Reduction of nitrogen (N) surpluses, indicating nitrate leaching, gaseous emissions, and low nitrogen use efficiency (Spiess & Liebisch, 2025) is necessary in Swiss agriculture. Over 8 years, multiple field trials across Switzerland evaluated the potential of data-driven decision support to optimize N fertilization while maintaining crop productivity. This contribution synthesizes several studies and a pilot project conducted across many sites and years, for winter wheat N fertilization. The evaluated approaches combined standard tools such as measuring soil mineral nitrogen (SMN) with digital monitoring. Different crop sensors from hand-held to satellite (Argento et al. 2025), weather information, and farm management data were used to support site-specific and crop demand-oriented N fertilization strategies. Combined approaches were compared with conventional fertilization practices and Swiss fertilization guidelines (Sinaj & Richner, 2017) under practical farming conditions. Performance indicators were N application rates, N surplus, crop yield, and nitrogen use efficiency. Across studies, data-driven decision support consistently reduced N inputs without significant yield losses. In many cases, N surpluses were reduced by 10–30% compared to reference treatments, primarily due to adjustment of N rates to actual crop demand and soil N supply (Argento et al., 2021; Grossrieder et al., 2022). Sensor-based and model-assisted approaches showed particular advantages in situations with high spatial or temporal variability of SMN. However, results also pinpoint limitations related to data availability, calibration effort, and farmer acceptance. Our findings demonstrate that data-driven decision support substantially reduces N surplus in Swiss agriculture. The synthesis underscores the importance of integrating multiple data sources to adapt decision support systems for regional conditions. Further development should focus on improving robustness, usability, and the integration of such tools into advisory services and farm management workflows. In particular, modelling the nitrogen mineralisation and understanding N release patterns from organic sources could improve spatial and temporal N fertilization.