Innovation of Customer Experience Value - Data Strategy of Shinsei Bank Group

Transcription

1 Innovation of Customer Experience Value - Data Strategy of Shinsei Bank Group February, 2018 Hiroyuki Torigoe President & CEO Shinsei Financial Co., Ltd.

2 Creation of Customer Experience Value and Growth with Lake ALSA Lake ALSA will keep pursuing the maximization of customer experiences and their lifetime value (LTV). LTV (Lifetime Value )Utilization of Digital Technology Web+Automated Stores (ACM) Marketing 1to1Marketing Smart-phone transaction UX/UI ChatBot/API Mass Marketing (TV) ACM Web Advertisement Convenience Store ATMs Lake ALSA Rating/ Collection AI Score Digitalization of accumulated human expertise YUI Platform Introduction of credit scores Manual credit ratings Operation IVR RPAxOCRxAI API Operator Support Calling centers s/SMS TV CM+Branches with Clerks TV CM 写真を探して張り付ける ATM Manual Ratings Branches with clerks Essential Depth of Customer Experience 2

3 YUI Platform to Support Shinsei Bank Group s Data Strategy A YUI Platform is an analytics platform created by combining an Artificial Intelligence (AI) and the Group-wide Customer Database. YUI Platform is capable of expecting the needs and credit ratings of each customer through AI, allowing us to keep improving our marketing and credit risk assessment capabilities to a higher level. The Group aims to continue upgrading the Platform so that we can always fulfill the true needs of the customers, targeting also the customers outside of the Group in the future. Consumer Finance Customers Intricate Ratings Appropriate Product Recommendation New Lendings to New Customers Varieties of Transaction Methods Shinsei Bank APLUS Shinsei Financial Affiliates with Individual Customers Individual Rating Marketing Scores Data Inquiries YUI Platform AI/Analytics Platform Group Customer Database (Identification, Anonymization) Total: about 10 million customers Corporate Scores Fraudulent Scores Shinsei Bank Showa Leasing APLUS Affiliates with Corporate Customers Transactionbased Lending Appropriate Product Recommendation Varieties of Lending Method Corporate Customers AI keeps learning though continuous utilization and acquisition of the new data 3

4 Innovate Business with Digital Technology and Acquire the New Customers Shinsei Financial will gradually upgrade the credit rating / digital marketing method with AI learning ability, acquiring the young generation with high digital literacy while retaining current customers of the Lake brand. Shinsei Financial, simultaneously, will optimize its operational cost with digital technology, namely, using OCR and RPA. C (II) Sophistication in Rating/Collection Additional acquisition of potentially qualified customers (Upgraded AI Scores) Lake Brand Potential Market Essential deepening of customer experience with digitally savvy customers (UX/UI) + D Realize cost optimization by lowcost operation using digital technology B (I) Sophistication in Digital Marketing 4

5 Branding and Marketing Concept of Lake ALSA B Shinsei Financial aims to create a brand value which can differentiate itself from competitors, by taking over the sense of safety and convenience of a big brand, Lake, and with advanced technology and customer support. The Value of Lake Sense of safety Lake as a big consumer finance brand Convenience No ATM handling charge, and no interest for the first 180 days + The New Value to be Created Agility On-time service with high-end AI technology Linked Build a new customer base utilizing the integrated database (YUI Platform) Security Always behind the customers According to our vision* *Build a proud financial service provider for ordinary citizens, sincerely engaging with customers and offering values needed by society. Enforcement of Multi-Function to Create the Brand Value Card-less withdrawal/repayment service with Seven Bank ATMs. Better response with Web-API, optimization of web members site and forms (AB Test, PDCA cycles) ChatBot services with Smart phones/tablets 24h wire transferring services (from Fall, 2018) 5

6 Upgrading the Rating and Collection Procedures with AI Scorings C-1 Expecting better performance through the application of sophisticated hybrid of AI and logit scores to the various financial products of the Group. Expecting higher accuracy of prediction by combining the big data and the AI s deep learning technology. Expecting better credit scoring performance by utilizing AI analysis of the credit position in fund collection area. Hybrid Ratings Utilizing AI and Logit AI Scoring the Group is Working On High AI Scoring Higher Scoring Performance Approve/increase when AI finds blue-chip customers, regardless of low logit scores Data Integrate and utilize not only credit and attribute information, but also all structured, non-structured and open data within the Group for scoring Medium Decline/decrease when AI finds inappropriate customers, regardless of the high logit scores Algorithm Maximize the impact of the big data by hiring the high-end mechanical learning models such as deep learning and ensemble learning Low Lower Scoring Performance Traditional Logit Scoring Area of Application Utilize AI for the credit ratings as well as for the fund collection area, upon evaluation of the credit position(as they are two sides of the same coin) Low Medium High 6

7 Digitalization of Human Expertise in Fraudulent Control C-2 Share the expertise of Shinsei Personal Loan s skilled staffs within the Group by replacing manual credit screening of fraudulent detection with AI. Digitalization of manual credit screening techniques in collection centers are carried out simultaneously. Current Status Future Application Application Scoring Extract Applications Subject to Manual Scoring Scoring Extract Applications Subject to Manual Scoring Toward the higher accuracy/credit rating support through AI s learnings of the past manual credit ratings Manual Operations Extraction Tool Manual Scoring Operations by Experienced Staffs Decisions of Experienced Staffs Manual Scoring Operations of Experienced staffs Decisions of Experienced Staffs Information Collection/ Checking Tools AI Ratings Contract Decline Contract Decline 7

8 Phases of Robotization with Robotics Process Automation (RPA) D The Group is promoting its robotization by introducing RPA and OCR technologies mainly in back office operations in Shinsei Financial The Group almost completed their preparation processes to implement them in FY2018 We are expecting 50% reduction of workload in the selected 19 operations in FY2018, and will expand its application through modularization going forward. Phase 2-1 Design Each Operation/ Carry Out Product Selection PoC Phase 2-2 New Process Introduction Phase 3 Start Utilizing Consideration Period October - November 2017 December 2017 March 2018 April 2018 Details of Tasks Concept of New Process Share Views Set Evaluation Axis Design Each Operation (List of New Operation, New Operation Flow Chart) Specify Tasks for sampling RPA Product Comparison OCR Product Evaluation /Selection Reflect Characteristics RPA PoC Operation Check OCR PoC Introduction Plan Trial Calculation of Effectivene ss Get Estimates and Select Tools Project Decision Making Check-point Support Detailed Designing of Operations Starting-up Program of RPA Preparation for Utilizing Phase/Adjustment of OCR The Operation Starts Consideration of Management Policy/ Set Introduction Support Establishment of Rules for Robotics Management Tools 8

9 Disclaimer The preceding description of Shinsei s Medium-Term Management Plan contains forward-looking statements regarding the intent, belief and current expectations of our management with respect to our financial condition and future results of operations. These statements reflect our current views with respect to future events that are subject to risks, uncertainties and assumptions. Should one or more of these risks or uncertainties materialize, or should underlying assumptions prove incorrect, our actual results may vary materially from those we currently anticipate. Potential risks include those described in our annual securities report filed with the Kanto Local Finance Bureau, and you are cautioned not to place undue reliance on forward-looking statements. Unless otherwise noted, the financial data contained in these materials are presented under Japanese GAAP. The Company disclaims any obligation to update or to announce any revision to forward-looking statements to reflect future events or developments. Unless otherwise specified, all the financials are shown on a consolidated basis. Information concerning financial institutions other than the Company and its subsidiaries are based on publicly available information. These materials do not constitute an invitation or solicitation of an offer to subscribe for or purchase any securities and neither this document nor anything contained herein shall form the basis for any contract or commitment whatsoever. 9