Internal linking: strategy and architecture for SEO

Why internal linking matters for SEO

Internal linking is the system of links between pages on the same site. It looks like simple user navigation on the surface. In reality it's one of the few tools you fully control, and it directly affects three things: crawling, indexing, and rankings.

Googlebot crawls a site by following links. A page with no incoming internal links is invisible to the crawler (unless it's in the sitemap). A page with a single link gets little link equity. A page that dozens of pages link to with relevant anchors ranks significantly better.

zoom_in

Internal linking by the numbers

PageRank and link equity flow

PageRank is Google's algorithm for assessing page authority based on the quantity and quality of inbound links. External links bring link equity from outside; internal links distribute it between pages. Smart internal linking directs PageRank to the right pages.

The principle is simple: every page has a certain PageRank. When it links to another page, it passes a portion of its equity. The fewer outgoing links a page has, the more equity each one passes. The homepage typically has the highest PageRank — so a link from it is more valuable than one from a deep internal page.

Links with rel="nofollow" don't pass PageRank. This is useful for unnecessary pages (login pages, filters, duplicate content) to avoid wasting link equity. However, overusing nofollow internally hurts the structure — Google still crawls such pages, just without equity.

Site architecture: depth and structure

Click depth is the number of clicks from the homepage to a page. Pages at depth 1–2 receive more crawling, more PageRank, and generally rank better. Pages deeper than 4 clicks are often crawled less frequently or not at all.

Site architecture types

For large e-commerce or content sites, silo architecture is optimal: the site is divided into topical clusters (silos), pages within a cluster actively link to each other, and clusters are connected through hub pages. This concentrates topical authority and simplifies crawling.

Example silo architecture for an SEO blog:

Homepage
├── /blog (hub)
│   ├── /blog/technical-seo (sub-hub)
│   │   ├── /blog/crawl-budget  ←─┐
│   │   ├── /blog/robots-txt    ←─┤ cross-links
│   │   └── /blog/xml-sitemap   ←─┘ within silo
│   └── /blog/on-page-seo (sub-hub)
│       ├── /blog/h1-tag        ←─┐
│       ├── /blog/meta-description ←─┤ cross-links
│       └── /blog/title-tag     ←─┘ within silo

Anchor text for internal links

An anchor is the text a user sees as a link. Google uses anchor text as a relevance signal: if a page about 'server log analysis' is linked to with the anchor 'crawler log analysis,' Google understands the destination page is related to that topic. Internal links with exact anchors strengthen rankings for target keywords.

• Descriptive anchor — best option: 'how to configure robots.txt,' 'crawl budget guide.' Helps Google understand the destination page's topic
• Partial match — fine: 'load optimisation' instead of 'how to optimise loading speed.' More natural and doesn't look like over-optimisation
• Bare URL — poor anchor: 'https://example.com/crawl-budget.' Use only for technical links (canonical etc.)
• 'Here,' 'more,' 'read' — useless for SEO: carry no information about the destination page. Acceptable in CTAs but not in content links
• Exact-match keyword anchor — use sparingly: if all links to a page share the identical anchor, it looks unnatural

Internal linking patterns

There's no single correct way to link — there are patterns that work in different contexts. Combining several patterns gives the best result.

zoom_in

Core patterns

• Breadcrumbs: show site hierarchy, create upward links to categories and homepage. Automatically solve the linking problem on large sites. Google explicitly recommends them.
• Related posts / Similar content: a block of links at the end of an article or page to similar content within the silo. Keeps users engaged and passes link equity within the cluster.
• In-text contextual links: most valuable from an SEO perspective — the anchor is embedded in a relevant context. Google treats surrounding text as an additional relevance signal.
• Navigation menus: provide links to key pages from across the site. Header and footer links get maximum reach, but for that reason each passes the least equity per link.
• Hub pages: a category page or overview article linking to all content on the topic. Collects PageRank and redistributes it across the cluster's content.
• Links to pillar articles: from each cluster article — a link to the main topic piece (pillar page). Forms a topic cluster and strengthens the main topic page.

Orphan pages: find and fix

An orphan page is a page with no incoming internal links. Googlebot may not find it at all (or will only find it via the sitemap, but without link equity). Such a page has almost no chance of ranking well, even if the content on it is excellent.

Orphan pages arise for several reasons: a category was removed but articles remained; a page was created and forgotten to be added to navigation; site migration lost part of the linking structure; landing pages were created outside the CMS without adding links.

Solution: export a list of all site URLs (from sitemap or Google Search Console), run a crawler (Screaming Frog, Ahrefs), compare the list of crawled pages with the full URL list — the difference is your orphan pages. For each, determine where a natural link can most logically come from.

Internal link audit

Regular link auditing is not a one-off task but part of the technical SEO process. At minimum quarterly, check several key metrics.

What to check in an audit

• Orphan pages — sitemap URLs with no incoming internal links (Screaming Frog → Reports → Orphan Pages)
• Click depth — pages more than 3 clicks from the homepage (Screaming Frog → Reports → Crawl Depth)
• Broken internal links — links pointing to 404 pages (GSC → Coverage → Errors; Screaming Frog → Response Codes → 4xx)
• Redirect chains — links going through 301→301→301 (replace with direct links to the final URL)
• Pages with no outgoing links — dead ends that don't pass PageRank further
• Anchors — overly generic or duplicate anchors pointing to the same pages (Ahrefs → Internal backlinks)
• Incoming link count on priority pages — commercial and pillar pages should have more incoming links than supporting pages

# Screaming Frog CLI — basic internal link audit
screaming-frog-seo-spider-cli \
  --crawl https://example.com \
  --headless \
  --save-crawl \
  --export-tabs "Internal:All,Response Codes:4xx,Orphan Pages"

# Альтернатива — через API Google Search Console
# (инструмент seohead.tech Scripts автоматизирует этот аудит)

Internal linking checklist

• Every page has at least one incoming internal link
• Priority pages (commercial, pillar) have the highest number of incoming links
• No important page is more than 3 clicks from the homepage
• Internal link anchors are descriptive and vary in wording
• Breadcrumbs are set up on all category and product pages
• Broken internal links (to 404 pages) are fixed
• Redirect chains are replaced with direct links to the final URL
• Pages with duplicate content don't receive link equity (rel="nofollow" or canonical)
• Related posts / similar products are configured for blog articles and catalogue products
• Orphan page audit is run at minimum quarterly

FAQ