The Secrets of Card Sorting: A UX Designer's Toolkit
What is card sorting, and why do UX designers use it?
Imagine stepping into your kitchen, teeming with an array of ingredients—some familiar, others a bit more mysterious. Your mission is to create the most user-friendly and wonderful dish. This is akin to the essence of card sorting in user experience (UX) design. Hand someone a stack of index cards, each one labelled with a piece of website content, and ask them to group them however makes sense to them. That's card sorting, and it's one of the simplest, most revealing exercises in a UX designer's toolkit.
It works a bit like sorting a spice rack. Left to guess, most people would put the cinnamon somewhere different to the next person. Card sorting is how we find out where your users actually expect things to live, rather than guessing on their behalf.
Why it matters
Most websites don't start as a blank page, they start as years of accumulated content, added page by page without much thought for how it all fits together. That's fine until a visitor can't find what they came for. Card sorting exists to solve that problem before it becomes an expensive one: it replaces internal assumptions about site structure with real evidence of how users think.
How the process actually works
It starts with a content audit, a proper review of what already exists on a site, what's still relevant and what needs to carry through to a new one. Each topic gets distilled down onto its own card.
Then we bring in participants, a mix of real users and stakeholders, because the two groups rarely categorise things the same way and both perspectives matter. Each person works through the cards and groups them however feels natural to them. There are two ways to run this: an open sort, where participants create and name their own categories from scratch, or a closed sort, where we give them a fixed set of categories to sort into. Open sorts tend to surface how people naturally think about a site; closed sorts are more useful once you already have a rough structure and want to test whether it holds up.
A session can take anywhere from a couple of hours to a full day depending on how much content there is and how many people are involved. It can be run physically, with real cards on a table, or digitally through online tools. Physical sessions tend to generate richer discussion, since people talk through their reasoning as they go. Digital tools make it easier to run sessions remotely and to collect data across a larger group.
Making sense of the results
Once the sorting is done, we ask participants to walk through their thinking, which is often where the most useful insight comes from, not the groupings themselves. The results get analysed for patterns, and from there we build a structure that reflects how users actually expect to find things, rather than how the business happens to be organised internally.
That structure feeds straight into navigation, content layout and, on larger projects, the site architecture as a whole. It's part of what we call the research phase of a project, sitting alongside tools like tree testing and click testing, all aimed at replacing guesswork with something we can actually point to.
Why it's worth the time
Card sorting takes longer than skipping straight to wireframes, and it's tempting to skip it under time pressure. But for any site with more than a handful of pages, particularly ones with complex offerings or multiple audiences, it's often the difference between a structure that makes sense to the business and one that makes sense to the people actually using it. Good design starts with understanding how people think, and card sorting is one of the clearest ways to find that out before a single screen gets designed.
