In this paper, we focus on quantification processes as processes of institutionalization. By drawing the consequences from a metaphor that underpins much of the research in the history of accounting and statistics, we aim at going beyond the usual considerations on the conventional nature of the resulting quantitative representations. We claim for the need to reflect on the nature of these processes rather than on their outcomes, focusing on the mechanisms at work, on the role of individual agents in shaping their direction, and on their unpredictable nature. When applied to quantification, this point allows to highlight the limitations of ethnographic approaches to the study of quantitative practices inside of organizations (ethnostatistics), limitations that result from the lack of a proper historical perspective. We then develop from microhistory a method that satisfies the need for logical rigour without giving up the idiosyncratic richness of historical evidence. We suggest that this frame may be useful for the analysis of a set of historical cases, focusing on the interaction between organizations and society in shaping measurable objects and their metrics.