Organizations can only absorb so much improvement at a given time before they reject the person providing the feedback. Being rejected while trying to help is painful. It is also a predictable outcome — and failing to predict it is a failure of the person giving feedback, not the system receiving it.
Simple Picture
Every relationship has a feedback pipe — a conduit with fixed bandwidth. Some pipes are wide: a trusted mentor, a long-time collaborator, someone who has earned the right to be blunt. Some pipes are narrow: a new colleague, someone in active conflict with you, someone above or below you in hierarchy. Some pipes are completely blocked: no feedback can successfully be transmitted, and every attempt widens the blockage.
The mistake is treating all pipes as if they were the same width. The importance drive explains what sizes the pipe: every person is starving for appreciation, and feedback that threatens their sense of importance closes the pipe regardless of its accuracy. A barber lathers a man before he shaves him. The person who has crucial, correct feedback but pushes it through a pipe too narrow to carry it does not improve anything — they become hard to work with.
The Accountability Trap
The most common failure pattern: you cannot do your job effectively because of a peer’s low performance. You escalate to the appropriate manager. The manager transforms the performance issue into a relationship issue — it is not that the peer is underperforming, it is that the two of you “don’t like each other.” Instead of the manager’s responsibility to resolve, it is now your responsibility to “work on the relationship.”
By attempting to drive accountability in their peer, the main character has blocked their own progress (“they’re just hard to work with”) without accomplishing anything. — Will Larson
The structural trap: you think you are bringing a new problem to that manager. What you are actually doing is trying to hold that manager accountable for not solving a known problem — and holding someone more senior accountable for a known failure almost always ends poorly. The manager already knows. They have chosen not to act. Your escalation is not information; it is an implicit accusation.
This is the expert-novice impasse in organizational form. The expert sees the problem clearly, escalates with conviction, and is baffled when the response is hostility rather than gratitude. But the expert’s first-order agenda — “let me educate you to appreciate me” — is exactly what makes them easy to dismiss. The Gervais Principle adds the power dimension: the Sociopath manager already sees the problem and has decided the cost of fixing it exceeds the cost of tolerating it. Your feedback is not news. It is a threat to a decision already made.
Commentary vs Feedback
A critical distinction: feedback is input directed at someone with the intent to change their behavior. Commentary is opinion about a situation expressed to the room. Most people think they are giving feedback when they are actually providing commentary — and commentary has almost no positive impact.
Commentary polarizes teams. It harms relationships, reducing the bandwidth of the feedback pipe. It makes you harder to work with, even when the commentary is accurate. The person who says “this codebase is a mess” in a team meeting is not giving feedback to anyone — they are performing frustration, and the performance costs them the credibility they would need to actually change anything.
The right to criticize is granted only from superabundance — when the critic has so much that sharing costs nothing. Commentary from a place of frustration is the opposite: it is criticism from scarcity, and it reads as neediness wearing the mask of standards.
The Power Gradient
There is a power dynamic in feedback that most people acknowledge in theory and ignore in practice. The less senior you are perceived to be, the more you are expected to have strong rationale before delivering feedback. A VP can say “I don’t love this direction” and the team pivots. A junior engineer saying the same thing needs data, precedent, and political cover.
This is not fair. It is also not going away. Ignoring it does not make you principled — it makes you ineffective. The legibility frame applies: the senior person’s feedback is legible because their position grants it weight. The junior person’s feedback must manufacture its own legibility through evidence and relationship, or it will be processed as noise.
Feedback Lawyers
The mirror failure: people who reject feedback by fixating on weaknesses in its delivery rather than engaging with its content.
They become feedback lawyers who fixate on the weakness in how feedback was delivered rather than trying to understand the content within the feedback itself.
This is disgust weaponized as a defense mechanism — the feedback lawyer ejects the input by finding it contaminated, regardless of the signal it carries. The tone was wrong. The timing was wrong. The context was wrong. Each objection is technically valid and functionally a refusal to hear anything that threatens the current self-model. This is paradigm-lock-in at the personal level: the framework that makes you wrong is the same framework you use to evaluate whether to listen.
The Absorption Rate
The deepest insight: even when feedback is correct, well-delivered, and aimed through a pipe with adequate bandwidth, organizations have a maximum absorption rate. Push more improvement than the system can metabolize and the system rejects you, not the improvement.
This means the strategically correct move is often to hold back correct feedback. Not because it is wrong, but because delivering it now will reduce your capacity to deliver future feedback that matters more. The pipe has bandwidth. Every message you send either maintains or degrades that bandwidth for the next message.
Work environments change slowly. It benefits your team more to give them feedback about how they can succeed in their current environment than to agree with them about how the environment fails to support them. The first is actionable. The second is commentary — and commentary, however accurate, burns pipe bandwidth without producing change.
Dimwit / Midwit / Better Take
The dimwit take is “if you’re right, say it — people need to hear the truth.”
The midwit take is “wrap feedback in a compliment sandwich and people will accept it.”
The better take is that the correctness of feedback is nearly irrelevant to whether it produces change. What matters is the bandwidth of the pipe, the absorption rate of the receiver, and whether delivering this feedback now increases or decreases your capacity to produce change over time. The best feedback-giver is not the most honest person in the room — it is the person who has invested enough in the relationship that their honesty can actually land. The door-with-hinges-on-your-side principle applies: you control what you say and when, not whether it is received. Accepting that is the precondition for effective influence.
Main Payoff
The person who pushes correct feedback through a pipe too narrow to carry it ends up branded as “hard to work with” — and the label sticks not because they were wrong but because they were right in a way the system could not absorb. The feedback’s accuracy becomes irrelevant once the messenger is discredited. The messenger’s credibility is the message’s delivery vehicle, and burning the vehicle to send one package means no future deliveries.
The fix is not to stop giving feedback. The fix is to invest in the pipe before you need it, to respect the absorption rate, and to accept that the most effective form of influence is often not saying the thing you most want to say — because the thing you most want to say is usually commentary, and commentary is the most expensive way to change nothing.
References:
- Will Larson, Hard to Work With
- Will Larson, Constraints on Giving Feedback