We use cookies for analytics and advertising. You choose what to allow. Privacy Policy.

Card Sort Analysis: How to Read a Similarity Matrix and Dendrogram

Running a card sort is one thing; analysing the results properly is another. After participants have grouped your cards, you're left with category names to clean, a similarity matrix to read, and a dendrogram that may look intimidating at first glance. This guide walks through the analysis step by step: standardising categories, reading the matrix, interpreting the dendrogram, and turning those clusters into navigation structure you can validate with a tree test.

Get your similarity matrix and dendrogram

Free plan: 3 studies. Unlock the full analysis for a single study for $15, or choose Starter at $19/mo.

1,800+ users

See an example analysis →

What a Card Sort Gives You

After participants have finished sorting, you have three main outputs:

  • Category names and groupings: Each participant created categories (open sort) or placed cards into your categories (closed sort).
  • A similarity matrix: Shows how often pairs of cards were grouped together across all participants.
  • A dendrogram: A tree diagram that visualises hierarchical clustering based on the similarity data.

Your job now is to turn these outputs into navigation structure that makes sense for your users.

Step 1: Standardise Category Names

In an open sort, participants create their own category names. You'll see variations like "About Us", "Company Info", "Who We Are" for the same grouping. Merge these synonyms into a single label for category-level analysis.

Look for:

  • Plurals vs singulars ("Product" vs "Products")
  • Synonyms ("Help" vs "Support" vs "Customer Service")
  • Abbreviations ("FAQ" vs "Frequently Asked Questions")
  • Spelling variants ("Colour" vs "Color")

The similarity matrix is calculated from card co-occurrence data (which pairs were grouped together), so it's independent of category names. Category name standardization matters when interpreting which categories your participants created, not when reading the matrix itself.

Step 2: Read the Similarity Matrix

The similarity matrix shows how often pairs of cards were grouped together. Each cell displays a percentage: 100% means every participant put those two cards in the same category, 0% means none did.

Similarity Matrix (Illustrative Example)A matrix showing how often pairs of cards were grouped together, with percentages from 0% to 100%. Higher percentages (darker blue) indicate stronger agreement.HomeAboutTeamCareersServicesPortfolioBlogContactHomeAboutTeamCareersServicesPortfolioBlogContact100%85%75%70%45%40%25%30%85%100%75%70%45%40%25%30%75%75%100%80%35%30%20%25%70%70%80%100%35%30%20%25%45%45%35%35%100%85%15%20%40%40%30%30%85%100%15%20%25%25%20%20%15%15%100%70%30%30%25%25%20%20%70%100%Agreement:70%+50-69%30-49%0-29%

Illustrative example. Darker blue = stronger agreement.

What to look for:

  • High percentages (70%+): Strong agreement. These cards belong in the same navigation category.
  • Moderate percentages (50-69%): Reasonable agreement, but check whether subcategories make sense.
  • Low percentages (below 30%): Cards are seen as separate. Don't force them into the same category.
  • Split cards: If a card has moderate agreement with two different clusters, it may need to appear in both places or be labelled more clearly.

Thresholds depend on your sample size: larger samples give more stable percentages. Use 70%+ as a cutoff for strong agreement and 50-69% for moderate agreement.

Step 3: Read the Dendrogram

A dendrogram visualises hierarchical clustering. Cards that merge lowest (closest to the card labels) were grouped together most often; merges higher up mean weaker agreement.

Dendrogram (Illustrative Example)A tree diagram showing how cards cluster together at different agreement levels. Cards that merge lowest (closest to the card labels) were grouped together most often; merges higher up mean weaker agreement.0%25%50%75%100%Agreement LevelHomeAboutTeamCareersServicesPortfolioBlogContact22.5%37.5%72.5%85.0%80.0%85.0%70.0%Strong clusters(merge lowest, near cards)Weak merges(merge higher, far from cards)

Illustrative example. Cards that merge lowest have the strongest agreement.

How to interpret it:

  • Lowest merges (high agreement): Form your main navigation categories. In the example, "Home + About + Team + Careers" is a strong company info cluster; "Services + Portfolio" is a strong work cluster.
  • Higher merges (low agreement): May be statistical artifacts rather than meaningful patterns. The final merge at 22.5% suggests these two clusters should stay separate in navigation.
  • Single-card branches: Cards that don't merge until very high are outliers. Check whether they were unclear or genuinely don't fit anywhere.

Different clustering algorithms (average linkage, complete linkage, Ward's method) produce slightly different dendrograms. Most tools default to average linkage, which balances sensitivity and stability for most sample sizes.

Closed Sorts: The Placement Matrix

In a closed sort, you measure placement agreement: what percentage of participants put each card into each predefined category.

What to look for:

  • Cards with high agreement (70%+) in one category: Good fit.
  • Cards with split placement: Participants disagree on where it belongs. Either the card label is unclear, or your categories overlap.
  • Cards that don't fit anywhere: Often placed in "Other" or spread thinly across categories. These cards may need clearer labelling or a new category.
  • Categories with very few cards: May be too narrow, or your content set doesn't support them.

Turn Clusters Into Navigation

Once you've identified strong clusters, design your navigation structure:

  1. Main categories from strong clusters. Use the lowest merges in your dendrogram or the highest-agreement groupings in your matrix.
  2. Subcategories from mid-level clusters. If a main category is large, check for subclusters at 50-69% agreement.
  3. Clear, consistent labels. Use the most common category names from participants, adjusted for clarity and your brand voice.
  4. Validate with a tree test. Card sorting reveals mental models; tree testing checks whether your navigation design actually helps users find things. Run a tree test with different participants to measure success rates and directness.

Worked Example

The diagrams above show a card sort for a small agency website. Here's what the analysis reveals:

  • Company info cluster (72.5% agreement): Home, About, Team, Careers. This becomes an "About" or "Company" navigation section.
  • Work cluster (85.0% agreement): Services, Portfolio. This becomes a "Work" or "What We Do" section.
  • Content cluster (70.0% agreement): Blog, Contact. Weaker than the others, but still a reasonable pairing for a "Resources" or "Connect" section.
  • Weak final merge (22.5%): Don't merge all three clusters into one mega-category. Keep them separate in the navigation.

Proposed navigation: About, Work, Blog, Contact. Tree test this structure to see if users can find specific pages reliably.

Automatic vs Manual Analysis

You can analyse card sort results manually in a spreadsheet: count placements, calculate co-occurrence percentages, build a similarity matrix, and draw a dendrogram. This works for small studies but becomes error-prone and time-consuming with more responses.

Automated tools (like Free Card Sort) generate the matrix and dendrogram from your data in seconds, letting you focus on interpretation rather than calculation. For most projects, automation is faster and more accurate.

Get your similarity matrix and dendrogram

Free plan: 3 studies. Unlock the full analysis for a single study for $15, or choose Starter at $19/mo.

1,800+ users

See an example analysis →

Frequently Asked Questions

What is a similarity matrix in card sorting?

A similarity matrix shows how often pairs of cards were grouped together across all participants. Each cell displays a percentage: 100% means every participant put those two cards in the same category, while 0% means none did. High percentages suggest strong mental associations; low percentages suggest the cards belong in separate areas of your navigation.

How do I read a dendrogram?

A dendrogram is a tree diagram showing how cards cluster together at different levels of agreement. Cards that merge lowest (closest to the card labels at the bottom) were grouped together most often; merges higher up mean weaker agreement. The strongest clusters appear at the bottom and should become your main navigation categories. Higher merges may be statistical artifacts rather than meaningful patterns.

What agreement threshold should I use?

Common thresholds are 70%+ for strong agreement, 50-69% for moderate agreement, and 30-49% for weak agreement; pairs below 30% are usually noise rather than a real grouping. Your sample size affects stability: larger samples give more confident percentages. Higher thresholds give you more confident clusters but may miss useful groupings; lower thresholds capture more structure but include noisier patterns.

Do I need to analyse card sort results manually?

No. Free Card Sort generates the full analysis automatically from your participants' responses (available with the per-study unlock or Starter plan). The manual spreadsheet method is time-consuming and error-prone. Automated analysis gives you the same statistical outputs in seconds, letting you focus on interpretation rather than calculation.

What do I do after I read the dendrogram?

Turn the lowest (strongest) clusters into a navigation structure, label them clearly, and validate the design with a tree test. Card sorting reveals mental models; tree testing checks whether users can actually find things in the structure you built from those models. Run both methods in sequence for best results.

Related Guides