Take a good amount of existing landing pages offline because of low traffic, cannibalism and thin content

[max.favilli]

Why not to redirect? If you don't you will keep seeing them in error in WMT, which is not a good thing. Also returning 410 in theory is an option, but I tried in the past and WMT ignores that.

[Sheena_Schleicher]

What you're really doing is consolidating 'orphan SEO pages' to fewer, higher value pages - which is a specific example Google providesas a "good reason to redirect one URL to another." I would 301 the pages to their most relevant, consolidated landing pages that remain.

Hope this helps!

[ChimplyWebGroup]

I'm not so sure.

Common sense tells me that pages without any Page Authority, or those that may have been penalised (or indeed not indexed) for having spammy, thin content, etc will only pass these **negative **signals on through a 301 redirect?

Also surely if there is as many as 250 potential landing pages all redirecting (maybe even to one single URL), it'd surely raise alarm bells for a crawler?

[Sheena_Schleicher]

It's possible, but it sounds like the ranking fluctuations are likely from multiple URLs competing for the same search queries ("Often really similar landing pages exist - just minor keyword targeting difference and I would call it "thin" content") rather than poor link profiles. He didn't mention any manual penalties either.

I agree that you would not want all 50 URLs redirecting to one or even just a few URLs. Only redirect the ones that are really related to the content of the remaining pages and let the rest drop off. Also make sure you have a killer 404 page that helps users get to the right pages.

[Sheena_Schleicher]

This article is about removing blog posts, but I think it's still relevant: http://www.koozai.com/blog/search-marketing/deleted-900-blog-posts-happened-next/

The 'removals/redirects' & 'lessons learnt' sections are particularly important to consider.

[_Heiko_]

Thanks for your opinions!

There are no manual actions against the pages, so shouldn't care about this! Like I said mostly they are generating no traffic at all (for these ones I cannnot see a good reason to redirect and not just delete them from the index and take them down) and some URL's are just competing against each other and the ranking fluctuations are quite high and therefore I want to put these competing pages together.

I guess I will redirect the pages which still have relevant similiar pages left, but don't redirect pages which basically had no traffic at all in 8 months and no real similiar page is existing.

[max.favilli]

Have you checked with OSE and other tools to see the page juice/authority they may have?

[_Heiko_]

Yes I checked, these pages don't have external backlinks, generating only link juice through internally linking. As I will change the internal linking and the pages I take down will not get any more internal links this should'nt make any difference...

I just want to avoid any redirect, which is not necessary to really make sure that only pages who have a relevant similiar page get a redirect. makes sense, right?

[max.favilli]

Yes make sense. It's also what the people at koozai describe in the link Sheena posted.

Yet, my personal seo-religion so far have dictated me to never remove, every time I asked myself if I should, I got to the conclusion was better not to.

Let me re-check your motivation to do so:

• These pages sent no organic traffic at all in this 8 months

That's horrible, but removing them is going to improve something else? Maybe, or maybe not. You can find out only trying out (testing).

• Often really similiar landing pages exist (just minor keyword targeting difference and I would call it "thin" content)

If you are worried about duplicate content penalization, there's no such thing as a duplicate content penalization, google doesn't penalize duplicate content, google just make a choice, choosing one among different duplicate page to rank. Matt Cutts on that here.

If you have multiple landing pages for similar keyword with thin content, improve the content. You can find authoritative voices advocating multiple landing pages for related keyword interlinking as a perfectly whitehat LSI SEO strategy.

• Moreover I had some Panda Issues in Oct, basically I ranked with multiple landing pages for the same keyword in the top ten and in Oct many of these pages dropped out of the top 50..

I doubt your algo penalization is due to those 0-traffic landing page mentioned above, remove them and see what happen, but I bet won't change it. Instead I would look honestly at all your website and ask myself what spammy, stuffing, nasty dirty little things did I in the past?

[max.favilli]

Question. If the fluctuations were due to the different pages competing with each other, shouldn't you see the different pages exchange places, one goes up, the other far down, then swap places and keep dancing?

[_Heiko_]

1. Well, yes I expect that the other pages will benefit from it, because I basically can overtake the good content parts to the similiar pages. Moreover I can set more internal links to the pages which are actually ranking and generating more traffic. Of course, I could just take off all internal links from the dead pages, but I see no sense in there existence any more.

2. I know that you don't get a penalty for duplicate content. But I think it makes more sense to have one (improved) page for a topic/keyword than having 2 pages and one is basically dead from traffic perspective. From their whole structure the pages are just to simiiliar beside the "content" and even if this cannot force manual actions, it can lead to panda/hummingbird issues you will never recognize.

3. Yeah this action has nothing to do with the dead pages, you are right, I just wanted to mention it, because for me I inptreted it in the way, that google tests similiar pages in there performance and this can lead to longterm decreases. That was for me just another reason for putting similiar websites together and think more in "topical hubs". I talk about really similiar websites like for 3 phrase keywords when just the last word differs and the content is unique but basically tells the user the same like on the other page...

[max.favilli]

It all make sense.

[Everett]

I've done this many times with good results. If the page has no traffic and no external links just remove it, and allow it to 404 so the URLs get removed from the index. If the page has traffic and/or external links, 301 redirect it to the most appropriate page about the topic. In either case remove/update internal links, including those within sitemaps.

Simple as that.

[_Heiko_]

Thanks Everett for your response, changes are in process and I will implement it this week. But it would be even better do remove the not redirected URLs in webmaster tools. right?

Technical question to the redirected URLs: Is there any difference if I leave the redirected webpages on the server or if I delete them?

[Everett]

If you remove a URL and allow it to 404 you can either remove it in GWT as well, or wait for them to update it. I would remove it in GWT as well just to be sure.

There is no difference whether you have the files on the server or not unless the redirect comes down someday for awhile (even for an hour), which could result in all of those pages being reindexed. Other potential issues are if you have the site available on another domain or sub-domain that points to the same folder, in which case your redirects might not work on the other domain, depending on how they were written.

For these reasons, I would go ahead and remove the files from the server just to be safe. You can back them up somewhere local or at some point before the "Public HTML" folder on the server.