Penguin May Not be What You Think

[zartania]

OK, here's some more Penguin feedback that has me a bit frustrated. I have an EMD with a few hundred links that was evidently penalized. Two of the top three sites that are now ranking are not EMD's (one of them is) and have less than 10 links for that keyword phrase. The interesting thing is that ALL 10 LINKS ARE EXACT MATCH. They have no anchor diversity. It's not a temporary thing either. They appeared right after April 24th and have stayed there.

What's up with that?

 

[assphuck]

Check their onpage SEO and compare to yours. Also check social signals.

Penguin sux if you ask me, but Google apparently wants to display a lot of garbage in some niches and default to big corporate brands in others.

It's a shame backlinks can hurt sites now. One of my sites I'm seeing a lot of .ru links popping up (not from me) and there is not a damn thing I can do about it except to wait for the hatchet to drop.

 

[zartania]

Check their onpage SEO and compare to yours. Also check social signals.

My onpage and their onpage is pretty much the same. Hell, they even have exact match titles. Their social signals are non-existent (I have social signals). Even their small link profile is not from themed sites - they are from crap directories.

The only difference in this scenario is that they have around 10 links and I have a few hundred.

And...I just checked one of my other sites that went from #1 to 200+. It's now back up in the 30's, but the top 3 sites are the same scenario as I just mentioned. Same everything except links. They have around 30 non-relevant links and I have around 300 on that site (of which about half of my links are in the same niche).

So, what I'm getting from this is that social signals haven't matter in this case, nor has diversified anchor text or onpage optimization. I even have better domain authority on one of them. The second example the #1 site has better domain authority than me. So, that's a wash too.

Another difference is domain age. Their domains are older. Mine are about a year old.

I cannot pin this down and I have analyzed relentlessly.

 

[thetroglodyte]

OK, here's some more Penguin feedback that has me a bit frustrated. I have an EMD with a few hundred links that was evidently penalized. Two of the top three sites that are now ranking are not EMD's (one of them is) and have less than 10 links for that keyword phrase. The interesting thing is that ALL 10 LINKS ARE EXACT MATCH. They have no anchor diversity. It's not a temporary thing either. They appeared right after April 24th and have stayed there.

What's up with that?

Google works on percentages but there has to be some numbers there too. 10 backlinks wouldn't be enough to trigger a penalty under any circumstances except maybe if the domains they came from were deindexed. Also, keep in mind there are backlinks there Google sees which you can't.

 

[Chicilikit]

Anchor diversity matter, if you have hundreds of links, but of course, when you have ten, 20 or so backlinks than not. I bought some services here few days before penguin with matching keywords and because I have only few links from them and those sites are new with no other links, those sites ranks great now, but I'm sure if I added hundred more they would hit me.

 

[txholdem]

my conclusion is just to forget about it..
so i have read a lot about it here at BHW.
some say reduce keyword density, which i did - in vain.
some say dilute anchor, which i did - in vain (took me 2 nights to SB "click here go there" lol).
some say 301 redirect which i have not tried because i wonder how long that can last - a brand new domain rank high? and penalty passing along?
some say social signals - which is inconclusive because i seem to have hard time getting +1 for some reason. other social signals so far have done nothing (after almost 1 week).

some say what penguin is is that google does not want to share money anymore. they want adwords!

i tried a couple keywords - almost NONE of the 1st page sites have ads AT ALL. forget about above or below the fold!
so what am i trying now? i took the damn ads out! they are not making me shit anyway. and i blasted SB using top sites as my anchor texts. WTF! like a Chinese expression: "you can do no wrong when you are trying to cure a dead horse!"

this is my last attempt on recovering my PR-5 site. so far the only benefit i see from PR-5 is new posting get indexed fast. Hell I even remove the site from my google analytics like some suggested.

btw, i also see #7 site in my niche - UNDER CONSTRUCTION!

 

[manny521]

These stories are so frequent its hard to know what to do post penguin. One person sees one thing, another person sees something else. Penguin is frustrating...

 

[HoNeYBiRD]

more fun yet to come...

 

[EvilPlankton]

Seems like the only thing consistent about Penguin is that it's inconsistent. I've got a handful of sites that I'm trying different techniques on and I have found no reliable indicators yet.

 

[zartania]

Seems like the only thing consistent about Penguin is that it's inconsistent. I've got a handful of sites that I'm trying different techniques on and I have found no reliable indicators yet.

Certainly, there have been some decent things published like the study from microsites that show data that says link diversification, links from the same niche, etc. However, my testing has shown absolutely nothing concrete. It's all over the board. The only thing I haven't done is a new site from scratch. That starts tonight just to see how I fare.

I've read a lot on this forum and others about what people think works and what doesn't. I don't believe that ANYONE has anything that we can take to the bank yet. If you do, be a saint and please share!

 

[SuperLinks]

Does the site have redirects?

If not, you have HUNDREDS of links, this site has 10. The chances of any type of statistical anomalies is rare at this point in the game. You on the other hand have hundreds of links, and what about your keyword diversity?

Ten exact match anchor texts are nothing crazy, 100s+ could be in your niche.

 

[caspka]

blast them and see what happens.

 

[zartania]

My keywords are diverse and so are my anchors.

 

[zartania]

Also, no redirects. 100's of links were quite the norm when I started the site about a year ago. It took that to beat all the other guys like me.

Perhaps in a normal world that niche wouldn't have many inbound links. If that's the case then every other niche site I own won't rebound either.

If this speculation turns out to be cold hard truth then no amount of "balancing" out a link profile with keyword diversity will work (unless of course your niche naturally has hundreds or thousands of links). Time would be better spent starting over with a fresh site.

 

[zartania]

Does the site have redirects?

If not, you have HUNDREDS of links, this site has 10. The chances of any type of statistical anomalies is rare at this point in the game. You on the other hand have hundreds of links, and what about your keyword diversity?

Ten exact match anchor texts are nothing crazy, 100s+ could be in your niche.

Wow, I just figured out how to quote using Tapatalk. Lol

Your post really has me wondering if perhaps the Penguin update really isn't an algorithm update in the truest sense, but rather a ban on any site that has done link building like we did in the days of old (i.e. last month). They've either deindexed them or given big penalties to push us way down in the SERPS.

I'm starting to believe that Gwaggle did a major Spring cleaning and they never want those sites to rise again as long as they have something "natural" to display in front of them.

 

[zartania]

blast them and see what happens.

I'm starting to believe that blasting won't help anymore. Time for some testing on a few throw away domains...

 

[scalehard]

I agree with one thing that is Google probably doesn't want to share the wealth anymore. Penguin is nothing but a highly controlled algorithm pushing people to adwords and favoring corporate owned websites. if you get booted from adwords as well its back to a McJob. Their entire "do no evil" slogan is complete bullshit. They are hurting people's finances and they know it!

 

[scalehard]

Has anybody read G's blog regarding the 50+ updates they made kinda sick the nicknames they are giving to some of the penguin updates like its some game.

Penguin Update List: http://insidesearch.blogspot.com/2012/05/search-quality-highlights-53-changes.html

 

[lagger]

Wow, I just figured out how to quote using Tapatalk. Lol

Your post really has me wondering if perhaps the Penguin update really isn't an algorithm update in the truest sense, but rather a ban on any site that has done link building like we did in the days of old (i.e. last month). They've either deindexed them or given big penalties to push us way down in the SERPS.

I'm starting to believe that Gwaggle did a major Spring cleaning and they never want those sites to rise again as long as they have something "natural" to display in front of them.

This theory actually makes sense. If they detected "unnatural link building" through out the sites life then all of sudden they roll out a algo that slaps a lot of sites in the index. This would make sense why no one can replicate a penguin strike on new sites. It would explain the randomness. This unexpected strike on the SERPS would mean negative seo is mostly ineffective, unless the roll out another penguin strike. This actually makes a lot of sense, because the algo behind the bmr and spam networks seemed a bit more manual/specific.

This is kinda action reminds me of the saying"kill them all and let god sort them out"

 

[caitlin]

I am thinking Penguin is only half of the algo that killed all these sites. I know I read, or heard, somewhere that Google had this algorithm that basically looked for a huge similarity between the content on your money page and the page that had the link to your money page. If the content was similar, the page that was linked to got a flag. The only purpose of this bot was to seek out these pages that were similar. Obviously it was looking for spun articles. When it found them, it marked what they were linked to for Penguin.

Then they deployed Penguin to be the executioner. I think the reason some sites survived and others were killed simply is because bot #1 didn't finish its crawling of everything. A few sites slipped through the cracks.

So I really doubt it is about the links, I believe it is all about the content around the links that are pointing to your money site. That was the indicator.

I think we assume it is the links because for the most part, in a spun article we are still using the same anchor links so we blame it on that.. But I honestly do not think they are the cause, they are just the links to the relationships between your money page and the page with the link.

Think about everything that went on with the blog networks.. why some of them were removed from the index... Lots of similar content pointing to similar content. Some of us, like me, basically had a lot of apples pointing to apples. The sites that seem to have made it through unscathed had oranges pointing to apples.

Hell, for all we know one of the aspects of Panda could be to search and flag sites as well as rate the on-site content for some sort of numerical value. If the page has no links out, then the page gets a zero, if Panda crawls a page with a link and follows the link and the content on the landing page is vastly similar the page it came from- get a 1... Well if you have 100 links pointing to your money page, each time you get visited as a result of that bot following a link.. it's pretty easy to rack up a "score" each time you get a 1 added to the tally from a visit... Expressed like this 0=45 (zero links out, 45 from similar content in)

Send out the penguin bot to kill every site it finds with an accumulated score over a set number... Could be 5, could be 10... That is the unknown. It could look for any site rated 0<20 or higher and give you some sort of penalty.. 5 places in results for every point over 20...

Anyway, just a theory and my 2 cents but it does seem to be easy to deploy something like that and it is possibly a reasonable explanation for a lot of things. In this theory it would be fairly simple to isolate a certain segment of link types and issue penalties for only those link types.

I am sure it is more complex than that, different columns for different types of links... but the principal could still be the same.