Many Weeks since the Panda Update - Any Improvements? [part 2]

[rustybrick]

< continued from [webmasterworld.com...] >

So, still, no one is seeing any significant improvements?

[edited by: tedster at 5:00 pm (utc) on Mar 25, 2011]

[Content_ed]

@tedster

I wonder if you're falling into the twin traps of "Google tells the truth" and "Google is always a step ahead".

Yes, Google claims high and low that it's about thin content, but that's not what I've seen and it's not what I've seen reported. eHow is the king of thin content. The article farms that they hit featured fatter content than eHow, though it was stolen or garbage. The only common element I've seen in the sites that were big Panda winners is that they were brand names.

And from my standpoint, the penalty acts exactly like a site-wide version of the old single page duplicate content penalty. Maybe I'm only familiar with how those worked because I had a single dominant page years ago that made it obvious when the duplicate content penalty hit, and when I got it removed.

I know several of the regular posters here have married Google's "it's the algo" concept, but what difference does it make how Google thinks it works if in practice it does something else. I'm sure many of us here have worked with some highly complex systems in our lives, and everything about the way Google went about creating this update screams that they don't really know how it works, they just believe the aggregate results justify using it.

I can't judge whether 90% of searches are returning slightly better results, but I can see that 10% or more are returning total garbage now, where pre-Panda, there were always a couple reasonable candidates above the fold.

[Shatner]

>>Since this re-ranking is applied site-wide I think any new content added will be treated with the score for the old content.

Re-ranking was applied site wide, but they could easily remove the penalty for any new content posted and leave it for existing content.

Has anyone noticed new content ranking a little better?

[crobb305]

I'm seeing something interesting. In my stats, I see some sporadic traffic from a few of my main phrases that were severely demoted after Panda. When I click the referring Google.com url in my stats, I go to the first page of Google results and there's my site -- with 4 site links ranked in the top 5. If I hit refresh, it's gone. I can hit refresh a few times and it will sometimes reappear. There is either some testing of different indexes, or a new index that could be emerging. To be clear, I am seeing Google U.S. -- not international.

[edited by: crobb305 at 8:35 pm (utc) on Mar 25, 2011]

[chrisv1963]

What are the chances that Panda has basically applied a permanent penalty to all the old content on your site

Then Panda should hurt Wikipedia and eHow badly. Some of the content on those sites is old too.

[tedster]

Yep - Danny Sullivan first called it the Farmer Update. In my opinion, that was a misstep, because it focuses attention incorrectly. Ecommerce sites have been hit, forums and blogs have been hit - and some sites people call content farms have even improved.

Since I realized this, I've been calling it "Panda" so I don't complicate things even more. I was just at SES in New York and noticed that most of the industry is now calling it Panda. That seems wise to me. As politicians know very well, you can misdirect a lot of people with your word choices.

[edited by: tedster at 8:59 pm (utc) on Mar 25, 2011]

[netmeg]

I still don't think all the shoes have dropped. Just because some sites <cough>eHow</cough> didn't get it this go-round don't mean they won't in the future. There's no reason in the everlovin' blue eyed world that Google has to do everything all at once, and lots of reasons for them not to. So I'd stop pointing to eHow as proof or non-proof of whatever theory you'd like to believe. In the grand scheme of things, eHow is anecdotal evidence, and not statistically significant.

[econman]

eHow is anecdotal evidence

By far the most interesting evidence I've seen concerning eHow, to my mind fully explained why eHow survived Panda, while their direct competitors were hit badly.

This study showed that the other content farmers had published thousands upon thousands of pages focused on a small group of topics that are associated with spam, whereas eHow had a much smaller set of pages related to those same topics.

My reaction to that study was that the topics or keywords identified in the study were probably not, of themselves, being used as direct "indicators" of low quality by Google, but rather, this pattern suggested a subtle but fundamental differences in their respective business models, internal rules, or editorial standards. Because of those differences, eHow was exercising tighter editorial judgment, which allowed it to excape Panda (perhaps just barely), whereas competitors who were willing to publish so many thousands of pages about such obviously spammy topics, were probably too loose in their judgments concerning other issues, as well.

This reaction was reinforced by the observed differences between the Squid site and the Hub site, with the former having tighter editorial control over spammers, and surviving Panda, while the latter site was not as careful and had been hit by Panda.

[Content_ed]

I don't know who did the eHow study, but I can tell you in that in a couple of my subject areas, they have thousands of pages to my one. They simply pin the same answer into a zillion how-to's. Want to know how to paint a house?

How to paint a house red
How to paint a house pink
How to paint a house red and pink
How to paint a house re with pink trim
How to paint a house pink with red trim

repeated over and over again for every possible combination that their algo says will pay.

[walkman]

Here's a satire of a content farm article, minus the ads: How to pour milk [thecontentfarm.tumblr.com]

No need for satire, some Einstein
even made a video on associated content "How to Pour Milk into a Mug" [associatedcontent.com...]

The user comments are hilarious and well worth:
"Hey man, you spilled some! Can't wait for the sequels, How to wipe up a small spill with a damp cloth, how to drink milk and how to put milk in the refrigerator."

Re-ranking was applied site wide, but they could easily remove the penalty for any new content posted and leave it for existing content.

They way I understood is that when Google re-indexes and calculates again, if your pages pass the threshold, that penalty is removed automatically. Say you had 100 thin pages and Google penalized your entire site because of them. You noindex or 404 them. Google grabs the site and sees that these are gone. When they run the algo you are no longer penalized.

[outland88]

That’s an excellent article Shatner highlighted from Search Engine Land(everybody should read it) and it’s what I’ve been trying to convey with my posts. You aren’t necessarily being penalized and the quality of your site is not necessarily bad. Google is basically re-shifting what they feel to be the best (more branding) to the top. That’s why people can’t explain the variations they are seeing. Plus if you analyze it you’ll understand why you’re seeing more duplicate content.

Improving your site has always been an option for increasing your rankings. This time though many will be in survivalist mode. Adsense on your site or the positioning had nothing to with it but it was the doomsday machine. Since Google seldom policed that program and the users were using every scheme imaginable Google was going to have to eventually deal with all the inferior and thin content in it. All small businesses whether they used it or not were going to have to pay because of it.

Also what you need to realize is Google is rightfully facing tremendous competition and resentment. In other words they’re trying give-backs to big business at the expense of small business to get that quality content. More and more companies on the other hand are saying we aren’t giving Google squat for free anymore period. No true business does.

[tedster]

I'd say it will be more like Google changes the score that the classifier has assigned to your pages. But it hasn't happened so far, so we're just shoveling smoke until there are some hard facts to work from.

[outland88]

Well what do you think of that research in that Search Engine Land article Tedster.

Its like I told IB, when was the last time you woke up to a headline reading "The Poor Getting Richer."

[tedster]

I assume you mean their coverage of the article at CNN, Google's code change shifts billions from losers to winners [money.cnn.com]. But there's no "study" in that, just some smoke shoveling by a financial blogger.

If you take out a lot of junk results, it seems inevitable that well-established sites will gain some of the traffic as a result. I doubt that such a consideration was in any way part of the plan - nor do I think it will be part of any future tweaks. I do own a fine collection of tinfoil, but nothing in that hat size ;)

Food for thought - it was almost definitely Adsense clicks that took a big piece of the loss, right?

[outland88]

Google has been positioning itself for a few years to separate not divorce itself from the woes of smaller Adsense publishers. Why because of the potential of lawsuits and the rising amount of DMCA’s that Congress might subpoena. In other words until Google captured a larger share with big business they were quite content to pursue a policy of “use em and loose em”. So to answer the question No, I don’t think Google necessarily saw any significant loss.

[magpete]

can someone explain me please what is "thin content" because I see on #1 results many pages with content "under construction" ! of a certain website that ranks for almost every targetet keyword .........

[snickles121]

"What are the chances that Panda has basically applied a permanent penalty to all the old content on your site"

I have one old site that has not been updated in years except minor coding changes, but content still is good and accurate. It took better rankings after Panda.

[walkman]

If we're to go into conspiracy theories, New York Times, WSJ and the likes can embarrass Google, and have done so, so Google has an extra incentive on that department.

But my gut tells me that they are trying to milk the cow for as long as they can. So if you want to buy something they'll send you to Amazon or Walmart, if you want med info to cdc.gov and Mayo clinic, insurance to Geico and All State....and call this the magic of brilliant PHDs. The rest can starve as far as google is concerned. A company with a 70% market share and some $10 billion in earnings has one employee answer a question or two a week that gives advice that actually harms you "Design for users not search engines" more than the advice Adsense gave to site owners "Why don't you flood your page with 3 ads" .

Their scoring system is so great that you can be #1 for a decade and all of the sudden, with no changes on your part, you can find yourself on page 8 and firing dozens of employees. Just because a guy named Panda decided that graphics mattered in a page.

[outland88]

I’ve never supported the idea that Google is in for the long run. They’re in it for it as far as they can take it before legislation catches up with them. Bottom line Google’s business models and profits seem to tamper a great deal with other people’s copyrights.

[browsee]

@walkman, well said! Its been 28 days, they are crawling everyday and I did not see any change in US search results. There are more than thousand replies in the GWT forum, no single answer from G till now. We need to either live with it or look out for alternative solutions.

[tedster]

can someone explain me please what is "thin content" because I see on #1 results many pages with content "under construction" ! of a certain website that ranks for almost every targetet keyword

Welcome to the forums, magpete. Thin content is a phrase originally created to describe sites that duplicate things like manufacturers description and affiliate feeds, but without anything much of an original nature to add value for the visitor. The Panda update is aimed at what Google engineers have described as "shallow content" - somewhat similar but without the duplicate factor necessarily being in place. Just filler content that doesn't add anything of value, in other words.

We can't base decisions on the occasional anomaly we see in the SERPs such as under construction pages. That happens, but it doesn't last - and we want our rankings to last.

[Shatner]

>>The Panda update is aimed at what Google engineers have described as "shallow content"

Correction. Google has HINTED that might be the case and a lot of tech media has speculated that's what it's aimed at... but we don't actually know if that's true and so far there's been no data that proves it's so.

Google has also HINTED that it might have a lot to do with branding more than anything. We also don't really know how true that is, except for some anecdotal evidence of big business coming out better in the update than smaller business.

[Shatner]

So I think I'm definitely starting to notice a trend in my content.

All of the content which used to be ranked well on the site before Panda remains completely and utterly buried, penalized, by Panda.

However I'm noticing that a lot of the new content I've posted in the past two or three days is being ranked fairly well, in some cases nearly as well as I might have expected it to before Panda.

Maybe it's just a coincidence, I need a few more days of monitoring to really see if it's anything...

But I'm really starting to believe in my theory that all the pre-Panda content which was penalized is just going to stay penalized no matter what, but new content going forward may not have the penalty applied to it if whatever it was that caused the Panda penalty in the first place has been fixed.

That's not really good news if true, since it means that 8 years of content is now garbage, and it's not like I can make that up with new content. I'm sure that's true of most people who were hit by Panda.

But at least it's something.

[tedster]

another thought. I also notice that URLs dinged by Panda aren't moving even when fresh content on the same domain is ranking. Maybe Panda is such a major change it's not fully integrated into the Google infrastructure right now - so it will only update in major waves or dances for a while. It's a thought, anyway.

[magpete]

Thanks Tedster.
I want to ask you guys (the ones that as they say did not worth a downgrade in US index)
Does your websites targeting US or they have a global content? , if the second occurs then what is your traffic outside US, like in Canada, UK, Australia, Europe.
Do you have content in other languages? and how do you rank there?
Do you host the websites in the US or somewhere else, there is the option of geotargeting in WT.
"If no information is entered in Webmaster Tools, we'll continue to make geographic associations largely based on the top-level domain (e.g. .co.uk or .ca) and the IP address of the web server from which the context was served.
If no information is entered in Webmaster Tools, we'll rely largely on the site's country domain (.ca, .de, etc.). If an international domain (.com, .org, .eu, etc) has been used, we'll rely on the IP address. If you change hosting provider for a country domain, there should be no impact. If you change the hosting provider of an international domain to a provider in another country, we recommend using Webmaster Tools to tell us which country your site should be associated with. "
What I try to say is that there are too many factors , I know great sites .co.uk that they have amazing global content but nowhere to be found in US results.
Finally I recon that this new algo takes as another option where in the world you have more traffic ( check Alexa or your Analytics).
Any ideas?

Finally I'v seen in google.com and google UK resurrection of some old and unupdated domains (some of them from 97) on top ten results.

[kd454]

A couple of people are posting in Webmaster Central that the U.K Panda is on loose.

[walkman]

another thought. I also notice that URLs dinged by Panda aren't moving even when fresh content on the same domain is ranking. Maybe Panda is such a major change it's not fully integrated into the Google infrastructure right now - so it will only update in major waves or dances for a while. It's a thought, anyway.

Looks like Google locked us in a Panda prison and said 'we'll take you before a judge some day' ;)
Or maybe some came back and aren't telling (unlikely but who knows)
Or maybe Google threw the key away for x months.
Or Google run the algo again and everyone failed.

It's getting very frustrating and very expensive.

[reblaus]

I also saw the first UK Panda reports. This would mean though Google thinks Panda works and is now rolling it out world-wide. It also means no relieve in sight for those who lost traffic.
It really hurts seeing how well I ranked on google.com when accessed from Europe versus accessing it from the US. With google rolling out Panda world-wide the international organic traffic will also die for my site.

[outland88]

But I'm really starting to believe in my theory that all the pre-Panda content which was penalized is just going to stay penalized no matter what, but new content going forward may not have the penalty

That’s what I’ve been wondering for about a week now even though I’ve long known you can be chasing your tail with any changes. Initially the url is seemingly penalized for the keyword phrase. The sign it is: If you even attempt minute changes it plummets even further. New content seems to go unpunished.

What I find so disturbing about the Google engineers is they talk in such vagaries and generalizations. They likely don’t even know what the truth is, or the concept is so alien to them they can’t express a clear and concise idea without fear of losing their cushy jobs. Take for instance Mu’ statement that non-quality content should be deleted. Assuming the fact you know what pure unadulterated junk is and you were designing with that totally in mind what actually does the word “quality” mean. What Google means is branding, large companies, educational institutes, news organizations, large media, original research etc.etc. The point is many of those engineers at Google need to “cowboy up” and start dealing in the truth. They just hope you will slowly accept they took the Internet from the little guy.

[maximillianos]

I would guess it is not permanent. Google doesn't permanently penalize. Take this site for example:

< link removed >

They had created hundreds and hundreds of fake blogs and used them to point thousands of links to their site. You can't get much worse than what they did... and they were only penalized for about 12 months or so. Beyond that they had scraped tens of thousands of pages from other sites to seed their site.

Now they are back and ranking at the top of serps again and pulling in close to a million uniques a month according to Quantcast.

So if Google is going to permanently penalize some sites for too many ads, or some thin content, then some thing is really wrong in this world when sites like the above get a free pass back into serps.

Don't even get me started on what the above site is doing right now to gain serps rankings. Pretty much the same thing they did before, just more covert.

They are the poster child for why this update failed to catch the worst offenders.

NOTE: Just wanted to clarify the link above that was removed. It was a link to an article on SEOMoz that supported my above statements. Just wanted to add this clause in case folks thought I was making this up... But since the article does mention the site, I completely understand why it was removed. =)

[edited by: tedster at 7:42 pm (utc) on Mar 26, 2011]

[walkman]

I sincerely HOPE that Google with their dupe-detection technology (alpha level at best) didn't give us a set time penalty. Just today a well documented case on Google Support forums shows how an original publisher ranked below a dozen or so sites that had copied her story-verbatim. Losing clicks is one thing, getting a site-wide penalty for Google's fault is totally another.

Thin content is not the problem, not by itself anyway, and some content is supposed to be thin. I see plenty of thin sites doing very well. What sucks is that Google has re-indexed pretty much all the sites in the post-Panda month and nothing happened.