How does our understanding of private security alter if we treat security consumption as consumption? In this article, we set out the parameters of a project which strives-theoretically and empirically-to do just this. We begin with a reminder that private security necessarily entails acts of buying and selling, and by indicating how the sociology of consumption may illuminate this central-but overlooked-fact about the phenomenon. We then develop a framework for investigating security consumption. This focuses attention on individual acts of shopping; practices of organizational security that individuals indirectly consume; and social and political arrangements that may prompt the consumption of, or themselves be consumed by, security. This way of seeing, we contend, calls for greater comparative enquiry into the conditions under which markets for security commodities flourish or founder, and close analysis of the social meanings and trajectories of different security goods. By way of illustration we focus on four such categories of good-those we term commonplace, failed, novel and securitized. The overarching claim of the article is that the study of private security currently stands in need of greater conceptual and empirical scrutiny of what is going on when 'security' is consumed.