How Can I Sign Massachusetts Banking Word

How to industry sign banking massachusetts word mobile

[Music] i'm gonna get going straight away and uh we can keep it um obviously this is a discussion we can keep it kind of informal just natural if you want to say something please do and i'm just gonna try and keep the conversation flowing i've got a few questions and obviously we're going to speak about our expertise our perspectives on financial inclusion and obviously thinking about africa as specifically and see what obviously financial inclusion is huge issue here but it's a it's a global issue as well so one can talk about um that is more broadly so um and and the other thing that i've obviously noted i i'm connected with swin and paul and i know your analytics background but moodle you also have a strong analytics background as a matter of fact maybe we can start with you where we we each of us just quickly introduce ourselves and and there's a starting point there and then we can tackle some questions so please go ahead moodly perfect thank you thank you pleasure to be part of this um session so my name is marilee voloswar i am the head of u.s consumer analytics for the city group or city bank and in that capacity i'm on a mission to think about how we could as a bank build deeper relevance and engagement with our customers on a real-time basis and understand them in the proverbial customer segment of one that's my mission statement in effect my career has spanned the intersection of where data meets strategy needs change over the last 20 plus years i was i was in the insurance sector for the longest time and started initially in consumer banking and now i've been back in the banking space with city for about two years oh fantastic i see you've been at the boston um uh consulting group as well spent some time with them as well yes yes so um so i was the chief science officer for aig for many years and when i left them thought i might do something a little bit less structured and different so worked with bcg and with a couple of startups for a few years and i'm still invested in those startups and then i thought i might make the leap back into the big corporate world what's interesting and challenging about the big corporate space is the scale the scale of the access to data and the scale of impact that one could have now that whole concept is also being debunked when when you've got firms that have been in existence for two three four years uh that are creating their own data assets and business models and are getting valued at hundreds of millions if not billions of dollars so so maybe i didn't quite get that right is it yet incredible opportunities i'm sure we're going to dig into that as well especially on financial services side site paul i just want to quickly go to you um so obviously we i've been also at joomla i think probably before you were there and my role there was just briefly for a year chief um uh i was the the data office chief data officer and also looking at data intelligence and so forth and i think you've got pretty much you're into that kind of role from that mistake yeah yeah yeah so yeah exactly yeah yeah so hi guys my name's paul welton i'm currently the chief information and analytics officer at gmo world and that that encompasses um all of our data and information um the entire university so data science and engineering but also our credit um and and business pipeline so all of our actual financial product design and the delivery of portfolio management maybe just a little bit of a background on what jumo is so we're a full stack technology company it brings financial services to the mass market so very much the i guess the easiest way to describe it is the intel inside into the into um some of the biggest partnerships in in africa so we sit in in between very big financial institutions and the mobile network operators and provide a full suite of products to their customers and you you specifically and i think very relevant to this chat using machine learning and artificial intelligence yes yeah that's it and your experience also standing you've been there for a while for a long time and also analytics and customer pricing so it's all very relevant to what you're doing right now that kind of experience yeah yeah very very very good um experience there standing back i think the biggest difference i guess it's the exact opposite of what merely is experiencing so coming from a big corporate where the focus was very large transactions specifically so ahead of customer pricing i was really looking at very big transactions and not a lot of volume so it's way more depth than with if you would call it that and ajumma it's the exact opposite where we've done a 100 million loans now in five years so it's it's very very quick and and and fast turning around but not as deep so obviously smaller smaller ticket sizes and a completely different challenge so very exciting to be on that side yeah wow nice fantastic and i want jimmy to also do well i think i'm still a little shareholder still there so i'm rooting for and we want success stories here in africa we want to create these kind of unicorn businesses so i'd love to dig into that um swing so voter phone michael you you're right currently um you've been at vodacom eight years and uh and now apparently i'd vote a phone so just give a little bit of background in terms of your experience and it seems like very much analytics and customer value management and stuff like that as well that's right jock so um at the moment i'm the base growth and retention marketing for vodafone and basically the transformation that i'm busy leading there is to drive fully automated ai driven marketing uh so a lot of work around personalization and recommendations um in terms of the actual work we do it's uh to cross and upsell our products and services to our existing customers uh turn in retention management and then also a lot of work around loyalty and rewards programs before i joined vodafone i was at vodacom i was responsible for ai and analytics so i started up the big data team at vodacom um and yeah i've been uh obviously for the last two years or so focused very much in the marketing space using ai to drive marketing but before that i've also worked on various ai use cases from mobile network investment optimization to the customer service space you know coursing to routine optimization problems chatbots and conversational agents and also fraud management i'm so very passionate about this topic this this just listening to you guys there's so much to to talk about here in terms of real world applications clearly you guys have lots of experience in that space as well maybe just a little quick background to myself before we get into questions and the topics as well so i've been um about my whole career has been building ai i did my phd machine learning ai in the 90s and then i actually did quite a bit of research and in terms of just applications chemical engineering even did some stuff we built neural network models of brain disorders got a book published on neural networks and psychopathology and i'm currently writing a book and almost pretty much done on democratizing ai to benefit everyone and that ties in with my massive transformative purpose around how to shape a better future in a small technology era and how can we use ion smart technologies to actually create that so it's a lot of thinking around those kind of things as well but earlier on for me it was it was all about the applications how can we unlock business value societal value with ai and my first iac company was season systems i i that was about 14 years and uh initially we did kind of everything in terms of applications so i was also financial there's some lump symptoms a lot of customers in south africa insurance and and bank so for banking customers retail and so forth but we went into we decided south africa was obviously mining and we went into minerals metals mining and manufacturing so i did quite a bit of work over probably two a decade and a half in that area and then eventually sold season systems to um to general electric in 2011 so it's a great experience and that not only obviously building that out internationally spend quite a bit of time across all the continents we had a distribution channel across the world in terms of just taking our software and and and delivering solutions on that and then worked for general electric spent quite a bit of time in the u.s as well it was in silicon valley and chicago specifically um because we had our asset management unit there and we were people all across the globe in india the uk and africa and us developing the software and stuff like that as well so i've been in that space so it was a wonderful experience and then i joined juma for for a year that was fantastic experience because it exposed me to the mobile network area specifically and working with vodafone vodacom and mtn airtel tigo all these kind of things that paul will know extremely well but learned about the applications there if you don't have any credit uh any history how do you build it up with uh obviously looking at the the uh the wallets wallet behavior thinking about uh then obviously the call direct records so much information there if you can combine that uh opportunities to unlock value uh but then after that to short circuit everything yeah i founded another company called cortex logic it's a engine for business which is part of the fortix group now and we are currently in financial wellness and health wellness we've got also some trading solutions that we've developed um so some very exciting stuff that's all relevant to our top discussion today so fantastic so let's dig into it so financial let's quickly talk about financial inclusion so i think the the topic is how ai will impact financial inclusion over the next decade so as we sitting here 2021 a lot of states it's a big question if you think about that and think about what will probably happen over the next decade or so what will realistically happen and i'm actually i've written a book now on that where i'm thinking about the future and from all perspectives and all industries but if we zoom in on financial inclusion and we can talk about financial services in general because i think you will probably get disruption from all sorts of angles um and even the tech giants are coming in to play here but let's just zoom in on what's your perspective on how i will impact financial inclusion over the next few years so moody let's maybe start with you what is your perspective there from new york thank you and it's um you know really quite an honor after having listened to each of your experiences i realized even more so what elite company i'm in so it's a it's a pleasure um so when i think about the role of ai and i'm going to abstract a little bit and and sort of put it in the context of advanced analytics using data more intelligently on a real-time basis to better understand a phenomenon and then to be able to predict it ultimately to be able to influence outcomes as well if i think about the u.s market in its own way over the last 25 years there's a couple of firms one in insurance and one in consumer banking that made a massive massive mark in disrupting the lending slash credit card sector as well as the auto insurance sector and those firms are capital one and progressive and i happen to know both of them very well because i did work with them in in years past and what they did which i think is very telling for where i expect the world of advanced analytics machine learning ai to take us is to get to this notion of d averaging so um you know where i think the problem of financial inclusion runs typically against a brick wall is that we tend to cluster customers in two broad groups of uh their potential and their risk and their financial health and and therefore their value as a customer i fully expect that where ai is going to have a big mark is in continuing to de-average our understanding of risk so getting to this notion of there is there is no such thing as a bad customer or a bad risk there's only a bad price so literally you know at an extreme getting to that notion of segment of one that is the big big big um you know phenomenon that i expect will will will be part and parcel of the innovation number one number two is um not thinking of risk and behavior as being um exogenous or uncontrollable but rather building more interaction and engagement with customers in a way that influences outcomes in a way that preempts the likelihood of a customer going down a spiral of debt and not being able to get out of it so being able to not just predict but influence outcomes and engaging with customers uh to take mitigating actions such that banks can continue to support them before before they get themselves into financial trouble the third piece of it to me is this notion of timeliness and and using more on credit bureau and the problem with credit bureaus it's sort of a chicken and egg where if you don't have enough of a history then you don't get access to credit quite as easily and and in order to build history you need to get access to credit so uh where the world is turning to is is recognizing that while that data and that history from a credit standpoint is important there's a lot more that is knowable about customers um that can that can help connect to a deeper understanding of risk and that can expand the access of risk of of financial products and and and credit to a broader base of customers and i think that were very early stages of that um globally and and even frankly within the us as well despite how developed it is it reminds me of this notion that we tend to over predict innovation in the near term and tend to under predict in the long term and last but not the least is this notion of digital which is uh i expect that the cost of delivering these products um and and the access to credit is going to go down dramatically over time because you're not going to have quite the same degree of a fixed cost overhead um and that infrastructure that that can be sometimes a little bit of a noose around some larger firms and i expect that the role of that will diminish over time and and and and that allows for the prices to drop as well so those are some of my thoughts on how i could imagine ai transforming the financial services sector in the next decade i'm very much alone with that i i would love to respond to all of those points but i think you're absolutely spot on i just want to give everyone opportunities so paul what's your what's your perspective how do you think about it obviously you're working on the in emerging markets as well so africa i love pakistan you're looking at indonesia without south east asia as well yeah yeah thanks so much and merely that that was really interesting to you that perspective i think it's something that we live and breathe every day in our company so very very interesting to hear those perspectives and how aligned we actually on that i think just tagging on to to what you've said there i think um one thing that we definitely see is that um there's no such thing as as inherent customer risk is what we believe like every customer as deep as you get to personalization there is a a product for them and getting to product fit is really what what it's all about so you can serve as many customers as possible if you can get to that product fit i think the the one thing that we truly believe is that pricing specifically is not an output but an input and if you think about traditional banking and specifically traditional lending you would kind of um develop your cost tax so what's your cost of funds where's your your fixed cost going to sell and eventually get to some sort of return on equity margin and price at that and that as being an ex-head of pricing for for a large bank that's how we used to do pricing where in the emerging markets and specifically the financial inclusion when you start thinking of pricing as an input and how that could influence a customer's behavior it becomes very powerful tool to change product fit so the base example of that would be agricultural lending so um if if there is a news of a drought coming in a specific country on a different area usually what a creative team would do is to s y risk is going up so let's minimize our exposure so let's not give these customers the same kind of level of exposure that used to have and increase the price slightly because we've got risk-based pricing because we think there's going to be a drought and then what happens is the drought happens the customer allows a smaller facility to actually get them through the year and it's price higher so you default and your create model gets a tick and says yeah we got this right but the customer is kind of down and what we've seen is like turning that around and actually when you see the stress on a customer trying to drive price down continuously even for your highest risk customer gives you an amazing outcome where a customer under distress is actually the one that needs a bigger discount so when we when we talk financial inclusion that's really like how deep do you want to go and not not formalizing or not focusing on lowest risk highest reward which i think was kind of where the entire world was at for for the last 20 years to see how do you minimize risk for the biggest return i think the other two points that also tied with what merely said for me is it's the first one is um unit economics so when we talk true financial inclusion we we need to you need to get to a fully digital space where you're leveraging the cloud and everything digital as far as you possibly can like the unit economics of most traditional financial institutions mean that anything under 10 000 um or let's call it a thousand dollar loan um is is not profitable and if you start looking at the real big market so jack you mentioned pakistan one of the markets we're in if you cannot make a profit of a five dollar loan um or a twenty dollar investment you're not going to be able to really financially include like nine percent of the the base which is which is really a very very important part of the equation so i think that's that's a few things that we've been really focusing on focusing on actually inclusion going as deep as you possibly can and then definitely i agree with mooley it's the approach of the to of how you um use ai i think the last point for me is just i think ai and fintech and a lot of these companies are associated with disruption quite often often so we hear a lot of great success stories of how big industries get disrupted but it it's it's something where i think it's not always necessary like um collaboration is is probably one of the most powerful ways that we can get to specifically financial inclusion if i look at the african continent specifically trying to disrupt a a existing market they would mean we would target the same customers go for the same small piece of the pie and trying to kind of get more um value from those customers which should once again be the low risk high profitable customers but collaboration where you've got very strong institutions with cheap balance sheets you've got big distribution channels and then using tech to kind of go to a new asset class that doesn't exist is i think what is key so i really like to see a lot of the new fintechs coming up in a collaborative way not looking at the biggest play in the market and trying to take them out but rather looking at big banks and big institutions that's been around for a hundred years and collaborating with them to financially institute paul that's interesting in terms of new asset classes what what do you suggest from a financial inclusion perspective any specifics on that well for for me i mean what i mean with the new asphalt is it's really just a a segment of customers that are unattainable to any of these traditional businesses right so just from a peer unit economics like if you can get to 100 million customers and we're talking a seven day twenty dollar loan that's something that a big bank i mean if i'm sure merely you sitting in manhattan thinking about a a ten dollar seven day loan is something that's that it's just not on the class but if you do that for 100 or 200 million people across the continent at a 30 roa it becomes something big right so it's it's asset classes where you don't have to compete with traditional banks that are already there or any traditional player where you can actually open a new pool of customers and for me that's the real definition of financial inclusion is can you get to those parts of the population that is actually not served instead of trying to just move customers from existing platforms fantastic swing what's your perspective you will you in the mobile network operator with all that rich data how do you think about things so i suppose i'm slightly outside of the banking industry so probably talking more from a customer's point of view but i think or maybe the first spot is uh i think digitization is key right so um i think this digital is almost made for ai and ai is made for digital because the futures go so well together and if we really want to want to use ai in a large scale you know we really need to do it in a digital way and that's how i believe you'll scale it but if i look at specifically financial inclusion and at the unbanked population i think there's two uh use cases that stand out for me one is very practical uh amongst the unbanked population they're still in some areas challenges around formal documentation so identity management right and i think this is where digital identity with the use of ai techniques like facial recognition combined with biometrics and creating a digital identity that's widely accepted between financial institutions and banks it becomes really powerful you know as a high technology to drive financial inclusion forward so i think that's one thing that you'll probably see within the next decade and hopefully sooner rather than later and then i'd say the second thing is really around affordability so i think one of the big customer needs amongst unbanked people is remittances right the ability to send money back home cross-border for migrant workers that sort of thing and um especially for smaller amount remittances you still see that fees make out a large portion you know of the transaction value and i think that's really where ai can play a big part in reducing uh banks and financial uh institutions cost space so if you take ai within the customer service environment you can definitely move towards combined with digitization you can move towards things like jackpots etc that reduces the cost based and then that should obviously uh hopefully find its way back into fees um and i think those are two really practical applications um you know that i that i think we'd see you know hopefully sooner rather than later yeah awesome um that touches on uh next topic and question that i want to quickly ask you if you think about artificial intelligence there's so many different use cases in terms even when you think about financial inclusion and all of that um clearly you can use ai within the first of all we know if you connect to lots of data if you instrument the world you get more data and you can mine it and you can get a 360 degree view of the customer etc but i think something like for instance since when you talked about chat box when you think about intelligent virtual assistant functioning has also a way to communicate with the customer trying to understand their needs but then also is a way to capture and get information that you can also mine that you can combine in collaboration with other data that you also have available so it's almost like a sensor in the one hand but it's also a communication channel um i think it's probably super untapped but it would be interesting to to just hear your perspectives on that chat on that specifically and your own experiences and strategies around that tapping into that and then also referencing what does it mean from a financial inclusion perspective even unbanked can you actually reach them you need probably a smartphone the kind of infrastructure needed around that type of thing as well so moody let's start with you yes um thank you jack i think your point of thinking about data more expansively is a very very powerful one and why that is important is at the end of the day if you're really trying to get to timely relevance um at a customer segment of one or close to one you've got to be able to understand the customer at that point in time in their situation in ways than ever before and that's the power of ai and that's critically linked to paul's point about going to cloud computing and being really able to think of data in much much more broader ways so a few things that we are doing at city is number one is we've made sure that we thread the customer experience across all channels so that the customer is experiencing us as a unified firm and as simple as that sounds historically in large institutions if you've had an interaction in the call center and then you walk into a branch those interactions don't typically speak to each other and so what we're doing is not only tapping into the data around mobile engagement or app app-based or online as well as a call center as well as a physical in-person interaction what we're doing is we're stitching that together and trying to draw deeper relevance and understanding and that requires the ability to do natural language processing and to translate that text into meaning both in terms of the sentiment and in terms of the actual words exchange it also means that we've got to build the ability and we have been to to to to think of vision and image as data as well to draw meaning from that so computer vision and deep learning related to natural language processing and then stitching together every interaction to be able to build that timely relevance so that we could either we could impute what's on your mind as a customer and the next time you log in that's the conversation we're having or if it is more of a crisis situation we're actually proactively reaching out to you via digital channels or other channels to engage on an issue that we see that you you're signaling to us so uh from from a large institution perspective um i think it's really sort of cutting across those channels and products in ways that historically haven't happened to get a 360 understanding of your customer to build relevance and engagement to what extent in the u.s with citibank and are you are you getting that right to stitch it together because i i find here in south africa and we have we've been doing quite a bit of work for some of the major banks retailers and it's it's like they still struggle to to put it all together and sometimes i've got these kind of broker channels and they don't have access to the actual customer data in a proper way and and but but in the us how do you experience that specifically is is there some banks that's getting that right or is it more the disruptors that's doing it from from the ground up from um i think that the plumbing is actually starting to be built quite nicely across many institutions and the raw materials for that are actually getting more and more ubiquitous i think the bigger challenge is how do you separate signals from noise so uh there's a lot of data that's getting thrown at you and it's it's it and it's easy to sort of over interpret the meaning from that and in our efforts to build relevance to customers um do perhaps quite the relevance in a non-positive way perhaps is what you could end up having so i think the bigger challenge is not the data flows necessarily but it's sort of that intelligence on the other end of that to say so what do i do now how do i do it and when do i do it and how do i then take that feedback loop and create sort of an iterative design of experiments type approach to make it a continuous learning culture and approach because as this data comes in and as you see an opportunity there's no single answer or you what the next action one should take isn't very obvious yeah so creating that iterative loop i think is the bigger challenge and making sure that you're not separate you're not confusing um uh noise for signal is is really the critical piece from a technology standpoint um being able to sort of stitch together the data the raw components are getting more and more ubiquitously available frankly yeah absolutely um paul just i'm curious now from even from a joomla perspective um because obviously you can use a on various places and i've seen how you do this with risk detection or affordability and those kind of things but i'm curious on the interaction side there's so much there's a big touch point there's a lot of information that you can get right there um and i'm not exactly sure we where joomo is right now in terms of tapping into that piece and to what extent are you making it more kind of ai enabled and tapping into that kind of data that's been generated as well because that's going to save your purpose in terms of what you just said in terms of just bringing to understand the user and bringing a product and solution to them and price simple yeah yeah i think it's a great question josh i love the way you you framed it earlier when you said that you're gonna get the cost benefit of cleaning up a call center and you kind of do a chat box you also get such great signal from individual customers i think for us the that is the the ultimate goal right like if you can if you can reach a customer directly it'd be great unfortunately and you'll know better than anyone on this call like 95 percent of our customers don't have smartphones right so you're dealing with a very difficult way to kind of in enhance which is which which in in the beginning of my career jimmy was quite challenging to think about we all understand that you want to mine some of that additional information but how do you get access to that and kind of put that those two things together i think the approach we took now is is is basically very similar that you would do with a normal chatbot as an example so where you use your cost efficiency and kind of match that with with with getting some more signal from the customer for us it's really about looking at any sort of very complex customer behaviors that you can can analyze and then the by not having the voice of the customer you've got two options right you can do surveys with the customer or you're going to have to do some really hard yard focus groups to then get your quant and your qual merged together so i'll give you a practical example is the way a customer goes through their screen so let's say there's five or six or ten screens before you get to the new loan the speed of what a customer does that the amount of times he goes in and out back fails trying to kind of borrow a million dollars then he drives a thousand eventually gets detained like all of those kind of interactions obviously those patterns are very important then understanding the why behind those patterns is what we are really trying to do so we've we've cracked it just to understand and segment those customers into very very granular level to seeing those kind of individual behaviors and it's all nonverbal and non-actual kind of comp rights it's just a customer's behavior on your platform which has become quite rich and you don't need partners that i thought you do that but then getting that match with some sort of qualitative information we actually understand what's going on here so a customer that keeps going and getting stuck at a certain space or keeps asking for a million dollars on a on a dumb phone interface you kind of like you have to start segmenting them with some sort of qual so it's been a a very interesting challenge for us over the last two years last two years but i think it's it's it's exciting to see how you can actually match those two and having a a good enough sample set with very simple surveys or or focus groups matching that back to a 100 million um data points is actually quite powerful so it's definitely a bigger challenge than in the in the smartphone world where customers can access but definitely the same principles um all too fast fantastic and to what extent are you guys still tapping into the cool direct records the mobile network type of data as well because when i was lost there we did quite a bit of analysis but it was not completely utilized properl so that's a big source of information because that's giving you supposedly real-time information transactional information even if it's aggregated but there's a lot of info there as well can you tap into that yeah it's very very big data so far so i think our um things are quite similar from from back in the day i'm sure you'll you'll recognize quite a bit what we're doing but the big difference is as we've now got to 100 million loans we've been layering that data and reached it so starting with with mobile network information so gsm if you want to call it voice and like real-time interaction with customers then you get transactions from the wallet and then it really becomes all about behavior and loan data and for everyone on the call you guys would be very aware that having a real deep training set of data to train on is very important and that's been one of the big focus for us over the last couple of years is making sure we've got the deepest training set that you've actually tapped into all these segments that you try to see so we've been layering our data quite extensively like starting with the quant making sure that you as a customer stays on your platform you learn more so there's no there's no customer for us where his first interaction and his 16th interaction with the platform that you have the same information you have to be learning more about who this customer is at every interaction otherwise you can never see separation from a risk perspective so definitely the fundamentals from the day that you were there still there but it's like now building onto that to make sure that information is getting richer and richer fantastic yeah no it's it's um swim so i know you are you've got also specifically that personalization experience you talked about offering customer experience and and also digital mobile first bank of the future and those kind of things you want to talk about but what is your perspective because even from a mobile network operator that customer engagement and how you can tap into that and the types of data that you've got available in terms of offerings what is your what is your perspective on that so um so i think with obviously as uh mobile network operators we've got uh very rich data sets right so there's a lot of interaction data there but it's also very important obviously one of the big things is to um is to actually work with customers permission and consent right so especially in europe we see gdpr just kind of the gold standard when it comes to uh dealing with customer data and customer privacy and permissions um and and i think that's something that uh you know as a company we kind of uh we've adopted those principles and those gold standards across you know all of our operations so when you've got that detailed low-level granular data of customers it's very important that you that you obviously uh treat it respectfully and in the right way where we have customer consent we are able to do amazing things right so when it comes to personalization i believe that every customer deserves to be treated as an individual you know and we can create real personalized experiences for customers i know mooley also talks about the segment of one a lot or you know going down to that that specific customer um level and and we've done that across various aspects so i suppose probably the easiest place to do it is across offers so you can take your whole product set that you have of all of the services and products that you offer and you can put those together with a price in a personalized solution for every customer right so you do a combination of things and uh and using uh using ai i mean that's quite easy to do but we want to effectively personalize the whole experience for our customers so if you if you look at app journeys as an example the way customers interact with their app journeys and how they prefer to see those journeys and we can personalize those as well so things like dynamic creative optimization creating different images for different products for different customers those are the sorts of problems that we work on and we've really seen great um great results from uh from personalizing more i suppose the the challenge we we have is that there's various decision points in order to deliver personalized experience for customers and what we found is if you try and do everything at once if you try and first last offers you try and uh you know personalize the time choose the right time for the customer the right channel and also the creative and the copy uh if you try and do all at once it's too big a task so you know it's really a case of breaking it down at these smaller problems um doing some pocs uh experimenting with things and then scaling them and then bringing the system together in an automated way yeah yeah there's so much so much on fact there um so i just let's just go into think about um challenges as well what what are what is the biggest obstacles that's preventing the banks from deploying these just tapping into the potential of of ai and we know it's on so many different levels it's obviously on the data level there's so much to unlock there i mean just stitching things together um what just give you an example some of the financial wellness and health wellness solutions we're building we are connecting to so many different types of data sources say in health wellness we connect them to health risk assessments the clinical measurement data the medical claims data all of that information is important also if you think about financial wellness because that also affects health wellness and and vice versa as well so we we're connecting to ouii data um and other things as well and then it's about mining that and then we build up a kind of a wellness profile of this particular individual so that you so that it can inform the backend intelligence to be relevant in that moment in that conversation um and and uh so important that personalized interaction so for us that intelligent virtual assistant as part of this whole application is such an important area of engagement and if you think about other things like these mental health issues as well and and how to deal with that which is more kind of qualitative as opposed to quantitative things like clinical measurements and stuff how do you bring it all together um so i i can see some challenges and obstacles and we are building now platforms where we bringing all everything together and building it up from scratch but if you've got a lot of legacy if you're looking let's say citibank there's obviously legacy so how do you move from what is the kind of obstacles that you specifically deal with in in that kind of environment and how do you what can one do to actually overcome that goodness um that's that's that's at a deeper level so how do cultures and and firms and people evolve is the question that you're really asking john i'd put it in three things i was i was taking my notes here as i was thinking about this number one is um culture and um and what i mean by that is um we as human beings tend to be victims of our own success um our natural inclination is to be expert oriented not not to be more dynamic and sort of agile and learning oriented and growth oriented not that we don't like growth but it's just that when it comes to our careers we tend to lean on what has worked for us in the past yeah and and try to extrapolate that into the future versus uh challenging our own assumptions and recognizing or understanding to what extent is that foundation also shifting so that's probably the the biggest thing and look at the end of the day when you think about ai and i don't care which firm uh which industry but if it's a sizable firm you're talking about the what and how of decision making is going to change yes the questions you ask and how you solve for them and how you make decisions is going to change that means that people's jobs are going to change and that means that their identity and their sense of self-worth and their confidence and comfort with making that leap is also affected it can be positive it can also not be positive so that's a massive massive um you know i'd say consideration when we think about the pace of change um the second is i would say um uh uh imagination is um in order for this to really kind of take off on scale in more established institutions that ability to sort of reimagine what the world should and could look like that ability to start with a clean sheet versus you know the beauty of many many of the entrepreneurial efforts is they start without that bias you know they don't start with that um i'm gonna use the word baggage but it can be both positive and negative and i think paul's going to speak to it in just a moment because he'll have some wonderful perspectives on this and then the third thing and this is also equally critical i think is going back to what paul was also alluding earlier is when you think about expansion um you know you could get a very good uh return on the on equity on that margin so you're it's a it can be a phenomenal return but if you think of it in sort of pure how much does it actually deliver to your earnings as a percentage it can be pretty small so it's almost a mindset shift around how you should actually think about that segment and making it profitable and you have to actually focus on the profitability not the profits and i think that's a shift that typically in in larger organizations while there is sort of good intention um it's it's hard to get that mind space of senior leaders i think to to to sort of uh switch gears and and and so frame the opportunity and the potential in slightly different light uh and so you end up and and i'm not talking about any particular institution here but the the the bias is to end up fine-tuning what you're already doing for the most privileged of customers that where there's an appropriate amount of scale versus thinking about that expansion yeah that's that's that's uh awesome i i fully feel you um paul what what's your take i think i think a really good good question jack and their point on legacy systems i think it's one that always comes up i think it's the the rate of change of technology at the moment is getting quite scary so if i listen to you to the the my two colleagues here on the on the cool like from both of you guys being a very big institution it's like five years ago um being a bank was scary because it was big legacy systems and you were seeing like ai and all of these systems kind of get get really big adoption and thinking okay how am i going to turn a 100 or 200 year old company into this but even now for a company like ours jumo being five years old like that massive advantage of starting it from a from from a blank page like legacy and that baggage you're referring to really it generates pretty quickly so i think you you're spot on where having that ability to see into the future where this is going is becoming very important because playing catch-up is just not possible anymore like the rate of of change has definitely increased so there was a time where you could wait and see what the technology does and this is where it's going and then play catch up if you had enough dollars to kind of get yourself there but it's so scary at the moment where like by the time someone catches up like it's completely changed again so data processing is one of the craziest ones for me like the things you've learned third normal form data warehouses it doesn't exist anymore there are people that studied it their entire life that a phd on how to optimize the data then i mean it's it doesn't exist in the world today so i think that rate of change is is making it it's quite difficult and and the ability for any organization to to move forward so once you see the opportunity i think the big organizations are really good at seeing what needs to be done right it's like they know what needs to be done like they're very very smart people knowing how to do this it's can we get out of our own ways and actually do that before it changes again because if if you if it takes two three years to kick it off and get the initial kind of plan altogether you might be too too late and we've we've even as a kind of startup started experiencing some of those growing pains as you get bigger and serve more customers that it's something to be careful of so i think it's definitely a job one of the biggest challenges is keeping up with the rate of change and then looking forward i i yeah i i absolutely agree i remember even the last decade or so helping customers even when i was at ge even in seasons but definitely at cortex doing a digital ai driven transformation for companies the industrial data corporation for example these many of these frameworks talks about obviously it starts with you the strategy will actually start with the business value drivers of the business as well you want to obviously create your own you know you to create that story but people process this um and and the technology the data all of it is very important elements and you need to say how do i move from a state where i'm optimizing all of those kind of pieces um and and as you say paul i think in a world where things are changing so quickly probably in all those kind of dimensions the moment you put in this ai solution it's going to affect the other task it's maybe going to have a downstream effect on other people and processes as well so you've got to think holistically about this and it's obviously the ethical trustworthy side of things as well the privacy things there's so many things taken consideration um but anyway so yeah lots of different obstacles swing what what what is your uh from a vodafone perspective you obviously deal with legacy and things like that as well um can you move that ship faster what what how do you disrupt in that space or how do you move quickly how do you think about being agile are you guys trying to do to to do that absolutely joe click i think um the legacy systems always be a challenge and i think that's really where digitization um you know and digitizing the business comes into play because once we move more into the digital space you know to we we use newer technology new systems etc and then to do ai becomes a lot easier um i always always say that you know i would rather take a 20 accurate model but with 80 execution then take a 80 percent accurate model with 20 execution because it doesn't matter if the model is 80 accurate if i'm not executing on it properly i'm not going to get the benefits from it in any case and what i see is um you know we had to make the shift from re-engineering processes and redesigning customer journeys around ai and once you start doing that then i think you get the true benefit of ai and so that's really i think one of the big obstacles and in order to do that you really need to buy into it it's not just starting a data science department and putting a few data scientists in the corner and then trying to put a model into a broken process it's actually redesigning your whole business and the way you do things um so so i think that's uh that's kind of the approach um that we try and take as well um and and i think that's kind of one of the obstacles that a lot of organizations face is not really going all in on ai obviously you do it in a responsible way the great thing about the work that we do is it's highly measurable and we can do experiments right so we can try a step at a smaller scale we can figure out whether it works and then we can really scale it and then we can make it big i i if i just think about all these legacy in corporates and and what what i currently experience we live in the api economy plug and play it's it's like legal blocks and and when we make decisions we do super super quick and it reminds me of uh when cece my first i company i saw the general electric just that experience from going from where you control almost every piece the whole platform everything and then going to a corporate where you you stack and things move it's almost like there's different you can just see they're trying to be quick but but they can't they even had eric the guys from edge from lead startup and they're trying stuff so i've exper enced it in a corporate how difficult it could be to to to actually change the shipment and really you talked about the culture and all of those kind of things imagination and even if you've got all of that to execute and that's now the swings point as well even if it's about a driven solution that you put in place execution is absolutely that's what it's about how can you can you can you actually do something with this um and uh so it's so it's actually uh it's fascinating um i i'm just fortunate now in terms of what we're building now we're putting all the pieces very quickly together we can move very quickly and adjust to customer needs and and bringing customer sources external ones and just in terms of partnerships the way we partner suddenly we've got access to this kind of data and we very seamlessly bring it in and i think you're going to see more of these kind of digital players that's going to move very quickly utilizing with even with collaboration lots of different types of data sources provide a a better solution for for the customer and and its speed is so important now if you've got a magic wand next question here so a lot probably the last one the only few minutes left well and we think about obviously the frame here is financial inclusion so it's financial services if you think about the world where we live in what what what kind of you've got a magic wand what would you like to see that look like um and just from your perspective what is what is your what what would you like to obviously you can frame it back to say each of you on your environments that you work in so obviously with citibank if you've got a magic wand what would you like to see but then also thinking about that more broadly um what we'd like to see in the world in terms of from a financial inclusion from financial services perspective and also thinking about how smart technology can support that goodness um it's it's really sort of a uh frankly an extension of the themes that we've already discussed it's this notion of understanding customers and their financial needs at points in time and being able to not just understand and predict and serve but also to be able to engage in ways where we could make them more systematically intelligent in their financial choices so kind of creating that cycle of predicting and understanding but then also being able to engage and influence outcomes in a way that is symbiotic that's where i'd like to go versus predicting to try and sell something without understanding why or by thinking of behavior as being exogenous so i think really kind of making that a closed-loop interaction um that is uh highly relevant at specific points in time when a customer has specific needs um and through that making it the proverbial segment of one i think would be a a thing of beauty because then that that that that financial net will be cast much much wider and essentially reach anybody and everybody when they need it not necessarily always at the price that they want but at least giving them access that's market dynamics yes yes that's right okay paul what do you say you've got a few minutes left yeah i'll be i'll be brief i think i think it was michael porter that coined the the term creating shade value being the kind of the the the real organization is going to make it i think if i had a magic wand it's really getting fast forwarding into that world where where all organizations get to the point where they know the customer needs to win there should be no such thing as a grudge bird that purchased i think if you think about the carrot or the stick like you mentioned merely a career bureau that that stick of not paying and what consequences in my mind like ai is the real thing that with everything you've already said guys like segment of one and all of that personalization gets us to the point where it becomes about the character i don't want to lose access to this product because it's real time it's for me it's for paul and i'm got share value every time i do something which is great for the company or for the shell of that company it's sure value for me that to me would be the ultimate way of using ai where it's not it's it's not creating value in a skewed way for any particular party but it's really truly shared where people can kind of leverage their own financial identities absolutely sween what's your what do you think what's your perspective so um i think i i definitely have to say that i spoke about digitization quite a few times and i think that's really uh like if i had a magic wand right that would be it it would be that everything would be digital because i think that would make it a lot easier for all of us um i think definitely in the future we'll see more things being mobile first extending to wearables as well um and and then obviously ai first right i think that's kind of the uh the ideal world and and then it ties up with all of the themes like merely said around personalization as well you know i believe that in an ideal world every customer will be treated as an individual right so we'll really understand the needs of one's of every customer and then we'll address that both from the solutions that we offer but also the experiences that we offer and they really get rid of inefficiencies so you know where we've got parts of the value chain that do not necessarily need to be there or they don't work as well as they should or where we're carrying legacy costs from systems etc all of that should disappear at the end of the day and it should be value for customers um so i think that would be our ideal outcome oh absolutely i think you guys were brilliant i just want to maybe just say i i absolutely agree with all of those kind of comments towards that question um if i think more broadly as well i would love to see a world where um that's probably more decentralized where people are more empowered where you can even monetize your own data and and services we put everybody their own agent um where that's looking after them um where you don't necessarily have these big tech joints or corporates that's manipulating that or at least utilizing your data for their benefit we can spread a little bit more um obviously we would never get a a world that's that's equal you want to reward um obviously positive contributions to society and and and people that's creative and and all the best that capitalism is offering but i think we probably need a little bit of a balance as well so and hopefully as a society as humanity will get there and in the book um democratizing ayah to benefit everyone but one chapter dealing talking about the beneficial outcomes for humanity and i'm actually proposing they're a massive transformative purpose for humanity and there's 14 mtp goals that's linking to the sdgs um as well and and and the other thing is ideas is around can we create a ai driven decentralized type of platform that that's that's really helping individuals on individual levels and even communities and and make the future more local human and lets smart technology support us on that level on the level of one and then a family on a family level community level and then you've got this network of communities connected in the world you still can have countries and stuff like that but but but i think there's i would love to see kind of decentralized technology with ai and smart technology coming together and providing that kind of solution where it's it's just beneficial for more people love to see more people benefit from ai and especially in financial financial inclusion i think technology can make a big difference guys this was awesome talking to you to you uh we over time now and it's love to engage in the future as well um the input here was brilliant so well done to the organizers here africa payment club i think they did a great job um and it was fun talking to you guys so thank you very much i hope you have a fantastic day weekend and i have to speak soon thank you a real pleasure thank you seriously thanks everyone thanks guys thanks guys you