The concept of thresholds of potential concern (TPCs) as implemented for the last decade in strategic adaptive management in South African National Parks (SANParks), has proved workable in practice in a number of instances, but in others appears beset by conceptual and practical limitations or barriers. Three common challenges relate to (1) situations where there is uncertainty about whether and where real thresholds exist, (2) whether and how preferences and other social constructs, as opposed to what were seen as objective biophysical variables only, can be used for TPCs and (3) whether it is admissible to adjust TPCs to allow for variations in societal behaviour, in particular rate of management response. All three challenges arise in the face of TPC objectivity implied by the original definition, and in the light of the original view that TPCs be set some distance prior to a presumed ecological threshold. This paper suggests that the three challenges can be partly or largely dealt with by the use of a wider socio-ecological view, rather than seeing TPCs in isolation or as being only biophysical. Also, while detection of abrupt changes is helpful, it makes little practical difference if some TPCs happen to describe linear processes. The very decision to intervene can induce an abrupt change. Once a wider socio-ecological approach is employed, it becomes necessary for the user to specify the particular usage envisaged for the TPC, for instance, whether it is considered a preference and whether that preference is believed in any way to be related to an ecological threshold. In all cases, it is recommended that some form of explicit representation of the socio- ecological view is constructed – we suggest a cause-and-effect diagram (and give an example generated through a thought experiment) which describes presumed relationships in the subsystem of interest. This provides a broader systemic context and a shared understanding, and has implications for considering scenarios and management alternatives. For practical reasons, from the several states and processes in such a subsystem, only a few links can be chosen on which to base particular TPCs. If we have understood the subsystem well enough, these few links, at each of which a TPC is developed, will act as diagnostic points at which we can monitor the performance of the subsystem adequately. A broadened definition of a TPC is presented, supporting this approach.Conservation implications: The concept of thresholds (initially ecological thresholds) has started influencing conservation management practice, a commonly-used formulation for management decision-making being the threshold of potential concern (TPC). Practical TPC usage can often be improved by moving away from its initially pure ecological outlook, rather framing understanding through an interlinked socio-ecological view.