Whether you’re designing a new website or redesigning an existing one, card sorting is a quick, reliable and inexpensive research tool that can significantly improve your information architecture. By improving your information architecture, you’re giving yourself the best chance at delivering a product that is accessible, usable and relevant.

So, what exactly is card sorting? In short, card sorting is a user research technique that helps you discover how people understand and categorize information. Since great information architecture is built on the premise of organizing and categorizing information, card sorting is a secret weapon for website and digital product designers around the world. Actually, the tool is super common, and for good reason.

In this article we’ll help you prepare and conduct card sorting research. We’ll also help you make sense of the data you find and how to apply it to design great information architecture.

Planning and Preparation

Card sorting delivers the best results when you clearly define your goals. The narrower your scope, the more insightful and practical your results will be. It’s important to focus on one goal at a time when planning a card sorting study. What part of your information structure do you need clarification on? Organizing FAQ’s, product categories in an online store, or submenus, are common examples of card sorting projects.

Next, how best can you feasibly recruit participants? Depending on your situation, you may prefer conducting remote card sorts or in-person. Card sorts in person allow you to read body language and you may be more comfortable asking qualitative “why” questions of your participants. Whereas the benefits of remote card sorts, like OptimalSort, is that you aren’t constrained by location or time - just set it up, share the link with participants, then quickly analyze the results. In either case, be sure to recruit participants that represent the demographics of your intended users.

The next step is to prepare the cards themselves. The cards will represent the elements/topics that you wish to organize. Typically, you should aim for between 30 and 50 cards in order to get enough useful data. It also forces you to include only the most relevant cards. Additionally, they should also be on the same conceptual level to avoid confusion and ambiguity.

Finally, decide if you’re asking participants to group the cards based on categories that you decide (closed card sorting), or if participants will be able to create their own groups for cards (open card sorting). You can also facilitate hybrid card sorting which starts off as a closed card sort, but gives participants the option to create additional categories themselves. When you’re deciding, think about your task list (how you’re asking using to sort the cards) and how open-ended you’re prepared for the answers to be. Closed card sorting will narrow your results, whereas open will broaden your results.

Conducting the Session

Now that you’ve done the preparation, it’s time for the fun part! How involved you’ll be depends on whether you’re conducting remote or in-person sessions. We’ll discuss in-person card sorting first, then we’ll point out how remote card sorting differs.

An overview of conducting in-person card sorts:

1. First, shuffle the cards and give them to your participant(s). Ask them to look at each card, then direct them to either organize into groups on their own (open) or into the groups you have provided (closed). It’s important to emphasize to the participant that there are no right or wrong answers. Remember, you’re looking for a real, unfiltered insight into how people organize your information. You can even ask them to think out loud while they’re sorting the cards to gain additional, qualitative insight. One benefit of group sessions is that they usually do this anyway via natural discussion.
2. Then, if you’re running an open card sort, ask your participant(s) to name the groups they have organized. This will help you to understand the rationale behind their decisions and will give you some pointers when you come to labeling information architecture.
3. Once the session is complete, ask participants some open questions. Did you find any cards difficult to place? Did some overlap? Were any left out entirely? This sort of questioning, along with your notes throughout the session, will prove invaluable when you come to analyze the results.
4. Carefully collect the cards and make a record of the groups - there’s nothing worse than clearing the table and messing up the cards before you do this!

Remote card sorts differ from remote sessions in that once you’ve set up the cards in a tool like OptimalSort, you’re good to go. No printing, no shuffling, no resetting. You simply send a link to your participants and ask them to complete the task within a defined timeframe. Online card sorts are generally quicker and less time consuming in this respect, and they may allow you to find more participants and therefore more data.

There are two key things to highlight when running a remote card sort session. Firstly, ensure your instructions are clear and concise. Unlike an in-person session you won’t get the opportunity to clarify any misunderstandings. Secondly, you may consider a follow up questionnaire to gather additional qualitative insights. Check out this facilitation guide for more pointers on remote card sorting.

Analyzing and Interpreting the Results

Now that you’ve got all of your juicy data, it’s time to analyze it! If you ran a remote card sort, there will be some manual processing of your results (usually translating data to excel) which can be time consuming, whereas online tools will generally have analysis tools built right in. This is great for getting quick insights and quick development of information architecture.

When analyzing results, you’re really looking for patterns by identifying similar groups and labels. Using a tool like OptimalSort, for example, you’ll be provided with a few reports that will help you identify patterns and themes:

• Participants Table: Review all of the people who took part in your card sort and segment or exclude them.
• Participant-Centric Analysis (PCA): See the most popular grouping strategies as well as the alternatives among those people who disagreed with the first strategy.
• Dendrograms: Quickly spot popular groups of cards and get a sense of how similar or different your participant’s card sorts were.

Strong patterns or themes that emerge from the data tell us that participants understood categories in a similar way. On the flipside, different or disperse patterns tell us that there was no clear consensus on how information should be categorized. Both insights support effective design of information architecture. The goal is to find common ground in order to create seamless user experiences.

So far we’ve discussed statistical analysis which is all about the hard numbers. But it’s important to infuse some of the qualitative data into your reporting too. If you find that there is confusion within your results or no clear themes, you need to understand why. This is where the interpretation of questionnaire feedback or notes from in-person sessions become so valuable.

Using a combination of your data and your insights, it’s helpful to pull a summary together of your findings in a report. This can be shared with the wider team who have influence on the design of information architecture. Check out this analysis guide for more information on interpreting your results.

Conclusion

Card sorting is a fairly quick and straightforward way to inform information architecture design. It allows us to put the user at the center of our decisions surrounding the categorisation and grouping of information. Why is this important? Because as designers we can often assume how things should be organized. It’s too easy to be influenced by internal factors, like organization structures and the status quo. Don’t fall into this trap - use card sorting to gather clear, unbiased feedback on your information architecture.

Effective card sorting has clear objectives and is best suited to answering specific, information-related questions. We recommend using it when you need clarification around specific information structure, such navigation, menus and product categorisation.

As we’ve discussed, there are a few different approaches to card sorting research. They all have their place, so hoose which one best suits your needs. There’s a lot of resources available if you want to learn more. A good place to start is our card sorting 101 article. Good luck and happy researching!

This morning Ieavesdroppeda conversation between Amy Worley (@worleygirl) and The SemanticWill™ (@semanticwill) on "the twitters".Aside from recommending two books by Donna Spencer (@maadonna), I asked Nicole Kaufmann, one of the friendly consultants at Optimal Usability, if she had any advice for Amy about reorganising 404 books into categories that make more sense.I don't know Amy's email address and this is much too long for a tweet. In any case I thought it might be helpful for someone else too so here's what Nicole had to say:In general I would recommend having at least three sources of information (e.g. 1x analytics + 1 open card sort + 1 tree test, or 2 card sorts + 1 tree test) in order to come up with a useful and reliable categorisation structure.Here are four options for how you could consider approaching it (starting with my most preferred to least preferred):

Option A

• Pick the 20-25 cards you think will be the most difficult and 20-25 cards that you think will be the easiest to sort and test those in one open card sort.
• Based on the results create one or two sets of categories structures which you can test in a one or two closed card sorts. Consider replacing about half of the tested cards with new ones.
• Based on the results of those two rounds of card sorting, create a categorisation structure and pick a set of difficult cards which you can turn into tasks which you can test in a tree test.
• Plus: Categorisation is revised between studies. Relative easy analysis.
• Minus: Not all cards have been tested. Depending on the number of studies needs about 80-110 participants. Time intensive.

Option B

• Pick the 20-25 cards you think will be the most difficult and 20-25 cards that you think will be the easiest to sort and test those in one open card sort.
• Based on the results do a closed card sort(s) excluding the easiest cards and adding some new cards which haven't been tested before.
• Plus: Card sort with reasonable number of cards, only 40-60 participants needed, quick to analyse.
• Minus: Potential bias and misleading results if the wrong cards are picked.

Option C

• Create your own top level categories (5-8) (could be based on a card sort) and assign cards to these categories, then pick random cards within those categories and set up a card sort for each (5-8).
• Based on the results create a categorisation structure and a set of task which will be tested in a tree test.
• Plus: Limited set of card sorts with reasonable number of cards, quick to analyse. Several sorts for comparison.
• Minus: Potential bias and misleading results if the wrong top categories are picked. Potentially different categorisation schemes/approaches for each card sort, making them hard to combine into one solid categorisation structure.

Option D

• Approach: Put all 404 cards into 1 open card sort, showing each participant only 40-50 cards.
• Plus: All cards will have been tested
• Do a follow up card sort with the most difficult and easiest cards (similar to option B).
• Minus: You need at least 200-300 completed responses to get reasonable results. Depending on your participant sources it may take ages to get that many participants.

On the fence about card sorting and why you should be using it to improve your user experience? Let’s take a look at why you should take advantage of this powerful user research method.

Simply put, card sorting can help you discover how your users think your content should be organized and categorized. Card sorting gives you insight into how people conceptualize, group and label ideas, enabling you to make confident, informed information architecture (IA) decisions.

What is card sorting?

In a card sort, participants sort cards containing different items into groups. You can use the results to figure out how to group and label the information on your website in a way that makes the most sense to your audience. 

Using card sorting at the start of your website build means you’re able to make decisions based on data, not assumptions. Being informed at the start of your website build can mean saving a lot of time later with revisions or rebuilds. Better to build something intuitive now, than be left wondering later why parts of your website aren’t working as you expect.

When should you use card sorting?

It’s best to do card sort research when you want to answer a specific, information-related question. For example, maybe you’re adding a new range of “natural products” to your Health and Beauty site. On the other hand, you may want to redesign how information is grouped together across your entire website.

Card sorting is at its most effective when you’ve got the information and detail you need but you just need guidance on how it's best (most intuitive) to organize it.

While card sorting is typically used in the early stages of the design process, when there’s no live IA, it’s also common to use the technique to make changes to a live IA down the line. 

Card sorting: A powerful way to understand users’ mental models

Card sorting is a powerful tool to understand your users and how they make sense of information when they arrive on your website. A good rule to keep in mind is that what makes sense to you and your colleagues may not make sense to your users.

Using an online card sorting tool like OptimalSort it can be useful to check in with users to understand where they think information should sit on your website. As product ranges increase or change over time, it can also be useful to undertake card sort research when updating your website. 

Let’s take a look at how card sorting might work for an e-commerce website. 

Imagine you run a health and beauty e-commerce business with an active and successful website with a vast range of products that can be grouped in many different ways. At worst the website is clunky and hard to search, making it difficult for shoppers to find the right product, quickly. At best everything is there but it doesn’t quite answer what the user is looking for. It can be incredibly powerful to have a fuller understanding of how our website is viewed from our shoppers (rather than just internally). The goal of our website should be to showcase our products in a way that makes shopping easy, quick and even intuitive.

We are introducing a full range of natural based products that include products intended for babies, children, women and men. These products have previously  been categorized by who they are intended for. But we want to know if there is a better way that these could be made available, especially with a market shift to an increased demand in  natural based products.

By doing some card sorting with OptimalSort, we gather data from users and the pattern that our audience use to group these products. Through the data analysis we have discovered that a large majority of our users would group by natural products first and then by who the product is intended for (baby, children, women, or men). Armed with this insight (amongst others) we can use it to influence our IA. Ultimately, we end up with a far more intuitive and streamlined user experience (UX).

Three ways to use card sorting

Did you know that there are multiple ways to use OptimalSort card sorting? Let’s take a look– you may be surprised.

1. Building a new website

This is by far the biggest use-case for card sorting. When looking at building a new website or making better use of an existing one, utilizing card sorting at the research stage can be insightful and informative. Seeing your website, products and/or navigation from your end user’s perspective can enlighten, inform and assist in creating an enhanced user experience.

2. Combine card sorting and tree testing

When combined with card sorting, tree testing, with Treejack, can help you to improve your navigation and give you a fuller understanding of how your website is used. Tree testing is a technique for evaluating the findability of topics on a website. It’s also commonly known as reverse card sorting and is the perfect technique to complement card sorting. After you’ve analyzed your card sorting results and transformed them into a draft IA, you can test these insights using a tree test. Using this technique, you task users with seeking as opposed to sorting. This technique aims to replicate the experience of using a website – without visual distractions.

Unlike usability testing, tree testing only focuses on the IA of your website. It makes the process of developing an IA much faster, as you can easily make refinements and tweaks without needing to get bogged down in costly redesigns. 

3. Make collaborative design decisions

You can use OptimalSort to get your team involved and let their feedback feed your designs — logos, icons, banners, images, the list goes on. By creating a closed image sort with categories where your team can group designs based on their preferences, you can get some quick feedback to help you figure out where you should focus your efforts.

Use OptimalSort to run your first card sort

Card sorting can take place in person, or online with a tool like OptimalSort. OptimalSort gives you the flexibility to conduct moderated and unmoderated card sorts online. Now, you can collect the data you need, how and when you need it. Plus it only takes a few minutes to design and launch your study. 

Not only is OptimalSort simple to use,  but it’s backed up with the strength of powerful analysis functionality. Taking the pain of trawling screeds of information, OptimalSort pulls out useful, usable insights from your card sorting data. This allows you to quickly identify common groups at a glance with comprehensive and vibrant visualizations and use this data to support design changes and recommendations. 

What is 3D Cluster View?

Part of our OptimalSort analysis is the 3D Cluster View (3DCV). While the addition of ‘3D’ may throw off red flags of being a gimmick, it’s actually entirely appropriate. 

The 3DCV basically allows you to visualize the similarity between cards as three-dimensional spatial relationships. Each point in the 3D visualization represents one of the cards from your original sort. Cards that are closer together were more frequently sorted into the same category. Likewise, when you see 2 cards that are quite far apart, they weren’t sorted together as frequently. If you’d like to find out more take a look at OptimalSort 3DCV, we think it’s pretty clever.

Wrap Up

If you’re now interested in a card sort of your own, we obviously recommend OptimalSort (which you can get started with for free). Or you want to find out more, take a look at our Card Sorting 101.

Happy sorting!