Useful Heuristics

Imagine two managers who want to develop a real estate project. At some point during the planning phase it becomes crucial to predict the future sales prices of houses in a certain area. The first project leader, Mr O, approaches this task by searching exhaustively for all available pieces of information that he knows will influence the selling price, such as property tax, lot size, total living space, age of the house, number of bathrooms and so on. Based on his past experience, he weighs all that information according to its importance and then integrates it to predict the selling price of each house, using some statistical software. The second manager, Mr F, makes a fast decision, relying on a simple strategy based on just one single piece of information that he regards as most important, such as total living space. Which of these two managers will make a more accurate forecast? Many people, researchers and lay persons alike, suppose that the outcome of a decision can be improved by (a) an exhaustive search for information, and the integration of many pieces of information, (b) having more time to think, and calculate possible outcomes, or (c) having more computational power and the use of complex forecasting software or decision tools. This seems to indicate that Mr O will make the better decision. However, this chapter demonstrates that less can sometimes be more, and that a strategy relying on very few pieces of information, and quickly deriving a decision based on a simple algorithm, may well outperform more sophisticated, supposedly rational decision strategies.