Testing SEO with a controlled experiment

Folks with a background in science should be familiar with experimental design. If you want to test something, a basic experiment divides subjects into 2 groups: a control and experimental group. The groups should be identical except that the experimental group has been subjected to some outside factor. Differences that arise between the two groups can be attributed to the outside factor.

I have applied a very rough experimental design to my SEO efforts on a new site and have been very surprised by the findings.

I launched 2 brand new sites at about the same time in 2/08:
1. Atlantaprice.com: This is “control” site with similar content to the page I was optimizing for:
2. localprice.com/atlanta/homesecurity.html. I have been optimizing for the search terms: “Atlanta home security”

For the past 2 months, I have engaged in a full fledged SEO effort and have built 500 inlinks according to Webmaster Tools and 100 of which are directed to this specific page. Not all are a high quality, but there are many with PR in the 2-5 range. I published 6 articles, did a press release, listed the site in most major directories, optimized the site content, purchased some links, etc. Essentially, I did everything that all the SEO sites and my SEO consultant recommended.

Admittedly, the techniques worked well for Yahoo results. But, after all this, my control site is well ahead of my experimental site in Google rankings. The site with no inlinks and no SEO effort ranks at about 100 and the site that got all the SEO attention is at about 200 on Google. Can someone explain this?

Wow. Very interesting. It is possible that you are experiencing a Google sandbox effect, because you did too much SEO in a short period of time. It could be the quality of the links or maybe purchasing links.

On the other hand, it might be domains that are making the difference.

KIDS....nuff said

Could be the sandbox, but it would surprise me if the sandbox was so severe that I would have been better off doing nothing at all rather than all the SEO efforts I put into this.

I suspect that you may be right about the domain name. In my experience, I have noticed that having the city name in the URL seems to have a very substantial impact. But this is rarely discussed in SEO forums.

Update: The control site is still ahead.