AI Forecasting: Identify the insights that drive profitable actions

Spend just a few seconds on the internet these days and you’re bound to stumble on something to do with artificial intelligence (AI). The darling of, well, pretty much everyone, AI has captured the collective imagination. From website recommenders to advert optimisers and image generators, there’s nothing it can’t do.

AI, in the form of Machine Learning, is a computer programme that uses historical data to predict likely outcomes and what the next action in a sequence might be; continually learning and improving as it receives new data. Which makes it the gold standard when it comes to forecasting.

AI forecasting for restaurants

AI analyses huge, varied data sets to identify patterns and correlations we might not have spotted before. As such, restaurant operators can use this tech to analyse past data on staff availability, weather, sporting events, and demand – among many other factors – to predict revenue for any given shift, on any day, week or at any restaurant location.

The intelligence part of AI comes from its ability to learn nuances, meaning an AI forecaster can not only use a restaurant’s past data to predict future revenue, but it can also learn the exact impact a sporting event on a rainy day might have on demand in one location, compared to a sunny day in another. In short, it can spot the trends that inform the actions you take to keep your business profitable.

Why do restaurants need AI forecasting?

The last few years have seen the world turned on its head. Supply chain issues, soaring energy costs, and the impact of rampant inflation on consumer spending have put extreme pressure on operators throughout the hospitality sector. Demand is increasingly volatile, and understanding it can be tricky.

Manual predictions take a huge amount of managers’ time and require them to consider everything from local events that might increase footfall to staff availability throughout the week. AI forecasting streamlines this process, allowing for almost instant analysis of datasets that would be too big for a single person or team to fully assess. This not only increases the efficiency of forecasting, but also the accuracy, since AI forecasts can take more factors into account than a manual prediction, at a scale humans cannot match.

Using AI forecasting to drive profitability

Revenue forecasts are the cornerstone of operations, and form many of the rules and guardrails that define business decision making. If you expect to take a certain amount on a Sunday lunch shift, you’ll schedule enough staff to cover that anticipated demand and match inventory accordingly.

An inaccurate forecast can, therefore, lead to excessive labour costs and wastage if shifts are quieter than expected, or missed sales if demand is underestimated. AI forecasting allows operators to strike the right balance, providing accurate forecasts that staff are confident of, and providing insight that can then be leveraged in other areas of the business:

• Financial planning: A more accurate and nuanced revenue forecast increases the accuracy of financial projections and budgeting, enabling better financial planning.
• Demand forecasting: Understanding likely revenue, and how that will fluctuate based on time of year and external factors such as weather or local events, means restaurants can anticipate demand at an item level, scheduling extra bar staff and reducing kitchen staff if beverages are expected to be the main revenue driver for a particular shift.
• Staff scheduling: With demand understood at a granular level, managers can match staff rotas to it precisely, creating efficiencies in labour costs without overburdening and burning out employees.
• Inventory management: Revenue and subsequent demand forecasting allows operators to optimise inventory, reducing wastage while ensuring popular items don’t sell out and become unavailable.

AI forecasting with Fourth

AI forecasting forms the basis of Fourth’s Intelligent Scheduling. It’s trained on one of the industry’s biggest datasets, with over 160k shifts scheduled every day, meaning it can deliver highly accurate forecasts almost instantly.

These revenue forecasts form the cornerstone of Intelligent Scheduling, providing insight into exactly what actions are most profitable for a business. This in turn can be leveraged to determine inventory and the optimal shift pattern for staff – how many people trained in each specialty will be required to meet the anticipated demand for every shift. In this way, AI forecasting increases productivity, reducing both food waste and limiting the instances of over or under staffing.