"The real search market share report is measured in terms of estimated unique U.S. visitors rather than number of queries performed. Because of automated rank-checking tools (and manual rank checking), Google’s estimated search market share based on queries performed is heavily inflated and therefore an unreliable indicator of potential search traffic."
Clipped to Evernote for future reference.
3 Comments
Clipped to Evernote for future reference.
3 Comments
Comments
Unique visitors is not a good measure. In fact, that’s sort of what we used to have before we had number of searches. The problem is that a particular visitor to a site might do more than one search. Moreover, automated queries are unlikely to be a major factor in skewing results given that two of the services use browser-metering software. I can’t even see that ISP data would be that skewed. In addition, rather than Google being overstated, it’s probably understated. Consider how Google is in the 60-70% range of share of US searches yet many site owners find it drives 80-90% of its traffic. Number of searches isn’t perfect, has plenty of problems, but nope -- unique visitors isn’t a better number.
I don’t see Michael’s data representation as an "instead" but an "also" view on search engine usage.Unique visitors vs. queries performed is a lot like unique visitors vs. page views: different stories about the same thing.Adding Michael’s view can help hold on to a realistic feel of search penetration outside our techno-world: 70% of all quries go to Google souns deafening (and it is) but it doesn’t equal that the majority of searchers go to Google.Just adding to that 70+% of queries picture PEW’s data that only a litle under half of people online will do a search in a given day changes my impression of their daily reach.Given that half of the people online today will *not* come to your site via any search engine, the fact that we often see 75-90% influence by Google could be saying a lot about a site; either about the type of site (one-time use) or about its quality/stickiness.Oversimplified, botht metrics and my use of PEW’s data, but puzzle pieces none the less: can’t discard one for the other.
My $0.02: how many people search for "Google" at all other SEs? If Ask is to be believed: http://about.ask.com/en/docs/2008/topqueries.shtml - a LOT!!!!If your a unique, and go to Google quickly, you get counted for both, but really, is the guy in a $7,000 suit really an Ask unique I mean, COME ON!