Classifying Complicated Information
There is clear and unambiguous information. The serial number on a computer is generally unambiguous information. While I have seen cases where a detailed product number was also needed in order to correctly identify the item, information like serial numbers are pretty exact.
This is the type of information that fits neatly into rows and columns. Creating reports based on this data can still be pretty complicated depending on what you want to call out. There can be hundreds of variables in a simple personal computer — probably thousands when you really look at it, but still this is information that is relatively easy to capture, store and reuse.
What is more difficult to deal with is the information in your averaging customer service ticketing system. There can be a lot of notes in those tickets, but just because there are notes it doesn’t mean that they are well documented. There is a lot of garbage information in those systems. Interim notes, partial notes, ambiguous notes, etc. It is ugly. There is information that is wrong, information that is out of date, information that sorts it all out and finally information that doesn’t fit at all. And yet, customer service reps have to sift through all the notes with this bad information in real time to help you solve a problem before you become really cranky.
The translation of your analag problem into a shared database system doesn’t always go well either. Do we have a classification and tagging system that is both flexible enough to accomodate what it needs, and structured enough so that you can find and report on items quickly. Is it constructed well enough so it will work well in the future?