Most learning teams know the feeling. A new platform arrives, a course is commissioned, people are trained, and within a couple of years the tool has changed, been replaced, or been quietly absorbed by something else. The effort was real. The half-life was short. And the cycle begins again.
There is a calmer, more sustainable way to think about development, and it starts with a simple distinction: some capabilities last, and some are built to be refreshed. Getting that distinction right is one of the most practical decisions a Learning and Development function can make.
The pace of change is not a rumour; it is measured. According to the World Economic Forum's Future of Jobs Report 2025, employers expect 39% of key skills required in the job market will change by 2030, a figure that represents significant disruption but is down from 44% in 2023. Encouragingly, a growing focus on continuous learning, upskilling and reskilling programmes has enabled companies to better anticipate and manage future skill requirements.
The scale of the shift is reflected in the Forum's broader ambition. Its Reskilling Revolution initiative is working with over 350 organisations to enable 1 billion people with access to better education, skills and economic opportunities by 2030.
The practical problem is that not all skills decay at the same rate. Research cited by IBM suggests that skills generally have a "half-life" of about five years, with more technical skills at just two and a half years. As IBM frames it, perishable skills, with a half-life under 2.5 years, are specific technology skills that are updated frequently, along with organisation-specific policies, tools and specialised processes.
Chasing every one of those perishable skills as though it were permanent is exhausting and, ultimately, a losing strategy. The alternative is to be deliberate about what you anchor to.
The useful mental model separates two layers.
Durable capabilities are the foundations that hold their value for years. Avilar describes durable skills as having a half-life of more than 7.5 years, semi-durable skills as 2.5 to 7.5 years, and perishable skills as a half-life of less than 2.5 years, being specific technical skills related to particular vendors, platforms, or programming languages that must be updated frequently. Durable capabilities include analytical thinking, problem-framing, communication, resilience and learning agility.
Perishable, tool-specific skills are the renewable layer: the particular software, platform or process that matters now and will be replaced later.
This is not a theoretical nicety. The labour market rewards the durable layer heavily. Analysis by America Succeeds, conducted with Emsi Burning Glass across more than 82 million job postings, found that seven out of the ten most-requested skills in job postings are durable skills, and employers seek these skills nearly four times more frequently than the top five technical or hard skills. A later update found that 76% of all job postings require at least one durable skill, up 12 percentage points from 2021.
This direction of travel aligns with where emerging skills are heading. The Future of Jobs Report notes that creative thinking and resilience, flexibility and agility, and curiosity and lifelong learning are rising in importance. These are durable by nature. They do not expire when a tool does.
A common mistake is to build a capability framework that tries to cover every desirable trait for every person. The result is a document nobody uses and a development plan nobody can prioritise.
Jobs tend to lean on a small cluster of capabilities for success. A useful discipline is to cap your framework: a handful of durable competencies that genuinely matter for the role, plus a few role-specific skills that need refreshing. The aim is clarity, not completeness.
A structured measure helps here. Rather than guessing which durable strengths matter most, assessment tools such as the TTI DNA assessment of transferable workplace competencies can help identify which durable capabilities a specific role genuinely leans on. The same logic underpins assessing learning agility when hiring for potential and the broader move toward skills-first approaches: define the capability precisely, then measure it fairly.
IBM offers a helpful image for sequencing this. It suggests thinking of skills as a tree, where durable skills form the roots, semi-durable frameworks form the branches, and more perishable skills come and go like the leaves with the changing seasons. You invest most heavily in the roots. You expect the leaves to turn over.
Once you separate the two layers, your design choices become clearer.
Keep the durable base stable. Treat foundations like analytical thinking, communication and resilience as long-term investments. These justify deeper, structured programmes and sustained practice over time.
Refresh the perishable layer little and often. Short, frequent updates beat annual crash courses for tool-specific skills. This is where microlearning and learning in the flow of work earn their place, because the content will change again soon anyway.
Audit your framework for layer confusion. Walk through each competency and label it durable, semi-durable or perishable. If a "core competency" is really a passing tool, move it to the refreshable layer and stop treating it as permanent.
This matters commercially as well as individually. The Future of Jobs Report 2025 projects that if the global workforce were represented by 100 people, 59 would require reskilling or upskilling by 2030, with 11 unlikely to receive it. Organisations that know where their durable strengths sit are far better placed to redeploy people as tools change, rather than starting from zero each time.
In a fast-moving environment, the temptation is to keep chasing the newest tool, as if staying current means staying trained. But perpetual tool-chasing drains energy and rarely compounds. Durable capabilities allow people and organisations to adapt to the next tool rather than be overtaken by it.
Build on what lasts. Refresh what doesn't. Keep the two clearly separated, and development becomes something you sustain rather than something that sustains you.
The reflective question to sit with is this: in your own capability framework right now, how much of what you call "core" is genuinely durable, and how much is simply a tool you will be refreshing again within two or three years?