Abstract. In recent years ontologies – shared conceptualizations of some domain – are increasingly seen as the key to further automation of information processing. Although many approaches for representing and applying ontologies have already been devised, they haven’t found their way into enterprise applications. In this paper we argue that ontology-based systems lack critical technical features, such as scalability, reliability, concurrency and integration with existing data sources, as well as the support for modularization and meta-concept modeling from the conceptual modeling perspective. We present a conceptual modeling approach that balances some of the trade-offs to more easily integrate into existing enterprise information infrastructure. Our approach is implemented within KAON, the Karlsruhe Ontology and Semantic Web tool suite.