As a key component in the building process of a smart city, smart water systems greatly contribute to relieving the pressure of monitoring and maintaining water resource systems as well as improving allocation rationality. However, the existing smart water systems have a wide variety of types and designs without a unified optimal evaluation method. Unlike other scientific models, the majority of end users for smart water system are the government and the public who may not be familiar with complex and academic principles, making it difficult to determine the applicability of the current system to local demands. Accordingly, we proposed an evaluation method for the smart water system based on the concept of fit-for-purpose. The proposed method enables a comprehensive evaluation in three dimensions: usefulness of the system, reliability in results, and feasibility in practice, to make sure that a smart water system can achieve the specified purposes required by the certain environment and users in its optimum state. In this study, we applied this approach to a typical smart water system with the main purpose of water system scheduling and flood control in Fuzhou, China to (1) evaluate the current system on three dimensions; and (2) reconstruct the system for identifying the improvement details that fit for intended requirements. We found that the new evaluation method showed good adaptability to the smart water system: (1) For a built smart water system, the method provides users with acceptance and improvement references through comprehensive evaluation and re-construction; (2) For a proposed smart water system, the method can contribute to pragmatic design suggestions that make the system fit for purpose from the beginning. Overall, the evaluation method based on Fit-for-Purpose can serve as the basis for future smart water and smart city, hence improving resource efficiency and facilitating the development of digital urban water management.