DesignallBeginner

Information architecture: Card Sorting

Method for understanding how users group information, helping to define or validate an information architecture.

What it is

Main objective

Understand how users organize information and help define navigation structure.

Detailed description

Card Sorting is an information architecture methodology that reveals how users mentally organize and categorize content, providing fundamental insights for structuring navigation and taxonomies. Participants group cards representing content or functionalities into logical categories, allowing identification of common patterns in mental models and organizational expectations. Research demonstrates its effectiveness in creating intuitive structures that reduce disorientation and improve findability (Nielsen Norman Group). It is essential to conduct before designing navigation and can be executed through open, closed, or hybrid approaches depending on the project's specific objectives.

When it fits

Information architecture, before final wireframes.

•When designing or redesigning navigation
•When defining information hierarchies

Use cases

WebMobile appsDesktop applications

Effort level

Medium

Recommended number of users

15–20 participants

What you gain and what you give up

Advantages

User-centered: It reveals how users structure information, instead of imposing the organization's structure (the business's mental model).
Cheap and simple: It is inexpensive and easy to run, especially in its manual format.
Sparks discussion: When run in a group, the debate among participants provides very rich qualitative information about their mental models.
Flexibility: It works both for generating new structures (open) and for validating existing ones (closed).
Input for labeling: It provides real user terms and vocabulary for navigation.

Disadvantages

Not final navigation: The results are logical groups, not ready-to-use navigation menus. Expert interpretation is required to turn the groups into a viable information architecture.
Surface-level bias: Participants may group cards based on superficial keywords (e.g., grouping everything that says "printer") instead of by the meaning or function of the content.
Lack of context: Organizing cards is a classification task, not a search task. It does not guarantee that the user will be able to find the information in a real task.
Complex analysis: Data analysis, especially in open studies with many participants, can be tedious and difficult without specialized software.