Card sorting is one of the most powerful — and underused — UX research methods. It reveals how real users think about your content, navigation, and information architecture.

Whether you're redesigning a website, building a new app, or trying to fix confusing navigation, card sorting gives you data-driven answers instead of guesswork.

What Is Card Sorting?

Card sorting is a UX research method where participants organize topics, labels, or content into groups that make sense to them. Each "card" represents a piece of content, feature, or navigation item.

The goal is simple: understand how your users mentally organize information so you can build navigation and structures that match their expectations.

Why Card Sorting Matters

Problem	How Card Sorting Helps
Users can't find what they need	Reveals natural mental models
Navigation labels confuse users	Discovers preferred terminology
Information architecture feels "off"	Maps content to user expectations
Redesign needs data, not opinions	Provides quantitative agreement data

Types of Card Sorting

Open Card Sort

Participants create their own categories and group cards into them. Best for discovering how users naturally organize information.

When to use: Early-stage research, new product, major redesign.

Closed Card Sort

You provide pre-defined categories. Participants sort cards into those fixed groups. Best for validating an existing or proposed structure.

When to use: Testing a proposed navigation, comparing alternatives.

Hybrid Card Sort

Pre-defined categories exist, but participants can also create new ones. Combines discovery with validation.

When to use: When you have a hypothesis but want to stay open to surprises.

How to Run a Card Sort Study

Step 1: Define Your Cards

Choose 30–60 items (too few = no patterns, too many = participant fatigue)
Use actual content titles or features from your product
Write labels clearly — avoid jargon participants won't know

Step 2: Choose Your Method

Decision	Open Sort	Closed Sort
You have no existing structure	✅	❌
You're validating a redesign	❌	✅
You want both discovery + validation	Hybrid	Hybrid

Step 3: Recruit Participants

15–20 participants for reliable patterns
Match your actual user demographics
Include a mix of experienced and new users

Step 4: Run the Sessions

Remote (unmoderated):

Write clear instructions explaining the task
Set a time estimate (usually 15–20 minutes)
Let participants sort at their own pace
Collect results automatically

In-person:

Prepare physical cards (index cards or sticky notes)
Explain the task, then step back
Observe silently — note hesitations and regroupings
Ask participants to explain their grouping logic after

Step 5: Analyze Results

Similarity Matrix: Shows how often participants grouped any two cards together. High similarity (>70%) means strong association.

Dendrograms: Tree diagrams that show hierarchical clustering of cards. Help identify natural category boundaries.

Category Naming: In open sorts, analyze the category names participants created. Common themes suggest intuitive labels.

Minimum viable analysis: Look at the similarity matrix. If two cards are grouped together by >70% of participants, they belong together. If <30%, they're in different mental categories.

Common Card Sorting Mistakes

Too many cards. Over 60 cards causes fatigue and random sorting.
Vague card labels. "Settings" means different things to different people — be specific.
Leading instructions. Don't hint at the "correct" answer.
Ignoring outliers. Sometimes the most interesting insights come from unexpected groupings.
Skipping the analysis. Raw data isn't insight — invest time in pattern recognition.

Card Sorting with Afkar

Afkar supports both open and closed card sorting with real-time results. Create a study, add your cards, recruit Arabic-speaking participants from the MENA region, and get dendrogram analysis automatically.

No need for spreadsheets or manual clustering — the platform does the analysis for you.