Link analysis systems are the Google ranking systems that read how pages link to each other to work out what pages are about and which are most useful for a query. PageRank is the best known of them. Google confirms both in its ranking systems guide and says PageRank has evolved a lot but remains part of its core ranking systems. For a site, this means crawlable links, descriptive anchor text and an internal link graph that points weight at the pages that matter are still worth auditing.
What are Google's link analysis systems and PageRank?
Link analysis systems are the family of Google ranking systems that use the links between pages to understand what a page is about and how useful it is likely to be for a query. Google lists them under one heading, "Link analysis systems and PageRank", in its guide to Google Search ranking systems.
PageRank is the named member of that family. It was described in 1998 by Sergey Brin and Larry Page in The Anatomy of a Large-Scale Hypertextual Web Search Engine, which treats a link as a citation and weights each citation by the importance of the page it comes from.
We have various systems that understand how pages link to each other as a way to determine what pages are about and which might be most helpful in response to a query. Among these is PageRank, one of our core ranking systems used when Google first launched.
Google, A guide to Google Search ranking systems, Google Search Central
How does PageRank work?
In the original paper, a page's PageRank is the probability that a "random surfer" who keeps clicking links lands on it. Links from pages that are themselves heavily linked pass more weight, and a page's outgoing weight is split across its links.
Two other ideas from the same paper still matter in practice. Anchor text is attached to the page a link points to, not just the page it sits on, and a damping factor lets the random surfer jump to a new page, which the authors noted could be biased towards chosen pages for personalisation. Later link systems, such as topic-sensitive PageRank and the reasonable surfer patent, are variations on these ideas.
PageRank extends this idea by not counting links from all pages equally, and by normalizing by the number of links on a page.
Sergey Brin and Lawrence Page, The Anatomy of a Large-Scale Hypertextual Web Search Engine, Stanford InfoLab
What evidence is there that Google uses PageRank today?
This is a Confirmed system. Google's ranking systems guide names PageRank as a core ranking system, links to the original paper and patent, and states that it has changed a great deal but is still in use. Google's own link best practices page also says links are used as a relevance signal and for discovery.
What Google does not publish is the current formula, its weight against other signals, or how link spam is discounted before it reaches PageRank. The public Toolbar PageRank score was switched off on 15 April 2016 according to Wikipedia's PageRank article, so no public PageRank number exists. Third-party metrics such as domain rating are estimates, not Google data.
How PageRank works has evolved a lot since then, and it continues to be part of our core ranking systems.
Google, A guide to Google Search ranking systems, Google Search Central
What should sites do about link analysis?
Treat links as two separate jobs: discovery and weighting. For discovery, every page you want ranked needs a crawlable <a href> link from somewhere Google already crawls. For weighting, the internal link graph decides where your own link equity flows, and you control it completely.
The falsifiable test: crawl the site, compute an internal PageRank style score for every indexable URL, and compare the top of that list with the pages that earn revenue or leads. If utility pages such as login, cart, terms or tag archives outrank your money pages internally, your own architecture is working against you.
- Use real <a href> links, not click handlers on spans or divs.
- Write anchor text that describes the target page.
- Link to key pages from pages that themselves receive links, not only from the footer.
- Qualify paid and user-generated links with rel values rather than hoping they are ignored.
Google uses links as a signal when determining the relevancy of pages and to find new pages to crawl.
Google, Link best practices for Google, Google Search Central
Which Laurelin audit checks test for link analysis problems?
The new check, Internal link graph concentrates PageRank on low-value pages, models your internal PageRank from a crawl and flags where it pools. It measures a proxy: a site-only model of link flow, not Google's PageRank, which includes external links and spam discounting we cannot see.
Related checks:
- Key pages with few internal links: the simplest symptom of weak internal link weight.
- Deep pages not linked from authority pages: pages that sit far from where external link equity lands.
- Orphan page: a page with no internal links cannot receive any internal PageRank.
- Internal nofollow on equity links: nofollow on internal links withholds a signal you want Google to use.
- Links as span or div with JS onclick: links Google cannot extract pass nothing.
- Low referring domain authority vs competitors: the external side of the same system.
What are the key dates for PageRank?
Frequently asked questions about PageRank
Does Google still use PageRank?
Yes. Google's ranking systems guide says PageRank has evolved a lot since launch and continues to be part of its core ranking systems.
Can I see my site's PageRank?
No. Google stopped showing Toolbar PageRank in 2016. Any score you see in an SEO tool is that vendor's estimate, not Google's number.
Is internal linking part of PageRank?
PageRank as described in the original paper is computed over all links, internal and external. Google does not publish how it weights internal links today, so an internal PageRank model is a useful proxy rather than a measurement.