On information markets, many suppliers and buyers of information goods exchange values. Some of these goods are data, whose value is created in buyer interactions with data sources. These interactions are enabled by data market services (DMS). DMS give access to one or several data sources. The major problems with the creation of information value in these contexts are (1) the quality of information retrievals and related queries, and (2) the complexity of matching information needs and supplies when different semantics are used by source systems and information buyers. This study reports about a prototype DMS (called CIRBA), which employs an ontology-based information retrieval system to solve semantic problems for a DMS. The DMS quality is tested in an experiment to assess its quality from a user perspective against a traditional data warehouse (with SQL) solution. The CIRBA solution gave substantially higher user satisfaction than the data warehouse alternative.