Analytical techniques and speciation models have been developed to characterize the speciation of Cd in soil solution. They provide an estimate of operationally defined species of Cd that need to be compared, especially for soil solutions highly concentrated in organic matter as are the solutions collected after soil rewetting. This work deals with the comparison between the speciation of Cd measured by anodic stripping voltammetry (ASV) and ion exchange and the speciation of Cd calculated using Visual MINTEQ. The aim of this study was to quantify and explain the differences in Cd speciation observed between the three approaches. Cd speciation was assessed in soil solutions collected 4, 8, 24, 48, 96 and 144h after the rewetting of an air-dried contaminated soil. To optimize the computed speciation of Cd, other physico-chemical parameters were followed (e.g. pH, ionic strength and the concentrations of major anions, major cations and dissolved organic carbon) and a brief characterisation of dissolved organic matter (DOM) was performed. The discrepancy between model predictions and analytical measurements highlighted the need for caution in the interpretation of geochemical speciated data for Cd. The major result of this study was that a characterization of DOM based on its specific UV-absorbance at 254 nm improved the accuracy of model predictions. Another finding is that labile Cd complexes, even organic, may have been included in the electrochemically labile fraction of Cd measured by ASV.