Masters in Public Affairs
Episode 13 · published 2 September 2026
64 min
13

Supercommunicators

Charles Duhigg, 2024
The person receiving your message isn't one person. They're a stack of identities, and only one of them is listening.
0:0064 min

Why This Book Matters

Duhigg wrote this to fix communications challenges in his own life. Despite being a Pulitzer-awarded writer and building a career explaining why people do what they do, he couldn't connect with his own team — or, when he got home, with his wife and his children. Notionally, the book is him working out why.

Which means it is, on its face, a book about one-to-one, or one-to-few conversations. Nothing in it was written for anyone running a public affairs campaign, even though it offers us several important lessons.

Ever felt like you followed every best practice for writing a campaign message and still fell flat? Me too. I wonder if the live identity challenge is what we faced in that moment?

Supercommunicators has a place in the canon because Duhigg surfaces something so fundamental: at any moment only one of a person's many identities is doing the listening. The interesting challenge is that the surrounding context has more to do with which identity is live, than the message itself.

The book's own claim about itself is that none of this is innate. Its opening exemplar is a shy kid who learned to connect by studying people who were good at it, and who spent two decades at the FBI talking fugitives out of buildings. "It's a set of skills," he tells a room of sceptical scientists. "There's nothing magical about it."

Several weeks of stewing on this and I still don't have the answer.

What the book does not do is scale. Duhigg's answer to nearly every problem feels like it's some version of getting closer to someone. Which might be scalable, but I struggle to see how. I do believe if we can scale what Duhigg teaches us here, we could transform our campaigns and our advocacy efforts. But the playbook isn't obvious. At least, not to me, so I welcome any feedback you have if the playbook is obvious to you.

What We Cover in the Episode

  • Behfar Ehdaie, a prostate cancer surgeon at Memorial Sloan Kettering, whose patients kept choosing an operation he had just shown them they didn't need. Around 40% of prostate cancer patients in the United States do the same — more than fifty thousand men a year.
  • Felix, the FBI's negotiator of last resort, and the hour he spent with a room of scientists who couldn't work out what he actually does. He answers them by demonstrating it on one of them.
  • The three kinds of conversation Duhigg says every discussion is one of — and why two people in different ones fail each other even when both are trying hard.
  • Why the identity doing the listening gets set by the environment rather than by the person. A pro-Obama t-shirt at a barbecue versus at an NRA rally; and a city in Iraq where bouncers read your religion off your ID card at the grocery store door.
  • Dr. Rima Shami, a patient who won't take the COVID vaccine because God will provide, and what happened in the thirty minutes after the appointment ended.
  • The maths test where reminding women of four of their identities did nothing at all, and reminding them of thirty closed the gap with the men completely.
  • A Milwaukee jury in 1985, a man who bought a pistol because a mail-order detective course told him to, and the juror who moved eleven other people without once arguing the case.
  • Salma Mousa's soccer league in a city recovering from ISIS, where nobody was taught to communicate and the effects were still holding six months later.
  • Four mental models: the Salience Dial, the Near-Universal Identity, the Silent Ledger, and the Directive Penalty.
  • Four ways the book gets misread — including the one Duhigg heads off twice, because he clearly saw it coming.
  • The mastery lesson, and the asymmetry between what this book does to your life and what it does to your work.

Bonus: ideas that didn't make the episode

Looping for understanding

This comes out of a chapter on conflict, where gun control activists and gun rights advocates were brought together to see whether people who despise each other's beliefs could hold a civil conversation. The technique that rates most highly is three steps, and then a loop back:

  1. Ask questions.
  2. Summarize what you heard in your own words.
  3. Ask whether you got it right.
  4. Then do it again until the other person agrees you've understood.
Listening that can't be seen does no work. I can’t tell you how this has positively impacted my own personal relationships since I started doing this more mindfully.

A 2020 study found that looping early in a conversation forestalls escalation later, and that people who do it are seen as better teammates and more desirable collaborators.

Mood and energy

Two important variables that work well in person, but that digital environments completely strip out.

Specific emotions turn out to be close to unreadable. A furrowed brow is anxiety or concentration and you cannot reliably tell which; a smile is pleasure or something considerably stranger. So the brain uses a cruder two-variable proxy instead — is this person positive or negative, and are they high energy or low? It's also why matching intensity matters as much as matching direction. A half-hearted chuckle against somebody's belly laugh reads as rejection, not agreement.

What the book is finally for

The afterword closes on the Harvard Grant Study — undergraduates followed for decades — and its finding that the strongest predictor of health and happiness at eighty was how satisfied people were in their relationships at fifty. Social isolation, the researchers found, was more dangerous than diabetes. Duhigg's last word isn't about persuasion or influence at all. It's that connection makes people live longer. So heck, even if you don't see how you can apply this in a public affairs context, you'll still find the time worthwhile.