Soil conservation service (SCS) is critical for maintaining ecological security and achieving sustainable development goals. As China implements ambitious national soil conservation policies, including the "Overall Plan for Major Projects of National Important Ecosystem Protection and Restoration (2021-2035)", scientifically evaluating the coordination between SCS supply and demand is essential for optimizing resource allocation across diverse landscapes. This study applied an evaluation framework integrating the revised universal soil loss equation (RUSLE), a coupling coordination degree (CCD) model, and an elasticity coefficient method to analyze supply-demand relationships across China from 1990 to 2019, and constructed a coupling patterns indicator to classify the spatiotemporal dynamics of SCS supply and CCD. Results show that national SCS supply increased by 16.1% while demand rose by 22.6% from the 1990s to the 2010s. Approximately 32% of China's area, identified as hotspot regions (HRs), contributed 94.1% of the national supply increase and 82.1% of the demand increase. Within HRs, average CCD improved from 0.55 to 0.58, with "moderate balance" areas expanding from 30.8% to 39.4%. Coupling patterns analysis revealed that 61.5% of HRs exhibited simultaneous increases in both SCS supply and CCD, while 15.5% displayed increased supply but decreased CCD. Elasticity coefficient analysis identified that rainfall erosivity dominated CCD changes in 66.3% of HRs, whereas vegetation cover and management dominated in 33.7%. These findings provide a scientific foundation for implementing China's differentiated soil conservation policies, demonstrating that supply-side improvements alone are insufficient without integrated demand management, and offering region-specific guidance for prioritizing vegetation restoration in C-factor-dominated areas and climate adaptation in R-factor-dominated regions.