷰션(Multi-Touch Attribution) 모델’ 두 가지가 있습니다.
👉[마케팅 성과를 다각도로 분석하기 위한 접근 방식 알아보기 1편 - 어트리뷰션(LTA, MTA)에서 더 자세히 살펴보기](https://blog.ab180.co/posts/marketing-measurement-1-attribution)
### MMP(모바일 측정 파트너, Mobile Measurement Partner)
MMP는 [Facebook/Meta 비즈니스 파트너(이전의 Facebook 마케팅 파트너)](https://www.facebook.com/business/marketing-partners)의 하위 집합인, 페이스북의 모바일 측정 파트너십 프로그램의 이름에서 시작되었습니다. 이후 시간이 흐르면서, SAN(Self Attributed Network)과 같은 수퍼 퍼블리셔에 대한 주요 모바일 측정 및 어트리뷰션 파트너와 동의어가 되었습니다. MMP는 문자 그대로, 다양한 기술을 사용하여 **모바일 상에서**발생하는 이벤트인, 클릭, 앱 설치, 참여, 수익 등을 **측정**하고, 어떤 터치포인트가 기여했는지 확인하도록 도와주는 파트너를 의미합니다. 다만, 일반적으로 모바일 어트리뷰션 솔루션과 같은 의미로 사용된다고 하더라도, 사실 페이스북의 MMP가 아니면 MMP라고 부를 수 없습니다.(예: 파이어베이스는 MMP가 아닙니다.)
MMP를 사용하게 되면, SAN을 포함한 다양한 매체에서 발생한 앱 성과를 측정하여 가장 효과적인 광고 채널을 발견할 수 있습니다. 예를 들어, 페이스북 SDK에서는 페이스북 광고의 성과만 확인할 수 있지만, 에어브릿지와 같은 MMP를 사용하면 페이스북은 물론 구글, 애플 등 다양한 매체의 성과를 함께 분석할 수 있게 되는 것이죠. 게다가 MMP에서 수집한 인앱 이벤트를 페이스북으로 전송하여 리타겟팅 및 최적화에 활용할 수도 있습니다. 에어브릿지는, 페이스북 MMP 뿐만 아니라 MTA(Multi-Touch Attribution) 파트너로도 선정되었는데요. MMP와 MTA 분야 파트너에 동시에 선정된 것은 에어브릿지가 세계 최초 사례입니다. ([에어브릿지의 프로필 보러가기](https://www.facebook.com/business/partner-directory/search?solution_type=measurement&ref=fmp_about_solution_card&capabilities=Mobile%20Measurement&id=3963759510351117§ion=overview))
👉[에어브릿지 MMP & MTA 페이스북 파트너 선정 소식 자세히 알아보기](https://blog.ab180.co/posts/airbridge-becomes-facebook-business-partner-for-measurement)
MMP를 활용 할 때 유의해야 할 점은, MMP에서 확인하는 광고 성과와 페이스북 광고 관리자에 표시된 결과가 다를 수 있다는 점입니다. 이는, 다음의 세 가지 경우로 인해 발생할 수 있습니다. 우선 페이스북과 MMP의 기여 모델이 다르기 때문에 발생하는 경우가 있으며, 기여 기간이 다르게 설정되어 있는 경우에도 수치에 차이가 발생할 수 있습니다. 또한, 전환 표시 시간 기준의 차이로 인해 발생하는 경우도 있습니다.
또한, iOS 14.5 업데이트 이후 Apple에서 광고로 인한 앱 설치 트래킹을 위해 제공하는 **SKAdNetwork 프레임워크**를 활용할 때에도 MMP의 도움이 있으면 좀 더 간편한 세팅이 가능합니다. SKAdNetwork를 통해 앱 설치가 일어날 경우, SKAdNetwork에서 광고주가 Conversion Value 형태로 인앱 이벤트를 최소한으로 트래킹할 수 있도록 데이터를 제공합니다. 이때, 무엇을 트래킹하는데 사용할지 설정하기 위해 MMP 혹은 앱 이벤트 API를 통해 Conversion Value를 세팅해야 합니다. 특히, 페이스북의 경우, MMP 혹은 앱 이벤트 API를 통해서 ‘전환 스키마(Configuration Schemas)’를 설정하여, 이벤트 관리자를 통해 어떤 전환 이벤트를 사용할 것이며, 그러한 이벤트의 가치를 어떻게 판단할 것인지를 페이스북에 알리는 작업이 필요합니다.
[👉Facebook의 전환 스키마 설정하는 방법 자세히 알아보기](https://www.facebookblueprint.com/student/activity/232676?sid=ae38229c-c3c6-4ec0-b5ca-4e240928bc00&sid_i=0)
[👉에어브릿지에서 SKAN Conversion Value 설정하는 방법 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/900006402223-SKAN-Conversion-Value)
### 서드파티 트래킹 툴(Third party tracking tool)
서드파티는 단어 그대로, ‘제3자’를 의미합니다. 각 마케팅 매체에서 제공하는 광고 성과 데이터를, 제3자의 입장에서 같은 기준으로 트래킹하여 성과를 측정하여 제공해주기 때문입니다. 이러한 동일한 기준을 바탕으로 한 광고 성과 분석은 두 가지 측면에서 중요합니다.
우선, **전반적인 플랫폼의 기준(Metric, 지표)을 통일**할 수 있습니다. 마케터는 표준화된 기준을 통해, 매체 간 성과를 비교하면서 개별 캠페인을 평가하고 최적화할 수 있습니다. 즉, 진짜 마케팅 성과를 분석할 수 있게 됩니다. 서드파티 트래킹 툴을 사용하면 전체 채널에 자체적으로 정한 기여 기간 및 기준을 적용할 수 있으며, 혹은 특정 채널에 따라 기여 기간을 다르게 설정할 수도 있습니다. 또한, 소재별 혹은 매체들 간의 성과를 분석하고 싶을 때, 서드파티 트래킹 툴을 사용하지 않는다면 각 대시보드에 나온 성과를 스프레드 시트 등으로 통합하여 확인해야 한다는 번거로움이 발생합니다. 게다가 매체별로 기여 기준이 다르기 때문에 제대로 된 성과인지도 확인하기 어려울 겁니다. 그렇기 때문에 제대로 된 마케팅 성과를 분석하기 위해서는 서드파티 트래킹 툴이 반드시 필요합니다. 에어브릿지의 경우, 웹 SDK 및 앱 SDK를 통해 웹과 앱에 대한 어트리뷰션을 확인할 수 있는 것은 물론, iOS와 Android, 데스크톱에서 발생한 마케팅 성과를 통합 대시보드에서 한 번에 확인할 수 있습니다.
[__wf_reserved_inherit]
두 번째로는, **매체별 광고 성과에 대해 정산에 사용**할 수 있습니다. 제3자의 입장에서 광고 성과를 분석함으로써 광고주와 매체가 서로 합의된 기준에 따라 공정하게 정산 할 수 있습니다. 만약 특정 매체에서 자사를 포함한 여러 매체의 광고 성과를 분석하게 된다면, 자신의 매체에 대해 유리한 방향으로 성과를 측정하는 문제가 발생할 수 있습니다. 또한, 서드파티 트래킹 툴은 제3자로서 [프로드](https://www.airbridge.io/ko/product/marketing-analytics#fraud_detection_kr)를 검증하는 규칙도 정할 수 있어서, 혹시나 발생할 수 있는 매체에서의 프로드를 효과적으로 예방할 수 있습니다.
이와 같은 이유로, 광고를 집행하는 매체가 아닌 에어브릿지와 같은 마케팅 성과 측정을 전문으로 하는 솔루션을 활용해야 합니다.
또한, 서드파티로서 트래킹을 하기 때문에, **개인정보보호 트렌드로 인한 변화에도 효과적으로 대응**할 수 있습니다. 지난 2021년 4월, iOS 14.5 버전의 개인정보보호 정책으로 인해 어트리뷰션 시장에 새로운 변화가 나타났는데요. iOS 플랫폼에서 개인화된 광고나 리타겟팅 광고를 집행하기 위해서는 유저의 추적 허용이 필요합니다. 이러한 변화에 대해 서드파티 트래킹 툴들에서 여러 가지 방안을 통해 제시하고 있기에, 개인정보보호 트렌드 변화 대응에 도움을 받을 수 있습니다. 에어브릿지에서는 앱 설치 트래킹 지연(Deferred Install Tracking), 컨버전 밸류(Conversion Value) 자동 세팅 및 컨버전 규칙 설정 등의 기능을 통해 iOS 14.5로 인한 변화에 완벽히 대응할 수 있습니다.
### 앱 트래킹 툴 / 앱 트래커 (App Tracking Tool / App Tracker)
이 용어의 경우, **앱에 대해 발생한 성과를 트래킹하는 툴(솔루션)**을 의미하는 용어임을 직관적으로 알 수 있습니다. 성과를 측정하기 위해서는, 집행한 각 매체에서 유저들이 마케팅 캠페인에 어떻게 반응했는지, ‘링크 클릭’, ‘클릭후 설치’, ‘설치 후 구매행동’ 등과 같은 유저들로부터 발생한 설치 및 인앱 이벤트를 ‘추적(Tracking)’해야 하기 때문입니다.
성과 측정을 위한 트래킹은, 에어브릿지와 같은 앱 트래킹 툴에서 제공하는 트래킹 링크를 통해 할 수 있으며, 또한 솔루션에 따라 동일한 링크를 [딥링크](https://www.airbridge.io/ko/product/deep-linking) 및 [브랜디드 링크](https://blog.ab180.co/posts/airbridge-branded-link)로도 활용할 수도 있습니다.
이와 더불어, 기존에 웹에서 유저의 행동을 추적할 때 사용하던 Google 애널리틱스와 같은 웹사이트 트래픽 분석 툴(웹 트래킹 툴, 웹사이트 추적 툴)이 있었기 때문에, 이 개념에 익숙하고 이 툴을 사용했던 사람들이, 모바일 앱이 활성화된 시기에 비슷한 용어로 사용하기 시작했음을 유추할 수 있습니다.
한 가지 유의해야 할 점은, [앰플리튜드(Amplitude)](https://www.ab180.co/solutions/amplitude)와 같은 프로덕트 분석 솔루션 또한 앱 내 행동을 추적하기 때문에 ‘앱 트래커’라고 부르는 경우도 있다는 점입니다. 하지만, 프로덕트 내에서 유저의 행동을 분석하는 앰플리튜드는, 광고 성과를 측정하는 어트리뷰션 툴과는 다른 목적을 지닌 솔루션입니다. 앱 트래커라고 지칭할 때는, 같은 카테고리로 혼동하지 않도록 용어 사용에 주의해야 합니다.
## 에어브릿지는 MMP이고, 서드파티 트래커이며, 앱 트래킹 툴이자, 모바일 어트리뷰션 툴입니다.
결론적으로는, 위와 같은 용어들이 모두 같은 솔루션을 지칭하는 용어임을 아는 것이 중요합니다. 그리고 이 용어들이 왜 이러한 명칭을 갖게 되었는지를 이해하면, 솔루션에 대한 이해도도 높아질 수 있습니다.
MMP이고, 서드파티 트래커이며, 앱 트래킹 툴이자, 모바일 어트리뷰션 툴인 [에어브릿지](https://www.airbridge.io/ko)는, 웹&앱 통합 분석과 인크리멘탈리티 분석, 멀티 터치 기여도 분석을 통해 진짜 마케팅 성과 분석을 제공합니다. 모바일 비즈니스에 최적화된 기능으로 진짜 마케팅 성과를 측정하고 개선해보세요!
---
## 성공적인 마케팅으로 이어지는 QR 코드 활용 팁 네 가지
URL: https://www.airbridge.io/ko/blog/airbridge-qr-code
Language: Korean
Published: 2022-08-31
몇 년 전만 해도 낯설었던 QR코드는, 코로나19 확산 방지를 위한 방안인 전자출입 명부 작성을 위해 활발하게 사용되며 일상 속 깊이 자리 잡았습니다. QR 코드가 익숙해지고 널리 활용되고 있는 지금 이때, QR 코드는 온˙오프라인 모두에서 매력적인 마케팅 수단이 될 수 있습니다. 단순히 오프라인 고객을 웹페이지나 앱 등 특정 모바일 링크로 연결시키는 것을 넘어서, 브랜드 인식 제고와 마케팅 리드 증가까지 다양한 목적을 달성할 수 있죠.
이번 포스트는 **QR 코드를 활용한 마케팅 팁**과 더불어, **강력한 딥링크 기능이 담긴 에어브릿지 QR 코드**를 활용해 인게이지먼트를 증가시키고 마케팅 효과를 높일 수 있는 방안을 알아보겠습니다.
## QR 코드의 역사 및 관심도 변화
QR 코드는 Quick Response Code의 약자이며, 흑백 격자무늬 패턴으로 정보를 나타내는 매트릭스 형식의 2차원 바코드입니다. 1990년대 중반에 처음 개발된 QR 코드는 2011년에 잠시 주목받았다가 최근 관심도가 더욱 증가하고 있습니다.
[__wf_reserved_decorative]
[__wf_reserved_decorative]
QR 코드에 대한 관심도 및 활용이 증가한 가장 큰 이유는, 안드로이드 및 iOS 기기에서 기본 카메라로 QR 코드를 바로 스캔하는 것이 가능해졌기 때문입니다. 게다가, 최근 코로나19의 전 세계적인 유행 이후 언택트 문화가 보편화된 것도 또 다른 이유입니다. 예를 들어, 터치스크린 대신 QR코드를 통해 개인 기기로 유저가 직접 인게이지하도록 유도할 수 있으며, 전자 출입 명부와 같이 개인 정보를 빠르고 안전하게 제공하기 위한 QR 체크인으로도 활용하고 있습니다. 미국의 경우, 질병통제예방센터에서 바와 레스토랑에서 버릴 수 있는 혹은 디지털 메뉴판을 활용하도록 권고했고, 많은 레스토랑들에서 QR 코드를 통한 메뉴 확인 및 주문 시스템을 사용하고 있기도 합니다.
[__wf_reserved_decorative]
## QR 코드를 활용한 네 가지 마케팅 팁
이렇게 우리의 일상에서 쉽게 찾아볼 수 있게 된 QR 코드는 활용도가 무궁무진합니다. QR 코드에 저장된 링크로 웹사이트나 앱으로 연결할 수 있는 것은 물론, 여러 미디어 형태를 저장하여 정보를 전달할 수 있습니다. 게다가 QR 코드는 어디에나 쉽게 넣을 수 있습니다. 영수증 하단, 제품 설명란, 티셔츠, 간판, 심지어 TV 광고에서도 QR 코드를 활용할 수 있습니다. 즉, QR 코드가 훌륭한 마케팅 수단이 될 수 있다는 것을 의미합니다.
다만, 아무리 좋은 콘텐츠를 담은 QR 코드라도, 이를 발견한 고객이 바로 구매로 전환될 가능성은 적습니다. 그렇다면 성공적인 마케팅 캠페인으로 이어지기 위해서는 어떻게 QR 코드를 활용해야 할까요? 마케팅에서 더 효과적으로 QR 코드를 활용할 수 있는 네 가지 팁을 소개합니다🚀
### 1) 타겟 오디언스가 위치한 곳에서 맥락에 맞게 활용해야 합니다.
QR 코드를 물리적 공간에서 활용할 경우, 타켓 오디언스가 있는 곳에 QR 코드를 배치해야 합니다. 이를 위해서는 다른 마케팅 캠페인과 마찬가지로 목표하는 오디언스에 대한 이해가 수반되어야 합니다.
먼저, 여러분의 브랜드가 어떤 고객과 어디에서 소통하고 있는지 파악하세요. 고객들이 무슨 목적을 가지고 특정 물리적 공간에서 QR 코드를 발견하게 될지 이해해야 합니다.
**오프라인 매장 혹은 식당을 방문하는 고객을 타겟으로, QR 코드를 비대면 마케팅에 활용할 수 있습니다.**
- QR 코드를 통해 상품 및 메뉴를 확인하고 주문까지 가능한 시스템을 구축하여, 고객에게 안전한 비대면 경험을 제공할 수 있습니다.
- 이외에도, QR 코드에 브랜드의 가치가 담긴 영상, 재료를 어디서 공급받았는지 등 다양한 콘텐츠를 담을 수도 있습니다. 이를 통해, 직접적인 커뮤니케이션이 어려운 상황에서도 브랜드 이미지를 제고할 수 있습니다.
**이외의 공간에서는 타겟 오디언스가 있을 법한 장소에서 QR코드를 맥락에 맞게 적절히 활용해야 합니다.**
- 하교하는 아이들을 데리러 가면서 빠르게 저녁을 준비하고 싶은 부모님들을 위해 학교 앞 혹은 어린이집 로비 등에 음식 광고와 함께 배달 주문을 할 수 있는 QR 코드를 배치할 수 있습니다.
- 저녁 시간 집으로 귀가하는 혹은 퇴근길 배고픈 사람들을 위해 버스 정류장 혹은 지하철역에 광고와 함께 배달 주문을 할 수 있는 QR 코드를 배치할 수 있습니다.
### 2) QR 코드를 통해 고객이 얻을 수 있는 부분을 명확하게 밝혀야 합니다.
타겟 오디언스를 위해 광고나 제품 설명란에 QR 코드를 적절히 배치하셨나요? 그렇다면 이제 다음 단계인 QR 코드 속 링크로 연결시켜야 합니다. 고객이 QR 코드를 발견하더라도 QR 코드 안에 어떤 정보가 있는지 알 수 없다면, 접속하지 않고 그냥 지나쳐버릴 것입니다. 혹은 QR 코드를 활용하는 방법을 모를 수도 있습니다. 그렇기 때문에, QR 코드 접속 방법과 QR 코드 속 정보를 확인해야 할 이유를 분명하게 설명하는 것이 중요합니다. 고객이 QR 코드에 접속함으로써 얻을 수 있는 부분을 적극적으로 표현하여 QR 코드의 링크를 열도록 만들어야 합니다.
예를 들어,
- 할인이나 프로모션 등을 언급하며 웹사이트나 앱으로의 접속을 유도할 수 있습니다.
- 제품 설명란에 더 자세한 사용 방법이 궁금하다면 QR 코드로 접속하여 영상을 통해 확인하라는 메시지를 넣을 수 있습니다.
- 가게를 홍보하는 팸플릿에 지도 이미지와 함께 구글이나 네이버 지도로 바로 연결되는 QR 코드를 삽입할 수 있습니다.
이처럼 적절히 활용된 QR 코드는 마케팅 퍼널의 최초 단계인 유저 유입을 늘리도록 도와줍니다. Paid 채널(검색어 광고, SNS 광고 등)을 통하지 않고서도 고객을 원하는 곳으로 바로 연결시킬 수 있기 때문에, PPC(Pay Per Click) 비용을 절감할 수 있습니다. 또한 유저와의 즉각적인 연결로 이어지기 때문에 Paid 채널에서 다른 경쟁자들과의 경쟁을 피할 수 있습니다.
### 3) 딥링크 QR 코드를 통해 개인화되고 자연스러운 연결 경험을 제공해야 합니다.
고객이 QR 코드를 통해 접속했더라도, 원했던 콘텐츠와 다르거나 모바일 친화적이지 않은 페이지로 연결된다면 유저의 경험을 해치기 때문에 즉시 이탈로 이어질 수 있습니다. 이 점을 유의하여 QR 코드를 생성한 후에는 항상 올바르게 연결되는지 테스트해야 하며, 업데이트가 있을 시에는 이를 QR 코드에도 바로 반영해야 합니다. 이때 딥링크 QR 코드를 사용하게 되면 유저에게 개인화되고 자연스러운 연결 경험을 제공할 수 있습니다.
[__wf_reserved_inherit]
에어브릿지(Airbridge)의 경우, 직접 연동되지 않은 Owned Media 혹은 Earned Media 채널인 Custom Channel(이메일, SMS, 블로그 포스트, SNS 게시물 등)에 대해 **트래킹 링크**와 더불어 트래킹 링크가 담긴 **에어브릿지** **QR 코드**를 생성하는 기능을 가지고 있습니다. 이 트래킹 링크와 QR 코드는 디바이스별로 유저를 랜딩시킬 최종 목적지를 다르게 설정할 수 있으며, 앱 내 특정 콘텐츠로 연결하는 딥링크도 설정 가능하기 때문에 맞춤 연결 경험을 제공할 수 있습니다.
트래킹 링크와 에어브릿지 QR 코드를 생성하는 방법은 아래와 같습니다.
- 채널, 캠페인, 광고그룹, 키워드, 매체 등의 정보를 입력하여 캠페인 파라미터를 정의합니다.
- 링크를 클릭한 유저의 최종 목적지를 앱 마켓, 앱(딥링크), 웹사이트 중 하나로 설정합니다. 이때 디바이스별로(Android, iOS, Desktop) 랜딩시킬 최종 목적지를 각각 다르게 설정할 수 있습니다.
- 트래킹 링크를 공유 혹은 게시하면 노출되는 섬네일을 미리 확인하고 설정합니다.
- 모든 설정을 완료하고 나서 \[링크 생성\] 버튼을 클릭하면, 상단의 화면과 같이 트래킹 링크 생성이 완료됩니다. 트래킹 링크와 함께, 해당 링크가 담긴 QR 코드도 동시에 생성됩니다.
이렇게 생성된 에어브릿지 QR 코드로 유저에게 자연스러운 연결 경험을 제공할 수 있을 뿐 아니라, 딥링크 경로나 리다이렉트 경로를 수정해야 할 때도 기존 QR 코드를 그대로 사용할 수 있어 유저의 경험을 해치지 않으면서도 빠르게 업데이트할 수 있습니다.
### 4) 에어브릿지 QR 코드는 트래킹 코드 역할을 해서, 고객 데이터 분석에 활용할 수 있습니다.
앞서 언급한 것처럼, QR 코드는 유저를 웹사이트 및 링크로 바로 연결할 수 있기 때문에 유입 채널은 물론, 유입 후 행동 등 다양한 데이터를 자체적으로 얻을 수 있습니다. 즉, 퍼스트 파티 데이터를 수집할 수 있게 됩니다. 써드 파티 데이터(Third Party Data)에 의존하지 않고도 더 많은 오디언스를 확보할 수 있게 되는 거죠. 이는 일반 QR 코드가 아닌, **에어브릿지 QR 코드**와 같이 트래킹 링크를 저장한 **딥링크 QR 코드** 일때만 가능합니다.
또한, 해당 데이터를 기반으로 에어브릿지와 같은 어트리뷰션 툴을 통한 기여도(Attribution) 분석도 할 수 있습니다. TV 광고 등 마케팅 성과 측정이 어려운 광고 채널이라도 QR 코드로 수집한 데이터를 기반으로 온˙오프라인 모두에서 크로스 디바이스에 대한 통합적인 마케팅 성과를 분석할 수 있죠.
딥링크 QR 코드로 수집할 수 있는 데이터 및 인사이트의 예시는 다음과 같습니다.
- 타겟 오디언스와 일치하는 유저 유입에 효과적인 채널 파악
- 특정 캠페인이 유저의 장기간 리텐션 및 인게이지먼트에 미치는 영향력
- 광고 채널 별, 유입 이후 나타나는 행동이나 관심사 차이
위와 같은 데이터는 [에어브릿지](https://www.airbridge.io/ko) 및 [Amplitude](https://www.ab180.co/solutions/amplitude)와 같은 프로덕트 분석 툴을 활용하여 더 효과적으로 분석할 수 있습니다. 분석을 통해 얻은 인사이트를 기반으로, 비즈니스의 성장을 이끌 수 있는 보다 성공적인 마케팅 전략을 기획해보세요!
## Conclusion
지금까지 QR 코드를 통해 마케팅 효과를 높일 수 있는 네 가지 방법을 알아보았습니다. 우선 딥링크 QR 코드를 생성하여, 타겟 오디언스를 위한 적절한 맥락에 배치해야 합니다. 그리고 고객이 얻을 수 있는 부분을 잘 표현하여 올바른 콘텐츠에 연결시키면, 유저 유입을 늘릴 수 있을 뿐 아니라 데이터를 트래킹해서 마케팅 성과도 개선할 수 있게 될 겁니다.
QR 코드가 일상 속 깊이 자리 잡은 지금, 에어브릿지 QR 코드와 함께 더 효과적인 마케팅 캠페인을 진행해보세요!
---
## [에어브릿지 in MGS 2022] 통합 측정 스택: MMP + MTA + MMM + Lift
URL: https://www.airbridge.io/ko/blog/mgs-2022-unified-measurement-stack
Language: Korean
Published: 2022-08-24
에어브릿지(Airbridge)는 지난 7월 13일 서울 삼성동 코엑스 그랜드볼룸에서 개최된 [모던 그로스 스택(Modern Growth Stack 2022, 이하 MGS 2022)](https://www.moderngrowthstack.com/) 컨퍼런스를 주최하였습니다. MGS 2022 속 에어브릿지가 전한 인사이트를 시리즈로 소개합니다.
> MGS 2022는 AB180(Airbridge)에서 개최한 애드테크&마테크 컨퍼런스로, 온오프라인 2,000여 명이 참여, 국내외 40여 명의 연사의 20여 개 세션으로 디지털 프로덕트 시대를 이끌어가는 주역들의 철학과 가치, 비즈니스 노하우를 공유하였습니다.
👉 MGS 2022 컨퍼런스의 전반적인 내용을 담은 현장 스케치는 [여기 링크](https://blog.ab180.co/posts/mgs-2022)에서 확인해보세요.
첫 번째로 공개하는 세션 스케치는 메인 컨퍼런스 홀에서 진행된 한용철 AB180 CSO의 “Unified Measurement Stack: MMP + MTA + MMM + Lift” 입니다. 새로운 트렌드를 마주한 광고 마케팅 영역에서 어떻게 성과를 측정해야할 것인지에 대해 알아보고, 에어브릿지가 제안하는 해결 방안을 소개했습니다.
## 마케팅 측정 영역이 마주한 새로운 도전
2022년 마케팅 측정(Marketing Measurement) 영역은 개인 정보 보호 트렌드와 기존 성과 분석 모델의 한계로 새로운 도전을 받고 있습니다.
### 개인 정보 보호 트렌드로 인한 유저 레벨 데이터 수집의 제한
이미 잘 알려진 것처럼, Apple의 iOS 14.5 업데이트 이후 사용자 식별값의 수집이 제한되고 있으며, Chrome과 Safari 등 브라우저에서 Cookie의 수집을 제한하고 있습니다. 또한 GDPR, CCPA 등의 제도적인 Privacy 보호 정책 등으로 인해 전반적으로 User-level Data의 수집을 제한받고 있습니다.
### 기존 라스트 터치 모델(Last-touch Model)의 한계
라스트 터치 어트리뷰션 모델은 기존에 가장 널리 사용되고 있는 성과 측정 모델로, 전환 직전 마지막 단 1개의 터치포인트에 모든 성과를 부여합니다.
> 👉[마케팅 어트리뷰션 모델(Marketing Attribution Model) 알아보기 포스트](https://www.airbridge.io/ko/blog/what-is-marketing-attribution-model)에서 더 자세히 알아보기
복잡한 어트리뷰션 모델을 단순화할 수 있기 때문에 적은 리소스로 합리적인 성과를 측정할 수 있다는 점에서 많이 사용하고 있는 모델이지만, 실제 광고 성과 증분(Incrementality)을 분석할 수 없다는 한계가 존재합니다. 또한, 전체 성과를 한 개의 터치포인트에 기여하기 때문에 멀티 채널 상황에서의 유의미한 분석이 어려우며, 마케팅 성과로 발생한 터치포인트 전체를 분석에 활용할 수 없습니다.
## 에어브릿지는 어떻게 대응하고 있을까
이러한 개인 정보 보호 환경과 기존 모델의 한계점을 극복하기 위해 에어브릿지는 다음의 새로운 두 가지 대응 방안을 제안합니다.
### 멀티 터치 어트리뷰션(Multi-Touch Attribution, MTA) 성과 분석
에어브릿지는 실제 광고 성과인 증분(Incrementality)을 확인하고 분석할 수 있는 서비스를 제공합니다. 에어브릿지의 멀티 터치 어트리뷰션 분석을 통해 마케터들은 각각의 터치포인트들이 유저 전환에 어떤 영향을 미쳤는지 더 명확하게 이해할 수 있습니다. 그리고 이를 위해 [메타(페이스북)와 멀티 터치 어트리뷰션 파트너십(MTA Partnership)을 체결](https://www.airbridge.io/ko/blog/airbridge-becomes-facebook-business-partner-for-measurement)하는 등 파트너십을 통해 라스트 터치포인트 이전의 데이터를 받아오고 있습니다. 또한, PET이라고 하는 프라이버시 보존 기술을 통해 개인 정보를 보호하면서 성과 분석을 제공하고 있습니다.
### 마케팅 믹스 모델링(Marketing Mix Modeling, MMM) 성과 분석
마케팅 믹스 모델링은 마케팅 활동으로 인해 발생한 매출 및 ROI(투자 수익률) 증가를 수량화하여 산출하는 거시적인 데이터 기반 통계 분석 모형입니다.
에어브릿지에서는 회귀(Regression) 혹은 베이지안(Bayesian) 모델 기반 분석을 통해 실제 광고 성과 (Incrementality) 분석이 가능합니다. 통계 모델 기반 분석 방법을 통해, 예산 분배에 따른 캠페인 성과를 예측할 수 있습니다. 또한, 유저 레벨의 데이터가 아닌 이미 집계된(Aggregated) 데이터를 사용한 분석이기 때문에, 사용자 식별값을 수집하기 어려운 환경에 대응할 수 있습니다.
### 단점은 보완하고 장점은 취하는 통합된 측정 스택을 만들어야 합니다.
[]
결론적으로, 한용철 AB180 CSO는 앞서 소개한 에어브릿지 성과 분석 솔루션과 기존의 라스트 터치포인트 모델의 성과 분석을 종합적으로 적용하여, 통합된 측정 스택(Unified Measurement Stack)을 세워야 한다고 제안합니다. 각각의 성과 모델은 한계점을 가지고 있을 수밖에 없기 때문에, 서비스 및 프로덕트의 상황과 특성에 맞게 세 모델을 적절히 조합하여 단점은 보완하고 장점은 취하여 통합된 측정 스택을 만들어가야 합니다.
### 에어브릿지의 관측적 분석 방법과 실험적 분석 방법을 상호보완하여 실제 광고 성과를 분석해보세요.
통합된 측정 스택의 세 가지 솔루션으로 제안하는 MMM, MTA, MMP는 기 수집된 데이터를 활용하여 모델링을 한 후 성과 분석을 진행하는 **관측적(Observational) 방법**입니다. 캠페인 집행과 성과 분석이 독립적으로 가능하고, 빠른 분석 결과를 얻을 수 있다는 장점이 있습니다. 하지만, 한편으로는 모델링이기 때문에, 이 분석 결과가 실제인지 아닌지를 확인할 수 없다는 한계도 존재합니다.
반면, **실험적(Experimental) 방법**은 실제 실험인 만큼 정확도는 높아지지만, 집행한 마케팅 성과를 분석하기 위해 실제 실험을 집행을 해야 하며, 실험이 끝날 때까지 성과를 확인할 수 없다는 단점으로 인해 쉽게 진행하기 어려웠습니다.
그렇기 때문에, 평소에는 관측적 방법을 통해 성과 분석을 빠르게 진행하고 반기 혹은 분기별로 실험적 방법 집행하여 실험을 통해 분석한 모델과 실제가 얼마나 일치하는지 확인하며 두 가지 방법론을 상호보완하여 사용하도록 제안합니다.
> 다노에서 에어브릿지의 관측적 방법인 멀티 터치 어트리뷰션(MTA) 분석과 Meta의 실험적 방법인 Lift 스터디를 함께 진행하여 Meta의 마케팅 성과를 분석하였습니다. 그 결과 Lift 스터디와 MTA 스터디 성과가 단지 근소한 차이만을 보였습니다. MTA 스터디에서 앱 설치는 2% 적었으며, 판매는 3% 적었습니다. 이를 통해 다노는 진행 중인 캠페인 전략에 확신을 가질 수 있었으며, 에어브릿지의 MTA 분석을 통해 빠르고 정확하게 성과를 분석하고 있습니다.
👉[다노의 케이스 스터디 더 자세히 알아보기](https://www.facebook.com/business/success/dano)
## 에어브릿지의 진짜 마케팅 성과 분석
에어브릿지는 기존 MMP/어트리뷰션 툴과 같이 라스트 터치포인트 모델의 성과 분석을 제공하고 있으면서도, 멀티 터치 어트리뷰션 모델(MTA)과 마케팅 믹스 모델링(MMM)과 같은 복수의 성과 분석 방법으로 진짜의 마케팅 성과 분석을 돕습니다.
한용철 CSO는 측정 솔루션을 만들고 판매하는 입장이지만, 측정 솔루션 하나만을 절대적으로 의존해서는 안 된다고 언급하며, “에어브릿지는 단지 측정만을 위한 것이 아닌, 측정을 통해 마케팅이 성공하는 과정에 일조하고 싶다.”고 이야기했습니다. 에어브릿지와 함께 진짜 성과 측정을 통해 마케팅 성공에 가까워질 수 있기를 바랍니다.
*마케팅 측정 영역이 마주한 새로운 도전과 에어브릿지의 해결 방안*에 대한 자세한 내용이 궁금하시다면, 아래 버튼을 클릭하고 **AB180 Academy**에서 MGS 2022 다시 보기 영상을 확인해 보세요! 간단한 회원가입 후 즉시 무료로 녹화 영상 시청이 가능합니다.
*AB180 Academy는 AB180에서 만든 마케팅 솔루션 교육 플랫폼으로, AB180이 서비스하는 솔루션 사용법 및 마테크 실무 활용법을 무료로 배우실 수 있게 동영상 자료를 지원하고 있습니다.
[]
---
## 에어브릿지 트래킹 링크로 심리스한 유저 경험 제공하기
URL: https://www.airbridge.io/ko/blog/airbridge-tracking-link
Language: Korean
Published: 2022-08-24
앱 마케팅의 궁극적인 목표 중 하나인 ‘유저 획득’을 위해서는 여러 플랫폼과 브라우저에 흩어져 있는 유저들을 원활하게 우리 앱으로 유입시켜야 합니다. 이를 위해 반드시 필요한 것이 ‘딥링크’입니다. 하지만, 기술적으로 구현하기에 복잡하며 모든 상황에서 원하는대로 작동시키기 쉽지 않습니다. 게다가 앱을 설치하지 않은 유저에게 앱 설치 후 딥링크를 연결해주는 디퍼드 딥링크 기능은 직접 구현하기에 더욱 까다롭죠.
그렇다면 이러한 문제를 어떻게 해결할 수 있을까요? 모바일 어트리뷰션 솔루션 / MMP에서 광고 성과를 측정하기 위해 제공하는 **트래킹 링크**를 활용하여 딥링크 기술을 적용하는 겁니다. 이번 포스트는 에어브릿지 트래킹 링크의 역할 및 동작 방법을 알아봅니다. 심리스한 유저 연결을 만들고 싶다면, **에어브릿지 트래킹 링크**를 확인해 보세요.
> 💡 딥링크에 대한 자세한 소개가 궁금하다면, \[딥링크 101\] 시리즈를 확인해보세요.[
> ](https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers)\- [\[딥링크101\] 마케터와 개발자를 위한 딥링크 시작하기](https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers)\- [\[딥링크101\] iOS Safari에서 딥링크 동작 시 경고 문구가 뜨는 이유가 무엇일까요?](https://www.airbridge.io/ko/blog/deeplink-101-ios-safari-alert)\- [\[딥링크101\] 디퍼드 딥링크로 앱을 설치하지 않은 유저도 앱 내 특정 페이지로 연결하기](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)
> 딥링크에 대한 가장 쉬운 가이드, <[에어브릿지 핸드북 - 딥링크](https://www.airbridge.io/ko/library/airbridge-handbook-deeplinks)\>를 다운로드하고 확인해보세요.
## 에어브릿지 트래킹 링크는?
**에어브릿지 트래킹 링크**는 유저로부터 발생한 행동에 어떤 채널이 기여됐는지 분석할 수 있는 것(어트리뷰션)은 물론, 링크를 클릭하는 유저들을 원하는 목적지에 도달할 수 있게(딥링크) 만들어줍니다. 트래킹 링크의 딥링크는 URI Scheme 유형 뿐만 아니라 App Links(Android) & Universal Links(iOS) 유형, 그리고 [디퍼드 딥링크](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking) 기술도 포함합니다. 트래킹 링크로 유저에게 앱 설치를 유도하거나, 앱 내 특정 페이지로 연결하는 등 다양한 광고 캠페인에 활용할 수 있습니다.
> 💡 간단하게 알아보는 에어브릿지 트래킹 링크의 동작 과정 :
> 에어브릿지 트래킹 링크 생성 → 링크 클릭 → 에어브릿지에 데이터 보내기 → 데이터 매칭 → 설정한 리다이렉트 경로로 리다이렉션 & 딥링킹 및 유저 트래킹 → 유저 이동 및 에어브릿지 대시보드에서 데이터 확인
>
> 👉 [에어브릿지 트래킹 링크 생성 방법 알아 보기](https://help.airbridge.io/hc/ko/articles/900001037886)
### 에어브릿지 트래킹 링크의 역할
에어브릿지 트래킹 링크의 역할은, 1) 유저 트래킹 그리고 2) 리다이렉션 및 딥링킹으로 나눠서 생각해볼 수 있습니다.
1\. 유저 트래킹
우선 ‘트래킹 링크’라는 이름이 의미하는 바 그대로 유저를 ‘트래킹’하는 것이 중요한 역할입니다. 광고주의 입장에서는 트래킹 링크로 수집한 데이터를 활용하여 각 매체별 정산에 활용하거나 미디어 믹스의 조정으로 마케팅을 최적화할 수 있습니다.
그리고 에어브릿지에서는 데이터 수집을 위해 유저가 트래킹 링크를 클릭하면, click event에 담긴 캠페인 파라미터를 태깅(tagging)하여 저장합니다.
이 과정에서 우선 에어브릿지 서버가 유저를 식별할 수 있는 고유의 값을 발급하여 브라우저 상에서 유저를 구분하고 유저 여정을 트래킹합니다. 또한, 웹사이트 상에서 이동하는 유저의 여정을 추적하여 기여 정보를 받기 위해 Airbridge Referrer 파라미터를 발급합니다.
그리고, 누가 어떤 매체의 어떤 광고를 보고 언제 이벤트가 발생했는지 정보를 확인하여, 각 채널에서 발생한 전환에 대한 기여를 매칭하기 위해 **터치포인트를 수집**합니다.
이렇게 수집한 데이터는 에어브릿지 대시보드에서 한눈에 확인할 수 있습니다.
2\. 리다이렉션 및 딥링킹
에어브릿지를 사용하는 광고주 입장에서는 트래킹 링크를 사용하여 링크를 클릭한 유저를 원하는 목적지로 안내하는 것도 중요합니다. 그 목적지는 리다이렉션 설정에 따라 앱, 앱 마켓, 혹은 웹사이트가 될 수 있습니다. 더불어, 딥링크 기술도 트래킹 링크에 기대하는 중요한 역할 중 하나일 겁니다.
여기서 _리다이렉션_은 클라이언트가 설정한 목적지로 유저를 이동시키는 것이며, _딥링크_는 그 중에서도 앱 내 특정 페이지로 이동시키는 경우를 의미합니다.
### 에어브릿지의 두 가지 트래킹 링크 형태
1\. abr.ge도메인의 Long과 Short 형태
에어브릿지로 트래킹 링크를 생성하면 기본적으로 _abr.ge도메인_을 사용하게 되며, Long과 Short 두 가지 형태로 생성할 수 있습니다.
[__wf_reserved_decorative]
우선 **Long** 형태는, 기여 정보가 모두 포함되어 링크의 길이가 길어지지만, 기여 정보를 필요에 따라서 마케터 혹은 매체에서 직접 변경할 수 있다는 장점이 있습니다. 매체 또는 지면을 제공하는 곳에서 동적으로(다이나믹하게) 기여 정보가 변경되거나, 혹은 링크의 텍스트가 직접적으로 보여지지 않는 경우에 주로 사용됩니다.
[__wf_reserved_decorative]
그리고 **Short** 형태는, 기여 정보가 랜덤으로 생성된 아이디로 변경되어 링크의 길이가 짧으며, 기여 정보를 변경하기 위해서는 에어브릿지 대시보드를 통해서만 변경이 가능합니다. 주로 텍스트가 직접적으로 보여지는 경우에 사용됩니다.
2\. 에어브릿지 브랜디드 링크
에어브릿지 트래킹 링크의 기본 도메인(abr.ge) 대신, 고객사에서 보유하고 있는 도메인을 이용하여 브랜디드 링크로 생성 및 활용할 수 있습니다. 즉, 커스텀 채널의 트래킹 링크에 사용되는 도메인을 브랜드와 어울리도록 새롭게 설정할 수 있습니다. 이 기능을 활용하면 [Abit.ly](http://abit.ly/) 혹은 bitly 등을 통해 URL을 축약했던 것 대신, 우리 브랜드명을 담은 깔끔하고 보기 좋은 URL로 생성할 수 있습니다.
[__wf_reserved_decorative]
> 👉 브랜디드 링크에 대한 더 자세한 내용은 [브랜디드 링크로 브랜딩과 성과 측정 한 번에 해결하기](https://www.airbridge.io/ko/blog/airbridge-branded-link) 포스트에서 자세한 내용을 확인해보세요.
> 👉 [커스텀 도메인 등록하는 방법 알아보러 가기](https://help.airbridge.io/hc/ko/articles/900002943346-Custom-Domain)
## 에어브릿지 트래킹 링크, 딥링크로 활용하기
트래킹 링크를 사용하면 다양한 플랫폼, 채널, 상황에서 링크를 클릭하는 유저들을 원하는 목적지에 심리스하게 도착하도록 만들 수 있습니다. 게다가 [\[딥링크101\] 마케터와 개발자를 위한 딥링크 시작하기](https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers)에서도 이야기한 것처럼 각 앱 프로덕트에서 딥링크 기술을 유지 보수 등의 이슈 없이 사용하기 어렵기 때문에, 에어브릿지와 같은 MMP의 트래킹 링크를 통해 딥링크 기술을 구현하는 것이 좋습니다.
[__wf_reserved_decorative]
### 에어브릿지 트래킹 링크 생성 화면에서 간편하게 딥링크 설정하기
[__wf_reserved_decorative]
에어브릿지 트래킹 링크는 **리다이렉트 경로** 설정을 통해 링크를 클릭한 유저의 최종 목적지를 앱 마켓, 앱(딥링크), 웹사이트 중 하나로 설정하여 지정할 수 있습니다. 이때, _App(Deep Link) 설정 화면_에서 앱 내 특정 페이지로 연결된 딥링크를 넣으면, 트래킹과 동시에 앱 내 특정 화면이나 콘텐츠로 유저를 문제 없이 이동하도록 만들 수 있습니다. 트래킹 링크를 통해 딥링크를 구현하면, MMP의 지속적인 업데이트 및 개발의 도움을 받아 딥링크를 유저의 다양한 환경에서 구현할 때 마주할 수 있는 여러 문제를 방지할 수 있습니다. 또한, 에어브릿지 트래킹 링크는 딥링크를 설정하면, 앱을 설치하지 않은 유저도 앱 설치 후 특정 페이지로 바로 연결시켜주는 _디퍼드 딥링크_기술도 디폴트로 구현할 수 있습니다.
### 딥링크를 셋팅한 에어브릿지 트래킹 링크는 아래와 같은 흐름으로 작동합니다.
1. 유저가 링크를 클릭하면 딥링크를 시도하여 앱이 열리는지 여부를 확인합니다.
2. 앱이 설치되어 있지 않아 앱 오픈에 실패하면 Fall Back으로 웹 브라우저가 열립니다.
3. 웹 브라우저에서 Javascript가 동작하여 User Agent 환경을 확인합니다.
4. 확인한 User Agent에 따라서 리다이렉트합니다.
한편, 각 프로덕트의 브랜딩이 더해진 커스텀 도메인(에어브릿지 브랜디드 링크)을 사용하고 Universal Links를 설정해 둔 경우, 아래와 같이 딥링크를 동작시킵니다.
1. 유저가 트래킹 링크를 클릭하면, Universal Links를 실행시킵니다.
2. 광고주가 에어브릿지에 설정된 경로로 Fallback 합니다.
이때, Universal Links를 지원하는 App 지면에서는 에어브릿지의 중개 페이지인 에어페이지를 거치지 않고 앱이 즉시 열리게 됩니다.
## 에어브릿지 트래킹 링크로 딥링크를 문제없이 구현해보세요.
이번 포스트를 통해 에어브릿지 트래킹 링크에 대한 소개와 더불어 딥링크로의 활용 방법을 알아보았습니다. 에어브릿지 트래킹 링크는 Redirection Path에서 이미 세팅된 딥링크를 잘 연결해두면 유저의 다양한 OS 및 브라우저 환경에 대해서 딥링크를 문제 없이 대응할 수 있기 때문에, 앱 마케팅에 반드시 필요한 딥링크를 편리하고 원활하게 구현할 수 있습니다.
에어브릿지의 심리스한 딥링크 연결을 지금 경험해보세요!
> ✔️ 이미 에어브릿지를 쓰고 계신 고객사라면?
> 👉 자세한 딥링크를 활용 방법을 담당 CSM에게 문의해주세요!
>
> ✔️ 자체적인 혹은 타사에서의 딥링크 구현 및 활용에 어려움을 겪고 계시다면?
> 👉 [에어브릿지의 앱 마케팅 전문가와 이야기 나눠보세요!](https://www.airbridge.io/ko/book-a-demo)
---
## 모바일 게임 산업 생태계의 주요 플레이어
URL: https://www.airbridge.io/ko/blog/mobile-game-ecosystem
Language: Korean
Published: 2022-08-22
2022년 글로벌 모바일 게임 시장의 수익은 1,035억 달러로 전체 글로벌 게임 시장의 53%를 차지합니다. 작년에 비해 PC 브라우저 게임 및 콘솔 게임 시장은 감소하고, 다운로드 PC 게임은 1% 정도 약간 증가한 것에 비해 모바일 게임 시장은 5% 이상의 증가세를 보이고 있습니다.
[__wf_reserved_decorative]
이와 같은 모바일 게임 업계의 성장에는 개발은 물론 운영, 마케팅, 시장 분석 등을 도와주는 다양한 플레이어들이 뒷받침하고 있습니다. 이 생태계는 계속해서 변화하고 고도화하고 있지만, 전체적으로 조망할 수 있도록 게임 시장 데이터 리서치 업체 뉴주(Newzoo)에서 ‘게임테크(Gametech)’라고 불리는 생태계로 연결하여 [GameTech Ecosystem 인포그래픽](https://newzoo.com/insights/infographics/gametech-ecosystem-map-technology-game-creation-supply-chain)을 발표했습니다.
이번 포스트에서는 뉴주에서 정리한 분류를 바탕으로 모바일 게임 산업 생태계와 주요 플레이어를 살펴봅니다.
## 개발사(Game Developers, 스튜디오)와 배급사(Game Publisher, 퍼블리셔)
‘게임사’라고 하는 경우, 게임을 개발하는 ‘개발사’와 시장에 유통하는 ‘배급사(퍼블리셔)’의 역할 모두를 포괄하여 지칭하는 것이 대부분입니다.
조금 더 상세히 살펴보면 **개발사**는 이름 그대로 게임을 제작하는 곳으로, 스튜디오라고 부르기도 합니다. 그리고 사업성 있는 게임을 발굴하고, IP 라이센스를 계약해서 시장에 유통하는 역할을 하는 **배급사(퍼블리셔)**가 있습니다. 글로벌 런칭, 마케팅, 디자인, 영상 등 게임 홍보 업무가 주된 조직입니다. 다만 여기서 주의해야 할 점은 최근에는 개발사와 퍼블리셔의 경계가 허물어져서 하나의 역할만 하는 경우가 드물어졌습니다. 개발 후 자체적으로 퍼블리싱하고 동시에 타사의 게임을 퍼블리싱하는 경우도 있으며, 게임 개발에만 집중하기 위해 개발사를 분사시켜서 자회사로 두는 경향도 있기 때문입니다. 이에 잘 알려진 게임사들의 경우 개발사인지 혹은 퍼블리셔인지 명확하게 구분하기 어려우며 특히 순수 ‘개발사’로 남아있는 경우가 드물어졌습니다.
국내의 주요 퍼블리셔로는 2022년 Data.ai에서 발표한 전 세계 소비자 지출 기준 APAC 본사 상위 30위 퍼블리셔에 포함된 넷마블(Netmarble), 엔씨소프트(NCSOFT), 넥슨(Nexon), 카카오(Kakao), 그리고 한국 소비자 지출 기준 한국 본사 상위 10위 팀 배틀(RPG) 퍼블리셔에 오른 데브 시스터즈(DEVSISTERS), 컴투스홀딩스(Com2uS Holdings), 스마일게이트(Smilegate), 클로버 게임즈(CLOVERGAMES), 게임펍(Gamepub), 시프트업(Shiftup), NHN, 베스파(Vespa Interactive) 등이 있습니다. ([출처](https://www.data.ai/kr/insights/data-ai-news/top-publishers-awards-2022/)) 앞서 언급한 것처럼 위 기업들 중 대다수가 동시에 개발사이거나 자회사로 개발사를 가지고 있습니다.
## 게임 개발 툴(Development)
게임 개발에는 다양한 기술이 필요하며, 이를 도와주는 플랫폼 및 툴들도 존재합니다. Newzoo에서는 게임 개발에 필요한 기술을 제공하는 게임 엔진(Game Engine) , 미들웨어(Middleware), 소프트웨어 관리 툴(Software Management Tool), 현지화 툴(Localization), 아트 툴(Art tool), 오디오 툴(Audio Tool) 등으로 세분화해두었습니다.
게임 엔진은 게임을 구동시키는데 필요한 핵심요소들을 담은 소프트웨어로, 개발사에서 자체적으로 개발하는 경우도 있지만 상용 엔진을 구매하여 사용하는 경우도 많습니다. 모바일 게임에서 많이 사용되고 있는 엔진으로는 유니티(Unity), 언리얼(Unreal), 게임메이커(GameMaker) 등이 있습니다.
미들웨어는 게임을 개선하고 서비스 및 기능을 추가하도록 도와주는 소프트웨어로, 사운드 컨트롤을 도와주는 Wwise, 게임 내에서 치팅(Cheating)을 방지하는 이지 안티 칫(Easy Anti-Cheat) 등이 있습니다.
이외에도 소프트웨어 관리 툴이나 현지화 툴, 아트 툴, 오디오 툴 등을 활용해서 적재적소에 필요한 툴을 통해 효율적인 게임 개발을 도움받을 수 있습니다.
## 플랫폼(Platform)
모바일 게임 앱 개발을 완료했다면 이를 플랫폼에 출시해야 합니다. 모바일 게임 앱을 다운로드하거나 결제를 할 수 있는 **앱 마켓**인 구글 플레이(Google Play), 앱 스토어(App Store), 원스토어(ONE store), 갤럭시 스토어(Galaxy Store) 등이 주요 플랫폼입니다.
일반적인 앱 마케팅과 마찬가지로 모바일 게임 앱 마케팅에 있어서 Paid 마케팅과 Organic 마케팅은 중요한 두 축입니다. 그중에서도 유저 획득 비용을 낮추기 위해서는 얼마나 많은 유저를 자연적으로 발생시킬 수 있는가가 중요합니다. 즉 유저가 우리 게임을 발견할 수 있는 가능성을 높여야 합니다. 위의 플랫폼은 Oraganic으로 유저를 획득할 수 있는 주요한 채널이며, **ASO(App Store Optimization)**를 통해 검색에 대한 노출 순위 및 클릭율을 높일 수 있습니다. ASO를 도와주는 툴/서비스로는 SplitMetrics, App Radar, StoreMaven 등이 있습니다. 또한 앱 마켓 내의 차트에서 순위권에 들어가는 것도 중요하기 때문에 게임 런칭 직후 순위를 높이기 위한 대대적인 마케팅 전략을 기획하기도 합니다.
## 수익화 플랫폼(Monetization)
한편, 모바일 게임 앱의 수익화를 도와주기 위한 다양한 플랫폼 및 솔루션도 존재합니다. 모바일 게임 앱의 주요 수익 모델로는 IAA(In-app Ads, 인앱 광고), IAP(In-app Purchase, 인앱 결제), Paid App(유료 앱), Subscription(구독) 등이 있습니다. 최근에는 Free-to-Play(F2P) 전략이 두드러지며 인앱 광고 및 인앱 결제 모델이 주로 사용되고 있으며, 두 모델을 혼합하여 사용하여 하이브리드 수익 모델을 구축하는 경우도 많습니다.
### IAA(In-app Ads, 인앱 광고)
인앱 광고 모델에서는 게임사들이 플레이어들에게 광고를 송출하면서 광고 수익을 얻는 구조로 이루어집니다. 플레이어들은 게임의 모든 기능을 무료로 이용할 수 있지만, 플레이하거나 아이템을 얻기 위해서 광고를 봐야 하는 경우가 있습니다. 이때 게임사들이 광고를 송출할 수 있는 공간인 인벤토리 생성 및 광고 공급을 도와주는 **SSP(Supply-Side Platforms)**와 **광고 미디에이션(Ad Mediation)** 플랫폼이 있습니다.
SSP는 퍼블리셔를 다양한 애드 네트워크(Ad Network), DSP(Demand-Side Platforms) 등에 연결하며, 게임 안에 배너, 동영상, 인터랙티브, 오퍼월 등 다양한 형태의 광고 인벤토리를 제공하도록 도와줍니다. 앱러빈(AppLovin), 아이언소스 탭조이(ironSource Tapjoy), 민티그럴(Mintegral), 유니티 애즈(Unity Ads) 등이 있습니다. 또한, 광고 미디에이션은 여러 애드 네트워크와 연동하여 하나의 SDK로 더 많은 광고 소스에 접근할 수 있고 광고 수익을 극대화할 수 있는 기능을 제공하는 플랫폼입니다. 앱러빈의 맥스(Max) 및 아이언소스의 레벨플레이(LevelPlay), 구글 애드몹(Google AdMob) 등이 있습니다.
### IAP(In-app Purchase, 인앱 결제)
게임 내에서 새로운 아이템 및 캐릭터, 게임 화폐를 구매하거나, 광고를 보지 않기 위해 구독 결제를 하는 등 모바일 게임 앱 내에서 결제가 필요한 경우가 있습니다. 이와 같은 인앱 결제 시스템 구축을 위해 각 OS에서 제공하는 기능을 이용하거나, 혹은 크로스 플랫폼에서의 인앱 결제 시스템 빌딩 및 관리를 도와주는 Unity IAP, 인앱 구독 결제를 도와주는 RevenueCat 등의 솔루션을 활용할 수도 있습니다.
## 유저 획득 플랫폼(User Acquisition)
소셜 미디어 플랫폼 및 애드 네크워크에서 Paid 광고를 통해 유저 획득을 도와주는 플랫폼들로는 메타(Meta, 구 Facebook), 틱톡(TikTok), 트위터(Twitter), 구글(Google), 스냅챗(Snapchat), 몰로코(Moloco), 리프트오프(Liftoff), 앱러빈, 아이언소스, 유니티 애즈 등이 있습니다.
유저의 전환에는 광고 크리에이티브(Ad Creative)가 중요한 영향을 미치기 때문에 뉴주에서는 광고 크리에이티브 제작을 도와주는 툴/플랫폼도 주요 하위분류로 구분하였습니다.
## MMP / 모바일 어트리뷰션 솔루션(Mobile Attribution Solution)
또한 앱 내외에서 진행한 광고의 성과를 측정해 주는 에어브릿지와 같은 **MMP / 모바일 어트리뷰션 솔루션**도 게임 앱의 성장을 위해 필요한 솔루션입니다. 광고 매체 별 성과를 측정하여 가장 효과적인 매체에 집중하여 매체 전략을 기획하는 데에 도움이 될 수 있습니다.
MMP는 유저들이 어떤 채널에서 유입되었는지, 인앱 구매는 얼마를 발생시켰는지, 어떤 인앱 이벤트를 했는지 등에 대한 정보를 가지고 있습니다. 그렇기 때문에 IAA 모델을 운영하고 있는 모바일 게임의 경우 MMP를 통해 광고 수익에 대한 어트리뷰션을 분석하여 어떤 채널에서 Ad Whale(광고를 많이 보는 유저)이 주로 유입되는지 알 수 있습니다. 또한, IAA, IAP 및 다른 수익 모델을 함께 사용하고 있는 경우에는 MMP의 대시보드에서 각 모델에서 발생한 수익을 비교하거나 모든 수익을 합쳐서 ROAS를 계산할 수 있습니다.
이와 더불어 모바일 게임 앱은 다운로드를 위한 Paid 마케팅을 대대적으로 집행하는 만큼 효과적인 마케팅 비용 집행을 위해서는 프로드 방지(사기 설치 방지)도 중요합니다. 이에 MMP에서 제공하고 있는 프로드 방지 솔루션을 이용하는 경우도 많습니다.
## 고객 인게이지먼트(Customer Engagement)
모바일 게임 앱이 지속 가능하기 위해서는 유저 획득 비용(CAC, Customer Acquisition Cost)은 낮추고 리텐션과 고객의 LTV(Customer Lifetime Value, 고객 생애 가치)는 높여야 합니다.
어렵게 획득한 유저를 성공적으로 온보딩 시키거나 리텐션과 LTV를 높이기 위한 푸시 알림 등을 CRM을 자동화해주는 솔루션 브레이즈(Braze), 음성 채팅 메신저 디스코드(Discord), 소셜 미디어 레딧(Reddit), 앱 리타겟팅 전략을 도와주는 리머지(Remerge) 등 다양한 서비스 및 플랫폼의 도움을 받을 수 있습니다.
## 시장 분석(Market Analysis)
이외에도 **시장 분석**을 도와주는 Data.ai, 뉴주, 센서타워(SensorTower), 모바일액션(MobileAction) 등과 같은 플랫폼들을 활용하여 업계 동향을 분석하는 데에 도움받을 수 있습니다. 뉴주에서 GameTech Ecosystem을 발행하거나, Data.ai에서 State of mobile 2022나 IDC와 협력한 [게임 스포트라이트 2022 리뷰](https://go.data.ai/2022_IDC_Download_KR.html)를 발행하는 등 다양한 인사이트를 제공해 주고 있습니다.
또한 Pocket Game, TouchArcade, Gamezebo, 게임메카, 게임동아, 경향게임스, 더게임스데일리 등 게임에 특화된 언론들을 통해 유저 및 시장의 반응을 살펴보는 것도 좋습니다.
## 모바일 게임 생태계의 주요 플레이어와 함께 성장해 보세요.
지금까지 살펴본 것처럼 모바일 게임 생태계는 복잡하고 다채롭습니다. 게임을 효율적으로 개발할 수 있도록 다양한 개발 툴의 도움을 받을 수도 있고 더 많은 유저를 획득하여 성장을 도와줄 수 있는 다양한 플랫폼과 툴들도 존재합니다. 하나의 게임이 성공하기 위해서는 잘 만들어지고 흥미로운 게임이라는 전제 조건이 필요하지만, 이를 플레이할 유저에게 다가가기 위한 마케팅 및 수익화가 뒷받침되어야 지속 가능한 게임으로 성장해갈 수 있습니다. 이번 포스트에서 살펴본 모바일 게임 생태계 속 주요 플레이어들을 적절한 시기 및 필요에 따라 활용해간다면 지속 가능한 성장에 도움받을 수 있습니다.
### 참고자료) Newzoo Gametech Ecosystem
[__wf_reserved_decorative]
---
## 마케터라면 반드시 브랜디드 링크를 사용해야 하는 이유
URL: https://www.airbridge.io/ko/blog/airbridge-branded-link
Language: Korean
Published: 2022-08-19
‘모든 비즈니스는 브랜딩이다’라는 서적이 있을 정도로 브랜딩은 중요합니다. 이미 많은 마케터들도 장기간에 걸쳐서 브랜드에 대한 인지도를 높일 수 있는 브랜딩의 중요성은 인지하고 있을 겁니다. 하지만 빠른 시간 내에 성과를 내기 위해, 브랜딩보다는 퍼포먼스 마케팅에 집중할 수밖에 없는 상황이 펼쳐지기도 합니다. 브랜딩과 퍼포먼스 마케팅은 우위를 따질 것 없이 모두 중요하다는 것을 알고 있음에도 말이죠. 하지만 단기간의 목표를 달성하기 위해 퍼포먼스 마케팅을 진행하는 과정에서, 이와 동시에 브랜드를 알리는 노력도 함께 수행되어야 합니다.
어떻게 하면 브랜딩과 퍼포먼스 두 가지를 모두 고려한 마케팅을 진행할 수 있을지 고민되신다면, 이번 포스트를 통해** 브랜디드 링크(Branded Link)** **활용법**을 확인해 보세요. 브랜디드 링크와 함께라면 **Owned 미디어 및 Earned 미디어**에서 **링크** **브랜딩**은 물론, **성과 측정**까지 모두 한 번에 가능합니다.
## 브랜디드 링크가 브랜딩에 미치는 영향
### 브랜딩이 중요한 이유
우선 **브랜딩**이 중요한 이유를 다시 한번 짚어보겠습니다. [Forbes의 기사](https://www.forbes.com/sites/forbesagencycouncil/2021/03/24/the-importance-of-branding-in-business/?sh=2b401f6e67f7) 중 아래와 같은 내용이 있습니다.
> “비즈니스에서 브랜딩은 당신이 생각하는 것보다 더 중요합니다. 겉으로 봤을 때, 브랜딩은 브랜드 로고나 색과 같은 일부 요소라고 생각할 수 있겠지만, 당신의 브랜드는 비즈니스의 전체 아이덴티티와 마찬가지입니다. 브랜딩은 비즈니스에서 항상 중요한 역할을 차지해왔지만, 지금은 이전 어느 때보다 더 중요해졌습니다. 소셜 미디어의 등장과 함께, 소비자들은 매일 같이 새로운 브랜드에 노출되고 있기 때문입니다. 고객 입장에서는 수많은 옵션들을 가지고 비교 분석하여 최고의 제품을 찾을 수 있게 되었지만, 비즈니스 상황은 더 치열하고 어려워졌습니다.
오늘날 엄청나게 많은 경쟁자가 계속해서 등장하고 있으며, 비즈니스들은 이런 무리 속에서 눈에 띌 수 있는 특별한 방법을 찾아야만 합니다. 사람들의 관심을 사로잡기 위한 강력한 브랜드를 만드는 데에 노력을 쏟아야 하죠. 브랜딩을 올바르게 진행한다면, 여러분 브랜드에 대한 대중들의 인식을 바꿀 수 있는 기회를 갖게 될 겁니다. 이 중요한 기회를 놓치지 마세요.”
대부분의 마케터들은 이미 이와 같은 브랜딩의 중요성을 인식하고, 광고 캠페인 집행에서 여러 가지 브랜드 요소를 고려하고 있을 텐데요. 하지만 고객들이 직접 마주할 ‘링크’에 대한 브랜딩은 간과했을지도 모릅니다. **‘링크’에 브랜딩을 더하면, 작은 노력으로도 큰 효과를 누릴 수 있습니다.**
### 브랜디드 링크(Branded Link)
**‘링크(Link)’**는 문자 그대로 온라인상의 모든 웹사이트 및 앱을 ‘연결’하고 ‘접속’하게 만들어줍니다. 우리 몸의 중추신경과도 같이 온라인에서 연결과 발견을 돕는, 근본적이면서도 중요한 역할을 합니다. 이처럼 링크를 통한 발견이 중요하다는 사실은 마케터들도 인지하고 있기 때문에 링크를 축약하여 깔끔하게 만들거나, OG 태그(Open Graph Meta Tags, 오픈 그래프)를 통해서 미리 보기 이미지, 제목 및 설명 등을 관리해서 클릭과 인게이지를 유도하기도 합니다.
그 중에서도 더 근본적인 방법으로, 링크에 브랜드 요소를 더한 **‘브랜디드 링크(Branded Link)’**로 만들어서 겉으로 드러나는 링크의 룩 앤드 필(Look and Feel)을 관리하는 것이 중요합니다. 브랜디드 링크는 수많은 비즈니스와 브랜드들 사이에서 우리 브랜드를 눈에 띄게 만들 수 있으며, 우리 브랜드가 고객에게 다가갈 수 있도록 도와줄 수 있습니다. 소셜 미디어에 공유한 포스트의 링크나 레퍼럴 링크(Referral link)들이 시간이 흐르면서 자연스럽게 브랜드의 인지도를 높여줄 것이며, 곧 브랜드의 성장으로도 이어질 겁니다.
## 브랜딩과 어트리뷰션을 위한 트래킹이 가능한 에어브릿지의 브랜디드 링크
이러한 브랜디드 링크는 에어브릿지를 통해 간단하게 만들 수 있습니다. 에어브릿지에서 **Custom Channels***의 **트래킹 링크**를 생성할 때, 도메인, 경로, 오픈 그래프를 브랜드 별로 설정하기만 하면 됩니다.
👉 [에어브릿지 트래킹 링크 생성 방법 자세히 알아보기](https://help.airbridge.io/hc/ko/articles/900001037886)
에어브릿지의 트래킹 링크는 클릭 및 광고 노출과 관련된 유저의 활동 정보를 에어브릿지에 전달하는 역할을 함과 동시에, 딥링크로서 유저의 리다이렉트 경험을 부드럽게 만들어줍니다. 즉, 에어브릿지의 브랜디드 링크를 통해 브랜딩 강화는 물론 어트리뷰션 및 딥링크까지 모두 한 번에 가능합니다.
> * Custom Channels: 사용자의 필요에 따라 커스텀하게 만들어 사용하는 채널로서, 사전에 연동되지 않은 채널입니다. 보통 마케터들은 *자체 홈페이지, 페이스북 페이지, 자체 블로그, 이메일, 메신저 그룹*과 같은 **'Owned 미디어' **혹은 *친구 초대 링크, 콘텐츠 공유 링크*와 같은 **'Earned 미디어'** 채널들의 성과를 측정하기 위해 Custom Channel을 생성하게 됩니다.
* (참고) Integrated Channels: 에어브릿지와 사전에 연동된 채널로서, 사전에 정의된 구분값을 가지고 있습니다. 에어브릿지와 연동된 채널 목록은 아래 링크에서 확인할 수 있습니다.
- Integrated Channels 리스트: [http://abit.ly/integrated-channels](http://abit.ly/integrated-channels)
## 브랜디드 링크의 다섯 가지 활용법
브랜딩 및 성과 분석을 더 효과적으로 진행할 수 있는, 에어브릿지 브랜디드 링크의 다섯 가지 활용법을 소개합니다.
### 1. 커스텀 도메인 설정으로 브랜디드 링크 만들기
에어브릿지에서 Custom Channels의 트래킹 링크를 생성할 때 기본 도메인(abr.ge) 대신, 브랜드명이 들어간 자체 도메인을 사용해서 커스텀 도메인을 가진 트래킹 링크를 생성할 수 있습니다. 즉, 커스텀 채널의 트래킹 링크에 사용되는 도메인을 브랜드와 어울리도록 새롭게 설정할 수 있습니다. 이 기능을 활용하면 기존에 Abit.ly 혹은 bitly 등을 통해 URL을 축약했던 것 대신, 우리 브랜드명을 담은 깔끔하고 보기 좋은 URL로 생성할 수 있습니다.
[]
첫 번째 예시와 같이 기존 도메인에 서브 도메인을 생성해서 커스텀 도메인을 생성할 수 있으며, 혹은 두 번째와 세 번째 예시와 같이 새로운 도메인을 구입하여 사용할 수도 있습니다. 원하는 형태의 브랜드 링크가 있다면 해당 도메인을 구입 혹은 서브 도메인을 생성한 후, 에어브릿지에 등록 하기만 하면 됩니다.
이와 같이 브랜드명이 들어간 링크를 사용하면, 링크를 공유함과 동시에 브랜드명도 함께 공유하는 것이기 때문에 브랜드를 더 많이 노출하여 브랜드의 인지도를 높일 수 있습니다. 그리고 브랜디드 링크를 마주한 유저는 해당 링크가 브랜드의 이름을 건 양질의 콘텐츠로 연결될 것이라는 신뢰를 가지게 되어, 반응을 더 잘 유도할 수 있습니다. 게다가 해당 콘텐츠로부터 도움을 받거나 긍정적인 영향을 받았다면, 이후 해당 브랜드에 대해서도 더 좋은 기억을 갖게 될 것이며 브랜드에 대한 신뢰도 높여줄 수 있습니다.
(단, 커스텀 도메인 기능은 고객사에서 보유하고 있는 도메인 중, 현재 사용 중이지 않은 서브 도메인으로만 설정 가능합니다.)
### 2. 링크 숏 아이디 설정을 통해 브랜딩 효과 높이기
미리 커스텀 도메인을 설정했다면, 트래킹 링크 생성 시 ‘링크 숏 아이디’ 설정값을 변경할 수 있습니다. (설정값을 비워두면 랜덤으로 생성되며, 트래킹 링크 생성 후에는 변경이 불가능합니다.)
‘링크 숏 아이디’는 ‘슬러그(slug)’라고도 부르는 콘텐츠의 고유 주소이며, 링크와 연결된 캠페인 및 콘텐츠 목적에 맞게 설정할 수 있습니다.
[]
[]
이렇게 생성한 링크는, 연결되는 콘텐츠의 내용이 무엇인지 쉽게 이해할 수 있도록 도와줍니다. 이는 SEO(검색 엔진 최적화)에도 도움이 될 수 있는데, 링크 숏 아이디가 콘텐츠의 핵심을 담은 단어들이기 때문입니다. 검색 엔진에서 더 빠르게 정보를 읽고 해당 페이지를 찾으면서 정확한 검색 결과 노출에도 도움이 될 수 있습니다.
브랜디드 링크 및 링크 숏 아이디를 효과적으로 사용할 수 있는 사례들은 아래와 같습니다.
- 온라인 교육 콘텐츠 회사: 특정 강사 및 강의에 대한 광고 캠페인에서 ‘브랜디드 링크’를 사용하고자 할 때, 도메인에는 회사명을 기입하고 링크 숏 아이디에는 해당 광고와 연결된 강사명 또는 강의명을 기입합니다. 이를 통해 특정 강사 및 강의 자체를 하나의 브랜드로 만들어서 광고 캠페인의 효과를 높일 수 있습니다.
- 온라인 쇼핑몰 회사: 특정 상품을 광고하는 캠페인에서 ‘브랜디드 링크’를 적용하여, 도메인에는 회사명을 넣고 링크 숏 아이디에는 해당 상품명을 넣어서 회사 및 상품에 대한 브랜딩을 강화할 수 있습니다.
### 3. 에어브릿지의 강력한 딥링크 기능을 통해 올바르게 트래킹 경로를 설정하고, 유저에게 일관된 경험 제공으로 브랜딩 효과 높이기
앞서 언급한 것처럼 에어브릿지의 브랜디드 링크는 트래킹 링크이자 딥링크로서 유저의 리다이렉트 경험을 부드럽게 만들어줄 수 있습니다. 디바이스별 유저를 랜딩 시킬 최종 목적지를 다르게 설정할 수 있으며, 딥링크 설정으로 원하는 앱 내 특정 콘텐츠에 바로 연결시킴으로써 더 나은 UX를 제공하고 전환율을 확보할 수 있죠. 이러한 설정은 유저에게 일관된 경험을 제공하여 브랜딩도 강화할 수 있습니다.
에어브릿지의 트래킹 경로는, 유저가 링크를 클릭하고 난 후 연결할 최종 목적지로 앱 마켓, 앱(딥링크), 웹사이트 중 하나로 설정할 수 있습니다.
- 앱 마켓: 앱 다운로드 유도를 위해 유저의 최종 목적지를 앱 마켓으로 설정하고 싶은 경우
- 앱의 딥링크: 유저의 최종 목적지를 앱 내 특정 화면이나 콘텐츠로 설정하고 싶은 경우
- 웹사이트: 유저의 최종 목적지를 웹사이트 URL로 설정하고 싶은 경우 (Android / iOS / Desktop 환경에서 각각 다른 URL 경로로 설정할 수 있습니다.)
### 4. OG 태그를 커스텀 설정하여 소셜 미디어에서 공유된 링크에 브랜딩 강화하기
OG 태그(Open Graph Meta Tags)는 소셜 미디어에 링크를 공유할 때 아래와 같은 미리 보기의 제목, 설명, 이미지를 결정하는 메타 데이터입니다.
[]
에어브릿지에서 생성한 트래킹 링크가 각 소셜 미디어 플랫폼 상에서 어떻게 공유될지 ‘소셜 미디어 썸네일’에서 미리 확인하고 원하는 형태로 수정할 수 있습니다. 상단의 이미지와 같이 오픈 그래프 태그(OG Tag)를 수정함으로써 개발자의 도움 없이도 링크 목적에 맞는 미리 보기 상태로 노출될 수 있도록 관리할 수 있습니다.
유저는 이 미리 보기 화면을 통해 연결된 콘텐츠를 직관적으로 이해할 수 있게 되어, 시선을 사로잡고 링크 클릭률을 높이는 데에 도움이 됩니다. 연결된 링크 속 콘텐츠의 핵심을 설명하기 때문에 SEO(검색 엔진 최적화)에도 도움이 될 수 있으며, 브랜드에 대한 신뢰도와 인지도를 높여줄 수 있습니다.
> (참고) Custom Channels의 트래킹 링크를 생성하면, QR 코드도 함께 제공됩니다. QR 코드를 활용한 마케팅 팁이 궁금하다면 아래 포스트에서 확인해보세요.
👉[ 성공적인 마케팅으로 이어지는 QR 코드 활용 팁 네 가지 알아보러 가기](https://blog.ab180.co/posts/airbridge-qr-code)
### 5. 에어브릿지 브랜디드 링크로 광고 캠페인 성과 측정하고 개선하기
위와 같이 브랜디드 링크를 광고 캠페인에 활용했다면, 이제 다음 단계로는 광고 캠페인의 성과를 측정해야 할 텐데요.
에어브릿지의 브랜디드 링크는 앞서 언급한 것처럼 **트래킹 링크**이기 때문에, **해당 링크를 통해 접속한 유저들의 행동 및 해당 광고의 효과 등을 확인할 수 있습니다**. 트래킹 링크인 브랜디드 링크를 통해, 클릭 및 광고 노출과 관련된 데이터가 에어브릿지로 전달되기 때문입니다. 에어브릿지는 이렇게 수집한 데이터를 처리하고 가공하여, 다양한 리포트들을 통해서 광고주들을 위한 정보를 제공해 줍니다. 트래킹하고 있는 데이터는 통계 리포트인 Actuals Report와 Trend Report 등을 통해 자유롭게 확인할 수 있습니다.
## 마치며
지금까지 에어브릿지 브랜디드 링크의 중요성부터 역할 및 기능, 그리고 활용법까지 알아보았습니다. 포스트 초반에 다룬 Forbes 기사의 내용처럼 ‘브랜드는 비즈니스의 전체 아이덴티티와 마찬가지’이기 때문에, 브랜디드 링크만으로 브랜딩에 필요한 모든 것을 해결할 수 있는 건 아닙니다. 브랜디드 링크가 브랜딩의 전부는 아니지만, 브랜디드 링크는 브랜딩에 효과적입니다. 게다가 트래킹 링크이기 때문에 성과 측정까지 한 번에 가능하다는 것도 강점입니다. 그렇기 때문에, 마케팅 캠페인에서 필수인 ‘링크’를 사용해야 한다면, 에어브릿지의 브랜디드 링크로 브랜딩 강화는 물론 성과 측정까지 한 번에 해결해보세요!
---
## [딥링크101] 마케터와 개발자를 위한 딥링크 시작하기
URL: https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers
Language: Korean
Published: 2022-08-17
[__wf_reserved_decorative]
‘링크’라고 하면, 아래 이미지와 같은 **URL**을 떠올리게 됩니다.
[__wf_reserved_decorative]
링크를 클릭하면 연결된 웹사이트로 이동하죠. 모바일에서 ‘딥링크'도 마찬가지입니다.
**딥링크**란 모바일 환경에 놓인 유저가 웹사이트 혹은 다른 앱에서 URL을 클릭했을 때, 앱을 실행시키고 특정 페이지로 이동하도록 돕는 기술입니다. 다시 말해, 유저가 광고를 클릭했을 때 앱을 열어 특정 페이지로 바로 도달하도록 만들어 줍니다. 유저가 하나의 링크를 클릭하는 것만으로도 앱 내부 콘텐츠를 바로 확인할 수 있죠. 이는 모바일 앱 마케팅을 할 때 반드시 필요한 기술로, 유저 경험을 최적화하여 전환율 및 리텐션을 높일 수 있습니다.
하지만, 딥링크 기술은 계속해서 변화하고 있으며, 유저가 딥링크를 호출하는 환경 및 방법에 따라 때때로 잘 구현되지 않는다는 문제가 발생하기도 합니다. 전환율 및 유저 리텐션에 중요한 영향을 미치는 딥링크, 어떻게 하면 문제없이 구현할 수 있을까요? 그 시작점으로서, 이번 포스트에서는 **딥링크의 역할, 그리고 URI Scheme 딥링크 및 Universal Links / App Links**에 대해 알아보겠습니다.
> 딥링크에 대한 더 자세한 내용이 궁금하시다면, <[에어브릿지 핸드북 - 딥링크](https://www.airbridge.io/ko/library/airbridge-handbook-deeplinks)\> 가이드를 확인해보세요.
## 딥링크가 필요한 이유
앱 마케팅을 하고 있다면, 딥링크의 중요성과 필요성에 대해 아무리 강조해도 모자라지 않습니다. 아래는 에어브릿지(Airbridge) 고객사인 펫프렌즈의 딥링크 사용 예시입니다.
- 웹페이지(Safari)에서 ‘APP 설치하기’를 눌렀을 때 앱이 설치되어 있는 경우, 우측과 같이 바로 앱이 오픈되며,
- 웹페이지(Safari)에서 ‘APP 설치하기’를 눌렀을 때 앱이 설치되어 있지 않은 경우, 좌측과 같이 앱 마켓(App Store)의 상세 페이지로 연결됩니다.
- Youtube에서 딥링크를 눌렀을 때 앱이 설치되어 있는 경우, **‘펫프렌즈’(앱)에서 열겠습니까?**라는 프롬프트(Prompt)가 뜬 후, ‘열기’를 누르면 앱으로 바로 연결됩니다.
- Youtube에서 딥링크를 눌렀을 때 앱이 설치되어 있지 않은 경우, 앱 마켓(App Store)의 상세 페이지로 연결됩니다.
[]
[]
유저가 웹에서 딥링크를 클릭했을 때 앱이 이미 설치되어 있음에도 앱 마켓으로만 연결된다거나, 특정 상품 광고를 클릭하여 앱이 열렸을 때 상품 페이지가 아닌 메인 화면으로만 연결된다면 유저는 이탈해버릴 겁니다. 어렵게 획득한 유저를 붙잡기 위해서는 제대로 된 목적지로 보내주는 ‘딥링크’가 반드시 필요합니다. 딥링크는 매끄러운 유저 경험을 제공하여 전환율과 리텐션을 높여줄 수 있죠. 웹과 앱 간의 모든 파편화된 여정에 있어서 유저가 훨씬 쉽게 이동할 수 있도록 만들어 줍니다.
## 딥링크의 세 가지 유형
이처럼 앱 마케팅에서 중요한 역할을 하는 딥링크에는 **URI Scheme, Universal Links, App Links**의 세 가지 표준 기술 유형이 있습니다. 모두 유저를 앱 내 특정 페이지로 이동시키는 딥링크 기능을 수행하긴 하지만, 딥링크 설정 방식 및 개발 주체에 따라 구분하고 있습니다. 즉, 전통적인 딥링크 방식(URI Scheme), 그리고 이 방식의 한계를 보완하기 위해 [iOS(Universal Links)](https://developer.apple.com/ios/universal-links/)와 [Android(App Links)](https://developer.android.com/training/app-links) 각각에서 제시한 해결책입니다. 이처럼 다양한 유형의 딥링크로 인해 앱 개발자는 물론 딥링크를 사용하고 싶은 마케터들에게도 혼란을 불러일으키곤 합니다. **딥링크를 제대로 사용하려면 딥링크의 유형 및 작동 방식에 대해 먼저 이해하는 과정이 필요합니다.**
[__wf_reserved_decorative]
### 1) URI Scheme(스킴 딥링크)란?
URI Scheme은 딥링크의 가장 초기 형태로, 가장 일반적으로 사용되는 딥링크 방식이라고 할 수 있습니다. 앱을 개발하는 주체가 각 앱 내 특정 페이지마다 고유한 주소(링크)를 자유롭게 설정하여, 해당 주소(링크)를 클릭하면 앱이 열리고 특정 페이지가 열리는 형태입니다. 인터넷에서 흔히 사용하는 URL 주소처럼 링크를 클릭하면 특정 웹사이트가 열린다고 생각하면 쉽습니다. 이때 URL은 대개 http:// 혹은 https:// 로 시작된다면, URI Scheme은 각 모바일 앱에서 지정한 Scheme 값으로 시작합니다.
\[참고\]
- [URI](https://developer.mozilla.org/ko/docs/Glossary/URI)(Uniform Resource Identifier, 통합 자원 식별자): 리소스(resource, 자원)을 식별하는 문자열을 의미합니다. URI의 하위 개념으로 URN과 [URL](https://developer.mozilla.org/ko/docs/Glossary/URL)이 있습니다.
- URL(Uniform Resource Locator): 인터넷에서 웹 페이지, 이미지, 비디오 등 리소스의 위치를 가리키는 문자열입니다.
더 이해를 돕고자, URI Scheme의 예시 링크를 통해 살펴보겠습니다.
[]
여기서 **airbridge://** 부분이 Scheme(혹은 프로토콜)입니다. 모바일 앱은 자신만의 Scheme 값을 등록할 수 있으며, 이를 통해 앱을 구분합니다. 앱 개발자가 우리 앱에 이 Scheme을 쓸 것이라고 결정하는 겁니다.
이어서, 특정 페이지에 도달하도록 만들기 위한 path인 **page/ko**, parameter(파라미터 / 쿼리스트링)인 **?product=1**, 그리고 리소스 자체의 다른 부분을 가리키는 앵커(북마크)인 **#content**로 구성되어 있습니다. ([MDN에서 URL 구문 더 알아보기](https://developer.mozilla.org/ko/docs/Web/HTTP/Basics_of_HTTP/Identifying_resources_on_the_Web#uniform_resource_identifiers_\(uris\)_%EA%B5%AC%EB%AC%B8))
- Scheme 값: 앱 개발자가 정한 값으로 앱을 구분
- Path: 앱 내 특정 페이지를 지정
### URI Scheme의 한계와 Universal Links 및 App Links의 탄생
이러한 URI Scheme 방식의 딥링크는 앱 개발자가 쉽게 등록하고 설정할 수 있기 때문에 널리 사용되었습니다. 다만, 점차 앱의 수가 증가하면서 **소유권을 증명할 수 없다**는 문제로 인해, Scheme 값이 중복되는 경우가 발생하게 되었습니다. 앱 개발자가 자신만의 Scheme 값을 자유롭게 설정할 수 있기 때문에 고유(unique)한 URI를 점유하기 어려워진 겁니다. 예를 들어, 2개의 앱이 중복된 Scheme을 가질 경우, 둘 중 하나의 딥링크는 하이재킹되어 유실될 수 있습니다. path 등까지 모두 동일한 주소를 사용한다면, Android의 경우 딥링크로 오픈할 앱 선택창이 노출되고, iOS는 가장 마지막에 설치한 앱이 자동으로 열립니다.
각 앱들이 Scheme 값을 공개적으로 드러내지 않기 때문에, 각 개발자들은 자신의 Scheme 값이 고유한지 확인할 수 있는 방법은 없는 상황입니다. 이러한 문제점에 대해, 애플과 구글 각각에서 해결 방안을 내놓은 것이 **Universal Links(iOS 제공)와 App Links(Android 제공)**입니다.
### 2) Universal Links(유니버셜 링크) & App Links(앱 링크)란?
Universal Links와 App Links는 각각 iOS와 Android에서 제공되는 딥링크 형태로, OS에 앱에 대한 도메인 주소를 등록함으로써 소유권을 증명하는 겁니다.
**도메인(Domain)**은 각 웹사이트마다 가지고 있는 고유한 인터넷 주소로, **http://www.airbridge.io/ko** 라는 주소가 있다면, 이 중, **www.airbridge.io** 부분이 도메인입니다. 특정 도메인은 유니크하고, 도메인 소유자만이 관리할 수 있기 때문에 소유권이 증명되며 보장됩니다.
앱 빌드 시 앱을 열 도메인을 등록하고, 보안을 위해 등록한 도메인의 특정 경로에 인증 텍스트를 심어 놓습니다. 그러면 앱이 설치된 후 OS 레벨에서 해당 도메인 경로의 인증 텍스트와 앱 빌드 시의 값이 동일한지 검증하여 소유권을 인지하게 됩니다. 일단 검증이 완료된 후에는 앱을 삭제하기 전까지 해당 검증은 유효합니다.
검증된 도메인으로 생성된 딥링크 클릭 시,
- **앱이 설치된 경우**에는 앱 및 특정 페이지가 바로 열리고
- **(앱이 설치되어 있지 않아) 딥링크 실패 시**에는, 해당 도메인의 웹페이지로 이동합니다. 이때, 해당 도메인을 관리하는 개발자의 구현에 따라서 웹페이지에서 다양한 Fallback이 가능하기 때문에, 웹사이트 혹은 앱 마켓으로 보내는 것도 가능합니다.
Universal Links 및 App Links의 예시 링크는 아래와 같습니다.
[]
## Android 및 iOS 딥링크 설정 방법
세 가지 딥링크 유형에 대해 이해했다면, 다음은 OS별로, URI Scheme 및 App Links(Android), Universal Links(iOS)를 설정하는 방법을 알아보겠습니다.
### Android에서 딥링크 설정하기
Androidmanifest.xml 파일 내에서 앱 내 각 Activity 마다 해당 Activity를 열 수 있는 딥링크를 설정합니다. 이때, 각각의 Activity를 설정하기보다는, 비어있는 하나의 Activity에서 여러 경로의 딥링크가 열리도록 구현해 놓은 후, 그 Activity의 화면에서 딥링크 URL 값에 따라 각각의 Activity로 이동시키도록 구현하는 경우가 많습니다.
Androidmanifest.xml 파일 내에는 앱 정보(패키지명, 버전 코드 등), 앱에서 사용할 권한, 앱에서 사용되는 컴포넌트(Activity, Service, BroadcastReceiver, ContentProvider)가 포함됩니다.
[__wf_reserved_decorative]
URI Scheme 및 App Links를 설정할 때는, 상단의 이미지와 같이, AndroidManifest.xml > activity > intent-filter 내에 해당 Activity를 열도록 허용할 Scheme 또는 Host 등을 등록합니다.
- [에어브릿지 개발자 가이드에서 Android 딥링크 설정하는 방법 자세히 알아보기](https://developers.airbridge.io/docs/android-sdk#%EB%94%A5%EB%A7%81%ED%81%AC-%EC%84%A4%EC%A0%95)
### iOS에서 딥링크 설정하기
iOS에서는 하나의 화면으로 딥링크가 열리도록 설정하고, 이후 해당 화면에서 각각의 상세 페이지로 이동하도록 구현하는 방법만 지원합니다.
**URI Scheme을 설정할 때는**, 아래 이미지와 같이 Xcode > Project 파일 > Info > URL Types에서 iOS URI Scheme을 입력합니다.
[__wf_reserved_decorative]
**Universal Links의 경우,** 아래의 이미지와 같이, Xcode > Project 파일 > Signing & Capabilities 로 이동 후 + Capability 를 눌러 Associated Domains 에 Universal Links로 설정하고 싶은 도메인을 추가하면 됩니다.
[__wf_reserved_decorative]
- [에어브릿지 개발자 가이드에서 iOS 딥링크 설정하는 방법 자세히 알아보기](https://developers.airbridge.io/docs/ios-sdk#%EB%94%A5%EB%A7%81%ED%81%AC-%EC%84%A4%EC%A0%95)
## 딥링크를 앱 마케팅을 위해 사용할 때 유의해야 할 부분
이렇게 소유권이 보장된 딥링크라고 하더라도, 앱 마케터가 원하는 완전한 형태의 딥링크라고는 할 수 없습니다. 딥링크를 앱 마케팅을 위해 사용할 때는 **URI Scheme 및 Universal Links, App Links를 앱 및 브라우저 별로 다르게 설정해두어야 한다**는 점을 유의해야 합니다.
Universal Links 및 App Links의 경우, 각각의 OS에서 제공하는 딥링크 형태이기 때문에, 각 OS의 브라우저인 Safari(iOS) 및 Chrome(Android)에서는 딥링크가 잘 동작할 수 있도록 설정되어 있습니다. Safari의 경우, URI Scheme의 동작은 제한되어 있기도 합니다.
그리고 이외의 앱 및 브라우저는, URI Scheme 만 동작하는 경우도 있고 혹은 Universal Links 및 App Links가 동작하긴 하더라도 한계를 가진 경우도 있습니다. 이는 각 앱 및 브라우저의 개발자가 OS 레벨에서 지원하는 API를 통해 얼마나 적절하게 구현해 놓는가에 따라 달라지기 때문입니다. 세 유형의 딥링크들의 대표적인 앱 및 브라우저에서 상황에 따른 동작 여부는 아래의 표에서 간단하게 확인할 수 있습니다.
아래의 표는 Universal links 및 App Links(https://www.airbridge.io) 또는 URI scheme(airbridge://)을 사용하여 딥링크를 동작시켰다고 가정하고 우리나라에서 주로 사용하는 앱 또는 브라우저들에서 어떤 동작이 발생하는지 정리하였습니다. 이때, Universal Links 혹은 App Links를 사용함에도 **웹으로 랜딩이 된다는 것은 제대로 딥링크 기능이 작동하지 않았다**는 것을 의미합니다.
[__wf_reserved_decorative]
또한, Universal Links의 경우 아래와 같은 상황도 염두해야 합니다.
- Universal Links는 주소 입력창에 링크를 직접 복사 붙여 넣기 할 경우 작동하지 않습니다.
- Universal Links는 JavaScript로 트리거 될 경우 리다이렉트가 되지 않습니다.
- Universal Links는 openUrl과 같이 앱 내에서 프로그래밍 방식으로 링크를 열 경우, 작동하지 않습니다.
이처럼 각 앱 및 브라우저마다 구현할 수 있는 딥링크의 유형 및 조건이 다르기 때문에, 각 유형별 딥링크가 앱을 열 수 있도록 각 앱 및 브라우저마다 다른 딥링크를 적용하도록 설정을 살펴보고 테스트 해야하는 번거로움이 발생합니다. 초기 형태의 딥링크 방식인 URI Scheme을 사용하고 싶지 않더라도, 각 앱 및 브라우저의 설정에 따라 반드시 사용해야 하는 경우가 발생할 수도 있죠.
이외에도, URI Scheme의 경우, 앱이 설치되어 있지 않을 때, 다른 곳으로 이동(Fallback)이 불가능하다는 점도 염두해야 합니다. 즉, 아무런 작동이 발생하지 않습니다.
> 👉 URI Scheme 딥링크를 매끄럽게 사용할 수 있는 방법은, [\[딥링크101\] iOS Safari에서 딥링크 동작 시 경고 문구가 뜨는 이유가 무엇일까요?](https://www.airbridge.io/ko/blog/deeplink-101-ios-safari-alert) 포스트에서 확인해보세요!
앱 마케팅을 할 때는 앱 다운로드를 통한 유저의 유입 및 유저 획득이 중요한데, 웹사이트로 랜딩 된 유저 혹은 딥링크가 작동하지 않은 유저는 이탈할 가능성이 높아지게 됩니다.
## 앱 마케팅에 딥링크를 문제없이 사용할 수 있는 방법
그렇다면, 어떻게 하면 앱 마케팅에 딥링크를 문제 없이 사용할 수 있을까요?
우선, 세 가지 유형의 딥링크에 대한 이해를 바탕으로 딥링크를 제대로 설정하는 것부터 시작해야 합니다. 앱에 도착한 유저를 앱 내 특정 페이지로 랜딩시키는 것은 앱의 역할이기 때문에, 딥링크 자체는 앱에서 반드시 먼저 설정되어야 합니다.
그러고 나서, 앞서 언급한 딥링크를 활용할 때 유의해야할 점은 에어브릿지와 같은 어트리뷰션 툴/MMP의 **트래킹 링크**를 활용하여 해결하는 겁니다. 어트리뷰션을 위한 터치포인트 데이터를 수집하는 **트래킹 링크**에 딥링크를 설정하는 방식으로, 트래킹 링크를 고도화하여 사용할 수 있습니다.
구체적으로는 1) 유저가 어떤 환경 및 방법으로 딥링크를 호출하였는지 등에 따라서 작동 여부 및 흐름이 달라질 수 있다는 문제를 해결할 수 있습니다. 다양한 상황을 모두 테스트하거나 설정을 변경하는 등의 과정 없이, 유저를 앱까지 잘 도착하도록 만들 수 있습니다.
2) 또한, 앱이 설치되지 않았을 때 OS(Android, iOS) 및 Desktop에 따라 Fallback을 각각 다르게 설정하는 과정도 MMP의 대시보드 상에서 쉽게 가능합니다. URI Scheme만 사용할 경우엔 앱이 설치되지 않은 경우 다른 곳으로 이동(Fallback)이 불가하다는 문제도, 트래킹 링크를 통해 해결할 수 있습니다.
[__wf_reserved_decorative]
3) 게다가 앱이 설치되지 않은 유저가 앱을 설치한 후 앱을 실행할 때까지 딥링크를 지연시키는 **디퍼드 딥링크**(Deferred Deep Linking, 지연된 딥링크) 기술은 Android OS 레벨 및 MMP레벨에서만 지원하기 때문에 이 기능을 활용하기 위해서는 MMP가 필요하죠.
다만, 이때 앱에 도착한 유저를 앱 내 특정 페이지로 랜딩 시키는 것은 앱의 역할이라는 점은 반드시 염두 해야 합니다. 에어브릿지와 같은 MMP는 앱에 설정된 딥링크의 설정값을 받아서 실행하는 역할을 수행할 뿐이기 때문에, 딥링크 자체는 앱에서 반드시 먼저 설정되어야 합니다.
유저를 앱 내 특정 페이지로 부드럽게 연결시키는 딥링크 기술. 에어브릿지를 통해 유저 여정을 더 완벽하게 만들어보세요!
> **\* 다른 딥링크101 시리즈 보러가기**
> 👉[\[딥링크101\] iOS Safari에서 딥링크 동작 시 경고 문구가 뜨는 이유가 무엇일까요?
> ](https://www.airbridge.io/ko/blog/deeplink-101-ios-safari-alert)👉[\[딥링크101\] 디퍼드 딥링크로 앱을 설치하지 않은 유저도 앱 내 특정 페이지로 연결하기](https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking)
---
## 모바일 게임 카테고리에 따른 특징 이해하기
URL: https://www.airbridge.io/ko/blog/mobile-game-genres
Language: Korean
Published: 2022-08-10
모바일 앱을 크게 두 개의 카테고리로 분류하면 게임과 비게임 앱으로 나눌 수 있을 만큼, 게임 앱이 차지하는 비중이 높습니다. 게다가 [2020년 기준 구글 플레이에서 발생하는 모바일 앱 수익 중 게임 앱이 차지하는 비중이 83%](https://www.statista.com/statistics/1010710/google-play-app-revenue-share-by-category-worldwide/)일 정도로 모바일 앱 시장에서 발생하는 수익의 근원은 대부분 게임 앱입니다. 이렇게 모바일 앱 시장에서 중요한 비중을 차지하는 게임 앱 중에서도 어떤 카테고리가 주목받고 있는지, 그리고 주요 카테고리별 특징은 어떤지 살펴보겠습니다.
## Google Play 및 App Store의 게임 카테고리 및 태그 분류
주요 앱 마켓인 Google Play 및 App Store에서 분류하고 있는 게임 카테고리를 살펴보겠습니다.
- **Google Play**: 액션, 어드벤처, 아케이드, 보드, 카드, 카지노, 하이퍼 캐주얼, 캐주얼, 교육, 음악, 퍼즐, 자동차 경주, 롤플레잉, 시뮬레이션, 스포츠, 전략, 퀴즈, 단어 등
- **App Store**: 액션, 어드벤처, 보드, 카드, 가족, 음악, 퍼즐, 레이싱, 롤플레잉, 시뮬레이션, 스포츠, 전략, 단어, 인디, 캐주얼, 퀴즈 등
이외에도 모바일 앱 분석 툴 [data.ai](https://www.data.ai/kr/)에서도 액션, 테이블톱(카드 및 보드 게임), 카지노, 하이퍼 캐주얼, 시뮬레이션, 키즈, 스포츠, 슈팅, 레이싱, 매치, 액션, RPG 등의 장르로 나누고 하위 장르 및 태그로 분류하고 있습니다.
## 지금 세계에서 가장 주목받는 게임 카테고리
data.ai에서 발행한 [State of Mobile 2022 리포트](https://www.data.ai/en/go/state-of-mobile-2022/)에 따르면, 다운로드 수로 본 세계에서 가장 주목받고 있는 게임 장르는 액션(하이퍼 캐주얼), 퍼즐(하이퍼 캐주얼), 시뮬레이션, 키즈, 드라이빙(시뮬레이션) 등의 순입니다. 그리고 모바일 게임에서 가장 수익성이 우수한 장르의 경우, 4X March-배틀(전략), MMORPG(RPG), 팀 배틀(RPG), 슬롯(카지노), M3 Meta(매치) 등의 순입니다.
또한, 페이스북에서 발행한 [Games Marketing Insights for 2022](https://www.facebook.com/fbgaminghome/marketers/find-new-players/advertising-hub/gaming-marketing-insights-2022) 리포트에 따르면, 다운로드 수에 따른 선호 장르는 미국은 하이퍼 캐주얼, 시뮬레이션, 액션 순이었으며, 한국은 하이퍼 캐주얼, RPG, 시뮬레이션 순입니다.
### 하이퍼 캐주얼 게임의 특징
하이퍼 캐주얼 게임은 조작이 쉽고 간단한 게임으로, 성별과 나이, 국적을 불문하고 게임을 즐길 수 있다는 특징을 가지고 있습니다. 그렇기 때문에 다른 장르에 비해 넓은 오디언스와 좋은 접근성을 가지고 있습니다. 누구나 쉽게 접근할 수 있는 만큼 리텐션이 낮을 수 있지만, 넓은 오디언스로 인해 수익화로 이어질 수 있는 기회가 크며 다른 비슷한 게임 앱으로 유저를 유도하기에도 유리합니다.
[__wf_reserved_decorative]
앞서 언급했듯이 data.ai의 State of Mobile 2022에서 발표한 가장 다운로드 수가 많은 게임 장르도 하이퍼 캐주얼(액션과 퍼즐)이었으며, 페이스북의 Games Marketing Insights for 2022에서 조사한 미국 및 한국에서 가장 많은 다운로드 수를 보이는 장르도 하이퍼 캐주얼 게임입니다. 게다가 2021년 3분기 전 세계 다운로드 수 기준 상위에 오른 모바일 게임 퍼블리셔 10곳 중 8곳이 하이퍼 캐주얼 게임에 중점을 두고 있을 정도로 유저 및 퍼블리셔에게도 주목받고 있는 카테고리입니다.
조금 더 구체적으로 성별 및 연령에 따른 특징을 살펴보면, 여성 게이머가 퍼즐(하이퍼 캐주얼)을 플레이할 확률이 일반 대중에 비해 1.21배 더 높으며, Z세대(16~24세)가 퍼즐(하이퍼 캐주얼)을 플레이할 확률이 일반 대중에 비해 1.6배 더 높습니다. 한국에서는 여성 게이머가 퍼즐을 플레이할 확률이 일반 대중에 비해 1.13배 더 높으며, Z세대가 퍼즐을 플레이할 확률이 일반 대중에 비해 1.74배 더 높습니다. 또한, 프랑스, 일본, 영국, 미국에서 액션(하이퍼 캐주얼) 게임을 플레이하는 유저는 퍼즐(하이퍼 캐주얼) 게임도 같이 플레이하는 경우가 많으며, 한국에서는 시뮬레이션 게임과 가장 많은 중복 유저를 공유하는 특징도 가지고 있습니다. (출처: [data.ai 성공적인 모바일 게임 전략 및 기능 로드맵 구축하기 - 2021년 4분기 에디션](https://www.data.ai/kr/go/create-a-winning-mobile-gaming-strategy/))
### 시뮬레이션 게임의 특징
시뮬레이션 게임은 현실과 비슷한 환경을 구현해 간접 체험을 누릴 수 있는 게임을 뜻합니다. 교육적인 목적으로 개발되는 경우도 있으나, 순수하게 오락을 위한 시뮬레이션 게임도 다수 존재합니다.
시뮬레이션 게임은 얼마나 현실을 사실적으로 반영했는가가 핵심입니다. 세부적으로는 놀이공원이나 농장을 운영하는 등의 건설/경영 시뮬레이션 게임, 피파 사커(FIFA Soccer)와 같은 스포츠 게임, 심즈(The Sims)와 같은 육성 시뮬레이션 게임, 연애 시뮬레이션 등의 서브 카테고리로 나눌 수 있습니다. 미국, 영국, 독일 및 한국에서 월간 실사용자 기준 Z세대(16~24세)가 플레이할 가능성이 높은 게임 1위에 오른 로블록스(Roblox)도 대표적인 시뮬레이션 게임 중 하나입니다.
Games Marketing Insights for 2022 자료에 따르면 미국, 영국, 독일 및 한국에서 다운로드 수에 따른 선호 장르로 시뮬레이션 게임이 모두 2~3위를 차지하고 있을 정도로 빠질 수 없는 주요한 게임 장르입니다.
[__wf_reserved_decorative]
#### RPG 게임의 특징
RPG 게임은 롤플레잉 게임(Role-playing games)으로, MMORPG(대규모 멀티 플레이어 온라인 롤플레잉 게임, Massive Multi-player Online Role Playing Game), 팀 배틀, 방치형 RPG 등으로 세분화할 수 있습니다.
Games Marketing Insights for 2022 자료에 따르면 한국은 다른 국가들과는 다르게 RPG 게임이 다운로드 수에 따른 선호 장르 2위에 위치합니다. 또한, 모바일 광고 플랫폼 팽글(Pangle)이 글로벌 시장조사업체 뉴주(Newzoo)와 발간한 [RPG Report: How Do Gamers Across East and West Engage with the Genre?](https://www.pangleglobal.com/resource/role-playing-games-report) 리포트에 따르면, 2020년 전 세계 모바일 게임 수익 중 RPG 게임이 21.3%를 차지하며, 전체 RPG 게임에서 발생하는 수익 중 약 72%가 한국, 일본, 중국의 세 개 국가에서 발생할 정도로 동아시아에서 큰 영향력을 발휘하고 있습니다. 특히 한국의 경우, MMORPG가 강세를 보이고 있습니다. App Store와 Google Play에서 발생한 2020년 상위 수익 50개 게임을 분석한 결과 한국은 MMORPG의 개수가 16개인 반면, 미국은 수집형 RPG가 6개, 영국은 수집형 RPG가 3개에 불과했습니다.
전체 모바일 게이머와 모바일 RPG 게이머 간 게임을 하는 이유에 대한 설문조사에 대해서도 흥미로운 점이 있습니다. 두 그룹 모두 휴식과 긴장 해소가 가장 많은 비중을 차지한다는 점은 동일하나, 시간 때우기라고 답한 비율은 전체 게이머가 44%, RPG 게이머가 34%로 차이를 보이며, RPG 게이머는 게임할 때의 즐거움과 스토리라인 및 세계관 경험, 캐릭터나 스토리에 대한 몰입감 등의 이유가 더 큰 비중을 차지하고 있습니다. (출처: [RPG Report](https://www.pangleglobal.com/resource/role-playing-games-report))
### 액션 게임의 특징
액션 게임은 플레이어가 버튼을 통해 캐릭터의 행동을 제어 및 조작하는 데 초점을 맞춘 게임 장르입니다. 플레이어의 신속한 의사결정과 동작, 그리고 그에 따른 즉각적인 결과가 특징으로 액션 영화를 보는 것과 같은 통쾌함이나 통제감의 재미를 느낄 수 있습니다. data.ai에서는 하위 장르로 대전, 대포 슈팅, 러닝, 배틀 로얄, MOBA(Multiplayer online battle arena), 진행형 슈팅 게임 등으로 분류하기도 합니다.
[__wf_reserved_decorative]
data.ai의 2021년 3분기 전 세계 게임 장르 분석에 따른 게임 시간 비중이 14.3%로 2위를 차지할 정도로 주요한 장르이며, 세부 장르 중에서는 쿠키런과 같은 러닝 액션 게임이 가장 많이 다운로드 되었습니다. 또한 2021년 미국에서 세 번째로 많은 다운로드가 발생한 게임 장르이기도 하며, 미국의 Z세대(16~24세) 게이머가 러닝 액션 게임을 플레이할 확률이 일반 대중보다 1.66배 더 높습니다. 한국에서도 Z세대(16~24세) 게이머가 러닝 액션 게임을 플레이할 확률이 일반 대중에 비해 1.75배 더 높습니다.
## 게임 카테고리 별 특성을 반영한 마케팅 전략이 필요합니다.
지금까지 모바일 게임 카테고리 및 주요 게임 카테고리의 특성을 소개하였습니다. 이번 포스트에서 살펴본 것처럼 게임 카테고리에 따른 오디언스 및 특징은 매우 다릅니다. 그렇기 때문에 모바일 게임 마케팅 전략을 기획할 때도 게임 장르 및 오디언스에 따른 특성을 고려하여 반영해야 합니다.
이번 포스트를 통해 모바일 게임 장르에 따른 특징을 이해해 보았다면, 다음 포스트에서는 모바일 게임앱을 둘러싼 생태계에 대해 알아보겠습니다.
---
## [에어브릿지 in MGS 2022] WWDC 2022 리뷰와 모바일 어트리뷰션
URL: https://www.airbridge.io/ko/blog/mgs-2022-wwdc-review
Language: Korean
Published: 2022-08-05
에어브릿지(Airbridge)는 지난 7월 13일 서울 삼성동 코엑스 그랜드볼룸에서 개최된 [모던 그로스 스택(Modern Growth Stack 2022, 이하 MGS 2022)](https://www.moderngrowthstack.com/) 컨퍼런스를 주최하였습니다. MGS 2022 속 에어브릿지가 전한 인사이트를 시리즈로 소개합니다.
> MGS 2022는 AB180(Airbridge)에서 개최한 애드테크&마테크 컨퍼런스로, 온오프라인 2,000여 명이 참여, 국내외 40여 명의 연사의 20여 개 세션으로 디지털 프로덕트 시대를 이끌어가는 주역들의 철학과 가치, 비즈니스 노하우를 공유하였습니다.
>
> 👉 MGS 2022 컨퍼런스의 전반적인 내용을 담은 현장 스케치는 [여기 링크](https://blog.ab180.co/posts/mgs-2022)에서 확인해보세요.
세 번째로 공개하는 세션 스케치는 브레이크아웃 룸에서 김양연 AB180 Product Design Team Lead가 모더레이터로 최수범 AB180 SDK Team Lead 및 손건 AB180 Backend Engineer와 함께 진행한 “WWDC 2022 리뷰와 모바일 어트리뷰션” 입니다. 애플의 개인 정보 보호 강화가 앱 마케팅에 미친 영향과 SKAN 4.0 업데이트 내용을 자세히 알아보았습니다.
## 애플의 앱 추적 투명성 정책과 유저 데이터 추적
애플의 앱 추적 투명성(AppTrackingTransparency, ATT) 프레임워크는 2020년 도입된 개인 정보 보호 강화 조치의 일환입니다. 앱이 광고를 목적으로 타사 앱 및 웹사이트에서 유저 데이터를 추적할 경우 반드시 유저의 동의를 받도록 하는 것이 그 핵심입니다. 앱은 유저 동의 없이 IDFA(iOS 기기용 광고 식별자)에 접근할 수 없으며 유저 또는 기기를 식별하는 기타 정보를 사용하여 유저의 활동을 추적할 수도 없습니다.
애플은 iOS 14.5 배포 당시 몇 가지 예시를 통해 ‘유저 추적’의 개념을 구체화했습니다.
### 유저 추적 사례 1: 맞춤형 광고 표시
커머스 앱에서 특정 상품을 구경한 뒤 다른 앱에서 해당 상품의 광고를 발견한 경험이 있으실 겁니다. 이는 해당 커머스 앱이 타사 앱 및 웹사이트에서 수집한 유저 데이터를 활용하고 있기 때문입니다.
광고주 앱은 지금 광고에 노출될 유저가 앞서 다른 상품을 봤던 유저와 동일한지 판별하기 위해 광고 식별자를 활용하여 리타겟팅 광고를 진행할 수 있습니다. 또는, ‘유저가 장바구니에 담은 상품’과 같이 앱 내에서 수집된 유저 데이터를 써드파티 및 매체로 전송하여 비슷한 취향을 가진 유저를 대상으로 유사 타겟을 구성하기도 합니다.
애플은 이러한 맞춤형 광고, 리타겟팅 및 유사 타겟팅과 같이 ‘유저를 식별할 수 있는 데이터를 활용한 광고’에 대해 유저의 동의가 반드시 선제되어야 한다는 점을 명문화했습니다.
### 유저 추적 사례 2: 광고 성과 측정
마케팅 캠페인을 최적화하여 ROI를 높이려면 각 마케팅 채널의 효과를 평가해야 합니다. 이를 위해서는 유저가 광고와 상호작용했다는 **터치포인트(touchpoint)** 데이터와 광고주 앱을 설치하거나 앱에서 구매 등을 했다는 **전환(conversion)** 데이터를 수집해야 하며, 이 두 종류의 데이터를 결합하면 어떤 전환에 어떤 터치포인트가 기여했는지 확인할 수 있습니다. 이러한 어트리뷰션 과정을 통해 매체별 광고 성과를 측정 및 분석할 수 있게 됩니다.
하지만 애플은 광고 성과를 측정하기 위해 앱의 유저 데이터와 타사 앱의 유저 데이터를 결합하는 써드파티 SDK를 앱에 배치하는 경우 역시 추적으로 정의합니다. 이는 IDFA에 크게 의존하고 있던 광고 성과 측정을 어렵게 하며, 에어브릿지와 같은 MMP에도 상당한 도전을 제기합니다.
### 유저 추적 예외 사례
추적에 동의하지 않은 유저에 대해 데이터를 수집할 방법이 전혀 없는 것은 아닙니다. 예외적으로, 앱의 유저 또는 기기 데이터가 유저 기기에서만 타사 데이터와 연결되어 있고, 유저 또는 기기를 식별할 수 없는 방식으로 광고주 앱에 데이터가 전송되는 경우는 유저 추적으로 간주하지 않습니다.
따라서 유저 데이터 기반 마케팅을 진행하고자 하는 광고주 앱에는 두 가지 선택권이 있습니다. 유저의 동의를 구하거나, 유저의 개인 정보를 보호하는 선에서 데이터를 수집하는 것입니다.
### 유저 추적 권한 요청하기
애플의 ATT 프레임워크 사용 시, 아래 그림과 같은 팝업창을 유저에게 노출하여 유저에게 IDFA에 대한 앱 접근을 허용할지 묻게 됩니다. 이를 통해 앱이 타사 앱 및 웹사이트에 걸친 유저의 활동을 추적하는 데 동의하는지 확인하는 것입니다. 추적 동의 관련 문구는 볼드체로 표시되며, 수정이 불가합니다. 대신 볼드체로 표시되어 있지 않은 데이터 수집 목적 관련 문구는 수정할 수 있습니다.
[__wf_reserved_decorative]
IDFA 접근 허용율이 높으면 유저 수준의 데이터를 수집하여 다양한 방식으로 활용할 수 있다는 장점이 있습니다. 그래서 ATT 프레임워크 도입 이후, 많은 기업이 조금이라도 ATT 동의율을 높이기 위해 데이터 수집 목적 관련 문구를 최적화하거나 팝업창을 띄우는 시점을 조정하는 등의 노력을 기울이고 있습니다.
### ATT 동의 외에 유저를 추적할 수 있는 방법은 없을까?
일각에서는 통합 인증(Single Sign-On, SSO)으로 IDFA 수집 없이 유저를 추적하는 방안에 관심을 갖기도 했습니다. SSO는 한 번의 로그인을 통해 여러 앱 및 웹사이트에 접근할 수 있도록 하는 유저 인증 방법입니다. 유저가 앱에 포함된 SSO 기능을 통해 구글, 페이스북, 카카오, 네이버 등의 계정으로 앱에 로그인하면, IDFA 대신 그 계정을 활용하여 유저를 추적한다는 발상입니다.
안타깝게도 이러한 경우에도 유저의 동의가 반드시 필요합니다. 애플은 “[개발자는 자신의 앱에 포함된 모든 코드에 대해 책임이 있으며, 여기에는 타사에서 제공한 SSO도 포함된다. 앱에 포함된 SSO 기능으로 인해 유저가 추적 대상이 될 경우, ATT 요청을 사용하여 먼저 유저의 허가를 받아야 한다.](https://developer.apple.com/kr/app-store/user-privacy-and-data-use/)”라고 명시하고 있습니다. 때문에 앱 개발자의 목적이나 의도와 상관없이, SSO를 통해 유저 동의 없이 유저를 추적하면 앱 심사를 거절 당할 수 있습니다.
그 뿐만 아니라, ATT에 동의해야만 앱을 사용할 수 있도록 하거나, ATT 동의 시 혜택을 제공하는 것 역시 앱스토어 심사 지침에서 금지되어 있습니다. 결론적으로 ATT에 동의하여 IDFA 접근을 허용한 유저만 추적할 수 있고, 이러한 지침을 비켜갈 방법은 없습니다.
## ATT 프레임워크 이전과 이후의 어트리뷰션
ATT 프레임워크 도입 이전에는 LAT(Limited App Tracking)라는 기능이 존재했습니다. LAT는 기본적으로 비활성화되어 있었고, IDFA 수집을 제한하고자 하는 유저는 아이폰 설정 앱에서 LAT를 활성화하는 수고를 들여야 했습니다. 또한, LAT는 모든 앱에 일괄 적용되는 설정이었습니다. 에어브릿지 데이터에 따르면 LAT를 활성화했던 iOS 유저는 15% 정도였으며, 약 85%의 유저에 대해 IDFA를 활용하여 어트리뷰션 할 수 있었습니다.
IDFA가 중요한 이유는 어트리뷰션의 정확도를 높여주기 때문입니다. IDFA를 활용하면 특정 기기를 식별할 수 있으며, 터치포인트 및 전환 데이터를 정확하게 연결할 수 있습니다. 즉, 한 대의 기기를 사용하는 유저라면, 여러 앱에서 활동했다 하더라도 관련 데이터가 해당 유저 한 명으로부터 기인했다고 확신을 갖고 판단할 수 있는 것입니다. IDFA가 매체 앱과 광고주 앱 사이의 연결고리 역할을 한다고 볼 수도 있습니다.
[__wf_reserved_decorative]
하지만 ATT 시대의 도래와 함께 IDFA 수집이 어려워졌습니다. 에어브릿지는 앱별 ATT 동의율을 평균 30%로 분석했습니다. 이를 바탕으로 계산해 보면, 매체 앱과 광고주 앱 양측에서 유저의 동의를 받아야 하는 **앱투앱 광고**의 경우 전체 유저의 9%에 대해서만 IDFA 활용 어트리뷰션이 가능하며, 광고주 앱에서만 동의를 받으면 되는 **웹투앱 광고**의 경우 약 30%의 유저에 대해 어트리뷰션 할 수 있습니다. LAT 사용 당시의 85%라는 비율에 비해 훨씬 작아졌다는 걸 알 수 있습니다.
### IDFA 없이 어트리뷰션하기
ATT 동의를 구하지 못해 IDFA를 사용할 수 없는 경우에는 Random ID를 식별자로 활용해야 하는데, Random ID는 단일 기기, 단일 앱에서 수집된 데이터에 대해서만 같은 값을 공유합니다. 이러한 경우 터치포인트 데이터끼리, 전환 데이터끼리 연결할 수는 있지만 터치포인트와 전환 데이터를 교차하여 연결하는 것은 불가하며, 어트리뷰션 역시 어렵습니다.
[__wf_reserved_decorative]
물론, 기술적으로는 핑거프린팅이라는 대안이 존재합니다. 핑거프린팅은 IP 주소와 기기명, OS, 이동통신사 등의 정보를 결합하여 유저를 특정하는 방법으로, Random ID만 있어도 어트리뷰션이 가능하도록 합니다. 하지만 애플은 ATT 프레임워크 도입과 동시에 핑거프린팅을 금지했으므로, 사실상 핑거프린팅을 통한 어트리뷰션 역시 불가능합니다.
## SKAN으로 어트리뷰션 하기
기존에는 MMP가 터치포인트와 전환 데이터를 수집해서 기여를 하는 방식으로 광고 성과 측정이 이루어졌습니다. 하지만 ATT 시대가 도래함에 따라, MMP가 추적 및 어트리뷰션 할 수 있는 유저는 소수에 불과하게 되었습니다. 이러한 문제에 대한 해답으로 애플은 SKAdNetwork(SKAN)라는 프레임워크를 도입했습니다. 개인 정보 보호 중심의 앱 설치 광고 성과 측정용 프레임워크를 대안으로 제시한 것입니다.
SKAN 프레임워크 사용 시, 애플이 MMP를 대신하여 직접 터치포인트와 전환 데이터를 수집하고 처리해서 기여 결과를 제공합니다. 다만 기존에 MMP들이 어트리뷰션 해 온 방식과는 일부 차이가 있습니다.
### 터치포인트 데이터 수집
MMP가 터치포인트 데이터를 수집하는 것과 유사하게, 애플은 SKAN 캠페인에 대해 광고를 클릭한 유저 및 광고에 노출된 유저를 구분할 수 있는 값을 제공합니다. 현재 도입되어 있는 SKAN 3.0 기준, 0과 99 사이의 숫자로 표현되는 Campaign ID를 통해 캠페인을 구분할 수 있으며, Source App ID를 통해 어떤 지면에 광고가 노출되었는지 확인할 수 있습니다.
다만, SKAN 캠페인에서 터치포인트 데이터를 수집하려면 지면에서 StoreKit이라는 프레임워크를 사용해야 하는데, 이는 아직 앱에 대해서만 지원됩니다. MMP는 웹과 앱을 아우르는 다양한 지면 및 캠페인에 대해 성과를 측정할 수 있는 데 반해 SKAN은 앱투앱 설치 캠페인에 대해서만 작동한다는 것이 한계점입니다.
### 전환 데이터 수집
MMP는 유저가 어떤 경로로 앱을 설치했는지 어트리뷰션 하여 광고 채널별 ROAS를 계산합니다. 하지만 애플은 이러한 분석을 유저 추적으로 간주하므로, 전환값(conversion value)이라는 제한적인 형태로 앱 설치 관련 데이터를 수집합니다. 유저를 식별할 수 없도록 하여 전환 데이터를 타사 앱 데이터과 결합할 수 없는 형태로 제공하는 것이라고도 설명할 수 있습니다.
전환값은 0부터 63까지의 정수로, 앱 개발자는 최대 64개의 유저 행동을 지정하여 측정할 수 있습니다. 예를 들어, 인앱 이벤트 중 캠페인 목표에 부합하는 특정 이벤트의 발생 여부 혹은 발생 횟수를 전환값으로 표현할 수 있습니다. 또 다른 예로 커머스 앱의 경우, 구매액을 구간별로 나누어 ROAS를 계산하는 데 활용할 수 있습니다. 만 원 이상 오만 원 미만 구매 유저에게는 3, 오만 원 이상 10만 원 미만 구매 유저에게는 4라는 임의의 값을 부여하는 식입니다.
이렇게 측정하고자 하는 유저 행동을 선별하고 그에 따른 전환값을 직접 설계하는 것이 어려울 수 있습니다. 이에 에어브릿지는 SDK와 대시보드에서 보다 쉽게 이런 설정을 조정할 수 있도록 지원하고 있습니다.
### 결과 데이터 제공
유저가 지면 앱에서 광고를 클릭한 다음 앱을 설치하면, 애플은 SKAN 포스트백을 통해 유저 데이터를 광고주 앱에 전달합니다. 포스트백에는 전환값과 더불어 앱 설치 여부, 노출 또는 클릭 여부, Campaign ID, Source App ID 등이 포함됩니다.
기본적으로는 이러한 데이터를 통해 유저 또는 기기를 식별할 수 없기 때문에 애플은 유저의 ATT 동의 여부와 관계없이 해당 정보를 전달합니다. 하지만 앱 설치가 충분히 발생하지 않아 유저를 특정할 수 있는 경우, 즉 애플의 정의해 둔 최소 달성 조건에 부합하지 못한 경우에는 포스트백이 불가합니다.
SKAN은 ATT 동의를 구하지 않고도 광고 성과를 측정할 수 있다는 점에서 유용하지만, 그만큼 많은 한계점을 수반합니다. 하지만 다행히 SKAN은 매년 조금씩 발전하고 있습니다. 2020년 iOS 14 배포 당시 SKAN 2.0이 발표됐고, 2021년 SKAN 3.0이 출시되었으며, 2022년에는 SKAN 4.0이 나올 예정입니다. 이렇게 새로운 기능이 꾸준히 추가된다면, 개인 정보 보호 강화로 인한 iOS 어트리뷰션에 대한 제약 역시 줄어들 것으로 기대됩니다.
> SKAN에 대한 더 자세한 설명 및 활용 방법은 아래 포스트에서 확인하세요!
> 👉 에어브릿지와 함께하는 iOS의 SKAdNetwork(SKAN) 성공 가이드
> 👉 [iOS의 SKAdNetwork(SKAN) 기본 개념과 측정방식 이해하기](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)
## WWDC 2022: Explore App Tracking Transparency
지난 6월 6일부터 10일까지 애플의 2022년 WWDC(Worldwide Developers Conference)가 전 세계의 뜨거운 관심 속에 진행됐습니다. 그 중 [Explore App Tracking Transparency](https://developer.apple.com/videos/play/wwdc2022/10166)라는 세션에서는 애플의 개인 정보 보호와 ATT에 관련한 시각을 엿볼 수 있었습니다.
### 핑거프린팅 금지 강조
핑거프린팅은 iOS 환경에서 특히 널리 쓰였던 기여 방식입니다. 안드로이드 환경에서는 리퍼러(Referrer) 방식으로 구글 플레이스토어를 통해 이뤄진 앱 설치에 대한 정보를 받을 수 있지만, iOS 환경에서는 유저가 어떤 채널을 통해 앱을 설치한 것인지 알 수 없기 때문이었습니다. 특히 웹에서 광고에 노출된 유저를 식별할 때에는 IDFA와 같은 광고 식별자도 부족했기 때문에 핑거프린팅의 필요성은 더욱 대두되었습니다.
하지만 애플은 ATT 프레임워크 도입 당시 핑거프린팅을 금지했으며, 이는 ATT 동의 여부와 무관하다고 밝혔습니다. 그리고 이번 WWDC에서 이러한 점을 재차 강조하며 iOS 앱 설치 캠페인에 대한 성과 분석에는 반드시 SKAN을 활용해야 한다는 입장을 보였습니다.
### 웹투앱 어트리뷰션에서의 SKAN 활용
지금까지 SKAN은 앱투앱 UA 광고에 대해서만 지원되었지만, WWDC 2022에서 발표된 SKAN 4.0에는 웹투앱 UA 광고에 대한 지원이 추가되었습니다. 이를 활용하고자 하는 광고주는 사용 중인 매체에 추가 개발을 요청해야 합니다.
이번 지원 방식은 애플의 웹투웹 및 앱투웹 어트리뷰션 프레임워크인 PCM(Private Click Measurement)과 유사한 부분이 많습니다. PCM이 뷰-쓰루(view through) 어트리뷰션을 지원하지 않는 것으로 보아, SKAN 4.0 역시 이 부분을 지원하지 않을 것으로 예상되는 점은 아쉬움으로 남습니다. 애플이 추후 SKAN 4.0에 대해 구체적인 정보를 발표하면 그 지원 범위도 더 정확하게 알 수 있을 것으로 보입니다.
[__wf_reserved_decorative]
### 세분화된 터치포인트 및 전환 데이터
Source App ID는 SKAN 포스트백에 포함된 정보 중 하나로, 앱스토어에서 다운로드를 유도한 앱을 식별할 수 있는 값을 의미합니다. 기존에는 Source App ID로 0과 99 사이의 정수만 사용할 수 있었지만, SKAN 4.0에서는 0과 9,999 사이의 값을 사용할 수 있도록 개선되었습니다. 경우의 수가 100개에서 10,000개로 늘어나면, 보다 많은 캠페인 데이터를 획득할 수 있을 것으로 보입니다.
전환 데이터 역시 한층 더 풍부해졌습니다. 기존에는 전환값으로 0부터 63까지의 정수(fine value)만 사용할 수 있었지만, 이제는 Low, Medium, High의 세 가지 구분값(coarse value)이 추가로 제공됩니다.
유관 데이터를 포스트백 받기 위해 달성해야 하는 앱 최소 설치 조건 또한 완화되었습니다. 기존의 임계값(privacy threshold)은 공식화되어 있지 않아서 마케터가 경험적으로 추측하는 수밖에 없었고, SKAN을 사용해 작은 캠페인을 테스트하기가 어려웠습니다. 하지만 SKAN 4.0에서는 앱 설치 수가 많을수록 더욱 풍부한 데이터를 보내주는 방식(crowd anonymity)으로의 변화가 있을 예정입니다.
애플은 앱 설치 수를 ‘적음(low)’, ‘보통(medium)’, ‘많음(high)’ 세 가지로 구분합니다. 터치포인트 데이터를 포함하는 Source App ID는 최대 네 자리 숫자로 구성되는데, 앱 설치 수가 적은 그룹에는 두 자리만, 보통인 그룹에는 세 자리까지, 많은 그룹에는 네 자리 모두 포스트백합니다. 전환값의 경우, 앱 설치 수가 적은 그룹에는 아예 제공하지 않고, 보통인 그룹에는 coarse value로, 많은 그룹에는 fine value로 데이터를 전송합니다.
### 포스트백 횟수 증가에 따른 추적 가능 기간 확대
기존에는 전환값이 마지막 업데이트 후 24시간 동안 변화가 없으면 자동으로 광고주에게 전송되었습니다. 하지만 무료 사용 기간을 제공하는 구독형 앱과 같이 유저를 오랜 기간에 걸쳐 추적해야 하는 경우에는 캠페인 성과를 측정하는 데 어려움이 발생했습니다. 전환까지 긴 시간이 걸리는데, 단 한 번의 전환 포스트백만 받을 수 있었던 것입니다.
이런 문제를 해결하기 위해 SKAN 4.0에서는 포스트백을 세 번에 걸쳐 전송하는 기능이 추가되었습니다. 첫 번째 포스트백은 앱 설치 후 이틀 안에, 두 번째 포스트백은 3~7일째에, 세 번째 포스트백은 8~35일째에 전송됩니다.
하지만 여전히 한계점은 존재합니다. 첫 번째 포스트백에서는 fine value를 받을 수 있지만, 두 번째와 세 번째 포스트백에서는 coarse value만 받을 수 있기 때문입니다. 포스트백이 여러 차례라 해도 각각 개별적인 시그널로 작용하기 때문에 세 번의 포스트백이 한 명의 유저로부터 발생한 건지 확인이 불가하며, 특정 기간 동안의 유저 추세를 파악할 수 있는 정도입니다.
### SKAN 4.0 도입 예상 시점
SKAN 2.0과 3.0은 각각 iOS 14, iOS 14.6과 함께 발표되었습니다. 따라서 SKAN 4.0은 iOS 16과 함께 2022년 9월 10~20일에 출시될 것이라고 예측하는 사람들도 있습니다. 하지만, iOS 16 개발자 베타 버전에 SKAN 4.0이 적용되어 있지 않고, 릴리즈 노트나 SKAN 공식 문서에도 관련 내용이 아직 추가되지 않은 것으로 미루어 보아 iOS 16.1 또는 그 이후 버전에 SKAN 4.0이 적용될 가능성도 배제할 수 없습니다.
## 어려워진 마케팅 환경, 어트리뷰션으로 헤쳐나가기
이전보다 iOS 마케팅에 제약이 많아졌지만, 다행히 애플은 SKAN을 꾸준히 업데이트할 것으로 보입니다. 이와 더불어 에어브릿지는 이 과정에서 고객사가 조금이나마 마케팅 고민을 덜 수 있도록, SKAN을 연구하고, 그에 맞추어 솔루션을 업데이트 해 나가고 있습니다.
변화하는 환경에서도 마케팅 성공을 이루려면 정확한 어트리뷰션을 통한 즉각적인 상황 개선이 필요합니다. 지금, 에어브릿지와 함께 진짜 마케팅 성과를 측정하세요!
애플의 앱 추적 투명성(ATT) 정책과 WWDC 2022에 대한 자세한 내용이 궁금하시다면, 아래 버튼을 클릭하고 AB180 Academy에서 MGS 2022 다시 보기 영상을 확인해 보세요! 간단한 회원가입 후 즉시 무료로 녹화 영상 시청이 가능합니다.
\*AB180 Academy는 AB180에서 만든 마케팅 솔루션 교육 플랫폼으로, AB180이 서비스하는 솔루션 사용법 및 마테크 실무 활용법을 무료로 배우실 수 있게 동영상 자료를 지원하고 있습니다.
[__wf_reserved_decorative]
---
## ROAS란? 에어브릿지 Revenue 리포트로 ROAS 측정하기
URL: https://www.airbridge.io/ko/blog/measuring-roas
Language: Korean
Published: 2022-08-04
마케팅 성과 측정, 어떤 지표를 사용하고 계시나요?
마케팅 예산은 한정적이고, 한정된 예산 속에서 최대의 효과를 발생시켜야 합니다. 성과를 개선하기 위해서는 우선 제대로 측정하고 분석하는 것이 중요합니다. 이때 마케팅 캠페인에 지출한 광고비 예산이 얼마나 가치가 있었는지 확인할 수 있는 지표인, **ROAS의 정의 및 특징**을 알아보고, **에어브릿지로 쉽게 측정하고 분석할 수 있는 방법**을 소개합니다.
## ROAS란?
**ROAS(Return on Ad Spend)**는 광고비 대비 매출액(수익)을 의미하며, 특정 광고를 집행할 때 투입한 광고비의 원 당(달러 당) 얼마만큼의 수익을 얻었는가의 비율을 보여주는 지표입니다. 이 지표는 마케터가 캠페인에 지출한 광고비 예산이 얼마나 가치가 있는지 가시적으로 보여줄 수 있습니다. 즉, 마케팅 캠페인 성과를 측정할 수 있는 중요한 지표로서 역할을 합니다.
- **ROAS 계산식: 광고를 통한 매출 / 광고비 * 100 (%)**
예를 들어서, 페이스북에서 아이폰 13 케이스에 대한 광고 캠페인을 하루에 10만 원씩 집행했고, 해당 캠페인으로 유입된 유저들이 20만 원의 구매를 발생시켰다면, ROAS는 200%입니다. 이때 하나의 캠페인이 아닌 여러 개의 캠페인을 대상으로 ROAS를 계산할 수도 있으며, 계산에 필요한 광고를 통한 매출 및 광고비 수치만 확인할 수 있다면 CPC 기반의 광고 외에도 이메일 마케팅 및 인플루언서 마케팅 캠페인 등 다양한 캠페인의 ROAS를 확인할 수 있습니다.
### ROAS가 중요한 이유
ROAS는 모바일 광고 시장이 활성화되면서, 모바일 웹&앱 및 데스크톱 웹&앱에서 발생하는 광고에 대한 전환(앱 설치, 인앱 이벤트) 및 클릭을 측정할 수 있게 되면서 주목받고 있는 지표입니다. 오프라인 광고에 대해서는 성과를 측정하기에 어려우나, 모바일 웹&앱 및 데스크톱 웹&앱에서 발생하는 광고의 경우, 전환율(앱 설치, 인앱 이벤트, 구매) 및 클릭률 등으로 성과를 측정할 수 있죠.
ROAS는 많은 마케터들이 캠페인의 성공을 판단하기 위해 근거로 사용하는 주요 지표입니다. 캠페인의 목표가 단지 브랜드 인지도를 높이기 위한 것이 아니라면, 마케팅 캠페인의 궁극적인 목표는 수익을 창출하는 것이기 때문입니다.
게다가 기업에서 자본을 효율적으로 사용하고 있다는 것을 보여주기에 적합한 근거로도 활용될 수 있습니다. 사용한 원 당(달러 당) 금액 이상의 수익을 거두는 것을 통해 비즈니스 성장을 증명할 수 있습니다.
### 좋은 ROAS란?
좋은 ROAS의 기준은 비즈니스별로, 캠페인별로, 플랫폼별로 다를 수 있습니다. 캠페인에 따라 브랜딩에 조금 더 집중하거나, 뉴스레터 구독자를 늘리는 등 다른 목적을 가진 경우, ROAS는 낮을 수밖에 없습니다. 하지만 ‘허슬, 멈추지 않는 추진력의 비밀’ 저자이자, 주목받는 온라인 마케터 [닐 파텔(Neil Patel)의 글에 따르면](https://neilpatel.com/blog/roas/), 대부분의 비즈니스는 일반적으로 **4:1 비율, 즉 400% 이상의 ROAS를 목표**로 합니다.
### ROI와 ROAS의 비교하여 이해하기
한편, 수치 기반의 마케팅 성과를 측정할 때, ROI를 지표로 활용하는 경우도 많습니다. **ROI(Return on Investment)**는 투자이익률로, 투자액에 비해 얼마나 이익이 올랐는지를 보여주는 지표입니다.
- **ROI 계산식: 이익(매출 - 마케팅 비용) / 마케팅 비용 * 100**
때때로 ROI와 ROAS를 혼동하는 경우도 있지만 제대로 구별할 필요가 있으며, 두 지표를 비교하여 살펴봄으로써 각각의 지표를 더 잘 이해할 수 있습니다.
ROAS는 특정 캠페인에서 발생하는 수익 기반의 지표이기 때문에, 기업 전체로 봤을 때는 실제 이익으로 이어지지 않을 수 있습니다. 또한, 특정 캠페인과 연관된 비용에 대해서만 측정하는 것이기 때문에, 전체적인 마케팅 비용에 대한 기업의 이익을 측정하기엔 무리가 있습니다. 그렇기 때문에 전체적인 마케팅 전략의 성공을 측정하기 위해 ROI를 장기적인 전략 기획 및 성과 측정에 활용하고, 동시에 ROAS를 단기적인 전략 조정과 캠페인 별 성과 측정에 활용하는 것이 좋습니다.
[]
## ROAS, 에어브릿지 Revenue 리포트에서 빠르게 확인해보세요.
[]
에어브릿지 Revenue 리포트는 분석 기간 내에 앱 설치 또는 딥링크 오픈 이벤트를 통해 앱으로 유입된 유저들에게 발생한 Revenue 성과를 다양한 메트릭으로 보여줍니다. 에어브릿지에서 자체적으로 개발한 Luft 엔진을 기반으로 유입 이후의 일자별로(N Day) 분석할 수 있습니다. 이때 에어브릿지에서 정의하는 ROAS의 경우 **사용한 비용에 대한 매출액 비중**입니다.
- **에어브릿지의 ROAS 계산식: Revenue / Cost *100 (%) **
메트릭을 ROAS로 선택하면, 지정한 **Revenue Event로 발생한 수익(Total Revenue)**과 Channel, Campaign, Ad creative 등 **지정한 그룹바이에서 발생한 비용(Total Cost)**을 기반으로 일자별 ROAS를 테이블에서 쉽게 확인할 수 있습니다.
이때, Revenue Event로 Order Complete (App), First Order Complete (App), Ad Impression (App), Ad Impression or Order Complete (App)의 네 가지를 선택할 수 있으며, 각각 이벤트에 대한 설명 및 계산 방법은 아래와 같습니다.
- **Order Complete (App)**: 유저가 구매 이벤트를 수행한 경우를 측정한 것으로, Order Complete 이벤트의 Event Value 값을 사용하여 메트릭을 계산합니다.
- **First Order Complete (App)**: 유저가 구매 이벤트를 처음으로 수행한 경우만 측정하며, 첫 Order Complete 이벤트의 Event Value 값을 사용하여 메트릭을 계산합니다. (* 이때 첫 구매 이벤트의 기준은 'Is First Event per Device ID'를 사용합니다.)
- **Ad Impression (App)**: 유저가 광고 노출 이벤트를 수행한 경우를 측정한 것으로, Ad Impression 이벤트의 Event Value 값을 사용하여 메트릭을 계산합니다.
- **Ad Impression or Order Complete (App)**: 유저가 광고 노출 이벤트 또는 구매 이벤트를 수행한 경우 모두를 포함하여 측정하는 것으로, Ad Impression 이벤트의 Event Value 값과 Order Complete 이벤트의 Event Value 값의 합산 값을 사용하여 메트릭을 계산합니다.
이외에도, 매출액, 이벤트 수, 그리고 유저 당 평균 매출액인 ARPU(Average Revenue Per User)도 메트릭으로 지정하여 빠르게 수익 관련 성과를 확인할 수 있습니다.
## ROAS는 성과 측정을 위해 반드시 필요한 지표입니다.
비용을 투자하여 마케팅을 집행했다면, 그 비용에 대한 성과를 측정하는 것은 당연합니다. 특히 ROAS는 특정 캠페인 혹은 여러 캠페인들의 성공 여부를 비용에 대한 수익으로 측정하도록 도와주며, 마케팅 성과 측정에 반드시 필요한 지표 중 하나입니다.
마케팅 효과를 증대화하기 위해서는 성과 측정이 먼저여야 합니다. ROAS를 포함한 마케팅 성과 측정을 위한 지표 및 데이터, 에어브릿지에서 모두 한 번에 확인해보세요!
---
## [에어브릿지 in MGS 2022] 웹투앱 전환을 극대화하기 위한 CTA 전략
URL: https://www.airbridge.io/ko/blog/mgs-2022-web-to-app-conversion
Language: Korean
Published: 2022-07-28
에어브릿지(Airbridge)는 지난 7월 13일 서울 삼성동 코엑스 그랜드볼룸에서 개최된 [모던 그로스 스택(Modern Growth Stack 2022, 이하 MGS 2022)](https://www.moderngrowthstack.com/) 컨퍼런스를 주최하였습니다. MGS 2022 속 에어브릿지가 전한 인사이트를 시리즈로 소개합니다.
> MGS 2022는 AB180(Airbridge)에서 개최한 애드테크&마테크 컨퍼런스로, 온오프라인 2,000여 명이 참여, 국내외 40여 명의 연사의 20여 개 세션으로 디지털 프로덕트 시대를 이끌어가는 주역들의 철학과 가치, 비즈니스 노하우를 공유하였습니다.
>
> 👉 MGS 2022 컨퍼런스의 전반적인 내용을 담은 현장 스케치는 [여기 링크](https://blog.ab180.co/posts/mgs-2022)에서 확인해보세요.
두 번째로 공개하는 세션 스케치는 브레이크아웃 룸에서 진행된 남성필 AB180 대표의 “웹투앱 전환을 극대화하기 위한 CTA 전략” 입니다. 에어브릿지의 방대한 데이터를 분석하여, CTA 버튼을 활용하여 웹투앱 전환을 극대화하고 비즈니스의 성장을 촉진하는 방법을 알아보았습니다.
## 웹투앱: 비즈니스 성장의 열쇠
통계에 따르면 미국에서만 매년 60만 개 이상의 기업이 새로 등장하지만, 그 중 10%가 1년 안에, 31%가 2년 안에, 50%가 5년 안에 비즈니스를 정리합니다.
어떻게 해야 치열한 경쟁을 뚫고 살아남을 수 있을까요? 남성필 대표는 ‘웹 유저를 앱에 온보딩시키는 것’에 집중해야 한다고 제안합니다. 보다 많은 사람들이 앱을 설치하도록 유도하고, 앱에서 회원 가입, 구매 등의 전환을 일으켜 가치를 창출하도록 하는 것입니다.
앱마이사이트(AppMySite)와 제이망고(JMango)에 따르면, 앱 유저는 웹 유저에 비해 전환율은 3배 이상, 세션 수는 4.2배 이상 높습니다. 이렇게 유저가 웹보다 앱에서 훨씬 높은 활동성을 보이기 때문에, 기업들이 유저를 앱에 안착시키고자 하는 것은 어쩌면 당연한 니즈라고 볼 수 있습니다.
하지만 에어브릿지 데이터 분석 결과, 여전히 많은 기업들이 모바일 트래픽의 60% 가량을 모바일 웹, 즉 랜딩 페이지로 보내고 있다는 것을 알 수 있었습니다. 앱으로 바로 다이렉트된 모바일 트래픽은 40%에 불과하다는 것입니다.
이에 따라 모바일 웹에 접속한 유저를 앱으로 전환시키는 것이 대단히 중요한 비즈니스 목표로 떠올랐으며, CTA(Call-to-Action) 버튼을 통해 이러한 목표를 효율적으로 달성할 수 있습니다.
## CTA 버튼 ‘똑똑하게’ 활용하기
웹 유저의 33% 이상이 최초 접속 후 72시간 내 적어도 한 번 웹투앱 CTA 버튼과 상호작용합니다. 웹에서 앱으로 이동하려는 유저가 그만큼 많다는 뜻입니다. 그러므로 웹과 앱을 모두 지원하는 프로덕트를 갖고 있는 기업은 아래 두 가지 요인을 고려하여 CTA 버튼을 장착하고 기회를 활용하기를 추천합니다.
### 적합한 종류와 적절한 타이밍을 선택하세요.
[__wf_reserved_decorative]
기업들이 가장 흔하게 사용하는 네 가지 CTA 버튼의 클릭률을 측정해 보니 서로 극명하게 다른 결과가 나타났는데, 그 이유의 핵심은 ‘**유저 경험**’에 있었습니다.
우선, Type A 버튼은 유저가 웹 페이지의 최하단까지 스크롤해야만 발견할 수 있습니다. 하지만 대개 콘텐츠 참여도(engagement rate)는 웹 페이지에서 스크롤을 내리지 않은 채 볼 수 있는 영역에서 정점에 도달합니다. 따라서 콘텐츠를 끝까지 꼼꼼하게 읽지 않는 대부분의 유저는 CTA 버튼이 있다는 사실조차 모른 채 웹 페이지를 떠나게 됩니다.
Type B 버튼은 CTA 버튼의 노출 자체를 늘리고 앱에 대한 인지도를 제고할 때에는 도움이 될 수 있지만, 클릭률을 높이는 데는 효과적이지 않습니다. 커머스 플랫폼을 예로 들어, 유저가 웹에서 상품을 검색하던 중 상세 페이지로 이동하는 버튼을 클릭했다고 가정하겠습니다. 그때 이런 모달(modal)이 띄워지면, 유저는 원하는 콘텐츠를 바로 확인하지 못한 채 시간을 낭비해야 합니다. 모달은 진행하고 있던 작업을 막아 세우는 경향이 강하기 때문에 유저 경험에 좋지 못한 인상을 남기고, 결국 유저 이탈을 유발합니다.
Type C 버튼은 언뜻 보면 Type B 버튼과 비슷하게 느껴질 수 있지만, 유저의 행동을 방해하지 않는다는 점에서 결정적인 차이가 있습니다. CTA 버튼이 콘텐츠를 소비하는 내내 노출되기 때문에 가시성은 높지만, 웹 페이지의 하단에 위치하고 있기 때문에 유저는 의도한대로 끊김없이 작업을 이어나갈 수 있습니다. 이러한 특징은 유저의 CTA 버튼에 대한 거부감을 낮추는 데 도움이 됩니다.
Type D 버튼은 유저가 특정 액션을 취했을 때 나타납니다. 커머스 플랫폼에서 유저가 ‘장바구니 담기’ 또는 ‘구매하기’ 버튼을 클릭했을 때 CTA 버튼을 노출하는 식입니다. 유저가 상품에 대한 긍정적인 관심을 표하는 순간에 “앱에서 더욱 편하게 구매하세요”라는 메시지를 던지면, 유저 입장에서 합리적인 가치 제안을 받은 것처럼 느껴집니다. 이에 따라 CTA 버튼에 대한 클릭률과 웹투앱 전환율도 자연스레 증가합니다.
### 다양한 공간에 자주, 꾸준히 노출하세요.
CTA 버튼의 종류를 선택했다면, 그 다음은 노출 빈도를 결정할 차례입니다. CTA 버튼은 유저로 하여금 앱 설치를 고려하도록 유도하기 때문에, 최대한 다양한 곳에 많이 설치하는 것이 좋습니다. 에어브릿지 데이터 분석에 따르면, 커머스 비즈니스가 아닌 기업들은 평균적으로 3.4개의 CTA 버튼을 사용합니다.
다만, CTA 버튼 배치 시 어떤 공간에 어떤 종류의 CTA 버튼을 두는지에 따라 클릭률이 상이해진다는 점을 기억해야 합니다. 또한, 앰플리튜드(Amplitude)와 같은 프로덕트 분석 솔루션을 사용하고 있다면, A/B 테스트를 진행하여 웹투앱 전환율을 높일 수 있는 최선의 방법을 찾는 것도 추천합니다.
결론적으로, 유저의 원래 행동을 방해하는 수단으로써 CTA 버튼을 활용하는 건 지양해야 합니다. 유저가 하고 있던 작업을 이어나갈 수 있도록 함과 동시에 언제든지 앱을 설치할 수 있도록 ‘선택권’을 주는 것이 중요합니다. 유저의 프로덕트 사용 경험을 방해하지 않는 선에서 여러 CTA 버튼을 설치하는 것은 유저의 웹투앱 전환 기회를 확대합니다. 또한, 유저가 웹에서 회원 가입, 구매 등의 유의미한 행동을 하거나 직접적인 의도를 드러낼 때 앱 설치를 유도하면 CTA 버튼과 상호작용할 가능성이 커집니다.
## 딥링킹으로 유저 경험 최적화하기
에어브릿지 데이터를 통해 확인한 또 하나의 사실은 CTA 버튼을 클릭하는 유저 중 25%는 이미 앱을 설치한 상태라는 점입니다. 일부 CTA 버튼은 유저를 무조건 앱스토어나 구글플레이스토어로 다이렉트하는데, 앱 기설치자는 앱 마켓을 거친 후 앱으로 이동하는 과정을 번거롭게 느낄 수 있습니다.
따라서 이러한 25%의 유저까지 사로잡으려면, 유저를 두 그룹으로 분류해야 합니다. 앱 미설치자의 경우, 앱 마켓으로 이동시켜 앱 설치를 유도하는 것이 먼저입니다. 반면 앱 기설치자의 경우, 딥링크를 활용하여 앱 내 특정 페이지, 가능하면 유저가 브라우징하고 있던 웹 페이지와 동일한 페이지로 보낼 수 있도록 설계해야 합니다.
### 커머스 플랫폼의 광범위한 딥링크 활용
[__wf_reserved_decorative]
웹투앱 전환에 활용되는 딥링크에 관련하여 또 한 가지의 놀라운 사실이 있습니다. 쿠팡과 같은 대형 커머스 기업들은 CTA 버튼과 함께 평균 10,000개 이상의 다양한 딥링크를 활용하고 있다는 점입니다.
유저를 무조건 앱의 홈페이지나 고정된 페이지로 다이렉트하면, 유저는 기존에 웹에서 보던 콘텐츠를 계속해서 소비하기 위해 즉각 앱을 이탈할 수 있습니다. 이러한 가능성을 최소화하려면 딥링크를 적극 활용하여 유저의 쇼핑 경험을 매끄럽게 만들어야 합니다. 커머스 기업들이 모바일 웹의 상품 상세 페이지와 앱의 상품 상세 페이지를 매우 비슷하게 디자인하는 것 역시 딥링킹을 염두에 둔 것이며, 이러한 노력에서 딥링크의 중요성을 엿볼 수 있습니다.
## 에어브릿지의 유연한 웹&앱 통합 분석
에어브릿지 데이터 분석에 따르면, 전체 앱 설치의 10%는 웹 캠페인에서 비롯됩니다. 그런데 만약 웹과 앱을 통합하여 성과를 분석하지 않으면, 이 10%를 페이드(paid) 전환이 아닌 오가닉(organic) 전환으로 오해할 수 있습니다. 결국 해당 페이드 웹 캠페인의 CPA와 CPI 성과를 잘못 평가하게 되는 것입니다.
이에 에어브릿지는 국내외 200여 개 기업들이 웹과 앱 모두에 대해 정확하게 마케팅 성과를 측정하고 캠페인을 최적화할 수 있도록 돕고 있습니다. 웹투앱 전환 캠페인을 진행하고 계시다면, 에어브릿지와 함께 효과적으로 캠페인 성과를 이해하고 마케팅 성공에 한 발 더 가까워지시길 바랍니다.
_CTA 버튼과 딥링크를 활용한 웹투앱 전환_에 대한 자세한 내용이 궁금하시다면, 아래 버튼을 클릭하고 **AB180 Academy**에서 MGS 2022 다시 보기 영상을 확인해 보세요! 간단한 회원가입 후 즉시 무료로 녹화 영상 시청이 가능합니다.
\*AB180 Academy는 AB180에서 만든 마케팅 솔루션 교육 플랫폼으로, AB180이 서비스하는 솔루션 사용법 및 마테크 실무 활용법을 무료로 배우실 수 있게 동영상 자료를 지원하고 있습니다.
[__wf_reserved_decorative]
---
## Touchpoints Overlap Report로 다이렉트 리스폰스 마케팅 최적화하기
URL: https://www.airbridge.io/ko/blog/direct-response-marketing-with-touchpoints-overlap-report
Language: Korean
Published: 2022-06-16
광고를 본 타겟 오디언스가 바로 앱 다운로드 혹은 구매하도록 만드는 캠페인을 운영하고 계시나요? 즉, 다이렉트 리스폰스(Direct Response)식 마케팅을 진행하거나, 혹은 다이렉트 리스폰스식 서비스를 운영하고 계시나요?
그렇다면, 가능한 한 많은 그리고 다양한 타겟 오디언스에게 광고를 노출하고 전환을 유도하고 싶을 텐데요. 동일한 유저에게 같은 캠페인의 광고를 여러 채널에서 보여주기엔 한정된 예산 속 기회비용이 커지기 때문에, 각 채널에서 오디언스가 얼마나 겹치는지 반드시 확인해야 합니다.
에어브릿지는 이를 효과적이고 빠르게 확인할 수 있도록, 같은 오디언스들 중 중첩적으로 광고를 보는 비율을 보여주는 **터치포인트 오버랩 리포트(Touchpoints Overlap Report)**를 제공합니다. 이번 포스트를 통해 에어브릿지가 제안하는 다이렉트 리스폰스 마케팅 캠페인의 성과를 최적화하는 방법을 확인해보세요!
## 다이렉트 리스폰스 마케팅(Direct Response Marketing)이란?
타겟 오디언스에게 앱을 다운로드하거나, 구매하는 등 즉각적인 반응을 유도하도록 만들기 위해 특별 제안을 제공하는 마케팅 캠페인을 의미합니다. 장기적으로 브랜딩을 강화하거나 인지도를 높이는 것과는 다르게, 빠르게 성과로 이어질 수 있기 때문에 마케터들에게 매력적으로 느껴질 수밖에 없습니다. 단 시간 에 ROI(Return On Investment)를 높이며 성과를 도출할 수 있죠.
이러한 전략은 직관적으로 유저들에게 다가가는 하이퍼 캐쥬얼 혹은 캐쥬얼 게임, 그리고 저 관여 제품을 판매하는 이커머스 플랫폼 등에서 주로 활용하곤 합니다. 특히 출시 초기에 모든 전략을 유저 획득에 집중하는 게임의 특성상, 다양한 채널의 넓은 타겟 오디언스에게 광고를 전달하여 가능한 한 많은 유저를 획득하고자 노력합니다. 또한, 이커머스 플랫폼에서 단기간에 매출 목표를 맞추기 위해 타임 세일과 같은 광고를 집행하는 경우, 많은 유저가 광고에 즉각적으로 반응하여 구매로 이어지도록 만들고 싶을 겁니다. 혹은 트렌드를 반영한 제품을 주로 판매하는 경우, 그 트렌드가 지나가기 전에 빠르게 많은 잠재 고객에게 도달하여 구매를 유도해야 합니다.
이렇게 시간에 민감한 다이렉트 리스폰스 전략이 최대한의 효과를 내려면, 타겟 오디언스가 겹쳐서는 안됩니다.
## 다이렉트 리스폰스 마케팅에서 주의해야 할 중복되는 타겟 오디언스 비율
다이렉트 리스폰스 마케팅에서 중요한 점은, 다양한 사람들에게 광고를 겹치지 않도록 송출하는 것입니다. 즉, 한 유저가 같은 광고를 여러 개의 채널에서 마주하도록 만들어서는 안됩니다. 여러 채널에서 오디언스가 겹치는 만큼 불필요한 광고 노출이 많아지고 비용이 늘어나게 됩니다. 캠페인에 대한 예산은 한정적인데, 같은 유저에게 광고를 노출하는 만큼 새로운 유저를 만날 수 있는 기회가 줄어들게 되죠. 다이렉트 리스폰스 마케팅에서는 광고 노출 빈도가 기회 비용이 되기 때문에, 같은 오디언스에게 중복되는 터치포인트가 세 개 이하로 내려갈 수 있도록 조정하는 것이 좋습니다.
## 터치포인트 오버랩 리포트로 중복되는 터치포인트 확인하기
그렇다면, 어떻게 하면 중복되는 터치포인트를 확인할 수 있을까요? 에어브릿지에서 제공하는 *터치포인트 오버랩 리포트(Touchpoints Overlap Report)*를 활용해보세요.
에어브릿지 터치포인트 오버랩 리포트는 설치, 구매 등과 같은 Conversion에 존재하는 Touchpoint들을 분석하여, 마케팅 채널 간 중첩되는 수치를 제공하고 있습니다. 즉, 특정 채널을 통해 전환된 유저가 우리 앱에 들어오기 전에 다른 채널에서도 광고를 클릭하거나 조회했는지에 대한 현황을 파악할 수 있습니다.
[]
또한, 아래의 표와 같이 채널별로 발생한 터치포인트가 다른 채널에서 발생한 터치포인트와 어떤 관계가 있는지, 중첩(Overlap) 정도를 파악할 수 있도록 직관적으로 보여줍니다. 이를 통해 멀티 터치 상황에서 광고 채널이 얼마나 겹치는지를 확인하고, 미디어 믹스를 조정하기 위한 인사이트를 얻을 수 있습니다.
[]
[👉 자세한 설정 및 분석 방법은 유저 가이드에서 확인해보세요.](https://help.airbridge.io/hc/ko/articles/900001042463-Touchpoints-Overlap)
### 터치포인트가 중복된다는 것이 의미하는 바는 무엇일까요?
위 스크린샷 이미지에 따르면, 특정 채널에서 발생한 전환에서 다른 채널과 7~80%가 터치포인트가 겹치는 경우가 다수 존재합니다. 이는 같은 오디언스를 대상으로 광고가 노출되고 있다는 것을 의미하므로, 해당 채널에 광고를 게재하는 것이 맞는지를 고민해 볼 필요가 있습니다. 다이렉트 리스폰스 마케팅 캠페인은 한정된 예산과 시간 속에서, 광고 노출 빈도가 즉 기회 비용이기 때문입니다. 혹은 터치포인트가 다수 겹치는 채널이 고객의 전환에 조력자 역할을 하고 있는 것일 수도 있지만, 검색엔진을 제외하고는 그런 경우는 많지 않다는 점을 유의해야 합니다.
(위 이미지 속 리포트는 데모 데이터로 실제 광고 캠페인의 성과를 의미하지 않습니다.)
## 에어브릿지 터치포인트 오버랩 리포트로 다이렉트 리스폰스 마케팅 예산을 효과적으로 활용해보세요.
[]
시간과 타겟에 민감한 다이렉트 리스폰스 마케팅을 운영하고 있다면, 여러 채널에서 타겟 오디언스가 겹치지 않도록 만드는 것이 중요합니다. 에어브릿지 터치포인트 오버랩 리포트로 오디언스의 중복을 빠르게 확인하고 캠페인 성과를 최적화하는 방법을 확인해보세요!
---
## Airbridge, 어디까지 써봤니? 에어브릿지 편의 기능 5가지
URL: https://www.airbridge.io/ko/blog/airbridge-5-features-to-watch
Language: Korean
Published: 2022-06-07
웹&앱 통합 광고 성과 어트리뷰션 솔루션 [에어브릿지(Airbridge)](https://www.airbridge.io/ko/)는, 캠페인 기획부터 진행, 성과 모니터링, 최적화까지 캠페인 성과 측정의 전 과정을 빈틈없이 지원합니다.
[__wf_reserved_decorative]
> 에어브릿지에서는 웹과 앱 등 다양한 플랫폼에서 발생한 유저 여정을 통합하여 트래킹 하여 마케팅 성과를 360도로 완벽하게 분석할 수 있습니다. 그리고 라스트 터치 어트리뷰션의 한계를 뛰어넘은 증분(Incrementality) 분석으로 광고의 진정한 성과를 확인할 수 있습니다.
이번 포스트에서는 에어브릿지의 사용자를 위한 편의 기능 5가지를 소개합니다.
## 1\. 대시보드에 접속하지 않아도 알 수 있어요!
\- 실시간 리포트 공유
에어브릿지의 대시보드에 접속 권한이 없는 관계자 혹은 파트너사에 실시간으로 통계 데이터를 공유하셔야 하나요? 이럴 땐 대시보드 접속 권한을 부여하지 않더라도 Actuals 리포트 상의 실시간 통계 데이터를 CSV 파일 혹은 Google Sheet를 통해 공유할 수 있습니다.
[__wf_reserved_decorative]
**Actuals 리포트**는 트래킹하고 있는 데이터를 자유롭게 확인할 수 있는 통계 리포트입니다. 에어브릿지에서 제공하는 다양한 데이터 필드를 활용하여 메트릭을 지정하고 특정 기준(그룹바이, 필터)들로 그 데이터를 세분화하거나 필터를 적용하는 등 원하는 형태로 리포트를 커스텀하게 구성할 수 있습니다.
이렇게 리포트를 생성하면 저장된 리포트에서 \[실시간 리포트 공유\] 옵션이 활성화되어 CSV 파일을 다운로드할 수 있는 URL이 제공됩니다. **이때 해당 CSV 파일은 특정 시점의 데이터를 저장하고 있는 것이 아닌, CSV 파일을 호출하는 시점의 실시간 데이터를 제공합니다.**
### \[실시간 리포트 공유\] 다운로드 받기
1. 리포트 상단의 \[Saved Reports\]를 클릭하여 저장된 리포트로 이동합니다.
2. 이동한 리포트의 통계 테이블 우측 상단 '공유하기' 버튼을 클릭합니다.
3. '실시간 리포트 공유' 토글을 ON 시키고 URL을 클릭하여 복사합니다.
4. 복사한 URL을 주소창에 복사하여 CSV 파일을 다운로드 받으면 해당 리포트의 실시간 통계 데이터를 확인할 수 있습니다.
[__wf_reserved_decorative]
보안을 위해, 동일한 리포트라고 하더라도 \[실시간 리포트 공유\] 설정을 ON/OFF 할 때마다 리포트 공유 URL이 변경됩니다. 리포트 공유 URL이 변경되면 이전에 사용하던 URL은 만료됩니다.
### Google Sheets를 사용하여 \[실시간 리포트 공유\] 기능 활용하기
구글 시트에서 제공하는 importdata 함수를 사용하면 .csv(쉼표로 구분된 값) 또는 .tsv(탭으로 구분된 값) 형식으로 주어진 URL에서 데이터를 가져올 수 있습니다.
**\=importdata("에어브릿지 대시보드에서 실시간 공유 기능으로 복사한 URL")**
importdata 함수는 설정한 URL을 정기적으로 호출하기 때문에 (약 15~30분 주기) 거의 실시간에 가까운 통계 데이터를 구글 시트에서 확인하고자 할 때 활용 가능합니다.
구글 시트에 불러온 에어브릿지 통계 데이터는 고객사 내부의 다른 데이터와 함께 결합하거나, 에어브릿지의 여러 리포트를 importdata 함수로 불러와서 통합된 리포트를 생성하는 용도로도 활용할 수 있습니다.
## 2\. 클릭 한 번으로 모든 설정 복사하기
\- 리포트 설정값 복사
에어브릿지의 리포트들을 살펴보다가 원하는 대로 데이터가 나오지 않아서, 혹은 데이터가 왜 이렇게 나오는지 궁금하셨던 적이 있으신가요? 문제를 확인하거나 의문을 해소하기 위해 AB180의 담당 고객 성공 매니저와 소통하게 되실텐데요. 이때 고객사와 AB180의 담당 매니저가 같은 데이터와 리포트를 확인하기 위해서는, 해당 리포트의 메트릭, 그룹바이, 필터와 설정 등이 모두 동일해야 합니다.
에어브릿지에서는 이 설정을 공유하는 과정이 매우 간편합니다. **\[리포트 설정값 복사\]만 클릭하면, config 형태로 복사되어 간단하게 전달할 수 있습니다.**
그리고, 복사된 내용은 \[설정\] 화면의 오른쪽 상단의 \[설정값 붙여넣기\]를 통해 바로 적용할 수 있습니다. 또는 리포트가 띄워져 있는 상태에서 Ctrl+v 를 통해 바로 동작시킬 수 있습니다.
[__wf_reserved_decorative]
## 3\. Ctrl+c & Ctrl+v? 더 간단합니다.
\- 버튼 하나로 차트 및 테이블 복사하기
평소 데스크톱이나 노트북에서 특정 요소를 복사해서 공유할 때, Ctrl+c와 Ctrl+v를 많이 활용하실 텐데요. 혹은 캡쳐 기능을 이용하실 수도 있죠. 하지만, **에어브릿지에서는 버튼 하나만 클릭하면 리포트를 복사할 수 있습니다.**
1. **차트 복사하기**
에어브릿지의 리포트 중, 차트가 함께 생성되는 리포트들에서는 차트의 오른쪽 상단에 위치한 \[차트 복사\] 버튼을 통해 쉽게 복사할 수 있습니다. 복사한 차트는 이미지로 붙여넣기 할 수 있습니다.
- 차트가 함께 생성되는 리포트: Trend 리포트, Active Users 리포트, Touchpoints Analysis 리포트, Touchpoints Overlap 리포트 등
2. **테이블 복사하기**
에어브릿지의 모든 리포트들의 테이블은, 테이블의 오른쪽 상단에 위치한 \[테이블 복사\] 버튼을 통해 쉽게 복사할 수 있습니다. 리포트 테이블에 있는 정보를 쉽게 복사하여 엑셀, 스프레드시트 등의 도구에 바로 붙여 넣을 수 있습니다. 테이블 서식을 그대로 유지할 수 있다는 장점이 있습니다.
## 4\. 트래킹 링크 대량 생성하기
에어브릿지의 트래킹 링크는 유저의 리다이렉트를 돕는 딥링크이자, 어트리뷰션을 위한 트래킹이 한 번에 가능한 All-in-One 링크입니다. 대시보드 상에서 캠페인 파라미터(UTM 파라미터)를 포함시킨 링크를 생성함과 동시에, 다양한 플랫폼, 채널, 상황에서 링크를 클릭하는 유저들을 원하는 목적지에 도달시킬 수 있죠.
그리고, **트래킹 링크를 한 번에 대량 생성하는 기능도 제공합니다.**
[__wf_reserved_decorative]
대시보드에서 일일이 트래킹 링크를 만들기 번거롭거나, 수천 개의 키워드별 링크를 만들어야 할 때 트래킹 링크 제너레이터(Tracking Link Generator)를 사용하여, 트래킹링크를 대량 생성할 수 있습니다. 트래킹 링크 제너레이터는 Google AppScript로 트래킹링크 생성 API를 사용하여 만든 Google Sheet 어플리케이션으로, 에어브릿지 고객이라면 누구든지 활용하실 수 있습니다.
1. 운영할 매체의 채널 타입에 따라 아래 템플릿을 복사합니다.
- [트래킹링크 제너레이터 템플릿 (Integrated Channel 용)](https://docs.google.com/spreadsheets/d/1I3oHv7gmjTTnbo-Ewcf2wTajUwC5qN0PlVGgk2Ktdig/edit#gid=0)
- [트래킹링크 제너레이터 템플릿 (Custom Channel 용)](https://docs.google.com/spreadsheets/d/1QTryw9E-lDMMXO89zY4ysQdrINhgV6kMhQFMSz8NQNU/edit#gid=0)
복사한 시트의 'Configs' 탭에서 app\_name과 tracking\_link\_api\_token을 채워줍니다. app\_name과 api\_token은 에어브릿지 대시보드 > Settings > Tokens 메뉴의 앱 이름 (Unique ID)과 트래킹 링크 API 토큰을 사용하시면 됩니다.
2. [Airbridge Tracking Link Generator Add On](https://gsuite.google.com/marketplace/app/airbridge_tracking_link_generator/9223133159)을 설치합니다.
3. 채널과 캠페인 파라미터, 유저 경로 등을 설정합니다.
4. 설정값 입력이 끝나면 \[부가기능 > Airbridge Tracking Link Generator > Generate Tracking Links\]를 눌러 링크를 생성합니다.
## 5\. 아직도 전체 Raw Data를 다운로드하시나요?
에어브릿지에서 원하는 데이터만 골라서 추출하세요!
에어브릿지에서 트래킹 한 마케팅 캠페인 성과의 Raw Data가 필요한 경우, Raw Data 추출 기능을 통해 **원하는 데이터를 맞춤 설정하여 다운로드**할 수 있습니다.
[__wf_reserved_decorative]
에어브릿지의 \[Raw Data > Data Export\] 메뉴에서 필요한 데이터를 선택하여 CSV 파일로 추출할 수 있습니다. Raw Data는 App Raw Data와 Web Raw Data로 구분되어 제공됩니다.
**Raw Data 추출 페이지 좌측 상단의 \[로우 데이터 추출\] 버튼을 클릭하여 추출할 Raw Data의 세부 조건들을 설정하고 추출을 요청하면 됩니다.**
\[이벤트 선택\], \[프로퍼티 선택\], \[필터 선택\] 단계에서 각각 원하는 데이터를 포함시키거나 제외할 수 있습니다. 그리고 \[데이터 프리뷰\] 단계에서는 앞서 진행한 \[이벤트 선택\], \[프로퍼티 선택\], \[필터 선택\] 단계의 설정 내역을 한 번에 확인함과 동시에, 해당 설정이 적용된 최근 5일 동안의 Raw Data를 미리 보기 형태로 확인할 수 있습니다. _여기서 미리 보기는 해당 설정이 적용된 최근 5일 동안의 데이터이므로, 다음 단계인 \[기간 선택\] 에서 설정한 추출 기간에 따라서 실제 추출되는 데이터와는 상이할 수 있습니다._
이후 \[기간 선택\] 단계에서, 원하는 기간을 설정한 후, \[추출 요청\] 버튼을 클릭하면 Raw Data 요청이 완료됩니다. 요청한 추출의 경우, 처리될 때까지 시간이 소요될 수 있으며 준비가 완료되면 요청 계정 메일로 알림이 전송됩니다.
요청한 Raw Data들은 ‘요청 기록’ 탭에서 세부 내역 및 히스토리를 확인할 수 있으며, 준비가 완료된 Raw Data를 다운로드할 수 있습니다. 또한, 상세 내역에서 \[데이터 추출\] 버튼을 클릭하여 기존 설정과 동일한 Raw Data를 요청하거나 이벤트 추가, 날짜 변경 등 세부 항목을 변경하여 유사한 Raw Data를 손쉽게 추출할 수 있습니다. \[템플릿으로 저장\] 버튼으로 기존 요청 내역을 저장하고 이후에 저장된 템플릿을 불러와서 Raw Data를 추출할 수도 있습니다.
## 웹 & 앱 통합 광고 성과 어트리뷰션 솔루션 에어브릿지
[에어브릿지](https://www.airbridge.io/ko/?utm_source=ab180_blog&utm_medium=link&utm_campaign=airbridge-5-features-to-watch)는 웹과 앱을 넘나드는 사용자의 유입을 정확히 파악하고 광고 채널, 캠페인 별 성과를 측정할 수 있는 국내 유일 웹 & 앱 통합 광고 성과 어트리뷰션 솔루션입니다.
자세한 내용이 궁금하시다면, 에어브릿지의 어트리뷰션 전문가들과 만나보세요.
[__wf_reserved_decorative]
---
## 에어브릿지와 함께하는 iOS SKAdNetwork(SKAN) 성공 가이드
URL: https://www.airbridge.io/ko/blog/skadnetwork-with-airbridge
Language: Korean
Published: 2022-06-02
[지난 포스트](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)에서는 SKAdNetwork(SKAN)의 기본 개념과 작동 방식을 이해하면서 SKAdNetwork에 대해 알아보았습니다. 이번 포스트에서는 SKAdNetwork의 고유한 작동 방식에 따른 전환 성과 데이터의 특징과 기존 어트리뷰션 모델과 달리 성과 측정 시 발생하는 제약 사항을 자세히 알아봅니다. 또한 이러한 SKAdNetwork 환경에 빠르게 대응할 수 있도록 어트리뷰션 툴 에어브릿지가 제시하는 SKAdNetwork 솔루션을 살펴보며 마케터가 SKAdNetwork 캠페인의 전환 성과를 성공적으로 측정하기 위해 꼭 알아야 팁들을 함께 확인해봅니다.
> \[주요 내용\]
> \* SKAdNetwork 전환 성과 데이터의 특징과 제약 사항을 자세히 알아봅니다.
> \* SKAdNetwork 생태계의 플레이어들을 확인하고 마케터가 놓치지 말아야 할 중요한 팁들을 확인합니다.
> \* 성공적인 SKAdNetwork 캠페인을 위한 에어브릿지의 SKAdNetwork 솔루션을 살펴봅니다.
## SKAdNetwork 전환 성과 데이터에 대한 이해
지난 포스트에서 SKAdNetwork는 **iOS 환경에서 개인정보를 준수(Privacy-centric)하며 앱 설치 캠페인의 성과를 측정할 수 있는 새로운 성과 측정 방식**으로 정리한 바 있습니다. SKAdNetwork는 기존의 어트리뷰션 모델과 달리 사용자 식별 정보 없이 전환 성과 데이터를 제공하기 위해 다음과 같은 두 가지 큰 특징을 가지고 있습니다.
- Device/User-level의 정보 없이 2진수 6자리의 Conversion Value를 통해 캠페인의 전환 성과(앱 설치와 설치 이후의 후속 인앱이벤트)를 측정합니다.
- Conversion Value를 담은 Postback 데이터는 전환 성과가 발생한 즉시 전송되는 것이 아니라 Conversion Value 업데이트가 종료한 이후 0~24시간 이내 랜덤한 시간에 지연 전송됩니다.
👉 [SKAdNetwork의 기본 개념 및 측정 방법에 대한 자세한 설명은 지난 포스트에서 확인해주세요!](https://www.airbridge.io/ko/blog/understanding-ios-skadnetwork)
이러한 SKAdNetwork의 고유한 측정 방식으로 인해 SKAdNetwork의 전환 성과 데이터는 기존 어트리뷰션 모델의 전환 성과 데이터와 달리 몇 가지 제약 사항이 발생하게 됩니다.
### SKAdNetwork 성과 데이터의 제약사항
\*SKAdNetwork로 측정한 전환 성과 데이터는 **Device/User-level 정보를 제공하지 않습니다.**
- 따라서 사용자 식별이 필요한 캠페인(리타겟팅, 리인게이지먼트 혹은 딥링크 오픈 등)은 지원하지 않습니다.
\*SKAdNetwork 캠페인의 전환 성과 데이터는 **실시간 성과 측정을 할 수 없습니다.**
- _SKAdNetwork의 전환 성과 데이터는 Apple에서 매체로 Postback 데이터를 전송하는 시점이 일정하지 않습니다. 그 시점이 Conversion Value 업데이트 완료 후 0~24시간 이내로 랜덤하기 때문입니다. 전환 성과에 해당하는 인앱 이벤트 발생 후 실제로 에어브릿지에서 해당 데이터를 확인하게 되는 시점까지 최소 24시간 이후 최대 48시간까지 소요될 수 있습니다._
\*SKAdNetwork 환경에서는 전환 성과의 **이벤트 발생 시점을 알 수 없습니다.**
- 전환 성과 데이터의 Event Timestamp는 매체에서 에어브릿지로 Postback 데이터가 전송되는 시점을 기준으로 합니다. 그러나 Postback 데이터는 이벤트가 발생한 뒤 바로 전송되는 것이 아니라 일정 시간 지연되어 랜덤하게 전송되기 때문에 정확한 이벤트 발생 시점을 알기 어렵습니다.
\*SKAdNetwork로 측정되는 전환 성과는 MMP의 SDK를 통해 기존 어트리뷰션 모델로도 측정되기 때문에 두 데이터 사이에 중복된 값이 존재할 수 있습니다.
- SKAdNetwork로 측정되는 인앱 이벤트는 기본적으로 앱 내에서 발생한 전환 이벤트입니다. 따라서 해당 이벤트는 MMP의 SDK를 통해 기존 어트리뷰션 모델로도 측정될 수 있으며, _광고주 앱 (Advertised App)과 지면 앱 (Source app) 모두 ATT의 상태가 opt-in인 경우_SKAdNetwork 방식으로 측정된 전환 성과 데이터와 기존 어트리뷰션 모델로 측정된 전환 성과 데이터 사이에 중복 값이 발생하게 됩니다.
👉 [SKAdNetwork 전환 성과 데이터를 적절하게 해석하는 방법](https://help.airbridge.io/hc/ko/articles/4412679790873-SKAdNetwork-Overview-#%EC%A0%84%ED%99%98-%EC%84%B1%EA%B3%BC-%EB%8D%B0%EC%9D%B4%ED%84%B0-%ED%95%B4%EC%84%9D%ED%95%98%EA%B8%B0) 자세히 보기
## SKAdNetwork 환경의 4명의 플레이어
SKAdNetwork 환경의 모바일 광고 생태계에는 플레이어에 해당하는 4명의 주체가 있습니다. 각각의 주체는 다음과 같은 역할을 합니다.
#### 지면 앱 (Source App)
SKAdNetwork 광고가 노출되는 앱(Publisher)입니다.
- 이때 지면 앱에 노출되는 광고는 두 종류로, View-through 광고와 [StoreKit-rendered 광고](https://developer.apple.com/documentation/storekit/skadnetwork/signing_and_providing_ads)\*가 있습니다.
\* _StoreKit-rendered 광고란 Storekit framework로 랜더링 되어 광고를 클릭하면 바로 앱 설치를 할 수 있는 광고를 말하며, Click-through 광고에 해당합니다._
#### 광고주 앱 (Advertised App)
광고의 대상이 되는 앱입니다.
- 해당 앱 설치 후 첫 번째 실행 시점부터 SKAdNetwork에서 Conversion Value를 측정하게 됩니다. Conversion Value 업데이트가 종료되면 랜덤한 지연 시간 이후 해당 앱이 설치되어있는 iOS 기기에서 앱 설치에 기여한 매체로 Postback을 전송합니다.
#### 매체 (Ad Network)
광고주(Advertiser) 와 지면(Publisher)를 연결하며, Apple로부터 전송받은 SKAdNetwork의 Postback 데이터를 다시 에어브릿지와 같은 MMP로 전달하는 역할을 합니다.
매체가 Postback 데이터를 전송하는 방식은 두가지가 있습니다. 1) API를 통해 집계된(Aggregated) 데이터를 전송하거나, 2) 개별 Postback을 전송합니다.
Meta와 Google Ads의 경우 1)의 방식을 사용하며 에어브릿지는 하루에 한 번 오전 1시에 데이터를 수집합니다. 데이터 수집을 위해 에어브릿지 대시보드 상에서 연동 설정 작업이 필요합니다.
- **Meta(Facebook Business)**: Conversion Value로 선택한 에어브릿지 이벤트를 페이스북 이벤트와 매핑한 뒤, 이 매핑 정보를 페이스북이 가져갈 수 있는 URL API에 담아서 Meta 측 대시보드에 입력하면, **Meta가 Conversion Value를 해석한 뒤 전송한 Postback 데이터를 집계된(Aggregated) 데이터로 제공**합니다.
- **Google Ads**: Conversion Value에 대한 **해석은 별도로 하지 않고 Postback 데이터를 집계된(Aggregated) 데이터로 제공**합니다.
👉 에어브릿지 대시보드에서 Meta와 Google Ads의 SKAdNetwork Postback 데이터 연동을 위한 설정 작업에 대한 자세한 안내가 필요하시면 담당 CSM에게 문의해주세요.
#### Mobile Measurement Partner (Airbridge)
에어브릿지와 같은 MMP는 SKAdNetwork 생태계에서 기여 분석(Attribution)이 아닌 전환 성과 데이터를 집계하는 (Aggregator) 역할합니다. 각각의 매체로부터 SKAdNetwork Postback을 전송받은 뒤 Conversion Value를 해석하고 (단, Meta는 직접 해석) 데이터를 집계한 뒤 SKAdNetwork의 전환 성과를 매체 별로 확인할 수 있는 대시보드 환경을 제공합니다.
- Conversion Value를 해석하기 위해서 SKAdNetwork 캠페인 집행 전 MMP(에어브릿지) 대시보드에서 Conversion Value로 측정하고자 하는 전환 성과에 해당하는 인앱이벤트를 설정해야 합니다.
\* 단, Meta는 광고 관리자 대시보드에서도 Conversion Value를 설정할 수 있습니다. 그러나 에어브릿지는 Conversion Value 설정 시 오류가 발생하는 것을 방지하기 위해 에어브릿지 대시보드에서 Conversion Value를 설정하도록 권장하고 있습니다.
## 효율적이고 직관적인 캠페인 관리를 돕는 에어브릿지의 SKAdNetwork 솔루션
에어브릿지는 복잡한 SKAdNetwork 환경에서 캠페인을 직관적이며 효율적으로 관리하고 전환 성과 데이터를 효과적으로 분석할 수 있는 SKAdNetwork 솔루션을 제공합니다. 에어브릿지가 고객사의 성공적인 SKAdNetwork 캠페인을 위해 제시하는 솔루션을 살펴보겠습니다.
### SKAN Conversion Value Model의 트래킹 세트로 Conversion Value를 손쉽게 설정
에어브릿지는 SKAdNetwork의 Conversion Value 측정 방식을 고려하여 직관적이고 손쉽게 Conversion Value를 설정할 수 있는 트래킹 세트 방식을 지원합니다. 에어브릿지의 SKAN Conversion Value 모델은 특히 커머스 분야에서 가장 효과적으로 캠페인 성과를 측정할 수 있도록 최적화되어있습니다.
[__wf_reserved_decorative]
👉 에어브릿지의 [SKAdNetwork Conversion Value Rule 설정 방법](https://help.airbridge.io/hc/ko/articles/900006402223-SKAN-Conversion-Value) 자세히 보기
- SKAN Conversion Value 모델의 트래킹 세트는 1) 구간별 구매액 (Revenue) 2) 구매 횟수 (Number of Order Complete) 3) 추가 이벤트 (Event Conversion) 실행 여부를 측정할 수 있는 옵션을 제공합니다.
- SKAN Conversion Value 모델의 트래킹 세트를 통해 Conversion Value 측정 시 인앱이벤트가 누락되는 경우를 방지할 수 있습니다.
SKAdNetworK의 Conversion Value는 높은 수로만 업데이트(Upper Update Only) 됩니다. 예를 들어 Conversion Value의 ‘10’에 설정한 인앱이벤트가 발생해 Conversion Value가 업데이트가 된 이후에는 Conversion Value 의 ‘1’~ ‘9’에 설정한 인앱이벤트가 발생해도 Conversion Value가 업데이트 되지 않기 때문에 해당 인앱이벤트는 전환 성과로 측정되지 않습니다.
- 에어브릿지의 SKAN Conversion Value 모델은 Conversion Value 설정 시 인앱이벤트의 중요도에 따라 [우선순위](https://help.airbridge.io/hc/ko/articles/900006402223-SKAN-Conversion-Value#%EC%9A%B0%EC%84%A0%EC%88%9C%EC%9C%84-%EB%8A%94-%EB%AC%B4%EC%97%87%EC%9D%B8%EA%B0%80%EC%9A%94)\*를 적용해, 중요하다고 여기는 이벤트를 더 높은 Conversion Value 숫자에 설정할 수 있습니다. _이 때 적용된 우선순위는 성과 측정 최적화에 사용됩니다._
> Tips!
> SKAdNetwork 캠페인 성과 측정을 위해서는 반드시 캠페인 집행 **전**에 Conversion Value을 완료해야합니다. Conversion Value를 설정되기 이전에 발생한 전환 성과는 측정이 되어 Postback 데이터가 발생해도 해석이 불가능해 활용할 수 없는 데이터가 되기 때문입니다.
캠페인의 성과로 측정해야 하는 프로덕트의 주요 성장 지표를 깊이 고민하는 과정을 에어브릿지와 함께 하세요. 복잡하고 어려운 SKAdNetwork 캠페인에 조금 더 쉽게 다가갈 수 있습니다.
### 앱 설치 후 24시간 측정으로 Postback 전송 지연 최소화
에어브릿지는 최적의 시점에서 성과를 측정 할 수 있도록 앱 설치 이후 24시간까지 Conversion Value를 측정해 Postback 데이터가 전송되는 시점이 지연되는 것을 최소화합니다. 따라서 에어브릿지에서는 SKAdNetwork 캠페인 집행 후 최소 24시간 이후 최대 48시간 이내에 캠페인의 전환 성과를 확인할 수 있습니다.
_SKAdNetwork는 앱 설치 이후의 후속 인앱 이벤트를 측정해 Conversion Value를 업데이트 하기 위해 24시간 동안 대기합니다._
- 그런데 이 Conversion Value 측정 타이머는 Conversion Value가 업데이트되면 초기화되어 다시 다음 인앱 이벤트가 발생할 때까지 Conversion Value의 업데이트를 대기합니다. Conversion Value의 24시간 타이머가 종료되기 전에 새로운 인앱 이벤트가 발생하지 않으면, 그제서야 Conversion Value가 인앱 이벤트 측정을 종료하고 0~24시간 내에 랜덤하게 기여한 매체로 Postback 데이터를 전송하게 됩니다.
- 따라서 이론적으로는 Conversion Value가 최대로 업데이트 되는 경우 앱 설치 후 65일 후에 전환 성과 데이터를 확인할 수 있는 경우가 발생할 수도 있습니다. 이를 방지하고자 에어브릿지는 최적의 시점에서 효율적인 성과 측정을 위해 **앱 설치 이후 24시간까지의 인앱 이벤트를 측정**합니다.
### SKAdNetwork와 non-SKAdNetwork 데이터를 한 번에
에어브릿지는 SKAdNetwork의 전환 성과 데이터와 기존 어트리뷰션 모델의 전환 성과 데이터를 한 눈에 확인할 수 있습니다. 에어브릿지의 Actuals 리포트는 단일 대시보드에서 특정 캠페인에 대해 SKAdNetwork로 측정된 성과 데이터와 기존 어트리뷰션 모델(에어브릿지 SDK)로 측정된 성과 데이터를 한 번의 조회로 한 눈에 확인할 수 있는 최적의 환경입니다.
- SKAdNetwork 전환 성과 데이터는 Actuals 리포트에서 [메트릭(Metric, 지표)](https://help.airbridge.io/hc/ko/articles/900003101763-Report-Metrics#app) 중 SKAN 메트릭을 선택해 조회할 수 있습니다. 데이터 조회 시, SKAN 메트릭과 일반 메트릭을 동시에 선택할 수 있습니다.
[__wf_reserved_decorative]
👉 [SKAdNetwork 데이터를 에어브릿지 Actuals 리포트에서 조회하는 방법](https://help.airbridge.io/hc/ko/articles/900006402223-SKAN-Conversion-Value#skan-%EB%8D%B0%EC%9D%B4%ED%84%B0-%EC%A1%B0%ED%9A%8C-actuals-%EB%A6%AC%ED%8F%AC%ED%8A%B8%EC%97%90%EC%84%9C-%EC%A0%9C%EA%B3%B5) 자세히 보기
## SKAdNetwork는 에어브릿지와 함께,
지금까지 두 편의 포스트를 통해 SKAdNetwork의 기본 개념과 특징, 그리고 이를 마케터가 좀 더 직관적으로 대응하도록 지원하는 에어브릿지의 SKAdNetwork 솔루션에 대해 알아보았습니다. 기존의 어트리뷰션 모델에 익숙한 상태에서 처음 SKAdNetwork의 개념을 접하면 어렵게 느껴질 수 있습니다. 그러나 SKAdNetwork의 핵심 개념인 Conversion Value와 Postback 전송 방식에 대해 이해하고, 이에 따른 전환 성과 데이터의 제약 사항을 이해하고 나면 SKAdNetwork 환경에서도 효과적인 캠페인 집행이 가능할 것입니다.
에어브릿지의 직관적이고 손 쉬운 SKAdNetwork 솔루션을 체험해보고 싶다면 담당 CSM에게 문의하시거나, 에어브릿지의 어트리뷰션 전문가와 이야기 나눠보세요!
---
## 어트리뷰션 툴 도입 전 체크해야 할 7가지 질문
URL: https://www.airbridge.io/ko/blog/attribution-tool-checklist
Language: Korean
Published: 2022-05-25
[앱 마케팅을 진행한다면 반드시 필요한 어트리뷰션 툴](https://blog.ab180.co/posts/why-mobile-attribution-tool)은, 한번 도입 후 다른 플랫폼으로 교체하기에 쉽지 않습니다. SaaS의 특성상 데이터의 이관이나 교육 등 전환에 드는 비용이 높아지기 때문이죠. 그렇기 때문에 도입하거나, 툴을 바꿔야 하는 상황이 왔을 때 의사 결정 시까지 많은 고민을 하게 됩니다. 그렇다면, 어트리뷰션 툴 도입에 앞서 어떤 부분을 반드시 고려해야 할까요?
각 프로덕트의 특성과 투자 및 성장 단계 등에 따라 고려해야 할 부분이 달라질 수는 있겠지만, 미리 체크해야 할 사항을 7가지 질문으로 정리했습니다. 아래의 질문을 미리 고려함으로써, 어트리뷰션 툴 도입 고민 과정에 도움이 되기를 바랍니다.
## 1. 로우 데이터(Raw Data)를 얼마나 유연하게 추출할 수 있는가?
어트리뷰션 툴은 앱 설치 및 인앱 이벤트의 원인이 되는 광고를 찾기 위해, 광고 클릭 데이터와 앱 설치 데이터를 서로 연결하여 그 성과를 보여주는 플랫폼입니다. 그렇기 때문에 어트리뷰션 툴은 SAN 매체로부터 받아온 데이터, 트래킹 링크를 통해 수집된 데이터, 그리고 앱에 설치된 SDK를 통해 수집된 앱 설치 및 인앱 이벤트 등에 광고 및 앱과 관련된 모든 데이터의 파이프라인 역할도 할 수 있습니다.
어트리뷰션 툴에서 이러한 데이터를 연결하여 광고 성과를 분석한 결과를 바로 대시보드에서 확인할 수도 있지만, 로우 데이터를 추출하여 원하는 형태로 가공하여 활용할 수도 있습니다. 따로 데이터를 수집하는 것보다 이미 어트리뷰션 툴로 수집한 데이터를 활용하는 것이 효율적이죠.
이때 특히 살펴봐야 할 점은, **데이터를 쉽게 추출할 수 있을지 여부, 그리고 얼마나 유연하게 원하는 데이터를 추출할 수 있을지**입니다. 툴에 따라서 정해진 데이터 스펙만 추출할 수 있는 경우도 있으며, 혹은 추가 비용이 부과되는 경우도 있기 때문에 이 부분은 미리 확인해 보는 것이 좋습니다.
> **에어브릿지(Airbridge)**의 경우, Web SDK, App SDK, 에어브릿지 트래킹 링크 및 에어브릿지 통합 ID 데이터에서 생성된 220여 종류의 로우 데이터를 추출할 수 있으며, 원하는 데이터만 커스텀하게 설정하여 빠르게 추출할 수 있습니다. 이는 타사에 비해 더 많은 데이터 필드이기 때문에, 더욱 방대한 양의 데이터 수집과 내재화가 가능합니다.
💡 에어브릿지 고객사인 짐싸는, 에어브릿지의 로우 데이터 추출 기능을 활용하여 특정 이벤트에 대해 디바이스 ID를 조회하여 중복을 제거함으로써 진짜 성과를 확인하고 있습니다.
[👉짐싸 인터뷰 전문 보러가기](https://blog.ab180.co/posts/zimssa-interview)
## 2. 서드파티 툴(Third-party tool)과 얼마나 잘 연동되어 있는가?
앞서 언급한 것처럼 어트리뷰션 툴의 로우 데이터를 추출하여 관리 및 자체적인 분석, 마케팅 등의 목적을 위해 활용하는 것과 더불어, 클라우드 스토리지나 타 솔루션을 직접 연동하여 더 고도화된 기능을 활용하는 경우도 많습니다. 사업 초기 단계여서, 혹은 다른 툴들을 사용하지 않는다는 이유로 이 부분을 고려하지 않을 수도 있지만, 언제라도 필요에 따라 활용하게 될 수 있죠. 이를 위해 다양한 서드파티 툴들과의 연동 여부는 물론, 쉽게 확장 가능한지 등을 미리 확인해두는 것이 좋습니다. 그리고 이미 여러 서드파티 툴을 사용하고 있다면, 어트리뷰션 툴 도입에 앞서 해당 툴과 쉽게 연동할 수 있는지는 반드시 체크해야 할 사항입니다.
어떤 서드파티 툴과 연동할 수 있는지와 그 방법에 대해서는 해당 어트리뷰션 툴의 사용 가이드, 개발 가이드, API 가이드에서 확인할 수 있습니다.
> 에어브릿지는 타 플랫폼으로의 데이터 이전 기능을 기본으로 제공하기 때문에 별도의 개발이나 추가 비용이 필요하지 않습니다. 간단한 설정만으로 아마존 S3(AWS S3)와 같은 클라우드 스토리지나 앰플리튜드(Amplitude), 브레이즈(Braze) 등의 타 솔루션과 연동할 수 있습니다. 쿼리 API를 통해 프로그래매틱하게 연동할 수도 있습니다.
💡 **앰플리튜드(Amplitude)의 HTTP API V2 연동?** 앰플리튜드를 사용하고 있는 경우, 어트리뷰션 데이터는 물론 수집한 이벤트 데이터도 전달할 수 있는지 확인해 보는 것이 좋습니다. 어트리뷰션 툴과 앰플리튜드를 SDK를 통해 연동한다고 하더라도 Attribution API의 스펙 차이 및 시간차로 인해 데이터가 유실되어 수치 차이가 발생할 수 있습니다. HTTP API V2 연동을 하게 되면 이러한 문제 없이 모든 데이터가 앰플리튜드에 기록되게 됩니다.** HTTP API V2 연동은 어트리뷰션 툴 중 에어브릿지만 가능합니다.[
👉 앰플리튜드와 앱 어트리뷰션 툴의 연동 방법 더 자세히 알아보기](https://blog.ab180.co/posts/amplitude-integration-app-attribution)
## 3. 세밀한 어트리뷰션 윈도우 설정이 가능한가?
앱 설치나 구매 등의 전환 이벤트에 대해 어떤 터치포인트가 기여했는지 확인하기 위한 방법론인 ‘어트리뷰션 모델’에는 다양한 모델이 존재합니다. 대부분의 광고주들은 복잡한 어트리뷰션 분석을 단순화하여 빠르게 진행하고, 매체와 광고비 정산 기준을 협의할 때 ‘라스트 터치 어트리뷰션 모델(Last-touch attribution model)’을 사용합니다.
👉 [마케팅 성과를 다각도로 분석하기 위한 접근 방식 알아보기 1편 - 어트리뷰션(LTA, MTA)에서 어트리뷰션 모델에 대해 더 자세히 알아보기](https://blog.ab180.co/posts/marketing-measurement-1-attribution)
각 어트리뷰션 툴에서는 이 기여 모델을 적용시키는 기간인, ‘어트리뷰션 윈도우’의 설정이 가능합니다. 다만, 얼마나 세밀하게 설정할 수 있을지는 플랫폼마다 다르기 때문에 다음의 질문들에 대한 확인이 필요합니다.
1. 설치의 기여를 결정하는 윈도우, 인앱 이벤트의 기여를 결정하는 윈도우 각각에 대해 설정 가능한가?
2. **디바이스 매칭 방법**일 때와 **확률적 매칭 방법**일 때 각각 다르게 설정할 수 있는가? (아래 내용은 에어브릿지 기준입니다.)
- 디바이스 매칭 방법: 구글 플레이스토어 레퍼러(referrer) 쿼리스트링 파라미터, 딥링크(deeplink), 광고 ID 매칭 (ADID 매칭), 플랫폼 매칭
- 확률적 매칭 방법: 고유하지 않은 값들의 조합(디바이스 정보, IP 등)을 통해서 광고를 조회하고 클릭한 사용자와 앱을 최초 실행한 사용자가 동일한 사용자인지 확률적으로 추정하는 방법
3. 전체 채널은 물론 각 채널별로 윈도우를 다르게 설정할 수 있는가?
에어브릿지의 경우, 앱 설치를 기점으로 어트리뷰션 윈도우 설정을 두 가지로 나눌 수 있습니다. 앱 설치 전에 설치에 기여한 광고 터치 포인트(클릭, 조회)를 다루는 **룩백 윈도우(Lookback Window)**, 그리고 앱 설치 후 사용자가 앱에서 한 활동(인앱 이벤트)를 다루는 **어트리뷰션 윈도우(Attribution Window)**가 있습니다. 또한, 디바이스 매칭 방법일 때와 확률적 매칭 방법일 때 윈도우를 다르게 설정할 수 있으며, 전체 채널(글로벌) 및 각 채널별 윈도우 설정도 가능합니다.
타 어트리뷰션 툴의 경우, 앱 설치에 대해서만 기여를 분석할 수 있는 경우도 있는데요. 그렇게 되면, 설정한 어트리뷰션 윈도우 기간 동안 처음 기여한 채널이 계속 유지되기 때문에, 이후 Organic 및 다른 채널로 다시 유입된 유저가 구매를 해도 기여가 변하지 않는다는 문제가 발생하게 될 수 있다는 점을 염두 해야 합니다.
## 4. 모든 광고 성과 데이터를 커스텀 대시보드 설정을 통해 한눈에 확인할 수 있는가?
마케터라면 광고를 집행하고 대시보드에서 데이터를 통해 성과를 확인하는 것이 하루의 중요한 일과 중 하나일텐데요. 매일 대시보드를 방문하는 만큼,** 얼마나 마음껏 커스텀 하게 설정하여 원하는 대로 구성할 수 있는지가** 업무 효율성에 영향을 미칠 겁니다. 캠페인에 따라 확인하고 싶은 데이터가 달라질 수밖에 없기 때문입니다.
게다가, 웹과 앱 모두 보유한 경우 **하나의 대시보드에서 웹과 앱의 데이터를 확인 가능한지, 그리고 모든 모바일 플랫폼(Android & iOS)의 데이터를 한눈에 살펴볼 수 있는지를** 확인하는 것도 좋습니다. 만약 앱과 웹을 트래킹하는 툴이 나눠져 있다고 하면, 앱은 어트리뷰션 툴(MMP)에서, 웹은 구글 애널리틱스와 같은 웹 로그 분석 툴 각각에서 데이터를 확인해야 할 겁니다. 심지어 어트리뷰션 툴에서도 Andriod 및 iOS를 확인하는 리포트가 나눠져 있고, iOS에서는 SKAdNetwork의 데이터를 다른 리포트에서 확인해야 한다면, 모든 데이터를 분석하기 위해서는 총 네 번의 서로 다른 리포트 조회가 필요하게 됩니다.
‘어트리뷰션 툴’이라고 한다면, 모바일 앱의 설치 및 인앱 이벤트에 대한 성과만 분석하는 것이라고 생각하기 쉽습니다. 하지만, 고객은 모바일 앱에만 존재하지는 않습니다. 모바일 웹/앱은 물론, 데스크톱의 웹까지 **모든 플랫폼에서, 광고가 가능한 모든 채널의 마케팅 성과를 확인해야 합니다.**그리고 이 모든 데이터를 한눈에 확인할 수 있는 어트리뷰션 툴을 사용한다면 진짜 마케팅 성과를 분석하는 데에 도움이 될 겁니다.
> 에어브릿지는 Actuals 리포트에서 모든 데이터를 원하는 형태로 커스텀 설정 가능합니다. 모든 플랫폼에서 전환 데이터를 수집할 수 있도록 앱 SDK와 웹 SDK를 제공하기 때문에, 웹과 앱(iOS-SKAdNetwork 포함, Android) 데이터를 한눈에 확인할 수 있습니다. 그리고 Groupby를 10개까지 지정할 수 있어 더 자유로운 설정이 가능합니다.
💡 에어브릿지 고객사인 엔코드는 자유도가 높은 대시보드를 통해 모든 데이터를 대시보드상에서 원하는 형태로 자유롭게 살표보고 있다는 점에 만족하고 있습니다. Actuals 리포트에서 웹과 앱을 같이 살펴보기도, 따로 살펴보기도 합니다.
[ 👉 엔코드 인터뷰 전문 보러가기](https://blog.ab180.co/posts/ncode-interview)
## 5. 계정 별 데이터 접근 권한을 나눌 수 있는가?
경우에 따라 마케터가 단독으로 마케팅을 기획하고 집행하는 곳도 있겠지만, 많은 회사에서 대행사 및 미디어 파트너와 함께 광고 및 마케팅을 기획하고 실행하게 됩니다. 이때, 민감한 데이터에 대한 보안 문제 등의 이유로 **사용자 계정에 따라 데이터 접근 권한을 다르게 부여**해야 할 필요가 있습니다.
이에 대해 각 어트리뷰션 툴에 따라 어떻게 권한을 나누는지, 그리고 각 권한에 따라 확인하거나 설정할 수 있는 데이터가 달라질 수 있는데요. [에어브릿지의 경우](https://help.airbridge.io/hc/ko/articles/900001060963#%EC%95%B1-%EC%82%AC%EC%9A%A9%EC%9E%90-%EA%B5%AC%EB%B6%84), 아래와 같이 앱을 처음 등록한 관리자인 ‘오너’, ‘사내 마케터’, ‘대행사’, ‘ 미디어 파트너’로 권한을 나누고 있습니다.
- **오너**: 앱을 처음 등록한 관리자로, 모든 앱 관리 기능을 이용할 수 있으며, 모든 리포트를 열람하고 데이터를 추출하거나 설정하는 것이 가능합니다. 계정 추가 및 삭제도 가능합니다. 유일하게 앱을 삭제할 수도 있습니다.
- **사내 마케터**: 관리자로, 모든 앱 관리 기능을 이용할 수 있으며, 모든 리포트를 열람하고 데이터를 추출하거나 설정하는 것이 가능합니다. 계정 추가는 가능하나, 앱이나 권한을 삭제하는 것은 불가능합니다.
- **대행사**: 대행사 계정을 통해 생성된 트래킹링크에 대해서만 데이터를 조회할 수 있습니다. 광고주로부터 권한을 획득하여, 허용된 광고 채널에 대한 데이터의 리포트 조회 및 데이터 추출이 가능합니다.
- **미디어 파트너**: Actuals 리포트에서 소속 매체(채널)에 기여된 데이터를 확인할 수 있습니다. 광고주가 설정한 해당 채널의 연동 정보는 확인할 수 있습니다.
## 6. 마케팅 컨설팅과 기술 및 고객 지원이 원활한가?
위와 같은 기능적인 차이점 외에도, **어트리뷰션 툴을 더 잘 활용하도록 도와줄 수 있는 지원팀의 존재 및 적극적인 지원 가능성**에 대해서도 확인해 보아야 합니다. 좋은 마케팅 성과를 내고 비즈니스의 성장을 위해서는, 전문적이고 체계적인 고객 성공 팀(Customer Success Team, CS Team)이 필요합니다.
혹시 해외 SaaS 툴을 사용하다가 CS 지원 요청 후 답변이 오기까지 한참을 기다렸던 적이 있으신가요? 요청하고 많은 시간이 흐른 뒤, 해외에 위치한 CS 팀과 문제 해결을 위해 새벽 시간에 화상 미팅을 해야 했던 불편함도 있었을 겁니다. 혹은 문제가 발생할 때마다 사용하고 있는 여러 가지 툴 각각의 담당자에게 따로 연락을 하며 리소스를 소모했던 경험을 했을 수도 있습니다.
에어브릿지 CS 팀은 한국에 소재하고 있기 때문에 보다 신속한 대응이 가능하며, AB180의 파트너사인 앰플리튜드(Amplitude) 및 브레이즈(Braze)에 대한 지원도 함께 수행하고 있기 때문에 전체 프로덕트 및 마케팅에 대해 체계적이고 전문적으로 지원하고 있습니다. 기본 사용법 교육은 물론, 커스텀 세팅 및 관리자 세팅 교육부터, 이벤트 분류 체계와 SDK 설치 안내 및 리뷰로 완전하고 효과적인 사용이 가능하도록 돕습니다. 이후에는 실질적인 성과까지 체계적으로 지원하기 위해, 솔루션 활용 정도를 트래킹하고 활용도가 낮은 기능에 대해 어드바이징하며, 콘퍼런스 및 콘텐츠를 통해 마케팅 테크놀로지 업계 최신의 트렌드를 제공하고 있습니다.
## 7. 비용 구조가 어떻게 이루어져 있는가?
어트리뷰션 툴마다 비용 구조가 다르게 구성되어 있습니다. 크게 Install, DAU/MAU에 따른 과금 구조를 가지고 있으며, 각각은 장단점이 있기 때문에 견적을 미리 받아보는 것이 좋습니다. 우선 DAU/MAU 기반의 과금 구조의 경우, 비용을 예측할 수 있다는 장점이 있습니다. 또한, 광고를 통해 설치만 하고 이탈하는 유저들에 대해서 지나친 비용을 지불할 필요가 없게 됩니다. Install 기반의 과금 구조의 경우, 마케팅을 하는 만큼 비용을 지불하게 되는데, 특히 Paid Install 광고 캠페인을 운영하여 Install이 급증하게 되면 갑작스러운 추가 비용이 발생할 수 있다는 점을 염두 해야 합니다.
또한, 어트리뷰션 툴에 따라 부가 기능에는 비용이 추가되는 경우도 있고, 올인원으로 한 번에 지불하는 경우도 있습니다. 그렇기 때문에 처음에 견적을 받을 때, 필요한 모든 기능을 포함한 비용인지 미리 확인해야 합니다. 대부분 비용의 경우 웹사이트보다는, 데모 신청 및 직접 문의를 통해 확인할 수 있습니다.
## 마무리하며,
어트리뷰션 툴은 한 번 도입하면 변경하기 쉽지 않은 경우가 많습니다. 그렇기 때문에 도입하거나 변경하기 전에 신중한 고민과 준비가 필요하죠. 그 준비의 시작은 우리 서비스/앱에 필요한 어트리뷰션 툴의 기능이 무엇인지 스스로에게 질문하는 것입니다. 그리고 더 나아가 각 툴을 비교해 보며 진짜 마케팅 성과를 분석하고 개선할 수 있도록 도와줄 툴을 찾는 겁니다. 어트리뷰션 툴의 도입 전, 위의 7가지 사항을 반드시 체크해 보시기 바랍니다.
[]
---
## 에어브릿지가 직접 개발한 OLAP DB, Luft를 소개합니다
URL: https://www.airbridge.io/ko/blog/introducing-luft
Language: Korean
Published: 2022-05-19
웹과 앱을 넘나드는 사용자의 유입과 행동을 정확히 파악하여 광고 성과를 측정, 분석하는 어트리뷰션 솔루션 에어브릿지(Airbridge)는 월 100억 건 이상의 데이터를 처리하고 있습니다. 이 데이터들을 실시간으로 분석한 애널리틱스를 제공하기 위해, 에어브릿지에서는 **Luft**라는 사용자 행동 분석 특화 OLAP 데이터베이스를 자체적으로 개발했습니다.
에어브릿지는 이 Luft 데이터베이스를 통해 코호트 분석 기능 및 각종 리포트를 실시간으로 제공할 예정입니다. 이번 포스트는 앞으로도 에어브릿지의 중추 역할을 담당할 Luft에 대해 자세히 알아보겠습니다.
## 코호트 분석 기능?
**코호트 분석**은 원하는 유저군(= 코호트)을 잡아서 해당 유저들의 행동을 분석하는 기법입니다. 예를 들어 "지난 6개월간 우리 앱에서 한 달에 10만 원 이상 소비한 30대 여성"이란 유저군을 잡은 후, 해당 유저군의 리텐션(재방문률)을 분석하는 것 등이 코호트 분석에 해당됩니다.
[__wf_reserved_decorative]
이를 위해선 특정 유저군을 먼저 잡고, 해당 유저들의 이벤트 데이터를 로드해 행동 분석을 수행하는 두 단계를 거쳐야 합니다. 문제는 두 단계의 작업이 모두 시간이 많이 걸리는 데이터 분석이라는 점입니다. 물론 결과를 나중에 이메일로 보낼 수도 있지만, 이럴 경우 데이터 분석가가 조건을 조금씩 바꿔서 결과를 실험해보는 게 사실상 불가능해진다는 UX적인 단점이 있습니다. 따라서 대시보드에 즉각 분석 결과가 보여져야 한다는 성능을 갖출 필요성을 느끼게 되었습니다.
## 기존 아키텍처의 한계
한편, 에어브릿지는 기존에 데이터 분석을 위해 Apache Druid를 사용하고 있었습니다. Druid는 OLAP 데이터 스토어로서, 정해진 메트릭을 시간 단위별로 미리 계산해 두는 Pre-Aggregation과 실제 쿼리시에는 필요한 값만 선택해 연산하는 Roll-up 방식을 통해 수 초 내로 통계 쿼리가 가능합니다.
Druid는 그간 에어브릿지에 필요했던 다양한 통계 리포트를 제공하는 데에는 최적의 솔루션이였지만, 실시간 코호트 분석 기능을 개발하려고 하면서부터 한계를 느끼기 시작했습니다. 에어브릿지에서 필요한 종류의 쿼리에 비해 한정된 쿼리만을 수행할 수 있다는 점과, 쿼리 성능이 상당히 떨어진다는 점이 (테스트 최소 15-20초) 발목을 잡았습니다.
### Pre-Aggregation의 한계: 정해진 종류의 분석만 가능
Druid를 포함한 상당수의 OLAP DB는 정해진 메트릭을 사전에 미리 계산해놓는 Pre-Aggregation을 사용합니다. Raw 데이터를 읽는 게 아니라 이미 계산된 메트릭만 읽기 때문에 빠르게 쿼리를 처리할 수 있습니다. 하지만,
- 사전에 미리 계산해놓지 않은 메트릭은 쿼리 할 수 없기 때문에, 새로운 종류의 분석을 수행하려면 데이터를 새로 인덱싱해야 합니다.
- 퍼널 분석 등의 고도화된 행동 분석은 Raw 데이터 없이는 불가능합니다.
따라서 Pre-Aggregation을 사용하는 OLAP DB들은 에어브릿지의 니즈를 충족하지 않았습니다.
### 대부분 솔루션의 성능적 한계 : High-Cardinality Shuffle
유저 행동 분석 쿼리들은 필연적으로 이벤트 데이터를 유저별로 Group By 하는 과정이 포함됩니다. 일반적인 [MapReduce](https://en.wikipedia.org/wiki/MapReduce) 기반 프레임워크에선 Group By 쿼리를 수행할 땐 [Shuffle](https://data-flair.training/blogs/shuffling-and-sorting-in-hadoop/)작업을 수행하는데요. 같은 키를 가진 데이터가 같은 파티션에 배치되게 분류한 후 네트워크를 통해 데이터를 전송합니다.
하지만 유저 ID는 매우 높은 Cardinality를 가지는 키기 때문에, Shuffle 과정에서 보틀넥이 크게 발생합니다. 예를 들어 하루에 수백만 이상이 사용하는 고객사가 리포트를 본다면 수백만 유저 ID가 네트워크에서 전송되어야 합니다. 이는 대부분의 솔루션에서 큰 성능 저하의 원인이 됩니다.
## 자체 개발의 필요성
기존 솔루션을 테스트해보면서 에어브릿지의 니즈를 충족하기 위해선 자체 개발해야 한다는 결론에 도착했습니다.
- **비용 효율성 문제**: Snowflake나 Redshift 등의 데이터 웨어하우스를 큰 규모로 스케일 하면 모든 분석 니즈를 커버하면서 성능도 챙길 수 있습니다. 하지만 상용 데이터 웨어하우스들은 OLAP에 최적화된 것이 아니라 범용적인 니즈를 커버하기 위한 솔루션이기 때문에 목표에 비해서 너무 큰 규모의 클러스터를 운영해야 했습니다.
하지만 유저 행동 분석을 위한 OLAP 쿼리라는 특수성과 데이터의 특성을 고려했을 때 최적화할 수 있는 여지가 무궁무진했기 때문에, 자체 개발로 클러스터 비용을 최대한 줄이고자 했습니다.
- **쿼리의 다양성**: 기존에 제공하던 트렌드 및 리텐션 분석 리포트뿐만 아니라, 퍼널 분석이나 ID 매칭 등의 다양한 니즈를 커버하기 위해서는 SQL 기반 데이터베이스로는 한계가 있었습니다. 지금까지는 Spark 등의 데이터 처리 프레임워크를 통해 이런 분석을 수행했지만, 이러한 솔루션들 역시 성능과 비용 효율성 문제가 있었기 때문에 에어브릿지를 위해 자체적으로 커스텀 할 수 있는 솔루션이 필요했습니다.
## Luft: 유저 행동 분석을 위해 설계한 데이터스토어
Luft는 이런 문제를 해결하기 위해 처음부터 유저 ID 기준 Group By된 유저 행동 분석 쿼리를 빠르게 수행할 수 있도록 설계한 데이터스토어입니다. **현재 에어브릿지에서는 수십 TB 규모의 유저 데이터로부터 코호트 분석을 5대 이하의 노드만으로 평균 3초 ~ 최대 10초 사이에 처리하고 있습니다.**
일반적인 RDBMS와는 달리 Luft의 데이터는 불변합니다. 일반적으로 OLAP 데이터스토어는 유저가 일으킨 행동 이벤트 데이터를 분석하는데, 이벤트 데이터는 한번 쌓이면 변하지 않기 때문에 많은 OLAP 데이터스토어들이 한번 데이터스토어에 유입(Ingestion) 된 데이터를 직접 수정될 수 없게 합니다. 대신 필요할 경우 같은 기간의 데이터를 새로 유입해 덮어쓰는 방식으로 보정할 수 있습니다.
이러한 **불변성**을 통해 DBMS가 필수적으로 해결해야 하는 많은 난제를 간단히 해결할 수 있었습니다. 예를 들어 각종 복잡한 일관성 문제를 해결하느라 클러스터 구조를 복잡하게 만드는 대신 Druid와 같이 심플한 클러스터 디자인을 채택할 수 있고, [Kafka처럼 데이터 캐싱 역할을 커널 페이지 캐시에 위임](https://docs.confluent.io/platform/current/kafka/design.html)해 복잡한 페이지 매니저 구현 없이도 오히려 더 높은 성능을 낼 수도 있습니다.
하지만 그중에서도 제일 큰 이점은, 에어브릿지에서 사용하고자 하는 형태의 데이터 저장 포맷을 설계할 수 있다는 점이였습니다.
## 스토리지 엔진
데이터 수정이 필요한 RDBMS에서는 B-Tree나 Skip-List 등의 자료구조를 사용합니다. 하지만 데이터 불변성 덕분에 자료구조를 유저 행동 분석에 맞게 최적화해 성능을 크게 끌어올릴 수 있었습니다. Luft는 Adroll 사에서 만든 TrailDB를 스토리지 엔진으로 채택했습니다.
TrailDB는 유저 이벤트를 저장하기 위해 설계된 로우스토어(Rowstore)입니다. 데이터는 유저별로 그룹화되고 시간순으로 정렬되어 저장되는데, 유저 이벤트 데이터의 특성을 고려한 다양한 인코딩을 적용합니다.
- _Delta Encoding_: 이벤트가 발생 시각(Event Timestamp) 순으로 정렬되기 때문에, 해당 발생 시각 정보는 이전 이벤트 대비 늘어난 만큼만 저장합니다.
- _Dictionary Encoding_: 실제 데이터 값을 저장하는 대신, 값을 사전화시켜서 그 ID만 저장합니다.
- _Edge Encoding_: 일반적으로 행동 데이터에서 유저에 관한 정보는 변경되지 않기 때문에 (e.g. 나이, 이메일, 디바이스 종류, ...) 이전 이벤트와 달라진 칼럼만 저장해 중복을 제거합니다.
[__wf_reserved_inherit]
이러한 데이터 구조 덕분에 유저 데이터를 TrailDB로 저장하면 디코딩 성능은 O(N)의 시간 복잡도를 그대로 유지한 채 **97%**의 놀라운 압축률을 기록했습니다. 실제로 에어브릿지에서 CSV로 추출한 13GB의 샘플 데이터가 TrailDB로 인코딩 시 300mb 수준으로 줄어들었습니다. 데이터 용량이 작다는 건 RAM에 더 많은 데이터를 캐시할 수 있다는 의미죠.
또한 이벤트 데이터가 저장 단계에서부터 이미 유저별로 파티셔닝되어 때문에 Group By 과정에서의 **셔플 성능 저하 문제를 해결할 수 있습니다.** 그룹화된 유저 데이터를 셔플링하는 게 개별 이벤트를 따로 셔플링하는것보다 부담이 덜하기 때문입니다. 심지어 유저별 데이터의 오프셋이 기록되어 있기 때문에, 원하는 유저의 데이터만을 읽을 수 있어 코호트 분석 시 읽어들일 데이터의 양을 크게 줄일 수 있습니다.
## You Only Scan Once 🏇
일반적인 사용자 행동 분석 쿼리는 많은 서브 쿼리로 이루어져 있습니다. 데이터 스캔 시간을 단축하기 위해, Luft에서는 한 번에 필요한 모든 데이터를 스캔하자는 디자인 원칙을 세웠습니다.
에어브릿지의 유저 이벤트는 시계열 데이터이며, 모든 쿼리에도 반드시 기간이 들어갑니다. 따라서 Luft도 이벤트 발생 시각에 따라서 시계열로 데이터를 파티셔닝합니다. 그런데 서브 쿼리들 간에 스캔하는 기간이 겹칠 경우 필연적으로 데이터를 중복해서 스캔하는 기간이 생기게 됩니다.
[__wf_reserved_decorative]
Luft는 쿼리 플래닝 단계에서 한 번의 스캔만 이루어질 수 있게 쿼리를 최적화합니다. 다양한 서브 쿼리들을 시간축에 따라 쪼갠 후, 유사한 조건을 가진 쿼리끼리 머지해서 최적화된 서브 쿼리를 만들어냅니다. 쿼리 실행 후 결과들은 조인 과정에서 다시 결합되기 때문에, 중복 없이 단 한 번의 스캔만으로 복잡한 쿼리를 빠르게 수행할 수 있습니다.
## Architecture for Consistency
한편 실시간 데이터 쿼리를 위해 기존 솔루션에서 **람다 아키텍처(Lambda Architecture)**를 차용했습니다. 람다 아키텍처는 배치 방식과 스트리밍 방식을 함께 활용하는 데이터 처리 아키텍처인데요. 데이터 파이프라인을 실시간인 **스피드 레이어**와 **배치 레이어**로 분리하고 실제 서빙 레이어에서 하나로 합쳐서 데이터를 처리합니다. 정합성과 일관성이 떨어지는 실시간 데이터를 나중에 배치 데이터로 덮어쓰기 때문에 **결과적 일관성(Eventual Consistency)**을 보장할 수 있다는 장점이 있습니다.
[__wf_reserved_decorative]
Luft는 람다 아키텍처를 사용하고 있는 Druid의 구조를 차용해 노드의 역할을 다음과 같이 나눴습니다.
- **리얼타임 노드:** Kafka 등의 데이터 스트림을 구독하며 실시간 데이터를 저장하는 샤드입니다. 실시간으로 들어오는 데이터를 바로바로 저장해 쿼리 될 수 있게 합니다.
- **히스토리컬 노드:** 배치 데이터를 저장하는 샤드입니다. 배치 데이터를 캐싱하고 쿼리를 수행합니다.
그럼에도 불구하고, 데이터를 여러 노드에 샤딩해 저장하다 보면 장애로 인해서 개별 노드가 가진 데이터가 유실되거나 노드 간에 일관성이 맞지 않는 문제가 생길 수 있습니다. 이를 방지하기 위해 데이터는 기본적으로 S3에 저장되고 필요한 데이터만 실제 노드의 디스크 / 메모리에 저장됩니다.
물론 이렇게 하면 만약 쿼리 될 데이터가 실제 노드에 없는 경우 S3에서 다운로드하느라 시간이 오래 걸릴 수 있다는 문제가 있지만, 쿼리 될 가능성이 높은 최근 데이터 위주로 노드에 저장시키거나 리눅스의 [User-Space Page Fault Handler를 이용해 S3에서 데이터를 빠르게 불러오는 방법](https://tech.nextroll.com/blog/data/2016/11/29/traildb-mmap-s3.html) 등을 통해 Cache Miss에 대비하고 있습니다.
## 이외에도... 기존 솔루션에서 얻은 인사이트 📚
Luft를 개발하면서 기존에 만들어진 좋은 솔루션으로부터 많은 인사이트를 얻었습니다.
- Spark와 Kafka의 디자인을 많이 참고해 Golang과 gRPC로 MapReduce 연산 레이어를 만들었습니다. 파티션 스케줄링 방식과 Pull-based Stream 구조를 통해 높은 성능으로 온라인 쿼리를 처리할 수 있었습니다.
- [etcd](https://etcd.io/)로 클러스터를 구성합니다. 일반적으로 ZooKeeper가 많이 사용되지만, ZooKeeper 대비 낮은 운영도와 제약 사항, 그리고 Kubernetes와 같은 프로젝트에서 이미 검증된 바 있다는 점으로 인해 etcd를 채택했습니다.
- 클라우드를 최대한 활용합니다. 데이터를 S3에 저장하고 메타데이터는 DynamoDB에 관리하며, Kubernetes 워크로드의 스케일링을 직접 관리하면서 클러스터 운영 비용을 효과적으로 절감할 수 있었습니다.
## Conclusion
에어브릿지를 통해 중소규모 클러스터만으로도 테라바이트 스케일의 유저 행동 분석을 실시간으로 수행하는 것을 목표로 하고 있습니다. 단기적으로는 수십 TB 규모의 유저 데이터 위에서 퍼널 분석을 40 CPU Core 미만으로 처리할 수 있도록 개발 중에 있으며, 이를 위해서 성능을 최적화하고 데이터 구조를 개선하는 데 집중하고 있습니다.
이 Luft를 기반으로 에어브릿지의 기능들을 계속해서 업그레이드할 예정입니다.
> 웹&앱 통합 광고 성과 어트리뷰션 솔루션 에어브릿지에 대한 더 자세한 내용이 궁금하시다면, 에어브릿지의 어트리뷰션 전문가들과 만나보세요.
> [👉_에어브릿지 데모 신청하러 가기_](https://www.airbridge.io/ko/book-a-demo?utm_source=blog&utm_medium=link&utm_campaign=introducing-luft)
---
## [딥링크101] 디퍼드 딥링크로 앱을 설치하지 않은 유저도 앱 내 특정 페이지로 연결하기
URL: https://www.airbridge.io/ko/blog/deeplink-101-deferred-deep-linking
Language: Korean
Published: 2022-04-08
[__wf_reserved_decorative]
앱을 설치하지 않은 유저에게도, 링크를 클릭하여 앱을 설치한 후에도 관련 페이지로 연결하고 싶으신가요? 그렇다면, ‘디퍼드 딥링크(Deferred Deep Linking)’가 필요합니다.
> 딥링크에 대한 더 자세한 내용이 궁금하시다면, <[에어브릿지 핸드북 - 딥링크](https://www.airbridge.io/ko/library/airbridge-handbook-deeplinks)\> 가이드를 확인해보세요.
## 디퍼드 딥링크란?
딥링크에 추가적으로 활용할 수 있는 기술로, 문자 그대로 ‘지연된 딥링크’를 의미합니다. 앱이 설치되지 않은 유저에게는 딥링킹을 할 수 없기 때문에, 해당 유저가 링크를 클릭한 후 앱을 설치하고 앱을 실행하기까지 기다렸다가 딥링크를 실행해 주는 기술입니다. 이를 위해서는 유저가 링크를 클릭했을 때의 정보와, 앱을 설치하고 오픈한 유저의 정보는 물론 딥링크를 보관하고 내려줄 서버가 필요하기 때문에 직접 구현하는 것이 쉽지 않죠.
이번 포스트에서 디퍼드 딥링크 기술이 필요한 이유, 그리고 이를 구현하는 방법을 확인해 보세요.
## 디퍼드 딥링크가 필요한 이유?
‘딥링크’는 모바일 환경에 놓인 유저가 링크를 클릭했을 때, 앱을 실행시키고 특정 페이지로 이동하도록 돕는 기술입니다. 즉, 앱으로 향하는 특정한 경로가 지정된 링크이죠. 하지만 딥링크를 클릭했을 때, 앱이 설치되어 있지 않다면 앱 마켓으로 연결되거나 오류가 발생하면서 해당 링크/경로가 제대로 작동하지 않습니다. 다시 말해, 딥링크를 클릭한 유저라도 앱 내 특정 페이지로 연결되지 않는다는 것을 의미합니다.
구체적으로는, 앱이 설치되지 않은 경우엔 다음과 같은 현상들이 발생할 수 있습니다.
- URI Scheme 딥링크를 사용했을 때는, 앱이 설치되어 있지 않다면 앱 마켓 등 다른 곳으로의 이동이 불가능하기 때문에 아무런 동작이 일어나지 않습니다.
- Universal Links 및 App Links의 경우 App Store 및 Google Play로 바로 랜딩 시키는 URL 사용이 불가능하기 때문에, 앱이 설치되지 않은 유저에게 앱 설치를 유도하기조차 어렵습니다. (Universal Links 및 App Links URL로 특정 페이지에서 자바스크립트를 동작시켜서 앱 마켓으로 이동시킬 수는 있으나, 이는 불완전할 수 있다는 점을 염두해야 합니다.)
- 위와 같은 상황에 대처하고자 MMP에서 제공하는 트래킹 링크로 딥링크만 사용하는 것도 완전한 유저 경험을 제공하지는 않습니다. 사용하는 MMP의 트래킹 링크에 디퍼드 딥링크가 기본으로 설정되어 있지 않은 경우, 리다이렉트 설정으로 앱 마켓까지는 이동할 수 있겠지만 앱 설치 후 앱을 오픈한 다음 의도했던 특정 페이지로는 연결되지 않기 때문입니다.
이러한 기능의 한계로 인해, 광고 클릭 후 아무런 동작이 발생하지 않거나 앱 설치까지 하는 번거로운 과정 끝에 앱의 메인 페이지만 나타나게 된다면, 광고에 소구된 잠재 고객이라도 즉시 이탈할 수 있습니다. 혹은 앱의 메인 페이지를 보고 당황한 고객이 다시 광고에서 본 페이지를 찾아 헤매는 경우도 있을 겁니다. 이는 결코 좋은 고객 경험이라고 할 수 없습니다.
**이러한 문제를 해결해 줄 수 있는 기술이 바로 ‘디퍼드 딥링크’입니다.**
앱을 설치하지 않은 유저라도 광고를 보고 앱을 설치했다면, 설치 후 앱 내에 관련 내용으로 연결되는 것이 자연스럽다고 느낄 겁니다. 충분한 전환 의사를 가지고 앱을 설치한 유저에게 기대했던 바를 즉시 보여주어야 하죠. 그리고 앱 설치 후 소구 된 광고 내용과 연관된 페이지로 연결되면 더 원활하게 상품 및 서비스를 탐색할 수 있기 때문에, 전환할 가능성까지도 높아집니다. 앱 마케팅을 한다면 이와 같은 더 좋은 유저 경험을 제공할 수 있는 기회를 놓쳐서는 안됩니다. 이것이 바로 앱 마케팅에 ‘디퍼드 딥링크’가 반드시 필요한 이유입니다.
%%CTA|데모 신청하기|/ko/book-a-demo|에어브릿지 딥링크로 앱 설치 전 유저도 원하는 화면으로 바로 연결하세요.%%
## 디퍼드 딥링크 구현하는 방법
다만, 디퍼드 딥링크 기술은 직접 구현하기엔 여러 어려움이 발생할 수 있습니다. Android OS 레벨에서는 Google Play Install Referrer API를 활용하여 구현할 수 있으나 제한적입니다. 게다가 iOS에서는 OS 레벨에서 디퍼드 딥링크를 구현할 수 있는 방법이 제공되지 않기 때문에 따로 서버를 구축하는 등 복잡한 환경이 필요합니다.
### 쉽지 않은 디퍼드 딥링크 기술의 직접 구현
우선, **Android에서는** Google Play 스토어의 [Google Play Install Referrer](https://developer.android.com/google/play/installreferrer) API를 활용하여 Google Play에서 설치된 패키지의 리퍼러 URL 및 리퍼러 클릭이 발생한 시간의 타임스탬프 등을 가져올 수 있기 때문에 디퍼드 딥링크를 자체적으로 구현할 수 있습니다.
하지만 이 방법은 앱이 설치되어 있지 않아야 하며, 플레이 스토어 URL을 구성해야 하므로 크롬 인텐트와 같이 사용할 수 없습니다. 또한, 구현 방법이 복잡하기 때문에 데이터 수집에 이슈가 생길 가능성이 높으며, Google Play에서 다운로드한 경우에만 동작한다는 제한도 있습니다.
이렇게 Google Play Install Referrer를 활용하여 Android에서 구현했다고 하더라도, **iOS에서는**비슷한 방법으로는 구현할 수 없기 때문에 결국 서버를 구축하는 추가적인 과정이 필요합니다.
### 모바일 어트리뷰션 솔루션으로 손쉽게 디퍼드 딥링크 구현하기
자체적으로는 구현하기 어려운 디퍼드 딥링크, 어떻게하면 효율적이고 완전하게 구현할 수 있을까요? 바로, 모바일 어트리뷰션 툴에서 제공하는 솔루션을 활용하는 겁니다.
디퍼드 딥링크는 유저가 링크를 클릭한 후, 앱을 설치하고 실행할 때까지 딥링크를 지연시켜주는 기술입니다. 다시 말하자면, **유저가 클릭했을 때의 정보를 보관하고 있다가 앱이 실행되었을 때 클릭 정보와 앱을 오픈한 유저를 매칭하고 딥링크를 내려줄 수 있는 서버**가 필요합니다. 이는 어트리뷰션 툴에서 트래킹 링크 클릭 후 클릭 데이터와 유저 데이터를 매칭하여 어트리뷰션을 하는 과정과 유사하죠. 실제로 어트리뷰션 툴에서 이 과정을 활용하여 디퍼드 딥링크를 제공하곤 합니다.
어트리뷰션 툴 에어브릿지를 통해 디퍼드 딥링크를 구현하는 과정을 알아보면 다음과 같습니다.
- 에어브릿지 트래킹 링크를 클릭하게 됨 -> 에어브릿지 서버에 클릭한 링크 정보를 서버에서 보관 -> 유저가 앱 설치 또는 SDK가 설치된 앱 최초 설치 -> SDK가 포함된 앱이 최초 실행된경우 서버에 해당 유저가 열렸어야 했던 링크 정보가 있는지 확인 -> 서버에서 열렸어야할 딥링크 주소를 SDK로 전달 -> 앱에서 다시 해당 링크의 딥링크 동작을 반복하여 지정된 앱 내 특정 페이지를 열어줌
구체적으로 딥링크와 디퍼드 딥링크를 설정하는 방법은 아래의 각 OS 별 SDK 개발자 가이드에서 확인할 수 있습니다.
- 에어브릿지 [Android SDK](https://developers.airbridge.io/docs/android-sdk#%EB%94%94%ED%8D%BC%EB%93%9C-%EB%94%A5%EB%A7%81%ED%81%AC-%EC%84%A4%EC%A0%95) 가이드
- 에어브릿지 [iOS SDK](https://developers.airbridge.io/docs/ios-sdk#deeplink-callback-%EC%84%A4%EC%A0%95) 가이드
### 디퍼드 딥링크를 손쉽게 적용할 수 있는 에어브릿지 트래킹 링크
에어브릿지의 경우, 트래킹 링크를 딥링크로 설정하면 디퍼드 딥링크 기능이 디폴트로 구현됩니다. 즉, 트래킹 링크 설정 한 번으로, 링크 클릭 후 앱이 설치되어 있다면 바로 해당 페이지로, 앱이 설치되어 있지 않다면 앱 설치 후 해당 페이지로 이동할 수 있습니다.
[__wf_reserved_decorative]
## 앱 마케팅을 원활하게 진행하기 위해서는 MMP의 트래킹 링크를 사용하는 것이 좋습니다.
이와 더불어, 마케팅을 할 때 활용하는 앱 설치 링크와 딥링크(URI Scheme, Universal Links, App Links), 디퍼드 딥링크 등의 기능을 OS 업데이트와 유지 보수의 이슈 없이 사용하기 위해서는, 자체 개발 보다는 모바일 어트리뷰션 툴(MMP)의 트래킹 링크를 사용하는 것이 좋습니다. MMP의 트래킹 링크를 통해 OS 별 복잡한 구현 과정 없이 편리하게 원하는 유저 여정을 설계할 수 있죠. 특히 에어브릿지는 어떠한 상황에서도 문제 없이 트래킹 링크로 원하는 기능을 구현할 수 있도록 지속적인 업데이트와 개발을 이어가고 있습니다.
앱 마케팅 시 반드시 필요한 기술인 ‘디퍼드 딥링크’, 에어브릿지 트래킹 링크를 활용하여 효율적으로 구현해보세요.
> **\* 다른 딥링크101 시리즈 보러가기**
> 👉[\[딥링크101\] 마케터와 개발자를 위한 딥링크 시작하기](https://www.airbridge.io/ko/blog/deeplink-101-for-marketers-and-developers)
> 👉[\[딥링크101\] iOS Safari에서 딥링크 동작 시 경고 문구가 뜨는 이유가 무엇일까요?](https://www.airbridge.io/ko/blog/deeplink-101-ios-safari-alert)
---
## 모바일 앱 유저 획득 마케팅의 A to Z 알아보기
URL: https://www.airbridge.io/ko/blog/ua-marketing-a-to-z
Language: Korean
Published: 2022-03-30
유저가 없는 프로덕트는 실패한 것이나 마찬가지입니다. 우리 프로덕트의 성공을 위해서는 ‘유저’가 필수이죠. 유저의 시간과 마음을 붙잡기 위해 수많은 앱들이 경쟁하는 속에서 반드시 준비해야 할 유저 획득(User Acquisition, UA) 마케팅의 모든 것을 알아보겠습니다.
각 제품의 진정한 성장을 위한 궁극적인 목표는 단순한 유저의 존재 자체가 아닌, **유저를 획득하고 그 유저를 활성화시켜서 진성 유저로 만들고, 더 나아가 파워 유저를 늘리는 것**입니다. 이를 위한 유저 여정의 첫 시작점인 유저 획득 앱 마케팅은, 단기적인 성과를 냄과 동시에 지속적인 성장을 고려한 전략을 수립해야 합니다.
이번 포스트의 주요 내용은 다음과 같습니다.
**1) 프로덕트 론칭 전 준비해야할 것 2) 프로덕트 초기에 유저 획득의 중요성 3) 유적 획득의 두 가지 전략 - Organic & Paid 4) 딥링크의 중요성 5) UA 캠페인 성과 제대로 측정하기**
## 프로덕트 론칭 전, 이미 마케팅은 시작됩니다.
모바일 앱 마케팅은 프로덕트 론칭 전부터 시작되어야 합니다. 브랜딩, 고객 이해, 경쟁사 분석 등이 포함되며, 이를 기반으로 전체적인 앱 마케팅 기획을 시작해야 하죠.
### 브랜드 및 프로덕트에 대한 인지도 높이기
브랜드 및 프로덕트를 가능한 많은 사람들에게 알리는 것은 마케터가 프로덕트 론칭 전에 수행할 수 있는 주요 전략 중 하나입니다. 이는 브랜딩의 영역일 수도 있겠지만, 우리 프로덕트에 우호적인 사람들은 실제 프로덕트 론칭 후에 전환되고 진성 유저가 될 가능성이 높아지기 때문에 마케팅에서도 놓쳐서는 안되는 부분입니다. 고객이 우리 브랜드의 가치와 목적을 인지하고 공감하는 것만으로도 추후 마케팅에 더 큰 효과를 불러일으킬 겁니다.
또한, 이는 **그로스 해킹**에서 중요하게 여겨지는 마케팅 방법론 중에 하나이기도 합니다. [라이언 홀리데이(Ryan Holiday)](https://ryanholiday.net/)는 그의 저서 [그로스 해킹(Growth Hacker Marketing)]에서 최초 이용자 집단에게 다가가기 위해 사용할 수 있는 여러 가지 방법을 소개했습니다. 잠재 유저들이 자주 방문하는 웹사이트에 우리 브랜드 및 프로덕트를 소개할 수도 있으며, 관련 분야에서 화제가 되는 주제를 다루는 블로그 포스트를 게재하는 방법도 제안했습니다. 그는 실제로 [그로스 해킹]을 출판하는 과정에 책에서 소개한 방법들을 적용하였는데, 출판 전에 간결한 전자책 형태로 먼저 출판하여 시장을 테스트하고 시장의 경쟁자들을 압도하였습니다. 또한 독자들의 피드백을 받아서 내용을 개선했을 뿐만 아니라 마케팅하기에도 쉬워졌습니다. 책 출시 전에는 관련한 글을 작성하여 영향력 있는 웹사이트에 기고 했으며, 이미 확보하고 있던 이메일 목록을 바탕으로 쉽게 홍보할 수 있었습니다.
### 고객 및 경쟁사 분석으로 페르소나 설정하기
앱 마케팅을 본격적으로 시작하기에 앞서, 우리 프로덕트가 어떤 유저 층을 타겟으로 하는지 확실히하고, 고객과 시장을 이해하고 적절한 페르소나를 설정하는 과정도 필요합니다. 고객을 이해하기 위해서는 정성적 및 정량적 데이터를 모두 종합적으로 활용하여 파악해야하며, 이는 비즈니스 목표를 성취하기 위한 중요한 열쇠가 될 겁니다.
고객 조사는 친구, 가족, 잠재 고객 대상의 설문 조사부터, 시장의 욕구와 관심을 정확한 수치로 분석하는 것 까지 넓은 스펙트럼 속에서 진행할 수 있습니다. 이때, 유저들이 우리 서비스 및 제품에 기대하는 바를 파악하여 새로운 기능을 더하거나 UI/UX를 개선함으로써 유저 중심의 프로덕트 발전으로도 이어질 수 있습니다. 이는 프로덕트 런칭 이후로도 고객에 대한 이해와 분석이 계속되어야 하는 이유이기도 합니다.
Forbes의 [신규 프로덕트 혹은 서비스 론칭 전에 니즈를 확인할 수 있는 14가지 방법(14 Ways To Determine Need Before Launching A New Product Or Service)](https://www.forbes.com/sites/forbesbusinesscouncil/2021/07/06/14-ways-to-determine-need-before-launching-a-new-product-or-service/?sh=757ef8c113f2) 기사에서 소개한 내용 중 고객 및 경쟁사 분석에 도움이 될 수 있는 방법 몇 가지를 발췌하여 소개합니다.
- [**Google 트렌드**](https://trends.google.co.kr/trends/?geo=KR)**를 사용해 보세요**: 구글 트렌드를 통해 전 세계 사용자들의 검색어 동향을 살펴볼 수 있습니다. 이 지표를 활용하여 우리 잠재 고객들이 어떤 정보를 검색하고 있는지 확인하고, 이에 대한 콘텐츠 및 서비스를 기획해볼 수 있습니다. 다만, 정보 검색의 양과 실제 관심도 및 구매 의사는 다를 수 있다는 점을 주의해야 합니다.
- **경쟁 제품 및 서비스를 분석하여 틈새를 공략해 보세요**: 특히 같은 업계의 1위 제품 및 서비스를 철저하게 분석합니다. 1위임에도 불구하고 디자인 혹은 기능 등의 측면에서 유저를 만족시키지 못하는 부분이 있을 겁니다. 해당 제품을 사용하는 유저를 대상으로 인터뷰 및 설문조사 해보는 것도 좋은 방법입니다.
- **제품의 베타 버전을 무료로 공개하거나, 사전 예약을 받아 보세요**: 소셜 미디어에서 무료로 론칭하거나 사전 예약을 받으며 잠재 고객들의 반응을 확인해보세요. 긍정적이든 부정적이든 그 원인을 분석하고, 긍정적인 요소를 강화하거나 부정적인 부분을 개선할 수 있는 방법을 고민해야합니다.
### 구체적인 앱 마케팅 전략 세우기
고객과 경쟁사에 대해 분석했다면, 이를 기반으로 구체적인 앱 마케팅 전략을 세워야 합니다. 전략을 어떻게 세워야할지 막막하다면, ‘앱 마케팅 퍼널’을 기반으로 구상해보는 것도 좋습니다. 각 퍼널 별로 수행해야 할 전략을 세우고, 퍼널을 통과하기 위한 조건을 명확하게 세워보는 겁니다. 단, 각 단계들을 완전히 독립적으로 놓고 고려하는 것이 아닌, 우리 제품 및 서비스에 맞는 퍼널의 흐름을 이해하고 단계들이 유기적으로 연결될 수 있도록 전략을 세우는 것이 중요합니다.
[👉 <앱 마케팅 시작을 위해 반드시 알아야 하는 앱 마케팅 퍼널> 포스트에서 자세한 내용 확인해보기](https://blog.ab180.co/posts/app-marketing-funnel)
또한, 이 전략은 수치로 된 KPI(Key Performance Indicator, 핵심성과지표)와도 연관지을 수 있어야 합니다. 주로 고려해야할 KPI 요소로는 DAU와 MAU, 설치당 비용(CPI, Cost per Install), 획득당 비용(CPA, Cost per Acquisition), 클릭률(CTR, Click-Through Rate), 전환율 및 이탈율 등이 있습니다. 각 유저 여정 단계에 맞는 적절한 메트릭을 트래킹하여, 유저가 유입부터 전환 및 리텐션에 이르는 과정이 매끄럽게 진행되고 있는지 확인해야하기 때문입니다.
## 프로덕트의 성패는 앱 론칭 초기의 유저 획득에 달려있습니다.
프로덕트 출시 전 유저 획득을 위한 준비를 마쳤다면, 다음은 본격적으로 유저를 우리 앱으로 유입시킬 차례입니다. 대부분 모바일 프로덕트의 경우, 초기에 유저들을 얼마만큼 많이 확보하는가에 따라서 이후의 성패가 갈리곤 합니다. **프로덕트 초기에 유입된 유저들이 장기간 지속적인 우리 고객이 될 가능성이 높기 때문입니다.** 실제로 많은 앱 및 서비스들이 출시 직후 대대적인 마케팅을 진행하고 있으며, 마케터라면 이 시기를 절대로 놓쳐서는 안됩니다. 최대한 많은 유저들에게 우리 앱을 알리고, 앱 설치를 유도하고, 성공적으로 온보딩을 시켜야하죠.
우리 앱을 알리고 유저를 획득하기 위한 본격적인 모바일 앱 마케팅 전략을 Organic 및 Paid 두 가지로 나눠서 살펴보겠습니다.
### Organic 전략
1. 우리 앱을 소개하는 랜딩 웹페이지 만들기
모바일 앱이 주력인 서비스라고 하더라도 랜딩 및 앱을 소개하기 위한 웹페이지가 반드시 필요합니다.
많은 유저들이 니즈가 있거나 어떤 문제가 있을 때면 웹에서 서치하며, 특정 제품 및 서비스를 알아볼 때도 앱을 다운로드 받기 보다는 웹에서 먼저 검색하기 때문입니다. 웹사이트는 우리 앱의 첫인상과도 같습니다. 웹 서치를 통해 우리 앱을 발견했다고 하더라도, 웹사이트가 없다면 ‘앱 마켓에서 다운로드’해야 한다는 장벽으로 인해 유저가 이탈할 수 있습니다. 또한, 프로모션을 진행한 경우에는 앱으로 이동하기 전에 상세한 안내를 위한 웹페이지가 필요하며, 특히 검색 광고(SA, Search Ad) 등 일부 광고의 경우 웹으로 리다이렉트 할 수 밖에 없기 때문에 웹페이지가 필수적이기도 합니다.
게다가 웹페이지는 Owned 미디어이기 때문에 별도의 광고비가 발생하지 않으며, 검색엔진최적화(SEO)를 통한 신규 유저 확보에 도움이 된다는 장점이 있습니다. 유저들이 이미 웹사이트를 통해서 서비스 및 프로모션에 대해 더 많은 정보를 인지한 상태에서 앱에 유입될 것이기 때문에 앱 설치 후 바로 이탈할 가능성도 줄어들게 됩니다.
이 웹페이지에는 유저가 앱을 설치한 후 경험할 수 있는 부분을 영상 및 이미지 등을 통해 보여주는 것이 중요합니다. 또한, 앱 마켓으로 이어지는 링크 및 명확한 콜투액션(CTA, Call to Action)을 포함하여 유저의 다음 행동을 명확하게 안내해야합니다. 혹은 앱 출시 전에 뉴스레터 및 소식 전달 용으로 이메일 주소를 수집하여, 앱의 출시와 업데이트 등을 알리며 유저와 소통의 기반을 만들 수도 있습니다.
1. 콘텐츠 마케팅
잠재 고객을 사로잡기 위한 방법으로는 콘텐츠 마케팅이 매우 효과적입니다. 콘텐츠 마케팅의 핵심은 목표로 할 **‘타겟 오디언스’**, 달성하고자 하는 **‘비즈니스 목표**’,** **그리고** ‘양질의 콘텐츠’**입니다.
우선, 우리 프로덕트의 유저 기반은 누구인지, 타겟 유저들이 주로 어느 웹사이트 및 소셜 미디어에 모여있는지, 관심사는 무엇인지, 유저들이 앱을 사용하는 이유 등 우리 유저와 업계 동향을 분석하여 콘텐츠 마케팅을 기획해야 합니다. 이 기획으로 달성할 구체적인 목표도 함께 정해놓는 것이 좋습니다.
기획을 완성했다면 다음은 실제 콘텐츠를 제작하여 유포해야 합니다. 여기서 ‘콘텐츠’는 소셜 미디어에 올라가는 숏폼 게시물은 물론, 블로그 게시물, 플레이북, 고객 사례 및 영상자료, 그리고 뉴스레터와 이메일 등을 모두 포함할 수 있습니다. 콘텐츠 마케팅 집행 시 주요 채널별로 주목해야할 점들을 간단히 소개합니다.
- **소셜 미디어**: 2020년 전 세계의 일간 평균 소셜 미디어 사용 시간은 2시간 25분입니다. 2012년 1시간 30분에 비해 150%이상 증가한 수치입니다. ([statista 통계](https://www.statista.com/statistics/433871/daily-social-media-usage-worldwide/)) 많은 사람들이 시간을 보내는 소셜 미디어는 광고 외에도 다양한 콘텐츠를 통해 자연스럽게 고객에게 접근하고 우리 프로덕트를 알릴 수 있는 효과적인 수단입니다. 게다가 단순히 콘텐츠를 게시하는 것 외에도, 커뮤니티를 만들거나 코멘트 등을 통해 고객의 피드백을 받을 수 있는 채널이기도 합니다.
- **블로그**: 상대적으로 짧은 내용 및 이미지와 영상 위주인 소셜 미디어에 비해, 블로그는 정보성을 띄며 길이가 긴 콘텐츠가 주를 이룹니다. 콘텐츠가 업계 및 트렌드와 연관된 가치있고 좋은 인사이트를 전할수록 더 많이 공유되며, 계속해서 더 많은 사람들이 찾게 될 겁니다. 또한, SEO를 통한 검색엔진 노출로 유저들을 확보할 수 있는 효과적인 방법입니다. 앞서 소개한 책 [그로스 해킹]에서도 블로그를 통한 잠재 유저의 확보 및 소통의 중요성에 대해 언급하기도 했죠.
- **뉴스레터 및 이메일**: 블로그를 통해 콘텐츠를 게재했다면, 뉴스레터를 통해 이 소식을 구독자들에게 전달할 수 있습니다. 우리 앱 프로덕트의 최신 업데이트 및 관련 업계 동향을 지속적으로 전해간다면 현재 활성화된 유저는 더 깊이 인게이지할 것이며, 휴면 유저라도 우리 프로덕트를 잊지 않게 될 겁니다. 게다가 뉴스레터 및 이메일은 앱 밖에서 우리 고객과 인게이지할 수 있는 중요한 수단 중 하나입니다.
1. 미디어 노출 및 보도 자료 활용하기
앱 출시에 맞춰서 보도 자료를 배포하여 프로덕트 론칭 관련 소식이 언론에 노출이 된다면, 많은 사람들에게 자연스럽게 앱을 알릴 수 있습니다. 규모가 작거나 초기 스타트업의 경우 프로덕트 론칭 외에 PR까지 진행하는 것이 쉽지는 않겠지만, 언론 보도도 놓쳐서는 안 될 초기 유저 획득에 중요한 마케팅 수단 중 하나입니다.
1. ASO(App Store Optimization)
ASO는 SEO(Search Engine Optimization, 검색 엔진 최적화) 및 LPO(Landing Page Optimization)와 동일한 원리를 기반으로 한, SEO 및 LPO의 앱 마케팅 버전이라고 할 수 있습니다. [2020년 모든 Apple Search Ads 국가 및 지역의 App Store 데이터에 따르면](https://searchads.apple.com/kr), App Store 방문자 중 검색을 통해 다음에 사용할 앱을 찾는 방문자의 비율이 70%이며, 전체 다운로드 중 검색 후 발생하는 다운로드의 비율이 65% 일 정도로 앱 마켓에서의 노출은 중요합니다. 그렇기 때문에 유저들이 앱 마켓에서 앱의 이름을 검색할 때는 물론, 특정 키워드를 검색할 때에도 앱을 확실하게 찾을 수 있도록 최적화해야 합니다. 앱 마켓에서 노출되는 앱 소개 및 설명 문구, 미리보기 이미지 등을 계속해서 업데이트하고 테스트하는 과정을 통해 가시성을 높이고 궁극적으로는 더 많은 다운로드를 유도해야 합니다.
ASO의 세부적인 전략에 대해서는 다음 포스트에서 더 자세히 알아보겠습니다.
### Paid 전략
앱을 런칭한 후, 유료 광고를 통해 유저를 앱으로 유입시키는 것은 당연하고도 필수적인 전략입니다. 다양한 채널을 활용하여 캠페인을 진행한 후, 시간이 지남에 따라 광고 비용을 조정하고 미디어 믹스를 재구성하여 성과를 높여야 합니다.
[]
위의 이미지는 Google과 Ipsos에서 공동 조사한 앱 다운로드에 영향을 주는 광고 유형의 비율을 보여주고 있습니다. 검색 광고, 소셜 미디어 광고, 인앱 배너/그래픽 광고 및 인웹 배너/그래픽 광고 등 모두 40~50%의 비중으로 앱 다운로드에 영향을 미치고 있기 때문에 Paid 마케팅 전략의 중요성을 간과할 수 없습니다. 이 광고 유형을 포함한 대표적인 Paid 전략들을 알아보겠습니다.
1. 앱 마켓에서의 광고 캠페인
대표적인 앱 마켓인 iOS App Store 및 Android Google Play에서 광고를 집행하여 앱 설치 및 참여를 유도할 수 있습니다. 앞서 언급한 것처럼 App Store에서 오가닉 검색 후 발생하는 다운로드가 65%를 차지하기 때문에, 앱 마켓에서 ASO 외에도 광고도 함께 진행하는 것이 효과적입니다.
우선, [Apple Search Ads](https://searchads.apple.com/kr)를 통해 App Store에서 앱을 검색할 때, 앱을 추천해주는 광고를 할 수 있습니다. iOS 기기 유저를 대상으로 앱 설치 광고를 진행할 때 빼놓을 수 없는 채널입니다.
또한, [Google Ads](https://support.google.com/google-ads/answer/6247380?visit_id=1-636549956449797142-1627836531&rd=2)를 통해 Google Play에서 앱 설치(ACi), 앱 참여(ACe), 앱 사전 등록(Android 전용) 세 가지 유형의 캠페인을 진행할 수 있습니다. Google Play외에도 검색, YouTube, Google 검색의 디스커버, Google 디스플레이 네트워크 등 Google의 다양한 서비스에서 앱을 홍보할 수 있기 때문에, iOS 및 Android 유저를 모두 대상으로 한 광고를 기획할 수 있습니다.
1. 소셜 미디어에서의 앱 설치 광고
[페이스북](https://www.facebook.com/business/goals/promote-app), [트위터](https://business.twitter.com/en/help/campaign-setup/create-an-app-installs-campaign.html), [틱톡](https://www.tiktok.com/business/ko/how-it-works) 등의 소셜 미디어 플랫폼을 통해 앱 설치 광고를 진행할 수도 있습니다. 이때 우리 앱의 주요 타겟 고객층이 모여있는 플랫폼을 활용하는 것이 중요합니다. 소셜 미디어에서 이미지와 영상을 통해 효과적으로 우리 앱을 소개하고 설치를 유도할 수 있습니다. 이때 Organic 전략에서 다루었던, 소셜미디어 및 Owned 미디어들의 콘텐츠들을 활용하여 시너지 효과를 발생시킬 수도 있습니다.
1. 오퍼월 등 리워드 광고
‘리워드 광고’는 앱을 설치하거나 상품을 구매하면 포인트를 적립받는 등 사용자에게 보상이 주어지는 광고 형태입니다. 특히 앱 이용자를 위한 보상 리스트를 보여주는 영역인 ‘오퍼월’을 통해 광고가 대표적입니다. 오퍼월은 쉽게 말하면 ‘무료 포인트 충전소’입니다. 앱 설치, 소셜 미디어 팔로우, 상품 구매 등의 행동을 완료했을 때 실시간으로 보상을 받을 수 있으며, 그렇게 적립된 포인트로 앱 내에서 유료 콘텐츠를 결제하거나 상품을 구매할 수 있습니다.
다양한 앱의 오퍼월 광고 지면을 활용하여 우리 앱의 설치를 유도하고 홍보할 수 있습니다. 이 리워드 광고의 경우 서비스 출시 초기 짧은 시간 내에 앱 마켓에서 순위를 올려야 하는 상황에 적합하다고 할 수 있습니다. 다만, 보상에만 목적을 가진 체리피커 유저들이 보상을 받은 후 앱을 삭제하여 이탈하는 경우도 상당수라는 점을 염두해야 합니다.
👉 [2021년 상반기 리워드 광고 시장 더 알아보기 (Platum 기사)](https://platum.kr/archives/165946)
1. 비보상형 광고(NCPI,Non-incentivized Cost Per Install 광고)
‘비보상형 광고’는 문자 그대로 특별한 보상 없이 앱 설치 및 앱 설치 후 행동을 유도하는 광고를 의미합니다. 배너 및 동영상 광고 등의 형태로 진행되며, 보통 CPM, CPI, CPC을 기준으로 과금이 진행됩니다. 보상형 광고와는 다르게, 비보상형 광고를 통해 자연스럽게 유입된 유저들은 우리 앱에서 더 오래 활성화되고 더 깊이 인게이지 할 가능성이 높습니다. 즉, 리텐션이 향상될 수 있다는 장점이 있습니다. 다만 앱 설치를 유도할 직접적인 넛지가 없기 때문에 유저 획득에 더 긴 시간 및 많은 예산이 소요될 수 있습니다.
NCPI 광고 집행 시에는 일부 매체에서 발생할 수 있는 인스톨 하이재킹(Install Hijacking), SDK Spoofing 등 광고 사기(AD Fraud)에 대비할 필요가 있습니다. 매체 내부에서도 필터링 기술을 갖추고는 있겠지만, 광고주의 자체적인 필터링 및 에어브릿지(Airbridge)와 같은 어트리뷰션 툴의 사기 설치 식별 기술을 활용하여 광고 사기를 더 효과적으로 방어할 수 있습니다.
## 딥링크로 전환된 유저 완벽하게 붙잡기
앞서 UA 마케팅의 다양한 전략을 활용하여 유저를 획득했다면, ‘딥링크(deeplink)’를 통해 앱을 설치한 유저를 더 확실하게 붙잡아야 합니다.
모바일 앱 마케팅이 일반적인 디지털 마케팅과 가장 구별되는 차이점은, ‘앱 마켓’을 통해 ‘앱 다운로드’를 한다는 과정이 추가된다는 점입니다. 광고 및 콘텐츠에서 유도하는 대로 앱 및 서비스에 유입된 고객이라고 하더라도, 앱 마켓으로 이동하는 순간 이탈할 수 있습니다. 혹은 특정 프로모션 광고에 이끌려 클릭 후 앱 마켓에서 다운로드 했을 때, 앱 설치 후 프로모션 페이지로 이어지지 않는다면 유저 경험을 헤치고 전환율이 낮아질 겁니다. 그렇기 때문에 ‘**딥링크**’, 특히 ‘**디퍼드 딥링크**’를 활용하여 유저를 광고 또는 웹페이지에서 관련 앱 내 특정 콘텐츠로 연결시켜야만 합니다.
[]
**디퍼드 딥링크**는 deferred deeplink, 즉 ‘지연된 딥링크’라는 뜻으로, 문자 그대로 딥링크 실행을 앱 설치가 완료될 때까지 잠시 지연시키는 것을 의미합니다. 디퍼드 딥링크를 활용한다면, 앱이 설치된 뒤에 딥링크가 실현되면서 앱 메인 화면 대신 이벤트 페이지나 제품 상세 페이지로 랜딩할 수 있게 됩니다. 유저가 기대하던 내용을 바로 마주하게 되면서 전환율도 개선할 수 있게 될 겁니다.
게다가 에어브릿지 딥링크의 경우, 디퍼드 딥링크로 유저의 리다이렉트를 돕는 동시에 어트리뷰션을 위한 트래킹 링크 역할을 하기 때문에 유저 행동을 추적하고 광고 성과 분석에 도움이 될 수 있습니다.
## UA 마케팅을 진행했다면, 성과도 제대로 측정해야합니다.
UA 마케팅 여정의 마지막 단계이자, 다시 시작점이 될 수 있는 전략은 **마케팅 성과 측정 및 분석**입니다. 마케팅을 집행했다면 실시간으로 데이터와 성과를 살펴보고 이를 기존 전략에 수정 반영해야하는 것은 당연합니다. 진행한 마케팅 캠페인들의 메트릭을 분석하고, 앞서 소개했던 앱 마케팅 전략 KPI와 비교하는 겁니다. 여기서 얻은 인사이트를 바탕으로 콘텐츠 전략 수정 및 미디어 믹스 재구성 등으로 성과를 개선해 나가는 것이 좋습니다.
특히 앱 마케팅의 경우, referral, 쿠키, UTM 파라미터 등을 통해 트래킹하는 웹과는 달리, 앱 마켓을 통과하여 다운로드 하는 과정이 추가되기 때문에, [에어브릿지](https://www.airbridge.io/ko)와 같은 ‘어트리뷰션 툴’, 즉 ‘광고 성과 측정 툴’이 반드시 필요합니다. 이러한 툴은 모든 앱 설치를 트래킹하고 각 설치마다 전환을 발생시킨 캠페인에 기여를 해주기 때문에 캠페인을 최적화하는 데에 유용한 데이터로 활용할 수 있습니다. 참고로 에어브릿지의 경우, 웹과 앱을 동시에 트래킹할 수 있으며, 웹&앱 광고 성과를 하나의 대시보드에서 확인할 수 있기 때문에 간편합니다. 각 매체별로 성과를 확인하는 번거로움 없이, 진짜 마케팅 성과 분석에만 집중할 수 있습니다.
## 마치며
모바일 앱 유저 획득을 위한 마케팅 전략의 모든 것을 알아보았습니다. 각 프로덕트 및 업계에 따라 가장 효과적인 마케팅 전략은 모두 다를 겁니다. 그럼에도 모든 프로덕트에서 유저 획득은 가장 중요한 부분이며, UA 마케팅은 우리 고객과 처음 만나는 지점이자 모바일 앱 마케팅의 시작점입니다. 이번 포스트에 소개한 전략들을 각 프로덕트에 맞게 적용 및 참고하여 성공적인 앱 프로덕트로 나아갈 수 있기를 바랍니다.
---
## 앱 스토어 최적화(ASO)란? 성공 가이드
URL: https://www.airbridge.io/ko/blog/aso-success-guide
Language: Korean
Published: 2022-03-17
# ASO(앱 스토어 최적화)란?
ASO(App Store Optimization)는 SEO(Search Engine Optimization, 검색 엔진 최적화) 및 LPO(Landing Page Optimization)와 동일한 원리를 기반으로 한, SEO 및 LPO의 앱 마케팅 버전이라고 할 수 있습니다.
앱 마켓의 검색 결과에서 더 높은 순위에 노출될 수 있도록 최적화하는 과정으로, 당연하게도, 앱 마켓에서 상단에 노출될수록 잠재 고객에게 더 많이 다가갈 수 있습니다.
> ASO의 목표는, 앱스토어 내의 앱 페이지로 더 많은 트래픽을 유도하는 것이며 이로 인해 검색하는 유저들이 특정한 액션, 즉 다운로드를 더 많이 수행하도록 만드는 것입니다. ([출처](https://neilpatel.com/blog/app-store-optimization/))
## ASO를 해야 하는 이유
<[모바일 앱 유저 획득 마케팅의 A to Z 알아보기](https://blog.ab180.co/posts/ua-marketing-a-to-z)\> 포스트에서도 다뤘지만, [구글의 리서치에 따르면](https://www.thinkwithgoogle.com/intl/ko-kr/marketing-strategies/app-and-mobile/mobile-app-marketing-insights/), 스마트폰 사용자의 40%는 앱스토어 탐색을 통해 앱을 찾으며, [2020년 Apple Search Ads 국가 및 지역의 App Store 데이터에 따르면](https://searchads.apple.com/kr), App Store 방문자 중 검색을 통해 다음에 사용할 앱을 찾는 방문자의 비율이 70%이며, 전체 다운로드 중 검색 후 발생하는 다운로드의 비율이 65% 일 정도로 앱 마켓에서의 노출이 중요합니다.
게다가 애플의 ATT 프레임워크 및 앞으로 예정된 구글의 크롬 서드파티 쿠키 제한 등 개인 정보 보호가 강화되는 트렌드 속에서 Organic 유저 획득 마케팅이 더욱더 중요해졌습니다. ASO는 Paid 마케팅과 다르게 무료라는 장점도 있으며, 장기적으로 유저들에게 우리 프로덕트의 노출 비율을 높이는 데에도 효과적인 전략입니다.
그렇기 때문에 유저들이 앱 마켓에서 앱의 이름을 검색할 때는 물론, 특정 키워드를 검색할 때도 앱을 확실하게 찾을 수 있도록 최적화해야 합니다. 그리고 발견한 앱 페이지를 매력적으로 구성하여 '앱 다운로드'로 이어질 수 있도록 만들어야 합니다.
## 성공적인 ASO를 위한 필수 전략 살펴보기
SEO를 해보신 분들이라면 그 기본 전략은 비슷하기 때문에 익숙하실 겁니다. 다만, '앱 마켓'이라는 특성에서 발생하는 차이점이 있다는 점은 주의해야 합니다. 성공적인 ASO를 위해 필요한 전략을 1. 키워드 2. 앱 다운로드 유도 3. App Store & Google Play의 가이드라인에서 얻을 수 있는 인사이트 순으로 살펴보겠습니다.
### 1\. 키워드
‘키워드’는 ASO에서 기본 중의 기본이라고 할 수 있습니다. 어떤 키워드를 사용하고 있느냐에 따라서 잠재 유저가 검색한 결과에 노출될지 여부가 결정되기 때문입니다. 같은 키워드를 사용하고 있는 프로덕트들과의 ASO 경쟁에서 프로덕트의 질, 다운로드 수 및 리뷰 등에 따라 노출 순위가 달라질 수 있겠지만, 절대적인 검색 노출량을 늘리기 위해서는 우리 프로덕트에 적합하면서도 잠재 유저가 검색할 만한 키워드를 제대로 선정하는 것이 중요합니다.
예를 들어, App Store 및 Google Play에 '이사, 이사 견적, 이사 비용' 등을 검색하면 에어브릿지(Airbridge) 고객사인 '짐싸'가 검색 결과가 최상단에 뜹니다. (2021.12.13 기준) 짐싸는 '이사, 이사 견적, 이사 비용' 등에 대한 키워드에 대해 우위를 선점하고 있음을 알 수 있습니다.
이렇게 상단에 노출되기 위해서는 잠재 고객 분석을 통해 검색할 만한 키워드를 알아내는 것은 물론, 우리 앱과 비슷한 프로덕트를 찾는 데에 어떤 키워드가 사용되는지 파악하는 것이 중요합니다. 다른 중요한 ASO 요소들에 비해 수정 및 테스트하기에 비교적 쉽고 직관적이기 때문에 정기적으로 관리해 주는 것이 좋습니다.
[__wf_reserved_decorative]
[__wf_reserved_decorative]
### 효과적인 키워드 선정을 위한 팁
이 키워드는 앱 마켓에서 메타 데이터(meta data)로 인정되는 앱 이름 및 제목, 설명, 키워드(App Store만 가능)에 모두 사용할 수 있습니다.
앱의 이름 및 제목 속에 키워드를 포함시키는 것이 좋지만, 글자 수에 한계가 있고(App Store - 최대 30자) 키워드만 마구잡이식으로 나열하는 것은 오히려 악영향을 끼칠 수 있기 때문에 적절한 키워드 선정에 주의를 기울여야 합니다. 효과적인 키워드를 선정하기 위한 몇 가지 방법을 살펴보겠습니다.
**1) 데이터 기반 리서치**
우선 앱과 함께 웹이 있는 프로덕트라면 [구글 서치 콘솔](https://search.google.com/) 및 [네이버 서치 어드바이저](https://searchadvisor.naver.com/)에서 우리 서비스를 어떻게 발견하는지 '검색어'를 리서치 할 수 있습니다. 웹과 앱 마켓에서 사용되는 검색어가 반드시 일치하는 것은 아니지만, 여태까지 쌓아온 소중한 퍼스트 파티 데이터를 통해 인사이트를 얻을 수 있습니다.
혹은, 처음 서비스를 런칭하거나 웹이 없는 프로덕트라면, [구글 키워드 플래너](https://support.google.com/google-ads/answer/7337243?hl=ko)를 활용하는 것도 좋습니다. ‘키워드 플래너’는 구글 애즈의 무료 도구 중 하나로, 비즈니스와 관련된 새로운 키워드를 탐색하고 해당 키워드의 예상 검색량 및 타겟팅 비용을 확인할 수 있습니다. 만약 새로운 키워드 아이디어가 떠오르지 않는다면, 아래에 간단히 정리한 [구글 애즈가 권장하는 사항 및 팁](https://support.google.com/google-ads/answer/9247190)을 참고해 봐도 좋습니다.
- 고객의 입장에서 생각해 보기: 고객이 우리 프로덕트나 서비스 카테고리를 생각할 때 떠올릴 것 같은 표현 사용하기
- 구체적인 표현을 사용하되, _너무_ 구체적인 표현은 사용하지 않기
- 웹사이트도 추가하기
- 프로덕트나 서비스에서 이미 인기있는 브랜드명이 있다면 함께 사용하기
- 프로덕트 또는 서비스를 잘 나타내는 표현 사용하기 : 형용사는 제한적으로만 사용하기
- 동일한 주제에 대해 여러 가지 키워드 및 문구 사용하기
혹은 [App Annie의 무료 키워드/ASO 탐색기](https://www.appannie.com/kr/insights/aso-app-store-optimization/how-to-improve-your-aso-app-store-optimization-by-maximizing-your-keyword-strategy/)를 이용하여 효과적인 키워드를 찾을 수도 있습니다.
**2) 우리 비즈니스에 적합한 키워드 선정하기**
데이터 기반 리서치를 통해 인사이트를 얻었다면, 우리 비즈니스에 적합한 키워드를 전략적으로 선정해야 합니다. 사용할 수 있는 키워드 개수에 한계가 있기도 하고, 키워드에 따라 경쟁 정도가 다르기 때문에 우리 프로덕트와 연관성이 있으면서도 빠르게 많은 유저들에게 노출될 수 있는 키워드를 사용해야 합니다.
키워드를 선정할 때는, 앞서 언급한 구글 키워드 플래너 및 App Annie의 키워드/ASO 탐색기와 같은 서비스를 활용하여 각 키워드의 검색량을 확인해 보는 것이 좋습니다. 주요 키워드에 대해 앱 마켓에 검색할 때 함께 제안되는 검색어들을 참고하는 방법도 있습니다.
예를 들어, '미세먼지' 키워드에 대해 구글 키워드 플래너에서 제안하는 아이디어는 다음과 같습니다.
이때 우리 비즈니스를 대표적으로 나타내는 키워드이면서도 동시에 일반적으로 사용되는 단어인, 알람, 날씨, SNS, 사진 등을 키워드로 사용하려고 한다면, 치열한 경쟁에 놓이게 될 겁니다. 반면 합성어 및 수식어가 덧붙여진 상세 키워드는 상대적으로 경쟁이 약할 수 있습니다.
처음 ASO를 진행한다면, 키워드를 선정할 때 대표적인 키워드만으로 구성하기보다는, 비교적 경쟁이 약한 롱테일 키워드를 먼저 주로 활용하는 것이 좋습니다. 경쟁이 약한 키워드를 통해 다운로드 수를 늘리고 나면 경쟁이 치열한 일반 키워드에 대한 우위도 함께 높아질 수 있기 때문입니다. 위의 '미세먼지' 키워드에 대한 예시를 생각해 본다면, '오늘 미세먼지', '초미세먼지', '내일 미세먼지' 등의 키워드를 고려해 볼 수 있겠습니다.
[__wf_reserved_decorative]
**3) 이외에 참고할 수 있는 몇 가지 팁**
- 키워드로 브랜드 및 프로덕트 명을 활용하는 것은 당연할 텐데요. 이때 해당 지역의 언어와 영문 키워드를 조합하는 것이 중요한데, 한국의 고객을 타겟으로 한다면 한글과 영문을 병기하는 것이 좋습니다.
- 가장 핵심이 되는 키워드는 제목에 넣고, 부가적인 내용은 부제목에 상세하게 넣습니다.
- 키워드 최적화만이 ASO의 전부는 아닙니다. 키워드를 아무리 잘 선정하여 넣는다고 하더라도 고객이 만족하지 못하는 프로덕트라면 상위에 노출될 수 없을 겁니다. 다만 우리 앱이 노출될 수 있는 키워드와 기회를 놓치지 않기 위해선 철저한 리서치를 해야 합니다.
- 키워드를 선정했다면, 해당 키워드들에 대한 검색 노출 성과를 계속해서 모니터링해야 합니다. 그리고 더 효과적인 키워드를 찾아 지속적으로 업데이트합니다.
### 2\. 발견 후 앱 다운로드로 유도하기
적절한 키워드 사용을 통해 앱의 발견 가능성을 높였다면, 다음은 유저가 앱을 다운로드할 만큼 매력적이면서도 직관적이고 빠르게 이해할 수 있는 상세 페이지(App Store: 제품 페이지 / Google Play: 스토어 등록정보 페이지)를 구성해야 합니다.
[__wf_reserved_decorative]
**1) 아이콘**
앱 마켓 검색 결과 및 카테고리 목록에서 가장 먼저 눈에 띄는 요소 중 하나는 '앱 아이콘'입니다. 때로는 앱 제목보다 더 먼저 눈에 들어와서 클릭을 유도하기도 합니다. 그뿐만 아니라 스마트폰의 메인 화면에서도 앱 아이콘은 다른 프로덕트들 사이에서 클릭에 중요한 영향을 미치기도 하죠. 그렇기 때문에 프로덕트의 아이덴티티를 잘 담고 있으면서도 앱의 목적 및 서비스와 직관적으로 연결될 수 있는 아이콘으로 디자인해야 합니다. 여러 옵션을 테스트하여 눈에 가장 잘 띄고 의미 있는 아이콘으로 결정해 보세요.
**2) 스크린샷 및 영상**
- App Store: 앱 미리보기 영상, 스크린샷
- Google Play: 그래픽 이미지, 스크린샷, 프로모션 동영상
프로덕트의 스크린샷 이미지 및 관련 영상은 앱 마켓 검색 결과는 물론 상세 페이지에서 시각적으로 매우 중요한 역할을 차지합니다. 잠재 유저에게 해당 앱 내에서 경험할 수 있는 바를 효과적으로 전할 수 있는 놓칠 수 없는 기회이죠. 유저가 프로덕트에서 원하고 얻고자 하는 가치 및 핵심 기능을 매력적으로 구성해야 하고, 가장 첫 두 이미지에 주요 콘텐츠가 노출될 수 있도록 만들어야 합니다.
스크린샷의 경우 정해진 사이즈가 있지만, 상단의 My Device ID by Airbridge 앱의 스크린샷 이미지와 같이 가로로 긴 이미지를 활용하고 싶은 경우, 두 개의 이어지는 이미지를 차례로 넣는 방법도 있습니다.
광고 캠페인 집행 시, 이미지 및 영상 크리에이티브에 대해 A/B 테스트를 진행하듯이, 앱 마켓의 스크린샷 및 영상에 대해서도 A/B 테스트는 중요합니다. 테스트를 통해 더 반응이 좋은 크리에이티브를 찾고, 정기적으로 업데이트하다 보면 우리 프로덕트에 최적화된 크리에이티브를 찾을 수 있게 될 겁니다.
스크린샷 이미지 제작에 [AppLaunchpad](https://theapplaunchpad.com/)나 [AppScreens](https://appscreens.com/), [Previewed](https://previewed.app/) 등과 같은 스크린샷 디자인 툴을 활용해 보는 것도 좋습니다.
**3) 설명 문구**
설명 부분은 앱 제목 및 키워드에 비해 넣을 수 있는 문구는 길지만, 그렇다고 해서 내용을 길게만 작성한다고 해서 더 성과가 좋을 것이라고는 할 수 없습니다. 유저들을 붙잡을 수 있는 문구는 가장 초반의 한두 문장일 겁니다. '더 보기'를 클릭해서 자세한 내용을 살펴볼 유저는 그다지 많지 않기 때문입니다. 설명 초반에 유저들에게 다운로드를 유도할 매력적인 문구를 넣는 것이 좋습니다. 이 또한, A/B 테스트를 통해 적합한 문구를 선정해도 좋습니다.
특히, 앱 설명 부분에 신규 업데이트 소식을 공유하면, 앱이 계속 업데이트되고 있음을 알리면서도 신규 기능을 잠재 유저에게 자연스럽게 홍보할 수 있습니다. App Store의 경우, 세일 및 신규 기능 등 앱 관련 최신 뉴스를 공유할 수 있는 홍보 문구 설명 상단에 기재할 수 있습니다.
**4) 평가 및 리뷰**
[Apps Store에서 **평가 및 리뷰는 앱의 검색 순위에 영향을 미친다**고 언급하고 있는 것처럼](https://developer.apple.com/kr/app-store/product-page/), 평가 및 리뷰는 ASO에 중요한 영향을 미칩니다. 이와 더불어, 잠재 유저가 앱 다운로드 여부를 고민할 때도 평가 및 리뷰를 통해 다른 유저들의 경험을 살펴보기 때문에 중요합니다. 같은 용도의 앱이 여러 개라면, 평점이 더 높고 리뷰가 많은 앱을 신뢰하게 될 것은 당연합니다. 그렇기 때문에 앱을 잘 사용하는 유저에게 적절한 시기에 평가 및 리뷰를 요청하여 가능한 많은 피드백을 받을 수 있도록 기획해야 합니다.
**5) 앱 상세 페이지 테스트하기**
이렇게 상세 페이지를 구성하는 데에 있어서, 가장 성과가 좋은 페이지로 만들기 위해서는 지속적인 테스트가 중요합니다. 각 앱 마켓에서 테스트 기능을 제공할 예정이고, 제공하고 있기 때문에 이 기능을 잘 활용하는 것이 좋습니다.
테스트는 앱 아이콘, 스크린샷 이미지, 설명 등에 대해 진행할 수 있는데요. 상세 페이지를 테스트할 때는 3가지의 다른 버전을 준비하여 테스트하는 것이 좋습니다. App Store의 경우, 최대 3개의 다른 버전까지 테스트할 수 있으며, Google Play의 경우 3개 이상의 대안을 준비하도록 권장하고 있기 때문입니다.
- **App Store의 경우**, [제품 페이지 최적화](https://developer.apple.com/kr/app-store/product-page-updates/)를 통해 다양한 앱 아이콘, 스크린샷, 설명 및 앱 미리보기를 활용해 보고 그 실적을 비교해 사용자들에게 더 반응이 좋은 페이지는 무엇인지 파악할 수 있습니다. [App Store Connect](https://developer.apple.com/kr/app-store/measuring-app-performance/)에서 앱 분석을 사용하여 노출 수, 전환율 및 개선 사항을 확인하고 테스트 기간(최대 90일) 동안 실적을 비교할 수 있습니다. 그런 다음 모든 고객에게 표시될 가장 실적이 우수한 버전을 만들 수 있습니다.
- **Google Play의 경우**, [스토어 등록정보 실험](https://developer.android.com/distribute/best-practices/grow/store-listing-experiments)을 통해 최적의 앱 아이콘, 이미지, 동영상, 앱 설명을 찾을 수 있습니다. 특정 사용자를 대상으로 스토어 등록정보 변경사항에 관해 A/B 테스트를 진행할 수 있으며, 데이터를 기반으로 스토어 등록정보를 변경하고 설치 수를 늘릴 수 있습니다. [Play Console](https://developer.android.com/distribute/console)에서 결과를 확인하고, 실험에서 가장 좋은 실적을 보인 항목을 기본 스토어 등록정보로 설정하면 됩니다.
### 3\. 양대 앱 마켓(App Store & Google Play)의 가이드라인에서 얻을 수 있는 인사이트 활용하기
유저들이 가장 많이 사용하는 양대 앱 마켓이라고 할 수 있는 App Store와 Google Play는 각각의 앱 상세 페이지에 대한 규격 및 기준을 가지고 있으며, 사용하는 유저의 특성도 다릅니다. 그렇기 때문에 기본적인 ASO 전략을 동일하게 가져가되, 각 앱 마켓에 따라 다른 특성 및 기준을 반영하여 ASO를 진행해야 합니다.
또한, 두 앱 마켓에서는 개발자용 페이지에서 지켜야 할 가이드라인 및 앱이 더 잘 노출될 수 있는 방법, 업데이트 사항 등을 소개하고 있기 때문에 앱 출시 전에는 물론, 앱 출시 후에도 주의 깊게 살펴봐야 합니다.
각 앱 마켓의 개발자용 페이지 및 참고할 수 있는 웹사이트를 정리하여 소개합니다.
#### **App Store**
[App Store의 개발자용 페이지](https://developer.apple.com/kr/)는 최신 업데이트 사항은 물론, App Store에 앱 등록 시 주의해야 할 사항 등 자세한 내용을 담고 있습니다. 또한, App Store에 앱을 제출하고 관리할 수 있는 [App Store Connect의 도움말](https://help.apple.com/app-store-connect/?lang=ko#/)도 참고해 보면 좋습니다. 특히 ASO와 관련된 부분은, [App Store 아이콘, 앱 미리보기 및 스크린샷 개요](https://help.apple.com/app-store-connect/?lang=ko#/dev910472ff2) 페이지입니다.
App Store에서 앱은 앱 이름, 앱 부제목, 키워드 및 회사 이름으로 검색할 수 있으며, 활용할 수 있는 크리에이티브로는 앱 아이콘, 최대 3개의 앱 미리보기와 최대 10개의 앱 스크린샷을 게재할 수 있다고 합니다.
특히, [개발자용 App Store의 제품 페이지 만들기](https://developer.apple.com/kr/app-store/product-page/) 페이지를 살펴보면, 앱 이름, 아이콘, 부제목, 앱 미리보기, 스크린샷, 설명, 홍보 문구, 키워드, 앱 내 구입, 새로운 기능, 평가 및 리뷰 등 제품 페이지를 구성하는 요소들에 대한 가이드라인과 더불어 제안하는 내용이 자세히 담겨 있습니다. App Store의 제품 페이지를 만들기 전 반드시 살펴보아야 합니다.
또한, [Apple의 발표에 따르면](https://developer.apple.com/kr/app-store/whats-new/), 올 하반기부터 iOS 15 또는 iPadOS 15 버전 이상의 디바이스에 대해 **제품 페이지 최적화** 및 **맞춤형 제품 페이지**를 통해 App Store에서 고객에게 훨씬 더 관련성 있고 효과적인 제품 페이지를 만들 수 있다고 하니, 활용해 보아도 좋겠습니다.
#### **Google Play**
Google Play에서도 [개발자용 Google Play 페이지](https://developer.android.com/distribute)를 제공하고 있습니다. 특히 ASO에 대해 참고하면 좋을 부분은, [앱 품질 및 검색 가능 여부 개선](https://developer.android.com/distribute/best-practices/launch/store-listing) 페이지입니다.
가이드에 따르면, Google Play가 추천하는 앱은 사람의 선별과 알고리즘이 함께 작동해 추천하는 것으로, 여기에서 고려되는 가장 큰 두 가지 요소는 '관련성'과 '품질'이라고 합니다.
특히, Google에서 앱의 '품질'을 평가하는 요소 중 하나는, **효과적인 스토어 등록정보 페이지**라고 합니다. 앱의 발견은 스토어 등록정보 페이지에서 시작되기 때문에, 효과적이고 정확한 스토어 등록정보 페이지를 통해 사용자와 신뢰를 쌓고 강한 첫인상을 남겨야 합니다. 등록정보 페이지에 제공된 앱 설명, 스크린샷 및 기타 메타데이터를 통해 앱에서 제공하는 기능과 콘텐츠를 유저가 명확하게 파악할 수 있어야 합니다. Google Play에서 앱이 더 잘 발견되기 위해 제안하는 몇 가지 팁은 다음과 같습니다.
- 앱의 등록정보 페이지에 동영상을 추가하고 16:9 비율의 스크린샷을 3개 이상 포함하면 전환율을 높일 수 있습니다.
- Google Play에서 앱을 올바르게 홍보할 수 있도록 정확한 [카테고리](https://support.google.com/googleplay/android-developer/answer/113475?hl=ko&ref_topic=3450987) 및 [콘텐츠 등급](https://support.google.com/googleplay/android-developer/answer/188189?hl=ko&ref_topic=3450987)을 제공하세요.
- [스토어 등록정보 실험](https://developer.android.com/distribute/best-practices/grow/store-listing-experiments)을 통해 실제 Google Play 사용자를 대상으로 다양한 아이콘, 설명, 스크린샷 등을 포함하여 등록정보 페이지의 여러 버전을 A/B 테스트하세요. 의미 있는 결과를 얻으려면 각 구성요소를 독립적으로 테스트하고 일주일 이상 테스트를 진행해야 합니다.
또한 [Play Console](https://developer.android.com/distribute/console)의 사용자 획득 탭을 통해 자연 획득 통계를 사용하면 검색어를 비롯하여 사용자가 Google Play에서 어떻게 내 앱을 발견하는지 확인할 수 있습니다.
## Conclusion
ASO는 SEO와 마찬가지로 한 번에 완벽하게 만들겠다고 정하기보다는, 지속적으로 관찰하고 개선하며 최적화해가는 것이 중요합니다. 수많은 앱들 속에서 우리 앱이 효과적으로 노출되고 다운로드 될 수 있는 최고의 방법을 찾기 위해, 앱 마켓의 콘텐츠 및 자산을 업데이트, 테스트, 측정 및 분석하는 지속적인 과정이라고 할 수 있죠.
본 글에서 소개한 ASO 인사이트를 바탕으로 앱 마켓에서 더 잘 발견되어 더 많은 유저를 유입시키고 성장할 수 있기를 바랍니다.
---
## 모바일 비즈니스에 앱이 반드시 필요한 4가지 이유
URL: https://www.airbridge.io/ko/blog/why-a-mobile-business-needs-an-app
Language: Korean
Published: 2022-03-17
## 모바일 시장의 발전과 앱 사용 시간의 증가
스마트폰 보급의 활성화로 인한 개인 기기의 확대는 모바일 비즈니스의 비약적인 발전으로 이끌었습니다. 유저들은 시간과 공간의 제약 없이 모바일 상에서 언제나 활동할 수 있죠. 특히 모바일 앱을 사용하는 데에 하루 중 꽤 많은 시간을 보내고 있습니다. 게다가 코로나19로 인해 모바일 앱 사용 시간이 더욱 증가했죠. [App Annie의 데이터](https://www.appannie.com/kr/insights/market-data/q2-2021-market-index-regional-rankings/)에 따르면 2021년 2분기 모바일 앱 소비 시간은 전년 대비 동일하거나 조금 감소하였습니다. 그래도 여전히 코로나19 대유행이 시작되기 전인 2019년 2분기 수치보다는 상당히 높습니다.
모바일에 대한 절대적인 수요와 앱 사용 시간의 증가는 모바일 비즈니스 기업들이 도달할 수 있는 유저의 수와 수집할 수 있는 데이터의 증가로 이어졌습니다. 모바일 시장 환경의 변화는 대부분 비즈니스가 웹사이트를 모바일에 최적화하고 직접 앱을 만들어야 할 필요성을 증가시키기도 했습니다. 이는 앱의 수가 지속적으로 증가하고 있는 이유와도 같습니다. 앱의 수 증가로 경쟁이 치열해졌고 고객의 시간을 사로잡기는 더 어려워졌습니다. 쉽지 않은 상황이지만 그럼에도 불구하고 모바일 비즈니스가 성공하기 위해서 앱을 만들어야 하는 이유는 무엇일까요?
## 모바일 비즈니스 전략 - 웹과 앱
앱을 만들어야 하는 이유를 이해하기 위해서, 우선 모바일 비즈니스의 전략부터 알아보겠습니다. 모바일 비즈니스는 모바일에서 자신의 제품을 선보이고 고객과 소통하기 위해 웹과 앱이라는 두 가지 플랫폼을 주로 활용합니다. 각각은 고객 여정에서 중요한 역할을 차지하기 때문에, 두 요소가 원활하게 협력하도록 전략을 구성하는 것이 중요합니다.
**모바일 웹**은 잠재 고객이 검색 등을 통해 처음으로 유입될 수 있는 플랫폼으로, 넓은 네트워크 기반 속에서 해당 브랜드와의 만남을 가능케합니다. ‘앱 다운로드’라는 허들을 거치지 않고도 바로 접속이 가능하기 때문에 다양하고 많은 고객들을 유입시킬 수 있는 기회가 되죠. 웹을 통해 접속한 고객이 브랜드와의 첫 경험이 만족스럽다면 충성 고객이 되는 시작점으로도 작용할 수 있습니다.
다만, 모바일 웹은 개인화된 맞춤 경험을 제공하는 데에 한계가 있으며, 각 브라우저에 따라 다른 유저 경험을 하게 될 가능성도 있습니다. 또한, 결제가 불안정하여 유저가 결제 과정 중간에 이탈할 수 있습니다. 이는 리텐션와 전환율에 좋지 않은 영향을 미치게 됩니다.
**모바일 앱**은 위와 같은 모바일 웹의 단점을 보완해 줍니다. 로그인이 지속될 가능성이 높기 때문에 앱 내에서 개인화된 경험 제공이 가능하며 브랜드와의 상호작용을 더 원활하게 만들어줍니다. 알림, 맞춤 이벤트 등으로 양질의 개인화도 가능합니다. 이는 충성 고객을 만들고 인게이지먼트 활성화에 유리해질 수 있죠. 다양한 모바일 웹 브라우저로 인한 유저 경험의 불확실성도 앱이 해결해 줄 겁니다.
모바일 웹만으로도 비즈니스를 운영할 수는 있겠지만, 이와 같은 앱의 단점 보완과 장점으로 인해 많은 기업들이 앱을 만들고 있다고 할 수 있습니다. 모바일 비즈니스에서 앱을 만들어야 하는 더 상세한 이유를 4가지로 정리해 보았습니다.
## 모바일 비즈니스에서 앱을 만들어야 하는 4가지 이유
### 1\. 모바일 앱을 통해 높은 전환율과 수익을 만들 수 있습니다.
모바일 앱이 필요한 가장 결정적인 이유 중 하나는 모바일 앱은 웹에 비해 높은 전환율을 보이기 때문입니다. 게다가 리텐션과 구매 전환율, LTV도 함께 향상시킬 수 있습니다. 고객에게 앱을 설치하도록 만드는 것은 어렵지만, 설치 이후 프로덕트의 역량에 따라 그 고객으로부터 창출해낼 수 있는 가치는 훨씬 더 큽니다. 앱을 통해 고객과 빠르고 편리한 상호작용이 가능한 것은 물론, 고객 데이터를 활용해 프로덕트를 개선하고 개인화된 유저 경험을 제공할 수 있기 때문이죠.
모바일 앱의 높은 전환율을 증명하는 리서치 결과를 알아보겠습니다.
우선, [크리테오의 리서치](https://www.criteo.com/blog/retail-travel-apps-higher-conversions-mobile/)에 따르면 특히 이커머스와 여행 업계에서 앱 내 전환율은 모바일 웹의 전환율에 비해 3~4배 정도 높습니다. 구체적으로는 이커머스 광고에서 모바일 웹의 전환율은 4%에 불과한 반면, 앱의 전환율은 18%에 달했습니다. 또한, 여행 광고에서는 모바일 웹의 전환율은 6%, 앱의 전환율은 20%였습니다. 아래 이미지는 글로벌 리테일 기업들의 모바일 웹과 앱에 따라 다른 퍼널 전환율을 비교한 자료입니다.
[__wf_reserved_decorative]
또한, Button의 [2019 모바일 커머스 리포트: Holiday Analysis](https://www.usebutton.com/2019-mobile-holiday-analysis)는 2018년 홀리데이 쇼핑 기간 동안 데이터를 분석하여 모바일 업계의 성장에 대한 인사이트를 아래와 같이 공유했습니다.
- 앱은 모바일 웹에 비해 14% 높은 전환율을 보였습니다.
- 앱에서 한 탭 당 수익이 모바일 웹에 비해 117% 높았습니다.
- 앱은 모바일 웹에 비해 구매자 당 108% 많은 주문을 했습니다.
### 2\. 우리 앱에서 고객과의 쉽고 빠른 상호작용이 가능합니다.
모바일 앱은 모바일 디바이스에 최적화되어 있기 때문에 빠르고 효율적이며 안정적이고 편리하게 접근할 수 있습니다. 홈 화면에서 아이콘 형태로 배치되기 때문에, 일단 설치만 된다면 URL과 브라우저를 거치지 않고도 바로 해당 프로덕트에 접속할 수 있어 접근성이 높아집니다. 앱에 대한 즉각적인 접근이 일반화되면서 다운로드 및 설치라는 앱이 가진 장벽도 점차 완화될 것입니다.
게다가 카메라, GPS, 생체 인식 인증(특히 터치 ID와 같은 지문 인식 및 안면 인식), 블루투스 등 디바이스의 자체 기능에 대한 접근성도 높아져서 유저와 원활한 상호작용을 하는 데에도 더 효과적입니다. 이 기능들은 모바일 웹보다 앱에서 안정적으로 구현할 수 있기 때문에 프로덕트 자체의 사용성과 편의성이 개선될 수 있습니다.
이와 같은 특징은 다음의 효과로도 이어질 수 있습니다.
- **일관된 고객 경험 제공으로 인한 브랜딩 강화**: 고객들이 프로덕트 내에서 브랜드로부터 빠르고 믿을 수 있는 경험을 제공받음으로써 신뢰도 및 충성도가 높아질 수 있습니다.
- **고객과 기업의 1:1 소통 창구로서의 앱**: 앱을 통해 고객과 브랜드가 지속적으로 소통할 수 있게 됩니다. 고객의 스마트폰 안에서 기업이 적극적으로 커뮤니케이션할 수 있는 ‘프라이빗한 장소'가 생기는 겁니다.
- **고객과 브랜드가 지속적으로 소통하는 습관 형성**: 앱을 통해 프로덕트와 서비스의 핵심 가치를 고객들이 항상 지니고 가까이하는 기기를 통해 언제 어디서나 전달할 수 있습니다. 자연스럽게 앱의 재방문율도 높아질 겁니다.
- **결제 최적화**: 모바일 앱은 상품을 장바구니에 담고 결제까지 이어지는 과정이 웹에 비해 안정적이고 간단합니다. 간편 결제 등을 활용해 결제에 필요한 수고도 덜어줄 수 있습니다.
이를 위해서는 고객을 위한 앱 경험을 제공하는 방법을 이해하는 것이 필수적이며, 지속적인 UI, UX 개선이 필요합니다.
### 3\. 고객 데이터를 바탕으로 프로덕트를 성장시킬 수 있습니다.
iOS 14.5 업데이트 및 구글의 쿠키 제거 등 개인정보보호 강화 트렌드 속에서 서드 파티 데이터(Third-party data)를 얻기 힘들어지고 있습니다. 그렇기 때문에 ‘로그인 기반’의 퍼스트 파티 데이터(First-party data)가 더욱 중요합니다. 앱은 로그인이 지속될 가능성이 높기 때문에, 유저는 편리함과 사용성이 좋아진다는 이점이 누릴 수 있고 브랜드에서는 데이터 기반으로 사용자의 행동을 분석하기에 용이해졌습니다. 게다가 쿠키 기반 데이터보다 더 정확한 사용자 데이터 수집이 가능합니다.
이렇게 수집한 데이터는 앰플리튜드(Amplitude)와 같은 애널리틱스 툴을 활용하여 유저들이 이탈하는 지점 발견, 리텐션을 높이는 특정 행동 확인, 코호트 기능을 활용한 특정 집단에 대한 분석 등을 진행하는 것이 효과적입니다. 이 데이터 분석을 기반으로 프로덕트의 UI 및 UX를 개선하거나, 핵심 기능 추가 등을 통해 전환율 및 리텐션을 높일 수 있습니다. 제품 주도 성장(Product-led Growth)의 핵심도 바로 유저들의 실제 행동 ‘데이터’입니다.
[__wf_reserved_decorative]
이와 더불어, 에어브릿지(Airbridge)와 같은 광고 성과 측정 툴을 활용하면, 광고를 통해 앱을 설치한 고객 데이터를 기반으로 리인게이지먼트 광고를 집행할 수도 있습니다.
### 4\. 양질의 개인화된 고객 경험 제공이 가능합니다.
앞서 언급한 고객 데이터는, 고객에게 개인화된 고객 경험을 제공하는 데에도 중요하게 활용됩니다. 각 프로덕트에서 수집한 고객의 정보 및 데이터를 통해 보다 세련되고 맞춤화된 유저 경험을 제공할 수 있습니다. 또한, 개인화된 형태의 마케팅 캠페인을 진행할 수도 있죠.
[Think with Google에서 제공한 리서치 결과](https://www.thinkwithgoogle.com/marketing-strategies/app-and-mobile/smartphone-user-personalization-preferences/)에 따르면, 58%의 스마트폰 유저들은 자신 혹은 자신의 행동을 기억하는 모바일 웹 혹은 앱에 대해 더 호감을 느낀다고 할 만큼 고객 행동을 바탕으로 한 개인화된 경험 제공은 중요합니다.
[__wf_reserved_decorative]
이때 특히 푸시 알림 및 인앱 메시지 등을 활용하면 적재적소에 고객에게 원하는 메시지를 전할 수 있습니다. 게다가 앱 자체적으로 보내는 알림은 무료로도 가능하죠. 더 고도로 개인화된 앱 내 캠페인을 진행하고 싶다면, 브레이즈(Braze)와 같은 마케팅 자동화 툴을 사용하는 것도 좋습니다. [브레이즈의 리서치에 따르면](https://www.ab180.co/library/braze-in-app-message-guide?utm_source=ab180-blog&utm_medium=cta&utm_campaign=braze-in-app-message-guide), 앱에서 인앱 메시지를 사용하여 고객과 소통한 경우 고객 인게이지먼트의 131%가 증가했다는 리서치 결과도 있었습니다. 이런 방법을 통해 프로모션을 집행하고 나면 유저의 즉각적인 반응이 나타나고 눈에 보이는 수익으로도 이어질 겁니다.
## 마치며
모바일 비즈니스 성공을 위해 ‘앱’을 만드는 것의 중요성에 대해 알아보았습니다.
정리하자면, 모바일 비즈니스에서는 모바일 웹과 앱을 모두 활용하는 것이 필요합니다. 앱 개발에 드는 리소스와 치열한 경쟁이라는 어려움도 있지만, 그럼에도 불구하고, 앱은 리텐션과 전환율 개선 및 수익 창출에 효과적이기 때문입니다. 이와 더불어, 앱을 기반으로 고객과의 지속적이고 개인화된 커뮤니케이션이 가능한 것은 물론, 앱의 퍼스트 파티 고객 데이터를 바탕으로 한 프로덕트의 성장 및 양질의 개인화를 통해 모바일 비즈니스에서 우위를 선점할 수 있습니다.
모바일 비즈니스의 성공을 위해, 앱은 중요한 수단 중 하나입니다.
---
## 모바일 웹투앱 어트리뷰션 제대로 알아보기
URL: https://www.airbridge.io/ko/blog/mobile-webtoapp-attribution
Language: Korean
Published: 2021-12-23
이번 포스트의 주제는 ‘모바일 웹투앱 어트리뷰션(Web-to-App)’입니다. 모바일 앱 어트리뷰션 및 웹투앱 어트리뷰션의 기초적인 개념 이해에서 시작해, 웹투앱 마케팅과 어트리뷰션 전략이 필요한 이유를 알아보겠습니다. 모바일 웹투앱 어트리뷰션이 무엇이고, 어떻게 시작해야 하는지 궁금하셨다면 이번 포스트를 끝까지 읽어주세요!
## 모바일 웹투앱 어트리뷰션 이해하기
### 모바일 앱 어트리뷰션과 웹 어트리뷰션
디지털 마케팅에서 ‘**어트리뷰션(Attribution)**’이란, 다양한 터치포인트로 이루어진 고객 여정에서 광고주가 의도한 행동(대표적으로 '구매')을 성공적으로 유도한 마케팅 채널/캠페인에 해당 성과를 기여시키는 작업을 의미합니다. 어트리뷰션을 통해 마케터는 신규 고객이 프로덕트를 찾아오는 과정에서 유효했던 터치포인트, 그리고 실질적인 전환 성과를 발생시킨 터치포인트를 파악할 수 있습니다. 어트리뷰션 분석은 해당 프로덕트 및 타겟 오디언스에게 적합한 미디어 믹스와 크리에이티브 전략을 수립하는 데 도움을 줍니다.
그 중에서도 **모바일 앱 어트리뷰션**은 앱 설치 및 특정 인앱 이벤트에 대해 어떤 마케팅 캠페인이 실질적으로 기여했는지 분석하는 것이라고 볼 수 있습니다. 다수의 앱 광고주들은 대표적으로 마케팅 퍼널 위쪽에 자리하는 ‘앱 설치’를 위한 UA(User Acquisition) 캠페인을 진행하게 되는데요. 이 때 앱 설치 광고를 클릭하여 앱을 설치한 경우, 어떤 광고를 통해 해당 유저가 유입되었는지 분석하는 것이죠. 에어브릿지(Airbridge)와 같은 어트리뷰션 툴을 사용하면 이와 같은 UA 캠페인은 물론, 딥링크를 통한 리인게이지먼트 캠페인도 트래킹할 수 있습니다. 한 가지 유의할 점은, 모바일 앱 어트리뷰션에서 ‘앱 설치’ 이벤트는 앱이 기기에 ‘설치'된 시점이 아니라, 앱이 설치된 후 유저가 앱을 ‘처음으로 실행'하는 시점으로 정의된다는 점입니다.
**웹 어트리뷰션**에서는, 웹상에서 일어나는 전환 성과에 기여한 마케팅 캠페인이 무엇인지 파악합니다. 모바일 웹 및 데스크탑 웹을 넘나들며 활동하는 유저들이 광고주가 집행 중인 광고에 인게이지할 경우, 웹 브라우저의 쿠키 ID를 기반으로 각 유저를 식별하여 전환된 유저에게 어떤 터치포인트가 유효한 영향을 미쳤는지 분석하게 됩니다. 에어브릿지는 웹사이트에서 발생한 모든 전환을 트래킹할 수 있도록 웹 SDK를 제공하고 있습니다.
### 웹투앱 어트리뷰션이란?
[__wf_reserved_decorative]
그렇다면 이 포스트의 핵심 주제인 **웹투앱 어트리뷰션**이란 무엇일까요? 표현 그대로, 웹에서 앱으로 이동한 유저들의 전환 성과에 대한 마케팅 캠페인 기여도를 분석하는 것입니다. 즉, 웹 캠페인을 통해 유입된 유저가 앱을 설치하거나 앱 내에서 전환 이벤트를 발생시킬 경우, 해당 앱 성과들을 유저 유입에 기여한 웹 캠페인과 매칭하게 됩니다. 또한 웹상에서 딥링크를 통해 유저가 앱으로 즉시 이동하는 경우에도 웹투앱 어트리뷰션 기술이 필요합니다. 에어브릿지 딥링크를 사용하면 유저를 원하는 앱 내 특정 페이지로 부드럽게 연결할 수 있으며, 이와 같은 딥링크 전환에 대한 웹투앱 어트리뷰션 데이터도 확인하실 수 있습니다.
👉 [에어브릿지를 활용한 웹&앱 통합 분석 더 자세히 알아보기](https://www.airbridge.io/ko/product/web-app-attribution)
## 앱 마케팅에서 웹투앱 전략이 필요한 이유
### 앱 마켓이라는 허들
가장 먼저, 모바일 앱이 웹에 비해 유저를 획득하기에 불리한 이유 중 하나인 ‘앱 마켓'의 존재 때문입니다. 웹 브라우저에서 URL 주소만으로 바로 접근 가능한 웹사이트와 달리, 앱은 구글 플레이 스토어나 애플 앱 스토어 등, 앱 마켓이라는 중간 단계를 거쳐야 합니다. 앱 설치 광고를 클릭한 유저는 앱 마켓으로 랜딩되어 앱 마켓 상세페이지만 살펴보고 앱 설치 여부를 결정하게 되며, 따라서 [앱 스토어 최적화(ASO)](https://blog.ab180.co/posts/aso-success-guide)를 통해 설치 전환율을 높이기 위한 노력은 필수입니다. 그러나 앱 마켓 상세페이지를 통해 유저에게 제공할 수 있는 경험이 제한적이기 때문에, ASO에만 의존할 수는 없습니다. 앱 마켓 앞단에서 웹 플랫폼을 활용해 풍부한 콘텐츠와 경험을 전달한다면, 유저들이 앱을 설치하게끔 만드는 유인을 더 효과적으로 제공할 수 있습니다. 상대적으로 유저 획득이 용이한 웹 플랫폼을 활용해 유저를 보다 효과적으로 앱으로 전환시키는 웹투앱 마케팅 전략이 필요한 이유입니다.
### 효과적인 진성 유저 확보
앞에서 설명한 대로 앱은 웹에 비해 유저 획득이 제한적이기 때문에, 사실 이미 많은 앱 마케터들이 웹상에서 캠페인을 집행하고 있습니다. 웹이 유의미한 유저 획득 플랫폼이라는 것에 다수의 광고주들이 공감하고 있죠. 어느 정도 니즈를 가진 유저들의 유입 창구인 키워드 검색 광고의 경우에는, 반드시 웹으로만 랜딩시킬 수 있다는 점도 기억하세요.
궁극적으로는 앱으로 유저를 유입시켜야 하는 이유는, 앱의 수익성이 더 높기 때문입니다. [Criteo의 조사](https://www.criteo.com/blog/global-commerce-review-map/)에 따르면 모바일 앱에서의 구매 전환율이 모바일 웹에 비해 3배 더 높습니다. 즉, 앱은 유저를 보다 강력하게 락인(lock-in)시킬 수 있고, 앱에서 꾸준히 활동하는 유저는 진성 유저가 될 확률도 높습니다.
특히 웹과 앱 모두를 보유한 광고주의 경우 웹에서의 행동 이벤트를 적극적으로 촉진하여 전환 성과를 거두고, 웹상에서 프로덕트에 만족을 느낀 유저들이 자연스레 스스로 앱을 설치하게 되는 선순환 구조를 만들기도 하는데요(에어브릿지 고객 디코드 성공 사례 확인하기). 이러한 유저들의 경우 이미 웹을 통해 제품의 가치를 발견한 진성 유저들이기 때문에 리텐션 뿐 아니라 LTV도 높습니다.
## 모바일 웹투앱 어트리뷰션의 필요성
### 파편화된 고객 여정
모바일 앱 마케터들에게 미디어 및 캠페인 최적화는 필수 과제입니다. 이를 위해 유저들이 어떤 플랫폼의 무슨 채널, 구체적으로 어떤 캠페인의 어떤 소재를 통해 우리 프로덕트를 찾아오는지 파악하고 각각의 전환에 얼만큼의 비용을 지출하고 있는지 정확하게 트래킹할 필요가 있습니다.
사실상 이미 유저들은 웹과 앱을 넘나들며 활동하고 있으며, 웹 캠페인을 통해 웹으로 랜딩되었더라도 이후 앱으로 이동하여 구매 등 실질적인 전환 성과를 발생시키는 경우가 많습니다. 즉, 오늘날 고객 여정은 매우 파편화되어 있기 때문에, 웹과 앱을 서로 연결하여 어트리뷰션 분석을 진행하지 않는다면 복잡한 고객 여정 간의 연결고리는 모두 사라지게 되고, 정확한 트래킹은 불가능해지는 셈입니다.
## 모바일 웹투앱 어트리뷰션, 어떻게 시작해야 할까?
### MMP 솔루션 활용하기
마케터들은 여러 매체를 통해 웹과 앱에 걸쳐 다수의 캠페인을 동시에 진행합니다. 이 때 여러 플랫폼 및 매체를 통해 발생하는 광고 성과를 통합적으로 측정하기 위해서는 어트리뷰션 툴, 즉 **MMP(Mobile Measurement Partner) 솔루션**의 활용이 필수적입니다. 각 매체별 데이터를 연동하여 MMP 대시보드를 통해 산재되어 있는 광고 데이터를 모아서 살펴보고, 전환에 대한 매체별 기여도를 중복 없이 분석할 수 있기 때문입니다. 즉, 정확한 광고 성과 분석에 있어 MMP 솔루션이 핵심적인 역할을 하는 것이죠.
### 모바일 웹투앱 어트리뷰션 분석 환경 셋팅하기
MMP 솔루션을 활용하여 어떻게 모바일 웹투앱 어트리뷰션 분석을 진행할 수 있을까요? 에어브릿지의 경우를 예로 살펴보겠습니다.
- 먼저, 앱에는 앱 SDK, 웹사이트에는 웹 SDK를 설치합니다. (에어브릿지는 웹 SDK와 앱 SDK를 모두 지원하며, 이를 기반으로 웹과 앱의 데이터를 통합적으로 분석합니다.)
- 웹사이트에 유저의 웹투앱 이동을 유도하는 배너 또는 버튼을 셋팅합니다(예: ‘앱에서 살펴보기' 버튼). 이 때 setDeeplinks 메소드를 활용하면 코드 몇 줄만으로 간편하게 셋팅이 가능합니다.
- 웹에서 앱으로의 이동 장치 셋팅이 완료되면 에어브릿지 트래킹 링크를 생성하여 웹으로 랜딩시키는 캠페인을 진행합니다. 웹사이트에 설치한 SDK가 웹 URL의 캠페인 파라미터를 통해 웹에서의 행동을 트래킹합니다. (혹은, UTM 파라미터가 포함된 랜딩 페이지 URL을 활용하고 싶은 경우, 에어브릿지에서 캠페인 파라미터 자동 분석 기능(utmParsing)을 설정하면 URL 파라미터의 UTM 값을 에어브릿지 캠페인 파라미터 값과 자동 매칭해 줍니다.)
위 과정을 모두 완료하면, 에어브릿지가 웹사이트를 통해 전환된 앱 설치에 대해서도 자동으로 트래킹을 시작합니다. 필요하다면 웹과 앱 데이터를 별도로 추출하여 자체적인 데이터 분석도 진행할 수 있습니다.
\*에어브릿지 웹투앱 어트리뷰션 셋팅 방법에 대해 더 자세히 알고 싶다면 [개발자 문서](https://developers.airbridge.io/docs/web-sdk#%EC%9B%B9%ED%88%AC%EC%95%B1-%EC%84%A4%EC%A0%95)를 참고해주세요.
## 에어브릿지에서 시작해보세요!
[__wf_reserved_decorative]
에어브릿지는 데이터 수집 및 측정에서부터 캠페인 분석, 어트리뷰션을 아우르는 통합 마케팅을 다방면으로 지원합니다. 특히 에어브릿지의 웹 및 웹투앱 어트리뷰션 기능은 많은 광고주들이 이미 적극적으로 활용하고 있습니다. 또한, 에어브릿지는 마케터들의 업무 효율을 실질적으로 향상시켜 줄 다양한 [편의 기능](https://www.airbridge.io/ko/blog/airbridge-5-features-to-watch)도 제공합니다.
다른 회사는 어떻게 에어브릿지를 활용하고 있는지 궁금하시다면, 아래 고객사 인터뷰 포스트를 확인해보세요.
🎤 [디코드 성공 사례 보러가기](https://www.airbridge.io/ko/case-studies/dcode-web-to-app-attribution)
🎤 [헬로우봇 성공 사례 보러가기](https://www.airbridge.io/ko/case-studies/hellobot-multi-channel-tracking)
🎤 다노 성공 사례 보러가기
---
## 앱 마케팅 유입 코호트별 매출액을 에어브릿지 Revenue 리포트로 분석해 보세요
URL: https://www.airbridge.io/ko/blog/airbridge-revenue-report-update
Language: Korean
Published: 2021-11-30
에어브릿지(Airbridge)에서 분석 기간 내에 앱 설치 또는 딥링크 오픈 이벤트를 통해 앱으로 유입된 유저가 발생시킨 Revenue 성과를 다양한 메트릭을 통해 일자별로 확인할 수 있는 코호트 분석 리포트, **Revenue 리포트**가 업데이트되었습니다. 특히 이번 업데이트에는 “[_Airbridge가 직접 개발한 OLAP DB, Luft를 소개합니다._](https://blog.ab180.co/posts/introducing-luft)” 포스트에서 다룬 Luft 데이터베이스를 기반으로 한 코호트 분석 기능이 도입되었습니다.
## 앱에서 유저가 발생시킨 Revenue를 확인해야 하는 이유
앱 설치 캠페인 및 리인게이지먼트 캠페인 등 앱 마케팅을 통해 앱으로 유입된 유저들의 앱 내 전환율은 트래킹해야 할 핵심 지표 중 하나입니다. 특히 이커머스 혹은 게임 서비스 등 인앱 결제가 중요한 프로덕트의 경우, 단순히 ‘구매 여부’뿐만 아니라 구매 및 결제로 인해 발생한 ‘수익’도 중요하게 확인해야 할 부분입니다. 유저들이 프로덕트와 얼마나 잘 인게이지하고 있는지를 확인할 수 있을뿐더러, 특정 캠페인이 수익에 미치는 영향도 파악할 수 있기 때문입니다.
> 많은 무료 모바일 게임의 경우 광고를 통해 수익을 창출하며, 광고 없이 게임을 하는 경험을 저렴한 가격에 판매하기도 합니다. (중략) 광고를 노출하지 않는 게임은 소수의 유저들이 매일 만들어 내는 수 천 번의 소액결제로 이윤을 창출하는 경우가 많습니다. 모바일 게임 개발자들은 가장 큰 손인 ‘고래(Whale)’ 유저들의 행동을 트래킹하고 분석합니다. 이 파워 유저의 경험을 최적화함과 동시에, 무료 유저의 경험과도 균형을 맞추기 위해 노력합니다. (출처: [Amplitude 인게이지먼트 마스터하기 플레이북](https://www.ab180.co/library/amplitude-engagement-playbook) 48쪽)
## 에어브릿지의 Revenue 리포트란?
에어브릿지에서 제공하는 Revenue 리포트에서는 ‘분석 기간 내에 Start Event를 수행한 후 Revenue Event를 수행한 유저 코호트’가 ‘N 번째 날’에 발생시킨 ‘매출액(Revenue)’, '이벤트 수(Event Count)‘, ‘ARPU(Average Revenue Per User)’, ROAS(Return on Ad Spend)를 확인할 수 있습니다.
[__wf_reserved_decorative]
👉[에어브릿지 Revenue 리포트 이용가이드 바로가기](https://help.airbridge.io/hc/ko/articles/4409764368409-Revenue)
## Revenue 리포트 설정 방법
[__wf_reserved_decorative]
Revenue 리포트의 상세한 설정 방법 및 용어들의 정의는 다음과 같습니다.
### 이벤트
Start Event 및 Return Event를 선택하여 분석 대상이 되는 유저 코호트 및 Revenue 이벤트를 설정할 수 있습니다.Revenue
**Start Event:** 설정한 분석 기간 내에 Start Event를 수행한 유저들을 대상으로 Revenue Event 수행 여부를 측정 및 분석합니다. _Target Events (App)_로만 선택 가능하며, 여기서 _Target Events_는 _Install (App)_ 혹은 _Deeplink Open (App)_ 이벤트를 수행한 경우를 모두 포함합니다.
**Revenue Event:** 설정한 분석 기간 내에 Start Event를 수행한 유저 중 Revenue Event까지 수행한 유저를 코호트로 정의합니다. Revenue Event에 대한 Revenue 값은 _구매 완료 (App)_, _Ad Impression (App)_ 모두 Event Value(앱에서 발생한 이벤트 값의 합)를 사용하여 계산합니다. 선택 가능한 이벤트로는 _구매 완료 (App)_, _첫 구매 완료 (App)_, _Ad Impression (App)_, _Ad Impression (App)_ or _구매 완료 (App)_가 있습니다.
### 메트릭
Revenue 리포트에서 선택 가능한 메트릭은 매출액(App), 이벤트 수(App), ROAS(App), 혹은 ARPU(App)입니다. 여기서 매출액(App)은 구매 이벤트에서 발생한 Revenue 이벤트 값이며, 이벤트 수는 Revenue 이벤트가 발생한 횟수, ARPU(App)는 유저당 평균 Revenue를 의미합니다. 여기서 ARPU를 계산할 때 User는 Start Event를 수행한 유저를 의미합니다. 그리고 ROAS(App)은 사용한 비용에 대한 매출액 비중으로 Revenue / Cost \* 100(%)로 계산합니다.
Revenue 리포트에서는 메트릭을 다음의 두 가지 방법 중 하나를 선택하여 계산할 수 있습니다.
**1) On day**: 코호트 유저가 해당 날짜에 발생시킨 값으로, 매출액(App) 메트릭을 지원합니다.
예를 들어, 메트릭으로 매출액(Revenue), 계산 방식으로 On day를 설정한 경우, Day 3의 값은 유입 후 3일째 날에 발생시킨 Revenue 만으로 계산된 값을 의미합니다.
_Day 3 = Day 3 Revenue_
**2) Cumulative**: 코호트 유저가 유입된 날로부터 해당 날짜까지 발생시킨 누적값으로, 매출액(App) 및 ARPU(App) 메트릭을 지원합니다.
예를 들어, 메트릭으로 매출액(Revenue), 계산 방식으로 Cumulative를 설정한 경우, Day 3의 값은 유입된 날짜로부터 3일째 날까지 발생시킨 누적 Revenue로 계산된 값을 의미합니다.
_Day 3 = Day 0 Revenue + Day 1 Revenue + Day 2 Revenue + Day 3 Revenue_
계산 시, Users는 Airbridge Device ID를 기준으로, 설정한 분석 기간동안 Start Event를 수행한 유니크 유저수입니다. 그리고 Total Revenue는 설정한 분석 기간동안 Start Event를 수행한 유저들로부터 발생한 Revenue Event의 Total Revenue입니다. 그리고 Total Cost는 설정한 분석 기간 동안 발생한 Start Event에 기여된 캠페인에 사용된 전체 Cost를 의미합니다.
### 그룹바이 및 필터
그룹바이는 최대 4개 항목까지 선택할 수 있으며, 그룹바이 설정을 통해서 유입 채널 또는 OS 등 특정 데이터를 기준으로 세분화하여 Revenue를 확인할 수 있습니다. 또한, 필터 설정을 통해서 특정 채널이나 특정 OS에서 발생한 이벤트만을 대상으로 Revenue를 확인하는 등의 설정도 할 수 있습니다. 설정된 그룹바이 및 필터는 테이블 상단에 위치하며 해당 박스에서 손쉽게 특정 필드를 추가하거나 삭제할 수 있습니다. 이때, 그룹바이와 필터 조건은 분석 기준이 되는 Start Event에만 적용됩니다.
👉 이외에도 에어브릿지의 대시보드를 더 잘 활용할 수 있는 방법은, [Airbridge, 어디까지 써봤니? - 에어브릿지 편의 기능 5가지](https://blog.ab180.co/posts/airbridge-5-features-to-watch) 포스트에서 확인해 보세요!
더 자세한 에어브릿지 Revenue 리포트 이용 방법은 아래 링크와 연결된 이용 가이드 문서에서 확인할 수 있습니다. 에어브릿지 Revenue 리포트를 통해 유저들이 매출 성과에 얼마나 기여하고 있는지 파악해 보세요!
[__wf_reserved_decorative]
---
## 에어브릿지, 업계 최초 MMP & MTA 페이스북 비즈니스 파트너 선정
URL: https://www.airbridge.io/ko/blog/airbridge-becomes-facebook-business-partner-for-measurement
Language: Korean
Published: 2021-10-07
빅데이터광고 성과 분석 전문 주식회사 에이비일팔공은 자사의 제품인 에어브릿지가 페이스북 모바일 측정(MMP)과 멀티터치 기여도분석(MTA) 분야의 비즈니스 파트너로 동시에 선정되었으며, 광고캠페인의 실제 성과를 측정할 수 있는 인크리멘탈리티 분석 솔루션(Incrementality)을 새롭게 출시한다고 밝혔다. 페이스북 MMP와 MTA 분야 파트너에 동시 선정된 것은 에어브릿지가 세계 최초 사례이다.
에어브릿지의 CEO이자 공동 창업자인 남성필 대표는 “인크리멘탈리티 솔루션을 통해, 마케터들에게 글로벌 최고 수준의 광고 성과 분석 서비스를 제공할 수 있게 되어 기쁘다"라고 밝히며, “에어브릿지는 공격적인 R&D와 지속적인 제품 개선을 통해 데이터 기반의 광고 성과 측정 전반에 필요한 모든 기능들을 빠르고 쉽게 제공하는 통합 솔루션으로 거듭나겠다"고 밝혔다.
에어브릿지의 신기능인 인크리멘탈리티 분석 솔루션은 기여도분석 시장의 패러다임이 전환되고 있음을 보여주는 대표적인 사례로, 빅데이터 분석으로 통해 웹과 모바일 앱의 크로스 플랫폼 상황 속, 광고 캠페인을 통한 실제 마케팅의 성과를 알려준다.
에어브릿지의 인크리멘탈리티 분석 솔루션(Incrementality)은 머신러닝과 빅데이터 통계 분석 방법에 기반하고 있으며, 분석을 위한 최소 데이터가 확보되는 즉시 모델링 및 분석 결과 제공이 가능하다. 또한, ‘앱 설치', ‘주문 완료' 등 핵심전환 이벤트에 대해 매일 갱신되는 분석 리포트를 통해, 별도의 데이터 분석팀 도움 없이도, 광고 캠페인의 실제 성과를 분석하고 마케팅 전략 및 미디어믹스를 최적화할 수 있다.
한국 최대 홈쇼핑&리테일 기업, GS 리테일의 박우현 마케팅부문장은 “에어브릿지의 인크리멘탈리티 분석 솔루션을 통해 라스트 클릭 분석 모델 하에서 과대평가 또는 과소평가된 광고 매체 확인을 위한 데이터를 빠르게 확보할 수 있었으며, 다양한 지표에 대한 인크리멘탈리티 분석을 진행하여, 광고 매체별 효용을 극대화 할 수 있었다”고 증언했다.
● 에어브릿지 더 알아보기: [https://www.airbridge.io/ko](https://www.airbridge.io/ko)
● 에어브릿지 데모 요청하기: [https://www.airbridge.io/ko/book-a-demo](https://www.airbridge.io/ko/book-a-demo)
## **에어브릿지란?**
에어브릿지(Airbridge)는 에이비일팔공의 광고 성과 분석 솔루션으로, 디지털 마케터가 마케팅 캠페인의 실제 광고 성과를 빠르고 정확하게 분석할 수 있도록 돕습니다. GS 리테일, SC 제일은행, 버거킹 코리아와 같은 국내외 엔터프라이즈 기업들에서 ‘웹&앱 통합 분석부터 인크리멘탈리티 분석까지’ 광고 성과 분석 전 영역을 빈틈없이 지원하는 에어브릿지 솔루션을 활용하고 있습니다.
---
## 이제 에어브릿지에서 구글 계정 로그인을 통한 구글 애즈 비용 연동이 가능합니다!
URL: https://www.airbridge.io/ko/blog/airbridge-x-google-ads-costs
Language: Korean
Published: 2021-09-29
이제 에어브릿지(Airbridge)에서 **구글 계정 로그인 한 번으로** **구글 애즈 비용 연동**이 편리하게 가능합니다!
구글 애즈 광고 캠페인을 운영 중인 구글 계정에 로그인하여 비용을 연동하면 전환 성과와 비용을 에어브릿지 대시보드에서 한 번에 조회할 수 있습니다. 비용 연동 후 앱 프로모션 **캠페인, 광고 그룹, 광고 레벨의 노출수(Impressions(Channel)), 클릭수(Clicks(Channel)), 지출 금액(Cost(Channel)) 데이터를** 아래와 같이 Actuals 리포트에서 확인할 수 있습니다.
[__wf_reserved_decorative]
### 구글 애즈 비용 연동 방법
아래 이미지는 에어브릿지 대시보드의 **Integrations - 구글 애즈 - 비용 연동** 메뉴 화면입니다.
- 우선, \[구글 애즈 비용 연동\] 토글을 켭니다.
- 하단의 \[구글 계정 로그인\]을 누르고 구글 애즈 캠페인을 운영하고 있는 구글 계정으로 로그인합니다.
- 구글 로그인을 성공적으로 진행한 후 에어브릿지 대시보드에서 연동할 구글 광고 계정을 선택하면 연동이 완료됩니다.
- 연동된 광고 계정이 하단의 이미지와 같이 리스트에 추가되어 연동 정보를 확인할 수 있습니다.
[__wf_reserved_decorative]
연동일 포함 7일 치 비용 데이터가 연동이 되며, 데이터 이동 과정에서 약간의 시간이 소요되기 때문에 구글 애즈 대시보드와 차이가 발생할 수 있습니다. 또한, 비용 데이터의 특성상 구글의 데이터가 변할 수 있으며, 에어브릿지는 가장 최근에 가져온 데이터를 사용합니다.
아래 가이드 문서 링크를 통해 더 자세한 에어브릿지와 구글 애즈의 앱 연동 및 비용 연동 방법을 확인해 보세요!
[__wf_reserved_decorative]
[]
## 구글 애즈(Google Ads)란?
[구글 애즈(Google Ads)](https://ads.google.com/home/?subid=ww-ww-et-g-aw-a-vasquette_ads_1!o2&authuser=0)는 구글이 운영하고 있는 실시간 입찰(RTB) 방식의 광고 플랫폼으로, 디지털 광고 캠페인을 진행할 때 빼놓을 수 없는 플랫폼입니다. 구글 애즈를 통해 웹사이트는 물론 앱 마케팅 광고를 할 수 있으며, 검색 결과, 유튜브, 디스플레이 및 구글과 파트너십을 맺고 있는 다양한 지면에 광고를 게재할 수 있습니다. 잠재 고객이 유튜브 동영상을 보거나, 구글에서 검색을 할 때 웹사이트 및 앱을 효과적으로 알릴 수 있죠.
## 에어브릿지와 구글 애즈의 연동
에어브릿지와 구글 애즈를 연동하게 되면, 구글 애즈의 '앱 프로모션' 광고 성과를 에어브릿지 대시보드에서 한눈에 확인할 수 있습니다. 연동 이후 구글 애즈로 발송된 이벤트를 전환으로 가져와서 캠페인 최적화를 위해 활용할 수도 있습니다.
구글 애즈 등 여러 채널들에서 진행한 앱 마케팅 캠페인의 성과를 분석하고 최적화하기 위해서는, 에어브릿지와 같은 어트리뷰션 툴과 광고 매체들을 연동하여 한눈에 성과를 파악하는 것이 좋습니다. 각 채널들의 캠페인 및 광고에 대한 노출, 클릭, 비용 모두를 어트리뷰션 툴과 연동하여 성과를 살펴보면, ROAS(광고비 대비 수익률), CPC(클릭당 비용), CPA(전환당 비용), CPM(1,000회 노출당비용) 등 주요 지표들을 개선하는 데에 도움이 될 수 있습니다.
이제 에어브릿지 대시보드에서 구글 애즈의 앱 캠페인 성과를 더 편리하게 한눈에 확인하고, 앱 마케팅 성과를 개선해 보세요!
---
## Airbridge, TikTok 공식 모바일 측정 파트너(MMP) 선정
URL: https://www.airbridge.io/ko/blog/airbridge-x-tiktok
Language: Korean
Published: 2021-06-24
Airbridge가 글로벌 숏폼 모바일 비디오 플랫폼 틱톡(TikTok)의 Marketing Partner Program 공식 마케팅 파트너로 선정되어 이 좋은 소식을 여러분께 알립니다. Airbridge는 이제 공식 모바일 측정 파트너(Mobile Measurement Partner)로서 TikTok과 함께하게 됩니다.
이 파트너십을 통해 마케터들이 TikTok에서 광고 캠페인을 생성하고, 타겟팅하고, 측정하고, 최적화하여 궁극적으로 고객 여정을 개선하는 것을 도울 예정입니다.
이제 TikTok Ads 집행시 Airbridge 를 연동하여 아래와 같은 기능을 통해 TikTok Ads 의 광고 캠페인을 정확하게 측정하고 최적화할 수 있습니다.
- 인앱 이벤트 포스트백: Airbridge Dashboard 에서 간단하고 쉬운 설정을 통해 Airbridge 에서 수집된 인앱이벤트 및 각종 이벤트 속성값들을 틱톡 플랫폼으로 전달할 수 있습니다. Airbridge 대시보드에서 포스트백 설정을 올포스트백(All Postback)으로 설정하여 운영 시 더 많은 인앱이벤트를 오디언스 생성 및 타겟팅에 효과적으로 활용할 수 있습니다.
- 인앱이벤트 최적화 기능 (App Event Optimization): Airbridge 가 포스트백으로 전송한 데이터를 기반으로 틱톡 광고 플랫폼에서 AEO 기능을 사용할 수 있습니다.(참고 : [https://ads.tiktok.com/help/article?aid=14175](https://ads.tiktok.com/help/article?aid=14175))
- Pangle 성과 추적 가능: 팽글은 글로벌 앱 수익화 및 사용자 확보를 달성하도록 도와주는 광고 플랫폼으로 Airbridge 대시보드에서 틱톡 플랫폼과 함께 성과 추적이 가능합니다.
Airbridge는 TikTok 전용 트래킹 링크를 통해 앱을 설치하고 인앱 이벤트를 발생시킨 유저들의 행동을 정확히 측정합니다. 이로써 TikTok을 통한 광고의 비즈니스 성과를 파악하는 것은 물론 미디어 믹스 전략에 반영하여 ROI를 극대화할 수 있습니다.
세계적인 모바일 광고 플랫폼인 TikTok For Business의 파트너십 프로그램에 참여하여 TikTok 과 Airbridge 의 고객들에게 효과적인 모바일 측정을 지원할 수 있게 되어 기쁩니다. 앞으로도 이와 같은 Integration 을 강화시킬 수 있도록 TikTok 팀과 지속적인 커뮤니케이션 및 공동 작업을 진행할 계획입니다.
아래 가이드 문서 링크를 통해 Airbridge와 TikTok For Business 플랫폼의 구체적인 연동 방식을 확인하세요!
[]
#### TikTok for Business 소개
[TikTok for Business](https://www.tiktok.com/business/ko)는 브랜드 및 마케터들이 창의적인 스토리텔러로 틱톡 커뮤니티와 연결되고 소통할 수 있는 솔루션으로 개발되었습니다. 틱톡을 통해 창의적인 비디오와 사운드를 즐길 수 있는 브랜드 환경을 구축할 수 있으며, 일방향 소통이 아닌 참여형 캠페인을 진행할 수 있습니다.
뿐만 아니라, TikTok for Business에서는 [틱톡 마케팅 파트너 프로그램](https://ads.tiktok.com/marketing-partners)을 통해 다양한 마케팅 파트너들과 함께 브랜드의 창의적인 마케팅 활동을 지원하고 비즈니스를 성장시킬 수 있도록 지원하고 있습니다.
---
## Airbridge Update News: 네이버 성과형 디스플레이 광고 채널이 연동되었습니다!
URL: https://www.airbridge.io/ko/blog/airbridge-update-news-naver-gfa
Language: Korean
Published: 2020-12-24
[__wf_reserved_decorative]
Airbridge가 **네이버 성과형 디스플레이 광고** 플랫폼과 연동되었습니다. 이제 네이버 성과형 디스플레이 광고 집행시 Airbridge를 연동하시면 아래 기능을 활용하실 수 있습니다!🥳
- **데이터 자동 포스트백:** 포스트백을 통해 Airbridge에서 수집된 광고 식별자/전환 데이터를 네이버 성과형 디스플레이 광고 플랫폼에 자동으로 전달할 수 있습니다.
- **맞춤 타겟/유사 타겟 생성:** Airbridge가 포스트백으로 전송한 데이터를 기반으로 네이버 성과형 디스플레이 광고 플랫폼에서 ‘MAT(Mobile App Tracker) 타겟’ 을 등록해 맞춤 타겟 및 유사 타겟으로 사용할 수 있습니다. 자동 연동 방식인 MAT 타겟 기능을 사용하려면 Airbridge 대시보드에서 포스트백 전송을 설정해주셔야 합니다. 타겟 소스 이름, 연동 업체, 수집 이벤트, 수집 기간 등 MAT 타겟의 정보는 네이버 성과형 디스플레이 광고 대시보드에서 수정 및 삭제할 수 있습니다.
아래 가이드 문서 링크를 통해 Airbridge와 네이버 성과형 디스플레이 광고 플랫폼의 구체적인 연동 방식을 확인하세요!
[__wf_reserved_decorative]
### **\+ 네이버 성과형 디스플레이 광고란?**
[__wf_reserved_decorative]
네이버 성과형 디스플레이 광고는 네이버 프리미엄 지면에서 보다 효율적인 광고 집행이 가능한 광고 플랫폼입니다. 네이버 소유의 다양한 글로벌 매체 이용자들을 대상으로, AI 기술을 통해 광고주의 니즈에 맞는 오디언스를 타게팅하여 실시간 광고 입찰 및 운영이 가능합니다. 광고 퍼포먼스를 모니터링하면서 입찰가/타겟 등을 실시간으로 변경할 수 있으며, CPM/CPC/CPV 등 과금 방식을 자유롭게 선택할 수 있습니다.
[__wf_reserved_decorative]
네이버 성과형 디스플레이 광고 관리자 화면에서 로그인 계정 정보 및 광고주 정보를 관리할 수 있으며, 계정 추가도 가능합니다. 또한 광고비/성과 등 광고 집행 결과 조회 및 다운로드, 타겟 오디언스 관리, 전환 데이터 추적을 통해 광고 운영 효율을 높일 수 있습니다.
[\> 네이버 성과형 디스플레이 광고 더 알아보기(링크)](https://gfa.naver.com/)
---
## Airbridge 업데이트: Trend Report, Raw Data Export
URL: https://www.airbridge.io/ko/blog/airbridge-update-news-trend-rawdata
Language: Korean
Published: 2020-09-02
에어브릿지 기능 업데이트 소식 전해드립니다!🎉
## 1. 'Trend' 리포트 신규 릴리즈
기존에 사용하고 계셨던 Actuals 리포트의 데이터를 시각화해 분석 가능한 Trend 리포트가 Reports 탭에 새로 추가되었습니다. 아래에서 구체적인 내용을 확인해 보세요!
### 이벤트(Metric)를 시간 단위로 분석 가능
[]
- Actuals 리포트 제공 데이터의 트렌드를 차트로 시각화해 시계열 분석이 가능
- 차트 하단에서 테이블 형태로도 시간에 따른 데이터 흐름 확인 가능
- 이벤트 속성에 따라 Count 혹은 Sum 값을 제공
- 테이블 좌측 체크박스에서 차트로 분석할 데이터 선택 가능
[]
- Legend(범례)의 각 시리즈(Series)에 마우스 커서 호버 시, 차트에서 해당 데이터만 강조하여 표시 가능
[]
- 마우스 드래그로 영역 선택시 해당 구간을 확대해 모니터링이 가능
### 데이터를 쪼개어 볼 수 있는 옵션 제공
- 차트 내 우측 상단 버튼을 통해 X축 Time 설정을 시간/일/주/월 단위로 변경 가능
- Metric 최대 8개, Groupby 최대 4개까지 선택 가능
### 분석 가능한 데이터 볼륨 안내
- 리포트 차트 하단 테이블에서 Metric별 최대 100행까지 데이터 확인 가능
- CSV 또는 Google SpreadSheet로 다운로드시 Metric별 최대 1000행까지 데이터 확인 가능
## 2. Raw Data Export 기능 개선
Raw Data 탭 내 Data Export의 세부 기능이 업데이트 되었습니다. 아래에서 자세한 내용을 확인하세요!
### Request Activity 탭
Export하고자 하는 Event부터 명확히 선택한 후에 데이터를 필터링하는 로직을 구축해 이전보다 직관적으로 기능을 이해하고 쉽게 사용 가능합니다.
[]
- Request 관련 기능: New Request, Request by History
- 개선된 로직: Export Raw Data 버튼 클릭 시 나타나는 Step Drawer → Event 선택 → Property 선택 → Filter 선택 → Data Preview 확인 → Date Range 선택 → Export 요청
(*기존 로직: Property 선택 → 필터 기능을 통해 Event 필터링)
- Request History 리스트 확인 가능
### Saved Template 탭
Saved Template, 즉 저장된 템플릿이 별도의 탭으로 구분되어 있으며, 이곳에서 커스텀 설정 값을 템플릿으로 저장하고 관리할 수 있습니다.
[]
- Template 관련 기능: Create, Duplicate, Edit, Delete Template
- 커스텀 설정값을 Template으로 저장 및 관리 가능
- Saved Template을 사용해 Raw Data 요청시 Request History에 기록됨
---
## 에어브릿지 Fraud Validation Rules를 소개해요
URL: https://www.airbridge.io/ko/blog/fraud-validation-rules-update
Language: Korean
Published: 2020-02-28
에어브릿지(Airbridge)에서 트래킹한 Touchpoint와 Conversion에 대한 Fraud 룰을 설정할 수 있는 Fraud Validation Rules 기능이 추가되었습니다!
Fraud Validation Rules은 Touchpoint와 Conversion에 대해서 원하시는 Fraud 조건을 설정하고 그에 따른 처리 방법을 설정할 수 있는 기능으로, Fraud Validation Rules 기능을 통해 유료 광고 집행 시 고객사 내부 기준에 부합하지 않는 Touchpoint와 Conversion 기준을 정하고 해당 기준에 해당하는 Touchpoint와 Conversion은 기여시키되 포스트백을 전송하지 않거나 애초에 기여를 시키지 않는 등 매월 정산 시마다 발생하는 추가적인 Fraud Deduction 작업 공수를 줄이고 Fraud로 의심되는 광고 성과에 대한 광고비 지출을 막는 데 활용하실 수 있습니다.
Fraud Validation Rules은 크게 Touchpoint에 대한 설정과 Conversion에 대한 설정으로 나눌 수 있으며, 개별 Fraud 룰은 독립적으로 적용되며, Touchpoint와 Conversion이 Fraud로 분류될 시 설정한 룰에 따라 Attribution 여부가 결정됩니다.
Fraud Validation Rules은 에어브릿지 대시보드 내 [Management] > [Fraud Validation Rules] 에서 설정하실수 있으며, Rule 설정 및 수정 완료 후 해당 설정이 실제 서비스에 반영되기까지는 약 10~15분 정도가 소요됩니다.
또한 Fraud Validation Rules 설정 내역도 다른 기능들과 마찬가지로 'Activity History' 메뉴에 기록되어 확인 가능합니다.
**광고비 정산에 영향을 줄 수 있는 기능인만큼 반드시 Fraud Validation Rules 설정 내역에 대해서는 대행사 및 매체 측과 사전 협의 후 광고 캠페인을 진행하시기 바랍니다.**
## Touchpoint Fraud Validation Rules
### Fraud Validation Rules로 설정할 수 있는 조건
- Lag Time : Touchpoint와 Install이 발생한 시간의 차이(Lag time)을 이용해, 설정된 기준 이하 또는 이상의 시간차를 갖는 Touchpoint를 Fraud로 분류합니다. Fraud Lag Time 룰은 Click과 Impression에 대해 독립적으로 설정할 수 있습니다.
[]
- Frequency Capping : 설정된 시간 동안 기준치 이상의 Touchpoint가 발생할 경우, 기준치를 넘는 Touchpoint를 Fraud로 분류합니다.
[]
- Blacklisting IP : Touchpoint가 발생한 국가 IP 값을 기준으로 Fraud 여부를 판단합니다. 타겟 이외의 국가에서 발생한 Touchpoint와 서버에서 발생한 Touchpoint를 제외 처리할 수 있습니다.
[]
### Fraud Validation Rules에 따른 처리 방법
- Fraud로 표기 (Mark as a fraud) : Fraudulent Touchpoint 로 Tagging, 정상 Touchpoint 들과 동일하게 Attribution 후보에 오를 수 있으며 기여된 경우 해당 이벤트를 Postback으로 전송합니다.
- Postback 미전송 (Disable Postback) : Fraudulent Touchpoint 로 Tagging, 정상 Touchpoint 들과 동일하게 Attribution 후보에 오를 수 있으나 기여가 되더라도 해당 이벤트를 Postback으로 전송하지는 않습니다.
- Attribution 제외 (Disable Attribution) : Fraudulent Touchpoint 로 Tagging, 해당 Touchpoint는 Attribution 후보에서 제외 됩니다.
## Conversion Fraud Validation Rules
### Fraud Validation Rules로 설정할 수 있는 조건
- Conversion Traffic : Conversion이 발생한 모바일 디바이스 정보(OS Version, Language, Country, Network Carrier 등)를 기준으로 Fraud 여부를 판단합니다.
[]
### Fraud Validation Rules에 따른 처리 방법
- Fraud로 표기 (Mark as a fraud) : Fraudulent Conversion 으로 Tagging, 정상 Conversion들과 동일하게 Attribution과 Postback 전송 처리됩니다.
- Postback 미전송 (Disabled Postback) : Fraudulent Conversion 으로 Tagging, Attribution은 정상 Conversion과 동일하기 이뤄지나 Postback을 전송하진 않습니다.
- Attribution 제외 (Disabled Attribution) : Fraudulent Conversion 으로 Tagging, Attribution에서 "Unattributed"로 처리됩니다.
Fraud Validation Rules 기능과 관련하여 궁금한 점이 있으신 고객사께서는 담당 CSM에게 연락 부탁드리겠습니다.
---
## 네이버 검색 광고를 통해 앱 마케팅을 하는 방법
URL: https://www.airbridge.io/ko/blog/naver-search-ad-app-marketing
Language: Korean
Published: 2019-12-16
네이버는 우리나라에서 디지털 마케팅을 진행함에 있어서 뗄래야 뗄 수 없는 광고 매체입니다. 가장 많은 유저가 사용하는 웹 서비스가 바로 네이버이기 때문이죠. 지금도 많은 광고주들이 네이버 광고를 활용하여 고객을 유치하기 위해 노력하고 있습니다.
당연히 마케터들은 미디어 믹스 전략을 세울 때 필수적으로 고려해야 하는 매체를 네이버로 꼽습니다. 그 중에서도 '네이버 파워 링크'라는 검색 광고 상품은 네이버의 주력 상품이라고 말할 수 있습니다. 네이버에도 디스플레이 광고를 포함한 다양한 광고 상품이 있지만, 그 중에서도 검색엔진이라는 아이덴티티를 나타내면서도 가장 강력한 광고가 바로 검색 광고이기 때문입니다.
그러나 사실, 검색 광고는 최근의 광고시장 트렌드와는 조금 맞지 않습니다. 디스플레이 광고는 물론이고 동영상 광고, 다이내믹 리타겟팅 등 애드테크의 발전과 함께 다양한 광고 기법이 시장을 선도하면서, 단순 검색 광고는 한물 간 광고 기법으로 인식되기 때문입니다.
그럼에도 불구하고 결론만 말하자면, 네이버 검색 광고는 아직까지 무조건적으로 활용해야 하는 광고 상품입니다. 네이버의 높은 트래픽은 물론, 목적성이 분명한 키워드를 활용하기 때문에 키워드 당 비용이 높다고 할지라도 높은 전환율을 기대할 수 있는 특성을 가지고 있기 때문입니다.
당연히 앱 마케팅에서도 검색 광고를 통해 높은 성과를 얻을 수 있다면 좋겠죠. 애플에서도 검색 광고의 장점과 앱스토어라는 플랫폼을 융합하여 검색 광고 상품인 Apple Search Ads를 정식 출시했습니다. 그 성과가 증명되면서 앱 마케터들이 필수적으로 활용하는 매체로 자리매김했습니다.
그런데, 네이버 검색 광고에서는 정책상 앱 설치 캠페인에 대한 트래킹이 불가능하다고 합니다. 그 이유가 무엇인지 그리고 이를 Airbridge에서는 어떤 식으로 해결했는지 정리해봤습니다.
## 네이버 정책상 앱 설치 캠페인이 불가능하다
우리는 기본적으로 앱 설치 캠페인을 진행할 때 랜딩 URL을 구글 플레이(Android)와 앱스토어(iOS)로 셋팅합니다. 앱을 설치하기 위해서는 필수적으로 앱스토어에 진입해야 하기 때문에 앱을 설치할 수 있는 페이지로 연결시키는 겁니다. 물론, 이 과정에서 Android와 iOS의 OS 간 랜딩 URL를 다르게 하고, 광고 성과를 분석할 수 있는 Airbridge(에어브릿지)와 같은 써드 파티 트래킹 솔루션을 활용하여 트래킹 링크를 랜딩 URL로 설정합니다.
하지만, 네이버에서는 정책 상 네이버 비즈채널에 등록하지 않은 URL로의 광고를 제한하고 있습니다. 이로 인해 앱스토어로 랜딩되는 URL의 경우 네이버 검색 광고를 집행할 수 없습니다. 또한 써드 파티 트래킹 솔루션을 통해서 생성한 트래킹 링크도 활용할 수가 없습니다. 해당 트래킹 링크를 클릭했을 때 비즈채널에 등록되지 않은 앱스토어로 랜딩되기 때문입니다.
[]
그래서, 대다수의 기업들은 네이버 검색 광고는 무조건 웹사이트로 랜딩시키고 있습니다. 그러나 웹사이트가 없는 앱 서비스는 어떻게 광고를 해야할까요? 네이버 검색 광고를 통한 앱 설치 캠페인을 진행하지 않는다면 전환율이 높은 네이버의 검색 트래픽을 포기하는 꼴이 됩니다.
그래도 정책 상 방법이 없으니 포기할 수 밖에 없겠죠. 그런데, 가끔 네이버 검색 광고를 클릭하면 앱스토어로 랜딩되는 경우가 있습니다. 분명 정책 상으로는 안된다고 했는데, 어떻게 앱스토어 랜딩이 가능한 것일까요?
## 리다이렉션 기능을 활용한 방법
클릭하면 바로 앱스토어로 연결되는 네이버 검색 광고. 이상적이지만 뭔가 이상합니다. 분명 정책 상 불가능하다고 했는데, 앱이 있는 기업임에도 불구하고 어디는 앱스토어로 랜딩시키고 또 어디는 웹사이트로 랜딩시킵니다. 정책 상 불가능하다는 말이 잘못된 것일까요? 아니면 네이버에서 특정 기업에만 열어준 기능일까요? 아니면 다른 방법이 있을까요?
네이버 정책 상 앱스토어로 바로 연결시킬 수 없는 것은 변함없는 사실입니다. 그렇다고 특정 기업에만 해당 기능을 열어준 것도 아닙니다. 마지막으로 남은 것은 제 3의 방법입니다. 바로, 리다이렉션 경로 변경입니다.
방법은 간단합니다. 기존에 비즈채널에 등록된 URL로 광고 검수를 통과한 이후에 경로를 변경하는 것입니다. 물론, 네이버 광고 시스템에서 랜딩 URL을 변경하면 당연히 검수를 다시 받아야 합니다. 그래서 리다이렉션 방법을 활용하는 것입니다.
[]
예를 들어, www.aaa.com 이라는 URL은 www.bbb.com 으로 리다이렉션되는 URL입니다. 최종으로 도착하는 URL인 www.bbb.com 은 네이버 비즈채널에 등록되어 있기 때문에 해당 URL로 검수를 신청하면 통과가 되는 구조입니다. 이 상태라면 www.bbb.com 웹사이트로 랜딩이 되고 정상적으로 네이버 검색 광고를 활용할 수 있습니다.
여기서 검수를 통과하여 정상적으로 진행되고 있는 광고 URL의 리다이렉션 경로를 변경합니다. www.bbb.com 으로 셋팅된 리다이렉션 경로를 www.ccc.com 로 바꾸는 것입니다.(앱 설치 캠페인을 진행하려면 www.ccc.com 대신에 구글플레이나 앱스토어로 셋팅하면 됩니다.)
[]
이제 www.aaa.com 링크를 클릭하면 www.bbb.com 가 아닌 새롭게 셋팅된 www.ccc.com URL로 이동됩니다. 물론, 네이버에서는 해당 방식을 '검토를 받은 사이트와 다른 사이트로 광고를 연결하는 경우'로 인식하여 광고를 제한하고 있습니다.
(참고 : [https://saedu.naver.com/help/faq/ncc/view.nhn?faqSeq=107](https://saedu.naver.com/help/faq/ncc/view.nhn?faqSeq=107))
그래서 데스크탑 버전의 리다이렉션 URL은 그대로 두고 모바일 웹사이트에서만 앱스토어로 연결되게 하거나, 시간 차를 둬서 데스크탑 URL에 랜딩된 이후 모바일 기기에서만 앱스토어로 리다이렉션되게 개발하는 등의 우회 수법을 활용하여 앱 설치 캠페인을 진행하는 것입니다.
그러나 이 방법은 네이버 광고의 검수 시스템을 속인 것이기 때문에 활용하면 안되는 잘못된 방법입니다. 그럼 네이버 검색 광고를 통한 앱 설치 캠페인은 아예 불가능한 것일까요?
## 웹사이트를 활용해야 한다
네이버 검색 광고의 정책만 잘 따라간다면 네이버 검색 광고를 통해서도 앱 설치 광고가 가능합니다. 정책에서 다뤄진대로 앱스토어로 랜딩시키는 것이 아닌 웹사이트로 랜딩시키는 것입니다. 비즈채널에 등록된 웹사이트에 제대로 랜딩시키고 그 웹사이트에서 유저가 앱스토어로 갈 수 있게 하면 됩니다.
물론, 이 방법은 광고 클릭 이후 퍼널을 길게 만들고 유저가 빠져나갈 수 있는 여지를 줄 수 있습니다. 일반적으로 전환율을 높이기 위해 불필요한 퍼널을 줄이거나 통합하는 최적화 작업을 하기 때문에, 퍼널을 길게 만드는 것은 불편한 부분으로 생각될 수 있습니다.
[]
그러나 앞서 말한 것처럼 검색 광고는 타 광고에 비해 전환율이 높다는 특징이 있습니다. 이미 목적(심지어 구매도 가능)을 가지고 키워드를 검색하기 때문에, 이후 이탈율도 타 광고 매체에 비해 상대적으로 낮은 편입니다. 그래서 중간에 웹사이트가 있다고 할지라도 앱 설치까지 이어지는 과정에 이탈이 많이 발생하는지는 따져봐야 하는 부분입니다.
그리고 오히려 검색 키워드를 입력했는데 바로 앱스토어로 연결되는 경우, 유저들이 서비스에 대해 제대로 이해하지 않고 연결되기 때문에 다음 퍼널인 앱 설치 버튼 클릭에 대한 진입 허들이 급격히 높아질 수 있습니다. 따라서 앱스토어 이전에 서비스에 대해 잘 설명할 수 있는 웹사이트를 두고 전환율을 테스트하는 것이 필요합니다.
결론적으로 네이버의 정책을 위반하지 않는 유일한 방법이기 때문에, 네이버 검색 광고를 통해 앱 설치 캠페인을 진행하고자 한다면 이 방법을 활용할 수밖에 없습니다. 하지만, 실제로 위와 같이 장단점을 따지기도 전에 이 방법을 포기하는 경우가 많습니다.
앱 설치 캠페인을 트래킹하기 위해서 써드파티 앱 트래킹 툴을 필수적으로 사용하는데, 대다수의 써드파티 앱 트래킹 툴의 경우 트래킹 링크를 통한 앱 설치 추적은 지원하지만 웹사이트에서 앱으로 이동하는 것은 트래킹하지 않기 때문입니다.
이를 해결하고자 보통 커스텀으로 발급받은 트래킹 링크를 버튼에 a 태그로 연결시키는데, 검색 키워드나 마케팅 채널에 따라 해당 링크를 동적으로 바꾸기 힘든 치명적인 단점이 존재합니다. 개발을 통해 파라미터를 동적으로 받아오게 처리를 할 수 있겠지만, 유지보수가 힘들고 특히 해당 버튼을 누르지 않고 앱스토어에서 따로 앱을 설치한 경우는 트래킹할 수 없습니다. 결국 트래킹 링크라는 방법을 활용했기 때문입니다.
## Airbridge(에어브릿지)를 활용한 웹 to 앱 트래킹
Airbridge(에어브릿지)는 이와 같은 문제점을 해결하고 웹사이트와 앱 사이의 User Journey를 추적하기 위해 WEB SDK를 활용합니다. WEB SDK를 통해 네이버 검색 광고로 들어온 유저들의 데이터를 기록하고, 이후 앱을 설치하면 동일 유저로 인식하여 해당 설치를 네이버 검색 광고에 기여하는 것입니다.
따라서 Airbridge(에어브릿지)를 활용하면 네이버 검색 광고를 통해 웹사이트를 방문하고 이후 앱을 설치하는 유저를 트래킹할 수 있습니다. 사실 네이버 검색 광고 뿐만 아니라 기존에 웹사이트에서 구글 애널리틱스를 통해 분석하였던 유입경로 트래킹이 가능합니다. WEB SDK가 UTM 파싱 기능을 지원하기 때문이죠.
마케팅 성과 분석에 있어서 웹사이트와 앱은 둘 간의 기술적 차이 때문에 피할 수 없는 트래킹 분절 구간이 발생합니다. 그래서 써드파티 트래킹 툴이 개발되었고, 트래킹 링크를 통한 트래킹이 일반적으로 사용되고 있지만 웹 to 앱은 또 다른 미지의 영역입니다. 특히, 네이버의 영향력을 기반으로 구축된 강력한 웹 기반의 사용자 환경은 다른 국가에 비해 우리나라 사용자들이 웹사이트와 앱을 복합적으로 사용하는 상황을 만들었습니다.
Airbridge(에어브릿지)는 웹과 앱의 성과를 함께 분석하고자 하는 마케터들의 니즈에 맞춰 Airbridge WEB SDK를 개발하였습니다. Airbridge WEB SDK는 해당 스크립트가 설치된 웹사이트로의 유입을 측정하는 것 뿐만 아니라, 웹사이트 내에서 발생한 전환을 기록하고 웹사이트에서 앱을 설치한 유저까지 트래킹합니다.
지금까지 네이버라는 광고 채널에서는 앱 설치 캠페인을 진행하지 못한다고 생각하셨나요? 아니면, 웹사이트에서 따로 개발을 통해 앱스토어로 리다이렉션을 하고 계셨나요? Airbridge WEB SDK를 활용하면 네이버 검색 광고는 물론 기존의 웹사이트 광고도 함께 트래킹 할 수 있습니다.
네이버 검색 광고를 통해 앱 설치 캠페인을 진행하고 싶으시거나, 웹 to 앱 트래킹 기능에 대해 궁금하시다면 언제든지 저희 AB180으로 연락주세요!
# Chinese Articles
---
## Airbridge × Liquid：Steam 营销实战指南
URL: https://www.airbridge.io/zh/blog/game-studio-steam-marketing-playbook
Language: Chinese
Published: 2025-07-02
**_Airbridge 与 Liquid Advertising 全球媒体副总裁 Nikki DePaola 的深度访谈实录_**
对于许多中型厂商而言，在 Steam 上发行游戏不是选择题而是必选项——这里聚集数千万 PC 玩家，具备原生社区生态，甚至能为主机版发售造势。
但预算有限、又具备一定规模的中型团队，在部署 Steam 时常常会遇到一个问题：目前大多数针对 Steam 的营销建议，要么是为零预算的独立开发者准备的，要么是手握全球 IP 的 3A 大厂专属打法。对于中型团队来说，常常因为得不到合适的建议而无从下手。
洞悉到这一痛点，Airbridge 与 Airflux 的 CEO [Roi Nam](https://www.linkedin.com/in/roinam/)，专门邀请了 Liquid Advertising 的全球媒体副总裁 [Nikki DePaola](https://www.linkedin.com/in/nikki-depaola/)，来聊聊她在服务 Blizzard、Nexon、Bethesda 以及一系列中型、独立游戏时总结的经验：哪些方法有效，哪些容易踩坑。
**🎥 \[点击观看完整采访视频\]**
**
**
## **「愿望单」很重要，但更重要的是挖掘市场的「真需求」**
愿望单（Wishlist）是 Steam 上少数可追踪的预售指标，能够算法曝光率，也能初步判断市场兴趣。但光靠愿望单，其实未必能撑起一款游戏。
> “我们见过一些客户因为在愿望单上名列前茅而兴奋，却在看到实际转化率时感到失望——因为很多时候愿望单上的排名，是靠激励或活动强推出来的结果。”
强推或激励获取的愿望单虽然能带来一个好看的排名，却不一定能带动实际销量。Nikki 的建议是：大家更需要关注「质」，而不是单纯的「量」，**70% 以上的愿望单都应该是真实用户自发添加的，而且要和目标地区相呼应。**
> “漂亮的愿望单排名能为你赢得平台曝光，但如果没有真实的用户口碑同时加码，这种曝光很难真正打动目标受众。”
换言之，如果你的游戏没有在 Discord、Reddit 或 YouTube 评论区引发话题，愿望单再多也难撑起热度。Steam 看重的是势能，不单纯只是数字。
[Roi Nam，Airbridge 与 Airflux CEO & Nikki DePaola，Liquid Advertising 全球媒体副总裁]
## **发行方案的「三架马车」：社群运营+ PR + 广告**
很多团队以为上线只要建个 Steam 页面、发个预告片、花点广告预算就行了。但实际上，大多数 Steam 玩家都不是看到一次广告就买单的「冲动型选手」，他们需要完整的认知路径：他们追寻口碑热议、查看媒体评测、浏览玩家反馈，最终通过他们信任的渠道，来了解游戏。
> "愿望单本质反映的是用户对游戏的期待感。想透这点后，关键问题就不再是「如何达成愿望单目标」，而是「如何让消费者对这款游戏的发售感兴趣」”。
这种思维转变，让愿望单从目标变为结果，并扩展了运营维度。Nikki 建议每个想认真对待 Steam 市场的发行，在上线新游戏时，都应该好好运营以下六大模块：
- **高质量的商店页：**强视觉冲击力 + 明确核心玩法 + 优质预告片
- **精准买量投放：**覆盖玩家主动发现游戏场景
- **深度社区运营：**重点深耕Discord与Reddit
- **PR 战略：**争取玩家信任的媒体深度评测
- **内容排期：**从预告、试玩到测试，分阶段释放情报
- **数据监测系统：**用Airbridge等工具追踪真实转化
简单说，在 Stream 上线并不是一个结果，而是一个需要长线运营的过程，需要持续打磨、用心经营，才能有收获。
## **上线日不是终点，而是「公测」的开始**
首发日绝非终点，而是公开测试的开始。从上线的一刻开始，Steam 和玩家都会密切关注你的作品。在这个阶段，你必须保持嗅觉灵敏，随时准备灵活调整你的策略。
除定制化策略外，还有几个通用要点需要牢记在心：
> “你得有 Plan B。比如如果上线后口碑一般，甚至差评偏多，团队要怎么应对？”
上线第一天很关键，一些游戏上线首日就成为爆款，而一些则举步维艰。无论首发爆火或遇冷，团队都需要有快速响应的机制。Nikki 建议大家能够提前部署衡量工具，捕捉早期趋势并迅速反应。
> “另一个建议是，在上线前就部署好归因平台，确保能及时看到 Day 1、Day 3 的变化——而且你的团队能快速反应，广告活动能够支持针对创意、受众和地区的精细化调整。”
越早知道每一个策略的效果，你就能越快矫正，并快一步走向正轨。
如果你的游戏支持账号登录或联机，不妨把这些第一方数据用起来，有助于后续做 LTV 建模、再营销和区域洞察。
[]
## **品牌传播与效果营销：不是「二选一」，而是「组合拳」**
过去做 PC 与主机游戏营销，更注重品牌塑造——讲故事、做预告、铺媒体。效果营销更像是移动营销的专属，一切以点击、安装和 ROAS 为导向，缺乏品牌叙事空间。
但效果广告现在已经逐渐成为 PC 与主机游戏营销不可忽视的一部分。特别是随着抢先体验、LiveOps 等运营模式流行，**品牌+效果双轮驱动**，已经成为很多成功团队的共识。
Nikki 也认为，品牌与效果广告搭配组合，能达到更好的效果。最有效的团队，往往是将两种营销人员都配备齐全，将移动与 PC 营销的优势相融合：
- 移动游戏营销：更擅长数据驱动与 ROI 优化
- 主机/PC营销：更精通品牌塑造与用户心理
> “顶尖工作室正融合双方优势：基于触达度、频次、注意力和声量份额等指标，平衡媒体类型选择。”
这种平衡对于中型团队尤为重要：移动营销人能负责精准的 KPI 追踪，主机营销人贡献创意直觉。当双方在全漏斗目标上对齐，广告活动就能推进更快、更有效。
## **PC 与主机玩家，要的是「真感受」，不是「花架子」**
与移动用户不同，PC 和主机玩家并不会看到广告就轻易下手。相比移动玩家，他们极少冲动下载，更不会单次曝光就转化，更习惯做功课、看评论、逛论坛。
> “耐心是第一要义。PC / 主机玩家转化周期长，他们需要看到足够多次，才会认真考虑是否购买。”
更重要在于：主机玩家不只看到广告，还关注其呈现环境——广告位本身、评测内容、预告片质感、上下文语境及社区讨论，这些都会塑造玩家对这款游戏的观感。
> “要让玩家在愿意探索新内容的时候看到你。场景价值在主机领域远超移动端。”
谨慎选择广告位：廉价低质网站（Nikki 称之为「为广告而生」的站点）可能能为你获得更多点击，但也有可能会伤害品牌形象。在主机领域，广告环境比以往更关键。
> “优化注意力与影响力，转化自会跟随——即便归因在这些广告位更具挑战。”
## **有效但容易被忽视的三个渠道**
如果你已经在 Meta、YouTube、Google 上投放，这很好，证明你已经达到了 PC 和主机营销的基本线。但实际上，还有很多被忽略的渠道值得中型团队（尤其是亚太团队）去尝试：
- **深度激活 Discord 和 Reddit 等社群运营**: 所有人都知道玩家聚集在 Discord 和 Reddit。但「知道存在」与「真正运营」是两回事。社群运营不仅仅是推送公告或转发预告片。社区是真实对话发生的交流场：玩家表达兴奋、发泄不满、交换攻略、传播梗图——最终定义你的游戏口碑。在玩家极度看重世界观、品质和社区的PC/主机领域，这种互动才是真正影响成败的关键。
- **高触达程序化广告**: 多数团队只盯着转化层——反复再营销、再营销、再营销。但 PC 和主机玩家不会点击即购。他们会暂停、比较、深思熟虑。这就是Nikki强调全漏斗程序化策略的原因：覆盖上层（认知）、中层（考虑）、下层（转化）全阶段。通过智能频次控制、用户路径规划和受众定向，程序化广告能自然引导玩家从认知到转化——整个过程不像广告轰炸，而像精准时机介入。
- **OTT/CTV 广告（如 Netflix、Hulu、Disney+**: OTT 和 CTV 渠道仍未被充分利用，部分是由于归因精度问题，或者是这些渠道在非西方市场并不占据主导地位。但在欧美市场，CTV 绝非小众。三星、LG、Vizio 等设备商所提供的数据洞察，能显示玩家正在观看什么内容，在某些情况下，甚至能追踪广告后的游戏行为。而随着 Netflix、Hulu、Disney+ 等流媒体平台开放更多广告位，开发者也能更容易地触达有探索欲的玩家，投放更精准，测量更科学。
[]
## **创意不是看数量，而是看「信息清晰度」**
不同于移动游戏里海量 A/B 测试素材的模式，PC/主机创意重在精准而非堆量——是否能够向玩家清晰传递关键信息，十分重要。Nikki 建议：
- **针对不同的目标受众，定制素材信息：**针对核心玩家群定制创意，突出游戏独特卖点，并分群体测试实际共鸣点
- **明确游戏类型和差异化卖点：**避免模糊表述。清晰传达游戏类型、目标受众和差异化亮点
> “有时候玩家可能看完一些素材，都不知道这是什么游戏。不要只是概括把游戏描述为「射击游戏」，而要描述「80年代风格、合作、恐怖题材、带 roguelike 要素的 FPS」，并且通过强力的视觉和明确的信息传递来辅助说明。”
- **视觉素材要体现品质感，传递品牌定位：**视觉素材应传递品质感、完成度和市场定位，尤其当你想让用户认为这是“3A潜力股”的时候
> “真正达到3A水准的关键在于传递出产品气质。”
## **跨平台营销策略：精细化运营，拒绝一刀切**
越来越多工作室选择多平台发行：Steam、主机、手游甚至HTML5/网页版。理论上这能扩大覆盖面、增加变现途径，并灵活触达玩家所在场景。
策略虽可行，但团队必须要认识到不同受众对平台的认知差异——尤其在欧美市场。
比如在亚太地区，移动游戏是主流的游戏平台。但在欧美，核心玩家会把「移动游戏」当作另一种类型，和他们在主机上玩的游戏完全不是一个概念。
> “如果你向主机核心玩家推送「游戏已在App Store上架」的广告，他们的认知可能是‘这是另一类手游’——而非和兄弟在PlayStation上玩的那款。”
这种认知错位可能带来损失，即使游戏跨平台内容完全一致。错误场景的错误信息，可能会让玩家感觉降低兴趣。如何能在跨平台营销的时候精准穿题正确信息呢？以下是一些参考做法：
- **受众分群：**将主机玩家、PC玩家、移动端用户视为需求完全不同的群体
- **信息调适：**根据不同平台提炼价值主张——向主机玩家强调高端画质和操控感，对手游用户突出便捷体验****
- **或保持中性**：避免在早期曝光中强调平台，让玩家自己选择在哪
## **别只看最后一次点击，漏斗全貌更重要**
末次点击归因是多数游戏营销人的起点。这固然是实用基准线，但随着战役升级，测量方式必须同步进化。
PC 和主机玩家极少在单次广告曝光后就转化，尤其在 Instagram 等渠道。这需要频次积累、信息清晰度和正确时机的精准触达。
> “全漏斗归因对所有游戏都至关重要。现实情况是：几乎没人第一次在 Instagram 看到广告就立刻转化。这需要长期培养——通过持续的频次建设，配合阶段化信息传递，逐步建立产品认知，并引导至购买和安装。”
如果你只追踪点击，那电视、户外、主机等渠道的「辅助效果」可能很容易被忽略，但这些曝光又的确在加深品牌印象。
无论是否采用跨平台营销，其中的核心逻辑其实是相通的：让信息匹配用户决策阶段，策阶段，并用工具揭示渠道协同效应使用能串联全漏斗的平台，才看得清、调得快、改得准。
[]
## **进军美国市场？警惕三大误区**
开拓美国市场，绝非只是简单翻译广告、照搬其他工作室策略那么简单。缺乏对本地行为习惯、预算基准和文化敏感点的理解，极易浪费预算错失时机。
Nikki 指出亚太厂商西进时的常见错误：
- **盲目复制策略：**亚太工作室大多数都是「风险厌恶型」选手，所以在进军美国时，不少选择了参考别人的成功模版。但实际上，每款游戏的预算、用户和策略都不一样，别人的打法未必适合自己。
- **低估预算门槛：**许多亚太发行商误认为半额预算就能参与高端市场竞争。但在溢价市场，缺乏足够投入或强力测量工具，很难掀起水花。
- **忽视文化细节：**美国市场由碎片化的亚文化驱动，潮流迭代迅速。若缺乏本地人才把控创意和媒体组合，必然错失关键机会点。
> “你的游戏不是其他的游戏。每款游戏所面临的挑战和掌握的预算都不一样。”
## **小预算独立游戏的制胜之道**
我们讨论了大量高预算发行案例。小工作室有成功样板吗？Nikki提到了独立游戏[《Squirrel with a Gun》](https://store.steampowered.com/app/2067050/Squirrel_with_a_Gun/)。
> “这款游戏顾名思义，本质是松鼠模拟器，只是主角有枪。你不能用 3A 标准做广告——比如首页强推或精致软文，这些做法完全不合适。”
Nikki 团队聚焦游戏核心特质：
> “这款游戏的营销关键，在于放大游戏的品牌个性——内核就是荒诞搞笑。如何强化这个特质？最终我们做了大量原生感强的社区驱动传播：比如与 Discord 社群共创梗图，在 Instagram 联动搞笑账号。”
团队将此与智能再营销和转化层支持结合。成果如何？广告活动既忠于游戏调性，又带来了真实都增长效果。
## **与 Liquid Advertising 一起打造突破性的游戏营销战役**
总而言之，Steam没有通用成功公式——但只要信息精准、测量可靠、策略适配游戏特质，中型工作室就能站稳脚跟。
- **愿望单需真实热度支撑：**自然口碑+区域热度永胜刷量数据
- **Steam 不仅是商店更是发现引擎：**通过PR/内容/社区/广告组合拳构建势能，而不仅仅是一个简单的宣传片和落地页
- **发行日只是起点：**追踪早期信号、准备 Plan B、为 D2+ 保持灵活和迅速响应
- **品牌与效果营销双管齐下：**最强战役=大胆创意+精准触达+全漏斗策略
- **创意要玩家导向：**通用信息无效，分群体定制沟通语言
- **西进须本土洞察**：杜绝复制粘贴，预算适配市场，文化理解优先
想了解更多资讯？欢迎查看[liquidadvertising.com](http://liquidadvertising.com) 和 [airbridg.io](http://airbridg.io)
### **想进一步优化营销表现？Airbridge 助你高效提升 ROI**
无论是 PC、网页、主机还是移动端，Airbridge 都能助你洞悉全平台玩家行为，实时洞察追踪跨平台用户流、优化效果表现：
- **高级 LTV 分析：**支持按国家、创意、关键词等多维度拆解
- **集成多方数据：**在一个面板，丝滑接入来自不同广告聚合平台的多方数据。现 Airbridge 已支持 Google AdMob、AppLovin MAX、Unity LevelPlay 等聚合平台
- **pLTV实时预测：**预判营收潜力，帮助快速判断广告效果
- **分钟/小时级留存分析：**捕捉玩家行为脉络，获得实际洞察
- **一价全包，无隐藏费用：**一次解锁所有专家级分析功能，无附加升级包
- **便捷受众与原始数据导出功能**：支持快速提取筛选后的目标用户群，并将原始数据无缝导出至CSV或谷歌表格，助力定制化仪表盘搭建及深度分析需求
无论您是新锐独立团队还是成熟发行厂商，Airbridge 都为您提供决胜市场的核心工具，助您在加速迭代的游戏市场中持续领跑。立即访问 airbridge.io 掌握解决方案。
### **与 Liquid Advertising 一起打造突破性的游戏营销战役**
- 访问 [](https://liquidadvertising.com)[liquidadvertising.com](https://www.liquidadvertising.com)，了解顶级游戏工作室是如何在 Steam、主机平台等多个渠道成功发行的。
- 欢迎关注 Liquid Advertising 的 [LinkedIn](https://www.linkedin.com/company/liquidadvertising) 与 [Instagram](https://www.instagram.com/liquidadvertising/?hl=en)，获取更多实战案例、行业洞察与创意灵感。
- 日本地区请关注我们的 [Note](https://note.com/liquid_ad_agency) 和 [X](https://x.com/liquid_ad_jp)
- 中国大陆用户可通过[微信公众号](https://mp.weixin.qq.com/s/6AJfTq5KltEtcdJINypAaw)获取内容更新
想为你的游戏定制一套更适配、更有力的营销策略？欢迎联系 [Nikki](https://www.linkedin.com/in/nikki-depaola/) 和 Liquid 团队，开启合作对话。
---
## Playio 如何切换到 Airbridge 并通过单一平台管理全球营销
URL: https://www.airbridge.io/zh/blog/playio-mmp-migration-global-marketing
Language: Chinese
Published: 2025-04-01
## 游戏玩家的最佳价值 — 游戏玩家专属奖励应用 Playio
### Q. 能否介绍一下您和公司？
大家好，我是 Playio 营销团队负责人 Hayoung Park。从职业生涯之初，我就对游戏充满热情——我的第一份工作就是在游戏公司。目前，在 GNA Company，我正与 Playio 一起朝着"Best Value for Gamers"的使命努力，构建一个健康的游戏生态系统。
### Q. 能否介绍一下 Playio？
从本质上来说，Playio 是一款基于游戏时长的奖励应用。用户根据游戏时长赚取积分，然后可以用积分兑换商品。然而，Playio 的目标不仅仅是一个简单的奖励应用——它是一个让游戏玩家的独特价值得到认可的平台。
例如，我们不只是提供奖励，而是根据用户的实际游戏时长将他们分为不同的等级和层级。Playio 也是一个多元游戏玩家聚集的社区。对用户来说，这是一个享受游戏和社交的空间；对游戏发行商来说，这是一个被公认拥有高活跃度"核心玩家"的营销渠道，能够驱动安装和留存。
[访问 Playio](https://playio.co/?LANG_CD=ENG)
[]
### Q. 我们了解到 Playio 在多个国家运营。
是的。Playio 目前在韩国、美国、日本和台湾运营。我们还计划拓展到香港、澳门等更多市场。
### Q. 营销团队的组织架构是怎样的？负责哪些工作？
我们的营销团队分为 B2B 和 B2C 两个方向。B2B 团队于 2025 年新成立，与销售团队紧密合作，向客户展示 Playio 作为营销渠道的效果。他们发布文章、发送电子通讯、挖掘潜在客户以获取新客户，并加强与现有合作伙伴的关系。
B2C 团队专注于帮助用户发现和享受 Playio。除了获取新用户外，他们还开展各种营销活动来保持现有用户的满意度和参与度。
## Playio 选择 Airbridge 的历程
### Q. 听说 Playio 已经用到了第三家 MMP，最终选择了 Airbridge。这背后有什么故事？
去年，我们进入了日本市场，开始与当地一家发行商合作，该发行商同时也负责营销。这意味着我们需要限制发行商的仪表盘访问权限，仅能查看日本的数据——他们不应该看到其他国家的业绩。我们之前的 MMP 表示，这种国家级的数据分割是不可能实现的。
因此，我们开始寻找能够按国家限制数据可见性的 MMP。大多数 MMP 都无法提供这一功能。Airbridge 是唯一一家确认可以做到的，这也是我们决定迁移的原因。
但这并不是唯一的原因。个人而言，我一直觉得 Airbridge 相比其他 MMP 拥有更加直观的 UI/UX。营销绩效应该是组织中每个人都能查阅的——不仅仅是数据分析师或效果营销人员——而 Airbridge 直观的界面使这成为了可能。
### Q. 迁移过程花了多长时间？过程如何？
大约一个月。正如我提到的，我们有按国家级分割数据权限的特殊需求，而且在集成过程中有很多额外的请求。我们频繁召开迁移会议，但得益于 Airbridge 团队的主动支持，一切都进展顺利。
MMP 迁移不是经常做的事情，而且往往困难且耗时。Airbridge 提供了一个系统化的流程和详尽的检查清单，使我们能够快速完成迁移。他们还进行了全公司范围的入职培训，让更多同事能够使用 Airbridge。
[]
## Playio 如何使用 Airbridge
### Q. Playio 在多个国家运营。国家级策略和效果衡量一定非常重要。
没错。我们针对不同国家采用不同的策略和媒体渠道。韩国于 2021 年上线，日本于去年 4 月上线，台湾于 11 月上线。在韩国，去年 D30 留存率超过了 30%，用户忠诚度非常高。我们不仅专注于获取新用户，还通过内容策略、角色策略和 CRM 策略来有效传达品牌愿景。
台湾和日本正处于快速增长阶段，更加侧重于新用户获取。我们积极投放 Google、Meta 等主要渠道以及 DSP，优化各国家的特定策略。
### Q. Playio 在移动营销优化方面最常使用 Airbridge 的哪些功能？
我们广泛使用 Airbridge 的指标和数据来衡量各媒体渠道和素材的广告效果。Playio 特别看重的一个功能是报告的多样性。其他 MMP 无法提供这样丰富的报告类型。Airbridge 让你可以按任何你想要的指标自定义报告——这是一个巨大的优势。
[探索 Airbridge 报告功能](https://help.airbridge.io/en/guides/airbridge-report-overview)
我们营销团队的目标与公司整体 KPI 保持一致——包括新用户增长和留存率提升等指标。我们利用 Airbridge 多样的报告功能创建和分享图表，让全公司一起跟踪这些目标。得益于入职培训，营销部门以外的同事现在也能直接自己创建报告了。
另一个关键功能是通过自定义渠道进行链接追踪。除了移动广告，Playio 还大量开展网红营销、内容营销和外部合作推广。对于这些活动，我们使用自定义渠道来按活动和网红衡量和优化效果。
[探索 Airbridge 自定义渠道](https://help.airbridge.io/en/guides/custom-channels)
### Q. 能否分享使用 Airbridge 取得的营销成果案例？
在韩国，我们有一个通过结合 Airbridge 漏斗报告和内部数据分析来提升效果的案例。提升留存率是韩国市场的关键目标。我们利用内部数据识别用户的"顿悟时刻"，制定各种行动方案引导用户到达这些时刻，进行 A/B 测试，并使用 Airbridge 漏斗报告来检验和优化结果。最终，通过优化到达顿悟时刻的路径，韩国 D30 留存率超过了 30%。
我们还有一个按国家优化营销的案例。在台湾，Airbridge Actuals 报告显示，在 UA 投放中，视频素材的表现明显优于图片素材。基于这一数据，我们提高了视频素材的比例，最终用户获取量提升超过 100%，UA CPA 降低了 40%。
> 想了解更多关于 Playio 使用 Airbridge 的移动营销成果？
[阅读完整案例研究](https://www.airbridge.io/en/customers/playio-d30-retention-cpa-optimization)
### Q. 作为营销人员，Airbridge 最大的优势是什么？
我认为最大的优势之一是，即使作为 MMP，Airbridge 的数据可视化设计也非常出色。有很多可自定义的报告、灵活的数据组合，设计也非常直观。在使用过多个 MMP 之后，Airbridge 是唯一一个在仪表盘中提供如此丰富和直观体验的。
[]
另一个重要优势是，非营销人员也能轻松使用。对于营销或数据岗位以外的人来说，查看数据可能会让人望而生畏。但 Airbridge 直观的 UX/UI 和便捷的操作方式，让即使没有营销背景的同事也能轻松上手。
## Playio 的愿景：游戏娱乐的未来（Reward 2.0）
### Q. Playio 不仅仅是简单的奖励渠道——它认可并尊重游戏玩家的价值。Playio 对游戏奖励平台未来的愿景是什么？
游戏玩家出于各种动机在玩游戏——从简单的娱乐到成就感、社交连接和认同感。Playio 计划提供更多功能、内容和奖励来放大这些多元化的动机。
我们相信 Playio 的愿景超越了仅仅做一个奖励平台。我们称之为"Reward 2.0"。在这一愿景下，我们旨在继续认可用户作为游戏玩家的价值，同时成为一个为游戏发行商提供更精准广告投放的平台。
### Q. 在 Playio 走向 Reward 2.0 的道路上，营销团队的目标是什么？
正如我所说，Playio 致力于在简单奖励之外创造一个健康的游戏生态系统。营销团队朝着同一个目标前进。团队中的每个人都是游戏玩家，我们努力创造更多符合产品方向、为游戏玩家带来价值的体验。这就是 Playio 做营销的原因。
[]
### Q. 最后还有什么想分享的吗？
经历了多个 MMP 之后，我很高兴最终选择了 Airbridge。复杂的迁移过程很快就完成了，现在全公司都在通过 Airbridge 的数据实现增长。凭借直观的仪表盘、响应迅速的 CSM 支持以及众多其他优势，我无法想象再回到其他 MMP。
Playio 的每一个人都在努力打造更好的服务。新功能不断推出，服务正在快速发展，请继续关注 Playio 的旅程。我们也在持续招募优秀人才，一起踏上这段精彩的征程。
[查看 GNA Company 招聘职位](https://gna.career.greetinghr.com/team)
想用 Airbridge 优化您的全球移动营销效果吗？
---
## Airbridge 2024 年度复盘：升级、突破、连接
URL: https://www.airbridge.io/zh/blog/airbridge-recap-2024
Language: Chinese
Published: 2025-01-14
2024，是充满机遇与变革的一年。
游戏、短剧、网文作为「出海新三样」惊艳了世界，《黑神话：悟空》成为传奇、刷屏的「霸总」与爽文成为了世界的多巴胺引擎；AI 重新定义了效率，也重塑了生活方式——与 AI 聊天、和纸片人约会、跟虚拟家教对话已成为越来越多人的日常；小游戏从「红海」变「血海」、女性向游戏遍地开花、PC 游戏市场焕发生机，机遇来临之际，挑战也相伴相随。
在希望和挑战交织的缝隙里，除了拥抱变化，Airbridge 也在思考中大步向前，在移动测量和跨端分析的领域里，努力走在引领潮流与革新的第一线。
辞旧迎新之际，我们想以数字为引，与你分享属于我们的 2024。
[__wf_reserved_inherit]
## 功能革新、技术升级，保持增长加速度
与 2023 年相比，2024 年 Airbridge 所服务的游戏发行数量增加了 4.3 倍，涵盖超休闲游戏、RPG、赛车游戏、AR 音乐游戏、互动视频游戏平台等各个品类。与此同时，Airbridge 的非游应用追踪数量也在 2024 年提升了 90%，越来越多的移动开发者选择与 Airbridge 一同，见证增长旅程。
「降本」与「增效」，依旧是全球开发者 2024 年的主旋律：
- 一方面，在面对众多的买量渠道、不同的操作系统、环境各异的区域市场和不同目标受众时，开发者要在零碎、分散的数据中，找准 ROI 更高的买量渠道组合，将「预算花在刀刃上」；
- 另一方面，开发者需要尽可能节省每个营销环节的运营时间与成本，用更好用、智能的平台与工具，提升增长 ROI。
面对这些痛点，Airbridge 不断精进归因与衡量技术，通过对全链路归因解决方案与服务能力的持续打磨，助力全球开发者提升增长势能。
- 在产品能力上，Airbridge 更新且优化 pLTV、跨端归因、深度链接等功能，用更智能、更精细的实时数据分析，帮助全球出海开发者应对行业挑战，提升增长 ROI。
- 在战略合作上，我们打通了 Mintegral Target ROAS 的数据通道，接入了 adjoe、ayeT、Kwai for Business、Microsoft Ads 与 Digital Turbine 等全球主流媒体渠道，并持续深化与 Amplitude 及 Braze 等主流营销技术解决方案的集成合作，助力客户实现更加精准和广泛的营销活动。
- 在团队支持上，Airbridge 始终保持「专业支持、高效响应」的服务原则，针对每一位国内出海客户，更是配备了 1 对 1 的专业服务团队，彻底解决不少开发者在归因过程中，面临的问题解决不及时、响应慢等问题。
当环境、规则、常态都在改变，夯实的技术能力与服务水平是平台站稳脚跟的底气与功底。而在 2024 年，越来越多的合作伙伴也选择通过 Airbridge 打通营销卡点，用更精准的数据与洞察，把握增长脉搏：
- 2024 年，全球客户通过 Airbridge 的 10+ 个在线报表，生成超 1400 万份报告，获得了更精准、系统的数据分析；
- Airbridge 用户在 2024 年生成了超 8.613 亿个跟踪链接，总点击量高达 2.628 万亿次、平均每月 200 亿次点击，使用量远超短链接平台 Bitly；
而在客户规模上，Airbridge 的游戏与非游应用均在全速增长：
[__wf_reserved_inherit]
## 新突破，新佳绩
助推游戏与应用开发者征战全球
与 2023 年相比，2024 年 Airbridge 所服务的游戏发行数量增加了 4.3 倍，涵盖超休闲游戏、RPG、赛车游戏、AR 音乐游戏、互动视频游戏平台等各个品类。与此同时，Airbridge 的非游应用追踪数量也在 2024 年提升了 90%，越来越多的移动开发者选择与 Airbridge 一同，见证增长旅程：
- 在 Airbridge 的帮助下， [Webzen](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization) 深度优化短视频创意素材，并通过 pLTV 功能，提前预测用户收入潜力最高的投放渠道，精准分配投放预算。在 Airbridge 团队的全力支持，Webzen 旗下《Mu Monarch》的玩家登录率提升了 90%。
> "_Airbridge 灵活、高度自定义的报表功能，让我们的数据监测变得更简单高效。最棒的一点是，无论何时何地，Airbridge 团队总能及时响应、高效支持，专业性与给力程度堪比我们的内部营销团队。Airbridge 不仅提升了我们的投放表现，也让我们的运营工作更加丝滑。我们对 Airbridge 团队无比满意！"_—— Hailey Kang，Webzen 营销负责人
- 出海发行厂商 VortexGravity 通过 Airbridge，实时掌握各个投放渠道的买量数据、高效优化投放策略，在有限的投放预算中，高效提升买量效率，轻松省下了50%的归因成本
> "_在迁移到 Airbridge 后，我们的投放效果追踪变得更加简单、可控。Airbridge提供了一体化、自定义的面板，让我们可以对不同国家与地区，各广告渠道的安装量、CPI、ROAS、收入、活跃用户等关键指标一目了然。此外，原始数据均可通过 API 接入，无需额外费用获取，大幅缩短了我们的跨团队决策链路。“——Anthea，VortexGravity的海外增长负责人_
- 被称为「抓狂闹钟」，全球下载超 7500 万的 Alarmy，通过使用 Airbridge 的**漏斗分析**功能，从投放渠道、广告活动为维度，分析用户从哪里来、为什么来，并进一步拆解用户的下载、互动、试用、订阅与付费行为，进而找到**转化效果最优、用户质量最佳的营销渠道组合**，以数据为导向，实时优化广告策略、获得了更高的 ROI。
> "_作为一个 MMP，Airbridge 非常清楚如何提供真正有价值的服务。Airbridge 客观、透明、准确的投放效果分析，能帮助我们准确把握营销效果，提升投放效益。”——Sean，Delight Room BD 总监_
- 心理健康应用 Shmoody 通过 Airbridge 的一体化集成面板，一站查看安卓、iOS SKAN 等多个端口的归因情况及营销效果，大幅减少在不同面板中来回横跳的时间与运营成本。
> "Airbridge 一体化的数据面板，让我们能够将零散的归因数据清晰地整合在一起，通过直观、全面的数据指标，轻松分析安卓、iOS 及 SKAN 等端口的数据。
> 在 Airbridge 透明的「一价全包」模式下，细分面板数据、买量成本数据等刚需功能都能直接使用，不用像别的归因平台一样，需要我们支付额外费用，这对我们来说意义重大。
> 在 Airbridge 的帮助下，我们实现了安装过百万、月活用户增长率超 500% 的目标。正如我们 CTO 所言，Airbridge 在各个方面的能力目前要优胜于其他平台."——Michael McSweeney, 联合创始人兼首席产品官
除了移动应用，《黑神话：悟空》里程碑式的成功，也让 PC 与主机游戏重新站在了行业聚光灯之下，让「跨端营销」再次成为热议话题。
当「大屏游戏、小屏营销」成为 PC 与主机厂商加速出圈、打开流量池的刚需，精准的跨端归因衡量，就自然成为了厂商提升投放效率的不二法门。
- Nexon 通过 Airbridge 的跨端归因解决方案，对其 PC 与主机游戏《第一后裔（The First Descendant）》在移动端上的跨端营销表现实现精准追踪，并基于平台数据，对多个渠道的在投素材进行调优，有效提升投放活动优化效率。
> "通过 Airbridge，我们在 PC 和主机平台上获得了与移动归因同样的可见度，大大减少了我们在 PC 和主机游戏投放活动中的不确定性."——Sunny Kim，Nexon 用户获取团队负责人
出色的助攻战绩，也让 Airbridge 逐步收到更多瞩目和肯定。2024 年，Airbridge 被欧美头部行业媒体 Business of Apps 评为全球头部领先 MMP；同时也被行业头部移动产品智能商业分析平台七麦数据评为「年度出海实力服务商」。
[__wf_reserved_inherit]
## 连接全球，出谋划策
与全球开发者一同见证增长
流量红利探顶、内卷程度升级，无论是出于版图扩张，还是迫于压力，对于全球开发者来说，破圈与出海在 2024 年都是一件自然又必然的事。
但在全球大环境整体低迷的情况下，海外市场的情况变得更加复杂。开发者是否能快速在异国市场找准机遇、避开暗雷、解决营销难题，成为每一个出海远征军的头号难题。
洞悉到这一点的 Airbridge，也在 2024 年为合作伙伴积极探路，成为连接游戏开发者与厂商的纽带
- 5 月， Airbridge与罗斯基在韩国举办“罗斯基 ×Airbridge 移动游戏行业交流·韩国站”移动游戏行业交流会，联结中、韩两地的行业精英进行深入交流。
- 9 月，Airbridge CEO Roi Nam 受邀参与由广州市商务局主办、广州市游戏行业协会承办、韩国手游协会协办的“文化出海，有服同享”广州数字文化服务交流会，在首尔为中、韩游戏开发者分享有关移动营销与归因分析的最新趋势。
- 10 月，Airbridge 与 Unity 旗下 Supersonic 一起，举办了一场主题为Speed Networking 的活动，集结了来自韩国本土、中国等全球多个市场的开发与发行游戏厂商，进行了深度交流。
此外，Airbridge 的足迹也遍布世界，深入本土市场，帮助开发与发行创造更多好游戏的同时，直接聆听与解决从业者在归因过程中的痛点：
- ChinaJoy
- GDC
- Gamescom
- TGS
除了深入本土，我们也在积极发声，与行业伙伴一同，分享来自第一线的市场洞察：
- AppGrowing 白皮书
- Adjoe 白皮书
## 2025：深度拥抱 AI，更智能、更高效
今年，Airbridge 推出了 AI 驱动的新功能，为您提供更强大的数据衡量能力。我们的增强型 AI 功能可利用仅三天的数据预测 180 天的客户生命周期价值，检测异常安装，并通过 UCA（未归因转化分析）模拟 iOS 归因，为您的应用业务增长带来更智能、更快速、更精准的洞察。
感谢您一路以来对 Airbridge 的支持与信任。让我们携手共创更加辉煌的成功故事，迈向更加卓越的 2025！
---
## Kwai（快手国际版）集成
URL: https://www.airbridge.io/zh/blog/kwai-integration
Language: Chinese
Published: 2024-10-02
通过 Airbridge 与 Kwai for Business（快手出海平台，简称 “Kwai”）的集成，您现在可以精确监测并优化 Kwai 广告活动的表现。这一强强联合，让您在触及 Kwai 覆盖 30 多个国家/地区、每月超 10 亿活跃用户的同时，还能充分利用 Airbridge 的高级分析功能，全面提升广告效果。
[__wf_reserved_inherit]
## **精确监测 Kwai 广告活动**
通过此次集成，您可以：
- 专为 Kwai 广告计划创建和管理追踪链接
- 为 Kwai 配置量身定制的归因规则
- 设置回传功能，实时发送应用内事件和转化数据
这些功能帮助您深入了解 Kwai 各种广告形式的表现，包括：
- 信息流广告
- 短视频广告
- 网红合作
## **增强分析 Kwai 用户群体**
Airbridge 帮助您更好地理解并与 Kwai 的核心用户群体——即 Z 世代和年轻的千禧一代——互动。您可以:
- 追踪用户与您广告内容的互动情况
- 监测基于流行趋势的广告计划的绩效
通过将 Kwai 广告活动集成到 Airbridge，您将获得全面的数据洞察，从而优化目标定位、广告素材策略和预算分配，更高效地与 Kwai 社区的高度活跃用户群体互动。
详细的分步集成指南请参阅 [Airbridge 用户指南](https://help.airbridge.io/zh/guides/kwai-for-business)。若想了解 Airbridge 与 Kwai 联合如何可以推动您的 App 增长，请立即联系 [Airbridge](https://www.airbridge.io/zh/book-a-demo)。
## **关于 Kwai**
Kwai 成立于 2021 年, 总部位于中国北京, 是一个全球性平台, 致力于为广告主提供全面的营销解决方案, 帮助他们以创意方式与 Kwai 社区互动。Kwai 覆盖 30 多个国家/地区, 拥有超过 10 亿月活用户, 帮助您真实且有效地与目标受众建立联系。
Kwai 提供多种社交广告产品, 包括信息流广告、品牌推广和用户生成内容（UGC）合作。您可以从零开始制作本地化视频广告, 并在同一平台上追踪其表现, 从而优化您的广告策略。通过 Kwai, 您可以探索受众社区中的热门话题, 抓住流行趋势, 并与顶级创作者合作, 提升广告效果。
---
## [Deep Link 101] 使用Deferred Deep Linking将未安装应用的用户引导至应用内特定页面
URL: https://www.airbridge.io/zh/blog/deeplink-101-deferred-deep-linking
Language: Chinese
Published: 2022-04-07
[]
想要在用户点击链接并安装应用后，仍能将其引导至应用内的相关页面？如果是，您需要deferred deep linking。
> 如需了解更多关于deep link的详细内容，请查看<[Airbridge Handbook - Deep Links](https://www.airbridge.io/en/library/airbridge-handbook-deeplinks)>指南。
## 什么是Deferred Deep Linking？
Deferred deep linking是一种扩展标准deep link功能的技术。顾名思义，它指的是"延迟的deep link"。由于deep linking无法对未安装应用的用户生效，该技术会等待用户点击链接、安装应用并打开它后，再执行deep link。自行实现并不容易，因为需要一个能够存储和传递deep link的服务器，还需要将用户点击链接时的信息与安装并打开应用的用户信息进行匹配。
在本文中，了解为什么需要deferred deep linking技术以及如何实现它。
## 为什么需要Deferred Deep Linking？
Deep link是一种技术，当移动端用户点击链接时，启动应用并导航至特定页面。换句话说，它是一个指定了通往应用内特定路径的链接。然而，如果点击deep link时应用尚未安装，用户会被重定向到应用商店或遇到错误，链接/路径无法正常工作。这意味着即使用户点击了deep link，也不会被引导至应用内的特定页面。
具体来说，当应用未安装时，可能会出现以下问题：
- 使用URI Scheme deep link时，如果应用未安装，无法导航至应用商店等其他目的地，因此什么都不会发生。
- 对于Universal Links和App Links，无法使用直接跳转到App Store或Google Play的URL，使得引导未安装应用的用户进行安装变得困难。
- 为了应对上述情况，仅通过MMP的tracking link使用deep link仍无法提供完整的用户体验。
由于这些功能限制，如果点击广告后什么都没有发生，或者在经历繁琐的应用安装过程后只显示主页面，即使是被广告吸引的潜在客户也可能立即流失。
**解决这个问题的技术就是deferred deep linking。**
即使对于未安装应用的用户，如果他们看到广告并安装了应用，安装后被引导至应用内的相关内容会感觉很自然。转化的可能性也会提高。这正是deferred deep linking对app marketing至关重要的原因。
## 如何实现Deferred Deep Linking
然而，自行实现deferred deep linking技术可能面临诸多挑战。在Android操作系统层面，可以使用Google Play Install Referrer API来实现，但存在局限性。此外，在iOS上，操作系统层面没有提供实现deferred deep linking的方法，因此需要搭建单独的服务器等复杂环境。
### 自行实现Deferred Deep Linking的困难
首先，**在Android上**，可以使用Google Play Store的Google Play Install Referrer API自行实现deferred deep linking。
然而，这种方法实现起来较为复杂，容易出现数据收集问题，而且只适用于从Google Play下载的情况。
**在iOS上**，无法使用类似的方法，因此搭建服务器成为额外的必要步骤。
### 使用移动归因解决方案轻松实现Deferred Deep Linking
既然deferred deep linking难以自行实现，如何才能高效且完整地实现？答案是使用移动归因工具提供的解决方案。
Deferred deep linking是一种延迟deep link的技术，直到用户点击链接、安装应用并打开它。**需要一个能够存储用户点击信息的服务器，当应用启动时，将点击数据与打开应用的用户进行匹配并传递deep link。**这正是归因工具所提供的。
通过Airbridge实现deferred deep linking的流程：
- 用户点击Airbridge tracking link -> 点击信息存储在服务器上 -> 用户安装应用 -> 首次启动集成SDK的应用时，服务器检查链接信息 -> 服务器将deep link地址传递给SDK -> 打开应用内的特定页面
SDK指南：
- Airbridge [Android SDK](https://developers.airbridge.io/docs/android-sdk)指南
- Airbridge [iOS SDK](https://developers.airbridge.io/docs/ios-sdk)指南
### Airbridge Tracking Link让Deferred Deep Linking轻松应用
使用Airbridge，当您将tracking link设置为deep link时，deferred deep linking功能默认启用。
[]
## 为了顺利开展App Marketing，建议使用MMP的Tracking Link
MMP的tracking link让您无需进行复杂的按操作系统实现，即可设计理想的用户旅程。
Deferred deep linking是app marketing必不可少的技术。使用Airbridge tracking link高效实现吧。
> 查看Deep Link 101系列的其他文章：
👉 [Deep Link 101：面向营销人员和开发者的Deep Link入门指南](https://www.airbridge.io/zh/blog/deeplink-101-for-marketers-and-developers)
👉 [Deep Link 101：为什么在iOS Safari中进行Deep Linking时会出现警告？](https://www.airbridge.io/zh/blog/deeplink-101-ios-safari-alert)
# Vietnamese Articles
---
## AppsFlyer vs Airbridge vs Adjust vs Branch: MMP nào phù hợp nhất cho app subscription trong năm 2026?
URL: https://www.airbridge.io/vi/blog/appsflyer-vs-airbridge-vs-adjust-vs-branch-best-mmp-for-subscription-app-in-2026
Language: Vietnamese
Published: 2026-04-21
Author: Luke
Trong năm 2026, việc chọn đúng MMP cho một app subscription không còn chỉ xoay quanh install và CPI. Các team tăng trưởng muốn hiểu rõ AppsFlyer, Airbridge, Adjust và Branch thực sự khác nhau ở đâu, nền tảng nào hỗ trợ tốt nhất cho user acquisition, retention và phân tích doanh thu, pricing sẽ thay đổi như thế nào khi scale, và mỗi nền tảng xử lý attribution trong bối cảnh ưu tiên quyền riêng tư dưới ATT và SKAdNetwork ra sao.
Bài so sánh này sẽ đi sâu vào độ chính xác attribution, cấu trúc chi phí, khả năng chống fraud và đo lường web-to-app, với mục tiêu giúp các team tìm ra một nguồn dữ liệu thống nhất cho install, in-app event và doanh thu subscription. Nếu bạn đang cần đánh giá MMP phù hợp nhất cho app subscription trong năm 2026, đây là cách để giúp bạn đưa ra quyết định tự tin và dựa trên dữ liệu.
## Key Takeaways
- App subscription cần một MMP có thể kết nối install với trial, renewals, retention và LTV, chứ không chỉ dừng ở click và CPI.
- Các mô hình attribution truyền thống không còn phù hợp với app subscription vì doanh thu đến muộn, lặp lại theo chu kỳ và phụ thuộc vào phân tích theo cohort.
- AppsFlyer và Adjust mạnh về UA ở quy mô lớn, Branch nổi bật ở web-to-app linking, nhưng cả ba đều gặp nhiều vấn đề khi đi sâu vào phân tích doanh thu subscription.
- Airbridge được xây dựng với tư duy subscription-first, xem trial, chuyển đổi, renewals và churn là những tín hiệu attribution cốt lõi.
- Trong bối cảnh ATT và SKAdNetwork, yếu tố tạo ra khác biệt thực sự nằm ở việc một MMP có thể map các tín hiệu an toàn về quyền riêng tư tới kết quả doanh thu tốt đến đâu.
- Nhiều đội ngũ đang thay thế các MMP cũ bằng Airbridge để có một nguồn dữ liệu thống nhất cho acquisition, retention và doanh thu — mà không phải đối mặt với dashboard rời rạc hay tình trạng upsell tính năng.
## Vì sao app subscription cần một MMP (Mobile Measurement Partner) phù hợp
Một **MMP (Mobile Measurement Partner)** là hệ thống giúp theo dõi nguồn hay kênh nào tạo ra install và các sự kiện sau cài đặt, sau đó kết nối chúng với doanh thu, retention và hiệu suất chiến dịch trên nhiều kênh.
Với app subscription, vai trò này vượt xa install, click và CPI. Một MMP tốt cần cho thấy kênh nào mang về trial, chiến dịch nào thực sự chuyển đổi thành người dùng trả phí theo thời gian, và ở đâu LTV thực sự vượt chi phí. Trong mô hình này, install chỉ là điểm khởi đầu — còn retention, renewals và doanh thu mới là các chỉ số quyết định tăng trưởng.
### Vì sao app subscription đang dần “phá vỡ” các mô hình attribution truyền thống
Có thể bạn từng nghe đến nhiều MMP hoạt động rất tốt trong các ngành như gaming hoặc e-commerce. Nhưng liệu chúng có thực sự phù hợp với app subscription không?
Lý do các mô hình attribution dùng cho gaming hoặc e-commerce không hiệu quả với app subscription là vì:
- Chuyển đổi thường diễn ra **sau vài ngày hoặc vài tuần kể từ khi cài đặt**
- Doanh thu là **doanh thu định kỳ**, không phải giao dịch một lần
- Hành trình **web-to-app** rất phổ biến (landing page → app → trial)
- ROI chiến dịch phụ thuộc vào **churn và renewal**, chứ không phải doanh thu ngày 0
Trong khi đó, nhiều setup MMP cũ vẫn tối ưu xoay quanh last-click installs, cửa sổ attribution ngắn hoặc số lượng event không gắn với bối cảnh doanh thu. Cách này chỉ phù hợp khi mục tiêu là volume UA. Nó không hiệu quả với tăng trưởng subscription. Xét về mặt vận hành, các đội subscription cần:
- Attribution window linh hoạt
- Khả năng mapping chính xác các subscription event
- Visibility về LTV và retention ở cấp cohort
### 3 sai lầm phổ biến khi chọn MMP
Dù đã nhận thức được điều này, nhiều đội vẫn gặp khó khăn khi chọn một MMP phù hợp. Sau khi làm việc với các app subscription đang scale, có 3 sai lầm lặp đi lặp lại rất rõ:
#### 1\. Chọn theo độ phổ biến trên thị trường, không phải theo mức độ phù hợp
Những MMP tên tuổi lớn thường được cân nhắc hơn. Nhưng sau đó, đội ngũ mới nhận ra họ đang trả tiền cho những tính năng được thiết kế cho UA quy mô gaming, chứ không phải cho analytics của app subscription.
#### 2\. Chỉ tối ưu cho UA
Attribution cho install có thể rất “sạch”, nhưng attribution cho renewal, churn và doanh thu lại bị phân mảnh giữa nhiều công cụ khác nhau.
#### 3\. Chấp nhận làm việc với dashboard rời rạc
Web analytics ở một nơi. Báo cáo SKAN ở nơi khác. Dữ liệu subscription ở một hệ thống khác nữa. Kết quả là tốc độ ra quyết định chậm lại, và mức độ tin tưởng vào dữ liệu cũng giảm xuống.
> Ở giai đoạn này, câu hỏi không còn là “MMP nào theo dõi install tốt nhất?” mà sẽ trở thành:
> “**MMP nào cho bạn một nguồn dữ liệu thống nhất cho tăng trưởng?**”
## AppsFlyer vs Airbridge vs Adjust vs Branch: So sánh về tính năng, giá và mức độ phù hợp
Đến đây, bạn đã hiểu vì sao MMP lại quan trọng. Bước tiếp theo là chọn một nền tảng phù hợp — và cũng chính từ đây mọi thứ bắt đầu trở nên rối hơn. Chỉ cần bắt đầu tìm hiểu các giải pháp MMP, bạn sẽ nhanh chóng thấy hàng loạt lựa chọn khác nhau.
**Tại thị trường Mỹ, có thể xem AppsFlyer, Airbridge, Adjust và Branch là nhóm 4 MMP toàn cầu nổi bật nhất. Cả bốn đều giải quyết bài toán attribution.** Nhưng câu hỏi thực sự mà nhiều đội ngũ đặt ra là: “Đâu mới là MMP phù hợp nhất với chúng tôi?”
Khác biệt không nằm ở việc nền tảng nào có attribution, mà nằm ở độ sâu dữ liệu, cấu trúc giá và khả năng hỗ trợ mô hình kinh tế của app subscription.
Dưới đây là phần so sánh thực tế dựa trên cách các đội ngũ thật sự sử dụng những công cụ này, chứ không chỉ dựa trên cách chúng được marketing.
| Hạng mục | AppsFlyer | Airbridge | Adjust | Branch |
| --- | --- | --- | --- | --- |
| Điểm mạnh cốt lõi | Attribution mạnh cho UA quy mô lớn | Attribution và analytics tập trung cho subscription | Theo dõi UA cấp enterprise | Deep linking và web-to-app |
| Phù hợp nhất với | Các chiến dịch tập trung vào install với volume lớn | App subscription và app AI | App enterprise toàn cầu | App ưu tiên nội dung và web-first |
| Theo dõi doanh thu subscription | Có hỗ trợ, nhưng thường dưới dạng add-on | Tích hợp sẵn và linh hoạt | Có hỗ trợ, nhưng setup phức tạp | Độ chi tiết còn hạn chế |
| Attribution web-to-app | Có hỗ trợ qua add-on | Tích hợp sẵn, thống nhất | Có hỗ trợ | Mạnh, nhưng thiên về web |
| Báo cáo SKAdNetwork | Hoàn thiện, nhưng còn phân mảnh | Tập trung, dễ theo dõi | Mạnh, nhưng UI nặng | Chỉ tập trung một phần |
| Phòng tránh gian lận | Nâng cao, nhưng nằm ở gói cao | Được tích hợp mặc định | Kiểm soát mạnh ở cấp enterprise | Cơ bản |
| Độ phức tạp của dashboard | Cao | Thấp đến trung bình | Cao | Trung bình |
| Độ minh bạch về giá | Giá theo tầng, dựa trên event | Minh bạch, dựa trên mức sử dụng | Phụ thuộc nhiều vào hợp đồng | Biến động |
| Mức độ phù hợp để scale app subscription | Trung bình | Cao | Trung bình | Thấp đến trung bình |
Cùng phân tích sâu hơn để thấy rõ sự khác biệt:
### Độ chính xác attribution và độ chi tiết của dữ liệu
Cả bốn nền tảng đều khẳng định mình có attribution chính xác. Khác biệt thực sự bắt đầu lộ rõ **sau Day 7.**
- **Appsflyer & Adjust:** Tối ưu ở attribution cấp install và có độ phủ network lớn. Tuy nhiên, độ chi tiết dữ liệu thường đi kèm với chi phí bổ sung hoặc yêu cầu setup phức tạp hơn.
- **Branch:** Rất mạnh trong việc kết nối hành vi người dùng từ web, email và nội dung sang app. Nhưng lại kém đáng tin cậy hơn khi đi sâu vào mô hình doanh thu subscription.
- **Airbridge:** Được thiết kế để kết nối liền mạch từ install → trial → subscription → renewal mà không cần ghép nhiều báo cáo lại với nhau.
Xét ở góc độ vận hành, Airbridge giúp thu hẹp khoảng cách giữa **dữ liệu attribution thô** và **quyết định dựa trên doanh thu thực tế**.
### Analytics cho subscription và doanh thu
Đây là nơi các app subscription bắt đầu cảm nhận rõ các vấn đề.
Với phần lớn các MMP, dữ liệu doanh thu mặc dù có tồn tại trong hệ thống — nhưng hiếm khi đủ để mang lại nhiều insights. Subscription event vẫn được ghi nhận, nhưng thường không được gắn chuẩn với cohort. Dữ liệu renewal thường bị đẩy sang BI tools vì đó mới là nơi mà marketers cho rằng họ có thể phân tích sâu. Tỷ lệ chuyển từ trial sang trả phí thậm chí còn không được xem trong dashboard của MMP.
[**Airbridge**](https://www.airbridge.io/solutions/subscription) **đi theo một hướng khác. Các sự kiện subscription được xem là dữ liệu cốt lõi, không phải tín hiệu phụ.** Doanh thu, retention và LTV được đặt cạnh các chỉ số UA, giúp đội ngũ nhìn thấy chất lượng acquisition mà không cần export dữ liệu hay đối soát giữa nhiều dashboard. Những câu hỏi cốt lõi như user nào chuyển đổi, user nào gia hạn, hoặc user nào churn đều có thể được trả lời ngay bên trong MMP.
### Mô hình giá và những chi phí ẩn
Pricing hiếm khi được đề cập đầy đủ, nhưng lại là yếu tố quyết định liệu một nền tảng có được sử dụng hay không.
**Appsflyer / Adjust**
- Giá thường tính theo event hoặc volume
- Các tính năng nâng cao thường bị khóa ở gói cao hơn
- Chi phí tăng rất nhanh khi khối lượng subscription event ngày càng lớn
**Branch**
- Mức giá đầu vào khá cạnh tranh
- Nhưng chi phí tăng lên khi độ phức tạp của attribution tăng theo
**Airbridge**
- Mức giá dễ dự đoán hơn, bám sát mức sử dụng thực tế
- Ít xảy ra những thời điểm “phải nâng gói” khi đội ngũ mở rộng
- Giảm ma sát khi bổ sung web, SKAN hoặc subscription events
Với CFO và growth leader, khả năng dự đoán chi phí quan trọng không kém gì chất lượng tính năng của nền tảng.
### Hệ sinh thái tích hợp và mức độ phù hợp với workflow
Cả bốn nền tảng đều tích hợp với các ad network và công cụ analytics lớn, nên nhìn bề ngoài có vẻ tương đương nhau. Sự khác biệt thực sự nằm ở cách các đội ngũ tận dụng dữ liệu.
AppsFlyer và Adjust được xây dựng dựa trên giả định rằng attribution chỉ là lớp dữ liệu đầu vào cho nhiều hệ thống phía sau. Chúng hoạt động tốt với vai trò nhà cung cấp dữ liệu, nhưng việc phân tích và ra quyết định thường diễn ra ở nơi khác. Branch nghiêng mạnh hơn về linking và routing user journey, còn analytics chỉ đóng vai trò hỗ trợ.
[**Airbridge**](https://www.airbridge.io/solutions/partner-network) được thiết kế theo hướng khác. Nền tảng này hướng đến việc trở thành **lớp vận hành trung tâm**, chứ không chỉ là một nguồn dữ liệu để xuất sang công cụ khác. Attribution, subscription metrics và performance insights cùng tồn tại trong một nơi, từ đó giảm số lần handoff và thu hẹp khoảng trống trong báo cáo.
## SKAdNetwork, ATT và attribution trên iOS: MMP nào làm tốt nhất trong năm 2026?
Trong bối cảnh **ATT** và **SKAdNetwork (SKAN)**, vai trò của một MMP đã thay đổi: không còn xoay quanh việc theo dõi người dùng ở cấp độ cá nhân, mà chuyển sang **tổng hợp dữ liệu an toàn cho quyền riêng tư, mô hình hóa và diễn giải tín hiệu** — nhưng vẫn phải đảm bảo quá trình ra quyết định không bị gián đoạn.
ATT đã loại bỏ khả năng theo dõi người dùng một cách deterministic trên iOS. Thay vào đó, SKAN cung cấp dữ liệu install ở dạng tổng hợp, có độ trễ, với cửa sổ conversion value bị giới hạn và không còn user-level identifier.
Về lý thuyết, tất cả MMP đều hỗ trợ SKAN. Tuy nhiên, trải nghiệm sử dụng thực tế rất khác nhau.
### Độ chi tiết báo cáo SKAdNetwork theo từng nền tảng
**Appsflyer**
- Hỗ trợ SKAN mạnh và có tài liệu đầy đủ
- Dữ liệu thường bị chia tách trên nhiều dashboard khác nhau
- Cần đội ngũ có kinh nghiệm để phân tích chính xác
**Adjust**
- Pipeline SKAN ở cấp enterprise
- Rất mạnh nhưng UI khá nặng
- Phù hợp hơn với các đội lớn có analyst chuyên trách
**Branch**
- Có hỗ trợ SKAN
- Chưa tối ưu bằng cho mô hình chuyển đổi của app subscription
- Trọng tâm vẫn nằm ở linking hơn là insight về doanh thu
**Airbridge**
- Dữ liệu SKAN được hiển thị cạnh các chỉ số non-SKAN
- Conversion value được map trực tiếp với subscription event
- Giảm nhu cầu đối soát thủ công giữa nhiều dashboard
### Dữ liệu vẫn thất thoát — và nền tảng nào xử lý tốt hơn?
Phần lớn thất thoát dữ liệu SKAN xảy ra **từ rất lâu trước khi báo cáo được động đến**.
Các điểm gãy quen thuộc thường là: conversion value được thiết kế chưa hiệu quả, postback window thiếu linh hoạt, và subscription event không được kết nối chặt với logic attribution. Khi những lựa chọn này đã được cố định, đội ngũ sẽ mất visibility đúng ở nơi quan trọng nhất — **sau khi người dùng đã cài app**.
Những setup cũ thường khiến vấn đề này nghiêm trọng hơn. Conversion value vẫn xoay quanh install. Thứ tự ưu tiên event giữ nguyên, thiếu linh hoạt. Việc tối ưu chậm lại vì mỗi thay đổi đều trở nên tốn kém hoặc rủi ro.
[**Airbridge tiếp cận SKAN theo hướng khác**](https://www.airbridge.io/product/ios-skan). Schema của conversion value có thể được cập nhật nhanh hơn. Subscription event được mapping có chủ đích, chứ không bị xem như phần bổ sung về sau. Tín hiệu SKAN được gắn với **kết quả doanh thu**, chứ không chỉ dừng ở xác nhận install.
### Phòng tránh gian lận trong kỷ nguyên ưu tiên quyền riêng tư
**Appsflyer & Adjust**
- Có hệ thống chống gian lận đầy đủ
- Nhưng các lớp bảo vệ nâng cao thường nằm ở gói trả phí cao hơn
**Branch**
- Có các lớp bảo vệ cơ bản
- Không phải nền tảng được xây với fraud prevention là trọng tâm
**Airbridge**
- Tính năng chống gian lận được tích hợp mặc định
- Tập trung vào invalid traffic và các bất thường trong attribution
- Mang lại mức độ bảo vệ thực tế mà không đi kèm rào cản về chi phí
Với app subscription, việc bảo vệ chất lượng trial ngay từ giai đoạn đầu thường quan trọng hơn cả việc chỉ chặn fraud ở mức install thô.
## MMP nào phù hợp với giai đoạn tăng trưởng của bạn?
MMP phù hợp không phụ thuộc quá nhiều vào danh sách tính năng, mà phụ thuộc vào việc doanh nghiệp subscription của bạn đang ở đâu trong hành trình tăng trưởng — và cần thích nghi nhanh đến mức nào. Dưới đây là hướng dẫn thực tế để lựa chọn theo từng giai đoạn.
### App subscription giai đoạn đầu (đang tìm product-market fit)
Ở giai đoạn đầu, mục tiêu là tốc độ và sự rõ ràng, chứ chưa phải sự phức tạp. Các đội ngũ cần triển khai tracking thật nhanh, kiểm soát chi phí chặt chẽ và hiểu được liệu người dùng có đang đi qua những bước đầu tiên của funnel hay không.
Tốc độ setup rất quan trọng vì các thử nghiệm có thể thay đổi theo từng tuần. Mức giá hợp lý cũng quan trọng vì volume vẫn chưa ổn định. Bên cạnh đó, khả năng nhìn rõ install và các conversion event ban đầu cũng rất cần thiết, vì đây chính là nơi product-market fit bắt đầu lộ diện.
**Best fit**
- **Branch** phù hợp nếu tăng trưởng của bạn đi từ web và phụ thuộc nhiều vào content.
**Airbridge** phù hợp với những đội muốn làm đúng bài toán subscription tracking ngay từ ngày đầu tiên, mà không phải đổi nền tảng về sau.
### Đội performance đang scale (tập trung vào UA + retention)
Ở giai đoạn này, volume install không còn là vấn đề chính nữa. Trọng tâm chuyển sang **chất lượng** — ai là người chuyển đổi, ai ở lại, và ai thực sự trả tiền theo thời gian.
Các đội ngũ cần attribution rõ ràng từ **trial sang paid**, cần **LTV theo cohort** đủ tin cậy, và cần báo cáo **SKAN** không bị tách ra thành một workflow riêng.
Quan trọng không kém, họ cần **ít dashboard hơn**. Khi dữ liệu performance, retention và doanh thu bị chia nhỏ giữa nhiều công cụ, việc tối ưu sẽ chậm lại và niềm tin vào dữ liệu cũng giảm xuống.
**Best fit**
- **Airbridge** nổi bật ở giai đoạn này nhờ hợp nhất UA, retention, doanh thu và SKAN trong một góc nhìn thống nhất, giúp các đội đang scale tối ưu theo **LTV dài hạn**, chứ không chỉ theo install.
- **AppsFlyer** và **Adjust** vẫn hoạt động tốt nếu mục tiêu là scale UA thuần túy, nhưng thường cần thêm công cụ xung quanh để trả lời các câu hỏi liên quan đến subscription.
### Đội ngũ lớn đang dần thay thế MMP mà họ đã luôn sử dụng
Với các đội ngũ đã trưởng thành, vấn đề không còn là có dữ liệu hay không. Vấn đề là có còn tin vào dữ liệu đó hay không.
Ở giai đoạn này, mức độ tin cậy của dữ liệu quan trọng hơn độ rộng tính năng. Khả năng dự đoán chi phí trở thành mối quan tâm ở cấp quản lý. Attribution phải đủ vững trong bối cảnh privacy mà không cần liên tục làm lại hệ thống. Hiệu quả vận hành cũng rất quan trọng, vì mỗi quy trình phát sinh thêm đều tạo ra áp lực nhân lên giữa các đội.
Các MMP cũ của họ thường mang lại nhiều vấn đề hơn — mặc dù rất hiệu quả, nhưng tốn kém để vận hành và khó thích nghi hơn. Quyết định thay thế thường đến từ nhu cầu về sự rõ ràng, ổn định và một hệ thống có thể scale mà không tạo thêm ma sát.
**Best fit**
- **Airbridge** phù hợp khi đội ngũ muốn có một lớp attribution hiệu quả mà không phải gánh chi phí lớn.
- **Adjust** phù hợp nếu bạn có yêu cầu compliance toàn cầu phức tạp và đội analyst đông.
## Kết luận: Chọn một MMP làm nguồn dữ liệu thống nhất duy nhất
Tăng trưởng của app subscription sẽ bắt đầu gãy khi dữ liệu nằm rải rác ở quá nhiều nơi.
Khi attribution cho install nằm ở một công cụ, doanh thu subscription ở một công cụ khác, web analytics ở nơi khác nữa, còn báo cáo SKAN thì đứng riêng biệt, các đội ngũ phát triển người dùng sẽ mất đi cả tốc độ triển khai dự án lẫn niềm tin vào dữ liệu. Quyết định trở nên bị động. Việc thử nghiệm cũng chậm lại.
AppsFlyer, Adjust và Branch vẫn là những nền tảng mạnh cho những use case cụ thể. Nhưng nhiều app subscription thường vượt qua mô hình attribution tập trung vào install nhanh hơn dự kiến. Điều họ cần tiếp theo là sự rõ ràng — xuyên suốt từ acquisition đến retention và doanh thu.
Đây chính là khi Airbridge đang ngày càng trở nên phù hợp hơn:
- Subscription event được xem là dữ liệu cốt lõi
- Hành trình giữa web, app và SKAN được kết nối trong một góc nhìn thống nhất
- Đội ngũ không còn mất thời gian đối soát số liệu, mà có thể tập trung tối ưu tăng trưởng
Nếu MMP hiện tại của bạn giống một gánh nặng hơn là một cỗ máy tăng trưởng, có lẽ đã đến lúc nhìn lại toàn bộ growth stack. Và đừng chỉ nghe lời hứa từ phía chúng tôi!
👉 [Cách Playio nâng D30 Retention lên 30% và giảm 40% Global UA CPA với Airbridge](https://abr.ge/jqz1p7)
👉 [Cách Rooster Games đạt mức tăng 15% ROAS nhờ đo lường cross-platform của Airbridge](https://abr.ge/g1y748)
👉 [Nightly giảm 18% CPA với Simulated iOS Attribution và vươn lên top 3 App Store tại Nhật Bản](https://abr.ge/ubscdbp)
---
## 4 giải pháp thay thế AppsFlyer tốt nhất năm 2026: Phân tích chuyên sâu về chi phí và độ chính xác attribution
URL: https://www.airbridge.io/vi/blog/4-best-appsflyer-alternatives-for-2026-a-deep-dive-into-costs-attribution-accuracy
Language: Vietnamese
Published: 2026-04-20
Author: Luke
Mobile measurement đang đứng trước một ngã rẽ lớn. Sau khi Firebase Dynamic Links chính thức ngừng hoạt động vào ngày 25 tháng 8 năm 2025, cùng với sự thất vọng ngày càng gia tăng trước việc nhiều tính năng bị khóa sau các gói trả phí của các MMP truyền thống, các mobile marketer buộc phải nhìn lại toàn bộ chiến lược đo lường của mình.
**Nếu bạn đang tìm một giải pháp thay thế AppsFlyer, thì bạn không phải người duy nhất — và chính sự xáo trộn này có thể là cơ hội để đội ngũ của bạn chuyển sang một lựa chọn tốt hơn.**
**Key Takeaways**
- Các đội ngũ tìm kiếm giải pháp thay thế AppsFlyer vì chi phí cao, tính năng bị giới hạn theo các gói giá và vận hành phức tạp, đặc biệt khi event volume và subscription tăng lên.
- Nhu cầu với một MMP hiệu quả không chỉ dừng ở install, mà còn phải bao gồm attribution toàn funnel, deep linking, doanh thu subscription và LTV — mà không yêu cầu trả thêm phí cho các tính năng cốt lõi.
- AppsFlyer và Adjust mang lại mức độ kiểm soát ở cấp enterprise, nhưng thường quá nặng về quy trình vận hành và tốn kém với các đội ngũ tinh gọn hoặc cần tốc độ triển khai cao.
- Branch mạnh về deep linking và hành trình web-to-app, trong khi Tenjin phù hợp với các đội giai đoạn đầu cần giải pháp đơn giản và tiết kiệm chi phí hơn.
- Airbridge định vị là một lựa chọn thay thế hiệu suất cao, cung cấp độ chính xác attribution tương đương AppsFlyer nhưng với mức giá minh bạch hơn và trải nghiệm sử dụng dễ hơn.
- Trong năm 2026, giải pháp thay thế AppsFlyer tốt nhất sẽ là MMP mang lại attribution chính xác, khả năng sẵn sàng cho privacy và time-to-value nhanh, mà không khóa những tính năng cốt lõi sau paywall.
## Vì sao các đội ngũ bắt đầu tìm giải pháp thay thế AppsFlyer
Một **Mobile Measurement Partner (MMP)** như AppsFlyer giúp các doanh nghiệp sở hữu app theo dõi nguồn người dùng, phân bổ install và in-app event, đồng thời đo lường hiệu quả marketing trên nhiều kênh.
Có thể bạn đã từng nghe đến AppsFlyer và tự hỏi liệu đây có thực sự là lựa chọn phù hợp cho app của mình không. Bạn có thể đã đọc review, tìm hiểu pricing, thậm chí chủ động tìm các giải pháp thay thế để xem liệu có công cụ nào phù hợp hơn với ngân sách, cấu trúc kỹ thuật và mục tiêu tăng trưởng của đội ngũ hay không.
Các đội bắt đầu tìm một giải pháp thay thế AppsFlyer khi họ vẫn muốn giữ **độ chính xác attribution ở mức cốt lõi**, nhưng với **chi phí thấp hơn, thiết lập đơn giản hơn**, hoặc một sản phẩm phù hợp hơn với giai đoạn tăng trưởng và mix chiến dịch hiện tại của họ.
### Độ phức tạp trong vận hành đối với các đội ngũ tinh gọn
AppsFlyer hoạt động hiệu quả với những đội có sẵn nguồn lực analytics hoặc vận hành chuyên biệt. Nhưng với các đội nhỏ hơn, mức độ ma sát sẽ rõ rệt hơn. Theo G2, phản hồi phổ biến từ các performance marketer có 3–5 năm kinh nghiệm cho thấy:
- Việc triển khai cần phối hợp khá kỹ với đội developer
- Quá trình duy trì về sau tạo thêm khối lượng vận hành
- Dashboard mạnh về tính năng nhưng khá nặng nếu chỉ cần ra quyết định nhanh
[Reviews of AppsFlyer. Source: G2]
### Vấn đề về chi phí
Nhiều đội ngũ ngày càng cảm thấy khó chịu với các MMP như AppsFlyer khi “mọi thứ đều bị upsell” và các tính năng quan trọng bị khóa sau các gói trả phí cao hơn, theo [Metacto](https://www.metacto.com/blogs/the-complete-guide-to-appsflyer-costs-setup-integration-maintenance). Việc khóa tính năng tạo ra những rào cản không cần thiết. Khi các khả năng như **attribution, phân tích nâng cao hay deep linking** đều nằm sau paywall, bạn buộc phải lựa chọn giữa **ngân sách và độ chính xác**.
Mô hình này khiến quyết định không còn dựa trên nhu cầu kinh doanh thực tế, mà bị chi phối bởi việc phải trả thêm tiền cho những tính năng đáng lẽ nên là tiêu chuẩn của một nền tảng đo lường toàn diện.
### Không phù hợp với nhu cầu chiến dịch hoặc sản phẩm cụ thể
AppsFlyer được thiết kế để phục vụ nhiều use case rộng. Tuy nhiên, [một số đội cần **độ linh hoạt cao hơn**](https://www.g2.com/products/appsflyer/reviews?qs=pros-and-cons). Ví dụ:
- Các chiến dịch tập trung nhiều vào **web-to-app hoặc deep linking**
- Sản phẩm ưu tiên **chỉ số vòng đời subscription** hơn là số lượng install
- Đội ngũ cần **insight nhanh** thay vì tùy chỉnh sâu phức tạp
Khi nền tảng không phản ánh đúng cách tăng trưởng thực sự diễn ra, các đội sẽ bắt đầu nghiêm túc so sánh và tìm kiếm các giải pháp thay thế AppsFlyer.
## Nên tìm gì ở một giải pháp thay thế AppsFlyer
Khi đánh giá các lựa chọn thay thế, hãy tập trung vào những năng lực giải quyết đúng các hạn chế cốt lõi khiến bạn bắt đầu tìm kiếm ngay từ đầu. Dưới đây là những yếu tố thực sự quan trọng:
### Attribution toàn diện mà không bị giới hạn bởi gói giá
Giải pháp thay thế bạn chọn nên cung cấp **attribution toàn funnel như một tính năng tiêu chuẩn**, không phải một tiện ích bổ sung chỉ có ở gói cao cấp. Hãy tìm những nền tảng có thể theo dõi chuyển đổi trên mọi nền tảng và mọi điểm chạm mà không giới hạn quyền truy cập theo bậc giá.
Mục tiêu ở đây là có được **bức tranh marketing đầy đủ**, chứ không phải các insight rời rạc bị giới hạn bởi mức phí bạn sẵn sàng chi trả.
### **Khả năng deep linking mạnh mẽ**
Deep linking không chỉ là đưa người dùng đến một trang cụ thể trong app, mà là tạo ra **trải nghiệm liền mạch trên toàn bộ kênh marketing**. Giải pháp thay thế bạn chọn cần xử lý được nhiều loại deep link khác nhau như **URI Schemes, Universal Links và App Links**, đồng thời cung cấp attribution đáng tin cậy cho người dùng đến từ các đường dẫn đó.
Ngoài ra, nền tảng cũng cần xử lý tốt các tình huống như **deferred deep linking** cho người dùng chưa cài app, để không đánh mất cơ hội chuyển đổi chỉ vì giới hạn kỹ thuật.
### Tích hợp và quan hệ đối tác với các nền tảng lớn
Khả năng tích hợp mượt mà với các nền tảng quảng cáo là điều **không thể thiếu**. Giải pháp thay thế bạn chọn phải duy trì **quan hệ đối tác chính thức** với Meta, Google Ads, TikTok và các nền tảng lớn khác mà bạn đang dùng cho user acquisition.
Những quan hệ đối tác này giúp dữ liệu luân chuyển trơn tru giữa MMP và nền tảng quảng cáo mà không cần workaround thủ công hay đối mặt với sai lệch dữ liệu.
### Hỗ trợ migration và tài liệu đầy đủ
Việc chuyển đổi nền tảng đo lường vốn khá phức tạp, nên hãy tìm những giải pháp cung cấp **hỗ trợ migration chuyên biệt**, tài liệu toàn diện và lộ trình triển khai rõ ràng.
Việc chuyển đổi nên giúp bạn **nâng cấp năng lực đo lường**, chứ không làm gián đoạn hoạt động marketing.
### Gói giá minh bạch
Nếu đã mệt mỏi với các chi phí ẩn, **sự minh bạch về giá** trở thành yếu tố rất quan trọng. Giải pháp thay thế nên cho biết rõ **mỗi mức giá bao gồm những gì**, không có phí ẩn hay upsell bất ngờ cho các tính năng thiết yếu.
Bạn cần có khả năng lập ngân sách dựa trên **nhu cầu kinh doanh thực tế**, chứ không phải dựa trên việc bị giới hạn quyền dùng tính năng.
### Đáp ứng yêu cầu về privacy
Một MMP tốt phải có khả năng xử lý các yêu cầu về mặt privacy của nền tảng. Tối thiểu, bạn nên tìm kiếm những giải pháp có:
- Hỗ trợ đầy đủ cho SKAdNetwork và Privacy Sandbox
- Framework đo lường dựa trên consent
- Tùy chọn attribution phía server và modeled attribution
## Các giải pháp thay thế AppsFlyer hàng đầu: So sánh tính năng và chi phí
Mặc dù trên thị trường có khá nhiều MMP, quá trình đánh giá thực tế thường nhanh chóng thu hẹp lại chỉ còn một vài cái tên. Dựa trên dữ liệu thị trường và hành vi của người mua, các đội ngũ thường đặt AppsFlyer lên bàn cân cùng **Adjust, Branch, Tenjin và** [**Airbridge**](https://abr.ge/d4zv0x). Dưới đây là cách những lựa chọn hàng đầu này được so sánh với nhau:
| Nền tảng | Điểm mạnh chính | Hạn chế phổ biến | Mô hình giá | Phù hợp với |
| --- | --- | --- | --- | --- |
| AppsFlyer | Attribution cấp enterprise, phân tích chuyên sâu, khả năng chống gian lận mạnh | Chi phí cao khi scale, dư thừa tính năng với nhiều teams, learning curve cao | Giá theo volume và tính năng | App quy mô lớn có đội analytics riêng |
| Airbridge | Độ chính xác attribution tương đương, analytics linh hoạt, deep linking mạnh, giá minh bạch | Nhận diện thương hiệu đang phát triển so với các nền tảng khác | Giá linh hoạt, minh bạch | App subscription, dịch vụ AI, team mid-market |
| Adjust | Attribution mạnh, hiện diện toàn cầu, báo cáo đáng tin cậy | Áp lực chi phí tương tự AppsFlyer, UI nhiều tính năng | Giá enterprise theo volume | App global với UA đã trưởng thành |
| Branch | Deep linking hàng đầu, mạnh về web-to-app flow | Độ sâu attribution kém linh hoạt trong một số use case | Giá theo package | Đội ưu tiên deep linking hơn attribution toàn funnel |
| Tenjin | Setup đơn giản, dashboard tập trung UA, tiết kiệm chi phí | Khả năng tùy chỉnh hạn chế, độ sâu analytics thấp hơn | Giá thấp, minh bạch | Indie studio và team giai đoạn đầu |
### So sánh chi phí của AppsFlyer trong các tình huống thực tế
Chi phí của AppsFlyer thường vẫn được xem là hợp lý ở giai đoạn đầu. Nhưng áp lực bắt đầu xuất hiện khi quy mô tăng lên. Trong thực tế, các đội ngũ thường cảm nhận rõ điều này khi:
- Khối lượng event tăng mạnh do mô hình subscription hoặc tần suất tương tác trong sản phẩm cao
- Nhiều kênh chạy song song, làm tăng tải attribution
- Các tính năng nâng cao không còn là “nice to have” mà trở thành yêu cầu bắt buộc
Đã mệt mỏi vì phải trả tiền cho những tính năng đáng lẽ nên là tiêu chuẩn? [Hãy xem cách Airbridge mang đến attribution cấp enterprise mà không đi kèm bất cứ rào cản về gói giá nào.](https://abr.ge/757220)
### Tính dễ triển khai là điểm khiến nhiều đội ngũ đổi hướng
AppsFlyer và Adjust được thiết kế để mang lại mức độ kiểm soát tối đa. Nhưng kiểu kiểm soát đó cũng đồng nghĩa với việc cần có nguồn lực chuyên biệt. Các team nhỏ hơn và những công ty AI vận hành nhanh thường ưu tiên những nền tảng:
- Cần ít sự tham gia liên tục từ đội developer hơn
- Có dashboard dễ sử dụng ngay từ đầu
- Rút ngắn khoảng cách giữa việc thu thập dữ liệu và ra quyết định
### Vì sao Branch và Tenjin cũng được đưa vào bàn cân
Branch thường được cân nhắc khi deep linking và hành trình web-to-app đóng vai trò trung tâm trong tăng trưởng. Điểm mạnh của nền tảng này nằm ở khả năng duy trì tính liên tục của user journey, hơn là độ sâu attribution trên toàn funnel.
Tenjin thường xuất hiện trong các so sánh khi yếu tố chi phí được đặt lên hàng đầu và nhu cầu chủ yếu là báo cáo user acquisition. Nền tảng này phù hợp với các setup đơn giản hơn, nhưng có thể trở nên hạn chế khi nhu cầu đo lường tăng lên.
### Airbridge định vị khác ở đâu
[Airbridge](https://abr.ge/757220) thường xuất hiện ở giai đoạn sau của quá trình đánh giá — **khi các đội ngũ nhận ra rằng họ không cần phải đánh đổi chất lượng attribution chỉ để giảm chi phí hoặc độ phức tạp**.
Thay vì cạnh tranh bằng cách có “nhiều tính năng hơn”, Airbridge cạnh tranh bằng:
- Độ tin cậy attribution tương đương AppsFlyer
- Analytics phục vụ tốt cho các quyết định hằng ngày
- Mô hình giá tăng theo giá trị kinh doanh thực tế
## Vì sao Airbridge là một lựa chọn thay thế AppsFlyer mạnh mẽ
Ngày càng nhiều thương hiệu chuyển [từ AppsFlyer sang Airbridge](https://abr.ge/lhs1gv) nhờ cách tiếp cận pricing hoàn toàn khác biệt trong mobile measurement, trực tiếp giải quyết những vấn đề cốt lõi khiến các đội ngũ tìm kiếm giải pháp thay thế ngay từ đầu.
### Triết lý đo lường không rào cản chi phí
Không giống các MMP truyền thống khóa tính năng sau các gói cao cấp, Airbridge cung cấp **attribution toàn diện, deep linking và analytics** mà không có giới hạn về gói giá.
Điều này đồng nghĩa với việc bạn có **toàn bộ visibility về hiệu suất marketing** dựa trên nhu cầu kinh doanh thực tế, thay vì bị giới hạn bởi mức phí bạn sẵn sàng chi trả cho từng tính năng.
### Deep linking mạnh mẽ và đo lường cross-platform web-to-app
Airbridge mang đến hệ thống **deeplink nâng cao**, được thiết kế cho những hành trình người dùng phức tạp và phân mảnh. Nhờ hạ tầng deeplink mạnh mẽ, marketer có thể kết nối liền mạch giữa **web, mobile web và app** mà không làm mất độ chính xác của attribution.
Điều này đặc biệt quan trọng trong các trường hợp:
- Hành trình **web-to-app**
- Flow onboarding của các app subscription
- Các sản phẩm AI có vòng lặp login và tương tác lặp lại
Với những đội phụ thuộc nhiều vào traffic từ web trước khi chuyển sang app, [**Deeplink Plan**](https://abr.ge/vigxykk) **chuyên biệt của Airbridge** đảm bảo routing ổn định, hỗ trợ deferred deep link và cung cấp đầy đủ visibility cross-platform.
### Quan hệ đối tác và tích hợp chính thức
Airbridge duy trì các [**quan hệ đối tác chính thức với Meta, Google Ads và TikTok**](https://abr.ge/2excxm), đảm bảo dữ liệu đo lường được tích hợp mượt mà với các nền tảng quảng cáo. Nhờ đó, bạn tránh được việc đối soát dữ liệu thủ công hay gián đoạn workflow — những vấn đề thường xảy ra khi chuyển đổi giữa các MMP.
### Theo dõi hành trình subscription và sản phẩm AI
Không chỉ giải quyết các nhu cầu trước mắt như deep linking hay attribution, Airbridge còn cung cấp hệ thống đo lường toàn diện bao gồm:
- Phân tích hành trình người dùng
- Insight về retention
- Bảo vệ gian lận 3 lớp
- Công cụ tối ưu chiến dịch
Điều này giúp đội ngũ không chỉ giải quyết vấn đề kỹ thuật hiện tại mà còn **xây nền tảng cho tăng trưởng dài hạn**.
## Final thoughts: AppsFlyer cũng đã làm khá tốt. Nhưng giờ là lúc để làm tốt hơn.
Không thể phủ nhận rằng trong nhiều năm, AppsFlyer là lựa chọn mặc định của thị trường, và trong một thời gian dài, điều đó hoàn toàn hợp lý. Nhiều đội ngũ đã xây dựng measurement stack xoay quanh nền tảng này, tăng trưởng cùng nó, và cũng trực tiếp trải nghiệm “enterprise-grade” thực sự vận hành như thế nào trong thực tế.
Nhưng thị trường đã thay đổi. Ngân sách ngày càng chặt hơn, đội ngũ ngày càng tinh gọn hơn, và việc phải trả thêm chỉ để mở khóa những tính năng cơ bản không còn là điều dễ chấp nhận.
Nếu bạn muốn độ chính xác attribution tương đương AppsFlyer, nhưng không đi kèm rào cản về chi phí nặng nề, gánh nặng vận hành hay cú sốc chi phí khi scale, thì Airbridge được xây dựng đúng cho giai đoạn này. Hãy xem cách các thương hiệu lớn đang tận dụng Airbridge để tạo ra kết quả ấn tượng:
👉[Cách Playio nâng D30 Retention lên 30% và giảm 40% Global UA CPA với Airbridge](https://abr.ge/jqz1p7)
👉[Nightly giảm 18% CPA với Simulated iOS Attribution và vươn lên top 3 App Store tại Nhật Bản](https://abr.ge/ubscdbp)
👉[Insight cross-platform hợp nhất của Airbridge giúp Shmoody mở rộng lên hơn 1 triệu lượt cài đặt app](https://abr.ge/a016utf)
Airbridge mang đến độ chính xác, khả năng deep linking và các mối quan hệ đối tác nền tảng mà bạn kỳ vọng ở một MMP hàng đầu — cùng với mức giá và trải nghiệm sử dụng thực sự phù hợp với cách các đội ngũ hiện đại vận hành.
## FAQ về các giải pháp thay thế AppsFlyer và so sánh MMP
### **Giải pháp nào phù hợp nhất cho startup nhỏ hoặc đội giai đoạn đầu?**
Các đội nhỏ thường ưu tiên thiết lập dễ dàng, giá dễ dự đoán và chi phí vận hành thấp. Những công cụ như Airbridge và Tenjin thường xuất hiện trong các so sánh nhờ sự đơn giản và rào cản gia nhập thấp, trong khi vẫn cung cấp độ chính xác attribution đủ tốt cho giai đoạn tăng trưởng ban đầu.
### **MMP nào phù hợp nhất cho app subscription?**
Với app subscription, MMP tốt nhất là công cụ có thể theo dõi install, conversion event, retention và doanh thu mà không khiến chi phí tăng quá cao khi số lượng event lớn.
Nhiều đội subscription lựa chọn Airbridge vì nền tảng này cân bằng được giữa độ chính xác attribution, pricing dễ dự đoán và hệ thống analytics được thiết kế xoay quanh lifecycle người dùng.
### **Airbridge có thể thay thế hoàn toàn AppsFlyer không?**
Có, trong nhiều trường hợp. Airbridge hỗ trợ attribution install, tracking event sau cài đặt, deep linking và đo lường tuân thủ privacy ở mức tương đương AppsFlyer.
Các đội chuyển sang thường không phải đánh đổi độ chính xác attribution, đặc biệt với các app subscription và sản phẩm AI.
### **Điểm khác biệt chính giữa AppsFlyer và các đối thủ là gì?**
Khác biệt lớn nhất không nằm ở logic attribution, mà ở cấu trúc chi phí, trải nghiệm sử dụng và độ phức tạp vận hành.
AppsFlyer và Adjust tập trung vào tính linh hoạt ở cấp enterprise, trong khi các nền tảng như Airbridge ưu tiên time-to-value nhanh hơn, workflow đơn giản hơn và pricing phù hợp với từng giai đoạn tăng trưởng.
---
## Gian lận trong Influencer Marketing: Cách click giả âm thầm làm hao hụt ngân sách app
URL: https://www.airbridge.io/vi/blog/influencer-marketing-fraud-fake-clicks-mobile-app
Language: Vietnamese
Published: 2026-04-20
Author: Luke
Một app fitness đã chi 8.000 USD cho ba influencer. Dashboard chiến dịch báo cáo 12.000 lượt click, 2.400 lượt cài đặt và tỷ lệ engagement 4,2%. Nhưng RevenueCat chỉ ghi nhận 11 người đăng ký mới. Hai trong số ba influencer không tạo ra bất kỳ doanh thu nào.
**Nếu không có dữ liệu chuyển đổi theo từng influencer, hầu hết các team sẽ không có cách nào biết liệu 2.400 lượt cài đặt đó đến từ người dùng thật hay bot** -- và thường chỉ nhận ra vấn đề sau nhiều tháng. Influencer marketing fraud không phải là rủi ro nhỏ lẻ, mà là một vấn đề mang tính hệ thống trong cách các đội mobile app đo lường hiệu quả influencer, âm thầm làm thất thoát ngân sách.
**Key Takeaways**
- **74% marketer từng gặp influencer fraud** ([AMRA & ELMA, 2025](https://www.amraandelma.com/influencer-fraud-statistics/)). Vấn đề này xuất hiện ở mọi ngành, nhưng các app subscription đặc biệt dễ bị ảnh hưởng vì fraud ẩn dưới các chỉ số top-of-funnel.
- **Fake followers, bot clicks và engagement pods** làm sai lệch các chỉ số mà đội ngũ thường dùng để đánh giá ROI influencer.
- **Các công cụ analytics phổ biến chỉ đo engagement, không đo doanh thu subscription.** GA4 và UTM có thể track click và install nhưng không nối được đến hành vi trả tiền trong app.
- **Tín hiệu fraud đáng tin nhất nằm ở dữ liệu sau cài đặt.** Nếu 500 install mà không có subscription, câu trả lời đã rất rõ — không cần tool chống fraud riêng.
- **Airbridge Core Plan cho phép thực hiện** [**attribution**](https://www.airbridge.io/blog/what-is-mobile-app-attribution) **theo từng influencer từ install đến subscription**, giúp các team quy mô nhỏ có visibility ở mức conversion mà không cần trả chi phí enterprise.
## **Quy mô của influencer fraud trong mobile app**
Influencer fraud là việc cố tình thổi phồng các chỉ số như follower, click, install hoặc engagement nhằm chuộc lợi, thu tiền các lượt chuyển đổi không có thật. Quy mô thiệt hại toàn cầu ước tính từ **1,3 tỷ USD** ([AMRA & ELMA](https://www.amraandelma.com/influencer-fraud-statistics/), 2019 baseline) **đến 4,6 tỷ USD mỗi năm** ([SociaVault, 2025](https://sociavault.com/blog/fake-follower-study-key-findings)). Vấn đề này tập trung chủ yếu ở nhóm influencer quy mô trung bình đến lớn, nơi động lực tài chính để thổi phồng số liệu cao nhất.
Với app subscription, rủi ro càng lớn vì sự khác biệt giữa một install giả và một subscriber thật là rất lớn — nhưng hoàn toàn **vô hình nếu không có dữ liệu chuyển đổi**.
Một nghiên cứu về follower giả năm 2025 cho thấy **37,2% follower của influencer là giả** trên các nền tảng lớn. Nhóm influencer quy mô lớn (100K–500K follower) có tỷ lệ gian lận cao nhất, lên đến **48,3%** ([SociaVault](https://sociavault.com/blog/fake-follower-study-key-findings)). Đối với các team app subscription trả phí theo mỗi lượt cài đặt hoặc mỗi lượt click, điều này đồng nghĩa với việc gần một nửa lượng reach được báo cáo có thể là không thật.
### 1\. Fake followers và lượng tiếp cận bị thổi phồng
Follower ảo giúp tăng lượng audience “trên bề mặt” của influencer. Một influencer fitness có 250K follower với 15K tài khoản giả trông không khác gì một người có 250K follower thật — cho đến khi bạn đo lường hành vi sau đó. Follower giả sẽ không bao giờ cài app, không bắt đầu trial và cũng không bao giờ subscribe.
### 2\. Click và install do bot tạo ra
Bot traffic không chỉ dừng lại ở follower giả. Các click farm tinh vi có thể tạo ra những lượt click và thậm chí cả install trông hoàn toàn hợp lệ, đủ để vượt qua các bộ lọc analytics cơ bản.
Theo dữ liệu từ TrafficGuard, **31% install trên iOS là gian lận**, và **15–25% ngân sách quảng cáo bị lãng phí mỗi năm do traffic không hợp lệ** ([TrafficGuard](https://www.trafficguard.ai/guides/invalid-traffic-in-the-mobile-app-industry)). Những install này vẫn xuất hiện trên dashboard như một nguồn acquisition bình thường — chỉ là chúng không bao giờ chuyển đổi.
### 3\. Engagement pods và social proof “giả lập”
Engagement pods là các nhóm tài khoản tương tác chéo với nhau để tăng like, comment và share một cách giả tạo. Với marketer đánh giá influencer dựa trên engagement rate, các nhóm này khiến influencer gian lận gần như không thể phân biệt với influencer thật.
Điểm chung của cả ba loại gian lận: **Ở top-of-funnel, mọi chỉ số đều trông rất đẹp. Nhưng khi đi xuống những funnel sâu hơn, lượt conversion sẽ ngày càng giảm mạnh**. Nếu việc đánh giá của bạn dừng lại ở click hoặc engagement rate, fraud sẽ hoàn toàn rất khó để xác định.
## Vì sao analytics thông thường không phát hiện được influencer fraud
Hầu hết các nền tảng influencer và công cụ analytics chỉ đo lường những gì xảy ra trên chính nền tảng đó — như impression, click và engagement. Chúng không đo lường được những gì xảy ra sau khi người dùng cài app.
[diagram-fraud-funnel.webp]
GA4 và tracking bằng UTM có thể phân bổ một lượt cài đặt về một link cụ thể. Nhưng với app subscription, câu hỏi quan trọng không phải là _“người dùng có cài app không?”_ mà là _“họ có đăng ký trả tiền không?”_ UTM sẽ “đứt gãy” tại App Store. GA4 cũng không thể kết nối một click trên web với một [subscription event](https://www.airbridge.io/blog/subscription-app-attribution) trong app nếu không có thêm hệ thống attribution hỗ trợ.
Điều này tạo ra một **blind spot** mà influencer gian lận có thể khai thác. Khi không có dữ liệu conversion sau cài đặt, **một kênh mang về 500 người dùng thật sẽ trông giống hệt một kênh mang về 500 bot**. 👉 Số install giống nhau. Nhưng kết quả doanh thu hoàn toàn khác.
Nền tảng của influencer chỉ báo cáo engagement. Ad network chỉ báo cáo click. Hệ thống billing chỉ báo cáo doanh thu. Nhưng **không có gì kết nối các lớp dữ liệu này theo từng influencer** -- **trừ khi bạn chủ động xây dựng hệ thống đó.**
[**_Phân bổ theo từng influencer từ install đến subscription. 15K install miễn phí. Bắt đầu ngay._**](https://www.airbridge.io/en/plans/core)
## Cách phân bổ theo từng influencer giúp xác định traffic gian lận
Nếu bạn đang tự hỏi làm thế nào để phát hiện influencer marketing fraud — từ click giả đến install do bot tạo ra — trong mobile app, thì câu trả lời thực ra đơn giản hơn nhiều so với những gì hầu hết nhà cung cấp giải pháp vẫn nói.
Điều bạn cần là dữ liệu chuyển đổi theo từng influencer, từ install cho đến doanh thu subscription. **Khi dữ liệu đủ minh bạch, fraud sẽ tự lộ diện.**
### Cách tiếp cận bằng attribution: Mỗi influencer một tracking link riêng
Nền tảng của phương pháp này rất đơn giản: gán một tracking link riêng cho từng influencer. Không phải một link cho cả chiến dịch — mà là một link cho mỗi đối tác. Cách làm này cho phép bạn đo lường riêng từng bước trong funnel:
[diagram-analytics-blind-spot.webp]
- Influencer A: 800 installs, 120 trials, 34 subscriptions
- Influencer B: 1,100 installs, 4 trials, 0 subscriptions
- Influencer C: 500 installs, 95 trials, 28 subscriptions
Đặc điểm của **Influencer B** — install cao nhưng gần như không có trial — là một dấu hiệu gian lận rất rõ. Không chỉ số engagement nào có thể cho bạn thấy điều này. Chỉ **dữ liệu chuyển đổi sau cài đặt** mới làm được.
Bài kiểm tra doanh thu ở đây gần như mang tính “có hoặc không”: nếu hàng trăm lượt cài đặt không tạo ra **0 đồng doanh thu subscription**, thì nguồn traffic đó hoặc là gian lận, hoặc kém chất lượng đến mức gây ra tác động tương đương lên ngân sách của bạn.
### Vì sao Airbridge Core Plan giúp mang tới khả năng trên, đặc biệt phù hợp với các team quy mô nhỏ
Các công cụ chống gian lận chuyên dụng thường có chi phí từ 500–2.000+ USD/tháng và chủ yếu hướng đến khách hàng enterprise. Với các đội phát triển app subscription quy mô nhỏ, mức giá này rất khó để áp dụng cho doanh nghiệp.
**Airbridge Core Plan** cung cấp lớp attribution giúp làm rõ influencer fraud — không phải thông qua các thuật toán phát hiện gian lận phức tạp, mà thông qua **sự minh bạch ở cấp độ chuyển đổi**.
Và đây là cách mà bạn có thể set up mọi thứ:
- **Tracking link theo domain riêng:** Tạo một link riêng cho từng influencer. Mỗi link đều đi vào cùng một hệ thống attribution, giúp bạn so sánh hiệu suất giữa các đối tác một cách trực tiếp.
- **25 sự kiện subscription tiêu chuẩn:** Bao gồm Start Trial, Subscribe, Unsubscribe… Các sự kiện này được cài đặt sẵn, chỉ cần tích hợp vào code app, giúp giảm đáng kể công sức thiết kế schema.
- **Tích hợp S2S với** [**RevenueCat**](https://www.airbridge.io/blog/abr-revenuecat-integration) **hoặc Adapty:** S2S (server-to-server) cho phép truyền dữ liệu theo thời gian thực mà không phụ thuộc vào trạng thái mở app của người dùng. Các sự kiện như gia hạn, hủy đăng ký hay thanh toán sẽ tự động được gửi về Airbridge.
- **Funnel và Revenue Report:** Theo dõi toàn bộ hành trình từ Install → Start Trial → Subscribe theo từng tracking link. Nếu funnel của một influencer “rơi về 0” ở bước trial, bạn sẽ biết ngay vấn đề nằm ở đâu.
Một [**Mobile Measurement Partner (MMP)**](https://www.airbridge.io/glossary/mobile-measurement-partner-mmp) đóng vai trò kết nối tất cả các lớp dữ liệu này. Core Plan tập trung vào 4 kênh quảng cáo chiếm 80–90% ngân sách phổ biến: **Google, Meta, Apple Search Ads và TikTok**, cùng tối đa 2 tích hợp bên thứ ba như RevenueCat.
Hơn thế nữa, Airbridge hiện tại đang loại bỏ rào cản chi phí với gói hỗ trợ hấp dẫn bao gồm:
- **15.000 install được phân bổ miễn phí**
- Sau đó: **$0.05/install**
- **Trả theo mức sử dụng**, không cần hợp đồng dài hạn
Với phần lớn các team ở giai đoạn đầu, **gói miễn phí đã đủ để test influencer marketing**.
## FAQ: Phát hiện influencer marketing fraud cho app marketer
### Tỷ lệ follower giả của influencer là bao nhiêu?
Trung bình khoảng **37,2%**, và có thể lên đến **48,3%** với nhóm influencer quy mô lớn (100K–500K follower) ([SociaVault](https://sociavault.com/blog/fake-follower-study-key-findings)). Trong khi đó, các nano và micro influencer (dưới 50K follower) thường có tỷ lệ gian lận thấp hơn. Đây cũng là lý do vì sao các hợp tác với nhóm nhỏ thường mang lại hiệu quả tốt hơn về **chi phí trên mỗi subscriber** đối với app subscription.
### Có thể phát hiện influencer fraud mà không cần tool chuyên dụng không?
Có. Các nền tảng chống fraud chuyên dụng sử dụng fingerprint thiết bị và thuật toán phát hiện bot — hữu ích nhưng chi phí cao (500–2.000+ USD/tháng).
Với app subscription, bạn có thể dùng một cách đơn giản hơn: So sánh **tỷ lệ install → trial → subscription giữa các influencer.** Nếu một đối tác có tỷ lệ chuyển đổi thấp hơn **10 lần so với trung bình chiến dịch**, tín hiệu đã đủ rõ để hành động — ngừng hợp tác hoặc yêu cầu giải thích trước khi thanh toán tiếp. Dữ liệu bạn cần là [**Lifetime Value (LTV)**](https://www.airbridge.io/glossary/lifetime-value-ltv) **theo từng influencer**, không phải “điểm fraud”.
### Tín hiệu đáng tin nhất cho thấy influencer gian lận là gì?
**Tỷ lệ install → trial theo từng influencer (không phải theo toàn chiến dịch)**. Traffic thật thường có 5–15% install chuyển thành trial. Nếu một influencer chỉ đạt 0,3% trong khi trung bình chiến dịch là 12%, đó là đối tượng cần kiểm tra ngay lập tức.
Sai lầm phổ biến nhất của các team là chỉ đo lường ở mức tổng (aggregate). Ví dụ: một chiến dịch có 3 influencer có thể có tỷ lệ trial trung bình là 10% — trông có vẻ tốt. Nhưng thực tế có thể là Influencer A đạt 15%, Influencer B đạt 14%, trong khi Influencer C chỉ đạt 0,4%. **Chỉ khi đo theo từng influencer, bạn mới phát hiện được vấn đề.**
## Click giả mạo làm mất tiền — giải quyết liền với dữ liệu chuyển đổi
Influencer fraud tồn tại vì hầu hết đội ngũ dừng lại ở bước đo lường install. Engagement trông có vẻ thật. Install cũng trông hợp lý. Nhưng sự khác biệt chỉ lộ ra ở phía sau — tại bước trial, subscription và doanh thu.
Giải pháp không phải là thêm công cụ mới, mà là kết nối những dữ liệu bạn đã có — install, trial, subscription — theo từng influencer trong cùng một góc nhìn.
[**_Bắt đầu miễn phí với Airbridge Core Plan — theo dõi theo từng influencer, từ install đến subscription, 15K install miễn phí._**](https://www.airbridge.io/en/plans/core)
---
## Meta AEM: Airbridge đã hỗ trợ View-Through Attribution
URL: https://www.airbridge.io/vi/blog/meta-aem-view-through-attribution
Language: Vietnamese
Published: 2025-12-04
Airbridge hiện đã hỗ trợ view-through attribution cho giao thức [Meta’s Aggregated Event Measurement (AEM)](https://www.facebook.com/business/help/721422165168355?id=1877298665783613) của Meta trên các chiến dịch iOS App Promotion đủ điều kiện. Tính năng này giúp mở rộng phạm vi đo lường vượt ra ngoài click-through attribution, mang lại khả năng quan sát rõ ràng hơn về mức độ đóng góp của lượt hiển thị (impressions) đối với lượt cài đặt (installs). Nhờ vậy, việc đánh giá hiệu quả chiến dịch Meta trở nên toàn diện và chính xác hơn.
Bản cập nhật áp dụng từ ngày 27/10/2025 cho tất cả chiến dịch iOS đủ điều kiện AEM.
## Meta AEM là gì?
Aggregated Event Measurement (AEM) là giao thức của Meta dùng để đo lường các sự kiện web và app trên thiết bị iOS 14.5 trở lên. AEM cung cấp dữ liệu gần thời gian thực, hỗ trợ quá trình tối ưu hóa hệ thống nhanh hơn và mang đến insight hiệu suất chi tiết hơn.
## Làm gì để kích hoạt view-through attribution cho AEM?
- Với các chiến dịch đã bật AEM: không cần thêm bất kỳ cài đặt nào.
- Với các chiến dịch chưa bật AEM: vui lòng tham khảo [Hướng dẫn tích hợp của Airbridge](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem) để hoàn tất thiết lập.
## Khi tạo chiến dịch App Promotion, có cần chọn AEM để xem kết quả view-through trong Airbridge không?
Không cần.
Airbridge sẽ tự động nhận dữ liệu view-through AEM của các chiến dịch đủ điều kiện. Để kiểm tra tính đủ điều kiện của app events, vui lòng tham khảo [Meta Business Help Center](https://www.facebook.com/business/help/617021863580582?id=1877298665783613).
## Sau cập nhật này, có những thay đổi nào trong Airbridge?
Airbridge sẽ bắt đầu thu thập và ghi nhận các touchpoint là view-through cùng với click-through attribution hiện có. Các lượt cài đặt được thúc đẩy bởi lượt hiển thị sẽ xuất hiện trong các mục sau:
- Attribution reports
- Analytics dashboards
- Raw data exports
Lưu ý: Dữ liệu SKAN sẽ tiếp tục được thu thập và phân bổ riêng biệt. Cập nhật này không ảnh hưởng đến báo cáo SKAN.
## Attribution Window
- Meta hỗ trợ view-through attribution window là 1 ngày đối với lượt cài đặt.
- Airbridge áp dụng probabilistic view-through window là 6 giờ.
Để giảm sai lệch số liệu giữa hai hệ thống, nhà quảng cáo có thể điều chỉnh Meta Ads attribution rules trực tiếp trên Airbridge Dashboard.
Để biết thêm thông tin chi tiết và quy trình thiết lập, vui lòng tham khảo [Hướng dẫn tích hợp AEM](https://help.airbridge.io/en/guides/meta-business-channel-integration#aggregated-event-measurement-aem) hoặc liên hệ đội ngũ hỗ trợ của Airbridge.
---
## Cách thiết kế một paywall hoàn hảo cho mobile app
URL: https://www.airbridge.io/vi/blog/design-paywall
Language: Vietnamese
Published: 2025-12-01
Khi xây dựng một app, đa số developer thường ưu tiên thiết kế tính năng và UI, và vô tình bỏ qua paywall – nơi người dùng thực sự đưa ra quyết định mua. 80% giao dịch diễn ra ngay sau onboarding khi người dùng nhìn thấy paywall lần đầu, vì vậy ấn tượng đầu tiên vô cùng quan trọng.
Paywall là điểm cuối trong hành trình ra quyết định “mua hay không mua”, nên mọi chi tiết đều phải được tính toán cẩn thận: số gói sản phẩm, mức gói giá, nội dung, thiết kế, và nhiều yếu tố khác. Bài viết này sẽ giúp bạn biết cần chú ý điều gì và cách tạo ra một paywall hiệu quả nhất cho app của mình.
# **Giá và ưu đãi dùng thử (Intro Offers)**
Trước khi bàn đến giao diện, hãy nhìn lại cách các mobile app định giá hiện nay. Hầu hết app sử dụng mô hình subscription theo chu kỳ thanh toán, trong đó các gói phổ biến nhất gồm:
- **Gói tháng. **Đây là tiêu chuẩn vàng. Với mức giá khoảng 5–10 USD, người dùng sẵn sàng chi trả khi thấy app có giá trị. Gói tháng phù hợp cho paywall có 2–3 offers hoặc thậm chí là chỉ có một lựa chọn gói giá duy nhất.
- **Gói 6 tháng. **Thường xuất hiện ở paywall với 3 offers, đôi khi xuất hiện ở cả paywall có 2 lựa chọn. Đây là “điểm giao” hợp lý giữa gói tháng và năm. Nếu định giá tốt, người dùng sẽ thấy đây là lựa chọn có lợi.
- **Gói năm. **Hiệu quả với app mang lại lợi ích hằng ngày hoặc cập nhật nội dung liên tục. Khi người dùng thấy đủ giá trị, họ chấp nhận mức giá cao hơn. Điều quan trọng là phải định vị gói năm sao cho “đáng tiền”.
- **Gói trọn đời (lifetime). **Là khoản thanh toán một lần dùng mãi mãi. Nó có thể phù hợp giai đoạn đầu để tạo doanh thu nhanh, nhưng cần hiểu rằng mỗi người dùng lifetime là một người dùng không trả subscription định kỳ.
Bạn cũng có thể bắt gặp các loại subscription khác như gói tuần, gói 2–3 tháng. Không phải app nào cũng phù hợp, nhưng rất đáng để thử nghiệm (nội dung sẽ được đề cập kỹ hơn ở phần A/B testing). Về mức giá, hãy bắt đầu bằng cách tham khảo đối thủ và đưa ra mức tương đương hoặc thấp hơn, đừng đặt cao hơn ngay lập tức, vì quá rủi ro.
[]
Một yếu tố quan trọng khác cần cân nhắc là các ưu đãi dùng thử (intro offers) mà bạn muốn áp dụng. Có ba loại chính: **free trial**, **pay as you go**, và **pay upfront**.
**Free trial** cho phép người dùng trải nghiệm app miễn phí trong một khoảng thời gian giới hạn trước khi thanh toán. Trong khi đó, hai lựa chọn còn lại cung cấp **mức giá ưu đãi ban đầu** giúp việc mua hàng trở nên hấp dẫn hơn.
Với **pay as you go**, người dùng được hưởng mức giá giảm cho subscription trong vài chu kỳ thanh toán đầu. Còn **pay upfront** thì cung cấp **một khoản giảm giá một lần** ngay ở kỳ thanh toán đầu tiên.
Những ưu đãi này kết hợp rất tốt với cách bạn cấu trúc subscription và tổ chức sản phẩm trên paywall — nội dung sẽ được bàn đến ngay dưới đây.
[]
# **Số lượng sản phẩm hiển thị trên paywall**
Sau khi thống nhất gói giá và ưu đãi, câu hỏi tiếp theo là: **paywall nên có bao nhiêu lựa chọn?** Thông thường mobile app có 1–3 sản phẩm. Mỗi cách đều có ưu – nhược điểm:
- **1 lựa chọn. **Rõ ràng, dễ hiểu, không gây phân tâm. Nhưng bạn sẽ khó thử nghiệm mức giá và khó đánh giá mức willingness-to-pay của người dùng.
- **2 lựa chọn. **Linh hoạt hơn. Tuy nhiên cần đảm bảo chênh lệch giá đủ rõ để tránh gây rối mắt hoặc khó hiểu.
- **3 lựa chọn. **Là cấu trúc phổ biến nhất. Bạn có nhiều không gian thử nghiệm và dễ định hướng người dùng vào gói mong muốn bằng cách làm cho gói đó “hấp dẫn” hơn hai gói còn lại.
[]
# **Các yếu tố thiết kế**
Khi đã xong phần lựa chọn mức giá và số lượng gói sản phẩm, đây là lúc bạn có thể sáng tạo, nhưng đừng “quá đà”, vì vẫn có những nguyên tắc cần tuân theo. Cách trình bày paywall sẽ thay đổi rất nhiều tùy từng app. Có thể phân loại ra khoảng [10 kiểu paywall phổ biến](https://adapty.io/blog/the-10-types-of-mobile-app-paywalls/), nhưng thực tế thì paywall giống như một bộ lắp ghép mà bạn có thể tùy chỉnh theo ý mình. Điều quan trọng là hiểu một paywall tốt cần những thành phần nào, rồi từ đó thiết kế phiên bản phù hợp nhất với app của bạn. Hãy đi theo thứ tự từ trên xuống dưới để xem những yếu tố nào là thiết yếu.
Trước tiên, để thu hút sự chú ý của người dùng, paywall nên có hình ảnh minh họa liên quan trực tiếp đến giá trị cốt lõi của app. Đặt một hình stock phù hợp hoặc một hình minh họa thiết kế riêng ở phía trên cùng luôn tạo hiệu ứng tốt. Một số app thậm chí sử dụng video, giúp truyền tải nhiều thông tin trực quan hơn về giá trị của sản phẩm.
Tiếp theo là **phần nội dung (copy)**. Đây là nơi bạn cần vừa sáng tạo vừa rõ ràng. Tiêu đề có thể mang chút “màu sắc riêng”, nhưng phần liệt kê lợi ích thì nên minh bạch và trung thực. Người dùng cần nhìn thấy lý do thuyết phục để trả tiền cho app của bạn, đồng thời paywall không được chứa nội dung gây hiểu nhầm nếu muốn vượt qua quá trình review của store.
Sau khi nêu rõ lợi ích, tiếp theo là phần **gói giá**. Thông thường chúng được thiết kế dưới dạng badge hiển thị thông tin subscription. Cách phổ biến nhất là dạng toggle, cho phép người dùng chạm để chọn gói mong muốn rồi bấm nút mua phía dưới. Một số app thiết kế badge như một nút mua trực tiếp, giúp rút ngắn thêm một bước chuyển đổi.
**Nút mua (CTA)** cũng rất quan trọng, vì đây chính là lời kêu gọi hành động. Hãy chọn từ ngữ rõ ràng, không gây hiểu nhầm. Ví dụ: “Purchase”, “Buy”, “Subscribe” đều minh bạch. Ngược lại, dùng “Start your free trial” cho một subscription có free trial có thể gây hiểu sai, vì người dùng sẽ bị tính phí sau khi hết trial.
Cuối cùng, đừng quên thêm các đường link cần thiết ở dưới cùng, gồm: **Terms of Use**, **Privacy Policy**, và **Restore Purchase**.
# **Analytics và A/B Testing**
Sau khi hoàn thành phần sáng tạo, bạn cần đảm bảo paywall hoạt động đúng như dự định. Để làm được điều đó, bạn phải theo dõi toàn bộ hành vi của người dùng như lượt xem trang, số lần nhấn nút, thao tác cuộn, giao dịch mua… và quan sát tỷ lệ chuyển đổi để đánh giá hiệu suất của paywall. Một giải pháp tiện lợi là sử dụng hệ thống phân tích dành riêng cho mobile app, ví dụ như Adapty. Công cụ này cung cấp dữ liệu chính xác, biến mọi hành vi người dùng thành các báo cáo dễ đọc và dễ diễn giải. Quan trọng hơn, dữ liệu này không chỉ để xem, mà còn có thể dùng trực tiếp để cải thiện paywall thông qua các thử nghiệm.
Nếu analytics cho thấy tỷ lệ chuyển đổi thấp, bạn cũng không cần thất vọng. Việc bạn đã biết được vấn đề nghĩa là bạn có thể bắt đầu tối ưu. Bạn có thể đặt ra một vài giả thuyết: có thể giá quá cao, hoặc nội dung chưa thể hiện đúng giá trị thật của app. Tuy nhiên, việc thay đổi mò mẫm rất tốn thời gian (chỉnh sửa app, xuất bản bản cập nhật mới, chờ review lại), thậm chí còn nguy hiểm, vì nếu giả thuyết sai, tỷ lệ chuyển đổi có thể giảm mạnh hơn nữa.
Cách an toàn và hiệu quả nhất để kiểm chứng giả thuyết là **chạy A/B test**, và Adapty hỗ trợ rất tốt. Việc thử nghiệm paywall không hề khó: bạn tạo hai phiên bản paywall, thường chỉ thay đổi một yếu tố (như giá gói tháng), chia traffic (thường 50/50), và sau 2–4 tuần xem kết quả để xác định giả thuyết đúng hay sai. Ưu điểm của cách làm này là bạn gần như không gặp rủi ro, vì traffic được chia đều. Quan trọng hơn, bạn không cần phát hành bản app mới, do Adapty SDK cho phép chỉnh sửa paywall từ xa bằng remote config, tiết kiệm cực nhiều thời gian.
Thực hiện A/B test thường xuyên cho paywall là cách chắc chắn để tăng doanh thu app lên đến **30% trong vài tháng đầu**.
[]
# **User Acquisition và Attribution**
Split-testing là cách tuyệt vời để tìm ra mức giá tối ưu và tăng tỷ lệ chuyển đổi, nhưng có một vấn đề thường gặp: **nếu app không đủ người dùng/tải xuống/cài đặt**, việc chạy A/B test sẽ rất khó vì không đủ dữ liệu để đạt độ tin cậy thống kê. Vì vậy, điều quan trọng là phải đảm bảo app được quảng bá đủ mạnh. Trong bối cảnh cạnh tranh hiện nay, kỳ vọng chỉ dựa vào organic traffic là điều khá ngây thơ. Để tăng lượng user acquisition (UA), bạn cần đầu tư quảng cáo, thông qua store ad managers, ad networks, social networks, hoặc in-app ads.
Bạn nên thử nhiều kênh để xem kênh nào đem lại người dùng phù hợp nhất. Và đây là lúc **Airbridge** phát huy sức mạnh. Một MMP như Airbridge giúp đo lường chính xác hiệu quả của từng chiến dịch trên từng kênh marketing, từ đó biết nên ưu tiên và đầu tư nhiều hơn vào kênh nào. Nếu app có phiên bản web, Airbridge cũng hỗ trợ cross-platform attribution, giúp bạn xem dữ liệu của cả web và app trên cùng một dashboard.
Khi bạn đã bắt đầu chạy A/B test và mua traffic, bạn rất dễ nhận ra một điều: **làm nhiều paywall khác nhau theo từng nguồn traffic thực sự rất hữu ích**. Mỗi kênh sẽ mang về một nhóm người dùng với hành vi và khả năng chi trả khác nhau, nên việc tùy chỉnh paywall theo từng nguồn là một chiến lược đáng thử, có thể là ưu đãi riêng, thiết kế riêng, hoặc cấu trúc giá riêng.
Trong trường hợp này, **Adapty** sẽ hỗ trợ đầy đủ nhu cầu A/B testing của bạn, còn **Airbridge** sẽ đảm nhiệm phần attribution và phân tích hiệu quả. Việc tìm ra nguồn traffic sinh lời tốt nhất và nhóm người dùng trả tiền trung thành nhất cần thời gian, nhưng nếu bạn liên tục thử nghiệm và kiểm chứng giả thuyết, tốc độ tăng trưởng của app sẽ vượt xa so với việc đưa ra quyết định một cách cảm tính.
# **Kết luận**
Chúng ta đã đi qua toàn bộ những yếu tố quan trọng để xây dựng một paywall hiệu quả cho mobile app. Tóm lại, đây là các điểm cốt lõi giúp paywall của bạn hấp dẫn hơn với người dùng và dễ dàng vượt qua vòng review của store:
- **Chọn mức giá và ưu đãi dùng thử phù hợp.** Hãy luôn xem người dùng mục tiêu là ai và đảm bảo mức giá của bạn tương thích với khả năng chi trả của họ.
- **Xây dựng cấu trúc hợp lý.** Số lượng gói trên paywall nên dựa trên giá trị mà app mang lại. App càng nhiều tính năng hữu ích thì bạn càng có thể thử nhiều loại subscription.
- **Thu hút bằng hình ảnh.** Người dùng xử lý thông tin trực quan rất nhanh và thường đưa ra quyết định dựa trên cảm nhận đầu tiên, nên hãy đầu tư vào hình ảnh và thiết kế phù hợp.
- **Truyền đạt mọi thứ thật rõ ràng.** Người dùng phải dễ dàng hiểu giá trị của app, mức giá họ sẽ trả, chu kỳ thanh toán, thời gian dùng thử, và mọi thông tin quan trọng khác.
- **Sử dụng analytics và chạy thử nghiệm.** Theo dõi chỉ số, quan sát conversion để biết điểm mạnh – điểm yếu của paywall. Tạo giả thuyết, kiểm chứng bằng A/B test, và tăng doanh thu của app.
**Đảm bảo đủ traffic.** Trong thị trường app đầy cạnh tranh, organic traffic không đủ để phát triển. Đừng ngần ngại đầu tư vào quảng cáo và theo dõi attribution cẩn thận để tối ưu kết quả.
---
## CPP đã thay đổi ra sao trong năm 2025: Cá nhân hóa trải nghiệm App Store để tăng trưởng chính xác hơn
URL: https://www.airbridge.io/vi/blog/the-cpp-revolution-in-2025-personalizing-app-store-experiences-for-precision-growth
Language: Vietnamese
Published: 2025-11-24
Trong kỷ nguyên marketing ưu tiên quyền riêng tư, khi việc mất tín hiệu đã trở thành điều “bình thường mới”, các đội ngũ buộc phải suy nghĩ lại về ý nghĩa thực sự của tối ưu hóa. Marketer không còn có thể chỉ dựa vào nhắm chọn đối tượng chi tiết hay dữ liệu chuyển đổi granular như trước. Thay vào đó, trọng tâm đang dịch chuyển sang **cá nhân hóa chính trải nghiệm người dùng** — từ nội dung quảng cáo cho đến trải nghiệm trên App Store.
Đây chính là lúc **Custom Product Pages (CPPs)** bắt đầu thay đổi cuộc chơi của user acquisition.
Từng được xem là một tính năng khá “ngách” trên App Store, CPPs nay đã trở thành một trong những đòn bẩy mạnh mẽ nhất để cải thiện tỷ lệ chuyển đổi, tăng mức độ ăn khớp giữa creative và thông điệp, cũng như nâng cao chất lượng người dùng. Với các marketer đang xoay xở trong bối cảnh hậu ATT, việc làm chủ CPPs đồng nghĩa với việc giành lại quyền kiểm soát — không phải bằng cách nhắm chọn nhiều hơn, mà bằng **kể câu chuyện thông minh hơn**.
## **Custom Product Pages là gì, và vì sao chúng quan trọng hơn bao giờ hết?**
Ra mắt vào cuối năm 2021, **Custom Product Pages (CPPs)** của Apple cho phép marketer tạo ra nhiều phiên bản khác nhau của trang App Store, mỗi phiên bản được tối ưu cho một nhóm người dùng, chiến dịch hoặc chủ đề creative cụ thể. Mỗi CPP có thể hiển thị bộ screenshot, app preview và nội dung quảng bá riêng, giúp thông điệp quảng cáo khớp chặt chẽ với trải nghiệm trên App Store.
Và chính những cập nhật tại [**Apple’s WWDC 2025**](https://developer.apple.com/videos/play/wwdc2025/328/)** **đã biến CPPs trở thành một đòn bẩy tăng trưởng thực sự quan trọng với marketer hiện nay. CPPs không còn bị giới hạn với paid traffic. Apple đã cho phép gán **từ khóa organic riêng cho từng CPP**, giúp các trang này có thể xuất hiện trực tiếp trong kết quả tìm kiếm trên App Store.
Nói cách khác, CPPs đã chính thức tiến hóa từ landing page phục vụ performance marketing thành **tài sản discoverability**, nơi paid user acquisition và App Store Optimization (ASO) giao thoa và hỗ trợ lẫn nhau.
> **Ví dụ đơn giản về CPPs:**
Một game thủ nhấp vào quảng cáo nhấn mạnh “PvP battles” sẽ được dẫn đến một CPP tập trung vào các chế độ chiến đấu. Trong khi đó, một người dùng tìm kiếm organic với từ khóa “city-building games” lại thấy một CPP làm nổi bật yếu tố sáng tạo và thư giãn. Cả hai trải nghiệm đều được cá nhân hóa theo intent, và đều được thiết kế để tối đa hóa khả năng chuyển đổi.
Chính sự ăn khớp giữa **thông điệp **và** động cơ người dùng** này là lý do khiến CPPs trở thành một trong những đòn bẩy chuyển đổi mạnh mẽ nhất trong năm 2025.
## **Tác động thực sự của CPPs: Kết nối creative và chuyển đổi**
### **1. Thu hẹp khoảng cách giữa creative và trải nghiệm App Store**
Trước khi có CPPs, marketer thường gặp một vấn đề quen thuộc: **creative và trải nghiệm trên App Store không khớp nhau**. Người dùng có thể nhấp vào một mẫu quảng cáo hiệu quả, nhưng lại được dẫn đến một trang App Store chung chung, không phản ánh đúng thông điệp mà quảng cáo đã hứa hẹn. Sự lệch pha này dẫn đến rơi rụng, tỷ lệ chuyển đổi thấp và chất lượng cài đặt kém.
CPPs loại bỏ ma sát đó bằng cách giúp marketer duy trì **tính nhất quán trong câu chuyện**, hay còn gọi là nguyên tắc *message match*. Theo [**Adaptly**](https://adaptly.dev/blog/ppc-message-match-guide), điều này không chỉ giúp cải thiện tỷ lệ chuyển đổi mà còn có thể **giảm CPC hơn 30%**, đồng thời nâng cao chất lượng người dùng nhờ thu hút đúng nhóm đối tượng thực sự phù hợp với giá trị cốt lõi của app.
### **2. Biến CPPs thành “phòng thí nghiệm” để tạo chuyển đổi**
CPPs ngày nay không còn chỉ là landing page — chúng là những thử nghiệm vi mô giàu dữ liệu.
Lớp phân tích sâu hơn được Apple giới thiệu trong năm 2025 cho phép các đội tăng trưởng vượt qua các chỉ số chuyển đổi bề mặt. Giờ đây, họ có thể đo lường hiệu suất của từng CPP thông qua **hơn 100 chỉ số mới trong App Store Connect**, từ đó theo dõi cách các cặp creative–CPP khác nhau ảnh hưởng đến retention, tương tác trong app và khả năng kiếm tiền.
> **Ví dụ:** một app thiền có thể nhận ra rằng các CPP nhấn mạnh “giảm căng thẳng” mang lại tỷ lệ chuyển đổi Day 1 cao hơn, trong khi các trang tập trung vào “cải thiện giấc ngủ” lại tạo ra retention dài hạn tốt hơn. Insight từ CPPs giờ đây được đưa thẳng vào chiến lược creative, thông điệp sản phẩm và tối ưu LTV.
### **3. Kết nối CPPs với toàn bộ funnel sau cài đặt**
Sức mạnh thực sự của CPPs được phát huy khi chúng được kết nối với **hành vi sau cài đặt**. Với một đối tác đo lường như **Airbridge**, marketer có thể phân bổ install, in-app conversion và các chỉ số retention ngược lại đúng CPP đã tạo ra chúng.
Quan trọng hơn, theo [**Apple**](https://developer.apple.com/help/app-store-connect/create-custom-product-pages/configure-multiple-product-page-versions), kể từ **iOS 18**, CPPs có thể tích hợp [**deep links**](https://www.airbridge.io/en/blog/deeplink-101-for-marketers-and-developers), đưa người dùng thẳng đến trải nghiệm trong app phù hợp ngay sau khi cài đặt. Điều này rút ngắn đáng kể “đường đến giá trị”, giúp tăng retention ngay từ phiên đầu tiên — một yếu tố then chốt để cải thiện các chỉ số D1 và D7 engagement.
## **Nhưng… chỉ 31% app hàng đầu sử dụng CPPs trong năm 2025**
Dù sở hữu nhiều khả năng mạnh mẽ như vậy, mức độ ứng dụng CPPs hiện nay vẫn thấp một cách đáng ngạc nhiên.
Theo [**AppTweak’s 2025 ASO Benchmarks Report**](https://www.apptweak.com/en/aso-blog/aso-app-store-trends-benchmarks-report), chỉ 31% các app hàng đầu và 26% game top đầu đang chủ động sử dụng Custom Product Pages. Trong khi đó, những app đã triển khai CPPs ghi nhận mức tăng chuyển đổi trung bình 8,6%, cho thấy tiềm năng cá nhân hóa vẫn còn rất lớn và chưa được khai thác đúng mức.
Với performance marketer, khoảng trống này chính là một lợi thế hiếm có: một tính năng vẫn chưa bị đối thủ tận dụng rộng rãi, nhưng đã chứng minh được ROI rõ ràng cho những thương hiệu sẵn sàng đầu tư vào trải nghiệm cá nhân hóa.
## **Cách các đội ngũ tăng trưởng hàng đầu sử dụng CPPs**
### **1. Căn chỉnh creative phù hợp với ý định người dùng**
Các app hiệu suất cao xem CPPs như phần mở rộng câu chuyện của quảng cáo. Một game RPG mobile có thể triển khai các CPP khác nhau theo từng chủ đề creative như “PvP combat”, “sưu tầm anh hùng” hay “idle rewards”. Với app fintech hoặc ngân hàng, CPPs có thể được phân tách theo các use case như chuyển tiền, đầu tư, tiết kiệm hay quản lý tín dụng.
Bằng cách ánh xạ mỗi creative quảng cáo với một CPP tương ứng, marketer xây dựng một **intent funnel liền mạch**, nơi người dùng luôn được đưa đến đúng trang phản ánh chính xác điều mà quảng cáo đã hứa hẹn.
[]
### **2. Chạy thử nghiệm dựa trên giả thuyết**
CPPs cho phép A/B testing ở quy mô lớn, xoay quanh giá trị cốt lõi, hình ảnh và giọng điệu truyền tải.
Ví dụ:
- Liệu việc hiển thị testimonial từ người dùng thật có giúp tăng CVR tốt hơn so với screenshot giao diện sản phẩm?
- Thông điệp “Tiết kiệm thông minh hơn” có hiệu quả hơn “Đầu tư với sự tự tin” không?
- Headline tập trung vào phần thưởng có hấp dẫn người dùng mới hơn không?
Theo thời gian, dữ liệu từ CPPs sẽ cho thấy đâu là yếu tố cảm xúc, hình ảnh hay giọng điệu chuyển đổi tốt nhất, từ đó quay ngược lại hỗ trợ chiến lược creative tổng thể và tối ưu ASO.
### **3. Triển khai trang theo chiến dịch hoặc theo mùa**
Không chỉ dừng lại ở việc căn chỉnh quảng cáo, CPPs đặc biệt hiệu quả trong các **chiến dịch ngắn hạn.** Các sự kiện theo mùa, hợp tác với influencer hay ra mắt tính năng mới đều hưởng lợi từ những CPP được thiết kế riêng để làm nổi bật nội dung liên quan.
Sự linh hoạt này giúp thương hiệu phản ứng nhanh với xu hướng thị trường, dịp lễ hoặc sở thích theo từng khu vực mà không cần thay đổi trang App Store mặc định. Đặc biệt, với cập nhật năm 2025 của Apple cho phép CPPs tham gia **organic discovery**, các CPP theo chiến dịch còn có thể xếp hạng cho những từ khóa theo thời điểm như “Black Friday deals”, “Holiday Shopping” hay “new season update”.
[]
## **Tương lai của CPPs: Cá nhân hóa theo ngữ cảnh và theo cohort**
Làn sóng đổi mới tiếp theo của CPPs nằm ở **cá nhân hóa động**, nơi nội dung được điều chỉnh dựa trên ngữ cảnh, hành vi người dùng hoặc cohort cụ thể. Khi cá nhân hóa theo ngữ cảnh và cohort trở nên dễ triển khai hơn, CPPs sẽ dần không còn là những trang tĩnh, mà hoạt động giống như **các “storefront” động** trên App Store.
> “Thay vì hiển thị cùng một bộ screenshot cho mọi người dùng, CPPs trong tương lai có thể tự động điều chỉnh dựa trên ý định, hành vi trước đó, hoặc thậm chí các tín hiệu theo thời gian thực như vị trí, loại thiết bị hay nguồn chiến dịch. Điều này đồng nghĩa với việc một người dùng đến từ TikTok creator sẽ thấy một CPP mang màu sắc vui nhộn, thiên về tính xã hội, trong khi một người dùng quay lại sau thời gian gián đoạn có thể thấy trang nhấn mạnh vào tính năng mới và phần thưởng.”
- Airbridge Marketing Team
Điều này có ý nghĩa gì với marketer?
- **Hiệu quả chuyển đổi cao hơn:** CPP theo ngữ cảnh hiển thị thông điệp phù hợp với ý định của từng người dùng, từ đó tăng khả năng cài đặt.
- **Chất lượng người dùng dài hạn tốt hơn:** CPPs theo cohort thu hút đúng nhóm người dùng có đặc điểm tương đồng với nhóm mang lại giá trị cao nhất, giúp cải thiện retention và LTV.
- **Giảm lãng phí ngân sách:** Marketer tránh chi tiền cho traffic không phù hợp bằng cách đảm bảo nội dung CPP khớp với kỳ vọng của người dùng.
- **ROAS mạnh hơn:** Trải nghiệm CPP liên quan trực tiếp đến hiệu quả kiếm tiền downstream, từ đó cải thiện ROAS tổng thể.
- **Mở rộng quy mô thông minh hơn:** Khi tín hiệu cohort rõ ràng hơn và chất lượng install cao hơn, đội tăng trưởng có thể scale chiến dịch với mức độ tự tin cao hơn.
## **Tăng trưởng cùng Airbridge: Biến insight từ CPPs thành tăng trưởng thực**
Trong khi CPPs giúp marketer cá nhân hóa trải nghiệm trên App Store, **Airbridge khuếch đại tác động đó** bằng cách kết nối dữ liệu CPP với attribution toàn funnel và phân tích hành vi người dùng sau cài đặt.
[]
Bằng cách liên kết từng CPP với nguồn cài đặt, các sự kiện post-install và chỉ số retention, Airbridge giúp marketer hiểu không chỉ CPP nào tạo ra lượt cài đặt, mà CPP nào thực sự nuôi dưỡng giá trị dài hạn.
> *CPPs cho bạn biết ai đã nhấp. Airbridge cho bạn biết ai ở lại, chi tiền và tiếp tục tương tác.*
Với Airbridge, marketer có thể:
- **Phân bổ install và post-install events** về đúng CPP đã tạo ra chúng, ngay cả trong môi trường SKAN.****
- **So sánh chất lượng người dùng theo từng biến thể CPP**, phân tích retention, LTV và ROAS theo cohort creative.****
- **Minh họa toàn bộ funnel** từ quảng cáo → CPP → install → mua hàng trong app, mang lại góc nhìn thống nhất về hiệu suất hành trình người dùng.****
- **Tận dụng deep linking** để dẫn người dùng từ CPP đến đúng trải nghiệm trong app, rút ngắn time-to-value và cải thiện mức độ tương tác.
---
## 3 phương pháp tối ưu luồng kiếm tiền cho ứng dụng đăng ký năm 2026
URL: https://www.airbridge.io/vi/blog/3-monetization-flow-best-practices-for-subscription-apps-in-2026
Language: Vietnamese
Published: 2025-11-14
## **Vì sao monetization flow đang dần thay đổi**
Nền kinh tế subscription trên mobile đang chuyển dịch rất nhanh, đặc biệt trong các lĩnh vực cạnh tranh cao như sức khỏe, fitness và lifestyle. Khi chi phí user acquisition ngày càng tăng, thành công phụ thuộc vào việc chuyển đổi người dùng mới thành người trả tiền càng sớm càng tốt trong hành trình.
Những chu kỳ chuyển đổi kéo dài không còn hiệu quả. App buộc phải nắm bắt ý định người dùng ngay từ khoảnh khắc họ vừa bước vào sản phẩm. Tính cấp bách là rất rõ ràng: theo [Business of Apps](https://www.businessofapps.com/guide/mobile-app-retention/), 77% người dùng hoạt động hằng ngày rời bỏ app chỉ trong vòng ba ngày, và phần lớn sẽ gỡ cài đặt trong chưa đầy một tuần. Trong bối cảnh này, một monetization flow hiệu quả không còn là lựa chọn “nên có”, mà đã trở thành **yếu tố sống còn** cho cả tăng trưởng lẫn khả năng tồn tại của app.
## **Monetization flow là gì và vì sao nó quan trọng**
Monetization flow là hành trình được thiết kế có chủ đích nhằm dẫn dắt người dùng từ lần tương tác đầu tiên với app đến thời điểm họ thanh toán. Hành trình này bao gồm toàn bộ các điểm chạm quan trọng như màn hình hiển thị onboarding, vị trí hiển thị paywall, cấu trúc gói giá và các chiến lược giữ chân người dùng.
Một subscription funnel được xây dựng tốt không chỉ nhắm đến chuyển đổi nhanh, mà còn tạo ra trải nghiệm liền mạch, tự nhiên và mang lại giá trị thực sự cho người dùng — đồng thời đảm bảo doanh thu bền vững cho doanh nghiệp.
[__wf_reserved_inherit]
Điều này quan trọng vì người dùng đến với app **đã mang sẵn ý định cao**. Tại thời điểm họ nhấn “install”, họ đã tìm hiểu trên App Store, đọc review, so sánh các lựa chọn và chủ động chọn app của bạn. Chính vì vậy, những tương tác đầu tiên bên trong sản phẩm là thời điểm vàng cho monetization. Nếu developer bỏ lỡ khoảnh khắc này, họ rất dễ mất người dùng trước khi kịp thể hiện giá trị thực sự của app.
Bên cạnh đó, có một sự thật đơn giản nhưng rất quan trọng trong các chiến lược subscription: “**người dùng trả tiền sẽ chú ý hơn.**” Người dùng trả phí không chỉ tạo ra doanh thu, mà còn tương tác sâu hơn, gắn bó lâu hơn và mang lại lifetime value cao hơn đáng kể so với người dùng miễn phí. Nói cách khác, monetization và engagement luôn song hành cùng nhau.
[__wf_reserved_inherit]
Đồng thời, một monetization flow được tối ưu tốt sẽ tạo ra sự cân bằng đúng đắn giữa chiến lược paywall mạnh mẽ và việc xây dựng niềm tin với người dùng. Flow này đảm bảo người dùng cảm thấy được dẫn dắt, không bị ép buộc, trong khi vẫn được khéo léo khuyến khích nâng cấp lên gói premium hoặc các tầng trả phí một cách tự nhiên.
Monetization không chỉ là tối đa hóa doanh thu. Nó còn đóng vai trò quan trọng trong việc cải thiện retention, tăng mức độ gắn bó với sản phẩm và cuối cùng là xây dựng một hệ sinh thái app lành mạnh hơn.
## **Best practices để tối ưu monetization flow**
Cách tiếp cận này đã được nhiều app thành công áp dụng, đặc biệt trong lĩnh vực Health & Fitness. Theo [Steve](https://www.linkedin.com/in/stevepyoung/), Founder của App Masters, app [**Flo Period & Pregnancy Tracker**](https://www.youtube.com/watch?v=ODWgul9tXus&t=1s) đã làm chủ monetization flow đến mức tạo ra 6 triệu USD doanh thu subscription mỗi tháng, chỉ từ khoảng **1 triệu lượt tải mỗi tháng**.
Dưới đây là một số chiến lược mà các đội phát triển app hàng đầu đang tận dụng hiệu quả:
### **1\. Onboarding & tâm lý người dùng**
- **Onboarding với từng câu hỏi một** → giảm quá tải thông tin và tăng tỷ lệ hoàn thành.
[__wf_reserved_inherit]
- **Tín hiệu cá nhân hóa** (như gọi tên người dùng) giúp tạo kết nối cảm xúc.
- **Social proof kết hợp với sự đồng cảm** (ví dụ: “bạn không hề đơn độc”) giúp tăng mức độ tin cậy.
- **Giới thiệu tính năng trước paywall** — việc xin quyền truy cập sớm giúp cải thiện tỷ lệ chuyển đổi.
### **2\. Chiến lược paywall**
- **Paywall đầu tiên xuất hiện trong onboarding** — 60–80% giao dịch diễn ra ngay lần đầu người dùng mở app.****
- **Paywall hai bước kèm ưu đãi** (paywall ban đầu → gợi ý gói năm có giảm giá) giúp tăng chuyển đổi.****
- **Tùy chọn free trial trên paywall** tạo cam kết tâm lý và giảm tỷ lệ hủy.****
- **Paywall trước bước đăng ký** — một khách hàng ghi nhận **tăng 52% doanh số** chỉ bằng cách đổi thứ tự này.****
- **Hiển thị paywall nhiều lần** — càng nhiều điểm chạm (onboarding, đăng ký, nội dung bị khóa, mở lại app), doanh thu càng cao.
[__wf_reserved_inherit]
- Micro-offer ở lần mở app thứ hai → mở khóa với $0.99, sau đó gia hạn $X/năm; hiệu quả vì nguyên lý “người dùng đã trả tiền sẽ chú ý hơn”.
### **3\. Content Locking & Engagement**
- **Chiến lược khóa nội dung** → cho người dùng xem trước giá trị (preview tính năng), nhưng khóa quyền truy cập để thúc đẩy nâng cấp.****
- **Nhiều tab bị khóa** (insight, chat bí mật, tin nhắn…) liên tục nhắc người dùng về những gì họ đang bỏ lỡ.
## **Tăng trưởng monetization flow cùng Airbridge**
Thiết kế một monetization flow subscription tốt mới chỉ là một nửa chặng đường; nửa còn lại là đo lường và tối ưu liên tục. Phần lớn app hiện nay vẫn đối mặt với chi phí UA tăng cao, hành trình người dùng bị phân mảnh giữa iOS, Android và SKAN, cùng bài toán xác định paywall hay ưu đãi nào thực sự hiệu quả. Nếu không có hệ thống đo lường phù hợp, ngay cả flow được thiết kế tốt nhất cũng có thể bỏ lỡ doanh thu.
Đây là lúc Airbridge phát huy vai trò của một MMP. Cụ thể, Airbridge hỗ trợ đội app bằng cách:
- Đo lường hiệu suất paywall đa nền tảng **(Android, iOS, SKAN)** thông qua Funnel Report, đảm bảo dữ liệu rõ ràng cho A/B testing.
- Với **phân tích funnel và ROAS chi tiết**, đội ngũ có thể xác định bước nào trong monetization flow (ví dụ: free trial toggle, paywall thứ hai, micro-offer) mang lại ROI tốt nhất.
- Tích hợp dữ liệu chi phí, giúp cân bằng **chi phí user acquisition** với kết quả monetization để đảm bảo lợi nhuận.
- Cung cấp insight **cross-platform và cross-channel**, cho phép nhân rộng mô hình thành công như Flo một cách bền vững và có hệ thống.
---
## Luồng Web-to-App và App-to-Web trong subscription app: Vì sao hybrid funnel luôn tạo ra hiệu suất vượt trội
URL: https://www.airbridge.io/vi/blog/web-to-app-vs-app-to-web
Language: Vietnamese
Published: 2025-10-27
Subscription app không chỉ bán tính năng, họ bán hành vi. Hãy nghĩ thử mà xem: không ai trả tiền cho một app thiền chỉ để nghe nhạc nền; họ trả cho nhịp độ thư giãn trước khi ngủ mỗi tối. Fitness app? Không chỉ là bài tập; đó là sự duy trì đều đặn. App học ngôn ngữ? Họ kiếm tiền từ những bước tiến nhỏ về kiến thức mỗi ngày.
Đó là lý do hành trình mà người dùng trải qua, từ cú nhấn đầu tiên đến khoản thanh toán đầu tiên, quan trọng hơn nhiều so với những gì các founder thường nghĩ. Một số app tập trung vào **funnel web-to-app**, trong khi số khác ưu tiên **app-to-web**. Nhưng với những app hàng đầu ? Họ kết hợp cả hai.
Vậy web-to-app và app-to-web thực sự nghĩa là gì, và vì sao ngày càng nhiều thương hiệu subscription chọn mô hình hybrid này? Cùng phân tích dưới đấy.
## **Web-to-App: Bắt đầu trên web, kết thúc trong app**
Hãy tưởng tượng bạn thấy một quảng cáo fitness trên Instagram với lời hứa “7 ngày để trở thành phiên bản khỏe nhất của chính bạn.” Thay vì đưa bạn thẳng đến App Store, quảng cáo dẫn bạn đến một landing page một cách hoàn hảo. Trang tải nhanh, giải thích rõ giá trị, và kết thúc bằng một CTA đơn giản: “Tiếp tục trong app.” Chỉ với một lần chạm, bạn tải app. Bên trong app, onboarding thu thập sở thích của bạn, đưa ra một “con đường giảm cân nhanh nhất,” rồi hiển thị paywall subscription ở thời điểm phù hợp.
Đó chính là mô hình **web-to-app**.
Quy trình vận hành:
1. Người dùng tìm thấy thương hiệu qua SEO, quảng cáo hoặc content marketing.
2. Họ truy cập vào landing page tối ưu cho mobile với thông điệp rõ ràng.
3. Trang nhấn mạnh lợi ích và dẫn dắt bằng một CTA mạnh mẽ.
4. Một deep link đưa họ đến App Store hoặc Google Play để tải và mở app.
5. Bên trong app, onboarding được cá nhân hóa theo hành vi và nhu cầu.
6. Paywall subscription xuất hiện đúng thời điểm trong hành trình trải nghiệm.
Đối với marketer, mục tiêu của funnel này rất rõ ràng: **acquisition**. Người dùng thường khám phá app thông qua công cụ tìm kiếm, quảng cáo trả phí hoặc nội dung blog. Web là nơi sự tò mò được khơi dậy, nhưng app mới là nơi hành vi được hình thành. Mobile app mang lại những thứ mà web khó làm được: push notification, streak khuyến khích duy trì thói quen, onboarding được cá nhân hóa, tất cả giúp người dùng gắn bó và quay lại nhiều hơn.
## **App-to-Web: Từ free trial đến thanh toán**
Giờ hãy đảo ngược kịch bản. Người dùng tải một productivity app, trải nghiệm free trial và sau đó nhận được thông báo nâng cấp. Thay vì thanh toán trực tiếp trong app, nút “Upgrade” đưa họ đến trang checkout trên web. Tại đây, họ thấy các gói linh hoạt hơn, nhiều phương thức thanh toán hơn và đôi khi còn có mức giá thấp hơn. Sau khi thanh toán, quyền truy cập premium được kích hoạt ngay lập tức trong app.
Đó chính là mô hình **app-to-web**.
Quy trình vận hành:
1. Người dùng tải app và bắt đầu sử dụng tính năng miễn phí hoặc bản trial.
2. App hiển thị nhắc nhở nâng cấp sau khi sử dụng hoặc khi trial kết thúc.
3. CTA dẫn người dùng đến trang checkout bảo mật trên web.
4. Người dùng lựa chọn giữa nhiều gói subscription khác nhau. Thanh toán được thực hiện bằng các phương thức linh hoạt như thẻ ngân hàng, PayPal, BNPL hoặc ví điện tử địa phương.
5. Quyền truy cập premium được mở khóa và đồng bộ trở lại trong app.
**Mục tiêu của funnel này là monetization.** Khi chuyển thanh toán sang web, thương hiệu tránh được mức phí **15–30% từ App Store hoặc Google Play**. Quan trọng hơn, họ có sự linh hoạt để thử nghiệm ưu đãi, tạo bundle, và cung cấp phương thức thanh toán phù hợp từng thị trường. Ở quy mô lớn, mức tiết kiệm và sự tự do tối ưu này tạo ra khác biệt đáng kể.
## Vì sao nhiều app kết hợp cả hai (và vì bạn cũng nên làm vậy)
Điểm mấu chốt là: phần lớn subscription app thành công không chọn một hướng duy nhất. Họ kết hợp cả hai funnel để tối ưu hiệu suất.
### **1. Spotify**
Ông lớn trong ngành streaming đã thành thạo hybrid funnel bằng cách ưu tiên đăng ký qua web để tránh mức phí hoa hồng cao từ các app store đối với người dùng mới.
- **Web-to-app strategy:** Spotify chạy quảng cáo trên công cụ tìm kiếm và mạng xã hội, dẫn người dùng về website. Tại đây, họ có thể đăng ký gói Premium, thường kèm ưu đãi như “€0 trong 1 tháng, sau đó €11.99/tháng”. Sau khi đăng ký, người dùng được khuyến khích tải app, với deep link đưa họ đến đúng màn hình cần thiết trong app.****
- **App-to-web strategy:** Chính app Spotify lại đóng vai trò chủ lực trong việc duy trì engagement và retention. Người dùng đang dùng bản miễn phí có thể nhận email hoặc xem quảng cáo được nhắm mục tiêu, mời họ nâng cấp lên Premium thông qua một liên kết dẫn đến website Spotify. Cách này chuyển billing khỏi app store, giúp Spotify giữ lại phần doanh thu cao hơn.
[]
### **2. Netflix**
Netflix là ví dụ kinh điển cho mô hình dùng web để đăng ký và dùng app để tạo engagement. Dù Netflix gần đây đã mở lại in-app purchase trên một số nền tảng, chiến lược hybrid của họ vẫn còn rất rõ rệt.
- **Web-to-app strategy:** Suốt nhiều năm, thậm chí tới nay vẫn phổ biến, Netflix luôn dẫn người dùng mới về website để bắt đầu subscription. Website xử lý toàn bộ quy trình thanh toán và tạo tài khoản. Sau khi hoàn tất, người dùng được mời tải app và đăng nhập vào tài khoản mới. Chiến lược này được thiết kế nhằm tránh mức phí hoa hồng cao từ app store.
- **App-to-web strategy:** App Netflix là giao diện chính để xem nội dung, tận dụng các tính năng native nhằm mang lại trải nghiệm mượt như xem TV. Dù một số chức năng thanh toán đã có thể thực hiện trong app, Netflix vẫn điều hướng người dùng về website cho các hành vụ quản lý tài khoản quan trọng, duy trì một hệ sinh thái web làm trung tâm cho các giao dịch then chốt.
### **3. Duolingo**
Ngôi sao trong lĩnh vực giáo dục về ngôn ngữ này tận dụng mô hình hybrid một cách rất hiệu quả để tối đa hóa conversion và giữ doanh thu, thậm chí đôi khi đưa ra mức giá khác nhau giữa web và app store.
- **Web-to-app strategy:** Duolingo sử dụng web funnel để thu hút người dùng mới bằng cách cho họ học trọn vẹn một bài ngay trên website. Khi người dùng được trải nghiệm một bài học thú vị, dễ hiểu và có tiến bộ thật, họ xây dựng được niềm tin và nhận thấy giá trị trước khi Duolingo đề nghị cam kết lâu dài. Kết thúc bài học, người dùng được mời tải app để tiếp tục hành trình, thường là qua deep link ghi nhớ tiến độ học của họ.
- **App-to-web strategy:** Với người dùng bản miễn phí trên app, đội ngũ retention của Duolingo thường gửi email ưu đãi độc quyền trên web để mời họ nâng cấp lên “Super Duolingo.” Chiến lược này không chỉ tránh phí app store mà còn mở ra cơ hội thử nghiệm các mô hình giá khác nhau, phù hợp từng nhóm thị trường.
### **4. Bumble**
Dating app này áp dụng chiến lược hybrid để triển khai mức giá khuyến mãi và xử lý thanh toán bên ngoài hệ thống app store trong một số ưu đãi nhất định.
- **Web-to-app strategy:** Bumble sử dụng các bài quiz và nội dung trên website để thu hút người dùng mới với khả năng attribution chính xác hơn và chi phí thấp hơn. Thông qua các chiến dịch web, họ có thể cung cấp mức giá ưu đãi cho lần subscription đầu tiên, điều không thể thực hiện trong in-app purchase vì bị ràng buộc bởi chính sách app store. Sau khi người dùng đăng ký trên web, họ được chuyển sang app để bắt đầu sử dụng dịch vụ.
- **App-to-web strategy: **Dù người dùng có thể mua tính năng premium ngay trong app, Bumble vẫn tận dụng email marketing để thúc đẩy gia hạn hoặc quảng bá ưu đãi đặc biệt thông qua website, giúp thu trọn doanh thu mà không mất phí chia sẻ. Ví dụ: người dùng đã hết hạn subscription có thể nhận được email với giá “welcome back” đặc biệt, dẫn họ đến web để đăng ký lại.
### **5. Noom**
App về sức khỏe và kiểm soát cân nặng này đã xây dựng giai đoạn tăng trưởng ban đầu dựa vào web-to-app funnel và biến nó thành trụ cột trong chiến lược marketing của mình.
- **Web-to-app strategy:** Noom nổi tiếng với bài quiz onboarding trên web. Bài quiz thu thập mục tiêu cá nhân và dữ liệu cơ bản, giúp xây dựng niềm tin trước khi người dùng tải app. Khi hoàn thành, người dùng được dẫn sang app để bắt đầu hành trình giảm cân của họ.
- **App-to-web strategy: **Dù app đóng vai trò trung tâm cho engagement hằng ngày, web funnel ban đầu cho phép Noom thu thập first-party data để nhắm mục tiêu và tối ưu chính xác hơn. Dữ liệu này giúp Noom re-engage người dùng qua email hoặc các kênh web khác, điều hướng họ trở lại website cho những ưu đãi nâng cấp đặc biệt.
[]
Tại sao phải “nhảy hai bước”? Vì acquisition và monetization là hai trận chiến hoàn toàn khác nhau. Bạn cần app để xây dựng hành vi bền vững, nhưng bạn cần web để bảo toàn biên lợi nhuận và thử nghiệm ưu đãi một cách tự do.
Giống như vận hành một quán cà phê: bạn muốn khách bước vào cửa (đó là web-to-app), nhưng bạn cũng không muốn trả phí cho bên trung gian mỗi lần ai đó mua một ly latte (đó là app-to-web).
### **Hybrid Funnel: Bức tranh trực quan**
Hãy hình dung ba funnel đặt cạnh nhau:
- **Web-to-App:** Discovery → Landing Page → App Install → Subscription trong app****
- **App-to-Web:** App Install → Free Trial → Web Checkout → Premium Access trong app****
- **Hybrid:** Discovery → App Install → Engagement trong app → Payment trên web → Retention dài hạn trong app
Mô hình này kết hợp khả năng mở rộng với sự kiểm soát tốt hơn. Thương hiệu thu hút người dùng trên web để tối ưu acquisition, sau đó monetize theo cách họ chủ động quyết định, trong khi app vẫn giữ vai trò động cơ tạo engagement mỗi ngày.
## **Thách thức đo lường**
Hybrid funnel mở ra nhiều cơ hội nhưng cũng kéo theo không ít rắc rối. Người dùng di chuyển liên tục giữa web và app khiến attribution trở nên phức tạp:
Quảng cáo nào mang lại lượt cài đặt? Subscription diễn ra trên web hay trong app? Funnel nào mang lại LTV cao hơn? Nếu thiếu dữ liệu rõ ràng, quyết định chi tiêu sẽ chỉ là phỏng đoán.
Đây là khi một measurement partner như [**Airbridge**](https://www.airbridge.io/)** **trở nên quan trọng. Airbridge kết nối dữ liệu web và app, cung cấp bức tranh toàn diện: chiến dịch nào tạo install, checkout flow nào tạo revenue, và hành trình nào mang lại người dùng trung thành. Khi dữ liệu được đo chính xác, hybrid funnel không còn là lý thuyết, mà trở thành một growth engine đáng tin cậy.
## **Vậy funnel nào hiệu quả nhất?**
Câu trả lời là: **không funnel nào mang lại giá trị lớn nếu đứng đơn lẻ.**
- Web-to-app là cánh cửa để có install và engagement ban đầu.
- App-to-web là đòn bẩy bảo vệ revenue và mở rộng khả năng thanh toán.
Câu chuyện tăng trưởng thật sự của subscription app nằm ở việc kết hợp cả hai. Đây cũng chính là chiến lược của những app bền vững nhất năm 2025, được hỗ trợ bởi hệ thống đo lường theo dõi trọn vẹn hành trình người dùng. Vì trong subscription app, **hành trình người dùng như thế nào quan trọng không kém điểm đến cuối cùng của họ.**
👉 Loay hoay tìm cách đo lường hybrid funnel để không bỏ sót bất kỳ touchpoint nào? Liên hệ [Airbridge](https://www.airbridge.io/) để được tư vấn thêm!
---
## Airbridge hỗ trợ báo cáo chuyển đổi theo thời gian thực cho TikTok iOS
URL: https://www.airbridge.io/vi/blog/tiktok-ios-realtime-conversion-reporting
Language: Vietnamese
Published: 2025-10-21
Airbridge hiện đã hỗ trợ Real-Time Conversion Reporting của TikTok cho các chiến dịch trên iOS 14.5 trở lên.
Với tư cách là TikTok Mobile Measurement Partner (MMP) được cấp giấy phép, Airbridge giúp marketer vượt qua độ trễ của SKAN và đưa ra quyết định nhanh hơn, chính xác hơn.
## **Real-Time Conversion Reporting là gì?**
Real-Time Conversion Reporting của TikTok sử dụng mô hình probabilistic từ MMP để bổ trợ cho SKAN. Không giống SKAN — vốn có thể làm chậm báo cáo lên đến 48 giờ — tính năng này cung cấp dữ liệu theo thời gian thực trực tiếp trong TikTok Ads Manager. Nhờ đó, marketer có thể theo dõi conversion, cost per conversion và conversion rate gần như ngay lập tức, thay vì phải chờ các báo cáo tổng hợp bị trì hoãn.
Tính năng này cũng là nền tảng cho Smart+, giải pháp tối ưu chiến dịch dựa trên AI của TikTok. Khi có tín hiệu mới và chính xác hơn, Smart+ có thể tối ưu nhanh hơn và mang lại hiệu suất tốt hơn.
## **Vì sao Real-Time Conversion Reporting hoạt động hiệu quả nhất khi kết hợp với SKAN?**
Airbridge cho phép chạy Real-Time Conversion Reporting song song với SKAN. Đây không phải là giải pháp thay thế SKAN, mà là lớp dữ liệu bổ sung.
Khi kết hợp cả hai trong Airbridge, marketer vừa có khả năng quan sát hiệu suất chiến dịch nhanh hơn, vừa duy trì được độ tin cậy của báo cáo SKAN. Cách tiếp cận hai lớp này mang lại insight sâu hơn và giúp đội ngũ tự tin điều chỉnh ngân sách, bid và creative đúng thời điểm cần thiết.
## **Cách kích hoạt TikTok Real-Time Reporting với Airbridge**
Việc bật Real-Time Conversion Reporting với Airbridge rất đơn giản:
- Tắt [Extended Privacy Control(EPC)](https://help.airbridge.io/en/guides/extended-privacy-control) khi truyền dữ liệu từ Airbridge sang TikTok.
- Điều này giúp đảm bảo tính nhất quán giữa phương thức postback của TikTok và cơ chế probabilistic matching.
Marketer có thể cập nhật cài đặt này trực tiếp trong dashboard Airbridge hoặc liên hệ đội ngũ Airbridge để được hỗ trợ. Sau khi kích hoạt, Airbridge sẽ giảm chênh lệch dữ liệu giữa TikTok và MMP, đặc biệt ở các chỉ số conversion, CPA và ROAS, đồng thời tăng hiệu quả cho các tính năng tối ưu tự động như Smart+.
## **FAQs: Những điều bạn cần biết**
**Q. Real-Time Conversion Reporting có thể thay thế SKAN không?
**Không. Real-Time Conversion Reporting là một luồng đo lường bổ sung. Nếu bạn tắt SKAN, dữ liệu SKAN sẽ không còn hiển thị trong TikTok Ads Manager cũng như trong MMP. Việc chạy song song cả hai mang lại góc nhìn đầy đủ và cân bằng nhất.
**Q. Tính năng này có dùng được nếu không có MMP không?
**Hiện tại thì không. TikTok SDK và Event API chưa hỗ trợ Real-Time Conversion Reporting độc lập. Việc tích hợp với một MMP là bắt buộc.
**Q. Làm thế nào để thiết lập trong Airbridge?**
Sau khi đã tích hợp TikTok với Airbridge, bạn chỉ cần tắt EPC transmission.
**Q. Hỗ trợ những phiên bản iOS nào?**
iOS 14.5 trở lên.
**Q. App nào cũng dùng được tính năng này không?**
Việc đủ điều kiện hay không phụ thuộc vào **lượng tín hiệu dữ liệu** của app. Trong TikTok Ads Manager, hãy kiểm tra thông báo:
*“Your app is eligible for iOS Real-Time Reporting.”*
## **Bắt đầu với Airbridge**
Airbridge giúp việc kích hoạt **TikTok iOS Real-Time Conversion Reporting** trở nên đơn giản và liền mạch. Marketer có thể đo lường hiệu suất chính xác hơn, ra quyết định nhanh hơn và cung cấp tín hiệu chất lượng cho Smart+ để tối ưu hiệu quả hơn.
Sẵn sàng nâng cấp hiệu suất chiến dịch TikTok? **Airbridge sẵn sàng đồng hành cùng bạn.**
---
## Cách phát triển subscription app của bạn: 5 yếu tố chính giúp tăng doanh thu hiệu quả trong năm 2025
URL: https://www.airbridge.io/vi/blog/how-to-grow-your-subscription-app-5-proven-elements-for-scaling-revenue-in-2025
Language: Vietnamese
Published: 2025-10-01
Nếu bạn làm việc trong lĩnh vực subscription app, hẳn bạn đã hiểu rõ thách thức. CAC ngày càng tăng, người dùng trở nên hoài nghi hơn khi phải trả phí, và churn có thể “ăn mòn” doanh thu bạn vất vả xây dựng chỉ sau một đêm. Dù vậy, subscription app vẫn là một trong những mô hình tăng trưởng nhanh nhất trên mobile. Theo SensorTower, mô hình này tăng **41% toàn cầu và 44% tại Mỹ** trong năm 2021, và xu hướng đó vẫn tiếp tục cho tới ngày nay.
Hãy nhìn vào những “ông lớn” như **Calm, Noom và Headspace**. Đây không chỉ là những lifestyle app, mà là những doanh nghiệp hoàn chỉnh với dòng doanh thu định kỳ ổn định và có thể dự đoán.
Vậy điều gì khiến những subscription app lớn này khác biệt? Tại Airbridge, chúng tôi nhận thấy cùng một công thức tăng trưởng lặp lại ở mọi lĩnh vực, xoay quanh 5 yếu tố. Mỗi yếu tố đều quan trọng, nhưng khi kết hợp, chúng trở thành động lực không thể ngăn cản.
## **1. Paywall tốt giúp tạo chuyển đổi**
Thành thật mà nói, phần lớn người dùng không vượt qua nổi paywall. Vấn đề duy nhất đó là ranh giới giữa churn ở free trial và doanh thu định kỳ.
Sai lầm phổ biến là nhiều đội ngũ xem paywall như yếu tố phụ. Họ chỉ xây một lần, đặt giá rồi để đó. Trong khi đó, các subscription app hàng đầu xem paywall như một phễu bán hàng sống động. Họ liên tục A/B test, thay đổi headline, chỉnh nội dung nút bấm, thử màu sắc, thậm chí sắp xếp lại thông tin để tối ưu tỉ lệ chuyển đổi.
[]
Ví dụ, chỉ một thay đổi nhỏ trong câu từ, từ “*Unlock lifetime insights*” sang “*Unlock everything for a lifetime*”, đã giúp tăng conversion rate lên **10,4%**. Đây là minh chứng rõ ràng rằng những điều chỉnh nhỏ có thể mang lại tăng trưởng doanh thu đáng kể.
**Đây là kinh nghiệm xương máu cho bạn**
- **Tần xuất:** Hiển thị paywall thường xuyên hơn.
- **Đề xuất:** sử dụng các phiên bản paywall khác nhau cho người dùng mới, người quay lại lần hai hoặc những người rời bỏ ở bước thanh toán.
- **Content:** thử các layout dài hơn với CTA nổi bật, thông điệp thu hút và social proof rõ ràng.
- **Hooks:** ưu đãi giới hạn thời gian, thông điệp cá nhân hóa hoặc testimonial từ người dùng thật có thể tạo thêm động lực để họ đăng ký.
👉 **Mẹo nhỏ:** Hãy xem paywall như một chiến dịch riêng biệt. Luôn thử nghiệm và điều chỉnh, đừng bao giờ để nó “tù đọng.”
## **2. Pricing tốt để phù hợp từng nhóm người dùng**
Pricing là nơi tâm lý học gặp doanh thu. Vấn đề không phải là bạn nghĩ app của mình đáng giá bao nhiêu, mà là mỗi nhóm người dùng *cảm nhận* giá trị đó như thế nào.
Ví dụ, Zach Shakked, nhà sáng lập [**Hashtag Expert**](https://www.hashtag.expert/), nhận ra rằng gói $20/năm hoạt động kém hiệu quả trên Snapchat và TikTok, nơi người dùng trẻ tuổi hơn và thích các khoản thanh toán nhỏ, linh hoạt. Khi chuyển sang mô hình **$2.99/tuần**, kết quả đã thay đổi hoàn toàn: conversion cao hơn và tổng doanh thu tăng đáng kể.
**Bài học rút ra**
- Luôn cung cấp ba gói pricing.
- Sử dụng hiệu ứng “chim mồi” để khiến gói bạn muốn người dùng chọn trở nên hấp dẫn hơn.
- Điều chỉnh pricing theo kênh acquisition, cấp quốc gia hoặc loại người dùng.
- Thêm các ưu đãi theo mùa, giảm giá cho người dùng mới và chương trình khuyến khích quay lại cho nhóm churned.
[]
👉 **Mẹo nhỏ:** Hãy xem pricing như một phần nội dung quảng cáo, liên tục test và tối ưu dựa trên hành vi người dùng.
## **3. Onboarding tốt giúp thu hút người dùng nhanh chóng**
Một sự thật đáng buồn: phần lớn người dùng đưa ra quyết định mua trong** lần truy cập đầu hoặc thứ hai**. Nếu onboarding thất bại, cơ hội chuyển đổi coi như mất.
[]
Đây là lúc marketer và product manager cần phối hợp chặt chẽ. Một màn hình đăng ký tải chậm, thông điệp free trial gây hiểu nhầm hoặc thiếu phần giá trị cốt lõi ngay từ đầu đều có thể khiến tỷ lệ chuyển đổi từ trial sang trả phí sụt giảm nghiêm trọng.
Một subscription app đã tăng gấp đôi doanh thu lên **2.000 USD mỗi tháng** chỉ bằng cách tối ưu luồng onboarding và paywall mà không cần thêm lượt tải hay chi tiêu quảng cáo nào.
[]
**Điều cần lưu ý **
- Giữ quy trình đăng ký ngắn gọn, trực quan.
- Làm nổi bật giá trị ngay từ đầu, chẳng hạn tính năng tracking tiến độ cho fitness app hoặc gợi ý kết đôi cho dating app.
- Sử dụng deep link để dẫn người dùng trial trực tiếp đến tính năng có giá trị cao.
- Tạo thói quen sử dụng thông qua thông báo hằng ngày, giúp người dùng gắn bó trước khi trial kết thúc.
👉 **Mẹo nhỏ:** Hãy xem onboarding như một chiến dịch tăng trưởng. Mỗi bước đều là một phần trong hành trình chuyển đổi.
## **4. ASO và ASA tốt để hỗ trợ lẫn nhau**
Mức độ hiển thị cao là yếu tố sống còn với subscription app. Bạn không thể tạo chuyển đổi nếu người dùng không tìm thấy bạn. Đây là lúc [**App Store Optimization (ASO)**](https://www.airbridge.io/glossary/app-store-optimization-aso) và **Apple Search Ads (ASA)** phát huy sức mạnh khi kết hợp.
ASO là chiến lược tăng trưởng bền vững theo thời gian, còn ASA là động lực tăng tốc. Khi được triển khai song song, ASA giúp cải thiện các chỉ số như tap-through rate và conversion rate, từ đó nâng cao thứ hạng ASO. Đây chính là một vòng tăng trưởng khép kín.
Với marketer của các subscription app, điều này có nghĩa là ngân sách không nên tách ASA và ASO thành hai mảng riêng biệt. Chiến lược từ khóa được test trong ASA có thể xác thực xem liệu từ khóa đó có nên được ưu tiên trong ASO hay không.
👉 **Mẹo nhỏ: **Đừng chỉ chạy ASA để thu hút người dùng. Hãy dùng nó như công cụ testing cho ASO, từ khóa chất lương trong ASA thường cũng sẽ có lợi về mặt organic.
## **5. Paid UA tốt tập trung vào đúng sự kiện của người dùng**
Đúng vậy! subscription app phát triển nhờ paid UA. Nhưng thời của việc chỉ tối ưu install đã qua. Các growth team giờ đây tập trung tối ưu những sự kiện sâu hơn trong funnel, như trial start, paywall view, hay predicted LTV.
**Những chuyên gia UA giỏi nhất không bao giờ chấp nhận các chiến dịch chung chung. Họ:**
- Xây dựng nhiều nhóm lookalike audience khác nhau để test.
- Liên tục làm mới nội dung quảng cáo (creative) để tránh tình trạng gây chán.
- Phân tách theo nền tảng, như Meta, TikTok hoặc Google, thay vì chạy quảng cáo đại trà.
- Tối ưu dựa trên những sự kiện thực sự liên quan đến doanh thu.
Một trong những sai lầm phổ biến nhất là chi mạnh để có được “cheap install” nhưng không chuyển đổi. Những gì trông có vẻ hiệu quả về CPI thường lại là khoản lỗ nếu xét theo góc độ subscription.
👉 **Mẹo nhỏ: **Hãy định nghĩa lại chỉ số thành công. Đừng chỉ ăn mừng lượt cài đặt, hãy tập trung đo lường trial-to-paid hoặc predicted LTV.
## **Hiệu ứng cộng hưởng khi tối ưu hóa**
Điều thú vị là: bạn không cần chi hàng triệu đô cho UA để đạt được tăng trưởng.
Trong buổi chia sẻ [Jetstream session ](https://www.youtube.com/watch?v=9yjiaeqc_-s)cùng Airbridge, **Steve P. Young từ App Masters** đã ghi nhận nhiều app tăng trưởng đáng kinh ngạc:
- **Từ $250 lên $1.000/tháng **chỉ với 15 lượt tải mỗi ngày
- **Từ $30 lên $2.000/tháng **chỉ với 50 lượt tải mỗi ngày
Bí quyết không nằm ở việc tăng chi tiêu quảng cáo, mà ở việc tối ưu paywall và onboarding.
Đó chính là hiệu ứng “multiplier”, khi mỗi yếu tố khuếch đại tác động của yếu tố khác. Onboarding tốt hơn giúp tăng trial start, paywall tối ưu hơn giúp cải thiện conversion, pricing thông minh giúp tăng ARPU, và UA hiệu quả quay lại tiếp sức cho toàn bộ vòng tăng trưởng.
👉 Nhấn [tại đây ](https://www.airbridge.io/vi/blog/subscription-app-growth-guide)để đọc toàn bộ bài “Subscription App Growth Guide: From Concept to Early Success to Scale.”
## Tổng kết
Subscription app là một trong những nhóm năng động nhất của hệ sinh thái app. Chúng có thể mở rộng quy mô toàn cầu mà không gặp rào cản logistics, mang lại dòng doanh thu định kỳ mà nhà đầu tư đặc biệt yêu thích. Đồng thời, việc đo lường cũng dễ dàng hơn so với thương mại, bởi công thức tính CAC và LTV rất rõ ràng.
Tuy nhiên, thành công không đến từ việc sao chép Calm, Noom hay Headspace, mà đến từ việc triển khai kỷ luật 5 yếu tố cốt lõi:
- Luôn test và tối ưu paywall
- Phân khúc và tinh chỉnh pricing
- Xem onboarding là một phần của chiến lược growth
- Kết hợp ASA và ASO để tạo hiệu ứng cộng hưởng
- Tối ưu UA dựa trên giá trị thật sự
👉 **Mẹo nhỏ:** Hãy xem 5 yếu tố này như những lực cộng hưởng, mỗi yếu tố đều khuếch đại tác động của yếu tố còn lại.
Khi làm đúng, bạn không chỉ đang chạy các chiến dịch, mà đang xây dựng một subscription business có khả năng tăng trưởng kép.
## Tăng trưởng bền vững cùng Airbridge
[**Airbridge**](https://abr.ge/1i8vxa) mang đến cho các đội ngũ growth của subscription app khả năng theo dõi toàn phễu (full-funnel visibility) để đo lường, tối ưu và mở rộng những gì đang hoạt động hiệu quả.
- Tối ưu Apple Search Ads (ASA) với dữ liệu từ khóa real-time để nhắm đúng đối tượng.
- Khởi chạy và theo dõi chiến dịch UGC trên TikTok, Meta và nhiều nền tảng khác.
- Hợp nhất hành trình người dùng web và app với Web & Mobile SDK mạnh mẽ của Airbridge.
- Theo dõi conversion từ install đến subscription trong một giao diện thống nhất.
- Sử dụng dữ liệu SKAN để tối ưu hiệu suất trial-to-paid trên iOS.
- Retarget người dùng có nguy cơ churn bằng segmentation dựa trên hành vi.
- Tích hợp với RevenueCat và Adapty để phân tích LTV chính xác và toàn diện hơn.
---
## Xu hướng mới 2025: Hybrid Monetization trong subscription app là gì?
URL: https://www.airbridge.io/vi/blog/what-is-hybrid-monetization-in-subscription-apps-2025
Language: Vietnamese
Published: 2025-09-29
## **Giới thiệu chung**
Mô hình hybrid monetization từ lâu đã phổ biến với các ứng dụng game, nơi in-app ads (IAA), in-app purchase (IAP) và lượt đăng ký (subscription) cùng tồn tại để tối đa hóa doanh thu. Ngược lại, các ứng dụng subscription trước đây gần như chỉ dựa hoàn toàn vào nguồn thu từ phí đăng ký định kỳ.
Tuy nhiên, bối cảnh kinh tế của ngành mobile app đang thay đổi. Chi phí user acquisition tăng cao, quy định về quyền riêng tư khiến dữ liệu trở nên hạn chế, và người dùng ngày càng “mệt mỏi vì phải subscribe.” Với các ứng dụng subscription, mô hình hybrid monetization giờ không còn chỉ ở bước thử nghiệm, mà đã trở thành điều kiện tiên quyết để tăng trưởng bền vững.
## **Vì sao mô hình kiếm tiền của subscription apps đang thay đổi?**
Trong nhiều năm, mô hình “subscription-first” hoạt động rất hiệu quả. Doanh thu định kỳ ổn định giúp nhà đầu tư yên tâm, và việc dự đoán churn cohort cũng khá dễ dàng. Tuy nhiên, ba yếu tố chính đang buộc ngành phải thay đổi cách tiếp cận:
1. **Chi phí UA tăng**: Trên iOS, chi phí acquisition tăng vọt sau khi Apple triển khai ATT và loại bỏ user-level identifier. Các chiến dịch giờ đây phải đạt LTV cao hơn mới có thể hòa vốn.
2. **Sự mệt mỏi vì subscription**: Người dùng đang phải trả quá nhiều khoản phí định kỳ, từ streaming, fitness đến productivity và wellness, khiến họ thận trọng hơn với việc đăng ký mới.
3. **Cạnh tranh từ nền tảng:** Apple One và Google Play Pass đã thay đổi cuộc chơi. Khi người dùng có thể trả phí cố định để truy cập nhiều app trong hệ sinh thái cùng lúc, các app độc lập phải chứng minh giá trị riêng để xứng đáng được trả tiền subcription.
Các xu hướng ngành đã xác nhận sự chuyển dịch này. Báo cáo của [Liftoff](https://info.liftoff.io/2025-non-gaming-ad-monetization-trends) dự đoán doanh thu quảng cáo ngoài game sẽ vượt qua game vào năm 2026, một cột mốc mang tính lịch sử. Trong khi đó, báo cáo [*State of Subscription Apps 2025* của RevenueCat](https://www.revenuecat.com/state-of-subscription-apps-2025) nhấn mạnh hybrid model là một trong những đòn bẩy tăng trưởng hàng đầu cho các doanh nghiệp subscription.
Tóm lại, những gì ngành game đã làm được từ nhiều năm trước giờ đây đang trở thành “playbook” cho các ứng dụng subscription.
## **Lợi ích của mô hình Hybrid Monetization cho các subscription apps**
Hybrid monetization mang lại sự linh hoạt. Thay vì chỉ dựa vào một nguồn doanh thu, các app có thể kết hợp nhiều hình thức như app trả phí, subscription, IAP, quảng cáo, affiliate partnership hoặc thương mại điện tử, để phù hợp hơn với hành vi người dùng.
[]
Không phải người dùng nào cũng sẵn sàng trả phí đăng ký. Một số chấp nhận xem quảng cáo, số khác sẽ mua nội dung riêng lẻ, và có những người chỉ nâng cấp sau vài tháng tương tác. Hybrid monetization mở ra nhiều điểm chạm trong revenue funnel, nhưng vẫn giữ subscription làm trụ cột.
- **Revenue diversification**: Reduce risk when one channel underperforms.
- **User-centric design**: Let free users create value through ads while giving engaged users premium paths.
- **Higher LTV**: With rising UA costs, maximizing monetization from every user segment matters more than ever.
- **Adaptability**: Hybrid allows apps to pivot quickly when one revenue stream underperforms, ads during downturns, and subscriptions during stronger engagement cycles.
- **Competitive edge**: Offering multiple ways to engage and pay helps apps expand reach while still monetizing effectively.
- **Đa dạng hóa doanh thu**: Giảm rủi ro khi một kênh hoạt động kém hiệu quả.
- **Thiết kế hướng tới người dùng**: Cho phép người dùng miễn phí tạo giá trị doanh thu qua quảng cáo, đồng thời cung cấp gói cao cấp cho nhóm người dùng gắn bó lâu dài.
- **Tăng LTV**: Khi chi phí UA tăng, việc tối ưu doanh thu từ từng nhóm người dùng trở nên quan trọng hơn bao giờ hết.
- **Tính linh hoạt**: Hybrid giúp app nhanh chóng chuyển hướng khi một nguồn thu suy giảm, tận dụng quảng cáo khi engagement thấp và đẩy mạnh subscription khi tương tác cao.
- **Lợi thế cạnh tranh**: Nhiều cách tương tác và thanh toán giúp app mở rộng tệp người dùng nhưng vẫn tối ưu doanh thu hiệu quả.
## **Ví dụ thực tế về Hybrid Monetization trong subscription apps**
Hybrid monetization không còn là lý thuyết mà đã được ứng dụng rộng rãi trên nhiều lĩnh vực như:
**Education Apps**
- **Duolingo** là ví dụ điển hình cho hybrid monetization. Phiên bản miễn phí của app được hỗ trợ bởi quảng cáo, giúp tạo doanh thu từ nhóm người học thông thường. Trong khi đó, gói trả phí Duolingo Super loại bỏ quảng cáo và bổ sung các tính năng nâng cao như lượt học không giới hạn và học offline. Mô hình này giúp Duolingo tạo doanh thu ổn định từ cả người dùng miễn phí và người dùng trả phí.
**Wellness & Fitness**
- **Strava **cung cấp gói subscription cao cấp với các tính năng như phân tích chuyên sâu, lập kế hoạch lộ trình và bảo mật nâng cao. Ngoài ra, ứng dụng còn kiếm tiền thông qua các thử thách được tài trợ bởi thương hiệu, mở rộng nguồn doanh thu vượt ra ngoài subscription truyền thống.
**Dating Apps **
- **Tinder** kết hợp nhiều gói subscription như Tinder Plus, Gold và Platinum với các giao dịch trả phí một lần như Boosts và Super Likes. Cách kết hợp này vừa đảm bảo dòng doanh thu định kỳ ổn định, vừa tận dụng được các khoản chi tiêu bổ sung theo từng giao dịch.
**Music & Entertainment**
- **YouTube** vận hành trên nhiều nguồn doanh thu khác nhau. Người dùng có thể truy cập miễn phí với quảng cáo, đăng ký gói premium để trải nghiệm không quảng cáo, hoặc thực hiện in-app purchase thông qua các tính năng như Super Chat và Memberships. Cấu trúc nhiều lớp này giúp YouTube khai thác doanh thu từ đa dạng nhóm người dùng và mục đích sử dụng.
- **Wattpad** cung cấp gói Premium cho phép đọc không quảng cáo và truy cập offline, đồng thời kiếm tiền từ mô hình coin, người dùng sử dụng coin để mở khóa truyện trả phí, cùng với quảng cáo dành cho người dùng miễn phí.
[]
Xu hướng này thể hiện rõ ràng: việc kết hợp nhiều hình thức kiếm tiền giúp ứng dụng đáp ứng hành vi và nhu cầu khác nhau của người dùng.
## **Best Practices khi áp dụng Hybrid Monetization cho subscription apps**
Hybrid monetization không chỉ đơn giản là thêm quảng cáo, cách triển khai mới là yếu tố quyết định. Các đội ngũ nên tập trung vào:
- **Phân khúc người dùng**: Người dùng thường xuyên có xu hướng đăng ký subscription, trong khi người dùng với tần suất ít hơn có thể chấp nhận quảng cáo.
- **Đề xuất theo ngữ cảnh**: Đề xuất nâng cấp (upsell) sau các cột mốc như hoàn thành khóa học hoặc duy trì streak.
- **Cân bằng quảng cáo**: Rewarded video và native ad mang lại hiệu quả tốt hơn banner gây gián đoạn.
- **A/B testing**: Thử nghiệm các gói bundle, thời điểm upsell và tần suất quảng cáo.
- **Minh bạch**: Truyền thông rõ ràng về giá và tần suất hiển thị quảng cáo giúp xây dựng niềm tin.
- **Hợp tác thương hiệu**: Các chương trình affiliate hoặc brand partnership có thể tạo thêm nguồn doanh thu mà không gây khó chịu (ví dụ, Calm hợp tác với American Airlines để cung cấp nội dung thiền và thư giãn trên chuyến bay).
Hybrid monetization đã trở thành bước tiến lớn của các subscription apps. Chi phí acquisition tăng, quy định quyền riêng tư nghiêm ngặt hơn, cùng sự thay đổi trong kỳ vọng của người dùng đòi hỏi một mô hình vừa linh hoạt vừa bền vững. Bằng cách kết hợp quảng cáo, subscription, in-app purchase và hợp tác thương hiệu, các apps có thể mở rộng tệp người dùng, tối đa hóa giá trị trên từng phân khúc và xây dựng tăng trưởng bền vững trong thị trường cạnh tranh.
### **Tăng trưởng bền vững cùng Airbridge**
- Tối ưu Apple Search Ads (ASA) ở cấp độ từ khóa bằng phân tích real-time để nhắm đúng đối tượng.
- Thúc đẩy chiến dịch UGC trên TikTok, Meta và các nền tảng khác nhằm tăng lượt cài đặt và conversion.
- Tối ưu hiệu quả web-to-app với Web và Mobile SDK mạnh mẽ của Airbridge, giúp theo dõi hành trình người dùng liền mạch.
- Theo dõi funnel conversion rate qua các giai đoạn chính như onboarding, sign-up, bắt đầu trial, subscription và re-subscription để phát hiện điểm cần cải thiện.
- Sử dụng SKAN để phân tích hiệu suất iOS, đặc biệt trong giai đoạn trial start.
- Retarget thông minh bằng cách xây dựng behavioral audience cho nhóm người dùng rời bỏ hoặc churned.
- Tích hợp với RevenueCat và Adapty để có góc nhìn toàn diện về funnel và phân tích chi tiết hơn.
---
## Hướng dẫn Tăng trưởng cho Subscription App: Từ Ý tưởng đến đến Mở rộng quy mô
URL: https://www.airbridge.io/vi/blog/subscription-app-growth-guide
Language: Vietnamese
Published: 2025-09-25
Thị trường ứng dụng subscription đang trở nên khốc liệt hơn bao giờ hết. Người dùng có vô số lựa chọn từ lĩnh vực thể hình, chăm sóc sức khỏe tinh thần, ứng dụng hẹn hò cho đến các app chuyên về hỗ trợ năng suất làm việc, trong khi các marketer lại phải đối mặt với chi phí thu hút người dùng (acquisition cost) ngày càng cao, quy định quyền riêng tư nghiêm ngặt hơn và áp lực chứng minh tăng trưởng bền vững.
Trong phiên Jetstream này, CEO [Airbridge](https://www.airbridge.io/) – Roi Nam đã có buổi trò chuyện cùng Steve Young, Founder của [App Masters](https://appmasters.com/) và là một trong những chuyên gia uy tín nhất trong lĩnh vực tăng trưởng ứng dụng subscription. Với hơn 10 năm kinh nghiệm về ASO, Apple Search Ads (ASA), định giá, và paywall, Steve đã cố vấn giúp hàng trăm ứng dụng phát triển dựa trên kinh nghiệm thực chiến của mình.
Trong cuộc thảo luận, Steve đã chia sẻ nhiều bài học về cách kiểm chứng ý tưởng, dấu hiệu cho thấy khả năng thành công ở giai đoạn đầu của app, các chỉ số để mở rộng quy mô, và đặc biệt là chiến lược mà anh gọi là “The Greatest Growth Hack in Da World.”
🎥 **Xem đầy đủ buổi phỏng vấn tại đây**
[]
## **Kiểm chứng ý tưởng cho ứng dụng subscription**
Mọi hành trình đều bắt đầu bằng một ý tưởng đúng đắn — một ý tưởng giải quyết được nỗi đau thực sự của người dùng, thay vì chỉ tạo ra một tính năng “có thì cũng tốt.”
> “Bạn có thể tạo một ứng dụng thiền, hoặc một ứng dụng giúp giảm lo âu. Cả hai đều có người cần đến, nhưng chỉ một trong số chúng thật sự giải quyết vấn đề. Build the painkiller, not a vitamin.”
— Steve Young, Founder App Masters
Những vấn đề mà người dùng thật sự có nhu cầu giải quyết thường thu hút người dùng sẵn sàng đăng ký dùng thử, trả tiền, và gắn bó lâu dài hơn. Để phát hiện các cơ hội này, Steve thường sử dụng Google Trends, AppFigures, và Sensor Tower để tìm những từ khóa đang tăng trưởng trước khi thị trường trở nên quá đông đúc.
> “Nếu bạn tìm được chỉ cần hai từ khóa có điểm tìm kiếm khoảng 30 và giành vị trí top, bạn đã nắm trong tay cơ hội kiếm tiền lên tới sáu con số.”
— Steve Young, Founder App Masters
Khi ý tưởng đã được kiểm chứng, câu hỏi tiếp theo là liệu ứng dụng có thể phát triển hơn chỉ là một ý tưởng tiềm năng không.
## **Dấu hiệu sớm cho thấy ứng dụng subscription có tiềm năng tăng trưởng**
Sau khi ra mắt, chỉ cần nhìn vào một vài chỉ số là có thể nhận ra liệu ứng dụng có đủ tiềm năng để mở rộng hay không. Những benchmark quan trọng nhất gồm có:
- Tỷ lệ chuyển đổi từ cài đặt sang dùng thử (Install-to-trial): 5–10%
- Tỷ lệ chuyển đổi từ dùng thử sang trả phí (Trial-to-paid): 30–40%
- Số lượt cài đặt tự nhiên mỗi ngày: 50–100 (để có đủ dữ liệu phục vụ tối ưu hóa)
[]
Ở giai đoạn này, Steve cho rằng chưa cần đến những công cụ subscription phức tạp, App Store Connect đã đủ để theo dõi toàn bộ tiến trình và kết quả.
> “Nếu tôi chi $1 cho mỗi lượt cài và đạt 70% ROAS, tôi đã có thể hài lòng. Dù chưa có lợi nhuận, điều đó cho thấy ứng dụng đã có đà để tôi tối ưu tiếp.”
— Steve Young, Founder App Masters
Khi các chỉ số trên đạt kỳ vọng, đó là dấu hiệu mạnh mẽ cho thấy ứng dụng đã sẵn sàng bước vào giai đoạn tăng trưởng thực sự. Và lúc này, thử thách tiếp theo chính là mở rộng quy mô mà vẫn giữ hiệu quả.
## **Mở rộng ứng dụng subscription với ASO, ASA và Meta Ads**
Khi những tín hiệu ban đầu trở nên khả quan, trọng tâm tiếp theo là mở rộng tăng trưởng. Steve nhấn mạnh rằng chiến lược cần đơn giản, dễ lặp lại và được xây dựng từng bước**.**
Hầu hết các ứng dụng thành công đều đi theo một lộ trình tương tự:
1. Bắt đầu với ASO để thu hút người dùng có nhu cầu cao thông qua tìm kiếm.
2. Sau đó thêm ASA để mở rộng khả năng hiển thị và khám phá.
3. Cuối cùng, khi hiệu quả tài chính cho phép, mở rộng sang Meta Ads để tiếp cận rộng hơn.
Tuy nhiên, một lỗi phổ biến mà nhiều team mắc phải là ngừng chạy Meta khi kết quả có vẻ chững lại. Steve cảnh báo rằng điều này có thể bỏ lỡ tác động gián tiếp nhưng rất mạnh mẽ mà Meta mang lại.
> “Luôn có một nhóm người thấy quảng cáo của bạn, không nhấp vào, nhưng lại lên App Store để tìm kiếm ứng dụng. Lượng tìm kiếm thương hiệu này sẽ giúp tăng thứ hạng ASO.”
— Steve Young, Founder App Masters
Steve gọi đây là “hiệu ứng cộng hưởng tự nhiên” (organic multiplier). Các chiến dịch trả phí tạo nhận diện thương hiệu, nhận diện này dẫn đến lượt tìm kiếm tự nhiên, và chính những lượt tìm kiếm đó lại đẩy thứ hạng ASO cao hơn.
Đó cũng là lý do tại sao ứng dụng Calm và Headway không chỉ đứng đầu kết quả tìm kiếm cho từ khóa “meditation” mà còn thống trị ngay cả với tên thương hiệu của chính họ.
> “Với Headway và Calm, lượng tìm kiếm thương hiệu trong ASO còn lớn hơn cả từ khóa ‘meditation’.”
— Steve Young, Founder App Masters
## **Danh sách kiểm tra Paywall cho ứng dụng subscription**
Paywall vẫn luôn là đòn bẩy mạnh nhất để thúc đẩy tăng trưởng doanh thu cho các ứng dụng subscription. Steve cho rằng, cách nhanh nhất để kiếm thêm tiền chính là đảm bảo mọi người dùng đều nhìn thấy paywall.
> “Cách nhanh nhất để kiếm nhiều tiền hơn là đảm bảo 100% người dùng đều thấy paywall.”
— Steve Young, Founder App Masters
Để làm đúng điều này, Steve khuyên nên bắt đầu bằng một vài câu hỏi đơn giản:
- Bạn đã giải thích rõ giá trị của ứng dụng chưa?
- Bạn có thêm yếu tố social proof (như đánh giá, số lượt tải, người dùng nổi tiếng...) không?
- Bạn đã thu thập thông tin cơ bản của người dùng (ví dụ: mục tiêu thể hình, ngôn ngữ ưa thích, lĩnh vực quan tâm) để cá nhân hóa paywall chưa?
- Paywall có xuất hiện với tất cả người dùng ngay khi họ cài đặt không?
Sau khi đảm bảo các yếu tố cơ bản này, bước đột phá nằm ở quá trình thử nghiệm**.** Dưới đây là một vài ví dụ mà Steve chia sẻ:
- **Cấu trúc gói giá:** Một ứng dụng gặp khó với gói $50/năm và 7 ngày dùng thử đã tăng doanh thu đáng kể khi chuyển sang gói tuần +3 ngày dùng thử, kèm một gói năm không dùng thử. Ngạc nhiên là sau thay đổi này, nhiều người dùng chọn gói năm hơn.
- **Cách hiển thị giá:** Thay đổi cách trình bày từ “$60/năm” sang “$5/tháng (trả hàng năm)” giúp tỷ lệ chuyển đổi tăng 38%.****
- **Paywall nhiều trang**: Hiển thị song song gói tháng và gói năm giúp người dùng dễ so sánh và ra quyết định nhanh hơn.
[]
## **Chiến lược thu hút người dùng (UA) tiết kiệm cho ứng dụng subscription**
Với các ứng dụng subscription giai đoạn đầu, ngân sách marketing thường rất hạn chế. Vì vậy, Steve khuyên nên bắt đầu với Apple Search Ads (ASA) — kênh này dựa trên từ khóa, có intent cao, và dễ quản lý. Từng có trường hợp một khách hàng chi $800 cho ASA và thu về $3,000 — trong đó $2,000 đến trực tiếp từ ASA và $1,000 từ tăng trưởng organic.
Khi ASA đã chạy ổn định, bước tiếp theo tự nhiên là Meta Ads. Steve gợi ý bắt đầu với chiến dịch broad, sau đó phân nhóm theo sở thích khi có thêm dữ liệu. Tuy nhiên, “chìa khóa thực sự” lại nằm ở creative.
> “Video hiệu quả nhất của một khách hàng tôi từng làm việc được tạo hoàn toàn bằng AI. Rẻ hơn mà hiệu quả không kém gì UGC.”
— Steve Young, Founder App Masters
AI giờ đây đã trở thành “vũ khí bí mật” cho các ứng dụng subscription muốn thử nghiệm nội dung quảng cáo nhanh mà không tốn kém.
Đội ngũ của Steve thường dùng Eleven Labs để chuyển văn bản thành giọng nói, Creatify và Arcads để tạo biến thể quảng cáo đa dạng, Reel.farm để thêm các cảnh reaction giúp video hấp dẫn hơn. Nhờ những công cụ này, ngay cả những team có quy mô nhỏ cũng có thể thử nghiệm hàng chục ý tưởng quảng cáo với chi phí chỉ bằng một phần nhỏ so với sản xuất truyền thống.
Với các ứng dụng có ngân sách hạn chế, AI chính là “mã gian lận” hợp pháp — giúp việc thử nghiệm sáng tạo trở nên vừa nhanh, vừa khả thi, vừa bền vững.
## **The Greatest Growth Hack in Da World**
Khi được hỏi về “bí kíp tăng trưởng” tâm đắc nhất, Steve không cần suy nghĩ lâu. Anh gọi nó là “The Greatest Growth Hack in Da World” — chiến dịch lifetime offer.
Cách làm rất đơn giản: tạo gói lifetime miễn phí hoặc giảm giá trong thời gian giới hạn, rồi quảng bá nó trên các nền tảng khám phá ứng dụng như *AppAdvice* hoặc *IndieAppSanta.* Và kết quả thường… bùng nổ.
Ví dụ, một khách hàng của Steve có ứng dụng ghi chú bằng AI đã thử chiến dịch này và đạt được kết quả mà hầu hết các chiến lược UA trả phí đều không thể sánh được.** **Chỉ trong 2 ngày, ứng dụng thu về 106.000 lượt cài đặt và 800 đánh giá mới. Doanh thu hàng tháng tăng từ vài trăm đô lên 7.000 đô, và vượt 90.000 đô sau một năm.
> “Còn kênh nào giúp bạn tăng trưởng mạnh đến vậy mà không cần chi tiền quảng cáo chứ?”
— Steve Young, Founder App Masters
Điều khiến chiến lược này mạnh mẽ không chỉ là doanh thu, mà là tác động lan tỏa của nó.
Khi lượng cài đặt tăng đột biến, lượng đánh giá cũng tăng theo, App Store visibility được cải thiện, và ASO được thúc đẩy nhờ tốc độ tải ứng dụng tăng mạnh.
Dù không phải lúc nào cũng thành công, nhưng nếu đúng thời điểm, chiến dịch này có thể thay đổi hoàn toàn quỹ đạo phát triển của một ứng dụng subscription chỉ sau vài ngày.
## **Nền tảng cốt lõi trong tăng trưởng ứng dụng subscription**
Khi nhìn về tương lai, Steve vẫn lạc quan về tiềm năng của các ứng dụng subscription. Tuy nhiên, anh nhấn mạnh rằng những nguyên tắc nền tảng vẫn không hề thay đổi kể từ hơn mười năm trước, khi anh mới bước chân vào ngành này.
> “Tập trung tốt hơn phân tán. Kiên trì thắng ngắn hạn. Đơn giản hiệu quả hơn phức tạp.”
— Steve Young, Founder App Masters
Dù bạn là nhà sáng lập tự thân hay marketer của một team đang mở rộng, playbook thành công vẫn giống nhau:
- Kiểm chứng ý tưởng bằng ASO
- Xác nhận đà tăng trưởng bằng ASA
- Tái đầu tư bằng Meta Ads
- Và không bao giờ ngừng thử nghiệm paywall.
Tăng trưởng bền vững không đến từ việc chạy theo mọi chiến thuật mới, mà từ việc làm thật tốt những điều cơ bản và để thời gian nhân sức chúng lên.
Bạn muốn có nhiều insights chi tiết hơn về subscription app, hãy truy cập[ App Masters YouTube channel](https://www.youtube.com/appmasters?utm_source=chatgpt.com) hoặc[ appmasters.com](https://appmasters.com?utm_source=chatgpt.com). Bạn cần đo lường và tối ưu tăng trưởng subscription app, đừng bỏ lỡ [Airbridge](https://www.airbridge.io/).
### **Tăng trưởng ứng dụng subscription của bạn nhanh hơn cùng Airbridge 🚀 **
- Tối ưu Apple Search Ads (ASA) theo từ khóa, phân tích real-time để nhắm đúng tệp người dùng.
- Tăng hiệu quả chiến dịch UGC trên TikTok, Meta và các nền tảng khác để thúc đẩy lượt cài đặt và chuyển đổi.
- Đồng bộ chiến dịch web-to-app với Web & Mobile SDK mạnh mẽ, theo dõi hành trình người dùng liền mạch.
- Giám sát tỷ lệ chuyển đổi Phễu ở các giai đoạn như onboarding, đăng ký, bắt đầu trial, đăng ký gói, và đăng ký lại để phát hiện điểm rơi.
- Khai thác dữ liệu SKAN để hiểu rõ hiệu suất iOS, đặc biệt ở giai đoạn trial start.
- Nhắm mục tiêu lại hiệu quả bằng cách xây dựng nhóm hành vi cho người dùng đã drop hoặc churn.
- Tích hợp với RevenueCat và Adapty để có cái nhìn toàn Phễu và tối ưu hóa chiến lược tăng trưởng.
---
## Xu hướng AI trong ngành game 2025: UA định hướng sản phẩm, creative và chiến lược monetization
URL: https://www.airbridge.io/vi/blog/2025-ai-trend-in-gaming-ua-driven-products-ad-creatives-monetization
Language: Vietnamese
Published: 2025-09-16
Mobile marketing luôn gắn với những biến động lớn. App Tracking Transparency (ATT) của Apple khiến việc theo dõi trở nên hạn chế, trong khi SKAN lại mang đến nhiều giới hạn và độ trễ. Giờ đây, AI đang định hình lại toàn bộ cách các đội ngũ xây dựng sản phẩm, triển khai marketing và khai thác doanh thu từ game. Những studio có thể tăng trưởng trong năm 2025 chỉ khi họ thiết kế hệ thống đủ linh hoạt để phát triển trong môi trường ngày càng thay đổi.
Trong podcast Brave New Digital World của Phiture, [Roi Nam](https://www.linkedin.com/in/roinam/), CEO của Airbridge và Airflux, đã trò chuyện cùng [Andy Carvell](https://www.linkedin.com/in/andycarvell/) về nhiều chủ đề nóng, bao gồm việc UA ngày càng trở thành động lực thúc đẩy phát triển sản phẩm, sự phức tạp ngày càng tăng của đo lường iOS, và cách AI đang thay đổi creative lẫn mô hình monetization.
🎧 Theo dõi toàn bộ cuộc trò chuyện trên [Apple Podcasts](https://podcasts.apple.com/us/podcast/brave-new-digital-world/id1814863716) và [Spotify](https://open.spotify.com/show/6PuaiCBX1I0u11ywEBMeJM?si=29fb16c604c14300).
## **Dựa vào UA để định hướng sản phẩm, thay vì phát triển mù quáng**
Trong nhiều năm, công thức đã luôn rất đơn giản: phát triển game trước, marketing sau. Tuy nhiên, mô hình này đã dần trở nên lỗi thời trong thời đại dữ liệu mới. Ngân sách phát triển ngày càng thắt chặt, cạnh tranh khốc liệt, chi phí dành cho một concept được đầu tư sai là quá lớn. Trong thế giới mobile gaming, quy tắc đã được viết lại bởi tốc độ và dữ liệu. Các đội ngũ giờ thường tung ra bản build rút gọn để kiểm tra CPI, retention D1, conversion từ thời lượng chơi game và free trial, trước khi đầu tư phát triển phiên bản đầy đủ hơn.
Các studio về dòng game hypercasual thậm chí còn tung nhiều prototype cùng lúc và dựa vào dữ liệu để quyết định sản phẩm win. Với subscription app, quá trình thường bắt đầu bằng việc test ASO để xác thực nhu cầu trước khi viết code. Logic ở đây rất rõ ràng: UA data có thể cho thấy chỉ trong vài ngày liệu người chơi có hứng thú hay không, vì thế chẳng có lý do gì phải mất hàng tháng cho một bản build sai hướng.
> *“UA không chỉ đơn thuần là marketing cho sản phẩm. Nó thực sự đã trở thành một phần trong quá trình phát triển sản phẩm. Bạn không cần phải phân vân trong việc chọn lựa concept phù hợp. Hãy xây dựng 5 concept khác nhau và chọn ra sản phẩm win.” *
— Roi Nam, CEO Airbridge & Airflux
AI đang giúp quá trình này trở nên nhanh hơn nữa. Từ coding, thiết kế đến testing giờ đều có thể tự động hóa, rút ngắn đáng kể khoảng cách từ ý tưởng đến giai đoạn launching. Các team vận hành với quy mô nhỏ giờ có thể vận hành với tốc độ mà trước đây phải cần đến cả một phòng ban lớn.
> *“Chúng tôi hay đùa gọi là vibe coding… tức là code theo vibe. Các framework như React Native hay Flutter, khi kết hợp với hệ thống AI như Cursor và Windsurf, giúp quy trình trở nên dễ dàng và nhanh chóng hơn nhiều. Một developer dày dạn kinh nghiệm giờ có thể launch app chỉ trong 5–7 ngày.” *
— Roi Nam, CEO of Airbridge & Airflux
Với đội ngũ phát triển, thông điệp rất rõ ràng: UA không còn chỉ là việc mở rộng sản phẩm đã hoàn thiện. Nó đã trở thành một công cụ khám phá, định hướng cho studio biết concept nào thực sự xứng đáng để đầu tư.
## **Xu hướng AI thúc đẩy sản xuất creative và chiến lược monetization**
AI đang làm thay đổi toàn bộ quá trình sản xuất ad creatives. Từ banner tĩnh, playable ads cho đến video UGC dạng ngắn, tất cả đều có thể do máy tạo ra, thường mang lại hiệu quả ngang bằng hoặc thậm chí vượt trội so với nội dung do con người làm. Những video theo phong cách của influencer đang tràn ngập TikTok, Reels, Shorts, rất nhiều trong số đó không phải người thật mà là persona do AI tạo dựng. Với công cụ như OpenAI Sora, ranh giới giữa thật và ảo đang dần biến mất.
> *“Quy mô sản xuất ad creative với AI đã và đang ngày càng mở rộng. Không chỉ là hình ảnh tĩnh, mà bao gồm tất cả: từ UGC, Playables đến video, tất cả đều do AI tạo ra.” *
— Roi Nam, CEO of Airbridge & Airflux
Sự thay đổi này cũng đang dần lan sang chiến lược monetization và live ops. Các yếu tố như độ khó, thời điểm chèn quảng cáo, hay cơ chế thưởng vốn được tối ưu thủ công, chậm chạp, đang dần trở thành quá khứ. Việc cá nhân hóa real-time sẽ sớm trở thành chuẩn mực, mang đến trải nghiệm phù hợp cho từng người chơi.
> *“Hiện tại, các yếu tố như độ khó màn chơi, in-game economy hay ad monetization vẫn đang được test thủ công, vừa chậm vừa rời rạc. Sắp tới, tất cả sẽ trở nên dynamic, cá nhân hóa và hoàn toàn tự động nhờ AI, giúp ta có thể điều chỉnh độ khó hoặc thời điểm quảng cáo theo thời gian thực cho từng người chơi.”*
— Roi Nam, CEO of Airbridge & Airflux
Điều này đã dần trở thành hiện thực. [**Airflux**](https://abr.ge/mqmtum) – sản phẩm do Airbridge phát triển – đã ứng dụng AI để tối ưu hóa monetization theo cách dynamic, phân nhóm người chơi dựa trên hành vi và điều chỉnh phân phối quảng cáo real-time nhằm tối đa hóa cả engagement lẫn doanh thu.
## **Đo lường iOS: Khi sự phức tạp trở thành chuẩn mực**
Phát triển định hướng dựa trên UA, ad creatives do AI tạo ra và mô hình monetization thông minh chỉ thực sự hiệu quả khi attribution đủ tin cậy. Đây vẫn là nền tảng của mọi hệ thống tăng trưởng, và cũng là nơi nhiều marketer gặp khó khăn nhất, đặc biệt trên iOS. Kể từ khi ATT ra mắt, dữ liệu đo lường trở nên phân mảnh, SKAN chỉ cung cấp tín hiệu hạn chế và chậm trễ. Những studio có thể tiếp tục scale chính là những đơn vị biết kết hợp nhiều chiến lược thay vì dựa vào một tactic duy nhất.
[]
> “Không có câu trả lời đơn giản. Bạn không thể chỉ dựa vào một chiến lược duy nhất. Điều bắt buộc là phải có một cách tiếp cận nhiều lớp, kết hợp web-to-app cùng các mô hình attribution khác nhau để luôn kịp thích ứng.”
— Roi Nam, CEO of Airbridge & Airflux
Đây chính là lúc vai trò của MMP trở nên quan trọng. Một MMP ngày nay không chỉ dừng lại ở việc track installs. Nó còn mô hình hóa conversions, đồng bộ dữ liệu SKAN với dữ liệu organic và kết nối trực tiếp với ad platforms để duy trì cung cấp feedback, tối ưu hệ thống. Đặc biệt, conversions được mô hình hóa giờ đã trở thành yếu tố then chốt, giúp marketer lấp đầy khoảng trống giữa báo cáo SKAN và performance thực tế.
> ““Dù SKAN có thể hiển thị 100 installs, bạn biết rằng organic multiplier chỉ là 1.5. Đó chính là lý do chúng tôi xây dựng framework conversion được mô hình hóa tại Airbridge, để giúp marketer hiểu rõ hơn về phân bổ installs trong thế giới privacy-first.”
— Roi Nam, CEO of Airbridge & Airflux
👉 Đọc thêm: “Vì sao Data là "siêu năng lực mới" của ngành game: Những gì bạn cần biết về Attribution”
## **Chuẩn bị cho kỷ nguyên hội nhập AI**
Việc ứng dụng AI đã đi qua giai đoạn thử nghiệm và đang bước vào quy trình vận hành doanh nghiệp hằng ngày. Với Airbridge, điều này bao gồm việc phát triển **Airflux** – nền tảng ứng dụng AI để tối ưu hóa monetization trong thời gian thực, cùng với marketing co-pilot, cho phép đội ngũ truy vấn dữ liệu MMP bằng hội thoại thay vì phụ thuộc vào dashboard.
> “AI không nên chỉ là một dự án phụ hay công cụ tiện ích thông thường, mà là một phần trong từng công việc hằng ngày. Trong tương lai, bạn sẽ có thể gõ và trò chuyện trực tiếp với chatbot của MMP thay vì phải đích thân mở dashboard để tra cứu và phân tích dữ liệu.”
— Roi Nam, CEO of Airbridge & Airflux
Đối với game marketers, định hướng đã rất rõ ràng. UA, ad creatives, monetization và measurement đều đang trở thành những tính năng do AI dẫn dắt. Tăng trưởng trong tương lai sẽ phụ thuộc ít hơn vào các chiến lược đơn lẻ, và nhiều hơn vào việc xây dựng hệ thống có khả năng thích ứng nhanh với mọi thay đổi.
👉 Khám phá cách [**Airbridge**](https://abr.ge/1i8vxa) và [**Airflux**](https://abr.ge/mqmtum) có thể đồng hành trong chiến lược tăng trưởng của bạn.
---
## Vì sao Data là "siêu năng lực mới" của ngành game: Những gì bạn cần biết về Attribution
URL: https://www.airbridge.io/vi/blog/why-data-is-gamings-new-superpower
Language: Vietnamese
Published: 2025-09-16
App Tracking Transparency (ATT) của Apple không chỉ làm thay đổi ngành mobile marketing, mà còn viết lại toàn bộ “playbook”. Bốn năm sau, các studio vẫn loay hoay trước tình trạng mất dữ liệu và tín hiệu từ các ad networks ngày càng yếu đi. Tuy vậy, một số vẫn tiếp tục mở rộng kinh doanh có lãi. Bí quyết là gì? Chính là xây dựng hệ thống chấp nhận dữ liệu bị thiếu hụt và dựa vào phương pháp đo lường thông minh hơn.
Như Roi Nam, CEO của Airbridge, chia sẻ trong buổi trò chuyện cùng Joseph Kim trên podcast GameMakers, thành công đến từ những đội ngũ không còn chạy theo “perfect data”, mà thay vào đó xây dựng hệ thống có thể xử lý vấn đề về dữ liệu. Một mobile growth manager tại một midcore game studio đã chia sẻ:
> “Thay vì chờ đợi dữ liệu đầy đủ, chúng tôi đã bắt đầu xây dựng một hệ thống có thể xử lý vấn đề thiếu hụt dữ liệu. Chúng tôi test nhiều hơn. Chúng tôi xác thực bằng MMM. Chúng tôi vận hành CAPI ngay cả khi gặp khó khăn."
Theo Roi, chính tư duy đó đã trở thành tiêu chuẩn cho những studio vẫn đang tăng trưởng hiệu quả
🎧 Nghe toàn bộ bài chia sẻ trên GameMakers, hiện có mặt trên [Apple Podcasts](https://podcasts.apple.com/td/podcast/why-data-is-the-new-superpower-in-gaming/id1541808441?i=1000719648426) và [Spotify](https://open.spotify.com/episode/7ljfun4uAvaMm6hiRxIg34).
[]
### 4. Ngày càng nhiều studio đầu tư vào MMM
Khi tracking ở cấp độ người dùng trở nên kém tin cậy, nhiều đội ngũ đã chuyển sang Marketing Mix Modeling (MMM). Khác với các phương pháp attribution gắn liền với user identifier, MMM an toàn về quyền riêng tư và phù hợp hơn với bối cảnh hiện nay. MMM giúp trả lời những câu hỏi mà MMP hoặc SKAN không thể:
- Doanh thu thực sự đến từ paid UA là bao nhiêu?
- Kênh nào đang tạo ra giá trị gia tăng thực sự, chứ không chỉ đơn thuần được ghi nhận conversion?
- Ngân sách có thể cắt giảm ở đâu mà không ảnh hưởng đến tăng trưởng?
Các nhà phát hành game casual thường có lợi nhất vì chiến dịch của họ đơn giản hơn, ít kênh hơn và dữ liệu sạch hơn. Điều này giúp MMM dễ dàng đưa ra insight đáng tin cậy.
Khi triển khai MMM, các studio thường có hai lựa chọn: thuê vendor hoặc tự xây dựng hệ thống self-serve. Một quy trình self-serve điển hình bao gồm:
1. Chọn một model như Robyn của Meta hoặc Meridian của Google.
2. Xác định biến số kết hợp input marketing (chi phí, impression, install) với các yếu tố ngữ cảnh như tính mùa vụ hoặc xếp hạng trên app store.
3. Hiệu chỉnh model bằng lift study từ Meta hoặc Google để đảm bảo kết quả có cơ sở.
4. Kiểm tra độ chính xác với các chỉ số thống kê như R², MAPE hoặc MCMC convergence để đảm bảo tính tin cậy.
5. Chạy model định kỳ, cập nhật input mỗi 1–3 tháng và hiệu chỉnh lại mỗi 3–6 tháng.
6. Rà soát và tối ưu ngân sách một cách từ tốn, áp dụng insight mà không điều chỉnh quá đà.
> "Một growth lead tại studio puzzle casual từng chia sẻ với tôi, MMM đã giúp họ phát hiện rằng chiến dịch influencer trên YouTube mang lại hiệu quả vượt trội so với rewarded video tại một số khu vực. Nếu chỉ dùng last-touch, chúng tôi sẽ không bao giờ nhìn thấy điều đó.''
— Roi Nam, CEO & Co-Founder của Airbridge & Airflux
[]
### Tối ưu hóa doanh thu thông minh hơn với Airflux
Đo lường chỉ là một nửa câu chuyện. Khi chi phí UA ngày càng tăng, câu hỏi thực sự là: làm thế nào để khai thác nhiều giá trị hơn từ chính những người chơi bạn đã có? Đó là lý do Airbridge phát triển [Airflux](https://www.airflux.ai/) – công cụ AI giúp nâng cao LTV bằng cách triển khai chính sách quảng cáo thông minh ở cấp độ phân khúc. Các studio đã bắt đầu thấy kết quả.
👉 Khám phá cách [Clegames](https://airflux.ai/blog/clegames-interstitial-monetization-ai) và [Treeplla](https://airflux.ai/blog/treeplla-ltv-increase-experiment-automation) ứng dụng Airflux.
### UA và Monetization: Một vòng khép kín
Trong thời gian dài, UA và monetization vận hành tách biệt. UA team tập trung giảm CPI và đạt mục tiêu ROAS, trong khi monetization team tối ưu vị trí quảng cáo và luồng IAP. KPI khác nhau, dashboard khác nhau, gần như không giao thoa.
Mô hình đó giờ không còn hiệu quả. Khi CPI tăng và tín hiệu giảm, tăng trưởng và doanh thu phải vận hành như một vòng khép kín. Những studio hiệu quả nhất đã nắm rõ điều này.
Các DSP hiện đại đang dẫn đầu. AppLovin, chẳng hạn, truyền dữ liệu bidding từ MAX mediation ngược trở lại UA engine. Vòng khép kín này giúp hiệu suất tăng tới 4 lần, không phải vì quảng cáo rẻ hơn mà vì chiến dịch được tối ưu dựa trên kết quả monetization thực tế.
Các studio có thể áp dụng cách tiếp cận tương tự bằng cách điều chỉnh chiến lược chiến dịch theo hành vi người chơi:
- **Người dùng tập trung IAP:** Trì hoãn quảng cáo và khuyến khích mua sớm.
- **Người dùng tập trung IAA:** Hiển thị quảng cáo ngay từ phiên đầu tiên cho nhóm có xu hướng tương tác với quảng cáo.
- **Blended ROAS:** Kết hợp cả hai hướng tiếp cận để tăng trưởng bền vững.
Thách thức nằm ở đo lường. MMP và SKAN được xây dựng trên last-touch attribution, cho biết ai được ghi nhận nhưng không thể hiện giá trị gia tăng thực sự. Đó là lý do ngày càng nhiều đội ngũ xác thực bằng Lift Study và MMM. Những lớp dữ liệu này giúp phân biệt yếu tố nào thực sự tạo doanh thu và yếu tố nào chỉ “đẹp trên giấy”.
> "Nguyên tắc rất đơn giản: DSP ngày càng giỏi trong việc nhận diện người chơi có tiềm năng về mặt IAP hoặc IAA. Công việc của studio là gắn đúng chiến lược chiến dịch với đúng đối tượng."
— Roi Nam, CEO & Co-Founder của Airbridge & Airflux
### Tương lai: Tín hiệu tốt hơn, hệ thống thông minh hơn
Giai đoạn tiếp theo của đo lường sẽ không đến từ một “single source of truth”, mà từ việc kết hợp nhiều phương pháp, mỗi phương pháp bổ sung một lớp insight.
**MMP** như Airbridge mang lại khả năng quan sát real-time trên campaign, creative, channel, country và xu hướng theo OS. **MMM** mang đến góc nhìn dài hạn, cho thấy online và offline media đóng góp vào doanh thu thế nào theo thời gian. **Lift Study** từ Meta, Google hoặc bên thứ ba xác thực tính incrementality, giúp team biết yếu tố nào thực sự tạo tác động.
Tuy nhiên, sự thay đổi lớn hơn nằm ở **kỹ thuật xử lý tín hiệu**. Nhiều dữ liệu hơn không còn đồng nghĩa với hiệu quả tốt hơn. Các studio dẫn đầu đang loại bỏ tín hiệu giả hoặc kém giá trị, tập trung vào tín hiệu chất lượng cao, đã được xác thực – chẳng hạn dữ liệu purchase trên App Store – để cải thiện tối ưu hóa nền tảng.
Đồng thời, AI đang tái định hình cách growth team vận hành. Chúng ta đã thấy:
- **Công cụ creative automation** như Poolday ML, Blay và Incymo
- **Nền tảng agentic **như RetentionX, AppRadar và Pollen VC cho CRM, ASO và ad ops
- **Công cụ ad monetization** như Airflux, sử dụng reinforcement learning để tối ưu chiến lược ở cấp phân khúc
Hãy nhớ rằng, những studio đang bứt phá tập trung vào ba ưu tiên:
- Tín hiệu tốt hơn để tăng sức mạnh phản hồi nền tảng
- Đo lường thông minh hơn với MMM, Lift Study và CAPI
- Sự gắn kết chặt chẽ hơn giữa UA, product và monetization team
> "Không phải là xây lại playbook cũ, mà là thích nghi nhanh hơn tất cả."
— Roi Nam, CEO & Co-Founder của Airbridge & Airflux
Tương lai thuộc về những studio biết thích nghi, thử nghiệm và tái tư duy cách đo lường và tối ưu hóa doanh thu.
👉 Khám phá cách [Airbridge](https://www.airbridge.io/) và [Airflux](https://www.airflux.ai/) trở thành một phần trong chiến lược tăng trưởng của bạn.
---
## Bứt phá tăng trưởng Affiliate Marketing tại Đông Nam Á với tích hợp Airbridge & ACCESSTRADE
URL: https://www.airbridge.io/vi/blog/accesstrade-integration
Language: Vietnamese
Published: 2025-08-14
Airbridge đã hợp tác cùng [ACCESSTRADE](https://accesstrade.vn/) nhằm nâng tầm chiến lược affiliate marketing tại khu vực Đông Nam Á. Sự tích hợp này kết hợp khả năng phân bổ nâng cao và phân tích dữ liệu ở cấp độ người dùng của Airbridge với mạng lưới publisher rộng lớn của ACCESSTRADE, giúp marketers khai thác insight sâu hơn, tối ưu chi phí, và gia tăng hiệu quả affiliate nhanh chóng, bền vững hơn.
## Mạng lưới của ACCESSTRADE thúc đẩy tăng trưởng lượt Conversions
ACCESSTRADE là nền tảng affiliate marketing hàng đầu tại Đông Nam Á, kết nối hơn 1.000 advertisers và hơn 1,5 triệu publishers tại Việt Nam. Mạng lưới này bao gồm cộng đồng trực tuyến, nhà sáng tạo nội dung và các nền tảng cashback, mang lại hiệu quả cho các chiến dịch dựa trên hiệu suất. Với mô hình win-win bền vững, ACCESSTRADE thúc đẩy tăng trưởng doanh thu bền vững cho cả advertisers lẫn đối tác.
## Nâng tầm phân bổ Affiliate với Airbridge
Tích hợp giữa Airbridge và ACCESSTRADE giúp hiệu quả chiến dịch affiliate được đo lường hiệu quả và dễ triển khai hơn. ACCESSTRADE cùng với Airbridge mang đến:
### Theo dõi lượt Conversions tức thì
Các events quan trọng như app installs, sign-ups, và in-app purchases được gửi ngay lập tức từ ACCESSTRADE sang Airbridge. Marketers nhận được insight chính xác theo thời gian thực mà không cần thao tác thủ công.
### Phân bổ chi tiết và hiệu suất hợp nhất
Airbridge phân bổ từng conversion từ ACCESSTRADE về đúng publisher, placement, và nội dung quảng cáo thực sự hiệu quả, giúp marketers có góc nhìn chi tiết để đưa ra quyết định chính xác. Với hiệu quả chiến dịch affiliate được hiển thị song song cùng paid media, owned channels và các touchpoint khác trong một bảng điều khiển duy nhất, marketers dễ dàng so sánh kết quả, tối ưu ngân sách, và hiểu rõ cách từng kênh đóng góp vào tăng trưởng mà không cần chuyển đổi giữa nhiều nền tảng.
# Bắt đầu ngay
Thiết lập tích hợp giữa Airbridge và ACCESSTRADE vô cùng đơn giản. Chỉ cần thực hiện theo các bước sau để bắt đầu theo dõi và tối ưu chiến dịch affiliate của bạn:
- Tạo tracking links trong Airbridge cho các chiến dịch ACCESSTRADE của bạn.
- Cấu hình postbacks để trao đổi dữ liệu theo thời gian thực dựa trên mục tiêu conversion.
- Sử dụng báo cáo của Airbridge để nắm bắt insight về hành vi người dùng và hiệu suất chiến dịch.
- Điều chỉnh chiến lược dựa trên dữ liệu đã thu thập.
Khám phá thêm về [cách Airbridge hỗ trợ tích hợp với ACCESSTRADE](https://help.airbridge.io/en/guides/accesstrade).
Sẵn sàng tăng trưởng chiến dịch affiliate?
Hãy liên hệ với đội ngũ [Airbridge](https://www.airbridge.io/) để tìm hiểu cách mà tích hợp với ACCESSTRADE giúp bạn theo dõi thông minh hơn, tối ưu nhanh hơn và tăng trưởng mạnh mẽ hơn.
---
## Đo lường hiệu quả chiến dịch In-Feed trên Mi Ads với hệ thống attribution của Airbridge
URL: https://www.airbridge.io/vi/blog/airbridge-x-xiaomi-integration
Language: Vietnamese
Published: 2025-08-11
Airbridge đã chính thức tích hợp với Mi Ads, cho phép đo lường hiệu quả các chiến dịch in-feed dựa trên cả lượt clicks và lượt hiển thị. Với tích hợp này, advertiser chạy chiến dịch in-feed trên Mi Ads có thể nhận postback install và in-app event trực tiếp qua Airbridge, đảm bảo tính nhất quán, minh bạch và khả năng mở rộng cho toàn bộ hoạt động marketing trên OEM inventory.
Là một trong những nhà sản xuất smartphone lớn nhất thế giới, Xiaomi mang lại giá trị đáng kể cho app marketer. Nhờ hỗ trợ đo lường in-feed ads của Mi Ads, Airbridge trở thành cầu nối giữa OEM inventory và dữ liệu hiệu suất đáng tin cậy, giúp marketer có insight chính xác để mở rộng tăng trưởng.
## In-feed ads là gì?
In-feed ads xuất hiện tự nhiên trong các luồng nội dung của Xiaomi như gợi ý biểu tượng, ứng dụng hệ thống, nội dung đề xuất và bảng tin chọn lọc, mang lại quảng cáo phù hợp và tự nhiên trong quá trình trải nghiệm của người dùng.
Không giống banner từ bên thứ ba, in-feed ads hiển thị trực tiếp trên giao diện gốc của Xiaomi như content hub hoặc widget đề xuất, thu hút sự chú ý mà không làm gián đoạn hành trình người dùng. Đây là điểm chạm hiệu quả cho việc khám phá app và tăng mức độ tương tác.
## Đo lường hiệu suất in-feed ads với Airbridge
In-feed advertising đang tăng trưởng mạnh, đặc biệt trong hệ sinh thái OEM như Xiaomi. Đối với app marketer, hình thức này mang lại:
- **Lượng traffic** chất lượng cao, được định hướng theo hành vi
- Tăng **độ hiển thị gốc** ngay trên thiết bị, thu hút sự chú ý một cách tự nhiên
- **Khả năng mở rộng** trên mạng lưới toàn cầu của Xiaomi
Với tích hợp Mi Ads, marketer có thể đo lường chiến dịch in-feed với độ chính xác tương đương bất kỳ kênh quảng cáo lớn nào.
Airbridge hỗ trợ cả **click-to-app** và** impression-to-app attribution**. Khi người dùng nhấn hoặc xem quảng cáo in-feed rồi cài đặt app, Airbridge ghi nhận tương tác, phân bổ attribution và gửi postback real-time về Xiaomi. Tất cả touchpoint, từ impression, conversion đến in-app event, đều hiển thị trực tiếp trên dashboard của Airbridge.
Nhờ đó, marketer có thể:
- Theo dõi chiến dịch thống nhất trên nhiều kênh quảng cáo, với dữ liệu Mi Ads hiển thị liền mạch trong cùng dashboard.
- Phân tích sự kiện chi tiết, không chỉ dừng ở install mà còn toàn bộ in-app event tùy chỉnh (signup, purchase…).
- Đánh giá cohort và retention, theo dõi giá trị dài hạn thay vì chỉ nhìn cuối phễu.
- Kết nối đa nền tảng, đo lường hành trình từ web đến app trong một luồng thống nhất.
Điều này giúp marketer tự tin mở rộng chiến dịch OEM mà không phải đánh đổi độ chính xác trong attribution hay khả năng quan sát hiệu suất.
## What’s next
Tích hợp này là bước tiến lớn, phản ánh mục tiêu của Airbridge nhằm củng cố quan hệ với các đối tác OEM. Khi giá trị của OEM inventory ngày càng lớn đối với ngành mobile advertising, sự hợp tác giữa hai bên sẽ tiếp tục được củng cố và mở rộng, mang lại một giải pháp toàn diện dành cho các advertisers.
Sẵn sàng chạy chiến dịch in-feed với Mi Ads?
Hãy liên hệ với Customer Success Manager của bạn hoặc tham khảo [Airbridge](https://www.airbridge.io/book-a-demo) ngay.
---
## Tăng trưởng Game toàn cầu 2025: Các Chiến Lược UA, Monetization và Đo Lường Để Tăng Trưởng Bền Vững
URL: https://www.airbridge.io/vi/blog/global-game-growth-2025-ua-monetization-and-measurement-strategies-that-scale
Language: Vietnamese
Published: 2025-07-02
Thách thức lớn hơn nằm ở việc mở rộng game ra toàn cầu. Để có thể thành công vượt ra khỏi biên giới, các studio cần làm chủ các chiến lược User Acquisition (UA), kiếm tiền (monetization), hạ tầng thanh toán quốc tế và khả năng đo lường dữ liệu chuẩn xác.
[]
Tại sự kiện Actionable Growth Passport 2025: Scale Vietnam Game Studios To Global Success, bốn công ty hàng đầu trong lĩnh vực tăng trưởng game là [Airbridge](https://www.airbridge.io/), [Bidroid](https://www.bidroid.io/), [Pingpong](https://vn.pingpongx.com/vn/index) và [TopOn](https://www.toponad.net/) đã chia sẻ những chiến lược cụ thể, có thể áp dụng ngay để giúp các studio Việt Nam mở rộng ra thị trường quốc tế. Từ dự đoán LTV, ad mediation cho tới localization, sự kiện cung cấp nhiều góc nhìn thiết thực mà các marketer có thể vận dụng ngay lập tức.
Dưới đây là tổng hợp những insight quan trọng để áp dụng trong thực tế.
## 5 bài học quan trọng để tăng trưởng game toàn cầu:
- Dự đoán LTV (Predictive LTV) và phân khúc người dùng bằng AI là chìa khóa để tối ưu monetization lâu dài.
- Lựa chọn thị trường thông minh và localization bài bản giúp tối ưu chi phí UA.
- Giải pháp thanh toán quốc tế phải nhanh, tiện lợi và tuân thủ chặt chẽ quy định địa phương để người chơi tin tưởng.
- Các nền tảng ad mediation nên tập trung tối ưu doanh thu quảng cáo nhưng không làm giảm trải nghiệm người chơi.
- Thường xuyên A/B testing nội dung quảng cáo, chiến lược target và monetization để tăng trưởng bền vững.
## Airbridge: Tăng trưởng vượt bậc với pLTV, Phân khúc khách hàng và A/B testing
CEO & Cofounder của Airbridge, [Roi Nam](https://www.linkedin.com/in/roinam/), khẳng định: Chỉ đo lường retention D7 hay D30 thì không đủ. Thay vào đó, Predictive Lifetime Value (pLTV) mới là thước đo tốt hơn để dự đoán giá trị người dùng trong dài hạn.
[]
### Predictive Lifetime Value (pLTV): Dự báo giá trị người dùng tương lai
Nhiều game casual đang có thời gian hoàn vốn (payback period) kéo dài, khiến việc mở rộng quy mô sớm trở nên rủi ro. Do vậy, Airbridge khuyến khích các studio nhìn xa hơn, thay vì chỉ dựa vào các chỉ số ngắn hạn. pLTV giúp marketer dự đoán chính xác giá trị người dùng trong khoảng thời gian từ 90 đến 180 ngày, giúp đưa ra quyết định ngân sách nhanh và chính xác.
Airbridge tùy chỉnh mô hình pLTV dựa theo kiểu monetization của game:
- Game casual dùng quảng cáo (IAA): Dùng phương pháp curve-fitting.
- Game dùng IAP hoặc hybrid: Ứng dụng machine learning.
Ngoài ra, marketer còn nhận được các insight chi tiết theo quốc gia, OS, kênh, chiến dịch hay creative. Những dữ liệu chi tiết này sẽ giúp đội ngũ của họ tìm ra nhóm người dùng giá trị nhất và tập trung nguồn lực vào các hoạt động thực sự hiệu quả.
Một ví dụ thực tế là ứng dụng Alarmy. Họ đã thay thế mô hình đo lường LTV thủ công và chậm chạp trước đây bằng giải pháp pLTV của Airbridge. Nhờ khả năng dự đoán gần như tức thì theo từng kênh và nội dung quảng cáo (creative), Alarmy có thể nhanh chóng tối ưu ngân sách quảng cáo và giảm đáng kể tình trạng lãng phí chi phí marketing.
👉 [Xem case study chi tiết về Alarmy.](https://www.airbridge.io/vi/case-studies/alarmy-roas-measurement)
### Cá nhân hóa monetization bằng Phân khúc khách hàng chuẩn, Thử nghiệm thường xuyên
Airbridge cũng nhấn mạnh rằng không thể áp dụng một chiến lược monetization chung cho tất cả người dùng. Mỗi người chơi đều có hành vi khác nhau, vậy tại sao lại chỉ dùng duy nhất một cách tiếp cận?
Bằng việc liên tục thực hiện micro-segmentation dựa trên hành vi, vị trí địa lý và nhiều đặc điểm khác, các studio hoàn toàn có thể cá nhân hóa chiến lược monetization một cách linh hoạt. Khi kết hợp thêm A/B testing theo thời gian thực, các đội ngũ sẽ dễ dàng tối ưu tần suất và thời điểm hiển thị quảng cáo xen kẽ (interstitial) phù hợp với từng phân khúc người dùng cụ thể.
[]
Airbridge đã chứng minh rằng chỉ cần tinh chỉnh hợp lý tần suất hiển thị quảng cáo interstitial, studio hoàn toàn có thể tăng thêm doanh thu IAP mà không làm ảnh hưởng tới tỷ lệ retention. Tuy nhiên, vì tần suất tối ưu nhất sẽ thay đổi tùy theo khu vực và thời điểm cài đặt, nên không thể áp dụng một công thức chung cho tất cả người dùng.
Để hỗ trợ việc này, Airbridge khuyên bạn nên sử dụng các công cụ như Firebase, Amplitude, Mixpanel và ThinkingData nhằm phân khúc người dùng thông minh hơn, đồng thời tiến hành các thử nghiệm ngay trong sản phẩm một cách dễ dàng và hiệu quả.
### Airflux – Giải pháp Monetization quảng cáo bằng AI dẫn đầu xu thế
Một điểm nhấn nổi bật khác tại sự kiện là [Airflux](https://www.airflux.ai/), giải pháp tối ưu hóa doanh thu quảng cáo sử dụng AI, được thiết kế dành riêng cho các tựa game mobile.
[]
Airflux không chỉ dừng lại ở việc dự đoán LTV. Nó còn có khả năng tự động học hỏi và thích nghi liên tục. Công cụ này tự động hóa việc phân khúc người dùng, thử nghiệm các chính sách hiển thị quảng cáo interstitial theo thời gian thực, đồng thời cập nhật định kỳ 7–10 ngày một lần dựa trên dữ liệu hiệu quả thực tế. Nhờ vòng lặp học hỏi và cải tiến liên tục này, các studio có thể tối ưu thời điểm hiển thị quảng cáo và mức thưởng quảng cáo chính xác theo từng phân khúc người dùng.
Kết quả là? Doanh thu tăng trưởng mạnh mà không làm giảm trải nghiệm người chơi. Các studio sử dụng Airflux ghi nhận mức tăng tổng hợp về LTV từ 13% đến 55% chỉ trong vài tuần đầu tiên triển khai. Và điều đặc biệt là hệ thống càng hoạt động lâu, hiệu quả tích lũy càng cao, giúp mang lại lợi nhuận ngày càng ấn tượng theo thời gian.
## Thay đổi tư duy chiến lược UA toàn cầu
Tại phiên chia sẻ của mình, Bidroid đã giúp các studio game nhìn nhận lại cách họ tiếp cận việc thu hút người dùng toàn cầu. [Santhosh Malleshwara](https://www.linkedin.com/in/santhosh-malleshwara-a584177/), Giám đốc Kinh doanh và Điều hành tại Bidroid, đã đưa ra một khung chiến lược thực tế giúp các đội ngũ tránh những sai lầm phổ biến và mở rộng quy mô một cách có mục tiêu.
[]
### UA không chỉ là marketing – mà là chiến lược sản phẩm
Một sai lầm thường gặp? Coi UA chỉ đơn thuần là chi phí marketing ngắn hạn. Bidroid kêu gọi các studio hãy xem UA như một phần của chu trình kiểm nghiệm sản phẩm rộng hơn. Nó không chỉ là đổ tiền để kéo lượt cài đặt, mà còn là cơ hội để kiểm tra độ phù hợp với thị trường, cải thiện retention và nâng cao chất lượng sản phẩm.
Mở rộng toàn cầu không có nghĩa là phải lao ngay vào các thị trường Tier 1 đắt đỏ. Mà là bắt đầu ở nơi sản phẩm của bạn có khả năng thành công cao nhất. Hãy thử nghiệm trước, rồi mới mở rộng – đó là cách để không "đốt tiền" vào những thị trường bạn chưa sẵn sàng.
### Chọn thị trường đúng, bản địa hóa có chiến lược
Việc chọn thị trường rất quan trọng, nhưng nhiều studio lại làm sai. Nhảy vào những khu vực có chi phí cao quá sớm hoặc chọn sai thể loại game với gu thị trường là cách nhanh nhất để phá hỏng một sản phẩm tiềm năng. Bidroid khuyên nên bắt đầu với những thị trường "test cân bằng" như Mexico hoặc Đông Nam Á – nơi chi phí thấp và khả năng học hỏi nhanh hơn.
Còn về bản địa hóa? Không chỉ là dịch ngôn ngữ. Nội dung quảng cáo cần phản ánh được văn hóa địa phương, thói quen chơi game và ngữ điệu phù hợp. Trong các case study của Bidroid, cách tiếp cận có chủ đích như vậy đã giúp tăng retention tới 30%, tăng ARPU và giảm CPI hơn 25%.
### Test với ngân sách đủ lớn, thu thập dữ liệu đủ sâu
Một sai lầm khác là phân tán ngân sách quá nhỏ trên quá nhiều kênh, dẫn đến dữ liệu hời hợt và khó phân tích. Bidroid khuyên các studio nên tập trung vào 1–3 thị trường test, mỗi thị trường từ 3.000–10.000 USD, đủ để tạo ra những insight thực sự có giá trị từ cohort.
[]
Những dữ liệu đó chính là nền tảng cho một chu trình “test – học – cải thiện” đúng nghĩa. Khi đó UA không còn là các chiến dịch rời rạc nữa, mà trở thành một quá trình liên tục, giúp chi tiêu hiệu quả hơn, target chính xác hơn và tăng trưởng nhanh hơn mà ít rủi ro hơn.
## Pingpong: Hạ tầng thanh toán toàn cầu cho game mobile
Để mở rộng game mobile toàn cầu, bạn cần nhiều hơn là chỉ UA tốt và monetization hiệu quả. Bạn cần đảm bảo người chơi có thể thanh toán – một cách đáng tin cậy, nhanh chóng và xuyên biên giới, thứ mà Pingpong có thể giúp bạn. Bill Liu, Giám đốc vận hành sản phẩm khu vực APAC tại Pingpong, đã chia sẻ cách nền tảng của họ giúp đơn giản hóa thanh toán quốc tế, nâng cao tỷ lệ thành công và giảm chi phí mà không làm phức tạp thêm hệ thống.
[]
### Tuân thủ luật pháp – Mở rộng thị trường – Sẵn sàng toàn cầu
Pingpong hiện có giấy phép hoạt động tại Trung Quốc, châu Âu, Bắc Mỹ và nhiều khu vực khác, đảm bảo nền tảng này có thể hỗ trợ tăng trưởng quốc tế một cách hợp pháp và bền vững. Với mạng lưới hơn 140 tổ chức tài chính toàn cầu, Pingpong hỗ trợ hơn 200 phương thức thanh toán bản địa – bao gồm thẻ tín dụng, cổng thanh toán địa phương và chuyển khoản ngân hàng.
Các studio sử dụng Pingpong được hưởng lợi từ:
- Thanh toán nhanh
- Rút tiền đa tiền tệ
- Phí giao dịch giảm tới 95%
### Tác động thực tế: Xiaomi đạt tỷ lệ thanh toán thành công hơn 90%
Pingpong chia sẻ câu chuyện thành công với bộ phận Internet toàn cầu của Xiaomi – vốn từng gặp vấn đề với hệ thống thanh toán phân mảnh và tỷ lệ top-up thấp.
Bằng cách tích hợp API của Pingpong với hơn 200 lựa chọn thanh toán bản địa vào hệ thống top-up trong ứng dụng của Xiaomi, tỷ lệ thanh toán thành công đã vượt 90% và cho phép người dùng top-up chỉ bằng một cú nhấp chuột. Kết quả: tỷ lệ chuyển đổi cao hơn, người dùng hài lòng hơn và monetization hiệu quả hơn trên toàn cầu.
### TopOn: Monetization hybrid bằng IAA và IAP
[Cooper Pi](https://www.linkedin.com/in/cooper-pi-604b6b5a/), Giám đốc phát triển kinh doanh toàn cầu của TopOn, đã chỉ ra một xu hướng rõ ràng: ngày càng nhiều studio game đang chuyển từ mô hình kiếm tiền đơn lẻ sang mô hình hybrid – kết hợp quảng cáo trong ứng dụng (IAA) và mua hàng trong ứng dụng (IAP).
[]
Từ Q2/2023 đến Q2/2024:
- Tỷ lệ game hyper-casual dùng cả IAA và IAP tăng từ 19% lên 26%
- RPG tăng từ 41% lên 51%
- Game chiến thuật tăng từ 59% lên 66%
Sự dịch chuyển này phản ánh nỗ lực cân bằng giữa trải nghiệm người dùng và tăng trưởng doanh thu. Hyper-casual vẫn phụ thuộc vào quảng cáo (thường hơn 90% doanh thu), còn các game mid-core và hard-core thì phần lớn doanh thu đến từ IAP, với doanh thu từ quảng cáo đóng góp dưới 25%. Các thể loại như word, arcade hay thể thao thường nằm ở giữa.
### Ad mediation và chiến lược định dạng quảng cáo thông minh
Để thực hiện tốt mô hình hybrid, studio cần ad mediation hiệu quả và chiến lược định dạng quảng cáo linh hoạt. Nền tảng TopOn giúp tăng doanh thu trên mỗi impression thông qua các tính năng như:
- Waterfall layering, header bidding và hybrid sorting
- Kết hợp định dạng quảng cáo – ví dụ: dùng splash ad ở vị trí interstitial để tăng eCPM
- Chế độ yêu cầu thông minh: ưu tiên theo lượt hiển thị hoặc theo giá
### Tối ưu monetization qua A/B testing
Với tính năng A/B testing tích hợp, studio có thể điều chỉnh chiến lược theo hành vi người dùng và mục tiêu doanh thu. Theo TopOn, các studio game mô phỏng và hyper-casual đã ghi nhận mức tăng eCPM lên tới 15% và doanh thu tăng 10%, nhờ các tính năng như sàn giá nhiều tầng, chuyển đổi vendor theo thời gian thực và SDK thống nhất giúp thử nghiệm dễ dàng hơn.
### Growth Clinic: Giải quyết vấn đề thực tế cùng chuyên gia
Sự kiện kết thúc bằng một phiên thảo luận mở – Growth Clinic – nơi các studio hàng đầu cùng nhau xử lý các bài toán thực tế trong 5 lĩnh vực chính liên quan đến tăng trưởng game mobile. Dưới sự dẫn dắt của các chuyên gia, người tham dự đã chia sẻ kinh nghiệm và rút ra những chiến lược có thể áp dụng ngay.
[]
Một số bài học thực tiễn nổi bật:
- Mở rộng kênh UA: Xây dựng checklist về nội dung quảng cáo, tech setup và onboarding. Ưu tiên các ad network phù hợp với thể loại game và danh tiếng, luôn test với ngân sách linh hoạt nhỏ.
- Tăng ROAS sau soft launch: Nếu D1 retention cao nhưng ROAS thấp, hãy chỉnh lại kịch bản quảng cáo, cải thiện UI/UX và tối ưu target theo hành vi hoặc khu vực.
- Tăng LTV với UA ổn định: Phân khúc theo kênh, khu vực, mức chi tiêu. A/B test liên tục để tối ưu monetization và retention.
- Tracking chính xác: Theo dõi theo lớp – từ lượt cài đặt và mức độ tương tác, đến loại thiết bị và hành vi trong game như điểm rơi trong funnel hoặc người dùng mệt mỏi vì quảng cáo.
- Monetization theo phân khúc người dùng: Dùng Firebase để xác định nhóm người dùng và cá nhân hóa offer dành cho người chi tiêu cao (whales), người hay xem quảng cáo, và người dùng chi tiêu thấp.
Sự thật là: không có một chiến thuật đơn lẻ nào có thể đưa game của bạn ra thị trường quốc tế. Việc mở rộng đòi hỏi sự kết hợp thông minh và hài hòa giữa UA, đo lường, monetization và thanh toán. Nếu bạn là một marketer cho game mobile sẵn sàng bứt phá tăng trưởng, hãy liên hệ với [Airbridge](https://www.airbridge.io/book-a-demo) ngay hôm nay.
---
## Meta AMM: Phân bổ dữ liệu chi tiết của Meta với Airbridge
URL: https://www.airbridge.io/vi/blog/meta-amm-support
Language: Vietnamese
Published: 2025-06-18
Airbridge hiện đã hỗ trợ Advanced Mobile Measurement (AMM) của Meta, tin vui cho các app advertisers đang phụ thuộc vào dữ liệu chi tiết để tối ưu hiệu suất chiến dịch.
Bắt đầu từ ngày 19 tháng 6 năm 2025, các advertisers ký kết điều khoản AMM của Meta sẽ có thể truy cập dữ liệu phân bổ last touch của Meta một cách chi tiết hơn, thông qua hệ thống báo cáo tiêu chuẩn của Airbridge.
## AMM là gì?
Advanced Mobile Measurement (AMM) là hình thức báo cáo thông qua Mobile Measurement Partners (MMPs) như Airbridge, cho phép advertisers truy cập dữ liệu phân bổ last-touch ở cấp độ từng dòng dữ liệu (row-level) gắn với các chiến dịch Meta. Khác với báo cáo dạng aggregated, dữ liệu AMM cho phép phân tích từng conversion đã được phân bổ ở cấp độ chi tiết, hỗ trợ các công cụ business intelligence và mô hình nội bộ.
Đây không phải là tích hợp mới mà là sự “tái kích hoạt”. Meta đã tạm dừng AMM vào năm 2021. Kể từ đó, advertisers chỉ có thể nhận dữ liệu hiệu suất đã được tổng hợp cho các chiến dịch Meta.
Giờ đây, điều này đã thay đổi.
## Các thay đổi & Những điểm giữ nguyên
Là đối tác Mobile Measurement Partner chính thức của Meta, Airbridge đã luôn nhận dữ liệu phân bổ last-touch cho các chiến dịch Meta. Tuy nhiên, kể từ khi Meta ngừng AMM năm 2021, dữ liệu này chỉ được cung cấp cho advertisers ở dạng aggregated.
Bắt đầu từ ngày 19 tháng 6 năm 2025h, advertisers chấp nhận điều khoản AMM của Meta sẽ có thể truy cập dữ liệu conversion chi tiết hơn ở cấp độ từng dòng, bao gồm:
- Thông tin phân bổ ở cấp độ conversion cho Meta ads
- Dữ liệu row-level ngay trong báo cáo MMP hiện tại của bạn
- Đo lường chính xác hơn trong các công cụ BI hoặc phân tích nội bộ
Điểm cần lưu ý:
- Tích hợp giữa Airbridge và Meta vẫn giữ nguyên. Không có thay đổi trong cách Airbridge nhận hoặc xử lý dữ liệu phân bổ.
- Bản cập nhật này chỉ ảnh hưởng đến mức độ chi tiết của dữ liệu advertisers nhận được từ Airbridge. Báo cáo hiệu suất quảng cáo sẽ không bị tác động.
## Điều bạn cần làm
Để bắt đầu nhận dữ liệu phân bổ ở cấp độ row-level từ các chiến dịch Meta:
1. Ký điều khoản AMM của Meta [tại đây](https://www.facebook.com/ads/manage/advanced_mobile_measurement/app_based_tos)
2. Đảm bảo bạn có quyền Administrator đối với các Meta developer app liên quan để có thể chấp nhận điều khoản của AMM
- [Tìm hiểu thêm về các quyền quản trị app](https://developers.facebook.com/docs/development/build-and-test/app-roles/)
- [Bài viết hướng dẫn quản lý các quyền quản trị](https://www.facebook.com/business/help/)
Sau khi điều khoản AMM được chấp nhận, bạn sẽ bắt đầu nhận dữ liệu phân bổ Meta chi tiết ngay trong dashboard của Airbridge.
## Key Takeaway
- Meta tái kích hoạt AMM: cho phép báo cáo last-touch chi tiết cho advertisers đã ký điều khoản AMM.
- Airbridge hỗ trợ hiển thị: Là MMP hợp tác với Meta, Airbridge sẽ tự động chuyển dữ liệu này đến advertisers, cập nhật mỗi 24 giờ sau khi opt-in.
- Yêu cầu: Ký điều khoản AMM (cần quyền App Administrator).
- Lợi ích: Gia tăng mức độ chi tiết trong đo lường hiệu suất quảng cáo Meta trên MMP và công cụ phân tích.
- Tác động đến hiệu suất hiện tại: Không. Việc đo lường và báo cáo phân bổ không thay đổi, chỉ khác ở mức độ chi tiết dữ liệu advertisers được truy cập.
Đối với advertisers đang tìm cách cải thiện modeling, phương pháp đo lường hoặc phân tích LTV, sự trở lại của AMM là một bước tiến quan trọng. Hãy liên hệ [Airbridge](https://www.airbridge.io/) để bắt đầu ngay hôm nay.
---
## Chiến lược thu hút người chơi game năm 2025 sẽ thay đổi ra sao? (The Future of Game UA – Góc nhìn từ sự kiện)
URL: https://www.airbridge.io/vi/blog/ua-seminar-sketch
Language: Vietnamese
Published: 2025-05-26
[__wf_reserved_inherit]
Từng tự hỏi chiến lược UA (user acquisition – thu hút người dùng) cho game sẽ trông như thế nào vào năm 2025? Câu trả lời sẽ hoàn toàn khác biệt so với trong quá khứ. Khi chi phí thu hút người dùng tăng vọt còn doanh thu lại không theo kịp, việc chỉ tập trung phát triển chất lượng gameplay thôi là chưa đủ. Thay vào đó, các studio cần chuyển mình sang hướng tiếp cận tối ưu hóa hiệu quả quảng cáo và doanh thu dựa trên dữ liệu người chơi – càng cá nhân hóa càng tốt. Tại hội thảo "**The Future of Game UA: Global Trends & Data-Driven Strategies**", các chuyên gia kỳ cựu đã chia sẻ toàn bộ “playbook” mới cho UA thời đại mới – từ AI định hướng target cho đến tối ưu App Store. Dưới đây là những điểm nổi bật nhất mà bạn không nên bỏ lỡ.
[]
### **Game Trung Quốc đang chinh phục thị trường Hàn Quốc như thế nào?**
Quay lại năm 2019, các nhà phát hành game Trung Quốc đầu tư mạnh vào thị trường Hàn Quốc: mở công ty con tại Seoul, chi đậm cho TVC với ngôi sao hạng A. Nhưng hiện tại, cục diện đã thay đổi.Thay vì vận hành cồng kềnh, giờ đây các studio Trung Quốc trực tiếp quản lý từ trụ sở Thượng Hải. Họ cắt giảm ngân sách marketing và tập trung tối đa vào quảng cáo AI-driven cùng các chiến dịch A/B test siêu tối ưu.
[]
Tuy nhiên, muốn mở cửa thị trường Hàn Quốc không dễ: hai thử thách lớn vẫn đang chắn đường.
- **Giấy phép phát hành (판호)**: Từng bị đóng băng từ năm 2018, nay việc xin 판호 đã dần hồi phục – nhưng vẫn mất 10–12 tháng để được cấp. Chưa kể môi trường cạnh tranh khốc liệt và nền kinh tế chững lại khiến ngay cả những hãng có giấy phép ICP nội địa cũng gặp khó.
- **Sàn thương mại và KOL mới nổi**: Khi chi tiêu trong nước giảm, các nền tảng mới như hệ thống KOL theo hội nhóm (streamer guild) và nền tảng giảm giá cực sâu nổi lên. Cái đầu giúp tăng nhanh reach và doanh thu – nhưng biên lợi nhuận mỏng (có khi chỉ 10% trên 10 tỷ won). Cái thứ hai có thể tạo bùng nổ doanh số nhờ giảm giá app đến 99% – nhưng lại “đốt content” và chóng lụi tàn sau 6–12 tháng.
Kết luận: Trung Quốc vẫn là thị trường khổng lồ đầy tiềm năng, nhưng để thành công, các studio cần những đối tác nhanh nhạy, hiểu luật, và biết ứng biến theo nhịp thay đổi nền tảng.
### **AI không chỉ là một trend nhất thời: Airbridge lột xác doanh thu in-app nhờ Airflux**
Dù thị trường game mobile chỉ tăng trưởng ~5% mỗi năm, sự cạnh tranh thì ngày càng khốc liệt. Doanh thu từ game casual và hyper-casual đã tăng 180% chỉ trong 5 năm, nhưng sự trùng lặp và nhàm chán khiến người chơi ngày càng khó giữ chân. Quảng cáo interstitial là “phao cứu sinh”, nhưng dùng quá tay sẽ khiến người chơi rời bỏ, còn quá nhẹ lại lãng phí doanh thu.
[]
Giải pháp của Airbridge: **Airflux** – công cụ AI tối ưu tần suất và thời điểm hiển thị quảng cáo theo từng người chơi để tăng tối đa LTV.
- Một studio toàn cầu tăng LTV thêm 12% và giữ chân người chơi tăng 31% chỉ nhờ tùy chỉnh tần suất quảng cáo theo từng khu vực.
- Một nhà phát triển Hàn Quốc ghi nhận LTV tăng 45,9% nhờ phân khúc người chơi và gán chiến lược quảng cáo riêng cho từng nhóm.
Việc tự động hóa những tối ưu này không hề đơn giản — đó là lý do Airflux liên tục thu thập dữ liệu hành vi người chơi, đề xuất cấu hình quảng cáo tối ưu và cải thiện hiệu suất theo thời gian. Một tựa game đạt 20 triệu lượt tải đã chứng kiến LTV tăng vọt 54% chỉ trong vòng 4 tuần; nhiều sản phẩm khác cũng ghi nhận mức tăng ổn định từ 13% đến 18%. Điều đáng nói là: không cần chi thêm một đồng cho quảng cáo, Airflux vẫn đang lọt vào tầm ngắm của những studio hàng đầu. Tính năng nổi bật sắp tới? Kết hợp chiến lược IP vào một nền tảng tối ưu hóa doanh thu toàn diện.
Điều thú vị là AI giúp vừa cải thiện trải nghiệm người chơi, vừa tăng doanh thu – không cần phải đánh đổi.
### **Dự đoán người dùng VIP: Appier và công thức cho các game thủ với mức chi tiêu cao**
Ngày xưa, bạn chỉ cần chạy ads trên Kakao Games, Google Play hoặc Facebook. Ngày nay? Hệ sinh thái marketing đã mở rộng với DSP programmatic, quảng cáo tặng thưởng, và hàng loạt kênh khác. Bởi vậy, chiến lược mới sẽ là dự đoán ai sẽ chi tiền nhiều – và ai sắp rời bỏ game.
[]
Appier kết hợp log từ bid-request và first-party data để nhận diện nhóm “đại gia tiềm năng”. Qua việc phân tích các tựa game casual tại Bắc Mỹ, họ phát hiện:
- **Playable Ads** thu hút nhóm người chơi có xu hướng chi tiêu cao.
- **Banner Ads** lại phù hợp hơn với nhóm LTV thấp.
Sử dụng các thuật toán như XGBoost và LightGBM, Appier còn xây dựng mô hình dự đoán churn – giúp tự động kích hoạt retargeting chỉ sau 3–14 ngày không hoạt động.
Bên cạnh đó, Appier còn trình diễn công cụ AI tạo ads chỉ trong 10–15 phút, sử dụng footage gameplay + thông tin App Store, kèm dự đoán chuyển đổi theo từng kênh phân phối.
Tóm lại, đây là một hệ sinh thái martech khép kín: dự đoán – cá nhân hóa – sáng tạo. Và đúng như cảm nhận: “nắm trước insight của ngày mai, ngay từ hôm nay”.
### **ASO – Yếu tố quan trong của chiến lược UA**
Khi CPM tăng cao và ngân sách quảng cáo ngày càng eo hẹp, App Store Optimization (ASO) trở thành cứu cánh cho lượt tải tự nhiên.
[]
Chiến thuật của AppTweak:
1. **Tối ưu từ khóa:** Lưu lượng truy cập (impression) × tỷ lệ chuyển đổi = lượt tải. Muốn tăng hiển thị? Hãy “cắm” từ khóa đúng vào tiêu đề, mô tả và trường từ khóa. Những thương hiệu như Bondee hay Alarmi đã chứng minh: làm ASO bài bản và kiên trì giúp bạn xuất hiện ở hàng chục từ khóa tìm kiếm. Ví dụ điển hình: Call of Duty đã cải thiện rõ rệt thứ hạng tìm kiếm và số lượt tải sau khi triển khai tối ưu từ khóa.
2. **Creative Testing:** Icon, ảnh chụp màn hình, video – từng yếu tố hình ảnh đều ảnh hưởng trực tiếp đến tỷ lệ chuyển đổi. Hãy A/B test liên tục và có chiến lược riêng cho từng quốc gia, từng nền tảng – và bạn sẽ khám phá đâu là creative giá trị.
[]
Và quan trọng hơn hết: hãy đo lường **số lần hiển thị, lượt xem trang, tỷ lệ chuyển đổi** thông qua công cụ App Store Console hoặc AppTweak để luôn điều chỉnh kịp thời. ASO tưởng chừng phức tạp, nhưng nếu biết dựa vào dữ liệu, bạn hoàn toàn có thể tăng lượt tải mà không cần thêm một xu ngân sách quảng cáo.
### **Không chỉ là quảng cáo: Làm chủ retention bằng loyalty programs**
Hiệu quả của performance marketing có giới hạn – nhất là trong thời kỳ quyền riêng tư bị siết chặt. Và đó là lý do các ứng dụng tích điểm (loyalty apps) đang “bùng nổ” trong giới game thủ Gen Z và Millennials. Hãy tưởng tượng chương trình tích điểm hàng không – nhưng dành riêng cho game.
[]
Lấy ví dụ từ Playio, Người chơi được thưởng điểm khi chơi game, nhận đề xuất game cá nhân hóa và hoàn thành các nhiệm vụ. Hệ sinh thái tính năng gồm: đăng ký trước, tuyển tester cho CBT, điểm thưởng theo thời gian chơi, retargeting, và chuỗi nhiệm vụ. Kết quả là họ đã tăng tần suất chơi lẫn mức chi tiêu. Các sự kiện cộng tác kiểu “dungeon” thu hút lượng lớn người chơi mới, trong khi các tựa game như BlockNyang và Berserker Parenting đều ghi nhận ROAS và retention vượt kỳ vọng.
Bởi vậy, khi phần thưởng được gắn với trải nghiệm gameplay thực sự có ý nghĩa, tuổi thọ của game có thể vượt xa sức hút ban đầu.
### **Talkshow cùng các chuyên gia UA: Góc nhìn thật – chiến lược thật**
Các chuyên gia từ Nexon, Kakao Games, Grampus và Team 42 đã khép lại sự kiện bằng một panel thẳng thắn xoay quanh xu hướng UA năm 2025
[]
- **AI xâm nhập sâu vào UA** – Nexon
- **Chiến lược đa nền tảng, ưu tiên quyền riêng tư** – Kakao Games
- **Tập trung xây dựng hệ sinh thái và chiến lược xây dựng nội dung quảng cáo** – Grampus
- **Kết hợp branding + performance thông qua AI** – Team 42
Họ cũng nhấn mạnh việc điều chỉnh độ dài chiến dịch và nội dung quảng cáo tùy từng thị trường. Nhật Bản yêu cầu chiến dịch dài hơi, xây cộng đồng bền vững. Còn phương Tây ưu tiên A/B test nhanh – thay đổi liên tục. Trong các chiến dịch “comeback”, những game hardcore RPG thường dùng mô hình dự đoán để thu hút lại người chơi VIP đã rời bỏ. Và dù AI tạo nội dung có thể giúp đẩy nhanh test, những creative có chiều sâu do đội ngũ kỳ cựu tạo ra vẫn mang lại ROAS dài hạn vượt trội – đặc biệt với nhóm người chơi nữ.
Lời khuyên cuối cùng dành cho những người mới bắt đầu: Hãy bắt đầu nhỏ, chứng minh hiệu quả, sau đó mới mở rộng quy mô – và luôn luôn ra quyết định dựa trên dữ liệu.
### **Vậy UA game sẽ phát triển như thế nào?**
Đây không chỉ là lý thuyết suông – mà là chiến lược thực chiến từ tiền tuyến UA toàn cầu. Từ AI tối ưu targeting, mô hình dự đoán người chơi giá trị cao, đến nền tảng marketing mới, ASO bài bản và chương trình loyalty tăng retention, sự kiện mang một thông điệp rõ ràng: User Acquisition đang chuyển mình thành một nghệ thuật số hóa, chính xác, và mang tính sống còn. Nếu bạn đang chuẩn bị chinh phục thị trường toàn cầu hoặc tìm kiếm một bước đột phá mới trong chiến lược game marketing, đừng bỏ qua những chiến lược và case study thực tiễn đã được chia sẻ hôm nay — chúng chính là kim chỉ nam giúp bạn đi đúng hướng trên hành trình tăng trưởng này.
---
## Đo lường incrementality cho re-engagement và retargeting game di động
URL: https://www.airbridge.io/vi/blog/mobile-game-reengagement-and-retargeting
Language: Vietnamese
Published: 2025-05-07
[Taoufik Lazrak](https://www.linkedin.com/in/taoufik-l-75640646/) – Giám đốc Doanh thu (CRO) tại [Adikteev](https://www.adikteev.com/) – là người đứng sau thành công của nền tảng retargeting hàng đầu dành cho game mobile. Adikteev đã hợp tác với hơn 400 tựa game đình đám như King, Blizzard, Rovio, hay Playtika và đánh dấu sự khởi đầu của khái niệm incremental value. Với hơn một thập kỷ kinh nghiệm, Taoufik đã chứng kiến mọi bước chuyển mình của ngành marketing game từ thời sơ khai cho đến kỷ nguyên retargeting đỉnh cao hiện nay.
Gần đây, [Roi Nam](https://www.linkedin.com/in/roinam/), CEO của [Airbridge](http://airbridge.io/) đã có buổi trò chuyện với Taoufik để cùng chia sẻ những bí quyết vàng trong các chiến dịch retargeting và re-engagement. Hãy cùng khám phá ngay những mẹo thực chiến, xóa tan các lầm tưởng thường gặp, và sẵn sàng làm chủ cuộc chơi mới trong bối cảnh quyền riêng tư được thắt chặt ngày nay.
**🎥 Xem toàn bộ cuộc trò chuyện**
## **Những lầm tưởng lớn nhất về game re-engagement và retargeting**
Cùng đi sâu vào các sai lầm trong tư duy thường gặp về các chiến dịch game re-engagement và retargeting.
### Đâu là 3 quan niệm sai lầm nhất mà anh đã từng gặp?
**Taoufik**: Đây là một câu hỏi khá thú vị! Tôi rất vinh dự khi có cơ hội được làm rõ vài lầm tưởng phổ biến nhất về re-engagement và retargeting.
#### Lầm tưởng 1: Chiến dịch Re-engagement không hiệu quả với game casual.
Anh chắc hẳn đã từng nghe đâu đó rằng re-engagement với các tựa game có vòng đời ngắn thì chỉ phí tiền vô ích. Nhưng điều đó không hề đúng. Chỉ cần chiến lược targeting và nội dung quảng cáo phù hợp, ngay cả các game casual và hybrid-casual cũng có thể đạt được những kết quả rất ấn tượng từ các chiến dịch re-engagement.
Chúng tôi từng chứng kiến nhiều chiến dịch thành công khi kết hợp các sự kiện in-game, cập nhật nội dung mới, hoặc vào các khoảng thời gian cao điểm có khả năng tạo ra doanh thu. Bí quyết cốt lõi của re-engagement nằm ở sự cá nhân hóa—người chơi casual sẽ phản hồi tích cực hơn nhiều khi họ cảm nhận thông điệp quảng cáo thực sự hữu ích và được gửi đến vào đúng thời điểm.
#### Lầm tưởng 2: Chỉ số view-through không có ý nghĩa.
Nhiều marketer thường bỏ qua view-through vì nghĩ rằng đây là chỉ số vô nghĩa. Tuy nhiên, trong bối cảnh quảng cáo ngày càng phức tạp hiện nay, mỗi tương tác nhỏ nhất cũng đều có giá trị riêng của nó. Việc áp dụng một mô hình **mixed attribution** là rất cần thiết, đặc biệt với game mobile, khi người dùng thường cần nhìn thấy quảng cáo nhiều lần trước khi đưa ra quyết định quay lại hay thực hiện mua hàng. Ngay cả khi một quảng cáo chất lượng không nhận được lượt click nào, nó vẫn hoàn toàn có khả năng tác động tới hành vi của người chơi. Vì thế, những tác động này cần được đo lường và đánh giá một cách chính xác.
[__wf_reserved_inherit]
#### Lầm tưởng 3: Incrementality không đáng tin cậy.
Một số người vẫn nghi ngờ rằng các chiến dịch re-engagement không thật sự tạo ra doanh thu gia tăng mới. Tuy nhiên, là đơn vị tiên phong về thử nghiệm incrementality trong mảng re-engagement, Adikteev đã chứng minh rằng incrementality hoàn toàn có thể đo lường được – miễn là khung thử nghiệm phải minh bạch và khách quan.
> “Bí quyết then chốt của re-engagement là cá nhân hóa – người chơi casual sẽ phản hồi tốt hơn rất nhiều khi thông điệp quảng cáo thật sự phù hợp với họ và đến vào đúng thời điểm.”
## **Thúc đẩy doanh thu incremental cho game casual và hybrid-casual**
Thử nghiệm incrementality thường yêu cầu chia người dùng thành hai nhóm rõ ràng: một nhóm thử nghiệm (test group) và một nhóm đối chứng (control group), và việc này không hề đơn giản. Theo Taoufik, thách thức lớn nhất là phải đảm bảo các thử nghiệm này hoàn toàn khách quan, không bị thiên vị, để có thể đo chính xác doanh thu incremental từ các chiến dịch re-engagement và retargeting.
### Anh đo lường incrementality như thế nào đối với chiến dịch re-engagement dành cho những người chơi giá trị cao? Làm sao Anh chắc chắn được kết quả đo lường là hoàn toàn khách quan?
**Taoufik**: Tại Adikteev, chúng tôi áp dụng phương pháp ITT (Intention-to-Treat) vốn xuất phát từ lĩnh vực y học, nhằm đảm bảo kết quả thử nghiệm hoàn toàn khách quan và minh bạch. Điều này tương tự cách các bác sĩ đánh giá hiệu quả của một loại thuốc mới qua các thử nghiệm lâm sàng. Chúng tôi tách riêng một nhóm đối chứng (control group) để so sánh với nhóm được tiếp xúc quảng cáo re-engagement (test group), sau đó đo lường xem liệu có sự gia tăng đáng kể nào về doanh thu và mức độ hoạt động của người dùng hay không.
> “Với một khung thử nghiệm incrementality minh bạch và khách quan, ngay cả các game casual và hybrid-casual cũng hoàn toàn có thể tận dụng chiến dịch re-engagement để tạo ra thêm doanh thu incremental.”
### Các chiến dịch re-engagement có thật sự mang lại doanh thu incremental, hay chỉ đơn thuần là phân bổ lại doanh thu vốn có?
**Taoufik**: Nhiều marketer trong lĩnh vực game vẫn lo rằng các chiến dịch re-engagement chỉ khiến họ phải trả thêm tiền để tiếp cận những người chơi đã và đang chi tiêu nhiều. Tuy nhiên, những gì chúng tôi quan sát được một cách nhất quán là những người chơi giá trị cao, kể cả các nhóm đã trả tiền, thường có xu hướng chi tiêu nhiều hơn nữa khi được tiếp xúc với chiến dịch re-engagement. Đây không phải là "cannibalization" – tức lấy doanh thu cũ phân bổ lại – mà là việc khai thác thêm những giá trị mới mà nếu không có re-engagement sẽ không thể thu được.
### Đâu là cách giải thích đúng về giá trị incremental thực từ các chiến dịch re-engagement và retargeting cho khách hàng của mình?
**Taoufik**: Chúng tôi tập trung vào các kết quả thực tế, dễ hiểu. Với nhóm người dùng đang hoạt động, re-engagement giúp giữ chân họ lâu hơn và hạn chế việc họ giảm chi tiêu. Với nhóm người chơi ngừng hoạt động (lapsed users), chiến dịch re-engagement có thể kích thích họ quay trở lại. Chúng tôi chứng minh điều này qua các chỉ số rõ ràng như doanh thu incremental, mức tăng ARPU và sự khác biệt rõ nét trong hành vi người dùng giữa hai nhóm test và control. Giá trị thật sự của re-engagement không chỉ dừng lại ở click hay impression, mà là những tác động trực tiếp và tích cực đến kết quả kinh doanh của khách hàng.
[__wf_reserved_inherit]
Ngoài ra, chúng tôi luôn đảm bảo tính minh bạch tuyệt đối, cho phép khách hàng tự thực hiện lại các thử nghiệm và đối chiếu kết quả. Toàn bộ dữ liệu đo lường incrementality đều dựa trên dữ liệu sự kiện và attribution từ các MMP (Mobile Measurement Partner) như [**Airbridge**](http://airbridge.io). Vì khách hàng hoàn toàn có quyền truy cập những dữ liệu này, họ có thể dễ dàng kiểm chứng lại các thử nghiệm của chúng tôi bất cứ lúc nào, từ đó đảm bảo mọi bên đều thống nhất và tin tưởng vào kết quả.
[__wf_reserved_inherit]
### Giả sử tôi là một marketer game và muốn chạy thử nghiệm incrementality với Adikteev. Anh có thể giải thích rõ quy trình này không?
**Taoufik**: Đầu tiên, chúng tôi sẽ phân tích dữ liệu thật kỹ trước khi triển khai chiến dịch. Sau khi chính thức hợp tác, chúng tôi sẽ nhận dữ liệu về event, attribution và hành vi người dùng từ các MMP (Mobile Measurement Partner) như **Airbridge**. Dữ liệu này giúp chúng tôi đánh giá liệu tựa game đó có phù hợp để chạy các chiến dịch re-engagement hay không. Chúng tôi thường dành khoảng từ 1 đến 2 tuần để phân tích kỹ hành vi người dùng và hiểu rõ các pattern của họ. Sau đó, chúng tôi đề xuất những phân khúc người dùng cụ thể, đồng thời tính toán ngân sách tối ưu cho từng phân khúc—thường dao động từ 500 đến 1.000 USD mỗi ngày ở phạm vi toàn cầu. Mục tiêu cuối cùng là mở rộng quy mô chiến dịch một cách hiệu quả nhất.
Với thử nghiệm incrementality, chúng tôi đo lường ngay từ ngày đầu tiên khởi chạy chiến dịch re-engagement. Nhờ vậy, khách hàng có thể thấy rõ giá trị thực sự của chiến dịch ngay trong thời gian chiến dịch diễn ra, đảm bảo tính minh bạch và chắc chắn rằng mỗi đồng chi tiêu đều tạo ra giá trị cụ thể.
> “Bản chất của re-engagement không phải là phân bổ lại doanh thu vốn có (cannibalization), mà là khai phá thêm giá trị mới, những doanh thu mà nếu không chạy re-engagement sẽ không thể thu được.”
## Chiến lược đã được kiểm chứng cho chiến dịch re-engagement và retargeting
Kể từ năm 2012, Adikteev đã phát triển vượt xa việc retargeting cơ bản, cung cấp hiệu quả ROI (Return on Investment) thực sự rõ ràng và đo lường được. Các thuật toán của họ được xây dựng dựa trên hàng tỷ điểm dữ liệu về game, giúp Adikteev giữ vững vị trí dẫn đầu thị trường. Taoufik Lazrak chắc chắn thực sự hiểu rõ điều gì sẽ hiệu quả nhất trong chiến dịch re-engagement và retargeting game.
### Đâu là điều giúp khách hàng đạt được kết quả tốt nhất trong lĩnh vực gaming?
**Taoufik**: 3 tips hiệu quả nhất luôn là: phân khúc người dùng chi tiết (audience segmentation), thử nghiệm nội dung (creative A/B testing) và nhắm mục tiêu theo khu vực địa lý (geo-targeting).
Thứ nhất, hãy ưu tiên phân khúc người dùng dựa trên dữ liệu một cách thông minh. Không chỉ dừng lại ở những người dùng đang hoạt động, anh còn cần tái tương tác với những người chơi đã bỏ dở (lapsed users), thậm chí chuyển đổi được cả những người chưa từng chi trả (non-payers). Càng chia nhỏ nhóm người dùng thì hiệu quả càng cao.
Tiếp theo, hiểu rõ điều gì có thể tạo sức hút với người chơi là vô cùng quan trọng. Đó là lý do tại sao anh phải thường xuyên thử nghiệm nội dung quảng cáo (A/B testing). Anh nên liên tục thử nghiệm các định dạng quảng cáo khác nhau như interstitial, rewarded video hay banner, cùng các chiến lược về thông điệp và vị trí hiển thị chuyên dành cho retargeting. Việc này cần làm liên tục nhằm xác định rõ những yếu tố nào thật sự thúc đẩy người dùng tương tác và chuyển đổi.
Cuối cùng, khi muốn tối ưu chiến dịch, việc triển khai toàn cầu nhưng tối ưu theo từng thị trường địa phương là yếu tố then chốt. Hãy bắt đầu với geo-targeting rộng, sau đó nhanh chóng tập trung nguồn lực vào các khu vực có tiềm năng LTV cao nhất.
### Nhưng geo-targeting diện rộng như vậy liệu có rủi ro khiến game marketer lãng phí ngân sách vào những vùng không hiệu quả không?
**Taoufik**: Những nền tảng như Adikteev sử dụng các thuật toán cá nhân hóa để dự đoán chính xác người dùng nào có khả năng chuyển đổi cao, và tự động ngưng target vào những người không hiệu quả. Nhờ vậy, anh sẽ không bị lãng phí ngân sách cho các khu vực hoạt động kém, và chắc chắn rằng quảng cáo luôn được hiển thị tới đúng người dùng giá trị nhất.
> “Chúng ta nên bắt đầu với geo-targeting diện rộng, sau đó nhanh chóng thu hẹp phạm vi vào những thị trường có LTV tiềm năng cao nhất.”
## Chiến lược hybrid monetization cho game casual và hyper-casual
Trước đây, các tựa game thường chọn một trong hai hướng kiếm tiền chính: IAP (mua vật phẩm trong ứng dụng) hoặc IAA (quảng cáo trong ứng dụng). Mỗi phương pháp hướng đến một đối tượng người chơi và dòng doanh thu riêng biệt. Tuy nhiên ngày nay, nhiều tựa game casual và hyper-casual đã chuyển sang áp dụng mô hình kiếm tiền kết hợp (hybrid monetization), tận dụng đồng thời cả IAP và IAA để tối ưu doanh thu.
[__wf_reserved_inherit]
### Việc chuyển từ mô hình kiếm tiền đơn thuần sang mô hình hybrid liệu có phải là một lựa chọn đúng đắn? Các marketer game nên tập trung vào chỉ số nào?
**Taoufik**: Chắc chắn anh nên cân nhắc việc chuyển sang mô hình hybrid monetization, đặc biệt là nếu anh đang vận hành các tựa game casual hoặc hyper-casual. Tôi khuyên việc sử dụng chỉ số blended tROAS (target ROAS) thay vì ROAS truyền thống. Lý do là vì ROAS truyền thống chỉ đánh giá kết quả trong quá khứ, còn blended tROAS thì chủ động hơn, cho phép anh tối ưu hóa chiến dịch liên tục trong thời gian thực. Khi đặt một mức tROAS cụ thể, ta cũng có thể đảm bảo các chiến dịch quảng cáo luôn bám sát mục tiêu kinh doanh của mình – dù đó là tối đa hóa doanh thu hay đạt được biên lợi nhuận nhất định.
> “Hãy thử dùng blended tROAS thay vì ROAS truyền thống. ROAS truyền thống nhìn về quá khứ, còn tROAS lại giúp ta chủ động tối ưu ngay khi chiến dịch đang diễn ra.”
## Chiến lược re-engagement và retargeting trong thời đại bảo vệ quyền riêng tư
Kể từ khi Apple giới thiệu khung quản lý minh bạch theo dõi ứng dụng (App Tracking Transparency – ATT) từ phiên bản iOS 14.5 vào năm 2021, các quy định về quyền riêng tư ngày càng trở nên nghiêm ngặt hơn. Người dùng iOS giờ đây nắm toàn quyền kiểm soát dữ liệu cá nhân, buộc các marketer game phải điều chỉnh lại chiến lược re-engagement của mình để duy trì hiệu quả.
### **Trong bối cảnh quyền riêng tư được ưu tiên hàng đầu như hiện nay, các game marketer nên triển khai re-engagement như thế nào để đạt hiệu quả cao nhất trên iOS?**
Taoufik: Trước khi ATT xuất hiện, việc retargeting dựa trên IDFA (Identifier for Advertisers) là tiêu chuẩn chung. Tuy nhiên, hiện nay, nhiều người dùng chọn không cho phép theo dõi (opt-out) trên app game của bạn, ngay cả khi họ đồng ý chia sẻ dữ liệu ở các ứng dụng khác. Đây là lúc phương pháp probabilistic attribution (phân bổ xác suất) phát huy tác dụng. Probabilistic attribution giúp chúng ta kết nối và nhận diện người dùng giữa nhiều nền tảng khác nhau. Ví dụ, nếu người dùng đã đồng ý chia sẻ dữ liệu ở một app, chúng ta có thể suy đoán với mức độ chắc chắn rất cao – khoảng 90% – rằng người dùng đó cũng chính là đối tượng ta muốn tiếp cận ở nền tảng khác.
Trong một năm qua, khả năng target người dùng iOS của chúng tôi đã được cải thiện rõ rệt. Trước đây, mức độ tiếp cận chỉ khoảng 20–30%, nhưng hiện nay, con số này có thể lên đến 60–70% ở một số đối tác. Cụ thể tại thị trường Mỹ, tỷ lệ opt-in tự nhiên của người dùng game là khoảng 40–45%. Nhờ các chiến dịch của Adikteev, chúng tôi giúp đối tác tiếp cận tới 60–70% nhóm người dùng mục tiêu trên iOS, từ đó tác động rất tích cực tới doanh thu.
> “Nhờ probabilistic attribution, nếu người dùng đồng ý chia sẻ dữ liệu ở một ứng dụng, chúng ta hoàn toàn có thể suy đoán với độ chính xác cao – khoảng 90% – rằng họ cũng là người dùng mục tiêu mà chúng ta đang cố gắng tiếp cận ở ứng dụng khác.”
## 6 nội dung cần ghi nhớ
Dưới đây là 6 điểm chính bạn nên ghi nhớ để triển khai hiệu quả các chiến dịch re-engagement và retargeting cho game:
- Phá bỏ lầm tưởng: Thử nghiệm incrementality, đo lường chỉ số view-through và chiến dịch re-engagement đều có hiệu quả, ngay cả với những tựa game có vòng đời người dùng ngắn như casual hay hyper-casual.
- Incrementality: Chỉ cần có khung thử nghiệm minh bạch và khách quan, các game casual và hyper-casual đều có thể tận dụng re-engagement để tạo thêm doanh thu incremental.
- Proven best practices: Use audience segmentation, creative A/B testing, and geo-targeting for effective re-engagement.
- Tips thực chiến: Luôn áp dụng bộ ba chiến thuật: phân khúc người dùng (audience segmentation), thử nghiệm nội dung quảng cáo (creative A/B testing) và nhắm mục tiêu theo khu vực (geo-targeting) để chiến dịch re-engagement hiệu quả hơn.
- Hybrid monetization: Với game casual và hyper-casual, hãy áp dụng mô hình kiếm tiền hybrid kết hợp cả IAP và IAA, đồng thời dùng chỉ số blended tROAS làm thước đo hiệu quả chủ động.
- Retargeting ưu tiên quyền riêng tư: Sử dụng probabilistic attribution giúp bạn tăng khả năng tiếp cận thêm 20–30% người dùng iOS dù không có IDFA.
### Thúc đẩy tăng trưởng mạnh mẽ cùng Airbridge
Nếu bạn muốn nâng chiến lược game marketing lên tầm cao mới, hãy liên hệ [Airbridge](https://www.airbridge.io/) ngay. Với Airbridge, bạn sẽ có:
- Phân tích cohort nâng cao để theo dõi LTV: Đo chính xác giá trị vòng đời người chơi và phân tích hiệu suất theo từng keyword, creative, quốc gia, hệ điều hành...
- Tích hợp dữ liệu quảng cáo (Ad Mediation Integration): Tổng hợp dữ liệu từ nhiều ad network lớn như Google AdMob, AppLovin MAX hay Unity LevelPlay trên một nền tảng duy nhất.
- Dự đoán giá trị vòng đời (Predictive LTV – pLTV): Đưa ra dự báo doanh thu tương lai theo thời gian thực, giúp bạn quyết định nhanh và chính xác.
- Phân tích retention chi tiết theo phút/giờ: Thấu hiểu hành vi người chơi tới từng phút, từng giờ để có các insight thực sự có giá trị.
- Gói giá cạnh tranh, trọn gói và minh bạch: Tiếp cận các tính năng phân tích cấp độ doanh nghiệp mà không có chi phí phát sinh hay phụ phí ẩn.
Cho dù bạn là một studio indie nhỏ hay một nhà phát hành game lớn, Airbridge luôn mang đến những công cụ quan trọng giúp bạn dẫn đầu thị trường game đầy cạnh tranh ngày nay. Truy cập [airbridge.io](http://airbridge.io) để tìm hiểu thêm.
### Tối ưu chiến dịch re-engagement cùng Adikteev
Luôn dẫn đầu với những chiến lược, insight và kinh nghiệm thực tế hàng đầu về re-engagement và retargeting từ Adikteev. Khám phá thêm các tài nguyên hữu ích cùng câu chuyện thành công từ các khách hàng tại [adikteev.com](http://adikteev.com), hoặc [liên hệ trực tiếp](https://content.adikteev.com/contact-us) ngay hôm nay để được tư vấn chiến lược phù hợp nhất với game của bạn.
---
## Giới thiệu Airflux: Nền tảng tối ưu quảng cáo ứng dụng bằng AI cho game mobile
URL: https://www.airbridge.io/vi/blog/introducing-airflux-mobile-game-ad-solution
Language: Vietnamese
Published: 2025-04-09
Điều gì sẽ xảy ra nếu một tựa game casual có thể tạo ra doanh thu nửa tỷ đô?
Đó chính xác là điều mà Survivor.io đã làm được, vượt mốc doanh thu 500 triệu đô tích lũy vào năm 2024. Với một trò chơi mà chỉ cần chạm để chơi (tap-to-play), con số này thật đáng kinh ngạc — và là minh chứng rõ ràng cho sự phát triển bùng nổ của thị trường game casual.
Chúng ta đang ở thời điểm vàng của game mobile casual: chỉ cần một tựa game được đầu tư đúng cách, lợi nhuận có thể vượt xa kỳ vọng.
Nhưng mặt trái là gì? Cạnh tranh ngày càng khốc liệt. **Chi phí thu hút người dùng (UA) ngày càng tăng cao, traffic người chơi đang chững lại, áp lực phải tối đa hóa doanh thu từ người chơi đang lớn hơn bao giờ hết.** Đây là một bài toán không hề dễ giải.
Đó chính là lý do đội ngũ Airbridge xây dựng **Airflux** — một nền tảng giúp các studio game chuyên về casual gia tăng doanh thu quảng cáo nhanh chóng hơn, thông minh hơn và hiệu quả hơn. Đây cũng là sản phẩm độc lập đầu tiên kể từ khi Airbridge được thành lập.
Để hiểu rõ hơn về hành trình ra đời của Airflux — và vì sao nó cực kỳ quan trọng trong thời điểm hiện tại — chúng tôi đã có cuộc trò chuyện với Noel Son (Data Scientist) và Roi Nam (CEO & Đồng sáng lập Airbridge).
## 1. Bạn có thể giới thiệu ngắn gọn về bản thân không?
Roi: Tôi là Roi, CEO và đồng sáng lập của Airbridge. Tôi bắt đầu công ty khoảng 10 năm trước với Airbridge — một giải pháp MMP hiện đang hỗ trợ hơn 500 doanh nghiệp tại 10 quốc gia, bao gồm Hàn Quốc, Mỹ, Trung Quốc, Việt Nam và Thổ Nhĩ Kỳ. Gần đây nhất, chúng tôi cho ra mắt Airflux, và tôi đang rất sẵn lòng để được chia sẻ câu chuyện đằng sau nó.
Noel: Tôi là Noel, trưởng nhóm Data Science tại Airbridge trong 3 năm qua. Năm ngoái, nhóm của tôi đã đề xuất ý tưởng về Airflux, và sau nhiều nỗ lực, sản phẩm đã chính thức đã ra mắt. Hiện tôi đang là product owner của Airflux. Chúng tôi thật sự rất vui khi thấy sản phẩm đi vào thực tiễn.
[]
## 2. Airflux là gì?
Noel: Trước tiên, ta cần hiểu cách các tựa game mobile kiếm tiền.
Đa số game — đặc biệt là game casual — dựa vào hai mô hình chính: quảng cáo trong ứng dụng (In-app ads - IAA) và mua hàng trong ứng dụng (In-app purchases - IAP). Tùy vào từng thể loại game, tỷ trọng doanh thu đến từ hai nguồn này sẽ khác nhau.
Game casual thường phụ thuộc nhiều vào quảng cáo, trong khi game hardcore thiên về IAP. Một số game còn sử dụng chỉ một mô hình. Ví dụ, game hyper-casual thường có 100% doanh thu đến từ quảng cáo, còn game MMORPG thì hoàn toàn dựa vào IAP.
Game midcore như thể loại idle thường kết hợp cả hai: có vài video có thưởng (RV) để giữ người chơi, trong khi IAP vẫn là nguồn thu chính. Nhưng với game casual, doanh thu đến từ việc mua hàng rất thấp. Vì lối chơi nhẹ nhàng, ít khoảnh khắc “đáng tiền” nên quảng cáo đóng vai trò chủ đạo đối với doanh thu của dòng game này. Nếu người dùng không mua gì cũng không xem quảng cáo, game sẽ không có doanh thu.
Vấn đề là: hầu hết game casual dùng video có thưởng và quảng cáo xen kẽ (interstitial) — loại quảng cáo thường khiến người dùng cảm thấy bị bắt buộc vì chúng làm ngắt quãng quá trình chơi. Nếu quảng cáo hiển thị quá nhiều, người dùng sẽ bỏ game. Còn nếu quảng cáo xuất hiện quá ít, game sẽ không có nhiều doanh thu.
Để cân bằng giữa trải nghiệm người chơi và doanh thu, các monetization manager thường thiết lập sẵn (hardcode) thời điểm và tần suất hiển thị quảng cáo xen kẽ — ví dụ như cài đặt để quảng cáo xuất hiện sau mỗi 1–2 màn chơi, hoặc hiển thị với xác suất 50% sau khi người chơi vượt qua một màn.
But here’s the catch: increasing effective CPM (eCPM) by a few cents—say, bumping your bid from 3¢ to 4¢—adds a small 1¢ margin. Getting a user to watch just two more ads without churning? That has a far bigger impact on revenue.
Nhưng nếu bạn chỉ tăng eCPM thêm 1 cent (từ 3¢ lên 4¢), biên lợi nhuận tăng rất ít. Trong khi nếu khiến người dùng xem thêm 2 quảng cáo mà không rời bỏ game, tác động sẽ lớn hơn rất nhiều.Tuy nhiên, khả năng “chịu đựng” quảng cáo của người chơi khác nhau ở từng khu vực, thiết bị, phong cách chơi, và từng giai đoạn trong game — nói cách khác, mỗi người dùng là một trường hợp riêng. Nhưng đa số game vẫn áp dụng các quy tắc cố định và phân khúc người dùng rộng, do đó bỏ lỡ nhiều cơ hội tối ưu.
Đây chính là điểm cần có sự can thiệp từ Airflux.
Airflux tự động cá nhân hóa tần suất và thời điểm hiển thị quảng cáo xen kẽ, theo từng nhóm người dùng. Airflux sẽ phân khúc người chơi theo quốc gia, thiết bị, hành vi và tiến độ chơi, rồi tối ưu thời điểm và tần suất hiển thị quảng cáo. Ví dụ: người có khả năng chi tiêu cao sẽ thấy ít quảng cáo hơn, trong khi người không chi nhiều nhưng chấp nhận xem quảng cáo sẽ thấy quảng cáo nhiều hơn.
Kết quả là gì? Người chơi không có xu hướng chi tiền sẽ xem nhiều quảng cáo hơn, người có khả năng chi trả sẽ thấy ít quảng cáo hơn, giữ chân người dùng tốt hơn và giá trị vòng đời người dùng (LTV) cao hơn.
[]
Roi: Noel và nhóm data science đã chứng minh rằng: thời điểm và tần suất hiển thị quảng cáo ảnh hưởng trực tiếp đến LTV.
Điều này đặc biệt quan trọng với các studio hyper-casual và hybrid-casual — nơi biên lợi nhuận nhỏ và mọi cải thiện nhỏ về LTV đều rất đáng giá.
Airflux tự động hóa quy trình kiếm tiền và vận hành game (LiveOps) — vốn trước đây phải làm thủ công. Airflux cá nhân hóa trải nghiệm và A/B testing tự động, thay thế phán đoán cảm tính bằng dữ liệu chính xác.
Chính vì vậy, chúng tôi tự tin khẳng định: giá trị lớn nhất mà Airflux mang lại cho các studio game là gia tăng doanh thu và lợi nhuận.
## 3. Tại sao lại tập trung vào ngành game mobile?Noel: Các studio game casual cần nền tảng SaaS như thế này hơn bất kỳ ngành nào khác.
Nhiều studio chỉ có <50 nhân sự nhưng tạo ra 20-30 triệu USD mỗi năm. Tuy nhiên, các bộ phận ngoài dev như phân tích dữ liệu hay LiveOps thì không được mở rộng tương xứng.
Khác với e-commerce – nhiều data scientist và marketer dùng các công cụ như Amplitude hay Braze – ngành game vẫn còn dựa vào Excel khá nhiều, đặc biệt trong monetization từ quảng cáo.
Roi: Vì không có đội ngũ chuyên phân tích dữ liệu, nên dù có tiềm năng tăng trưởng, các studio vẫn chưa khai thác được hết tiềm năng đó.
Thị trường game mobile rất lớn. Theo [Statista](https://www.statista.com/outlook/amo/media/games/in-game-advertising/worldwide?utm_source=chatgpt.com), doanh thu quảng cáo trong game sẽ đạt 124 tỷ USD vào năm 2025. Nhưng do thiếu công cụ tối ưu, các studio đang bỏ lỡ phần lớn con số này.
Điều mà chúng tôi đang xây dựng là một giải pháp giúp các studio tận dụng AI và machine learning để gia tăng doanh thu quảng cáo từ chính lượng người chơi sẵn có. Đó chính là giá trị mà chúng tôi muốn mang đến — giúp các đội ngũ kiếm được nhiều tiền hơn từ người dùng hiện tại.
Noel: Đây chính xác là điểm cần đến các nền tảng SaaS.
Việc xây dựng và kiểm thử các mô hình tối ưu hóa đòi hỏi cả một đội ngũ data science — những chuyên gia về toán học, thống kê và học máy. Nhưng hầu hết các studio không có điều kiện để đầu tư như vậy.
## 4. Vậy các studio game mobile có thực sự coi đây là vấn đề nghiêm trọng không?Noel: Chắc chắn rồi. Hiện tại, chúng tôi đang triển khai thử nghiệm Airflux với hơn 10 studio và kết quả ban đầu rất khả quan.
Ngay cả trước khi ra mắt, nhiều studio đã chủ động nói rằng họ muốn dùng thử ngay khi sản phẩm sẵn sàng. Toàn bộ slot thử nghiệm miễn phí đã nhanh chóng được lấp đầy, và hiện chúng tôi đã chuyển sang các POC trả phí. Dù vậy, phản hồi nhận được vẫn vô cùng tích cực — điều này thực sự tiếp thêm động lực cho cả team.
Hiện chưa có nhiều công cụ nào tương tự trên thị trường, vì vậy chúng tôi đang tập trung toàn lực xây dựng một giải pháp thực sự giúp các studio gia tăng doanh thu và cải thiện lợi nhuận.
[]
Roi: Chúng tôi đặc biệt nhận được nhiều quan tâm từ các studio hyper-casual, nơi biên lợi nhuận rất thấp và họ buộc phải chạy volume lớn mới có lãi. Những team này luôn tìm cách tối ưu hóa tối đa giá trị từ inventory quảng cáo của mình.
Ngoài ra, chúng tôi cũng thấy sự quan tâm ngày càng lớn từ các studio hybrid-casual, đặc biệt khi Airflux mở rộng khả năng tối ưu hóa đến cả việc phân bổ doanh thu từ video có thưởng.
Một yếu tố lớn thúc đẩy điều này là vì tăng trưởng eCPM trong năm 2024 khá khiêm tốn. Đa số team đã thử mọi cách quen thuộc như tăng số vị trí đặt quảng cáo, nâng bid floor, kéo dài độ dài video, thêm nhiều đơn vị đấu thầu… nhưng những cách đó không còn hiệu quả như trước.
Tối ưu hóa quảng cáo đang ngày càng khó khăn. Vì vậy, ngày càng nhiều team chuyển sang các giải pháp tự động và thông minh hơn — như Airflux.
Không một studio nào có nguồn lực vô hạn. Mỗi điểm chạm trong quy trình kiếm tiền đều phải mang lại giá trị. Trong ngắn hạn, chúng tôi giúp tối ưu quảng cáo xen kẽ (interstitial) và video thưởng. Về dài hạn, chúng tôi đang mở rộng sang tối ưu IAP — tất cả đều được vận hành bởi cơ chế học máy và tự động hóa độc quyền của chúng tôi.
Vì khi LTV tăng lên, các team có thể tái đầu tư vào UA, thu hút thêm người chơi mới, và duy trì vòng quay tăng trưởng liên tục. Đó chính là ví dụ điển hình của việc LTV thúc đẩy CAC (customer acquisition cost).
## 5. Airflux khác gì so với việc tự chạy A/B test bằng Firebase?Roi: Với Firebase, các studio phải tự đặt giả thuyết, tự setup test và cấu hình các biến — ví dụ như có nên hiển thị quảng cáo sau Stage 1 hay Stage 2, xác suất là 50% hay 70%.
Những test như vậy tuy đơn giản nhưng lại đòi hỏi nhiều công sức nội bộ. Mà trong game mobile, nơi có hàng trăm biến số, việc test thủ công không thể mở rộng được.
Bạn còn phải quản lý thời gian chạy test, cỡ mẫu, và phân tích sau test. Mà nếu phải làm điều đó ở hơn 100 quốc gia, trên nhiều nền tảng, với hành vi người chơi khác nhau? Đó là hàng trăm, thậm chí hàng ngàn biến thể — gần như không thể kiểm soát nổi.
Đó chính xác là điều mà Airflux giải quyết. Hệ thống tự động hóa A/B testing trên từng phân khúc nhỏ — không cần dev, không cần viết code. Chỉ cần cắm vào là chạy, và thiết kế để mở rộng theo cách mà các công cụ truyền thống không thể làm được.
Firebase rất tuyệt để thử nghiệm những thay đổi lớn như cập nhật game design hay gói IAP mới. Nhưng nếu mục tiêu là tối ưu hóa quảng cáo ở quy mô lớn, Airflux là công cụ được sinh ra để làm việc đó.
Noel: Việc chạy các A/B test có ý nghĩa thống kê thực sự phức tạp và tốn thời gian hơn mọi người nghĩ. Có quá nhiều biến số để quản lý thủ công, đặc biệt trong game mobile.
Airflux dùng machine learning để tự động hóa toàn bộ quy trình — từ tạo giả thuyết, chạy test, phân tích kết quả cho đến triển khai tối ưu hóa. AI sẽ phân khúc người dùng, chạy hàng nghìn test song song và liên tục tối ưu doanh thu.
Đối với các studio, điều này đồng nghĩa với việc không cần thuê thêm đội ngũ data science, không cần tiêu tốn hàng giờ để chạy test thủ công. Vừa tiết kiệm thời gian, vừa tiết kiệm chi phí.
Quan trọng hơn, Airflux có tiềm năng mang lại doanh thu gia tăng mà các thiết lập cố định không thể mang lại.
Roi: Tóm lại, đây là những giá trị cốt lõi mà Airflux mang lại cho các studio game:
- Tự động điều chỉnh tần suất và thời điểm hiển thị quảng cáo xen kẽ dựa trên hành vi và đặc điểm người dùng
- Tăng LTV từ chính lưu lượng người dùng hiện có, với kết quả được chứng minh bằng số liệu đáng tin cậy
- Tự động hóa các công việc LiveOps vốn tốn nhiều giờ thủ công
- Thay thế A/B testing thủ công bằng hệ thống thử nghiệm quy mô lớn và có thể mở rộng
- Tối ưu quảng cáo dựa trên ngưỡng chịu đựng của từng người dùng — tối đa hóa doanh thu mà không ảnh hưởng đến UX hay retention
- Mang lại kết quả rõ ràng, đo lường được và có căn cứ dữ liệu
## 6. Airflux đã có những thành công thực tế nào chưa?Noel: Chắc chắn rồi. Chúng tôi hiện đang hợp tác với hơn 10 studio game và kết quả ban đầu rất ấn tượng.
Với một tựa game FPS chủ đề zombie có hơn 10 triệu lượt tải toàn cầu, chúng tôi ghi nhận mức tăng LTV người dùng đến 50%. Trong một trường hợp khác, một studio với hơn 10 triệu MAU đã ghi nhận mức tăng trưởng 14% chỉ sau hai tuần.
Một đối tác lớn hiện tại đang đạt mức tăng 16%, vô cùng triển vọng.
[]
## 7. Các studio cần chiến lược gì để giữ vững tính cạnh tranh?Roi: Trong e-commerce, chiến lược ban đầu tập trung vào việc thu hút người dùng mới và bán hàng nhanh. Nhưng theo thời gian, trọng tâm đã chuyển sang tăng LTV — nhờ vào CRM, phân tích dữ liệu, cá nhân hóa sâu và các chương trình giữ chân người dùng.
Chúng ta đang chứng kiến xu hướng tương tự trong ngành game mobile. Thời kỳ “mua user giá rẻ và scale nhanh” đã qua. Giờ điều quan trọng là bạn kiếm được bao nhiêu giá trị từ mỗi người chơi.
Và theo tôi, Airflux chính là giải pháp tốt nhất để làm được điều đó.
Noel: Tôi hoàn toàn đồng ý với Roi. Khi rào cản gia nhập giảm xuống, thị trường game mobile ngày càng cạnh tranh.
Vì vậy, trọng tâm phải chuyển sang việc tăng LTV mà không ảnh hưởng đến trải nghiệm người chơi. Về bản chất, điều đó rất đơn giản: bạn đã tốn nhiều tiền để có được những người dùng này — giờ hãy xem họ như tài sản giá trị dài hạn. Hãy bảo vệ trải nghiệm của họ, nhưng cũng đừng bỏ lỡ bất kỳ cơ hội kiếm tiền nào về lâu dài.
---
## Cách Playio chuyển sang Airbridge và quản lý marketing toàn cầu từ một nền tảng duy nhất
URL: https://www.airbridge.io/vi/blog/playio-mmp-migration-global-marketing
Language: Vietnamese
Published: 2025-04-01
## Giá trị tốt nhất cho game thủ — Ứng dụng phần thưởng dành cho game thủ, Playio
### Q. Bạn có thể giới thiệu về bản thân và công ty không?
Xin chào, tôi là Hayoung Park, Trưởng nhóm Marketing tại Playio. Tôi đã đam mê game từ khi bắt đầu sự nghiệp — công việc đầu tiên của tôi là tại một công ty game. Hiện tại, tại GNA Company, tôi đang cùng Playio hướng tới sứ mệnh "Best Value for Gamers" để xây dựng một hệ sinh thái game lành mạnh.
### Q. Bạn có thể cho chúng tôi biết về Playio không?
Về cơ bản, Playio là một ứng dụng phần thưởng dựa trên thời gian chơi game. Người dùng kiếm điểm dựa trên thời gian chơi game và có thể đổi điểm lấy sản phẩm. Tuy nhiên, Playio hướng tới nhiều hơn một ứng dụng phần thưởng đơn giản — đây là một môi trường nơi game thủ được công nhận giá trị riêng của mình.
Ví dụ, thay vì chỉ cung cấp phần thưởng, chúng tôi xếp hạng người dùng theo cấp độ và bậc dựa trên thời gian chơi thực tế. Playio cũng là một cộng đồng nơi các game thủ đa dạng kết nối với nhau. Đối với người dùng, đây là không gian để thưởng thức game và kết nối; đối với nhà phát hành game, đây là kênh marketing được công nhận có những "game thủ nòng cốt" tham gia tích cực, thúc đẩy cài đặt và giữ chân người dùng.
[Truy cập Playio](https://playio.co/?LANG_CD=ENG)
[]
### Q. Chúng tôi biết Playio hoạt động tại nhiều quốc gia.
Đúng vậy. Playio hiện đang hoạt động tại Hàn Quốc, Hoa Kỳ, Nhật Bản và Đài Loan. Chúng tôi cũng đang lên kế hoạch mở rộng sang Hồng Kông, Ma Cao và các thị trường khác.
### Q. Đội ngũ marketing được tổ chức như thế nào và làm những gì?
Đội ngũ marketing của chúng tôi được chia thành hai mảng B2B và B2C. Đội B2B, mới thành lập năm 2025, phối hợp chặt chẽ với đội ngũ bán hàng để chứng minh Playio là kênh marketing hiệu quả như thế nào. Họ xuất bản bài viết, gửi bản tin, tạo khách hàng tiềm năng để thu hút khách hàng mới và củng cố mối quan hệ với các đối tác hiện tại.
Đội B2C tập trung giúp người dùng khám phá và tận hưởng Playio. Ngoài việc thu hút người dùng mới, họ triển khai nhiều hoạt động marketing khác nhau để giữ cho người dùng hiện tại hài lòng và gắn bó.
## Hành trình của Playio đến với Airbridge
### Q. Chúng tôi nghe nói Playio đang dùng MMP thứ ba và đã chọn Airbridge. Bối cảnh là gì?
Năm ngoái, chúng tôi gia nhập thị trường Nhật Bản và bắt đầu hợp tác với một nhà phát hành địa phương, người cũng đảm nhận marketing. Điều này có nghĩa là chúng tôi cần hạn chế quyền truy cập dashboard của nhà phát hành chỉ với dữ liệu Nhật Bản — họ không nên nhìn thấy hiệu suất từ các quốc gia khác. MMP trước đó của chúng tôi cho biết việc phân tách dữ liệu theo cấp quốc gia như vậy là không thể.
Vì vậy, chúng tôi bắt đầu tìm kiếm một MMP có thể giới hạn khả năng xem dữ liệu theo quốc gia. Hầu hết các MMP đều không thể cung cấp điều này. Airbridge là đơn vị duy nhất xác nhận họ có thể làm được, và đó là lý do khiến chúng tôi quyết định chuyển đổi.
Nhưng đó không phải lý do duy nhất. Cá nhân tôi luôn cảm thấy Airbridge có UI/UX trực quan hơn so với các MMP khác. Hiệu suất marketing là thứ mà mọi người trong tổ chức nên tiếp cận được — không chỉ các nhà phân tích dữ liệu hay performance marketer — và giao diện trực quan của Airbridge đã giúp điều đó trở nên khả thi.
### Q. Quá trình chuyển đổi mất bao lâu và diễn ra như thế nào?
Khoảng một tháng. Như tôi đã đề cập, chúng tôi có yêu cầu đặc biệt về phân quyền dữ liệu theo cấp quốc gia, và chúng tôi có nhiều yêu cầu bổ sung trong quá trình tích hợp. Chúng tôi tổ chức các cuộc họp chuyển đổi thường xuyên, nhưng nhờ sự hỗ trợ chủ động từ đội ngũ Airbridge, mọi thứ đều diễn ra suôn sẻ.
Chuyển đổi MMP không phải là việc bạn làm thường xuyên, và chúng thường khó khăn và tốn thời gian. Airbridge cung cấp một quy trình có hệ thống với danh sách kiểm tra chi tiết, cho phép chúng tôi hoàn thành nhanh chóng. Họ cũng tổ chức đào tạo onboarding cho toàn công ty, giúp nhiều đồng nghiệp hơn có thể sử dụng Airbridge.
[]
## Cách Playio sử dụng Airbridge
### Q. Playio hoạt động tại nhiều quốc gia. Chiến lược và đo lường hiệu suất theo từng quốc gia hẳn rất quan trọng.
Chính xác. Chúng tôi triển khai chiến lược và kênh media khác nhau theo từng quốc gia. Hàn Quốc ra mắt năm 2021, Nhật Bản vào tháng 4 năm ngoái, và Đài Loan vào tháng 11. Tại Hàn Quốc, D30 retention vượt 30% vào năm ngoái, cho thấy mức độ trung thành của người dùng rất cao. Chúng tôi không chỉ tập trung vào thu hút người dùng mới mà còn triển khai chiến lược nội dung, nhân vật và CRM để truyền tải hiệu quả tầm nhìn thương hiệu.
Đài Loan và Nhật Bản đang trong giai đoạn tăng trưởng nhanh, với trọng tâm mạnh hơn vào việc thu hút người dùng mới. Chúng tôi tích cực chạy các kênh lớn như Google và Meta cùng với DSP, tối ưu hóa chiến lược theo từng quốc gia.
### Q. Playio sử dụng những tính năng nào của Airbridge nhiều nhất cho việc tối ưu mobile marketing?
Chúng tôi sử dụng rộng rãi các chỉ số và dữ liệu Airbridge để đo lường hiệu suất quảng cáo theo kênh media và creative. Một tính năng mà Playio đặc biệt đánh giá cao là sự đa dạng của các báo cáo. Các MMP khác không cung cấp mức độ đa dạng báo cáo như vậy. Airbridge cho phép bạn tùy chỉnh báo cáo theo bất kỳ chỉ số nào bạn muốn — đó là một lợi thế rất lớn.
[Khám phá báo cáo Airbridge](https://help.airbridge.io/en/guides/airbridge-report-overview)
Mục tiêu của đội marketing được gắn kết với KPI toàn công ty — những chỉ số như tăng trưởng người dùng mới và cải thiện retention. Chúng tôi sử dụng các tính năng báo cáo đa dạng của Airbridge để tạo và chia sẻ biểu đồ để toàn công ty có thể theo dõi các mục tiêu này cùng nhau. Nhờ buổi đào tạo onboarding, các đồng nghiệp ngoài bộ phận marketing giờ tự tạo báo cáo trực tiếp.
Một tính năng quan trọng khác là tracking link thông qua custom channel. Ngoài quảng cáo mobile, Playio triển khai rất nhiều influencer marketing, content marketing và các chương trình hợp tác đối tác bên ngoài. Với những hoạt động này, chúng tôi sử dụng custom channel để đo lường và tối ưu hiệu suất theo từng chiến dịch và từng influencer.
[Khám phá Airbridge Custom Channel](https://help.airbridge.io/en/guides/custom-channels)
### Q. Bạn có thể chia sẻ ví dụ về kết quả marketing đạt được với Airbridge không?
Tại Hàn Quốc, chúng tôi có một trường hợp cải thiện hiệu suất bằng cách kết hợp báo cáo phễu Airbridge với phân tích dữ liệu nội bộ. Cải thiện retention là mục tiêu chính tại Hàn Quốc. Chúng tôi sử dụng dữ liệu nội bộ để xác định "khoảnh khắc aha" của người dùng, tạo nhiều hạng mục hành động để dẫn dắt người dùng đến những khoảnh khắc đó, chạy A/B test, và sử dụng báo cáo phễu Airbridge để kiểm tra và tối ưu kết quả. Cuối cùng, bằng cách tối ưu con đường đến khoảnh khắc aha, D30 retention nội địa đã vượt 30%.
Chúng tôi cũng có trường hợp tối ưu marketing theo quốc gia. Tại Đài Loan, báo cáo Airbridge Actuals cho thấy video creative có hiệu suất tốt hơn đáng kể so với image creative trong các chiến dịch UA. Dựa trên dữ liệu này, chúng tôi tăng tỷ lệ video creative, dẫn đến tăng hơn 100% lượng người dùng thu hút được và tiết kiệm 40% chi phí UA.
> Bạn muốn tìm hiểu thêm về kết quả mobile marketing của Playio với Airbridge?
[Đọc case study đầy đủ](https://www.airbridge.io/vi/customers/playio-d30-retention-cpa-optimization)
### Q. Với tư cách là một marketer, những ưu điểm lớn nhất của Airbridge là gì?
Một trong những ưu điểm lớn nhất mà tôi thấy là, ngay cả với tư cách là một MMP, Airbridge có khả năng trực quan hóa dữ liệu được thiết kế đặc biệt tốt. Có nhiều báo cáo có thể tùy chỉnh, kết hợp dữ liệu linh hoạt, và thiết kế rất trực quan. Sau khi sử dụng nhiều MMP, Airbridge là nơi duy nhất cung cấp mức độ đa dạng và trực quan như vậy trong dashboard.
[]
Một ưu điểm lớn khác là người không phải marketer cũng có thể sử dụng dễ dàng. Đối với những người ngoài lĩnh vực marketing hoặc phân tích dữ liệu, việc nhìn vào dữ liệu có thể khiến họ e ngại. Nhưng UX/UI trực quan và khả năng sử dụng dễ dàng của Airbridge khiến ngay cả những đồng nghiệp không có nền tảng marketing cũng cảm thấy thoải mái.
## Tầm nhìn của Playio: Tương lai của giải trí game (Reward 2.0)
### Q. Playio không chỉ là kênh phần thưởng đơn giản — mà công nhận và tôn trọng giá trị của game thủ. Tầm nhìn của Playio về tương lai của nền tảng phần thưởng game là gì?
Game thủ chơi game vì nhiều động lực khác nhau — từ giải trí đơn giản đến thành tích, kết nối xã hội và sự công nhận. Playio dự định cung cấp thêm nhiều tính năng, nội dung và phần thưởng để khuếch đại những động lực đa dạng này.
Chúng tôi tin rằng tầm nhìn của Playio vượt xa việc chỉ là một nền tảng phần thưởng. Chúng tôi gọi đây là "Reward 2.0." Theo tầm nhìn này, chúng tôi hướng tới việc tiếp tục công nhận giá trị của người dùng với tư cách là game thủ, đồng thời trở thành nền tảng cung cấp quảng cáo có mục tiêu chính xác hơn cho nhà phát hành game.
### Q. Mục tiêu của đội marketing khi Playio bước trên con đường Reward 2.0 là gì?
Như tôi đã đề cập, Playio nỗ lực tạo ra một hệ sinh thái game lành mạnh vượt xa phần thưởng đơn giản. Đội ngũ marketing cùng hướng tới mục tiêu này. Mỗi thành viên trong đội đều là game thủ, và chúng tôi làm việc để tạo ra nhiều trải nghiệm hơn mang lại giá trị cho game thủ phù hợp với định hướng sản phẩm. Đó là lý do Playio làm marketing.
[]
### Q. Bạn có muốn chia sẻ điều gì thêm không?
Sau khi trải qua nhiều MMP, tôi rất vui khi đã ổn định với Airbridge. Quá trình chuyển đổi phức tạp đã được hoàn thành nhanh chóng, và giờ toàn công ty đang đạt được tăng trưởng thông qua dữ liệu Airbridge. Với dashboard trực quan, đội ngũ CSM hỗ trợ nhanh nhạy, và nhiều ưu điểm khác, tôi không thể tưởng tượng việc quay lại MMP khác.
Mọi người tại Playio đều đang nỗ lực xây dựng một dịch vụ tốt hơn. Các tính năng mới liên tục được bổ sung và dịch vụ đang phát triển nhanh chóng, vì vậy hãy tiếp tục theo dõi hành trình của Playio. Chúng tôi cũng liên tục tuyển dụng những người tài năng để cùng đồng hành trên con đường thú vị này.
[Xem vị trí tuyển dụng tại GNA Company](https://gna.career.greetinghr.com/team)
Bạn muốn tối ưu hiệu suất mobile marketing toàn cầu với Airbridge?
---
## Cách Tính Giá Trị Vòng Đời Dự Đoán (Predictive Lifetime Value - pLTV) Cho Ứng Dụng Di Động
URL: https://www.airbridge.io/vi/blog/calculate-pltv-predictive-lifetime-value-for-mobile-app
Language: Vietnamese
Published: 2025-03-06
## Vì Sao pLTV Đang Trở Nên Quan Trọng Hơn Bao Giờ Hết
Trong hệ sinh thái ứng dụng di động ngày càng cạnh tranh khốc liệt, việc hiểu rõ Giá trị vòng đời dự đoán (Predictive Lifetime Value - pLTV) là yếu tố then chốt. Khi chi phí thu hút người dùng (UA) không ngừng tăng, các marketer cần có được cái nhìn sớm về giá trị người dùng để ra quyết định dựa trên dữ liệu, tối ưu ngân sách và thúc đẩy tăng trưởng bền vững.
> 💡 pLTV là gì?
Predictive Lifetime Value (pLTV) là một chỉ số hướng tới tương lai, ước tính tổng doanh thu mà một người dùng có thể mang lại trong suốt quá trình gắn bó với ứng dụng hoặc dịch vụ.
### Vì Sao pLTV Là Một Chỉ Số Quan Trọng:
- Thời gian hoàn vốn kéo dài: Chi phí thu hút người dùng (UA) tăng khiến thời gian thu hồi vốn từ quảng cáo cũng lâu hơn. Nếu không hiểu rõ giá trị dài hạn của người dùng, bạn khó có thể tối ưu chiến lược UA.
- Doanh thu được ghi nhận dần theo thời gian: Các mobile app, cụ thể là dạng subscription hoặc quảng cáo trong ứng dụng (in-app ad) thường không tạo ra doanh thu ngay. Vì vậy, các chỉ số ngắn hạn không phản ánh đúng tiềm năng của người dùng.
- Tối ưu ngân sách theo thời gian thực: Marketer cần ra quyết định nhanh chóng, chính xác để phân bổ ngân sách hiệu quả. Dự đoán pLTV chính xác giúp tối đa hóa ROI trên mọi chiến dịch và kênh.
Nhờ pLTV, marketer có thể chủ động điều chỉnh chiến lược thu hút và giữ chân người dùng, ngay cả khi doanh thu ngắn hạn chưa phản ánh đầy đủ giá trị dài hạn của họ.
## 2 Cách Tính pLTV
Cùng khám phá hai phương pháp phổ biến để tính pLTV: Phương pháp hồi quy đường cong (Curve Fitting) và Phương pháp học máy (Machine Learning).
### Phương pháp Curve Fitting
Sử dụng dữ liệu retention trong quá khứ để vẽ đường cong toán học, dự đoán xu hướng giữ chân và doanh thu trong tương lai.
> Cách thực hiện:
1. Phân tích đường cong retention: Chọn hàm toán học (ví dụ: mũ, lũy thừa) phù hợp với xu hướng retention và tích phân trong khoảng thời gian nhất định (30, 60, 90 ngày).
2. Tính ARPDAU: Phân nhóm người dùng theo cohort (quốc gia, hệ điều hành, kênh...) và tính trung bình doanh thu mỗi người dùng hoạt động hàng ngày, bao gồm cả IAP và IAA.
3. Tính pLTV: Nhân kết quả tích phân retention với ARPDAU để ước tính tổng LTV.
Ưu điểm:
- Hiệu quả khi phân tích hành vi người dùng ở quy mô lớn.
- Phù hợp với app có tương quan chặt giữa tần suất hoạt động và doanh thu.
Hạn chế:
- Kém hiệu quả với app có mô hình kiếm tiền phức tạp (ví dụ: game nặng IAP).
- Khó cá nhân hóa và không phù hợp với mô hình retention không ổn định.
### Phương pháp Machine LearningPhương pháp Machine Lea
Học máy cho phép dự đoán LTV chi tiết và chính xác hơn bằng cách phát hiện các mẫu hành vi phức tạp.
> Cách thực hiện:
1. Chọn biến đặc trưng: Bao gồm lịch sử giao dịch, tần suất phiên, thông tin nhân khẩu học và chỉ số tương tác.
2. Huấn luyện mô hình: Sử dụng các thuật toán ML như Gradient Boosting, Neural Networks để dự đoán LTV dựa trên dữ liệu quá khứ.
Ưu điểm:
- Dự đoán cá nhân hóa, chính xác hơn.
- Phù hợp với app có hành vi giữ chân hoặc doanh thu không đều.
Hạn chế:
- Cần dữ liệu lớn và thường xuyên tinh chỉnh mô hình.
- Phức tạp và tốn tài nguyên khi triển khai.
### Vai Trò Của MMP Trong Việc Tính pLTV
Nhiều marketer nghĩ rằng pLTV có thể tính được chỉ từ dữ liệu nội bộ (như hệ điều hành, quốc gia, thiết bị). Nhưng khi kết hợp với dữ liệu phân bổ từ MMP (Mobile Measurement Partner), độ chính xác được cải thiện đáng kể.
Tại Sao Dữ Liệu Phân Bổ Quan Trọng?
- Mỗi nguồn quảng cáo mang lại người dùng với giá trị vòng đời khác nhau.
- Biết người dùng đến từ đâu giúp dự đoán chính xác hành vi, mức độ giữ chân và doanh thu dài hạn Ví dụ, người dùng đến từ quảng cáo video ngắn thường có hành vi khác với người dùng từ quảng cáo tìm kiếm.
### Tính Năng pLTV Của Airbridge: Chính Xác Và Quy Mô Lớn
Airbridge sử dụng phương pháp Curve Fitting Bayesian kết hợp với[ Luft Engine](https://www.airbridge.io/en/blog/introducing-luft), cơ sở dữ liệu phân tích hành vi người dùng độc quyền – để cung cấp giải pháp pLTV chính xác và đáng tin cậy dành riêng cho marketer di động.
[]
#### Ưu Điểm Nổi Bật:
- Phân tích chi tiết: Theo dõi pLTV theo quốc gia, hệ điều hành, chiến dịch và nội dung quảng cáo.
- Dự đoán dài hạn với dữ liệu ngắn hạn: Chỉ cần 3 ngày dữ liệu để dự đoán pLTV trong 180 ngày.
- So sánh chỉ với 1 cú nhấp: Theo dõi pLTV, CAC và ROAS trực tiếp trên dashboard của Airbridge.
- Không tính phí thêm: Đã bao gồm trong gói Airbridge tiêu chuẩn – không phát sinh chi phí.
Với Airbridge, marketer có thể ra quyết định nhanh chóng và chính xác, phân bổ ngân sách hiệu quả theo khu vực và chiến dịch.
### Case Study: DelightRoom (Alarmy) Tối Ưu Hóa Chiến Dịch Toàn Cầu Với Tính Năng pLTV
[]
#### Thách Thức
Trước khi dùng Airbridge, DelightRoom gặp khó khăn khi phải tự tính pLTV cho chiến dịch toàn cầu. Họ cần dữ liệu chi tiết phân theo kênh, chiến dịch, nội dung quảng cáo, quốc gia, hệ điều hành – và theo tuần, tháng, quý.
#### Kết Quả
Bằng cách áp dụng tính năng pLTV của Airbridge, DelightRoom đã ngay lập tức truy cập được các phân tích chi tiết, loại bỏ quy trình thủ công trước đó tốn hơn 3 tiếng mỗi lần. Giờ đây, họ có thể theo dõi pLTV và pROAS (ROAS dự đoán) một cách linh hoạt, từ đó tối ưu hóa ngân sách theo thời gian thực.
Giải pháp của DelightRoom: Tái phân bổ ngân sách thông minh
- DelightRoom kết hợp dự đoán pLTV với chỉ số CPI và retention để đưa ra quyết định chính xác hơn.
- Với các nội dung quảng cáo có pLTV cao nhưng CPI hoặc retention thấp, họ điều chỉnh ngân sách hợp lý để tránh lãng phí.
Nhờ tận dụng tính năng pLTV của Airbridge, DelightRoom đã tối ưu hóa quy trình phân bổ ngân sách, nâng cao hiệu quả marketing và cải thiện đáng kể ROI.
### Tối Ưu Tăng Trưởng Với Dự Đoán LTV
Dự đoán LTV nhanh chóng và chính xác giúp marketer tự tin phân bổ ngân sách, nâng cao hiệu suất chiến dịch và thúc đẩy tăng trưởng bền vững trong thị trường ngày càng cạnh tranh.
Hãy liên hệ với đội ngũ Airbridge để khám phá cách mà tính năng pLTV có thể giúp bạn ra quyết định thông minh và tối đa hóa lợi nhuận dài hạn của ứng dụng.
---
## Top 3 Chiêu Trò Gian Lận Quảng Cáo Di Động Phổ Biến Nhất 2025 Trong Ngành Fintech Việt Nam & Giải Pháp Phòng Chống Hiệu Quả
URL: https://www.airbridge.io/vi/blog/top-3-mobile-ad-fraud-tactics-in-vietnams-fintech-industry-2025-and-how-to-prevent-them
Language: Vietnamese
Published: 2025-02-13
## Tổng quan
Ngành công nghệ tài chính (Fintech) tại khu vực APAC đang chứng kiến những thay đổi đáng kể, với sự tăng trưởng vượt bậc ở các chỉ số như lượt cài đặt ứng dụng, số phiên hoạt động trên di động và doanh thu từ các ứng dụng tài chính. Tại Việt Nam, thị trường này đang phát triển mạnh mẽ với sự xuất hiện của hàng loạt ứng dụng cung cấp các dịch vụ thanh toán, vay tín chấp và quản lý tài chính cá nhân.
[]
[]
Tuy nhiên, sự bùng nổ của thị trường cũng kéo theo sự gia tăng của các hành vi gian lận quảng cáo ngày càng tinh vi, đặc biệt gây ảnh hưởng nặng nề đến lĩnh vực Fintech. Tỷ lệ gian lận cài đặt cao trong các ứng dụng tài chính và Fintech xuất phát từ nhiều nguyên nhân, bao gồm ngân sách tiếp thị lớn; sự thiếu hiểu biết sâu sắc về các chỉ số KPIs trong ngành quảng cáo, đặc biệt ở các ngân hàng truyền thống hoặc các công ty đầu tư mới bắt đầu bước chân vào quảng cáo kỹ thuật số; cùng với chi phí cài đặt trung bình (CPI) cao nhất trên thị trường, tạo động lực lớn cho các hành vi gian lận.
Ngoài ra, một lý do quan trọng khác khiến tỷ lệ gian lận cao là ngành Fintech vẫn đang ở giai đoạn đầu của quá trình chuyển đổi từ nền tảng desktop sang di động và hiện tập trung nhiều hơn vào việc tăng số lượng người dùng thay vì tối ưu hóa doanh thu. Bên cạnh đó, chi phí để có được một người dùng trong lĩnh vực tài chính thường cao hơn so với các ứng dụng khác, do kỳ vọng người dùng có tỷ lệ giữ chân lâu dài hơn và tần suất giao dịch cao hơn.
Theo các báo cáo gần đây, Việt Nam là một trong những thị trường chịu ảnh hưởng nặng nề nhất bởi gian lận quảng cáo di động, với ước tính khoảng 10-30% ngân sách quảng cáo bị thất thoát mỗi năm. Thực trạng này không chỉ gây lãng phí tài nguyên mà còn tác động tiêu cực đến hiệu quả kinh doanh cũng như niềm tin của người dùng đối với các ứng dụng Fintech.
## 3 Hình Thức Gian Lận Quảng Cáo Phổ Biến Trong Ngành Fintech
Bot luôn đứng đầu trong các loại gian lận ảnh hưởng đến thị trường di động của Fintech, chiếm hơn 50% tổng số gian lận. Các bot gây thiệt hại nghiêm trọng cho doanh nghiệp trong ngành khi thực hiện các “cuộc tấn công” với quy mô lớn trong thời gian rất ngắn, khiến dữ liệu bị ảnh hưởng nhanh chóng.
Hiện nay, tỷ lệ gian lận quảng cáo trong lĩnh vực Fintech vào khoảng 31,8%, tức cứ 1 trong 3 lượt cài đặt là gian lận. Vì vậy, hãy cùng Airbridge tìm hiểu những loại gian lận quảng cáo phổ biến trong ngành Fintech, cách thức hoạt động và cách nhận biết để bảo vệ doanh nghiệp của bạn.
### Điểm chạm & lượt chuyển đổi giả mạo (Device Farms)
Device Farms là gì?
Device farms, hay còn gọi là [Phone farms](https://www.airbridge.io/vi/glossary/phone-farms), sử dụng các bot thông minh cùng với hàng loạt những thiết bị đời cũ giá rẻ, có thể tạo ra một lượng lớn các điểm chạm (touchpoints) hoặc lượt chuyển đổi (conversions) giả mạo. Theo số liệu thống kê, Device farms có thể sử dụng hàng nghìn thiết bị để giả mạo các sự kiện sau cài đặt trong thời gian lên đến một tháng sau mỗi lượt cài đặt.
[]
Số liệu từ Interceptd cho rằng hình thức gian lận này thường chiếm tới 25% tổng các trường hợp gian lận, và còn số này thậm chí còn lớn hơn với ngành Fintech, chiếm tới 31%, theo như nguồn tổng hợp của Bussiness of Apps. Bởi vậy, theo như Jupiter Research, Device Farm thường tiêu tốn lượng lớn ngân sách quảng cáo của các doanh nghiệp Fintech, dao động từ 12% cho tới 20%.
[]
“Triệu chứng” của Device Farms
Loại gian lận này thường được nhận biết qua tỷ lệ chuyển đổi và mức độ tương tác bất thường thấp, do phần lớn người dùng ứng dụng tài chính này đều là giả mạo từ các thiết bị. Điểm khác biệt so với gian lận phân bổ (attribution fraud) chính là chất lượng người dùng cực kỳ tệ, với tỷ lệ giữ chân người dùng không mấy khả quan.
Cụ thể hơn, tỷ lệ giữ chân thường giảm mạnh từ Day 1, và gần như bằng 0 từ Day 5 đến Day 7 trở đi do tài khoản giả mạo thường không có động lực thực hiện các hành vi tự nhiên như đăng nhập hoặc tương tác với ứng dụng sau khi được tạo.
[]
Sự thiếu động lực để tiếp tục sử dụng ứng dụng này cũng được thể hiện qua mức độ chuyển đổi trong ứng dụng thấp. Thông qua các thông số chuyển đổi qua từng bước trong eKYC (electronic Know Your Customer), các doanh nghiệp có thể nắm bắt được hành trình khách hàng sau khi thực hiện cài đặt theo từng kênh, từ đó phát hiện và loại bỏ các kênh có xu hướng hành động trong ứng dụng (in-app event) đáng nghi.
Ví dụ, sau khi cài đặt, tỷ lệ người dùng thông thường thực hiện các bước trong eKYC sẽ khá đồng đều do họ có nhu cầu thật trong việc sử dụng ứng dụng, nên họ sẽ có động lực để thực hiện các bước tiếp theo, giống như ở kênh “tradingworks” dưới đây.
[]
Tuy nhiên, với các lượt cài đặt được tạo bởi Device Farms, do phần lớn được tạo nhằm mục đích ghi nhận lượt cài đặt, sẽ không có động lực để các thiết bị thực hiện những hành động khác trong ứng dụng, giống như các cài đặt từ kênh “Discord” như sau
[]
Ngoài ra, device farms thường dựa vào các thiết bị lỗi thời, giá rẻ với các hệ điều hành đời thấp. Với ví dụ dưới đây, các hệ điều hành đều đã khá cũ, thường thấy ở các dòng máy cũ giá rẻ và không còn được hỗ trợ để nâng cấp lên IOS 18 mới nhất hiện nay.
[]
Airbridge có thể giúp gì?
Thông thường, các doanh nghiệp đều gặp khó khăn lớn trong việc xác định các hình thức gian lận nêu trên, họ thường mất phần lớn thời gian, công sức, và tài chính trong việc nhận dạng và loại bỏ các lượt cài đặt gian lận. May mắn thay, với các tính năng Phát hiện gian lận của Airbridge, các marketer có thể triển khai chiến dịch an toàn hơn mà không lo chi phí phát sinh, tất cả đều được gói gọn trong một mức giá trọn gói.
Cụ thể hơn, với tính năng Abnormal Install Report của Airbridge được hỗ trợ hoàn toàn tự động bởi trí tuệ nhân tạo (AI), họ có thể quan sát số lượng lượt cài đặt đáng nghi theo từng nhóm gian lận, giúp cho quá trình nhận biết gian lận trở nên dễ dàng và tiết kiệm hơn bao giờ hết
[]
Với các vấn đề liên quan tới Device Farms, các nhà quảng cáo trong lĩnh vực Fintech nên kiểm tra các sub-publishers và vị trí quảng cáo có tỷ lệ giữ chân, chuyển đổi, và tương tác thấp bất thường. Ngoài ra, đừng quên chú ý xem có sự xuất hiện của thiết bị lỗi thời hoặc nhóm nhỏ địa chỉ IP lặp lại nhiều lần không. Tất cả các yếu tố trên nằm gói gọn trong tính năng Conversion Traffic và Retention, cho phép việc xác định các chuyển đổi đáng ngờ dựa trên đa dạng các tiêu chí như IP, Quốc Gia, hệ điều hành,.... Bởi vậy, doanh nghiệp Fintech sẽ không còn phải trả phí cho những lượt chuyển đổi cài đặt tới từ địa chỉ IP bị trùng lặp bất thường, các quốc gia khác không phải Việt Nam, hay những lượt cài đặt có tỷ lệ giữ chân thấp một cách đáng ngờ.
[]
### Gian lận phân bổ
Gian lận phân bổ là gì?
[Gian lận phân bổ](https://www.airbridge.io/vi/glossary/attribution-fraud) thường được nhận biết qua hiện tượng "organic cannibalization". Nếu số lượt cài đặt không được phân bổ (= organic installs) giảm trong khi số lượt cài đặt được phân bổ (= paid installs) tăng, mặc dù ngân sách quảng cáo vẫn duy trì ổn định, ta có thể cho rằng các lượt cài đặt tự nhiên (organic installs) bị "chiếm đoạt" và ghi nhận sai thành lượt cài đặt có trả phí (paid installs).
[]
“Triệu chứng của” Gian lận phân bổ
Organic Cannibalization thường xảy ra khi Thời gian từ nhấp chuột tới cài đặt (CTIT) quá ngắn hoặc có tỷ lệ bất thường giữ lượt nhấp và lượt cài đặt, khi có nhiều lượt nhấp nhưng không xảy ra cài đặt. Loại gian lận này được thực hiện bởi [click injection](https://www.airbridge.io/vi/glossary/click-injection) hoặc [click spamming](https://www.airbridge.io/vi/glossary/click-spam), đều là các hình thức gian lận phân bổ (Attribution Fraud).
#### Click injection
Click injection xảy ra khi một bên trung gian gian lận tìm cách xác định thời điểm ứng dụng được tải xuống và chèn một điểm chạm hay một tương tác giả ngay trước khi MMP ghi nhận hoàn tất cài đặt ứng dụng để chiếm quyền ghi nhận lượt cài đặt (winning touch point).
[]
Đây là một hình thức gian lận đặc thù trên các thiết bị Android, lợi dụng hệ thống broadcast của hệ điều hành này. Khi một ứng dụng mới được cài đặt trên thiết bị Android, một tín hiệu sẽ được gửi đến các ứng dụng khác. Hệ thống này được tạo ra nhằm cải thiện kết nối giữa các ứng dụng thuộc lĩnh vực Fintech trên thiết bị của người dùng, chẳng hạn để hỗ trợ deep linking hoặc đơn giản hóa quy trình đăng nhập.
#### Click Spamming
Click Spamming, hay click flooding, là một hình thức gian lận quảng cáo, trong đó kẻ gian lận tạo ra lượt nhấp giả để chèn vào hành trình của người dùng tự nhiên mà không có sự nhận biết hay đồng ý từ họ. Hành vi này có thể thực hiện thông qua các ứng dụng hoạt động ngầm, như công cụ dọn dẹp bộ nhớ hoặc tiết kiệm pin, tạo lượt nhấp bất kỳ lúc nào. Kết quả là dữ liệu bị sai lệch, làm mất độ tin cậy và dẫn đến đầu tư lãng phí vào các chiến dịch không hiệu quả.
[]
Airbridge có thể giúp gì?
Không chỉ dừng lại ở việc nhận biết, Airbridge còn sẵn sàng đồng hành với các đối tác Fintech phòng chống các loại gian lận, giúp họ hợp tác chặt chẽ với đội ngũ tư vấn của Airbridge để thiết lập các Fraud Validation Rules phù hợp với đặc thù chiến dịch. Đây là giải pháp mà mọi doanh nghiệp, bất kể lĩnh vực, đều có thể tận dụng tối đa. Điển hình như [Go2Joy](https://www.airbridge.io/en/case-studies/go2joy-fraud-prevention-detection), họ đã loại bỏ thành công 90% lượt cài đặt gian lận nhờ Airbridge.
Cụ thể hơn, các gian lận phân bổ như Click Injection hay Click Spamming có thể được giải quyết triệt để bởi tính năng Lag Time và Frequency Capping. Với Lag Time, doanh nghiệp có thể cài đặt từ chối gửi thông tin (hay còn gọi là postback) cho các lượt cài đặt có thời gian từ lúc nhấp chuột đến khi cài đặt (CTIT) dưới 10 giây, bởi người dùng không thể nhấp chuột, tải xuống, cài đặt và mở ứng dụng trong khoảng thời gian ngắn như vậy.
[]
Với Frequency Capping, họ có thể phòng tránh các địa chỉ IP hoặc các kênh có lượt nhấp giả mạo với tần suất lặp lại quá lớn trong một khoảng thời gian tuỳ chỉnh, do người dùng thông thường khó có thể thực hiện nhiều lượt nhấp trong thời gian ngắn như vậy.
[]
### SDK Spoofing
SDK Spoofing là gì?
[SDK Spoofing](https://www.airbridge.io/vi/glossary/sdk-spoofing) xảy ra khi kẻ gian lận hack các SDK của MMP (bao gồm thông tin đăng nhập, gói dữ liệu, v.v.) để gửi các tín hiệu giả mạo. Khi đó, kẻ gian sẽ tạo ra các lượt cài đặt giả nhưng trông giống như thật bằng cách sử dụng dữ liệu thiết bị thực. Đây còn được gọi là cuộc tấn công man-in-the-middle.
[]
Cụ thể, kẻ gian lận phá vỡ mã hóa SSL giữa SDK của phần mềm theo dõi và các máy chủ backend của nó để tạo ra hàng loạt lượt cài đặt thử nghiệm cho ứng dụng mục tiêu. Sau khi kẻ gian lận xác định được các URL đại diện cho một số hành động trong ứng dụng nhất định (in-app events), chúng sẽ thử nghiệm các phần động của URL để tạo ra các lượt cài đặt giả.
Khi đã thu thập đủ thông tin, kẻ gian có thể lặp lại quy trình này vô thời hạn, gây thất thoát lớn cho nhà quảng cáo.
“Triệu chứng” của SDK Spoofing
Tình trạng SDK Spoofing dễ thấy nhất khi xuất hiện nhiều transaction IDs hoặc user IDs không tồn tại trong hệ thống nội bộ (CRM, ERP) nhưng lại được ghi nhận trong dữ liệu từ MMP. Trong ví dụ dưới đây, ID user_161110 và tx_97776154 có thể được giả mạo, do chỉ có ở MMP mà không xuất hiện ở dữ liệu nội bộ gốc của CRM
[]
Bên cạnh đó, tỷ lệ chuyển đổi thấp bất thường và thời gian giữa các sự kiện như cài đặt ứng dụng, mở ứng dụng và thực hiện hành động đầu tiên quá ngắn hoặc không hợp lý, khác biệt so với hành vi người dùng thực tế cũng là yếu tố không thể bỏ qua trong việc xác định hành vi SDK Spoofing.
Một số tác động đến doanh nghiệp của hình thức này:
- Lãng phí ngân sách UA (User Acquisition): Chi tiêu quảng cáo bị tiêu tốn cho các lượt cài đặt hoặc tương tác giả mạo.
- Ảnh hưởng đến mọi chỉ số KPI kinh doanh: Đặc biệt là các chỉ số quan trọng như chuyển đổi (conversions) và tỷ lệ giữ chân người dùng (retention), làm sai lệch dữ liệu và ảnh hưởng đến hiệu quả chiến lược quảng cáo.
Airbridge có thể giúp gì?
Nhằm phát hiện tình trạng SDK Spoofing, Airbridge còn cung cấp báo cáo Raw Data, không chỉ cung cấp đầy các thông số đo lường giúp các doanh nghiệp hiểu hơn về tổng quan chiến dịch, mà còn có thể so sánh user_id hoặc transaction_id để chắc chắn rằng số liệu của MMP khớp với số liệu từ các hệ thống nội bộ như CRM, ERP. Tính năng SDK Signature sau đó sẽ giúp họ loại bỏ các trường hợp giả mạo MMP nêu trên một cách nhanh chóng và hiệu quả.
[]
## Kết luận
Gian lận quảng cáo trong ngành Fintech ngày càng phức tạp với các hình thức như bot fraud, mistargeting, device farms và gian lận phân bổ, gây thất thoát ngân sách, sai lệch dữ liệu và ảnh hưởng đến hiệu quả chiến dịch. Airbridge cung cấp các giải pháp mạnh mẽ như sử dụng dữ liệu thô, thiết lập quy tắc xác thực gian lận, và kiểm tra chặt chẽ trên nền tảng MMP, giúp doanh nghiệp Fintech nhanh chóng phát hiện và ngăn chặn các hành vi gian lận, bảo vệ ngân sách và tối ưu hóa hiệu quả quảng cáo.
---
## Đột Phá Tăng Trưởng Game Hybrid & Hyper-Casual với Chiến Lược UA, Monetization và Quảng cáo Sáng tạo AI năm 2025
URL: https://www.airbridge.io/vi/blog/hybrid-hyper-casual-game-strategy-2025-with-ua-monetization-ai-creatives
Language: Vietnamese
Published: 2025-02-11
## **Thẳng thắn và không sáo rỗng - podcast gaming dành cho những ý tưởng & insight sắc bén bên tách cafe lúc 4 giờ sáng. **
Đó là châm ngôn của [Two and a Half Gamers](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH), nơi bộ ba diễn giả [Matej](https://www.linkedin.com/in/matejlancaric/), [Jakub](https://www.linkedin.com/in/jakubremiar/), và [Felix](https://www.linkedin.com/in/felix-braberg-7a732b51/) cùng thảo luận mà không có bất kỳ sự chỉnh sửa cắt ghép nào. Họ kết hợp những góc nhìn sắc bén với những màn tranh luận nảy lửa, tạo nên một không gian đối thoại vừa thực tế, vừa sâu sắc. Nhưng trên hết, họ luôn giữ vững sự trung thực tuyệt đối. Trong thời đại mà tính chân thực là yếu tố cốt lõi, phong cách thẳng thắn và không khoan nhượng của họ đã thu hút đông đảo cộng đồng game mobile – những người luôn tìm kiếm lời khuyên thực tế về UA (thu hút người dùng), cách kiếm tiền từ game và tối ưu hóa nội dung quảng cáo.
Trong hơn một thập kỷ làm nghề, họ đã chứng kiến sự phát triển của đủ mọi thể loại game, từ bom tấn triệu đô và indie game đến casual và hardcore game – tất cả đều hướng đến mục tiêu bứt phá trên thị trường toàn cầu. Không chỉ đứng ngoài quan sát, họ đã trực tiếp làm việc với các studio, nhà phát hành và đội ngũ phát triển game để giúp họ tìm ra công thức thành công.
[]
[Roi](https://www.linkedin.com/in/roinam/), CEO của [Airbridge](http://www.Airbridge.io), đã có buổi trò chuyện cùng Two and a Half Gamers để chia sẻ những chiến lược giúp game hybrid và hyper-casual vươn lên dẫn đầu trong năm 2025. Bạn đã sẵn sàng hoà vào cuộc phỏng vấn giàu giá trị – có thể chính là cú hích giúp game của bạn bùng nổ trên thị trường hay chưa?
### 🎥 **Xem cuộc phỏng vấn đầy đủ với Two and a Half Gamers trên Youtube **
[]
Trong những năm qua, xu hướng toàn cầu hóa game hybrid và hyper-casual ngày càng rõ rệt. Các tựa game này đang xuất hiện ngày càng nhiều trong khu vực Á-Âu và châu Á, đặc biệt là từ các quốc gia như Thổ Nhĩ Kỳ, Cộng hoà Síp, Việt Nam, Trung Quốc và Hàn Quốc. Các nhà phát triển game ở đây không chỉ nhắm đến thị trường nội địa mà còn hướng đến người chơi trên toàn cầu, với tốc độ mở rộng và mức độ ảnh hưởng vô cùng mạnh mẽ. Matej chia sẻ:
> "Chúng tôi thực sự kinh ngạc trước sự tiên phong của Trung Quốc. Ví dụ, khi Lucky Defense ra mắt toàn cầu, một nửa ngành công nghiệp game ở Mỹ và châu Âu còn chưa biết đến nó. Nhưng ở Trung Quốc, chỉ hai tuần sau khi game phát hành, đã có 10 công ty bắt tay vào phát triển phiên bản clone hoặc bản nhái hoặc cải tiến, thậm chí đã có nguyên mẫu sẵn sàng. Thật điên rồ!"
## **Đa Dạng Hóa UA: Bắt Đầu Ngay, Mở Rộng Dần**
User Acquisition (Thu hút người dùng) ngày càng trở nên thách thức hơn bao giờ hết, nhưng khi thị trường trở nên khắc nghiệt, những người giỏi nhất lại càng phát huy khả năng sáng tạo hơn. Trong bối cảnh cạnh tranh khốc liệt của ngành game mobile hiện nay, đa dạng hóa chiến lược UA không còn là một lựa chọn – mà trở thành điều bắt buộc. Matej giải thích:
> "Giờ đây, bạn cần triển khai UA trên nhiều kênh ngay từ đầu, ý tôi là ngay từ ngày đầu tiên – vài năm trước đây điều này không phải là tiêu chuẩn."
Một điểm khởi đầu vững chắc là tập trung vào bốn nền tảng chủ chốt: Facebook, TikTok, AppLovin, và Unity, sau đó dần mở rộng sang các kênh khác như Mistplay và [adjoe](https://www.airbridge.io/en/blog/adjoe-integration). Cách này giúp bạn tiếp cận đối tượng rộng hơn, đồng thời giảm phụ thuộc vào một kênh duy nhất – yếu tố có thể dẫn đến CPI cao và ROI thấp. Matej tiếp tục nói:
> "Tôi đã thực hiện một khảo sát với các subscriber của mình, và trung bình mỗi nhà quảng cáo đang sử dụng từ 8 đến 12 kênh UA, với ngân sách dao động từ 100.000 USD/tháng đến vài triệu USD."
Đa dạng hóa UA không chỉ dừng lại ở việc chọn nền tảng, mà còn phải thử nghiệm nhiều loại chiến dịch khác nhau. Một chiến lược cân bằng sẽ kết hợp nhiều phương pháp tối ưu lợi nhuận như: chiến dịch ROAS mua hàng trong ứng dụng (in-app purchase) cùng với những phương pháp tiết kiệm chi phí như chiến dịch tối ưu hóa theo sự kiện trong ứng dụng.
Thành công trong UA ngày nay đòi hỏi bạn phải hành động từ sớm, thử nghiệm đa kênh, và điều chỉnh nhanh chóng dựa trên dữ liệu. Những ai không theo kịp tốc độ thay đổi sẽ sớm bị bỏ lại phía sau.
## **Hybrid Monetization: Điểm Cân Bằng Hoàn Hảo Cho Doanh Thu**
Two and a Half Gamers định nghĩa hybrid monetization là mô hình doanh thu trong đó 40-60% đến từ quảng cáo và 40-60% từ in-app purchases (IAP). Khi kết hợp hai nguồn doanh thu này, mô hình hybrid không chỉ giúp tạo ra lợi nhuận nhanh chóng từ quảng cáo mà còn đảm bảo sự tăng trưởng bền vững nhờ doanh thu từ IAP.
Để đạt được thành công với hybrid monetization, game của bạn cần có một hệ thống kinh tế vững chắc với "spend depth" đủ sâu—tức là mức chi tiêu tối đa mà người chơi có thể đầu tư để tăng tiến độ hoặc sức mạnh trong game. Điều này đòi hỏi xây dựng nền kinh tế game hợp lý, thiết kế hệ thống phát triển nhân vật cuốn hút và mở khóa các tính năng theo lộ trình rõ ràng. Khi người chơi tiến xa hơn, bạn có thể tích hợp các yếu tố monetization (kiếm tiền) một cách có chiến lược, bao gồm quảng cáo xen kẽ (interstitial ads), video thưởng (rewarded videos) và banner ads. Những công ty như Voodoo và SayGames đã áp dụng thành công chiến lược này. Jakub nhấn mạnh:
> “Với một tựa game có vòng đời dài, khoảng 3-4 năm, việc xây dựng một nền kinh tế dựa trên IAP là điều tối quan trọng. Nó giúp người chơi luôn có cơ hội chi tiêu trong game, từ đó duy trì sự phát triển lâu dài."
[]
Tuy nhiên, thành công của hybrid monetization phụ thuộc vào chiến lược cá nhân hóa và khả năng thu hút người dùng sắc bén. Felix bổ sung:
> “Điều quan trọng nhất cần nhớ đó là: user acquisition chính là mũi nhọn của chiến lược—nó ảnh hưởng trực tiếp đến thiết kế game và cách phân khúc quảng cáo của bạn. Nếu không làm tốt bước này, bạn sẽ không thể tối ưu hóa doanh thu từ quảng cáo và duy trì CPI thấp."
Với phân khúc người dùng thông minh, các ứng dụng game có thể tùy chỉnh trải nghiệm người dùng dựa trên mục tiêu chiến dịch UA. Người chơi đến từ các chiến dịch tập trung vào IAP có thể sẽ không thấy quảng cáo ngay từ đầu, trong khi những người đến từ chiến dịch tối ưu ROAS từ quảng cáo sẽ sớm tiếp cận với banner ads và interstitial ads để đẩy nhanh quá trình monetization.
Các công cụ phân bổ như [Airbridge](https://www.airbridge.io/) giúp game mobile thực hiện phân khúc người dùng theo thời gian thực, đảm bảo áp dụng chiến lược kiếm tiền phù hợp với từng nhóm người chơi. Jakub cảnh báo:
> "Đừng lãng phí toàn bộ chiến lược kiếm tiền (monetization) của bạn bằng cách đưa ra ưu đãi cao cho những người dùng chi tiêu ít. Họ sẽ cảm thấy không phù hợp và thậm chí không muốn mua nữa."
## **Quảng cáo Sáng tạo bằng AI: Tăng Trưởng với Tốc Độ Và Số Lượng**
Thành công trong UA không chỉ đến từ việc tạo ra quảng cáo hay—mà còn là tạo ra số lượng lớn quảng cáo chất lượng trong thời gian ngắn. Và đây chính là lúc các mẫu quảng cáo sáng tạo được hỗ trợ bởi AI phát huy tác dụng.
Playables được xem là một trong những định dạng quảng cáo hiệu quả nhất cho game hybrid và hyper-casual. Tuy nhiên, dù mang lại hiệu suất cao, việc sản xuất và triển khai playables trên quy mô lớn vẫn là một thách thức. Matej nhấn mạnh:
> "Bạn cần quảng cáo playables cho tất cả các mạng quảng cáo SDK. Nếu bạn chạy Facebook và Google thì không bắt buộc, nhưng với AppLovin, playables là yếu tố không thể thiếu.Các nền tảng như Mintegral và Moloco cũng cực kỳ coi trọng điều nay."
Hiện nay, các nhà phát triển có thể đơn giản hóa quy trình này bằng cách bắt đầu từ game tutorial và mở rộng quy mô hiệu quả với các công cụ như [playablemaker.com](http://playablemaker.com), giúp tạo ra quảng cáo tương tác hấp dẫn từ nội dung video có sẵn.
AI không chỉ là công cụ hỗ trợ—mà đã trở thành yếu tố bắt buộc. Matej giải thích:
> "Trên Facebook và TikTok, bạn cần cập nhật nội dung quảng cáo liên tục—khoảng 2-3 ngày một lần. Nếu không có AI hỗ trợ, việc này gần như không thể hoàn thành."
Cách tiếp cận tối ưu là cân bằng giữa chất lượng và số lượng. Khi các công cụ như Midjourney phát triển hình ảnh tĩnh sang video, từ 2D sang 3D, việc tạo ra nội dung quảng cáo có tính kể chuyện, kết hợp AI voice-over trở nên dễ dàng hơn. Matej mô tả:
> "Cách cân bằng lý tưởng là bạn có một đội ngũ sáng tạo chuyên làm ra nội dung chất lượng, độc đáo. Đồng thời, bạn tận dụng AI để gia tăng số lượng sản phẩm một cách nhanh chóng."
Với những công cụ như Poolday AI và Runway ML, việc sản xuất nội dung sáng tạo giờ đây không chỉ dành riêng cho đội ngũ thiết kế mà cả các UA Manager cũng có thể thực hiện dễ dàng. Matej chia sẻ:
> “Tôi sử dụng Poolday AI. Bạn có thể xuất một video với 10 nhân vật AI khác nhau chỉ trong khoảng thời gian mà 100 người cần đến 5 phút để làm. Năm nay, yếu tố then chốt chính là tốc độ sản xuất nội dung quảng cáo. Bạn có thể tạo ra bao nhiêu quảng cáo? Vì thuật toán của các nền tảng hiện nay đang 'đói' nội dung hơn bao giờ hết."
[]
## **UA Manager: Người Đảm Nhận Nhiều Vai Trò**
UA Manager giờ đây không chỉ đơn thuần là người mua quảng cáo. Vai trò của họ đã mở rộng hơn rất nhiều. Ngoài việc tối ưu ROAS, họ còn thực hiện các bài test marketability (kiểm tra tiềm năng thị trường), phối hợp với product team trong các đợt soft launch (phát hành thử nghiệm(, và thậm chí đảm nhiệm cả việc thiết kế quảng cáo. Matej nhận xét:
> "Là một UA Manager, bạn cần đảm nhận được nhiều vai trò, làm nhiều nhiệm vụ cùng một lúc."
Những ngày chỉ dựa vào test CPI đã qua. Ngày nay, các chiến lược UA bao gồm thử nghiệm nhiều phiên bản onboarding được thiết kế dưới dạng mini-game để nhanh chóng xác định phương pháp hiệu quả nhất. Jakub chia sẻ về xu hướng này:
> "Những tựa game thành công thường lấy 60-70 phút đầu tiên của game, chia thành các phiên bản A, B, C, D và chạy thử nghiệm. Vẫn là cùng một game, nhưng họ có thể thấy phiên bản onboarding nào hoạt động tốt nhất."
## **Predictive LTV: Ra Quyết Định Theo Thời Gian Thực**
Mô hình Predictive Lifetime Value (pLTV) đã trở thành công cụ thiết yếu giúp các nhà phát triển game mobile ra quyết định nhanh hơn dựa trên dữ liệu. Matej nhấn mạnh:
> "Bạn cần có số liệu cập nhật hàng ngày, và lý tưởng nhất là theo nhóm cohort, ví dụ trong 3 ngày đầu tiên. Nếu bạn chờ đến 7 ngày mới phân tích dữ liệu và điều chỉnh, thì đã quá muộn rồi."
Dù pLTV rất quan trọng trong việc dự báo hiệu suất và tối ưu hóa chiến dịch, nhiều công ty game vẫn gặp khó khăn trong việc duy trì nó. Matej chỉ ra một lỗi phổ biến:
> "Hầu hết các công ty thất bại vì họ xây dựng mô hình pLTV rồi bỏ đó, không cập nhật hoặc xem xét dữ liệu lịch sử. Sau một thời gian, mọi thứ trở nên lộn xộn. Bạn cần hiệu chỉnh lại mô hình theo 3 tháng một lần để đảm bảo tính chính xác."
Vì việc duy trì mô hình pLTV nội bộ khá phức tạp, nhiều nhà phát triển game mobile tìm đến các giải pháp như [Airbridge](https://www.airbridge.io/vi/product/airbridge-ai), giúp họ thực hiện pLTV, phân tích cohort, và theo dõi tỷ lệ giữ chân theo từng phút/giờ một cách hiệu quả.
## **Bản Địa Hóa: Tinh Chỉnh Game Cho Từng Khu Vực Người Dùng**
Để vươn tầm toàn cầu, chỉ dịch thuật thôi là chưa đủ—bản địa hóa (localization) chính là chìa khóa giúp tựa game phù hợp với từng thị trường. Jakub nhận xét:
> "Các tính năng trong game tại châu Á phức tạp hơn gấp 3 lần so với phương Tây, bởi vì người chơi tại đây có xu hướng thích những trò chơi đa tính năng hơn."
Người chơi châu Á thường ưa chuộng các game có cơ chế phức tạp, hệ thống phát triển đa tầng và nhiều tính năng chồng chéo. Trong khi đó, các thiết kế nặng tính năng này có thể khiến người chơi phương Tây cảm thấy quá tải. Vì vậy, để thành công tại thị trường phương Tây, game cần được tinh giản và dễ tiếp cận hơn. Jakub tiếp tục:
> "Khi đưa game sang thị trường phương Tây, bạn cần làm cho nó mượt mà hơn, thân thiện hơn và dễ tiếp cận hơn. Ví dụ, Rush Royale của MY.GAMES thực chất là một phiên bản tinh giản hơn của Random Dice."
## Work Smarter, Not Harder
Những insights từ Two and a Half Gamers vạch ra một con đường rõ ràng cho ngành game mobile: Thích nghi nhanh hoặc bị bỏ lại phía sau. Thành công trong năm 2025 không chỉ nằm ở việc sở hữu một tựa game xuất sắc, mà còn là làm chủ ba yếu tố: đa dạng hóa UA, tối ưu hóa hybrid monetization và sáng tạo nội dung quảng cáo bằng AI. Matej khẳng định:
> "Thành công đến từ việc cân bằng giữa chiến lược UA, chiến lược kiếm tiền sáng tạo và quy trình sản xuất nội dung quảng cáo có thể dễ dàng mở rộng."
[]
Để khám phá nhiều insight hơn, hãy theo dõi [Two and a Half Gamers podcast](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH) và [Airbridge](https://www.airbridge.io/vi)
### **Đột Phá Tăng Trưởng Với Airbridge**
Nếu bạn muốn nâng tầm chiến lược marketing game của mình, hãy khám phá [Airbridge](https://www.airbridge.io/). Nền tảng của Airbridge cung cấp:
- **Phân tích Cohort nâng cao cho LTV:** Theo dõi giá trị vòng đời của người chơi một cách chính xác, với khả năng phân tích theo từ khóa, nội dung quảng cáo, quốc gia, hệ điều hành và nhiều yếu tố khác.
- **Tích hợp Ad Mediation: **Tổng hợp dữ liệu từ nhiều mạng quảng cáo vào một nền tảng duy nhất (Google AdMob, AppLovin MAX, Unity LevelPlay, v.v.).
- **Predictive LTV (pLTV): **Dự đoán doanh thu trong tương lai theo thời gian thực, giúp đưa ra quyết định nhanh chóng và chính xác dựa trên dữ liệu.
- **Phân tích Retention (Tỷ lệ giữ chân) theo phút/giờ: **Hiểu rõ hành vi người chơi ở từng giai đoạn, từ đó rút ra các insight có thể áp dụng ngay.
- **Gói giá minh bạch và cạnh tranh**: Cung cấp phân tích chuyên sâu cấp doanh nghiệp mà tuyệt đối không có phí ẩn hay chi phí bổ sung phức tạp.
Dù bạn là một studio indie hay một nhà phát hành lớn, Airbridge mang đến công cụ thiết yếu để bạn duy trì lợi thế cạnh tranh trong thị trường đầy biến động. Truy cập [airbridge.io](https://airbridge.io/) để tìm hiểu thêm.
### **Trở Thành Chuyên Gia Với Two and a Half Gamers**
Để cập nhật những chiến lược mới nhất, nghiên cứu điển hình và các cuộc thảo luận chuyên sâu về user acquisition, monetization và tối ưu hóa nội dung quảng cáo, hãy:
1. Theo dõi [kênh YouTube Two and a Half Gamers](https://www.youtube.com/@2.5gamers) để xem các hướng dẫn chuyên sâu, phỏng vấn chuyên gia và các case study thực tiễn.
2. Nghe podcast [podcast Two and a Half Gamers](https://open.spotify.com/show/00q9jH5ATTsklO5VUjGeqH) để cập nhật những phân tích và góc nhìn chuyên sâu về ngành game.
3. Thưởng thức [album Two and a Half Gamers](https://open.spotify.com/artist/2o80XDfSqkDIqmuc5tOZdt?si=67Hn_PzAS_Sw6w-QS-XWRw), nơi bạn có thể tìm thấy những ca khúc bắt tai về cách thúc đẩy tăng trưởng game.
4. Kết nối với [Matej](https://www.linkedin.com/in/matejlancaric/), [Jakub](https://www.linkedin.com/in/jakubremiar/) và [Felix](https://www.linkedin.com/in/felix-braberg-7a732b51/) trên LinkedIn để thảo luận sâu hơn về những thách thức tăng trưởng mà bạn đang gặp phải.
Hãy nhanh chóng cập nhật, thích nghi và sẵn sàng bùng nổ với game hybrid & hyper-casual trong năm 2025!
---
## Affiliate Marketing là gì? Chiến lược thành công chuẩn 2025
URL: https://www.airbridge.io/vi/blog/what-is-affiliate-marketing
Language: Vietnamese
Published: 2025-02-03
## **Affiliate Marketing là gì và tại sao nó đang định hình xu hướng marketing 2025?**
Bạn từng thấy các chương trình “Giới thiệu bạn bè – nhận ưu đãi” trong ứng dụng mua sắm? Hay từng click vào một đường link giới thiệu sản phẩm từ influencer trên YouTube hoặc Instagram?
Đó chính là affiliate marketing – hình thức tiếp thị liên kết, nơi sản phẩm được quảng bá thông qua các đường link gắn mã theo dõi. Trong bối cảnh mobile marketing cạnh tranh khốc liệt, affiliate marketing đang nổi lên như một chiến lược giúp gia tăng ROI hiệu quả và tiết kiệm chi phí quảng cáo.
Trong bài viết này, chúng ta sẽ cùng tìm hiểu khái niệm cơ bản của affiliate marketing, lý do bạn nên bắt đầu ngay trong năm 2025, và gợi ý cách triển khai thành công.
## Hiểu đúng về Affiliate Marketing
> 💡 Affiliate Marketing là gì?
> Affiliate marketing là mô hình tiếp thị dựa trên hiệu suất, nơi doanh nghiệp chi trả hoa hồng cho bên thứ ba (gọi là affiliate – đối tác liên kết) khi họ tạo ra chuyển đổi như lượt truy cập, đơn hàng hoặc lượt cài đặt app.
Affiliate có thể là công ty, cá nhân hoặc influencer, và họ được trả tiền khi người dùng thực hiện hành động qua link giới thiệu có gắn mã theo dõi riêng biệt.
Doanh nghiệp chỉ trả tiền khi có kết quả thực tế, nên chi phí quảng cáo được tối ưu hơn rất nhiều so với các phương thức truyền thống. Một số nền tảng affiliate nổi bật có thể kể đến như [**Amazon Associates**](https://affiliate-program.amazon.com/)**,** [**Walmart Affiliate Program**](https://creator.walmart.com/?ebebhghdefhbdjldreibbfljnbirebktvvuvighttigg&gad_source=1&gclid=Cj0KCQiAq-u9BhCjARIsANLj-s0MT8Ec7Vze-Ij9f4AjbCnEzffcweOmWvfgghkcQ87xnuu27WtXQD0aAkAdEALw_wcB&gclsrc=aw.ds), hay [**Rakuten Advertising**](https://rakutenadvertising.com/affiliate/).
### Ví dụ về Affiliate Marketing
**Trường hợp A – Affiliate nhận hoa hồng:
**Một YouTuber đăng video review sản phẩm kèm theo link mua hàng. Khi người xem nhấp vào link và mua sản phẩm, YouTuber nhận được hoa hồng.
**Trường hợp B – Người dùng nhận phần thưởng:
**Một app mua sắm triển khai chương trình “Giới thiệu bạn bè”. Người dùng nhận phần thưởng khi bạn bè cài app hoặc mua hàng lần đầu thông qua link giới thiệu cá nhân.
**Lợi ích của Affiliate Marketing
**Affiliate marketing đang dần trở thành xu hướng chủ đạo trong marketing mobile nhờ vào những ưu điểm nổi bật:
1. Quản lý chi phí hiệu quả
Chỉ trả tiền khi có chuyển đổi (mua hàng, cài đặt, đăng ký...), giúp tối thiểu hóa lãng phí ngân sách.
2. Tận dụng đa kênh
Dễ dàng triển khai trên blog, mạng xã hội (YouTube, Instagram, TikTok...), ứng dụng và các nền tảng khác.
3. Mở rộng nhanh chóng
Tăng tốc độ tiếp cận thị trường bằng cách mở rộng mạng lưới đối tác liên kết.
4. Tối đa hóa ROI
Theo dõi hiệu suất chính xác, từ đó phân bổ ngân sách hợp lý và mang lại lợi nhuận cao hơn.
## Bắt đầu như thế nào?
1. **Tìm đúng đối tác – Xây đúng mô hình**
Bước đầu tiên là xác định mục tiêu và lựa chọn đối tác affiliate phù hợp. Bạn có thể hướng đến tăng đơn hàng, cải thiện doanh số theo danh mục sản phẩm, tăng số lượng thành viên đăng ký hoặc tỷ lệ chuyển đổi gói subscription.
2. **Thiết kế hệ thống trả thưởng minh bạch**
Xây dựng cơ chế trả thưởng rõ ràng dựa trên hiệu suất. Có thể lựa chọn giữa các mô hình như CPC (cost per click – tính theo lượt nhấp), CPA (cost per acquisition – tính theo lượt chuyển đổi), và CPS (cost per sale – tính theo đơn hàng)
3. **Tạo và phân phối link affiliate**
Khi đã xác định được mục tiêu, đối tác và cơ chế trả thưởng, bạn cần tạo ra các link affiliate để đối tác quảng bá. Link phải có mã tracking riêng để hệ thống có thể đo lường chính xác hiệu suất từng chiến dịch. Nếu link không hoạt động hoặc không được theo dõi đúng cách, bạn sẽ mất dữ liệu chuyển đổi – đồng nghĩa với thất thoát doanh thu.
4. **Theo dõi hiệu suất bằng công cụ tracking**
Sau khi chiến dịch bắt đầu, bạn cần theo dõi hiệu quả chiến dịch liên tục để đảm bảo trả thưởng chính xác cho đối tác. Hãy sử dụng các công cụ tracking để đo lường số lượt cài đặt, mua hàng... đến từ mỗi link và tính toán phần thưởng tương ứng.
**Bắt đầu với Airbridge ngay hôm nay!**
Tăng tốc chiến dịch affiliate marketing của bạn với Airbridge. Gói DeepLink Plan từ Airbridge giúp bạn tối ưu hóa toàn bộ quy trình bằng cách tạo và quản lý link tiếp thị một cách linh hoạt và hiệu quả.
Nhờ công nghệ deep linking, Airbridge mang đến trải nghiệm liền mạch cho cả người dùng và đối tác affiliate – từ click đến chuyển đổi không hề gián đoạn. Với hệ thống Airbridge DeepLinks, bạn có thể theo dõi hiệu suất affiliate theo thời gian thực, phân phối hoa hồng chính xác, và tối đa hóa ROI của toàn bộ chiến dịch
- Tự động tạo link tracking qua API
- Tương thích hoàn toàn với mọi hệ thống và nền tảng — từ iOS, Android đến web và cả offline
- Theo dõi hiệu suất siêu chính xác và phân tích dữ liệu theo thời gian thực
- Mô hình định giá linh hoạt theo MAU — tăng trưởng đến đâu, trả phí đến đó
---
## Nhìn lại 2024 cùng Airbridge
URL: https://www.airbridge.io/vi/blog/airbridge-recap-2024
Language: Vietnamese
Published: 2025-01-14
Với bối cảnh số hóa tiếp tục phát triển với tốc độ chóng mặt, năm 2024 đã trở thành một năm đánh dấu sự tăng trưởng chưa từng có, với các mối quan hệ đối tác chiến lược và những đổi mới đột phá đối với Airbridge. Chúng tôi không chỉ thích nghi mà còn dẫn đầu, mở rộng giới hạn trong đo lường di động và phân tích đa nền tảng.
Hãy cùng nhìn lại những cột mốc và con số đã tạo nên thành công của chúng tôi.
[]
## Tăng trưởng 4,3 lần số lượng khách hàng trong ngành game
Năm 2024, Airbridge vươn lên mạnh mẽ trong ngành game, khi số lượng khách hàng của chúng tôi trong lĩnh vực này tăng 4,3 lần so với năm trước. Từ hyper-casual, RPG, racing, AR music đến các video game tương tác, Airbridge được các nhà phát triển và studio tin tưởng lựa chọn làm đối tác MMP hàng đầu.
### Giải pháp tối ưu hóa tăng trưởng game
Thành công trong lĩnh vực game của Airbridge đến từ khả năng đo lường hiệu suất vượt trội. Các studio game thường gặp khó khăn trong việc theo dõi và phân tích chiến dịch trên nhiều kênh, nền tảng, quốc gia và nhóm đối tượng khác nhau. Chúng tôi hiểu rõ vấn đề và đã cung cấp giải pháp phù hợp cho khách hàng theo từng chiến dịch của họ. Với bộ công cụ chuyên biệt dành riêng cho tăng trưởng game, Airbridge mang đến khả năng phân tích dữ liệu thời gian thực chi tiết, giúp các studio và nhà phát triển đưa ra quyết định tối ưu hóa chiến dịch hiệu quả.
[Webzen](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization) đã sử dụng Airbridge để tối ưu hóa quảng cáo dạng ngắn ở cấp độ sáng tạo và áp dụng tính năng Predictive Lifetime Value (pLTV) để xác định trước các kênh có doanh thu kỳ vọng cao nhất, giúp tối ưu hóa ngân sách marketing.
Khi nhu cầu dành cho game PC và console ngày càng tăng, Airbridge đã mở rộng khả năng phân tích đa nền tảng, hỗ trợ đo lường hiệu suất trên tất cả các kênh, bao gồm ứng dụng di động, PC và console.
[Nexon](https://www.airbridge.io/en/case-studies/nexon-cross-platform-tracking) đã áp dụng giải pháp của Airbridge để đo lường hiệu suất quảng cáo cho game PC và console "[The First Descendant](https://www.playstation.com/en-us/games/the-first-descendant/)", với các chỉ số đáng tin cậy tương đương như trên game mobile.
### Mở rộng sức ảnh hưởng trong ngành game
Năm 2024, Airbridge đã tham gia và tạo dấu ấn tại nhiều hội nghị game hàng đầu thế giới như G-Star, Vietnam Games Forum, ChinaJoy, GDC và Gamescom.
Không chỉ dừng lại ở các hội nghị, chúng tôi còn hợp tác chiến lược với hơn 70 phương tiện truyền thông và nhà cung cấp giải pháp, bao gồm các tích hợp đáng chú ý như Mintegral tROAS, AVOW, ayeT và adjoe. Airbridge cũng đã cung cấp những hiểu biết sâu về thị trường game Hàn Quốc để cùng adjoe phối hợp thực hiện báo cáo "[Mobile Gamers of Korea 2025](https://adjoe.io/mobile-gamers-of-korea/)" dựa trên dữ liệu kéo hơn một thập kỷ.
[]
## Tăng 90% số lượng ứng dụng sử dụng Airbridge
Năm 2024 chứng kiến sự gia tăng đáng kể với 90% số lượng ứng dụng sử dụng Airbridge để theo dõi hiệu suất so với năm 2023. Điều này phản ánh sự tin tưởng ngày càng cao vào khả năng cung cấp thông tin chi tiết và tối ưu hóa chiến dịch của chúng tôi.
### Mở rộng tích hợp và quan hệ đối tác
Chúng tôi tiếp tục mở rộng hệ sinh thái bằng cách tăng cường tích hợp với các giải pháp martech hàng đầu như Amplitude và Braze, đồng thời mở rộng phạm vi thông qua các nền tảng quảng cáo toàn cầu như Microsoft Ads và Digital Turbine.
MyRealTrip đã tăng tỷ lệ chuyển đổi lên gấp 12 lần, một con số đáng kể ghi có Airbridge đồng hành cùng họ. [Greenlabs](https://www.airbridge.io/en/case-studies/greenlabs-crm-data-integration) tận dụng tích hợp Airbridge-Braze để tăng gấp đôi tỷ lệ mở của chiến dịch tái kích hoạt, sử dụng tin nhắn cá nhân hóa dựa trên phân tích chính xác từ Airbridge.
Khả năng tích hợp đa dạng với các nền tảng của Airbridge giúp các doanh nghiệp dễ dàng triển khai chiến lược marketing toàn diện và hiệu quả.
### Hỗ trợ khách hàng vượt trội
Đội ngũ hỗ trợ của Airbridge tiếp tục mang đến những chiến lược tùy chỉnh, giúp khách hàng đạt được mục tiêu tăng trưởng của mình. Webzen tăng 90% tỷ lệ đăng nhập người chơi [Mu Monarch](https://www.airbridge.io/en/case-studies/mu-monarch-creative-optimization) bằng cách tận dụng khả năng đo lường mạnh mẽ của Airbridge kết hợp với sự hỗ trợ tận tình từ đội ngũ của chúng tôi. [Vortex Gravity](https://www.airbridge.io/en/case-studies/vortexgravity-cut-mmp-cost) cũng đã gặt hài được nhiều lợi ích, không chỉ thành công trong việc cắt giảm 50% chi phí MMP hàng năm mà còn nhận được tư vấn hỗ trợ chiến lược nhanh chóng thông qua kênh trao đổi mà họ ưu tiên, WeChat.
[]
## 14,15 triệu báo cáo được tạo ra
Năm 2024, khách hàng của chúng tôi đã tạo ra hơn 14 triệu báo cáo, sử dụng 10 loại báo cáo khác nhau của Airbridge. Những báo cáo này cung cấp thông tin chi tiết chính xác và có hệ thống về tăng trưởng ứng dụng, hỗ trợ phân tích dữ liệu theo thời gian thực và tối ưu hóa chiến dịch hiệu quả hơn. [Shmoody](https://www.airbridge.io/en/case-studies/shmoody-cross-platform-tracking) tận dụng bảng điều khiển tùy chỉnh của Airbridge để hợp nhất hiệu suất Android, iOS và SKAN trong một giao diện duy nhất, giúp đơn giản hóa quá trình phân tích dữ liệu và đẩy nhanh tiến độ đạt 1 triệu lượt tải. [Buddy.ai](http://buddy.ai) sử dụng tích hợp chi phí và các chỉ số ROAS của Airbridge để đạt được vị trí Top 10 trên App Store, minh chứng cho sức mạnh của hệ thống phân tích toàn diện trong việc thúc đẩy hiệu suất ứng dụng. [Segari](https://www.airbridge.io/en/case-studies/segari-overview-dashboard) cắt giảm 50% thời gian tổng hợp và phân tích dữ liệu nhờ bảng điều khiển tổng quan của Airbridge.
## 86,13 triệu Tracking Link được tạo
Người dùng Airbridge đã tạo hơn 86 triệu liên kết tracking trong năm 2024, mang lại con số ấn tượng 2.628 nghìn tỷ lượt nhấp trong suốt năm. Điều này tương đương với trung bình 20 tỷ lượt nhấp mỗi tháng, gấp đôi so với mức trung bình của công cụ rút gọn URL miễn phí [Bitly](https://bitly.com/pages/resources/press/bitly-wraps-2022-surpassing-100m-in-arr-and-over-500k-global-customers).
Liên kết tracking của chúng tôi không chỉ dừng lại ở việc tạo liên kết, mà còn cung cấp khả năng deep linking và điều hướng liền mạch trên mọi nền tảng và hệ điều hành. Đặc biệt, sự ra mắt của [Airbridge DeepLink Plan](https://www.airbridge.io/deeplink-plan) vào quý 4 đã thu hút sự chú ý lớn, nhất là trong bối cảnh Firebase Dynamic Links dự kiến sẽ ngừng hoạt động vào ngày 25 tháng 8 năm 2025.
## Hướng tới năm 2025: Đổi mới với AI
Năm nay, Airbridge sẵn sàng mang đến cho bạn những công cụ đo lường mạnh mẽ hơn với các tính năng hỗ trợ bởi AI. Chúng tôi đang giới thiệu các khả năng AI nâng cao để dự báo giá trị vòng đời khách hàng (LTV) trong 180 ngày chỉ với dữ liệu của 3 ngày đầu tiên, phát hiện cài đặt bất thường và mô phỏng phân bổ iOS thông qua Unattributed Conversion Analysis, giúp cung cấp thông tin chi tiết thông minh hơn, nhanh hơn và chính xác hơn để thúc đẩy hành trình tăng trưởng doanh số của bạn.
Cảm ơn bạn đã đồng hành cùng Airbridge trong hành trình này. Chúng tôi hi vọng sắp tới sẽ là một năm với nhiều thành công lớn hơn nữa. Cùng Airbridge hướng tới một năm 2025 ấn tượng và rực rỡ hơn nhé!
---
## 5 nhóm đối tượng cần tới DeepLink của Airbridge
URL: https://www.airbridge.io/vi/blog/5-team-who-need-airbridge-deeplink-plan
Language: Vietnamese
Published: 2024-11-26
Deep link là một công cụ mạnh mẽ cho mobile app marketing, có vai trò điều hướng người dùng đến những điểm đến cụ thể trong một ứng dụng. Deep link đóng vai trò quan trọng đối với CRM (Customer Relationship Management), quảng cáo, giới thiệu (referrals), và tiếp thị liên kết (affiliate marketing). Tuy nhiên, việc xây dựng và quản lý deep link có thể là một thách thức lớn.
Gần đây, Google đã thông báo [ngừng cung cấp Firebase Dynamic Links ](https://firebase.google.com/support/dynamic-links-faq) khiến nhiều ứng dụng gặp khó khăn. Các ứng dụng phụ thuộc vào Dynamic Link hiện nay phải lựa chọn giữa việc tự xây dựng hệ thống deep link hoặc chuyển sang một nhà cung cấp dịch vụ mới.
DeepLink Plan của Airbridge sẽ là một giải pháp thay thế đơn giản cho Firebase Dynamic Links. Bài viết này cung cấp thông tin về 5 nhóm đối tượng có thể khai thác được nhiều lợi ích nhất từ DeepLink của Airbridge. Nếu ứng dụng của bạn thuộc một trong những nhóm này, đã đến lúc bạn cần tới sự đồng hành của Airbridge.
💡 Bạn cần thêm thông tin chi tiết?
Khám phá [Airbridge Deeplink Plan](https://www.airbridge.io/deeplink-plan) tại đây.
## 1. 🚚 Migration | Đội nhóm đang sử dụng Firebase Dynamic Links
Vào ngày 25/08/2025, Firebase Dynamic Links sẽ chính thức ngừng hoạt động. Tất cả các liên kết được tạo bởi Firebase Dynamic Links (bao gồm liên kết được lưu trữ trên tên miền tùy chỉnh và subdomain page.link) sẽ ngừng hoạt động, và bạn sẽ không thể tạo liên kết mới hoặc di chuyển các liên kết cũ được nữa.
Nếu đội nhóm của bạn đã từng phụ thuộc vào Dynamic Links, bạn cần chuyển sang một dịch vụ deep linking mới trước thời hạn trên. Để quá trình chuyển đổi diễn ra dễ dàng, bạn sẽ cần tìm kiếm một giải pháp đơn giản hóa việc chuyển đổi deep link.
Airbridge DeepLink Plan cung cấp khả năng chuyển đổi không cần mã hóa (no-code migration) chỉ trong 3 bước. Plan này giữ nguyên tất cả các tính năng của Dynamic Links, bao gồm deferred deep linking và các link tuỳ chỉnh theo từng brand, đồng thời bổ sung các lợi ích như:
- Tối ưu hóa trải nghiệm người dùng trên cả iOS và Android
- Tạo mã QR tự động
- Thêm các tính năng tùy chỉnh cho brand
Airbridge DeepLink có khả năng cung cấp mọi tính năng của Dynamic Links – và hơn thế nữa.
## 2. 💻 Development | Đội nhóm gặp khó khăn khi tự xây dựng deep link
Việc xây dựng và duy trì hệ thống deep link nội bộ là một quá trình phức tạp và tốn kém tài nguyên. Việc này đòi hỏi đội ngũ Dev phải xây dựng được hệ thống link cơ bản, có tích hợp các tính năng nâng cao như dynamic và deferred deep link, liên tục cập nhật để đảm bảo các liên kết hoạt động được trên nhiều hệ điều hành, trình duyệt và ứng dụng khác nhau.
Airbridge DeepLink sẽ đơn giản hóa quy trình này với chuyên môn và tính năng hoàn thiện, dễ dàng sử dụng ngay:
- Đảm bảo deep link hoạt động liền mạch trên mọi nền tảng
- Hỗ trợ dynamic và deferred deep linking mượt mà
- Giải phóng đội nhóm Dev để họ tập trung vào các nhiệm vụ cốt lõi
👉 Tìm hiểu thêm qua [Deep Link Handbook](https://www.airbridge.io/library/airbridge-handbook-deeplinks) của Airbridge.
## 3. 🔗 User Experience | Đội nhóm đang tìm kiếm hiệu suất deep link tốt hơn
Trải nghiệm người dùng mượt mà là yếu tố then chốt cho sự thành công của ứng dụng. Khi người dùng nhấn vào một link mà không được dẫn đến đúng nơi, hoặc tệ hơn, ứng dụng không mở được, điều này dẫn đến gián đoạn hành trình người dùng và thậm chí khiến họ từ bỏ ứng dụng.
Hiệu suất deep link hoàn hảo là yếu tố quan trọng đối với các ứng dụng phụ thuộc nhiều vào quảng cáo hoặc liên kết được chia sẻ. DeepLink của Airbridge hoạt động mượt mà trên mọi kênh, mọi nền tảng, mang lại trải nghiệm liền mạch và đồng nhất cho người dùng.
## 4. 📆 App Launch | Đội nhóm phát hành ứng dụng mới
Phát hành ứng dụng mới là một sự kiện thú vị nhưng cũng đi kèm những quyết định quan trọng, bao gồm cả cách xử lý deep link. Việc Google ngừng cung cấp Firebase Dynamic Links khiến cho các ứng dụng chuẩn bị phát hành không thể sử dụng công cụ này được nữa.
Bắt đầu với một dịch vụ deep link chuyên dụng ngay từ đầu có thể giúp bạn tiết kiệm thời gian và tránh những rắc rối sau này. **Airbridge DeepLink Plan** cung cấp mức giá linh hoạt dựa trên MAU (Monthly Active Users), một lựa chọn tuyệt vời cho các ứng dụng mới với nhiều mức ngân sách khác nhau.
## 5. 💵 Pricing | Đội nhóm tìm kiếm giải pháp deep link tiết kiệm
Khi lựa chọn một dịch vụ deep link mới, vấn đề chi phí quan trọng không kém các vấn đề về tính năng. Airbridge DeepLink Plan cung cấp mức giá cạnh tranh, linh hoạt dựa trên MAU, phù hợp với sự tăng trưởng của ứng dụng. Bạn chỉ cần chi trả cho những gì bạn thấy cần thiết!
👉 Truy cập [Pricing Page](https://www.airbridge.io/deeplink-plan) của Airbridge để lựa chọn gói phù hợp nhất với ngân sách và yêu cầu của bạn.
—
Dù doanh nghiệp của bạn thuộc một trong các đối tượng trên hay có nhu cầu riêng biệt về deep link, Airbridge đều có thể hỗ trợ bạn. Liên hệ ngay với chúng tôi để khám phá cách Airbridge DeepLink có thể giúp ứng dụng của bạn vươn xa hơn!
---
## Tăng trưởng Game với tích hợp Airbridge + Hive
URL: https://www.airbridge.io/vi/blog/hive-platform-integration
Language: Vietnamese
Published: 2024-11-26
Chúng tôi trân trọng giới thiệu sự tích hợp mới giữa Airbridge và Hive Platform - một giải pháp toàn diện giúp các nhà phát triển game dễ dàng đo lường và tối ưu hóa chiến dịch marketing của mình. Giờ đây, bạn có thể nhanh chóng tích hợp SDK của Airbridge và bắt đầu đo lường, phân tích, cải thiện hiệu suất của các chiến dịch đa kênh, đa quảng cáo, và tại mọi khu vực mà không cần phải lo lắng về việc cài đặt phức tạp.
## Sự tích hợp này mang đến lợi ích nào?
1. Tích hợp với Airbridge với thao tác đơn giản:
- SDK Hive phiên bản v4.24.4.0+ (dành cho di động) hoặc v24.4.0+ (dành cho PC) đã được tích hợp sẵn với Airbridge
- Bạn chỉ cần cập nhật SDK lên phiên bản mới nhất để kích hoạt Airbridge trong vài giây mà không cần cài đặt gì thêm.
2. Chia sẻ dữ liệu hai chiều – Tính năng sắp ra mắt
- Dữ liệu phân bổ từ Airbridge sẽ sớm được đồng bộ với Hive, giúp bạn thu thập thông tin chi tiết theo thời gian thực, sử dụng cho việc tối ưu hóa các chiến lược thu hút người dùng, cải thiện tương tác, và tạo doanh thu một cách hiệu quả.
## Về Hive Platform
Hive Platform ([https://hiveplatform.ai/](https://hiveplatform.ai/)) của Com2uS là giải pháp phát triển Game toàn diện, được các hãng Game lớn như Gravity, StoneAge, Mistgames, và Blue Potion tin dùng. Nền tảng này cung cấp:
- Developer SDK: Đẩy nhanh quá trình phát triển Game trên Unity và Unreal Engine.
- LiveOps Portal: Quản lý Game theo thời gian thực với các tính năng như event promotion (khuyến mãi theo sự kiện) và user engagement (tương tác người dùng).
- Game Backend: Cung cấp các dịch vụ backend, từ xác thực người dùng, thanh toán, đến thông báo đẩy (push notification) và tích hợp blockchain.
- Tạo doanh thu từ quảng cáo: Tích hợp quảng cáo hiệu quả mà không làm gián đoạn trải nghiệm chơi game, tạo doanh thu qua các định dạng quảng cáo đa dạng.
- Data & Phân tích: Cung cấp dịch vụ phân tích big data theo từng game, mang đến những insights sâu sắc về hành vi người chơi và hiệu suất Game.
## Tóm lại
Sự kết hợp giữa các công cụ phát triển Game mạnh mẽ của Hive và công nghệ đo lường tiên tiến của Airbridge sẽ mang đến cho bạn một giải pháp toàn diện: đo lường chính xác, tối ưu hóa quy trình, thúc đẩy thu hút người chơi, và tăng cường hiệu quả doanh thu - tất cả gói gọn trong một nền tảng duy nhất.
## Về Airbridge
Airbridge là đối tác đo lường di động (MMP) toàn cầu, giúp thu thập insights người dùng theo thời gian thực trên nhiều nền tảng như di động, web, PC và console. Với hơn 300 chỉ số dùng cho thu hút người dùng (user acquisition), tương tác (engagement), và tạo doanh thu (monetization), Airbridge đã được các hãng game hàng đầu như Nexon, Webzen kiểm chứng và tin dùng. Các tính năng nổi bật của Airbridge bao gồm: phân tích theo cohort, chống gian lận, deep link, xuất dữ liệu thô không giới hạn, và dự báo LTV chuẩn xác - tất cả tính năng dễ dàng được tiếp cận thông qua một dashboard hợp nhất với sự hỗ trợ tận tình từ đội ngũ chuyên gia.
---
## Chuyển đổi Deeplink không cần mã lập trình từ Firebase sang Airbridge
URL: https://www.airbridge.io/vi/blog/how-to-migrate-deeplink-in-airbridge
Language: Vietnamese
Published: 2024-10-25
Google sẽ ngừng cung cấp Firebase Dynamic Links từ ngày 25/08/2025. Sau thời điểm này, các liên kết mà bạn đang sử dụng sẽ không hoạt động và việc chuyển đổi lúc đó là quá muộn. Bạn sẽ cần phải chuyển đổi sang một giải pháp deeplink mới nhanh chóng trước thời hạn trên.
Khi chuyển sang một giải pháp deeplink mới, việc di chuyển các liên kết đang dùng hiện tại là một phần quan trọng của quá trình này.
Với Airbridge, việc chuyển đổi Firebase Dynamic Links sẽ trở nên dễ dàng và hoàn toàn no-code. Bạn có thể chuyển tất cả link của mình nhanh chóng và hiệu quả thông qua công cụ **Tracking Link Migrator và Generator của Airbridge.**
Dưới đây là hướng dẫn chuyển đổi Dynamic Links sang Airbridge chỉ với 3 bước đơn giản!
## **Bước 1: Xuất Firebase Dynamic Links**
Sử dụng tính năng Firebase Takeout để xuất các Dynamic Links từ Firebase, sau đó tạo một tệp CSV chứa tất cả các link đang hoạt động.
## **Bước 2: Tiến hành mapping các tham số sang Airbridge Tracking Links**
Dán các link đã xuất vào tab **Dynamic Link Input** trên thanh công cụ **Tracking Link Migrator của Airbridge**. Sau đó, sao chép các giá trị tham số đã được chuyển đổi từ tab **Parameter Mapping Results.**
## **Bước 3: Tạo Airbridge DeepLinks**
Dán các link đã được chuyển đổi vào tab **Link Generation** trên thanh công cụ **Tracking Link Migrator** của Airbridge. Nhấp vào “Generate Tracking Link” để tạo Airbridge DeepLink. Liên kết mới sẽ xuất hiện trong danh sách và trên bảng điều khiển Airbridge của bạn.
## **Bắt đầu triển khai DeepLink của Airbridge!**
Thay thế các Dynamic Links cũ bằng các Airbridge DeepLink mới đã được nâng cấp. Việc chuyển đổi đến đây là hoàn tất!
Airbridge giúp việc chuyển đổi liên kết trở nên nhanh chóng và dễ dàng hơn bao giờ hết, bất kể bạn cần chuyển bao nhiêu. Bạn muốn tìm hiểu rõ chi tiết hơn về Airbridge? Hãy liên hệ với Airbridge ngay hôm nay và nhận một bản demo cá nhân hóa cho riêng mình. Đội ngũ chuyên gia của Airbridge cam kết hỗ trợ bạn để quá trình chuyển đổi diễn ra mượt mà nhất.
---
## Tăng hiệu quả chiến lược đa kênh với Conversion API
URL: https://www.airbridge.io/vi/blog/capi-for-omni
Language: Vietnamese
Published: 2024-10-24
Các quy định về quyền riêng tư đang định hình lại thế giới quảng cáo kỹ thuật số. [Privacy Sandbox (Hộp cát quyền riêng tư) của Google sẽ sớm cho phép người dùng Chrome chặn cookie của bên thứ ba](https://privacysandbox.com/news/privacy-sandbox-update/), tương tự như chính sách App Tracking Transparency (ATT) của Apple. Vào năm 2023, chỉ có 30% người dùng iOS chấp nhận chia sẻ dữ liệu, và không khó để dự đoán, dữ liệu cookie dùng trong việc cá nhân hoá sụt giảm đáng kể.
Khi các quy định về quyền riêng tư ngày càng nghiêm ngặt, Conversion API (API Chuyển đổi) càng trở nên quan trọng hơn bao giờ hết. Những API này cung cấp cho marketer một cái nhìn tổng thể về đối tượng của họ, giúp họ tạo ra các chiến dịch hiệu quả và nhắm mục tiêu đến khách hàng trên các nền tảng trực tuyến và ngoại tuyến.
## Vậy chính xác Conversion API là gì?
Conversion API, còn được gọi là CAPIs hoặc server-side APIs (API máy chủ), gửi dữ liệu event (sự kiện) của bên thứ nhất trực tiếp từ máy chủ của người dùng, tới các nền tảng quảng cáo bên thứ ba. Conversion API hoạt động trên phía máy chủ nên có thể thu thập toàn bộ dữ liệu từ các máy chủ nội bộ chứa các website hoặc ứng dụng mà bạn đang dùng. Điều này giúp bỏ qua các rắc rối từ phía khách hàng như pixel web, cài đặt trình duyệt và hạn chế cookie.
Conversion API có thể truyền nhiều loại dữ liệu từ bên thứ nhất, bao gồm hành vi người dùng (user behavior) và sự kiện chuyển đổi (conversion events) – chẳng hạn như mua hàng và các lượt đăng ký – từ nguồn web, ứng dụng, và cả nguồn ngoại tuyến. Điều này đặc biệt hữu ích trong bối cảnh các hạn chế về quyền riêng tư trong quảng cáo ngày càng tăng. Các API này cung cấp nhiều phương thức để duy trì độ chính xác của dữ liệu và hiệu quả của chiến dịch, khắc phục các thách thức từ chính sách quyền riêng tư mới.
Nhận thấy sự chuyển đổi này, các nền tảng quảng cáo lớn như Meta, Snapchat, TikTok, và X đã phát triển và ra mắt các giải pháp Conversion API của riêng mình.
[]
## Tại sao chiến lược đa kênh của bạn cần phải có Conversion API?
Conversion API hoạt động song song với các phương pháp theo dõi tiêu chuẩn để cung cấp cho marketer dữ liệu chuyển đổi và dữ liệu người dùng theo thời gian thực và đáng tin cậy. Từ đó, bạn có thể đưa ra quyết định dựa trên dữ liệu và tối ưu hóa các chiến dịch quảng cáo của mình.
Với Conversion API, bạn có thể tối đa hóa việc sử dụng:
- Dữ liệu từ bên thứ nhất (First-party data):
Nếu như pixel truyền dữ liệu qua trình duyệt người dùng có thể bị chặn, hay postback dựa vào ID quảng cáo (ADID) đang trở nên ít đáng tin cậy hơn do thay đổi trong chính sách quyền riêng tư, thì Conversion API thu thập thông tin trực tiếp từ máy chủ của bạn. Chúng thu thập các điểm dữ liệu quan trọng như ID của các cú nhập (clicks), email và số điện thoại đã mã hóa, từ đó cung cấp các tham số để kết nối dữ liệu và đo lường chính xác ở phạm vi rộng hơn.
- Dữ liệu đa kênh:
Conversion API có thể gửi dữ liệu từ nhiều nguồn khác nhau, bao gồm ứng dụng, trang web, PC, console, TV, và cả các tương tác ngoại tuyến như mua hàng tại cửa hàng. Điều này cho phép bạn:
Tối ưu hóa các chiến dịch bằng dữ liệu từ tất cả các môi trường trực tuyến và ngoại tuyến, tại mọi nơi mà bạn tiến hành dịch vụ của mình.
- Đo lường hiệu quả chiến dịch chính xác hơn qua nhiều kênh.
- Lập kế hoạch tương lai dựa trên bức tranh tổng thể về sự tương tác của người dùng.
Ví dụ, Nexon đã đạt thành công trong lĩnh vực game PC và console với “The First Descendant” bằng cách sử dụng Conversion API thông qua Airbridge cho chiến dịch thu hút người dùng (user acquisition - UA). Điều này giúp Đội ngũ UA của Nexon gửi trực tiếp các sự kiện chuyển đổi (conversion events) trong game đến các nền tảng quảng cáo như Meta và X. Bằng cách tích hợp Conversion API với pixel web trên trang đích, Nexon đã cải thiện đáng kể chiến dịch đa nền tảng của mình, dẫn đến gia tăng Lợi tức chi tiêu quảng cáo (ROAS).
> 📚 [Xem cách Airbridge hỗ trợ Nexon thúc đẩy hiệu suất marketing cho game trên PC và console tại đây ](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta1-nexon-the-first-descendant&og_tag_id=102968490&routing_short_id=kthfj3&sub_id=abr-en&tracking_template_id=d0c8d0332ea8449d6e592a2c22e1a591&ad_type=click)
## Kết nối Conversion API thông qua Airbridge
Hãy cập nhật những Conversion API mới nhất từ các nền tảng quảng cáo. Airbridge có tích hợp với các đối tác lớn như Meta, Snapchat, TikTok, X, và sắp tới là Google. Với Airbridge, bạn có thể quản lý nhiều API và theo dõi các chỉ số hiệu suất từ một bảng điều khiển duy nhất, giúp đơn giản hóa việc phân tích dữ liệu và tối ưu hóa chiến dịch.
Liên hệ với [Đội ngũ Airbridge](https://abr.ge/@ab180themodel/blog?ad_group=insight&campaign=capi-for-omni&content=cta2-book-a-demo&og_tag_id=102968689&routing_short_id=mhmecs&sub_id=abr-en&tracking_template_id=9b28522cf38b99dabb7bcf9baf66fdfd&ad_type=click) ngay hôm nay để khám phá cách sử dụng Conversion API và tăng cường chiến lược marketing đa kênh của bạn.
---
## Cùng Airbridge x Kwai đo lường quảng cáo video ngắn
URL: https://www.airbridge.io/vi/blog/kwai-integration
Language: Vietnamese
Published: 2024-10-02
Sự tích hợp giữa Airbridge và Kwai for Business giờ đây mang đến cho bạn khả năng đo lường chính xác và tối ưu hóa hiệu suất chiến dịch trên nền tảng Kwai. Nhờ quan hệ đối tác chiến lược này, bạn có thể tận dụng sức mạnh của hơn 1 tỷ người dùng hoạt động hàng tháng (MAU) của Kwai tại hơn 30 quốc gia, đồng thời khai thác các tính năng phân tích tiên tiến của Airbridge để tối ưu hóa chiến dịch quảng cáo một cách hiệu quả. Điều này không chỉ giúp bạn tiếp cận một lượng lớn người dùng mà còn đảm bảo rằng các chiến dịch của bạn đạt được hiệu suất tối ưu nhất.
[]
## Đo lường hiệu quả các chiến dịch trên Kwai
Với sự tích hợp mạnh mẽ giữa Airbridge và Kwai for Business, bạn có thể dễ dàng theo dõi và tối ưu hóa hiệu suất các chiến dịch quảng cáo trên Kwai. Các tính năng nổi bật bao gồm:
- Tạo và quản lý các tracking link được thiết kế riêng cho chiến dịch Kwai
- Thiết lập các quy tắc đo lường phân bổ (attribution) phù hợp với nền tảng Kwai
- Cấu hình postback để gửi dữ liệu real time về các sự kiện in-app và các chuyển đổi
Nhờ những tính năng này, bạn có thể nắm bắt chính xác hơn hiệu suất của các định dạng quảng cáo khác nhau trên Kwai, bao gồm:
- Quảng cáo in-feed
- Quảng cáo video ngắn
- Hợp tác với các influencer
## Phân tích nâng cao cho cộng đồng Kwai
Với nền tảng Airbridge, bạn có thể hiểu rõ hơn và tối ưu hóa việc tương tác với cộng đồng người dùng Kwai, đặc biệt là nhóm Gen Z và thế hệ Millennial, những đối tượng chủ yếu tạo nên sự sôi động và tiềm năng của nền tảng. Nhờ các công cụ phân tích nâng cao:
- Theo dõi mức độ tương tác của người dùng (user engagement)
- Đo lường hiệu quả của các chiến dịch theo xu hướng (trend-based)
Bằng cách tích hợp chiến dịch Kwai vào Airbridge, bạn có thể truy cập vào một lượng dữ liệu toàn diện, từ đó tối ưu hóa việc nhắm mục tiêu, sáng tạo nội dung, và phân bổ ngân sách cho cộng đồng có mức độ tương tác cao trên Kwai.
Để biết thêm chi tiết từng bước thực hiện, hãy tham khảo [Airbridge User Guide](https://help.airbridge.io/en/guides/kwai-for-business). Nếu bạn muốn tìm hiểu cách Airbridge và Kwai for Business có thể giúp thúc đẩy tăng trưởng ứng dụng và tối ưu hóa các chiến dịch quảng cáo của bạn, hãy liên hệ với [chúng tôi](https://www.airbridge.io/vi/book-a-demo) ngay hôm nay.
## Về Kwai for Business
Kwai for Business, ra mắt vào năm 2021, là một nền tảng toàn cầu có trụ sở tại Bắc Kinh, Trung Quốc, cung cấp các giải pháp marketing toàn diện cho các marketer nhằm sáng tạo và tương tác sâu với cộng đồng người dùng Kwai. Với hơn 1 tỷ người dùng hoạt động hàng tháng (MAU) trên hơn 30 quốc gia, Kwai for Business mở ra cơ hội kết nối chân thực và hiệu quả với đối tượng mục tiêu của bạn.
Kwai mang đến nhiều lựa chọn quảng cáo, bao gồm: quảng cáo in-feed, chiến dịch tương tác (brand challenge), nội dung do người dùng tạo (UGC). Với Kwai for Business, bạn có thể nâng cao chiến lược sáng tạo của mình thông qua các video quảng cáo được địa phương hóa, đồng thời theo dõi hiệu suất của chiến dịch trong một nền tảng duy nhất. Bạn cũng có thể khám phá các xu hướng thịnh hành trong cộng đồng và hợp tác với các nhà sáng tạo hàng đầu để đưa chiến dịch của mình lên một tầm cao mới.
---
## Khởi động chiến dịch iOS với Apple Search Ads và SKAN
URL: https://www.airbridge.io/vi/blog/asa-and-skan-to-optimize-ios-campaign
Language: Vietnamese
Published: 2024-09-23
Việc tối ưu hóa chiến dịch marketing trên iOS đã trở thành một thách thức lớn đối với nhiều marketer kể từ khi chính sách App Tracking Transparency (ATT) của Apple được triển khai vào năm 2021. Chính sách này hạn chế quyền truy cập dữ liệu nhận diện thiết bị, chỉ cho phép thu thập dữ liệu từ người dùng đã đồng ý chia sẻ.
Nhưng thực tế, có bao nhiêu người dùng iOS sẵn sàng chia sẻ dữ liệu của họ? Theo phân tích của Airbridge vào năm 2023, chỉ 3/10 người dùng iOS đồng ý chia sẻ, khiến việc thu thập đủ dữ liệu để tối ưu hóa chiến dịch trở nên thách thức hơn khi nào hết. Điều này đã khiến nhiều marketer phải từ bỏ iOS marketing hoặc tập trung hơn vào Android, hệ điều hành có đến 69% người trưởng thành tại Hàn Quốc sử dụng thiết bị Android.
Tuy nhiên, khi xét theo độ tuổi, iOS marketing vẫn đặc biệt quan trọng. Trong thống kê, 64% người dùng smartphone trong độ tuổi 20 tại Hàn Quốc sử dụng iPhone, và tỷ lệ này lên đến 72% ở phái nữ. Nếu ứng dụng của bạn hướng đến đối tượng người dùng trẻ, đặc biệt là trong các lĩnh vực như thời trang và game, thì iOS marketing vẫn là một thị trường tiềm năng cần được chú trọng. Vậy làm thế nào để tối ưu hóa chiến dịch iOS?
## Bước đầu tiên: Tận dụng Apple Search Ads và SKAN
Để bắt đầu hành trình tối ưu hóa marketing trên iOS, bạn cần đến Apple Search Ads (ASA) và SKAN (SKAdNetwork). Apple Search Ads là kênh quảng cáo trực tiếp của Apple, mang lại hiệu quả cao nhờ hiển thị quảng cáo ngay trên App Store, thu hút trực tiếp người dùng iOS. Đặc biệt, Apple cung cấp dữ liệu nhận diện thiết bị cho marketer ngay cả khi người dùng không đồng ý cho theo dõi, mở ra nhiều cơ hội tối ưu hóa chiến dịch cho marketer.
Đối với những marketer đang quảng cáo trên nhiều kênh, SKAN là công cụ không thể thiếu. Đây là một framework bảo mật từ Apple, cho phép thu thập dữ liệu quảng cáo ngay cả khi người dùng từ chối chia sẻ, giúp bạn tối ưu hóa chiến dịch một cách an toàn và hiệu quả.
- [Tìm hiểu thêm về SKAN 3](https://help.airbridge.io/en/guides/skadnetwork-3)
- [Tìm hiểu thêm về SKAN 4](https://help.airbridge.io/en/guides/skadnetwork-4)
⚠️ Lưu ý
Apple chỉ cung cấp dữ liệu về lượt tải ứng dụng thông qua SKAN.
## Tối ưu hóa iOS với MMP dễ sử dụng
Mặc dù Apple Search Ads và SKAN mang lại nhiều cơ hội, nhưng không ít marketer vẫn gặp khó khăn trong việc phân tích và tích hợp dữ liệu. Đặc biệt, nếu bạn cũng đang chạy các chiến dịch trên Android, việc kết hợp dữ liệu từ Apple Search Ads với dữ liệu Android để có cái nhìn toàn diện về hiệu suất marketing là vô cùng cần thiết. Ngoài ra, việc diễn giải dữ liệu từ SKAN cũng không đơn giản. Đây là lúc cần đến sự hỗ trợ từ một Mobile Measurement Partner (MMP) dễ sử dụng, giúp bạn tối ưu hóa chiến dịch một cách trơn tru và chính xác.
### 1. Phân tích tổng hợp dữ liệu từ Apple Search Ads
Dữ liệu nhận diện thiết bị từ Apple Search Ads chỉ cung cấp một phần nhỏ trong toàn bộ hành trình người dùng. Để hiểu rõ hơn về hành vi sau khi người dùng tương tác với quảng cáo, bạn cần tích hợp dữ liệu này với thông tin trong ứng dụng. Nếu không có sự tích hợp, bạn sẽ khó xác định liệu người dùng có thực hiện các hành động quan trọng, như mua hàng, sau khi tiếp cận quảng cáo hay không.
Giải pháp là sử dụng Mobile Measurement Partner (MMP) để phân tích toàn bộ hành trình từ quảng cáo đến tương tác trong ứng dụng. MMP không chỉ cung cấp dữ liệu về tỷ lệ chuyển đổi mà còn đo lường các chỉ số quan trọng khác như retention (giữ chân người dùng) và giá trị vòng đời dự đoán (pLTV).
### 2. Tối ưu hóa mobile marketing
Việc tối ưu hóa marketing iOS chỉ là một phần của chiến lược tổng thể. Để có cái nhìn đầy đủ về hiệu suất marketing trên cả hai nền tảng iOS và Android, dữ liệu cần được phân tích một cách đồng nhất. Tuy nhiên, bảng điều khiển của Apple Search Ads chỉ cung cấp những con số cơ bản và thiếu đi khả năng đánh giá hiệu quả của các kênh quảng cáo khác.
Khi bạn chạy quảng cáo trên nhiều kênh, MMP là công cụ không thể thiếu. MMP không chỉ cung cấp quy trình phân bổ (attribution) chính xác mà còn giúp xác định kênh nào mang lại hiệu quả cao nhất và có khả năng giữ chân người dùng lâu dài nhất.
- [Tìm hiểu thêm về Phân bổ (Attribution)](https://help.airbridge.io/en/guides/attribution-overview)
### 3. Cài đặt SKAN dễ dàng
Apple cung cấp dữ liệu qua SKAN (SKAdNetwork), nhưng để truy cập và phân tích dữ liệu này, bạn cần một MMP phù hợp.
Việc thiết lập SKAN có thể phức tạp, nhưng với một MMP hiệu quả, quá trình này sẽ trở nên đơn giản hơn. Sau khi thiết lập, MMP sẽ tự động phân tích và trình bày dữ liệu theo cách dễ hiểu, giúp bạn đưa ra các quyết định tối ưu hóa nhanh chóng.
## Bắt đầu tối ưu hóa iOS với Airbridge – MMP dễ sử dụng nhất
Airbridge cung cấp một loạt tính năng giúp việc tối ưu hóa iOS trở nên dễ dàng hơn bao giờ hết. Chỉ với một vài thao tác, bạn có thể tích hợp Airbridge với Apple Search Ads để thu thập, sắp xếp và phân tích dữ liệu một cách chi tiết. Airbridge mang đến hơn 200 chỉ số quan trọng, từ lượt cài đặt ứng dụng cho đến retention và pLTV, giúp bạn nắm bắt đầy đủ thông tin cần thiết cho việc tối ưu hóa chiến dịch.
Nếu bạn sử dụng giải pháp Apple Search Ads Advanced, Airbridge sẽ liên kết chi phí quảng cáo với các chi tiết chiến dịch như tên chiến dịch, nhóm quảng cáo, từ khóa và nội dung quảng cáo, cung cấp cái nhìn toàn diện hơn về hiệu suất. Dữ liệu cũng được cập nhật mỗi 4 giờ, giúp bạn luôn theo dõi được kết quả theo thời gian thực mà không cần phải kiểm tra liên tục.
Việc thiết lập SKAN trên Airbridge rất trực quan. Trước khi chạy chiến dịch iOS, bạn có thể cấu hình SKAN và Airbridge sẽ phân tích dữ liệu hàng ngày, kết hợp với các chỉ số khác để cung cấp bức tranh toàn diện về hiệu suất chiến dịch:
- [Tích hợp Apple Search Ads](https://help.airbridge.io/en/guides/apple-search-ads)[](https://help.airbridge.io/en/guides/preparing-for-skan)
- [Chuẩn bị chiến dịch iOS với SKAN](https://help.airbridge.io/en/guides/preparing-for-skan)
Nếu bạn đang tìm kiếm một MMP để tối ưu hóa chiến lược marketing iOS hoặc gặp khó khăn trong quá trình này, Airbridge sẽ là lựa chọn lý tưởng hỗ trợ bạn tập trung vào những yếu tố quan trọng nhất.
---
## Firebase Dynamic Links ngừng hoạt động. Airbridge sẵn sàng hỗ trợ bạn.
URL: https://www.airbridge.io/vi/blog/introduce-deeplink-plan
Language: Vietnamese
Published: 2024-09-05
Chúng ta đều đang bước vào giai đoạn đếm ngược đến ngày Firebase Dynamic Links chính thức dừng hoạt động. Vào ngày 25 tháng 8 năm 2025, dịch vụ [Firebase Dynamic Links](https://firebase.google.com/docs/dynamic-links) của Google sẽ dừng lại. Tất cả các đường liên kết do Firebase Dynamic Links cung cấp sẽ ngừng hoạt động, và bạn sẽ không thể tạo đường liên kết mới. Điều này đồng nghĩa với việc bạn có chưa đến 12 tháng để tìm, đánh giá và hoàn tất [việc chuyển đổi sang một nhà cung cấp dịch vụ deep link mới](https://www.airbridge.io/en/blog/firebase-dynamic-links-shutdown).
Vậy, câu hỏi đặt ra là: Đâu là lựa chọn thay thế tốt nhất? Xin giới thiệu **Airbridge DeepLink Plan**.
# **Airbridge DeepLink Plan: Giải pháp thay thế đáng tin cậy cho Firebase Dynamic Links**
Airbridge tự hào mang đến **DeepLink Plan – giải pháp thay thế đơn giản và đáng tin cậy cho Firebase Dynamic Links sắp ngừng hoạt động**. Là một công cụ hàng đầu trong lĩnh vực đo lường và cung cấp deep link, Airbridge tin rằng DeepLink Plan sẽ là lựa chọn tốt nhất cho bạn. Bài viết này sẽ chia sẻ 3 lý do vì sao DeepLink Plan của Airbridge xứng đáng là giải pháp bạn đang tìm kiếm.
## **1\. Airbridge DeepLinks vượt xa tiêu chuẩn thông thường.**
Chuyển sang DeepLink Plan của chúng tôi không chỉ giúp bạn duy trì các liên kết deep link mà còn mang đến những tính năng ưu việt hơn hẳn Firebase Dynamic Links, giúp bạn cung cấp trải nghiệm người dùng tuyệt vời để tiếp tục thúc đẩy tăng trưởng, tỷ lệ giữ chân và doanh thu. Hãy cùng đi sâu cụ thể hơn:
### **Trải nghiệm người dùng mượt mà**
Tại Airbridge, trải nghiệm người dùng luôn là ưu tiên hàng đầu. Người dùng không nên bị ngắt quãng ở các trang trung gian khi mở các liên kết deep link. Vì vậy, DeepLink Plan của chúng tôi cung cấp khả năng tương thích trên **phạm vi toàn cầu**, cho phép mở ứng dụng trực tiếp trên hầu hết các nền tảng và hệ điều hành, bao gồm cả truy cập online và offline, web và ứng dụng, thiết bị Android và iOS, và các nền tảng phổ biến như Instagram, X, YouTube, Facebook Messenger, Slack và nhiều hơn nữa. Bằng cách giảm thiểu các trang trung gian, Airbridge DeepLinks đảm bảo trải nghiệm mượt mà và hiệu quả cho người dùng.
### **Hỗ trợ toàn diện iOS và Android**
Trong khi Firebase Dynamic Links có một số hạn chế về hỗ trợ iOS, Airbridge DeepLinks cung cấp **khả năng đo lường hiệu suất hoàn chỉnh cho cả Android và iOS** mà không cần các bước bổ sung để thu thập dữ liệu hiệu suất trên iOS.
### **Liên kết sâu chờ tải (deferred deep links) trơn tru**
Tận hưởng các [tính năng liên kết sâu chờ tải](https://help.airbridge.io/en/guides/retargeting-with-deep-links) mạnh mẽ của chúng tôi. Chỉ với một lần nhấp, Airbridge DeepLinks có thể đưa người dùng trực tiếp đến đích trong ứng dụng mà họ mong muốn, ngay cả khi họ chưa cài đặt ứng dụng của bạn.
### **Trình tạo các liên kết theo dõi một cách hàng loạt**
Bạn cần nhiều liên kết theo dõi cho các mẫu quảng cáo và vị trí quảng cáo khác nhau? DeepLink Plan của chúng tôi bao gồm mẫu [Trình tạo Liên kết Theo dõi nhanh và API](https://help.airbridge.io/en/guides/creating-bulk-tracking-links), cho phép bạn tạo liên kết theo dõi hàng loạt trực tiếp trong Google Spreadsheets.
### **Tuỳ chỉnh Deep links có thương hiệu**
Làm nổi bật thương hiệu của bạn với các **liên kết ngắn tùy chỉnh**. Với DeepLink Plan của chúng tôi, bạn có thể sử dụng tên miền của riêng mình để tạo các liên kết cá nhân hóa và nâng cao hiệu quả của previews khi chia sẻ trên mạng xã hội bằng cách tùy chỉnh tiêu đề, mô tả và hình ảnh của URL.
## **2. Airbridge DeepLinks cung cấp nhiều hơn các lựa chọn khác.**
Airbridge cung cấp nhiều tính năng hơn so với các lựa chọn thay thế Firebase Dynamic Links khác trên thị trường. Với DeepLink Plan của chúng tôi, việc tối ưu hóa chiến lược deep link của bạn trở nên dễ dàng và hiệu quả hơn. Hãy xem điểm khác biệt của chúng tôi:
### **Tạo 10,000 liên kết theo dõi mỗi tháng**
Dễ dàng tạo ra số lượng lớn, **lên tới 10,000** [**liên kết theo dõi**](https://help.airbridge.io/en/guides/tracking-link-management) hàng loạt thông qua một bảng điều khiển cực kì dễ dùng và thân thiện với người dùng của chúng tôi—vượt xa giới hạn 3,000 liên kết của Hãng A. Thực sự là gần như không giới hạn!
\*_Bạn cần nhiều hơn 10,000 liên kết mỗi tháng?_ [_Liên hệ ngay với Airbridge_](https://www.airbridge.io/vi/book-a-demo)_._
### **Quản lý mã QR dễ dàng**
Mã QR trở nên vô cùng thiết yếu trong việc theo dõi các hoạt động marketing ngoại tuyến. Airbridge DeepLinks **tự động tạo mã QR cho mỗi liên kết mà bạn tạo**. Khác với các dịch vụ deep link khác như Hãng B (không cung cấp mã QR) hoặc Hãng C (chỉ cung cấp với một số tùy chọn hạn chế), chúng tôi làm cho việc quản lý mã QR trở nên đơn giản và tập trung.
### **Quản lý các tình huống phức tạp**
Trải nghiệm hệ thống quản lý tình huống đáng tin cậy và hỗ trợ kỹ thuật với DeepLink Plan của chúng tôi. Chúng tôi đảm bảo rằng các deep link của bạn hoạt động hoàn hảo trên mọi nền tảng—ứng dụng di động, trình duyệt trong ứng dụng và thậm chí khi được dán trực tiếp vào thanh địa chỉ.
## **3\. Chuyển đổi sang Airbridge đơn giản và liền mạch.**
Chuyển đổi không chỉ đơn giản là thay đổi nhà cung cấp mà còn là đảm bảo rằng các liên kết và dữ liệu hiện có của bạn vẫn hoạt động bình thường. Nếu xử lý không cẩn thận, các liên kết có thể bị đứt gãy trong chiến dịch marketing—và Airbridge sẽ giúp bạn tránh điều này. Với Airbridge, quy trình chuyển đổi từ Firebase Dynamic Links của bạn trở nên dễ dàng. Tải dữ liệu Firebase của bạn lên bảng chuyển đổi của chúng tôi, và hệ thống sẽ tự động tạo ra các liên kết deep link mới để thay thế các liên kết cũ. Chỉ cần sao chép và dán để hoàn tất chuyển đổi.
# **Quan tâm đến DeepLink Plan của chúng tôi?**
Bạn đang tìm kiếm giải pháp thay thế cho Firebase Dynamic Links? Hãy nhấp vào bên dưới, chia sẻ một vài thông tin, và chúng tôi sẽ sớm liên hệ với bạn. Cùng khám phá cách Airbridge có thể đáp ứng các yêu cầu về deep link của bạn!
---
## Chinh phục 10.000 lượt cài đầu tiên nhờ TikTok
URL: https://www.airbridge.io/vi/blog/cracking-the-first-10-000-app-installs-using-tiktok
Language: Vietnamese
Published: 2024-04-24
Các marketer app đều thấu hiểu nỗi khổ này: đạt được 10.000 lượt cài đầu tiên giống như một nghệ thuật phức tạp. Hàng loạt thử nghiệm tiêu tốn ngân sách mà vẫn không rõ đâu mới là yếu tố quyết định thành công. Nhưng nếu bạn có thể khám phá ra một công thức tăng trưởng có thể lặp lại thì sao?
Xin giới thiệu Hamza Alsamraee, nhà sáng lập của NewForm.ai, cùng “bí kíp” TikTok của anh.
Hamza đã giải mã thành công TikTok và có đầy đủ số liệu để chứng minh: hơn 1 tỷ lượt xem, giảm trung bình 48% chi phí chuyển đổi (CAC), ROAS đạt 5.6 lần, và hơn 100 triệu USD ngân sách quảng cáo được quản lý.
Trong bài viết này, bạn sẽ được nghe chia sẻ chuyên sâu từ Hamza về chiến lược chiến dịch và từ Airbridge – đối tác tiếp thị của TikTok – về đo lường và tối ưu hóa chiến dịch.
## Đa số marketer app hiểu sai điều gì về TikTok?
Thành thật mà nói – không có “golden tip” duy nhất, nhưng có một quy trình mà chúng tôi luôn áp dụng – và nó hiệu quả. Tất cả xoay quanh một yếu tố: nội dung tốt. Mọi thứ khác đều phụ thuộc vào đó.
Không có một công thức nội dung cố định nào (chúng ta sẽ nói về các tips sau), nên chìa khóa là lặp đi lặp lại thật nhanh và hiệu quả. Và đa số các đội ngũ đều thất bại ở đây.
Hầu hết các team (agency và in-house) đều tìm content sai cách. Họ liên hệ hàng loạt influencer hoặc UGC creator, hoặc thuê một agency làm việc đó thay. Cách làm này vừa đắt đỏ vừa khó kiểm soát chất lượng và deadline.
Tôi hiểu điều đó vì tôi từng là một creator. Và creator thường khá “chậm deadline” và không dễ để scale. Một video ngắn có thể khiến thương hiệu tiêu tốn hàng ngàn đô. Trong khi đó, các UGC creator giá rẻ thì rất khó chọn được người có chất lượng. Thương hiệu thường phải thử hàng chục người mới chọn được một người ổn.
Tôi từng đứng ở phía bên kia – điều hành growth cho Faves, một app subscription đã vượt mốc 1 triệu lượt tải. Vì quá khó tìm creator nên tôi đã tự quay hơn 1000 video cho Faves.
Tất nhiên, không phải ai cũng có thời gian/quyết tâm làm vậy. Giải pháp là tìm creator nội bộ đáng tin cậy, có tư duy phân tích và có thể sản xuất nội dung theo chu trình. Đó chính là điểm khác biệt của NewForm: tất cả creator đều là người nội bộ và có background kỹ thuật/phân tích.
Và “growth hack” lớn nhất chính là: sở hữu creator nội bộ có thể sản xuất hàng chục video mỗi tháng một cách có hệ thống (dù là làm việc với NewForm hay thuê riêng). Thay vì brief ngẫu nhiên, bạn có thể kiểm soát chất lượng nội dung và tập trung vào định vị sản phẩm.
Chính việc thử nghiệm và tối ưu hóa định vị nội dung mới là chìa khóa giúp app bạn tăng trưởng bền vững.
## Sau hàng nghìn video TikTok, đâu là insight bất ngờ giúp tối ưu ROI và giảm CPI?
Từ “ghét” (“hate”). Không đùa đâu – gần một nửa video hiệu quả nhất của chúng tôi bắt đầu bằng cụm “Tôi ghét ___”.
“Thuốc giảm đau” luôn bán chạy hơn “vitamin” – không chỉ trong hiệu thuốc mà còn trong thế giới app. Hãy định vị app như một giải pháp giảm đau, và bạn sẽ thấy kết quả cải thiện rõ rệt.
Ngoài ra, nói ở ngôi thứ nhất (“Tôi gặp vấn đề này”) hoạt động tốt hơn hẳn. Ngôi thứ hai (“Bạn cần…”) nghe không thật. Dùng “tôi” giúp nội dung chân thực và dễ đồng cảm hơn.
## Hiệu quả Spark Ads so với các loại quảng cáo khác trên TikTok ra sao?
Spark Ads chỉ hiệu quả nếu nội dung đã có tương tác organic tốt từ trước – vì nó giúp tăng thời gian xem và giảm CPM (người dùng TikTok rất thích đọc comment). Các nhà quảng cáo lớn gần như không dùng Spark Ads cho các video mới chưa có tương tác.
Tuy vậy, đừng bỏ qua Spark Ads hoàn toàn. Rất nhiều app đã “gây bão” TikTok và dùng Spark Ads để khuếch đại hiệu ứng. Hãy tìm các video về app của bạn trên TikTok, rồi liên hệ creator để xin quyền sử dụng làm Spark Ads.
## Tăng trưởng là một chuyện, giữ chân và chuyển đổi người dùng mới là mục tiêu dài hạn. Làm sao để biến người dùng thoáng qua thành fan trung thành?
Đừng chạy theo trend (nhạc, thử thách, etc) – chỉ hút người dùng thoáng qua. Hãy giới thiệu app đúng với giá trị thực, để thu hút đúng người dùng trung thành và có khả năng trả tiền.
Chúng tôi thấy nội dung “cầm điện thoại trên tay” hiệu quả hơn nhiều so với chỉ quay màn hình. Thay vì screen record, hãy làm demo trực tiếp với điện thoại thật. Ví dụ như [thế này.](https://drive.google.com/drive/u/1/folders/1fXkJ7LnslREtoTmrJmZ2744CxgdZwHo1)
[]
Cuối cùng, vấn đề retention và monetization thường nằm ở onboarding – mà onboarding lại bắt đầu từ nội dung. Nếu nội dung ban đầu khiến người dùng hiểu sai về sản phẩm, họ sẽ rời đi ngay.
## Marketer thường sai lầm gì khi chạy TikTok để tăng trưởng?
Một vài lỗi điển hình:
Nhắm sai đối tượng khi chạy install campaign. Nếu app bạn có mục tiêu doanh thu và là app subscription, hãy tránh nhóm dưới 25 tuổi. Họ thường có chỉ số đầu phễu đẹp (CPI thấp) nhưng lại không chuyển đổi.
Nhắm mục tiêu quá hẹp: thuật toán TikTok hiểu người dùng tốt hơn 99% media buyer. Trong hầu hết các trường hợp, bạn nên để targeting rộng.
> *Gợi ý từ Airbridge*: Tối ưu hóa đăng ký app bằng phân tích funnel
Tái phân bổ ngân sách quảng cáo hoặc tối ưu phễu đăng ký bằng cách phân tích chi tiết tỷ lệ rớt người dùng theo từng bước: kênh, chiến dịch, nội dung quảng cáo, quốc gia, thiết bị – từ Cài đặt → Đăng ký → Màn hình thanh toán → Dùng thử miễn phí → Đăng ký → Thanh toán lần 2/3, v.v.
[]
Chi quá nhiều cho Android hoặc các quốc gia T2. Người dùng Android có chi phí cài đặt tương tự iOS nhưng lại khó chuyển đổi hơn. Các quốc gia T2 thì CPI thấp nhưng doanh thu tiềm năng lại hạn chế. Chúng tôi chi 90% ngân sách vào người dùng iOS, đặc biệt là ở Mỹ. Các quốc gia khác nên thử là: UK, Ireland, Canada, Úc, New Zealand, Đức, Pháp, Ý, Tây Ban Nha, Bắc Âu, Singapore, Hàn Quốc, Nhật Bản, v.v.
> *Gợi ý từ Airbridge*: Khai phá insight về giữ chân và doanh thu sau khi cài đặt
Dù bạn đang phát triển app game, subscription hay mạng xã hội, hãy đào sâu vào các chỉ số in-app chi tiết như ROAS/retention ngày 1, 3, 7 (d1, d3, d7), ARPU, CPI, thời gian đến lần mua đầu tiên, vòng đời dự đoán và giá trị vòng đời dự đoán – tất cả đều được hiển thị chi tiết theo từng kênh, chiến dịch, thiết bị và quốc gia trên bảng điều khiển của Airbridge.
[]
### Về NewForm
NewForm là đối tác tăng trưởng toàn diện cho ứng dụng của bạn. Khác với các agency tập trung vào ecommerce, NewForm xây dựng nội dung ngắn dạng lập trình sẵn để tăng trưởng app có lãi. Họ đã giúp hàng chục app tăng trưởng vượt bậc và tạo ra doanh thu 8 chữ số.
[Hamza](http://linkedin.com/in/hamza-alsamraee/) (founder và CEO) sẵn sàng trò chuyện với bạn.
### Về Airbridge
Các marketer app hàng đầu toàn cầu đang dần từ bỏ các công cụ attribution đắt đỏ để chuyển sang Airbridge – nền tảng đo lường thống nhất. Bảng điều khiển tùy chỉnh siêu chi tiết của chúng tôi cung cấp cái nhìn real-time về hành trình người dùng đa nền tảng: mobile, web, PC, console – từ lần click đầu đến lúc trở thành user trung thành.
200+ chỉ số, đo lường iOS, deep link, chống gian lận, chia sẻ dữ liệu – tất cả trong một mức giá tối ưu giúp bạn tiết kiệm ngân sách thay vì “đốt tiền.”
Khám phá bản demo miễn phí tại [airbridge.io](https://www.airbridge.io/book-a-demo)
---
## [Deep Link 101] Sử dụng Deferred Deep Linking để điều hướng người dùng chưa cài ứng dụng đến trang cụ thể trong app
URL: https://www.airbridge.io/vi/blog/deeplink-101-deferred-deep-linking
Language: Vietnamese
Published: 2022-04-07
[]
Bạn muốn điều hướng người dùng đến trang liên quan trong ứng dụng ngay cả sau khi họ nhấp vào liên kết và cài đặt app? Nếu vậy, bạn cần deferred deep linking.
> Để tìm hiểu thêm chi tiết về deep link, hãy xem hướng dẫn <[Airbridge Handbook - Deep Links](https://www.airbridge.io/en/library/airbridge-handbook-deeplinks)>.
## Deferred Deep Linking là gì?
Deferred deep linking là công nghệ mở rộng khả năng của deep link tiêu chuẩn. Đúng như tên gọi, nó có nghĩa là "deep link bị trì hoãn." Vì deep linking không thể hoạt động với người dùng chưa cài đặt ứng dụng, công nghệ này sẽ đợi cho đến khi người dùng nhấp vào liên kết, cài đặt ứng dụng và mở nó — sau đó mới thực thi deep link. Việc tự triển khai không hề dễ dàng vì cần có máy chủ có khả năng lưu trữ và phân phối deep link, cũng như khớp thông tin từ thời điểm người dùng nhấp vào liên kết với thông tin của người dùng đã cài đặt và mở ứng dụng.
Trong bài viết này, hãy tìm hiểu tại sao công nghệ deferred deep linking là cần thiết và cách triển khai nó.
## Tại sao cần Deferred Deep Linking?
Deep link là công nghệ giúp khi người dùng trên thiết bị di động nhấp vào liên kết, ứng dụng sẽ được khởi chạy và điều hướng đến một trang cụ thể. Nói cách khác, đó là một liên kết có đường dẫn được chỉ định dẫn vào ứng dụng. Tuy nhiên, nếu ứng dụng chưa được cài đặt khi nhấp vào deep link, người dùng sẽ bị chuyển hướng đến cửa hàng ứng dụng hoặc gặp lỗi, và liên kết/đường dẫn không hoạt động đúng. Điều này có nghĩa là ngay cả người dùng nhấp vào deep link cũng không được điều hướng đến trang cụ thể trong ứng dụng.
Cụ thể, khi ứng dụng chưa được cài đặt, các vấn đề sau có thể xảy ra:
- Khi sử dụng URI Scheme deep link, nếu ứng dụng chưa được cài đặt, không thể điều hướng đến nơi khác như cửa hàng ứng dụng, vì vậy không có gì xảy ra.
- Trong trường hợp Universal Links và App Links, không thể sử dụng URL dẫn trực tiếp đến App Store hoặc Google Play, khiến việc khuyến khích người dùng chưa có ứng dụng cài đặt trở nên khó khăn.
- Để giải quyết các tình huống trên, chỉ sử dụng deep link thông qua tracking link của MMP vẫn không mang lại trải nghiệm người dùng hoàn chỉnh.
Do những hạn chế về chức năng này, nếu không có gì xảy ra sau khi nhấp vào quảng cáo, hoặc sau quá trình cài đặt ứng dụng phức tạp mà chỉ hiển thị trang chính, thì ngay cả khách hàng tiềm năng bị thu hút bởi quảng cáo cũng có thể rời đi ngay lập tức.
**Công nghệ giải quyết vấn đề này chính là deferred deep linking.**
Ngay cả đối với người dùng chưa cài đặt ứng dụng, nếu họ nhìn thấy quảng cáo và cài đặt app, việc được điều hướng đến nội dung liên quan trong ứng dụng sau khi cài đặt sẽ cảm thấy tự nhiên. Khả năng chuyển đổi cũng tăng lên. Đây chính là lý do tại sao deferred deep linking là thiết yếu cho app marketing.
## Cách triển khai Deferred Deep Linking
Tuy nhiên, việc tự triển khai công nghệ deferred deep linking có thể gặp nhiều thách thức. Ở cấp hệ điều hành Android, có thể triển khai bằng Google Play Install Referrer API nhưng với nhiều hạn chế. Hơn nữa, trên iOS, không có phương pháp cấp hệ điều hành nào để triển khai deferred deep linking, vì vậy cần thiết lập phức tạp như xây dựng máy chủ riêng.
### Khó khăn khi tự triển khai Deferred Deep Linking
Đầu tiên, **trên Android**, bạn có thể tự triển khai deferred deep linking bằng Google Play Install Referrer API của Google Play Store.
Tuy nhiên, phương pháp này phức tạp trong triển khai, dễ gây ra vấn đề trong việc thu thập dữ liệu, và chỉ hoạt động với các lượt tải xuống từ Google Play.
**Trên iOS**, không thể sử dụng cách tiếp cận tương tự, vì vậy việc xây dựng máy chủ trở thành bước bổ sung cần thiết.
### Dễ dàng triển khai Deferred Deep Linking với giải pháp Mobile Attribution
Với việc deferred deep linking khó triển khai một mình, làm thế nào bạn có thể thực hiện hiệu quả và đầy đủ? Câu trả lời là sử dụng giải pháp do công cụ mobile attribution cung cấp.
Deferred deep linking là công nghệ trì hoãn deep link cho đến khi người dùng nhấp vào liên kết, cài đặt ứng dụng và mở nó. **Cần có máy chủ có thể lưu trữ thông tin nhấp chuột của người dùng và khi ứng dụng được khởi chạy, khớp dữ liệu nhấp chuột với người dùng đã mở ứng dụng và phân phối deep link.** Đây là những gì công cụ attribution cung cấp.
Quy trình triển khai deferred deep linking thông qua Airbridge:
- Người dùng nhấp vào tracking link của Airbridge -> Thông tin nhấp chuột được lưu trên máy chủ -> Người dùng cài đặt ứng dụng -> Khi khởi chạy ứng dụng lần đầu với SDK được tích hợp, máy chủ kiểm tra thông tin liên kết -> Máy chủ gửi địa chỉ deep link đến SDK -> Trang cụ thể trong ứng dụng được mở
Hướng dẫn SDK:
- Hướng dẫn Airbridge [Android SDK](https://developers.airbridge.io/docs/android-sdk)
- Hướng dẫn Airbridge [iOS SDK](https://developers.airbridge.io/docs/ios-sdk)
### Tracking Link của Airbridge giúp áp dụng Deferred Deep Linking dễ dàng
Với Airbridge, khi bạn thiết lập tracking link dưới dạng deep link, tính năng deferred deep linking sẽ được bật mặc định.
[]
## Để App Marketing diễn ra suôn sẻ, tốt nhất nên sử dụng Tracking Link của MMP
Tracking link của MMP cho phép bạn thiết kế hành trình người dùng mong muốn mà không cần triển khai phức tạp cho từng hệ điều hành.
Deferred deep linking là công nghệ thiết yếu cho app marketing. Hãy triển khai hiệu quả bằng tracking link của Airbridge.
> Xem các bài viết khác trong series Deep Link 101:
👉 [Deep Link 101: Bắt đầu với Deep Link cho Marketer và Developer](https://www.airbridge.io/vi/blog/deeplink-101-for-marketers-and-developers)
👉 [Deep Link 101: Tại sao xuất hiện cảnh báo khi Deep Linking trên iOS Safari?](https://www.airbridge.io/vi/blog/deeplink-101-ios-safari-alert)