Homepage hero image matched to buyer intent: +7.50% RPV
Test run byDave DiederenSyntra Labs
Commentary byLiftArchiveOur analysis, not theirsSwapping the homepage hero image for one closer to buyer intent lifted revenue per visitor by 7.50%. Dave Diederen ran this A/B test on a Shopify store and estimated the extra profit at $15,330 a month. The ranges were wide, so the result is promising but not certain.
- Revenue per visitor rose 7.50%, from $1.35 to $1.45, estimated at $27,289 a month.
- Conversion rate rose 3.92%, from 1.34% to 1.40%.
- Profit per visitor rose 5.84%, from $0.97 to $1.03, estimated at $15,330 a month.
- Average order value rose 3.45%, from $100.46 to $103.92.
- The second new image also edged out control ($1.43 RPV), but the ranges were wide (RPV -25% to +55%), so neither win is conclusive.
The change
Only the homepage hero image changes.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The homepage opened with the store's existing hero image.
Variant
The hero image was replaced with a new one chosen to match what buyers came for and to read more clearly. A second alternative image was also tested.
Why it likely won
Our reading of the result, not the practitioner's claim.
- the hero is the first thing visitors see, so an image that matches why they came keeps them moving.
- a clearer image helps people grasp what the store sells in a second.
- both new images beat the old one, which suggests the original hero was the weak point.
Caveats: The confidence ranges were very wide and include losses, and no sample size or duration was shown. The post does not show either image.
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 replace the homepage hero with an image that shows what our main buyers come to get, then revenue per visitor will rise, because visitors instantly see they are in the right place.
- Primary metric
- Revenue per visitor
- Guardrails
- Homepage bounce rate
- Average order value
- Click-through from hero
- Who should test it
- Stores that get a lot of homepage traffic and have a hero image chosen for looks rather than message. It matters less if most traffic lands directly on product pages.
Questions
Does the homepage hero image affect sales?
In this test a new hero image lifted revenue per visitor by 7.50%. The range was wide, so treat it as a strong hint rather than proof.
What makes a good homepage hero image?
Dave Diederen's winning image was chosen to match buyer intent and be visually clear. It should show what your typical buyer came for.
Should I test more than one hero image?
This test ran two new images against the original, and both did better. Testing several options helps you learn what kind of image works.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon@oliverkenyon@DaveDiederen@DaveDiederen