Quick answer
Apple accepts up to 10 screenshots per device class and Google Play accepts up to 8 per device type, with a minimum of 2 on Play. Almost nobody scrolls past the first three, so those three carry nearly all the conversion weight. Five to six well made screenshots is the practical sweet spot for most apps, because it fills the visible row and gives search browsers something to swipe without diluting your best work.
Short Answer
> Quick answer: The hard limits are 10 screenshots per device class on the App Store and 8 per device type on Google Play, where the minimum is 2. Those are ceilings, not targets. The behaviour that matters is that most people never scroll the screenshot carousel at all, so the first two or three images do nearly all the work.
Five to six is the practical answer for most apps. That's enough to cover your main features for the minority who swipe, without spreading your design effort so thin that the first three suffer. If you only have three genuinely strong screenshots, ship three, because a weak fourth image is worse than no fourth image.
Five to six is the practical answer for most apps. That's enough to cover your main features for the minority who swipe, without spreading your design effort so thin that the first three suffer. If you only have three genuinely strong screenshots, ship three, because a weak fourth image is worse than no fourth image.
What the Store Actually Shows First
The count that matters isn't the count you upload, it's the count a browsing user sees before deciding.
On the App Store, a search result shows a preview of your first one to three screenshots depending on orientation, right there in the results list, before anyone taps into your page. Portrait screenshots show three across, and a single landscape image takes the same space on its own. That's your real shop window.
On the product page itself, the carousel shows roughly two and a half portrait screenshots on a typical phone, with the edge of the third visible to signal that there's more. Everything past that requires a deliberate swipe.
Google Play behaves similarly on the listing page, showing the first few and cutting off mid image to hint at scrolling. Play also weights the feature graphic and video heavily in some placements, which changes where attention lands first.
So the practical structure is that screenshots one to three are the pitch, four to six are for people already interested, and seven onward are seen by almost nobody. Our first screenshot guide covers what to put in the most valuable slot.
On the App Store, a search result shows a preview of your first one to three screenshots depending on orientation, right there in the results list, before anyone taps into your page. Portrait screenshots show three across, and a single landscape image takes the same space on its own. That's your real shop window.
On the product page itself, the carousel shows roughly two and a half portrait screenshots on a typical phone, with the edge of the third visible to signal that there's more. Everything past that requires a deliberate swipe.
Google Play behaves similarly on the listing page, showing the first few and cutting off mid image to hint at scrolling. Play also weights the feature graphic and video heavily in some placements, which changes where attention lands first.
So the practical structure is that screenshots one to three are the pitch, four to six are for people already interested, and seven onward are seen by almost nobody. Our first screenshot guide covers what to put in the most valuable slot.
Why More Is Not Better
The temptation to fill every slot is strong, and it usually costs conversion rather than adding to it.
Design effort is finite. Ten screenshots made in the same afternoon as five means each one got half the thought, and the ones that suffer are usually the early ones because you start with your obvious features and improvise the rest. Diluting slot one to fill slot nine is a bad trade at any level of traffic.
Filler is visible. When screenshots four through eight are settings screens, an empty state and a generic testimonial, the message a browsing user gets is that the app ran out of things to show. Fewer strong images read as more confident.
Every extra image is another chance to break your visual system. Consistent background, caption position and type size across screenshots is what makes a listing look professional, and consistency is harder to hold across ten than across five.
There's also a maintenance cost that nobody budgets for. Every screenshot has to be redone for each localization and re exported when your UI changes, so ten screenshots across six languages is sixty images to keep current. That's the reason most listings drift out of date.
I checked a batch of listings that were clearly built with effort at launch, and the pattern was consistent. The first two or three screenshots had been updated at some point, and slots six through ten still showed a UI that no longer existed in the app. Nobody had deliberately abandoned them, they just quietly stopped being worth the export time, which tells you what the owners actually believed those slots were worth.
I'd rather ship four screenshots that stay current than nine that go stale within a release or two, because a screenshot showing an old interface is worse than a missing one. It sets an expectation the app then fails to meet in the first thirty seconds after install.
Design effort is finite. Ten screenshots made in the same afternoon as five means each one got half the thought, and the ones that suffer are usually the early ones because you start with your obvious features and improvise the rest. Diluting slot one to fill slot nine is a bad trade at any level of traffic.
Filler is visible. When screenshots four through eight are settings screens, an empty state and a generic testimonial, the message a browsing user gets is that the app ran out of things to show. Fewer strong images read as more confident.
Every extra image is another chance to break your visual system. Consistent background, caption position and type size across screenshots is what makes a listing look professional, and consistency is harder to hold across ten than across five.
There's also a maintenance cost that nobody budgets for. Every screenshot has to be redone for each localization and re exported when your UI changes, so ten screenshots across six languages is sixty images to keep current. That's the reason most listings drift out of date.
I checked a batch of listings that were clearly built with effort at launch, and the pattern was consistent. The first two or three screenshots had been updated at some point, and slots six through ten still showed a UI that no longer existed in the app. Nobody had deliberately abandoned them, they just quietly stopped being worth the export time, which tells you what the owners actually believed those slots were worth.
I'd rather ship four screenshots that stay current than nine that go stale within a release or two, because a screenshot showing an old interface is worse than a missing one. It sets an expectation the app then fails to meet in the first thirty seconds after install.
When More Screenshots Genuinely Help
There are real cases for going long, and they have a common shape: the app has distinct audiences or distinct modes that a single sequence can't serve.
Apps with many separate features benefit, because a user searching for one specific capability wants to see that capability. A productivity app where different people arrive wanting calendars, notes and tasks can reasonably show all three rather than picking.
Games often justify more, since players scroll further and want to see variety in levels, characters or environments. The browsing behaviour is genuinely different from utility apps, and the two categories deserve different screenshot strategies rather than one template applied to both.
Apps that need to establish trust before download, like anything financial or medical, can use later slots for credentials, security details and social proof that would waste an early slot. A regulatory badge or a security certification means nothing to someone who hasn't decided they want the app yet, and it means a lot to someone who has and is looking for a reason not to.
Onboarding heavy apps are another case. If a real objection is that the app looks complicated, showing the setup flow across three later screenshots answers that objection directly, and it's the kind of thing that only helps the people who scrolled far enough to be worried about it.
And if you're running experiments, more screenshots gives you more to test. Though in practice, testing the first image against an alternative first image moves the number far more than anything you do in slot seven.
Apps with many separate features benefit, because a user searching for one specific capability wants to see that capability. A productivity app where different people arrive wanting calendars, notes and tasks can reasonably show all three rather than picking.
Games often justify more, since players scroll further and want to see variety in levels, characters or environments. The browsing behaviour is genuinely different from utility apps, and the two categories deserve different screenshot strategies rather than one template applied to both.
Apps that need to establish trust before download, like anything financial or medical, can use later slots for credentials, security details and social proof that would waste an early slot. A regulatory badge or a security certification means nothing to someone who hasn't decided they want the app yet, and it means a lot to someone who has and is looking for a reason not to.
Onboarding heavy apps are another case. If a real objection is that the app looks complicated, showing the setup flow across three later screenshots answers that objection directly, and it's the kind of thing that only helps the people who scrolled far enough to be worried about it.
And if you're running experiments, more screenshots gives you more to test. Though in practice, testing the first image against an alternative first image moves the number far more than anything you do in slot seven.
A Structure That Works
Rather than picking a number in the abstract, fill these roles and let the count fall out of it.
Slot one states what the app is and who it's for, in a caption readable at thumbnail size. This is the single highest value image in your entire listing.
Slot two shows the main thing the app does, the action a user came to perform. Slot three shows the payoff, the result or output they get from doing it.
Slots four and five cover secondary features, ranked by how often people ask about them rather than by how proud you are of building them.
Slot six is where social proof, awards or a scale claim belongs if you have something credible. If you don't, stop at five and don't invent one.
Every image needs a caption. A screenshot with no text is a picture of a UI, and nobody browsing a store is studying your UI. Build the whole set in the editor with a consistent device frame so the sequence reads as one thing rather than six exports made on different days. Apple's own screenshot specifications list the exact sizes each device class needs.
Slot one states what the app is and who it's for, in a caption readable at thumbnail size. This is the single highest value image in your entire listing.
Slot two shows the main thing the app does, the action a user came to perform. Slot three shows the payoff, the result or output they get from doing it.
Slots four and five cover secondary features, ranked by how often people ask about them rather than by how proud you are of building them.
Slot six is where social proof, awards or a scale claim belongs if you have something credible. If you don't, stop at five and don't invent one.
Every image needs a caption. A screenshot with no text is a picture of a UI, and nobody browsing a store is studying your UI. Build the whole set in the editor with a consistent device frame so the sequence reads as one thing rather than six exports made on different days. Apple's own screenshot specifications list the exact sizes each device class needs.
Getting the Number Right for Your App
Start with three and add only when you can name what a fourth image would say that the first three don't. That question kills most filler before it gets designed.
Check your own analytics rather than trusting averages. Both stores report conversion, and if you're running custom product pages on the App Store you can see how different sets perform against each other directly.
Test the order before testing the count. Moving your strongest image into slot one reliably moves conversion more than adding a seventh screenshot ever will, and it costs nothing to try. The A/B testing guide covers how to run that properly rather than swapping images and guessing.
Match the count across platforms only if it makes sense. There's no rule that says your Play listing and your App Store listing need the same number, and Play's 8 cap versus Apple's 10 already breaks the symmetry for anyone using every slot.
And revisit it when the app changes. A listing built for version one frequently still shows version one's features two years later, which is a bigger problem than having five images instead of six. Put a reminder against every major release to look at the set, and delete anything showing a screen that no longer looks like that. Deleting a stale screenshot takes a minute and usually helps more than adding a new one.
Check your own analytics rather than trusting averages. Both stores report conversion, and if you're running custom product pages on the App Store you can see how different sets perform against each other directly.
Test the order before testing the count. Moving your strongest image into slot one reliably moves conversion more than adding a seventh screenshot ever will, and it costs nothing to try. The A/B testing guide covers how to run that properly rather than swapping images and guessing.
Match the count across platforms only if it makes sense. There's no rule that says your Play listing and your App Store listing need the same number, and Play's 8 cap versus Apple's 10 already breaks the symmetry for anyone using every slot.
And revisit it when the app changes. A listing built for version one frequently still shows version one's features two years later, which is a bigger problem than having five images instead of six. Put a reminder against every major release to look at the set, and delete anything showing a screen that no longer looks like that. Deleting a stale screenshot takes a minute and usually helps more than adding a new one.
Frequently asked questions
How many screenshots does the App Store allow?
Up to 10 per device class and localization. That is a ceiling rather than a target, since most people never scroll past the first three, so five or six well made images usually outperform ten rushed ones.
What is the minimum number of screenshots on Google Play?
Two phone screenshots is the minimum and eight is the maximum per device type. Two is enough to publish but not enough to convert, so three is the realistic floor for a listing you care about.
Do people scroll through app store screenshots?
Most do not. The carousel shows roughly two and a half portrait images and everything beyond needs a deliberate swipe, so the first three carry nearly all the conversion weight. Our [first screenshot guide](/blog/first-screenshot-app-store-conversion-2026) covers slot one specifically.
Should I use all 10 screenshot slots?
Only if you have ten genuinely distinct things worth showing. Filler images read as an app that ran out of features, and every extra screenshot is another asset to redo for each language and each UI change.
Do games need more screenshots than utility apps?
Usually yes. Players scroll further and want to see variety in levels and characters, while utility app users decide fast on the first two or three. The browsing behaviour genuinely differs between the two categories.
Build the mockup in your browser.
Drop a screenshot into a real device frame and export at the exact store size — free, no signup.