Are we building what people need, or just what we want?

You did not set out to serve yourself. No one does. It happens slowly — a feature you were proud of, a design you fought for, a roadmap that quietly bent toward what impressed you rather than what helped anyone. This is not a moral failing. It is what happens when skill goes unchecked. Notice what you are actually asking. Not "are we good people" — that question flatters you and changes nothing. Ask instead: in the last month, how many decisions were made because a user asked for them, and how many because they were interesting to build? Count honestly. You already suspect the answer. You cannot do this alone, and you should not try to. One person checking their own motives will always find reasons to trust them. A team checking each other will not. This is why you do this work together — not as a group therapy session, but as a standing discipline, the way you'd check a budget.