Insurance PoliciesSales

2

Project Background, cont.– Financing system

• Company sells financing and insurance to home owners through independent dealers

Financial Services

CompanyDealers Customer

Loan/insurance Loan/insurance

3

Six Sigma Project– Company decided to initiate a Six Sigma project

– Two issues for consideration:

• Improve the accuracy rate—the rate at which the policy and loan package accurately represent what the customer specified—primary focus

• Increase the number of policies sold - some loans processed without attempting to sell accompanying insurance policy—secondary focus— potential future project

4

Charter– Purpose

• Problem/Opportunity Statement: The division produces 17% of all insurance policies with inaccuracies as compared to customer specifications.

• Goal Statement: Improve set up accuracy rate from current rate of 83% to 99%.

– Importance

• Business Case: Improving the accuracy will:

– Increase customer satisfaction by meeting specifications—this will translate into increased revenue because of more policies sold and more referral business

– Increase efficiency by eliminating rework and non-value added activities from the process—this will translate into decreased costs

• Total estimated annualized project impact = $525,000

5

Map the ProcessSIPOC Model

Loan process

Insurancesold

Policy info to Loan

Processing

Documentrequest

Documentsprepared

Loan closed/policy set up

Suppliers

Customer/Dealer-Agent

Dealer-Agent

Financial ServicesLoan Processor

Inputs

Applicationinformation

Insuranceapplication

Loan/insurancepackage documents

Process Outputs

Loan proceeds

Insurancepolicy

Insurance commission

Revenue

Customers

Customer

Customer

Dealer-Agent

Financial ServicesLoan/InsuranceDivision

6

Customer Voices

I want my insurance coverageto include natural disaster and flood insurance

I want to be able to finance my insurance with my loan

I don’t want my down payment to increase becauseI am financing insurance

Reworking Policies is eating up time

Errors in policies and closing documents are creating Unhappy customers

I never agreed to finance insurance, and it shows up on my loan agreements

The insurance raised my payment by more thanI was expecting

My down payment went up and I was not expecting that

I did not realize my payment would go up by $35

The closing package was already printed whenI noticed the increased payment

I want the insurance on my loan to reflect whatthe agent sold me

I got a better deal from another companyThe insurance raised the payment too much

My down payment went up and I was never told

I can get a better deal somewhere else

7

Customer Requirements KJWhat are the Key Customer Requirements for purchasing insurance?

My insurance coverage includes natural disaster and flood insurance only when required

I am able to finance my insurance in the manner I desire

The coverage I receive is the coverage I need

I want to know that this is the best deal I can get with this company

This company offers the best value compared to all other options

This company provides the service with minimum rework

I really believe that price incurred is worth the value I receive

The Paperwork accurately reflects what was promised and sold

I don’t have to worry about changes that are

wrong or sudden

I am comfortable with my understanding of

the product , cost, and implications on

accepting insurance

Errors in terms and payments are noted before the closing package is to be signed

Changes in terms and payments are identified and agreed to ahead of changes

being implemented

Insurance raises are tied to an attributable reason and notice is given in advance

The insurance on the loan reflects what my agent sold me

My payments are reflected in the manner that I agreed to

Initial down payment expected matches what the company

communicated

My expectations for payment and potential changes match

what the company actually does

My understanding of insurance options and coverage is complete

My understanding of insurance implications is

complete

The customer feels that they understand and are comfortable with the insurance and believe the value created by our company is real

8

Critical To Quality Metrics

Customer Requirements Quality Metrics

Paperwork Accurately reflectsWhat was sold

Coverage matches expectation

Financing matches expectation

Deductibles match expectation

Customer understands policy options

Dealer answers questions in first meeting

Coverage receivedCoverage matches expectation

Best value Mode of financing matches expectationPercent Completion right first time

9

Project Components– Data Collection Plan

– Process Flow Diagrams

– Baseline Data on Defects

– Stratified Data Charts

– Calculate Process Sigma

10

Key Questions– What does the current detailed process look like?

– Does the defect rate show trends over time?

– Does the defect rate depend on the day of the month?

– What are the types of defects?

– Do any regions (clusters of dealers) account for the majority of defects? (80/20 Pareto principle)

– Do any agents account for the majority of defects?

11

Process Flow

Loan process

Insurancesold

Policy info to Loan

Processing

Documentrequest

Documentsprepared

Loan closed/policy set up

SIPOC Model

Insurance SoldThrough Dealer

Potential “missed

policy” if not caught

prior to loan signing

Potential point where “inaccurate policy” could occur

Potential points where

“inaccurate policy” could

occur

Complete application

Customer Dealer-Agent

FS Loan Processor

Sends application and docs to processor

Checks agent remarks in

system

Insurance sold by agent?

Documents prepared with

insurance

Loan documents

signed

Loan funded w/o insurance

Change in collateral since

application

?Supplies new numbers

Accept new

terms?Update screens

Yes

No

Yes

Yes

Loan funded w/ insurance

Loan/insurance documents

signed

Documents prepared without

insurance

No

Loan Process

FS Agent

Contact Dealer-Agent

No

12

Process Flow

Loan process

Insurancesold

Policy info to Loan

Processing

Documentrequest

Documentsprepared

Loan closed/policy set up

SIPOC Model

Changes madeafter the sale

Potential impact of “inaccurate policy”

Potential point where “inaccurate policy” could occur

FS Agent CustomerDealer-Agent

Contacts customer

Updates screens

Calls dealer

Calls FS/Agent

Calls FS/Agent

Customer wants to change

insurance

Change terms/coverage

and requote

Accepts new

insurance

?

Customer wants to cancel

insurance

Insurance sold

Updates screens

Insurance not sold

No

Yes

No

Yes

Customer wants to cancel

insurance

Yes

No

Changes After Sale

Customer wants to change

insurance

13

Others

Repo-sold as 5 years

No agencyremarks

Licensed Dealer

Customer changed term

Notes different

Agencychanged

Not on loan assold

81111171719221523.24.34.36.56.57.58.659.1

100.096.892.588.281.775.367.759.1

2702402101801501209060300

100

80

60

40

20

0

DefectCount

PercentCum %

Perc

ent

Coun

t

INACCURATE POLICIESPARETO CHART OF DEFECTS ON

Data collected from May 2001

Types of Defects?– 60% of “Inaccurate Policies” due to the paperwork not

reflecting what was initially sold to the customer

14

Others

Repo-soldas5years

Noagencyrema

rks

LicensedDeale

r

Customercha

ngedterm

Notesdifferent

Agencychang

ed

Not onloanasso

ld

81111171719221523.24.34.36.56.57.58.659.1

100.096.892.588.281.775.367.759.1

2702402101801501209060300

100

80

60

40

20

0

Defect

CountPercentCum %

Per

cent

Cou

nt

INACCURATE POLICIES

PARETO CHART OF DEFECTS ON

Data collected from May 2001

Types of Defects?

F i n an c i ng

D educ ta b le s

Cove rage

Ot he r

105 31 10 669.1 20.4 6.6 3.969.1 89.5 96.1 100.0

0

50

100

0

20

40

60

80

100

Defect

CountPercentCum %

Per

cent

Cou

nt

PARETO CHART OF DEFECTS “NOT ON LOAN AS SOLD”

Data collected from 05/01/01 to 05/31/01

150

– 69% of “not on loan as sold” due to errors in “financing terms”

15

Difference by Region?– Pareto Principle does not hold…apparently no significant

different between regions

Others43403863259078239787747360218072527550302827209557537616

1233344555666666888910101112121518181920

51112222222222233334445567778

1009594939290878583817977757371696663605753494540352922158

250

200

150

100

50

0

100

80

60

40

20

0

Region#Count

PercentCum %

Perc

ent

Coun

t

Pareto Chart of Inaccurate Policies by Region

Data from May 2001

16

Customer Impression– Telephone survey to a random sample of 150 of the 1775

total loans written in May, 2001

– Survey questions:

• “Was the agent able to provide information and answer all of your questions in the first meeting?”

• “If no, approximately how many times did you need to contact the agent in order to resolve any questions you had?”

• “If you could make the loan application and approval process better, what would that mean to you?”

17

Customer Impression, cont.

Telephone Survey Question #1 - Did Dealer-Agent provide information/answers in first meeting?

0

10

20

30

40

50

60

70

80

90

Yes No No response

Category of reponse

Num

ber o

f res

pons

es

Telephone Survey Question #2 - How many times did you have to contact the Dealer-Agent to resolve your questions?

0

5

10

15

20

25

2 3 4 or more

Number of contacts made

Num

ber o

f res

pons

es

– 31 out of 114 responses (27%) could not get the information at the first meeting

– Of those 31, 23 resolved the questions during the second contact

18

Customer Impression, cont.

Telephone survey Question #3 - How to improve the process?

0

10

20

30

40

50

60

70

better terms faster process more knowledgeableagent

more self-service

Category

Num

ber o

f res

pons

es

Addressed bythis project

Future project

Outside of scope

19

Difference by Agent?– No significant difference agent to agent

– Every agent, overall, seems to produce defects at a rate of 15% of total written

0.3

0.2

0.1

0.0

Pro

porti

on

31771

31571

31511

31401

3131

531

314

31145

3111

431

080

30941

3085

3

30243

30035

3002

2

22992

22118

21469

6105

1966Agent ID

over Total Policies Written by AgentP Chart for Proportion of Policies with Errors

P=0.1454

UCL=0.2643

LCL=0.02639

20

Missed Policies

Missed in Process

Customer c

ancelled

No remarks

Dealer cancelled

122 78 64 1843.3 27.7 22.7 6.4 43.3 70.9 93.6 100.0

0

100

200

0

20

40

60

80

100

Defect

CountPercentCum %

Per

cent

Cou

nt

PARETO CHART OF DEFECTS ON MISSED POLICIES

Data collected from 03/01/01 to 05/23/01

– 70% of “Missed Policies” missed by dealer in the loan process or cancelled after the sale

21

Tree Diagram

Inaccurate financing

terms set up on loans

sold

Information entered

incorrectly by Loan

Processor

Dealer explained incorrectly

Application filled out

incorrectly

Customer change during

the process not entered

into the system

Information passed

between FS Agent and

Dealer inaccurate

FS Agent explained incorrectly

Not aware of program

Not aware of updated financing terms

Directions unclear

Dealer not aware of program

Procedures for updating screens do not exist

Too many applications in process at the same

time

Agent remark screen not filled out

Procedures for updating screens do not exist

Too many applications in process at the same

time

Procedures for updating screens do not exist

Different knowledge base for FS and Dealer

Agent

No training provided

No standard communication of

updates

High turnover of Dealer-Agents

Long cycle time to process

Procedures for updating screens do not exist

No training provided

Long cycle time to process

Unclear who should train the Dealer Agents

FS Agents viewed as experts

Potential Causes– Tree

Diagram constructed to investigate…why financing terms inaccurate?

Suspected Causes

22

– The relationship between cycle time and defects was also analyzed using a control chart

Verification of Causes, cont.

50403020100

100

50

0

Observation Number

Indi

vidu

al V

alue

I Chart - Number of days from sale to funding

1

Mean=36.88

UCL=86.53

LCL=-12.77

Non-Defect Defect

UCL=38.08

Mean=13.48

LCL=-11.12

23

Redesigned Process• Reduced cycle time

– If customer not offered insurance initially, FS Loan Processor calls Dealer direct instead of calling FS Agent

– FS Loan Processors switched to a mind-set of processing an application when it came in instead of batching applications to process together

• Reduced cycle time = less applications in process at one time = fewer defects

• Reduced cycle time = less time during which changes can occur = fewer defects

Complete application

CustomerDealer/Agent

FS Processor

Sends application and docs to processor

Checks agent remarks in

system

Customer offered

insurance

?

Documents prepared with

insurance

Loan documents

signed

Loan funded without insurance

Update screens

Loan funded with insurance

Loan/insurancedocuments

signed

Calls Dealer

Loan Process

Call customer

Yes

Customer buys

insurance

?

No

Agent attempts to contact customer

Agent updates screens Yes

No

24

Failure Modes & Effects Analysis– Recommended Actions

• Small group training for dealers

• Ongoing training and communication for dealers

• Ongoing communication for Regional Managers

• Periodic memos on updating screens

• Auto reminder for Processors

FMEA Team: Dealer Reps and Loan Processors (Revised):

FMEA PROCESS RESULTS ACTION

Item / Process

Step

Potential Failure Mode

Potential Effects of

Failure Seve

rity Potential

Causes of Failure

Occ

urre

nce

Current Controls

Det

ectio

n

RPN

Recom-mended Actions

Respon-sibility and

Comple-tion Target

Date

Action Taken

Seve

rity

Occ

urre

nce

Det

ectio

n

RPN

Sends applic-ation to LP

WrongInfo 10

Training not Successful andnot continuous

10Doc's Out

Report 1 100Smaller

Staff and Training

RegionalManagers/Dealer

Smaller Team

Completed 8/01/01

10 5 1 50

Sends applic-ation to LP

5 NewDealer 8 Doc's Out

Report1 40 On Going

TrainingRegionalManagers

On Going Training

5 4 1 20

Document Processing

Processor MakesError

Low customersatisfaction/Lost revenue

5No screen

Updateprocedures

2Doc's Out

Report 1 10Periodicmemos onscreen updates

LoanSupervisor/

Procedure Change 5 2 1 10

ContactsCustomer

Processordoes not follow up onmissed call

5 No followup procedure

6Doc's Out

Report 1 30 LoanSupervisor

Procedure Change 5 2 1 10

Low customersatisfaction/Lost revenue

Low customersatisfaction/Lost revenue

Auto reminderto Processor

Wronginfo Low customer

satisfaction/Lost revenue

25

Results– May Defect Rate = 15%

– June Defect Rate = 10%

– July Defect Rate = 9%

– August Defect Rate = 6%

0.3

0.2

0.1

0.0

Pro

porti

on

8/31

/200

18/

30/2

001

8/29

/200

18/

28/2

001

8/27

/200

18/

24/2

001

8/23

/200

18/

22/2

001

8/21

/200

18/

20/2

001

8/17

/200

18/

16/2

001

8/15

/200

18/

14/2

001

8/13

/200

18/

10/2

001

8/9/

2001

8/8/

2001

8/7/

2001

8/6/

2001

8/3/

2001

8/2/

2001

8/1/

2001

5/31

/200

15/

30/2

001

5/29

/200

15/

28/2

001

5/25

/200

15/

24/2

001

5/23

/200

15/

22/2

001

5/21

/200

15/

18/2

001

5/17

/200

15/

16/2

001

5/15

/200

15/

14/2

001

5/11

/200

15/

10/2

001

5/9/

2001

5/8/

2001

5/7/

2001

5/4/

2001

5/3/

2001

5/2/

2001

5/1/

2001Date

Total Policies Written by DayP Chart for Proportion of Policies with Errors over

P=0.05995

UCL=0.1439

LCL=0

Before After

26

Updated Financials

– Savings from project

• Increased customer satisfaction resulting in increased revenue

• Decreased rework and increased efficiency

– Total savings

• Dollar impact

= $478,000

= $160,300

= $638,300