Our Methodology

Research-Backed AI
Vizibility Scoring

GetVizible’s Vizibility Score analyzes 58 real signals across 5 categories using weights derived from peer-reviewed research and industry data. Here’s exactly how it works.

Research Foundation

Grounded in Real Research

Our scoring methodology isn’t based on hunches or best guesses. Every category weight and signal threshold is derived from published research and real-world data:

  • Princeton GEO Study (KDD 2024) — Aggarwal, Murahari, Rajpurohit et al. demonstrated that Generative Engine Optimization techniques can improve source visibility by up to 115%. Their research established that citations, quotations, statistics, and fluency are the primary factors AI engines use to select sources.
  • Seer Interactive SearchGPT Study — Analysis of over 10,000 SearchGPT results revealed that traditional SEO ranking correlates weakly with AI citation. Content structure, semantic clarity, and authority signals matter significantly more in AI search contexts.
  • Digital Bloom AI Visibility Report — Found that only 2–3% of websites are regularly cited by AI engines, and those that are share common structural and content patterns: comprehensive definitions, structured data, and clear entity information.
  • Semrush AI Search Visibility Study — Analyzed how AI-powered search features select and present sources, confirming that structured content and topical authority are primary selection criteria.
  • Google E-E-A-T Documentation — Google’s Experience, Expertise, Authoritativeness, and Trustworthiness framework provides the foundation for how search engines (including AI-powered ones) evaluate content quality and source credibility.
Scoring Framework

5 Categories, 58 Signals

Each website is scored across five categories with research-backed weights.

20%

SEO Foundations

12 signals
25%

Content Quality

14 signals
20%

GEO Optimization

12 signals
25%

AI Readiness

12 signals
10%

Authority

8 signals

Category Breakdown

Category Weight Key Signals Research Basis
SEO Foundations 20% Meta tags, canonical URLs, robots.txt, sitemap, page speed, HTTPS, mobile-friendly, heading structure, image alt text, internal links Google E-E-A-T, Semrush
Content Quality 25% Word count, reading level, citation density, definition patterns, structured data, FAQ presence, original statistics, comparison tables Princeton GEO, Seer Interactive
GEO Optimization 20% Fluency optimization, quotation integration, statistic citation, technical terminology, authoritative tone, source attribution Princeton GEO (115% visibility improvement)
AI Readiness 25% Schema.org markup, llms.txt, structured tables, clean HTML, entity markup, API accessibility, content machine-readability Digital Bloom, Semrush
Authority 10% Author attribution, organizational schema, contact information, about page, editorial standards, external citations, trust signals Google E-E-A-T, Digital Bloom

Signal Deduplication

Our v3.0 methodology includes rigorous signal deduplication to prevent double-counting. For example, “has structured data” (SEO) and “schema markup depth” (AI Readiness) are distinct signals measuring different aspects of structured data implementation. Each of our 58 signals measures a unique, independent aspect of AI visibility.

Scoring Scale

The Vizibility Score is a composite score from 0 to 100:

Score Range Grade Interpretation
90–100 A Excellent AI visibility. Likely cited by AI engines regularly.
75–89 B Good foundation with clear optimization opportunities.
50–74 C Significant gaps. AI engines may overlook this brand.
25–49 D Poor AI visibility. Urgent optimization needed.
0–24 F Effectively invisible to AI search engines.
Case Study

Self-Audit: 37 → 87 in 3 Rounds

We use our own tool to continuously improve our own AI visibility. Here’s the real data from our self-improvement experiment.

37
Baseline
77
Round 1
89
Round 2
87
v3.0
Round Score Key Changes Methodology
Baseline 37/100 Initial website with minimal optimization v1.0 (47 signals)
Round 1 77/100 Added structured data, FAQ schema, citation density, expert quotes, comparison tables v1.0 (47 signals)
Round 2 89/100 Added llms.txt, improved content depth, more inline citations, image alt text v2.0 (66 signals)
v3.0 Baseline 87/100 Scored with stricter v3.0 methodology; identified new gaps in authority, SEO, citations v3.0 (58 signals, deduplicated)

Note: The score dropped from 89 to 87 between Round 2 and the v3.0 baseline because v3.0 introduced a stricter, deduplicated methodology with 58 unique signals (down from 66 which had some overlap). The v3.0 score is more accurate and harder to game.

This self-audit process demonstrates the core value of continuous AI visibility optimization: audit, identify gaps, implement fixes, re-audit. Each round takes hours, not weeks, because Agent Zero automates the entire cycle.

Transparency

Our Commitment to Honest Scoring

🔍 Built by AI. Audited by AI. Continuously Improved by AI.

Our tool is built by AI, audited by AI, and continuously improved by AI. Every Vizibility Score is generated from real website analysis — no fake data, no simulated results, no random number generators. When you run an audit, our engine actually fetches your website, parses your HTML, analyzes your content, checks your structured data, and scores each of the 58 signals against research-backed thresholds.

Our methodology is research-backed and transparent. We publish our scoring categories, weights, and research sources so you can understand exactly how your score is calculated. If you disagree with a signal or weight, we welcome that feedback — that’s how the methodology improves.

We also practice what we preach: GetVizible regularly audits itself with its own tool and implements the recommendations. The self-audit case study above uses real data from real audits.

Editorial Standards

All content on GetVizible is created with the following standards:

  • Research-backed claims: Every statistic and finding is cited with a link to the original source.
  • No fabricated data: All scores, metrics, and case study data are real and verifiable.
  • Transparent methodology: Our scoring weights, signal counts, and research bases are published publicly.
  • Regular updates: Our methodology evolves as AI search engines evolve. Version history is maintained.
  • AI-created, human-reviewed: Content is generated by Agent Zero and reviewed by human co-founders for accuracy and relevance.

See Your Vizibility Score

Run a free audit and see how AI engines perceive your brand across all 58 signals.

Run Free Audit → About GetVizible