Internal Links for SEO: How to Use Them (and Get Cited by AI)
- Level: Intermediate
- #SEO
- #On-page SEO
- #Content

An internal link is any link from one page of your site to another page on the same site. It helps readers keep browsing and helps Google discover pages and understand how they relate. To work, the link must be crawlable (<a href>), carry a descriptive anchor and start from relevant pages. A page that no other page links to is an "orphan" and is less likely to be found.
This guide is part of the path that starts with What is SEO: the complete guide. If terms like crawling and indexing still sound unfamiliar, start there and come back.
Quick summary
- Google says every important page should have at least one link coming from another page on the site.
- Only a link built with an
<a>element and anhrefattribute is crawlable. Buttons,onclickhandlers androuterLinkwithouthrefdo not count. - A good anchor is descriptive, short and relevant. "Click here" tells nobody anything.
- There is no official number of links per page. The criterion is helping the reader.
- For AI, the effect is indirect: internal links help a page get found and understood, but there is no public proof that more links lead to more citations.
How to read this guide
If you are new to SEO, read the sections on what internal links, anchors and orphan pages are, and follow the audit steps. If you already work in SEO, jump to click depth, the pillar and cluster structure and the AI section, where I separate what is documented from what is inference. Technical terms are in the SEO, AEO and GEO glossary.
What are internal links and why does Google care?
They are the links that connect the pages of your site to each other: menu, footer, links inside the text, breadcrumbs. Google follows them to discover new pages and uses the link text to understand what the destination page is about. In its documentation on crawlable links, the guidance is direct: every page you consider important should receive a link from at least one other page.
In plain English: think of your site as a building. Each page is a room and each internal link is a door. A room with no door exists, but nobody gets there, not the visitor and not Google's robot.
Google's SEO starter guide adds that link text tells both users and Google something about the destination page. That is why the anchor matters.
For SEO professionals: a sitemap helps discovery, but Google itself warns it does not guarantee crawling or indexing. An internal link does what a sitemap cannot: it carries context (anchor and source page) and shows which pages you treat as central. Google does not publish weights for internal links, so any "authority distribution formula" is a working model, not a documented rule.
What makes an internal link crawlable?
Google only crawls links built with the <a> element and the href attribute. The documentation lists as non-crawlable the links that rely only on onclick or routerLink, and elements such as span. It sounds like a detail, but on JavaScript-built sites it is the most common cause of "invisible" pages.
<!-- Crawlable -->
<a href="/blog/topic-cluster-guide">build a topic cluster</a>
<!-- Not crawlable by Google -->
<span onclick="location.href='/blog/topic-cluster-guide'">build a topic cluster</span>
<a routerLink="/blog/topic-cluster-guide">build a topic cluster</a>
| How the link was built | Can Google follow it? |
|---|---|
<a href="/page">text</a> | Yes |
Image inside <a href> with a descriptive alt | Yes, and the alt works as the anchor |
<span onclick="..."> | No |
<a routerLink="..."> without href | No |
For SEO professionals: frameworks such as Next.js, React Router and Nuxt ship link components that render a real
<a href>. The problem appears when someone swaps that component for a button withonClick. Check the rendered HTML of a listing page (the URL Inspection live test in Search Console shows it) and confirm the links appear as<a href>.
How do you write a good anchor?
A good anchor is descriptive, reasonably short and relevant to the destination page. That is what Google recommends, and it is what helps readers: they know where they are going before they click.
| Avoid | Prefer |
|---|---|
| click here | on-page SEO checklist |
| learn more | how to measure the return on backlinks |
| https://yoursite.com/blog/post | Search Console guide for beginners |
| the same exact keyword in every link | natural variations: "local SEO", "show up on Google Maps", "business profile" |
Google does not document a limit or a penalty for exact-keyword internal anchors. Even so, mechanically repeating the same phrase across dozens of links makes the text feel artificial to readers, and that is the criterion that matters. For the logic of anchors on links from other sites, see the anchor text guide.
How many internal links should each page have?
There is no official number. In Google's material on links, the criterion is usefulness: link when the destination helps the reader at that point in the text. As a rule of thumb of ours (not Google's), a 1,500 to 2,000 word article usually holds 4 to 10 contextual internal links, without repeating the same destination more than twice.
What matters most is the two-way street. When you publish a new post, follow these steps:
- Link early to the page that summarizes the topic (the "pillar", explained below).
- Link to 3 to 5 related posts, in sentences where the link answers a natural question from the reader.
- Go back to 2 or 3 older posts on nearby topics and add a link to the new one. Without this step, the new post is born almost orphaned.
- Confirm that the anchor describes the destination and that the link opens a page that answers what the anchor promises.
- Log the relationship (source, destination, anchor) in a spreadsheet so you do not repeat yourself by accident.
How do you find orphan pages and broken links?
Compare the list of URLs that exist with the list of URLs a robot reaches by following links from the home page. The difference is your orphans. This audit takes an afternoon on sites of up to a few hundred pages.
- List the URLs that should exist: export the sitemap or the CMS list.
- Crawl the site like Google does: a crawler such as Screaming Frog crawls up to 500 URLs at a time in the free version. Start from the home page and follow links only.
- Compare the two lists: a URL that is in the sitemap but did not show up in the crawl is an orphan.
- Decide case by case: if the page matters, link to it from a relevant place; if it is redundant, merge it with another; if it should not appear in search, treat it as
noindexor remove it. - Fix links that return a 404 error or go through a redirect: replace the link with the final destination instead of relying on the redirect.
The Search Console Links report helps too: it lists the most internally linked pages. But Google itself warns that it shows a sample, limited to 1,000 rows per table, and does not indicate whether a link is nofollow. Use it as a clue, not as an inventory. If you are just starting with Search Console, the beginner's guide shows where these reports live.
What is the ideal click depth from the home page?
The fewer clicks separate a page from the home page, the easier it is for Google to see it as important. In 2018, John Mueller said in a conversation with webmasters that click distance from the home page matters, and that the number of slashes in the URL does not. The source is the Search Engine Journal report on that session; it is an informal and old statement, not documentation, so treat it as a hint.
The "3 clicks" rule is an industry heuristic, not Google's. What holds is this: pages that make money or that you want to rank should sit close to the home page or to strong pages.
For SEO professionals: click depth is different from folder depth.
/blog/guide/may be one click from the home page and/guide/may be six away if it sits behind pagination, filters or infinite scroll without crawlable links. In e-commerce and portals, check pagination and facets before touching URLs.
How do you structure internal links with a pillar and cluster pages?
A pillar and cluster structure is a group of pages on the same subject: one broad page (the pillar) and several specific ones (the cluster pages, also called satellites), all linked to each other. The pillar links to every cluster page, and each cluster page links back to the pillar. See the full walkthrough in how to build a topic cluster.
Hypothetical example: an online specialty coffee store builds a cluster on "specialty coffee for beginners".
| From | To | Suggested anchor |
|---|---|---|
| Pillar: specialty coffee guide | Cluster page: arabica vs robusta | difference between arabica and robusta |
| Pillar: specialty coffee guide | Cluster page: grind size by method | the right grind for each brewing method |
| Cluster page: arabica vs robusta | Pillar | specialty coffee guide for beginners |
| Cluster page: grind size by method | Cluster page: how to store coffee | how to store ground coffee |
| All cluster pages | Product page, only when there is buying intent | whole bean coffee for drip brewers |
Notice that the last link points to a product. An informational page should not turn into a storefront, but it can offer the next step when that makes sense.
Do internal links help you get cited by AI?
There is no public proof that more internal links produce more citations in ChatGPT, Perplexity or Google AI Overviews. What exists is a chain of requirements and a reasonable inference. According to Google, for a page to show up as a supporting link in AI Overviews or AI Mode it must be indexed and eligible to appear with a snippet, with no special optimization. Internal links take part in that chain because they are one of the paths through which a page is discovered.
Google also says these features may use "query fan-out", which means firing several related searches to build the answer. If your site has linked pages that cover the subtopics of a subject, there is a better chance that one of them matches one of those searches. That is inference, not a guarantee. How AI products from other companies choose sources is not known.
In plain English: an internal link is the index at the back of a book. It is not what gets the book cited, but without it nobody finds the right chapter. What gets you cited is a clear, trustworthy chapter. See how to build that in content that LLMs can cite.
How does Contentor help with internal links?
Contentor's article generation uses templates with validated structure and works with internal links, publishing straight to WordPress. The review is still yours: the person who knows the site decides whether the anchor is right and whether the destination deserves the click. To see how it works in practice, generate an article on Contentor and check the links before publishing.
Frequently asked questions
Do internal links help SEO?
Yes, in two documented ways: they help Google discover pages (every important page should receive at least one link) and they give context through the anchor. Google does not disclose the exact weight of each link, so do not promise to "climb positions" just by adding links. The safest gain is getting important pages out of orphan status.
How many internal links should I put on each page?
There is no official number. Link where the destination helps the reader. As a rule of thumb, 4 to 10 contextual links in an article of 1,500 to 2,000 words, with at most two links to the same destination. If a paragraph starts to look like a list of links, cut the least useful ones.
Should I use nofollow on internal links?
Generally, no. Google advises using nofollow for links you do not want to endorse, such as untrusted sources, sponsored for paid links and ugc for user-generated content. Internal links are yours: if you do not want a page to be found, the way is noindex or removing it, not marking the link.
Can I repeat the same anchor in several links?
You can, with judgment. If two different links use the same anchor for different destinations, the text stops being descriptive. For the same destination, repeating is natural, as long as it does not become a mechanical formula. Vary it when the sentence calls for another way of saying the same thing.
What is an orphan page and how do I fix it?
It is a published page with no internal link pointing to it. To fix it, find it by comparing the sitemap with a crawl of the site, and add links from related, strong pages. If the page has no purpose, merge its content into another page or remove it.
Next step
With links in place, the next step is organizing them into groups: build a topic cluster, with a pillar and cluster pages that complement each other. Once the site has some volume, run an SEO audit step by step to find what was left behind.
Read also:
Prepare your content for AI answers
Create AEO/GEO briefs, entities, schema, and structured articles for Google AI Overviews, AI Mode, and traditional SEO.
Try Contentor freeStrengthen your authority for AI Search
Combine source-worthy content with contextual backlinks on vetted sites to reinforce trust, discovery, and topical authority.
View contextual backlinks


