Competitor Overlap & Semantic Dominance Framework
A technical guide by AEObility, built for Australian businesses seeking clarity in how modern AI systems interpret, rank, and recommend local service providers. This framework shows how businesses across Perth, Australia, and New Zealand can measure competitor overlap, semantic dominance, and vector proximity using advanced AEO diagnostics such as cosine similarity, entity clarity scoring, and RAG survival testing.
Important Diagnostic Principle:AEObility's framework is an intentional query-specific diagnostic, evaluated per target search intent, rather than a generic site-wide SEO score. It assesses real-time search competition, content structure, and machine retrieval survival for specific commercial queries.
WHAT AEOBILITY CHECKS
AEObility compares your business against the competitors currently showing up for your target search queries. It checks whether your content is clear enough, structured enough, and relevant enough for AI systems—like ChatGPT, Google Gemini, and Perplexity—to understand and recommend your services.
Under the hood, the framework evaluates high-dimensional vector embeddings, Schema.org completeness, content depth ratios, and synthetic query retrieval survival.
The Four Core Questions We Answer
Instead of relying on keyword density or link counts, our framework answers four essential commercial questions:
1. Visibility Check
Does your business surface when prospective clients ask AI search engines for local recommendations?
2. Machine Understanding
Do AI models accurately parse your core services, locations, and value propositions?
3. Competitor Benchmarking
Is your content clearer and better structured than rival websites in your target market?
4. Prioritised Actions
What structural changes, schema additions, or content rewrites should you execute first?
Who This Framework Helps
Small Business Owners
Plain Language & Practical Outcomes- Higher Visibility: Get found across AI search engines, voice assistants, and map packs.
- Stronger Brand Trust: Ensure AI tools describe your services accurately without hallucinations.
- Local Discovery: Optimise coordinate maps and location signals for Australian service areas.
- Clear Priorities: Focus on straight-forward execution steps with no locked-in agency contracts.
Agency Teams & Consultants
Repeatable Telemetry & Diagnostics- Competitor Benchmarking: Measure relative vector distance using text-embedding models.
- Schema & Entity Auditing: Validate nested Schema.org microdata and RDF graph density.
- RAG Retention Testing: Test atomic answer block survival rates under prompt fan-out.
- Objective Metrics: Deliver standardized AI readiness scorecards for client reporting.
How the Audit Works: Step-by-Step Sequence
Instead of dumping raw technical data, AEObility follows a clear 6-step audit process:
Define the Target Search Intent
We pinpoint the exact question or local service query your ideal clients ask AI systems.
Identify Active Competitors
We pull the top competitor websites currently showing up for that search in real-time.
Compare Content & Structure
We evaluate how clearly your pages answer the search intent compared to your rivals.
Test AI Retrieval Survival
We run related prompt variations to check if AI search engines keep your content when generating answers.
Calculate Your Readiness Score
We aggregate relevance, structure, and depth into a clear 0–100 rating for your domain.
Build a Prioritised Action Plan
We turn your score into step-by-step recommendations on what to fix first.
Framework Execution Flow Chart
Search Intent
Target commercial query vector & user intent specification.
Competitor Set
Real-time organic SERP ingestion & active domain URL extraction.
Semantic Comparison
Cosine similarity vector scoring & net dominance differential.
Feature Review
Schema microdata audit, page depth & RDF graph density.
Readiness Score
Weighted 0–100 AI search readiness classification tier.
Recommended Actions
Prioritised 90-day AEO Sprint execution roadmap.
Standardised Component Modules
Each module follows a structured diagnostic schema: What It Does, How It Works, Inputs, Outputs, and Action Implications.
Why Cosine Similarity Matters for Vector Proximity
Cosine similarity computes the dot product of two normalized 768-dimensional vectors divided by their magnitudes. Because it measures the angle between vectors rather than their spatial distance, it evaluates pure semantic direction—ensuring content length variations do not distort relevance measurement against competitors.
1. Competitor Ingestion Module
Identifies the active competitor websites currently winning organic search attention for your target service terms.
Executes real-time SERP parsing using Brave Search and Gemini API endpoints, filtering out generic directories and aggregators to isolate true commercial domain nodes.
Target search intent query string (e.g. 'digital marketing specialist Perth').
Array of verified competitor homepage and service page domain URLs.
Establishes the realistic market benchmark you must outperform to gain citation priority.
2. Vector Proximity Module
Measures how closely your content's underlying meaning matches what prospective clients are asking AI search engines.
Converts text content into 768-dimensional vector embeddings using Google text-embedding-004 and calculates normalized Cosine Similarity.
Search intent string, client page text, and competitor page text chunks.
Individual similarity scores (0.00 to 1.00) for client and each competitor node.
Reveals whether your content is conceptually relevant or if your messaging drifts away from user intent.
3. Semantic Dominance Module
Determines whether your website commands stronger AI clarity than rival local businesses.
Calculates the net vector differential between client similarity and average competitor similarity: Math.max(0, Math.min(100, (ClientSim - CompAvgSim) * 100)).
Client vector similarity score and competitor vector similarity array.
Semantic Dominance Score (0 to 100% net differential).
A positive score indicates market advantage; a negative score signals risk of being overlooked in AI summaries.
4. Feature Extraction Module
Benchmarks your website's structural depth, Schema.org completeness, and trust signals directly against competitors.
Extracts Content Length Ratio (ClientLen / CompAvgLen), Schema.org entity count, internal link depth, and RDF subject-predicate triple graph density.
Scraped client HTML & DOM tree alongside competitor DOM structures.
Structured feature matrix (Content Ratio, Schema Completeness %, Technical Advantage, Entity Authority score).
Uncovers technical blind spots giving competitors an unfair advantage in machine indexation.
5. RAG Survival Testing Module
Tests whether AI search engines retain your content when generating conversational answer summaries.
Generates 3+ synthetic query prompt variations (query fan-out) and tests paragraph chunk retrieval retention against a strict 0.62 similarity threshold.
Target search intent and chunked client content paragraphs.
RAG survival rate (%), attribution rank, and selected passage chunks.
Shows if your content is written in direct atomic answer blocks or if key service facts get dropped during retrieval.
6. AI Search Readiness Scoring Module
Synthesises all diagnostic metrics into a single, actionable 0–100 rating with clear execution priorities.
Combines weighted categories based on observability config (Semantic: 40%, Technical: 20%, Entity: 15%, Competitor: 15%, Authority: 10%).
Engineered feature matrix, RAG simulation runs, RDF triples, and performance metrics.
Overall Readiness Score (0–100), Classification Tier (Dominant, Competitive, Emerging, At Risk), and Score Explanations.
Maps directly to a prioritised 90-day AEO Sprint execution plan focused on immediate high-impact fixes.
Module Summary Matrix
| Module Name | Code File | Primary Input | Key Output | Action Result |
|---|---|---|---|---|
| Competitor Ingestion Module | src/lib/telemetry/search.ts | Target search intent query string | Array of verified competitor homepage and service page domain URLs. | Establishes the realistic market benchmark you must outperform to gain citation priority. |
| Vector Proximity Module | src/lib/telemetry/proximity.ts | Search intent string, client page text, and competitor page text chunks. | Individual similarity scores | Reveals whether your content is conceptually relevant or if your messaging drifts away from user intent. |
| Semantic Dominance Module | src/lib/telemetry/features.ts | Client vector similarity score and competitor vector similarity array. | Semantic Dominance Score | A positive score indicates market advantage; a negative score signals risk of being overlooked in AI summaries. |
| Feature Extraction Module | src/lib/telemetry/features.ts | Scraped client HTML & DOM tree alongside competitor DOM structures. | Structured feature matrix | Uncovers technical blind spots giving competitors an unfair advantage in machine indexation. |
| RAG Survival Testing Module | src/lib/telemetry/rag-sim.ts | Target search intent and chunked client content paragraphs. | RAG survival rate | Shows if your content is written in direct atomic answer blocks or if key service facts get dropped during retrieval. |
| AI Search Readiness Scoring Module | src/lib/telemetry/scoring.ts | Engineered feature matrix, RAG simulation runs, RDF triples, and performance metrics. | Overall Readiness Score | Maps directly to a prioritised 90-day AEO Sprint execution plan focused on immediate high-impact fixes. |
Code Implementation Mapping
For developers and technical auditors, the framework maps directly to AEObility's open-source telemetry engine modules:
Queries Brave Search and Gemini APIs to harvest top organic competitor domain URLs for the target intent.
Generates 768-dimensional text-embedding-004 vectors and computes Cosine Similarity between intent and page chunks.
Calculates Semantic Dominance, Content Length Ratio, Schema Completeness %, and Entity Authority scores.
Executes synthetic query fan-out prompt passes and checks passage retention against a 0.62 survival threshold.
Applies observability platform weights to output the final 0–100 Readiness Score and diagnostic explanations.
Defines system weights: Semantic (40%), Technical (20%), Entity (15%), Competitor (15%), Authority (10%).
RAG_RETENTION_THRESHOLD = 0.62EMBEDDING_MODEL = text-embedding-004VECTOR_DIMENSION = 768WEIGHTS = Semantic 40%, Technical 20%, Entity 15%Competitor Overlap & Prioritised Blind Spots
When evaluating competitor overlap, AEObility distinguishes between Semantic Overlap (topic similarity) and Feature Overlap (structural capabilities). High semantic overlap without structured microdata causes content dilution, where AI models view your page as a generic duplicate.
Prioritised Blind Spot Action Matrix:
Competitors provide explicit JSON-LD data while your site relies on plain text prose.
Long prose paragraphs drop out during RAG retrieval passes.
Competitor sites cover intent sub-topics with greater topical detail.
Missing coordinate metadata prevents local map pack discovery.
Instructional Worked Example: Perth Local Business
Local Service Provider (Perth, WA)
The audit scraped real-time organic search results for "AEO consultant Perth", isolating the top 2 competing local domain homepages.
Although the client possessed positive semantic dominance (+9.3%), feature extraction revealed two key blind spots: initial Readiness Score was capped at 62 due to 0% Schema Completeness and unformatted prose paragraphs dropping out during synthetic prompt tests.
Deployed nested JSON-LD microdata and restructured key service sections into atomic answer blocks. Outcome: Readiness Score climbed from 62 to 88, securing top citation placement in ChatGPT and Gemini local recommendation summaries within 30 days.
| Metric | Before Audit | After AEO Sprint | Net Delta |
|---|---|---|---|
| Semantic Similarity | 0.785 | 0.862 | +0.077 |
| Schema Completeness | 0% | 100% | +100% |
| RAG Chunk Survival Rate | 45% | 92% | +47% |
| Overall Readiness Score | 62 (Emerging) | 88 (Dominant) | +26 pts |
What Your AI Search Readiness Score Means
Your final telemetry score (0–100) maps directly to four clear visibility tiers:
Consistently cited and recommended across conversational search engines and map assistants.
Strong overall foundation, but minor structural gaps allow top competitors to capture citations.
Moderate relevance, but at risk of being missed by AI search due to missing schema or thin content.
AI search models struggle to parse your business, resulting in zero citation share and low discovery.
Frequently Asked Questions
Q1.What is the Competitor Overlap & Semantic Dominance Framework?
It is a query-specific diagnostic methodology that measures how strongly your site competes for a given intent by analysing competitor overlap, semantic coverage, technical readiness, and authority signals.
Q2. Why does competitor overlap matter for AEO?
Competitor overlap shows which sites consistently appear for the same intent across engines, revealing who answer systems trust and where your site is missing from the dominant cluster.
Q3. What goes into the AEO readiness score?
The score blends semantic coverage, technical implementation, entity clarity, competitor presence, and authority signals into a single readiness metric for a specific query.
Q4. How does AEObility turn this framework into practical actions?
AEObility converts diagnostic findings into concrete content, schema, and entity recommendations so your site can move from at-risk or emerging into competitive or dominant tiers for the target query.
Run Your Free AI Visibility Diagnostic
Discover how your website performs across conversational search engines, maps, and AI recommendation systems. Get your custom AI Search Readiness Score.