ADVANCED COACHING AND ARTIFICIAL INTELLIGENCE REFINEMENT
Systems and methods are provided to assess a communication conducted over a network. The systems monitor the communication and submit a survey to the customer and agent. The agent may be a human agent or an artificially intelligent (AI) agent. The survey is selected or generated, such as in response to communication content, and presented to the customer and agent. If the scoring is significantly different, an automated agent initiates a remediation action, such as to coach a human agent or to provide feedback or a corrective prompt to an AI agent.
1 . A method, comprising:
monitoring, by a microprocessor of a communication monitoring system, a communication, via a network, between a customer device utilized by a customer and an agent device utilized by an agent;
presenting, by the microprocessor through a graphical user interface rendered on the customer device, a customer survey comprising user interface elements configured to solicit feedback on the communication;
presenting, by the microprocessor through a graphical user interface rendered on the agent device, an agent survey comprising user interface elements configured to solicit feedback on the communication;
generating, by the microprocessor, a customer survey score from responses from the customer device by converting the responses into a numerical score and storing the numerical score in a database associated with the communication;
generating, by the microprocessor, an agent survey score from responses from the agent device by converting the responses into a numerical score and storing the numerical score in the database associated with the communication;
computing, by the microprocessor, a difference value between the agent survey score and the customer survey score; and
responsive to determining that the difference value exceeds a programmable threshold, the microprocessor automatically initiating a remediation action comprising triggering a notification message to a supervisory system.
2 . The method of claim 1 , further comprising:
generating, by the microprocessor, a training set comprising at least one of the communication, customer survey, agent survey, the customer survey score, the agent survey score, and the difference; and
wherein the agent comprises an artificial intelligence (AI) agent;
wherein the agent device comprises a neural network executing the AI agent; and
wherein the remediation action comprises providing the training set to the neural network.
3 . The method of claim 1 , further comprising:
generating, by the microprocessor, a prompt comprising at least one of the communication, the customer survey, the agent survey, the customer survey score, the agent survey score, and the difference; and
wherein the agent comprises an artificial intelligence (AI) agent;
wherein the agent device comprises a machine learning algorithm executing the AI agent; and
wherein the remediation action comprises providing the prompt to the machine learning algorithm.
4 . The method of claim 1 , wherein:
the agent comprises a human agent; and
the supervisory system automatically connects to the communication and either takes over the communication or provides assistance to the human agent.
5 . The method of claim 1 , wherein at least one of the customer survey and the agent survey are conducted while the communication is ongoing.
6 . The method of claim 1 , wherein the customer survey is a portion of a master customer survey for a plurality of communications comprising the agent, and wherein at least one other portion of the master customer survey is presented to at least one other customer during a prior communication.
7 . The method of claim 6 , wherein the customer survey comprises a random portion of the master customer survey.
8 . A method, comprising:
monitoring, by a first artificial intelligence (AI) agent, a communication, via a network, between a customer device utilized by a customer and a second artificial intelligence (AI) agent;
presenting through a graphical user interface rendered on the customer device a customer survey comprising user interface elements configured to solicit feedback on the communication;
causing the second AI agent to respond to an agent survey to provide feedback on the communication;
generating a customer survey score from responses from the customer device by converting the responses into a numerical score and storing the numerical score in a database associated with the communication;
generating an agent survey score from responses from the agent device by converting the responses into a numerical score and storing the numerical score in the database associated with the communication;
computing a difference value between the agent survey score and the customer survey score; and
responsive to determining that the difference value exceeds a programmable threshold, automatically initiating a remediation action comprising triggering a notification message to a supervisory system.
9 . The method of claim 8 , further comprising:
generating a training set comprising at least one of the communication, the customer survey, the agent survey, the customer survey score, the agent survey score, and the difference; and
wherein the second AI agent comprises a neural network executing the second AI agent; and
wherein the remediation action comprises providing the training set to the neural network.
10 . The method of claim 8 , further comprising:
generating a prompt comprising at least one of the communication, the customer survey, the agent survey, the customer survey score, the agent survey score, and the difference; and
wherein the agent device comprises a machine learning algorithm executing the second AI agent; and
wherein the remediation action comprises providing the prompt to the machine learning algorithm.
11 . The method of claim 8 , wherein the second AI agent comprises a neural network trained with a dataset comprising a plurality of past communications.
12 . The method of claim 8 , wherein the first AI agent comprises a neural network trained with a dataset comprising a plurality of past communications, corresponding past customer surveys, and a set of communication features.
13 . The method of claim 12 , wherein the communication features comprise at least one of pace of speech, volume of speech, intonation, accent, word choice, and pauses that are absent speech.
14 . A system, comprising:
at least one microprocessor coupled with a computer memory comprising computer readable instructions that, when read by the at least one microprocessor, cause the at least one microprocessor to:
monitor a communication, via a network, between a customer device utilized by a customer and an agent device utilized by an agent;
present through a graphical user interface rendered on the customer device a customer survey comprising user interface elements configured to solicit feedback on the communication;
present through a graphical user interface rendered on the agent device an agent survey comprising user interface elements configured to solicit feedback on the communication;
generate a customer survey score from responses from the customer device by converting the responses into a numerical score and storing the numerical score in a database associated with the communication;
generate an agent survey score from responses from the agent device by converting the responses into a numerical score and storing the numerical score in the database associated with the communication;
compute a difference value between the agent survey score and the customer survey score; and
responsive to determining that the difference value exceeds a programmable threshold, automatically initiate a remediation action comprising triggering a notification message to a supervisory system.
15 . The system of claim 14 , further comprising instructions to cause the at least one microprocessor to:
generate a training set comprising at least one of the communication, the customer survey, the agent survey, the customer survey score, the agent survey score, and the difference; and
execute the agent comprising an artificial intelligence (AI) agent;
wherein the agent device comprises a neural network executing the AI agent; and
wherein the remediation action comprises providing the training set to the neural network.
16 . The system of claim 14 , further comprising instructions to cause the at least one microprocessor to:
generate a prompt comprising at least one of the communication, the customer survey, the agent survey, the customer survey score, the agent survey score, and the difference; and
execute the agent comprising an artificial intelligence (AI) agent; and
wherein the agent device comprises a machine learning algorithm executing the AI agent; and
wherein the remediation action comprises providing the prompt to the machine learning algorithm.
17 . The system of claim 14 , wherein:
the agent comprises a human agent; and
the supervisory system automatically connects to the communication and either takes over the communication or provides assistance to the human agent.
18 . The system of claim 14 , wherein at least one of the customer survey and the agent survey are conducted while the communication is ongoing.
19 . The system of claim 14 , wherein the customer survey is a portion of a master customer survey for a plurality of communications comprising the agent, and wherein at least one other portion of the master customer survey is presented to at least one other customer during a prior communication.
20 . The system of claim 19 , wherein the customer survey comprises a random portion of the master customer survey.