This latest Google update, which appeared at the end of April, 2012, seems to have created quite a lot of confusion. In what can only be described as a ‘haphazard’ update, SEO’s everywhere, are doing their best to come to grips with possible rule changes that may be applied. Due to the confusion, there is many contradicting conclusions in relation to these very rules, and no one seems to be able completely agree on any one particular set of factors.
Basically, SEO ‘experts’ are basing their opinions on minimal data, rumor, and hearsay. There simply doesn’t seem to be enough information available to form a definite conclusion. Having said this, I am going to present some of our data and conclusions in an effort to try and clear up some of the confusion. The conclusions presented, run parallel to and in conjunction with informed opinions of several other major SEO testers I work regularly with, and who also operate with large-scale data from tens of thousands of sites.
Link Profiles
All sites/pages that lost rankings from the Penguin update, had at least 50% of anchor texts using ‘exact match’ main-term phrases. There were also plenty of comparative sites that weren’t getting hit due to a high concentration of exact-match anchors, but his can tend to happen; they just haven’t been caught yet. This only goes to demonstrate how ineffective Google really is at detecting such data algorithmically. It can be a case of ‘chance’ as to whether you’ll be actually discovered. The over-riding factor here is, that hardly any negatively hit sites with a diverse anchor text profile, could be found. The more diverse the profile, the less likely it was to be hit by Penguin. In the case where a site’s page had less than 1/3 of its anchors from that page’s key term, there was no obvious ranking drops at all.
Taking all this into consideration, it is apparent that anchor text profile remains one of the key factors of this update. A lot of damage was cause during Panda 3.3 for the very same reason, however, the ‘tolerances’ have changed considerably. It could be that Google has turned up the ’sensitivity dial’ somewhat for this update, as many sites that escaped Panda 3.3 were also caught by this.
This is also most likely one of the many reasons that so many good-quality ‘white-hat’ sites are being hit. If you don’t have many links, and are are using precise/fixed anchor texts for your article/press release strategy, then you may find your percentages are going to fail. It is therefore proof that sticking to ‘white-hat’ methods offers no guarantee against the
It certainly goes to show how sticking to ‘white-hat’ methods is absolutely no guarantee against the unpredictability of Google’s changes.
Content
Making sense from Google’s perspective, we saw some evidence that ‘power-links’ from highly relevant ‘themed’ sites, now had slightly increased ranking effects. Due to the fact that anchor texts are so easily manipulated, it is a possibility that Google has widened the gap in ranking-power between links from non-relevant sites, compared to links from tightly-themed sites. By trying to classify a site into a theme, Google is attempting to assign a different ranking value based on site topic.
I am NOT saying that ‘non-relevant’ links don’t work, or aren’t required, it just seems that Google has either turned down the ranking-effect of links from sites that it determines to be non-topic relevant, or has turned up the power of closely topic-themed links.
We certainly haven’t come across any evidence supporting there being a penalty or negative-ranking where links are achieved from non-relevant sources. This is possibly due to the fact that this would be impossible to implement, as it would virtually hit everyone. This would include the major corporate sites that Google only ever seem to pay any attention to, being their main AdWords revenue stream.
The bottom line here is, there just isn’t enough topic-relevant sites available to provide us with the IP spread, diversity, and volume required in our link-building. There is just no way to compete commercially with the quantity of links available from on-topic sites. These other links are needed to keep your glass full!
I firmly believe that content relevance is one of the biggest misconceptions to come out of the Penguin update. Matt Cutts’ example on the Google blog has lead some people to believe they can’t have links from non-relevant content. (Always bear in mind that they will take every possible chance to misinform SEO’s and create confusion). Out of hndreds of sites tat we are optimizing none of these have been affected by Panda, and only a few by Penguin. The few that were dropped with Penguin, were the very sites that we were pushing the envelope with straight after the Panda update, in order to investigate and test the threshold/limits. As soon as Penguin hit and ‘moved the bar’, they got slapped a bit. These are some of the indicators of the 50% threshold as discussed above.
Moving Forward
I could now argue that you may want to start looking for somewhat relevant topic sites as opposed to your more expensive link-building expenses from a quality private blog network, and sites that don’t have as wide an overall keyword/topic distribution in order for Google to find them easier and ‘classify’ them according to theme. I anticipate private networks popping up in the not to distant future, which are topically themed for a smaller quantity of power-links to be added into your mix.
This also presents the argument of point for themed manually-built asset-quality 2.0’s and buffer sites. This also implies, building your own mini private networks of web assets. Incorporating volumes of links into these, in addition to using a large variety of themed keywords to pass the juice onto your key pages, you will be able to make the most of all sides of this update. Ideally, you should utilize a different outbound anchor on every buffer site you build, so you don’t risk over-optimization with your key pages.
The ongoing message here is to keep coping with things that I recommended in my last update on dealing with Google penalties and ranking drops. Keep building, or start building, high diverse links, adding fresh content and start expanding your web estate incrementally. Do not contact Google. I won’t reiterate further, as my last few posts have already covered these areas in detail.
Penguin - A animal that lives in a Chilly Climate
It must be understood that Penguin is purely algorithmic in nature. There was no increase in GWT Notices, and Matt Cutts is instructing people not to file ‘Reconsideration Requests’. This means, there is no manual penalty that can be removed. It is a fundamental change to Google’s rank ordering algorithm. However, the good news is that any corrective changes you achieve should reflect in a short space of time in the results.
Some SEO’s are assuming that Penguin is all about content and not linking. If this is the case, then why are we seeing Amazon, and other big corporate sites ranking everywhere with a great deal of obscure exact-match domains featuring no content? This is due to the fact that these big sites don’t build back-links like IM’ers do. So, what little linking they do have, is very natural and their ranking comes mainly from their overall authority and on-page content, page titles and headers.
As the obscure sites we’ve seen have either no or very little SEO, they haven’t been hit by any of the algorithm changes. This means, because they’re doing nothing ‘wrong’, they don’t suffer any of the manual penalties either. They’re now ‘left’ there at the top because most of the sites that have been building back-links over the years, have been doing so in accordance with the old mas keyword-linking rules. These rules have been misunderstood for a long time, so their link-profiles are usually terrible. It can often be a case of ‘last man standing’.
Another example is, why are so many 2.0’s (and we have thousands of these), out-ranking ‘proper’ sites at the moment? A lot of these are constructed on reasonable article-quality content, however, are small in comparison to a proper site, and the link profiles in their thousands, are predominantly from off-topic content and linking.
They do, however, have a large varied inbound anchor text profiles, so they pass the Google algorithmic filters and subsequently end up out-ranking the larger, and better sites simply due to the fact that they have the right link-profile and are sitting on high authority domains.
A Degree from The University Of Diversity
Diversity is usually the key, as it creates a natural looking randomness, together with the disorder that is required. We always aim for 100% variety in our link-profile, even though we never achieve it, however, having this goal means you will never end up with large percentages of the same anchors. We’ve all heard people talk about their ‘diverse’ anchor profiles, and how they use highly varied terms. The truth is, they’ve usually used 5 different terms for the page, with most of the links to 1 or 2 of those terms, and perhaps added a few ‘click here’ generic terms in there somewhere, and maybe even a few URL/brand anchors if they’re really stretching it.
When I am discussing anchor diversity, I mean starting with 3 to 5 keywords, and dressing that up to 10 to 15 good LSI/long-tail terms. Each of those is then multiplied by 100 by adding 50 varying prefix modifiers, and 50 different suffix modifiers. We now have 1000 to 1500 different anchors with a central ‘keyword mass’ themed towards our required terms. I then add 100 to 200 various generic terms, as well as the bare URLS, and company/brand names. That all mixes together and used randomly. That is the essence of a mixed anchor profile. I will stress here that when we build links at Local Site Submit we are mixing anchor text and using a wide range of sources to get links.
Final Conclusions
A very worrying factor of this update, is that it renders negative SEO even easier. As Google appears to be working with narrower tolerances on anchor-text profiles, it makes it very easy for an unethical person to blast your site with main-term anchors from content-less sources, which will push your percentages into the red. This is one of the main areas I am hoping that Google will address in the very near future.
The biggest fall out from this entire update, is a reminder as to why everyone really needs to diversify their sales channels to ensure their income is protected. Never before, has it been more imperative to build and maintain volumes of micro-sites, 2.0’s, and buffer pages. Channeling the majority of your link-building through these, you will greatly reduce the chances of getting negative feedback, reduced direct-linking, (which is where all penalties arise from), and protect yourself by virtue of sheer scale.
As we try and tweak our programs to adjust to this algo change I cant help to think how easy it would be to crank up SENuke and hit our competitors to get them a penalty and move them off the first page. Of course I would never do this but this does not mean that someone else might. I hope Google realizes that this is not a very good update and has allot of potential issues. I would love to say “If that’s you how you want to play Google then I will just go play with Bing instead”…..but I can only do this if this is where the masses go. So time will tell if people start to leave based on a poor user experience with Google.
and more……………….
Todd Bryson