Side-by-side bar comparison page, no discount: +10.25% revenue/visitor
Commentary byLiftArchiveOur analysis, not theirsA landing page that compared POSSIBLE's meal bar and snack bar side by side, with an "I want both" button and no discount, raised revenue per visitor by 10.25% and conversion rate by 9.91%.
VaynerCommerce ran the test for the health food brand, and it was published as a Replo case study.
The new page changed many things at once, including dropping a 20% discount, so the win belongs to the whole page.
- Revenue per visitor: +10.25% for the comparison page.
- Conversion rate: +9.91%, even though the winning page had no discount and the control offered 20% off.
- Average order value was not reported on its own. Sample size and statistical confidence were not shared.
The change
Separate product sections with a 20%-off timer vs a side-by-side comparison with an add-both button
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Version 1 had separate sections for the meal bar and snack bar, carousels of ingredients and reviews, a section promoting protein powder, and a banner countdown timer for 20% off.
Variant
Version 2 compared the two bars side by side and added an "I want both" button that put both in the cart, lifestyle images and videos, and the tagline "Never go hungry again". It offered no discount.
Why it likely won
Our reading of the result, not the practitioner's claim.
- Putting the two bars next to each other makes the choice easier than scrolling between separate sections.
- The "I want both" button turns indecision into a bigger order with one click.
- Dropping the protein powder section kept the page focused on the product the ad promoted.
- Lifestyle videos showed the bars in use, which may have mattered more to these shoppers than a countdown discount.
Caveats: The two pages differed in layout, copy, imagery, the add-both button and the discount, so no single change can be credited. The test ran about one month as a 50/50 URL split from the same ad, and no sample size was shared.
Who ran it
VaynerCommerce ran this test for POSSIBLE and published the results (Replo case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on VaynerCommerce →
How to run this test on your store
- Hypothesis
- If we show our two best-selling products side by side with a one-click option to buy both, then revenue per visitor will rise because shoppers can compare quickly and are nudged toward a bigger order.
- Primary metric
- Revenue per visitor
- Guardrails
- Conversion rate
- Average order value
- New customer share
- Who should test it
- Brands with two closely related hero products that shoppers often choose between. It is less useful with one hero product or a wide catalog. Test the discount removal separately if you want to know its effect.
Questions
Can removing a discount increase sales?
It can when the page explains the product better.
In this test the page without a discount won, but it changed many other things too, so the discount's own effect is unknown.
Do "buy both" buttons work?
They can raise order value when shoppers are torn between two similar products.
Watch conversion rate too, since a bigger default order can put some people off.
Similar tests
@oliverkenyon@oliverkenyon@oliverkenyon
Vasta