Maybe it all boils down to finding out who has the least bias. I there really such a thing as 100% accurate these days? Sounds like wishful thinking to me.

Best.
alain
http://www.mor.ph

Look, Nate, it’s crystal clear that Quantcast’s tagging program is not endorsed nor used by the vast, vast majority of important sites that carry advertising and who account for virtually all online ad dollars. So, all you get if you use Quantcast is a mishmash of data that is unquestionably biased toward the smaller and less relevant sites on the Internet. Ad agencies are simply not going to use such biased numbers.

None of the tools (ComScore, Nielsen, Quantcast, etc.) are going to be 95-100% accurate on web traffic for each property out there. In fact, you can almost count on any one of them being off at some point.

As a media planner, I’m more interested in being able to slice and dice the data (i.e. defining a target audience) that will get me to the most appropriate sites for my online marketing campaign. Google just doesn’t add up in this area. Unless they make significant improvements, ComScore/@plan will still be the best tools for online media planning.

Nowhere near the 10 million you claim.

Let me summarize some other relevant numbers for you that you can verify from Quantcast’s site. Of the 1000 largest sites in the U.S. that accept advertising, fully 80% DO NOT ACCEPT Quantcast’s tagging. Here are some of the more important sites that DO NOT COOPERATE with Quantcast: Google, Yahoo, MSN, AOL, Myspace, mapquest, amazon, facebook, craigslist, information.com, go.com, blogger, paypal, CNN, all.com, imb, flickr, geocities, yellowpages, whitepages, weather.com, nextag, classmates, nytimes, webmd, shopzilla …. need I continue through the other 800 sites in the largest 1000 that DO NOT cooperate with Quantcast? You should get the point already.

I think it’s particularly telling that the largest site that DOES carry Quantcast tags is wordpress while the fifth largest site that DOES carry Quantcast tags is … wait for it .. YOUPORN.com! That’s one heck of a negative reflection on the quality of Quantcast’s program, don’t you think?

Look, Nate, it’s crystal clear that Quantcast’s tagging program is not endorsed nor used by the vast, vast majority of important sites that carry advertising and who account for virtually all online ad dollars. So, all you get if you use Quantcast is a mishmash of data that is unquestionably biased toward the smaller and less relevant sites on the Internet. Ad agencies are simply not going to use such biased numbers.

My recommendation to you if you are a publisher is to buy comScore or Nielsen data. It will cost you a few bucs more than the free Quantcast numbers you use today. But, at least you’ll get something that advertisers and their agencies are using to place ad dollars today.

Your belief as I understand it is: Quantcast data is less accurate than panel-only services because tagging a site significantly increases Quantcast’s people UV estimates; a site that doesn’t permit tagging will be significantly undercounted by Quantcast relative to tagged sites.

The point in my last post is that both of these are valid arguments. And without someone (being Quantcast, ComScore, IAB, etc) providing data to support either of these positions, you will have people on both sides of this argument.