Affinity Clustering
Developed by Japanese anthropologist Jiro Kawakita (hence KJ Method), affinity clustering is the process of grouping a large set of ideas, observations, or data points into natural thematic clusters — driven by intuition and similarity rather than predefined categories. The result surfaces the underlying structure of complex information and is a core step in research synthesis, brainstorming, and sensemaking.
How to run it
- 1
Generate a large number of individual items — one idea, observation, or data point per sticky note.
- 2
Place all notes randomly on a large surface.
- 3
Silently (no talking), begin moving notes that 'belong together' into clusters. Everyone moves notes simultaneously.
- 4
When a note keeps being separated from a cluster, it may need its own cluster or a new one.
- 5
When movement stabilises, give each cluster a name that captures its essence.
- 6
Look at the cluster map: what patterns emerge? What's central? What's isolated?
- 7
Use the clusters to identify themes, priorities, or gaps.
Tips
The silence during clustering is important — it prevents verbal consensus from overriding genuine affinity.
If two people keep splitting a note, that's a signal the theme needs to be split.
Clusters of one are valid — don't force them to join a larger group.
Variations
Run virtually using digital sticky notes (Miro, FigJam) with simultaneous movement. Combine with dot voting on the cluster level to prioritise themes rather than individual items.
Where it fits
Plan your next workshop with AI
Workshop Weaver helps you combine methods like Affinity Clustering into a complete, timed agenda in minutes.
Try it freeMethod descriptions on Workshop Weaver are original content written by our team, based on established facilitation practices.