Card Sorting: Designing Usable Categories
By Donna Spencer, with a foreword by Jesse James Garrett
Rosenfeld Media; 2009
A whole book on card sorting? Perhaps others shared my initial surprise when they learned that Donna Spencer was planning a whole book on the topic. After all, what’s to know? Get a few index cards, print things on them, and have representative users (or domain experts) sort them into meaningful categories. The author obviously heard the same comment from others, as Chapter 1 opens with the same question: “A whole book on card sorting?”
Spencer allayed any skepticism I had about how one might fill such a book with solid content. This book is replete with solid advice, clearly written and illustrated, and well supported by case studies and examples.
Chapter 1 introduces card sorting, but also alerts the reader to the availability of complementary, and sometimes more appropriate, user research activities.
Chapter 2 is a succinct exploration of the ways in which information can be categorized, organized, and classified. Without avoiding underlying concepts such as centrality and the classical view of categorization, Spencer manages to explore these in a way that is practical, simple, and well illustrated. The chapter concludes with a brief discussion of the fact that coming up with a perfect categorization scheme is an unachievable goal, despite occasional client expectations that information architects should be able to do things “in the right way.”
Chapter 3 is a discussion on defining the need to conduct a card sort. I found this to be rather light and unsatisfying—particularly as this is an area where newcomers may need significant assistance.
The following chapters jump into the practicalities—whether and when to use open or closed card sorting (including explanations of each), whether to use individuals or teams (Spencer prefers teams), how to locate, select, and name content, and how to choose participants.
Spencer discusses the use of software tools for card sorting, although she makes it clear that her preference is to use physical cards.
Chapter 8 is a hands-on tour through the practicalities of preparing for, and executing, a card-sorting session, and the following two chapters discuss how to manage and analyze the material collected during the session. Beginning with instructions and examples on how to enter data into a spreadsheet, Spencer then covers cluster analysis and multidimensional scaling. This part of the book gives the impression that while the material is included in the interest of completeness, it is not the author’s own preferred approach to analysis, and indeed, she cautions that statistical analysis, “can produce such nice outputs that it is easy to get drawn into a simple answer and forget about the underlying reason for the output.”
This sentiment is also echoed towards the conclusion of the book, in the exhortation: “Don’t rely on a technique to do your thinking.”
An appendix entitled “Documentation” was extremely light on detail, but did suggest that example reports were available on the publisher’s website (www.rosenfeldmedia.com) although, ironically, I could not find them. (The author tells me they should be online by the time this review is in print.)
I felt the index was, perhaps, not comprehensive. I was certain I had come across a reference to facets (surely I did?) but word “facet” does not appear in the index.
Overall, the book is easy reading, and will be of benefit both to those who are new to the card sorting (who will find useful step-by-step guidance and straightforward advice) and to experienced practitioners who all too often fail (if I am any example) to question and re-evaluate the practices that they feel they know thoroughly. For example, the statement that individuals can handle more cards than groups surprised me, although on reflection it confirms my own experiences.
I did find myself not entirely in agreement with the author on some points, and at times found her advice overly prescriptive. For example, the instruction not to mix function and content in a card sort is one that I would dispute, although her underlying concern was a valid one.
Purists may feel that the books is not supported by sufficient theory or supporting references, but given the strong practitioner focus, I did not find this to be a weakness.
The comprehensive use of real-life case studies to illustrate concepts and practices and the frank admissions of instances where the author (and others) “got it wrong,” provide a way to learn that is almost as good as making one’s own mistakes.
Minor quibbles aside, this book is a must-read for anyone who uses, or intends to use, card sorting.
Use the code “UXMAG” for a 15 percent discount on this book or any other item at http://rosenfeldmedia.com