.png)
Tuesday: 250 meals, Friday: 140, long weekend: 60: No sector of the food service industry fluctuates as systematically as the employee cafeteria. The number of diners depends not on the cafeteria’s own offerings, but on the company’s work schedule—and that has changed fundamentally with the rise of remote work and flexible work models.
Anyone who plans for these fluctuations using average figures will produce too little on Mondays and waste too much on Fridays. This article explains the causes—and how data-driven planning can address them.
The Five Drivers of Fluctuations
1. The Work-from-Home Routine. The most significant trend right now: Tuesday through Thursday are office days, while Monday and Friday are work-from-home days. The difference between the busiest and least busy days of the week can be a third or more—and the pattern varies from company to company.
2. Vacations and long weekends. School vacations thin out the staff, and long weekends leave the building almost completely empty. If you don’t plan around the holiday calendar down to the day, you’ll end up serving “phantom guests” on long weekends.
3. The weather. Sunny weather draws guests outside to neighborhood take-out spots and restaurants, while rain fills the staff cafeteria. In addition, menu choices shift: salad bar when it's hot, stew when it's cold.
4. Company-wide events. Town halls, workshops, customer events, and team events result in hundreds of meals—sometimes eaten at a restaurant, sometimes not.
5. Your own menu. Popular menu items attract guests who would otherwise eat out. Menu planning affects foot traffic—and should therefore be included in the analysis.
From the Cause to the Solution
The key point: Four of these five drivers are predictable. Work-from-home schedules, holiday calendars, weather forecasts, and menu cycles are all embedded in data that already exists today. " GoNina " learns these patterns from the company’s sales data and provides daily forecasts for each component over the next seven days—including the effects of weekdays, holidays, and weather.
For the fifth driver—internal company events—humans remain in charge: Management can adjust the forecast in response to known events. The division of labor is thus clear: The machine calculates the patterns, and humans contribute the knowledge that isn’t contained in the data.
Practical experience shows just how effective this is: GoNina — companies are reducing their food waste by up to 52%. The article on company cafeterias explains exactly how planning works in these settings.
Three Immediate Measures
Analyzing Weekday Profiles: Sales data from the past few months, broken down by weekday, clearly demonstrate the impact of working from home—the basis for setting differentiated initial quantities rather than a one-size-fits-all figure.
Mark long weekends on the calendar: Every long weekend of the year is already known—and should be included in your rough planning right now.
Track daily consumption: Without daily surplus data, planning lacks feedback. What gets measured improves.
Frequently Asked Questions
Our work-from-home pattern is changing right now—so is historical data even helpful?
Yes, because the model is constantly learning: New sales data is used every week to update the patterns. This allows for structural shifts to be captured within a few weeks—significantly faster than it takes for our mental models to adjust based on experience.
We don't know the number of guests in advance—is point-of-sale data sufficient as a basis?
Yes. The number of meals and components sold per day is exactly the metric that needs to be forecasted—a separate headcount is helpful, but not required.
Conclusion
Fluctuations in demand at the staff restaurant aren’t chaos—they’re a set of predictable patterns: working from home, vacations, the weather, and menu cycles. If you plan based on data, you’ll cook enough on Tuesdays and avoid wasting food on Fridays. The rest is a matter of skill.
To learn how AI forecasting works in the restaurant industry as a whole, check out the complete guide for the restaurant industry.
