## [Ontology](/en/glossary/ontology) ## What Is an Ontology An ontology is a formal model of the concepts in a domain and the relationships between them. It defines the kinds of things that exist, their properties, and how they relate, providing a shared understanding that people and software can use. In enterprise data management, an ontology might define Customer, Account, and Contract as concepts. It could specify that a customer owns an account, a contract belongs to a customer, and a business customer is a type of customer. These definitions describe the domain independently of how individual systems store the data. For example, a bank may store customer information in its CRM, account details in its banking platform, and contracts in a document system. An ontology provides a common model for understanding those concepts and their relationships across the three systems. It does not, by itself, connect to those systems or keep their data current. ## What Problems Does an Ontology Address Different systems and teams often describe the same business concept differently. One system may use “client,” another “customer,” and another “account holder.” The differences may reflect equivalent terms, or genuinely different roles. Without an explicit model, consumers must infer which meaning applies and how the information connects. An ontology helps make those distinctions clear. It can support: - Shared understanding – Define concepts and relationships so teams can interpret information consistently. - Integration across systems – Provide a conceptual reference for mapping different data structures to agreed-upon business concepts. - Connected information – Describe relationships that help consumers understand how customers, products, contracts, and other concepts fit together. - AI interpretation – Give AI applications explicit domain concepts and relationships to support retrieval and interpretation of relevant information. As more applications and AI agents use enterprise data, undocumented assumptions become harder to manage. A shared domain model reduces the need for each new consumer to reconstruct the same business understanding. ## What Makes an Ontology Formal An ontology specifies concepts and relationships explicitly, rather than leaving them only in descriptive text. It may define classes, such as Customer; properties, such as Customer Identifier; and relationships, such as Holds Contract. Depending on the modeling language, it can also express logical conditions that support automated inference. For example, software could infer that a Business Customer is also a Customer. Languages such as the Web Ontology Language (OWL) support machine-readable models. The level of formality should fit the use case. Not every organization needs a comprehensive ontology or automated reasoning to establish useful shared meaning. ## How Ontology Relates to Other Data Concepts The clearest distinction is the question each concept answers. The following examples all use the same customer domain. ConceptQuestion it AnswersCustomer Example[Data Semantics](https://www.denodo.com/en/glossary/data-semantics-definition-importance-and-benefits)What does this data mean?“Active customer” means a customer with at least one current contract.OntologyWhat kinds of things exist and how can they relate?Defines Customer, Account, and Contract, and specifies relationships such as a Customer owns an Account.[Knowledge Graph](https://www.denodo.com/en/glossary/knowledge-graph)Which specific things are connected?Represents Acme Corporation as owning Account 123 and holding Contract 456.[Semantic Layer](/en/glossary/semantic-layer-definition-benefits-modern-applications)What is the standard way to represent shared entities?Makes such entities as the Active Customer definition and Contract Value calculation reusable across dashboards, applications, and AI agents. An ontology can guide the structure of a knowledge graph, while a semantic layer can make shared business definitions usable across data consumers. These roles can overlap in particular implementations, but they are not interchangeable. A knowledge graph does not always require a formal ontology, and a semantic layer can operate without either. ## How an Ontology Differs from a Taxonomy or Database Schema A taxonomy primarily organizes concepts into categories and hierarchies. An ontology can describe additional relationships and logical conditions, such as the distinction between owning an account and being authorized to use it. A database schema defines how data is structured in a particular database; an ontology describes domain meaning that may span several databases and applications. ## Where Denodo Fits Defining a domain model is only part of making enterprise information useful. Consumers also need access to relevant data, a way to apply shared meaning, and controls that govern how the information can be used. Denodo’s [universal semantic layer](https://www.denodo.com/en/solutions/by-capability/universal-semantic-layer) makes business definitions, metrics, and relationships reusable across distributed data sources. This addresses the need to apply consistent meaning when enterprise information is delivered to analytics tools, applications, and AI agents. Denodo’s [AI data layer](https://www.denodo.com/en/data-management/ai-data-layer) combines semantic trust with connectivity, live delivery, and governed access to create [active context](https://www.denodo.com/en/data-management/active-context). This brings trusted business meaning together with current enterprise information and policies appropriate to the consumer and task. The shared foundation supports governed, reusable enterprise data products and reduces the need to rebuild data access and semantics for each new use case. Explore how Denodo makes shared business meaning usable across systems and consumers. [Explore Denodo’s Semantic Layer solution](https://www.denodo.com/en/solutions/by-capability/universal-semantic-layer) [Back to Glossary page](/en/glossary)