{{text}}
Share
Internal linking is the set of links connecting the pages of a single site to one another. It is the cheapest SEO lever available, the fastest to fix, and the one nearly everybody lets decay on its own.
This is not about adding links at random. A useful internal link structure answers three questions: which pages Google discovers, where the authority the site has accumulated ends up, and what each page is meant to be about.
The three jobs internal links do
Discovery. A page with no inbound link exists only in the sitemap. It may get crawled, it will rarely get prioritised. This is the classic case of the article that ships and is then never linked from anywhere.
Authority distribution. A site's authority naturally concentrates on the homepage and on the handful of pages that attract external links. Internal linking is the only mechanism you have for pushing it back down towards the pages that need to convert.
Meaning. The anchor and the text around it tell Google what the target page is about. Five links from articles on the same theme, with consistent anchors, describe a subject. Fifty footer links describe nothing.
What Google actually says, and what it does not
Two pages of Google's documentation settle arguments that have run for years.
On link format, the page about crawlable links, last updated on 10 December 2025, leaves no room: Google can generally only crawl a link if it is an HTML a element carrying an href attribute. Links driven by an onclick, by a framework attribute such as routerLink, or by a tag with no href, are not reliable. Google notes it may still attempt to parse them, which is not a guarantee.
On crawl budget, the documentation is equally clear in the other direction: the topic applies to sites above one million unique pages whose content changes weekly, or above ten thousand pages whose content changes daily. For everyone else, Google states plainly that optimising crawl budget is not necessary. In other words, if your site has two hundred pages, stop justifying internal linking decisions with crawl budget. The stakes lie elsewhere.
What Google does not say: the ideal number of links per page, a maximum acceptable click depth, or any rule about anchor text distribution. Any numeric claim on those points is an industry convention, not documented guidance.
The rules that hold
| Rule | Why | How to check |
|---|---|---|
| At least three contextual inbound links on any page meant to rank | A page with one inbound link is treated as secondary | Count inbound links page by page, not outbound ones |
| No internal link pointing at a redirect | Pointless hop, diluted signal, unreadable audits | Crawl the site and filter links whose target returns a 3xx |
| Descriptive anchors, never a click here | Google asks for descriptive, concise, relevant anchor text | Read anchors out of context: is the destination obvious? |
| Links inside body copy, not only in the footer | Contextual links carry meaning, global links carry structure | Count template links and editorial links separately |
| One intent per page, with linking to match | Two pages on the same query starve each other of signals | Find the duplicates, then merge or redirect |
| Links going down and coming back up | The pillar points to its variations, which point back to the pillar | Check that every article in a cluster links to its hub |
The mistakes you find everywhere
We regularly audit sites that produced content for two years without ever touching their internal linking. The same defects come back.
Orphan and near-orphan pages. The most frequent case, and the easiest to fix. An article published, shared once, then never linked to anything. It stays in the sitemap and is supported by nothing.
Links pointing at redirects. After a redesign, old internal links keep pointing at old URLs. Everything works for the user, so nobody notices. On our own site, a recent audit found more than fifty internal links pointing at a redirected URL, a direct inheritance from a change in site structure.
Broken links left in place. A link to an article moved back to draft returns a not-found page. It is invisible in normal browsing, because nobody clicks, and perfectly visible to a crawler.
A global link to a page that no longer exists. The worst scenario, because it multiplies. One element present across the whole site pointing at a deleted page produces as many broken links as there are pages. We have seen a single link of this kind generate errors on more than three hundred pages.
Linking that ignores commercial pages. The blog links to the blog, and the pages that sell never receive anything. Structurally this is the most expensive defect, because it cancels the entire commercial benefit of the content.
Anchors: write for the reader
Google asks for anchor text that is descriptive, reasonably concise and relevant to the target page, and explicitly advises against generic formulas such as click here as well as chains of links stuck together. The practical test is simple: reading the anchor alone, without the sentence around it, do you know where it goes?
The second rule, less well known, concerns context. An anchor placed in the middle of a paragraph about the same subject beats a perfect anchor dropped into a block of links at the end of an article. The surrounding text counts as much as the anchor.
Finally, avoid varying anchors artificially just to avoid repetition. The real problem with an anchor repeated fifty times is not the repetition, it is that it reveals linking applied mechanically rather than written.
Internal linking and cannibalisation are the same subject
When two pages target the same query, internal linking always makes it worse. Links split between the two, neither pulls ahead, and Google picks for you, often the weaker one.
The correct sequence never changes: identify the duplicates, decide which page survives, 301 the other, then repoint every internal link at the surviving page. Skipping the last step is the classic error, and it leaves the redirect sitting in your link graph for years. We cover it in our guide to SEO cannibalisation.
Auditing your internal links in an hour
No expensive tool and no complex methodology required. Four passes are enough to produce an action list.
- Crawl the site and export the URL list with the number of inbound internal links. That is the number that matters, and the one reports surface least.
- Isolate pages with zero or one inbound link. That is your immediate worklist. On a content site running for two years it often exceeds a hundred rows.
- Filter links whose target returns a 3xx or 4xx. These are mechanical fixes with no editorial judgement involved, so they are the quickest to clear.
- Cross-reference with your commercial pages. How many contextual internal links does your main service page receive? If the answer is under five, you have found your priority.
That fourth pass is the one that moves results. It follows the same logic as a technical SEO audit, except that it requires no development skills.
Where to start
In decreasing order of return:
1. Fix broken links and links to redirects. No judgement calls, an immediate gain in clarity, and often a side effect on crawl speed.
2. Link to your commercial pages. Three to five contextual links from your most visited articles, with descriptive anchors. This is the move that turns a blog producing traffic into a blog producing enquiries.
3. Bring orphan pages back into circulation. One inbound link from a relevant article is enough to put them back in the loop.
4. Deal with duplicates. Slower, because it requires a decision, but it is what unblocks pages stuck between eleventh and thirtieth place.
On the Garden Reclaimer engagement, the second point produced the clearest difference: the content already existed, it simply pointed at no conversion page.
A last word on the temptation to chase external links before doing this work. Buying links for a site whose internal structure is incoherent is filling a leaking bucket. The correct order is set out in our article on link building: clean up the inside, then reinforce the outside.
Frequently asked questions about internal linking
What is internal linking?
The set of links connecting the pages of a single site. It decides what Google discovers, how authority flows from one page to another, and what subject each page is understood to cover.
How many internal links should a page have?
There is no official number. The better question is the reverse: how many inbound links does each page receive? A page with a single inbound link is one you have implicitly declared secondary. We aim for at least three contextual inbound links on any page meant to rank.
Does Google follow every kind of link?
No. Its documentation, last updated on 10 December 2025, is explicit: Google can generally only crawl a link if it is an HTML a element with an href attribute. Links carried by a JavaScript event, a framework attribute or a tag without href are not reliable.
Is internal linking about saving crawl budget?
For almost every site, no, and the confusion is widespread. Google reserves the crawl budget topic for sites above one million unique pages, or above ten thousand pages changing daily. Below that, internal linking is about discovery and hierarchy, not about saving resources.
Should anchor text be varied?
Yes, but not for the reason people think. Google asks for anchor text that is descriptive, reasonably concise and relevant to the target page. Repeating the exact same anchor across fifty links is not penalised as such, but it signals links written for an algorithm rather than a reader, and it produces stilted sentences.
Is a navigation link worth the same as an in-text link?
They do different jobs. Navigation handles discovery and overall structure; the contextual link carries meaning, because it sits inside text about the subject. A page linked only from the footer is weakly supported, even though it is technically reachable.
What should you do about links pointing to a redirect?
Fix them at source. An internal link to a redirecting URL works for the user but adds a pointless hop, dilutes the signal and makes audits unreadable. It is one of the most common defects after a website redesign.
How do you choose which pages to support first?
The ones that earn. In practice, commercial pages and any page ranking between eleventh and thirtieth, which is within reach of page one. Our article on SEO KPIs covers how to spot them.
Synqro is a Webflow Premium Partner agency in Paris. Want to know how many internal links your commercial pages actually receive? It is the first thing we look at in an audit. A Webflow expert can run that check with you. See how we work or request an audit.
Synqro's latest achievement
