The primary search keywords for our news release network have dropped like a rock in Google... we are not sure why.

[jarrett.mackay]

Hi,

On April 11th, a month after the farmer update was released for U.S. users of Google, the primary keywords for ALL our sites significantly dropped in Google.  I have some ideas why, but I wanted to get some second opinions also.

First off, I did some research if Google did anything on the 11th of April... they did.  They implemented the farmer update internationally, but that does not explain why our ranks did not drop in March for U.S. Google users... unless they rolled out their update based on what site the domain is registered in... in our case, Canada.

The primary news release site is www.hotelnewsresource.com, but we have many running on the same server.  EG. www.restaurantnewsresource.com, www.travelindustrywire.com and many more.

We were number 1 or had top ranks for terms like ¨Hotel News¨, ¨Hotel Industry¨, ¨Hotel Financing¨, ¨Hotel Jobs¨, ¨Hotels for Sale¨, etc... and now, for most of these we have dropped in a big way.

It seems that Google has issued a penalty for every internal page we link to.

Couple obvious issues with the current template we use... too many links, and we intend to change that asap, but it has never been a problem before.  The domain hotelnewsresource.com is 10 years old and still holds a page rank of 6.

Secondly, the way our news system works, it´s possible to access an article from any domain in the network.  E.G. I can read an article that was assigned to www.hotelnewsresource.com on www.restaurantnewsresource.com... we don´t post links to the irrelevant domain, but it does sometimes get indexed.  So, we are going to implement the Google source meta tag option.

The bottom line is that I think we put too much faith in the maturity of the domain... thinking that may protect us... not the case and it´s now a big mess.

Any insight you can offer would be greatly appreciated.  Do you think it was farmer or possibly something else?

Thanks,

Jarrett

[AlanBleiweiss]

I have a client who saw big drops beginning the 11th even though they're U.S. based.  A portion of their visits come from international sources.  We're still in evaluation mode though because the site had no significant drops on the initial Panda update here in the U.S.

Preliminary checks of other data show it might partially be a Panda issue, but not completely - oddly enough beginning on the 11th they also had a serious lag time in content loading due to advertiser scripts, which I'm thinking is a big factor as well.

Something that I have been speculating on with several new clients is the gradual degradation concept.  My theory is many of their links coming from other sites that got hit has caused a rolling drop effect.  Since none of my longer term clients have seen any significant drop though it's a more difficult assessment.

[jarrett.mackay]

Did you see any cases where the Panda update went live, but it did not impact a given site till later on... like 30 days plus, which is our case?  I do see a downward trend in analytics starting the day Panda was implemented, but it was´t until the 11th of April did it really drop.

[AlanBleiweiss]

That's your biggest challenge - to find the best ways to monetize enough to justify the editorial uniqueness.  Yet it's that uniqueness that will ultimately bring you back.  Be aware though - I have yet to hear of a single site that's bounced back significantly.  All the sites I've audited that were hit by the Panda update require tremendous changes, especially in the unique content aspect, and in building quality links to category level and deep internal pages.

[jarrett.mackay]

Thanks Alan,

Yes, the number of ads is another issue I intend to address... that will be a tough one to overcome.

Yes, it appears that every site hosted on our server has suffered... even the non related news websites like www.2casarealty.com, which does not have any news stories posted on, but is in the same IP range as the news sites and does have links to it from the news websites.

I would say 50% of the paid releases sent to us are first distributed in our network and then filtered down to other relevant websites... some of whom use widgets to link back to the full story, others use our API and then there are the content scrapers of course.

We have around 50 niche news release sites in our network as well as a number of websites we have launched to further monetize our business... E.G. hospitalityexpos.com, hotelsupplyresource.com

Yes, I full intend to implement that source meta tag and assign credit to the most appropriate site in out network.  Not sure what to do with stories that have come to us after the customer posted on their site or another news release network first.  I suppose the safest method is to find out where it was posted first and give them credit.

I was thinking we could balance the daily submissions with more editorial content that is written by us.  That will help I´m sure.

Part of the problem is the business model itself... paid news release submissions haven´t necessarily hit our service first.

J

[AlanBleiweiss]

Jarrett

There are potentially several other factors as well, beyond what you've already pointed to yourself.

Panda/Farmer issued in a new era where many old "rules" were tossed by the wayside.  Some of these were already questionable, as the seedlings of Panda were planted with last year's May Day update.  if any single site in your network saw significant drops but not the others, it might well have been something else that caused it.  Given the breadth and severity, this clearly does sound like Panda.

Taking a brief look at some of your content, site structure and duplicate content issues gives me even more confidence that's the cause.

Example 1:  Duplicate content severity

I randomly picked one article and searched Google.  Found a dozen sites where the same article shows up, including the site that's currently #1 in organic rankings in my result, which is the actual author's own site (Daniel Edward Craig).  http://www.google.com/search?q=In+this+post+I+discuss+how+newly+available+reputation+metrics+are+paving+the+way+for+the+integration+of+the+online+reputation+management+and+revenue+m&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a

How many sites are in your network where you share the content?  So that begs the question - do you intend to find a way to give credit to the original news source for articles shared across sites?  That's a big recommendation.

Obviously just this factor alone shouldn't necessarily cause bottoming out.  News articles from major wire services routinely spread across the net in massive numbers every day.

Yet it's just one indicator to now consider.  Especially when considering other major factors.  Like the sheer volume of ads on pages.  Too many ads is a big factor now.  Amount of content that's unique to each site is a big part of the puzzle.  If most of the content is duplicated across all sites, what makes each site uniquely valuable?

I could go on, but those are some high level areas for consideration.