Lapse predicting tool and scoring mechanism to triage customer retention approaches

• US 10,096,068 B1
• Filed: 12/19/2014
• Issued: 10/09/2018
• Est. Priority Date: 12/23/2013
• Status: Active Grant

• Current Assignee: Massachusetts Mutual Life Insurance Company
• Original Assignee: Massachusetts Mutual Life Insurance Company
• Inventors: Ross, Gareth, Walker, Tricia
• Primary Examiner(s): Chang, Edward
• Application Number:
  US14/577,402
• Time in Patent Office: 1,390 Days
• Field of Search: 705 4
• CPC Class Codes: G06Q 30/01 Customer relationship services
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

Disclosed here is a triaging system, including a value engine, an analytical engine, a scoring engine, and databases storing internal, external, and retention value data. A value engine may be configured to receive information and determine a value associated with a policy; an analytical engine may be configured to receive information related to one or more factors associated with a policyholder, as well as actuarial relationships between the one or more factors; and a scoring engine may be configured to receive information associated with a policyholder from one or more value engines and analytical engines, and may output a scalar score associated with the retention value of the policyholder. The system may determine the value of retention associated with the policyholder and use this value to build and store a client score, where the client score may be a scalar representation of the value of retaining business with the policyholder.

Claims

1. A computer-implemented method comprising:

querying, by a server, one or more internal and external databases comprising data of a plurality of customers and users to retrieve information corresponding to each of the plurality of customers and users, wherein the one or more external databases comprise unstructured data;

determining, by the server, a cash flow value for a policy dataset corresponding to each of the plurality of customers;

generating, by the server, a heuristic computer-generated model based at least on data corresponding to a performance value of one or more users of the plurality of users, the heuristic computer-generated model configured to utilize a support vector machine algorithm to emulate a resolution pattern of the plurality of users having the performance value greater than a pre-determined threshold value;

continuously querying, by the server, the one or more internal databases and the external databases to determine any modification of the information corresponding to the one or more users having the performance value greater than the pre-determined threshold value;

continuously updating, by the server, the heuristic computer-generated model based on a modification of the information corresponding to each of the one or more users having the performance value greater than the pre-determined threshold value, whereby the heuristic computer-generated model improves quality of an output of the resolution pattern emulation with each instance of updating the heuristic computer-generated model;

re-generating, by the server, the policy dataset for each of the one or more customers by applying the policy dataset to the updated heuristic computer-generated model;

determining, by the server, a value of retention by comparing the determined cash flow value associated with each of the one or more customers and a value ascribed to each of the one or more customers after re-generating the policy dataset;

generating, by the server, a client score for each of the one or more customers based upon the value of retention for each of the one or more customers;

receiving, by the server in connection with a call center, a call from a first customer of the one or more customers;

identifying, by the server, the first customer based on a unique identifier associated with the call; and

during the call and while a representative is interacting with the first customer, presenting, by the server, the client score corresponding to the first customer identified based on the unique identifier to the representative.

2. The method according to claim 1, wherein the client score is a scalar representation of the value of retention of each of the one or more customers.

3. The method according to claim 1, wherein generating comprises emulating resolution patterns of the one or more users.

4. The method according to claim 3, wherein the one or more users are highest performing one or more users in the plurality of users.

5. The method according to claim 1, wherein the heuristic computer-generated model is based upon a linear relationship between factors and risk classes.

6. The method according to claim 5, wherein the factors include at least one of age, current health, and occupation of the one or more customers.

7. The method according to claim 1, wherein the heuristic computer-generated model is based upon a nonlinear relationship between factors and risk classes.

8. The method according to claim 7, wherein the factors include at least one of age, current health, and occupation of the one or more customers.

9. The method according to claim 1, wherein re-generating the policy dataset based on the heuristic computer-generated model generates a resolution, and wherein the resolution comprises a risk classification.

10. The method according to claim 1, wherein re-generating the policy dataset based on the heuristic computer-generated model generates a resolution, and wherein the resolution comprises a risk of loss assessment.

11. The method according to claim 1, further comprising periodically updating, by the server, a set of standards to improve the heuristic computer-generated model.

12. The method according to claim 1, wherein the value of retention is proportional to a difference of the value ascribed to each of the one or more customers.

13. The method according to claim 1, wherein determining the flow value associated with each of the one or more customers comprises quantifying an amount of distress of each of the one or more customers.

14. The method according to claim 13, wherein the amount of distress of each of the one or more customers is derived from data selected from a group consisting of a credit score, historical lapse data set, census data, and customer support logs.

15. The method according to claim 13, wherein the amount of distress of each of the one or more customers is derived from information regarding a source of income of each of the one or more customers and a state of a market from which each of the one or more customers obtains income.

16. A computer executed method comprising:

determining, by a server, a cash flow value associated with a product dataset, the product being issued to a customer based on execution of a heuristic computer-generated model, wherein the heuristic computer-generated model is generated based at least from data collected from a database of one or more users of a plurality of users having a performance value greater than a pre-determined threshold value, wherein the computer-generated model is configured to utilize a support vector machine algorithm to emulate the one or more users having the performance value greater than a pre-determined threshold value;

wherein the server periodically update the heuristic computer-generated model based on modification of information regarding the one or more users having the performance value greater than the pre-determined threshold value, wherein the server periodically queries the database to determine modification information, whereby the heuristic computer-generated model improves quality of an output of the emulation with each instance of updating the heuristic computer-generated model;

re-generating, by the server, the product dataset for the customer by applying the policy dataset on the updated computer-generated model;

determining, by the server, a value of retention by comparing the determined flow value associated with the customer and a value ascribed to the customer after re-generating the policy dataset;

generating, by the server, a client score for the customer based upon the value of retention for the customer;

receiving, by the server in connection with a call center, a call from the customer;

identifying, by the server, the customer based on a unique identifier associated with the call; and

during the call and while a representative is interacting with the customer, presenting, by the server, the client score corresponding to the customer identified based on the unique identifier to the representative.

17. A system comprising:

a server comprising;

a value engine of the server configured to query one or more internal and external databases comprising data of a plurality of users and customers to retrieve policy information corresponding to each of the plurality of customers and determine a cash flow value for a policy dataset held by each of the plurality of customers, wherein the one or more external databases comprises unstructured data;

an analytical engine of the server configured to generate a heuristic computer-generated model based at least on data corresponding to a performance value of one or more users of the plurality of users, the heuristic computer-generated model configured to utilize a support vector machine algorithm to emulate the one or more users having the performance value greater than a pre-determined threshold value and periodically update the computer-generated model based on modification of information regarding the one or more users having the performance value greater than the pre-determined threshold value, wherein the analytical engine of the server periodically queries the internal database and the external database to determine modification information, whereby the heuristic computer-generated model improves quality of an output of the emulation with each instance of updating the heuristic computer-generated model;

a triaging engine of the server configured to re-generate the policy dataset for each of the one or more customers based by applying the policy dataset to the updated computer-generated model, determine a value of retention by comparing the determined flow value associated with each of the one or more customers and a value ascribed to each of the one or more customers after re-generating the policy dataset, generate a client score for each of the one or more customers based upon the value of retention for each of the one or more customers;

receive a call from a first customer of the one or more customers;

identify the first customer based on a unique identifier associated with the call; and

during the call and while a representative is interacting with the first customer, present the client score corresponding to the first customer identified based on the unique identifier to the representative.