The Idea You Haven't Tested Is Just an Expensive Opinion

There's a particular kind of Sunday afternoon that a lot of would-be founders know well. You've got a notebook open, a half-cold coffee, and an idea that has been rattling around your head for three weeks. It feels good. It feels obvious. And the next thing you do — almost every time — is start building. You buy the domain. You sketch the logo. You spend four weekends on a website nobody has asked for.

Six months later, the site is live, the inbox is quiet, and the only thing you've learned is that you can build a website.

The uncomfortable truth is that the excitement you feel about an idea tells you nothing about whether strangers will pay for it. Research on startup failure keeps landing on the same finding: roughly 35% of failed startups say they didn't fill a market need, and around 43% point to poor product-market fit. Those aren't engineering failures. They're failures to ask a question early enough.

An untested idea isn't a business. It's an opinion you've decided to fund.

The good news is that testing an idea is cheap, fast, and mostly unglamorous. What follows is a 30-day approach you can run on evenings and weekends, with no code, no investors, and a budget that fits in a single grocery run.

Week 1: Write down the bet you're actually making

Before you talk to anyone, write one sentence in this shape: "I believe [specific person] has [specific problem] and would pay [specific amount] to solve it."

Most ideas fall apart at this step, which is exactly why it's worth doing. "An app for busy people" isn't a bet — it's a mood. "Freelance graphic designers lose two hours a week chasing late invoices and would pay $12 a month to automate the chasing" is a bet. It names a person, a cost, and a price. It can be proven wrong, which is the whole point.

Then list the assumptions hiding inside that sentence. There are usually three or four: that the problem happens often enough to hurt, that people currently solve it in some clumsy way, that they have budget, and that you can reach them without spending a fortune. Rank them by how badly you'd be sunk if you were wrong. The one at the top is what you test first — not the fun one, the fatal one.

A quick way to check whether you've written a real bet: could a friend read it and immediately name someone who fits the description? If they can't picture the customer, neither can you.

Week 2: Have ten conversations, and resist pitching

Talk to ten people who match your customer description. Not ten friends. Ten people who actually live the problem — found through a niche subreddit, a Facebook group, a professional Slack, an old colleague, the counter of a shop you frequent.

The hard part is that you must not pitch. The moment you describe your solution, the conversation dies. People are polite, they say "oh, that's a great idea," and you walk away with a warm feeling and zero information. Instead, ask about the past:

• "Walk me through the last time this happened."
• "What did you do about it?"
• "How long did it take, and what did that cost you?"
• "Have you ever paid for anything to make it better? What happened?"

Past behavior is evidence. Future intentions are not. Someone who says "I'd definitely use that" has told you nothing. Someone who says "I paid a virtual assistant $200 last quarter to do exactly that, and she quit" has told you almost everything — the problem is real, it recurs, and there's already money moving.

Listen for two signals in particular. First, existing spend: what people already pay, in money or hours, to work around the problem. Second, emotional temperature: mild annoyance rarely converts into a purchase; a problem that makes someone visibly tired is a problem worth solving. If ten conversations produce ten shrugs, that's not failure. That's you saving a year of your life for the price of ten coffees.

Week 3: Manufacture demand before you have a product

Now put a claim in front of strangers and see if anyone reaches for it. You need something that looks real and costs almost nothing to make.

The classic version is a one-page site: a headline that states the promise, three lines of explanation, a price, and one button. The button doesn't go to a checkout — it goes to a short form or a waitlist. Drive a small amount of traffic to it (a $50 ad test, a post in the community where you found your interviewees, a comment thread where the problem is being discussed) and measure what fraction of visitors act.

The concierge test deserves special mention because it's the one people skip. Instead of building software, you do the work by hand for three to five customers. If your idea is an automated invoice chaser, you personally send the reminder emails. It doesn't scale, it's tedious, and it is by far the fastest way to find out that the real problem was something adjacent to what you assumed. Founders routinely discover in week three that customers don't want the thing they described — they want the boring half of it, done reliably.

Set your threshold before you look at the results. Decide in advance what "this is working" looks like: for example, 20 email signups from 300 visitors, or three of five interviewees agreeing to a paid trial. Deciding afterward is how people talk themselves into continuing.

Week 4: Ask for money, because that's the only real vote

Everything up to here is warm-up. The test that matters is whether someone will part with money before the thing exists.

This feels rude to a lot of first-time founders. It isn't. You're not tricking anyone — you're offering a clear deal: a discounted price, an early-access slot, a refundable deposit, a paid pilot. Say plainly that it launches in six weeks and that they can have their money back at any point. Then watch what happens.

Ten people saying "I love this" is a conversation. One person handing over $30 is data.

Small numbers are fine. Three pre-sales from ten serious conversations is a strong signal for a side business. Zero pre-sales from thirty conversations is also a strong signal, and it's telling you something useful: either the problem isn't painful enough, the price is wrong, or you're talking to the wrong people. Those are three very different fixes, and your interview notes will usually tell you which one applies.

If you can't charge yet — some services are regulated, some products need a prototype — get the next best thing: a signed letter of intent, a calendar booking, a deposit held in escrow, a commitment to a specific start date. The principle is the same. Make people spend something they'd rather not spend: money, time, or reputation.

What to do with a "no"

Here's the part nobody puts on the motivational poster: most ideas fail this process, and that's the process working. The cost of finding out in 30 days is a few hundred dollars and a month of evenings. The cost of finding out after you build is a year and a chunk of your savings.

When a test fails, don't throw away the whole idea reflexively. Look at which assumption broke. Wrong customer is a pivot you can make in a week — the same product sold to agencies instead of individuals often changes everything, because agencies have budgets and individuals have credit cards. Wrong price is a smaller fix than founders think; underpricing kills more small businesses than overpricing does. Wrong problem is the one that means starting over, and it's also the one you'll be grateful to have learned cheaply.

Keep a running document of what you tested, what you predicted, and what actually happened. After three or four cycles you'll notice something: your predictions get better. That skill — being able to guess what the market will do and be right more often than not — is worth more than any single idea you'll have.

The short version

Validation isn't a bureaucratic step between having an idea and starting a business. It is the start of the business. You're just doing sales and customer research before you've built inventory, which is the order that costs the least.

Four weeks, four questions: What exactly am I betting? Do ten real people have this problem? Will strangers reach for it? Will anyone pay before it exists? Answer those honestly and you'll either have a business with early customers already attached, or you'll have your Sunday afternoons back — with a much better idea of what to do with them.

Either outcome beats spending a year building something into silence. Start with the conversation you've been avoiding. It's usually the one that tells you the most.