Copy-paste prompts for defining brand voice, building tone matrices, rewriting content in brand voice, generating style guides, checking consistency, and training teams. Each prompt includes exact text, expected output, and implementation tips.
Last updated: March 2026 · Reading time: 15 min
A ChatGPT brand voice prompt is a structured instruction that generates a specific brand voice deliverable: voice attribute definitions, tone matrices, content rewrites, style guide sections, or consistency audits across channels.Brand voice consistency increases revenue by up to 23% according to Lucidpress’s brand consistency report (2024). Yet 77% of brands produce off-brand content regularly (Marq, formerly Lucidpress). The gap isn’t about intention. It’s about documentation. Teams that have a written voice guide produce more consistent content. Teams that don’t, drift. These prompts build the documentation.
Analyze the voice and tone of this brand based on these content samples: Sample 1 (website homepage): [paste copy] Sample 2 (blog post): [paste copy] Sample 3 (email newsletter): [paste copy] Sample 4 (social media post): [paste copy] Sample 5 (product description): [paste copy] For each sample, identify: 1. Formality level (1-5 scale, 1 = very casual, 5 = very formal) 2. Sentence length average (short/medium/long) 3. Jargon usage (none/moderate/heavy) 4. Emotional tone (clinical/warm/enthusiastic/authoritative) 5. Person used (first/second/third) 6. Active vs. passive voice ratio (estimate %) Then synthesize: What are the 3-4 consistent voice traits across ALL samples? Where does the voice drift or contradict itself between channels? What's the overall "personality" of this brand in 3 adjectives?What it produces: A voice audit that reveals your current voice patterns and inconsistencies. Most brands discover they have 2-3 different voices across channels. The audit shows where you’re consistent and where you need alignment. Pro tip: Include at least 5 content pieces from different channels. A brand that only analyzes its website misses voice drift that happens on social media or in sales emails.
Define our brand voice using 4 core attributes. For each attribute: Brand context: - Company: [name, what you do] - Target audience: [who you serve] - Brand personality: [how you want to be perceived] - Competitors: [2-3 competitors and how they sound] For each of the 4 attributes: 1. The attribute name (e.g., "Direct," "Warm," "Expert," "Witty") 2. Definition in one sentence 3. "This means..." (3 specific behaviors: sentence structure, word choice, content approach) 4. "This doesn't mean..." (3 specific anti-patterns to avoid) 5. Example sentence written in this voice 6. Same message written WITHOUT this attribute (for contrast) 7. Where on the spectrum this falls: [formal ← → casual], [serious ← → playful], [technical ← → simple], [reserved ← → bold] Make attributes specific enough that two different writers would produce similar content using them.What it produces: A 4-attribute voice framework with “this means” and “this doesn’t mean” guardrails. The contrast examples (with and without each attribute) are the most useful training tool for writers because they show the difference in concrete terms. Pro tip: Four attributes is the sweet spot. Three is too few to be distinctive. Five or more is too many for writers to hold in their heads while working. Mailchimp uses 4. Stripe uses 4. Follow the pattern.
Compare our brand voice against 3 competitors and identify differentiation opportunities. Our current voice samples: [paste 2-3 pieces] Competitor 1 [name] voice samples: [paste 2-3 pieces from their website/blog] Competitor 2 [name] voice samples: [paste 2-3 pieces] Competitor 3 [name] voice samples: [paste 2-3 pieces] For each brand (including ours), rate on these dimensions: - Formality (1-5) - Technical depth (1-5) - Warmth (1-5) - Boldness of claims (1-5) - Use of humor (1-5) - Data vs. narrative orientation (1-5) Create a comparison table. Then identify: (1) where all competitors sound the same (voice commodities), (2) where our voice is distinct, (3) 3 voice positioning moves that would differentiate us further without alienating our audience.What it produces: A competitive voice analysis with differentiation recommendations. In markets where every brand sounds the same (looking at you, B2B SaaS), voice is an untapped differentiator. If all competitors are formal and technical, there’s an opening for a brand that’s direct and conversational. Pro tip: Don’t differentiate for the sake of being different. Differentiate in a direction your audience would welcome. A law firm audience wants authority, not comedy. Differentiate on warmth or accessibility, not humor.
Map our brand to a personality archetype and translate that into voice guidelines. Brand information: - What we do: [describe] - Our mission/purpose: [why we exist beyond profit] - Our customers' emotional state when they find us: [frustrated, curious, ambitious, anxious] - What we want customers to feel after interacting with us: [confident, understood, excited, safe] - Brands we admire (even outside our industry): [list 3-5] - Brands we don't want to sound like: [list 2-3] Based on this, identify our primary archetype (from the 12 brand archetypes: Innocent, Sage, Explorer, Outlaw, Magician, Hero, Lover, Jester, Caregiver, Creator, Ruler, Regular Guy). Justify the choice. Then translate the archetype into: 1. Voice characteristics (5 specific traits) 2. Vocabulary preferences (20 words we should use often) 3. Vocabulary restrictions (20 words we should never use) 4. Sentence style guidelines (length, complexity, structure) 5. Content approach (how we open articles, handle objections, close CTAs)What it produces: An archetype-to-voice translation. Brand archetypes provide a psychological framework that gives voice guidelines emotional depth. The vocabulary lists (use/avoid) are immediately actionable for content teams. Pro tip: Most brands are a blend of 2 archetypes (primary + secondary). For example, Sage (expertise) + Caregiver (helpfulness) produces a different voice than Sage + Ruler (authority). Identify both.
Build our brand voice starting from our audience, not our brand. Our primary audience: - Role/title: [describe] - Industry: [describe] - Daily challenges: [list 3-5] - How they talk about their work: [paste examples from LinkedIn, reviews, forums if available] - Content they consume and share: [blogs, podcasts, influencers they follow] - What they respect in communication: [directness? depth? brevity? data?] - What annoys them: [jargon? hype? condescension? vagueness?] Based on this audience profile, reverse-engineer the voice they'd trust and engage with. Create: 1. Voice positioning: "We sound like a [role/relationship] who [description]" (e.g., "a senior colleague who respects your time and gives straight answers") 2. 4 voice attributes derived from audience preferences 3. Writing rules: 5 specific dos and don'ts that serve this audience 4. Test: Write the same message 3 ways (too formal, just right, too casual) so writers can calibrateWhat it produces: An audience-derived voice framework. Starting with the audience’s communication preferences (rather than the brand’s desired identity) produces a voice that resonates rather than one that merely expresses. The “we sound like a…” positioning line is the single most useful calibration tool. It gives writers an instant mental model for every piece of content. Pro tip: Interview 5 customers and ask: “How would you describe the way we communicate? What do you like and dislike about our tone?” Their answers often reveal voice perception gaps you’d never discover internally.
Create a tone matrix for our brand across these channels: Our voice attributes: [list the 3-4 attributes from the voice definition] Channels to map: 1. Website (homepage, product pages) 2. Blog / educational content 3. Email (newsletters, onboarding sequences) 4. Social media (LinkedIn, Instagram, X) 5. Sales collateral (decks, proposals) 6. Support/help documentation 7. Error messages and notifications For each channel, specify: - Tone adjustment (e.g., "more conversational than website, same authority level") - Formality shift (+1, 0, or -1 from baseline) - Humor allowance (none / light / moderate) - Sentence length target (average words per sentence) - Person/POV (we/you/they) - Example: one sentence written in the channel-specific tone Format as a table.What it produces: A comprehensive tone matrix that writers can reference before creating content for any channel. The example sentences are the most useful column because writers learn tone by example, not by description. Pro tip: Post this matrix where your content team can see it daily (Notion page, shared doc, printed on the wall). A voice guide that lives in a forgotten PDF doesn’t improve consistency.
Create a situational tone guide for our brand. For each situation below, define how our tone should shift while maintaining our core voice. Our voice: [paste voice attributes] Situations: 1. Celebrating customer success (case study, social shout-out) 2. Announcing a new feature or product 3. Apologizing for an outage or error 4. Responding to negative feedback publicly 5. Handling a crisis or controversy 6. Onboarding a new customer 7. Following up after a sale 8. Requesting customer feedback 9. Pricing and billing communications 10. End-of-contract or churn situations For each situation: - Emotional register: the primary emotion to convey - Words to use: 5 specific words appropriate for this situation - Words to avoid: 5 words that would be tone-deaf here - Example message: one paragraph written in the appropriate tone - Common mistake: the tone error most brands make in this situationWhat it produces: A 10-scenario tone playbook. The “common mistake” column is preventative. Knowing that most brands over-apologize in outage comms (which signals incompetence) or sound robotic in churn communications (which guarantees the customer leaves) helps your team avoid these patterns. Pro tip: Pressure-test the crisis communication tone with your PR or legal team. The voice guide says “be human and direct,” but legal may require specific language. Find the overlap before a crisis hits.
Create an emotional tone spectrum for our brand showing the range of emotional expression we use. Our voice: [paste voice attributes] Our brand personality: [describe] Map our tone across these emotional dimensions, each on a 1-10 scale: Dimension 1: Serious (1) to Playful (10) — where do we typically land? Where's our range? Dimension 2: Reserved (1) to Enthusiastic (10) — same Dimension 3: Formal (1) to Casual (10) — same Dimension 4: Technical (1) to Simple (10) — same Dimension 5: Objective (1) to Opinionated (10) — same For each dimension: - Our default position (e.g., 4/10) - Our acceptable range (e.g., 3-6/10) - When we move toward each end (specific situations) - Our "never zone" (positions we should never occupy on this spectrum) Include an example sentence at position 3, 5, and 7 on each spectrum so writers can see the gradient.What it produces: A calibrated tone spectrum with examples at multiple positions. The “never zone” is the most valuable part. Knowing that your brand never goes below 3 on playfulness (too corporate) or above 7 (too flippant) gives writers a safe range to work within. Pro tip: Have 3 writers independently place your recent content on these spectrums. If they disagree by more than 2 points on any dimension, your voice isn’t consistent enough yet.
Create tone templates for these 6 common content types: Our voice: [paste voice attributes] Content types: 1. Blog post (educational/thought leadership) 2. Product launch announcement 3. Customer email (nurture sequence) 4. Social media caption (LinkedIn vs. Instagram) 5. Help center article 6. Sales proposal/deck For each content type, provide: - Opening sentence template (the first sentence pattern) - Tone description in one line - Average sentence length - Paragraph length - Use of contractions (always/sometimes/never) - Use of questions (frequently/occasionally/rarely) - Data density (high: stat every paragraph / medium / low) - CTA style (direct/suggestive/embedded) - Example: first paragraph of each content type written in our voiceWhat it produces: Specific writing templates per content type. The first-paragraph examples are particularly useful for writers who struggle with openings. They set the tone for the entire piece. Pro tip: Build these templates into your content management workflow. If your team uses a content brief template, include the relevant tone template in each brief so the writer sees it before they start writing.
Adapt our brand voice for an AI assistant/chatbot that interacts with customers on our website and in our product. Our brand voice: [paste voice attributes] Our audience: [describe] Assistant's purpose: [support, sales, onboarding, or general Q&A] Create: 1. PERSONALITY BRIEF: A 3-sentence personality description for the assistant 2. GREETING MESSAGES: 3 options for different times/contexts 3. RESPONSE PATTERNS: How the assistant should structure answers: - Answer length (target word count) - When to use bullet points vs. prose - How to handle "I don't know" gracefully - How to escalate to a human 4. LANGUAGE RULES: 10 specific dos and don'ts 5. EXAMPLE CONVERSATIONS: 3 sample exchanges (user question → assistant response) covering: - A simple factual question - A complaint or frustration - A complex question requiring escalation 6. BANNED PHRASES for the assistant (things that sound robotic or off-brand) Make the assistant sound like our brand, not like a generic chatbot.What it produces: A chatbot/AI assistant voice specification. With 85% of customer interactions expected to be handled without a human by 2026 (Gartner, 2024 prediction), your AI touchpoints need to sound like your brand. A chatbot with a different personality than your website creates a disjointed customer experience. Pro tip: Include your brand’s banned words list in the AI assistant configuration. If your brand never says “sorry for the inconvenience” (because it sounds corporate), your chatbot shouldn’t say it either.
Rewrite this content in our brand voice. Our voice attributes: 1. [Attribute 1]: [definition, dos, don'ts] 2. [Attribute 2]: [definition, dos, don'ts] 3. [Attribute 3]: [definition, dos, don'ts] 4. [Attribute 4]: [definition, dos, don'ts] Content to rewrite: [paste the content] Rules: - Keep the same structure (headings, paragraphs, lists) - Keep all factual claims and data points unchanged - Change: word choice, sentence structure, tone, and opening lines to match our voice - Do NOT add information. Do NOT remove information. Only change HOW it's said. - Highlight the changes you made by marking modified sentences with [CHANGED] After the rewrite, list the 5 biggest voice changes you made and why each change better aligns with our attributes.What it produces: A voice-aligned rewrite with change tracking. The change list teaches the writer (or future AI sessions) what “in our voice” specifically means in practice. Over time, this builds a pattern library. Pro tip: Rewrite 10 pieces of existing content through this prompt and create a before/after swipe file. This swipe file becomes the most effective voice training tool for new hires.
Here's a piece of content from a competitor [Competitor Name]: [paste their content]. Rewrite this content to cover the same topic but in OUR brand voice. The rewrite should: - Cover the same key points (don't skip important information) - Replace competitor-specific references with our perspective - Apply our voice attributes: [list attributes] - Be better: more specific, more actionable, more honest, or more concise (pick the dimension where we can genuinely outperform) - Add our unique perspective or data where relevant Our voice: [paste attributes] Our expertise/differentiator on this topic: [describe] After the rewrite, note: (1) where our voice differs most from theirs, (2) what we added that they missed, (3) where their content was genuinely better and we should learn from it.What it produces: A competitive content piece rewritten in your voice. The note about where their content is better is important. Honest competitive awareness improves your content. Ignoring competitor strengths creates blind spots. Pro tip: Don’t plagiarize structure. Rewrite the topic, not the article. Your content should have a different angle, different examples, and different takeaways, all delivered in your voice.
Rewrite this [blog post / web page / email] for a different channel while keeping our brand voice consistent. Original content (from [channel]): [paste content] Target channel: [LinkedIn post / email newsletter / Instagram caption / support article / landing page] Adjustments for the target channel: - Length constraint: [character or word limit] - Tone shift: [describe how tone should change for this channel] - Format: [list, prose, conversational, Q&A] - CTA appropriate for this channel: [describe] Our voice attributes (maintain throughout): [list attributes] Our tone matrix position for this channel: [paste the relevant row from tone matrix] After the rewrite, explain what you changed and why for the channel shift, vs. what you kept constant for voice consistency.What it produces: A channel-adapted version of existing content. This is how one blog post becomes a LinkedIn post, an email teaser, and an Instagram caption without sounding like three different brands wrote them. Pro tip: Build a content repurposing workflow using this prompt. Every blog post can spawn 5-7 channel-specific pieces. Voice consistency across those pieces builds brand recognition faster than any single channel.
This content was written by [freelancer / AI / team member who didn't have our voice guide]. It's factually correct but doesn't sound like us. Rewrite it in our brand voice. Original content: [paste] Our voice attributes: [list all 4 attributes with dos/don'ts] Our banned words/phrases: [list] Our preferred terminology: [list industry-specific terms we use vs. generic alternatives] Specific fixes needed: - Replace generic language with our preferred terminology - Adjust sentence length to match our target (average [X] words per sentence) - Change passive voice to active where appropriate - Add or adjust personality elements per our voice - Ensure the opening sentence passes the "would our audience keep reading?" test Track changes: mark every changed sentence with [FIXED] and categorize each fix as: word choice, tone, structure, or terminology.What it produces: An on-brand rewrite with categorized changes. The categorization reveals which voice elements the original writer missed most frequently. If 80% of fixes are “word choice,” your team needs a vocabulary list. If they’re “tone,” they need the emotional spectrum tool. Pro tip: Use this prompt to quality-check every piece of outsourced content before publishing. A 5-minute voice pass through ChatGPT catches off-brand language that would otherwise go live and dilute your voice over time.
Rewrite this executive communication in our brand voice while maintaining the gravitas expected of [role: CEO, VP, Head of Product]. Original message: [paste draft] Author: [name, title] Audience: [customers, employees, investors, press] Context: [product launch, company update, crisis response, thought leadership] Our brand voice: [paste attributes] Additional requirements for executive communications: - Must sound like a real person (not a committee-written corporate statement) - Include one personal perspective or anecdote - Data point within the first 100 words - Clear position taken on the topic (no hedging) - Sign-off that feels natural to this executive's personality Write 2 versions: (1) formal (for investor/press), (2) conversational (for social/blog). Both should be unmistakably in our brand voice.What it produces: Two versions of executive communication calibrated for different audiences. Executive voice is the hardest to get right because it needs to be both personal (this specific person) and brand-aligned (this specific company). The two-version approach lets the executive choose the version that feels more natural to them. Pro tip: Record 20 minutes of the executive speaking extemporaneously about the topic. Transcribe it and use key phrases, cadences, and perspectives in the rewrite. This makes the written version sound like them, not like a ghostwriter’s interpretation of them.
Create a complete brand voice style guide for [brand name]. Brand information: - What we do: [describe in 2-3 sentences] - Target audience: [describe] - Brand personality: [3-4 adjectives] - Mission: [one sentence] - Competitors: [list 3 and how they sound] - Brands we admire: [list 2-3] Generate these sections: 1. VOICE OVERVIEW: Our voice in 3 sentences 2. VOICE ATTRIBUTES: 4 attributes with definitions, dos, don'ts, examples 3. TONE MATRIX: How voice adapts across 6 channels 4. VOCABULARY: 20 preferred words, 20 banned words, industry terminology guide 5. GRAMMAR & STYLE: Contractions policy, Oxford comma, capitalization, number formatting 6. SENTENCE GUIDELINES: Target average length, maximum length, variety requirements 7. PARAGRAPH GUIDELINES: Length, opening patterns, transition approaches 8. CONTENT OPENINGS: How we start blog posts, emails, social posts, landing pages (with templates) 9. CTA VOICE: How we write calls-to-action (with 10 examples) 10. EXAMPLES: 3 before/after rewrites showing generic → on-brand Keep the guide under 3,000 words. It should be reference-quality, not reading material.What it produces: A complete, usable brand voice style guide. The 3,000-word constraint forces specificity over verbosity. Guides over 5,000 words rarely get read. Guides under 3,000 get referenced regularly. Pro tip: Put the vocabulary list (preferred and banned words) on page 1. It’s the section writers reference most frequently. Everything else is context; the word lists are daily tools.
Create a comprehensive writing dos and don'ts list for [brand name] content. Our voice: [paste attributes] Our audience: [describe] Common mistakes we see in our content: [list 3-5 recurring issues] Generate 15 DOs and 15 DON'Ts, organized by category: WORD CHOICE (5 each) - DO: [specific word/phrasing preference] - DON'T: [specific word/phrasing to avoid] SENTENCE STRUCTURE (5 each) - DO: [specific structural preference] - DON'T: [specific structural issue] TONE & PERSONALITY (5 each) - DO: [specific tone behavior] - DON'T: [specific tone mistake] For each DO and DON'T, include a concrete example showing the right and wrong way. The list should be printable as a one-page reference sheet.What it produces: A 30-item checklist writers can reference while writing. The one-page constraint makes it a practical tool rather than a document. Print it, pin it next to the monitor. Consistency improves when the reference is visible. Pro tip: Update this list every quarter. Add new DON’Ts when you see recurring off-brand patterns. Remove DOs that your team has fully internalized. A living list is more useful than a static one.
Create a brand vocabulary dictionary for [brand name] covering how we talk about our industry, products, and audience. Brand context: - Industry: [describe] - Products/services: [list] - Audience: [describe] - Voice attributes: [list] Create three sections: 1. PREFERRED TERMINOLOGY (20-30 entries): Term | Instead of | Why (e.g., "growth engineering firm" | "agency" | "Agency implies service provider. We're positioned as engineers who build systems.") 2. BANNED WORDS (15-20 entries): Word/Phrase | Why Banned | Alternative (e.g., "leverage" | "Corporate jargon that adds nothing" | "use" or "apply") 3. PRODUCT & FEATURE NAMING: Official Name | Acceptable Short Form | Never Call It (e.g., "Organic Growth Engine" | "OGE" | "SEO tool" or "SEO product") This dictionary should be searchable and maintained as a living document.What it produces: A vocabulary reference that prevents the most common brand voice violations: wrong terminology. Consistent terminology is the baseline of brand voice. If your team calls the same product three different names, no amount of tone guidance fixes the resulting confusion. Pro tip: Share this dictionary with freelancers, agencies, and any external partners who write on your behalf. Terminology consistency is the first thing that breaks when multiple writers are involved.
Create a library of content templates for [brand name], each pre-written in our brand voice. Our voice: [paste attributes] Create fill-in-the-blank templates for: 1. BLOG POST INTRO (3 templates for different topics): "[Opening hook]. Here's [what this post covers]. We've [credibility statement]. [Preview of takeaways]." 2. EMAIL SUBJECT LINES (5 templates): Pattern + example for each 3. SOCIAL MEDIA POSTS (LinkedIn + Instagram, 3 templates each): Framework + example 4. PRODUCT DESCRIPTION (2 templates): Short version (50 words) + long version (150 words) 5. CTA PHRASES (10 templates): Button text + supporting line 6. ERROR MESSAGES (5 templates): What went wrong + what to do + tone 7. CUSTOMER REPLY (3 templates): Positive feedback response, negative feedback response, feature request response Each template should be immediately usable by a writer who fills in the [brackets].What it produces: A ready-to-use template library. Templates reduce voice drift because writers start from a branded base rather than a blank page. The error message and customer reply templates are often the most neglected but most seen brand touchpoints. Pro tip: Test every template by having 3 different writers fill in the same template for the same topic. If the outputs sound like they came from the same brand, the template is working. If they sound different, the template needs more specific guidance.
Create a brand voice training exercise for new writers joining [brand name]'s content team. Our voice guide: [paste or summarize key elements] Design a 30-minute exercise with: 1. QUIZ (5 questions, 5 minutes): Show 2 versions of the same message. The writer picks which is on-brand and explains why. 2. REWRITE EXERCISE (10 minutes): Provide 3 off-brand paragraphs. Writer rewrites each in our voice. 3. CREATION EXERCISE (10 minutes): Given a brief (topic, audience, channel), writer produces a first draft of [content type] in our voice. 4. SELF-ASSESSMENT (5 minutes): Checklist the writer uses to evaluate their own work against our voice attributes. Include answer keys for the quiz and model answers for the rewrite exercise. Scoring rubric: how to evaluate voice accuracy on a 1-5 scale across our 4 voice attributes.What it produces: A structured onboarding exercise for content team members. Voice training that uses exercises (not just reading a guide) produces 3x faster voice adoption. The quiz builds recognition skills, the rewrite builds editing skills, and the creation exercise builds writing skills. Pro tip: Run this exercise during week 1 of any new writer’s onboarding. Then review their first 3 published pieces against the rubric and provide feedback. Early correction prevents voice drift from becoming a habit.
Audit this content for brand voice consistency. Our voice attributes: 1. [Attribute 1 with definition] 2. [Attribute 2 with definition] 3. [Attribute 3 with definition] 4. [Attribute 4 with definition] Our banned words: [list] Our preferred terminology: [list key terms] Content to audit: [paste] Evaluate: 1. OVERALL VOICE SCORE: 1-10 (10 = perfectly on-brand) 2. ATTRIBUTE SCORES: Score each attribute 1-10 individually 3. VIOLATIONS: List every instance of: - Banned word usage (with line reference) - Wrong terminology - Tone drift (sections that feel different from the rest) - Passive voice overuse - Sentences over [X] words 4. POSITIVE EXAMPLES: 3 sentences that are excellent examples of our voice 5. RECOMMENDED FIXES: Specific rewrites for the top 5 issuesWhat it produces: A scored voice audit with specific line-level fixes. The attribute-level scoring reveals which voice elements drift most. If “Direct” consistently scores 8/10 but “Warm” scores 4/10, the team knows where to focus improvement. Pro tip: Run this audit on content from the past 6 months to establish a baseline. Track scores over time to measure whether voice consistency is improving. Treat it like a KPI.
You are our brand voice checker. Review this content before publication. Voice guide summary: [paste the one-page version] Content type: [blog post / email / social / landing page] Target channel: [where this will be published] Content to review: [paste] Check for: 1. Does the opening match our pattern for this content type? (yes/no + fix if no) 2. Are there any banned words? (list them with replacements) 3. Is the tone appropriate for this channel per our tone matrix? (yes/no + adjust if no) 4. Average sentence length: is it within our target range? (calculate + flag outliers) 5. Does it use our preferred terminology correctly? (check each instance) 6. Is the CTA in our voice? (yes/no + rewrite if no) 7. Would this pass as [brand name] content without a logo? (the ultimate test) Provide a PASS / NEEDS REVISION / FAIL verdict with the 3 most important fixes if not PASS.What it produces: A quick pre-publication voice check. The “would this pass without a logo” test is the gold standard. If content is truly in your voice, someone familiar with your brand should recognize it without seeing the brand name. That’s the level of distinctiveness to aim for. Pro tip: Set up a custom GPT with your voice guide built into the system prompt. Then every piece of content can be pasted in for a quick voice check without re-entering the voice guide each time.
Compare these content pieces from different authors on our team and check for voice consistency. Our voice guide: [paste summary] Piece 1 (by [Author A]): [paste] Piece 2 (by [Author B]): [paste] Piece 3 (by [Author C]): [paste] For each piece, score on our 4 voice attributes (1-10 each). Then analyze: 1. CONSISTENCY SCORE: How similar are the 3 pieces in voice? (1-10, 10 = identical) 2. DIVERGENCE POINTS: Where do the authors differ most? (specific examples from each piece) 3. INDIVIDUAL TENDENCIES: Each author's strengths and weaknesses relative to our voice 4. ALIGNMENT RECOMMENDATIONS: Specific feedback for each author to improve consistency 5. TEAM-LEVEL FIX: One training action that would improve consistency across all authorsWhat it produces: A multi-author consistency analysis. Content teams with 3+ writers always have drift. This prompt identifies where the drift happens so you can address it through training, not constant editing. Pro tip: Run this quarterly with content published in the previous quarter. Share results (anonymized if needed) with the team. Making voice consistency visible and measured motivates improvement.
Audit our brand voice consistency across channels by comparing these content samples: Website copy: [paste hero section or key page] Blog post: [paste recent article excerpt] Email: [paste recent email] LinkedIn post: [paste recent post] Instagram caption: [paste recent caption] Support article: [paste help center article] Our voice attributes: [list all 4] Our tone matrix: [paste or summarize expected tone per channel] For each channel: 1. Voice consistency with core attributes (1-10) 2. Tone appropriateness for this channel per our matrix (1-10) 3. Biggest voice violation in this sample 4. What this channel does best (voice strength to preserve) Overall assessment: Does our brand sound like ONE brand across all channels? What's the #1 cross-channel consistency issue?What it produces: A cross-channel voice audit. Most brands are fairly consistent within a channel but inconsistent between channels. The social media team writes differently than the blog team. This audit catches that disconnect. Pro tip: The fix for cross-channel inconsistency is almost always organizational, not editorial. When different teams own different channels without a shared voice guide and regular calibration, drift is inevitable. Centralize voice governance.
Compare our brand voice over time and track its evolution. Content from [6-12 months ago]: Sample 1: [paste] Sample 2: [paste] Content from [current]: Sample 3: [paste] Sample 4: [paste] Our current voice guide: [paste attributes] Analyze: 1. How has our voice changed? (specific shifts in formality, tone, vocabulary, structure) 2. Are the changes intentional improvements or unintentional drift? 3. Which changes improved our voice? (brought us closer to our defined attributes) 4. Which changes degraded our voice? (moved us away from our attributes) 5. Is our voice guide still accurate, or does it need updating to reflect how we actually sound now? Recommend: (1) voice guide updates if needed, (2) corrections if we've drifted off-brand, (3) one voice improvement to prioritize next quarter.What it produces: A voice evolution report. Brand voice should evolve, but it should evolve intentionally. This prompt separates deliberate improvement from unintentional drift. Running it twice a year keeps your voice guide aligned with your actual content. Pro tip: Voice evolution is healthy when it’s driven by audience feedback or strategic repositioning. It’s unhealthy when it’s driven by writer turnover or content team shortcuts. This prompt helps you distinguish between the two.
“Most brand voice guides fail because they describe what the brand WANTS to sound like without showing what it actually sounds like in practice. The guides that work have specific word lists, before/after examples, and channel-specific templates. I’ve never seen a team improve voice consistency from a guide that says ‘be authentic and conversational.’ I’ve seen teams improve from a guide that says ‘average 14 words per sentence, use contractions, and never say leverage, synergy, or at the end of the day.'”
Hardik Shah, Founder of ScaleGrowth.Digital
| Voice Element | Vague (Useless) | Specific (Usable) |
|---|---|---|
| Personality | “Be friendly and professional” | “Sound like a senior colleague who respects your time” |
| Sentence length | “Keep it readable” | “Average 14 words/sentence. Max 25. Vary between 5 and 20.” |
| Vocabulary | “Avoid jargon” | “Never use: leverage, synergy, paradigm. Instead: use, teamwork, model.” |
| Tone | “Be conversational” | “Use contractions. Start 1 in 5 sentences with ‘You.’ End sections with a question.” |
| Data | “Include data when relevant” | “One specific number per 200 words. Name the source and year.” |
Three principles we follow for brand voice work:
40+ prompts for SEO workflows. Use your brand voice guide to ensure every SEO content piece matches your voice. View Prompts →
30+ prompts for landing page copy. Apply your brand voice to headlines, CTAs, and benefit sections for consistent conversion pages. View Prompts →
Every content brief should include voice guidelines. Our template has a dedicated voice section for each piece of content. Get Template →
Not automatically across separate sessions. Each new ChatGPT conversation starts fresh. To maintain consistency, either paste your voice guide at the start of each session or build a custom GPT with your voice guide as system instructions. Custom GPTs are the better long-term option because they embed your voice rules permanently and every conversation inherits them.
Four is the sweet spot. Three attributes are too few to create a distinctive voice. Five or more are too many for writers to remember while working. Each attribute should be specific enough to guide real decisions (word choice, sentence structure, opening patterns) and different enough from the others that removing one would noticeably change the voice.
Voice is constant. It’s your brand’s personality expressed through language: word choice, sentence structure, and perspective. Think of it as WHO you are. Tone is variable. It’s how you adjust your expression based on context: channel, audience, situation. Think of it as your MOOD. You’re always the same person (voice), but your energy changes depending on whether you’re at a funeral or a party (tone).
With these prompts, you can generate a working voice guide in 2-3 hours. The process: analyze existing content (30 min), define attributes (30 min), build the tone matrix (30 min), create vocabulary lists (30 min), and write examples and templates (30-60 min). Refinement happens over the first month of use as writers apply the guide and you adjust based on what works.
Yes, but intentionally. Review your voice guide every 6 months. Update it when your audience shifts, your market position changes, or your brand matures. A startup voice (scrappy, urgent, personal) naturally evolves into a growth-stage voice (confident, polished, authoritative). The change should be documented, not accidental. Unintentional voice evolution is just drift.
Our AI Visibility practice ensures your brand voice appears accurately and consistently across ChatGPT, Google AI Overviews, and Perplexity. Explore AI Visibility Services →