Affinity diagram
Definition
An affinity diagram (also called an affinity map or KJ diagram) is a technique for organizing a large collection of qualitative data — user research observations, brainstorming ideas, customer feedback — into naturally emerging groups or themes to surface patterns and insights.
The affinity diagram was created by Japanese anthropologist Jiro Kawakita in the 1960s and is sometimes called the KJ Method after his initials. It entered UX and design practice through the work of American researchers and has become a standard technique for synthesizing large volumes of qualitative research.
When to use an affinity diagram:
- After user interviews or usability tests — to synthesize observations across multiple sessions
- After brainstorming — to find natural clusters in a large set of ideas
- During retrospectives — to group feedback themes before prioritizing
- After surveying customer feedback — to identify recurring pain points or desires
How to build one:
- Capture: Write each individual observation, idea, or data point on a separate sticky note. One thought per note. Don't pre-sort as you write.
- Read aloud: Each person reads their notes to the group — no discussion yet.
- Sort: Working in silence, participants move related notes near each other. If two notes feel related, cluster them together. Anyone can move any note.
- Name clusters: Once the natural groups emerge, give each cluster a header note describing the theme.
- Discuss: Review the clusters together. Are they meaningful? Should any be merged or split?
Key rules:
- Sort in silence. Talking during sorting creates anchoring and slows the process.
- Anyone can move any note. Ownership of ideas is off the table.
- Emerge, don't impose. Let clusters form naturally — don't predetermine the categories.
Affinity diagrams and whiteboards: Physical affinity diagramming on a whiteboard wall with sticky notes is the classic format. The tactile, collaborative sorting process is one of its strengths. The result — a wall covered in clustered sticky notes — is exactly the kind of artifact BoardSnap AI was built to read and structure.
Examples
- UX team sorts 120 user interview observations into 8 clusters — three clusters dominate, pointing to the core design focus
- Sprint retrospective: 40 feedback items clustered into 'Communication', 'Process', and 'Tooling' groups
- After a two-day design workshop, the affinity diagram reveals that users talk about 'trust' three times more than any other theme
- Product team uses affinity diagram to organize 6 months of support tickets into priority problem areas
- Workshop wall covered in 200 sticky notes, sorted and photographed with BoardSnap — AI extracts the cluster names and top items from each
Snap a affinity diagram. Ship its actions.
BoardSnap turns any whiteboard — including this one — into a summary and action plan.