How to do pre-orders before you manufacture a product

To do pre-orders, list the product at the full price with a ship-by date you can meet, charge or save each customer's card at checkout, and place the factory order once enough people have ordered to pay for it. Run the first two weeks as a demand test.

This guide is for a founder with a physical product, a factory quote and a plan to sell online. Pre-orders are the most direct way of testing demand before you build, because people commit the full price to something that doesn't exist yet. The ads and the page have their own guides.

How pre-orders work for a product you haven't made

A customer orders against a promised ship date and pays at once or saves their card for later. You collect orders until they cover the factory order, place it, and ship to everyone when the stock arrives. A waitlist signup costs the customer nothing, so only orders at the full price count as demand.

How to do pre-orders: a worked example with 300 coffee drippers

The numbers are invented and round, carried on from What is MOQ? and What is landed cost?. A founder sells a ceramic pour-over coffee dripper at $45. Her first order is 300 units, costing $3,420 landed. Each sale costs $9 in payment fees, packaging and postage. Her plan assumed winning a customer costs $15, which leaves $21 a sale and a break-even of 163 pre-orders.

The $3,420 is within the $5,000 she is prepared to lose, so she can pay the factory without pre-order money. She has also set aside $3,800 for ads across the whole pre-order run.

1. Work out the ship date from the factory's timeline

Build the date from your factory's figures, counted from the day pre-orders close, and only once you've approved the final sample. If samples are still going back and forth, add each remaining round.

In Ethan Mollick's study of Kickstarter projects, over 75% delivered later than expected. Adding slack as long as your longest stage means a production run that takes twice as long as quoted still ships on time. Add more if production runs over Chinese New Year, when factories in China shut for the holiday.

The founder opens pre-orders on Monday, March 1, 2027, runs the first two weeks as the test, and closes on Sunday, April 11. Her page says the dripper ships by September 5, 2027, 21 weeks after the close, and that date holds for the last pre-order as well as the first.

2. Choose between charging now and charging if the order goes ahead

If you charge at checkout, the money is in your account when you place the factory order, which matters if you need it for the deposit. You then hold it for months under refund duties, and each refund costs you the card fee, which Stripe keeps. Stripe charges 2.9% plus 30 cents for a payment with a US card, $1.61 on a $45 pre-order.

If you save the card, the customer pays nothing unless the order goes ahead, so trying an unknown brand is less of a risk for them, and if you stop, nobody is charged. But the charge comes weeks later, without the customer at the checkout. Stripe's guide to saving a card and charging it later says that charge can fail, for example for lack of funds or because the bank wants the customer to confirm it, and you then have to ask them to come back and pay. Stripe tells you to state in your terms when you'll charge, and to keep a record of the customer's agreement.

A saved card isn't an authorization hold. Stripe says a hold on an online card payment is usually valid for 7 days, far shorter than a pre-order run.

Charge at checkout if you need the pre-order money to pay the factory. Save the card if you can pay the factory yourself and production depends on the pre-orders, as with the founder. Either way, use the full $45: a discount would test demand at a price she won't charge later, so settle the price first.

3. Write the pre-order terms and refund policy

In the US, pre-orders fall under the FTC's Mail, Internet, or Telephone Order Merchandise Rule (16 CFR Part 435). Its clock starts when you receive an order with payment or permission to charge the card, so saved cards are covered too. It requires the following:

• You need a reasonable basis for the ship date you state. If you state none, you must have a reasonable basis to ship within 30 days.
• If you can't ship on time, you must offer customers the choice of a new date or canceling for a full refund, before the original date passes.
• After a cancellation, you have seven working days to refund a charged card. For a card you saved and never charged, you can instead confirm in writing within that time that it won't be charged.

The FTC's business guide to the rule describes an advisory opinion that allowed testing a product not yet available, on conditions that include saying plainly that it is only planned and may not ship, and refunding promptly if it doesn't. So if production depends on the pre-orders, say so on the page.

4. Set go and stop lines before the ads start

Before spending anything, write down which results mean go, which mean stop, and what you'll do in between, working from your break-even and your cost per customer.

The stop line is the ceiling from the MOQ article: $45 − $9 − $11.40 = $24.60. At that cost each pre-order only pays back its own unit, so the order breaks even only if all 300 sell.

The plan's $15 was a guess made before any ad ran, and some people won't wait months for a product, so don't use it as the go line. Use the highest cost per pre-order at which reaching the break-even fits inside your ad money. At $18.90, each sale leaves $17.10, the break-even is $3,420 ÷ $17.10 = 200 pre-orders, and winning 200 costs $3,780, inside the founder's $3,800. At $19 it would take 202 pre-orders and $3,838.

Between the lines, the order can still pay back, but getting there costs more than she set aside. She decides now that a result there means changing one thing, such as the ad or the page, and testing two more weeks, and a second such result means stopping.

The test budget has to buy enough pre-orders for the lines to mean something: her $500 buys about 26 at the go line. With only a handful, a few good or bad days would decide it. The counts below assume she spends all $500:

Count only full-price pre-orders, charged or with a saved card, and divide everything spent to win customers in the test, boosted posts and creator fees included, by them. That is your customer acquisition cost. Draper's Design Your Demand Test workshop works out a test budget and the result to aim for from your own quote and costs.

5. Run a two-week test with regular checks

Keep the ads, the page and the price the same for the whole two weeks, so the result describes one offer. Every two or three days, check that the ads are spending and that an order placed on your phone from the ad link goes through as the page says, then cancel it. Leave the numbers alone until the end.

Count pre-orders in your store, not the ad platform, and give each source its own tagged link: ads, your own posts, and creators you sent the product to.

6. Read the result against your lines

The founder spent $500 and took 31 pre-orders: 27 through the ads and 4 from friends. That's $16.13 each, under the $18.90 go line. Most remaining orders will come from ads, so she reworks the break-even with the ads alone: $500 ÷ 27 = $18.52, which leaves $17.48 a sale and a break-even of 196. She needs 165 more, about $3,056 in ads, or $3,556 with the test, inside her $3,800.

• Go: keep pre-orders open until the close. With saved cards, charge them all the next day, email anyone whose charge fails with a link to pay another way, and place the order if the payments that go through reach the break-even. If they fall short, decide in advance whether to pay the gap yourself or cancel. Don't extend the close, because the ship-by date counts from it.
• Change: compare clicks with orders to choose what to change. Plenty of clicks and few orders points at the page, the price or the wait. Few clicks points at the ads.
• Stop: cancel every order, tell customers why, and refund anything charged within seven working days. Talk to some of them before you pay for another test.

When to skip pre-orders

• If you can buy a small batch outright, as with a white-label product or print-on-demand, sell from stock instead, so buyers don't have to wait.
• If the product still needs development work, you can't give a ship date with a reasonable basis. Get to an approved sample first, and test the offer with conversations and a priced page in the meantime.
• If you sell outside the US, the rules differ. In the UK, a customer can cancel a pre-order up to delivery and for 14 days after it arrives.

Related questions

Whether the numbers can work at all, before any pre-orders, is what a feasibility study checks. Order size and import costs are in What is MOQ? and What is landed cost?.