Card Sorting

Imagine you’re clocking in for your shift at the library and the owner asks you to sort and organize all books in a way that feels right to you. You start by sorting them by genre and then from A-Z. But when you ask your colleague for some input, they tell you they would have organized the books by color and alphabetically by the author's last name. Later that day, a friend of yours tells you about a bookstore that organizes their books by the country the story takes place in (yes, this library actually exists). This makes you reflect on what customers expect—what’s logical to you may not be logical to them. 

Card sorting operates similarly. It's a method used to understand how people logically categorize information. But not just any type of information, card sorting is commonly used to make sure a website aligns with its users’ expectations. In other words, they help designers create intuitive and easy-to-navigate websites.

During a card sorting session, participants are given cards representing website content including topics, links, images, videos, and other sections used to navigate through a website. They are then asked to sort these into categories that make sense to them. 

For example, a gourmet e-commerce website might provide participants with the cards: olive oil, chocolate, spices, and tea. A participant might create three categories, sorting olive oil and spices into "Cooking Essentials," chocolate into "Sweets," and teas into "Beverages."

User Experience (UX): UX describes the overall experience a user has when interacting with a product, system, or service. The goal of UX design is to create simple, effective, relevant, and overall enjoyable experiences for the user. UX design considers the why, what, and how of a product from the user's perspective.⁴

Information Architecture (IA): Specific discipline within UX. IA requires organizing, structuring, and labelling content effectively and sustainably. The goal of IA is to help users find information and complete tasks by providing a clear navigation path and intuitive placement of items.⁵

Mental Models: How individuals process information and form expectations based on their understanding and experience. They guide our perception and behavior.⁶

Open card sorting:  The most common method. Participants come up with their own overall categories. It allows you to fully understand the mental models of your users.^1,2

Closed card sorting:  This method is chosen when you want to analyze if existing data/websites work. Participants are given pre-selected categories and asked to place all cards into those groups.^1,2

Hybrid card sorting: This is the least common method. It involves giving participants pre-selected categories but also allowing them to create new ones. It could often lead to status quo bias, as it exploits the user’s tendency to stick to the way things already are (preselected groups), rather than performing actions that require change.

Card sorting tasks were built on principles of cognitive psychology, more specifically, the study of pattern identification and information arrangement. This method was formally introduced during the late 1880s and early 1890s.⁷ These first experiments looked at reaction time, memory, and imagination, which ultimately led to the development of the Wisconsin Card Sorting Test—a neuropsychological test that measures cognitive processes.⁸

As card sorting demonstrated its utility in mapping out human mental models, its application extended beyond psychology into fields like criminology, social sciences, semantics, and notably, market research—a field that exploded after the creation and increased accessibility of the internet. The growing emphasis on UX in the 1990s led to a shift towards designs that focused on the user. It was in this context that card sorting became a critical tool in developing market research and UX strategies, enabling a deeper insight into consumer preferences and behaviors.⁷

The evolution of technology further transformed card sorting from a manual, time-consuming process to a common digital practice. Modern software now allows remote and large-scale studies that have expanded and facilitated the use of card sorting. The fast-paced introduction of artificial intelligence (AI) has promising applications, as it might be able to provide deeper insights into card sorting findings. In other words, card sorting evolved from a simple research tool into a sophisticated technique.

Improved Information Architecture 

Card sorting directly aids in structuring or restructuring a website so it’s intuitive and easy to understand. Having a clean and easy-to-navigate website might also cause people to infer the level of organization or quality of a business.

Enhanced User Understanding and Navigation

For clients to purchase a product, they need to understand it. By employing card sorting, we can gain insight into how users think, enabling us to identify and comprehend their needs more effectively. With its findings, you can make informed decisions that promote ease of use and limit frustration points. 

Identifies Gaps in Content

After analyzing card sorting results, it’s not uncommon to find some information missing, especially if there isn’t space for individualized insights or feedback. By noticing and addressing these gaps you might come up with innovative and creative ideas and solutions.

Reduced Development Time

Insights from card sorting can be quickly incorporated into prototypes or designs, allowing for rapid, cost-effective testing and iteration of IA concepts.

Although the benefits of card sorting sound promising, they shouldn’t be used in all cases. Card sorting usually focuses on a particular section of a website, which can result in a lack of overall context (since it doesn't encompass the entire site or business concept). Also, this type of methodology typically involves creating categories and not subcategories which results in a knowledge gap. This could likely be solved by qualitative research, and by asking participants further questions, but it's not common practice.

Additionally, users might get tired of sorting cards and find it tedious. This leads to biased results that might not accurately capture their needs and thoughts. Also, if your target demographic encompasses a wide variety of people you might encounter really different views and preferences. While in general, this is a positive, it may not be the most conducive to effective card sorting. It may lead to difficulty interpreting results and knowing what to do with them. Finally, the digital world is moving quite fast which is also causing user preferences to shift drastically in short periods. This concern begs the question as to whether card sorting strategies are the most straightforward method. 

Despite these controversies, card sorting remains a valuable tool in the UX and IA toolkit, offering unique insights into user cognition and behavior. Addressing its limitations through careful planning, diverse participant selection, and combining it with other research methods can mitigate some of these concerns, leading to more effective and user-centered designs.

This Barbie is organized

When Mattel noticed a decline in visits to the Doll Showcase section—with over 2,000 dolls and accessories—of their Barbie collector’s website, they realized they needed to spice things up. The website was no longer resonating with collectors' perspectives and navigation preferences. Consequently, Mattel decided to outsource and find a way to restructure its page so it would align with the collector's preferences.

The consultant on the case, Dave Rogers, was keen to perform a card sorting session but he knew making participants categorize over 2,000 cards would not work. He decided to choose “fashion” as a subset or topic to reduce the number of cards to 130. This would make everyone’s job easier and lead to more specific insights explaining the collectors’ mental models. Rogers and his team recruited nine participants which included newbies to the most dedicated Barbie collectors. They performed both open and closed card sorting sessions.

The open sort analyzed how collectors naturally organized the 130 dolls into their own categories. The closed sort aimed to understand if the existing organization into collections/series within themes made sense to users. The results were quite valuable to Mattel, serious collectors categorized and found value in groups that involved expert knowledge of doll attributes. However, those who were less experienced preferred categories based on doll names and/or concepts and on the feelings that the doll names sparked.

The biggest finding or takeaway message was that all collectors had personalized preferences. Having that in mind Mattel’s new website enabled users to search by doll name, year of release, SKU numbers  (identity number above a barcode), Collector LabelsTM, and past themes. Their new design considered the diverse preferences of collectors and allowed personal browsing experiences. This was a great example of how card sorting can be quite useful and successful as the website’s visits increased.

Note: This website was launched in the late 2000’s and is no longer available. But you can visit its updated version Mattel Creations if you’re interested in analyzing how they organize their collections.⁹

The Elements of Choice with Eric Johnson 

This podcast episode explores the concepts and insights from Eric Johnson's work on decision-making, choice architecture, and how different elements influence the choices people make.

1. Whitenton, K. (2024, February 2). Card Sorting: Uncover Users' Mental Models for Better Information Architecture. Nielsen Norman Group. Retrieved March 1, 2024, from https://www.nngroup.com/articles/card-sorting-definition/
2. Optimal Workshop. (n.d.). Card Sorting 101 - How To Run An Online Card Sort. Optimal Workshop. Retrieved March 1, 2024, from https://www.optimalworkshop.com/learn/card-sorting-101-introduction-to-card-sorting/
3. UX Tweak. (2023). UX Quotes. Retrieved March 1, 2024, from https://blog.uxtweak.com/ux-quotes/
4. Interaction Design Foundation. (n.d.). UX Design. Retrieved March 1, 2024, from https://www.interaction-design.org/literature/topics/ux-design
5. Usability.gov. (n.d.). What & Why of Usability. Retrieved March 1, 2024, from https://www.usability.gov/what-and-why/information-architecture.html
6. Clear, J. (n.d.). Feynman Technique. Retrieved March 1, 2024, from https://jamesclear.com/feynman-mental-models
7. Interaction Design Foundation. (2014). Card Sorting. In The Encyclopedia of Human-Computer Interaction, 2nd Ed. Retrieved March 1, 2024, from https://www.interaction-design.org/literature/book/the-encyclopedia-of-human-computer-interaction-2nd-ed/card-sorting
8. ScienceDirect. (2017). Information Architecture. Retrieved March 1, 2024, from https://www.sciencedirect.com/science/article/abs/pii/B9780128022191000055
9. Spencer, D. (2009). Card Sorting: Designing Usable Categories. Retrieved March 1, 2024, from https://www.ipcinfo.org/fileadmin/user_upload/drm_matrix/docs/CardSorting-for-printing.pdf