The Conversation You're Having Isn't the One They Need

What Supercommunicators Teaches PMs About Stakeholder Communication

When a client asks "how's the project going?" My instinct used to be to open Jira, pull the latest status, and give them the honest answer: velocity is trending down, I have concerns about two deliverables, but the team is working through it.

Thorough. Accurate. And sometimes it seemed to leave the client unsatisfied.

I didn't understand why for longer than I'd like to admit. I was giving correct information. The meetings still felt off afterward. A slight disconnect I couldn't locate, a sense that the client had walked away unsatisfied even though nothing I said was wrong. It took years of accumulating those moments before I could name what was actually happening: they weren't asking about the project. They were asking whether they were in good hands, pressure-testing trust, and I kept missing the actual question.

The Framework I Wish I'd Had Earlier

Charles Duhigg's Supercommunicators isn't a project management book, but it gave me language for something I'd been fumbling toward by instinct. His argument is that most communication failures aren't about content. They're about type mismatch. People are having different kinds of conversations and don't know it.

He identifies three types: practical (what do we do?), emotional (how do we feel about it?), and social (who are we to each other?). Worth being honest about the limits here. These are a useful organizing lens from a well-observed business book, not a clinical framework. And the limitation that matters most for PM work is this: the types aren't reliably visible on the surface. Someone can look like they're having an emotional conversation when they actually need structure. Someone can sound like they want data when what they need is reassurance. The framework tells you what question to ask about a conversation. It doesn't answer it for you.

Which means the framework is a starting point, not a diagnosis. You still have to watch the person in front of you, what they're moving toward, what they're trying to put down, whether they want you to organize the room or get out of it.

Reading the Room, and Getting It Wrong

The hardest part isn't understanding that conversation types exist. It's figuring out which one you're in before you've already responded badly, and I want to be specific about what that actually looks like, including when I've gotten it wrong.

Mid-sprint on a large enterprise engagement, a stakeholder pivoted priorities on us. The way it came out was a mess: jumping between concerns, contradicting himself, circling back to things we'd already resolved. My first read was that he was anxious and needed space to process. I gave him that space. It made things worse.

He wasn't anxious. He was trying to construct an argument for his leadership and couldn't get the pieces to fit. He needed someone to pull in all the facts, organize his thinking, not witness it. When I stopped mirroring his disorganization and started structuring it: here are the three options as I understand them, here are the tradeoffs, his confidence settled. We got our working relationship back on track with a lot less friction.

I turned that into a rule too quickly: structure over space, clarity over processing room. But situations don't hold still for rules.

Recently, I worked as project manager for a political candidate during a particularly ugly campaign cycle. Mud-slinging, social media trolling, public attacks, the kind of sustained pressure that accumulates. Our weekly syncs kept drifting off the agenda. She'd come in with energy that had nothing to do with the campaign plan and everything to do with what the week had thrown at her. My response was to redirect us back to tactics. It didn't land, and I kept thinking the problem was her focus.

It wasn't. She needed to unload first. To name what had been hard, get it out of her head, and reorient before she could think clearly about strategy. The practical conversation was fine once the emotional one had happened. I was just refusing to let it happen. When I finally stopped redirecting and listened, the actual meeting took less time than the ones where I'd been fighting the dynamic.

Same surface presentation as the stakeholder situation. Completely different needs underneath. I misread it initially because I was pattern-matching instead of paying attention to what "done" would actually look like for each of them. For the stakeholder, done meant having something he could present upward. For the candidate, done meant feeling confident, and feeling steadier. Those require different responses.

Three Situations Where the Mismatch Is Most Costly

The client who keeps reopening settled decisions.

You have documentation. You have a project tool like Asana, Jira, or Monday.com. You have a sign-off. And still, every few weeks, something closed comes back open.

The practical response is to point them to the decision log. I've done that, it protects the project, so it's not wrong. But it rarely fixes the pattern, because the pattern isn't about the decisions. It's about confidence. They're not questioning the choice; they're questioning whether they made the right one. The PM who says "that makes sense to revisit, I hear your concern, and here's why I still feel good about the direction we chose" gives a different answer than the one who says "per our project plan, approved on xyz date." Pointing to the record protects the project. Acknowledging the concern rebuilds the relationship. Both matter, and they're not the same move.

The teammate who goes quiet.

This one I know from the inside. I've quiet quit companies. Not loudly, not dramatically, but in the way Gallup describes in their 2023 State of the Global Workplace report: psychologically disconnected, doing the minimum, filling the seat. The tells were subtle. Details I wouldn't normally miss started slipping. My standards dropped in ways that were hard to name but easy to feel. The founder I worked with noticed. We never had the hard conversation about it. And that silence made everything worse. It wasn’t a surprise when we parted ways.

Gallup puts roughly 59% of workers in this state at any given time. That figure is worth treating as a data point, not a verdict. It doesn't tell you why a specific person disengaged or what would bring them back. But I believe the scale of it, because I've lived it. From my perspective, a process change wouldn't have helped me. Shorter standups, async updates, a different format. None of that gets at the real question, which is whether someone still feels like they belong to the work they're doing.

What fixes a belonging problem is a private conversation where you ask how things are going and actually listen. Not a performance review framing, which is evaluative and produces defensive answers. A conversation with enough safety in it that someone can say something true. When I see a teammate go quiet now, that's what I try to create. I ask how things are going and I don't fill the silence that follows.

The team member who doesn't know who she is in the room.

I had a VA on my team who wanted to demonstrate value to a project sponsor. She built a ten-page deck, thorough, well-researched, genuinely good work. It sat unread for weeks.

The first fix was practical: shorter, one call to action, formatted to match how that sponsor actually communicated. She rebuilt it to a single page. It got a response and the conversation opened up. Format mattered. The original deck was asking the sponsor to do work to extract the point, and he didn't have the bandwidth for that.

But something else came out of that conversation that the format fix didn't touch. As we retroed the engagement, we started noticed what she'd left out. Her own read on the situation. Her recommendation. The thing she actually thought the sponsor should do.  When I asked about it, she said she hadn't thought that was her place.

She wasn't confused about format. She was operating from the wrong answer to a question she hadn't consciously asked: who am I to this person, and what am I allowed to say?

She saw herself as someone who executed tasks and reported up. That self-concept shaped everything, what she included, what she omitted, why she built a deck in the first place rather than walking into a conversation. The format change got her in the door. The identity shift was what changed how she worked going forward. Once she understood herself as a thought partner trusted to bring a point of view, not just synthesize information but actually hold one, her communication changed in ways a template never would have produced.

This is why I build the social layer deliberately, before a project is in motion. Ways of working sessions, kickoff conversations, framing early check-ins as peer-to-peer rather than manager-to-report. Not because it's good for morale, but because when people don't know who they are to each other, even accurate and well-formatted communication falls flat. Nobody's quite sure what they're allowed to say, or to whom. The VA situation took weeks to surface and cost the sponsor's attention in the meantime. Most of that was recoverable. Some of it wasn't.

What to Do With This

Think about the last conversation that felt slightly off. Not a blowup, just the quiet kind of wrong where you gave a complete answer and the other person still seemed unsatisfied. What type of conversation were they having? What type were you having?

When I can't locate the mismatch in the moment, I run three questions in my head: Is there a decision or next step being sought, or is that not actually the point right now? Is this person trying to build toward something or trying to put something down? Would they feel better if I organized this, or if I got out of the way? They don't always give me a clean answer, but they usually tell me what I'm missing, that I'm reaching for structure when someone needs space, or waiting for an emotional signal that was never coming.

The framework doesn't make you right. It gives you something to recalibrate against when the conversation isn't working and you don't know why yet. You still have to watch the person, ask better questions, and sometimes just admit mid-conversation that you've been answering the wrong thing and start over.

That last move, "I think I've been answering the wrong question, can we back up," is underrated. It's awkward for about ten seconds, and then it usually changes the whole vibe of the conversation. Most people are relieved someone named it.