The issue of herb-drug interactions has generated significant concern within the pharmaceutical industry and among regulatory authorities in recent years. Therefore, accurate models of predicting metabolic herb-drug interactions would be useful tools in efforts to avoid toxic adverse events. However, the majority of pharmacokinetic interactions listed for herbal medicinal products are based on theoretical predictions of the in vitro pharmacological effects of known constituents, which do not necessarily have to be the active ingredients. The prediction of herb-drug interactions is further complicated by the fact that pharmacokinetic data on active or (at least) known ingredients are often not available. The present article discusses the potential of pharmacokinetic profiling for detecting herb-drug interactions, using the most frequently cited interactions in the literature as examples. In particular, common mechanisms of herb-drug interactions are summarized, and the available experimental methods for detecting such interactions, as well as the limitations of these models, are critically evaluated. In addition, we discuss the question of whether the existing methods of detecting herb-drug interactions correlate with the clinical relevance. Effective screening tools that accurately predict metabolic herb-drug interactions would offer a tremendous advantage because it is not possible to study all potential herb-drug interactions in clinical trials.