Localized upsell paywalls: +50% upsell screen conversion
Commentary byLiftArchiveOur analysis, not theirsLocalizing its upsell paywalls lifted upsell screen conversion by 50% for Photoroom, an AI photo editing app.
Photoroom ran the tests using Superwall, and the result was published as a RevenueCat customer story.
- Upsell screen conversion rose 50%.
- The number of variants and rounds behind the figure was not stated.
- Which languages or markets were included was not shared, nor whether prices also changed.
The change
Upsell paywall copy is localized for each market.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Upsell screens were not localized for each market.
Variant
Upsell paywalls were localized for users' markets, then tested and refined over several rounds.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Users trust and understand a paywall more in their own language.
- Local wording can make benefits feel relevant instead of translated.
- The figure reflects several iterations, so part of the lift may come from later design tweaks.
Caveats: The story describes testing and iterating, so the 50% may be the result of several rounds rather than one clean split test. It was published by RevenueCat, which integrates with Superwall, and sample size and duration were not shared.
Who ran it
Photoroom ran this test. Source: RevenueCat customer story. The numbers are theirs; the summary and commentary on this page are ours.
How to run this test on your app
- Hypothesis
- If we localize the upsell paywall copy for each major market, then more users will upgrade because the offer is clearer and feels made for them.
- Primary metric
- Upsell screen conversion by market
- Guardrails
- Refund rate by market
- Trial-to-paid
- Proceeds per user
- Who should test it
- Apps with a large share of users outside their home language market. It matters less if nearly all users read the app's main language.
Questions
Is translation enough, or do I need local pricing too?
The story only says paywalls were localized. Whether prices changed was not stated, so test copy and pricing separately if you can.
Which markets should I localize first?
Start with markets that send a lot of paywall views but convert below your average.
That is where clearer copy has the most room to help.