# Internal links

Inline anchor links pointing at other pages on the same domain. The density of those links is the authority topology — an orphan page reads as weak even when its content is strong.

By AgentSite · 2 min read · Updated 2026-05-24

Internal links are anchor links from one page on your domain to another. Their density forms an authority topology engines read alongside the content. A page densely embedded reads as central; an orphan reads as peripheral, regardless of content quality. The floor for a cited page is three contextual internal links.

## What "contextual" means

[Google's links-crawlable documentation](https://developers.google.com/search/docs/crawling-indexing/links-crawlable) is direct on the format: "Google can only crawl your link if it's an `<a>` HTML element with an `href` attribute." JavaScript-only navigation (`onclick` handlers, hash routes, `<div>`\-based menus) is invisible to non-rendering crawlers. The `<a href="">` is the protocol.

Contextual means the link sits inside body prose, not in the footer or a sidebar component. The anchor text describes the target — "see [direct answer](/direct-answer)" not "click here." Google's guidance: "Anchor text tells people and Google something about the page you're linking to. Effective anchor text should be descriptive and specific."

Three contextual links per page is the floor because each one does work — it embeds the page in the citation graph, gives the agent a follow-up path, and tells the engine which other pages on the site share authority with this one.

## What [schema.org](http://schema.org) names

Structured-data has parallel concepts. [Schema.org/WebPage](https://schema.org/WebPage) defines `relatedLink` ("a link related to this web page, for example to other related web pages") and `significantLink` ("one of the more significant URLs on the page… typically, these are the non-navigation links that are clicked on the most"). [Schema.org/SiteNavigationElement](https://schema.org/SiteNavigationElement) marks navigation regions explicitly so extractors don't conflate menu chrome with editorial links.

A site that ships all three (inline `<a>` plus `relatedLink` for editorial cross-refs plus `SiteNavigationElement` for menu items) gives every consumer a clean separation between "what to read next" and "site furniture."

## Anti-patterns

Three failures show up reliably:

1.  **Orphan pages.** A page with zero inbound links from anywhere else on the site. Even strong content reads as peripheral; the engine has no path in.
2.  **JavaScript-only navigation.** The menu uses `onclick="navigate('/x')"` instead of `<a href="/x">`. Non-rendering crawlers walk away with zero of the site's structure.
3.  **All-footer linking.** Every internal link sits in a global footer; nothing inline in the body. Engines treat footer links as chrome and weight them lower than contextual editorial links.

## Where this fits

Internal links is a Layer 4 signal — content-quality territory because the link graph is part of how the engine assesses authority. It pairs with [llms.txt](/llms-txt) (the curated index of the most-quotable pages) and with the catalog of [AEO problems](/aeo-problems) (a graph of failure modes that all link to each other). The layered place is in [the five layers of AEO](/five-layer-aeo).