When shopping for shoes online, selecting the right size and colour should be seamless. In practice a clunky interface costs you the sale twice: once when the customer gives up, and again when the ones who persevere order the wrong size and send it back.
The task was to redesign and rebuild the size and colour selection tool so it worked properly on both mobile and desktop.
What was actually going wrong
Analysis and customer feedback pointed at four distinct problems, not one.
- Size confusion. Customers struggled to pick the correct size, which led to abandoned purchases and expensive returns.
- Regional sizing. Some shoppers thought in EU sizes, others in US sizes. The system had to serve both without making either feel like the afterthought.
- Stock visibility. It was hard to tell which sizes and colours were actually available, so customers hunted through combinations that were never going to work.
- The interface itself. The existing controls were awkward, particularly on a phone, where most of the traffic was.
The rebuild
The new interface prioritised clarity over cleverness. Customers could switch between any number of size formats rather than a fixed pair, so nobody had to convert in their head. Availability was shown directly against each size and colour, which removed the guesswork and the wasted clicks. The layout was designed for the phone first and then allowed to expand, rather than the reverse. Discounts appeared alongside the size and colour choice instead of surfacing later in the basket.
None of that is a novel idea in isolation. The value was in treating all four problems as one decision the customer is trying to make, rather than four features to be added separately.
What changed
Over the three months after launch, add-to-basket actions rose by 18% against the previous year’s average for the same period. Customers and suppliers both commented on how much easier the new selection process was.
The wider lesson is the one I keep returning to. The interface was never really about size and colour. It was about whether somebody could answer a simple question — will this fit, can I have it, what does it cost — without being made to work for the answer.