Two covers at the same table most Tuesdays, and a first-timer on Friday who liked it, said so, and never came back. Every restaurant has both, and the difference between them is not the food. It is whether anything existed to make the second visit easier than the first.
This page is the design: what the card should count, where the goal goes, what staff say when they drop the bill, and what the first week looks like.
What to reward
Reward the visit. A spend-based program pays out fastest to the table of eight celebrating something, and that table was never going to become a regular; a visit-based program pays out to the person who comes for lunch most weeks, and that person is the entire argument for having a program at all.
Restaurant visits arrive unevenly, which is exactly why the rule has to be simple. A lunch trade and a dinner trade have different rhythms in the same room, and a card that counts visits works in both without you having to model either. The behaviour worth encouraging is the return itself, at whatever interval your room actually produces, and counting it is the only version of that rule you can explain in one sentence at a table.
Stamps or points, and where the goal goes
Stamps for a lunch trade, points where the bill genuinely swings. A neighbourhood restaurant with a set lunch and a full evening menu can see a bill move by a factor nobody wants to explain on a card, and points handle that honestly by earning against value. A café-restaurant where most tickets look alike should stay on stamps, because a stamp needs no explanation and the card reads itself.
The goal is a waiting time in disguise. Divide it by how often a regular actually eats with you and you have the weeks until the reward. A lunch regular reaches a goal of twelve inside a season; the same twelve for somebody who comes for a birthday is a promise measured in years, and it should be much shorter or the program is aimed at the wrong table.
Two things are worth saying about that number before you copy it. It is a design choice, and the only measured figure on this page is what other businesses picked. In Fideliya’s production data on 4 September 2026 there were fifteen live stamp cards, and their goals ran five on two cards, six on two, seven on one, ten on eight, twelve on one and fifteen on one, a median of ten. There were nine live points cards with a median goal of 750. The sample is small and concentrated, and that belongs in the same sentence: one restaurant is 84% of the wider holder sample those cards sit behind. Read it as what a handful of real businesses chose, never as a benchmark for yours. The same reasoning applied across fifteen trades is on fifteen punch card ideas.
The reward ladder
The distance a restaurant has to close is between visit one and visit two. Build the ladder for that gap, not for the loyal diner who was coming anyway.
- Give the first stamp at the table where they join. The card should never be handed over empty.
- Put something on the second visit. An appetiser or a coffee, small enough to give without a discussion and early enough to matter.
- Put a middle rung where the habit forms. A dessert somewhere in the middle keeps a long card from feeling long.
- Keep the named reward at the goal. A main course, or whatever the card is called after, stays at the end.
- Price the whole ladder against the covers it takes to earn. Not against the menu price of the item you are giving away.
In plan terms the card, the scanner and the wallet pass are on the free plan, which carries twenty customers at one location. Pro is the plan a restaurant reaches quickly, because it opens four team seats and a thousand customers, and a program only one manager can run is a program that stops on their day off.
The counter script
It is said once, at the bill, and it has to be shorter than the pause while the card machine thinks:
“We’ve got a card if you’d like it. It’s free, it goes on your phone, and it starts today.”
Three clauses, each removing an objection: no cost, no app, no empty card. Say it while handing over the bill presenter with the code facing them, and then stop talking. The failure mode here is a server explaining a rewards structure to a table that has already decided; the code does the explaining, and the first stamp does the persuading.
Launch week
Seven days, and the middle three are the ones restaurants skip.
| Day | What you do | Why it is that day |
|---|---|---|
| Day 1 | Decide visits or spend, the goal and the reward | Change this later and you move the finish line for everyone |
| Day 2 | Design the card and print bill-presenter codes | The bill is where the phone already is |
| Day 3 | Enrol the floor team and scan each other | Servers who have used it will offer it |
| Day 4 | Agree the one sentence at pre-service briefing | A script agreed once is a script said the same way |
| Day 5 | Run it on a full service, not a quiet one | You want the awkward version early |
| Day 6 | Put the same code in takeaway and delivery bags | It is the only way a platform order becomes yours |
| Day 7 | Read the funnel and count second visits | Enrolments are the easy half |
None of the seven days needs a full morning, and the seventh is the one that decides whether week two is worth planning. The step-by-step version, written for the counter rather than for the design, is on how to create a restaurant loyalty program.
What to watch
Three numbers do the work, and they already have names on the dashboard. The Loyalty Funnel shows how many diners joined in a period and how far along they are, which is the difference between a card people take and a card people use. Repeat Visit Rate is the share who came back more than once, and it is the only honest test of whether the program changed anything. And Not in for 8+ days is the attention list, whose own subtitle reads customers who have not visited in eight days or more.
Behind those sit four health bands. Active requires a real visit; cooling, at risk and drifting follow, and drifting is cut at thirty days, the same window the customer roster uses for its quiet filter. Check the funnel and the attention list on the same day each month. A card that fills on a busy Friday and never gets scanned again is a card that told you something, and it told you early.
One limit belongs here rather than in a footnote. A Fideliya broadcast is addressed: to everyone holding that card, to the customers seen in the last 30 days, to those not seen for over 30, or to named individuals. What the product will not do is send on its own. Nothing in Fideliya notices that somebody has gone quiet and writes to them, so every message is a person deciding to send one. And Fideliya reports that Apple and Google accepted a broadcast, never that anyone saw it.
What a card does not solve here
A loyalty card cannot bring back a table that left unhappy. If the food was wrong or the service was slow, the card is a reminder of an evening somebody would rather not repeat, and the honest thing to say is that no wallet pass has ever fixed a kitchen. The program works on diners who liked it and simply did not think about coming back.
The second limit is narrower and worth naming. A delivery-app customer becomes yours only if the code is physically in the bag; the platform will not hand you a relationship, and the card cannot reach through it. Everything else about that trade-off sits on why restaurants replace stamp cards.
Sources
Search Console recorded the query "restaurant loyalty" at 94 impressions and average position 66.7 in the three months to 1 September 2026, which is the demand this page is written against; the figure lives in the source note rather than in the prose. The goal distribution is Fideliya production data measured on 4 September 2026 across external accounts, with its sample size and concentration caveat stated beside it. Every plan limit and product behaviour named above is cited to the file that implements it in the source note on this article. The suggested goal and the ladder are design choices for this page, not measurements, and this page quotes no third-party statistic.
Fideliya is wallet-native loyalty program software for small businesses. Customers add a stamp or points card to Apple Wallet or Google Wallet in one tap, with no app to download and no account to create. Fideliya runs stamps, points, gift cards and referrals on one platform in English, French, Spanish and Arabic with full right-to-left support, and the free plan issues a real wallet pass for up to 20 customers. What the product does at the counter is described on the restaurant loyalty program, and the plans are on the pricing page.