The use of precursor ion and constant neutral loss scanning as a means of rapidly detecting drug metabolites is evaluated. Four clinically useful drugs, namely (i) cyclophosphamide, (ii) mifentidine, (iii) cimetropium bromide and (iv) haloperidol, were subjected to microsomal incubations to afford phase I metabolites. Aside from a minor clean-up procedure involving zinc sulfate precipitation of microsomal proteins and solid-phase extraction of metabolites using a Sep-pak C-18 cartridge, the mixtures were analysed directly by fast atom bombardment tandem mass spectrometry. It is demonstrated that such screening strategies are important in detecting novel metabolites. However, there are some problems associated with only using such methods, including (i) the possibility of not detecting metabolites that undergo unusual collision-induced dissociation fragmentation pathways, (ii) the non-detection of metabolites that have undergone metabolic change at unusual sites of reactivity, and (iii) production of artifacts derived from the parent drug by the primary ionization process. Examples are discussed that highlight both the strengths and weaknesses of such an approach.