Presumptive drug testing is commonly used in both the clinical and forensic fields to allow rapid identification of the presence and/or usage of drugs. Because the tests generally have a high sensitivity and specificity (often >90%), then a positive test result may be taken to mean there is a high probability that a targeted drug is present. This assumption is, however, incorrect. This paper demonstrates how, in order to assess the positive predictive value (PPV) of a test, it is necessary to take into account, along with the sensitivity and specificity of the test, the prevalence of the drug in the population being investigated. We demonstrate how an alternative, Bayesian approach to assessing the posterior probability of a drug being present mimics the conventional calculation of PPVs but, because a Bayesian approach requires case-specific prior probabilities, the posterior probabilities are more meaningful than PPV in any one specific case. The effectiveness of presumptive test results in cases such as drink-spiking, drug-driving, testing of drugs during seizures and the confirmation of initial presumptive test results is explored.In order to exploit the potential of presumptive drug testing, it is important that the prevalence of the targeted drugs in relevant populations is understood but, more importantly, it is important to consider using a Bayesian approach in order to tailor results to the specific individual or drug batch being tested.