Put your hands up who remembers their first card sort? Of course you do, because it probably went horribly wrong!
Because card sorting (like many user-centered design and research tools) is grounded in common sense, it’s hard to imagine what could possibly go wrong when asking people to sort a pile of cards into sensible groups.
But if you’ve ever run a card sort before, you’ll know that although the principles behind it are simple, it’s the thoughtful details in preparation, execution, and analysis that will make or break a card sort.
A real life example: My first card sort was for a website with about 1,000 pages. We enticed friends and family to come in for an evening with drinks and pizza and had teams of people sort a subset of content into groups. Each team came up with more or less the same result – which coincidentally closely matched the existing site structure.
Our management were relieved – we must already have the best structure, so there was no need to look at rearranging content. I left the exercise not convinced with this, but at the same time confused since “how could the cards lie?”
Unfortunately I didn’t have Donna’s book to help answer that question, but on reflection realised that the card labels I had used were introducing a strong bias in suggesting groups that matched the existing site.
You could learn by trial-and-error (like me), or piece together snippets of many useful (although sometimes contradictory) articles and posts on the internet (BTW some of the best are also written by Donna), but there is nothing that gives such a complete, and practical reference to every step of running a card sort than Card Sorting by Donna Spencer.
The book is divided into three main sections, covering preparation, execution, and analysis of a card sorting activity.
A nice touch is the two page FAQs before even the contents that gives quick answers to some common card sorting questions (like “should I let people put cards in more than one place?”) and send you deeper into the book for more in-depth information.
The best chapters were in the ‘preparation’ and ‘execution’ sections, which make up the bulk of the book. Donna writes with confidence and clarity, and also humility – several anecdotes of card sorts that haven’t run quite perfectly are provided, these serve as good learning devices by pointing out common mistakes, but also remind readers who may have had negative experiences with card sorting in the past to keep on persevering.
My tip for making the cards that isn’t included in the book: when you’re numbering the cards write the number on the back of the card instead of on the label with the card title. I have a deep set fear that the numbers may somehow influence the groups that people make.
The last chapters (which cover analysis of data collected from card sorting) are good at explaining a spreadsheet-based approach that Donna developed and uses, but start to get a little vague in explaining some other methods of analysis where the focus changes from practical advice to more of a summary of different methods that are available.
These other methods could have been expanded into much more detail (maybe Donna has a sequel planned – “Advanced Card Sorting” ;-), or perhaps could have just been included as an appendix. If you just stick with the spreadsheet method and ignore the other stuff you will probably be fine.
There are heaps of gems of useful advice and ideas in this book, and as I was reading I was happily agreeing to myself “wow, this is so great stuff” but at the same time some ideas aren’t always spelt out. My recommendation would be to read this book with your highlighter and marking bits that stand out, then after each card sort dive in again because some things will become clearer in retrospect (doesn’t everything?).
Recently I’ve been experimenting with using card sorting for goals outside of content organising, specifically for running requirements prioritisation workshops. I’m sure other people have found other uses of applying card sorting and I would have liked to have seen some discussion of this also included in the book.
Another area to might have been to walk people more specifically through examples of how analysing data, complemented with other research, had influenced specific content organisation decisions, but perhaps this was intentionally omitted because including this might have lead people to try and inappropriately apply these situations to their own projects.
But to wrap up (and slight repeat): This is a great book for someone tasked with organising content on a website who is either unfamiliar or unconfident with card sorting. Read this book and absorb it’s advice and you’ll avoid countless mistakes that people often face with card sorting!
For people who feel they are confident with card sorting (like me), you might also get some value it might help give clarity in discussing card sorting with your client or stakeholders, or in my case I’m glad to say it’s lead me to question ideas that I’d previously taken for granted, for example this book has helped challenge my opinion of team based card sorts.
Many successful card sorts to you!
You can buy Card Sorting from the publisher, Rosenfeld Media, in digital and print editions.
Update: Use the promo code MSANDERS to receive a 20% discount off Card Sorting, and all other products at www.rosenfeldmedia.com.
And of course you can buy a print edition from Amazon.
Tags: Book Review, Card Sorting
Update 2: This morning I was writing a peer review of a research study that was using a closed card sort activity to test how easy it was for people to find information in different categories.
As well as explaining why this approach isn’t really testing how easy it is for people to find content, it was great being able to reference Card Sorting to back up my justification.
So that’s is another reason for experienced practitioners to be familiar with the contents of Card Sorting – when it comes to convincing people, it helps to have a published book to back you up!
It’s a little disappointing that the analysis chapters got short-changed. For me that’s the key part.
Hi Eva, thanks for your comment!
It’s true that the analysis chapters are brief compared to the preparation and execution sections, but having said that, they do provide a good amount of detail around the spreadsheet approach that Donna uses.
A few years back I had some interesting discussions with a colleague. At the time I was pretty much convinced that the best approach to card sorting was to run remote evaluations, so that you could run more card sorts to get enough data to run statistical models. He was the opposite, and argued that the time you had one-to-one with the participant was the most valuable, and that he rarely analysed the raw results, instead taking notes he’d make at the time of the card sort.
Although he didn’t convince me to that extreme (I’m still a sucker for card-pair analysis) I’d like to think that I’ve reached a happy medium where the process of running card sorts are just as valuable as the actual results.