iOS 14 attribution changes UK: what advertisers need to know

What to take away

What changed when iOS 14.5 shipped

Apple began enforcing App Tracking Transparency (ATT) in iOS 14.5, released in April 2021. Apps must ask permission before tracking a person across other companies' apps and websites, and the prompt stays off until the user turns it on. Apple's App Store privacy and data use rules set out what counts as tracking and what developers must disclose.

On a UK iPhone, the prompt appears in Settings under 'Allow apps to request to track'. A refusal leaves the app usable, minus the identifier that links a click to an install, and the IDFA an app reads back is zeroed. Apple's UK support pages explain where the switch lives and what changes when it is off.

SKAdNetwork in plain terms

SKAdNetwork is Apple's privacy-preserving attribution framework for app campaigns. It reports at campaign level, hides small counts behind privacy thresholds, and sends results after a delay. The value an app writes back is a six-bit number, so it runs from 0 to 63 rather than carrying a user-level path, and the postback goes out once the conversion window closes, after a random delay of up to 24 hours. Treat it as a reporting feed, not a user-level ledger.

Apple later shipped SKAdNetwork 4.0 with iOS 16.1. It can send more than one postback over time, returns a coarse conversion value — low, medium or high — for every install, and releases a finer value only when install volumes clear Apple's privacy threshold. That is why measurement platforms ask you to define a conversion schema — the set of events the conversion value stands for — rather than a single install event.

Which attribution platforms handle the postbacks

Most UK advertisers never read a postback by hand. Mobile measurement partners including AppsFlyer, Adjust, Branch, Singular and Kochava act as the SKAdNetwork partner that receives the postback and map the conversion value back to the events in your conversion schema. On the web side, Google Analytics 4 and Google Ads model iOS conversions on the same aggregated basis, and the Meta Conversions API takes server-side events so browser restrictions do not remove the signal twice. All of them work from delayed, aggregated data: none can recover an IDFA Apple has zeroed.

What it means for Facebook Ads attribution

Meta could not keep the old per-user view, so it introduced Aggregated Event Measurement. Advertisers verify a domain in Business Manager — with a meta-tag in the site head, an HTML file or a DNS TXT record — and then rank the web events that matter most for iOS traffic; Meta allows eight events per domain, in priority order. Events that fall outside those eight may go unrecorded, so the ranking decides what Meta can credit.

Reports now mix modelled and observed conversions, and recent-day figures keep moving as postbacks arrive. Signal loss reaches your web measurement too, so read cookieless attribution in the UK before you buy new tracking. A UK plan that blends app installs, web checkout and phone leads is hit hardest.

The awkward part is the handover. SKAdNetwork can attribute an install, but if the first purchase happens on the web, nothing carries that user across. App-to-web journeys become estimates, so treat a blended cost per acquisition as a planning assumption.

Before and after, side by side

The table below compares the two counting regimes on the items UK advertisers check most often.

What UK advertisers should do first

Apple's rules do not replace your own obligations. Under UK GDPR you need a lawful basis for tracking, and under PECR regulation 6 you need consent before you store or read information on a person's device; that consent must be specific and as easy to withdraw as to give. The ICO's guidance on the use of cookies and similar technologies sets out how that works in practice, and the ICO has already told the UK's most-visited websites to fix banners that make refusal harder than acceptance. Its guidance on lawful basis covers the tests, including the limits of legitimate interests.

Then fix the mechanics. Verify the domain and rank conversion events before you touch budgets, because those settings govern what Meta can record for iOS traffic. Revisit reporting windows, and stop comparing this month with a pre-2021 baseline as if the rules were unchanged.

Finally, treat platform-reported conversions as credited rather than incremental. When windows shrink and events are modelled, incrementality testing against attribution is the honest way to judge whether a campaign added sales or merely claimed them.

Common questions

Does iOS 14 still affect UK advertisers?

Yes. ATT applies to iPhones running iOS 14.5 or later, so the opt-in rule still governs app-based tracking. The effect lands at the app and app-to-web boundary rather than in your CRM.

Does SKAdNetwork replace Facebook Ads attribution?

No. SKAdNetwork covers Apple-mediated app install attribution, while Meta reports on its own terms with modelled conversions and wider date ranges. Neither gives a user-level path.

What should UK businesses check first?

Your lawful basis, your privacy notice and your withdrawal route. Apple's data-use requirements apply on top of that for anything distributed through the App Store.

Can attribution software close the gap alone?

No. Aggregated frameworks and platform reports describe credited events, not causal lift, so add holdouts or incrementality tests before moving budget.