As a business person, it’s invaluable to understand what might happen to you in the future.  But – unless your crystal ball is crystal clear, most managers must come up with a way to guess at what will probably happen.  The practice of predicting the future in a business scenario is called forecasting.

Forecasting can be extremely complex – utilizing many variables and situational factors, but good forecasting techniques can also be pretty simple and based upon common sense.  A good rule of thumb is to get only as complex as business needs dictate.  Many businesses don’t need to forecast down to the hour or half-hour, so understanding business needs prior to creating forecasting methods is key to success.

In the words of the Marquis of Halifax, "The best qualification of a prophet is to have a good memory."  The first forecasting technique is called the Time Series technique. This method requires a quantity of past data.  If one can get three years worth of historical data, he can then perform some averages based upon the nature of the specific business.  A farmer’s business is definitely cyclical based upon the seasons, whereas a grocery store manager is cyclical based upon holidays and time of month (at the start of the month, food stamp customers are in abundance).  A farmer should be able to analyze the past 3 summers to get a pretty good forecast of how much corn seed will be needed this summer.  A grocery store manager should be able to preview the previous three year’s Novembers to understand how many turkeys to order this year.

“To expect the unexpected shows a thoroughly modern intellect.”  This brings us to the second forecasting technique, which is more qualitative in nature.  This can include market research or panel consensus.  These techniques are very often used in conjunction with the time series techniques or used when past data is not available.  Market research means understanding the external variables that impact the business.  In the example above, if a drought is expected this summer, the farmer may purchase fewer corn seeds.  Or – in the grocery store example, if the main competitor went out of business, the manager might purchase more turkeys to make up for higher demand.In the event of a brand new business, many researchers send out surveys or form panels or focus groups to understand consumer needs and unmet demands.

“A good forecaster is not smarter than everyone else, he merely has his ignorance better organised.”  This basically describes the causal techniques of forecasting.  By analyzing past data with as many attributes as possible and searching for cause and effect indicators, one can better understand the outcomes.  One example of causal techniques is to look at all customers who have filed a complaint within the past year and trying to find anything that links them together.  If 85% of these customers complained about the exact same product, there might be a linkage between a product defect and customer satisfaction.  Products such as SAS and SPSS are great to help find attribute weights in relation to the primary piece of data.

Simply put – there is no one right method to forecasting.  Each business must determine which technique to use in which situation.  Many businesses use all three techniques, many times without realizing they are actively engaging in forecasting techniques. To close with one last quote: C.F. Kettering once stated “My interest is in the future because I am going to spend the rest of my life there.”