Analyze Panda Losers That Don't Fit The Mold

[Shatner]

So we've had two iterations of Panda now, and with each iteration has come a publish list of the biggest losers. We all know, if we're honest, that a lot of the losers on those lists deserved to lose and lost for obvious reasons.

The point of this thread is to pick out the sites from those lists which DO NOT fit that mold, sites which it's not obvious why they lost, and figure out why they were hit.

In doing so, maybe we'll understand why Panda has hit so many here who don't seem to deserve it either. Here's the list of sites to discuss, I suggest we take them one at a time and simply go down the list one at a time and each list reasons we think each site might have been Pandalized. Once we think we've come up for an explanation for that site, we check it off and move on to the next one:

prnewswire.com
blogcritics.org
cinemablend.com
digitaltrends.com
technorati.com
daniweb.com
popcrunch.com
techradar.com
reghardware.com
pcadvisor.co.uk
techwatch.co.uk
just-food.com
computerweekly.com

[Shatner]

Here's one we should DEFINITELY analyze. Sistrix just released a list of Panda 2.0 losers in the US, and on that list was:

Spike.com

That one definitely does not fit the mold.

[Shatner]

Interesting thing about Spike, search for their URL on Google and you'll get literally about a million category pages like this:

[secure.spike.com...]

I mean to me that's valid. If you're for some reason actually looking for "Hooters Snow Angels" then that page is a good resource full of articles on Spike on the subject.

But it's basically the same as the tag category pages we've been questioning whether Google is misinterpreting as Thin Content on other sites.

Could that be what took Spike.com down?

[brinked]

Shatner, lets please stay on topic and go through the current list of sites that were effected by panda. We will be able to get to others once we review all of the sites on the first list.

For spike, its probably is something stupid like those collection or tag pages, I noticed the same thing when I did a site:www.spike.com in google.

[walkman]

One thing: let's not buy the nonsense from Sistrix and 'visibility. ' Apparently it means jack. This was already debunked by ezinearticles and other that looked at stats. A 90% loss of visibility was 10%-15%-20% of total traffic. I am using Alexa and see less than a 20% drop for all the content farms supposedly killed by Panda. Unless Alexa changed their methodology post Panda, which I doubt, then they are accurate to see the changes.

Small sites got killed by losing that occasional but valuable long tail referral, big ones will be making a bit less money but survive and do very well.

Alexa for suite101: [alexa.com...]

For Mahalo [alexa.com...]
and Mahalo can't even rank for their own name

[Shatner]

@walkman Alexa isn't accurate.

I've personally spoken to at least 4 webmasters on one of the previous losers lists, and all reported 40 - 50% loss of Google traffic, one as much as 85%, and in each case it was not reflected in their Alexa rankings at all.

[Whitey]

Unless Alexa changed their methodology post Panda, which I doubt, then they are accurate to see the changes.

Alexa is an indicative "sniff" at the very best.

[walkman]

"Alexa is an indicative "sniff" at the very best. "
"@walkman Alexa isn't accurate."

Did the level of accuracy change after Panda? :)

[brinked]

Looking at alexa to get a sense of a websites traffic stats is like trying to take your temperature with a scale. You will get a number, you'll probably be disappointed and you will end up scratching your head.

I love how when someone tries to tell me what my traffic is because they saw my alexa stats and they will try to tell me site x gets more traffic than me because thats what alexa says.

[incrediBILL]

I thought this thread was serious about Panda... Alexa? NEXT!

This was already debunked by ezinearticles and other that looked at stats.

ezinearticles, another example of what I said they seem to have in common above, syndication, mash-up, duplication of content pure and simple.

What is it Panda's have a hard time doing?

Multiplying, live Panda births in zoos are almost as rare as unscraped content, unique.

Has anyone considered Panda literally stands for UNIQUE content?

Some sites hit obviously look like other factors at work, but then again, to a computer doing content analysis, just how unique was it?

[Whitey]

Did the level of accuracy change after Panda? :)

Walkman - I take your point on consistancy, but still Alexa wouldn't be a tool I'd use for analysis, beyond very loose indications. Even Google Trends has some question marks on it ( and is not up to date to help here ).

Actually what I'm uncomfortable with is the reliability of any statistical lists that have been presented, since it has been challenged by discountvouchers revelation of their own stats ( and we're taking that at face value also ). If we can't be sure of the facts, it's very hard to analyse.

I even saw a BBC report today, but how could they be sure of the integrity of the claims - check this out from the Editor , James Holland over at Electricpig.co.uk which is quoted in the drop :

Searchmetrics analysed Google results in response to a range of keywords, both before and after the Panda update.

Alongside Ciao's 94% reduction in visibility, it found that hubpages.com fell by 85% and eHow.co.uk dropped 53%.

A similar analysis by Sistrix found a 81% drop in visibility for Ciao.co.uk, 72% reduction for hubpages.com and an 84% fall for eHow.co.uk.

While a sharp drop in visibility may constitute a crisis for some websites and their search engine optimisation (SEO) engineers, it does not necessarily spell disaster.

Technology news website Electricpig.co.uk was downgraded by 94% by the Panda update, according to Searchmetrics.

Site editor James Holland told BBC News: "We haven't seen an immediate impact.

"Comparing our traffic from Google for that week, we're actually only down 0.5% versus the week before Panda took effect. [bbc.co.uk ]

i think we have to be careful to conserve focus on areas of meaningful data , and the only folks to have that are the site owners experiencing traffic drops themselves and the rhetoric between them at this stage. Please update me if I'm wrong.

[walkman]

Has anyone considered Panda literally stands for UNIQUE content?

Wow! Or it may be the family name of the engineer that came up with the criteria for the mess? Mighty conspiracy :)

[incrediBILL]

Or it may be the family name of the engineer that came up with the criteria for the mess?

That was just the cover story.

[tedster]

to a computer doing content analysis

We do keep slipping away from the content itself - and that was the expressed target - "shallow content" or, as the rest of the world named it, "content farms."

Content analysis by computer has come miles in recent years. There are even applications that do pretty well at deciding whether the author is male or female.

Google must have a version of that technology. They may be doing direct textual analysis of some kind on pages. It may even be the area where "the breakthrough" occurred.

[edited by: tedster at 3:09 pm (utc) on Apr 17, 2011]

[indyank]

Pls. don't again use the term "content farms". It was a confusion created by searchengineland. There is nothing to prove that panda acted only on content farms.Sites of varying sizes have been impacted and this algo has nothing to do with content farms.

[tedster]

Agreed - but no matter what sites actually got hit, Google did say they were targeting "shallow content". So my point is that as we look at multiple factors, we should not forget the content itself. It seems to me it has to be in the mix directly, in some form.

[indyank]

Tedster, I agree. But it also means that it will be very difficult to recover unless a lot of pages are cleaned up or removed.Moreover, site owners would find it hard to understand which content google dislikes.

[SouthAmericaLiving]

@Nano - makes sense, thanks.

@Tedster - other thread was excellent, clear in explaining why bounce rate is not used in rankings.

[incrediBILL]

Google did say they were targeting "shallow content".

Defining shallow content as duplicate content, like an affiliate site, a syndicated site, a scraper site, product comparison sites, down to ecommerce sites with the same copied manufacturer product specs repeated over many sites.

What do they all have in common?

Same old stuff over and over again just showing up in new pages, shuffled around in new orders, but at the end of the day just another copy of the same old stuff.

Google has definitely killed the concept of article marketing, because using those articles could now be a simple death sentence for any site including that free content.

[brinked]

So I have been thinking about spike all day. Why was it hit by panda?

There content is not copied from anywhere, they arent plastered with ads everywhere.

Then I read incrediBILLs post.

"Has anyone considered Panda literally stands for UNIQUE content? "

and then it dawned on me. All of this time we are thinking about unique content as an article that was not directly copied from another source. But how about we change the word from unique to original for this panda update. If we look at spikes content, none of it is original. Go down the list, JLO is worlds most beautiful woman, kobe bryant fined 100k, miley cyrus feels unloved etc. All of these stories are all over the web. They are just essentially rewriting the same article thats already on every other gossip site.

I see a lot of people selling spun articles, I believe these are software meant to rewrite an article automatically in order to trick google into thinking it is a whole new article. I think this is something google would want to put an end to right?

So my ads theory still makes sense. In many of the cases of the pandalized sites, most of them are probably just rewriting the same story but plastering there ads all over it. So google panda comes in and says "well this story is all over the web written a million different ways, we are going to give credit to a site that offers a better readability experience so the user doesnt have to get so distracted with ads.

[Shatner]

>>Agreed - but no matter what sites actually got hit, Google did say they were targeting "shallow content". So my point is that as we look at multiple factors, we should not forget the content itself. It seems to me it has to be in the mix directly, in some form.

@tedster the Websites I picked for analysis in this thread were picked in large part because they do NOT contain "shallow content". That's why they don't fit the mold.

[helpnow]

Content is the key. Everything else is just correlating or supporting or coincidental factors.

[Shatner]

@helpnow The purpose of this thread is not to make broad statements.

Please support your statement using one of the sites above as an example. Do you find that any of those sites don't contain high quality content? Because I don't.

[incrediBILL]

Do you find that any of those sites don't contain high quality content? Because I don't.

What does the quality of the content have to do with this?

It's obviously the rampant replication of the content.

Once it's replicated all over the place the quality of the sites containing replicated content has been diminished to the end user now encountering it on multiple sites.

[brinked]

@incrediBILL,

It does make a lot of sense, as a lot of these sites do syndicate there content. But I ask you this...what about a site that reproduces great content and that content then gets copied all over the web? Is it there fault for others copying them?

[Shatner]

>>Once it's replicated all over the place the quality of the sites containing replicated content has been diminished to the end user now encountering it on multiple sites.

Not all of these sites in the list are doing that though. Unless you're counting scrapers? One thing about all of these sites is they are HEAVILY scraped by content thieves.

Nothing they can do about it, and it isn't their fault, but it is happening and it is duplication.

[Dan01]

Spike.com

That one definitely does not fit the mold.

They have a lot of videos on their front page. Not much written content. Their written content seemed shallow.

BTW, there was an article in Website Magazine that found that most vendors don't add proper tagging to their videos and they are not index by the SEs.

They don't have any advertising on their home page, and only two ads on the content pages. Ads don't seem to be a factor.

[wanderingmind]

What about the heavy footer on Spike? Too many links?

[wanderingmind]

And continuing IncrediBill's line of thought - is it that Google has given up trying to figure out quality of content entirely - because there are thousands of articles about the same topic, each differing to some extent - so they say, these are all the same (even though actual content is different), not unique enough, so when it comes to them, we focus on site experience?

[incrediBILL]

Not all of these sites in the list are doing that though. Unless you're counting scrapers? One thing about all of these sites is they are HEAVILY scraped by content thieves.

If you provide an RSS feed the site republishing it isn't scrapers, they're aggragators.

I went through many of the sites listed and I see tons of identical results in the SERPs.

Who's the original author, who syndicates, who scrapes, who cares?

Apparently the Google algo doesn't care and is making buggy as hell decisions to boot.

what about a site that reproduces great content and that content then gets copied all over the web? Is it there fault for others copying them?

Yes, if they provide automated means like RSS for allowing it to happen without using DMCA's to stop the republishers.

The only way to have unique content is to defend unique content and not allowing it to be syndicated or scraped for that matter, both pretty easily accomplished for the most part.

[arikgub]

We do keep slipping away from the content itself - and that was the expressed target - "shallow content" or, as the rest of the world named it, "content farms."
..............
Google must have a version of that technology. They may be doing direct textual analysis of some kind on pages. It may even be the area where "the breakthrough" occurred.

Honestly, it sounds too complicated to be true. Even if Google finds some correlation between the text semantics/statistics and the human perception of "shallow" and "deep" - well it is hard to believe it is strong enough to base any algo changes on it ...

It must be something much more simple. May be it is just a distribution of incoming links?

Content farms tend to have very "uneven" inbound links distribution, with a very few deep links. On the other hand, high quality content attracts more deep links. This is a very simple metric to apply and could explain a lot of what we see after Panda. All the news site usually have a lot of deep links, so they are up. Ezinearticles and similar do not have much deep links and the deep links that they have are usually spam put in place by the article writer.