The Hidden Economics of Loyalty: 2026 Trends from High-Performing Loyalty Programs

Get the report!

Loyalty Program Breakage: 20 Questions Finance and Loyalty Teams Ask

Breakage is the most consequential number in a loyalty program and the least discussed. It decides the liability on the balance sheet, the revenue deferred each month, and the cost of every point issued. It also shapes whether finance sees the program as an investment or a liability.

Most teams inherit the number rather than derive it. Someone built a spreadsheet, the auditor accepted it, and it has been rolling forward ever since.

We get asked the same questions about breakage constantly. Here are twenty of them, answered plainly.

Want the mechanics of the cost calculation itself? The three bases of ultimate redemption rate, mix shift, and the monitoring checklist are covered in Loyalty Program Cost Calculation: Why Breakage Is the Assumption That Decides It.

Breakage, defined

1. What is breakage in a loyalty program?

Breakage is the share of issued points that will never be redeemed. If 10% of the points you issue will expire, lapse, or sit unused forever, your breakage rate is 10%.

It matters because you only pay for points that get redeemed. Points that break cost you nothing. Breakage is the assumption that converts “points outstanding” into “money owed,” and it sits underneath every cost and liability figure the program produces.

2. How is loyalty program breakage calculated?

Breakage is one minus the Ultimate Redemption Rate (URR). If 90% of points will ultimately be redeemed, breakage is 10%.

The liability that follows is roughly:

The formula

Liability = Outstanding points × (1 − Breakage rate) × Cost per point

Equivalent to: Outstanding points × Ultimate Redemption Rate (URR) × Cost per point

The arithmetic is trivial. The hard part is the ultimate redemption rate, which is a forward-looking prediction about behavior that has not happened yet, for points that may not get redeemed for another several years.

3. What is the difference between Ultimate Redemption Rate (URR) and redemption rate?

A redemption rate describes what has already happened. Of the points issued in some period, how many have been redeemed so far. It is a historical count.

Ultimate Redemption Rate (URR) describes what will happen. Of the points issued in some period, how many will eventually get redeemed. It is a prediction.

The two get conflated constantly, and the conflation is expensive. A program with a 40% redemption rate to date may have a 90% URR, because the remaining points simply have not been redeemed yet.

4. What is the difference between breakage rate and ultimate redemption rate?

Two views of the same quantity. Ultimate redemption rate (URR) is the percentage of points that will ultimately be redeemed. Breakage is the remainder.

A URR of 92% means breakage of 8%. Nothing is lost in the translation. Finance teams tend to speak in breakage because that is the side that reduces the liability. Actuaries tend to speak in URR because that is the side being modeled.

5. Which breakage number should we be using?

This is the question behind most breakage errors, and the answer depends on which formula you are filling in.

There are three bases. They produce three different numbers, and they are not interchangeable.

BaseWhat it answersWhere it belongs
Outstanding pointsOf the points sitting unredeemed today, what share will be redeemed?The liability on the balance sheet
Cumulative earned pointsOf every point ever issued, what share has been or will be redeemed?Program-lifetime economics
Current-month issuanceOf the points issued this month, what share will be redeemed?Revenue deferral under ASC 606 and IFRS 15, and trend analysis

Using one where another belongs is the most common mistake we see. The comparison we keep reaching for is Fahrenheit and Celsius. Both are temperature, neither substitutes for the other, and 70 in the wrong one will do real damage.

We have seen a liability overstated by 50% from exactly this. Right formula, wrong base. The full mechanics are here.

Reading a breakage rate

6. What is a typical breakage rate for a loyalty program?

There is no single number, and anyone who gives you one without asking about your program is guessing.

Breakage varies by industry, by program design, by expiration rules, and by how much of the point supply sits with co-brand card members rather than occasional earners. A travel program with no expiration and a large dormant base behaves nothing like a coffee program with a 90-day reward window.

Benchmarks are a gut check. We hold them, and they will tell you quickly whether a number is implausible. They are not an answer. The useful question is whether the number is right for your program, and answering that means looking at your own members’ behavior.

7. What does it mean if our breakage rate is falling?

Usually, that the program is working.

A falling breakage rate means a larger share of your points are owned by people who intend to use them. That happens when you accumulate engaged, high-value members, which is what a loyalty program is for. Those members hold more of the point supply over time, and they redeem.

It costs more. It is also the signature of a healthy program, which is an uncomfortable thing to explain to someone reading only the cost line.

8. What does it mean if our breakage rate is rising?

Short-term relief, long-term warning.

Rising breakage lowers your redemption costs immediately, which looks like a saving. What it usually indicates is thinning engagement. More points are being held by members who will never come back. That shows up in customer lifetime value years before a quarterly report would catch it.

You are optimizing the short game and eroding the long one.

9. Why does our breakage rate change when we haven’t changed the program?

Mix shift. It is the single biggest driver of movement in ultimate redemption rates and breakage rates, and it happens whether or not you touch anything.

The mix of members earning points twelve months from now will not be the mix earning them today. New members join. Some lapse. A co-brand partnership shifts who is accruing. A tier skews the distribution. Each group redeems at a different rate, so the blended rate moves even when every individual behavior stays constant.

There are dozens of dimensions along which mix can shift at once. A model that looks at the program in aggregate, or across one or two segments, cannot see it. That is why so many teams find out at year end.

Mix shift also accelerates when you do make changes. A new card, a new redemption partner, an expiration rule, a push on acquisition. No program worth running sits still, so mix shift is permanent rather than occasional.

Breakage and the financial statements

10. Is breakage recognized as revenue?

Indirectly, yes. When you issue points you defer a portion of the transaction’s revenue into a liability. In most interpretations of the accounting standards, breakage determines how much you defer, because you only defer against the points you expect to be redeemed.

So the breakage assumption releases revenue you would otherwise have held back. The higher the estimate, the less you defer, and the more revenue you recognize now. That is why auditors look at it closely. It is an estimate with a direct line to the top line.

It’s worth noting that there are instances where the accounting standards specify a slight different treatment for breakage.

11. Where does breakage show up on the financial statements?

In three places. On the balance sheet, as the loyalty liability, which is outstanding points net of expected breakage. On the income statement, through revenue deferred against each month’s issuance and released on redemption. And in the disclosures, where a material loyalty program is often named as a critical accounting estimate.

For large programs the loyalty liability is among the biggest estimated liabilities on the balance sheet. In financial services, hotel and airlines, the balances run into the hundreds of millions to billions of dollars.

12. Can a brand-new program assume any breakage at all?

Yes, and most should.

The standards expect your liability to reflect expected redemptions, which means taking a haircut for breakage, provided you believe the estimate will not lead to a material restatement. A new program has no history, which makes the estimate harder. That does not make zero the right answer.

Plenty of companies assume zero for the first year or two, then move to a non-zero figure once data arrives. It is a defensible, conservative choice. It also overstates the liability, understates revenue, and sets up a sizable correction later. Make that trade deliberately rather than by default.

13. Why do auditors treat breakage as a critical accounting estimate?

Because it is highly leveraged. You are multiplying a single percentage by an enormous balance of outstanding points, so a small movement in the rate moves a large number of dollars.

A program with a billion points outstanding and a one-point move in breakage has just moved its liability materially, with no change in member behavior that anyone noticed. Auditors flag it for the reason we do. The sensitivity is out of proportion to how much attention the number usually gets.

Expect your auditor to bring in their own actuary. For a material program, they will.

14. Do breakage rules differ between the US and the rest of the world?

Barely. The US applies ASC 606. Most other jurisdictions apply IFRS 15. The two were written to align, and on loyalty points they are near-identical in substance. Neither has changed materially since the 2018 to 2019 transition.

Where programs diverge is interpretation. How the standard is applied varies by industry and, within an industry, from one company to the next. Two airlines on the same standard can treat the same question differently and both be defensible.

So a multinational is rarely managing two rulebooks. It is managing one rulebook and several interpretations, which is harder to spot and harder to reconcile.

15. Who owns the breakage liability in a franchise program?

It depends on the model, and the difference is consequential.

In most hotel programs there is a central legal entity. A share of each stay’s revenue flows into it, it funds the points issued, and it reimburses the property that takes the redemption. The liability sits centrally, with the entity that has the resources to model it, audit it, and defend it.

In much of quick-service restaurant industry, there is no such entity. The franchisee books the liability on their own balance sheet and funds their own redemptions, with no central reconciliation. That creates three problems. A small-business owner has neither the time nor the actuarial expertise to estimate the number properly. A large and possibly wrong liability distorts financial statements that banks read when extending credit. And the franchisee experiences redemptions as a cost they absorb rather than a program they benefit from.

It is a structural weakness rather than an accounting detail, and very little of it is disclosed publicly. We think centralizing it is the better design.

Breakage, redemption and value

16. Should we add friction to redemption to keep breakage high?

No. It is one of the most expensive things a program can do to itself.

Redemption is an inflection point, not a leak. Our benchmark work found that members who make a first redemption go on to be materially more profitable than closely matched members who were equally engaged and able to redeem but simply did not. The lifetime value gap can run into the hundreds of dollars per member.

So restricting redemption to manage the liability line forfeits the single event most associated with a member becoming more valuable. You bank a visible cost saving and pay for it in revenue you never see, because it arrives as an absence.

The same logic argues against devaluation under cost pressure. Devaluing a program erodes trust with the exact members you most want to keep, which defeats the purpose of having one.

17. How does breakage relate to customer lifetime value?

Directly, and in the opposite direction to intuition.

A loyalty program trades margin for volume. You give away points, which costs margin, in exchange for members coming back more often over a long horizon, which adds revenue. The question is whether the volume gain exceeds the margin given up.

Breakage is the dial on the margin side. Lower breakage means higher redemption costs and thinner margin. It also means more engaged members, which lifts long-term revenue. What we see repeatedly is that the revenue gain more than offsets the margin cost, so the net effect on customer lifetime value is positive.

Read in isolation, falling breakage looks like a problem. Read against CLV, it is usually the program doing its job. No good business decision has ever been made by looking at costs alone.

18. What does getting breakage wrong actually cost us?

Three things, in rising order of pain.

The first is financial volatility. A breakage estimate held flat through the year, then revised at year end, produces a variance against plan. Multiplied through a large outstanding balance, that variance routinely runs to tens of millions of dollars. We have met teams who had come to treat this as weather, an annual surprise nobody enjoyed and nobody thought was solvable.

The second is decisions made on wrong numbers. Redemption cost is the single largest line in the loyalty P&L, and breakage is the assumption that sets it. Get it wrong and every downstream decision, from earn rates to tier thresholds to partner economics, rests on a figure that is not true.

The third compounds. A loyalty team that cannot defend its numbers cannot win the investment argument, and a program that cannot win the investment argument eventually gets cut or devalued. The cost of a wrong breakage number is not only the restatement. It is arriving at the budget conversation without a case.

Getting the number right

19. What data do you need to model breakage properly?

More granularity than most teams expect, and the granularity is the point.

A defensible model predicts at the level of the individual point. When it was issued, to whom, under what earn mechanic, with what expiration, alongside everything known about the member holding it. That is what lets the model see mix shift as it happens rather than after it has distorted the number.

It also matters for compliance. The standards require you to defer revenue against each month’s issuance specifically, so a model that cannot isolate this month’s points from last month’s cannot produce the figure the standard asks for.

A spreadsheet working in aggregate will produce a number. It will not produce the right number.

20. Does loyalty breakage need a different kind of actuary?

Yes. Finance leaders already do this instinctively elsewhere. A pension obligation is a long-term liability estimated by a credentialed actuary who applies professional standards, documents the method, and opines on the result, so the finance leader signing the statements does not have to carry the uncertainty personally. Nobody asks whether someone in the finance team should have a go at valuing the pension.

Loyalty liabilities are structurally the same kind of problem and almost never get the same treatment. Most loyalty programs have never spoken to an actuary at all.

Not every actuary fits either. Insurance and pension methodology assumes a portfolio that is not actively trying to change its own behavior. A loyalty program is doing exactly that, deliberately and continuously, which demands models that respond to behavioral change rather than smoothing over it.

What this means for loyalty and finance leaders

Most of these twenty questions have the same answer underneath them. Breakage is a forward-looking prediction being treated as a backward-looking measurement.

A rate derived from how members redeemed in the past is reliable only if the program is stable and the membership is unchanging. Neither is true of any program worth running. The whole purpose of a loyalty program is to change behavior, and if it is succeeding, the mix of members holding your points is different this quarter than last.

So the question worth asking is where your breakage is going, and whether you will see it coming.

Teams that can answer it get stable, predictable financial results and a defensible position in front of their auditor. Teams that cannot find out at year end, when there is no time left to do anything about it.

If you are not confident in your breakage number, or you are confident but have never had it independently reviewed, book a conversation with the KYROS actuarial team.

Subscribe for Updates

Related Resources