The fastest way to learn what a feature is worth is to put it in front of people who didn't build it. Polish delays that lesson; shipping collects it.
Rough edges are feedback magnets
Users forgive an honest rough edge far more readily than a missing capability. What they won't do is imagine the feature you kept on a branch.
A released rough edge teaches you more than an unreleased perfect one.
What we do instead
- Ship the smallest slice that answers a real question.
- Watch how it's used for a week before touching it again.
- Polish only the paths people actually walk.
None of this is an argument against quality. It's an argument about sequence: quality earns its keep after usefulness is proven, not before.