An approach to enterprise revenue forecasting as a decision support system

Abstract

Successful competitions in the marketplace require the ability to make accurate management decisions, which often call for a decision support system. In the finance function of an enterprise, a significant portion of decision-making involves quarterly revenue forecasts and corresponding gap analyses updated on a weekly basis. This paper describes a statistical framework based on an enterprise quarterly revenue forecasting system that provides information directly relevant for management decision-making. This approach utilizes a statistical framework that can issue accurate metrics-based forecasts and can delineate the influences of specific business drivers. Some of the drivers may be acted on by the enterprise, while others may represent the business environment and are unactionable. Built on such a framework, analytics can be developed for specific business functionalities such as performance gap assessment, root-cause investigation, and possible action suggestions to achieve an objective. We demonstrate the value of our framework on a real-world problem: forecasting IBM's corporate revenue.