Academic journal article The Journal of Business Forecasting Methods & Systems

New Developments in Business Forecasting

Academic journal article The Journal of Business Forecasting Methods & Systems

New Developments in Business Forecasting

Article excerpt


(This is an ongoing column in The Journal, which is intended to give a brief view on a potential topic of interest to practitioners of business forecasting. Suggestions on topics that you would like to see covered should be sent via email to Ed)

Several issues back I wrote a column that extolled the virtues of collaborative forecasting pointing out that because "two or more heads can be better than one" it could lead to greater forecast accuracy, as well as organizational buy-in of forecasts. However, two or more heads might not be better, especially if those heads are "butting heads" over whose forecast is right. I stated that in collaborative forecasting a consensus approach needed to be set up in which a reconciliation process had to be established to develop a single forecast from several. The multiple forecasts might be obtained via different sources, such as from statistical baseline forecasting techniques, the marketing and sales departments, customers, and suppliers. ( See Figure 1). However, I failed to provide advice on what methods could be used to reconcile multiple forecasts. Essentially to help address two questions. Whose forecast should you trust? How might you consolidate forecasts? Herein, I offer some suggestions and opinions on these two questions.


The answer to this question depends on what portion of a set of demand forecasts you are considering. Some organizations understand various aspects of a forecast better than others. Below list the different types of possible sources and the aspects in which their forecasts are better. It is about these aspects that these sources should have the most credible views:

Marketing: This department is best able to understand a forecast at a national rather than regional level, since they have little to no visibility into local market dynamics. Since it is also responsible for new product introductions and promotional/pricing programs it is likely to have a better understanding of promotional uplifts and new product cannibalizations that might impact future demand. In addition, this department is often better at long-term rather than short-term forecasts.

Sales: As `eyes and ears to the market,' this department has the best understanding of local market dynamics, such as relating to individual sales accounts or local competitive environments. For example, Sales is better able to know which local customer accounts are growing faster and what competitors are coming into a local market. For these reasons it best understands the trends taking place on a geographical and customer account basis. For example, a sales rep would know that a competitor is opening a warehouse in an area, likely to steal business from your company, and would know that a retailer is opening more stores in an area, impacting your sales. This department is often better at short-term rather than longterm forecasting.

Statistical Baseline: This refers to the forecast produced by quantitative methods that is used as the baseline from which to develop a final forecast. It is obtained from projecting past demand patterns and includes the quantitative impact of planned events such as pricing changes, promotions, and other events that might significantly impact demand. The value of this forecast stems from its being an objective, unbiased, and unemotional view of the future. As such, it often provides a good `sanity check' to other forecasts to ascertain whether they are reasonable and to determine `real' factors that cause them to differ. The statistical baseline forecast, however, is rarely the last word, since it cannot incorporate every factor that might impact demand.

Customers: These are forecasts provided by customers on their anticipated demand. Since most businesses work on the premise that "the customer is always right" it is hard not to believe a customer does not know its own needs as well as you. …

Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.