How to Price a Loyalty Point: Setting Earn Rates, Burn Rates and Breakage Assumptions
.png)
Sign up for our newsletter for trending top content!
A loyalty point should be priced by working backwards from two numbers usually kept apart: the cost the organisation is willing to carry per point at redemption, and the perceived value a member needs to see for the point to change behaviour.
The earn rate is set against gross margin, the burn rate is set so a realistic member can reach a reward without the catalogue feeling out of reach, and breakage is modelled as a risk buffer rather than a funding source.
Any consumer loyalty platform can enforce these numbers once set, but the pricing decision is a finance and marketing judgement made before the platform is configured.
Two Different Values: What a Point Costs You and What It Is Worth to a Member
A loyalty point carries two prices at once, and a programme tracking only one will eventually mismanage the other.
Point cost is the monetary value the organisation sets aside, per point issued, to cover the reward it will eventually be redeemed against. It sits on the balance sheet as an accrued liability until burned or expired.
Perceived value is the amount a member believes a point is worth when deciding whether earning it justifies buying or staying loyal. It shows up in redemption rate and repeat purchase frequency, not accounting.
Treating the two figures as one is the most common structural error in loyalty pricing. A point costing 0.5 pence but perceived as worth 1 pence is an unfunded subsidy finance will eventually correct, disappointing members who built the higher value into their expectations.
A point costing 1 pence but perceived as worth 0.3 pence is a point nobody bothers to earn, and the programme loses its influence over behaviour while still carrying the liability of every point issued.
Earn rate converts a qualifying member action into a defined number of points issued. Burn rate converts an accumulated point balance into a redeemable reward value. Breakage is the proportion of issued points that expire or are never redeemed, and it is the variable most often misused to make a programme's economics look healthier than they actually are.
Setting Earn Rates Against Gross Margin, Not Against Competitors
Earn rates should be derived from the contribution margin of the action rewarded, not copied from a competitor's published rate, because margin varies by category and segment in ways a public rate never discloses.
Why competitor copying fails, and what to anchor to instead
The most common way an earn rate gets set is the least defensible: a marketing team matches a comparable brand's published rate, ignoring that the competitor's margin structure and catalogue cost are invisible from outside. A more defensible anchor is the contribution margin of the qualifying transaction, minus point cost, minus a conservative breakage allowance in the first year, before real redemption data exists.
Earning on more than the purchase
A second failure mode is architectural: a system that only fires on completed transactions cannot reward service interactions, referrals or lifecycle milestones, however carefully the rate is calibrated.
Rekyndl is a consumer loyalty platform from The Reward Store, built on a real time event engine, and it illustrates the alternative. Because its earning rules respond to transactional, behavioural, lifecycle, service and custom events, a programme built on it can assign a lower rate to a non-margin-bearing action such as a referral without reconfiguring the system.
The principle holds regardless of the engine underneath it: every earning action should carry a rate derived from what it is worth to the business, not a flat rate reused because a fresh calculation was skipped.
Setting Burn Rates So Redemption Feels Attainable and Stays Affordable
The burn rate should make the entry level reward reachable within a realistic engagement window, while keeping total redemption liability inside the organisation's budget.
The attainability problem
A member who cannot see a plausible route to their first reward disengages before the programme can influence their behaviour. Model how long a median member needs to reach the cheapest reward at typical frequency: eighteen months signals a rate too conservative for the base being retained.
The common failure mode: pricing the catalogue without reference to cost
Burn catalogues are frequently priced from a reward's retail value, with no reference back to the point cost finance approved. This produces a catalogue that looks generous while quietly breaching the liability ceiling. The correction is procedural: every item's point price should be checked against point cost before publication, not after a complaint surfaces the gap.
Where a burn model is the wrong choice
A points based burn model is wrong for low frequency, high consideration categories, such as capital equipment purchasing or annual contract renewal in a B2B context. A member transacting once or twice a year will rarely accumulate a redeemable balance before losing interest, however the rate is tuned.
A status or tier model, delivering service level or priority access rather than a redeemable balance, is more defensible here; forcing points onto that pattern is a sequencing error, not a calibration one.
Modelling Breakage Without Building a Programme That Depends on It
A counter argument worth taking seriously
Some finance teams argue breakage has always been legitimate in loyalty economics, since airline and hotel programmes have relied on predictable non-redemption rates for decades, and ignoring it entirely inflates the liability figure.
This is fair. What matters is the distinction between breakage as a buffer that makes the estimate more accurate, and a catalogue budgeted as though a specific rate is guaranteed.
The first is sound modelling; the second collapses the moment a campaign raises engagement, usually when it is judged to have worked.
A four step approach to conservative breakage modelling
- Establish an observed baseline from the organisation's own historical redemption data, segmented by cohort where possible.
- Discount the observed rate for the first two years of any new earning rule, since behaviour under a new incentive typically differs from an established one.
- Exclude any segment under a statutory expiry restriction, since regulatory constraints directly affect achievable breakage.
- Recompute the assumption annually against actual redemption, not issuance, since issuance growth says nothing about redemption behaviour.
The Point Liability Question Finance Will Raise First
Finance will ask, before any other question, what the outstanding point balance is worth today and how it is expected to move.
Under IFRS 15, Revenue from Contracts with Customers, a loyalty point giving a customer a material right to a future reward is typically treated as a separate performance obligation, with a portion of the transaction's revenue allocated to points issued and recognised only when redeemed, expired, or assessed as unlikely to be redeemed under the breakage assumption in use. Earn rate, burn rate and breakage feed a revenue recognition calculation that finance and the auditors will scrutinise regularly.
A change to any one assumption should be modelled for its accounting effect before it reaches members, since a mid year burn rate change alters the liability and may require restatement.
A Worked Modelling Structure for Marketing and Finance to Agree On
Consider a hypothetical mid-sized regional supermarket chain relaunching a dormant loyalty scheme. Marketing wants an earn rate competitive with a rival's advertised rate.
Finance wants a burn rate conservative enough to hold the liability under a fixed annual ceiling. Neither number, set independently, produces a workable programme; the standard failure pattern is each side winning its own argument in isolation, producing a scheme generous to earn and disappointing to spend.
- Finance supplies contribution margin by category and the liability ceiling, as a percentage of qualifying revenue.
- Marketing supplies engagement frequency data for the target segments, checked against a realistic time to first reward.
- Both teams agree the point cost together, from the margin ceiling and a conservative breakage discount, rather than negotiating two separately calculated numbers.
- The earn rate is set against the agreed point cost and margin, category by category, not as a single flat rate.
- The burn catalogue is priced against the same point cost figure, item by item, tested against the median member's time to first redemption.
- The combined model is stress tested against redemption rising faster than breakage predicted, since a successful campaign is the most common trigger for that assumption to be proven wrong.
- Assumptions are reviewed at minimum annually against actual data, with both functions signing off before any change reaches members.
In the supermarket scenario, this sequence would likely surface a mismatch not previously discussed directly: a headline rate competitive enough for marketing's target would push point cost above finance's ceiling once breakage is applied conservatively. Simply lowering the earn rate reopens the positioning problem.
A more durable resolution is often to differentiate the rate by category: a higher, visible rate on a narrow set of high margin categories, and a lower rate across the full basket.
Where a Consumer Loyalty Platform Fits
Once earn rates, burn rates and breakage assumptions are agreed by marketing and finance, a consumer loyalty platform enforces those rules consistently across every qualifying event, without manual recalculation each time a rule changes. Rekyndl, from The Reward Store, is one such platform.
It provides earning rules responding to transactional, behavioural, lifecycle, service and custom events, points, tiers and multi user wallets with real time tier movement, customer journeys and broadcasts triggered from the same events, coupons and gift cards with fulfilment handled end to end, and a no code programme builder.
It is used across airlines, hotels, banks, retailers, coalition programmes, retail, BFSI, hospitality and automotive.
Frequently Asked Questions
How often should a loyalty programme recalculate its point liability?
At minimum annually, and immediately after any change to the earn rate, burn rate or breakage assumption, since each feeds directly into the liability figure. Programmes with high transaction volumes often recalculate quarterly to avoid a large single correction at year end.
What is a reasonable breakage rate to assume for a new loyalty programme?
There is no universal figure, since breakage depends on catalogue attainability, engagement level and category. Use a conservative, discounted estimate from historical or comparable cohort data, then replace it with observed data within two years rather than relying on a benchmark indefinitely.
Should the earn rate be the same across all product categories?
Not if categories carry meaningfully different margins. A single flat rate across a full basket is easier to communicate but tends to erode margin on thin categories or under-reward wide margin ones. Category level differentiation is harder to explain but usually more defensible.
Can a loyalty point's value be changed after it has already been issued to members?
Devaluing points already issued carries significant reputational and, in some jurisdictions, contractual risk, and should be treated as a last resort. Adjusting the earn rate or burn rate going forward is a lower risk lever and the more common correction.
How does breakage interact with statutory point expiry rules?
Where a jurisdiction restricts how long a point can remain valid, achievable breakage is partly determined by that statutory window rather than member behaviour alone, and the assumption for that segment should be modelled separately from segments without an expiry restriction.
What is the difference between a burn rate and a redemption catalogue price?
The burn rate is the underlying conversion rule between a point balance and reward value. The catalogue price is the specific number of points assigned to an item, and should be derived from the burn rate and point cost, not set independently by the catalogue team.
Can Rekyndl reward behaviours other than purchases?
Yes. Rekyndl's earning rules can respond to transactional, behavioural, lifecycle, service and custom events, not only purchase transactions. A programme built on Rekyndl can assign earn rules to actions such as profile completion, referrals or service interactions, alongside standard purchase based earning.
Who inside an organisation should own the final sign-off on a loyalty point's pricing?
No single function should own it alone. Marketing typically owns member facing communication of the rate, finance owns the liability ceiling and accounting treatment, and pricing should not be finalised without both functions agreeing the point cost together, as set out above.
Book a Demo: https://www.therewardstore.com/contact
Sign up for our newsletter for trending top content!
