Follow-up Data on Org vs Com vs Net

I wanted to release a few more stats regarding the comparison of Google rankings for .org, .com, and .net

Unfortunately, I do have a sneaky suspicion that over the last day there has been some fluctuation due to a few factors.

1. Additional noise as this story has spread across the web
2. Google’s unintential indexing of the actual domains and not simply the experimental domains
3. Potential “mention” bias from listings on newly registered domain sites (no links, however)

Nevertheless, here are some of the raw numbers coming out of the earliest of reporting. These only include the previously revealed experimental subdomains, and not the other domains / subdomains we are running the test now. We will not reveal those domains until we have completed all experimentation, so as not to run the risk of tainting data which I believe we may already have for these first 9 subdomains.

The two tables examine the subdomains with adjustments made by removing non-indexed subdomains. We checked 25+ datacenters in the process and, in some instances, one or more of the subdomains were not included in the index. There is no fair way to weight these equations, so they were removed. Unfortunately, this cuts down our available data set by 33%, so the statistical significance of these results, I am certain, is quite low.

As you will see, on average, the .orgs enjoyed the lowest rankings across all datacenters compared to  .net and .org. Once again, these results represent a small subsection of the total study and are not statistically significant.

URL	Raw	Adjusted	Unindexed
com1	3.14814814815	3.14814814815	0
com2	4	4	0
org1	3.92592592593	3.92592592593	0
org3	2.40740740741	2.40740740741	0
net2	6.51851851852	6.51851851852	0
net3	1.51851851852	1.51851851852	0

Extension	Raw	Adjusted
ORG	2.11111111111	2.11111111111
COM	2.38271604938	2.38271604938
NET	2.67901234568	2.67901234568

No tags for this post.