Architecture essentially defines the structural design of a system that has to fulfill its purpose over a long period of time. Star architect Norman Foster once put it this way: “Architects design for the present, conscious of the past, for a future that remains largely unknowable.” This describes a dilemma faced by business, IT, and corporate architects. On the one hand, architectural design is influenced and constrained by a legacy environment that has developed over many years. This includes aspects such as the organizational structure, a historically grown system and IT landscape, organizational skills as well as corporate culture. Awareness of these factors is important, and long-term architectural decisions and designs cannot neglect these boundary conditions. For this reason, large-scale architectural designs are usually not developed on a greenfield basis. On the other hand, while architecture is concerned with long-term stability, it must also be able to adapt to changing requirements and future needs. This applies in particular to business intelligence architecture. BI architectures need to be able to provide timely updates to new data and information requirements and furthermore extend the reach to new sources of data to meet unknown future requirements without major disruption, while at the same time meeting current business needs. In the following chapter, we will explore architecture patterns and elaborate how a reference architecture for business intelligence can be designed using an enterprise architecture approach. In the context of this book, architecture is not the main focus, but we will provide some guidance on how archtitectural aspects of enterprise-wide business intelligence can be addressed as part of a BI strategy to be able to effectively deliver on the strategic vision and value contribution of business intelligence.