360 Web Consulting

SEO, Analytics and Web Design Consultancy

  • Home
  • About 360 Web Consulting
  • Main 360 Thinking Website
    • Edit
    • Delete
    • Tags
    • Autopost

    Analytics - should be a treasure map, rather than a JCB

    Web analytics is all about clicks.

    Google Analytics, which is used by about 80% of the world's websites (yes I made that stat up), is a tool for counting clicks, tracking clickstreams.  That's the simple bit, accumulating bucket loads of data.

    But not all clicks are equal.  Some clicks are much, much more important than others.  Most of them are irrelevant facts, a few are clues leading to buried treasure.

    So the hard bit is to ignore all the unimportant stuff and identify the clicks which mean something.  The clicks which are clues, which represent trends, changes and  exceptions.

    Then comes the harder bit - looking at these important clues, interpreting them and finding the opportunities to make you money; or build you a mailing list; or get you more phone calls; or get you more loyal readers.

    Good analytics = counting + sifting + analysing = making the right decisions

    The trouble is most analytics packages are great at counting the clicks and gathering related data.  They are pretty good at displaying the clicks, and even slicing and dicing it up however you want.  But they suck at giving you clues about which bits to slice, where to sift for the gold.

    Which gets in the way of you doing the analysis stuff.

    Don't get me wrong, Google Analytics has some features which are great for picking out the clues.  The automatic intelligence alerts are pure magic (though hidden far too far out of sight).  Weighted ranking is fantastic.  Advanced segments and custom reports are very powerful.   All of these can save a whole lot of effort and labour.

    But these are added on to the main experience.  They require switching view, changing the level your thinking on, stopping to work out the best way to ask the right question.  I want to conjure them up on a whim, with a simple click of a mouse, so intuitively I don't have to stop to think about them.

    Why can't I quickly build up my segments iteratively as I navigate the data?

    Why do I have to guess which keywords are causing my growth in traffic over the course of a week?

    Why aren't the automatic intelligence alerts there all the time, so whenever I'm looking at a new trend I'm just one click away from understanding the underlying changes?

    Why can't I quickly compare the keywords of two arbitrary dates?

    Why doesn't it automatically group keywords and analyse them en masse?

    Why do I have to export it all to excel so that I can run statistical checks and look for correlations and patterns?

    Surely I can't be the only one asking these questions?

    Anyone found a solution?  Or planning to build one?  Post a comment, let us know!
    Tags » SEO analysis analytics clickstream decision making excel google analytics keyword analysis keyword clusters long tail keyword analysis online strategy rant traffic analysis web analytics website optimisation
    • 22 October 2010
    • Views
    • 0 Comments
    • Permalink
    • 0 responses
    • Like
    • Comment
  • Michael Hayes's Space

    I'm a contractor specialising in website development and optimisation, focussing on Search Engine Optimisation and analytics (getting more visitors then analysing what they do on your website).

    Click here to find out more about my services, or post any questions you have in a comment and I'll email you back.

    Contributed by Michael Hayes

    • Contributors
    • Michael Hayes
  • About Michael Hayes

    I'm a contractor specialising in website development and optimisation, focussing on Search Engine Optimisation and analytics (getting more visitors then analysing what they do on your website).

    Click here to find out more about my services, or post any questions you have in a comment and I'll email you back.

  • Subscribe via RSS
  • Follow Me

Theme created for Posterous by Obox