Policymakers and communities are increasingly looking to body-worn cameras to increase accountability and fix the legitimacy crisis affecting American police. Empirical research on the effectiveness of body-worn cameras is therefore an important avenue of study. Although some research shows that body-worn cameras may influence officer behaviour, there is no research examining whether officers will use the device and how usage behaviour may depend on administrative policies. Thus, the relationship between officer preferences, policies regarding camera activation, and camera use remains unknown. The current study examines whether officers’ activation of body-worn cameras depends on two different policy conditions. Integrating research on administrative policy and officer behaviour with studies of technology use in organizations, we test key hypotheses using longitudinal data for 1,475 police-citizen encounters involving 50 officers over a 9-month period. Our study yields two key findings. First, body-worn camera activation is more prevalent under a mandatory use policy relative to a discretionary use policy. Second, although camera activation declined under the discretionary use policy, this was much less likely among officers who volunteered to wear cameras. The lowest levels of activation occurred among officers who were compulsory-assigned to wear cameras. We discuss the dual role of officer preferences and administrative policy on compliance with technological innovations within police organizations.