In the preclinical field, preliminary estimation of the number of experimental units to be included in a study is an important step in the setting up of the study. This estimation must be based on a dialogue between the experimenter and the statistician. This article, intended for both experimenters and statisticians, is designed to facilitate their dialogue by describing certain elements that are involved in the calculation of the optimal sample size in addition to the factors that influence their quantification. The proposed approach will result in better analysis of each experiment and, therefore, more globally, in rationalization of the use of experimental units.