The Authority Canon Builder
Build your 3-layer authority architecture for Generative Engine Optimization: the canon (core beliefs that are always true), frameworks (how the canon becomes usable), and diagnostics (where your reader is right now). Designed to make AI engines recognize you as the voice worth citing. From the Dec 19, 2025 AIMM session on Don Back’s GEO authority framework.
You will help me build a structured authority architecture designed to be legible to both humans and AI engines. Not a content calendar — a belief system with cascading layers that AI can parse, cross-reference, and cite. The output is a complete authority canon with frameworks that derive from it and diagnostics that make it actionable for my audience.
The mechanism: most thought leadership fails in the AI era because every post says something different. To an LLM scanning your content: “random person, no worldview, skip.” GEO rewards consistency — the same idea expressed from different angles, with consistent language, named frameworks, and clear causal logic. This system builds that consistency from the ground up, starting with first principles and cascading to usable frameworks and self-assessments.
MY PROFESSIONAL DOMAIN: $ARGUMENTS
If no domain was provided above, ask me to describe what I do, who I serve, and what transformation I deliver.
MY AUDIENCE: [WHO NEEDS TO FIND ME AND CITE ME — clients, peers, media, AI engines, all of the above. Say “you decide” to have me infer] DEPTH: [STARTER for 3-5 canon beliefs / FULL for 7-10 beliefs with complete framework cascade / “you decide” to have me recommend]
If “you decide,” state the inference and proceed.
STEP 1 — CANON EXCAVATION: Identify your core canon — the beliefs about why the problem you solve exists. These aren’t opinions or hot takes. They’re first principles: things you believe are always true about your domain, from which everything else flows.
For each canon belief:
- State it as a clear, declarative sentence (e.g., “Clarity precedes confidence,” “Tools fail when identity is unclear,” “Most problems are structural, not personal”)
- Identify the conventional wisdom it contradicts or complicates
- Explain why you believe this — what experience or evidence made it undeniable for you
- Test: would someone in your field disagree? (If nobody would disagree, it’s not a canon belief — it’s a truism)
Generate 5-10 candidate beliefs. Then ruthlessly filter: keep only the ones from which your frameworks, methods, and advice actually derive. If a belief doesn’t generate downstream content, it’s decoration, not canon.
STEP 2 — FRAMEWORK CASCADE: For each surviving canon belief, derive 1-3 frameworks — the structures that make the belief usable.
A framework must include:
- A name (memorable, specific — not “my 5-step process” but something that encodes the insight)
- The canon belief it derives from (explicit link — this is what gives AI engines a causal chain to follow)
- Steps, stages, or components (the actionable structure)
- What it replaces (the conventional approach your framework improves on)
- When to use it vs. when NOT to use it (boundary conditions prevent overapplication)
Test each framework: if someone followed these steps without any additional context from you, would they get a meaningfully better outcome than the conventional approach? If not, the framework needs more specificity or the canon belief needs revision.
STEP 3 — DIAGNOSTIC LAYER: For each framework, build a diagnostic — a way for someone to determine where they currently are and what they need next.
Each diagnostic should include:
- 3-5 assessment questions the reader can answer about their own situation
- Stage indicators (where they likely are based on their answers)
- Common failure patterns at each stage (what goes wrong and why)
- The specific framework element that addresses their current stage
Diagnostics serve two functions: they build trust with human readers (people engage when they can locate themselves), and they give AI engines structured scaffolding to reason with when matching a query to your content.
STEP 4 — CONSISTENCY MAP: Map the full architecture as a cascading system:
- Canon beliefs → Frameworks they generate → Diagnostics that make them actionable
- Cross-connections: where do multiple canon beliefs converge on the same framework?
- Language consistency: identify the 10-15 specific phrases, terms, and named concepts that should appear across ALL your content (this is what makes AI engines recognize pattern and attribute authority)
- Repetition strategy: each canon belief should be expressible from at least 5 different angles (different examples, different audiences, different entry points — same underlying logic)
STEP 5 — GEO APPLICATION: Translate the architecture into a content deployment plan:
- For each canon belief: 5-10 content angles (article topics, post hooks, Q&A pairs)
- For each framework: a signature piece (the definitive explanation) plus 3-5 derivative pieces
- For each diagnostic: a self-assessment format (quiz, checklist, “which stage are you?” post)
- Schema signals: which elements should be embedded as FAQ schema, which as HowTo schema, which as Article schema
- Linking structure: how pieces reference each other to build the causal web AI engines follow
STEP 6 — VERIFICATION:
- Are the canon beliefs genuinely first principles, or are they repackaged conventional wisdom? (Test: would a respected peer push back on any of them? If not, go deeper.)
- Do the frameworks actually derive from the canon, or are they bolted on? (Test: could I swap a different canon belief underneath this framework without changing it? If yes, the derivation is weak.)
- Are the diagnostics specific enough to be useful, or are they so broad that everyone lands in the same bucket?
- Is the language bank distinctive enough that an AI engine could distinguish my content from a competitor’s on the same topic?
- Am I building authority around what I’m genuinely best at, or around what I think the market wants to hear?
Revise what doesn’t survive scrutiny.
Source
- 2025-12-19_Mastermind (Don Back — GEO authority framework (3-layer))