Guide 04 · Finance
Loyalty program ROI
Loyalty program ROI is the extra profit the program creates, minus everything it costs, divided by that cost. Measure the extra profit by comparing members with similar customers who did not join, not by adding up what members spend. Count the full cost: rewards, the liability for unredeemed points, technology and the people who run the program.
Loyalty programs are easy to defend with the wrong numbers. Member revenue looks impressive, enrollment grows every month, and redemptions feel like proof. None of these shows whether the program pays for itself. This guide sets out how to measure the return honestly, from the full cost to the extra profit the program actually creates.
Define ROI
What does loyalty program ROI measure?
It measures the extra profit the program creates, minus everything the program costs, divided by that cost. The key word is extra. Starbucks, for example, said Rewards members drove nearly 60% of its US company-operated revenue in fiscal 2025, and McDonald’s CEO said loyalty members drive about 30% of its sales. Those figures show how much spending runs through a program. They do not show how much of it the program caused, because the most loyal customers are the ones most likely to join.
Count the cost
What does a loyalty program cost?
More than the rewards. A complete cost line includes the rewards members redeem; the liability for points earned but not yet redeemed; discounts given through member prices; the loyalty platform, integrations and app; marketing to recruit and keep members; customer service for program questions; fraud prevention; and the team that runs it. Partner programs add what you pay partners when members redeem with them. Leave any of these out and the return looks better than it is.
Measure the lift
How do you measure the revenue the program creates?
Compare members with similar customers who did not join, over the same period. The best comparison is a control group: customers who were eligible but were not offered the program, or members who did not receive a specific offer. When a control group is not possible, compare members with non-members who looked similar before joining, for example in visit frequency and spend, and follow both groups over time.
Segment the results. McDonald’s describes its members by how often they visit: more than once a month, about once every 90 days, and those who have not visited in a few months. A program usually changes these groups differently, and the average hides it.
Account for points
How should points be accounted for?
A point is a promise. Many businesses record unredeemed points as a liability and defer part of the revenue from the purchase that earned them, recognizing it when members redeem or when the points are expected to lapse. Starbucks, for example, defers revenue for the Stars members earn toward free products and records a liability for them. That treatment moves profit between periods, so the ROI should use the same numbers as the finance team.
The estimate of points that will never be redeemed, called breakage, matters most. A high breakage assumption lowers the liability and flatters the return; if members then redeem more than expected, the cost comes back. Keep the assumption conservative and review it at least once a year.
Track the metrics
Which metrics should you track?
| Metric | What it shows | What to watch |
|---|---|---|
| Incremental revenue | Extra spending caused by the program | Measure against a control group, not all customers |
| Active member rate | Share of members who bought or engaged recently | Enrollment alone hides inactive members |
| Redemption rate | Share of points or offers actually used | Too low means rewards miss; too high can burn margin |
| Points liability | What the program owes members | Read it with the breakage assumption behind it |
| Churn | Members who stop buying or engaging | Rising churn erodes lifetime value |
| Total program cost | Everything the program costs to run | Include technology, people and partners |
Report activity alongside the money. White Castle, for example, said its Craver Nation members saved a combined $5.1 million in one season, earned 32 million points and redeemed more than 1.2 million offers. Numbers like these show engagement; the incremental revenue line shows whether that engagement paid off.
Sources for this section
White Castle launches Craver Nation Rewards season 3 after members save $5.1M (QSR Web) F-WC-01
Avoid the usual mistakes
Where do ROI calculations usually go wrong?
The most common mistake is counting all member revenue as program revenue. A second is leaving costs out: the platform, the team and the liability for unredeemed points are often missing from the cost line, while every redemption is counted as a win. A third is choosing the comparison after the results are in, which makes it easy to find a group the members beat.
Timing causes quieter errors. A program measured in its first months has issued many points and seen few redemptions, so its costs look low. Measured a year later, the redemptions arrive and the return falls. Measure over a full year, and keep the liability in the calculation from the start.
Finally, watch for changes the program did not cause. A price cut, a new store or a competitor closing can lift member spending at the same moment the program launches. A control group protects you from most of these, because both groups feel the same outside changes. When you cannot run one, write down what else changed during the period and say how it could affect the result.
Agree on the method with the finance team before launch: which costs count, which comparison group you will use, how points are valued and how often you will report. A number everyone accepted in advance is far more useful than a better-looking one negotiated afterward.
Do the math
How do you calculate the ROI?
Take the incremental profit, not revenue, that the program created over the period, subtract the total program cost, and divide by that cost. In a hypothetical program that creates $2 million of incremental profit at a total cost of $1.5 million, the ROI is ($2 million minus $1.5 million) divided by $1.5 million, or about 33%. Run the same calculation with a lower and a higher breakage assumption, and with the most cautious comparison group you have. If the program only pays off under the most generous assumptions, it is not yet paying off.
More answers
Frequently asked questions
What is a good ROI for a loyalty program?
Any return above the cost of running the program, measured against customers who did not join. There is no universal benchmark, because programs differ in rewards, margins and how they are counted. Set your own target before launch and measure against it.
Why is member revenue not the same as loyalty ROI?
Because your best customers join first. Members would spend more than average even without the program, so total member revenue overstates what the program caused. Only the difference between members and similar non-members is the program's effect.
How are loyalty points accounted for?
Many businesses treat unredeemed points as a liability and defer part of the revenue from the purchase that earned them, recognizing it when the points are redeemed or expected to lapse. Starbucks, for example, defers revenue for the Stars members earn. Your accountants decide the exact treatment.
How long before a loyalty program pays off?
It depends on how often customers buy and how quickly behavior changes. Programs with frequent purchases show effects sooner. Plan to measure at least a full year, so seasonal patterns and the first wave of redemptions are included.
Which metrics should a loyalty program report every month?
Active members, incremental revenue against a control group, redemption rate, the points liability and the breakage assumption behind it, churn, and the total cost of the program. Each one should reconcile with the finance team's numbers.
Does breakage improve ROI?
On paper it can, because points that are never redeemed are never paid for. But breakage is an estimate, and high breakage often means members are disengaged. Count it conservatively and do not design a program that depends on it.
Sources
- F-SBX-02Starbucks reimagines its Rewards program with new levels and exclusive perks (about.starbucks.com)
- F-SBX-04Starbucks Corporation Form 10-K, fiscal 2025 (sec.gov)
- F-MCD-01McDonald's to create tiered loyalty program to lure repeat visitors (Reuters)
- F-MCD-02McDonald's to create tiered loyalty program to lure repeat visitors (Reuters)
- F-WC-01White Castle launches Craver Nation Rewards season 3 after members save $5.1M (QSR Web)