Most of what you know about an important counterparty is real, useful, and has never been written down — which means it leaves when you do, and has to be re-derived by everyone else.
Writing a brief fails, because you do not know what to say — the knowledge is associative and it surfaces in fragments, over more than one sitting. Talk about how they decide, what they have historically rewarded, what they are protecting, and where you have misjudged them before.
Not a personality type. Useful shapes are: how trust is actually built with this person; what story they are currently telling themselves about the situation; what a win would sound like in their own words; and what all of that implies for how an agreement should be framed.
Read it back. The only passing grade is “yes, that is exactly it.” Anything less means it is plausible rather than true, and plausible is worse than nothing because it will be trusted.
From then on, every communication with that person is checked against the profile instead of being reasoned out from scratch. That is the whole return: the thinking happens once.
What you know about someone who matters is genuine knowledge — accumulated over years, usually accurate, and almost never recorded. It fails to transfer for a mundane reason: nobody sits down and writes it, because a blank page asks a question you cannot answer directly.
Talking works where writing fails, because recall is associative: one fragment pulls the next. The structuring step turns fragments into something re-readable, and the recognition test is the quality gate — you cannot fake “that's exactly it”, which is precisely why it is the test.
Worked exampleThe hardest possible subject: a founder, by his own son, inside the family business.
No brief was ever written. The knowledge came out in fragments across several sessions — how he decides, what he has always protected, what he would never say out loud but clearly believes. Those fragments were structured into four named parts: how trust is built with him, the story he is currently telling himself, a draft of what a win would sound like in his own words, and what that implies for how any agreement should be framed.
He recognised it as accurate. Every communication with him has been governed by it since — the language it is written in, the format, what belongs in it and what does not — without anyone re-deriving those judgments. The thinking happened once.
Anti-patterns