In many companies, centralisation does not look like a problem.
It often looks like proof of efficiency.
The leader knows. The leader decides. The leader follows up. The leader knows the important clients. The leader keeps the priorities, emergencies, exceptions, promises and risks in their head.
At first, this centralisation makes it possible to move quickly.
When the company is small, that is normal. There are few people, few processes, a great deal of intuition and direct proximity to the work. The leader compensates for gaps, absorbs uncertainty and turns energy into movement.
Then the company grows.
There are more clients. Issues multiply. Sales channels accumulate. Teams expand. New tools are introduced. Decisions become more frequent, more nuanced and more interdependent.
But the organisation remains built around one reflex: ask the leader.
That is when the hidden cost begins to appear.
Centralisation slows the company before it blocks it
An over-centralised company does not seize up all at once.
It slows down first.
A decision waits for approval. A follow-up waits for a judgement call. A proposal waits to be reviewed. A priority waits to be clarified. An opportunity waits for the leader to decide whether it deserves attention.
Each delay seems reasonable. Every question can be justified. The problem is their accumulation.
Over time, the leader becomes the mandatory route for too many flows.
They are not deliberately blocking the company. They have simply become its natural point of congestion.
And the more capable they are, the longer the system can remain in this state. Because they respond quickly, they compensate. Because they understand everything, they absorb. Because they care about quality, they check. Because they want to protect the company, they intervene.
The problem is therefore not an individual weakness. It is a fragile architecture.
The cost does not always appear in the accounts
The cost of an over-centralised company is not always visible in the profit and loss statement.
It hides elsewhere.
In sales opportunities lost because nobody followed up. In decisions made too late. In teams waiting for approval. In issues escalated unnecessarily. In projects that move only when the leader pushes them. In clients who are served properly, but whose relationships are never developed into growth opportunities.
This cost is rarely measured.
Yet it weighs on performance.
A centralised company can be busy and still lose considerable value through its blind spots. It can win clients while neglecting opportunities. It can be profitable while exhausting its leader. It can grow while becoming more fragile.
Centralisation does not always destroy growth. It may simply make growth heavier, slower and more dependent.
The leader’s fatigue becomes a system indicator
When everything depends on the leader, fatigue is no longer merely personal.
It becomes an organisational indicator.
A leader may be tired because they work long hours. But they may also be tired because the company uses their attention badly.
Their energy is spent on micro-decisions. Their time is spent retrieving information. Their concentration is consumed by issues that should be clarified elsewhere. Their memory becomes a production tool. Their intuition becomes an emergency system.
This is often where the mental load becomes abnormal.
The leader is not simply busy. They are carrying too many invisible elements.
They know that a prospect needs following up. They know that a client relationship is sensitive. They know that an offer has been misunderstood. They know that a colleague lacks key information. They know that a project is not defined clearly enough. They know that a number looks good but conceals a problem.
This knowledge is valuable. But when it remains concentrated, it makes the organisation dependent.
Teams become cautious
An over-centralised company often creates a secondary effect: teams become cautious.
They do not necessarily take fewer initiatives because they lack ability. They take fewer because the system teaches them that final approval sits elsewhere.
When important decisions systematically return to the leader, the team learns that reflex. People wait. They ask. They seek reassurance. They avoid taking the risk of making the judgement themselves.
This behaviour can easily be mistaken for a lack of autonomy.
But autonomy cannot simply be declared. It has to be built within a framework.
For a team to make better decisions, it must understand the rules of the game: the priorities, criteria, limits, responsibilities, signals that must be escalated and decisions that can be taken without approval.
Without this framework, asking for more autonomy amounts to asking teams to guess.
And when teams have to guess, they naturally return to the leader.
Tools do not solve centralisation on their own
Many companies try to solve this problem with tools.
A CRM, a dashboard, project management software, a knowledge base, an automation or a shared note-taking space.
These tools can help. But they are not enough.
If decision criteria remain implicit, the tool merely moves the ambiguity elsewhere. If responsibilities are unclear, the tool displays tasks without genuine ownership. If the sales pipeline is not structured, the CRM becomes a contact database. If management cadences do not exist, dashboards are viewed without prompting decisions.
Centralisation is not solved simply by adding an interface.
It is solved by making explicit what was previously implicit.
What needs to leave the leader’s head
The first step is not to delegate everything immediately.
The first step is to identify what lives too strongly in the leader’s head.
There are usually five categories.
Priorities: what genuinely matters now, what can wait and what must be refused.
Decision criteria: what makes a prospect attractive, a client risky, an opportunity important or a request acceptable.
Sensitive information: promises made, points requiring attention, weak signals, and commercial or operational risks.
Next actions: who must do what, by when and with what degree of urgency.
Implicit rules: the things the leader knows but nobody has properly formalised.
As long as these elements remain invisible, the company depends on the person carrying them.
Decentralising does not mean withdrawing
Decentralising does not mean that the leader should disappear.
Quite the opposite.
A useful leader is not absent. They intervene at the right level.
They do not spend their time restating the obvious, retrieving information, correcting oversights or approving every micro-decision. They apply judgement where their attention genuinely matters: strategy, positioning, structural trade-offs, recruitment, offers, key accounts, investments and major risks.
To make that possible, the company needs relays.
Human relays, of course. But also methodological ones: a clear pipeline, short management cadences, visible criteria, explicit responsibilities, accessible information and documented decisions.
This combination allows the leader to regain perspective without losing control.
The two-week test
A simple test reveals the true degree of centralisation.
What happens if the leader disappears for two weeks?
Are important prospects followed up? Do sales decisions move forward? Do teams know what takes priority? Are sensitive clients looked after? Are blocked issues identified? Are the useful figures readable? Are genuinely necessary decisions distinguished from requests made merely for reassurance?
If everything slows down immediately, the company does not merely have a delegation problem.
It has a systems problem.
This is not a theoretical test. It reveals the organisation’s invisible level of dependency.
Begin with one recurring point of friction
Trying to decentralise everything at once is unnecessary.
It is often counterproductive.
The right place to begin is a recurring point of friction: somewhere the leader is repeatedly asked to intervene for the same reasons.
For example: approving quotations, prioritising opportunities, following up prospects, qualifying inbound enquiries, arbitrating urgent requests, monitoring projects or interpreting performance indicators.
Then formalise one simple rule, criterion, cadence or responsibility.
Observe whether the issue is escalated less often, decisions move forward more effectively and the team gains clarity.
A good system rarely begins with a sweeping transformation. It begins by making one dependency visible and then reducing it.
The real issue: making the company less fragile
An over-centralised company can operate for a long time.
It may even give the impression of being firmly under control.
But its resilience depends too heavily on one person.
The real objective is not to remove the leader. It is to turn their intelligence, experience and decision criteria into a transferable system.
That transition changes the nature of growth.
The company no longer relies solely on one person’s permanent presence. It begins to rely on rules, relays, cadences and responsibilities.
It becomes clearer for teams, easier for the leader to understand, more robust for clients and better able to grow without exhausting itself.
That is often the moment when a company stops merely holding together.
It begins to build structure.