Ever wondered why finding info on some websites is a breeze, while others make you want to pull your hair out? The secret often lies in two key UX research techniques: tree testing and card sorting. 

In this article, we'll explain what these methods are, when to use them (both alone and together), and the benefits they can give. By the end, you'll know how to make the structure of your website intuitive and user-friendly. 

Let’s begin by first understanding tree testing.

Definition and purpose of tree testing

Tree testing is a method used to evaluate the findability and usability of a website or app’s hierarchical structure. It focuses on how easily users can navigate to specific content within an information architecture (IA).

For example, imagine you run an e-commerce website with categories like "Men's Clothing," "Women's Clothing," "Accessories," and so on. Tree testing helps you understand if users can intuitively find the "Winter Jackets" category under "Men's Clothing" or "Women's Clothing" without getting lost.

The primary goal is to identify any navigational obstacles or confusing elements that may hinder the user experience.

Your go-to user research platform

The best teams use Lyssna so they can deeply understand their audience and move in the right direction — faster.

Sign up for free

Sign up for free

Sign up for free

How to conduct tree testing 

Setting up a tree test is straightforward. Let’s go through the key steps to get you started. 

Define goals for your research

Before anything else, clearly define what you want to achieve with your tree test. Are you trying to improve the findability of specific information? Or perhaps you're looking to streamline your website's navigation? Take time to establish your goals, as this will help to guide the entire testing process.

Design the tree

Start by making a simpler version of your website or app's IA. Focus on the order of categories and subcategories. This structure, often called the "tree," represents the backbone of the site's navigation.

When building a tree test in Lyssna you can easily map out categories and subcategories. For example, on a banking website, you might have ‘Accounts’, ‘Transfers & Payments’, ‘Investments’, and ‘About Us.’

Create tasks 

Develop specific goals that participants need to achieve, such as finding a page or piece of information. These tasks should mimic real-world scenarios users might encounter. 

Continuing with the example above, you might give participants the task: ‘Where would you go to transfer money between accounts?’.

Recruit participants

Recruit users who match your target demographic to get relevant results.

If you’re not sure how many people you should recruit, starting with at least 30 participants is a good bet. Although when it comes to recruiting, the more the merrier to get a clearer picture of how users navigate your site. Aiming to recruit roughly 40 to 60 participants will give you enough variety in user experiences without breaking the bank or taking forever.

Analyze the results

During the tree test, participants navigate through the structure to complete tasks, and you track how well they find what they're looking for. Analyzing the results involves looking closely at whether users struggled or breezed through the navigation. This can help you identify areas that need improvement, like tweaking categories or making labels clearer.

Analyzing tree testing results is more than just looking at numbers. It's about understanding how users really use your site. By asking follow-up questions and blending quantitative and qualitative data, you get a fuller picture of their habits and preferences.

Again, using the same example, you can see below that 80% of users easily found the money transfer button.

But when we dive into the qualitative feedback, we discover why some couldn't. One person was unsure about what to expect under the "Services" button. In this case, it’s an outlier result, but it could be an important thing to consider. 

Benefits of tree testing

Here are a few good reasons why you should consider doing a tree test.

• Objective insights: Tree testing generates clear and objective data on where users encounter difficulties within the IA structure, without the bias of visual design elements.

• Identify navigation issues: You can pinpoint specific paths that users take – or struggle to take – through the IA, identifying potential issues at any stage of the design process. This can help you refine navigation paths and make sure users can find information efficiently.

• Optimize IA: Through systematic testing, you can assess the effectiveness of your IA. Making adjustments based on user behavior data can lead to clearer categorizations, better labeling, and improved overall structure.

• Cost-effective: Tree testing is known for its speed and simplicity, making it a practical and efficient way to gather valuable user feedback. Plus, compared to the hefty price tag of redesigning entire interfaces, tree testing allows for iterative improvements at a lower cost. 

This is just the tip of the iceberg. For a deeper understanding of tree testing, visit our ultimate guide to tree testing.

Definition and purpose of card sorting

Card sorting is a great technique for figuring out how users think about and organize information. 

Imagine you have a pile of recipes, each with a different dish, like pasta, salad, soup, cake, etc. You ask your friends to group these recipes based on how they think about meal categories, like main course, sides, or desserts. This exercise helps you understand how people naturally categorize things in their minds. 

It’s the same for websites and apps. Card sorting works similarly, but with information or features instead of recipes. It helps designers make layouts and menus that users understand. This makes it easier for them to find what they want and move around easily.

Types of card sorting 

There are three types of card sorting: open, closed, and hybrid. Each type offers unique insights depending on what you’re looking to find out.

• Open card sorting: This is where participants group cards however they like and come up with their own category names, based on their own understanding and logic. This method will show you how users naturally think about and categorize information.

• Closed card sorting: Here, participants sort items into predefined categories you provide. This approach helps validate your existing IA structures and assess how well they align with user expectations.

• Hybrid card sorting: Like the name suggests, hybrid card sorting combines parts of both open and closed approaches. Participants may initially categorize content freely and then refine their groupings based on predefined categories.

Benefits of card sorting

Card sorting offers plenty of benefits. Here are a few of the key pros.

• User-centered IA design: The data you gather from card sorting will show how people naturally organize information. Aligning your site structure with this information ensures that your design puts users first, making it simpler for them to find what they're looking for.

• Improves content organization: By identifying common patterns and groupings among users, card sorting helps create intuitive categories and labels that enhance user navigation and satisfaction. It also helps uncover any gaps in the site structure that may need attention.

• Validate your IA: Running a card sorting test will give you hard data on whether the site's structure matches how users expect things to be.

Looking for more information on card sorting? Visit our card sorting guide.

What's the difference between tree testing and card sorting?

Now that you have a clearer grasp of tree testing and card sorting, can you distinguish between the two? While both methods aim to enhance the user experience, they do so from distinct perspectives. Let’s dive a little deeper into this comparison.

When to use tree testing vs card sorting

Each method offers unique insights and benefits depending on your research goals and the stage of your design process. We cover a few key use cases below.

When to use tree testing

Checking your existing IA setup

• Purpose: Use tree testing to evaluate the effectiveness of an already established IA structure.

• Scenario: When you want to make sure users can navigate through your site or app without any hiccups, tree testing gives you solid data on where things might be confusing.

Navigation path optimization

• Purpose: Identify and refine navigation paths within the existing IA.

• Scenario: If you suspect users are getting lost in your menus or categories, this method shows you where to smooth things out.

Comparative IA analysis

• Purpose: Compare multiple versions or iterations of IA structures to determine which performs better.

• Scenario: You’re redesigning your website’s navigation and want to test different IA concepts to see which one users find most intuitive and usable.

When to use card sorting

IA design and development

• Purpose: Involve users in the creation or refinement of IA structures.

• Scenario: You’re starting from scratch with a new website or application, and you want to understand how users naturally categorize and organize content to inform your IA design.

Understanding user logic

• Purpose: Gain an understanding of how users perceive content relationships and categorization.

• Scenario: You’re unsure about the best way to structure your content or how users would expect to find information on your platform, and you need qualitative data to guide your decisions.

Testing labeling and terminology

• Purpose: Test and refine the labels and terminology in your menus or categories, so you can speak the same language as your users.

• Scenario: Make sure the words and phrases on your site match how people think. You could use card sorting to make sure your labels make sense and are clear for everyone using your site.

Advantages and limitations of tree testing and card sorting

Combining tree testing and card sorting 

While tree testing and card sorting have distinct roles, using them together can give you a full picture of how to improve your IA. Let’s explore how these methods can complement each other and be used in tandem to achieve the best results.

To make things easy, we’ll be using an invented e-commerce site, ‘Tony’s Electronics’, which is looking to improve its product categories.

Step 1: Start with card sorting

Objective: Understand how users naturally group and label content.

Example: The team at Tony’s conducts open card sorting sessions with their target demographics. They ask the participants to organize various products into categories and label them as they see fit.

What we learn: The users create natural groupings like ‘Computers,’ ‘Home Appliances,’ and ‘Gaming,’ and they provide labels that make sense to them.

Step 2: Create an initial IA draft

Objective: Develop a draft IA structure based on card sorting insights.

Example: Tony’s UX team takes the categories and labels from the card sorting session and creates a draft navigation structure. Products are organized under the user-generated categories with their preferred labels.

What we learn: The initial IA draft includes the categories ‘Computers,’ ‘Home Appliances,’ and ‘Gaming,’ which reflect how users think. 

Step 3: Validate with tree testing

Objective: Test the effectiveness of the draft IA structure.

Example: The team then conducts tree testing with a new set of users. Participants are given specific tasks, such as finding a particular type of camera or a specific kitchen appliance, to see how easily they can navigate the new structure.

What we learn: The results show that while most users can find items under "Electronics" easily, they struggle with "Home Appliances" due to some subcategories being unclear.

Step 4: Refine, iterate, and validate

Objective: Use the feedback from tree testing to refine the IA structure. Then retest. 

Example: The "Home Appliances" category is refined by reorganizing subcategories and improving labels. In the second test, users are given similar tasks to ensure they can now navigate the "Home Appliances" category more effectively.

What we learn: The second round of tree testing shows improved task completion rates and quicker navigation times, confirming that the refinements were successful.

Step 5: Implement and monitor

Objective: Roll out the final IA structure and monitor user interactions.

Example: Tony’s Electronics implements the refined IA on their live site and monitors user feedback and interactions to make sure everything is working well.

What we learn: The final IA structure is user-friendly, intuitive, and aligned with user expectations, leading to better overall user satisfaction and reduced bounce rates.

Your go-to user research platform

The best teams use Lyssna so they can deeply understand their audience and move in the right direction — faster.

Sign up for free

Sign up for free

Sign up for free

Branch out in UX: Start using tree tests and card sorting today

Tree testing and card sorting are invaluable tools in your UX toolkit, each offering distinct insights into IA, usability, and user preferences. Now that you understand their differences and know when to use each method, you can be on your way to creating intuitive and user-centered information architectures that enhance overall user satisfaction and engagement. 

Our suggestion: start experimenting with both methods, as it not only improves IA but also ensures that designs are grounded in user behavior and expectations. To get started, check out our card sorting templates and tree testing templates.

Try out Lyssna’s tools for tree testing and card sorting to elevate your UX research efforts and deliver exceptional user experiences.