The rapid scale-up of research on coronavirus disease 2019 (COVID-19) has spawned a large number of potential vaccines and immunotherapies, accompanied by a commensurately large number of in vitro assays and in vivo models to measure their effectiveness. These assays broadly have the same end-goal — to predict the clinical efficacy of prophylactic and therapeutic interventions in humans. However, the apparent potency of different interventions can vary considerably between assays and animal models, leading to very different predictions of clinical efficacy. Complete harmonization of experimental methods may be intractable at the current pace of research. However, here we analyse a selection of existing assays for measuring antibody-mediated virus neutralization and animal models of infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and provide a framework for comparing results between studies and reconciling observed differences in the effects of interventions. Finally, we propose how we might optimize these assays for better comparison of results from in vitro and animal studies to accelerate progress.