Internal Link Spread Optimization
Most sites starve their own pages of internal links; fixing spread is SEO's biggest quick win.
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 92%
Google crawls and credits pages through internal links, yet ~90-95% of sites concentrate links on a handful of pages, leaving most with zero or one. Standard link algorithms (faceted, popular, recent, related) all skew toward a subset. Spreading body links so every page has enough, and sits few hops from a crawl point, reliably lifts traffic 25-100%.
Origin
Ethan Smith's finding, first tested at a 2007 e-commerce site with complete Google Search Console internal-link data, later rebuilt at Thumbtack and Upwork.
Core principles
- 01Google needs multiple internal paths (not just a sitemap) to find and regularly crawl a page
- 02Body links count far more than navigation/footer links
- 03You want a flat graph with short paths from crawl points, not a deep hierarchy
- 04Faceted, popular, recent and related link algorithms all leave large gaps
How to run it
- 1
Audit link distribution
Crawl the whole site with Screaming Frog and read the 'unique inlinks' count per page. Expect a power curve where ~5% of pages get ~95% of links.
Pro tip Look for pages with zero or one internal link; aim for at least 5, ideally 10, links per page.
- 2
Correlate links to traffic
Join internal-link counts with traffic data to confirm the pattern that more-linked pages get more traffic, then treat added links as a causal test.
- 3
Spread body links to starved pages
Add contextual body links (not footer/nav) pointing to under-linked pages, placed at the bottom of articles or category pages, above the footer.
Pro tip Body links count a lot; navigation and footer links are treated as navigational and count much less.
Watch out Don't bury the links in the footer; Google separates out repeated navigational chrome and discounts it.
- 4
Flatten paths from crawl points
Ensure every page is few hops from a crawl point (home page plus any high-authority pages Google enters through), rather than deep in a tree.
In the wild
With millions of products, Smith downloaded every URL and its internal-link count from Search Console, correlated links with traffic, then added internal links to link-starved product pages.
→ Roughly a 300% increase in organic traffic to those product pages, a clear causal result.
Common mistakes
Relying on a sitemap to get pages crawled
A sitemap is one path; Google needs many internal paths (≥~10 links) to crawl a page regularly enough to credit it, so sitemap-only pages stay uncrawled.
Putting the fix in the footer
Footer and nav links are treated as navigational and heavily discounted, so footer-based 'link spreading' fails to move rankings.
Is it for you?
Best for
Large programmatic or content sites with thousands of pages and uneven internal linking
Not ideal for
Small sites with few pages where every page is already well-linked
From the transcript
“pages with more links got more traffic fewer links got less traffic”
“usually at least 10 links Google needs to to find a page”
“you'll see a power curve where you know five percent of pages are getting 95 of links”
“you want a shortest path from crawl point”
“I wouldn't put it in the footer I would put it above the footer”
From the episode
The ultimate guide to SEO
Ethan Smith (Graphite)