Why Your Executives Seem to Be Doing Nothing
If you spend enough time a few layers down in a company, you eventually notice something at the very top that looks a lot like idleness. The executives take a handful of meetings, make the occasional call, and spend a surprising amount of energy on some version of “let’s wait and see.” From below it can read as coasting, a group of well-paid people who have somehow arranged their jobs so that very little is asked of them on any given day.
I want to make the case that, for a certain kind of leader, that appearance isn’t absence. It’s the job, done well. The thing that looks like doing nothing is usually restraint, and restraint at the top is a discipline rather than a vacancy. It’s also genuinely hard, frequently correct, and nearly impossible to tell apart from laziness if you’re only watching the surface.
The Job Isn’t Activity#
It helps to be precise about what an executive is actually for. Strip away the theater and the calendar and the role comes down to three things: point the organization in a direction, make the small number of high-impact decisions that direction requires, and (the part nobody writes on a job description) avoid the wrong ones.
Notice what’s missing from that list: volume. Leverage inverts as you move up an org chart. Down at the leaves, output tends to scale with activity. An engineer who writes more good code, or a rep who makes more good calls, generally produces more. Near the root that relationship mostly breaks. An executive’s output is dominated by the quality of a few decisions and by the damage they manage not to cause. Measuring a leader by how busy they look is measuring close to the wrong variable; at the top, “busy” and “effective” are only loosely related.
A quick clarification, because it matters later: when I say “executive” I don’t only mean the person at the very top who sets the whole company’s direction. The same logic runs down the stack. Every layer of leadership owns some slice of direction, and every layer faces the same choice between setting that slice and then leaving the people beneath it alone, or reaching in to churn them. The slice shrinks as you descend; the discipline doesn’t.
This isn’t a new observation, it’s just an unpopular one. Peter Drucker built The Effective Executive around the idea that effectiveness is “doing the right things,” and that effective executives do first things first and “second things not at all.” Andy Grove’s High Output Management is largely about managerial leverage: finding the one to three activities where your involvement actually multiplies output, and being ruthless about ignoring the rest. Richard Rumelt’s Good Strategy / Bad Strategy makes the same point from the strategy side: a real strategy is mostly about focus, which is to say it’s mostly about what you’ve decided not to do. Even Jeff Bezos has said that if he makes three good decisions a day, that’s plenty.
If you’ve read my earlier piece on engineering levels, this is the same leverage curve, just extended past the point where you’re still shipping the work with your own hands.
Most of What They Could Do Would Make It Worse#
Here’s the mechanism behind “point it, then don’t touch it.” An organization running along its happy path is, most of the time, a reasonably stable system. It has normal day-to-day variation (a bad week, a missed deadline, a metric that wobbles, etc), and the majority of that variation is noise, not signal.
The executive who reacts to every wobble is doing something with a name. In quality control it’s called tampering: adjusting a stable process in response to its normal variation. Deming demonstrated it with his funnel experiment, and the result is counterintuitive but robust. Tampering with a stable process doesn’t reduce error, it adds variance. You reliably make the output worse by trying to correct noise you should have left alone.
Now layer the asymmetry of executive leverage on top of that. When someone near the top “improves” something, the effect doesn’t stay local; it propagates across the whole org. A correct intervention aimed at that happy-path noise buys only a marginal gain, while a wrong one cascades, and most reactive interventions are wrong precisely because they’re responding to noise. Add up the expected value of most day-to-day meddling and it tends to come out negative. (The high-impact exceptions, where decisive action genuinely pays, are the signal case I’ll come back to.) The discipline is knowing all of that, and sitting on your hands anyway.
Reacting to noise isn’t leadership, it’s variance. The hard part of the job is telling noise from signal and, most days, doing nothing about the noise.
There’s a name for the healthy version of this posture, too. Taleb borrows a medical term, iatrogenics (harm caused by the healer), and I’ll admit I’m extending it somewhere he mostly didn’t, from doctors to org charts. The related idea he calls via negativa travels more cleanly: often the reliable way to improve a system is subtraction, and a large share of wisdom is knowing what not to do. The oldest version of the same instinct is Chesterton’s fence: before you tear down a fence whose purpose you don’t understand, first work out why it was put there. Executives stand in front of a lot of fences they didn’t build.
I should be honest that “stable system” is an approximation. Some wobbles genuinely are signal: the market really did shift, or a competitor really did ship something that changes the game. Telling signal from noise is most of the skill, and I’ll come back to it. But the reflex that treats every wobble as a summons to act is exactly the bias that gets organizations tampered into mediocrity.
Why They Touch It Anyway#
If inaction is so often the right call, why do so many executives meddle constantly?
Because doing nothing feels like negligence, to the executive and to everyone watching them. This is action bias: the well-documented human preference for doing something over nothing even when nothing is the better move, paired with the tendency to feel worse about a bad outcome that followed inaction than about the identical outcome that followed action.
My favorite illustration comes from soccer. Bar-Eli and colleagues studied penalty kicks and found that goalkeepers almost always dive left or right, even though the distribution of shots says that staying put in the center would stop more of them. So why dive? Because a goal conceded while you stand still feels far worse, and looks far worse, than the same goal conceded mid-dive. The dive is a hedge against appearing to have done nothing. (Here’s a decent general summary of the bias if you want one.)
Executives dive for the same reason. Visible activity is insurance against looking negligent, not a contribution to the outcome. The good ones feel the same pull toward the dramatic dive and, more often than not, hold the center anyway.
Restraint vs. an Empty Chair#
Now the uncomfortable part, and the reason none of this is a charter for laziness. From below, masterly restraint and total disengagement look identical. In both cases the executive “isn’t doing much.” You can’t tell them apart by visible activity, because visible activity is precisely the thing both of them lack.
The difference is everything underneath. The restrained executive has pointed the org at something clear, is actively watching the few signals that carry real information, and moves decisively the moment a genuinely high-impact situation appears. The disengaged one has set no direction, is watching nothing, and is equally still only because stillness is the path of least resistance. Same surface, opposite substance.
So inaction is only valuable when it sits on top of a clear direction and live attention. It has to be chosen, not defaulted into. That’s why the deliberate version earned its own old name in medicine: “masterly inactivity,” the discipline of therapeutic waiting in place of reflexive intervention.
Investors have their own vocabulary for it, and underneath it sits the same asymmetry: when the cost of a wrong move dominates the gain from a right one, waiting beats acting. Warren Buffett has been unusually blunt that Berkshire’s stillness is a decision and not a lapse. “Lethargy bordering on sloth remains the cornerstone of our investment style,” he wrote in 1990, and six years later that “inactivity strikes us as intelligent behavior.” His partner Charlie Munger is widely quoted, though without a clean primary source, so treat it as attributed, for the idea that the big money isn’t in the buying or the selling, it’s in the waiting.
But hold onto the tension, because it’s real. “I’m staying out of it” is a genuine discipline and the most convenient disguise available for having checked out. What separates the two isn’t anything you can see on the surface; it’s whether there’s a real direction underneath.
“But What About Founder Mode?”#
If you’ve been anywhere near startup discourse lately, an objection has probably been forming. Doesn’t Paul Graham’s Founder Mode argue the opposite, that the best founders stay intensely hands-on, run skip-level meetings, refuse to blindly delegate, and that being told to “act like a manager” is exactly what wrecks founder-led companies?
It does, and I don’t think it actually contradicts any of this, for two reasons.
The first: even in founder mode, the leverage is still a small number of high-impact, context-rich calls. Founder mode is about a founder staying connected to their own company and their own vision, and refusing to let a layer of professional managers blur it into mush. It isn’t a license to thrash execution at random. Staying deeply informed so that your rare interventions are good ones is close to the opposite of tampering; reaching in to adjust every wobble because you can is the thing this whole post is warning about. (The Oxide writeup makes the point well: founder mode isn’t a blank check to meddle.)
The second reason is sharper, and it’s the one I actually care about. Founder mode is for founders. The executives people tend to complain about, the VPs and directors a few rungs down, are mostly not the ones setting the vision. Their job is largely to implement someone else’s direction. A director running “founder mode” on a strategy they didn’t set isn’t being visionary; they’re meddling with a flattering name for it. If anything, that layer should be more restrained, not less: set your own slice of the direction, then resist the constant urge to churn the teams beneath you. This is the old direction-versus-implementation split, who owns the “what” versus the “how,” and it matters a great deal here.
Check the Direction, Then Check the Chair#
So the next time you look up and leadership seems to be doing nothing, resist the obvious read. The executive who appears idle may be doing the genuinely hard part of the job: having pointed the org somewhere, and then having the discipline not to keep touching it. Activity at the top is cheap and, in expectation, frequently negative. Judgment and restraint are the actual work.
Just don’t let anyone hide inside that. The question to ask is never whether leadership looks busy. It’s whether the direction is clear, and whether they move when it actually counts. That, not the meeting count, is the line between masterly inactivity and an empty chair.
Hopefully this works out for you, or was at least helpful. If you have any questions, don’t hesitate to shoot me an email, or follow me on twitter @nrmitchi.