Be specific. Vague input produces useless output.
Select everything that applies. Don't be flattering to yourself.
Not what you plan to test. What you've already done. Specific experiments, conversations, changes.
Answer honestly. This is the question that separates signal from avoidance.