Building a Strong Data Architecture

Keep data available

Portal, Web-based and mobile application deployments keep employees in close contact with data entry.

Database replication techniques allow for broad data distribution while enforcing data quality rules.

Security should be implemented as close to the data as possible using row- and column-level permissions.

Keep data accurate

Maintain a centralized, consistent data dictionary. Try to initially create departmental databases by building subsets of larger operational databases. Enforce business rules at all points of data entry using shared business logic. Keep data in context When possible, fold warehouse and analysis tasks into operational databases to allow all data to be manipulated in one place. OLAP, data mining, data visualization and statistical analysis are all highly effective (although underused) techniques for understanding data. Web services allow data from enterprise resource planning and other vertical applications to be integrated into line-of-business applications far more easily than ever before.