Scholars have traditionally considered government-initiated votes of confidence a powerful tool that parliamentary executives can leverage to achieve policy and electoral goals. Yet, these votes remain a relatively understudied procedure, partly because theoretical efforts at making better sense of their use have not been matched by efforts at gathering comparative evidence vis-à-vis their employment. I draw upon new data from 14 parliamentary democracies (1945–2021) to propose a more clear-cut definition and classification of government-initiated votes of confidence, with the goal of refining our understanding of how they are employed in practice. I review the institutional rules (type of vote, initiating power, decision rule, consequences of failure), present some figures on the use of the procedure and cases of failure by country, and distinguish four applications and the related benefits/costs. Two scenarios in which executives can exploit this instrument to facilitate the emergence of or foster cohesion in legislative majorities are identified (“majority-making” and “whipping/cohesive” confidence), as well as two additional ways in which this mechanism can aid executives (“signaling” and “technical” confidence).