Imagine you run an online store with hundreds of products. You group them the way your team thinks. Shoes go here. Bags go there. Then customers show up and cannot find anything. They look for “gifts under $50,” but you have no such group.

This gap happens often. The people who build a site think one way. The people who use it think another. Card sorting is a simple method that closes this gap.

This guide explains what card sorting is, the main types, and how to run a study. You will also find examples and tips for your next project.

What Is Card Sorting?

Card sorting is a research method that shows how people group and name information. You give participants a set of cards. Each card holds one topic, page, or product. They sort the cards into piles that make sense to them. In some studies, they also name each pile.

The results show how real users expect content to be organized. You then use these insights to shape menus, categories, and page structure.

The method started in psychology, where researchers used it to study how people classify ideas. UX teams later adopted it as a core tool for information architecture. It is cheap, fast, and easy to learn.

Card sorting works best when you want to answer questions like these. How do users group our content? What words do they use for each group? Which items do they think belong together?

Why Card Sorting Matters

People do not all think like your team. Internal staff know the company chart, the product names, and the jargon. Users know none of it. They sort things by their own needs and habits.

Card sorting brings that view into the open. It helps you build a structure that feels natural instead of forced. This leads to easier navigation, fewer lost visitors, and better findability.

It also settles debates. When team members disagree about menu labels, real data from users gives a fair answer. And since the method is low cost, you can run it early, before design and development spend real money.

For SEO, a clear structure helps too. Logical categories make it easier for search engines to crawl your site and understand how pages relate. They also help users move through related content, which keeps them engaged.

Types of Card Sorting

There are three main types. Each suits a different stage of the project.

Open Card Sorting

In an open sort, participants create their own groups and name them. You give no categories in advance.

This type is best at the start of a project. It shows how users naturally organize content and which words they choose. Use it when you build a new site or plan a major redesign.

Closed Card Sorting

In a closed sort, you provide the categories. Participants place each card into one of them.

This type is best when you already have a structure and want to test it. It shows whether users place items where you expect. It also reveals which cards cause confusion.

Hybrid Card Sorting

A hybrid sort mixes both styles. You give a set of categories, but participants may also create new ones if nothing fits.

This works well when you have a draft structure but want room for fresh ideas.

Moderated vs Unmoderated

You can also run sorts with or without a facilitator. In a moderated session, a researcher watches, asks questions, and listens to the participant think aloud. This gives rich detail on why people choose certain groups.

In an unmoderated session, participants work alone, often online. This reaches more people at lower cost, though you lose the chance to ask follow-up questions.

A Simple Example

Suppose a bank wants to simplify its website menu. The team writes 30 cards, such as “Open a savings account,” “Home loan calculator,” “Report a lost card,” and “Branch locator.”

They run an open sort with 15 customers. Many place “Report a lost card” and “Branch locator” together in a pile named “Help.” The team had planned to put the lost card page under “Cards.” The data shows that customers see it as an urgent help task.

Several participants also group “Home loan calculator” with tools rather than with loans. The team now sees that a “Tools” section might serve users better.

After updating the structure, they run a closed sort to confirm it. Most participants place cards where the team expects. The new menu is ready for the next stage.

How to Run a Card Sorting Study

You do not need special skills. Follow these steps.

1. Set a Clear Goal

Decide what you want to learn. Are you building a new structure or checking an existing one? Your goal decides which type of sort to use.

2. Choose Your Content

Pick the items to sort. These could be page titles, product names, or features. Aim for 30 to 60 cards. Too few give weak results. Too many tire participants and lower the quality of their work.

3. Write Clear Cards

Use short, plain labels. Avoid jargon unless your users know it. Keep the level of detail even. Do not mix a broad item like “Insurance” with a narrow one like “Update my phone number.” Do not use words that give away a category, since that sways how people sort.

4. Pick Your Method and Tool

For in-person sessions, use paper index cards and a big table. For remote studies, use an online tool. Optimal Workshop, UXtweak, Maze, and Miro all support card sorting. Pick one that fits your budget and team.

5. Recruit the Right People

Find participants who match your real audience. Do not rely on coworkers, who know too much. For a reliable pattern, aim for 15 to 30 participants in an open sort and fewer in a closed one. Studies with five users can still reveal major issues, but larger groups give clearer data.

6. Run a Pilot

Test your study with one or two people first. Check that the cards are clear and the instructions make sense. Fix any problems before the full run.

7. Run the Sessions

Explain the task in simple terms. Tell participants there are no wrong answers. In moderated sessions, ask them to think aloud. Take notes on hesitation, debates, and cards they struggle with. Avoid helping too much, since you want their natural thinking.

8. Analyze the Results

Look for patterns. Which cards often land together? Which group names come up again and again? Online tools build charts for you. A similarity matrix shows how often two cards were grouped. A dendrogram shows clusters in a tree-like view.

Also note the outliers. Cards that people place in many different spots may need a clearer label or a place in more than one category.

9. Turn Findings Into Structure

Use the patterns to draft your site map and menu labels. Then test the draft with a closed sort or a tree test. This confirms that users can find items in the new structure.

Card Sorting Best Practices

A few habits make your study stronger.

Start with real content. Use actual page titles and product names, not made-up ones. This keeps the results useful.

Keep the card count manageable. Around 40 cards is a sweet spot. If you have more content, split it into several studies by section.

Write each card in the same style. If some cards are full sentences and others are single words, people will group by style instead of meaning.

Recruit people who match your audience. If you serve several user groups, run sorts with each. Their mental models may differ.

Use open sorting first, then closed sorting. Open sorts help you generate a structure. Closed sorts help you check it. This two-step approach gives strong results.

Ask follow-up questions. After a sort, ask why they made certain groups. Their reasons often reveal more than the piles themselves.

Do not copy results blindly. Card sorting shows patterns, not final answers. Balance the data with business goals, content needs, and common sense.

Pair it with other methods. Tree testing checks whether people can find things in your structure. Usability tests show how the full design works. Analytics reveal what real visitors do.

Include accessibility. Make sure online tools work with keyboards and screen readers. Offer other ways to take part if someone cannot use the tool.

Common Mistakes to Avoid

Some errors come up often. Using internal jargon on the cards is one of the biggest. Participants then sort by guesswork instead of real understanding.

Another mistake is using too many cards, which leads to fatigue and rushed answers. Teams also pick the wrong participants, such as employees, which hides real user thinking.

Some teams treat the results as final. People may sort items in ways that do not fit business needs or technical limits. Use the data as a guide, then test your design.

Finally, be careful with leading labels. If several cards share the same key word, people will group them by that word alone. Vary your wording so you capture meaning, not matching text.

Card Sorting vs Tree Testing

These two methods often work together, but they answer different questions.

Card sorting helps you build a structure. It shows how users group and name content. Tree testing helps you check a structure. Participants get a text-only menu and are asked to find certain items. You see whether they succeed, how fast, and where they get lost.

A common plan is to run an open card sort, draft a structure, and then run a tree test to confirm it works.

Final Thoughts

Card sorting is a simple way to see how your users think. It turns guesses into evidence and helps you build menus and categories that feel natural.

Start with a clear goal. Choose a fitting type. Write clear cards, recruit real users, and look for patterns in the results. Then test your new structure before you launch.

When your content is grouped the way people expect, they find what they need faster. That makes for a better experience for users and better results for your business.

is this related to ui ux?

Yes, card sorting is closely tied to UI/UX design. It is a UX research method, and it feeds directly into information architecture, which is a core part of the UX process.