FEATURE : ANALYTICS
move over their data warehouses into the cloud are limiting themselves .”
According to Gill , deploying the right business analytics strategy is not something that you can achieve overnight . “ However , there are specific best practices that can help ensure that organisations are successful with their business analytics implementation
CIOS NEED TO CHOOSE THE TYPE OF BUSINESS ANALYTICS BASED ON THE STAGE OF THE WORKFLOW AND THE REQUIREMENT OF DATA ANALYSIS .
as much as possible . These best practices include : establishing the business use case and the goal before implementing business analytics .
Types of analytics
Given the many types of business analytics in the market today , how should CIOs pick the type that suits their business environment ?
Gill said the three primary types of analytics – descriptive analytics , predictive analytics and prescriptive analytics – are interconnected solutions that help companies leverage data . “ Each type of business analytics provides a different kind of insight and choosing what to employ depends on the business situation at hand ,” he said . “ Descriptive analytics is the interpretation of historical data to discover trends and patterns while predictive analytics is the usage of statistics to predict future outcomes . Prescriptive analytics on the other hand is the application of testing and other techniques to establish what outcome will produce the best result in a particular scenario .”
Gill explained that CIOs need to choose the type of business analytics based on the stage of the workflow and the requirement of data analysis . “ While the different kinds of analytics are usually implemented in stages , no one type of analytics is better than the other ,” he said .
Determining the specific criteria for success and failure , validating models using the chosen criteria for success and failure and designing a methodology , narrowing down data , and identifying the internal and external factors needed for making a precise projection ,” he said .
Rakesh Jayaprakash , Product Manager , ManageEngine , said business analytics can be broadly split into descriptive , diagnostic , predictive and prescriptive analytics . “ Descriptive analytics involves interpretation of data in an understandable form , diagnostic analytics helps answer why something happened , predictive analytics helps forecast the most likely outcome based on historical data and prescriptive analytics suggests favourable outcomes for various courses of action ,” he said .
Jayaprakash said for an organisation beginning its analytics journey , a good starting point would be to get descriptive analytics right because that forms the basis for the other stages . “ Humans are good at looking at historical data and figuring out why something turned out the way it did ; however , they ’ re not great at forecasting the future . Forecasting future trends helps a lot in making crucial decisions ,” he said . “ Predictive analytics can fill this gap . These are the two main areas CIOs should focus on while designing their analytics strategy and choosing the right analytics platform that suits their business needs .”
According to Ghazi , the transformation to a datadriven enterprise often starts in different places for different businesses – from the management board to individual departments . “ Data analysts might see the growing need to expand self-service analytics within the company , for example . Companies should see
36 INTELLIGENTCIO AFRICA www . intelligentcio . com