Loyalty cost: what a reward really takes out

A regular redeems a free main course worth eighteen pounds. The owner thinks of it as costing six pounds, because that is roughly what the ingredients cost. The accountant thinks of it as costing eighteen pounds, because that is the sale the restaurant did not make. Both are right in different situations, and the difference between them decides whether a loyalty scheme is a sensible investment or a quiet leak.

Loyalty cost is not a single number. A reward costs the food it uses, but it may also cost the sale it replaces, the time it takes to process and the habit it creates. Working out what a reward really takes out of the business means asking one question above all: what would the guest have done without it?

This guide covers how to cost rewards properly, the difference between food cost and menu price, the figure that decides the whole design, the hidden costs of a scheme and how to design rewards that cost less than they appear to. Every venue and figure in it is an example.

There are two ways to cost a reward. At food cost, a free main course costs what its ingredients cost, perhaps a third of its menu price. At menu price, it costs the full price the guest would otherwise have paid.

Food cost is the right measure when the reward is extra: a dessert the guest would not have ordered, a drink on a night they would not have come. Menu price is the right measure when the reward replaces something the guest would have paid for: a main course they would have ordered anyway, on a visit they would have made anyway.

Most rewards fall somewhere in between, and that is where the honest costing lies.

The figure that decides the whole design is the share of rewards that replace a purchase the guest would have made anyway. If most rewards go to regulars on visits they would have made regardless, the scheme's real cost is close to menu price. If most rewards bring guests in more often, or add items they would not have bought, the cost is close to food cost.

This is the economist's idea of opportunity cost: the true cost of a choice is what you give up to make it. A free main course costs the sale it displaces, if it displaces one.

Gross profit per dish, the menu price minus the food cost, is the right lens for costing rewards. A free main that would have sold for eighteen pounds with six pounds of ingredients takes twelve pounds of gross profit if it replaces a sale, and six pounds of ingredients if it does not.

Our guide to gross profit per dish beats food cost percentage covers why the money a dish makes matters more than its percentage. The same applies to rewards: what matters is the profit given away.

Some rewards cost far less than their value to the guest. A dessert, a coffee, a glass of house wine or a side dish often has a low food cost and a high perceived value. Guests enjoy them, and they rarely replace something the guest would have paid for, because many guests skip dessert or a second drink.

Rewards that cost the most are main courses and expensive drinks, especially if guests would have ordered them anyway. A scheme built on desserts and drinks is usually far cheaper than one built on free meals.

Rewards that bring guests in on quiet days or at quiet times cost the least of all. A free dessert redeemable on a Tuesday, or a reward that encourages a booking in the first week of January, fills a table that would otherwise be empty.

The food cost is real, but the table, the staff and the room were already paid for. If the reward brings a guest who also buys a main course and a drink, the reward has more than paid for itself.

The most expensive rewards are the ones that change nothing. A regular who visits every Friday, spends the same and redeems a free main every six weeks has received money back without any change in behaviour. The scheme has cost the full menu price of each reward and gained nothing.

That does not mean regulars should be ignored; they deserve recognition. But recognition can be given in ways that cost less: a warm welcome, their favourite table, an invitation to a tasting.

Watch how and when rewards are redeemed. If most redemptions happen on the busiest nights, by the most frequent guests, on the most expensive dishes, the scheme is costing near menu price. If they happen on quiet nights, by occasional guests, on desserts and drinks, it is costing near food cost.

Loyalty redemption patterns tell you whether the design is working as intended. Adjust the rules, such as which rewards are offered or when they can be used, if redemptions cluster where they cost most.

Beyond the rewards, a loyalty scheme has hidden costs. Software and cards. Staff time explaining the scheme and processing rewards. Marketing to recruit members. The time spent dealing with disputes about points. And the outstanding balance of points not yet redeemed, which is a promise the venue must one day honour.

Add them up. A scheme whose rewards cost two per cent of members' spend but whose software and staff time cost another two per cent is a four per cent scheme.

In points based loyalty, the reward rate, the share of spend returned in rewards, sets the starting cost. One point per pound, with a hundred points buying a five-pound reward, returns five per cent of spend at menu prices before breakage. Our guide to points based loyalty: the maths before the launch works through the numbers.

The real cost then depends on which rewards guests choose and whether they replace purchases. The same scheme can cost one per cent or four per cent of members' spend.

Discounts and rewards cost differently. A discount, such as ten per cent off the bill, costs the full value at menu price every time it is used, and it applies to everything on the bill. A reward, such as a free dessert, costs a fixed item, usually at a lower food cost, and the guest still pays for everything else.

Discounts also train guests to expect lower prices. Our guide to discounts or rewards: which habit are you building? covers the difference in behaviour.

A reward is worth giving if it brings back more profit than it costs. Work out how much extra a guest must spend, or how many extra visits they must make, to cover the reward's real cost.

If a reward costs six pounds in food and brings a guest back for an extra visit worth forty pounds, with a gross profit of twenty-six pounds, it has paid for itself many times over. If it costs eighteen pounds in lost sales and changes nothing, it is pure cost.

Measure the real cost by comparing members with similar guests who are not members, and members before and after joining. Look at visit frequency, spend per visit and what they buy.

If members visit more often after joining, part of each reward is paying for itself. If they visit no more often, every reward is costing close to menu price. That comparison, more than any formula, tells you what the scheme really takes out.

Record every reward on the till, as a reward, with its reason, rather than simply leaving it off the bill. A reward that is not recorded cannot be counted, and the venue cannot see how much it is giving away.

A reward recorded properly shows in the reports alongside comps and discounts, by member of staff and by reason, so the cost of the scheme is visible every week.

If the figures show a scheme costs too much, change it carefully. Shift rewards from main courses to desserts and drinks, make some rewards redeemable only at quieter times, or raise the threshold for the most expensive rewards.

Tell members in advance, explain the change honestly and keep existing points valid for a reasonable time. A scheme that is changed fairly keeps its members; one that is changed suddenly loses their trust.

The hardest part of costing rewards honestly is estimating what guests would have done without them. You cannot ask every guest, but you can look at patterns. Did members visit as often before they joined as after? Do members order the reward dish at full price on other visits? Do rewards cluster on nights the venue is full anyway?

Each answer gives a clue. If members visited just as often before joining, many rewards are replacing purchases. If rewards are mostly desserts that members rarely order otherwise, they are mostly extra.

A reward can bring a guest in, but the visit itself has costs and benefits beyond the reward. A guest who comes in to redeem a free dessert usually buys a main course, drinks and perhaps a coffee. The reward's cost should be weighed against the whole visit's gross profit.

That is why rewards that bring in a visit, rather than simply discounting one, can be very cheap overall. A six-pound food cost that brings in a forty-pound visit is a good trade.

Rewards that bring groups are especially valuable. A birthday reward for one guest often brings four or five others who pay full price. The reward's cost is spread across a table, while its value is concentrated on the occasion.

Design some rewards to encourage groups: a cake for a birthday table, a bottle of wine for a party of six on a quiet night. The cost per guest at the table is tiny.

Rewards affect the kitchen too. A scheme that sends many free main courses through the kitchen on a busy Saturday adds work without adding revenue. A scheme that sends desserts on quiet nights uses spare capacity.

Talk to the kitchen when designing rewards. The chef knows which dishes are easy to add at quiet times, which have a low food cost and which would strain the line on a busy night.

Some rewards build better habits than others. A reward for visiting on a quieter day builds the habit of coming midweek. A reward for trying the new menu builds curiosity. A reward for bringing friends builds groups.

A reward that only ever brings regulars on Saturdays builds nothing new. Think about the habit each reward creates, as well as what it costs.

Compare the cost of a loyalty scheme with other ways of spending the same money: marketing to find new guests, better ingredients, more staff on busy nights. A scheme that costs three per cent of members' spend is a significant investment.

Ask whether that money would bring more guests back if spent another way. Sometimes the answer is a better welcome or a quicker bill, not a reward at all.

You do not need complicated models to cost rewards well. A simple sheet with each reward, its food cost, its menu price, how often it is redeemed and roughly what share replaces a paid purchase gives you most of what you need.

Review it every few months. The sums change as the menu, prices and guests change, and a reward that was cheap last year may be expensive now.

Staff often see rewards as free, because the guest does not pay. Explain to the team what rewards really cost, in simple terms. A server who understands that a free main course on a busy night costs the venue a full sale is more careful about processing rewards correctly.

Understanding also helps staff explain the scheme to guests honestly, without overpromising.

Sometimes a reward that looks expensive is worth it: a reward that keeps a valuable regular coming back after a bad experience, or a reward that wins a large group booking. The question is not whether the reward costs money, but whether it brings back more than it costs.

Keep the big picture in mind: loyalty is an investment in guests coming back. Cost it honestly, and spend it where it brings the most return.

The cost of a reward is not fixed across the year. A dessert made with summer berries costs more in winter, a free bottle of wine costs more after the supplier raises prices, and redemptions themselves cluster around the festive season and the summer holidays, when guests have more reason to go out.

Look at when rewards are redeemed and what they cost at those times. A reward that is cheap in March may be expensive in December, when the kitchen is busy and every table could be sold at full price.

Choosing rewards whose cost stays steady, or that can be swapped for a seasonal dish of similar value, keeps the scheme's cost predictable. So does limiting redemptions on the few nights a year when the venue will be full regardless.

Drinks usually cost less to give away than food, compared with their menu price. A coffee, a soft drink or a glass of house wine has a low cost against what the guest would pay for it, needs no kitchen time and can be served in a moment, even on a busy night.

Food rewards, on the other hand, often feel more generous and more personal, especially a dish the chef is proud of. A dessert or a starter sits somewhere in between: lower in cost than a main course, but still a taste of the kitchen.

Alcohol rewards need care. Responsible service still applies, some guests do not drink, and a reward should never encourage a guest to drink more than they intended. Offer an alternative of similar value, so every member can enjoy the reward.

When menu prices rise, rewards change cost without anyone deciding it. A reward of a free main course now costs more at menu price, and usually at food cost too, because ingredient prices often rose first. A reward worth a fixed amount, such as ten pounds towards the bill, costs the same but means less to the guest.

Neither is wrong, but the choice should be deliberate. Review the rewards each time the menu is repriced, and decide whether to keep their value to the guest or their cost to the venue.

Tell members if a reward changes. A guest who saved towards a reward and finds it worth less than they expected feels the scheme has been changed behind their back.

An invented bistro runs a scheme where every tenth visit earns a free main course. The owner costs it at food cost, about six pounds a reward, and thinks it is cheap. The figures show otherwise: most rewards go to guests who visit weekly and would come anyway, and almost all are redeemed on Fridays and Saturdays, when the bistro is full.

The bistro changes the reward to a dessert and a glass of wine, redeemable from Monday to Thursday, and adds an invitation to a quarterly tasting for its best guests. The food cost of each reward falls, it rarely replaces a paid purchase and redemptions move to quieter nights.

The scheme's real cost falls by more than half, and the bistro's midweek covers rise as members come in to use their rewards.

servearound does not run a loyalty scheme, so it does not count points, rewards or redemptions, and there is no field for a dish's cost, so a reward's food cost and gross profit are worked out in your own sheet. What servearound does show is the sales side: the menu report gives the units sold, revenue and share of every dish over any range of dates, which tells you how often a reward dish sells at full price.

A reward given at the table can be recorded on the till as a comp, with its reason, and comps then appear in payment analytics by member of staff, so the giveaways are visible each week. Guest analytics show the repeat rate and how many guests came back, which is the behaviour a reward is meant to change. See Analytics.

It depends on what the guest would otherwise have done: food cost if the reward is extra, close to menu price if it replaces a purchase.

The share of rewards that replace something the guest would have bought anyway.

Desserts, drinks and sides, especially redeemable at quieter times.

Main courses and expensive drinks redeemed by regulars on busy nights.

Software, cards, staff time, marketing, disputes and the outstanding balance of unspent points.

Usually not. Discounts cost the full value at menu price on everything on the bill.

Compare members with similar non-members, and members before and after joining.

Yes, as rewards with a reason, so their cost is visible every week.

Loyalty cost is not one number. A reward costs its food if it adds to what the guest buys, and close to its menu price if it replaces a purchase they would have made anyway. That share, what the guest would otherwise have done, decides whether a scheme is an investment or a leak.

Cost rewards by gross profit, choose rewards with low food cost and high perceived value, make some redeemable at quieter times, count the hidden costs, record every reward on the till, measure whether members actually come back more often and change an expensive scheme fairly and with notice.