A burgeoning area of social science research examines how state capabilities and bureaucratic effectiveness shape economic development. We study how the management practices civil service bureaucrats operate under correlate to the delivery of public projects, using novel data from the Ghanaian Civil Service. To do so, we combine hand-coded progress reports on 3600 projects with a management survey in government Ministries and Departments responsible for these projects, following the methodology of Bloom et al [2012] Management matters: practices related to autonomy are positively associated with project completion, yet practices related to incentives/monitoring of bureaucrats are negatively associated with project completion. The negative impact of incentives/monitoring practices is partly explained by bureaucrats having to multi-task, interactions with their intrinsic motivation, their engagement in influence activities, and project characteristics such as the clarity of targets and deliverable outputs. Finally, we discuss the interplay between management practices and corruption, alternative methods by which to measure management practices in organizations, and the external validity of our results by comparing key findings to those in Rasul and Rogger [2018]. Our findings suggest the focus of many civil service reform programs on introducing stronger incentives and monitoring may backfire in some organizations, and that even countries with low levels of state capability may benefit by providing public servants with greater autonomy in some spheres.