A/B Test5 min read
Shopify Collection Page A/B Test: How Product Cards Lifted Revenue per Visitor by 28.2%
Product cards that show the saving, machine fit and key features lifted revenue per visitor by 28.2% for a European sleep therapy store.
Last updated: October 22, 2026
Last updated: October 22, 2026
TL;DR
Choosing sleep therapy equipment online is harder than it looks. Many models share a similar shape, similar names, and similar prices, and the collection page is where shoppers decide which products deserve a closer look.
For a European online store selling sleep therapy devices, masks and supplies, that first comparison was difficult to make. The product cards showed a photo, a rating, a name and a price, and little else to help shoppers tell one mask from another.
As part of our Shopify A/B testing work for the store, we redesigned the product cards to show what makes each product different, whether it fits the shopper’s machine and how much they save. Over 19 days, the new cards lifted revenue per visitor by 28.2% and conversion rate by 20.6%.
The challenge: similar-looking products and nothing on the card to separate them
For sleep apnea equipment, the buying decision is practical. Shoppers want to know how one mask differs from the one next to it, whether it will work with the machine they already use, and whether the price is a good deal.
The old cards answered none of these questions. Sibling masks had near-identical names and photos, so the only way to compare them was to open each product page one by one.
Discounts were easy to miss: a generic “Sale” label did not say how much a shopper saved, and the sale price looked very similar to the full price.
The research: shoppers were leaving to compare products elsewhere
Our Shopify CRO research showed that the biggest loss happened before the cart.
Customer voice research explained why:
- Products were hard to tell apart: sibling masks had near-identical names and photos, and shoppers left to research the difference elsewhere.
- Compatibility was the top question: “Will it fit my machine?” came up 13 times in support chats.
- Discounts did not read as discounts: sale and full prices used similar colours and font weights.
Our insight: the collection page had to do the comparison work that shoppers were doing on their own, across several product pages and support chats.
The hypothesis: put the decision on the card
If each product card shows what makes the product different, whether it fits the shopper’s machine and how much they save, shoppers will choose faster and more of them will buy, increasing revenue per visitor.
We kept the scope to the product card itself rather than redesigning the whole collection page.
The variation: product cards that answer the shopper’s three main questions
The product photo, name and price stayed in place. Around them, we added:
- A discount badge on the image showing the saving in %, instead of a generic “Sale” label.
- The sale price in red next to the struck-through full price.
- A “Compatible with all machines” line with an info tooltip.
- Up to four feature tags per product, such as “Ultra compact”, “Under the nose design” or “For side sleepers”.
- The rating score out of 5, placed on the product image.
Each change supports a decision the shopper has to make:
- Feature tags let them compare masks without opening several product pages.
- The compatibility line answers the most common pre-purchase question before it reaches support.
- The percentage badge and red price make the saving visible at a glance.
Because all five elements changed together, the test measured their combined effect and does not isolate any single one.
The results: +28.2% revenue per visitor in 19 days
We ran the test in Intelligems for 19 days, comparing the original cards with the redesigned version.
Most of the gain came from more shoppers buying: conversion rate rose by 20.6%, while average order value grew by a more modest 6.3%. We use revenue per visitor as the primary metric because it captures both effects in one number.
After the test, we implemented the winning cards on the store’s live collection pages.
What this test taught us about product cards
The lesson wasn’t that every Shopify store needs feature tags on its product cards. It was that when products look alike, the card has to carry the comparison.
For this store’s customers, that meant three things visible before the click:
- what makes this product different from the one next to it
- whether it works with the machine they already own
- how much they save
What could A/B testing uncover in your store?
We plan, build and run A/B tests for Shopify stores, grounded in customer research rather than guesswork. Get in touch to find out what to test next.
Let us take your business further than it has ever been.
All 8 steps as a printable worksheet — 74 checkpoints, scoring template included.
Share this article
