Realizing A Real-Time Enterprise With Business Performance Management
08/02/2004
By John Medicke, Executive IT Architect; Architect, SMB and Mid-Market Architecture, IBM
Reacting with speed to changing business conditions, occurring either internally or in the marketplace, requires insight and agility. More than ever, businesses need to leverage their business intelligence systems to fuel this responsiveness. However, many legacy data warehouse environments are holding organizations back with their crusty structures and stale data.
There is good news. Business Performance Management solutions offer a new approach to corporate informational alignment. Business performance management derives performance from the business process coalescing the operational and the analytical environments. The business process management environment, as the superstructure of operational activity, is the perfect channel for the fluid execution of performance management. Business measures are streamlined into the analytical environment and actionable knowledge is turned into action by driving business process operations.
Constrained by Your Legacy Infrastructure
Many organizations are inhibited by their existing data warehouse infrastructure. Throughout the creation of the warehouse, it has had many masters. It has been driven by very specific point requirements and becomes a hodgepodge of varying informational content jumbled together through some very bazaar ETL techniques. ETL, which is almost always the most expensive part of any data warehouse upgrade, severely limits growth in functionality. Of course, the pace of the analytical thirst is not slowing down to accommodate this legacy, but is ever-increasing.
Cost and complexity are not the only inhibitors. The sluggishness of the warehouse population can leave the reports as day-old news. To meet business goals, business intelligence needs real-time information. It is no good to find out that a particular unit missed its profit target for the day at the close of business. You need that information while you can still do something to change the outcome. But many legacy warehouses are too cumbersome in their ETL processing and too slow in the warehouse and data mart roll-ups to meet the requirement for real-time information.
Lost in Disorganization
Many data warehouses are a convolution of data models that do not reflect the organizational or process structure of the business. Multiple data marts and often multiple data warehouses exist in one enterprise. The data models of the data warehouses are often fragmented, reflecting the data islands of information that exist in the operational environment. Changes in business requirements and changes in IT strategy have brought about this state of disorganization.