PageRank Explained: How Google Ranks Pages

PageRank is the algorithm that put Google on the map. Named after co-founder Larry Page, it remains a foundational pillar of how Google evaluates the authority and importance of every page on the web — even two decades after its invention. Understanding how it works is essential for anyone serious about SEO.

What Is PageRank?

PageRank is a link analysis algorithm developed by Larry Page and Sergey Brin at Stanford University in 1996. It was the core innovation behind the original Google search engine, and the insight was elegantly simple: a page is important if other important pages link to it. The more links a page receives — and the more authoritative those linking pages are — the higher its PageRank score.

The algorithm models a hypothetical "random surfer" who browses the web by clicking links at random. PageRank represents the probability that this random surfer lands on any given page at any given time. Pages that attract many links from many places end up with a higher probability of being visited — and a higher PageRank score. This probabilistic framing is why PageRank remains mathematically elegant and surprisingly robust even under adversarial conditions like link spam.

How PageRank Is Calculated

The PageRank formula is iterative. At its core: PR(A) = (1 - d) + d * (PR(T1)/C(T1) + ... + PR(Tn)/C(Tn)), where d is the damping factor (typically set at 0.85), PR(Ti) is the PageRank of pages linking to page A, and C(Ti) is the number of outbound links on each of those linking pages. The damping factor represents the probability that the random surfer continues clicking rather than jumping to a random new page — the 0.85 value means there is an 85% chance they keep surfing and a 15% chance they teleport elsewhere.

PageRank is calculated iteratively across the entire web graph. Google starts with an initial value for every page and runs the calculation repeatedly until the scores converge. Each iteration redistributes PageRank from linking pages to linked pages. A link from a high-PageRank page with few outbound links passes far more authority than a link from a low-PageRank page with hundreds of outbound links. This is why a single editorial link from a major newspaper can be worth more than hundreds of links from obscure directories.

PageRank vs. Modern Google Ranking

PageRank is one of over 200 ranking signals that Google uses today. It has been joined by signals covering content relevance, user experience, Core Web Vitals, E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness), entity understanding, and dozens of other dimensions. Google's systems are now a complex ensemble of machine-learned models — including neural networks trained on human rater feedback — that go far beyond link counting.

Despite this complexity, PageRank remains foundational. Google's own engineers have confirmed it continues to run across the full web graph. It functions as a prior: before Google even evaluates content quality, PageRank gives each page a baseline authority score that influences how much trust the content-quality signals are allowed to contribute. A high-PageRank page does not automatically rank well, but a low-PageRank page faces a much steeper climb no matter how good its content is.

Internal PageRank Flow

PageRank flows through your site's internal link structure just as it flows across the external web. Your homepage typically has the most PageRank because it attracts the most external backlinks. From there, internal links distribute that authority to deeper pages. A page that is only reachable through four or five clicks from the homepage — or an orphan page with no internal links pointing to it at all — receives very little PageRank, even if the page itself contains excellent content.

This is why site architecture decisions are also PageRank decisions. A flat hierarchy — where important pages are reachable within one or two clicks from the homepage — ensures your best content receives meaningful authority. Deeply nested pages, pagination-buried archives, and orphan pages are all PageRank dead-ends. Auditing your internal link structure with a tool that maps link depth and identifies orphans is one of the highest-ROI technical SEO activities you can do.

External Links and PageRank

External backlinks are the primary source of PageRank for most websites. When an authoritative external site links to one of your pages, it transfers a portion of its PageRank to you. Quality matters far more than quantity: one link from a highly authoritative, topically relevant domain can outweigh hundreds of links from low-authority or off-topic sites. Google has become increasingly good at discounting links that appear to be artificially manipulated.

The anchor text of an external link also signals topical relevance to Google, though over-optimized exact-match anchor text can trigger algorithmic penalties. The best backlinks are editorially earned — they appear naturally within content because the linking author found your page genuinely useful. These links tend to come with diverse, natural anchor text and from pages that are themselves well-linked, maximizing the PageRank they transfer.

PageRank Sculpting: Dead or Alive?

PageRank sculpting was a popular technique in the mid-2000s: webmasters would add rel="nofollow" to internal links they considered less important — login pages, privacy policies, tag archives — in an attempt to concentrate PageRank toward their most valuable content pages. In 2009, Google's Matt Cutts revealed that this approach had stopped working. When you nofollow an internal link, the PageRank that would have flowed through it is now simply lost rather than redistributed to the followed links.

The current best practice is to avoid nofollowing internal links entirely. Use nofollow only for paid external links or external links you genuinely do not want to vouch for. Instead of trying to sculpt PageRank through nofollow, sculpt it through your site architecture: create a clear, flat hierarchy where important pages receive many internal links from contextually relevant pages, and unimportant pages are consolidated, redirected, or pruned.

What Toolbar PageRank Was

For many years, Google provided a public-facing PageRank score through the Google Toolbar — a browser extension that displayed a score from 0 to 10 for any page you visited. This toolbar PageRank was a simplified, discretized, and deliberately delayed snapshot of the internal PageRank score Google used for ranking. SEOs obsessed over it, entire businesses were built around it, and it created enormous incentives for link buying and selling since a single high-toolbar-PR link could be measured and monetized.

Google began winding down toolbar PageRank updates in 2013 and stopped updating it entirely in 2016. The Google Toolbar itself was deprecated shortly thereafter. Google's stated reason was that the public score had become more of a distraction and a target for manipulation than a useful signal. Removing it forced the SEO industry to develop more nuanced link quality metrics, though it also gave rise to a cottage industry of third-party link authority metrics designed to fill the gap.

Modern Proxies for PageRank

Since Google no longer publishes PageRank, the SEO industry relies on third-party metrics as proxies. Ahrefs provides Domain Rating (DR) — a measure of a domain's overall backlink profile strength — and URL Rating (UR), which estimates the authority of a specific URL. Moz offers Domain Authority (DA) and Page Authority (PA). SEMrush provides Authority Score. All of these metrics are broadly correlated with Google's internal PageRank but are computed from each tool's own crawled link graph, so they are estimates rather than ground truth.

These proxy metrics are useful for relative comparisons — comparing your site against a competitor, evaluating the quality of a prospective link target, or tracking the impact of a link-building campaign over time. They should not be treated as absolute measures. A site with DR 60 is not precisely twice as authoritative as a site with DR 30; the scales are logarithmic and tool-dependent. Use them directionally, not doctrinally.

How to Build PageRank

Building PageRank means earning external links and distributing authority effectively through your site. On the external side, focus on creating content that other webmasters have genuine reason to link to: original research, comprehensive guides, free tools, and data-driven studies consistently attract editorial links. Outreach, digital PR, and broken link building are proven tactics for amplifying this process. Avoid any tactic that purchases links or attempts to manipulate anchor text at scale — Google's SpamBrain system is highly effective at detecting and nullifying these.

On the internal side, audit for orphan pages — pages that receive no internal links and therefore inherit no PageRank from anywhere. Fix broken internal links that create dead ends in your link graph. Ensure your most commercially important pages are reachable within two clicks from the homepage and receive contextual links from related content. A site crawl combined with a PageRank flow visualization can reveal which pages are starved of authority and where structural changes would have the greatest impact.

Common PageRank Misconceptions

The most persistent misconception is that buying links is a viable PageRank-building strategy. Paid links that pass PageRank violate Google's guidelines, and Google actively targets link networks, private blog networks (PBNs), and link farms. Sites caught participating in these schemes face manual penalties that can remove them from the index entirely. Even if a paid link temporarily boosts rankings, the risk-adjusted return is deeply negative compared to earning links legitimately.

Another misconception is that exact-match anchor text in external links is a reliable ranking booster. While anchor text is a PageRank signal, manipulated anchor text — particularly over-optimized external anchors pointing to a single page — is a strong spam signal. Google's Penguin algorithm, now running in real-time as part of core, specifically targets unnatural anchor text patterns. The safest and most sustainable approach is to let anchor text vary naturally across your backlink profile, with brand name and URL anchors making up the plurality of your links.