Abstract: 【Purpose/significance】Public data classification and grading is a prerequisite for building a sound data element market sys⁃ tem, and is of great significance for effectively promoting data security and data utilization. Constructing a theoretical framework and implementation path of public data classification and grading can provide a reference framework for formulating strategies and solu⁃ tions for public data governance.【Method/process】Through systematic literature research and analysis of relevant policy texts, draw⁃ ing on the holistic governance theory, this study constructed a governance framework for public data classification and grading.【Re⁃ sult/conclusion】From the perspective of holistic governance, public data classification and grading faced inherent requirements for co⁃ ordinating the relationship between open sharing and security protection, between top-level design and grassroots exploration, between risk control and value realization. In view of the above requirements, the governance framework was conducted from the perspective of practical foundation, implementation path and expected goals.【Innovation/limitation】This study inevitably has some limitations such as insufficient theoretical research and few practical research cases, future research can further verify the robustness of this framework