Every new product development (NPD) project should not only deliver a successful new product but also should generate learning for the organization. Postproject reviews (PPRs) are recognized by both practitioners and academics as an appropriate mechanism to stimulate and capture learning in NPD teams. However, relatively few companies use PPRs, and those that do use them often fail to do so effectively. Although they are widely perceived to be a useful tool, empirical research on how PPRs are typically organized and the learning that results is limited. The present article addresses this gap in the extant knowledge and describes five in‐depth case studies, which were conducted at leading companies in Germany. A detailed investigation was made of how PPRs are conducted and of the type of learning that can result. Three main sources of data were used for each case: company documentation, in‐depth interviews with managers responsible for NPD, and observation of an actual PPR. The different data sources enabled extensive triangulation of data to be conducted and a high degree of reliability and validity to be achieved. The analysis enabled a number of key characteristics of the way PPRs are managed to be identified. Various characteristics of PPRs influence their utility, such as the time at which they take place and the way discussions are moderated. In addition, the data show that participants in the discussions at PPRs often use metaphors and stories, which indicates that PPRs have the potential to generate tacit knowledge. Interestingly, the data also show that there are various different ways in which metaphors and stories appear to stimulate discussions on NPD projects. Based on the cross‐case analysis, a wide range of implications are identified. Researchers need to investigate PPRs further to identify how they can generate tacit and explicit knowledge and support project‐to‐project learning. The generation of tacit knowledge in NPD is a topic that particularly needs further investigation. The research also led to a range of recommendations for practitioners. Companies need to strongly communicate the purpose and value of PPRs, to run them effectively to stimulate the maximum possible learning, and to disseminate the findings widely. PPRs have the potential to create and transfer knowledge amongst NPD professionals, but, as they are seldom currently used, many companies are missing an important opportunity.