Building on a longitudinal dataset of 245 small firms covering the period of the Global Financial Crisis, this study uses, in combination with fuzzy clustering, the N-State Classification and Ranking Belief Simplex (NCaRBS) technique. This technique, able to deal with ambiguous outcome variables, small datasets, incomplete data and relationships that have the potential to be non-linear, is used to explore the relationship between learning and the resilient performance of small firms. Our findings provide a fine-grained picture of the complex relationships between strategic, cognitive and behavioural learning mechanisms and three resilient performance clusters – sustained performance, stability, and survival – which has implications for theory, as well as practice. By examining learning at the level of the individual owner-manager and also the organisation, we contribute to a better understanding of the role of specific learning mechanisms, a role that is still not well understood