A multivariate version of the classical univariate standard addition method is tested as a proof of concept for the voltammetric analysis of complex samples generating overlapped signals in the presence of significant matrix effects. The proposed strategy applies a multivariate calibration method such as partial least squares (PLS) to the full voltammograms measured for the sample alone and after combined additions of a series of standard solutions (one for every analyte). Then, a calibration model is built and further applied to the prediction of the concentration added to a blank, i.e., a full voltammetric signal measured in the absence of analytes. The absolute value of such predicted concentration is taken as the concentration of the analyte in the sample. The method has been successfully tested in different natural water samples spiked with hydroquinone and catechol and appears to be a promising tool for the analysis of overlapped signals in complex matrices.