OVERVIEW: This technique is part of a set of techniques that are useful in situations where past data do not exist, causal relationships have not been identified, or some major change has occured in the forecasting context which is not accounted for by other techniques (such as a Gulf War, trade agreement, etc.). Evidence as to the validity of using these methods by themselves is mixed, although using them correctly can provide very good forecasts, especially in uncertain environments. The objective of these techniques is to provide logical, unbiased, and systematic quantitative estimates.
BASIC IDEA: Combines input from key information sources.
PROCEDURE: Very simple. Have "forecast meetings", perhaps beginning with initial estimates of the forecast variables. Discussion ends when consensus is reached.
EXAMPLE: Mainframe computer forecasting is done by conducting a series of meetings between the two mainframe analysts at a company, the Service director, and a Research Operations analyst. Typically, three or four meetings are required in order to arrive at a forecast consensus. In between meetings, the forecasts are examined by colleagues, both domestic and abroad, for feedback and reaction.
COMMENTS: This approach is used widely because it is relatively quick and uses the solid understanding of participants who are presumed to have relevant knowledge. However, this approach is subject to severe judgmental biases that can result in poor forecasts. The next technique specifically addresses these potential pitfalls.