Could this be Panda - or is it something else?

[thebigsicilion]

This is my first post on the board so I couldn't think of a better way to make an introduction then to jump in with both feet first! That being said here is my dilemma.

Feb 3rd at noon, I was literally running through my rankings and writing down the positions when POOF 3 years of hard work vanishes!

I run a company that lives off the traffic I get to the site so this drop in rankings has really punched me and my staff in the stomach.

We are a pr5 site 150k back-links that has taken years to build. We had over 25 number 1 positions and about 65 first page positions or strong keywords. The niche is competitive with the main keyword bringing about 30000 exact search in the USA monthly.

Previous Rankings before the abuse began - Keywords ranked in order of highest traffic to least,

Keyword 1 - Position 3 now position 85
Keyword 2 - Position 1 Now position 12
Keyword 3 - Position 1 Now position 12
Keyword 4 - Position 1 Now position 14
Keyword 5 - Position 1 Now position 13


We made some changes to our blog url structure on Jan 20th 2012 and I am curious if this is where we ran into trouble.

Here is what we did,

The url and title tag structure were changed. We changed it from http://www.mysite.com/keyword/ to http://www.mysite.com/key-word-blog

Then we made each article page in the blog appear like this http://www.mysite.com/key-word-blog/{article title}.php

We then had the title tag become the title of the article in the blog

Finally we redirected the blog to the new blog url. So in the end we redirected over 3000 pages since the blog was large. We had lots of redirect errors and a few soft 404s in WMT come up shortly after. We did our best to fix this when we noticed all prior to the ranking drop.

Could these types of changes bring attention to our site? Could this new attention be the cause of my issues which damaged my rankings?

Our blog is attached to the static site and is not responsible for very much traffic at all. As a manner of fact I dont like much about the blog, the pages are thin and dont offer much to anyone. Mainly regurgitated info.

We also found out that all of the 3000 pages in the blog had the rel=canonical tag pointing to the home page of the site. This seemed like a large mistake on our parts as well. The home page is where all the rankings were and got nailed.

Does this sound like Panda to anyone? What would you do if it was your site?

[edited by: Robert_Charlton at 12:18 am (utc) on Feb 13, 2012]
[edit reason] disabled auto-linking for example urls [/edit]

[Sgt_Kickaxe]

We made some changes to our blog url structure on Jan 20th 2012 and I am curious if this is where we ran into trouble.

This could absolutely be the cause if you mean you changed your internal link structure, changing the url structure in effect shifts the incoming link value of each page in ways you probably can't predict. While I'm sure you kept your most important pages less than two clicks from the index page, hopefully many of them a single click away, the impact of wholesale changes can't be known until the dust settles. It's also possible that just making changes has, at least temporarily, given Google reason to want to re-evaluate the site before trusting it on page one again.

Not all is lost. Because it takes time for all changes to propagate fully it's likely you'll see some pages slowly rise back to their prior positions all on their own after google is done re-evaluating the site and it's new internal link structure.

The only major change I would even consider is a full reversal back to the original. If that's not on the meny proceed slowly right now and make only minor changes with a good 10-14 days or more in between.

The url and title tag structure were changed.

If I understand that correctly you changed the actual url to be different from the old urls. By doing so you have essentially created all NEW pages that Google does not know. A 301 redirect now passes all incoming link value to the new page but you don't get a 100% value through a redirect so you will take a rankings hit if all your pages suddenly did this.

In either case I don't think this is Panda, it's a case of too many changes at once and an actual change of urls (always a last resort).

[Pjman]

I have to back Sgt 100%. It's the change in URL structure for sure.

[Robert Charlton]

We also found out that all of the 3000 pages in the blog had the rel=canonical tag pointing to the home page of the site.

Additionally, this was would have caused major problems by itself, and in this case could well have overridden everything else. I'm trying to think what Googlebot scratching its head would look like. This would have caused a lot of confusion.

[thebigsicilion]

Sgt and Pj,

Let me try to clarify a bit---We have a static site and then a large blog attached to it. We capture all of our traffic on the static index page mainly. We have a few subpages we are developing but 90% of our customers are coming off the index page. The url structure of the index page or subpages never changed. What we did change was the url of the blog. Im sure the link juice does flow through the entire site into the blog as well. I guess google views the blog and the index page as 1 in the same but im not sure.

[aakk9999]

Then we made each article page in the blog appear like this http://www.mysite.com/key-word-blog/{article title}.php

We then had the title tag become the title of the article in the blog

So if I understood well, you made <title> element feed into <h1> of the article and <h1> feed into part of article URL? Someting like this:

Page HTML:
<title>Great widgets overview</title>
...
<h1>Great widgets overview</h1>

Page URL:
http://www.mysite.com/key-word-blog/great-widgets-overview.php

I think these are too aligned with each other.

Also, what happens if you change <title> element? Does it create a new URL and does article title <h1> change automatically? Because if so, then you will also have a future problems with this architecture.

What was the main reason for this change? Was there a problem with the old architecture?

[thebigsicilion]

Robert,

I was under the impression that in some cases google will actually ignore the rel= issue when there is thousands of pages. If rel= worked then my pages would not be indexed and they all were. I wish I could hang my hat on this being the culprit but im not too sure at this point.

[thebigsicilion]

AAKK,

Its not exactly like that from what I can see. Im not to clear on the technical parts of it b/c that's more my partners area. He changed it in an effort to clean up the structure and make things better but now I wish we never made the changes.

[aakk9999]

Apart from URL and <h1>, has the navigation path to blog article changed? Also, have you changed all internal links that point to various articles of the blog to go directly to new URL or do internal links perform 301 redirect?

All these can result in changes in ranking.

[Planet13]

What would you do if it was your site?

Check, double check, and then TRIPLE check for any possible structural errors.

Do you have an XML sitemap? Is it pointing to the correct URLs? Is your robots.txt file correct? Is it possible that one or more of your significant pages has a noindex tag on it?

Also, if the blog is as distasteful as you say it is, why keep it around? Does it generate much revenue for you?

I don't have any direct experience with Panda fever, but it seems like it is a bug you are likely to catch eventually if you have a lot of "regurgitated" material.

[suggy]

In my experience, the blog may well be attracting Panda's attention if it's just written for longtail searches and largely uninspired and uninspiring.

[thebigsicilion]

Our site map is updated and in good order.
The robots.txt is also correct
We use seomoz to scan the site for noindex tags and the like. All is ok there.

If you Google search our title with spaces in it and without the tld we show up at number 5 which makes me think there is a filter in place. We still have some first page rankings for the index page, one of the keywords hasnt even lost position while 99% of them dropped back.

[Planet13]

Also, please do a search for a unique string of text, maybe 10 to 15 words long, that appears on the home page. Where do you rank?

If you Google search our title with spaces in it and without the tld we show up at number 5 which makes me think there is a filter in place.

When you do this, Are the sites that rank ABOVE your site using your content?

Has your home page been scraped by a lot of other sites?

[driller41]

What was your goal when you made those chsnges, it would be useful to understand what you were trying to achieve.

[thebigsicilion]

I copied the first few words from my index page and I appear in the number 1 spot. I see one person who did copy my text but the site has no links and looks like a dud.

Driller-I guess we were just trying to separate the blog url so it didnt look so much like some of my sub pages. Hindsight is 20/20 as they say

[Planet13]

I copied the first few words from my index page and I appear in the number 1 spot.

...for the index page, one of the keywords hasnt even lost position while 99% of them dropped back.

If you Google search our title with spaces in it and without the tld we show up at number 5 which makes me think there is a filter in place.

We are a pr5 site 150k back-links that has taken years to build.

Hopefully some of the moderators will jump in here, but putting these four things together, I think there might be some keyword-level penalty in place against your site.

You said that your site title shows up at number 5. does your site title have one or more of the keywords in it that lost position in the SERPs?

What does the anchor text in your backlinks look like? Is it primarily an exact match for the keywords for which you are trying to rank?

Also, what type of link building did you do? How natural are the 150K backlinks to your site?

Also, how diverse is the anchor text for the INTERNAL linking on your site? Does the INTERNAL linking use all exact match for the keywords?

Again, if you have triple checked that the technical errors are all taken care of (including the canonical errors), then personally, I would look at the inbound links and the internal linking being over optimized. (Maybe someone blasted your site with exact match spammy links?)

But that is one man's opinion, so hopefully others will chime in here.

[thebigsicilion]

Planet,

My url is an exact match for a long tail kw. Its not a major kw though. We had number 1 position for it for many years. Now we are at position 5 so that to me shows some type of penalty.

My anchor text does look a bit intense trending toward my exact match terms that we were aiming for. It is common to see this in my vertical but after looking it over I feel it could possibly be an issue. Having said that, we are starting to break up the profile with domain links, "click here" and "for more info" as well as lots of non kw phrases.

Link building comes from lots of areas. Guest posts, Blog posts some homepage links and the standard stuff everyone does as well. We don't do reciprocal linking and the index page doesn't link out to other sites outside of some of the no follow social sites.

Internal linking is pretty much exact match to the respective title of the sub page in question.

Technical issues have been checked over by several unrelated professionals. I used 3 different parties to tally up opinions and thoughts. The general consensus is that things are clean and we should sit for a week or so to see what happens. No site is ever perfect but there isn't any more glaring mistakes.

How can I check for spammy link blasts from competitors? I ran a Majestic and Seomoz report but I didn't see anything there. Whats the best way to find this info out? I also don't see much unrelated anchor text popping up. It could also be that these applications haven't updated yet either. We do have some unscrupulous people in my space.

[maximillianos]

Quadrupal check those redirects. One small bug can create chaos.

[Planet13]

Link building comes from lots of areas. Guest posts, Blog posts some homepage links and the standard stuff everyone does as well.

I have to ask what "and the standard stuff everyone does as well" means exactly. (I'm not trying to make a moral judgement; just seeing how likely a penalty is.)

Also, when you say, "some homepage links," would it be fair to say that you paid for those?

How can I check for spammy link blasts from competitors? I ran a Majestic and Seomoz report but I didn't see anything there.

Yeah, Majestic and Seomoz would be my first choices for checking spammy backlinks. Although you might search for your domain name in quotes along with, say, the name of a particular male medication that starts with a V in quotes, and see if they exist on the same page / blog somewhere.

So do a search for:

"yourdomainname.com" "v*****"

and see what comes up (pardon the pun).

[thebigsicilion]

Standard stuff to me is blog comments, social bookmarking, posting articles on lower level blogs, twitter and fb sharing.

I have provided some services from my business in exchange for some advertising on similar niche sites. I haven't done this to much though.

[Planet13]

Well, it's hard to say, in my humble opinion.

It certainly doesn't sound like you have done anything egregious.

Maybe it isn't so much a filter as much as the whole internal page rank flow shifted with your blog restructuring?

But if the blog still LINKS the same way to your home page / main pages, then that wouldn't really explain it either.

I hope this helps but I don't know if I have helped you get any closer to the truth or not.

[Robert Charlton]

Technical issues

I'm not seeing answers to any of the specific questions that were asked here about navigation path change, internal links, etc... so it's really hard to comment about the technical issues on this thread. IMO, the site map would have nothing to do with it.

I was under the impression that in some cases google will actually ignore the rel= issue when there is thousands of pages.

I agree more with your original reasoning...

We also found out that all of the 3000 pages in the blog had the rel=canonical tag pointing to the home page of the site. This seemed like a large mistake on our parts as well. The home page is where all the rankings were and got nailed.

It might take Google a while to figure all this out.

[thebigsicilion]

Robert,

I cannot comment on the technical questions to well b/c that is not my forte. The technical stuff is best left to my partner and supporting staff. Im still a bit of a novice.

I do appreciate all of the advice and comments on the board.

[MikeNoLastName]

thebigdsicilion,
Go to G and type in some of the urls of the OLD pages which have been 301 redirected to new pages (i.e. "www.example.com/abc.htm". Try at least 10-20 of them to get a good sampling. If they come up in the index with the NEW title and description and with the OLD URL then you are probably a victim of the phantom 301 redirect bug. It hurt us for almost a year. To confirm, type in the G search field: "site:www.example.com". If the results show totally random pages from your site on the first page, of results... you most likely have the duplication penalty. If you get a fairly good representation of your home page FIRST and other primary pages in the top 3 pages, then you are probably not a victim of this G bug.

[onebuyone]

MikeNoLastName, there must be something about this. I did some mistakes one year ago, which caused duplicate content penalty. Home page was not on first page when I was doing site:mywebsite.com, recently after more than a year this penalty got removed and home page is first result for site: search.

[mirrornl]

Internal linking is pretty much exact match to the respective title of the sub page in question.

[thebigsicilion]

Mike - When I type my site:www.myexmaple.com in "g" the index page is number 1 followed by lots of the important pages. Nothing weird is displayed. So would this mean duplication penalties might not be an issue? Is that the same thing as the 301 bug you mentioned?

[edited by: Robert_Charlton at 6:56 pm (utc) on Feb 21, 2012]