Some studies examining the impacts of invasive plant species in native plant communities have demonstrated non-linear damage functions, whereby community components, such as species richness, are seemingly unaffected by the presence of an invader until it has attained relatively high levels of abundance, whereupon there is a marked decrease with further increases in abundance. Given chronic limitations in the resources available for managing invasive species, it has been argued that the most damaging invaders would be controlled most efficiently by maintaining their abundances below such threshold levels. Because many impact studies do not involve sampling over a wide range of invader abundances it is not possible to estimate the prevalence of threshold relationships. Furthermore, studies that have employed appropriate sampling methods have shown that different life forms exhibit different threshold responses, indicating that maintenance management for biodiversity values should be designed to protect the most sensitive species or groups of species. Since control costs increase with invader abundance, economic and ecological considerations are aligned when invaders are sustainably maintained at relatively low abundances. Adopting such an approach should also minimise negative impacts where damage functions are linear.