The most common way to lose money on a business idea is not bad execution. It is building something carefully, competently and on time — that nobody was waiting for.
CB Insights has spent years collecting startup post-mortems, essays written by founders explaining why their company died. Across 110-plus of those post-mortems, the single most cited reason was no market need, at 42%. When they rebuilt the study in 2024 across 431 venture-backed companies that had shut down since 2023, the same cause came top again at 43%.
Two different samples, a decade apart, landing within one percentage point of each other. That is about as stable as a finding gets in this field.
What "no market need" actually means
It is a soft phrase for a hard outcome. In practice it resolves into one of four things, and they are not equally difficult to detect:
- The problem is real but not painful enough. People agree it is annoying. They will not pay to remove it. This is the most common and the most expensive, because everyone you describe it to nods.
- The market is smaller than the deck assumed. Usually because a top-down number was taken from an industry report and never narrowed to the segment that could actually be served.
- Someone already serves this customer. Often not a direct competitor — a spreadsheet, a WhatsApp group, or the person who has always done it manually and is good enough.
- The buyer is not who you think. The person with the problem, the person who chooses, and the person who pays are three different people more often than founders expect.
The uncomfortable part: most of this is checkable
None of those four require a launch to discover. They require asking. The reason they get discovered after the money is spent is not that the information was hidden; it is that asking felt slower than building, and building felt like progress.
A useful discipline is to write down, before you spend anything, the single assumption that would kill the idea if it were false. Not a list of ten. One. Then ask what evidence would settle it, and how much that evidence costs. In our experience the answer is usually a few weeks and a fraction of one month of burn.
What good evidence looks like
There is a hierarchy, and most people stop too early on it.
- Weakest: friends and family say it is a great idea. This is social, not commercial.
- Weak: a survey where people say they would pay. Stated intent is famously unreliable; people are polite.
- Better: structured interviews about what they do today, what it costs them, and who they already pay. Past behaviour beats future intention.
- Strong: what is physically on the shelf, at what price, moving how fast. What distributors will actually stock.
- Strongest: someone gives you money, or a deposit, before the thing exists.
A note on market-size numbers
Industry reports are useful for direction and dangerous for decisions. They are usually top-down, often two or three years old by the time they are quoted, and frequently disagree with each other by wide margins. A figure with no date and no named source attached to it is not evidence, it is decoration. If a number is going to justify spending, it should carry the organisation that published it and the year it refers to.
The cheapest thing you will ever buy
A validation exercise that ends in "do not do this" feels like a wasted fee for about a week. Then it feels like the best money you ever spent, because the alternative was finding out the same thing eighteen months later with a team, a lease and a loan attached.
Testing this properly
We run this as a fixed-price service. A desk read with a written verdict starts at ₹10,000 and takes five working days; the full engagement puts one of our people in your target market for three days. See the four validation plans.
Why 42% of startups fail, and the part you can check in five days