Live Agent Recommendation via Human Expertise Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current live agent recommendation systems in contact centers face challenges such as limited working hours, capacity, and knowledge levels, leading to inconsistent and inadequate assistance, especially when customers interact with different agents for the same issues, and existing AI solutions struggle with complex queries, resulting in poor user satisfaction and resource inefficiency.
Innovation Solution
A method and system utilizing deep neural networks to construct a human expertise matrix based on average net promoter scores (NPS) and category assessments, predicting the best-suited live agent for user interactions, ensuring consistent and skilled assistance by selecting agents with high overall NPS and relevant skills.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional communications mechanisms (live agents, IVR units, email) are used to handle user communications, then basic customer service functions can be provided, but the system suffers from limited working hours, limited capacity, limited knowledge levels, and inconsistent service quality
Solution Approach 1:
The patent segments the service system into multiple specialized agents, each with specific expertise in different domains (e.g., technical support, billing, sales). This segmentation allows the system to handle diverse customer inquiries with specialized knowledge while maintaining consistent service quality through structured agent assignment based on customer needs and agent capabilities.
Solution Approach 2:
The patent creates a universal agent management system that can handle multiple types of communications (voice calls, emails, chat messages) through a single integrated platform. The system universally applies agent expertise matching, availability management, and skill assessment across all communication channels, eliminating the limitations of traditional single-function systems.
2Productivity
If more live agents are hired to increase capacity and availability, then service coverage improves, but operational costs and management complexity increase
Solution Approach 1:
The patent implements continuous feedback mechanisms where customer interactions, agent performance metrics, and skill assessments are systematically collected and analyzed. This feedback loop enables dynamic agent assignment, performance optimization, and skill development planning, allowing the system to efficiently manage agent capacity without proportional increases in management complexity.
Solution Approach 2:
The patent transforms agent management from a static administrative task to a dynamic optimization problem by continuously adjusting parameters such as agent availability, skill levels, and assignment priorities based on real-time system state and performance data, thereby increasing productivity without linearly increasing management overhead.
3Productivity
If AI solutions are deployed to handle complex queries, then automation efficiency improves, but the system struggles with nuanced customer issues resulting in poor user satisfaction
Solution Approach 1:
The patent introduces human agents as intermediaries between AI automation and complex customer queries. The system uses AI to handle routine tasks and preliminary analysis, then seamlessly transfers complex or nuanced queries to appropriately matched human agents who provide the empathy and judgment needed for difficult situations, combining automation efficiency with reliable resolution accuracy.
Solution Approach 2:
The patent applies preliminary AI processing to customer queries before human agent involvement, performing initial classification, information gathering, and solution preparation. This preliminary action filters out routine queries that can be fully automated while pre-processing complex queries to reduce the burden on human agents and improve overall resolution accuracy.
Data Source
AI summary
A computer-implemented method is presented for selecting a preferred live agent from a plurality of live agents. The method includes constructing, via the processor, a human expertise matrix pertaining to each of the plurality of live agents by determining an average net promoter score (NPS) for each of the plurality of live agents for each category of a plurality of categories, and in response to a voice call by a user, determining, via the processor, a predicted human expertise on average by collectively assessing the human expertise matrix, a predicted NPS derived from a first deep neural network, and a predicted category derived from a second deep neural network. The method further includes, based on the predicted human expertise on average determined, triggering communication via the live agent communication network between the user and the preferred live agent to initiate a conversation between the user and the preferred live agent.