We have built a 700GB data warehouse. Now we want to initial a meta data project, what is the first step?


Ross Armstrong’s Answer: So the easy part is done, now the project gets interesting. Meta data is one of the greatest mysteries of our industry. In my practice, I usually recommend running these projects in parallel because the process of sourcing and defining the transformations, no matter how simple, provide a significant amount of meta data. The first objective is to find out what the roles and responsibilities are of the meta data. Is it only used to define data elements in the warehouse? Is it to record the state of the data in the warehouse (i.e., date of last refresh, number of rows added, etc.)? Is there a legacy system tie-in? Can the business user use meta data to construct a query? Is the meta data to be the single authority on definition, ownership, where-used, rules of integration, etc for the corporation? Due diligence at this level will set the scope and success criteria of the project and saves tens or hundreds of thousands of dollars. Don’t get dragged down into a technology or meta-model discussion until you are very sure of the role and responsibilities of the meta data in your environment.

Sid Adelman's Answer: The first step is understanding what meta data you need, who will use it, how it will be used, who is responsible for it and where will it come from.

David Marco’s Answer: Start by creating a project scope document. Within this document you will need to define the specific business and technical drivers that your meta data repository project will attain. When identifying business and technical drivers, it is important to highlight exactly how the driver will benefit the company. The job of all senior executives is to make their company more profitable. They generally accomplish this by "reducing expenses" or "increasing revenues." Therefore, each driver should show how it either reduces expenses or increases revenues or both. This is especially important when speaking to high-level executives. Keep in mind that anytime you speak to these executives and you are not talking about reducing expenses or increasing revenues, you will be viewed like the teacher that was in the old Charlie Brown Peanuts cartoons… "Blah, Blah, Blah, Blah, Blah." Examples of the most common technical and business drivers include:

  • Reduce future IT development costs and shrink the system development life cycles
  • Drive the business users access to the decision-support information by creating a meta data-driven decision support front-end to help them find the information that they need
  • Reduce turnaround time for production related problems
  • Record and provide access to the business users for historical changes in the decision support system to allow them to make better business decisions

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access