As society becomes more data-driven, new alternative technologies to the more traditional relational database model arise. One of these technologies is the graph database model, which stores the data as nodes and edges, where the edges define the relationships between the nodes representing entities. Graph databases are a great fit for data with dense and dynamic relationships, such as social networks, fraud detection and recommendation engines. At the Swedish Pensions Agency, the unit working with fraud detection has rapidly increased personnel due to increased demand for fraud prevention. New technologies are being investigated to improve the efficiency in both detection of current fraudulent activities and also to detect suspicious activities and persons to prevent fraud from even happening. The scope of this thesis is to develop a tool as a proof of concept to import data through the web browser into a graph database, namely Neo4j. The evaluations show that importing nodes and relationships with the tool is slower than importing with Neo4j LOAD CSV. However, the performance is still reasonable for data sets of a few hundred thousand nodes or relationships. The evaluations with the Swedish Pensions Agency show that the tool could bring value to them by removing the complexity of writing code.