App Store A/B testing: product page optimization and Google Play experiments

Product page optimization tests up to three versions of your app icon, screenshots and app previews for an iOS or iPadOS app. Each treatment starts as a copy of your original page; whatever you do not change is taken from the original. Treatments can be localized into any language the app supports and are shown on the App Store on iOS 15 and iPadOS 15 or later. Icons must already be in the app binary currently on the App Store, built with Xcode 13 or later with alternate icons in the asset catalog, sized 1024 × 1024; if someone downloads from a treatment, its icon shows through the download and on their device. Tests are not available for custom product pages or for the App Store on Apple Watch or iMessage, and the app must be Ready for Distribution and live on the App Store, or in Pre-Order Ready for Distribution.

In App Store Connect, open Product Page Optimization in the app’s sidebar and create a test: give it a reference name, choose up to three treatments, and pick a traffic proportion, the share of visitors who see a treatment instead of the original. With three treatments and 30%, each treatment gets 10% of traffic. Choose which localizations to include; visitors in excluded localizations are left out. An optional estimate shows how long a test might take to reach the improvement you want, based on your impressions and downloads, and Apple notes that more treatments can take longer to reach a conclusive result. New screenshots or previews need App Review approval, which you can submit without a new app version; reordering existing ones or changing only the icon needs no resubmission. The test runs for 90 days or until you stop it, and a stopped test cannot be restarted. Releasing a new app version during a test may affect results if it changes the assets being tested.

Results appear in App Analytics once the test has five first-time downloads. Apple judges the best treatment by the estimated lift in conversion rate, with algorithms that indicate how reliable the result is. You can apply one treatment per test to your original page, and to versions in Ready for Distribution or Prepare for Submission; this cannot be undone, and applying it while the test runs stops the test. Only the screenshots and previews are applied: to adopt a winning icon, make it the default icon in your next app version.

In Play Console, go to Grow users, Store presence, Store listings and set up an experiment on your default or a custom store listing. A default graphics experiment tests the icon, feature graphic and screenshots in your default language; visitors who see a localized graphic in their language are excluded. A localized experiment tests graphics and descriptions in up to five languages. You can run one default graphics experiment or up to five localized experiments at the same time. Choose a target metric: unique user install clicks, open clicks or pre-registration clicks. Advanced settings let you pick up to two variants, the audience percentage, the minimum detectable effect and the confidence level; Google says the defaults give results fastest with reasonable accuracy, and recommends testing one asset at a time. Results show which variant performs best, may say “More data needed” or “Draw”, and you can apply a variant or keep the current listing. Experiments stop automatically after six months, and traffic returns to the current listing.

Test what people see before they tap: the icon and the first screenshots. App Store tests cannot change any text, and Google does not list the app name among what an experiment can test, so work on text separately with App Store keywords and subtitle and promotional text. For asset sizes, see app icon size, screenshot sizes and preview video specs. To show different pages to different audiences rather than test one page, use custom product pages; for the full checklist, see the ASO checklist.

A better store page converts the visitors you already get; Censuus brings visitors from the web. We rank apps by the real visit traffic they draw, and adding your app is free on the submit page. Sponsorship raises placement and adds a followed link to your website; details are on the sponsor page. We do not promise conversions, downloads or store rankings.