The Organisation for Economic Cooperation and Development (OECD) has developed impressive machinery to produce international comparative data across more than 70 systems of education and these data have come to be used extensively in policy circles around the world. In many countries, national and international comparative data are used as the bases for significant, high-stakes policy and reform decisions. This article traces how international comparability is produced, using the example of equity measurement in OECD's Programme for International Student Assessment (PISA). It focuses on the construction of the objects of comparison and traces the struggles to produce equivalence and commensurability across diverse and complex worlds. Based on conversations with a number of measurement experts who are familiar with the OECD and PISA, the article details how comparability is achieved and how it falters and fails. In performing such an analysis, this research is not concerned with "exposing" the limitations of comparison or challenging their validity. Rather, based on the work of Steve Woolgar and other scholars, it attempts to mobilise a "sociology of measurement" that explores the instrumentalism and performativity of the technologies of international comparisons.