I've interviewed a lot of people on stage over the years, but this one was different. Sitting across from Professor Brian Cox, the particle physicist justifiably famous for making the complex physics of the universe feel comprehensible to the rest of us, I half expected us to geek out and float off into galaxies and quantum mechanics, never coming back down to anything I could use back at my desk.
We did the opposite. The more we talked about how he explains the hardest ideas in science, the more it sounded like he was describing the job of leadership itself - communicating clearly, earning trust, and staying honest under genuine uncertainty. And the single most useful thing he said had nothing to do with physics at all.
The "I don't know" moment
He told a story about being on stage in Melbourne, in front of about 6,000 people, running a supercomputer simulation of the early universe. He was mid-sentence, saying "the input to this simulation was…", when he realised he didn't actually know what the input was. Random seeds? Real data? He wasn't sure.
So he said exactly that. "I actually don't know. I don't know the answer to that question." Then he went home, read the paper that night, and came back with the answer.
"The key to presenting complex ideas," he said, "is don't bullshit."
It's such an unglamorous piece of advice and so completely right. The higher you climb, the more pressure there is to project certainty. Certainty about the strategy, certainty about the market call, certainty about why the numbers moved, certainty about what the board wants to hear. And certainty performs well in the room: it's reassuring, it ends the meeting faster, it looks like leadership. But it's also the fastest way to erode the trust you most need, because the people around you can eventually tell the difference between someone who knows and someone who is performing knowing. And once a team catches a leader performing, they quietly discount everything else you say.
Brian's version of credibility is the opposite. It's earned precisely because he's willing to say "I don't know" in front of 6,000 people. The "I don't know" is what makes you believe the rest. Authority that can admit its limits is far more durable than authority that can't.
Clarity is a test, not a talent
The reason he can simplify so well, it turns out, has little to do with a gift for metaphor. It comes from refusing to fake his own understanding first.
He quoted Richard Feynman: to truly explain something complex in a simple way, you have to genuinely understand it. But what really struck me was another Feynman line: "Don't kid yourself. You're the easiest person to fool." You can gloss over a gap in your own knowledge, nod along to your own explanation, and never notice the hole until someone falls into it.
That's the part worth sitting with as a leader. We treat "communicating clearly" as a packaging problem: the all-hands narrative, the board deck, the one-liner that's supposed to align a thousand people. But clarity is really a diagnostic. If you can't explain your strategy simply and honestly, it's rarely because you haven't found the right words. It's because the thinking underneath isn't finished. The muddled strategy and the muddled understanding are the same problem wearing different clothes. The communication just exposes it, which is exactly why the discipline of clearly communicating it is so valuable, and so often avoided.
This is why I treat clarity as a test rather than a finishing flourish. The real measure of a strategy isn't whether it survives the board deck; it's whether your team can say it back to you, in their own words, without the footnotes. If a direction only holds together with three caveats, five slides, and a bit of borrowed jargon, you haven't finished thinking. You've just decorated the confusion. So when a direction comes out tangled, I've stopped reaching for a better wordsmith. I go back and ask whether we actually understand the problem yet. The cost of getting that wrong scales with your seniority: a leader who can't explain the "why" simply will watch it mutate into a dozen contradictory interpretations on the way down their decision stack.
Who do you trust when you're not the expert?
I asked him how, as a science communicator, he decides who to listen to in fields where he isn't the authority, like quantum computing or AI. His answer was disarmingly simple: when genuine experts disagree, treat it as a signal that nobody knows. "Who am I to take a side? I'm not an expert."
For anyone making consequential bets at the edge of their own knowledge (which is most strategic decisions, most of the time), that reframing is freeing. A leader's real job here is to hold the uncertainty honestly, weigh the disagreement, and make the call knowing you own it, rather than hunting for the one expert who's definitely right and borrowing their conviction. Confidence isn't pretending the fog isn't there. It's deciding anyway, with your eyes open, and being willing to update when you're wrong. The senior version of this is harder, because the higher up you are, the fewer people will tell you the fog exists at all.
Wanting to know, not wanting to be right
My favourite thing he said was this: there's a difference between wanting to be right and wanting to know. The first is ego. The second is progress. He talked about another physicist who got in touch to tell him he'd got something wrong on air, and how they ended up becoming friends over dinner, because being corrected wasn't a humiliation, it was the whole point.
That distinction is the dividing line between leaders who compound and leaders who stall. Because most strategy, in practice, is run as an exercise in being right. We arrive at a plan, fall in love with it, and then spend our energy defending it: marshalling the data that supports it, explaining away the data that doesn't. The plan becomes an extension of our judgment, and challenging it starts to feel like challenging us. That's exactly backwards. A strategy isn't a verdict to be protected. It's a hypothesis to be tested, a set of bets about the future that are, by definition, not yet known to be true.
The shift sounds small and changes everything. Instead of asking "How do I prove this is the right plan?", you ask "What would have to be true for this to be the right plan?" That's Roger Martin's question, and it's one of the most useful tools I know for taking the ego out of strategy. It quietly flips the conversation from defending a conclusion to listing the assumptions the conclusion depends on: the things you'd need to believe about your customers, your market, your capabilities, your competitors. And once those assumptions are on the table, you can do the genuinely useful thing: figure out which ones you're least sure of, and go test those first. You're no longer arguing about whether the strategy is right. You're agreeing on what would make it right, and then going to find out.
This is the muscle every leadership team needs and most under-train. We get attached to our own conviction at exactly the moment the organisation needs us to be testing it. Our seniority makes it dangerously easy to win the argument and lose the truth. The most valuable thing a leader can model is the visible, public willingness to say "here's what would have to be true, and here's what we don't yet know," and to treat the day someone proves part of it wrong as a good day rather than a bad one.
Protecting the wonder
There's a thread of all of this I don't want to lose, even in a post aimed at leaders, because Brian made all of us feel it in the room: the sheer wonder of the thing. Two trillion galaxies. Every point of light in tonight's sky with its own solar system. Two collections of atoms that, 13.8 billion years after a featureless Big Bang, can sit on a stage and have a conversation about physics and leadership.
When I asked how he keeps a sense of wonder alive in people, he flipped the question. "It's more, how do you not destroy it?" His example stuck with me: Andre Geim, who won a Nobel Prize for graphene, discovered it "messing around on a Friday afternoon," because he ran a lab that insisted people have a Friday afternoon to mess around in. The curiosity was always there. The leader's job was simply not to crush it.
That's the strategic responsibility of running an organisation. Curiosity, creativity, the willingness to sit with a half-formed idea: you don't install these with a values poster or an innovation initiative. You keep them by not flattening them with process, quarterly pressure, and the relentless executive demand to look certain. The bigger the organisation you lead, the more of that flattening you do without noticing, and the more deliberate you have to be about protecting the space where the next idea actually comes from.
So here's what I'm taking from an hour with someone who studies the universe for a living, and pointing it squarely back at the work of leading: understand the thing deeply enough to explain it simply. Say "I don't know" out loud, especially when everyone expects you to have the answer. Want to know more than you want to be right. And protect the Friday afternoons: yours, and everyone's you're responsible for.
Thank you, Brian. I came for the physics and left with a leadership philosophy.
