Many people, who are interested in sales forecasting, are familiar with the book Predictably Irrational by Dan Ariely. The abstract of the book states: “Dan Ariely refutes the common assumption that we behave in fundamentally rational ways. From drinking coffee to losing weight, from buying a car to choosing a romantic partner, we consistently overpay, underestimate, and procrastinate. Yet these misguided behaviors are neither random nor senseless. They're systematic and predictable—making us predictably irrational.” The irrationality of human decisions is the basis of behavioral economics. Predictably Irrational, and many other books, give many examples which argue the idea of consumer rationality. Businesses are irrational too, as they are guided by human beings. Do these theories affect sales forecasting methodologies? I feel that today they don’t, because often the decision makers are either economists, or hired economists, who were educated long before behavioral economics became well known. Results of classic marketing research present a static situation. So then, how can the future behavior of consumers be forecasted? The classic researchers extrapolate trends in the future. Their opinions are based on current situations and previous experiences. How can one be sure that extrapolation of rapidly changing trends will make a precise forecast? You are probably wondering why I am writing about this in the AnyLogic blog? It is because Agent Based modeling is a great way to present the markets in dynamics. If you understand the logic of the decision making of a particular consumer, then, using Agent Based modeling, you will be able to recreate the system as a whole, simulating the decisions of hundreds or thousands of independent consumers. Let me present some materials on Agent Based modeling in marketing that we have collected on our website: Case Study: Simulation of Telecom Market in Argentina Article: The Impact of Human Decision Makers’ Individualities on The Wholesale Price Contract’s Efficiency: Simulating The Newsvendor Problem
