How to Validate Your Business Idea Before You Start It
Adebayo Adebisi28 June 202623 min readThe cheapest place to discover you are wrong is before you have built the thing.
The most expensive mistake you can make in business is to build a product before validating that people want to buy it. In this post I will show you how to validate your business idea before you start it, whatever you are planning to sell.
By the end you will have a clear order of operations: what to do first, what to do second, and the point where you are finally allowed to build. Do this, and you will never again spend your months, your savings and your energy building something that nobody pays for after you launch it.
The path almost everyone takes
Let me describe how most people start a business or launch a new product.
You get an idea. It feels like a good one, and the more you think about it the better it gets. So you start building. If it is a product, you find manufacturers and you build the product. If it is a service, you map out the whole service, end to end. You buy the domain, pay someone to build the website and sort out the logo too. Maybe you register the company and spend even more money buying fancy software you think you might need. And in a lot of cases, I've seen people rent an office space. So after many weeks, sometimes months, sometimes a year, you are still busy doing something, so you assume you are making real progress.
Then you finally launch. And almost nobody buys. Maybe a few friends and that's it. The customers you pictured while you were building do not show up. All the money you were projecting to make didn't come either.
Now you have a problem, because you still do not know what went wrong. You cannot tell whether the idea was wrong, or the product was wrong, or the price was wrong, or you built the right thing for the wrong person, or your new social media account doesn't have enough followers. The list goes on.
At this stage a lot of founders have spent everything they had, and are facing the prospect of having to spend more just to work out what the issue is.
So what actually went wrong? In almost every case it is the same thing. You built the whole business, product or service before you asked a single person to pay for it. The test of whether anyone actually wanted it came during and after you launched. The fix is to make that test one of the first things you do. Before you build or launch anything, you find out whether people will really pay, and you have to get them to commit. If you change your mind about the idea, it will cost you almost nothing in comparison. You'll have real data to base your decisions on. Then you only build what people have already shown you they will buy.
The mistake that kills most businesses
There is a research firm called CB Insights that studies startups and businesses for a living, and one of the things they track is why companies fail. When a business shuts down, the founders will often write publicly about what went wrong. CB Insights gathers those write-ups, along with the notices companies put out when they close, and sorts the failures by cause. Their latest study looked at 431 venture-backed companies that shut down since 2023.
The reason that sits at the top of the list is running out of money, at about 70%. That is the one most people would guess. But CB Insights made an important point about it. Running out of money is almost always how a business finally dies, but it is never the real why.
Let me be a bit dark for a second and think of it as the line written on a death certificate. It tells you the business ran out of cash, but not what was actually wrong with it. The cash runs out because something underneath was already broken, and most of the time that something is simple: not enough people were buying. So when CB Insights look past the money to the real causes, the biggest one is poor product-market fit, at 43%. In simple terms, the company built something and then found out not enough people wanted to pay for it.
That number has been steady for over a decade. The older version of the same research, many years prior, put "no market need" at 42% and listed it as the single most common reason startups failed. Different studies, years apart, landing on the same conclusion. The most common root cause of business failure is building something the market did not want.
So think about what that means for you. You can have a problem with your idea, or a problem with your product or service, and you will not find out which until you have built the whole business and put it in front of real buyers. By then, your time is gone, your money is gone and your soul is gone. The reason this keeps happening to good people with good ideas is that they find out too late.
Why we build before we sell
If selling your product or offer first is the better approach, why does almost everyone build first? It's simple: because building lets you feel productive without ever experiencing rejection. Selling is the truth, where you find out whether anyone will actually pay. I tell people all the time. No one cares about you, your product or your service. They care more about what pain it can solve for them.
Let me be fair to building your product or service for a second. When you are making the thing, you are getting somewhere real. The product gets better. The website starts to look like an actual business. Your messaging starts to take shape. At the end of the day you can point at something you made. And nobody can take that away from you. However, it's a false sense of security.
Selling is where that safety ends. Selling is putting your offer in front of a real person who can look at it and walk away. It is uncomfortable, it's painful, it's rejection, and it tells you your idea was a dud. This is the reason people avoid it early on. We expect every idea we have to have a 100% success rate. Selling and money in the bank is the only real test. It is the one moment the market actually tells you the truth.
The longer you build on your own, the more certain you become in your mind. You have founder bias. Every week you spend improving the thing makes you more sure it will work. You start treating your own confidence as evidence. But confidence is just a feeling. The only evidence that counts in business is someone handing over money. Someone saying they will buy or use it is not enough evidence.
Be careful who you listen to here, because the most comfortable feedback is usually the most useless. Your friends and family telling you they love the idea is not validation. They love you. They are never going to look you in the eye and say they would not spend a penny on it. The only feedback worth anything is the kind that costs the other person something, and the cleanest version of that is cash. Real money from a real buyer is the test.
There is a useful way Alex Hormozi puts this. He says you are not trying to create demand, you are trying to channel it. People either already want the result you are selling or they do not, and you cannot build your way into a market that was never there in the first place. Building harder, building better does not magically create buyers. It just makes you waste far more time and money before you stumble across the eventual truth.
The Proof-Before-Build Sequence
So how do you actually find out if people will pay, before you build anything? You run five questions, in order. I call it the Proof-Before-Build Sequence.
The logic behind it is simple. Each question costs you a little more than the one before it, and every single one of them costs almost nothing next to building the whole thing. For example: a conversation is cheaper than building a landing page or a webpage. A simple web page or landing page is cheaper than a presale. A presale, or even just taking a deposit, is cheaper than building the whole product or service. So you go through the questions in order, trying to get it wrong while getting it wrong is still cheap. If your idea falls apart after a few conversations, that is a good day. You just saved yourself a year.
Here are the five questions.
Step 1. Whose problem is this, and is it real?
Before you build anything, you go and talk to the people you think you are building for. Not to sell to them. To understand them.
There is a short book on how to do this properly, written by a founder named Rob Fitzpatrick, called The Mom Test. The name comes from a simple problem. If you ask your mum whether your business is a good idea, she will say yes, because she loves you. And it turns out almost everyone you talk to does a softer version of the same thing. They do not want to hurt your feelings, so they tell you what you want to hear. The book is about how to talk to people so they cannot do that, even when they want to.
The mistake nearly everyone makes is asking the wrong question. You ask someone, would you buy this? And they say, yeah, that sounds great, I would definitely use it. That may feel like a yes to you. But they are being polite, and they are guessing about their own future behaviour, which people are terrible at. So you do not ask what they would do. You ask what they already do:
- What do they use right now to deal with this problem?
- How much time does it cost them?
- How much money have they already spent trying to fix it?
- What did they buy last to solve it, and did it actually work?
That is the whole rule of The Mom Test: what someone has already done is the truth, what they say they will do is just talk.
This is exactly where I started with my own consulting business. I did not start with a service or a website or anything fancy. I started with the customer in mind. Before I had anything to sell, I worked out who actually had the problem, which of those people could afford to pay me to fix it, and what fixing it was worth to them in money. I picked three types of business I understood, service businesses, tech startups with funding and ecommerce businesses, and I went and learned their problem properly before I decided what I was going to offer them.
Step 2. Can you describe their problem better than they can?
This is how you know whether Step 1 worked. After enough of those conversations, you should be able to say the person's problem back to them, in their own words, so clearly that they go, yes, that is exactly it.
If you cannot do that yet, you do not understand the problem well enough, and anything you build from here is built on your best guess. This is the cheapest checkpoint in the whole thing. Stay here until the problem is so clear in your own mind and mouth that you could explain it better than your target customer can. The more people you speak to here the better, because it is easy to have three good chats, get excited, and run off to build too soon.
For me, this is what all the conversations were really for. Every meeting and every call made my eventual offer a little sharper, because I kept hearing the same frustrations come up in the same words. I was not improving the offer in my own head and inventing potential problems. I was shaping it against what real buyers actually said to me.
Step 3. Will they raise their hand?
Now you test the offer itself, and at this stage you still have not built the product. You put the offer in front of people and count how many actually move toward it. People in business call this a smoke test. You put up something that looks real and see if anyone reacts. The simplest version is a basic landing page.
Let me give you two real examples.
The first is a company called Buffer, a tool for scheduling social media posts. Before the founder, Joel Gascoigne, built any of it, he put up one simple web page describing what Buffer would do, with a button that said Plans and Pricing. When people clicked it, expecting to buy, they landed on a short page that said it was not ready yet and asked for their email. That click was the test. It told him people wanted it enough to try to pay. Then he added a page in the middle that showed actual prices, so he could see what people were willing to pay before he built a thing. From that little two-page test, he went from idea to his first paying customer in about seven weeks, and the first payment landed within three days of him opening it up. He knew what people wanted and what they would pay before he wrote a single line of code for the product.
The second is Dropbox, the file storage company. Early on, the founder, Drew Houston, could not easily build the full product just to test it, so he made a short video showing how it would work and posted it with a link to join the waiting list. Overnight, that list went from about 5,000 people to 75,000. He had proof that people wanted this, at a scale he could never have guessed, and he had it before the product was finished.
Step 4. Will they pay before it exists?
A raised hand is good. Money is better. The strongest proof you can get, short of a finished business, is someone paying you before the thing is even built. This is pre-selling, and it is the heart of the whole sequence.
Alex Hormozi did this with his first gym. Before the doors even opened, he ran ads and pre-sold memberships, and took in enough cash in that first month to cover his rent. He found out people would pay before he had a finished gym to hand them, so the scariest question, will anyone actually buy this, was answered before he was on the hook for the full cost of being wrong.
Years before that, a company called Zappos started the same way. The founder, Nick Swinmurn, wanted to know whether people would buy shoes online. Instead of spending a fortune on stock, he went into local shoe shops, took photos of their shoes, and put them on a simple website. When someone ordered, he walked back to the shop, bought that exact pair at full price, and posted it himself. He made no money on those early sales, and that was the point. He was paying out of his own pocket to find out if the demand was real, and he found out it was before he ever risked big money on a warehouse full of shoes.
This is the part of my own story that matters most, so let me give it to you straight. When I started, I built the offer before the service behind it existed. I had no website. I had nothing built. What I had was a way to find out if the offer was real, so I went and found out. I ran small ads. I cold called businesses. I sent emails. I sat in people's DMs. I showed up in person. I flew to other countries to have the conversations face to face and still got a holiday out of it. None of it was efficient, and I did it the slow, unglamorous way on purpose, because every one of those conversations taught me something about the market I could not have learned any other way.
For about three weeks, nothing happened. I had no clients, no concrete yes, just activity and silence. That is exactly the point where most people decide it is not working and quit. I almost quit but I kept going because there was interest. And then it started raining clients. The thing that mattered most is that I kept doing the same activities after the clients came that I was doing when nothing was working. I kept consistent with the ads, the calls, the emails, the DMs, the showing up. The activity did not change when the results did. In this type of business I realised one conversation was not enough. It took three to five conversations, because I was a high-value investment for any business.
And I priced it on the revenue I could bring in, not on my hours. My base was 20 to 30% of the revenue I would generate for the client, with a bonus on top when we hit the targets, so I only won when they won. The actual delivery, the work I would do for them, I built after they said yes and paid. I pre-sold first and delivered a few weeks later. The demand came before the product, every single time for me.
Hormozi has a line that sums up the logic:
Pre-selling is not a trick to grab money early. It is the most honest test you can run, because nothing tells you the demand is real like someone paying for a thing that does not exist yet.
Step 5. Now build the smallest version that delivers.
When people pay, only now do you build. And even now, you do not build the whole dream. You build the smallest version that over-delivers the result you already sold.
This idea comes from a book called The Lean Startup, by a founder named Eric Ries, and it is worth getting right, because it is one of the most misunderstood ideas in business. Ries called it the minimum viable product. Most people hear minimum and picture a cheap, stripped down version of their product. That is not what he meant. He meant the smallest thing you can put in front of real customers that teaches you the most about them for the least effort. It is a tool for learning. The way it works is a loop, and it repeats. You build the smallest version you can. You measure what real paying customers actually do with it. And you learn enough from that to decide whether to keep going or change direction. Then you go round again, a little less wrong each time. Build, measure, learn.
So at this stage you are not trying to build something impressive. You are trying to deliver the one result your buyers already paid for, in the simplest way that works, and then learn from how they actually use it. Anything you add after that, you add because a paying customer showed you it was needed, not because you imagined it would be.
That is the Proof-Before-Build Sequence. Talk to them properly. Understand their problem better than they can. Test the offer in the real world. Take the money before you build. Then build the smallest thing that over-delivers. Five questions, in order, and every one of them is a cheaper way to find out you are wrong than the one after it.
Where to find the people to test on
Everything so far assumes you can get in front of the right people. So let me deal with the obvious question: where do you actually find them, and where do you run the test? A lot of people know they are supposed to talk to customers, and they have no idea who to talk to or how to reach them.
Start with one rule. You are not looking for everyone. You are looking for the specific people whose problem you described back in Step 1. The narrower you are about who that is, the easier all of this gets, because you start going to where those exact people already are.
The first place to look is wherever those people already gather. Every type of customer has somewhere they spend their time. There are online groups and forums, subreddits, LinkedIn, the Slack and Discord communities, trade bodies, local meetups, and industry events. Find the few places your specific customer actually shows up, and go there. Do not walk in pitching yourself or your product. Find what they complain about, and then talk to them about it, the Mom Test way.
The second route is to go straight to them, even when they do not know you yet. This is the route I took earlier, the cold outreach, and it is worth saying why it works. When you can name the exact type of person or business you are after, you can almost always find a direct way to reach them, by email, by phone, in their DMs, or in person. It is slow and it is uncomfortable, and it puts you in front of the exact people you want instead of waiting for them to find you.
If you already have an audience, a list or a following, that is a head start, but handle it with care. These people already like you, which makes them warm, and warm people go easy on you. So do not read their encouragement as proof. Push them to a real action, a call, a deposit, a pre-order, and watch how many actually move. A warm audience is a great place to find your first buyers. It is a poor place to get an honest rejection.
If you want a faster read from complete strangers, put a small amount of money behind an ad pointing to a simple page, the smoke test from Step 3. You do not need a big budget. You need enough clicks to see whether cold people, who owe you nothing, will hand over their email, book a call, or try to pay. A stranger reaching for their wallet is the cleanest signal there is, because none of them are being kind to spare your feelings.
Use a mix of warm and cold outreach. Your own audience and your network will move fastest, but they are the most biased toward you. Cold strangers move slowest, and they give you the most honest answer. If the only people interested are the ones who already love you, you have not validated anything yet. You want to see strangers have the same level of interest, too.
You have to do all of this at volume. One or two conversations will not tell you much. You are looking for a pattern, the same problem and the same words coming up again and again, and a pattern only shows up across enough people. So always talk to far more of them than you feel is necessary.
What validation is not
Before you go and run this, I want to put the false signals that make you think your idea is amazing all in one place, because this is exactly where people fool themselves. We touched some of this earlier. Here is the full list, so that whatever type of validation you get back, you can hold it up and know whether it is real.
Let's start with compliments. For example, "That is a great idea." "I love this." "You should build that." It feels good to hear but it means nothing, because it costs nobody anything to say. Praise is the cheapest thing in the world to give, and people give it to be kind. Your friends and family are the obvious version of this, and we covered them earlier, but it is wider than that. Most people you talk to will hand you a compliment instead of the truth, because the compliment is easier for both of you.
Next, when someone says, "I would definitely use that." This one feels stronger, because it sounds like a decision. It is still just a guess about the future, made by someone who wants to encourage you, and people are wrong about their own future behaviour all the time. What a person says they will do, and what they actually do when real money is on the table, are two different things.
Then there are likes, followers and views. This is the trap that fools the most people today, because the numbers are large and they feel like real demand. A like costs nothing. A follow costs nothing. The people who enjoy your content are not automatically the people who will pay you to solve a problem. You can put a post in front of a hundred thousand people and have no buyers in there at all, because they came for the entertainment, not the solution. Attention is not demand. I lived this one myself, in my consulting business.
After that, the email list or waitlist with nobody paying. This one is closer to real, because a signup is at least a small action, a raised hand. I call this a warm lead. But a free signup is not a sale. People will give you an email address for almost anything, because it costs them nothing. Some are real buyers and most are just curious, and the list alone will not tell you which is which. A signup is a hint. A payment is an answer. Do not confuse the two.
You can probably see the pattern now. Every fake signal is free for the other person to give, and every real one costs them something, so you can actually rank them. A compliment costs nothing. An email address costs almost nothing. A proper call costs them a slice of their time. A deposit or a payment costs them real money, and that is the one you can trust completely. The more a yes costs the person giving it, the more that yes is worth to you.
That is the same point Rob Fitzpatrick makes in The Mom Test, and it is worth saying in plain words: politeness is not data. The only feedback you can safely build a business on is the kind someone paid something to give you.
What to do this week
Knowing all of this changes nothing until you act on it. So here are five things you can do this week, whatever stage you are at.
- 01Find the one belief everything depends on. Every idea rests on a single belief that, if it turned out to be wrong, would take the whole thing down. For most people it is the same one: that enough people feel this problem badly enough to pay real money to fix it. Write that belief down in plain words, because it is the thing all your testing has to go after.
- 02Pick exactly who has that problem, and where they gather. Not everyone. One specific type of person or business, the kind you could actually go and find. Then list the few places they already spend their time, online or in person, and how you could reach them directly. That is your testing ground for the week.
- 03Have real conversations, the Mom Test way. Go and talk to those people, and do not pitch. Ask what they do about this problem now, what it costs them, and what they have already paid to fix it. Listen far more than you talk. Talk to more of them than feels comfortable, and make sure some of them are strangers, not just people who already like you.
- 04Put the offer in front of them and ask for a commitment. A simple page, a call, a direct message, whatever fits. Then ask for something real: a deposit, a pre-order, a paid pilot. Real money or a firm commitment, given before the thing exists, is the only answer that counts.
- 05Only build what passed. If an idea cleared those tests, build the smallest version that delivers the result you already sold, and nothing more. Then let the people paying you tell you what to add next.
The bottom line
So that is the whole idea. You find out people will buy before you build a single thing. You prove the demand is real first, and then you build for the people who have already shown you they want it.
Building something nobody wanted is the most common way a business dies, and it is also the most avoidable, because finding out is cheap if you do it at the start. A few conversations and a week of discomfort now is always better than wasting a year of your life and your savings. The cheapest place to discover you are wrong is before you have built the thing. So go and test it on real people, with real asks, before you pour your life into building for a market that might not be there. Don't build it until you've sold it.