Understanding the metabolic answer and the adaptation of plants towards heavy metal exposure opens the way to future phytoremediation of polluted sites. For this reason, we studied the impact of cadmium, a toxic heavy metal, on the metabolome of the model plant Arabidopsis thaliana. The analytical methodology (liquid chromatography coupled with mass spectrometry) that was used within the framework of a designed experiment conducted on A. thaliana cells exposed to cadmium generated an important volume of data. Multivariate statistical analyses appeared relevant to compare the metabolic fingerprints in order to isolate and identify some discriminating metabolites. Three types of data pretreatment, i.e., reduction of dimensionality, bucketing and automatic processing by the MetAlign™ software were compared for efficiency in extracting the information. The pretreated data were then subjected to multivariate statistical analysis by principal component analysis (PCA) and partial least square regression (PLS). Finally, an OSC (Orthogonal Signal Correction)-PLS2 approach performed on kinetic and dose ranging studies allowed to visualize time- and cadmium dose-induced changes on the metabolism of A. thaliana cells.