Why Your Best Expert Gives Your Worst Pitch

Your most senior person is often your worst presenter. Not nerves, and not because nobody sent them on a course.

There's a well-documented bias working against them: the curse of knowledge. The fix isn't more coaching. It's a format that gives the bias nowhere to hide.

What the curse of knowledge actually is

Once you know something, you find it very hard to imagine not knowing it. That's the whole bias, and it's remarkably hard to switch off.

Colin Camerer, George Loewenstein and Martin Weber described it formally in a 1989 paper in the Journal of Political Economy. Better-informed people, they found, consistently fail to predict what less-informed people will understand, even when they're paid to get it right. The knowledge leaks into the guess and skews it.

The cleanest illustration came a year later. In 1990 at Stanford, Elizabeth Newton had "tappers" tap out the rhythm of a well-known song on a table while "listeners" tried to name it from the taps alone. Tappers predicted the listeners would get it about half the time. The actual score was 3 out of 120. Roughly 2.5%.

The tapper hears the whole song: melody, lyrics, the drums. The listener gets a knuckle on a table. That gap is the whole problem in one experiment.

Your specialist is the tapper

Swap the tune for a specialism and the experiment runs itself in every meeting room. A senior partner walking through a regulatory change, a technical lead through an architecture: both are tapping out a rhythm they can hear perfectly and assuming the client hears it too.

The connecting steps get skipped because those steps are invisible to someone who spent fifteen years building the mental model. The jargon rarely comes from ego. It's usually vocabulary that stopped registering as jargon years ago, worn smooth by daily use.

I've sat through a lot of pitches, ours and other people's, and the pattern holds with almost embarrassing reliability. The more senior the person holding the clicker, the more steps go missing.

The audience knows it too. In interviews we ran with buyers last year, a senior bid manager at a global engineering consultancy gave us the sharpest version of the test:

if a child can understand it everyone else in the room will

Her logic was blunt. In most buying committees, only one or two people actually know the service being pitched. Everyone else is a finance director, a procurement lead, an operations head. A pitch that assumes the whole room shares the expert's context has lost most of the room before it starts, and the people it loses first tend to be the ones controlling the budget.

None of that means dumbing it down, and it doesn't assume the room is thick. The child test is about explaining the thing better, not explaining less of it. Plain speech is harder to produce than jargon, not easier. Jargon is the path of least resistance, which is exactly why the expert reaches for it. Saying something simply means understanding it well enough to cut everything that isn't load-bearing, then finding words that carry it to someone who hasn't spent years inside the subject. That is real work, and most experts skip it without noticing they've skipped anything.

Why training doesn't fix it

Communication training makes people more confident and better at delivery. It doesn't touch the bias, because the bias isn't a delivery problem.

The expert genuinely can't tell which parts of their explanation need justifying and which don't. Their own knowledge has erased the memory of not knowing, so there's nothing for a presentation-skills course to grab onto. You can teach someone to stand still and project. You can't teach them to un-know the thing that's making them skip step four.

A tax partner explaining a new relief will often walk straight past the part the client cares about most, whether it applies to their specific structure, because to the partner that qualifying step is so automatic it no longer registers as a step. More polish on the delivery just makes the gap more fluent.

None of this is about partners specifically. It shows up wherever the most senior or most technical person gets handed the microphone on the assumption that seniority produces clarity. It doesn't, not reliably. The longer someone has worked in a specialism, the more time they've had to build exactly the blind spot the curse of knowledge describes.

Why a 90-second script beats a 20-page deck

A short video script imposes a discipline a long deck doesn't. At 90 seconds there's nowhere to hide. An unexplained acronym doesn't get a free pass. There's no dense context slide to disappear behind, and no ten-minute runway before the actual point arrives.

Every sentence has to answer one blunt question: would this land with someone who wasn't in the room for the last five years of context? A deck can absorb an unclear slide, because twenty others sit around it to carry the meaning. A tight script can't absorb an unclear sentence. There's nowhere else for the point to live.

That constraint is worth more than it looks. Writing the script becomes the debugging step for the curse of knowledge, not a deliverable you produce after the thinking is done. By the time an expert has narrated a version that holds up at 90 seconds, they've usually sharpened the longer versions too: the meeting, the proposal, the deck it came from.

So the script isn't a smaller version of the message. It's the exercise that finds where the message was broken.

There's a catch, and it's the one this whole piece has been circling. Writing the plain-spoken script is the hard part. Getting an idea down to its core without losing what makes it true takes the exact skill the curse of knowledge quietly removes.

This is where the pairing does the work. The expert's knowledge on one side, a set of tested storytelling prompts on the other. It's what we built Hyperframe's scripting around: the AI drafts the plain version and the expert corrects it, which is a much smaller job than facing a blank page and trying to un-know fifteen years. The expertise stays with the person who has it. The tool handles the translation the expert was never going to do on their own.

What this looks like on a real sentence

Take a sentence like "the relief operates on a de minimis basis, subject to associated company aggregation." Accurate. Also close to useless to a client hearing it for the first time.

Forced through a 90-second script, the same idea comes out closer to: "this saves you money automatically, unless you're linked to other companies that push your combined size over the limit."

Nothing about the underlying advice changed. What changed is that someone who isn't already an expert can follow it. And in most buying committees, that someone is the person who signs.