The 'great flattening' is costing more than organisations bargained for

HR leaders are facing a defining question as artificial intelligence accelerates the restructuring of corporate hierarchies: when you remove middle management, where does the leadership actually go?

The so-called "great flattening" is reshaping workplaces at pace. Manager headcount at public companies fell 6.1% between 2022 and 2025, according to a Lepaya study, with companies including Meta, Amazon, Google, Intel and more flattening their hierarchies in pursuit of faster decisions and lower costs. But the pitch for leanness is increasingly complicated by what gets left behind.

Star Levandowski, senior vice president of business and people operations at Deputy, argues that the hidden costs of flattening are only now beginning to surface.

"The great flattening can absolutely create meaningful efficiency gains, but only when organisations are honest about the role middle managers actually play," Levandowski said.

"Too often, companies remove layers without redesigning the systems that support communication, coaching, decision-making, and culture. When that happens, the 'savings' tend to resurface elsewhere – through burnout, role confusion, stalled development, and erosion of trust."

The connective tissue problem

The crux of the issue, Levandowski argues, is that middle managers are rarely just administrators. They are the connective tissue between strategy and execution – the people who translate organisational priorities into daily action on the ground.

That function becomes especially acute in frontline environments, where managers are simultaneously coordinating schedules, managing compliance, supporting customers and making real-time operational calls. Strip that layer without replacing what it did, and the pressure does not disappear.

"When those layers shrink without the right systems and support in place, the pressure doesn't disappear – it shifts," Levandowski said. "Often onto already stretched frontline workers and senior leaders."

The data bears this out. The average number of direct reports per manager reached 12.1 in 2025, according to Gallup – up from 10.9 in 2024 and a 50% increase from 2013, when Gallup first collected this data.

Meanwhile, 37% of employees say the lack of managers has left them feeling directionless, and nearly half of senior executives doubt their ability to manage everything that has landed on them, according to Korn Ferry research from 2025.

For HR leaders thinking through how workforce restructuring affects retention and engagement, these numbers represent more than an operational headache – they are a signal that structural decisions are creating culture problems.

What doomjobbing is really telling you

One of the more visible symptoms of this pressure is the rise of "doomjobbing" – the behaviour of employees who, resigned to uncertainty about their future, spend increasing time passively searching for other roles while still showing up for work.

Levandowski sees it as a diagnostic rather than a discipline problem.

"When doomjobbing becomes widespread, it's usually signalling something much deeper than employees casually exploring the market," she said. "It's often an indicator that people are burned out, uncertain about their future, and losing confidence in their ability to grow within the organisation."

The triggers, she explains, are specific: career stagnation, job instability, reduced access to manager support, and a lack of clear communication about how AI adoption will affect their roles. When those factors converge, job searching becomes a way for employees to reclaim a sense of agency.

"The real question leaders need to ask is not, 'Why are employees looking elsewhere?' but, 'What clarity, support, or trust are employees currently missing here?'"

HR leaders navigating the operational and human dimensions of AI implementation will recognise the warning: doomjobbing is not a productivity issue to be managed; it is a trust deficit to be addressed.

AI as a tool, not a substitute for leadership

Levandowski is clear-eyed about what AI can and cannot do in a flattened structure. Deputy's own research – the Better Together Report – found that 98% of workers believe human connection is essential to their jobs. Empathy, judgment, coaching and trust are not functions that can be automated.

The most effective use of AI in frontline environments, she argues, is eliminating administrative friction so that managers can spend more time on people – not replacing the manager altogether.

"At Deputy, we're seeing strong adoption of AI when it helps managers turn everyday operational decisions into fast, controlled actions," Levandowski said. "Whether it's adjusting shifts, comparing timesheets to schedules, approving matching timesheets, or identifying who is best positioned to fill an open shift – the goal is the same: reduce manual work so leaders can focus on the parts of the job that matter most."

That distinction matters for organisations watching how leading companies are using workforce technology to navigate this transition. AI adoption and structural leaning are not the same strategic lever.

Deputy's Big Shift 2026 Report, which drew on more than 41 million shifts and 268 million hours worked, found that frontline work is structurally resilient to the displacement risks threatening office-based roles – but only when businesses pair technology adoption with transparency and clear communication about how tools are being introduced.

That transparency gap, Levandowski said, remains one of the most pressing challenges for HR.

"Eighty per cent of workers in our research said their workplace does not clearly communicate how AI tools are introduced or used," she said. "Efficiency alone is not a strategy. Organisations that flatten successfully are the ones pairing technology adoption with transparency, clear communication, and intentional leadership."

The implication for HR leaders is not to resist restructuring, but to own it. Removing layers without rebuilding what those layers provided – coaching, clarity, connection – creates organisations that look leaner on paper but feel far less stable to the people doing the work.