SAS Banking And Insurance Day May 2014 Andy Scherpenberg, Fraud Expert, SAS

Transcription

Fighting online payments fraudSAS Banking and Insurance Day – May 2014Andy Scherpenberg, Fraud Expert, SAS

Fraudiseverywhere

edit CardClaimsInvoiceValue ChainDarknetProducts/Services/SalesCyber g Functions (HR, Procurement, Legal)PayrollCall CenterRogue TradingProcurementTheft

edit CardClaimsInvoiceValue ChainDarknetProducts/Services/SalesCyber g Functions (HR, Procurement, Legal)PayrollCall CenterRogue TradingProcurementTheft

SOME HISTORY2007300 B.C.200420152012

WHAT HAPPENS?

No single layer of fraud prevention or authentication is enoughto keep determined fraudsters out of enterprise systems.Avivah LitanVP & Distinguised Analyst – Gartner Research

LANDSCAPEComplexityAverage FraudDeliberate FraudOrganized Criminal Gangs

BALLOON EFFECTFRAUD IS MOVINGCompanies implementmanual/basic fraud detection

BALLOON EFFECTFRAUD IS MOVINGCompanies implementmanual/basic fraud detectionFraud MO’s change,detection becomes more complex

FRAUD LOSSES FEBELFIN FIGURESOnline Fraud

MISSION HAS CHANGEDFacilitating Payments

MISSION HAS CHANGEDFacilitating Secure Payments

A NEW ERAA NEW ERA Customers want faster payments (i.e. mobile) Fraudsters adapt People are not robots, they change theirbehavior

A NEW ERABRINGS NEW CHALLENGES‼ Payment processing times decline‼ Modus Operandi changes rapidly‼ False positive reduction remains a challenge

APPROACHDETECTING , TREATING, IMPROVINGAnalyticalOptimizationReal TimeDetectionEngineCaseManagement

APPROACHDETECTING, TREATING, IMPROVINGAnalyticalOptimization

MATURITY MODELOutsetMaturationLayer 1EndpointLayer 2SessionLayer 3ChannelMaturationLayer 4Multi-ChannelLayer 5Social Network

MATURITY MODELLayer 1EndpointLayer 2SessionLayer 3Channel

IT TAKES TWO TO TANGOExpert KnowledgeAdvanced AnalyticsYesterday’s methods are insufficient toaddress tomorrow’s fraud risk

WE DELIVER INSIGHT“SAS uses embedded analytical models to uncover things that the humanbrain cannot. Our investigators can focus on cases with the highestprobability of fraud or highest potential financial damage.”Maya MaškováHead of Internal AuditAllianz

MODERN FRAUDDETECTIONAnomalyDetectionHYBRID APPROACH TO Social Network Analysis

Social Network Analysis Demo- Tom StockFraud is not commited by machines,but by humans. Cunning as they might be, they can rarely do it alone.The Social Network Analysis shows the often hidden links between customers, companies and employees

DIFFERENTMETHODOLOGIESWHY SHOULD YOU USE ANALYTICS? Advanced analytics with SNAFRAUD Advanced analytics Business rules Random samplesPOPULATION

THE MULES

WHO ARE THEY?

ANALYST vs. PREDICTIVE 18%Marital StatusLanguage14%Marital status17%Monetary inflow22%Postal Code2%Education Level3%Client relationship age2%99% accurate

WHAT DOES ITLOOK LIKE?TransactionsProcessingEngineReal TimeDetectionEngineAlert / Case ManagementScenario Designer

VISUAL SCENARIODESIGNPUTTING CONTROLS WHERE THEY BELONGDesign & deploy

Visual Scenario Designer Demo- Tom StockDemonstrates how an end user can tune an existing scenario with instant feedback on the effect of the modifications.Also shows how the user can deploy a modified scenario into production. All of this without IT intervention and with full audit trail

Visual Analytics Demo- Tom StockImportant as a good starting point might be, nothing is more important than continuous adaptation.Graphical report-builing and data mining give insight in the current state of affairs, all in the blink of an eye,allowing you to react to changes in the outside world faster than ever before.ContinuousImprovement

A solution integrated bydesignCapability can bedeployed over time, atthe speed and extentyou chooseEach next step builds onthe previous, no wasteExpand scope to othertypes of fraud or othersilosSTEP-BY-STEP ROAD TO MATURITYFocus on rapid deployment of a line of defence, and prepare for the ADMAPNetwork AnalysisAnomalyDetectionPredictive ModelsRule-basedAlert generationManualDetectionTime

KEYDIFFERENTIATORSLeader inGartner'sMagicQuadrantMARKET LEADER ACROSS THE BOARDForrestercites SAS as aLeaderLeader inGartner'sMagicQuadrantForresterCites SAS asa LeaderChartisResearchLeadersCategoryOnly Vendor in this space that is leading in each of these sectorsDataIntegrationBig DataPredictiveAnalyticsBusinessIntelligence ering5

OUR VISIONAs promised, a look into the future .

OUR VISIONInformation theftThreats on mobile devicesCybersecurity attacks on backendSophisticated malware, built specifically for your institutionExperts are convinced major institutions already have malware inside

Sounds unlikely? Unbelievable?

If I would have been standing here before you 5 years ago,and told you that your customers would give away theirbanking details and credentials over the phone, that wouldhave sounded more unlikely than my closing message just now.

Closing Message

We’re watching intently how the sea gently touches our feet

But we don’t realize what is coming

Gartner's Magic Quadrant Anti-Money Laundering Chartis Research Leaders Category Enterprise Fraud Management Forrester Cites SAS as a Leader Business Intelligence & Analytics Leader in Gartner's Magic Quadrant Big Data Predictive Analytics Forrester cites SAS as a Leader Only Vendor in this space that is leading in each of these sectors 5