The Flattening Game

Dieser Artikel ist auf Englisch verfasst. / This article is written in English.

49,000 managers out of work. Reason: efficiency. My first association when I heard about "The Great Flattening"? A computer game from my youth. Populous. You play a god who reshapes the landscape — flattening mountains, filling valleys, forming the terrain. Terraforming. Piece by piece you level the hill until flat land appears.

That's what it feels like when organisations remove their hierarchy layers. A divine intervention from above. The mountains — the management levels — are stripped away. Flat land shall emerge. More efficient. Faster. More agile.

But in the game you learn quickly: flat land alone isn't enough. Without drainage it becomes swamp. Without paths it stays inaccessible. Without structure nobody settles.

Abolishing hierarchies sounds like a liberation strike for autonomy. In reality it's often just cheaper.

The Handelsblatt reports: 49,000 managers were registered as job-seeking in 2025. 14 percent more than the previous year. The highest level in ten years. Microsoft, Google, Meta cut middle management roles. Bayer halved its management positions under CEO Bill Anderson. The justifications always sound the same: agility, customer proximity, faster decisions, less bureaucracy. Nobody says: fewer salaries.

And nobody asks: What happens to value creation?

Because that's really what this is about. Organisations don't exist to maintain hierarchies — they exist to create value. For customers. For stakeholders. For society. Leadership is not an end in itself. It is a means to fulfil that purpose.

The actual question is therefore not: How many managers do we need? But rather: Which leadership structure maximises our ability to create value?

The Thinking Error: Position is Not Function

Here begins the actual thinking error that underlies the Great Flattening.

A position is a box in the organisational chart. You can delete that. A function, on the other hand, is what that box did: giving feedback, resolving conflicts, making decisions, developing people, coordinating. You cannot delete that. It doesn't disappear.

When you remove a management level, the functions migrate somewhere. Either to other leaders who become overloaded. Or to teams who weren't trained for it. Or to processes and structures that still need to be built. Or to nobody — and that's what we then call chaos.

The question is not whether leadership disappears. The question is where it goes. Whoever doesn't answer that before cutting has no strategy. This endangers the entire organisation.

The Two Forms of Leadership

To understand what's at stake in flattening, we must first distinguish two fundamentally different forms of leadership. Both are necessary. Both have limits.

The first form I call living leadership. A person observes, judges, decides. Situationally. Flexibly. Humanly. Living leadership sees that Sarah is burning out. It recognises the potential in Thomas. It leads the conversation that hurts and yet helps. This form of leadership responds to the unpredictable, the individual, the human.

The second form I call consolidated leadership — frozen leadership. By this I mean decisions that were once made and then frozen into structures. Processes. Rules. Standards. Systems. A good example is ISO 27001: in a security incident everyone knows what to do. Nobody needs to ask. Nobody needs to decide. The process provides the answer. Consolidated leadership scales. It relieves cognitive load. It creates consistency.

The image: Living leadership is the gardener walking through the greenhouse — he sees which plant needs more water, which needs repotting. Consolidated leadership is the greenhouse itself — it protects against frost, regulates temperature, creates the conditions for growth. Both are needed. The greenhouse alone lets plants wither. The gardener without a greenhouse fights the elements.

A formula for efficiency that emerges from this:

As much consolidated leadership as possible. As much living leadership as necessary.

That sounds uncomfortable. Perhaps even freedom-restricting. But the reverse formula — as much living leadership as possible, as little consolidated as necessary — is equally conceivable. The difference lies in the understanding of autonomy. The first formula enables end-to-end responsibility, higher digitalisation and flat hierarchies — through structure. The second requires more coordination, more alignment, more people who lead.

The Mathematics Behind Flattening

There is a mathematical reason why flattening so often fails. Communication doesn't grow linearly with team size. It grows exponentially.

With 5 people there are 10 communication paths. With 8 people there are already 28. With 20 people the number explodes to 190 communication paths.

The jump from 5 to 20 people costs roughly 4 to 8 percentage points of productivity — depending on how well coordination is structured. In an organisation of 500 employees that's 20 to 40 full-time equivalents going purely towards additional coordination.

Flattening promises efficiency. The mathematics show: it depends. With clearer interfaces, defined processes and consolidated leadership, flattening can work. Without this preparation, coordination overhead eats up a significant part of the savings.

Die detaillierte Herleitung findest du im Anhang (Zahlenbasis).

The Three Tipping Points

From this mathematics, three simple rules of thumb can be derived that show whether the reduction process is going well or poorly.

The first tipping point lies at around 150 people. This is the so-called Dunbar limit. Above this, informal coordination collapses. Gore-Tex has understood this and builds a new factory at 150 employees. Not on principle. From experience.

The second tipping point lies at 5 to 8 people per team. This is the team sweet spot. Above this, coordination load explodes. Jeff Bezos' Two-Pizza-Rule is very pragmatic here and can also be derived mathematically in terms of team efficiency.

The third tipping point lies at 8 to 15 people. This is the span-of-control boundary. A person cannot effectively manage more than that in a disciplinary sense. Except: when consolidated leadership takes over part of the load.

Whoever ignores these limits isn't building a flat organisation. They're building a dysfunctional one.

The Three Pathologies of Naive Flattening

What happens specifically when management levels are cut without explicitly reassigning the functions? Three predictable and damaging patterns emerge.

The first pathology is the shadow hierarchy. Whoever is loud leads. Whoever has relationships decides. Without transparency, without legitimation, without accountability. That is arbitrariness with a friendly face. In a formal hierarchy you at least know who decides. You can question it, you can complain. In a shadow hierarchy, informal power rules — and nobody is responsible.

The second pathology is the coordination vacuum. Nobody takes over the functions. Teams work past each other. Conflicts smoulder. Decisions are avoided. The no-man's-land between two areas of responsibility grows and grows.

The third pathology is overload. One or two people per team do everything — in addition to their actual work. Burnout. Bottleneck problem. Single point of failure. You abolished the hierarchy. And created informal managers — without title, without authority, without time.

What these pathologies have in common: they all reduce the organisation's ability to create value. Shadow hierarchies waste energy on politics instead of customers. Coordination vacuums lead to duplicate work and missed opportunities. Overload burns out exactly those people who contribute most to value creation.

Naive flattening optimises the cost structure — and destroys the value creation structure in the process.

The Counter-Position: Where Flattening Works

At this point I must fairly present the counter-position. Because flattening can work. Valve, Morning Star and Haier show it.

Valve has a 50-page employee handbook. Peer reviews. Transparent salary formulas. Morning Star is based on written agreements between employees. Hundreds of pages. Per person. Haier has equipped every micro-enterprise with performance metrics, market mechanisms and governance structures.

If you look more closely, you recognise the pattern: these companies have no managers. But they have massive consolidated leadership. That is the difference nobody articulates: they haven't abolished leadership. They've cast it in structures.

What is Really Needed

If flattening is to succeed, it needs seven elements. Without all seven, flattening is not a liberation strike. It is organisational suicide.

• Small teams: 5 to 8 people. Everything above drowns in coordination.
• Clear roles and ownership: Who is responsible for what? From start to finish. No grey zone.
• Defined interfaces: What does Team A deliver to Team B? In which format? The coordination decision is made once. Then it applies.
• Explicit assignment: Somebody — or something — must take over the leadership functions. The question is not whether. The question is who or what.
• Clear empowerment: Everyone knows what they can decide alone. No ambiguity. Autonomy without clarity is paralysis.
• Decision architecture: Who decides what? With whom? By when? What happens in case of dissent?
• Balance: Consolidated leadership for the recurring. Living leadership for the human. Regular review: does the balance still fit?

The Final Calculation

An organisation with 500 people cuts 145 management positions. The salary saving: around 12.6 million euros per year. That sounds like a clear business case.

But the calculation has hidden items. Productivity losses through coordination chaos. Salary increases for those who take on more responsibility. Turnover costs when the best people leave. Depending on the scenario, these hidden costs eat up 30 to over 100 percent of the savings in the first year.

Long-term, flattening can still pay off. The savings accrue every year, the transition costs are one-time. Over five years a positive result remains even in the middle scenario. The decisive question is: does the organisation survive the transition period? And at what price?

But the calculation is still incomplete. It only considers internal costs. What about value creation outward? While the organisation is preoccupied with itself — with coordination chaos, shadow hierarchies, overload — something is happening outside: customers wait. Innovations are delayed. Competitors move ahead. Market opportunities expire.

The actual costs of naive flattening are not in the balance sheet. They are in the orders not won. In the products not developed. In the customers who left.

What Really Happens

Leadership work doesn't disappear. It becomes invisible.

It migrates to the shoulders of informal leaders who have no time for it. To the shoulders of the remaining managers who become overloaded. To the shoulders of teams who weren't trained for it. Or it simply no longer happens.

Productivity doesn't drop immediately. It drops gradually. Over months. Through turnover of the best people. Through burnout of the informal leaders. Through conflicts nobody resolves anymore. Through decisions nobody makes anymore.

After 18 to 24 months, everyone wonders why nothing works anymore.

The actual question is therefore not: hierarchy or no hierarchy? And not: more or fewer managers?

The actual question is: Do you have an effective leadership system? One that fulfils all functions. Whether through people or through structures.

The Great Flattening is no threat to real leadership. It is a threat to pseudo-leadership. For leadership that builds on titles instead of trust. For leadership that builds on position instead of competence.

49,000 managers out of work. That is not per se wrong. It is a bet that the functions land somewhere else. The question is whether this bet was made consciously — with clear planning for where the functions should go. Or whether it was made naively, in the hope that everything will somehow sort itself out.

The god in Populous had an advantage: he could reshape the terrain at any time. Organisations rarely get this chance a second time. Whoever tears down the mountains without knowing what will carry things afterwards is no longer playing — they are losing.

Reflection Questions to Close

Imagine tomorrow all titles disappear from your organisation.

Who would continue to lead? Whom would people continue to follow? Where would coordination continue — because processes exist? Where would chaos emerge?

The answers show where real leadership lives. And where it is missing.

Would anyone notice if your management level wasn't there for two weeks?

If yes: the functions are with people. If no: the functions are consolidated. Or they are missing.

The attention is on the 49,000 who have gone. The actual question is: what happens to those who stayed?

Quellen

1. Obmann, C., Fröhndoff, B., & Scheppe, M. (2026). Manager: Zahl arbeitsloser Führungskräfte erreicht Höchststand. Handelsblatt. handelsblatt.com
2. ETtech (2025). Tech layoffs: Google, Microsoft, Meta among companies leading 2025 layoffs. The Economic Times. economictimes.indiatimes.com
3. Baltzner, S. (2024). Bayer-Chef Bill Anderson will viele Führungspositionen streichen. FAZ.NET. faz.net
4. Bayer (2026). Our Mission & Strategy -- Dynamic Shared Ownership. bayer.com
5. Care, D.S.A.F.O.-E. Can Bayer CEO Liberate Pharma from Stultifying Bureaucracy? American Enterprise Institute. aei.org
6. Brooks, F. P. (1975). The Mythical Man-Month: Essays on Software Engineering. Addison-Wesley.
7. Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22(6), 469--493.
8. Hackman, J. R. (2002). Leading Teams: Setting the Stage for Great Performances. Harvard Business School Press.
9. Lanaj, K., & Hollenbeck, J. R. (2015). Leadership Over-Emergence in Self-Managing Teams. Academy of Management Journal, 58, 1476--1494.
10. Berger, C. (2023). 2023 broke middle managers' spirit. Fortune. fortune.com
11. Hamel, G., & Zanini, M. (2018). The End of Bureaucracy. Harvard Business Review, 96(6), 50--59.
12. Amerland, A. (2021). Warum flache Hierarchien Unternehmen nutzen. Springer Professional. springerprofessional.de
13. Eseryel, U. Y., Crowston, K., & Heckman, R. (2021). Functional and Visionary Leadership in Self-Managing Virtual Teams. Group & Organization Management, 46, 424--460.
14. McGovern, M. (2025). "Great Flattening": 5 Big Ways This Trend Will Impact Work. HR Morning. hrmorning.com
15. Redaktion Markt und Mittelstand (2025). Deutlicher Rückgang bei Top-Führungspositionen in Deutschland. marktundmittelstand.de
16. Valve Corporation (2012). Handbook for New Employees. PDF
17. Morning Star Company (2012). Colleague Letter of Understanding. opensource.com
18. Bonchek, M. (2016). Power, Pay, Purpose -- With No Managers. platformdesigntoolkit.com
19. Microsoft (2022). Work Trend Index: Great Expectations -- Making Hybrid Work Work. microsoft.com