Docs

ATT Opt-In Rate Tracking Under Regulatory Pressure

Regulators in France and Italy have fined Apple over ATT — track opt-in rate, prompt wording, and regional variance before your attribution model breaks.

Acquisition

Apple’s App Tracking Transparency prompt has been fined twice in the last year — France’s competition authority imposed a €150 million penalty in March 2026, and Italy’s AGCM followed with a €98.6 million fine in December 2025, both ruling that Apple’s ATT framework unfairly disadvantaged third-party developers relative to Apple’s own tracking. Neither ruling forces an immediate change to the prompt itself, but they signal that ATT’s design — and by extension, opt-in behaviour — is now a live regulatory target across the EU. Teams that built attribution models around a single, stable opt-in rate are about to find that assumption tested region by region.

Most teams still treat ATT opt-in as a single global number, checked occasionally and rarely broken down further. That was defensible when the prompt was static and uniform. It isn’t anymore. Regulatory pressure is pushing toward regional variants of the prompt, publishers are lobbying for changes to when and how it’s shown, and any future adjustment Apple makes in response to fines will shift opt-in rates unevenly across markets — which quietly degrades attribution coverage in some regions while leaving others untouched, with no obvious signal unless you’re already tracking the rate by geography.

Data Points to Track

  • ATT prompt outcome (accepted, denied, not_determined, restricted), captured via ATTrackingManager.trackingAuthorizationStatus immediately after the prompt resolves
  • Opt-in rate segmented by country/region, since regulatory and cultural differences already produce meaningfully different baselines and will diverge further as enforcement continues
  • App version and iOS version at time of prompt, so a shift in opt-in rate can be attributed to an OS-level prompt change rather than assumed to be organic
  • Time-to-prompt from first app open, since pre-prompt priming screens measurably affect opt-in and are themselves a target of regulatory scrutiny over “dark pattern” framing
  • Downstream attribution match rate for opted-in vs opted-out cohorts, to quantify in dollars — not just percentage points — what a given opt-in rate is actually costing in attributed revenue

Setup Steps

  1. Log the ATT authorization status as its own event the moment the prompt resolves, rather than inferring it later from whether SKAdNetwork or IDFA-based attribution data arrives.
  2. Tag every ATT event with country code, app version, and OS version so a change in opt-in rate can be isolated to a specific cohort instead of read as a single blended trend.
  3. Instrument any pre-permission priming screen separately from the system prompt itself, capturing whether the user saw priming and what they did before the OS-level prompt appeared.
  4. Build a weekly opt-in rate dashboard segmented by region, not just a single global figure, so a regulatory-driven shift in one market doesn’t get averaged away.
  5. Reconcile opt-in rate against attributed install volume monthly, so a drop in opt-in shows up as a quantified revenue-attribution gap, not just a smaller percentage.

Actionable Insights

A sudden change in opt-in rate isolated to one country is very likely a prompt or regulatory change in that market, not a shift in your own users’ behaviour — check for a recent iOS update or local ruling before assuming anything about your app changed. If opt-in rates diverge meaningfully between regions and stay diverged, your attribution model needs region-specific correction factors rather than a single global multiplier, particularly for markets like France and Italy where the prompt itself is under active challenge. And if the gap between opted-in and opted-out attribution coverage is widening, that’s a growing blind spot in spend efficiency reporting that server-side or probabilistic attribution should be covering — not one to leave unmeasured until a fine or an OS update forces the question.

Expert help

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