GEO Scoring Methodology — How CiteFlow Measures AI Visibility

Overview

CiteFlow's GEO scoring framework evaluates businesses on a 0 to 100 scale based on how visible they are across AI-powered search engines. The score is calculated by auditing 15 AI engines across 3 tiers — a methodology we publish publicly so you can understand exactly how your score is calculated.

This transparency is intentional. We believe GEO can only be effective if the metrics are clear and verifiable.

The 3-Tier Engine Architecture

CiteFlow audits AI engines in three tiers, reflecting how users actually discover businesses:

Tier	Engines	Weight in Score	Update Frequency	Why It Matters
Tier 1 — Primary AI Search	ChatGPT (OpenAI), Perplexity AI, Google AI Overviews (SGE), Claude (Anthropic), Microsoft Copilot	50%	Real-time	Primary discovery tools for most users
Tier 2 — Emerging AI Platforms	Grok (xAI), DeepSeek, Meta AI (Facebook), Samsung Galaxy AI, Xiaomi AI	30%	Weekly	Growing usage, especially in specific demographics
Tier 3 — Reference & Research	Perplexity Pages, NotebookLM, Phind, You.com, Komo AI, Phind	20%	Monthly	Used by power users and professionals for B2B decisions

The 15 AI Engines Audited

Why 15 engines? AI visibility isn't just about ChatGPT. Different audiences use different AI platforms. A Moroccan business traveler might use Grok or DeepSeek. A marketing manager uses Copilot. A researcher uses Perplexity Pages. CiteFlow tracks all of them so you're not optimizing for one platform at the expense of the others.

Engine	Company	Type	Language Support
ChatGPT	OpenAI	Primary AI Search	FR, EN, AR + 95 languages
Perplexity AI	Perplexity	Primary AI Search	FR, EN + 30 languages
Google AI Overviews	Google	Primary AI Search	FR, EN, AR
Claude	Anthropic	Primary AI Search	FR, EN, AR + 10 languages
Microsoft Copilot	Microsoft	Primary AI Search	FR, EN + 40 languages
Grok	xAI (Musk)	Emerging AI	EN, FR
DeepSeek	DeepSeek	Emerging AI	EN, ZH, FR
Meta AI	Meta	Emerging AI	FR, EN, AR
Samsung Galaxy AI	Samsung	Emerging AI	FR, EN, AR
Perplexity Pages	Perplexity	Reference & Research	EN, FR
NotebookLM	Google	Reference & Research	EN, FR
Phind	Phind	Reference & Research	EN, FR
You.com	You.com	Reference & Research	EN, FR
Komo AI	Komo	Reference & Research	EN, FR
Arc Search	Arc Browser	Reference & Research	EN, FR

Scoring Criteria

Each engine audit evaluates 4 dimensions:

1. Entity Recognition (25%)

Does the AI know your business exists as a distinct entity? Checked by asking category queries ("best accounting firm in Casablanca") and entity queries ("is [Business Name] a real company"). Scored: 0 if never mentioned, 25 if occasionally mentioned in context, 25 if reliably mentioned and contextually accurate.

2. Content Quality Signals (30%)

Is your content structured in a way AI engines can extract and cite? Evaluated through: Schema.org markup presence, structured FAQ content, factual/statistical claims in content, recency of information (publication date), and semantic clarity (clear headings, direct answers). Scored 0-30 based on how many quality signals are present.

3. Authority & Credibility (25%)

Does the AI trust your business as a credible source? Evaluated through: mentions in authoritative sources, consistency of information across the web (NAP), presence of reviews and testimonials in AI-accessible content, and year/founding date in entity description. Scored 0-25 based on number and quality of authority signals.

4. Semantic Relevance (20%)

Does your content directly answer the queries your target audience asks AI? Evaluated by running your target category queries ("best hotel Marrakech", "accounting firm Rabat") and checking if your business appears in the answer. Scored 0-20 based on match rate across representative queries.

Score Calculation

Final score = weighted average of all 15 engines:

Tier 1 engines: Each scored 0-100, averaged, then multiplied by 0.50
Tier 2 engines: Each scored 0-100, averaged, then multiplied by 0.30
Tier 3 engines: Each scored 0-100, averaged, then multiplied by 0.20

Example: If Tier 1 avg = 30, Tier 2 avg = 25, Tier 3 avg = 20 → Score = (30×0.50) + (25×0.30) + (20×0.20) = 15 + 7.5 + 4 = 26.5/100

Score Interpretation

Score Range	Level	What It Means	Action
0-10	Invisible	AI does not know your business exists	Start from basics: Schema.org + About page
11-25	Weak	Mentioned occasionally but not in context	Add structured content + FAQ pages
26-40	Developing	Appears in some queries with partial context	Authority building + content strategy
41-60	Moderate	Regularly cited with decent context	Fine-tune + expand multilingual coverage
61-80	Strong	Reliable AI citation with good context	Maintain + add original data/Research
81-100	Authority	Go-to citation for category queries	Protect + expand into adjacent categories

Moroccan Market Benchmarks

Based on 138 real audits in Morocco (January-May 2026):

Average score across all Moroccan businesses

Average ChatGPT score for Moroccan businesses

Average Google AI score for Moroccan businesses

13%

Businesses that score above 50 (strong level)

Why publish this? CiteFlow believes in transparent, verifiable metrics. If you want to estimate your score without using our platform, apply the criteria above to your website and content. The scoring is deterministic — the same inputs always produce the same outputs. This framework is what CiteFlow uses internally; we publish it so clients understand exactly what we're measuring.

Audit Frequency

CiteFlow runs AI visibility audits at three frequencies depending on plan level:

Monthly (Starter): Full audit of all 15 engines once per month
Weekly (Pro): Full audit every 7 days with delta reporting
Daily (Enterprise): Continuous monitoring with anomaly alerts

AI engines update their knowledge bases at unpredictable intervals. Monthly is the minimum for tracking meaningful change. Weekly is recommended for competitive industries. Daily is for businesses where AI citations directly impact revenue (tourism, hospitality, high-ticket B2B).

The GEO Scoring Framework