Card Sorting: How to Uncover Mental Models & Inform UX Decisions

We’ve all heard the infamous statistic that says you have three seconds to capture a user before they ditch your website or app.

And it’s true—users expect intuitive navigation that helps them find what they need, fast.

Card sorting is an effective UX research method to ensure your information architecture does just that.

Card sorting is a UX research method used to discover how people understand and categorize information. In a card sort, participants group ideas or information written on cards into different categories in a way that makes sense to them. You can use virtual cards, pieces of paper, or an online card sorting tool.

Researchers and designers most commonly use card sorting to:

Here’s a quick rundown of how to conduct card sorts:

So, that’s your whistlestop tour of conducting card sorts—but there’s plenty more to think about: you need to recruit the right participants, select the right card sorting method, and effectively analyze your results.

Luckily for you, we cover it all in this guide—starting with the fundamentals of card sorting and how it can help you build better UX.

Card sorting has many benefits within UX research and design, namely due to its ability to comprehensively understand a user’s way of thinking. It’s a powerful tool to grasp how users categorize information and determine the best way to organize and label the content on your website or product.

Let’s get into more detail about a few of the main benefits of card sorting.

A mental model is your user’s internal map of how something should work based on their experiences and expectations. Card sorting helps you identify these mental models and make informed decisions based on users' expectations rather than assumptions.

Understanding your users’ mental models helps inform a variety of decisions, including how to:

Understanding your users’ mental models can then in turn inform a variety of decisions, including how to:

Information architecture (IA) is how you organize, structure, and label your content so that users can easily navigate your product and find the information they need. Creating an IA that's easy to understand and intuitive is key to providing a good user experience. That's where card sorting comes in.

With card sorting, your participants group product elements into categories that feel natural to them. This categorization is an excellent way to understand how users expect the information architecture of a website or product to be structured.

These insights will allow you to make informed decisions and organize your content in a way that makes sense to your audience.

Card sorting provides you with a glimpse into the perspective of your user. You can learn how people understand different concepts or ideas, and how they feel about them.

Whether it’s a selection of potential features or different naming conventions, card sorting is the ideal method to gain feedback from users based on their first impressions and instincts.

The best products are created with their user’s needs and pain-points in mind. Card sorting can allow you to start understanding these early on, before too much time has been invested in ideas that don’t align with your users.

Card sorting is an excellent way to gather new ideas from your target user base, from understanding naming conventions and colloquialisms to generating new categories or sub-groups.

By providing a group of potential users with the space and freedom to suggest ideas and categorize information in a way that makes sense to them, you are provided with a plethora of diverse thoughts and perspectives which are then yours to shape to your product.

Last but not least, card sorting is a simple and cost-effective method of quickly gaining insights about your users and how they think. Tests are easy to arrange and provide fast, reliable results.

As we explore in the next section of this guide, there are multiple ways to run a card sort—allowing you to easily connect with the users that will provide the most value for your research. You can run a card sort in person, using physical cards, or remotely using an online tool like Maze.

Card sorting offers several benefits as a user research method—here’s where you should consider card sorting:

There are three main types of card sorting, and while each type serves the same broad purpose—understanding how your users interpret and categorize information—the different types do have their own ideal uses:

Determining the right type of card sorting for your research needs is a key part of ensuring you get the results you’re looking for and gathering high-quality, relevant data.

Of course, as with any UX research, each method has its ideal circumstances, and its pros and cons. Let’s take a look.

Open card sorting is a generative research exercise, rather than an evaluative one—meaning, it helps uncover and define the problem at hand, rather than evaluate a solution.

In this card sorting type, participants organize topics into categories that make sense to them and then create their own category names.

This method helps reveal not only how users classify the cards but also what labels they use for each group, which generates new ideas and category names, and provides a deeper understanding of your user.

If you're unsure how to organize your website or app information, an open card sort can help. The results from an open card sorting test are helpful in understanding how a target audience structures information, identifies potential bottlenecks, and can help to better label categories and sub-categories.

Vaida Pakulyte, UX Researcher and Designer at Electrolux, notes that “An open sort exercise is good at the beginning of the project because it helps you to understand how people naturally categorize information.”

In particular, open card sorts are commonly used to:

Closed card sorting is an evaluative research method—it’s used to assess and validate potential ideas or solutions. In a closed card sorting session, you present participants with predefined categories and ask them to prioritize and sort cards as it makes sense to them, within those defined categories.

This type of card sorting study doesn't reveal how your group of participants naturally categorize a set of topics. Instead, it tests if existing category labels are clear.

Closed card sorting is a great way to test an existing structure, validate decisions when adding new items to an existing system, reorganize or rank content. You can use this method to:

Hybrid card sorting combines open and closed card sorting methods. Participants can group a set of cards into predefined categories, but they can also create new categories.

This card sorting technique is ideal for when you already have certain predefined categories, but you need input into how the remaining ones should be labeled, or aren’t positive on what goes where.

When improving a live website, the hybrid method might be most suitable. First, you can conduct a hybrid card sorting exercise to see how participants label and sort your website’s structure. You can then follow that up with a closed card sorting session that will help you validate the website’s structure and test if people can find the relevant information. Vaida explains: “Once the initial structure is set, then you can test it a second time with closed card sorting or a tree testing exercise.”

Hybrid card sorting is a great format if you’re looking to validate existing ideas and information, without closing off the potential for new ideas. It’s particularly useful if:

Similarly to how you can choose from various types of card sorting depending on your objectives, you can also choose from different card sorting formats:

Let’s take a closer look.

As the name suggests, moderated card sorting has a moderator guide participants through the process, asking questions and digging into their decisions as they sort. It’s more time-consuming, but it’s worth it when you’re designing something complex—like a new navigation system or a feature-rich product.

Moderated card sorting is your go-to when you need detailed feedback and want qualitative insights on the rationale behind participants' grouping.

Unmoderated card sorting is all about efficiency and autonomy. Participants are provided with instructions then left to their own devices. Running your card sort as an unmoderated study gives participants the freedom and flexibility to do what feels right, without feeling as though they are being watched, which could make them reconsider their natural decisions.

Unmoderated sessions can be quicker and easier to organize as they don’t require facilitators, and you can use digital card sorting tools such as Maze for conducting remote card sorting.

This is perfect when you need to validate existing designs or test how well a broader user base understands your current information structure. You’ll get a good sense of general trends and patterns without the heavy time commitment of moderated sessions.

In digital card sorting, you use online card sorting tools to simulate the drag-and-drop activity of dividing cards into groups.

This method is generally easier because it requires less resources—the card sorting software does the heavy lifting of analyzing the card sorting results and identifying common groupings.

While card sorting can be conducted in person, digital card sorts can provide more data more quickly, so are often the go-to for teams due to their flexibility, convenience, and speed.

Paper or in-person card sorting is a more traditional technique, where you write down the topics on physical cards and ask participants to group them. This format naturally lends itself more towards moderated card sorting, but can also be conducted as unmoderated.

Ultimately, identifying the card sorting type that’s best for you comes down to the type of project you’re conducting, your user research goals, and available resources.

Whether you're creating a new website or redesigning an existing one, following these best practices can help you make the most of your card sorting session and ensure reliable, actionable results.

If you have a large number of items for the card sort, focus on what's most important to start with. Group similar cards and ideas into broader categories. This helps participants concentrate on the critical aspects of your website or application, leading to more meaningful and actionable results. You can always run follow-up sessions later to delve into more detail.

Just like the way you word research questions can impact your answers, the wording of cards can heavily influence a card sorting study. Ensure the wording on each card accurately represents the content it refers to, making it easy for participants to understand.

Matthieu shares an example where you’d want to understand how runners categorize their running shoes, so you create cards like minimalist shoes, trail shoes, racing flats, and stability shoes. “What if you conduct a closed card sort and ask them to sort these cards into two categories: ‘Good running shoes’ or ‘Bad running shoes’?”, says Matthieu. “You’re leading participants to categorize the running shoes based on perceived quality rather than actual use cases.”

In the same way you want to avoid leading questions when interviewing or surveying users, you want to avoid leading cards when conducting card sorting. Instead, use neutral instructions like: “Please sort these cards into categories that make sense to you when thinking about different types of running activities.”

For example, if you’re organizing content for a pet website, focus either on parent categories like ‘Dogs’ and ‘Cats’ or child categories like ‘Dog Breeds’ and ‘Cat Breeds.’

If you need to explore both levels of hierarchy, consider conducting separate card sorting sessions for each. This allows participants to focus on one level at a time and avoid any confusion. For example, if you’re organizing content for a pet website, focus either on parent categories like ‘Dogs’ and ‘Cats’ or child categories like ‘Dog Breeds’ and ‘Cat Breeds.’

For example, you could use an open card sorting session if you're designing a new online bookstore and want ideas for organizing virtual shelves according to customer interest. Participants could create their own categories, such as ‘Dark thrillers’ or ‘Cheesy romance’, giving you insights into their preferences.

On the other hand, if you're working on improving the navigation of an existing news website with predefined categories like ‘Politics’, ‘Sports’, ‘Entertainment’, and ‘Technology’, a hybrid card sorting approach might be more suitable.

In this case, you could feature existing categories while also allowing participants to suggest new categories or place cards in a ‘Miscellaneous’ or ‘I don't know’ bucket. This approach helps you validate and refine the current information architecture, as well as identify potential gaps and generate new ideas.

Matthieu recommends consciously switching up how cards are given out. “Varying the order of the cards between participants will prevent any bias that the order of the cards might introduce.”

If all participants receive the cards in the same sequence, cognitive bias might subconsciously push them to group cards in a way that aligns with that order rather than their natural thought process.

This approach helps ensure your card sorting results accurately reflect your users' mental models and preferences, leading to a more intuitive and user-centric information architecture.

Matthieu also recommends pre-testing the card set with a small group of participants to double-check your instructions and chosen card sorting type work.

When you set up your card sorting, you need to recruit research participants who actually represent the target users of your website or product. This includes getting a good mix of those who know your product well and those who might be seeing it for the first time.

Recruiting participants using UX research tools is great when you need lots of participants. Maze offers two solutions to help you locate and manage the right participants:

Maze offers an intuitive drag-and-drop interface for card sorting tasks, making it easy for users to categorize information. This seamless experience allows participants to engage with the task from the get-go.

You can choose between two types of card sorting: open card sorting and closed card sorting. This flexibility ensures that your UX team gathers valuable insights into how users naturally group information, or can validate existing structures during the UX design process.

With Maze’s ready-to-use card sorting templates, you can create tailored instructions, adjust task settings, and add images or descriptions to improve the research experience. You can also limit the number of cards a participant can sort into each category.

Once the sorting is complete, use Maze’s reporting tools to visualize the results. You can track metrics like:

With Maze, card sorting becomes a streamlined process, offering both flexibility and in-depth insights to enhance your design decisions.

Card sorting is an excellent way to get ideas, understand how users think, gather insight on potential IA, and discover your users’ mental models.

Whether you prefer to work with physical index cards or online card sorting software, with a moderator or not, every card sort will answer a different set of questions and provide new understanding of your users, so you can lay a better foundation for a user-centered design.

Now we’ve discussed what options you have when it comes to card sorting, let’s take a look at just how you can plan and execute a successful card sort within UX research.