As demand planners, have we added value through forecasting? Is your forecasting effort efficient?
You can answer in the affirmative, if your diligent efforts in demand planning has resulted in improving forecast accuracy. Most best practice companies measure accuracy trends or evolutions. They reward planners for improving forecast accuracy from a base level – either last year corresponding month or an average baseline accuracy over last year.
From a planner’s perspective, let us say you are evaluating the power of a forecast model. How do we determine if this model or a forecasting process is particularly better than a baseline. One of the measures used is the Geometric Mean Relative Absolute Error (GMRAE) which measures the improvement of the forecast over a naive model.
In this context, the Forecasting Efficiency quotient simply measures forecasting improvement over the average as the baseline forecast. If you use a simple arithmetic average as the forecast, then the forecast error is the Co-efficient of Variation (CV).
The Forecasting Efficiency Quotient (FEQ) = (CV – WAPE)/CV.
The FEQ measures how much of an improvement you were able to attain with your forecasting over using just the simple average as the forecast over the same period of time.
More detailed illustration is available in the Demand Metrics Template at