Wider Spans, Thinner Support: The Middle-Manager Squeeze Isn't Really About Team Size

Two numbers from Gallup's newest span-of-control research refuse to sit comfortably beside each other. The first is the one every CHRO has already seen: the average manager went from 10.9 direct reports in 2024 to 12.1 in 2025, a jump that puts team sizes nearly 50% above where they were in 2013. The second is the one nobody quotes: the median manager still oversees just five or six people, and 66% of managers have fewer than 10 reports at all.

Both figures come from the same January 13, 2026 Gallup analysis by chief workplace scientist Jim Harter, built on a meta-analysis of 92,252 teams across 104 organizations, 26 industries, and 46 countries, covering 897,971 employees. Read together, they describe something more specific than "managers are overloaded." The average is being dragged upward by a minority carrying very large teams: 22% of managers now have 10 to 24 reports, and 13% have 25 or more.

That distinction matters, because the story companies are telling themselves about flattening is a story about headcount, and headcount turns out to be a poor predictor of whether the manager in question is about to break.

The Flattening Is Real. The Diagnosis Isn't.

Start with the scale of the cuts, which is not in dispute. A Korn Ferry survey of 15,000 professionals reported by Fortune in June found that roughly 41% said their companies trimmed management layers in the past year. Forbes contributor Cindy Rodriguez Constable, writing on June 30, pulled together the supporting ledger: manager headcount at public companies down 6.1% between May 2022 and May 2025 per Live Data Technologies, Google running 35% fewer managers of small teams year over year, Amazon shedding 30,000 corporate roles across two rounds. Gartner's October 2024 forecast that one in five organizations would use AI to flatten structures and eliminate more than half of existing middle-management roles by 2026 has aged into something closer to a description.

The engagement damage is real too. Gallup's State of the Global Workplace 2026 report, drawn from 263,810 respondents surveyed through 2025, puts global manager engagement at 22%, down from 31% in 2022. Managers fell further and faster than any other worker category. In organizations Gallup classifies as best-practice, 79% of managers are engaged, which is a useful reminder that 22% is a choice rather than a law of physics.

Here is where the standard diagnosis goes wrong. When Gallup broke engagement down by team size, the relationship was not linear and it was not primarily about the number of people. What separated the managers who held up from the ones who didn't was how much of their week went to their own individual contributor work. Managers spend a median of 40% of their time on non-managerial tasks. Those who stay below that 40% line maintain engagement above the 37% average regardless of how many people report to them. Those above it decline as their teams grow. Manager talent and work location moved the results too: high-talent managers held engagement even with 25-plus reports, while remote and hybrid managers struggled as spans widened no matter how capable they were.

Gallup's own conclusion is blunt about the trade-off. "The ideal span of control isn't universal," the analysis states, before landing on the test that matters: larger spans can improve efficiency without weakening performance, but only if organizations can honestly answer yes to a set of questions about manager capability and time. If they can't, "short-term savings may undermine the daily manager-employee relationships that support long-term results."

Nobody Is Training the People Absorbing the Work

The flattening logic assumes the remaining managers can simply do more. What it rarely funds is the capability that assumption requires. Skillsoft research surveying 1,000 HR and L&D professionals across the U.S., UK, Germany, and Australia found leadership among the three most significant skill shortages, with 26% reporting that employees are being promoted without adequate preparation and 31% citing increased manager stress from a distorted view of workforce capability.

"Business transformation depends on the strength, adaptability, and skills of the workforce. Skills gaps are already impeding growth, and traditional approaches to talent development are only worsening the issue." – Ciara Harrington, Chief People Officer, Skillsoft

The tension is straightforward once you line the numbers up. Organizations are removing the layer where most managers historically learned the job, widening the spans of the people who remain, and then asking those people to coach more employees with less proximity to each one. Korn Ferry's respondents noticed: 43% said leadership alignment suffered after the layer cuts. DDI's 2025 Global Leadership Forecast, covering more than 10,000 leaders across 50 countries, found trust in immediate managers at 29% and leadership bench strength at 20%, with 40% of stressed leaders having considered leaving leadership altogether.

The Bench You Stop Building Now Is the One You'll Need in 2036

Every efficiency argument for flattening is an argument about this quarter. The counter-argument is about a decade from now, and it is the one most likely to be right.

"The biggest risk isn't necessarily fewer middle managers today, it's what happens 10 years from now if fewer people are getting the experience they need to become senior leaders and experts." – Max Martina, President, Cambridge Leadership Associates

That risk compounds quietly, because nothing visibly fails in the year you cut the layer. The org chart gets cleaner, the salary line drops, and the surviving managers absorb the difference by working longer and coaching less. Three or four years later the succession slate is thin, the internal candidates have less real management reps than their predecessors did, and the company starts paying outside-hire premiums for skills it used to grow.

The practical response is not to refuse to flatten. It is to stop treating span of control as the only lever and start managing the variable Gallup's data actually implicates:

• Measure individual contributor load, not just report counts. If a manager is spending more than 40% of the week on their own deliverables, widening their span is a bet against Gallup's findings.
• Fund the first-time manager transition before the promotion, not after. With 26% of HR and L&D professionals reporting employees promoted without adequate preparation, the training gap is a known input, not a surprise.
• Treat remote and hybrid managers as a distinct risk group. Gallup found they lose engagement as spans widen regardless of talent level, which makes distributed teams the wrong place to test a bigger span.
• Protect the roles where future leaders get their reps. Bench strength at 20% in DDI's forecast is the leading indicator of the hiring bill you'll face in 2030.
• Set a weekly feedback floor and staff to it. If a manager cannot hold a meaningful conversation with every direct report on a regular cadence, the span is too wide for that manager, whatever the benchmark says.

The flattening wave was sold as removing a layer of overhead. What the 2026 data suggests is that companies removed a layer of apprenticeship and handed the bill to the people left standing. Twelve reports is not the problem. Twelve reports, forty percent of the week spent on your own deliverables, and no one who ever taught you how to do either, that's the problem.