---
title: Internal Link Equity Graph
category: product
entity_type: skill
price: $15
canonical: https://forgehouse.ai/skills/internal-link-equity-graph/
lang: en
hreflang_alt: https://forgehouse.ai/tr/skiller/internal-link-equity-graph/
last_updated: 2026-06-20
---

# Internal Link Equity Graph

> Map your content as a directed graph and distribute authority with PageRank-style analysis, finding orphan pages and recommending relevance-validated inbound links.

A systematic internal-link engine that maps your content as a directed graph and distributes authority with a PageRank-style analysis. It tokenizes your corpus, scores page authority, finds orphan pages, and recommends new inbound links with relevance-validated anchor text, replacing slow manual audits with a repeatable pipeline.

## Use cases
- Auditing internal link structure across a 100-to-10,000-page content site
- Finding orphan pages that receive too few inbound links
- Identifying which pages deserve pillar status by authority score
- Generating anchor-text recommendations validated by topical similarity
- Measuring topic-cluster coherence and hub-and-spoke link density
- Setting up incremental link-graph updates triggered by CMS webhooks

## Benefits
- Replace a multi-hour manual audit with a 30-second graph build
- Surface every orphan page and the exact authority pages that should link to it
- Strengthen pillar pages with data-driven internal link distribution
- Stop anchor-text spam with a relevance threshold that rejects weak links

## What’s included
- Corpus tokenizer with TF-IDF vectorization and Turkish stopword support
- Directed-graph builder with PageRank authority scoring
- Orphan-page detector and low-authority page flagging
- Anchor-text recommendation engine scored by authority times relevance
- Cluster coherence analysis with hub-and-spoke link-density reporting
- Markdown dashboard output and incremental CMS-webhook update support

## Who it’s for
SEO engineers and content strategists managing large sites who want PageRank-grade internal-link authority analysis instead of manual spreadsheets.

## How it runs
A manual internal-link audit takes 4 to 6 hours; this graph pipeline does it in about 30 seconds. Pages become a directed graph, PageRank surfaces the authority, and orphans leave with anchor-ready link suggestions.
1. Loads the content corpus (Sanity export or MDX files) and vectorizes each page from title, headings and the first 2000 body characters with TF-IDF, Turkish stopwords and 1-2 n-grams included.
2. Builds a NetworkX directed graph: every existing internal link becomes an edge weighted by semantic similarity (TF-IDF cosine x 0.7) plus temporal proximity (x 0.3); links below 0.15 similarity get flagged as low-relevance.
3. Runs PageRank (alpha 0.85) across the graph, marks the top 5 percent as pillar candidates and lists every published page with fewer than 2 inbound links as an orphan.
4. Generates inbound link recommendations for weak pages: candidate sources need TF-IDF similarity above 0.4, scored by source authority x relevance, with a 2 to 5 word anchor suggestion pulled from the target title. Anchors like 'click here' are blacklisted.
5. Detects topic clusters via Louvain community detection and checks coherence: intra-cluster link density should run about 2x cross-cluster density, low ratios raise a restructure alarm.
6. Renders a markdown dashboard built for a 30 second decision: top 10 authority pages, orphan list with top 3 suggested inbound links each, and cluster coherence in plain 1 to 10 scores instead of raw PageRank decimals.

## FAQ
### Our content lives in a headless CMS. Can it still build the graph?
Yes. The engine tokenizes your corpus from a content export, so where the content is hosted doesn't matter. It also supports incremental graph updates triggered by CMS webhooks, so you don't rebuild from scratch on every publish.

### Couldn't I just track internal links in a spreadsheet?
At a hundred pages maybe; at a thousand, no. The engine builds a directed graph, scores authority PageRank-style, and validates anchor suggestions with TF-IDF topical similarity. A multi-hour manual audit becomes a 30-second graph build, with orphan pages and their best linking candidates listed automatically.

### Does it insert the recommended links into my pages?
No. The output is a recommendation list with anchor text and targets, plus a markdown dashboard. Placing the links is your editorial or CMS step, and suggestions below the relevance threshold are rejected before they ever reach you.

## Price
$15, one-time, no subscription. VAT included.

Related guide: [How to automate SEO and AEO with Claude](https://forgehouse.ai/guides/automate-seo-claude/)