App store screenshots and creative
App store creative — icon, screenshots and preview video — decides conversion rate, and conversion rate is a ranking input on both stores. The rules that matter are consistent across categories: the first two screenshots carry almost all the weight because most viewers never swipe; captions must be legible at thumbnail size, which in practice means around six words at a large weight; the icon has to stay distinct at 60 pixels with no readable text in it; and the preview video must show the product within its first two seconds because it autoplays muted. Test creative with Apple's product page optimization on iOS and store listing experiments on Google Play rather than shipping on taste.
Key takeaways
- The first two screenshots do most of the work — the majority of store viewers never swipe past them.
- Caption text should be readable in the search-results thumbnail, not just on the full product page.
- Icons must survive being shrunk to 60 pixels; text inside an icon is almost always wasted.
- Preview videos autoplay muted, so the first two seconds must communicate without sound.
- Portrait screenshots show fewer panels but bigger detail; landscape shows more context. Test rather than assume.
What each store requires
| Asset | App Store | Google Play |
|---|---|---|
| Screenshots | Up to 10 per device size | Up to 8 per device type |
| Visible without swiping | First 3 (portrait) or 1 (landscape) | First 3 (phone) |
| Preview video | Up to 3, 15–30 seconds, autoplays muted | 1, up to 30 seconds, YouTube-hosted |
| Icon | 1024×1024, no alpha, no rounded corners | 512×512, 32-bit PNG |
| Feature graphic | — | 1024×500, required for some placements |
How to build a screenshot set
- Decide the single promise of screenshot one — the reason someone downloads, stated as an outcome, not a feature.
- Make screenshot two the proof: the interface that delivers that outcome.
- Use screenshots three to five for secondary jobs, social proof and differentiation.
- Keep captions to roughly six words, in a weight heavy enough to read at thumbnail scale.
- Show real UI. Mostly-lifestyle panels with no product get rejected and convert worse anyway.
- Localize the caption text, not just the app language — untranslated captions are one of the most common conversion leaks in non-English storefronts.
The icon
The icon is the only creative asset that appears in every surface: search results, charts, ads, and the user's home screen after install. That reach makes it the highest-risk thing to change and the reason icon tests should come last, after screenshots have been optimized.
Design for the small size first. One shape, two colours, high contrast against both light and dark backgrounds, and no words. If a competitor grid at thumbnail scale makes your icon hard to pick out, that is the problem to solve.
Testing creative rather than debating it
- iOS: product page optimization, up to three treatments against the original, one variable each.
- Google Play: store listing experiments, which can vary graphics and text and support localized tests.
- Both: change one element per treatment, run through at least one full weekly cycle, and only ship when the reported range excludes no change.
Worked example: a screenshot set rebuilt around the first two frames
A scanning app had six screenshots, each showing one feature with a caption naming it: 'OCR', 'Cloud Sync', 'Batch Mode'. The set was tidy, comprehensive, and answered a question nobody had asked, because the visitor deciding in two seconds from a search result does not know what batch mode is or why they should want it.
The rebuild changed the first two frames only. Frame one became 'Turn any paper document into a searchable PDF' over a single legible before-and-after; frame two became 'Scan a 40-page contract in under a minute'. Both are outcome statements, both are readable at thumbnail size, and both can be understood without swiping — which matters because the majority of viewers never do.
The feature tour survived intact, moved to positions three through six where it serves the people who have already been convinced and now want detail. Total effort was two frames; the conversion difference was larger than any of the metadata work done that quarter, because the metadata was already fine and the creative had been carrying the loss.
Creative errors that cost conversion
The failures that show up most often in creative audits:
- Captions naming features rather than stating outcomes.
- Text sized for a design review rather than for a search result thumbnail.
- A first frame that is a title card or logo instead of the product.
- Six unrelated frames with no narrative order, so the set has no argument.
- An icon that only works against a light background, or that loses its detail at small sizes.
- Screenshots left unchanged after a redesign, which suppresses conversion and generates refund complaints.
Frequently asked questions
How many app store screenshots should I use?
Fill the first three to five slots properly rather than all ten thinly. Viewers rarely swipe beyond the third panel, so effort spent past that point returns little.
Do app preview videos increase installs?
They can, but not universally. Because they autoplay muted and delay the screenshots, a weak video can reduce conversion. Test it as a treatment rather than assuming a lift.
What size should App Store screenshots be?
Apple requires the current largest iPhone display size and scales it down for others; you can upload up to ten per device size. Google Play accepts up to eight per device type.
appXL Research
App Store Optimization Research Team
The appXL research team analyzes App Store and Google Play ranking data across the apps our agent manages, and publishes what it finds.