We have discussed the use of DSS for making decisions. Now let's go one step further, and understand how the role of DSS has evolved into business intelligence systems. The term BI, like DSS, is a content-free expression. So it means different things to different people, and no one can agree on the exact same definition. The Data Warehousing Institute defines BI as the processes, technologies, and tools needed to turn data into information, information into knowledge, and knowledge into plans that drives profitable business action. BI is a umbrella term that encompasses data warehousing, analytical tools, and applications. These are leveraged to create business intelligence. The BI process is based on the transformation of the data, to information, then to decisions, and finally to action. BI is outcome from this blending process. How does BI defer from DSS? BI mainly uses integrated data from data warehouses with many possible data sources and formats. While DSS can use any data source including a data warehouse. BI systems provide information that leads to better decisions. DSS supports individual and institutional decision-making directly. BI orients itself toward executives and strategic decision-making. DSS are usually oriented toward end users, analysts, and middle level managers. BI tend to be developed with commercially available tools, DSS tends to use custom applications to focus on structured and unstructured decisions. And finally, BI came from software industry, while DSS largely oriented in academia. Imagine a timeline with Leo 1 back in 1950s and the third generation of DSS appearing in the 1990s. BI is actually an outgrowth of the same process that created EIS systems in 80s and the mid 90s. This is where dynamic multidimensional reporting forecasting and predictive analytics, and data mining were developed. Then in about 2005, artificial intelligence, and powerful analytical capabilities were also introduced. These developments provided information for executives with broader access to deeper intelligence at the drop of a hat. While the term business intelligence has been around since mid-90s, system like you see today have been around since about 2005. BI systems have four major components, a data warehouse, business analytics tool for manipulating and analyzing data, a business performance management systems for monitoring and analyzing performance, and finally user interface in forms of portals, dashboards and score cards. Here is a generic BI architecture environment. Source systems provide data to data warehouses and data marks. Sources can be ERP systems, point of sale, web data, legacy system, a spreadsheet or transactional databases. These source systems can be on different platforms and store data in many different formats. Data integration services includes extract, transform and load services. Enterprise application integration services, enterprise information integration services and operation data feeds services. Data management services employ a variety of architectures, technologies and data models, including federated data marts, data warehouses, and OLAP cube data. A collection of the tools are available for manipulating, mining and analyzing the data for reporting via business performance management dashboards and scorecards. Finally, different information delivery tools and applications can communicate the BI to many different users. Including IT developers, analysts, information workers, managers, executives, front line workers, suppliers, and customers. As such, the trend towards pervasive BI means extending the reach of the intelligence to other organizations. BI traditionally has been implemented around achieving long term enterprise wide goals such as increasing revenue, or market share, reducing costs, improving customer retention, and of course, increasing profit. Providing support for strategic issues like these are BI's bread and butter, but enterprises now expect that a more effective use of information will lead to better execution of numerous decisions. These include tactical issues that focus on short term initiatives, like helping a sales manager optimize the region-wide sales campaign. In these cases, progress is measured against preset goals on tactical dashboards. Another realm of BI applications is in operational category. In this case, BI helps to identify process-centric solutions. For optimizing a specific processes, such as product managers deciding about a discount, a schedule, or human resource managers configuring a new work and break schedule. Operational BI, however, is mainly used at the departmental level, where progress can be measured and tracked using real time charts and reports. These are all small items compared to those big strategy questions, but that doesn't mean they are without value. BI's value proposition stretches across an organization, and can help answer questions about customer segmentation, propensity to buy, customer profitability, fraud detection, customer attrition, and channel optimization