How to: Card Sort

Card sorting helps understand how users mentally organize information, allowing you to optimize the information hierarchy and improve the user experience.

The sorting type depends on the research goals and the tasks it aims to solve. For example,

Open card sort is typically used in early-stage research when you want to understand how users naturally group information.

Closed card sort can be used to optimize an existing structure when you aim to check how understandable the current structure is. You can also use it to validate hypotheses—when there are proposed suggestions for changes to check.

Hybrid card sort combines elements of both methods: testers can use existing categories as well as create their own. It allows for testing the current structure with added flexibility—if there is partial confidence in the structure but a need to ensure it fully meets users' needs.

01 Open or Closed Sort: In an open card sort, testers can create their own categories. The researcher may create a few categories in advance or choose not to create any at all. In a closed sorting, categories must be defined in advance.

02 Creating a Task: As a first step, you need to define the task for testers. For example, if it is an open card sorting, the task might look like this:

03 Creating Cards and Categories: To add cards, click the "Card" button in the card editing section—you can add as many as you like. There is also an option to shuffle the cards so that they are displayed in a random order.

The same can be done with categories—in a closed card sort, creating categories is mandatory, while in an open sort, you can allow testers to create their own.

04 Additional Settings: You can add images and descriptions to each card in the card creation menu. You can also allow testers to skip sorting some cards. If this option is enabled, testers can complete the sorting even if not all cards have been sorted. Additionally, you can shuffle the categories so that they are displayed in a random order.

In the card sort report, you can find three main sections: Matrix, Categories, and Cards. Each section provides valuable insights for analyzing how testers organized and categorized the information.

Matrix

Here, you can see how testers distributed the cards into categories:

Each cell has the percentage of testers who categorized a specific card into a category.

This approach helps quickly assess user agreement in grouping information and identify contentious cards that were sorted into different categories.

Categories

This section contains information about all the categories used in the card sorting. Here you can find:

Cards

Here, you can find:

This tab allows to analyze how testers sorted the cards into categories, which cards posed difficulties, and which categories they were most frequently assigned to. This helps assess the grouping logic and identify potential inconsistencies in the structure.

01 Always create an Other category to understand which cards pose difficulties for testers.

02 Shuffle the cards and categories so that testers do not see the relation between them.

03 If you have a complex system, you can break it down into several blocks. This makes the process less tedious and allows testers to focus on specific areas.

04 A closed card sort can also be used as a follow-up to an open sort to analyze whether the categories defined in the first round seem logical to most users.

05 It’s best to create around 30-50 cards—this amount should not overwhelm testers while still providing enough information to confirm or refute a hypothesis.

06 Do not mix child and parent categories—this can lead to misunderstandings in analyzing the relation between categories. If both levels of hierarchy need to be tested, you can create one block for each and conduct separate research on each hierarchy.