Insight

One Badge Treatment, Opposite Device Results

Stone Creek Coffee wanted shoppers to identify new, popular and limited-release products more quickly on collection pages. Its initial blue-badge test increased mobile conversion but reduced desktop performance, raising a practical CRO question: was the device split caused by the badge treatment or by blue competing with the brand’s purchase-path colour? The client approved a retest using charcoal. Our CRO audit had shown that mobile generated 69% of visits while desktop converted more strongly, so an all-device rollout carried real risk. The retest needed to preserve the existing Sold Out and Sale badge logic while isolating colour. This approach reflects Blend’s CRO process: turn a plausible design concern into a controlled experiment and use device-level evidence to decide where a change belongs.

Jump to results

A/B Test

Charcoal Product Badges on Collection Pages

We hypothesised that changing the New Arrival, Best Seller and Limited Release badges from blue to charcoal would reduce competition with purchase CTAs while preserving faster product recognition. Using Intelligems, we ran a Shopify A/B test across desktop and mobile collection pages. The control showed the current experience without the new charcoal badges. Variant 1 displayed the three badges in charcoal, while Sold Out and Sale treatments remained unchanged. Conversion rate was the primary metric, supported by revenue per visitor, orders, subscription performance, average order value, product-page views and add-to-cart rate.

The Results

Variant 1 produced a clear device split. On mobile, revenue increased 4%, conversion rate 4%, revenue per visitor 7%, subscription revenue per visitor 6% and subscription orders per visitor 22%. Average order value rose 3% and add-to-cart rate 1%, while product-page view rate fell 4%. Desktop moved sharply in the opposite direction, pulling the blended result negative: conversion rate fell 32%, revenue per visitor 31%, subscription revenue per visitor 43%, subscription orders per visitor 57% and add-to-cart rate 22%. Because a previous badge design had shown the same mobile-positive, desktop-negative pattern, the repeated result points to device behaviour rather than badge colour as the stronger factor. Mobile shoppers appeared to benefit from the extra product information, while the same treatment likely added visual distraction on larger screens. No sample size or confidence figure was supplied, so the uplifts should be treated as directional. The client approved the mobile-only recommendation, and the charcoal badges were made permanent on mobile while desktop retained the control experience.

  • 22%

    increase in mobile subscription orders per visitor

  • 7%

    increase in mobile revenue per visitor

  • 6%

    increase in mobile subscription revenue per visitor
Book a CRO discovery call

See More A/B Tests