Find your real customers before you build a thing

Rip the plaster off. The instinct is to build first and find customers later, because building feels like progress and talking to strangers feels like risk. But the goal isn’t to protect the idea — it’s to find out if it’s real, fast enough to either commit or move on. Fail fast, or spend two years floating. Discovery is the skill that decides which.

The four questions

Customer discovery is really one ordered list of questions:

1. Who could possibly care? Everyone — paying customers and non-payers who’d still show demand by signing up or sharing.
2. Who’s most likely? Rank them. You can’t chase everyone, so chase the warmest first.
3. Where are they? Online and in person. Go to where the problem happens and where they already spend time.
4. When are they receptive? The goal of the whole exercise is one thing: get a conversation.

Then experiment. Try different ways of explaining the value, see what makes people lean in, and pivot toward whatever sticks.

Don’t forget the champion

If you’re selling to organisations, the person who loves your idea usually isn’t the person who signs the cheque. Find your champion — someone in mid-to-upper management who feels the problem — and help them sell it internally. The org chart is part of the map.

Build the map with AI in an hour

This used to be a whiteboard afternoon. Now it’s a prompt. Your AI can brainstorm every customer type you’d never have listed, rank them, guess where each gathers, and draft the outreach — leaving you to do the only thing it can’t: have the conversations.

Act as a customer discovery strategist. My idea: [one line]. List every type of person or organisation who could plausibly care about this — paying customers AND non-paying people who'd signal demand. Rank them most-to-least likely to be interested, with a reason. For the top 3, tell me: where they spend time online, where they gather in person, and what they actually care about (the "why should they care"). For any B2B buyer, name the likely champion and the decision-maker.

Treat the output as a first draft of reality, not reality. It gives you somewhere to start knocking — and the next step is to actually talk to those people without fooling yourself, which is its own skill: the Mom Test.