It's all so very
subjective. And therein lies the problem.
Jeremy Zawodny's numbers from his blog
would agree with Jill's 80%. Though Tim Converse
offers some interesting thoughts on the validity of the data in his blog, which is where I first ran across it. (Tim C is basically in the same position with Y! as Matt C is with G for those who don't recognize the name.)
Then again, if you look at the numbers from last July provided by a more formal study by SEW and ComScore's
relatively new qSearch, the numbers are a lot
closer. Noting that they have a much larger sampling of data and that they merge channels together.
My own numbers are somewhere between, with a slight favoring towards the ComScore numbers. I get more Google traffic than the ComScore data would indicate, but it's not nearly as skewed as Jeremy's or Jill's numbers are.
FWIW, everything I've read about the qSearch module are promising. Sure they don't separate out Google searches from Google Image searches, or Yahoo! Search from any of the other Yahoo! properties, but it's still pretty useful. It's the only independent measure out there that looks at a large enough sample (1.5 million users
) to be anything close to relevant.
The only major problem still being that each site is obviously different. And obviously targets a different audience sector. So will have totally different market share numbers.