Work the numbers first
A stamp card is a discount with a delay. The only question that matters at setup is what that discount costs you per completed card, and whether the visits it buys are worth it. Fill in your own figures — the ones below are placeholders chosen to be legible, not benchmarks.
| Input | Illustrative value | What it drives |
|---|---|---|
| Average qualifying purchase | $4.00 | Revenue per stamp issued |
| Stamps to a reward | 8 | Revenue per completed card ($32.00) |
| Cost of the reward to you | $1.50 | Not the menu price — the cost of goods |
| Effective discount | 4.7% | $1.50 against $32.00 of qualifying spend |
Two things fall out of that table. First, the reward's cost to you is the cost of goods, not the menu price — a free drink that sells for $5.00 and costs $1.50 to make is a $1.50 decision. Second, the threshold is the discount dial: eight stamps is a 4.7% programme on these assumptions, six is 6.3%, ten is 3.8%. Pick the threshold from the discount you are willing to fund, then check that it is reachable inside a month for a regular. A card a customer cannot finish is a card they abandon.
Assumption, not a result: nothing here predicts how many customers will join, return, or complete. Those depend on your shop, and you should measure them rather than accept an estimate.
What goes on the card
Four decisions, and they should fit on an index card before they go into software.
- The qualifying purchase. "Any drink" is easy to say and easy to honour. "Any drink over $4" creates a conversation at the till every time someone orders a small coffee.
- The threshold. From the discount you chose above, not from a round number.
- The reward. One reward, named exactly. "A free drink of the same size" avoids the argument that "a free drink" invites.
- The remainder rule. If someone buys three coffees at once, they earn three stamps and the overflow carries into the next card. Say so on the card, because the alternative is a customer who feels shorted.
Rewardfinity supports a required stamp count, an optional number of stamps per purchase, and an optional minimum purchase amount. The wording customers read, and the sentence your staff say, are still yours to write — and they matter more than the settings.
Paper, an app, or a Wallet card
The format decides your join rate, and join rate decides everything downstream. This is the trade-off, stated plainly.
| Paper punch card | A customer app | Wallet member card | |
|---|---|---|---|
| What the customer does to join | Takes a card | Downloads, installs, creates an account | Scans a QR code at the counter |
| Survives a lost wallet | No | Yes | Yes |
| You know who your regulars are | No | Yes | Yes |
| Cost per card | Printing, reprinting, forgery | Development or subscription | Subscription |
| Friction at the till | Lowest | Highest | Low |
Paper wins on friction and loses everything else — you cannot tell a regular from a stranger, and you cannot reach anybody. A dedicated app has the opposite problem: the install is asked for at the exact moment a customer wants their coffee and to leave. A card that lives in the phone's own wallet is the compromise that keeps the counter fast, which is why it has become the default recommendation for cafés.
Rewardfinity issues a hosted member card. Google Wallet is a configured path; Apple Wallet public launch remains dark pending its physical canary. Confirm the exact state before you print counter signage that promises either wallet.
The staff side
A loyalty programme is executed by whoever is on shift. If signing a customer up is invisible work, it stops happening in week three — and no amount of customer-side design fixes that.
This is the part of the category that is genuinely thin.Plenty of tools suggest running a staff signup competition; far fewer ship the per-person attribution you would need to run one, which is why most of these end up tracked on a whiteboard. Rewardfinity tracks counter actions per staff member and can turn them into a staff-facing spin wheel with an owner-set allowance, so recognition is a mechanic rather than a memo.
- Give staff one sentence to say, and put it where they can see it.
- Make the ask part of the transaction, not an addition to it.
- Attribute signups to the person who got them, so recognition is possible at all.
- Review it after two weeks with the team, not at them.
Launch checklist
- Write the qualifying rule in one sentence and read it aloud.
- Set the threshold from the discount you costed, not from a round number.
- Name the reward precisely, including size or variant.
- Decide and publish the remainder rule.
- Put the QR code where a customer already looks while paying.
- Run one real transaction end to end, including a refund.
- Brief every shift, including the person who works Sundays.
- Choose the date you will review it, and what result would make you change the threshold.
What café owners ask before launching
How many stamps should a coffee shop loyalty card have?
Choose the number from the discount you are willing to fund. On a $4.00 average purchase and a reward that costs you $1.50, eight stamps is roughly a 4.7% discount, six is 6.3%, and ten is 3.8%. Then sanity-check that a regular can finish the card within about a month — an unreachable card is abandoned.
Are stamp cards better than points for a café?
For most cafés, yes, at the start. Stamps match one repeatable purchase and a fixed reward, and they need no mental arithmetic. Points earn their extra complexity when baskets vary — retail bags of beans, food, merchandise — or when you want several rewards at different costs.
Do customers need to download an app?
Not necessarily. A card that lives in the phone's built-in wallet is added by scanning a QR code at the counter, with no separate install. That matters in a café specifically, because the install request lands exactly when the customer wants to take their coffee and go.
What does a coffee shop loyalty programme cost to run?
Two costs, and they are different in kind. The software is a subscription; Rewardfinity has a free plan for up to 100 members and prices paid plans to local purchasing power rather than one global sticker. The larger cost is the reward itself — the cost of goods on every completed card, which is the number to model before you choose a threshold.
Can I move my existing paper punch cards over?
Yes, and the honest way is to publish the conversion before you change anyone's progress. Decide what a partially filled paper card is worth, honour it visibly for a defined period, and record conversions as auditable adjustments rather than silently resetting people. Customers forgive a rule; they do not forgive losing progress.
Will a loyalty programme actually bring people back?
That depends on your shop, and anyone who quotes you a number without seeing it is guessing. What a programme reliably does is tell you who your regulars are, which is the thing you cannot do with paper. Decide in advance which measurement would convince you it is working, and set a review date.
Cost the reward, then set the threshold.
Start with the free plan and a single stamp card before adding anything else.