The area around Elands Bay and the adjacent interior landscapes west of the Doring River have been subject to intense archaeological investigation over the last ~50 years. The result is a region with great depth and diversity of archaeological information. In this paper I discuss three general observations that arise from the integration of data across this region. The first is that redundancy in site occupation is limited: even where many sites are excavated in a small area, understanding of the regional sequence cannot be assumed to be complete. The second is that humans did not live in rock shelters: a focus on rock shelters alone, even where these are abundant, produces a skewed picture of occupational and demographic histories. The third is that the coast and its hinterland are intimately bound: interaction between the two zones is variable, and even where it is limited this observation is important to the understanding of both.