We have a successful data warehouse (which is to say that business users are asking for more at an accelerating rate) and I am having a real problem getting the business users and other IT groups to understand the difference between analysis and operational reporting. Once the users saw the BI tool, they wanted to get all their reports that way. Now I am trying to explain to the business and other IT managers why it doesn't make sense to load every OLTP system data element into the data warehouse and convert our thousands of operational reports. We store the data in the warehouse at the atomic level to support very detailed analysis requirements, which makes the difference harder to explain. My fear is that the price we pay in attention and performance to make the system support all operational reporting (which is already available, if in a less sexy format), will keep us from providing real analysis capability. Any suggestions?


Sid Adelman’s Answer: Your question deals with a number of issues. The first one I’ll address is the desirability of rewriting the operational reports with the new BI tool only to make them sexier. This probably provides no value to your organization, has no cost justification and could impact other initiatives. Don’t do it.

Your question also deals with what data is being accessed for the operational reports. This question is a major topic for debate that has gone beyond the debate stage to name calling and some minor physical injury. There are some who want to use the data warehouse as a source for operational reports since it’s typically cleaner and better documented through the use of meta data. If you do decide to go this way be prepared for frustrating years of explaining to the operational people why the results are different than what they have been getting in the past.

Dust off your mission for the data warehouse. What are the data warehouse goals and objectives? If operational system reporting doesn’t fit, either change the mission or reject operational reporting.

Doug Hackney’s Answer: Replicating the same data in a new UI is not politically sustainable. In order for the system to succeed politically in the long term, you need to be providing unique answers to politically meaningful questions. Focus on the strengths of the DW system’s unique capabilities. And, BTW, welcome to the curse of success!

Chuck Kelley’s Answer: The data warehouse was originally defined to house data for strategic analysis. Operational and tactical reporting should be done via the operational data store (ODS). I would look at the Corporate Information Factory (a lot of that can be found at billinmon.com) environment to create an architected environment that provides for the needs of the organization, not just slam all the data in the data warehouse.

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