Walk through five organisations that work with skills (a school, a vocational training provider, a hospital, a council, a defence force) and you'll find five different ways of naming and grouping skills, sometimes wildly different. The school might describe what students develop in terms like creativity, collaboration, and critical thinking, while the hospital lives inside a clinical competency framework that names dozens of specific capabilities by procedure. The defence force has its own scheme entirely, and the council uses something else again.
People often ask why we don't just pick one and standardise. Wouldn't it be easier? Wouldn't it make everything line up?
Easier in some respects, probably. But less useful for most of the people who actually rely on these frameworks day to day.
Skills aren't universal because work isn't universal
The reason organisations build different taxonomies is that the people they're describing are doing different things, in different contexts, for different reasons. A taxonomy isn't really an abstract description of human capability so much as a description of what a particular group of people are actually doing, organised in a way that's useful to whoever's going to use it.
The school's taxonomy needs to make sense to fifteen-year-olds, their parents, and the staff guiding them, and it needs to map onto what a school day actually contains and what young people are learning to be capable of. The hospital's taxonomy needs to be specific enough that a nurse manager can look at it and know whether a graduate is cleared to do a procedure unsupervised. The defence force needs categories that survive contact with operational reality. Each is shaped by who it has to serve and what decisions it has to support.
Forcing all of them onto a single shared list would mean either flattening everything to the lowest common denominator (so the school taxonomy stops being useful to schools), or piling so much detail into one master scheme that it becomes hard for anyone to navigate. Neither outcome tends to serve the people the taxonomy was meant to help.
Different framings answer different questions
The differences between taxonomies usually aren't mistakes, they reflect real choices about what each taxonomy is for. One framework might call something 'communication' while another splits it into 'presenting', 'negotiating', and 'active listening'. One might group all clinical procedures under 'patient care' while another fans them out into thirty separate competencies. They're answering different questions, and each tends to be doing a job that's appropriate to the context it was built for.
The same pattern shows up at the international level. ESCO is built for European labour mobility, so it's organised around occupations and the skills that move between them across borders. O*NET is built for the United States workforce and reflects how American employers describe roles. SFIA is specific to digital and IT work and goes deep on technical specialisation. Each is the right tool for the job it was built for, and none of them was ever intended to be the right tool for every job.
What this means for organisations building their own
If you're a school, a college, a vocational training provider, an industry body, or a workforce planner thinking about your own taxonomy, the upshot is fairly straightforward: build (or adopt) one that reflects your people, your context, and the decisions you need it to support. Picking someone else's framework off the shelf because it happens to be there is rarely the strongest move. What tends to work better is choosing or building one that genuinely describes what your students, staff, members, or community are actually doing.
In practice that might mean adopting an existing framework if it fits well, starting from one and adapting it, or building something fresh when nothing on the market reflects the work you're describing. All of these are reasonable approaches, and what tends to cause problems is using a framework that distorts the work simply because it was easier than building one that actually fits.
There's also a quieter benefit worth naming. A taxonomy that fits its context is one the people inside it can recognise themselves in. A nurse looking at a competency framework that names what nurses actually do tends to feel seen by it, and a student looking at a skills framework that names the things they're developing in the school they attend can actually put it to use. The reverse is also true: taxonomies that people don't recognise themselves in tend to gather dust.
So what about comparison?
The natural follow-up question is: if everyone has their own taxonomy, how do we compare across them? How does a graduate from one system get recognised by another? How does a workforce planner make sense of skills data that came in from twelve different sources?
That's the question we exist to answer, and it's a different question to the one this post is about. The point here is that the diversity of frameworks isn't really a problem to be solved by pushing everyone onto the same list. It's a feature of a world where different work happens in different places, and where people are best served by frameworks that fit what they actually do.
Comparison and translation between frameworks is where most of the real work sits.