comScore Claims Online Spending up 15%

I woke up to read in the morning paper and read everywhere online that data from tracking firm comScore Inc showed a 15% in crease in online sales on the Monday following Thanksgiving from 2007 to 2008. That seems reasonable. Then I read on...
"And on Cyber Monday...online spending reached $846 million, up 15 percent from a year earlier, comScore found." (link here)
This $846 million figure seemed drastically low to me plus I'm always dubious of "analysts" claims so I thought I'd check a couple things out to see if this figure is reasonable.
Amazon.com reported $5.67 billion in 4th quarter sales in 2007. (Link here) Let's give comScore a huge gimme and assume that the Monday after Thanksgiving is an average sales day within the quarter. That puts Amazon sales at $63 million on the day (I'll also give them another gimme and not extrapolate their 15% figure and add it to the $63 million).
Next let's look at heavy hitter eBay. eBay had gross merchandise volume of $16.21 billion in the 4th quarter of 2007. Let's do the same math on that one and we get $180 million for Cyber Monday sales.
So now with just 2 website we've already reached $243 million in sales of the supposed $846 million total. Don't forget that we're assuming that this is an average day within the 4th quarter.
I don't have the time to go hunting around for loads of internet companies sales records but at this point for my own thirst for knowledge I'm going to assume that the claimed $846 million figure is very low.
Think of all the big online sites where people spend money that I haven't included:
Dell, Office Depot, Staples, Nike, United Airways, every other airline, Zappos, WalMart, Target, Buy.com, and my favorite MileHighAthletic.com. I think it's safe to assume that big sellers along with the tens of thousands of smaller ones will put us way over the total.
What I find annoying is how readily the media eats up "analysts" data. The "analysts" and "experts" are wrong so often that you think there'd be more analysis of their findings or at least the media wouldn't trumpet their findings with such zeal. Companies collapse all the time yet the expert analysts who track these companies for a living often don't predict their demise.
Search Google for "online spending 15%" and you get an idea of all the respected publications that publish comScore's data with a hint of research.