The term ‘Business Intelligence’ (BI), has been around since long before computers were invented. In his book, ‘Cyclopædia of Commercial and Business Anecdotes’ (1865), Professor Richard Millar Devens describes it as the initiative to collect data and the ability to react to it based on the results collected, an explanation which is still the core definition of BI today. With terms like ‘Big Data’ and ‘Data Mining’ in play in today’s technology market, BI can play a considerable part in the process of the storage, analysis, visualisation and querying of such data.
An integral part of BI is data warehousing. In simple terms, data warehousing is the storage of data to be used for reporting and data analysis. Data stored within a data warehouse comes from source (possibly legacy) systems, which is then passed to an Operational Data Store (ODS) database for cleansing and then put through a process known as Extract, Transform, Load (ETL). During the ETL process, data is transformed and stored in a way which makes sense for business intelligence before being stored in a separate database.
Data collected for warehousing purposes must be proper and qualitative for the BI implementation to be successful. It is ideal to create a data profile, an analysis of the data available, preferably at the early stages of the project to make sure to select the best features BI offers for the business scenario.
Before implementing a BI project, one also needs to take into account the benefits of its implementation. A solution which is implemented and designed correctly provides the bigger picture of key data which is not available in any other system within the entity.
The data stored in the warehouse database is presented in clean and interactive reports, which might include charts and tables which allow the business professionals to extract the required information and also the ability to gather in one-page information from multiple systems. This makes room for a centralised view across the firm or entity.
With BI reports, even though data comes from different sources, information from each source can be presented in a uniform manner by providing a common data model regardless of the source. It can be restructured to make sense to the end business users and deliver high-level results without taking a toll on the current systems in operation.
BI portals are the entry point for users to access the data within the warehouse. Although desktop applications still exist, technology is shifting towards a browser application where the user can access the reports and any additional data analysis functionality that is supplied.
The accuracy of the data collected in the data warehouse affects the business decisions to be taken. It is crucial to make sure the data presented in the BI reports is as correct and accurate as possible - to make an intelligent business decision using business intelligence.