...
Instead of finalizing the structure then and there, she wants some proof that a topic-based tree is better than her initial idea of an audience-based tree. So she creates an audience-based tree as well, and decides to test them against each other. (The project sponsor happens to prefer the audience scheme, so this is an additional reason to get objective data on both before deciding.)
She takes a week to run side-by-side tree tests (one for each tree), and gets about 100 people to try out each tree:
...