Asking whether somebody likes a page seems less useful than watching them try to do something on it. But the wording of the task can give too much away.
What would you ask a researcher to prepare? How should they report unexpected routes that still end successfully?
.
