You've got an idea. It's 2 AM and you can't sleep because you keep imagining what it could become. You've already named the company in your head, designed the logo, planned the Series A pitch. The only problem: you have no idea if anyone actually wants it.
This is where most startups die — not because the idea is bad, but because the founder never bothered to ask. They spend six months, a year, or longer building something in isolation. Then they launch to silence. Not because the product is broken, but because they were solving a problem nobody has.
The good news? This is preventable. And the solution is simpler than you think.
Why Your Intuition Is Probably Wrong
You believe in your idea. That's a requirement. But belief and customer demand are not the same thing. Your mom thinks your idea is great. Your friends say "yeah, that's cool." Your internal monologue has already crowned you the next unicorn founder. None of that means you have a business.
The trap is subtle. When you tell someone your idea and they respond positively, you interpret that as validation. What you're actually getting is politeness. People are nice. They don't want to hurt your feelings. "That's a great idea" costs them nothing to say, and it makes you happy, so they say it.
What you need to hear is: "I would pay money for that" or even better, "I would change my behavior right now to use that."
Those are rare. Your job is to find them before you start building.
The Mom Test: Asking Questions That Matter
Rob Fitzpatrick's "The Mom Test" reframes customer conversations. The core idea is simple: stop pitching. Stop asking leading questions. Ask about their actual problems and behavior instead.
Bad question: "Don't you think it would be cool if there was an app that helped you organize your recipes?"
Good question: "Tell me about the last time you cooked something new. How did you find the recipe? What was frustrating about that process?"
The difference matters. The first question invites agreement. The second reveals reality.
When you're interviewing potential customers, you're not selling. You're investigating. You want to understand their world well enough that your product idea seems obvious to them, not the other way around.
Here's the framework: Start broad. Ask about their day, their challenges, how they currently solve the problem you think you can solve. Listen for specifics. When they mention something relevant, dig deeper. Ask "why" and "then what" until you understand the full context.
Take notes. Not on their feedback about your idea, but on what they reveal about their actual behavior. Do they actually care about this problem enough to spend time solving it? How much would they pay? What's their current workaround?
Finding Your First 10 Potential Customers
You can't validate a business idea by talking to nobody. But you also don't need 1,000 interviews. Ten good conversations will tell you more than you think.
Start with people who have the problem most acutely. If you're building project management software for freelancers, talk to freelancers. Not entrepreneurs in general. Not "people who might use project management tools someday." Actual freelancers who are actively juggling multiple clients right now.
Where do you find them? LinkedIn, Reddit, Facebook groups, industry forums, Twitter. Attend meetups or conferences. If there's a community around your target customer, spend time there. You're not recruiting. You're just looking for people who fit the profile.
When you reach out, be honest. Don't pretend to be a researcher or hide that you're exploring a business idea. Say: "Hey, I'm looking to talk to freelancers about how you manage projects. I'm exploring an idea in this space and I'd value your perspective. Could we grab 20 minutes this week?" Most people will say yes. Some will even be excited to help.
Aim for ten conversations. Spread over two weeks. After that, you'll see patterns. You'll hear the same problems repeated. You'll start to understand the real opportunity.
The Gap Between "Cool" and "I'll Pay For That"
This is the hardest distinction to make, and it's where most founders deceive themselves.
Listen for evidence of genuine problem intensity. Does this person complain about the problem? Do they spend time on workarounds? Have they already tried to solve it? Do they talk about it multiple times in the conversation? These are signs the problem is real enough to sell against.
Then ask directly, without ego: "How much would you pay to solve this?" or "Would you buy this if it existed?"
Most people will give you a nice answer. What you're looking for is the number that makes them think twice. "$50?" might get a nod. "$500?" reveals their true valuation. If they balk at a number that's meaningful to you, that's information.
Even better is a commitment. "Would you be willing to be an early customer?" Not "would you consider it?" but "are you in?" The difference is behavioral. One is hypothetical. One is real.
The Smoke Test: Before You Code
Once you've validated the problem exists, you need to test if your solution is the one people want.
A smoke test is simple: create a landing page that describes your solution. Drive traffic to it. See if people sign up for a waitlist, pre-order, or express interest. You're testing if the positioning and messaging resonate, not if you have a perfect product.
As an illustrative example: a founder believes contractors need better time tracking. Instead of building an app, they put up a landing page describing the solution and run Google Ads targeting contractors. $2,000 in ad spend, 250 waitlist signups from people actively searching for "time tracking for contractors." That isn't proof of product-market fit. It is proof that the messaging and the problem resonate before any code is written.
You don't need a perfect landing page. You need one that accurately describes what you're building and why it matters. Then drive real traffic to it. Real traffic means real people, not your network.
When to Pivot vs. When to Persevere
You've had ten conversations. You've run a smoke test. You've learned something you didn't expect. Now what?
If the problem validation is strong but your solution isn't resonating, pivot the solution. Keep the problem, change the approach.
If the problem validation is weak — people don't really care, they're solving it fine without you — step back. You can either find a different problem this customer segment has, or you can find a different customer segment that has this problem more acutely.
What you don't do is keep building. Every week you spend coding before validation is a week you could have spent talking to ten more customers. The biggest advantage young founders have over big companies is speed. Don't waste it building in a vacuum.
The goal of validation isn't to de-risk everything. It's to de-risk the biggest assumptions: Do people care? Would they pay? Is there a viable path to customers? Once you can answer those, you're ready to build.
Your idea has survived the first real test. Now the work actually begins.
If you want to compress the time between idea and a validated product customers actually want, Arepa builds your full business plan and grounds it in real market research, so you start your first ten conversations with a point of view instead of a blank page.