Corrected on September 17, 2026
The first version of this guide attributed things to Google that Google never said. John Mueller didn't confirm a "three or four clicks" limit, the 100-links rule didn't come from Matt Cutts in 2008, and no experiments showed that only the first anchor counts. It also claimed AI search engines follow internal links during retrieval, called Search Console's link report complete (it's a sample), showed breadcrumbs in mobile results where they no longer appear, said Lumina's Link Analyzer finds broken links and gave link-count thresholds with no source. The "live audit of 11 guides" had no saved data, so I removed it.
Internal linking is the SEO work most teams skip. Backlinks get outreach campaigns, schema gets validators, page speed gets dashboards. Internal links are something everyone agrees matter and few people audit, so the same problems show up again and again: pages nothing links to, "read more" anchors, important content that's hard to reach, and hub pages that don't link to their own topic.
Where Google has said something, I link or quote it. Where something is my judgment, I say so.
What Internal Linking Actually Is
Internal links connect pages on the same domain. They do three things: they let crawlers discover pages, their anchor text tells search engines what the destination is about, and they pass part of the linking page's importance along. A link from page A to page B is both a path and a signal.
The mechanism is simple. Googlebot fetches a page, extracts every <a href>, and adds the linked URLs to its crawl queue. The same happens on every page it fetches. Pages that many important pages link to are easier to find and look more important. A page nothing links to, an orphan, can still be found through the sitemap, but it gets no importance or context from the rest of your site.
Google keeps the advice short. Its link documentation says every page you care about should have a link from at least one other page on your site.
Why Internal Links Still Matter
Google still uses links to discover pages and to understand how they relate. Internal links are the part of that graph you fully control. You can't decide who links to you from outside, but you can decide which of your own pages get support from the rest of the site.
In 2018 John Mueller described it like this: when a page is linked from the homepage, Google treats it as more important than a page that takes several clicks to reach. He didn't give a number of clicks. The point is relative: whatever you link prominently, you mark as important.
How Google Actually Treats Internal Links
A few things are often repeated with the wrong source or the wrong detail. Here's what's documented.
There's no fixed link limit
The old advice to keep a page under 100 links came from Google's own early webmaster guidelines, at a time when Google only indexed about the first 100 KB of a page. Matt Cutts explained that background in a 2009 blog post. Today Google's documentation says plainly that there's no magic ideal number of links on a page. What still matters is whether a link helps: a page with thirty links that each point somewhere relevant gives clearer signals than one with two hundred boilerplate links.
Click depth matters, but not as a rule of three
The "3-click rule" is a usability heuristic that goes back to Jeffrey Zeldman's 2001 book Taking Your Talent to the Web, and Nielsen Norman Group has shown it doesn't hold even for users. Google never set a click limit. In practice, deep pages often have few internal links, and that's the part worth fixing.
Nofollow became a hint
Google announced in September 2019 that rel="nofollow" would be treated as a hint. For ranking that applied immediately, and for crawling and indexing from March 1, 2020. Using nofollow on internal links to steer PageRank ("PageRank sculpting") stopped working much earlier, which Cutts wrote about in 2009. There's rarely a good reason to nofollow your own internal links.
The 5 Internal Linking Patterns That Work
They aren't exclusive, and most well-linked sites combine several.
1. Hub and spoke
One hub page covers a topic broadly and links to more specific pages. Each specific page links back to the hub and to closely related siblings. This makes the topic structure obvious to readers and crawlers, and the hub is a natural target for external links because it's the most complete entry point.
2. Contextual links in body text
Links inside paragraphs, with anchor text that describes the destination, pointing to related articles or product pages. Google's documentation says it uses anchor text and, to some extent, the words around a link to understand the destination. That makes contextual links more informative than a generic menu entry.
3. Breadcrumb navigation
Breadcrumb trails (Home → Category → Subcategory → Page) show where a page sits in the hierarchy and add links up the tree on every page. With BreadcrumbList structured data Google can show the path in search results, but since January 2025 only on desktop; mobile results show just the domain. Breadcrumbs are still useful for navigation and structure either way.
4. Related-content blocks
The "related articles" section at the end of a page. Chosen by hand, it connects pages that really belong together. Filled automatically by date or random selection, it links unrelated pages and adds little. Curate them for your most important pages, and let automation handle the long tail only if its relevance logic is decent.
5. Footer and sidebar navigation
Sitewide links reach every page and make key sections easy to find. Because they're identical on every page, they say little about any specific page's topic. Keep them for genuinely central destinations, and don't count on them to replace contextual links.
The 6 Mistakes Most Sites Make
None of these is exotic. They build up over time when content gets published without anyone checking how it connects to the rest of the site.
1. Orphan pages
A URL that no other page on your site links to, so readers have no way to reach it. Orphans usually come from migrations, deleted category pages or articles that were published and never linked. For each one, decide: link it from a relevant page, redirect it or remove it.
2. Generic anchor text
"Click here", "learn more" and "read more" tell neither readers nor Google what's behind the link. Google's documentation recommends descriptive anchor text that's reasonably concise. The fix is usually to rewrite the sentence so the link sits on the words that describe the destination: "the schema markup guide" instead of "read more".
3. Important pages buried deep
A page that matters commercially but is only reachable through several layers of navigation gets little support from the rest of the site. You don't need to flatten your whole architecture. Add contextual links from strong pages, such as popular articles or category hubs, so the important page is closer to where readers and crawlers already are.
4. Many links to the same page from one page
Linking the same destination five times from one article doesn't add much and makes the text harder to read. How Google weighs multiple anchors to the same URL isn't documented. In 2021 Mueller said there's no simple rule like "only the first link counts" and that Google tries to understand the structure the way a user would. Link once or twice where it's most useful, with the clearest anchor.
5. Broken internal links
Internal links that return a 404 frustrate readers and send crawlers to dead ends. They typically appear after URL changes, migrations or restructures. Check them regularly, and either update the link or redirect the old URL to its current equivalent. Lumina's Broken Link Checker requests each link and reports the status code.
6. Links to redirects, noindex pages or non-canonical URLs
Links that point to a redirect, a noindex page or a URL whose canonical points elsewhere send mixed signals. Google usually sorts it out, but you're spending your own internal links on a detour. Link directly to the final, canonical URL, and check whether noindexed pages that receive a lot of internal links shouldn't actually be indexable.
How to Audit Your Internal Linking (3 Methods)
Each method finds a different kind of problem. For a small site, one is often enough. For a larger one, combine them.
Method 1: Lumina Link Analyzer
Lumina's Link Analyzer has two modes. Single page lists every internal and external link on one URL with its anchor text and rel attributes. Whole site reads the pages in your sitemap and counts how many other crawled pages link to each URL, separated into all links and links in body content. It also lists orphans (sitemap pages no other crawled page links to) and pages that are only linked from navigation. Two limits: the counts only cover the pages in that crawl, and the tool doesn't request each link, so it can't tell you which ones are broken. Use the Broken Link Checker for that.
Method 2: Search Console Links report
In Google Search Console, the Links report shows your top internally linked pages, based on Google's own crawl. It's a good way to see which pages your site emphasizes most. Google describes the data as a sample, not a complete list of every link, and the export is capped at 1,000 rows. So a sitemap URL that's missing from the report isn't proof that it's an orphan. Treat it as a lead and confirm with a crawl.
Method 3: Full-site crawl with Screaming Frog or similar
Screaming Frog (free up to 500 URLs), Sitebulb and Ahrefs Site Audit list inlinks per URL, crawl depth from the start URL and broken links. For large sites this is the most complete view. Useful filters: URLs with no internal inlinks, URLs with a high crawl depth, and important URLs with only a handful of inlinks compared with similar pages.
Anchor Text: What Google Actually Says
Google's link documentation asks for anchor text that's descriptive, reasonably concise and relevant to the page it points to. It lists "click here" and similar phrases as bad examples, and also warns against stuffing anchors with keywords or making them excessively long.
For internal links, that means writing anchors the way you'd naturally describe the destination. Using the destination's main keyword sometimes is fine, and so is using a synonym or the page title. Mueller has said keyword-rich internal anchors aren't a problem. I haven't seen evidence of Google penalizing internal anchor text, so there's no reason to invent percentage rules for how often an exact keyword may appear.
The real risk sits on the other side: anchors that say nothing. A quick check is to list the anchors pointing to your most important pages. If many of them are "read more", "here" or a bare URL, rewrite those sentences.
Internal Linking and AI Search
It's tempting to call internal links an "AI citation signal". There's no documentation that supports that. OpenAI's, Anthropic's and Perplexity's crawler pages describe what each bot is for (training, search indexing or fetching on request) and how to control it in robots.txt. None of them says how internal links affect which page gets cited.
What does apply is more basic. Search crawlers like OAI-SearchBot and PerplexityBot have to find your pages before they can index them, and links and sitemaps are how pages get found. A page that's linked from relevant articles is easier to discover than an orphan. On-request fetchers like ChatGPT-User load a specific URL when a user's question calls for it; they don't need your link graph for that.
So the useful work is the same as for Google: make sure important pages are linked from relevant content, avoid orphans, and let each page answer its question clearly.
A 5-Step Internal Linking Workflow
Run the Link Analyzer in Whole site mode, or a full crawl, and list pages that no other page links to. For each one: link it from a relevant page, redirect it or remove it from the site and the sitemap.
Run Link Analyzer →List the 10 to 20 pages that matter most for traffic or revenue. Compare how many body-text links each gets with similar pages on your site. The ones clearly below the rest come first.
Compare inlinks →Check the anchors pointing to those key pages. Replace "read more", "here" and bare URLs with words that describe the destination.
Audit anchors →For important pages that are only reachable through several levels, add contextual links from strong articles or category hubs. Then crawl again and compare the depth.
Check links for errors →Give pages below the homepage a breadcrumb trail with BreadcrumbList structured data. It helps readers and crawlers understand the hierarchy and can show the path in desktop results.
FAQ
Where to Start
If you only do one thing this week, find your orphan pages. Run the Link Analyzer's Whole site mode on your sitemap, list the pages nothing links to, and give each one a contextual link from a relevant page, a redirect or a clean removal.
If you have more time, take your 10 most important pages, count how many body-text links each gets, and add links from related articles to the ones that clearly fall behind. Then watch those pages in Search Console over the following weeks. Internal links are one signal among many, so judge the result on those pages' own data rather than on a promised number.
Audit your internal linking now
Lumina's Link Analyzer lists every link on a page with its anchor text, or crawls your sitemap to count inlinks and find orphan pages. Free, no signup.
Run Link Analyzer →