4:47 PM
Unknown
If you are familiar at all with search engine optimization, you probably know that the last couple of years have brought some major changes to the industry. Google has not only accelerated their updates to their algorithms, but they are also making bigger splashes with them as well. This recent, as of yet unknown, algorithm update on December 17th, 2013 is no different.
As the head of the SEO department here at Foremost Media I often find myself seeing these changes the very day they happen across the data we collect from our many clients' rankings. For instance, many SEO's out there keep repeating the mantra that "nobody noticed" when the new Google Humingbird algorithm was released a month before it was announced in September (It was implemented on August 20th). Well, they do not speak for Foremost Media. I can tell you at the end of August we were all noticing MAJOR rank volatility and were wondering what was going on. Now, I can tell you that another major update just occurred on December 17th. This time, I figured I better get out ahead of it before the internet starts buzzing with SEO's claiming that nobody noticed.
Whenever I suspect an update is occurring I check Mozcast. Mozcast does a great job at telling us when something is going on. While you can't tell WHAT is going on specifically, you can tell when volatility is up. Well, it was up very heavily on December 17th. In addition, on the same day, Foremost Media watched as rankings across a half dozen clients dropped for a couple select terms each. We aren't talking your standard two or three position drops either. We are talking about two or three pages. So, what happened? We started digging.
Mozcast showed that on December 17th, this update caused exact match domain influence to drop a little, but heavily increased both partial match domain influence and diversity. The first thing we noticed was that for about a dozen clients or so, we were seeing major rank declines for just a couple of keywords, usually isolated by topic. It wasn't all bad news. There was a little upswing, and in the case of one client their rankings skyrocketed. So, the most obvious question from looking at this data was, "Were the terms that dropped, terms that were ranking do to an exact match URL? The answer? No. There was no correlation. Some were EMD, but some were also PMD or no match at all. The interesting thing was that every drop was for an interior page of a site, whereas the client who got a big boost got that boost on the home page only. Coincidence? It could be. I don't have enough data to tell for sure. Let's look further...
Naturally, there are a lot of factors here. However, If EMD became extremely volatile, that usual means (simplistically) on some level they are either increasing/decreasing the effectiveness of EMD/PMD, or the end result of their change was EMD volatility. Since it doesn't appear that I can substantiate the former, let's look at the latter. What could cause EMD volatility? Well, while the most obvious cause is usually an alteration of EMD/PMD influence over rankings, it could be that the net result of whatever the change was is simply something that causes diversity in domain names and weakens EMD. At this point, I am on a wild goose chase because there are many things that could correlate with EMD domain names. I am inclined to think though that Google is still after spammers. So, I will start to look in that direction.
So, naturally I took to looking at the link profiles of every page that dropped. I am pleased to say that their profiles were all very weak. Why am I pleased? Because I am hoping this indicates that increasing the strength of the profile will help bring the rankings back to normal. On the other hand, this opens another Pandora's box. Since there was not direct correlation with any of the pages that dropped regarding partial or exact match domains, but the profiles were very weak, either Google is analyzing the page and matching it with the link profile in a new way, or this is all about the link profile. I think it is the latter, and here's why...
At Foremost Media we use a number of very sophisticated tools that helps us with on-page optimization. We have a very high success rate with these tools and in every case, my tools are telling me that the pages are still very well optimized. So, I am inclined to say that this update has something to do with back link authority and how it is calculated. It may have something to do with "Hummingbird", mainly because even with the declines in rankings traffic did not decline as a result (though the holidays made that harder to determine). This might indicate that while the term we were tracking decreased, they gained equally in another term we were not tracking.
Of course, I could be completely off base. Algorithm changes don't occur in a vacuum and the "butterfly effect" exists heavily in the sea of the web. There are so many factors involved that it is very difficult to guess with any accuracy what the algorithm update actually was unless we are told. Chances are, we will know in a month or so. In the mean time, I will keep crunching data because I enjoy it. I also didn't see anyone else actually trying to tackle the issue at this point with any meaningful data. So take all this with a grain of salt, and please let me know if you have anything to share regarding this update (Nfitzgerald@foremostmedia.com). If you're an SEO and you are reading this, know that the sky is not falling and SEO is not "dead". This isn't the first time a major update has caused some havoc and it won't be the last. Rankings will return and we will will continue to learn how to please the almighty Google for our clients' sakes.
Update: We ran an experiment by changing some links that we had control over to a couple of pages on some affected sites, for the exact match terms. We found that by either eliminating the link with the exact match or changing the destination of the link to the home page of the site, instead of the page of the site, almost completely reversed the rank drop. In this case, it almost appears that this could be some sort of Penguin refresh or update with the way it is behaving.
As the head of the SEO department here at Foremost Media I often find myself seeing these changes the very day they happen across the data we collect from our many clients' rankings. For instance, many SEO's out there keep repeating the mantra that "nobody noticed" when the new Google Humingbird algorithm was released a month before it was announced in September (It was implemented on August 20th). Well, they do not speak for Foremost Media. I can tell you at the end of August we were all noticing MAJOR rank volatility and were wondering what was going on. Now, I can tell you that another major update just occurred on December 17th. This time, I figured I better get out ahead of it before the internet starts buzzing with SEO's claiming that nobody noticed.
Whenever I suspect an update is occurring I check Mozcast. Mozcast does a great job at telling us when something is going on. While you can't tell WHAT is going on specifically, you can tell when volatility is up. Well, it was up very heavily on December 17th. In addition, on the same day, Foremost Media watched as rankings across a half dozen clients dropped for a couple select terms each. We aren't talking your standard two or three position drops either. We are talking about two or three pages. So, what happened? We started digging.
Mozcast showed that on December 17th, this update caused exact match domain influence to drop a little, but heavily increased both partial match domain influence and diversity. The first thing we noticed was that for about a dozen clients or so, we were seeing major rank declines for just a couple of keywords, usually isolated by topic. It wasn't all bad news. There was a little upswing, and in the case of one client their rankings skyrocketed. So, the most obvious question from looking at this data was, "Were the terms that dropped, terms that were ranking do to an exact match URL? The answer? No. There was no correlation. Some were EMD, but some were also PMD or no match at all. The interesting thing was that every drop was for an interior page of a site, whereas the client who got a big boost got that boost on the home page only. Coincidence? It could be. I don't have enough data to tell for sure. Let's look further...
Naturally, there are a lot of factors here. However, If EMD became extremely volatile, that usual means (simplistically) on some level they are either increasing/decreasing the effectiveness of EMD/PMD, or the end result of their change was EMD volatility. Since it doesn't appear that I can substantiate the former, let's look at the latter. What could cause EMD volatility? Well, while the most obvious cause is usually an alteration of EMD/PMD influence over rankings, it could be that the net result of whatever the change was is simply something that causes diversity in domain names and weakens EMD. At this point, I am on a wild goose chase because there are many things that could correlate with EMD domain names. I am inclined to think though that Google is still after spammers. So, I will start to look in that direction.
So, naturally I took to looking at the link profiles of every page that dropped. I am pleased to say that their profiles were all very weak. Why am I pleased? Because I am hoping this indicates that increasing the strength of the profile will help bring the rankings back to normal. On the other hand, this opens another Pandora's box. Since there was not direct correlation with any of the pages that dropped regarding partial or exact match domains, but the profiles were very weak, either Google is analyzing the page and matching it with the link profile in a new way, or this is all about the link profile. I think it is the latter, and here's why...
At Foremost Media we use a number of very sophisticated tools that helps us with on-page optimization. We have a very high success rate with these tools and in every case, my tools are telling me that the pages are still very well optimized. So, I am inclined to say that this update has something to do with back link authority and how it is calculated. It may have something to do with "Hummingbird", mainly because even with the declines in rankings traffic did not decline as a result (though the holidays made that harder to determine). This might indicate that while the term we were tracking decreased, they gained equally in another term we were not tracking.
Of course, I could be completely off base. Algorithm changes don't occur in a vacuum and the "butterfly effect" exists heavily in the sea of the web. There are so many factors involved that it is very difficult to guess with any accuracy what the algorithm update actually was unless we are told. Chances are, we will know in a month or so. In the mean time, I will keep crunching data because I enjoy it. I also didn't see anyone else actually trying to tackle the issue at this point with any meaningful data. So take all this with a grain of salt, and please let me know if you have anything to share regarding this update (Nfitzgerald@foremostmedia.com). If you're an SEO and you are reading this, know that the sky is not falling and SEO is not "dead". This isn't the first time a major update has caused some havoc and it won't be the last. Rankings will return and we will will continue to learn how to please the almighty Google for our clients' sakes.
Update: We ran an experiment by changing some links that we had control over to a couple of pages on some affected sites, for the exact match terms. We found that by either eliminating the link with the exact match or changing the destination of the link to the home page of the site, instead of the page of the site, almost completely reversed the rank drop. In this case, it almost appears that this could be some sort of Penguin refresh or update with the way it is behaving.
