Social science research has demonstrated that new efforts to quantitatively evaluate performance in the U.S. and Europe effectively alter organizational priorities and staff behavior, yet little attention has been paid to performance metrics in transnational evaluation. Sociological theories of quantification contend that evaluation systems serve as inescapable universalizing agents for Western cultural attitudes and behavior. Rather, I hypothesize that the impacts of performance metrics are culturally mediated during the everyday labor of evaluation and are particularized in unforeseen ways. My dissertation draws on the mandatory evaluation of 'women's empowerment' in a bilateral donor's agricultural development program to ask: How are accountability, development worker and beneficiary identities, and knowledge production affected when development organizations are contractually obligated to demonstrate quantitative results about socially-oriented outcomes? This ethnography focuses on the development project implemented by a bilateral donor's sub-contractor ValueWomen (pseudonym) in a recipient country. This multi-sited ethnographic research mimics the process of data aggregation in evaluation: moving from farms, ValueWomen and donor field offices, to the respective international headquarters of both. Methods include semi-structured interviews throughout the evaluation system, textual analysis of evaluation documents, and participant observation at the project site and field offices. My research will extend social science by theorizing the effects of performance metrics on identities, staff and organizational work practices, and knowledge production in transnational settings. This research will utilize the everyday labor of evaluation to theorize how a diverse set of people comply, resist, and/or renegotiate performance metrics as they work together to turn the messy realities of projects into fact-like knowledge.