How to Script an App Preview Video That Converts

Why most app previews do not convert

Most app previews are built as feature tours. They open with the app logo, show a list of screens, and end with the app icon. They are technically compliant and visually acceptable — and they do almost nothing for conversion.

The problem is structure. Without a deliberate narrative arc, even well-produced footage fails to move users from interest to install.

The Conversion Narrative Framework

Every preview we produce at W. App Videos is built around a four-part structure called the Conversion Narrative Framework. It works because it mirrors how users actually make install decisions — not by evaluating features, but by feeling confident that this app will solve their problem.

Part 1 — Hook (0–3 seconds)

The hook has one job: stop the scroll and create immediate relevance.

In practice, this means opening with either the problem your app solves or the outcome your app delivers — in the clearest, most specific terms possible. Not "the best productivity app." Something like: a user in a recognizable situation, the app opening, the friction disappearing instantly.

Three seconds is not much time. But it is enough to create a moment of recognition: this is for me.

Part 2 — User Benefits (3–18 seconds)

The middle section is where you show the app working. But the goal is not to show everything — it is to show the right things in the right order.

Choose two or three user outcomes, not features. The difference: a feature is "real-time sync across devices." The outcome is "open the app on your phone and everything is already there." Show the outcome. The feature is implicit.

Each benefit should be demonstrated in no more than four or five seconds. If a screen requires more than that to understand, it is probably not the right screen for a preview.

Part 3 — Future Positive State (18–24 seconds)

This is the most underused part of the structure. The goal is to show the user what life looks like after they use the app — not just what the app does, but what becomes possible.

For a fitness app, this might be a completed workout log, a streak reached, a goal hit. For a finance app, it might be a clear dashboard with everything organized. For a note-taking tool, it might be an inbox at zero.

This section does not have to be long — six seconds is enough. But it shifts the video from "here is the app" to "here is who you become."

Part 4 — Call to Action (24–30 seconds)

The closing frames should reinforce the core promise one more time, clearly and simply. This is where many previews end abruptly or fade out on a logo. Instead, close with the app's clearest value statement and let the last frame be in-app footage — not a title card.

Voiceover vs text overlays

Both work. The decision depends on your app and your audience.

Voiceover is effective when the benefit needs explaining — when watching the app without audio context leaves users uncertain about what they are seeing. It adds personality and can guide the eye.

Text overlays are effective when the app is visually self-explanatory and the overlay just needs to name what is happening. They work well in competitive categories where users mute videos by default.

A common approach is a light voiceover combined with on-screen text for key moments — redundancy that works in both muted and unmuted states.

Practical scripting tips

• Write the script before you record. Trying to find a narrative in existing footage is much harder than recording to a plan.
• Time your script aloud. Read it at the pace you would want a voiceover to deliver it. Adjust until it fits 28 seconds comfortably.
• Choose your single strongest user outcome and build everything around it. Sub-narratives weaken the core message.
• Assume the user will watch muted. Make sure the footage communicates value even without audio or text.
• Design the opening frame for a thumbnail. The first two seconds will autoplay in search at a small size. If that composition does not read on its own, rework it before doing anything else.
• Cut ruthlessly. Every shot needs to justify its seconds. If you can remove a shot without losing the story, remove it.
• Review the script at 1.5x speed. If it still makes sense at that pace, pacing is right. If it feels rushed, the script is too dense.

A worked example: scripting a habit-tracking app

To make the framework concrete, here is the kind of script we would write for a generic habit-tracking app:

• 0–3s (Hook): Close-up of a calendar with a broken streak. Hand taps phone, app opens, streak is restored. On-screen text: "Never lose your streak again."
• 3–8s (Benefit 1 — Effortless logging): User taps three habits in under two seconds. Visual feedback confirms each. Text overlay: "Log in seconds."
• 8–14s (Benefit 2 — Visual progress): Progress ring fills. Calendar grid lights up showing consistency. Text overlay: "See your progress build."
• 14–18s (Benefit 3 — Smart reminders): Quick sequence showing the app surfacing a reminder at the right moment. Text overlay: "Nudges when you need them."
• 18–24s (Future Positive State): Calendar now full of green checkmarks. 30-day streak. User smiles. Text overlay: "Become the person who follows through."
• 24–30s (CTA): App icon centered. Tagline: "Start your streak today." Fade on the streak counter, not the logo.

Notice what is not in the script: the onboarding flow, the settings screen, premium upsells, a feature list, an app architecture overview. None of that belongs in a 30-second preview. The script exists to surface the user's future self and make it feel reachable. Everything else goes in the app.

Common scripting mistakes that kill conversion

After reviewing hundreds of preview scripts, these are the failure modes we see most often:

• Starting with the logo or app open screen. The first two seconds are the most valuable real estate in any preview. Burning them on branding tells the viewer nothing and gives them a reason to scroll away. Open in the product, not in a title card.
• Showing the onboarding flow. Nobody is excited to watch someone sign up. Skip past the friction and open where the value lives.
• Listing features without context. "Real-time sync. Offline mode. Cross-platform. AI-powered." Features stacked without a narrative read as noise. Every feature the preview shows needs a "so what?" answer framed around the user.
• Over-relying on voiceover. If the script only makes sense with narration, the 60%+ of viewers watching on mute get nothing. The visual story must stand on its own.
• Writing for the team, not the user. Scripts that emphasize how proud you are of a feature you just built, rather than why the user should care, never convert. Test every sentence against "is this about the user's life or about our roadmap?"
• Cramming the CTA into a static title card. The closing frame should remind viewers of the core benefit in product context, not abruptly cut to a plain-text "Download now."
• No throughline. Five separate mini-stories instead of one coherent arc. Viewers cannot hold multiple narratives in 30 seconds — they need one.

How to test if your script works before production

You can validate a script before spending a cent on production:

1. Read it aloud, on a timer, at normal speed. It must fit within 28 seconds with a pause at the end. If it does not, cut.
2. Mute-test your storyboard. Sketch or stand-in-frame the key shots. Show them to someone cold without audio. Ask: "What is this app for?" If they cannot tell you in one sentence, the visual story is not strong enough.
3. Run the 3-second test. Show only the first three seconds to five people who have never heard of your app. If fewer than four can guess the category, the hook is failing.
4. Check the benefit claims. Every claim in the script should be demonstrable with real app footage. If you have to stage it, soften it, or simulate it, either the claim is too aspirational or the app does not yet support it.

This validation costs nothing and catches most script problems before they get baked into expensive video. Skip it and you pay twice.

Adapting the framework for different app categories

The Conversion Narrative Framework is structure, not a strict template. The shape changes slightly depending on what you sell:

• Productivity and utilities: Lead hard with the before/after. The hook is friction, the benefit section is the workflow compressed, and the Future Positive State is a clear workspace or completed task list.
• Games: The hook is usually gameplay, not a problem. Benefits become mechanics. Future Positive State is progression — a character leveled up, a map revealed, a leaderboard position. Keep the framework; replace the emotional beats with game-specific ones.
• Health and fitness: Lean into identity. Hook with a relatable moment, benefits show activity logging and progress, Future Positive State is the healthier self. Emotional arc matters more here than in any other category.
• Finance: Clarity and calm beat excitement. Hook with "where does my money go?", benefits show categorization and insights, Future Positive State is organized, controlled, understood. Slow pacing works if it feels reassuring.
• Social and dating: Hook with a real user moment, benefits show connection, Future Positive State is the human outcome — a conversation, a match, a friend added. People matter more than UI here.
• Creative tools: Hook with raw input to polished output. Benefits are the workflow. Future Positive State is the finished artifact being shared. Let the craft speak.

FAQ

Should the script be 30 seconds or 15 seconds?

Apple allows 15–30 seconds. 30 is the standard because it fits the full four-part structure. A well-written 15-second cut can work for very simple apps, but you usually sacrifice the Future Positive State section to get there. For most apps, 28–30 seconds is the right target.

Do I need a voiceover?

No. Plenty of top-converting previews have no voiceover and rely entirely on visuals plus text overlays. If the app is visually self-explanatory, skip narration. We dig into this tradeoff in can you use voiceover in an App Store preview.

How many benefits should I show?

Two or three. One is usually too light, four or more turns into feature soup. Pick the outcomes your audience cares about most and leave the rest for the product page copy and screenshots.

Should I show my app icon in the preview?

The app icon already appears above the preview on the product page. You do not need to show it inside the video. If you use it anywhere, use it only in the final frame — and only if it reinforces the story, not as a default close.

Can I use text overlays with voiceover?

Yes, and it is often the strongest approach. Redundancy between audio and on-screen text means viewers understand the preview whether they are muted or not. Just keep overlays short — two to five words per screen.

How do I know if my script is working?

Ship it, wait two weeks, check App Store Connect conversion rate. If it moved up, the script is working. If it is flat or down, iterate. The framework is a starting point — your specific audience tells you what lands.

The practical takeaway

A great preview script is not about being clever. It is about being clear. The Conversion Narrative Framework works because it forces you to answer the three questions every potential installer is silently asking: Is this for me? Can I trust that it will work? What do I become if I install it?

Write your script to answer those three questions in that order, with real app footage and a single confident throughline, and you will produce a preview that lifts conversion without gimmicks.