Are you listening? Everyone Needs a Steel Man.

I pride myself on being an excellent listener. Which is why I was completely taken aback yesterday when my six-year-old son dissolved into tears in the back seat of the car on the way home from school.

"No! You're not LISTENING, Mummy!"

I was listening. I genuinely was. I heard every word about the complex social dynamics currently unfolding in his Year One classroom. Who said what to whom at lunch, who was and wasn't invited to the game at the morning tea break - the full political landscape of a six-year-old's social world, rendered in considerable detail.

The problem was simple.

I didn't agree with his read of the situation.

I thought, gently and carefully, that perhaps the other child hadn't meant it the way he'd taken it. That there might be another explanation. That sometimes things aren't quite as they seem from inside the moment.

To him, in that back seat, that was indistinguishable from not listening at all.

I drove the rest of the way home thinking about it. About how this pattern runs deeeep.. and what it means for every leader who has ever walked out of a difficult conversation absolutely certain they'd listened well, while the other person walked out certain they hadn't been heard at all.

The gap between listening carefully and being experienced as a good listener is real, measurable, and has almost nothing to do with your listening technique. It has everything to do with whether you agreed.

What the research actually found

In 2024, Ren and Schaumberg published a paper in Psychological Science that will be uncomfortable reading for anyone in a leadership role. Across eleven studies with 3,396 adult participants, using multiple communication formats including face-to-face, video, and text - they found something striking: people are almost completely unable to separate their perception of listening quality from their perception of agreement. 🎤 mic drop.

When a listener agreed with them, participants rated that person as an excellent listener. When a listener disagreed, even a listener who had done everything right, who had reflected back what was said, who had asked clarifying questions, who had been fully present, participants rated them as a poor listener.

The researchers called this "disagreement getting mistaken for bad listening." But I'd reframe it slightly for a leadership context because it's not just that disagreement looks like bad listening. It's that for the person on the receiving end of a disagreement, the two experiences are functionally identical. The emotional result is the same. The sense of dismissal is the same. The damage to trust is the same.

As a leader who disagrees with people for a living, doing all the things you’ll be told make a “good” leader (giving feedback, challenging assumptions, pushing back on proposals, making decisions that not everyone agrees with) this is not a minor finding.

Why this happens and why it's so hard to fix

The mechanism, as best as the research currently explains it, is rooted in how we infer the internal states of other people.

When someone listens to us, what we're really trying to detect is whether we mattered to them. Whether our perspective was understood. Whether they took it seriously. We can't see inside another person's head, so we make assumptions based on what we CAN see. The most obvious cue that "they took my view seriously" is "they updated toward my view."

Agreement is visible evidence that something you said changed the landscape of the conversation. Disagreement, by contrast, looks like the landscape stayed the same. Like your words went in and nothing moved. And if nothing moved, the natural inference is: they weren't really listening. (#my-life-as-a-parent)

When you're delivering feedback someone doesn't want to hear, agreeing isn't an option. When you're making a decision your team disagrees with, changing your mind to avoid the perception of poor listening would be worse than useless. When you're pushing back on a proposal because you can see what the person in the room can't, the whole point is that your view is different from theirs.

And in every one of those situations, the research says, you are likely to be experienced as someone who didn't listen. Even when you did.

The most listened-to leader in any room might simply be the most agreeable one. And those are very different things to be.

What IS fixable

Here's what I find genuinely hopeful about this research, because there is something hopeful in it.

The problem isn't that people are irrational or that the bias is insurmountable. The problem is a sequencing and signalling failure. One that can be deliberately interrupted.

What typically happens in a disagreement conversation is this:

• a person shares their view, the listener processes it, forms a different view, and then expresses that different view.

• The disagreement arrives before the understanding has been demonstrated.

• And so ,the person on the receiving end never gets the felt experience of being heard, because the moment they realise you see it differently, the whole exchange gets reinterpreted.

• They replay the conversation through the lens of your disagreement, and it colours everything that came before it.

The fix is NOT to agree more.

The fix is to demonstrate understanding visibly, specifically, and verifiably BEFORE your position is known.

By that I don’t mean “oh yes, I hear you”. That doesn’t count. Instead:

• A genuine, detailed reflection of the other person's position, in words they would recognise as their own.

• Name the concern underneath their argument, not just the argument itself.

• Acknowledge the logic of their view, not just its existence. Show that you can represent their position as well as they can.

Then, and only then - show your hand.

Reflect before you redirect.

The order matters enormously. When understanding is demonstrated first, disagreement that follows feels like a considered response to something that was genuinely received. When understanding is absent, or arrives after the disagreement, it feels like retrospective justification for a decision that was already made.

I know this because I got it backwards in the car. I moved to reframe before I'd shown my son I understood why it had felt the way it felt. I skipped the step. And the conversation, predictably, went the way conversations go when you skip that step.

The steel man test: a practical tool for leaders

There's a simple diagnostic I've started using with clients, with teams, and now apparently with my six-year-old, that I think is more useful than most of the listening frameworks I've encountered.

Before you express a disagreement, ask yourself: could I pass a steel man test right now?

A steel man is the opposite of a straw man. That means representing the other person's position in its strongest, most compelling form. If you can't do that yet, you're not ready to disagree. You haven't finished listening.

If you can articulate their position as well as they can, and they would recognise it as fair, then you've done the work that earns you the right to a different view. You've demonstrated the understanding before the disagreement. And the research suggests that sequence changes everything about how the conversation goes.

Practically, this often means slowing the conversation down. It means saying, before you share your perspective: "Let me make sure I've understood what you're saying." And then actually doing that (out loud, in front of them), in enough detail that they can correct you if you've got something wrong.

That correction, incidentally, is not a failure of the technique. It's the point of it. You want them to correct your understanding, because every correction is the conversation moving toward the moment where they finally feel heard.

What this means for feedback conversations specifically

The feedback context is where I see this play out most damagingly in organisations.

A leader gives careful, considered, well-intentioned feedback. The recipient leaves feeling dismissed, unheard, like the leader had already made up their mind and the conversation was theatre. The leader, baffled, reports that they listened carefully and gave the person every opportunity to respond.

Both accounts are, in their own terms, accurate.

The leader did listen. The recipient wasn't heard. These things can be simultaneously true when the sequencing goes wrong (i.e. when the feedback arrives before the leader has visibly demonstrated that they understand how the person sees their own performance.)

The most effective feedback conversations I've observed share a structure that looks, on the surface, almost backwards. The leader spends the first portion of the conversation asking the person to give their own account of the situation: what happened, why, what felt hard, what they think went well. The leader listens. And then - critically - the leader reflects back not just the content of what was said but the logic of it. The frame the person is using. The standards they're applying to themselves.

Only then does the leader introduce a different perspective.

The feedback lands differently when it's built on a demonstrated understanding of the person's own account. It doesn't feel like a verdict being handed down. It feels like a conversation between two people who are looking at the same situation and seeing it differently, which is, in fact, what it is.

You don't earn the right to a different view by being senior. You earn it by showing you genuinely understood the view you're about to challenge.

The leadership implication nobody talks about

If your team consistently rates you as a poor listener, if feedback suggests people don't feel heard in conversations with you:

• The first question to ask is not "am I listening enough?" It's "am I disagreeing too quickly?"

• And the second question, which is harder, is: "have I created an environment where my disagreement arrives before my understanding is visible?"

Leaders who disagree frequently and confidently (which is often exactly what good leadership requires) are fighting a perception battle that listening skill alone won't win. The technique matters. The paraphrasing, the eye contact, the questions - all of it matters. But none of it survives the moment your disagreement surfaces if the understanding hasn't been visibly established first.

The most listened-to leaders I know have a habit that looks, from the outside, almost inefficient. They're not being slow. They're sequencing deliberately.

My son didn't need me to agree with him about what happened at lunch. He needed me to show him I understood why it had hurt.

Two questions to carry into your next difficult conversation:

1. Before you express your disagreement, can you pass the steel man test? Can you represent the other person's position in terms they'd recognise as fair?

2. After a conversation where someone seemed to feel unheard, ask yourself honestly: did my disagreement arrive before my understanding was visible? Because if it did, that's not a listening problem. That's a sequencing problem. And sequencing is entirely within your control.