1. Pick the audience before anything else
Every other decision follows from this one, and it is the decision with the largest effect on the outcome.
For a first campaign, mail your own customers — specifically the ones who have gone quiet. They are the cheapest list you will ever have and the most likely to respond, which means your first campaign is also your best chance to learn something.
Do not buy a list for a first drop. You will not know whether a poor result came from the list, the offer, or the card, and a purchased list adds a variable you cannot yet evaluate.
2. Decide what a response is
Write down, before you design anything, what you want the recipient to do. Book online. Call. Come in with the card. Scan a code.
One action. If you cannot choose, that is a sign the campaign's purpose is not settled, and the card will be vague in proportion.
This also defines how you will measure, which is why it comes this early rather than at the end.
3. Write the offer
Specific, with a reason, with a deadline. "$40 off your next visit, because it has been over a year, before March 31" has all three.
Make sure you would be happy if it worked. An offer priced so thin that a flood of responses would hurt you is a trap — think about the hundredth response, not the first.
4. Choose a size and check the budget
A 4×6 is the default and the cheapest at $1.19 all-in. A 6×9 at $1.35 gets noticed more in a stack of mail and gives you room for a longer message. A 6×11 at $1.69 is hard to ignore and appropriate when the offer justifies the cost.
Multiply by your list size and decide whether the number is one you would be comfortable losing entirely. If not, cut the list, not the card.
5. Clean the list
Duplicates, blanks, malformed postal codes, and anyone who asked not to be contacted. An hour here saves you from paying to mail cards that could never arrive, and from a distorted response rate afterward.
Read ten records at random as though you were delivering them. Printed mail is unrecoverable, and this is the last cheap moment to catch an error.
6. Design both sides
Front: one message, readable at arm's length, worth two seconds. Back: the offer, one action, who you are, and the address block left clear.
Resist adding a list of services. The instinct to use the space is strong and almost always makes the card worse.
7. Proof it on paper
Order or generate a physical proof and hold it. Colors shift between screen and print, type that looked fine at 100% zoom can be unreadable at actual size, and the address area problem is invisible on a monitor.
Scan your own QR code from the printed proof and follow it all the way to booking, on a phone. This catches the single most expensive category of mistake.
8. Send, and write down the details
Record the drop date, the exact count, the list description, and the offer. You will need all four to interpret the result, and you will not remember them in six weeks.
Pace the send if your capacity is limited. Five hundred cards arriving at once can produce more calls than you can answer, and an unanswered call is a wasted card.
9. Wait longer than feels natural
Postcard response is not immediate. Cards sit on counters and get acted on when the moment is convenient.
Give it four to six weeks before you judge anything. Then look at scans, calls, and bookings, calculate what each new customer cost, and compare that to what a customer is worth to you.
That number, not the response rate, tells you whether to do it again.