Over the past 10 years, I’ve noticed a growing dialogue around the subject of forecastability. It seems to be one of the hot topics these days, and represents what some believe to be futuristic thinking—the next step-change advancement—in the field of demand planning. I recently attended the latest APICS/IBF Best of the Best S&OP conference where the topic surfaced again in side conversations and hallways. I have always been hesitant to comment in such moments, because I believe the exercise of understanding and learning about demand attributes is worthwhile regardless of whether or not I agree with the wisdom of a concept. For some reason, this time I came away with a different perspective—I wondered whether it was time to answer the question that has been on my mind for a while now: Is forecastability a real thing?
Every forecasting professional knows that there are no “perfect” forecasts, but there is a way to see if you are getting it right. There is a better way to measure what forecasting can be expected to achieve and what it might take to achieve it. By utilizing forecast value added (FVA) analysis, one can determine the effectiveness of an organization’s forecasting efforts and help streamline its forecasting process. In this article, I will discuss FVA as a forecasting key performance metric, and explain how it can help organizations and forecasting professionals meet the primary objective: to improve forecast accuracy, while adding value to the process and the organization.
As demand planners and supply planners, we are committed to supporting the business decisions that are essential to making the company successful, making growth possible, and creating value for all of the company’s stakeholders.
I recently attended a John Galt Systems User’s Group event to train and speak. Before leaving the next day, I attended a morning session where the company discussed its Forecastability Analysis that provides information on how well various methods are able to forecast items in a product line. The analysis, for example, identifies high versus low “forecastable” products. The presenter showed a two-by-two matrix that is similar to the Alan L. Milliken (from BASF) matrix described in my Fall 2009 Journal of Business Forecasting (JBF) column, titled “Volume-Variance Analysis.” That column depicted a vertical axis labeled Sales Volume versus a horizontal axis labeled Variability of Sales (co-labeled: Coefficient of Variation = [(Standard Deviation of Period Sales or Forecast Error)/(Average Period Sales)]). BASF uses it to apply differing inventory management strategies depending on what quadrant a product falls into when the matrix is split into four High- Low quadrants.
There has always been considerable debate about the best forecast metric to use. All forecast metrics have a place, and there is not a universal perfect fit measure of forecast performance. The author describes 10 guidelines that should be used when considering the most appropriate metric for your organization.