geetanjali chakraborty advanced analytics & predictive ... · cool analytics for the insurance...

34
Cool Analytics for the Insurance Industry IASA Conference November 21, 2014 Geetanjali Chakraborty Advanced Analytics & Predictive Modeling Practice Deloitte Consulting 0

Upload: others

Post on 08-Jul-2020

7 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Cool Analytics for the Insurance Industry

IASA Conference November 21, 2014

Geetanjali Chakraborty

Advanced Analytics & Predictive Modeling Practice

Deloitte Consulting

0

Page 2: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Agenda

Analytics is all around us…

What is analytics?

How analytics is being used in insurance?

Lifestyle Based Analytics

So many have already done it…

The savings potential

Questions

Page 3: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Analytics is all around us…

Page 4: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Analytics is all around us…

Page 5: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Why is analytics such a hot topic?

“…Perhaps the most important cultural trend today: the explosion of data about every aspect of our world and the rise of applied math gurus who know how to use it.“

The increase in digital footprint, the rise of cheap computing power and digital storage, and the seamless integration of

networks are allowing the accumulation of huge amounts of data.

Data is information about the past. Analytics can make it about the present and the future. Knowledge and insights about the future

can drive significant business value

“…the world contains an unimaginably vast amount of digital information which is getting ever vaster ever more rapidly.”

Page 6: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Analytics in the insurance industry

Insurance is a data rich industry and has long

mined its data to improve pricing and

underwriting activities

Page 7: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Pricing

Underwriting

Customer

Service

Marketing and

Agency Management

Claims

Management

• Targeted Lead Generation

• Cross-Selling Potential

• Agency/Agent Management, Training, Servicing

• Automated Processing and Triage

• Fraud/Salvage/Subrogation Potential

• Duration Improvement and Litigation management

• Tiering, schedule plan

• Class plan optimization and optimal scheduled credits/debits

• Enhanced underwriting decision making

• Risk selection, retention strategies, automated underwriting

• Resource allocation, straight-through processing

Traditional Applications

Emerging Trends

• Queue Prioritization

• Service Offerings

• Resource Allocation

Predictive Analytics in Insurance

Page 8: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

What is analytics?

Page 9: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

What do you think of when you hear “Analytics”?

Analytics is the discovery and communication of meaningful patterns in data; relying on the

simultaneous application of statistics, computer

programming and operations research to quantify insights.

Analytics imply a wide range of possibilities in its definition, its business application, and its delivery.

Statistics

Numbers

Business Intelligence

Analysis of data

Data and Computers

Page 10: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Basic Ingredients of Analytics

Basic ingredients of analytics include Data that contains insights, intelligence to extract those insights and act on them and Technologies to

implement appropriate actions.

External

Data

Internal

Data

Synthetic

Data

Data

Intelligence

Technologies

Page 11: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.10

Example of Data Sources – P&C

3rd Party Data

Marketing and Sales

Claims Data

Weather

CustomerData

PolicyInformation

Coverage Information

AgencyInformation

BillingData

Customer Data

Policy Records Correspondence

Policyholder InfoExperience DataPolicyholdersInsuredLoss Control Data

Claims Data

Losses and FrequencyTiming / PatternsJurisdictionClaimant informationInjury/DiagnosisTreatment patternsSettlement dataClaims NotesMedical Billing DataLegal Bill Data

Agency Information

RetentionRecruitingProfitabilityAdjusted Premium RatioNew Business VolumeContinuing Education

Weather

Heat / Cold ExtremesPrecipitation ExtremesHailWind / StormsEvent Extremes

3rd Party DatabasesBusiness CreditPersonal CreditCLUE / MVR / ISO CIBCheck CashingSub-Prime LendingCredit BureausReal EstateGeographic / GeocodeDemographicPsychographicBureau Data SourcesConsumer / LifestyleMedical and PharmacyBehavioralLitigation

Marketing / Sales

Campaign, PromotionCust Response ScoresCust Segmentation

Coverage Information

Product Coverage Options

Billing Data

Billing / Payment HistAccepted ApplicationsRejected Applications

Page 12: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Rel

ativ

e cl

aim

sev

erity

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

100%

< 25 25-30 30-35 35-40 40-45 45-50 50-55 55-60 60-65 65+

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

< 1 1 to 3 3 to 5 5 to 7 7 to 10 10 to 15 15 to 20 20 to 25 25 to 30 30+

Rel

ativ

e cl

aim

sev

erity

Claimant Age

Distance: Claimant Home and Employer

Insights can be revealed through both traditional and non-traditional risk characteristics.

Examples of Internal Predictive Variables

Page 13: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Examples of External Predictive Variables

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

Low HighLo

ss a

nd E

xpen

seR

elat

ivity

Percentage of Sports Ultility Owners

-30%

-20%

-10%

0%

10%

20%

30%

Low High

Loss

and

Exp

ense

Rel

ativ

ity

Percentage of Population with High School or Less Education

Public domain data on the financial condition of the parties involved in the claim can provide new insights into loss and expense severities.

External public database variables provide new insights. Even variables based on the claimant’s address have proven predictive.

The populations with lower education levels were over 20% higher in terms of loss and expense severity.

Page 14: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Representation of a Claims Multivariate Model

John Smith1 Circle Ave.Anytown, NY

92 Reason Messages:• Multiple co-morbidities• Claim history• Employment characteristics• Distance from work

Several hundred internal and external variables are tested to identify the 50 -

100 with greatest predictive power

w1(Claimant Age) + w2(Dist_H_W)+w3(Emerg_ Rm) + w4(Occupation) +

w5(CoMorbidity) + w6(Report_lag) +….

Sample Model Equation

Claim Complexity

Low High

Claim Segmentation Curve

Out

com

es

Model Inputs

Model Outputs

Predictive modeling combines and converts available internal and external claim characteristics into a score with corresponding reason messages. In workers’ compensation, output may also be “normalized” by injury group to better understand high severity claims relative to those with similar diagnoses.

Page 15: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Social Media Data

Every claim investigation typically starts with a visit to social networking websites such as Facebook and Twitter to assess the validity of the claim.

Page 16: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Analytics on the Cloud

Cloud computing is used by top insurance carriers to manager their

claims better and faster.

Page 17: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Lifestyle Based Analytics

Page 18: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Effic

ienc

y

Business valueCustomized analysis

Marketing and Sales•Movement beyond traditional “likely to buy” models

•Improvement in morbidity by selling only to best risks

Customer Retention•Identify compounding components of at risk customers

•Develop, deploy data driven intervention strategies

Data aggregation

and data cleansing

Predictive Analytics

Evaluate and

create variables

Develop

predictive

models

Score individual

profilesNon-traditional data sources unlock

new insights into employee populations

Traditional internal

data sources

Non-traditional external

individual or household level

data sources

Co-morbidity

data

Benchmark data

EASI census

Household data

Consumer data

Historical Claims

Customer data

Applicant data

Lifestyle Based Analytics (LBA)

Traditional data is augmented with non-traditional data to create stronger correlations to the target

Innovative data sources

Lifestyle base analytics can be used to add efficiency across critical business areas

Pricing•More accurately price products in situations where you have no or limited medical experience

Marketing and Sales•Movement beyond traditional “likely to buy” models

•Improvement in morbidity by selling only to best risks

Customer Retention•Identify compounding components of at risk customers

•Develop, deploy data driven intervention strategies

Medical Management / Wellness•Improved targeting of health events within a population; based on predicting propensity of having a certain clinical condition

•Deeper understanding of the current & potential risks of the customers

•Understand the behaviors creating the risks and monitor and develop behavior related strategies to change customers risks

Pricing•More accurately price products in situations where you have no or limited medical experience

Page 19: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

LBA provides Better Answers To Difficult Questions

Lifestyle Based Analytics (LBA) can be used to better understand the member and prospect populations

Managing Health Risk:• Which members will likely be afflicted with

a specific disease?

Health Plans, using a new generation of lifestyle-based analytical models, may be able to predict the likelihood of significant life events with more accuracy than ever before, and it starts with

something as simple as a name and an associated address

Retention:• Which members of a relatively unknown

population should we target for retention?

Efficiency:• Which members are most likely to

comply with health engagement programs?

• Which members have a higher probability of having positive outcomes from medical management programs?

• Which groups would it make sense to offer wellness initiatives to?

Acquisition:• Which consumers are most likely to buy?

• Who are the best candidates to target with a specific product?

Future Health Risks:• What are the future health risks for

members with unknown claims data?

Page 20: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

The black arrow points to a random distribution. In this case, 20% of the people will have 20% of the future cancer claims.

The red arrow points to traditional underwritings ability to predict cancer claims in this healthy population. In this case, 20% of the highest risk members accounted for 30% of the future cancer claims.

The blue arrow points to LBA’s ability to predict future cancer claims in this same population. In this case, 20% of LBA’s highest risk members accounted for over 60% of the future cancer claims.

Examples of lifestyle-based diseases include: diabetes, cardiovascular, cancer, and respiratory.

This chart demonstrates LBA’s ability to identify future cancer claims in a healthy female

population.

Lifestyle-based analytics (“LBA”) focuses on identifying increased morbidity and mortality risks for “lifestyle” based diseases.According to the US Surgeon General, lifestyle based diseases account for over 70% of US of healthcare expenses and subsequent deaths.

Lorenz Curve for Neoplasm Female Sample

LBA and Improved Morbidity Risk Evaluation

Page 21: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

-80%

-60%

-40%

-20%

0%

20%

40%

60%

80%

1 2 3 4 5 6 7 8 9 10

Actu

al C

laim

s R

elat

ivity

Predicted Claims Decile

Algorithm was constructed using a 40/30/30 train/test/validate methodology

Lift above demonstrated on blind validation after final algorithm was chosen

Age/Gender correction made (neutral)

Individual variables:‒ Selected disease state algorithms (both binary and cost-weighted)‒ Selected 3rd party ailment indicators‒ Selected individual characteristics

Average Claims Relativity

Observations

Members with the worst algorithm scores

experienced actual claims 60% higher

than average

Result for Claims Cost

Page 22: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Data Visualization

Page 23: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Front End Tools

We have the ability to display modeling results in graphic, front-end tools that allow users to select different dimensions for additional analyses. The exhibit below depicts member risk levels for Cardiovascular Disease for a sample of individuals in the greater New Jersey area.

MEMBERS

Page 24: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Front End Tools (continued)

The exhibit below shows the trend of policies and premiums across 10 buckets grouped by high to low loss ratio for Auto insurance renewal business.

Page 25: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

So many have already done it…

Page 26: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

What Underwriters & Claims Executives are Saying

Science, enabled by technology, now plays an integral role in our value proposition. Pricing risks

and establishing optimal claim outcomes for our Insureds are being aided by sophisticated analytics

such as predictive modeling

Our claim scoring models review new claim notices daily to identify red flags and suspicious claims for investigation.

… better outcomes, through enhanced automation from first notice of loss to

claims resolution

…first Workers’ Compensation Company to apply advanced analytics to

claims

Page 27: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

• Insurance Company of the West (ICW)• Companion Property & Casualty• ACE Ltd.• Acuity• XL• Westfield Group• Grange Mutual• Louisiana Workers Compensation Corp• Fireman’s Fund• California State Fund• Secura• Allstate• State Farm• QBE• Farmers

• Utica National• WR Berkley• Sentry• NJM• Auto Owners• Main Street America• RLI• Unitrin• AFICA• Plymouth Rock• Beazley Group• American Family• Meadowbrook• Nationwide• Church Mutual

Who’s Attending Predictive Modeling Seminars?

Page 28: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

The savings potential

Page 29: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Right claim, right resource

Improve routing to auto-adjudication

Increase triage consistency through automation

Claim Routing & Assignment

Reduce lag time of SIU referrals

Improve mix of claims referred to SIU

Deterrence of “soft-fraud”

Fraud Detection

Prompt assignment of nurses on those cases that need it most

Integrate behavior issues into nurse assignment

Cost effective use of field case management

Medical Management

Demonstrated ability to close claims faster and cheaper leads to competitive market advantage

Improved client satisfaction strengthens the relationship and brand

Top Line Growth

Projected Business Impact

4-8% reduction in loss and expense

5-10% improvement in SIU managed claims

3-7% improvement in nurse managed claims

20-25% redeploymentof supervisory resources

Deloitte successfully designed and implemented Workers’ Compensation claim severity predictive model into multiple clients’ claims operations to help injured workers return to work sooner.

Benefits Realized

Page 30: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Benefits Case – Calculation & Allocation Tool

Once broad benefit target areas, amounts and associated metrics are defined, we use our Benefits Calculation Tool to provide a highly tailored and approach/framework to refine, allocate and aggregate benefits.

Illustrative Benefits Calculations

Page 31: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Questions?

Page 32: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Questions

31

Page 33: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

Copyright © 2014 Deloitte Development LLC. All rights reserved.

Biography

Geet specializes in the development and application of predictive analytics and business intelligence for the financial services and insurance industries. With a background in mathematics, Geet has worked with many Fortune 500 companies to leverage data analytics and technology to contain costs and gain operational efficiencies. She has lead various analytics teams through the end-to-end process of model design, build, and implementation, and co-develop Deloitte’s Advanced Analytics solutions for Healthcare Insurance. She has publications in Claims P&C magazine, Claims 360 degree magazine and through IIMA Analytics conference

Geetanjali ChakrabortyEmail : [email protected]

Tel (US) : +1 617 437 2393

32

Page 34: Geetanjali Chakraborty Advanced Analytics & Predictive ... · Cool Analytics for the Insurance Industry. IASA Conference November 21, 2014. Geetanjali Chakraborty Advanced Analytics

© 2009 Deloitte Development LLC

End of Presentation

33