How many stamps should a loyalty card have?
Set it too low and you are paying regulars for visits they were making anyway. Set it too high and the card stops feeling winnable, which removes the only reason it works. The tool below shows what each threshold gives away and what it leaves, on your own numbers rather than a coffee shop's.
On these numbers, 15 extra visits a month covers the £39 subscription in gross profit.
Why the discount and the cost are different numbers
A six stamp card looks like a one in seven discount, or about 14%. That is what the customer sees and it is the number most owners judge the scheme by. It is not what the card costs.
The reward costs you what it costs you to make. On a coffee at 75% gross margin, a £3.70 giveaway is about 93 pence of actual cost, against the £22.20 that customer spent to earn it. The scheme is not a 14% discount. It is closer to 4% of that customer's spend, and it is only ever paid out to customers who came six times.
That gap is the single most useful thing on this page. A shelf wide discount costs you on every sale including the ones that would have happened anyway. A completed card costs you nothing on the visits that never occurred.
Choosing the number
The arithmetic narrows the range and does not pick for you. What picks for you is how often your customers come.
- A card should be finishable in about two months of ordinary visiting. A daily customer can carry a ten stamp card without noticing. A salon client on a six week cycle cannot: ten visits is more than a year, which is not a goal, it is a filing cabinet.
- Match the threshold to the visit rhythm, not the ticket. Bakeries and coffee shops tend to sit at eight to ten. Barbers at six to eight. Salons, restaurants and nail bars at five or six.
- Round numbers are not automatically better. Six and eight read as achievable. Ten is fine where visits are daily and reads as a slog where they are not.
The trade pages for coffee shops, barbers and hair salons each set out the rhythm that trade actually runs on, and the longer piece on thresholds works through the reasoning without the sliders.
The head start is not part of the threshold
Whatever number you choose, the card should not be handed over blank. Nunes and Drèze demonstrated why at a car wash in 2006: one group needed 8 stamps on an empty card, another needed 10 on a card with 2 already filled in. Both groups had exactly eight washes to do. 34% of the head start group finished, against 19% of the other, and they came back sooner in between.
So a six stamp card is better given as a seven stamp card with the first already earned. It is the same six visits. It is a materially different card. The endowed progress effect is defined in the glossary and set out in full, with what the study does and does not show, on the evidence page.
What this calculator will not tell you
It will not estimate how many extra visits a loyalty card would earn you, because nobody can know that in advance and a number produced here would look more authoritative than it deserves. Every output above is arithmetic on figures you typed in.
For the other half of the question, what those extra visits would be worth against the cost of running the scheme, the ROI calculator takes the visits as your input and does the same honest arithmetic on them.
No sign up, no email, no result withheld
The calculator runs in your browser and sends nothing anywhere. There is no form in front of the answer. If it tells you the threshold you had in mind gives away more than you thought, that is worth knowing before you print the signs rather than after.