A Measure for Dialog Complexity and its Application in Streamlining Service Operations

TL;DR

This paper introduces a dialog complexity measure to analyze customer-service interactions, enabling better understanding, streamlining operations, and improving agent performance through data-driven insights.

Contribution

It proposes a novel, general method to quantify dialog complexity and demonstrates its application in analyzing and improving service operations.

Findings

Dialog complexity correlates with customer needs and interaction types.

The measure helps identify routine versus extraordinary interactions.

Application of the measure improves service efficiency and agent evaluation.

Abstract

Dialog is a natural modality for interaction between customers and businesses in the service industry. As customers call up the service provider, their interactions may be routine or extraordinary. We believe that these interactions, when seen as dialogs, can be analyzed to obtain a better understanding of customer needs and how to efficiently address them. We introduce the idea of a dialog complexity measure to characterize multi-party interactions, propose a general data-driven method to calculate it, use it to discover insights in public and enterprise dialog datasets, and demonstrate its beneficial usage in facilitating better handling of customer requests and evaluating service agents.