You launched a landing page. You posted it on X, dropped it in three Slack groups, and 1,400 people gave you their email address. It felt like the ground moved.
Here's the uncomfortable part. You've learned almost nothing about whether anyone will pay you.
A waitlist signup costs a person six seconds and zero dollars. That's the entire problem. You built a test where the cost of a "yes" is roughly the same as the cost of a "no," and then you read the results as if they meant something. They don't. Waitlist validation is the most popular self-deception in early-stage startups right now, and 2026 is the year founders started saying so out loud.
The fix isn't complicated. It's just harder. You have to make saying yes cost something.
Why doesn't a waitlist prove anyone wants your product?
A waitlist proves people are curious. It doesn't prove they have a problem, that the problem hurts enough to pay for, or that they'll change what they currently do. Curiosity and demand feel identical on a dashboard and behave nothing alike in a checkout flow.
Think about what you're actually asking someone to do. You're asking them to type an email they already use for newsletters into a box, in exchange for the possibility of something interesting later. There is no downside for them. There is no decision. Most people who join a waitlist aren't evaluating your product, they're bookmarking a vibe.
That's why the numbers look so good and convert so badly. Industry benchmarks put landing-page-to-waitlist conversion around 3.4% for SaaS, 4.1% for consumer apps, and 4.6% for AI tools, and the median waitlist turns about 11% of its page visitors into signups. Those are healthy top-of-funnel numbers. But waitlist-to-paying-customer sits somewhere between 5% and 25%, averaging near 20% only if you convert people within a month, and dropping under 10% once you make them wait longer than three months.
Run that math on your 1,400. If you take three months to launch, which you will, you're looking at maybe 100 paying customers in the best case and 70 in the realistic one. If your price is $20/month, that's a business doing $1,700 MRR at peak hype. Worth knowing before you spend eight months building.
And that's the optimistic version, where everyone on the list is a real prospect. Some of them are your friends. Some are competitors. Some are founders who joined to study your onboarding email.
What is the money test?
The money test is a single question: did this person give up something scarce to say yes? Money is the cleanest version of scarce, but time, reputation and access count too. If the answer is no, you have interest. If the answer is yes, you have evidence.
One founder building a Stripe failed-payment recovery tool wrote a postmortem that puts it better than any framework: he'd written a 50-page product spec and shipped a landing page before talking to a single customer, and got zero signups in five days. His conclusion: "Validation is a person handing you money."
That's the bar. Not a survey response. Not a "this is sick, ship it" reply on X. Not a signup. A transfer of something the other person can't get back.
The useful thing about this framing is that it scales down. You don't need a product to run it. You need an ask that has teeth.
What signals actually count as validation?
Rank every piece of evidence you have by what it cost the other person. Anything free sits at the bottom and should never drive a build decision. Here's the ladder, weakest to strongest:
Tier 1: free and anonymous. Page views, social likes, poll votes, "would you use this" survey answers. Cost: nothing. Use: traffic testing only. Never quote these numbers to yourself or an investor as demand.
Tier 2: free and identified. Waitlist signups, newsletter subscribes, fake-door clicks, replies to a cold email. Cost: a few seconds and an email address they may not check. Use: testing which message lands, not whether the product should exist.
Tier 3: time and access. A 30-minute discovery call they showed up to on time. An intro to their boss. Sending you their actual spreadsheet. Cost: real, because calendar time is finite. Use: problem confirmation. This is where you learn whether the pain is real.
Tier 4: reputation and commitment. A signed letter of intent, a referral to a peer, agreeing to be a named design partner. Cost: their credibility is now attached to your thing. Use: strong B2B signal, especially at higher price points.
Tier 5: money. A pre-order, a deposit, a paid pilot, a subscription that starts before the product is finished. Cost: obvious. Use: this is validation. Everything else is a step on the way here.
Most founders live in tiers 1 and 2 for months and call it traction. The jump from tier 2 to tier 5 is where you find out if you have a company.
And when you do move money into the equation, the behavior on the other side changes immediately. Deposit-based waitlists have been reported converting 3x to 5x better than free lists, with email open rates in the 60% to 80% range versus 15% to 25% for free signups. Same people, same product, different filter. Paying $10 turns a stranger into someone who remembers your name.
How do you run a money test without a product?
You pre-sell. The mechanics depend on your model, but the principle is identical: describe the thing precisely enough that someone can decide, then ask for the transaction before the thing exists.
Waseem Daher, a three-time YC founder, did this before Pilot wrote a line of code. He went to small business owners and got them to commit to paying for the accounting service once it launched. Those commitments were the validation. The product came after.
Five versions that work depending on what you're building:
The deposit. $20 to $100, refundable, holds your spot in the first cohort. Works for consumer and prosumer products. Stripe Payment Links take about ten minutes to set up and you don't need an app to accept them.
The pre-order. Full price, charged now or on launch, usually at a discount. Works when the value is obvious and the product is concrete. This is the whole Kickstarter model compressed into a landing page.
The paid pilot. For B2B: charge $500 to $5,000 for a scoped, time-boxed engagement you deliver manually. You get paid to learn. The customer gets the outcome. You get a case study and a reference before you've written any software.
The letter of intent. When procurement makes a real payment impossible, get a document stating they intend to buy at a stated price on a stated timeline. It's weaker than cash and much stronger than enthusiasm.
The concierge. Do the job by hand for one paying customer. Notion, spreadsheets, and your own labor. If nobody will pay you to solve the problem manually, they won't pay software to solve it either.
The hardest part of all five isn't the setup. It's that you have to ask, and asking is where founders discover their idea was a hypothesis. That discomfort is the test working correctly.
Keep the results somewhere structured rather than in your head. You want the ask, the response, the objection, and the amount, per person, so you can see patterns instead of remembering the two conversations that went well. A spreadsheet is fine. So is Notion, or a planning tool like Foundra that walks first-time founders through validation before the build phase. The tool matters far less than writing down the nos.
How many paid commitments do you need before you build?
Enough that the pattern can't be luck, which for most early products means somewhere between 5 and 20 paying commitments from strangers. The word doing the work in that sentence is strangers.
Three signals that mean you're clear to build:
- Five or more paid commitments from people you didn't already know. Friends and former colleagues buy out of loyalty. Their money is real and their signal isn't.
- A close rate above roughly 20% on qualified conversations. If you're pitching 50 people to get one deposit, the problem is either the problem or the price, and you should find out which before scaling the pitch.
- At least one unprompted referral. Someone paid, then told a peer without you asking. That's the earliest form of the growth loop you'll live on later.
There's a companion heuristic worth keeping: if you can't find ten strangers who describe the problem in their own words without you prompting them, you don't have demand yet. You have a hypothesis with a nice landing page.
What if nobody pays?
Then you've saved yourself the eight months, which is the entire point. A failed money test is the cheapest useful outcome available to a founder. But before you kill the idea, separate the three things it might be telling you.
It might be the ask. You buried the price, apologized for it, or made the commitment vague. "Would you pay for this?" is not an ask. "It's $49, here's the link, first cohort starts October 6" is an ask.
It might be the audience. You pitched people who have the problem mildly to people who have it severely. Severity is what opens wallets. Go find the ones who've already built a broken workaround for this, because a workaround is proof of pain.
It might be the idea. Sometimes people are polite, interested, and completely unwilling to pay. That's a real answer. Take it.
The order matters. Most founders conclude "the idea is dead" when they actually ran a bad ask at a soft audience, and most of the rest conclude "I just need better messaging" when the idea is dead. Test the ask twice with a sharper audience before you decide.
How do you turn a waitlist you already have into a real signal?
Email it with a paid offer. Today. You're sitting on the cheapest test available to you and it takes one message.
Send the list a note that says the first cohort is opening, it costs X, and there are a limited number of spots. Include a payment link. Then read three numbers: how many opened, how many clicked, how many paid. That third number is your actual validation, and you'll have it by Friday.
Expect it to be brutal. A 1,400-person list that produces four purchases is telling you something clear, and it's better to hear it now than after the build. A 1,400-person list that produces sixty is a business.
If you can't bring yourself to charge yet, run the intermediate version: ask for a 20-minute call. Count how many book it and show up. Time is tier 3, which is weaker than money but far stronger than the signup you already have.
One more thing worth doing regardless of the result: ask the people who don't buy what would have to be true for them to say yes. The non-buyers explain your pricing, your positioning, and your roadmap better than the buyers do.
Key takeaways
- Waitlist signups measure curiosity. They cost the signer nothing, so they predict almost nothing about revenue.
- Waitlist-to-paid conversion runs 5% to 25%, and falls below 10% if you make people wait more than three months.
- Rank every signal by what it cost the other person. Free and anonymous is worthless, money is validation, and time and reputation sit in between.
- You can run a money test without a product using a deposit, a pre-order, a paid pilot, an LOI, or a manual concierge offer.
- Five to twenty paid commitments from strangers, a close rate above 20%, and one unprompted referral is a reasonable bar to start building.
- If nobody pays, check the ask and the audience before you blame the idea.
- If you already have a waitlist, email it a real paid offer this week. That one message is worth more than the last six months of signups.
FAQ
Is a waitlist ever useful?
Yes, for distribution and messaging. A waitlist gives you a list to launch to and a cheap way to test which headline pulls. Treat it as a marketing asset, not as evidence that your product should exist.
What's the difference between a smoke test and a pre-sale?
A smoke test measures intent: someone clicks a button for a product that isn't built. A pre-sale measures commitment: someone enters a card. Both are useful, but only one of them survives contact with an investor's questions.
How much should I charge for a pre-order or deposit?
Charge something close to your intended price. A $1 deposit filters out nobody and teaches you nothing. Most consumer and prosumer products land between $20 and $100 for a refundable deposit, and B2B pilots start around $500.
Is it dishonest to sell something I haven't built?
Not if you're clear. Say plainly that it doesn't exist yet, give a date, and offer a full refund if you miss it. Founders get into trouble by implying the product is ready, not by pre-selling.
What if my product is free or ad-supported?
Substitute a different scarce resource. Ask for 30 minutes of their time, a piece of their data, or an invite to three friends. The test isn't money specifically, it's cost.
How long should a money test take?
Two to three weeks. If you need longer, you're building instead of testing. Set a date, make the ask to a defined list of people, and read the result on the date.
