Four Types of Data Analytics

In today’s information age , organizations have access to more forms of data than ever before, with new data and information coming from multiple sources by the minute.For different stages of business analytics huge amount of data is processed at various steps.

There are four main categories of data analytics — descriptive, diagnostic, predictive and prescriptive. These four types together answer everything a company needs to know- from what’s going on in the company to what solutions to be adopted for optimizing the functions.

• Descriptive Data Analytics : What happened:
  This is the process of describing or summarizing the existing data using existing business intelligence tools to better understand what is going on or what has happened.This type of analytics involves analyzing historical data and identifying Key Performance Indicators that can be part of the business objective.This can be further separated into two categories: ad hoc reporting and canned reports. A canned report is one that has been designed previously and contains information around a given subject. e.g. monthly report sent by the advertising team that details performance metrics on your latest ad efforts.Ad hoc reports, on the other hand, are specifically designed , useful for obtaining more in-depth information about a specific query. e.g. social media profile looking at the types of people who’ve liked your page along with what other pages in your industry they’ve liked as well as any other engagement and demographic information.
  Techniques like data aggregation, data mining, clustering and/or summary statistics all serve to provide analytics that describe a past state.
• Diagnostic Analytics: Why it happened:
  Diagnostic data analytics is the process of examining data to understand cause and event, or why something happened. The result of the analysis is often an analytic dashboard.n particular, diagnostic data analytics help answer why something occurred. Diagnostic Data Analysis is broken down into two specific categories: discover and alerts and query and drill-downs.
  i. Query and drill-downs are what you’ll use to get more detail from a report. For example, let’s say that one of your sales reps closed significantly fewer deals last month. A drill-down could show fewer work days, reminding you that they had used 2 weeks vacation that month explaining the dip.
  ii. Discover and alerts can be used to be notified of a potential issue beforehand, such as alerting you to a low amount of man hours which could result in a dip in closed deals. You could also use diagnostic data analytics to “discover” information like who the best candidate for a new position at your company is.
  Diagnostic analytics can also provide guidance by helping to:
  i. Identify outliers : e.g. a sudden drop in sales or an explosion in website traffic that can’t be explained and may indicate a need for additional examination.
  ii. Isolate patterns : in this case the analysts may need to look outside the existing data-set to identify the source of the pattern. e.g. a sudden drop in sales may have stemmed from the launch of a competitor product.
  iii. Uncover relationships : using more complex analytics, analysts may employ probability theory, regression analysis, or time series to isolate cause and effect relationships.
• Predictive Analytics: What might happen if:
  Predictive data analytics emphasizes on predicting the possible outcome using statistical models and machine learning techniques. Predictive analytics focuses on determining “what will happen” in the future based on analysis of historical data. Prediction is accomplished by applying techniques such as principle components analysis, sensitivity analysis and training algorithms for classification and regression on historical data. Predictive analytics can also provide a diagnosis i.e. server as a tool for diagnostic analytics.
• Prescriptive Analytics: How to make it happen:
  Prescriptive analytics is a type of predictive analytics that is used to recommend one or more course of action on analyzing the data. Prescriptive analytics builds on predictive analytics by helping determine recommended (prescribed) actions based on desired potential (predicted) outcomes, helping organizations achieve their business objectives.
  Prescriptive analytics models are constantly “learning” through feedback mechanisms to continuously analyze action and event relationships and recommend the optimal solution. By simulating the solution, prescriptive analytics can examine all the key performance criteria to ensure the outcome would achieve the correct metric goals before anything is implemented.
  Artificial intelligence, machine learning and neural network algorithms are often employed to support prescriptive analytics by helping to make specific suggestions based on nuanced patterns and perceptions of organizational goals, limitations and influencing factors.