Abstract
Co-production, or integrating a range of perspectives from researchers and non-researchers into the knowledge development process, is considered valuable for increasing the potential that scientific results will be applied by practitioners. While the literature around co-production often touts the benefits of such work, there is less focus on the potential costs of co-production, and discussions often lack nuance about the specific meaning of co-production in a particular context. In this perspective essay, we use an example co-production process focused on the development of a science agenda for a federal research program to consider the ideal of co-production. Specifically, we reflect on the appropriate level of non-researcher involvement throughout the full cycle of research, and position our process within the diverse range of existing co-production approaches. We suggest that the ideal of co-production is not necessarily one that integrates the maximum amount of non-research involvement throughout the full cycle of research at all costs, but one that focuses on mitigating the research-management gap while limiting the risks to those involved.