Predictive Analytics in Insurance Getting it right when your customers need you most

Transcription

1 Predictive Analytics in Insurance Getting it right when your customers need you most Rob McCullagh Tony Boobier Dr Claire Jordan 16 November 2016

2 Today s Speakers Tony Boobier Published Author Analytics for Insurance linkedin.com/in/tonyboobier Rob Mc Cullagh Strategic Account Director Dr Claire Jordan Senior Analytics Consultant linkedin.com/in/robmccullagh linkedin.com/in/clairejordan 2

3 Today s Topics The Analytical Insurer Using Predictive Analytics to drive customer experience and increase profitability 3 Predictive Analytics in Insurance injecting intelligence into the claims process

4 About me: Independent consultant and industry commentator Previously World Wide Executive at IBM 30 years insurance industry experience Author of Analytics for Insurance 4

5 5 The analytical insurer is an insurance company using analytics throughout its organization to improve business performance.

6 Market Changes require Greater Insights Economy Incl Brexit Distribution Digital and More Savvy Customers Competitive Environment Risk Management 6 Demand for Growth

7 The Mega Technology Trends in Insurance Cloud Mobile Social FinTech Analytics 7

8 The New Mega Trends in Insurance The New Mega Trends in Insurance The Connected Car /Home/Person Innovating to Zero Urbanization Geo-socialisation 8

9 The Era of Big Data The Era of Big Data Un-Structured % % Structured A Complex Network of Over 1 Trillion Devices 9 Generating 2.5 Billion Gigabytes of Data Daily

10 Analytics Create Create Actionable Actionable Insight Insight Strategy Tactics Best Practices 10 Distribution Sales Resources Supply Chain Operational Management Financial Management Marketing Asset Management Risk & Compliance Claims Advantage Differentiation Profitable Growth

11 Business Value The Fourth Age of Analytics 11 Alerts Unusual activity Simulation Impact of rising rates Reporting Last quarter s results Cognitive learns Prescriptive Optimization Highest returning portfolio based on risk appetite Understands natural language, hypothesizes, adapts & Stochastic Optimization Managed exposure to Cat Risks Predictive Forecasting Trending analysis Modeling Predicting elasticity of insurance rates Descriptive Analysis Product profitability Discovery # security breaches this month vs. last

12 By % of all businesses will be making decisions using artificial intelligence and prediction Gartner 12

13 An Enterprise View on Analytics is Essential Asset and Supply Chain Management Servicing including Claims Management Risk Appetite and Solvency Optimisation 13 Marketing and Distribution Underwriting And Pricing

14 Converged Insights are Critical Business data, weather data, location data, contextual data, via analytics will transform Enterprise-wide decision making 14

15 What will Your Future World Look Like? Redesigned business models New skills, roles and professions Different organisational Structures New styles of leadership 15

16 Four Key Building Blocks of Change Technology Talent Management Implementation Leadership 16

17 About me: passionate about data and the application of mathematical and statistical techniques to solve real business problems 17

18 Journey Satisfaction impacts Revenue Growth 8 Revenue Growth % - Insurance Industry 2011 v 2010 A 6 N Revenue Growth % 4 2 C M Company Name 0-2 E B K Journey Satisfaction 2011 Average 18 of satisfaction with each company s three key journeys Source 2011 McKinsey Multi-Industry Survey company financial statements

19 Success Stories Increase customer satisfaction BES (Banco Espirito Santo) reduced attrition by 15-20% FBTO increases marketing ROI by 29% Increase customer attrition Reduce claims cost Infinity Insurance increased subrogation recoveries by $12m ING Reduced Call Centre Costs by 20-40% Improve key to key KPI performance 19

20 What is Predictive Analytics? The right solution for the right business problem Reporting Analysis & Monitoring What happened? Why did it happen? What should I do next? What will happen? Prescriptive Analytics Predictive Analytics 20 Copyright Presidion Ltd

21 The Predictive Insurance Company Customer LTV Supplier Analysis Customer Retention Claims Processing Complaints Up-Sell/Cross Sell Risk X X Behavioural/Descriptive Information Claim History Call Centre Interactions Online Activity Product Information Demographics 21 External data Weather data Social Media data Regulatory Information Attitudinal Information Complaints Surveys Call Centre Notes s

22 Your customers 22

23 The Predictive Insurance Company Customer LTV Supplier Analysis Customer Retention Claims Processing Complaints Up-Sell/Cross Sell Risk X X Behavioural/Descriptive Information Claim History Call Centre Interactions Online Activity Product Information Demographics 23 External data Weather data Social Media data Attitudinal Information Complaints Surveys Call Centre Notes s

24 Improving the Claims Handling Process Service : Fast Track Fast Track Go through regular First notification of loss PA Request additional information processing and assignment Escalation Refer to SIU 24 Suspect

25 Predictive Analytics and the Claims Process 25

26 Example Motor Claims Process Meet Michael Jones: 26

27 Michael s Experience Michael calls Insurance Company Sarah asks some questions Autocorrect Garage Michael s car is fixed 27 Michael is a happy customer

28 Sarah s Experience Sarah takes a call from Michael Sarah is prompted by the system to ask certain questions Autocorrect Garage Sarah tells Michael where to bring his car to be fixed 28 Sarah is confident she has dealt with Michael efficiently and correctly

29 Predictive Analytics at work 78% Michael makes contact with his insurance company he damaged the front bumper of his car when he veered off the road Sarah is the claims handler dealing with his call 29

30 Predictive Analytics at work 78% Michael makes contact with his insurance company he damaged the front bumper of his car when he veered off the road Sarah is the claims handler dealing with his call 83% Prompted by the system Sarah asks Michael were there any witnesses to the accident. Michael says yes, the car behind him stopped to see if all was okay and gave Michael his name and number 30

31 Predictive Analytics at work 78% Michael makes contact with his insurance company he damaged the front bumper of his car when he veered off the road Sarah is the claims handler dealing with his call 83% Prompted by the system Sarah asks Michael were there any witnesses to the accident. Michael says yes, the car behind him stopped to see if all was okay and gave Michael his name and number 91% Again, prompted by the system Sarah asks Michael was anyone injured in the accident. Michael replies no, there were no injuries. 31

32 Predictive Analytics at work 78% 83% Michael makes contact with his insurance company he damaged the front bumper of his car when he veered off the road Sarah is the claims handler dealing with his call Prompted by the system Sarah asks Michael were there any witnesses to the accident. Michael says yes, the car behind him stopped to see if all was okay and gave Michael his name and number 91% Again, prompted by the system Sarah asks Michael was anyone injured in the accident. Michael replies no, there were no injuries 91% Sarah is directed by the system to fast track Michael s claim and information relating to most appropriate garage is relayed to Michael 32

33 Michael s Options Approved Repairer Average Lead Time Average Repair Time Average Satisfaction Score Distance NPS Autocorrect Garage 3 days 4.5 days 89% 3 miles High Repair Centre 5 days 4 days 74% 2.5 miles Mid BRC 7 days 5 days 71% 1.5 miles Low 33

34 Real Time Front End System 34

35 Benefits of Predictive Analytics Without Predictive Analytics With Predictive Analytics Call Received Call Received Agent Inputs Claims Predicted claim duration Claim Classification Agent Inputs Claims Documents Sent Claims Handler Clarification Call Repair Approval Request further information Claim Classification Recommended fast track Supplier Analysis Garage Waiting List Resolution Repair Approval Resolution 35 Time to resolution: 2 Weeks Number of contacts: 2 Time to resolution: 3 Days Number of contacts: 1

36 Unstructured Data? Text Analytics What do your customers really think? 36

37 Journey-led transformations deliver impact across multiple dimensions 37

38 Questions and Answers 38

39 Thank you for watching Follow Us and check out our Blog (0) (UK) +353 (0) (Ireland)