The addition of a solid to a mixture of small molecules introduces variations on the average translational mobility of these latter, proportional to the affinity of a given moiety for the less-mobile phase. This effect is at the basis of chromatographic separation, but can be used to simplify the NMR analysis of mixtures as well. In fact, as the induced differences in molecular mobility can span orders of magnitude, it becomes a much easier task to split the overall NMR spectrum of the mixture into one of the pure components using pulsed field gradient (PFG) methods. We have demonstrated recently this approach, in the context of HRMAS (high-resolution magic angle spinning) NMR, as required to recover high-resolution spectra in solid/liquid mixtures. In this review, we shall cover some of the principles and the practicality of this HRMAS-PFG approach for the study of mixtures. A comparison of the actual (in LC) and virtual (in NMR) separation capabilities of a few combinations of solid phase material/solvent composition shows similarities but also some striking differences: bare (not functionalized) silica expresses a superior potential for resolving spectral components than expected on the basis of the LC outcome.