Few people would be crazy enough to let a bunch of architects each design a different room of their house without ever talking to each other and without having any common foundation on which to base their design. Oddly enough, that's exactly how most companies implement their document and records management solutions; one room at a time. It's no wonder people have little confidence in their ability to find the information they're looking for within their own companies. With all of that chaos created by the numerous unique information models and disorganized file shares, more and more companies are looking towards taxonomy development in order to improve their odds of finding information. The problem is that the term "taxonomy" seems to mean different things to different people. For some, taxonomy means folder structure. For others, it's a way of navigating a Web site.

In reality, what companies really need is a well designed information architecture that incorporates a number of elements such as:

• Metadata standards that incorporate standard "dropdown" values when possible to drive information quality.
• Standards related to the creation of folder structures/ file plans that allow for information to be inherited, reducing the time spent tagging content and improving data quality.
• Faceted classifications to improve search and browse results.
• Custom thesauri to capture the unique "language" of the company and industry.

While many consulting firms focus on gathering information only on the documents themselves, Kestral Group takes a holistic Master Data Management (MDM) approach to analyze not only the content, but also the database structures of key applications such as ERP or other business specific systems. This allows us to pull existing information table structures and corresponding values, which users are familiar with, to quickly build an information standard that are also easily adopted by the organization. In addition, this approach facilitates the data integration process by ensuring that data moving from one application to another does so with a common metadata definition including key elements such as name, type and length to further ensure data consistence and quality.

[Information Architecture: An Energy Case Study]