Complete firstContent inventory
The content relevant to the structure question is listed and can be described as individual cards.
Missing or duplicated items distort grouping patterns.
Method guideUX Research
Card Sorting
A practical method guide: clarify prerequisites and materials, understand the complete process, and look up outputs, pitfalls, and stop criteria.
ComplexityLow
Time20-45 min
ParticipantsBased on research question
FormatWorkshop + async
MaturityCanonical
01
Prerequisite
What needs to be finished first
Complete firstSuitable sort format
Open, closed, or hybrid sorting matches the specific discovery or evaluation question.
A prescribed category set can prevent discovery.
Complete firstParticipant context
Audience experience and use context are understood.
Sorts from people outside the intended audience may not answer the question.
02
Preparation
What needs to be ready before start
Materials
Research question, one card per content item, neutral sorting instructions, and a capture grid. Use a piloted digital sorting tool for remote studies.
People / roles
A research or moderation lead explains the format neutrally and records questions. Participants sort individually; a group discussion can obscure individual results.
Pre-read
Record the content inventory, audience, sort question, intended use of findings, and known internal categories.
Time needed
About 20–45 minutes for an individual sort. Plan preparation, recruitment, and comparative analysis separately; remote studies may run over several days.
Setup
Shuffle cards and check readability. For moderated work, explain the rules without examples that imply the expected taxonomy.
03
Core question
The one question this method answers
Which content groups and terms fit the mental models of the relevant audience?
04
Flow
Phases with goal, actions, output and pitfall
Card sorting makes mental models and useful terms visible as participants group clearly worded content cards using their own logic. The results provide structure hypotheses for later information architecture.
Sort analysis
Fictional teaching example
Track content cards, groups, participant labels, and open structure questions.
| Sort finding | Content cards | Participant group | Participant label | Analysis question |
|---|---|---|---|---|
| Example groupRecord a group from one individual sort. | Search for a tool; check availability | P1: grouped together | Find a suitable tool | Does this grouping recur in other sorts? |
| VariationKeep a single card or alternative placement visible. | View pickup location | P1: separate group | Plan pickup | Is pickup its own task or part of reservation? |
1. Content cards
- Example group
- Search for a tool; check availability
- Variation
- View pickup location
2. Participant group
- Example group
- P1: grouped together
- Variation
- P1: separate group
3. Participant label
- Example group
- Find a suitable tool
- Variation
- Plan pickup
4. Analysis question
- Example group
- Does this grouping recur in other sorts?
- Variation
- Is pickup its own task or part of reservation?
Keep group labels in the participant’s wording.
- 1
Frame
Bound the question
Goal
Decide which structure decision the sort should inform.
Inputs for this step
- Name the content whose expected grouping or label remains unclear.
Actions
- Set the audience, content area, and open question: discover groups, assess existing categories, or compare both.
Output
Focused information-architecture question.
Focused information-architecture question.
Quality check
The method fits a content-grouping question, not a visual-design review.
Pitfall
The team wants to assess navigation, page layout, and labels all at once. Separate grouping from later tree tests or interface tests.
Timing: 20–30 minutes for focus and scope
- 2
Cards
Prepare cards and format
Goal
Write clear cards and select a sorting format.
Inputs for this step
- Focused information-architecture question.
Actions
- Create a representative, plainly worded set; remove duplicates, pilot the terms, and choose open, closed, or hybrid sorting.
Output
Versioned card set, sort format, and moderator notes.
Versioned card set, sort format, and moderator notes.
Quality check
Each card represents content; open sorting does not prescribe categories.
Pitfall
Ambiguous cards or an overloaded set produce noisy sorting. Bound the content, simplify labels, and pilot the set.
Timing: 30–60 minutes for card set and pilot
- 3
Sort
Run the sort
Goal
Let participants sort independently and explain their labels.
Inputs for this step
- Versioned card set, sort format, and moderator notes.
Actions
- Give the question and rules neutrally; record groups, labels, multi-placement, and uncertainty without suggesting a structure.
Output
Individual sorts with participant-created group names and comments.
Individual sorts with participant-created group names and comments.
Quality check
Sorts remain separate; the moderator does not correct a participant’s logic.
Pitfall
The moderator suggests a category or reinterprets groups. Ask why items belong together and capture the participant’s wording.
Timing: 20–45 minutes per moderated sort; unmoderated varies
- 4
Synthesize
Analyze patterns and variation
Goal
Compare sorts and turn similarities and differences into structure hypotheses.
Inputs for this step
- Individual sorts with participant-created group names and comments.
Actions
- Review recurring groups, competing labels, single cards, and multi-placement; keep alternative explanations visible.
Output
Structure hypotheses with open label questions and follow-up needs.
Structure hypotheses with open label questions and follow-up needs.
Quality check
Interpretations trace to individual sorts; disagreement is retained.
Pitfall
A cluster is adopted directly as the final category. Form multiple hypotheses and check findability with tree testing.
Timing: Analysis time in addition to participant time
05
Artifact
What comes out at the end
Sketch: what the artifact looks like
Form
Versioned card set, sort format, individual sorts, participant labels, analysis notes, and structure hypotheses.
Versioning / ownership
Date the card set, sort format, and question. Preserve original sorts; add derived clusters as analysis instead of overwriting raw data.
Tool alternatives
- Paper cards or accessible digital cards
- Content-item catalog and wording variants
- Comparison grid for groups, labels, and divergent sorts
card-sorting-working-template.md
Prepare content cards, sorting format, and analysis of individual sorts.
Card Sorting: Worksheet
Question
To be confirmed
Desired outcome
To be confirmed
Scope
One focused sorting study Time: Multi-session research initiative; duration depends on sorting format, recruitment, and comparative analysis.
Complete this worksheet on paper or in your own document during the work.
Preparation for this scope
Method setup
- Materials: Research question, one card per content item, neutral sorting instructions, and a capture grid. Use a piloted digital sorting tool for remote studies.
- Roles: A research or moderation lead explains the format neutrally and records questions. Participants sort individually; a group discussion can obscure individual results.
- Advance information: Record the content inventory, audience, sort question, intended use of findings, and known internal categories.
- Overall time needed: About 20–45 minutes for an individual sort. Plan preparation, recruitment, and comparative analysis separately; remote studies may run over several days.
- Setup: Shuffle cards and check readability. For moderated work, explain the rules without examples that imply the expected taxonomy.
Prepare for the work steps
Frame: Bound the question
- Name the content whose expected grouping or label remains unclear.
Cards: Prepare cards and format
- Focused information-architecture question.
Sort: Run the sort
- Versioned card set, sort format, and moderator notes.
Synthesize: Analyze patterns and variation
- Individual sorts with participant-created group names and comments.
Sort analysis
| Sort finding | Content cards | Participant group | Participant label | Analysis question |
|---|---|---|---|---|
| Example group | ||||
| Variation |
- Example group: Record a group from one individual sort.
- Variation: Keep a single card or alternative placement visible.
Blank template: add your own cards, groups, labels, and analysis questions.
Frame: Bound the question
Expected artifact: Focused information-architecture question.
Focused information-architecture question.
Entry:
...
- The method fits a content-grouping question, not a visual-design review.
Cards: Prepare cards and format
Expected artifact: Versioned card set, sort format, and moderator notes.
Versioned card set, sort format, and moderator notes.
Entry:
...
- Each card represents content; open sorting does not prescribe categories.
Sort: Run the sort
Expected artifact: Individual sorts with participant-created group names and comments.
Individual sorts with participant-created group names and comments.
Entry:
...
- Sorts remain separate; the moderator does not correct a participant’s logic.
Synthesize: Analyze patterns and variation
Expected artifact: Structure hypotheses with open label questions and follow-up needs.
Structure hypotheses with open label questions and follow-up needs.
Entry:
...
- Interpretations trace to individual sorts; disagreement is retained.
Open questions and next steps
...
Method guide: https://methodatlas.meierhoff-systems.de/en/methods/card-sorting/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.
card-sorting-beispiel.md
Concrete filled scenario, fictional example
Fictional teaching example: neighborhood tool-lending content
All content and sorts are fictional. No study was conducted.
Question: Which content do borrowers expect to see together when planning a tool loan?
Sort format: Fictional open sort with four cards: “Search for a tool,” “Check availability,” “View pickup location,” and “Find repair help.”
Example sort P1: “Search for a tool” and “Check availability” are grouped under the participant-created label “Find a suitable tool.” “View pickup location” is a separate group, “Plan pickup.” “Find repair help” remains separate.
Observation: P1 moved cards individually and grouped pickup only after checking availability.
Hypothesis: Search and stock may belong together; pickup appears to be a later planning task. One sort is insufficient to decide a category.
Next step: Compare alternatives across more individual sorts and test the resulting navigation with tree testing.
07
Pitfalls
Recognize symptoms and steer against them
Quick overview, the full text is in the matching phase in the flow
- Bound the questionThe team wants to assess navigation, page layout, and labels all at once.View in phase
- Prepare cards and formatAmbiguous cards or an overloaded set produce noisy sorting.View in phase
- Run the sortThe moderator suggests a category or reinterprets groups.View in phase
- Analyze patterns and variationA cluster is adopted directly as the final category.View in phase
Trap
Bound the question
Symptom
The team wants to assess navigation, page layout, and labels all at once.
What to do
Separate grouping from later tree tests or interface tests.
Trap
Prepare cards and format
Symptom
Ambiguous cards or an overloaded set produce noisy sorting.
What to do
Bound the content, simplify labels, and pilot the set.
Trap
Run the sort
Symptom
The moderator suggests a category or reinterprets groups.
What to do
Ask why items belong together and capture the participant’s wording.
Trap
Analyze patterns and variation
Symptom
A cluster is adopted directly as the final category.
What to do
Form multiple hypotheses and check findability with tree testing.
08
Stop criteria
When to pause, resolve prerequisites, or choose another approach
Participants do not know or use the content: revisit audience or question.
Cards are ambiguous: pilot the terms and clarify item descriptions.
The decision requires findability evidence: plan a follow-up tree test.
Finished the method guide?
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.