Skip to content

sidchaudhary/gtm-skills / skills/brand-designer/tone-of-voice

tone-of-voice

Unverified

Analyses real content samples and extracts a brand voice profile as a checkable artifact: tone dimensions each carrying at least one followable rule, required and banned vocabulary, sentence and paragraph patterns, channel adaptations, and a weighted scorecard with a publish-ready threshold to grade drafts against. Use when onboarding a writer or agency, when AI-generated copy is drifting off-brand, or when nobody on the team can say concretely what on-brand means. Boundary: `.agents/product-context.md` carries a short voice summary, while this skill produces the full profile and the scorecard.

FileKindSize
SKILL.mdskill md16.9 KB
references/agent-security.mdmarkdown7.2 KB
references/brand-voice-dimensions.mdmarkdown19.2 KB
references/house-rules.mdmarkdown13.3 KB

references/brand-voice-dimensions.md

View on GitHub
# Brand Voice Analysis


> **Provenance.** The numbers in this file are **pack benchmarks**, not the reader's own data, and
> most carry no named study behind them. When you state one in output, label it inline as a pack
> benchmark so nobody mistakes it for a figure derived from their business. Where a number is doing
> real work in a decision and you cannot name a source for it, write `[NEED: source]` instead of the
> number. House rule 4b covers this and binds every skill that reads this file.

Reference for analyzing and defining brand voice: six core dimensions, vocabulary profiling, sentence patterns, tone modifiers, consistency rules, channel adaptations, anti-patterns, and comparison examples.

---

## Voice Dimensions

Each dimension is rated on a 1-10 scale representing a spectrum between two poles. A brand's voice is the unique combination of positions across all six dimensions.

### 1. Formality: Casual (1) --- Formal (10)

How structured and conventional the language is.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Very Casual | Slang, contractions, sentence fragments, emoji, internet shorthand | "hey, we just dropped something cool" |
| 3-4 | Casual | Contractions, conversational tone, relaxed grammar, first person | "We've been working on something you'll love" |
| 5-6 | Balanced | Professional but approachable, some contractions, clear structure | "We're excited to introduce our latest feature" |
| 7-8 | Formal | Complete sentences, no contractions, structured paragraphs, third person | "The company is pleased to announce the release of its latest innovation" |
| 9-10 | Very Formal | Academic/legal tone, complex sentence structure, passive voice acceptable | "It is with considerable enthusiasm that we present this advancement" |

### 2. Energy: Calm (1) --- Energetic (10)

The intensity and pace of the communication.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Very Calm | Measured, contemplative, quiet confidence, long sentences | "Take a moment to explore what's possible" |
| 3-4 | Calm | Steady, reassuring, unhurried, thoughtful | "We've thoughtfully designed every detail" |
| 5-6 | Balanced | Clear energy without excess, confident statements | "Here's what you can accomplish with this" |
| 7-8 | Energetic | Exclamation points (sparingly), action verbs, momentum | "Get ready to transform your workflow" |
| 9-10 | Very Energetic | Excitement, rapid pace, urgency, bold punctuation | "This changes EVERYTHING. Let's go!" |

### 3. Humor: Serious (1) --- Playful (10)

How much levity and wit appears in the communication.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Very Serious | No humor, facts-only, gravitas | "Data security requires unwavering vigilance" |
| 3-4 | Serious | Occasional dry wit at most, substance-focused | "Your data deserves better protection (and so do you)" |
| 5-6 | Balanced | Light touches of humor, relatable observations | "Because nobody loves spreadsheets (well, almost nobody)" |
| 7-8 | Playful | Regular wordplay, pop culture references, fun metaphors | "Your old CRM? It's giving 'fax machine in 2025'" |
| 9-10 | Very Playful | Humor-led, memes, self-deprecating, entertainment-first | "We made analytics fun. Yes, we're surprised too." |

### 4. Authority: Peer (1) --- Expert (10)

The power dynamic between brand and audience.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Peer | "We're figuring this out together," humble, learning alongside | "We've been exploring this too, here's what we found" |
| 3-4 | Friendly Guide | Helpful without being preachy, suggestive not prescriptive | "You might want to try this approach" |
| 5-6 | Knowledgeable | Confident opinions backed by evidence, balanced perspective | "Our data shows this approach works best" |
| 7-8 | Expert | Definitive statements, research-backed, thought leadership | "The most effective strategy is X. Here's the evidence." |
| 9-10 | Ultimate Authority | Categorical, authoritative, "the last word" on the topic | "This is the definitive guide to customer engagement" |

### 5. Warmth: Distant (1) --- Intimate (10)

The emotional closeness and personal connection in the communication.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Very Distant | Institutional, no personal pronouns, corporate | "The platform provides enterprise-grade capabilities" |
| 3-4 | Professional | Polite but not personal, "you" used generically | "Users can configure settings to match their needs" |
| 5-6 | Warm | "You" and "we" used naturally, empathetic touches | "We understand the challenges you're facing" |
| 7-8 | Personal | Direct address, emotional resonance, relatable | "We built this because we've been in your shoes" |
| 9-10 | Very Intimate | Like a close friend, deeply empathetic, vulnerable | "We get it. We've struggled with this too. Here's what helped us." |

### 6. Complexity: Simple (1) --- Sophisticated (10)

The vocabulary and conceptual depth of the communication.

| Score | Label | Characteristics | Example |
|-------|-------|----------------|---------|
| 1-2 | Very Simple | Short words, short sentences, no jargon, grade-school reading level | "Send better emails. Get more clicks." |
| 3-4 | Simple | Common vocabulary, clear explanations, accessible | "Track how customers interact with your product" |
| 5-6 | Moderate | Some industry terms (explained), varied sentence structure | "Behavioral segmentation lets you target customers based on their actions" |
| 7-8 | Sophisticated | Industry jargon assumed, complex concepts, nuanced arguments | "Leverage RFM-based cohort analysis to optimize lifecycle engagement" |
| 9-10 | Very Sophisticated | Academic/technical language, multi-clause sentences, dense concepts | "The stochastic optimization of multi-armed bandit allocation strategies..." |

---

## Vocabulary Profile

### Preferred Words

Words the brand actively chooses to use. These create consistency and distinctiveness.

| Category | Examples |
|----------|---------|
| Action verbs | "build," "launch," "grow," "unlock," "discover" |
| Descriptors | "smart," "intuitive," "powerful," "seamless" |
| Value words | "insight," "growth," "impact," "results," "clarity" |
| Relationship words | "partner," "team," "together," "community" |

### Banned Words

Words the brand never uses due to tone, association, or brand positioning.

| Category | Words to Avoid | Reason |
|----------|---------------|--------|
| Overused tech | "synergy," "paradigm," "disruptive," "scalable" (unless technical context) | Cliche, meaningless |
| Aggressive | "crush," "dominate," "destroy," "kill" | Too aggressive for most brands |
| Vague | "solutions," "best-in-class," "world-class," "cutting-edge" | Empty buzzwords |
| Hedging | "maybe," "possibly," "we think," "sort of" | Undermines confidence |
| Competitor names | Naming competitors directly in marketing copy | Gives them attention |

### Jargon Policy

| Stance | Description |
|--------|-------------|
| **No jargon** | Explain everything in plain language. For consumer brands. |
| **Jargon with explanation** | Use technical terms but define them on first use. For mixed audiences. |
| **Jargon assumed** | Use industry terms freely. Audience is experts. For technical/B2B brands. |

### Acronym Handling

| Stance | Description |
|--------|-------------|
| **Spell out always** | Always write the full term, never abbreviate |
| **Spell out first, then acronym** | "Customer Relationship Management (CRM)" on first use, "CRM" thereafter |
| **Acronym only** | Use established acronyms without expansion (assumes audience familiarity) |

---

## Sentence Patterns

### Average Sentence Length

| Category | Words per Sentence | Effect |
|----------|--------------------|--------|
| **Short** | < 12 words | Punchy, scannable, urgent. Good for CTAs and headlines. |
| **Medium** | 12-20 words | Balanced, clear, conversational. Good for body copy. |
| **Long** | 20+ words | Detailed, nuanced, explanatory. Good for thought leadership and documentation. |

**Best practice:** Vary sentence length. Short sentences create impact. Medium sentences carry the narrative. Occasional long sentences add depth and nuance. The rhythm of varying lengths keeps readers engaged.

### Active vs. Passive Ratio

| Ratio | Description | Example |
|-------|-------------|---------|
| **90%+ active** | Almost entirely active voice. Direct, clear, engaging. | "We analyzed 10,000 campaigns" |
| **70-90% active** | Mostly active with occasional passive for emphasis or variety | "10,000 campaigns were analyzed" (passive for emphasis on scale) |
| **50-70% active** | Balanced. Common in academic or formal writing. | Mix of both |
| **< 50% active** | Predominantly passive. Feels institutional or bureaucratic. | Avoid for marketing copy |

### Question Frequency

| Frequency | Usage | Effect |
|-----------|-------|--------|
| **Rare** (< 1 per 500 words) | Questions only for dramatic effect | Declarative, authoritative |
| **Moderate** (1-2 per 300 words) | Questions to engage and transition | Conversational, engaging |
| **Frequent** (1+ per 200 words) | Questions as a primary rhetorical device | Interactive, Socratic, can feel uncertain if overdone |

---

## Tone Modifiers by Context

The base voice stays constant, but tone shifts across contexts.

### Marketing Tone (Aspirational)

- Emphasize vision and possibility
- Future-oriented language ("imagine," "what if," "when you")
- Emotional resonance over technical detail
- Benefits over features
- Confidence without arrogance

### Support Tone (Empathetic)

- Lead with acknowledgment ("I understand this is frustrating")
- Clear, step-by-step guidance
- Warm but efficient
- No blame language ("instead of X, try Y" not "you did X wrong")
- Patient and reassuring

### Sales Tone (Confident)

- Direct value statements
- Evidence-backed claims (stats, case studies, ROI)
- Urgency when appropriate, never desperation
- Consultative, not pushy
- Mirror the prospect's language and formality

### Product Tone (Precise)

- Accurate, specific language
- Technical terms used correctly and consistently
- No marketing superlatives ("blazingly fast" becomes "50ms p99 latency")
- Clear information hierarchy
- Task-oriented, helping users accomplish goals

---

## Voice Consistency Rules

### What Stays Constant (Across All Channels)

| Element | Description |
|---------|-------------|
| Core personality traits | The fundamental character of the brand (e.g., always curious, always honest) |
| Values expression | How brand values manifest in language (e.g., transparency = specific metrics, not vague claims) |
| Vocabulary preferences | Preferred and banned word lists apply everywhere |
| Point of view | First person plural ("we") vs. third person ("the company") |
| Ethical guardrails | Never misleading, never condescending, always inclusive |

### What Flexes (By Channel/Context)

| Element | How It Flexes |
|---------|--------------|
| Formality level | +/- 2 points from base by channel (more casual on social, more formal in contracts) |
| Sentence length | Shorter on social, longer in blog posts, moderate in email |
| Humor level | More playful on social, less in support tickets |
| Technical depth | More sophisticated in docs, simpler in marketing |
| Urgency | Higher in sales and promotions, lower in thought leadership |

---

## Channel Adaptations

### Website (Full Voice)

The canonical expression of brand voice. All dimensions at their defined levels.
- Long-form content with full depth
- Complete sentences and paragraphs
- All brand vocabulary and messaging frameworks applied

### Social Media (Casual Variant)

Formality: -1 to -2 from base. Energy: +1. Humor: +1.
- Shorter sentences and fragments acceptable
- More colloquial language
- Emoji allowed (if brand permits)
- Questions and engagement hooks
- Platform-specific conventions (hashtags, mentions, threads)

### Email (Warm Variant)

Warmth: +1 to +2 from base. Formality: same or -1.
- Personal address ("Hi {{first_name}}")
- Conversational tone within professional bounds
- Clear structure with scannable formatting
- Direct CTAs

### Documentation (Precise Variant)

Complexity: +1 to +2. Humor: -2. Warmth: -1.
- Technical accuracy is paramount
- Consistent terminology throughout
- Step-by-step structure
- Code examples and specific instructions
- No marketing language

---

## Anti-Patterns

### Cliches to Avoid

| Cliche | Why It Fails | Better Alternative |
|--------|-------------|-------------------|
| "Take it to the next level" | Meaningless, overused | State the specific improvement |
| "Synergy" | Corporate jargon, vague | Describe the specific collaboration benefit |
| "Leverage" (as verb) | Overused in B2B | "Use," "apply," "build on" |
| "Game-changer" | Hyperbolic, overused | Describe the specific impact |
| "Unlock" (overused) | Fine occasionally, tired if repeated | "Access," "enable," "gain" |
| "Seamless" | Every product claims this | Describe what specifically works well |
| "Robust" | Meaningless qualifier | Describe the specific strength |
| "Best-in-class" | Unsubstantiated claim | Provide specific evidence of superiority |
| "End-to-end" | Vague scope claim | Describe the specific scope |
| "At the end of the day" | Filler phrase | Remove entirely |

### Competitor Phrases

Avoid using language so closely associated with a competitor that readers will think of them.

- Do not echo a competitor's tagline or signature phrases
- Do not position against a competitor by name in marketing copy (permissible in sales collateral)
- If a term is "owned" by a competitor, find your own language

### Overused Qualifiers

| Word | Problem | Fix |
|------|---------|-----|
| "Very" | Weakens the word it modifies | Remove or use a stronger word ("very fast" → "blazing") |
| "Really" | Same as "very" | Remove or use a stronger word |
| "Just" | Minimizing or hedging | Remove unless it means "only" or "fair" |
| "Actually" | Implies surprise at something obvious | Remove or restructure the sentence |
| "Basically" | Condescending or filler | Remove |
| "Simply" | Dismissive of complexity | Remove, or use "straightforward" if needed |
| "Literally" | Misused as emphasis | Use only when distinguishing from figurative |

---

## Voice Comparison Examples

The same message written at different dimension levels to illustrate the spectrum.

### Message: "Our new feature helps teams collaborate better"

**Casual + Energetic + Playful (2, 8, 8):**
> "New collab features just dropped and your team is gonna love this. Work together like never before. Let's gooo!"

**Balanced + Balanced + Balanced (5, 5, 5):**
> "We've launched new collaboration features designed to help your team work more effectively together. Here's what's changed."

**Formal + Calm + Serious (8, 2, 2):**
> "We are pleased to announce the release of enhanced collaboration capabilities, designed to improve cross-functional coordination and team productivity."

### Message: "Fix a common data integration problem"

**Simple + Warm + Peer (2, 8, 2):**
> "Data sync issues? We totally get it, we've dealt with the same headache. Here's the fix that worked for us."

**Moderate + Professional + Expert (5, 5, 8):**
> "Data integration failures are the #1 cause of pipeline disruption. Based on our analysis of 5,000 implementations, here's the proven resolution."

**Sophisticated + Distant + Authority (9, 2, 9):**
> "Deterministic resolution of event-driven data synchronization failures requires systematic identification of schema drift, temporal misalignment, and idempotency violations within the integration pipeline."

### Message: "Invite someone to a product demo"

**Casual + Energetic + Playful (3, 8, 7):**
> "Want to see what all the fuss is about? Grab a spot in our live demo, it's 30 minutes of 'aha moments' with zero sales pitch. Promise."

**Balanced + Balanced + Balanced (5, 5, 5):**
> "Join us for a 30-minute product demo to see how Intempt can help your team drive engagement. We'll walk through the key features and answer your questions."

**Formal + Calm + Serious (8, 3, 2):**
> "We cordially invite you to attend a comprehensive product demonstration, during which our team will present the platform's capabilities and address your specific requirements."

---

## Why Voice Profiles Fail, and What to Ship Instead

Consistent brand presentation across channels is associated with **23-33% higher revenue**, so the value
of getting voice right is not in dispute. What fails is almost never the analysis. It is what the
analysis becomes.

**Voice guidelines fail on execution and governance, not on content.** The recurring pattern: the
guideline is written like a marketing artifact and used like a fire drill. The people who need it most
receive a PDF after the fact and are told to "stay on brand", which is not an instruction anyone can
follow. Brand inconsistency then shows up wherever a decision was never actually defined, and the team
falls back on asking for approval instead of applying a rule.

Two consequences worth designing around:

- **A profile that only describes is not usable.** "Warm but direct" cannot be applied or disputed. A
  rule can: "contractions yes, exclamation marks no, never open with a question."
- **Verbal guidelines evaporate with departures.** The voice lives in whoever internalised it, and it
  leaves when they do. Only a written, checkable artifact survives a handover.

### One more reason this matters now

Teams using AI content tools report a specific asymmetry: **visual brand consistency holds up, verbal
consistency collapses.** Logos, colour and type are enforced by design systems and templates. Voice has
no equivalent enforcement layer, so generated copy drifts toward a generic register while everything
around it stays on brand. The drift is easy to miss precisely because the page still looks right.

That makes the checkable artifact the point of the exercise, not a nice-to-have.

### Ship a scorecard, not a description

The output that survives contact with a team is a weighted scorecard someone else can apply to a draft
without asking a question. Five dimensions:

| Dimension | What is checked |
|---|---|
| Vocabulary compliance | Required words present, banned words absent |
| Tone alignment | Register matches the profile on the stated axes |
| Structural patterns | Sentence length, paragraph shape, opener and closer habits |
| Readability consistency | Complexity in the brand's actual range, not just "clear" |
| Identity compliance | How the brand refers to itself, the customer, and the problem |

Score a draft against them and treat roughly **85%+ as publish-ready**. The threshold matters less than
having one, because a threshold turns "does this sound like us" from an opinion into a check.

### And what an audit should actually produce

Three artifacts, not a report:

1. **A pattern summary**, the strengths and failures that repeat across channels.
2. **A gap list**, the specific places where current language contradicts the intended voice.
3. **A north-star statement**, how the brand sounds at its best, in one paragraph, drawn from real
   sentences rather than aspiration.

Where recognition itself is in question, a **blind content test** settles it: strip the branding, ask
whether people can pick the brand's copy out of a set. That is the only honest measure of whether a voice
exists yet.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall