COMMUNITY OBSERVED📈 Jev for SEO

Internal Link Optimizer - 500-Page Batch Semantic Graph

Automated cross-page relevance scoring and contextual internal link recommendation across 500+ URLs.

Internal Link Optimizer - 500-Page Batch Semantic Graph

Overview & Result

SEO authority Ian Nuttall published an automated internal link builder using Jev. Traditional internal linking tools rely on brittle exact-match keywords or expensive LLM prompt chains. By using Jev's Score API, this tool evaluates 500 crawled articles against target keyword siloes in under 3 minutes, recommending high-intent contextual links and exporting structured CSV/JSON audit reports for under 50 cents total.

How Jev fits in the loop

  1. Crawls target website sitemaps to build an in-memory index of URLs, H1s, and paragraph excerpts.
  2. Streams paragraphs through Jev Score API to compute contextual relevance (0.00 - 1.00) against target pillar pages.
  3. Applies a configurable threshold (>0.85) to filter high-confidence link insertion candidates.
  4. Generates exact in-text anchor placement suggestions and produces a downloadable CSV mapping.

How to reproduce

  1. Clone iannuttall/internal-link-jev and install dependencies.
  2. Configure sitemap.xml endpoint and set JEV_API_KEY in .env.
  3. Execute node run.js to start asynchronous crawl and Jev Score scoring stream.
  4. Inspect generated internal_links_audit.csv with recommended anchor placements and relevance scores.

Why this build matters

Democratizes programmatic SEO link audit automation. What previously required $30+ in OpenAI API calls and complex scripts now runs on desktop node scripts in minutes for pocket change.

Reported performance

Reported by author

Cost: $0.48 / 500 pages

Latency: 45ms / comparison

Evaluated over 3,200 paragraph-to-target comparisons across 500 crawled HTML pages in 3 minutes total for $0.48.

Limitations

  • Does not automatically modify live CMS databases; outputs audit proposals requiring editorial sign-off.
  • Requires target site to have accessible clean paragraph markup without aggressive anti-scraping blocks.

Patterns