Josh Cutler
← Back to writing

Don't Juggle Other People's Balls

A few months ago I stumbled across Murray Bowen’s work on overfunctioning and underfunctioning in family systems. It’s a framework from therapy, not management, but the pattern it describes hit uncomfortably close to home. It gave me a name for something I’d been watching (and doing) for years across every team I’ve been part of.

The Hero’s Entrance

I’ve been both the hero and the manager of heroes at different points in my career. This piece is informed by making (and cleaning up) the same mistakes from both sides of the table. The title comes from advice I got early in my career that I wish I had internalized sooner.

Every organization I’ve worked in has had at least one. The person who always knows the answer. The one who picks up the slack when the deadline is slipping, who debugs the production issue at midnight, who somehow makes the demo work ten minutes before the meeting. For a long time, I was that person. And for a longer time after that, I managed people like that and mistook their heroics for high performance. It took watching the same pattern play out across multiple teams before I started seeing what was actually happening.

The hero is good at their job. That’s what makes this so insidious. They’re succeeding, visibly and consistently, in a way that slowly poisons the system around them. The work gets done. The launches happen. The metrics look fine. And underneath all of it, the organization is becoming more fragile with every crisis the hero quietly absorbs.

The conventional take on hero culture focuses on sustainability: heroes burn out, so we should be nicer to them. True, but incomplete. Yes, heroics are unsustainable. The bigger issue is that they mask people problems and process failures, and in doing so prevent organizations from ever becoming genuinely high-functioning.

Overfunctioning and Underfunctioning

Family systems therapy has a concept for this, and it maps with uncomfortable precision.

Beginning in the 1950s, psychiatrist Murray Bowen developed a theory of how families operate as emotional systems. One of his key observations was what he called the overfunctioning/underfunctioning reciprocity. In any relationship system (a family, a marriage, a team), when one person chronically takes on more than their share of the responsibility, they don’t just compensate for the people around them. They actively create the conditions for those people to do less. The two roles are locked together. One cannot exist without the other.

I mean this literally. It’s a description of how human systems actually work. The overfunctioner steps in, and the underfunctioner steps back. The overfunctioner does more, and the underfunctioner does less. Over time, both roles become entrenched. The overfunctioner genuinely believes the others can’t handle it. The underfunctioner genuinely believes they don’t need to. Both are wrong, but the system never gets tested because the overfunctioner never stops.

As Michael Kerr and Murray Bowen wrote in Family Evaluation: “The fixer’s Achilles heel is underestimating the resources of the people he intends to ‘help.’”

Both sides are getting something out of the arrangement. As Kerr writes in Bowen Theory’s Secrets: “The underfunctioning one is freed from the anxiety of responsibility and decision making, and the overfunctioning one is freed from the anxiety of not being in control and in charge.” Forget the narrative of a selfless hero carrying a lazy team. Both sides are running anxiety responses that reinforce each other until the system breaks.

Consider the senior engineer who always fixes the build. Each time they jump in, they’re making a trade: the build works today, but nobody else learns how it works. After six months of this, they haven’t just saved time. They’ve become the only person who can do it. The PM who rewrites every spec runs the same play. Junior PMs learn fast that their drafts are just raw material for someone else’s rewrite, and they stop investing accordingly. I have been this person, more than once, and I can report that the view from inside the pattern is remarkably self-justifying. It feels like competence. It looks like something else entirely from the outside.

The pattern is self-reinforcing and, left alone, it only moves in one direction. Mike Bowler describes a client whose Slack kudos channel told the whole story. Early on, people were recognized for “creativity” and “always having our backs.” Over time, the recognition shifted to praising those who “worked all through the weekend” and “put in the hours even when sick.” The culture had normalized heroics so thoroughly that unsustainable behavior became the explicit standard for recognition.

The Organizational Damage

So what happens to an organization that runs on heroics? The damage compounds in ways that are hard to see from inside the system.

Heroics Are an Analgesic

The most immediate harm is that heroics numb the organization to pain it needs to feel. When the deploy works because one person stayed up until 3am nursing it through, leadership sees a successful release. They don’t see a broken deployment pipeline. The feedback loop that would drive investment in better tooling, better processes, better training never fires. The hero absorbed the pain, so the organization never felt it.

Sam Schillace frames this well: “Hero culture means your systems are broken.” I’d extend that slightly. Hero culture means your systems are broken and nobody can tell. The hero is a painkiller, and painkillers are useful when the patient is also receiving treatment. When the painkiller replaces the treatment, the underlying condition just gets worse.

Schillace borrows a line from chef Thomas Keller that captures this nicely: “If you have to rush, you’re already lost.” The same applies to heroics. If someone needs to perform an extraordinary feat to make a routine process work, the process is already broken. The heroic effort just obscures that fact.

Heroics Create Learned Helplessness

The second harm is subtler and more corrosive.

In 1967, psychologist Martin Seligman conducted a series of experiments that became foundational to our understanding of motivation and depression. He found that when animals were repeatedly subjected to adverse conditions they couldn’t control, they eventually stopped trying to escape, even when escape became possible. He called this learned helplessness: the phenomenon where organisms learn, through experience, that their actions don’t affect outcomes, and so they stop acting.

The concept has since been extensively studied in humans and in workplaces. When problems reliably get solved by one person, everyone else’s problem-solving capacity atrophies. The easy label is laziness. The accurate one is rational adaptation to an environment where individual effort has been decoupled from outcomes. Why spend hours debugging the deployment when Sarah is going to jump in and fix it in twenty minutes anyway? Why write a thorough spec when the PM lead is going to rewrite it from scratch? The team has learned, correctly, that their efforts don’t matter.

The result is a team that looks low-performing but is actually functioning exactly as the system has trained it to function.

Heroics Concentrate Organizational Risk

The third harm is the one that should keep leadership up at night.

Most engineering leaders are familiar with the concept of bus factor: the number of people who would need to be hit by a bus (or, more charitably, win the lottery) before a project stalls. A bus factor of 1 means one person’s absence would cripple a critical system. Healthy teams aim for a bus factor of 5 or higher on critical systems.

Hero-dependent organizations often have a bus factor of 1 across multiple systems simultaneously, and they don’t know it. That’s the dangerous part. When the hero is present and performing, everything looks fine. Dashboards are green. Deadlines are met. Leadership has no signal that the organization is sitting on a single point of failure. The hero makes the risk invisible right up until the moment it materializes.

But What About Startups?

Early-stage companies often require heroic effort. There are five people, the building is on fire (metaphorically, sometimes literally), and everyone needs to do everything. The founder who codes all night to ship a demo, the first engineer who holds the entire infrastructure in their head because there’s nobody else to hold it. These aren’t pathologies at this stage but mostly just a form of survival.

The distinction that matters is between situational heroics and cultural heroics. Situational heroics are a response to a genuine crisis with a defined end. The production database is corrupted. A major customer is about to churn. The fundraise closes in 48 hours and the demo is broken. These are real emergencies that warrant extraordinary effort.

Cultural heroics are what happens when the crisis response becomes the default operating mode. When “all hands on deck” isn’t an exception but a standing expectation. When the sprint retrospective always ends with “we need to be more disciplined next time” and nothing changes. When people are routinely working nights and weekends not because of a specific emergency but because that’s just how the team operates.

Every growing organization needs to make the transition from heroics to systems. The same behaviors that make an early-stage startup succeed will make a scaling company fail. The founder who could hold the whole product in their head becomes a bottleneck. The instinct to just jump in and fix it, the very thing that made the early days work, becomes the thing preventing the team from ever learning to fix things themselves.

Think of it like technical debt. Sometimes taking on debt is the right call. Ship the feature, win the customer, survive to fight another day. But if the organization never pays down the debt, the interest compounds. Heroic effort works the same way. Every time someone solves a problem through sheer will instead of building a system that prevents the problem, the organization takes on a little more hero-debt. And like technical debt, hero-debt is invisible on the balance sheet until it isn’t.

None of this means “never be a hero.” It means that if heroics are the steady state, something is deeply wrong, and the heroics are preventing anyone from seeing it clearly enough to fix it.

The Personal Cost (Which Becomes the Org’s Problem)

Most writing about hero culture treats the personal cost as a separate concern from the organizational one. A “take care of yourself” appendix. That framing misses the point entirely. The personal cost is not separate from the organizational problem. It is the organizational problem, on a delay.

Heroes burn out. The timeline varies but the trajectory doesn’t. It starts with genuine motivation: this is important work, I’m good at it, people are counting on me. Then the load increases, because load always increases when someone demonstrates they can carry it. Recovery time shrinks. The intrinsic motivation that once fueled the effort gets slowly replaced by obligation, then resentment, then exhaustion. Mike Bowler puts it plainly: “Hero culture, allowed to continue to its natural conclusion, is a failure for everyone.”

What makes the cycle particularly pernicious is the normalization. The hero’s extraordinary effort gets absorbed into baseline expectations. Nobody celebrates the midnight deploys anymore because midnight deploys are just how things work now. The hero went from being appreciated for going above and beyond to being expected to go above and beyond. Dependency masquerades as recognition. I have watched this happen to good people, and the most frustrating part is that by the time they can see the pattern, they’re usually too deep in it to stop without feeling like they’re abandoning the team.

W. Edwards Deming, the statistician and management theorist whose work transformed manufacturing quality in the 20th century, argued that roughly 85% of the variability in work performance is attributable to the system, not the individual. The hero has been applying a 15% solution to an 85% problem. The math was never going to work.

And then the hero leaves. Or gets sick. Or simply stops caring. And the organization experiences what feels like a sudden crisis but is actually the accumulated consequence of years of deferred system-building. Every process the hero was holding together by hand comes apart at once. Every knowledge silo they embodied becomes a black box. The organization discovers, too late, that it was not high-functioning. It was one person functioning highly while everyone else adapted to not needing to.

The Uncomfortable Truth

I want to end with the uncomfortable version of this argument, because the comfortable version has been made plenty of times before.

The comfortable version says: heroes burn out, so let’s build better systems. True. But incomplete. The uncomfortable version is this: many heroes do not want to stop.

The identity of being the person who saves the day is powerful. It provides certainty in an uncertain environment. It provides control. Bowen’s framework explains why: overfunctioning is an anxiety management strategy. Stepping back means trusting others, which means tolerating the discomfort of watching them struggle and sometimes fail. It means sitting with the anxiety of not being in charge. For a lot of high-performers (and I include myself in this), that is genuinely harder than just doing the work.

Hero culture persists because it’s a collision between organizational incentives (reward the person who ships) and individual psychology (I need to be the person who ships). Both sides of that equation need to change.

When someone on a team is consistently heroic, that is not a performance review highlight. It’s a diagnostic signal. Something in the system is broken, and one person has been masking the symptoms. The appropriate response isn’t gratitude alone. It’s investigation. Celebrate the firefighter, absolutely. But then ask why there are so many fires.

An organization that needs heroes to function is not a high-performing organization. It is a fragile one that has been getting lucky.

Things I read while thinking about this