Set up your loyalty program without destroying your margin

An attractive reward must also be payable

A loyalty program can increase visit frequency and at the same time become a silent drain on margin. This occurs when the restaurant chooses a reward based on intuition, calculates its cost at the wrong price, or rewards customers who would have returned anyway without checking whether the benefit is generating additional business.

The problem is not solved by offering something cheap. A low-value reward can cost almost nothing and still waste team effort because no one will have any interest in pursuing it. Nor is it resolved by setting the goal too far away: if the benefit seems unattainable, the program stops influencing the next visit.

The sustainable setup is in the middle of those two fault lines. The customer must perceive a clear benefit, the team must be able to deliver it without complicating the service, and the entire cost of the program must fit within the margin that the restaurant is willing to invest to generate recurrence.

In this guide you will learn how to:

• calculate the economic cost of a reward, not just its prescription cost;
• decide how many visits to request before delivering a benefit;
• evaluate programs by visits and level programs;
• estimate how much additional activity the program needs to produce to pay for itself;
• configure relevant rules in Fudi Rewards;
• review results and correct the program before the cost becomes permanent.

Start with the behavior you want to change

Before choosing a drink, dessert, or discount, define what the customer should do differently thanks to the program. The reward is not the goal; It is the cost you agree to pay to produce a behavior.

A restaurant can search, for example:

• an occasional customer adds a visit during the month;
• that a frequent customer chooses the restaurant instead of a nearby alternative;
• that a long-term relationship moves towards level benefits;
• that diners maintain a constant frequency after receiving their first reward.

The distinction matters because two programs with the same reward can have very different economics. If the majority of customers already visit the restaurant each week, providing a benefit for a behavior that occurred without the program can reduce margin without changing frequency. On the other hand, if the program gets some of those clients to move forward or add a visit, there is an incremental contribution that can pay for the benefit.

That's why it's a good idea to write a simple hypothesis before configuring anything:

We hope that this program will ensure that participating customers make an additional visit within a normal purchasing period, without significantly reducing the average ticket or displacing sales that were already occurring.

The hypothesis will change depending on the restaurant, but it should be testable. “We want to pamper our customers” may be a valid intention, although it does not serve on its own to establish a budget, a visit goal, or a success criterion.

Protect the margin with the correct measurement

The selling price of a reward is not its cost. If a dessert sells for $120, delivering it doesn't necessarily cost the restaurant $120. The direct cost can be $32 between ingredients, packaging and consumables. However, it would also not be correct to conclude that the benefit costs only $32.

To make a useful decision you need to separate three concepts.

Contribution margin per visit

It is the part of the ticket that remains after subtracting the costs that change with that sale. A practical version of the account is:

Contribution margin per visit = net sale - variable costs of the visit

Use sales net of taxes and discounts. Variable costs may include ingredients, packaging, collection fees, third-party commissions, and any labor costs that increase directly from producing or delivering that sale. Rent and other fixed costs do not disappear, but they do not usually change for delivering an additional reward; That is why they are not the best basis for deciding how much each exchange costs.

If the average net ticket is $240 and the associated variable costs are $132, the average contribution margin per visit is $108. That figure helps estimate how many additional visits are needed to recover the cost of the program.

Economic cost of the reward

It is the cost that really causes the exchange. May include:

• ingredients, packaging and consumables;
• additional work that only occurs when preparing the benefit;
• expected shrink;
• replacement cost when the benefit replaces something the customer would normally have purchased;
• capacity cost when the reward complicates the operation in times of high demand.

A drink that you would almost never have bought can cost almost the same as producing it. The same drink, delivered to a customer who always buys it, also eliminates the margin on that sale. In that second case, using only the prescription cost underestimates the impact.

Total program cost

In addition to the redeemed rewards, consider the proportional part of the platform, materials, training, operational time, incidents and any variable communication or administration costs. Not all of these expenses should be allocated to each redemption, but they should appear when you calculate whether the entire program pays for itself.

Correct accounting does not seek perfect accounting precision. It seeks to avoid a decision built with an incomplete cost.

##Distinguish between low-cost rewards and low-risk rewards

A cheap reward is not always safe. The risk depends on how it affects purchasing behavior and operation.

Additional product with little substitution

It could be a side dish, a simple drink, a small dessert, or an upgrade that the customer would not normally have purchased. It usually combines a perceived value greater than its direct cost and a controllable impact on the ticket.

It works best when:

• its preparation is stable;
• there is sufficient availability;
• does not displace a regular sale;
• can be delivered in the same way at all participating branches.

Product that replaces a frequent purchase

“Your usual free coffee” is easy to understand, but it can eliminate a sale that was already happening. The economic cost is closer to the lost margin than to the cost of the beans and the glass.

It doesn't mean it should be discarded. It can work when the additional visit brings other consumption, when the customer rarely buys only that product or when the frequency generated compensates for the substitution. The difference is that it must be modeled with that risk included.

Discount on the ticket

A percentage seems simple, but it grows with consumption and directly reduces the contribution of each visit. A 10% discount on a $300 bill costs $30; on a bill of $1,200 it costs $120. If the program is based on visits and tickets vary greatly, the restaurant may end up providing a very different benefit to customers who made the same progress.

Permanent tier discounts require even more care because their cost doesn't end with a single redemption. Each future customer visit can continue to reduce the margin.

Operating or experience benefit

A customization, special access, or upgrade may have a low direct cost, but it is not free. They must be checked against the actual capacity of the restaurant. A benefit that seems cheap in the office can be difficult to fulfill on a Friday night and end up causing delays, exceptions or inconvenience for the team.

The best reward is not the lowest prescription cost. It is what maintains a favorable difference between perceived value and economic cost, without creating a promise that the operation cannot sustain.

Model the complete cycle before setting up a visit program

In Fudi you can create a program based on visits and place rewards at different milestones. To calculate it, don't just review the final prize. Add up all the benefits that a customer can obtain by completing the cycle.

Follow this process.

1. Calculate the economics of an average visit

You need, at least:

• average net receipt;
• average variable cost;
• contribution margin per visit;
• usual frequency of the clients you want to attract.

When there are relevant differences between branches, make the account separately. The same dish can have different costs, losses, and sales mix depending on the location.

2. Calculate the economic cost of each reward

Don't use the menu price as a cost, but don't limit yourself to the recipe either. For each benefit write down:

• direct cost;
• additional preparation or packaging cost;
• probability of substituting a purchase;
• delivery difficulty during times of high demand;
• cost variation between branches.

If you can't estimate the substitution yet, prepare two scenarios: one in which the benefit is completely additional and another in which it replaces a regular purchase. The difference will show how sensitive the program is to that assumption.

3. Add up the cost of the entire cycle

The basic account is:

Cycle cost = sum of the economic cost of all rewards in the cycle

Then compare that cost with the contribution margin accumulated during the required visits:

Accumulated cycle margin = contribution margin per visit × visits in the cycle

Program burden on margin = cycle cost ÷ accumulated cycle margin

There is no universal percentage that is correct for all restaurants. Instead of copying a figure, prepare scenarios with different internal limits and review how much additional behavior each would need to produce. The limit of a high-frequency cafe does not have to be the same as that of a special occasion restaurant.

4. Take a trial with 100% redemption

You can also prepare an expected scenario with a lower redemption rate, but the setup must survive a high participation scenario. Relying on many customers forgetting or letting their rewards expire creates a fragile economy: If the program improves and more people redeem, the supposed savings disappear.

An illustrative example

Let's suppose a cafeteria with these data:

• average net ticket: $180;
• average variable cost per visit: $76;
• contribution margin per visit: $104;
• cycle of 12 visits;
• drink on visit 4 with an economical cost of $27;
• dessert on visit 8 with an economical cost of $34;
• dish selected on visit 12 with an economical cost of $78.

The full cost of the cycle would be $139. The accumulated contribution margin over 12 visits would be $1,248. In a full redemption scenario, rewards would consume approximately 11% of that accumulated margin.

That figure does not prove that the program is profitable. It only indicates that the promise fits within the economics of the cycle under the assumptions used. It remains to be seen whether the rewards generate enough additional visits to recover those $139 and the other costs of the program.

All amounts in this example are illustrative. They must be replaced by actual restaurant costs and tickets.

Don't confuse a payable reward with a profitable program

A program can be payable and still produce no return. The reason is simple: visits that occur during the cycle are not necessarily caused by the program.

To evaluate the return, use the incremental margin, that is, the contribution generated by visits or consumption that would not have occurred in the absence of the program. The logic is:

Incremental result = additional margin generated - total program cost

And a useful way to calculate the break-even point is:

Additional visits required = total program cost ÷ contribution margin per additional visit

Let's assume that during a month the restaurant incurs $4,500 between redeemed rewards and costs allocated to the program. If one additional visit leaves $110 in contribution, it takes approximately 41 truly incremental visits to cover that cost.

The word “truly” avoids a misleading reading. Recording 300 visits on Fudi does not mean that all 300 are new. Part of them may belong to customers who already visited the restaurant. That is why it is convenient to compare periods, branches or similar groups and cross Fudi's activity with sales and costs of the point of sale system or accounting.

A BCG analysis of the economics of loyalty programs shows why a cost that seems small as a percentage of sales can absorb almost all of the incremental profit when the margin is limited. The lesson is not to copy its percentage, but to require the change in behavior to pay for the program. See the reasoning in Leveraging the Loyalty Margin.

Decide how many visits to ask for without pushing the reward too far

Asking for more visits reduces the reward cost per visit, but may also reduce the number of customers who perceive the benefit as achievable. A stretch goal does not protect margin when no one changes their behavior to complete it.

Research on the goal proximity effect has observed that people may speed up their behavior as they get closer to a reward. In a field study with a coffee program, intervals between purchases decreased as customers moved toward the reward. The full study is available in The Goal-Gradient Hypothesis Resurrected.

To put this idea into a practical decision, estimate how long it takes a repeat customer to complete the cycle:

Expected completion time = required visits ÷ usual visit frequency

A client who visits twice a week may perceive a reward on the eighth visit as close. For those who visit once a month, the same goal implies eight months. The setup is identical, but the experience is not.

Before setting the final number, answer:

• How long would a frequent customer, an average customer and an occasional customer take?
• Does the first reward come early or is all the value concentrated at the end?
• Is the cumulative cost of intermediate rewards still sustainable?
• Does the customer easily understand what they get and when?
• Can the restaurant keep the promise throughout the cycle?

When the frequency is high, a longer cycle with intermediate milestones may operate. When visiting is infrequent, a closer initial reward can help the program make sense before asking for a long-term relationship. In both cases, it models the cumulative cost, not just the distance to the first benefit.

Fudi shows the diner their progress and the next reward, while the program report allows you to see how many people are on each visit and how many are close to obtaining a benefit. This distribution helps detect a goal that is too far away: if the majority concentrates at the beginning and almost no one advances, the problem may be distance, the value proposition, the natural frequency of the restaurant or operational adoption.

Be more careful with permanent level benefits

Visit-based tier programs allow you to recognize longer relationships and configure requirements and benefits by tier. The main economic risk appears when the benefit is maintained for each future visit, especially if it is a discount.

To model it, don't calculate a single swap. Choose a review horizon, for example three or six months, and estimate:

Tier cost for the period = eligible customers × expected visits × benefit cost per visit

Imagine that 80 customers reach a tier with a 10% discount, visit twice a month, and have an average net ticket of $500. The discount would reduce the contribution by $8,000 per month:

80 × 2 × $500 × 10% = $8,000

If the pre-discount contribution margin was $175 per visit, those customers would generate $28,000 in contribution before the benefit. The discount would absorb about 29% of that amount. To justify it, the level would have to produce a relevant and demonstrable increase in frequency, permanence or consumption.

Tiers can be more sustainable when they combine recognition with cost-controlled benefits. Some options are a specific reward upon reaching the level, a limited cost upgrade, or an operating privilege that the restaurant can consistently fulfill. Any non-automated benefits must be clearly described and understood by the team.

Avoid accumulating several permanent benefits without calculating their joint cost. A discount, a drink and an upgrade may seem modest separately, but together they apply to the most frequent customers, precisely those who can use them the most.

Take the decision to Fudi with rules that the team can operate

Once the account is finished, the configuration in Fudi should reflect the exact promise you modeled. Don't use a broad description on the platform and a different interpretation on the box.

In a visit program

You can define the visits that provide rewards, the validity and what happens when the diner completes the cycle. Before activating it:

• place each reward on the milestone used in your calculation;
• describes the product, size or benefit without ambiguity;
• check the accumulated cost of all rewards;
• decides whether progress restarts upon completion of the cycle or ends;
• confirms that inventory and preparation are feasible at participating locations.

If the cycle restarts, the economy repeats itself. A one-time sustainable program may no longer be sustainable when repeat customers complete several cycles per year. Project at least how many cycles a high-frequency customer could complete.

In a level program

Define the requirements, benefits and rewards of each stage. For each level, record internally:

• how many customers you expect to arrive;
• how many visits they could make while keeping the benefit;
• how much the benefit costs on a typical visit and on a high ticket visit; *how much of the cost depends on manual operation.

Fudi allows you to consult an explanation of how the program works and preview the card that the diner will see. Take advantage of this review to verify that the benefit understood by the customer coincides with the profit budgeted by the restaurant.

Rules and validity

The diner can consult terms, validity, rewards and exchange rules. Write conditions that staff can explain in one sentence. At a minimum, clarify:

• what product or benefit is delivered;
• if there are specific options or substitutions;
• where it can be redeemed;
• what happens when the product is not available;
• what is the validity;
• how the exchange is confirmed.

Excessive restrictions can protect the paper cost and destroy the perceived value in the experience. If you need a lot of exceptions to make the reward sustainable, it's probably worth choosing another reward.

Fudi maintains a single public program visible to diners. It also allows you to pause, resume or archive programs. This makes it easier to stop a problematic setup, but it is no substitute for planning: rewards already earned and expectations created should be addressed with clear and consistent rules.

Try the program during one or two normal purchasing cycles

Don't evaluate a coffee shop and a special occasion restaurant on the same timeline. The trial period should allow a reasonable portion of customers to progress far enough to reveal distance, value, and cost issues.

Before activating, save a baseline with:

• total sales and visits per branch;
• average net receipt;
• estimated contribution margin; *frequency of returning customers, if you can get it;
• cost and usual sale of the products chosen as a reward.

During the test, Fudi can provide:

• visits accepted;
• different diners;
• rewards awarded; *redeemed rewards;
• distribution of diners by visit or level;
• number of diners close to receiving a benefit;
• filters by date and branch.

The point of sale or accounting must provide:

• net sales;
• average ticket;
• variable costs;
• actual cost of delivered products;
• sales of products that could be substituted;
• inventory or preparation incidents.

With both sets of data you can calculate:

Tasa de canje = recompensas canjeadas ÷ recompensas otorgadas

Cost per participating diner = total program cost ÷ program participants

Burden on margin = total program cost ÷ contribution margin for the period

Break-even point in visits = total program cost ÷ contribution per additional visit

Don't use a single metric to declare success. A high redemption rate may indicate that the reward is attractive, but also that it is too generous. A low rate may indicate little value, a distant goal, poor communication, exchange difficulties, or simply that a sufficient cycle has not yet passed.

Interpret the signs before changing the reward

The results usually point to different problems.

Many people enter, but few advance

There may be initial enthusiasm without a sufficient reason to return. Review the natural frequency, distance to first reward, and consistency with which the team logs visits. Don't automatically lower the goal: first confirm that the program is being offered and operated in the same way.

Many rewards are awarded, but few are redeemed

The reward may not be relevant, the conditions may be confusing, or the product may not be available when the customer tries to use it. There may also be friction in the exchange. Talk to the team before concluding that the benefit is cheap and convenient.

Redemption is high, but sales or total visits are not improving

It is a sign of possible replacement. The program may be rewarding behavior that already occurred or giving away a product that was previously sold. Recalculate the economic cost using a more conservative scenario and check whether the benefit can become a complement rather than a replacement.

The cost is concentrated in one branch

Compare sales mix, costs, availability and way of operating. The same reward may be sustainable in one location and problematic in another. Fudi's branch filters help locate activity; The economic explanation must be completed with the operational data of each establishment.

Many clients stay close to the reward for too long

There could be a loss of frequency, a poorly aligned goal, or an inconsistent experience. Before giving away a visit or increasing the benefit, check to see if the normal time between visits has changed and if the team keeps the registration active.

Level profits grow faster than incremental activity

Reduce dependence on permanent discounts. Evaluate a new structure with more controlled cost benefits and a clear review horizon. Don't wait for the majority of repeat customers to reach the level to discover the accumulated cost.

Adjust without breaking customer trust

Changing the program every week makes it difficult to measure and also reduces confidence. The client needs to understand that his progress has stable rules. Therefore, before modifying, identify if the problem is in the reward, the goal, the operation or the measurement.

When the cost is clearly unsustainable, Fudi allows you to pause the program while you review the settings. If the adjustment can wait, complete a sufficient observation period and document the reason for the change.

When preparing a new version:

• preserve clarity about rewards already earned;
• communicate any changes before they affect the client;
• avoids silently reducing a promised benefit;
• updates team training and visible conditions;
• compares the new economy with the previous version using the same criteria.

The platform maintains consistent rules for participating diners. Still, the business decision on how to treat advancements and rewards must prioritize an understandable transition, not just a cost reduction.

Errors that usually erode the margin

Calculate with sales price

It makes some low direct cost benefits seem prohibitive and may lead you to dismiss them for no reason. Use the full economic cost, not the menu price.

Calculate with prescription cost only

Ignores replacement, packaging, additional work and capacity. It can make a reward seem much cheaper than it really is.

Choose the cheapest reward

A reward with no perceived value does not modify the next visit. The program retains operating cost and loses the ability to influence behavior.

Concentrate all the value at the end

It reduces the cost of quitters, but it can also make the goal feel far away. See if a controlled first milestone produces a better experience without accruing too many rewards.

Trust that few customers will redeem

The lack of exchange is not a sustainable strategy. Model a high participation scenario and treat a lower rate as a variation, not a requirement for survival.

Give permanent discounts without horizon

The cost is repeated for each future visit and is concentrated on the most active customers. Project the eligible population and their visits before activating it.

Use the same calculation for all branches

Costs, tickets and operating conditions change. A unique program needs a sustainable reward under the less favorable economics of participating branches.

Confusing adoption with increase

More visits recorded on Fudi may mean that the team uses the platform better, not necessarily that the restaurant receives more real visits. Cross program activity with sales and operation.

Approve Configuration Worksheet

Before publishing, complete this review with real data:

1. Behavioral Objective: What additional visit or relationship the program should produce.
2. Average net ticket: per restaurant and, when necessary, per branch.
3. Contribution margin per visit: Net receipt less variable costs.
4. Economic cost of each reward: direct, operational and replacement.
5. Cumulative cost of the cycle: sum of all possible benefits.
6. Estimated time to reach the first reward: according to usual frequency.
7. Estimated time to complete the cycle or reach the level: for frequent, average and occasional customers.
8. Full redemption scenario: cost if all awarded rewards are used.
9. Monthly program cost: rewards, platform and assigned operation.
10. Incremental breakeven visits: monthly cost divided by contribution per additional visit.
11. Visible rules: exact benefit, validity, branches, substitutions and exchange.
12. Trial period: enough to observe one or two normal purchasing cycles.
13. Review indicators: activity in Fudi, sales, margin, exchanges and incidents.
14. Correction threshold: specific condition that will force you to pause or redesign.

If you can't complete the first ten points, you don't have a setup yet; You have an idea of ​​reward. The difference between the two is the account that shows how much you can invest and what behavior that investment should pay for.

The next step is to test a promise that you can keep

Margin protection is not about making the program less attractive. It involves choosing a reward whose value to the customer is greater than its economic cost to the restaurant, placing it within reach, and demanding that the change in behavior pay for the investment.

First calculate the contribution per visit. Then model the complete cycle, including intermediate rewards and permanent benefits. Finally, configure in Fudi the same rules that you budgeted and review the activity by date and branch along with the sales and costs of your operation.

A sustainable program can become more generous when it demonstrates results. A poorly measured one usually does the opposite: it starts by giving away too much and ends up withdrawing benefits when customers already expect them. The best first version is not the most aggressive; It is the one that you can operate, measure and improve without putting the margin or the diner's trust at risk.

Sources consulted

• Leveraging the Loyalty Margin, Boston Consulting Group
• The Goal-Gradient Hypothesis Resurrected, Journal of Marketing Research
• Turn loyalty into a tech-powered marketing engine, National Restaurant Association