Abstract: A systematic study on effective interaction depth (EID) theory is conducted in an effort to achieve parametric normalization for peak count rates taken from different counting positions. The results show that the errors introduced by EID prediction are generally less than 3% for analytical peaks with neg-legible summing effect. S0(Eγ) functions for two Ge detectors have been established and successfully used for 3 years. With the use of EID law, each sample (or standard) can be measured at its most suitable position (not necessarily at the same position for all the samples and standards in one batch). As a result, comparative analysis of activated samples and standards with extremely different gross radioactivities can be easily accomplished with no cost in precision and counting time.