Penguin is not constantly running so when is the next update?

Expertpeon said:

I don't think this is "well documented" at all... I've watched thousands of sites over the last month... link anchors seem to have absolutely nothing to do with if a site keeps its gains or not.
People are just parroting what Matt Cutts has said, along with a bit of "fill in the blanks"

Click to expand...

Well, i've read loads of posts from Google and other popular forums. Here's a good post talking about it http://www.micrositemasters.com/blo...need-to-act-smarter-and-5-easy-ways-to-do-so/ (a thread was posted on BHW a few weeks back about this post)

I do exactly the same onsite SEO and link cycles for all of my sites. The only site of mine that was effected by penguin was the one that i was concentrating on 1, very competitive keyword. I'm running further tests to confirm this but at the moment i think it's very likely that anchor text diversity is a big part of the penguin update

Expertpeon said:

I have around 25 sites running... 1 was hit.
The sites that were not hit all had
1) Extremely high anchor similarities (usually only 2-3 keywords) and tons of spam links (usually d0follow blog comments)
2) Tons of spun paid blog posts

The site that was hit, had 200+ unique pages, tons of HQ links with 20-25 different anchors, and a PR of 5... major difference? That site was ranking really really well on a niche lionbridge probably manually checks for google.

Click to expand...

I very much doubt anchor text diversity is the only thing that changed and noone is saying it is. In any case though, if say the threshold was 40% anchor text diversity, your sites that were not hit would have been within that threshold as exact match anchors for 3 keywords would give you a 33.3% anchor text diversity. I can't see how this proves or disproves anything. I certainly don't think you can conclude that anchor text diversity is not a factor in penguin

Based on my findings Penguin works with tresholds. They do not give the same penalty to everything. They look at how badly you screwed up your link profile. They compare it with the trust/authority of the site and the number of searches for the keyword. If you are going for a low competition keyword chances are you will not hit the treshold and the filter does not get triggered even though the penalty is applied to your site. If you are going for tougher keywords the number of searches is like a magnifier.

I have been watching Panda closely and it works similarly. That is why everybody is saying it makes no sense. It does.

rugbyjack2005 said:

Well, i've read loads of posts from Google and other popular forums. Here's a good post talking about it http://www.micrositemasters.com/blo...need-to-act-smarter-and-5-easy-ways-to-do-so/ (a thread was posted on BHW a few weeks back about this post)

I do exactly the same onsite SEO and link cycles for all of my sites. The only site of mine that was effected by penguin was the one that i was concentrating on 1, very competitive keyword. I'm running further tests to confirm this but at the moment i think it's very likely that anchor text diversity is a big part of the penguin update

Click to expand...

rugbyjack2005 said:

I very much doubt anchor text diversity is the only thing that changed and noone is saying it is. In any case though, if say the threshold was 40% anchor text diversity, your sites that were not hit would have been within that threshold as exact match anchors for 3 keywords would give you a 33.3% anchor text diversity. I can't see how this proves or disproves anything. I certainly don't think you can conclude that anchor text diversity is not a factor in penguin

Click to expand...

navimeister said:

Based on my findings Penguin works with tresholds. They do not give the same penalty to everything. They look at how badly you screwed up your link profile. They compare it with the trust/authority of the site and the number of searches for the keyword. If you are going for a low competition keyword chances are you will not hit the treshold and the filter does not get triggered even though the penalty is applied to your site. If you are going for tougher keywords the number of searches is like a magnifier.

I have been watching Panda closely and it works similarly. That is why everybody is saying it makes no sense. It does.

Click to expand...

As the author of that MicrositeMasters article, I wholeheartedly agree with everything quoted above.

You cannot say "in this one example it didn't happen to me, therefore everything you are talking about is invalid". You need more data than that to be able to make conclusions like that.

All of this at best is conjecture. We can all agree on one thing: Google has turned seo on it's head.
Aside from that, we may all see different examples due to randomness or filters based on tons of factors. All kwp/niches/sites aren't equal, and as such can show different results.
I've studied this new algo change(s) for the past 30 days across 10+ big and small verticals, and one thing holds true by my account:
1) Almost each tested vertical has 1-2 sites that rank with very weak link BL profiles.
2) Aged sites/large sites don't hold alot of weight (aside from big brands) on their own merit.
3) Serps don't seem to bounce or change often- less googlebot crawling/updating of content.
4) New site indexing seems to be slower (Days instead of minutes in many cases)
5) Some sites with a high main kw anchor % still rank.
6) 2 way link exchanges seem to be working

My loosely educated guess is Goog is slowing down spam opportunities by slowing down the crawl.
You can blast away/301/hack or whatever you want, but if G slows the machine down, the effects are heavily diluted.

affarmy said:

All of this at best is conjecture. We can all agree on one thing: Google has turned seo on it's head.
Aside from that, we may all see different examples due to randomness or filters based on tons of factors. All kwp/niches/sites aren't equal, and as such can show different results.
I've studied this new algo change(s) for the past 30 days across 10+ big and small verticals, and one thing holds true by my account:
1) Almost each tested vertical has 1-2 sites that rank with very weak link BL profiles.
2) Aged sites/large sites don't hold alot of weight (aside from big brands) on their own merit.
3) Serps don't seem to bounce or change often- less googlebot crawling/updating of content.
4) New site indexing seems to be slower (Days instead of minutes in many cases)
5) Some sites with a high main kw anchor % still rank.
6) 2 way link exchanges seem to be working

Click to expand...

No doubt, for some reason link exchanges and link directories are working well.
However SERPs seem to change hourly across a variety of niches... google will display different results throughout the day for many niches I monitor.

I recovered about three weeks after removing duplicate content from my back linking websites (the one's that I could). There were 2 sites that had a lot of spun content backlinking and they have not recovered.

canadainsurance said:

I recovered about three weeks after removing duplicate content from my back linking websites (the one's that I could). There were 2 sites that had a lot of spun content backlinking and they have not recovered.

Click to expand...

So you had first page ranking before the 25/4 which dropped right down on the 25th right and that same domain last week jumped back up to where it was from simply removing duplicate content from pages linking to your money site?

I haven't had to concern myself with this just yet. However, this is a good summary of what i might need to be aware of when I finally get my sites set up the way I want.

rugbyjack2005 said:

So why would it be different for every site? If say there was a threshold of 30% anchor text diversity, it would be consistent for all sites. Algorithms are set and there are no considerations. You are either within the parameters or you are not. I would find it hard to believe that the changes are not consistent across all sites. I am sure Google will have very strict fair play policies and it will be the same for everyone and no randomness.

Click to expand...

Simple you could have 3% onsite keyword and 27% anchortext,that's why it'll be different for every site.So it would be the same for everyone if it was 30%.What better way to have people scratching their heads when certain things work for some and not others.Do you see anything unfair with that?

artificial_genius said:

As the author of that MicrositeMasters article, I wholeheartedly agree with everything quoted above.

You cannot say "in this one example it didn't happen to me, therefore everything you are talking about is invalid". You need more data than that to be able to make conclusions like that.

Click to expand...

I'm not using my own sites as sole guidance, but monitoring several hundred competitors at any given time. There has been little insight into the new algorithm, but one thing is clear, google is bouncing around things like crazy regularly.
Ultimately I don't see why this is good for search results, or their customer.

Highly spammed 301'd feeder sites (fresh domains) are outranking the best white hat efforts made by the SEOmoz crowd.
If you think de-optimization is going to lead to stable top ranks for competitive keywords, you're terribly wrong.

Google will fix Penguin, google cannot maintain its reputation and result quality with this update. Maybe at the end of this month we will see a new update.

Many of the site's, we worked are all recovered, when we create backlinks with main url and other anchor text to level the percentage.
Hopefully, we might seen another update soon, as the last update had hit many aged sites and rank blank websites.

linksmaster said:

Many of the site's, we worked are all recovered, when we create backlinks with main url and other anchor text to level the percentage.
Hopefully, we might seen another update soon, as the last update had hit many aged sites and rank blank websites.

Click to expand...

Do you think it was mainly anchor text related then?

Expertpeon said:

I'm not using my own sites as sole guidance, but monitoring several hundred competitors at any given time. There has been little insight into the new algorithm, but one thing is clear, google is bouncing around things like crazy regularly.
Ultimately I don't see why this is good for search results, or their customer.

Highly spammed 301'd feeder sites (fresh domains) are outranking the best white hat efforts made by the SEOmoz crowd.
If you think de-optimization is going to lead to stable top ranks for competitive keywords, you're terribly wrong.

Click to expand...

Overoptimization was a very very large factor in the Penguin update. I'm not sure how you could look at any reasonably sized dataset and not come to that conclusion.

A site with links that
1. Have diverse anchor text
2. Come from niche sites
3. Come from sites with some sort of authority

were almost completely unaffected by Penguin. I am not comparing whitehat vs blackhat or BHW vs SEOMoz. If your links meet the above criteria you will be able to have stable top ranks for competitive keywords.

You can get those links through blackhat methods and you can get them through whitehat methods. But your backlink profile better start looking like that if you want your rankings to last. Even if 301'ing spam works now, eventually it will be overused and you'll see another animal related update and wonder where all your rankings are.

The focus for any SEOer who wants long term rankings should be creating backlinks that Google cannot accurately differentiate as being whitehat vs blackhat and would appear a lot on sites with a natural link profile. De-optimization fits this mold.

301'd spam does not fit that profile so I do not expect that it will have this ranking power a year from now or even several months from now.

Matt Cutts just announced via Twitter that they just ran the Penguin Algo an hour ago. https://twitter.com/mattcutts/status/206232437427154944