# App Store screenshots that convert: examples, templates and rules

*Three frames do most of the selling, so make them tell one story.*

> How to design App Store screenshots that convert: the promise, proof, how it works story, caption rules, layout templates, 6 storyboards and testing.

Source: https://www.appgrowthmarketer.com/app-store/app-store-screenshots
Author: Ar.Bhavesh Panse, AI App Growth Marketer (https://www.arbhaveshpanse.com)
Published: 2026-09-26 · Updated: 2026-09-26

App Store screenshots convert when the first three frames tell one short story: frame 1 makes a promise, frame 2 proves it, and frame 3 shows how it works. each frame carries one big caption and one real screen from the app. everything after frame 3 is detail for people who are already interested.

this page is about what goes inside the frames. exact sizes, file types and counts for both stores are in [App Store marketing assets and sizes](https://www.appgrowthmarketer.com/app-store/app-store-marketing-assets), so i will not repeat them here.

## Why the first three screenshots matter most

the first three screenshots matter most because they are what people see before they decide to tap or scroll. Apple's [product page guidance](https://developer.apple.com/app-store/product-page/) says the first one to three images appear in search results when no app preview is available. most visitors never reach frame 6, so the story has to land early.

so frame 1 must work alone, captions must be readable at thumbnail size, and the first frame should be your best reason to download, not your best feature.

## The three frame story: promise, proof, how it works

the three frame story gives each opening screenshot one job. frame 1 states the outcome the user wants. frame 2 shows evidence that the outcome is real, usually the result screen. frame 3 shows the simple action that gets them there. together they answer "what do i get", "is it real" and "is it easy".

| frame | job | caption pattern | screen to show |
|---|---|---|---|
| 1 | promise | the outcome, in the user's words | the moment of value, not the home screen |
| 2 | proof | what the result looks like | a finished result, report or answer |
| 3 | how it works | one action, how simple it is | the input step: snap, type, tap |
| 4 to 6 | key features | one feature, one benefit each | the feature in use |
| 7 to 10 | objections | privacy, pricing, platforms, trust | settings, plans, offline mode |

a quick test: read only the captions, then look only at the screens. both halves should tell the same story.

## How to write screenshot captions

write captions as outcomes the user wants, in the plain words they already use, in five to eight words. lead with a verb or a result, not your feature name. the best source of those words is your own reviews, support messages and the hooks from ads that already won, because people have already shown you what they respond to.

caption rules i use:

1. one idea per caption. if you need "and", you need two frames.
2. outcome before feature. "understand any homework question" beats "AI homework solver".
3. no claims you cannot back up: no "#1", no invented numbers, no competitor names.
4. say which features are paid if a frame shows one, as Apple's accurate metadata rules require.

a copy-paste worksheet for your first three captions:

> frame 1 (promise): [outcome the user wants] in [time or effort].
> frame 2 (proof): see [the result] for [their specific situation].
> frame 3 (how it works): just [one action]. we do the rest.

## Layout templates, described in words

a layout template is a repeatable arrangement of caption, device and screen that you reuse across frames so the set looks like one story. pick one or two templates for the whole set. mixing five styles makes the page look like five apps. the five below cover most consumer apps and are easy to build.

| template | how it looks | best for |
|---|---|---|
| big caption + device | caption across the top third, phone below showing the real screen | frame 1 for most apps |
| big caption + device + one highlighted UI element | same, with one button or result enlarged and pulled out of the phone | frame 2 proof |
| full bleed screen + caption bar | no device frame, the screen fills the image, a solid bar holds the caption | content heavy apps like reading or video |
| before and after split | left half the problem, right half the result, one caption across both | photo, fitness, cleanup apps |
| checklist frame | caption plus three short benefit lines, small screen | frames 7 to 10, objections |

## Six example storyboards (illustrative)

these six storyboards are illustrative examples i wrote for common app types, not copies of real apps. use them to see the promise, proof, how it works pattern in different categories, then rewrite every caption in your own users' words. none of them has been tested, so treat each one as a starting hypothesis.

1. **AI study app.** frame 1: "stuck on homework? get unstuck in minutes." frame 2: a step by step solution screen, one step highlighted. frame 3: "snap a photo of the question."
2. **habit tracker.** frame 1: "build one habit that actually sticks." frame 2: a 30 day streak calendar. frame 3: "one tap a day. that's it."
3. **budgeting app.** frame 1: "know where your money went this month." frame 2: a spending breakdown by category. frame 3: "connect your account in two minutes."
4. **language learning app.** frame 1: "speak your first real conversation." frame 2: a chat with corrections shown inline. frame 3: "ten minutes a day, on your commute."
5. **AI photo editor.** frame 1 as a before and after split: "fix bad lighting in one tap." frame 2: three results side by side. frame 3: "pick a photo, pick a look."
6. **sleep and meditation app.** frame 1: "fall asleep faster tonight." frame 2: a sleep story player with a timer. frame 3: "press play and put the phone down."

## How to test screenshots with product page optimization and store listing experiments

test screenshots with the stores' own tools. on the App Store, product page optimization runs up to three treatments against your original for up to 90 days. on Google Play, store listing experiments run up to two variants against your current listing. change one idea per test, usually the first caption, and let the test finish.

- **Apple.** [product page optimization](https://developer.apple.com/app-store/product-page-optimization/) can test icons, screenshots and previews. you choose what share of traffic goes to the test, and it is split evenly across treatments. Apple reports conversion rate, improvement and a confidence level. new screenshots need App Review approval before the test runs.
- **Google.** [store listing experiments](https://support.google.com/googleplay/android-developer/answer/6227309) can test the icon, feature graphic and screenshots in your default language, or descriptions too in localized experiments across up to five languages. you set a confidence level and a minimum detectable effect before launch.

a simple test plan: write two new first captions, one from your best ad hook and one from your best review, and build them as treatments with frames 2 to 10 unchanged. run the test untouched, keep other big changes out of the same window, and keep the winner only if it reaches the confidence level you set.

## Matching screenshots to ad creative and custom product pages

match screenshots to ads so the promise in the ad is the first thing people see on the store page. if a video promises "homework help at midnight" and frame 1 says "your all in one learning platform", you paid for a click and lost the install. custom product pages give each ad angle its own first frame.

Apple allows up to 70 [custom product pages](https://developer.apple.com/app-store/custom-product-pages/), each with its own screenshots, previews, promotional text and, if you want, keywords. you can point Apple Ads (formerly Apple Search Ads) variations at them. Google Play allows up to 50 [custom store listings](https://support.google.com/googleplay/android-developer/answer/9867158), which can target countries, search keywords, ad traffic and groups such as churned users.

when i ran growth at ZuAI, an AI study app that went from 10K to 2M users in eleven months at a $0.02 blended CAC, paid only scaled behind creatives that had already proven themselves. the store page is the other half of that creative, so each winning ad angle deserves a matching first frame. how the ad side works is in [app creative testing](https://www.appgrowthmarketer.com/paid/app-creative-testing).

## Tools people use to make screenshots

most teams design screenshots in Figma, using one template frame per display size and swapping captions and screens inside it. developers who ship often automate capture and framing with fastlane. browser generators with templates suit founders without a designer. whichever you use, the story and captions matter far more than the tool.

- **[Figma](https://www.figma.com/):** full control and easy localized copies of every frame.
- **[fastlane](https://docs.fastlane.tools/actions/frameit/):** snapshot captures screenshots automatically through UI tests, and frameit puts device frames and caption text around them.
- **[AppLaunchpad](https://theapplaunchpad.com/), [AppScreens](https://appscreens.com/) and [Screenshots Pro](https://screenshots.pro/):** browser based generators with templates.

## What to measure

measure product page conversion rate, meaning the share of page views that become downloads, before and after every screenshot change. App Store Connect and Play Console both report it. also watch whether the users a new set brings stay as long as before, because a caption that overpromises can lift installs and hurt retention.

- conversion rate by source: search, browse and each custom product page
- day 1 and day 7 retention for users who arrived after a change
- a dated log of every change, so you can read cause and effect

screenshots are one lever in a bigger listing. the full metadata checklist is in the [ASO guide](https://www.appgrowthmarketer.com/app-store/app-store-optimization), and the step by step plan for turning store traffic into early users is in [first 100 users from the App Store](https://www.appgrowthmarketer.com/playbooks/first-100-users-from-app-store).

## Frequently asked questions

### What should the first App Store screenshot show?

the single biggest promise your app makes, in a caption someone can read at thumbnail size, next to the real screen that delivers it. Apple says the first one to three images appear in search results when there is no app preview, so frame 1 has to work alone, without the rest of the page.

### How many App Store screenshots should I upload?

Apple allows up to 10 per display size, and you should use most of them, but put nearly all your effort into the first three. those carry search results and the first scroll. frames 4 to 10 are for people already interested, so use them for features, social proof you can back up, and pricing clarity.

### Can I A/B test App Store screenshots?

yes. on the App Store, product page optimization tests up to three treatments of your screenshots, icon and previews against the original for up to 90 days. on Google Play, store listing experiments test up to two variants. change one idea per test, such as the first caption, so you know what moved.