One page cut from registration: +117% free trials, +130% revenue
Test run byMarketingExperimentsResearch publisher
Commentary byLiftArchiveOur analysis, not theirsRemoving one whole page from Encyclopaedia Britannica Online's subscription registration path lifted conversion to free trials by 117% and revenue by 130%.
MarketingExperiments (MECLABS) ran the test as part of a two-year program with Britannica and published it on MarketingExperiments.
- Conversion to free trials: +117%.
- Paid subscriptions: +61%, so the extra trials did turn into paying customers.
- Revenue: +130%.
The change
Variant removes one whole page from the registration path
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The subscription registration path with its original number of pages.
Variant
The same registration path with one whole page taken out. The source does not say which page was removed.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Every extra page is another chance to drop out, so cutting one removes a point of loss.
- A shorter path gets people into the product sooner, while they are still motivated.
- Paid subscriptions rose too, which suggests the removed page was not doing useful qualifying work.
Caveats: The test dates from 2008, and the source does not say which page was removed. Sample size and duration were not shared.
Who ran it
MarketingExperiments ran this test and published the results (MarketingExperiments). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on MarketingExperiments →
How to run this test on your site
- Hypothesis
- If we remove or merge a step in our sign-up flow, then more visitors will start a trial because there are fewer points to drop out.
- Primary metric
- Free trial starts per visitor
- Guardrails
- Trial-to-paid conversion
- Revenue per visitor
- Data completeness for onboarding
- Who should test it
- SaaS and subscription businesses with multi-step sign-up flows. It may not help if your flow is already one or two steps, or if the step you cut collects data you truly need.
Questions
How many steps should a SaaS sign-up have?
As few as you can manage while collecting what you need to start.
Each page you add is another place people can quit.
Will a shorter sign-up bring in lower-quality users?
Not in this test. Paid subscriptions rose 61% alongside the jump in trials. Still, track trial-to-paid when you test it.