Card sorting in UX - How to perform card sorting?

Do you know a UX research method called card sorting? Today, we will tell you how card sorting can come in handy for researchers and user experience designers. You will also show the pros and cons of card sorting, demonstrate its example, as well as present its different types and their characteristics.

Card sorting in UX – table of contents:

What is card sorting?
Advantages and disadvantages of card sorting
How to perform card sorting?
Types of card sorting
Summary

What is card sorting?

Card sorting is a research method in which participants group topics in a way that is logical to them. Typically, UX researchers conduct card sorting by writing topics on individual cards, asking users to sort related topics into groups, and then naming each group (or assigning it to a category). Card sorting provides insight into a user’s thought process, its workings or how they think it should work. This research method helps UX researchers and designers formulate an information architecture that is more intuitive to the user and easy to navigate.

Advantages and disadvantages of card sorting

Like any other research method – card sorting is not perfect, it has pros and cons worth keeping in mind. One of the main advantages is its simplicity – card sorting is a simple exercise, relatively easy for researchers to organize, and easy for participants to understand.

The method is also relatively inexpensive – whether it is a desktop study (e.g., with colored post-it notes) or conducted using online tools, the cost of conducting it is usually low. Card sorting is also very efficient, as it provides enough useful data in a short period. The last major advantage we want to mention is that the mode is based on accurate insights from our actual users.

On the downside, card sorting may provide inconsistent data – the results obtained can vary widely from one participant to another. It also has some limitations – by focusing on labels and categorization, it does not provide feedback on how users perceive the actual content or how it relates to the actual tasks they perform. It’s also worth remembering that analyzing the results of this survey is time-intensive. While users quickly sort through the cards, analyzing the data takes much more time.

How to perform card sorting?

To carry out card sorting, you should start by selecting a set of topics/words to be grouped. This set should contain between 40 and 80 items, and each of them should be written on a separate card. Try to avoid topics containing the same words – otherwise, participants may tend to subconsciously combine these items into one group. Your task is to combine the cards (topics) into groups that are logical to you. Start by shuffling the cards and handing them to the participants. Ask them to look at the cards one at a time and combine them into groups. Some groups may be large, others small.

If the participants are unsure about a card or don’t know what it means, you can set it aside or possibly come back to it later. Let them know that they can change their minds during their work – they can, for example, move cards from one stack to another, combine two stacks into one larger one, or divide a stack into several new ones. In the next step, the user gives the group’s names.

After dividing the topic cards into groups, ask the attendees to write down their proposed names for the groups. This will give you some ideas for navigation categories, but keep in mind that these are only guidelines for you as a designer – don’t expect participants to create final category names for the site.

After dividing the cards and naming the groups, you may (but don’t have to) ask the participants to explain why they created and named the groups that way. Ask if any items were particularly difficult to assign to a category if any items fit into more than one group, and what they think about any items left unsorted.

If you are interested in learning more about users’ reasoning, you can also ask them to “think aloud” throughout the survey. This will provide detailed information, but will also require more time for analysis. Remember that you can request your subjects to create more practical group sizes if necessary. You should not impose anything on them during the initial sorting, but after the initial summary, you can suggest, for example, breaking one large group into several smaller ones.

Repeat card sorting with about 15-20 subjects. This is an adequate number of participants to detect patterns in users’ thinking patterns.

Once the survey is complete and all the data is collected, it’s time to analyze it. Look for common groupings, category or theme names, and items that were often combined. If you notice that some items were often set aside, determine whether this is due to unclear card names or content that seemed unrelated to the other themes. Card sorting will help you understand what system of organization will be most effective for your users.

Types of card sorting

Important differences in card sorting include whether users can create their category names or have them imposed; whether a moderator leads the session or it is an unmoderated survey; and whether the survey is conducted on paper or with an online tool. Each type has certain advantages and disadvantages, of course.

Open card sorting vs. closed card sorting

Open card sorting is the most common variant of this study (it is the course of classic open card sorting that we described in the previous subsection – “how to conduct card sorting?”). In open card sorting, users are free to assign their names to the groups – stacks of cards that they arranged a moment earlier.

Closed card sorting is a variation in which users are given a predetermined set of category names and are asked to organize individual cards into these predetermined categories. Closed card sorting does not reveal how users perceive and group a set of topics. Instead, it is used to assess whether the existing category structure connects logically to the content from the user’s perspective. A criticism of closed card sorting is that it only tests users’ ability to match content to the correct category – which, for users, can be more like solving a puzzle than naturally matching content to categories. Thus, the method does not reflect the natural way users browse content.

Moderated vs. unmoderated card sorting

The second division is between moderated and unmoderated card sorting. Moderated card sorting includes the summary described above and thinking out loud during sorting – so during the survey, the participant makes ongoing comments, tells what they are doing, shares their thoughts, and at the end explains why they made the decisions they did and not others. Including this step during the survey is a great way to gain qualitative insight into users’ rationale for their grouping. If necessary, you can ask additional questions and inquire about specific cards to better understand his thought processes.

Unmoderated sorting, on the other hand, involves users assigning cards to groups on their own, often with an online tool and without any interaction with the moderator. This is certainly faster and cheaper than moderated card sorting, as it does not require the researcher to talk to each user individually. Non-moderated card sorting, on the other hand, can be useful as a complement to moderated card sorting sessions.

Paper versus digital card sorting

The last division we want to present is “paper” card sorting vs. performing with the help of technology. Paper card sorting is by far the most traditional form of card sorting. Topics are written on fiches, and users are asked to form groups. The biggest advantage of paper card sorting is that study participants don’t have to learn how to use any tool, since all they have to do is stack the cards on a table.

It’s a simple flexible process: users can move cards around at will, and even start over as needed. It’s also easier for people to manage a very large number of cards on a large table than to manage many objects on a computer screen (which often can’t even show everything in one view). The disadvantage of sorting paper cards is that researchers must manually document each participant’s choices – making the analysis potentially long and tedious.

Digital card sorting applies software or a web-based tool to simulate topic cards that subjects drag and drop into selected groups. This method is generally the easiest for researchers, as the software can automatically analyze the results of all participants and reveal which items were most often grouped, what category names users created, and the likelihood of pairing two items. The downside is that the usability of the tool can affect the success of the session – a lack of technical skills or technological problems can cause frustration and even prevent users from creating exactly the groups they want.

Summary

Card sorting can, therefore – depending on the needs of the project – take very different forms. In this way, the survey can be tailored to the needs and capabilities of the project and the research team, depending on your time resources, budget or established research goals and questions. This method can be a great way to explore the user’s thought process and see if categories and groupings that seem logical and orderly to product developers, researchers and UX designers are equally intuitive to their real users.

If you like our content, join our busy bees community on Facebook, Twitter, LinkedIn, Instagram, YouTube, Pinterest, TikTok.

Author: Klaudia Kowalczyk
A graphic & UX Designer which conveys into design what cannot be conveyed in words. For him, every used color, line or font has a meaning. Passionate in graphic and web design.
View all posts