# Product page optimization: A/B testing on the App Store

> Product page optimization (PPO) is the App Store's built-in A/B test: you show up to three alternate versions of your icon, screenshots and app previews to a share of App Store visitors on iOS and iPadOS 15 or later for up to 90 days, one test at a time, and compare their conversion rate with your original page in App Analytics.

Uptika team · Published: 2026-11-05 · Last updated: 2026-11-05 · https://uptika.app/en/guides/product-page-optimization

## What are the rules of product page optimization?

| Rule | Limit |
|---|---|
| Treatments per test | Up to 3, plus your original page |
| What a treatment can change | App icon, screenshots, app previews (and header and search creative assets) |
| Tests at the same time | 1 |
| Maximum length | 90 days, or until you stop it |
| Editing after start | Not possible; stop and create a new test |
| Who sees treatments | A random share of App Store visitors on iOS 15 / iPadOS 15 or later |
| Icon treatments | Icons must be alternate icons in the binary of the live version, 1024×1024 px, built with Xcode 13 or later |
| App status | Ready for Distribution (or Pre-Order Ready for Distribution) and live on the App Store |
| Not available for | Custom product pages, Apple Watch and iMessage product pages |
| Results appear | In App Analytics after five first-time downloads tied to the test |
| Reference name | Up to 64 characters |

Text, such as the name, subtitle, description and promotional text, cannot be tested with PPO. Treatments can be localized into all or some of the languages your app supports.

## How does traffic split work?

You choose a **traffic proportion**: the share of visitors who see any treatment instead of your original. It is split evenly between treatments. With 30% traffic and three treatments, each treatment gets 10% and the original 70%. With 40% and two treatments, each gets 20% and the original 60%. Each person sees the same treatment for the whole test. If a treatment has an alternate icon, that person also sees that icon on their Home Screen after downloading.

Treatments, like your original, can appear in search results and on the Today, Games and Apps tabs.

## How do I create a test, step by step?

1. In App Store Connect, go to **Apps**, select your app, and click **Product Page Optimization** in the sidebar.
2. Click **Create Test** (or the add button next to Product Page Optimization if you have run tests before).
3. Enter a **reference name**, for example "Icon colour test, 3 icons".
4. Choose the **number of treatments** (up to three). More treatments need more time to reach a result.
5. Choose the **traffic proportion**.
6. Select the **localizations** to include. Visitors in excluded localizations are not part of the test.
7. Optional but useful: open **Estimate Your Test Duration** and pick the improvement in conversion rate you hope to detect. App Store Connect uses your daily impressions and downloads to estimate the days and impressions needed.
8. Click **Create Test**.

Then configure treatments:

1. Rename **Treatment A, B, C** with **Edit Name** so results are readable later.
2. For icons, open the **App Icon** tab (it appears only when your binary has more than one icon) and click **Change** under a treatment.
3. For screenshots and previews, open **App Previews and Screenshots** (or **Header and Search Results** for creative assets), upload new files or use **Browse Assets**. Device sizes you leave alone use the original page's assets.

Finally, start it: open the test and click **Start Test**. If new metadata needs review, click **Add for Review**, add it to a submission and **Submit for Review**. Tests without new icons can be reviewed without a new app version. Reordering existing screenshots or switching to an already-shipped icon needs no new review.

## How long should a test run?

As long as it takes to reach a confident result, and no longer than 90 days. Apple recommends waiting until at least one treatment is shown to be better or worse than the baseline with **at least 90% confidence** before you apply a treatment or stop the test. The estimate in step 7 tells you in advance whether your traffic can detect the change you care about within 90 days. If the estimate says it cannot, test a bigger change (a new first screenshot rather than a colour tweak) or use fewer treatments.

> Avoid releasing a new app version that changes the tested assets while a test is running. Apple notes it may affect the results.

## How do I read and apply results?

Open **App Analytics**, choose the test, and read impressions, conversion rate, improvement over baseline and confidence for each treatment. You can change the baseline to compare treatments with each other.

To apply a winner, apply the treatment to the original product page. Doing this while the test is running ends the test. If the winning treatment uses an alternate icon, make it the primary icon in your next app version too. A stopped test cannot be restarted; create a new test with the same treatments.

## What should I test first?

Test what most visitors see and what differs most:

1. **The first screenshot.** It shows in search results and at the top of the page.
2. **The order of the first three screenshots.** Cheap to run: reordering approved screenshots needs no review.
3. **An app preview versus none.** Previews play before screenshots; test whether yours helps.
4. **The icon.** It shows everywhere, but icon tests need the icons in the live binary.

Change one idea per treatment. If a treatment changes the icon and all screenshots at once, you will not know which change mattered. More ideas on what to put in frames are in [How to design store screenshots that convert](https://uptika.app/en/guides/design-app-store-screenshots).

## How does it compare with Google Play store listing experiments?

| | App Store PPO | Google Play store listing experiments |
|---|---|---|
| Variants per test | Up to 3 treatments | Up to 2 variants |
| Tests at once | 1 | 1 default graphics experiment, or up to 5 localized experiments |
| What you can test | Icon, screenshots, previews, header and search assets | Icon, feature graphic, screenshots; short and full description in localized experiments |
| Length | Up to 90 days | Stops automatically after 6 months; Play estimates the time needed |
| Text tests | No | Yes, in localized experiments |
| Target metric | Conversion rate | Unique user install clicks, open clicks or pre-registration clicks |
| Name length | 64 characters | 50 characters |

Google Play lets you set the minimum detectable effect and confidence level in advanced settings. Its experiment pages are under Grow users > Store presence > Store listings; see [Google Play store listing experiments](https://uptika.app/en/guides/store-listing-experiments).

## Common mistakes

- Running a test on a page with little traffic and stopping it after a week.
- Testing an icon that is not in the live binary; the App Icon tab will not offer it.
- Shipping a new version mid-test that changes the same assets.
- Declaring a winner without the 90% confidence mark.
- Expecting PPO to test text. Use [custom product pages](https://uptika.app/en/guides/custom-product-pages) or release-by-release changes for that, and [measure before and after](https://uptika.app/en/guides/measure-aso-impact).

## How do I watch the effect after applying a winner?

Conversion changes show up in App Analytics. Uptika adds the search side: the Keyword radar records your rank for every tracked keyword every day, and the ASO score has a listing conversion part, so you can see whether a new page changed more than conversion. The Rival log shows when competitors change their screenshots or icon, which can shift your results while a test runs. The Free plan tracks one app in one storefront: [start free](https://uptika.app/app/#/?signup=1).

Related: [App Store screenshot sizes](https://uptika.app/en/guides/app-store-screenshot-sizes), [App preview video specs](https://uptika.app/en/guides/app-preview-video-specs) and [App icon sizes](https://uptika.app/en/guides/app-icon-sizes).

## Questions

### What is product page optimization?

An App Store Connect feature that tests up to three alternate versions (treatments) of your product page against the original. Each treatment can change the app icon, screenshots and app previews. Visitors are assigned at random, keep seeing the same treatment for the whole test, and results appear in App Analytics.

### How do I A/B test the App Store icon?

Include every icon you want to test as an alternate icon in the binary of the version currently on the App Store (built with Xcode 13 or later, 1024×1024 px). Then create a test, open the App Icon tab of a treatment and choose the alternate icon. Icon-only tests use icons that are already approved, so they do not need new metadata review.

### How long should a PPO test run?

A test runs for up to 90 days or until you stop it. Apple advises waiting until at least one treatment is declared better or worse than the baseline with at least 90% confidence before applying it or stopping. Use the Estimate Your Test Duration option when creating the test to see how long your traffic needs.

### Product page optimization vs custom product pages?

PPO tests versions of your default page on the same traffic to find the best one. Custom product pages are separate, permanent pages with their own URLs (and optionally their own keywords) for specific audiences or ads. PPO tests are not available on custom product pages.

## Sources

- [Product page optimization (Apple Developer)](https://developer.apple.com/app-store/product-page-optimization/)
- [Overview of product page optimization (App Store Connect Help)](https://developer.apple.com/help/app-store-connect/create-product-page-optimization-tests/overview-of-product-page-optimization)
- [Create a test (App Store Connect Help)](https://developer.apple.com/help/app-store-connect/create-product-page-optimization-tests/create-a-test)
- [Configure test treatments (App Store Connect Help)](https://developer.apple.com/help/app-store-connect/create-product-page-optimization-tests/configure-test-treatments)
- [Run a test (App Store Connect Help)](https://developer.apple.com/help/app-store-connect/create-product-page-optimization-tests/run-a-test)
- [Run A/B tests on your store listing (Play Console Help)](https://support.google.com/googleplay/android-developer/answer/12053285)

---

Uptika is an ASO and store-intelligence tool for the App Store and Google Play. https://uptika.app/en