Avinash Kaushik has a great blog post about trying to do predictive analytics with web data:
“Data Mining And Predictive Analytics on Web Data Works? Nyet!”

Avinash is one of the truly brilliant minds in web analytics, so it’s great to see him put his brainpower behind explaining this assertion. And, it’s timely, in light of the new book by Ian Ayres, a Yale Law School professor and econometrician. I really need to order the book and read it, as I’ve got preconceived notions based on watching an interview with Ayres. Fortunately, I’ve got a B&N gift card and we’re B&N members, so get an additional discount. Hmmm…my link above is to Amazon…yet I’m going to buy through B&N. Why is that? Topic for another post…on someone else’s blog, I suspect!

For now, I’m going to view these two sources as representing two schools of thought / approaches to data. On the one hand, we actually have, oddly enough, highly trained statisticians, academics, AND casual users of business data. This is a group that, overly simplistically, sees “more data is better,” albeit for different reasons. The statisticians want more data so they can find increasingly subtle correlations in the data. the casual users want more data because, like revenue, more data is better, right?

The other school of thought is a bit more grounded in reality. As I’m prone to do on this blog, I’m once again blatantly showing where I fit. This school of thought recognizes that more data brings along the need for more discipline — more *business* discipline — to actually get actionable information from the data.

Now, I’ve got to go and see if I can figure out how to use a BN gift card AND get my discount through their online site.