Will more features get you customers?

Almost never, when nobody's buying yet. Here's how to tell when a feature is the answer, and what to do this week instead.

2 min read · Updated 1 Nov 2026

The short answer

Adding features rarely fixes a product nobody's buying. If people who match your customer aren't paying, the problem is usually who sees it, how much the problem matters to them, or the offer. Build a feature when a specific customer has told you it's what stands between them and paying. Otherwise spend the week finding ten people to talk to.

When a launch goes quiet, the instinct is to build. One more feature, a cleaner onboarding, a redesign. It feels like progress because you’re making something. For a product nobody is buying yet, it’s usually the wrong move.

Why more features rarely fix a quiet launch

Pieter Levels described a meetup of indie hackers in June 2026 with impressive automated AI pipelines, where “almost none of them have money or traffic” (levels.io). Building is the easy part now, which is exactly why founders reach for it first.

If nobody is paying, something is broken in one of three places: the people seeing it, how much the problem matters to them, or the offer. A new feature only helps with a narrow part of the third. If the wrong people are seeing the product, or the problem isn’t costly enough, a better product meets the same silence.

When a feature is the right answer

Build it when all three of these are true:

A specific customer asked for it. A name, not a hunch or an anonymous survey answer.
It’s what stands between them and paying. They’ve said they’d buy, or renew, once it exists.
Their behaviour backs it up. They already do the job some other way today, with a spreadsheet, a workaround or another tool.

Then build the smallest version and show it to that person first.

The question to ask before you build

Before the next feature, ask: has a specific person told me they’d pay for this? If the answer is no, the feature is a guess, and guesses are expensive when every week counts.

What to do this week instead

Write down ten people who match your customer, by name, with how you’d reach each one.
Message three of them and ask for fifteen minutes about the last time they had the problem you solve.
Write what you hear as a Problem Receipt, in their words.
Rewrite your offer from it, and ask each person for a decision.

If you can’t name ten people, that’s the real problem, and it’s a more useful thing to know than another feature. For the full process, see what to do when you’ve launched and nobody’s buying.

The Sunday email

Get one of these every Sunday.

One founder problem and one thing to do before Monday.

Questions

When is a new feature the right call?

When a named customer, or several, has told you it's what stops them paying or renewing, and you've checked the request against what they actually do. Then build the smallest version and show it to them first.

What about features competitors have?

Matching a competitor's feature list rarely wins early customers. They choose you because you solve their specific problem better, or because you reached them first.

How do I say no to feature requests from users who don't pay?

Write them down, thank them, and ask what they do today instead. The answer tells you how much the request matters. Prioritise requests from people who pay or who've committed to.

Next step

Is your idea worth 30 days? Find out free.

Five questions about your idea and a straight answer. It's week zero of Humble's 30 days, and it costs nothing.

Take the free first step