Affinity Mapping in UX: How to Turn Research Data Into Insights

Affinity mapping (also called affinity diagramming) is a synthesis method for organising large amounts of qualitative research data into emergent themes. After user interviews, usability testing sessions, or observational research, you are typically left with dozens or hundreds of individual observations, quotes, and findings. Affinity mapping gives you a structured way to make sense of that material -- moving from raw data to patterns, and from patterns to insights that can actually drive design decisions.

The method was developed by Japanese anthropologist Jiro Kawakita in the 1960s and is often called the KJ Method after his initials. Kawakita originally used it in field anthropology to synthesise large bodies of qualitative field data. UX and product design teams adopted it because it solves the same problem: too much raw material, not enough structure, and a need to find signal in the noise.

If you are new to the research process that feeds into affinity mapping, UX research methods covers the methods you will most commonly draw from.

Want to practise affinity mapping on a real brief? UX Academy (myuxacademy.com)'s Intermediate UX Design course covers synthesis and research methods with live tutor feedback -- you will run real sessions, not just read about them. Cohort 1 starts 5 Sep 2026. Or try the free masterclass first.

When Should You Use Affinity Mapping?

Affinity mapping belongs in the synthesis phase of the design process -- after research, before ideation. Concretely, that means after you have completed a round of user interviews, usability testing, or contextual enquiry, and before you start writing problem statements, creating user personas, or generating solutions.

It is most useful when you have data from multiple participants or sessions. If you have interviewed eight people, you will have somewhere between 80 and 200 individual observations to make sense of. Without a synthesis technique, you risk anchoring on the most memorable or most recent interview and ignoring quieter signals that cut across the full dataset.

You do not need affinity mapping for every research activity. A quick round of five-user usability testing on a specific flow might surface clear, actionable problems that do not require collaborative synthesis. Affinity mapping earns its place when you are dealing with exploratory, generative research -- discovery interviews, early-stage user research, or post-study synthesis where the findings are open-ended.

What You Need Before You Start

Before running an affinity mapping session, you need your research data captured at the right level of granularity. That means:

One observation per note. Each sticky note should contain a single, discrete piece of data -- a direct quote, an observed behaviour, a stated frustration, a moment of confusion. Notes like "generally found the interface confusing" are too vague to cluster meaningfully. Notes like "couldn't find the save button -- assumed progress would auto-save" are specific enough to reveal a real pattern.

Enough participants. Affinity mapping on data from two interviews will surface coincidences, not patterns. You typically need at least six to eight participants before clustering becomes genuinely revealing.

The right people in the room. The value of affinity mapping is partly in the collaborative sense-making -- having designers, researchers, and product people looking at the data together. That does not mean everyone needs to have been present for the interviews, but bringing in people who were not involved at all can mean the session becomes a debate about the data rather than a synthesis of it.

How to Run an Affinity Mapping Session

Step 1: Prepare the raw data

Transfer every observation from your research into individual sticky notes -- physical or digital. Colour-code by participant if it helps you track where data came from, but do not impose categories yet. Resist the urge to pre-sort. The whole point is to let themes emerge from the data, not from your assumptions about what the themes should be.

Step 2: Put everything on the wall

Post every note on a shared surface -- a whiteboard, a wall, a Miro board -- without any structure. The initial state should look like chaos. That is correct.

Step 3: Sort in silence (initially)

Ask participants to start moving notes that feel related close to each other -- working in silence. The silence is deliberate. It prevents one person's framing from dominating early and forces everyone to work from the data rather than from each other's interpretations.

After ten or fifteen minutes, open the floor and let people discuss groupings, move notes between clusters, and flag disagreements. Disagreements are often where the most interesting insights live.

Step 4: Name the clusters

Once groupings have stabilised, give each cluster a descriptive header. A good cluster header captures what the theme means, not just what it is. "Navigation" is a label. "Users cannot predict where settings will be, so they give up rather than explore" is a theme. The second version already points toward a design implication.

Step 5: Identify second-order patterns

Look at your named clusters. Are there clusters that belong together at a higher level of abstraction? You may find that several clusters all relate to trust, or onboarding anxiety, or a specific mental model mismatch. These second-order groupings often become the pillars of your synthesis -- the three or four big things you learned from the research.

Step 6: Move from clusters to insights

A cluster is not yet an insight. An insight is a statement about what the data means and why it matters for design. Take each cluster and ask: what does this tell us about users' goals, mental models, or frustrations? What design implication follows?

This is the step most teams skip or rush. The clusters feel like the output, but they are actually just organised data. The insight -- the statement that tells you what to do next -- requires interpretation.

Physical Sticky Notes vs Digital Tools

The original affinity mapping method uses physical sticky notes and a wall. There is something genuinely useful about the tactile, spatial quality of moving paper around -- it slows the process down in a way that prevents premature closure on themes.

For distributed teams, digital tools are the practical alternative. Miro and FigJam are the most commonly used platforms. Both support virtual sticky notes, real-time collaboration, and colour coding. Miro has more sophisticated organisation and navigation features; FigJam integrates naturally into a Figma-based design workflow. Either works.

A few practical tips for digital sessions: use colour to distinguish participants or data sources, not to pre-categorise themes (that defeats the purpose). Keep video on so you can see when someone is about to move a note. Use a time-boxed silent sorting phase even in digital sessions -- it is easy to lose this discipline when everyone is looking at the same screen.

Moving From Clusters to Insights

The synthesis step -- moving from organised clusters to usable design insights -- is where affinity mapping often stalls. Here is a framework that helps.

For each cluster, complete the sentence: "We learned that [type of user] [does/feels/needs X] because [underlying reason]." The "because" is the part that most teams omit. It forces you to explain the mechanism, not just describe the observation.

From that statement, ask: what does this mean for our design? Some clusters will point toward clear implications (a navigation pattern needs to change, a concept needs better onboarding). Others will surface questions that require further research. Both are legitimate outputs -- knowing what you do not yet understand is itself a research finding.

If you are using personas to communicate research findings, the insights from affinity mapping will become the raw material. User personas work best when they are grounded in synthesis like this, not assembled from demographic assumptions.

Affinity Mapping vs Thematic Analysis vs Journey Mapping

These three methods are often confused because they all involve organising qualitative research data. They serve different purposes.

Affinity mapping is a collaborative, bottom-up workshop technique. The team physically or digitally sorts data together in real time, letting themes emerge from the material. It is fast, participatory, and well-suited to design teams who need actionable insights within a day or a sprint.

Thematic analysis is a more formal qualitative research methodology involving systematic coding, reviewing, and defining of themes, typically done by one or two researchers working through interview transcripts methodically. It is more rigorous and auditable than affinity mapping -- appropriate for academic research or high-stakes contexts where the analysis needs to withstand external scrutiny. For most UX design teams, affinity mapping offers the right balance of structure and speed.

Journey mapping is a different activity altogether. Where affinity mapping organises findings by theme, a customer journey map organises them by the user's sequence of actions and experiences over time. Journey mapping is an output of synthesis, not a synthesis technique itself -- you might use affinity mapping to identify themes first, then use those themes to inform the emotional arc of a journey map.

Common Mistakes

Starting with categories. Defining your clusters before you sort the data turns affinity mapping into a filing exercise. You will find what you expect to find and miss what you did not know to look for.

Mixing data and interpretation. Sticky notes should contain data (what participants said or did), not your interpretation of what it means. Keep the interpretation for the cluster-naming and insight-generation steps.

One person driving the sort. If one person moves all the notes while others watch, you lose the collaborative sense-making that makes affinity mapping useful. Everyone should be moving notes.

Stopping at the clusters. Clusters are organised data. Insights are what you do with them. If your synthesis deliverable is a photograph of a clustered whiteboard, you have done half the work.

Skipping the "because." The most common failure mode in synthesis is describing what happened without explaining why. A finding that explains the underlying cause is ten times more useful to a design team than one that just names the problem.

Affinity Mapping in the Broader Design Process

Affinity mapping sits between research and definition in the UX design process. It is the bridge that connects what you observed to what you will design. Used well, it means your problem statements, personas, and design decisions are grounded in evidence rather than assumption.

For career-changers learning UX, affinity mapping is often one of the first synthesis techniques that makes the design process feel concrete. You go from a stack of interview notes that feel overwhelming to a clear picture of what users need and why. That transformation -- from raw data to structured insight -- is one of the core skills that separates designers who can do research from designers who can use research.

At UX Academy, our Intermediate UX Design course covers affinity mapping alongside the full research and synthesis toolkit, with hands-on practice on real client briefs. Tutor-led sessions with Natalia Veretenyk give you direct feedback on your synthesis -- not just on whether you ran the session correctly, but on whether your insights are strong enough to drive design decisions. Cohort 1 opens 5 Sep 2026. Register your place with a £99 deposit, or join the free masterclass to get a feel for how we teach.