A register is not a brand book

Brand books tell you whether the headline is sentence case. They do not tell you whether “independent security review in progress” is true, who may say it, in which market, and what happens if the engagement letter is withdrawn. A claims register is closer to a blotter than to a style guide. It is a list of statements the house is prepared to defend.

Defence, here, does not mean litigation theatre. It means: can we put our finger on the artefact? Can we name the approver? Can we say, without improvising, which audiences are in and which are out? Can we take the sentence down in twenty-four hours if the artefact moves? If the answers are no, the sentence is not a claim. It is a hope.

Founders resist this for a reason that is not vanity. A register feels like it will slow the work. In the first week it does. The team discovers that half the homepage is unsourced, that “audited” was a conversation with a firm that has not been engaged, that “global community” is a substitute for naming a market, that the litepaper still contains a return illustration someone meant to delete. That discovery is the point. It is cheaper on day two than on day forty.

There is a second resistance, quieter: the register makes it obvious who decides. Many token teams run on consensus-by-chat. A claim with a named approver ends that. The founder, counsel, or a designated controls lead signs the line. If they will not sign it, it does not ship. This is not bureaucracy. It is how you stop a well-liked contractor from becoming the unofficial spokesperson of the offering.

Treat the register as the spine and other artefacts become cheaper. The landing page is a view of live claims. The founder briefing is a view of live claims. The creator brief is a view of live claims plus a prohibited list. The weekly report is: what ran, who approved, which claims, which archives. When a journalist asks for the basis of a sentence, you hand them a row, not a vibe.