Document Your Brand Voice in 5 Components

Most brand voice documents are useless within ninety days.

They sit in a Notion page nobody opens, made of adjectives that can't be falsified — friendly, professional, expert, approachable, human. Six months later the marketing team has turned over, three freelancers have written copy under the brief, and the homepage, the cold email sequence, and the LinkedIn feed read like four different companies. The brand manager who wrote the document blames execution. The writers blame the brief. Both are right and neither is the cause.

The cause is that the document never specified anything testable in the first place.

A brand voice that survives team turnover, freelancer cycling, and AI-assisted drafting documents five components, each of which can be checked against a sentence. Not a vibe. Not a personality test. Five components, five tests.

What brand voice actually is — and what most documents get wrong

Brand voice is the pattern of choices a brand makes — and refuses — across every piece of writing it ships. Word choices. Sentence shapes. Frame moves. Stake levels. It is not how the brand "sounds." It is what the brand consistently does and consistently doesn't do.

Most brand voice documents collapse that pattern into adjectives. "Confident but not arrogant. Expert but approachable. Witty but never flippant." The problem with adjectives is that two competent writers reading the same brief will write copy that looks nothing like each other — and both will be defensible. Adjectives produce paragraphs that pass review and bear no family resemblance to anything else the brand has shipped.

The frameworks behind the 5-component approach below are grounded in 860 peer-reviewed papers on personality, persuasion, and consumer behavior. The fix isn't more adjectives. It's moving the document from adjectives to behavioral specifications.

The 5 components

A brand voice document earns its place if it specifies these five components at a level where a sentence can be tested against them.

1. Stance

What does the brand believe about the category that nobody else is saying? Stance is not the value proposition. It is the contrarian position — the place where the brand disagrees with the prevailing wisdom of its market.

A brand voice without stance is decorative. A brand voice with stance refuses to write certain sentences because they would betray what the brand actually thinks.

Example specification: "Most B2B copy fails from personality mismatch, not poor writing. We refuse copy that diagnoses the problem as a writing-quality issue when the data shows it's a targeting issue."

Test against a sentence: Does this sentence carry the stance? If a competitor could publish it without changing a word, the stance isn't in the sentence.

2. Personality

The OCEAN profile of the brand voice — not the customer's. Where on Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism does the brand sit, and what do those positions imply for the copy?

A High-Conscientiousness brand voice will not write "transform your marketing overnight." A High-Openness brand voice will not write the same six-bullet listicle template every week. The personality specification predicts what the brand will and won't ship before any specific copy exists.

Test against a sentence: Could this sentence have come from a brand sitting two standard deviations away on a key trait? If yes, the personality didn't carry through.

3. Lexicon

The vocabulary the brand uses and the vocabulary it refuses, with examples in both columns and the reason for each choice. Not "avoid jargon." Specific phrases the brand has decided are in, and specific phrases the brand has decided are out, with the reason behind each.

Lexicon is the component that survives team turnover best because it's the most mechanical to enforce. New writers can read a 50-line lexicon list and produce copy that recognizably belongs to the brand within hours.

Example specification (note the reasoning column — that's the part most lexicon lists skip):

Test against a sentence: Does any word in this sentence violate the don't-use list? Does the sentence reach for a banned phrase to fill a slot? Can the writer name the reason the choice was made?

4. Cadence

The rhythm of the prose. Sentence-length distribution, paragraph density, the ratio of declarative to compound sentences, the placement of short punchy lines.

Cadence is the component most often skipped because it feels intangible. It isn't. It can be measured: average sentence length, distribution variance, paragraph length distribution, the position of the shortest sentence in each paragraph. Two pieces of copy with the same lexicon but different cadence read as two different brands.

Example specification: Average sentence length 14–18 words. Maximum 25. Each paragraph contains at least one sentence under 10 words. Paragraph length 2–4 sentences. Lists used for parallel items, not for breaking up dense prose.

Test against a sentence: Does this paragraph match the cadence signature? Run a word count on each sentence and compare against the spec.

5. Stakes

The consequence the brand exposes the reader to in every piece of writing. The thing the reader is being asked to weigh, lose, or risk if they don't pay attention to what comes next.

Most copy is stakeless. It describes features, promises outcomes, and asks the reader to trust the assertion. Copy with stakes specifies what's actually at risk: the loss, the cost of inaction, the gap between current state and the alternative. Stakes are not urgency. Stakes are the seriousness of the consequence.

Example specification: Every piece of writing surfaces what's at risk for the operator who ignores the insight. Examples: "Six months of growth that doesn't convert." "A cold email list that responds at half the rate it should." "A positioning that drifts because the team can't tell when it's drifted."

Test against a sentence: What's at stake here for the reader? If the answer is "nothing — they just learn a thing," the stakes are missing.

The 5-component voice test in one table

Before and after, on the same brief

The brief: write 90 words on why our analytics tool helps B2B marketers.

Brand voice document — adjectives version
"We are confident, approachable, expert, and human."

Copy generated under it:
"We help marketers create amazing content that connects with customers and drives results. Our analytics tool gives you the insights you need to make data-driven decisions and stay ahead of the competition. Join thousands of marketers who trust us with their content strategy."

That copy is defensible against every adjective. It is also indistinguishable from twelve hundred competitors.

Brand voice document — 5-component version
Stance: B2B copy fails from personality mismatch, not poor writing.
Personality: High-C, moderate-O — analytical, framework-driven, occasionally dry.
Lexicon: Use measurement, audit, score, coverage gap, operator. Don't use leverage, synergy, transform, world-class, AI-powered.
Cadence: Average sentence 14–18 words. Each paragraph has one short punchy line.
Stakes: Show what's lost when copy doesn't fit the audience.

Copy generated under it:
"Most B2B copy reaches two of five personality dimensions on your buyer list. That's why response rates underperform whatever your CRM benchmark says they should. The fix isn't better writing. It's measurement: which dimensions your current copy covers, which it misses, and what each gap is costing you. We score copy against the personality of who you're trying to reach. The audit catches the gap before you ship."

Same brief. Different document. Different copy. The first paragraph can survive a freelancer hand-off. The second one already has.

What to do this week

If your brand voice document is more than three pages of adjectives, the highest-leverage move is rewriting it in the 5-component shape above. Keep it short. Make every line testable. Then run two existing pieces of copy through the document and check whether a writer who has never met the brand could produce the same output.

If most lines fail the test, the document is still aspirational. Keep cutting until each line specifies something a sentence can be checked against.

Test the framework on one piece of copy

Start small. The Email Subject Line Analyzer scores any subject line against a personality target you specify. Thirty seconds, no login. If the output tells you something you didn't already know about a piece you thought was on-brand, the 5-component shape is worth investing in. Run a longer piece through the Ad Copy Analyzer to see what the audit catches on a paragraph rather than a single line.

The Brand Voice pillar walks through the OCEAN-to-brand-voice mapping in detail — particularly how to derive Component #2 (Personality) from your audience-fit profile rather than guessing at it.

You're measuring brand-voice fit, not arguing about whether the writing "feels right."