The affinity diagram, also called the KJ method after its creator Jiro Kawakita, is a collaborative synthesis technique that transforms large volumes of raw qualitative data into organised, meaningful clusters of insight. After research fieldwork generates hundreds of observations, quotes, and data points, the affinity diagram provides a structured process for making sense of it all. It is one of the most commonly used analysis tools in design research precisely because it converts chaos into clarity without losing the richness of the original data.
What It Is
An affinity diagram involves writing individual observations, quotes, or data points on separate sticky notes and then clustering them into groups based on natural relationships. The process is done collaboratively, with the whole team participating rather than a single analyst working alone. This shared sense-making is one of the method's most important features: it builds a common understanding of the research findings across the team, which is as valuable as the diagram itself. Once clusters are formed, each is given a header note that captures the underlying theme or insight.
How to Run It
Print or transcribe all raw research data onto individual sticky notes, one observation or quote per note. Assemble the team around a large wall or whiteboard. Begin by asking each person to silently read and place notes on the wall, grouping related items together without discussion. Once all notes are placed, the team reviews the groupings together, moving notes, splitting clusters, and merging overlapping ones through quiet consensus. Header notes are then written to name each cluster. A second pass creates higher-level themes from the clusters if the dataset is large enough to warrant it.
When to Use It
Affinity diagramming is most useful immediately after a research phase, when the team needs to move from raw data to patterns and insights. It is particularly effective after a series of user interviews, contextual inquiry sessions, or diary studies where a large volume of qualitative data has been collected. It can also be used in co-design workshops to synthesise ideas generated by stakeholders or participants. Use it whenever you need a collaborative, visual method for finding structure in complexity.
Tips for Success
Write observations and quotes verbatim rather than already-interpreted summaries. The more direct the language, the richer the discussion during clustering. Avoid creating clusters around data collection methods or participant demographics: group by theme and meaning instead. Leave ambiguous notes in a separate parking area rather than forcing them into a category where they do not quite fit. Photograph the final diagram before it comes down and use it as the foundation for the insights and opportunity statements that will drive the next phase of the project.
