Can you help with this mystery? Old site's links attributed to new site.

[Dr.Marie]

The site, http://bit.ly/WB3S68 was affected by Penguin. The site owner wanted to start over and did the following to create a new site (http://bit.ly/11xqetB):

-Used the Google URL removal tool to remove the old site from the index.
-Used Robots.txt on the old site to disallow search engines
-When the old site was gone from the index, placed the new site online. The new site has a different home page than the old but the inner pages are all the same.

A couple of weeks later, according to WMT the new site suddenly had 10,000+ links that previously had pointed at the old site's home page. There is a line underneath each link saying "Via this intermediate link: [old site]"

There was definitely never any redirect from the old to the new.
I am figuring the problem is related to what is described here, http://dejanseo.com.au/mind-blowing-hack/ where Dejan SEO noticed that if you copied a page completely Google, that page's links would appear in your WMT.

It turns out that when you remove a site from the index using Google's URL removal tool, if you choose to remove the whole site it removes it from the index and not the cache. When I searched for "webcache.googleusercontent.com/[old site]" it would display the cache for the new site! So, I went in and removed each individual url from the cache using the tool. (you only get the option to remove from cache if you remove individual URLs).

Now, 2 days later, the cache search for the old site throws a 404. Yet, all 10,000+ of the links to the old are still pointing at the new according to WMT.

My guess is that as these links get crawled they will disappear, but I would have expected at least some of them to be gone by now. I know that WMT is slow to update and I'm hoping that this is why we're still seeing them.

Questions:

1.What do you guys think of these "via another site" links appearing in WMT? Do you think Google is passing link juice through them? In the Dejan SEO article, those via links appeared even if a low pagerank site copied a high pagerank site's article, so I am guessing that the links aren't really passing juice and therefore don't really count as pointing to your site.

2. I know no one can answer this for sure, but how likely do you think it is that this site would be affected by Penguin if it refreshes soon?

[Dr.Marie]

Thought I would bump this as some of the regulars may not have been on on the weekend. Can anyone help?

[KernelPanic]

good question, I'll bump also. My guess is that it will be sorted on the next refresh, but that's just a guess.

How sure are you that it's a penguin issue?

[Dr.Marie]

Thanks KP. I'm quite certain this was a Penguin issue. I didn't have any analytics or server logs to dig through, but there was a distinct drop in traffic at the end of April. The site lost positions for the keywords that they had been building links for. There were thousands of links from blog comments, articles and paid links all with keyword anchor text. I also didn't find any obvious Panda issues on the site.

[Kevin H]

This is a good one.

Assuming there is nothing in WMT to 'set up' the re-direct etc and the 'old' page has not been set to feed one page to Google and another to people then I have only one 'guess' at the moment.

That G has got clever enough to recognise that the text on the 'old' page has no content other than the "We have moved. Please visit us at <new URL>"

This sort of fits with the 'new' idea that Google awards weight to attributions as well as links from other websites.

[KernelPanic]

This is a good one.

Assuming there is nothing in WMT to 'set up' the re-direct etc and the 'old' page has not been set to feed one page to Google and another to people then I have only one 'guess' at the moment.

That G has got clever enough to recognise that the text on the 'old' page has no content other than the "We have moved. Please visit us at <new URL>"

This sort of fits with the 'new' idea that Google awards weight to attributions as well as links from other websites.

haha consider the ramifications of a penalty flowing through a citation! What's google doing??

[Chedders]

The new site has a different home page than the old but the inner pages are all the same.

IMHO
The inner pages are the problem here.
The issue is although the pages are de-index, and out of the cache this is only 2 layers of the google beast. Think of it as just the customer facing layers.

Google has become quite clever over the years on recognizing duplicate content even if spun. It can only do this if its data core has a bit of history in there as well. If that article is correct and I have no doubt it is then I am pretty sure its matching the old inner pages up with the new but same inner pages and treating them as the same.

The old pages where subject to a penalty so this new but duplicate content should be treated the same as google has already identified them as being bad in their eyes.

The only solution in my view is to re-write a couple of the pages and see if the links drop in GWT, my guess is they will once they get indexed.

Pain in the arse I know and to be honest a bit unfair in this case but unless you can get someone at google to manually fix things I would say the inner pages need re-writing as well.

[DarrenHaye]

The old pages where subject to a penalty so this new but duplicate content should be treated the same as google has already identified them as being bad in their eyes.

If the problem was penguin then the old pages were subject to penalty because of the links pointing at them, not the content.

So if they move to get rid of the links, then google shouldn't punish them for that.. because they're trying to do the right think by dumping all the bad manipulative links.

But if there's panda issues then yes you would be right, moving poop around different domains would still be poop, and that would make them dung beetle esk I guess lol.

[Dr.Marie]

Thanks guys. Good thoughts here. Sorry I couldn't weigh in earlier. A stupid stomach bug has thrown our family for a loop the last couple of days.

This is not a Panda issue for sure. And as Darren mentioned, if it was Penguin then it's the links that are the culprit and not the content so the site itself doesn't deserve to be punished.

At this point we'll give the site a bit more time to see if the "via this intermediate link" messages go away once the links get crawled again. WMT is notoriously slow at updating so I am hoping that one day we'll just wake up and see that those links are gone.

The next option is to disavow those links. But that would be a pain. Plus the links will always be there clouding up WMT even if there is a disavow.txt present.