The skill that got you hired is already expiring
What you are good at now decays on a clock. That does not mean you are becoming obsolete. It means the question worth asking has quietly changed.
IBM puts the half-life of a professional skill at about five years, and the half-life of a technical one at roughly two and a half. Read that as a physicist would. A half-life is the time it takes for half of something to decay. So the specific, hard-won competence that got you hired is, by this measure, half gone before your third work anniversary. Not because you stopped being good at it. Because the ground it stood on moved.
This is a genuinely new fact about work, and most of us are still carrying an older one around. The older fact was that skill compounds. You learned a trade, got better at it every year, and the value stacked. Seniority meant accumulated mastery, and mastery was a kind of savings account you drew down slowly toward the end. That model was true for a long time. It is quietly ceasing to be true, and the people it is failing hardest are not the unskilled. They are the competent ones who assumed the account would keep earning interest on its own.
The evidence that the ground is moving
LinkedIn, which watches more job transitions than any other single source, estimates that by 2030 70 percent of the skills used in most jobs will change, with AI as the accelerant. In the same analysis it notes that people entering the workforce today are on pace to hold twice as many jobs over a career as someone who started fifteen years ago. The unit of a working life is shrinking from the career to the stint, and each stint asks for a partly different toolkit than the last.
The employers see it too. In the World Economic Forum's 2025 Future of Jobs report, companies expect 39 percent of workers' core skills to change or become outdated by 2030. That figure is often quoted as alarming, and it is, but the more interesting number is the direction: it was 44 percent in the 2023 edition. Even as the survey ticks down slightly, two of every five things you know how to do are on a countdown. The Forum estimates that 59 of every 100 workers will need retraining by the end of the decade to stay useful, and that 11 of those 100 will not get it.
None of this reads like a story about lazy or fragile workers. It reads like a story about a shelf life that used to be measured in decades and is now measured in single-digit years. The competence is real. It is just perishable in a way it did not used to be.
What actually expires
Here is the part the half-life number hides. Not all of you decays at the same rate. A skill is really two things wearing the same word. There is the perishable layer: the specific tool, the current framework, the exact way this year's software wants you to do the thing. And there is the durable layer underneath: the judgment about what matters, the pattern recognition you only get from having been wrong a hundred times, the taste for a good answer over a clever one. The perishable layer is what has the two-and-a-half-year half-life. The durable layer barely decays at all.
The Forum's own ranking of what employers want most says the same thing without meaning to. The skills climbing the list are not this quarter's technology. They are analytical thinking, resilience, flexibility, curiosity, and the willingness to keep learning. Those are not skills so much as the machinery that acquires skills. The market is quietly repricing the ability to replace expired knowledge above any particular piece of knowledge you currently hold.
The durable asset was never the skill. It was the capacity to grow the next one, and the room to do it in.
This matters because of what people do when they feel their expertise slipping. The instinct, especially for someone who has been good at their job for a long time, is to grip the perishable layer harder. To defend the tool you mastered, to treat the new way as a fad, to read every shift as evidence that you are being left behind. That instinct is the actual danger, and it has a name in these pages already: the slow atrophy of being overqualified on paper and underused in practice, where the expensive skill you are proud of is exactly the one the world stopped needing.
The question changes
If your specific expertise is always halfway to expired, then the anxious private question, am I still good enough, is close to meaningless. You are always simultaneously good enough at the durable layer and behind on the perishable one. Everyone competent is. The question that actually predicts how the next five years go is a different one, and it points outward, at the job rather than the self: does this place renew me, or is it quietly spending me down?
Some jobs are renewal engines. They put you next to problems you have not solved before, give you time and cover to be a beginner again, and treat your learning as an asset they are building rather than a cost they are tolerating. Other jobs are extraction engines. They found you at peak competence and have been drawing on that stored expertise ever since, asking you to run the same play at higher volume, never once refilling the account. From the inside the second kind can feel safe, even flattering. You are the expert. You are needed. But if the skills you are being valued for have a half-life, an employer who values only your existing stock, and never your capacity to build the next one, is holding a depreciating asset and paying you like it appreciates. That arrangement ends, and it rarely ends on your schedule.
The mobility data suggests it ends sooner than people plan for. US median job tenure has fallen to 3.9 years, the lowest since 2002. Whether you leave or the job does, the average working relationship now lasts less time than a technical skill's full decay curve. You cannot wait out a shift by staying put and hoping the old competence holds. The environment that keeps replenishing you is not a luxury on top of the job. In a world of shrinking half-lives it is the job's single most important feature, worth more than the title and, over a decade, worth more than the pay.
So the reframe is not that you are becoming obsolete, and it is not that you must frantically re-skill to stay ahead of a machine, which is its own kind of tax on your attention and nerves. It is quieter and more practical. Stop auditing your skills for signs of decay, because decay is now the baseline condition and it says nothing about you. Audit the job instead. Ask whether the place you spend your days is renewing your capacity or drawing it down. A role that keeps teaching you fits the era you actually work in. A role that only harvests what you already knew has stopped fitting, and no amount of gripping the old skill tighter will change that. The skill was always going to expire. The right environment is the thing that lets you keep making new ones, which is also, not by accident, what makes starting over in the middle of a career feel normal rather than like failure.