Big Data, Big Banks and Unleashing Big Opportunities

Transcription

1 Big, Big Banks and Unleashing Big Opportunities

2 Big, Big Banks and Unleashing Big Opportunities Big, Big Banks and Unleashing Big Opportunities A retailer using Big to the full could increase its operating margin by more than 60 percent. Additionally, with the use of Big, users of services enabled by personal-location data could capture $600 billion in consumer surplus. McKinsey Research McKinsey Research clearly highlights how enterprises could benefit if they embrace Big to know their consumers better. However, the reality is enterprises are not there yet. When consumers engage with channels such as Amazon or Yelp, they get a very intimate experience and they receive personalized product offers that are relevant to them. Based on these experiences, consumers have similar expectations from other consumer-centric companies such as banks. Without deep knowhow about their consumers, banks may not be able to meet consumer expectations and this can have a profound impact on their business, in terms of: Lost revenue opportunity due to poor targeting and irrelevant product offers, Inability to proactively detect frauds and manage credit risks, Limited consumer view that results in poor targeting and wasted marketing spend, churn due to false positives and irrelevant product offers, Lost competitive edge by allowing competition to identify business trends faster. The world is increasingly interconnected, instrumented and intelligent. And in this new world the velocity, volume, and How many times do we receive variety of data being created is unprecedented. creation coupons for shops that are 50 has exploded with the growth of social, mobile, cloud and multimedia computing. Eighty percent of the data being created miles away that we rarely visit? is either unstructured or semi-structured which cannot be We hear stories about massive easily stored and analyzed using traditional database systems. amounts of data enterprises are As the amount of data created about a consumer is growing, collecting and yet we receive the percentage of data that businesses can process is going totally irrelevant offers that we down fast. This is due to the fact that traditional systems are never sniff. inadequate to store, process and analyze massive amounts of structured and unstructured data. They are not designed for today s unstructured data, rapidly changing schema and elastic scaling of storage. Additionally, banks have a very complex IT architecture with their existing mission critical applications and it makes the paradigm shift to managing unstructured data monumental. 2 Copyright 2015 NGDATA

3 Big, Big Banks and Unleashing Big Opportunities Without robust Big management solution, banking and financial services companies are unable to gain business insight from massive internal and external data about customers and are unable to deliver personalized offers based on one-to-one marketing to increase loyalty and use of their products such as credit cards and loans. In this paper, we will examine in spite of having a massive amount of consumer data at their disposal -- why banks do not know their customers, and the business opportunities they are unable to capitalize on because of lack of consumer intelligence. We will also showcase, key use cases and business problems that are being solved by progressive banks using Big technologies. Additionally, we will look at how they can partner with retailers to drive usage of their credit cards, detect fraud quicker and increase consumer intimacy by delivering consumers what they want, when they want it. Without Big, Big Opportunities are Lost In a mature market, such as the US, Europe or Canada, where credit is a mature industry, it is naïve for a bank to believe that the way it is going to grow revenues is simply by issuing more credit cards. The issue for a bank is not to increase the amount of credit cards, but to ask, How do we get the user to use our card? Chief Analytics Officer at a Financial Company, Computing Magazine, June 2012 With the lack of consumer intelligence, banks are unable to gain business insight from massive internal and external data about their customers. The business impact? Lost revenue opportunities, low coupon redemption rates, lower share of customers wallet and lost competitive agility. With a 360 degree consumer view, banks could team up with retailers, deliver personalized product offers, increase coupon redemption rates and drive usage of their credit cards. Without a 360 degree view of their consumers, banks and retailers are unable to identify potential warning signs in terms of fraud or creditworthiness of consumers and expose themselves to greater business risk. Lack of complete view of a consumer can result in false positives in case of a legitimate transaction and create customer relationship nightmares. Even relatively straightforward tasks such as checking consumer names against a blacklist can become very complicated as banks may have hundreds of consumers with the name Jane Smith in their database. Continuously updating the consumer database in the midst of hundreds of interactions across many channels and providing an up-to-date single view of each consumer becomes a source of false positives. Identifying a very small number of fraudulent transactions in the ocean of millions of legitimate ones is a huge problem. A false positive in case of a legitimate transaction can often lead to customer deflection to a competitor. In a nutshell, not being able to gain insights from the goldmine of data means banks and retailers are allowing their competition to identify critical business trends and act on those before they can. Basically leaving business on the table. 3 Copyright 2015 NGDATA

4 Big, Big Banks and Unleashing Big Opportunities So Much and So Little Intelligence! Why? Banks have more data about their consumers than companies such as Google and Amazon, who have revolutionized consumer intelligence and one-to-one targeting. In addition to web clickstream data and data from social networks such as Facebook, banks possess very personalized data on credit card purchase histories and buying preferences. Enterprise applications such as CRMs or ERPs are other big sources of consumer data that banks own. Banks also can access a large amount of loyalty program data they receive from merchants. With the growth in mobile banking, banks also possess contextual data such as location of consumers. And this data is growing exponentially. Growing in volume, growing in velocity at which the data is created and growing in the variety of the data both structured and unstructured. When banks have so much data, why do they still not know their consumers well? Here are the key challenges banks are facing in getting a 360 degree view of the consumer: Hundreds of internal data sources: The number of data Banks have more data about sources is ever growing. After every M&A, banks add their consumers than companies new systems and many more data sources into their IT such as Google and Amazon who infrastructure. Each system and data store holds different information and a limited view of the consumer. The have revolutionized consumer data could be stored in many forms including relational intelligence and one-to-one databases, XML data, Warehouses and enterprise targeting. applications such as ERP and CRM. This creates hundreds of data silos, each reflecting a small slice of the consumer. This fragmented consumer view clearly does not help banks in knowing their consumers. Growth in external and unstructured data: Banks also have a large amount of external and unstructured data about their customers in the form of tweets, Facebook wall posts, searches, website visits, streams, videos and so forth. In fact, 80% of the data being created is either unstructured or semi-structured which cannot be easily stored and analyzed using traditional systems. At the same time, the percentage of data that businesses can process is going down fast as traditional systems are inadequate since they are not designed for today s unstructured data, rapidly changing schema and elastic scaling of storage. Storing, indexing and analyzing massive data: Dynamically scaling the storage capacity without any disruption to mission critical applications is a big challenge. Finding actionable insight among the massive structured and unstructured datasets, and delivering that with sub-millisecond latency is like finding a needle in a haystack. Being able to query data across multiple clusters of commodity servers and aggregate the results into meaningful insights is increasingly difficult with traditional technologies. Velocity of data creation: The speed of data creation across multiple channels is unprecedented. If banks continue to process data in a batch mode they can lose precious time that competition could gain to act faster and wipe away profits. It has become critical to not only process static data and consumer profiles, but their interactions with the data in real-time so banks can gain actionable insights to make moves before the competition. Fragmented view of consumer: Even if banks were to aggregate data from hundreds of internal and external sources and put it into a unified system such as Hadoop, the information will still be in multiple silos. Additionally, how do you match information about a consumer Jane Smith from multiple data sources? 4 Copyright 2015 NGDATA

5 Big, Big Banks and Unleashing Big Opportunities First of all, there could be multiple Jane Smiths. Secondly, it is difficult to correlate information from multiple data sources to a single individual. In a nutshell, simply integrating and aggregating data from multiple sources does not provide a single view of the consumer that is essential for more sophisticated personalized marketing and loyalty programs. Organizational readiness and skillsets: The volume, velocity and variety of the unstructured data makes it impossible for organizations to store, index, search and analyze massive amounts of data using traditional systems. In fact the traditional systems are inadequate for unstructured data, rapidly changing schema and elastic scaling of storage. On the other hand, most organizations including banks do not have organizational expertise and skillsets to deal with the complexities of document store type Big management systems such as Hadoop. The learning curve, complexity of data management and need to integrate different modules from the Hadoop stack makes it very difficult for banks to undertake meaningful Big projects, let alone implement them successfully to gain consumer intelligence. The Conceptual Solution About NGDATA NGDATA is the customer experience management solutions company that enables organizations to maximize the value of their customer relationships. Through its breakthrough solution, Lily Enterprise TM, companies can create individual and extensive customer profiles, in real time, resulting in highly effective targeting for more personalized customer experiences. NGDATA empowers enterprises to Listen bigger to all customer interactions utilizing Big technologies, Learn faster from behavior and contextual information, creating more effective actionable insights, and Execute smarter on these insights to better find, optimize and engage targets. The Solution NGDATA proposes to use Lily Enterprise to implement the strategy for customer centricity that is able to listen, learn and execute, using the available data sources and business workflows. With Lily, companies can better: Listen across many different digital channels, collecting every interaction related with behavioral, operational and socio-demographic observations. Learn based on individual customer behavior (such as offer responses) to generate an individual profile - customer DNA - and individual customer preferences. Lily learns is a key differentiating function and is the most effective adaptive learning engine. Execute upon customer activity based on simple instructions for how to find, optimize and engage targets. It adapts to real-time input to deliver highly relevant offers. 5 Copyright 2015 NGDATA

6 Big, Big Banks and Unleashing Big Opportunities statements IRA investments online credit cards debit balance teller savings withdraw transfers checking deposits Security mobile wallet mortgage wire transfer phone annuity interest ATM Loans Rates savings bonds direct deposit no fees CDs branch Figure 1: Lily Listen - Learn - Execute Approach. The Lily DNA The heart of the system is the Lily DNA variables and values that describe each and every user in a very detailed way. The Lily DNA contains 1000s of (out of the box) socio-demographic values, scores describing mobility, communication preferences, spending behavior, customer engagement and more. The DNA is built in an automated and real time manner based on source data and interactions ingested into Lily. The DNA contains values coming from source records, or scores, derived from the ingested data, calculated at ingest time, rather than batch-mode. Those scores are calculated for each individual customer, and kept and enhanced over time. Apart from predefined metrics, Lily also allows companies to define additional company specific metrics that are relevant for the business of the company. 6 Copyright 2015 NGDATA

7 Big, Big Banks and Unleashing Big Opportunities 35% 87% $11350 $ $ months 11 75% 18 80% 2d 2 36d 5 63% Figure 2: Lily DNA. All DNA variables are kept up to date, for each and every individual in the system. As the DNA evolves over time, the data is kept, enabling: Trending and prediction of future values; Alerting whenever a DNA variable (e.g. Churn score) trends at a certain rate, or reaches a certain value. The DNA information is available to review through the Lily Interface or using third party software using Lily s open API s. 7 Copyright 2015 NGDATA

8 Big, Big Banks and Unleashing Big Opportunities Lily Preference Learning While the Lily DNA are attributes or scores that belong to a customer, Lily also allows companies to predict the propensity a customer might have for a new or existing product, a service or a particular offering. This propensity is calculated automatically and based on a number of machine learning algorithms. It is also updated in real time, using all incoming interactions. Because of the rich Lily DNA, the Lily Preference Learning feature has more data to start from and can come up with better recommendations and analytics, simply because it knows more about individual customers. Figure 3: Lily Preference Learning. 8 Copyright 2015 NGDATA

9 Big, Big Banks and Unleashing Big Opportunities High Level Architecture The high level architecture is illustrated in Figure 4. Lily Enterprise will host all customer information. All the data is ingested into Lily and used for customer insight and profiling. Back Office Systems Transactions Reporting / Analytics Enterprise BI and Reporting Applications Social Company and Activity Operations ERP/CRM DWH External 3rd Party Reference External Systems Lily Enterprise Connector - ETL Tools Enterprise Analytics Applications Lily Enterprise Web and Mobile Website And Online Apps Mobile App Server Channel Campaigns Marketing Campaign Mgt Mail SMS Print Broadcast SCORES 3rd Party Master 3rd Party Operational Interaction base Single View DNA Targeting Service Desk CRM Systems Figure 4: High Level Architecture. Lily integrates with a number of other tools in the ecosystem: ETL tools to facilitate the ingest of data, third party reporting and modelling. Underlying API s are based on REST, Java, JDBC or Hive protocols. The Lily Enterprise solution has the following key features: Lily not only stores structured and unstructured information - out of the box - it also adds the following insights: Lily organizes the data in a customer centric model, based on industry specific data models. Although those models can be extended if desired by the company, the models do not have to be designed or built - they are already in the software. Based on the available data, Lily generates a Lily DNA profile - for each and every end customer in the system. Apart from social demographic information, that profile contains life stage events, affinities and interests (for sports, leisure, science, business ), lifestyle values, mobility information, communication preferences, predefined segments, customer status, etc., and a number of scores, including risk scores, etc. Lily also predicts the propensity for certain events - product purchases, churn, reacting to an advertisement, redeeming a coupon, etc. 9 Copyright 2015 NGDATA

10 Big, Big Banks and Unleashing Big Opportunities Lily is a real-time and interactive system - reference implementations include real time interactions with e.g. a mobile wallet, a website, etc. As scores are updated in real-time, Lily offers the latest information and does not require nightly batch jobs. As a result, Lily does not only help understand why something has happened (as more classic BI tools do), Lily can also drive your actions by suggesting what to do when customers interact with the company. Lily learns and builds and stores a detailed individual customer profile (DNA), which gets richer every time new information is ingested. Lily builds a 100% objective customer detail. As opposed to rule based systems where more general rules (often human assumptions) are applied. Those systems don t learn, and tend get complicated and hard to maintain after a while as more rules get introduced. The Technology Platform Lily is an application built using Big software components such as Hadoop (HDFS) and Hbase. Lily runs on top of Cloudera s distribution, and on commodity hardware or Cloudera based appliances. The storage of all the interactions, the raw data, the DNA and the preferences is done in a customer centric database, built on the above technologies and using an industry specific data model, including the DNA model and the Preference model. The central database consolidates data coming from different sources, ingested in streams (real-time) or in batch, as necessary, often via ETL tooling. As such, Lily combines all the siloed data into one central system using Big technology. This is illustrated in Figure 5. From Channel Silos towards Integrated Channels 1. Centric 2. Centralized 3. Real-Time Interactions 4. Open for 3 rd Party Systems CRM ERP OPS Various DWH Transactions Partner External Know your customer Centralize data into a single customer view Figure 5: From siloed data to a customer centric system. 10 Copyright 2015 NGDATA

11 Big, Big Banks and Unleashing Big Opportunities NGDATA in Action: Sample Use Cases The Big market is in its infancy but is growing exponentially. While banking executives agree that Big has potential to transform their businesses, they are often unsure which use cases they should consider when implementing a Big solution. While working with a number of Fortune 500 type banks, we have identified key use cases that are leading the wave and delivering significant business value: Mobile wallet for one-to-one marketing Fraud detection and prevention of false positives Risk management based on unique risk profiles Segmentation, targeting and risk based pricing In this section, we will drill down further into couple of these use cases. 1. Mobile Wallet and One-to-One Marketing Mobile commerce and payments is a fast growing trend. Many large banks have hundreds of thousands of customers who us their mobile wallet application for day to day banking and purchase transactions. Banks have a lot of data about consumers their purchase histories, shopping habits, shopping frequency and so forth. Merchant loyalty programs make additional consumer data available to banks - in addition to merchant profiles, locations and their products. Banks also can gather data from external sources such as website visit clickstreams, social interactions, and enterprise applications. And lastly, through the mobile wallet application, banks can get contextual data such as consumer s location. See Figure 6 below. Social direct mail Company and activity Channel Campaigns Web & Mobile SCORES Website and Online Apps Mobile App Server ATM web agent, IVR Mail SMS TIMES Broadcast Print Service Desk mobile CRM Systems chat Figure 6: The Personalization Experience 11 Copyright 2015 NGDATA

12 Big, Big Banks and Unleashing Big Opportunities In fact, banks have more data about consumers than Amazon and Google but they are not able to unleash the power from this goldmine of massive data. Lily can change this. Using Lily, a bank can get a continuous, 360 degree view of the consumer and know their consumer better. These insights can allow banks to team up with retailers and deliver personalized product offers and coupons based on purchase histories, buying habits, brand loyalties and consumer location. With one-to-one targeting and personalization banks can drive the coupon redemption revenues by over 100% and increase usage of their credit cards and loans. Banks have more data about consumers than Amazon and Google but they are not able to unleash the power from this goldmine of massive data. 2. Fraud Detection & Prevention of False Positives Consider the case of the consumer Jane Smith, who while at a retail banking conference in New York, her credit card was being swiped for a purchase in Los Angeles. Keep in mind, this is Jane s backup credit card and she does not use it very often. The velocity tracking algorithms would not be able to detect this potential fraud since the card is only being used in Los Angeles and the bank has very limited view of Jane. Additionally, that view is solely based on her credit card transactions. On the other hand, if the bank has more complete information about Jane and her public data from the social networks it will be able to detect from her tweets that she is having a great time at the New York conference while the card is being swiped in Los Angeles. Using Lily, a bank can first get the missing social data aggregated and matched with the right Jane Smith. So that as Jane is tweeting from the conference session, the thief is swiping the card in Los Angeles, and the bank s view of Jane Smith is being updated on a continuous basis. This continuous 360 degree view about Jane Smith can help banks detect and prevent subsequent fraud. Additionally, banks can also leverage the 360 degree view of a consumer to prevent false positives. For example, Jane Smith, during her next conference in Brussels, after securing large customer deals, has decided to indulge in some expensive diamond shopping. Unfortunately, she could not find time to notify her bank that she would be traveling abroad and her credit card transaction is denied. Using Lily this false positive and bitter customer experience could be avoided since the bank has complete data including her tweets from the conference and ability to match that data to her unique profile to let the bank know that Jane is indeed in Brussels and is carrying out a legitimate transaction. In summary, banks can get a more complete view of the consumer based on, not just the credit card or enterprise data, but also based on location, social and clickstream data. The data from multiple channels is matched and correlated to create unique consumer profiles that are being continuously updated based on new interactions. These always up-to-date profiles give continuous 360 degree view of the consumer to the bank. Using the Lily SDK (Software Development Kit), and high level API s, banks can get single view of consumer and update their fraud detection applications to be more effective. 12 Copyright 2015 NGDATA

13 Big, Big Banks and Unleashing Big Opportunities Conclusion Knowing their customers better is critical for banks to gain competitive agility. Banks can leverage Big to know their consumers better, deliver personalized offers, coupons and drive net promoter scores. Banks can also drive usage of their products such as credit cards or loans with a 360 degree view of the consumer. When the consumers receive offers they were just thinking about, the customer satisfaction improves dramatically. With Lily, banks can deliver targeted retention offers to consumers based on their preferences and propensity to switch. In a nutshell, banks can gain significant competitive edge as they can identify trends and changes in consumer behavior faster than their competition by embracing Big without massive changes to their organization, IT architecture, and skillset. To discover more about NGDATA or how Lily Enterprise can help you solve your customer experience management challenges, please go to or contact us at info@ngdata.com. 13 Copyright 2015 NGDATA

14 NGDATA, Inc. All rights reserved.