Onboarding rebuilt on Jobs-to-be-Done research: +20% ARPU
Commentary byLiftArchiveOur analysis, not theirsA redesigned onboarding flow built on Jobs-to-be-Done research lifted ARPU by 20% for Five Minute Journal (p = 0.04).
Applica, a growth agency, ran the A/B test and wrote it up on the RevenueCat blog.
Several changes shipped together, so no single change gets the credit.
- ARPU: +20%, with p = 0.04.
- Trial starts: about +10%.
- Trial conversions: about +14%.
- All three metrics moved up together, so the gain was not just more low-quality trials.
The change
Feature-led screens become goal questions, early testimonials and benefit-led copy.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Five Minute Journal's older onboarding flow, which had not been updated in a while and lost users along the way.
Variant
A new flow shaped by research into why people use the app. It tailored content to the user's goals, showed testimonials early and framed benefits from the user's side instead of listing features.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Framing the app around the user's own goal, such as building a daily habit, makes the trial feel relevant.
- Testimonials early in the flow may build trust before the paywall.
- Showing outcomes instead of features helps users picture the benefit. These effects are bundled, so we can't rank them.
Caveats: Bundled: personalisation, social proof and new copy shipped together. Sample size and duration were not shared, and it is an agency case study on a vendor blog.
Who ran it
Applica ran this test for Five Minute Journal. 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 rebuild onboarding around the jobs users hire our app to do, with testimonials and benefit-led copy, then ARPU will rise, because users see how the app solves their problem before the paywall.
- Primary metric
- ARPU per new user
- Guardrails
- Onboarding completion rate
- Trial-to-paid conversion
- Day-1 retention
- Who should test it
- Habit and wellness apps with an outdated onboarding and room for user research. Expect two to three months of work, so it fits less well if you need a quick win.
Questions
What is Jobs-to-be-Done in onboarding?
It means asking what goal users hire the app for, then building screens around that goal instead of the feature list.
How long does an onboarding redesign like this take?
Applica says research, design and build can take two to three months.
Which part of the redesign drove the lift?
We can't tell. The changes were tested together, so test them one at a time if you need to know.