Methods of forecasting

We will cover the following:

(a) incremental budgeting;

(b) researching;

(c) extrapolating;

(d) using 'leading' indicators;

(e) modelling;

(f) guaranteeing the figures.

(a) Incremental budgeting

It may well be that traditional budgets are all that is needed for a pedestrian business. The forecast is simply the existing figures for sales and costs altered by forecasts from others; for example, published increases in energy costs, costs of labour (the minimum wage for next year) and so on. The weakness in this approach is that the forecast and resultant budgets are self-perpetuating and incremental (with ever-increasing incremental errors!).

Budgeting almost inevitably depends on sales forecasts and, for some stable businesses with a reliable customer base, sales forecasts can be fairly reliable and thus costs reliably estimated and matched to the sales.

What drives a business? The textbook answer is: 'We are a sales-driven company.' Indeed, this is correct for the majority of businesses - if you do not chase and secure sales you have no business. But it is not always the case.

However, the principle of the 'sales-driven' mantra is indeed that sales are chased. Sales are vital, but the reality is that the business has another driver - a traditional one was production; maybe for very good reasons (survival), cash flow is chased.

In one case I spent the morning with the board of an obviously sales-driven company (it was the most innovative in its sector), trying to fathom why sales/margins were down from trend. The casual remark as we walked to lunch, 'Of course, we have to keep the six factories at full capacity', said it all!

In a growing business, in a growing sector, in a growing economy, the successful business can rightly say it is 'sales driven'. I would challenge this. If the company hits a downturn, it is likely that the business may find that it has been a 'sales follower'.

These comments are made to raise the point that if your forecasts depend on sales, do check that sales are indeed the driver.

Sales forecasting is the process of estimating what a business's sales are going to be in the future. Sales forecasting for an established business is easier than sales forecasting for a new business. Established businesses already have a sales forecast baseline of past sales. A business's sales from the same month in a previous year, combined with knowledge of general economic and sector trends, may work well for predicting a business's sales in future months. An example of this approach is given in the section on the use of graphical extrapolation.

A sales forecast could be the result of detailed research (see the section on research below) or more likely from informed opinions and views. Models can help by showing the outputs of a 'what if?' analysis. Such models are meant to be used in a balanced way, but it does seem that some users (the sales team?) get dazzled by the possibilities. Business failures can be traced back to, let us say, naive views on sales growth. A recent example is of two passenger rail franchises that failed, as they had assumptions about year-on-year passenger-number growth that were absurd - ending with more passengers travelling than existed!

Commodity costs are often very difficult to forecast. This is certainly an area where outside research or opinion should be sought. The classic exemplar of the difficulties of forecasting costs must be for the price of oil. There is an abundance of opinions and charts on the web. One thing to appreciate, not that it greatly helps annual budgeting, is that long-term price movement predictions are often more reliable - have a look at the oil, coffee or iron-ore long-term price charts.

< Prev   CONTENTS   Next >