Card Sorting

Card sorting is a user research method that helps designers understand how users categorize and organize information. Participants categorize content on index cards, physically or digitally. Designers use card sorting in the information architecture (IA) phase to build structures that align with users’ mental models.

In this video, Design Consultant and author Donna Spencer states, “The most important thing about this [card sorting] is that you learn about how people think about the ideas under question. You understand what things people think go together. You learn a lot about what people call those things…” 

Why Is Card Sorting Important in Design?

Card sorting is important because it helps designers understand user mental models, improves IA, and is a key part of a user-centered design approach. It contributes to the overall success of a design because it integrates the final product with user expectations and behaviors.

Card sorting impacts design in the following ways: 

1. User-Centered Design: Card sorting allows designers to understand users’ unique perspectives and helps to design interfaces that align with users' mental models.

2. Informs Information Architecture: It helps designers organize information and content in the most intuitive way. When designers analyze how users categorize content, they can create IAs that represent user expectations which, in turn, improves navigation and usability.

3. Content Relevance: Card sorting helps identify and prioritize what content or features users consider most important so they are easily accessible.

4. Usability Improvement: When designers understand how users expect information to be organized and presented, they can make informed decisions to improve the overall usability of a design. Usability enhances the user experience and reduces cognitive load.

5. Validation of Design Decisions: Card sorting can validate existing IAs or test proposed structures. This ensures that design decisions are grounded in user preferences and allows for more effective designs.

6. Iterative Design Process: Design is an iterative process, and card sorting fits into this since it provides feedback that can be used to refine designs. Continuous testing and adjustment based on user input lead to designs that continually improve over time.

7. Collaborative Decision-Making: Card sorting can involve stakeholders, including clients and team members, in the decision-making process. It facilitates collaboration and considers diverse perspectives.

8. Optimizing Navigation: Designers can use the insights from a card sort workshop to optimize navigation paths—crucial for websites or applications with a large amount of content. 

9. Enhancing Findability: Card sorting helps designers with discoverability. Users can find relevant information more easily when the design aligns with their mental models.

10. Adapting to User Diversity: Different user groups have different ways to cluster and label information. Card sorting allows designers to accommodate the diverse needs and preferences of various user segments.

Types of Card Sorting

There are several types of card sorting: 

• Open Card Sort: Participants organize website content into groups based on their understanding and name each group to reflect its content accurately. Designers opt for an open card sort to see how users naturally label and group content without predefined categories. This method is beneficial in the early stages of a project.

• Closed Card Sort: Participants categorize website content into predetermined groups. This method is suitable when working with a set of pre-established categories and aims to understand how users sort content items into each.

• Mixed Card Sort: Designers can combine open and closed card sorting when they have some established categories but want to explore new ones. This hybrid approach lets participants use provided categories and create new ones. It's useful to expand or restructure a partially established IA.

Designers should choose different types of card sorting depending on the project’s specific requirements. 

How to Run a Card Sort

Here are some guidelines to successfully run a card sort.

Set Objectives

For any research methodology, it's crucial to clearly define the objectives and determine whether card sorting is the most effective approach to gain the intended insights. 

Example:

Consider a card sorting session which focuses on improving a community-based platform that connects volunteers with local service opportunities.

The organization wants to revamp its website. The current platform is complex, and users find it challenging to discover relevant volunteer opportunities.

In this case, an objective could be “Enhance the website's menu and navigation structure to make it more user-friendly for volunteers seeking community service opportunities.”

Select Tools

Designers can use plain index cards or sticky notes for in-person card sorting, with a range of a few to about 100 cards or more, depending on the project. For a virtual workshop, however, limit the number of cards to around 50.

Various online card sorting tools exist, and they offer a key advantage—they’re readily available and easily facilitate analysis without the need to transfer results from cards to a spreadsheet.

Prepare Instructions 

Meticulous preparation of instructions is essential. The instructions must be clear to ensure participants comprehend the task at hand. This becomes particularly crucial in an online card sort.

Designers should anticipate potential questions participants might pose. If the card sort is online, incorporate answers within the instructions. For an in-person card sort, prepare responses in advance.

Potential participant questions may include:

• Can a card be placed in more than one group?

• Can I exclude a card if I don't understand it?

• Am I allowed to add a card if I sense content is missing?

Example:

For the volunteer platform, participants organize the cards into groups based on how they would expect to find volunteer opportunities on a community service platform.

The designer should encourage participants to vocalize their thought process and share any specific preferences.

They should also answer any questions the potential users might have. 

Select Participants 

It’s important to select the right people for the card sort. Determine the ideal number of participants—this is typically more straightforward for in-person sessions due to space and time constraints; 3 to 5 participants is ideal. 

In online card sorting, it's recommended to involve only a select few individuals initially. Handling an extensive volume of responses can be overwhelming during the analysis phase. Begin by distributing the card sorting exercise to a small group, providing space for refinement and adjustment. If necessary, consider expanding the participant pool to a larger audience in subsequent rounds or phases.

Example: 

For the volunteer card sort, the designer or researcher running the session should recruit participants with varying levels of experience in volunteering, including newcomers and frequent volunteers. They should limit the number of participants to 3-5 per card sort.

Incentive

As with most research activities, incentivize the participants to ensure they are motivated to participate. This is especially true for online card sorts.

Example: 

The volunteer platform could offer a year of free access to the platform or $10 gift cards to a charitable organization, etc. 

Content Preparation 

It’s essential to evaluate the content thoroughly before the card sort. The content participants receive will determine the quality of the results.

For a larger content set, divide the process into multiple card sorts for the best results. Avoid using jargon, as unclear content can negatively impact effective card grouping. When individuals can't understand the information, they tend to force connections that aren't genuine. To achieve the best outcome, ensure the content is clear and direct.

Example: 

In the volunteer website example, the lead designer creates physical or digital cards for categories like "Education," "Environmental Projects," "Health and Wellness," etc. They leave a few cards blank with markers available in case participants have an idea they’d like to add to the sort.

The designer prepares a set of sample volunteer opportunities within each category and identifies key tasks users often perform, such as finding tutoring opportunities or environmental cleanup projects.

Workshop

It’s important to provide participants with context and explain the instructions. In physical card sorts, communicate verbally; for virtual sessions, written instructions are essential.

Avoid excessive explanations or over-assistance. Participants may seek guidance, but refrain from delving into extensive detail. Encourage them to proceed independently for optimal results.

Refrain from telling participants that they’ll label the groups they create during the workshop. While impractical for online card sorts, this ensures the users focus on categories rather than being influenced by predefined labels, preserving the organic card sorting process.

Throughout the card sort, make note of participant comments. Maintain distance to facilitate natural participant conversation without disruption. 

Example:

In the case of the volunteer site, the designer should explain the purpose of the session and emphasize that there are no right or wrong answers.

They can provide a brief overview of the current platform and its challenges in connecting volunteers with opportunities.

The lead designer presents specific tasks to participants, such as finding a tutoring opportunity or discovering environmental cleanup projects, and asks them to locate the relevant category in their card groups.

The designer observes and takes notes on how participants group the cards, the labels they assign to each group, and any specific comments they make.

After the sorting, the designer should ask participants about why they made certain choices, what terms they find most intuitive, and if they encountered any difficulties.

Analysis

“The first thing you want to do is just poke around… What did people do? What’s kind of interesting? What ways did they do things?”

—Donna Spencer

In the video, Donna Spencer speaks about how to analyze the data of a card sorting workshop and pinpoint what’s most interesting or relevant to reveal the final results. 

First, get an initial overview. It’s more convenient to analyze an online card sorting session since results are directly available on your computer. However, for in-person card sorts it’s best to streamline the process with numbers and a spreadsheet. Assign a number to each card and enter it into the spreadsheet along with the card's content to simplify data input.

Then delve deeper into the analysis—scrutinize the participants' results for detailed comparisons and identify those similarities and differences. Examine variations to extract valuable insights. Note instances where similar groupings contain different cards. Additionally, evaluate specific cards and their placement by each participant.

This is a qualitative research method—focus on the discovery of interesting aspects within the results that align with the card sort’s objectives.

Finally, integrate your findings into the project, preferably along with other research methods.

Example:

Following the volunteer organization, the designer analyzes the collected data to identify common patterns, groupings, and potential improvements for the website's menu and navigation.

They present the results to the design team and relevant stakeholders.

From there, the designer can recommend changes for the menu structure based on user insights to improve the volunteer experience.

If the design team has completed all of the above, they should be be able to establish the following outcomes after their analysis:

• Insights into how volunteers naturally categorize and expect to find community service opportunities.

• Identification of intuitive groupings and potential improvements for the website's menu structure.

• Greater understanding of user preferences in the platform’s navigation which adds to a more effective and user-friendly community service hub.

Card Sorting: Benefits and Drawbacks

Card sorting provides valuable insights into information organization and user preferences. Designers everywhere praise the practice because it improves usability, coordinates designs with how users think, and makes it easier to prioritize content. However, like any methodology, card sorting is not without its limitations. 

Benefits

Card sorting offers several benefits, including but not limited to:

• Information Organization: Card sorting helps organize and structure information and enhances accessibility and user understanding.

• User-Centered Design: Facilitates a user-centric approach by involving participants in the categorization process; it helps to align the design with user mental models.

• Content Prioritization: Identifies which content is most relevant or important to users during design and development.

• Navigation Design: Card sorting informs the design of intuitive navigation systems that are aligned with how users naturally group and categorize information.

• Reduced Cognitive Load: It minimizes cognitive load through logical content groupings and intuitive navigation.

Drawbacks

While card sorting is a valuable strategy, it does have some drawbacks:

• Limited Quantitative Data: Card sorting provides qualitative insights but lacks quantitative data, which makes the prioritization of results objectively a challenge.

• Time-Consuming: Sessions can be time-intensive, especially when they involve a lot of content to process or a large number of participants.

• Difficulty with Divergent Thinking: Some participants may struggle with open-ended categorization tasks. This limits the effectiveness of capturing diverse perspectives.

• Not Suitable for All Content: Card sorting may not be the best fit for highly complex or technical content; other methods like tree testing or mind mapping may be more appropriate.

• Dependency on Participants: Participants’ knowledge, backgrounds, or cognitive biases can influence results and potentially lead to skewed outcomes. Too many participants also presents a problem, as does too few.

Learn More about Card Sorting 

Read more about card sorting in Chapter 22 of IxDF’s The Encyclopedia of Human-Computer Interaction, 2nd Ed. by William Hudson. 

If you want to learn more about how to run a card sorting session, read IxDF’s Card Sorting: The Ultimate Guide (in 2024). 

Watch Donna Spencer’s Master Class, How To Use Card Sorting For Better Information Architecture.

Interested in learning more about card sorting and UX Design? Enroll in Interaction Design for Usability with user-centered design expert, William Hudson, today!

Read about card sorting on Usability.gov’s website.