To pre-sell a service before you build the business, write a one-page offer, talk to about twenty people who have the problem, and ask the interested ones for a deposit or a signed commitment before you've set anything up. Money is the only honest test of demand. People will say they'd buy something all day long and then not pay.
Pre-selling doesn't mean taking money for something you can't deliver. It means selling a clearly described service that you will deliver by hand, with real terms, to a small number of people first. The example is Yusuf, a former hospital cook who wants to start a weekly meal-prep service for night-shift nurses and has not bought a single container yet.
Why a pre-sale beats a survey
Surveys measure politeness. Ask a hundred people if they would use a meal-prep service for shift workers and most will say yes, because yes is free. A deposit measures whether the problem hurts enough to act on. If five out of twenty people pay $60 up front, you have information. If twenty people say they love it and none pays, you have information too, and it's cheaper than finding out after renting a kitchen.
A pre-sale also tells you what to build. Customers who commit tell you what they expect: delivery times, portion sizes, dietary needs, how often. That specificity is worth more than any market-size estimate. Design partners before product describes the same idea for software, and it applies to services even more directly, since you can deliver the first version yourself.
Write the offer in one page
The page should say, in plain words: who it's for, what they get, when, how much, and what happens if they're unhappy. Yusuf's:
Night Shift Kitchen. Ten ready-to-heat meals each week, cooked on Sunday, delivered to your door or your hospital's staff room by Monday at 3 p.m. Lower-sodium and high-protein options. $78 a week. Cancel any week before Friday. First delivery starts the week of November 3, with room for 12 customers. Not happy with the first week? Full refund.
It's concrete, with a price, a start date and a cap. The cap is useful: "room for 12" is honest and gives people a reason to decide. It also limits what you have to deliver by hand.
Keep claims modest and true. If you will be preparing food for sale, find out what permits and rules apply where you live before you take a single payment. The same applies to any regulated service, such as financial advice, legal help, health care or childcare. Check the rules for your field and area first. A pre-sale does not exempt you from them.
Talk to 20 people who have the problem
Make a list of twenty people who fit, by name. For Yusuf: nurses, techs and paramedics he knows or can reach through his old hospital's staff groups, shift workers in local community groups, and the managers who organize staff meals. Where you have no contacts, how to find your first customer when you have no network shows how to build a list of strangers.
Keep these as conversations, not pitches. Ask how they eat on a night shift, what they do now, what it costs them, what they've tried. Listen for the specific pain: no time to cook, bad vending-machine options, hating the cafeteria. Describe the offer only after they've said what the problem is. Then ask: "If this existed starting in three weeks at $78 a week, would you sign up?" Watch what they do with that question. Hesitation is useful. So are objections: "I'd want it cheaper", "I have kids and need bigger portions".
If you dislike the thought of this part, how to sell without being salesy covers how to do it without pressure. And nobody feels ready is for the moment when you're convinced you should build more first. You shouldn't.
Ask for a deposit
When someone says yes, ask for money, right then or within a day. A deposit can be a full first week ($78), a smaller amount ($25) or a refundable reservation. Send a payment link from your payments provider and a short note of terms: what they're paying for, when it starts, and how to get a refund if you don't deliver or they change their mind before a stated date.
Be clear and fair. Put the refund terms in writing and honor them promptly. Don't call it an investment, and don't promise anything you aren't sure you can deliver. If you can't start on the date you gave, say so early and offer a refund. Taking a deposit is a promise, and breaking it costs you a reputation that you can't rebuild with the next twenty people.
Taxes and licensing still apply to early sales, so keep records from the first payment. If it's unclear how a pre-sale should be handled in your state, ask an accountant before you take the money.
What a yes and a maybe look like
A yes pays. It also usually asks a practical question: when does it start, can I change the order, can my coworker join. A maybe says "sounds great, keep me posted". A maybe is a no that has not been spoken yet. Count only the yeses.
Decide your threshold in advance. Yusuf might say: if 6 of 20 pay a deposit, he proceeds; if 2 do, he rethinks the price or the audience; if none do, he asks the last ten what stopped them. Writing the number down before you start keeps you from talking yourself into a verdict.
Between a yes and a maybe sits the person who says "it's too expensive". Ask what price would work, and whether they'd pay it. The answers often point to a different version of the offer: smaller portions, five meals instead of ten, a workplace-paid plan. That is useful too.
What to do if nobody pays
If twenty conversations produce no payments, don't build more. Look at what you learned. Did people not have the problem? Not enough? Did they have it but not trust you with it? Was the price wrong, or the format? Go back to the five who showed the most interest and ask directly what stood in the way. Change one thing, such as the price, the audience or the offer, and run it past ten new people.
A "no" at this stage costs you a few weeks. A "no" after you've leased a kitchen and printed menus costs far more. If you find a different problem, you've just learned something your competitors haven't. From waitlist to paying customer is worth reading if you collected interest without payments and need to find out why nobody converted.
Deliver the first one by hand
When you do get paying customers, deliver the service as simply as possible, personally, without a system. Yusuf cooks in a rented kitchen on Sundays, packs by hand, and drives the route himself. Message each customer on day two: how was it? Write down what they say. Fix the top complaint before the next week.
Do the first ten deliveries manually, even if you imagine automating them later. You'll find out what really matters and what doesn't, and your customers will see an owner who cares. Then decide what to systematize.
The conversations you had to pre-sell are the same ones you need to keep selling. Finding the next twenty people and following up with each of them is where help pays off. Rocketship finds the buyers, emails them from your own Gmail, follows up, and books the meeting on its own. Plans that get you customers start at $24.99 a month (Launch), and every monthly plan starts with a 7-day free trial, card required. See what the AI workers take over, and if you're also starting from a standing start, how to get customers for a side business with a day job is a close cousin.