Fingertips drumming on a bare wooden table, warm side lighting and long shadows Brain Science
AI-generated, Working Theory
Brain Science · the neuroscience of building · ◉ Evergreen

You can't un-know your own product

by · ·4 min·Working Theory

Once you understand your own product, you cannot un-understand it. Introspecting harder about what a new user needs just produces a tooltip written by someone who no longer needs one. The fix is getting the signal from outside your own head.

There’s an old experiment where people are split into tappers and listeners. A tapper picks a song everyone knows and drums the rhythm on the table with a finger. A listener tries to name it. Before they start, the tappers guess how often they’ll be understood, and they land around one in two. The real number is closer to one in forty.

The gap isn’t about rhythm. It’s about what the tapper can’t turn off. As they tap, they hear the whole song, the melody, the words, the horns in the second verse. The listener hears knuckles on wood. The tapper knows too much to imagine not knowing, and the not-knowing is the exact thing they’re trying to model.

Psychologists call this the curse of knowledge, and it is the quiet reason your onboarding is worse than you think. Once you understand your own product, you cannot un-understand it. Your brain has already done the expensive work — it has taken a screen full of separate elements and compressed them into a few meaningful chunks. Where a newcomer sees eleven unlabeled things, you see “the nav, the canvas, and the one button that matters.” You are not looking at your interface. You are looking at your memory of it.

The same screen, two brains What you see nav canvas the button What a first-timer sees ?
Expertise is compression. It also erases your ability to see the thing you built the way a stranger will. Original diagram · Working Theory

This is why the standard fix fails. The standard fix is to sit down and imagine the new user, to introspect your way to empathy. But introspection runs on the same brain that’s cursed. You’ll write the tooltip you would have needed, which is a tooltip written by someone who no longer needs it. You reach for a word like “workspace” or “sync” or “instance” and it feels neutral, obvious, free of jargon, because for you it is. The curse isn’t that you use hard words on purpose. It’s that you’ve lost the ability to feel which of your words are hard.

So the decision this forces is not “write clearer copy.” It’s “stop trusting your own judgment about clarity, and get the signal from outside your head.” A few ways that actually work:

Watch one real first-timer touch the product without helping them, no narration, no “just click there.” The discomfort you feel in the first ninety seconds is your onboarding backlog, ranked by pain. It is worth more than a month of your own re-reading.

Read your first screen and circle every noun. For each one, ask whether a person who has never used the product could point to the thing it names. “Project,” “board,” “run,” “deploy,” “connect” — half of these are borrowed from a mental model the user doesn’t have yet. The curse hides inside nouns.

And write the first-run copy before you’re an expert, if you can, the day you’re still confused is the only day you’ll ever share a brain with your user. Save what confused you. It expires fast.

The tappers weren’t bad at tapping. They were bad at forgetting the song. You are not bad at explaining your product. You just can’t hear the table anymore, so you have to let someone else tell you what they heard.

The science, to look up: the curse of knowledge (Camerer, Loewenstein & Weber, 1989); Elizabeth Newton’s “tappers and listeners” study (Stanford, 1990); chunking and expert perception in memory (Chase & Simon’s chess studies). All well established.

Sources

  • The science, to look up: the curse of knowledge (Camerer, Loewenstein & Weber, 1989)
  • Elizabeth Newton's 'tappers and listeners' study (Stanford, 1990)
  • chunking and expert perception in memory (Chase & Simon's chess studies).

Liked this? Get the next one in Working Theory.

Going weekly in August (it's in beta now). One genuinely interesting read on building, the brain, and the science most people missed.

Subscribe →
Got a reaction, a counter-example, or something I missed? Reply by email — I read everything.
◉ join in

Where have you hit this — in a product you use, or one you're building?

Threads open here soon. For now, the conversation lives two clicks away — discuss on GitHub, or just reply by email. I read and answer everything.