New data from ComScore shows that Google users run more searches than Yahoo and Microsoft users… combined. According to the research, Google users conduct an average of 54.5 searches. Yahoo users perform 24.5 searches and Microsoft users run 26.9 queries.
More interesting than that, though, are the loyalty numbers. Google users are far more loyal, conducting 68.9% of all their searches on Google. Users at Microsoft and Yahoo only conduct 32.6% of their searches exclusively on those engines, choosing to also perform a number of searches on Google and other competitor sites.
In plain English, that means that Google users are more loyal to Google than Yahoo and Bing users are to their brand.
Once again I must point out how frustrating it is to have such a lack of information about this survey. First, the Reuters story I read originally (republished on Yahoo) doesn't cite any source whatsoever. And I'd like to take this moment to wonder aloud (in print) how this can possibly still go on? Haven't we learned our lesson over the past couple years with hoax stories that we should maybe cite a source? Isn't ComScore respected enough to earn a freaking link from Reuters? I, for one, am not inclined to believe something just because some faceless newswire journalist wrote it.
It's even more annoying when I learn that ComScore has their press release on their own website, publicly available. Good grief. And it's the news organizations that are threatening to charge for stories because of page views, right? (Yes, I know that's AP and Murdoch… but still… here's Reuters, doing what I would call a blatant attempt to keep the page views instead of letting users navigate to ComScore.)
Anyway, let's get back to the study. The last thing I want to do is gain a reputation of being the guy who questions all surveys, studies, and polls. But I'm about to ask some more questions. First of all, how does ComScore determine who is and is not a Google user? Their press release doesn't say one way or another.
And the loyalty stats—while probably right on target—are really confusing when you start wondering how they determine who is a "Google user" and who is a "Microsoft user”.
Do they simply ask the respondents if they consider themselves a Google or Yahoo user? Do they base it on toolbars or on which engine a user visits first? (Wikipedia suggests there is some level of disagreement on the accuracy of ComScore data, but then… I don't trust Wikipedia either.)
As I understood it, ComScore uses software installed on many computers, tracking users habits, to gather their raw data… then they use some weighting and advanced math to make sure they account for variances. So… if they're monitoring users' behavior, then how do they know which engine someone affiliates themselves with? Let me give you an example. Here are two possible statements about Bob, based on his web browsing data:
- Bob is a Google user. He uses Google 68% of the time. However, 32% of the time he goes to Yahoo to search.
- Bob is a Yahoo user. He uses Yahoo 32% of the time. However, 68% of the time he goes to Google to search.
Couldn't I draw either conclusion from one set of data?
Okay, so ComScore's numbers are probably pretty accurate. But it all hinges on how they determine who is a Google user and who is a Yahoo/Bing user. And I can't find any information on that. Please don't misunderstand my frustrations and my questions for a lack of trust in these numbers. I just wish there was more disclosure so that I could have a bit more context behind the numbers.
I sometimes wonder why we can't just say "more people use Google than the other two combined" and call that an important piece of data.