Better Forecasts, Better Plans, Better Results: Enhance the Validity and Credibility of Your Forecasts by Structuring Them in Accordance with the Five Different Ways People View the Future

Article excerpt

Society's long-term ambivalence about projections of future developments is illustrated by the intriguing fact that at the time Greek city states were basing their military strategies on the predictions of the Oracle at Delphi, the Roman Emperor Justinian decreed the death penalty for anyone engaged in forecasting. (Given Rome's subsequent conquest of Greece, it doesn't appear that the entrails-readers provided much help.)

In reality, of course, all important business decisions are based, in large measure, on how decision-makers foresee developments in market demand, competitive threats, new technologies, financial realities, regulatory restrictions, social mores, and a host of other influencing factors.

The key question, therefore, is not whether executives use forecasts, but rather, how they formulate their views of the future and how they act on these views. The value that decision makers place on formal forecasts--and the extent to which they act on these forecasts--depends on their conviction that credible data, treated in a logical manner, support the forecasts: credible and logical in the mind of the decision-maker, not necessarily in the mind of the forecaster. Therefore, to be useful, a forecast must be both valid--based on solid facts and proven analytical techniques--and credible, meaning convincing to the people making decisions.

The term "valid" instead of "accurate" is used deliberately here. Although all forecasters would prefer that their projections be accurate, actually, the value of a forecast depends not on its specific accuracy but, rather, on the extent to which it contributes to better decision-making. In forecasting, being approximately right is always better than being precisely wrong. In fact, the best forecasts are often never borne out because the decision makers have acted to prevent unfavorable predictions from happening.

Structuring forecasts in accordance with the different ways that people view the future can enhance both the validity and credibility of the forecasts. These views can be classified into five categories:

* Extrapolators.

* Pattern analysts.

* Goal analysts.

* Counter-punchers.

* Intuitors.

Each of these views has both strengths and shortcomings that motivate their use under certain circumstances and discourage their use in others. There are specific methods and techniques recognized in the forecasting field that can be associated with each category, and each has its supporters and detractors. Almost everyone feels most comfortable with one or another of these views. Although there are no hard-and-fast rules, it seems that most engineers are basically extrapolators, most pure scientists are primarily pattern analysts, and most marketing people are goal analysts. The majority of executives appear to be intuitors, although, interestingly, many characterize themselves as counter-punchers.

These five approaches, individually and in concert, can provide the foundation for a powerful forecasting program. Following are descriptions of these approaches and some associated techniques, along with illustrations of their usefulness.


Extrapolators believe that the future will represent a logical extension of the past. They believe that large-scale, inexorable forces will drive the future in a continuous, reasonably predictable manner and, therefore, the future can best be forecast by identifying past trends and extrapolating them in a reasoned, logical manner.

Extrapolators normally base their forecasts on straight-forward logic that is easily reproducible and that usually provides quantitative results. However, this approach often fails to take into account the fact that changes in driving forces can result in rapid and dramatic changes in trends. For example, the explosion in Internet usage in the 1990s drove the rapid expansion of e-commerce-based businesses. …


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.