Insights on Data and Analytics by Gourab Nath

 


What are the leadership skills needed to drive effective change in organizations to achieve data and analytics excellence?


While the core Data Scientists has a highly technical role the leaders are expected to play very different roles. We expect them to be a good business leader having a good understanding of what Data Science can do. I believe the skills include (1) being able to explore problems and demonstrate how data science may be applied to gain better insights, (2) understanding business strategies, (3) being able to talk more about what data science can do and what their teams are doing to bring in more projects, (4) To be able to explore various fields of Data Sciences and explore which fields are useful for solving what problems, (5) being able to motivate a team of good Data Scientist, to specify a few of them.


Can artificial emotional intelligence be a substitute for real emotional intelligence?


Anita is a young and determined saleswoman. She was selling handmade chocolates. One day when she was sitting inside a coffee shop she observed three young girls enjoying each other’s company over cups of coffee and cakes. Identifying some happy moods and a potential opportunity to sell some of her chocolates she walked to their table and said “Hey ladies! How often do you have chocolates?”.


“Quite often, why??”, said one of them with a smile on her face while two other girls kept staring at her with curious eyes.


“Then I have got something very interesting for you”, said Anita with delightful eyes and with a wonderful smile on her face. She put her hand inside her bag to take out a few samples while maintaining that same smile on her face and the glitters in her eyes and put them on the table. She continued saying, “Chocolates! Absolutely new! Tastes amazing and all made with natural sugar. So a perfect solution for all chocolate lover who wants to save some calories.”


“Haha! That’s nice”, said one of the girls as she laughed. “But then that won’t taste the same, right?”


“Yes! Maybe a bit different but these pieces of deep browns taste really amazing!”, said Anita with a confident voice.  “Why don’t you just try one. Please, I insist. And if you don’t like it I will take them away. No questions asked!”, said Anita with a persuasive voice and before anyone could say anything she cut opened one of the samples and offered it to them.


The girls took the sample to take a bite and when they almost started to like it Anita with a wonderful smile said, “So, which ones would you like to take?”.


This was usual for Anita. She did not find it difficult to sell the chocolates. She is a confident person with high emotional quotient and therefore she has got a very persuasive nature. Can this be substituted using artificial emotional intelligence? Well, maybe in the future but today this branch of AI is at an experimental stage. We do not yet have any device which has got such high emotional quotient.


Now, let’s imagine an application of artificial emotional intelligence. A fashion retail outlet displayed a number of digital posters. The posters are artificial emotional intelligence enabled. The digital posters display the items from the new arrival categories. It is enabled to recognize human emotions using a face recognition algorithm. When a potential customer looks at the poster the algorithm understands if the customer likes the poster or not. If the customer doesn’t like it then it generates a different poster. Can we employee humans to do the same task? Of course, we can. However, that won’t be cost-effective and of course not scalable. So, when we talk about substituting real emotional intelligence with artificial emotional intelligence we need to understand when and where it should be done and can be done. At this point in time, the scope of an entire substitution is limited.


What role does analytics play in measuring the engagement level of employees?


Before we discuss this, let us understand what we mean by employee engagement. The word employment engagement may be perceived in different ways by different people because of its subjective nature. ‘Vijay is more engaged at work than Amit.’, said the HR. Does that mean Vijay is happier at work than Amit? Or, he is more energized at work than Amit? If so, then how can we quantify happiness or energetic level of an employee? These are intangible and are difficult to measure. 


On the other hand, by employee engagement if we mean the number of hours worked, hours of overtime, etc. then we have some quantifiable variables. Employee engagement, however, should not be seen as a single number which can be measured based on one of these variables. It is a concept which is measured using well-designed surveys, which are usually designed by good psychometricians. In such engagement surveys, employees are usually enumerated using a series of questions. The aggregate of their responses is taken as a measure of their work engagement. UWES-9 and UWES-3 are examples of such measures. Analytics has some different roles to play here. It can help us to identify factors which may affect work engagement, e.g., high increment rates may increase employee engagement, or it may help us to identify the effect of work engagement, e.g. increase in employee engagement may increase the company’s turnover rate.


What does sales analytics mean?


Sales analytics is a branch of Business Analytics which focus on developing essential methods to discover relevant insights related to market and unlock insights to increase revenue and profitability and improve brand perception. Precisely, it helps us to answer questions like – what is the size of the market for the products and services being sold? Is there sufficient growth potential? (Market size analytics). How many ski jackets should we order for the upcoming ski season? (Demand Forecasting). Is the market growing or declining or stagnant? (Market Trend Analytics). What should be the price of the products? Should we give discounts? How much should be discounts? (Pricing analytics). Who are the right customers? Who are the most valuable customers? How to retain customers? Which customers should be targeted and with which products? (Customer segmentation analytics). Which are the most effective sales and marketing channels? (Marketing and Sales Channel Analytics), etc.


How can data scientists and analysts improve their communication skills?


First and foremost, learn to interpret almost everything in the simplest possible way. There is a tendency of Data Scientists and Analysts to interpret their results in a very technical way, mostly maybe because of the kind of job roles they usually have. However, being a solution provider one must understand how to put things in simple ways and interpret the results more from the business point of view. I also believe that storytelling is an essential skill which every Data Scientists and Analyst must-have. Data always speak whenever you communicate with it and the data scientists and analysts know the best possible ways to do that. 

However, how one put those pieces of information across really makes a big difference. 


Therefore, one must give special attention to be better at this skill. Next comes presentation. This is the time when you share your ideas, thoughts and results with your peers or clients. So, you have the spotlight on yourself and therefore it is a huge opportunity for you to create some impression. Plan out your presentation with a storyline. Choice of sequence can make a huge difference. Moreover, try to include different elements like facts, humour, suspense, opinions, etc., otherwise, your presentation may look dull and monotonous. Be short and precise. Be clear, be genuine, transparent and real. Your verbal communication plays an important role and you must take special care of it. However, understand that communication is not just what you speak or write. It’s how your entire self responds to the environment around you.



Gourab is an Assistant Professor at Praxis Business School, Bangalore. He has 5 years of experience in teaching Data Science. His focus areas include Statistics, Machine Learning and Natural Language Processing (NLP). He has guided multiple Capstone Projects and his research interest is in the areas of Applied NLP. He has been a visiting lecturer across Colleges and Universities and had delivered Management Development Programs for Senior Leadership teams of large corporates.

Post a comment

0 Comments