Since September 1, 2026, Apple’s AdServices API adds an exact engagement timestamp to every iOS attribution claim it returns — including claims that previously carried no timestamp at all. Every claim is now evaluated against your app’s configured attribution lookback window using that timestamp. A claim that would have been accepted under the old, timestamp-optional behaviour can now be rejected if the actual engagement falls outside the window you’ve configured, which means some measurement partners and in-house attribution stacks are seeing a small but real drop in installs credited to Apple Ads, purely from claims that no longer clear the window check.
The change itself is a precision improvement — a claim with a real timestamp is more trustworthy than one without — but precision cuts both ways. Teams that never audited their lookback window configuration, because a loosely-timed claim used to get through anyway, are now finding out what that window actually does. A campaign that looked stable for months can show a step-down in reported installs this month with no change in real user behaviour, purely because attribution got stricter about when an engagement has to have happened.
This lands on top of Apple Ads’ earlier move to view-through attribution, which already blurred the line between “someone tapped an ad” and “someone saw an ad and installed anyway.” Now that every claim carries a real timestamp, teams finally have the data to separate those cases properly — if they update their reporting to look at it.
Data Points to Track
- Engagement timestamp per attribution claim, captured and stored, not discarded after the pass/fail lookback check
- Lookback window configuration, documented per app and per measurement partner, since a mismatch between your intended window and what’s actually configured is now directly visible in accepted-vs-rejected claim counts
- Claim rejection rate before and after September 1, 2026, as a distinct metric from raw install volume, to isolate the effect of the timestamp change from any real change in campaign performance
- View-through vs. tap-through split, now more reliably computable with real timestamps, rather than inferred
- Cross-MMP consistency, since AppsFlyer, Adjust, Singular, and Kochava each surface this change slightly differently — confirm your dashboards agree with the raw AdServices API response
Setup Steps
- Pull your app’s current attribution lookback window setting from Apple Ads and confirm it matches what you actually intend to measure, not a legacy default.
- Compare Apple Ads-credited installs week-over-week around September 1, 2026, isolating any step change and attributing it explicitly to the timestamp rollout before assuming a real performance shift.
- Update your MMP or in-house attribution pipeline to store the new engagement timestamp field rather than dropping it once the lookback check passes.
- Rebuild view-through vs. tap-through reporting using the newly reliable timestamp data, if you haven’t already separated these since Apple Ads introduced view-through credit.
- Set an internal alert on claim rejection rate, so a future tightening of the lookback logic shows up as a metric change you catch, not a mystery someone has to reverse-engineer later.
Actionable Insights
A metric getting more precise is good news dressed as a bad-looking chart. The teams that handle this well are the ones who can point to the exact date the measurement methodology changed and explain the resulting dip without touching the campaign itself. Everyone else will spend a budget cycle chasing a performance problem that was actually a measurement-accuracy improvement — and may cut spend on a channel that didn’t get worse, it just got honestly measured.
Related Resources
Need help tracking this in your app?
Our team sets up analytics pipelines for mobile and web teams every day. Talk to us and get your first events flowing in under an hour.
Talk to an expert