Live Webinar - AI: The new power for marketers
Good afternoon ladies and gentlemen. And except you are very warm welcome to today's session. I sincerely hope all of you and the best of your health. Today's panel discussion will be on the topic AI, the new powerful marketing deals, organized by email@example.com. But before we get started, I would take a couple of minutes to set the context and then bring in our speakers to get the discussion rolling for TV. Marvin is trying to position their products to their specific customers by m Simon de departments. Hence it is critical for decision makers to understand and use the power of artificial intelligence very effectively by making decisions on marketing campaigns, budgets, channels, identifying opportunities, and ultimately creating marketing USP. In today's session, we aim to unfold the face possibilities of using artificial intelligence in the marketing process. And to enlighten us further on this topic we have with us today three very experienced veterans. First we have with us sir jack Ma. Professor, I am Calcutta. professional service research work is primarily in the nexus of public policy, development and marketing is the subject matter experts in marketing analytics, AI, machine learning, artists and marketing. He has also authored several papers in journals of international recommend. Professor surbana has developed and served as the incumbent director for analysis and digital marketing programs for executed. He has also won several research awards including the outstanding research on development by global development network. Next, we have with us selection RMG. He's a management consultant and a business leader who love solving business problems with the help of data, AI and digital technologies. is a management graduate from I am Lucknow, and a postgraduate diploma in AI and machine learning from shut it down below is an experienced management consultant with extensive experience in providing AI analytics consulting to many fortune 500 companies. These four areas include digital strategy, delivery excellence, analytics, digital media, pre sales, omni channel marketing campaigns, is also visiting faculty in Ireland, Calcutta, I am Lucknow I am 30 engaged in teaching modern the tools and techniques to students is an even speaker or different industry forums on a regular day to several msmes across the country. Finally, we have a raging chocolaty Chief Business Officer sees entries. This is a topic over 20 years of experience in the technology space, often business leaders during strategy and operations leadership in challenging situations. Over the years he has involved in diverse projects related to marketing, business development, sales, new product development, as well as Innovation Ventures, which is a customer experience and contact center solution, PSTN global marketing, new product visioning and business expansion in North America amnesia is also a combination of the AI powered marketing program from IBM, Calcutta and talentsprint. But that's a very warm welcome to all the gentlemen today, a very quick announcement before we move on to our discussion further, as the session progresses, I would request our audiences to keep sharing their questions with us in the q&a tab towards the end, where we pick up the questions for our speakers to answer them exclusively for you. So without any further delay, I would sincerely testify as a server now and actually narrow in on Mr. Charging to start off today's discussion on a high pitch over to you.
Yeah, thank you so much by psyche for the introduction. And thanks, everyone for joining us this evening port to talk about artificial intelligence and machine learning. And especially thanks to Mr. lakshminarayanan, from Deloitte and also rgq, who's been who's completed this program a couple of cohorts ago. Right, he was part of the first cohort. So thank you for taking your time to join and join us and also to, you know, help other prospective students as well. Right, let me start off today's talk for giving you a brief intro. I'll just start with a brief intro give you a background, then I'll hand it over to Mr. lakshminarayana, who will take you through key aspects of what is happening in the industry in terms of, you know, the current prospects for AI and machine learning and different roles. Then I'll again, come back and I'll talk about the post that all of you are interested in. So let me start with giving you a real life illustration of what a can do. Right. So this was a news article published in January of 2020. When we were actually envisioning something, putting together a course of this sort. This was a timely article, and we actually showed it in the first batch as well. So here, you can say that this was a front page of times and you can see that a artificial intelligence actually beats the doctors at breast cancer detection, right. So we're talking about Results published in a top journal in the world, which is journal Nature, which is peer reviewed. So there are a lot of checks and balances into, you know, the results that have been published in this particular journal. Right. So here, it's kind of like, you know, you're comparing doctors who are actually treating people who have their own patients, and they couldn't actually detect in a timely manner, looking at the reading the different radiologists, they review from outputs, but a could actually beat the doctors at breast cancer detection. So this kind of gives you a better idea about how far and how advanced we become in terms of using artificial intelligence and machine learning in detecting certain patterns and anomalies that are invisible to human eye. Right, just to get you started on this. So since we're talking about, you know, the power, how he can actually add value to a marketer, right? So first few things to think about would be how he can change marketing. Right. So when we, I think we had this discussion in several cohorts, the, you know, first the initial few sessions on, you know, driverless cars, right. So people think of driverless cars, the first thing that comes to their mind is, you know, yeah, the automotive industry would change, the insurance process will change, you know, even to some extent, the traffic and you know, how traffic is being maintained the signals, the traffic, police department, all of those would actually go through a drastic change. But that's not the only field that's going to be affected. Right. So breathalyzers, for instance, the market for breathalyzers are going to go down the market for cloud computing, because people have a lot more time to commute a lot more time to be productive during commute. So people who want to actually use cloud computing services, a lot of businesses especially want might want to integrate their systems with the vehicles. So there are a lot more fields that are going to actually be affected, it's not just restricted to that automotive segment, if you think outside the box, real estate might also be affected because, you know, people don't really care about commute anymore, it's become a little bit easier. On the downside, you know, on one side, number of accidents will come down drastically, but even if one accident happens, that people are gonna react or because nobody is actually willing to take responsibility for it, when humans are at the wheel wheels, you know, take responsibility they think they are in control, and do the number of actions are really high with human drivers. Even if one accident happens with an AI that's going to create a huge negative impact as well. So, there are these different things that different aspects that one needs to take into account, when you think about where you know, we are going in terms of implementing a ml in our daily lives. So that just to give an illustration of you know, how a can change several one particular drastic change, which is a driverless car could have an impact ripple effect on almost all the industries that you can think about the second sales process. So, sales process, typically, you know, now, you would have heard, you know, Google convention on voice recognition and how particular sales process is being handled by Google AI. So, with the advent of AI, the entire sales process can go through a drastic change. We look at few examples, when we have time. In the course we look at several examples like you know, how the sales process how different conversational ASAP actually changed the way sales process actually carry on. And last but not the least shipping then shopping model, typically, what we are aware of is shopping and then shipping right? So but with the advent of AI and predictive analytics, you can ship it directly to the consumer or ship it to the barrows in anticipation of certain orders, and then the shopping can be done. That also brings in a lot of changes in the, you know, in how we look at ecommerce as a whole or how we look at even inventory management, right. So this just to give an example of the sales process. So one example is conversa ai, was actually taken over the entire industry of sales process. And, you know, they actually have a customized a for each process where it takes over the sales process. And it brings in all the sales reps, only when it's actually extremely necessary. Otherwise, it handles many of these queries, many of the lead management and lead conversions, it kind of follows up follows up with them also regularly and able to convert this one example of, you know, sales process. The other example where it's already in action is for a model, which is shipping and then shopping is Stitch Fix. So we do a case study in the process will on Stitch Fix how to use data write about consumers about what their preferences about what they share on social media, using all of those data at Stitch Fix is able to predict what kind of designs that what kind of apparel we're going to like. And they get to contract with all the customers that six months subscription model and they ship the dress first night and the customer likes, then he or she may keep the dress if they don't want they can return right? So they rely on the entity the power of data to reduce the returns because they know exactly Well, what their what the customers are going to like. So that's another example of where machine learning has played a huge role. Even Amazon is coming up with Pino tried coming up with Amazon wardrobe where they're trying to follow the same model a Stitch Fix when they will ship it first. And then you know, let the customer shop with the boyfriend. So imagine like, you know, you don't ever have to place an order, right? So, for instance, big basket or Amazon, they send you on a monthly basis based on the data that they have based on what they know about you. They're able to ship everything that you require before and and you never have to place remember a to do list or a shopping list and it just automatically comes to you at the doorstep. Enter in a canvass of ecommerce changes, right the process changes consumer behavior changes so there's going to be a lot of differences. So that's how what that is a power of AI, right. So this just to give a brief overview of three different aspects, one is driverless cars, the sales process, the third is shipping, then shopping model, all these three can have a huge impact on not just the process, but also the consumer behavior, how consumers behavior, how the babies are going to change in the future. So we do discuss all of these in a conceptual manner. Yeah. So now having given the brief background on you know, about AI, and ml and a few illustrations of how we can have an impact. Let me hand it over to Mr. lakshminarayanan. So he will talk about where the market is setting and take to take you through the different aspects of the current market. What would you like? Yeah.
Thank you, processor owner. So I may request to rail for watching the slides alone. So good afternoon, everyone. Now, it's a pleasure to connect with all of you today to speak a bit on how here is changing the market, changing the marketing world. So I'm going to touch upon two topics, one, where is the market heading. And the other one is some applications of how AI is helping in the marketing space. So moving to the next slide. So you need to understand there is a changing landscape that is happening around us so many times when we are part of the industry, we never realize that there are so many things happening around. We have lots of things that are getting automated today. Automated. Initially, when we are speaking about automation, it was more about process automation, isn't it? But today, if you notice, we are talking about automating the process plus capturing the data and using the data, can I drive better decisions? Can I understand what is happening in the market? And I understand reduced ambiguity a bit? Because it's a vuca. world outside volatile, uncertain, ambiguous. So how can you bring a bit of certainty with the help of the data that you can bring across? So yeah, machine learning is definitely playing a role in this particular direction. And it helps brands, products and decision makers to take better decisions, automation, and personalization is the new age mantra. I'm sure all of you know this. On one side, you're trying to automate every aspect of marketing, digital transformation of marketing is happening. And on the other side, we are talking about, how can we make use of the data to customize and personalize everything that we offer to the customers. So the real world of personalized marketing in the era of internet is something which is happening today, internet plus AI is bringing a revolution in the space of marketing. And of course, thanks to the advent of technology, big data, data science, machine learning algorithms, all these aspects of ensure that we're able to derive better results for the marketing people to take better decisions for the customers. So yeah, MLS enabled marketing decisions, which is transforming the marketing, you will realize that you would have seen it already you would have been on the other side of the table, trying to look at it. But now you have an opportunity to be on the other side of the table drive decisions, use AI ml data, everything to solve problems. So look at the industry reports that have come in. So 72% view AI as a business advantage. That's what PwC says. And MIT says the Titan out of 10 companies already use AI to improve their customer journeys, understand about what the customer is doing. In fact prasarana and myself we have always discussed in the class with the previous course that how behavioral targeting can be mapped today, the customer journeys because for a conventional market, you're doing a behavioral targeting slightly challenging, but thanks to the AML thanks to the data that we capture the online world today. Behavioral targeting is something which is making a difference in mapping the customer journey, segmenting them and doing a better job with the customers. And new Vantage partner says that 92% of the Fortune 1000 executives surveyed indicated that companies are accelerating the pace of the big data and AI investments. In fact, here I would like to share this popular joke, I'm sure all of you would have bought it in WhatsApp, who is driving the digital transformation. Is it the CIO, the CTO or the COVID-19. I'm sure all of you know that COVID-19 has really accelerated the digital transformation of the market today. Your organization, my organization, every organization is going through this journey, of course at a rapid pace now, thanks to the work from home culture and thanks to the new age ecommerce, your parents on my parents have started shopping online, isn't it? That's the change that we're seeing in the market today. So more into the next slide process. So it is critical for the decision makers to understand use the power of AI effectively while making decisions on marketing campaigns. Because not everyone wants to spend so much right they want to be very clear. In fact, viotti use the term surgical strikes on the customers when to do what only with the help of data the marketing people can achieve that. optimizing the budgets choosing the right channel, identifying the right potential opportunities where they can connect with the customers or prospects and creating a marketing USP everywhere. data is available today. The moment you have data from the data, you can recognize patterns from the patterns, you can take better decisions predict and do better decisions or the committee's. So that's where AI is making a difference in the market, helping organizations to achieve the right process the targeted intended results. So that's a big difference that's happening in the market today. And moving on.
So let's look at some of the successful AI applications in the marketing domain. Definitely, you have a lot to understand I'm sure some of the products or solutions that we're going to talk about, again, very well relate to it. So let's look at the first application. Netflix, of course, when we are in COVID, when we are in work from home, we cannot ignore this app. Right? All of us who have subscribed to this What have you noticed some changes in Netflix, compared to any other wotd app, I'm sure you would have seen a lot of personalization happening in the fits the movie recommendations that we get the thumbnails that you get. Everything is done with the help of data, your behavior, whatever you as a consumer, when you're using net Flix as an app, they keep monitoring your behavior. They understand what you're trying to do what you like your preferences, and then they try to bring across the personalized thumbs up thumbnails so that you can choose and watch new series and spent a lot of time in Netflix. So location scouting for movie production, Netflix uses AI ml data, and movie editing streaming quality. And in fact, I can ask a trivia question here. Do you know which series was launched by netflix purely based on the customer data? That is what they like, what kind of scenes they like, which more screenplays they would enjoy watching which actors they would enjoy watching. I'm sure some of you would have guessed it, if you have watched House of Cards, and if you're a fan of it, that's a case study for you. Oh, Netflix has used house of cards and create a house of cards with the help of the consumer data. Okay, so Netflix is something a wow factor for all the people who are watching Ott platforms. Moving on Amazon. Another popular example we always discuss segment we talked about digital marketing, Amazon as an example, marketing. Amazon is an example a driven marketing, we cannot ignore them because they are the technology pioneers who are able to bring the best of technology in the marketing world. Take recommendation engine. Today, most of the companies are able to use recommendation engine, they're able to take a cue from Amazon, I'm sure you would have noticed that whenever you buy some product, there is apparently a bunch of products that comes up. People who bought this also bought this. That's a typical recommendation engine machine learning is behind it. When you talk about future of marketing, Alexa plays a significant role, isn't it? You have Alexa to Alexa is going to play a big role processor was talking about automating everything, isn't it, the coming days, how things are going to be in the coming days. So voice search, visual search all these aspects the future of marketing, everywhere, if you notice, Alexa is going to play a big role. Amazon go stores I'm sure some of you who have been to us would have visited it. That's the future of marketing and AI ml plays a big role in that. So it's all about connecting the dots and understand how the future is going to be and ml is playing a role in it. So moving on,
dynamic pricing over color, what we are using today, almost everyone uses dynamic pricing, isn't it. Hilton uses machine learning based system to increase the demand forecasting, understand the demand supply cap, price it accordingly, armies on eBay uses dynamic pricing to offer customer prices. In fact, Uber is trying to do a bit of steady on understanding how much you're willing to pay. So on top of the demand supply gap based on the dynamic pricing on top of it, they're also trying to absorb use some data to understand how much you will be willing to pay when there is a cab requirement for you. Based on that they would like to adjust the price and ensure that they do not lose you even during the peak pricing. So lots of data is being collected. And of course with the help of machine learning it makes one does hyper personalization. I'm sure you would have seen you handle geo conflict few day a few years back, it started a few years back and you all would have seen how companies were losing customers, isn't it? In fact, that was the time when telecom companies picked up the churn analysis, why people are leaving? How can we target them? What kind of data I have about that utilization? Can I offer something exclusive based on their pattern based on their behavior, so they were able to think a lot about it. They made they understood more about the customers and they were able to personalize everything offering then packages everything they could offer it at a very, very customized level. So that's something which has been possible made possible with the help of pa and machine learning. So Yang retail definitely has a lot of role chatbots and he will say that he will say today that is a chatbot you have normal chatbots you have a power chatbots Ai powered chatbots are like digital workforce for you 24 cross seven, they can enable lead generation, they can target customers, and bring in the power of all aspects, like tracking the behavior, CRM, everything, if you bring it together, he can do wonders in carving out the products, creating better professional services for you. And of course, programmatic advertising, predictive merchandising, all these are specific solutions in the retail space, which is creating wonders for the marketers and able to drive better ROI with whatever it must want that they're making in this space. So yeah, and retail is another good one. So job prospects, if I were to quickly touch upon it, I mean, not only by this time, you would have realized that the market is full of applications, just that we are trying to realize that Okay, there is a bit of data, there is a bit of AI, a lot of AI behind all these applications, you can see that every organization is investing in EA, thanks to COVID. In the last one of years, all the companies are trying to move to look at Cloud plus AI cloud plus machine learning and how they can apply it in different domains. So the data is very clear. If you are going if you want a data driven decision maker, the destiny is very clear, 10 years down the line, everything is going to be data driven, the moment it is data driven, is going to be AI or machine learning driven, so that we can predict the future in a much better way. So every organization is investing in AI cloud, be ready for the market, be head of the curve, and try to grab the opportunities. That's one bit of a tip, I would say, because any time when you're choosing a career option, education option, you should always look at the technology that is getting into the introduction stage, and try to see how you can grab that and how you can be ahead of the curve. So definitely here is one that's something you can see from the data and moving on. So yes, most wanted, which artificial integral skills are adopters, most actively seeking. So when researchers are doing the research, of course, you can find that the focus has shifted from the researchers to the business leaders, especially in the years, he has jumped out of the labs, these days, everything is available in your mobile, connecting data for the brands collecting data for the marketeers and they are able to interpret the results, make decisions and take appropriate action. So that's something which has changed in the recent times. And every industry report that you read Gartner consulting firms, reports, you all can observe this pattern that is happening in the market. So one final, if I have to quickly summarize, I will say that AI has a huge role to play. Because data is something we are part of the information revolution today, with the information revolution, definitely, that's something which has ensured that the data is getting created, player information is coming out with the help of insights, the better predictions can be done with the help of machine learning and other technologies. So what do you process are now back to you. So for the next?
Yeah, thank you so much for the great insights. So I think he's covered most of the aspects in terms of, you know, applications of AI. So let me give you a brief overview, and then we can jump into details about the program and we can discuss a little bit more in detail about what aspects come into the program, right. So this screen, most of you would be, you know, would have seen, I mean, different products, obviously, inspired by shopping trends, lecture and and already covered it in terms of you know, how Amazon comes up with recommendation engines, one of the key facts is 35% of Amazon's revenue actually comes from recommendations that they generate, right. So one is, you know, people can shop through cat shop through the traditional browsing channel, which is categories, or they can search shop to search, or they are going to shop through recommendations that come up. So recommendations generate about 30%. And the core of recommendation engine is actually the recommendation engine is a machine learning Anyway, there's an algorithm. So you can imagine the importance of, you know, a ml when it comes to digital marketing and in the retail space as well. So Facebook, obviously, you know, there are a lot of majority of Facebook operations actually going based on AML. And you also have ami zones, which we saw their example. And the other example that was already explained to you was Netflix. So these are the you know, kind of products that we use on a regular basis. And they all actually run based on AML. And it's not very far for, you know, all the other organizations to catch up on to start understanding the importance of the power of having a lot more data and utilizing the data in a meaningful way. So that's where the market is actually headed. Right? So recommendation engines are not just, you know, unique to a Amazon's, we have several different companies, even smaller companies that started to use powerful recommendation engines and trying to, you know, generate a lot more revenue out of that from the baseline. And you also have logistics and deliveries, routing recommendations also being generated based on AML. So all of these different aspects tell you that AI plays a huge role. And as a potential to play a huge role in several companies that have not yet adopted a ml at all. This just to give you an example of you know, something that AML and marketing automation, what kind of insights they can give you like we look at this one particular transaction, you're looking at one transaction from a family and Walmart and you're seeing that they're buying a Diet Pepsi, they're buying an Atkins bar They're buying slimfast, you can understand that there's someone out there in the family, probably a female, because slimfast is targeted as females, and you have someone who's counting calories, but then you also see ice cream popsicle Cheetos, and probably you kind of understand maybe there is a child that is a kid in that particular household. So one transaction has given you to put two insights. And let's say you're tracking them over a period of time, because of the phone number, you're actually tracking them. The second transaction tells you they're buying some sort of baseball cards, this kind of confirms your hypothesis that there is a kid in the house. And then they're also buying dog treats, maybe they have a dog in the house, there's a second transaction, now put together another transaction after we do a transaction tells you the bank fishing knows, maybe they have a hobby, and they like fishing, right? They like fishing as a hobby. They're also buying marine oil, which tells you that they own a boat, right? based on just these three transactions, right, you can put together an entire DNA of a household. So that's a small family would like to go fishing and there is a dog in the household. And the kid likes basketball, baseball. So you can put together an entire idea of what kind of what group of customers do they belong to. And then you apply predictive analytics to that, and what kind of requirements are they going to have likely going to have in the future, you combine that with marketing automation, automatically identifying, you know, what automatically predicting, and then automatically reaching out to them at the right point in time, now a powerful powerful marketing tool to reach out to your customers, even before you know they reach out to the company or before the any milestone milestone change that happens in the family, the system will be able to predict and will be able to actually come up with a recommendation. There are several examples of target using several different algorithms. Just to give you an illustration of the power of data with just three transactions, were able to understand a lot more about the family. So think about 100 transactions in a year. So that you know, you know a lot more in detail about the family, what kind of guests they are getting, what, what vacations, they take a lot more information to be gained out of the transactions, the data that you're actually generating. So far, all this data have been used in a way that you know, they bought this, so let's send them a coupon, they bought a particular they typically buy a particular item. So let's send them another discount voucher for that particular item or send them offers on that particular item. But now it's more about using those transactions to understand the DNA of the family. And then to predict what exactly is going to happen in that particular group of households you put together a cluster of households are very similar. And then you're trying to predict what exactly is going to happen with that group of households? And can you reach out to them in a timely manner when they're looking for a particular product or service? So that's the basic, you know, one example of the power of a can add to detail in other fields this I'm going to keep it a little brief, because there are several fields that is actually playing a huge role nowadays. And one is financial services that obviously you can for fraud detection, you're using a ml algorithms for a credit card default, or, you know, should they offer a loan or not? How likely is the customer likely to sign up for a loan, so all of all those predictive analytics, right, so AML can play a huge role and help the financial services in a great extent. So you already have several implementations in terms of algorithmic trading in terms of chatbots. In terms of advisors, robo advisors are coming up and giving investment advice, all of this base is actually a and ml net. And if you look at healthcare again, so healthcare again, there is a huge amount of, you know, a change, huge drastic change in the healthcare space as well. So you can see the potential annual value of these different AML applications in the healthcare field with the robot assisted surgery, that's a different implementation of AML. But you can see fraud detection, there is a huge investment going in, and clinical trial participation, preliminary diagnosis, automated image diagnosis, all of these actually work based on AML. And there is a huge potential in the healthcare field, especially to apply all of these technologies now.
Going into marketing field, so this one another article that I would like to show this other example on the other example I'm going to use here, so where Harley Davidson publicized and they said that, you know, they used AML, to actually increase New York sales, because we were actually declining at one point in time around 2015 16. And they looked at the data, the amount of data that they had dumped of leads they were generating, they started using a month, and they're actually they see the sales leads increased by 2009 30%. That's a huge chunk, right? So they're able to, you know, get the right crowd and they were able to convert that crowd as well. So that MLS actually playing a huge role. Right. So why do we need to look at AML from a as a business person or as a budding CXO? Right? So why do you need to look at AI? Right? So you have a data science team, and they're really, really well qualified, and they know what they're doing in terms of, you know, executing all the algorithms and giving you the results. But as a business person, it's your responsibility to actually come up with those particular, the right questions to ask, what kind of problems can the data solve? What kind of predictive analytics can the data that I'm collecting now can? How can they help? Or what kind of data Do I need to collect so that I can actually predict certain things and I can actually put them into put them into business use, right? So those kind of insights that is a huge gap, data science, you have a data science team, we're really, really well qualified and well equipped to, you know, to execute any sort of algorithms that you can think of, but that then you have the business case where they're really not, you know, the business, my business might be in the company, right? So your you guys are the ones who are running the business and whether you can understand what are the things that are key one possible within this realm. And you're able to translate that to the data scientist team to get that work done. So that need for the translators, when you mean, but what we mean by translators is people who know the business objectives understand the business really well, for very specific roles to play. But they're not necessarily, you know, hardcore coders, they don't understand they don't have to code, but they need to be able to converse with the data scientists or even if you are engaging, external organization, you need to be able to, you know, converse with the organization to get what can be useful for your business. So that's a in marketing. So that's the starting point for any person to look at the kind of role that they can play. So this kind of tells you how they have, you know, boundaries between data scientists and domain specialists. So if you look at you know, Facebook and Instagram, you'll see a lot of means also floating in terms of what data scientists are saying, and domain specialists or marketing csoc, they're not able to, you know, they're not conversing in a really smooth manner. Right. So that's, that's exactly the kind of issue that we're trying to solve in terms of making you well aware of, you know, what are the different methods available in AML? What are its capabilities? And what can you achieve? And how can you convert that into some sort of business insight and implement that. So once you are able to, you know, grasp that particular aspect of how predictive analytics can add value? And what are the capabilities, how do we actually, you know, implement that in a particular organization, then you will be better equipped to actually be deal with the data scientists, in some cases, you might go on, and you might start coding yourself, but that's not required, right. So you can, you many of you will have a data science team in your own organization, and you will be able to get the documents actually translated in a neat manner. And you'll be able to interpret the results that they're actually given to you. And you're able to take business decisions on whether that needs to be deployed or not. So that's exactly what that's the need of the are in terms of, you know, people who understand technology, who also have domain knowledge, really strong domain knowledge, so they'd be able to execute that really well. So that's the gap in the market that we saw. So that's where this entire program comes in, where we talk about, you know, artificial intelligence powered marketing, right. So we had two cohorts over, there's a third cohort that you're planning to launch in a couple of months, in a month or so. And so here, I'm sure many of you would have gone through the entire detail. So what we're trying to achieve here is take you through the different, you know, analytics, Ascension or maturity was we call it. So right now, you know, many of the organizations in India, almost all the top organizations as well, if you look at them, they will have, you know, the past month data, they will just provide the descriptive analytics, this is what happened last month, this age group that typically bought, now this average conversions that we had in the last one month, this averaged into more people who bought a product. So that's kind of like, you know, summarizing what happened in the past. So you have some past data, you're just summarizing it. And every month they call, you know, typically, it's called a strategy team. And we'll send the reports every month to several divisions, and people look at the you know, averages, and they say, okay, it's all going in a particular way. And then that's it. That's how the data is being used, right? Typically, majority of us are stuck in the descriptive analytics space. Some companies go one step further and say, okay, age group, 35, to 40 is buying a product a lot more, can we figure out why? So that's diagnostic, that's a little bit more qualitative in nature. But people do actually ask, but this is where we are at this moment, right? But how can I use the data? Whatever has happened in the past? Can I train something I can train a machine learning algorithm? Can I predict what will happen when new customers come in? Or will they convert? Will they not convert? That is using the data one step further and go into predictive analytics? So I have the past data I know who can go to do not convert? Can I use the data and build a model so that when a new customer comes in, can they predict how likely are they to convert? Because if I know before, and how likely they're going to convert, if I'm able to predict if a particular customer is going to convert or not, that gives me a lot of information. And I can actually approach the customer with a with a different aspect every time. We even in case for landing page, right. So before even I load the landing page, if I know that this customer, there's high likelihood that they're going to convert and immediately focus on conversion, and the first fold the landing page, the first fold is going to focus on call to action at convert, right. But if I know that this person landing, coming to my landing page is not likely to convert, they need a lot more information, they're on the edge, they're 5050, you know, percentage when it comes to conversion, then I can actually give them an offer and say, you know, you need to pick us up over others and this offer that we are providing, and you load a different landing page altogether. So you kind of you when you're able to predict you're able to come up with prescriptive analytics, I can make it happen. If a customer is not likely to convert, how can I make it happen? That is prescriptive. So what we're trying to do in this course, is you know, get you to move all the way from descriptive to prescriptive. So predictive is going to play a huge role there, understanding what kind of predictions you can come up with, what are the shortfalls while you work around certain limitations. So we cover all of that. And then we give you examples of we ask you to think along the lines of how to make it happen prescriptive analytics as well. So the idea is to take you up to this maturity curve, the analytics maturity curve, Right, I hope it's clear. So it's a very, very simple, you know, concept of past data as you're just describing it right now. But what we want to do is we want to predict using the past data so that I can predict the future, what will happen? That's foresight, optimization, right? Anyway, so objectives of this program, I won't go through it in detail. But yeah, we all understand the importance of taking a data driven approach when it comes to marketing. So that's one of the main reasons why, you know, companies are looking for data scientists, they're trying to build a ramp up their data science capabilities. But there is a huge lack of talent on the other side, where people who understand data science and you know, machine learning and Raman, artificial intelligence, and also the domains, domain knowledge, so combine those two, you have a huge cover in terms of what you can achieve. So that's exactly what we call as martec professionals, people who understand the domain really well. And also the technology also really well, we're not essentially hardcore into coding, right? So we're talking about my marketing tech professionals, right. So the objective is to, you know, provide this program to current and aspiring cxos. Right, so who want to actually lead the business in this in this domain. And we have a lot of talent in our time, we have a lot of expertise and time together, we're really, you know, experts in this particular field, and when it comes to quantitative courses, and we would also combined with few people in the industry, and we will actually provide, you know, provide you will equip you with all that is necessary to understand predictive analytics to understand artificial intelligence and machine learning, and to take it one step further and go to prescriptive analytics as well. Right. So overall, in this program, we have about seven modules. And so I'll just take you through briefly, each of these modules better, okay, it's a completely interactive, we use technology, it's going to be online, we use a lot of real life case studies. So I'll give you the game. So the case studies, what case studies we're going to use, we will have few quizzes and assignments. But evaluation is, you know, part of the program, but the focus is going to be on case studies and in class discussion. And, you know, we also have something called happy hours. So what we do is in on Friday evenings, sometimes we do catch up with students and over zoom, and we have, you know, informal discussion on what is happening in their company, anything that along the lines of technology and marketing, the program duration is for six months, it's 132 hours of classroom contact, shedule will be on Sundays and Tuesdays Sundays would be two to 5pm, and Tuesday's will be 648 945. So all this information you might already have, so I'm just investigating it here. Cause details we have the initial start will be slightly non technical in nature. So that, you know, we kind of cover a few concepts in marketing few data driven marketing patients, because we have people from different backgrounds with different some of you might have postgraduate some of you might be coming directly from undergrad with a lot of work experience. To get everybody on the same ground, we start with a few sessions, like five to six sessions on data driven marketing decisions, where we kind of lay the background or the context for a multicam. And so we talk about causality. We talk about prediction, we talk about different market research. So in case you need to collect data, you also know the different methods and the pros and cons. So we cover all of this in the first module. So it's not too overwhelming in the beginning. So it's kind of like, you know, comfortable for you to get, you know, hit the ground running in terms of marketing, learning about marketing strategy and marketing decisions. So that's typically what we'd covered, this goes on for a couple of weeks on the max, two to three weeks at the max, then immediately we jump into fundamentals of AML. So here, you don't need to have math background, you don't don't need to have an IT background, we're not going to be giving you a lot of equations to learn, we're not going to give you a lot of mathematical derivations, none of those, we're going to be talking about only pure intuition behind each of these methods, how to each of these methods work? What are the intuition behind these methods? And what what are the pros and cons, right? And that's basically the fundamentals of AML, we kind of make sure that you learn the terminology, you kind of understand in simple terms, right? How do you differentiate between apples and oranges, that's typically how we talk about you know, different problems in the class. But anyway, so this module will be covered by a faculty revenues dopamine sorry, this name is poison. Landry Sham is literally written the book on artificial intelligence and marketing, which will hit the shelves, I think, this year and right. So you will be learning from the best in the domain. And he would not be going into mass too much of mathematical equations or, you know, derivation, so it's going to be purely intuition being these what is the logic being these methods. And you see, it's an, it's not necessary that you need to have any math background to understand these. So we will actually talk about a lot more from, from an intuition perspective, right. The third module, which will run parallel to the second and third module will run parallel. So we'll talk about AML. Using our we have chosen art, because art is much simpler language to learn even for beginners. So we would, you know, we would take it really slow in the beginning, because you have one session only for setup and one session for just to prepare the data. So how do I load the data so we take you step by step, and we will also give you templates so you don't have to write the code from scratch. You don't have to remember any syntax You just have to know the steps and the logic behind it, right. So we will take it up with everything in our so you will understand what are the different steps involved when I actually have to use a data and get something, some sort of insights out of that. So we will take you through every one of those aspects, different algorithms how to render, typically, there are two three line algorithms, so two lines of code, and you will be able to run that algorithm. And you don't even have to remember the code because we'll be giving the code in class and explain what needs to be changed. If you have your own data you uploaded, you just need to change the parameters, the parameters and you'll be able to execute the code. So, coding background is not necessary for this particular module. So, we will be taking you by hand and you know, explaining every single aspect through the center pro program. fourth module is AML applications in marketing here, we will jump into few more applications, we will introduce two more algorithms, but we will talk mostly from an application perspective, we have several case studies. One is, you know, Indian apparel store chain, another one is Apollo hospitals in Chennai, then net promoter score at manipal, hospitals, all of them in Indian context. So, we will also give you certain data, and I'll talk about this module and the case studies on how you need to work on it in a little while. But yeah, so AML applications in marketing will cover a few more aspects. Apart from predictive analytics, we also talk about clustering and forecasting and stuff like that. So we like if you would, you know, understand the intuition behind all these methods, even get your hands dirty with some data, and then, you know, write the code and see how it how you're able to execute, that we'll get, we'll get back to you know, how we have changed over the last 234 weeks, I'll get back to this as well. And the core focus will be on AML, and digital and social media marketing. Here, we have five case studies, along with a lot of examples that will be discussed in class selection, and will come in for AML applications in digital marketing, we will also talk about marketing, automation and marketing in the world of Alexa, then I'll be taking you through a little bit about, you know, because not all of you might be from the digital marketing background, we'll cover a little bit of background in digital marketing. And then we'll go into a little bit applied, were applying machine learning to how to, you know, deploy recommendation systems, what kind of recommendations Can you generate, you also have a case study on with data from bigbasket. We'll also be looking at VMware, which is another company from India. So we look at customer digital journey and VMware what kind of data is generated online? And what can they do with the data? And again, another example is Eureka, Forbes, they also collect data online, and how are they actually using ml to improve customer experience? So we will go through all of these case studies in this particular module. And we'll also be looking at, you know, explaining the intuition behind recommendation systems, how do they work? What are the best practices, how to go from point A to point B? So your you don't have a website? No. So how do you generate a website? And how do you go all the way up to a data driven website, so we cover end to end in that particular manner. We also have some sort of, you know, text analytics, which can be applied to social media, and user generated content. So we will cover all of those in detail in this particular module. And you will be doing five case studies over here as well. All those are real time, I mean, real life case studies from India with data from those companies. And finally, we have one short module on where the future of marketing is going beyond AML. So we so here we allow the other program that the person they wish to come in. So he will talk about blockchain. He's an expert in blockchain and smart contracts, will talk about their applications in marketing, and also talk about Internet of Things, augmented reality and virtual reality, how they can be applied in marketing in different examples are how to what to what to expect in the next few years, right. So this is basically what we have in mind Futuristic Marketing, and only in the in the second and fourth and fifth modules we have, we have four case studies, and three plus five. So there are eight case studies from the Indian context, you will also get the access to what we call us. Tech scribes, right. So from talentsprint, so basically, they are well versed in coding, so you will be reaching out you your job would be to as a group, we will be put into different groups, obviously. So you don't need to actually have a strong coding background, you don't have to code many of these things. So here, what we'll be doing is you will be reaching out to the consider them as a data science team. So you're working in a company and they are your data science team, you're going to give them instructions, they're going to write the code based on your instructions purely based on your instructions, they're going to write the code, translate that into poor, they won't write the code, they will give you the output. So you will have to figure out whether that output is acceptable or not what kind of insights can be developed for all these case studies. So models four and five, you will have access to your go to groups, and each group will have access to one particular data center resource. So consider that you know you're working in a company and you have a data science team. How do you actually Converse when you have an issue of this sort, and you have a data data and then so that will actually give you a lot more experience and you don't have to worry too much about polling. Now, of course, if you're interested in polling that yourself feeling good, you can do that. But you will have access to a data science team as well. That's the idea. So anyway, so that's the basic overview of the entire program. So what we did in the initial first cohort is once the first quarter was done, the students were still in touch with this. So we have one of our you know, ex, who completed the course in the first gordmans strategy strategy. So he will take you through what we did even after the course was done. So RJ, Can you shed a light on? You know what that is you've been doing after the course? Yeah. Okay, well, that's been sorted out, let me give you a brief of what we have done. So one of the key one major part of the part of the course is a capstone project, where you will be put into different groups, and each group will be working on, you know, either their own company's issue, or they can come up with, you know, any of the companies that they have contact with, they could take up that particular issue, and, you know, apply the learnings, what they have bought in the water received in this particular course, and then apply a model to that particular issue. So, that's a real life problem that we'll be working on. And you will be put into groups, obviously, you'll be working as a group, and you can choose, you know, either to work with one company or a couple of companies, depending on how many taxes you have. And what happens here is people work on that particular that particular capstone project at the end of the course, they make the presentations, the capstone project will also have Yeah.
So, Professor, as you were mentioning about what we post the course. So, first of all, the course is very engaging. And we got a very good perspective from this course, about application of AI in marketing, because here whenever it is perceived, it is more perceived from a product perspective, from a inside the organization perspective, but outside the organization from sales and marketing, how it can be useful. So we are taking a capstone project, and because can project is short duration, so we were looking at a broad brick and mortar company, and how they can apply marketing and the AI in marketing, to understand acting integration and how it boosts the sales. And we did it but it was a limited exercise inside the course. And we took that and we are continued writing and working more on detail about that project, which is actually committing towards kind of a white paper. And it is also a very good application of what we learned in the class, and then trying to do the same thing in the real world and taking it forward towards a real applicant so that we can see how this predictions that is the AI model is providing really how it is going to boost the sales. Yeah.
Can you share a little bit more about your experience in the course. And the prospect is students? Yeah. So thanks for sharing that. So what will so the capstone project, like you mentioned is a is a short duration project, it's like three to four months that we'll be working on that project. And as a group, you'll actually you'll be interacting with the group members. And you'll have to, you know, work on it on your own, and you need to come up with a solution, you need to come up with an architecture and present the present results in the last few sessions, of course, right. So overall, there is a lot of interaction during the entire course with the faculty, and among your peers. And peer learning is actually one of the major aspects of this particular course. Because in the first part, also, there are a lot of people who are still in touch and who actually work together on several, several different marketing issues as well. So anyway, so that's the basic overview of the course and what you can expect from this. If there are any questions. Can you see their questions? Can you please list them out? By cycling? So I'll turn it over to you. Yeah.
Yeah, sure. So thank you, Professor Sarah. Now Mr. Chakravarthy, and lakshminarayana for such an insightful session, may lead garnered quite some intriguing and interesting learnings today. So we'll move on to a q&a session where a lot of questions has come in from various industry verticals. I keep the patients open for us so that any one of you can take it up on the first day, I have two questions for Mr. Chocolate came from the audience specifically for him. The first question that we have is I'm a sales professional with 10 years of experience in b2b and b2c vertical. So how will AI help me in my career?
So see more interesting that you mentioned that you have 10 years experience in the sales field. So because we have several profiles, from the sales with exactly 10 years of experience in Cohort Two, as well as in Cohort One. So what we talked about where this can add value is, let's see, even in b2b, it's not just limited to b2c, even if it's in b2b, or b2b, ca ml can add a lot of value in terms of predictive analytics. So how much more your sales process could be streamlined and could be made more effective, if we can actually figure out which of your leads are actually going to convert? What is the likelihood that they will convert with the high accuracy, right? So if you know beforehand, someone is actually on the fence 5050, right. So then you know that you need to make an offer to them, they are cool, they're a warm lead, and you need to actually nurture them a little bit more and you need to convert, on the other hand, you know, that certain group of certain leads are actually the probability of conversion is 90 to 95. Right? Then you automatically know that you don't have to push any offer to them. All you need to do is get the conversion done, right. So AML can add a lot of value in terms of optimizing your efforts in terms of how to how to actually maximize the conversions, right? You know, before and particularly promising the look, the probability of conversion, the base predicting is about point two, right? Then it kind of gives an information that a cold lead, and you're not going to put in a lot of a lot of effort, you're going to give them a lot more information for them to go through, and try to nurture them to it to the next level, right. So that kind of adds a lot of value when you can predict something about your about your sales conversions. That's number one. Number two, you can also pay forecast sales right to a great to a much better accuracy level, when you're actually using machine learning to understand from different territories, what exactly are likely to happen, and different different marketing mix, how it's going to actually result in, what kind of sales are they going to result in, so you'll be able to predict a lot more along those lines as well. So it's not just one application, there is a wider range of applications you can think of when it comes to sales roles. Right. So there are a lot of other examples, but let me keep it short. So I think that that gives you some idea about predictive analytics, and that can add value.
We have such Apogee with us once again. So if I may request you to see something so that we get to know.
Yeah, can you hear me now?
Yeah. So I'll just share the patient that we have exclusively for us one of our audiences interested to know that how was your experience throughout the program by you join this program on the first place? And also Is there any prerequisite in terms of technical knowledge to join this program?
So, first of all, we all know that AI in business is becoming more of a norm now, it is not an option anymore. So, and being into a business, where we are providing software technology for customer experience. So, AI is a very important thing and we are doing it for some time. But what was very interesting to understand that how AI can help in the in the realm of sales and marketing, because traditional industries and even new new Gen industries know all oral do not use it for for their own sales and marketing, as well as more or so ecommerce companies have been doing it now for a long time. But restaurant industry is not doing so that was the Indiana interest to join the course to understand how AI can be applied in sales and marketing. In terms of the prerequisite, or technical understanding, yeah, one thing is very important, which I would say, you know, even people may deny is that the person should have no interest, or at least the inquisitiveness to understand this technology, or it's not necessary that you have to learn the programming, though, in this programming is very simple using our and and that can be done easily. But still the inquisitiveness to know that technology is one predictors that I would expect people to have any other questions Misaki?
No, so I have a couple of questions. So as I mentioned, I'll keep it open for us. Anyone can take it up. Thank you for sharing your experience. With that. Move on to the next questions. The next question is he is a technical subject, but I'm not a technical person. So how would I be able to grasp the technical aspects of this program?
Yeah, so like I mentioned, so you don't have to have a master an IT background, you don't have to have coding experience. So we will talk mostly in terms of, you know, intuition behind these methods. So, we are not expecting you to become experts, we are not expecting to become data scientists who are going to go into coding. So what we would want you to know is what are the different methods that are available? Right? So if you look at 20 years ago, customer relationship management, CRM was a buzzword, and people are thinking, How do I learn CRM? How do I, you know, will companies adopt CRM or not? Now it's kind of like, you know, it's, it's everywhere without CRM, nothing works, right. So it's more in from that perspective. So you don't have to actually program and build a CRM system, but you need to know how to use a CRM system. So think along those lines, right. So later on, as we, as the market progresses, you know, a lot of these terms will become a will become mainstream. And you would have to know what these terms would mean. And we would actually be covering those mostly in a non technical manner, so that you understand the logic and the intuition behind them, and how to apply them. So more focus will be on interpreting an application. That is one aspect number two, we will be taking you through our right from the basics, even to install our How to get it done. And we will have a dedicated session to kill installer. And we will be giving you the coding template. It's like you know, you have five, six lines of code, you have it with you, you just change the name of the file and you will be able to run any algorithm there. That's it. So we'd be actually hand holding you over there in terms of initial stages of getting you comfortable with the technology. The course is actually designed for people from non technical background. So having said that, modules four and five We will be giving you real life cases, and they might be a little bit complex. So what you are, that's where we give you access to a data science team, right? So you will be talking to the data science team. And we'll be telling them your requirements. So similar to what your role is right now. So you look at your current business, you look at the data that you have this is the customer data that they have, how do we use it? So now that you know the terminologies? Can I go to the data science team and tell them can you execute and give me the results. So they might use any technology they want, they might or any platform they want, they will execute and give you the results, you should be able to based on the course you should be able to read it and say, you know, this is the kind of decision I need to take based on the results come this kind of, you know, prescriptive insight I can come up with, so that will be a job. So it will be more like, you know, training for actually implementing it in real life. So you don't need to have a lot of training, coding background or technical background at all. Right. I hope that answers your question.
Yeah. It's a great opportunity to learn actually. So moving on to an expression of business analyst and my firm. So how AI in marketing was useful
to you, like I mentioned, so I showed you the, to this particular question. So I showed you the different analytics, Ascension or analytics maturity, right. So when you're a business analyst, analyst, look back and see, where exactly do most of your work line. So it might be in the descriptive, or diagnostic analytics, or at the max, you might go into, you know, basically describing what has happened in in the past, or as a business analyst, you might be working with the requirements of your clients and trying to, you know, convert that into a platform that could actually be so there are different roles that business analysts may play in any of these roles, if you're able to have a good grip on AML. So even if it's a client facing role, right, so you are actually when you look at the client's business, you will be able to come up with several applications of machine learning, and you'll be able to build a better rapport and better, you know, better kind of grasp the requirements of the clients, and you will be able to suggest better outcomes as well. That is one side of the coin, the other side of the coin is you'll be able to translate those requirements effectively to your technical team as well, in terms of what kind of output the client really needs. That is from from one perspective, as a business analyst, as a business intelligence analyst, you might be working in descriptive analytics. So can I move to predictive analytics? So both these would actually be addressed by this particular course anyway?
Moving on, the next question is, according to you, what are the biggest challenges in applying artificial intelligence in marketing?
Yeah. lecture and you want to take your question, take this one. Yeah, I can take the processor, man. Thank you.
So to quickly answer this question, our primary challenge, I would say that the availability of the data, the quality data, so this is a common problem across domains, I wouldn't say only marketing. But of course, with the digital online marketing coming into picture, we are capturing a lot of data. But still, the data availability is something which is a challenge. One second thing, I would always describe EA machine learning as kind of angry, Godzilla, I'm sure if you have seen if your mind is good, or 80 skips, you all remember the Godzilla movie at our deeds. So yeah, he's always like Angry Godzilla looking for data more and more data, better the algorithms better the accuracy of the model. So with more the data, the obviously the meta models are going to give you better results over a period of time. So we are in the nascent stages, the organizations are trying to capture the data at this point of time. So I would say that data capturing is something which is very, very important. And organizations are trying to put it in processes to capture the data, use it with other raw data internally captured already within the systems like AARP CRM and other data and trying to make more sense out of it. So that's one challenge. And the second thing is obviously, as processor I mentioned, the domain plus technology knowledge, in fact out to answer connected with the previous answer that he was talking about. Finally, how do you apply technology to solve business problems? That perspective is something which is kept in the market slightly. We are engineers, we are business people, somebody who has to bridge the gap between techno functional people who can bridge the gap. So that's another challenge that we see in the market today. So courses like this will definitely help people to become specialized in technology, solving the business problems with the help of data and technology. So these are the two primary challenges I can think about. Thank you so much.
So we have three more questions, which are very industry specific. So the next question is, how is the pharma industry using AI to increase ROI and ensure better CRM effectiveness? Right compared to conventional methods?
Yeah, so specifically in pharma, right, so I've worked with the farmer for some time as a consultant, at least. So I'm aware on the ground while we exactly they were using data and the initial stages, like you mentioned, the initial stage data would be every single division that they have. The strategy team would send the monthly report like you know, this is the average sales that we could get from the IMS Health and this is a, you know, average prescriptions that we get. So all those would be descriptive in nature. So where AML could play a role is number one, it can predict or forecast the demand. So at the stock level, what would be the demand at the retail level? What would be the demand? And if we change the marketing mix, so what exactly would be the impact on sales. So number one is coming up with demand forecast. So there are obviously a machine learning algorithms can add a lot more value, because their accuracy of forecast is really, really high. And number two is, since we don't have direct to consumer advertising in India, right, so the main vehicle of, you know, promoting something would be through a medical representatives, right? So that territory allocation, territory optimization, and what kind of response can I expect from different doctors, who are the doctors who are my core, who are currently who are the new doctors that I've acquired, and who are likely to go into the poor doctor group of mine who can be evangelists, for me, all those predictions, right can be done effectively with the AML, because pharma companies have a huge amount of data. So sometimes they do their own market research and collect data on prescriptions done by different doctors, sometimes they take the buy the data from other organizations that provide them this prescription level data, so AML. So rather than just describing what has happened in the market, you if you're able to apply AML, that you'll be able to come up with a lot of predictions that could actually optimize your marketing efforts, marketing efforts, typically be your detailing and you know the order of presentation of your molecules that also could be determined using AML. So to that extent, the data could be useful. So there are several applications the applications are endless. So once you are aware of conceptually, what are the different methods? What are the different algorithms and what are the capabilities, right, then you will realize that the possibilities are endless, because the amount of data generated in pharma is quite high, are held by any pharma company is actually quite rich in nature. So what I just mentioned, were also a few examples. One, predicting the sales in the stock is level number two, optimizing your marketing efforts. Number three, would be predicting prediction about doctors at the individual level at the individual doctor level, what is the percentage prediction prescriptions Can you predict who are likely to be evangelists? So all those could actually be done with AML to create a much greater accuracy level?
Yeah. The next question is, I'm in the media advertising sales so far in this program helped me in media or advertising industry?
Yeah. If you want to take it off, right?
You can go ahead. Yeah,
yeah. So in the media space, so glad that you asked one of the in Cohort One as it might be aware, in Cohort One, we had a person from z z channel, who was actually it was working with z consulting or working with the channel covered the scenarios, he was actually working with us. And he was able to apply this particular learnings from this particular course. And he was able to actually take the meta data from different ads, like who are the key characters in these ads? Who are the key? What was the kind of scenario so you could prepare an entire data set? They also had the viewership data on how many people actually viewed it? How many seconds did they view it, and they were able to build a model and predict what kind of headline what kind of caption and what kind of content and what kind of characters are typically getting a lot more views in Facebook, what on what kind of the source characteristics are they are getting a lot more views in, let's say, Instagram, so they were able to figure out build a model. And they're able to come up with a model, which could tell them what combination actually resulted in highest number of views. So cross a five second threshold cross 15, second threshold, so when they were coming up with a new clip, they could actually plug it into the model and see what kind of a target segment would result in, you know, more than five seconds of so when you create a Facebook audience and push it to them, you can actually predict and say, This Facebook audience, you're much more likely to cross it five second threshold. So this model was actually built by the student as a capstone project in in the first cohort. So that was a very good application in the media and advertising space. And, again, like I mentioned, the applications are limitless. So depending on the context that you're talking about, you'll be able to apply that in that particular scenario. There's just one example anyway. Yeah.
Two more quick questions. We are in a chemical distribution company. So how will this program help us? And what kind of infrastructure Do we need to run AI? What kind of sorry, infrastructure Do we need to run vi?
C, in terms of infrastructure, so I'll ask lehkonen to respond in the chemical industry, he will talk about preventive maintenance, and we'll talk about the other terms in the b2b segment. But in terms of infrastructure, so all you need is a system right? That's it's all you need is normal, you know, any Intel system that could actually handle our in our studio, which is any system that comes out in the market, right now. So infrastructure wise, when it comes to, you know, modeling, you don't need a huge computing power or anything, your normal world desktop, a laptop will be able to handle it, right? In terms of deploying it in real time. You have cloud services, where you can, you know, where you can get this uploaded, and then the results will be automatically sent to the different systems. So those are all simple to manage. So we're not talking about infrastructure in terms of a huge server and Huge workstations, infrastructure is bad, minimal. We are not building robots, by the way. So let me make it clear. You're not building robots, we're using the data to make predictions we're using machine learning to predict, make predictions about what could happen in the future. And based on the data that you have. So infrastructure wise, you don't need a lot your system is more than sufficient to start working on different AML applications. Yeah. lakshminarayana.
Yeah. Yeah. So if you have to look at the practical applications, in your context, I will, I will try to keep it in simple terms. At the end of the day, any data that you have across domains, finance, marketing, HR operations, any data that you have, you are able to derive a bit of intelligence, use your experience. And with the help of that experience, you have intelligence to derive insights from it. That's what we call it as gut feeling right? In your way, gut feeling is also an insight coming from your experience based on the data. So whatever you are able to derive from your experience in the industry today, from the data, the same thing mission can do with the help of AI. As prasarana rightly pointed out, bringing cloud bringing the power of data on the cloud, with all the analytics and machine learning applications on top of it, it can create better insights, it can help you to derive better versus solutions or 20 re IQ or I mean, if I were to generalize it for manufacturing assets, as serve it pointed out, it's more about digital twin as a concept, predictive maintenance, you can bring that across demand forecasting everything you would be able to do it with the help of data. So if we have capture data is your data in your organization in some form it excel sheet or in a proper systems, you have the power to use machine learning to take organization to the next level, no doubt about it. Yeah.
And to add to that, right, so usually what happens is people have this misconception that you know, I work in a b2b setting, so can machine learning be useful for me. So we have had people from mission from b2b, they're in the first cohort, as well as in the second cohort, and their projects have been taken up with the capstone project where the groups and they have been able to implement that as well. So just to add to what Miss Lakshman and said, so you can, when you have, you know, when you let's say, you go through this course, and then you try to figure out how to actually implement it. Right? So we talk about something called a bottom up approach, right? You go with the data that is currently there in the organization. So what is the data that is neatly available? And what can you do with the data from the learning that you already have in the course, when you start implementing it and showing the results to the management, right in terms of how the process can be improved? That's what we call is bottom up approach, because you already have the data in that you're trying to reverse engineer and figure out what can be done with the data? And how can it add value to the business at the top level. So that would be a good starting point to make a change in the organization. The other way is to look at the top down approach and say, you know, this is a business objective. This is the kind of predictive analytics that can help me with the business, these are the predictions I need to make for these marriage predictions, I need this kind of data, that will be a top down approach, which requires a different kind of mindset altogether. So what we what we make you well accustomed to is both approaches. So you start with the data, identify what is the pattern in the data? What can you predict? And how can we employ? How can you connect it to the business objective? other one is look at the business objective, what data do you need? How do you collect the data, so we kind of cover both these approaches as well. So in many cases, people, you know, use the data in the capstone project itself. And you know, you have a group of people working on it for three months, and you'll be able to come up with a lot more richer insights on how to actually apply that in your company. Right, any?
And the last question that we have is, can we use Python code instead of art?
Yeah. So the reason we this major question that we get all the time, the reason we chose R is people who are not, who have not had any coding background, will be able to pick up art on the first day, because it's not an object oriented programming, it's very simple to, you know, two, three lines, and you'll be able to pick it up. Again, the idea here is to not meet your data scientists, but to get you, you know, get you some sort of hands on when it comes to implementing these algorithms. So you get a feel of what the data scientists do. The reason we have gone with our simple, the coding is very, very simple. And you can make keep it really, really simple. And if you already have Python knowledge, you would be able to implement that on your own. Having said that, when I mentioned you have data scientists access to data scientists in the fourth and fifth module, the data scientists will actually be solving the problems that you're mentioning to them in Python only envy, they'll be giving you the results in a proper format in a in a table or in a in a tabular format. So actually, the, the technology behind it is actually Python in the model four and five, the idea is to train you in R, but when you talk about AR we are not talking about you know, you know, remembering the codes. So what we will be telling you is, you know, when there is missing data, this is how you handle it, right? So whether it's Python, or the way you handle it should be the same, right? So should I drop them? Should I actually do something else with them to actually include them in the analysis. So those, that's the kind of insights you will get out of the modules. Number two is, you know, I load the data, I deal with the missing values, then I deal with, you know, what kind of algorithm I need to provide. So that's a step by step that will actually teach you are is just a vehicle to run that type. So you can follow the same steps and you will be able to run that in Python as well. The results will be the same, the answers will be the same, irrespective of whether you're going to use Python or R But we've chosen our The reason is, is much more simple for beginners to, you know, not get intimidated by too much of coding language here and there. It's just simple English. So that's why the reason we went to that, but if you're comfortable with Python and you're able to resolve, you know, solve the problems in Python, you can very well go into Word with that. And just to reiterate, models four and five, you will have a data science team and the data science team will work on Python, and they will give you the results. So it doesn't matter what is going on in the background, your job will be to give them the right instructions, get the output and make the interpretation. So yeah,
thank you that despite some amount of information that we have, you know, garnered in the last one. So with that, find two speakers once again for taking our time and being with us today. With that we come to the end of today's webinar. Polar audiences found this event to be pretty insightful and engaging as much as we have enjoyed putting it all together. But before we have to anomic time, which is all our audience's safety and the rest of it, have a great evening ahead of you.
Watch the entire interview here https://www.youtube.com/watch?v=FF6FvvB3JtY