Loyalty Program Metrics: How to Tell If Yours Is Working

By Alva Chew · Updated 2026-11-02

In short Track five loyalty program metrics: the share of sales tied to a member, active members, repeat purchase rate, redemption rate and reward cost as a share of member spend. Compare members with their own history before joining, not only with non-members, and judge after about three months. A points program makes this simple because every number comes from one balance per customer.

Which loyalty program metrics should you track?

Five numbers are enough for a café, restaurant or shop. Each one answers a question you can act on, and each can be worked out from a list of member purchases.

Metric	Formula	Question it answers
Capture rate	Sales made to identified members / all sales	Is the program reaching the counter?
Active members	Members who earned or redeemed in the last 90 days	Is the member list alive or just long?
Repeat purchase rate	Members with 2+ purchases / members who bought	Are people coming back?
Redemption rate	Points redeemed / points issued	Do members use the reward?
Reward cost	Value of rewards redeemed / member spend	What does the program cost you?

Source: Gimify calculations. Pick the 90-day window to suit your visit pattern and keep it the same every time.

Skip vanity totals such as “1,200 sign-ups”. A list that large with 80 active members is a small program. The active count and capture rate tell you more than the total.

How do you work out each metric?

Use a worked example. A café issued 100,000 points in a quarter at 1 point per US$1, and 100 points are worth US$5.

Figure	Value
Total sales in the quarter	US$100,000
Sales tied to members	US$40,000
Capture rate	40%
Points redeemed	30,000
Redemption rate (30,000 / 100,000 points issued)	30%
Value of rewards redeemed (30,000 / 100 x US$5)	US$1,500
Reward cost as a share of member spend	3.75%

Source: illustrative US$ figures. The points issued here match the total of sales, so the example assumes all spend earned points. Use your own.

The reward cost line is the one owners skip. It tells you what the program costs before any extra sales. Compare it with the break-even sum in our guide on whether a loyalty program is worth it. For the repeat measure, the definition and a worked case sit in the glossary entry on repeat purchase rate.

What does a healthy redemption rate look like?

It depends on the reward and your visit pattern, so read the direction and the reason, not a target.

Research on a real points program shows why a low rate is common and why it is not automatically good. Stourm, Bradlow and Fader followed customers of a home and hardware retailer whose program earned 1% of spend. Only 3% of purchases had a redemption, and 60% of customers never redeemed in 43 months. The authors attribute the hoarding mostly to psychological motives, not to rational planning. Members who redeemed had baskets about 25% larger and visited far more often, though the paper describes that as an association, not proof the reward caused it.

A rate near zero usually means the reward is too far away, too small or too hard to read.
A rate where most points are spent within a few weeks usually means the reward is too rich.
A steady rate with members redeeming after a handful of visits is the pattern to aim for.

Source: Stourm, Bradlow and Fader, Journal of Marketing Research, 2015. One international retailer, one cohort. The glossary entry on redemption rate covers the two meanings of the term, and our guide to rewards customers actually redeem covers how to raise it.

How do you avoid fooling yourself with the numbers?

Regulars join first. If you compare members with non-members, the members will look better even if the program did nothing. Three habits fix that.

Compare members with themselves. Take each member’s visits and spend in the three months before joining and the three months after. This is the cleanest test a small shop can run.
Compare with a similar group. Pick customers who visit as often as members did before joining but have not joined. If both groups moved the same way, the program did not cause it.
Watch the first three months separately. In a two-year study of a convenience store franchise, Liu found light and moderate buyers gradually bought more, with the largest change in the first three months, while heavy buyers did not raise their spending (Liu, Journal of Marketing, 2007, read on the author’s summary). If your gain shows up only in your heaviest buyers, the reward is mostly paying for visits you already had.

Also hold the rest steady where you can. A price rise, a new menu or a road closure moves visits as much as a loyalty program does. Note the date of each change next to your monthly numbers.

Why do members drift away, and what do you check first?

Look at the metrics in this order when the numbers disappoint.

Low capture rate. Staff are not offering it, or sign-up takes too long. The fix is a one-sentence prompt at payment. KPMG found that 60% of consumers in 20 countries said programs were too hard to join or earn in (KPMG via PR Newswire, 2019, 18,000+ consumers).
Good sign-ups, few active members. The reward is unclear or too far away. Check the earn rate and make sure a first reward is reachable in a few visits.
Active members, flat repeat rate. The program is rewarding visits that would have happened anyway. Reduce the reward rate before you add rules.
High redemption, high cost. The reward is too rich for your margin. Lower the reward share of spend, and announce it in advance.

Notice what the list does not include: adding stamps, tiers, vouchers or games. Each adds a rule customers must learn, which is the opposite of what a weak-capture or weak-activity problem needs. We set out the evidence in the case for points only.

What should you do this week?

Write down total sales and member sales for the last 90 days, then work out capture rate.
Count members who earned or redeemed in that window.
Work out redemption rate and reward cost with the table above.
Pick 20 members and list their visits before and after joining.
Set a date three months out to look again, and write down what result would make you change the reward.

Gimify records points per member on one balance, earned in proportion to each purchase, so the earn and redeem history that these formulas need sits in one place, even across outlets. One outlet is free with unlimited members, so you can run this check before paying anything.

Questions owners ask

What are the most important loyalty program metrics?

For a small shop, five: the share of sales made to identified members, active members, repeat purchase rate, redemption rate and reward cost as a share of member spend. Together they show who joined, who returns, who uses the reward and what it costs you.

How long before I can tell if my loyalty program is working?

Give it about three months. Compare each member's visits and spend with their own record before joining. Earlier than that, early joiners are mostly your existing regulars and the numbers flatter the program.

What is a good redemption rate?

There is no universal figure. A rate near zero means the reward is too far away or too hard to understand. A rate where most points leave within weeks means the reward is too rich. Look at whether your rate is steady and members can see a reward within a few visits.

Why do members spend more than non-members?

Often because they were already your best customers before they joined. That is why you compare members with themselves before and after, and with similar customers who did not join.

Do I need software to measure a loyalty program?

You need a record that ties each purchase to a person. A spreadsheet works for a few dozen members. Past that, a points program that records every earn and redeem against one member account saves the counting.