Contact Center Agent Assistance System Using Sentiment Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face challenges in optimizing efficiency and quality of call handling, particularly when silver-rated agents handle calls, leading to potential customer dissatisfaction and increased operational stress due to the lack of real-time assistance and training for new agents.
Innovation Solution
An agent assisting system that monitors call parameters and user sentiments, determines scripts to improve call handling, and automatically provides instructions to agents, enabling real-time assistance and training without the need for continuous supervisor intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If silver-rated agents handle calls to increase throughput, then productivity improves, but call handling quality deteriorates
Solution Approach 1:
The system enables self-service by providing automated real-time monitoring and assistance to agents. The monitoring module automatically detects call parameters and sentiment changes, while the script publishing module delivers relevant scripts without requiring supervisor intervention, allowing agents to independently maintain quality while handling increased call volumes
Solution Approach 2:
The system implements continuous feedback loops by monitoring call parameters and user sentiments in real-time, analyzing whether current scripts are effective, and automatically publishing updated scripts when improvements are detected. This closed-loop feedback mechanism ensures quality maintenance through data-driven adjustments
2Manufacturing precision
If supervisors provide continuous assistance to agents, then call handling quality improves, but device complexity and operational stress increase
Solution Approach 1:
The system replaces the mechanical system of continuous human supervisor intervention with an automated electronic monitoring and assistance system. The monitoring module, sentiment detector, and script publishing module work together to automate quality assurance, eliminating the need for constant supervisor involvement while maintaining or improving call handling quality
Solution Approach 2:
The system introduces an intermediary automated assistance layer between supervisors and agents. Rather than direct supervisor-agent interaction, the system acts as an intermediary that monitors calls, analyzes sentiment, and automatically publishes appropriate scripts, reducing the complexity of human coordination while maintaining quality standards
3Manufacturing precision
If more training is provided to new agents, then call handling quality improves, but loss of time increases
Solution Approach 1:
The system applies preliminary action by pre-processing and analyzing call data to identify patterns and optimal scripts before agents need them. The monitoring module continuously analyzes calls in advance, and the script publishing module makes scripts available proactively, allowing agents to access relevant information exactly when needed rather than through extended training periods
Solution Approach 2:
New agents benefit from self-service learning where the system automatically monitors their calls, analyzes performance, and publishes personalized script recommendations in real-time. This on-the-job adaptive training eliminates the need for lengthy pre-training programs while maintaining quality standards
Data Source
AI summary
An agent assisting system for providing assistance to agents of contact center during calls is provided. The agent assisting system includes a monitoring module configured to monitor parameters of a call coming to the contact center. The agent assisting system further includes a sentiment detector module configured to detect sentiments of a user during the call. The agent assisting system further includes a pattern matching module configured to determine a script for an agent handling the call, based on the parameters of the call and the sentiments of the user during the call, wherein the script providing instructions to the agent to improve handling of the call. The agent assisting system further includes a script publishing module configured to publish the script to the agent handling the call.