In Australia, academic workloads are increasing, and university funding is decreasing. Academics and university managers are engaging in risk adverse behavior and tending to focus on customer satisfaction and student retention, potentially at the expense of academic standards. Conventional approaches to pedagogy minimize adverse student feedback, but may not prepare graduates for the workplace (Savery, 2006). By contrast, PBL (problem-based learning) is generally thought to produce better student outcomes and performance, but can be more demanding in terms of academic workloads. PBL uses realistic scenarios which may not be well defined and the methods students use to solve the problems are often as important as their answers. PBL is grounded in constructivism (Ernest, 1993) and student-centered learning (Ally, 2004). This paper critically evaluates the claims made for PBL and provides arguments for its use in university teaching. Two research approaches were employed to inform this paper: the first is a meta-level review of PBL-oriented papers and the second is hermeneutic phenomenology in which the author gives an account of his experiences using PBL in university education and makes recommendations for its deployment. It is concluded that there are no academic or logistical grounds that support the use of PBL, but that it may still be regarded as a worthwhile exercise.