PageRank is Google's original algorithm for measuring the importance of a webpage based on the number and quality of other pages linking to it. Developed by Larry Page and Sergey Brin at Stanford in 1998, it adapted academic citation analysis to the web. Google stopped showing public PageRank scores in 2016, but patent filings and engineer statements confirm evolved versions continue operating inside their ranking systems.
PageRank is why backlinks matter. Every link from another page is a vote — but votes from high-authority pages count far more than votes from low-authority ones. Understanding PageRank explains why one link from a single high-authority source can outrank dozens of weak ones.
What is PageRank?
PageRank treats the web as a citation graph. Each page starts with a base score and receives fractional authority from every page that links to it. A link from a page with high PageRank passes more authority than a link from a page with low PageRank. The key insight: not all links are equal.
The original paper, "The PageRank Citation Ranking: Bringing Order to the Web" (Page et al., Stanford, 1998), described the algorithm as simulating a "random surfer" who clicks links indefinitely. The probability they land on any given page equals that page's PageRank.
Google confirmed in 2017 patent filings that modified PageRank variants continue operating within their ranking systems. Backlinks remain a confirmed top-3 ranking factor. Public scores disappeared in 2016 to prevent manipulation, not because the underlying concept became less important.
Why PageRank matters for SEO
PageRank is the mechanism behind why backlinks affect rankings. Four practical implications:
- Not all links are equal. One link from a prestigious editorial source (major publication, university, government site) outweighs hundreds of directory submissions. Quality of linking domain matters more than raw link count.
- Internal links distribute authority. Every internal link you add from a high-PageRank page to a priority page passes real authority. You can dramatically improve rankings for key pages by restructuring internal links — no external backlinks required.
- Unreachable pages get no PageRank. Pages buried more than 3-4 clicks from the homepage receive minimal PageRank. Orphaned pages — those with no internal links — receive effectively zero.
- Link dilution is real. A page with 50 outbound links passes far less authority per link than a page with 5 outbound links. The available PageRank is split equally among all outbound links.
How PageRank works — the mechanics
The PageRank formula has four core components:
1. The voting model
Each page votes for the pages it links to. The vote weight equals the linking page's PageRank divided by its total number of outbound links. A page with PageRank 100 and 10 outbound links passes 10 units per link. A page with PageRank 100 and 100 outbound links passes 1 unit per link.
2. The damping factor (0.85)
The random surfer doesn't follow links forever — they occasionally type a new URL directly. The damping factor (d = 0.85) models this. It prevents infinite concentration of PageRank on a single page and ensures even pages with no incoming links receive a baseline score.
PR(A) = (1-d) + d * (PR(B)/C(B) + PR(C)/C(C) + ...)
# Where:
# d = damping factor (0.85)
# PR(B) = PageRank of linking page B
# C(B) = number of outbound links from page B
3. Iterative recalculation
PageRank cannot be calculated in a single pass — it requires iterative computation across the entire web graph until scores stabilise. Google recalculates continuously as new pages and links are discovered.
4. Topic-sensitive weighting
Modern PageRank variants weight links by topical relevance, not just raw authority. A backlink from a topically relevant source in the same industry passes more effective PageRank than an equal-authority link from an unrelated domain.
PageRank variants and third-party proxies
| Metric | Creator | Scope | Scale | Public access |
|---|---|---|---|---|
| PageRank (internal) | Per page | Internal (0-10 historically) | No (since 2016) | |
| Page Authority (PA) | Moz | Per page | 0-100 logarithmic | Yes (Moz tools) |
| URL Rating (UR) | Ahrefs | Per page | 0-100 logarithmic | Yes (Ahrefs tools) |
| Domain Authority (DA) | Moz | Per domain | 0-100 logarithmic | Yes (Moz tools) |
| Domain Rating (DR) | Ahrefs | Per domain | 0-100 logarithmic | Yes (Ahrefs tools) |
Real PageRank examples
1. Content cluster — internal PageRank without external backlinks
An HVAC services blog published 30 articles per month, interlinking all 180 articles over 6 months into a tight content cluster. Despite having few external backlinks, the internal PageRank distribution pushed category pages to page-1 rankings for competitive local keywords. Internal links did the job that external link building typically requires.
2. Single high-authority backlink — position 14 to 6 in 8 weeks
An accounting firm's service page sat at position 14 for "small business accountant [city]". A Forbes guest post mentioning and linking to the service page moved it to position 6 within 8 weeks. The Forbes domain passes significant PageRank — one link from a domain ranking for millions of terms outweighed dozens of directory links the firm had built over years.
3. Site architecture — orphaned page recovery
A real estate site had a high-quality neighbourhood guide indexed but stuck at position 23. Analysis revealed the page had only one internal link from a buried category page, 4 clicks from the homepage. Adding 12 contextual internal links from relevant blog posts and the homepage sidebar moved it to position 8 within 6 weeks. Content didn't change — only PageRank flow did.
PageRank vs. Domain Authority — which one should you track?
This comparison matters because SEO tools show DA/DR constantly, but Google uses neither. Here is the practical distinction:
PageRank (what Google uses)
- Per-page scoring — not domain-level
- Reflects actual Google ranking input
- Accounts for link equity dilution per outbound link
- Updated continuously as Google recrawls
- Not publicly accessible since 2016
Domain Authority (third-party estimate)
- Domain-level scoring — not per-page
- Moz's prediction of Google rankings, not Google's own signal
- Useful for comparing sites at a high level
- Manipulable through link schemes
- Not correlated with Google rankings as cleanly as PA/UR
5 PageRank best practices for 2026
- Target high-authority pages, not just high-DA domains. A link from a top-ranking page on a mid-DA domain can pass more PageRank than a buried page on a high-DA domain. Check the linking page's URL Rating in Ahrefs, not just domain-level authority.
- Flatten your site architecture. Keep every important page within 3 clicks of the homepage. Each additional click depth reduces the PageRank a page receives from your homepage's authority pool.
- Link strategically from your highest-PageRank pages. Identify your top-authority pages (highest inbound links, oldest pages, most linked-to content). Add contextual internal links from those pages to your conversion or priority ranking pages.
- Never add nofollow to internal links. Nofollow on internal links blocks PageRank flow to linked pages. Internal nofollow is rarely justified and typically represents a configuration error.
- Publish content that earns links passively. Original data studies, comprehensive guides, free tools, and contrarian analysis attract editorial backlinks. Each editorial backlink passes real PageRank that you cannot replicate through internal linking alone.
DA is a domain-level metric. PageRank is a page-level metric. A domain with DA 80 can have individual pages with minimal PageRank if they have few internal or external links pointing to them. Always evaluate the specific page you want a link from, not just the domain it's on. Check Ahrefs URL Rating or Moz Page Authority for the page-level signal.
Common PageRank mistakes to avoid
- Building links to the homepage only — PageRank built on the homepage must flow through internal links to reach product and category pages. Build links directly to the pages you want to rank.
- Ignoring internal linking — internal links are free PageRank distribution. Most sites under-invest here while spending heavily on external link building.
- Chasing DA over page-level authority — a DA-90 domain's 500th blog post passes less PageRank than a DA-60 domain's heavily linked cornerstone piece.
- Allowing redirect chains to absorb PageRank — each redirect hop in a chain leaks a fraction of link equity. Redirect directly to the final URL.
- Treating all links as equal — 500 blog comment links pass less PageRank combined than a single editorial mention in a well-linked industry publication.
- Removing pages without redirecting — deleted pages lose all earned PageRank. 301 redirect them to the closest relevant page to preserve equity.
Frequently asked questions
Yes. Patent filings and Google engineer statements confirm evolved versions of PageRank remain operational within Google's ranking systems. Public scores disappeared in 2016 to reduce manipulation incentives, but the underlying link-authority concept continues to influence rankings.
Real PageRank scores are not publicly accessible since Google retired Toolbar PageRank in 2016. Third-party estimates include Moz's Page Authority (PA) and Ahrefs' URL Rating (UR), which are the closest available proxies for page-level link authority.
Officially no. Since 2019, Google treats nofollow as a "hint" with possible rare application, but dofollow links reliably transmit PageRank. Nofollow links may provide referral traffic and brand signals but should not be counted in link equity calculations.
Absolutely. Strategic internal linking distributes PageRank to priority pages across your site without requiring external backlinks. Pages deep in site architecture receive less PageRank from the homepage — flattening site depth is one of the most underused PageRank tactics.
No direct Google replacement exists. Third-party tools — Moz's Page Authority, Ahrefs' URL Rating, and Semrush's Authority Score — serve as proxies. None are as authoritative as the original Toolbar PageRank since they estimate Google's internal score rather than measuring it directly.
Related glossary terms
Sources
- [01]Page, L. et al. (1999) — The PageRank Citation Ranking: Bringing Order to the Web (Stanford InfoLab)
- [02]Google Patent US9165040 — Ranking search results (2017 filing confirming evolved PageRank)
- [03]Google Search Central — How Google Search works (ranking signals overview)
- [04]Ahrefs — Google PageRank: What it is and how it works in 2026
- [05]Moz — Page Authority: what it is and how to improve it
