More collection page filters: -1.0% conversion, inconclusive
Test run byClean CommitCRO agency
Commentary byLiftArchiveOur analysis, not theirsReworking the collection page filters and adding more filter options changed conversion rate by -1.0%, which is within normal variation.
Clean Commit ran this test for an anonymized Shopify store and rated it inconclusive.
Revenue per visitor rose 8.9%, but that was not the metric the test was judged on.
- Conversion rate: -1.0% (primary metric). Inconclusive, with a 42.1% probability that the variant beats the original.
- Revenue per visitor: +8.9%, driven by higher order value. A secondary metric, so treat it as a lead, not a result.
- Average order value: +9.9%, while orders fell 2.0%.
- Devices split: desktop conversion rate rose 9.7% and mobile fell 5.1%. Neither segment is a proven result.
The change
The variant swaps the original filters for a more responsive set with more options.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The collection page used the store's original filters.
Variant
The filters were rebuilt to be more usable and responsive, with more options. Price range, customer rating, product category and availability were among the filters proposed.
Why it likely came out this way
Our reading of the result, not the practitioner's claim.
- Many shoppers on a collection page browse by scrolling, so better filters may only matter to a minority who use them.
- Extra filter options can add clutter on a small screen, which may explain the weaker mobile numbers.
- Shoppers who did filter may have narrowed in on pricier items, which would fit the higher order value without more orders.
Caveats: Based on 64,314 visitors with a 42.1% probability to beat the original, so the overall result is a coin flip. Test duration was not shared, and the device split is a sub-segment that needs its own test to confirm.
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 collection filters easier to use and add filters for price, rating and availability, then conversion rate will rise because shoppers find a suitable product faster.
- Primary metric
- Conversion rate
- Guardrails
- Revenue per visitor
- Mobile conversion rate
- Filter usage rate
- Who should test it
- Stores with large catalogs where shoppers need to narrow down many products. It probably won't matter for stores with small collections that fit on one or two screens.
Questions
Do more filters on a collection page increase conversions?
Not in this test. Conversion rate moved -1.0% with a 42.1% chance of beating the original, so there was no clear effect.
Why would filters help desktop but hurt mobile?
Filter panels take more space and more taps on a phone.
A layout that works well on desktop can feel cluttered on mobile, so test the two separately.
Which filters should an online store add first?
Start with what shoppers search for most, often price, size or availability.
Check your site search and filter usage data before adding more.
Similar tests
@oliverkenyon@oliverkenyon
ConvertiblesConvertiblesConvertiblesConvertibles