I have just returned from the 2007 DAMA Conference in Boston. It was an intense conference with over 750 attendees. Topics ranged from Metadata Management to data modeling challenges to developing your enterprise data strategy.
I sat in several worthwhile sessions and came away with one excellent suggestion that BAs can begin to incorporate right away. It comes from one of the many sessions on Data Quality. With organizations amassing more and more data of various types, the quality of the data stored in databases is often questionable. Many of our applications contain redundant data. We have thousands of null or missing data values and much of our data is old and obsolete. Inaccurate data really causes problems when a business stakeholder asks for a new report or query from our databases. When we are assigned to a new project that involves the use of existing data, one of our first questions should be how accurate is the data? Business stakeholders may not be aware of the number of errors that exist in production databases. In our company we are always amazed at how many bad email addresses that we have when we try to send out a notification. Many of our contacts do not notify us when they get a new email address so when we send out messages, we gets lots of errors.
If we go forward with projects, utilizing existing data and the data is not accurate, then the time spent on the project has been wasted. As an excellent Business Analyst, we can add value at the beginning of projects by asking and researching the quality of existing data. If we find that the data quality is poor, we should recommend a clean up project first, before designing new reports and queries. These “data cleansing” projects can be expensive but improving the quality of our data, which is a corporate asset, increases our ability to make good business decisions that rely on the data.