How do I test an app idea before building it?

Testing an idea means finding out which of your guesses hold up while they're still cheap to change. You don't have to finish testing before you build anything. With AI builders like Lovable, Bolt or Claude Code, a rough version you can put in front of customers takes an afternoon, and it can be one of the best ways to learn. What matters is that each thing you do teaches you something about the people you'd sell to.

The five things to find out about

Your idea depends on five guesses. Together, they make up your Business Hypothesis:

Customer: who it's for, who needs it most, and how you'll reach them.
Problem: what they're struggling with, what they do about it now, and what it costs them.
Promise: the outcome you're offering, and why they'd choose you.
Solution: what the first version must do to keep that promise.
Commercial: what you'll charge, who pays, and whether that covers what each customer costs you to serve.

A rhythm for finding out

It helps to work in a loop with five steps:

Choose what to find out. Start with the guess that's most likely to be wrong and would hurt most if it were.
Make a useful move. Have a conversation, build something rough, make an offer or deliver the outcome yourself.
See what customers do. Give the right people a real chance to respond, try it, buy it or come back to it.
Make sense of what happened. What does it tell you, what surprised you and what's still unclear?
Decide what comes next. Keep going, change something, look into it further or stop.

You can talk, build, sell and deliver in the same week, with each one feeding the others.

Useful moves for each guess

Customer: can you reach them? Try to book five conversations in a fortnight with people who fit, and get at least two from outside your own network. If you can't, you've learned how hard selling will be once your contacts run out.

Problem: is it real, and what does it cost? Ask about the last time it happened, what they did about it and what it cost them. Better still, ask them to show you how they handle it today.

Promise: does it make sense to them? Put a one-page website with your promise and a price in front of people who fit your customer, or show them a rough version. Notice what they do next: ask questions, book a call, ask when they can start, or move on.

Solution: does it work? Let a few customers try a rough version on their real work, or deliver the outcome by hand for them. Sitting with someone while they use it shows you what the product really needs, often things nobody would have mentioned in a conversation.

Commercial: will they pay? Ask for money: a payment, a pre-order or a signed letter of intent. See how to pre-sell software.

Before you build with Lovable or another AI builder

A rough version is one of the most useful moves, and AI builders make it cheap. Before you start prompting, write down four things: who it's for, the problem it solves for them, what this version needs to do for them to react to it, and the price you'll ask. That gives the builder a clear brief, tells you what to watch for when someone tries it, and keeps the cost down, because most of what founders spend on these tools goes on fixing and rebuilding. See what it costs to build an app with Lovable, Bolt or Replit.

Decide beforehand what would change your mind

When you're running a specific test, write down the result that would change your mind before you start. Otherwise every result looks encouraging, and "two people seemed interested" becomes a reason to carry on. For example:

If fewer than [3] of the [5] [people who fit my customer] I speak to describe [the problem] from the last [fortnight] without me prompting them, I'll look again at whether it's common enough.

It won't always give you a clear answer, and it isn't meant to. It keeps you honest about what you're seeing.

Expect messy results

Results are rarely clean. People sound interested and then go quiet, feedback conflicts, and one surprising reaction can change what you thought you were testing. Look at what people did more than what they said, check whether conflicting reactions come from different kinds of customer, and treat surprises as the most useful thing you've learned. It's normal to come back to the same question two or three times as your offer changes.

Then keep moving. Trying things costs far less than it used to, for your competitors too, so waiting has become the riskier choice. When you're unsure what to do next, pick the move that puts something real in front of customers sooner.