GEO Metrics Reference
Primenza GEO Intelligence scores crawled pages using 15 metrics grouped into four pillars: Readability, Citation, Structure, and Trust. This guide documents each metric as implemented in the product — not as a generic industry definition.
How GEO metrics are organized
Pillar scores are aggregated from page-level metric measurements during crawls. Each metric is normalized to a 0–100 scale for prioritization inside Primenza. Pillar scores are ESTIMATED aggregates from OBSERVED page measurements.
Readability pillar
Measures how easily AI systems can parse and summarize page content.
- ai_readability — sentence clarity, jargon density, and extractable prose structure
- semantic_clarity — consistent terminology and unambiguous topic statements
- topic_coverage — whether the page addresses its stated subject with sufficient depth
- knowledge_density — substantive information per section without filler or repetition
Citation pillar
Measures whether content is structured for attributable references in AI answers.
- citation_readiness — definitions, statistics, and quotable statements with context
- answer_quality — direct responses to likely user questions on the page
- source_credibility — references, authorship signals, and verifiable claims
- question_coverage — explicit FAQ or Q&A patterns that match user intent
Structure pillar
Measures machine-readable organization of entities and information hierarchy.
- entity_recognition — clear brands, products, people, and topics in text and markup
- structured_information — JSON-LD and schema coverage aligned with page intent
- information_hierarchy — heading order, sections, and logical content flow
- content_chunking — paragraph and list structure suited to retrieval snippets
Trust pillar
Measures signals that reduce misinterpretation risk for AI systems.
- hallucination_resistance — explicit limitations and qualified claims
- fact_consistency — alignment between titles, body copy, and structured data
- expert_signals — methodology notes, authorship, and first-party evidence
Methodology and limitations
Metric formulas are deterministic checks against fetched HTML and configured rules. GEO scores indicate measured readiness on audited pages — they do not guarantee AI citations or rankings. See /methodology for OBSERVED vs ESTIMATED labels and score semantics.
Where to use these metrics in Primenza
- GEO Intelligence audits with per-metric breakdowns
- GEO Wizard prioritization by pillar and metric
- API: run GEO audits and read metric payloads per page
- Public methodology at /methodology
References
Frequently asked questions
Are these metrics industry-standard?
They are Primenza product metrics documented from the live catalog. Other tools may use different names or formulas.
Can I export metric history?
Yes. Domain workspaces store timestamped audit results with per-page metric values where your plan allows.