Background/Question/Methods The Great Lakes Restoration Initiative (GLRI) Priority Watersheds program is an interagency effort with a goal to accelerate ecosystem restoration in the Great Lakes. The USDA Natural Resources Conservation Service, US-EPA, and the USGS have partnered to track GLRI program accomplishments on a select number of privately-owned farms applying conservation practices. However, evaluating the effectiveness of conservation effort requires consideration of not only the conservation activity, but also soil conditions, field practices, hydrologic conditions, and weather. An evaluation of suspended sediment and nutrient runoff concentrations and loads was conducted at 4 edge-of-field (EOF) sites (catchment areas 0.02-0.1 km2) on 4 working farms in Wisconsin within the Fox River Priority Watershed. All sites have >5 years of continuous water-quality monitoring between 2012 and 2020, and each farm implemented at least one conservation practice mid-way through the study. Monitoring data were used to build empirical models of concentrations and loads of suspended sediment and nutrients in surface runoff based on season, weather, field conditions, and pre/post conservation actions, which were used to estimate changes associated with each site’s conservation effort. An additional assessment at the receiving stream site downstream of the EOF sites was used to evaluate if conservation activities across the 116 km2 basin translated into reduced sediment and nutrient loading to the Fox River and Green Bay over the study period. Results/Conclusions Water quality was highly variable through time and among the EOF sites, likely reflecting the timing and intensity of runoff events, soil conditions, and farm management. Varying improvements were observed from the field-level evaluations, with 1 of 4 EOF sites showing statistically-significant reductions of suspended sediment, total phosphorus, and total nitrogen concentrations by 77%, 53%, and 73%, respectively. However, at the other 3 EOF sites concentrations and loadings either increased or changes were not statistically significant. Similarly, the downstream site also did not indicate statistically-significant improvements of water quality despite larger conservation efforts adopted across the basin. The lack of improvements in water quality at the EOF- or stream-scale may reflect legacy nutrient sources, unanticipated changes taking place on the landscape, or an inability to account for variability within the model framework. Monitoring efforts at a greater number of long-term sites are necessary to evaluate changes in water quality on working farms to account for temporal and spatial variation and to better assess if and where water-quality goals are being met.