Tuesday, November 18, 2008

Web Analytics - An Hour A Day: Book Review


Web Analytics: An Hour A Day
Avinash Kaushik

I am a numbers geek. My current job consists mainly of sifting through numerous amounts of data, mining and modeling it, and then telling a story with it. Since I deal with huge amounts of data, I was interested in reading about analysis strategies for a field that seems to equate to an almost endless amount of it.

The author, Avinash Kaushik, has an extensive background and experience in the subject. He currently authors a well known blog in the web analytics community called Occam’s Razor at http://www.kaushik.net/avinash/. Additionally, he regularly gives presentations on the topic. I also read in the forward that 100% of the proceeds for this book are being donated to two charities for cleft lip and palate surgeries in underdeveloped countries and Doctors Without Borders. If that doesn’t show that his main agenda is in sharing his knowledge and enthusiasm with readers, I don’t know what would.

The layout of the book is broken out into what I would consider two main parts. It first builds the foundation that includes basic things you need to know about web analytics and then goes into some business theory as a basis for interpreting the results. The second part of the book then breaks out all of your web analysis tasks into a seven month exercise that uses a different topic each day or two and encourages you to spend an hour a day on each task, hence the title.

The audience the book speaks to is very broad. Every business oriented person can learn from this book; however the entire book would not necessarily be applicable. For someone who would be more of a user of the data, which would be myself at this point since I don't have access to this specific type of data, the beginning sections where much more valuable then when the books starts delving into the technical applications.

One of the main points of this book is to be able to create actionable analyses with the data. This really resonates with me, because I also believe that is one of the biggest disconnects when you are in a data centric job. You can look at things a million different ways, create charts, graphs, ratios, percentages that are all really nice and can make a great report to upper management, but the bottom line is that’s great, now what.

Overall, I would highly recommend this book. It has a great foundation of concepts that are relevant to all audiences and then the detail needed to get to the hard core analytics for those who need it.

No comments: