What Percentage of iPhone Users Opt Into App Tracking in 2026?

About 50% of iPhone users globally opt into app tracking when prompted by Apple's App Tracking Transparency (ATT) prompt, according to AppsFlyer's most recently published measurement, released in April 2025 to mark four years since ATT's rollout. That's up from roughly 30-42% in ATT's first months in 2021, and no more recent comprehensive AppsFlyer update has been published as of this writing — treat 50% as the current best-available figure, not a live 2026 reading.

How AppsFlyer Measures This

AppsFlyer defines the opt-in rate as authorized users divided by the sum of authorized and denied users whose initial tracking status was "not determined," pulled from its own attribution platform across apps that log at least 30,000 installs per quarter and country. That's a meaningful sample, but it's AppsFlyer's own customer base, not every iOS app in existence, so treat it as a strong directional benchmark rather than an exact market-wide census.

The Rate Has Climbed Steadily Since Launch

When Apple rolled out ATT in 2021, AppsFlyer's early analysis of more than 300 apps found opt-in rates of roughly 42% for apps and 30% for games — already showing that games, which tend to rely more heavily on broad-reach advertising, saw lower consent than utility or lifestyle apps. By April 2025, the global blended figure had climbed to 50%, a meaningful 10-point increase over roughly four years as users grew more accustomed to the prompt and some apps improved how they framed the ask.

The Rate Still Varies by Country

AppsFlyer's 2025 release broke the global figure out geographically: France sat just under 51% and Germany came in at 47%, both close to the global blended average, suggesting the gap between major Western markets has narrowed since ATT's early years, when opt-in behavior varied more widely by region.

What a 50/50 Split Means for Measurement

With only about half of iOS users opting into IDFA-level tracking, a campaign that measures iOS performance solely through device-level attribution is blind to roughly the other half of its actual install volume. That's the practical reason SKAdNetwork and other aggregated, privacy-preserving measurement methods exist — they're not an optional upgrade, they're the only way to see the non-consenting half of an iOS campaign's results at all.

Why Android Comparisons Can Mislead

Android has no equivalent system-level prompt gating device identifiers the way ATT does on iOS, so a campaign that looks far more efficient on Android than on iOS isn't necessarily performing better there — it may simply be measured more completely, since Android attribution isn't split roughly in half by a consent prompt the way iOS attribution is. Comparing blended iOS and Android cost-per-install figures without accounting for that measurement gap risks shifting budget toward the platform that's easier to measure rather than the one that's actually performing better.

Bottom Line

Budget for roughly half of iOS install volume being fully attributable at the user level and the other half only visible through aggregated measurement, and don't benchmark a single app's opt-in rate against the 50% global figure without checking AppsFlyer's category breakdowns first — games and apps consistently land on opposite sides of that blended average.

Sources: AppsFlyer, "Market Adaptation to Privacy-First Ecosystem Four Years After ATT" (April 2025 newsroom release); AppsFlyer 2021 preliminary ATT opt-in analysis.