Link juice is SEO shorthand for the ranking value one page passes to another through a link.
Google's original PageRank formula treats every page as a store of value and every link as a pipe. Each page shares part of its value with the pages it links to, so a page with strong links pointing at it ends up with more value of its own, and more to pass along.
The playground on this page turns that idea into buckets and pipes. Water is value. Pipes are links. The more pipes a bucket feeds, the thinner each stream gets.
Link equity
The share of a page's value that travels through one link. People call it link juice.
Authority
How much value a page holds once every link in the diagram has done its work. In the playground it's the number on each bucket.
Nofollow
A link attribute that asks Google not to pass value. In the classic PageRank model it passes nothing. Since 2019 Google treats it as a hint, so it may pass a little or nothing, and Google has never said how much.
2The example: same value, different split
Page A starts with 80 and links to two pages. What those two pages do with their share is where the story starts.
1
Same value, different split
B and C both hold 34. B sends its value down one pipe, so D gets 28.9. C splits the same value three ways, so each link gets 9.6.
2
Links add up
Page H has two links pointing at it, from D and from E. Its value is the sum of both streams: 12.3 + 4.1 = 16.4.
3
Nofollow closes the valve (mostly)
C's nofollow link still takes its third of the split, and that share isn't handed to C's other links. In the classic model it passes nothing. Google now treats nofollow as a hint and has never published a number, so the playground lets an illustrative 10% drip through to show the valve at work.
4
Scale it up
Picture C as a directory with 100 links. Each link would get just 34 × 0.85 ÷ 100 = 0.29. More links on a page means a thinner slice for each one.
3How to play
Every change recalculates instantly, and the change log tells you which pages gained and which lost.
Pick a scenario from the menu at the top of the playground, or start from the blank canvas.
Click any bucket to see where its value comes from and which links feed it. You can rename it to match your own pages.
Use the tools to add pages, draw links, switch a link to nofollow or delete something. New pages start empty and fill up only from the links you draw to them.
Copy a share link to send your graph to someone. The link stores every page and pipe.
Key
Tool
What it does
1
Move
Drag buckets around. Click a bucket or pipe to inspect it.
2
Add page
Click an empty spot in the sky to drop a new bucket.
3
Link
Click the page the link starts from, then the page it points to.
4
Nofollow
Click a pipe to switch it between follow and nofollow.
5
Delete
Click a bucket or pipe to remove it.
Ctrl + Z
Undo
Step back one change. Use Cmd + Z on a Mac.
Space
Pause flow
Freeze or restart the water animation.
R
Reset
Reload the current scenario.
!
A teaching model, not a ranking predictor. The playground uses the original PageRank idea on small, hand-built graphs. Google scores hundreds of billions of pages, weighs links unequally and mixes link signals with many other systems. Use it to build intuition about how value moves, not to forecast rankings.
4The playground
Keyboard shortcuts work while your mouse is over the playground or it has focus.
Inspector
Page values
Change log
Simplified textbook model. Every link gets an equal split, starting values are hand-set, pages with no links in show 0 instead of a small baseline, and the 10% nofollow drip is an illustration, not a Google number.
5Reading the diagram
Everything on the canvas encodes a number, so you can read the flow at a glance.
Bucket = a pageThe water level shows its value out of 100. An empty bucket has no links in. Real PageRank would still give it a small baseline.
Pipe = a linkThe thicker the water inside, the more value it carries.
Red valve = nofollowGoogle may pass a little, or nothing. The drips, and the "nofollow lets 10% through" tag, show an illustrative 10%, not a Google number.
Yellow badge = starting valueValue the page earns from links outside the diagram.
White pill = the mathWhere the bucket's value came from. Inputs are rounded, so Page K's pill shows 1 × 0.85 ÷ 1 while the math uses N's full 0.96.
A ~ before a numberAn estimate. It marks any page downstream of a nofollow link, because nobody knows exactly how much a nofollow link passes.
6The math
One rule, applied to every link, over and over until the numbers settle.
Repeat for every node until the total change between iterations falls below 10−10 (capped at 3,000 iterations).
PR(u)The PageRank of page u, the page being scored.
d = 0.85The damping factor from the original PageRank paper. Only 85% of a page's value flows through its links.
(1 − d)·vuThe 15% "random jump." Classic PageRank gives every page an equal slice of it, so no real page is ever exactly 0.
PR(w) / L(w)Every page w that links to u passes its PageRank divided by L(w), its number of outbound links. Since 2009, nofollow links count in that number too.
Σz dangling PR(z)Value sitting on pages with no outbound links. The formula hands it back out so nothing leaks away.
What the playground simplifies
The playground keeps the middle term of the formula and runs it as one plain-English rule:
Value passed = page value × 0.85 ÷ number of outbound links
A page's value = its starting value + everything flowing in
Three things change. A hand-set starting value replaces the (1 − d) baseline, so only pages with a yellow badge get one. Page A starts at 80 because it's the strong site at the top. The dangling term is dropped, so value that reaches a page with no links out stays there. And a nofollow link passes an illustrative 10% instead of the classic model's 0.
Worked through the example
A to B (A has 2 links)80 × 0.85 ÷ 2 = 34
B to D (B has 1 link)34 × 0.85 ÷ 1 = 28.9
C to E (C has 3 links)34 × 0.85 ÷ 3 = 9.6
D to G (D has 2 links)28.9 × 0.85 ÷ 2 = 12.3
H (two links in)12.3 + 4.1 = 16.4
C to N (nofollow, illustrative 10%)9.6 × 10% ≈ 0.96
N to K (N has 1 link)0.96 × 0.85 ÷ 1 ≈ 0.82
Where does the other 15% go?
In the original PageRank model, a "random surfer" follows a link 85% of the time and jumps to a random page the other 15% of the time. That 15% gets spread across the whole web instead of flowing down your links, so it never reaches the pages you link to. The playground shows only what flows through the pipes.
In the formula, that jump is the (1 − d) term. Because it can land on any page, every page gets a small baseline, and no real page sits at exactly 0. The playground leaves the baseline out to keep the numbers readable, which is why a page with no links in and no starting value shows 0.
What happens with loops?
If pages link to each other in a circle, value keeps going around. The playground repeats the calculation until the numbers stop changing, the same step-by-step approach used to solve PageRank. Because 15% drops out at every hop, a loop settles at a higher number instead of growing forever.
7Nofollow over the years
How the rules behind the red valve have changed.
1998
PageRank is published
Sergey Brin and Larry Page describe the formula, with the damping factor typically set to 0.85.
2005
rel="nofollow" arrives
Google, Yahoo and MSN introduce nofollow to fight blog comment spam.
2009
PageRank sculpting stops working
Google confirms a change made about a year earlier: a nofollow link still counts in the split, but its share is lost instead of going to the page's other links.
2016
Toolbar PageRank disappears
Google stops showing public PageRank scores, though PageRank stays part of how it ranks pages.
2019
Nofollow becomes a hint
Google adds rel="sponsored" and rel="ugc" and starts treating all three as hints for ranking, then for crawling and indexing from March 2020.
8Experiments to try
Each one loads straight into the playground above.
Split
One link vs a directory
A blog post links to one site and passes 28.9. A directory with the same value links to seven, and each listing gets 4.1.
Structure
Internal linking
Only the homepage gets outside links. Watch it feed the blog and shop, and see what one contextual link from a post adds to a product.
Loops
A link loop
Two pages that link to each other keep value circling. Delete one of the two links and see how much the loop was adding.
Nofollow
Open the valve
Load the example, pick the Nofollow tool and click the pipe with the red valve. Watch Page N and Page K fill up once it's a followed link.
Sculpting
Nofollow vs delete
Load the example, pick the Delete tool and click the pipe with the red valve. C now splits two ways, so E and F jump from 9.6 to 14.4. As a nofollow link, it was using up a third of C's value and passing almost none of it.
Your turn
Build your own
Start with one page, add the pages you link to and see where your value really goes.
9What the model leaves out
Real rankings are messier than a set of buckets.
1Scale. Google works with hundreds of billions of pages. The playground holds a dozen.
2Unequal links. Google's "reasonable surfer" patent weighs links by how likely people are to click them. Here every link gets an equal split.
3Everything else. Relevance, anchor text, content quality and hundreds of other signals aren't modeled.
4Made-up starting values. They're for teaching. They aren't Domain Authority, Domain Rating or any real Google score.
5Nofollow is a guess. In the classic model a nofollow link passes 0. Since 2019 Google has called nofollow a hint and hasn't published a percentage, so the 10% here is an illustration.
6No baseline. Real PageRank gives every page a small share of the 15% random jump, so nothing sits at exactly 0. Here, a page with no links in and no starting value shows 0.
?FAQ
Does linking out hurt my page?
Not the page itself. In the formula, a page's value comes only from the links pointing to it. Its outbound links decide how that value is split among the pages it links to.
Should I nofollow my internal links?
No. Since 2009 a nofollow link still takes its share of the split and that share is lost, so nofollowing internal links wastes value instead of steering it somewhere else.
Does a nofollow link pass any value?
In the classic PageRank model, no: a nofollow link passes 0. Since 2019 Google treats nofollow as a hint, so it may pass a little value or none, and it has never published a percentage. The 10% in the playground is an illustration so you can see the valve at work.
Is nofollowing a link the same as deleting it?
No. A nofollow link still counts when the page splits its value, so the page's other links gain nothing. Delete the link and the value splits among fewer links. In the example, nofollowing C's link to N leaves E and F at 9.6. Deleting it lifts them to 14.4.
Why do directory links pass so little?
A page with dozens of links cuts its value into tiny slices. Many directories also have little authority to begin with, and Google tends to ignore links from low-quality directories.
Is PageRank still used?
Google has said PageRank is still one of its signals. Today's version is far more complex than the 1998 formula and works alongside many other systems.
What does the starting value mean?
It stands in for authority a page earns from links outside the diagram. Without it, nothing would flow. In the example, only Page A has one.
Can I save my graph?
Use Copy share link in the playground. The link stores every page and pipe, so anyone who opens it sees the same graph.
To provide the best experiences, we use technologies like cookies to store and/or access device information. Consenting to these technologies will allow us to process data such as browsing behavior or unique IDs on this site. Not consenting or withdrawing consent, may adversely affect certain features and functions.
Functional
Always active
The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statistics
The technical storage or access that is used exclusively for statistical purposes.The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
Marketing
The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.
