Algorithmic Knowledge Gap Detection for Contact Center Coaching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Supervisors in contact centers face challenges in manually identifying knowledge gaps and relevant coaching content for agents, leading to inefficiencies and potential biases, which affect the quality and effectiveness of coaching sessions.
Innovation Solution
An algorithm analyzes agent interactions to identify performance gaps and knowledge gaps, using statistical methods to determine significant differences, and then provides relevant knowledge base articles to supervisors or agents based on mutual information scores, automating the coaching content selection process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If supervisors manually analyze agent interactions to identify knowledge gaps and coaching content, then coaching quality can be customized and tailored, but it consumes a large amount of time and is prone to human error and bias
Solution Approach 1:
The system enables self-service by automatically analyzing agent interactions and generating coaching content without requiring supervisor intervention. The algorithm independently processes interaction data, identifies knowledge gaps, and retrieves relevant coaching materials, allowing the system to serve itself rather than relying on manual supervisor analysis.
Solution Approach 2:
The patent replaces the mechanical manual analysis process with an automated algorithmic system. The algorithm uses computational methods to analyze interaction data, calculate knowledge gap scores, and retrieve coaching content, substituting human cognitive and manual work with automated information processing and retrieval mechanisms.
2Loss of information
If supervisors manually search for coaching content, then relevant materials can be found, but the process becomes complicated and inaccurate due to lack of indexing
Solution Approach 1:
The algorithm acts as an intermediary between the interaction data and the coaching content repository. It processes interaction data to identify knowledge gaps, then queries the knowledge base for relevant articles, serving as a mediator that connects agent performance data with appropriate coaching materials through automated information retrieval.
Solution Approach 2:
The system changes parameters by transforming unstructured interaction data into structured knowledge gap scores, then using these scores to retrieve coaching content with specific relevance thresholds. The algorithm modifies the state of data from raw interaction transcripts to quantifiable performance metrics that can be systematically matched with coaching materials.
3Productivity
If automated algorithms are used to identify knowledge gaps and provide coaching content, then supervisor time is reduced and consistency is improved, but the system complexity increases
Solution Approach 1:
The system segments the coaching process into distinct automated components: interaction data collection, knowledge gap score calculation, topic identification, and coaching content retrieval. Each segment is handled by specific algorithmic functions, dividing the complex overall process into manageable, automated stages that can be executed systematically without human intervention.
Data Source
AI summary
Coaching systems and methods, and non-transitory computer readable media, include analyzing an agent's interactions to identify knowledge gaps and specific topics where an agent has difficulties. An algorithm uses bootstrap sampling to verify that an agent's scores are significantly different from other agents' scores. The algorithm further uses a mutual information score to find topics that are associated with interactions having a high knowledge gap score.