Over the last decade, much research has focused on the potential health benefits of antioxidants and indeed many synthetic and natural compounds have been evaluated for their antioxidant profile. However, in several studies only a limited number of assays, often poorly validated, are used and the techniques available frequently lack specificity. These limitations may incorrectly influence the results. This review will therefore focus on several pitfalls that may emerge in vitro and in vivo antioxidant research. First, different in vitro techniques to determine antioxidant potential are discussed, including radical scavenging assays and fingerprinting methods. As a rule, a panel of different assays is indispensable to characterize and establish in vitro antioxidant activity. Furthermore, as problems of absorption, distribution, metabolism and excretion are only accounted for by in vivo studies, the need for in vivo antioxidant research is pointed out. Several methods to characterize the in vivo activity of antioxidants, including major drawbacks and pitfalls of some assays, have been discussed. The availability of both a representative “oxidative stress” animal model and a battery of well-validated assays to assess the broad diversity of oxidative damage and antioxidative defence parameters, are crucial for antioxidant research in vivo.