System and method for prioritizing agent intervention into automated customer engagements

• US 9,621,728 B2
• Filed: 10/11/2013
• Issued: 04/11/2017
• Est. Priority Date: 10/11/2013
• Status: Active Grant

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Denby, Paul, McCormack, Tony, O'Connor, Neil, D'Arcy, Paul
• Primary Examiner(s): SHAH, ANTIM G
• Application Number:
  US14/051,932
• Publication Number:
  US 20150103995A1
• Time in Patent Office: 1,278 Days
• Field of Search: 37926502, 37926505, 37926509, 37926512, 37926513
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements
• CPC Class Codes: H04M 3/5191 interacting with the Internet

Abstract

A system for prioritizing intervention of live agents into automated customer engagements in a communication system is disclosed. The system includes an intervention prioritization module configured to identify a live agent to intervene into an automated customer engagement in a communication system based on a confidence factor corresponding to the automated customer engagement and one or more live agent attributes corresponding to the live agent. The system further includes a live agent conference module configured to cause the identified live agent to intervene into the automated customer engagement.

Claims

1. A system for prioritizing intervention of live agent communication devices into automated customer engagements in a communication system comprising a contact center, the system comprising:

a first server including a processor coupled to memory, wherein the processor executes program code stored in the memory to;

dynamically adjust a confidence factor threshold associated with an automated customer engagement, based on an availability of live agents in a pool of live agents;

identify an available live agent from the pool of live agents to intervene into the automated customer engagement based on a confidence factor corresponding to the automated customer engagement, and the confidence factor threshold; and

selectively implement a first configuration where a live agent communication device associated with the identified live agent intervenes with the automated customer engagement, wherein the automated customer engagement provides responses that are at least one of directed by, and assisted by, the identified live agent operating the associated live agent communication device, such that a customer associated with the customer engagement is unaware of the live agent intervention; and

a second server that executes program code stored in memory to;

receive the automated customer engagement; and

generate the confidence factor threshold in response to the received automatic customer engagement;

wherein the first server dynamically adjusts the confidence factor threshold by updating the confidence factor threshold generated by the second server.

2. The system of claim 1, wherein the automated customer engagement is an automated chat session.

3. The system of claim 1, wherein the automated customer engagement is a select one automated customer engagement of a plurality of automated customer engagements, wherein:

the processor executes further program code stored in memory to;

identify the select one automated customer engagement based on a plurality of confidence factors related to each of the plurality of automated customer engagements and one or more strategy rules.

4. The system of claim 1, wherein:

the processor executes further program code stored in memory to;

identify the live agent based on engagement metadata related to each of the plurality of automated customer engagements.

5. The system of claim 1, wherein the engagement metadata includes at least one of a domain of the engagement or a customer metric.

6. The system of claim 1, wherein:

the processor executes further program code stored in memory to;

identify the available live agent from the pool of live agents to intervene into the automated customer engagement based on at least one live agent attribute; and

the at least one live agent attribute includes at least one of a domain of the live agent or a proficiency of the live agent.

7. The system of claim 1, wherein:

the processor executes further program code stored in memory to;

identify the available live agent from the pool of live agents to intervene into the automated customer engagement based on at least one strategy rule; and

the at least one strategy rule includes at least one of a desired utilization of live agents or an optimal confidence factor threshold.

8. The system of claim 1, further comprising:

an automated engagement system that provides the automated customer engagement by engaging a customer communication device of the customer in an automated chat session; and

a natural language processor that receives textual queries from the customer communication device and provides textual responses to the customer communication device;

wherein;

the processor executes further program code stored in memory to;

determine the confidence factor based on the textual queries received by the natural language processor and the textual responses provided by the natural language processor.

9. The system of claim 8, wherein:

the processor executes further program code stored in memory to;

implement a second configuration where the identified live agent communication device intervenes by passing a communication between a customer communication device operated by the customer and the live agent communication device of the identified live agent such that the customer is aware of the identified live agent.

10. A computer-implemented method for prioritizing intervention of live agents into automated customer engagements in a communication system comprising a contact center, the method comprising:

adjusting dynamically, by a processor on a first server, a confidence factor threshold associated with an automated customer engagement, based on an availability of live agents in a pool of live agents;

receiving by a second server, the automated customer engagement;

generating by the second server, the confidence factor threshold in response to the received automatic customer engagement, wherein the first server dynamically adjusts the confidence factor threshold by updating the confidence factor threshold generated by the second server;

identifying, by the processor, an available live agent from the pool of live agents to intervene into the automated customer engagement based on a confidence factor corresponding to the automated customer engagement, and the confidence factor threshold; and

executing by the processor, a first configuration where a live agent communication device associated with the identified live agent intervenes with the automated customer engagement, wherein the automated customer engagement provides responses that are at least one of directed by, and assisted by, the identified live agent operating the associated live agent communication device, such that a customer associated with the customer engagement is unaware of the live agent intervention.

11. The computer-implemented method of claim 10, further comprising implementing the automated customer engagement as an automated chat session.

12. The computer-implemented method of claim 10, wherein the automated customer engagement is a select one automated customer engagement of a plurality of automated customer engagements, the method further comprising identifying the select one automated customer engagement based on a plurality of confidence factors related to each of the plurality of automated customer engagements and at least one strategy rule.

13. The computer-implemented method of claim 10, further comprising identifying the live agent based on engagement metadata related to each of a plurality of automated customer engagements.

14. The computer-implemented method of claim 10, further comprising defining engagement metadata to include at least one of a domain of the engagement or a customer metric.

15. The computer-implemented method of claim 10, further comprising defining live agent attributes to include at least one of a domain of the live agent or a proficiency of the live agent.

16. The computer-implemented method of claim 10, further comprising defining the a strategy rule to include at least one of a desired utilization of live agents or an optimal confidence factor threshold.

17. The computer-implemented method of claim 10, further comprising:

automatically engaging a customer in the automatic customer engagement by engaging a communication device of the customer in an automated chat session;

receiving a textual query from the customer communication device; and

generating a textual response that is communicated to the customer communication device.

18. The computer-implemented method of claim 17, wherein causing the identified live agent to intervene into the automated customer engagement comprises implementing a second configuration where the identified live agent intervenes by passing a communication between the customer communication device of the customer and the live agent communication device of the identified live agent such that the customer is aware of the identified live agent.

19. The computer-implemented method of claim 17, further comprising determining a confidence factor based on the textual query and the textual response.

20. Computer readable hardware storing computer readable instructions that when executed by a processor perform:

adjusting dynamically, by the processor on a first server, a confidence factor threshold associated with an automated customer engagement, based on an availability of live agents in a pool of live agents;

receiving by a second server, the automated customer engagement;

generating by the second server, the confidence factor threshold in response to the received automatic customer engagement, wherein the first server dynamically adjusts the confidence factor threshold by updating the confidence factor threshold generated by the second server;

identifying an available live agent from the pool of live agents to intervene into an automated customer engagement based on a confidence factor corresponding to the automated customer engagement, and the confidence factor threshold; and

implementing, selectively, a first configuration where a live agent communication device associated with the identified live agent intervenes with the automated customer engagement, wherein the automated customer engagement provides responses that are at least one of directed by, and assisted by, the identified live agent operating the associated live agent communication device, such that a customer associated with the customer engagement is unaware of the live agent intervention.