.svg)
You've launched your app install campaigns across Meta, Google, and TikTok. The ads are running, the budget is flowing, and installs are coming in. But here's the problem: you have no idea which campaign actually drove that user who just subscribed for $99/month. Was it the Meta carousel ad? The Google App Campaign? That TikTok video that went semi-viral last week?
Without accurate conversion tracking, you're flying blind. And with iOS App Tracking Transparency limiting traditional tracking methods and Android privacy updates following suit, the old ways of measuring app installs simply don't work anymore. You need a system that captures every install, attributes it correctly, and feeds that data back to your ad platforms so their algorithms can optimize for the users who actually matter.
This guide walks you through building that system from the ground up. You'll learn how to configure a mobile measurement partner, implement tracking SDKs, connect your ad platforms, and handle the complexities of iOS privacy restrictions. By the end, you'll have accurate attribution data flowing into a central dashboard where you can finally see which campaigns drive not just installs, but revenue.
Let's build a tracking foundation that turns your app marketing from guesswork into a data-driven growth engine.
Your first decision is selecting a Mobile Measurement Partner, or MMP. Think of your MMP as the neutral referee that sits between your app and all your ad platforms. When someone installs your app, the MMP determines which ad campaign gets credit for that install and sends that attribution data back to the ad platform.
Why do you need an MMP instead of just using each ad platform's built-in tracking? Because ad platforms are biased toward their own data. Meta wants to claim credit for every install, Google wants to claim credit for every install, and when the same user saw ads on both platforms, they'll both try to take credit. An MMP provides a single source of truth that all platforms trust.
The major players are AppsFlyer, Adjust, Branch, and Kochava. AppsFlyer and Adjust are the most widely adopted, with robust fraud prevention and extensive ad network integrations. Branch excels at deep linking and user journey mapping. Kochava offers strong privacy compliance features. For most marketers, AppsFlyer or Adjust will cover your needs with the best balance of features and platform support.
Once you've chosen your MMP, create an account and add your app. You'll need to provide your app's bundle ID for iOS (found in your Xcode project) and package name for Android (found in your AndroidManifest.xml file). The MMP will generate SDK credentials including an app ID and SDK key that you'll use in the next step.
Next, configure postback URLs for each ad network you're running campaigns on. A postback is how your MMP sends attribution data back to the ad platform. In your MMP dashboard, navigate to the integrations section and enable each ad network you plan to use. The MMP will provide pre-configured postback templates for major platforms like Meta, Google, and TikTok. Understanding mobile app attribution tracking fundamentals will help you configure these settings correctly.
Finally, verify your MMP account is properly linked to your app store listings. For iOS, you'll need to add your Apple App Store credentials. For Android, connect your Google Play Console account. This linkage allows the MMP to verify that installs are legitimate and pull additional metadata about your app.
Before moving on, double-check that your MMP dashboard shows your app as "Active" and that all planned ad network integrations show as "Connected." This foundation is critical because everything else builds on top of it.
Now comes the technical work: adding your MMP's SDK to your actual app code. This is where many marketers hand things off to their development team, but understanding the process helps you troubleshoot issues and verify the implementation is correct.
For iOS apps, you'll typically add the MMP SDK using either CocoaPods or Swift Package Manager. If your team uses CocoaPods, they'll add a line to your Podfile specifying the MMP SDK and version. For Swift Package Manager, they'll add the SDK's GitHub repository URL in Xcode's package dependencies. The MMP's documentation will provide exact instructions for both methods.
For Android apps, the SDK is added through Gradle. Your developers will add the MMP's Maven repository to your project-level build.gradle file, then add the SDK dependency to your app-level build.gradle file. Again, the MMP provides exact syntax in their documentation.
Once the SDK is added to your project, it needs to be initialized in your app's startup code. For iOS, this typically happens in your AppDelegate's didFinishLaunchingWithOptions method. For Android, it's in your Application class's onCreate method. The initialization code includes your SDK credentials from Step 1 and configuration options for attribution windows and data collection.
Here's the critical part: you must initialize the SDK before any other tracking or analytics code runs. If you initialize your MMP SDK after other SDKs, you risk missing the attribution data that identifies which campaign drove the install. Following best practices for tracking conversions accurately from the start prevents costly data gaps.
Before deploying to production, test the integration in sandbox or debug mode. Most MMPs provide a debug mode where you can see real-time logs of SDK events. Install your app on a test device, and watch the MMP's dashboard for a test install event to appear. You should see the install attributed to your test device within seconds.
Common integration errors include incorrect SDK credentials (causing authentication failures), missing privacy permissions in your Info.plist or AndroidManifest.xml files, and SDK initialization happening too late in the app lifecycle. The MMP's debug logs will help you identify these issues before they affect real user data.
To verify successful implementation, perform a test install from each major ad platform you'll be running campaigns on. Click a test ad, install the app, and confirm that the MMP dashboard shows the install attributed to the correct campaign. If attribution isn't working, check that your postback URLs from Step 1 are configured correctly.
With your MMP SDK live in your app, you need to connect each ad platform so they can receive attribution data and optimize their campaigns. Each platform has its own requirements and integration steps.
Start with Meta. In your Meta Events Manager, create an app event dataset for your app. You'll connect this to your MMP by providing Meta's app ID to your MMP dashboard and configuring the MMP's postback URL in Meta's partner integrations section. Meta also requires server-side event tracking through the Conversions API for optimal performance, especially post-iOS 14.5. Your MMP should handle sending these server-side events automatically once configured. For detailed guidance, review conversion tracking tools for Facebook Ads.
For Google Ads, you'll need both Firebase integration and MMP postbacks. First, add Firebase to your app following Google's setup guide. Firebase provides Google with first-party data about app events. Then, in your MMP dashboard, enable the Google Ads integration and configure conversion postbacks. This dual approach gives Google the most complete picture of your app install performance.
TikTok requires connecting their Events API to receive app event data. In TikTok Ads Manager, navigate to Assets, then Events, and create an app event source. You'll get an Access Token and Pixel Code that you'll add to your MMP's TikTok integration settings. TikTok's system is similar to Meta's Conversions API in that it relies on server-side data transmission. Learn more about tools for tracking TikTok ads to maximize your campaign performance.
Across all platforms, maintain consistent naming conventions for your campaigns, ad sets, and ads. When attribution data flows back from your MMP, inconsistent naming makes it nearly impossible to analyze performance. Establish a naming structure like "Platform_Campaign_Objective_Date" and enforce it across all team members who create campaigns.
One often-overlooked step: verify that attribution windows match between your MMP and each ad platform. If your MMP uses a 7-day click attribution window but Meta is set to 1-day click, you'll see discrepancies in reported conversions. Align these settings to ensure data consistency across your stack.
Test each integration by running a small test campaign on each platform. Install your app from each test ad and confirm that the MMP shows the correct attribution source. Then check that each ad platform's dashboard also shows the conversion event. If data appears in your MMP but not in the ad platform, your postback configuration needs adjustment.
An install is just the beginning of the user journey. What you really care about is what happens after the install: do users sign up, make a purchase, subscribe, or complete other valuable actions? Tracking these post-install events is essential for understanding which campaigns drive quality users, not just download numbers.
Start by identifying your key conversion milestones. For most apps, these include registration or account creation, first meaningful action (like creating a profile or completing onboarding), first purchase or subscription, and repeat engagement events. Map out your ideal user journey and pinpoint the moments that indicate a user is becoming valuable.
Each of these events needs to be tracked in your app code. When a user completes a registration, your app should fire an event to your MMP SDK. This typically looks like a single line of code that calls the MMP's event tracking method with an event name and optional parameters. For example, when a user completes a purchase, you'd track a "purchase" event with parameters for revenue amount, currency, and product details.
Revenue tracking deserves special attention. When tracking purchase or subscription events, always include the actual revenue value and currency code. Your MMP will aggregate this data to show you true return on ad spend (ROAS) for each campaign. Implementing revenue tracking through attribution platforms ensures you're optimizing for profitability, not just conversion volume.
Create event funnels to understand user progression. Most MMPs let you build funnels that show what percentage of users move from install to registration to first purchase. These funnels reveal where users drop off and help you identify which traffic sources bring users who complete the full journey versus those who install and immediately churn.
Don't track everything. Focus on events that directly correlate with business value. Tracking too many events creates noise in your data and makes it harder to identify meaningful patterns. A common framework is to track 5-7 core events: install, registration, first key action, first purchase, repeat purchase, subscription start, and subscription renewal.
Test your event tracking thoroughly before launching campaigns. Go through your app's user journey yourself, completing each tracked action, and verify that events appear in your MMP dashboard with correct parameters. Revenue values should match exactly what you entered. Event names should be consistent and properly formatted.
iOS privacy changes have fundamentally altered how app install attribution works. Understanding and properly configuring SKAdNetwork (SKAN) is no longer optional if you run iOS campaigns.
First, implement the App Tracking Transparency (ATT) prompt. Since iOS 14.5, apps must request explicit user permission to track activity across other apps and websites. This prompt appears when your app first calls the AppTrackingTransparency framework's requestTrackingAuthorization method. The timing of this prompt matters significantly. Show it when users understand your app's value and have a reason to grant permission, not immediately on first launch. Understanding the full iOS App Tracking Transparency impact helps you develop strategies to maintain attribution accuracy.
For users who grant ATT permission, your MMP can use traditional deterministic attribution with device-level data. For users who decline or don't see the prompt, attribution falls back to SKAdNetwork, Apple's privacy-preserving framework that provides limited but privacy-compliant conversion data.
SKAdNetwork works differently than traditional attribution. Instead of tracking individual users, it provides aggregate conversion data with a 24-72 hour delay. The key constraint is conversion values: you only get 64 possible values (0-63) to represent all your post-install events. This limitation requires careful mapping of your most important events to these conversion value slots.
Configuring SKAN conversion values is strategic work. You might map value 0-10 to represent installs with no post-install activity, 11-30 to represent registration milestones, 31-50 to represent purchase tiers, and 51-63 to represent high-value subscription events. Your MMP provides tools to configure these mappings, and many offer recommended schemas based on your app category.
The conversion value can update during a conversion window (typically 24 hours), but only to a higher value. Once a user triggers a high-value event, you can't downgrade them. This means your mapping should progress from low-value to high-value events in numerical order.
Balance probabilistic modeling with deterministic data. Most MMPs offer probabilistic attribution methods that use statistical modeling to attribute installs even without device-level tracking. These models analyze patterns like IP addresses, user agent strings, and timing to make educated guesses about attribution. While less accurate than deterministic tracking, probabilistic methods help fill gaps in your attribution data.
Monitor your ATT opt-in rate in your MMP dashboard. Industry averages hover around 25-30% globally, but rates vary significantly by app category and geography. Higher opt-in rates give you more deterministic data and better campaign optimization capabilities.
Your MMP captures attribution data, but that's only part of the picture. To make truly informed decisions, you need to connect app install data with web conversions, CRM events, and revenue data from your backend systems.
Most MMPs offer data export features that let you send attribution data to external analytics platforms. This might be a direct integration with platforms like Google BigQuery, Snowflake, or Redshift, or it might be a scheduled export of CSV files. Configure these exports to run daily so your central analytics system always has fresh attribution data.
The challenge is unifying data from multiple sources. A user might see your Facebook ad, visit your website, then later install your app and make a purchase. Traditional analytics tools treat these as separate events across different platforms. But they're all part of the same customer journey, and understanding that full path is critical for accurate attribution. Solving multiple ad platforms tracking issues requires a unified approach to data collection.
This is where platforms like Cometly can unify app install data with web conversions and CRM events. By connecting your MMP data to a central attribution platform, you can see the complete customer journey across every touchpoint. A user's path might start with a Google search ad, continue through email nurture, include an app install from a Meta retargeting campaign, and end with a subscription purchase tracked in your CRM. Cometly captures all these touchpoints and shows you which channels truly drive revenue, not just which ones get last-click credit.
Build dashboards that show true customer acquisition costs across channels. Your MMP shows cost per install, but what's the cost per paying customer? Connect your revenue data to your attribution data to calculate metrics like CAC (customer acquisition cost), LTV (lifetime value), and ROAS by campaign, ad set, and even individual ad creative.
Set up automated reporting for campaign performance. Configure daily or weekly reports that show your key metrics: installs by source, conversion rates for post-install events, revenue by campaign, and ROAS trends over time. Automated reporting ensures you catch performance changes quickly instead of discovering issues weeks later.
The goal is a single dashboard where you can see app performance alongside web performance, understand cross-platform attribution tracking, and make budget decisions based on complete data rather than siloed metrics from individual platforms.
Your tracking system is live, but your work isn't finished. Ongoing validation and optimization ensure your data stays accurate as platforms change, SDKs update, and your app evolves.
Run test installs across different attribution scenarios. Test organic installs (no ad click), paid installs from each major platform, installs after clicking multiple ads from different platforms, and installs with various time delays between ad click and install. Each scenario should attribute correctly in your MMP dashboard. If you see unexpected attribution or missing data, investigate immediately.
Check for data discrepancies between your MMP, ad platforms, and analytics tools. Some variance is normal due to different attribution windows and methodologies, but large discrepancies indicate a problem. If your MMP shows 1,000 installs but Meta only reports 700 conversions, your postback configuration might be broken or your attribution windows might be misaligned. Using conversion tracking software for multiple ad platforms helps identify and resolve these discrepancies quickly.
Set up alerts for tracking failures or unusual data patterns. Most MMPs let you configure alerts for scenarios like zero installs in a 24-hour period, sudden drops in event volume, or postback failures to ad platforms. These alerts help you catch tracking issues before they impact significant ad spend.
Perform regular audits of your event tracking code. As your app updates with new features, ensure that event tracking code is properly implemented for new user actions. A common mistake is launching a new feature without adding event tracking, creating blind spots in your conversion funnel.
Monitor SDK version updates from your MMP. Mobile measurement platforms regularly release SDK updates that improve attribution accuracy, add new features, or address privacy compliance requirements. Schedule quarterly reviews of your SDK version and update when new releases offer meaningful improvements.
Test your tracking after major app updates. Before releasing a major version update to production, run through your full testing protocol to verify that attribution and event tracking still work correctly. App updates can inadvertently break tracking if SDK initialization code is moved or event tracking calls are removed.
Keep documentation of your tracking setup. Document which events you track, how conversion values are mapped for SKAdNetwork, which attribution windows you use for each platform, and any custom configurations in your MMP. When team members change or issues arise months later, this documentation is invaluable for troubleshooting.
You now have a complete conversion tracking system for app installs. Your MMP SDK captures every install, attributes it to the correct campaign, and sends that data back to your ad platforms. Post-install events show you which campaigns drive valuable users, not just download numbers. SKAdNetwork configuration ensures you maintain attribution data even from privacy-conscious iOS users. And your attribution data flows into a central analytics dashboard where you can see the complete customer journey.
Quick checklist to confirm everything is working: Your MMP SDK is live and receiving install events in real time. All ad platform integrations show active postbacks with recent data transmission. Post-install events are firing correctly for key conversion milestones. SKAdNetwork is configured with conversion value mappings for iOS campaigns. Attribution data is flowing to your central analytics platform with daily updates.
The real power of this system emerges when you use it to make better decisions. Look beyond install volume to understand which campaigns drive users who actually convert. A campaign with fewer installs but higher conversion rates to paid subscriptions is more valuable than one with high install volume and low engagement. Use your attribution data to identify these patterns and shift budget accordingly.
Watch for trends in your conversion funnels. If users from Google App Campaigns consistently have higher registration rates than users from TikTok, that insight should inform your creative strategy and budget allocation. If iOS users convert at higher rates than Android users for certain campaigns, adjust your bidding strategies to reflect that value difference.
Feed better data back to your ad platforms. As your tracking captures more post-install events and revenue data, ad platform algorithms get better at finding similar high-value users. The conversion data you send through postbacks directly improves targeting and optimization. This creates a virtuous cycle where better tracking leads to better campaign performance, which generates more revenue to reinvest in growth.
Ready to elevate your marketing game with precision and confidence? Discover how Cometly's AI-driven recommendations can transform your ad strategy. Get your free demo today and start capturing every touchpoint to maximize your conversions.
