This article analyses how formalization of promotion criteria and procedures influences clarity and transparency of academic assessment. Based on a longitudinal, structural micro‐study of a new tenure track system in a Swedish higher education institution, we find that inequality was reproduced through the choice of explicitly gendered metrics across all areas of assessment (research, teaching and service). We further demonstrate how the formalization of a ‘good enough’ standard, in addition to a standard of ‘excellence’, reinforced the scope for interpretational flexibility among assessors. This combination of explicitly gendered metrics and dual standards of performance gave gatekeepers’ broader discretion in hiding or communicating failure, with gendering effects. Finally, we conclude that the choices made about how to formalize assessment work placed a small group of senior academics firmly behind closed doors, thus ensuring that gatekeepers’ discretion and power was entrenched rather than restricted by formalization.