Forum Moderators: Robert Charlton & goodroi
So what happens when you have a page with “ten PageRank points” and ten outgoing links, and five of those links are nofollowed? Let’s leave aside the decay factor to focus on the core part of the question. Originally, the five links without nofollow would have flowed two points of PageRank each (in essence, the nofollowed links didn’t count toward the denominator when dividing PageRank by the outdegree of the page). More than a year ago, Google changed how the PageRank flows so that the five links without nofollow would flow one point of PageRank each.
I also noticed this confirmation:
I wouldn’t recommend closing comments in an attempt to “hoard” your PageRank. In the same way that Google trusts sites less when they link to spammy sites or bad neighborhoods, parts of our system encourage links to good sites.
So everybody - let's all kill the myth and get back to a nicely interconnected web!
a nicely interconnected web!
This goes to the point Matt has made several times about good site structure. I'm sure we have all seen many sites that have an "about us" page with a paragraph on it, and *second* "contact us" page with just an address and phone number. Making two such pages is a waste of pagerank. It's notmally easy to make a single link like "About Us - Contact".
"... for example, we made significant changes to the PageRank algorithm in January [2008]."
There were discussions of this last year, and a new way of looking at nofollow was one obvious possibility.
If I understand correctly, nofollow has just become useless?
Useless for sculpting PageRank, maybe. But PR sculpting or hoarding was never the reason behind nofollow: it was merely a side effect.
we’ve changed other, larger aspects of how we look at links and people didn’t notice that either
Clearly we (webmasters in general) need to focus more on testing. Makes you wonder what he's talking about, doesn't it?
So the times of URLs in blog comments are over?
So what happens to the PageRank that belongs to those nofollowed links? For example you have a page with 50 “points” of PageRank, 50 links, and 25 of them are nofollow. So that page passes 25 points of PageRank. What happens to the other 25? Does it get discarded? Redistributed to the rest of the web?
To which Matt replied:
it’s a bit complicated, esp. since Google doesn’t view pages exactly in the framework as “classic PageRank” any more. You can think of that PageRank going into the reset vector without being too far off.
Which means that the extra PR basically just gets distributed evenly throughout the rest of the web, modulo whatever cool stuff they are doing with the reset vector, which he mentioned earlier. Note that this is essentially equivalent to the answer put forth by Shaddows -- this simply increases the base "vote" value of a single page.
So I stand corrected. Apparently they do indeed have a clever way of dealing with normalization, and they felt this change was worth the trouble of implementing it.
Danny Sullivan suggested the following as a replacement for everything Matt said:
Google itself solely decides how much PageRank will flow to each and every link on a particular page. The number of links doesn’t matter. Google might decide some links don’t deserve credit and give them no PageRank. The use of nofollow doesn’t “conserve” PageRank for other links; it simply prevents those links from getting any PageRank that Google otherwise might have given them.
To which Matt agreed except for the "the number of links doesn't matter" bit. This is very interesting, and confirms that PR is not divided evenly among the links on a page anyway, but is distributed according to G's perceived importance of the link. This has been long suspected, but it's interesting to get another confirmation of it.
That opens the door to another form of PR sculpting, by changing the placement and presentation of your preferred links to try to get more PR passed to them.
That opens the door to another form of PR sculpting, by changing the placement and presentation of your preferred links to try to get more PR passed to them.
So we get back to the theory that links within an article might have more juice than links found at the bottom of the page? I've never bought into the belief that links higher up in the HTML carry more weight with Google, because that just seems wrong to me. People read pages from top to bottom, so when they've finished the article, they're expected to scroll back up the page to go elsewhere?
That doesn't make sense to me. I've always placed links that go to main site navigation, and to other related pages at the bottom, because I think that's the most convenient place for them for my site visitors. I don't like having to scroll back up to find navigation to other pages on a site, nor do I like having only an ad to click on at the bottom of the page.
By presentation, are we suggesting bolded links might be deemed more important? Or does it need to be an H4 link to warrant more importance? It sure seems like Google is making webmasters configure their pages to suit Google and not their visitors, which supposedly goes against the Google guidelines.
Am I completely wrong about this?
Am I completely wrong about this?
Google thinks it is pretty good at analysing content. It sees this as its job, analysing and, er, organizing the world's information
What you are supposed to do is write for visitors, and TRUST google to index your page correctly. You are in fact actively DISCOURAGED from helping G with this task. Any attempt to help is seen as manipulative, and may or may not be punished. Any success may or may not be transient.
You may not think this is fair, but from G's POV, they are simply trying to QC their web property, which is under attack from spammers on the one hand, and over zealous SEOs trying to over-rank their content-poor site. Sites should rise by merit of content, not skill of SEO. If G was perfect at their job of analysis, and webmasters were perfect at optimising for visitors, content really would be king.
Its just that G is not perfect. As the current SERPs thread atests.
larger aspects of how we look at links and people didn’t notice that either
[edited by: Shaddows at 5:03 pm (utc) on June 16, 2009]
By presentation, are we suggesting bolded links might be deemed more important? Or does it need to be an H4 link to warrant more importance? It sure seems like Google is making webmasters configure their pages to suit Google and not their visitors, which supposedly goes against the Google guidelines.
I have no evidence one way or the other, but that's the sort of thing Danny seems to suggest.
On the contrary, since we don't know what parameters G uses to determine the importance of a link, your best bet is probably to place important links wherever you think they would be most useful to visitors. This is just incentive to think more about which links are most important to you.
In Google, nofollow links don’t pass PageRank and don’t pass anchortext
Don't pass anchortext, that's also interesting...or is this old news?
[edited by: Pico_Train at 5:39 pm (utc) on June 16, 2009]
Don't pass anchortext, that's also interesting...or is this old news?
A nofollow link has been traditionally explained as "not even used for url discovery" and "completely removed from the webgraph". The fact the no anchor text influence gets passed seems implied by those statements.
Many have noticed that links in various segments of a page are treated differently - navigation area and footer links carrying less punch than content area links. Seems to me this is part of the "PR variation" picture. I would also conjecture that on standard blog pages, comment area links get a lot less punch, even if they are dofollow.
So can we use java links to sculpt PR now?
If you mean javascript (java is a different technology) I'd say probably not. Google has made a lot of changes in how they handle javascript links, and how they examine javascript altogether.
As with anything web related Google's system serves to equalize all websites. When one performs too well it receives the NASCAR equivalent of a restrictor plate. When one is so full of problems that it stands no chance on it's own it is given a chance by Google. All we can do is put the odds in our favor by doing everything that is allowed and recommended.
Nofollow sculpting still works, based on my limited testing over the past 2 weeks on an older site, but I think it has just been tossed into the black hat arena and so it's no longer safe to test with. I find that a waste of a good tool, Google does NOT get page importance right all the time.
[edited by: JS_Harris at 9:04 pm (utc) on June 16, 2009]
First, I wonder how they determine the difference between a paid ad and a valid partner exchange of related content in the form of links.
Second, and most important, does this mean that paid ads using the recommended "nofollow" (if you have many on your page) will considered a negative factor?
Finally, if you have an information-intensive site which links out to 20,000 other sites as part of its intrinsic nature as a service--such as Wikipedia (pre nofollow)--will you then not be penalized for your outbound links which use "do follow?" (assuming you have some quality inbound links as well, of course.)
I have seen sites which use "nofollow" on all outbound links and rank very high--if not #1.
This raises a lot of questions.
yes it is old news, but nice to see it reiterated. Nothing just changed. Nofollow did not just become "useless".
nofollow's most clearly useful function for most webmasters in the world (since their websites are designed non-optimally with duplicate navigation) is to nofollow a "Home" link and leave a "Red Widgets" link in the clear. Thus you get anchor text benefit for the useful terms and not the useless term of "home".
----
(In terms of the above though, given Matt's other comments a link positioned at the top of the page may send more PR than one at the bottom of the page, so nofollowing a "home" link at the top will sacrifice some PR to get the anchor text benfit from a bottom link saying "red widgets". The best solution is to use accurate, descriptive anchor text where both bots and humans will find it useful.)
This is an important message in the present case, as well as for many other things people see discussed on SEO boards. Do not overreact - do not make drastic changes based on nothing more than what you read.
nofollow a "Home" link and leave a "Red Widgets" link in the clear
That makes sense assuming that, if both links were normal links, they would still only count as one link for purposes of PR distribution. Do we know that for sure?
My understanding was that, in the scenario you describe, the link to the homepage would count as a single link. I also would have thought that G would not give a lot of weight to the useless term "home" in any case, and would pass anchor text relevance for "red widgets" even if all links involved are normal links. You seem to be indicating otherwise. So maybe the whole issue should be rethought in this context.
What if, by nofollowing the home link, you are now throwing away half of the PR you would otherwise have passed back to the homepage? Or, alternately, what if G would have handled this correctly (by passing relevance for "red widgets") in any case, so the use of nofollow here is just a waste of time?
-------------------------------------------
(Dang it, steveb, you edited your post while I was typing and now you address this issue. Oh well, I'll leave my post as it stands in case anyone wants to comment further. Overall, I'm thinking the best bet is just not to use nofollow on internal links, and go back to doing whatever you would have done before nofollow was invented. It would still be quite useful for avoiding "bad neighborhood" penalties on outbound links, though.)
