Affinity Diagrams

Summary

Affinity diagrams organize subjective data, concerns, and ideas into categories based on intuitive relationships so patterns and commonalities can be discovered. The tool is best used by cross-functional groups of four to eight people and is useful in several innovation phases, especially creation, value proposition, resourcing, and production. The process starts with a defined issue and brainstorming output, then groups related ideas, assigns headings, refines categories, and can feed into a cause-and-effect diagram.

Important Keywords

• Affinity diagram: a method for organizing many ideas or observations into natural groups based on perceived relationships.
• KJ method: the affinity diagram method associated with Jiro Kawakita.
• Brainstorming: generating many ideas or observations before grouping and interpreting them.
• Subjective data: qualitative impressions, concerns, and statements that affinity diagrams help organize.
• Concern categories: clusters of related concerns identified through grouping.
• Intuitive relationships: relationships among items discovered by sorting based on felt similarity rather than predefined categories.
• Cross-functional teams: teams from multiple areas who can jointly interpret grouped information.
• Creation phase: the phase where ideas or observations are generated and grouped.
• Value proposition phase: the phase where grouped concerns are interpreted for value or opportunity.
• Resourcing phase: the phase where needed resources are considered from the affinity results.
• Production phase: the phase where insights are translated into implementation or action.

• Cause-and-effect diagram: a related tool for organizing possible causes of a problem.

• Code: a label for a meaningful segment of data.

• Theme: a higher-level pattern of meaning across coded data.
• Inductive analysis: bottom-up analysis driven by data.
• Deductive analysis: top-down analysis guided by theory or prior concepts.
• Saturation: point where further grouping adds little new insight.

Important Concepts

• Affinity diagrams help organize large numbers of ideas or facts that appear chaotic, including brainstorming output and survey results.
• The method groups ideas by perceived relationships, then identifies category headings that summarize each group.
• Team discussion and rearrangement are part of the process, because groupings often change as relationships become clearer.
• The tool supports innovation work by comparing options, evaluating relationships between outputs and outcomes, comparing financing approaches, and supporting production decisions.

• Preparation requires simple materials such as marking pens, a large work surface, and cards or posted notes.

• Affinity mapping is a practical qualitative analysis method: write observations on cards or sticky notes, group related items, name groups, refine them, and derive insights.

• It is useful after interviews, diary studies, field studies, and probe returns because it helps teams move from many fragments to patterns.
• Thematic analysis is a more systematic approach to identifying, organizing, and interpreting patterns of meaning across qualitative data.
• Braun and Clarke's six phases are familiarization, generating initial codes, searching for themes, reviewing themes, defining/naming themes, and producing the report.
• Good themes answer the research question, have a clear focus, are supported by data, and together tell a coherent analytic story.

Examples

• The text shows an affinity diagram example for the issue, "What are the issues involved in completing a purchase order?"
• Brainstormed issues include slow computer systems, approval delays, missing information, damaged boxes, missing labels, complex data entry, unavailable product substitutions, and handwritten information that is difficult to read.
• These ideas are grouped into concern categories such as receiving process, order processing, and PO preparation, then further separated into actionable categories such as PO policy, receiving process/equipment, order processing system, internal, and external.
• A production-phase example asks whether output should be produced internally or outsourced to Asia.