1. Tinton Falls, NJ Charlotte, NC www.predictivemetrics.com
732-530-9303
2. About PredictiveMetrics (PMI) Turning Probabilities Into
Profits
Established to provide higher quality analytics and predictive
decision scoring in a customer focused environment.
Analytical staff is comprised of Ph.D. and master level
econometricians and statisticians.
Experts in developing and implementing statistical-based
predictive decision systems for collections, recovery and debt
buying/selling.
Have proprietary software and state-of-the art hardware in a
SAS 70 II environment that is designed specifically for production
of sophisticated analytical applications.
Innovators of utilizing accounts receivable and placement data,
which is proven to be the most predictive data and its FREE.
Successfully pioneered and continuously develop more efficient
and effective predictive analytical decision solutions.
Celebrating 14 years of predictive scoring excellence
3. Manufacturers Integrity Portfolio Funding LP Investments
Wholesale / Dist Debt Buyers Collection Agencies Utilities
Transportation Financial Services Medical MOUNT CARMEL HEALTH
SYSTEMS Southern Credit Recovery PMIs Customers - A Tribute to Our
Success
4. Benefits of Using PredictiveMetrics A Consultative Approach:
Your business expertise combined with leveraging information,
technology, and statistics, PredictiveMetrics provides you with
superior predictive decision solutions. Success Drivers Data and
Modeling Expertise We know statistics and how to apply data from
any source, i.e., your data, bureau data, etc. providing you with
the most predictive, cost-productive solutions. Cutting Edge
Technology We utilize advanced hardware and software designed for
the most complex analytics. Innovators of more effective
statistical techniques. On Going Customer Service We are there
before, during, and after the analytical process begins. Your
models are well maintained and predictiveness kept intact through
on-going validation. Just Ask Our Customers They are reducing
costs, better utilizing resources, accurately targeting accounts
based on risk or liquidation, and maximizing profits. Consumer
Collection Customers Achieving substantially higher collections by
knowing payers and dollars to be collected or for portfolio
management knowing who will likely go delinquent. Financial
Services, B2B Trade, Utilities, Leasing Customers Reducing
write-offs and delinquencies. Streamlining operations. Developing
strategies to control risk. Proven Results
5. Focus in Predictive Analytics for Collections A large
portion of PredictiveMetrics solutions are for collections
Data Sources
A/R Data
Placement Data
Collection Data
Socio-Economic Data
Demographic Data
D&B
Experian Commercial
InfoUSA
PayNet
Experian Consumer
Equifax
Trans Union
Financial Statement
6. PMIs Consumer Scoring Products
DebtBuyerScore SM - Quickly determines the price to bid on a
bad debt portfolio maximizing profits
CardCollectionScore SM - Identifies dollars and payers for
collecting credit and charge card bad debt
CollectionStrategyScore SM Identifies the probability that a
customer will go severely delinquent or to loss for more
cost-effective collection prioritization
LegalCollectionScore SM Suits are costly! Knowing who will pay
and how much are crucial for cost-effective decisions
MedicalCollectionScore SM - Liquidate more medical debt by
knowing who will pay and how much
PriorityScore for Collections SM - Suite of collection scoring
models designed for late-stage and post charge-off collections that
is co-branded with Experian and blends your account-level data and
Experian's credit data
UltraCollectionScore SM - Optimizes collections by identifying
payers and dollars
UtilityScore SM for Collections - Work the right accounts by
knowing which accounts are likely pay and which are likely to go
severely delinquent or to write-off
Specializing in providing industry/finance specific models
7. PMIs Commercial Scoring Products
Net30Score SM - B2B portfolio management tool that effectively
streamlines credit and collection decisions; Multiple versions to
best fit your needs
ScoreMiner SM - A web-based credit/collection scoring and data
mining application that leverages the predictive power of our
Net30Score, UtilityScore, LeaseRiskScore and custom portfolio model
output
UtilityScore SM for Collections - Work the right accounts by
knowing which accounts are likely pay and which are likely to go
90+ days delinquent or to write-off
AccessLink SM SAP NetWeaver Process Integration - Helps
customers easily extract the necessary data to take advantage of
utilizing PredictiveMetrics' industry- and finance-specific and
custom behavior scores for pro-active decisioning.
LeaseRiskScore SM - Collection prioritization model for B2B
leasing companies
Lease Decision Scores - Suite of decision tools based on lease
performance data from several major leasing companies combined with
Experian commercial and/or consumer data
Specializing in providing industry/finance specific models
8. There are numerous benefits of using internal performance
data Greater predictiveness Higher hit rate Segmenting accounts
Receive two scores Leverages the predictive power of your FREE,
easily accessible A/R or placement data = Yields higher returns
Uses clients A/R or placement data allowing creditors and
collectors to score thin or no file accounts = Decision more
accounts Receive a segmented score based on risk or type of debt,
and age of debt = Target the right accounts Receive traditional
payer score and a unique dollar score for bad debt = More
profitable collection prioritization Advantages of Utilizing A/R
and Placement Data
9.
All of PMIs models are empirically derived multivariate
statistical models
No bureau data or personally identifiable information is
required to produce accurate scores - leverages internal A/R or
placement data, which is proven to be the most predictive in this
type of model and its FREE
Several reports are available: Pre Charge-Off - Dollars at
Risk; Trend; Watch List, Distribution of Accounts by Aging and Risk
Class Post Charge-Off - Segmentation for Payers and Dollars;
Summary of the state of the portfolio of submitted accounts;
demographic breakdown; Quality of data reporting; Cost benefit and
expected profitability
No IT resources required to implement, file transfer through
secure encrypted FTP in SAS 70 II environment
Score all accounts to determine overall risk or collectability
and liquidation on an account and portfolio basis
Clients include, but are not limited to: Several divisions of
GE Capital, Florida Power & Light, 3M, Edward Don, Wright
Express, Mount Carmel Health, Southwest Credit, Penn Credit,
National Enterprise, Capio, Blue Tarp, and Boulder Credit
Focus in Predictive Analytics for Collections
10. Case Study: Pre Charge-Off Collection Scoring Recession
Period
Prior to working with PMI, customer was using internally
developed judgmental based scoring model that was more than 9 years
old and never validated
Based on 3 data variables: Pay weight Length of service Age of
arrears
Variables and weights were subjectively determined for the
scorecard
Customer decided to implement statistical behavior collection
models, developed by PredictiveMetrics and the results were:
A savings of $1.2 million in losses over 12 months
Reduction in collection costs by creating a process to allocate
resources more efficiently
Created a segmentation of their accounts into different risk
groups
Implemented targeted collection strategies for each risk
group
Used scores to develop a process to forecast write-off
rates
Monitored changes in customer behavior over time
Customer was looking to reduce their high operational collection
costs and headcount as well as improve cash flow by prioritizing
collections based on risk
11. Case Study: Pre Charge-Off Collection Scoring Results
Recession Period Customer saved $1.2 million in write-offs in the
first year compared to their competitors utilizing PMIs scoring
technology and their own internal data Pre charge-off collection
scoring Customer experienced substantially better results
12. Greatest $ Collected Least $ Collected Value of using
internal placement data blended with socio-economic data Focus in
Predictive Analytics for Post Charge-Off Collections
13. Properly segmenting accounts is important because you make
your treatment decisions based on knowledge of the segment Focus in
Predictive Analytics for Post Charge-Off Collections $10,243,686
$20.51 1.84% 7.51% $556,059,274 100.00% 499,529 Total: 100.00%
100.00% $99,560 $1.79 1.13% 2.08% $8,820,016 11.11% 55,483 F 99.03%
88.89% $156,528 $2.32 0.68% 3.08% $23,133,566 13.50% 67,447 D
97.50% 75.39% $253,254 $3.97 1.06% 4.64% $23,940,079 12.78% 63,821
C3 95.03% 62.61% $651,278 $7.54 1.45% 5.62% $44,988,001 17.29%
86,366 C2 88.67% 45.33% $1,218,156 $12.64 1.39% 6.91% $87,868,863
19.29% 96,343 C1 76.78% 26.04% $2,472,452 $30.76 1.85% 10.49%
$133,532,931 16.09% 80,366 B3 52.64% 9.95% $1,055,923 $56.03 1.96%
16.19% $53,887,194 3.77% 18,846 B2 42.33% 6.18% $717,277 $72.72
1.95% 20.91% $36,867,638 1.97% 9,863 B1 35.33% 4.20% $878,713
$99.70 2.22% 25.70% $39,532,227 1.76% 8,814 A3 26.75% 2.44%
$1,080,638 $155.71 2.58% 31.15% $41,825,617 1.39% 6,940 A2 16.20%
1.05% $1,659,908 $316.78 2.69% 34.96% $61,663,141 1.05% 5,240 A1
(J) (I) (H) (G) (F) (E) (D) (C) (B) (A) Payments Number Payments
per Account to Balance of Payment Balance Distribution Number
Rating Cumulative Payment(s) Payments Incidence Number Dollar
Actual Results Account Scoring Distribution
14. Cost benefit and profitability analysis
Portfolio was scored and segmented into 11 groups, A through F;
A highest liquidation per account and F the lowest
Profit associated with providing equal collection effort and
costs are $5,248,406
After applying a collection strategy to minimize costs on less
profitable segments, segments C1 through F, and increasing costs or
remaining at $10 on higher scored segments, A1 through B3 = GREATER
PROFITS
Focus in Predictive Analytics for Post Charge-Off Collections
15. Statistical Collection Scoring Advantage
Statistical-based models typically deliver 10% to 300%, or
more, improvement in predicting risk or liquidity over
non-scientific methods
Segmenting accounts based on risk or liquidity allows for
different collection treatments to be applied based on cost, effort
and profitability
PMI conducts a Free historical back test (retrospective
analysis) on your own portfolio using actual payment performance
information allowing you to see how the scores would have worked,
if you had used the scores to segment, prioritize and create
strategies and tactics
Score all accounts using your internal data (even bureau no
hits)
Allows for increased collection department performance without
adding staff
Advanced reporting is provided with the scores to set up
cost-effective collection strategies
Our models bring greater efficiency and effectiveness to the
collection process, substantially reducing the cost of collections
and optimize resources