.png)
Whether it’s a staff restaurant, hospital cafeteria, takeout shop, or catering service: wherever food is prepared in advance, daily volume planning determines whether there will be leftovers or empty buffets. Demand fluctuates depending on the weather, the day of the week, holidays, and the percentage of employees working from home—and anyone who misjudges it pays the price twice: either through food waste or disappointed guests.
This guide explains how AI sales forecasts work in the restaurant industry, where they have the greatest impact, and how to implement them.
Why the Restaurant Industry Is So Hard to Plan For
The number of guests is a moving target. In the staff restaurant, work-from-home days, meetings, and the weather determine how busy it gets. At the take-out counter, a rainy day shifts demand from salads to soup. In the cafeteria, school breaks and exam periods change everything.
Food is prepared in advance of demand. Meals, salad bars, and sandwiches are prepared hours before it’s clear how many guests will be coming. So the quantity must be estimated—every day, for each item.
Freshness is unforgiving. Much of the inventory is no longer saleable by the next day. Any overproduction results in a direct loss; any shortage means lost revenue and an unhappy guest.
Experience alone isn't enough. An experienced management team develops a good sense of things—but when dealing with dozens of components, multiple locations, and varying calendar effects, daily forecasting becomes error-prone and time-consuming.
How AI Sales Forecasts Work
1. Integration: GoNina links the relevant data sources—sales and transaction data from the point-of-sale system, as well as weather, holidays, vacation periods, and location-specific factors. The integration is handled via automated interfaces, and the onboarding process takes just a few days.
2. Analysis: The AI identifies the patterns behind the fluctuations: How does a sunny Friday affect buffet traffic? How does the first day of vacation impact menu sales? How does Location A differ from Location B? The model continuously learns from each new day of sales.
3. Forecast: This generates specific figures for planning and production—on a daily basis, by item or component and by location, for the next seven days. The recommendations are incorporated into the usual workflow, and the team retains the final say in special situations such as events or large orders.
Where the effect is greatest
Institutional Food Service and Company Cafeterias: High volume, daily prep work, and highly fluctuating foot traffic—the prime example. In a pilot project with a leading Swiss provider of institutional food service, up to 48% less waste was recorded across ten locations. Read more about this in the article on company cafeterias.
Employee Cafeterias: Working from home, meeting days, and the weather drive fluctuations—this article on demand fluctuations in employee cafeterias highlights the causes and solutions.
Fresh foods and takeout: Sushi, salads, and sandwiches have the shortest shelf lives of all—read the in-depth article to learn how forecasts for highly perishable goods work.
Catering: Regular corporate clients and predictable orders are combined with spontaneous additional orders—this article on quantity planning in catering shows how to balance both.
What It Actually Offers
The same effects are evident across all areas of application: up to 52% less surplus, up to 6% more revenue due to improved availability, and up to 12 hours saved in planning time per month. Added to this are softer factors: less stress in day-to-day operations, greater planning certainty for production and staff, and a measurable contribution to reducing food waste—an issue that clients and guests alike are increasingly demanding in the restaurant industry.
Here's how the rollout works
Demo: In our initial consultation, we’ll use real-world examples to show how the forecasts fit into your operations. Trial Period: This is followed by a risk-free launch—four weeks of everyday use, with a money-back guarantee. Rollout: After the trial period, we’ll fine-tune the system and provide ongoing support. The service is hosted in Switzerland and the EU.
Frequently Asked Questions
Does this work even without a traditional point-of-sale system?
The system is based on recorded sales or expenditure data—whether from a cash register, the cafeteria’s point-of-sale system, or an ordering system. We’ll discuss the specific integration during our initial consultation.
How does the forecasting system handle menu changes?
Changing menus are a daily occurrence in the restaurant industry. The AI learns demand patterns at the component and category levels—ensuring that forecasts remain reliable even with rotating menus.
What distinguishes forecasts for the restaurant industry from those for the bakery industry?
The principle is the same, but the influencing factors differ: In the restaurant industry, customer traffic and calendar effects carry more weight, while in the bakery industry, the product assortment is the key factor. The article “From Bakery to Restaurant” shows what both industries can learn from each other.
Conclusion
The foodservice industry produces in anticipation of demand—and that is precisely why volume planning is its biggest hidden cost factor. AI sales forecasts turn daily estimates into data-driven decisions: by component, by location, seven days in advance. The result is full buffets without overflowing trash cans.
To learn how this same principle works in a bakery, read the guide to AI sales forecasts for bakeries.
