Hybrid Service Recovery: Design for Seamless Inquiry Handovers between Conversational Agents and Human Service Agents

Abstract

The effort of companies to deploy conversational agents (CAs) for customer self-service has been renewed due to their recent technological improvements. Despite their efficiency in processing recurring simple customer inquiries, limited capabilities of CAs to handle complex inquiries still lead to service failure and unsatisfied customers. Therefore, we propose a hybrid service recovery strategy with real-time handovers of inquiries from CAs to human service agents (HSAs), if CAs' capabilities are exceeded. Following a Design Science Research (DSR) approach, we present design principles (DPs) for the inquiry handover scenario, based on meta-requirements (MRs) derived from literature and expert interviews. By evaluating the DPs via prototype instantiation and process modulation, the suitability and interdependence of CAs' information collection activities and information presentation for handover could be verified.

Supporting Online Customer Feedback Management with Automatic Review Response Generation

“…Still, the user is supposed to believe that they are interacting with a genuine business representative. This has led to intensified research on smooth or seamless handovers from human agents to bots [26] and how the perception of interacting with a bot develops [27].…”

“…Information Systems (IS) research has recognized the potential of applying NLP-based bots-as-a-service [21,22,23]. The research explores the use of such technologies in face-to-face settings [24,25], synchronous chat [26,27], as well as asynchronous communication like answering emails [24].…”

Claim success, but blame the bot? User reactions to service failure and recovery in interactions with humanoid service robots

“…In order to retain users, organizations need to assess which recovery strategy is most effective in interactions with service robots [9]. Prior studies that have investigated different recovery strategies in interactions with service robots have thus far mainly focused on robot-initiated actions [26,31]. Most commonly, research has examined the effectiveness of apologies and explanations [5,9].…”

Task Delegability to AI: Evaluation of a Framework in a Knowledge Work Context

“…Hybrid collaboration involves at least two actors (at least one being human and one AI) having the same understanding of a shared goal and working together toward this mutual goal [18,19,6]. Examples of Human-AI collaboration can be seen in clinical decision support systems and customer service chatbots [19,20]. There are many advantages to Human-AI collaboration, e.g., it improves group performance [21,22], inter alia.…”