I was talking with a colleague the other day about putting multiple codes on a bunch of web pages to see if different web stat packages treat information differently. Even before I started my own exercise he categorically said that of course the stats would be different as each treats the information with its own set of rules.

One package will ignore one set of data as it will see it as ‘not a true picture’. Another package will see things in a different light and want to know more about a particular element, breaking it down further into its component parts and then ignoring some of the sub sets of data.

It would be interesting to know which package people use and why they prefer them over others. It may well be that they use what came with their web site package by default and continue to use it because they have got used to it and so even if another system was proved far more “accurate” in its accounting process, actually they’d rather stay with the system they know. Perhaps they haven’t got time to learn a new way of doing things or they view the data as it is supposed to be viewed, as a comparison chart.

We get more visitors to this page than that. This page gets a bunch of attention from the search engines. How about if I just tried adding, or subtracting, words from the title. Will that get me more traffic or doesn’t it matter? The visitor is only hanging around 1.3 seconds so hey, I’ll focus on telemarketing anyway!!

<b>What’s your experience of web analytics?</b>

hi, Jonathan, how are you?

i've used coremetrics, omniture, hbx, & google analytics along with a number of the old 'log analysis' tools. i've used webtrends in the past too, but only for a couple of months so i'll leave that out of here.

here's my descending order of preference:

1. coremetrics
2. omniture
3. google analytics
4. hbx

• coremetrics: most complete, intuitive package with the greatest range of tools. simply fantastic account management.
  
• omniture: huge functionality, not quite as flexible as coremetrics.
• google analytics: ultra-intuitive. only downsides are lack of customisation (some level through filters & goal reporting, but not close to the other 3)
• hbx: not intuitive, slow interface, heavily dependent on implementation

i've overlapped a couple of them & they've never matched up. I have big problems getting google analytics to report e-commerce stats correctly - never been able to figure that one out.

if you're using one of the others, my advice is to also run google analytics alongside it just to balance out the picture & find out where there is an over/underreport.

working to improve data

coremetrics (& omniture too if I remember) has some nice functionality that allows you to automate a batch of your e-commerce orders across to them to reconcile against their stats. as those are the most important business stats, that's extremely useful.

I hope that's useful - not quite sure if it's what you were after!

daniel