The way most companies evaluate forecasting performance tells them the magnitude of their error, but does little to identify causes of the error or potential for improvement, says Michael Gilliland of SAS. Gilliland explains how the addition of a few simple analytic tools can provide fuller and more useful evaluations.