What you owe the first person who pays you

Published September 2, 2026 · 7 min read

A first sale is celebrated as evidence and then discovered, some weeks later, to have been a commitment. Somebody paid, which means somebody is now waiting: for the thing to work, for a reply, for the product to still exist next month. Almost none of that gets decided in advance, and all of it gets decided eventually, usually at the worst possible time and in the worst possible mood.

The sale is the beginning of the obligation, not the end of the experiment

Charging early is good advice and it is usually given as an experiment: ask for money and see whether anyone says yes. That framing is useful right up to the moment somebody does. From there the experiment has produced a person, and the person has expectations that exist whether or not you have thought about them.

This is the quiet transition that catches solo operators. Before the first sale you own a project, and a project can be neglected without anyone being wronged. After the first sale you own a small obligation, and neglect now has a victim. Nothing about the code changed. The moral situation did.

None of this is an argument for waiting longer to charge. It is an argument for spending an hour, once, deciding what you are promising, so that the promise is something you can keep on a bad week rather than something invented on a good one.

Four things to decide before the second sale

Each of these will be decided at some point. The only real choice is whether you decide it calmly, in advance, or under pressure in front of the person affected.

• Response time. Not the fastest you could reply, the slowest you are willing to be. Say it publicly and let it be unimpressive. A stated slow answer is kept; an unstated fast one is broken the first time you are ill.
• Refunds. What qualifies, for how long, and how someone asks. Write it in one sentence. The purpose is to stop you negotiating against yourself at midnight.
• Absence. What happens to the product while you are away, and what the customer sees. If nothing keeps running without you, that is a fact about the product and you should know it before a holiday teaches it to you.
• Ending. What you owe people if you shut it down: how much notice, whether their data leaves with them, whether anything is returned. Deciding this while enthusiastic is the only time you will decide it generously.

The obligation is continuity, not features

Operators tend to assume the debt they owe a customer is improvement, so they respond to a first sale by building. It is a comfortable assumption because building is the part they enjoy, and it is usually wrong. What somebody who paid actually wants is that the thing keeps doing what it already did, reliably, for as long as they are relying on it.

Continuity is unglamorous and it is what gets skipped. The certificate that has to be renewed, the payment provider that changes a rule, the dependency that stops being maintained, the inbox nobody reads. None of it is visible to a customer until all of it is, at once, in the form of something that no longer works.

The practical version of this is uncomfortable: a new feature that increases the number of things which can silently break is not obviously a gift to the person paying you. It might be. It is worth asking rather than assuming, because the answer changes what you should do with the week.

Support is a design decision, not a personality trait

The first few customers are usually served heroically. Replies come within minutes, at any hour, from the founder, because there is one of them and it feels good to be needed. This is generous and it sets a precedent that cannot survive the tenth customer, let alone the hundredth. The people who received the heroic version experience its withdrawal as a decline in service, which, from where they sit, it is.

The alternative is not to be cold. It is to promise the level you can hold indefinitely and then, when you have time, exceed it quietly. Under-promising is not a trick here. It is the only way the promise survives contact with a bad month.

It also matters where support arrives. A channel you check because you happen to be in it is not a channel, it is a coincidence. Pick one place, put it where a customer can find it without asking, and let everything else be somewhere you are not obliged to be.

Ending well is part of what you sold

Most small products are not shut down, they are abandoned: the domain lapses, the last email goes unanswered, and the customer works out from silence that it is over. This is the cheapest possible ending for the operator and the most expensive one for everybody else, and it is a large part of why people are wary of buying from a one-person business at all.

A clean ending is not difficult, it is just decided too late. Tell people it is stopping, tell them when, stop taking money before you stop providing the thing, let them take their data with them, and say plainly whether anything is being refunded. Doing that once is worth more to your name than anything you will ever write about yourself.

There is a reason to care about this beyond decency. An operator who runs several small things will end some of them, and the way the first ending is handled is what the next audience finds when they look you up. Reputation for a small operator is not built by the products that worked. It is built by what happened to the ones that did not.

The hour that is worth spending

All of this fits on one page: how fast you answer, what you refund, what happens when you are away, what happens when it ends. Write it before the second sale, put it somewhere a customer can read it, and treat it as a floor rather than an aspiration.

The reason to do it early is not tidiness. It is that every one of these questions has an honest answer and a convenient one, and which you give depends almost entirely on whether you are answering in the abstract or answering a specific person who is annoyed with you. Write down the honest one while it is still cheap.

Common questions