Understanding the user satisfaction and loyalty of customer service chatbots

The artificial intelligence (AI) chatbot is emerging as a significant corporate customer-facing application, potentially increasin customer service efficiency while reducing costs. However, little work has sought to assess the quality of service they provide consumers. This study applies the e-service quality by incorporating conversational AI quality to predict users' satisfaction and loyalty to customer service chatbots. The proposed model was empirically evaluated using survey data collected from 219 users responding about their perceptions of customer service chatbots. The findings indicate that AI chatbot service recovery quality and AI chatbot conversational quality significantly influence user satisfaction. On the other hand, core AI chatbot service quality and satisfaction significantly influenced chatbot user loyalty. This study contributes to researchers and practitioners by proposing and evaluating a more comprehensive chatbot e-service quality that combines both fundamental (core service and service recovery qualities) and human-like (conversational quality) aspects of e-service. The results are of value in devising future AI chatbot services and related strategies.     

Dissecting the mixed effects of human-customer service chatbot interaction on customer satisfaction: An explanation from temporal and conversational cues

Human-customer service chatbot interaction has had mixed effects on customer satisfaction in sales-related services. To dissect these conflicting results, three 2 (temporal cues: instant versus anthropomorphic delayed response) × 2 (conversational cues: task- versus social-oriented) between-group experiments were conducted. The results revealed the complex effects of temporal cues, conversational cues, and their interactions on perceived warmth, perceived competence, and satisfaction. And, perceived warmth and competence mediated the effects of temporal and conversational cues on satisfaction. Furthermore, the product attribute type (functional versus experiential) affected the relationships between these variables. These findings provide two explanatory perspectives (temporal and conversational cues) for dissecting the mixed effects of human-customer service chatbot interaction on customer satisfaction, and provide practical insights for improving customer service chatbots and enhancing customer satisfaction.     

Whether to trust chatbots: Applying the event-related approach to understand consumers’ emotional experiences in interactions with chatbots in e-commerce

Chatbots can be used in marketing services to substantially improve the consumer experience. Based on cognitive appraisal theory, this study applied an event-related potential (ERP) approach to investigate consumers’ emotional experiences and consumer trust in passive interaction with chatbots versus humans, taking into account objective or subjective tasks in e-commerce. The results showed that chatbot (vs. human) service interactions automatically drew more consumer attention at the subconscious stage (i.e., a larger P2); consumers purposefully allocated more resources to regulate the negative emotions elicited by chatbots at the conscious stage (i.e., a larger LPP); and there was a lower trust in chatbots than in humans. Moreover, under subjective tasks, the differences between chatbots and human agents in emotional experience (as reflected by LPP) and trust were amplified. The findings will encourage e-retailers to improve the emotional service experience of their chatbots and prioritize the application of chatbots for objective tasks in customer service.     

It is not merely a chat: Transforming chatbot affordances into dual identification and loyalty

Artificial intelligence (AI)-powered service chatbot functionality is changing to more effectively connect with customers in the era of digital marketing. As such, an understanding of how to enhance user perceptions and behaviors through interface design has become crucial. Using affordance actualization theory and social identity theory as a theoretical lens, this study explored how chatbot affordances affect consumer brand loyalty. Data were collected from 369 respondents who had at least one conversation with AI chatbots from a predetermined list of banks in Taiwan; the collected data were then analyzed using structural equation modeling. The results revealed that anytime–anyplace connectivity, information association, visibility, and interactivity affordances provided by chatbots positively influenced chatbot exploitation, which in turn affected perceived brand competence and perceived brand warmth. Furthermore, perceived brand competence influenced customer–chatbot identification and customer–brand identification, whereas brand warmth influenced customer–brand identification. Customer–chatbot identification also influenced brand loyalty both indirectly—through customer–brand identification—and directly. These findings are useful for assisting managers in various industries in the application of AI technologies to implement digital transformation strategies and improve customer services.     

Exploring the impact of chatbots on consumer sentiment and expectations in retail

Retailers are increasingly using conversational AI (chatbots) for customer service due to the perceived benefits and reduced operational costs of this emerging technology. Yet our understanding of how consumers perceive interactions with chatbots, and how these interactions may influence other consumer service programs, remains limited. This paper investigates the differences in consumers' sentiments towards chatbots across retail sectors, and the influence chatbots have on consumers’ sentiments and expectations towards other service interactions with online human agents. Using a hybrid automated sentiment analysis approach, we identify that (1) overall sentiment towards bots are less negative than sentiment towards online human agents; (2) these sentiments differ across fashion and telecommunications sectors, and finally (3) sentiments towards online human agents in both sectors become more negative after a retailer implements a chatbot.     

An Integrated Model of Quality for Mass Services in the Context of the Ghanaian Retail Banking Sector

This study draws upon the service literature and operationalizes the investment model in the services domain to examine factors that contribute to customers’ desire to maintain service relationships with firms (i.e., retail banks) in an emerging economy in sub‐Saharan Africa: Ghana. We empirically test the explanatory power and robustness of the investment model using 218 customers of various banking institutions in Ghana. Results from partial least squares—structural equation modeling (PLS‐SEM)—reveal that (1) service quality is positively associated with customer satisfaction, investment size, and customer commitment; and (2) while customer satisfaction is positively associated with customer commitment, both investment size and quality of attractive alternatives are not associated with customer commitment. Interestingly, we found investment size to have a positive association with customer commitment only when fully mediated by customer satisfaction. Our model reveals that service quality and customer satisfaction predict 79.3% of the variation in customer commitment toward maintaining a service relationship with their retail banks. Discussion and managerial implications conclude the article. © 2015 Wiley Periodicals, Inc.     

Appreciation vs. apology: Research on the influence mechanism of chatbot service recovery based on politeness theory

Failures in human-chatbot interactions are becoming inevitable such as failure of chatbots to maintain contextual awareness. Thus, an effective service recovery strategy is essential for e-commerce enterprises to restore customers and resolve their complaints. Therefore, based on politeness theory, this study explores the fundamental mechanism and boundary conditions for a chatbot politeness strategy (appreciation vs. apology) on consumers' post-recovery satisfaction using four scenario-based experiments. The results indicate that establishing a good human-chatbot relationship (appreciation) is a more effective recovery strategy than admitting the chatbot's limited competence (apology) in redressing service failures; face concern mediates the effect of the politeness strategy on post-recovery satisfaction; and time pressure plays a moderating role in the effect of the politeness strategy on face concern and post-recovery satisfaction. This study extends the research of politeness theory and face concern in the field of chatbot marketing, and provide practical guidance for e-commerce enterprises to deal effectively with chatbot failure.     

When do AI chatbots lead to higher customer satisfaction than human frontline employees in online shopping assistance? Considering product attribute type

The increasing adoption of AI chatbots in online shopping assistance, as a complement or substitute for human frontline employees (HFLEs), leads to the question whether HFLEs perform better than AI service robots and why. From the perspective of product attribute type (experiential/functional) and focusing on customer satisfaction, this study explores how the impact of service agent on customer satisfaction varies along with product attribute type. A scenario-based experiment was designed and completed by 567 participants. Although HFLEs lead to higher customer satisfaction when the product attribute is experiential, AI chatbots perform better than HFLEs when the product attribute is functional. We make use of perceived information quality, perceived waiting time, and positive emotions, three determinants of customer satisfaction, to explain the variation of the role of different service agent types. The findings offer useful implications for companies when selecting service agent types in online shopping assistance.     

Forecasting artificial intelligence on online customer assistance: Evidence from chatbot patents analysis

The main goal of this research is to provide a comprehensive understanding of the actual progresses in artificial intelligence, with emphasis on chatbots as emerging forms of customer assistance in online retailing. Drawing upon an analysis of the chatbot patents in the past 20 years, our findings show the increasing technology push towards the adoption of new conversational agents based on natural language. Findings also highlight the extent to which the research and development efforts are attempting to improve artificial intelligence systems that characterize chatbots. To this end, technology advancements are mainly focusing on: (i) improving chatbot ability to automatically draw inferences on users starting from multiple data sources, and (ii) using consumers’ knowledge adaptively to provide more customized solutions. Finally, results show the tight relationship between the digital assistants’ analytical skills and their ability to automatically interact with the users.     

The happy versus unhappy service worker in the service encounter:Assessing the impact on customer satisfaction

This paper assesses the impact of the service worker's display of emotions (i.e., one aspect of functional service quality) on customer satisfaction under the conditions of different levels of technical service quality by means of an experimental approach (N=600), in which display of emotions (unhappiness vs. happiness) and technical service quality (poor vs. good) were manipulated. The results indicate that the impact of the service worker's emotional display behavior on customer satisfaction is contingent on the level of technical service quality, in the sense that customer satisfaction is affected only when technical service quality is good rather than poor. Encouraging a positive display (e.g., by a smile policy), which many service firms do, is thus not a panacea for improved customer satisfaction. The moderating effect is explained in terms of service encounter congruency, which influences the mediated process by which emotional displays by service workers come to affect customer satisfaction.     

The interrelationships among service quality,customer satisfaction,and customer loyalty: Examination of the fast-food industry

The purpose of this study was to use customers’ perspectives to develop a conceptual model for improving service quality and enhancing customer satisfaction and loyalty. The 197 samples were selected from customers of leading fast food franchise enterprises in Taiwan. The research results indicate that service quality positively influences customer satisfaction and customer loyalty. Service quality positively influences customer loyalty through customer satisfaction. The research findings indicate that managers need to ensure they have efficient service quality to enhance customer satisfaction and customer loyalty levels.     

Employee encouragement of self-disclosure in the service encounter and its impact on customer satisfaction

In the service encounter, the employee must often encourage customer self-disclosure (i.e., revealing of personal information) to be able to match the customer's needs with what the firm has to offer. This study uses an experimental approach to manipulate employee encouragement of self-disclosure (low vs. high) to explore its impact on the customer. It was found that encouraging self-disclosure enhanced customer perceptions of customization, employee effort, own effort, privacy concerns, and employee humanness, and that these responses influenced customer satisfaction. In addition, because many firms are beginning to replace human employees with various forms of virtual agents (and it has been argued that we humans may find it less threatening to self-disclose to such agents), the identity of the employee (virtual agent vs. human employee) was manipulated, too. The identity factor, however, did not influence customers' responses.     

Who's got my back? Comparing consumers' reactions to peer-provided and firm-provided customer support

This study demonstrates that when an individual encounters a product-related problem, fellow consumers (i.e., one's peers) have a unique advantage in providing social support to the affected consumer. Specifically, we find that social support can be a dominant driver of consumer satisfaction when the risk of customer defection is at its highest (i.e., following an unsuccessful attempt to solve the consumer's problem). Using real-world data from an online support community, a pilot study finds that if the problem that a consumer faces goes unsolved, satisfaction is greater when consumers receive peer-provided versus firm-provided support. Study 1 replicates this finding in a controlled experiment that realistically simulates an actual customer support incident in real-time. Study 2 identifies social support as the mechanism that underlies this effect and investigates whether firm employees can take steps to appear more customer-like and thereby replicate the advantage of peer-provided support. Finally, Study 3 reveals an alternative strategy (i.e., utilizing multiple employees) that firms can use to enhance social support and provides evidence that peer-provided support not only enhances satisfaction but also positively influences consumers' behavioral intentions.     

In bot we trust: A new methodology of chatbot performance measures

Chatbots are used frequently in business to facilitate various processes, particularly those related to customer service and personalization. In this article, we propose novel methods of tracking human-chatbot interactions and measuring chatbot performance that take into consideration ethical concerns, particularly trust. Our proposed methodology links neuroscientific methods, text mining, and machine learning. We argue that trust is the focal point of successful human-chatbot interaction and assess how trust as a relevant category is being redefined with the advent of deep learning supported chatbots. We propose a novel method of analyzing the content of messages produced in human-chatbot interactions, using the Condor Tribefinder system we developed for text mining that is based on a machine learning classification engine. Our results will help build better social bots for interaction in business or commercial environments.     

Hospice Informal Caregivers' Perception of a Good Death Experience: A SERVQUAL Analysis

ABSTRACT

This research study explores hospice informal caregivers' perceptions of service quality and a good death experience during end-of-life care. It demonstrates how service dimensions of the SERVQUAL analysis affect overall customer satisfaction. This study addresses contrast and dissonance theory as relative to marginal gaps in actual service performance and the effect on overall customer satisfaction. The research indicated that while reliability was the core of the service outcome, peripheral variables (e.g., assurance, empathy, responsiveness, and tangibility) integrated emotions and feelings into the hospice service process that equated to a positive disconfirmation of expectations and a good death experience.     

Effect of social support on customer satisfaction and citizenship behavior in online brand communities: The moderating role of support source

Customers often join online brand communities to seek support from others when they encounter product problems. Some customers who receive good social support exhibit customer citizenship behavior. This study develops a theoretical model to investigate how social support influences customer citizenship behavior through customer satisfaction with firms in online brand communities. Moreover, the moderating role of support source (i.e., firms vs. other customers) is measured. The research model is tested by using the Partial Least Squares technique. The results show that informational and emotional support significantly affects the customer citizenship behavior of providing feedback to the firm, recommendations, and helping other customers through customer satisfaction in online brand communities. Moreover, informational and emotional support from firms and other customers exert different effects on customer satisfaction.     

The Moderating Influence of Broad‐Scope Trust on Customer–Seller Relationships

Trust relates not only to customer trust in individual companies (i.e., narrow‐scope trust) but also to the broader business context in which customer–seller relationships may develop (i.e., broad‐scope trust [BST]). Based on two surveys comprising 1155 bank consumers and 817 insurance consumers, respectively, this study investigates the moderating influence of BST on relationships between satisfaction, narrow‐scope trust, and loyalty and also examines the direct influence of BST on these variables. The results indicate that whereas BST negatively moderates relationships between satisfaction and narrow‐scope trust and between narrow‐scope trust and loyalty, BST positively moderates the relationship between satisfaction and loyalty. In addition, it is demonstrated that BST positively influences customer satisfaction and narrow‐scope trust. © 2012 Wiley Periodicals, Inc.     

Examining the Customer Experience of Using Banking Chatbots and Its Impact on Brand Love: The Moderating Role of Perceived Risk

In today’s digitally enabled world, banks offer chatbots to improve customer service. This research looks at chatbots as a form of an information system. It observes the effect of the three quality dimensions of the information system success model on customer experience, further leading to love for the bank brand. As chatbots are a new technology platform, consumers may find some risk in its use. Hence, the moderating role of perceived risk between the three quality dimensions and customer experience is also observed. The survey method was employed to conduct the study. Data were collected from 258 respondents. The results provide a definite direction for banks to strengthen the consumer–brand relationship by offering chatbots suiting their customers’ expectations.     

Differentiation Via Technology: Strategic Positioning of Services Following the Introduction of Disruptive Technology

Introducing a “disruptive” technology into an existing service market provides new opportunities for firms and customers, often altering the nature of the market. Consequently, new technology often destabilizes market equilibrium, forcing firms to consider the role new technology will play in determining the new market structure. In this paper, we focus on understanding how significant new technology forces firms to evaluate and renegotiate their market positions in a service industry newly infused with technology (i.e., brokerage market). Specifically, we examine how firms integrate technology into the service process to create new forms of interaction. We used content analysis and a combination of structuralist methods including multidimensional scaling analysis (MDS), clustering methods, and property-fitting regression techniques (ProFit) to examine how technology changes the service process, and how firms leverage these changes for positioning. Our research demonstrates a strategic approach to positioning that suggests (1) introducing significant technology opens new positioning avenues by providing new service interactions, (2) technology must be leveraged via links to customer value sets, and (3) firms that leverage technology through customer values outperform their competitors.     

Employee versus supervisor ratings of performance in the retail customer service sector: Differences in predictive validity for customer outcomes

Based on data from Maxham and Netemeyer [Maxham, J. G. and R. G. Netemeyer (2003). “Firms Reap What They Sow: The Effects of Employee Shared Values and Perceived Organizational Justice on Customer Evaluations of Complaint Handling,” Journal of Marketing, 67, 46-62], the authors present two field samples to examine predictive validity differences of service employee ratings of their performance versus supervisor ratings of employee performance with respect to customer satisfaction and customer likelihood of spreading positive word-of-mouth (WOM) after a service recovery attempt. The results generally show that supervisor ratings are more strongly positively related to customer satisfaction and WOM than are employee ratings of their own performances. The results also show that both supervisor ratings and employee ratings are related to customer satisfaction and WOM in a curvilinear fashion (as well as linear fashion). Employee extra-role performances (toward customers and the firm) show increasing returns at the higher levels of performance, and employee in-role customer performance generally shows a decreasing return at the higher level of customer in-role performance.These results suggest two managerial implications. First, supervisor ratings of customer service employee performances may be the preferred form of measurement for predicting customer outcomes. Second, maximizing in-role performance inputs may have decreasing returns for customer evaluations in the service recovery context; but maximizing extra-role performance inputs may actually “delight” customers, i.e., increasing returns for customer evaluations.