Analytics

We live in a world of data where data is produced every millisecond. Many business organisations have humongous amount of data, which as it is does not produce any value. This is because facts and figures are futile if we can’t extract any valuable insights from it. In order to get information from a data set, advanced level of computational tools and expertise are essential. This process is termed Analytics which can be defined as the systematic study of data or statistics using computational tools and techniques.

Analytics is widely used in the areas of Finance, Marketing and HR. Some of the examples are forecasting future financial performance, assessing the risk of investments, determining the mix of skillsets that are mandatory in accomplishing a project, hiring talents of the highest quality, ensuring better use of advertising budgets, assessing customer satisfaction and loyalty and the like. Apart from business, the medical field also resorts to analytics in diagnosing and prescribing better treatments.

Based on the outcome that a particular analysis could provide, analytics is mainly divided into four namely – Descriptive Analytics, Diagnostic Analytics, Predictive Analytics and Prescriptive Analytics.

Descriptive Analysis: As the name suggests, it deals with describing and interpreting the data using data aggregation and data mining techniques. This interpretation can be in the form of diagrams, tables, charts or dashboards. It mainly answers “What” a dataset is trying to convey. This analysis can be considered as the starting point of analytics, as any analysis has to start with understanding the data that we have in our hands. Description of the sales and the revenue of a product, surveys and feedbacks, reporting general trends and patterns, etc are examples of Descriptive analysis.

Diagnostic Analysis: This type of analyses is for discovering the root cause, anomality or flaws in a process using data. To identify this, it makes use of data discovery and data mining techniques. It asks the question “Why” to a data set. For example, cause of a product failure, fall in employee performance, etc

Predictive Analysis: This analysis attempts to fill gaps in a data or in other words predicts what can we expect in future, which means it questions the data on “what to expect”. This is done by feeding previous data set into machine learning models that identifies specific trends and patterns in the data set. This is followed by applying this model into current data to predict what future outcomes could be. For example, risk assessment, forecasting the sales of a product, surge in demand for a product etc

Prescriptive Analysis: Prescriptive analysis prescribes a remedy .i.e., what could be done in case something happens. Artificial Intelligence systems is a perfect example for this method of analysis. It churns data to study a data and use them to prescribe. It suggests various courses of action and prescribes corresponding information on the result of each action. For example, price modelling, optimizing production and inventory making in a supply chain etc.