Pre-Sell

You sell the thing before it exists. Real money in, real card, real commitment. You deliver later, or you refund if it was only a test, so nobody pays for something they never get. It's the most honest demand test there is, because money is the currency that's hardest to fake.

The question it answers

Will the interest convert to payment? Not "do they like it." Not "would they use it." Will they pay.

That makes this a desirability test, the will-anyone-want-it lens, asked in its strongest form. A survey tells you people approve of the idea. A click tells you a stranger was curious for half a second. A credit card tells you they'll spend their own money on it. Those are three different facts, and only the last one still counts once it's their own money on the line.

Underneath every demand signal sits a hierarchy. Say it flat: money beats time beats an email address. An email costs nothing, so it proves almost nothing. Time costs a little. Money costs the most to give, which is exactly why a yes paid in money is worth the most. When someone pays, you've learned something. When someone nods, you've learned they're polite.

The taxonomy this experiment sits inside comes from David Bland and Alex Osterwalder's Testing Business Ideas, which catalogs these tests by the kind of evidence each one buys. Pre-sell buys the strongest evidence on the desirability shelf.

How to run it cheaply

You don't need the product. You need a way to take money for it.

1. Build a buy page, not a build. One page that sells the thing as if it ships. The promise, the price, a real checkout. Stripe is the cheap way in here, the service that lets you take a card without building your own checkout. An afternoon, not a sprint.
2. Make the offer concrete. A price and a date. "$49, ships in 30 days" pulls a real decision out of a person. "Coming eventually" pulls nothing.
3. Write the success number down first. How many paid yeses by what date tell you to build it, and below what count you refund everyone and walk away from the idea. Pick it before launch so it can't bend afterward.
4. Send real traffic and take real cards. A small ad budget, a list, a post. Charge the card. Refunding later doesn't weaken the test, because the answer arrived the moment they paid.

Cap the whole thing inside two weeks. If building the pre-sell takes longer than that, you've started building the product you were supposed to be testing.

A worked example

Here's one I ran myself, on my own book, 5 Weeks to Hired. I built a buy page for it before I'd written a single page, told people it would ship in 30 days, and watched whether cold strangers would actually pay. My rule of thumb is to aim for roughly 1% of cold traffic converting, so that was the bar I was watching against. I even pre-sold the name itself, running ads behind more than 50 candidate titles to see which one pulled hardest before I committed to it.

Get the money to move before you build. A paid yes from a cold stranger is the cleanest read you can buy, and a refund afterward costs you nothing but the answer.

When to reach for it, and when not

Reach for it when the cheaper demand signals already came back positive and the only question left is whether interest converts to payment. This is a top-of-the-ladder rung. Climb to it after a landing-page test or a fake door, a button or page that advertises a thing that doesn't exist yet to measure how many people click to buy, not before.

Don't reach for it first. It costs more to set up than a fake door and asks more of the customer than a landing page, which only collects an email or a click and never makes anyone reach for a card. Use those cheaper rungs to learn if anyone's curious, then pre-sell to learn if anyone's serious.

Don't confuse it with its neighbors. A concierge test asks for effort, not money. You deliver the service by hand first, before building anything, and if enough people want it that the manual version stays busy, you've learned the automated version is worth building. A Wizard of Oz test fakes a working feature to learn if people use it. Both buy a softer yes than a paid one. And if you're selling to businesses on large deals, where nobody swipes a card on a buy page, the right tool is a letter of intent, a signed non-binding statement of intent to purchase. Pre-sell is for the consumer-scale yes you can actually charge for.

The neighbor recipes (landing-page test, fake door, concierge, Wizard of Oz, letter of intent) each get their own catalog page; the cross-links go live once those routes are wired.