Summary

We have learned to accept the reality that 90 percent of business users are unlikely to ever build their own query from scratch, and they need a more appropriate access method. Business intelligence applications provide the bridge most business users will take to access the data warehouse. In this chapter we described the nature and purpose of BI applications, ranging from standard reports to more complex, business process-specific analytic applications. We also talked about some of the tradeoffs involved in buying packaged analytic applications.

The bulk of the chapter described the Lifecycles two-step process of specifying and building an initial set of standard reports to be included in the initial deployment of the BI system. The initial application specification step follows closely on the heels of the requirements definition process. The application development step must wait until the data is available, the tools are selected and installed, and the specs are completed. As a result, the two steps are separated by a fairly large time gap, usually measured in months.

Both the specification and development processes involve participation from the user community. This helps build an understanding of the strengths and weaknesses of the BI system, and an ownership of the decisions around which reports to build and the tools and data choices that went into the initial design.

The next chapter presents many of the techniques described here in the context of Microsofts Reporting Services.