It is generally accepted that bioactivation of relatively inert functional groups (toxicophores) to reactive metabolites is an obligatory step in the pathogenesis of certain idiosyncratic adverse drug reactions (IADRs). IADRs cannot be detected in regulatory animal toxicity studies and, given their low frequency of occurrence in humans (1 in 10,000 to 1 in 100,000), they are often not detected until the drug has gained broad exposure in a large patient population. The detection of IADRs during late clinical trials or after a drug has been released can lead to an unanticipated restriction in its use, and even in its withdrawal. To date, there is neither a consistent nor a well-defined link between bioactivation and IADRs; however, the potential does exist for these processes to be causally related. Thus, the formation of reactive metabolites with a drug candidate is generally considered a liability in most pharmaceutical companies. Procedures have been implemented to evaluate bioactivation potential of new drug candidates with the goal of eliminating or minimizing reactive metabolite formation by rational structural modification of the lead chemical class. While such studies have proven extremely useful in the retrospective analysis of bioactivation pathways of toxic drugs and defining toxicophores, their ability to accurately predict the IADR potential of new drug candidates has been challenged, given that several commercially successful drugs form reactive metabolites, yet, they are not associated with a significant incidence of IADRs. In this article, we review the basic methodology that is currently utilized to evaluate the bioactivation potential of new compounds, with particular emphasis on the advantages and limitation of these assays. Plausible reasons for the excellent safety record of certain drugs susceptible to bioactivation are also explored. Overall, these observations provide valuable guidance in the proper use of bioactivation assessments when selecting drug candidates for development.