How to Validate a Business Idea Before Spending Money — How-To #022
Go deeper (affiliate link — we may earn a commission): The Mom Test by Rob Fitzpatrick — the short book on asking customers questions they can’t lie to you about — the skill this whole post depends on.
To validate a business idea before spending money, say it in one sentence, ask ten strangers about the problem, sell it before you build it, do the math, and set a kill line. Five cheap tests give you proof while it still costs almost nothing. About one in five new businesses closes within its first year, according to the Bureau of Labor Statistics. Most of them didn't die from bad work. They died from building something nobody asked for — and the owner only found out after the money was gone.
Disclosure: some links on this page are affiliate links. As an Amazon Associate, Changenaire earns from qualifying purchases made through links on this page, at no extra cost to you.
Key takeaways
- Say the idea in one sentence a stranger understands: whose problem, and how they pay for the fix.
- Ask ten strangers about the problem, not the product — friends will tell you it is great because they like you.
- Sell it before you build it: a one-page site and pre-orders or deposits are proof; compliments are worthless.
- Do the math out loud: price minus delivery cost is your margin, and monthly costs divided by margin is how many sales you need.
- Set the kill line before you fall in love: write the go and no-go numbers down before the results come in.
Here's the rule this whole post runs on: proof before purchase. Not enthusiasm — proof. Enthusiasm is free, and friends hand it out by the bucket. Proof is a stranger describing a problem they already pay to solve, or a stranger handing you money for something that doesn't exist yet. Everything below is a way to get proof while it's still cheap.
Yesterday's installment — How to Trademark Your Business Name (How-To #021) — put a fence around the name. This one asks whether the thing inside the fence is worth building at all. It should have come first in the series, honestly. It comes now because it's the step that saves you from everything after it.
Five tests, in order. None of them needs a dollar you don't already have.
Say it in one sentence a stranger understands
Before you test an idea, you have to be able to say it. One sentence. Three parts: whose problem, and how they pay for the fix.
"I help busy parents in my neighborhood get a week's worth of home-cooked dinners delivered on Sundays for $85." Problem, buyer, price — all in one breath. Compare that with "I'm building a food-tech platform leveraging AI to disrupt meal prep." The second one sounds impressive at a dinner party and means nothing to a customer. If you can't say it plainly, you don't understand it yet — and no amount of building will fix that.
The test is simple: say the sentence to a stranger and ask them to repeat it back. If they can, the idea is clear enough to test. If they can't, rewrite it until they can. Most ideas need three or four rewrites before they survive this step, and every rewrite is free. The expensive version of this step is a logo, a website, and a lease for an idea nobody can repeat back to you.
Ask ten strangers about the problem — not the product
Now go talk to ten people who match your buyer. Not your friends, not your family — strangers. Friends will tell you the idea is great because they like you. Strangers have no reason to lie, which is exactly why their answers are worth something.
And ask about the problem, not the product. "Tell me about the last time dinner fell apart on a Sunday" beats "Would you buy my meal service?" every time. The first question gets you a story — what happened, what it cost them, what they did instead. The second gets you a polite yes from someone who will never buy. There is a whole book about this trap, Rob Fitzpatrick's The Mom Test, and its one idea is worth the whole cover price: compliments are worthless; evidence is everything.
What counts as evidence: the person describes the problem in their own words, unprompted. They already pay for a workaround — a worse product, a hack, their own time. They can name what it costs them. Our working bar: if five of the ten describe real pain and already spend money or time on it, you have a signal worth building on. Fewer than that, and you either have the wrong buyer or the wrong problem — both fixable now, while fixing them is free.
Go deeper (affiliate link — we may earn a commission): The Mom Test by Rob Fitzpatrick is the short book on asking customers questions they can't lie to you about — the skill this whole step depends on. Optional reading; the steps above stand on their own.
Sell it before you build it
Conversations are evidence. Money is proof. The step most builders skip is the one that matters most: ask someone to pay before the thing exists.
Put up a simple one-page site describing the offer — what it is, what it costs, when it arrives — and take pre-orders or deposits. A page costs almost nothing: a free form, a $19-a-year page builder like Carrd, a domain for about $12 a year. Then send it to the people you interviewed, post it where your buyer already hangs out, and watch what happens. Ten pre-orders is a business with a pulse. A hundred email signups and zero payments is a compliment with a mailing list. Those pre-orders are the start of your first ten customers — the same conversations, carried one step further.
Yes, taking money for something unbuilt feels forward. That's the point — it's the only test that can't be faked. A deposit is a stranger saying "I believe this enough to risk money on it." Refund it cheerfully if you don't hit your threshold; the refund costs you nothing, and the information is worth more than the sale. The builders who skip this step don't skip the test. They just take it later, after the inventory is bought and the lease is signed, when failing it costs everything.
Do the math out loud
An idea can pass the conversation test and the pre-order test and still be a bad business — because the numbers don't work. So do the math before you fall in love with the revenue.
Write it down, on paper, where you can see it: the price of one sale, minus the cost of delivering one sale, equals your margin. Then divide your monthly costs by that margin. That's how many sales you need every month just to stand still. If the answer is 200 and your pre-order test found 10 buyers, you don't have a pricing problem — you have a math problem. Fix the price, cut the cost, or pick a different idea. Setting the price itself — the floor, the lanes, the worked example — is How to Price Your Product or Service.
Do it with real numbers, not hopeful ones. The cost of delivering includes your time, the packaging, the payment fees, the delivery — all of it. A $50 product that costs $45 to deliver and an hour of your time is a hobby with a website, not a business. The math test is the least romantic step in this post and the most protective one. It takes twenty minutes. It has saved more builders than any business plan ever written.
Set the kill line before you fall in love
Here's the step that separates builders who validate from builders who perform validation: write down your go and no-go numbers before the results come in.
Decide now, while you're honest: "If fewer than five of ten interviews show real pain, I stop. If fewer than ten people pre-order, I change the offer or walk away. If the math needs more than fifty sales a month to break even, I reprice or I don't build." Write it down. Show it to someone. Then run the tests and honor what you wrote — even when the answer hurts, especially when it hurts.
Because by test three, you will be in love with the idea. That's human. The kill line is the part of you that was honest before love arrived, and it's the only part that can save you from building a beautiful thing nobody wants. A "no" from the tests is not a failure. It's the test working. A pivot now — a different buyer, a different price, a different problem — costs you two weeks. A pivot after launch costs you a year and whatever you spent. The cheapest funeral in business is the one you hold before the first dollar goes out.
Validate first. Then build. The order is the whole lesson.
Go deeper (affiliate link — we may earn a commission): The Startup Owner's Manual by Steve Blank and Bob Dorf is the full manual for this process — customer development, step by step, with the worksheets. Optional reading; the steps above stand on their own.
At a glance
| Test | The proof it gives |
|---|---|
| One sentence | Problem, buyer, price — in one breath |
| Ten strangers | Real pain, in their words, about the problem |
| Sell before you build | Pre-orders or deposits — money is proof |
| The math out loud | Margin, and how many sales cover monthly costs |
| The kill line | Go / no-go numbers, written before the results |
Common mistakes
Asking friends and family instead of strangers. Friends will tell you it is great because they like you. Fix: talk to ten people who match your buyer and have no reason to lie to you.
Asking about the product ("Would you buy this?"). A yes/no about a product that does not exist tells you nothing. Fix: ask about the problem — "Tell me about the last time dinner fell apart on a Sunday" beats a yes/no about a product that does not exist.
Building before selling. The builders who skip the pre-order step take the test later, after the inventory is bought and the lease is signed, when failing it costs everything. Fix: put up a one-page site and take pre-orders or deposits first — money is proof.
Skipping the math because the conversations went well. An idea can pass the conversation test and the pre-order test and still be a bad business, because the numbers don't work. Fix: write it down — price minus delivery cost is the margin, and monthly costs divided by margin is the sales you need to stand still.
Falling in love before setting the kill line. Without numbers written down first, every result looks like a reason to keep going. Fix: write the go and no-go numbers down before the results arrive, and honor them.
Questions people ask
How many people do I need to talk to to validate an idea?
Ten is our working number — enough to see a pattern, few enough to finish in a week. If five of them describe real pain and already spend on a workaround, that is a signal. Thirty interviews is procrastination dressed as research; three is a coin flip.
What if everyone says they love the idea?
Then you have learned nothing yet. "Love it" is a compliment, and compliments are free. Ask about the last time the problem actually happened — and ask them to pay before you build it. Money is proof; praise is not.
How much does it cost to validate a business idea?
Almost nothing — $0 to $50. Conversations are free, a one-page site costs about $12 a year for a domain, and a page builder like Carrd runs about $19 a year. The expensive mistake is building first and finding out later.
What is a kill line?
The go and no-go numbers you write down before the results come in — for example, if fewer than five of ten interviews show real pain, stop; if fewer than ten people pre-order, change the offer or walk away. Deciding now, while you are honest, is what separates validating from performing validation.
What comes next
The idea passed. Now it needs an audience — a place where strangers can find it, trust it, and come back for the next thing. Tonight, the series turns to the cheapest distribution a builder has ever had: How to Start a YouTube Channel From Zero — How-To #023. No studio, no budget, no excuses — a channel built on the phone already in your pocket, and a schedule that turns viewers into customers.
Sources
Figures in this guide are as of November 2026 unless noted. Prices change — check the source page linked here before you spend.
- U.S. Bureau of Labor Statistics — Business Employment Dynamics — about one in five new businesses closes within its first year; roughly half survive five years.
- Carrd — the one-page site builder named in this guide.
- Rob Fitzpatrick, The Mom Test — the book behind the "compliments are worthless; evidence is everything" rule in this guide.
Subscribe for more free step-by-step how-to guides to help you get more things done to grow your business or project everyday.
Email required. Unsubscribe anytime.
Read our Privacy Policy.
