Every Feature Is a Forever Decision
When you ship a feature, you tend to think of it as a finish line. It is closer to a marriage. From the day it goes live, someone has to maintain it, test it, document it, support it, account for it in every future design, and explain it to every new engineer for as long as the product lives. The cost of a feature is not the cost to build it. It is the cost to carry it, paid in small installments, forever.
This is why your best feature is often the one you decided not to build. It generates no bugs. It needs no docs. It never breaks at 2 a.m. It does not sit in the settings menu confusing new users or forcing the next three features to design around its edge cases. It costs nothing because it does not exist, and yet it delivered exactly the value its absence was meant to deliver.
The real enemy is not the missing feature. It is complexity. Every addition increases the surface area of the product: more states, more interactions, more ways for things to go wrong, more cognitive load on the user trying to find the one thing they came for. Complexity compounds quietly. No single feature feels like too much. The hundredth one buries the product.
So treat subtraction as real product work, not cleanup you get to if there is time. Removing a confusing option, killing a feature nobody uses, collapsing three flows into one. These do not show up well in a "what we shipped this quarter" deck, which is exactly why they are undervalued and exactly why disciplined teams pull ahead. Less product, more clearly aimed, beats more product almost every time.
Scope discipline is the daily version of this. The instinct on every project is to add: one more option, one more edge case handled, one more "while we are in here." Resist it. The question is never "would this be nice?" Almost everything would be nice. The question is "is this worth carrying forever?" That is a much higher bar, and most additions do not clear it.
The mature product is not the one with the most features. It is the one where every feature still earns its keep, and where someone had the discipline to keep the rest out.
The hardest thing to build is the thing you were brave enough to leave out.