The purpose of analytics is to derive actionable insights from data, helping make
smarter decisions, and thus bringing in competitive advantage for an organization.
The approach we take to architect and design strategies to derive these insights var-
intelligence (BI) architecture in place that can efficiently ingest and analyze large and
diverse data sets.
Here are three paradigms of data analytics:
• Descriptive analytics
• Predictive analytics
• Prescriptive analytics
Descriptive analytics is all about taking data and analyzing past actions for intuitions
to help identify an approach for the future. Historical data related to past failures or
success is collected and mined/processed for the actual reasoning behind a success
The variable that is measured is usually referred to as a dependent variable and
all other variables that determine its value or state are referred to as independent
variables. Every independent variable in data is analyzed to identify its relationship
with the dependent variable. For example, a customer buying or not buying a fitness
product is determined by factors such as BMI, demographic details, and age. Here,
buying or not buying of the fitness product is the dependent variable and all the
factors that determine this decision are the independent variables.
This analysis can be categorized as a post-mortem process and involves collection
of quantitative data, and usually provides hindsight that can be used for future or pre-
Descriptive models quantify dependencies and relationships in data in a way that is
often used to categorize prospects. Descriptive modeling techniques focus on all as-
pects of the data as against a single outcome and simulate all possible dependen-
cies. Some descriptive models and statistics do make assumptions about the data
being measured. For example, the assumption that the data set is normally distrib-
uted, or that the data set is linear.