What Is PageRank? Google's Link Algorithm Explained

PageRank is an algorithm that measures how important a web page is, based on the links pointing to it. This guide explains the idea in plain words. It walks through the formula with a tiny example. It also shows how PageRank still matters for SEO today.

Updated 26 Oct 20268 min readChecked against the original 1998 paper

Quick answer

PageRank treats a link as a vote. A page linked by many important pages gets a high score. It then passes some of that score to the pages it links to. Larry Page and Sergey Brin created it at Stanford in the late 1990s. Google still uses it today, as one of many ranking signals.

Written by Usman AliGrowth and SEO specialist · 7 years in SEO

Reviewed by Amir ShahidSenior SEO strategist · 9+ years in SEO

The basics

What Is PageRank, in Plain Words

PageRank scores a page by the number and importance of the pages that link to it.

Before PageRank, most search engines judged pages mainly by the words on them. That old method was very easy to fool. A page could repeat a keyword hundreds of times to jump to the top.

PageRank added a new question: who links to this page? A link from a respected site suggests the page is worth reading. The idea came from academic papers, where a study cited by many other studies is seen as important. It quickly became the foundation of Google Search. Later it turned into a core part of search engine algorithms in general. To see where link analysis sits in the wider process, read our introduction to search engines.

Count matters

A page with many incoming links usually earns more PageRank than a page with few.

Better links

Quality matters more

One link from a highly ranked page can be worth more than dozens from unknown pages.

The math

The PageRank Formula, Made Simple

PageRank comes from one formula, repeated across every page until the scores settle.

PR(A) = (1 - d) / N + d × ( ~~PR(T1) / C(T1)~~ + ... + ~~PR(Tn) / C(Tn)~~ )PR(A) the PageRank of page A T1..Tn the pages that link to page A C(T) how many links page T has in total d the damping factor, usually 0.85 N the total number of pages

Part	What it means in plain words
PR(T) / C(T)	A page splits its score evenly across all its links. More links means a smaller share each.
Damping factor d	The chance a random surfer keeps clicking links. At 0.85, they click on 85 percent of the time. The other 15 percent of the time, they jump to a random page.
(1 minus d) / N	Every page gets a small base score, so no page is ever exactly zero.

The formula is run again and again across the whole web. Each round, scores move a little closer to their final values, until they stop changing.

Worked example

PageRank on a Tiny Web of Three Pages

Here is the formula run on a web of three pages, with d set to 0.85.

APage A: 0.388Links out to pages B and C. It gets only one link, from page C.

BPage B: 0.215Its only outgoing link points to page C. It gets half of the vote from page A.

CPage C: 0.397Its only outgoing link points to page A. It gets votes from both A and B.

We start every page at the same score of one third. After the formula runs many times with d set to 0.85, the scores settle at the numbers above. Together, the three scores always add up to 1.

Page C wins because two pages link to it. Page A comes a close second, since its only link comes from the strongest page, C. Page B is last, because it receives just half of A’s vote. The lesson for your site is clear. Pages that get links from your strongest pages tend to rank better.

Real-world case

The Rise and Fall of the PageRank Score

Case study

How a public number started a link market

PageRank built Google, but showing the score in public caused real trouble.

1996Larry Page and Sergey Brin build a search project at Stanford, first called BackRub.
1998They publish the PageRank paper before founding Google the same year.
2000The Google Toolbar shows every page a public score from 0 to 10.
2000sA market grows for buying or selling links from high-score pages. Google fights back with link spam rules.
2013 to 2016Google stops updating the toolbar score before removing it for good.

Inside Google, PageRank never stopped being part of ranking. Only the public number disappeared, because it encouraged people to chase links rather than build good pages.

Lesson: earn links by being worth linking to. Bought links break Google's spam policies, which can cost you rankings.

Use it today

How PageRank Affects Your SEO Today

You cannot see your PageRank, yet you can still put the idea to work.

Link to key pages

Link your most important pages from your homepage or popular posts. Those pages pass the most value to others.

Internal links

Earn real links

Useful guides tend to attract links naturally over time. Free tools or original data do the same. One trusted link beats many weak ones.

External links

Avoid orphan pages

A page with no internal links gets almost no PageRank. Such a page may not even be found by crawlers.

Site structure

Canonical tags help with link value here too. They gather links that point at duplicate URLs onto one page. Our canonical tag guide explains how this works. For the full list of other signals, see our guide to Google ranking factors.

Go deeper

Dangling Pages, Convergence and Link Equity

These terms appear in technical PageRank explanations.

Term	What it means
Random surfer model	The idea of a person clicking links at random. PageRank is the chance this surfer lands on each page.
Dangling page	This is a page with no outgoing links. The classic fix spreads its score evenly across all pages, so value is not lost.
Iterations	Each full round of the formula across every page. Early rounds change scores a lot, later rounds only a little.
Convergence	The point where scores stop changing in a useful way, so the calculation can stop.
Eigenvector	In math terms, the final PageRank scores form the main eigenvector of the web’s link matrix.
Link equity	The SEO name for the value a link passes, sometimes called link juice.
PageRank sculpting	An old trick of adding nofollow to internal links to save value. Google changed how nofollow works in 2009, so it no longer helps.
Reasonable surfer	This is a later idea from a Google patent. Links a person is likely to click, such as links in main content, may pass more value than footer links.

The reasonable surfer idea explains a practical tip. A link inside a helpful paragraph is likely worth more than the same link buried in a footer. Our guide to canonical tags shows how to keep that link equity on one URL.

Wider uses

PageRank Beyond Web Search and Link Spam

The idea behind PageRank reaches well beyond Google, while its weak spots explain why link spam became such a stubborn problem.

Mathematically, the random surfer is a Markov chain, which is a process where the next step depends only on the current page. PageRank is that chain’s long-run distribution, meaning the share of time the surfer spends on each page after countless clicks.

A rank sink is a group of pages that link to each other but never link out, which traps score inside the loop. The damping factor limits that effect, because the surfer can always jump to a random page instead. Link farms exploit the same maths, using networks of throwaway pages to point backlinks at one target. Researchers answered with TrustRank in 2004, which spreads trust outward from a hand-picked set of reliable seed sites.

Other fields borrowed the idea too, starting with HITS, which Jon Kleinberg described in the late 1990s. It gives each page two scores, one as a hub with useful outbound links and one as an authority with valuable inbound links. TextRank applies PageRank to words or sentences, which helps software pick key phrases or build automatic summaries. Citation analysis inspired PageRank in the first place, while variants of it now rank journals or authors by influence.

FAQ

Questions About PageRank

Does Google still use PageRank?

Yes, Google lists link analysis systems, including PageRank, among its ranking systems. Today it is one signal among many, not the main deciding factor.

Can I check my PageRank score?

No, Google stopped updating the public Toolbar PageRank in 2013. It removed the score completely in 2016. Third-party authority scores are only rough estimates of it.

Who invented PageRank?

Larry Page and Sergey Brin created it as Stanford University students. The name comes from Larry Page, though it also fits the idea of ranking pages.

Do nofollow links pass PageRank?

Google treats a nofollow link as a hint. In most cases these links do not pass ranking credit, so they add little PageRank to the target page.

Do internal links pass PageRank?

Yes, links between your own pages share value too. That is why important pages should get links from your homepage or popular posts.

What replaced Toolbar PageRank?

Nothing official replaced it, since Google keeps its internal PageRank private. Site owners now rely on the Links report in Search Console or third-party authority scores, which are only rough estimates.

Wrapping up

Final Thoughts on PageRank

PageRank changed the way search works forever. It turned links into votes before using those votes to sort the web by importance. More than 25 years later, the same simple idea still lives inside Google, working next to many newer systems.

You do not need the formula to benefit from it. To use it well, just remember these three simple points. Links from strong, relevant pages count most. A page shares its value across every link it has. Pages with no links pointing to them get almost nothing. So build a clear internal link structure that points your best pages at your most important ones. Then earn outside links with content people want to share. That is PageRank working for you, without a single calculation.

Keep learning

Where to Go Next

/ranking/Search engine rankingHow results are ordered./crawling/web-crawler/What is a web crawler?The bots that follow links./ranking/search-engine-positioning/Search engine positioningHow positions are counted./serp/What is a SERP?Where ranked pages appear./search-engines/history/History of search enginesWhere PageRank came from./resources/glossary/Search engine glossaryEvery ranking term, defined.