Affinity mapping

What it is

“A collaborative sorting exercise”: Teams group qualitative data (e.g., user insights, ideas, pain points) into themes based on natural relationships to reveal patterns.

How to use

Checklist:

  1. Sticky notes (100+), markers, large wall/whiteboard.
  2. Timer (visible to all).
  3. Camera (to document the final map).
  4. Scripts: Clear instructions: "Group similar items. No talking during sorting!"

Pre-work:

  • Gather raw data (e.g., user quotes from interviews, survey responses).

Time:

  • Prep: 1-2 hours (compile data).
  • Sorting: 45-90 mins (with 5-8 participants).
  • Analysis: 1-3 hours (theme refinement).

Participants:

  • Ideal: 5-8 (mix of researchers, designers, stakeholders).
  • Minimum: 3. Avoid >10 to prevent chaos.

Steps:

  1. Prep:
    • Write each data point (e.g., user quote, observation) on a sticky note.
  2. Silent Sorting:
    • Stick all the notes randomly on a wall.
    • Participants group related notes without speaking.
    • Set a timer (45-60 mins).
  3. Theme Naming:
    • Discuss groupings. Label each cluster with a descriptive header (e.g., "Checkout Frustrations").
  4. Refine:
    • Merge overlapping groups. Discuss outliers.
    • Identify high-level insights (e.g., "Users need simpler payment options").
  5. Synthesise:
    • Identify high-level insights (e.g., "Users need simpler payment options").

Tips & Variations

  • Remote? Use digital tools (Miro, FigJam, Trello).
  • Large datasets? Split into rounds: First sort broadly, then subgroup.
  • Stuck? Ask: "Would users group these the same way?"
  • Avoid:
    • Over-rationalising during silent sorting.
    • Letting hierarchies influence groupings.

Why this method

Pros:

  • Uncovers hidden patterns in messy qualitative data.
  • Democratises analysis (non-researchers contribute meaningfully).
  • Creates shared understanding across teams.

Cons:

  • Time-intensive with large datasets.
  • Subjective groupings if participants lack context.
  • Risk of surface-level themes without deep discussion.

Find out more

Back to

Next Step