How does your restaurant forecast demand? And how accurate is your system?
Every manager forecasts differently. Some rely on their gut instincts. Others on a combination of guesswork and copying and pasting past weeks’ approaches. In an industry where tight margins are getting even tighter, pure guesswork won’t cut it. The real secret to seeing the future is a comprehensive understanding of the past. The right tools (and data) can help your manager make better predictions.
The more accurate the forecast, the more profitable the business. What happens when you know what you are going to sell (and when)? You know what you need to buy, and what schedule to set to ensure the right number of people are on at the right time. It’s all about striking the right balance between expectation and reality. This will generate less waste, more profit, and happier employees and guests.
The key to achieving an accurate forecast is using the data you already have at your fingertips. An integrated back-office system will make this easier. Combine internal data (like historical sales analysis and recent trends) with external information (such as weather, national events, and holidays). Then, managers should overlay their local knowledge (about local events, or nearby construction). Self-learning algorithms can use this data to forecast what items will sell. The right tool can do this right down to 15-minute increments. This approach is called demand forecasting.
Demand Forecasting for Stronger Schedules
Labor is one of the highest possible costs for a hospitality business. More accurately planning and scheduling labor should be top priority.
A demand forecast shows the shape of the day and the anticipated demand. That way, managers can see exactly how many employees they need in each area to meet that demand. It also takes into account exactly how much time employees need to deliver each activity. This even includes non-revenue-generating (but necessary) activities, like prep and cleanup. Forget designing schedules around SPLH targets or predicted covers. Managers can now set a schedule based on what will sell. They can even stagger start and end times to limit over- or under-staffing. Employees have better shifts and the guests experience the best-possible level of service. And, in states where predictive scheduling legislation exists, setting more accurate initial schedules helps limit liability and last-minute changes. This further protects the business from change fees.
Inventory Management and Prep Planning
Using the demand forecast to manage prep planning means saving time, money, and effort. “Managed prep planning” is the science of purchasing the right inventory and managing kitchen production using analysis and calculation. Restaurants that manage their prep planning are able to have the right products and labor available, so they can meet customer demand.
A robust demand-forecasting engine will create predictive orders. It can also deliver a plan to optimize kitchen prep. This helps you figure out how many people you need when scheduling kitchen staff. And, your team will spend less time prepping items that only end up in the trash. They will be more efficient in their prep work, and better equipped to meet customer demand. The benefit? Improved forecasting accuracy both lowers your environmental impact and strengthens your bottom line.
The demand forecast also helps guide your purchasing strategy by ensuring you only buy what you need. It helps avoid the trap of buying in bulk to get a discount, only to end up with inventory that goes bad before it can be used.
Full-System Integration and Limiting Data Errors
As busy operators know, it is impossible to get ahead if you’re always scrambling to catch up. With manual processes and teams working in silos, inaccurate and outdated data is everywhere. It’s easy to accidentally pull the wrong data. The errors only compound as they flow through the system. Cobbled-together systems are rife with error, too. Many managers miss out on opportunities to optimize as a result.
Fully integrating your systems helps achieve the single version of the truth. Using the most accurate, up-to-date data is critical. Tools like Fourth Analytics can simplify the process of aggregating and correlating data. Intuitive, pre-built dashboards and reports make it easy to uncover opportunities for improvement. And, give you the chance to act on them quickly.
With a back-office solution that allows your menu, HR, scheduling, purchasing and inventory management systems to seamlessly integrate with your POS and other integrated solutions, a single data set can flow among them. This helps you drastically reduce error, save time and money, and streamline operations. Correlated data gives complete visibility into the organization. You’ll be able to make better decisions faster and with less of an administrative burden.
Measuring Forecasting Accuracy
At Fourth, we use Weighted Absolute Percentage Error (or WAPE) to understand the true levels of forecast accuracy. WAPE ensures that instances of under or over forecasting do not cancel each other out in weekly figures or regional aggregations. And, because it is a weighted measure, operators are able to see the true cost to the business, proportional to the size of each location. WAPE helps reveal hidden insights and clear opportunities for improvement.
To learn more about how you can achieve the most accurate forecast possible, get in touch. Want to see how accurate demand forecasting can help transform the scheduling process? Download our complimentary white paper, The Science of Scheduling.
Just a few things happened in 2020 – an election, a pandemic, an economic shut down. Believe it or not, the world is still spinning, and we are still building features in HotSchedules to help you manage your workforce.
HotSchedules was created in 1999 to solve one of the hospitality industry’s most complex problems: keeping teams connected.
Hoteliers from independent bed-and-breakfast concepts, global hotel chains, and everything in-between have faced immense challenges throughout the pandemic.