BACKGROUND: Perioperative red blood cell (RBC) transfusion is a common intervention in patients undergoing surgery but there is marked variation in practice. Key performance indicators (KPIs) are central to identifying deviation from agreed standards and improving clinical outcomes. We aimed to identify KPIs which can potentially be measured from routinely collected electronic healthcare records. METHODS: We undertook a three-stage process. First, we completed a scoping review to identify potential KPIs from relevant literature and clinical guidelines. Next, we conducted a modified RAND consensus process with a multidisciplinary panel including medical professionals, patients and public involvement members. The consensus panel rated these KPIs according to importance and feasibility. RESULTS: We identified 28 candidate KPIs covering the entire perioperative RBC transfusion process. The majority of the KPIs focused on improving patient care around the time of decision to transfuse RBCs and transfusion safety. Clinical outcome KPIs included hospital length of stay, hospital acquired infection, mortality, and hospital readmission at 30 and 90 days. Five candidate KPIs were judged as unimportant whilst there were concerns around the feasibility of measurement using routine data for 14 candidate KPIs. The panel identified nine potential KPIs for future testing. CONCLUSIONS: Using a systematic, stepwise, transparent approach, we have identified a set of 28 KPIs for assessment, monitoring, and improvement of perioperative RBC transfusion. Future research is needed to further validate this set for external use and benchmarking between hospitals and departments.