Can You Actually Afford to Discount? A Margin-First Guide to Loyalty

Most loyalty programs are designed backwards. A brand decides it wants a program, picks a headline number that sounds competitive — “earn 5% back,” “10% off your next order” — and launches. The rate gets set by what the market seems to expect, not by what the business can actually sustain. Then, a few quarters in, finance looks at the give-back line and asks the uncomfortable question nobody asked at the start: can we even afford this?

The answer, worked out properly, should come first — not last. Set the give-back rate too high and you quietly destroy margin on every order. Set it too low and you leave real repeat-purchase revenue on the table. Getting it right isn’t guesswork; it’s a calculation. Here’s the margin-first way to run it.

Start with contribution margin, not gross margin

The single most common mistake is funding a loyalty program out of gross margin. Gross margin flatters you — it ignores the variable costs that a repeat order actually incurs.

The right number is contribution margin (CM): what’s left of a sale after subtracting every variable cost — COGS, shipping, fulfilment, payment fees, and returns. It’s the honest measure of what a sale contributes to the bottom line, and it’s the number your program should be funded from. Using CM instead of gross margin makes the whole program more financially defensible and more resilient when costs move.

Most DTC brands sit somewhere between 25% and 40% contribution margin. That range matters, because everything downstream is a slice of it.

Decide how much of that margin you’ll reinvest

You are never going to reinvest 100% of contribution margin into loyalty — that would mean running the program at zero profit. The real question is: how much of your margin are you prepared to reinvest into retention?

This is a strategic posture, not a formula, and it should map to how margin-sensitive and growth-hungry the business is:

Most brands land in the 10–20% band. From here, your affordable spend falls straight out of one line of arithmetic:

Affordable give-back rate = contribution margin × reinvestment rate

Two worked examples make the range concrete:

So a brand with a healthy 40% CM that’s willing to reinvest a fifth of it can afford a program that costs roughly 8% of eligible revenue. That 8% is your ceiling. Every design decision from here has to fit underneath it.

And to be clear about framing: a loyalty program funded this way is not a cost centre. It’s an investment that returns multiples over time. The discipline of the affordable rate isn’t about spending less — it’s about making sure every dollar you do spend is one you can defend.

The number you advertise is not the number you pay

Here’s the piece most brands miss, and it’s where the margin math gets genuinely interesting. There are four different “rates” in a discount program, and confusing them is what wrecks P&Ls:

You advertise the headline rate. You only pay on what’s redeemed. The gap between the two is breakage — the difference between the value of discounts awarded and the value actually claimed — and it works in your favour.

This is the lever that lets you advertise a rate more generous than your affordable ceiling. If your affordable cost is 6% but only 40% of issued value ever gets redeemed, you can advertise a headline rate well above 6% and still land inside budget. The relationship is:

Headline give-back rate = affordable rate ÷ modelled redemption rate

Model your redemption rate honestly

Everything hinges on that redemption assumption, so don’t guess it. Several factors push it up or down:

• Currency. Points redeem less often than cashback — they carry friction, because customers have to accumulate, understand the conversion, and choose to redeem. Cashback is a dollar and everyone understands a dollar, so it redeems at high rates.
• Member tier. Higher tiers redeem more. Engaged, high-value members claim what they earn; disengaged base-tier members let it lapse.
• Communications. Redemption climbs sharply with active nudges — balance reminders, expiry alerts, tier-progress prompts. A neglected program decays into invisible, unredeemed balances.
• Vertical. Frequent-purchase categories — beauty, supplements, food — give customers more chances to accumulate and redeem, so they run higher.

Okendo benchmarks put the redemption bands roughly here:

Now you can see how affordable rate, redemption, and headline rate move together. A few reference points from the model:

Read across the 6% row: the same affordable budget supports a 20% advertised rate if redemption sits at 30%, but only 12% if half your issued value gets claimed. Same cost to you — very different offer to the customer, entirely driven by the redemption assumption. This is why the currency choice is a margin decision, not just a UX one.

A margin warning on cashback

That last point deserves emphasis, because it’s counterintuitive. Cashback feels efficient — it’s simple, it’s understood instantly, and it’s close to set-and-forget. But it is also the most expensive currency you can run.

Because cashback has low redemption friction, breakage is low — which means your effective cost sits close to the full headline rate. You get predictable margin erosion with very little of the breakage cushion that points give you. Points, by contrast, go unredeemed more often, so the effective cost runs well below the headline rate, and the currency is far more versatile for bonus events, multipliers and non-purchase rewards. The trade-off is complexity and a larger liability on the balance sheet.

Neither is wrong. But “cashback is simpler” is not the same as “cashback is cheaper” — and confusing the two is an expensive mistake.

Let tiers carry your most generous rate

The headline rate you calculate is a blended number — the weighted average across your whole member base. And blending is where tiers earn their keep.

The strategic move is to reserve your richest rates for your top tiers. Because top tiers hold only a small share of total revenue, you can advertise a genuinely generous rate to your best customers while the blended cost stays comfortably inside your affordable ceiling:

Blended headline = Σ (tier revenue share × tier rate)

This is what makes tiering economically powerful. A flat program can never advertise a rate richer than what’s affordable across the entire base. A tiered program can promise 15% to Superfans — the customers most likely to notice and respond — and still land at a defensible blended cost, because most members sit in lower tiers earning less. You concentrate the generosity where it changes behaviour and dilute it everywhere it wouldn’t.

The four numbers to walk away with

Run properly, the whole exercise is four steps and three numbers:

The three numbers you hold at the end: your affordable rate (what you can spend), your redemption rate (what customers will claim), and your headline rate (what you advertise — always higher than affordable, because redemption is never 100%). The gap between headline and affordable is funded by breakage.

The point of all this isn’t to talk yourself out of discounting. It’s to make sure that if you do discount, you’re doing it from a number you can defend to your CFO — not one you picked because a competitor did. Loyalty that erodes margin unexamined is a liability. Loyalty built on a proper give-back model is one of the highest-return investments a brand can make. The difference is entirely in the math you do before you launch.