There is little consensus among scholars on whether targeted killings of members of terrorist and militant groups work, though some have argued that they do, at least under certain circumstances. Most of the work so far has focused on the ability of targeted killings to disrupt targeted groups’ ability to function. In many cases, work has centered on whether they work and not addressed how they work if they do. There has been insufficient study of the coercive effects of targeted killings and these studies have often produced mixed results, with violence sometimes increasing and sometimes decreasing after targeted strikes. This article argues that the focus on disruption and a failure to differentiate between deterrence and compellence and properly to account for the timing of attacks may be at least partly to blame for the mixed record observed from the use of targeted killings. It asserts that targeted killings are inherently compellent and can therefore only be used to change a status quo and potentially establish new “rules of the game.” They cannot be used to deter (i.e. to maintain current rules). Taking this into account could shed new light on discussions of the appropriateness of targeted killings in given situations and their effectiveness overall.