Internal Linking Strategy
Internal linking is the highest leverage thing most sites ignore, mostly because doing it properly by hand is miserable. So I stopped doing it by hand and built a tool that maps the link graph and tells the team where the next link should go.
Where the authority actually goes
Same pages, same content, same site. The only thing that changes between these two is which pages link to which. The number above each column is how many clicks that page sits from the homepage.
What I work on
Internal linking audits
Which pages are orphaned, which money pages are buried four clicks deep, and which pages are collecting links they do not need. Usually the answer surprises people who have been running the site for years.
Link graph mapping
The full picture of what links to what, rather than a sample. This is the thing the dashboard was built to do, and it is the part almost nobody has because doing it by hand on a large site is not realistic.
Hub and cluster architecture
Designing which pages are hubs, what sits under them, and how the clusters connect. I built this for cricket match pages at Parth Tech and for editorial and commercial pages together at MoveUp Media. Related to the crawl and architecture work.
Anchor text that reads like a sentence
Anchors should tell a reader where they are going. If your internal links all say the exact same four words, that is a pattern, and patterns that exist only for search engines tend to age badly.
Linking rules writers will follow
A framework nobody follows is worse than no framework, because you think it is handled. I write rules that fit how the desk already works, which is part of my editorial SEO systems.
Crawl efficiency
Internal linking and crawl budget are the same problem seen from two ends. Log files tell you which paths Googlebot is actually walking, and the linking structure is how you change that.
The internal linking dashboard
A Streamlit application connected to Google Sheets that maps how internal links are distributed across a site and recommends where the next one should point. I built it for a team I was running, which is the only reason it gets used.
Reads the link graph as it actually is
Not how the navigation suggests it should be. Every internal link on the site, which page it comes from and which page it lands on, counted per page.
Finds the pages nothing points at
Orphans are invisible in every report that starts from a list of URLs, because the URL exists. They only show up when you look at the graph rather than the sitemap.
Recommends the next target
Given a page about to be published, it says which existing pages should link to it and which pages it should link out to, based on where authority is currently pooling.
Runs on Google Sheets
Because that is where the content team already works. A tool that needs a login nobody has is a tool that gets used twice.
Built with Claude and MCP
From idea to something the team opens daily. I built it for my own team rather than as a demo, which is the only reason it survived contact with them.
How I work through it
Five steps, in this order, and the order matters more than usual here.
Decide what matters before touching a link
Internal linking is a way of telling Google what is important. If nobody has decided what is important, the linking will just make the existing accident louder.
Map the graph, do not guess it
How pages are connected in practice is rarely how the navigation implies. On any site past a few hundred pages, this is the step where the surprises are.
Find what nothing points at
Orphans and near-orphans are the cheapest wins available. The page already exists, someone already wrote it, and it is currently invisible.
Shorten the path to the pages that pay
Depth is a signal and it is also a user problem. A page three clicks in gets a fraction of the crawl attention and a fraction of the humans.
Leave rules, not a spreadsheet
A one off linking pass decays the moment publishing continues. What lasts is a rule the desk can apply, and ideally a tool that tells them where to point the next one.
"Internal linking is authority routing. Treat it like plumbing, not decoration."
Linking work I can point at
Three examples with the company named. The full write ups are on the homepage.
The internal linking dashboard
A Streamlit app that maps the link graph and recommends where the next link should point.
Rebuilding architecture after a manual action
Log file analysis showed where Googlebot was really going, and the crawl paths were rebuilt around it.
The cricket match hub model
Live match pages linked into team, player and editorial content so each one strengthened the others.
Frequently Asked Questions
You probably have pages nothing links to
Send me the site and I will tell you where I think the authority is pooling, whether or not we end up working together.