Here's what I wrote:

Comscore reports around 5B total searches/month in the US across all search engines in July 2005.

...

However, I think comScore and Nielsen are wrong in these estimates. Not wrong by a little, but wrong by a lot. And it throws into question a lot of the analysis that people do about Google and the search engines.

...

Of course, both Yahoo and Google subscribe to these services. I wonder what they think of the accuracy of these reports?

Well, now we know what Google Director of Research Peter Norvig (he's someone who would know) thinks of the accuracy of those numbers. Not much. Take a look at this article lambasting the typical journalist's math skills on Norvig's "blog".

Peter writes:

Finally, another report on Nielsen/NetRatings says that "The activity at more than 60 search sites makes up the total search volume upon which percentages are based -- 5.1 billion searches in this month." Of these, Google gets a 46.3% share of the searches. 46.3% of 5.1 billion is 2.4 billion, so Nielsen/NetRatings (as quoted by these two reporters) is saying that each Google search results in 7,000 clicks.

Exercise for the reader: is the Nielsen/NetRatings estimate of 2.4 billion Google searches per month (which translates to 80 million per day) accurate? Consider research such as ["million google searches per day"] and note the year those estimates were made.

In case you are too lazy or disinclined to do the math, (Norvig would say you'd do fine as a typical journalist), he's implying that those numbers are way off.

Thank you from me to Peter Norvig. I only wish I didn't have to steal bartnagel.com's bandwidth to put up a picture to display of him in one of his loud shirts. Oh what the hell, Norvig probably has $546M in Google cash lying around, and he's stealing bandwidth from Bart, so maybe I'm ok...

¶ 2/16/2006