What Is Brand Voice (and How to Actually Define It)
RedHub AI Editorialupdated September 7, 20267 min read

Jump to a section12
- TL;DR
- What is brand voice?
- Why "we're friendly" isn't a brand voice
- The real test: could a stranger tell?
- How to actually define your brand voice
- Vibe words vs. checkable rules
- Try it: is your voice word checkable?
- Voice stays constant, tone flexes
- Turning rules into a guide people use
- What strong voice looks like in the wild
- The hardest part: keeping AI on-voice
- Where the four deep-dives fit
TL;DR
- What it is: brand voice is the consistent personality behind everything you write — the same "you" whether it's a tweet or a terms page.
- Who it's for: anyone whose writing sounds like ten different people (or a generic AI) instead of one brand — see the Brand Voice Engine.
- How it works: you define it by pulling rules out of writing you already did well, not by picking adjectives out of thin air.
- Bottom line: a voice you can only feel isn't a spec. A voice you can check is.
What is brand voice?
Brand voice is the consistent personality in how a company writes, no matter who's writing or where it shows up. It's not tone — tone changes with the moment, voice stays the same underneath it. Most teams think they have a brand voice because everyone agrees on a vague vibe ("we're friendly and bold"). A real brand voice is specific enough that two different writers, or a human and an AI, produce copy that reads like the same person wrote it. You define it by extracting rules from writing that already worked, not by brainstorming adjectives.
Best for: any team where writing passes through more than one person or one AI tool. The Brand Voice Engine turns your best samples into a portable, checkable spec.
Ask ten people on a team to describe the company's brand voice and you'll get ten different adjectives — "friendly," "bold," "professional," "authentic." Ask them to point at a paragraph and say why it's on-voice, and most go quiet. That gap is the whole problem. Brand voice isn't a list of adjectives everyone nods at. It's a set of rules specific enough that anyone — a new hire, a freelancer, an AI assistant — can follow them without asking you first.
Why "we're friendly" isn't a brand voice
Every brand thinks it's friendly. Every brand thinks it's authentic. Adjectives like these feel true and mean nothing, because they don't tell a writer what to actually do differently. "Friendly" doesn't say whether to use contractions, whether to crack a joke, or how long a sentence should run. It's a mood, not an instruction. A brand voice only becomes usable the moment it turns into something you can point to on the page — a rule a writer (or a model) can actually check a draft against.
The real test: could a stranger tell?
Here's a simple test. Take a paragraph with your company name blacked out. Could someone who knows your brand well still tell it's yours? If the answer is no — if the paragraph could belong to any competitor with the name swapped in — you don't have a voice yet, you have a category-standard tone. A real brand voice survives the blackout test. That's the bar to write toward.
How to actually define your brand voice
Skip the whiteboard session where everyone throws out adjectives. Start from evidence instead — writing you already did that felt unmistakably like you.
- Pull 3-5 pieces of writing that felt unmistakably like your brand — not writing you wish sounded like you, writing that already does.
- Find the patterns across them: sentence length, words you keep using, words you'd never use, how you open a paragraph, how you talk to the reader.
- Turn each pattern into a rule that can be checked, not a mood that can't. "Short sentences, second person, no jargon" beats "conversational."
- Write "we are X, not Y" contrast pairs for anything that could be misread. "Direct, not blunt" tells a writer exactly where the line is.
- Name the filler words that make everything sound the same — and ban them outright. A banned-words list does more work than any adjective ever will.
Vibe words vs. checkable rules
This is the single most useful swap you can make. Every vague adjective has a checkable version hiding behind it — you just have to write it down.
| Vibe word | Why it can't be checked | Checkable rule instead |
|---|---|---|
| "Friendly" | Every brand thinks it's friendly | Uses contractions, second person, no jargon |
| "Professional" | Professional compared to what? | No exclamation points, one idea per sentence, no slang |
| "Bold" | Bold is relative, not absolute | Makes a claim in the first sentence, cuts hedge words like "maybe" and "could" |
| "Authentic" | Unmeasurable by definition | First person, admits limits, no third-person brand-speak |
Try it: is your voice word checkable?
Type an adjective your team uses to describe your voice and see whether it passes the checkable test.
Voice-word checker
Key insight: the fix for almost any vague voice word is the same move — replace the adjective with the specific behavior a writer would need to copy. If you can't name the behavior, the word isn't defined yet, no matter how many times it's used in a meeting.
Voice stays constant, tone flexes
One more distinction matters before you start writing rules down: voice and tone aren't the same thing. Voice is the constant personality; tone is how that personality flexes by situation — a launch announcement and an apology email use different tones but should still sound like the same brand. Getting this wrong is why some teams write one voice guide and then feel like it "doesn't work" for support replies. It's not broken — tone just needs its own layer on top. Tone of voice vs. brand voice goes deep on where that line sits.
Turning rules into a guide people use
Rules on a sticky note don't scale past one person. Once you've got a handful of checkable rules and contrast pairs, the next step is putting them into a document a new hire, a freelancer, or an AI assistant can actually follow without you in the room. How to write a brand voice guide people actually use covers the anatomy of a guide that survives contact with a real deadline.
What strong voice looks like in the wild
Rules are easier to write once you've seen them in action. Brand voice examples walks through five illustrative archetypes — generic patterns, not real companies — so you can see how contrast pairs and banned words show up in real sentences.
The hardest part: keeping AI on-voice
Adjectives fail hardest with AI. Ask any assistant to "write in a friendly, bold voice" and you'll get the same corporate filler every other brand gets, because the model has nothing specific to check itself against. This is the biggest reason brand voice work matters right now — the more writing that touches an AI tool, the more a checkable spec pays off. How to keep AI writing on-brand is the deepest dive on this, and it's the exact problem the Brand Voice Engine was built to solve.
Where the four deep-dives fit
| Deep-dive | What it answers |
|---|---|
| Tone of voice | How voice stays constant while tone flexes by moment |
| Brand voice guide | How to write rules into a doc people actually use |
| Brand voice examples | What strong, checkable voice looks like in real sentences |
| Keep AI on brand | Why AI defaults to generic, and how a voice spec fixes it |
Codify your voice into rules a machine can check
The Brand Voice Engine turns 3-5 writing samples into a portable Voice Spec, then scores any draft against it on a 100-point rubric — flagging every off-brand line with a rewrite.
Get the Brand Voice Engine — $59 →What defining your voice doesn't fix
A voice spec makes your writing sound consistent. It doesn't invent your positioning, decide what to post, or make weak ideas land. If the deeper problem is "we don't know what to write about," pair voice work with a real posting system — the Content Engine for One Person handles the pillars-and-calendar side. Voice is how it sounds; a content system is what gets said and when.
Decision Guide
Define your voice now if: more than one person (or any AI tool) is writing for your brand and the results don't sound consistent.
Skip it for now if: you're a true one-person operation with no plans to hand writing off — though even then, a spec keeps your own AI drafts from drifting.
Best first step: pull 3-5 samples of your best existing writing and find the patterns before you write a single adjective.
Common Questions
What is brand voice, in one sentence?
Brand voice is the consistent personality in how a brand writes, no matter who's writing or where it appears — the part that stays the same underneath a changing tone.
Is brand voice the same as tone of voice?
No. Voice is constant; tone is how that voice flexes by situation. A brand keeps one voice across a hundred different tones.
How do I define my brand voice from scratch?
Pull samples of your best existing writing, find the patterns, and turn each pattern into a checkable rule — not an adjective. Contrast pairs and banned words do the heaviest lifting.
Why do adjectives like "friendly" or "bold" not work?
Because they don't tell a writer what to actually do. Every brand thinks it's friendly. A checkable rule ("uses contractions, second person, no jargon") replaces the guess with an instruction.
Does having a brand voice matter more with AI writing?
Yes. AI defaults to generic filler unless something tells it not to. A specific, checkable voice spec is what keeps AI-drafted copy from sounding like every other AI-drafted copy.
How long should it take to define a brand voice?
An afternoon, if you start from real samples instead of a blank whiteboard. The Voice Extractor prompt in the Brand Voice Engine turns 3-5 samples into a structured spec in one pass.


The gate this post refers to, drawn from the tool’s own logic. See the tool.