Do not test your idea by building it. Test it by trying to get 5 people to book time with you about a specific problem, at a specific price, in the next 3 weeks. You are looking for people who move their own calendar to make time for it. Compliments about the idea count for nothing.
If 5 conversations produce 2 people asking what it would cost to start, you have something. If they produce compliments, you have a hobby with an audience.
Why the usual advice fails here
Most advice about validating a business idea was written for people who have already left. It assumes you have entire days to run experiments, a budget to spend on ads, and the emotional distance to hear a no without it landing on your mortgage.
You have none of that yet. You have evenings, a salary that is still arriving, and a decision you are trying to de-risk before it becomes irreversible. That is a different problem, and it has a smaller answer.
The other failure is subtler. Asking people whether they like your idea is a question about them being polite. Almost nobody will tell a colleague they respect that their plan sounds thin. So you collect a stack of encouraging replies, read them as demand, and resign on the strength of good manners.
There are only 2 categories of response worth counting, and neither of them is enthusiasm.
- Somebody spent something. Time on their own calendar, an introduction that costs them social capital, or money. All 3 are real.
- Somebody described their own problem back to you, unprompted, in detail. People only do this about things that are actually costing them.
The test, in 5 conversations
The whole thing fits in about 3 weeks of evenings. You are not selling. You are finding out whether a specific problem is expensive enough that somebody would rather pay to have it gone.
- Name 1 problem rather than a service. “Help with marketing” is not a problem. “Your launch emails go out late because nobody owns the calendar” is a problem. Somebody has to recognize themselves in the sentence.
- Make a list of 15 people who plausibly have it. Former colleagues, people 1 step removed, anyone who has complained about this in your hearing. You need 15 to reach 5.
- Ask for 20 minutes about their situation. A specific ask with a specific length gets a real yes or a real no, where an open invitation to coffee gets a maybe.
- In the conversation, ask what they have already tried. This is the whole test. Somebody with an expensive problem has a history of attempts. Somebody without one has opinions.
- Say what you would charge, and stop talking. Give a number and a scope, then wait. The silence is uncomfortable and it is where the information is.
5 is not an arbitrary number. It is small enough to finish in 3 weeks around a job, and large enough that a clean result is unlikely to be luck. If 4 of 5 conversations go the same way, that direction is telling you something. For the sequence that turns those answers into an actual offer and a price, that is what the first toolkit is built to run.
What actually counts as a yes
This is where most people mislead themselves, so it is worth being blunt about the ranking. Not all positive responses are the same weight, and the gap between the top and bottom of this table is the gap between a business and a year of wasted evenings.
| What they did | What it means | Weight |
|---|---|---|
| Asked what it would cost to start | They have moved from your idea to their budget. This is the strongest signal short of payment. | Very strong |
| Paid a deposit | The only signal that cannot be walked back politely. | Very strong |
| Introduced you to someone else with the problem | They spent their own credibility. People do not do this casually. | Strong |
| Described a failed attempt in detail | The problem has already cost them money or time. | Strong |
| Asked you to follow up after a specific event | Real, but conditional. Put a date on it and see if it survives. | Moderate |
| Said the idea was great | They are being kind. Count it as zero. | None |
| Offered to help you think it through | Generous, and still zero. They are buying nothing. | None |
2 very strong signals out of 5 conversations is a green light to keep going. 1 is worth another 5 conversations before you decide. Zero, with 5 people who all had the problem, means the problem is real and your version of the solution is wrong, which is a much better position than it feels like.
The question is never whether the idea is good. It is whether the problem is expensive.
What to do with a no
A clean no in week 3 is worth more than a maybe in month 9, and it costs you nothing you were not already spending. The mistake is treating it as a verdict on the whole plan rather than on 1 specific framing.
Before you throw the direction away, check which of these you actually got, because they point in completely different directions.
- Wrong problem. Nobody recognized the sentence you opened with. Change the problem, keep the audience.
- Right problem, wrong buyer. They agreed it was painful but it was not their budget. Move up or across, and try the person who owns the number.
- Right problem, right buyer, wrong price. They flinched at the number rather than the idea. That is a scoping question and it is the easiest of the 3 to fix.
- Right everything, wrong moment. They have a reason it cannot be now. Get the date and go back.
Only the first one is a reason to change direction. The other 3 are adjustments, and people quit on them constantly because a no feels the same regardless of which kind it was.
When to hand in your notice
Demand is 1 of 2 numbers you need, and it is the one people focus on because it is the exciting one. The other is how long you can operate without it, and that number decides how much room the first one has to be wrong.
A reasonable bar before you resign: 2 very strong signals, 1 of them with a date attached, and enough runway to survive twice as long as you think you need. Most first businesses take longer than planned to reach a first payment, and the founders who make it are usually the ones who could afford to be late.
If the runway half is the part you have never actually counted, the runway toolkit is the one that turns it into a real figure instead of a feeling.
What not to wait for
Do not wait for certainty, a finished website, a business name you love, or the moment the idea stops feeling presumptuous. None of those arrive on a schedule, and none of them are evidence. The 2 numbers are the evidence.
Common questions
What if I cannot talk to people in my industry without tipping off my employer?
Run the test 1 step out. Former colleagues who have left, people at companies you have never worked for, or the adjacent market that has the same problem. The signal you are looking for does not require insiders, and a slightly colder list often gives you a more honest read anyway.
Is 5 conversations really enough to bet a year on?
It is enough to decide whether to keep going, which is the only decision in front of you right now. It is not enough to build a forecast on. Treat it as a gate rather than a guarantee: 5 conversations tell you whether the next 15 are worth having.
What if I do not know what to charge yet?
Pick a number you can say without apologizing and use it as an instrument rather than a commitment. You are measuring the reaction rather than signing a contract. A number that is obviously too high still tells you more than refusing to name one.
Everyone was positive but nobody has paid. What does that mean?
Usually that the problem is real and cheap. People will happily agree that something is annoying without it ever reaching the threshold where they spend money on it. Look for the version of the same problem that has a cost attached, and talk to whoever pays that cost.
Should I build a small version first to make it easier to sell?
Almost never at this stage. Building is the most expensive way to ask a question you can ask in a conversation, and it tends to make you defensive about the answer. Build once somebody has told you what they would pay for.
