Citation and Evidence Density

What this parameter measures

This parameter checks whether your content backs its claims with verifiable evidence. friendly4AI parses your page and counts five evidence signals: outbound links to external domains (same-domain and # anchors are excluded, and only http/https hosts count), statistical markers such as percentages, large comma-grouped numbers, and quantified figures ("3.2 million", "45%"), quotes (both <blockquote> elements and attribution phrases like "according to" or "said"), data tables (<table> elements), and a references or sources section detected by labels like "references", "sources", "bibliography", or "citations". From these it derives an internal evidence_density_score and an overall band.

Why it matters for AI-readiness

AI engines prefer to cite content that is itself well-sourced. Outbound links to credible domains, quantified data, and a references section signal that your claims are verifiable rather than asserted, the same trust markers a human fact-checker looks for. When a model can trace your statements to external evidence, it treats your page as a citable secondary source instead of an unverifiable opinion. Thin, unsourced pages are far less likely to be quoted in an AI answer, even when the underlying claim is correct.

How we score it

The v4.4 methodology scores this Authority and Trust parameter in three tiers drawn from the evidence signals. A page passes (100) only when it has 3 or more outbound external links AND at least one statistical marker, the combination that demonstrates genuine sourcing. It earns a partial (50) when any single signal is present (at least one outbound link, statistic, quote, or data table) but the full pass combination is missing. It fails (0) when none of those four signals appear. Separately, the processor reports an evidence_density_score that adds weighted points (up to 40 for outbound links, 25 for statistics, 10 each for quotes and tables, 15 for a references section) for richer diagnostics, but the pass/partial/fail score is governed by the outbound-link-plus-statistic rule above.

How to fix common issues

• Add at least 3 outbound links to credible external sources (research, official docs, primary data) on substantive pages — same-domain links do not count.
• Include quantified, dated statistics rather than vague claims, so the statistical-marker signal fires alongside outbound links to reach a pass.
• Use <blockquote> for quoted material and attribute it ("according to ...") so the quote signal is detected.
• Present comparative data in real <table> elements and add a "References" or "Sources" section for longer pieces.
• Re-scan and check the outbound_link_count, statistic_markers, quote_count, data_table_count, and reference_section evidence fields.