Automated Contact Center Testing via Persona-Based Query Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contact centers face challenges in effectively testing automated customer response systems, which often lead to inadequate system testing, customer frustration, and retention issues due to the difficulty in evaluating the accuracy and understandability of chat and voicebots.
Innovation Solution
An automated system and method for testing automated customer response systems that uses a 'conversation multiplier' to generate queries based on 'personas' to mimic real-world interactions, analyzing responses for context, dialect, and formality, and producing results on the appropriateness of the responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual scripting of testing routines is used for each desired test function, then testing can be performed, but the process becomes complex and time-consuming
Solution Approach 1:
The testing system performs self-service by automatically generating test queries and evaluating responses without requiring manual scripting for each test function. The system uses a conversation multiplier to automatically create varied queries and a response evaluator to automatically assess accuracy, eliminating the need for manual test routine scripting while maintaining comprehensive testing coverage.
Solution Approach 2:
The system changes parameters by using a conversation multiplier that generates queries with varying complexity, context, and phrasing based on persona characteristics. This allows the same testing framework to evaluate multiple scenarios by modifying query parameters rather than creating separate manual test scripts for each scenario.
2Productivity
If automated response systems are used to handle customer queries, then productivity is improved, but the systems often perform inadequately for most customers
Solution Approach 1:
The system implements feedback by using a response evaluator that automatically assesses the accuracy and understandability of automated responses. The evaluator compares system responses against expected outcomes and provides feedback on performance, allowing continuous improvement of the automated response system's accuracy while maintaining high productivity.
3Productivity
If automated response systems are deployed, then operational efficiency is improved, but customer frustration increases due to difficulty in understanding and use
Solution Approach 1:
The system applies dynamics by using a conversation multiplier that generates queries with varying levels of complexity and phrasing based on different personas. This allows the testing system to evaluate whether automated responses adapt to different customer communication styles and difficulty levels, ensuring operational efficiency while maintaining ease of operation across diverse user scenarios.
Data Source
AI summary
A system and method for testing of automated contact center customer response systems using a customer response testing system and a real time conversation engine, wherein the customer response testing system generates simulated human queries using persona profiles, sends test cases containing those queries to a contact center under test, and receives and analyzes the responses to determine whether the contact center's automated response systems understand the queries and respond appropriately.