How to test a business idea before you spend money

Start by writing the idea in one sentence that a stranger could follow: who has the problem, what it costs them today, and what you would offer. If you need a paragraph, the idea is not yet clear enough to test. Then add what they do about it at the moment, including doing nothing or asking a cousin. That existing habit is your real competitor, and it is usually cheaper and more familiar than anything you will offer at first.

Next, find people, not a market. Write down ten names of individuals or businesses who have this problem right now. Start with people who know you, because they will agree to a conversation, but include some who do not, because friends are generous and strangers are honest. Ask each for fifteen minutes to talk about their problem. Say clearly that you are not selling anything yet.

In the conversation, ask about the past, not the future. People are poor at predicting what they will buy and reasonably good at describing what they did. Ask when they last faced the problem, what they did about it, what it cost them in money or time, what annoyed them, and whether they have ever paid anyone to solve it. 'Would you buy this?' invites a polite yes. 'What did you do the last time this happened?' invites a fact. Listen more than you speak, and write down their words afterwards, not your interpretation of them.

Then look for commitments, not compliments. Something said in a meeting such as 'this is a great idea, let me know when it is ready' is a goodbye, not a sale. Better signs include an advance payment, even a small one; a written order or letter of intent where that is normal in the trade; a booked time for a trial, a visit or a delivery; an introduction to someone else with the same problem; and a real objection, which usually means the person took the idea seriously enough to argue with it. A flat refusal with a reason is also useful. It tells you what to change.

Now run a small real test that involves a customer and money. The cheapest version depends on the business, but the principle is to do by hand what you would later build. If you plan a service, do it for two or three customers before building a system. If you plan a product, make a small first batch or a sample and sell it rather than give it away. If you plan an online business, take orders through a simple form or a messaging app before there is a website. If you plan to sell to businesses, ask one for a paid pilot. Whatever the form, keep it small enough that failure costs little and real enough that a customer has to decide.

Set the rules before you start, because afterwards it is easy to explain away what you hear. Decide how many people you will approach, how many yeses of what kind would make you go ahead, how many refusals would make you change the idea or stop, the date the test ends, and what you will not spend until it has. Write it down and tell someone who will remind you. A test with no end date is not a test. It is a delay.

When the test ends, read the result honestly. Which customers said yes, and what do they have in common? Which said no, and what reason did they give? Was the problem real but your offer wrong? Was the price the obstacle, or trust, or timing? Sometimes the correct response is to go ahead. Sometimes it is to change who you sell to or what you offer. Sometimes it is to stop, and that is a good result too, because it cost you a few weeks and not your savings.

Some cautions. Do not pitch during the conversations; a founder who talks for ten minutes has learned nothing. Do not take enthusiasm from friends and family as demand, however kind it is. Do not test only the idea you are most attached to; if the answers keep pointing elsewhere, follow them. And do not skip the legal and practical basics because it is only a test. If you sell anything, give a proper bill or invoice, and check with a chartered accountant what, if anything, needs registering before you take money regularly. If you are still employed, read your contract first, because some restrict outside work or approaching the employer's customers.

A test does not guarantee success. People who said yes may not repeat the order, and people who said no may change their minds. What it does is turn a hope into a few facts and shrink the amount you risk on the first decision. That is a better position from which to spend your first rupee.