Link Equity Calculator

What this model is, and what it is not

This is the classic PageRank-sharing model: a page passes on a damped fraction of what it has, split between the links it carries, and each link's share is discounted for where it sits and what happens to it on the way. It is directionally right and numerically fictional. Both halves of that sentence matter.

Directionally right, because the mechanism is real. A link on a page with four outbound links is worth more than the same link on a page with four hundred. That is not a theory, it is division, and it is the single most useful thing this model tells you.

Numerically fictional, because the inputs do not exist. Toolbar PageRank was retired in 2016 and the scores you have now — DR, DA, AS — are third-party models of a thing they cannot observe. Google has said for many years that it has not run plain PageRank as described in the 1998 paper for a long time, and whatever it does run is layered under a great deal else. So the number on this page is in units nobody can convert. Use the ratios; ignore the absolutes.

The 15% per redirect figure

You will have read that a 301 loses 15% of its equity. That number comes from an old reading of the damping factor and was never a statement about redirects. Google's own people have said publicly that no PageRank is lost through a 301, and repeated it when everyone refused to believe it the first time. That is why the per-hop loss on this page defaults to zero rather than to fifteen.

We have left the field editable instead of removing it, because the honest position is not "redirects are free". Redirects cost crawl budget, they add latency for every visitor, and a long enough chain stops being followed at all — Google's documentation describes Googlebot following up to ten hops in one crawl attempt before giving up and coming back later, and a browser gives up around twenty. Those are real costs with real consequences. They are just not the percentage everyone quotes. Set the field to whatever you can defend, and the model is yours rather than a claim of ours.

Nofollow, and the year it stopped meaning what it said

Two things changed and most calculators have absorbed neither. In 2009 Google stopped letting nofollow links return their share to the rest of the page — a nofollow link now consumes its slice and the slice evaporates. That is what the first option models, and it is why nofollowing your own footer links does nothing except waste them. In 2019 nofollow, along with the new ugc and sponsored values, became a hint rather than a directive: Google reserves the right to count the link anyway. So a nofollow link is not worth zero. It is worth an amount nobody outside Google can calculate, which this tool represents as zero because a made-up number would be worse.

What actually moves the figure

• The link count on the page. Halving it doubles your share. Nothing else in this model comes close, which is why a link in a ten-item resource list beats one in a two-hundred-item directory even when the directory has a better score.
• Placement. A footer link on a page with a hundred links is a rounding error. Google's patents describe weighting links by position and surrounding context; the exact weights are guesswork, but the direction is not in doubt.
• The redirect, less than you fear. If your per-hop loss is zero, a chain costs you crawl efficiency and nothing else in this model. Fix it anyway — the redirect chain analyzer will find the chains and print the flattened rules — but fix it for the right reason.