One Bestseller badge instead of four: +8.89% revenue per visitor
Test run byDave DiederenSyntra Labs
Commentary byLiftArchiveOur analysis, not theirsShowing a "Bestseller" badge on only the single top-selling product lifted revenue per visitor by 8.89% and conversion rate by 6.03%. Dave Diederen ran this as a follow-up after an earlier version, with badges on three to four products, lost 4.76% revenue per visitor. Same idea, one badge instead of several.
- Revenue per visitor: +8.89% ($2.06 to $2.24) for the single badge, with a 95.1% chance to beat the control.
- Conversion rate: +6.03% (3.69% to 3.91%), and order value rose 2.70% ($55.76 to $57.26).
- Profit per visitor: +9.25% ($1.32 to $1.45), estimated at about $96,856 extra revenue per month.
- Negative first version: badges on three to four products cut conversion rate by 4.72% and revenue per visitor by 4.76%, with order value flat.
The change
A Bestseller badge on just the single top-selling product card.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The collection page showed products with no Bestseller badges.
Variant
The first version put a Bestseller badge on three to four top products and lost. The winning version put the badge on only the actual number one seller.
Why it likely won
Our reading of the result, not the practitioner's claim.
- when several products all claim to be the bestseller, the label stops meaning anything.
- one badge points shoppers to a single safe choice, which can make deciding easier.
- a rare badge reads as earned, so it works as a trust signal rather than decoration.
Caveats: The two versions had different baselines (3.69% vs 5.32% control conversion), so they likely ran at different times and are not a direct comparison. Ranges for the winner crossed zero slightly (-1% to +20% revenue), and sample size and test length were not shared.
Who ran it
Dave Diederen (@DaveDiederen) shared this test on X. The numbers and screenshots are theirs.
View the original post on X
How to run this test on your store
- Hypothesis
- If we put a Bestseller badge on only the single top-selling product, then revenue per visitor will rise because the badge keeps its meaning and guides shoppers to a safe pick.
- Primary metric
- Revenue per visitor
- Guardrails
- Conversion rate
- Average order value
- Sales mix across other products
- Who should test it
- Stores with a clear number one product on a busy collection page. It may not help if sales are spread evenly across many products, or if the badge goes on a product that is not really the top seller.
Questions
How many products should have a bestseller badge?
In these tests, badges on three to four products lost 4.76% revenue per visitor, while a badge on just one product won 8.89%. Fewer badges kept the label meaningful.
Can bestseller badges hurt conversion?
Yes. The version with several badges cut conversion rate by 4.72%.
Is it worth retesting a losing idea?
This is a good example. Changing how the badge was used turned a loss into a clear win.
Similar tests
@oliverkenyon@oliverkenyon@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen