For many companies, the same data/metric term has different meanings for different departments.  An example of this would be “Customer”.  To the sales department, it may mean they have a new customer that has been entered into the system and now has an account number.  To Finance, “Customer” may mean that this person has been sent at least one shipment within the last 12 months.  If you want to learn about a Data Dictionary, read on.

For example, during your design and build phase, you have company agreement on the following:

• Names and definitions of your data/metric terms
• Department ownership of term
• Source of data
• Update frequency
• Reporting usage, etc

Then, you should be creating a cube to store this information and have it viewable for all to see.  You may want to add a time dimension, as business rules may change over time and that will allow for a history of term usage.

You can also use a cube to maintain Reporting housekeeping.  Creating a cube that lists out your reports and monitor it periodically to see if they are still being used or have become obsolete.  With this cube, I would suggest having a column for each department to answer “Yes/No” on usage.  Periodically, change the view so that a new “Yes/No” column appears blank and have departments answer if they still use the report.  You can filter on the blanks to then contact that department to make certain of usage, but if everyone is in agreement that it’s not being used, it is then one less thing to maintain.  Otherwise, you might want to find out what the report is missing that would make the report useful.  By doing this, you may find that you just need to modify existing reports and not create new ones that add to more maintenance.

As a tip, don’t show the column from their prior answer, else you get the copy/paste by the users and renders this exercise pointless.