your brand voice doc is fan fiction.
most tone-of-voice decks die in week two. why they don't survive contact with the calendar, and the three-page replacement we now write for every account that crosses 50 posts a month.
the 40-page voice deck is aspirational. the post still has to ship by 4pm.
the 40-page voice deck is a one-time artifact masquerading as an operating manual. we replaced it with three pages: words we use, words we never use, and the substitution table. it survives weekly contact with the calendar. the deck did not.
every agency, ours included, has at some point written a 38-slide brand voice deck. it had persona archetypes. it had a "voice spectrum" with little pegs along it. it had a paragraph called "what we sound like when we're at our best." it was, on the day it shipped, the best deliverable in the building.
six weeks later it was a pdf on a shared drive that nobody opened. the writer who needed it on a tuesday at 3:40pm did not have the time to scroll through 38 slides to find out whether "y'all" was on-brand. they made a call. the post shipped. the deck did not influence it. multiply by 200 posts a year.
— section onethe deck was for the pitch.
here is the uncomfortable thing. the voice deck has a customer, and the customer is not the writer who has to use it. the customer is the cmo who paid for it, the agency partner who shipped it, and the new business deck that includes a slide saying "we developed your voice." it is a sales artifact dressed as an operating one.
a tool that nobody uses on a tuesday is not a tool. it is a souvenir from a kickoff.
we kept making them anyway because clients kept asking for them. we kept making them longer because length felt like rigour. the longer they got, the less they were used. the correlation was nearly perfect.
— section twothe three pages that replaced it.
- 01page one: words we use. a list of about 40 words and phrases. real ones. not adjectives like "warm" — actual words, like "honestly" and "low-key" and "field notes."
- 02page two: words we don't. a list of about 60 banned ones. "elevated." "curated." "world-class." "passionate." "unparalleled." the corporate-poetry shortlist.
- 03page three: the substitution table. two columns. left: the cliché the writer reaches for. right: the brand-version we use instead. 25 rows. updated quarterly.
the entire doc fits on three pages. the writer can read it before lunch. when they get stuck at 3:40pm, they consult one specific table, not a 38-slide narrative. the doc is now a tool, not a thesis.
— section threewhat the deck was hiding.
here's the second uncomfortable thing. most voice decks are written before there are enough posts to write a voice deck from. the agency is interpreting a brand it has barely worked on. the deck is therefore a guess about what the voice should be, dressed up as a description of what it is.
this is unpopular. clients want the voice deck on day twelve. we explain that we can give them one on day twelve, but it will be aspirational, and we will rewrite it in month four anyway. about half agree. of the half that don't, we ship the deck and we still rewrite it in month four. the second one is always better. the first one is always a costume.
— closingthe test for whether your deck is fan fiction.
open your voice doc. find the most flowery slide. now open your last ten posts. does the slide describe them, or does it describe a brand the agency wished you were. if the second one is true — and it usually is — the doc is fan fiction. lovely, well-art-directed fan fiction. and not, in any meaningful sense, your voice.
cut it down to three pages. write it from the actual work. update it every quarter. and stop pretending the deck does the writing.
this essay is part of the dispatch — one piece of writing on social, content, and the operations of running both, sent on the last friday of every month. no roundups, no listicles, no spam. subscribe here →