18 checkout usability fixes at Devold: +73% checkout conversion
Test run byConversionistaCRO agency
Commentary byLiftArchiveOur analysis, not theirsDevold's checkout conversion rate rose 73% after 18 small, research-led checkout fixes went live.
Conversionista ran the work for the Norwegian merino wool brand, working with Devold's tech partner Columbus.
This was a before/after comparison of a bundle of changes, not an A/B test.
- Checkout conversion rate: +73% in the first month after launch, compared with the same month a year earlier.
- Traffic in the later month was much lower, which the source notes. Lower traffic can raise conversion rate on its own if the visitors who remain are keener buyers.
- The 18 changes shipped together, so no single fix can be credited. The source does not list each change.
The change
Same checkout, with 18 small usability fixes across its steps
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Devold's existing checkout. Heuristic review and usability testing had found several pain points that were losing sales.
Variant
18 small design changes to the checkout, each aimed at a pain point found in research and each needing little development work. They shipped over six weeks in two sprints.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Usability testing pointed at real friction points, so fixes went where buyers were getting stuck.
- Many small fixes in one flow can add up, because each removes a reason to abandon.
- Part of the jump may come from the comparison itself: a year-on-year month with much lower traffic is not a controlled test.
Caveats: Before/after, not a split test: one month after launch versus the same month a year earlier, with much lower traffic. 18 changes were bundled, and the agency reported the result.
Who ran it
Conversionista ran this test for Devold and published the results (Conversionista case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Conversionista →
How to run this test on your store
- Hypothesis
- If we fix the checkout pain points found in usability testing, then more shoppers will complete checkout because fewer will hit confusing or broken steps.
- Primary metric
- Checkout conversion rate (orders / checkout starts)
- Guardrails
- Average order value
- Payment errors
- Customer service contacts about orders
- Who should test it
- Stores whose checkout has known usability problems and who can't wait for a full redesign. If traffic allows, roll fixes out as an A/B test or in stages so you can tell which ones help. A checkout already in good shape will see much less.
Questions
Can small checkout fixes really move conversion?
Yes, when they target friction buyers actually hit.
Usability tests and session recordings show where people get stuck, and fixing those spots often needs little development work.
Is a year-on-year comparison a reliable way to measure checkout changes?
Not very. Traffic mix, season, promotions and site speed all change between years.
An A/B test or a staged rollout gives a much cleaner read.
Should I ship checkout fixes one at a time or together?
Together is faster, but you won't know which fixes helped.
If you have the traffic, test the larger changes on their own and bundle only the obvious bug fixes.
DRIP Agency