The Over-Manage / Under-⁠Manage Cycle

An observation on why most management is really firefighting, and why the harder craft is knowing where to let the fire burn

Damian Fozard

There is a pattern I have watched repeat itself in every business I have worked in, founded, advised, or sat on the board of. A function (treasury, quality, a regional sales operation, a customer success team) runs along quietly beneath the level of conscious attention. It is not ignored because it is unimportant. It is ignored because it is not currently causing pain. Then it slips. A miss becomes a trend, the trend becomes a board slide, and suddenly the function is the centre of gravity. A senior leader, often one with no prior involvement, descends on it with the conviction that they are the only person who can untangle the mess. Dashboards appear. Standing meetings are scheduled. Process is rewritten, usually with more rigour than the function ever needed. And then, predictably, the spotlight moves on. The next fire is more interesting. What is left behind is a patched problem, a new layer of process built to solve a problem that no longer exists, and a team that has learned attention from above is something to be endured rather than welcomed.
I do not think this is a personnel-management problem, and I do not think it is even really a process-management problem. I think it is what an organisation looks like when it has more managers than leaders: when the work of holding a long direction, of deciding what is worth fixing and what is worth letting fail, has been replaced by the work of responding to whatever is currently bleeding.
The question I want to take seriously here is whether the management literature supports this view or contradicts it. I have looked into it. The honest answer is that it supports the diagnosis almost without qualification, but the prescription it converges on is harder than I had initially framed it. The hardest thing a leader has to do is not to set the vision. It is to decide which fires to let burn

I.
The Mechanism

The first time I saw the cycle clearly I was a graduate, in my first proper job, at Clerical Medical International. One of my managers ran his team in what I can only describe as two modes. For weeks at a time he was low-key, hands-off, easy to be around. Then, intermittently, he would change. Reports that had never existed would suddenly be required. Work that had been trusted would now be double-checked. People who had been colleagues a week earlier were now treated with a low-grade suspicion, given to understand that they had been falling short and would need to work harder and later until things were back on track. After a few days, sometimes a couple of weeks, the mood would lift and the original manager would return. As a graduate I assumed something had changed in the work, and that I was simply too junior to see what. It took me a while to notice that nothing had changed in the work. What had changed was that he had become aware of a problem, and his response to becoming aware of a problem was to undergo a temporary personality change. Once the problem passed out of his attention, he reverted. The team’s actual performance, as far as I could tell, was unaffected in either direction.
That observation has stayed with me for thirty years, partly because it was so cleanly drawn (the cycle made visible in a single individual), and partly because I have since seen versions of it in almost every organisation I have worked in, including, with some reluctance, in myself.
The most rigorous treatment of the cycle comes from Nelson Repenning and John Sterman at MIT Sloan, who have spent two decades formalising what they call the capability trap. Their 2001 paper, with the perfect title Nobody Ever Gets Credit for Fixing Defects That Didn’t Happen, and its 2002 successor in Administrative Science Quarterly, lay out the dynamics in system-dynamics form.
The mechanism is simple enough to state. Organisations under pressure direct effort towards fixing the visible problem rather than improving the underlying capability that would have prevented it. Problem-fixing produces faster, more visible results than capability-building, and gets rewarded accordingly. Over time, the reward system selects for managers who excel at heroic rescue rather than at quiet prevention. As Repenning and Sterman put it, in a line that has stayed with me ever since I first read it:
“As organisations grow more dependent on firefighting and working harder to solve problems caused by low process capability, they reward and promote those who, through heroic efforts, manage to save troubled projects or keep the line running.”
The engineer they quote captures it more bluntly: nobody ever gets credit for fixing problems that never happened.
The cycle, in their telling, is structural rather than personal. The over-zealous manager who descends on a struggling function is not a bad manager. They are the natural product of a system that pays for visible rescue and ignores invisible prevention. The patched fix, the bolted-on process, the new committee: each is, in Repenning’s more recent framing, a small piece of arson laid down in the course of fighting yesterday’s fire. He and Donald Kieffer have started calling these people firefighting arsonists, and the phrase is exact. The intervention itself produces the conditions for the next intervention. The system never recovers; it descends.
This is the mechanism. It is well documented, it is mathematically modelled, and it is depressingly familiar to anyone who has watched a business mature.

II.
The Diagnosis

The strategy literature gives this its proper name. Henry Mintzberg has spent most of a career making a distinction that bears directly on what I have been describing: between strategic planning and strategic thinking. Planning, in Mintzberg’s account, is analysis. It is the breaking down of strategies into actionable steps, the building of control mechanisms, the formalisation of process. Thinking, by contrast, is synthesis. It is the holding of a coherent perspective on where the organisation is heading and why. The two are different cognitive acts, and Mintzberg’s claim, argued at length in The Rise and Fall of Strategic Planning in 1994, is that most organisations have mistaken one for the other.
The over-managing executive who descends on a broken function is doing analysis. They are diagnosing, programming, formalising. They are not doing synthesis. They are not asking whether the function should exist in its current form given the direction of the business, whether the failure is a symptom of strategic misalignment elsewhere, or whether their intervention is going to create dependencies that constrain future flexibility. They are solving the visible problem in front of them with the tools they happen to have to hand. That is management. The leader’s job, in Mintzberg’s framing, is to hold the perspective that makes individual operational problems interpretable and individual operational failures survivable. Without the perspective, every problem looks equally urgent, and the only available response is the firefighting hero.
Mintzberg goes further than this, and the further point is the one that matters to me. He argues that the separation of “thinkers from doers” is itself a pathology. The over-managing executive is, in this sense, a thinker temporarily playing doer, and they do neither role well. The doers below them stop thinking, because someone else is now handling it. And the thinker stops thinking, because they are now in the weeds. The cycle hollows out both functions. By the time the executive moves on to the next fire, the team has lost both its capacity to manage its own problems and its sense that anyone above it is holding the larger picture.
The pull is real. There is a particular kind of executive satisfaction in being the person who solves the visible problem. The quieter satisfaction of having built an organisation that does not need you to solve it is harder to feel, because you mostly feel its absence.

III.
Why It Persists

The attention-based view of the firm, developed by William Ocasio and others, gives the structural reason for why the cycle persists even when people inside it can see it clearly. Managerial attention is scarce. There are always more potential issues in any organisation than there is bandwidth to address. Organisations therefore develop implicit rules about what gets attended to, and crisis is the most attention-grabbing signal available. Functions learn (quickly, and not always consciously) that the way to get resources is to fail visibly. The pre-crisis state, where strategic investment could prevent the crisis, is by definition not attention-grabbing. There is no narrative.
This is why “fixing problems that never happened” is structurally unrewarded. The counterfactual is invisible. A leader who quietly maintains a function so that it never breaks produces no story, no rescue, no promotion-worthy moment. The cycle is, in part, a story problem. Organisations reward people whose work generates narrative, and the firefighting hero generates the cleanest narrative in business: it was failing, and then I arrived, and now it is not. The leader who let it run untouched because it was, on balance, the right thing to let run untouched, has nothing to put on a slide.
There is one honest counter-argument worth engaging with. The evolutionary-biology literature has given organisational theory the model of punctuated equilibrium (long periods of stability interrupted by short bursts of intense change), and a reasonable reading of the cycle I have been describing is that it is not a pathology but a feature. Organisations cannot sustain high-intensity attention on every function simultaneously, so they cycle attention to wherever it is most needed. Crisis-driven intervention, in this view, is how organisations actually learn and adapt. The COVID-era literature on adaptive leadership leans this way. The managers who delivered for their organisations during the pandemic were the ones who could mobilise quickly in response to disruption.
I think this view contains a real truth, but it does not refute the diagnosis. The question is whether attention cycling is directed by strategy or driven by symptoms. A leader can deliberately decide that this quarter, treasury gets attention, and next quarter, sales operations. That is strategic prioritisation. What the cycle does is the opposite. Attention is dragged by whatever is currently bleeding, with no governing logic for why this problem now. The same observation appears in Karl Weick’s work on high-reliability organisations. The organisations that handle crisis well are not the ones that respond fastest. They are the ones that have invested in capability during the calm, so that response, when it comes, is bounded and proportionate and does not damage the underlying system. Adaptive leadership, properly practised, is the disciplined version of what the cycle does badly. The undisciplined version is what most organisations actually do.

IV.
Where to Fail

I had originally planned to end the essay at the previous section. The diagnosis was clear, the literature was supportive, and the conclusion, that the cycle is a strategy deficit dressed up as a management problem, seemed to do enough work. But the more I have sat with it, the more I think the diagnosis is the easy half of the argument. The hard half is the prescription, and the prescription is much harder to sit with than the diagnosis.

The framing I want to offer is this: the hardest thing a leader has to do is to know where to fail, and which fires to let burn. No organisation can be excellent on every dimension simultaneously. Over-optimisation in one place produces under-resourcing somewhere else. The strategic question is not how to fix everything that is broken. It is how to choose, deliberately and in advance, the places where some degree of underperformance is acceptable, because the cost of fixing it is higher than the cost of carrying it.

Eli Goldratt made this point as forcefully as anyone in The Goal in 1984, and it is the central insight of his Theory of Constraints. The most useful, and least intuitive, claim in Goldratt’s framework is that most local improvements are not improvements at all. They optimise a local optimum that does not contribute to the global optimum, and they often do so at the cost of the constraint that actually governs the system’s throughput. Goldratt’s mantra, which I have come back to in my own businesses more times than I can count, is that an hour saved at a non-bottleneck is worthless. The corollary, less often quoted, is that an hour spent fixing a non-bottleneck is worse than worthless. It consumes attention, energy, and political capital that should have been spent on the constraint. The locally underperforming function that an over-managing executive descends on is, very often, a non-bottleneck. It is underperforming because it can afford to underperform, given the larger system. The intervention is correct at the local level and wrong at the system level, which is the most expensive kind of wrong.
Getting comfortable with this is, I think, the actual work. It is easy in theory and hard in practice, because it requires the leader to live with visible imperfection in a part of the business that everyone in the room is looking at. The discipline is not the analytical one of identifying which fires can burn; that part is, with some patience, tractable. The discipline is the temperamental one of holding the position when the room is asking you to act. Every other person in that room is participating in the over-manage / under-manage cycle in their own work. They are not bringing you a problem to think about. They are bringing you a problem to solve. The cultural pressure to perform a rescue is enormous, and most leaders, including some very good ones, eventually give in to it.

I have, with some embarrassment, watched myself do the same thing. As a younger entrepreneur, I carried everything on my own shoulders, on the basis that no one else could be trusted with it, until eventually the load made the position untenable and I had to learn to delegate. Learning to delegate turned out to be much harder than I had expected, because it was not really a logistical problem. It was a temperamental one. Delegating meant giving people the dignity of making their own mistakes, and trusting them to fix them. It meant, in practice, watching things go less well than they would have if I had done them myself, in the short run, in the service of an organisation that could actually function in the long run. The over-managing executive who descends on a struggling function is, in part, a person who has never properly made that trade. They are still operating on the early-stage founder’s logic, that the work cannot be trusted to anyone else, long after the organisation has outgrown it.
The leaders who appear, in retrospect, to have been unusually strategic were, in many cases, simply leaders who let things fail in the right places. Their judgment was visible only in the absence of action. They did not get credit for what they did not do, because nothing they did not do generated a story. The leaders who appear, in retrospect, to have been chaotic were, in many cases, the ones who solved every problem in front of them. Their judgment was indistinguishable from their activity. They got credit for every fire they put out, including the ones they should not have noticed.
This is uncomfortable advice. It tells you that the most valuable work you do as a leader may be invisible to the people you lead, and that the actions you take that look like leadership are often the ones that erode the thing you are trying to build. It also tells you that the cycle I have been describing is, in the end, sustained not by the over-managing executive but by the organisational culture that rewards them. As long as the rescue narrative is the only one that generates promotion, the cycle reproduces itself, regardless of how clearly anyone in it can see what it is doing.

The prescription, then, is not really a management prescription at all. It is a temperamental one. You have to become the kind of leader who can sit, visibly, with a function that is not performing, and decide, deliberately and at some emotional cost, that this is not the problem worth your attention this quarter. You have to be willing to be seen doing nothing about a thing that everyone in the room thinks should be fixed. You have to get comfortable living with failure, because the alternative is the cycle.
This is, I suspect, what the older generation of business writers meant by judgment. It is not strategic planning, in Mintzberg’s sense. It is what the strategic perspective is for. The plan tells you what to do. The perspective tells you what to leave alone. The first is teachable. The second, in my experience, is not, though it can be earned, slowly, by anyone willing to be wrong in public long enough to develop the stomach for it.