How to Use Card Sorting to Better Understand Your Users

Are you confident that your website‘s navigation makes sense to your users? Can they easily find what they need or are they getting lost and frustrated?

If you‘re not sure, it‘s time to consider conducting a card sorting study. Card sorting is a powerful user research technique that reveals how your users think your content should be organized and labeled.

By watching users sort topics into groups, you‘ll uncover their mental models and expectations for your site‘s information architecture. You can then use these insights to build a site structure that aligns with users‘ natural ways of thinking. The result? Greatly improved navigation usability and findability.

In this comprehensive guide, we‘ll walk you through everything you need to know to run an effective card sort, from choosing the right methodology to analyzing the results. Plus, we‘ll recommend some of the best tools on the market to streamline the process.

Whether you‘re building a site from scratch, restructuring an existing one, or just want to validate your navigation labels, card sorting is an invaluable technique to have in your UX toolbox. Let‘s dive in.

Card Sorting 101

Before we get into the nitty-gritty, let‘s cover the basics. Card sorting is a research method where participants organize topics into groups that make sense to them. The topics are written on cards (either physical index cards or virtual cards in an online tool) and participants sort them into piles or categories.

There are three main types of card sorting:

  1. Open card sorting: Participants create their own category names and sort topics into those categories. This helps uncover users‘ natural mental models for how information should be grouped.

  2. Closed card sorting: Participants sort topics into predefined categories provided by the researcher. This is useful for testing how well a proposed structure resonates with users.

  3. Hybrid card sorting: A combination where participants sort into predefined categories but can also create their own if needed.

The beauty of card sorting is its simplicity. It‘s an easy, cost-effective way to gather insights from real users at scale. And the results can drastically improve your site‘s navigation usability and overall user experience.

The Power of Card Sorting: Real-World Examples & Results

To illustrate the impact card sorting can have, let‘s look at some real-world examples and statistics:

  • The financial services company State Farm conducted an open card sort with 50 participants before redesigning their website. The results revealed that users thought about the company‘s products very differently than how they were categorized on the current site. By restructuring the site‘s IA based on the card sort findings, State Farm saw a 42% increase in policy purchases and a 15% boost in customer satisfaction scores.

  • Walmart used closed card sorting to test their proposed e-commerce site navigation labels with over 200 shoppers. The results showed that the label "Home" was ambiguous to users – some expected it to contain furniture while others thought it meant the home page. Based on this feedback, Walmart renamed the category to "Home & Furniture" and saw a 25% increase in clickthrough to that section.

  • The US General Services Administration ran a 20-person open card sort before overhauling the IA of their USA.gov site. They found that users clustered topics differently than the existing site structure and used simpler category labels. After implementing the changes, USA.gov‘s bounce rates decreased by 35% and navigation-related support requests dropped by 50%.

These examples show the tangible, bottom-line results that card sorting can deliver. By aligning your site‘s structure and labels with users‘ expectations, you can dramatically improve usability, engagement, and conversions.

How to Run an Effective Card Sort

Now that you‘re sold on the benefits, let‘s walk through how to actually conduct a card sort study:

1. Define your objectives.

Are you trying to design a new site structure, validate proposed navigation labels, or identify areas of confusion in your current IA? Get specific on what questions you need the card sort to answer.

2. Choose the right type of sort.

Based on your objectives, determine if an open, closed or hybrid sort makes the most sense:

  • If you want to discover users‘ natural way of categorizing content, go with an open sort.
  • If you want to see how well a proposed IA resonates, use a closed sort.
  • If you have a draft structure but want to allow some flexibility, do a hybrid sort.

3. Select your cards.

Create a list of 30-60 key topics or pages from your site. Write each on a separate index card or in an online card sorting tool. Be sure to use clear, unambiguous labels that users will understand.

4. Recruit participants.

Look for 15-20 participants per card sort type (open and closed). Aim for a representative sample of your target users – consider demographics, familiarity with the subject matter, and web savviness. If your user base is large and diverse, you may need more participants to capture all perspectives.

5. Run the study.

If doing an in-person sort, give participants the stack of cards and have them sort into piles that make sense to them. For open sorts, have them label each pile. For closed sorts, provide the category labels. Encourage participants to think out loud as they work so you can probe on their rationale.

If using an online tool, send participants a link to the study. Include clear instructions on what they need to do. Most tools will automatically capture the time taken, category labels, and any comments from participants.

6. Analyze the results.

For open sorts, look for common category labels and groupings across participants. Create a standardized set of categories and tally how many participants put each card in each category. Note any outliers or disagreements.

For closed sorts, measure the percentage of agreement for each card‘s placement. So if 18 out of 20 participants put "Product Returns" under a "Customer Service" category, that‘s a 90% agreement rate. Also note any unexpected or inconsistent placements.

Look for patterns in the data – where is there consensus versus disagreement? What categories were used most often? What new ideas emerged?

7. Translate findings into action.

Use your analysis to inform your site‘s IA and navigation design. But don‘t treat it as gospel – complement card sorting with data from analytics, user interviews, and usability testing to get a complete picture.

Some tips for applying card sort results:

  • Use the card sort to generate ideas and a "first draft" for your site map, then refine and validate with additional research.
  • For open sorts, aim for 5-9 primary categories based on the most common groupings.
  • Choose clear, specific, and differentiating labels for your categories based on user terminology.
  • If a card was sorted in multiple places, consider cross-listing it or using clearer labels.
  • Run a closed sort or tree test to validate your proposed structure before implementing.

Best Practices For Effective Card Sorting

To ensure your card sort delivers reliable, actionable insights, keep these best practices in mind:

  1. Keep the number of cards manageable. 30-60 cards is ideal. Too few and you won‘t have enough data points. Too many and you‘ll overwhelm participants.

  2. Use clear, jargon-free labels. Write card labels in plain language that users will unambiguously understand. Avoid brand or internal terminology.

  3. Provide clear instructions. Explain succinctly what you want participants to do and what kind of feedback you‘re looking for. Do a practice round if needed.

  4. Encourage thinking out loud. Hearing participants‘ gut reactions and rationale is often more insightful than the final card placements.

  5. Be strategic in your recruiting. Aim for a representative sample of your core user types. Don‘t just grab whoever is convenient.

  6. Consider context of use. Will your users typically be navigating your site on desktop or mobile? Looking to complete specific tasks or casually browsing? Tailor the instructions and cards to match.

  7. Combine with other user research methods. Card sorting is most powerful when combined with methods like user interviews, tree testing, A/B testing, and web analytics.

Recommended Card Sorting Tools

While you can run card sorts with physical cards, online tools offer significant advantages, especially for remote participants and large sample sizes. Here are our top picks:

  1. KardSort – A free, open-source online card sorting tool with a simple drag-and-drop interface. Results are exported in CSV format for analysis in Excel or any data viz tool of your choice.

    Price: Free

  2. OptimalSort – Offers open, closed, and hybrid sorting with real-time results and rich analytics like dendrograms and participant-centric analysis. Integrates with their wider suite of user research tools.

    Price: $99/month for individuals, $166/month for teams

  3. UsabiliTEST – An enterprise-grade tool with a slick UI for participants and a range of analysis features like similarity matrices, cluster analysis, and filtering by demographics.

    Price: Starts at $199/month

  4. UserZoom – An advanced tool with participant recruitment services and a range of UX research capabilities beyond card sorting.

    Price: Custom pricing, request a demo

When choosing a tool, consider:

  • Ease of use for both researchers and participants
  • Range of analysis and visualization options
  • Integrations with your existing UX research workflow
  • Scalability and security for your organization‘s needs
  • Pricing and support

Go Forth & Sort

We‘ve covered a lot of ground in this guide, but the most important takeaway is this:

Card sorting is a UX research power-up that helps you build more intuitive, user-friendly site structures.

By conducting even a small card sort study, you can glean invaluable insights into your users‘ mental models and expectations. You can use these learnings to architect an information hierarchy that aligns with users‘ natural way of thinking.

And the payoff is huge – improved navigation usability, better engagement, more conversions, and happier users. All from a simple exercise of sorting cards.

So what are you waiting for? Pick a card sorting tool, grab some (virtual) index cards, and start uncovering how your users really think about your content. Your site navigation – and your bottom line – will thank you.

Similar Posts