What most people do to solve a problem:
Lock in plans early
Diverge without clarity of problem
Run (& forget about) off-the cuff poorly designed experiments
Hope that the magical “how might we” generates superior ideas from a bland statement
What you should do:
Steep yourself in research, together with the team. The problem has a shape, you’re the archaeological team.
Try before you move further. I start light throwaway prototyping very early to understand the problem better.
Get serious about the word “experiment”. It doesn’t stand for “f*ck around and find out” — be deliberate about what it is, how it’ll work. Even for a small round of prototyping.
Make your mental models explicit & curate them. No one knows what’s in your head. But models all over the place is as bad as no models.
Why is this hard? I don’t know! Thinking is hard.