What’s the REAL Organic Search Market Share? Part II
May 15th, 2012 by
A few months ago, I wrote a blog post about the difference between comScore’s organic search market share numbers and what I’ve seen in Google Analytics accounts over the years. In the numbers I’ve seen, Google’s share of search was always much more dominant than the approximately 65% market share reported by comScore.
So, what does comScore say this time?
In the graph above, you will see that Google, Bing and Yahoo had 66%, 15% and 14% of the search market share respectively in February 2012 (read the official report here).
For a little background on the comScore numbers, you can consult my previous blog post or check out their official word here, but as I’ve said before, they seem to go to incredible lengths to get valid, representative data — unlike me, who just has a spreadsheet featuring data from 69 websites.
Last time, I speculated that the inflated organic search visits from Google that I observed could have been due to traffic from Search Influence, since everyone here uses Google and we’re always viewing SERPs relevant to our clients (and likely clicking without regard). Since then, we have installed filters on every Analytics account that blocks data collection from Search Influence’s static IP address. In spite of this, the results were pretty much the same.
We no longer have access to 4 of the 73 (a random number… you get tired after 73 exports) Analytics accounts that we had access to in September of 2011. In spite of that, we still have an approximately 450,000 visit sample, and the sites that we lost were a very small percentage of the visits from last time. Out of that many visits, in a wide variety of industries, Google has an 88% share of search.
Why? DOES ANYONE KNOW WHY!? I still don’t know. I do believe that the comScore numbers are probably accurate, but I can’t think of many reasons outside of coincidence that our numbers differ from theirs so much. Maybe Google under-reports referrals from other search engines? Maybe we optimize solely based on Google results and our clients rank best in Google, therefore we don’t get a representative share of visits from Bing and Yahoo? That would ostensibly seem to hold water, but there are very few differences in SEO for Bing and Google that I know of.
The data is from Feb 1, 2012 – March 2, 2012 (to match the number of days of the September data). As you can see below, directories, health and beauty and non-profit make up a significant portion of the industries represented. Even if you remove the directories data from the mix, though, the numbers are approximately the same.
So once again, does anyone have any ideas why this information doesn’t mesh with comScore reporting? I’d truly love to hear your feedback on this issue. If anyone has any data they would like to contribute, that would be great — but even better would be a tool to extract GA data on a massive scale so that we could have more representative information.
You can go here to see the data.