The Most Expensive Word in a Startup Is “Yes”

I have spent most of my working life saying yes. Yes to the customer who wanted one more feature. Yes to the hire who dazzled in the room. Yes to the partnership that sounded strategic when someone said it out loud. For a long time I believed this was what building looked like — an open door, a bias toward action, a founder who never let an opportunity walk past.

It took me forty years to understand that yes was quietly the most expensive word I owned.

A startup does not run out of ideas. It runs out of time, focus, and cash, usually in that order and faster than anyone forecasts. Every yes spends all three at once. It commits engineering hours you can never re-earn, attention you cannot redirect once it’s fragmented, and runway you priced as if the yes were free. The invoice does not arrive the day you say it. It arrives three months later, when the thing you actually needed to build is late and you cannot quite explain why.

Every yes is a mortgage, not a purchase

The reason yes is so seductive is that it feels like momentum. You leave the room having agreed to something, and agreement reads as progress. A purchase is honest — you hand over money, you get a thing, the transaction closes. A yes in a startup is a mortgage. You take on an obligation now and pay it down for months, long after everyone has forgotten the conversation that created it. The cost hides inside familiar situations.

A large customer asks for a feature and hints that the deal depends on it. Saying yes feels like closing revenue. What you have actually done is let one account edit your roadmap, and the next large customer will ask for a different feature with the same leverage. You have not won a customer. You have hired a product manager who does not work for you.

An investor suggests a pivot with the easy confidence of someone spending your years, not theirs. The suggestion carries weight because they hold the checkbook. But most pivots proposed in a single meeting are pattern-matching against someone else’s portfolio, not conviction about your business. Yes here can cost you the one thing a founder cannot outsource, which is knowing why the company exists.

An impressive candidate walks in — a pedigree, a presence, a story about scale. The team is stretched, and you want the relief. The wrong senior hire is among the most expensive yeses of all, because it compounds: the wrong leader hires the wrong people and sets the wrong tempo, and the damage takes far longer to unwind than the hunt for the right person would have taken.

A partnership arrives dressed in the word strategic. Two logos, a press line, a sense that you are playing a bigger game. Most consume quarters of integration work for a distribution promise that never materializes. Strategic is the adjective people attach to deals that cannot justify themselves on the numbers.

None of these are obviously bad in the moment. That is the whole problem. A costly yes almost never looks like a mistake going in. It looks like an opportunity, which is exactly why the discipline has to be structural rather than emotional.

“We could do that” is the most dangerous sentence in the room

Somewhere in every roadmap conversation, someone says “we could do that.” They are always right. You could. A capable team can do almost anything, and that capability is precisely the trap. The question is never whether you can. It is what dies quietly while you do — the feature that does not ship, the segment you stop serving well, the focus that thins across too many fronts until nothing is excellent.

I have learned to treat “we could do that” as a yellow light, not a green one. The honest follow-up is not “how would we build it” but “what are we willing to stop doing to make room for it.” If the answer is nothing, then the real answer is no, and everyone in the room already knows it.

Here is the test I use now. A real opportunity moves the one thing the company has decided matters most this year. A costly yes moves something adjacent, something you could write a reasonable memo about — but not the main thing. Good ideas are cheap and endless. The scarce resource is a team pointed in one direction long enough to get somewhere. Almost every yes I regret was a yes to something genuinely good that pulled us off the one thing that was essential.

The hardest version is saying no to good people with good ideas who are not wrong. Saying no to a bad idea requires no character. Saying no to a good idea, from someone you respect, when you can feel the room leaning toward yes — that is the whole job.

The disciplined no is a form of respect

I want to grant the counterpoint honestly. Say no to everything and you become the founder nothing gets past, the one who confuses caution for judgment and slowly starves the company of the surprises that turn into the whole business. Some of the best things I ever built started as a yes to something that made no sense on paper. A no reflex is just as lazy as a yes reflex. The skill is not saying no. It is knowing which few yeses are worth the mortgage.

What changed for me was learning to say no slowly and with respect, instead of yes quickly and with relief. A good no explains the tradeoff out loud. It tells the customer what you are choosing to be excellent at instead. It tells the investor you heard them and here is the conviction you are betting on. It tells the candidate this is not the right seat, which is kinder than hiring them into a mistake. A disciplined no treats the other person as someone who can handle the truth.

I still say yes. I say it less, later, and only after I have named what it costs and what I am willing to stop doing to afford it. The founders I admire most are not the ones who moved fastest. They are the ones who protected their focus like it was the only capital they had.

Yes is the word that feels like building. No is the word that actually does the building. It took me forty years to tell the two apart.