Internal Linking Suggestions

SEO Dash semantically searches your crawled content to find pages that should link to each other but don't, then scores each suggestion by topical relevance. You get a sortable list of (source page → destination page) pairs with anchor-text suggestions and the body-text snippet to drop the link into.

How it works

The builder runs against your latest crawl + (optionally) your latest topical map. For each page:

Find candidate destination pages that are semantically similar but not yet linked.
Boost candidates that share a topical-map cluster with the source page.
Generate an anchor text and locate a snippet in the source page where the link would read naturally.
Score by combined semantic similarity + cluster fit + structural opportunity.
Persist as suggestion rows you can dismiss or mark as applied.

Re-running the builder reuses your dismissals and applied marks, so you only see fresh open suggestions next time.

Where to see them

/utilities/linking-suggestions - the full sortable list with bulk actions.
Agent - get_linking_suggestions tool surfaces them with optional filtering by source page URL. The agent uses these when answering "what internal links should I add to /pricing?".

What makes this different from a generic keyword-match tool

Semantic, not literal. A page about "keyword research" can be linked to from a page that says "finding the right phrases" without sharing keywords.
Topical-aware. Suggestions across cluster boundaries get downweighted; suggestions within a cluster get boosted.
Grounded in your actual crawl. No external corpus, no sample data. Only pages on your own site.

Per-tier access

Available on every tier. Quality scales with crawl size and embedding coverage.