 Hello and welcome people. My name is Vashar Chinchankar. I am the Chief Digital Officer at Madison and at our media agency So before we move on to our topic, which is data-driven mindsets Let me just do a quick introduction of the lovely panel that we have today We've got Sujata She's Sujata has over a couple of Decades of experience in sales and marketing with companies like PNG, Coke, Nokia, Google And now she heads marketing for India and South Asia at Visa We've got Pankaj Gupta, senior executive vice president sales and CMO HDFC live So we've got a couple of them coming in from PFSI as a as an industry with wide experience from other industries as well. We also have Sagar Boke head of marketing food businesses of Tata Amnip, consumer products Sagar is a proud guy to sell Deshka Namak, Tata Namak And he also is responsible for the business of Tata Sampan Masales So he's one of the great guys that we've managed to get on board today We are from the publishing side. We've got Tarika Soni Tarika heads commercial strategy and Admonitization for Snapchat India and she's been over a year She's previously worked with Twitter for about five years and managed video suits at With RTP in programmatic space. We've got one person who's really into this data marketing stuff for more than about seven eight years So that's going to be a lovely piece to hear from them and Shweta Shweta leads marketing at APAC level for Tabula. Tabula is the world's largest publishing and the native platform So welcome on board Tarika Shweta. You guys It's going to be wonderful to listen to you guys from from the publishing perspective and Last but not least we've got Archana Sinha. She's a senior marketing directors at sales force one of the technology partners and she brings in a couple of decades of B2B marketing in the past. She's worked with leadership roles like Oracle India IBM and Microsoft So that's going to be a massively Fantastic view coming in from various different Industry perspectives and before I move on I like to wish Archana a very very happy birthday So happy birthday to you Archana. Thank you so much. Thank you So we have The best marketing minds coming out of BFSI FMCG publishing and technology company I think I think that's a wonderful You know at a to get into this, right? Let's Moving on before what sessions, you know, I'd like to just talk about what is happening in this industry over a last decade We've witnessed a seismic shift of how our mom our marketing has outpaced Conversations have been changing from GRPs SOVs SOEs To programmatic and audience cohorts There's a significant amount of marketing budgets which are shifting from traditional media to data-driven marketing Right and digital as seen implementation of machine learning and Google and Facebook are one of the finest examples Of how they really turn out data and they learn out of it While India remains to be a duopoly market with walled gardens Take maybe a search or social for an example most of the trash Machines Get most data and then the machine learning algorithm generates real great insight about the users Right, so that's what the wood marketing is more headed towards. It's also a trip to a business to a large industries And when you when I talk about a trip to the businesses, you know, it's all about the funnel marketing There's stock of the funnel which is more awareness and equipment of new users and there's a bottom of the funnel very hard One-on-one sharp targeted Marketing really comes to the player and optimizes to hear more generally converges So this is just to up still upgrade their cell their cells and their teams. Yeah, we live up to this dynamic change All right, so that's that's what has really happening in this Context and so let's find out from this eminent panel where the data-driven marketing mindsets can really help them make informed decisions quickly deliver higher marketing outputs understanding consumer journeys and audience trends and many more such questions, right? so let's let's move on to our first question and So I'd like to just open up this question more from the publishing side and then later move on to Okay, and the marketing side of what does it really take for companies to create strong data driven cultures, right? So from a team's point of view, how do you really you know that they want to be I'm gonna bring this question more into the part One is the culture of how do you really build the culture inside your company? How do you really upskill your managers? How do you really bring them to speed? With the capability and the data knowledge that stuff and and on the other hand another challenge remains is how do you really even? You know get your management to invest more into these data products, right? I can imagine the publishing side and the and the technology guys who really live out of data in and out Must not be extremely difficult But it will be very exciting to hear a point of from marketing side of the businesses. So I let this question open to you guys Start with Swita. Do you want to step up? Thank you Vishal. So from tabula's angle Um tabula being a publisher and a native advertising platform data is really our bread and butter And I think really to create a data driven culture. It starts from the top So we all like to joke at tabula that our president has a huge poster in his office It's custom-made and it says in God we trust but for everyone else, please bring data So that's pretty much how Tabula's culture is so internally we really measure everything Nothing no actions get approved without data to back it up usually three pieces from three different sources internal and external and Everything is measured. So from the basics of business like ROI business metrics Even our HR function is measured. So we have an MIT graduate who leads our people analytics team based in New York So internally it's just drilled into our DNA Externally since we're an ad platform Data is that definitely live and die for us. That's why we have Brands that always spend with us and that's the ROI the heart of ROI is what drives? retention out there and It's just the part of the culture driven down from the top its role modeled a lot of our Meetings start with something provocative like a data nugget. For example, we recently had a F1 pit stop that was world-breaking. It was under two seconds So the question to all the tabula is was how many of tabula videos? Do you think it was seen in those one point eight seconds and we had to guess so it's very it's just Something that's driven from the top and something in our DNA. I Can I can agree with that? I think from our point of view on snapchat again. We are a data heavy Advertising platform like any other tech giants that exist out there. So it's clearly similar to Smith's story It's quite drilled in in our DNA and I want to take it a step forward The way we look at data is data scheme. The rest of it is all just an opinion. That's the way we use the language internally An opinion is an opinion everyone has a right for it, but data is scientific, right? So the way we use it first, I want to start first with advertising and how we use that and I know Back in the day when marketing with FNCG or any other vertical for that matter existed It was a lot to do with gut checks. You knew your audience better than anyone else did, right? That was the notion that everyone on the marketing side knew like you know your audience the best That's the assumption now the interesting thing is when you go into an advertising world Which is so skewed towards data driving that a lot of the times you have to a lot of my exercises are myth-busting exercises where I talk to gamers or I talk to FMCG partners and I say I know you're specifically Targeting a 35 plus for a reason But open your horizon a little more and maybe test a 25 year old Maybe test over a 35 year old. So I'll try and expand the horizon Again with the nugget of proving post campaign when they get that data that actually 24 to 35 May outperform your audience that you think is your audience, right? So I think the interesting thing we've learned and I've learned this a lot a lot with my past experience as well I've been in the app ecosystem For a long time now and I've realized that especially when gaming became the most sophisticated app Vertical right and you would think well somebody like me. I'm fond of gaming I'm a casual gamer, but I'm also fond of cooking and baking now if I spoke to a gamer and said Target interests based on cooking for a 30 plus year old woman. They would laugh at me and say, no, that's not my audience At all. That's absolutely not my audience. Why would I go after cooking? They're not they're not like Interested in puzzle games or I would only target games, right now that They can actually see once they've enabled a Wider audience to reach out to that they actually drives ROI. I think that horizon has changed over time Where it's less gut-driven and more data-driven. So the appetite to test has increased Which is interesting So you could be a gamer and you could be fond of cooking and that's fine I think the advertisers have learned to understand that what they understand about their audience is one thing But data should speak louder. So that's one way in the advertising stack, right? Now not just that we are collecting data like for specifically a snapchat in my case We don't have followers or You know you follow me and I follow you and I like food on I'm like, you know following us Mexican handle because I like Mexican food That's not how we're collecting data, right? Like you telling me is not authentic enough for us So what we're doing on snapchat is actually building cohorts of data set based on the snaps you're sending your friends So we are saying you're actually taking a picture of the Mexican food that you're eating outside of Mexican restaurant I believe you're you're actually a Mexican fan So I think that's the way we are pulling data based on Actuality because pictures don't lie, which is the data context that we're pulling from so that's one angle to think of Is it's not about what you're telling me? It's actually what I'm pulling and understanding from your behavior So it's a lot more organic and authentic So that's one way of thinking about marketing and then in terms of the culture, which is the last piece I want to touch on a data-driven mindset when we do rigorous hiring right like pro I'm completely aligned with what Smith had said top down But even from a bottom-up standpoint when I'm hiring teams right now There is a point in their final stages of the interview where we'll give them a case study of trying to solve a problem for a brand and it is Essentially just data. That's all they're doing is crunching numbers and telling me if they increase the CPM by this much and Increase that daily budget by X. What is the click-through ratio that they would be able to see? What is the install ratio they would be able to see and for how long and those are really technical data-driven questions Right like you have I don't necessarily believe you have to come from a stat or a math background But you need to understand that when you're coming into a dynamic Data-driven environment so that you're not left behind but you can actually cope with that dynamic dynamic nature of the business that we're living in That's wonderful. Thanks, Tareqa I'd love to hear from Pankaj as well as Sujata, you know your side Because you guys both of you all come from heart for BFSI. You all eat, breathe memos It's all about sales and you know more into Visa stuff or the insurance area that you guys get into How important is it for your marketing and I'm assuming you know Sujata for you There's been a dramatic change. I mean you've been into PNG and coke and this language of GRPs right and from there now into a visa where you really using data So how's been your journey over there? Thanks Vishal. So firstly I would say as marketers we have to balance You know the whole in gut feel and how we use insights along with what we use with data-driven and more and more I see especially now in the BFSI It's really critical to have a data-driven approach Because what we've seen you know with more and more campaigns is when we have data-driven Personalization the returns on investment in marketing are far far higher. So that is something that's you know, we've seen campaign over campaign So therefore it is critical to have that. However, I would always say that it's very important to balance that You know with understanding what your consumer insights are because you know if you just go with data And I understand where Sujata and Dharika are coming from but as marketers when you think about customers and when you think about consumers You have to balance both Having said that like we said, you know in marketing earlier We used to do a lot more just generic above the line a lot of spray and pray as I would like to say and Now we're seeing the merits of doing data-driven. So how does the organization equip itself and in visa? So we actually have a really great internal learning and upskilling portal It's called visa university and you'll be interested to know we actually have data driven marketing as from grade 1 to 10 So people can actually upskill themselves So whether you come in as a you know fresh management trainee or you come in at a senior level You go and assess what level you're at and you're able to upskill yourselves Of course thing in their external courses and stuff that people take We read a lot of reports that come out But I think the two big things that really contribute to developing a data-driven mindset for marketing is number one like I said doing a combination of insights and understanding a consumer and doing data-driven marketing and Using that and at the truth test is actually to see what your return on investments on your marketing spend So that's what we do and we see that as and more we do more and more data-driven Personalization to our campaigns. We're getting higher MRO is I Wanted to get in here and I represent the life insurance industry and you look at the structure of the life insurance industry we are in the business of getting customers for the very long term and We are in the business of taking a small amount of premium and paying huge amount of claims at a certain point of time the way the Industry works in life insurance is that data permeates all aspects of our operations One of the most important persons in the life insurance company is our Actuary and the actual team and actually is in a way are the real experts in using manipulating and manipulating in a positive way using and Implementing various action items based upon a very deep solid understanding of the data So when I say data permeates all aspects of our company's operations It starts from even before a customer is acquired into our system Which is the customer segment that we should go after? Whatever the customer segment we go after in what geography do we go after that customer? What is the message we take to that customer so that the message is registered if you are going for a certain customer? What are the sales channel to use if we are approaching a customer with a different offer versus rather a b a Versus a b offer which offer makes sense to which customer so the take-up rate might be different for different offers if I'm processing that customer through my system whether I do it digitally or physically or Through a phone-based Approach which approach works better with which customer once the customer is within my setup In terms of engagement with a customer What is the customer looking for as a message from me as part of the life cycle that I have with the customer? Is it a message about Something that is relevant to his child because he's worried about his child's education and what he can do Through our solutions for his child or is he rather thinking about his retirement planning or is he worried about his health situation and health Cost escalation all of these elements become very important in our industry the customer comes for the real long term so something called persistency is very very critical and persistency again is Determined with a lot of scientific data usage Including aspects like which product has a higher persistency Which cohorts have a higher persistency with geographies have got impacted In one way or the other with channel partners are acquiring a certain type of a customer for us and so on and so forth So not just marketing when I talk about data in life insurance context It permeates all aspects of our operations if I have to come to marketing side in particular So from a marketing perspective alone There are so many different things where data becomes a very critical aspect of our decision-making process One of the projects that my team led recently was Deciding the regional language strategy for the company now when we decide the regional language strategy for the company What is right in Tamil Nadu may not be right in Karnataka What is right in Chennai may not be right in Salem and what is right in Bengal may not be the right Solution in Orissa. So, how do we decide each one of those is there a trend that we can observe because of the ticket sizes Is there a trend we can observe because of the product stream? Is there a trend based upon whether the customer signed his initial application in a regional language versus the English language Is there a difference when we approach the customer with a written communication versus a phone communication or Versus let us say formal document like a policy document versus a promotional communication, which I'm sending to him on email So when I'm talking about data permeating all aspects of the life insurance companies operations It is a very very exciting journey for us. And the way I have seen it evolve is Somebody mentioned that culture and beginning from the top is very very important and I fully agree with that I think Svita mentioned that it has to begin from the top because the type of sophistication that we need in our industry in our context about data as a very high level sophistication and That means that we need the right quality right talent in our system Which it essentially is not very easy to get we need to have that technological backbone to manipulate Again in a positive way that much of data and if you are able to create that backbone that means investments If you want to operate at a certain level that against means having that Ability to manage that much of data flowing through our organizations can we manage the peak data that is flowing through our organization Which may be a hundred thousand transactions in a in a second or in a couple of seconds So all of those things mean that it begins from the top along with the culture the hiring process the talent process as well as the Investment approval that are needed at a very very top And I'm so glad that you touched upon You know all the aspects of my question that I was trying to ask now I want you to wear a marketing hat and tell me that For you and your business, you know generating leads on a regular basis is still To me warm performance lead and trip somewhere for you, right? So what portion of your budgets? Do you allocate more towards? You know data driven marketing versus your You know your your generic traditional way of gutsy way of advertising. I would say almost 80 to 90% of our decision-making is Born by some data that are underlying our decision-making process There is a subjectivity in certain parts of the decision-making for example Whether you go with a creative versus be creative is a function of not just which creative will work better But which creative may have in today's social media's active context a level of risk Which we have to be accepting of if you want to go ahead with a certain type of a creative and that part of Subjectivity when it comes to taking a certain risk for example with a creative strategy That is the gut part or the decision-making part but if I have to decide if I have a product a to sell which Creative I go with I know with a fair degree of accuracy through my data analysis Which creative will work better if I have to resign my media plan. I know where my customer is What are the demographics? I am going after which is the media being consumed more? Is it print? Is it TV? Is it digital? Is it hot star versus YouTube all of data is available with me So I would say if I have to quantify 80 to 90 percent of my decision-making in marketing today is database But yes, there is a 10 to 20 percent element of subjectivity Which primarily comes to either creative differences or a risk type of an approach What approach we take to risk with a certain creative or a certain concept or a certain communication? Fantastic. Thank you. Thank you. I'm gonna jump in on Pankaj's point actually very quickly On the creative example, I think that's where publishers like me and tabula would probably play a role where we have dedicated data science teams in house that are built from ground up where To encourage a more efficient gut for Pankaj's business, for example Is where we will actually run blinded case studies internally which are first-party driven and externally through third parties so that when the likes of Insurance companies or any marketer for that matter wants to know a versus b as an a b desk Either we give them our internal first-party data, which are blinded a lot of the times That's more for our knowledge than asked by the advertiser But if it went asked by the advertiser it is always driven by the third party for like the likes of kantha Romilly brown or whoever nielson Where they'll actually do these brand-news studies where It's very hard right like for even for insurance when you're sitting between tv and digital I understand not just digital but even mobile to that matter is very intimate Because my device is just my device my husband and then my kids so and none of them are using it right like so It's very intimate data. So that's fun. But what is the guarantee that the ad That converted on snapchat wasn't actually an offshoot of tv and that's where these brand lift studies really act as that That primer as to whether it's worth investing in this digital platform. Are you just leaking your dollars right? So I think that's where we play a role in giving that And sarika thanks for touching on the major ability attribution But we want to do a deep dive on that and we'll come to you a bit later But before I come to you I love saga to step in and saga you run a fantastic show of You know one of the most popular brand called as talus salt. It's more modern trade It's more general trade you reach out to tear to tear fees. You have premium brands in that you have the masalas And you come from a very very hard for typical fmc g where you know grp's also trip your business, but you've been taking some great moves You've invested into a lot of data lakes you invested into your own e-commerce There's a lot of content that's getting created and there's a lot of data. That's also getting generated So what's your vision and I want you to touch upon how have you been successfully managed to really build that culture within your system Keeping in mind that you've come from You know an fmc g fast-moving environment Yeah, so I will answer the second part of your question first I think the first thing everything starts really with the culture So other than having commitment from top, which I think everyone has spoken about There are two things which we really need to Which I think in a company like ours and largely would apply to most of the fmc g companies That need to be done. The first thing is I think you have to commit You have to show There is there are going to benefits on the table like say for example if you're investing in data management platform It's not only about the money It's also about the time that you're investing right if you show that you're going to sell 15 percent of your money you are going to Make your funnel much sharper. You're actually you're turning you're almost running a funnel to a cylinder, right? I think that's when people will start buying in in what you believe That's the first thing. I think you have to be committed. You have to show that it works. That's the first thing I think the second thing that That is important is really that if you look at the marketer, so the fmc g is really where So the marketing originally started right and that's an advantage as well as that's a disadvantage Because most of the people also have been trained in you know television in terms of you know We say there are three things in fmc g. First thing you have a great product You get a great communication and you put in the And you put that commercial on the best possible Tenue television channels and get into the outlets and either you succeed or you don't right? I think that's changing And hence investing in in educating your team because you know, there are so much of changes happening Happening here. I think that's that's a very important and that education Also has to happen both upwards and downwards. I think these are the two things that you if you really do You know that that that really would help us to become looking at like, you know, it's it's it's it's not very obvious That someone selling a salt and a spice to be really investing so much in data I would say that it really should start from what you really would want to do So if you look at salt per se, I think I think data has to be subset of your business strategy So if you're looking looking at salt, so tata salt, obviously they're just 75 grow indians So every single person in this country almost use this I mean, you know every every two out of three business people in this country use this tata salt Maybe the rule of data is not as strong there. But if you're looking at Tata salt light, which is a low sodium salt which is targeted people Who can develop lifestyle disease? I think that's a very sharp Audience that you're talking about I think there is a rule for a rule for data and that's where we really invest significant amount of money in there Other thing that we are also realizing is that Brands the way they are getting popular the changing Pre digital and post digital it was a one way traffic essentially Marketer was disseminating a message and consumer had an option to accept or reject it Consumer never had an option to respond what he or she paid about it, right? I think with digital that digital that is fundamentally a change So one significant advantage that most of the marketers will have Post digital is to build a community of fans. So if you look at the whole neutral corner strategy, I think I would say We are possibly one of the pioneers to really create a content performance platform there We are on five to six likes visitors coming every single month Is to really create a club of fan and that can only happen if your data So, you know a customer who has been coming regularly what he or she is interested in what kind of content you should give How you should create opportunities what for that consumer to buy That really comes from our belief that if you are tomorrow, we are able to create 100,000 fans of data something as a brand Our ability to create advocacy and sale is going to be significantly higher So I would say it it finally boils down to what you really would want to do and I think it has to be strategy first And execution would would really follow Well said Sagar well said absolutely appreciate that point of view I want to ask you one specific very very specific question. You've done serious levels of investments When it comes to sharp targeting and uh, you in fact invested more even in the space of your own DMP, right? How difficult or how easy does it become to you know, uh, convince your management or convince your managers To really come up to the speed of understanding the importance and the value of your investments Well, I think every single every every new thing almost always Passes through this phase, right? I mean ignorance Uh, ridicule and finally acceptance, right? I think it's a process if you really believe in it It really pays off. I would say uh, that it's been around three four years. We have been on the journey. Uh Where we currently are possibly we are we are significantly ahead of most of the peers You know in the industry, I would say I would say upwards and downwards everywhere. I think people appreciate the value that that that you are adding Uh, and finally, I think the numbers and the results I think that's that's one Uh, one I wouldn't say mantra but one experience that I would like to share with anyone who is in financial industry Where it is believed that, you know, digital doesn't really have data doesn't have a role to play Uh, I think start with small steps demonstrate and and that's when the magic starts Fantastic well said now i'm going to move to archana Archana, you come from a serious tech background you've been into this b2b you work with organizations like microsoft You've been living that culture you've been living with that culture not just with your managers But also your top management, right? But having said that you still wear a marketing hat Right and you still talk to various different organizations Or you know, you still need to promote your product, uh, which is still more data later So how do you uh, you know look at that uh piece of uh business where you look at it data to sell your own data products Sure. Thanks vishal. I think I'll answer the question in two parts. I think the first, uh, you know bit which is really important is that Um, I think as a company sales force believes completely in building a data first culture themselves So internally we have a data first culture Uh, this is something that I've seen evolve over the years working with various different companies Um, you know seeing this culture evolved I think there are three things which are very important. One is building a data first culture Uh, second looking at marketing as performance based. So, uh, I think there's a lot of arguments against and for by marketing should be performance based And the third aspect being how do you actually build a data driven team? I think, uh The culture of uh of an organization making that uh, you know data heavy or data oriented is really about You know driving data driven decision making Making this the norm. So I think this creates a culture which encourages critical thinking and curiosity Um, and uh, you know people at every level, uh, whether you know, you speak to junior most or senior most person in the organization All of these people have conversations that start, uh with data and they actually develop their own data skills through practice and application I think, uh, this is another thing that's really critical. Uh, so at sales force We actually have a very very strong onboarding model where, uh, you know people when they Um, you know join the company they actually go through and I'm talking about the marketing environment go through very rigorous You know onboarding training, uh, where they actually get to understand how we use our own solutions Um, these solutions to create campaigns which perform How do you actually think about measurement as you're not start? I think that's that's really important. Uh, and that's something that happens to ensure in sales force. Um, I think another thing which is foundational is that this actually requires a self-service model of data Where people can actually access the data they need Balance that with security and governance and that's something that is built into our platform Um, and uh, you know, I already talked about the proficiency, uh, you know being done through creating training and development now Again sales force leads with, uh, you know a lot of uh with a growth mindset So learning and development are part of the culture again We have what is called a strail head Which is something that we offer to customers as well where you know, that's actually, uh, self-paced e-learning And we use that internally as well. So employees learn data skills by actually spending time And they are encouraged by their managers to invest time in their training and development Other than this, uh, you know when I talk about how sales force uses sales force for data driven marketing We actually use our own product called sales cloud where we set up campaigns to track the success of each activity Analyze what is driving those activities We use campaign history to enable the sales teams to understand what audiences are responding to this This actually helps drive alignment between sales and marketing, which is really truly integral to marketing's roi The other thing that uh, you know, I want to point out is we use marketing cloud Which actually helps to use different data points in order to segment audiences across channels We use data rama, which actually integrates digital campaign activity And also takes in marketing cloud data to empower all campaign marketers to have insights to have, uh, You know discussions around roi and to think about how they can actually personalize their messages through different channels Of course we use analytics because that's actually what helps us to measure and understand the results of the marketing campaigns You know that uh, we are driving which we do through using campaign setup in the tools Um, I think another thing which is really important to point out is the sales force dmp That's the data management platform Which actually houses first party data segments for cross channel targeting So if we you know, if we send uh audience segments to our cms our content management system It serves personalized content We also use the dmp segmenting for customer separation and personalization where we can actually target social lookalikes and also display messaging We also have a dynamic email engine, which is again built on sales force marketing cloud. So I can go on but you know The thing is that we actually use uh all of this internally and and in fact We can we also at times present this to customers where we talk about how sales force uses sales force Another very important aspect of data is trust and trust is the number one value for sales force. In fact, we have Uh a site which is called is trust dot sales force dot form where actually there is uh, you know transparent data sharing Which is actually the sales force community is home for real-time information and system performance and security So it's it's actually You know part of the dna of the company Fantastic archana. Thank you so much. So I think you know net net we've we've sort of spoken about every industry's culture Uh, how does the top management really look at it? We've looked at some conversations about gut feel versus decision making Using quantifiable research and data. We did speak about uh stuff like the dmp's and dynamic creative optimization personalization communications and messaging you know, uh There's one fantastic question that I've got from an audience and I I just couldn't stop myself to pick it up But there's one great question which says that uh, would you please suggest the importance? Uh, sorry. I'm just Sorry. Yeah So what's the bridge between human learning and machine learning in decision making and that probably would give a Of a peak on gut feel versus really decision making stuff I just want to keep this question open To all that where wants to pick this up because this has come from minus mohan and I thought it's a lovely question So a bridge between human learning versus machine learning Sure So, um, you know, I think that there is a you need to have a balance of both Absolutely, like we said, we need to have gut feel you need to have data But over and above all I think there's one thing which we have we're you know missing which is content Whether you do is the gut feel route or whether you're using data I think content is the king and you need to make sure you have that beyond then you then need to look at what the Sort of product is and and from a marketing standpoint. I'll talk about two things So for example, you have like saga talked about products like salt, which is very, you know, generic and you know widespread In visa, I would say credit cards would be one of those because you know, that's relevant for everybody Especially in today's day and age when we are seeing people are going more and more off-cached and into more You know contactless ways of paying that's something that's relevant for everybody There I see a greater role of you know gut feel understanding. What is the right content? What is the right insights to put across and come across with a mass campaign? You obviously have a you know a tv leg, but there is a very strong role for digital to play in that Then there are very very targeted products such as um, you know cross border travel, which Like I said it varies from person to person and there is not one answer that fits all So in something like this, you would have to have a very very strong data-driven approach Um, I think I mentioned to you earlier, but I think for the wider group. It's very relevant Like you say there is you vishal who's traveling to say dubai in 90 days Versus an ajna who's traveling to singapore in six days There are very very different needs that you both have and I cannot give you a same messaging So with dco and you touched about dco it actually helps us to actually target you Saying that you're 60 to 90 days You will actually need much more planning help which is like, you know, where should I stay? Or what's the best rate? It's one of the tickets while an ajna who's going to go to reaching in six days would have probably already done that What she would need is actually one of the deals for shopping for dining And that's where I feel the right programmatic using dco and using data driven You can actually target the right audience your right consumer with the right message And that is it and of course for you know overall products, whether it's you know FM cg or products like like we said credit cards You use a much more mass But you add in the data aspect to sort of segment between different target groups different age groups Millennials versus elders, but there is a need for both and like I said content has to be the key You know, I completely appreciate and that's the right chord you touched upon. I think we've lost him again Yeah, if I have to give my perspective on human learning and how it transitions to machine learning It's a very very important and burning question right now The way I see it There are two parts In which it can be dissected One is which are those elements of your decision making whether it is marketing decision making or even other decision making Which is at a high frequency and in a relatively stable environment So for example, whether you approve a credit card transaction when it is swiped at a machine or whether I approve A customer's underwriting when I am getting the risk to be signed off at my end Which is a normal situation high frequency situation and there has not been any systemic impact for example I might be doing a high frequency transaction through machine learning in a usual scenario, but after pandemic because of covid Do I suspend it for the time being and supplement with expert insights? So that Is the way I see it if it is a high frequency decision In a relatively stable context for most of the usual transactions I'll go with machine learning if it is a highly specialized decision Which is a low frequency decision or say decision which is coming after a systemic shock like let us say covid situation Or it could be any other situation in any industry It could have been the 2008 2009 global financial crisis also So I would have to suspend my judgment one example is for example How you use credit scores credit score in a way is an aggregation of a lot of knowledge And I can use it to take my decisions But after a systemic shock is the same credit score is still valid or not for my decision making through my algorithms All that is still in the air. I would say that Deep level of machine learning is still some time Well said and I'm going to move to the publishing side and now the technology side You know Svita, Tareka and Archana you guys spoke a lot about the culture of data and You guys live pretty deep data. You know, I want Each one of you to step up and say that what's the importance of human learning In your business because end of the day, whatever machine gives you a learning. It's it's still built by humans, right? So Svita you run a great Network or for native advertising. You have great data coming in from new sites and Consumers What's your human learning coming into the play and how does it really impact the machine? Honestly, I completely agree with sujatha and hankash where they have to coexist. There's it's a happy coexistence of What sujatha said before which was God feel is really just really strong knowledge of your consumer and the best tabula can do. Yes, we have You know NDTV India today all these and we have click data from half the world's internet users, but That's no substitute for what a brand manager can do. So the best we can do is guide the brand manager To use what they already know. So for example Let me pick on Let's see. So jar for And the food industry salt for example We have a site called trends dot tabula dot com which just compiles week by week data from her country Her category and what creatives are getting clicked on. So if we were to go with that His creatives would be featuring women this week. Let me look A close-up of her face and eating indoors So her creative the creative would be targeted that way, but you know, that's over So many brands and his the salt the specific salt might not be like that at all So it's no use really just to rely on the data. It's a guide post, but it has to coexist Thank you very well said Okay, Tariqa you won learning versus machine learning. You have so much of machine learning. Tell me what are you doing? So I will be honest. So in the programmatic space, right? I think I agree that I agree with pankaj. We are A few years away from machine learning being democratized in the way that it will get there. Oh sagas back guys So I think the interesting thing is It's coming, but I'm not I I agree with pankaj. We are quite a few decades away if not more where Right now we are in a hybrid. We're leaning more and more on data That's not to say that it would completely veto out One thing I want to touch on right like which I was thinking about right now And sujata was talking about content now on our side, right like on snapchat I don't know how many of you saw yesterday's tweet around The u.s president's not going to be curated on snap now How do we as a snapchat as a platform take that the reason we could take a stand there? And I'm being honest here is as candid as I can be is That's curated data And what does curated data mean curated means that there is human interaction where we choose whether this is appropriate or inappropriate Right. So that's why we're not heavily leaning 100% on machine learning if we were there's nothing controversial on my platform That was tweeted. We're not twitter Right. So the fact that we took that stand is because we have the right as a platform where we curate publisher content It's not an open platform like a lot of platforms exist out there. You if I want to create Homegrown video and publicize it. I can't But if I if that was what was allowed, then I am not human impacting and curating the data or promoting it based on what my audience wants to see So I think this is where it's a really good example from yesterday where The fact that we're curating data With human interaction like my I have a team that vets content that looks at content on a daily basis Guides publishers and what we work and not work that is all human interaction That is not just based on behavior that we're seeing right like and this is another example that I've given you earlier A lot of the times we see in platforms especially in social media across All platforms where you're when you're a microphone to the world. It's a lot about gloating It's a lot about how many people are following me How many friends do I have how many likes do I have like that's that's become a part of our dna in the way We consume content now, right and it's interesting to see is I may not actually like fashion But I'm kind of under pressure because all my friends like fashion. So now I have to like fashion That's not really your private personal interest So now if I'm targeting you based on fashion and you've never ever made a previous purchase That's inaccurate data. So I don't know a hundred percent of the machine is there yet Where human interaction is not valid I think human interaction might be more valid in a lot of times in the instances that I've given you in the last few days especially Uh than human day than machine learning only Thank you. Thank you. Thank you. Ajna. What's your point of view? You sell a lot of high-end technology Products you use a lot of data, right? But how do you see? You know, there's there's a great amount of machine learning. You build a lot of cohorts You do a lot of dco. You do a lot of uh tool management stuff, but where's the The the the experience and the human learning coming into the play for you as as as your industry I think, uh, you know, I'd like to just start quickly with talking about, you know, the very interesting debate between trusting your gut and trusting data which comes from machine learning and how do you actually combine the two? I think, um, in terms of You know decision-making sometimes when we are facing Information overload or we are you know in a situation like the pandemic currently Uh often there is uh because of the pressure we Or to make the right decisions. We actually delay decision-making. We put the choice off rather than deciding I think trusting your gut really allows leaders freedom to move forward and while intuition can actually provide a hunch or a spark that You know starts you down a particular path or you know reasoning It's actually data that you verify understand and quantify. I think uh, you know As uh, the world is evolving. We're actually going through an intelligence revolution Everything is smart. Everything that we use is smarter. Everything is driven by artificial intelligence. We can communicate with our speakers Our watches our cars, uh, you know infrastructure in the world talks to each other And actually machine learning is driving recommendations that are surprisingly predictive out of all this um, and smart experiences are some Because you and me, you know, we have come to expect this Having said that I think a very important point is that you know with the world in crisis such as you know, the situation is right now the use of customer data for Hetic marketing has come to the forefront and I bring this point up because I think it's really essential To have clear understanding of the customer's needs as they evolve and ai actually helps marketing has achieved You know personalization at scale by actually pulling out insights From data and taking action very quickly In fact, I will cite reference to uh, uh, six state of marketing report Which has just been released by sales force where we've actually In reached out to about seven thousand plus global, uh, you know marketeers including a large percentage from india And 84 percent marketeers report using ai which is about 30 percent higher than it was in 2018 So as you know, the amount of available data grows and the platforms that have to manage growth Um, I think it's really important to use human learning Where you have data and insights and information But actually making the decision really involves cut field and as a leader you should use cut field because that's really you know That is where your experience actually comes handy. So I think it's a it's a mix of the two Fantastic. Well, sir. Achana. We are currently at 455 up closer to 455. We are just about five minutes I'm going to go around the table and I'm just wanting each uh cmo To leave one advice for this at this audience is wonderful audiences that we've got today What is that one thing? As a data-driven mindset that you guys have what is that one thing that you want to tell the audiences Maybe a short sentence. You've got five minutes and we have six panelists Acharnan you want to go first One very hard to pick one I think uh, you know, what is really important is uh, I think I would what I would say is engage customers at the right moment with the right message Uh through personalized marketing by knowing what the customer wants or the consumer wants you can actually exceed expectations By serving them what they want and this actually comes through data-driven mindset Fantastic sujata I'd like to say that you know and as a really traditional Marketer from over 20 years is don't get overwhelmed with data and you know Have so many filters that you finally end up You know with a small subset and you lose the larger picture of why we are marketers Which is to understand the the the consumer insight develop the right content So I think data is great, but we shouldn't get overwhelmed with it Fantastic punkage My message is that having a data-driven mindset is a must today and we all If we have not already done it align ourselves to that reality However, as it has been said here We must leave some room for the human intuition and judgment to play that final role It is not going away anytime soon Yeah, saga Uh, I would say We have you Yeah, can you can you hear me? I think in between there was some issue So Look at where data fits in your role as oral strategy. I think the data should start up Essentials of sort of a strategy experiment Uh, sometimes it's going to surprise you if It's cheap. I think that's what I would say That's a very nice one. Thank you saga really appreciate Can we have Siva swita Um, so I think in terms of as marketers We've been told that data is really the new oil and we got to get the infrastructure start gathering it But once we have it, we don't necessarily know what to do with it and we get all this analysis paralysis So the best thing to do like as all my panelists peers have said before is have a Data hypothesis driven approach. So, you know your customer more than a new one Start from there then get the data that's only if it will influence your decision or not If it doesn't it's useless to collect it That's it Erica So off of the same way and I would say Track everything and then a lot It's a it's a common symptom symptom people collect too much data and don't know what to do with it It needs to derive a decision but I think Not tracking is equally harmful, which is not even influencing your decision. So I think Track and then decide how you are Thank you. Thank you. Lovely panelists and I think that was a wonderful session I'm just want to quickly sum it up in about a couple of minutes of what all that Today covered for those audiences We did speak about the culture. We did speak about the tools and the data lakes and the data stacks and the DMPs and the dco's You know during this course. I did keep getting some questions and there was audience asking about what are the kind of tools that you suggest Uh, you know, that's that's quite a very narrow question So I I probably wouldn't want to take it but you guys, uh, there is a lot of stuff that we've spoken probably that will address your questions Uh, and and wonderfully what was really appreciated by uh, by more by most of y'all is The human versus the machine. I think uh, that was so remains and While we may want to bend toward machine today We still need to be quite balanced between these two because it is the gut which also helps us really create that business So I think it was really wonderful. I really thank you. I think Uh, I'm sure out of this session today. We've heard a lot of You know data driven marketing stuff of How can it really help you take informed decisions? We how can it understand the consumer journey? really create audiences and How do you really look at? LTV lifetime value for the customers We did touch upon the fact that it delivers much much higher ROI So just summing that up, uh, I'd like to personally thank everyone for this session I think you guys were wonderful. The audience was wonderful though. We couldn't see them But I think I'm assuming that this audience has surely been giving us a good year Uh, so on that note. I really wish you good luck and thank you very much Thank you. Thank you everyone. Thank you everyone. Thank you guys