Open vs Closed Card Sorting: Complete Guide

The short version: open card sorting lets participants make up their own categories, and closed card sorting gives them categories to sort into. They answer different questions, and picking the wrong one wastes your time.

Open sorting is exploratory. You're trying to learn how people naturally think about your content — what groups make sense to them, what words they'd use. Closed sorting is a gut-check. You already have a structure in mind and you want to know if real users can work with it. Most projects benefit from doing both, in that order. Start by discovering patterns, then validate the ones that look promising.

When to Use Open Card Sorting

Open card sorting is your starting point when you don't yet know how users think about your content. You're not testing a hypothesis — you're generating one. Participants look at your items and group them however makes sense, using whatever category names feel right to them.

This is the right call when you're:

• Building something new and you genuinely don't know how users would organize it
• Redesigning an IA that clearly isn't working but you're not sure why
• Trying to understand the language your users actually use (not the language your team uses)
• Early in the project, before you've committed to a structure

Here's a real-world scenario: an e-commerce team skips open sorting and just organizes products the way their warehouse does. Users can't find anything. A team that runs open sorting first learns that customers think in terms of occasions ("gifts for kids," "weekend projects") rather than product categories. That insight changes everything about the navigation.

When to Use Closed Card Sorting

Closed card sorting is for when you've got a structure and you need to pressure-test it. You give participants your proposed categories and ask them to sort items into them. The results tell you where users agree with your thinking and where they get confused.

Use it when you're:

• Testing whether your current navigation labels make sense to users
• Comparing two or three candidate structures against each other
• Checking if specific category names are clear or misleading
• Refining an IA that's mostly working but has some trouble spots

For example, say your SaaS product has three possible ways to organize its settings panel. Instead of debating internally, you run a closed sort for each option. The one where users consistently put items in the right place wins. Much faster than arguing in a meeting.

Pros & Cons Comparison

Open Card Sorting

Pros:

• Shows you how people actually think, without your assumptions getting in the way
• Surfaces groupings and relationships you'd never have come up with yourself
• Tells you what words real users reach for when naming categories

Cons:

• Takes longer per session — expect 20-30 minutes per participant
• Analysis is significantly more work because everyone's categories are different
• You need more participants to see clear patterns, since responses vary widely

Closed Card Sorting

Pros:

• Quick sessions — most participants finish in 10-15 minutes
• You get clean, quantitative data like agreement percentages right away
• Directly answers the question "does this structure work?"

Cons:

• You'll never discover a better structure than the ones you thought to test
• Participants are boxed into your framework, even if it doesn't fit their thinking
• Can give you false confidence if none of your options were great to begin with

Hybrid Card Sorting Approach

There's a middle ground. Hybrid card sorting gives participants your proposed categories but also lets them create new ones if nothing fits. It's a good option when you have a strong hypothesis but want to leave the door open for surprises.

In practice, the hybrid approach gives you most of the validation data you'd get from closed sorting, while still catching cases where your categories have blind spots. If participants keep creating the same new category, that's a signal you missed something important.

Sequential Methodology Recommendation

The strongest approach for most projects is to run both methods in sequence. It takes more time, but you end up with an IA built on real user thinking rather than educated guesses.

1. Run open card sorting first with 12-15 participants. Look for the natural groupings that emerge, pay attention to the category names people invent, and note where mental models diverge.
2. Follow up with closed card sorting using 15-20 participants. Take the most promising structures from round one and see how well they hold up when you put them to the test.
3. Consider a hybrid round if your open sorting results pointed in a clear direction but you want to confirm you're not missing an alternative.

This two-phase approach costs more upfront, but it's significantly cheaper than redesigning a navigation that doesn't work after launch. You catch problems when they're still just sticky notes on a screen instead of code in production.

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

What is the main difference between open and closed card sorting? In open card sorting, participants come up with their own categories and decide what to call them. In closed card sorting, you provide the categories and participants just sort items into them. Open sorting tells you how users think. Closed sorting tells you whether your proposed structure matches how they think.

Which card sorting method should beginners use first? Start with open card sorting. It's more forgiving of inexperience because you're not locking yourself into a hypothesis upfront. You'll learn how users naturally group your content, what language they prefer, and where their mental models overlap. That foundation makes your follow-up closed sort (if you run one) much more useful.

How long does each type of card sorting session take? Open sorting runs about 20-30 minutes per participant. They're doing more cognitive work — reading everything, deciding on groups, and coming up with names. Closed sorting is faster, usually 10-15 minutes, because participants just need to match items to the categories you've already defined.

Can you combine open and closed card sorting methods effectively? Yes — that's what hybrid card sorting is. You provide suggested categories but let participants create new ones when nothing fits. It's a practical compromise when you have a reasonable starting structure but don't want to miss something better. You get validation data on your existing ideas plus a safety net for catching gaps.

When do closed card sorting results become statistically reliable? You'll typically see clear patterns with 15-20 participants, though it depends on how many categories and items you're testing. Because the responses are structured (everyone sorts into the same buckets), you can calculate agreement percentages and run clustering analysis even with moderate sample sizes. Open sorting needs more participants to reach the same level of confidence because the responses are so varied.