Great Flattening

Great Flattening

Great Flattening refers to the removal of middle management layers in large companies, a trend accelerated since around 2023 by cost pressures and automation. Instead of many intermediate bosses, a few managers now oversee significantly larger teams.

Large companies are usually structured like a pyramid. At the very top sits the executive leadership, at the very bottom work the employees, and in between lie several layers of bosses and sub-bosses. These intermediate layers manage small teams, pass on tasks, and report results upward. Great Flattening is the umbrella term for a trend in which exactly these middle layers are being eliminated. Translated, it roughly means “the great flattening”: the pyramid keeps its width but loses floors. The term emerged in business media starting in 2023, when corporations like Meta, Amazon, Google, and Intel simultaneously announced plans to cut management layers.

What the elimination of middle layers means for careers

For companies, the appeal is obvious: management positions are expensive. In the tech industry, a middle manager quickly earns double what an entry-level position pays. Cutting such positions reduces costs without touching the actual work. Amazon CEO Andy Jassy even set a goal in 2024: 15 percent fewer managers relative to the rest of the workforce.

For employees, this changes the path upward. Promotion used to mean climbing rung by rung: first team lead, then department head, then more. When rungs disappear, jumps become bigger and rarer. Anyone wanting to advance has to wait longer or switch companies. Because of this, some firms now offer specialist career tracks where you can earn more without taking on personnel responsibility.

The trend doesn’t affect everyone equally. Particularly affected are layers whose main job was to pass along information and summarize reports. Such tasks can be relatively easily replaced by software. Managers who make substantive decisions or resolve conflicts are less at risk.

How software and cost pressure are flattening the pyramid

Two forces are working together. The first is economic: after the hiring waves of the pandemic years, many tech companies grew too fast. When interest rates rose afterward and investors pushed for profits, headcount was cut. Management was a particularly visible target.

The second force is technical. Much of what middle managers used to do is now handled by software. Project status is visible to everyone in tools like Jira or Asana. Weekly reports can be compiled by AI systems from chat logs and tickets. This removes part of the justification for why a manager can only oversee eight people.

In practice, this means the so-called span of control is growing. This technical term describes how many people report directly to a manager. Six to eight used to be considered normal; today companies report fifteen or more. Critics warn, however, that beyond a certain size, no one gets real feedback anymore. A manager with twenty direct reports can barely help each individual develop.

Great Flattening in quarterly earnings and job postings

In the news, you usually encounter the term around quarterly earnings reports. When a corporation talks about “efficiency,” “streamlining,” or “reducing management layers,” Great Flattening is what’s meant. Investors often react positively, because falling personnel costs boost profits in the short term.

The trend is also visible in job postings. Phrases like “flat hierarchies” or “high degree of ownership” sound like freedom, but often simply mean there are few supervisors. For entry-level workers this can be a good thing, since you get to talk to decision-makers early on. But it can also mean no one has time to onboard you.

A common misconception is equating Great Flattening with a general wave of layoffs. What’s actually meant is something more specific: a redistribution within the structure, not merely its shrinkage. Some companies simultaneously hire specialists and cut management positions. Whether the flat structure works in the long run remains an open question. Earlier waves of this kind often ended with the eliminated layers returning under a new name after a few years.

Related Products

Latest News

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.