Most founders validate the wrong way. They ask friends, run a survey, get a bunch of "that's a great idea!" responses, and start building. Six months later they have a product nobody uses.
Real validation is about testing whether strangers — people with no reason to be polite — will actually do something to get what you're offering. Not say they like it. Do something.
Here are 6 methods ranked by how much real signal they give you, from weakest to strongest.
The 6 methods — ranked by signal strength
Check whether people are searching for the problem you're solving. Use Google Keyword Planner, Google Trends, or Semrush. If the problem keywords have zero search volume, the market is either too early or doesn't exist. If there's volume, you have a starting point — not proof.
How to do it: Go to Google Trends and compare your problem keywords against related terms you know have real demand. Look for steady or growing interest, not spikes.
What it tells you: Whether the problem exists at scale. Not whether people will pay you to solve it.
If multiple companies are building in your space, making revenue, and growing — the problem is real and people pay for solutions. No competitors isn't a good sign; it usually means the problem isn't worth solving commercially.
How to do it: Find 3–5 competitors. Check their revenue (G2, Capterra reviews mention pricing), their growth (LinkedIn company growth, job postings), and their customer complaints (1-star reviews, Reddit threads).
What it tells you: The market exists and someone is making money. Where the current solutions fall short — which is your opportunity.
Talk to 10–15 people who match your target customer profile. Ask about their problem, their current workflow, and what they've already tried. Do not pitch your idea. Listen for: how often they face the problem, how much it costs them, and whether they've ever paid for something to fix it.
The Mom Test rule: Ask about the past ("Tell me about the last time this happened") not the future ("Would you use something that..."). People lie about the future. They can't lie about the past.
Red flags: "That's a great idea, you should build it!" Green flags: "I've tried [tool X] and it doesn't do [specific thing] — I'd pay anything for that."
Build a one-page site describing your product. Drive cold traffic — from Reddit, community Slack groups, LinkedIn outreach, or small paid ad spend — and measure how many strangers click your sign-up CTA. A 5%+ click-through rate from cold traffic is a meaningful signal.
Build it in: Carrd (free), Framer (free tier), or raw HTML. It should have a clear headline, 3 benefit bullets, and a single CTA button. Take one day maximum.
Critical rule: Traffic must come from people who don't know you. Your Twitter followers and email list are not cold traffic. They're biased toward supporting you.
If you can get strangers to pay you — even a small amount — before the product exists, you have strong validation. A pre-order at even 30% of your intended price is more meaningful than 10,000 email signups.
For B2B: Ask companies to sign a Letter of Intent (LOI) — a non-binding document saying they'd buy the product if built. Even a verbal "yes" on a recorded call where they give a specific number is meaningful. See the full B2B validation guide →
For consumer: Use Gumroad, Stripe, or a simple PayPal link. Set a price. If people pay, you have validation. If they don't, you have information.
The ultimate validation is a paying customer who keeps paying. Before writing complex code, ask: what is the smallest thing that delivers the core value? Build that. Sell it manually. If 5–10 people pay for the manual version, automate it.
This is how many great SaaS companies started — doing things manually at first (Concierge MVP), using existing tools like Airtable and Zapier, or delivering the output of their "software" as a Google Sheet.
Common validation mistakes that waste months
Asking the wrong question
"Would you use this?" is not a validation question. It's a politeness question. Ask: "When did you last try to solve this? What did you try? Did you pay for anything?" The past reveals real behavior. The future reveals wishful thinking.
Validating with warm audiences
Your followers, newsletter subscribers, and LinkedIn connections are pre-disposed to support you. Validation from them is 10x weaker than the same signal from cold strangers. Build the discipline to test on people who don't know you.
Confusing interest with intent
1,000 email signups is interesting. 10 people who clicked "Buy Now" even though the product didn't exist is validation. Email collection measures curiosity. CTA clicks on a real-looking product page measure intent. Pre-orders measure commitment.
Building during validation
Once you start building, your brain switches into execution mode and you stop listening. Do all your validation before you write significant code. It's much easier to pivot a landing page than a half-built product.
How to know when you're done validating
| Stage | Minimum validation signal |
|---|---|
| Problem validation | 8+ interviews confirm the same painful problem exists |
| Demand validation | 5%+ CTR on landing page from cold traffic, OR 50+ cold email signups |
| Willingness-to-pay | 3+ pre-orders, OR 2+ LOIs (B2B), OR 1 paying pilot customer |
| Solution validation | First paying customers use the product and come back voluntarily |
You don't need to complete all stages before starting to build. Most founders start building after demand validation, then use early customers to validate the solution.
Frequently asked questions
How long does validation take?
Problem validation (interviews) takes 1–2 weeks. Demand validation (landing page + traffic) takes 2–3 weeks. Willingness-to-pay validation takes another 1–2 weeks. Total: 4–6 weeks if you move fast. That's a rounding error compared to the 6–12 months you'd spend building the wrong thing.
What if my idea has no competitors?
Usually a red flag, not a green one. It typically means either the problem isn't big enough to build a business around, or people have already tried and failed. Dig into why before taking it as a sign of "untapped opportunity."
How do I validate without showing my idea to competitors?
Be vague about the solution, specific about the problem. "I'm researching how companies handle [workflow X]" reveals nothing proprietary while still giving you real signal from interviews. Your idea isn't what gives you an advantage — your execution does.
Get honest feedback on your idea or landing page
Submit your project to HelpMarq and get structured written feedback from real people in 48 hours — on whether your value prop is clear, what's missing, and what would stop someone from signing up.
Get feedback — free →