Ontology
An ontology is a structured map of the concepts, entities, and relationships in a domain, used to give a language model a consistent vocabulary and schema for reasoning about source data.
An ontology defines the building blocks of a knowledge domain and how they connect. For customer intelligence, the building blocks include interactions, accounts, people, objections, competitors, phrases, churn signals, deal stages, and products. The relationships include who said what on which call, which objection surfaced against which competitor, which phrase predicts which outcome.
Without an ontology, a model treats every document as an unstructured blob. Retrieval becomes keyword matching. Reasoning becomes pattern matching on surface text. The model cannot answer questions that require joining across sources because it does not know which pieces are joinable.
With an ontology, the same source data becomes queryable. The model can reason about objections as a class, not as isolated mentions of the word 'objection'. It can compare competitor mentions across accounts. It can surface every call where a specific deal killer appeared. The ontology is the schema that turns a pile of text into a knowledge base.
The Amdahl view
The ontology is the single highest-leverage investment a team can make in their AI GTM stack. A raw RAG pipeline gets you to 60 percent. An ontology gets the remaining 40 percent and makes the first 60 percent dramatically more reliable. Teams that skip the ontology step and go straight to 'connect sources, ask questions' are rebuilding the same brittle prototype that everyone else has. The ontology is the moat.
Frequently asked
Related terms
- AI InfrastructureContext EngineeringContext engineering is the discipline of deciding what a language model should know at inference time, including the source data, structure, and ordering of its working memory.
- AI InfrastructureRetrieval Augmented Generation (RAG)Retrieval Augmented Generation (RAG) is a pattern where a system retrieves relevant documents from an external source, injects them into the model's prompt, and has the model answer from the retrieved material rather than from parametric memory.
- AI InfrastructureCitation GraphA citation graph is the structure that traces every claim in a generated output back to the specific source material it came from, creating a verifiable audit trail.
- The IntersectionCustomer intelligenceCustomer intelligence is the structured, queryable layer of meaning a B2B company builds from every conversation, signal, and interaction it has with buyers and customers.
- The IntersectionAgent-ready dataAgent-ready data is customer data structured, cited, and queryable by an AI agent without human translation.
See customer intelligence running on your own customer conversations.