Admission testing is an integral part of interpreter training, yet it is surrounded by much controversy and scepticism. This paper first looks into some general principles of designing an admission test (its purpose, link with curriculum, effectiveness and efficiency of testing), emphasizing the dissociation between skills to be tested and tests used for tapping them. The authors then present data from a small-scale study in which 18 interpreting schools provided information on their admission testing practice and answered questions related to the skills tested and tests employed, and to the overall efficiency of their admission procedure. Based on this analysis, they conclude that there is a reasonable consensus among schools in terms of which tests are best suited to test which skill, but that new approaches to admission testing are needed to improve their efficiency. More specifically, the authors suggest that schools improve their recording systems, that soft skills are included in formal testing, and that new testing methods are explored.