The Leader Who Plans to Disappear: Building Work That Survives You

The old CEO came back on a Wednesday.

Not for a board meeting, not for a grand farewell. That part was over. Two weeks earlier there had been the speeches, the champagne, the obligatory slide deck cycling through his greatest hits. This time he was just “dropping by,” or so he told himself, to return his badge and clear out the last personal debris.

He stepped out of the elevator into the same open-plan floor he had ruled for fifteen years. The same bank of monitors, the same standing desks, the same unreliable ficus tree in the corner.

And something inside him recoiled.

Because nothing was wrong.

The support queue was getting worked through. Engineers were arguing over a whiteboard. The new CEO was in his old office on a Zoom call, laughing. No one had paused the world to mourn his absence. The place hummed with a faintly impersonal competence. Things were… fine.

He felt a sharp, almost embarrassing emotion.

Is this what success feels like, or erasure?

It’s a strange test, the one that really measures whether you’ve led well: you don’t get to take it until you’re gone. And if you pass, it can sting.

We think of leadership as the art of being indispensable. The linchpin. The rainmaker. The one they “couldn’t do this without.”

But seen from the point of view of the work itself—the hospital ward, the codebase, the nonprofit, the family business—that is, in fact, a failure case. A bug, not a feature.

The most quietly radical thing a leader can do is plan to disappear. Not in the dramatic sense of storming out or retiring to a cabin in the woods, but in the mundane sense: to deliberately build systems, people, and culture so that the work not only survives without them, but barely stumbles.

That sounds noble in theory. In practice, it’s emotionally weird and structurally hard. The world keeps rewarding the opposite.

So let’s stay for a while with this uncomfortable question:

If you vanished tomorrow, what would actually break?

The invisible deal of being indispensable

Imagine a small nonprofit that exists because one person cared enough not to look away.

She started it out of a spare bedroom ten years ago, after watching a relative bounce between shelters. She raised the first donations, did late-night grant writing, answered the crisis line herself. The organization survives today because she has poured herself into it.

You join as a staff member. At first you’re inspired. Over time, you notice some things.

The founder ignores board advice. Every staffing decision, every social media post, every spending decision over fifty dollars is run through her. Directors can’t get meetings. Any challenge to her ideas is received as a personal attack, on both her and “her baby.” The charity and the founder feel indistinguishable.(theguardian.com)

The staff has a whispered phrase for it: founder syndrome.

In the nonprofit world, that term has a specific meaning: a pattern where the founder has too much influence, decisions bottleneck through them, and the organization’s identity fuses with their personality. Boards find it almost impossible to hold them accountable. Succession planning—if anyone dares utter the phrase—is treated as betrayal.(501c3.org)

This is not a nonprofit-only problem. Corporate founders get mythologized to similar effect. So do star professors, ten-x developers, legendary chefs, visionary pastors.

From the outside, it looks like devotion. From the inside, everyone is walking on eggshells.

The paradox is that this pattern often starts in virtue. Of course the founder feels indispensable: for years, they were. They were the one answering emails at midnight, patching the server, soothing the biggest donor. The organization really did ride on their back.

Competence is gravitational. The more often you’re the one who can solve the hard problem, the more people bring you hard problems. The more hard problems you solve, the more you start to believe the story: this place works because I hold it together.

Over time, that story calcifies into something darker:

Decisions feel risky in other hands, so you pull them back to yourself.
Delegation becomes a polite fiction; every file lands on your desk anyway.
When board members or senior staff suggest you groom a successor, you hear: we’re tired of you.

It feels like loyalty to the mission; from a systems point of view, you’re quietly signing its death warrant.

Because here is the statistical backdrop no amount of heroic identity can outrun.

In recent surveys, up to three-quarters of nonprofit leaders say they plan to leave their roles within the next five to ten years.(wipfli.com) At the same time, only about 29 percent of nonprofits report having a written succession plan.(councilofnonprofits.org) That’s not a gap; it’s a cliff edge.

Business is no calmer. Research on CEO changes in public companies finds that leadership transitions reliably move firm performance. In one study of Malaysian firms, post-succession performance actually improved on average—especially after forced turnovers where an underperforming CEO was replaced by an outsider—but cases with poorly managed internal handoffs saw performance decline in the years after.(ejournal.usm.my) Other literature reviews find a consistent positive association between serious succession planning—pipeline-building, talent development, not just picking a name—and firm performance over time.(gprjournals.org)

Put bluntly: leaders like to imagine they are the savior; the data suggest they are often the risk.

There’s a name technologists use for this risk that I wish we used more widely: the bus factor.

The bus you never see coming

People in software have a dark little metric they joke about in planning meetings: “What’s our bus factor?”

The idea is simple. The bus factor is the minimum number of people on a project who, if they suddenly disappeared—got hit by a bus, won the lottery, left for a new job—would cause the whole thing to stall because critical knowledge or capabilities went with them.(en.wikipedia.org)

A bus factor of one is a single point of failure. If that database whisperer goes down, no one else can deploy. If that engineer quits, nobody really understands the billing system. If that founder leaves, donors and staff evaporate.

When researchers looked at 133 popular open-source projects on GitHub a few years ago and calculated their truck factor (same idea, gentler vehicle), they found that about 65 percent had a bus factor of two or less, and fewer than ten percent had a bus factor above ten.(arxiv.org) In other words, most important software we rely on every day is effectively “two-people-away” from being in serious trouble.

We got a vivid illustration of this fragility in 2016, during the now famous left-pad incident.

A single developer, frustrated after a naming dispute involving one of his packages, unpublished all 273 of his JavaScript modules from npm, the dominant package registry for Node.js. Among them was a tiny utility called left-pad—a mere 11 or 17 lines of code, depending how you count—that simply padded strings with spaces or zeros. That trivial fragment happened to be a transitive dependency of thousands of widely used tools like Babel and Webpack. When it vanished, builds started failing all over the world; major tech companies suddenly couldn’t ship. Within hours, people were joking that “one developer just broke the internet.” npm ended up restoring the withdrawn package from backup and changed its policies so that widely used modules couldn’t be unpublished so casually again.(theregister.com)

This is what a bus factor of one looks like, made painfully concrete. One irritated programmer, a few deleted lines of code, and suddenly you can’t deploy half your apps.

We recognize that as bad engineering. We grumble about brittle dependencies, undocumented APIs, keys nobody else has. Product managers put “reduce key-person risk” into their Q3 OKRs.

But we rarely turn the same lens on ourselves.

If the bus factor on your leadership is one—if your exit would leave your team in chaos, your customers in the dark, your mission wobbling—then you are, from the system’s perspective, as dangerous as that unmaintained core library.

The deeper, quieter problem is that the very things we call “leadership potential” often reward lowering the bus factor:

Being the one who can swoop in and close the biggest deal.
Being the only person who knows the real donor behind the donor-advised fund.
Holding the relationships with key regulators personally.
Fixing crises through personal heroics instead of changing the process that creates them.

The more you do this, the more the system routes around everyone else and through you. Your influence grows—and the bus factor plummets.

Uncomfortable as it is, that’s the correct place to start: treating “they can’t do this without me” not as a compliment, but as a diagnosis.

Once you see it, you can ask more interesting questions.

Not: How can I make myself indispensable?

But: How do I build this so that it doesn’t need me?

The answer lives in a boring-sounding discipline that turns out to be strangely existential: succession.

Succession as the operating system, not the afterthought

Say “succession planning” in most organizations and people picture a dusty binder in HR, updated in a panic when the CEO announces their retirement.

That’s succession as paperwork. It’s mostly theater.

Real succession is less about “who takes over the big chair” and more about a habit of structuring work so that critical roles, relationships, and knowledge are never single-threaded.

Think of it less as naming your replacement and more as rewiring the operating system of the organization around a belief Peter Drucker captured in a line that old-school managers still quote to each other: “There is no success without a successor.”(htweston.co.uk)

That belief has some concrete implications.

First, it means you accept, at a deep level, that you will leave. Death, illness, burnout, opportunity—something will get you. Nonprofits are finally waking up to this; trade groups now describe leadership continuity planning as a basic form of risk management, not a luxury.(councilofnonprofits.org)

Second, it means you stop treating leadership as embodied in one heroic figure and begin treating it as a collective capacity. Consultants working with nonprofits talk about this explicitly: the point of succession planning isn’t just to pick the next executive director, it’s to ensure that “knowledge, relationships, and institutional memory don’t walk out the door” when any one person does.(zimconsulting.com)

Third, it means weaving continuity into the everyday fabric of how you run things, not bolting it on in an emergency.

High-reliability organizations—airlines, nuclear plants, certain parts of healthcare—understand this intuitively. They have to. The cost of a single failure can be a plane full of people or a poison spill in a river.

Aviation culture, for instance, has long been obsessed with “What if the captain goes down?” They build redundancy everywhere: two pilots, cross-trained; checklists shared and read aloud; standard call-outs; cockpit voice recorders. The point is not that the captain is unimportant. It’s that the system should not depend on any one person’s memory or mood to land the plane safely.

Medicine has been slowly importing this mindset. Atul Gawande’s work on surgical safety checklists came from watching how complex and error-prone modern surgery had become. What killed patients wasn’t ignorance but complexity: too many steps, too much tacit coordination. Borrowing from aviation, his team created simple, explicit checklists—confirming the site, counting sponges, verifying antibiotic timing—that dramatically reduced complications in trials across multiple countries.(atulgawande.com)

Checklists, standardized briefings, and explicit handoff rituals are not exciting. They are the opposite of the lone genius surgeon improvising heroically. But that’s the point.

At scale, continuity beats virtuosity.

In a Toyota plant, new employees headed to a fresh factory don’t just get a stack of manuals. The company sends them to established plants to work alongside experienced operators for months, absorbing tacit production know-how that’s too complex to write down. Only then are they sent back to seed the new plant, training the next cohort.(en.wikipedia.org) It’s an institutional commitment to keeping crucial knowledge “out in the open” rather than locked in the heads of a few lifers.

Succession, in this broader sense, is not an event. It’s the constant, deliberate work of increasing the bus factor everywhere—on the shop floor, in the boardroom, in the back office.

And it turns out to pay.

A 2023 review of empirical studies on succession planning and firm performance found a consistent positive relationship: companies that take succession seriously, and marry it to talent development and strategic alignment, tend to do better on measures like return on assets and innovation.(gprjournals.org) Old-school research on CEO transitions tells a similar story: how a transition is prepared for, and how deep the leadership bench is, predicts whether the change unlocks new performance or drags the place into a slump.(ejournal.usm.my)

So if the organizational case is strong, what gets in the way?

In my experience, three things: the emotional cost, the design challenge, and the time horizon. Let’s walk through each—not in a theoretical way, but through the grain of how this feels to a person trying to lead well.

The emotional cost: letting go of being “the one”

We should be honest: planning your own obsolescence can feel like a small death.

There’s a quote from Sheryl Sandberg that floats around leadership trainings: “Leadership is about making others better as a result of your presence and making sure that impact lasts in your absence.”(reddit.com)

It’s the second half that is hardest.

Making others better while you’re there is gratifying. You see the coaching land, the offsite unlock a new way of working, the promotion you fought for change someone’s life.

Making sure the impact lasts after you leave means confronting, viscerally, that you are not the point. The work is.

That’s easy to say when you’re posting platitudes on LinkedIn; it’s harder at 11 p.m. when you’re considering whether to copy your deputy on that sensitive email, or whether to walk a promising staffer through a difficult negotiation you could easily handle yourself.

Ego is only part of it. Identity is the deeper root.

If you’ve spent a decade building your sense of self around being the person who can fix it, paradoxically, building systems that don’t need you can feel like tearing out your own wiring.

This is one reason founder syndrome is so persistent. In one nonprofit advisory article, a consultant described how often they see organizations where “the universe revolves around the founder.” The person isn’t necessarily a villain; more often they simply have too much influence and no one has successfully challenged the pattern.(501c3.org) The trauma of imagining the place going on, maybe thriving, without them is so acute they unconsciously sabotage any move in that direction.

If you’ve ever known a retiring professor who “just can’t quite let go” of their lab, or a founder who stays on as “chairman” and quietly undermines the new CEO, you’ve seen this dynamic at work.

On the flip side, there is a particular, quieter joy available to those who manage the passage.

I think of the manager who jokes that her best performance review is “my team didn’t notice I was gone for three weeks.” Or the nonprofit ED who says, with real pride, that the fundraising gala she used to virtually live inside now runs beautifully with her checking in only once or twice.

But getting to that place demands that you reframe what winning looks like.

Instead of asking, “How important am I here?” you start asking, “How replaceable is what I do?”

That sounds like a path to irrelevance; in practice it’s the opposite. The leaders whose impact endures are those who have repeatedly asked: “If I weren’t in this seat, what would break? And what would have to exist so it didn’t?”

You will grieve some things as you do this. You will watch meetings proceed where nobody asks your opinion first. You will watch someone younger get applause for modernizing a process you quietly nudged them toward. You will feel, sometimes, like you are giving away your sharpest tools.

But if you stick with it, something liberating happens.

You start to be known, not as the hero who saved the day, but as the architect who made it so nobody had to.

The design challenge: embedding yourself into systems, not systems into yourself

Once you’ve made peace with the emotional weirdness of being replaceable, the question becomes practical: how, exactly, do you build a higher bus factor into the work you lead?

There are no one-size-fits-all answers—succession in a two-person startup looks very different from succession in a civil service or a family farm—but there are recurring design moves.

One is painfully basic: moving from “oral tradition” to shared artifacts.

Lots of organizations look, from the outside, like they have processes; from the inside what they have is people who are the process.

A senior revenue ops manager once described taking over a team where there was essentially one person who knew how to do the quarterly forecast. The spreadsheet model lived on her laptop. The logic lived in her head. When asked why nothing was documented, she said, a bit sheepishly, “Every quarter something changes; by the time I finished writing it down it would be out of date.”

Meanwhile, she was exhausted.

They started small: a shared repo, a README that exported just the core workflow, a policy that every modeling change required at least one peer review. She took a Friday to walk her second-in-command through every tab, even the ones she was embarrassed by.

Six months later, the team could survive her cancelling a flight. A year later, she could imagine taking a sabbatical.

That’s what “improving the bus factor” looks like when you zoom in. It’s not grand strategy; it’s a hundred choices to externalize knowledge and authority.

Organizational learning scholars have a formal vocabulary for this. They distinguish between tacit knowledge—skills and intuitions learned through experience, like knowing how an engine “feels” when it’s about to stall—and explicit knowledge, which can be written down or codified. Tacit knowledge is hard to transfer. Toyota’s solution, as we saw, is apprenticeship: they literally embed new operators in existing factories so they can absorb the fine-grained know-how before seeding new plants.(en.wikipedia.org)

Leaders face the same problem. Much of what you know is tacit: how to read a donor’s silences, when to push your CFO, which informal network actually runs the place. But you can do a surprising amount to surface and share it, if you treat that as core work rather than as a nice-to-have.

That often looks like:

Narrating your thinking aloud in meetings (“Here’s why I’m leaning toward this option…”).
Turning your personal spreadsheets and one-off emails into templates and wikis.
Inviting others to sit in on your most delicate conversations—not as scribes, but as future owners.

Another design move is distributing relationships.

Anyone who’s waited tables knows the stress of “sections”—when a whole room full of diners’ needs route through you. Leadership often ends up like a restaurant with one overburdened waiter and a dozen bored bussers.

In nonprofits, this usually shows up in donor and partner relationships concentrated in the ED. Articles on nonprofit succession hammer on this point: if your major funders only know one face, your mission is hostage to that person’s continued existence.(councilofnonprofits.org) In business, sales teams quietly panic when all the enterprise accounts “belong” to one rainmaker.

Rewiring that is slow, gritty work:

Introduce your deputy as the primary contact on some key accounts and actually defer to them.
Share your personal cell number less, shared inboxes more.
Bring a second chair into critical negotiations and let them speak first.

There is a moment every leader knows where a stakeholder looks past your colleague and directs the critical question back at you. In that instant, you get to decide whether to validate the old pattern or the new one.

A third move is structural: building mentorship and rotation into the way roles flow.

Think of the way some consulting firms move people across practices, or how airlines require copilots to get thousands of hours in the right-hand seat before they get the left. That’s not just skill-building; it’s risk management. If the captain collapses, the person next to them knows the plane.

Nonprofits are starting to catch up. Guidance from boards and sector bodies now explicitly encourages cross-training key staff, identifying “critical roles” beyond the CEO, and updating continuity plans annually.(councilofnonprofits.org) An article for nonprofit leaders put it bluntly: leadership continuity planning “should not be reserved just for the CEO” but extended to board officers and senior managers, with a focus on building a “deeper bench” of people who can carry critical responsibilities in an emergency.(cbiz.com)

There is also something bracingly humble in the best of these practices: they assume you will drop the baton occasionally, and they ask, then what?

Aviation and medicine talk about “crew resource management”—the discipline of flattening hierarchies enough that anyone can speak up about anomalies, and designing checklists and briefings so that tasks are cross-checked. Multiple studies report that when health systems import aviation-style practices—shared mental models, pre-briefs, explicit debriefs, standardized handovers—rates of serious errors drop.(pubmed.ncbi.nlm.nih.gov)

Behind all this is a simple reframe: it’s not your job to be the irreplaceable one. It’s your job to design roles and relationships so no one has to be.

Of course, none of this matters if you don’t tackle the third obstacle: time.

The time horizon: playing for the scene you won’t be in

In The Checklist Manifesto, Gawande describes a tension that surfaces every time you try to bring “boring” safety practices into professions built on heroics. Checklists and handoffs and boring redundancy offend a certain image we have of professionals as autonomous, audacious experts.(atulgawande.com)

Leadership has a similar hero story. We like tales of the CEO who turned the company around, the founder who “willed into existence” a new product, the turnaround artist. They are cinematic. They compress easily into a keynote.

Succession is not cinematic. It’s more like gardening.

You invest hours mentoring someone who may or may not become your successor, who may well leave, taking all that development to a competitor. You spend a year building documentation nobody thanks you for, only to discover that half of it is obsolete three months later. You introduce your deputy to all your best contacts and then watch as they build a relationship that no longer needs you.

From the vantage point of this quarter’s results, succession work is a cost center with no obvious upside. From the vantage point of ten years out, it’s most of what will matter.

Joe Bower, a Harvard professor who spent decades studying CEO succession, came to the view that how a company manages succession is inseparable from how it manages itself overall. Companies that treat it as a last-minute scramble tend to have other strategic and cultural problems; companies that invest in developing leaders at multiple levels are the ones that sustain performance over long stretches.(bigthink.com)

His favorite type of successor was the “inside outsider”: someone who’d grown up in the firm and understood its people and products intimately, but who had also maintained enough distance to see what needed to change. They had not drunk the Kool-Aid completely. They could use the system’s strengths while still challenging its dogmas.

Producing inside outsiders takes time. It means letting promising people take on messy, career-risky assignments years before you “need” them to. It means supporting their forays into other industries or roles that might broaden them in ways that don’t directly benefit you. It means resisting the short-term temptation to hire a smooth outsider to “fix things” when your real problem is that you never grew anyone internally.

Time horizon is also why so many family businesses and farms struggle with succession. Studies of family firms repeatedly find that lack of early, explicit planning leads to conflict, fire sales, or simply closure when the founding generation retires or dies.(en.wikipedia.org)

Everyone means to get around to writing down the plan. It just never quite outranks fixing the tractor, closing the big contract, helping with the grandkids.

So the question for you, if you lead anything—a team, a product, a parish, a lab—isn’t “Do I care about succession?” but “Over what time horizon am I genuinely willing to act like I care?”

If you measure your success over this quarter or this election cycle or your own tenure, it will always make sense to stay at the center. If you measure it over the life of the mission, you start making very different choices.

You start optimizing, not for how good it feels to be needed now, but for how unnecessary you can make yourself by the time you walk out of the building for the last time.

The quiet satisfaction of disappearing

Let’s go back to that retired CEO standing on the edge of his old open-plan office, watching his former company hum without him.

If you linger in that scene, there’s another way to read it.

Yes, the place is fine. The releases are going out, the new CEO is getting their hands dirty, the marketing roadmap is advancing. The world has not paused in deference.

But look more closely and you can see ghosts of his decisions everywhere.

The staffer running the weekly standup used to be an intern he argued for against HR’s doubts. The cross-functional incident review they’re following grew out of a painful outage five years ago where he insisted they talk less about “who” and more about “what in the system allowed this.”

The fact that nobody is looking anxiously over their shoulder for a last-minute sign-off is not an accident. It’s the residue of hundreds of choices to delegate cleanly, to let someone else own the room, to build simple defaults instead of complicated personal exceptions.

He is absent in the best sense.

Most of us will not be remembered with monuments. Our legacies will live, if at all, in those kinds of absences: in systems that work, in people who grew under our shade and now cast their own.

There’s a wonderful phrase in some leadership circles: “multipliers.” Liz Wiseman popularized it in her research on how some leaders seem to make everyone around them smarter. In contrast to “diminishers,” who hoard decisions and focus on displaying their own intelligence, multipliers operate with a basic conviction that the people around them are capable—and then design their leadership to stretch and amplify that capability.(en.wikipedia.org)

A multiplier leader, by definition, is one who expects to be outgrown.

They hire people better than them in specific domains. They ask questions instead of giving answers. They give away credit easily and take responsibility when things break. When they notice they’ve become the sole owner of something important, they treat it as an engineering problem.

They’re playing, consciously or not, for the scene where they are gone and things keep working.

We rarely celebrate that in the way we do heroics. There are no Netflix dramas about the COO who spent ten years quietly cross-training their entire team so that a cancer diagnosis didn’t sink the company. There are few op-eds praising the founder who stepped aside a year before they burned out, while they still had the energy and clarity to coach their successor.

But talk to people inside healthy organizations and you quickly discover how much their stability and sanity owe to those kinds of choices.

A hospital where shift handovers are disciplined, where juniors can speak up, where “I’ve always done it this way” is not a trump card—that’s not an accident. It’s the product of leaders willing to flatten hierarchy just enough, to standardize just enough, to repeat the unglamorous message that the process matters more than the person.

A software team where no one fears taking a vacation because “only I know X” is not a happy coincidence. It’s the result of months of pairing, writing runbooks, refusing to ship code that only one person understands, insisting that documentation is “done” when someone else can successfully follow it.

A community group that survives the death of its founder, or the departure of its charismatic minister, or the retirement of its matriarch, does so because somewhere along the line someone decided to stop being the whole story and start being a good ancestor.

That’s what building work that survives you really is: the practice of being a good ancestor in real time.

You will almost certainly not get a standing ovation for it. Planning your own irrelevance does not trend on social media. It might even cost you some short-term prestige.

But consider the alternative.

Consider the project that can’t ship because the one engineer who understands the deployment scripts ghosted. Consider the nonprofit that folds because the founder clung too long, poisoned the board, and left a crater where a pipeline should have been. Consider the family business that dies in a tangle of resentment because nobody could bring themselves to name, in time, who would actually take over.

There is a bus in all of our futures. The question is not whether it comes, but whether we let its shadow distort how we lead—or whether we use it as an oddly clarifying design constraint.