A link-analysis algorithm developed by Larry Page and Sergey Brin at Stanford University and later patented by Google. It provides a mathematical method for measuring the relative importance of hyperlinked documents (web pages) based on the voting power of the incoming links.
Core Concept: Links as Votes
The fundamental premise of PageRank is that a link from Page A to Page B is an endorsement. However, PageRank differs from a simple link count by applying two critical filters:
- Weighted Authority: A link from an authoritative, high-ranking page carries more weight than a link from a low-ranking or unknown page.
- Dilution of Influence: If an authoritative page links to hundreds of other sites, the “vote” (PageRank) it passes to each individual site is diluted.
The PageRank Formula
The algorithm models the behavior of a random surfer who follows links until they get bored and jump to a random new page.
Loading formula...
Variables:
- PR(A): The PageRank of Page A.
- d: The Damping Factor (typically set at 0.85), representing the probability that a user will continue clicking links.
- 1-d: The probability that a user jumps to a random page.
- N: Total number of pages in the index.
- PR(Tn): The PageRank of page T which links to A.
- C(Tn): The number of outbound links on page T.
Technical Characteristics
| Feature | Description |
| Iterative Process | Because a page’s rank depends on the rank of its neighbors, the scores are calculated repeatedly until they converge (stabilize). |
| Eigenvector Centrality | Mathematically, PageRank is a variant of eigenvector centrality in graph theory, where a node’s importance is based on its connections to other important nodes. |
| Random Walk | The algorithm simulates a Markov Chain; the PageRank value is essentially the probability distribution of a random surfer landing on a specific page. |
| Damping Factor | Prevents “Rank Sinks” (loops where pages link only to each other) from hoarding all the value by ensuring a 15% chance of the surfer teleporting elsewhere. |
Operational Evolution
While the original PageRank was the primary ranking signal in early search, its role in 2026 has evolved from a standalone score to a foundational component of a larger algorithmic ensemble.
- Public vs. Internal: Google retired the public Toolbar PageRank in 2016 to prevent manipulation, but it remains a core internal signal used to determine crawl priority and baseline authority.
- The Nofollow Attribute: Introduced to allow webmasters to link to a page without passing PageRank (commonly used for advertisements or untrusted user-generated content).
- AI & Semantic Filtering: Modern iterations of the algorithm (often integrated with systems like RankBrain) can now distinguish between an earned link in an editorial context and a manipulated link in a footer or sidebar.
Key Success Factors
To maximize internal PageRank distribution, technical SEO relies on:
- Internal Linking: Using a hub and spoke model to direct authority from high-performing pages (hubs) to deeper content (spokes).
- Backlink Quality: Acquiring links from high-trust entities (e.g.,
.gov,.edu, or major news outlets) which possess massive amounts of accumulated PageRank. - Eliminating Bottlenecks: Fixing Orphan Pages (pages with no internal links) that cannot receive any PageRank flow.
