During the twentieth century, a number of actors and institutions acrossthe global north set out to develop hierarchical social taxonomies of their national populations. Mainly used for the making of statistics, these divisions soon came to be influential in policy and public debates. Using mainly Swedish examples, this article offers new ways of understanding social taxonomies, thereby adding insights into an understudied research object within the field of history of knowledge. Social taxonomies connect mundane and practical aspects of knowledge in the making – in terms of how actors order empirical material to through these create statistics – with larger public debates on society. They are, moreover, linked to different epistemic and political projects. I argue that social taxonomies should be understood as difference technologies; that is, ways of ordering and studying the social by producing differences between and sameness within its classifications.