Loyalty & Retention

How to Set Your Cafe's Loyalty Earn Rate & Rewards

By The Tany Team 8 min read

Almost every cafe knows it should run a loyalty program. Far fewer have actually done the math on how generous it should be. Too stingy and no one bothers chasing the reward; too generous and you’re quietly handing away margin on every regular you already had. The right answer isn’t a gut feel — it’s a number you can calculate from your own food cost and average ticket.

This guide gives you that number. We’ll set an earn rate and reward threshold that reliably drives repeat visits, then prove it pays for itself using the two facts owners most often forget: a coffee’s real cost is a fraction of its price, and not every earned reward gets redeemed.

The one rule: think in real cost, not menu price

The mistake that makes owners nervous about loyalty is pricing the reward at its menu value. A free $5 latte feels like giving away $5. It isn’t. It costs you what the latte costs you to make — beans, milk, a cup, a lid — which for most cafes is $1.20 to $1.80.

So the real question is never “can I afford to give away a $5 drink.” It’s “does the extra visits this reward drives more than cover the ~$1.50 of ingredients it costs me?” Framed that way, loyalty is one of the cheapest marketing levers you have — you’re paying wholesale to buy retail-priced repeat behaviour.

We go deeper on the full budget in what a coffee-shop loyalty program actually costs; here we’re focused on setting the earn rate itself.

Step 1: Pick your reward structure

Two structures dominate cafe loyalty. Pick based on how consistent your tickets are.

  • Per-visit / per-item (the digital punch card). “Buy 9 drinks, get the 10th free.” Simple, instantly understood, and ideal when most tickets are one or two drinks. This is the classic coffee model.
  • Points per dollar. “Earn 1 point per $1, redeem 50 points for a reward.” Better when ticket sizes vary a lot — big food orders, group orders, catering — because it ties the reward to spend rather than trips.

For a straight coffee shop, per-visit almost always wins on simplicity. For a cafe with a real food menu and wide ticket variance, points-per-dollar is fairer to you. We compare them head-to-head in punch card vs points for cafe loyalty.

Step 2: Set the earn threshold

Now the core number: how many paid visits (or dollars) before the reward unlocks?

The nominal discount of a “buy N, get 1 free” program is simply 1 ÷ (N + 1):

Reward thresholdNominal discountFeels like
Buy 5, get 6th free16.7%Very generous
Buy 8, get 9th free11.1%Generous
Buy 9, get 10th free10.0%Standard
Buy 11, get 12th free8.3%Conservative

“Nominal” is the discount on menu price. Your real cost is lower — see Step 3.

A threshold of 8 to 12 paid drinks is the sweet spot for most cafes: generous enough that a twice-a-week regular hits the reward roughly monthly (which keeps it motivating), conservative enough that you’re not discounting the whole business. Below 5 gets expensive fast; above 12 the reward feels too far away to change behaviour.

If you want to be more generous without raising the headline discount, do it with targeted boosts — a birthday reward, a double-points day on your slowest afternoon — rather than a richer base rate. That concentrates the giveaway where it actually moves a visit. See birthday rewards for coffee-shop loyalty.

Step 3: Convert nominal discount to real cost

This is where loyalty stops looking scary. Two adjustments turn that 10% nominal number into your true cost.

Adjustment 1 — food cost, not menu price. The free drink costs you its COGS. If a $5 latte costs $1.50 to make, the reward is $1.50, not $5.

Adjustment 2 — breakage. Not every earned reward gets redeemed. Customers forget, move away, or never quite finish the card. Across loyalty programs, unredeemed rewards (“breakage”) are commonly cited in the 20–40% range, though it varies widely. Every unredeemed reward is a giveaway you booked but never paid out.

Put them together for a “buy 9, get 10th free” program:

  • Revenue to earn one reward: 9 paid lattes × $5 = $45
  • Cost of the reward (food cost): $1.50
  • Nominal reward cost as % of that revenue: $1.50 ÷ $45 = 3.3%
  • Apply, say, 30% breakage → effective cost ≈ 2.3%

So a program that looks like a 10% discount actually costs you roughly 2–3.5% of the revenue it generates — and that’s before counting the extra visits it drives that wouldn’t have happened otherwise. Target a design whose real reward cost lands around 3 to 6% of earning revenue and you have wide margin to work with.

A free drink priced at menu value is a 10% discount. Priced at food cost, after breakage, it’s a 2–3% cost of acquiring repeat behaviour. Same program, completely different decision.

Step 4: Sanity-check against your margin

One guardrail: make sure the base product you’re rewarding still carries healthy margin, because the paid visits are what fund the free one. If your latte is already thin on margin, a rich reward can tip a customer’s blended margin negative. The fix is usually to reward the high-margin item (drip coffee, espresso drinks) rather than the low-margin one (a loss-leader food item), and to keep the threshold at 9+ rather than 5.

If you also run discounts and promos on top of loyalty, make sure they don’t stack into negative margin. Our guide to running cafe promotions without losing margin covers how to keep layered offers profitable.

Step 5: Measure the two numbers that matter

Once it’s live, ignore vanity metrics (sign-ups) and watch two things:

  1. Visit frequency of enrolled vs non-enrolled customers. The whole point is more trips. If members visit meaningfully more often than non-members, the program is working, and its real cost is easily covered by the extra visits.
  2. Redemption rate. Very low redemption (say under 10–15%) means the reward is too far away or people forgot they were enrolled — shorten the threshold or add reminders. Very high redemption with no lift in frequency means you’re discounting behaviour you’d have gotten anyway — lengthen the threshold.

This ties directly to customer lifetime value at a coffee shop: a good loyalty earn rate should lift frequency enough that each member’s lifetime value rises by far more than the reward costs.

A note on Square Loyalty and where an app helps

If you’re on Square, Square Loyalty lets you build exactly these structures — points per visit, per item, or per dollar, plus VIP tiers that multiply earning (1.5×, 2×, 3×) once a customer crosses a spend threshold, up to six tiers. As of 2026 Square simplified its loyalty pricing (bundled with the Square Plus plan at around $49/month per location in the US; confirm current pricing and your country). It’s a solid, native option covered in Square Loyalty for cafes and regulars.

The limit of a card-in-the-wallet or a POS-only program is reminding people. A reward no one remembers they’re close to doesn’t change behaviour. That’s where a branded app earns its place: the reward progress lives on the customer’s phone, and a push notification — “you’re one drink away from a free latte” — turns a dormant card into a visit. That’s the niche Tany fills: a branded order-ahead app plus web ordering on your existing Square POS, with self-running loyalty and push built in, live in about a day for $99 CAD/month per location.

You don’t need an app to set a smart earn rate — the math in this post works with any tool, including a paper punch card. But the earn rate only pays off if customers remember to chase it, and a reminder in their pocket is what closes that gap.

The takeaway

Don’t guess at generosity — calculate it. Pick per-visit for coffee or points-per-dollar for variable tickets, set the threshold at 8 to 12 paid drinks, then convert the nominal discount to real cost using your food cost and expected breakage. A standard “buy 9, get the 10th free” looks like 10% off but costs most cafes 2 to 3.5% of the revenue that earns it — cheap fuel for repeat visits. Set it there, measure frequency and redemption, and adjust.

Sources

Frequently asked questions

What is a good loyalty earn rate for a coffee shop?
A common and safe structure is a free drink after 8 to 12 paid drinks, which nominally discounts about 8 to 12 percent but costs far less in real margin because a coffee's food cost is low and not every earned reward is redeemed. Aim for a program whose real reward cost lands around 3 to 6 percent of the revenue required to earn it.
Should I give points per dollar or per visit?
Per visit (or per item) suits cafes with small, consistent tickets like coffee, because it rewards frequency and is easy for customers to understand. Points per dollar suits venues with variable ticket sizes, like restaurants or cafes with big group and food orders, because it ties the reward to how much a customer actually spends.
How much does a free-drink reward really cost my cafe?
Much less than the menu price. A $5 latte might cost you $1.20 to $1.80 in ingredients, so the free reward costs you that food cost, not $5. Spread across the 9 or 10 paid drinks required to earn it, plus the share of rewards customers never redeem, the effective cost is typically in the low single-digit percentages of revenue.
What percentage of loyalty rewards go unredeemed?
A meaningful share — often cited in the 20 to 40 percent range across loyalty programs, though it varies widely by design and audience. This 'breakage' means your real reward cost is lower than the nominal discount suggests. Don't design around breakage as a goal, but do account for it so you don't over-budget the program's cost.