What is Tree Testing?

Hidden in plain sight. Just because the location of an item on your site is apparent to you does not mean that it is as readily apparent to your user. Tree Testing asks users to find an item or complete as task using the navigational structure or “tree” we’ve created as a way of determining if an item is easy to locate or “hidden” to the user. Tree Testing is also known as reverse Card-Sorting. You’ve allowed the users to tell you how they think things should be organized through Card-Sorting. What is the next step? Tree testing can be used to test the “trees “ developed in Card-Sorting to ensure that you translated the input of the users accurately in the creation the site’s navigation.

Why Use Tree Testing? 

Whether you are testing an existing site or a tree developed in card-sorting, the only way for you to validate that your structure is effective is through tree-sorting. A counter intuitive or confusing navigation can prevent users from achieving their goal on your site. It can negatively affect the user’s perspective of your product or service by “hiding” their goals in obscurity.  Tree testing allows you to identify potential navigation issues and to analyze where your users would expect to find different things.  As you are presenting realistic scenarios to your users, it provides you with insight into real-world behavior. Using the feedback received from your users you can ensure that your site is providing your user with as helpful and as streamlined an experience as possible.

How to Conduct a Tree Test:

Tree testing can be done in one of two ways: Remotely (Users conduct the test at home without moderation) or in a Moderated session(Users are in a controlled environment, where a moderator can interact directly with the participants and ask clarifying questions). Ideally you should involve no fewer than 50 participants. The test should take no more than 15- 20 minutes and you should be sure to be sure to ask no more than 10 tasks. (Over 10 tasks will run the risk of creating additional familiarity with your site. Familiarity with your site will skew results.) 

The steps involved are as follows:

  1. Establish your categories and sub-categories
  2. Provide a tree that is 3-5 levels deep and allow users to access the lowest level of the sub-categories to gather the most amount of feedback.
  3. Ask participants to click through the provided categories and sub-categories to locate a particular item or complete a particular task.
  4. Analyze the following results from the testing:
    1. Paths chosen to get to an item
    2. Length of time it took for someone to find an item
    3. Number of attempts to locate an item
    4. Percentage of successful vs failed attempts to locate an item
    5. Percentage of individuals who did not hesitate and were correct.
  5. If you are conducting a moderated session, also follow up with questions concerning the thought behind the paths the users chose to get to an item.
  6. Use the answers to finalize the tree that is most intuitive and helpful to your users.




<iframe width="100%" height="315" src="https://www.youtube.com/embed/v0F-mb9x67o" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>


The BetterUX Breakdown - Tree Testing

This clip provided by UserZoom, | Published on Nov 29, 2018






<iframe width="100%" height="315" src="https://www.youtube.com/embed/cSHiu_m6vCs" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

Card sorting and tree testing: how do they work together?

Optimal Workshop | Published on Jun 17, 2018