This recipe produces a complete Tone of Voice Guide that defines your brand's voice attributes and positions your tone across four measurable dimensions — formal/casual, serious/playful, respectful/irreverent, and enthusiastic/matter-of-fact. The output includes slider positions for each dimension, do/don't writing examples, a content-type tone mapping matrix, and before/after writing samples — ready for team distribution and AI tool configuration. [src1]
Which path?
├── Solo creator or small team, no existing guidelines
│ └── PATH A: Lean Guide — 1-page tone card + examples
├── Small-to-medium team, some informal guidelines
│ └── PATH B: Standard Guide — full document with all 4 dimensions
├── Large team, high content volume, needs governance
│ └── PATH C: Comprehensive Guide — full guide + decision matrix + training materials
└── Enterprise, multiple sub-brands or regions
└── PATH D: Scalable System — modular guide + per-channel tone profiles + AI tool config
| Path | Tools | Cost | Time | Depth |
|---|---|---|---|---|
| A: Lean Guide | Google Doc | $0 | 2-3 hours | Core voice + 2 dimensions |
| B: Standard Guide | Google Doc + Sheets | $0 | 4-6 hours | All 4 dimensions + examples |
| C: Comprehensive Guide | Notion/Confluence + Sheets | $0 | 6-10 hours | Full guide + matrix + training |
| D: Scalable System | Notion/Confluence + AI tools | $0-5K | 10-20 hours | Modular, multi-brand, AI-ready |
Duration: 45-60 minutes · Tool: Document editor
Define 3-5 core voice attributes that describe your brand's personality. Voice is the constant — it does not change across content types or situations. [src3]
Voice Attribute Workshop:
──────────────────────────────────────────────────
1. List 5 adjectives that describe how your brand SHOULD sound:
_________, _________, _________, _________, _________
2. List 5 adjectives your brand should NEVER sound like:
_________, _________, _________, _________, _________
3. "We are _________ but not _________" (complete 3-5 pairs):
• We are confident but not arrogant
• We are _________ but not _________
• We are _________ but not _________
4. If the brand were a person, how would colleagues describe them?
_____________________________________________
Example outputs:
Mailchimp: "plainspoken, genuine, has a sense of humor"
Slack: "clear, concise, human — like a friendly, intelligent coworker"
GOV.UK: "direct, informative, authoritative but never bureaucratic"
Verify: Voice attributes pass the "but not" test — each attribute has a clear boundary. · If failed: Interview 3 team members about what makes the brand's communication different from competitors.
Duration: 45-60 minutes · Tool: Document editor + visual slider
Position your brand on each of the four NN/g tone dimensions. Each dimension is a spectrum from 1-10. Most brands do not sit at extremes. [src1]
TONE DIMENSION POSITIONING
══════════════════════════════════════════════════
Dimension 1: FORMAL ←──────────→ CASUAL
Position: ___/10 (1 = very formal, 10 = very casual)
1-3 (Formal): "We regret to inform you that your transaction
could not be processed at this time."
4-6 (Neutral): "Your transaction didn't go through.
Here's what you can do."
7-10 (Casual): "Hmm, that didn't work. Let's try again —
here are your options."
Dimension 2: SERIOUS ←──────────→ PLAYFUL
Position: ___/10 (1 = very serious, 10 = very playful)
Dimension 3: RESPECTFUL ←──────────→ IRREVERENT
Position: ___/10 (1 = very respectful, 10 = very irreverent)
Dimension 4: MATTER-OF-FACT ←──────────→ ENTHUSIASTIC
Position: ___/10 (1 = very matter-of-fact, 10 = very enthusiastic)
Verify: Each dimension has a specific numerical position AND at least one example sentence written in your voice. · If failed: Collect 10 existing content samples and independently rate each on all four dimensions. Average the ratings.
Duration: 60-90 minutes · Tool: Document editor
For each dimension, create 3-5 paired examples showing what your tone sounds like versus what it does not sound like. Use your actual products, features, and scenarios. [src3]
DO/DON'T EXAMPLE FORMAT
══════════════════════════════════════════════════
DIMENSION: [Formal/Casual — Position: 7/10 (casual)]
Context: Welcome email
✓ DO: "Hey [Name]! Welcome aboard. Here's the quick
tour to get you started."
✗ DON'T: "Dear [Name], Thank you for registering your
account. Please find enclosed your onboarding
documentation."
WHY: We're casual and direct. Formal language creates
distance we don't want with new users.
Context: Payment failure notification
✓ DO: "Your payment didn't go through. Update your card
details and we'll try again."
✗ DON'T: "Whoopsie! Your payment took a tumble.
No worries though!"
WHY: Even at position 7/10 casual, payment issues require
clarity over personality.
Verify: Every do/don't pair references a real content type you actually produce. · If failed: If examples feel forced, your dimension positions may be wrong. Return to Step 2.
Duration: 45-60 minutes · Tool: Spreadsheet or table
Different content types require different tone positions. An error message and a marketing email should not sound the same. [src4]
CONTENT-TYPE TONE MAP
══════════════════════════════════════════════════
Baseline tone: F:7 S:6 R:4 E:7
Content Type │ F/C │ S/P │ R/I │ M/E │ Notes
────────────────────┼─────┼─────┼─────┼─────┼──────────────
Marketing landing │ 7 │ 6 │ 4 │ 8 │ Enthusiasm up
Blog post │ 8 │ 6 │ 5 │ 7 │ Most casual
Social media │ 8 │ 7 │ 5 │ 8 │ Most playful
Product UI copy │ 6 │ 4 │ 3 │ 5 │ Clarity first
Onboarding flow │ 7 │ 5 │ 3 │ 7 │ Warm, guiding
Error messages │ 5 │ 2 │ 2 │ 3 │ Serious, clear
Support emails │ 6 │ 3 │ 2 │ 5 │ Empathetic
Legal/compliance │ 2 │ 1 │ 1 │ 2 │ Formal, precise
API documentation │ 3 │ 2 │ 2 │ 3 │ Technical
Press releases │ 3 │ 3 │ 2 │ 5 │ Professional
Sales proposals │ 4 │ 3 │ 2 │ 6 │ Confident
Key: Tone adapts to user emotional state.
• User is happy → enthusiasm OK
• User is frustrated → serious + empathetic
• User is deciding → clear + confident
• User is worried → respectful + precise
Verify: Every content type in the map corresponds to content your team actually produces. · If failed: Write the same message three ways and ask 3 team members which feels right.
Duration: 60-90 minutes · Tool: Document editor
Take 5-10 real pieces of existing content and rewrite them using your new tone guide. This validates the guide and creates training material. [src7]
BEFORE/AFTER REWRITE FORMAT
══════════════════════════════════════════════════
Content type: [Error message]
Tone position: F:5 S:2 R:2 E:3
BEFORE (existing):
"Error 404: The requested resource was not found on this
server. Please verify the URL and try again."
AFTER (new tone):
"We couldn't find that page. It may have been moved or
the link might be outdated. Try searching for what you
need, or head back to the homepage."
WHAT CHANGED:
- Removed technical jargon
- Added helpful next action
- Shifted from passive to active voice
- Kept serious tone — no jokes in error states
Verify: Show before/after pairs to someone unfamiliar with the project. They should identify the tone difference without explanation. · If failed: If rewrites feel indistinguishable, push dimension positions further from center.
Duration: 30-60 minutes · Tool: Document editor
Compile all components into a single, accessible document with sections: introduction, voice attributes, tone dimensions, do's and don'ts, content-type tone map, writing samples, word lists, and quick reference card.
DISTRIBUTION CHECKLIST:
[ ] Share with all content creators
[ ] Present to marketing/product/support leads
[ ] Pin in team communication channels
[ ] Add to onboarding materials
[ ] Configure AI writing tools with tone settings
[ ] Schedule quarterly review (set calendar reminder)
Verify: At least 3 people from different teams have reviewed and confirmed the guide reflects the brand accurately. · If failed: Run a calibration workshop with 10 content samples rated independently.
{
"output_type": "tone_of_voice_guide",
"format": "document",
"sections": [
{"name": "voice_attributes", "type": "object", "description": "3-5 brand personality attributes with but-not boundaries", "required": true},
{"name": "tone_dimensions", "type": "array", "description": "4 dimensions with position (1-10), rationale, and example sentences", "required": true},
{"name": "do_dont_examples", "type": "array", "description": "3-5 do/don't pairs per dimension with context and reasoning", "required": true},
{"name": "content_type_map", "type": "table", "description": "Matrix of content types x 4 dimensions with numerical positions", "required": true},
{"name": "writing_samples", "type": "array", "description": "5-10 before/after rewrites demonstrating the tone in practice", "required": true},
{"name": "word_lists", "type": "object", "description": "Preferred and avoided words with context", "required": false},
{"name": "quick_reference", "type": "document", "description": "One-page summary card for daily use", "required": true}
],
"expected_sections": "7",
"sort_order": "logical flow (voice then dimensions then examples then application)",
"deduplication_key": "voice_attributes"
}
| Quality Metric | Minimum Acceptable | Good | Excellent |
|---|---|---|---|
| Voice attributes defined | 3 attributes | 3-5 with "but not" pairs | 5 with examples + anti-examples |
| Dimension coverage | 2 of 4 dimensions | All 4 with positions | All 4 with positions + do/don'ts |
| Do/don't examples per dimension | 1 pair | 3 pairs | 5+ pairs across content types |
| Content types mapped | 3 types | 6-8 types | 10+ with emotional state notes |
| Before/after samples | 2 rewrites | 5 rewrites | 10+ across all content types |
| Team validation | 1 reviewer | 3 reviewers (same team) | 3+ reviewers (cross-functional) |
If below minimum: The guide is too thin to be useful. Prioritize adding do/don't examples and content-type mapping — these are the sections teams reference daily.
| Error | Likely Cause | Recovery Action |
|---|---|---|
| Team cannot agree on dimension positions | Different mental models of the brand | Audit 10 existing content pieces — rate independently, average scores to find de facto positions |
| Do/don't examples feel artificial | Examples not grounded in real content | Replace generic examples with rewrites of actual content from your website, app, or emails |
| Guide completed but nobody uses it | Distribution failure or guide too long | Create a 1-page quick reference card and integrate into existing workflows |
| Tone sounds inconsistent across channels | Content-type map not specific enough | Split broad categories into sub-rows (e.g., "marketing" into "landing page", "ad copy", "email campaign") |
| AI writing tools ignore the guide | Tone instructions not formatted for AI | Extract key rules as explicit prompt instructions with numerical tone levels |
| New team members produce off-brand content | Onboarding gap | Add guide to onboarding checklist, assign a "tone buddy" for first 2 weeks |
| Component | Free Tier | Paid Tier | At Scale |
|---|---|---|---|
| Document editor | Google Docs ($0) | Notion Team ($8/mo) | Confluence ($6/user/mo) |
| Content audit tools | Manual review ($0) | Grammarly Business ($15/user/mo) | Acrolinx/Writer ($50K+/yr) |
| Stakeholder workshops | Self-facilitated ($0) | External facilitator ($1K-3K) | Brand agency ($5K-15K) |
| AI tool configuration | Manual prompts ($0) | Jasper/Writer ($50-100/mo) | Custom fine-tuning ($2K+) |
| Total for initial guide | $0 | $1K-3K | $10K-20K |
Many guides use "voice" and "tone" interchangeably, producing documents that either change personality per channel (inconsistent) or apply the same rigid tone everywhere (robotic error messages, inappropriately cheerful billing warnings). [src3]
Define voice attributes once. Then map tone adaptations per content type and user emotional state. Mailchimp exemplifies this — their voice is always plainspoken and genuine, but tone shifts from enthusiastic in onboarding to serious in account issues.
A slider showing "formal to casual" at position 7 is meaningless without sentences showing what position 7 actually sounds like. Teams interpret abstract scales differently, producing inconsistent output. [src1]
Each dimension position must include do/don't sentence pairs using your actual product names, features, and scenarios.
Applying the same playful tone to a marketing landing page and a payment failure notification alienates users. Inappropriate humor in error states significantly damages brand perception. [src2]
Error states require seriousness and empathy regardless of overall brand playfulness. Build a content-type tone map that accounts for the user's likely emotional state at each touchpoint.
Use this recipe when a brand needs to create or overhaul its written communication guidelines. It produces a structured, actionable tone guide — not a branding document. The output is designed for daily use by writers, designers, product managers, and AI writing tools. Essential when scaling content production beyond one writer, onboarding new team members, or configuring AI content generation tools to match brand voice.