Systems and methods for identifying potentially inaccurate data based on patterns in previous submissions of data

• US 9,659,277 B2
• Filed: 05/27/2011
• Issued: 05/23/2017
• Est. Priority Date: 05/27/2011
• Status: Active Grant

• Current Assignee: Hartford Fire Insurance Company (The Hartford Financial Services Group, Inc.)
• Original Assignee: Hartford Fire Insurance Company (The Hartford Financial Services Group, Inc.)
• Inventors: Friedman, Kurt L.
• Primary Examiner(s): AMELUNXEN, BARBARA J
• Application Number:
  US13/117,409
• Publication Number:
  US 20120303389A1
• Time in Patent Office: 2,188 Days
• Field of Search: 705 1, 705 2, 705 4, 705 9, 705 10, 705 1425, 705 37, 705 35, 705 38, 707999104, 707999107, 1 1, 345419
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

According to some embodiments, a system may receive, from a remote insurance submitter, insurance data submitted via an electronic quoting and submission system in connection with behavioral data associated with the submitter. A processor may then automatically analyze the submitted insurance data based at least in part on information stored in a historic underwriting analytics database and the behavioral data associated with the submitter. Potentially inaccurate insurance data submitted by the insurance submitter may then be identified based on the analysis. According to some embodiments, a plurality of potentially inaccurate insurance data values may be used to flag the submitted insurance data, the insurance submitter, one or more input portions of the electronic quoting and submission system, and/or an insurance class of business.

Claims

1. A data analytics system, comprising:

an electronic quote submission system configured to receive, via a communications network, a plurality of policy application data submissions for potential customers from remote agent devices, each of the policy application data submissions including (i) revisions data indicative of one or more revisions of submitted policy application data made by a corresponding remote agent during entry of data for each of the policy application data submissions, (ii) backdating data indicative of requests for backdated coverage;

(iii) cancellation data indicative of policy cancellations; and

(iv) classification data indicative of classification changes; and

a policy analytics platform coupled to the electronic quote submission system and configured to;

store, in a data storage device, the plurality of policy application data submissions in a database of historical underwriting data;

continuously analyze the historical underwriting data to identify trends in the database of historical underwriting data;

receive, from the electronic quote submission system, a policy application data submission for a potential customer from a remote agent device associated with an agent;

receive, from the data storage device, agent behavioral data corresponding to the remote agent in connection with other policy applications submitted by the remote agent through the electronic quote submission system, wherein the agent behavioral data includes the revisions data, the backdating data, the cancellation data, and the classification data;

automatically analyze the submitted policy application data based at least in part on the agent behavioral data corresponding to the remote agent and the identified trends in the historical underwriting data;

identify, based on the analysis, potentially inaccurate policy application data submitted by the remote agent in the submitted policy application data;

responsive to identification of the potentially inaccurate policy application data submitted by the remote agent, generate a flag corresponding to at least one of the input portion of the electronic quote submission system, and a class of business;

responsive to identification of a plurality of potentially inaccurate policy application data submitted by the remote agent, generate one or more flags corresponding to the remote agent based on application of;

(i) a revisions rule threshold, (ii) a cancellations rule threshold, (iii) a backdating rule threshold, and (iv) a classification change rule threshold, to the plurality of potentially inaccurate policy application data; and

render for display on an administrator response device a real-time graphical display which identifies the one or more flags generated in relation to potentially inaccurate policy application data and which displays a recommendation for addressing each of the one or more flags, thereby permitting review of the potentially inaccurate policy data before generation of an underwriting determination.

2. The system of claim 1, wherein the electronic quote submission system is configured to receive one of:

(i) a policy application being submitted for a quote or approval, (ii) a previously issued policy, and (iii) a plurality of previously issued policies.

3. The system of claim 1, wherein the electronic quote submission system is configured to receive submitted policy application data for at least one of:

(i) a commercial small business owner'"'"'s policy, and (ii) a workers compensation policy.

4. The system of claim 1, wherein the electronic quote submission system is adapted to collect behavioral data associated with the remote agent.

5. The system of claim 1, further comprising:

a response device adapted to receive a signal associated with the potentially inaccurate data and to maintain a queue of actions to be taken in connection with the potentially inaccurate data.

6. The system of claim 1, wherein the memory storing program instructions executable by the computer processor to receive agent behavioral data further comprises the memory storing program instructions executable by the computer processor to receive one or both of (1) data relating to a set of answers submitted by the remote agent over a period of time in connection with the other submitted policy applications, and (2) data relating to endorsements requested by the remote agent after one or more of the other submitted policy applications has been submitted or issued.

7. The system of claim 1, wherein the policy analytics platform is further configured to generate modeling information indicative of rule weighting and whether to employ certain rules in generation of one or more flags based at least in part on steadiness of rate of flagging.

8. A computer-implemented method, comprising:

receiving, by an electronic quote submission system via a communications network, a plurality of policy application data submissions for potential customers from remote agent devices, each of the policy application data submissions including (i) revisions data indicative of one or more revisions of submitted policy application data made by a corresponding remote agent during entry of data for each of the policy application data submissions, (ii) backdating data indicative of requests for backdated coverage;

(iii) cancellation data indicative of policy cancellations; and

(iv) classification data indicative of classification changes; and

storing, by a policy analytics platform coupled to the electronic quote submission system, in a data storage device, the plurality of policy application submissions in a database of historical underwriting data;

continuously analyzing, by the policy analytics platform, the historical underwriting data to identify trends in the database of historical underwriting data;

receiving, by the policy analytics platform from the electronic quote submission system, a policy application data submission for a potential customer from a remote agent device associated with an agent;

receiving, by the policy analytics platform from the data storage device, agent behavioral data corresponding to the remote agent in connection with other policy applications submitted by the remote agent through the electronic quote submission system, wherein the agent behavioral data includes the revisions data, the backdating data, the cancellation data, and the classification data;

automatically analyzing, by the policy analytics platform, the submitted policy application data based at least in part on the agent behavioral data corresponding to the remote agent and the identified trends the historical underwriting data;

identifying, by the policy analytics platform based on the analysis, potentially inaccurate policy application data submitted by the remote agent in the submitted policy application data;

responsive to identification of the potentially inaccurate policy application data submitted by the remote agent, generating, by the policy analytics platform, a flag corresponding to at least one of the input portion of the electronic quote submission system, and a class of business;

responsive to identification of a plurality of potentially inaccurate policy application data submitted by the remote agent, generating, by the policy analytics platform, one or more flags corresponding to the remote agent based on application of;

(i) a revisions rule threshold, (ii) a cancellations rule threshold, (iii) a backdating rule threshold, and (iv) a classification change rule threshold, to the plurality of potentially inaccurate policy application data; and

rendering, by the policy analytics platform for display on an administrator response device, a real-time graphical display which identifies the one or more flags generated in relation to potentially inaccurate policy application data and which displays a recommendation for addressing each of the one or more flags, thereby permitting review of the potentially inaccurate policy data before generation of an underwriting determination.

9. The method of claim 8, wherein receiving the submitted policy application data comprises receiving the submitted policy application data from a party entering information on behalf of the agent.

10. The method of claim 8, wherein receiving the submitted policy application data comprises receiving the submitted policy application data for a renewal of a previously issued policy.

11. The method of claim 8, further comprising storing in the historic underwriting analytics database information comprising at least one of:

(i) agent large loss information, (ii) an audit team input, and (iii) a field agency assessment.

12. The method of claim 8, wherein generating the flag further comprises generating the flag based on a rule threshold for at least one of:

(i) hard stops, (ii) exposures, (iii) pricing, and (iv) large loss rates.

13. The method of claim 8, further comprising transmitting agent recommendation in connection with the flag, the recommendation being at least one of:

(i) no action being required for an agent, (ii) an agent audit recommendation, (iii) a monitor status for an agent, (iv) an increase in an agent quality assurance process, (v) a verbal or written agent warning, (vi) an agent training plan, (vii) an agent action plan, and (viii) a financial penalty to an agent.

14. The method of claim 8, wherein the electronic quote submission system includes a plurality of input portions and further comprising flagging, based upon the data about a plurality of potentially inaccurate data values, at least one of the input portions.

15. The method of claim 14, wherein the input portion comprises a Standard Industrial Classification (“

SIC”

) code question or selection.

16. The method of claim 8, further comprising flagging, responsive to identification of a plurality of potentially inaccurate data values, a class of business.

17. The method of claim 8, wherein receiving the submitted policy application data comprises receiving submitted policy application data comprising at least one of:

(i) an analysis in process, (ii) a monitor status, (iii) an open status, (iv) a delivered recommendation, and (v) an in progress recommendation.

18. The method of claim 8, wherein receiving the submitted policy application data comprises receiving the submitted policy application data for at least one of:

(i) a commercial small business owner'"'"'s policy, and (ii) a workers compensation policy.

19. The method of claim 8, further comprising storing, in the historic underwriting analytics database, at least one of:

(i) leading indicators of misinformation, (ii) agency profile data, (iii) declination information, (iv) endorsement activity, (v) exposure discrepancy, (vi) hard stop overrides, (vii) large loss and unacceptable risk values, (x) risk alerts, and (xi) third-party data.

20. The method of claim 8, wherein continuously analyzing the historical underwriting data to identify trends in the database of historical underwriting data comprises periodically performing an analysis on the information stored in the historic underwriting analytics database to determine at least one of:

(i) a trigger, (ii) an unusual distribution or frequency of negative attributes, (iii) a tolerance, and (iv) a threshold.

21. The method of claim 20, wherein the periodic analysis comprises at least one of:

(i) scoring data, (ii) data mining, (iii) outlier identification, (iv) weighting factor generation, and (v) a recent activity analysis.

22. A non-transitory computer-readable medium storing computer executable instructions, which, when executed by a computer processor, cause the computer processor to:

receive a plurality of policy application data submissions for potential customers from remote agent devices, each of the policy application data submissions including (i) revisions data indicative of one or more revisions of submitted policy application data made by a corresponding remote agent during entry of data for each of the policy application data submissions, (ii) backdating data indicative of requests for backdated coverage;

(iii) cancellation data indicative of policy cancellations; and

(iv) classification data indicative of classification changes; and

store the plurality of policy application submissions in a database of historical underwriting data;

continuously analyze the historical underwriting data to identify trends in the database of historical underwriting data;

receive, from a remote agent, a policy application data submission for a potential customer via an electronic quoting and submission system,

receive, from a historic underwriting analytics database, agent behavioral data corresponding to the remote agent in connection with other policy applications submitted by the remote agent,

wherein the agent behavioral data includes the revisions data, the backdating data, the cancellation data, and the classification data;

automatically analyze the submitted data based at least in part on the insurance agent behavioral data corresponding to the remote agent and the identified trends in the historical underwriting data;

identify, based on the analysis, potentially inaccurate data submitted by the remote agent in the submitted data;

responsive to identification of the potentially inaccurate insurance policy application data submitted by the remote agent, generate a flag corresponding to at least one of the input portion of the electronic quoting and submission system, and a class of business;

responsive to identification of a plurality of potentially inaccurate policy application data submitted by the remote agent, generate one or more flags corresponding to the remote agent based on application of;

(i) a revisions rule threshold, (ii) a cancellations rule threshold, (iii) a backdating rule threshold, and (iv) a classification change rule threshold, to the plurality of potentially inaccurate policy application data; and

render for display on an administrator response device a real-time graphical display which identifies the one or more flags generated in relation to potentially inaccurate policy application data and which displays a recommendation for addressing each of the one or more flags, thereby permitting review of the potentially inaccurate policy data before generation of an underwriting determination.

23. The medium of claim 22, wherein the instructions which cause the computer processor to receive submitted policy application data comprises instructions which cause the computer processor to receive submitted policy application data for a renewal of a previously issued policy.

24. The medium of claim 23, wherein the instructions which cause the computer processor to receive submitted policy application data comprises instructions which cause the computer processor to receive submitted policy application data for at least one of:

(i) a commercial small business owner'"'"'s policy, and (ii) a workers compensation policy.