No index alone is not working - so what else is required - Google Search and SEO forum at WebmasterWorld

Forum Moderators: Robert Charlton & goodroi

Message Too Old, No Replies

No index alone is not working - so what else is required

Whitey

12:12 pm on Oct 6, 2011 (gmt 0)

Tedster - I've been dubious about the "noindex" advice that was being pushed out, even by some Googlers. Somehow it just doesn't make solid sense to me - after all, the pages are still "there", just not being used as search landing pages.

How can this be a good user experience in Google's Panda eyes. Linking out to no indexed pages or indeed having them on your site has got to signal, these pages are still no good and they are part of the bad user experience.

Surely Google is taking these into account? Any views on how to better approach the de indexing element to assist a Panda release. Get rid of them completely?

Whitey

12:15 am on Oct 8, 2011 (gmt 0)

I'm going to bow out with me tail between my legs and stay away from these discussions in the future

No way ! Things happen. Your community needs your most valued inputs ( that at least caused a qualified reaction) .... more needed :)

mslina2002

1:14 am on Oct 8, 2011 (gmt 0)

How should one link to those no index pages ? Do you still send PR through to them or adjust the link code to preserve the juice or remove them completely or apply some other option?

Whether you like it or not you will be linking to those noindex pages via your navbar, internal pages, etc. However, I must say I have not totally noindexed all those pages. I have them grouped by widget category and each category has a "main page". Those mainpages are all index,follow. Any paginated pages are noindex,follow. So I am actively linking to those mainpages that are index, follow.

Subsequently these mainparent pages also have child category pages. Also any child category of those parent pages are noindex,followed because when Panda hit, I had to do something quick to get back the traffic, and all the child pages were generated with a query. Eventually I will look at these child widgets to see their value and popularity with visitors and then index, follow those that are impt.

When Panda hit I also looked at minimal use of nofollow linkage, stopped link sculpting, focused on tightly theming content on my pages, tightly theming my sidebar, tightly theming the site and reducing the number of ads/affiliate products links. Don't know if that or any of that made a difference. And yes, on some pages I even have freely dofollow links to sources I thought were valuable. Funny enough, those pages are ranking well. G seems to like that.

lucy24

1:14 am on Oct 8, 2011 (gmt 0)

Yes, I really did mean <head>, not HEAD, since the original issue was about <meta> tags. But computers, unlike humans, can purge their memories. Tell a robotic jury to disregard the last two hours of testimony, and it will be as if they never heard it.

"noindex" to me means "This page has no meaning in isolation, and will make sense only if you come in through the front door" (which is indexed). Or "This person's life story is frankly none of your business unless you already know them, in which case a full-name search will lead you to a page containing links that you-- the human visitor-- will recognize as pertaining to the person you're interested in."

noindex pages are always introduced by nofollow links. Belt and suspenders. Or belt and braces, depending on which side of the Atlantic you are on.

Robert Charlton

2:30 am on Oct 8, 2011 (gmt 0)

noindex pages are always introduced by nofollow links.

No! I think you're conflating several different things here. We're not talking about the rel="nofollow" link attribute, which blocks the front door but allows side doors into a page. We're talking about the robots meta tag.

If you want to keep a page out of the index but allow it to acquire and pass on PageRank, you can use the robots meta tag in this format...

<meta name="robots" content="noindex,follow">

Even though follow is default behavior, I'm including it in the above just to remove ambiguity.

Note in the above situation that Googlebot has to index the rest of the page (beyond the <head></head> section)if the PageRank is to circulate through the links in the body of the page. If it did not, Google wouldn't know about the links.

With regard to "nofollow"... you can also use the robots meta tag in this format, with a "nofollow" attribute...

<meta name="robots" content="noindex,nofollow">

When the robots meta is "nofollowed", it will stop PageRank from circulating through the page to the rest of the site.

That is not the same as rel="nofollow" link attribute, which is what you're thinking about, which stops PR from going through a link to a page but may allow other access to the page via other unblocked links.

lucy24

5:20 am on Oct 8, 2011 (gmt 0)

It depends on what leads to what, doesn't it? I'm dealing with dead ends: pages A and B are indexed, pages C and D aren't, so let's just close all possible access points to C. It's a whole different scenario if you're no-indexing C but leaving D wide open. And also more likely to be ignored by g### on the grounds that you can't possibly have meant them to ignore this vital transfer point. Which I guess is where this thread started.

Zivush

6:50 am on Oct 8, 2011 (gmt 0)

I still have a problem with the concept of retaining no index pages that are good for users, but bad for Google where large amounts of data are involved.

Available options (and sorry for my English):
1. Complete removal 404
2. Meta No-index, Follow
3. Complete removal coping to sub-domain and 301 redirection
4. Complete removal coping to different domain and 301 redirection
5. Complete removal coping to different domain with no 301 redirection but linking out.

Panda eyes to these above 1-4 options:
1. Could help (need some other activities on the remaining quality pages such as, social networks, site's look n feel etc)
2. Probably would not work by itself - Pages are still part of the site. Still Panda affected.
3. Still correlating the "bad" pages to the site. Might be wrong?
4. Same as 3.
5. Though, in a way, same as 3 and 4 because still linking to "bad" pages, it is not as 301.

Hoping I made these above options clear.

This 36 message thread spans 2 pages: 36