Methods and systems for ranking leads based on given characteristics

• US 10,242,068 B1
• Filed: 12/19/2014
• Issued: 03/26/2019
• Est. Priority Date: 12/31/2013
• 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): Singh, Gurkanwaljit
• Application Number:
  US14/576,774
• Time in Patent Office: 1,558 Days
• Field of Search: None
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 3/048 Interaction techniques based on graphical user interfaces [GUI]
• CPC Class Codes: G06F 7/00 Methods or arrangements for processing data by operating upon the order or content of the data handled logic circuits H03K19/00
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Systems and methods for ranking and appraising leads according to their quality are disclosed. The disclosed method operates within a systems'"'"' architecture configured to rank and auction leads. One or more client computing devices allow an agent to request for the ranking and appraisal of a set of leads. Following the request, a ranking module implements one or more software modules for assessing the quality of each lead and ranks the set of leads according to a quality criterion. Next, a price modeling module defines a floor price for the set of leads. The solutions derived from the software modules are stored in an internal database where they are available to other software modules operating within the system architecture for ranking and auction leads. In some embodiments, the ranking results are used for tracking results and developing insight about the value of leads and the effectiveness of the ranking method.

Claims

1. A computer-implemented method comprising:

scanning, by a first processor of a computer comprising at least two processors, one or more social networking web documents associated with one or more leads within one or more databases and extracting lead information from the one or more social networking web documents, the lead information comprising one or more characteristic values;

classifying, by the first processor of the computer, the lead information into categories of lead information based on the one or more characteristic values of the lead information;

upon receiving from a computer of an agent, a selection of a first category of lead information and an attribute of the first category, filtering, by the first processor of the computer, the lead information to obtain a set of filtered lead information comprising only a subset of leads containing the attribute;

assigning, by the first processor of the computer, a score to each attribute associated with each lead from the filtered lead information based on a measure of how each attribute satisfies a predetermined set of criteria;

executing, by the first processor of the computer, a machine-learning algorithm technique to calculate a quality score for each lead, wherein the machine-learning algorithm technique is configured to calculate the quality score of each lead by computing a mean score for each lead based on a learning dataset comprising the each scored attribute;

while the first processor of the computer is executing the machine-learning algorithm, iteratively updating, by a second processor of the computer, the learning dataset based on modified data associated with each lead having a score for each attribute greater than the predetermined set of criteria;

periodically querying, by the second processor of the computer, the one or more databases to receive inputs on modified data associated with each lead and, in an event that the computer determines that the score of the attributes associated with each lead is changed, adjusting, by the second processor of the computer, the learning dataset;

ranking, by the first processor of the computer, each lead based on their corresponding quality score, wherein the computer ranks each lead in order of their corresponding implied quality score and a propensity to close a transaction, wherein the computer determines the propensity to close the transaction for each lead based on the one or more characteristic values associated to each lead;

and

updating, by the first processor of the computer, a graphical user interface of the computer of the agent with auction information comprising the ranked leads and the quality score.

2. The method according to claim 1, wherein the lead information comprises at least one of:

identity, age, ethnicity, place of residence, number of dependent persons, identity of dependent persons, expenditure, savings, approximate market value of assets and their composition, education, professional situation, contact information, shopping preferences, travel preferences, hobbies, social activities, active lifestyle, online behavioral patterns, medical and health records.

3. The method according to claim 1, wherein the categories of lead information comprises at least one of:

geography, demographics, asset value, and credit score.

4. The method according to claim 1, wherein the set of attributes associated with the lead from the filtered lead information comprises at least one of:

economic stability, credit card usage, liquid assets, education level, discretionary spending, and retirement assets.

5. The method of claim 1, wherein the lead a potential new customer.

6. The method of claim 1, wherein the lead is a group of potential new customers.

7. The method of claim 1, wherein computing the mean score for each lead based on the each scored attribute is based on at least one of:

naive bayes, logistic regression, support vector machine and random forest.

8. The method of claim 1, wherein the criteria is based on at least one of:

a company standard, and agent preference, the selection from a computer of an agent of a first category of lead information and an attribute of the first category.

9. The method of claim 1, wherein the quality score additionally weighted based on at least one of:

a company standard, and agent preference, the selection from a computer of an agent of a first category of lead information, and an attribute of the first category.

10. The method of claim 9, wherein additionally weighing is based on at least one of:

multiplying the quality score by a standard factor, adding a standard factor to the quality score, dividing the quality score by a standard factor, and subtracting a standard factor to the quality score.

11. A system for ranking leads, the system comprising:

a computer having a first and a second processor;

a memory containing a program that, when executed by the first and the second processor, is configured to perform an operation comprising;

scanning, by the first processor, one or more social networking web documents associated with one or more leads within one or more databases and extracting lead information from the one or more social networking web documents, the lead information comprising one or more characteristic values;

classifying, by the first processor, the lead information into categories of lead information based on the one or more characteristic values of the lead information;

upon receiving from a computer of an agent, a selection of a first category of lead information and an attribute of the first category, filtering, by the first processor, the lead information to obtain a set of filtered lead information comprising only a subset of leads containing the attribute;

assigning, by the first processor, a score to each attribute associated with the leads from the filtered lead information based on a measure of how each attribute satisfies a predetermined set of criteria;

calculating, by the first processor executing a machine-learning algorithm technique, a quality score for each lead, wherein the machine-learning algorithm technique is configured to calculate the quality score of each lead by computing a mean score for each lead based on a learning dataset comprising the each scored attribute;

iteratively updating, by the second processor, the learning dataset based on modified data associated with each lead having the score for each attribute greater than the predetermined set of criteria;

periodically querying, by the second processor, the one or more databases to receive inputs on modified data associated with each lead and, in an event that the computer processor determines that the score of the attributes associated with each lead is changed, adjusting, by the computer processor, the learning dataset;

ranking, by the first processor, each lead based on their corresponding quality score, wherein the computer processor ranks each lead in order of their corresponding implied quality score and a propensity to close a transaction and determines the propensity to close the transaction for each lead based on the one or more characteristic values associated to each lead;

and

updating, by the first processor, a graphical user interface of the computer of the agent with auction information comprising the ranked leads and the quality score.

12. The system according to claim 11, wherein the lead information comprises at least one of:

identity, age, ethnicity, place of residence, number of dependent persons, identity of dependent persons, expenditure, savings, approximate market value of assets and their composition, education, professional situation, contact information, shopping preferences, travel preferences, hobbies, social activities, active lifestyle, online behavioral patterns, medical and health records.

13. The system according to claim 11, wherein the categories of lead information comprises at least one of:

geography, demographics, asset value, and credit score.

14. The system according to claim 11, wherein the set of attributes associated with the lead from the filtered lead information comprises at least one of:

economic stability, credit card usage, liquid assets, education level, discretionary spending, and retirement assets.

15. The system of claim 11, wherein the lead a potential new customer.

16. The system of claim 11, wherein the lead is a group of potential new customers.

17. The system of claim 11, wherein computing the mean score for each lead based on the each scored attribute is based on at least one of:

naive bayes, logistic regression, support vector machine and random forest.

18. The system of claim 11, wherein the criteria is based on at least one of:

a company standard, and agent preference, the selection from a computer of an agent of a first category of lead information and an attribute of the first category.

19. The system of claim 11, wherein the quality score additionally weighted based on at least one of:

a company standard, and agent preference, the selection from a computer of an agent of a first category of lead information and an attribute of the first category.

20. The system of claim 19, wherein additionally weighing is based on at least one of:

multiplying the quality score by a standard factor, adding a standard factor to the quality score, dividing the quality score by a standard factor, and subtracting a standard factor to the quality score.