How do I calculate discounts without losing money?

1. The discount does not start with a percentage, but with the minimum price

I calculate discounts so that after the price reduction, at least my planned contribution margin remains in place, and I do it on one consistent net basis. In practice, that means I never start with “10 percent sounds good,” but with my minimum net selling price. That comes from product cost, directly attributable costs and the amount each sale still has to contribute toward my fixed costs. A simple bistro example: a pasta dish has a net recipe cost of 3.20. If I normally sell it for 12.80 net, I have a gross margin of 9.60 before further overhead. If the promotion is not supposed to slip into a loss, I define a floor of 10.90 net in advance. If I show that as a guest-facing gross price for food in Germany at 7 percent VAT (Section 12 UStG), that would be about 11.66 and 13.70 respectively. Any discount that pushes the net price below that is no longer marketing, but giving away margin.

I explain the same logic very clearly to my team: a discount is not a nice feeling, but a planned exception to the standard price. Guests can expect us to act fairly and transparently, and staff can expect not to negotiate prices from the gut. That is why I define three values for every promotion: the regular price, the promotional price and the absolute floor price for complaint or goodwill cases. If, for example, a guest receives a reduction because of a long wait, the decision should not depend on the server’s mood that day, but on a range discussed in advance. That way I protect revenue, avoid debates at the table and still leave room for genuine hospitality when a situation truly calls for it.

2. The right calculation order: costs and target contribution first, discount second

To avoid making a loss, I always calculate discounts in the same order: first the product cost per sold item, then the target contribution per sale, and only after that the maximum possible discount. A burger menu shows why. Burger, bun, sauce, side and garnish cost 4.80 net in total. Packaging or extra material for takeaway could increase that amount; in the restaurant there may also be a small dip bowl. If I need at least 6.70 net contribution after product cost for this menu, my minimum price is 11.50 net. If the menu normally sells for 13.90 net, the largest discount I can give is 2.40 net. In percentage terms, that is about 17.3 percent. Anything above that eats into my planned contribution and turns a controlled promotion into a weaker result than I intended.

The typical mistake is using the wrong reference point. Many people calculate 20 percent off the selling price and assume there is still “enough room” afterwards. In reality, the discount applies to the full price, while product cost hardly changes. The same discount is therefore much more dangerous on a dish with a high material share than on coffee, water or house lemonade. That is why I separate product groups carefully. I can look very differently at a dessert with a cost of 1.10 than at a steak dish with a much higher purchase value and possible weight loss during cooking. This is exactly why I keep recipes and ingredient costs structured, so I decide on a reliable basis instead of by feel: Understand what goes into every dish on your menu.

3. Percentage, euros or a bundle: separate net and gross first, then choose the discount format

Whether I offer a discount as a percentage, a fixed euro amount or a bundle makes a big difference, but only if I separate net and gross prices properly first. For calculating product cost and contribution, I compare prices on the same net basis; the visible guest price can differ from that. A simple example shows it clearly: 10.70 gross equals 10.00 net at 7 percent, and 11.90 gross also equals 10.00 net at 19 percent. That is exactly why I first check the net floor price and only then decide on the form of the discount. Percentage discounts look simple, but they are often imprecise because they treat expensive and cheaper items alike. A fixed reduction of 2 euros is easier to control once I know the net floor. Bundles are often even better, because the advantage stays visible without pushing the stronger margin core unnecessarily deep.

I organise that decision deliberately by product logic. Items with high product cost or fluctuating day quality are more likely to stay out of blanket discounts. I use bundles where add-on items can raise the check without causing much extra production stress. An afternoon example is coffee and cake sold as a package, which often works better than a general reduction across the whole menu. That way I do not accidentally cut the price of items that already sell well on their own. For my team, that is much clearer in service: instead of calculating for themselves, they sell a cleanly defined promotion. For guests, it does not feel like arbitrary haggling, but like an honest, understandable offer with a visible benefit.

4. Give discounts only on suitable items, not across the whole menu

Losses often come not from the individual discount itself, but from applying it to the wrong range of items. That is why I keep discounts as selective as possible and never spread them blindly across the whole menu. Good candidates are items with solid margin, predictable production or a useful steering effect. Bad candidates are products with high product cost, strong waste risk or already tight calculation. A realistic example: my lunch menu is supposed to bring in more traffic. In that case I would rather discount a pasta dish with stable recipe cost than a fish dish whose purchase price changes from day to day. The guest still sees an attractive offer, but I keep financial control on my side and avoid unpleasant surprises in product cost.

I also check which operational consequence a discount creates. A low price on a labour-intensive pass item can overload the kitchen during peak time and create indirect cost through longer ticket times, more mistakes or unhappy guests at neighbouring tables. So I do not calculate only the plate, but also the production capacity behind it. If a promotion only runs cleanly when service, kitchen and bar all know the same framework, I set it up as a clear process and communicate it to the team beforehand. Especially with several stations, it helps when everyone works from the same order instead of reinterpreting every request from scratch: Give service, kitchen and bar a shared view of orders.

5. This is how I work through an individual hospitality example properly

Take a concrete example from an urban evening bistro: I sell a house lasagne at a regular price of 14.50 net. The recipe cost is 4.10, and including portioning, cheese finish and side salad, the total product cost comes to 4.90 net. If I want each sale to contribute at least 7.00 toward wages, rent, energy and other overhead, the minimum price is 11.90 net. My maximum discount is therefore 2.60 net, or roughly 17.9 percent. If I plan a weekly promotion, I do not simply apply 15 or 20 percent. I decide based on the goal. If I only want to improve occupancy on a weak Tuesday, a 2 euro discount may be enough. If I want to push a menu deliberately, I would rather combine the lasagne with a drink than reduce it further.

The same example also shows the difference between a discount and extra sales. If I drop the lasagne to 11.50 net, I move below my floor price and lose 0.40 contribution per portion against my minimum target. If I sell 40 portions that evening, I am short 16.00 in contribution, even if the dining room feels fuller. In practice, that kind of misdirection is dangerous because it can feel like success. That is why I do not evaluate promotions only by revenue or by units sold, but by what remains after product cost. For my team, I put the rule simply: full occupancy is only good when the extra plates do not just create more work, but also deliver the planned contribution. Otherwise I am discounting business instead of building it.

6. Organise goodwill, team authority and limits at the table fairly

Not every discount is a marketing action; some price reductions happen as goodwill. To prevent losses caused by uncertainty, I define clear decision levels. For example, up to a small amount the shift lead may decide independently in understandable cases, such as a clearly delayed dish or a proven misunderstanding in the order. Beyond that, a second approval is required. What matters to me is that staff should not be squeezed between hospitality and fear of making a mistake. Whoever is responsible at the table should have a fair framework, but should not have to conduct open price negotiations. That is why I define which cases are better solved with a dessert, a drink or a small euro reduction rather than a large percentage discount on the whole bill.

This clarity is also fair for guests. If two tables have the same issue, they should not receive completely different reductions just because different people are on shift. I organise this with a simple matrix: the occasion, the appropriate remedy, the maximum amount and the approval level. One everyday example: after a long wait, a hot drink or dessert can be more sensible than 15 percent off the whole table, because the guest gets a noticeable make-good and I do not unnecessarily cheapen high-margin items that were delivered correctly. At the same time, I document internally why the goodwill was granted. That helps me spot patterns, such as recurring bottlenecks at one station, and fix the cause instead of permanently treating the symptom with discounts.

7. Whether a discount works only becomes clear in the follow-up review

A discount is only good when I can prove afterwards that it achieved its goal. After every promotion, I therefore review at least four points: quantity sold, average check, effect on product cost and the contribution that remained. One example: an afternoon promotion brings in 18 additional orders, but lowers the average check because guests choose the discounted individual item instead of full menus. In that case the discount may have been visible, but economically weak. That is exactly why I do not compare the promotion day only with my feeling, but with a suitable reference, such as the same weekday without the promotion. Only then can I see whether the extra guests actually created extra return or whether I simply sold regular purchases more cheaply.

I base my next decision on that review. If a promotion works only with a very deep discount, the real problem is often not the reduction itself, but the offer, the timing or the communication. Then I would rather change the format of the offer, limit it to quieter times or switch to a more attractive bundle. For my team, I record the result briefly and concretely: what was discounted, where the floor price sat, what the outcome was and what we will do differently next time. That takes the arbitrariness out of discounting. Guests experience understandable offers, staff act more confidently, and I protect the business from drifting into losses through well-meant price actions.