We present a comparison between the sum frequency generation (SFG) and the difference frequency generation (DFG) at the electrochemical interface. We show that the phase matching conditions are not always satisfied for DFG and analyse the experimental difficulties of DFG. We then present a modelling of the second order non linear susceptibilities (both in SFG and DFG) at the electrochemical interface. As this modelling gives two mathematical solutions of the fit of the experimental data, we demonstrate, thanks to experimental results obtained for cyanide on platinum, that DFG could be the essential complement of SFG for a quantitative analysis of the experimental data. In the particular case of noble metals for which the visible beam lies in the spectral range of interband electronic transitions (case of CN- on gold system), we show the possibility to overcome the obscuring of the signal by the non resonant contribution of the substrate, by switching from SFG to DFG.