Intent-Based Skill Classification for Contact Center Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face inefficiencies due to inaccurate and incomplete data in agent skill level determination, leading to improper routing of interactions and inefficient resource utilization, as current methods rely on manual audits and assumptions about tool-based skills correlating to actual knowledge-based skills.
Innovation Solution
A method that uses content analysis to determine the intent of customer interactions and dynamically updates agent skill levels based on successful resolutions, allowing for accurate and granular skill administration without extensive administrative involvement, by assigning point values to intents and updating skills accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual audits and assumptions about tool-based skills are used to determine agent skill levels, then administrative involvement is required, but the determination becomes inaccurate and incomplete
Solution Approach 1:
The system enables self-service by automatically determining agent skill levels through content analysis of interactions. The content analyzer independently evaluates interaction content and updates skill levels without requiring manual administrative audits, eliminating the trade-off between accuracy and administrative complexity
Solution Approach 2:
The patent replaces the mechanical manual audit process with an automated content analysis system using natural language processing. This substitution transforms the skill determination from a manual administrative task to an automated computational process, improving both accuracy and eliminating administrative burden
2Reliability
If manual skill audits are performed periodically, then administrative resources are consumed, but skill level data becomes outdated between audits
Solution Approach 1:
The system implements continuous skill level determination by analyzing interaction content in real-time as interactions occur. Rather than periodic audits, the content analyzer continuously updates agent skill levels based on each new interaction, ensuring data remains current and eliminating time delays between updates
Solution Approach 2:
The system establishes a feedback loop where each interaction analyzed provides immediate feedback to update agent skill levels. The content analyzer continuously processes interaction content and feeds results back to update the skill level database, creating a dynamic, always-current skill representation
3Measurement precision
If skills are highly granularized to accurately identify agent expertise, then routing precision improves, but the number of skills to maintain increases significantly
Solution Approach 1:
The system applies dynamics by making the skill taxonomy flexible and adaptive rather than static. New skills are automatically created and added to the taxonomy as new interaction topics emerge, allowing the system to maintain high granularity and precision without requiring manual planning for all possible skill categories
Solution Approach 2:
The content analyzer segments interaction content into specific intent topics, automatically identifying and categorizing nuanced skill requirements. This segmentation approach enables highly granular skill identification by breaking down interactions into specific topic areas, maintaining precision without manual overhead
4Productivity
If best-effort skill determination is used for routing decisions, then system complexity is reduced, but interaction routing efficiency decreases
Solution Approach 1:
The patent replaces best-effort heuristic routing with automated content analysis using natural language processing. This substitution introduces computational complexity but eliminates the inefficiency of best-effort approaches, achieving both high productivity through accurate routing and managed complexity through automation
Data Source
AI summary
The present invention is directed toward a method and system for determining the skill levels of an agent in a contact center. The invention creates a relationship between contact intent and agent skill to help determine agent skills based on successfully completed contacts.