Validating Your Idea Before You Build
The fastest way to know if your business idea is worth pursuing — before you spend months building something nobody wants.
The Most Common Startup Mistake
Most founders build too much before they know if anyone wants what they're building. They spend months — sometimes years — creating a product, then discover the market doesn't want it, or won't pay for it.
Validation is how you avoid this. The goal is to answer one question as cheaply and quickly as possible: "Will people pay for this?"
The Validation Hierarchy
Not all positive signals are equally meaningful. Here's the hierarchy from weakest to strongest — and why each one is worth what it's worth.
From weakest to strongest evidence
| What it looks like | Why it's weak or strong | |
|---|---|---|
| 1. Verbal enthusiasm | "That's a great idea!" | Means almost nothing — people are polite, and agreeing costs them nothing |
| 2. Email signups | "Sign up for early access" | Slightly better than words, but a free click is still a very cheap commitment |
| 3. Letters of intent / pre-orders | A specific commitment to buy, even if payment isn't collected yet | Real validation starts here — the person had to actually think about committing |
| 4. Someone pays you | Money changes hands, even a deposit | The only real proof — everything else can be politeness in disguise |
Move toward money as fast as possible. Every step up this hierarchy filters out people who were just being nice.
Five Ways to Validate Quickly
1. Smoke Test / Landing Page. Build a simple page describing your product with a "Get early access" or "Pre-order" button. Send a small amount of traffic to it. What you're measuring: does anyone care enough to sign up or pay?
2. Concierge MVP. Do the service manually before automating it. If you're building software that matches freelancers to clients, do the matching yourself by hand first. This tests the value proposition without building anything.
3. Wizard of Oz Test. The customer sees a normal-looking product. Behind the scenes, humans are doing the work that will eventually be automated. You validate the experience without building the technology.
4. Pre-Sales. Sell the product before it exists. Take deposits or full payment. If people won't pay for it before it's built, the idea needs work — this is the fastest route to tier 4 of the hierarchy above.
5. Customer Interviews (with a specific question). Run 10–15 conversations focused on one question: "Tell me about the last time you had [the problem you're solving]. What did you do about it?" You're not pitching — you're listening to find out if the pain is real and how people currently deal with it.
Which Validation Method Fits Your Idea?
Which Path Fits You?
Do you already understand the problem well, or are you still not sure it's real?
Good Signals vs. Bad Signals
| Good signals | Bad signals | |
|---|---|---|
| Unprompted commitment | People try to pre-pay without being asked | "That's interesting — I'd definitely use that" |
| How they describe the problem | Multiple people describe their problem in almost the same words | "A lot of people would want this" |
| Existing behavior | People are currently paying for an inferior solution to the same problem | Enthusiasm with zero willingness to pay, share contact info, or commit |
| Their reaction to your offer | "When can I get it?" | Vague future interest with no present-tense commitment |
How Long Should Validation Take?
Aim for 2–4 weeks of focused validation effort before committing to building. A reasonable target is 3–5 paying customers (or pre-orders), or 50+ qualified signups for a free product. If you can't hit that with active effort in 4 weeks, the idea likely needs to change — and finding that out now, not six months from now, is a genuine win, not a failure.
Key Terms
Key Terms
- Concierge MVP
- Manually delivering a service by hand before building any automation, to test the value proposition cheaply.
- Wizard of Oz test
- A test where the customer experiences what looks like a finished product, while humans do the work behind the scenes manually.
- Smoke test
- A simple landing page or offer used to measure real interest before a product exists.
- Validation
- The process of gathering evidence — ideally paid evidence — that people actually want what you're planning to build.
Discussion & questions
Ask a question about this lesson or share your take.
Loading…