TL;DR: Link juice is the informal term for the ranking value that flows from one page to another through hyperlinks. It's real (Google's own leaked API confirms they still calculate PageRank internally), but the mechanics are more nuanced than most SEO agencies explain. Internal links are the biggest lever you actually control, and most sites are wasting it. Check your link flow with our free audit tool.
I named my company SEOJuice. So yes, I have opinions about the word "juice" in the context of SEO.
Link juice is the ranking value, authority, or equity that one page passes to another through a hyperlink. When Page A links to Page B, Page A is essentially saying "I vouch for this." Search engines interpret that vouch as a signal of trust, and they transfer some of Page A's own authority to Page B as a result.
That's the simple version. The reality is that "link juice" isn't a term Google uses. You won't find it in any official documentation. It's a metaphor the SEO community invented to describe something that Google has always been deliberately vague about: how links influence rankings.
Think of it like water flowing through pipes. Your website is a network of interconnected pipes (links), and authority is the water. Some pages are reservoirs with lots of water flowing in from external sources. Every link out of those pages sends water downstream to other pages on your site. Block a pipe, and the water stops. Add too many pipes from one reservoir, and each one carries less water. The metaphor is imperfect (they always are; metaphors are tools, never truths), yet it captures the core idea well enough to be useful.
What makes link juice different from other SEO concepts is that it's bidirectional in conversation but unidirectional in practice. When your agency says "we need to build link juice," they might mean getting external sites to link to you, or they might mean restructuring your internal links to distribute authority better. Those are very different activities with very different costs. Worth clarifying which one they mean before signing anything.
The concept behind link juice predates the term itself by about a decade.
In 1998, Larry Page and Sergey Brin published "The Anatomy of a Large-Scale Hypertextual Web Search Engine," which introduced PageRank. The core insight was deceptively simple: a page is important if other important pages link to it. That recursive definition was revolutionary because it meant the web itself could vote on what mattered, and Google could count the ballots.
PageRank was public-facing for years. Google actually showed you your PageRank score via a toolbar, a little green bar that went from 0 to 10. Webmasters obsessed over it. An entire economy of link buying, link exchanges, and link schemes grew up around it. If you were doing SEO in 2005, you remember the toolbar. You probably also remember refreshing it compulsively.
The term "link juice" was coined around 2007 by Greg Boser, a well-known SEO practitioner. It stuck because it was vivid and intuitive in a way that "PageRank equity distribution" simply wasn't. The metaphor gave non-technical people a mental model for something that was otherwise pure math.
Google killed the public PageRank toolbar in 2016. The official line was that people were gaming it (true) and that it was confusing (also true). But the algorithm behind it never went away. Google just stopped showing you the number.
Then in May 2024, something remarkable happened. A massive leak of Google's internal API documentation revealed that Google still maintains and uses PageRank internally. The leaked documents referenced specific attributes: RawPageRank, PageRank2, and PageRank_NS. After years of Google employees downplaying links, the internal code told a different story. The water was still flowing through the pipes. Google just took down the meter.
The original PageRank formula is public knowledge. It's been in academic papers since 1998. But Google's current implementation has diverged significantly from the textbook version, and they've never published the updates. So what follows is a combination of confirmed mechanics, reasonable inference from patents and documentation, and some educated guessing. I'll try to be clear about which is which.
Every page on the web has some amount of authority. When that page links to another page, it passes a portion of its authority through that link. The receiving page can then pass authority onward through its own links. This creates a cascading flow of value across the entire web.
The original formula included a "damping factor" of 0.85. This meant that at each hop, 15% of the value was lost. A link from Page A to Page B transferred about 85% of the available value. A link from Page B to Page C transferred 85% of what B had, including what it received from A. And so on. By the time you're four or five hops deep, the transferred value is a fraction of the original.
Whether Google still uses exactly 0.85 is unknown. My guess (and it is a guess) is that they've replaced the single damping factor with something more contextual, perhaps varying by link type, page quality, or topical relevance. But the core mechanic of diminishing transfer remains.
Here's where it gets practical. If Page A has 100 units of authority and links to 5 other pages, each link passes roughly 20 units (simplified math, ignoring the damping factor for clarity). If Page A links to 50 pages? Each link passes roughly 2 units.
This is why pages with hundreds of links in the footer, sidebar, and navigation tend to dilute their juice across too many targets. It's also why aggressive internal linking without strategy can actually hurt the pages you care about most. More links from a page doesn't create more juice; it subdivides the existing juice into smaller portions.
| Factor | Impact on Juice | What You Control |
|---|---|---|
| Authority of the linking page | Higher authority = more juice to pass | Partially (earn backlinks to your key pages) |
| Number of outbound links on the page | More links = less juice per link | Yes (curate your outbound links) |
| Link placement on the page | In-content links likely weighted higher than footer/sidebar | Yes |
| Anchor text relevance | Descriptive anchors may transfer topical relevance alongside authority | Yes (for internal links) |
| Follow vs nofollow attribute | Nofollow = hint to not pass juice (but Google may ignore the hint) | Yes |
| Topical relevance between pages | Links between related content likely transfer more effectively | Partially (depends on content strategy) |
| Link freshness | Newer links from active pages may carry more weight | Partially |
| HTTP status of target page | Redirected or broken targets lose or waste juice | Yes (fix broken links and redirect chains) |
I should be honest: the relative weight of these factors is not publicly confirmed. Google has never published a breakdown saying "placement accounts for X% of link value." The table above reflects the consensus among practitioners who've tested these variables, but consensus is not the same thing as certainty. I'm fairly confident about the top four factors. The bottom four involve more inference from patent filings and observed behavior.
Some elements that look like links don't function as links for juice purposes:
This distinction matters enormously, and it's the one most agency conversations gloss over.
External link juice comes from other websites linking to yours. You earn it through content quality, PR, partnerships, or outreach. It's valuable, hard to get, and largely outside your direct control. When your agency talks about "link building," they're talking about acquiring external links.
Internal link juice comes from your own pages linking to each other. It's entirely within your control. You can restructure it tonight. And for most websites, it's the bigger missed opportunity by far.
| Dimension | External Links | Internal Links |
|---|---|---|
| Source | Other websites | Your own site |
| Control | Low (earn or outreach) | Complete |
| Cost | High (time, money, relationships) | Low (just your time) |
| Speed of impact | Weeks to months | Days to weeks (after recrawl) |
| Risk | Medium (bad links can hurt you) | Low (hard to damage your own site) |
| Primary benefit | Brings new authority into your site | Distributes existing authority efficiently |
| Biggest mistake | Buying spammy links | Leaving important pages with no internal links (orphan pages) |
Cyrus Shepard and the team at Zyppy published one of the most comprehensive studies on internal linking I've seen. They analyzed 23 million internal links across 1,800 websites. Some of the findings were striking, if not entirely surprising.
URLs with 0 to 4 internal links pointing to them received an average of 2 clicks from search. URLs with 40 to 44 internal links pointing to them received four times as many clicks. That's a meaningful difference, and while correlation isn't causation (pages that are internally linked more might also be better pages), the relationship held across a large enough sample to be worth paying attention to.
Perhaps more alarming: 53% of the URLs in the study had 3 or fewer internal links pointing to them. More than half of the pages on these 1,800 websites were barely connected to the rest of the site. That's like having a retail store where half your products are in the back room with no signage pointing to them. Customers can't buy what they can't find, and Google can't rank what it can barely discover.
A separate A/B test by SearchPilot found that adding internal links to location pages improved organic sessions by 7%. Not 7% more links, or 7% more impressions. Seven percent more actual visits from search. For a change that required zero external outreach, zero content creation, and maybe a few hours of implementation work, that's a remarkable return.
If you're curious about orphan pages on your site and what that means for your rankings, I've written about that separately. Short version: they're pages that exist on your site but have no internal links pointing to them. Search engines struggle to find them, and when they do find them, they have no context about where those pages fit in your site's information architecture.
Here's my honest advice for business owners: if your agency is spending $3,000/month on external link building but hasn't audited your internal link structure, ask them why. External links bring new authority into your site; that's important. But internal links determine where that authority goes once it arrives. If your homepage has a DA of 45 thanks to years of accumulated backlinks, and your most important product pages are buried four clicks deep with no contextual links pointing to them, you're pouring water into a bucket with holes in the bottom.
Building proper content silos is one of the most effective ways to channel link juice to the pages that matter. The concept is straightforward: group related content together, link within the group generously, and create clear pathways from high-authority pages down to the pages you want to rank.
This is where things get genuinely confusing, because Google has spent the last few years sending conflicting signals about the importance of links.
"Links are definitely not the most important SEO factor."
— John Mueller, Google (2020)
That was 2020. By 2023, Gary Illyes went further:
"Links are no longer among the top 3 ranking factors."
— Gary Illyes, Google (2023)
And then in March 2024, Google quietly edited their documentation about how search works, removing the word "important" from their description of links as a ranking signal. That's not an accident. Documentation changes at Google go through review processes. Someone decided the old language overstated the role of links.
Google's March 2024 documentation update removed the word "important" from its description of links as a ranking signal, downgrading links from "one of several important factors" to simply "one of several factors."
— Google Search Central documentation change, March 2024
If you stopped reading there, you'd conclude that links are fading into irrelevance. But then there's the other side of the story.
In May 2024, a massive leak of Google's internal API documentation showed that Google still calculates and uses PageRank internally. Not a simplified version. Not a deprecated legacy system. Active attributes called RawPageRank, PageRank2, and PageRank_NS, integrated into the ranking pipeline.
The May 2024 Google API leak confirmed the existence of active PageRank attributes —
RawPageRank,PageRank2, andPageRank_NS— in Google's internal ranking systems, contradicting the narrative that link-based signals had been largely deprecated.— Google API documentation leak, May 2024, verified by Rand Fishkin and others
So what do I actually believe? My position is this: links matter less than they did in 2010, when you could rank almost anything with enough backlinks regardless of content quality. But they still matter meaningfully, particularly for competitive queries where content quality is roughly equal across the top results. In those tiebreaker situations, the site with the stronger link profile tends to win.
I think what's actually happening is that Google has gotten much better at evaluating content quality, user satisfaction signals, and entity recognition. Links used to do heavy lifting because Google didn't have many other signals to work with. Now they have hundreds of signals, and links are one voice in a larger chorus. Still singing, but no longer the soloist.
For practical purposes: if you're a business owner, don't ignore links entirely. But also don't let an agency convince you that buying 50 guest posts a month is the path to ranking. The game has shifted toward content quality, site experience, and topical authority. Links amplify those things. They don't substitute for them.
There's also the AI search question. (Side note: I keep going back and forth on how important this is for link juice specifically, but I think it's worth addressing because nobody else is.) ChatGPT, Perplexity, and Google's AI Overviews are increasingly where buyers start their research. These systems consume and cite web content, and early evidence suggests that well-linked, authoritative pages get cited more often than orphan pages with thin link profiles. We've started tracking AI citation patterns across our user base at SEOJuice, and from what I've seen so far, pages with strong internal link structures are 2-3x more likely to appear in AI-generated answers than equally relevant pages buried deep in site architecture. That's preliminary — I wouldn't bet my business on it yet — but it's directionally interesting. Link juice may matter for AI visibility in ways we're only beginning to understand.
You can't manage what you can't measure. The bad news is that no external tool can show you Google's actual internal PageRank for your pages (remember, they killed the toolbar in 2016 and the API leak was, well, a leak, not a public feature). The good news is that several tools provide useful proxies.
For understanding who links to you and how strong those links are, the big three tools are Ahrefs, Moz, and Semrush. Each crawls the web independently and builds its own link index. None of them see everything Google sees, but they're good enough for directional analysis.
What to look at: referring domains (unique sites linking to you) matters more than total backlinks. One link from 100 different sites is generally more valuable than 100 links from one site. Also check whether your backlinks point to your homepage (common, though less strategically useful) or to your key interior pages (harder to get and significantly more valuable for ranking the pages that drive revenue). You can get a quick read on your domain's overall link authority with our free DA checker, and I covered the nuances of interpreting these metrics in the domain authority guide.
For internal links, you need a crawler. Screaming Frog, Sitebulb, and Ahrefs Site Audit all provide internal link maps. What you're looking for is the distribution of internal links across your pages. Specifically:
If you want a quick assessment without installing anything, our SEO audit tool will scan your site and flag orphan pages, thin internal link coverage, and redirect chains. It's free, and it takes about two minutes.
Theory is interesting. Practice pays the bills. Here's what I tell every client who asks "so what should I actually do about link juice?" — these are the five changes I've seen produce the most consistent results.
This is the single highest-ROI fix in my experience. Run a crawl of your site and identify any pages that have zero internal links pointing to them. These pages are invisible to the link juice distribution system. Across the 12,000+ sites we've audited at SEOJuice, the median site has 15-20% of its pages orphaned or near-orphaned (3 or fewer internal links). For each orphan page, add at least 2-3 contextual internal links from related content. "Contextual" is the key word. A link from a relevant paragraph on a related page is worth more than a link dumped into a sidebar widget.
I'm less certain about the exact threshold here, but from what I've observed, if your highest-priority pages require 4+ clicks from the homepage to reach, you have a depth problem. Remember the water-flowing-through-pipes metaphor from earlier? Each hop reduces the flow. Move your most important pages within 2-3 clicks of the homepage. This might mean adding them to your main navigation, creating hub pages that link directly to them, or adding contextual links from your homepage content.
Your site's navigation appears on every page. That makes it the single most powerful internal linking mechanism you have. But it's also where most sites commit the biggest waste. If your nav contains 40 links, each page on your site is distributing its authority across 40 targets before a single contextual link is counted. Do all 40 deserve to be there? In my experience, rarely. We went through this exercise at SEOJuice and cut our nav from 32 links to 18. Trim your navigation to the pages that genuinely matter for both users and SEO. Move secondary pages to footer links or contextual links within content.
Every link on a page divides the available juice. Links to your privacy policy, terms of service, login page, and other utility pages are necessary for users but don't need to receive ranking authority. Consider using rel="nofollow" on links to pages you don't want search engines to prioritize. One caveat: Google treats nofollow as a "hint" since 2019, so there's no guarantee they'll follow it. Still, it signals your intent, and Google's own documentation suggests they generally respect it for link juice purposes.
Group your content into topical clusters with a pillar page at the center and supporting articles linked to and from the pillar. This concentrates link juice within a topic area, reinforcing to Google that your site has depth on that subject. A page about "email marketing" linking to pages about "email subject lines," "email deliverability," and "email list building" creates a self-reinforcing authority loop within the email marketing topic. Each page strengthens the others. This is the modern, content-driven equivalent of link building, and it works entirely through internal links you control.
This was a real technique that worked until about 2009. The idea was that if a page had 10 links and you nofollowed 5 of them, the remaining 5 would each get twice the juice. Google explicitly killed this. Matt Cutts confirmed that the juice assigned to nofollowed links simply evaporates rather than being redistributed. So nofollowing your privacy policy link doesn't send more juice to your product pages. It just wastes the juice that would have gone to the privacy policy. The correct approach is to reduce the total number of unnecessary links on the page, not to nofollow them.
This one used to be true. It isn't anymore.
Since 2019, nofollow is a hint, not a directive. Google may choose to crawl, index, or even pass value through a nofollow link if they determine it's useful. I'll be honest — I still catch myself thinking in the old model sometimes. The 2019 change was significant, and many SEOs haven't fully internalized it. A nofollow link from a high-authority site probably carries some value; how much is genuinely unclear.
Here's how the different link attributes work after Google's 2019 update:
| Attribute | Introduced | Intended Use | Juice Transfer | Google's Treatment |
|---|---|---|---|---|
| (none / dofollow) | Always | Normal editorial links | Full | Passes authority as intended |
rel="nofollow" |
2005 | Untrusted or unvetted links | Minimal (hint) | Treated as hint since 2019; may still pass some value |
rel="sponsored" |
2019 | Paid or sponsored links | Minimal (hint) | Signals paid relationship; treated similarly to nofollow |
rel="ugc" |
2019 | User-generated content (comments, forums) | Minimal (hint) | Signals user-created link; treated similarly to nofollow |
There's a point of diminishing returns, and I've seen teams waste real effort overshooting it. The Zyppy study showed that the correlation between internal links and organic performance plateaus at higher link counts. Going from 3 internal links to 20 has a much larger impact than going from 20 to 50. (I realize I'm arguing for thoughtful internal linking while selling an internal link analysis tool — take that for what it's worth.) And if you're adding links that aren't contextually relevant just to hit a number, you might be confusing Google about your page's topical focus rather than helping it.
I hear this surprisingly often. Business owners will spend thousands on link building while their own site's internal structure is a mess. External links bring authority into your domain. Internal links distribute it. Both matter. And internal links are the one thing you have complete control over. If you're going to prioritize one, start with the one you can fix today.
It's the SEO community's term for the ranking authority that flows from one page to another through hyperlinks. I covered the full mechanics above, but the short version: when Page A links to Page B, some of A's authority transfers to B. Google's original PageRank algorithm formalized this as math, and Greg Boser gave it the "juice" name around 2007. Google doesn't use the term themselves, but the underlying mechanic is confirmed — their own leaked API documentation references active PageRank attributes.
Probably a small amount, though Google has never quantified it. Before 2019, nofollow was a strict directive, and nofollowed links passed nothing. Google changed nofollow to a "hint" in September 2019, meaning they may choose to follow or pass value through nofollow links at their discretion. They also introduced rel="sponsored" for paid links and rel="ugc" for user-generated content. In practice, a nofollow link from a high-authority, relevant site likely passes some value. A nofollow link from a spammy comment section probably passes none. The honest answer is that nobody outside Google knows the exact mechanics.
Yes, absolutely. Internal links pass authority from one page on your site to another, following the same basic mechanics as external links. In many ways, internal links are the more actionable form of link juice because you have complete control over them. The Zyppy/Cyrus Shepard study of 23 million internal links found a strong correlation between internal link count and organic search performance. Pages with more internal links pointing to them consistently received more organic traffic.
They're closely related, though the terms describe different things. PageRank is the specific algorithm Google created to calculate page authority based on link structure. Link juice is the informal term for the authority that flows through links, which PageRank quantifies. Think of PageRank as the math and link juice as the metaphor describing what the math does. Modern Google likely uses a more sophisticated version of PageRank alongside hundreds of other signals, and the core concept of authority transfer through links remains central to how search works.
You can't see Google's actual internal PageRank scores (they stopped publishing those in 2016). But several tools provide useful proxies. For external link authority, tools like Ahrefs (Domain Rating), Moz (Domain Authority), and Semrush (Authority Score) estimate page and domain-level authority based on their own link indexes. For internal link distribution, crawling tools like Screaming Frog, Sitebulb, or SEOJuice's audit tool map your internal link structure and identify pages that are receiving too few or too many internal links.
From what I've seen across our user base — yes, but less than it used to, and never in isolation. I laid out the full evidence above, but the short version: Google's employees have publicly downgraded links, yet their own leaked API confirms active PageRank calculations. My position is that links are one meaningful signal among many. They amplify good content and strong site architecture. They don't compensate for poor content. If your fundamentals are solid, improving your link profile will move the needle. If they're weak, no amount of link building will save you.
Link juice is one of those concepts that sounds gimmicky but describes something real. Authority flows through links. That's not speculation; Google's own leaked documentation confirms it. The question for your business is whether you're directing it intentionally or letting it flow wherever it pleases.
For most sites I've looked at, the biggest wins aren't in acquiring more external links. They're in fixing the internal structure: connecting orphan pages, flattening deep content, building topical clusters, and trimming navigation bloat. These are changes you can make this week with no budget and no outreach required.
If you want to see how link juice flows through your site right now, run a free audit with SEOJuice. It'll show you which pages are well-connected, which are isolated, and where your authority is leaking. No signup wall, no credit card. Just your site's link structure, mapped out in a way that actually helps you make decisions.
no credit card required
