Why run a tree test?

We already have card sorting, usability testing, and a variety of other UX methods. What does tree testing offer that these other methods don’t?

It’s quick.
We can set up, run, and analyze a tree test in a week or less.
We can test before we have a site or even a prototype.
All we need is a tree of topics, some representative tasks, and some willing participants.
It focuses on site structures only.
By testing organization and labeling by themselves, tree testing provides a solid foundation that we can build on later, adding page design, visual design, and search.
It’s better at testing structures than closed card sorting.
Closed card sorting was the traditional way to test site categories, but tree testing is more realistic and deals with all levels of the tree.
We get an objective evaluation of our headings and terms.
Qualitative feedback from users and opinions from stakeholders are good, but checking that against quantitative results from our target audience is much, much better.

For more on how tree testing fits into UX design (and how it compares to other UX methods), see Chapter 3 - IA in the design process.