Abstract: The governance of rural black and odorous water bodies is a crucial part of the battle for clean water. In recent years, special financial funds at various levels have been increasing continuously. However, in some regions, problems such as mismatch between fund allocation and governance needs, low budget execution rate, and poor governance durability have emerged. It is therefore imperative to further refine the performance evaluation indicators based on the actual conditions of governance projects, building upon the Measures for Performance Evaluation of Pilot Funds for Rural Black and Odorous Water Body Governance, so as to enhance their compatibility with governance projects. Building on the existing evaluation indicator system and targeting its deficiencies including a weak sustainability dimension, strong subjectivity in indicators, and insufficient quantification, this study screened indicators using the Delphi method, and determined the combined weights by integrating the analytic hierarchy process (AHP) with the entropy weight method. The performance evaluation system was optimized from the four dimensions: fund management, project management, development management, and public participation, with a framework covering 11 criterion layers and 27 tertiary indicators, so as to refine and supplement the existing indicator system. The results show that the optimized indicator system raises the weight of the development management dimension to 0.28, and increases the proportion of quantitative indicators from approximately 40% in the original indicator system to over 70%. Targeted indicators such as zero-occurrence rate of black-odor rebound, completeness rate of routine maintenance records, and the rate of villager participation in target setting are newly added, enabling a more comprehensive reflection of the long-term effectiveness and social benefits of the governance. Compared with the existing indicator system, the optimized indicator system shows significant improvements in evaluation comprehensiveness, indicator quantifiability, and technical adaptability.