Sales forecasting practice in small firms: the application of Bayesian theory to strategic decisions

This paper examines the sales forecasting practices of small firms, defined here as firms with no more than 50 employees. This paper discusses the application of Bayesian decision theory in the production of sales forecasts for small firms particularly in relation to longer term strategic decisions. A specific case study based on the author’s consultancy experience is used to illustrate the application of the procedure. The management of many firms feel unable to use formal objective forecasting techniques because of lack of information to start the forecasting procedure off. For example there may not be any historical data available or even if there is management may not have any objective probabilities in relation to initial starting conditions. As discussed in this paper evidence from the authors own work suggests that the management of many small firms make no formal sales forecasts at all. Where formal forecasting procedures are used by the smaller firms they often tend to be subjective methods based on the managers own experience or the collective experience of others. A more robust procedure is available which overcomes the lack of initial starting conditions and this is based on Bayesian decision theory. Such a procedure should be well within the competence level of the majority of small business managers. A computer can carry out the calculations and the basic principles of Bayesian forecasting procedures are relatively easy to grasp and apply by small business managers.