Customer retention rate: how to calculate it and tell whether your loyalty card works

What is customer retention rate?

Customer retention rate measures how many of the customers you had at the start of a period you still have at the end. It is a number between 0 and 100: at 100% you lost nobody, at 0% none of your starting customers stayed.

It does not measure goodwill or satisfaction, it measures a behaviour: coming back. That is why it is the most useful number for judging whether a loyalty card or a rewards programme is doing its job. The national picture, such as how many people in Italy use loyalty cards, is in our customer loyalty statistics; this guide is about the numbers in your own shop.

Retention rate has two close relatives: churn rate, which counts the customers you lost, and repeat customer rate, which counts the customers who came back at least a second time. Below you will find how to calculate each and when to use which.

Customer retention rate formula: how to calculate it

Customer retention rate is calculated from three numbers for the same period: customers at the start (S), customers at the end (E) and new customers gained in between (N). The formula is ((E − N) ÷ S) × 100. New customers are taken out because they were not there at the start: without that step, a month full of new faces would hide the customers you lost.

The worked example in the table is the one Salesforce publishes in its guide to customer retention rate (read on 30/10/2026): a month that starts with 107 customers, loses 8 and gains 21 ends with 120 customers, and the retention rate is 92.5%.

How to adapt the formula to a café, a salon or a gym

The classic formula was built for subscriptions and contracts, where you always know who is a customer. In a café or a salon nobody cancels: a customer simply stops coming. For a local business, then, «being a customer» means having visited at least once within a given stretch of time.

The practical version compares two periods of the same length: take the customers you saw in the first period and count how many of them you saw again in the second. A hypothetical example: in October a café recorded 200 customers with the card; in November 88 of them came back. The retention rate from October to November is 88 ÷ 200 = 44%.

Choose the period to match your regular customer’s rhythm: it should last at least twice the usual gap between two of their visits, or you will count as lost someone who was simply between appointments. The table applies that rule to different rhythms; the rhythms are examples, not industry averages.

Repeat customer rate, visit frequency and the gap between visits

Repeat customer rate is the share of a period’s active customers who came back at least a second time. The calculation is: customers with more than one visit ÷ customers with at least one visit × 100. If 150 card holders came in during a month and 57 came at least twice, the repeat customer rate is 38%.

Unlike retention rate it looks at one period instead of two, so it is quicker to read and useful even in the first weeks of a new card. Its weakness is that it rises on its own as the period gets longer: always compare periods of the same length.

Two more numbers complete the picture. Visits per customer is the period’s visits divided by active customers: if it grows, the same customers are coming more often. The average gap between two visits tells you when a regular is starting to run late, and it is the basis for deciding when to get in touch.

Customer churn rate: the other side of retention

Churn rate is the percentage of customers you lost in a period. On the same base as retention rate it is the complement: if retention is 44%, churn is 56%.

In a local business it pays to calculate churn on your regulars, not on everyone: someone who dropped in once as a tourist is not a lost customer. A practical rule is to treat a regular as lapsed when you have not seen them for more than twice their usual gap. A hypothetical example: of 60 regulars in September, 5 are lapsed by the end of October, so the churn rate among regulars is 5 ÷ 60 = 8.3%.

Rising churn is the signal to act at once, while the customer still remembers you. What to send and when is covered in the guide on how to win back lost customers.

Loyalty programme KPIs: which numbers to watch

A loyalty programme is judged on a handful of indicators, each answering one specific question. The table lists them with the formula and an example; the example figures are hypothetical.

Redemption rate deserves attention: if many customers reach the reward but few collect it, the reward probably does not appeal. Alternatives are in the loyalty card reward ideas, and you can work out what each reward costs you with the loyalty card reward calculator.

How to tell whether your loyalty programme works

A loyalty programme works if, all else being equal, more customers come back or come back more often than before. The simplest way to check is to note the numbers for a normal period before launch and compare them with a period of the same length and the same season after launch.

Even then, the comparison does not prove on its own that the card deserves the credit: weather, holidays, roadworks, promotions and changed opening hours all move visits. The programme’s numbers tell you what happened; for the why you need to look at everything else too. Some patterns, though, are clear enough to suggest a fix:

• Lots of sign-ups, few second visits: the first reward is too far away or the rule is unclear. Try a shorter target or a welcome stamp.
• Many rewards earned, few redeemed: the reward does not appeal, or customers do not know they have it. Change the reward or mention it at the counter.
• Steady repeat rate, falling visits per customer: customers stay but come less often. Look at the gap between visits and contact the ones running late.
• Sign-ups have stalled: the staff are not offering the card. The QR at the counter and one set sentence at the till matter more than any discount; ideas are in the guide on how to promote a loyalty card.
• Numbers that differ from one shift to the next: not everyone applies the rule the same way. Before changing the programme, make the counter consistent.

Common mistakes when measuring customer retention

Most wrong numbers come not from the formula but from how customers are counted. These are the most frequent mistakes, and almost all of them disappear once you set your counting rules and keep them fixed.

• Comparing periods of different lengths, such as a week with a month: the repeat rate of a longer period is always higher.
• Comparing different seasons, such as August with October, without saying so: for many shops the whole difference is seasonal.
• Counting stamps or receipts instead of people: ten stamps can be ten customers or one customer who comes in every day.
• Leaving new customers in the retention calculation: they inflate the number and hide the people you lost.
• Keeping test cards, staff cards or duplicates in: with few customers they skew the results.
• Drawing conclusions from small numbers: with 12 customers a single person is worth more than 8 percentage points. Wait until you have at least a few dozen customers or several periods to compare.

How Iterum (by FIDUS) measures retention

In Iterum (by FIDUS) these numbers are already calculated in the Results section of the dashboard, included on every plan. They start from verified visits, meaning visits confirmed by staff at the counter: undone visits are left out of every count, and customers use the card on their phone without downloading an app.

Iterum’s return rate compares returning customers with the customers active in the period: a customer counts as returning with at least two visits in the period, or one visit after an earlier one. Alongside it you will find verified visits, active customers, average visits per customer, new customers, sign-ups, rewards earned and redeemed, the redemption rate of each card, visits by day of the week and sign-up channels.

Cohorts show, for the customers who joined in each month, how many came back after one, two and three months; months with fewer than 5 sign-ups are not shown, out of respect for privacy and because a percentage says little on numbers that small. You choose the period from 7, 30 or 90 days, this month, last month or custom dates up to 366 days, per location if you like, and the data exports to CSV.

Iterum does not read receipts or the till: average spend stays in your own POS system. How the card is set up is explained in how Iterum works; you can see your own numbers from the first customer with the 7-day trial.

Sources and method

The classic formula and its worked example come from Salesforce’s page «Customer retention rate: formula & improvement tips» (salesforce.com/eu, read on 30/10/2026). The definitions of Iterum’s indicators are the ones used by the dashboard’s Results section on the same date.

Every other number on this page is a hypothetical example chosen to show the calculation: none is an industry average or a result from an Iterum customer. For national figures on loyalty cards, with the source and year of each one, see our customer loyalty statistics.