I still remember the first time I observed a concert, I stood by enraptured, watching users group little cards with items found in a grocery store, bread, shampoo, pet food crackers into these little piles. Some users made groups that were just what I would have done, while others created categories that were delightfully surprising to me. It blew my mind this simple technique for uncovering how target users think about an information space. And I was hooked that first time was a traditional paper based card. Sorry, but what if you can’t do a card or in person? What if you need to do a card search remotely? Well, you can, and it really isn’t terribly different. The basic formula is the same as with traditional paper sorting. You, the researcher, choose a set of topics. These are 40 60 items that represent the main content on the site and that are all roughly on equal hierarchical levels. So broccoli and carrots, not broccoli and vegetables. You make sure you vary the terms and the order on the cards to prevent users from keyword matching. So instead of party invitations, party favors and party decorations, you might try invitations, party favors and decorations. Choose your card sorting software, instead of putting those topics onto paper cards, you’re going to load them onto digital cards using the tool that you’ve chosen. So there’s two main options, a specialized card sorting tool, which is designed specifically for this purpose or a DIY do it yourself approach.
Specialized tools are the most common and also the easiest to set up and to use for analysis. They involve a drag and drop interface where users will move cards from one column into a work area where they can then group the cards and eventually create labels. Now, the real place that these tools shine is in their analysis capabilities, which show data and the form of similarity matrices. Which percentage of participants grouped any two cards together. They also show dende diagrams which help you see popular groups of cards. These tools have a big advantage because the calculations and summaries are available immediately after you finish collecting your data, unlike with paper cards where you have to manually enter in all of those cards getting results. The second option is your DIY approach. Now, this is a good fallback. If you absolutely cannot access a specialized tool in this approach, you pick any online tool that will allow you to do some basic affinity mapping, moving virtual stickies around. Then to do the quantitative analysis, you can find tutorials online to help you make those calculations using your standard spreadsheet applications. Next, just like in-person studies, you recruit your target users. Now it’s essential that they reflect your typical users, the same knowledge level and familiarity with your topics and your content as people who would actually be using your website or product. So you invite them to a virtual meeting where they can then share their screen and ideally a camera so that you can see their facial expressions as they’re working and you direct them to your card sorting tool where they can then begin the task. Now, when I introduce a task, you know, similar to in person, I’ll explain to them, OK, one by one, you’re going to take each of these items and just begin to organize them into groups that make sense to you. If you’re not sure about what a card is or where to put it, just leave that off to the side, create a separate area. Then I explain. I’m going to sit back and let you do the work on your own. I’ll mute myself, but I’ll still be able to hear you. So if you have a question or you won’t just let me know when you’re done, we’ll we’ll talk afterwards. Now, when the users finished, this is when you get to debrief, you get to ask them questions. You know, I’ll often just start very open ended and say, tell me about how you created these groups. Why did you decide to put some items together? Were there some items that seemed like they could belong in two places or more? Once you’re finished with all of your questions, you can thank the user and, you know, tell them how they’re going to receive their incentive. Typically in a remote session, this might be an online gift card. Now, you might be thinking, wow, this all sounds great. Why would anybody do paper? Well, remote card sorting does have one big drawback or one risk, which is that the usability of the tool can impact the success of the session.
For users, sorting digitally is harder than sorting paper cards. And if you think about it, wouldn’t you rather deal out a deck of 52 cards than drag and drop 52 items? Some software, it doesn’t work as well on certain devices or it gets buggy in the browser or there’s inline scrolling. That makes it hard for users to see, you know, all of the groups that they’ve created digitally. And all of this can mean that users are more likely to get fatigued in this digital environment. And if that happens, they might end up sorting cards just to be done with working in the interface. And lastly, you might wonder, well, can we do remote unmoderated card sorting where there’s no facilitator trainee? Can I recommend first you get that qualitative feedback with those de-brief so you can understand a little bit more about why people are grouping things together and then do some unmoderated sessions by adding those unmoderated sessions. You can get faster feedback and a little bit more data. You know, just a larger number of sessions run without necessarily using the researcher time. So there you have it. If you’re thinking about remote card sorting, it’s not too different from traditional paper based sorting. Pick your terms, pick your tool and your typical users and you’re on your way.