Most articles about loyalty programs describe what large chains do, then leave you to work out what any of it means for a shop with one till and four staff. This one goes the other way. Below are the structures that small businesses actually run, what each one costs, and the type of business each suits.
The six structures worth knowing
Almost every loyalty program in the wild is a variation on one of six ideas.
1. The stamp card
Buy nine coffees, get the tenth free. The customer collects one stamp per qualifying purchase and redeems when the card fills up.
This is the oldest format and still the most common, because it survives contact with a busy counter. Staff do not calculate anything. Customers do not need to understand a conversion rate. A new member understands the deal in about four seconds.
It suits businesses where most transactions are close in value: coffee shops, sandwich bars, bakeries, car washes, nail bars, takeaways. It suits them badly if your average order swings from £4 to £60, because a stamp for a flat white costs you far less than a stamp for a full colour treatment.
Cost to you: one free item per completed card. If your margin on a coffee is roughly 70 percent, giving one away every ten visits costs you about 3 percent of that customer's spend. That is cheap for a reason to come back.
2. The points program
The customer earns points per pound spent and exchanges them for rewards. One point per pound, one hundred points for a £5 credit, that sort of thing.
Points handle uneven basket sizes properly. A customer who spends £80 earns eight times what a £10 customer earns, which feels fair to both. The trade is complexity. Someone has to decide what a point is worth, and customers have to do arithmetic to know where they stand.
It suits salons, garages, pet groomers, independent retailers, anywhere the bill varies a lot.
The failure mode is setting the exchange rate so conservatively that nobody ever reaches a reward. If a customer needs to spend £500 to earn a £5 voucher, you have built a program that quietly does nothing.
3. Tiers
Bronze, silver, gold. Spend or visit more, unlock better perks permanently.
Tiers work on repeat behaviour rather than one purchase. The pull is status as much as discount: people do not want to fall back to bronze. This is why airlines and hotels lean on them so heavily.
For a small business tiers are usually overkill. They need enough customers for the tiers to feel meaningful, enough perks to differentiate each level, and enough admin to track it. A gym, a wine club or a members' space can carry it. A bakery cannot.
4. Paid membership
The customer pays a fixed amount per month and gets something unlimited or heavily discounted in return. Pret's coffee subscription is the example most people know.
The economics are different from every other structure here. You get predictable revenue up front and the customer gets a reason to walk past your competitor. It only works if you have enough regulars to convert and enough margin to absorb heavy users, and you need to be comfortable that your best customers will pay less per visit than they do today.
5. Cashback or store credit
A percentage of each purchase comes back as credit spendable in your shop.
Cashback is easy to explain and the credit stays with you rather than becoming cash in someone's pocket. It reads as more generous than it is, because 5 percent back sounds larger than one free item in twenty. The drawback is that it rarely creates a moment of delight. Nobody screenshots a £1.40 credit balance.
6. Referral rewards
Existing customers get something when they bring someone new who joins or spends.
This is the only structure on the list that grows your customer base rather than deepening the ones you have. It works best bolted onto a program that already exists rather than as your only mechanic. If you want the detail, we wrote about how a referral offer should be structured separately.
What the big programs get right
Three things worth copying from chains, and one worth ignoring.
Starbucks Rewards is worth studying because the progress bar is always visible. Customers open the app and see exactly how close they are. The mechanic is not clever. The visibility is.
Tesco Clubcard is worth studying for member pricing. The reward is not a voucher you save up for, it is a lower price at the shelf, right now. Instant beats deferred for a weekly shop.
Nando's card is worth studying for how small the reward ladder is at the bottom. You reach the first reward quickly, which teaches you the card is real before you have invested much.
What to ignore: the point valuations. Chains run loyalty at a scale where a fraction of a percent matters and where breakage, the rewards nobody claims, is part of the model. If you have four hundred customers rather than four million, you want people redeeming. A redeemed reward is a visit. An unredeemed one is a customer who forgot about you.
Examples by business type
Coffee shop. Nine stamps, tenth drink free. Add a bonus stamp on the quietest weekday to move traffic. Most cafes we see run a coffee shop loyalty program on exactly this shape and change nothing for years.
Hair salon. Points on spend rather than stamps on visits, because a cut and a full colour should not earn the same. Add a fixed perk at a spend threshold, such as a free treatment.
Bakery. Stamps, but on a category rather than the whole shop. Ten coffees earns a coffee. Otherwise a customer buying a £1 roll earns the same as one buying a £25 celebration cake.
Barber. Stamps at six rather than ten. Haircuts are roughly monthly, so a ten-stamp card takes most of a year to fill, which is too far away to feel real.
Takeaway. Spend-based, with the reward set slightly above your average order so redeeming pulls the basket up rather than down.
Nail bar or beauty studio. Stamps per appointment, plus a birthday reward. Appointment businesses have a booking system that already knows the date, so it costs you nothing to use it.
How to pick one
Answer two questions honestly.
Do your customers spend roughly the same amount each visit? If yes, stamps. If no, points.
Do you want more visits from existing customers, or more customers? Stamps and points do the first. Referral does the second. Most independents need the first and reach for the second too early.
Then set the reward so a regular reaches it within about six to eight weeks. That number matters more than the structure. A well-designed points program with a reward twelve months away performs worse than a crude stamp card someone can finish by the end of the month.
Where most programs go wrong
They are not mentioned. A loyalty program that lives on a poster by the door and nowhere else will sign up the customers who were already loyal. The single biggest lever is a member of staff saying one sentence at the till, and that is a training problem rather than a design problem. We covered it in training staff to bring it up naturally.
They are too generous at the start and get clawed back later. Reducing a reward feels like a takeaway to customers even when the new deal is reasonable. Start slightly tighter than you think and loosen it if uptake is slow.
They have no expiry and no clean-up. Cards from three years ago that nobody will ever finish sit in your data and distort every number you look at.
If you want to go from picking a structure to running one, the next step is how to create a loyalty program, which covers the reward maths, the launch, and the first month.