 So ladies and gentlemen, thank you so much for, you know, figuring out time on a Friday afternoon and joining this manual. And thank you so much for being kind enough to spend time and, you know, like enrich the entire process of putting this together. I will just quickly introduce my panelists here and then we straight away jump to our conversation without really wasting too much time. We've got Ramakanth Khandilwal. Ramakanth is the CMO of Payback India. He has about two decades plus experience across loyalty management programs, digital marketing and payment system. Hello Ramakanth. I've got Sujoy Ghulan. He is the CMO of Apple and he also heads Omnichannel SaaS platform businesses, you know, and he's got experiences across B2B, B2C, businesses in fintech, at-tech and e-commerce. I've got Prashant. Prashant, thank you for joining us. He's a seasoned digital marketing professional, extensive experience with the likes of Microsoft, Naukri, Visa. I don't know if there's the right order, Nielsen, Times Room, etc. But and currently he runs a company called Torq AI and he's responsible for its revenue, business growth and expansion. So pure suit. Next one is another suit lady, Nina Dasgupta, I think we all know her. She's the founder CEO for Zarka and also CEO for EDM. Passion is conversion marketing. She has been instrumental in making Zarka invest in the spaces, you know, which are futuristic and are building products which are absolutely unique to marketing. Last but not the least, Namrita, thank you so much for joining us. And she's currently the chief digital officer, you know, of Aditya Billa Group chemical sector. But we know Namrita for a very, very long time. She has been around for about two decades now and predominantly again in marketing space, US, UK, India. I think I was in touch with her on one of her assignment long back for auto sectors. I think from there to luxury to retail and specializing in marketing. So, so I think it's a fantastic panel. You know, thank you so much guys. You know, Nina, there is a second entry from your panel somewhere. We can see two Nina Dasgupta's on the screen, but you are twice as much anyway. So two of you would be four. You're on mute, Tina. You're on mute. I'm twice as much literally also. Nothing to do with the food, but I'll try and sort that. Okay, so now there are four Nina. There's a different person over there. Who did look like Nina? Nina's daughter, I guess. Okay, cool. So quickly, you know, without really wasting much time. Guys, interesting topic, right? Conversation and marketing. Have I been reduced to one? Yes, yes. Yeah, half almost. I've been reduced to, yeah, okay. One. Yeah, so, so yeah, so, so conversation and marketing, right? I mean, we want to touch the tech segment of conversation and marketing. We're going to talk about futuristic conversation and marketing. We want to talk about how NLP, how artificial intelligence, et cetera, are driving conversation. I think it's amazing, right? We have been all taught about a traditional funnel, which is funnel, sales funnel, marketing funnel. And I think currently, what digital has been able to do is kind of break those funnels completely and make it a lot more shorter, a lot more, I would say engaging for consumers to directly connect to the brands or even brands to target their customers. So if we just keep the physical conversation about social media, those hashtags where we use, you know, where we get on board with conversation that happens across various product. If we just look at, you know, just the one-on-one conversation, you know, from brands and consumers. And I will start, for no particular reason, I'll start with Prashan because, you know, I think he is right on the money and he's doing this for a living. So Prashan, please tell us your definition of conversation and marketing. I will keep it timed, okay? You have only 180 seconds to talk. Quickly, conversation marketing, I would rather look at a definition where you are seeing each and every action of yours is communicating to the end user or consumer. The earlier era where you were looking at a marketing where outdoor and display and digital, everything was different. Right now, everything has to go through one lens because your user at the end is evaluating you across different parameters. The user could be watching your TV series, could also be looking, reading blogs and reviews about your product, could also be commenting about them. And it's well-known fact that people obviously research a lot about any product nowadays. Even if you're ordering on Swiggy, you end up reading the reviews for it. So everything that an organization today actually is doing should be to communicate one piece. Your strategy of content, your strategy of digital marketing, as well as your product strategy should all be focused on and have a one-way view. So once you kind of attain that, all your data sets are together, all your communications, that's what a right definition of communication marketing because then you are conversational marketing because then you are not talking to different messages to the end user. What would be to your partner's users or the friend's users, you need to have one single line of communication. That's how I want you to hold there and move to Ramakanth actually. This is again purely your interest and your area expertise when it comes to one single line of communication from a loyalty payment, fintech kind of a background. So how do you guys see that? What is your definition of conversation marketing? So our sense about conversational marketing is that it is obviously being a conversation, it's a two-way communication. First thing, it's not like the standard push communication that we have been traditionally engaging in more in the legacy environment. Conversational marketing is where you listen to the customer and you respond to the customer. There could be some key aspects to it like contextual, real time and personalized. These are actually the very, very straightforward buzzwords as you know in the marketing automation kind of environment. I think these are very critical that apart from what Prashant has already talked about, it's really a two-way listening and it is something where you listen to your customer more and respond and probably most of the response is going from the machine. I mean that's where in your opening note you talked about AI, ML and NLP and those things. At the backend, machine is talking to the customer a lot more. That's how I would like to put it. So I think it's a good time to kind of bring in Neena and Sujai in no particular order but I'll start with Neena and Sujai. Now how do you look at this automation and how do you think that is empowering brands because you are at the castle where multiple brands are using your platform and your support to do this for various of their requirements. Is there a bucket, is there a formula, is there anything specific if you mind or it's just one size that fits all. How do you guys look at it? So if you take a step back, I think conversation is the new paradigm and communication and advertising is going to actually be under conversation. It is the new paradigm given the way technology has moved and conversation marketing is the typical space where the brand and consumer connect as if they are only made for each other and they are the only two people talking to each other. You feel special in every possible way of the connect. A conversation happens only after a series of contacts have happened. Now I think that conversation will happen only if you are connecting, if a brand understands the consumer and if a consumer feels that understanding coming from the brand. So therefore it is not linear in my opinion. It starts possibly in the offline world totally where you need to understand the consumer mindset. What we call psych, psychographics today very simply or what we call humans are based on personalities. What was derived by Carl Jung ages ago or the Australian psychology. So it is about what we do is about connecting mindset of consumers to a single brand conversation and brand agenda so that every consumer feels special. It's like I said it's about feeling that you are the only two people talking to each other feeling special. So technology of course plays a massive role then in terms of making sure that the continued communication by the mindset is happening on an ongoing basis. You are learning more about your consumer and you are responding in relation to the consumer's mindset. So this group for instance is a homogenous group. But how do I find the heterogeneity? You know if I ask Ramakanthi or the marketer or I ask Namretha, Namretha and Ramakanthi tell me the points of heterogeneity in this homogenous group. What is my mindset? What is Anup's mindset? That for me is conversation marketing and that's where the role of technology comes in my opinion. That's where AI starts building. It is a step before your marketing. It is planning. It is like I said a new paradigm. So that's what we have been working on and that's what we have had tremendous success with that. So it's a broader perspective about what the overall piece is. I'm going to move to Shijaya and ask where it comes to mobile because I will keep my questions about the platform that you guys specialised in. So tell me do you think that for mobile is it easier to identify psychographics because it is a lot more personalised device for at least in a country like India, you know in comparison to laptop where of course other countries it is probably personalised both devices are personalised. But do you think mobile is a much more personalised device and therefore the way you guys use the intelligence from mobile is any different or easier or like how do you guys operate? Another mobile is a huge enabler and has been so for years. So in the way and if we wear our consumer hats and look at or think back at the evolution of media and marketing. So anything which starts small starts offline. It's easy to drive conversations. It's easy to make it conversational with mass media coming in which was one way. There was a marketer sitting on TV or in the print ad communicating one way. Digital brought out a difference not at the beginning of digital but also with desktop advertising at the beginning of digital again it was one way largely banner ads you could click through and possibly discover content and explore. I think both bringing mobile and for obvious reasons. So again when you look at social media, social media is a layer or is an interface. Conversation possibly has much far greater interfaces and mobile is a channel. So the way I look at mobile is a channel and there are interfaces. Today my mother uses Alexa to time her cooking. She uses Alexa to listen to music. For her that's conversational commerce. That's conversational marketing and what she's engaging in. My daughter has a bunch of use cases. I have a bunch of use cases and I think all these are fantastic enablers and mobile for sure. And we've seen just the nature of the device, the ability to personalize conversation takes it to a very different level. The ability of access and that's the second point. Earlier to access conversations at scale or enable conversations at scale was just not possible or you had to be of a certain socio-economic strata to access it but obviously mobile has changed all of that and we've seen that just happened with such speed. And as it's cutting through and that brings me to a question. Again on the same format of conversational marketing. You have had a very exciting career right from an automobile to luxury to currently even with ABG. Have you seen conversation changing with these categories and if it has then where do you think we stand today and does it therefore empower you to probably predict a future for a lot more I mean a lot of us. Namrata you're muted. We can't hear you. So you're right. I think it's interesting over the last decade seen conversation marketing really involved. So I think about a decade back at least here in India when we were looking at it was predominantly around you know a lot of social and what you were doing on social. And towards I would say about five to six years back I saw that first transition happen where we started saying social is great. Performance is great. All of those aspects are great but you know what they're actually customers given that I was working with our details and we had a website and people were coming and making bookings over there we're like what can we do over here because we have analytics which is giving us some information. How do we start looking at combining that analytics to start creating different kinds of conversations on our website. Can we actually look at our lead generation data which can come from multiple sources including social search etc. Combine that with Siebel which was our CRM system that we had. And then look at that in terms of building a certain intelligence around the customer we could play back across multiple channels. It wasn't just across one channel. From there if I fast forward to Mahindra days and that's you know the last three four years all we've been doing is using the AI elements a lot more. And over there I think I'll share one or two examples which would potentially resonate with people. One is really around the whole area about saying that you know how do you create meaningful conversations with these customers. So if you take a category like farmers and I'm going to take not the traditional way to see categories because we all have examples but we were doing this entire sort of you know piece of work around saying what can we do with farmers we want to impact their lives. We make equipments for them we make you know products for them what can we do in that space. And we went out and like Nina said we did the primary research the understanding of the psychology and the mindset of the farmer. And then started building a platform whereby we said okay can we use this platform as a starting point for information dissemination and that's where we had chatbots so we partnered with IBM and we created those chatbots which would start giving basic information like you know if I'm if I'm showing a potato plant what's the right time to harvest it what kind of fertilizers need to go in and stuff like that. And so you started having very meaningful conversations which eventually dovetailed into e-commerce. That's just one example of saying you know how the conversation so the conversation instead of being a push became a pull. The second example to take very briefly is what we did with the auto business and what we found over there was again multiple channels of interaction and how do you create the differentiation. So here we went into looking at saying can we actually do a lot more customer profiling. So one was the customer profiling that we did because we were operating at a group level so we said we have the opportunity of taking second party data and looking at that and adding that as a unique value add to our understanding of the customer. And then we aligned with a consumer profiling company which was a Boston based company then and there you know more which have emerged since then but this was about 3-4 years back and at that point in time the intent was really to say can we do consumer profiling and use that again back on our websites can we use that on the other you know channels of communication so that we can actually start nurturing those leads and pushing them down the funnel and once they have bought the car one of the challenges we had was because the purchase happens with the dealer and not with the OEM and we were the OEM how do you in that scenario keep that conversation going with the customer. So by Namrita sorry to cut you so you know from this first party, second party information I would actually ask the question to Ramakanth you sit on significant amount of information from consumers I mean that's your core business right I mean you that's what you do for other brands which are using your platform so is there specific kind of exercise that you know organization like yourself do to kind of funnel psychographical analysis of consumers like say what Nina said about offline research when Namrita mentioned so do you guys conduct a lot of those online as well so as you know that we work with some of the big retail brands and bank and you know with each of these partnerships we get certain data and that data even goes down to the level of the SKUs that the customers are purchasing so not only do we have a holistic data you know about their spends you know using a credit card or debit card to their travel spends to their retail spends even fuel purchases so it's a very good set of data for us to deduce out of it as to what would be the customer let's say customers wallet potential what would be the customers life stage because you know these are things that you can deduce from the behavioral data I mean the standard examples like somebody's buying a diaper and you know life stage and so on and so forth the question that's asked about the psychographic thing because you know our two people are buying the same things but I think what appeals to them is that it's different it could be slightly different their motivations are different their motivations could be different right their ambitions could be different so this yeah yeah so we have a platform which is in partnership with a company called Unumur it's a digital market research company and we have integrated with them so all the data that we have at our end which helps us in segmenting the audience that you want to further analyze or interact with or you know get a sense of I mean this platform allows you know brands to do that it's an opt-in kind of survey you release it on the mobile app and the customers answer your questions and you can further enrich the data set that you have so the platform is very much there and a lot of companies use it for various reasons including their let us say when they are coming with an ad they want to see whether the ad appeals to a particular section of the audience or not the advantage of digital is that your cohort size could be as small as one and you can get data across that so yes that is possible but not as a part of the core business core business is really about the behavioral data real purchase behavior data and then building on it a layer of let's say more insights on how the customers are thinking cool so from enriching experience into enriching bank balances how profitable is this business how do you guys monetize conversation and marketing how do you foresee I won't say how do you guys but how do you foresee as a business leader who is the forefront of making this change working with this change makers how do you foresee this becoming a formal business line and therefore the partnerships etc etc so I want to respond to that by narrating a small story that I had heard years ago so there is this man who goes for an event at a marriage in Lucknow Lucknow Nawabi place and a big man asks this young man who is a journalist he says what do you do so that man says what do you do for salary what do you do in a nutshell that's conversational marketing and that's what happens when you're investing in a business it's about your ability about your ability to hold on to your idea and hold on to the thought because you know the shift is coming so in terms of making money you don't make money the day you start that business right you plow money because you're building a technology you're building layers of the product but are you making money in the future yes because when you make money in the future for instance when simply when Ramakan said laid out his first party data and he said the need is behavioral meaning for me in my mind is well I've solved that behavioral problem for many brands great right it's just a matter of integrating so there is definitely money it's about being in it for the long run so you just need to make sure that the role that you play in the entire ecosystem remains significant conversational marketing if you just say that I'm someone who will create headlines it is not a sustainable business model if you say that I'll do performance it's not a sustainable business model it's need it needs to be something that is integrating data intent interest on a layer of technology and that is constantly improving only then it's profitable do we have any regrets no are we profitable yes did we make money on they were absolutely not true but it's I think it's worth every penny that goes in because it's multiples are huge yeah yeah because it's just an offshoot and you also have personally selected that question from Rajanath Kamath would you take that question it's conversational so I'll just read it for everybody conversational is personal in an automated impersonal communication it's an integrated message delivery of brands communication strategy it's not really a question but I thought it's very important and interesting point of view and I have selected to answer it live or add to it would you do it now absolutely I'd love to so a brilliant point Rajan I know you are in that space significantly yourself it is it is personal and that is the point that I made in the beginning conversational marketing is when every consumer feels that the brand is talking only to her or him it is personal you don't even feel the platform actually because your device in my opinion in my personal experience like you say it's personal my device is an extension of my own self today yeah it is so it's absolutely a valid point if I I think one needs to also focus on the part that you are saying is about brands using this as an effective means you know I think the brand now needs to learn about saying that this is not performance marketing this is about building a relation to have the conversation so it is far larger than performance marketing one of its outcomes is better performance efficient more you know there is greater efficacy there is greater effectiveness but I think the second part is which needs to be taken for more seriously I think that is less of a comment and more of a direction and advice possibly Raj you know it's something that is of significant value and requires significant attention from brands today they it can you know I find it a bit disturbing for instance when everything gets bucketed under performance as a parameter you know almost saying I want innovation so when you take innovation to them saying this is innovative they say okay can you share precedence trees yeah so then you wonder okay fine you know how can I bring precedence to innovation so it is you know copied with pride innovation kind of a thing and then we'll say okay fine this is fantastic but you know the last person generated a lead for me in 137 rupees so because we are bringing in conversation will that cost per lead go down to 122 rupees so I think we have to stop commoditizing and marginalizing the impact because as individuals we have to see if the conversation is not great we get bored and we switch off and a brand today can no longer afford that because you know what we are researching on the net and we are eliminating on the net as users so you cannot even generate boredom so Raj I think your second part of that statement is a brilliant direction and something I resonate with totally yeah so I mean I'll just also ask on that same point that you made that you know it is an extension the device is an extension of yourself and the technology therefore is an intelligent layer that is creating this experience for people so I'll ask Prashant actually Prashant you've been building a tech yourself right so tell us a little more I mean and in probably language that were non-tech savvy people would understand that what do you keep in mind how do you build this I wouldn't call them emotions but the sentiments in an NLB in a process so there's one statement that I've read long back and which completely encompasses what how we look at technology and what is it it's mind over mouse so digital and others eventually brand marketers also start looking whether you're clicking whether what where the user is going where the mouse click is happening but essentially it's mine I mean a user today would and trying to simplify and as much words the evolution of AI into the marketing space is more in terms of telling looking at the insights of millions and millions of users and page views and giving you data sets saying that okay this is the best approach for a brand to take I mean a clear example there is now your AI can actually tell you what kind of your newsletter subject line should be I mean for a brand like let's say payback or any other brand which is sending marketing communication and AI can actually understand testing and tell you that okay these kind of subject line will help users open your email now that's very important now if you are kind of running a campaign or you kind of doing a marketing outreach it's not important for you is to be where you are being seen or what action is being taken but is your user actually user is integrating interacting with you I mean evolution of marketing happened where users were saying that okay so many people should see my ad to move that somebody should act on it now we're talking about a user should engage with you now it's typical direction evolution of a communication that you want the user to engage but now you don't only want the user to engage and buy but you want repeat user you want loyalty you want the user to be your mouth please not to achieve that you need to kind of track and have a single actually not only communicate to the user but also listen to the user everywhere the data sets is being I mean today brands and marketers have hundreds and thousands of touch points with users how do you capture all of these touch points and understand and take the next measured step is what actually AI and ML today is helping you right one example which I said is newsletters it also tells you which sites to go and advertise to what should be a subject line so all of those can be done through I think I'll bring like an example which is closer to the time we are in during this COVID-19 there's this company called health they actually created an entire health platform which was built on top of messengers so chatbots though Namrata strictly told me not talked about chatbots but I will just take one example is because I think it's very humiliating so they actually allowed people to during this time where mental health is a big cue wanting to be sure whether you are fit enough to dwell in the real world or not I think health that built a fantastic business and which kind of flourished in the last six or months is all about helping people ask these questions which are intuitively right and therefore kind of guide them to a direction right and so the next question is actually more about use cases I think we have all spoken about use cases from the past but primarily to the marketers here Namrata, Ramakan, Sujoy examples of such work around us I want this panel to give some real work examples so that we set the right context who will take the first Namrata, Ramakan, Sujoy Sure Anab just day to day life is so full of examples and so many of us like to try out new tools so eager to share I think both having been both B2B and a consumer marketer and continue to wear both hats just tools help us do and today AIML does not always necessarily come into a tool that's only priced at say $5000 a month but anything which we use everyday super cheap could be a Chrome browser plugin that's helping enable this as well so at least on the B2B side for conversational it helps to understand because there's so much conversation that's digital either on zoom meetings like this or on email we for example use sales tools anything that can help us understand intent and sentiment from for example the sales emails that we exchange we're working to trial a tool called humantic.ai I know someone who's built it launched on product and about a week ago rave reviews but it helps do that tools like saleskin to understand the conversation with the prospect you're having is the prospect serious other words he or she uses indicate that they're concerned they eager that they don't want to move ahead on the consumer side on the other hand at the risk of plugging our own product we use something we build something called shopper which helps retailers build O2O workflows online offline workflows set surface their offline store inventory and obviously the easiest way how all of us shop is through conversation there is a conversational interface today through chat but also available through web interfaces which understands today when I say I'm looking for and it's easier said than done all of us have gone to any site and said I'm looking for a brown leather belt and seen brown leather bags and said oh no this is not what I want so it takes a lot of effort to get that right I think today there's both effort and technology but the products are there all waiting for us marketers and consumers to use I'm sure of the products so one case I would talk about and frankly it is when you look at it it looks more like an iceberg at the top of it what comes out you might feel it's very simple but the fact is so much has to go behind the scenes to make it happen so a simple use case that we have built over the years is let's say what we call a wait time ratio now I know that there are set of customers who go for high frequency purchase in this case let's take the example of pure purchase by petrol or diesel now if we study that trend for a long period of time across millions of customers and all the other variables put together we have come up with something called a wait time ratio so for the cohorts we know that this customer now he comes to the petrol bank again in 15 days to even a month right so if somebody is not coming back and since we are in the business of loyalty our core driver is that the customer has to come back again and again and again so if the customer is not coming back then we are potentially looking at the risk of losing that customer and at that point we need a nudge that nudge could be a simple information that could even be embedded with an offer for that customer and the more the delay is you need to really sweeten the offer offer more now to achieve all this and because there is audience here and I don't want to make it sound like it's a very simple exercise I think all of us here have done it the implementation of a typical marketing cloud kind of a solution which is banking heavily on data data is the lowest layer the richer the data the better your modeling is going to be so we have to first create a 360 degree or single source of truth for the data that data could be starting with demographics to transactional to the response to your previous campaigns and whatever other external ingestion you can do this data is just a starting point on top of it you have to build layers of deterministic deterministic let's say equations and probabilistic modeling and then AI and other things which are only going to hone this the more data that you get data at the behavior level your true behavior and the models are to be defined so I just wanted to make sure that the use cases may not sound like out of this world but even to achieve those simple use cases it really takes incredible amount of effort and things to fall in place I think Arthur C. Clarke said Clarke's third law that any sufficiently advanced technology is indistinguishable from magic so I agree with you that sometimes it just looks very simple but it takes a lot behind it so from a lot to setting on a lot of data Nina I'll come to you Namrita on a more specific case setting on a lot of data of so many brands and I think Ramakan made a point that data is indeed the lowest layer and the richer the data is you can build such a huge structure on it there are so many questions about data security so many issues of and now we know from September the world is moving to a cookie less tea party as we call it so then what happens with so much data and consumers are more reluctant to share data and you want to build an entire business on top of not ready to share data and I think Apple is ensuring people to ask every time there is an app that you know do you want them to get your data or you don't want to get the data so what happens then and what do you how do you foresee a world like that it's God okay it just went a bit techno Anna went a bit techno was almost being breakdanced towards the end of the question so interesting you asked that so we anticipated a part of this cookie less world a while back right and we were working on our entire platform with with use an application of data that happens actually at more at a panel level and at a script level with the companies the first party data but it is more about estimation and use of it without revealing identities so the way our data is structured is actually structured on pure behavioral what should I say the offline personality buckets which are derived via psychometric tool that we have and we take it live via blue guys you know we activate the data via TMP so the blue guy right the beauty of our data is that it's going to be panel based so we're going to get panel based information more than actual first party cookie let data so the panel information which is the shift which is where which is where the beauty of our tool is which is what is creating excitement around it is that everything is going to go on for instance if payback has a panel of their entire people I can estimate the entire universe for payback using one layer of psychometric tools and there is no data transfer and there is no cookie enablement after that there is no cookie based conversation that happens right so our data has been structured on the future of data which is panel actually very less on cookies we don't do cookie less because cookie in any case fades away and dies in three months your cookie is not longer so I think how you define and structure your data is what makes it to be personality and psychographic based and mindset based and intent based you know like Sujai said true sorry to cut you but interesting point because I think the entire digital entire point of view when it started was about IP based data and cookie based data and that was the differentiator because the world survived on panel and extrapolation of panel on the entire population right are we going back there well we never you know the interesting part we never gave up the panel the entire study that we do today we look at com score anyone any planner who is doing the plan still bloody clicking on that one battle button of com score we are still doing Bach so we haven't moved from panel we are still doing Nielsen scope we are still doing PGI we are still doing IRS we are still doing panel world unless you have offline research which is the most deterministic you cannot have probabilistic in the online world because online world is a series of activities that are happening right so yes we are going back to panel the only thing that we get added is layers of validation that we get added so blockchain that's why blockchain is going to begin getting used so which is saying that this estimation getting validated right but your your data is today sitting hidden yeah the company is like Infosum that got by Xander when I met them doing back he told me he said you know everyone shows me the door and today is he does he giggles all the way to the bank each day so Xander is bought it right so I think the signals where they are 2 years back we are going back in the panel world because we need real people auto innovation are separate you see innovation happens probably still with the mind correct so to come back to so then we're going back to panel and I think Prashant mentioned about a more you know individualistic conversation where a machine is talking to a human being so while marketing goes back so many years to the panel and consumers move to talking to a machine where does it come together I think I see a very good synergy between the two and let me share an example actually because I think that's a classic example of you know the synergy between man and machine in that sense and I am going to though I said we won't talk about chatbots but we are now getting into that he's a great moderator he drew you so you know this was when I was at Mahindra and what we realized as we were looking at transforming the car buying experience something that came out of our primary research was that customers who used to typically come three times a year another year say three times in their car buying process into a dealership to check out the color, the configuration actually taking the test drive and all of those things over a period of you know a couple of years that number dropped from thrice a year to once thrice in a purchase cycle to once in a purchase cycle now the minute we cut that data point we said okay fine this is a make up grade because this is we always knew that the you know the in showroom experience was important we also knew within that the test drive experience was really important but then this coming down to just one particular you know sort of walk in or one interaction with the customer in their entire car buying journey we would get face to face with them was really critical now what we also realized was that we are in an industry where you have an attrition rate of 30% plus and that's an industry average that you have so you may have somebody attending to you as a sales executive who is three days old in the system or you may have somebody who is in a three years old and the experience will get is going to be so different the conversation you are going to have with the customer for that 30 minutes that you are giving him the test drive is going to be of such a different nature so that is where the man and machine connect happened and we said can we actually augment the capabilities of our you know sales people and give an opportunity for the customers to start you know getting their questions answered in the first go over a period of two three years we launched this about three years back over a period of three years we've collected a lot of data in terms of modifying in terms of making the entire what far more intelligent but also using those questions as an understanding of how do we need to sort of moderate those conversations so we've got six archetypes you know over there all those six archetypes and I'm not going down to an individual customer level because we haven't reached that level of maturity but at this archetype level now we have a fairly defined understanding of how the conversation may flow it is still fluid it's not scripted it's still fluid but I think that is where the man machine sort of you know convergence has happened very beautifully and that's just one example of you know seeing how that intelligence can play a role in actually building a far more meaningful conversation for the customer and making sure that obviously it has a direct impact on the conversion ratios you know so it's sort of a hit ball I just wanted to jump in here Arnav probably ask P.D. to talk about the project they're doing for us which is actually exactly what Namrita said that once you have derived your consumer personas how does it actually how is it that data is getting integrated and P.D. and all have worked really really well I want to do it a little bit more dramatically though so I want to read out Alan Turing okay he said that I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted okay so to that from that point P.D. I want to ask you that how do you kind of inculcate that knowledge or what are the kind of signs, what are the kind of rocket signs that goes on from the algorithm experts or people who are pros and enemies to build these machines which literally talk to you know consumer at the same level of or similar level of intellectual quotient so let me answer it because I'm not an engineer I'm a sales guy and my understanding would probably be at the bottom tier so the way what Nina mentioned your question I really Alan Turing is something and that still is a standard test to break AIs so what is data it's about data density the more data you have you are able to let the algo decide and a piece where the man machine conflict happens is when you start telling AI to generate a specific recommendation that won't work I mean you can't tell the technology to give you a certain result let the technology achieve the result I mean as Nina said and a project that we're doing with Nina is is where the machine is telling you that okay this is your consumer and this is a user and this is what their behavior that likes just likes archetype it you can't take and tell that a user is coming on to a sports page has to be a sports lover no maybe I came there trying to read something else so you don't put those rules so there used to be a rule based jargon that we used to do that anybody landing on technology side is a tech lover or as marketing evolved that anybody on sports and tech side is a male so that used to be the one and saying that automobile ads always were targeted at men but no now you see automobile ads displaying as the driver and the person entirely the only person in the ad because you let if your technology is working for you let the AI let the elbow tell you who your consumer is and what is the best for that consumer you cannot then put in a rule that will clash I mean that's where the man and machine disruption is but just to add one more bit is the more of the data the more measurement of each and every touch point is very critical here so you let those thing come into place are you measuring the right thing are you removing the bots and frauds out of it because today even if you run a camp and you'll end up finding 60 to 80% users landing on your sites are just bots so is your recommendation talking to a machine talking to machine or is it machine talking to user I believe machine talking to user is a much better scenario than machine talking to machine so that need we you know I can see Ramakan smile because I'm sure he would be using a lot of these elements across the board cool so I mean just quickly two minutes to be politically correct since we promote a little Zarka's Emotions innovation we'll also let Sujai speak about Apple's innovation quickly in two minutes we'll take some questions after that so lots of them will try to be quick so today we just announced a strategic investment into Indus OS India's indigenous App Store which is exclusive on the Samsung Galaxy and other other devices in this domain Sujai just to in this conversation marketing anything so what that allows us to do is to do two things vernacular and go into video both of which are the I think which makes or democratizes democratizes conversations to a large extent and doing that through easy ways on apps based on data backed by a DMP for example that understands both user profiles and engagements and therefore is able to with and is able to help attack live up to its true potential right we have a question on this Ankit has asked that how important is regional language to all of this you did this with ADM long back even with Zarka so how important is the regional languages in a country like India and otherwise also it's if I can take that it's the language we converse in India there's not people in this room are not extremely representative of larger India if I can say that and it's the natural course of things where innovation started so Bharat as we say is estimated to be 3 billion 3 billion dollar market in the next two years language is very critical because of the levels one is level of comfort of language towards education ability to reach entertainment, viewing and audio and access to cheaper devices and cheap data and I think Geo has done that and a country referred to only as a country has enabled the cheap phone today so it is significantly important it's value can only be determined when we move to start talking to them specifically are they a great consumer segment from a marketer's point that needs to be determined still and for which buckets it's a massive piece like from investment significant investment happening in that part today so another massive point actually made by Rahul on questions asking and I'll ask to Ramakanth do you feel the government you know I will actually put two parts is doing enough on the data protection laws and do you think that they need to be making it more stringent and to make it more stringent they actually need a lot more clarity themselves is what I would presume so what is your point of view on that so you know it's a very tricky matter as you know and if there are some references to be drawn we have the GDPR and those kind of things in Europe and the other markets I think for scaling up any business it is important that if there are any loopholes or leakages or risks they have to be covered I don't deny that at all you know the data is getting misused if the data has been you know data has been procured without customers consent I think these things need to be addressed over a period of time right the thing is that the entire process has to be gradual it should not be like one you know one fine day everything changes because so many systems have evolved like that they have been created in a certain fashion so it needs to be you know there should be enough time given customers have to be educated because you know when it comes to marketers procuring consent the customers also have to understand their rights and their their risks so this whole thing needs to be done in a gradual process okay so I think another interesting question on this topic and I actually love this question by Rupin he says that is there something like too much data where we have so much data about consumers that we lose sight of the consumers themselves and optimize for numbers and not humans how do we avoid that trap so I find it it's a very interesting philosophical question but it's a real to the core because sometimes when we get so close to you know reading something we often lose the sight of what we are reading or trying to read so I don't want to specifically tell anybody but all of you can give a quick you know point to this that are we becoming too friendly to PD's machine human connection or are we going to the other side of this panel is important and then have another leg on the other front by saying I will promote automation and conversation marketing so anybody can take this and we can soon kind of wind it yeah that you have made me the antagonist I just want to say that there's nothing like too much data rather there's a challenge of very little useful data so you need to swift through tons and tons of data to find the right pieces to act on but they see data as the new drug right and they say there is an overdose of drugs so why not there is something called too much of data I was saying that it already is an overdose stage you need to find the right to fix it it's not about how much data you are capturing on usage it's what you are doing with it and whether you have the right set now classic example as you rightly said was by September with iOS 14 coming in the device in Apple Google is ending the cookie it will be cookie less era after within 18 months so how now you already are doing away with the only measuring measurement or trackable metric you had now how do you go about that so you need to start looking beyond that you need to start looking at behavioral and other data to derive on those pieces so it's not a question of too much data it's again a challenge of useful data no PD is taking the fifth I know Nina is a direct person Nina what is it is it too much data it is FOMO it is it is just fear of thinking that if I don't talk about data I want to lose out second is I think it's rather simple we are just in the place of complicating it because complicating things gives us commercial benefits time complicated the more intellectual it sounds and it gives us commercial benefit third is I think we are not having honest conversation so there is too much of it there is huge amount of data I think more than that it is not about too much there is a lack of clarity on what data means itself and why am I seeking this data you know I think that is the need of the hour honest conversation is what's the agenda and why do I need the data and what's the data for just having data there is a time I don't know I don't know if you remember there is a time someone would say 70 million unique said MSNK 50 million unique we give business to Yahoo and then you would ask them will you bloody reach all the 70 million unique why is the number of unique important isn't what you are doing they will buy 1 million they will not even buy that 1 million right I think it's the needs to no basis it is that we want to have information but there is FOMO they are scared but we also have immense lack of clarity on how to use and the people like us also are responsible we will go and get this complicated layer of conversation and we will give them the fear of God we are like the pundit the pundits of the old time I think that is the data conundrum today true so from Brahmin to grassroots the last question we would take it says that often the linkages between analytics and execution this question is still very poor so Namrata you are now dealing in this category do you see a similar implementation from your sophisticated cars that you once sold to your current shift do you see the amount of data transferring to the grassroots and quickly we have to answer this okay I'll give you a contrast I'll give you the B2B space and the B2C space so in the B2B world which is the chemical part of the business that I work on there are no conversations happening there is very little data and it's pretty much like a customer coming to our website walking into an empty store there is nobody to attend to them so the bots that we are talking about the conversations that we are talking about the personalization that we are talking about all of that at least in the chemical space in the B2B sector is absent it's beginning to maybe start happening but by and large it's absent now let me move to the fertilizer side of the business and on that side the conversations with the farmers that we are having I think again there is scarce data we are beginning to make sense of it I gave you the example from Mahindra's we are trying to do similar stuff over here we have started collating data but I think meaningful data even trying to have those conversations creating meaningful data is going to be a journey so we know we are close to I guess the slightly more evolved world of B2C luxury auto property retail etc thank you so quickly just wrap guys thank you so much I think I feel a lot more confident about not making money after this session and I think we still have future because we are still at the top of Diceburg as Ramakan put it so thank you for boosting my you know what do I say confidence in it having said that quickly quick points that came out I think this man and machine existence this cohesive existence is only to get stronger and deeper and richer I think it's going to be there for sure this is the future this is the new funnel it will just going to break the the entire customer purchase funnel that we are all aware of it's also going to continue to get richer with time because we just know a little bit about this as yet as Nina said nobody asked the right questions our some of our you know viewers did and we hope that together as an industry we would continue to make the right questions ask the right questions and then find out the answers to those questions ourselves and we would do that without being the Brahmins we will involve everybody you know together so on that note thank you so much ladies and gentlemen and really appreciate your time thank you thank you Priyanka thank you everyone thank you so much it was a very insightful session thank you