Banking 4.0: The Industrialised Bank of Tomorrow

Foreword by Sashidhar Jagdishan
Foreword by Hans Tesselaar
Preface
Calibrated and Coordinated Delivery of the Digital Experience
Scaling and Timing of the Technology Design and Business Model Benchmarked to External Innovation
Developing and Embedding Banking AI
Industrialise Data
Capital Allocation as a Measure to Align the Entire Bank on Industrialisation Initiatives
Acknowledgements
Contents
1 Industrialise and Innovate to Deliver Banking 4.0 Services
1.1 Introduction: Evolution of Banking
1.1.1 Banking 1.0: System of Record on Mainframe-Enabled Business Model
1.1.2 Banking 2.0: System of Record on 4GL Technology-Enabled Business Model
1.1.3 Banking 3.0: Self-Service for Payments-Enabled Business Model
1.2 Major Shortcomings of Banking 3.0
1.2.1 Third-Generation Banks Invested in Self-service Rather Than Delivering Digital Experience
1.2.2 Third-generation Banks Are Under-invested in the Back Office and Mid Office
1.2.3 Third-generation Banks Are All-pervasive With Data and Application Duplication
1.2.4 Inability to Create a Business Case for Technology Investment
1.3 Banking 4.0: Services Delivered by Tomorrow’s Industrialised Bank
1.3.1 The Banking 4.0 Vision Statement
1.3.2 The Industrialised Bank of Tomorrow
1.3.3 Delivering Digital Experience to Customers, Partners, and Employees
1.4 The Six Engines of Industrialisation and Innovation
1.4.1 Technology Investments
1.4.2 Deliver Digital Experience
1.4.3 Embedded Innovation
1.4.4 Intelligent Process and Technology
1.4.5 Industrialised Data
1.4.6 Banking AI As a Service
1.5 Conclusion—Banking 4.0 Is an Iterative Process
2 Tactical Approach to the Industrialisation
2.1 Introduction—Manual Tasks in Banking Processes
2.1.1 The Genesis of Manual Processes
2.1.2 Legacy Applications Use GUI for Data Capture
2.1.3 15–20% of Employees at Banks Manage Data and Documents for Credit Management
2.1.4 15–20% of Employees at Banks Manage Financial Crimes Prevention Data
2.1.5 5–10% of Employees at Banks Manage Data for Financial Risk and Regulatory Reporting
2.2 Automation of Manual Tasks Versus Industrialisation of the Bank
2.2.1 Third-Generation Technology Automates High Volume, Low Complexity Manual Tasks
2.2.2 Robotic Process Automation (RPA) Automates Low Volume Low Complexity Manual Tasks
2.2.3 Machines at Industrialised Bank Manage High Volume High Complexity Processes
2.2.4 To Industrialise the Bank—Across the Valley of Investment, Innovation, and Alignment
2.2.5 Preliminary Approach to Legacy Application Management Through RPA
2.2.6 Tactical Approach to Legacy Application Modernisation
2.3 Robotic Process Automation (RPA) Automates Manual Tasks
2.3.1 RPA Tool Automates Granular Tasks
2.3.2 Case Study: Survey of Commercially Available RPA Tools for Capabilities to Automate a Banking Process
2.3.3 Case Study: RPA Delivers Suboptimal Benefits
2.3.4 Case Study: Automate the Process and Not the Tasks
2.3.5 Automation in Handling Unstructured Data Needs Cognitive Capabilities
2.3.6 Cognitive Process Automation Is Contextual to the Banking Process
2.3.7 Case Study: Dictionary and Domain Covered
2.3.8 Cognitive Computing Bots Available on Cloud Platforms
2.3.9 Case Study: Cloud-Provided Bots
2.3.10 Case Study: Pre-built Integration and Connector Services
2.3.11 Industrialisation of Bots Through Orchestration on Cloud
2.3.12 State Bank of India YONO Case Study—Digital Retail Banking Platform Embedding Automation into the Repurposed Legacy Applications
2.4 Conclusion–Leverage Industrialisation Engines for Tactical Realisation of Automation Benefits
3 Investing in Alignment of Industrialisation and Innovation Engines
3.1 Introduction—By Default Industrialisation and Innovation Engines Are Never Aligned
3.1.1 Misaligned Engines Start Accumulating Technical Debt
3.2 Build-Up of Technical Debt
3.2.1 Technical Debt Built up Due to Factors Beyond the Control of the Bank
3.2.2 Technical Debt Built up Due to the Decisions and Choices Made by the Bank
3.2.3 Technical Debt Build-Up Creates Existential Level Questions for the Bank
3.3 Managing Technical Debt
3.3.1 Case Study: UBS Has Budgeted to Spend 10% of Revenue on Technology
3.3.2 Empirical Study of Managing Technical Debt at 12 Banks
3.4 Adopting Industrialisation and Innovation Engine Alignment Framework to Manage Technical Debt
3.4.1 The Project-Based Approach to Managing Technology Debt
3.4.2 Establish a Bank-Wide Vision for Industrialisation and Innovation
3.4.3 Mapping the Project-Based Approach to Industrialisation and Innovation Engine Framework
3.5 Managing the Alignment Process
3.5.1 Continuously Rediscover Banking 4.0 Vision
3.5.2 Promote and Adopt Banking Technology Standards
3.5.3 Build Banking AI for the Bank
3.5.4 Adopt ISO 20022 Standards Internally in the Bank for Every Data and Process
3.5.5 Augment IT Governance to Encourage, Enable, and Enforce Paying off Technical Debt
3.5.6 Make Technology All-Pervasive
3.5.7 Continuously Benchmark Innovation
3.6 Conclusion—Continuously Rediscover, Benchmark, and Align
4 Industrialising Data
4.1 Introduction—Changed Data Requirements at Banks
4.1.1 Deliver Digital Experience in Real Time
4.1.2 Process High Volumes of Data
4.2 Key Indicators of Data Industrialisation
4.2.1 Aligned Industrialisation and Innovation Engines with Data Industrialisation Engines
4.2.2 Deliver Data in a Managed Service Mode
4.2.3 Better Certainty in Data Service Delivery
4.2.4 Deliver Greater Consistency and Accuracy in Models and Analytics
4.2.5 Deliver Industrialised Model and Analytics Development, Training and Deployment
4.2.6 Manage All Data Types
4.2.7 Deliver Conversational Banking at Scale
4.2.8 Better Data Governance
4.2.9 Industrialised Data Privacy Compliance
4.2.10 Empowered Business Managers
4.3 Industrialise Data Integration
4.3.1 Convergence of Data Preparation and Data Integration Tools
4.3.2 Industrialise Data Preparation
4.3.3 Data Wrangling for Machine Learning Models
4.3.4 Case Study: Google Cloud Data Wrangling Tool
4.3.5 Industrialised Data Preparation Tools Are Embedded with Statistical and Visualisation Capabilities
4.3.6 Case Study: Google Cloud Datalab Tool to Explore, Analyse, Transform, and Visualise Data for Building Machine Learning Models
4.3.7 Case Study: BigQuery Geospatial Visualisation Techniques
4.3.8 Embedding Geospatial Visualisation into Applications Empowers Business Managers
4.3.9 Case Study: Data Stream Processing Tools on Google Cloud
4.3.10 Embed Data Integration into Applications
4.3.11 Case Study: Building a Data Pipeline to Migrate Data to BigQuery
4.4 Cloud Data Platform as the Data Industrialisation and Innovation Engine
4.4.1 Cloud Data Platform Optimises Data Integration
4.4.2 Cloud Data Platform Augments Data Storage
4.5 Case Study—Cloud Data Platform
4.5.1 Case Study: Google Cloud Data Platform High-Level Architecture
4.5.2 Data Ingestion and Integration Tools from GCP, AWS, and Azure
4.5.3 Building and Orchestrating Data Pipelines on GCP, AWS, and Azure
4.5.4 Data Search on Data Catalog Capabilities on GCP, AWS, and Azure
4.5.5 SQL Data Modules on GCP, AWS, and Azure
4.5.6 NoSQL Data Capabilities in GCP, AWS, Azure
4.5.7 Case Study: BigQuery—Industrialised Big Data on Cloud
4.5.8 Data Storage on GCP, AWS, and Azure
4.5.9 Analytics and Visualisation on GCP, AWS, and Azure Platform
4.5.10 AI Machine Learning and AI as a Service Platform on GCP, AWS, and Azure
4.6 Industrialise Data Catalog to Deliver Data Governance
4.6.1 Case Study: Data Catalog to Implement Enterprise Data Governance Policy
4.6.2 Data Catalog as a Tool for Data Governance
4.6.3 Deliver Re-Usable Tag Templates to Build Consistency in Technical and Business Contexts
4.6.4 Implement Enterprise-Wide Data Governance on Data Catalog
4.6.5 Tag Metadata at Data Asset Hierarchies
4.6.6 Establishing Business Lineage to Empower Business Managers
4.6.7 Industrialise the Data Quality Process
4.7 The Future of Data Services
4.7.1 Case Study: Monitor Customer Behaviour in Real Time
4.7.2 Case Study: An Investment Bank Creating and Selling Thousands of Insightful Datasets Every Year
4.7.3 Case Study: The BBVA Bank Data Lab
4.8 Conclusion: Industrialised Data Enable Better Alignment of Industrialisation Engines
5 Deliver Digital Experience
5.1 Introduction: Digital Banking
5.2 Deliver Digital Experience of Banking 4.0 Services
5.2.1 Digital Customer Experience
5.2.2 Deliver Digital Experience in the Front Office
5.2.3 Industrialised Front Office
5.2.4 Deliver Digital Experience in the Back Office
5.2.5 Industrialised Back Office
5.2.6 Deliver Banking 4.0 Services by Integrating the Industrialised Back Office with the Industrialised Front Office
5.3 Hyper-Personalisation is the Foundation for Deliver the Digital Customer Experience
5.3.1 Personalisation of the Digital Customer Experience Has Matured at Banks
5.3.2 Hyper-personalisation to Deliver Digital Experience
5.4 BIAN Framework to Design Application Modernisation
5.4.1 Case Study: The BIAN Framework
5.4.2 BIAN Framework Helps to Build Consistency and Reusability into Microservices
5.4.3 BIAN Framework Helps in Identifying Wrappers to Modernize Underlying Applications
5.4.4 BIAN Framework to Build Microservices
5.4.5 BIAN as Standard for Designing Microservices
5.4.6 BIAN Service Domain as a Container
5.4.7 Standardised Integration
5.4.8 Standardised SLA and Intelligence
5.4.9 BIAN Delegated Service is a Framework to Build Agility
5.5 Modernise Applications to Deliver Digital Experience
5.5.1 Three Approaches to Application Modernisation
5.5.2 Aims of Application Modernisation
5.5.3 Scoping an Application Modernisation Program to Deliver an Industrialised Bank
5.6 Application Modernisation Through Microservice-Based Architecture
5.6.1 A Microservice Is Small Enough to Enable Agility, It Is Large Enough to Deliver Business functions
5.6.2 The Microservice Has to Be Agile, Flexible, Resilient, and Quickly Adjustable to Customer and Market Realities
5.6.3 Microservice Design Builds Agility into the Banking Business
5.6.4 Core Benefits of Microservice Architecture
5.6.5 Case Study: Microservice Architecture for Regulatory Reporting
5.6.6 Applications Fit for the Adoption of Microservice Design
5.7 Industrialisation of Microservices
5.7.1 Manage Version Control of Codebase to Re-use Code Repository
5.7.2 Managing Dependency on Another Library or Package
5.7.3 Separation of Configuration from Code
5.7.4 Backing Services
5.7.5 Managing the Services Version
5.7.6 Segregate Process from the State
5.7.7 Port Binding
5.7.8 Scale-Out
5.7.9 Disposability
5.7.10 Development Production Parity
5.7.11 Treat Logs as Event Streams
5.7.12 Run Admin Tasks as a One-off
5.8 Conclusion: Deliver Digital Experience by Modernising Applications
6 Cloud Adoption: A Foundational Engine
6.1 Introduction: Cloud adoption—A Foundational Engine
6.1.1 Cloud Adoption Lays Foundation for Industrialisation and Innovation at the Bank
6.1.2 Iterative Cloud Adoption Process Aligns Other Engines
6.2 Cloud-Native Applications (CNAs) Designed to Deliver Business Outcomes
6.2.1 Cloud-Native Applications Are Modular
6.2.2 Cloud-Native Applications Leverage APIs as an Integration Tool
6.2.3 Containerisation Makes Cloud-Native Applications Horizontally Scalable
6.2.4 Cloud-Native Applications Enable Continuous Deployment
6.2.5 Cloud-Native Applications Are Built-On Reusable Components
6.2.6 Cloud-Native Applications—Designed to Deliver Business Outcomes
6.2.7 BIAN Is a Design Template Used to Build Cloud-Native Applications
6.3 Managing Risks in the Cloud
6.3.1 Security Design and Architecture at Banks Must Complement the Security and Privacy Posture Provided by CSPs
6.3.2 Continuous and Real-Time Control and Risk Monitoring by the Bank and CSP
6.3.3 Clearly Defining and Action Items and Contractual Terms
6.4 Conclusion—Cloud Adoption Enables Alignment of Other Engines
7 Industrialise to Manage Changes in the External Environment
7.1 Introduction
7.1.1 An Inside-Out View of the Industrialisation of Banks
7.1.2 Outside-In View of the Industrialisation of Banks
7.1.3 Banks not Industrialising Are Likely to Lose Market Share
7.2 Digital ID Creates Trust in the Marketplace
7.2.1 Data Coverage of Digital ID
7.2.2 Digital ID Builds Trusted Marketplaces
7.2.3 Digital ID Delivers Industrialisation
7.2.4 Case Study: Aadhar ID Has Industrialised Customer Identification and Authentication
7.3 Industrialise to Deliver Innovation as BAU
7.3.1 Case Study: PNC Bank, USA Is Industrialising to Embrace Innovation
7.3.2 Case Study: To Embrace Innovation UBS Is Replacing 20% of the Application Portfolio
7.3.3 An Illustrative List of Innovation Vectors on the Industrialisation Foundation
7.3.4 Industrialising the Processing of Petabytes of Data Added Every Year
7.3.5 Blockchain Platform for Cross-Border Commerce, Payment, Banking, Clearing, and Settlement
7.3.6 Integrating Alternative Data Sources to Improve the Predictive Power of Models
7.3.7 ML Models to Enhance the Discriminatory Power of Credit Risk Models
7.3.8 Cloud Technologies Moving to the Mainstream
7.3.9 The Level of Disruption and Innovation May Vary Across the Product Portfolio
7.4 Industrialise to Manage Low-Interest Rate Regimes
7.4.1 Create Scale Through Industrialisation to Service the Cost of Capital Under Low-Interest Rate Regimes
7.4.2 Low-Interest Rate and Bank Profitability Regimes with High Disruptable Business Segments
7.4.3 Skill Distribution at Industrialised Banks
7.5 Industrialise to Engage Regulators
7.5.1 Technology Is Breaking Down the Barriers to Collaborate with Regulators
7.5.2 The Regulator Has the Dream of Monitoring Every Transaction in an Economy
7.5.3 Regulators Are Digitising: A Vision Statement on Digitisation by Ravi Menon, MAS MD While Establishing the Data Analytics Group (DAG)
7.5.4 Build Agility to Incorporate Regulatory Mandates on the Use of Machine Learning Models in Banking
7.6 Industrialise Risk, Finance, and Compliance
7.6.1 Containing Risk, Finance, and Compliance Annual Costs to Within 10–25 Basis Points of Assets
7.6.2 Building a Banking Ontology, Data Dictionaries, and Data Taxonomies to Manage Voluminous Data Through Semantics and Machine Learning
7.6.3 The Financial Industry Business Ontology (FIBO)
7.6.4 The ISO 20022 Data Model with Business Concepts
7.6.5 Managing and Monitoring Changes in Credit Risk: Credit Risk 3.0
7.6.6 Technology and External Data Is Available Now
7.6.7 Monitoring High-Frequency Indicators of Change in Credit Risk
7.6.8 Monitoring Theme-Based Actionables and Alerts
7.6.9 Operational Risk Management 3.0
7.6.10 Elevated Regulatory Mandates to Manage Non-financial Risks Need ORM 3.0
7.6.11 Forward-Looking Automated Risk and Control Assessment
7.6.12 An Illustrative List of Forward-Looking Risk and Control Monitoring
7.6.13 Capability of Monitoring, Preventing, and Recovering from a Cyber Attack
7.7 Conclusion: Banks Need to Industrialise to Manage and Exploit the Changes in the External Environment
8 Platform Banking Business
8.1 Introduction
8.1.1 Banking Customers and Partners Have Shifted Their Residence to the Cloud and the Marketplace
8.1.2 Industrialisation Enables a Bank to Reach Its Customers and Partners in the Cloud and Marketplace
8.1.3 Platform Banking Helps Participation in the Marketplace
8.1.4 Platform Banking Helps Banks to Become Technology Firms
8.1.5 Case Study: Capital One Has Created a Vision and Implemented a Strategy to Become a Technology Firm
8.1.6 Case Study: DBS Aspires to Be in the League of Big Tech Players like Google and Facebook
8.2 Platform Banking Is the New Capital for Banks
8.2.1 Develop a Strategy to Build Platforms and not Implement Projects
8.2.2 Innovative Banking Technology Is Disrupting Marketplaces
8.3 Platform Banking Business
8.3.1 Case Study: DBS Distributes Manulife Products
8.3.2 Case Study: Citi Launches IKEA Family Credit Card in Partnership with Ikea
8.3.3 Platform Business Case Studies
8.4 Replicate the Big Tech Business Model
8.4.1 Innovation Is the New Oxygen for Banks
8.4.2 Innovate to Disrupt Cost and Build Non-interest Revenue Streams
8.4.3 Cost to Income Ratio as an Indicator of How Far the Bank Is from Becoming a Tech Firm
8.4.4 An Illustrative List of Innovations Being Pursued by Banks
8.4.5 Banks Are Discovering Data Is the New Oil
8.4.6 Case Study: Digital Investment Bank of the Future as Data Provider to the Customer
8.4.7 Embedding the Bank in the Customer Journey
8.4.8 Case Study: Innovative and Differentiated Value Propositions for SME and Commercial Customers
8.4.9 Number of Banking Platforms as an Indicator of the Industrialisation of a Bank
8.5 APIs Are the Building Blocks of the Platform Banking Business Model
8.5.1 A Platform Banking Business Model Runs on a Low Latency Agile Technology
8.5.2 Platform Banking Services Marketplaces Through Published APIs
8.5.3 Closed or Internal APIs Industrialise Integration of the Front Office with the Back Office
8.5.4 Open APIs: Banks Provide Four Types of APIs
8.5.5 Typical Open APIs Provided by a Mature Bank’s API Hub
8.5.6 Partner APIs Open only to the pre-selected partner
8.5.7 Case Study: Integration with Accounting Software
8.5.8 APIs Open to Authorised Members of a Pre-selected Organisation or Community
8.5.9 APIs Are Made Available by Registering with the Organisation Subject to Certain Conditions
8.5.10 APIs Are Made Available to the Authenticated Parties by Registering with the Organisation
8.5.11 APIs Are 24/7 Invitations to Customers and Partners to Consume the Bank’s Services
8.5.12 Regulatory Mandated Open Banking Versus API Banking
8.5.13 Open Banking Is Driven by Regulatory Mandate
8.5.14 Open Banking Coverage Varies Widely Across the Jurisdiction
8.5.15 Open Banking APIs May not Be an Indicator of the Industrialisation of the Bank
8.5.16 API Banking
8.5.17 Mutual Contracts Drive API Banking
8.5.18 Case Study: BBVA Open Platforms Provide API Banking
8.5.19 Open APIs Are not Standardised
8.5.20 Currently, Banks Use Disparate, Bank-Specific Formats, and Definitions for APIs
8.5.21 Case Study: The Working of the API Register of the Monetary Authority of Singapore
8.5.22 Risks in Open API
8.5.23 Regulatory Framework to Manage Data Shared Through APIs
8.5.24 Open API Regulatory Technical Standards for Stronger Customer Authentication and Secured Sharing
8.5.25 Increased Cyber Risk Management Due to API Provision at Banks
8.6 Blockchain Platforms Are Disrupting the Revenue Model of Banks
8.6.1 Blockchain is an Emerging Technology for Building Immutability and Trust
8.6.2 Players in the Blockchain Business Model
8.6.3 Blockchain Business Model
8.6.4 Case Study: Trade Finance Platforms on Blockchain
8.7 Conclusion: Platform Banking for New Revenue Streams
9 APIs Are the Public Persona of an Industrialised Bank
9.1 Introduction:—APIs Industrialise Integration and Collaboration with Customers and Partners
9.2 Designing APIs
9.2.1 API Business Model Drives the API Design
9.2.2 API Monetisation Strategy Drives the API Design
9.2.3 Legacy Application Wrapper APIs
9.2.4 Process APIs
9.2.5 Experience APIs
9.3 Developing APIs
9.3.1 Deliver integration with a very large number of endpoints
9.3.2 Agile Integration
9.3.3 Stateless API to make Integration Scalable
9.3.4 Deployment in Seconds
9.3.5 Continuous Integration
9.3.6 Easier Retirement
9.3.7 Minimum Interdependencies
9.3.8 Decentralised API Development Teams
9.3.9 Leveraging Automated Development Tools
9.3.10 Consistent and Standardised API Design and Architecture Framework
9.3.11 Deployment of APIs
9.3.12 Discoverable REST APIs
9.3.13 Agile API Operationalisation
9.4 APIs Are the Public Persona of the Bank
9.4.1 APIs Across the Two Banks Are Different
9.4.2 Establish API Gateway to Manage APIs
9.4.3 Functions of an API Gateway
9.4.4 API Gateway Enables Internal Multi-Tenancy and Decentralised Ownership
9.4.5 Federated API Gateway to Build Consistency Across the Business Units of the Bank
9.4.6 Building API Security Blocks
9.5 Publish APIs
9.5.1 Case Study–APIs as a Tool to Deliver Digital Experience at DBS
9.5.2 The Cost of Building APIs is the Major Challenge for Smaller Banks
9.6 Conclusion: APIs Are the Tool to Disrupt the Business Model of the Bank
10 BIAN Framework to Build Banking AI and Semantic APIs
10.1 Introduction: Developing Banking AI
10.1.1 Existing AI Platforms Are Incapable of Learning Banking Processes
10.1.2 Develop Banking AI to Enhance the Accuracy of AI Systems
10.1.3 BIAN Service Domain Structure is a Template to Build Banking AI
10.2 BIAN Artefacts Are Benchmarks for Building Banking AI
10.2.1 Leverage BIAN Artefacts to Build Banking AI Machines
10.2.2 Business Context is the Key to Industrialise Banks
10.2.3 BIAN Metamodels as a Benchmark of Banking Process Hierarchies and Process Aggregation
10.2.4 BIAN Service Domain Represents a Unique and Exclusive Business Context
10.3 ISO 20022 Dictionaries Are Benchmarks for Banking Definition, Processes, and Concepts
10.3.1 BIAN is the Benchmark of Business Definition
10.3.2 BIAN Leverages a Standardised ISO 20022 Data Model for Products and Processes and Computation
10.3.3 ISO 20022 Business Object Model is Extensible
10.3.4 BIAN and ISO 20022 Are Extendible
10.4 BIAN Service Domain as a Template for Identifying Data Ownership of Microservices and APIs
10.4.1 How to Identify Original Data from a Set of Electronic Data?
10.4.2 Role of a BIAN Service Domain for the Owned Control Record
10.4.3 BIAN Service Domains as a Template to Identify Duplicate Application Functionality in Banks
10.4.4 Template to Identify Duplicate Data in the Bank
10.5 BIAN Templates
10.5.1 BIAN Templates Are Applied to the Deployment
10.5.2 BIAN Functional Patterns
10.5.3 BIAN Service Operations
10.5.4 BIAN Action Terms that Characterize the Purpose of an Offered Service
10.5.5 BIAN Action Terms Are the Foundation to Build Semantic APIs
10.5.6 BIAN Action Terms as a Template for Granularity-Level Decision Making
10.5.7 An Indicative Proportionality Level of API Granularity
10.5.8 BIAN is a Benchmark Semantic API and not a Design Standard
10.5.9 Compliance to BIAN Framework Over Iterative Cycles as a Template for IT Governance and Architecture
10.6 BIAN Semantic APIs as Templates
10.6.1 BIAN Template and Attribute Directory for Control Record Help in Creating Semantic APIs
10.6.2 BIAN Framework to Build Wrapped/Repurposed Legacy System APIs
10.6.3 BIAN Framework to Build Distribution and Customer Experience
10.6.4 BIAN Semantic API Examples as a Benchmark
10.6.5 Defining Semantic APIs: Learning from 180+ Semantic REST APIs Published by BIAN
10.6.6 The Purpose of Semantics is to Consistently Interpret the Nature or Purpose of the Service Domain
10.7 Conclusion: Building Banking AI by Aligning with BIAN Templates and Benchmarks is an Iterative Journey
11 Conversational Banking
11.1 Introduction to Conversational Banking
11.1.1 Conversational Banking is About Addressing the Limitations of Call Centre and IVR Technology
11.1.2 Conversational Banking Requires Establishing Conversational CoE at the Bank
11.2 Conversational Banking CoE
11.3 Conversational Banking Design Principles
11.4 The State of Conversational Technology
11.4.1 Case Study: A Virtual Assistant on Amazon Lex
11.4.2 Case Study: Azure Cognitive Services
11.4.3 Case Study: Kasisto
11.5 Conversational Banking Service Use Cases and Intents
11.6 Implementing Conversational Banking at a Scale
11.6.1 Understanding Existing Conversation Services in the Bank
11.6.2 Segment Audience Conversing with Bank
11.6.3 Capture Audience Experience in a Database
11.6.4 Build Conversational Models
11.6.5 Build Models to Personalise Conversations
11.6.6 Design Conversations
11.6.7 Reimagine Customer Journeys
11.7 The Future of Conversational Banking
11.7.1 Challenges in Building Conversational Banking
11.7.2 Limitation of Conversational Banking Technology
11.7.3 Designing Conversational Banking User Interfaces
11.7.4 Measuring the Accuracy and Effectiveness of Conversational Banking
11.7.5 Calibrate Customer Intent and Financial Obligation Based on Authentication Strength
11.7.6 Relative Frequency of Conversational Banking Service Use Cases
11.7.7 Success Measures of Conversational Banking
11.8 Conclusion: Building Conversational Banking at a Scale
12 Banking AI as a Service
12.1 Introduction: Banking AI Takes Forward the Industrialisation Agenda
12.1.1 Banking AI as a Service is a Basic Building Block in the Industrialisation Agenda
12.2 Establish Banking AI CoE
12.3 Establish Banking AI Organisation
12.3.1 Banks Are Appointing Chief AI Officers
12.3.2 Case Study: Chief AI Officer at JPMC
12.3.3 Availability of AI Skills is a Major Area of Constraint
12.3.4 Case Study: Augmenting AI Capabilities Through the Acquisition of an AI Consulting Firm
12.3.5 Case Study: Augmenting AI Capabilities Through Partnerships with Academia
12.3.6 Case Study: Augmenting AI Capabilities Through Investment in AI Startups
12.3.7 Augmenting the AI Team Through a Training Partnership with Academia
12.3.8 Banks Are Collaborating with Cloud Service Providers on AI Initiatives
12.3.9 Case Study: HSBC Partners with Element AI
12.4 Banking AI as a Service (BAIaaS) on a Cloud Data AI Platform
12.4.1 The Development of AI as a Service (AiaaS) for Banks Will Be a Very Long Journey
12.4.2 General Purpose AI Platform Capabilities Have Matured in the Past Three Years or so
12.4.3 Case Study: GCP AI Platform—A Platform to Provide AI as a Service
12.4.4 Case Study: General Purpose AI Platform
12.4.5 AI Services on a General Purpose AI Platform
12.4.6 Managing Cloud AI Platforms
12.4.7 Banks Need to Develop Narrow AI Very Specific to the Banking Process
12.5 Build a Semantic Technology Foundation for Banking AI
12.5.1 ISO 20022 Has Published a Banking Semantic Dictionary
12.5.2 ISO 20022 Dictionary for Banks is at Three-Level—Semantic, Logical and Physical
12.5.3 Further, Augment ISO 20022
12.5.4 Context Discovery Drives the Approach to Building Semantic Technology
12.6 Building Banking Knowledge Organisation and Representation
12.6.1 Establish Knowledge Architect and Knowledge Engineer
12.7 Case Study: Fintech AI Use Cases
12.7.1 Robo-Advice
12.7.2 Customer Complaints
12.7.3 Credit Scoring
12.8 Conclusion: Banking AI as a Service is not a Technology Project
13 Fintech: The Innovation Benchmark
13.1 Fintech 1.0: Wakeup Call for the Banks (2014–2018)
13.1.1 Regulators Were in a Dilemma
13.2 Fintech 2.0: Challenger to the Banks (2016–2020)
13.2.1 Regulatory Sandbox: A Landmark Initiative Across the Jurisdiction
13.2.2 Case Study: RBI Report of the Working Group on Fintech and Digital Banking
13.2.3 Digital Banks as an Extension of Fintechs to Regulated Banks
13.3 Fintech 3.0: Partnering with Banks (2018–2022 Onwards)
13.3.1 Fintechs as Technology Partners of Regulated Entities
13.3.2 Most of the Financial Services Firms Are Likely to Be a White Label in Their Offerings of Fintech Services and the Services of Other Financial Services Firms
13.3.3 Provision of Business Outcome for Partner Banks Is Emerging as a Business Model for Technology Fintechs
13.3.4 Fintechs Are Innovation Partners for Banks
13.3.5 Disruptive Marketplaces Are Emerging as the Next Bed of Innovation
13.3.6 Banks Have a Shortage of AI Skills: AI Fintechs Are Likely to Get Better Traction
13.3.7 Case Study: Fintech Upgrading to Become a Financial Services Firm
13.4 Fintech 4.0: Level Playing with the Industrialised Banks (2021–)
13.4.1 Fintechs Cannot Replace Banks, but They Will Make Banking Better
13.4.2 Developing AI/ML Models by Standalone Fintechs Will Be Challenging Going Forward
13.4.3 Industrialised Banks Are Becoming Equipped to Compete with Fintechs Even at a Technology Level
13.4.4 Regulators Are Establishing a Level Playing Field Between Traditional Financial Services Firms and Fintechs
13.4.5 Level Playing Regulation Case Study: Fintech Balance Sheet Lending
13.4.6 Level Playing Regulation Case Study: Crowdfunding Platform
13.4.7 Case Study: Emerging Regulatory Regimes in China for Fintechs in a Financial Service Business
13.4.8 Case Study: In India Digital Lending Is Being Brought Within the Ambit of Banking Regulation
13.5 Conclusion: An Industrialised Bank Is Converging with Fintech
Further Reading Sources on the Web
Banks and Financial Services Firms
Fintechs, Digital Banks, Marketplace, APIs, and AI Firms