How to organize website navigation for better user experience

Your website's navigation is the first thing visitors rely on to find what they need. If it doesn't match the way they think, they leave. Card sorting is a simple research technique that shows you how real people would organize your content -- so you can build navigation around their expectations instead of guessing.

Difficulty: Intermediate Time Required: 2-3 hours

Most websites organize navigation around internal team structures or whatever made sense to the person who built the site. The problem? Your visitors don't think that way. Card sorting flips the process: you let actual users group your content into categories that feel logical to them, then use those patterns to reshape your navigation. It's one of the most practical UX research methods out there, and you can do it in an afternoon.

What You'll Need

• List of your website content/pages
• Sticky notes or a digital card sorting tool
• 5-10 test participants (ideally from your target audience)
• Spreadsheet for analyzing results
• CardSort account (recommended for remote testing)
• Pen and paper for sketching navigation layouts

Step 1: Audit Your Current Content

Time: 30 minutes

Before you can reorganize anything, you need to know what you're working with. Go through your entire site and list every page that needs a home in your navigation.

Pull up your analytics and note which pages get the most traffic. Look at your existing categories and ask yourself whether they still make sense. Flag any content that's outdated or redundant -- this is a good time to clean house. Put everything in a spreadsheet with the page title, URL, and a short description of what the page covers. If you run an e-commerce site, you'll want to think about product categories, policy pages, support content, and informational pages separately since they serve different purposes.

Step 2: Create Cards for Each Content Item

Time: 20-30 minutes

Now turn your content inventory into cards. Each card should represent one piece of content with a short, clear label -- no jargon.

Aim for 30-50 cards total. Going much beyond that tends to wear out participants, and you'll get lower quality results. If a page title isn't self-explanatory, add a one-line description. For digital card sorts, enter these into your tool of choice. Think labels like "Size Guide," "Summer Collection," "Customer Reviews," or "Payment Methods" -- specific enough that someone unfamiliar with your site can make a quick judgment about where it belongs.

Step 3: Choose Your Card Sort Method

Time: 15 minutes

There are three ways to run a card sort, and the right one depends on where you are in the process.

• Open card sort -- Participants create their own category names and group cards however they want. This is the best choice when you're starting from scratch or doing a full redesign, because it shows you how people naturally think about your content.
• Closed card sort -- You give participants a set of predefined categories, and they assign cards to those categories. Use this when you have an existing structure you want to validate.
• Hybrid card sort -- You provide some categories but let participants add their own. This works well when you're tweaking an existing site and want to see if people would add or change anything.

If you're unsure, start with an open sort. You'll learn more.

Step 4: Recruit and Brief Participants

Time: 30 minutes

You'll want 5-10 people who represent your actual audience. That might mean a mix of ages, technical comfort levels, and familiarity with your product. Five participants will usually surface most of the major patterns; going beyond ten rarely adds much new insight.

Send clear instructions about what you're asking them to do. Each session should take 15-30 minutes. If you're running this remotely, set up your study in a tool like CardSort. Give participants a simple brief: "We're rethinking how our website is organized and want to understand how you'd group these pages. There are no right or wrong answers -- just go with what feels natural to you."

Step 5: Conduct the Card Sort Sessions

Time: 2-3 hours (depending on participant count)

This is where the real data comes in. Whether you're running sessions in person or remotely, encourage participants to think out loud as they sort. Hearing why someone puts "Returns Policy" next to "FAQs" is often more valuable than the grouping itself.

Take notes on hesitations and unexpected choices. For in-person sessions, photograph the final arrangements. Remote tools like CardSort collect results automatically, which saves a lot of time. Pay special attention when someone groups things in a way you wouldn't have predicted -- those moments often reveal the biggest gaps between your assumptions and your users' reality.

Step 6: Analyze the Results

Time: 30-45 minutes

Look for items that most participants grouped together. If the majority of people put two items in the same category, that's a strong signal they belong near each other in your navigation.

Build a similarity matrix showing which items were frequently paired. Notice the category names people used -- their language is usually more intuitive than whatever you'd come up with internally. Also pay attention to items that ended up in wildly different places across participants. Those are the tricky ones, and you might need to make them accessible from multiple spots in your navigation. For example, if half your participants put "Sizing Guide" under "Help" and the other half put it under "Products," consider linking to it from both.

Step 7: Design Your New Navigation Structure

Time: 30 minutes

Now take everything you've learned and sketch out a new site map. Start with your top-level categories, then fill in the secondary and tertiary levels.

Try to keep your top-level navigation to around five to seven items. More than that and people start to feel overwhelmed. Use labels that match the language your participants used -- not your internal terminology. Think about adding breadcrumbs for deeper pages so users always know where they are. A typical structure might include main categories like "Shop" (with subcategory dropdowns), "Help," "About Us," and "My Account," with the specific groupings inside each one shaped directly by your card sort findings.

Tips and Best Practices

• Use clear labels. Specific and descriptive beats clever or cute every time.
• Limit top-level categories. Five to seven main items is the sweet spot for most sites.
• Be consistent. Navigation should look and behave the same way on every page.
• Think mobile-first. If your structure doesn't work on a phone screen, simplify it.
• Prototype before you build. Test your new navigation with a clickable prototype before committing to development.
• Don't forget search. Some visitors will always prefer typing over browsing -- make sure search is easy to find.
• Run card sorts more than once. Doing a second round after your redesign can catch issues you missed the first time.

Common Mistakes to Avoid

• Organizing by department. Your users don't care about your org chart. Build navigation around their goals, not your team structure.
• Using jargon. If your audience wouldn't use a term in everyday conversation, don't put it in your nav.
• Overcomplicating things. When in doubt, go simpler. A deep menu with clear top-level labels beats a sprawling mega-menu.
• Ignoring mobile. Always check that your navigation works on small screens.
• Treating one card sort as the final answer. Card sorting gives you a great starting point, but you should still validate with usability testing.
• Stuffing too many items at the top level. Every item you add to the main nav makes every other item harder to find.

Next Steps

• What is Card Sorting? Complete Guide
• Navigation (UX Glossary)
• Information Architecture (UX Glossary)
• Card Sorting Information Architecture

Once your new navigation is live, set up analytics to track how people actually use it. Watch for pages with high exit rates or navigation paths that seem roundabout -- those are signs something still isn't clicking. Plan to revisit your structure every six to twelve months, especially if you're adding a lot of new content. And if you have the traffic for it, A/B testing different labels or category arrangements can help you fine-tune things over time.

Frequently Asked Questions

How many participants do I need for reliable card sorting results? Somewhere between five and ten is the usual recommendation. With five people, you'll catch most of the major patterns in how users think about your content. Adding a few more participants helps fill in the edges, but beyond ten you'll mostly see the same groupings repeated.

What's the difference between open and closed card sorting methods? With an open sort, participants make up their own categories and group cards however they want -- it's great for uncovering how people naturally think about your content. A closed sort gives them preset categories to sort into, which is more useful when you already have a structure and want to see if it holds up.

How do I analyze conflicting results when participants organize content differently? Start with the areas where people agree. Items that most participants grouped together are your safe bets. For the contentious ones, look at the reasoning behind different placements. Sometimes the answer is to make a piece of content reachable from more than one place in your navigation, or to do a quick follow-up conversation to understand why people disagreed.

Should I conduct card sorting for mobile and desktop navigation separately? Usually not. Card sorting is about how you organize your content, not how it's displayed on a particular screen size. The groupings you discover apply regardless of device. How you present those groupings visually on mobile versus desktop is a separate design question you'll tackle afterward.

How often should I repeat card sorting studies for my website? A good rule of thumb is every twelve to eighteen months, or whenever you're making a big addition to your site's content. If you start hearing from users that they can't find things, or if your business has shifted direction, that's also a good time to run a fresh study.