Sell it before it is ready

The advice to wait until the product is good enough before showing it to anybody is the most expensive reasonable-sounding thing in software.

Start selling the moment it does enough to solve the customer’s pain. Not when it’s polished, not when the edges are smooth and not when you’d be comfortable demoing it to a peer because none of those are the bar. The bar is whether the thing it does is worth more to someone than the problem they currently have and that point arrives a great deal earlier than the point where you stop being embarrassed by it.

The worry is that a bad first experience costs you the customer permanently and sometimes that’s true. What happens far more often is that someone has a bad experience, you talk to them about it and you find out precisely which part was bad and why which is information you were never going to get any other way. That conversation is worth more than the sale. It’s certainly worth more than another six weeks of improvements chosen by you, alone, based on what you imagine matters.

Because that’s the actual risk in waiting. Making changes in isolation means making the wrong changes and you can’t know which ones are wrong without someone outside the building telling you. Every week spent polishing before contact is a week of decisions made on guesses and the guesses compound and the product drifts steadily towards being an excellent solution to a problem nobody described to you.

The other outcome is worth naming too because it happens. Sometimes people don’t have a bad experience at all. Either you got lucky and built roughly the right thing first time or the product is really horrible and still less horrible than the problem it removes which is a much lower bar than anybody wants to admit and clears far more often than it should. In that case you now have paying customers, some traction, something to show an investor and, most importantly, money coming in.

That last part is the whole argument. Selling early isn’t really about learning faster, though it does that. It’s about not running out of money which is the thing that actually kills companies and every week of polish before revenue is a week of runway spent on an assumption. Sell as soon as you can. That’s usually earlier than you think and it’s a lot earlier than the moment the product feels ready.

There’s one exception worth stating. If you’re in a market with essentially one customer, defence contracting being the obvious case, none of this applies in the same way. But you’ll have started the sales conversation before you wrote any code at all because if that single customer doesn’t want the thing then there was never any point building it and you’d want to know that in the first week rather than the eightieth.

The thing that makes this hard isn’t the reasoning which most people accept as soon as they hear it. It’s that showing someone an unfinished thing you made feels bad in a way that’s hard to argue yourself out of, and waiting feels like craft rather than fear. Six weeks later the product is better and you still don’t know whether anyone wants it.

That’s the trade you’re actually making. Not polish against speed, but comfort against knowing.

What you learn in the first month of selling something rough is worth more than what you learn in three months of building something good, and the reason is that only one of those two activities involves a person who is not you.