Task Design - Paper Prototyping: The Fast and Easy Way to Define and Refine User Interfaces - page 97

Step 3: Prioritize Your Questions
As a rule, the specific questions are easier to answer, but the broad questions are usually more
important to the success of the product. (Specific issues that are costing the company money, like the
forgotten password example, may be exceptions.) So guess which questions get made into tasks first?
From what I've seen, it's often the specific ones, because they're easier to investigate. But remember
what's at stake — the success of the product, perhaps the entire company, may hinge on the broader
issues.
So look at your list and ask, "Which of these are most important to the success of the next release?"
(Or the company — the best way to phrase this question depends on the nature of the business you're
in.) One way to prioritize the issues is to have people vote for the three questions they think are most
important. Often, you can't answer all of the top questions in a usability study, but at least you'll be
looking at the forest, not just the trees.
Sometimes the broad questions pertain to patterns that will emerge only after you've watched a number
of users do various tasks. For example, if your broad questions are "How do users navigate?" and
"Why do customers complain that it's too slow?" you can't really create one specific task that will
answer them. In this case, make sure that your set of tasks as a whole will let you watch users do the
things that you think will yield clues, and eventually you'll start to see the patterns.
Remember what's at stake—your company.