The amount of information in today’s information society is immense, which creates a need for intuitive and effective search functionalities and applications. In addition to openly available search applications, organizations need internal search functionalities for optimizing their information management. This thesis provides an implementation suggestion for JoutseNet semantic search application. JoutseNet is a case management system used by the authorities and the employees of the city of Turku. Thesis begins by introducing some relevant fundamentals of natural language processing and search engines. Literature review is utilized to find semantic search implementation methods from previous research papers. Case JoutseNet is introduced with some background information on the case management process and with a brief user research and examination on the current state of the system. Learnings from the fundamental guidelines and conducted research are combined to implement the search application. After the implementation documentation, guidelines for optimizing and testing the application are given. The value and performance of the implementation is yet to be determined because the production data of the JoutseNet system could not be used for research purposes. A comprehensive suggestion is provided, but further research and development is still needed before delivering it to the production environment.