Blog/How-to

Brand Voice Audit: How to Measure Your Tone

A brand voice audit measures how your brand actually sounds across every channel against the voice you intend to have — surfacing where your tone drifts so you can fix it. Here's how to run one, with a scoring method, an example scorecard, and real-brand examples.

BA

Brand Audit Editorial

2026-03-11

7 min read
Brand Voice Audit: How to Measure Your Tone

A brand voice audit is a structured review of how your brand actually sounds across every piece of writing you publish — your homepage, emails, product UI, social captions, support replies — measured against the voice you intend to have. The output is simple: a map of where your tone is consistent, where it drifts, and exactly what to fix. Done right, it turns a fuzzy feeling ("our copy feels off") into something you can point at and correct.

Most teams have a voice on paper and a different voice in the wild. The paper version lives in a brand guidelines deck nobody opens. The wild version is whatever the last person to touch the copy felt like that day. A brand voice audit closes that gap. Below is how to run one — what to collect, how to score it, and how to read the results — with real examples you already know.

Key idea

Voice isn't what your guidelines say. It's the average of every sentence a customer actually reads. The audit measures the average — not the aspiration.

What a brand voice audit actually measures

Voice breaks down into a few concrete dimensions you can score. The classic framing comes from the Nielsen Norman Group, whose research on the Nielsen Norman Group "tone of voice" framework maps copy along four sliders: funny vs. serious, casual vs. formal, irreverent vs. respectful, and enthusiastic vs. matter-of-fact. You don't have to use those exact four, but you need sliders — not vibes.

Think about how different two well-known brands sound. Mailchimp is warm, plainspoken, a little playful — it will crack a joke in an error message. A private bank like Coutts is the opposite: measured, formal, never winking. Neither is "better." But each is consistent, and that consistency is what you're auditing for. The question is never "is this voice good?" It's "does this sound like us, everywhere?"

This is the close cousin of a broader brand messaging audit, which evaluates what you say. A voice audit is narrower: it evaluates how you say it. The two belong in the same project, but they answer different questions.

The four traits worth scoring

  • Personality — Is there a recognizable character behind the words, or could this be any company?
  • Tone range — Does the voice flex appropriately (lighter in marketing, calmer in support) without losing its core?
  • Vocabulary — Are there words you always use and words you never use? Or does it change page to page?
  • Consistency — The big one. How much does the voice drift between your best-written page and your worst?

How to run a brand voice audit, step by step

Here's the practical sequence. You can do this with a spreadsheet and a free afternoon, or you can let software pull and score the samples for you. Either way, the method is the same.

1

Pull a representative sample

Grab 15–30 real pieces of live copy across channels: homepage, a product page, three emails, ten social posts, your 404 page, a few support replies, app onboarding text. Don't cherry-pick your best work — that defeats the point.

2

Define your target voice on sliders

Before you judge anything, write down where you want to sit on 3–5 traits (e.g. "casual 7/10, playful 4/10, expert 8/10"). Now you have a ruler instead of an opinion.

3

Score each sample against the ruler

Rate every piece on each trait. Flag anything more than two points off-target. Patterns emerge fast — usually support and legal copy drift coldest, social drifts loudest.

4

Find the drift, not the defects

You're hunting for spread. A voice that's a consistent 6/10 casual everywhere beats one that swings from 2 to 9. Inconsistency reads as a brand that doesn't know itself.

5

Write the fix into your guidelines

Translate findings into do/don't examples with real before-and-after lines. Abstract rules ("be friendly") get ignored. Concrete rewrites get copied.

Skip the cherry-pick trap

If you only audit your hero pages, you'll conclude your voice is great. The drift always hides in the places nobody proofreads twice — receipts, error states, auto-replies, the footer.

Reading the results: an example scorecard

Here's what a finished voice audit looks like at a glance. Each row is a channel; the score is how closely that channel's copy matched the target voice. The low scorers are your work list.

/ Voice consistency by channel · sample scorecard

Homepage91%
Marketing emails84%
Social captions68%
Support replies52%
Error & system text41%

The story this scorecard tells is common: the voice is strong where the brand team writes (homepage, marketing) and collapses where other people write (support, system messages). That's not a copywriting problem — it's a brand consistency problem, and it's the single most valuable thing a voice audit surfaces.

The right way vs. the way most teams do it

❌ Common mistake

Judge a few pages by gut feel
"This feels off" with no ruler
Audit only the polished marketing copy
Deliver abstract rules ("sound human")
Do it once, never again

✓ Better approach

Score samples on defined traits
Numbered sliders you set in advance
Sample every channel, especially the boring ones
Deliver before/after rewrites
Re-check after every brand or copy refresh

Where a brand voice audit fits in the bigger picture

A voice audit is one slice of a full brand review. If you're starting cold, it helps to understand what a brand audit is first, then run voice as a focused module inside it. The complete sequence — visual identity, messaging, voice, positioning, perception — is laid out in our step-by-step guide to doing a brand audit, and you can work straight from the brand audit checklist if you'd rather not build the process yourself.

Voice tends to be the most-skipped module because it feels subjective. It isn't — it's just rarely measured. The teams that win on voice (Duolingo, Oatly, Liquid Death) didn't get there by accident. They defined a voice, then audited relentlessly to keep it from sliding back to the mean. The Harvard Business Review's long-running work on brand building, available at Harvard Business Review, repeatedly lands on the same point: distinctiveness compounds only when it's consistent.

15–30Copy samples to pull for a solid audit
3–5Voice traits to score against
2ptOff-target drift worth flagging

Manual audit vs. automated audit

You can run this by hand or have software do the heavy lifting. The trade-off is mostly time and coverage.

DimensionManual auditAutomated audit
SetupSpreadsheet, an afternoonPoint the platform at your site and channels
CoverageWhatever you can read in the time you haveEvery page it can crawl, scored uniformly
Consistency of scoringVaries by who's judgingSame ruler applied to every sample
Best forOne channel, deep nuanceWhole-brand drift across many channels
RepeatabilityTedious to redoRe-run instantly after changes

If you're weighing options, our roundup of the best brand audit tools and our explainer on how automated brand audit tools work both go deeper. The honest answer: manual is fine for a single channel; for whole-brand consistency, automated coverage wins because it scores everything the same way. You can see a sample brand audit to get a feel for the output before committing.

Do this today

Pull your last ten published things — including a support reply and an error message — and read them back to back, out loud. If they don't sound like the same brand, you just found your audit's first finding.

Frequently asked questions

What is a brand voice audit?

It's a structured review that compares how your brand actually sounds across live copy against the voice you intend to have. You collect real samples from every channel, score them on a few defined traits, and identify where the tone drifts so you can fix it. The goal is consistency, not a single "correct" voice.

How do you measure brand tone?

Define 3–5 traits as sliders — for example casual-to-formal, playful-to-serious, expert-to-approachable — and set a target number for each. Then score every copy sample against those numbers. Anything that lands more than about two points off-target is a drift worth flagging. The spread between your highest and lowest scores is your real consistency metric.

How long does a brand voice audit take?

A focused manual audit of one or two channels is an afternoon's work. A full-brand pass across many channels takes longer by hand because you're reading and scoring everything yourself. Software shortens the collection and scoring step considerably by applying the same ruler to every page at once, which is why most teams automate the wide passes and reserve manual review for nuance.

What's the difference between a voice audit and a messaging audit?

A voice audit measures how you say things — tone, personality, vocabulary, consistency. A messaging audit measures what you say — your value proposition, claims, and positioning. They overlap and usually run together, but they answer different questions and produce different fixes.

Run your voice audit the easy way

Scoring every channel by hand is the part everyone dreads — and the part that gets skipped. BrandAudit pulls your live copy and scores voice consistency across your whole brand automatically, so you spend your time fixing the drift instead of hunting for it. See pricing or browse a sample report to see exactly what you'd get.

Tags

brand voice auditbrand voicetone of voicebrand auditbrand consistencymessaging

Apply it to your brand

Ready to audit your brand?

First strategic read in minutes, full report ready to act on. No credit card required.

Start your brand audit →