How to Organize Large Ecommerce Product Catalogs Using Card Sorting

Difficulty: Intermediate Time Required: 4-6 hours (spread across 1-2 weeks)

Here's the thing about product categories: they almost always make more sense to the people who built the store than to the people shopping in it. Your team knows that "Performance Footwear" and "Active Lifestyle Accessories" are different departments. Your customers just want to find running shoes.

Card sorting fixes this. You put product names in front of real shoppers and let them group things however feels natural. The categories they create tell you exactly how your navigation should work. If you're managing more than a couple hundred products, this is one of the most practical ways to build a catalog structure people can actually use.

What You'll Need

• Complete product inventory list (exported from your current system)
• Access to 15-20 target customers or user testers
• Card sorting tool (CardSort recommended)
• Spreadsheet software (Google Sheets or Excel)
• 2-3 internal stakeholders for validation
• Current website analytics (optional but helpful)

Step 1: Audit and Prioritize Your Product Inventory

You can't ask people to sort thousands of products. It would take forever, and the results would be a mess. Instead, pick 200-300 items that represent your full catalog.

Export your product list and build a simple prioritization spreadsheet. Include columns for product name, current category, sales volume, search frequency, and profit margin. You're looking for products that cover the breadth of what you sell while leaning toward your best sellers.

Example: An outdoor gear retailer with 2,000+ SKUs might pick 50 hiking products, 40 camping items, 30 climbing pieces, 25 water sports products, and 35 winter sports items. The mix is proportional to how the catalog actually breaks down by sales, not split evenly.

Get this selection wrong and your whole study suffers. If you over-represent one corner of your catalog, the resulting categories won't translate to the rest of your inventory.

Step 2: Create Representative Product Cards

Now rewrite those product names so they sound like something a customer would search for, not something from your warehouse management system.

Strip out internal codes, excessive brand names, and technical specs that only make sense to your team. Keep each card title under 8 words. Check your site search logs and Google Search Console data to see what language people actually use when looking for these products.

Example: "NK-Trail-Runner-GTX-M-10.5" becomes "Men's Waterproof Trail Running Shoes." "Pro-Series Carbon Fiber Trekking Poles - Adjustable 110-140cm" becomes "Lightweight Adjustable Hiking Poles."

If participants can't tell what a product is from the card title alone, they'll sort it randomly, and that noise pollutes your data.

Step 3: Design Your Card Sorting Study

For an initial catalog reorganization, go with open card sorting. That means participants create their own groups and name them however they want. You're not testing whether your existing categories work—you're discovering how shoppers actually think about your products.

A few practical constraints to keep in mind:

• Cap it at 50-80 cards. More than that and people get tired and sloppy.
• Set a 45-60 minute time limit.
• Write clear instructions. Something like: "Group these products in ways that make sense to you for online shopping. Create category names that would help you find these items quickly on a website."

Add 2-3 screening questions about shopping frequency, familiarity with your product types, and basic demographics. You want 15-20 qualified participants—enough to spot real patterns without blowing your budget.

Step 4: Recruit and Run Your Card Sort Sessions

Who you recruit matters as much as how many. You need actual shoppers from your target audience, not your coworkers. Pull from your customer email list, social media, or a user research recruitment platform.

Once the study is live:

• Send reminder emails after 3-4 days to keep completion rates high
• Offer a reasonable incentive—$15-25 gift cards work well for a 45-minute task
• Watch for outliers: anyone finishing in under 15 minutes probably rushed through it, and anyone taking over 75 minutes may have gotten distracted

Example: A home improvement retailer recruited from their email newsletter, targeting customers who'd bought something in the last six months and had recent DIY project experience. That way, participants actually knew what the products were and could sort them based on real shopping habits.

Step 5: Analyze Results and Identify Category Patterns

Look for products that most participants grouped together consistently. If 60% or more of your participants put the same items in the same pile, that's a strong signal. Below that threshold, the grouping is too unreliable to build on.

Pull your results into a spreadsheet and map each product to its most common groupings. Pay attention to:

• Item pairs that showed up together across 10+ participants
• Products that got scattered all over the place (these might need to live in multiple categories)
• Surprising groupings that differ from your current structure but have strong consensus

Example: Say 14 out of 18 participants grouped "Running Shoes," "Athletic Socks," and "Fitness Tracker" together under names like "Running Gear" or "Cardio Equipment." That's strong agreement, and it tells you those items belong together in your navigation—even if they currently sit in three different departments.

Step 6: Design Your New Category Structure

Take your strongest clusters and shape them into a workable hierarchy. Aim for 5-7 main categories. That's enough to cover a large catalog without overwhelming people in the nav bar.

Within each main category, build out 3-8 subcategories based on your card sort clusters. Try to keep individual categories between 8 and 50 products—too few and the category feels empty, too many and it's hard to browse.

Example structure for sporting goods based on card sorting results:

• Running & Fitness (shoes, apparel, accessories, electronics)
• Outdoor Adventures (hiking, camping, climbing gear)
• Water Sports (swimming, surfing, boating equipment)
• Team Sports (equipment organized by sport type)
• Winter Sports (skiing, snowboarding, cold weather gear)

Before you commit, test your proposed structure against 50-100 products that weren't in the original card sort. If you're constantly struggling to place items, the structure needs reworking.

Step 7: Validate and Refine Your Organization System

Don't skip this step. Your card sort told you how people group products in the abstract. Validation tells you whether the structure actually works for finding things.

Run a closed card sort or a tree test using your proposed categories. Give participants 40-50 new products (ones they haven't seen before) and ask them to place each one in your structure. Or run tree testing tasks like "Find winter hiking boots" or "Locate camping cookware" and watch how people navigate.

You're looking for high success rates on main categories and reasonable success on subcategories. Where people stumble, you'll usually find one of these issues:

• A category that's too broad and needs splitting
• Two similar subcategories that should be merged
• A category name that doesn't match how people think about those products
• Uneven product distribution that makes some sections feel bloated

Fix those problems now, before you rebuild your whole site around a structure that doesn't quite work.

Tips and Best Practices

Stick with customer language everywhere. Check your site search logs and use the words shoppers actually type. Internal jargon and manufacturer terminology make products harder to find, not easier.

Pick category names that won't go stale. "Water Sports" works year-round. "Summer Sports" forces you to rethink your navigation every season.

Plan for products that fit multiple categories. A waterproof jacket could live under "Hiking Gear" or "Rain Gear" or "Winter Apparel." Rather than picking one, use cross-links, featured placements, or tags so people find it no matter where they look.

Think about mobile early. A five-level deep category tree might look fine on desktop, but it's miserable to tap through on a phone. Keep the hierarchy shallow.

Write down your reasoning. Six months from now when someone asks why camping hammocks are under "Sleep Systems" instead of "Camping Furniture," you'll want the card sort data to point to.

Common Mistakes to Avoid

Organizing around your org chart. Just because you have separate buyers for footwear and accessories doesn't mean customers think of them as separate things. Your internal structure is irrelevant to shoppers.

Too many categories too soon. It's tempting to create a category for everything. Resist that urge. Start with broader groupings backed by strong card sort data, then split them later based on how customers actually browse.

Ignoring your search data. Card sorting is powerful, but it's not the whole picture. Cross-reference your results with what people search for on your site. If hundreds of people search "gift ideas" every month, that might deserve a spot in your navigation even if it didn't emerge from the card sort.

Skipping validation. A card sort tells you how people group products mentally. That doesn't guarantee they can find them in a navigation tree. Always test the structure before building it out.

Thinking categories solve everything. Good organization helps, but it's not a replacement for solid filtering, a decent search bar, and smart product recommendations. These systems work together.

Further Reading

• What is Card Sorting? Complete Guide
• Card Sorting (UX Glossary)
• Information Architecture (UX Glossary)
• How To Run Your First Card Sort Study

Frequently Asked Questions

How many participants do I need for reliable card sorting results? For most ecommerce catalogs, 15-20 participants is the sweet spot. That's enough people to see clear grouping patterns without spending a fortune on recruitment. Go below about 12 and you risk missing important patterns. Go above 25 and you're mostly just confirming what you already found.

What's the difference between open and closed card sorting for ecommerce catalogs? With open sorting, participants make up their own groups and labels from scratch. With closed sorting, you give them pre-set categories and they just place items into them. Start with open sorting when you're rethinking your catalog from the ground up—it shows you how customers naturally organize things. Use closed sorting later to test whether your proposed structure actually works.

How often should I reorganize my product catalog using card sorting? A good rule of thumb is every 6-12 months, or whenever you add a significant chunk of new products (say, 20% or more of your catalog). If you notice customers constantly using search instead of browsing, or your support team keeps fielding "where do I find X?" questions, those are signs it's time for a fresh look.

Can card sorting work for very large catalogs with thousands of products? Absolutely. You don't sort the entire catalog—you sort a representative slice of it. Pick 200-300 products that cover your major product types, price ranges, and customer segments. The category patterns that emerge will apply to the rest of your inventory. Just make sure you validate the structure against products that weren't in the original sort.

What metrics should I track after implementing a new category structure? Keep an eye on category page bounce rates, how long people spend browsing, conversion rates by category, and how often people fall back to site search. Also watch your support tickets—if "I can't find X" complaints drop, that's a great sign. Direct customer feedback, even informal, is worth paying attention to as well.