Commonly encountered problems related to modelling bacterial growth in food are analysed from a mathematical point of view. Modelling techniques and terms, some misused, are discussed and an attempt is made to clarify how, and under what conditions, they may be used. A theoretical framework is given to provide a basis in which mathematical models having been used in predictive microbiology can be embedded. By using several simplifying idealizations as a compromise between the complexity of the biological system and the available data, a practically usable model becomes available.