Most national or regional initiatives aimed at managing biological invasions lack objective protocols for prioritizing invasive species and areas based on likely future dimensions of spread. South Africa has one of the most ambitious national programmes for managing plant invasions in the world. There is, however, no protocol for assessing the likely future spread patterns needed to inform medium- to long-term planning. This paper presents an assessment of the climatic correlates of distribution of 71 important invasive alien plants, and an analysis of the implications of these findings for future invasions in different vegetation types in South Africa, Lesotho and Swaziland over the next few decades. We used a variant of climatic envelope models (CEMs) based on the Mahalanobis distance to derive climatic suitability surfaces for each species. CEMs were developed using the first three principal components derived from an analysis of seven climatic variables. Most species are currently confined to 10% or less of the region, but could potentially invade up to 40%. Depending on the species, between 2% and 79% of the region is climatically suitable for species to invade, and some areas were suitable for up to 45 plant invaders. Over one third of the modelled species have limited potential to substantially expand their distribution. About 20% of the vegetation types have low invasion potential where fewer than five species can invade, and about 10% have high invasion potential, being potentially suitable for more than 25 of the plant invaders. Our results suggest that management of the invasive plant species that are currently most widespread should focus on reducing densities, for example through biological control programmes, rather than controlling range expansions. We also identify areas of the region that may require additional management focus in the future.