Multichannel Voice-of-Customer Framework for Taxonomy Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions fail to integrate and analyze voice-of-the-customer (VOC) data across multiple channels, preventing comprehensive understanding of customer interactions and trends.
Innovation Solution
A multichannel framework that integrates VOC data from various channels using a machine learning model, applies a multi-level taxonomy for clustering and mapping themes, and stores interaction reasons in a database for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate solutions are used for different interaction channels, then each channel can be analyzed independently, but integration of all VOC data is prevented
Solution Approach 1:
The patent merges multiple separate VOC analysis solutions into a single unified multichannel framework that processes data from phone, online chat, email, virtual assistants, and social media channels simultaneously. The system integrates these diverse channels through a common architecture using machine learning models and multi-level taxonomy to achieve comprehensive VOC analysis across all customer interaction touchpoints.
2Adaptability or versatility
If a unified multichannel framework is implemented, then comprehensive VOC data integration is achieved, but system complexity increases
Solution Approach 1:
The patent segments the complex multichannel VOC analysis system into distinct functional modules: data collection module for gathering VOC data from multiple channels, machine learning model execution module for processing transcripts, multi-level taxonomy clustering module for organizing themes, and interaction reason mapping module for generating insights. This modular segmentation manages system complexity while maintaining comprehensive multichannel integration capabilities.
Solution Approach 2:
The patent introduces a multi-level taxonomy as an intermediary structure that bridges diverse VOC data from different channels. The taxonomy provides a standardized hierarchical framework (with at least four levels) that mediates between raw multichannel data and final analysis outputs, enabling systematic organization and comparison of themes across phone, chat, email, and social media interactions without requiring complex custom integration logic for each channel pair.
3Measurement precision
If machine learning models are used for transcript analysis, then analysis depth is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by implementing multi-level taxonomy clustering before final interaction reason mapping. The system pre-organizes themes into a hierarchical structure with at least four levels during an initial processing phase, creating a ready-to-use framework that accelerates subsequent analysis. This preliminary organization of themes reduces the computational burden during real-time or near-real-time VOC data processing, balancing accuracy with processing efficiency.
Data Source
AI summary
A method for integrating customer interaction data from a plurality of channels and for a plurality of customers includes receiving a plurality of customer interaction records, each record associated with a channel and an identifier of a customer, each record including a customer interaction transcript; providing the plurality of customer interaction transcripts to a machine learning model; causing execution of the machine learning model, resulting in a model output including an interaction theme and an interaction summary associated with each one of the customer interaction transcripts; clustering the plurality of themes using a multi-level taxonomy, resulting in a plurality of clustered themes associated with each one of themes; mapping the pluralities of clustered themes and the plurality of interaction summaries, resulting in an interaction reason associated with each one of the customer interaction records; storing the interaction reason associated with each one of the customer interaction records in a database.