Card Sorting : Uncover user's mental models

Card Sorting is a user research technique used in UX and Information Architecture (IA) to help understand how users group, label, and navigate content. In a card sorting exercise, participants organize topics into categories that make sense to them and, in some cases, label these groups. It’s a simple yet powerful method to improve your website or app’s navigation, menu structure, and overall content discoverability.

Blog image
Blog image

💡 Why Use Card Sorting?

As designers, we often design based on assumptions or internal business logic. But your users may think differently — and card sorting helps you bridge that gap.

Benefits:

  • Improves information architecture and navigation

  • Uncovers users' mental models

  • Highlights confusing labels or categories

  • Validates your content grouping decisions

📦 Types of Card Sorting

There are three main types of card sorting, each with a specific purpose:

🔓 1. Open Card Sorting

  • Users create their own categories and groupings.

  • Best for exploratory research, when you don't have predefined categories.

  • Useful in early-stage projects.

Example: You give participants a set of product names (e.g., t-shirts, jackets, heels, jeans), and they organize them into their own groups and name them — like “Tops,” “Bottoms,” “Footwear.”

🔒 2. Closed Card Sorting

  • You provide predefined categories, and users sort content into them.

  • Best when you're validating an existing IA or redesigning.

  • Helps test how well your categories make sense to users.

Example: You provide categories like “Electronics,” “Home & Garden,” “Beauty,” and ask users to sort product items accordingly.

🔄 3. Hybrid Card Sorting

  • A mix of open and closed.

  • Users can use predefined categories and create new ones if needed.

  • Best when you want structure but are open to new insights.

Example: Users can sort content into existing labels, but if they feel a new category is needed, they can add it.

🧠 When Should You Use Card Sorting?

Use it when you:

  • Are building a new website or app

  • Need to redesign or restructure menus, navigation, or IA

  • Want to understand how users expect to find content

  • Are naming or categorizing content-heavy elements like product catalogs or help centers

🧰 Tools for Card Sorting (Remote & In-Person)

Online Tools:
In-Person:
  • Sticky notes & pens

  • Index cards

  • Table space or whiteboards

📋 Step by Step Process

  1. Define the Goal : Are you exploring how users group content or validating an existing structure?

  2. Choose a Method : Open, Closed, or Hybrid?

  3. Prepare Your Cards

    • Each card = one topic, item, or piece of content.

    • Keep labels clear, jargon-free, and concise.

  4. Recruit Participants : Ideally 15–20 participants from your target user group.

  5. Conduct the Session

    • In person: Ask participants to sort cards into groups.

    • Online: Use tools to track sorting, naming, and patterns.

  6. Analyze Results

    • Look for common groupings, outliers, and patterns.

    • Use dendrograms, similarity matrices, or visual maps (provided by tools like OptimalSort).

  7. Apply Learnings : Use the insights to build or improve your navigation or content structure.

🧾 Real World Example

Let’s say you're designing the IA for an e-commerce fashion website. You want to organize your product catalog in a way that makes sense to your customers.

You list items like:

  • T-shirts

  • Jeans

  • Skirts

  • Sneakers

  • Sunglasses

  • Jackets

  • Heels

  • Scarves

In an Open Card Sort:

Users might group them like this:

  • Tops: T-shirts, Jackets

  • Bottoms: Jeans, Skirts

  • Footwear: Sneakers, Heels

  • Accessories: Sunglasses, Scarves

They might even rename categories — for example, “Shoes” instead of “Footwear.”

These groupings reveal how your actual users think, helping you create a more intuitive menu.

📊 How to Analyze Card Sorting Data

You can analyze card sorting data by:

  • Identifying most common groupings

  • Using similarity matrices (tools like OptimalSort auto-generate this)

  • Reviewing category labels users apply (especially in open sort)

  • Spotting confusion or disagreement across participants (important outliers!)

⚠️ Common Pitfalls to Avoid

  • Using too many cards (stick to 30–40 max)

  • Adding unclear or ambiguous labels

  • Ignoring outlier groupings — they often point to unmet user expectations

  • Forgetting to analyze label choices in open sorting

📋 What is important during a Card Sorting session?

  • It is important to avoid overlapping terms or concepts between the cards, as they are often grouped together, potentially influencing the research results.

  • Ensure an adequate number of participants (15-25) for representative outcomes.

  • Provide participants with the option to add comments for a deeper understanding of their thought process.

  • Utilize a combination of open and closed card sorting for a more comprehensive perspective.

✅ Key Takeaways

  • Card sorting is a low-cost, high-value UX method.

  • Helps align your content with user expectations.

  • Useful for menu structures, site maps, and product categories.

  • Combine it with tree testing for validating the structure afterward.

📥 Want to try it yourself?

MIRO templates https://miro.com/miroverse/card-sorting-template/

📚 Good Reads

Designroadmap

A carefully curated collection of design resources, tools, and inspiration.

© 2024 Designroadmap. Curated with ❤️ by a designer for the design community.

Designroadmap

A carefully curated collection of design resources, tools, and inspiration.

© 2024 Designroadmap. Curated with ❤️ by a designer for the design community.