Do loyalty programs work for small businesses? What the data shows
Do loyalty programs work for small businesses? Rewardly data on 40,000+ cards, one outside study, when they fail and an 8 week test for your shop.

Often yes, but a card does not work by itself, and nobody can promise you the result in advance. In Rewardly's data, 42.5% of new loyalty members came back within 30 days. That describes how members behave, not what the card caused. One published study of a convenience store chain found that a program changed little for heavy buyers and lifted light and moderate ones. Below you will find the numbers with their caveats, five situations where programs fail and a test you can run in your own café, bakery or salon.
The short answer
- Members do come back. 56.0% of new members were back within 90 days in Rewardly's data. We cannot say how many would have come back without a card.
- A few members carry the visits. The top 20% of members make about two thirds of member visits, so a small core of regulars carries most of your traffic.
- The effect is not the same for everyone. In the outside study below, regulars did not change their buying, and occasional customers did.
- The only proof that counts is your own. A card has a price, so run a test with a pass mark you set before you start.
What Rewardly's data shows
We have looked at member behaviour in four studies. Each has a sample, a window and one definition of a visit: a card scanned on a given day.
- How soon members return (the full study): more than 40,000 digital loyalty cards across 50+ businesses, first used between May 2025 and mid July 2026. 21.9% of new members came back within a week, 42.5% within 30 days and 56.0% within 90 days. The typical returning member was back after 15 days. By trade the typical wait was 11 days in bakeries and sweet shops, 15 in coffee shops, 18 in restaurants and bistros and 27 in beauty, nails and fitness.
- Who stays (the 12 month study): 9,528 cards first used between May and September 2025, about seven in ten at coffee and matcha bars. 54.3% of members who came back in their first full month were still visiting 10 to 12 months later, against 20.1% of those who did not. That is 2.7 times as many.
- Who makes the visits (the top 20% study): 41,187 members of 60 businesses over twelve months to September 2026. In the typical business the top 20% of members made 65% of member visits, and 39% of members had made one visit.
- Who joins (the QR code study): of 48,848 people who opened a join page, 66% became members.
What these numbers cannot tell you matters as much. They describe members, not the card's effect. Some of the members who return quickly were regulars before they joined, and the 12 month study says so itself. Some visits are never scanned, so real return rates are probably a little higher. And the data has no before and after comparison, so it cannot show how many of those visits the card added. Use the figures to see what a normal return looks like in your trade, not as a result to expect.
One more number, with its limits. Rewardly clients report 30% more repeat visits. That is a company figure from our clients, not an independent study, and the data above cannot confirm or refute it.
What an outside study found
We went to one primary source. Yuping Liu published a study on the long term impact of loyalty programs on purchase behaviour in the Journal of Marketing in 2007 (abstract, read on 5 November 2026). It follows customers of a convenience store franchise over time after a loyalty program started. Customers who were heavy buyers at the start were the most likely to claim rewards, but their buying did not change. Customers who bought little or a moderate amount at the start gradually bought more and became more loyal.
Take care with what it proves. It covers one retailer, and it is a 2007 paper, so the data predates wallet cards. We read the abstract, not the whole paper. But it gives a useful rule: a reward aimed only at your best customers may pay them for visits they would make anyway. If your regulars are about two thirds of the visits, as in the data above, that is a cost you should count. The room to grow is in the occasional customer, so the second visit is the one to design for.
Say a café tries a card: the arithmetic
Illustrative numbers, so replace them with yours. Say a café has an average ticket of €6 and keeps 70% of it after ingredients, which is €4.20 a visit. It pays €29.90 a month for Basic (prices excluding VAT). The question is how many visits a month the card has to add, which means visits that would not have happened without it.
- 5 extra visits: 5 × €4.20 = €21.00. Minus the plan, a loss of €8.90.
- 12 extra visits: 12 × €4.20 = €50.40. Minus the plan, a profit of €20.50.
- 30 extra visits: 30 × €4.20 = €126.00. Minus the plan, a profit of €96.10.
Break even is €29.90 ÷ €4.20 = 7.1, so 8 extra visits a month. Those are before the cost of rewards. A free €3 coffee that costs €0.60 to make, given for every 10 coffees, takes about 2% of what members spend. If the card only brings back customers who would have come anyway, the reward is a pure cost. The ROI calculator does the same sum with your own ticket and also subtracts the rewards.
When loyalty programs do not work
- Customers buy once a year or less. Furniture or a one off service gives a stamp card nothing to count. Gift cards or a referral ask fit better.
- The reward is too far away. A customer who comes every two weeks needs 20 weeks to fill 10 stamps and will forget the card first. How many stamps on a loyalty card shows the sum to run.
- Nobody offers it. If one person at the till mentions the card, your results follow the shift rota. Every customer, every payment, the same one line.
- The reward pays people for visits they would make anyway. See the study above. Give regulars recognition, not a bigger discount.
- There are too few customers to see anything. Say a shop serves 15 customers a day, about 450 a month. A lift of 8 visits is under 2% of that and disappears in normal day to day swings. The test below runs for about eight weeks for that reason.
A digital card also does not fix a dull reward, a slow till or a product people do not need to buy twice.
How to test it in your own business
- Write the baseline first. The number of receipts a week over the last four weeks, your average ticket, your margin and the plan price. Work out your break even in extra visits a month, and divide by four for a weekly figure. In the café above, Advanced (€49.90) breaks even at 12 extra visits, about 3 a week.
- Launch one card with one reward and one line. How to launch a loyalty card without an app has the rollout. Design the card in the free loyalty card maker first.
- Check the first two weeks. New members a day, and the share of customers who were offered the card. If the offer rate is low, the test has not started yet.
- Read day 30. The second visit rate: members who joined and came back at least once, divided by members who joined. Compare it with the trade figures above for context, never as a target.
- Compare at week 8. The average weekly receipt count against your baseline, same weekdays. Note holidays, weather and price changes. If it is up by at least your weekly break even, and rewards cost less than the margin gained, keep the card. If not, change the reward or stop.
A before and after comparison cannot separate the card from the season. With two locations you can do better: run the card in one for four weeks and compare it with the other. Plan for it up front, because the 14 day trial is too short to read a 30 day number. Every plan starts with a trial (a payment card is required and nothing is charged during it), and after that a monthly plan stops when you cancel, while a yearly plan is not refunded. For a test, stay on monthly. 5 loyalty program numbers to check every month gives the monthly routine afterwards.
Frequently asked questions
Are loyalty programs worth it for a small business?
They can be, when the reward is cheap to give and the card brings extra visits. Check it with a break even: at a €6 ticket and 70% margin, a €29.90 plan needs 8 extra visits a month. Rewardly's data shows how members behave, not what the card adds, so test it in your own shop.
Do loyalty programs work for restaurants and cafés?
Members of both come back, but at different speeds. In Rewardly's data the typical wait for a second visit was 15 days in coffee shops and 18 in restaurants and bistros. The guide to getting repeat customers covers what to do in between.
What is the simplest loyalty program for a new shop?
One stamp card with one reward that a regular can reach within a month or two, and one sentence at the till. You can set up a card in about 2 minutes and customers add it to Apple Wallet or Google Wallet in about 10 seconds, with no app to download.
How long before I see results?
You see joins in the first week and second visits from about day 30. A fair read of sales needs about eight weeks, because one month is easily swallowed by seasons and holidays.
Does a loyalty program just give away discounts?
If it pays people for visits they would make anyway, yes. That is why the reward should be small compared with the ticket, and why you count only the extra visits in your break even.
Test it on your own customers. Try Rewardly for €0
No upfront payment, 14 days free.


