The police are widely recognized as an important deterrent of crime, with investments in policing broadly associated with lower crime rates. Much less is known about how investments in policing contribute to crime reductions, or the relative merits and risks of specific activities in which officers engage, leaving policing as an area in which benefit-cost analysis methods stand to make a significant substantive contribution. However, the implementation of these methods involves challenges related to measuring both the quantity and quality of policing involved in a given dosage of policing, as well as the causal effects of police practices on crime. This essay lays out several of the challenges inherent in understanding the benefits and costs of policing practices, with a specific eye toward practices commonly known as “proactive policing” practices, and their effects on crime rates. I lay out potential strategies for resolving these challenges, which focus largely on identifying exogenous discontinuities in policing practices that can be used to assess outcome differences. Thorough assessment of policing practices should apply as many of these strategies as possible, in order to understand how robust or sensitive substantive conclusions may be to strategies used.