Login paywall: a medium amount of explanation beat the minimal version
Commentary byLiftArchiveOur analysis, not theirsOn its login paywall, Alma Media found that a moderately informative message won over both a minimal version and the most detailed version.
The team at Alma Media, a Finnish digital media company, ran this test itself, and it was published as a VWO case study.
No numbers were shared, and the team had expected the minimal version to win.
- The moderately informative paywall got the most logins and registrations, so it was declared the winner.
- The minimal version, which the team predicted would win, did not.
- No uplift figures, sample size, duration or confidence levels were published.
The change
The winning paywall adds a moderate amount of explanation to the minimal version.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The login or registration paywall showed a minimal explanation of why readers should log in or register.
Variant
Two alternatives were tested: a moderately informative version and a most detailed version. The moderate one won.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Too little text may leave readers unsure why they are being asked to log in or what they gain from registering.
- Too much text can slow people down and make the step feel heavier than it is.
- a paywall works best when it gives just enough context to answer the obvious questions, then gets out of the way.
Caveats: No numbers were published, so the size of the win is unknown. The test was on a news subscription paywall, which may not match other sign-up flows.
Who ran it
Alma Media 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 Alma Media →
How to run this test on your site
- Hypothesis
- If we add a short explanation of what users get by logging in or registering to our minimal gate, then more users will complete it because they understand the value without being slowed down by long copy.
- Primary metric
- Login and registration completions at the gate
- Guardrails
- Bounce or exit rate at the gate
- Paid subscription starts after registration
- Return visits from registered users
- Who should test it
- Publishers, SaaS products and membership sites with a registration wall or login gate. If your gate already explains the value clearly and converts well, gains will likely be small.
Questions
Is shorter copy always better on a login or sign-up wall?
No. In Alma Media's test the minimal version lost to one with a bit more explanation.
Readers needed some context before they would log in or register.
How much copy should a registration wall have?
Enough to say what the user gets and why the step is needed, and no more.
Testing two or three lengths, as Alma Media did, is a good way to find the balance.