Calculating Demand
The purpose of calculating demand is to establish a baseline for forecasting future demand. Whereas data on workforce supply is typically in plentiful supply within an organisation, data on demand is often scant and fragmented. This delta, and the complexity of the subject, is why we conduct a baselining activity as part of the demand stage rather than as part of the baseline stage. If we are operating at a macro level, there would be little disadvantage to conducting this exercise earlier. At a meso and micro level, we risk the economy of our effort by collecting and analysing too much demand data before we understand fully the nature of the workforce supply who will be servicing that demand.
Last month’s web page on the workforce performance model gave us a clear framework that a greater level of resource is required than is indicated by volumes and processes when we convert from contract time to productive time. The product of labour is based on the productivity of a single worker, not the productivity of the productive time of a single worker. This provides us with two approaches that are output based or input-based.
An Output-Based Approach
This approach seeks to identify the relationship between the workforce and the outputs they produce, which are known as products of labour and illustrated in the following diagram:
The average product of labour (APL) is the average individual output for the workforce. For 10 workers producing 100 units, the APL is 10. The APL increases at a slow rate as the organisation benefits from the advantages of additional workers. These economies of scale reach a peak and then become diseconomies of scale as average output improves. The 10 workers with an APL of 10 may operate a production line where increasing hands allow the average output to increase. When that team reaches maximum capacity for space, and workers start bumping into each other on the production line, the APL starts to reduce. Whereas APL is the average, MPL is the marginal product of labour: the change in output from adding an additional worker. Initially, there are increasing marginal returns, the benefit of adding a worker is greater than the benefit of adding the previous worker. This quickly hits a peak before descending into diminishing marginal returns where each additional worker continues to increase the overall output but made less of a difference than the last. There are two key factors in play at this point: communication and divisibility.
- As more workers are added to the process, there is a polynomial growth in lines of communication as more people need to pass information to each other. This combinatorial explosion in demand erodes the benefit created by the marginal return.
- The second factor, divisibility, is that some activities are not easily divided and so the addition of workers does not add the same benefit as the initial workforce. At the end of the MPL curve, we hit the worst case, diminishing returns, where the organisation is saturated and the addition of an extra worker results in a decrease in output. This is the point, perhaps, where the organisation reaches the Malthusian catastrophe of having too many workers for the number of machines or space they have, and work suffers as a result. 2
Aligning high-level output data with worker levels will allow us to map a number of points on both the APL and MPL curves to understand current levels of derived demand for workers and how that varies.
The benefits of this approach are that the data on outputs is often the most readily available and is much easier to calculate. The limitations of this approach are that it does not connect with input data and is less useful at meso and micro levels of the organisation.
