🗺️Trusst Based Routing (AI powered Decision Engine)
This page describes TrusstGPT's AI driven decision and routing engine - Trusst Based Routing (TBR).
Overview
TrusstGPT's TBR feature is designed to enhance the efficiency and effectiveness of customer-agent interactions within contact centers by intelligently routing customer engagements to the most suitable agent based on historical performance and call context.
The Trusst Based Routing algorithm leverages outputs from Language Learning Models (LLM) to assess and categorize customer calls in real-time. By analyzing elements such as call content, customer intent, and emotional tone, the system determines the call type. Post-call, the agent’s performance is evaluated and assigned a Trusst Score, reflecting their effectiveness in handling specific types of calls. For future interactions, this score helps predict and select the best available agent for a particular customer query, optimizing outcomes for both the customer and the organization.
Agent Performance Evaluation (Trusst Score):
Input: Post-call data including call duration, resolution status, customer satisfaction scores, and LLM-analyzed call details.
Functionality: Each agent’s handling of calls is scored based on predefined criteria tailored to different types of queries. Factors include resolution efficiency, customer feedback, and suitability to call type.
Output: A comprehensive Trusst Score that updates the agent’s profile with new performance metrics after each call.
Intelligent Routing Engine:
Input: Incoming call data, available agents’ current Trusst Scores, and their historical performance metrics.
Functionality: Utilizes a decision-making algorithm to assess and match incoming calls with the optimal agent. The matching is based on factors like predicted call complexity, agent expertise in a particular domain, and past performance with similar calls.
Output: Decision on which agent should receive the call, aimed at improving overall customer satisfaction and efficiency.
Feedback Loop for Continuous Improvement:
Mechanism: Regular audits and recalibrations of scoring criteria and algorithm parameters based on ongoing performance data and evolving business needs.
Goal: Ensure the routing algorithm remains accurate and effective in matching customer calls with the best-suited agents.
Additional Considerations:
Privacy and Compliance: All features must comply with relevant privacy laws and regulations, ensuring that customer data is handled securely and ethically.
Scalability: The system should be designed to scale effortlessly with increasing call volumes and expanding agent pools without degradation in performance.
Integration: Seamless integration with existing customer relationship management (CRM) systems and databases to utilize comprehensive data insights.
Implementation Roadmap:
Prototype Development: Initial development focusing on core functionalities of call analysis and basic routing.
Pilot Testing: Limited deployment within select teams to gather data and refine the algorithm.
Full-scale Rollout: Gradual deployment across all teams following thorough testing and optimization.
Ongoing Optimization and Support: Regular updates and maintenance to enhance capabilities and address emerging needs.
By focusing on strategic agent-customer matching, the Trusst-Based Routing feature aims to transform customer interactions into more personalized and efficient experiences, ultimately fostering greater customer loyalty and operational efficiency.
Last updated