Star ratings on collection cards: -10% revenue per visitor
Commentary byLiftArchiveOur analysis, not theirsAdding star ratings to collection page product cards cut revenue per visitor by 10% at Gym King, a UK streetwear brand.
Gym King's in-house team ran the test, which was published as a Shoplift case study.
They stopped it after six days, when the variant was about £10,000 behind.
- Revenue per visitor: -10% for the variant. This was a clear loss.
- Running cost: about £10,000 in lost revenue before the test was stopped at six days.
- The ratings were removed and the original cards kept.
The change
The variant adds star ratings to each collection product card.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Collection pages showed product cards with no star ratings.
Variant
Each product card on the collection page showed a star rating. Only 60-70% of products had reviews, so many cards showed an empty rating.
Why it likely lost
Our reading of the result, not the practitioner's claim.
- With 30-40% of products unrated, empty or missing stars may have looked like a warning sign rather than proof.
- Shoppers may have skipped good unrated products in favour of the few with ratings.
- Gym King has many returning customers who already know the brand, so ratings added little for them.
- Extra elements on each card add visual clutter to a grid that was built around product images.
Caveats: The test was stopped early after six days, and no sample size or confidence level was shared. The result is tied to patchy review coverage and may differ on a catalog where every product has reviews.
Who ran it
Gym King ran this test for Gym King and published the results (Shoplift case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Gym King →
How to run this test on your store
- Hypothesis
- If we show star ratings on collection page product cards, then revenue per visitor will rise because shoppers can spot well-liked products faster.
- Primary metric
- Revenue per visitor
- Guardrails
- Collection page click-through to product pages
- Add-to-cart rate
- Share of clicks going to unrated products
- Who should test it
- Stores where nearly every product has a solid number of reviews. If a third of your catalog has no reviews, fix coverage first or hide ratings on unrated cards, as empty stars can hurt.
Questions
Should I show star ratings on collection pages?
Often yes, but only when most products have reviews.
Gym King lost 10% revenue per visitor when many cards showed empty ratings.
What should I do with products that have no reviews?
Hide the rating on those cards instead of showing zero stars.
An empty rating can read as a sign that nobody buys the product.
When should I stop a losing A/B test early?
Set a loss limit before launch. Gym King stopped this test at six days once it was about £10,000 behind.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon