The subject of metabolites in safety testing has had much debate in the recent past and has shown itself to be a complex issue with no simple solutions to providing absolute assurance of drug safety. Much of the attention has focused on the ability to identify metabolites and then demonstrate that their risk has been adequately characterized, either through their exposure in toxicology species or, failing this, by direct safety testing. In this review, we summarize our forward operational strategy that combines the principles summarized in the FDA Guidance, together with discussions at scientific meetings and literature opinions. It is a balance between the primary goal of assuring patient safety with one of reasonable investment. A key principle in striking this balance is to build stepwise information on metabolites through the drug discovery and development continuum. This allows assessments to be made from early nonclinical studies onward as to whether or not metabolite safety is underwritten by exposure in toxicology species. This strategy does not require absolute quantitation of the metabolites in early clinical trials but relies upon comparison of relative exposures between animals and humans using the capabilities of modern analytical techniques. Through this strategy, human disproportionate metabolites can be identified to allow a decision regarding the need for absolute quantitation and direct safety testing of the metabolite. Definitive radiolabeled studies would be initiated following proof of pharmacology or efficacy in humans, and nonclinical safety coverage would be adequately assessed prior to large-scale clinical trials. In cases where metabolite safety is not supported through the parent compound toxicology program, approaches for the direct safety testing of metabolites with regard to general and reproductive toxicology, safety pharmacology, and genetic safety have been defined.