Card sorting shows how customers group a messy list of concepts and needs. Run a study, watch the themes emerge, and read the result even on a large deck.
Type your concepts, needs, features or benefits. Run an open sort, a closed sort against your own categories, or a hybrid of the two.
Each person groups the cards in a way that makes sense to them, and names their groups. On a phone they get a sort built for the small screen.
See the themes on a similarity matrix, a dendrogram and a 3D cluster view. Pick the number of groups and Graham names them.
Most of your participants are on a phone, so the sort is designed for the phone from the first tap. They flip through the cards, file each into a group with a tap, and name the groups as they go. It is quick, needs no account, and that is why more people finish.
Where other tools stop at a tangle, SortedResearch keeps a 60 card study legible. A similarity matrix shows which cards were grouped together, a dendrogram shows the structure, and a 3D view lets you explore it. Pick the number of groups and the cards reorganize.
Pick the number of groups and Graham reads what participants created, then names and describes each one, so a long deck becomes a clean set of themes you can act on.
A card sort is qualitative: it gives you the themes in customers' own words. One click turns those themes into a pairwise study, a Best–Worst study or a Findability study, and now they are quantitative. Graham carries the group names and descriptions across, so nothing is retyped and nothing is lost in the handover.
Grouping and finding are not the same test. People can agree on where something belongs and still walk past it in a live menu, because the label that made sense on a card has to compete with everything around it. Build a Findability study straight from a finished card sort, give people the structure as words alone, and watch where they actually go looking.
Card sorting asks people to group a list of items into piles that make sense to them, and then name those piles. It shows you the structure customers already carry in their heads rather than the one your team assumed. In consumer research it is used on concepts, needs, benefits and features, not just website menus.
In an open sort, participants create and name their own groups, which is how you discover a structure. In a closed sort you supply the categories and they file the cards into them, which is how you validate a structure you already have. A hybrid sort gives them your categories and lets them add their own. SortedResearch runs all three.
About 15 participants gives a result that correlates around 0.90 with what a much larger group would tell you, according to Nielsen Norman Group guidance. At 20 to 30 that rises to roughly 0.93 to 0.98 and then flattens out (Tullis and Wood, 2004). So 15 is a defensible floor for a directional read and 30 is the practical ceiling for most studies.
Yes. The sorting flow is built for a phone screen first, which is where most panel respondents are. That is a large part of why completion rates hold up on long decks.
Decks of 60 cards stay readable for participants and stay legible in the analysis. The similarity matrix, the dendrogram and the 3D cluster view are all built to hold their shape at that size, which is where many tools start to blur.
You get a similarity matrix, a dendrogram and 3D clusters, and Graham suggests names for the groups people created. From there one click carries those groups into a pairwise, Best-Worst or Findability study, so you go straight from structure to priority without retyping anything.
Set up a card sort today. Graham will help you get going.
Start your free study