Identifying High Spend Consumers with Equifax Dimensions

Similar documents
UNDERSTAND & PREDICT CONSUMER BEHAVIOUR WITH TRENDED DATA SOLUTIONS

Accelerating Revenue with Customer Centric Offers

The Unique Credit Characteristics of Healthcare Patients. An Equifax Predictive Sciences Research Paper December 2003

Equifax Canada Reports-National Delinquency Rates Decline to Their Lowest Levels Ahead of the Holidays

Automotive Services. Tools for dealers, lenders and industry service providers that drive profitable results in today s economy

Driving Growth with a New Measure of Credit Capacity

Implementing a New Credit Score in Lender Strategies

White paper. Trended Solutions. Fueling profitable growth

CREDIT SCORE USER GUIDE

How much can increased predictive power impact profits?

Attract and retain more high-quality customers while reducing your risks.

CRIF Lending Solutions WHITE PAPER

Executing Effective Validations

Equifax Canada Reports: Consumer Appetite for Credit Grows as Total Debt Climbs to $1.718 Trillion

CREDIT REPORT USER GUIDE

Harnessing Traditional and Alternative Credit Data: Credit Optics 5.0

2008 VantageScore Revalidation

How Are Credit Line Decreases Impacting Consumer Credit Risk?

Making More Informed Decisions

Using alternative data, millions more consumers qualify for credit and go on to improve their credit standing

Top US Bankcard Issuer Validates the Power of FICO 8 Score Key metrics exceed client expectations in originations testing

Universe expansion. Growth strategies in the evolving consumer market

2017 Experian State of Lending. Who s Winning? Community Banks vs. Credit Unions

FICO s analysis indicates:

Analytic measures of credit capacity can help bankcard lenders build strategies that go beyond compliance to deliver business advantage

Boost Collections and Recovery Results With Analytics

Abstract. Estimating accurate settlement amounts early in a. claim lifecycle provides important benefits to the

Trended Credit Data Attributes in VantageScore 4.0

The Stark Reality of Synthetic ID Fraud How to Battle the Leading Identity Fraud Tactic in The Digital Age

SEGMENTATION FOR CREDIT-BASED DELINQUENCY MODELS. May 2006

For competitive advantage, refresh more frequently. FICO s analysis indicates:

Maximizing the Credit Universe

GET SOCIAL WITH US. #vision2016. Tweet, follow, share throughout the session.

In a credit-hungry economy, how much is too much?

December 2015 Prepared by:

Are today s market pressures reshaping credit risk?

Scoring Credit Invisibles

Forecasting Portfolio Performance in an Uncertain Economy

A new highly predictive FICO Score for an uncertain world

Using data mining to detect insurance fraud

Credit Score Basics, Part 3: Achieving the Same Risk Interpretation from Different Models with Different Ranges

Alternative Credit Scores: The Key to Financial Inclusion for Consumers

Presentation to August 14,

Get Smarter. Data Analytics in the Canadian Life Insurance Industry. Introduction. Highlights. Financial Services & Insurance White Paper

We are experiencing the most rapid evolution our industry

SF ROX. Investing in Execution Pays Dividends. Authors: Global Research. Copyright 2010 SuccessFactors, Inc.

Score migration strategies for turbulent times

Maximizing predictive performance at origination and beyond!

Another Approach to Managing High Risk Customers Vision and Strategy Document

PERSPECTIVES ON RETIREMENT

Federal Reserve Bank of Philadelphia

Life After Foreclosure and Hidden Opportunities

Research shows opportunities for lenders who act quickly and leverage sophisticated scoring and analytic tools

VANTAGESCORE SOLUTIONS INTRODUCES VANTAGESCORE 3.0 MODEL

A Balanced View of Storefront Payday Borrowing Patterns Results From a Longitudinal Random Sample Over 4.5 Years

Emerging Opportunities in Home Equity Lending. Joe Mellman Senior Vice President, Mortgage Business Lead

Business Identity Report PlusTM

Which creditors get priority when businesses face a financial burden?

Managing Risk in the Credit Crunch

Lazard Insights. The Art and Science of Volatility Prediction. Introduction. Summary. Stephen Marra, CFA, Director, Portfolio Manager/Analyst

Making Predictive Modeling Work for Small Commercial Insurance Risk Assessment

LexisNexis RiskView Report

The power of plan wellness

Actionable Intelligence December 2017

HEARTLAND VALUE FUND

FIGHTING FRAUD & CHARGEBACKS 5 STRATEGIES FOR WINNING

DriverRisk Guide: Violation insight to fuel your business

Market Variables and Financial Distress. Giovanni Fernandez Stetson University

W H I T E P A P E R. Sabrient Multi-cap Insider/Analyst Quant-Weighted Index DAVID BROWN CHIEF MARKET STRATEGIST

Model Maestro. Scorto TM. Specialized Tools for Credit Scoring Models Development. Credit Portfolio Analysis. Scoring Models Development

JACOBS LEVY CONCEPTS FOR PROFITABLE EQUITY INVESTING

HELOC end-of-draw analysis

CECL CONSIDERATIONS FOR CREDIT CARDS

A Renewed Focus on Risk Management at US Public Pensions

OTC Derivatives Valuation and Data Services Technology-enabled solutions for derivatives and complex instruments

White Paper. Who s Getting Paid During the Subprime Crisis?

Understanding HELOC end of draw

Mapping the Member Journey

The MarketGrader China A-Shares Size Indexes:

Monograph. Competitive Intelligence An Insurance Policy for Pricing Kathryn A. Walker, FCAS, MAAA, CPCU ABOUT THE AUTHOR KEY POINT

Key findings. FICO s analysis indicates: A notable percentage of scores migrated up or down more than 20 points over just one month.

Event Performance Indices (EPI) Report

银行管理学 作业习题. Chapter 1 Fundamental Forces of Change in Banking

Session 73 PD, Predictive Modeling for the Marketing Actuary. Moderator: Maria Patricia Marcelo Arellano, FSA, CERA, MAAA

Forecasting More Profits For You and Your Clients

Improving Tax Administration with Data Mining

White Paper. Liquidity Optimization: Going a Step Beyond Basel III Compliance

CreditVision New Account Risk Score study

Is Growing Student Loan Debt Impacting Credit Risk?

RESEARCH Stock Scoring System. An in-depth look at Burney's stock selection process

Your Credit Score What It Means to You as a Prospective Home Buyer

Universe Expansion: Is the Way You Score Customers State of the Art or State of Denial?

A Decade of Validation Demonstrates Superior Performance

2015 Letter to Our Shareholders

Universe Expansion: Is the Way You Score Customers State of the Art or State of Denial?

Account Management for Risk or Profit? Eva Neves Senior Consultant PIC Solutions

A distinctive solution for your plan and employees. TIAA-CREF Lifecycle Funds

D-U-N-S Number 3. Corporate Linkage 4. D&B Global Database. Quality Assurance

Consistent, immediate, cost-effective decisions about your customers

ESG Policy & Process. 1. Overview and Philosophy

Transcription:

Identifying High Spend Consumers with Equifax Dimensions April 2014

Table of Contents 1 Executive summary 2 Know more about consumers by understanding their past behavior 3 Optimize business performance by knowing more 3 Methodology 3 Results 6 See it in action 8 Conclusion About Equifax Equifax is a global leader in consumer, commercial and workforce information solutions that provide businesses of all sizes and consumers with insight and information they can trust. Equifax organizes and assimilates data on more than 600 million consumers and 81 million businesses worldwide. The company s significant investments in differentiated data, its expertise in advanced analytics to explore and develop new multi-source data solutions, and its leading-edge proprietary technology enable it to create and deliver unparalleled customized insights that enrich both the performance of businesses and the lives of consumers. Headquartered in Atlanta, Equifax operates or has investments in 18 countries and is a member of Standard & Poor's (S&P) 500 Index. Its common stock is traded on the New York Stock Exchange (NYSE) under the symbol EFX. In 2013, Equifax was named a Bloomberg BusinessWeek Top 50 company, was #3 in Fortune's Most Admired list in its category, and was named to InfoWeek 500 as well as the FinTech 100. For more information, please visit www.equifax.com. www.equifax.com Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions ii

Identifying High Spend Consumers with Equifax Dimensions Executive summary For businesses, the ability to predict individual consumer spend amount within a specific time frame is essential. It impacts their marketing strategies across the customer life cycle, particularly in the areas of customer acquisition and account management. But, their ability to understand spending preferences is also a key factor in developing customer loyalty. To provide that wider view of individual consumer credit behaviors, Equifax Dimensions was developed. The solution delivers unique consumer insights around credit usage and payment behavior signals transpiring over 24 months of history. By leveraging aggregated time series account-level consumer credit attributes, Equifax Dimensions can help businesses identify consumer behavior patterns and trends by industry and account type and use that data to better segment and target consumers with more meaningful offers. Proving the value of Equifax Dimensions, a data-driven comparison study within this document reveals how the solution can help businesses drive stronger, more profitable marketing and sales strategies by more accurately targeting high-spend consumers in the market. The study shows how Equifax Dimensions can: More powerfully rank spend amounts compared to using traditional credit attributes alone More efficiently separate high-spenders from low-spenders Help improve revenue opportunity throughout a large portfolio Based on the sample population outlined in the study, the results show how using Equifax Dimensions can help generate an estimated $111 million in increased spending and $4.2 million in potential new revenue within the top 30% percent of the population, providing businesses with a fresh strategy for increasing revenue while also reducing marketing costs. Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 1

Trended or Time-Series Data Trended, historical, longitudinal or timeseries information captures the trends within a data set for a specific period of time. A sequence of data points is typically measured at successive points in time spaced at uniform time intervals. Time-series data can be analyzed to extract meaningful statistics and help predict future values based on previously observed values. When measuring trended data within a consumer credit profile, analysis on direction, velocity, tipping points and magnitude of changes within the profile can deliver valuable information on credit behavior such as spending and payment patterns. Know more about consumers by understanding their past behavior Often, the best predictor of future behavior is past behavior. It was on this premise that Equifax Dimensions was developed. As a complement to basic tools such as consumer reports, which provide a snapshot of credit activity at one specific point in time, Equifax Dimensions helps reveal consumer credit trends that measure the general direction in which the data is developing, such as increasing or decreasing spend levels. Specifically, Equifax Dimensions provides 500 consumer attributes built from 24 months of account-level data that complies with Fair Credit Reporting Act guidelines. In practical applications, the trended data provided by Equifax Dimensions can help businesses identify: Propensity to open new accounts within a period of time Accounts more likely to be active Likelihood of balance transfer within a period of time Best candidates for line increases and likelihood to stay current Capacity to take on more debt without becoming past due How credit is used over an extended period of time Past credit performance and behaviors Credit stress and risk before delinquency and charge-off Accounts best suited for efficient collections and recovery efforts The solution includes flexible controls that allow users to set unique consumer risk levels, behaviors and characteristics, so businesses can more accurately identify and address specific behaviors that align with their business model. Likewise, the solution attributes are available both online for real-time decisioning, and offline for custom analytics and account management functions. Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 2

Optimize business performance by knowing more Study shows Equifax Dimensions significantly expands predictive insight To demonstrate how Equifax Dimensions can better predict consumer spend, this study used a two-stage testing methodology to compare the time-series attributes within Equifax Dimensions to traditional credit attributes, which use static data. The first step calculated spend on an account level first and then aggregated it to the consumer level. The second step distinguished high spend versus low spend. Observation status was defined as follows: Data Exclusion Criteria Industry Use Case A random 1% sample of the Equifax National Credit Database - Deceased consumer - Consumer with bankruptcy - Consumer with fraud - Not a tradeline Revolving accounts Spend: total spend amount on revolving accounts within 12 months is defined by using a combination of balance and payment information Spend was modeled as binary target variable: high spend/low spend created at the median (around $500) Methodology Two predictive models (logistic regression models) were built to forecast high spend: Champion model only utilized standard credit attributes, a set of attributes providing effective solutions for risk and profitability goals. Challenger model utilized standard credit attributes plus Equifax Dimensions attributes. Incremental lift is measured on statistical performance, based on a Kolmogorov-Smirnov (KS) test, 1 and 10 percent, 20 percent, and 30 percent target capture rates 2. Result Overview Based on the sample population studied, adding Equifax Dimensions in the Challenger model captured more high-spenders and generated more spending in top deciles compared to the Champion model, capturing about 1,800 more high-spenders and generating about $38 million more spend and $1.4 million in potential new revenue in the top decile. Result Details More powerfully ranked spend amounts compared to using traditional credit attributes alone Capturing consumer spend is a top priority for lenders. To better illustrate the capabilities of Equifax Dimensions attributes, Equifax used the sample population to further examine the relationship between score values and average Total Spend Value within 12 Months. 1 The Kolmogorov-Smirnov (KS) statistic is the measure of the maximum distance (greatest separation) between the cumulative % of high spend (goods) and the cumulative % of low spend (bads) across all score ranges. It represents the model's ability to differentiate "goods" from "bads" in the sample. Higher values indicate better overall separation and a stronger model. 2 10% capture rate: what percent of bads are captured within the 10% highest scores. In this comparison study, it means the rate to capture high-spend consumers. Similar for 20% and 30% capture rate. Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 3

Average Spend by Score Decile Champion: Credit Attributes Only Challenger: Credit Attributes + Dimensions Attributes $3,000 $2,500 Average Spend $2,000 $1,500 $1,000 $500 $0 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Score Decile Figure 1: Average Total Spend Within 12 Months by Score Decile Figure 1 demonstrates that credit attributes and Equifax Dimensions used together can powerfully rank order the spend amount the higher the score, the higher the spend. More efficient separation of high-spenders from low-spenders Figure 2 demonstrates how Equifax Dimensions more efficiently separates highspenders from low-spenders compared to using credit attributes alone. After adding Dimensions attributes, the percent of high-spenders captured in the top three deciles (or top 30%) of scores is higher in the Challenger model, illustrating the value of Dimensions attributes to capture high-spenders. Champion: Credit Attributes Only High Spenders by Score Decile Challenger: Credit Attributes + Dimensions Attributes 20 % of High Spenders 15 10 5 0 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Score Decile Figure 2: High spenders by Score Decile Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 4

Improved revenue opportunity throughout a large portfolio To optimize offers, terms and promotions, businesses need as much reliable insight as possible into how much consumers are spending. Figure 3 shows that using credit attributes alone in the Champion model provided strong model performance with a KS of 43.77. However, adding Equifax Dimensions further improved the model's predictive performance, with the Challenger model hitting a KS of 46.78 - a 7 percent lift over the Champion model. More importantly, the target capture rates across the top three score deciles in the sample population improved by more than 4 percent, demonstrating that Equifax Dimensions can deliver deeper insight on more consumers. Comparison Study Testing Methodology KS 10% Target Capture Rate 20% Target Capture Rate 30% Target Capture Rate Credit Attributes Only 43.77 18.05% 33.74% 47.85% Challenger Model Credit Attributes Plus Equifax Dimensions 46.78 18.71% 35.45% 50.20% Improvement by Equifax Dimensions 3.01 0.66% 1.71% 2.35% Lift from Equifax Dimensions 7% 4% 5% 5% Figure 3. Model Results Comparison by KS and Target Capture Rate Figure 4 shows the percent of increased spend amount for all score ranges after adding Equifax Dimensions attributes to the Champion model. Equifax Dimensions attributes effectively push high-spend consumers to the top score ranges, as spend amount is increased by 17 percent, 11 percent and 6 percent for the top three score deciles. Percent of Increased Spend Amount by Score Decile 25% 20% % of Increased Spend 15% 10% 5% 0% 17% 11% 6% 10% 20% 30% Score Decile Figure 4: Percent of Increased Spend Amount by Score Deciles Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 5

Considering Figures 3 and 4 together, it shows how Equifax Dimensions can help businesses target high-spend consumers more accurately in the market by putting a spend-amount increase analysis into model comparison results. After adding Equifax Dimensions attributes, the top score decile's target capture rate increased by 0.66 percent, but spend amount increased by 17 percent compared with the Credit-Attributes-Only model. The same pattern can be found on the top 20 percent and 30 percent deciles with a diminishing trend. This indicates that Equifax Dimensions can not only push highspend consumers to the top score deciles, but it can also suppress the low-spender rate, thus helping to increase the revenue opportunity. See it in action To demonstrate the increased number of high-spenders and total spend in top deciles captured by the Challenger model over the Champion model, Figure 5 presents a detailed example scenario based on a sample portfolio of 1 million consumers. Adding Equifax Dimensions helps capture more high-spenders in top deciles, capturing about 1,800 more high-spenders and generating about 38 million dollars more spend in the top decile in the sample portfolio. This strategy helps increase the revenue opportunity throughout a portfolio, while also reducing the long-term costs to identify and market to those high-spenders. Champion Model Total Spend Champion/Challenger Models Challenger Model Total Spend Champion Model High-Spender Count Challenger Model High-Spender Count 600 6% 250 Amount of Increased Spend ($ in millions) 500 400 300 200 100 17% 11% 200 150 100 Number of High-Spenders (in thousands) 0 10% 20% 30% Score Decile 50 Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 6

Champion Model Challenger Model Top Decile # of Consumers # of High-Spenders Total Spend # of High-Spenders Total Spend % Increased Spend 10% 100,000 89,189 $229 Million 90,909 $267 Million 17% 20% 200,000 164,619 $400 Million 174,201 $442 Million 11% 30% 300,000 232,555 $539 Million 244,472 $570 Million 6% Figure 5. Increased Spend of Top Deciles by Challenger Model over Champion Model The sample portfolio was also used to gauge potential incremental interest and transaction fees. Using Equifax Dimensions to identify more consumers that spend more over a 12-month period can result in incremental revenue of up to $1.4 million in the top 10% of scored consumers. This example assumes no other fees besides APR and transaction fees, and that half of the new consumers are revolvers making only the minimum payment. Potential Incremental Revenue by Decile Interest Fees Transaction Fees Total Revenue $1,800,000 $1,600,000 $1,400,000 $1,200,000 $1,000,000 $800,000 $600,000 $400,000 $200,000 $0 $1.44M $1.59M $1.17M 10% 20% 30% Score Decile Net New Spend Transaction Fees Interest Fees Total Revenue 10% $38,000,000 $760,000 $682,813 $1,442,813 20% $42,000,000 $840,000 $754,688 $1,594,688 30% $31,000,000 $620,000 $557,031 $1,177,031 Figure 6. Potential Incremental Revenue by Decile Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 7

Conclusion To maximize business performance, marketing strategies and return on investment, and improve customer response and loyalty, businesses are redesigning their strategies to identify, target, acquire and keep profitable customers. Equifax Dimensions can help by offering deeper, trended insight that shows past consumer credit behaviors over 24 months to provide businesses with more precise predictive power. Moreover, comparison results prove that pairing Equifax Dimensions with traditional credit attributes can help businesses effectively capture the high-spend consumers in the market. This enhanced insight can help businesses focus their marketing efforts on the most profitable consumers and perform more accurate targeting, better risk mitigation and enhanced segmentation. Contact Us Today For more information, please contact: www.equifax.com/consumer There are references throughout this publication/website to various trademarks or service marks and these, whether registered or not, are the property of their respective owners. Equifax is a registered trademark of Equifax Inc. Inform > Enrich > Empower is a trademark of Equifax Inc. Insert all other trademark, registration, text here. Copyright 2014. Equifax Inc., Atlanta, Georgia. All rights reserved. Equifax Inc. Identifying High Spend Consumers with Equifax Dimensions 8

www.equifax.com EFX-00201-3-14