homepage Welcome to WebmasterWorld Guest from 54.237.98.229
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Pubcon Platinum Sponsor 2014
Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

This 49 message thread spans 2 pages: < < 49 ( 1 [2]     
Using No-Follow on Internal Links Cause a Drop in Rankings?
Noticed a site dropping in SERPS with no-follow on internal links
braap

5+ Year Member



 
Msg#: 3296478 posted 4:17 pm on Mar 29, 2007 (gmt 0)

I have noticed a competitor's site placing no-follow on internal links to pages that are not on topic/theme, i.e. privacy policy, contact, user agreement, etc...

I assume they are doing this to preserve or funnel PR to their most important on-topic pages.

I have noticed the site has gradually been losing positions in SERPS over the past several months. No, they are not being out SEOD, if that is a word, by their competitors. I also realize that we can't hold and analyze this in a vacuum, so there is no need to speculate about other factors that could cause of a drop in rankings.

I would like to know others thoughts on whether or not placing no-follow on internal links could hurt a site in the eyes of Google.

For example, could placing no-follow on internal links hurt a site's trust rank, thus possibly hurting their rankings?

Also, if you donít feel that it would hurt a site, would placing no-follow on internal links serve much of a purpose at all?

 

activeco

10+ Year Member



 
Msg#: 3296478 posted 8:14 pm on Mar 30, 2007 (gmt 0)

Here we go again.
[webmasterworld.com...]

It seems that, for some reason, major consensus is that meta "index,nofollow" and rel="nofollow" behave differently on individual links, which I don't agree with.
It does the same, with 'rel' having finer granularity as Matt puts it:

The nofollow attribute on links is the most granular because itís on a link level, but something like a sponsor page is a fine opportunity to use the nofollow meta tag instead of marking each link.

soapystar

WebmasterWorld Senior Member 10+ Year Member



 
Msg#: 3296478 posted 8:42 pm on Mar 30, 2007 (gmt 0)

!

Drew_Black

10+ Year Member



 
Msg#: 3296478 posted 9:36 pm on Mar 30, 2007 (gmt 0)

I started blocking lots of pages on my sites via robots.txt to prevent duplicate content problems. (Resorted search results and different user-selectable views of content.) Webmaster Tools reports these blocked URLs as "Web Crawl Errors". (URLs restrcited by robots.txt sub-section.) I took this to mean that Google doesn't like links that appear on the site but are blocked by robots.txt to I started modifying my pages to include rel="nofollow" on those internal links. (My position within the serps has since tanked but I have no idea if it's related in any way.)

So Webmaster Tools calls it an error we can rel=nofollow to prevent those errors subsequently to be frowned upon for manipulating PR?

Is it an error or isn't it? Are errors bad?

Quarfelburg

10+ Year Member



 
Msg#: 3296478 posted 9:43 pm on Mar 30, 2007 (gmt 0)

I'm in a situation similar to mattrix - I have a navigational structure that allows for various methods of viewing my products.

While the categorical pages are different, the individual product pages have the exact same content in 3 sections.

I figured Google would view this as duplicate content so being paranoid, I denied the links in my robots.txt, meta robots tag, and used rel=nofollow.

Mattrix and others -- do you think that although the content is the same (aside from crumbtrails), it would be ok to reveal these pages to Google?

And if not, you really believe that rel=nofollow could be hindering the site in such a situation?

braap

5+ Year Member



 
Msg#: 3296478 posted 10:05 pm on Mar 30, 2007 (gmt 0)

...I started modifying my pages to include rel="nofollow" on those internal links. (My position within the serps has since tanked but I have no idea if it's related in any way.)

Drew_Black, did your rankings tank across the board? If not, did your rankings tank for perhaps either the keyword in the anchor text or for the keywords targeted on the destination page?

arieng

5+ Year Member



 
Msg#: 3296478 posted 10:19 pm on Mar 30, 2007 (gmt 0)

"First, you don't need to do both nofollow and noindex."

Annej, can you elaborate on this? Penalties aside, I use nofollow to keep from passing pr and use noindex to keep from being spidered. Why are those two attributes mutually exclusive?

tictoc

10+ Year Member



 
Msg#: 3296478 posted 10:45 pm on Mar 30, 2007 (gmt 0)

I disagree with Tedster on this in a way because when you look at google, msn, yahoo, nbc, cnn, etc you notice they all use no-follow links to some internal pages which says to me we can do it to help the bots not get confused on internal privacy pages and forms, etc.

activeco

10+ Year Member



 
Msg#: 3296478 posted 11:22 pm on Mar 30, 2007 (gmt 0)

Let me provide some short 101 about meta tags named 'robots', for better understanding and avoiding confusion for some people.

As a part of Robots Exclusion Protocol aimed on 'granulating' permissions to bots on a single page basis (having robots.txt on a domain basis), html meta tags named 'ROBOTS' are used.

<META NAME="ROBOTS" CONTENT="argument1, argument2">

where argument1 is related to the content of the page and argument2 to the links on the page.

argument1 can have one of the following values:
'INDEX' - meaning content can be indexed
'NOINDEX' - the opposite

argument2:
'FOLLOW' - bots can follow (crawl) links on the page
'NOFOLLOW' - don't crawl it

One can make any combination of the arg1, arg2 or use shorthand directives:
'ALL' meaning 'INDEX,FOLLOW' (default, if not defined)
'NONE' = 'NOINDEX,NOFOLLOW'

Obviously, web needed even finer granularity on a link by link basis, which was achieved with 'rel="nofollow"' directive.

I see no reason why would this one differ from previous implementations.

Miamacs

5+ Year Member



 
Msg#: 3296478 posted 12:32 am on Mar 31, 2007 (gmt 0)

It has been admitted that the chosen name for the attribute was a stupid move. If I remember clearly there was a statement from Google as well. Saying "sorry" or something like that, between the lines of course. Or perhaps Yahoo! beat them to it.

The rel="nofollow" does NOT mean that Google or Yahoo or as a matter of fact any other bot would ingore the link. It means that the given link will be followed and crawled, but the source page ( the one which has the link ) won't pass any parameters ( PageRank, TrustRank ) to the target page ( the one it links to ).

Check your linking profile on Google Webmaster Tools or Yahoo! Site Explorer. Or type in just about anything to find out if Wikipedia still is in the serps with 28.000.000 inbounds and 103.000.000.000 rel="nofollow" outbounds.

They follow these links. They know of them. Funny thing is they even make note of the anchor used, only that they say, it won't count. Best think of them as PR0, neutral trust image links with the anchor as the alt attribute. They get crawled. They will be on your record. As for linking to bad neighbourhoods, I don't even think using it would prevent you from being penalized, but never had the chance to actually test it. And finally, they were meant to be used between sites and not pages.

Answering the original question whether adding rel="nofollow" can hurt your rankings...

- Yes.

Not because there's a penalty for it, but because using it within the navigation will have a good chance of neutering the weight and trust network of your site.

Use NOINDEX, NOARCHIVE in the META to keep a PAGE out of the index.
Use NOFOLLOW in the ROBOTS META to make the bots ignore the links.
Add them to your robots.txt
Or better yet, don't have anything to hide.

The argument whether Google would distribute the PR on a page, based on the number of links without "nofollow" is interesting, but I don't think it holds much water. A page with 100 links will pass the same amount of PR with each link, as the same 100 links page with 25 links using the "nofollow" attribute.

...

This all seems splitting hairs to me.
The two noindexes are incidental namesakes.

[edited by: Miamacs at 12:33 am (utc) on Mar. 31, 2007]

Drew_Black

10+ Year Member



 
Msg#: 3296478 posted 1:44 am on Mar 31, 2007 (gmt 0)


Drew_Black, did your rankings tank across the board? If not, did your rankings tank for perhaps either the keyword in the anchor text or for the keywords targeted on the destination page?

I haven't done a decent analysis yet. We only have a dozen or so two-word phrases that we monitor regularly. Four of these have been top 3 in Google for close to 5 years. Now we're floating around 180 - 190 on these four and nowhere for the others. (I'm excluding serps for our domain name or its direct three-keyword derivatives, we still rank well on those.)

I suspect it's an across-the-board thing though since Google referrals have dropped from 6k - 8k visits/day in January to barely over 1k this past week. (The top four were only responsible for 500-600 visits a day combined in January.) The rel=nofollow changes were implemented around Feb 22nd and 23rd looking at the dates on the files. I don't think it would have taken this long to have an impact but how can you tell for sure? Our traffic has been slowly dropping since late December but the past 7 days has seen the most significant drop in Google traffic during that period.

I have reversed the rel=nofollow changes. We'll see what happens.

In the interest of full disclosure I'm also experimenting with navigation menu placement within the HTML. 60% of the site has the navigation block near the top of the page. The other 40% has the exact same block of navigation at the bottom of the HTML positioned via CSS. I have no idea if this is what's killing me.

I do think it's at least notable that Google reports pages it finds yet are excluded in robots.txt as "errors".

annej

WebmasterWorld Senior Member annej us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3296478 posted 2:01 am on Mar 31, 2007 (gmt 0)

I use nofollow to keep from passing pr and use noindex to keep from being spidered

I see what you are saying. I just meant that you don't have to do both to keep a page from being indexed.

Or better yet, don't have anything to hide.

A printable page shouldn't be something you have to hide but with the duplicate content problem it is.

A page with 100 links will pass the same amount of PR with each link, as the same 100 links page with 25 links using the "nofollow" attribute.

Am I write in understanding that this means that nofollow doesn't affect PR leakage at all?

Webnauts

5+ Year Member



 
Msg#: 3296478 posted 2:51 am on Mar 31, 2007 (gmt 0)

When I want to prevent links being followed, I edit this:

<a href="http://www.example.com/sample.html?googlebot=nocrawl" rel="noindex,nofollow,nsfw">Example</a>

and I add in my robots.txt:

User-agent: Googlebot
Disallow: /*?

User-agent: *
Disallow: /*?

If I do not want the link to be visible to Google at all, then I do this:

<!--googleoff: anchor-->
<a href="http://www.example.com/example.html?googlebot=nocrawl" rel="noindex,nofollow,nsfw">Non-sense</a>
<!--googleon: anchor-->

and I add the rules I mentioned above in my robots.txt.

[edited by: Webnauts at 2:52 am (utc) on Mar. 31, 2007]

activeco

10+ Year Member



 
Msg#: 3296478 posted 5:39 am on Mar 31, 2007 (gmt 0)

Hello Webnauts,

You mix a lot of things here, which produces incorrect usage of the code. Please don't additionaly confuse some readers here.

"Rel" is related to individual link only and can have only one attribute, that is "nofollow". "Index, Noindex" doesn't exist here.

"GoogleOn/GoogleOff" tags don't exist outside Google Search Appliances, so they don't work on Internet.

Miamacs

5+ Year Member



 
Msg#: 3296478 posted 11:19 am on Mar 31, 2007 (gmt 0)

Or better yet, don't have anything to hide.

A printable page shouldn't be something you have to hide but with the duplicate content problem it is.

Sorry annej to not have mentioned, it was on my mind actually.
I wasn't referring to your situation, your intention is clear.
But I see so many people claiming to keep their content out of the index with only a rel="nofollow" that I had to keep my comment generic.

In short, rel="nofollow" will not keep your pages out of the index.
If it does help duplicate issues, it does so becuase of this:

- Two or more versions of the same content exist, web and printer-friendly
- These only differ in layout/css so they are seen as duplicates
- If both pages have the same weight, Google will prefer either at random
+ If one has significantly less weight ( all its links are rel="nofollow", web version is level/tier 3 while printer version is tier 4, one has links on many pages, the other is linked only from a single place, etc. ) ...Google may receive a clear signal on which to keep of the two. The rest will be supplemental or drop out.

- If someone links to the "unwanted" page with a single inbound, your fragile system collapses.

And finally an important note on this issue, according to Adam Lasnik...

+ There are no automated penalties for duplicate content
+ Google will drop all but one version of the same content pages

A page with 100 links will pass the same amount of PR with each link, as the same 100 links page with 25 links using the "nofollow" attribute.

Am I write in understanding that this means that nofollow doesn't affect PR leakage at all?

Again, the Wikipedia example. ( or any WordPress / blog site )

If PR leakage would be stopped by a rel="nofollow" where do you think that a site with 26 million inbounds and ( by this logic ) no outbounds would be? PR 11+

This does not happen. The votes go out even with the rel="nofollow" tag, and arrive at the target allright, it's just the target page can't cash them in.

...

In the areas I monitor Wikipedia has been losing ground to pages that slowly adapted to the changes of the last 6-9 months, no big jump, no big dive at all. When I saw that he actually "welcomed" the change, I've asked MC Google several times whether the Wikipedia "nofollow" move would cause a huge shift in their linking profile, to which a large number of penalties ( some even automated ) are connected to. I received not a single answer. Which to me means I was spot on. He was cheering because he knew nothing would change for Wikipedia. Which means the linking profile of the sites are unaffected by their use of rel="nofollow".

...

The devaluing takes place at the target and not at the source, which is only common sense. But if someone can contradict me with an exact example or test results, I'm open to ideas.

activeco

10+ Year Member



 
Msg#: 3296478 posted 1:46 pm on Mar 31, 2007 (gmt 0)

Miamacs wrote:
In short, rel="nofollow" will not keep your pages out of the index.

Not exactly correct.
Again, rel="nofollow" behaves in the exactly the same way as "nofollow" meta tag, which means the linked pages will not be crawled at all.

So, if that is the only link on the web, the linked page will never be crawled.
The problem arises when the page is normaly linked from some other place, where the bot gets the green light for crawling, unless the indexing of the linked page is forbidden in robots.txt or in meta's using "noindex" attribute.

Miamacs

5+ Year Member



 
Msg#: 3296478 posted 2:45 pm on Mar 31, 2007 (gmt 0)

You know, I'm going to test it again.

Perhaps I'm wrong, perhaps something changed since I last did so.
I'll keep you posted on what happened.

Mind you it wouldn't change a thing on what I said about their use ( I'd still advise against relying on using them for anything else than community sites' outbounds ), your statements made me curious. Whatever the outcome would be, these links are on the linking profile of a site, along with their anchors used. They are only devalued when calculating rankings and relevance of the target page and not anytime sooner.

And seeing this, I'd be really surprised if Google didn't crawl them. As a matter of fact the last time I tested, it did.

As I said, we'll see.

[edited by: Miamacs at 3:03 pm (utc) on Mar. 31, 2007]

annej

WebmasterWorld Senior Member annej us a WebmasterWorld Top Contributor of All Time 10+ Year Member



 
Msg#: 3296478 posted 3:08 pm on Mar 31, 2007 (gmt 0)

the linking profile of the sites are unaffected by their use of rel="nofollow".

I can see why Google would do it that way. Then there would be no way to use nofollow either as a meta or in the anchor to improve one's page or site standing.

It still makes sense to use it when you don't want to give the page you are linking to a vote. Not just blog situations would fit. Sometimes you just want to suppress a subpage in the serps so it won't come up above a more important page. Like a how to article might have a sub page of specific instructions. You want the main how to article to be seen first.

OTOH I've been using nofollow in the links to Amazon. I figured they have enough votes anyway. As some of you here have found it suppressed your page results in the serps maybe I should just take the nofollows off of the links.

I have a question for the people who used a lot of nofollows and found it hurt their site in search results. Were You using nofollow extensively? What % of the links in your site were nofollow?

braap

5+ Year Member



 
Msg#: 3296478 posted 3:41 pm on Mar 31, 2007 (gmt 0)

I have a question for the people who used a lot of nofollows and found it hurt their site in search results. Were You using nofollow extensively? What % of the links in your site were nofollow?

The site I referenced in my initial post has 46% of the links on the home page using nofollow.

Drew_Black

10+ Year Member



 
Msg#: 3296478 posted 9:19 pm on Mar 31, 2007 (gmt 0)


I have a question for the people who used a lot of nofollows and found it hurt their site in search results. Were You using nofollow extensively? What % of the links in your site were nofollow?

On my site it was only 1% to 4% depending on the page.

This 49 message thread spans 2 pages: < < 49 ( 1 [2]
Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved