The lack of information concerning energy flows can be seen as a significant barrier to increase energy efficiency in an industry. The extension of energy measurement systems often fails due to the difficult appraisal of expected reflux. The suggested model presents a way to overcome this barrier by forecasting potentials and including soft facts into the appraisal as well. Learning effects objectify the presented model through time and make it more stable. As a result an expected net present value can be calculated and used to estimate the benefit of installing a measurement device into a given technical system.