Loading ...
Sorry, an error occurred while loading the content.

Re: Taxonomy Relevance

Expand Messages
  • Clint
    It really depends on what you are classifying and the purpose of the taxonomy as well as the tools that you are using. Is it an enterprise taxo? Is it an
    Message 1 of 9 , Dec 16, 2009
    • 0 Attachment
      It really depends on what you are classifying and the purpose of the taxonomy as well as the tools that you are using. Is it an enterprise taxo? Is it an ecommerce taxo?

      In my world of ecommerce taxo - there are two different taxos that I use. The first is a product data collection taxo while the second is a product display taxo. They are two very different taxos with two very different purposes. On the data collection side, we want to be as granular as possible to get the most and best data possible. But on the display side, we may not need to or want to be as granular (number of clicks on an ecomm site is so important to the metrics people).

      The use TOE also depends on whether you are talking about nodes or attributes. And how easy it is to modify either. When I am creating an attribute with limited responses, I almost always put some sort of an 'out' as a response. Exceptions would be where there is a definitive list of responses - there are only so many styles of refrigerators or so many numbers of burners available on a stove or cooktop. But there is an almost unlimited number of fabric combinations for clothing. We then go by the 90/10 rule with an 'out' for the 10%. The 'outs' are monitored and when they get too large action is taken.

      If modification down the road isn't easy or is cost prohibitive, then a decision has to be made as to how complete you want the taxo to be. In cases like that, it is my position that it is always easier to roll up into simpler terms than it is to expand after the fact.

      As to a 'static' taxo - in ecomm there is no such thing. At least not that I have experienced. Item assortments are always changing. Someone is inventing the next widget. Marketing strategies change. Product managers think they are taxonomy experts. Any number of situations happen on a regular (read daily) basis.

      My best piece of advice (for ecomm taxo) is to build the foundation of the system as flexible as possible so that it can change and do so without it being cost prohibitive.

      --- In TaxoCoP@yahoogroups.com, Eric Erickson <AxelDC@...> wrote:
      >
      > You always need a "miscellaneous" category for just such emergencies. You
      > can classify it grossly in the area you think it fits, and if that doesn't
      > work, just put it in a generic basket.
      >
      > Once your miscellaneous categories get too big, it's time for another
      > iteration of your taxonomy.
      >
      > On Thu, Dec 10, 2009 at 10:10 AM, John O'Gorman <jogorman@...>wrote:
      >
      > >
      > >
      > > Good morning;
      > >
      > > The topic is awkwardly worded, but here's the issue:
      > >
      > > We've all run into the problem, and I've wondered how others in our
      > > profession handle it. No sooner have the last bytes dried on a beautifully
      > > rendered taxonomy than somebody in the company 'discovers' a platypus in the
      > > attic that doesn't fit the hierarchy.
      > >
      > > We have limited options all of which seem to work until the next outlier
      > > (or merger, or acquisition or...) comes along.
      > >
      > > How do we keep an established taxonomy from becoming an overhead nightmare?
      > >
      > > John O'
      > >
      > >
      >
    Your message has been successfully submitted and would be delivered to recipients shortly.