With the rapid development of economy and society, decision-makers need a deep understanding of the role of sustainable ecosystem management in practical application, which is crucial to the sustainable development of ecosystems. Understanding the relationship between ecosystem supply and social demand is key in overcoming the contradiction between humans and the ecosystem, as well as achieving successful ecological restoration and healthy regional development. However, there is still a lack of multi-scale and multi-method research on the supply–demand relationship of ecosystem services for key ecological-restoration zone. In this study, a typical ecological barrier (i.e., Ganzhou City of southern China) was selected as the research area, and five ecosystem services (ESs) including water yield (WY), carbon sequestration (CS), crop production (CP), soil retention (SR) and recreation service (RE) were investigated at the pixel, township and county scale in 2020. The balance and coordination analysis of ESs supply and the quantitative analysis of the supply–demand relationship (using the coupling coordination degree (CD), matching degree (MD), and supply–demand ratio (SDR)) were executed under multiple scales. Finally, based on the MD at the township scale, Ganzhou City was divided into four areas: high-demand, low-demand, low-supply and high-supply areas. The results reveal a trade-off relationship between CP and CS at the pixel scale, with no trade-off relationships observed among the ESs at the other scales. The supply–demand relationships of different districts and counties in Ganzhou City are highly uncoordinated, with distinct spatial mismatching. Areas with high CD values are located in Chongyi County and Longnan County, while the low/value SDR area is located in Zhanggong District, with a relatively high population and GDP. The results of MD division based on township level in Ganzhou City show 17 townships are classified as high-demand areas (mainly in urban areas) and 23 townships are classified as low-demand areas (mainly in the suburbs); 77 townships are classified as low-supply areas and 167 townships are classified as high-supply areas, which are mainly distributed in forest areas. This study provides guide for regional ecological space planning, enhances the understanding of ESs management, and provides a more objective reference for decision makers to plan future ecological restoration.