What is Card Sorting? Complete Beginner's Guide
Card sorting is a UX research method that reveals how users naturally group and label information. If you've ever wondered "where should this feature go?" or "how should we organize our navigation?", card sorting gives you the answer—directly from your users.
Card Sorting in 60 Seconds
What it is: Users organize cards (representing content, features, or pages) into groups that make sense to them.
Why it matters: Shows you how users think, not how you think.
When to use it: Designing navigation, organizing content, or structuring information.
How long it takes: 2-3 days to run a study, 1 day to analyze.
Cost: Free to $30 (using online tools + small incentives).
The Card Sorting Metaphor
Imagine you're organizing a bookstore.
Your way (insider perspective):
- Fiction by publisher
- Non-fiction by publication date
- Technical books by ISBN
Customer's way (user perspective):
- Fiction by genre (mystery, romance, sci-fi)
- Non-fiction by topic (cooking, business, history)
- Quick finds (bestsellers, new arrivals, staff picks)
Card sorting reveals the customer's way.
How Card Sorting Works
The Traditional Method (Physical Cards)
Historical approach:
- Write each item on an index card
- Give participants a stack of cards
- Ask them to group cards into piles
- Have them name each pile
- Record results manually
- Look for patterns across participants
Problems:
- Time-consuming to set up
- Limited to in-person sessions
- Hard to analyze (manual counting)
- Small sample sizes (5-10 people max)
The Modern Method (Digital/Online)
Today's approach:
- Create digital "cards" in online tool
- Send link to 20-40 participants
- They drag-and-drop to organize
- Software automatically analyzes results
- Get patterns, agreement scores, visualizations
Benefits:
- Setup in 5 minutes
- Remote participants (worldwide)
- Automatic analysis
- Larger sample sizes (20-40 people)
- Results in days, not weeks
Types of Card Sorting
Open Card Sort
How it works: Users create their own category names.
Best for:
- ✅ Starting from scratch
- ✅ Don't know the best categories yet
- ✅ Want unbiased input
- ✅ Discovering how users think
Example:
You give participants:
├─ 30 feature cards (no categories)
They create:
├─ "My Stuff" (5 cards)
├─ "Shopping" (8 cards)
├─ "Help & Support" (4 cards)
└─ "Settings" (3 cards)
Output: Natural groupings + category names users would understand
Closed Card Sort
How it works: You provide category names, users sort cards into them.
Best for:
- ✅ Validating existing structure
- ✅ Testing specific hypotheses
- ✅ Comparing option A vs. option B
- ✅ Making sure your labels work
Example:
You give participants:
├─ Account
├─ Shop
├─ Support
└─ Company
They sort 30 cards into these 4 categories
Output: Which cards fit which categories, agreement scores
Hybrid Card Sort
How it works: Provide suggested categories, but users can create new ones.
Best for:
- ✅ You have ideas but want flexibility
- ✅ Testing + discovering simultaneously
- ✅ Iterative refinement
- ✅ Balanced approach
Example:
You suggest:
├─ Account
├─ Shop
└─ Support
Users can:
├─ Use your categories
├─ Rename them
└─ Create new ones like "My Orders"
Output: Validation of your ideas + new suggestions
When to Use Card Sorting
Perfect Use Cases
1. Website Navigation
- Redesigning main menu
- Too many pages, unclear structure
- Users can't find what they need
Real example: E-commerce site with 50 product categories. Card sort revealed users wanted to browse by occasion (work, casual, athletic) not just product type.
2. App Information Architecture
- Organizing features in mobile app
- Deciding what goes in which menu/tab
- Reducing complexity
Real example: Banking app with 30 features. Card sort showed users wanted "Move Money" and "See My Money" categories instead of "Transactions" and "Accounts."
3. Content Organization
- Structuring blog or knowledge base
- Organizing documentation
- Creating content taxonomy
Real example: Help center with 150 articles. Card sort revealed users wanted task-based categories ("Getting Started," "Troubleshooting") not feature-based.
4. Product Categorization
- E-commerce category structure
- Filtering systems
- Faceted search
Real example: Furniture store. Card sort showed customers think by room (bedroom, living room) AND by style (modern, rustic), leading to dual-axis navigation.
When NOT to Use Card Sorting
❌ Testing visual design
- Card sorting is about organization, not aesthetics
- Use: Usability testing or A/B testing
❌ Testing workflows or processes
- Card sorting shows grouping, not sequential steps
- Use: Task analysis or journey mapping
❌ Getting feedback on ideas
- Card sorting assumes items are final
- Use: Concept testing or surveys
❌ Testing findability directly
- Card sorting shows groupings, not necessarily where users look first
- Use: Tree testing (validates card sort results)
❌ Less than 15 items to organize
- Too few items don't provide useful patterns
- Better: Just ask users directly
The Science Behind Card Sorting
Why It Works
1. Mental Models
Humans organize information in their minds (mental models). Good design matches user mental models, not designer mental models.
Card sorting reveals mental models.
2. Cognitive Psychology
People naturally categorize to make sense of complexity. Card sorting taps into this innate ability.
Example: You instinctively group:
- Fruits (apples, oranges, bananas)
- Vegetables (carrots, broccoli, spinach)
Card sorting leverages this.
3. Collective Intelligence
20-30 people reveal patterns better than 1 expert. Patterns with 70%+ agreement indicate strong shared mental models.
What the Research Says
Sample Size Research:
- 15 participants reveal ~90% of patterns
- 20-30 participants give confidence
- 30+ has diminishing returns
Reliability Studies:
- Test-retest reliability: 85-90%
- Remote vs. in-person: 85-95% similarity
- Open vs. closed: Complement each other
Validation Research:
- IA based on card sorting = 30-40% improvement in findability
- Reduces navigation confusion by 50%+
- Increases task completion rates significantly
Step-by-Step: Your First Card Sort
Step 1: Define Your Goal (5 minutes)
Ask yourself:
- What am I trying to organize?
- What decision will this inform?
- Who are my users?
Example goal: "Understand how users would group our 30 product features for the app navigation redesign."
Step 2: Create Your Cards (15 minutes)
Select items:
- 20-40 items to organize
- Representative of full content
- Specific, clear names
Example cards (SaaS product):
- Dashboard
- Analytics Reports
- Team Chat
- File Sharing
- Task Board
- Calendar View
- Time Tracking
- Notifications
- User Permissions
- Integrations
... (20 more)
Tips:
- Use actual labels users will see
- Avoid jargon unless your users use it
- Keep names short (2-5 words)
- Make each card distinct
Step 3: Choose Study Type (2 minutes)
First time? → Start with Open card sort
Why: Reveals how users naturally think without your bias.
Step 4: Write Instructions (5 minutes)
Template:
Welcome! Thank you for helping us improve [Product].
Please organize these features into groups that make sense to you.
Create category names that describe each group.
This should take 10-12 minutes. There are no right or wrong answers!
Thank you!
Step 5: Set Up Study (5 minutes)
Use a card sorting tool:
- Free Card Sort (free, easy) [recommended]
- Optimal Workshop (expensive, comprehensive)
- UserZoom (enterprise-focused)
Step 6: Recruit Participants (1-2 days)
How many: 20-30 for open sort, 30-40 for closed
Who: Target users (not colleagues!)
Where to find them:
- Customer email list (best)
- User research panels (UserTesting, Respondent)
- Social media
- Friends/family (last resort)
Incentives: $5-$10 gift cards (optional but recommended)
Step 7: Launch Study (Instant)
Soft launch:
- Send to 5 participants first
- Check first few responses
- Fix any issues
- Send to everyone else
Monitor:
- Check responses daily
- Look for confusion
- Send reminder after 3 days
Step 8: Analyze Results (4 hours)
Look for:
- Cards grouped together 70%+ of time (strong relationships)
- Most common category names
- Surprising groupings
- Cards with under 40% agreement (confusing or don't fit)
Most tools auto-generate:
- Similarity matrix (visual heatmap)
- Dendrograms (hierarchical grouping)
- Agreement scores
Step 9: Implement Findings (Ongoing)
Create structure based on patterns:
Example results → Implementation:
Users grouped:
├─ "My Work" (Dashboard, Tasks, Calendar)
├─ "Team Stuff" (Chat, Files, @Mentions)
├─ "Reports" (Analytics, Time Tracking, Export)
└─ "Settings" (Permissions, Integrations, Account)
Becomes app navigation:
├─ Work
├─ Team
├─ Insights
└─ Settings
Real Example: From Card Sort to Design
Before Card Sort
Internal team's proposed navigation:
├─ Features
├─ Data Management
├─ User Administration
├─ Configuration
└─ Tools & Utilities
Problem: Technical language, unclear what's where.
Card Sort Results
30 participants sorted 35 features. Patterns emerged:
Strong groupings (over 75% agreement):
├─ "My Projects" / "Work Space"
│ └─ Project list, Tasks, Files
├─ "Team" / "People"
│ └─ Members, Chat, Activity
├─ "Reports" / "Analytics"
│ └─ Dashboards, Data Export, Charts
└─ "Settings" / "Account"
└─ Profile, Permissions, Billing
Implemented Design
Final navigation (based on card sort):
├─ Projects (most common user label)
├─ Team
├─ Insights (more approachable than "Reports")
└─ Settings
Results
- 45% reduction in navigation confusion
- 62% faster to find features
- 28% increase in feature adoption
- 4.1 → 4.7 star app rating
Common Beginner Mistakes
Mistake #1: Testing with Wrong People
Wrong: Ran card sort with 10 coworkers Right: Ran with 25 actual users
Impact: Coworker results reflect internal thinking, not user thinking.
Mistake #2: Too Many Cards
Wrong: 80 cards, 45-minute study Right: 35 cards, 12-minute study
Impact: Fatigue leads to rushed, unreliable results.
Mistake #3: Vague Card Names
Wrong: "Resources," "Platform," "More Options" Right: "Video Tutorials," "Dashboard," "Account Settings"
Impact: Users don't understand cards, create random groupings.
Mistake #4: Expecting 100% Agreement
Wrong: "Only 65% agreed, the study failed!" Right: "65% shows the primary pattern. 35% indicates some variation, which is normal."
Impact: Unrealistic expectations lead to ignoring valuable data.
Mistake #5: Skipping Validation
Wrong: "Card sort showed this structure, ship it!" Right: "Card sort suggests this. Let's validate with tree testing before building."
Impact: Card sorting shows groupings, not necessarily findability.
Card Sorting Tools
Free Card Sort (Recommended for Beginners)
Pros:
- ✅ Free plan (3 studies)
- ✅ 5-minute setup
- ✅ No credit card required
- ✅ Easy for participants (no login)
- ✅ Mobile-friendly
- ✅ Automatic analysis
Cons:
- ❌ Free plan limited to 50 responses per study
Best for: First-timers, small teams, freelancers
Optimal Workshop
Pros:
- ✅ Industry standard
- ✅ Advanced analytics
- ✅ Multiple research methods
Cons:
- ❌ Expensive ($149-$449/month)
- ❌ Steeper learning curve
- ❌ Overkill for simple projects
Best for: Large organizations, research agencies
DIY (Google Forms + Spreadsheets)
Pros:
- ✅ Completely free
- ✅ Full control
Cons:
- ❌ Manual everything
- ❌ Poor participant experience
- ❌ Time-consuming analysis
- ❌ No similarity matrix
Best for: Extremely tight budgets (but not recommended)
Frequently Asked Questions
Q: How is this different from surveys? A: Surveys ask opinions. Card sorting reveals mental models through behavior. More reliable for structure/organization questions.
Q: Can I run card sorting in-person? A: Yes, but online is faster, cheaper, and gets more participants. Reserve in-person for when you need follow-up questions during the activity.
Q: How long does a study take to complete (for participants)? A: 8-15 minutes for 30-40 cards. Keep it under 20 minutes to avoid fatigue.
Q: What if people disagree on groupings? A: Some variation is normal and valuable. Look for patterns with 60-70%+ agreement. Disagreement under 40% means cards may be unclear.
Q: Do I need to pay participants? A: Not always. Engaged customers often participate for free. General public usually needs $5-$10 incentive.
Q: Can I test multiple navigation ideas? A: Yes! Run multiple closed card sorts (one per option) or use hybrid sort.
Q: What software do I need? A: Online card sorting tool (Free Card Sort recommended), no other software needed.
Q: How many participants is enough? A: 20-30 for open sort, 30-40 for closed. Patterns emerge around 15-20, more gives confidence.
Next Steps
Learn More
Want to dive deeper?
- Card Sorting Examples - 15 real-world case studies
- Card Sorting for IA - Complete IA guide
- Remote Card Sorting - Best practices for online studies
- Card Sort Templates - Ready-to-use templates
Run Your First Study
Ready to try it?
- Create free account (2 minutes)
- Add your cards (5 minutes)
- Send to participants (1 minute)
- Get results (2-3 days)
No credit card required. No risk. Just valuable insights into how your users think.
Summary
Card sorting reveals how users naturally organize information.
Why it matters: Matches your design to user mental models.
When to use: Designing navigation, organizing content, structuring information.
How long: 2-3 days for study, 1 day to analyze.
Cost: Free to $30.
Getting started: Run your first study now →