“Something” is Wrong with Google (since 2004)

SMS Text
“Something” is Wrong with Google (since 2004)

Concentrated power has always been the enemy of liberty – Ronald Reagan

It never ceases to amaze me how Google has become such an integral part of our lives. Search has evolved in such colossal proportions and especially with the advent of Google instant my belief that Google can actually read my mind has only been fortified. Search Marketers around the world are always on their Toes and as Google states “Don’t try to follow our algorithm but try to think about the direction in which we are heading, build content for the users and not the search engines”. I sometimes wonder if Google was actually a person he would be an amazing election candidate.

But then again the high and mighty also have their share of secrets, secrets that should never come up, what I am about to reveal can be touted as one of the many chapters of the Da Vinci of Google.

Disclaimer – I am NOT one of those ANTI Google Freaks. As part of my research I came across this unique search result which I thought should be shared with the entire SEO community. I am also not claiming that the website is involved in any black hat SEO techniques. Over the years the link profile of this website has evolved in such a way that the website has been ranking despite going against some of the most conventional SEO rules.

Having made my point very clear let’s get started. Make a search for “something” and check the result that is ranking at #1.

It looks like a normal search result and at first glance everything seems just fine, but this search result is going to break (or possibly make) some of the biggest SEO myths.

Before we dig into this website, let’s have a look at the statistics of the search term “something” –

One would wonder, why the hell would someone search for something like well… “something”. The answer is simple, English is a funny language and there are millions of people who are trying to learn this language and would probably want to know the meaning of the word “something”. There might also be a case where people are searching for Beatles song “something”, in simple words there can be “n” number of reasons why people search for such a term. Also it’s needless to say that “something” happens to be a very competitive keyword with very high search volume even in exact match.

As soon as I saw this website ranking, I decided to do an analysis of this website and identify what is really working for this website, also the fact that it is ranking above the almighty WIKIPEDIA all the more increased my anxiety. As I dug more and more into the data I realized that this website was somehow able to defy all SEO laws and believe it or not FOOL Google. Yup you heard me right fool Google. One by one I will be enlisting those laws and then highlight how the website was doing the exact opposite of what ideally should be done.

Law 1 – (Relevant) Content Is King

If content is “King” then the kingdom which this content is ruling will definitely die in poverty. It is a well-known fact that relevant content plays a very important role as far as rankings are concerned. The website under consideration has exactly one instance of “something” and nothing else. Surely this does not qualify as good content.

Forget good content, I am pretty sure that this content cannot be classified as relevant content as well.

On a lighter note I believe spammers will love this version of Google’s relevant content

Law 2 – The URL Should Be Optimized

There has been a lot of debate on this topic where clearly there are two stands. Some experts believe that including keywords in the URL does help whereas others believe that relevant content and a good back link profile should outdo the advantage that websites with exact match domain have.

Nevertheless in our case, apparently the exact match domain is helping the website.

Hmmm “something” to think about.

Law 3 – The Website Should Be Spider Friendly

I will definitely not argue with the fact that this website is spider friendly, perhaps too friendly?

I had my initial doubts whether the website was into cloaking where content was written only for the search engine spiders. Alas I was so disappointed. Not only are they not doing cloaking but they actually have nothing in the source code.

Except the search engine spiders everyone is disappointed.

Law 4 – The SEO Elements Like Titles, Metas Etc Should Be Well Optimized

This is probably the first chapter in every SEO book out there. The role of on page SEO elements like Page Title, Header tags, Alt tags etc cannot be stressed enough.

However, except the Page Title the website does not have any of the above mentioned elements.

Again “something” to think about.

Law 5 – The Website Should Have A Good Back Link Profile

Analyzing the backlinks data for this website was like opening a Pandora’s Box. It was like being amidst corrupt officials of a Government who took pride in the fact that they were the ruling party despite their horrible background and one could do nothing about it. Sad but true story.

For the sake of simplicity and to have a broader perspective I classified the backlinks into various categories and then segmented them further so that one can view the bigger picture. I classified them on the basis of Page Rank (Google Tool Bar), Type of links, Website Category and Linking URL Status

External Link Analysis

Page Rank

 

I used Open Site Explorer to extract the data for back links.

As per the data, the website has a total of around 18647 backlinks pointing to it which includes around 970 root domains.

Page Rank wise data for the root domains is as follows –

Type of Link

I still remember those days when spammers used to make life of bloggers a living hell with their constant blog comment spamming.

Hell it became so popular that people started offering it as a service. I mean can you believe it? Anyways the point that I want to highlight here is that no follow tag was introduced to help webmasters to in a way control their link juice.

However with the advent of no follow tag also came the argument of whether no follow links help SEOs in any way or not. The SEO community was again divided into two parts where one side said that as long as links are coming from relevant websites it shouldn’t matter whether the link has a no follow tag or not and the other side stated that no follow links in simple words are not followed by search engine spiders and hence are of no use.

As far as the back link profile of this website is concerned, around 5,485 links out of 18,647 links have no follow tags. Did it help the website? Not sure if anyone can answer this question with utmost accuracy.

Error Pages –

According to Matt Cutts, showing a 404 message is one of the ways of telling Google to remove the respective pages from Google’s index. As per the data shown by open site explorer all the URLs for which external links have been built, except the home page, all others return a 404 error.

From an SEO perspective getting links built for pages that return a 404 error is not at all recommended. In my years of experience I have also noticed that whenever links are built for 404 pages they seem to have a negative impact on pages that return 200 ok as well.

However in case of our website, apparently the home page is unaffected despite the many links built for pages that return a 404 error.

Country Wise Distribution of Links

On analyzing the links from a GEO perspective, something just clicked me; I decided to check the result for the same keyword on Google.co.uk, Google.ca, Google.com.au and Google.de. Below are the ranking statistics for the keyword “Something” on these search engines –

As you can see, except for Germany, in all the other search engines the website is ranking at #1. Also, the website does not rank well in countries like Japan, Russia etc. These are the same geo locations from which there are no external links pointing to our website. A clear co relation between Geo Specific links and performance in Geo specific Local search engines.

Also the links can be broadly classified into three types –

  • Blog Comments
  • Forum Links
  • Others (content links, links from scraping websites etc)

As one can see, majority of the links are in the form of blog comments and forum links most of which are no follow links.

Anchor Text

Ok before you guys think that I have gone bonkers, please believe me when I say that I am perfectly fine. The Tag cloud that I have shown above consists of keywords that have been used as Anchor Text to point to our website. It is so obvious that except the home page URL none of the other anchor text makes any sense.

But wait, I thought anchor text plays an important role in helping the performance of the search terms? or does it?

Social Statistics

For http://www.something.com

 

For http://something.com

As per the data from SEOMOZ, it is pretty evident that the website (for some God Forsaken reason) has been performing well on the social front as well.

Some Interesting Links

Google has always been very clear about links from link farms i.e. in simple words they hate it. Somehow the website always got exempted from this. For example consider the link


If there was ever a link farm, this has to be by far the best and the easiest link farm to detect. On examining the source code you will realize that the page has only links and nothing else.

Number of Pages Indexed and Doorway Pages –

It is a well-known fact that the number of pages indexed by Google is also a crucial factor in helping the performance of any SEO campaign. When you look at the home page it is evident that there is absolutely no content or links on the page however on checking the site: for the website the results were pretty surprising –

I remember getting blacked out for almost 2 seconds, soon after I gained my senses I realized something was horribly wrong.

I took the courage to actually go and have a look at these webpages, I have provided snapshots below to some of the pages –

Crazy freaking pictures aren’t they? Nevertheless apart from the images what caught my attention was the fact that these pages had no links or if at all they had links they were pointing to third party websites.

Hmmm now what do we call such pages? I am sure that such pages are called Doorway pages, but wait aren’t doorway pages bad from an SEO perspective?

On another note – On checking the internet archives, back to the year 2000, something.com was actually an internet service provider  and their new website is http://westnet.com/.  Being an ISP might explain why there are images hosted with no links from the main page. The whois entry shows that it was created in 1995 and expires in 2015, so this site has a lot of domain age to it. But then again why would Google not filter this website out, just because it has domain age does not mean that it should rank.

Traffic Stats –

Of course I do not have the GA access to the website, in fact the website does not have Google Analytics tracking code, but still we can get some insights from Compete.com –

As per Rand Fishkin’s latest post one should not rely on traffic statistics by Compete.com, unfortunately for us we have no other data to work with.

Ill assume that the reported traffic is 30% of the actual traffic, hence in that way the website actually receives on an average at least 15,000 unique visitors.

Now that’s just bad, imagine the user’s plight when they actually visit the first page that is ranking on Google to be an empty page.

Now for the Ultimate Test – Google v/s Yahoo v/s Bing –

I wanted to see the performance of Google’s arch rivals Bing and Yahoo and below are the snapshots of the results –

If you ask me for a feedback I would surely rate Bings results above Yahoo and Google.

In Bing the first result is Wikipedia and second is Dictionary. These search results make a lot more sense for a highly generic keyword like “something”.

What Next?

As I stated above I am not running an Anti-Google campaign nor am I preaching Black Hat SEO techniques. It’s the global picture that I am trying to highlight. Despite so many things going wrong for the website, it’s just not possible that everything is wrong. There are certain lessons that one can learn from this or at least I surely did.

  • Exact Match Domain Name Matters

One cannot be oblivious to the fact that one of the main reasons the domain had an edge over other domains was the presence of the exact match keyword in the URL

  • Number of Pages Indexed Matters

As and when more pages get added to the search engine index, it’s a signal that fresh content is added to the website on a regular basis and hence beneficial from an SEO perspective

  • Geo is Very Much Important

In order to perform well in local search engines it is important to get localized links and should always be preferred over non localized links.

  • Social Matters a Lot

I am probably the trillionth person on this planet to state this but yes Social presence Matters

  • Text to Code Ratio

Now this inference was not drawn from the ranking result above but something that I have noticed over the years. IpullRank wrote a brilliant post on how search engine spiders can actually see flash content and can also crawl through JavaScript content. The bottom line of his post was that “Google and other search engines are SUPER INTELLIGENT”

All that is great but logically speaking search engine spiders will prefer to crawl pure text or more of textual content over JavaScript or Flash. Again I am not trying to argue or point fingers at anyone but am only putting forth my opinion.

  • Anchor Text Relevancy

Although Anchor Text is an important factor, as long as we are getting links from relevant websites, search engines are becoming smart enough to pass on the SEO value.

On a lighter note – I am not sure how many of you noticed but I consciously avoided writing the domain name throughout the post and there is a genuine reason for it. While analyzing the links I realized one common user behavior pattern of the internet users. So many times it has happened that while writing a comment for a blog post or a forum, the URL field has to be filled as it is a compulsory field. I asked myself if I am filling this field and if I don’t have a website what will I write and immediately the answer was somethingdotcom. I would like to highlight that at least 40% of backlinks for this website showed this tendency.

Lastly, many of you must be wondering how did I stumble upon the year 2004 or why did I include it in the title in the first place. While analyzing the backlink profile, I stumbled upon http://www.dslreports.com/forum/r9483427-Something.com

So I said to myself, if this link has been there since 2004 then obviously the website must be in existence since 2004 and probably even ranking. Funny isn’t it? Also if you ever decide to analyze the back link profile you will find lot of blog comments where people have actually stated that they have absolutely no clue why the website is ranking.

Hope this post was useful and could provide you with some insights. Would love to hear your thoughts.

 

Sajeet Nair

Sajeet Nair

Senior Project Manager at SMG Convonix
With over 5 years of work experience in digital marketing, Sajeet Nair is a Senior Project Manager at SMG Convonix, a Digital Marketing Firm based... Read Full Bio
Sajeet Nair
Sajeet Nair
Sajeet Nair

Latest posts by Sajeet Nair (see all)

Subscribe to SEJ!
Get our weekly newsletter from SEJ's Founder Loren Baker about the latest news in the industry!