Adaptive Support Guidance For Intent-Based Routing And Summaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional customer support systems often route customers to inappropriate representatives, leading to wasted time, frustration, and increased costs due to inefficient call handling and lack of access to past interaction data, resulting in repeated issues and poor customer satisfaction.
Innovation Solution
Implementing a system that determines customer intent through machine learning models, provides summaries of past interactions, and adapts recommended troubleshooting steps based on previous encounters, while using multi-modal sentiment analysis to enhance support agent guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional call routing systems are used, then customers can reach customer support, but they are routed to inappropriate representatives leading to wasted time and repeated issues
Solution Approach 1:
The system performs preliminary analysis of customer intent and past interactions before routing the call, using machine learning models to predict the most appropriate representative. This advance preparation ensures accurate routing from the first call, eliminating repeated transfers and reducing overall resolution time.
Solution Approach 2:
The system continuously learns from past call outcomes and customer interactions, using feedback loops to improve routing accuracy over time. By analyzing successful resolutions and customer satisfaction data, the system refines its routing decisions to prevent future mismatches between customers and representatives.
2Productivity
If customer support representatives lack access to past interaction data, then they can focus on current issues, but they repeat troubleshooting steps already tried
Solution Approach 1:
The system extracts relevant information from extensive past interaction records and presents only the most critical details to the support agent. By selectively extracting key troubleshooting steps already attempted and outcomes achieved, the system provides concise context without overwhelming the agent, maintaining efficiency while preventing repetition of failed approaches.
3Reliability
If detailed past interaction data is provided to support agents, then they can avoid repeating steps, but the information is not presented in a readily ingestible form
Solution Approach 1:
The system segments past interaction data into distinct, manageable components such as troubleshooting steps attempted, outcomes achieved, and relevant customer context. This segmentation allows agents to quickly scan and absorb critical information without being overwhelmed by raw data volume, improving both reliability and ease of information consumption.
4Adaptability or versatility
If customers select from limited dropdown options, then the system can categorize issues, but customers may be frustrated and press 0 without providing clear issue indication
Solution Approach 1:
The system introduces an intermediary natural language processing layer between the customer's free-text input and the issue categorization system. This mediator automatically interprets customer statements, extracts intent, and maps to appropriate categories without requiring customers to navigate limited dropdown menus, thereby maintaining categorization accuracy while dramatically improving ease of operation.
Data Source
AI summary
A method can include receiving, from a customer, a request for a support encounter. A method can include determining a customer intent indicative of an issue experienced by the customer. A method can include routing the customer to a support agent based on the intent. A method can include generating, using a first large language model, summaries of previous support encounters. A method can include providing the summaries to the support agent. A method can include determining, based on the intent and/or the summaries, one or more support documents related to the issue. A method can include generating, using a second large language model, a summary of each of the support documents, wherein an input to the second large language model comprises a support document. A method can include providing the generated summaries of the support documents to the support agent for use in mitigating the issue.