Video added to the paywall: about +80% revenue
Commentary byLiftArchiveOur analysis, not theirsFitnessAI added a video to its paywall and reported roughly 80% more revenue.
FitnessAI, an AI workout app, shared the result at App Promotion Summit, and it was retold on the RevenueCat blog. FitnessAI still uses a paywall video today.
- Revenue: up about 80%. This is the only number shared.
- Lasting change: FitnessAI kept the video on its paywall, which suggests the gain held up.
- Not shared: trial start rate, trial-to-paid, sample size, duration or whether it was a split test.
The change
A product video takes the place of static visuals at the top of the paywall.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The paywall had no video.
Variant
A video was added to the paywall. An earlier RevenueCat article describes it as a short, silent demo of the app in use.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Video shows what the app does faster than text or screenshots, which matters on a screen people decide on in seconds.
- Seeing the app work can lower the risk of starting a trial or paying.
- Motion draws the eye on an otherwise static screen, so more users may look at the offer before closing it.
Caveats: The result was shared at a conference and retold, and the test design was not stated. An earlier RevenueCat article says the video was added together with a move of the paywall before onboarding, so the lift may not come from the video alone.
Who ran it
FitnessAI ran this test. Source: RevenueCat blog. The numbers are theirs; the summary and commentary on this page are ours.
How to run this test on your app
- Hypothesis
- If we add a short product video to the paywall, then revenue per paywall view will rise because users understand the app's value faster.
- Primary metric
- Revenue per paywall view
- Guardrails
- Trial-to-paid conversion
- Paywall load time
- Refund rate
- Who should test it
- Apps whose value is easy to show visually, like fitness, photo, video or design apps. It may help less for apps whose value is hard to show in a few seconds, or where a heavy video slows the paywall on weak connections.
Questions
Does a paywall video need sound or captions?
FitnessAI's video, as described on the RevenueCat blog, had no sound or text.
A silent demo also works for users who browse with sound off.
Will a video slow down my paywall?
It can. Keep the file small, preload it, and track how often users leave before the paywall finishes loading.
How do I test a paywall video fairly?
Change only the video and keep placement, copy and prices the same.
Otherwise you can't tell what caused the result.
Similar tests
Kovalee