Method
We parsed visible server-rendered text after excluding script, style, and noscript blocks. Counts describe the fetched HTML only; they do not attempt to score writing quality.
Word and heading counts are paired with page purpose. An index or tool can be useful with fewer words, but it should not be presented as equivalent to a documented field report.
Observed evidence
823 words · 0 H2Most text came from repeated card summaries, not a single argument.246 words · 6 H2Short trust-method page.428 words · 5 H2Concise guide with external references.1248 words · 7 H2Longer, but length alone does not establish originality.Findings
- The longest sampled article was generated from a fixed six-section schema.
- The listing page's word count was inflated by repeated card excerpts.
- The fixed guide was concise, but contained no first-party measurement artifact.
Decision and resulting changes
Temporarily noindex legacy guides and generated articles while retaining access for readers. Make first-party field studies the indexable editorial core and evaluate each legacy page individually before restoring it.
- Removed legacy guides and automated articles from the sitemap.
- Added noindex metadata without deleting URLs or D1 records.
- Rebuilt the guides index around measured field studies.
Limits and reproduction
Re-run automation/evidence_collector.py against the public origin and compare the resulting JSON fields. A difference is evidence of a changed response, not automatically an improvement.
- Word count is not a Google quality threshold.
- This comparison does not judge prose accuracy or reader satisfaction.
Primary references
— Baseline captured from the public site, limitations documented, and remediation decision published.
Prepared by SiteSignal Hub Field Lab. Evidence collection and structural checks are automated; no personal use, customer result, or human test is claimed.