Apple Search Ads measurement just got harder

Apple Ads has been reporting through SKAdNetwork since April 10, 2025, which put Apple Search Ads on the same attribution framework every other network has used since iOS 14.5, across all four placements: search results, the search tab, the Today tab and app listing pages. Apple renamed the business from Apple Search Ads to Apple Ads that same month, though most teams still say ASA. Apple’s own Attribution API still reports as it always did. The difference is that ASA now also generates SKAN postbacks, and Apple put SKAN at 77% of all referral-based conversions to the App Store when it announced the change.

The change arrived quietly and plenty of teams have not acted on it. If you run user acquisition on iOS, this is one of those updates that looks like a reporting detail and turns into a budget argument three months later. ASA can finally be compared with Meta, Google and programmatic on one methodology, which also means ASA has to hold up under that comparison. The first thing most teams see when they run it is a lower ROAS number than the Apple dashboard has been showing them all year.

What actually changed in ASA reporting

Apple registered Search Ads as a SKAN ad network with click-through attribution, on SKAN versions 1 to 3. SKAN 4 and the full AdAttributionKit feature set are still not supported. Three details matter more than the headline:

• Postbacks are not forwarded automatically. Apple does not push SKAN data to your measurement partner on your behalf. You configure the endpoint with AppsFlyer, Adjust or whichever MMP you run, or the data simply does not arrive.
• The effective measurement window is shorter. On SKAN 1 to 3 there is one postback, fired after an activity timer that restarts each time the user opens the app and then a randomized delay on top. Apple’s own Attribution API window runs longer, so users who install and convert late fall out of the SKAN view. The two numbers will not match, and the SKAN one will be lower.
• The comparison is not yet like for like. Channels on SKAN 4 report across three windows: 0 to 2 days, 3 to 7 days, and 8 to 35 days. ASA is still on a single SKAN 3 postback, so a straight ROAS comparison flatters the SKAN 4 channels until you normalize for the difference.

Why this matters for app marketers

1) Your ASA ROAS will look worse, and that is not a performance change

The drop you see when you move ASA into SKAN reporting is the window, not the channel. Late converters never make it into the postback. A team that reads that as declining performance and moves budget away from ASA is cutting spend on a measurement artifact. Rebaseline first, then judge.

2) ASA now has to defend itself in the same room as every other channel

A 2026 Kochava survey of iOS marketers put Apple Ads at 36% of iOS attribution volume, the single largest attributed channel on the platform. The reason is structural: it is the only channel where a keyword still ties deterministically to a device without an ATT opt-in. A large share of that spend is brands bidding on their own brand terms, which is exactly where the incrementality question gets uncomfortable. The same survey found only 21% of iOS marketers confidently understand iOS attribution, and only 11% describe their SKAN setup as fully implemented and actively optimized. Unified reporting will not fix a configuration nobody owns.

3) More placements are coming, and they make the read coarser

Apple starts adding ad positions below the top slot in App Store search results in March 2026, beginning in the UK and Japan. Existing campaigns are eligible automatically, there is no control over which slot you land in, and pricing does not change. You still pay per tap or per install.

What changes is the read. SKAN postbacks map to Apple’s four ad formats, and search results is one of them, so unless Apple extends that mapping, every one of those new inline positions reports in the same bucket as your top-slot installs. More inventory on the same keyword, one line in the postback. We covered what the expansion does to competition in our breakdown of the March 2026 Apple Ads update, worth reading alongside this post.

What should you do now? A practical playbook

Here is our approach to this change if we are running your user acquisition:

Configure the SKAN postback endpoint and keep the Apple API alongside it.
Set up the endpoint with your MMP so ASA data actually flows, and do not retire Apple’s Attribution API reporting. Run both, and be clear which question each answers: the Attribution API for in-channel optimization, SKAN for cross-channel comparison.

Rebaseline before you compare anything.
Set fresh ASA benchmarks off the SKAN postback, not off the Attribution API numbers you have been using. Comparing a SKAN 4 channel to a SKAN 3 ASA number without adjusting for the window difference is how good channels get defunded.

Test incrementality on brand keywords.
Brand terms are where paid ASA and organic App Store search overlap most, so that is where to run a pause test. Baseline your total search installs, paid and organic together, pause the brand campaign for at least a week in a period with no seasonality or promotions running, then compare the observed total against the total you forecast without the pause. The shortfall is the incremental contribution. Read it off total install volume, not off SKAN, which only ever sees the attributed paid side.

Split brand defense from generic discovery in your reporting.
Judge generic keywords on incremental installs. Judge brand keywords on what it costs to keep someone else off your term. Two different jobs, and one blended CPA hides both.

How Zoomd helps

We run Apple Search Ads campaigns for global brands, and we run them next to the other channels those brands buy, which is the only way a cross-channel comparison is worth anything. Our integrations with AppsFlyer and Adjust put ASA data in the same reporting layer as everything else, alongside Apple’s Attribution API, and the optimization decisions get made against roughly 250 million devices a day rather than one dashboard at a time.

If you want someone to audit how your ASA account is set up and measured, or you are starting from zero and want it built properly the first time, talk to our team.

The bottom line

ASA did not get better or worse this year- it got comparable, and comparable is harder.
The app marketers who come out ahead here are not the ones spending more on keywords. They are the ones who:

1. Get the postbacks configured and flowing
2. Rebaseline before drawing conclusions from a ROAS drop
3. Answer the incrementality question on brand terms before someone in a budget meeting asks it
4. Plan for placement reporting getting coarser as the new slots go live

Measurement is where the next round of ASA advantage gets won. It’s up to us to get you there, and beyond.

Reach out to our team today.