Reference library · Glossary
Product page optimization: Apple's A/B test, and how much traffic it really needs
Apple's free A/B testing for icons, screenshots and previews — powerful for apps with traffic, and easy to waste on small tweaks if you don't have much.
Checked against Apple's product page optimization guide, App Store Connect Help and App Store Connect Analytics help, Apple's What's New page and Developer News (Aug 5 and Oct 5, 2026), on Oct 10, 2026.
Definition
- Product page optimization
- Product page optimization is Apple's built-in A/B test for your default App Store page: up to three treatments with different icons, screenshots or app previews, shown to a random share of visitors for up to 90 days.
Also called: PPO · App Store A/B testing
The short answer
Product page optimization lets you test up to three alternate versions of your default App Store page — different icons, screenshots or app previews — against the original. Apple shows them to a random share of visitors on iOS 15 or later for up to 90 days and labels a treatment better or worse once it reaches 90% confidence. Low-traffic apps should test big changes.
- Treatments
- Up to 3
- Against your original page
- Test length
- Up to 90 days
- One test at a time
- Winner threshold
- 90% confidence
- Apple uses Bayesian methods
- Who sees it
- iOS 15+
- A random share of visitors
01What product page optimization does
Compare different app icons, screenshots, and app previews on your App Store product page to find out which resonate with people most. Test up to three alternate product page versions against your original, view the results in App Analytics, and set the best performing version to display to everyone on the App Store.
| Rule | What Apple says |
|---|---|
| What you can test | App icons, screenshots and app previews — and, since 2026, the new header and search-result assets |
| How many | Up to three treatments against the original; one test at a time |
| Who sees it | A random share of visitors on iOS and iPadOS 15 or later, who keep seeing the same treatment |
| Where | Treatments can appear in search results and on the Today, Games and Apps tabs |
| Duration | Up to 90 days; a test can't be changed once started or restarted once stopped |
| App Review | New metadata in treatments must be approved; tests without alternate icons can be submitted without a new app version |
| Icons | Every alternate icon must already be in your app's current binary |
It tests your default page only. Apple's help says tests “aren't available for custom product pages” — see our custom product pages entry for how those differ.
According to ASOitis's sample-size estimate at Apple's 1.6% average product page conversion rate, an app with 500 product page views a day splitting traffic evenly would need about 84 days to detect a 20% lift and roughly 319 days to detect 10% — longer than Apple's 90-day test limit — so small apps should test bold changes, not tweaks.
02How Apple decides a winner
Apple's results help says it identifies winning treatments “using Bayesian techniques”, and that “when a treatment reaches 90% confidence, it may be labeled as Performing Better or Performing Worse compared to the baseline.” If the data won't be enough within 90 days, a treatment is marked “Likely to be Inconclusive”. Results start appearing after five first-time downloads associated with the test.
- Wait for a label. Apple recommends not stopping or applying anything until a treatment is declared better or worse with at least 90% confidence.
- Applying is one-way. You can apply one treatment per test and can't undo it; the screenshots and previews switch over, and a winning icon has to be set as the default in your next app version.
- Mind your releases. Apple says releasing a new version during a test “may impact the results” if it changes the assets being tested.
03How much traffic a test needs
Apple's own guidance hints at the problem: “Larger improvements (e.g., 30% lift) require fewer weeks of data than smaller improvements (e.g., 1% lift).” To put numbers on it, we ran a standard two-proportion sample-size calculation at 90% confidence and 80% power, starting from the 1.6% average default-page conversion rate Apple cites on its custom product pages page. Apple's Bayesian method differs, so treat these as orders of magnitude:
| Relative lift to detect | At 1.6% conversion | At 5% conversion |
|---|---|---|
| +5% | ≈ 312,000 | ≈ 96,000 |
| +10% | ≈ 80,000 | ≈ 25,000 |
| +20% | ≈ 21,000 | ≈ 6,400 |
| +30% | ≈ 9,700 | ≈ 3,000 |
| Product page views per day | +10% lift | +20% lift | +30% lift |
|---|---|---|---|
| 500 | ≈ 319 days — beyond the 90-day limit | ≈ 84 days | ≈ 39 days |
| 2,000 | ≈ 80 days | ≈ 21 days | ≈ 10 days |
| 10,000 | ≈ 16 days | ≈ 4 days | ≈ 2 days |
The practical rule for indie apps: with a few hundred page views a day, only big differences can win inside 90 days. Test a new icon concept or a completely different first screenshot, use one treatment instead of three, and give it most of your traffic. Save colour and caption tweaks for when you have thousands of views a day.
04Product page optimization vs custom product pages
| Product page optimization | Custom product pages | |
|---|---|---|
| Purpose | Find the best default page | Match a page to a link, ad or search |
| Audience | A random share of visitors | People who follow its link or search its keywords |
| What changes | Icons, screenshots, previews | Screenshots, previews, promotional text, keywords, deep link |
| Limit | Up to three treatments, 90 days | Up to 70 pages, no expiry |
Apple's 2026 creative assets work with both: its What's New page says you can use the new header and search-result visuals “on your custom product pages and even test them using product page optimization.”
Frequently asked questions
What can you test with product page optimization?
App icons, screenshots and app previews on your default App Store page, plus the new header and search-result assets Apple added in 2026. Alternate icons must already be included in your app's current binary.
How long does an App Store A/B test run?
Up to 90 days, and you can run one test at a time. Apple recommends waiting until a treatment is declared better or worse with at least 90% confidence before applying or stopping it.
Do product page optimization tests need App Review?
New screenshots and previews in a treatment must be approved by App Review. Tests without alternate app icons can be submitted independently of a new app version; icon tests need the icons in your published binary.
Why did my test say “Likely to be Inconclusive”?
Apple uses that label when there likely won't be enough data within 90 days to tell whether a treatment performs differently. Low-traffic apps hit it often; test bigger changes with fewer treatments.
Can I test custom product pages?
No. Apple's help says product page optimization tests aren't available for custom product pages. Compare custom pages by changing them over time and reading their App Analytics.
Where to go next
Sources
- Product page optimization — Apple Developer
- Product page optimization results — App Store Connect Analytics Help
- Configure test treatments — App Store Connect Help
- Overview of product page optimization — App Store Connect Help
- What's new for App Store — Apple Developer
- Custom product pages (1.6% default conversion figure) — Apple Developer
Published Oct 11, 2026 · last checked Oct 11, 2026. Found something out of date? Tell us and we'll fix it within a day. Machine-readable version: /glossary/product-page-optimization.md