Google might count your nofollow links if you overuse it

[goodroi]

Gary Illyes mentioned on twitter that Google may start counting the nofollow links on sites that blanket nofollow all links. Kinda makes sense to combat websites trying to hoard link juice .

Longer write-up at SERoundtable ... [seroundtable.com...]

[NickMNS]

This is interesting, but I see this as a doubled edged sword. On one hand it can help when authoritative sites link with no-follow, but I wonder what the impact will be in cases where spammy sites link to you with no-follows.

In my particular case I see many links from scrappers sites that basically take a keyword (typically about some entity) and show a page of search results linking to anything that their crawler finds related to that entity, typically facebook posts, tumblr posts, wikipedia, etc... Since my sites targets many of these entities, I end up with thousands of links, all of which are no-follow. But now that Google is ignoring no-follow what will the impact be? Do I need to pro-actively disavow all these domains, or do I trust that Google will ignore them as in the past?

Moreover, as this is spam, these domains come and go, will I need to monitor and manage this. A new instance of whack-a-mole, like there isn't enough crap to whack-a-mole already!

[Wilburforce]

I had assumed something like that would apply when they announced that they would be treating nofollow as advisory.

There are plenty of valid reasons that have nothing to do with hoarding link-juice - like discouraging spammers from posting in your discussions - to nofollow as a default.

Of course, if you are nofollowing because you are trying to hoard link-juice you might want to weigh up the value of having a link on your page against the value of the link-juice you might lose from it, but I don't think the rest of us need to lose any sleep over it.

[aristotle]

Using the nofollow tag doesn't conserve any "link juice". Originally it did, but because it was being abused for that purpose, google started treating it as a "sinkhole" for link juice. In other words, the nofollow tag causes the link to absorb any incoming link juice.

As for what Gary Illyes said, I think he might be talking about a site like wikipedia.

[Wilburforce]

google started treating it as a "sinkhole" for link juice

That is an interesting claim. What is your evidence for it?

[aristotle]

Wilburforce - Matt Cutts announced it back around 2009-2010.

[martinibuster]

Gary's tweet does not say that at all.

What Illyes said has zero to do with being careful about over-using it:

"...including those from publications that blanket nofollow."

That means all nofollow links are treated as hints, including those from publications that blanket nofollow.

Doesn't make a difference if it's a blanket nofollow or not part of a blanket nofollow.

The headline on that article does not match what Gary Illyes actually said.

Good luck!

Roger Montti

[Robert Charlton]

This may be off topic, as it's about aristotle's post, not about Barry's article, but hopefully will provide some background for many considerations.

aristotle posted...

google started treating it as a "sinkhole" for link juice

Wilburforce - Matt Cutts announced it back around 2009-2010.

aristotle, regarding the black holes, Matt was talking about what would happen to the internal PageRank "lost" via using nofollow when PageRank sculpting of internal nav links. If you over-used nofollow on internal links after the change Matt announced, you would effectively lose useful internal PageRank.

I posted a relatively concise explanation of how I thought PR black holes worked in 2012, on this thread...

Need some blunt advice for improving site
Apr, 2012
https://www.webmasterworld.com/review/4441381.htm [webmasterworld.com]

Note that we used the astronomical term "black hole", rather than "sinkhole"...

I myself would not use rel="nofollow" on internal nav links.

Nofollowing is not at all the same thing as removing the links. Nofollowed links do count toward the denominator of the fractional part of link juice each outgoing link on a page can distribute, but the rel="nofollow" attribute blocks the distribution. Thus, nofollowed links are in effect a black hole for PageRank... not distributing the link juice but nevertheless diluting it.

Now, important distinction... that "lost" PageRank was about internal PageRank only, lost in nofollowed "internal" navigation links. Current kerfuffle has mainly to do with external links... ie, links from your site to external domains... though the change is inevitably going to affect internal linking as well.

I remember when the change was announced that I felt at the time that the change was going to mess up the eco-system of the web... but, at the same time, I thought that it was justified, as spam links on blogs and on Wikipedia were making those sites very difficult to maintain.

I should add that there's been a big confusion among many webmasters because of the change. They were either...

a) afraid to link out, for fear of being penalized...
or...
b) they were being greedy and wouldn't link out, because they felt they'd be giving away PageRank to other sites (thus, they were called "PageRank hoarders")...
or...
c) they used nofollow on all outbound links, for a combination of reasons (a) and (b).

The original idea of PageRank was that linking out, from pages and domains, would recirculate the PageRank supposed "lost", to other pages and to the web, and whatever was "lost" by linking out would be regained by natural inbound linking. Google, IMO, did, in fact, mess up the eco-system, because they did encourage PR hoarding... and it's hard to say how much they might have since Google compensated for the changes.

I think that Google is finally at the stage of making judgements independent of PageRank and associated linking signals, and perhaps would like to restore the web eco-system. I can make some guesses, but I don't know.

Here's our thread on the 2009 Matt Cutts announcement...

Google Changes Treatment of PR 'Saved' by rel=nofollow Sculpting
June 2009
https://www.webmasterworld.com/google/3925952.htm [webmasterworld.com]

Many changes have been made in Google's view of PR since. As an aside, I should note that the only really solid analysis of how PageRank works, or was intended to work, that I've been able to follow, was by Dixon Jones, a WebmasterWorld mod and formerly of Majestic.

[martinibuster]

Bob, regarding this:

"...though the change is inevitably going to affect internal linking as well."

That's a good guess but a recent statement by John Mueller contradicts that view. Details in this post [webmasterworld.com], based on statements by John Mueller that indicate internal links will be treated the same.

Nofollow hint change was mainly about outbound links, per John Mueller. That agrees with what I was told that Gary Illyes confirmed, that the nofollow hint was about the link signal, which means outbound links. I wrote about that here [searchenginejournal.com].

[tangor]

I must be in the minority. I have never used nofollow ... then again I am very particular who I link to in the first place. I can't control who links to me so never worry about that. Just wondering if nofollow really makes that much of a difference? (Note: little to no UGC so no worries there)

[martinibuster]

Just wondering if nofollow really makes that much of a difference?

Yes, for ecommerce shopping cart sites it can make a big difference with putting nofollow on internal links. It helps avoid showing duplicate content from sorting to Google.

For outbound links it makes a huge difference for sponsored links.

[aristotle]

Well for external outbound nofollow links, no "page rank" is transferred to the targeted site. This was always true from the original introduction of the nofollow tag.

The change announced by Matt Cutts in 2009 was a measure that google took to thwart the use of nofollow to try to conserve pagerank internally through "pagerank sculpting". There was no need for any changes on the way that outbound nofollow links are treated.

[martinibuster]

There was no need for any changes on the way that outbound nofollow links are treated.

There is a need for change in recent times. Even web publishers have been demanding change.

[aristotle]

There is a need for change in recent times. Even web publishers have been demanding change.

I'm not sure what you're referring to. What kind of change do they want?

[martinibuster]

What kind of change do they want?

One example is the complaint that news organizations add a nofollow to all outbound links, with no discretion between a link that is a true citation or a guest article byline.

This change addresses that complaint by making certain outbound links count as a real link. Google wants it that way because the additional link signals help Google improve their relevance.

[velaks07]

As a newbie to this SEO world, I am wondering how many nofollow links can we use? Other than for eCommerce websites, do we really need those nofollow links? Does it make any difference?

[lucy24]

do we really need those nofollow links?

In a word: it depends.

OK, two words.

Are there areas of your site where links are out of your control? If you use "ugc" and "sponsored" appropriately, what's left for "nofollow"?

I’m thinking of one former WebmasterWorld site regular who never used "nofollow"; her position as I understand it was that if you don’t personally trust the link, it doesn’t belong on your site.

[Kendo]

I use rel="nofollow" extensively, but only on internal links to login, password, next and contact form pages.

I believe that is what it was designed for. So what's all this other nonsense about?

[Wilburforce]

I believe that is what it was designed for.

Actually, Google's advice is to use robots.txt for that purpose:

"If you need to prevent Google from following a link to a page on your own site, use the robots.txt Disallow rule" ([support.google.com ]).

For pages that really have no reason to be in the SERPs at all, you can also/instead place <meta name="robots" content="noindex"> in the page's Head seaction.

However, for internal links to bookmarks (e.g. mysite.com/widgets.html#top) there isn't really any alternative to nofollow.

Also, although this is a Google thread, there are other search-engines, and not all of them have the same policy on nofollow.