An internal link points from one page on a domain to another page on the same domain. That is the dictionary line, and it hides the part that matters: an internal link is the only ranking signal you fully control. You do not negotiate it, you do not buy it, you do not wait for an editor to grant it. Every link in your navigation, your contextual paragraphs, your breadcrumbs and your related-posts modules is a vote you cast for your own pages, and Google folds those votes into the same PageRank computation it runs on external links, with a different weight but the same mechanics.

That framing changes how you treat them. A site with weak internal linking is not missing a nice-to-have, it is throwing away the one piece of off-page math it can edit at will. The two families are worth naming: navigational links live in menus, footers and sidebars and define structure, while contextual links sit inside body content and carry the strongest relevance signal because they are surrounded by topical text. When SEOs talk about internal linking as a lever, they almost always mean the contextual kind. The anchor text and the surrounding sentence tell search engines what the destination page is about, in your own words, which is precisely why this signal is trusted.

Mechanically, three things travel along an internal link, and conflating them is the most common conceptual error. First, crawl discovery: Googlebot finds a URL because something links to it, and a page with zero inlinks is effectively invisible unless it sits in a sitemap, and even then it is deprioritized. Second, link equity, the modern descendant of the original PageRank patent (Google patent US6285999B1): authority flows from page to page proportional to how the linking page itself is ranked, divided across its outbound links. Third, context: the anchor and its surrounding text classify the destination, which is why a contextual link inside a relevant paragraph outperforms the same URL dropped in a footer.

In 2026 the measurement is straightforward with any crawler. Screaming Frog, Ahrefs Site Audit and Semrush all expose an inlinks column and a crawl-depth column, and those two numbers tell you most of what you need. A page four or more clicks from the homepage is, in practice, a page Google visits rarely and ranks reluctantly. The legacy assumption that internal links barely move the needle died years ago: Google's own documentation in Search Central states plainly that the number of internal links pointing to a page is a signal of its relative importance. This is not folklore, it is documented behaviour, and it is why a deliberate internal structure beats a flat one even when external links are identical.

The reason internal links matter is that they let you decide which pages deserve to rank, instead of letting your site architecture decide by accident. A page that earns many relevant internal links from authoritative pages on the same site is being nominated, repeatedly, as important. That nomination compounds with external authority rather than competing with it. Google's own guidance reinforces this, and the clip below from Google Search Central walks through how internal linking shapes both crawling and ranking.

For a working illustration of the official line, this short video is worth the three minutes:

The operational consequence is that internal linking is the cheapest authority gain available to most sites, far cheaper than acquiring new external links and entirely within your control. When you plan a content campaign or a structured editorial push, the internal mesh you build between the new pieces is what turns a set of disconnected articles into a topical cluster Google can rank as an entity. We see this directly when we run a structured editorial program: pieces that are tightly interlinked outrank isolated pieces of identical quality, because the cluster signals topical coverage that a lone page cannot. Anchor text is part of this too, since a descriptive internal anchor tells the engine what the target page should rank for, in language you wrote rather than language a third party imposed.

Building an internal linking strategy

A strategy starts with hierarchy, not with link count. Map your site into pillar pages and topic clusters: a pillar covers a broad subject, the cluster pages cover its sub-topics, and the cluster links up to the pillar while the pillar links down to each cluster page. That single pattern does most of the heavy lifting, because it both concentrates equity on the pages you want to rank and tells search engines how your topics relate. The video below covers practical placement and anchor choices, and it pairs well with the structural view.

From there, three habits separate a real strategy from a plugin running on autopilot. Link from your strongest pages, the ones that already attract external authority, into the pages you are trying to push, so that inbound equity gets redistributed instead of pooling on a single URL. Link deep, into individual articles and product pages, not just into category hubs, because deep pages are where most sites starve. And write anchors that describe the destination in natural language: a varied, descriptive anchor profile reads as editorial, while the same exact-match phrase repeated across forty links reads as manipulation. When you are also running paid acquisition, the same logic applies to how you brief a campaign: we plan the internal mesh around the landing pages before we calibrate any external push, and you can see how that connects when you structure a campaign over the long run with a clear topical map underneath. The internal links and the cross-glossary signals, including how you write your page titles for each linked target, should all point in the same topical direction.

This is where internal linking and off-page work meet, and where most operators leave value on the table. When you acquire an external backlink, the equity lands on one URL. If that URL is an orphan or a thin page with no onward internal links, most of the value stops there. The page that receives a strong inbound link should already be wired into your internal graph so the authority flows onward to the commercial pages you actually want to rank. An expensive external link pointing at a dead-end page is a partial waste, and it is a waste you caused, not the publisher.

Stringer Network operates 28 owned media in-house, so we treat the internal graph of each property as a first-class asset rather than an afterthought, and the same discipline applies whether you build links yourself or source a placement directly from the publisher. The sequence we follow is simple: identify the money page, confirm it is reachable in two or three clicks, ensure the page receiving any external link redistributes equity toward it, and only then evaluate the external source. Internal first, external second. A coherent topical content program, the kind you would run to cover a subject across many connected angles, multiplies the effect, because each new piece adds inlinks to the cluster and the cluster lifts the pillar. The cross-references between pages, including how each anchor and each heading on the destination page reinforces the same theme, are what make a network of content behave like a single authoritative entity rather than a stack of pages.

Common mistakes and how to audit them

Most internal linking damage is invisible until you crawl. The recurring failures we see in audits are orphan pages with zero inlinks, money pages buried beyond four clicks, contextual links pointing to irrelevant destinations because someone matched a keyword instead of an intent, and exact-match anchor over-optimisation that looks templated. The fix for each is mechanical once you have the data. The walkthrough below shows the audit moves on real sites.

Run the audit in four passes. Crawl the full site with Screaming Frog or Ahrefs and export inlinks plus crawl depth. Filter for pages with zero or one inlink, those are your orphans and near-orphans, and they are the fastest wins. Cross-reference your crawl against your sitemap or analytics to find URLs that exist but are never linked. Then look at link distribution: a single page hoarding hundreds of inlinks while your commercial pages sit at three is a misallocation, not a structure. On the volume question, ignore the old myth of a hard cap at a hundred links per page, Google retired that guidance years ago, but a page stuffed with several hundred links still dilutes the equity each one carries, so restraint is a math decision, not a rule. The two thresholds worth holding in your head: every important page should be reachable within three clicks of the homepage, and every published page should carry at least one relevant contextual inlink. Anything failing those two tests is a ticket, and clearing that backlog usually moves rankings faster than the next batch of external links you were about to buy.