Treejack helps you to discover if people can find content in your information architecture, so every task needs at least one correct destination, and you're unable to select parent nodes as correct. The results show you participant paths in detail, so you'll still be able to see how the parent node performs.

Why you can't select parent nodes as correct answers

Dave O'Brien provides a great explanation as to why you shouldn't select parent nodes as correct answers.

At the moment, Treejack doesn't allow you to mark a parent node as a correct answer. Even if your tree testing software allowed you to, marking parent nodes as correct isn't always a good idea. This is because in your results you want to see the actual path your participants took to arrive at their destination.

It's difficult to follow this down to lower levels if the parent node is the correct answer, as you won't be able to see their entire path! This simplifies the experience for your participants and the results you analyze. Furthermore, you won't learn a lot if all your correct answers were parent nodes.

Often, not all answers in the subtree are correct, so marking the parent as correct is sloppy.

Here's a couple of suggestions for what to do if you want select a parent node as the correct answer because:

  • every node underneath it could be correct

  • the page contains the answer but it has child nodes.

Everything in the category is correct

If it's because the node and any child of that node is correct, then one solution is to simply remove all child nodes from that node in your tree. For the purposes of your testing, the path ends at that node.

As an example, consider a product taxonomy for an auction site like eBay. If you set a task to find golf clubs, and you decide the category "Golf" is the correct answer, there is no need to further include "Drivers" and "Irons" etc. as child nodes.

Alternatively, you could get more specific with your tasks, and ask participants to find a five iron.

The page contains the answer but it has child nodes

This is usually because the node represents a landing page of some kind linking to further content on other pages, and it is this page which contains the content you consider the correct answer for a task. Because it has children, it can't be selected as the correct answer for a task. The solution is to explicitly include the piece of content you consider the answer as a leaf child of the "header" node in the tree.


As an example, if we use a fictional website and the site map looks like this:




But the structure of the content is actually

But now you can't select "Wellington office" as a correct answer. This is because when you add children to "Wellington office", it no longer reflects a piece of content, but instead a container for content. To better reflect how the content is structured, you could build the tree like this:

Did this answer your question?