SYSTEM AND METHOD FOR AUTOMATED DETECTION OF NEVER-PAY DATA SETS
Data filters, models, and/or profiles for identifying and/or predicting the never-pay population (for example, those customers that make a request for credit and obtain the credit instrument but over the life of the account, never make a payment) can be useful to various commercial entities, such as those issuing mortgages, home equity lines of credit, consumer or business lines of credit, automobile loans, credit card accounts, or those entities providing services, such as utility services, phone services, and the like.
This application claims the benefit of U.S. Provisional Application No. 60/931,902, titled METHODS AND SYSTEMS FOR MODELING NEVER-PAY POPULATION and filed on May 25, 2007, which is hereby incorporated by reference in its entirety, specifically the systems and methods for modeling the never-pay population as disclosed therein.
BACKGROUND1. Field of the Invention
This disclosure generally relates to data filters for modeling and processing credit report data and other data, and more particularly to improved systems and methods for generating and using data filters configured to conduct customer profiling and customer analysis relating to modeling, identifying, and/or predicting the never-pay population.
2. Description of the Related Art
Various financial service entities provide credit accounts, such as, for example, mortgages, automobile loans, credit card accounts, and the like, to consumers and or businesses. Prior to providing a credit account to an applicant, or during the servicing of such a credit account, many financial service providers want to know whether the applicant or customer will be or is likely to be within the “never-pay” population. The never-pay population includes without limitation those customers that make a request for credit, subsequently obtain the credit instrument, and over the life of the account, never make a payment or substantially never make a payment. Although the never-pay population is not always large (however, it can be a large population for certain financial firms, for example, those firms serving the sub-prime market or the like), it is a costly population to financial service providers and other entities. Most financial service providers can attribute a certain percentage of their losses to the never-pay population.
Traditional scoring models do not provide the necessary insight to identify the never-pay population. In part, this is due to the diversity of profiles that underlie these populations. Additionally, the attributes and/or reasons that contribute to the never-pay population are difficult to identify for some financial service providers because of their limited resources and the complexity of analyzing the never-pay population. Accordingly, these never-pay accounts are not identified early in the process, and are treated as typical credit loss and are often written off as bad debt.
SUMMARYNever-pay data filters, models, and/or profiles can be generated and applied to both data for potential and actual customers (for example, individual consumers, businesses, or the like) to determine their propensity to never make a payment on a credit account.
In an embodiment, a never-pay automated detection system, the system comprising: a processor configure to run software modules; a data storage device storing a plurality of consumer records comprising credit bureau data, tradeline data, historical balance data, and demographic data, the data storage device in electronic communication with the computer system; and a never-pay module configured to: identify a subset of the plurality of consumer records from the data storage device; receive a first never-pay data profile from a storage repository, the first never-pay data profile identifying consumer records that are likely or substantially likely to never make a payment; apply the first never-pay data profile to each of the subset of the plurality of consumer records to generate a first never-pay score for each of the subset of the plurality of consumer records; and store in a database an aggregate never-pay score associated with the subset of the plurality of the consumer records, the aggregate never-pay score comprising at least the first never-pay score; the processor able to run the never-pay module.
In another embodiment, the never-pay module further configured to receive a second never-pay data profile from the storage repository, the second never-pay profile identifying consumer records that are likely or substantially likely to never make a payment, and apply the second never-pay profile to each of the subset of plurality of consumer records to generate a second never-pay score for each of the subset of plurality of consumer records to be included in the aggregate never-pay score. In an embodiment, the never-pay module further configured to receive a third never-pay data filter from the storage repository, the third never-pay profile identifying consumer records that are likely or substantially likely to never make a payment, and apply the third never-pay filter to each of the subset of plurality of consumer records to generate a third never pay score for each of the subset of plurality of consumer records to be included in the aggregate never-pay score.
In an embodiment, a computer implemented method for maintaining a database comprising: electronically identifying a plurality of consumer records, wherein the consumer records comprise credit bureau data, tradeline data, historical balance data, and demographic data; electronically receiving a first never-pay data filter from a storage repository; electronically applying the first never-pay data filter to each of the plurality consumer records to generate a first never pay score for each of the plurality of consumer records; and electronically storing in a database an aggregate never-pay score associated with each of the consumer records, the aggregate never-pay score comprising at least the first never-pay score.
In an embodiment, the computer implemented method further comprising electronically receiving a second never-pay data filter from the storage repository and electronically applying the second never-pay filter to each of the plurality of consumer records to generate a second never-pay score for each of the plurality of consumer records to be included in the aggregate never-pay score. In an embodiment, the computer implemented method further comprising electronically receiving a third never-pay data filter from the storage repository and electronically applying the third never-pay filter to each of the plurality of consumer records to generate a third never pay score for each of the plurality of consumer records to be included in the aggregate never-pay score.
In an embodiment, the never-pay automated detection system comprises a processor configured to run software modules; a data storage device storing a plurality of credit data records, the data storage device in communication with the processor; and a never-pay module configured to: identify records in the data storage device that are defined as never-pay records which are likely to indicate consumers that are likely or substantially likely never to make a payment; track the identified records over a time period; and develop a first never-pay data profile that predicts the propensity of a consumer to be a never-pay record using the tracked identified records, the processor able to run the never-pay module.
In an embodiment, a computer implemented method of developing a data filter for automatically identifying never-pay database records comprising: electronically identifying records of a database that are defined as never-pay records which are likely to indicate consumers that are likely or substantially likely never to make a payment; electronically tracking the identified records over a time period; and electronically developing a data filter that predicts the propensity of a consumer to be a never-pay record using the electronically tracked identified records.
For purposes of this summary, certain aspects, advantages, and novel features of the invention are described herein. It is to be understood that not necessarily all such aspects, advantages, and features may be employed and/or achieved in accordance with any particular embodiment of the invention. Thus, for example, those skilled in the art will recognize that the invention may be embodied or carried out in a manner that achieves one advantage or group of advantages as taught herein without necessarily achieving other advantages as may be taught or suggested herein.
The foregoing and other features, aspects and advantages of the present invention are described in detail below with reference to the drawings of various embodiments, which are intended to illustrate and not to limit the invention. The drawings comprise the following figures in which:
Embodiments of the invention will now be described with reference to the accompanying figures, wherein like numerals refer to like elements throughout. The terminology used in the description presented herein is not intended to be interpreted in any limited or restrictive manner, simply because it is being utilized in conjunction with a detailed description of certain specific embodiments of the invention. Furthermore, embodiments of the invention may comprise several novel features, no single one of which is solely responsible for its desirable attributes or which is essential to practicing the inventions herein described.
As used herein the terms “individual(s),” “customer(s),” “consumer(s)”, “applicant(s)”, or “business(es)”, as used herein, are broad terms and are to be interpreted to include without limitation applicants, consumers, customers, single individuals as well as groups of individuals (for example, married couples or domestic partners or the like), business entities, organizations, or the like.
In general, the term “module,” as used herein, refers to logic embodied in hardware or firmware, or to a collection of software instructions, possibly having entry and exit points, written in a programming language, such as, for example, Java, Lua, C or C++. A software module may be compiled and linked into an executable program, installed in a dynamic link library, or may be written in an interpreted programming language such as, for example, BASIC, Perl, or Python. It will be appreciated that software modules may be callable from other modules or from themselves, and/or may be invoked in response to detected events or interrupts. Software instructions may be embedded in firmware, such as, for example, an EPROM. It will be further appreciated that hardware modules may be comprised of connected logic units, such as, for example, gates and flip-flops, and/or may be comprised of programmable units, such as, for example, programmable gate arrays or processors. The modules described herein are preferably implemented as software modules, but may be represented in hardware or firmware. Generally, the modules described herein refer to logical modules that may be combined with other modules or divided into sub-modules despite their physical organization or storage.
In general, the terms “data filter,” “model,” and “profile” as used herein are broad terms that are interchangeable, and generally refer without limitation to systems, devices, and methods for amplifying, selecting, filtering, excluding, predicting, and/or identifying subsets of a dataset that are relevant, substantially relevant, and/or statistically relevant to the user.
As used herein, the terms “financial entity,” “credit providers,” “credit issuers,” “financial institutions,” “clients,” “utility providers,” “utility service providers,” “phone service providers,” “financial service providers,” are broad interchangeable terms and generally refer without limitation to banks, financial companies, credit unions, savings institutions, retailers, utility (telecommunications, gas, electric, water, sewer, or the like) providers, bankcard issuers, credit card issuers, mortgage (for example, sub-prime) lenders, and the like.
Generally, the terms “never-pay” and “straight roller” as used herein are broad terms that are interchangeable, and generally refer without limitation to those customers that make a request for credit, subsequently obtain the credit instrument, and over the life of the account, never make a payment or substantially never make a payment. In an embodiment, the terms “substantially never make a payment” or “substantially likely never to make a payment” are based on various factors including without limitation type of credit/loan, number of credit/loan payments, duration of credit/loan period, amount of credit/loan, size of payment of credit/loan, or the like. Additionally, the foregoing broad terms can also refer without limitation to a booked account that rolls straight to loss without the lender, credit issuer, or the like collecting any fund from the consumer.
Data filters, models, and/or profiles for identifying and/or predicting the never-pay population (for example, those customers that make a request for credit and obtain the credit instrument but over the life of the account, never make a payment) can be useful to various commercial entities, such as those issuing mortgages, home equity lines of credit, consumer or business lines of credit, automobile loans, credit card accounts, or those entities providing services, such as utility services, phone services, and the like.
Some acquisition risk data filters/models and fraud data filters/models identify the respective subsets of the never pay population that align with acquisition risk or fraud data filters/models that they are configured to predict. Such risk models tend to focus on the macro level of risk (for example, 90+ days past due and bankruptcy), while such fraud models attempt to identify some form of fraud, typically identity fraud. The never pay population is, however, comprised of multiple models and/or profiles, some of which do not entirely resemble those of acquisition risk and/or fraud data filters/models of credit risk consumers. Accordingly, in an embodiment, the never-pay data filters and/or profiles include without limitation, the following, and those skilled in the art will recognize other possible data filters, models, and/or profiles without limiting the scope of the disclosure herein.
a) Credit risk data filter and/or profile/model—consumers whose credit profiles include multiple delinquent or derogatory tradelines. These consumers tend to score poorly on risk models, such as, for example, VantageScoreSM or other scores such as, generic risk scores.
b) No intent to pay data filter and/or profile/model—a behavioral pattern in which a consumer seeks and obtains credit with no intention of ever paying the debt obligation.
c) Synthetic credit data filter and/or profile/model—the combining of real and fictitious identification data in order to establish a consumer credit profile. These profiles may not resemble those of a risky consumer. Therefore, risk scores tend to have difficulty identifying these profiles.
d) True name fraud data filter and/or profile/model (for example, second party fraud or third party fraud)—assuming another person's identity in order to open a new credit account. This is typically referred to as “second party” or “third party” fraud. Second party fraud (or familiar fraud) is generally committed by someone known by or close to a genuine customer, usually a relative or employee. Third party fraud is generally fraud committed by an unrelated third party.
e) Credit manipulation data filter and/or profile/model (for example, first party fraud)—providing false information to obtain credit on more favorable terms.
Because the accounts for the never-pay population tend to have above average balance amounts, the losses attributed to such accounts are higher than the losses attributed to normal bad credit accounts. To identify and/or limit the liability incurred by the above, methods and systems are disclosed herein to identify the never-pay population using a never-pay data filters/models and scoring system that complements risk scores.
With reference to
The computing system 100 includes, for example, a personal computer that is IBM, Macintosh, or Linux/Unix compatible. In an embodiment, the computing device comprises a server, a laptop computer, a cell phone, a personal digital assistant, a kiosk, or an audio player, for example. In an embodiment, the exemplary computing system 100 includes a central processing unit (“CPU”) 105, which may include a conventional microprocessor. The computing system 100 further includes a memory 130, such as, for example, random access memory (“RAM”) for temporary storage of information and a read only memory (“ROM”) for permanent storage of information, and a mass storage device 120, such as, for example, a hard drive, diskette, or optical media storage device. Typically, the modules of the computing system 100 are connected to the computer using a standards based bus system. In different embodiments, the standards based bus system could be Peripheral Component Interconnect (PCI), Microchannel, SCSI, Industrial Standard Architecture (ISA) and Extended ISA (EISA) architectures, for example.
The computing system 100 is generally controlled and coordinated by operating system software, such as, for example, Windows 95, Windows 98, Windows NT, Windows 2000, Windows XP, Windows Vista, Linux, SunOS, Solaris, or other compatible operating systems. In Macintosh systems, the operating system may be any available operating system, such as, for example, MAC OS X. In other embodiments, the computing system 100 may be controlled by a proprietary operating system. Conventional operating systems control and schedule computer processes for execution, perform memory management, provide file system, networking, and I/O services, and provide a user interface, such as, for example, a graphical user interface (“GUI”), among other things.
The exemplary computing system 100 includes one or more commonly available input/output (I/O) devices and interfaces 110, such as, for example, a keyboard, mouse, touchpad, and printer. In an embodiment, the I/O devices and interfaces 110 include one or more display device, such as, for example, a monitor, that allows the visual presentation of data to a user. More particularly, a display device provides for the presentation of GUIs, application software data, and multimedia presentations, for example. The computing system 100 may also include one or more multimedia devices 140, such as, for example, speakers, video cards, graphics accelerators, and microphones, for example.
In the embodiment of
In the embodiment of
In the embodiments described herein, the computing system 100 is configured to execute the never-pay module 150, among others, in order to create models/profiles and/or to provide assessment information regarding certain customers, individuals or entities. For example, in an embodiment the computing system 100 creates models that determine the propensity of an individual to be a never-pay record and assesses a never-pay score of an individual or customer that comprises a never-pay record or comprises attributes of a never-pay model. As another example, in an embodiment the computing system 100 applies the created models to determine the propensity of a particular individual/customer or set of individuals/customers to be a never-pay record and assesses the never-pay score of the individual/customer or set of individuals/customers assessed or deemed to be never-pay records. Various other types of scores, related to other types of market opportunities, may also be generated by the computing system 100. As noted above, although the description provided herein refers to individuals or customers, the terms individual and customer should be interpreted to include applicants, or groups of individuals or customers or applicants, such as, for example, married couples or domestic partners, organizations, and business entities.
Referring to
With reference to
In
Referring to
With reference to
Referring to
In the illustrated embodiment depicted in
Referring to
In
Referring to
In
Referring to
With reference to
With reference to
Referring to
In
Referring to
With reference to
Referring to
In reference to
It is recognized that a variety of scoring methods may be used including numeric scores where the lower number indicates a never-pay or where a higher number indicates a never-pay. In addition, other scores may be used such as, for example, letters scores (for example A, B, C, D or F) or categories (for example good, bad), and so forth.
The never-pay model and/or score can be used in or applied to several markets including but not limited to the sub-prime lending market, finance companies, credit unions, savings institutions, retailers, telecommunications companies, bankcard issuers, student loans, other markets wherein credit issuers face risk and/or fraud dilemmas, or any other markets. The never-pay model and/or score is a useful tool for both risk management by allowing risk managers to discriminate on the front-end, and for fraud management by providing fraud managers a better idea on where to focus their efforts.
Additionally, the never-pay data filters, models, profiles, and/or scores can be bundled with a variety of other products and scores including but not limited to VantageScoreSM or any other generic score used to improve account acquisition, reduce account acquisition costs, justify credit line adjustments, predict loss rates, predict risks such as bankruptcy, fraud, and so forth, mitigate liability, or the like. A variety of pricing strategies can be applied to the never-pay model and/or score including but not limited to using the never-pay model and/or score as a value added solution, a loss-leader promotion, a free add-on service, a cross-sell opportunity, or the like. Additionally, the never-pay model and/or score can be offered at various price points depending on different factors including but not limited to speed of response, the number of profiles/models applied, the number of records reviewed, or the like.
There are several advantages in using various embodiments of the never-pay data filters/models generation system including without limitation: reducing account acquisition costs by helping to eliminate high-risk prospective consumers that do not fit a credit criteria; gaining better intelligence on consumer behavior and motivation by providing access to the most accurate data to show the most complete picture of the right consumer; gaining greater control over risk by more accurately and precisely identifying the never-pay population; automating decision making processes based on non-judgmental, uniform variables selected based on the internal data and/or client external data; allowing lenders, financial entities, and other entities to better discriminate traditional credit risk more finely to address and meet financial reporting and risk management regulatory requirements; or the like.
In some embodiments, the acts, methods, and processes described herein are implemented within, or using, software modules (programs) that are executed by one or more general purpose computers. The software modules may be stored on or within any suitable computer-readable medium. It should be understood that the various steps may alternatively be implemented in-whole or in-part within specially designed hardware. The skilled artisan will recognize that not all calculations, analyses and/or optimization require the use of computers, though any of the above-described methods, calculations or analyses can be facilitated through the use of computers.
Although this invention has been disclosed in the context of certain preferred embodiments and examples, it will be understood by those skilled in the art that the present invention extends beyond the specifically disclosed embodiments to other alternative embodiments and/or uses of the invention and obvious modifications and equivalents thereof. Additionally, the skilled artisan will recognize that any of the above-described methods can be carried out using any appropriate apparatus. Thus, it is intended that the scope of the present invention herein disclosed should not be limited by the particular disclosed embodiments described above.
Claims
1. A never-pay automated detection system, the system comprising:
- a processor configure to run software modules;
- a data storage device storing a plurality of consumer records comprising credit bureau data, tradeline data, historical balance data, and demographic data, the data storage device in electronic communication with the computer system; and
- a never-pay module configured to:
- identify a subset of the plurality of consumer records from the data storage device;
- receive a first never-pay data profile from a storage repository, the first never-pay data profile identifying consumer records that are likely or substantially likely to never make a payment;
- apply the first never-pay data profile to each of the subset of the plurality of consumer records to generate a first never-pay score for each of the subset of the plurality of consumer records; and
- store in a database an aggregate never-pay score associated with the subset of the plurality of the consumer records, the aggregate never-pay score comprising at least the first never-pay score;
- the processor able to run the never-pay module.
2. The never-pay automated detection system of claim 1, the never-pay module further configured to receive a second never-pay data profile from the storage repository, the second never-pay profile identifying consumer records that are likely or substantially likely to never make a payment, and apply the second never-pay profile to each of the subset of plurality of consumer records to generate a second never-pay score for each of the subset of plurality of consumer records to be included in the aggregate never-pay score.
3. The never-pay automated detection system of claim 2 the never-pay module further configured to receive a third never-pay data filter from the storage repository, the third never-pay profile identifying consumer records that are likely or substantially likely to never make a payment, and apply the third never-pay filter to each of the subset of plurality of consumer records to generate a third never pay score for each of the subset of plurality of consumer records to be included in the aggregate never-pay score.
4. The never-pay automated detection system of claim 1, the never-pay module further configured to electronically select the first never-pay data profile from a plurality of never-pay data profiles, the criteria being based at least in part on a customer selection based on price.
5. The never-pay automated detection system of claim 1, the never-pay module further configured to electronically select the first never-pay data profile from a plurality of never-pay data profiles, the criteria being based at least in part on a customer selection based on response speed.
6. A computer implemented method for maintaining a database comprising:
- electronically identifying a plurality of consumer records, wherein the consumer records comprise credit bureau data, tradeline data, historical balance data, and demographic data;
- electronically receiving a first never-pay data filter from a storage repository;
- electronically applying the first never-pay data filter to each of the plurality consumer records to generate a first never pay score for each of the plurality of consumer records; and
- electronically storing in a database an aggregate never-pay score associated with each of the consumer records, the aggregate never-pay score comprising at least the first never-pay score.
7. The computer implemented method of claim 6 further comprising electronically receiving a second never-pay data filter from the storage repository and electronically applying the second never-pay filter to each of the plurality of consumer records to generate a second never-pay score for each of the plurality of consumer records to be included in the aggregate never-pay score.
8. The computer implemented method of claim 7 further comprising electronically receiving a third never-pay data filter from the storage repository and electronically applying the third never-pay filter to each of the plurality of consumer records to generate a third never pay score for each of the plurality of consumer records to be included in the aggregate never-pay score.
9. The never-pay automated detection system of claim 6, further comprising a criteria for electronically retrieving the first never-pay data filter from a plurality of never-pay data filters, the criteria based at least in part on a selection by a customer based on price.
10. The never-pay automated detection system of claim 6, further comprising a criteria for electronically retrieving the first never-pay data filter from a plurality of never-pay data filters, the criteria being based at least in part on a selection by a customer based on response speed.
11. A storage medium having a computer program stored thereon for causing a suitably programmed system to process computer-program code by performing the method of claim 6 when such program is executed on the system.
12. A never-pay automated detection system, the system comprising:
- a processor configured to run software modules;
- a data storage device storing a plurality of credit data records, the data storage device in communication with the processor; and
- a never-pay module configured to:
- identify records in the data storage device that are defined as never-pay records which are likely to indicate consumers that are likely or substantially likely never to make a payment;
- track the identified records over a time period; and
- develop a first never-pay data profile that predicts the propensity of a consumer to be a never-pay record using the tracked identified records,
- the processor able to run the never-pay module.
13. The never-pay automated detection system of claim 12, the never-pay module further configured to store the data profile in a data repository.
14. The never-pay automated detection system of claim 12, the data profile comprising a first party fraud attribute.
15. The never-pay automated detection system of claim 12, the data profile comprising a third party fraud attribute.
16. A computer implemented method of developing a data filter for automatically identifying never-pay database records comprising:
- electronically identifying records of a database that are defined as never-pay records which are likely to indicate consumers that are likely or substantially likely never to make a payment;
- electronically tracking the identified records over a time period; and
- electronically developing a data filter that predicts the propensity of a consumer to be a never-pay record using the electronically tracked identified records.
17. The computer implemented method of claim 16 further comprising electronically storing the data filter in a data repository.
18. A storage medium having a computer program stored thereon for causing a suitably programmed system to process computer-program code by performing the method of claim 16 when such program is executed on the system.
19. The computer implemented method of claim 16, further comprising electronically applying the data filter to a selected consumer record to generate a never-pay score for the selected consumer record, the never-pay score being predictive of the consumer to be a never-pay record.
20. The computer implemented method of claim 19, further comprising electronically sending marketing material to a consumer linked to the never-pay score if the never-pay score is below a certain threshold.
Type: Application
Filed: May 22, 2008
Publication Date: Nov 27, 2008
Inventors: Christopher J. Celka (Suwanee, GA), Cristian R. Rojas (San Diego, CA)
Application Number: 12/125,820
International Classification: G06Q 40/00 (20060101);