The Google PageRank Patent — What It Actually Says (In Plain English)

US Patent 6,285,999 B1 is the original Google algorithm. It's still running. Most websites were built without any understanding of what it says — and it shows in their rankings.

What PageRank Is — The Original Idea

PageRank is not just a metric. It's a mathematical model of how trust propagates through a network. A page that is linked to by many trusted pages inherits some of that trust. A page that links to many other pages distributes its trust across them. The architecture of your website — which pages link to which — directly determines how trust accumulates and where it concentrates.

Most websites were built without any consideration of this. The result is a site where authority sits on the homepage and goes nowhere — while the service pages that actually need to rank are starved of the trust they need.

What US Patent 6,285,999 B1 Actually Says

The patent describes a method for assigning a numerical weight to each element of a hyperlinked set of documents. The weight — PageRank — is calculated iteratively based on the weights of the documents that link to it.

The plain English explanation: imagine every page on the internet as a person in a room. Every link from page A to page B is person A telling person B something important. The more people tell you something, the more credible you become. The more credible the people telling you are, the more their endorsement matters. PageRank is the algorithm that calculates how credible each person in the room is, based on who endorses them and how credible those endorsers are.

The formula includes a damping factor — typically 0.85 — which models the probability that a random web surfer continues clicking links rather than jumping to a new random page. This prevents PageRank from concentrating infinitely in closed loops and ensures every page receives some baseline authority.

Why PageRank Isn't Dead — It's the Foundation Everything Else Builds On

Google stopped displaying the public PageRank toolbar metric in 2016. This led many SEOs to conclude that PageRank was dead. It wasn't. Google confirmed in 2020 that PageRank remains a core ranking signal. The toolbar number is gone. The algorithm is not.

Every subsequent Google algorithm — the Reasonable Surfer Model, the Graph Distance patent, the Information Gain patent— builds on top of PageRank. It's the foundation. Understanding it is not optional for anyone who wants to build a site that ranks.

How PageRank Flows Between Pages — The Practical Implications

PageRank flows through links. Every link on a page distributes a fraction of that page's PageRank to the pages it links to. The more links on a page, the smaller the fraction each destination receives. A page with 100 outbound links passes less PageRank per link than a page with 5 outbound links.

This has direct implications for site architecture:

• A homepage that links to 200 pages passes very little PageRank to each one.
• A homepage that links to 8 primary pillar pages passes substantial PageRank to each.
• Those pillar pages then link to spoke pages, distributing authority further down the hierarchy.
• Spoke pages that link back to the pillar create a recursive loop that concentrates authority at the pillar level.

PageRank and Internal Linking — Why Your Site Architecture Matters

If your homepage has high PageRank — because lots of external sites link to it — and you never link from your homepage to your service pages, your service pages don't benefit from that trust. This is why most WordPress sites have high-authority homepages and near-zero-authority service pages. The trust sits on the homepage and goes nowhere.

PageRank in the AI Search Era

LLMs don't directly use PageRank — but they were trained on the web, and the web is shaped by PageRank. The pages that accumulated the most PageRank over time are the pages that appear most frequently in high-authority contexts. Those are the pages that disproportionately shaped LLM training data.

Building a site with correct PageRank architecture doesn't just improve Google rankings. It improves the quality and authority of the content that gets indexed — which improves the probability of that content appearing in LLM training data and AI citations. The GEO framework is built on this foundation.

How a Sovereign Build Is Engineered for PageRank

A Sovereign Build is engineered around Patent 6,285,999 B1 from the ground up. Every architectural decision — the number of primary navigation links, the pillar-spoke hierarchy, the in-content cross-linking, the footer link isolation — is made with PageRank flow in mind.

The Google Patent Compliancepage documents exactly how each build decision maps to the patent requirements. This is what “Blueprint Brutalism” means: structural integrity based on the actual documents that govern how Google works, not on marketing trends or plugin recommendations.

Frequently Asked Questions