lets get started so now that we know a little bit more about our panelists lets go into the professional introductions so Murr if you could just start us off with a self introduction just so people can know your relevance to you know the theme at hand of AI marketing yeah so I currently run the global marketing data science team at uber my teams are supporting 800 plus marketers spending hundreds of millions of dollars annually in terms of measurements attribution optimization targeting and insights that’s include both 8 channels paid digital channels.
also crm own channels and before that I was in a start-up in LA the multi-touch attribution TV attribution cross channel before that I was in an another company doing scenario simulation forecasting budget optimization products nice all right so it sounds like we had made a good choice and then run each of this panel and an in jabba how about you whats your professional professional relevance here so Im a head of marketing at gradient Ventures gradient Ventures is googles.
AI focused venture capital firm and so what we do is we focus on we make early-stage investments typically at the series a and its sometimes seed and Series B as well and you know that’s checks anywhere between one and 10 million and as companies grow you know we we grow with them and so we are an AI focused fund you may know cap G & G V as well within Google were exclusive focus on AI.
companies and we can do that because most of our investment partners are highly technical and a pettersen or our managing partner has spent a lot of time you know on the on Google search itself which is is very strongly I focused that’s what I do as a head of marketing there’s two components of the job or three components one of them is working with portfolio companies on their go to market challenges and marketing challenges one of them is actually think about the fun.
itself and how were positioned as a fund within the venture capital landscape and then third you know working on diligence for companies in the working space specifically cool that’s cool so I bet you have a pretty broad view of how many different companies are are thinking about AI when it comes to marketing that’s that’s cool yeah begin you like yeah that’s awesome and if she show you are you good to go shes tech tech good i I think so if you.
guys can hear me okay and see me fine then I guess if you’re good yeah you sound good you look good welcome alright so before you give us your professional introduction since I missed you for the basic questions just tell us where did you grow up Oh so Im originally from India so born and raised in India and not just one city in India my dad was in a job that took us across the length and breadth of the country so hard for me to pinpoint.
pinpoint one particular city but yeah originally from India and have been now in u s came for graduate school like a lot of you know immigrants like myself and then have been here for last 25 plus years nice awesome Wow so we have Turkey India in the Netherlands oh wow yeah that that’s awesome alright and so now give us your you know professional introduction why does the sheesh makes.
sense on on this panel yeah sure well why am I here what make sense that’s up to you guys right but I can I can I can tell you why I responded or why this is of interest to me so look background wise I mean right now Im at Humana I lead enterprise marketing for what we call our health care services care services part of the business which is everything other than insurance services part of the business I mean like a lot of health care or big health payers Humana is also transitioning from a pure pierside into.
more of a full-service you know care care business where there’s a pair side of the business and then there is you know health care side of the business I am the head of marketing on for the business and have been in health care for last 16 months or so now before this all of my background was in performance marketing and financial services so most recently before Humana was runnin marketing at digital payments startup venture-backed startup called remotely its no longer a start-up I.
joined router for CDs a and by the time I left we were CDs II almost so that was a great fun journey and and before that I was with Capital One for a long time so in all my rules I I would say data analytics consumer insights like has been a theme to drive the business drive the PNL and then that’s why now you know.
for here we you great I think I just froze for a moment but Im back alright awesome so now let us begin so hello everybody welcome just to you know sit the lay of the land this conversation for the next about 50 minutes is focused on how you know AI is coming to life within the domain of marketing the couple ground rules Im going to set is Im about to send in the.
chat to all of you the handful of preset questions that we have just so that you can understand kind of the journey were about to go on and then if you have any questions throughout you know the whole conversation just click the Q&A button and submit it and well try to get to as many audience questions as we can all right here we go so from a high level were starting you.
know 30,000 foot view how is a I changed marketing and you know explain it as if Im 10 years old this question is not meant to get caught up in the weeds somebody jump in I can go first perfect I mean what Im seeing like recently is especially on targeting sites and especially elective for existing users you could be very targeted with AI and understand its certain behaviors and predict them for example we can now build models very targeted models for cross sell upsell or reengagement meaning you can predict if.
someone is gonna do another order or another trip not because they will anyway but they will do it because you show them an impression or you send them an email or you send them a promotion so you can really understand the incremental ti of those behaviors given that there’s a marketing intervention and these models a lot of money especially if you are giving people promotions you can only send it to people who would otherwise wouldnt take an action and you dont waste your promotions that way you dont send it to everyone and you can also.
understand some of the harder problems with the same type of methodology its like what is the value of for example someone like Im talking in the context of uber ride-sharing business what is the value of someone taking it Airport trip on the long term lifetime value and there was the first premium right trip and you can tease that out and that you can done that way you can determine how.
much to pay for that impression so if you know the value and in that way you can actually optimize for your bottom line and on top line revenue so I think that’s a with the investment of the big data and modeling capabilities like retargeting becomes very accurate yeah you’re definitely with you and then yeah shes youve been in marketing for a while what did it used to be like before you know personalization and AI and data science was part of this field well what it used to be like yeah so first of all.
you know before I give my point of view and what you as I want caveated by saying that I have been looking back very fortunate that even 15 18 years back I was fortunate to work for organizations be part of organizations that were always at the cutting edge of analytics even back in the day right so back in the day when analytics a lot of.
analytics used to happen in Excel and you know some of the mainframe and things like that other language like SAS and all of that beavers still like Capital one back in the day or HSBC long time back in the day in financial services or Li bang that I was a part of they were always at the cutting edge back in the day so I have always been used to you know being a part of cultures where you are doing your response modeling pricing modeling target sophisticated.
targeting to really you know drive your strategies where I think a IIE has made a difference is in my experience building on from one Mort said AI has enabled the analytics at scale so you can not only think about optimizing five different point solutions but those point solutions also on its own can handle some really large complex problems so I think that is one aspect I see and theyre examples in my experience are across pricing across creative testing like you know versus a human looking at past wins in marketing.
messaging and creatives you know you actually feed it into a model and then model tells you that you know what looks like for this particular segment this is the kind of choice of words that you should be going for so in my experience I would say its really enabling analytic self scale now going to your question earlier that you know how how it used to be earlier I talking to my peers you know organizations where analytics was not back in the day at at.
a cutting edge I think what a AI has done is with the democratization of computing power and some of the AI tools and packages I think this is just broad analytics to organizations of various sizes like Im still involved all the NASS the adviser capacity with a some early-stage startups the kind of analytics they are able to do when you are like at a seed stage or even you know pre a stage or C is a stage its something that 15 years back you you.
couldnt yeah yeah jab I mean you you know are immersed in this the AI startup world specifically and I guess you use is there like a large gap between the marketing capabilities of one of your portfolio companies versus the marketing capabilities of you know Google for example your parent company so actually in a way no I think there’s a lot of disciplines within marketing that even with AI stay very similar and that are that are very core to the.
discipline like messaging and positioning that are quickly like very strategic and so you know a lot of our companies arent marketing companies but they apply AI in a way that it makes something you know particularly better like for example Elsa is one of her portfolio companies that helps you learn a language and not only learn the language itself and speak it better but also pronounce it better so because all these data can be super quickly analyzed and it can be processed on the cloud it.
can now instantly analyze your sub byte and return to you what you need to pronounce differently so for a company like that you know the actual like marketing capabilities that are driven by AI theyre kind of playing in the same space but obviously the marketing field has also itself been revolutionized by AI and you know there are some obvious tools that everyone has to their disposal right like in digital marketing there’s a lot more capabilities like for example with them dynamic search ads like within Google there’s a lot more like Auto generated.
content and experimentation that that can automatically happen similarly you know there’s tools in the in the content marketing space for example their stools like market Muse or phrased Ohio where basically it scrapes like all the content from from the web and looks where are the gaps in terms of what is out there and what people are searching for now that is not something that would.
have been possible before cloud computing at all so you know as a marketer there’s definitely different tools to your disposal that you’re an out kind of super super charged with and so obviously because of that another way that I would have impacted marketing as a discipline is actually the way that you hire right so it rather than in your.
hiring only looking for talent that has you know that the brand and communication side of marketing in the performance marketing side there is a much stronger analytical component so yeah those are some of the ways that I think about it yeah yeah that’s that’s super helpful yeah its crazy to think that you know a small seed series a startup could at all compete in terms of like.
marketing you know capability with a large fortune 500 company but yeah I guess I guess maybe that’s that’s become the case all right lets jump on to our next question which i think is an interesting one so you know how does AI make it into your marketing program is it models built in-house third parties both and I want to give people a sense for like you know yeah how do you build.
a a I powered marketing program in the present day whoevers excited jump in I can I can since March was gracious enough last time to to start the conversation I can I can volunteer for us you know from based on my again you know past few years or even the current kind of engagement its a mix of both its a mix of you know some internal development and then that times industry.
digit outsourcing so its its a mix of both really comes down to you know that that in-house versus kind of outsourcing kind of build versus by decision comes down to you know what are you trying to solve for is there a solution that either you know internally you dont have the data or talent or just resources bandwidth to focus on or sometimes you know there is a partner.
that comes with a unique capability proven capability and and and that’s where you decide to partner so its really there’s no I would say there’s no standard kind of answer on that and and and in terms of how do you build a program like data-driven I think it really starts with culture in my experience I mean having word that organizations where there that are best.
in class in data-driven being data-driven marketing performance marketing organizations to start up where we we scaled are really exciting data driven customer centric business or some of the transition stories that have been a part of and I think it all starts with the culture it starts with the right talent right either ship competencies the tools the data the technology that follows but but if the culture is not there that data-driven culture how the decisions are made how we talk about decisions how.
we think about frameworks then I think everything else you can have the best data you can have the best technology but that that data-driven marketing powerhouse doesnt get built yeah yeah that’s interesting Im ready digging a little bit of shoes so you know so you’re the CMO of the marketing function so seam of marketing redundant you’re the CMO for you know all the non insurance stuff that Humana does and so you’re sitting up top you’re making decisions lets off a budget key on a.
budget here but then you know as part of your department like do you have you have full time data scientists that are building models and and figuring out how to get those into production in the marketing function do you have people you know that are just focused on the you know third party vendor piece and trying to find like cool AI startups that you want to actually leverage what would like the personnel be youve.
actually assembled into your your department yeah yeah so we have I mean we are a large shipping you know Freeman is a fortune 56 company so as you can imagine both because of the the size of the business and in the complexity of healthcare which Ive come to appreciate in the last 16 months its its a complex business because there’s so many provide you know stakeholders in the value chain we have both you know we have in-house kind of analytical resources smart data scientists and on businesses marketing analysts who sit within marketing and then we have you.
know enterprise wide kind of you know data data analytics function as well so it really all starts with the the what is the marketing problem what is a business problem what is the customer problem we are trying to solve in pursuit of the of the business objectives and then its its these people right I mean whether these analysts regardless of where they are sitting in the organization that are making their those assessments and recommendations that what do we kind of you know decide to build internally and what do we what do we decide to partner.
with an external body co-kura thank you yeah how do you how do you build an AI AI driven marketing organization whos a part of it how do you structure it how do you make decisions was it for me sorry no that was for murder Jackie thank you sure I think I think the build frizzes by a decision that issues raises definitely makes sense I think there’s there’s a lot of stuff that you would only be able to do with your proprietary.
in-house data right like you cant you cant buy a tool necessarily that’s gonna for example make a recommendation engine with the data that your company has and so that wouldnt need to be internal now its really important to to make sure that even if you want to adopt AI into your company into your business that you really stay focused on the end result of what it is that you’re going towards so you know if that means you’re.
actually going to be able to provide better recommendation and for for your consumers and that’s gonna uplift you know the firms improve your business significantly they you should definitely do that and that may require an in-house solution then you know you can obviously implement AI solutions by Licensings existing software which is fairly easy and in fact there is actually a lot of tools out there that leverage AI in some way and in fact in probably whatever it is that you’re already using there is already a AI components in it and you.
know when I truly as a marketer your job is to realize customer value right you have a bunch of existing customers and potential customers and how do you make sure that they become valuable for the company now that’s become a lot easier with good analytics um you know things like customer lifetime value of analyses etc have become a lot easier even though the theory may have been out there long before like cloud computing became you.
know more widespread now you know those analysis can actually run so you just know a lot more about your customer and um you know whether or not theyre gonna churn when to upsell very different from other customers how did you segment your market in much greater detail yeah this is interesting and then Jeff Im curious so I know you had the portfolio gradient its all its all like a I focused companies but then are there you know like a Im marketing focused companies that are selling you.
know products specifically to marketing departments we currently dont have a portfolio companies where the primary buyer would be a marketing department but we certainly evaluate companies in that space and and you know my bet is that at some point we will make an investment in that space got it cool yeah yes interesting yeah I guess uh the reason for the question which you can still probably shed light on is like you.
know one of the big stresses which you guys alluded to slightly is okay were the big company we have all these customers and all this data heres the cool startup they’ve really amazing models and smart people but they dont have the data how is that navigated you know like do you guys have clear policies around which data sets youll actually give to a third party to use or is it totally case-by-case and POC by POC what do you mean to give to a third.
party to use so I guess in your case with your portfolio companies like whats whats their experience like working with customers and actually having access to the customer data well well you can you can only really be an effective AI company if you have a certain data set that is really huge right that you have access to and you know the data in general is extremely.
protected its not something that’s freely shared amongst company and so you know the example that I just called out which is Elsa which is where you actually speaking to a phone the data is all those sound bites of how someone pronounces the English language you know we have other we have other types of companies where for example accounting and bookkeeping is automated right well those are the data that specific company that they they can then analyze and.
automate it as much as possible and much further with with AI tools and so there’s lots of industry verticals in which we operate where companies you know run run analytics on a certain specific data set but really the amount of like data sharing between companies is is fairly limited which makes sense right because that’s kind of their um their raw raw ingredient in a way yeah thanks Jeff Im Eric so right you’re helping these I think you said 800 marketers at uber you.
know market the various products you guys have hows it set up how are you guys successful with that we actually over its all the Isaac in-house and there are a couple reasons for it like most of the data is sensitive and we cannot share with third parties but the main reason is the more sophisticated you get and it is easier to build everything in-house and exact so it operates in all around the world and every every country and city has nuances so it would be really difficult to build something with a third party and to.
capture all that nuances and as we switch to more value based optimization and when we do performance marketing meaning we calculate what is the profit from each of those users and what is the top-line revenue and then that set targets CPAs its not something we want to share we only share with Wall Street every quarter that for that reason we are like all the impasse in the performance marketing side on the brand side its more like agency SEC more like.
a mayor and his campaigns but the I dont the AI is like much less on the brand marketing side but performance is like all the way I drew them got it got it yeah that’s interesting that’s there’s almost like an internal agency and and then assigns team engineers project manager that gives us works God so the agency comes up with some gold they have some problem theyre.
noticing and then they can salt with the data science team and they spec about its most of the good about the automation it is as she said like we they are in a certain number and search campaign is hundreds of thousands of keywords and we automate that how much should we pay for each of them so and based on real-time data based on so there are no budgets right we just say the targets cost per acquisition and we need to calculate it regularly all the time so its more about like automation.
and skill scaling their campaigns got it cool thank you all right let us proceed so now you know you guys have mentioned some kind of use case examples but I want to dig a little bit deeper into you know what what are you finding its like the most valuable sort of Wow example of you know AI and marketing in the current year 2020 I can I can go first I would say theyre.
actually all across the you know spectrum and marketing right starting with I think targeting used to be the most kind of you know I mean has been around the longest like you just get smarter and smarter at targeting working with large large amounts of data I was surprising is one that at the power of AI in making pricing decisions or recommendation decisions not necessarily price or what is the next kind of best action or best decision or best you know cross sell opportunity I.
think I think those are those are some of the things that Im seeing a lot of promise and n value being continuously unlocked and then I would say on the other end would be also things in messaging optimization like we we worked with a with a partner and its anis their platform and model Presario and they helped us based on years and years of knowledge and insights that they’ve gained on our customer segments on their platform with all the testing we do it helps us you know really increase the.
speed to market in creative generation messaging generation because you can basically say hey I need to send this particular type of message 200 people like me and then the platform based on prior learning said this is the choice of words then the brand guidelines and you can spit out a complex test to pick the winner so I think its those cancers its all across the marketing value chain that AI is unlocking value and.
really excited about it that’s cool yeah that the copy generation example you just gave is interesting and like right so empresario you know spits out a draft a recommended draft are you finding that you know a human is then having to usually edit it or its coming out totally ready to send you know I mean there is there is still I mean you know going to be some human intervention but its less about but more just around making sure that you know hey from our brand perspective and everything like is there anything that you know we need to just you know.
discuss a review but now its said that the changes are minimal its more you know the human intervention is more to just a low touch review to say what is going out is it still kind of in line with what we wanted to do that’s yeah yes for most of a I in our portfolio as well um there’s a lot of you know predictive analytics that’s being done in a wide range of industries where at the end of the day you’re still gonna.
require humans in the loop which is probably good thing to do to a sense of reviews yep yeah as far as humans go though I mean so now that you know AI is reading the emails do you in theory have maybe at least one less copywriter yeah I mean there are there are going to be you know all kinds of productivity kind of gains right because of that you’re right and maybe the I in our case or the way I.
like to think about it is that copywriter can actually then focus more on some of their you know create a creative intent generation or really looking at you know the results from some kind of you know what these tests are telling us so they havent you just you know your your your your your focus and resource allocation changes yes yeah yes that’s interesting your murder how about you whats uh whats maybe you go.
wow when it comes to you know flying III in the marketing domain I agree with actually I think the pricing and personalizing pricing you by the use of promotions is you can like have a lot of growth with it and you can target your promotions to only the price sensitive people and that will leave to incremental revenue and you can do it at the right time and you can in many different ways you can also like do it.
for people who have churned can bring them back for you pretty someones going to churn and you only send it to people who needs it and you can prevent churn and you can again cross sell upsell and we made this like context you like band a so you can do it at the right time and and so the premium product like black car premium rights and those are you can do this in real time those are at scale really recent things and also again like scaling up at.
running global campaigns with dynamic creatives like in hundreds of countries and showing the right creative for example we have a display product where we can join auctions around the world through is works and you can show impressions and we can show the impression at real time based on what we predict is gonna ill that’s the most value it could be a order from Eagle eats we take it right got it yet that that’s cool yeah Im curious you guys kind of made me think of a new question so you know Ashish you mentioned Prasad.
oh and you just mentioned beeswax Birds but what are some other like your just tools in your toolkit that you find super valuable ideally AI related so I I Im also seeing Lindas comment in the chat about profitability segmentation and so in in terms of tools like from an analytics perspective just the amount of analytics you can do in your on your customer base.
is so much higher and so so it turns out that if you look at your customers that their recency with which they purchased the frequency with which if they purchased and the amount of money that they spent are the biggest predictor of how this customer will be in the future and so you know those are there’s a lot of research actually done into this area in terms of predicting the profitability of your existing customers so that’s something that now can be done better.
and there’s plenty of tools in that space that are I think at the very core of marketing right because once you know more about your customers then you can actually go to the University Communications and it contains etc that have emerged and so you know some examples of those are like gain side and bolster ah and um catalyst and and some of these companies are also being acquired like zodiac is a company that I think was acquired about one and a half years ago they cant came out of Wharton that did exactly this mostly for.
e-commerce players to understand like who are my most valuable customers and so who should I focus my campaigns on or my communications and in what way um so those those are some examples and I think just analytics in general yeah but yes so that’s some good examples of profitability segmentation tools yeah that’s awesome Thank You Linda how about how about you guys murder sheesh any other like tools that are just you know.
cant live without obviously you know Adwords and Facebook but what are some more more nuanced ones yeah Im definitely what most of our spending on Google and Facebook and we try to use their optimization capabilities as much as we can and we also control other things like our beats and campaign setup and also smartly to Liam this allows us like to set up our campaigns faster cool.
smartly tell you nice thanks have you see Jenny is prasada your your horse or you have any other yeah we have a variety I mean helium as well I mean more just mentioned and then we extensively they have a deep partnership with Microsoft for their suit of products and then there are some niche kind of you know products and partnerships as well that are very health care use case specific there are examples Im forgetting the name of the particular partnership but this example and you know some of the call center.
analytics voice analytics emotion analytics to predict turn things like that so its across the entire customer journey ranging from your you know your big kind of big 800-pound gorillas in the tech space to focus on each players like personal cool nice and then someone sue here has a specific question do you guys have an opinion on sales forces unsigned product and its okay if you dont just wanted to give Michael Tuohy.
sue we have no opinion on that I took someone else asks a question yeah theyre asking a question sort of sales questions around like you know how do you avoid information overload like sometimes access to too much data at least the arguments of the leads and this was a source of truth but Im interested to ask you just about data in general and just maintaining you know customer data and whats that pipeline.
like yeah interested to hear just about your database and how best best practices that youve learned to maintain a good customer database I mean in yeah its definitely the most important thing because I think any of the best models but garbage in garbage out and we have like a lot of data engineers maintain data pipelines but its usually a there’s a source data like individual events and then there are other.
pipelines that aggregate daily hourly and then there’s like layers of layers of pipelines and we have like a very good infrastructure team here but we also do guardrails and checks I get something baby eights a lot from expected when we get alerts and the issue is on the performance sites there are striking issues sometimes like you’re we are especially self attributing networks you have to accept the comers back to them and then they.
actually come back to you and actually another company for that emissions branch it handles our App attributions so then that’s out of your control if they have a problem and you have to work in terms of again guardrails and monitoring its very important nice any thoughts on data management no I mean I would I would agree with more just said about you know garbage in garbage out and again my earlier comment around how culture is the most important part because culture and talent that leads to that appreciation for data and then investments in data engineering.
data infrastructure to ultimately have you know the right you know not only the right data infrastructure decisions but data governance decisions as well because at the heart of all of this AI and machine learning and all of it is data right so having clean actionable data that is easily accessible and the data a taxonomy is consistent across the organization I mean that all becomes so critical so I mean +1 towards a smaller said that’s how even we are kind of investing in structured within Humana which is you know a big group of smart people across data engineering data.
scientists data governance folks putting all the data in one place so that its easily accessible but also being a regulated industry you know your access is obviously controlled because not everyone needs to have one important thing is Shoom higher data scientists I should be more engineers than later scientists maybe one to four ratio and even more and really good data hygiene yep absolutely nice yeah totally totally get that you.
mentioned sheesh right now health care super regulated but actually is there any like right regulation on like the marketing side for I guess in general across you guys work or I guess specifically is there like marketing regulation in the healthcare domain that inhibits you from using certain like maybe AI strategies I I wouldnt say you know inhibits I would just say its its more around that you have to be just.
mindful of you know things around like your your model might make a recommendation Im just gonna take a pic of non-healthcare example to make my point here your model might make a recommendation and say give up price one tomorrow you know a certain price to motor and certain price to ashish but you know because of the nature of the business and a combination of the business that put is in the best in the consumer interests and the regulatory landscape that’s not feasible so because that’s not how kind of thing that.
doesnt hold good so I mean its its some of those things that you have to be mindful of and also you know some of the things around regulations what they mean from our testing perspective its kind of somewhat some of the a be some of the things that you can do a classically a/b test in a financial services you can do.
in health both because of the way business and clans are structure but also sometimes you know it doesnt even make sense I mean its one thing to give two different price points but its never a good thing to recommend two different you know health health outcomes or health strategies right so there are just nuances around regulations and the me and the nature of the business that you have to be mindful of and what you can.
do and what you should not do because at the end of the day its all about the customer we are a very very member focused business and need to make sure that what whatever we are doing it has to be fair to the consumer so to add to that we also for the on the burner sites for uber that drivers couriers we we dont do any like differentiation because those are highly regulated areas.
not only for the consumer like writers and heaters here so you have to usually you cannot use AI that much exactly nice driving permits and from a regulatory perspective I would just recognize that that that that landscape has definitely changed a lot in in recent years starting I think with gdpr in the EU that is definitely a very strongly influence the way that.
companies think about leveraging customer data for marketing purposes and you know following that there is other regulations like the consumer Californian consumer Privacy Act um and and I think you know a lot of them are gonna follow suit so its definitely something that I think a lot of people are thinking about in fact its kind of you know its given rise to a whole separate sector which is called reg tech.
where you know companies help other companies deal with this and also consumers take charge around data so definitely Im definitely a whole different space which is something to to be thinking about its very like keeping the data for long term there’s always data termination Poseys yeah interesting and that kinda also leads me so someone here asks you know with analytics and predictions to upsell or cross-sell does.
the panel test for neuro diversity and ml bias around the LGBTQ community and to broaden that slightly how do you guys think about bias you know when it comes to getting models in production that were showing people products in trying to convince people of stuff how are you thinking about is in those programs so for us again and for marketing to drivers or owners couriers we make sure the audience is balanced like sponges you actually use minimal AI.
and as much as we can we dont use any attributes like for example related to sex race ethnicity religions and that kind of and we dont have most of those attributes but maybe sex sex and age you know using your our models nice yeah that makes sense that makes sense cool thanks and then sown here Hallel is wanted to confirm did someone say Tilly amor tiller helium helium there you go hello gotcha all right so let me go to my last official question.
here and then maybe we can get back to a couple of Q&A if we have time my last official question as you might imagine is future-focused so whats the future its 2025 I mean sorry its 2020 whats it gonna be in 2025 you know what what are your marketing programs gonna look like I I would say yeah what we are working towards and what I think keeps me really excited is a scenario again going back to my earlier point where your different point.
solutions that we are already using we have been using marketers have been using along the customer journey theyre actually talking to each other and a lot of things like next best action next most message and all of that actually have become a reality and and the reason they they are kind of still in a nascent stage is because some of some of the data infrastructure opportunities that more talked about right I mean those are.
the things I think in the next five years given the pace of disruption in just the investments that organizations are making I think you are seeing basically customer journeys marketers thinking about customer journeys and managing customer journeys and the data flowing longitudinally and across decision models and actually leading the nudging them customer towards the next best action versus right now point solutions working a different point I think that’s that’s that’s how it looks like cycle a greater that I think in any company especially larger organizations that um that sequence its very disjointed and there’s a lot of companies looking to.
you know tie those pieces together especially from the analytics perspective and I think we have we see a lot of emerging companies you know working on very specific marketing solutions like the communications themselves it contains themselves but I think at the end of the day starting with the actual customer analytics is where really what the intelligence and what the most valuable pieces for the marketer and then from there onwards if from within those types of engines you.
can find plugins to actually run the actual campaigns that would be really strong combination but its a its a complex landscape and there’s a lot of change recently in terms of what is technologically possible so I think were gonna see a lot of movement in terms of new companies coming up and growing in the next few years cool nice.
yeah I agree with both of you its about the customer journey and intervening at the right time and right over and my guiding them and doing this at scale real-time at scale that’s the key do you guys think voice is gonna be a bigger part of your marketing efforts like you know I have multiple Alexas for example yeah it is voice part of what you think about for the future like are people.
already more rides on uber you know buying healthcare stuff through voice I I believe it will be however there’s a caveat around there I think from a consumer behavior perspective the classical tension between here I know its very convenient but I dont understand the privacy part like I dont know what all Alexa is hearing all right the the the the classical concerned that you know any any coffee table conversation so yes I mean there is.
there is something around that which I think um Zimmer is you know businesses have to inspire and educate consumers and inspire confidence but yes I tighten engagement with voice devices and engagement through voice I think that’s the future that’s where its going yeah yeah I think absolutely I think I think there will be a lot around voice that’s.
that I think will they see will be seen as like one you know one other channel and at the end of the day like voice itself its not is not worried the actual intelligence lies where again where you can actually um do the customer segmentation that is so important to marketing as it as a discipline so I see it more as a a different channel that can be activated and something that’s um really changing marketing as a disc Linnet at its very.
core cool cool and then someone that’s kind of follow up Sean just asked where do you see chat bots going so I guess you know Ill just text text AI versus voice do you guys use chat pots yes yeah right its a super super crowded space like there’s tons and tons of chat bot companies and the premise of courses makes a lot of sense and I think there’s.
a lot of companies that have well I have a lot of success adopting chat BOTS so you know absolutely but I think you know there is gonna be a lot more creative ways in which a company can become an interface and interactive interface on a web page on the Internet and that doesnt necessarily need to be a child chat bot itself right it can it can also.
just be on the web page more interactive interfaces so chat bot in its in its you know forum that weve seen it emerge in the past years I think that that space has some challenges but I think its its very much of loving and I think were gonna find see other ways emerging which that um inter interactive web pages may actually be more helpful than than a an example chat bot cool cool thanks alright so just about a time weve worked through quite a few questions so before we officially sign off so the.
three of you you know you’re you’re each experts in in your domains and you’re talking to a bunch of marketing people its 20 20 a eyes a thing what piece of advice would you have for people whether its what they should learn think about having mind get end with one piece of advice from each of you start with data start the data and culture that’s to me.
that’s that’s the fundamental everything else I dont want to say jargon but everything is you know like buzzwords and you know is the easier part but if you dont have the right culture right talent because culture precedes talent and then theyre clean data everything else you dont get then you dont get any ROI out of platforms models every culture is the most important thing because you can make data say anything analytical and critical thinking culture and I guess on a low-tech note I think just in marketing in general even though.
there’s a lot of data and I stuff going on like dont lose sight of your messaging that’s still like very core to the discipline and that’s one of the first things we have to work on with the companies so um dont lose sight of those sides of that of marketing as well nice well curious end focused on mostly human things culture and and messaging but uh that’s maybe fitting the human piece is not gonna go away even though AI is becoming more of a thing all right.
massive thank you to the three of you for this awesome discussion I think people learn some good stuff and seem very engaged and thank you everybody for joining murd job Ashish Godspeed have an amazing rest of your week and yeah I look forward to talking to the three of you again soon invisible claps for everybody thank you cream this see you guys bye everybody thank you.
Leave a Reply