A few years ago, many large companies announced that they would stop asking for a university degree for a large share of their roles. They would hire for skills. The decision made headlines, was repeated at HR conferences and became one of the most quoted ideas in the field.
In February 2024, Harvard Business School and the Burning Glass Institute measured what had actually happened. Their report, Skills-Based Hiring: The Long Road from Pronouncements to Practice, concludes that the change showed up in fewer than one in 700 hires the year before. About 45 % of the companies that dropped the requirement went on hiring just as they had before. Almost all the real movement came from 37 % of them.
Dropping the degree is not enough without another proof
The simplest explanation is also the most uncomfortable. The degree was a shortcut: someone outside had already checked something about that person. When the shortcut is removed and nothing takes its place, the interviewer falls back on intuition, and intuition tends to prefer the CV it already knows.
Hiring for skills requires being able to see them. An interview mostly measures how someone talks about what they can do. What the company needs is to watch them do it.
The same happens inside the company
The problem does not stop at the front door. In-house training carries the same gap under another name. Course hours, attendees and satisfaction surveys are counted, and on the day of the annual review nobody can say what a manager can do today that they could not do a year ago.
LinkedIn's Workplace Learning Report 2025, based on 937 learning and HR professionals and 679 employees, shows it clearly. Only 36 % of organisations have strong career development programmes that deliver business results. Even in that group, the most common ways of measuring impact are employee engagement (72 %) and retention (48 %). They are valuable indicators, but they measure the climate training leaves behind, not the competence it produces. And only 15 % of employees say their manager helped them build a career plan in the last six months.
If a company cannot show which competencies its people have developed, it cannot use them to promote, move someone to another role or decide whom to train next. Training stays a well-meant expense.

Three steps to make competence leave a trace
At Kudzu we have spent years working with more than 20 universities and business schools on a very similar problem: how to show an accreditation committee that a student can do something, not just that they attended class. We sum it up in three verbs, and they carry over well to the company.
Practise in an environment where decisions are recorded. A business simulation has a team set prices, hire, invest or negotiate over several rounds, with consequences that build up. Every decision is stored, as is the reasoning the team presents afterwards. That is already evidence: it shows how someone prioritises with little information, how they react when a result goes wrong and how they share the work with others. In traditional training none of that is visible.
Assess from more than one angle. What is observed in a simulation is one part. The other is what the people who work with that professional see every day. A competency-based 360° assessment gathers the views of their manager, peers and team on specific behaviours, not on general impressions. Done before and after a programme, it lets the company stop assuming the training worked and start seeing where something changed and where it did not.
Accredit what has been demonstrated. The last step is to give that evidence a shape that is useful outside the classroom. The European Union has been pushing in this direction for some time: in June 2022 the Council adopted a Recommendation on a European approach to micro-credentials, which certify the learning outcomes of short learning experiences. The key word is outcomes. An internal credential that states which competency was assessed, by what method and at what level is worth far more than a certificate of attendance, because another manager can read it and make a decision with it.
Where to start
There is no need to redesign the whole training plan. It works better to start with a single programme and a single business question.
First, choose a competency with visible consequences. For example, a manager's ability to prioritise when the budget is cut, or a sales team's ability to negotiate without giving away margin. The closer it is to a real decision, the easier it is to observe.
Second, design the practice so that the competency shows up. If negotiating is what matters, the simulation has to force negotiation with incomplete information and a cost for getting it wrong. A case that is only read and discussed does not leave enough of a trace.
Third, measure before and after with the same questions and the same people. A short 360° at the start and another three or four months later say more than any end-of-course satisfaction survey.
Finally, write down what has been demonstrated, in a form that serves the next development or promotion conversation. It is the step most often forgotten, and the one that turns training into information for decisions.
The evidence is for the person too
All of this is usually framed as a control tool for the company, and that falls short. Whoever goes through a programme like this takes away something a course rarely gives: knowing what they do well, what they find hard and how others see them, with concrete examples. That information is the basis of a serious career plan, exactly what the LinkedIn report finds missing.
We discuss this in more detail, from the university side, in our CLADEA 2026 recap, and how AI is changing what is worth training in the company in AI adoption at work: the capacity is already paid for. If your organisation is already measuring training in a different way, we would very much like to hear how you do it.