The Problem

The Australian VET sector is being asked to do something its qualification architecture was not designed for. As the National Skills Taxonomy takes shape and industry begins to think about workforce in skills-based rather than role-based terms, the system that translates skills demand into trainable provision is increasingly out of alignment with what is being asked of it. The problem is not capability or commitment. It is structural.

Take communication. The skill appears, in some form, in hundreds of qualifications across childcare, hospitality, aged care, retail, business administration, and dozens of other sectors. A worker moving between any two of these sectors typically has to re-demonstrate the unit, even when their existing capability is obvious. The system treats the underlying skill as a different thing depending on which package it sits within. Multiply that decision across every transferable foundation skill, across every worker considering a sector move, and the cost to the worker, the employer, and the public training dollar becomes significant.

The system is not broken. It is doing what it was designed to do. The problem is what it was designed to do, and the fact that the world it was designed for has moved on.

The system is not broken. It is doing what it was designed to do.

The Diagnosis

The Australian VET system was designed for an economy that thought about workforce in terms of occupations. The training package model assumes that work is bundled into roles, roles are defined by industry, and qualifications credential people into those roles. Skills are described, assessed and funded as components of qualifications, not as units that travel independently of them. 

This architecture made sense when most workers moved within a single industry across their working lives, when role definitions were stable, and when the rate of skill change was slow enough for the system to keep up.

None of those conditions hold any more. The labour market has become more mobile. Roles are being reshaped by technology faster than qualifications can be updated. Workers move between sectors more often, and increasingly need to, as workforce shortages emerge unevenly across the economy.

The case for treating skills as the unit of analysis, rather than qualifications, has been made repeatedly by the Productivity Commission, by Jobs and Skills Australia, by industry bodies, and by the Strategic Review of the Apprenticeship Incentive System. It is no longer a contested position. It is becoming the operating assumption of every part of the workforce system except the one that has to deliver against it.

The training package architecture has been reformed, refashioned, and reorganised multiple times. The most recent round, completed in 2025, introduced a new Training Package Organising Framework and a new unit of competency template. These are useful incremental changes.

They are not structural reform.

The fundamental relationship between skills, qualifications, and industry sits roughly where it has sat for two decades. Skills are still subordinate to qualifications. Qualifications are still organised around occupations. The system can describe skills more cleanly now, but it still cannot move them.

The system can describe skills more cleanly now, but it still cannot move them.

The Future Skills Organisation has named this directly. In its concept paper for the Digital Capability Training Product Trial, the FSO observed that the national system that develops and deploys new VET training products is unable to keep pace with the speed of change, and that the process to review, design and update national training products can take years, which is too slow.

That is not an external critic making an argument.

It is the Jobs and Skills Council responsible for the cross-sector digital workforce stating it plainly. The system is too slow because it was designed for stability. The world it operates in is not stable.

The FSO Precedent

The Future Skills Organisation’s Digital Capability Training Product Trial provides a live precedent for building cross-industry, skills-based units of competency. This is not a conceptual proposal. It exists, it is operating inside the national system, and it is being evaluated.

The trial comprises thirty-nine units covering generalist digital capability, including cyber security and introductory artificial intelligence. The units align to DigComp 3.0 and are deliberately designed across AQF Levels 2 to 5, creating a clear progression of capability rather than a single, fixed skill set. This is not an IT qualification by another name. It is transferable digital capability intended to apply across roles and sectors.

The units sit within the BSB Training Package and are already being delivered through selected providers as part of the training product trial. Industry stewardship, public consultation, and independent evaluation are built into the model by design.

In short, this is a working example of how the VET system can separate capability from industry context while still operating at national scale.

Why the model works

The significance of the FSO trial is not the subject matter. It is the structure.

What the model demonstrates is that capability can be described independently of industry without stripping it of meaning. Each unit defines the underlying capability in a deliberately industry-agnostic way, while expecting context to be supplied at the point of delivery and assessment. Digital collaboration remains the same capability whether it is demonstrated in health, finance, or retail. What changes is the workplace environment in which it is applied.

Digital collaboration remains the same capability whether it is demonstrated in health, finance, or retail.

The model also takes progression seriously. By designing units across multiple AQF levels, it shows how capability deepens over time. Increased autonomy, complexity, judgement, and responsibility are built into the structure rather than implied.

Industry stewardship is central. These units are governed and maintained through an industry-led Jobs and Skills Council mechanism, with ongoing oversight rather than one-off consultation. Combined with the training product trial approach, this allows speed without abandoning national consistency.

The Extension Argument

If this model works for digital capability, the obvious question is why it stops there. Communication, problem solving, interpersonal skills, teamwork, work health and safety, and critical thinking appear in hundreds of qualifications across the AQF. They are repeatedly redefined, reassessed, and re-funded, even when the underlying capability is effectively the same.

A worker moving between sectors does not lose these capabilities, but the system behaves as though they might have. Foundation skills are embedded inside qualifications and treated as non-transferable by default.

If these capabilities were built using the same cross-industry, vertically articulated model as the FSO digital units, foundation capability would travel with the worker. Duplication would fall, recognition of workplace learning would improve, and sector movement would become faster and less punitive.

The Conditions for Getting It Right

Extending the model only works if it is done with discipline. Capability units must reflect real work, not abstract taxonomies. Each capability area requires genuine industry stewardship, appropriate granularity, and governance that allows evolution as work evolves.

Context must be deliberately layered onto capability at the point of delivery and assessment. Separating capability from context only works if context is reintroduced intentionally, not ignored.

Close

The National Skills Taxonomy is becoming the language of the workforce. Industry is articulating demand in skills terms. Workplaces remain where capability is applied. VET is the bridge between them.

The Future Skills Organisation’s digital capability trial shows that the bridge can be built differently. The question now is whether the sector treats it as a precedent worth extending, or as an exception to be admired and then ignored.

Disclosure

The Learning Resources Group builds tools in this space, including RPL Robot, which maps workplace evidence to units of competency. A system that treats foundation capability as portable across industries would expand the relevance of those tools. That interest is disclosed directly. The argument for extending the FSO model stands on its merits regardless.

About the Author

Matt Peachey