Exec Meta Eval — SEO / AEO / GEO + Executive Audience Scorer
Evaluates page meta information and returns a structured quality report: a scored scorecard across 9 dimensions, evidence-backed dimension analysis, and a prioritised fix plan. Designed for B2B pages targeting senior buyers and AI-driven search surfaces.
Step 1 — Ingest the Meta Information
Accept input in any of these forms:
- URL — use Bash with
curlto fetch raw HTML (not WebFetch, which strips the<head>). Extract:<title>,<meta name="description">, OG tags (og:title,og:description,og:image), Twitter card tags,<link rel="canonical">, H1 (note if JS-rendered/absent), first visible body paragraph, hero image alt text, and any<script type="application/ld+json">schema blocks - Pasted meta tags or HTML snippet — parse directly from the prompt
- File path (
.html,.md,.txt) — read with the Read tool
Always check character counts for title (target: 50–60) and meta description (target: 145–160).
Display the raw extracted fields before scoring:
Title tag: [value] ([N] chars)
Meta description: [value] ([N] chars)
OG title: [value]
OG description: [value]
OG image: [URL or "not set"]
Twitter card: [type or "not set"]
Canonical URL: [value or "not set"]
H1: [value or "JS-rendered / not found"]
First paragraph: [first 200 chars… or "JS-rendered / not found"]
Schema types: [e.g., Organization, FAQPage — or "none found"]
Schema issues: [e.g., empty logo field, missing required properties — or "none"]
If a field is missing, mark it [not present] — absence is a finding, not a gap.
Step 2 — Score Across 9 Dimensions
Score each dimension 0–10 (integers only). Use the rubrics below.
Calibration anchors: 10 = best-in-class (rare), 7 = genuinely good, 5 = mediocre
but functional, 3 = significant problems, 1 = nearly absent.
See references/scoring-examples.md for annotated examples per dimension.
Scoring Rubrics
Track 1 — Technical Search Quality
| Dimension | Weight | What it measures |
|---|---|---|
| SEO Fundamentals | 12% | Title 50–60 chars with primary keyword near front; meta description 145–160 chars with keyword and value hook; canonical set; H1 present and crawler-visible (not JS-only); SERP snippet earns the click |
| AEO Readiness | 13% | Structured to win featured snippets and People Also Ask; Q&A framing in content; definition clarity; FAQPage or HowTo schema present; entity (brand, product, topic) clearly named and disambiguated |
| GEO / AI Citability | 12% | Would Google AI Overviews, Perplexity, or ChatGPT surface and cite this page? Authoritative specific language; data-backed claims; quotable sentences; Article/Organization/Speakable schema; no vague filler |
Track 2 — Executive Audience Quality
| Dimension | Weight | What it measures |
|---|---|---|
| Executive Resonance | 14% | Language speaks to VP/C-suite concerns — business outcomes, strategic advantage, P&L, growth, risk — not feature lists or technical specs; framing is boardroom-ready in the SERP snippet itself |
| CXO Specificity | 10% | Explicit signals targeting CXOs: ROI quantification, competitive positioning, risk mitigation, market leadership, board-level imperatives; are CXO callouts present and credible? |
| Outcome Focus | 13% | Results are specific and quantified ("reduced X by 34% in 90 days") not vague ("improves performance"); before/after or benchmark framing; "so what?" is answered for a time-pressed executive |
| Industry / Domain Alignment | 12% | Clear vertical targeting; correct domain vocabulary; sector-specific pain points addressed; a reader from the industry immediately recognises this as relevant to them |
| Action Orientation | 8% | Meta description and title drive a clear, motivated next step; high-intent language; value proposition justifies the click; urgency or specificity is present |
| Brand & Trust Signals | 6% | Credibility markers in meta/OG content: client scale, awards, analyst recognition, proprietary data, proper OG hero image (not just a logo), schema completeness |
Weighted overall score = Σ(score × weight) × 10, rounded to nearest integer.
Step 3 — Find Evidence for Weak Dimensions
For any dimension scoring 6 or below, quote at least one specific field value (verbatim, in quotation marks) that illustrates the weakness. If a field is entirely absent, call that out explicitly as the evidence. For strong dimensions (7+), briefly note what's working.
Step 4 — Produce the Report
Use this exact structure:
Exec Meta Evaluation Report
Page / asset: [URL, filename, or inferred title] Page type: [Landing Page / Product Page / Home / Blog / Case Study / Other] Intended audience (inferred): [e.g., CMO, CTO, VP Sales — note "senior/executive" if applicable] Industry/vertical (inferred): [e.g., B2B Tech Services, FinTech, Healthcare IT]
Extracted Meta Fields
Title tag: [value] ([N] chars)
Meta description: [value] ([N] chars)
OG title: [value]
OG description: [value]
OG image: [URL or "not set"]
Twitter card: [type]
Canonical URL: [value]
H1: [value or "JS-rendered / not found"]
First paragraph: [first 200 chars…]
Schema types: [types or "none found"]
Schema issues: [issues or "none"]
Scorecard
| Dimension | Score /10 | Weight | Weighted |
|---|---|---|---|
| SEO Fundamentals | X | 12% | X.X |
| AEO Readiness | X | 13% | X.X |
| GEO / AI Citability | X | 12% | X.X |
| Executive Resonance | X | 14% | X.X |
| CXO Specificity | X | 10% | X.X |
| Outcome Focus | X | 13% | X.X |
| Industry / Domain Alignment | X | 12% | X.X |
| Action Orientation | X | 8% | X.X |
| Brand & Trust Signals | X | 6% | X.X |
| OVERALL | — | 100% | XX / 100 |
Grade: [A (85–100) / B (70–84) / C (55–69) / D (40–54) / F (<40)] One-line verdict: [e.g., "Strong executive framing but invisible to AI search — no schema, no AEO structure, no outcome data."]
Dimension Analysis
Group under two subheadings: Technical Search Quality and Executive Audience Quality.
For each dimension, write 2–4 sentences:
- What the score reflects
- A specific quoted field value as evidence (mandatory for scores ≤6)
- The core reason for the score
Top Recommendations (Priority-Ordered)
List 4–8 recommendations ranked by impact. Format each as:
#N — [Short action title] (addresses: Dimension Name) One paragraph: exactly what to change, why it improves the score, and a concrete rewrite example where applicable (revised title tag, meta description, schema snippet).
Quick Wins (under 30 minutes)
3–5 field-level fixes that can ship immediately:
- Missing canonical → add it
- Title 72 chars → trim to 58
- OG image is SVG logo → replace with 1200×630 PNG hero
- Twitter card
summary→summary_large_image - Schema logo empty string → set to full URL
Tone and Calibration Notes
- Be direct. This is an audit, not a compliment. Soft scores mislead.
- Anchor to real-world benchmarks: a best-in-class Fortune 500 B2B landing page targeting enterprise buyers is the reference for a "9" on Executive Resonance. A generic "solutions" homepage that could belong to any company is a "4."
- Always quote actual field values when citing evidence — never paraphrase.
- Recommendations must be specific: provide a rewritten title tag or meta description, not "improve keyword placement."
- For GEO/AEO: ask whether the content would survive a generative AI summary without losing meaning. Vague positioning language ("innovative solutions," "deliver value") gets omitted entirely — it is not cited.
- If a page is deliberately industry-agnostic (e.g., a global enterprise homepage),