>>Welcome to the customer. Obsession begins with data session. Uh, thank you for, for attending. Um, at Cloudera, we believe that a custom session begins with, uh, with, with data. Um, and, uh, you know, financial services is Cloudera is largest industry vertical. We have approximately 425 global financial services customers, uh, which consists of 82 out of a hundred of the largest global banks of which we have 27 that are globally systemic banks, uh, four out of the five, uh, top stock exchanges, eight out of the 10 top wealth management firms and all four of the top credit card networks. Uh, so as you can see most financial services institutions utilize Cloudera for data analytics and machine learning. Uh, we also have over 20 central banks and it doesn't or so financial regulators. So it's an incredible footprint, which glimpse Cloudera, lots of insight into the many innovations that our customers are coming in up >>With >>Customers have grown more independent and demanding. Uh, they want the ability to perform many functions on their own and, uh, be able to do it. Uh, he do them on their mobile devices, uh, in a recent Accenture study, more than 50% of customers, uh, are focused on, uh, improving their customer experience through more personalized, uh, offers in advice. The study found that 75% of people are actually willing to share their data for better personalized offers and more efficient and intuitive of services >>Together. And >>A better understanding of your customers use all the data available to develop a complete view of your customer and, uh, and better serve them. Uh, this also breaks down, uh, costly silos, uh, shares data in, in accordance with privacy laws and assists with regulatory adherence. So different and organizations are going to be at different points in their data analytics and AI journey. Uh, there are several degrees of streaming and batch data, both structured and unstructured. Uh, you need a platform that can handle both, uh, with common, with a common governance layer, um, near real time and real real-time sources help make data more relevant. So if you look at this graphic, looking at it from left to right, uh, normal streaming and batch data comes from core banking and, uh, and lending operations data in pretty much a structured format as financial institutions start to evolve. >>Uh, they start to ingest near real-time streaming that comes not only from customers, but also from, from newsfeeds for example, and they start to capture more behavioral data that they can use to evolve their models, uh, and customer experience. Uh, ultimately they start to ingest more real-time streaming data, not only, um, standard, uh, sources like market and transaction data, but also alternative sources such as social media and connected sources, such as wearable devices, uh, giving them more, more data, better data, uh, to extract intelligence and drive personalized actions based on data in real time at the right time, um, and use machine learning and AI, uh, to drive anomaly detection and protect and predict, uh, present potential outcomes. >>So this >>Is another way to look at it. Um, this slide shows the progression of the big data journey as it relates to a customer experience example, um, the dark blue represents, um, visibility or understanding your customer. So we have a data warehouse and are starting to develop some analytics, uh, to know your customer and start to provide a better customer 360 experience. Uh, the medium blue area, uh, is, uh, customer centric or where we learn, uh, the customer's behavior. Uh, at this point we're improving our analytics, uh, gathering more customer centric information to perform, uh, some more exploratory, uh, data sciences. And we can start to do things like cross sell or upsell based on the customer's behavior, which should improve, uh, customer retention. The light blue area is, uh, is proactive customer inter interactions or where we now have the ability, uh, to predict customers needs and wants and improve our interaction with the customer, uh, using applied machine learning and, and AI, uh, clap the Cloudera data platform. >>Um, you know, business use cases require enabling, uh, the end-to-end journey, which we referred to as the data life cycle, uh, what the data life cycle, what is the data life cycle that our customers want to take their data through to enable the end-to-end data journey. If you ask our customers, they want different types of analytics, uh, for their diverse user bases to, to help them implement their, their, their use cases while managed by a centralized security and governance later layer. Uh, in other words, um, the data life cycle to them provides multifunction analytics, uh, at each stage within the data journey, uh, that, uh, integrated and centralized, uh, security, uh, and governance, for example, uh, enterprise data consists of real-time and transactional type type data. Examples include, uh, clickstream data, web logs, um, machine generated, data chatbots, um, call center interactions, uh, transactions, uh, within legacy applications, market data, et cetera. >>We need to manage, uh, that data life cycle, uh, to provide real enterprise data insights, uh, for use cases around enhance them personalized customer experience, um, customer journey analytics, next best action, uh, sentiment and churn analytics market, uh, campaign optimization, uh, mortgage, uh, processing optimization and so on. Um, we bring a diverse set of data then, um, and then enrich it with other data about our customers and products, uh, provide reports and dashboards such as customer 360 and use predictions from machine learning models to provide, uh, business decisions and, and offers of, uh, different products and services to customers and maintain customer satisfaction, um, by using, um, sentiment and turn analytics. These examples show that, um, the whole data life cycle is involved, um, and, uh, is in continuous fashion in order to meet these types of use cases, uh, using a single cohesive platform that can be, uh, that can be served by CDP, uh, the data, the Cloudera data platform. >>Okay. Let's, uh, let's talk about, uh, some of the experiences, uh, from our customers. Uh, first we'll talk about Bunco, something there. Um, Banco Santander is a major global bank headquartered in Spain, uh, with, uh, major operations and subsidiaries all over Europe and north and, and south America. Uh, one of its subsidiary, something there UK wanted to revolutionize the customer experience with the use of real-time data and, uh, in app analytics, uh, for mobile users, however, like many financial institutions send them there had a, he had a, had a large number of legacy data warehouses spread across many business use, and it's within consistent data and different ways of calculating the same metrics, uh, leading to different results. As a result, the company couldn't get the comprehensive customer insights it needed. And, uh, and business staff often worked on multiple versions of the truth. Sometimes there worked with Cloudera to improve a single data platform that could support all its workloads, including self-service analytics, uh, operational analytics and data science processes in processing 10 million transactions, daily or 30,000 transactions per second at peak times. >>And, uh, bringing together really, uh, nearly two to two petabytes of data. The platform provides unprecedented, uh, customer insight and business value across the organization, uh, over 80 cents. And Dera has realized impressive, uh, benefits spanning, uh, new revenues, cost savings and risk reductions, including creating analytics for, for corporate customers with near real-time shopping behavior, um, and, and helping identify 7,000 new corporate, uh, customer prospects, uh, reducing capital expenditures by, uh, 3.2 million annually and decreasing operating expenses by, uh, 650,000, um, enabling marketing to realize, uh, 2.4 million in annual savings on, on cash back on commercial transactions, um, and protecting 3.7 million customers from financial crime impacts through 95, new proactive control alerts, improving risk and capital calculations to reduce the amount of money. It must set aside, uh, as part of a, as part of risk mandates. Uh, for example, in one instance, the risk team was able to release a $5.2 million that it had withheld for non-performing credit card loans by properly identifying healthy accounts miscategorized as high risk next, uh, let's uh, talk about, uh, Rabo bank. >>Um, Rabobank is one of the largest banks in the Netherlands, uh, with approximately 8.3 million customers. Uh, it was founded by farmers in the late 19th century and specializes in agricultural financing and sustainability oriented banking, uh, in order to help its customers become more self-sufficient and, uh, improve their financial situations such as debt settlement, uh, rebel bank needed to access, uh, to a varied mix of high quality, accurate, and timely customer data, the talent, uh, to provide this insight, however, was the ability to execute sophisticated and timely data analytics at scale Rabobank was also faced with the challenge of, uh, shortening time to market. Uh, it needed easier access to customer data sets to ensure that they were using and receiving the right financial support at the right time with, with, uh, data quality and speed of processing. Um, highlighted as two vital areas of improvement. Robert bank was looking to incorporate, um, or create new data in an environment that would not only allow the organization to create a centralized repository of high quality data, but also allow them to stream and, uh, conduct data analytics on the fly, uh, to create actionable insights and deliver a strong customer service experience. >>Rabobank >>Leverage Cloudera due to its ability to cope with heavy pressures on data processing and its capability of ingesting large quantities of real-time streaming data. They were able to quickly create a new data lake that allowed for faster queries of both historical and real-time data to analyze customer loan repayment patterns, uh, to up to the minute transaction records, um, Robert bank and, and its customers could now immediately access, uh, the valuable data needed to help them understand, um, the status of their financial situation, this enabled, uh, rebel bank to spot financial disasters before they happened, enabling them to gain deep and timely insights into which customers were at risk of defaulting on loans. Um, having established the foundation of a modern data architecture Rabobank is now able to run sophisticated machine learning algorithms and, uh, financial models, uh, to help customers manage, um, financial, uh, obligations, um, including, uh, loan repayments, and are able to generate accurate, uh, current liquidity overviews, uh, no next, uh, let's, uh, speak about, um, uh, OVO. >>Uh, so OVO is the leading digital payment rewards and financial services platform in Indonesia, and is present in 115 million devices across the company across the country. Excuse me. Um, as the volume of, of products, uh, within Obos ecosystem increases, the ability to ensure marketing effectiveness is critical to avoid unnecessary waste of time and resources, unlike competitors, uh, banks, w which use traditional mass marketing, uh, to reach customers over, oh, decided to embark on a, on a bold new approach to connect with customers via a ultra personalized marketing, uh, using the stack, the team at OVO were able to implement a change point detection algorithm, uh, to discover customer life stage changes. This allowed OVO, uh, to, uh, build a segmentation model of one, uh, the contextual offer engine Bill's recommendation algorithms on top of the product, uh, including collaborative and context-based filters, uh, to detect changes in consumer consumption >>Patterns. >>As a result, OVO has achieved a 15% increase in revenue, thanks to this, to this project, um, significant time savings through automation and eliminating the chance of human error and have reduced engineers workloads by, by 30%. Uh, next let's talk about, uh, bank Bri, uh, bank Bri is one of the largest and oldest, uh, banks in Indonesia, um, engaging in, in general banking services, uh, for its customers. Uh, they are headquartered in, in Jakarta Indonesia, uh, BR is a well-known, uh, for its, uh, focused on financing initiative initiatives and serves over 75 million customers through its more than 11,000 offices and rural outposts, >>Um, Bri >>Needed to gain better understanding of their customers and market, uh, to improve the efficiency of its operations, uh, reduce losses from non-performing loans and address the rising concern around data security from regulators and consumers, uh, through enhanced fraud detection. This would require the ability to analyze vast amounts of, uh, historical financial data and use those insights, uh, to enhance operations and, uh, deliver better service. Um, Bri used Cloudera's enterprise data platform to build an agile and reliable, uh, predictive augmented intelligence solution. Uh, Bri was now able to analyze 124 years worth of historical financial data and use those insights to enhance its operations and deliver better services. Um, they were able to, uh, enhance their credit scoring system, um, the solution analyzes customer transaction data, and predicts the probability of a customer defaulting on, on payments. Um, the following month, it also alerts Bri's loan officers, um, to at-risk customers, prompting them to take the necessary action to reduce the likelihood of a Vanette profit lost. Uh, this resulted in improved credits in, in improved, uh, credit scoring system, uh, that cut down the approval of micro financing loans, uh, from two weeks to two days to two minutes and, uh, enhanced, uh, fraud detector. >>All right. Uh, this example shows a tabular representation, uh, the evolution of a customer retention use case, um, the evolution of data and analytics, uh, journey that, uh, that for that use case, uh, from aware, uh, text flirtation, uh, to optimization, to being transformative, uh, with every level, uh, data sources increase. And, uh, for the most part, uh, are, are less, less standard, more dynamic and less structured, but always adding more value, more insights into the customer, uh, allowing us to continuously improve our analytics, increase the velocity of the data we ingest, uh, from, from batch, uh, to, uh, near real time, uh, to real-time streaming, uh, the volume of data we ingest continually increases and we progress, uh, the value of the data on our customers, uh, is continuously improving, allowing us to interact more proactively and more efficiently. And, and with that, um, I would, uh, you know, ask you to consider an assess if you are using all the, uh, the data available to understand, uh, and service your customers, and to learn more about, about this, um, you know, visit cloudera.com and schedule a meeting with Cloudera to learn more. And with that, thank you for your time. And thank you for listening.

>>Uh, hi, I'm Joe Rodriguez, managing director of financial services at Cloudera. Uh, welcome to the fight fraud with a data session, uh, at Cloudera, we believe that fighting fraud with, uh, uh, begins with data. Um, so financial services is Cloudera's largest industry vertical. We have approximately 425 global financial services customers, uh, which consists of 82 out of a hundred of the largest global banks of which we have 27 that are globally systemic banks, uh, four out of the five top, uh, stock exchanges, uh, eight out of the top 10 wealth management firms and all four of the top credit card networks. So as you can see most financial services institutions, uh, utilize Cloudera for data analytics and machine learning, uh, we also have over 20 central banks and a dozen or so financial regulators. So it's an incredible footprint which gives Cloudera lots of insight into the many innovations, uh, that our customers are coming up with. Uh, criminals can steal thousands of dollars before a fraudulent transaction is detected. So the cost of, uh, to purchase a, your account data is well worth the price to fraudsters. Uh, according to Experian credit and debit card account information sells on the dark web for a mere $5 with the CVV number and up to $110. If it comes with all the bank information, including your name, social security number, date of birth, uh, complete account numbers and, and other personal data. >>Um, our customers have several key data and analytics challenges when it comes to fighting financial crime. The volume of data that they need to deal with is, is huge and growing exponentially. Uh, all this data needs to be evaluated in real time. Uh, there is, uh, there are new sources of, of streaming data that need to be integrated with existing, uh, legacy data sources. This includes, um, biometrics data and enhanced, uh, authentication, uh, video surveillance call center data. And of course all that needs to be integrated with existing legacy data sources. Um, there is an analytics arms race between the banks and the criminals and the criminal networks never stop innovating. They also we'll have to deal with, uh, disjointed security and governance, security and governance policies are often set per data source, uh, or application requiring redundant work, work across workloads. And, and they have to deal with siloed environments, um, the specialized nature of platforms and people results in disparate data sources and data management processes, uh, this duplicates efforts and, uh, divides the, the business risk and crime teams, limiting collaboration opportunities between CDP enhances financial crime solutions, uh, to be holistic by eliminating data gaps between siloed solutions with, uh, an enterprise data approach, uh, advanced, uh, data analytics and machine learning, uh, by deploying an enterprise wide data platform, you reduce siloed divisions between business risk and crime teams and enable better collaboration through industrialized machine learning. >>Uh, you tighten up the loop between, uh, detection and new fraud patterns. Cloudera provides the data platform on which a best of breed applications can run and leverage integrated machine learning cloud Derrick stands rather than replaces your existing fraud modeling applications. So Oracle SAS Actimize to, to name a few, uh, integrate with an enterprise data hub to scale the data increased speed and flexibility and improve efficacy of your entire fraud system. It also centralizes the fraud workload on data that can be used for other use cases in applications like enhanced KYC and a customer 360 4 example. >>I just, I wanted to highlight a couple of our partners in financial crime prevention, uh, semi dine, and Quintex, uh, uh, so send me nine provides fraud simulation using agent-based modeling, uh, machine learning techniques, uh, to generate synthetic transaction data. This data simulates potential fraud scenarios in a cost-effective, uh, GDPR compliant, virtual environment, significantly improved financial crime detection systems, semi dine identifies future fraud topologies, uh, from millions of, of simulations that can be used to dynamically train, uh, new machine learning algorithms for enhanced fraud identification and context, um, uh, connects the dots within your data, using dynamic entity resolution, and advanced network analytics to create context around your customers. Um, this enables you to see the bigger picture and automatically assesses potential criminal beads behavior. >>Now let's go some of our, uh, customers, uh, and how they're using cloud caldera. Uh, first we'll talk about, uh, United overseas bank, or you will be, um, you'll be, is a leading full service bank in, uh, in Asia. It, uh, with, uh, a network of more than 500 offices in, in 19 countries and territories in Asia, Pacific, Western Europe and north America UA, um, UOB built a modern data platform on Cloudera that gives it the flexibility and speed to develop new AI and machine learning solutions and to create a data-driven enterprise. Um, you'll be set up, uh, set up it's big data analytics center in 2017. Uh, it was Singapore's first centralized big data unit, uh, within a bank to deepen the bank's data analytic capabilities and to use data insights to enhance, uh, the banks, uh, uh, performance essential to this work was implementing a platform that could cost efficiently, bring together data from dozens of separate systems and incorporate a range of unstructured data, including, uh, voice and text, um, using Cloudera CDP and machine learning. >>UOB gained a richer understanding of its customer preferences, uh, to help make their, their banking experience simpler, safer, and more reliable. Working with Cloudera UOB has a big data platform that gives business staff and data scientists faster access to relevant and quality data for, for self-service analytics, machine learning and, uh, emerging artificial intelligence solutions. Um, with new self-service analytics and machine learning driven insights, you'll be, uh, has realized improvements in, in digital banking, asset management, compliance, AML, and more, uh, advanced AML detection capabilities, help analysts detect suspicious transactions either based on hidden relationships of shell companies and, uh, high risk individuals, uh, with, uh, Cloudera and machine learning, uh, technologies. You you'll be, uh, was able to enhance AML detection and reduce the time to identify new links from months 2, 3, 3 weeks. >>Excellent mass let's speak about MasterCard. So MasterCard's principle businesses to process payments between banks and merchants and the credit issuing banks and credit unions of the purchasers who use the MasterCard brand debit and credit cards to make purchases MasterCard chose Cloudera enterprise for fraud detection, and to optimize their DW infrastructure, delivering deepens insights and best practices in big data security and compliance. Uh, next let's speak about, uh, bank Rakka yet, uh, in Indonesia or Bri. Um, it, VRI is one of the largest and oldest banks in Indonesia and engages in the provision of general banking services. Uh, it's headquartered in Jakarta Indonesia. Uh, Bri is well known for its focus on financing initiatives and serves over 75 million customers through it's more than 11,000 offices and rural service outposts. Uh, Bri required better insight to understand customer activity and identify fraudulent transactions. Uh, the bank needed a solid foundation that allowed it to leverage the power of advanced analytics, artificial intelligence, and machine learning to gain better understanding of customers and the market. >>Uh, Bri used, uh, Cloudera enterprise data platform to build an agile and reliable, predictive augmented intelligence solution, uh, to enhance its credit scoring system and to address the rising concern around data security from regulators, uh, and customers, uh, Bri developed a real-time fraud detection service, uh, powered by Cloudera and Kafka. Uh, Bri's data scientists developed a machine learning model for fraud detection by creating a behavioral scoring model based on customer savings, uh, loan transactions, deposits, payroll and other financial, um, uh, real-time time data. Uh, this led to improvements in its fraud detection and credit scoring capabilities, as well as the development of a, of a new digital microfinancing product, uh, with the enablement of real-time fraud detection, VRI was able to reduce the rate of fraud by 40%. Uh, it improved, uh, relationship manager productivity by two and a half fold. Uh, it improved the credit score scoring system to cut down on micro-financing loan processing times from two weeks to two days to now two minutes. So fraud prevention is a good area to start with a data focus. If you haven't already, it offers a quick return on investment, uh, and it's a focused area. That's not too entrenched across the company, uh, to learn more about fraud prevention, uh, go to kroger.com and to schedule, and you should schedule a meeting with Cloudera, uh, to learn even more. Uh, and with that, thank you for listening and thank you for your time. >>Welcome to the customer. Obsession begins with data session. Uh, thank you for, for attending. Um, at Cloudera, we believe that a custom session begins with, uh, with, with data, um, and, uh, you know, financial services is Cloudera is largest industry vertical. We have approximately 425 global financial services customers, uh, which consists of 82 out of a hundred of the largest global banks of which we have 27 that are globally systemic banks, uh, four out of the five top stock exchanges, eight out of the 10 top wealth management firms and all four of the top credit card networks. Uh, so as you can see most financial services institutions utilize Cloudera for data analytics and machine learning. Uh, we also have over 20 central banks and it doesn't or so financial regulators. So it's an incredible footprint, which glimpse Cloudera, lots of insight into the many innovations that our customers are coming up with. >>Customers have grown more independent and demanding. Uh, they want the ability to perform many functions on their own and, uh, be able to do it. Uh, he do them on their mobile devices, uh, in a recent Accenture study, more than 50% of customers, uh, are focused on, uh, improving their customer experience through more personalized offers and advice. The study found that 75% of people are actually willing to share their data for better personalized offers and more efficient and intuitive services to get it better, better understanding of your customers, use all the data available to develop a complete view of your customer and, uh, and better serve them. Uh, this also breaks down, uh, costly silos, uh, shares data in, in accordance with privacy laws and assists with regulatory advice. It's so different organizations are going to be at different points in their data analytics and AI journey. >>Uh, there are several degrees of streaming and batch data, both structured and unstructured. Uh, you need a platform that can handle both, uh, with common, with a common governance layer, um, near real time. And, uh, real-time sources help make data more relevant. So if you look at this graphic, looking at it from left to right, uh, normal streaming and batch data comes from core banking and, uh, and lending operations data in pretty much a structured format as financial institutions start to evolve. Uh, they start to ingest near real-time streaming data that comes not only from customers, but also from, from newsfeeds for example, and they start to capture more behavioral data that they can use to evolve their models, uh, and customer experience. Uh, ultimately they start to ingest more real time streaming data, not only, um, standard, uh, sources like market and transaction data, but also alternative sources such as social media and connected sources, such as wearable devices, uh, giving them more, more data, better data, uh, to extract intelligence and drive personalized actions based on data in real time at the right time, um, and use machine learning and AI, uh, to drive anomaly detection and protect and predict, uh, present potential outcomes. >>So this is another way to look at it. Um, this slide shows the progression of the big data journey as it relates to a customer experience example, um, the dark blue represents, um, visibility or understanding your customer. So we have a data warehouse and are starting to develop some analytics, uh, to know your customer and start to provide a better customer 360 experience. Uh, the medium blue area, uh, is a customer centric or where we learn, uh, the customer's behavior. Uh, at this point we're improving our analytics, uh, gathering more customer centric information to perform, uh, some more exploratory, uh, data sciences. And we can start to do things like cross sell or upsell based on the customer's behavior, which should improve, uh, customer retention. The light blue area is, uh, is proactive customer inter interactions, or where we now have the ability, uh, to predict customers needs and wants and improve our interaction with the customer, uh, using applied machine learning and, and AI, uh, the Cloudera data platform, um, you know, business use cases require enabling, uh, the end-to-end journey, which we referred to as the data life cycle, uh, what the data life cycle, what is the data life cycle that our customers want, uh, to take their data through, to enable the end to end data journey. >>If you ask our customers, they want different types of analytics, uh, for their diverse user bases to help them implement their, their, their use cases while managed by a centralized security and governance later layer. Uh, in other words, um, the data life cycle to them provides multifunction analytics, uh, at each stage, uh, within the data journey, uh, that, uh, integrated and centralized, uh, security, uh, and governance, for example, uh, enterprise data consists of real time and transactional type type data. Examples include, uh, click stream data, web logs, um, machine generated, data chat bots, um, call center interactions, uh, transactions, uh, within legacy applications, market data, et cetera. We need to manage, uh, that data life cycle, uh, to provide real enterprise data insights, uh, for use cases around enhanced them, personalized customer experience, um, customer journey analytics next best action, uh, sentiment and churn analytics market, uh, campaign optimization, uh, mortgage, uh, processing optimization and so on. >>Um, we bring a diverse set of data then, um, and then enrich it with other data about our customers and products, uh, provide reports and dashboards such as customer 360 and use predictions from machine models to provide, uh, business decisions and, and offers of, uh, different products and services to customers and maintain customer satisfaction, um, by using, um, sentiment and churn analytics. These examples show that, um, the whole data life cycle is involved, um, and, uh, is in continuous fashion in order to meet these types of use cases, uh, using a single cohesive platform that can be, uh, that can be served by CDP, uh, the data, the Cloudera data platform. >>Okay. Uh, let's talk about, uh, some of the experiences, uh, from our customers. Uh, first we'll talk about Bunco suntan there. Um, is a major global bank headquartered in Spain, uh, with, uh, major operations and subsidiaries all over Europe and north and, and south America. Uh, one of its subsidiaries, something there UK wanted to revolutionize the customer experience with the use of real time data and, uh, in app analytics, uh, for mobile users, however, like many financial institutions send them there had a, he had a, had a large number of legacy data warehouses spread across many business use, and it's within consistent data and different ways of calculating the same metrics, uh, leading to different results. As a result, the company couldn't get the comprehensive customer insights it needed. And, uh, and business staff often worked on multiple versions of the truth. Sometime there worked with Cloudera to improve a single data platform that could support all its workloads, including self-service analytics, uh, operational analytics and data science processes, processing processing, 10 million transactions daily or 30,000 transactions per second at peak times. >>And, uh, bringing together really, uh, nearly two to two petabytes of data. The platform provides unprecedented, uh, customer insight and business value across the organization, uh, over 80 cents. And there has realized impressive, uh, benefits spanning, uh, new revenues, cost savings and risk reductions, including creating analytics for, for corporate customers with near real-time shopping behavior, um, and, and helping identify 7,000 new corporate, uh, customer prospects, uh, reducing capital expenditures by, uh, 3.2 million annually and decreasing operating expenses by, uh, 650,000, um, enabling marketing to realize, uh, 2.4 million in annual savings on, on cash, on commercial transactions, um, and protecting 3.7 million customers from financial crime impacts through 95, new proactive control alerts, improving risk and capital calculations to reduce the amount of money. It must set aside, uh, as part of a, as part of risk mandates. Uh, for example, in one instance, the risk team was able to release a $5.2 million that it had withheld for non-performing credit card loans by properly identifying healthy accounts miscategorized as high risk next, uh, let's uh, talk about, uh, Rabobank. >>Um, Rabobank is one of the largest banks in the Netherlands, uh, with approximately 8.3 million customers. Uh, it was founded by farmers in the late 19th century and specializes in agricultural financing and sustainability oriented banking, uh, in order to help its customers become more self-sufficient and, uh, improve their financial situations such as debt settlement, uh, rebel bank needed to access, uh, to a varied mix of high quality, accurate, and timely customer data, the talent, uh, to provide this insight, however, was the ability to execute sophisticated and timely data analytics at scale Rabobank was also faced with the challenge of, uh, shortening time to market. Uh, it needed easier access to customer data sets to ensure that they were using and receiving the right financial support at the right time with, with, uh, data quality and speed of processing. Um, highlighted as two vital areas of improvement, Rabobank was looking to incorporate, um, or create new data in an environment that would not only allow the organization to create a centralized repository of high quality data, but also allow them to stream and, uh, conduct data analytics on the fly, uh, to create actionable insights and deliver a strong customer experience bank level Cloudera due to its ability to cope with heavy pressures on data processing and its capability of ingesting large quantities of real time streaming data. >>They were able to quickly create a new data lake that allowed for faster queries of both historical and real time data to analyze customer loan repayment patterns, uh, to up to the minute transaction records, um, Robert bank and, and its customers could now immediately access, uh, the valuable data needed to help them understand, um, the status of their financial situation in this enabled, uh, rebel bank to spot financial disasters before they happened, enabling them to gain deep and timely insights into which customers were at risk of defaulting on loans. Um, having established the foundation of a modern data architecture Rabobank is now able to run sophisticated machine learning algorithms and, uh, financial models, uh, to help customers manage, um, financial, uh, obligations, um, including, uh, long repayments and are able to generate accurate, uh, current real liquidity. I refuse, uh, next, uh, let's uh, speak about, um, uh, OVO. >>Uh, so OVO is the leading digital payment rewards and financial services platform in Indonesia, and is present in 115 million devices across the company across the country. Excuse me. Um, as the volume of, of products within Obos ecosystem increases, the ability to ensure marketing effectiveness is critical to avoid unnecessary waste of time and resources, unlike competitors, uh, banks, w which use traditional mass marketing, uh, to reach customers over, oh, decided to embark on a, on a bold new approach to connect with customers via, uh, ultra personalized marketing, uh, using the Cloudera stack. The team at OVO were able to implement a change point detection algorithm, uh, to discover customer life stage changes. This allowed OVO, uh, to, uh, build a segmentation model of one, uh, the contextual offer engine Bill's recommendation algorithms on top of the product, uh, including collaborative and context-based filters, uh, to detect changes in consumer consumption patterns. >>As a result, OVO has achieved a 15% increase in revenue, thanks to this, to this project, um, significant time savings through automation and eliminating the chance of human error and have reduced engineers workloads by, by 30%. Uh, next let's talk about, uh, bank Bri, uh, bank Bri is one of the largest and oldest, uh, banks in Indonesia, um, engaging in, in general banking services, uh, for its customers. Uh, they are headquartered in, in Jakarta Indonesia, uh, PR is a well-known, uh, for its, uh, focused on micro-financing initiative initiatives and serves over 75 million customers through more than 11,000 offices and rural outposts, um, Bri needed to gain better understanding of their customers and market, uh, to improve the efficiency of its operations, uh, reduce losses from non-performing loans and address the rising concern around data security from regulators and consumers, uh, through enhanced fraud detection. This would require the ability to analyze the vast amounts of, uh, historical financial data and use those insights, uh, to enhance operations and, uh, deliver better service. >>Um, Bri used Cloudera's enterprise data platform to build an agile and reliable, uh, predictive augmented intelligence solution. Uh, Bri was now able to analyze 124 years worth of historical financial data and use those insights to enhance its operations and deliver better services. Um, they were able to, uh, enhance their credit scoring system, um, the solution analyzes customer transaction data, and predicts the probability of a customer defaulting on, on payments. Um, the following month, it also alerts Bri's loan officers, um, to at-risk customers, prompting them to take the necessary action to reduce the likelihood of the net profit lost, uh, this resulted in improved credit, improved credit scoring system, uh, that cut down the approval of micro financing loans, uh, from two weeks to two days to, to two minutes and, uh, enhanced fraud detection. >>All right. Uh, this example shows a tabular representation, uh, the evolution of a customer retention use case, um, the evolution of data and analytics, uh, journey that, uh, that for that use case, uh, from aware, uh, text flirtation, uh, to optimization, to being transformative, uh, with every level, uh, data sources increase. And, uh, for the most part, uh, are, are less, less standard, more dynamic and less structured, but always adding more value, more insights into the customer, uh, allowing us to continuously improve our analytics, increase the velocity of the data we ingest, uh, from, from batch, uh, to, uh, near real time, uh, to real-time streaming, uh, the volume of data we ingest continually increases and we progress, uh, the value of the data on our customers, uh, is continuously improving, allowing us to interact more proactively and more efficiently. And, and with that, um, I would, uh, you know, ask you to consider and assess if you are using all the, uh, the data available to understand, uh, and service your customers, and to learn more about, about this, um, you know, visit cloudera.com and schedule a meeting with Cloudera to learn more. And with that, thank you for your time. And thank you for listening.

>> Announcer: Live from Las Vegas, it's the Cube. Covering Informatica World, 2018. Brought to you by Informatica. >> Okay welcome back everyone, this is the Cube live here in Las Vegas for Informatica World 2018 exclusive coverage of the Cube. I'm John Furrier co-host of the Cube with Jim Kobielus, my co-host this segment and with that we'll keep on continue with the Cube. Our next guest is Vira Shanti who is the chief data officer at Lippo Digital Group, welcome to the Cube. >> Thank you so much, very excited to be here. >> Thank you for coming on, but people don't know before we came on camera, you and Jim were talking in the native tongue. Thanks for coming on. I know your chief data officer, we've got a lot of questions we love these conversations because we love data, but take a minute to explain what you guys are doing, what the company is, what the size is and the data challenges. >> Okay, maybe let me introduce myself first, so my name is Vira, my role is the chief data officer. Responsibility, that actually is cover for the big data transformation for the Lippo group data. Lippo group is actually part of the one of the largest in Indonesia, we serve a middle class for the consumer services, so we are connecting I think more than 120 million of the customers. What's Lippo as a group doing is actually we do many things. We are the largest of the hospital in Indonesia or just super market, we do department stores, coffee shop, cinema, data centers. We on bang as well, news, cable TV, what else? >> You have a lot of digital assets. >> What you do is you drive to any state in Indonesia and you see Lippo everywhere. >> Yeah, education as well, from the kindergarten to the university, that's why it's a lot of diversity of the business, that owned by Lippo. But recently we're endorsing a lot in the digital transformation, so we're releasing a new mobile app, it is called OVO, O, V, O. Actually it's like centralized loyalty E money to providing the priority bills to all the Lippo group customers, so they're not going to maintain their own membership loyalty program, it's going to just like the OVO, so it's not only being accepted by Lippo ecosystem, but also to the external ecosystem as well. We start to engage with the machine partner, we just today sorted like reaching out 30000 machine outlets. >> Let's get Jim's perspective, I want you to connect the dots for me, because the size and scope of data, you talk about deep learning a lot. And let's connect the dots, cuz we've heard a lot of customers here talking about being having data all over the place. How does deep learning, why do you catalog everything? If you've always diverse assets, I'm sure there are different silos. Is there a connection, how are you handling? >> Okay, differently it's not easy job to do, implementing big data for this kind of a lot of diversity of the business, because how to bring all of this data coming from the different source, coming from the different ecosystem to the single analytical platform is quite challenging. The thing is, we also need to learn first about the business, what kind of the business, how they operate, how they run the hospital, how they run the supermarket, how they run the cinema, how they run the coffee shop. By understanding this thing, my team is responsible to transform, not start from the calling the data, cleansing the data, transform the data, then generate the insight. It has to be an action inside. Then we also not only doing the BI things, but also how from their data we can developing the analytical product on top of the technology big data, that we own today. What we deliver is actually beyond the BI. Of course we do a lot of thing, for example, we really focusing in doing the customers 360 degree profile, because that's the only reason how we really can understand out customers. Today, we have more than 100s of customer attribute teaching for individual customers. I can understand what's your profile for the purchasing behaviors, what kind of the product, that you like. Let's say for the data coming from the supermarket, I know what's your brands, your favorite, whether you're spending is declining. How you spend your point, part of the loyalty program. Then many things, so by understanding very deep these, that we can engage with customers in the better way in providing the new customer experience, because we not only let's say providing them with the right deals, but also when would be the right time, we should connect to them providing something, that they might need. This is the way how from the data we try to connect with our customers. >> Yeah, provided more organic experience across the entire portfolio of Lippo brands throughout the ecosystem. It doesn't feel to the customer and so it isn't simply a federation of brands, it's one unified brand in some degree from the customer's point of view delivering value, that each of the individual components of the Lippo portfolio may not be able to provide. >> Yes, yes, so many things actually we can do on top of that 360 degree of the customers. Our big data outcome in the form of the API. Why it has to be in the API, because when we interact with the customer, there could be unlimited customer touch point to call this API. It could be like the mobile apps after smart customer touch point or could be the dashboard, that we develop for our Lippo internal business. Could be anything or even we can also connect to the other industry from the different business, then how we can connect each other using that big data API, so that's why-- >> Is it an ecosystem, isn't that one API, or it's one API, when unified API for accessing all the back end data and services? >> For something like this, there are to type of the API, that we develop, number one is the API, that belong to the customer 360 degree. Every entry would then attach to your profile and say we can convert it to the API. Let's say smart apps, as part of customer touch point, for example like OVO, we would like to engage with our customers, meaning, that the apps can just designing their online business orchestration, then calling a specific API by understanding let's say from the point of view of loyalty or product preference, that you like, so that then what kind of offers, that we need to push to the customer touch point general using the OVO apps. Or even let's say other supermarket have their on apps, so the apps can also following our API based on their data to understand what kind of the brand or the preference probably they like. Let's run in their apps, when the customer connects, it's going to be something, that really personalized. That's why it's in order to manage the future, actually it's very important for us to deliver this big data outcome in the form of the API. >> It scales too, not a lot of custom work, you don't have to worry about connecting people and making sure it works, expose an API and say, there it is and then. >> Different countries, in terms of privacy in the use of personally identifiable information, different countries and regions have their own different policies and regulations, clearly the European union is fairly strict, the European union with GDPR coming along, the US has its own privacy mandates, in Indonesia, are there equivalent privacy regulations or laws, that we require for example. You ask the customers to consent to particular uses of their data, that you're managing with your big data system, that sits behind OVO. Is that something in your overall program, that you reflect? >> Yes, there are some regulation in Indonesia governed by the government, they'll call having their own regulation, but we let's say part of the thing, that, yes, there is a specific regulation. But regulation for the retail is not really that clear yet for now, but we put ourself in the higher restricted regulation, that we put in place as part of our data protection, part of our data governance compliance as well. If until we do this demonetization or consolidating this data, there is no data, that's being shared outside the entity of the organization. Because let's say, when we do that demonetization everything's done by system to system, when it's called the API, so there is no hands off for other customer in individual data. Let's say if our partner FMCG digital agency or even advertiser, future wise they would like to call our API, what they can see, but that target lead of the customers, that they would like to connect is actually not individual of the data. It's going to be in the aggregated format. Even though many segmentation, that we can deliver is not going to expose every individual customer. >> You have a lot of use cases, that you can handle, because of the control governance piece. How about, by the way, that's fantastic and I know how hard it must be the challenge, but you have it setup nicely. Now that the setup with Informatica and the work you're doing, how are you interfacing with developers, cuz now you have the API. Is it just API based, are you looking at containers, kubernetes, clout technologies? Are you guys looking at that down the road or is that part of the, or is it just expose the API to the developers? >> For today, that actually who's going to consume our API actually? Definitely it's going to be the ecosystem of the Lippo internals, how the customer touch point can leverage the API. Then for the external, for example, like FMCG, the digital agency, when they call our API, usually it's like they can subscribe, there could be some kind of the business model divine there, but once again, like I mentioned to you, let's say it's not going to reveal any individual customer information, but the thing is, how we deliver this API things? We develop our own API system, we develop our API gateway, in simple thing, that actually how to put the permission or grant the access of any kind of digital channel, when they consumer our API and what kind of subscription meta? What we did for the big data actually is not really into, we investing a lot of technology in place for us to use. The thing, that makes my team so exciting about this transformation, because we like to create something, that's we create our own API gateway. We create some analytic product on top of the technology, that we have today. >> When they subscribe to the API, you're setting policy for the data, that they can get and you're done. >> Something like that. >> You automated that. Cool, well we see a lot of AI, any machine learning in your future, you, guys, doing any automation, how are you guys thinking about some of the tools we've been seeing here at the show around automation and AI, Clair, you tapping into any of the goodness? >> Yes, if everybody like to talk what AI right? >> John: You got API, you're good, you don't need anything. >> Many organization, when they're really implementing big data, sometimes they start jumping, I need to start doing the AI things. But from our point of view, yes, AI is very important, definitely we will go there, but for now, what's important for us is how we really can bring the data to single analytical platform, developing that 360 degree customer profile, because we really need to understand our customer better. Then thinking about how we can connect with them, how we can bring the new experience and especially at the right time. >> Actually let me break down AI, cuz I cover AI for Wiki bond, it's such an enormous topic, I break it down in specific things, like for example, speech recognition for voice activated access to digital assistance, that might be embedded in a mobile phones. Indonesia is a huge diverse country, it's an acapela, you have many groups living under the unitary national structure, but they speak different languages, they have different dialects, do you use or are you considering speech recognition? How you would tailor speech recognition in a country, that is so diverse as Indonesia. Is that something an application of AI you're considering using in terms of your user interface? >> Okay, for now we not really into there yet, because you are definitely correct. Developing that kind of library for Indonesia, because different dialect, different accent, it's tough, so the AI things, that we're looking for is actually going to be product recommendation engine. Because you know, let's say, that a lot of things on top of this customer 360 degree, that we can do, right? Because meaning it's going to open unlimited opportunity how I can engage to the customers, what kind of the right offer. Because there's a lot of brand owners, like FMCG, that they would like to connect, also getting in touch, reach out our customers. By developing this kind of product recommendation engine, let's say using the typical machine learning, so we can understand when we introduce this thing, customer like it, introduce that thing, they don't like it. >> Let me ask the next logical question there, it's such a big diverse country, do you, in modeling the customer profile, are you able to encode cultural sensitivities, once again, a very diverse country, there's probably things you could recommend in terms of products to some peoples, that other people might find offensive or insensitive, is that something, that in terms of modeling the customer, you take into consideration? It doesn't just apply to Indonesia, it applies here too or anywhere else, where you have many people. >> Of course can to do that the modeling, but we're doing right now, let's say once again, speaking about the personalized offer, from that point of view, what we see is to create the definition based on customer spending power first, buying power, we need to understand, that this customer's actually in which level of the buying power. By understanding this kind of buying power level, then we really can understand, that should we introduce this kind of the offers or not. Because this is too expensive or not. Because customer spending level can be also different. Let's say when our customers spend in our supermarket, maybe it's going to medium spending level, but let's say when they spend their money to purchase the coffee, maybe it's regular basis, so it's more spending. Could be different spending, so we also need to learn this kind of thing, because sometimes the low spending or medium spending or high spending, sometimes it's not something, that we put in the effort level for everything, sometimes it could be different. This is the thing, that also very exciting for us to understand this kind of spending, buying power. >> Great to have you on the Cube, thanks for coming, so I got to ask you one final question. I heard you were in an honorary Informatica innovation award honoree, congratulations. >> Thank you. >> What advice would you have for your peers, that might want to aspire to get the award next year? >> The thing is, our big data journey just start last year. Really start from the zero, so when yesterday we get an award for the analytics, so actually what we really focus on to do something, that actually is very simple. Some organization, when they're implementing big data sometimes they would like to do everything in the phase one. What we're planning to do is number one, how to bring the data very fast, then understand what kind of value of the data, that we can bring to the organization. Our favorite one is developing the customer 360 degree profile, because once you really understand your customer from any point of view, it's going to open unlimited opportunities how you can engage with your customers, it also open another opportunity how you can bring another ecosystem to our business to engage with our customers, that one point of view is already opening a lot of thing, huge. Either that thinking what would be the next step. Of course, that API is going to simplify your business in the future scale so on. That's becoming our main focus to allow us to deliver a lot of quick low hanging effort at the same time. I think that's a thing, that makes us really can, within a short period of time, can deliver a lot of things. >> The chief data officer at Lippo digital group, thanks for sharing your story, it's the Cube, we're here live in Las Vegas. They're going to be bonding here talking about all the greatness going on there. This is the Cube here in Las Vegas, stay with us for continuing day two coverage of Informatica world 2018, we'll be right back.