The Epic v. Apple order and what it means for web-to-app funnels
A US court order and Apple's May 1 guideline update let apps link to outside purchases on the US storefront. Here is what to plan for, and what is still unsettled.
On April 30, Judge Yvonne Gonzalez Rogers found Apple in violation of a 2021 injunction requiring it to let developers point customers to purchase options on the web. On May 1, Apple published updated App Review Guidelines to comply. For app advertisers the headline is simple: on the US storefront, apps can now link out to web checkout without the restrictions Apple had imposed. The harder part is deciding what, if anything, to change in your funnel while an appeal is expected.
What the order does, and where it applies
As summarized in coverage of the order, Apple may not charge a commission or fee on purchases made outside the app, may not restrict the style, language, formatting, quantity, flow or placement of external purchase links, and may only show a neutral notice when a user is leaving for the web. The order also bars Apple from excluding certain categories of apps or developers from using links.
Apple's developer news post of May 1 says the guideline changes apply only to apps distributed on the US storefront, and that the affected guidelines are 3.1.1, 3.1.1(a), 3.1.3 and 3.1.3(a). Outside the US, nothing changes. If your user base is global, you are looking at a US-only experiment.
Apple's CEO said on the company's earnings call that Apple has complied with the order and will appeal. An appeal could change the rules again, so treat the current state as live but not permanent.
Why a web checkout matters to paid acquisition
For a subscription or in-app purchase business, a web checkout does two things. It can improve unit economics, because a payment processor typically charges less than a platform commission. Suppose a $10 monthly subscription loses 30 percent to a platform commission but only 3 percent to a processor; the difference is $2.70 of every $10. Those rates are hypothetical, and your actual terms and fees will vary. It also moves the revenue event out of the app, which is where most of your measurement was built.
That second effect is the one that bites paid media. Ad platforms and attribution tools learn from conversion events. If purchases now happen on the web, you need to get those events back to the platform you are optimizing.
What to prepare before changing anything
- Revenue events. Decide how a web purchase will be tied to the install that caused it. That usually means passing a user identifier from the app into the checkout and sending purchase events server-to-server to your attribution partner and, where relevant, to Meta and Google. Without it, campaigns optimized for purchase will see revenue disappear.
- Consent and privacy. A web checkout is a different surface with its own consent requirements. Do not assume the app's permissions carry over. Check your obligations by region and by tool.
- Paid ads that drive to web. Sending ad clicks straight to a web purchase flow, rather than the App Store, is a separate question from linking out from inside an app. We would not rely on this order alone to rewrite your ad destinations, and each ad platform has its own policies, so read them before launching.
- Apple Ads. Apple's own campaigns drive App Store product page views and downloads. Their reporting will not see a web purchase that happens later, so your cost-per-purchase comparison across channels needs a common source of truth.
A sensible test design
- Pick one US cohort and one offer, and show the web option next to, not instead of, the in-app option.
- Track conversion to paid, refunds, support contacts and first-renewal rate, not just the first payment. Fewer steps in the app can mean different customer quality.
- Compare against a holdout group seeing only the in-app purchase. Without a holdout you will credit the web flow for conversion changes that came from elsewhere.
- Keep acquisition budgets steady during the test, so you are not changing two variables.
The practical takeaway: the opportunity is real, but it is limited to one market and may be revised on appeal. Run a contained US test with the event plumbing in place first, and keep your existing in-app purchase path working while you learn.