Every researcher knows the moment. Fieldwork is done, your head is full, and in front of you is the wall: eighty sticky notes, forty transcript highlights, a dozen half-legible observations, and a team waiting for "the findings". Affinity mapping is the bridge from that wall to something you can act on.
The idea behind affinity mapping is simple: write every observation on its own note, group the notes that belong together, and only then name the groups. Done properly, it is a fast, collaborative way to let themes rise out of the data rather than being imposed on it. Done badly, it is a room full of people sorting notes into the categories they walked in with. The difference is technique, and technique is what this guide covers.
What affinity mapping is
The method traces back to the KJ method, developed by the anthropologist Jiro Kawakita for making sense of field data. That pedigree is worth remembering: this was born as a serious analysis technique, not a workshop energiser.
Three rules do most of the work:
- One observation per note. Not "onboarding problems" but "P4 abandoned signup when asked for a phone number". Specific, atomic, and traceable back to a participant.
- Group bottom-up. Move notes next to the notes they seem to belong with, and resist arriving with ready-made buckets. The clusters should emerge from the data.
- Name the groups last. Labels come after the clusters have formed. A label written first becomes a magnet, attracting notes that do not really belong to it.
The output is a set of clusters, each with a name that says something ("people distrust the estimate because it once let them down", not "trust"), and a visible trail from every theme back to the observations underneath it.
From messy notes to themes, step by step
- Prepare the notes before the session. Observations, quotes, and behaviours, one per note, written so that a stranger could understand each one without you standing next to it.
- Sort in silence first. Everyone places and moves notes without talking for fifteen or twenty minutes. Clusters form from the data, not from the most persuasive voice in the room.
- Discuss and refine. Talk through the clusters together. Split the big blobs: a cluster of twenty-five notes is a topic, not a theme. Merge the overlaps.
- Name each cluster with a full sentence. If you cannot write the sentence, the cluster is not a finding yet.
- Mark the tensions. Notes that contradict a cluster are not noise; park them visibly, because they are often the most interesting finding on the wall.
- Step back and prioritise. Map the themes against the decisions the team actually faces, and agree what happens next.
Doing it collaboratively without groupthink
The strength of affinity mapping is that the team builds understanding together, which beats any report at changing minds. The risk is that the room converges on the loudest or most senior voice. A few guardrails keep the data in charge:
- Write before you read. Everyone produces their notes before seeing anyone else's, so first impressions are independent.
- Silence before discussion. The quiet sorting pass is the single best defence against anchoring.
- Let duplicates stand. Five people noticing the same thing is signal; merging those notes too early hides how strong it is.
- Make moves cheap. Anyone may move any note, and a note that keeps bouncing between two clusters is telling you the clusters are drawn wrong.
- Settle arguments with data. When people disagree about a note, go back to the transcript or recording, not to whoever argues best.
- Sequence the seniority. Have the most senior person share their reading of the wall last.
Physical or digital
A physical wall is still the better thinking environment: the scale is bodily, everyone stands and moves, and the limits of the space force prioritisation. Its weaknesses are practical: the room gets rebooked, remote colleagues are excluded, and the map dies as a photo in a slide deck.
Digital whiteboards win on logistics: distributed teams, saved state, searchable text, a map that stays alive after the workshop. Two cautions, though. The infinite canvas removes the useful pressure of limited space, so sprawl can quietly replace synthesis. And copy-paste makes it tempting to dump transcript chunks instead of writing atomic observations, which breaks rule one without anyone noticing.
Choose based on where your team sits, then import the discipline of the wall either way: bounded space, real observations, a silent first pass.
Affinity mapping is not thematic analysis
It is worth being precise about this. Affinity mapping is rapid, collaborative synthesis: a few hours, a team, shared understanding by the end of the day. Thematic analysis is slower and more systematic: full familiarisation with the data, careful coding, themes reviewed against the entire dataset, and an audit trail that survives scrutiny.
They sit on a spectrum rather than in opposition. The map excels when speed and shared ownership matter most: a sprint, a workshop, a first pass over fresh fieldwork. When the study is large, the stakes are high, or the findings will face hard questions, graduate from the map to full thematic analysis, and let the map serve as your head start on searching for themes rather than your final answer.
The wall of notes is intimidating exactly once. After that, you learn to trust the constraints: atomic notes, silence before discussion, labels last, disagreements settled by the data. Those constraints are the method. Everything else is just sticky notes.
People of Research is a community and newsletter for UX researchers, market researchers, and anyone who has ever stood in front of a wall of sticky notes looking for the story. Join us and get the good stuff in your inbox.