The existence of extensive and updated datasets is useful for model calibration and validation. In this work, the CS-Model was applied at IJmuiden coastline, The Netherlands, evaluating the evolution of berm and dune toe positions by comparing the model results with a long series of observations (1979–2020). The model was calibrated considering the first 11 years of existing data and the model performance was evaluated every decade, considering the simulation of 2 scenarios: using the initial conditions; and updating the profiles characteristics, according to the observed positions along the decades. The results indicated better model performance when the profile characteristics are updated every decade, with the berm and seaward dune toe positions projections being nearest the observations.