Smart Underwriting: Are You Ready for the Next Wave

1
Smart UnderwritingAre You Ready for the Next
Wave?
2
Agenda

• Smart Underwriting An Industry View
• What is Smart Underwriting
• Evolution
• Benefits
• Risks
• Factors for success
• Smart Underwriting Case Study

3
Smart Underwriting An Industry View
4
Speaker Lindsay Fassett

• Product Strategy, Oracle
• Seven years rating / underwriting technology
  experience

5
What is Underwriting?

• The risk qualification process
• This logic could include
• Knockout questions
• Tier qualification
• Policy approval, denial, clarification

6
What is Smart Underwriting?

• Automating much of the underwriting process with
  the goal of reducing human touch to only those
  policies that are an exception from the norm.
• Also referred to as exception-based underwriting

7
What is required for Smart Underwriting?

• In-depth look into current underwriting process
  and determining which underwriting criteria can
  be quantified and automated.
• Needs a thorough understanding of both
• Underwriting rules
• System and processes

8
Smart Underwriting and Information Requests

• Information request automation
• Automate 3rd party requests
• Should rules be built around when to make these
  call?
• Gather more granular information in order to make
  informed underwriting decisions

9
Role of Analytics

• How do analytics impact the UW process?
• Constantly update underwriting rules based on
  analytical data.
• React to industry changes, weather changes, etc.
• The faster you can progress around this circle,
  the more responsive you are to industry changes.
  
• Systems/automation speed this cycle

10
Progression of Smart Underwriting

• Movement of Smart Underwriting from personal to
  commercial lines
• Web-based, consumer-focused rating applications
  require real-time underwriting
• Increase in available information to use in the
  UW process
• Example Google Earth
• Rapidly changing risk information

11
Measurable Results

• Metrics (Examples from actual insurers)
• Reduced cycle time for a commercial package
  policy from two weeks to 24 hours
• Small commercial business insurers have increased
  avg premium per underwriter by 100 (from 5M to
  7.5-10M)
• 50 reduction in leakage (improvement in loss
  ratio)
• source Automating Commercial Underwriting,
  Celent, Oct. 2007

12
Benefits of moving to Smart UW

• Improved underwriting consistency
• More resources available to spend on
  exceptional policies
• Policies are bound more quickly
• Decrease cost

13
Risks of not adopting Smart UW

• Inconsistent risk acceptance
• Resource intensive
• Lengthy time between quote and bind

14
Factors for Success

• Willingness to innovate
• Think creatively
• What is ideal?
• Define what is the end goal
• Plan practically

15
Factors for Success

• Right people at the right time
• Formalized project
• Dedicated and defined project team
• Funding and management buy-in

16
Factors for Success

• Right tools
• Flexible underwriting system
• Who is the user?
• How frequently can those users make changes/react
  to market?
• Predictive analytics
• BI tool that allows you to see those changes and
  act upon them
• Metrics
• How to judge success if nothing is measured?

17
Smart Underwriting A Case Study
18
Speaker Dean Kimball

• Project Manager, Penn National Insurance
• 13 years experience
• Claims underwriting systems enhancements
  projects

19
Case Study Agenda

• Company Background
• History/Issues
• Objectives/Approach
• Project Overview/Solution
• Project Pieces
• Lessons Learned
• Project Results
• Future Outlook

20
About Penn National Insurance

• PC Mutual Carrier
• Lines of Business
• States
• Written Premium

21
History Issues

• Wanted New Entry Portal
• Consolidate Dual Rating Engines
• Integrate with Policy Admin System
• Business Rules
• Desire to Write More Business
• Current Rating vs. New Rating

22
Prior Environment

• Vendor Front-End Entry System
• Running Dual Rating Engines
• Policy Admin System Entry
• Developer Rating Work
• Limited Tiers and Rating Ability

23
Buy-in

• Sell the project to business management
• Sell the project to IT management
• Sell the project to Executives

24
Project Overview

• GOAL 1
• Develop and implement systems that will allow for
  increased net written premium and profitability
  through
• Enhanced price segmentation
• Improved ease of business with agency systems
• Greater operational efficiency
• GOAL 2
• Develop and implement systems that will allow for
  straight through processing for new business and
  policy changes.

25
Project Overall Objectives

• User-friendly
• Reuse existing technology when possible
• Component-based (SOA)
• Rules-based
• Enterprise Best of Breed Tools
• Scalable

26
Project System Objectives

• Automate Upload of New Business Policy Changes
• Automate Many Tasks with Pre-Renewal Processing
• Allow Processing of New Business/Policy Changes
  from Agency Systems Comparative Raters into
  Agency Portal
• Create New Entry Portal
• Obtain Implement a Single Rating Engine
• Obtain Implement a Business Rules Engine
• Interface with Policy Admin System
• Implement SOA Approach
• Implement Predictive Analytic Rates to Write
  More Business

27
Project Approach

• Dedicated project team
• Separate project pieces
• Executive support
• Business IT management support
• Clearly defined plan and goals

28
Penn National Insurance Project Diagram
29
Project components

• Agency Portal
• Agency Upload
• Business Rules
• Underwriting Workflow
• Rating Engine
• Legacy Integration
• Enterprise Data Warehouse Integration
• Predictive Analytics Scorecard

30
Hardware / Software

• SQL 2005 Platform
• BOM
• BizTalk
• C
• Oracle Insbridge
• Exstream
• Pega
• ImageRight
• Web Services

31
Agency Portal

• PennPlus
• Web application
• Designed and developed by Penn National Insurance
  
• Allows quoting and submission of personal lines
  policies.
• Strategic Differentiator
• Interface with agents was too important to trust
  to an external vendor.
• Integration
• Bridges from the agency management systems via
  Transformation Station or TransactNOW
• Bridges from comparative rating vendors
  SilverPlume and EZLynx

32
Agency Portal Smart Underwriting

• Enhanced ease-of-doing business capabilities
• Automation of previous manual tasks
• Multiple screen edits
• Field edits for data choices and data quality
• Get it right the first time

33
Agency Portal Benefits

• Automatic ordering and integration of IBS, MVR,
  and CLUE
• IBS report is ordered immediately upon entry of
  the client
• User has the ability to dispute losses/violations
  without first submitting for approval.
• The entry/submission system rating engine will be
  one and the same,
• Will not have to maintain separate rates
• No manually rated coverages in the Agency Portal
• Agent does not have to have a rating manual
  available to do a quote
• A proposal is available in Agency Portal that can
  be printed and presented to the client
• Billing and installment information is calculated
  and displayed within the quote and on the proposal

34
Penn National Insurance
35
Rating Engine

• Oracle Insbridge Rating Underwriting
• Single rating engine
• RateManager app used to build the rates
• SoftRater engine that runs rating requests
  against the built rates
• Replaced both the Rackley PMS rating engines
• All lines of business and all states
• Rating (tables algorithms), stats, forms and
  driver assignment
• Multiple sets of rates for each state/line of
  business combination
• Effective and expiration dates for each set of
  rates (new business renewals)
• Rating new business and policy changes entered
  into Portal.
• Rating new business, renewals and policy changes
  entered into or upload to PAS

36
Smart Underwriting Rating Engine

• Business analysts can make and test rate changes
  on their own
• Tools available to test an entire book of
  business before a rate change is implemented
• Same rating engine called from all places

37
Rating Engine Benefits

• One rating engine for front-end and back-end
  rating
• No dual maintenance needed
• Single rating outcome
• We control all parts of rating engine
• Same premium calculations used
• Rate changes always in sync
• Quicker turnaround for rate changes

38
Penn National Insurance
Rating is called from multiple components.
39
Business Rules Engine

• PegaSystems Rules Engine
• Rule Applications
• Underwriting rules (Prohibited and Referral)
• Report Ordering Decisions
• Scorecard and Tiering
• ChoicePoint data mapping
• Systems using Rule Applications
• WebRating (Scorecard only)
• Agency Portal
• Pre-Renewals
• Standalone Scorecard
• Policy Admin System Nightly Cycle
• Business Impact
• Allows PNI to automate underwriting decisions
• Allows PNI to make changes to those rules without
  causing system changes

40
Smart Underwriting Rules Engine

• Business rules to prohibit and refer new business
  policy changes
• Users receive immediate feedback if rules are
  encountered
• Get it right the first time

41
Business Rules Engine Benefits

• Separates business rules from the applications
  that use them allowing maintenance of business
  logic without modifying software
• Increases flexibility and reduces time-to-market
• Rules are all in one place
• Promotes visibility and understanding of business
  policies and procedures
• Promotes consistent decision making
• Forces order to the rules and policies that
  govern business

42
Document Management/Workflow

• ImageRight
• A document management and workflow product
• Used to store all policy related documents in a
  logical order
• Used to assign and route all work tasks to
  internal staff
• Used to document the underwriters disposition
• Used to set follow-ups (diaries)

43
Smart Underwriting Document Automation

• All documents in one place to make approval
  decisions
• Approval decision is a few clicks
• Automated upload fed from ImageRight
• Get it right the first time

44
Document Management Benefits

• Improved customer service
• Expedite the decision making process with all
  data readily available in one system
• Improved work management tool
• Underwriting workstation concept -- one system
  vs. many to manage work items (diaries,
  documentation, tasks, etc.)

45
Legacy Interface Apps

• BizTalk
• First app takes entry from Agency Portal,
  converts to policy admin file format, and places
  into nightly cycle to hit the policy admin
  system.
• Second app takes uploaded entry on pending
  section of policy admin system, calls VIN, rates
  it, runs error checks and puts it on the
  permanent section of policy admin system.

46
Smart Underwriting Legacy Interface Apps

• Automated upload of entry from the agency portal
• Eliminates re-entry of new business policy
  change by PNI staff
• Policies from different front-end systems all
  sent to same policy admin system
• Agency Portal and Legacy Interface Apps both
  using same Rating and VIN services

47
Lessons Learned Design

• Work collaboratively with business to define
  business services require sign-off
• Involve IT architects from all sub-projects in
  the design process require sign-off
• Include sponsors in all sign-offs.
• Introduce new technologies only as necessary use
  experienced consultants in these areas
• Allow enough time we used 50 time more than
  usual (for design)
• Finish design before starting development.

48
Lessons Learned Development

• Do plenty of unit testing.
• Make the best possible architecture designs, as
  long as time allows.
• Assign architects to conduct code reviews.
• Have the right tools to test components
  separately.
• Require change control documentation and
  approval.
• Discuss and plan for version control

49
Lessons Learned Testing Implementation

• Have a defined Integration Testing period.
• Create a test matrix and tests cases for
  everything.
• Make sure that the test environment (hardware)
  matches the production environment as closely as
  possible.
• Begin performance and stress tests early.
• Conduct production parallel tests.

50
Project Results

• Increased profitability due to improved pricing
  precision
• Increased premium writing due to ease of use and
  pricing points
• Reduced processing expenses (less human
  intervention)
• Reduced report ordering expenses
• Improved time to market with price and product
  changes
• Reduced system maintenance expenses (ie one
  rating engine)
• Improved relationships
• Commitment to customers
• Competitive advantages

51
Measurable Results

• 80 straight through processing.
• Quotes up 90 in September, from previous year.
• Written Premium up 80 in September, from
  previous year.
• Policy changes into Policy Admin System down from
  7 days to 1 day.
• Automate report ordering process for renewals.
• Reduce number of renewals where Underwriter
  review is needed.

52
Future Outlook

• Implement the Same Project Pieces for Homeowners
• Review the Possibility of Using Insbridge for
  Other LOBs

53
Key Takeaways for Smart Underwriting Project

• Review current business processes
• Identify areas for automation
• Create rules that minimize human
  interaction/error
• Identify where duplicate entry can be eliminated
• Consolidate and simplify core systems
• Centralize all documentation

54
Conclusion

• Questions
• Feedback