Skills Ontology
TLDR
A structured map of skills, their relationships, and how they map to jobs and people.
Definition
A skills ontology is a taxonomy of competencies organized by hierarchy (broad to specific), adjacency (related skills), and prerequisite (what you need first). It is used in HR technology to match people to roles, identify capability gaps, and plan workforce development. Unlike a simple skills list, an ontology encodes relationships between skills so a system can reason about transferability and substitution.
Why it matters
Most HR data is text: job descriptions, CVs, LinkedIn profiles, internal performance reviews. A skills ontology turns this text into structured data that can be queried, compared, and scored. Companies with a strong ontology can do things that keyword matching cannot: "find people who could move into this role with three months of training" or "which teams lack redundancy on critical skills."
Belgian TechWolf is the most visible example of a company built around an ontology; its product extracts skills from CVs and job posts and matches them against an internal graph maintained by the company.
Mechanism
Ontologies are maintained through a combination of human curation, automated extraction from corpora, and feedback from user corrections. The quality bottleneck is typically the edge data, not the nodes: defining "Python" as a skill is easy; defining when "Python with Django" is transferable to "Python with FastAPI" requires domain judgment.