Method and Apparatus for Real Time Online Credit Approval
A system and method are described for providing real time approval of credit over a network. The method includes obtaining applicant data from an applicant. The applicant data is analyzed into a form suitable for directly obtaining a credit report from a credit bureau for the applicant. A credit report having credit report data is obtained from a credit bureau for the applicant. It is then determined whether to accept the applicant using the credit report data and it is communicated to the applicant that the applicant has been approved.
This application is a continuation of U.S. patent application Ser. No. 10/782,311,” filed Feb. 18, 2004, which is a continuation of U.S. patent application Ser. No. 09/595,556, filed Jun. 15, 2000, now U.S. Pat. No. 6,718,313, which is a continuation-in-part of U.S. patent application Ser. No. 09/185,201, filed Nov. 3, 1998, now U.S. Pat. No. 6,405,181, and of U.S. patent application Ser. No. 09/858,878, now U.S. Pat. No. 6,567,791, all of which are herein incorporated by reference.TECHNICAL FIELD OF THE INVENTION
The present invention relates generally to electronic commerce. More specifically, the invention relates to methods and apparatuses for providing real time credit approval to an applicant online by obtaining data from an applicant, verifying and formatting the data so obtained in a manner that permits accessing the applicant's credit report, and making an underwriting decision to grant or deny credit to the applicant in real time based on data from one or more credit bureau reports.BACKGROUND
With the advent of electronic commerce on the Internet, applicants have begun to expect decisions that have historically required a period of days or weeks to be made instantly when processed on line. Numerous transactions such as purchases of consumer goods, airline tickets, and movie tickets have been adapted for execution on line in a matter of seconds. What has not been perfected is the ability to make a credit decision and grant credit to a party on line in real time. (For the purpose of this specification, “instant” or “real time” credit means within a short period of time within less than about five minutes.) As a result, virtually all Internet commerce to date requires some previously secured method of payment such as a credit card obtained by conventional means or other previously ananged payment source such as a bank account or electronic money.
One factor that has prevented Internet applicants from providing information and receiving instant approval for credit is the difficulty of interfacing with the various credit bureau databases (Equifax, Trans Union, and Experian). Personal information must be entered by a party authorized by the credit bureaus to communicate with the credit bureaus for the purpose of accessing credit bureau reports. Such information must be in exactly the correct form in order for an individual's credit report to be retrieved. Another difficulty has been that the decision to grant credit carries with it significant risk and systems have not been successfully designed that can make a sufficiently reliable underwriting decision using data provided directly by an applicant. Many credit card issuers provide applications on line that may be filled out by applicants. However, data from those applications must be entered manually into the credit card issuer's system for processing before a credit report is obtained and an underwriting decision can be made. Other applicants may be preapproved by an existing card issuer's system before an offer is made and accepted online. However, the underwriting process has not been sufficiently automated to allow a credit decision to be made in real time for an applicant who has entered personal data into an application system.
What is needed is a system and method for obtaining personal data from a credit applicant, parsing the data into a format that is compatible with that used by the credit bureaus, obtaining credit bureau information and making an underwriting decision in real time. Such a system would be useful for conveniently obtaining a credit card on line. Automation of a process for obtaining a credit report and making
an underwriting decision without human intervention would be beneficial because credit approval decisions could be made faster and more cheaply. The true power of such a system would be realized when the system is accessed in the midst of a transaction to obtain credit specifically for the purpose of that transaction.SUMMARY
The present invention provides a system and method for obtaining information from an applicant, accessing credit bureau information and making a real time underwriting decision to accept or reject the applicant. A parsing engine parses the information provided by the applicant so that it may be sent directly to a credit bureau. Information obtained from one or more credit bureaus is used by an underwriter engine to make a decision whether to grant credit to the applicant. It should be appreciated that the present invention can be implemented in numerous ways, including as a process, an apparatus, a system, a device, a method, or a computer readable medium. Several inventive embodiments of the present invention are described below. In one embodiment, a method of providing real time approval of credit over a network is disclosed. The method includes obtaining applicant data from an applicant. The applicant data is analyzed into a form suitable for directly obtaining a credit report from a credit bureau for the applicant. A credit report having credit report data is obtained from a credit bureau for the applicant. It is then determined whether to accept the applicant using the credit report data and it is communicated to the applicant that the applicant has been approved. These and other features and advantages of the present invention will be presented in more detail in the following specification of the invention and the accompanying figures which illustrate by way of example the principles of the invention.
The present invention will be readily understood by the following detailed description in conjunction with the accompanying drawings, wherein like reference numerals designate like structural elements, and in which:
Reference will now be made in detail to the preferred embodiment of the invention. An example of the preferred embodiment is illustrated in the accompanying drawings. While the invention will be described in conjunction with that preferred embodiment, it will be understood that it is not intended to limit the invention to one preferred embodiment. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be practiced without some or all of these specific details. In other instances, well known process operations have not been described in detail in order not to unnecessarily obscure the present invention.
The parsing engine 106 parses the data into an exact format that may be used to directly access credit bureau data. The applicant is given an opportunity to view how the data submitted has been parsed and to make corrections to parsed data, if necessary. The parsing engine 106 is described in further detail in
Underwriter 110 receives data from the parsing engine and evaluates the data to determine if the applicant should receive an offer for credit. In one embodiment, the Underwriter sends the parsed data to at least two credit bureaus, receives data from the credit bureaus, and makes an underwriting decision based on an analysis of the credit bureau data. The analysis may include, but is not limited to, comparing the applicant's Fair Isaac Risk Score (FICO score) to certain thresholds. Underwriter 110 is described in further detail in
Whether an offer is extended and accepted or not, information about the offer or the rejection is passed to a creditor module 112 that finalizes the offer and builds a data file that is in the proper form to be sent to First Data Resources, Inc. (FDR), or another such entity that provides a similar service to FDR's service. During the finalization of the offer, FDR data is built for all approved and declined applications. FDR handles the embossing of the card and delivering it to approved applicants. FDR also handles sending rejection letters to rejected applicants.
If, at any time during the process, a system error occurs that interrupts the process, then an application object loader 114 loads the appropriate application for reentry into the system. It should be noted that in one embodiment, the data that is processed and stored by each module is stored as an application object as is described further in
The application data structure further includes a set of credit report objects 214 associated with each credit report ordered. In one embodiment, the Underwriter requires at least two credit reports from two of three credit bureaus before a decision to grant credit is made. This rule effectively enables a real time credit decision to be made without incurring an unacceptable amount of risk. Since credit reports are preferably ordered from more than one credit bureau, the application data structure will likely include several credit report objects. Each credit report object 214 includes a plurality of attributes 216. An attribute is an item of data provided by the credit bureau in the credit report. For example, one such attribute is a 90 day attribute that indicates the number of times that the applicant has been more than 90 days late in payment of a debt. Similarly, a 60 day attribute may be provided. Other attributes may include a FICO score, the number of times the applicant has been severely delinquent, existence of a derogatory public record, whether the applicant is now delinquent, the applicant's total revolving balance, and the amount of time that a credit report has been on file for the applicant (also referred to as “thickness of file” or “time on file.”
As is described below, in one embodiment, the Underwriter bases its decision on the FICO score alone when the FICO score is below a rejection threshold. In some embodiments, there may be automatic approval when the FICO score is above an approval threshold.
The application data structure further includes FDR data object 218 associated with the application. FDR data is created by the creditor module for the purpose of sending application information to FDR so that FDR may send credit cards to successful applicants and send rejections to unsuccessful applicants, when that is required.
The application object also includes a status object 220. The status of the application object is determined at various times by the modules. For example, the Validator module may determine that the application is invalid based on an invalid social security number or zip code. The Underwriter module may also determine that the application is a duplicate, as will be described below. The Underwriter may also change the status of an application to accepted or declined. In addition, certain applications may be tagged with a fraud status flag indicating that there is a likelihood of fraud. The application data structure also may include a set of offers 222 to be provided to the applicant.
Thus far, the software architecture and data structure used to make a real time credit decision in one embodiment have been described. Next, the processes implemented in the modules will be described.
If the information is valid, then control is transferred to a step 310 where the data entered is displayed along with the field assigned to each part of the data by the parsing engine. This step is important to ensure that the data will be readable when it is sent to a credit bureau by the Underwriter. An exact match is required by the credit bureaus for the correct credit report to be sent. Various ambiguities in the way that an address may be expressed can cause difficulties. Such difficulties have been a significant factor in preventing other systems from allowing individuals to directly access credit bureau data. For example, it is necessary to distinguish a street direction that is part of a street address from a street name that happens to be a direction, such as “North.”
To make certain that such distinctions as well as other distinctions are made correctly, the parsing engine categorizes each part of the entered address and presents the field names along with that portion of the address that it has assigned to each field name. So, for example, the applicant can move “North” from a street direction field to a street name field if that is appropriate. Thus, by parsing the address and assigning the different parts to fields and then allowing the applicant to check and edit the assignment, the parsing engine enables applicants with no knowledge of the Byzantine structure required by the credit bureaus to enter personal data in a manner that allows a credit report to be obtained without human intervention. Initial parsing is achieved by analyzing the form of the address and dividing, for example, the street number, street name, city and state. However, regardless of the care taken in designing initial parsing, some miscategorization will likely occur. Displaying the parsing to the applicant and allowing the applicant to correct parsing errors enables the imperfect output of the parsing engine to be corrected. At the same time, the process is much more user friendly and less tedious for the user than if the user had been asked to enter each field that the address is divided into by the parsing engine separately. By having the parsing engine parse the address and present the result of the parsing to the user, tedium is minimized and accuracy is achieved.
If the applicant responds that the data and parsing is correct instead of editing the parsing of the data into the displayed fields in step 310, then a step 311 transfers control to a step 312 where pre-credit bureau tests are run on the data. If the applicant edits the data, then control is transferred back to step 306 and the data is re-checked for validity. If the applicant fails the pre-credit bureau test, then the applicant's status is changed to rejected in a step 313 and if the applicant passes the pre-credit bureau test, then the credit bureaus are accessed and credit bureau tests based on the data obtained from the credit bureau and other applicant data are performed in a step 314. If the applicant passes the credit bureau tests, then post credit bureau tests are run in a step 316. If the applicant passes the post credit bureau tests, then the applicant is accepted to receive an offer for credit and the approval process ends at 320.
If the applicant fails the credit bureau tests, then the application status is changed to rejected in a step 315. As described below, an on line rejection process is executed for applications with a rejected status. Thus, the applicant information is input to a series of tests and the result of the tests determines whether the applicant is accepted or rejected.
In this manner, the form of the data entered by the applicant is checked to determine whether the data entered is at least potentially correct. For example, if a social security number that does not exist for anyone is entered, it can be determined that the entered data must be invalid. In other embodiments, additional validation tests may be performed. Specifically, validation tests that help detect fraud may be implemented. In one embodiment, the validation status associated with each test result object includes a time stamp. Multiple applications with the same or similar names may be tracked and a history may be saved. Fraud tests may be implemented that track the number of applications submitted by a given individual and check the consistency of applicant data between multiple submitted applications.
If the applicant meets the minimum age and income requirements, then control is transferred to a step 514. Step 514 checks whether the application entered is a duplicate application. If the applicant has previously entered the information in the application database, then the current application is a duplicate application. It is important to recognize such duplicate applications so that a single applicant cannot require multiple credit reports to be obtained. In one embodiment, duplicate applications are recognized by checking for duplicate social security numbers, duplicate names and/or duplicate addresses. In order to be rejected by the system, an application must match two of the three criteria. A rule is established that an applicant may reapply for a credit card after a specified time period has elapsed (e.g., 60 days). Such a rule is implemented in a step 516 that checks whether the application submission date exceeds a specified time period since the submission date of the found duplicate application. If the application is submitted prior to the specified time period, the status of the application is changed to duplicate in a step 518 and the process ends at 520.
When a duplicate application is submitted, then the applicant is notified and a message is provided that informs the applicants that duplicate applications may not be submitted within a certain time period of each other. In addition, the applicant may also be prompted to go to a re-entry screen that allows the found duplicate application to be processed if processing of that application was previously interrupted. In this manner, if an applicant quit in the middle of the application process, then the application process can be completed for the previously submitted application.
It should be noted that a specific series of pre-credit bureau tests have been shown for the purpose of illustration. Other tests can be used within the scope of this invention. Also, it should be noted that if one test is failed, then remaining tests are skipped in some embodiments. Alternatively, all of the pre-credit bureau tests may be performed and the pre-credit bureau test results may be stored in separate question objects. This may help detect potentially fraudulent applicants who create many duplicates. If an application is determined potentially to be fraudulent, the status of the application is changed to fraud. Alternatively a separate flag may be set to indicate the potential fraud.
Once it is determined the applicant has entered data that is at least potentially valid and the applicant has approved the output of the parsing engine, the application is ready to be checked by the Underwriter to determine whether credit should be approved for the applicant. The Underwriter makes such a determination based on the information obtained from credit bureaus. Since the decision made by the Underwriter is made without human intervention, it is particularly important that the method of determination made by the Underwriter is reliable. For this reason, it is preferred that, in order for an applicant to be approved, at least two credit bureaus must provide information about that applicant that passes a series of tests. In some embodiments, this rule may be relaxed, but a process that requires data from at least two credit bureaus for approval has been shown to have superior reliability to processes without such a requirement. In particular, it has been determined that requiring data from at least two credit bureaus for approval is an important factor in enabling the real time credit approval system to make sufficiently reliable determinations.
Because at least two credit reports from two different credit bureaus are required, it is possible that certain applicants may be rejected because they are only included in the records of a single credit bureau. When this occurs, that reason for rejection is given to the applicant instead of a reason based on the failure of the applicant to pass a test based on credit bureau data.
If two credit bureau tests have not been passed, then control is transferred to a step 616 where it is determined whether one credit bureau test has been passed. If one credit bureau test has not been passed, then the application is rejected in a step 618 for not having a record in at least two credit bureaus. The third credit bureau is not checked since it is not possible to get at least two credit reports at that point. If one credit bureau test has been passed, then a third credit bureau is consulted in a step 620. If the third credit bureau test is failed, then the application is rejected in a step 622 and the process ends at 606. If the third credit bureau report does not have a record for the applicant, then the application is rejected in step 618 for not having enough credit records and the process ends at 606. If the third credit bureau test is passed, then the application is accepted in a step 624 and the process ends at 606.
Thus, the Underwriter only accepts applications that pass at least two credit bureau tests. It should be noted that a special reason for rejection may be given to applicants who are rejected because they do not have a record in at least two credit bureaus. Also, it should be noted that in some embodiments, it is distinguished whether a credit report is not obtained because a credit bureau is temporarily unavailable or whether a credit report is not obtained because there is no record for the applicant. In the event that a credit bureau is unavailable, an applicant that cannot be found in the remaining two credit bureaus may be given a special rejection notice indicating that a later attempt should be made by the applicant when the unavailable credit bureau is functioning. Also, when two credit bureaus are unavailable at the same time, all applicants may be requested to reapply when the credit bureaus return on line.
The credit bureau tests may be separately performed to avoid performing the remaining tests once the failure of the application to pass a test results in a determination that the application will be declined. However, each of the set of attribute tests and credit report specific tests are preferably performed so that the best basis for rejection may be identified and provided to the applicant. Determining an appropriate basis of rejection to display to the applicant is described further below in connection with
As described above, the process of performing the various tests may generally be considered as performing various attribute tests and credit specific tests and combining the results of those tests in some fashion to make a decision to pass or fail an applicant.
It should be noted that in other embodiments, other methods of determining whether to accept or reject an applicant are used. For example, in one embodiment, an applicant is accepted automatically if he or she has a FICO score that is above a certain threshold.
The attribute tests performed in step 688 may take on various forms. In one embodiment, a list of attributes is checked including attributes such as whether the applicant is severely delinquent, currently delinquent, has a derogatory public record, or has been delinquent a certain number of times in a past period. A test may be defined for each attribute such as a maximum number of times delinquent above which the test is failed. In one embodiment, a list of tests is defined and all of the tests must be passed. In another embodiment, a list of tests is defined and certain subsets of the list are also defined. At least one subset must be passed for the applicant to pass.
Once the decision is made to accept or reject an applicant, the status of the applicant is set to be accepted or rejected. Rejected applications are processed in a rejection process described in
If the status of the application determined by the Underwriter is that the application is a duplicate of a previously entered application, then control is transferred to a step 706 and a message indicating that the application is a duplicate is displayed to the applicant. Next, in a step 708, a link to a reentry screen is provided to the applicant. The reentry screen allows the applicant to execute a process that finds the earlier application and allows the applicant to review or resume the earlier application. For example, if the earlier application was accepted but the applicant did not accept an offer, then the process may resume at that point and the applicant may be given another opportunity to accept. This is preferable to allowing the application process to be repeated from the beginning since that could needlessly cause a new credit report to be obtained. After the reentry screen is displayed, the process ends at 720.
If the status of the application indicates that the application has been accepted, then control is transferred to a step 714 and an offer process is executed. The offer process is described in further detail in
If the status of the application indicates that the application has been rejected, then control is transferred to a step 704 and a rejection process is executed. The rejection process is described in further detail in
Certain credit bureau codes that indicate positive factors that would be inappropriate bases for rejection such as home ownership are mapped by the data structure to a general rejection reason such as “Applicant rejected based on FICO score” or “Applicant rejected based on credit bureau data.” Although such general reasons may be provided to the applicant as a last resort, it is preferred that a more specific reason be given. To that end, a step 806 checks whether any of the FICO reasons have been mapped to any specific rejection reasons. If all of the FICO reasons map only to the general reason, then control is transferred to a step 808.
In step 808, the rejection process begins to attempt to find a more appropriate reason for rejection of the applicant. First, the results of the various attribute tests generated by the Underwriter are obtained. In a step 810, it is checked whether any of the attribute test results map to an appropriate rejection reason. If an attribute test result maps to an appropriate reason, then control is transferred to a step 812 and the attribute reason is assigned as the reason given to the applicant upon rejection. If the attribute test does not map to an appropriate reason, then control is transferred to a step 816 and a general reason is assigned as the reason given to the applicant upon rejection. If, in step 806, it was determined that one or more of the FICO score factors identified by the credit bureau correspond to an acceptable rejection reason other than the general rejection reason, then that reason is assigned as the reason to be given to the applicant in a step 814. Whether or not a specific reason is identified by that above mentioned steps, control is transferred to a step 818 where the reason is displayed to the applicant and the process then ends at 820.
Once an appropriate rejection reason is selected, it is necessary to display the reason to the applicant. In one embodiment, the reason is displayed on a web page along with an acknowledgement button that allows the applicant to acknowledge that he or she has read the rejection message.
It should be noted that other methods of verifying that a rejection has been received are used in other embodiments. For example, in one embodiment, an applet is sent along with the rejection that sends a message back to the credit approval system when the rejection message page is completely downloaded by the applicant. In this manner, the fact that a rejection was delivered to the applicant can be verified without requiring any action by the applicant.
Once the rejection has been sent and acknowledged or not, the rejection or acknowledgement status may be provided to an entity such as FDR for the purpose of generating hard copies of rejection letters and either sending such hard copies as confirmations to all rejected applicants or else, in some embodiments, only sending hard copies of rejection letters to applicants that have not acknowledged an on line rejection.
Accepted applications have an accepted status and they also contain important applicant information supplied by the applicant and obtained from the credit bureau reports that can be used to design a custom account level offer for the applicant. Preferably, multiple offers are presented to the applicant, allowing the applicant to select an offer that includes terms that the applicant desires to accept.
Next, in a step 1004, offer selection criteria are obtained from the credit report object. In one embodiment, the offer selection criteria include FICO score, income and a balance transfer requirement. Offer selection criteria also may include data entered by the applicant. The offer selection criteria also may include other attributes such as time on file. In general, the offer selection criteria are selected from information obtained from the applicant and from the credit bureaus for the purpose of estimating the applicant's risk of default to determine an expectation of future loss as well as an expected future total revolving balance (TRB). In this manner, an appropriate offer may be determined. In one embodiment, the balance transfer requirement is calculated as a selected percentage of the applicant's TRB. As described below, different offer terms may be provided for different balance transfer requirements. As noted above, in other embodiments, other data structures than the application object are used to store this information.
Next, in a step 1006, a set of offers is derived from the credit report data and other applicant information stored in the application object. In a step 1008, the set of offers is displayed. In one embodiment, the offers are derived from the FICO score and income of the applicant, which determine the risk of default, and also from a balance transfer amount specified in the offer. The balance transfer amount may be determined as a percentage of the total revolving balance that the applicant has on all outstanding credit cards in the credit report for the applicant. Both the credit limit offered to the applicant and the interest rate offered to the applicant may vary according to the amount of the total revolving balance that the applicant chooses to transfer to the new account.
In addition offers may present incentives such as frequent flier miles, cash back on purchases, or favorable interest rates.
In a step 1010, the system notes the selected offer and balance transfer amount. Next, in a step 1012, the system obtains the balance transfer amount from the applicant. Preferably, the balance transfer is actually executed while the applicant is on line. The process for obtaining and executing the balance transfer in real time on line is described further in
In different embodiments, the system uses different methods of determining the terms of the offer extended to the applicant based on the information derived from the credit report.
In one embodiment it is desired that a dollar charge off rate be kept within a determined range for different applicants. To accomplish this, it is desirable to extend smaller amounts of credit to applicants with a higher probability of defaulting. It is also useful to extend different amounts of credit based on a total outstanding balance transferred by the applicant since the balance transfer influences the likely future total outstanding balance of the account. Conventional offer systems have been able to extend offers to applicants with credit limits that are controlled by the applicant's predicted average dollar loss. However, prior systems have not been able to extend credit and determine a credit limit based on a predicted total outstanding balance for the client because they have failed to be able to present offers and condition the acceptance of the offers in real-time on a balance transfer made by the applicant.
Next, in a step 1026 the system determines one or more balance transfer amounts based on the total revolving balance that the applicant has in various other credit card accounts. In one embodiment, the balance transfer amounts are calculated based on different percentages of the total revolving balance determined from all of the applicant's accounts found in the credit report. Then, in a step 1028, the system calculates for each total balance transfer amount choice that will be presented to the applicant, a predicted estimated revolving balance for the future that the applicant would be expected to maintain. The estimated total revolving balance may be equal to the balance transfer amount or may be a function of the balance transfer amount. In one embodiment, the estimated total revolving balance does not depend on the balance transfer amount. In one embodiment, four possible percentages of the applicant's total revolving balance as determined by the credit report are presented to the applicant. Those choices are none of the balance, one-third of the balance, two-thirds of the balance, and the full balance. Depending on which of those amounts is selected by the applicant, the system calculates a predicted total revolving balance for the future. Then, in a step 1030, the credit limit for the applicant is set to achieve a target dollar charge off rate based on the amount of the total revolving balance that the applicant elects to transfer and the risk of default. The process then ends at 1032.
The process described in
In the embodiment described above, separate tables are prepared for applicants of different incomes. In addition, separate tables may also be prepared for applicants having other different characteristics such as time on file for the applicant. It should be noted that the tabular representation of the data is presented as an example only and the data may be represented in many ways including in three-dimensional or four-dimensional arrays, linked lists or other data representations optimized for a particular system. By allowing the account credit limit to be a function of FICO score, balance transfer, and income, a credit limit may be selected for each individual account that enables the dollar charge off rate for all applicants to be controlled.
Another component of the offer granted to the applicant that may be varied based on the balance transfer selected is a teaser rate or annual rate. A teaser rate is an interest rate that is temporarily extended to the applicant either on the amount transferred or on the amount transferred and purchases made for a certain period of time. The teaser rate is intended to incent the applicant to transfer a greater balance to a new account. In one embodiment, the teaser rate is determined based on the percentage of the applicant's total revolving balance that the applicant elects to transfer. Thus, the amount transferred by the applicant controls not only the applicant's credit limit but also determines a teaser rate extended to the applicant.
It should be noted that in some embodiments, the accounts chosen for display by the underwriter are selected in a manner to facilitate a simpler balance transfer. For example, the largest account balances may be displayed first so that amounts may be efficiently transferred to meet the required transfer. Also, a group of balances to transfer may be presented to the applicant by highlighting certain accounts.
Next, the applicant is given an opportunity to indicate a balance transfer by selecting one of the accounts and indicating the amount to be transferred. It should be noted that the applicant in this manner does not need to provide account information to execute a balance transfer. If a transfer is indicated, control is transferred to a step 1306 and the amount of the user balance transfer is obtained. Next, in a step 1307, it is determined whether the sum of the balance transfers is greater than or equal to the required transfer amounts for the offer selected by the applicant. If the amount is not greater than or equal to the required-transferred amount, then control is transferred back to step 1304 and the applicant is given an opportunity to select further balances to transfer. If the amount of the balance transfers is greater than or equal to a required transfer amount, then control is transferred to a step 1308 and the system requests final confirmation from the applicant of the balance transfers. If it is determined in a step 1310 that a confirmation of the balance transfer has been received, then control is transferred to a step 1312 and the balance transfers are executed. The process ends at 1314.
If in step 1304, it is determined that the applicant has elected to exit the balance transfer screen instead of indicating a balance transfer, or if it is determined in step 1310 that the applicant elects not to confirm the balance transfer amounts selected, then control is transferred to a step 1316 and the applicant is returned to the offer selection screen so that the applicant will have an opportunity to select another offer that either does not require a balance transfer or requires less of a balance transfer. The process then ends at 1314.
A real time credit approval system has been described herein primarily for the purpose of determining whether a credit card should be issued to an applicant. Software written to implement the system may be stored in some form of computer-readable medium, such as memory or CD-ROM, or transmitted over a network via a carrier wave in the form of Java® applets, other forms of applets or servlets, and executed by a processor. The system may be implemented on a PC or other general purpose computer known in the computer art.
It should be recognized that the system described may also be used for the purpose of granting credit to an applicant for the purpose of making a single transaction. In such a system, a transaction is interrupted and the application for credit is made. Based on the real time approval decision made, credit may or may not be granted for the purpose of completing the transaction.
Referring now to
In one embodiment, a method of offering credit to an applicant includes determining a plurality of offers using information about the applicant. A displayed offer is displayed and a withheld offer is withheld. An indication that the displayed offer is unacceptable is received and the withheld offer is displayed.
In one embodiment, a method of offering credit to an applicant includes determining a plurality of offers using information about the applicant. A displayed offer is displayed and a plurality of withheld offers are withheld. An indication that the displayed offer is unacceptable is received including an indication that an attribute is unacceptable. A selected withheld offer is selected using the attribute that is unacceptable and the selected withheld offer is displayed.
In one embodiment, a method of offering credit to an applicant includes determining a plurality of offers using information about the applicant. A displayed offer is displayed and a plurality of withheld offers are withheld. An indication that the displayed offer is unacceptable is received including an indication that a plurality of attributes are unacceptable. A primary unacceptable attribute is determined and a selected withheld offer is selected using the primary unacceptable attribute. The selected withheld offer is displayed.
In one embodiment, a method of offering credit to an applicant includes determining an offer using information about the applicant and displaying the offer to the applicant. An indication that the displayed offer is unacceptable is received. An attribute of the offer that is unacceptable is determined. In the event that the unacceptable attribute is the amount of the credit limit; the credit limit is recalculated for the applicant.
In one embodiment, a method of offering credit to an applicant includes determining a first offer using information about the applicant and displaying the first offer to the applicant. A chat interface is activated between the applicant and a customer service agent. A second offer for the applicant is determined based on chat between the applicant and the customer service agent and the second offer is displayed to the applicant.
In one embodiment, an application server for providing a counter offer of credit includes an applicant interface configured to receive applicant data from an applicant browser and to communicate an offer of credit to the applicant and the counter offer of credit to the applicant. A processor is configured to determine the offer of credit based on the applicant data and the counter offer of credit based on an unacceptable attribute of the first offer of credit and an agent interface is configured to receive the unacceptable attribute from an agent.
In one embodiment, a chat server for providing a counter offer of credit includes an applicant interface configured to receive chat from an applicant. An agent interface is configured to receive chat from an agent and an unacceptable attribute determined from the chat from the applicant. An application server interface is configured to send the unacceptable attribute and to receive a counter offer.
In one embodiment, an applicant client for obtaining a counter offer of credit includes an application server interface configured to send applicant information and to receive and offer of credit and a counter offer of credit. A chat server interface is configured to be activated upon an indication that the offer of credit is not acceptable.
In one embodiment, an applicant interacts with a web server and receives a web page containing offers of credit that may be accepted by the applicant. At any point during the interaction with the web server, an online chat button or process may be activated that sends an applicant ID to a chat server and opens a chat window so that the applicant can receive help. In one embodiment, the help takes the form of the applicant describing why a displayed offer is unacceptable and a counter offer being generated for the applicant.
Web server 1102 provides a web page 1104 to a browser 1106. Typically, the web server and browser communicate over the Internet using HTTP. Web page 1104 is shown for the purpose of illustration as an offer web page that includes three offers made to the applicant for a credit card as well as an on-line chat button that may be activated by the applicant to obtain help or to discuss the offers. Other web pages provided by the web server include forms that the applicant fills out to provide information so that a credit report may be obtained and an offer of credit generated based on the applicant's personal information.
An online application process for a credit card is described in detail in U.S. patent application Ser. No. 09/185,201, entitled: “Method And Apparatus For Real Time Online Credit Approval”, filed Nov. 17, 1998, which was previously incorporated by reference; and U.S. patent application Ser. No. 09/185,878, entitled: “Method And Apparatus For A Verifiable Online Rejection Of An Applicant For Credit”, filed Nov. 17, 1998, which was previously incorporated by reference; and U.S. patent application Ser. No. 09/185,000, entitled: “Method And Apparatus For An Account Level Offer Of Credit And Real Time Balance Transfer”, filed Nov. 17, 1998 which was previously incorporated by reference.
It should be noted that the process described herein will refer to the online credit application as being an application for a credit card. The process can also be applied to other offers of credit including an offer of instant credit for the purpose of consummating a single pending online transaction. In addition, the system and processes disclosed herein may be applied to other types of business transactions over the Internet. However, the particular architecture and processes described are especially useful for processing online credit card applications and the benefit of their application to online credit card applications is particularly strong.
Web server 1102 and browser 1106 continue to interact in a standard fashion with web pages being provided by web server 1102 and the applicant filling out information as needed. At some point, an applicant may activate the online chat button included on the web page and a chat window 1104a opens up for the chat application and a connection is established with a chat server 1108. As is described further below, the chat window is opened and the connection with chat server 1108 may be initiated by events other than just the activation of the online chat button. Chat server 1108 implements a standard chat environment such as the chat environment available from e-share. Other chat environments may be used that include the ability to pass a variable to the chat server from the browser.
The various servers shown in
In some embodiments, the chat process is initiated when the applicant cancels out of the application. In other embodiments, the chat process is initiated when the applicant lingers on a page for an amount of time that exceeds a threshold. In other embodiments, the chat process is initiated when the applicant's response to a request for information is somehow inadequate. For example, it may be detected that the answers provided by the applicant are incomplete or in the wrong form. The chat process may be initiated for the purpose of providing the applicant more detailed instructions or pointing out to the applicant the information that is required to complete the application.
In addition to opening the chat connection to chat server 1108, browser 1106 also sends the applicant identifier to the chat server. The chat server then uses the applicant identifier to access information about the application in the application database. It should be noted that the applicant identifier may be used as an application identifier in circumstances such as would be expected for an online credit card application where there is one and only one application per applicant. In other embodiments, an application identifier that is unique for each application is assigned and used. In this description, wherever an applicant identifier is mentioned, an application identifier could also be used.
Sending the applicant identifier to the chat server instead of sending the current web page or other information to the chat server is preferable from a security standpoint because the applicant identifier can only be used to obtain information about the application by accessing application database 1103. In addition, preferably, the applicant identifier is encrypted, adding a further level of security.
In the embodiment shown, chat server 1108 does not have a direct link to the application database 1103. Chat server 1108 is connected to a customer service agent 1110. Customer service agent 1110 handles the chat session, responding to requests made by the applicant. Other customer service agents 1112 and 1114 are also standing by to handle other chat sessions generated by chat server 1108. In one embodiment, requests made to the chat server are queued and the next available customer service agent is assigned to the first chat session request found in the queue.
Customer service agent 1110 is connected to a counter offer server that is connected to application database 1103. By passing the applicant identifier from the chat server to the customer service agent to the application database through the counter offer server, information about the applicant can be obtained from the application database. Connections from the customer service agent to the counter offer server and from the counter offer server to the application database may be made over the Internet or may be a dedicated secure connection.
In the embodiment shown, which is adapted specifically for implementing a counter offer strategy as is described below, a separate web server 1102 and counter offer server 1120 are shown. This divides the processing demand generated by normal communication with a browser from the processing demand generated by interaction initiated by chat with a customer service agent. This architecture is particularly useful since the two types of traffic are isolated. In other embodiments, the functions of the web server and the counter offer server are performed by a single application server. Dashed box 1122 represents a single application server that may include both the web server and the counter offer server. In general, the term application server is used to describe either the web server and counter offer server operating collectively or to describe a single server performing both the function of the web server and the counter offer server.
Additionally, in a system where a counter offer is not generated, counter offer server 1120 may be referred to as a customer service agent server or some other term describing its primary function. The important point is that both the web server and the counter offer server both access the application data base to obtain information about the status of the application. In addition, both the web server and the counter offer server may write data to the application database in some embodiments. The common access to the application database enables the customer service agent to obtain information about the status of the application using the applicant identifier received through the chat server and also allows the customer service agent to alter the status of the application based on information received from the applicant through the chat server by sending that information to the counter offer server for posting to the application database.
Thus, an applicant provides information to database 1103 via the Internet using web pages in a standard manner. In addition, the applicant may communicate via chat with a customer service agent who also is connected to the application database and may change the state of the application according to information received by the applicant via chat. In the embodiment shown, the customer service agent interacts with a special purpose counter offer server that uses the information provided by the applicant to determine a counter offer using information in the application database. The counter offer is stored in the application database and provided to the applicant's browser via the web server. As noted above, the counter offer server and the web server may be implemented on a single machine referred to as the application server. The various processes operating on the application server, the chat server and the browser are described below for the purpose of illustrating how the chat window may be activated and a counter offer generated for the applicant.
In an embodiment where a counter offer is generated by the customer service agent, a display is provided showing various offer terms that the applicant may indicate are not acceptable in the chat between the applicant and the customer service agent. The customer service agent may check one or more of the terms and the terms checked by the customer service agent are sent to the counter offer server to be used in generating a counter offer. The terms or attributes of the offer that the applicant considers to be unacceptable are obtained in a step 1712 and the initial process for receiving applicant information and providing information to the counter offer server ends at 1714.
It should be noted that a number of different methods of obtaining the unacceptable terms from the applicant may be used. In one embodiment, as described above, a set of offer terms are shown to the customers service agent and the customer service agent selects terms identified by the applicant in chat that are unacceptable. In other embodiments, a display of terms is provided to the applicant and the applicant picks the unacceptable terms with the aid of the customer service agent. In yet another embodiment, the chat generated by the applicant is automatically analyzed by a program which generates the list of unacceptable terms for the counter offer server.
Many different methods may be used by the counter offer server to generate a counter offer based on attributes or terms identified by the applicant as being unacceptable. In one embodiment, a number of potential offers are identified based on the applicant information provided and an assessment of the risk associated with extending credit to the applicant. Some of the generated offers are withheld while others are displayed to the applicant. A number of schemes may be used to decide which offers are displayed and which offers are withheld. Some methods may include a statistical selection or a selection according to a marketing scheme designed to increase the rate of acceptance. It may also be the case that the best offer is withheld and kept in reserve to use as a counter offer. In general, certain potential offers are withheld.
The identification by the applicant of an unacceptable term is used by the counter offer server to identify a better offer for the counter offer. In one embodiment, offer strategies are identified and the counter offer is identified by simply looking up an offer strategy associated with the applicant and the identified unacceptable term. In one embodiment, an offer strategy may include a set of offers shown to the applicant as well as offers that are not displayed and that correspond to various unacceptable terms. When an unacceptable term is identified, the offer corresponding to the strategy and the unacceptable term is used as the counter offer.
In some embodiments, the counter offer strategy is dependent on characteristics of the applicant. For example, the applicant may be classified as a “surfer” or “non-surfer”. A “surfer” is a person who shifts or surfs balances among credit cards, taking advantage of low teaser rates. A determination that an applicant is a surfer is made based on an analysis of the applicant's credit report. A counter offer strategy designed for such an applicant may adopt the strategy of extending the period of an introductory rate if requested by the applicant, but requiring the applicant to make a certain number of purchases or not transfer the balance for a certain period of time.
In general, added terms and conditions such as purchase requirements or a length of time that a balance may not be transferred from the card may be added to counter offers for the purpose of creating a perceived barrier to receive the counter offer. Such a barrier or condition prevents the applicant from deciding that the first offer should always be rejected. In some embodiments, the conditions are determined based on characteristics of the applicant. As described above, surfers may receive balance transfer time restrictions.
In addition to selecting a withheld offer based on a pre-determined offer scheme, the counter offer server may also recalculate a customized offer based on the identification of an unacceptable term and an actual requested term by the applicant. For example, the applicant may express that the credit limit is too low, either for a desired balance transfer that the applicant wants to make or new purchases. The amount of the credit limit minus the amount of the balance transfer is referred to as the amount of credit that is “open to buy”. The information sent to the counter offer server may include a requested credit limit and a requested balance transfer amount. From that information, the counter offer server can determine that the offer credit limit is too low either for the balance transfer requested or for the amount that the applicant wants open to buy. To minimize risk, it is desirable that the credit limit be as low as possible. Therefore, it is desirable not to simply select a withheld offer with a higher credit limit, but instead to customize an offer that conforms to the applicant's request but does not exceed the applicant's request.
Accordingly, a new credit limit may be calculated that incorporates the requested balance transfer and the amount that the applicant wants open to buy. The calculated new offer is of course checked versus the risk profile of the user and it is verified that the higher credit limit is appropriate for the user. Any of the various techniques well known in the art of assigning credit may be used to assess the risk of the applicant and determine an appropriate upper credit limit. Significantly, the counteroffer in the case of a requested higher credit limit is specifically customized for the applicant based on what the applicant requests. In general, any counter offer provided is based on the applicant's identification of an unacceptable term. In some embodiments, if no counter offer is available that improves an unacceptable term identified by the applicant, then a message is returned to the applicant either directly or through the chat interface that indicates that no counter offer can be provided at that time. For example, in one embodiment, the offer strategy may include an offer with the best annual percentage rate available in the set of offers initially displayed to the applicant. In such a case, if the applicant identifies the annual percentage rate as the unacceptable term, then no counter offer improving that term can be generated.
If the credit limit is too low for a requested balance transfer, then control is transferred to a step 11070 and the credit limit is adjusted based on the balance transfer requested if allowed by the credit line assignment being used. After the credit limit is adjusted in steps 11060 or 11070, the counter offer is defined and the counter offer determination process ends at 11080. Whether a precomputed offer is determined for display in 11030 or the credit limit is recomputed in step 11060 or 11070, if no better offer can be generated, then a message noting that no better offer can be generated is sent either to the chat server or to the applicant directly.
Once a counter offer has been defined or it has been determined that no counter offer that improves the unacceptable terms can be generated, the applicant is notified of the counter offer terms. In different embodiments, notification may be accomplished in various ways. For example, in one embodiment, a new offer page is generated in the application server based on data written to the application database by the application server. In the embodiment where the application server is split into a web server and a counter offer server, the counter offer server writes data to the application database and the web server generates a counter offer page based on the data written to the application database. In addition, the application server also provides information to the chat server indicating what counter offer, if any, has been generated. The customer service agent then discusses the counter offer with the applicant via the chat interface. In order to view the counter offer page generated by the web server, the applicant is asked to refresh his browser. Refreshing the browser causes the offer page to be requested from the web server and the web server responds with the counter offer page. In one embodiment, a button labeled “view offer” is provided on the displayed page. When the button is selected, the page is downloaded again and any changes are then viewed by the user.
In other embodiments, the displaying of the counter offer page to the applicant is handled somewhat differently. In one embodiment, the chat server enables the display of the page through the applicant's browser automatically, without requiring the applicant to refresh the screen. This can be accomplished in a variety of ways. In one embodiment, the chat server writes a variable to a memory location that the browser checks periodically. When the browser checks the location and finds a variable indicating that the counter offer page should be downloaded, the browser automatically refreshes itself. The applet that enables the browser to check the location and refresh itself may be used in some cases but not others. When such an applet is not used, the process of instructing the applicant through the chat interface to refresh his own browser or to select a button to view the offer may be implemented.
In some embodiments, the new offer is confirmed in the database concurrent with it being displayed to the customer service agent. Then, whenever the applicant's browser refreshes, the counter offer will be displayed. In some embodiments, it is desired that the display of the counter offer not be enabled until customer service agent has an opportunity to chat with the applicant about the new offer and confirm that display is appropriate.
A system and method for activating a chat interface with a customer service agent that has access to information about an application for credit has been described. In one embodiment, the chat interface is used to obtain information about why an applicant is rejecting an offer of credit and to identify unacceptable terms. Those unacceptable terms are communicated to a counter offer server and the counter offer server generates a new offer that improves the unacceptable term. The new offer is communicated to the applicant using the chat interface and a web page showing the new offer with an opportunity to accept the offer is displayed to the applicant when the applicant's browser is refreshed.
Although the foregoing invention has been described in some detail for purposes of clarity of understanding, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. It should be noted that there are many alternative ways of implementing both the process and apparatus of the present invention. Accordingly, the present embodiments are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.
1. A method implemented on one or more computers for providing approval of credit over a network, comprising:
- receiving via a computer network applicant data from an applicant for a credit application;
- prior to obtaining credit report data from a credit bureau for the applicant, determining compliance of the applicant with one or more one or more requirements, the one or more requirements comprising a duplicate check comprising comparing one or more elements of the applicant data with data previously submitted by previous applicants;
- transmitting electronically to applicant a determination to decline approval for credit to the applicant if the applicant data fails to meet one or more of the one or more requirements;
- if the applicant is not declined: processing one or more elements of the applicant data into a predetermined electronic form; electronically transmitting the one or more elements of applicant data in the predetermined electronic form to a credit bureau for obtaining a credit report from a credit bureau for the applicant; receiving electronically the credit report data from a credit bureau for the applicant; and determining, without intervention of a human, whether to approve or reject the applicant based for credit based at least in part on the credit report data; and
- if it is determined to approve the applicant for credit, communicating electronically to the applicant that the applicant has been approved for credit.
2. The method of claim 1, wherein the determination of whether to approve the applicant for credit occurs in real time.
3. A system for providing approval of credit over a network implemented on one or more computer processors, comprising:
- an application engine configured to obtain applicant data from an applicant;
- an address parser configured to format the applicant data into a form suitable for directly obtaining a credit report from a credit bureau for the applicant; and
- an underwriter module configured to: determine whether to continue to process or to reject the applicant based on the applicant data prior to obtaining a credit report from a credit bureau for the applicant, said determining whether to continue to process comprising: checking based on the applicant data entered by the applicant and prior to obtaining a credit report whether some or all of the applicant data is a duplicate of applicant data previously entered by the applicant; and declining the applicant, in the event it is determined that the applicant data is a duplicate of applicant data previously entered by the applicant after a predetermined duplication cutoff date; and obtain, in the event it is determined based on the applicant data to process the applicant, a credit report having credit report data from a credit bureau for the applicant; determine whether to accept the applicant using the credit report data; and, communicate, if it is determined to accept the applicant for credit, to the applicant that the applicant has been approved for credit.
4. The system of claim 3, wherein the determination of whether to accept the applicant for credit occurs in real time.
5. A computer readable medium having program code embodied therein for providing approval of credit over a network, which, when read by a computer, causes the computer to perform a process, the program code comprising:
- program code for receiving via a computer network applicant data from an applicant for a credit application;
- program code for determining, prior to obtaining credit report data from a credit bureau for the credit application, compliance of the applicant data with one or more one or more requirements, the one or more requirements comprising a duplicate check comprising comparing automatically one or more elements of the applicant data with data previously submitted by applicants;
- program code for declining approval for credit to the applicant if the applicant data fails to meet one or more of the one or more requirements;
- program code for, if the applicant is not declined: processing automatically one or more elements of the applicant data into a predetermined electronic form; electronically transmitting the one or more elements of applicant data in the predetermined electronic form to a credit bureau for obtaining a credit report from a credit bureau for the applicant; receiving electronically a credit report having credit report data from a credit bureau for the applicant; and determining automatically, without intervention of a human, whether to accept or reject the applicant for credit based at least in part on the credit report data; and
- program code for communicating, if it is determined to accept the applicant for credit, to the applicant that the applicant has been approved for credit.
6. A computer system, the computer system comprising:
- means for receiving via a computer network applicant data from an applicant for a credit application;
- means for determining, prior to obtaining a credit report from a credit bureau for the credit application, compliance of the applicant with one or more requirements, the means for determining compliance with one or more requirements including means for determining whether applicant has previously made application for credit within a predetermined period of time;
- means for processing automatically, if the applicant is in compliance with the one or more requirements, one or more elements of the applicant data into a predetermined electronic form and automatically electronically transmitting to a credit bureau for obtaining a credit report from a credit bureau for the applicant;
- means for receiving electronically a credit report having credit report data from a credit bureau for the applicant;
- decision means for determining, without intervention of a human, whether to accept or reject the applicant for credit based at least in part on the credit report data; and
- means for communicating automatically with the applicant the determination of the decision means.
7. The computer system of claim 6, wherein the decisions means makes the determination in real time.
Filed: Oct 31, 2007
Publication Date: Oct 30, 2008
Inventors: Jeremy R. Lent (Corte Madera, CA), Mary Lent (Corte Madera, CA), Eric R. Meeks (San Francisco, CA), Yinzi Cai (Fremont, CA), Timothy J. Coltrell (Danville, CA), David W. Dowhan (Mountain View, CA)
Application Number: 11/932,498
International Classification: G06Q 40/00 (20060101);