There is a pattern I see in nearly every founder-led business I work with. It starts subtly — a decision that should have been made by someone else, a question that could have been answered without escalation, a project that stalled because one person hadn't signed off yet. Individually, each instance is unremarkable. Collectively, they are the signature of an organisation that has quietly organised itself around a single point of failure.

That point of failure is usually the founder. And they almost never see it.

This essay is about why that happens, what it costs, and — drawing on the best thinking from cognitive science, systems theory, and military doctrine — what it takes to lead in a way that doesn't make you the ceiling of your own organisation.

Part One: Why It Happens

The Technician's Trap

Michael Gerber, in his seminal work The E-Myth Revisited, identified a founding tension that lies at the heart of almost every small business. Most companies, he argued, are not started by entrepreneurs in the classical sense — they are started by technicians who had an entrepreneurial seizure. A skilled engineer, a talented consultant, a gifted craftsperson — at some point they decided they could do what their employer did, and do it better. So they left. And in doing so, they brought with them the one tool they knew best: the ability to do the work themselves.

The tragedy Gerber diagnosed is not that these founders lack ambition. It is that their competence becomes a trap. The technician is most comfortable doing. The work of managing systems, developing people, and deliberately stepping back from decisions runs against the grain of everything that made them successful in the first place. The business, in Gerber's framework, never grows beyond the founder — it merely accumulates more of the founder's work.

This is more than a habits problem. It is, at a deeper level, a cognitive one.

Pattern Recognition and the Expert Trap

Gary Klein's research on Naturalistic Decision Making offers a revealing account of how experienced leaders actually make decisions. In high-stakes environments — from military commanders to ICU nurses — experts don't deliberate between options in the way formal decision models assume. Instead, they pattern-match. They scan a situation, recognise it as an instance of something they've seen before, and act on the first workable option that presents itself. Klein calls this Recognition-Primed Decision Making, and it is extraordinarily fast and often extraordinarily accurate.

But this capability, when carried into an organisational context, has a corrosive side effect. The expert founder who has spent a decade pattern-matching in their industry genuinely does make better snap judgements than most of their team. The decision to be in the room, to be consulted, to stay close to the work — each of these feels rational, because experientially it often is. Delegation feels like a downgrade. Handing off a decision feels like accepting a slower, less accurate answer to a question you could answer yourself in thirty seconds.

David Rock's SCARF model adds the neurological dimension. Rock's research in organisational neuroscience identifies five domains that the human brain treats as status threats: Status, Certainty, Autonomy, Relatedness, and Fairness. For a founder, ceding control activates multiple threat responses simultaneously. Autonomy diminishes when someone else owns a decision. Certainty drops when outcomes become less predictable. Status feels implicated when a junior team member's judgement is treated as sufficient. The result is that letting go does not merely feel strategically uncomfortable — it feels, at a neurological level, like danger.

The founder who cannot delegate is not failing to apply a principle they understand. They are fighting a nervous system that has been rewarded, for years, for doing exactly the opposite.

This is important to understand. The bottleneck is not a knowledge problem. Founders who stall their organisations generally know they should delegate more. The problem is that their pattern recognition screams competence, and their neurology screams threat, and the intellectual argument for letting go has to overcome both. This is why behavioural nudges and delegation frameworks, used alone, so rarely produce lasting change.

Part Two: What It Costs

Decision Velocity as a Strategic Asset

Kathleen Eisenhardt's 1989 study in the Academy of Management Journal remains one of the most rigorous investigations of how fast decisions get made — and what separates organisations that decide quickly from those that don't. Working with high-velocity technology firms, Eisenhardt found that speed of strategic decision-making was not, as might be assumed, a function of having less information or fewer people involved. The faster firms, paradoxically, used more information, more advisors, and considered more alternatives. What distinguished them was not the quantity of input but the quality of the decision architecture — who could decide what, and under what conditions.

The slower firms shared a common characteristic: decisions accumulated. They waited for more data, or for the right people to be available, or for the situation to become clearer. In founder-led companies, this dynamic is almost always traceable to a single source. Information flows upward. Sign-off is required. And the founder — who is also running sales, also reviewing product, also handling the difficult client — is the junction point through which all of it must pass.

Marcia Blenko, Michael Mankins, and Paul Rogers, in their work on decision effectiveness, are more blunt about the cost. Their research with Bain found that decision-making effectiveness — the speed and quality with which an organisation makes and executes key decisions — was among the strongest predictors of total shareholder return. Companies in the top quartile for decision effectiveness generated returns nearly six percentage points higher than those in the bottom quartile. Decision bottlenecks are not a process problem. They are a value-destruction problem.

The System Constraint

Donella Meadows, in Thinking in Systems, offers a perspective that reframes this entire problem at a structural level. Systems, she argues, have leverage points — places where a small change can produce large shifts in behaviour. But systems also have constraints: components that limit the throughput of the whole, not because they are poorly designed in isolation, but because the system has inadvertently arranged itself around them.

The founder-as-bottleneck is precisely such a constraint. And Meadows makes a point that is particularly uncomfortable for the people trying to fix it: in a system with a binding constraint, improving everything except the constraint achieves almost nothing. You can hire better people, implement better tools, build cleaner processes — but if all information still routes to one decision-maker, and that decision-maker's bandwidth is finite, the system's output is capped at that person's capacity. The constraint determines the ceiling.

What makes this especially difficult to diagnose is that constrained systems often appear to be working. The founder is getting through the decisions. The projects are moving, if slowly. The organisation is surviving, and in some cases thriving by certain metrics. The costs are invisible until they become visible all at once — when a key employee leaves because they haven't had meaningful autonomy in three years, or when a competitor with a faster decision cycle takes a market position that the founder's company should have occupied months earlier.

In a system organised around one person, the single highest-leverage change is not to make that person smarter or faster. It is to redesign the system so it no longer requires them in every loop.

There is a further wrinkle that Meadows' framework illuminates. Systems are self-reinforcing. The longer a founder remains the primary decision authority, the more the organisation adapts to that reality. People stop developing the judgment muscles they aren't being asked to use. Problems that could be solved locally get escalated by default, because that is what the culture rewards. The team learns helplessness — not from weakness, but from rational adaptation to a system that has, consistently, not required them to do otherwise. By the time the founder becomes aware of the bottleneck, they are often looking at a team that has been de-skilled in exactly the areas where it needs to grow.

Part Three: How It Is Solved

The Lesson from the Santa Fe

In 1999, Captain L. David Marquet took command of the USS Santa Fe, a nuclear submarine ranked last in the US Pacific Fleet on nearly every measure of performance and morale. What he found was an organisation built entirely on a leader-follower model: orders flowed down, information flowed up, and nothing moved without authorisation. The crew were technically competent people who had been systematically trained not to think for themselves.

Marquet's response was not to try to become a better commander in the traditional sense. Instead — as he recounts in Turn the Ship Around — he set about dismantling the structure that made the crew dependent on him. His central insight was that the leader-follower model is seductive precisely because it provides the leader with the feeling of control. But that feeling is an illusion. Control exercised through constant approval is not resilience — it is fragility. It means that the system degrades every time the leader is unavailable, overwhelmed, or simply wrong.

In place of the leader-follower model, Marquet built what he called a leader-leader structure. Rather than asking his crew to execute orders, he shifted the locus of decision-making to the person with the most relevant information — which, on a submarine, is almost always the person closest to the action, not the officer at the top. The phrase he trained his crew to use was not 'request permission to...' but 'I intend to...'. This subtle linguistic shift carried a profound structural implication: the crew member was not seeking approval; they were communicating intent. The captain's role became one of providing clarity on objectives, building competence and trust, and removing friction — not ratifying individual decisions.

Within two years, the Santa Fe had gone from last to first. It produced more officers promoted to command than any other submarine in the fleet. The lesson Marquet drew was not about naval management. It was about where authority belongs in any system that has to perform under pressure.

Mission Command: Doctrine as a Design Principle

Marquet's framework did not emerge from nowhere. It is, in essence, a civilian articulation of a doctrine that military forces — particularly the German and later NATO armies — had been developing for over a century. Auftragstaktik, or mission command, begins from a premise that sounds simple and proves radical in practice: no commander can anticipate everything, so no commander should try to control everything.

Under mission command doctrine, the role of the senior leader is to communicate the intent — what needs to be achieved and why — with enough clarity that subordinates can make sound decisions without further guidance when circumstances change, as they always do. The emphasis falls not on compliance with a plan, but on alignment with a purpose. The soldier, the junior officer, the team lead knows the commander's intent well enough to act in its spirit even when the letter of the orders no longer applies.

General Stanley McChrystal's account in Team of Teams shows what this looks like when applied to a contemporary organisation facing a complex, adaptive threat. The Joint Special Operations Command he led in Iraq was, in his description, extraordinarily capable — but organised for an industrial-era model of command that could not match the tempo of a decentralised enemy. His solution was to build shared consciousness across the organisation: radical transparency about what was known, what was uncertain, and what the overall intent was — combined with empowered execution at every level. Leaders remained in the loop not to approve decisions, but to understand them. The shift, he noted, felt to many of his senior staff like a loss of control. In practice, it produced a step-change in the organisation's ability to act.

Practical Architecture: From Insight to Structure

The question for founders and the advisors who work with them is how to translate these principles into concrete organisational design. Three moves matter most.

The first is to create and communicate intent with genuine precision. Mission command fails when intent is vague. 'Grow the business' is not intent. Intent means: here is the outcome we are pursuing, here is the constraint we are operating within, here is what a good decision looks like at the boundary. When a leader has articulated intent clearly enough that a team member can make a difficult call without asking — and get it right — the intent is clear enough.

The second is to build decision rights explicitly, rather than letting them accumulate by default. Blenko, Mankins and Rogers' RAPID framework — which assigns distinct roles in any given decision to those who Recommend, Agree, Perform, are given Input, and Decide — is one useful tool. The point is not to use any particular model, but to have the conversation at all. Most organisations have never explicitly mapped who decides what. The result is that every novel situation defaults to whoever has the most authority, which is usually the founder. Making decision rights explicit gives the team a structural permission to act.

The third, and perhaps the hardest, is to change the leader's own behaviour in a way that signals the shift is permanent. Marquet stopped giving orders — not occasionally, but as a rule. When a member of his crew asked him what to do, he refused to answer. Instead he asked what they thought they should do, and why. Over time, the crew stopped asking. This is the most uncomfortable part of the transition for most founders: the period in which they must tolerate slower decisions, more visible mistakes, and the anxious feeling that things are not being done quite right. That discomfort is not a sign that the change is failing. It is the sound of the organisation learning.

Conclusion: The Highest Form of Control

There is a version of this conversation that frames it as being about letting go — about founders releasing their grip on the business they built. I don't think that framing serves them well. What the best evidence suggests is something more precise: that the founder who remains a bottleneck is not exercising control. They are performing it, at great cost, while the real work of building a resilient organisation goes undone.

The leader who has communicated intent clearly enough that decisions happen well in their absence, who has built a team capable of acting with initiative inside a coherent framework, who has designed a system that does not require their presence in every loop — that leader has more control over outcomes than the one who insists on signing everything off. They just have it in a form that is harder to see.

That is the move this essay has been pointing toward. Not from order to chaos, not from command to abdication, but from the illusion of control to its substance. From the technician who cannot let go, to the officer who has learned that the highest form of leadership is to build an organisation that can hold the mission without you.

Key Sources

  • Gerber, M. E. The E-Myth Revisited: Why Most Small Businesses Don't Work and What to Do About It. HarperCollins, 1995
  • Klein, G. Sources of Power: How People Make Decisions. MIT Press, 1998
  • Rock, D. Your Brain at Work: Strategies for Overcoming Distraction. HarperBusiness, 2009
  • Eisenhardt, K. M. Making Fast Strategic Decisions in High-Velocity Environments. Academy of Management Journal, 32(3), 1989
  • Blenko, M., Mankins, M. & Rogers, P. Decide & Deliver: Five Steps to Breakthrough Performance in Your Organization. Harvard Business Review Press, 2010
  • Meadows, D. H. Thinking in Systems: A Primer. Chelsea Green Publishing, 2008
  • Marquet, L. D. Turn the Ship Around!: A True Story of Turning Followers into Leaders. Portfolio/Penguin, 2013
  • McChrystal, S. Team of Teams: New Rules of Engagement for a Complex World. Portfolio/Penguin, 2015