Krucoff nails it here:
My problem, and question, has to deal with with the individual breakout of specific blogs and determining their rank by unique visitors and visits. Publicly available stats highly contradict these rankings and I’m curious as to how you would explain a blog like Gawker.com, among others, (is it a coincidence they co-sponsored this report?) would rank so much higher than Engadget when Engadget’s traffic dwarfs Gawker.com’s in reality?
How reliable is this panel data when projected on blogs with considerably less traffic than, say, the top 100 sites on the Internet overall?
Anyone out there have background on Comscore and how they make money/operate? I called and told them I wanted to pay to be in the report and have Weblogs, Inc. consolidated in the rankings. They said they will have someone get back to me. If you have to pay to be represented correctly in a research report that seems a little wrong to me. I’d at least like to know what Nick paid for this and how much it would cost me to make the report accurate!
I know Rick Bruner, and I know he’s friends with Nick Denton, so I would think at the very least the report would have a disclosure that Nick’s friend was paid to work on this report. Additionally I told Rick that this data was way off months ago and that I would provide our log files to comscore and he never got back to me. So, Comscore knew this data was wrong and published it anyway.
Update: Rick talks about the stats here:
I’m also talking to Rick about it right now.
Update2: OK, I talked to Rick about the results. This could have been done better and he is getting me on the phone with Comscore today. My hope is that we can get all the data from Comscorewithout payingand make that available to everyonenot just the person who paid for the survey (and their friend who wrote it)! I’m pysched the report is being done, but Weblogs, Inc. was blocked from being involved in this report because of Gawker’s sponsorship and that makes the entire report suspect. I’d like to be involved in the report, and maybe Corante and Blogger/Google would be too. If this needs to be paid for that’s fine let’s hire someone who is not good friends with one of the two sponsors, let’s split the cost across 10 companies, and let’s make the data public.
In other words: let’s make it a fair survey.