A scientist at a synthetic-biology company I worked with once spent most of our first meeting explaining how his team had rewired ordinary brewer's yeast to make a chemical the world otherwise pulls from petroleum. He was good at it, or thought he was. He drew the pathway on the whiteboard, walked me through the fermentation, laid out the competing approaches and why each had failed. Forty minutes in, I understood two things clearly: the work was important, and the man was brilliant. I still could not have told my mother what the company did.

This happens to me constantly, and it is the most expensive problem I see. The people doing the most important work are frequently the worst at explaining it. They'll tell you the work is just complicated. I think it's closer to the opposite: the work doesn't go unheard in spite of being important, but because of it. Three things tend to be happening at once.

You can't unknow what you know

Psychologists call it the curse of knowledge. Once you understand something deeply, you lose the ability to remember not understanding it, and you start talking to everyone as though they already stand where you do. Ask the founder a simple question and the answer comes back four levels down, somewhere in the chemistry and the edge inside the edge, while the person across the table is still waiting to hear what the thing is and why it matters. The answer isn't wrong. It's built for an audience of one.

I'm useful in that room for an unglamorous reason: I don't get it yet. The whole job is to keep asking the dumb question until the smart answer is short enough to repeat.

Everyone is afraid of the wrong mistake

People who do technical and mission work are trained to fear one error above all others. They're afraid of being slightly wrong: the claim that overshoots, the caveat left out, the nuance flattened to fit a headline. So they hedge. I once watched a founder revise a single sentence in a press release a dozen times the night before a launch. Each version was more precise than the last. Each was a little less likely to be read.

There's a second mistake, and almost nobody fears it: being ignored. A perfectly accurate sentence that no one finishes is worth exactly nothing, and it is far more common than the sentence that goes too far. You don't have to choose between being right and being read. You have to stay with the thing until you find the one true line a stranger will carry out of the room.

The story gets bolted on at the end

In most companies, communication is the last thing anyone thinks about. The science is done, the product is built, the round is nearly closed, and somebody finally says: we should probably do some PR. A release goes out the week before launch, wrapped around a decision that was made months ago.

Narrative isn't the wrapping paper. It's the load-bearing wall.

The teams whose stories travel did it the other way around. They worked out what they were saying, and to whom, long before they had to say it in public. Leave it for the end and you aren't telling people what you built; you're explaining why you waited so long to mention it.

None of this is a marketing problem, and none of it is about being smarter. Usually the people in the room are the smartest ones in their field. What they need isn't volume. It's a translator: someone who can hold the real science in one hand and a stranger's attention in the other without dropping either, and who has sat through enough interviews to feel the exact moment a listener drifts.

The scientist with the yeast got there. A few months later a trade reporter finally wrote the piece he'd been waiting years for. The science in it hadn't changed. The first sentence had. Most of the time, that's the whole job.

— T.G.