Most new ventures do not fail from bad execution. They fail because nobody checked whether anyone wanted the thing, at a price that works, through a channel the founder could actually reach.
Four tests. Run them in this order. Each one can kill the idea, which is the point.
Test one: does the problem exist, in their words?
Talk to fifteen people who fit your target. Do not pitch. Ask how they handle the problem today, what it costs them, what they have already tried, and what happened.
The bar: they describe the problem unprompted and can quantify the pain. If you have to explain why they should care, you are educating a market — that is a much more expensive business than the one you thought you were starting.
Test two: will they pay, before it exists?
Interest is free. Money is information.
Pre-sell. Take deposits. Sign a letter of intent with a number in it. Sell the service manually before building anything to deliver it at scale.
The bar: a stranger — not a friend, not a former colleague being supportive — commits money or a signature. Three of those beat a hundred people saying it sounds great.
Test three: do the delivery economics work?
Deliver it manually, at least a few times, and track everything: your real hours, direct costs, the rework, the parts you did not anticipate.
The bar: at the price people actually paid in test two, there is margin left after honest cost — including paying someone other than you to do the work. A business that only works because the founder is unpaid is a job with extra risk.
Test four: can you reach them repeatedly and affordably?
One sale from your own network proves nothing about scale. Find the channel: search, referral partners, trade relationships, local presence, outbound.
The bar: you can name a repeatable way to reach ten more of these buyers and you have tested it once. "We will do social media" is not a channel, it is a hope.
What "validated" actually means
All four cleared, not one. The common trap is passing test one enthusiastically — everyone agrees the problem is real — and treating that as permission to spend. Real problems with no willingness to pay, no margin, or no reachable channel are still bad businesses.
How long this should take
Weeks, not quarters. If validation is taking six months, you are building instead of testing. The entire purpose is to spend small amounts of money and time to avoid spending large amounts.
When to kill it
Kill it when a test fails twice and you cannot explain why the next attempt would be different. Founders are not usually short on persistence — they are short on the discipline to stop, which is what makes the next idea affordable.
When to go
All four pass, you have money in the bank from real buyers, you know your unit economics, and you know where the next ten customers come from. Now build. Now you are funding something with evidence under it instead of enthusiasm.