So let me give you a bit of background behind the whole idea, I’m very technically minded and because of this I always want to know proven steps or facts behind things. I have been involved in search marketing for the last 5 years and in that time I have never been able to find research, which shows how certain SEO factors are influencing on rankings.
While there are several sources such as SEOMoz, SEOBook and thousands of other blogs, I have never found a resources that has setup lab conditions to test SEO ranking factors. So what I would like to do is setup a test environment in which I can set out to test each factor on it’s own.
The methodology that I thought of following was the classical A/B split testing that is used frequently in the on-line marketing environment. I would adapt this methodology slightly as I won’t be using a baseline to compare to but comparing two websites to each other where the factor has been implemented differently.
Now to setup a lab for testing – this is where the tricky part comes into play. I thought of using two different newly registered domains that are hosted on the same servers with all the factors kept as similar as possible except for the one I would be testing.
Now one of the first Ranking Factors that I would like to test out is the difference between having an .html extension and having no extension. So to do this I’m going to do some research using the Adwords keyword tool to find an industry that isn’t too competitive and setup a list of 3 keywords to optimize the site for. These website will both have domain names that have nothing to do with the industry to exclude this as a ranking factors.
As far as content and copy is concerned I will use the same amount of pages with content that share similar SEO factors such as length, keyword density etc. I will be using the SEO ranking factors as set out by SEOMoz’s Ranking Factors and try to ensure that none of these impact on the specific factor that I’m currently testing.
So after the initial test I would like to test factors as listed below
1. .com domain name against country specific such as .co.uk
2. Keyword density
3. Keywords used in internal link structure
4. Using subdomains compared to directories
5. Text bolding and none bolding of keywords
6. Dynamic vs Static urls
7. Click through rate of results on search engine result pages
This list will can be updated in the future but for now this is what I though of starting off with.
Now the next question is how will I judge the results? In the SEO industry it is common practice to look at the traffic generate determine the success of an SEO campaign. This is however not what I’m going to use. We want to determine if a certain factor has improved the rankings of a website. For this reason the ranking of each site on several major search engines will be the aspects that will be looked at.
Thus rankings will be monitored through out a period of 2 months. Now I know this is a shorter period of time than SEO’s would usually advise their clients to wait for results but because we are only testing single factors, I feel this should be sufficient time. If you however you disagree with anything, please leave a comment below as I would like to improve this research wherever I can.
Now as I mentioned before please leave comments to help me to improve this research. I would love to hear from anybody, which can help me improve this. Unfortunately I can’t release any specific info about the domain, industry or content as readers could scew the results. I will however release them when the research is concluded.
Please also share this with your friends and collogues to get more people involved. I really believe in the saying – two minds are better than one and would love as much input as possible.
Well now I need to get cracking the first test scenario so let me know what your thoughts are. Have an awesome day!