Restaurant Loyalty Programs Guests Actually Use
Most loyalty programs fail for the same reason: they ask the guest to do the remembering.
Why punch cards and bolt-on apps get ignored
Almost every loyalty programme fails for one reason: it asks the guest to do the remembering. Carry this card. Download this app. Give the code at the till. Every one of those is a small tax on a visit, and guests pay it for about three weeks.
The card is in another coat. The app was deleted after two visits because it wanted notifications. The code was on a receipt somebody threw away. None of that is the guest being difficult — it is a system that put its own bookkeeping on the wrong side of the counter.
What recognition actually requires
A programme that works has to do the counting itself, from something the guest was going to do anyway. That means it lives where the orders already are, rather than beside them.
It counts without being asked
If a visit only counts when somebody remembers to scan something, most visits will not count, and the programme is measuring staff diligence rather than loyalty.
The reward arrives by itself
Earned should mean applied. If a guest has to claim it, a good proportion never will, and the ones who do had to ask — which is exactly the friction you were removing.
They are told when it happens
Rewards are usually earned at the end of a meal, by which point the phone is away. Without a message the guest finds out on their next visit, if ever, and the moment of goodwill is lost.
It survives a change of mind
If you rename a reward mid-season, the guest who was promised the old one should still be shown the old wording. A receipt that contradicts what they were told is worse than no programme.
Visits or spend — they reward different behaviour
The choice looks administrative and is strategic. Counting visits rewards frequency, which suits a café or anywhere the same person could plausibly come twice a week. Counting spend rewards size, which suits a restaurant where somebody comes monthly but brings four people.
Pick the one that matches the behaviour you actually want more of. A visit-based card in a fine dining room quietly asks guests to come more often than the format supports, and then nobody reaches the threshold and the programme dies of irrelevance.
Automatic versus manual perks
Not everything should be automatic, and pretending otherwise is how venues end up giving away things they meant as gestures. There are two categories and they want different machinery.
Automatic
The stamp card. Defined threshold, defined reward, applied without anyone deciding. Its value is that it is predictable — the guest can count on it, which is the whole point.
Discretionary
The drink for the couple whose starter was late, the dessert on a birthday. These are judgement calls and should stay judgement calls, but they should still be recorded, because a venue that gives away four hundred euro a month without knowing it has a different problem from the one it thinks.
Segments beat blasts
Once visits are being counted automatically, the useful thing is not the card. It is that you now know how often each guest comes, and that divides a guest list into groups that want completely different things said to them.
New
One visit. The only message worth sending is one that gets them to a second, and it should arrive while they still remember the first.
Regular
The people the business runs on. Mostly these should be left alone and thanked occasionally. The main risk with regulars is over-contacting them.
Occasional
Come now and then. This is where a well-timed reason to visit moves the most revenue, because the intent already exists and only the prompt is missing.
Lapsed
Used to come, have not for a while. The highest-value list in the building and the one almost nobody has, because it requires having recorded the visits in the first place.
A campaign sent to all four of those groups says something wrong to at least three of them. The point of counting visits was never the free coffee — it was being able to tell these apart.
Consent, and why frequency is the whole game
The moment a loyalty programme touches email it stops being loyalty and starts being marketing, and the rules change. Under GDPR, consent has to be given rather than assumed, and it has to be given for something specific — a guest who agreed to hear about an event did not agree to receive offers.
The practical constraint is stricter than the legal one. The fastest way to lose a regular is to contact them too often, and every venue underestimates its own frequency because each message is sent by a different person for a different reason. A birthday from one system, an event invitation from another, an offer from a third, and the guest received three emails in nine days from a place they visit monthly.
A hard cap across everything — a specific number of contacts per guest per month, enforced by the system rather than by good intentions — is the single most useful rule you can put on this. Marketing covers how that is handled here, and Guest Loyalty covers the card itself.
Common questions
Should the reward be a discount or a free item?
A free item usually reads as more generous than a discount of the same value, and it costs you the ingredient rather than the margin. It also lets you point people at something you want them to try.
How many visits should a card be?
Short enough that a new guest can imagine reaching it. A ten-visit card in a place people visit monthly is a card nobody finishes.
Do guests need an account?
They need to be identifiable somehow. The less that identification asks of them at the table, the more of them will be counted.
Can we email everyone about a promotion?
Only the ones who agreed to that kind of message, and only within whatever frequency cap you have set. Both of those are worth having the system enforce rather than a person.