Early on, centralisation works. The founder knows the customer, offer, team and constraints. They can decide quickly, correct immediately and protect coherence.
The problem is not that the founder controlled too much from the first day. The problem is that the company grew without evolving its decision system.
Success initially reinforces centralisation
Early success teaches the organisation that going through the founder is effective. They know the history, see the consequences and can resolve exceptions.
Habits gradually form:
- ask for their view before committing to a decision;
- copy them in for reassurance;
- wait for approval on sensitive work;
- give them important customers or partners;
- escalate conflicts between functions.
Each behaviour looks prudent. Together, they turn the founder into a mandatory point of passage.
The bottleneck does not always look like overload
Some founders work extremely long hours and can see the issue clearly. Others appear available, while the organisation slows down around them.
Signals may be indirect:
- meetings are postponed until they can attend;
- managers present options without recommending one;
- work remains “awaiting approval”;
- teams avoid decisions that could be criticised;
- important information moves through private conversations.
The bottleneck is not only the founder’s time. It is the uncertainty created by the absence of another decision mechanism.
Why “delegate more” is not enough
Telling people to be more autonomous does not create autonomy. Genuine delegation requires:
- an expected outcome;
- a decision scope;
- limits or escalation thresholds;
- the necessary information;
- a review cadence;
- permission to make an imperfect decision within the agreed framework.
Without those elements, the founder delegates a task but retains the trade-off. The team moves until the next ambiguity, then returns to them.
The role of middle managers
When a manager consistently passes problems to the founder, there are two possible interpretations. They may lack maturity. They may also have learnt that decisions will be retaken above them in any case.
An overly centralised system weakens managers:
- they develop less judgement;
- their teams see them as messengers rather than decision-makers;
- they take fewer considered risks;
- the best may eventually leave because they have no genuine space.
Reducing founder dependency therefore requires rebuilding managerial authority, not simply moving tasks.
Interfaces create most escalation
Simple issues within one function are relatively easy to delegate. Difficulty appears at the boundaries: sales versus capacity, marketing versus product priority, quality versus speed, important customer versus general rule.
Those compromises cannot be resolved by one procedure. They require governance: who decides, with which data, according to which principles and at what cadence?
This is often where a COO becomes useful. The role owns the interfaces and prevents every cross-functional conflict from automatically escalating to the founder.
Three mechanisms for moving beyond centralisation
1. A clear decision map
Responsibilities should be described as decisions, not only duties. Who can change a price? Stop a project? Grant an exception? Recruit? Change a tool?
2. Stable operating cadences
When trade-offs have a predictable place and time, they stop invading every day. Genuine emergencies remain possible, but everything no longer presents itself as urgent.
3. Sufficiently reliable data
Founders often retake control because they do not trust the information. Improving measures, definitions and freshness is therefore a condition for delegation.
The founder must change role, not disappear
Moving out of the bottleneck does not mean abandoning vision, key customers or important decisions. It means reserving attention for the decisions that genuinely require it.
The transition is often uncomfortable. The founder must accept that some decisions will differ from those they would have made, while building a framework that protects the company.
Success is visible when teams make better decisions without waiting, managers own the compromises and the founder can focus on the future rather than continually repairing the present.