Most of what gets called "constructive criticism" in a boardroom is neither constructive nor criticism. It's discomfort wearing a lanyard. I've sat across from leaders delivering what they were certain was helpful feedback, and watched the person on the other side of the table shut down mid-sentence. The feedback was accurate. It still did damage. That gap — between being right and being useful — is the whole subject of this article.
Here's my actual position, the one I'll defend in any room: feedback and critique are not two ends of the same spectrum, softened or sharpened by tone. They are different acts with different purposes. Feedback is future-facing — it exists to change what happens next. Critique, in its harmful form, is past-facing — it exists to establish what went wrong and who's responsible for it. Confuse the two and you'll deliver a technically accurate assessment that nobody can act on, because you've spent all your words proving a point instead of opening a path forward.
I've coached executives who were brilliant analysts of failure and hopeless architects of improvement. They could dissect a missed deadline with surgical precision — sequence of events, who dropped what, exactly where the plan broke. What they couldn't do was turn that dissection into a next move the other person could actually take. That's not a communication-skills gap. It's a category error: they were doing critique when the moment called for feedback.
My lens for telling the two apart
- Direction of attention: Feedback points forward, at what happens next. Critique points backward, at what already happened and whose fault it was. If you can't name the next action inside the first thirty seconds, you're delivering critique, not feedback — no matter how calmly you're saying it.
- Who the message actually serves: Helpful feedback serves the recipient's growth. Harmful critique serves the deliverer's need to be right, to vent, or to be seen as tough. Ask yourself honestly, before you open your mouth, whose interest this conversation is really protecting.
- Specificity of behaviour, not identity: "You missed the deadline because the handoff to design happened two days late" is feedback. "You're not someone I can rely on" is critique dressed as feedback. The first names an event; the second names a person. Only the first is fixable.
- What happens to the room after: Good feedback leaves the recipient with energy to act. Harmful critique leaves them with energy to defend. Watch the body language in the ninety seconds after you deliver the message — that's your actual scorecard, not your intent going in.
- Whether it survives being written down: I test my own feedback by imagining it typed into an email with no tone of voice to soften it. If it still reads as respectful and specific in plain text, it was feedback. If it only worked because of a warm delivery, the words themselves were doing harm and the tone was covering for them.
There are two types of criticism, and only one of them is worth your time
Destructive criticism is easy to spot in its crude form — the sarcasm, the public dressing-down, the comment that makes fun of the mistake rather than addressing it. What's harder to spot is destructive criticism wearing professional clothing: the calm, measured, entirely "appropriate" comment that still leaves no room for the other person to do anything except feel bad. Tone isn't the test. Direction is the test. Does the comment point at a fixable behaviour, or does it point at the person's worth?
Constructive criticism does the opposite work. It offers a specific behaviour to change and, ideally, a next step to try. It treats the mistake as information rather than as a verdict. This is the distinction I actually coach to, and it's worth restating plainly: helpful feedback is not criticism delivered nicely. It's a completely different mental model of what the conversation is for.
I want to be precise about why this distinction keeps getting missed, because it isn't stupidity or malice — it's a shortcut the brain takes under pressure. When something goes wrong, the fastest cognitive move is to locate a cause, and the fastest cause to locate is usually a person. "This slipped because Sarah didn't chase the vendor" is quicker to produce than "this slipped because we don't have a clear ownership handoff between procurement and delivery." The first is critique. The second is feedback that can actually change the system. Leaders under time pressure default to the first because it's cognitively cheaper, not because it's more true.
I've watched this shortcut do real damage in performance reviews specifically, because the format itself invites it. A review is structured as a look backward over a defined period, which nudges even well-intentioned managers toward cataloguing what happened rather than architecting what happens next. The best reviews I've been part of flip that ratio deliberately — five minutes on what occurred, twenty-five on what to do with it. The worst mirror the destructive pattern almost exactly: a list of shortfalls with no attached next move, delivered as though naming the gap were itself the solution.
There's also a power dynamic hiding inside this that most leaders don't examine closely enough. Critique flows easily downhill — from manager to report, from board to CEO on a bad day — because the person delivering it holds the leverage and doesn't need the other person's cooperation to make the point stick. Feedback, by contrast, only works with the recipient's cooperation; it requires them to actually take the next step you've offered. That asymmetry is exactly why critique is the lazier default under stress: it doesn't need buy-in to feel complete. Feedback does, which is precisely what makes it harder to deliver well and more valuable when you manage it.
What makes feedback land — five things I actually look for
- It's specific enough to act on immediately — Vague feedback ("be more strategic") leaves the recipient guessing, which is its own quiet cruelty — it looks like feedback but functions as critique, because there's nothing to do with it. Specific feedback names the exact behaviour and the exact context it happened in.
- It's oriented toward solutions, not verdicts — The best feedback I give sounds like "what I'd try differently next time," not "what went wrong." Both may be true. Only one moves the person forward. Harmful critique stops at the diagnosis and calls that the whole job.
- It's delivered at a pace the recipient can process — Calm delivery isn't about being gentle for its own sake — it's about keeping the other person's thinking brain online. The moment someone feels attacked, they stop listening for content and start listening for threat. Nothing you say after that point actually lands.
- It targets behaviour, not character — "The report was late" is behaviour. "You're disorganised" is character. The SBI model — situation, behaviour, impact — is the cleanest tool I know for staying on the behaviour side of that line, because it forces you to describe what happened rather than who someone is.
- It survives the recipient repeating it back — I sometimes ask the person I've just given feedback to summarise what they heard. If what comes back sounds like an indictment, I didn't deliver what I thought I delivered — regardless of how carefully I'd chosen my words going in.
Why calm delivery isn't the safeguard people think it is
Most guidance on giving difficult feedback focuses almost entirely on tone: stay calm, don't raise your voice, pick the right moment. I don't disagree with any of that as basic professional conduct, but I think it's solved the wrong problem. Tone determines whether the conversation feels comfortable. It does not determine whether the content is useful. You can deliver a completely content-free, character-based judgement in the gentlest voice in the room and it will still leave the recipient with nothing to do except absorb the blow.
I ask the leaders I coach to separate two questions they habitually merge: "was I professional?" and "was I useful?" The first is about your conduct. The second is about their outcome. It's entirely possible to score highly on the first and fail completely on the second, and I'd argue that's the more common failure mode among senior leaders specifically, because most of them have already learned not to shout. They've solved the easy problem and stopped there, mistaking the absence of visible aggression for the presence of genuine help.
The line I actually draw
If you want one sentence to carry out of this: harmful critique explains a problem: helpful feedback explains a path. That's not a rhetorical flourish — it's the operational test I use in every difficult conversation I have with a client or a team I'm coaching. Before I speak, I ask myself which of those two things I'm about to hand the other person. If I can't answer that question, I'm not ready to speak yet.
This matters more at senior levels, not less. The higher someone rises, the fewer people are willing to give them a straight assessment of anything — which means the feedback that does reach them carries outsized weight, and the harm from getting it wrong compounds faster too. A junior employee getting harsh, vague critique from a manager has other input in their life to correct for it. A senior executive getting harsh, vague critique from a board member or a CEO often doesn't; that comment can sit in their head for months, unresolved, because nobody offered the second half of the conversation — the path forward.
Where this breaks down in practice — three patterns I see repeatedly
The first pattern is what I call feedback laundering: a leader delivers pure critique but wraps it in feedback-shaped language so it feels compliant with whatever training programme they last attended. "I just want to give you some feedback" followed by an unbroken list of everything that went wrong, with no proposed next step, isn't feedback because it opens with the word. The label doesn't change the direction of the message. If the sentence that follows only describes the past, relabelling it doesn't make it future-facing.
The second pattern is the feedback sandwich, which I actively coach people away from, and I know that puts me at odds with a lot of conventional advice. Praise, then criticism, then praise again sounds considerate on paper. In practice, recipients learn the pattern within a few rounds and start listening only for the middle slice, discounting the praise on either side as padding. Worse, it trains the deliverer to soften the actual content to make the sandwich feel balanced, which is how specific, actionable feedback quietly turns into vague, comfortable feedback. I'd rather see a leader give one clear, well-aimed piece of feedback than three wrapped in enough cushioning that none of them land.
The third pattern is delayed delivery disguised as tact. A leader notices a problem, decides the timing isn't right, and sits on it — sometimes for weeks — telling themselves they're being considerate. What's actually happening is that the feedback is decaying. By the time it's delivered, the recipient can't reconstruct the specific behaviour being referenced, so the conversation collapses into generalities and character judgements almost by default, because there's no longer a concrete event to anchor it to. Timeliness isn't a nicety in this model. It's a structural requirement for the feedback to stay feedback rather than curdling into critique.
What I ask leaders to do differently
The practical shift I coach is small in mechanics and large in effect: before any difficult conversation, write down the single next action you want the other person to walk away with. Not the problem. The action. If you can't complete that sentence before you walk into the room, you're not ready to have the conversation — you'll default to describing the problem in detail because that's the part you've actually prepared, and the recipient will feel the absence of the rest even if they can't name what's missing.
I also ask leaders to notice their own physiological state before delivering anything difficult. Frustration is data, not a delivery mechanism. If you're currently annoyed about the missed deadline, that's useful information about your own threshold — it is not useful information to hand to the person in front of you unfiltered. The conversation you have while frustrated and the conversation you'd have twenty minutes later, once you've named the actual behaviour and the actual next step to yourself first, are rarely the same conversation. I've re-run this exact experiment with clients: same issue, same person, twenty minutes of preparation in between, and the entire tenor of the exchange changes because the content changes, not just the tone.
I'd also push back on a comfortable myth: that harmful critique is always loud or cruel. The most damaging feedback I've watched land in a boardroom was delivered in a completely flat, professional, even kind tone. It was still critique, not feedback, because it never once pointed anywhere except backward. Calm delivery makes bad feedback easier to sit through. It doesn't make it good feedback. Don't let politeness fool you into thinking the job is done.
None of this means feedback has to be softened to be kind. Some of the most useful feedback I've ever received was blunt to the point of being uncomfortable, and I didn't experience it as harmful, because it was unmistakably aimed at a behaviour I could change and it arrived with an implicit next step attached. Bluntness and harm aren't the same axis as directness and vagueness. You can be direct and kind at once, and you can be gentle and cruel at once — the softness of your voice was never the variable that mattered.
So my closing position is this: stop asking whether your feedback was "nice enough." Ask whether it was navigable. Could the person who just heard it stand up, walk back to their desk, and know the very next thing to try differently? If yes, you did your job, however plainly you said it. If no, you delivered critique — accurate, maybe even fair, but not feedback — and the real, harder work of actually helping that person hasn't even started yet.
Further reading: 10 Essential Characteristics for Successful Leaders, 10 Cross-Functional Leadership Skills Every Manager Needs, 5 Creative Managerial Strategies to Solve Poor Performance