methodatlas
Method guideUX Research

Affinity Diagramming

A practical method guide: clarify prerequisites and materials, understand the complete process, and look up outputs, pitfalls, and stop criteria.

ComplexityLow
Time45–90 min
Participants3-10
FormatWorkshop
MaturityCanonical
Plan the work
01

Prerequisite

What needs to be finished first

Complete firstConcrete frame

Situation, purpose, and sources are named.

Without a frame, the artifact is arbitrary.
02

Preparation

What needs to be ready before start

Materials

Individual sourced notes, a free surface, cluster and outlier markers, and versioned snapshots.

People / roles

Facilitation, research or subject experts, and multiple perspectives; actively balance dominance.

Pre-read

Prepare the question, complete raw material, source markers, and clustering boundaries. Split compound statements before the session.

Time needed

45–90 minutes for up to about 30 notes; split larger sets into rounds.

Setup

One movable note contains one observation or statement. Clusters first emerge from relationships in the material; preset categories stay off the workspace.

03

Core question

The one question this method answers

What patterns emerge from individual notes, which outliers matter, and what question follows?

04

Flow

Phases with goal, actions, output and pitfall

Affinity Diagramming groups individual observations by shared relationships. Named clusters, visible outliers, and sources make the emerging structure traceable and revisable in review.

Affinity work matrix

Fictional teaching example

Sourced notes, inductive clusters, and open questions.

Affinity work matrix
Work rowSource or caseObservation or statementProvisional patternOpen question or follow-up action
Working noteOne note with context.F1: calendar noteF1: Saturday slot missingAvailability before travelQuestion open
Cluster reviewTest the label and record counterexamples and gaps.F2: fictional interview noteBorrows spontaneously without booking; counterexample to F1Access needs planned and spontaneous pathsWhen does walk-in borrowing suffice, and when is booking confirmation needed?

  1. 1. Source or case

    Working note
    F1: calendar note
    Cluster review
    F2: fictional interview note

  2. 2. Observation or statement

    Working note
    F1: Saturday slot missing
    Cluster review
    Borrows spontaneously without booking; counterexample to F1

  3. 3. Provisional pattern

    Working note
    Availability before travel
    Cluster review
    Access needs planned and spontaneous paths

  4. 4. Open question or follow-up action

    Working note
    Question open
    Cluster review
    When does walk-in borrowing suffice, and when is booking confirmation needed?

Keep classifications traceable and revisable.

NN/G

  1. 1

    Prepare material

    Prepare material

    Goal

    Each note can be moved and checked independently.

    Inputs for this step

    • Observation set and source scope

    Actions

    • Split combined statements into sourced, movable notes.

    Output

    Note · context · source

    Evidence, context, and source

    Quality check

    Can each note stand on its own?

    Pitfall

    Preset categories steer the grouping. Start with the notes and discover relationships.

    Timing: 10–20 minutes

  2. 2

    Cluster inductively

    Cluster inductively

    Goal

    Clusters emerge from the material.

    Inputs for this step

    • Movable sourced notes

    Actions

    • Group by observed relationship; keep outliers visible.

    Output

    Notes · provisional cluster · outlier

    Evidence, context, and source

    Quality check

    Are clusters grounded in note relationships?

    Pitfall

    A cluster hides differences between notes. Split mixed clusters and keep outliers visible.

    Timing: 15–30 minutes

  3. 3

    Review clusters

    Review clusters

    Goal

    Labels fit the notes and remain revisable.

    Inputs for this step

    • Provisional clusters and counterexamples

    Actions

    • Review labels, split mixed clusters, and record counterexamples.

    Output

    Label · supporting notes · counterexample

    Evidence, context, and source

    Quality check

    Do labels survive a counterexample review?

    Pitfall

    A label reflects one note only. Review the label against every note and counterexample.

    Timing: 10–20 minutes

  4. 4

    Derive questions

    Derive questions

    Goal

    The synthesis leads to explicit next questions.

    Inputs for this step

    • Reviewed clusters and open questions

    Actions

    • Document patterns and gaps; plan prioritisation separately.

    Output

    Pattern · source · open question · action

    Evidence, context, and source

    Quality check

    Is the next question traceable to the cluster?

    Pitfall

    Cluster size is treated as priority. Plan prioritisation with explicit criteria afterwards.

    Timing: 10–20 minutes

05

Artifact

What comes out at the end

Sketch: what the artifact looks like

Affinity Diagramming · Synthese-WandAus Rohdaten entstehen Cluster, Insight Statements und nächste Opportunity Areas
Affinity Diagramming Synthese-WandDie Skizze zeigt, wie einzelne Notizen zu Clustern gruppiert und daraus Insights sowie Opportunity Areas abgeleitet werden.Von Notizen zu handlungsfähigen InsightsRohmaterial wird gruppiert, benannt und in Opportunity Areas übersetzt.RohdatenZitate, Beobachtungen, TicketsClusterÄhnliches still gruppierenInsightsMuster in Handlung übersetzenZitatTicketBeob.PainIdeeFrageFindabilityFehlermeldungenTime-to-ValueInsight StatementCluster als Aussage formulierenOpportunity AreaProblemraum benennenTop-3 nächste SchritteOwner und Folgeaktion
Form

Affinity diagram with raw notes, inductive clusters, labels, outliers, and open questions.

Versioning / ownership

Preserve raw notes and save a dated snapshot after each clustering round. Keep renamed and dissolved clusters traceable.

Tool alternatives
  • Paper cards on a large work surface
  • Digital whiteboard without preset columns
  • Source list and versioned snapshots

affinity-diagramming-working-template.md

Turn source material into movable notes and prepare an inductive clustering round.

Affinity Diagramming: Worksheet

Question

To be confirmed

Desired outcome

To be confirmed

Scope

Create one affinity diagram Time: 45–90 minutes

Complete this worksheet on paper or in your own document during the work.

Preparation for this scope

Method setup

  • Materials: Individual sourced notes, a free surface, cluster and outlier markers, and versioned snapshots.
  • Roles: Facilitation, research or subject experts, and multiple perspectives; actively balance dominance.
  • Advance information: Prepare the question, complete raw material, source markers, and clustering boundaries. Split compound statements before the session.
  • Overall time needed: 45–90 minutes for up to about 30 notes; split larger sets into rounds.
  • Setup: One movable note contains one observation or statement. Clusters first emerge from relationships in the material; preset categories stay off the workspace.

Prepare for the work steps

Prepare material: Prepare material

  • Observation set and source scope

Cluster inductively: Cluster inductively

  • Movable sourced notes

Review clusters: Review clusters

  • Provisional clusters and counterexamples

Derive questions: Derive questions

  • Reviewed clusters and open questions

Affinity work matrix

Work rowSource or caseObservation or statementProvisional patternOpen question or follow-up action
Working note
Cluster review
  • Working note: One note with context.
  • Cluster review: Test the label and record counterexamples and gaps.

Blank worksheet: enter your own material.

NN/G

Prepare material: Prepare material

Expected artifact: Note · context · source

Evidence, context, and source

Entry:

...

  • Can each note stand on its own?

Cluster inductively: Cluster inductively

Expected artifact: Notes · provisional cluster · outlier

Evidence, context, and source

Entry:

...

  • Are clusters grounded in note relationships?

Review clusters: Review clusters

Expected artifact: Label · supporting notes · counterexample

Evidence, context, and source

Entry:

...

  • Do labels survive a counterexample review?

Derive questions: Derive questions

Expected artifact: Pattern · source · open question · action

Evidence, context, and source

Entry:

...

  • Is the next question traceable to the cluster?

Open questions and next steps

...

Method guide: https://methodatlas.meierhoff-systems.de/en/methods/affinity-diagramming/run-sheet

06

Example output

Concrete filled scenario, fictional example

Fictional example for orientation. Counts and timings describe this case and are not universal requirements.

affinity-diagramming-beispiel.md

Concrete filled scenario, fictional example

Fictional teaching example: tool lending

All details are invented and are not research. Four notes are grouped without preset themes.

  • F1: “The Saturday slot was in the calendar but missing at pickup.”
  • F2: “I had to call before travelling.”
  • F3: “The return time was only in the small print.”
  • F4: “The repair needed an attachment that was unavailable.”

Inductive clusters

  • Availability must be confirmed before travel: F1, F2
  • Return information is hard to find: F3
  • Suitable accessories remain unclear: F4 as an outlier

The outlier stays visible. A next question is: What information do households need before pickup and return? Frequency is untested.

07

Pitfalls

Recognize symptoms and steer against them

Quick overview, the full text is in the matching phase in the flow

  1. Prepare materialPreset categories steer the grouping.View in phase
  2. Cluster inductivelyA cluster hides differences between notes.View in phase
  3. Review clustersA label reflects one note only.View in phase
  4. Derive questionsCluster size is treated as priority.View in phase
08

Stop criteria

When to pause, resolve prerequisites, or choose another approach

Start without categories and group from the material.
Keep deviations visible and explain treatment.
Prioritize later with explicit criteria.

Where to go next?

Paper illustration of shared options, independently placed voting dots, and a highlighted shortlist.
Facilitation
Dot Voting

Condenses the resulting clusters into a prioritized shortlist.

How Might We
Innovation
How Might We

Turns a recognized pattern into an open design question.

View in playbookPlan and Evaluate UX ResearchMove from unclear research questions to reliable qualitative insights and next decisions.

Finished the method guide?

Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.

Go to profileMore methods