Software

The Critical Elements of an Effective Business Intelligence System

To derive the many benefits of BI, organizations must assemble the necessary tools.

Grant Fjermedal is a science and technology writer with more than two decades of experience. He is the author of four nonfiction books, including The Tomorrow Makers.

Business intelligence is a powerful tool that can help organizations survive and thrive in a challenging and competitive business environment. But to use this technology effectively, IT managers must understand how the parts come together.

Generally, the heart of a BI stack is the data warehouse, the central repository for information. A data warehouse simply brings together all relevant information from across an enterprise. Before organizations established data warehouses, information tended to be trapped in silos — different departments within an enterprise deploying point solutions designed to meet specific needs, but that may not have been designed to integrate easily with others.

Once data has been compiled in a centralized data warehouse, all kinds of value can be extracted. Here’s a look at some of the basic elements of the BI stack:

Central repository: Usually called a data warehouse, this is where an organization collects data from a variety of applications and data stores from across the enterprise.
Data integration services: This layer serves as the integration point for importing data from multiple sources into a unified data warehouse. The power of data integration is that it can take information from an Oracle database in sales and combine it with feeds from a SQL Server database in customer support, as well as mainframe files from operations. This layer serves to extract, transform and load data from any source so that it is seamlessly available from the data warehouse.
Master data services: These provide synchronization and deduplication to protect the integrity of the central data repository, helping to ensure that, among other things, insertions, updates and deletions from source locations are posted to the central repository.
Reporting services: Once data from throughout an organization has been fed into a data warehouse, reporting services are used to extract its value. Reporting tools, whether created internally or purchased from a vendor, should support the creation of recurring reports — such as weekly or monthly sales figures or operational expenses. The tools should also support self-service reporting so that users can create their own reports.
Analytics services: Analytics tools can be used to create analytical databases that make it faster and easier to run custom queries or perform data mining. Enterprises can use analytics tools to create the multidimensional cubes of a data mart specifically designed to meet the needs of a particular group or function.
Predictive analytics services: Predictive analytics uses techniques such as statistical, regression, correlation and cluster analysis. By leveraging these measures, along with text mining, data mining and social media analytics, organizations can learn what to expect in a given area. They can use the models and patterns created, along with real- time data, to improve decision-making in situations such as loan approvals or product development