Measuring iOS App Campaigns After ATT
For app marketers and analysts running user acquisition on iPhone

ScaleMind Media
App Performance, powered by FuelMyApp · 13 November 2026 · 2 min.
Since Apple introduced App Tracking Transparency in 2021, apps must ask permission before tracking users across other companies’ apps and websites. Many users say no. That ended the easy, user-level attribution that iOS campaigns once relied on, and it changed how iOS campaigns are bought, optimised and judged.
What changed with App Tracking Transparency?
Apps must show a permission prompt before they can access the device’s advertising identifier for tracking. When users decline, which many do, ad networks and MMPs cannot match an ad click to an install for that user in the old way. Deterministic, user-level attribution became partial.
How does SKAdNetwork work?
SKAdNetwork is Apple’s privacy-preserving attribution system. When an ad leads to an install, Apple sends the ad network a postback saying an install happened from a campaign, with a conversion value that the app can set to describe what the user did early on. Postbacks arrive with a delay, carry limited detail, and do not identify the user. Newer versions added multiple postbacks over time and coarser values when volumes are small. Apple has since introduced AdAttributionKit, which builds on the same ideas.
What does that mean in practice?
- Plan conversion values carefully. You choose what early in-app behaviour the values represent, such as registration or a first purchase, and partners optimise to them. Your MMP can help design the schema.
- Expect delays. Results arrive later than on Android, so judge campaigns over longer windows.
- Expect less detail. You will see less by creative, placement and day.
- Use more than one view. Combine SKAdNetwork data, consented user-level data, your own revenue data, and tests such as turning a channel off in one region to see what changes.
Can you still pay per result on iOS?
Yes, but agree how results are counted. Partners can be paid on events reported through SKAdNetwork or AdAttributionKit, on consented attributed events, or on modelled results agreed in advance. The less direct the measurement, the more important it is to agree the method in writing before launch.
Should you shift budget from iOS to Android?
Only if the numbers say so. iOS users are often more valuable in some markets and categories, so less precise measurement does not mean worse results. Compare cost per verified business outcome, such as subscriptions or deposits from your own data, across platforms rather than comparing attributed installs.
Is this legal or policy advice?
No. Apple’s rules and tools change; check Apple’s current developer documentation and your MMP’s guidance before you change your setup.
Work with ScaleMind
ScaleMind’s App Performance campaigns, powered by FuelMyApp, work alongside your MMP and bill only on events it verifies. Our guide to CPI, CPA and CPE pricing covers choosing the event to pay for.
Running app campaigns for clients?
App Performance, powered by FuelMyApp, gives agencies pay per result CPA traffic from OEM, CTV, premium in-app and programmatic sources in 40+ countries, priced on your client’s event and billed only on what their MMP verifies, such as a KYC, a first deposit or a first order.