Data Warehousing

I'm taking a break this week from the on-going discussion of the Business Intelligence Performance Index to include a topic shared with me by INFOBRIGHT's newest Senior Sales Eningeer Carl Gelbart.  Thanks, Carl!

Metadata is information about data and is critical for not only the business user but also data warehouse analysts and developers. Without metadata, business users will be like tourists left in a new city without any information about the city.
 
Databases, ETL tools, and reporting tools all contain metadata about how data is used in their environment. For example, a database has metadata about the data type and size of each column in a table.  An ETL tool has metadata on how source columns are mapped to target columns.  Reporting tools have metadata that defines table relationships and field definitions.
 
In my experience, the business user can become confused about where data in the data warehouse came from.  This confusion leads to a loss of confidence and credibility.  For example, marketing wants to do a mailer and they see an address in the data warehouse.  "Where did the address come from?  The billing system or the order entry system?  Is this address as it was entered by a customer service rep or has it been cleansed?   Does the address contain the house number or is it just street name?"
 
This is why having metadata that allows the business to get information about every attribute in the data warehouse, on reports and brings credibility to the data warehouse. This starts with where the data come from (i.e. database/table/column of origin, input screen field name), any transformations the data goes through, a definition of the attribute in business terms, the target table(s) and attribute(s) in the data warehouse and how that field is redefined in the reporting tools.