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]

[incrediBILL]

Amit said about the engineer named Panda: "He basically came up with the breakthrough a few months back that made it possible." That doesn't sound like their just mashing up traditional factors to me.

I'm definitely seeing multiple factors at work so focusing on just the "thin content" probably isn't enough, it's definitely more complicated than some of the past updates that were easily side-stepped.

I put up a test I'm still waiting for Google to add to their index today, very odd, it's been hours, hope we're not updating slow again.

[tedster]

The panda handle sounds like a misdirection

When the Wired interview was first published, I did a search for this engineer. Now that so many update articles have flooded the web, it's hard to find the same results I did at that time for a last name ;)

At the time, I found at least two people associated with Google with the last name Panda (probably pronounced Pond-Ah). One of them, Biswanath Panda, has a focus on machine learning. See this Google Research page [research.google.com] for some of his published work.

[bluntforce]

From what I have read, people are sacrificing parts of their sites to appease the Panda god.

Perhaps it wasn't clear I have pages that are both low on content and indeed, affiliate feed based.

My low content pages and my affiliate pages do not have any major connection points and I saw no Panda impact.

"Buyers also bought this" is one of my major turnoffs as a buyer, if it works for a given person running a site, power to them. Matt Cutts in that interview said "adding value for users versus not adding value for users". That kind of opportunity to link to other pages on a site reeks like SEO, not value added for the user.

[incrediBILL]

That kind of opportunity to link to other pages on a site reeks like SEO, not value added for the user.

Not at all.

Linking to helpful information is adding value, the more links to get users to what they're probably interested in the better.

Look at Amazon, linking like crazy to sell products, not SEO

[Robert Charlton]

"Buyers also bought this" is one of my major turnoffs as a buyer...

Interesting. I like it. I think it's one of the smartest things they do, and it helps in a variety of areas. I think it's adding value for users, a good way of increasing sales, and it's good SEO.

[chrisv1963]

Google needs to learn a thing or two from Bing.

Everyone thinks of Google as the best search engine and Google thinks they own the internet. This is no longer true. A couple of engineers broke Google in February and it's about time we see a "Larry sacks Panda" update.

Still no update on Matt Cutts' blog. It has been a month now. Why? Too many internal problems at Google? No longer allowed to communicate after messing things up?

[Pjman]

This would be a perfect opportunity for Bing to flex some Microsoft muscle. This is the first time ever, I am rooting for MS to do anything. Over the years, they have become less evil then Google.

They rushed this algo. update because of all the bad PR they were getting. They got beat up in the press, the worst I have ever seen, in January. All targeting their core business, i.e. Google results aren't as great as they say.

Think about it:

1. Is it just a mere coincidence that they happen to put the most unpredictable algorithm ever, they next month? I think not.

2. They have a new CEO who is trying to say "Back to Basics", i.e. "We rule Search! Our product is light years more advanced then everyone else!"

Panda is probably something they were working on for release in the future, but no where near ready for the production environment.

Business 101

They got embarrassed by the press. What happens to a stock when some one in the media attacks your core product? Stock dives!

How do you fix it? Make a better product or at least convince the press that you have.

That's what this is. I can not believe that no one in the press has the brains to look into how bad the results have gotten. As opposed to say what ever Google says in a press release.

We need an Aaron Wall or somebody to get the attention of big stream media about this. Hey Mr. Paperman, Google has the wool pulled over your eyes!

[dazzlindonna]

""Buyers also bought this" is one of my major turnoffs as a buyer.."

I purposely look for this when I'm buying something. If it exists, I'm glad. It's basically a part of my buying research.

[ascensions]

Here's what I don't get... isn't the point of a search engine to contain as much data as possible, and then rank it accordingly... If we're deleting content, (which apparently is the only fix) -no matter how thin it is, it's possible we're deleting something based on its size not its quality... meaning there's data or content going away because of this update which seems very counter-Google, atleast the way they stood in the past. Wouldn't they want to stop this travesty before it's too late? How many years mankind's web-work is going to disappear due to a single poor decision?

I'm also very curious about these "people" who they used to tell them their results were "better". I can go to my local supermarket and ask people what the best cheese and 90% of them will say "American" because they're IDIOTS or if I'm to be nice... ignorant... they've never been to France and had Gruyere. Likewise ask most people about search quality and index quality and they're likely to give you answers based on what they know- not what is really better.

Now... I'm going to go buy my some a baguette, some cheese, and a big bottle of French wine while I still can and crawl into the fetal position in the corner till this whole thing blows over.

[Content_ed]

I don't understand the confusion over what the Panda update consists of. I think Cutts and Singhal explained what the engineer "Panda" came up with, and we've discussed it several times already.

Everything beyond the new human "quality" filter appears to be symptoms of the penalty caused by loss of curbside appeal, not new factors. Sure, there may be some other minor tweaks, Google is always making those, but either your site looks (to an algorithm) like something people will have confidence in or it doesn't.

Those of us with deep content sites who've spent a month sending out DMCA requests are doing it because the junk sites are beating us in the results with our own content, but the duplicate content issues are only hitting because the underlying site has been weakened by a Panda mauling.

It's like all of the diseases that kill people who have been starving for months on end. The particular disease isn't really the cause of death, it's that the body has been weakened to the point that anything would do the job.

Once the body is weakened, the you can only try to cure the disease at hand. In the long run, getting back on a healthy diet can prevent other diseases from stepping in, but I have no reason to believe that's doable for existing sites unless they want to be somebody else.

It's not as simple as eating more (generateing better content), many of the sites that have been hit already have the best. It's more like becoming more acceptable to the 20-something society at Google, getting a nose job, becoming sexually promiscuous (social networking), seeking therapy (SEO?).

Some people may be willing to do these things, I'm not. And if you do those things, you'll probably just find that the cool kids in the plex still don't like you, because they've gone on to digging people whose websites autodetect the language and do do text-to-speech over cellphones.

[walkman]

ascensions, you make some very good points. On Google you cannot even delete 4 year old 'deleted' adwords (in fact you have to still update them to meet the new standards) and email on gmail was never meant to be deleted. But, there's nothing we can do, we must do whatever we think Panda likes.

Of course a month from now we'll hear that Google now penalizes for deleting too much stuff at the same time ('bad quality signal') :)

[dazzlindonna]

Of course a month from now we'll hear that Google now penalizes for deleting too much stuff at the same time ('bad quality signal') :)

LOL! And this is the real heart of the matter, imo, and it's one that can make a grown person laugh and cry at the same time.

It's what makes the statement "design for users, not search engines" a basic joke. That should probably read, "design for users, until the search engines slap you silly, then design for search engines, until users slap you silly, then design for users until........"

[incrediBILL]

Note that Panda still hasn't rolled out in the UK best I can tell.

Probably Google fears a big EU regulatory beat down if what happened in the US happens in the EU as well.

That fact alone makes me think they're watching and retooling Panda because it would probably be worldwide now if it was really ready for prime-time.

[dazzlindonna]

Yeah, us poor US slobs get to be the slaughtered guinea pigs. Fun times.

[ascensions]

Q: What do Google engineers eat for breakfast?

A: Pandacakes!

[kd454]

This could be a sign of hope.

My site has returned this A.M for a keyword that was pushed down to #15-18 when Panda started, it is now at #8 for this big kw, pre-Panda it had been sitting any where from 3-8.

I did remove what I considered "thin" pages from this site and replaced with unique content all over 500 words.

2 days ago I removed a banner ad that was above the fold on the homepage, it was the homepage that was pushed down for the major keywords for the most part.

I will report back if there are further improvements or declines.

[mromero]

kd454

"2 days ago I removed a banner ad that was above the fold on the homepage, it was the homepage that was pushed down for the major keywords for the most part."

Does this mean you are dumping your direct advertisers?

Are you painting yourself into a corner?

What does this portend when the next algo, maybe Mr. Pinata, decides no direct advertisers is the best quality signal?

[mromero]

dazzlindonna

"Yeah, us poor US slobs get to be the slaughtered guinea pigs. Fun times."

Not really.

Many manufacturers / producers are located overseas but their primary market (and site), think export, is based in the U.S. and the bulk of consumers are U.S. nationals. They too have been pandalised ;o)

[crobb305]

What does this portend when the next algo, maybe Mr. Pinata, decides no direct advertisers is the best quality signal?

They have specifically mentioned ad-to-content ratios. The problem they (Google) have created is that through the nature of penalty/filter/demotion (whatever you want to call it), scrapers that have a slightly higher score than your penalized site will be allowed to rank higher for your content. As webmasters become aware of this, they begin to think it's a battle they can't win (with respect to content), and so they stop creating it (for the duration of time they see others outranking them for their material). It then becomes a rational solution to start removing ads (including Adsense) to reduce the ad/content ratio. It is a negative feedback loop that Google has created that may ultimately hurt THEIR revenue. Google, if your scoring mechanism is going to allow scraped content from my site to outrank me, then I will remove Adsense ads. I'll play your game. Algorithm fail.

[edited by: crobb305 at 4:54 pm (utc) on Mar 27, 2011]

[incrediBILL]

I think Cutts and Singhal explained what the engineer "Panda" came up with

Yes, yes, we've all read their theory of Pandativity

P=MC²

Panda = Matt Cutts²

But not all the clues match their simplified blanket statement.

[kd454]

Does this mean you are dumping your direct advertisers?

Are you painting yourself into a corner?

What does this portend when the next algo, maybe Mr. Pinata, decides no direct advertisers is the best quality signal?

No I am not dumping all direct advertisers, I just am stopping this particular one for the time being.

I have 2 sites that are almost identical in every aspect except for the niche. One of these sites was hit with Panda and the other got a 35% increase after Panda.

The differences between the 2 are, the site that was effected had 18 pages of thin content (resolved), index page has banner ad above the fold (resolved), the menu is much more loose on the site that was effected (not resolved).

I am waiting change the menu on effected site, but that would be my next step.

It will take a couple days to see how traffic has changed, I do not want to jump to any conclusions at this point, but it does look positive when from 18 to 8 overnight.

On a unrelated notes or it could be filters have lifted, a site of mine that got a -350 or so penalty finally came back to the 3rd page yesterday, I was shocked to see there after 8 months or so.

[edited by: kd454 at 5:32 pm (utc) on Mar 27, 2011]

[superclown2]

I'm not convinced that Panda isn't here in the UK. I've watched some tremendous changes in my sector over the past two weeks.

[kd454]

I'm not convinced that Panda isn't here in the UK. I've watched some tremendous changes in my sector over the past two weeks.

From what I am been reading I am thinking your right several people reporting some Panda like hits over the last few days/week.

They could be putting on select data centers to see the impact.

[magpete]

"Note that Panda still hasn't rolled out in the UK best I can tell. "
1)How do you know that? BTW I see changes in UK

[mromero]

superclown2

"I'm not convinced that Panda isn't here in the UK. I've watched some tremendous changes in my sector over the past two weeks."

Agree with you.

One site we monitor has had a cautious panda update, pushed from #1 to #3 in the U.K.

A government controlled and funded site (and a government high on the U.N. corruption index at that) and the ever pervasive wikipedia are 1 and 2 respectively - play it safe kind of thing. So far the CIA and the state department have not been rolled into this niche area.

[incrediBILL]

How do you know that? BTW I see changes in UK

I monitor changes in my niche worldwide and so far it looks like pre-Panda in the UK.

The upshot to that is I have 2 Google data sets to compare against which provides lots of interesting details about all this Pandamonium.

Of course they could be doing a partial roll-out or the data center I monitor could be last of the old code, there are reasons, but I've tested Canada, Aussieland, etc. and they looked pre-Panda as well.

One site we monitor has had a cautious panda update, pushed from #1 to #3 in the U.K.

That sure doesn't sound like Panda, you'll know it if you see it.

The web as seen thru a Google lens suddenly turns topsy-turvy.

[crobb305]

the top-ranking site across several major phrases in my competitive industry has 14 affiliate links on the homepage, and all pages linked from the main nav (on every page) has 3 to 10 affiliate links. Each affiliate link (even on the homepage) has several bulleted ad descriptions. Talk about high ad-to-content ratio. I count at least 25 bullets on the homepage alone.

All affiliate links are nofollow and most navigational links at the top of each page are images. While this could be helping, it seems the algorithm would pick up on the excessive and repetitious ad wording in the bullets.

Perhaps there is something to the notion of excessive linkage to inner pages? Like I said, this particular site uses images, therefore there is no overuse of anchor text. Just a guess. I am shocked by the number of affiliate links and ads.

[edited by: crobb305 at 6:43 pm (utc) on Mar 27, 2011]

[tedster]

isn't the point of a search engine to contain as much data as possible, and then rank it accordingly

If so, then Google wins and Bing loses. Bing's indexation of almost any site I work with is FAR shallower than Google's. In fact, that may be why Google results seem so strange at times - they are trying to sort through so very many pages.

I'd say Google considers the point of their search engine is to match the intention of as many users as is possible. That's essentially what matters to them.

Those who lost rankings with the Panda update think it sucks. Others think it is an improvement. It's "courses for horses", as they say. In my personal searches, I find the results better than they were in January, and I'm less troubled when I'm looking for pages that I know exist.

not all the clues match their simplified blanket statement

Yes, exactly. Matt Cutts and Amit Singhal were describing how the training set was created, but not all the factors that machine learning put together to create the actual algorithm.

[BrianS]

Those who lost rankings with the Panda update think it sucks. Others think it is an improvement. It's "courses for horses", as they say. In my personal searches, I find the results better than they were in January

Now that Panda has hit the UK, Ive finally had a chance to look at its effects, and I concur with Tedster, they are presently (significantly) better than they were previously.

[walkman]

After reading this (last comment about an Indian portal) and trying some searches: [google.com...]

I'm 100% convinced that Tedster is right. Some sites have such a high G quality score that they will rank higher than the original even when they link to a sentence from another site's story. So google is going for safety and for brands. They have lost control of the search and are sending users for all searchers to more popular sites. So people will get screwed but who cares.

This is another blatant example: [google.com...]

Now the million $ question: How often does Google run their quality score? I doubt it's on the fly.