Forecasting: It’s Getting Better

My business career started back in the industrial “Dark Ages”—the 1960s. One of my first promotions was to head up the company’s sales forecasting group. But don’t be misled: this was definitely not a big deal. The department was small and possessed not a lot of talent, being viewed by many as a semi-clerical group. When I told a friend that I got promoted into sales forecasting, her reaction was, “Gee, that’s too bad. What did you do wrong?”

After a short while on the job, I began to understand what she meant. Forecasting is most often the ultimate no-win game. You’re almost always wrong (except on those rare, random occasions when actual sales come in right on forecast); you get beat up routinely for your “lousy forecasts” and, unlike another non-fun activity—the once-per-year budgeting process—you must go through the forecasting cycle every month, perhaps more often.

After more time on the job, I concluded that whoever called economics the “dismal science” never had a job doing sales forecasting. It was not a lot of fun. And in some companies, things aren’t much different today.

Well, why not? After all, there’s a great deal of forecasting software now on the market, and that should certainly help, right? And we’re better educated today than we were back then, with many more M.B.A., M.S. and Ph.D. degrees available. Plus, we’ve had all those years to learn from our mistakes. Things surely should have gotten better, right?

Yes, they have. The good news is that the companies still in the Dark Ages are becoming fewer and, in an increasing number of companies, the forecasting situation is getting better. Things are improving—in some cases, quite a bit.

And it’s a darned good thing, because complexity and rate of change in most of our businesses has increased sharply over the years. It’s tougher out there today, and that’s not going to change. An increasing number of companies are making great strides with forecasting by utilizing better concepts and techniques. I’d like to explore three of them.

  • A concept called “three-legged stool.” I first heard of this from my partner, Bob Stahl, and it’s spot on.
  • Forecast Value Added (FVA), a breakthrough technique as developed by Mike Gilliland from the SAS Institute.
  • Our old friend Sales & Operations Planning, specifically Executive S&OP – the volume planning process as opposed to mix.

The Three-Legged Stool

The Dark Ages approach to forecasting was to run the statistical forecast and stop there. No human judgment, no customer input, no recognition of economic conditions. That was it: have the computer spit out the forecast, fix the items with busted tracking signals and send it to the inventory replenishment system—for better or worse. One reason for this was the oft-stated belief of many senior managers, that “You can’t forecast this business.”

To which I respond that they’re doing it already. When asked, “What do you mean?” I would reply that they do it once each year: the annual financial plan for the next fiscal year and the out years.

Today, many companies recognize three major factors—the “Three-Legged Stool”—in constructing a valid forecast, as follows:

  • The Historically Modeled View is what comes from the folks in Demand Planning. Often called Forecast Analysts or Demand Planners, they are responsible for the statistical forecasting system. They manage the data and verify that the resulting forecasts are valid based on the historical data in the system.
  • The Customer View is usually best captured by the field sales folks, because they’re the closest to the customers. They, more than anyone else, should know the customers’ business conditions, current plans and intentions for the near future.
  • The Market View is normally longer-term and is typically the responsibility of people in departments such as Marketing or Merchandising. It’s their job to factor in economic conditions, future promotions, price changes, new product launches and the like.

Once the three views are on the table, people can synthesize them and generate a consensus forecast agreed upon by all or most of the key Sales & Marketing people.

Forecast Value Added

This is the Lean Manufacturing approach applied to Sales Forecasting. Lean says that if it’s not adding value, why are you doing it? FVA says compare your current forecasts to a “naïve” forecast—for example, a simple moving average, or same as last year plus X%, or something similar. Your may find that your current forecasts are no better than the naïve ones, or perhaps not as good. If they’re somewhat better, then ask, “How much is it costing us to generate these forecasts? Is it worth the time, effort and money it’s costing?” The answer may be no.

This is a superb tool to segment the products to be forecasted, to put some “on autopilot” by using naïve forecasts, and then devote more time, effort and IQ to the ones that respond to superior forecast models and the other elements of the three-legged stool. (For more on FVA, I recommend a superb book written by Mike Gilliland, developer of the FVA process: The Business Forecasting Deal, 2010, Michael Gilliland, Wylie & SAS Business Series.)

Executive S&OP

I’ve said more than once that S&OP is forecasting’s best friend, one of the most positive things that has ever happened to forecasting. Why? Well, in many companies, forecasters can see for the first time that their efforts in developing forecasts have value, that they’re no longer being ignored. They can see that their forecasts:

  • play a major role in how production rates and inventory levels are derived—instead of being ignored
  • serve as the basis for financial projections up to and including earnings calls to Wall Street or the corporate office—instead of being ignored
  • create the foundation for simulations on how to best structure the supply chain and how to best deal with risk, both in anticipation and recovery

Yep, Executive S&OP is just about the best thing that ever happened to sales forecasting. Along with the three-legged stool and FVA, it’s making forecasters’ work lives better, more productive and, dare I say, more fun.


Tom Wallace is a teacher and writer specializing in Sales & Operations Planning. He is a distinguished fellow of the Ohio State University’s Center for Operational Excellence, and currently writes and speaks in conjunction with the Institute of Business Forecasting. He’s taught in Australia, Belgium, Canada, China, France, Great Britain, Mexico, New Zealand and the U.S.

Tom has written twelve books, including Sales & Operations Planning: the How-To Handbook, 3rd Edition (2007), Sales & Operations Planning: the Executive’s Guide (2006) and Sales & Operations Planning: the Self-Audit Workbook (2005). Tom has also produced a number of educational videos, most recently, The Education Kit for Sales & Operations Planning (2009), a multi-media teaching resource for delivering S&OP education.

Join us!

The best of BONEZONE content delivered to your inbox, twice each month.

RELATED ARTICLES



CONTACT BONEZONE

 

CONTACT BONEZONE