There are a number of different methods for labour demand forecasting that can be used. The best labour management software will often take into account several of these methods, for instance, using historical analysis and managerial judgment.
Historical analysis is the process of referring to staffing records from previous months to create an upcoming schedule.
- Which days or times does the company experience the most customers?
- Which days or times are the slowest?
- Are there specific seasons when the store becomes busier or slower?
- What holidays or events affect customer demand?
- Has the specific venue being forecast had an increase or decrease in activity?
These models can often be run on a spreadsheet.
The Delphi model
This forecasting method involves having key stakeholders answer anonymous surveys about labour needs in the future. The collective responses are used to make the final scheduling decision. This method is recommended for long-term schedules or forecasting total demand for a large number of stores or departments.
This is a process involving multiple dependent variables resulting in one outcome. For example, taking into account historical sales data, weather forecasts, expected promotional offers and local events to predict the number of transactions a business will do in a given period in the future. Often done using machine learning or AI-powered tools.
Local management will often have a good understanding of the staffing level required for their store. They can often highlight issues with forecasts generated by the other methods mentioned in this blog post. Management teams also work directly with customers and staff members and will have a deeper knowledge of their team members preferences and needs. That is why even with the most sophisticated AI-powered forecasting and auto-scheduling tools, the local manager gets to have the final approval and is given the ability to make manual changes to the rota built by the AI.
Often workforce management solutions come with tools that take in historic sales data, already include historical employment and shift records and can be used to predict the level of staff required to meet customer demand.