App previews vs screenshots: which actually earns the slot

An app preview does not sit alongside your screenshots. It sits in front of them, in the first slot, and it autoplays muted. That is the whole reason this decision matters: adding a video does not add to your listing, it replaces the thing that was working hardest.

If your first screenshot is good and your video is mediocre, adding the video makes your listing worse.

What a preview actually is

iOS: 15–30 seconds, captured from the app itself, up to three per display class. Autoplays muted in the listing. Apple requires the footage to come from the app — no external footage, no device frames inside the video, no hands holding a phone.

Google Play: a YouTube link rather than an uploaded file, shown above your screenshots, and it does not autoplay the same way. Play's version is closer to an optional extra than a replacement for your first screenshot.

That difference matters: on iOS the decision is genuinely "video or your best screenshot first". On Play it is much closer to a free addition.

When a preview earns the slot

Motion is the product. Video editors, drawing apps, games, anything where the feel of an interaction is what you are selling. A still cannot show responsiveness.

The value is in a sequence. If understanding your app requires seeing three steps in order, video does that and eight disconnected panels do not.

The result is surprising. Before-and-after transformations — a cluttered photo becoming clean, a messy document becoming structured — land far harder in motion.

You have the skill or budget to do it well. This is the real gate.

When it costs you

Your app is mostly static. A note-taking app, a reader, a tracker. Video of someone typing is less informative than a well-designed still.

The video is mediocre. It replaces your best screenshot in the most valuable slot on the page. A bad preview is strictly worse than no preview, and this is the most common way teams lose conversion while thinking they improved the listing.

The poster frame is an afterthought. Most people see the poster frame rather than the video — it is what appears before playback and in contexts where autoplay is off. If your poster frame is a blurry mid-action moment, that is your first impression.

You cannot maintain it. A video showing a two-version-old UI is worse than screenshots showing the current one, and re-shooting video is much more work than re-rendering a panel.

Making one that works

1. Open on the outcome, not the launch. The first two seconds decide whether anyone watches the rest. Do not spend them on your splash screen.
2. Design for muted. Most views have no sound. Anything essential must be visible, not spoken.
3. Show one flow. Fifteen to thirty seconds is one task done well, not a feature tour.
4. Use real captures. Apple requires it, and staged footage looks staged.
5. Caption on screen if there is anything to say, since audio is not available.
6. Pick the poster frame on purpose.

The honest default

For most apps, most of the time: make the screenshots excellent first.

They are cheaper to produce, cheaper to maintain, cheaper to test, and they are what the majority of visitors actually look at. A strong set with no video beats a weak video in front of a mediocre set, every time.

Add a preview when you have a specific reason — motion is the product, or there is a sequence that needs showing — and when you can keep it current.

Do not guess — test it

This is one of the better uses of Product Page Optimization: run a treatment with your preview against one without it. It is a clean single-variable test and it settles an argument teams otherwise have forever.

On Play, Store Listing Experiments do the same job, though the lower stakes there — the video does not displace your first screenshot — make it a less urgent question.