Understanding Link Equity: A Complete Guide

Here's a fact that might catch you off guard: Google has never once used the term "link equity" in any official documentation, blog post, patent filing, or public statement. Not once. A phrase that millions of SEO professionals throw around daily in meetings, in client reports, in blog posts and conference talks — it doesn't exist in Google's vocabulary. They've talked about PageRank. They've discussed "link signals" and "link value" in broad terms. But "link equity"? That's an industry invention, a shorthand that took on a life of its own until people started treating it like an established technical concept with a precise definition. It doesn't have one. Not from Google, anyway.

Key Takeaways

• Where the Term Came From
• What Google Actually Says About Links Passing Value
• Comparing the Industry Model vs. What We Can Actually Verify
• The Decay Question
• Where the Model Probably Breaks Down
• The Tools Problem

And that's what makes this topic so interesting to dig into. We're dealing with a concept that everyone uses, everyone thinks they understand, and almost nobody can define with actual precision — because the entity that controls how it works won't confirm or deny most of what the industry believes about it.

I've spent the better part of six months trying to trace the origins of this term, cross-referencing industry claims against Google's actual statements, and what I found is... messy. Sometimes reassuring. Sometimes contradictory. What emerges isn't a clean framework you can plug into a spreadsheet. It's more like a collection of strongly held beliefs, some supported by evidence, some supported by correlation studies, and some that appear to be myths that got repeated so often they became accepted wisdom.

Where the Term Came From

The earliest uses of "link equity" I can find in SEO literature date back to roughly 2005-2007, right around the time the industry was moving away from the cruder term "link juice." Both phrases attempt to describe the same idea: that when one page links to another, some portion of ranking power — whatever that means in practice — flows through that link. On the other side, the linked-to page gets some benefit. That benefit can be further distributed through that page's own outbound links. And so on.

This idea isn't wrong, exactly. It's rooted in the original PageRank paper from Larry Page and Sergey Brin, published in 1998. Brin and Page describe a model where pages accumulate importance based on the links pointing to them, and where the importance of the linking page factors into how much value gets passed. That much is real. That much is documented. But PageRank — the specific mathematical formula described in that paper — hasn't been the whole story for a very, very long time. For a deeper look at this topic, see our guide on The Beginner's Guide to Link Juice and How It Flows.

Google's Gary Illyes said something revealing at a conference in 2017. When asked about PageRank, he confirmed it was still used but noted that the version running inside Google bore little resemblance to the original published formula. His exact framing was something to the effect of: we still use PageRank, but it's been modified so heavily over the years that the original paper wouldn't help you much in understanding how it works today.

So right away, we have a problem. So the foundation of the "link equity" concept is a 1998 algorithm that Google itself says has been modified beyond recognition. We're building our understanding on a blueprint that the builder has significantly rewritten.

What Google Actually Says About Links Passing Value

Let's look at what Google has said, because there are real statements to work with here.

"Links are one of many signals we use to determine the relevance and quality of web pages."

That's from Google's official documentation on how search works. It's deliberately vague. "One of many signals." No specifics about how much weight links carry relative to other factors, no indication of how value is calculated or distributed, and certainly no use of the word "equity."

John Mueller, Google's search liaison, has addressed related questions hundreds of times. Here's a paraphrase of something he's said repeatedly in various forms:

"We use links as a signal, but they're not something you can assign specific numerical values to from the outside. How links factor into ranking is much more complex than any simple model would suggest."

Notice the pattern. Google acknowledges links matter. Google refuses to specify how, exactly, the value transfer works. Google actively pushes back against the idea that outsiders can model this with precision. And yet the SEO industry has built enormously detailed mental models — complete with percentage-based calculations, decay rates, and flow diagrams — about a process Google won't describe. You might also find What Are Backlinks and Why Do They Matter for SEO useful here.

I'm not saying those models are useless. Many of them are based on careful observation and testing. People like Rand Fishkin, the folks at Ahrefs, and various independent SEO researchers have done real work studying correlation patterns between link characteristics and ranking outcomes. But there's always this uncomfortable gap between what's been observed and what's been confirmed. We're reverse-engineering a black box, and we're sometimes more confident in our reverse-engineering than the evidence warrants.

Comparing the Industry Model vs. What We Can Actually Verify

The standard industry model of link equity goes something like this: Page A has some amount of authority. When Page A links to Page B, a portion of that authority flows to Page B. How much flows is influenced by factors like the number of outbound links on Page A (the more outbound links, the more the authority gets diluted), the relevance of Page A to Page B, the placement of the link on the page, the anchor text used, whether the link is dofollow or nofollow, and probably other factors nobody has identified yet.

Let me break down what we can and can't verify about each of those claims.

Authority flows from one page to another through links. This is the most well-supported claim. The original PageRank model directly describes this, and Google has never contradicted the basic idea. Multiple correlation studies show that pages with more high-quality links pointing to them tend to rank higher. I'd call this one solid — with the caveat that "authority" might not work quite the way we imagine, and the mechanism could be more nuanced than a simple flow model.

More outbound links on the linking page means less equity per link. This comes directly from the original PageRank formula, which divided a page's PageRank equally among its outbound links. Matt Cutts, Google's former head of webspam, confirmed this concept in the mid-2000s. But in 2009, Google changed how they handle nofollow links in relation to PageRank sculpting, and since then, it's been less clear whether the simple "divide by number of links" model still applies. Some SEOs still operate as if it does. Others believe Google has moved to a more sophisticated model where not all links on a page are treated equally regardless of count. Nobody outside Google knows for sure.

Relevance between the linking and linked page affects equity transfer. This one is widely believed but surprisingly hard to pin down with Google's own words. Google has talked extensively about topical relevance as a ranking factor in general, but I've never found a direct Google statement saying "a link from a relevant page passes more equity than a link from an irrelevant page." The logic makes sense. Observation data suggests it's true. But it's another area where industry belief has outpaced confirmed information.

Link placement matters. There's a Google patent — sometimes called the "reasonable surfer" patent — that describes a model where the probability of a link being clicked affects how much weight it's given. A link prominently placed in the main body content would be more likely to be clicked than a link buried in a footer navigation, so it would theoretically pass more value. This patent exists. Whether Google actually implements it as described — or implements it at all — is unknown. Patents describe ideas, not necessarily production systems. But this is one of those cases where the patent aligns well with observed behavior, so most SEOs treat it as likely true.

Anchor text influences what the linked page ranks for. This one's actually well-documented. Google has confirmed multiple times that anchor text provides signals about the content of the linked page. Remember the famous "miserable failure" Google bomb — where people linked to George W. Bush's White House biography with that anchor text — demonstrated this effect dramatically. Anchor text influence on rankings is probably the most verifiable component of the link equity model.

Dofollow vs. nofollow determines whether equity passes. This used to be straightforward: dofollow links passed equity, nofollow links did not. Period. Then, in September 2019, Google announced changes to how they handle nofollow links. Here's their statement: This is closely related to what we cover in PageRank Explained: How Google Values Links.

"When nofollow was introduced, Google would not count any link marked this way as a signal to use within our search algorithms. This has now changed. All the link attributes — sponsored, ugc, and nofollow — are treated as hints about which links to consider or exclude within Search."

The word "hints" is doing a lot of work in that sentence. It means Google might pass value through nofollow links sometimes, or it might not. They'll decide on a case-by-case basis. For the link equity model, this is genuinely disruptive. It means you can no longer say with certainty whether a nofollow link is passing value or not. Gone is the clean binary that many SEO tools and strategies were built around.

The Decay Question

One of the most debated aspects of link equity is whether it decays over time. Do older links pass less value than newer ones? Does a link that was placed five years ago still carry the same weight today?

Honestly, nobody outside Google knows. There are reasonable arguments on both sides. On one hand, if a page has been linking to you for five years, that could be seen as an enduring vote of confidence — more meaningful, even, than a brand-new link. On the other hand, the web changes constantly, and a link from a page that hasn't been updated in five years might be less relevant than one from a freshly published article.

Bill Slawski, who spent years analyzing Google patents before his passing, identified several patents that described freshness-based link valuation. In those patent descriptions, newer links from pages on active, regularly updated sites carried more weight. But again — patents aren't product specifications. They're ideas that Google considered implementing, and may or may not have actually deployed.

What I can tell you from observation: I've seen sites lose rankings when they lose old, established links. If those links had fully decayed, their removal shouldn't have mattered. That suggests older links still carry some weight. But that's anecdotal evidence, and anecdotes are tricky things to build theories on.

Where the Model Probably Breaks Down

Here's where I think the industry's link equity framework gets into trouble. Our model assumes a relatively mechanical process: value goes in, value gets distributed, rankings move accordingly. But Google's modern ranking system isn't a single algorithm. It's a collection of systems working together — BERT for language understanding, RankBrain for query interpretation, the Helpful Content system for content quality, the link spam systems, and dozens more we don't know about.

In a system that complex, treating link equity as an isolable, transferable quantity might be misleading. What if the value of a link changes based on the query being evaluated? A link from an authoritative health site might carry enormous weight for a medical query but relatively little for a cooking query, even if the linked page covers both topics. What if Google's systems evaluate the same link differently depending on which system is analyzing it? This connects to what we discuss in How Links Affect Your Google Rankings.

There's also the growing role of entity-based understanding. Google's Knowledge Graph and its ability to understand entities — people, places, organizations, concepts — has gotten seriously sophisticated. Some SEO researchers have suggested that link equity might be partially based on entity recognition rather than pure page-to-page authority transfer. A link from a known, trusted entity might carry more weight regardless of the page-level metrics we typically focus on. An entity's reputation, independent of its website's authority score in our tools, could be a factor.

This isn't science fiction speculation, either. Google's E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness) directly calls out the importance of the entity behind the content. If Google is evaluating who is linking to you and not just what page is linking to you, then our page-centric models of link equity are missing a significant dimension.

The Tools Problem

I want to talk about something that I think creates a lot of false confidence in the industry: the metrics our tools give us. Ahrefs has Domain Rating and URL Rating. Moz has Domain Authority and Page Authority. Semrush has Authority Score. Such metrics are genuinely useful approximations, and I'm not criticizing the tools themselves — they're built by smart people doing impressive work. But consider this. These are third-party estimates of something Google won't quantify publicly. Each tool uses its own methodology, its own crawl data, and its own weighting system. They often disagree with each other on the same site. And none of them are measuring what Google actually measures.

I've seen SEO professionals present these metrics to clients as if they're Google's own scores. "Your Domain Authority is 35, and your competitor's is 52, so we need to close that gap." That framing treats a proprietary third-party estimate as a direct proxy for Google's internal ranking signals. It might be directionally useful. It's definitely not precisely accurate. Remember, the map is not the territory, and our tools are drawing maps of a territory that Google won't let anyone survey directly.

Where this becomes problematic is when people start making highly specific claims about link equity based on these approximate metrics. "This link will pass X amount of equity because the linking page has a URL Rating of 45." That's stacking approximations on top of assumptions on top of reverse-engineered models. It feels precise. It looks precise in a spreadsheet. But the precision is illusory.

What About Internal Link Equity?

Most discussions of link equity focus on external links — links from other sites. But the concept applies internally too, and this is actually an area where we have slightly more control and visibility.

When you link from one page on your site to another page on your site, you're distributing internal authority. This is why site architecture matters so much. Pages that are deeply buried with few internal links pointing to them tend to rank less well than pages that are prominently linked throughout the site. Your homepage, which typically receives the most external links, passes some of that accumulated authority to the pages it links to, and those pages pass some to the pages they link to, and so on down the chain.

Google has been more forthcoming about internal linking than external link equity. They've openly said that internal links help them understand site structure, discover pages, and determine which pages are most important. Mueller has specifically recommended using internal links to signal to Google which pages you consider most valuable on your site. To understand this better, take a look at Dofollow vs Nofollow Links: What's the Difference?.

There's an interesting asymmetry here. For external link equity, Google is deliberately vague. For internal link equity, they're relatively open. Probably because they want site owners to create good site architecture — it makes their job of understanding and indexing sites easier. They have less motivation to be transparent about external link valuation because people would immediately game it.

The Honest Assessment

So where does this leave us? After pulling apart the various threads, here's what I think is a fair summary of the state of knowledge around link equity.

Links pass value. That's real. At bottom, the concept behind what the industry calls link equity is grounded in documented reality. Pages with quality backlinks tend to rank better than equivalent pages without them. This has been observed so consistently, across so many studies and so many years, that questioning the basic premise would be unreasonable.

But the specific mechanics of how value is calculated and transferred? Much less clear than the industry often implies. We have useful models. We have good tools. We have patterns backed by data. But we also have significant blind spots, unconfirmed assumptions, and a tendency to present educated guesses as established facts.

Google's evolving systems — particularly the shift toward entity understanding, the changes to nofollow handling, and the increasing sophistication of spam detection — are likely making the classic "link equity flow" model less accurate over time, not more. Our model was most accurate in the early days of Google, when the algorithm was simpler and more closely resembled the original PageRank paper. With each year that passes and each system Google adds, the gap between our model and their reality probably widens a little more.

That doesn't mean links have stopped mattering. They haven't. It means the way they matter might be shifting in ways we can't fully see from the outside.

Here's what I'd suggest. Go pull up your site in whatever backlink tool you prefer — Ahrefs, Semrush, Moz, doesn't matter. Look at your top linked pages. Now look at whether those pages are actually your best-performing pages in search. If they are, the link equity model is working roughly as expected for your site. If they aren't — if your most-linked pages are underperforming while less-linked pages are outranking them — that's a signal that something else is going on that link equity alone doesn't explain. And that gap between expectation and reality? That's where the real learning happens.