Explaining the app at checkout in 3 steps: +10% progression
Commentary byLiftArchiveOur analysis, not theirsAdding visuals, explaining the app's benefits and breaking the download into three steps lifted progression from the app phone-number page to checkout by about 10%.
Plymouth Rock Assurance ran this server-side A/B test on its car insurance purchase flow, and it was published as a VWO case study.
Copy, visuals and step layout changed together, so no single change can take the credit.
- Progression from the app phone-number page to the checkout page: about +10%.
- The test ran server-side for two weeks. Sample size and confidence were not shared.
- The team's main goal was more app downloads after purchase, but the download rate itself was not reported.
The change
A bare phone-number ask gains visuals, app benefits and a three-step guide.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The final checkout step asked customers for a mobile number to get a link to the YUBI driving app, with little explanation of why it mattered.
Variant
The step added visuals, explained what the YUBI app does and why it helps, and split the download into three actions: give a mobile number, grant app permissions, and avoid policy cancellation.
Why it likely won
Our reading of the result, not the practitioner's claim.
- People share a phone number more readily when they understand what they get in return.
- Three numbered steps make an unfamiliar task feel manageable.
- Mentioning policy cancellation makes the stakes clear, which may push people to act.
- context matters most at the end of a long purchase flow, when patience is low.
Caveats: Bundled test: visuals, copy and step layout changed together. The ~10% figure is a page-to-page step, not completed policies or app installs, and sample size was not shared.
Who ran it
Plymouth Rock Assurance ran this test and published the results (VWO case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Plymouth Rock Assurance →
How to run this test on your site
- Hypothesis
- If we explain why a required post-purchase step matters and break it into a few clear actions, then more customers will complete it because they understand the benefit and the effort.
- Primary metric
- Progression from the step to the next page
- Guardrails
- App install or activation rate
- Policy purchase completion
- Support contacts about the step
- Who should test it
- Insurers, fintechs and SaaS products that ask for an app download, phone number or setup task during sign-up or checkout. It may matter less if the step is optional and rarely used.
Questions
How do I get more customers to download a companion app?
Explain the benefit in plain words at the moment you ask, and show the steps.
Plymouth Rock saw about 10% more people move past this step after doing so.
Should I break a task into numbered steps?
Usually yes for unfamiliar tasks. Numbered steps show the effort up front and reduce worry about what comes next.
Why run a checkout test server-side?
Server-side tests render the same page for each visitor without flicker, which helps on dynamic, data-heavy flows such as insurance quotes.
Online Dialogue