Measures designed to control erosion serve two purposes: on site (reduce soil loss) and off site (reduce sediment delivery to streams and lakes). While these objectives often coincide or at least are complementary, they could result in different priority areas when spatial planning is concerned. Prioritising for soil loss reduction at the field level will single out areas with high erosion risk. When sediment flux at the catchment scale is concerned, sediment pathways need to be identified in ex ante analyses of soil conservation plans. In Norway, different subsidy schemes are in place to reduce the influx of solutes and sediments to the freshwater system. Financial support is given to agronomic measures, the most important of which is reduced autumn tillage where areas with higher erosion risk receive higher subsidies. The objectives of this study are (1) to assess the use of an index of connectivity to estimate specific sediment yields, and (2) to test whether conservation measures taken in critical source areas are more effective than those taken at where erosion risk levels are the highest. Different modelling approaches are combined to assess soil loss at catchment level from sheet and gully erosion and soil losses through the drainage system. A calibration on two parameters gave reasonable results for annual soil loss. This model calibration was then used to quantify the effectiveness of three strategies for spatial prioritisation: according to hydrological connectivity, sheet erosion risk level and estimated specific sediment yield. The latter two strategies resulted in a maximum reduction in total soil loss due to reduced autumn tillage of 10%. Both model performance and the effectiveness of the different prioritisation strategies varied between the study catchments.