Checkout-first minicart without recommendations: -9.9% conversion
Test run byClean CommitCRO agency
Commentary byLiftArchiveOur analysis, not theirsRedesigning the minicart around a high-contrast checkout button, while removing recommendations and chat, cut conversion rate by 9.9% and revenue per visitor by 18.2%.
Clean Commit ran this test for an anonymized Shopify store and published it in its A/B test library.
Several changes shipped together, so the loss can't be pinned on one of them.
- Conversion rate: -9.9%, a negative result. The original had a 77.6% probability of being better.
- Revenue per visitor: -18.2%, and orders -8.8%.
- Average order value: -13.2%, which may reflect the missing recommendations.
- Click-through rate: +8.5%. More shoppers left the minicart for the next step, but fewer completed an order.
The change
Checkout button moved up and restyled; recommendations and chat removed.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The minicart showed the cart items, a product recommendations widget and a chat widget alongside the checkout button.
Variant
A high-contrast checkout button with a lock icon moved to the most prominent spot. The recommendations and chat widgets were removed and spacing was adjusted.
Why it likely lost
Our reading of the result, not the practitioner's claim.
- Removing recommendations took away a source of extra items, which likely cut order value.
- Some shoppers may have used chat to settle doubts before buying.
- A louder checkout button may have pushed people forward before they were ready, so more clicks did not become more orders.
Caveats: Bundled test that ran only 7 days, and sample size was not shared. Conversion fell on both mobile and desktop, but the size of the drop is uncertain.
Who ran it
Clean Commit ran this test for a Shopify store (anonymized) and published the results (Clean Commit A/B test library). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Clean Commit →
How to run this test on your store
- Hypothesis
- If we make the minicart checkout button more prominent and remove side widgets, then more shoppers will complete orders because there are fewer distractions.
- Primary metric
- Conversion rate
- Guardrails
- Average order value
- Revenue per visitor
- Who should test it
- Stores whose minicart widgets get little use. If shoppers add recommended items or use chat, test each removal on its own instead of all at once.
Questions
Should I remove product recommendations from my cart drawer?
Be careful. In this test, removing them alongside other changes came with a 13.2% drop in average order value.
Does a bigger checkout button increase sales?
Not on its own here. Clicks rose 8.5%, but completed orders fell, so judge cart changes on orders and revenue, not clicks.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon
ConvertiblesConvertibles