Every internal linking guide converges on the same checklist. Define internal links, list the SEO benefits, hand over eight to twenty best practices, done. None of them answers the question a writer standing in front of a live page actually has. Should this specific link exist, from this page, to that one, with this anchor text?
The fix is not another checklist. It is a yes/no decision rule, applied page by page and anchor by anchor, after an audit that finds the orphans and the under-linked pages first.
This applies once a site has enough pages that habit and memory can no longer track what links where, in practice anywhere from about 100 pages up. It will not fix a page that has nothing worth linking to. Internal linking cannot replace content work, and it cannot reverse a ranking drop Google has already decided on for other reasons.
If your crawl budget and internal link structure need a closer look than this guide covers, TRM's Technical SEO and AI Readiness work finds the orphaned and under-linked pages holding your site back. One technical review tells you where to start.
Table of Contents
- What Internal Links Do For Crawling And Ranking
- Types Of Internal Links You Are Already Using
- Audit Your Internal Links: Find Orphans, Click Depth And Under-Linked Pages
- The Decision Rule For Every New Internal Link
- Anchor Text Rules That Actually Hold Up
- How Many Internal Links Is Too Many?
- Fixing Broken Links, Redirect Chains And Nofollow Mistakes
- Build Hub-And-Spoke Clusters So Pillar Pages Carry The Right Pages
- The Publish-Day Checklist
- When To Automate Internal Linking Instead Of Doing It By Hand
- FAQs
What Internal Links Do For Crawling And Ranking
Google finds new and existing pages primarily by following links from pages it has already crawled. Google's guidance on crawlable links states plainly that a page that matters should be reachable through an ordinary link from another crawled page, and that a page carrying far too many links on it is probably overloaded. That single sentence from Google is most of the technical case for internal linking. Everything else in this guide is working out what to do about it on a real site.
A page with no internal links pointing to it is called an orphan page. Crawlers that follow links to find new content may never reach it at all, however good the page is, however well it targets a keyword, however much work went into writing it. If nothing on your site points to a page, that page is effectively invisible to the system that decides what gets crawled, indexed and ranked.
Link equity is the mechanism underneath that same Google guidance. Every page on a site holds some share of authority, built from the backlinks pointing to the domain and the structure of the site itself. In general terms, that authority moves between pages through internal links, loosely in relation to how many links a page sends out and which pages it sends them to. A page that receives links from several already-authoritative pages tends to carry more weight than a page that receives none. A page buried four or five clicks from the homepage, reachable only through a chain of other links, tends to get a thinner share of that equity than a page linked directly from the homepage or a major hub.
None of this is exotic. It is the same logic that made PageRank work in the first place, just applied inside one domain instead of across the open web. The practical upshot is that internal linking is not a cosmetic choice about reader navigation. It decides which pages Google can find, how quickly, and how much authority each one gets to work with once it is found.
This matters more, not less, as a site grows. A 20-page site can get away with sloppy linking because every page is a click or two from every other page by accident. A site with hundreds or thousands of pages cannot. At that scale, the pages that should be ranking and the pages that are actually well-linked start to diverge, quietly, without anyone deciding that should happen. That divergence is usually invisible until someone runs the kind of audit in the next section, which is also where a deliberate approach to internal linking separates itself from SEO strategies for B2B SaaS companies that treat linking as an afterthought rather than infrastructure.
Types Of Internal Links You Are Already Using
Every site already has all four types of internal link in place, usually without anyone naming them. The rest of this guide refers to these terms without redefining them, so it's worth being precise once.
- Navigation links: The links in your header and main menu, present on every page, pointing to the handful of top-level sections of the site.
- Contextual links: Links placed inside the body text of a page, pointing to another page because the content genuinely relates to it. These are the links this guide is mostly about.
- Footer links: Links in the site-wide footer, usually pointing to secondary pages like legal, contact or sitemap pages that don't belong in the main navigation.
- Breadcrumb links: The trail near the top of a page showing its position in the site hierarchy, such as Home > Blog > Technical SEO, each segment a link back up the structure.
Navigation, footer and breadcrumb links are structural. They exist on every page of a given type and don't change based on the specific content of that page. Contextual links are the opposite. They are decided one at a time, based on what a specific page is actually about, which is why the rest of this guide focuses mainly on them.
Audit Your Internal Links: Find Orphans, Click Depth And Under-Linked Pages
Before deciding what to change, find out what you actually have. This is the step almost every competing guide skips, in favour of jumping straight to best practices. Run it as a sequence, in order.
- Pull the Internal Links report. Search Console's Internal Links report lists how many internal links point to each URL, and it's the fastest free way to see where you actually stand. Export the full list rather than scrolling it; you'll want it in a spreadsheet for the next steps.
- Sort by link count, ascending. Anything at zero is a confirmed orphan. Anything in the low single digits on a site of meaningful size is an orphan candidate, worth a second look even if it technically has one link in.
- Check click depth from the homepage. For each page you're worried about, work out how many clicks it takes to reach it starting from the homepage, following only the links a normal visitor would click. A widely used rule of thumb says to keep your most important pages within three clicks of the homepage, though this is guidance rather than a measured finding, and it will vary with how big and how deep your site's structure already is.
- Build a list of money pages. These are the pages you most want to rank, typically product pages, pricing, comparison pages, or your highest-value content. If you don't already have this list for other reasons, this audit is a reasonable excuse to make one.
- Cross-reference the two lists. Take your money pages and check their link count from step 2. A money page sitting at five or ten internal links, on a site where other pages of similar importance sit at thirty or forty, is under-linked relative to its priority, even if it isn't technically an orphan.
- Keep orphans and under-linked pages as two different fixes. An orphan needs any relevant link at all. An under-linked money page needs more links from pages that already carry weight, not just any link from anywhere. These are two different problems, not one broad problem applied twice.
- Log what you find. A page, its current link count, its click depth, and a one-word verdict (orphan, under-linked, fine) is enough. Add a column for the money pages list too, so the sheet doubles as your priority ranking. You'll use this list directly in the decision rule in the next section.
There is no single correct number of internal links for every page. The right count depends on the page's length, the size of the site, and how important that specific page is. The point of this audit isn't to hit a target number. It's to find the pages where the gap between how important a page is and how linked it actually is runs largest, because those are the pages losing the most from being under-linked.
If clicking through a spreadsheet of URLs one by one is more than you want to do by hand, Semrush's Site Audit flags orphan pages, broken internal links and redirect chains automatically, which is worth setting up before this audit becomes a recurring chore rather than a one-off.
For sites with a genuinely large content footprint, running this same audit logic at scale is also where AI tools for technical SEO audits start to earn their keep, catching orphan and depth problems faster than a manual export can. It's also worth pairing this internal-link audit with a broader look at the content itself. An internal-link audit and a content audit are usually run together, and TRM's content audit services cover the latter.
The Decision Rule For Every New Internal Link
Every guide tells you internal links matter. Almost none of them tell you whether a specific link, from this page to that one, is actually worth adding. Here is the rule, applied one candidate link at a time.
- Does the source page already have some relevance? According to SE Ranking's internal-linking guide, a new page typically gets its first internal links from a small number of already-established, topically relevant pages, rather than from everywhere at once. If the source page ranks for, or at least covers, something genuinely related to the target, it passes this test. If the only connection is that both pages happen to be on the same website, it doesn't.
- Is the target under-linked relative to its priority? Pull the number from your audit. If the target is already well-linked for a page of its importance, one more link adds very little. If it's a money page sitting at five links when it should be at twenty, this link does real work.
- Would the link read naturally in context? A link forced into a sentence that didn't need it reads as forced, to readers and to anyone auditing the page later. If you have to invent a reason for the link to exist in that paragraph, that's your answer.
- If one and two pass, and three doesn't fail outright, add the link. If the source page has no real relevance, skip it, no matter how much the target needs links. A link with no topical reason to exist doesn't pass equity in any meaningful way; it just adds noise.
- Decide the anchor text as a separate step, not an afterthought. Once you've decided to add the link, write anchor text that describes the destination specifically, not a generic "click here" or "learn more," and not the identical phrase you used the last time you linked to this same page from somewhere else.
A quick worked example makes the shape of this clearer than another abstract rule would. Say you've just published a guide on technical SEO audits and you're deciding whether to link it from an older post about site migrations. The migration post already ranks for a handful of technical SEO terms, so step one passes. Your new audit guide is brand new and currently orphaned, so step two passes easily; it needs any relevant link right now. The migration post already discusses crawl budget in a paragraph that naturally leads into audits, so step three passes too. Add the link, with anchor text describing what the audit guide actually covers, not the phrase "this guide."
The rule also works in reverse, when you're deciding whether a page you're updating should link out to something older. A page can only send a limited number of genuinely relevant links before every link starts looking the same to a reader. If three different candidate targets all pass the first three steps, prioritise the one that's most under-linked relative to its own priority first, since that's the link doing the most marginal work.
Hubstaff's content strategy teardown is a useful real example to look at, because it shows a SaaS site making exactly these source-relevance and priority calls deliberately, rather than linking by habit.
Run this rule every time you publish something new, and periodically against your existing under-linked pages from the audit. It will reject most candidate links, which is the point. A page that passes this test for every link pointing to it has a defensible reason for each one. A page that doesn't is the page that ends up over-linked, under-linked, or linked for reasons nobody remembers six months later.
Anchor Text Rules That Actually Hold Up
The anchor text decision from the previous section deserves its own treatment, because most sites get it wrong in the same two ways.
The first mistake is generic anchor text. "Click here," "learn more," "read this" tell a reader nothing about what they're about to land on, and they tell a crawler even less. Descriptive, specific anchor text tells both readers and crawlers what the linked page actually covers, which is reason enough on its own to stop using generic phrases, independent of any SEO argument.
The second mistake runs the opposite direction. Using the exact same anchor text, word for word, every single time a page links to the same target, feels consistent, but it reads as repetitive and, more practically, it throws away the chance to describe the destination from a slightly different angle each time, which is useful for readers arriving from different contexts. Varying the wording while keeping it accurate beats repeating one exact phrase across every link to a given page.
Neither mistake is about keyword density. Exact-match keyword anchors on every single link to a page is not a stronger version of good anchor text; it's the first mistake and the second mistake stacked on top of each other, generic in the sense that it ignores what the surrounding sentence is actually saying, and repetitive in the sense that it never varies.
The practical version of this rule is short. Write the anchor text to describe what a reader will find on the other side of the click, in the words that fit the sentence you're already writing. Do that for every link, and vary the exact phrasing across the handful of times a given page gets linked from elsewhere on the site. If you're applying AI SEO tools for content optimization to anchor-text suggestions at scale, the same two rules still apply. Descriptive and varied beat generic and repeated, regardless of who or what is writing the sentence.
How Many Internal Links Is Too Many?
There is no fixed ceiling on internal links per page. Anyone who gives you an exact number is rounding a guideline into a rule it was never meant to be.
What is real is the failure mode at the other end. Per Google's own guidance, a page carrying far too many links on it is probably overloaded, and an overloaded page dilutes whatever authority it had to pass on, spread thin across dozens of destinations instead of concentrated on the ones that matter. A page with 150 links scattered through it, half of them navigation, the rest crammed into a "related content" block, isn't linking generously. It's linking carelessly.
A commonly cited guideline puts the figure at roughly five to ten internal links per 1,000 words, though no study backs the specific number, and it should be read as a reasonable starting point for a typical blog post rather than a target to hit on every page regardless of length or purpose.
Auto-generated "related posts" widgets and tag-cloud footers are a frequent source of link bloat, because they add links without anyone deciding each one deserves to be there. If your highest-link-count pages are high mostly because of templated blocks rather than deliberate contextual links, that's over-linking even if the raw total looks unremarkable next to the five-to-ten guideline.
The more useful question is the one the decision rule in this guide already answers, link by link. Does each individual link on the page pass the relevance and priority test? A page where every link earned its place through that test will usually land somewhere near the sensible range on its own, without anyone counting. A page padded with links added out of habit or obligation will usually overshoot it, and that's the actual problem, not the raw count.
Fixing Broken Links, Redirect Chains And Nofollow Mistakes
Four problems account for most of the maintenance work an internal-link audit turns up, and each has a specific, boring fix.
Broken internal links and 404s. A link pointing to a page that's been deleted or moved without a redirect sends both readers and crawlers to a dead end. Find these by crawling your own site (most crawlers flag internal 4xx responses by default) and update the link to point at the current URL, or remove it if there's genuinely no replacement.
Redirect chains. A link that points through two or three redirects before landing on the final page still works for a reader, but it wastes crawl budget and can dilute the equity passed through it. Update the link to point directly at the final destination instead of the first hop in the chain.
Accidental nofollow on internal links. Internal links should almost always be dofollow, the default. A nofollow attribute on an internal link, usually added by a plugin, a template, or a well-meaning rule meant for external links that got applied too broadly, blocks that link from passing any equity at all. Audit your templates for this specifically, because it tends to affect every page using that template at once, not just one.
Links that are technically present but uncrawlable. A link implemented as a JavaScript click handler on a <div>, rather than a real anchor tag, can work perfectly for a human visitor and still be invisible to a crawler. Google's crawlable-links guidance is clear that pages need to be reachable through ordinary, crawlable links, not script-only navigation that depends on a click event firing. If a link matters, it needs a real anchor tag underneath whatever interaction you've built on top of it.
None of these four show up in a normal reading of the page. They all show up in a crawl report or the Search Console coverage data, which is the actual reason to run one of these on a schedule rather than only when something feels off. A page that looks fine to a visitor can still be quietly leaking crawl budget and link equity through any one of these four problems, for months, before anyone notices the traffic didn't move the way it should have. Fixing all four is rarely more than an afternoon's work once you know where to look.
Running this audit and decision rule every month works, right up until the site outgrows what one person can track by hand. The GEO + SEO Growth Programme folds internal linking into an ongoing process, so it gets rechecked on a schedule instead of only when someone remembers.
Build Hub-And-Spoke Clusters So Pillar Pages Carry The Right Pages
Audits and decision rules fix individual links. Hub-and-spoke structure is what makes the whole site concentrate its link equity on the pages meant to win competitive terms, instead of spreading it evenly across everything.
The structure is simple to describe and genuinely useful to build. A pillar page covers a broad topic comprehensively. Pillar pages link out to the narrower subtopic pages that go deeper on one piece of that topic, and those subtopic pages link back to the pillar and to each other, forming a hub-and-spoke cluster. Internal linking guidance generally agrees this is how pillar pages are built, linking out to subtopic pages that link back and across, rather than leaving a broad page to stand alone with no supporting content around it.
Why this matters more than it looks like it should. A pillar page on, say, "technical SEO," competing for a broad, competitive term, benefits from every subtopic page underneath it (one on internal linking, one on crawl budget, one on Core Web Vitals) sending a link back up to the pillar. Each subtopic page is narrower and usually easier to rank for on its own. Collectively, they hand the pillar a concentration of internal links that a standalone page, with no supporting cluster, simply doesn't get.
Building one from scratch, or retrofitting it onto content you already have, follows the same pattern either way. Pick the broad topic that deserves a pillar. Identify, or write, the subtopic pages that cover its individual pieces in enough depth that they can stand as their own pages; this guide on internal linking is exactly that kind of subtopic page for a broader technical SEO pillar. Link every subtopic page back to the pillar, using anchor text that describes the pillar's scope rather than a bare "learn more." Link subtopic pages to each other where they genuinely relate, not exhaustively, only where a reader of one would plausibly want the other.
The direction that gets skipped most often is spoke back to pillar, and spoke to spoke. Teams tend to remember to link down from a new pillar page into every subtopic the day it launches, because that happens in one sitting. They forget to go back and add the return link from each subtopic once the pillar exists, because that means editing pages that were already finished and published. The audit step from earlier catches this specifically. A subtopic page with normal link counts but zero links back up to its own pillar is this exact gap, and it's usually a five-minute fix once you've found it.
The audit from earlier in this guide also tells you which pillar candidates are currently underperforming this pattern, a broad page with plenty of narrower content sitting around it on the site, none of which actually links back. That's usually the fastest structural win available, because the content already exists; it's only the links that are missing.
Topic clusters built this way also tend to reinforce the same entities and terminology across every page in the cluster, which is a separate but related piece of work. If that side of it is new to you, entity SEO and knowledge graph tools cover the part of this that's about consistency of meaning across pages, rather than the links between them. Building and maintaining clusters like this at scale, across dozens of pillars rather than one or two, is usually the point where it stops being a weekend project and becomes something closer to an ongoing SEO and AI-visibility programme run on a schedule.
The Publish-Day Checklist
Run this before a new page goes live, every time, regardless of how confident you are that you remembered everything.
- Confirm it isn't an orphan the moment it publishes. A brand-new page should launch with links already pointing into it from a small number of established, relevant pages, not wait for the next audit cycle to get discovered.
- Check the anchor text on every one of those links. Descriptive, specific to this page, not the same phrase copy-pasted from wherever you found a spot to add a link.
- Place it in its cluster, if it has one. If this page is a subtopic of an existing pillar, link it from the pillar and link it back, both directions, on day one rather than eventually.
- Check for a money page it should be supporting. If this new page is relevant to an under-linked priority page from your last audit, this is the moment to add that link too, while you're already editing both pages' context.
- Verify every link actually resolves. A typo in a URL, or a link copied before a page's final slug was set, is an easy broken link to create and an easy one to catch if you check once before publishing.
- Note the click depth from the homepage. If this page sits more than three clicks deep and it's meant to be a priority page, that's worth fixing now, with an additional link from somewhere higher up, rather than waiting to notice it later.
- Flag it for the next audit cycle. Even a page that passes every item above today can drift as the pages around it change. Add it to whatever list or spreadsheet your next audit will run against.
Seven items, five minutes once it's a habit. None of it requires a tool, a budget, or anyone's permission, which is exactly why it's a reasonable thing to run on every single page regardless of team size. It's the same discipline behind organic growth without a big SEO budget. A short repeatable checklist instead of an expensive tool nobody asked for, run consistently. Pair it with a SEO copywriting best practices guide for the writing side of the same publish-day routine, and the two together cover most of what a page needs before it goes live.
When To Automate Internal Linking Instead Of Doing It By Hand
Everything in this guide works by hand on a spreadsheet, up to a point. Past that point, it stops being realistic, and pretending otherwise just means the audit quietly stops happening.
The threshold is less about a hard page count and more about whether one person can still hold the site's structure in their head, or in a spreadsheet they actually keep updated. In practice, that starts breaking down somewhere around a few hundred pages, sooner if the site publishes several new pages a week rather than a few a month. At that point, a manual audit that used to take an afternoon starts taking the better part of a week, and the gap between audits stretches from monthly to "whenever someone has time," which is the same thing as not auditing at all.
Publish frequency matters as much as raw page count. A 150-page site that adds one page a month can run this guide's checklist and audit by hand indefinitely, because nothing changes fast enough to outrun a human checking it. A 150-page site adding ten pages a week outgrows manual tracking at a much smaller total size, because the publish-day checklist alone becomes a full-time job before the page count gets anywhere near four figures.
What changes at that point isn't the logic. The decision rule, the anchor text guidance, the hub-and-spoke structure, none of it stops being correct. What changes is that applying it to every new page and re-checking it against every old one stops being something a person can do reliably alongside everything else on their plate. That's a tooling problem, not a strategy problem, and it's worth treating it as one rather than trying to out-discipline a scale problem.
This guide deliberately doesn't review individual products for that job, because which one fits depends on your CMS, your team size and your budget more than on anything general this guide could say. AI tools for internal linking automation is where that comparison actually belongs, once you've decided the manual version has stopped being realistic for your site.
FAQs
What Is The Difference Between Internal Links And Backlinks?
What Is A Dofollow Link, And Does It Matter For Internal Links?
Do Navigation Links Count The Same As Links Placed In Body Text?
How Often Should I Re-Run An Internal Link Audit?
Can Internal Linking Alone Fix A Page That Is Not Ranking?
Should Every Internal Link Use Exact-Match Keyword Anchor Text?
Does The Number Of Links On A Page Change How Much Equity Each One Passes?
Is A Three-Click Depth Rule The Same For Every Site?
Do Internal Links Help AI Answer Engines The Same Way They Help Google?
If you would rather talk through your site's specific linking problems than run the audit alone, Contact us. We will tell you plainly whether it is a few hours of manual work or something bigger.
📑 Disclosure!
Some links in this article are affiliate links. If you sign up through them, The Rank Masters may earn a commission at no additional cost to you. This does not influence which tools we recommend or what we say about them.





