App Demo Videos: The 90-Second Version of Your Onboarding

Introduction

An app demo video does a job no other asset can do: it shows a stranger what using your app feels like before they commit to installing it.

That’s a narrower job than it sounds, and getting it wrong is expensive in a specific way. App install is a low-friction decision followed by a high-friction one — downloading takes ten seconds, actually opening the app and getting somewhere useful takes real effort. Most apps lose the majority of users in the first session. A demo video that oversells the app makes that worse, because it sets expectations the first session doesn’t meet.

Gisteo has produced 3,000+ video projects since 2011, including app demos for companies at every stage. This guide covers what an app demo video should actually contain, how it differs from a desktop product demo, the app store constraints that shape it, and the three versions most apps need.

An App Demo Is Onboarding, Shown Early

The most useful way to think about an app demo video: it’s your onboarding flow, compressed and moved earlier.

Your actual onboarding happens after install — account creation, permissions, setup, the first meaningful action. The demo video runs that same sequence in ninety seconds, before install, for someone deciding whether it’s worth their time.

That framing settles most content questions. What goes in the video is whatever gets a user from nothing to first value in your real app — the same path, minus the friction, plus the payoff shown clearly. What stays out is everything that doesn’t sit on that path: the settings, the edge cases, the features that matter in month three.

It also explains the most common app demo failure. Videos that showcase feature breadth — look at all fourteen things this app does — sell an app that doesn’t exist for a first-time user. Their first session touches maybe two of those fourteen. If the video promised the other twelve, the first session feels thin by comparison.

If you’re building the post-install sequence too, our guide to video for customer onboarding covers how the two connect.

How App Demos Differ From Desktop Product Demos

They’re often treated as the same deliverable. They aren’t:

The two constraints that matter most: **vertical and muted**. A demo produced horizontally with a voiceover carrying the explanation will fail in both places it most needs to work — the app store listing and the paid social feed. Design for vertical and silent first, then produce the horizontal and narrated variants from that, not the other way around.

What Belongs in the 90 Seconds

A structure that works for most apps:

1. The situation, in one beat (0–5s). Not your logo. The moment the app exists to fix, shown or stated fast enough that the right person recognizes themselves. In a feed, this is the whole ballgame.
2. The core action (5–35s). The single most important thing a user does in your app, performed at realistic speed on real screens. If your app has one killer interaction, this is where it lives.
3. The payoff (35–55s). What the user gets. Show the finished state — the completed workout, the sent invoice, the organized inbox. Ending on a settings screen is the app demo equivalent of ending on a form.
4. One supporting capability (55–75s). At most one. The thing that makes someone think “oh, it does that too” without implying they need to learn fourteen features on day one.
5. The ask (75–90s). Download, start free, try it — one action, stated plainly, with the app store badges or button visible.

Four things to leave out, in descending order of how often they show up anyway:

• Onboarding friction. Account creation, permission prompts, email verification. Nobody was ever persuaded to install by watching someone sign up.
• The logo intro. A ten-second animated logo reveal is ten seconds of your ninety spent on the least persuasive content available.
• Features that don’t exist yet. If it isn’t in the app yet, showing it converts installs into disappointed first sessions.
• Impossible speed. Real apps have loading states, empty states, and taps that take a moment. A demo that’s impossibly smooth reads as fake, and viewers are good at spotting it.

App Store Requirements Shape the Whole Video

Both major stores accept preview video, and both impose constraints that should inform production from the start rather than being handled in the edit.

Broadly: Apple’s App Store previews are short — in the region of 15 to 30 seconds each, with multiple allowed per listing — and Apple requires the footage to be captured from the app itself rather than animated or heavily stylized. Google Play is more permissive on both length and content, hosting the video via YouTube and allowing more produced material.

Verify current specifications against Apple’s and Google’s own developer documentation before production — both change these periodically, and the details matter enough that guessing is expensive.

Two production implications regardless of the exact numbers:

• Design the short version first. If Apple’s preview is capped near 30 seconds, your 90-second video has to contain a coherent 30-second version inside it. Plan that cut during scripting, not after the master is locked.
• Real app footage may be mandatory. A demo that’s entirely animation may satisfy Google Play and fail Apple’s requirements. Hybrid approaches — real screen capture with animated framing, callouts, and transitions — tend to satisfy both while still looking produced.

Screen Capture, Animation, or Both

Three approaches, each with a real use case:

Screen capture

Actual footage from the actual app. Most credible, mandatory in some store contexts, and the fastest to produce. The weakness: it ages the moment your UI changes, and small screens shot literally can be hard to read at thumbnail size.

Animation

Stylized recreations of the interface, or fully illustrated sequences. Better for explaining concepts, showing things that are hard to film (data syncing, matching, background processes), and holding up when the UI shifts. The weakness: it isn’t proof, and store rules may not accept it.

Hybrid — usually the right answer

Real screen capture as the substance, with animated or AI framing, motion callouts, kinetic type, and device mockups doing the explaining around it. You get credibility from the real footage and clarity from the animation, and updating a UI change means recapturing one segment rather than rebuilding the video.

For apps specifically, the hybrid also solves the legibility problem: a real screen is often too dense to read in a vertical thumbnail, and animated callouts can zoom, highlight, and isolate the part that matters.

The Three Versions Every App Needs

Producing these together from one shoot and one script costs a fraction of commissioning them separately later — and it prevents the common outcome where a company has one horizontal 2-minute demo and nothing that works in the two places installs actually come from.

Measuring an App Demo Video

That last row is the one most teams skip and the one that matters most. An app demo that lifts installs but lowers day-one retention is doing damage — it’s attracting people whose expectations the app doesn’t meet. Installs are the visible metric; expectation accuracy is the valuable one.

If retention drops after a new demo video ships, the video is overselling. That’s a content fix, not a targeting fix.

FAQs: App Demo Videos

What is an app demo video?

A short video showing what using an app is actually like, before someone installs it. It compresses the first-session experience — setup, the core action, and the payoff — into roughly 60 to 90 seconds, and needs to work vertically, on mute, at thumbnail size, and within app store constraints.

How long should an app demo video be?

It depends on placement, which is why most apps need three versions: 15–30 seconds for app store previews, 60–90 seconds for a website or landing page, and 6–15 seconds for paid acquisition creative. Produce them together from one script — the short cuts should be planned during scripting rather than salvaged from a finished master.

What should an app demo video include?

The situation in one beat, the core action performed on real screens, the payoff shown clearly, at most one supporting capability, and a single plain ask. Leave out account creation and permission prompts, animated logo intros, features that don’t exist yet, and impossibly fast interactions that read as fake.

How is an app demo video different from a product demo video?

App demos are vertical-native, watched at thumbnail size and usually on mute, and they support a low-friction install decision rather than a considered purchase. Desktop product demos are horizontal, viewed larger, tolerate two to four minutes, and handle denser interfaces. The same asset rarely serves both jobs well.

Do app store preview videos have special requirements?

Yes, and they differ by platform. Apple’s App Store previews are short and generally require footage captured from the app itself rather than animation; Google Play is more permissive on length and content and hosts video via YouTube. Both change their specifications periodically, so verify against current developer documentation before production.

Should an app demo use screen recording or animation?

Usually a hybrid: real screen capture as the substance, with animated framing, callouts, and kinetic type doing the explaining around it. That satisfies store requirements where real footage is mandatory, keeps small dense screens legible at thumbnail size, and means a UI change requires recapturing one segment rather than rebuilding everything.

How do you measure whether an app demo video is working?

Store conversion rate for the listing version, install clicks for the website version, and cost per install plus thumbstop rate for paid creative. The most important measure is post-install: day-one retention and first-session completion for users who saw the video. A demo that lifts installs but lowers retention is overselling the app.

Can Gisteo produce app demo videos?

Yes. Gisteo has produced 3,000+ video projects since 2011, including app demos delivered as a set — store preview, website version, and paid cuts — from a single production, in hybrid screen-capture and animation formats. See our app demo video production page or request a free consultation.

Conclusion: Show the First Session, Not the Feature List

The best app demo videos are honest previews of the first ninety seconds a user will actually spend in the app. They show the path from nothing to something, at realistic speed, and they stop there.

The temptation is always to show more — the full feature set, the roadmap, the power-user workflows. But an install is a decision about whether the first session is worth trying, and every promise beyond that first session is a promise the first session has to keep.

If you’re planning an app demo and aren’t sure which ninety seconds to show, request a free consultation. Working out what the core action actually is tends to be the useful part of the conversation.