Use of analytics in financial services to combat Fraud and Anti Money Laundering. The Data Warehouse Institute (TDWI) 3 August 2011 Richard Moore Head of Security Advisory & Investigations, Group Security Commonwealth Bank of Australia
Disclaimer The material that follows is a presentation of general background information about the Bank s activities current at the date of the presentation, 3 August 2011. It is information given in summary form and does not purport to be complete. It is not intended to be relied upon as advice to investors or potential investors and does not take into account the investment objectives, financial situation or needs of any particular investor. These should be considered, with or without professional advice when deciding if an investment is appropriate. 2
Themes Evolution of fraud and Anti Money Laundering in banks Technology and analysis working together Some high level successes executing strategies Benefits of the approach including Increased customer service Reduced fraud Reduced regulatory risks 3
Commonwealth Bank Group scale Australia s leading provider of integrated financial services Employs 42,500+ staff globally Customer base 11+M 7 million+ domestic transactions each day 70,000 international transactions each day More points of access than any other Australian bank 4
Firstly, Group Security s Operating Model Chief Security Officer Fraud Risk Advisory Investigations Protective Security Business Continuity Management Financial Crime Shared Services New Zealand Indonesia Bankwest Fraud policies and standards Fraud Risk Assurance Program Training & Awareness Intelligence & Planning Strategy & Capability Investigations policy & standards Staff investigations Major fraud investigations Internal control reviews Protective security policy & standards Physical Security Risk Assessment Program Global Incident Monitoring Centre Business continuity management policy BCM systems & reporting Crisis & emergency management Fraud detection policy & standards AML, fraud and Sanctions operations Rules, modelling, data analytics & information Measurement Others 5
2005: Group Security IT Strategic Plan Access to timely and accurate information about current & emerging risks to the bank, including: fraud money laundering terrorist financing; and other criminal activities Scaleable ability to monitor activity across a wide range of products and channels, both in Australia and overseas Help all business units to monitor and respond to their risks 24 x 7, minimising our customer s fraud exposure Deliver and demonstrate an efficient and valuable service to our customers & the Group Reduce the total cost of ownership for our fraud detection solutions 6
Our strategy Bespoke Scaleable (siloed) Systems Solutions Data Re-use for Multiple BI Data Marts Improved Customer Service Different Fraud Departments Holistic Disconnected Financial Approach Crime Management High Quality Metrics Lower Costs Higher Costs Centralised Case Management Poor Case Management 7
Financial Crime Platform 8
How it works Fraud AML Sanctions 9
Filter approach targeting the right activity Analytics Case Management Detection Engines Fraud AML Sanctions 10
Our Evolution of Analytics Off the shelf solutions neural and parameter driven Vendor supported models Basic rule writing capability Converted operations people to analysts 1999-2003 Started recruiting analytics resources. Enhanced rule writing Use of analytical software 2003-2007 Data management platform rich, clean data. Advanced analytical resources. Advanced rule capability Integrated filters, data matching, decision tree, behavioural modelling capability. Bank driven models 2007 to now Driven by; Increasing business volumes & fraud Investment in people & technology Vendors not able to keep up with change 11
Goal of Event Prevention: Minimise Cost & Risk Finite resources to action alerts require efficiency Right transaction in a high volume/low risk environment Reduce customer impact whilst minimising loss Manage total cost Law of diminishing returns 12
Customer Risk & Value Analytics helps us prioritise focus in higher risk segments those that hurt. Customer service priority to reduce impact and avoid loss Reduce reputation & financial loss to Group Maintain public confidence in the payment channels 13
Optimising Detection Program Analytics and Operations team working together capturing more fraud with less alerts Understanding the environment, customer response and rule/model performance Prioritising alerts within operational constraints 14
Showing the Benefits (Loss to Turnover Ratios) Improvement in all categories relative to volume Actual dollar fraud losses reducing year on year in the main 15
Lessons learnt Invest in developing a strategy for financial crimes control, including: IT systems organisational capability critical success measures Good analytical systems demand higher quality, granular data. Enterprise-wide strategy is required Leverage existing & emerging operational, technological or regulatory requirements (i.e. AML, privacy, technological changes) 16
Emerging Threats & Our Response Criminals morph and target weakest point Analytics, agile systems and flexible operations allow us to adapt to changing trends New products and channels are brought into our monitoring program Supporting our customers in strengthening their controls 17
Questions. 18