Many people refer to the data warehouse as a redundant data store. A properly engineered data warehouse is anything but redundant because the data warehouse is the only repository that provides an integrated view of enterprise data presented in a business semantic. In other words, the data warehouse is the only source for an enterprise view of information in business terms. The data stored in the ERPs, CRMs, MRPs, billing platforms, etc. are silos of operational data presented in a system semantic.
Your EDI challenge surfaces one of the most often overlooked critical success factors in data warehouse implementations, data stewardship. In the April 1998 issue of DM Review, Claudia Imhoff defines the primary objective of data stewardship as: ?Data stewardship has as its main objective the management of the corporation's data assets in order to improve their reusability, accessibility and quality.? The business needs to own the quality of the content of the data. It is a natural and necessary extension of the business owning ?process stewardship?. The definition of a valid process is one that produces data. Thus, the process is the means to the end, data. I recommend the business community signup for owning the content of the data stewardship process. That is, ensure quality of the data content. The IT community is better equipped and experienced with owning the context of the data stewardship process. IT, the data warehouse team in particular, can build tools and utilities to enable the data stewardship process for the business. Errors and exceptions can be flagged, routed and brought to the attention of the business process/data owners for correction and resubmission. The business defines the rules for clean data. IT helps enforce those rules. The business defines the procedures for ensuring compliance to those rules. IT monitors and reports exceptions and compliance violations. It must be a partnership between the business and IT communities.
Your EDI application appears to produce data as a result of the execution of the EDI modules. This makes the EDI application the system-of-record for that data. Even if the information made available via the execution of the EDI application comes from outside the company or is the result of a derivation of other information, it sounds like the EDI application created the data. If you were to associate the EDI application to the enterprise process model, you would find it is an enabling tool for several business functions thus several business departments/organization units. Facilitate a meeting with those business units to understand their dependency on the EDI results and the risks to their organization when those results are inaccurate. Get them to think in terms of their business processes rather than IT applications. This will set the stage for impressing on them the importance for making sure accurate data is produced from those processes, via enabling tools like the EDI application.
Dig Deeper on Oracle data and application integration
Related Q&A from Mike Lampa
When trying to design a data warehouse, we often try to model the database on the operational data model. Are there any guidelines in trying to ... Continue Reading
What is a surrogate key in a table? Continue Reading