Clean water is key for sustainable development. However, large gaps in monitoring data limit our understanding of global hotspots of water quality and their evolution over time. We demonstrate the value added of a data-driven approach to provide accurate high-frequency estimates of surface water quality worldwide over the period 1992-2010. We assess water quality for six indicators (temperature, dissolved oxygen, pH, salinity, nitrate-nitrite, phosphorus) relevant for the Sustainable Development Goals (SDG). The performance of our modelling approach compares well to, or exceeds, the performance of recently published process-based models. The model's outputs indicate that poor water quality is a global problem that impacts low-, middle-and high-income countries but with different pollutants. When countries become richer, water pollution does not disappear but evolves.