Free resource · use it on this page
Loyalty Program Planning Worksheet
Define the counter before the reward: fill this in top to bottom and you have a launchable loyalty program on one page.
How to use this worksheet
A loyalty program is two designs, not one: the counter (how a visit reliably becomes a recorded credit) and the reward (what the credits eventually buy). Most first programs are designed in the wrong order, reward first and counter never, which is why they die at the till rather than in the spreadsheet. This worksheet runs in the honest order: counter workflow first, then the reward math, then expiry and caps, then how people join, then the one number you will judge it by.
Work top to bottom in one sitting. Every answer should fit in a line or two; if a question takes more than a minute, that is the worksheet doing its job, because it just found a decision you had not actually made. Print the page or fill it in on screen, and keep the finished copy where whoever runs the launch can see it.
Part 1: The counter workflow (do this first)
Before you decide what customers earn, decide how earning physically happens. If any answer here is vague, the reward section does not matter yet.
System of record
Name the one place the count lives: a POS customer field, a loyalty platform, a wallet pass balance, or a spreadsheet. One system only. Two counts always drift, and the customer always believes the higher one.
Who increments it
The cashier keys it in, the customer scans a code, or it happens automatically on payment. If the answer involves a person remembering, write down what reminds them.
The exact moment it happens
Before payment, after payment, or at the receipt screen. Name the till step. Vague timing is how stamps get forgotten during a rush.
Seconds added per transaction
Time it at your busiest hour, not your quietest. Any step slow enough to hold up a queue will quietly stop being performed within a month.
Identification method
How the customer proves who they are: phone number spoken aloud, a barcode or QR on a wallet pass, a physical card, an app scan. Pick exactly one; a second method doubles training and halves consistency.
The failure fallback
What the cashier does when the scan fails or the system is down: a paper log entered later, a manager code, or apologise and skip. Decide it now, in writing, not mid-rush.
Part 2: Mechanic, threshold, reward
Now that credits can reliably exist, decide what they are and what they buy.
Mechanic
Stamps per visit, points per amount spent, visit count, or spend tiers. Stamps are the easiest to explain and to run; points give finer control but demand arithmetic at the till. Pick one and write it down.
Earn rule in one spoken sentence
Write it the way a cashier would say it: one stamp per visit with any drink purchase, one per day. If the rule needs a second sentence, simplify the rule, not the sentence.
Threshold
How much earning unlocks the reward, for example 9 stamps. Test it against a real regular: at their actual visit rate, how many weeks to the first reward? If the answer is months, shrink the threshold.
The reward, named exactly
An exact item, not a category: any medium hot drink, not a free treat. Beside it, write what it costs you in ingredients or margin, not its menu price.
Effective give-back rate
Reward cost divided by the spend required to earn it. Compute it before launch; this is the discount you are silently committing to give your most loyal customers, indefinitely, so it must be a number you can defend.
First-stamp policy
Do new members start at zero or with a visible head start? Granting the first stamp on enrolment is a standard nudge toward a second visit. Either answer is fine; decide it deliberately and apply it to everyone.
Part 3: Expiry and caps
These are the rules nobody reads until there is a dispute, which is exactly why they must be written before launch.
Credit expiry
How long stamps or points survive without a visit, for example 6 or 12 months of inactivity. Never is a valid answer, but it means balances accumulate as a standing obligation; choose it on purpose, not by omission.
Reward expiry after it is earned
Once unlocked, how long the customer has to redeem, for example 30 or 60 days. Write the number and where the customer sees it.
Earn cap per visit or per day
Usually one stamp per visit regardless of basket size; without a cap, one large order becomes a finished card. Points programs cap per transaction or per day instead. Write the cap and the reason.
Redemption rules
Excluded items, minimum spend, whether the reward stacks with other discounts, and whether the free item itself earns a stamp. Every rule you leave unwritten becomes an argument at the till.
Staff and edge cases
Do employees earn? Refunded orders? Split bills? Purchases made as gifts? One line each; these are the loopholes someone will find in week one.
Pause and change policy
How you would raise the threshold or end the program without confiscating earned balances. Honouring existing progress is the difference between a change and a betrayal.
Part 4: Enrolment channels
A program nobody joins is a rules document. Go down the table, fill in an owner for every channel you intend to use, and cross out the ones you do not. An unowned channel is a decoration.
| Channel | How it works | Prepare before launch | Owner and launch-day status (fill in) |
|---|---|---|---|
| The ask at the till | The cashier offers enrolment at payment. For most shops this is the highest-volume channel by far. | A one-sentence script, rehearsed out loud: Want your first stamp? It takes ten seconds. | |
| QR code at the counter | The customer self-enrols while waiting; costs no staff time. | A printed stand or sticker at eye height, placed where the queue actually forms. | |
| Receipt or bag insert | Reaches customers after they have left, when there is no queue pressure. | A printed slip or receipt footer carrying the join link. | |
| Wallet pass install link | The customer taps a link and the loyalty card lands in Apple or Google Wallet. No app to download, and the pass doubles as their scan code at the till. This is the model wallet platforms such as PushNotice are built around. | A working install link, tested on one iPhone and one Android phone before day one. | |
| Existing customer list | An email or SMS announcement to people who already know you; usually the fastest first hundred members. | One announcement message, checked against the permissions those contacts actually gave you. | |
| Paper card conversion | Existing paper-card holders trade their progress into the new program instead of losing it. | A written trade-in rule, for example: full card equals one free reward plus three starting stamps, and a dated trade-in window. | |
| Website and social bios | Low volume but always on; catches people researching you before a first visit. | The join link placed in the bio and the site footer. |
Part 5: The single KPI
One number, chosen before launch. More than one KPI on a first program means none of them gets acted on.
The one number
Pick exactly one, for example the share of new members who make a second visit within 30 days of joining. Signups alone is a poor choice; it measures enrolment effort, not loyalty.
Baseline before launch
Write down today's value, dated, before the program starts. If you cannot measure your candidate KPI today, pick one you can; an unmeasurable KPI is a wish, not a metric.
Target and review dates
A number and two dates, for example a first review at day 30 and a judgment at day 90. Loyalty moves slowly; judging the program in week one measures the launch, not the loyalty.
Where the number is read, and by whom
Name the report or screen and the one person who checks it weekly. A KPI with no owner and no cadence stops being read by week three.
The pre-committed response if it is flat
Decide now what you change first: usually the till ask or the threshold, not the reward. First programs rarely fail because the reward is too small; they fail because nobody is being asked to join.
Before you launch: the sanity check
Run this against the completed worksheet. Every unchecked box is a launch risk you are choosing to accept, so accept it in writing or fix it.
Part 1 was completed before Part 2
If you skipped ahead to the reward, go back. The counter workflow is where programs actually fail.
The till test passes
A staff member who was not in the planning meeting can complete the earn flow without asking a single question.
The failure test passes
Unplug the scanner or kill the connection and run a transaction. The fallback you wrote in Part 1 actually happens.
The threshold survives the regular test
A genuine regular reaches the first reward inside their existing habit, without changing how often they visit.
There is one identification method, not two
The fallback is a written process, not a second method. Backup methods double training and produce inconsistent counts.
The earn rule fits in one spoken sentence
And that sentence, verbatim, is what staff say. If staff paraphrase it differently, the rule is too complicated.
Caps close the split-order exploit
Ringing one order as three separate transactions must not produce three stamps.
Expiry rules are visible to customers before they matter
On the pass, the card, or the terms page. An expiry customers first learn about at forfeit costs more goodwill than it saves.
Every live enrolment channel has a named owner
Check the table in Part 4. A channel with a blank owner column is crossed out, not launched.
The KPI baseline is written down and dated
Recorded before day one. You cannot show improvement against a number you never captured.
Staff know the why, not just the how
One line on what the program is for. Staff who privately think it is pointless will quietly stop making the ask.
The pause plan exists
You can stop or change the program while honouring every earned balance, and you know who makes that call.
From the guide: POS Loyalty Program: Add Loyalty Without Switching POS
This resource accompanies the full article — worth reading before you commit to a tool.