When a hotel restaurant sells a dish at half price, the guest gets a deal, the app gets its fee, and the hotel gets the bill.
Hotels have discounted food and beverage for decades. What has changed is who benefits.
Picture a quiet Tuesday. A couple book a table through an app at 50% off. The chef cooks their dish with the same care as every other plate. The server gives them the same warm welcome. At the next table, two regulars are celebrating their anniversary. They are paying full price for the same meal.
The app takes its fee. The couple earn loyalty points with the app, not with the hotel.
The hotel got a booking. But did it make any money? Will that couple ever come back? Will the hotel even know who they were?
More hotel owners should be asking these questions.
The new middlemen in hotel dining
So who now controls the dining offer?
American Express owns Resy, the restaurant booking app. Its Platinum cardholders can get up to US$400 a year back when they pay at US restaurants on Resy. Chase gives Sapphire Reserve cardholders up to US$300 a year at a selected list of restaurants on OpenTable.
The card companies pay for these credits, not the restaurants. But the message is clear. Restaurants are being used to build loyalty to a card, not to the restaurant or the hotel.
OpenTable belongs to Booking Holdings, the group that also owns Booking.com.
Rooms show where this can lead. In Europe, online travel agencies took 19.7% of hotel bookings in 2013. By 2025 they took 29.9%. In 2024, the European Commission named Booking a “gatekeeper” under its Digital Markets Act. That is the EU’s label for a platform that stands between businesses and their customers.
The delivery apps are moving in too. In June 2025, DoorDash completed its purchase of SevenRooms, a booking and guest-data system used by many hotel groups. Uber has partnered with OpenTable to take table bookings inside Uber Eats. In India, Swiggy promotes flat discounts of up to 50% through its dine-out service.
Finding a restaurant, booking it, paying for it and staying in touch with the guest used to be separate steps. More and more, one app does all of them.
So when a guest decides where to eat next weekend, whose app will they open? Probably not the hotel’s.
More bookings do not always mean more profit
Take a main course that sells for US$100 (net of VAT). The maths is the same in any currency: euros, pounds, rupees or riyals.
The ingredients cost US$30, so the food cost is 30%. That leaves US$70, or 70% of the price.
That US$70 is not profit. It is what is left after the food cost. It still has to pay for the team, the rent, the energy and every other cost. Only what remains after all of that is profit.
Now sell the same dish at half price. The price drops to US$50. The kitchen still uses the same ingredients, so they still cost US$30.
Your food cost percentage has doubled, from 30% to 60%. After food cost, only US$20 is left, or 40% of the new price. That US$20 has to cover the same team, rent and energy. And that is before the platform takes its fee.
To match the US$70 left from one full-price dish, you need to sell three and a half discounted dishes (3.5 × US$20 = US$70). That is three and a half times the covers, the plates, the kitchen work and the service.
If the table would have stayed empty, that US$20 may still be worth having, because it helps pay costs you carry anyway. If the table would have sold at full price, you have given away US$50 that would have gone towards your costs and your profit.
Some platforms discount the food and leave drinks at full price. That protects some margin, but it is the wrong way round. It teaches guests that the dish is worth half.
Platforms measure success in bookings and new customers. Hotels should measure extra profit and repeat visits.
A booking is not the same as a profitable guest.
Teaching guests to wait for a discount
I have always worried about discounts becoming what a venue is known for.
Guests learn fast where the offers are. Many will still pay full price. Others start to follow the deals, moving from one venue to the next as the offers change.
There is nothing wrong with guests looking for value. The problem starts when a hotel teaches its guests that the menu price is not the real price. That lesson is very hard to unteach.
In India, restaurants pushed back. In 2019, and again in 2022, the National Restaurant Association of India urged its members to log out of dine-out apps over deep discounting. Its warning was blunt: once that discount culture takes hold, it cannot be reversed.
Seoul shows the other end of the market. There, the leading hotel buffets charge more than ₩200,000 per person for weekend dinner. They sell the food, the service and the occasion. When their costs went up this year, they raised their prices.
Did guests walk away? In 2023, Lotte Hotel’s La Seine raised its weekend dinner price by 38.5%. Most of its weekend bookings for that month were already sold out.
Protecting the price matters, especially for luxury hotels.
How I used discounting
I am not against discounting. I used it throughout my career. But I used it with rules.
Only when demand was missing. I started with the monthly sales reports, the year-on-year comparisons and the booking patterns. I also watched the weather. Rain, sunshine and temperature all change demand, especially for venues with outdoor seating.
The aim was to find the quiet periods where extra guests would add profit. I never discounted the nights that would fill anyway.
The drink, not the food. Where I could, I built offers around drinks.
Food is the heart of a restaurant’s reputation. I did not want guests getting used to paying half price for a dish the kitchen had to deliver at full quality.
Drinks packages (alcoholic and non-alcoholic), upgrades and added experiences gave us room to create value without cutting the food price in public.
Hotel brunches across the Middle East show how this works. Every guest eats the same food. The drinks package sets the price. In Bahrain, one hotel’s Friday brunch has been priced from BD15 with soft drinks to BD40 with its premium drinks package. In Doha, another has been priced at QR345 with soft drinks and QR425 with its premium package. Same food, very different spend.
Chosen guests, not everyone. I aimed offers at people who could become regulars or bring others with them. Depending on the venue, that meant HR leaders from nearby companies, doctors, nurses and other professional groups.
I also worked with model agencies and up-and-coming influencers who wanted to grow their following. Their people could shape where friends and wider circles chose to spend their time. My advice: don’t waste your time on the ones who have already made it and are only about themselves.
These were deliberate business decisions. They were not discounts handed to anyone who opened an app.
Who owns the guest?
Hotels already have big advantages: guest databases, corporate clients, loyalty programmes, events, and teams who know their regular guests by name.
Loyalty programmes matter here. When a member dines, they stay within the hotel’s own brand, and the hotel knows who they are. But a loyalty discount is still a discount. It should pass the same tests as any platform offer.
The platforms will say the restaurant owns its guest data, and that is partly true. OpenTable shares US guests’ email addresses with restaurants, along with whether each guest has agreed to receive the restaurant’s marketing. But the guest’s points and reviews stay in the app. And OpenTable charges a fee for each guest who books through its app or website. Bookings made on the restaurant’s own website or by phone cost nothing on its main plans.
The guest who books with you directly costs you no fee, and you can invite them back.
Guests are ready for that direct relationship. In a 2025 SevenRooms survey of 1,000 US diners, 83% said they would join a restaurant’s marketing programme. But they wanted more than discounts. The perk they wanted most was a birthday or anniversary offer.
That tells us something important. Guests do not only want a lower price. They want to be remembered.
That is how I always worked. When I was a restaurant General Manager, I encouraged my team to ask guests for reviews that named the waiter or waitress who looked after them. We used the in-house CRM to keep notes on every guest. And I gave guests my own number, so they could call me to book.
When I went on to manage larger regions, I asked every manager and host to follow the same principles. The guest had to feel cared for by the team, not processed by a system. And they had to know how to book with us next time.
But that only works if someone answers. Today, too many hotels let calls go to voicemail, or reply on WhatsApp with an automated message. A guest who cannot reach you is one tap away from booking through an app instead.
The data points the same way. In a 2025 Harris Poll of more than 2,000 US adults, commissioned by an AI phone-answering company, 63% said the phone is their preferred way to contact a restaurant. And when Revinate mystery-shopped 308 reservation calls to luxury and upscale hotels in 2026, 92% of the calls that did not end in a booking finished with no guest contact details taken.
I have seen the opposite happen. When I took over restaurants, or consulted for them, I taught the teams to capture guest data and use their CRM properly. We built guest recognition into service, with small, unexpected touches around food or drink. Moments of delight.
The results followed. Guests came back more often, and fewer drifted away. No-shows fell. Covers went up, average checks went up, and the team earned better tips.
Recognition pays everyone.
Speed of response is now part of hospitality. A guest who wants to book with you should reach a person, quickly.
A platform can be a good partner. The danger is depending on it without building any relationship of your own.
If the hotel carries the cost and the platform keeps the guest, whose business are we really building? I don’t think it’s the hotel’s.
Four questions before the next promotion
Before approving another discount, I would want the team to answer four questions:
- Would this table really stay empty without the offer?
- What profit is left after the discount, the direct costs and the platform fee?
- Could we create demand another way, through timing, drinks packages or added value?
- Will we know who the guest is, with their permission, so we can invite them back?
I am not saying hotels should stop using booking platforms or stop running promotions.
I am saying they should be far more selective about when they discount, who gets the offer, and what the hotel gains in return.
Discount the quiet night, not the busy one. Discount the drink, not the dish. And think twice before you discount a guest you can never invite back.
Filling a restaurant at any price is not a strategy. It quietly damages what the venue can earn for years to come.
And here is the part that should worry owners. In the US, some of the most sought-after restaurants now hold tables for Amex Platinum and Chase Sapphire Reserve cardholders. The best seats in the house are being used to build someone else’s loyalty.
When did your best table last go to your own best guest?
