Adapting To Customer Initiative: Insights From Human Service Encounters

As more and more service channels migrate from "live" human providers to automated "bot" systems, there is a need to understand the interactional machinery behind human-human service encounters upon which to best design semi-automated interactions. This chapter uses conversation analysis to examine some of the interactional patterns occurring in call center service encounters that are consequential for their outcomes. We discuss how the first few turns of the interaction projects a trajectory based on how the customer adheres to or deviates from the structure of the institutional service opening. We discuss how information is interactionally structured which informs the design of more natural exchanges and unpackages collaborative understanding. Finally, we examine the call's closing and analyze the practices of disengagement that signal customer satisfaction or discontent. The findings point to the ways that natural human service interaction practices can be applied to conversational system design at two levels: the turn level to enable appropriate next responses, and the activity level to understand the interactional trajectory.