There is a person in nearly every delivery organization I have worked in who functions as the institutional map. They know which system feeds which, and which vendor contract has the odd indemnity clause buried on page nine. New hires get walked over to their desk inside the first week. Their manager calls them indispensable and means it warmly. Their title has not changed in years.

We tend to read that as a matter of temperament, a question of whether somebody wanted more. A report released in June by the Burning Glass Institute and the NYU School of Professional Studies argues it is closer to a structural condition, and that the scale of it is hard to overstate. Roughly one in four mid-career professionals in the United States, 24.2 percent, is stalled, meaning five or more years with no meaningful promotion and negligible wage growth while still fully employed.1

The gap is too small for your review cycle to catch

The number that ought to bother an executive sits further down the summary. At the ten-year mark, the people who would go on to stall had already fallen behind, averaging 1.5 promotions and 45 percent wage growth against 1.9 promotions and 66 percent growth for the peers who never stalled.1 Four tenths of a promotion. Spread across a decade, that difference disappears into the noise of any single year. No rating captures it. Compensation bands are wider than the gap itself, so nothing flags it there either, and the person living it has no way to see it from the inside, because their own year looked fine.

Which is where the real trouble sits. Every instrument an organization owns for looking at a person takes an annual reading. The performance review asks how the last twelve months went, and the merit cycle asks the same question with money attached. What the research describes accrues across ten of those cycles, and it accrues as an absence. There was no bad quarter and no failed project. The person simply did not move, and nothing in the process was built to treat stillness as an event worth logging.

Every instrument you own reads one year at a time. The thing you are trying to catch takes ten years to become visible, and it arrives as a nonevent.

Where the money is, the stall is worse

The industry breakdown runs against intuition. Some of the highest rates of stalled talent sit in well-paying, high-growth sectors, with public administration and finance among them.1 A company can be adding headcount and posting roles at a healthy clip while the people already inside it go nowhere, because expansion and internal movement run on separate machinery and only one of them belongs to anybody in particular.

Here is the piece that connects them. Research from The Josh Bersin Company and AMS, published in 2023 and drawing on roughly 600,000 hire records across five years, found internal hires had fallen to about 24 percent of all hires, down from a pandemic peak of 40 percent in 2020 and below the 30 to 32 percent that had been more usual before that.2 Both firms sell talent advisory and recruitment services, so weigh the framing accordingly, though the finding is not especially flattering to the market they sell into. Three quarters of the roles your company opened went to someone who did not work there.

A couple of weeks ago I wrote about managers who hold on to their strongest people because the incentives reward it, and I am not going to relitigate that here. It happens, and it is part of this, though it cannot account for a stall that runs five years across whole industries. Something duller is at work. When a role opens, the first move is almost always to write a job description, and a description written from a blank page describes a stranger. The person two floors down who has been doing four fifths of that work under a different title never enters the comparison, because the comparison was not built to include them.

That is the same reflex I described when a company strips a degree requirement out of its postings and keeps applying it in the room. Written criteria are easy to change. The decision underneath them keeps running on whatever it was already running on.

The intervention costs less than the search

The report also traces how stalled workers get moving again, and the answer is almost embarrassing in how ordinary it is. An adjacent move does most of the work. Computer programmers who shift into data science roles inside the same company reduce their stall risk by as much as 86 percent.1 That is a change of seat in a building they already badge into, without a new employer or a new credential.

What it costs to skip that lands on both parties. For the worker, the report puts the penalty for an average stalled software developer at more than $43,000 in cumulative lost wages across fifteen years.1 For the employer, the cost is less visible and considerably larger, because it takes the form of experienced people who have stopped growing sitting inside a company that is paying external recruiters to find people with roughly their skills.

If you run an organization of any size, you can probably recite your open requisitions from memory. Try the other number. How long has each person on your team held the same seat, and when did each of them last change title or band? Most leaders I work with cannot answer that without asking somebody to pull it, which tells you where it sits in the operating rhythm. The report is not hard to build. Nobody has asked for it.

Here is what I would want to know from anyone who has run a team for more than five years. The last time you looked hard at your people, were you looking at their year or at their trajectory? And if it was the year, as it is for nearly all of us nearly all of the time, what would you have found if you had looked at the decade instead?