Why Competence Looks Broken From the Inside

May 26, 2026

11 min read

5 views

88%

The strange thing about expertise

What if the biggest trap in leadership is not that incompetent people think they are geniuses, but that the people who are closest to real competence often feel least satisfied with the systems around them?

That is a more unsettling idea than the usual cartoon version of overconfidence. We like the neat moral tale: the clueless overestimate themselves, the skilled remain humble, and merit eventually reveals itself. But real life is messier. People who are truly good at something often see the gap between what is possible and what is being done so clearly that they become impatient, even angry. Meanwhile, people who are merely adequate can mistake standard procedure for wisdom.

That tension matters because it shows up everywhere, from engineering teams to classrooms to companies. A founder looks at a professionalized management system and thinks, this is broken. A manager looks back and thinks, this is how serious organizations operate. Both can be sincere. Both can be wrong in different ways.

The deeper question is not who is confident and who is humble. It is this: how do you tell the difference between a broken system and a perspective that has simply outgrown the system?

Confidence is not the same thing as calibration

The popular version of the confidence problem says that the least skilled people think they are the best. That story is too simple. The more interesting pattern is that people at every level tend to rank themselves above those below them, while still recognizing that the truly excellent are better. In other words, confidence usually preserves the shape of the hierarchy, but warps the distance.

That distinction is important. It means overconfidence is not usually a flat denial of reality. It is more like a distortion field. A person who is bad at a task may still know enough to see that some people are better. What they lack is the ability to see the full size of the gap between their own performance and those who are genuinely strong.

This has a practical consequence: the danger is not always that weak performers think they are top tier. Often, the danger is subtler. They think they are a little better than they are, and because that error feels modest, they never correct it. The system rewards them for being decent at self-presentation, not for having accurate self-knowledge.

The most dangerous illusion is not total delusion. It is small miscalibration that never gets forced to update.

Now connect that to leadership. In many organizations, the people who rise through management are not the best at the underlying craft. They are the best at operating the machine as it exists. They know how to schedule, coordinate, delegate, review, and communicate in a way that keeps the institution stable. That is real skill. But it can also create a false sense that the institution itself is optimized.

A founder, by contrast, may be better at seeing the mismatch between reality and process. Because they built the thing, they can often sense when a rule exists because it is useful and when it exists because it merely allows the organization to continue looking orderly. To a manager, that can look like impatience or recklessness. To a founder, it can feel like sanity.

The disagreement is not just about preference. It is about what level of the system you believe is being judged.

The manager’s lens and the founder’s lens

Professional management is built to scale predictability. It asks, in effect: how do we make this organization run acceptably even when no one person can hold the whole thing in their head? That is why managers value process, reporting, delegation, and consistency. These are not merely bureaucratic habits. They are tools for keeping a complex machine from fragmenting.

But founders often judge the same organization by a different standard. They ask: does this still feel like the thing that should exist? Is this process serving reality, or is reality now being forced to serve the process? That difference sounds subtle until you see it in practice.

Imagine a restaurant that grew from a single beloved kitchen into a chain. The manager worries about standardization: every plate should taste the same, every shift should run smoothly, every location should meet target margins. The founder walks into one branch and notices that the chef no longer tastes the soup before serving it because the checklist says the soup has already been approved. The manager sees compliance. The founder sees that the living intelligence of the kitchen has been replaced by paper.

This is why founder mode often feels not like an alternative style, but like a correction. The founder is not necessarily rejecting structure. The founder is rejecting structure that has become detached from first principles.

And yet here is where the Dunning-Kruger insight matters. The founder’s dissatisfaction does not automatically make the founder right. A person can be deeply competent and still misread the domain they are entering. They can be better than the average manager at seeing rigidity, while still underestimating the value of systems they no longer have to personally maintain.

That is where many organizations get stuck. Managers can become overconfident in process. Founders can become overconfident in instinct. The result is not simply a clash of styles. It is a clash between two kinds of miscalibration.

One side says: if you trust judgment over process, chaos follows. The other says: if you trust process over judgment, decay follows. Both are partly right. The problem is not choosing one forever. The problem is knowing when each is becoming the costume of its own success.

A useful framework: three levels of competence

To make this more concrete, it helps to think in three layers.

1. Task competence

This is the ability to do the work itself. Writing code, closing deals, designing products, diagnosing problems, teaching a class. People can be excellent or mediocre here, and often they know enough to recognize some differences in quality.

2. System competence

This is the ability to design the environment in which the work happens. Hiring, incentives, review cycles, communication norms, decision rights, quality control. A person can be weak at the craft but strong at the system, or vice versa.

3. Reality competence

This is the hardest layer. It is the ability to tell when the system is serving reality and when it has begun substituting for reality. A team may have immaculate metrics but be solving the wrong problem. A company may have excellent meeting discipline but be losing contact with customers. A founder often has unusually high sensitivity at this level, because they remember the original reason the organization exists.

This framework explains why people can talk past each other so easily.

A manager who is strong in system competence may say, “We need a more disciplined process.” A founder who is strong in reality competence may say, “We need to stop pretending the process is the product.” Neither statement is enough by itself. Structure without reality becomes performance theater. Reality without structure becomes heroic improvisation.

The best organizations are not built by choosing one and abolishing the other. They are built by keeping the tension alive.

Healthy organizations do not eliminate friction between instinct and process. They make that friction legible.

That is the hidden connection between confidence bias and founder mode. In both cases, the critical error is not simply being wrong. It is failing to understand the level at which you are wrong.

A mediocre performer may think they are better than they are, but not by enough to trigger self-correction. A manager may think the company is healthier than it is, because the process metrics look good. A founder may think the machine is more broken than it is, because they can no longer tolerate the compromises required to make it run at scale. The common failure is miscalibration of scope.

Why strong people often sound unreasonable

One reason excellent people are often dismissed is that they violate the social expectations of their role. A good founder may interrupt a meeting to ask a question that seems naive, because they are testing whether the organization has drifted away from something basic. A good engineer may reject a “clean” architecture because it hides a practical failure. A good teacher may abandon a polished lesson plan because the students are not actually learning.

To someone inside the formal system, this can feel disruptive. Why not trust the process? Why not let the expert function within the lanes already defined?

Because sometimes the lanes are the problem.

This is where the conversation about competence becomes psychologically interesting. The more adept someone is, the more likely they are to notice hidden costs that others normalize. That does not make them automatically correct, but it does make them more likely to feel the system is off. The system is often optimized for average cases, while strong people are trying to solve for edge cases, exceptions, and future failure modes.

A founder sees the restaurant chain and notices that every branch is safe but uninspired, consistent but declining in soul. A manager sees the same chain and notices that profitability and predictability are holding. Which is the truth? Both, depending on the horizon you care about.

This is why many leadership disputes are actually disputes about time scale. Managers optimize for stability across the quarter. Founders often optimize for viability across years. A process can look good in the short run while quietly degrading the organism that depends on it.

And yet the reverse also happens. Founders can chase purity and break the machinery that makes excellence repeatable. A company cannot live on inspiration alone. A beautiful insight that cannot survive contact with scale is not a strategy, it is a mood.

So the question is not whether instinct should defeat process, or process should defeat instinct. It is whether the organization can tell when one has begun to impersonate the other.

The real skill: knowing when to trust your irritation

One of the most valuable forms of intelligence is not knowing the answer. It is knowing what your irritation means.

If you are a founder or highly competent operator and you feel that a system is broken, ask a sharper question: is this a true signal of degradation, or is this the friction of scale?

Here is a simple test.

If the process creates reliable outcomes but feels emotionally unsatisfying, you may be encountering necessary bureaucracy.
If the process creates acceptable metrics but bad real-world results, you may be seeing system theater.
If the process annoys strong performers but protects the organization from hidden failure, it may be doing its job.
If the process protects weak performers while blocking real judgment, it may have become defensive camouflage.

This test is useful because it avoids a common mistake: treating discomfort as proof. Strong people are often right to be dissatisfied, but dissatisfaction alone does not distinguish between genuine dysfunction and the discomfort of being constrained by scale.

The same applies to confidence. A person who is doing poorly may still have enough insight to know they are not doing well, but not enough to know how far off they are. A person who is doing well may be tempted to assume that because they can see flaws in a system, they should be able to redesign it instantly. Both need calibration.

The goal is not to eliminate self-confidence. The goal is to build structured self-correction. In practice, that means surrounding intuition with evidence, and surrounding process with people empowered to break it when reality demands it.

Key Takeaways

Confidence is usually distorted, not inverted. People often know roughly where they stand, but underestimate the size of the gap between themselves and true experts.
Founder frustration is often a signal, not just a mood. People who built a system can detect when process has drifted away from purpose.
Process and judgment solve different problems. Process scales reliability. Judgment rescues organizations from stale rules.
The hardest skill is reality competence. It is the ability to tell when an organization is serving the world and when it is serving its own procedures.
Use irritation as data, not as verdict. If something feels broken, test whether it is actually broken, or merely constrained by the demands of scale.

The deeper lesson: competence is not just about being right

The most important lesson here is not that some people are overconfident and others are visionary. It is that competence changes what you can perceive, and what you can no longer tolerate.

Incompetence tends to create illusion because it cannot see the gap. But high competence creates a different problem: it can make the existing system look intolerably narrow. That is why the same person may be both right and difficult. They are not merely complaining. They are sensing the gap between what the system currently rewards and what reality actually requires.

But that gift has a danger. The more clearly you see the system’s limits, the more likely you are to forget why those limits were created in the first place. Every process was once a solution to a prior mess. Every manager who looks conservative may be preserving hard-won stability. Every founder who looks reckless may be defending the living core of the organization.

So the mature posture is neither deference nor rebellion. It is the ability to ask, repeatedly: what level of the system am I seeing clearly, and what level am I blind to?

That is a much better question than, am I confident or humble? It recognizes that competence is not a personality trait. It is a relationship between perception, judgment, and the systems we build to survive our own limitations.

And once you see that, the apparent contradiction dissolves. The people who think the system is broken are not always deluded. Sometimes they are the only ones still able to see that it was built for a world that no longer exists.