The problem with web analytics is that it is a low barrier to entry. I have a saying - 'if you can't code then do SEO' - and far too many companies have some in

The problem with web analytics is that it is a low barrier to entry. I have a saying - 'if you can't code then do SEO' - and far too many companies have some inexperienced, non-technical squeaky wheel insistent on the overbearing analytics. It is not just the tracking scripts, there is a whole universe of snakeoil built on top of that. In ecommerce there are tracking script things that promise to deliver you the ultimate newsletter - if you sign up for it - and if you don't then all your interactions will be recorded anyway. This is all fantastic, however, there really just needs to be some compelling CTA for the newsletter signup, which is doable. Instead though, this stuff you could ask the customer has to be magically inferred. Another favourite is some stalking module for ecommerce that does personalised recommendations. This adds to the bloat and does 'jumble sale recommendations'. To do it properly, with code, is probably a simple SQL query but you still need to build, test, deploy such a thing in a professional way. But a third party script that magically does it, well, sign me up...! (Says the non-technical guy). On top of that is the instant help widget that someone in marketing expects. There is no need for it if the website actually works and information expected is provided. Again this widget needs to stalk 2000 people for the 1 person that uses it. So how do I face the challenge? Sometimes due to squeky wheel noise it is not possible. However, if you can do a better job of delivering what the required end information is, then you are good to go. Some people use analytics for actual sales information, again because some report can be produced easily without any knowledge required. Again, to do a proper report probably needs some knowledge of SQL and coding. So the result is business decisions made on data that is not correct. There is also a cost to some of these things, you can have a pop up giving customers 10% off their first order with yet more script going on. The script will charge a 8% cut to make it a 18% payout if anyone puts their email address in the popup box. Then there is affiliate marketing, which is okay but another level where more script is added and more 'fee' paid out for the privilege. So those scripts you don't see are costing money, real money. You as the consumer pay 10% of the product price to some mystery javascript bloaters every time you 'Buy Now' (which is always next day at the earliest).