Regulatory guidance on impurities is becoming increasingly comprehensive and complex. The advent of ICH M7(R1) on mutagenic impurities has introduced a significant and sophisticated toxicological component that can easily be underestimated by the unwary. The term "genotoxic impurity" was used in guidelines that predated the current guidance but is now outdated, although it is still often (mis)used in publications on impurities. ICH M7(R1) applies only to mutagenic impurities, which are defined as compounds that are DNA-reactive and test positive in a bacterial reverse mutation assay or are predicted to do so using appropriate in silico structure−activity software. A tentative indication of mutagenic activity is provided by so-called structural alerts, which are certain electrophilic moieties within a chemical structure. It is now well-established that many conventional alerts are associated with a significant number of false-positive predictions of mutagenic potential. Consequently, caution is required when an alert is used to tag a particular impurity as "genotoxic" with no further checks. Such an approach might lead to the development of unnecessarily sensitive impurity assays, which may or may not be a deliberate choice, and possibly to wasted additional process development costs. This review is intended to provide pragmatic guidance on the evaluation of the mutagenicity status of impurities, on the basis of which it is possible to determine appropriate limits. In addition, a series of published examples are reviewed where analytical method development has been compromised by mistakes concerning mutagenicity status and where incorrect mechanistic assumptions have been made regarding the potential for the formation of particular impurities.