Forum Moderators: Robert Charlton & goodroi
Could you say that the introduction of the nofollow attribute has allowed Google to implement more strict borders for over optimisation? Now we can nofollow "home" links without removing them as a user feature and Google has left us no excuse to have over optimised sites!
Nofollow is a blessing and a curse!
[edited by: tedster at 9:27 am (utc) on Dec. 10, 2008]
[edit reason] moved from another location [/edit]
A question for someone extremely comfortable on the subject: Have you ever considered using nofollow on EVERY link to your BEST page? If it has enough backlinks from other sites to push it's value to #1 in the serps you're just wasting your internal pagerank by linking to it anyway. It doesn't get better than #1, I hope that illustrates the potential power of micro-managing internal pagerank for you.
This is exactly what is done on a site I am involved with (a site of several million pages). We have sub-sites for several key locations in our country that have great PR on their own without getting link juice from the main super-site. In these instances instead of "over egging the mixture" we pass PR back UP through the directory structure so it can get used more sucessfully elsewhere. Lesser performing sub-sites with lower Page Rank get passed the link juice instead, this way we even have individual listing pages (which are far down the directory structure) with significant Page Rank.
I'm sure use of the nofollow attribute was Google's way of finding most, if not all of the SEO wantabes. If you are using it for anything other than what it was originally intended for, I say there are some associated risks.
I share the same opinion, even I cannot say I'm 100% SURE. I also think this doesn't apply only on wannabes, but SEO people in general.
One thing I am 100% sure is that using robots.txt/meta noindex is much safer way to go.
I'm sure use of the nofollow attribute was Google's way of finding most, if not all of the SEO wantabes. If you are using it for anything other than what it was originally intended for, I say there are some associated risks.
A conspiracy theory which i dont buy into at all. Wasnt it created as a way for webmasters to tell Google that they didnt endorse a particular link? If you read the matt cutts interviews you can see direct endorsements, by Google, for the use of nofollow to tell them which pages should have the most PR.
>> Find the SEO wannabes
What would they achieve by that? Google - "We know all these sites use nofollow, therefore the webmaster has hired or wants to be an seo"... what would be their next course of action after finding that out?
Using robots.txt and meta noindex wont stop links passing PR it will only stop pages being indexed. Using a meta nofollow is EXACTLY the same as adding nofollow to ALL the links on the page. So PR sculpting isnt possible without using nofollow at link level.
Well, actually im kinda wrong... you can use other methods to sculpt pr. But they take a lot more work to implement.
Using robots.txt and meta noindex wont stop links passing PR it will only stop pages being indexed. Using a meta nofollow is EXACTLY the same as adding nofollow to ALL the links on the page. So PR sculpting isnt possible without using nofollow at link level.
Why would I want to disrupt the internal flow of PR within my site at that granular level when I can just stop the page from obtaining it? Or, just remove the page completely with a noindex, nofollow directive?
Ya, I know, I just don't understand this. I'll pick it up one day. I do know that I will not recommend the use of rel="nofollow" unless we are dealing with unmoderated UGC, I see no other reason to use it.
If your originating site architecture is built using a "natural flow" I see no need to disrupt that at all. I can easily tell the SEs to just ignore the page (noindex) and follow the links so that it does continue to pass PR in the "natural flow" of things. Or, I can tell the SEs to both ignore the page and not pass any PR using noindex, nofollow. It's a simple process and it keeps me out of trouble. I don't have to sit there and build these complex graphs that show me how the PR is flowing through the site. What if I told you I rely on "gut instinct" to perform some of that? ;)
It's an interesting concept and I see quite a bit of discussion surrounding the topic. Heck, it even has its own nomenclature. I seriously don't think this is something that should be experimented with unless you "absolutely know what you are doing". I wouldn't do it and I think there are more failed implementations than there are successful ones. Whether those who did the implementation know it or not. :)
A conspiracy theory which i dont buy into at all.
Heh! My Tin Hat (I've moved up from Tin Foil), has quite a few dings and dents in it. I sometimes don't buy into my own theories but after I go through the process again, I convince myself. ;)
Well, actually im kinda wrong... you can use other methods to sculpt pr. But they take a lot more work to implement.
If you are working on an existing site where there isn't a budget for a redesign/re-architecture then the nofollow attribute has to be one of the best methods to rebalance/sculpt/whatever a sites PR?
Why would I want to disrupt the internal flow of PR within my site at that granular level when I can just stop the page from obtaining it? Or, just remove the page completely with a noindex, nofollow directive?
Because this is not just about the target page and the PR rank IT receives. This is as much about the source page of the nofollow and what PR it WANTS to pass on and WHERE. The difference is subtle but important.
I wouldn't do it and I think there are more failed implementations than there are successful ones. Whether those who did the implementation know it or not. :)
The number of corporate sites me and my colleagues see ever day with bad/incomplete/ineffective link sculpting is quite shocking. The problem is that even if you do think you know what you are doing (I am talking about myself here: I think I know what I am doing LOL!) each site is different in its application! It is not something you ever get complacent with: PR flow graphs are imperative for anything but the basics.
[edited by: tootricky at 5:28 pm (utc) on Dec. 11, 2008]
Why would I want to disrupt the internal flow of PR within my site at that granular level when I can just stop the page from obtaining it? Or, just remove the page completely with a noindex, nofollow directive?
I thought we'd decided that noindex took the page out of the index but didnt take it out of the algorithm/PR calculation process, so pages which are "noindexed" still have PR (like matt c says)... In which case... whats the point in a page that isnt going to appear in the SERPs having PR? The only time when that would be appropriate is if that page was instrumental in passing PR on to other pages. So if I had a login page that everyone linked to (an example matt c uses in the interview) that page should be noindex, followed. But what confuses me is... why have it noindexed? What positives does that bring?
If I rel="nofollow" every link on my site to, say, about-us.asp, that tells Google 1) not to follow the links to about-us.asp for discovery AND 2) not to pass PR from other pages on my site to about-us. However, if any external sites are linking to my about-us.asp page and those external links to about-us.asp are followed then Google can still discover my about-us.asp page, index about-us.asp, pass PR from the external links to my about-us.asp page, AND pass PR to all outbound links from about-us.asp to other pages on my site or elsewhere. I really see no problem with using this for PR sculpting as long as it is not overused and you are consistant in it's use. So if I nofollow links to about-us.asp on one page then I would do it on every link to that same page sitewide.
Adding a <meta name="robots" content="nofollow"> on my about-us.asp page is totally different. It tells Google 1) not to follow any of the outbound links from about-us.asp and 2) not to pass PR from about-us.asp to any page pointed to by about-us.asp's outbound links. In this case, however, the about-us.asp page still accumulates PR, it's just never passed out from about-us.asp. This is a PR sink or deadend. The only time I could think of when I might use a meta nofollow tag would be possibly on a directory page where I might have hundreds of outbound links to other companies.
Similarly, noindex on a link and meta noindex are totally different beasts. The meta noindex should guarantee the page never makes it in the index. I'm not sure why you would every use rel="noindex" on a link as it only says, "Don't index that page I'm linking to because you followed this link, however you CAN index it if you follow a link to that page from somewhere else which is rel="index".
There is a time to use nofollow at the link level, and a time to use meta nofollow on the page. They are for very different purposes IMO, and as with everything, as long as they are not abused and used consistantly I don't see any issues with doing either.
[edited by: ZydoSEO at 7:04 pm (utc) on Dec. 11, 2008]
Wasnt it created as a way for webmasters to tell Google that they didnt endorse a particular link?
So many people I talk to think that sticking a meta nofollow tag on a particular page is the same as adding nofollow to all inbound links to that page, and these are two very different things mostly because you cannot control the external links to that page.
Do you mean sticking a meta noindex tag on a particular page?
If you've 'followed' the nofollow saga since it's origin, it should be quite clear that Google is happy to change the use and meaning of the nofollow tag as it desires.
It wasnt really google that changed anything. Nofollow works exactly the same as it used to. Its SEOs that have changed its implication, which Google responded to by simply saying "do it, if you want. at least we wont have to index all your poor quality pages"?
use it to detect SEO or other things right now, that therefore they won't in the future.
But i dont see how "detecting seo" can be something that Google would want to do? It would harm their index if they somehow started penalising sites which used nofollow.
>>>>>>>>
No. I mean the results of <meta name="robots" content="nofollow"> existing in the <head></head> of page #*$! are very different from the results of using <a href="SomeCoolURL" rel="nofollow"> on all of the inbound links on your site which link to page #*$!.
[edited by: ZydoSEO at 10:36 pm (utc) on Dec. 11, 2008]
That is what the article says. Matt has made the point that you can make some pages not rank -- like duplicates, or sign in pages -- but nowhere has any Google employee stated this sculpting leads to doing a new mathematical calculation to increase pagerank to other links.
On the contrary, in this article Matt states that nofollow is the granualar equivalent of noindex and nofollow page-level meta tags.
A noindex meta tag leads to a page not being indexed, but it gets PR and passes it to other pages.
A nofollow tag added to a noindex page means that URL gets PR but does not pass it, and thus the PR dies.
This is the clearest statement from Matt that granular nofollow on links leads to PR dying, not being reassigned.
This is consistent with statements by Matt and others that nofollow links are "dropped" from the link graph.
What some are suggesting here runs smack in the face of Matt's statements, and also common sense regarding Google's use of its processing power... Googlebot hits a page, knows its PR, counts the hrefs, divides the PR, discards the nofollows. As some have posted, Google does not merely ignore nofollows totally, since they do appear in webmaster tools, so if they would do the redividing thing they would have to do another calculating step, and in general Google does things the easy way and doesn't needlessly cost itself effort.
It seems that webmaster FUD on nofollow is possibly the greatest amount of FUD created on any topic before. Personally I don't know for sure how they handle things this minute, and haven't tested it for quite some time, but just based on the statements Google employees have made, the central thesis presented in this thread is flat out wrong (probably).
The best evidence is that nofollow is the link equivalent of the noindex+nofollow meta tags, which means PR dies when you use it.
Because this is not just about the target page and the PR rank IT receives. This is as much about the source page of the nofollow and what PR it WANTS to pass on and WHERE. The difference is subtle but important.
i think this is key.
if what MC says, that Google simply drop the link out of the graph is true, then that only leaves the other links on the page, and their increased share of wellie down their line/s.
it's valid with simply reducing links off page, and seems the same with nofollow IMO
Yes, as I said they could do that. But it would be one extra step, which all other things being equal nobody does one extra step if they don't have to. Also, doing the extra step would sem to be strictly for the benefit of webmasters attemptng to scuplt PR, which strikes me as the mosquito on the tail wagging the elephant.
This means non-duplicates, not indexing useless login pages, etc.
It's fantasy to suddenly make the leap that he means anything like boosting the blue widgets article at the expense of the red widgets article.
Read what he says, not what you wish he said.
Googlebot hits a page, knows its PR, counts the hrefs, divides the PR, discards the nofollows
The opposite makes no sense. Adding more links to a page obviously should impact the other links on a page. None of this, and the random walk idea, makes any sense otherwise.
I don't think you understand how some of us use this tool.
For example, if I didn't want the page indexed, I'd use something else. What I want to do when I pagerank sculpt with rel="nofollow" is to focus the spidering from specific pages.
For example, one of my biggest projects is a custom community software that we made to replace phpbb. It's pure SEO candy and I use rel="nofollow" tags to minimize the links on some of the grids and make sure the right pages get the most weight.
So on a forum grid, there's the topic title link, the username link, the last poster link, and because we made this based on tags there's a lot of tags as well.
We use rel="nofollow" to bring the total links on that page from several hundred down to 100, and make sure the pages that we care about get the most of our own link love.
Now all the other links we nofollowed are still indexed, and I don't mind at all if you link to them. The point was to manage my site's link love, and if you want to toss your site's link love my way for those pages it's all good.
Within the Googleplex nofollow is viewed as a legit way to quickly identify the content that webmasters are attempting to optimize for.
Surely PR is calculated using the link graph that Google builds for your site/page. If nofollow drops the link from the link graph (which is what matt says -- word for word) then surely that PR flow and allocation is calculated in exactly the same way as the nofollow links not being their at all?
Until Google says differently
How can we get matt c to answer that question?!
Within the Googleplex nofollow is viewed as a legit way to quickly identify the content that webmasters are attempting to optimize for.
Silly Universal DNA Helix has us repeating the same EFV discussion from 2 years ago.
Or are we at the higher octave yet?
Don't NEED any Official proclaimations, MC pronouncements, or Goog employee #1-#3 confirmations about how Google handles rel=nofollow NOW.
Testing rel=nofollow NOW gives you the answers.
(Assuming one knows what to test)
and the answers come from the ONLY thing that matters...
What and how the ALGO views it.
How and what Google does with nofollow TOMORROW is as silly as worrying about how and what Google does with EXTERNAL LINKS or TITLE TAGS in a future time.
Are you going to STOP getting external links NOW, out of fear that Goog MIGHT change the algo a year from now?!
FUD-thinking at it's best.
I'll make everyone a 100% prediction.
Google WILL change how they value something that WORKS now, at some point in the future.
And it will either be:
* more effective
* less effective
* non effective
* cause a penalty
Worrying about WHAT and WHEN that will be, is an utter waste of time.
Thus, a 'follow' link to a "noindex, notfollow" page is precisely equivalent to a nofollow link (at the link level). Sorry Steve, I realise you have said this, but I'm just working over it myself.
I would imagine that that was how nofollow originally worked, when all it was meant to do was disavow a link from UGC (basically making it profitless to spam links). It's certainly worth some testing to find out if it still works that way (my suspicion is there will be context involved, and G does not handle it the same every time).
I've never done PR sculpting, preferring a logical and efficient page heirarchy. I know I keep banging on about this but ... :-)
Also, G does some fairly complex calculations already, I think an (x minus y) calc to discard nofollows before disributing PR would use negledgible resources
Right, all you who say it isnt in Gs interest to penalise SEO. You're 100% wrong. G is for users. G thinks IT is the best adjudicator of what content is best for its customers. It sees SEO as an attempt to 'game' the system. Most SEO is therefore BAD*.
G wants SEO to be about usability, accessability and content. Gaining links is also fine (think 'marketing') as long as you are not doing excessive reciprocal linking (again back to the manipulation point). Thats why penalties exist- to stop SEOs from outranking 'valuable' non-SEO sites.
Some people think they are somehow 'helping' G identify the right content. G does not see it that way. G sees it as trying to distort the REAL value (as defined and calulated by G).
*For clarity, any of the above (such as the word 'bad') that can be interpretted as a value-judgement should not be. Bad in this case is shorthand for 'disapproved of by Google'
(all written an hour ago but not posted for some reason)
[edited by: Shaddows at 10:39 am (utc) on Dec. 12, 2008]
Someone mentioned earlier about there being a link on every page to the shopping cart (View Cart/Check Out), and that's fairly standard procedure. But with some shopping carts, every time a bot hits that shopping cart link, it generates a URL with an added parameter with an error message - same page, different URL, same error message. And it happens a LOT. And yes - those get indexed.
Given that there's an economy (often an economy of scarcity) of the number of pages that'll be indexed for any given site, keeping all those dups with different, useless URLs ignored is reason enough to use rel="nofollow" in that particular instance.
Rogue bots and scrapers may not honor the restriction, but they aren't the ones who (individually) hammer the site with repeated crawls.
...SEO...
how do we keep getting back to whether using rel=nofollow is an ACT OF SEO (like an act of terrorism?)
It's a red herring argument at best.
Two very small examples using very LARGE segments of Goog's view of rel=nofollow.
Wordpress.com uses rel=nofollow for internal links.
Let me know when they fall of the Goog party of ultimate rankings.
BLOGGER.COM uses rel=nofollow.
They are OWNED by Google, yes?
Don't let the FUD campaigners scare you.
All of a sudden no one believes MC's "official" stance?!
Lol, I don't know whether to shout for joy or
yell at the irrational swing to the "rel=nofollow conspiracy" that is supposedly coming....
No, I'm not sold on nofollow as an 'act of SEO', nor do I know (having not tested) if PR gets dropped or redistributed. I can see the LOGIC either way, so it worth a test. I don't use it, apart from boilerplate pages, anyway. But, it is a tool worth understanding so I can use (or chose not to use) it on the future.
The second point (mini-rant on SEO) is unconnected with my views on nofollow. I think a lot of confused thinking goes on about how SEOs see themselves, or percieve google to see them.
SEOs and Google are not on the same side, but nor are they opponants. G is trying to return relevant results, SEOs are trying to make it appear that their pages are more relavant than they actually are (or at best, to fully showcase their relevance). These are not compatible ends.
edit- and anyone who takes corporate announcements at face value from ANYONE (including MC of Google) is asking to be deceived. Do you tell the warts-and-all truth about your own company? How about announcing your trade secrets? No? Thought not.
[edited by: Shaddows at 10:54 am (utc) on Dec. 12, 2008]
