How AiVIS Cite Ledger scores AI visibility

The scoring formula (avs-v3)

The AiVIS Cite Ledger composite score is a weighted sum of ten independent scoring families, each scored on a 0–100 scale. Weights were derived from observed citation patterns across major answer engines and reflect how heavily each signal influences whether a page gets extracted and quoted in a generated response.

Each family score is computed independently before weighting. Hard-blocker caps apply first: one hard blocker caps the composite at 79, three or more cap it at 59. Hard blockers are absent JSON-LD, absent title tag, absent H1, or AI crawlers blocked in robots.txt.

Family weights and signals (avs-v3)

Score tiers and citation readiness

AiVIS Cite Ledger maps composite scores to five citation readiness tiers. These thresholds reflect observed behavior across Perplexity, ChatGPT Browse, and Google AI Overviews - not theoretical ideals. Pages below 50 face structural extraction barriers that content improvements alone cannot resolve.

The BRAG trail protocol

BRAG is AiVIS Cite Ledger's internal evidence chain standard. Every finding in an AiVIS Cite Ledger Cite Ledger audit report must pass a four-step BRAG verification before it is surfaced as a recommendation. This prevents generic advice - the kind that applies to every site and helps none of them - from appearing alongside evidence-grounded findings.

In practice, the BRAG trail means every recommendation in an AiVIS Cite Ledger report answers three questions simultaneously: what exactly is wrong on this specific page, where is the evidence in the crawl output, and what specific change will move the score. Teams that implement fixes without this chain typically address symptoms while missing root causes.

Validation pipeline

AiVIS Cite Ledger audits run through a multi-stage validation pipeline before scores are finalized. The pipeline is designed to distinguish high-confidence findings - those grounded in directly observable page structure - from advisory findings that reflect best practice patterns but cannot be verified by crawl alone.

The optimization loop

A single audit is a diagnostic, not a solution. AiVIS Cite Ledger is designed for iterative improvement cycles where teams fix a cluster of related findings, re-audit, and measure category-level delta rather than overall score movement alone. Overall score can mask improvement in one dimension while another degrades - category tracking prevents this.

The most common optimization failure mode is fixing all technical issues first while leaving content depth and structured data untouched. Technical Trust accounts for only 15% of the composite score. A page can achieve a perfect technical trust score and still sit at 30 overall if content depth and structured data are near zero. Always prioritize dimension weight when sequencing fixes.

What answer engines actually extract

AiVIS Cite Ledger scoring is grounded in the extraction behavior of retrieval-augmented generation pipelines. When Perplexity, ChatGPT, or Gemini answers a question using web sources, the selection and extraction process follows a predictable pattern:

Entity resolution first. The model identifies what the page is about by reading the title, H1, first paragraph, and any schema with a name or description field. Pages with ambiguous or inconsistent entity signals are assigned lower retrieval priority regardless of content quality.

Passage extraction second. The model scans for short, self-contained answer units - sentences or paragraphs that fully answer a question without requiring surrounding context. This is why direct answer blocks and FAQ-style sections outperform long narrative content in citation selection even when the narrative is higher quality.

Trust verification third. The model cross-references the source against signals that indicate authority: presence of a methodology or about page, internal links to trust documents like privacy and terms, external corroboration from other indexed sources, and schema that asserts organizational identity. Pages without these signals are cited less frequently on contested or consequential queries.

AiVIS Cite Ledger scores each of these extraction phases through its dimension framework. Content depth and AI readability measure extractability. Structured data and heading structure measure entity resolution quality. Meta tags, technical trust, and security & trust measure reliability and accessibility. Improving all ten scoring families together is the only reliable path to consistent citation across answer engines.

Deterministic output contract

The methodology enforces a deterministic query-layer contract: same input URL + same entity seed + same configuration produces the same query set and scoring rules. Model variance is isolated to model output generation, not to query construction or weighting.

Formal clause: All evaluations are deterministic at the query layer. Variance is isolated to model output, not system design.

Verification loop definition

• Single pass: observation only.
• Double pass: same result repeated under identical query conditions = soft confirmation.
• Cross-query confirmation: agreement across independent query intents = verified signal.

Confidence scoring logic

Confidence combines four numeric inputs: frequency across queries, position stability, source authority weighting, and cross-query agreement. Output is normalized to High (>=0.80), Medium (0.55-0.79), and Low (<0.55).

Failure modes and limits

• Entity ambiguity (brand token collides with common-language usage).
• SERP personalization variance across location/session state.
• Temporal volatility during news spikes and event windows.
• Model hallucination vs retrieval mismatch.
• Thin-source environments with low-quality or sparse source coverage.

CITE LEDGER, from BRAG evidence to citable data

BRAG (Based-Retrieval-Auditable-Grading) is the evidence protocol. CITE LEDGER is the structured record each audit produces, the transformation layer that converts messy scraped content into deterministic, citable ground truth. ML in this pipeline operates as a deterministic filter: its job is to reject any scraped evidence that does not meet the structural requirements for a valid citation.

ledger_entry {
  query
  timestamp
  result_snapshot
  entity_detected
  position
  source_url
  serp_feature
  confidence_score
}

The audit-fix ML feedback loop

The ML model is trained on audit failures (where citations were rejected or incorrect) and manual fixes (human corrections to data). Each phase transforms the data closer to citation-ready state.

Transforming scraped evidence into citations

For data to be citable, it must transition from subjective interpretation to objective reference. The pipeline enforces this through three layers:

Upstream, Immutable capture. The system captures the DOM snapshot and Computed CSS at the moment of extraction. This evidence is non-negotiable and forms the immutable base layer of every CITE LEDGER record.

ML Auditor, Stability classification. A classifier trained on citable vs. non-citable data evaluates DOM element stability. If an element's ID or class varies too much across audit logs, the ML flags it as volatile and requests a more stable anchor.

Downstream, Reliability scoring. ML generates a reliability score from 0.0 to 1.0. Only data scoring above 0.98 earns a citation handle, a unique, auditable reference that resolves back to the original rendered source.

Pipeline principles

Related documentation

The AiVIS Cite Ledger Guide covers how to interpret audit output and sequence implementation. The FAQ addresses common questions about score interpretation, category grades, and optimization sequencing. The Compliance page documents data handling and crawl governance policies. Teams running the full optimization loop can track progress in Report History.