Since a lot of posts have been written on ETL in Informatica, today I thought , lets write a piece on Business Intelligence, the term everyone in the non IT industry today is talking about. in IT, BI is a norm.
So what exactly is Business Intelligence? – In lay man’s term we can define BI (I will be using the short form throughout this article) as combining applications, technologies and data warehouse to gather and analyse the data, which in turn helps organizations better understand their operation, gain more insight into factors that affect their business and hence make better decision.
A BI application can be classified into:
- Data Visualization: Makes data analysis easier and faster by providing insight into the underlying patterns and relationships contained in the data. Scorecards, dashboards and reports are presentation media types widely used.
- Analytics: Analytic application allows the user/analyst complete access to the cube (will be explained in the later part of the post), wherein the data can be drilled down to the lowest level, sliced and diced it in any way. Some of the analytic applications are data mining, predictive analysis and OLAP.
Architecture of Business Intelligence system:
Components of Business Intelligence.
Business intelligence is composed of 3 components
- ETL–The ETL tools perform 3 functions namely extract data from different sources, transform the data to format compatible with the destination data source and load into either a data warehouse or a data mart.
- Data warehouse –Stores historical data from multiple sources.
3.OLAP (Online Analytical Processing) – This is the most important component of BI system. Using an OLAP tool, users can perform querying, reporting and multidimensional analysis.
Multidimensional Analysis is analysing the data in more than one dimension at a time. For e.g. analysing sale of a product by store, by region, by month and by quarter. This complex analysis can be achieved by creating a cube.
Cube –Is a data structure that aggregates measures/facts by every level andhierarchy in the dimensions which needs to be analysed.
For e.g. a cube for analysing product sale by region, by month is shown below.
More on Business intelligence tools from – Passionned Group DAAN VAN BEEK