In brief
Start with the promise, not the software. Name the repeat visit you can recognize, choose a reward you can honor on a busy day, set a target that matches how often those customers already come in, decide who validates a visit, then test the whole journey yourself before you put a poster on the counter.
Start with the promise, not the platform
The most common way a loyalty program fails is being designed inside a settings screen. Software asks you for a target and a reward as though those are configuration; they are the commercial decision, and they deserve an afternoon before anything is installed.
Seven decisions, in order.
1. Name the qualifying visit
Write the sentence a staff member will use to decide. Not “a purchase” — something they can judge in a second while doing three other things.
- “Any hot drink.”
- “Any haircut or beard trim.”
- “Any order over £5.”
- “Any appointment they turn up to.”
The test: could a new starter on their first shift apply this rule without asking? If not, it will be applied inconsistently, and inconsistency is what customers complain about.
2. Choose a reward customers can picture
A free item beats a percentage. “A free coffee” needs no arithmetic; “15% off your next visit” needs a calculation and a purchase to apply it to.
Three constraints:
- It has to cost you less than the margin on the visits that earn it. Nine coffees of margin comfortably covers a tenth coffee. Nine coffees of margin does not cover a hamper.
- You have to be able to hand it over at 8:40am on a Friday. A reward requiring preparation will be resented by whoever is on shift.
- It has to be worth mentioning. If a customer would not tell a friend about it, it will not change a decision.
3. Set a target that matches real frequency
Look at how often your typical customer actually comes in, then pick a target they will reach in roughly six to eight weeks.
| Typical visit frequency | A sensible target | Time to reward |
|---|---|---|
| Most days | 10 | About two weeks |
| Twice a week | 8–10 | Four to five weeks |
| Weekly | 6–8 | Six to eight weeks |
| Every two weeks | 5–6 | Ten to twelve weeks |
| Every six weeks | 4–5 | Six months |
A target too far away produces cards abandoned at stamp three. That is worse than no program, because the customer has now had an experience of not getting there.
4. Write down the exceptions
Decide these before launch, and put them on the card:
- A visit limit. One stamp per day is the usual answer, and it prevents a customer collecting three in one transaction.
- A cooldown, if a limit is not enough.
- An expiry, if you need one. Many small programs do not, and not having one is a real trust advantage.
- What happens at other outlets, if you have more than one.
Rules a customer meets after they have earned something are the ones that cause arguments. Rules printed on the card before they start are just rules.
5. Decide who validates
The customer must not be able to add their own credit. Once they can, the count records intent rather than visits, and the reward is being given away on the honor system.
Decide which roles can issue a stamp, which can confirm a redemption, and whether that differs by outlet. Then check that the tool you are considering can actually express that distinction.
6. Prepare the counter, not the campaign
The single biggest determinant of enrolment is whether your team asks. Give them one sentence, agreed in advance:
“Do you want me to start you a stamp card? Ten coffees and the next one’s on us.”
Print the join poster and put it where a customer’s hands already are — beside the card reader, not on the door behind them. Brief every shift, including weekend staff, and tell them the reward so they can answer the follow-up question.
7. Test the whole journey before you promote it
Do it yourself, on your own phone, at your own counter:
- Scan the poster and join as a customer would.
- Have a staff member stamp a real visit.
- Reach the target, honestly, over however many visits it takes — or use a test card.
- Request the reward and have staff confirm it.
- Check what the dashboard recorded.
Anything awkward in that sequence will be ten times more awkward during a rush.
The first month
Watch two numbers. How many cards are being started tells you whether the counter conversation is happening. How far cards get before stalling tells you whether the target is right. Fix the target if most cards die early — do not add features.
The NeoLoyal approach
NeoLoyal takes the seven decisions above as its setup: choose the target from 2 to 50 stamps, set the reward, add limits, cooldowns, or expiry if your program needs them, decide which staff can validate, and print the QR join poster.
Customers scan the poster and the card opens in their phone browser. Staff issue each stamp. The dashboard shows enrolments, stamps, and redemptions, which is where the two numbers above come from.