Internal linking: how to get authority to the pages that need it

Internal links are the clearest signal you fully control. External links are earned or negotiated. Your own links are a decision you make, and they tell search engines which of your pages you consider important.

Most sites get this wrong in the same two ways: the pages that matter most commercially have the fewest links pointing at them, and the pages nobody needs have hundreds because they sit in the main navigation.

Click depth is the number to watch

Click depth is how many clicks it takes to reach a page from your homepage.

It matters because crawlers work outward from the pages they already know. A page three clicks deep gets visited often. A page seven clicks deep gets visited rarely, and on a large site may effectively never be revisited after the first time.

The rule of thumb that holds up well: every page you care about should be reachable within three clicks of the homepage. Not because three is magic, but because if a page needs more than that, it usually means no section of your site is claiming it.

To check yours, crawl your own site and sort by depth. Anything commercially important sitting at depth four or more needs a link from somewhere shallower, and the honest question to ask is which existing page should have been linking to it all along.

Find the orphans

An orphan page has no internal links at all. It exists, it may be in your sitemap, and almost nothing points at it.

This happens more than people expect, and usually for boring reasons: a page was added outside the normal navigation, a section was redesigned and old links were dropped, or a page was built for a campaign and never wired into the site.

To find them, compare two lists:

1. Every URL in your sitemap.
2. Every URL your crawler reaches by following links from the homepage.

Anything in list one but not list two is an orphan. That difference is one of the highest-value SEO reports you can run, because the fix is cheap and the pages are usually ones you already paid to produce.

Navigation links and editorial links are not the same thing

A link in your header or footer appears on every page. It is excellent for discovery: a crawler will find the target quickly. It is weak as a relevance signal, precisely because it is everywhere and therefore says nothing about the page it appears on.

An editorial link sits inside the body of a page, surrounded by related text, placed because a reader at that point might genuinely want it. That context is the signal. It says: this page is about X, and the page I am linking to is a relevant part of X.

The practical consequence is that adding a page to your footer is not the same as linking to it. If a page matters, it needs body links from pages about the same subject, not just a footer slot.

It also means the fix for an orphan page is not a block of links dumped at the bottom of the homepage. That is the same weak signal with extra steps, and it reads as spam to a person. Find the page that should have mentioned it, and mention it there.

Write anchor text that says something

Anchor text is a description of the destination. Treat it as one.

• Bad: “click here”, “read more”, “this page”
• Better: “our commercial cleaning rates”
• Also good: “how often an office should be cleaned”

Vary the phrasing across links to the same page. Five links reading “commercial cleaning Sydney” look manufactured; five links using natural variations tell search engines about the whole cluster of phrases the page should rank for.

One caveat worth knowing: when a page has several links pointing at the same destination, search engines typically pay most attention to the first one. So if you link to the same page twice, put the descriptive anchor first and the casual one second, not the other way round.

Watch for your own pages competing

Cannibalisation is when two of your pages target the same query. Google picks one, often not the one you would have picked, and your internal links get split between them.

Symptoms are easy to spot once you know them:

• Two URLs swapping positions for the same query week to week
• A page ranking for a query that a different, better page was written for
• Impressions spread thinly across several near-identical URLs

Internal linking both causes and cures this. It causes it when you write two similar pages and link to them interchangeably. It cures it when you pick the one that should win, point every relevant internal link at it, and either merge the loser into it or narrow the loser to a genuinely different question.

A practical order of work

1. Crawl your site and export every URL with its click depth and inbound internal link count.
2. List the orphans by diffing that against your sitemap.
3. Rank your pages commercially. Which ten actually earn money?
4. Compare the two lists. Any commercially important page with few inbound links or a depth above three is your work queue.
5. For each one, find two or three existing pages whose content genuinely relates, and add a body link with descriptive anchor text.
6. Re-crawl in a month and check the depth and link counts moved.

Argrow runs steps one, two and four as part of the weekly technical audit and drafts the specific link additions for you to approve, which mostly saves you the spreadsheet. The judgement in step three is still yours, and it should be: only you know which pages pay the bills.