Live search with category links: +13.7% conversion, not significant
Test run byClean CommitCRO agency
Commentary byLiftArchiveOur analysis, not theirsAdding live predictive search with clickable category links and product previews came with a 13.7% rise in conversion rate and a 14.7% rise in revenue per visitor.
Clean Commit ran this test for an anonymized Shopify apparel store and published it in its A/B test library.
The metric the test was built to move, click-through rate, rose only 1.2% and fell short of significance, so we treat the result as inconclusive.
- Click-through rate, the primary metric: +1.2%, with a 93.0% probability of beating the original. Not significant.
- Conversion rate: +13.7%, with gains on both mobile and desktop.
- Revenue per visitor: +14.7%, and orders: +11.3%.
- Add-to-cart rate: +5.4%.
The change
Live search adds category links and product previews as shoppers type.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The store's original search, without category suggestions or live product previews.
Variant
Search updated as shoppers typed, showing matching category links above up to six product previews with image, name, price and availability. It opened as a dropdown on desktop and full screen on mobile.
Why it likely came out this way
Our reading of the result, not the practitioner's claim.
- Category links give shoppers a quick path to a whole collection when a broad term like 'jackets' is typed.
- Live previews with price and availability let people judge products without opening each one.
- A full-screen layout on mobile gives search results room to be read and tapped.
Caveats: The test ran for 15 days with 53,622 visitors but did not reach statistical significance. The conversion gains were not the metric the test was built to measure.
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 add live search with category links and product previews, then click-through rate from search will rise because shoppers find the right products and collections faster.
- Primary metric
- Conversion rate of visitors who use search
- Guardrails
- Revenue per visitor
- Site speed on mobile
- Who should test it
- Stores with large catalogs and many category pages, where search is a common route to products. Small catalogs that fit on one page will see little difference.
Questions
Does predictive search increase conversions?
It can help. In this test, live search with category links came with a 13.7% conversion lift, but the result was not significant.
Should search results include category links?
Worth testing for stores with many collections. A category link helps shoppers who type a broad term like 'jackets' see the full range.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon
Conversion FanaticsClean Commit