Charge before it is ready

Published August 28, 2026 · 7 min read

The standard advice is to charge once the product is good enough to deserve it. That sounds like integrity and it is mostly cowardice, because a free product never has to answer the only question that matters. An operator running small products has very little to spend on research, and exactly one instrument that returns an honest reading. It is the payment form.

There is a particular kind of week that every operator of small products eventually has. The thing works. People are using it. The graphs go gently up. And you cannot say, with any confidence at all, whether it matters to anyone.

That uncertainty is not a data problem and more analytics will not touch it. It is that you have never asked anyone a question they could not answer politely.

Free feedback is contaminated by kindness

When something costs nothing, everyone you show it to is a guest in your house. They will find something encouraging to say, because that is what decent people do. They will sign up because signing up is free and refusing feels rude. They will use it twice and drift away without telling you why, because there is no relationship to end.

So you accumulate a body of evidence that is uniformly warm and completely uninformative. Nothing in it is a lie. It simply was not produced under conditions where lying would have cost anything, which means it was not produced under conditions where truth had to surface either.

Asking for money changes the conditions. It is the one question a person cannot deflect with politeness, because paying and not paying are visibly different acts and both of them are on the record.

What the payment form is actually measuring

It is easy to read an early price as a revenue decision and then reason about it badly, because at small scale the revenue is not the point and everyone knows it. The money will not change your life. What it does is force a set of questions into the open that you have otherwise been allowed to keep vague.

• Whether the problem you picked is one people already know they have, or one you have to convince them of first
• Whether the person using it is the person who would pay, which is a distinction free products let you ignore entirely
• Which single feature gets named when someone justifies the spend to themselves
• Whether your explanation of the product survives contact with someone deciding, rather than someone browsing
• Whether you believe in it enough to ask, which you will find out about yourself in the moment you try

That last one is not a joke. A surprising number of products stay free because the operator does not want to hear the answer, and has found a principled-sounding reason to avoid asking.

The polish trap

The argument against charging early is always some version of: it is not good enough yet. Charge when it deserves to be charged for. This sounds responsible and it inverts the actual sequence, because you cannot know what would make it good enough until somebody who is paying tells you what is missing.

Left alone, you will polish the parts you enjoy polishing. Everybody does. The paying user is the only force that reliably points you at the parts you have been avoiding, and they only point once they have standing to complain. Standing is what the payment buys them.

So the cycle you are actually choosing between is not build-then-charge versus charge-then-build. It is being corrected early and cheaply, or being corrected late, after you have spent months on the wrong surface.

How to ask without pretending

Charging early does not mean overselling. The honest version is simple to run and it protects both sides.

• Say plainly what the product does today, and do not describe the roadmap as though it exists
• Charge for the thing that already works, not for the thing you intend to build
• Make leaving trivially easy, because friction on the way out corrupts the signal on the way in
• Talk to the first people who pay, individually, and treat that conversation as the actual output of the exercise
• Keep the price where you can defend it out loud, and change it when you learn you were wrong

None of that requires a finished product. It requires being straight about which part is finished.

What you do with a no

The outcome operators fear is that nobody pays. It is worth being clear that this is a successful experiment, not a failed one. You have learned, in a week or two, something that a free product could have hidden from you for a year while you kept building on top of it.

A no is also rarely total. Usually some people pay and most do not, and the interesting work is in the difference between those groups. What did the payers have in common that you had not noticed? What were they using it for that you did not design for? That question has an answer, and the answer is frequently a better product than the one you were planning.

The uncomfortable version

Keeping a product free is often described as generosity toward users. Sometimes it is. More often it is generosity toward yourself, because a free product can never fail clearly. It can only fade, slowly, in a way that never requires you to admit anything.

Charging is how you make failure legible while it is still cheap. That is uncomfortable, which is precisely why it is worth doing early, and precisely why so few people do.

Common questions