How we score what AI sees.
GEOlens reports two top-line scores: SEO (classical, weighted Lighthouse) and AEO (Answer Engine Optimization, also called GEO). The vocabulary and weights below match the consensus that has emerged across the AEO tooling space since 2024.
SEO score
Hosted Lighthouse via Google PageSpeed Insights. Average across submitted URL plus up to 5 internal pages. Per spec §6.1:
- Performance · 25%
- Accessibility · 25%
- Best Practices · 20%
- SEO · 30%
AEO score
Three sub-scores combine into one AEO score:
- Engine Visibility · 60%
- AEO Hygiene · 25%
- Citability · 15%
Engine Visibility
We probe ChatGPT, Claude, Perplexity, and Gemini with three queries each:
- Brand recall — does the engine know the brand at all?
- Category placement — does the engine list the brand in its category?
- Citation behavior — does the engine cite the URL when asked about it?
Each probe is parsed into structured fields and scored:
probe_score = base × position × sentiment × accuracy position: primary 1.0 · secondary 0.7 · tertiary 0.4 · none 0.0 sentiment: positive 1.2 · neutral 1.0 · negative 0.5 accuracy: accurate 1.0 · partial 0.7 · misattrib 0.3
AEO Hygiene
On-page checks aggregated to a 0–100 score: presence of llms.txt, robots.txt rules for major AI crawlers, JSON-LD coverage (Organization, WebSite, Article, FAQPage, Product), Open Graph and Twitter cards, heading hierarchy, semantic HTML density.
Citability
Quantitative content shape: clean-text-to-boilerplate ratio, paragraph and sentence length distribution, structured-list density, FAQ pattern detection (in JSON-LD or HTML), statistical density.
What we don't do
- Live Google AI Overviews probes (deferred; requires SerpAPI-class data)
- 200M-prompt search-backed databases (deferred; that's incumbent moat)
- Continuous monitoring (one-shot reviewer for v1)
- Persist raw HTML beyond the scan (only computed signals; PRD §13)