What iOS privacy changes have actually removed
Apple's App Tracking Transparency framework, launched with iOS 14.5, requires apps to ask users permission before tracking them across other apps and websites. The opt-in rate for this permission request, when apps ask it plainly, is approximately 25 to 30 percent globally. That means 70 to 75 percent of iOS users are generating no device-level cross-app tracking data for Meta, TikTok, or Snapchat.
Additionally, Safari's Intelligent Tracking Prevention (ITP) limits cookie lifespans, blocks cross-site tracking, and restricts certain browser-based tracking mechanisms. A user who visits your Shopify store via Safari on iPhone generates a measurably weaker tracking signal than the same user on a desktop Chrome browser.
The combination of these two changes has reduced Meta's signal coverage on iPhone users by an estimated 30 to 50 percent depending on the user's age, location, and app usage patterns. Older users who installed apps before ATT and never updated permission settings may have stronger tracking profiles. Younger users with fresh installs are almost entirely opted out.
What still works on iOS
Server-side CAPI bypasses the browser and app-level restrictions entirely. Events sent from your server to Meta's CAPI endpoint using first-party data (email hash, phone hash, customer ID) are not subject to ATT restrictions because they do not involve device-level tracking. They use customer-provided data to match a purchase event to a Meta user.
The limitation is match rate, not coverage. Server-side events using email hash can match to a Meta user profile only if that user has a Meta account linked to the email address they used for their purchase. For most DTC brands, this match rate is 60 to 80 percent of server-side events, leaving 20 to 40 percent of server-side purchase events unmatched.
Google's approach versus Meta's approach to iOS
Google's approach to the iOS tracking gap is fundamentally different from Meta's. Google controls both the search engine and the Chrome browser on Android, which together account for a large majority of DTC search conversions. Safari users searching on Google generate click data even without cross-site tracking cookies, because the click is on Google's own platform.
Google's Enhanced Conversions work through email matching on the user's Google account, similar to Meta's CAPI mechanism but with higher match rates in many markets because more users are logged into Google accounts. For DTC brands where Google search is a primary acquisition channel, the iOS tracking gap is less disruptive than for brands heavily dependent on Meta's behavioral targeting.
Android and the separate picture
Android users are not subject to ATT and have historically generated stronger tracking signals than iOS users. In European markets, the Android and iOS user split is roughly 60 to 70 percent Android and 30 to 40 percent iOS, which is different from the US where iOS has a larger share.
For Dutch DTC brands, this means a higher proportion of your audience is on Android than in the US, which means your tracking gaps from iOS changes are somewhat smaller on a market-level basis. This does not mean iOS tracking gaps are negligible. It means the base rate of the problem is lower.
Building a measurement strategy that accounts for signal gaps
Accept that your measurement will never achieve 100 percent coverage. The goal is maximizing coverage through first-party data collection, minimizing coverage gaps through server-side tracking, and modeling the gaps you cannot close through probabilistic attribution and incrementality testing.
A measurement strategy built on this foundation is more honest and more durable than a strategy built on maximizing platform tracking coverage through workarounds that Apple regularly closes. Invest in the infrastructure that survives platform changes: first-party customer data, server-side event pipelines, and regular incrementality testing.
---