Databases Reference
In-Depth Information
Embedded business intelligence may also be implemented as a separate
step in a workflow, but not directly embedded into an operational application
system. For example, after the order scheduling process has finished, there
may be exceptions flagged for further investigation. A dashboard or BI portal
may be used to check on these exceptions to help determine what action needs
to be taken. The key point here is that business intelligence is used as an
integral part of operational business processes and taps into historical data
from the data warehouse.
Operational Business Intelligence
For the purposes of this topic, the term ''business intelligence'' has been used
to describe the delivery of data from the data warehouse to the appropriate
business audience. Operational business intelligence is quite different. Oper-
ational business intelligence changes the focus to address more immediate
business issues that deal with daily business operations. Operational BI is
used by frontline managers and customer-facing employees to measure and
monitor operational processes.
Typical business intelligence would help track and monitor claim payments,
the number of claims being handled, and the amount of time spent on the
phone, each compared to target values. It is often necessary to do this more
frequently than a weekly or daily claims report. This operational tracking is
needed to provide more immediate feedback of performance so that behavior
can change in real time. A bigger push can be made to complete sales or to
follow up on lingering claims to get to a final resolution.
This operational performance could be displayed on a big screen that shows
the number of calls on hold and the number of calls handled so far on the
current shift. This information can then be reported by a team or an individual
claims handler. Some organizations prefer to publicly show results by team,
and then show each person's results on his or her own individual screen.
Sometimes these measurements can be pulled directly from the operational
system. However, it may be necessary to include historical trending and data
from other systems that the operational system does not have. In this case, this
is more in line with the requirements for real-time data warehousing, which is
discussed in the next section.
When exploring operational BI, you need to consider a number of things,
which revolve around the data in the operational systems. The following char-
acteristics of data need to be evaluated to determine whether the operational
system can support the business need:
History
Does the historical data needed to support analysis already exist in the
operational database?
