A chief executive lays out a plan to the executive committee, and the room nods before she has finished the sentence. Nobody interrupts to question the assumption buried on slide four, or to mention that a similar initiative stalled two years ago under a different name. By the time the meeting ends, the plan is approved, and not one person present has said the thing they actually think. This is not a failure of the individuals in the room. It is what reliably happens to information as it travels upward through a hierarchy, and it happens to almost everyone who reaches the top of one.
The Filtering Effect
Subordinates rarely lie to their superiors outright. They curate. Bad news arrives softened, hedged, or wrapped in a recovery plan that makes it sound smaller than it is. A missed target becomes behind but recoverable. A colleague’s failing project becomes facing some headwinds. None of this is dishonest in the ordinary sense of the word. It is self-protective, and it is entirely rational: raising an uncomfortable truth with someone who controls your career carries a cost that raising it with a peer does not. The result is a predictable gradient. The further a fact travels from where it originates to where a decision gets made, the more of its rough edges get sanded off along the way.
The effect compounds with seniority rather than easing with it. A junior manager still hears plenty of unfiltered complaint from the people below them, because several layers of the organisation remain willing to speak plainly at that level. A chief executive, by contrast, sits at the convergence point of every filtering chain in the business at once. Direct reports edit their own accounts. Their direct reports had already edited before that. External advisers who want repeat engagements tend to tell chairs and chief executives what keeps the relationship comfortable. By the time information reaches the top job, it has often passed through five or six rounds of diplomatic revision, each one individually reasonable and cumulatively distorting.
Why the Problem Is Structural, Not Personal
It is tempting to read this as a character flaw: a leader who secretly does not want the truth, surrounded by people too timid to offer it. Sometimes that is part of the picture. But the pattern shows up just as reliably around leaders who explicitly ask for candour, reward it in principle, and would say, quite sincerely, that they want to know when something is wrong. The filtering happens anyway, because it is built into the structure of hierarchical authority rather than into any one person’s character. Whoever controls someone else’s budget, promotion or continued employment will receive a filtered account of reality from that person, however open-minded the listener genuinely is, because the incentive to filter sits with the speaker, not with them.
Boards experience a related version of the same dynamic at one remove. A supervisory or non-executive board typically learns about company performance through the same executive whose performance it is meant to be judging, and whose account of the year understandably emphasises what went well. Two-tier board structures, common across the Netherlands and Germany, were partly designed to address this by separating the management function from the oversight function into distinct bodies with no overlapping membership. Even there, though, the underlying dynamic persists in a milder form: the people charged with judging performance still depend heavily on the account supplied by the people being judged.
What Restores the Signal
Three things reliably help, and none of them are complicated, though all require deliberate effort to sustain. The first is access to people with no stake in the organisation’s internal politics: no promotion to chase, no budget to protect, no rivalry with whoever is delivering the news. An outside voice of this kind can say what an internal one cannot, precisely because their livelihood does not depend on the listener’s continued goodwill.
The second is a norm, actively maintained rather than merely announced, that rewards the bringing of bad news instead of quietly punishing it. This is harder than it sounds, because the punishment is rarely explicit. It shows up as a flicker of irritation, a slightly cooler tone in the next one-to-one, a meeting invitation that stops arriving without explanation. Leaders who genuinely want honest input have to watch for these small signals in their own behaviour and correct for them, because their team will register the signals long before the leader consciously notices sending them.
The third is a relationship built specifically for confrontation rather than comfort. Executive coaching, done well, occupies exactly this space. Arvid Buit’s TRUE Leadership practice frames this as a critical-friendship approach to executive coaching, borrowing a concept long used in education research, where a critical friend combines the loyalty of a friend with the honesty of a critic: someone unambiguously on the leader’s side who therefore has standing to say that the story the leader is telling themselves does not hold up. The friendship half of the term matters as much as the criticism. Confrontation from someone with no evident loyalty reads as an attack and gets defended against; confrontation from someone whose commitment to the leader’s success is not in doubt is far harder to dismiss.
The Limits of Fixing It Alone
None of this removes the need for a leader to build systems that actively surface dissent, because a single outside relationship cannot substitute for an organisational culture. Skip-level meetings, anonymous surveys, exit interviews conducted by someone outside the immediate chain of command: each of these exists because it interrupts the filtering chain at a different point. What an outside thinking partner adds is not a replacement for these mechanisms but a check on the leader’s own interpretation of what they produce, since even accurate data can be explained away by a mind with strong reasons to explain it away.
There is a fair objection to all this, which is that an organisation stacked with formal listening mechanisms and outside advisers can drift toward paralysis, second-guessing every decision against an ever-expanding chorus of inputs. The answer is not more voices but a small number of trusted ones, chosen precisely because they will say the uncomfortable thing rather than the popular one. The seniority that isolates a leader from the truth is not going away, and no framework will make it go away. What can change is whether anyone in a leader’s orbit is actually positioned, structurally as well as personally, to tell them what they need to hear rather than what will keep the meeting pleasant.
