Website Measurement Hacks: Tops and Tools to Help Optimize Your Online Business, by Eric T. Peterson

Yesterday, my review copy of Website Measurement Hacks hit the mailbox, and without ado I started reading it, taking breaks for coffee and updating the website. I am, after all, very interested in this sort of thing and have self studied a lot of it over the years - this month, with renewed interest. 'A Recipe for Newspaper Survival In The Internet Age?' had some of the influence, though with more of a web analytics bend. Anyone who has a website eventually wonders who is reading visiting their site; this includes people with weblogs, businesses, government offices and even non-governmental organizations (NGOs). What has been lacking has been a decent, basic explanation of how these numbers are to be had, and what they are to be used for. This particular book goes into getting the numbers themselves - and even has some code fragments within the book in javascript and Perl. The beginning... started at the beginning, explaining how web measurement statistics were actually an afterthought - and crediting the first web statistics package (getsites 1.4) to Kevin "Kev" Hughes of Honolulu Community College back in 1993. That's 12 years ago. Web measurements have come a long way since then.

For the absolute beginner and the layperson (such as myself), the beginning is important - though as a layperson, I wanted to get past the explanation between the differences of unique visitors, page views and hits - thankfully pointing out to everyone that a page hit can mean any file downloaded - an image on your webpage counts as a hit, folks! While I was sorely tempted to rush past those parts within the first chapter and get to what I consider to be the meat, it was a pleasantly fast read and helped me get my mind more focused on the topic at hand. Various types of measurements are looked at, and what I found really valuable was the table (1-2) of Criteria to determine which data source for statistics one should use. This table really makes it simple to compare the Logfile, which just about every web host has, as well as the Packet Sniffer type and the Page Tag type.¹

Around page 32, I kicked down a gear with Hack #9 - the 'Write a Useful Web Measurement Request for Proposal (RFP)' Hack. Why? Sure, I know about RFP, and I have my opinions on requesting proposals from vendors - but there is some solid advice in there which I was trying to find fault with. Instead, I found myself nodding at how well the author tied together what is really needed in a RFP - or more importantly, what an organization needs to know to write a RFP. That set the tone for the rest of the book.

Altogether, there are 100 Hacks in the book. Some hacks that O'Reilly make available in PDF are: Hack 2: Best Practices for Web Measurement Hack 47: Measure Content Syndicated via RSS Hack 58: Use the Entry, Exit, and Single-Access Page Report Hack 86: Manage Lifetime Value Using the Visitor Segment Value Matrix Hack 94: Use Key Performance Indicators These are, of course, teaser hacks. The hacks I found most worthwhile were Hack 13 (Build your own RSS Tracking Application: An Overview and Data Collection), Hack 34 (Measure Your Mistakes), Hack #35 (Build Your Own Web Measurement Application: The Core Code)... all of which probably say more about my interests than they say about the book. Still, the book had the answers to some of the questions I didn't even know I had!

In the end, there's just a lot of well thought out information in the 389 pages. The real beauty of the book is that one can hop into any one of the hacks, and flip back to previous hacks as needed, when something you don't know pops up. If you know it all, start on the last hack and work your way backwards!

This is a useful book for people who want to know more about measuring the statistics on the websites under their supervision - and in this it is overflowing with information. If even you think you have an idea of what you are doing, this book may allow you to correct what you are doing wrong - and perhaps even get better data to analyze. Many times I have seen organizations have different statistics from different tools for the same sites which do not jive, and I have a better understand of why this happens and how to avoid it now.

I do not often keep 'hack' books around long, but this one has already found a place on the well referenced, self dusting part of my bookshelf. I'd have to say that it would be a good gift for an aspiring blogger or website administrator that you know. Maybe even you.

Meanwhile, I expect that Web Analytics Demystified is in my near future.

¹If it seems that I am speaking a foreign language, you probably should read the book