Guide · last reviewed 31 August 2026
How to make a digital punch card without an app.
One location, one reward, one afternoon. Here is exactly how — and where it stops being an afternoon.
We sell this as a service, so read the last section with that in mind. Everything before it is the real method, not a teaser — a single-location shop with a simple reward can follow it and get a working card. What the method does not survive is scale, and that part is at the bottom.
Step one: decide the reward before you touch any software
This is the step people skip and the one that decides whether the whole thing works. Four questions, in order:
- What earns a stamp?
- One visit, one stamp for a café — frequency is what you are buying. Points per dollar only makes sense where ticket sizes vary a lot, like a restaurant where lunch is $19 and dinner for four is $95.
- How many stamps?
- Ten in a café. Five in a salon or barbershop, because at a five-week interval ten visits is more than a year away and nobody chases that. Match the number to how often people actually come.
- What does the reward cost you?
- A $4.75 drink costs about $1.10 in product. Give one away every ten visits and you are handing back roughly 2.5% of that customer's spend — cheaper than most discounts you already run.
- What does it have to produce?
- Write the number down before you start. If your platform costs $40 a month and you keep about $3.40 of a $4.75 sale, it needs twelve extra visits a month to break even. Decide now whether that is plausible.
Step two: get three things together
Your logo, transparent
PNG or SVG with a transparent background. If all you have is a JPG off a sign painter, that is a problem worth fixing once — you will use it everywhere.
Three photographs
Of your own product, shot on a phone, near a window, in daylight, landscape, from slightly above. Not the room, not the staff, not the storefront. This is the single biggest difference between a card that looks like yours and one that looks like a template.
Your details
Business name exactly as you write it, brand colours, full address of every location, and opening hours. Ten minutes of gathering that saves an hour later.
Step three: pick a platform and build the pass
Self-serve platforms in this category run roughly $25–$80 CAD a month for one location. They issue the passes under their own Apple certificate, so you do not need an Apple Developer account unless you want the passes issued under your own.
| Get this right | Why |
|---|---|
| A dark card body | It survives both light and dark lock screens, and it looks right sitting next to a bank card in the same wallet. |
| A photo behind the balance | A flat colour reads as a template. Your own product reads as your shop. Keep the left third of the image dark, because the balance renders over it in white. |
| Stamps, not points | A row of circles that fill is legible at arm's length. “237 points” is not. |
| The reward named in words | “Free 10th coffee”, not “10 points to reward”. People finish what they can see themselves finishing. |
| Your locations, with a geofence | Apple allows up to ten. This is what lets the card surface on the lock screen when someone is near your door. |
| A barcode with readable alt text | When a scanner fails — and it will — your staff need a number they can key in. |
If you want Google Wallet passes under your own brand you will need a free Google Wallet issuer account and publishing access, which is a form and a short wait rather than any real work.
Step four: the QR code, and where to put it
Generate the enrolment link and print it. Give each placement its own code — a different one on the counter card, the table tent, the window and the takeout bag. It costs nothing and after a month you will know which surface is actually recruiting customers and which is decoration.
Laminate the counter card, or put it in a stand. A QR taped to a till on fading receipt paper is the single most common reason these programs quietly die.
Step five: the part that actually decides it
Train your staff at a shift change, not by email. Everyone enrols their own phone and scans it once. Muscle memory beats instructions, and a barista who has the card in their own wallet will mention it without being told to.
Check the scan count on day three and day seven. If enrolments are under ten in the first three days, something is wrong at the counter and it will not fix itself. Falling scans in week one means the program is finished by week six.
Put something small behind it. Thirty dollars for whoever signs up the most customers in month one is the difference between working and not.
- Time to build
- An afternoon — per location
- Platform cost
- ~$25–$80 CAD/month
- Print cost
- Under $30
- Apple Developer
- Usually not needed
- Google issuer account
- Free
- Hardware
- None
Questions people ask
Can I really do this myself?
Yes. Nothing on this page requires a developer, and the whole thing is doable in an afternoon if you already have a logo and a photo of your product. The honest catch is not the setup — it is month two, when the campaign needs writing and the numbers need reading and you are three deep at the till.
What does it cost to do it myself?
A self-serve wallet platform is roughly $25–$80 CAD a month for one location. If you want to issue Apple Wallet passes under your own certificate rather than the platform's, add the Apple Developer Program at $99 USD a year — though most self-serve platforms issue under theirs, so you can skip it. Google Wallet issuer accounts are free. Printing a counter card and a table tent is under $30. Call it $40–$110 a month all in.
How long does it take?
An afternoon to build the pass and a week of thinking about the reward beforehand — the reward is the part worth the time. Add a couple of days if you need to get Google Wallet publishing access, which is a form and a wait rather than any work.
Do my customers need an app?
No, and that is the whole point. Apple Wallet and Google Wallet are already on the phone. Your customer scans a QR code, taps Add, and the card is in their wallet — no download, no account, no password.
What is the most common mistake?
Setting the reward too far away. Ten stamps works in a café because people come often. In a salon at a six-week interval, ten visits is over a year and nobody chases something that far off. The second most common mistake is putting the QR code somewhere nobody looks.
Why would I pay someone instead?
Because the software was never the hard part, and because the afternoon above only stays an afternoon at one location. Designing a reward that pays for itself, getting staff to actually scan, writing an offer every month, pushing it to the card, and reading the numbers back is four or five hours a month, every month, per brand. Add a second site, a second banner, franchisees or a POS to sit alongside and it stops resembling the guide above entirely.
Where this stops being an afternoon
The method above holds for one shop with one reward. Every one of these breaks it, and they break it in ways that do not show up until month two.
| What changes | What it actually costs you |
|---|---|
| A second and third location | Each site needs its own geofence perimeter, its own analytics and its own campaign timing — the quiet Tuesday in Mount Pearl is not the quiet Tuesday downtown. Self-serve platforms price per location and most cap the geofences well below what a chain needs. Apple allows ten locations on a pass, full stop. |
| More than one brand or banner | A separate Apple Pass Type Identifier and signing certificate per brand, and a separate Google Wallet class per brand. Certificates expire on their own schedule and a lapsed one silently stops passes updating. |
| Franchisees rather than staff | Somebody has to decide who controls the reward, who pays for it, and what happens when one franchisee wants a different offer. That is a governance problem, not a software problem, and no platform solves it for you. |
| A POS or CRM you already run | Integration is where the afternoon becomes a project. Matching a wallet member to an existing customer record is real work with real edge cases. |
| Staff turnover across sites | Retraining is the recurring cost nobody budgets. One shop is a shift change. Twelve shops is a programme. |
| Somebody has to own the numbers | Four or five hours a month, per brand, forever — designing the offer, sending it, reading the result honestly enough to change something. |
None of that is exotic and none of it is secret. It is simply a different job from the one described above, and it is the job we are actually in.
So: who should do it themselves?
Do it yourself if you run one location, you already have a reward that works on paper, someone on your team genuinely enjoys this kind of thing, and you can protect a few hours a month for it indefinitely. You will get a good card and you will keep your money. That is a real outcome and we would rather you had it than bought something you cancel in month three.
Talk to us if you have more than one site, or more than one brand, or franchisees, or a POS you want this to sit alongside — or if the honest answer to “who owns this in month six” is nobody. Those are the situations where the afternoon above turns into a standing commitment, and where doing it badly costs more than not doing it.
Keporia is $99 CAD once and $149 a month for one location, plus $49 a month for each additional site — the pass, the geofences per location, the printed pack, the staff training, a campaign every month and a one-page report. Multi-site and multi-brand operators, talk to us about scope rather than the list price; the work is different and so is the arrangement.
- One location
- Do it yourself, or $99 + $149/mo
- Two to ten
- Per-location geofences and reporting
- Multi-brand or franchise
- Scoped engagement — talk to us
- The recurring cost
- 4–5 hours a month, per brand
More than one location? Talk to us first.
We’ll build you a card free either way — look at it on your own phone, then decide which route makes sense for the size you actually are.
Book a 15-minute demo