Blog

Data Science No Comments

Data Science Use Cases in Retail : Part 2

Welcome back to Data science use cases in retail series.
To get the overview of how data science is helping in retail, please refer to the first article Data Science Use Cases in Retail : S1E1

In this article we will see how Data science is helping in understanding customer behavior.

Results from Mckinsey study demonstrate what this means for businesses: After a positive customer experience, more than 85 percent of customers purchased more. After a negative experience, more than 70 percent purchased less. So getting this wrong can prove a costly exercise.
To understand customer behaviour following things are done:

  • Customer sentiment analysis
  • Merchandising via visual channels

Customer sentiment analysis:

Its been there for a long time however with Data science tools it has become quick and less costly. Now without explicitly asking customer about their experience , machine learning algorithm can determine whether customer is happy or not.
Sentiment analytics uses language processing. In language processing, it tracks the “positive” and “negative” words which determines sentiments of the customer. This anlaysis become a background for services improvement.
The analysts perform sentiment analysis on the basis of:

  • Natural language processing (NLP)
  • Text analysis to extract defining positive
  • Neutral or negative sentiments.


The algorithms go through all the meaningful layers of speech. Based on that customers are categorized.
The output of sentiment analysis is the sentiment rating and the overall sentiment of the text.

Merchandising via visual channels:

It’s all about how we are promoting the merchandise. Smart merchandising includes:

  • Which products to show
  • Keeping product list fresh
  • Attractive packaging


Now, to do this lot of data science goes behind the scene.
What Data science merchandising mechanisms does is :


In the next article we will see how data science is helping in Influencing customer decision.

Comment