The performance of reed bed systems is usually presented following a long-term evaluation of case studies. However, short-term water content analyses may be useful to better understand the dynamics of the dewatering process and hence to rationally describe the system, with the ultimate aim of improving the system overall. In this study we performed daily water content analyses in the interval between subsequent sludge loadings in a real-scale reed bed system situated in Pisa (central Italy). Water content analyses were performed at different periods of the year: in winter, drainage quickly led to a maximum of about 8% dry matter while in summer the dewatering process significantly continued after the first week thanks to evapotranspiration. A comparative analysis of planted and unplanted beds confirmed that the mechanical function of plants is essential in the winter, when the unplanted bed did not present any dewatering capability. A semi-empirical equation describing the dynamics of the dewatering process is then proposed and calibrated. This equation forms the basis for the technical and economical optimization of the resting time in terms of required surface area and drying efficiency. An optimum of I I days resting time was finally evaluated in the presented case study.