Imagine a piano that's 90% in tune. How much would a concert pianist pay to rent it? Nothing. It's not 90% of a good piano. It's a broken piano. Even one out-of-tune key in a Chopin nocturne and the whole thing collapses.
This is how most professional work behaves. Code that's 90% correct crashes. A contract that's 90% right has a clause that costs you millions. Compliance documentation that's 90% complete fails the audit. Value doesn't degrade gracefully with quality. It falls off a cliff.
But not everything is a Chopin nocturne. Most work lives in a different zone entirely.
Internal documents, first drafts, research summaries, brainstorming, rough plans. For these, 90% is great. Better done than perfect is genuinely true for a huge portion of daily work, and anyone who insists on perfection across the board is just slow.
The market figured out something smart here: generate ten variations, let a human pick the best one and fine-tune it. This works. It's fast, it's cheap, and for the majority of tasks the result is good enough. Nobody needs a hand-crafted artisanal status update.
So 90% is often fine and sometimes it's zero. The interesting question is what happens beyond that.
Because there's a third category of work where the problem isn't getting to 90% or even 99%. It's that no number of variations gets you there at all.
Making something that truly stands out. Something novel. Something that moves the needle in a market that's seen everything. You can generate a thousand variations and pick the best one and fine-tune it and the result is still... fine, competent, defensible. Meaning nobody would remember it.
The gap between competent and extraordinary isn't a quality gap. It's a different kind. Competent comes from doing the expected thing well. Extraordinary comes from a starting point nobody expected. And often you can't iterate your way from one to the other.
Nowdays most of us spend most of our time in the variation-and-pick zone, and that's correct. That's where most work lives. The mistake is believing you can stay in that mode and occasionally produce something extraordinary by generating more variations. You can't.
The few things worth making extraordinary require you to stop picking from options and start from somewhere nobody offered you. That's a different skill, a different mode, and a different kind of effort. It's uncomfortable because there's nothing to react to, nothing to evaluate, nothing to fine-tune. Just a blank page and whatever you bring to it.
Most things deserve 90%. Ship them. A few things deserve everything you have. Pick wisely.