Free resource · use it on this page

Loyalty ROI Calculator

Check that a loyalty program pays for itself before you commit to it, using an incremental-profit-versus-program-cost model you can audit line by line.

Incremental revenue / month

$1,080

Incremental profit / month

$648

Return per $1 of program cost

22.3×

Break-even uplift for these numbers: 0.7% — if you believe the program moves visits less than that, it does not pay for itself yet. Incremental means vs. your pre-program baseline, not total member revenue.

What this calculator actually measures

The calculator above answers one question: does the extra gross profit your loyalty program generates exceed everything the program costs to run? The left side of that comparison is incremental customer lifetime value, the additional profit from visits the program causes. The right side is total program cost: rewards, software, and the operational overhead that rarely makes it into the pitch deck.

Incremental is the load-bearing word. Revenue from visits that would have happened anyway is not a return on the program; it is the same revenue with a stamp on it. Worse, you pay reward costs on those visits too, because members earn stamps on every purchase, not just the extra ones. A model that counts all member revenue as program revenue will approve almost any program. This one deliberately does not.

A good result here is not a spectacular ROI multiple. It is a net monthly return that stays positive when you feed in conservative assumptions. If the program only works when every input is optimistic, the calculator has done its job by telling you that before you upgraded to a paid plan.

The inputs, defined

Each field in the calculator maps to one of these. Where the number lives in your business is noted for each. If you are unsure of a value, enter the pessimistic end of your range first.

Active members

Customers actually holding your card or pass and still visiting, not total signups ever. A member who has not been seen in about 90 days should not count.

Average spend per visit

Total revenue divided by transaction count from your till or POS. Use the everyday average, not your best ticket. If you can filter to loyalty customers, use that figure.

Gross margin

What remains after cost of goods, as a percentage of price. A 4.00 coffee that costs 1.20 in ingredients and cup carries a 70 percent gross margin.

Baseline visits per member per month

How often a typical customer visited before the program existed, or how often comparable non-members visit now. This is the number the lift applies to, so do not inflate it.

Expected visit lift

The extra visits you attribute to the program, as a percentage of baseline. This is the assumption the whole model hinges on. Enter it low, then test a range rather than a single guess.

Purchases to earn a reward

The threshold of your offer, for example buy nine get one free means nine. A lower threshold pays out faster and costs more per visit.

Reward cost to you

Your cost of fulfilling the reward, not its menu price. A free 4.00 drink costs you roughly its ingredients, around 1.20 in the example above. Using menu price here overstates cost.

Redemption rate

The share of earned rewards customers actually claim. Unclaimed rewards, called breakage, cost you nothing but also build no goodwill. If unsure, model 100 percent as the worst case for cost.

Software cost per month

The plan tier you would genuinely need at your member count, not the entry tier. Include any per-message fees your channel charges on top of the subscription.

Other monthly costs

Staff time explaining and stamping, printed cards or signage, and a monthly slice of any one-off setup or design work. A rough monthly average is fine.

The formula, line by line

The calculator runs these lines in order. Nothing else happens under the hood, so you can reproduce any result on paper.

LineHow it is computedWhy it matters
Incremental visits per memberBaseline visits multiplied by expected visit liftThe extra visits the program causes. Every downstream benefit flows from this line and nowhere else.
Incremental revenue per memberIncremental visits multiplied by average spend per visitThe top-line gain, before margin. On its own this number flatters the program.
Incremental gross profit per memberIncremental revenue multiplied by gross marginRewards are funded from margin, not revenue, so this is the real benefit line.
Reward cost per visitReward cost divided by purchases to earn, multiplied by redemption rateEach stamp is a small installment toward a future free item. This spreads that liability across visits.
Monthly reward cost per memberBaseline plus incremental visits, multiplied by reward cost per visitRewards accrue on every visit, including the ones you would have received anyway. Skipping this is the most common way ROI gets overstated.
Net return per member per monthIncremental gross profit minus monthly reward costWhether one member is worth having in the program at all.
Program net per monthNet return per member multiplied by active members, minus software and other monthly costsThe bottom line. Fixed costs mean small programs need proportionally more lift to clear zero.
ROI multipleTotal incremental gross profit divided by total program costAbove 1.0 the program pays for itself. Below 1.0 it is a customer subsidy, which can be a deliberate choice but should be a knowing one.
Break-even liftThe visit lift at which program net is exactly zeroThe single most useful output. Instead of guessing your lift, ask whether the break-even lift sounds achievable for your shop.

How to read your result

  1. 1

    Check the sign before the size

    A positive program net per month under your conservative inputs is the pass mark. Ignore the temptation to tune inputs until the multiple looks impressive; the sign is the decision, the size is detail.

  2. 2

    Interrogate the break-even lift

    If the program breaks even at a very small lift, your downside is limited even if your forecast is wrong. If it needs a dramatic behavior change from every member to break even, treat the plan as speculative.

  3. 3

    Stress-test with two changes

    Halve your expected lift and set redemption to 100 percent, then look again. A program that survives both is robust. A program that flips negative on either is running on hope.

  4. 4

    Watch the reward-on-baseline line

    If reward costs on visits you already had swamp the incremental profit, the reward threshold is too generous for your visit frequency. Fix the threshold before blaming the software cost.

  5. 5

    Compare payback to commitment

    If moving to a paid plan involves an annual contract or setup work, check how many months of program net it takes to cover that. A payback longer than the commitment period means you are betting, not investing.

  6. 6

    Decide, and write the assumption down

    Record the lift figure you used to justify the decision. In 60 to 90 days, measure the real difference between member and non-member visit frequency and re-run the calculator with that number instead.

Costs and leaks the simple model misses

Before trusting a positive result, walk this list. Each item either belongs in your cost inputs or should temper your lift assumption.

  • Reward valued at your cost, not menu price

    Entering menu price overstates cost and can wrongly kill a viable program. Use cost of goods for the reward item.

  • Rewards earned on baseline visits

    Members collect stamps on visits they would have made anyway. The calculator includes this; your mental math probably does not.

  • Software tier jumps

    Many platforms price by member count or message volume. Cost the tier you will need in six months, not the one you qualify for today.

  • Per-message channel fees

    SMS and some app platforms bill per message sent, so reminder costs scale with your list. Wallet pass updates, the channel PushNotice uses, carry no per-message fee, which changes the cost line as you grow.

  • Seconds at the till

    Stamping, scanning, or looking up members takes staff time on every transaction. Small per-visit, large across a month of transactions.

  • Printing and replacement

    Physical stamp cards need printing, and lost cards mean either re-issued stamps or annoyed customers. Digital passes shift this cost but do not always remove setup effort.

  • Discount stacking

    If rewards can combine with other promotions, the effective giveaway per redemption is larger than the reward cost you entered.

  • Self-stamping and fraud

    Honor-system cards leak. If staff can stamp friends or customers can stamp themselves, your real redemption cost runs above the model.

  • Redemption instead of purchase

    Some rewards replace a purchase the customer was about to make rather than adding a visit. This cannibalization quietly shrinks your true lift.

  • Setup and design time

    Configuring software, designing the card or pass, and training staff are real hours. Amortize them into other monthly costs rather than pretending they were free.

  • The frequent-visitor subsidy

    Your most loyal customers earn rewards fastest while needing incentive least. If most redemptions come from people already visiting at ceiling, incremental lift is lower than average behavior suggests.

  • Outstanding liability at switchover

    If you later change systems or thresholds, customers hold earned-but-unredeemed progress. Honoring it costs money; voiding it costs trust.

Where this model bends

The lift is assumed, not measured. This calculator cannot tell you how much extra visiting your program will cause; it can only tell you what a given lift is worth. There is also a selection effect to respect: members often visit more than non-members partly because your most frequent customers are the ones who join, not because the program changed them. Comparing member and non-member frequency overstates lift for exactly this reason, so treat that comparison as a ceiling, not an answer.

Run ranges, not points. A single set of inputs produces a single, false-precision answer. Running the calculator three times, with pessimistic, expected, and optimistic lift, tells you something more honest: whether the decision changes across the plausible range. If all three runs agree, decide with confidence. If they disagree, the program design needs work before the software choice matters.

Mind the time horizon. Program costs start on day one; habit change arrives over months. A program that looks negative in its first six weeks may simply be early, and one that looks positive because of a launch-promotion spike may simply be discounting. Judge the model over a quarter or more, and re-run it with measured numbers once you have them. The calculator is a pre-commitment check, not a substitute for watching what your customers actually do.

From the guide: Free Loyalty Software: The 2026 Guide

This resource accompanies the full article — worth reading before you commit to a tool.