Web analytics, web metrics, and how to use them to track and improve conversions

Segmenting web traffic has been an enjoyable occupation for me and I just read Avinash Kaushik's post called Excellent Analytics Tip#2: Segment Absolutely Everything.

In a way, I did the same thing with Newsletters for www.thehousedesigners.com and www.mascord.com (without giving the details I'll talk about segmenting traffic).  

I usually decide prospects in a mailing list are hot if they looked at more than 4 pageviews from a newsletter link AND spent more than a minute on the site doing so.  The Urls are tagged with the customer ID and often there's clickstream analysis that tells us what plans were looked at by each customer.   I get a kick out of doing stuff like this and it's not high tech - it just takes some planning.

That's why I like Avinash Kaushik's  point about segmenting traffic:

New Nirvana Rule: Never report a metric (even God’s favorite KPI) without segmenting it to give deep insights into what that metric is really hiding behind it.

 

Just giving someone the number of visitors to a site during a certain period of time is not very helpful.

This is how Avinash Kaushik represents the data in Excel

Again, this is actually not that hard to do, and it's low tech - which goes along with Avinash Kaushik's first rule to throw away the expensive analytics and hire someone with that money and use less expensive analytics but better people.  

But I notice Avinash does not say, get a cheap solution that costs 10 bucks - he says get rid of your 100K solution and get a 10K solution and then spend the other 90K on me (or someone like me).   I like that idea but I would not want a 30 buck solution as that would be way too hard to get much useful information no matter how good the analyst.  Some packages are free and you get what you pay for.   As one of my high school teachers used to say:  "this is not good enough to wrap fish in".