Build an ontology

Build an ontology

Describe what your data means: business concepts, definitions and relationships that people and AI agents can reason over.

Outcome: You can turn a data model into a concept graph with clear definitions and meaningful relationships.

1. 1

   Name the concepts and define them

   An ontology starts with the things the business talks about, defined in one sentence each.

   Build concepts and relationships · 10 min · instant checks

2. 2

   Connect concepts with verb phrases

   'Customer places Order' says something. 'Customer has Order' doesn't.

   Build concepts and relationships · 10 min · instant checks

3. 3

   Classes and subclasses

   Retail customers and business customers are both Customers. A taxonomy says so, and says what makes each kind different.

   Build concepts and relationships · 10 min · instant checks

4. 4

   Synonyms and a business glossary

   Sales says Client, support says Account holder, the warehouse says Item. Record those words on one concept instead of creating three.

   Build concepts and relationships · 10 min · instant checks

5. 5

   Walk relationships to answer questions

   A question is a path through the concept graph. Find the path and you have the joins.

   Build concepts and relationships · 12 min · instant checks