The selection of spatial and temporal scales of analysis is an important and challenging activity at the start of any Policy Analysis or Strategic Environmental Assessment (SEA). The scale sets bounds on the types of problems to be addressed, the solutions to be found and the impacts to be evaluated. Particular scale choices may be more or less beneficial to different actors and are therefore politically loaded. Despite its importance, this topic has rarely been specifically addressed through empirical research. This paper presents an assessment of the impacts of scale choice from varying points of view – those of political actors, commissioners, analysts and scientists – focusing on one type of scale choice: the spatial boundary of the study.An examination of the completed transboundary Long Term Vision Study (LTV) of the Scheldt Estuary, commissioned by the Flemish and Dutch governments, is used to illustrate how actors vary both in the boundaries they choose and the assessment they make of the implications of that boundary choice. Consequently, no perfect spatial scale choice appears to exist. The paper recommends ways to structure the problem of scale selection to facilitate rational deliberation. It is suggested that the methods used to analyse different actors perspectives for the completed LVT study, in future could be employed up-front at the start of other projects. Scale-dependent decision trade-offs could then be made more transparent.