Guide
How to measure loyalty program ROI
Most loyalty reporting shows you how many stamps you gave out, which tells you how busy you were, not whether the programme did anything. These are the numbers that answer the actual question.
· 7 min read
The one number that matters
Repeat rate: of the customers who joined in a given month, how many came back at all. If that figure was 40% before you launched and it is 40% after, your programme is a discount and nothing else. If it moves to 55%, you have changed behaviour and can work out what that is worth. Everything else in this guide is a way of understanding why that number moved.
Measure frequency against a baseline
Take your regulars and ask how many days sat between their visits before the card and after. A shift from ten days to eight across a few hundred customers is a large amount of extra trade and almost invisible in daily takings. This is where loyalty schemes actually earn their keep, and it is undetectable without a customer record — which is the honest argument for digital over paper, more than fraud or convenience.
Incremental trade, and being honest about it
The temptation is to count every stamped visit as a win. Most of them would have happened anyway. The rigorous version is to compare the behaviour of customers who joined against those who did not, over the same period — if joiners visit more than non-joiners by a margin that appeared after they joined, that difference is your incremental trade. It will be a smaller number than you hoped. It is also the only one worth putting against the cost.
What to do with lapsed customers
The most valuable output of a loyalty programme is often not the reward at all — it is the list of people who used to come weekly and have not been in for a month. That group is knowable only if you have a customer record, and reaching them is usually the highest-return action available to a café. If your programme cannot tell you who has lapsed, it is doing half a job.
Metrics that look good and mean nothing
Total stamps issued rises when you are busy. Cards created rises when staff push signups. Rewards redeemed rises when you are generous. None of them says whether customers changed their behaviour. If a dashboard leads with those and buries repeat rate, it is designed to make you feel good rather than to help you decide.
Questions
- How long before I can tell if it is working?
- Long enough for a customer to complete a card twice — typically two to three months in a café. Judging in week two measures novelty, not loyalty.
- What is a good repeat rate for a café?
- It varies enormously by location and trade. The useful comparison is not against an industry number but against your own figure before you launched, which is why it is worth recording it now.
- Should I count a reward redemption as revenue?
- No. Count it as cost, at food cost rather than menu price. Counting giveaways as sales is the fastest way to convince yourself a programme is working when it is not.
Related guides
How much does a loyalty program cost?
What a café loyalty program really costs: software, the rewards you give away, staff time, and the one figure most owners forget to calculate.
How to increase repeat customers in a café
Retention beats acquisition in a café, and most of it is operational rather than promotional. What actually moves repeat visits.
How to start a café loyalty program
A practical guide to launching a loyalty program in a café: choosing the reward, setting the target, launching it at the counter, and knowing whether it worked.
Put your loyalty card in your customers’ pockets
Publish a branded card in about five minutes. 14-day free trial, cancel any time.