Last updated: September 2026
Ontologies form the knowledge scaffolding of the enterprise and provide reference values for both structured and unstructured information. They consist of the various taxonomies that describe enterprise information (products, services, solutions, processes, business units, regions, roles, interests, document types, and more) along with the relationships between these elements. (Services for a product or solutions to problems, interests of a role and more) This is reference data - the official terms and labels for information) it is important because without it, your systems and technologies will not know the names of your products, terminology of your organization and processes that are powered by data. Large language models also require ontologies since they are trained on public data - not information from behind your firewall.
An ontology is a set of concepts and categories in a subject area or domain that shows their properties and the relations between them.
Knowledge bases are sources of truth for AI powered applications. Ontologies form the the structure on which knowledge is contextualized. Ontology-based AI allows the system to make inferences based on content and relationships and can produce personalized results by relating customer data to the products they would be most interested in.
Types of ontology knowledge models
Knowledge models run along a continuum, beginning with the simplest level in which a controlled vocabulary is developed to encourage the use of the same word for a particular meaning (such as always using “client” rather than a mix of “client,” “customer,” and “purchaser”). The next step is a thesaurus, which allows identification of terms that relate to a single concept. The next stage is a taxonomy, which defines a hierarchy with parent-child relationships. The parent child relationship might be a specialization of a product category, or one item being classified as part of another, such as an engine being part of a car. They are often used as a navigational construct on websites to get the user from one piece of information to another. An ontology is a representation of the relationships among multiple taxonomies. Finally, a knowledge graph can be used to capture specific instances of the relationship, such as a particular sales transaction between entities, whereas the ontology is generic.
Ontology is therefore the knowledge scaffolding in an enterprise, and a lot of value can be achieved by leveraging that structure across multiple systems and data sources. Knowledge of relationships between concepts allows identification of solutions to problems, or knowledge about which products or services go with another set of products.
Types of relationships within the model
Different types of term relationships are applied to different knowledge models.
- Equivalence is used in thesauri, and it does not necessarily refer to synonyms. For example, transparency and opacity both refer to the same concept, although they describe it in different ways.
- Hierarchical relationships are used in taxonomies to categorize items or concepts, and associative terms are used for concepts or entities in ontologies.
- Associative terms are context- and audience-specific, and are used to relate multiple taxonomies to each other.
The figure below shows each of these in the context of game manufacturers. Once the ontology has been established, it becomes a very versatile tool. A brand manager might it to reveal one set of connections while a salesperson might use others, and a product developer still others.
Ontologies and taxonomies have some elements in common, but ontologies provide a richer set of information. A taxonomy is a tree structure, and one item or concept can have only one parent. In an ontology, an object such as a smartphone can inherit features from multiple parents. This quality is useful because in the real world, objects and concepts are related to each other in many ways, not just in a single hierarchy.
As a simple example, two individuals might be identified as co-presenters in a conference session. A human could readily understand that both were presenting at the conference, but without a defined relationship between the two, a computer system could not. The use of ontologies allows a system to infer conclusions and answer unanticipated questions that have not been programmed into them. It functions as a Rosetta stone that allows systems to communicate with each other, providing a richer context and knowledge base.
One way to think of ontologies is that they are master data management (MDM) for AI, in that they provide an overall structure for knowledge, but they differ in that they are dealing with more than data. They could be dealing with workflows, business processes, and other constructs that are not present in data itself.
Ontologies must work with multiple systems in order for their value to be fully realized.
Knowledge of product relationships and solution relationships can be embedded--for example, the components needed for a product, and what alternate parts can be purchased. These can be added to the ontology, which becomes an asset of increasing value.
Ontology real world example
The Cleveland Museum of Art case
Ontologies can be applied to situations in which customer behavior is being assessed. The Cleveland Museum of Art wanted to understand its visitors’ preferences and patterns of interacting with the museum’s collection. In order to do this, they needed to identify the characteristics of the collections, their themes, locations, and specific points of interactions by visitors. Terminology was developed to represent locations because those were needed in order to set up the correlations between visitor reactions and the exhibits. Visitors opted in to allow the museum to record their utterances. An ontology was developed to connect geo-spatial data to behavioral analytics based on very specific pieces of content. The museum gained a better understanding of its visitors’ preferences and was able to offer an improved experience.
This example is discussed in Seth Earley's book, The AI-Powered Enterprise: Harness the Power of Ontologies to Make Your Business Smarter, Faster, and More Profitable (LifeTree Media, 2020).
Applying the same pattern beyond one institution
An ontology allows an enterprise to reuse existing knowledge and resources by breaking them up into components that can be fed into multiple systems and applications and sent to multiple consuming systems, such as Slack, Facebook or other user-selected interface. The repository needs to be very granular in order to bring back the precise piece of content that reflects what the user was looking for. Conversational commerce requires a semantic deconstruction of utterances in order to respond appropriately to the user.
Ontology and digital transformation
For a full digital transformation, the enterprise will need hundreds or thousands of AI projects, so setting the stage for scaling up is essential. Rather than having a series of disconnected projects, a repeatable framework should be developed.
Why change management gets easier
Moreover, with an ontology in place, changes can be made to the data in one location and propagate through the existing associative relationships. So if there is a change in the price of a service, for example, the system does not require re-coding in multiple applications; the data can be changed once and carry through to all of them.
A robust information architecture is required in order to support sophisticated solutions that are emerging as organizations move forward with their digital transformations. Ontologies provide a re-usable, adaptive structure for organizations that want to power their AI initiatives. The more detailed the ontology, the more meaningful will be the responses that users receive.
Frequently Asked Questions
What is an ontology in the context of AI?
An ontology is a set of concepts and categories in a subject area that shows their properties and the relationships between them. In an enterprise setting, an ontology connects the taxonomies that describe products, services, processes, business units, and roles, giving AI systems the reference data they need to recognize what a term means and how it relates to everything else in the organization.
How is an ontology different from a knowledge graph?
An ontology defines the generic structure of concepts and their relationships in a domain, while a knowledge graph captures specific instances of those relationships, such as one particular sales transaction between two named entities. The ontology is the reusable template; the knowledge graph is the populated record built on top of it. Enterprises typically design the ontology first, then use it to structure the knowledge graph.
How is an ontology different from a knowledge base?
A knowledge base is the source of truth an AI application draws on, such as a repository of product or support content. An ontology is the structure that gives that content context: it defines the relationships between entities in the knowledge base so an AI system can reason about them rather than treating each record as an isolated fact. That structure is what lets the system infer which products a given customer is most likely to want.
Why do large language models still need ontologies?
Large language models are trained largely on public data, so they have no built-in knowledge of an organization's own products, terminology, processes, or business rules. An ontology supplies that missing enterprise context: it gives an AI system the reference data and relationships it needs to work with information that lives behind a company's firewall. Without it, a model can produce fluent answers that use the wrong terminology or miss how the organization's own products and services actually relate to each other.
Why does an enterprise need an ontology to scale AI across the organization?
A full digital transformation requires hundreds or thousands of AI projects, and building each one as a disconnected effort does not scale. An ontology gives those projects a shared, repeatable structure: because concepts and relationships are defined once, a change made in one place, such as a service price update, propagates automatically through every connected system rather than requiring separate re-coding. That reusable structure is what lets an enterprise move from isolated AI pilots to AI at scale.
What does it mean to call an ontology the enterprise's "knowledge scaffolding"?
Knowledge scaffolding describes the role an ontology plays in an organization: the underlying structure of taxonomies, terms, and relationships that everything else, from search results to AI-generated answers, is built on. Just as scaffolding supports a building under construction without appearing in the finished structure, an ontology supports every system that needs an enterprise's products, processes, and terminology. Earley Information Science uses the term to stress that this structural work has to happen before an AI system is built.
What is a controlled vocabulary, and why is it the starting point for building an ontology?
A controlled vocabulary is the simplest level of a knowledge model: a rule that one agreed-upon term is used for a concept, such as always using "client" rather than switching between "client," "customer," and "purchaser." It is the starting point because every later stage, a thesaurus, a taxonomy, and eventually an ontology, depends on the underlying terms already being consistent. Without a controlled vocabulary, the relationships an ontology defines will connect inconsistent labels instead of one concept.
How can an ontology make AI-powered customer engagement more personalized?
Ontology-based AI can infer connections a system was never explicitly told to make, so it can relate a specific customer's data to the products or services that customer is statistically most likely to want. Because the ontology already defines how products, categories, and customer attributes relate to one another, the system does not need a separate rule written for every combination; it reasons from the existing relationship structure instead. This is what lets personalization work at enterprise scale rather than through manually maintained rules.
Is there evidence that ontology-based or graph-based retrieval actually improves AI accuracy?
Yes. GraphRAG-Bench, a 2026 benchmark study presented at ICLR, found that graph-guided retrieval outperformed standard vector-based retrieval by roughly 10 percentage points on complex reasoning tasks and by roughly 13 points on contextual summarization, while performing about the same on simple fact lookup. The gain is concentrated exactly where ontology-based structure matters most: multi-step reasoning across related concepts, not single-fact retrieval.