Skip to content

1. Readiness Rubric for “Brand Brain-ability”

Use a 1–5 score per dimension:

  • 1 = Missing / unusable
  • 3 = Partial / needs work
  • 5 = Strong / near-production-ready

1. Brand Fundamentals

1.1 Brand Identity Clarity

Does the PDF clearly define who the brand is and what it stands for?

Check for:

  • Clear positioning (what space it owns)
  • Purpose / mission / vision
  • Values and personality traits
  • Target audience at a conceptual level

Score:

  • 1 – fuzzy platitudes only, no real positioning.
  • 3 – decent story, but fragmented or inconsistent.
  • 5 – sharp, coherent, and internally consistent; easy to summarise in 3–5 bullets.

2. Verbal Identity Readiness

2.1 Tone of Voice Definition

Is there a machine-digestible description of tone and style?

Look for:

  • Tone adjectives (e.g. “optimistic, direct, no jargon”)
  • Positive & negative examples (good/bad copy)
  • Guidance for different situations (support vs marketing vs sales)

Score:

  • 1 – “We’re friendly and professional” and that’s it.
  • 3 – decent tone definition, a couple of examples.
  • 5 – clear tonal dimensions + scenario guidance + examples.

2.2 Rules, Must-Do & Must-Not-Do (Verbal)

Does the document give explicit rules that could become ContentGuideline nodes?

Look for:

  • Explicit “do” guidelines (must include CTA, benefit first, etc.)
  • Explicit “don’t” guidelines (no swearing, no competitor mentions, no medical claims)
  • Constraints around claims, disclaimers, and legal phrases

Score:

  • 1 – mostly inspirational, very few explicit rules.
  • 3 – some explicit rules, but scattered or incomplete.
  • 5 – clear, enumerated do/don’t lists that can translate almost directly into machine rules.

3. Visual & Audio Identity Readiness

3.1 Visual Identity (Metadata-ability)

Is the visual system expressed in a way that’s indexable and referenceable?

Look for:

  • Precise color codes, logo usage rules, minimum sizes
  • Layout guidance, image style (photo vs illustration, mood), iconography
  • Examples of good/bad executions

Score:

  • 1 – mood boards only, no specs.
  • 3 – some specs, some example layouts.
  • 5 – clear specs + examples, easy to turn into VisualIdentityToken and Asset nodes.

3.2 Audio Identity

If relevant, is there anything structured about sound?

Look for:

  • Audio logo / jingle descriptions or references
  • Voice style guidance (pace, tone, vocabulary) for voiceover/TTS
  • Do/don’t for audio (e.g. “no cheesy stock music”, “avoid shouting”)

Score:

  • 1 – no audio at all.
  • 3 – some mention but not systematic.
  • 5 – defined audio cues and rules.

4. Context, Channels & Audiences

4.1 Channel-Specific Guidance

Does the PDF differentiate behavior by channel/context?

Look for:

  • Clear sections for email, social, web, support, ads, etc.
  • Differences in tone, length, formality per channel
  • Any channel-specific rules (e.g. no emojis in transactional emails)

Score:

  • 1 – “same tone everywhere” and nothing else.
  • 3 – some channel notes, but patchy.
  • 5 – well-defined channel guidance that maps naturally to Context nodes.

4.2 Audience / Archetype Readiness

Are audiences expressed in a way that can become Segments / Archetypes?

Look for:

  • Defined segments or personas (who they are, needs, motivations)
  • Differences in messaging or tone per audience
  • Any explicit “we talk differently to X vs Y”

Score:

  • 1 – “our audience: everyone who loves quality”.
  • 3 – some personas, but shallow / vague.
  • 5 – rich, differentiated audience descriptions with messaging implications.

5. Examples & Few-Shot Material

5.1 Canonical Good/Bad Examples

Does the deck include examples we can reuse as BrandExample nodes?

Look for:

  • Marked GOOD examples (headline, email, social, etc.)
  • Marked BAD or “don’t do this” examples
  • Explanation of why each is good/bad

Score:

  • 1 – almost no examples.
  • 3 – some examples but not annotated.
  • 5 – multiple annotated examples per key channel/use case.

6. Structure & Machine-Readiness

6.1 Structure & Extractability

How easy will it be for GPT-5.1 to translate into the Brand Brain schema?

Look for:

  • Clear sectioning (headings, lists)
  • Minimal contradictions
  • Repeated, consistent terminology (e.g. same word for segments/channels)
  • Tables or checklists (gold for extraction)

Score:

  • 1 – dense prose, inconsistent labels, contradictions.
  • 3 – mostly structured but with some ambiguity.
  • 5 – clean, consistent, highly extractable.

6.2 Gaps vs Brand Brain Schema

Are any core Brand Brain elements obviously missing?

Check against your v1 slice:

  • Brand fundamentals
  • Verbal identity
  • Visual/audio identity
  • Rules (mustDo/mustNotDo)
  • Context (channels/regions)
  • Segments/archetypes
  • Examples

Score:

  • 1 – multiple critical dimensions missing.
  • 3 – a few gaps but workable with some inference.
  • 5 – almost all major elements present.

7. Safety, Risk & Governance

7.1 Safety / Compliance Coverage

Are there clear constraints regarding risk / regulated topics?

Look for:

  • Legal / compliance sections (claims, regulated industries)
  • Sensitive topics (health, finance, kids, politics, etc.)
  • Escalation / approval notions

Score:

  • 1 – no explicit risk/compliance guidance.
  • 3 – some, but high-level only.
  • 5 – clear, explicit risk rules ready to become RiskDomain + ContentGuideline.

7.2 Governance & Updateability

Does it describe how guidelines change and who approves them?

Look for:

  • Ownership (brand team, legal, markets)
  • Update cadence or process
  • Versioning hints

Score:

  • 1 – “here’s the book, good luck”.
  • 3 – some hints at ownership.
  • 5 – clear governance, roles, and evolution process.

2. Due Diligence Prompt for GPT-5.1 (Template)

Here’s a prompt you can drop into GPT-5.1 when you upload a brand guidelines PDF.

2.1 System Prompt

You are an expert brand ontologist and AI readiness assessor.

Your task is to evaluate one or more brand guidelines documents for their readiness to be turned into a "Brand Brain" – a machine-readable repository that powers LLM-based tools for copy, design and personalisation.

You may receive:
- A single master brand guideline, or
- Multiple documents (e.g. master brand, sub-brand, channel, region, audio, social playbooks, etc.).

Treat all provided documents as a corpus. Use filenames, covers or obvious titles as each document’s ID.

The Brand Brain (v1) needs to support:
- Brand fundamentals (positioning, values, purpose)
- Verbal identity (tone of voice, style guidelines, explicit do/don't rules)
- Visual identity (logo, typography, colour, imagery, layout) with explicit visual rules
- Audio identity (audio logo, music, voiceover) with explicit audio rules
- Messaging framework (pillars, key messages, reasons-to-believe, claims) with clear rules
- Content guidelines (mustDo / mustNotDo / advisory rules) across modalities
- Contexts (region, channel, audience archetype / segment)
- Canonical examples (good/bad examples for different channels and modalities)
- Clear structure so the content can be mapped into a graph schema and APIs

You must:

1. For **each document**, score the guidelines on a 1–5 scale on each readiness dimension:
   - Verbal identity readiness and verbal rules
   - Visual identity readiness and visual rules
   - Audio identity readiness and audio rules
   - Messaging framework and messaging governance/claims rules

2. For **each document**, explain the evidence for each score with references to specific parts of that document (section titles or short quotes).

3. Then, at an **overall / corpus level**:
   - Provide combined scores (1–5) for each dimension.
   - Identify overlaps, gaps and contradictions between documents (e.g. conflicting tone rules, different logo specs, missing audio guidance in some docs, etc.).
   - Highlight where a master rule exists in one document but is missing or contradicted in another.

4. Identify gaps and ambiguities that would block or weaken Brand Brain implementation, clearly marking whether they are:
   - Global (affecting the whole brand system), or
   - Local to a specific document, region, channel or sub-brand.

5. Suggest concrete actions to improve readiness, by dimension (verbal, visual, audio, messaging, governance, structure), and where possible indicate:
   - Which document should be updated,
   - Whether a new document/module is needed (e.g. “Audio Guidelines”), or
   - Whether rules should be centralised into a master brand guideline.

6. Explicitly extract candidate rules for:
   - Verbal identity (copy)
   - Visual identity (design)
   - Audio identity (sound/voice)
   grouped into mustDo, mustNotDo, and advisory,
   and annotate each rule with:
   - The source document ID
   - The section or page where it came from.

7. Output both:
   - A human-readable summary (with per-document and overall views), and
   - A structured JSON block for downstream automation.

Do not use web search or external browsing. Base your assessment only on the provided brand guidelines documents and your own reasoning.

2.2 User Prompt (with PDF)

When you upload the PDF, use something like:

I am going to paste or attach multiple brand guidelines documents.

They may include, for example:
- Master brand guidelines
- Sub-brand guidelines
- Channel or region playbooks
- Audio/sonic identity guides
- Social or content playbooks

1. Read **all** documents.

2. For each document, treat the filename or cover title as its ID.
   Use that ID when you refer to evidence or rules.

3. Evaluate each document against the following readiness dimensions, scoring 1–5 for each:

   A. Brand Identity Clarity
   B. Verbal Identity Readiness (tone & style definition)
   C. Verbal Rules (explicit must-do / must-not-do / advisory for copy)
   D. Visual Identity Readiness (system clarity & structure)
   E. Visual Rules (explicit must-do / must-not-do / advisory for visual design)
   F. Audio Identity Readiness (system clarity & structure)
   G. Audio Rules (explicit must-do / must-not-do / advisory for audio / voice)
   H. Messaging Framework Readiness (pillars, key messages, RTBs)
   I. Messaging Governance / Claims Rules (what we can/can’t say, disclaimers)
   J. Channel / Context Guidance (regions, channels, situations)
   K. Audience / Archetype Readiness
   L. Canonical Examples (good/bad, annotated, across modalities)
   M. Structure & Extractability (for graph / API)
   N. Safety / Compliance Coverage
   O. Overall Brand Brain Readiness

   Pay particular attention to any explicit or implicit messaging framework:
   - brand/value pillars,
   - key messages,
   - reasons-to-believe (RTBs),
   - claims and disclaimers,
   - and how messages differ by audience, product, or channel.

4. For each dimension **per document**:
   - Give the **score (1–5)**.
   - Explain why you gave that score, citing sections or examples from that specific document.
   - Note any **clear gaps**, contradictions, or vagueness.

5. Then:

   - Produce a report title:  
     `# {Brand Name}: Brand Brain AI Readiness Assessment`

   - Write a short **executive summary** explaining what you have done, why, and the benefits to the client, plus up to **10 bullet points** describing overall readiness and key risks across all documents.

   - Provide a **prioritised gap list**:
     - what we must add or change in the guidelines **before** building the Brand Brain,
     - explicitly covering verbal, visual, audio, messaging, and governance,
     - and stating whether each gap is global or local to one document/region/sub-brand.

   - Provide a list of **immediately extractable elements** that could be turned into:
     - VerbalIdentityToken
     - VisualIdentityToken
     - AudioIdentityToken
     - ContentGuideline (mustDo / mustNotDo / advisory)
     - BrandExample (good/bad examples)
     - Context definitions (channel / region / audience)
     - Messaging elements (pillars, key messages, RTBs, claims/disclaimers)

6. Finally, output a JSON object with this shape:

{
  "meta": {
    "brand_name": "…",
    "brand_id": "…",
    "assessment_date": "YYYY-MM-DD",
    "assessment_version": "1.0",
    "description": "Short description of this Brand Brain AI readiness assessment.",
    "assessor": "Brand Brain Readiness Assessor",
    "source_system": "Manual upload / internal tool / etc.",
    "notes": "Any additional context or notes about this assessment."
  },

  "documents": [
    {
      "id": "string",
      "scores": {
        "brand_identity_clarity": 0,
        "verbal_identity_readiness": 0,
        "verbal_rules": 0,
        "visual_identity_readiness": 0,
        "visual_rules": 0,
        "audio_identity_readiness": 0,
        "audio_rules": 0,
        "messaging_framework_readiness": 0,
        "messaging_governance_rules": 0,
        "channel_context_guidance": 0,
        "audience_archetype_readiness": 0,
        "canonical_examples": 0,
        "structure_extractability": 0,
        "safety_compliance_coverage": 0,
        "overall_brand_brain_readiness": 0
      },
      "findings": {
        "brand_identity_clarity": "…",
        "verbal_identity_readiness": "…",
        "verbal_rules": "…",
        "visual_identity_readiness": "…",
        "visual_rules": "…",
        "audio_identity_readiness": "…",
        "audio_rules": "…",
        "messaging_framework_readiness": "…",
        "messaging_governance_rules": "…",
        "channel_context_guidance": "…",
        "audience_archetype_readiness": "…",
        "canonical_examples": "…",
        "structure_extractability": "…",
        "safety_compliance_coverage": "…",
        "overall_brand_brain_readiness": "…"
      }
    }
  ],

  "overall": {
    "scores": {
      "brand_identity_clarity": 0,
      "verbal_identity_readiness": 0,
      "verbal_rules": 0,
      "visual_identity_readiness": 0,
      "visual_rules": 0,
      "audio_identity_readiness": 0,
      "audio_rules": 0,
      "messaging_framework_readiness": 0,
      "messaging_governance_rules": 0,
      "channel_context_guidance": 0,
      "audience_archetype_readiness": 0,
      "canonical_examples": 0,
      "structure_extractability": 0,
      "safety_compliance_coverage": 0,
      "overall_brand_brain_readiness": 0
    },
    "findings": {
      "brand_identity_clarity": "…",
      "verbal_identity_readiness": "…",
      "verbal_rules": "…",
      "visual_identity_readiness": "…",
      "visual_rules": "…",
      "audio_identity_readiness": "…",
      "audio_rules": "…",
      "messaging_framework_readiness": "…",
      "messaging_governance_rules": "…",
      "channel_context_guidance": "…",
      "audience_archetype_readiness": "…",
      "canonical_examples": "…",
      "structure_extractability": "…",
      "safety_compliance_coverage": "…",
      "overall_brand_brain_readiness": "…"
    }
  },

  "rules": {
    "verbal": {
      "must_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "must_not_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "advisory": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ]
    },
    "visual": {
      "must_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "must_not_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "advisory": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ]
    },
    "audio": {
      "must_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "must_not_do": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "advisory": [
        { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
      ]
    }
  },

  "messaging": {
    "pillars": [
      { "name": "…", "description": "…", "source_document": "…", "source_section": "…" }
    ],
    "key_messages": [
      {
        "name": "…",
        "description": "…",
        "pillar": "…",
        "audience": "…",
        "channel": "…",
        "source_document": "…",
        "source_section": "…"
      }
    ],
    "reasons_to_believe": [
      {
        "message_name": "…",
        "rtb": "…",
        "source_document": "…",
        "source_section": "…"
      }
    ],
    "claims_rules": {
      "allowed_claims": [
        { "description": "…", "conditions": "…", "source_document": "…", "source_section": "…" }
      ],
      "forbidden_claims": [
        { "description": "…", "source_document": "…", "source_section": "…" }
      ],
      "required_disclaimers": [
        { "text": "…", "applies_when": "…", "source_document": "…", "source_section": "…" }
      ]
    }
  },

  "gaps": [
    { "priority": 1, "level": "global", "area": "verbal", "description": "…", "suggested_action": "…" },
    { "priority": 2, "level": "document", "document_id": "…", "area": "visual", "description": "…", "suggested_action": "…" }
  ],

  "extractable_elements": {
    "verbal_identity": [ "…" ],
    "visual_identity": [ "…" ],
    "audio_identity": [ "…" ],
    "content_guidelines": [ "…" ],
    "brand_examples": [ "…" ],
    "contexts": [ "…" ],
    "messaging": [ "…" ]
  }
}

Now I will provide the brand guidelines document. Please wait to respond until you have read the full document.

3. How You’d Actually Use This

Practical workflow:

  1. Upload PDF → run the due-diligence prompt above.
  2. Capture the JSON output and store it alongside the doc.
  3. Use the scores to compare brands / markets.
  4. Use the gaps list to brief the brand team:
  5. “We need explicit ‘do/don’t’ lists for social & email.”
  6. “We’re missing audience-specific guidance; we can’t build Archetypes yet.”
  7. Use extractable_elements as the first cut of data to feed into your Brand Brain ETL.