By Stephen Putman, Senior Consultant
The prospect of implementing metadata in most organizations is somewhat daunting. The types of metadata that the typical enterprise system produces are quite varied, and the value of collecting and reporting it is not immediately obvious. Many times these programs run out of steam before they even really get out of the starting gate because it is difficult to know how to start collecting this data.
I recently experienced this sort of issue at a large financial services client. The staff understood the basic value of metadata collection – end-user documentation, elimination of data element duplication, and data traceability through their process – but had a hard time getting started. I explained to them that as business analysts they already had the tools and experience to gather the metadata they needed, and were already doing it! The subject of requirements documents and mapping spreadsheets is quite familiar to business analysts, so the key to pitching an ongoing metadata process is to show that the techniques used to collect data for these documents are exactly the same as the techniques used to create a sustained metadata management system, just in a different format.
I illustrated this by citing examples from a human resources outsourcing project that was in progress at the time. I took the data mapping spreadsheets that were being used and translated them into a homegrown but industry-standard database system. By doing this, I demonstrated that data could be used to produce an ad hoc reporting capability to produce end-user data dictionary documentation, data lineage reports, and reference material for other business analysis activities. This is accomplished by standard database query techniques, which is much easier than extracting data from spreadsheets and other standalone documentation.
Convincing business people and end users that a metadata management system is worthwhile involves two main factors, neither of which involves the tools themselves: (1) the realization that the analysts are already doing the collection work, and (2) the idea that a centralized metadata repository can be leveraged across business requirements and applications to deliver common data definitions and rules. Implementing these two concepts puts you well on your way to a successful metadata management system—and evolving metadata maturity!
Stephen Putman has over 20 years experience supporting client/server and internet-based operations from small offices to major corporations. He has extensive experience in a variety of front-end development tools, as well as relational database design and administration, and is extremely effective in project management and leadership roles.
Recent Comments