Internal Linking for SEO: How to Build a Site Structure That Ranks

The links between your own pages are often the cheapest SEO lever you have, and also the most neglected one. In this guide I explain internal linking the way I have seen it work in the field since 2012. We will cover anchor text, the hub and spoke model, orphan pages and click depth. You will also get an audit routine and a step by step plan you can apply to your site this week.

What is internal linking in SEO and why does it matter?

Internal linking is the deliberate practice of connecting pages on the same domain with hyperlinks. Google uses these links to discover new pages, to understand how topics relate to each other and to see which pages you consider important. Visitors also use the same links to find their next step on your site.

If I had to sum up its value in one line, I would say this: internal linking lets you decide where traffic and authority flow inside your own website. Earning external backlinks can take months, and the outcome largely depends on other people. Internal links, by contrast, are entirely under your control. A link you add today can be picked up on Google's next crawl.

Here is the pattern I see most often. A company launches a new service page and writes solid copy for it. However, almost nothing else on the site links to that page, so it stays weak for months. With three or four well placed contextual links, the same page usually recovers much faster. That is why internal linking is one of the first things I look at in any SEO consulting engagement.

How are internal links different from external links?

An internal link points from one page on your site to another page on the same site. An external link either points to a different domain or comes to you from a different domain. We usually call the incoming ones backlinks.

In short, the key difference is control. You can only influence the number and quality of your backlinks indirectly. You publish good content, you build relationships and then you wait. With internal links, you decide which page links to which, and with what words. In other words, backlinks determine how much strength reaches your site, while internal links determine how that strength spreads across it.

Put simply, think of it like a city. Backlinks are the highways leading into town, and internal links are the streets inside it. A wide highway does not help much if the local streets are blocked. In practice, when a site has a strong homepage but weak inner pages, the problem is usually in the streets.

How does Google use internal links?

In its link best practices documentation, Google states that it uses links for two purposes. First, links help it find new pages to crawl. Second, they act as a signal when Google determines how relevant a page is. So an internal link is both a road and a label.

The same document sets an important technical condition. Google can only reliably crawl a link if it is an a HTML element with an href attribute. Buttons that rely on JavaScript event handlers, or clickable areas without an href, do not meet that condition. For example, some single page applications route menu items purely through JavaScript. Visitors will not notice, but a crawler may never see that path.

The sentence I find most useful in that guide is simple. Every page you care about should have a link from at least one other page on your site. It sounds obvious. Still, on a large share of the sites I audit, some pages fail this basic test. We call them orphan pages, and I cover them in their own section below.

Also keep in mind that Google reads the text around a link as context. Instead of squeezing a link into an unrelated paragraph, place it inside a sentence that actually discusses the topic.

How should you write anchor text?

Anchor text is the visible, clickable part of a link. Google's description is short: good anchor text is descriptive, reasonably concise and relevant. Put simply, readers should know where a link leads before they click it.

These are the rules I follow in practice:

• Avoid "click here" and "read more". They say nothing about the target page.
• Describe the destination. A phrase like "our custom web design service" gives both readers and Google a clear signal.
• Keep the anchor inside the sentence. It should make sense out of context without breaking the natural flow.
• Do not stuff keywords. Google reminds us in the same document that keyword stuffing violates its spam policies.
• Vary your wording. Using the exact same phrase in every link to one page looks unnatural.

Exact match anchors are risky for external backlinks. With internal links, however, you have more freedom. That said, my own test never changes. If I read the sentence aloud to a client, would it sound odd? If so, I rewrite it.

How do you build a hub and spoke internal linking structure?

The hub and spoke model connects one broad hub page with several supporting pages that cover subtopics in depth. The hub links out to every spoke. Each spoke links back to the hub and, where it makes sense, to neighbouring spokes.

For example, picture a company selling project management software. The hub could be a page about "project management software for agencies". The spokes might cover time tracking, client approvals, resource planning and invoicing integrations. As a result, Google sees a site that covers the topic as a coherent whole rather than as scattered posts.

This is the order I follow when I set it up:

1. Pick the topic with the highest commercial value and choose a hub page for it.
2. List the questions and subtopics around it, since each one is a spoke candidate. Solid research into keywords that drive sales helps a lot here.
3. Add a descriptive link to the hub early in every spoke.
4. Introduce the spokes on the hub with a short summary list or brief paragraphs.
5. Connect spokes that genuinely complement each other, but do not link every spoke to every other spoke.

For service businesses with a blog, this internal linking pattern is the fastest way I know to turn scattered content into a clear topical cluster.

What is an orphan page and how do you find one?

An orphan page is a page that receives no internal links from anywhere on your site. It is live and may even sit in your sitemap. However, no menu, category or article leads to it, so visitors cannot reach it by browsing.

Orphans usually appear in one of three ways. The first source is campaign landing pages that someone removed from the menu after the campaign ended. Next, there is old content that survived a redesign but never made it into the new navigation. Finally, there are older blog posts that lost their links after changes to pagination or filters.

To find them, you then compare two lists. The first list holds every URL that exists on your site, taken from your XML sitemap, a CMS export or server logs. The second list holds the URLs a crawler reaches by following links alone. Any address on the first list that is missing from the second is an orphan candidate. If your sitemap is outdated, our XML sitemap generator can produce a fresh one quickly.

Once you have the list, the decision is therefore simple. If the page has value, link to it from relevant places. If it has no value and no traffic, remove it or redirect it to the closest current page. Leaving it orphaned is never a good option.

Why does click depth matter?

Click depth is the minimum number of clicks needed to reach a page from the homepage. For example, the homepage has a depth of zero, and a page in the main menu usually has a depth of one. A product inside a category might sit two or three clicks away.

Why does this matter? Because on most sites the homepage attracts the most backlinks, and that strength flows downward through links. Every extra layer makes it harder for that flow to reach a page. Visitors also run out of patience long before they reach deeply buried pages.

I use a starting benchmark based on field experience, not a guarantee. Aim to keep commercially important pages within three clicks of the homepage. When I find a service or category page buried deeper, I look for a way to bring it forward.

In practice, the usual fixes are straightforward. Add a featured categories block to the homepage, link subcategories from category pages, link from blog posts to relevant services and use a sensible breadcrumb trail. Also avoid endless "next page" chains in paginated archives, because older posts in a long archive end up very deep.

What role do category pages play in your link structure?

On e-commerce stores and large content sites, category pages form the backbone of the internal link structure. They receive strength from the navigation and pass it on to subcategories, products and articles. Consequently, a poorly built category page drags down everything beneath it.

Specifically, I want to see three things on a good category page. First, plain text links to its subcategories. Second, a sort order that puts best sellers or the most searched products up front. Third, a short introduction about the category itself, with links to related guides that give both readers and search engines useful context.

How to design the category hierarchy itself is a separate and bigger topic. I cover it in detail in my article on category structure for large websites, so I will not repeat it here. In short, if the category structure is the skeleton, internal links are its nervous system. Unless you plan both together, one cannot make up for the weaknesses of the other.

On the other hand, watch your filter pages. When every colour, size and price filter multiplies into crawlable links, crawl resources get wasted and the signal to your real category pages gets diluted.

How do you link from blog posts to service pages?

For service businesses, this is usually where the biggest opportunity lies. Blog posts attract people looking for information, and they often earn more backlinks than service pages do. Yet if those posts never link to a service page, the attention they earn fades before it turns into revenue.

In practice, my method is simple. For every blog post I define a single primary target service page. Then I place a natural link at the point in the article where a reader might genuinely need that service. For instance, in an article about site speed, it makes sense to point readers with a slow theme to a web design service page.

A few rules I stick to:

• Do not rely only on a button at the end of the post. In-text links get more clicks and carry more context.
• Do not link to the same service page in every paragraph. One or two strong placements are enough.
• Do not turn an informational article into a sales brochure. If the reader feels pushed, both the post and the link stop working.
• Link back from the service page to relevant guides, so readers in the decision stage can find supporting information.

This two way structure is where your internal linking strategy connects directly to sales.

How useful are menu, footer and breadcrumb links?

Above all, these three areas repeat on every page, which makes them sitewide links. They are useful, but they have limits. Because menu and footer links look the same on every page, they cannot explain on their own why one page matters more than another.

For the menu, therefore, my advice is restraint. Putting everything in the menu means highlighting nothing. Main services and main categories belong there. Second and third level pages should instead earn links from category pages and from content.

The footer works well for legal pages, contact details and a handful of key pages. Still, lining up hundreds of city or keyword links in the footer adds nothing for users and looks spammy.

Breadcrumbs show visitors where they are and give every page a consistent link to its parent categories. In addition, if you mark them up according to Google's breadcrumb structured data documentation, the page's position in your site can appear more clearly in search results. In short, breadcrumbs are an internal link layer that pays off well for little effort.

How many internal links should a page have?

There is no official rule with a fixed number. Google's link documentation puts it in practical terms: if you think it's too much, then it probably is. So the measure is not a count. It is whether each link actually helps the reader.

Here is my starting range, based on field experience and not a guarantee. In a blog post of about 1,500 words, four to eight in-text internal links usually read comfortably. In comprehensive guides of 3,000 words, that can rise to between eight and twelve. On service pages I prefer fewer, more carefully chosen links.

Moreover, three things matter more than the number. Each link needs a reason to exist. Three or four links crammed into one paragraph make reading harder. And do not expect extra benefit from repeating the same link on one page. Even the Search Console links report combines duplicate links from the same source to the same target.

Therefore, I suggest replacing "how many links?" with a better question. Where will the person reading this page want to go next? The answer usually produces the right number on its own.

Should you use nofollow on internal links?

My short answer is no, because nofollow has a different purpose. The nofollow attribute exists for external links you do not trust and for paid relationships. Using it between your own pages is a bit like telling Google you do not trust your own site.

For instance, years ago some SEOs practised what they called PageRank sculpting. They added nofollow to internal links pointing at minor pages, hoping to push more strength toward important ones. That tactic stopped working a long time ago, and today it only adds clutter.

So what about login, cart or account pages that you do not want in search results? The right tools there are different. Add a noindex tag to pages you want kept out of results, and block areas you never want crawled with robots.txt. Leave the links themselves as normal links. That way the visitor experience stays intact, and Google receives a consistent message.

The one exception is areas where users can add content. For example, you can mark links in forums or comment sections with the ugc attribute. But that is about managing user content, not about your internal link plan.

How do you audit internal linking on an existing site?

An internal linking audit starts with mapping your site's links and spotting strong and weak points. Your free starting point is Google Search Console. Its Links report includes a table of the pages most linked from within your own site, and you can drill into each one to see which pages link to it.

Keep two limits in mind when you read that report. According to the Search Console help page, tables show at most 1,000 rows, and the report is a sample rather than a complete list. For that reason, I recommend a full crawl with a desktop crawler on larger sites.

During an audit I work through these questions in order:

1. Are the pages with the most internal links really my most important pages?
2. How many clicks deep are the service and category pages that drive revenue?
3. Which pages receive no internal links at all?
4. Are there broken links or redirect chains?
5. Is the anchor text descriptive, or full of empty phrases like "click here" and "details"?

On most sites, the answers to these five questions already produce the first month's task list. If you put the results in a simple spreadsheet, you can also compare progress month by month.

Internal link types compared

As a rule, different kinds of internal links do different jobs. The table below shows at a glance what each type is good for and where it reaches its limits.

The takeaway is clear: no single type is enough. Menus and breadcrumbs build the skeleton, while in-text links carry the real meaning.

Common internal linking mistakes

Over the years I have seen the same mistakes on site after site. The good news is that most of them take only a few hours to fix.

• Few links to key pages, many links to minor ones. For example, the privacy policy gets a link from every page while the main service page only sits in the menu.
• Broken links. Old links to deleted or moved pages lead both visitors and crawlers into dead ends.
• Redirect chains. A link goes to an old URL, then to a new one, then to the HTTPS version.
• Empty anchor text. Words like "click here", "here" or "details" carry no context.
• JavaScript only links. Clickable elements without an href are not a reliable path for Google.
• The same anchor pointing to two different pages. This sets up two pages to compete for one query.

The last point matters most, because internal linking and keyword planning meet right there. If you have not decided which page is the main page for a keyword, your internal links will stay undecided too. So settle the keyword to page mapping first, then link according to it.

How do redirects and broken links weaken internal links?

When an internal link points to a redirected URL, the visitor still ends up on the right page. However, each redirect adds another request, another wait and another step for the crawler. The longer the chain, the slower the page loads and the more crawl resources go to waste.

Broken links, on the other hand, are a clearer loss. A link to a page returning 404 passes value nowhere and gives visitors a poor experience. This problem grows quickly after migrations and redesigns.

The fix is simple, but it also needs discipline. Whenever you set up a redirect, also update the internal links so they point straight to the new URL. Treat redirects as a safety net for old external links and bookmarks only. To check individual URLs, use our redirect checker, which shows where each URL ends up and in how many hops.

For a migration, my website migration SEO checklist helps you avoid skipping these steps. For a redesign, read my guide on protecting SEO during a website redesign. Choosing short, permanent URLs for new pages also reduces the need for redirects later, and a slug generator makes that easier.

An internal linking plan, step by step

Let me turn everything above into a plan you can use. When I start with a new client, my internal linking work follows roughly this order:

1. Build a priority list. Identify the 10 to 20 pages that drive sales or leads. They are not all equally important.
2. Measure the current state. Note how many internal links each page gets, how deep it sits and which anchors point to it.
3. Find source pages. List related articles that already get traffic and backlinks. These are the pages you will link from.
4. Add the links. Give every priority page at least three or four contextual links, with varied anchors.
5. Clean up orphans and broken links. Connect or remove orphan pages and fix broken links.
6. Set a rule for new content. Every new post should link to its hub and receive at least one link from an older post.
7. Track results. Watch impressions and average position for the target pages in Search Console over the following weeks.

Step six turns internal linking from a one-off project into a habit. Ask your team to add it to the publishing checklist. Otherwise, new orphan pages will appear within a few months.

How do you measure the impact of internal links?

It is hard to isolate the impact of internal links with a single metric. Content changes, new backlinks and algorithm updates all happen during the same period. Still, a few indicators show the direction clearly.

First, my main indicator is impressions and average position for the target pages in Search Console. Note the date you added links, then compare those two values over the following weeks. My second indicator lives in GA4: do readers actually move from the source articles to the target pages?

Additionally, the third indicator is on the crawling side. If newly published pages get indexed faster than before, your link structure is probably speeding up discovery. The URL Inspection tool in Search Console also gives hints about how Google found a page.

One caveat. Internal linking is not magic. You cannot push a thin page to the top just by adding links to it. What it does is help a good page reach the visibility it deserves sooner, and in my experience that is where its real value lies.

Do internal links matter for AI search?

There is no documented "AI internal link factor". Yet the indirect role seems clear to me. Systems such as ChatGPT, Gemini and Perplexity rely on their own crawlers or on search indexes when they build answers. Those crawlers also discover pages largely by following links.

As a result, a well planned link structure makes it easier for AI systems to see how thoroughly your site covers a topic. A cluster of consistent, interlinked pages around a hub reinforces the impression that your brand knows the subject in depth. I explain the wider picture in my article on generative engine optimization.

There is also a technical angle. Many AI crawlers process JavaScript in a limited way or not at all. That makes it even more important than in classic search to have your links in plain, server rendered HTML. I go deeper into this in my article on technical SEO after AI.

In short, internal linking is not becoming obsolete in the AI era. On the contrary, it remains one of the basic conditions for a site that machines can understand.

Where should you start today?

Do not try to fix everything at once. My advice is to do just three things this week. First, write down your five most important pages and check in Search Console how many internal links each one gets. Next, pick the one with the fewest links and add in-text links to it from three related articles. Finally, search your site for empty anchors like "click here" and "read more", and replace them with descriptive wording.

Together, these three steps take a few hours. On most sites, they create noticeable movement within a few weeks. After that come the orphan page scan, the hub and spoke structure and the rule for new content.

If you would like to map your site's internal links together and build a plan for your priority pages, you can reach me through my SEO consulting page. I work directly with clients, without intermediaries, and I prepare both the audit and the link plan myself.