The susceptibility of a land surface to wind erosion is highly sensitive to changes in soil erodibility. Nonetheless, the performance of wind erosion models continues to be affected by the accuracy of their erodibility representations. There is thus an ongoing need for robust approaches for assessing and modelling soil erodibility dynamics. This paper provides a critical review of research into the controls on soil erodibility dynamics. The review focuses on progress in understanding temporal changes in soil aggregation and crusting as they influence the erodibility of agricultural and rangeland soils, and identifies deficiencies in approaches for resolving the nature and causes of spatio-temporal patterns of erodibility change. A conceptual model of soil erodibility dynamics is developed to represent erodibility changes within a single erodibility continuum. The model is used to identify ongoing research questions that are central to developing new measures and a deeper understanding of soil erodibility dynamics, and representations of soil erodibility for wind erosion and dust emission models. Finally, available soil erodibility metrics are evaluated in the context of their application in addressing these research needs, and new and alternate approaches for reducing the complexity of soil erodibility assessments and models are identified.