Validation is the cheapest insurance a founder can buy. It's the difference between spending a weekend to learn an idea is weak and spending six months to learn the same thing. Yet most “validation” is theater — asking friends if they'd use it and hearing a polite yes.
Real validation produces evidence that strangers will exchange something valuable for what you're offering, before you've fully built it.
The three things you're actually validating
- Demand: do enough people have this problem, badly enough to act?
- Willingness to pay: will they part with money (not just attention) to solve it?
- Reachability: can you get in front of those people repeatably and affordably?
An idea can pass one and fail another. Plenty of “wanted” products die because nobody will pay, or because acquiring each customer costs more than they're worth.
Signals ranked by how much they mean
Not all validation is equal. From weakest to strongest:
- “That's a great idea” (means nothing).
- An email signup (weak — attention is cheap).
- Time committed: a demo call, a survey, a waitlist with intent questions.
- A scarce resource committed: their team's data, a pilot slot, a referral.
- Money up front: a pre-order, a paid pilot, a deposit. This is the gold standard.
The goal of any validation experiment is to move up this ladder as fast as possible.
Cheap experiments that produce real signal
- The pre-sell: a landing page describing the outcome with a real “Buy / Reserve” button. Drive a little traffic; measure intent-to-pay, not just clicks.
- The concierge test: deliver the outcome manually for a handful of paying users before automating anything.
- The competitor teardown: if rivals have paying customers, their reviews and pricing already validate demand — your job is to validate your wedge.
- The outreach test: can you book five conversations with target buyers in a week? If you can't reach them now, distribution will be your real problem.
Borrowed validation: the fastest shortcut
The quickest way to de-risk an idea is to start from one that's already validated by someone else's revenue. If a competitor is doing real MRR, demand and willingness-to-pay are proven; you only need to validate that your specific angle — a niche, a feature, a price — is better for a segment they underserve.
That's why studying how successful startups work is itself a validation step. Trustyr does this at scale, turning revenue-verified businesses into the proof and the wedge so you validate the delta, not the whole idea.
When to stop validating and start building
Validation has diminishing returns. Once you have (a) evidence of demand, (b) at least one form of money-or-scarce-resource commitment, and (c) a repeatable way to reach buyers, you've learned what research can teach you. The rest is learned by shipping — see From Idea to MVP in 30 Days.