What's the one rule from a 17th-century science club that, broken, has cost you a contract — and followed, would've saved one?

On November 28, 1660, a knot of men filed out of a lecture hall at Gresham College, a London center for public science lectures, after listening to the resident professor of astronomy talk about the heavens, and instead of going home, they kept talking. They weren't credentialed scientists, there was no such category yet. They were lawyers, merchants, physicians, aristocrats and landowners, bound by nothing but a shared itch to understand how the physical world actually worked. That after-lecture conversation became a weekly habit, the habit became a Fellowship, and within two years King Charles II had granted it a royal charter under a name almost nobody uses anymore: The Royal Society of London for Improving Natural Knowledge. The lecturer that night, Christopher Wren, didn't want a debating society. He wanted an engine that converted knowledge into "profit, and health and the conveniences of life", proof you could touch, not theory you could merely admire.

To make that engine run, the Fellowship assigned roles that read like a modern org chart wearing a powdered wig. Henry Oldenburg became Secretary, managing the correspondence that let the group pull in observations from across Europe. Robert Hooke became Curator of Experiments, whose entire job was to walk into the weekly meeting carrying a demonstration, something that happened, in front of witnesses, that could be repeated. And over the doorway of the whole operation hung the motto that still defines it: Nullius in verba. Take nobody's word for it. Not the King's, not a bishop's, not a famous man's. If you wanted the room's belief, you brought the experiment, not the reputation.

That single discipline, demand the demonstration, refuse the narrative, has cost me a contract, and a different application of the exact same discipline has saved one.

The cost came from skipping it. A prospect describes their business in the first call with total fluency, clean numbers, a clear bottleneck, a confident diagnosis of their own problem. It's tempting to build the proposal on their account, because their account is articulate and they're paying you to listen, not interrogate. But an articulate account is not a verified one. Build a solution on a story instead of on the underlying data, and the gap between what they said and what's actually happening in their books, their pipeline, their team surfaces exactly when it's most expensive, mid-engagement, after the scope is signed, after trust has a crack in it. The deal that fell apart did so because I took the word, not the demonstration.

The save came from the opposite move: insisting on the artifact before the agreement. Pull the actual numbers. Watch the actual process run. Let the claim survive contact with evidence before a dollar moves. It's slower in the room and faster everywhere downstream, because nothing built afterward has to be unwound.

The Society itself shows this isn't a one-time gate, it's a discipline that has to survive its own founders' egos. Isaac Newton sat as President for twenty-seven years and, in that same span, was the center of one of the institution's bitterest feuds, brilliant men competing for credit over who discovered what first. The motto that governed the experiments did not automatically govern the politics. Verification protected the data; it took no effort to protect the room from vanity. That's worth sitting with, because the instinct in any high-trust relationship, a coaching client, a long partner, a founder you respect, is to extend them the same exemption Newton's peers extended him: surely his word is good enough. It rarely is, and the cost shows up later, dressed as a surprise.

But the sharpest lesson sits in what the Society got wrong for nearly three centuries. The first women weren't elected Fellows until 1945, when the crystallographer Kathleen Lonsdale and a handful of others finally broke a 285-year barrier. An institution built entirely on refusing unverified claims spent that entire stretch accepting, completely unexamined, the claim that competence had a gender. Nullius in verba governed the lab bench and never once turned around to examine the gatekeeping.

That's the upgrade worth taking from the whole story: the demand for proof has to point at your own filters, not just at the world's claims. It's not enough to ask "did they demonstrate the result?" You have to ask who you're letting bring a demonstration in the first place, which prospects you take seriously before they've shown anything, which advisors you defer to because of tenure rather than outcome. That's the real test, the one that catches what nullius in verba alone misses: judge the tree by its fruit, not by the credential stapled to the trunk, and don't grant yourself an exemption from the rule you apply to everyone else's claims. The Royal Society proved nature would not lie to a careful observer. It took 285 years to prove the harder version, that the observer has to be just as careful about who gets handed the instrument.