How would you define Analytics & Why is it important?
We’re living in an increasingly data-driven, digital world. There’s so much data generated every day from multiple sources. Analytics helps us make sense of this data and more. But Analytics is a big word, and it’s not always easy to understand what it means.
So to help us all understand analytics, in this episode of Qcast, we’ll discuss what analytics is and why it is important for your business.
Sahil Brij Malhotra: Hi, I'm Sahil Brij Malhotra, director of business development at Qualetics Data Machines. Welcome to Qcast. Where we bring in our leaders, experts, and guests to talk about AI challenges and opportunities. Today we are going to interact with Sumanth Vakada, founder and CEO at Qualetics. And we are going to talk to him about the impact of AI on business growth.
Hi, Sumanth. Welcome to Qcast. Today you know we would want to know from you, how do you define analytics and why is it so important?
Sumanth Vakada: Sure, Sahil. So let me answer that question now using a parallel. There's an interesting fact that I recently learned. Which is that the skin is the largest organ in the human body. The reason I bring up that example is because the skin in front of us all day, that's the first thing that we look at ourselves or when we look at somebody else. However, it is invisible to our eyes as an independent organ and we look past it. The same way, data is present all around us whether we are using it or whether we are producing it. In the context of an organization, tons of data is being produced by the organization and is being used by that organization continuously, however, the amount of information that can be extracted from the data is probably minuscule.
That is where the concept of dark data comes in. So if you want to understand the stories that your data is going to tell you then analytics is the first step for it. Analytics to me is the process of extracting useful information from the data that you possess and if the question is why do we need to hear the stories, that's an entirely different topic. But I hope we all agree that understanding the stories or hearing the stories that your data is about to tell you is important, then analytics is the first step and it is the fundamental step for you to start on so that you can achieve bigger things down the line with artificial intelligence or machine learning.
Sahil Brij Malhotra: So, Sumanth, that's a very interesting analogy that you bring up between you know data and the human skin, of course, as skin is all-pervasive so is data. So could you bring up an illustrative example in terms of how analytics moves towards the next step that is AI (artificial intelligence) and also towards ML (machine learning)?
Sumanth Vakada: Sure, Sahil. I can give you two examples, one that is 20 years old and another one that is more recent. I'll start with the first example, where I was involved directly in that project.
20 years ago I was working on a small project for a cigarette company in India. So this company was making a very popular brand of cigarettes and this was heavily in demand so they were working continuously throughout the week to produce that cigarette in eight-hour shifts throughout the day. The challenge that they were seeing was that the cost of packaging or the cost of producing the cigarettes was high and the problem they zeroed down was on the wastage of the cigarettes. What they noticed was that with each batch of cigarettes that were being produced they were seeing a high amount of cigarettes going into the trash, being wasted, and not going to the packaging. So they were trying to solve this problem using data but, I would say in a crude manner. There was a human factory worker standing at the end of the packaging line and taking account of all the cigarettes that would go into the packaging versus going into the waste. what they would do then is take these numbers and at the end of the day or at the end of the week they would try to identify the trends and patterns that would show what particular batch of the production or what time of the year, what day of the week, were they seeing the highest amount of wastage. And once they understood the detail they would then go about remediating the problem.
So my involvement in this project was to digitize that whole mechanism where instead of collecting that data on a sheet of paper we would put that into a computer system and then that system would produce the reports automatically. The reason I bring up that analogy is because of the question that you asked, is analytics important? Well even though we didn't call it analytics then, this was an analytics project where we were trying to extract information out of raw data. Raw data is in its purest form basically. So we were trying to extract information from this raw data and then identify the problem areas and the opportunities that we could implement to improve the whole process. So data analytics has always been around, it's always important. As I mentioned we may not have been calling it analytics 20 years ago but it has been the fundamental process of a business trying to understand its own inner workings and it's still being used today. So analytics to me, in my opinion, is an extremely important step and it's always important.
Now Sahil, let me come up with a more recent example. So let's take the case of Amazon. Amazon as you know is the global leader of shipping products virtually through e-commerce and what they have identified is that they are producing a lot of packaging waste to be precise they are producing 915,000 tons of packaging waste every year, and this is a huge problem. Not only from an economical standpoint but even from an environmental impact standpoint as well. So the greater you're able to reduce that packaging waste the more environmentally friendly, your product is or you know your processes. So again analytics here is important because analytics helped Amazon identify that this is a problem. They identified that packaging is where they are they're seeing the biggest impact in their process to make it more environmentally friendly and then as I mentioned that was the first step which led to successive implementations of the improvisation of the product. Right, so what they did was using that information they created a model that would help them identify the right kind of packaging for the product that is being shipped.
And how did they do that? They used the information that they have access to, which is the product information and the description to help them identify the right kind of dimensions, the right kind of weightage, and the right kind of material that can support the fragility of the product to recommend the right kind of overall packaging that needs to be used when that product is shipped. Now implementing that is a scaling problem because you know for the millions of products that amazon is shipping it would be virtually impossible to have that many number of humans that would identify that in real-time and then use the right packaging so what they did was using machine learning and AI they were able to create a model that would train on all this information and then as the product starts rolling down the packaging line, this model would identify the product and then identify the packaging that needs to be used for the product. Now whether it is fully automated or if it's just even supplying that information to a human worker there, it greatly improves the overall efficiency of the product because they're able to make that decision in real-time and then pack it and ship it. So I hope they implement this solution worldwide so that their goal is achieved.
But the more important fact that I wanted to highlight here is how amazon used analytics to identify a problem, and then started implementing a solution to that problem with the successful implementation of the information derived from analytics. What is also interesting is the fact that the problem that we were trying to solve 20 years ago is the same problem Amazon is also trying to solve, what has changed in the 20 years is the number of tools that are available to collect useful information. Where somebody was collecting that information manually 20 years ago, now we have sensors and OCR tools or image processing tools that can extract information and convert it into digital data that can be immediately used by machine learning models to provide some kind of resolution. So I hope this trend continues and we see a lot more growth and acceleration in the technologies that we are using, but again the key here is you have to start with analytics and analytics is the fundamental step to understand the story your data is telling you so that you can create the right kind of solution from that information.
Sahil Brij Malhotra: That's really interesting that you come up with two very different examples from two very different points in time. How yesterday’s statistics have turned into analytics and we can safely say that analytics is the first step or the prime step to move on to the journey of implementing sophisticated techniques such as artificial intelligence or machine learning.
Sumanth Vakada: Thank you, Sahil.
Sahil Brij Malhotra: Thank you Sumanth for your insightful thoughts on analytics and its impact on businesses.
We just heard from Sumanth Vakada, founder and CEO at Qualetics Data Machines and we will be back with more thoughts from Sumanth, and our other team members soon.