Validation guides
Market Validation: How to Test Demand Before Writing Code
Code is the most expensive way to find out nobody wants something. Market validation answers that question first, with signals that already exist and small tests that need no product at all.
Market validation is testing whether enough buyers want what you plan to sell, at a price that works, before you build it. Start with signals that already exist (money spent on the problem today, competitors, hiring for the manual work), then run a demand test that asks buyers to act: click a price, join a paid pilot, or pre-order.
It is worth doing before the code because poor product-market fit is one of the most common ways companies die. In CB Insights' study of 431 venture-backed companies that shut down from 2023 onward, 43% failed on poor product-market fit (CB Insights, 2026).
Key takeaways
- Market validation is about buyers, not features. It asks whether a group of people will pay, not whether your product is good.
- Start with signals that already exist: current spend, competitors, hiring for the manual version, and where buyers talk about the problem.
- Then run a demand test that asks for an action, such as a click on a price, a reply to outreach, a paid pilot or a pre-order.
- Competitors are usually a good sign. They show people pay to solve the problem.
- Decide your pass mark before you test, or any result will look like a yes.
What is market validation?
Market validation is the evidence that a reachable group of buyers wants a solution to a specific problem badly enough to pay for it. It sits between having an idea and building a product.
It helps to separate three questions founders often blur together:
| Question | What it tests | Typical evidence |
|---|---|---|
| Idea validation | Is the problem real, for someone specific? | Interviews about recent, specific experiences |
| Market validation | Will enough of those people pay, and can you reach them? | Current spend, competitors, responses to a priced offer |
| Product validation | Does your solution actually solve it? | Usage, retention and repeat purchase of a working version |
This guide covers the middle row. If you have not confirmed the problem yet, start with our business idea validation checklist.
Which market signals can you check before building anything?
Look for evidence that money already moves around the problem. People paying for a workaround, companies hiring someone to do the job by hand, and competitors charging for a solution are all stronger signals than general interest.
- Current spend. What does the buyer use today: a tool, a contractor, a spreadsheet and three hours a week? Something they already pay for, in money or time, is a budget you can compete for.
- Competitors and their prices. Published pricing tells you what buyers already accept. Paul Graham's view is blunt: "Even if you find someone else working on the same thing, you're probably not too late. It's exceptionally rare for startups to be killed by competitors" (Paul Graham, "How to Get Startup Ideas").
- Hiring for the manual version. Job postings for the role your product would help or replace show that companies spend real money on the problem today.
- Where buyers talk about it. Forums, communities and review sites where people complain about current tools tell you what is missing and give you a place to find people to talk to.
- Timing triggers. A new regulation, a platform change or a price rise can create demand that did not exist a year ago.
None of these proves that you can sell. They tell you whether there is a market worth testing.
How do you test demand without a product?
Describe the product as if it exists, put it in front of the buyers you identified, and measure an action that costs them something. Four tests work well, from cheapest to strongest.
| Test | How it works | What it tells you |
|---|---|---|
| Outreach reply rate | Send a short, specific message about the problem to buyers you can name | Whether the problem is urgent enough for strangers to answer |
| Explainer or offer page | A video or page describing the product, with a price and a sign-up | Interest at a stated price, not just interest |
| Concierge pilot | Deliver the outcome by hand to a few buyers | Whether they come back, and what they actually need |
| Pre-order or paid pilot | Ask for money, or a signed commitment, before building | The strongest evidence short of a repeat sale |
Dropbox is the best-known example of the second test. Before the product was broadly available, Drew Houston posted a three-minute demo video aimed at early adopters. In his words, "Our beta waiting list went from 5,000 people to 75,000 people literally overnight" (TechCrunch, Eric Ries).
Note what that did and did not show. A waitlist proves interest from people who can be reached through a particular channel. It does not prove they will pay. For a paid product, pair it with a price, then follow up personally with the people who respond.
How big does the market need to be?
Big enough that the buyers you can realistically reach, at your price, support the business you want. The headline market size matters much less than that reachable slice.
A useful way to size it is bottom-up: count the buyers you could actually reach, multiply by what each would pay a year, and compare that with the revenue you need. A large top-down figure from an industry report rarely tells you whether your first 100 customers exist.
Graham makes the case for starting narrow: "You can either build something a large number of people want a small amount, or something a small number of people want a large amount. Choose the latter" (Paul Graham). A small group that wants the product urgently is easier to reach, easier to learn from and more likely to pay.
The free TAM, SAM, SOM calculator walks through the bottom-up version, from total market to the share you can serve and win.
How do you read market validation results honestly?
Decide the pass mark before you run the test, test with the buyer you actually plan to sell to, and count actions rather than opinions. If you set the bar after seeing the result, almost any result looks like a yes.
A few rules keep the reading honest:
- Write the threshold down first. For example: "If fewer than 5 of 40 cold messages to operations managers get a reply, the problem is not urgent for this buyer." The numbers are yours to choose; the point is choosing them in advance.
- Keep the buyer constant. Replies from friends, other founders or a general audience do not count toward a B2B buyer's demand.
- Weight actions over words. A clicked price, a booked call, a deposit or a signed letter of intent outranks any number of "sounds great" replies.
- Treat a weak result as information. It usually points at the wrong buyer, the wrong channel or the wrong price, and each of those can be tested again.
Where does NELL fit in market validation?
NELL does the desk research for the signals above, so your time goes into the demand tests only you can run.
Quick Validate is free and unlimited, and gives a first read on an idea: a viability score, a verdict and next moves. It runs without live web research, so treat it as a filter for which ideas deserve a test.
DeepValidate adds live web research: named competitors, sourced market sizing, demand signals and a score across 11 dimensions in 5 pillars, including "Is the market worth it?" and "Can we win?". Every claim links to its source. See what that looks like in the sample report.
Neither replaces the demand test. A report can tell you the market exists; only a buyer can tell you they will pay you.
Frequently asked questions
What is the difference between market validation and market research?
Market research describes a market: its size, its players and its trends. Market validation tests whether buyers in that market will act on your specific offer, for example by replying, paying or signing a commitment.
Can you validate a market without a product?
Yes. Outreach, an offer page with a price, a demo video, a concierge pilot delivered by hand and pre-orders all test demand before anything is built.
Is a waitlist proof of market demand?
Not on its own. A waitlist shows interest from people reachable through one channel. Pair it with a price and personal follow-up to learn whether those people will pay.
Are competitors a bad sign for a new startup?
Usually not. Competitors show that people already pay to solve the problem. The question is whether you can serve a group of buyers they serve badly, and why those buyers would switch.
How long does market validation take?
It depends on how easy your buyers are to reach. The existing-signal checks take days. Demand tests run as long as it takes to get enough responses from your target buyer to compare against the threshold you set.
Start where you are
The signals are public. Collecting them is the slow part.
Sources
- CB Insights, Why Startups Fail: Top 9 Reasons (March 2026)
- Paul Graham, How to Get Startup Ideas (November 2012)
- Eric Ries in TechCrunch, Dropbox Minimal Viable Product (October 2011)
Example thresholds in this guide are illustrations, not benchmarks. Set your own based on your buyer, channel and price.