Your loyalty program’s first hundred members will be your regulars.
They will sign up in the first week, because they are in often enough to be asked, and because they already like you. Every one of them was coming back anyway.
So the first thing your program does is apply a discount to revenue you already had. That is not an argument against loyalty schemes. It is the thing you have to beat, and most programs never get past it, because they were designed to reward the behavior rather than to change it.
Which means there is one question to answer before you design anything, and it is a measurement question rather than a creative one.
The number you need first
How many of your customers come back within ninety days, and how would that number move if you did nothing?
The first half is countable. Pull your last two quarters, group customers by the month of their first purchase, and work out what share of each group bought again inside ninety days. That is repeat rate by cohort, and it is the number that decides whether your acquisition spend is survivable in the first place, which I argued about email flows.
The second half is the hard one, and I would rather be straight about it than pretend otherwise. Measuring whether a program caused anything requires a holdout group: some customers offered it, some not, compared. Almost no small business will run that, and I am not going to tell you to.
So you are estimating, and there is a right and a wrong way to estimate.
The wrong way is comparing your members with your non-members. Members are self-selected regulars, so that comparison will always flatter the program and tells you nothing.
The right way is watching your overall repeat rate, across everybody, before and after launch. If a thousand people sign up and the business-wide repeat rate does not move, the program is a rebate scheme. That is the only honest read available to you, and it takes one spreadsheet.
Two mechanisms, and only one of them looks like a discount
Rewarding loyalty is not a mechanism. These two are.
Shorten the gap between visits. If your customers come every seven weeks and you can make it five, you have added a visit and a half a year per person without acquiring anybody. This is where a reward with an expiry does real work, and it is the only context in which I would use expiry at all.
Raise the basket at the moment of purchase. A reward unlocked by an amount rather than by a visit count changes what goes in the basket while the customer is standing there deciding.
Notice that both of those are about a decision happening now. A points balance accruing silently towards something in the future influences nothing today, which is why most points schemes change behavior less than the spreadsheet predicted.
Why the punch card beats points for a business your size
Two reasons, and the second one has real research behind it.
The first is legibility. “Buy nine, the tenth is free” is understood instantly and the customer can do the math themselves. A points system where 400 points equals an unspecified reward requires the customer to care enough to work out the exchange rate, and almost nobody does. Points work for airlines because the balance gets big enough to dream about. Yours will not.
The second is the head start, and this is worth getting right.
The underlying finding is Joseph Nunes and Xavier Drèze’s paper The Endowed Progress Effect: How Artificial Advancement Increases Effort, in the Journal of Consumer Research in 2006. In a car wash field study, some customers got a card requiring eight purchases and others got a longer card already stamped to the same effective requirement. The group with the artificial progress was more likely to finish.
You will see specific completion percentages quoted for that study all over the internet. I could not verify them against the paper itself, so I am not going to repeat them. The direction is what you need.
Two practical consequences. Print a ten-space card with two stamps already on it rather than an eight-space blank card, because the requirement is identical and the completion is not. And give a reason for the head start, because the reported detail that matters most is that the effect went away when no reason was offered. “Two free because you came in during our first month” works. Silent free stamps read as a gimmick.
The break-even, on your own margin
Here is the arithmetic, and you should do it before you print anything.
Take the cost to you of the reward, at your own cost rather than its retail price. Divide that by the gross margin on an average purchase. That is how many extra purchases the program has to generate per completed card just to break even.
A coffee shop giving away a drink that costs them 80 cents, on a two dollar margin per visit, needs less than half an extra visit per card. That works comfortably.
A service business giving away a treatment that costs them sixty dollars in time, on an eighty dollar margin, needs most of an extra appointment per card, and now the whole thing rests on whether the scheme genuinely moves behavior rather than whether it feels nice.
The pattern is blunt. Low cost of goods and high visit frequency makes loyalty schemes easy. High cost of delivery and low frequency makes them expensive, and in that second case a better-run reminder sequence usually beats a reward.
And put the administration in the sum. Somebody reconciles this, somebody handles the lost card, somebody decides whether the person who swears they had nine stamps gets the tenth. That is real time, and it is the cost most owners discover in month four.
The thing that actually kills these
Not the design. The till.
A loyalty program lives or dies on whether your staff mention it, every time, without being reminded. Most do not, and the reason is not laziness. It is that nobody gave them an easy sentence to say.
“Are you on our card?” is awkward because it invites a no. “I’ll put two on your card to start you off, is that alright?” is easy, takes four seconds, and does the sign-up and the head start in one move.
Write the exact sentence down. Put it where they stand. Then check in two weeks whether it is being said, by standing there, because the sign-up rate per transaction tells you whether you have a marketing problem or a staffing one, and those have completely different fixes.
Before lunch
Ninety minutes, and most of it is the first item.
Pull your last two quarters and calculate repeat rate by cohort within ninety days. Write the number down somewhere you will find it in six months, because it is your baseline and without it you can never tell whether any of this worked.
Then do the break-even sum on one reward idea at your own cost. If it needs more than half an extra visit per card, pick a cheaper reward.
Then write the sentence your staff will say, with the head start in it, and stick it next to the card reader.
If you’d like a second opinion on whether a loyalty scheme or a better follow-up sequence is the right call for your margins, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on lead generation cover the acquisition side. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.