WhatsApp Chat Sentiment Analysis in R

Introduction to Descriptive Analytics

"The simplest class of analytics, one that allows you to condense big data into smaller, more useful nuggets of information." 
- Dr. Michael Wu  

Introduction to Descriptive Analytics

Suppose you're a college student and you are interested in measuring the anxiety level of your classmates. Your class consists of 40 students and you rate every individual's anxiety level on a scale of 1 to 10, 1 being the least ie, no anxiety and 10 being the most anxious.

You collect all the 40 ratings and evaluate them.

Now, how will you summarise this data?

These days a lot of collection and analysis of big data is outsourced to third party companies who specialize in these things such a scanner data in groceries stores.

Descriptive analytics is the heart of the quantitative analysis. 90% of the organizations use this technique. It is used to describe the basic features of data in a study.

Descriptive analytics can be defined in many ways, Descriptive analytics is the way of linking the market to the firm through decisions. It's the information that's needed to make actionable decisions. It's a principle of systematically collecting and interpreting data. It juggles raw data from multiple data sources to give valuable insights into the past.

In general descriptive analytics is a way of fetching good data. It can also be described as the choice between data and decisions that managers have to make for good analytics.

Types of descriptive analytics?

Exploratory Research

Consider a brand manager who is looking at his brand sales and he notices that the sales have suddenly started to drop. He might now wonder why are they dropping?

Is it because the customers' choices have changed or the customers are benefitting from the rival company?

There could be a variety of reasons for the sales to drop. All the manager will have to do is explore the reasons. Exploratory research is the research on theoretical and hypothetical data. It involves gathering general information about the data.

Thus here, the brand manager will have to look into his data and gain valuable information from it. He will then understand the reason behind the sales dropping and can then rectify the situation.

Exploratory research is done to develop initial hunches or insights, provide broad guidelines on what to test, etc.8

Descriptive Research

Consider the same example about the brand manager.

Let's say that the brand manager is interested in finding his customer share of wallet ie, how much are the customers spending on the competing companies.

Also, he would want to know who his customers are.

Such type of questions requires understanding how much customers are spending on his products or the products.

The descriptive research attempts to explore and explain information often discovered through exploratory research. It is the act of exploring the thing in the dark, creating a fuller picture of what you are looking at.

One psychological example is the use of CT scans to describe the living brain. It allows us to have the clearest picture of the most complicated organ, the living brain.

Causal Research

Consider an example wherein you are to change the landing page of your website.

You would now want to determine how this change will affect the customer behavior. Would it increase or decrease the access rate of the website?

Such type of questions are defined under causal research wherein you determine the impact of changes made to the customers.

Causal research requires systematic collection of data by carefully understanding and examining how to collect the data.

It is the investigation into an issue or topic that looks at the effect of one thing on another.

Causal research can be used in a business environment to quantify the effect that a change to its present operations will have on its future production levels to assist in the business planning process


Summarizing past events such as regional sales, customer attrition, or success of marketing campaigns.

Tabulation of social metrics such as Facebook likes, Tweets, or followers.

Descriptive analytics is used to determine an individual's social media record . This includes a number of friends on Facebook, number of likes per post, number of followers on Instagram, likes received on Instagram, followers, and reposts on twitter etc.

Reporting of general trends like hot travel destinations or news trends. This includes analyzing data and determining various offers based on climate and oh their factors, one can determine appropriate travels destinations or based on the data one can determine the news that is currently popular.

For example, for the upcoming political elections in Karnataka, through analyzing and creating polls, one can determine the party that is most probable to rule or dominate.

Industrial applications :

Descriptive Analytics has made great strides in supply chain mapping manufacturing plant sensors GPS vehicle tracking to gather, organize and view past events.

Descriptive Analytics and Business Intelligence:

Descriptive analytics is one of the most important technique in a business intelligence a company. Return on invested capital (ROIC) is a descriptive analytic technique. T is implemented by taking three parameters — net income, dividends and total capital.

Thus descriptive analytics will use these and turn the return on investment into an easily understandable percentage that can be used to compare one company's performance with respect to other companies.

Generally speaking, the larger and more complex a company is, the more descriptive analytics it will use to measure its performance.