Free resource · use it on this page
Reward Threshold Calculator
Work out how many visits your reward should take, what each redemption really costs you, and how much liability your program is building up.
Effective discount rate
5.6%
Gross profit per completed card
$61
Reward liability per 100 members
$420
Effective discount = reward value ÷ (visits-to-earn × average spend). Most sustainable stamp programs land between 5% and 12%; above that, check the margin column before launching. Unredeemed rewards (breakage) reduce real cost — the liability figure assumes your expected redemption rate.
What this calculator tells you
Every loyalty program has one number that decides whether it works: the threshold. Set it too high and customers stop counting; the card sits at three stamps out of twelve forever. Set it too low and you give away margin faster than the program earns it back. The calculator above works both problems at once: it tells you how long a typical customer will take to reach the reward, and what that reward will actually cost you once you account for how many people redeem.
It also answers a question most small programs skip until it hurts: liability. Every stamp you issue is a small promise. The calculator forecasts how many earned rewards will be outstanding at any time and what they would cost to honor, so the free-coffee line item never surprises you.
The sections below define each input, show the formulas the calculator runs, and walk through how to read the results. You can work the same math by hand with the formula table if you prefer paper.
The inputs, defined
The calculator asks for nine numbers. Most come straight from your till data; two are assumptions you will replace with real data after launch.
Average spend per visit
Your real average qualifying ticket, not your best customer's. This converts visits into revenue.
Typical visits per month
How often a regular actually comes in. Be honest here; this single input drives time-to-reward.
Stamps or points required (the threshold)
How much progress earns the reward. This is the number the whole page exists to help you set.
Stamps or points earned per visit
Usually one stamp per qualifying visit, or a points rate per unit of spend. Keep the earn rule simple enough to say in one sentence.
Reward retail value
What the reward would cost the customer if they paid for it. This is the value they perceive, not what it costs you.
Reward type: free product or money off
They cost you differently. A free product costs you its cost of goods; a discount costs you the full face amount.
Gross margin on the reward item
Used to convert a free product's retail value into its true cost to you. Ignored for money-off rewards.
Expected redemption rate
The share of earned rewards you expect to be claimed. If you are unsure, budget with a high rate; the unclaimed remainder is breakage.
Active members
How many customers are currently earning. Used to forecast monthly redemption volume and outstanding liability.
The formulas, in plain language
Everything the calculator outputs comes from these eight lines. No hidden weighting, no proprietary index.
| Output | How it is calculated | What it tells you |
|---|---|---|
| Visits to reward | Stamps required divided by stamps earned per visit | How many qualifying visits a customer needs to earn one reward |
| Spend to reward | Visits to reward multiplied by average spend per visit | The revenue you collect before you owe one reward |
| Face-value reward rate | Reward retail value divided by spend to reward | The generosity the customer perceives, as a share of what they spent |
| Cost per reward | For free product: retail value times (100 percent minus gross margin). For money off: the full face amount | What one redemption actually costs you |
| Redemption-adjusted cost rate | Cost per reward times redemption rate, divided by spend to reward | Your true program cost as a share of member revenue, after breakage |
| Time to reward | Visits to reward divided by typical visits per month | How many months a typical customer waits for their first reward |
| Rewards earned per month | Active members times visits per month, divided by visits to reward | The redemption volume your counter staff should expect |
| Outstanding liability | Earned but unredeemed rewards, times redemption rate, times cost per reward | The probability-weighted cost of every promise currently in circulation |
How to read your results
Face-value rate versus cost rate. These two numbers are deliberately different. A buy-nine-get-the-tenth-free coffee card returns roughly one visit in ten as free product, a face-value rate around ten percent, which sounds generous out loud. But the drink costs you its ingredients, not its menu price, and not every earned reward gets claimed, so the redemption-adjusted cost rate is usually a fraction of the face-value rate. That gap is the whole trick of free-product rewards: high perceived generosity, modest real cost. Money-off rewards have no gap, which is why they need more caution.
Time to reward is the reachability test. There is no universal correct number, but there is a useful thought experiment: say the timeline out loud as a customer would experience it. If a genuine regular needs a quarter of a year to see their first reward, the program is competing with their memory, and losing. Shorten the threshold, increase the earn rate, or pick a cheaper, closer reward.
Liability is a forecast, not a scare. Outstanding liability grows in the early months of any program because earning starts before redeeming does. That is normal. What matters is that the steady-state number fits comfortably inside your margin and that you tracked it on purpose rather than discovering it.
One direction-of-error rule: when you budget, err toward a high redemption rate, and when you forecast the customer experience, err toward low visit frequency. Both errors make the model stricter than reality, which is the side you want to be wrong on.
A sensible order of operations
- 1
Pick the reward before the threshold
Choose something with high perceived value and low marginal cost to you, usually a product you make yourself. The reward decides whether anyone cares; the threshold only decides when.
- 2
Set the threshold from real visit frequency
Decide how soon a typical regular should earn their first reward, then multiply their honest monthly visits by that timeline. Work backward from the customer's calendar, not forward from a round number.
- 3
Check both cost rates
The face-value rate tells you whether the offer sounds generous; the redemption-adjusted cost rate tells you what it does to your margin. You need both to be acceptable, and they answer different questions.
- 4
Stress-test at full redemption
Rerun the numbers with a one hundred percent redemption rate. If the program only works because you assume many people forget to claim, the threshold is wrong, not the customers.
- 5
Write the liability and expiry rules before launch
Decide now how long earned rewards stay valid, how you track earned-minus-redeemed each month, and who watches the number. Changing expiry rules after launch reads as a broken promise.
Pre-launch sanity checks
Run this list once your numbers look right. Each item is a way real programs quietly fail.
The threshold fits the customer's existing habit
A regular should reach the reward without changing their behavior. Rewards that require extra visits to become reachable feel like homework.
The first reward arrives within a memorable timeframe
Check your time-to-reward output against how long your customers plausibly keep one merchant's progress in mind.
The offer passes the say-it-out-loud test
Read the earn rule and reward to an imaginary customer in one sentence. If it needs a second sentence, simplify it.
Cost per reward uses cost of goods, not retail price
Pricing a free product reward at its menu price overstates your cost and pushes you toward stingier, weaker thresholds.
The program survives one hundred percent redemption
Breakage should be a bonus, never the reason the math works.
Redemption takes seconds at any till
If claiming the reward needs a manager, a code lookup, or an apology, redemption friction will do your breakage for you and cost you the goodwill the program was meant to buy.
The earn rule has no fine print
Exclusions, blackout days, and minimum-spend asterisks each cost more trust than they save in margin.
One earn action per transaction is enforced
A simple fraud guardrail: one stamp or points credit per receipt, however large the order.
A head start is considered
Issuing the first stamp at signup starts the customer with visible progress, and progress already underway is easier to continue than progress not yet begun.
Expiry and dormancy rules are written and disclosed
Decide validity windows before launch and state them where customers enroll. Silent expiry is the fastest way to turn a fan into a critic.
Liability is tracked monthly as earned minus redeemed
One number in a spreadsheet, updated monthly, compared against the calculator's forecast. Drift in either direction is information.
A ninety-day review is on the calendar
By then you have real redemption and frequency data. Replace every assumption in the calculator with observed numbers and rerun it.
Caveats and limits
This is a steady-state model. It does not capture launch spikes, seasonality, or the behavior change the program itself may cause; regulars who start visiting more often will reach the reward faster than the inputs suggest. Rerun the numbers once early data arrives rather than trusting the first pass for a year.
Threshold math is not ROI math. This tool prices the reward and forecasts what you will owe; it deliberately says nothing about whether the program generates enough incremental revenue to pay for itself. That is a separate question with its own model, covered by the companion loyalty ROI calculator in this resource library.
On accounting: earned, unredeemed rewards are a genuine liability, and how breakage is recognized in your books varies by jurisdiction and accounting standard. Use the liability output for operational planning and talk to your accountant about how to carry it formally. Nothing here is accounting advice.
Finally, calibrate your redemption assumption to your medium. Paper stamp cards lose members to lost cards and forgotten wallets; a card that lives in Apple or Google Wallet, the model PushNotice is built on, removes that failure mode entirely. If you are moving from paper to digital, assume meaningfully higher redemption than your paper-card history suggests, and let the ninety-day review tell you the real number.
From the guide: Free Loyalty Software: The 2026 Guide
This resource accompanies the full article — worth reading before you commit to a tool.