Buying an engine removes the risk of building the wrong software. It doesn’t remove the risk of aiming it at the wrong market. Before you license anything, or before you spend a dollar on ads for the one you just bought, run it through a validation funnel first. It takes a weekend, not a quarter.
The funnel
Think of validation as four narrowing stages, each one a real test, not a hypothetical one:
1. 100 people hear the idea. Not a survey — a plain description of what you’re building and who it’s for, said out loud or written in a DM, to real people who match your target customer. You’re not asking “would you use this?” (everyone says yes to be nice). You’re watching how they react when you describe the problem, before you’ve even mentioned the solution.
2. ~20 talk to you. Of the people who reacted with genuine recognition — “oh, that’s annoying, tell me more” — a fraction will actually get on a call or reply with detail. This is your real signal group. If nobody from 100 wants to talk more, the problem isn’t the solution — it’s that you haven’t found people who feel the pain yet.
3. ~8 try a smoke test. A smoke test is the smallest possible commitment that isn’t free: a waitlist with a deposit, a pre-order, a “pay $10 now to get access when we launch.” This is where wishful thinking dies and real intent shows up. Words are cheap; a card number is not.
4. ~2 pay for the real thing. Once the engine is live under your brand, convert smoke-test believers into actual paying customers. This is your first real cohort — the ten to fifty people from the first 90 days playbook.
The ratios above are a rough starting expectation, not a law of physics — but if your numbers are wildly worse at every stage, that’s the market telling you something before you’ve spent real money finding out the hard way.
What to do if the funnel breaks early
If stage one goes flat — nobody reacts when you describe the problem — the fix isn’t a better pitch. It’s a different audience or a different problem. Don’t polish your description; change who you’re describing it to.
If stage one works but stage two doesn’t — people nod, nobody wants to talk further — you likely found a problem people acknowledge but don’t prioritize. Painful-but-not-urgent problems are notoriously hard to sell into, no matter how good the software behind them is.
If stage two and three work but stage four stalls after launch — people believed you before seeing the product, but the live thing doesn’t match the promise. That’s a product-market fit gap you can usually close fast, since the engine itself is proven; it’s almost always messaging, onboarding, or price.
Why this matters even more when the product is already built
It’s tempting to skip validation entirely once you own a working engine — the software risk feels solved, so why bother testing anything? But the funnel isn’t testing the software. It’s testing you, in your market, with your price and your voice. That’s the one part Good Kulture can’t sell you pre-validated, because it’s genuinely yours to prove.