The App Tracking Transparency prompt was never neutral. Germany just made Apple prove it.

Anyone who has been working in native app analytics and marketing over the last few years will be familiar with the challenges that Apple’s App Tracking Transparency Framework (ATTF) has brought, along with the perceived “unfairness” of Apple’s own ability to track and advertise.

Now a regulator has officially confirmed that they do not see Apple’s tracking prompt as a neutral privacy control, rather a competition problem, and Apple has just agreed to fix it.

What happened

On 17 August 2026, Germany’s Bundeskartellamt closed a four-year investigation into Apple’s App Tracking Transparency Framework (ATTF). The result was not a fine, but legally binding commitments from Apple to redesign how the prompt works. The case ran under the special “paramount significance” abuse-control powers Germany applies to the biggest digital gatekeepers, and the finding was straightforward: if Apple sets the rules for how apps can ask to track users, those rules can’t quietly favour Apple’s own advertising business over everyone else’s. As Bundeskartellamt president Andreas Mundt put it, “if Apple sets up additional rules within its ecosystem for the use of data, these rules must not treat its own offerings better than those of its competitors.”

This isn’t the first time that ATTF has been reviewed by regulators. France’s competition authority fined Apple €150m over the same complaint in March 2025, and Italy’s AGCM fined Apple €98.6m in December 2025. Germany is the first to land on a structural fix rather than a fine, which arguably matters more: a fine is a cost of doing business, a redesigned prompt is a redesigned prompt, and Apple surely isn’t going to build one version for Germany and another for everywhere else.

Why it was seen as unfair

The complaint the ad industry has been making since ATT launched in 2021 comes down to three specific asymmetries, not a vague sense that Apple was being self-interested.

Apple’s own apps ran on a separate, lighter consent mechanism. Third-party apps had to show the full ATT prompt on top of whatever privacy consent flow they already had under GDPR. This meant that users were often hit with two stacked consent asks, one of which Apple’s own services never had to clear.

The prompt design itself pushed behaviour in one direction. France’s regulator was specific about the mechanics: users needed only one tap to refuse tracking, but had to actively confirm through more steps to allow it. That’s not neutral copy, that’s asymmetric friction built into the interaction itself, on top of wording (“allow this app to track you across apps and websites owned by other companies”) and icons the German regulator itself called discouraging.

The result was predictable: third-party opt-in rates fell from roughly the 60-70% range pre-2021 to somewhere in the 20-30% range after, while Apple’s own advertising business, Apple Search Ads in particular, wasn’t dragged through the same friction. Apple was simultaneously the referee setting the tracking rules for the entire App Store and a player competing for the same ad targeting data.

What actually changes

Apple has four months from the decision being served to implement the fix, and has to test it with app publishers before rollout. Three things change:

• The prompt has to look and read the same regardless of whether it’s Apple or a third party asking. This must mean the same wording, same layout, no more discouraging symbols on one side only.
• Publishers get more flexibility to fold Apple’s mandatory ask into their own consent flow instead of stacking two separate prompts, and can explain to users in that same flow why personalised ads matter to their business.
• The commitments are binding for seven years with an independent monitoring trustee watching compliance, so this isn’t a one-off patch, it’s a standing constraint on how Apple can design consent UI on iOS going forward.

What this means if you work with mobile attribution or app data

Realistically you’re looking at early-to-mid 2027 before the new prompt is live and opt-in behaviour actually shifts. But it’s worth flagging now, for two reasons.

First, whatever iOS attribution modelling, SKAN setup, or MMM you’ve built around “most users say no” assumptions was built around a prompt that’s about to change. If Apple’s redesign genuinely levels the playing field, expect opt-in rates to move, and your baseline comparisons across that transition period are going to be noisy in the same way GA4’s conversion-window change or Merchant Center’s reporting overhaul make trend lines noisy. You can expect more of your favourite conversations to happen: “it’s not performance that has changed, it’s the underlying measurement”. As always, it’s important to ensure these changes are flagged appropriately in your data.

Second, it’s a useful case study for a point worth making more broadly: the platform holding the data doesn’t get to be neutral just because it frames its rules as privacy protection. That’s true of Apple’s tracking prompt, and it’s worth asking the same question of any platform setting the terms for how your data gets collected, labelled, or reported.

---

Sources: Bundeskartellamt press release, 17 August 2026; Autorité de la concurrence, France, March 2025; AGCM, Italy, December 2025; VideoWeek; MediaNama