What is a skills ontology, and how is it different from a taxonomy?

They are two different layers of the same system, and the words get swapped constantly. There are three layers, and separating them is what makes the rest of the conversation possible.

The vocabulary is the bottom layer: a flat list of skills, each with a name and a description. No structure, no relationships, just the words. TechWolf maintains one shared vocabulary across our whole customer base.

The ontology is the layer above it, and it is the one most often skipped. An ontology holds the relationships between skills: which ones are adjacent, which co-occur, which sit close enough that someone who has one can plausibly reach the other. Java is adjacent to JavaScript. That relationship is what lets a system match a person to a role they have never held, or spot that a skills gap is smaller than it looks. Without it, matching is literal string overlap, and every gap analysis reads as a wall of red. We maintain the ontology centrally, because these relationships are a property of how skills relate in the world rather than of any one organisation.

The taxonomy is the top layer, and it is yours. Skills roll up into clusters, and clusters roll up into domains, forming the hierarchy that reflects how your organisation thinks about its own capability. You own and govern that structure, and you can rename, recluster and assign owners to keep it current.

The practical consequence: when someone offers you a skills taxonomy, ask what sits underneath it. A hierarchy with no ontology beneath it is a filing system. It will organise skills neatly and it will not help you match anyone to anything. A longer treatment of all three layers is on our blog.