Prominent, guided search with suggestions: -12.0% conversion rate
Test run byClean CommitCRO agency
Commentary byLiftArchiveOur analysis, not theirsMaking the search bar more prominent and opening it with examples, product previews and suggested collections lowered conversion rate by 12.0% and orders by 11.3%.
Clean Commit ran this 22-day test for an anonymized Shopify store and published it in its A/B test library.
The drop was not statistically significant, and a rise in order value kept revenue per visitor close to flat.
- Conversion rate: -12.0%. The original had an 83.4% probability of performing better.
- Orders: -11.3%.
- Average order value: +11.5%, which kept revenue per visitor nearly flat at -1.6%.
- Add-to-cart rate rose 3.8%, but the extra adds did not turn into orders.
- Flag: not statistically significant after 22 days.
The change
The variant highlights search and opens it with examples, previews, collections and a banner.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The store used its original, standard search bar.
Variant
The search bar was highlighted and opened with search examples, product cards, suggested collections and a 'Customize Your Own' banner linking to that collection.
Why it likely lost
Our reading of the result, not the practitioner's claim.
- A search panel full of suggestions can feel busy and pull shoppers into browsing instead of buying.
- The 'Customize Your Own' banner may have sent people toward a slower, higher-priced path. That fits the rise in order value and fall in orders.
- Several changes shipped in one variant, so it is hard to say which part caused the drop.
Caveats: Not statistically significant after 22 days, and sample size was not shared. Several search changes were tested together as one variant.
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 search more prominent and show suggestions when it opens, then conversion will rise because shoppers reach high-intent products faster.
- Primary metric
- Conversion rate
- Guardrails
- Revenue per visitor
- Average order value
- Search exit rate
- Who should test it
- Stores with large catalogs where many visitors use search. Test each addition on its own, since a crowded search panel can distract shoppers.
Questions
Does making search more prominent increase conversions?
Not always. In this test a highlighted, guided search lowered conversion rate by 12.0%, though not significantly.
Why did order value rise while orders fell?
The source does not say. One reading is that the custom-product banner drew some shoppers to pricier items while others dropped off.
What should I test first when improving search?
Change one thing at a time, such as suggested searches or product previews.
That way you can tell what helps and what hurts.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon
Conversion FanaticsClean Commit