Sage Journals: Discover world-class research

Abstract

Service organizations, emboldened by the imperative to innovate, are increasingly introducing robots to frontline service encounters. However, as they augment or substitute human employees with robots, they may struggle to convince a distrusting public of their brand’s ethical credentials. Consequently, this article develops and tests a holistic framework to ascertain a deeper understanding of customer perceptions of frontline service robots (FLSRs) than has previously been attempted. Our experimental studies investigate the effects of the (1) role (augmentation or substitution of human employees or no involvement) and (2) type (humanoid FLSR vs. self-service machine) of FLSRs under the following service contexts: (a) value creation model (asset-builder, service provider) and (b) service type (experience, credence). By empirically establishing our framework, we highlight how customers’ personal characteristics (openness-to-change and preference for ethical/responsible service provider) and cognitive evaluations (perceived innovativeness, perceived ethical/societal reputation, and perceived innovativeness-responsibility fit) influence the impact that FLSRs have on service experience and brand usage intent. Our findings operationalize and empirically support seminal frameworks from extant literature, as well as elaborate on the positive and negative implications of using robots to complement or replace service employees. Further, we consider managerial and policy implications for service in the age of machines.

Keywords

service robots service experience brand usage intent service innovativeness ethical/societal reputation

To improve frontline service encounters, service providers increasingly utilize autonomous service robots infused with artificial intelligence (AI) to augment or replace the role of human employees. Traditionally, customer-employee interactions at the social interface have been solely responsible for developing service encounters that create brand equity by enhancing the customer experience (Brakas, Schmitt, and Zarantonello 2009; Hepola, Karjaluoto, and Hintikka 2017) and driving brand usage intent (Hollebeek, Glynn, and Brodie 2014). It has long been asserted that “the people make the brand” (Hurrell and Scholarios 2014, p. 54) and that frontline employees play an important role in defining frontline service encounters (De Keyser et al. 2019; Voorhees et al. 2017). In this article, we attempt to answer the following questions:

Research Question 1: How will customers react when employees’ roles are augmented or substituted by frontline service robots (FLSRs)?

Research Question 2: What are the implications for service in terms of the innovativeness and ethical/social responsibility aspects of such augmentation or substitution?

While it is true that we do not yet have commercially viable robots that can walk dogs or run errands, it has been predicted that by 2025, service providing robots “will be melded into numerous service experiences” (van Doorn et al. 2017, p. 44). Their introduction will fundamentally change the interactions customers have with service organizations and the functions and responsibilities of all actors involved in service encounters (De Keyser et al. 2019; Larivière et al. 2017; Wirtz 2019). For the purpose of this article, service robots are defined as “system-based autonomous and adaptable interfaces that interact, communicate and deliver service to an organization’s customers” (Wirtz et al. 2018, p. 909). In contrast to other forms of AI, which are beyond the scope of this article, intelligent physically embodied FLSRs can have meaningful social interactions with customers and can therefore be considered as service agents (Jörling, Böhm, and Paluch 2019; Mende et al. 2019; van Doorn et al. 2017). FLSRs can be categorized from humanoid (anthropomorphized robots imbued with humanlike characteristics) to nonhumanoid (e.g., an artificially intelligent reception FLSR at a restaurant; Wirtz et al. 2018). In this article, we refer to a nonhumanoid FLSR that possesses the same ability, functionality, and intelligence as a humanoid FLSR, as a self-service machine.¹ They are intuitive, interact, and communicate with customers in a similar way to employees. Humanlike features can inspire trust and bonding, however, uncanny valley theory (Mori et al. 2012) suggests that the introduction of highly humanlike robots might create “feelings of eeriness or a threat to (a customer’s) human identity” (Mende et al. 2019, p. 539). The deployment of self-service machines that perform the same functions as humanoid FLSRs may not engender the same response. Therefore, service providers seeking to introduce FLSRs face the challenge of understanding both positive and negative implications that may follow and need to gain insights into how FLSRs will influence a customer’s service experience or intent to use a brand (Hollebeek, Glynn, and Brodie 2014). Furthermore, in seeking to answer these questions, rather than isolate our research to a particular service setting, we adopt a holistic approach by investigating how customer perceptions change according to service type (e.g., for a credence service in comparison to an experience service).

FLSRs conversant with big data analytics and biometrics are providing a variety of innovations that significantly alter service settings (Mende et al. 2019; Wirtz et al. 2018). Nonetheless, they are simultaneously threatening human jobs (Harris, Kimson, and Schwedel 2018) and creating ethical and societal challenges that might lead to public distrust, inhibiting broader adoption and customer engagement with FLSRs (Huang and Rust 2018). This may have a negative influence on the service provider. Given the above it is surprising that from a customer-centric perspective, a dearth of empirical academic research focuses on whether the introduction of FLSRs may create a dichotomy between negative perceptions relating to potential ethical and societal challenges and more positive perceptions of innovative service encounters. This article addresses these gaps in the literature and responds to demands for more research prioritizing frontline service encounters and the role that technology plays in advancing service provision (e.g., Hollebeek, Andreassen, and Sprott 2018; Ostrom et al. 2015).

In two experimental studies, we extend and refine the work of Larivière et al. (2017) by exploring how the expected positive impact of innovativeness shaped by FLSRs influences customer perceptions of a frontline service encounter. In Study 1, we simultaneously investigate the potential negative effects on the perceived ethical/societal reputation and positive effects on the perceived innovativeness of a service organization that uses an FLSR to either replace or complement a human employee in frontline service encounters. Subsequently, in Study 2, we build upon our initial results, incorporating new service types (credence vs. experience) and compare responses from customers who have been exposed to two AI types (humanoid FLSR vs. nonhumanoid self-service machine). We focus on perceived innovativeness-responsibility fit in an attempt to develop a holistic approach to understanding customer perceptions of FLSRs. We consider the influence of individual characteristics, particularly an individual’s openness-to-change (Schwartz 2003), as well as an individual’s preferences for an ethical/responsible service provider (Ramasamy and Yeung 2009), as the adoption of innovative services may be influenced by an individual’s preferences for innovation (Hoffmann and Soyez 2010). To the best of our knowledge, no other study has comprehensively explored these factors.

In this article, we make several important contributions to the literature. First, we address the urgent need to better understand the relationship between FLSRs, service providers, and brands (Mende et al. 2019). Second, we also answer calls for further research to explore the net effect that automated service interactions have on customers, including the influence of positive as well as negative factors (Čaić, Odekerken-Schröder, and Mahr 2018; Hollebeek, Jaakkola, and Alexander 2018). Third, not only does our work address the need for further research into the factors that drive customer acceptance or conversely mistrust of AI and FLSRs, it also provides more knowledge about how FLSRs can be better integrated into the servicescape (Wirtz et al. 2018). Specifically, we emphasize the importance of jointly considering the barriers and drivers of AI adoption in the service context. These factors may be technology-related such as innovativeness and/or customer-related characteristics such as values. Fourth, we develop and test a holistic conceptual framework by extending and refining Lariviere et al.’s (2017) conceptual Service 2.0 model. Larivière et al.’s (2017) model provides a comprehensive overview that considers the impact of automated technologies and provides a strong foundation for an empirical investigation into the effects of robots in frontline service encounters. Finally, we outline important managerial implications that highlight the importance of adopting a holistic approach to the introduction of FLSRs.

Conceptual Foundations

Role of Robots in Frontline Services

Larivière et al.’s (2017) ambitious conceptual paper recently sought to revamp and update service marketing’s conceptualization of the service encounter. They highlight two important roles that robots can play in customer-facing service scenarios: (a) augmentation (assisting and complementing human employees) and (b) substitution (replacing human employees) which in this article we refer to as role of FLSR. By building on extant literature, they also identify two different business models that create value (which we refer to as value creation model) where robots may complement or replace humans: asset-builder (businesses/service organizations that deliver physical goods including retailers) and service provider (e.g., hotels, restaurants and airlines or airports). We focus on both asset-builders, which through physical infrastructure and marketing typically deliver value much as a retailer does—and on service providers who deliver value for the most part through the skill of their employees. It is in these types of organizations, where technological augmentation is most likely to flourish, that employees with a strong sense of role clarity, ability, and motivation are seen as key progenitors of innovation. As Larivière et al. (2017, p. 241) state “authentic human touch can help differentiate offerings in the marketplace and display unique brand-building behaviors” which create an experience driven by “sensations, feelings, cognitions, social and behavioral responses that result from interacting with other parties—employees, technology etc.” (Larivière et al. 2017, p. 242).

Customer Cognitive Evaluations of FLSRs

Customers’ perceptions of an FLSR are informed not only by their cognitive evaluation of the service encounter but also by their broader knowledge and understanding of the positive and/or negative consequences of introducing robots to the servicescape. Successful service innovations, for example, can provide real value for customers (Hollebeek and Andreassen 2018; J. Kim et al. 2015). In their study of service design and value creation, Andreassen et al. (2016, p. 22) state “innovation is the new ticket” for organizations seeking to play and stay in the service industry. Therefore, a better understanding of innovative service provision represents a growing priority for both researchers and practitioners (Antons and Breidbach 2018; Patricio, Gustaffsson, and Fisk 2018), particularly in the area of radical service innovations (Goduscheit and Faullant 2018) such as FLSRs. Our focus is microlevel and customer-centric, since customers are ultimately responsible for the success of an innovation (Kunz, Schmitt, and Meyer 2011).

Although FLSRs will be perceived as innovative and provide benefits, their use in frontline service settings also creates a number of ethical and societal implications that may influence customer perceptions. Wirtz et al. (2018) highlight how customers may be concerned about “privacy and security” as robots can gather and store data and remotely connect and share it with other sources. Sensitive customer data collected by FLSRs and stored in the cloud could be hacked by criminals. They also discuss “dehumanization and social deprivation” issues. For example, substituting human carers with robots may dehumanize care, cause emotional concerns, and lead to social isolation, particularly for the elderly (Čaić, Odekerken-Schröder, and Mahr 2018). Intuitively, an apathetic, emotionless, innately cold robot does not seem like the ideal caregiver (Stahl and Coeckelbergh 2016).

Over the last decade, robots have replaced humans by performing automatable tasks on manufacturing assembly lines, raising concerns that there may be job losses in the service sector as increasingly intelligent robots gain the ability to perform cognitive nonroutine manual tasks (Decker, Fischer, and Ott 2017). Huang and Rust (2018) highlight how AI will increasingly take over analytical, intuitive, and eventually empathetic tasks in the future. However, it is not known whether and how customers will react to service providers that are replacing service staff with FLSRs and whether they will consider such behavior as unethical or create concerns that they act poorly in terms of ethical and societal reputation. We explore three different measures which differ conceptually and enable us to gain a holistic understanding of key issues: (1) perceived ethical/ societal reputation, (2) preference for ethical/responsible service providers, and (3) perceived innovativeness-responsibility fit.

Customer Engagement Outcomes

FLSRs have the potential to revolutionize customer engagement by transforming the service experience and influencing the extent to which customers intend to use a brand (brand usage intent). For example, Natwest Bank is testing Cora and Finistra has developed Sophia, which are both highly lifelike digital human bots empowered with AI and deep learning that can detect human emotions and physically react with their own facial expressions (Joyce 2018). In the United States, Lowe’s hardware stores are testing FLSRs that answer customers’ questions and help them navigate around a store (Rafaeli et al. 2017).

For frontline service encounters, robots are likely to play an increasingly important role in enhancing the customer experience in the future; however, to the best of our knowledge, no extant research has focused on FLSRs and the brand. We focus on two different customer-centric outcomes, the service experience and brand usage intent. Service experience is conceptualized based on a customer’s perceptions of the experience they have with a service provider and draws from the brand experience literature (Brakus et al. 2009), while brand usage intent is defined as “customers’ differential response between a focal brand and an unbranded product when both have the same level of marketing stimuli and product attributes” (Hollebeek, Glynn, and Brodie 2014, p. 163). Scholars have reported that an innovative service experience and brand usage intent increase customer engagement (Hollebeek, Glynn, and Brodie 2014; Lin 2015).

Hypothesis Development

The introduction of FLSRs to service settings is still a relatively novel experience for customers; therefore, the pathways that influence service outcomes such as the service experience and brand usage intent have not been completely mapped. A variety of sometimes opposing theoretical foundations can be used to explain parts of the picture; however, there is a need for an overarching framework to gain a more holistic understanding. Therefore, based on the preceding foundations, we propose a holistic conceptual framework that is presented in Figure 1. Additional key components of the framework are described in the paragraphs that follow.

Figure 1.

A holistic framework for understanding customers’ perceptions of frontline service robots.

Perceived Innovativeness, Perceived Ethical/Societal Reputation, and Role of FLSR

Perceived innovativeness involves a customer’s receptiveness and predisposition to a service provider adopting new ideas and launching new products and/or services (e.g., Hurley and Hult 1998) that result in “novel, creative, and impactful ideas and solutions” (Kunz, Schmitt, and Meyer 2011, p. 817). Existing studies suggest that service providers who demonstrate innovativeness create positive behavioral intentions (Eisingerich and Rubera 2010; Jin, Line, and Merkebu 2016) and that perceived innovativeness increases value in service settings (J. Kim et al. 2015; Lin 2015). Conceptually, it can be argued that customers will perceive that introducing FLSRs is innovative, particularly if FLSRs completely replace (substitute) rather than complement/augment existing employees. Therefore:

Hypothesis 1: The effect of FLSRs on perceived innovativeness is stronger for human employee substitution than human employee augmentation.

The concept of perceived ethical/societal reputation relates to customer perceptions of a service provider’s level of engagement in ethical practices and adherence to socially responsible principles, which contributes to long-term success (Fukukawa, Balmer, and Grey 2007; Stanaland, Lwin, and Murphy 2011). Thus, perceived ethical/societal reputation combines customer perceptions of a service provider’s fulfilment of ethical standards and societal responsibilities (Stanaland, Lwin, and Murphy 2011). Such practices are particularly important in the context of AI implementation. Davenport et al. (2020) note the importance of carefully considering ethical implications such as issues surrounding data privacy, biases, or the purpose of AI applications. Qureshi and Syed (2014) suggest that the introduction of robots in the health care sector is “killing off jobs” and could easily turn the perceptions of health workers and patients against such a development. Moreover, the installation of FLSRs on one side of the service interaction effectively removes the relational interplay between two human beings that previously characterized such encounters. This interplay was invariably governed by universal norms and unwritten moral codes (Abela and Murphy 2008). Thus, substituting employees with FLSRs in a previously human-human dyad may be perceived by customers as innovative but perhaps could inadvertently damage a service provider’s ethical credentials. At the same time, there is an expectation that service providers should act in the best interests of society. Substituting willing workers for robots, an act that will leave many unemployed and possibly destitute, is unlikely to be regarded as socially responsible (Barrat 2013; Ford 2015). Therefore, we propose:

Hypothesis 2: The effect of FLSRs on perceived ethical/societal reputation is stronger for human employee substitution than human employee augmentation.

Openness-to-Change

In general, customer adoption of innovative services is influenced by individual characteristics. Their values that are an essential psychographic trait reflect the motivational foundation that guides individual behavior across situations (Schwartz 2012). For example, the level of customer engagement in a brand is dependent on individual values such as uncertainty avoidance (Hollebeek 2018). Schwartz’s (2012) value theory organizes 10 value types in a motivational structure, which can be described by two axes: self-transcendence versus self-enhancement and openness-to-change versus conservation. The circumplex structure reflects a motivational continuum, in which similar value types are located close to each other (Schwartz and Boehmke 2004).

The influence of openness-to-change on an individual’s acceptance of and reasons for adopting innovative services, which in turn stimulates actual adoption behavior, has been established for different contexts. For example, Wang, Dou, and Zhou (2008) show that new product adoption is positively related to the degree of openness-to-change a customer holds and negatively to their preference for traditional products. Hence, high levels of openness-to-change should be associated with favorable behavioral outcomes as a consequence of FLSR implementation, while low levels should be related to negative outcomes. Specifically, as individuals with high levels of openness-to-change form more positive attitudes toward the implementation of innovation (Claudy, Garcia, and O’Driscoll 2015), the mediation via perceived innovativeness in such cases should be stronger. Additionally, as low levels of openness-to-change are accompanied by a preference for traditional and/or conservative products or service provision (Pepper, Jackson, and Uzzell 2009; Wang, Dou, and Zhou 2008), these individuals may be more skeptical toward FLSR implementation. As such, the negative mediation effect through perceived ethical/societal reputation should be weakened for customers with high levels of openness-to-change. Therefore, we propose:

Hypothesis 3: High openness-to-change strengthens the positive effect of FLSRs on perceived innovativeness.

Hypothesis 4: High openness-to-change weakens the negative effect of FLSRs on perceived ethical/societal reputation.

Service Experience, Perceived Innovativeness, and Perceived Ethical/Societal Reputation

The results of previous research in a service context suggest that customer experience is an antecedent of brand engagement and brand equity (Ding and Tseng 2015; Hepola, Karjaluoto, and Hintikka 2017; Lin 2015) and drives brand loyalty (van der Westhuizen 2018). Given the influence of the role of service robots on perceived innovativeness (Hypothesis 1), and perceived ethical/societal reputation (Hypothesis 2), it is likely FLSRs may provide an overall mediated effect on service experience. In their study of service robots, Čaić, Odekerken-Schröder, and Mahr (2018) suggest there is a need to consider both the positive and negative consequences of introducing robots simultaneously in a single study. On the basis of the conceptualization of our framework derived from our literature review, we expect that customer experience with a service organization is subconsciously influenced by both factors during a frontline service encounter. We subsequently propose the following mediating hypotheses:

Hypothesis 5a: FLSRs have a positive indirect effect on service experience via perceived innovativeness.

Hypothesis 5b: FLSRs have a negative indirect effect on service experience via perceived ethical/societal reputation.

Perceived Innovativeness-Responsibility Fit and AI Type

Perceived innovativeness-responsibility fit relates to customer perceptions of apparent incongruences between the innovative aspects of cutting-edge technology (AI) and service providers’ adherence to sound ethical, socially responsible principles. Brand congruency theory suggests that it is important for customer brand associations to be consistent with the behavior of brand owners (Arbouw, Ballantine, and Ozanne 2018; Sjödin and Törn 2006). In a similar manner to brand owners attempting to ensure that product extensions have a good fit with the parent brand (Carter and Curry 2013), service providers must ensure that there is congruency between new FLSRs and their existing brand—in particular, a fit or congruence with ethical and socially responsible activities (de Jong and van der Meer 2017). However, technology providers have frequently been criticized for their poor responsibility records and questionable ethical practices (Vaidhyanathan 2018). For example, Facebook has been condemned for selling personal data to Cambridge Analytica who potentially influenced the results of the U.S. election. In Europe and the UK, Google has faced scorn for not paying tax (Delfanti and Arvidsson 2019). Debate is continuing regarding AI and weapons, where robots and drones could save our armed forces, but create ethical challenges if robots are given the power to kill people without human intervention (Marr and Ward 2019). Such examples create challenges in many customers’ minds regarding FLSRs and perceived innovativeness-responsibility fit. Drawing on conceptual underpinnings from the brand congruency and socially responsible consumption literature (Ramasamy and Yeung 2009; Stanaland, Lwin, and Murphy 2011), as well as recent studies that highlight how uncanny valley theory suggests customers may feel apprehension and unease with humanoid robots (e.g., S. E. Kim, Schmitt, and Thalmann 2019; Mende et al. 2019), we hypothesize:

Hypothesis 6: Humanoid FLSRs have a weaker effect on perceived innovativeness-responsibility fit than self-service machines.

Perceived Innovativeness-Responsibility Fit and Service Type

The credence-experience service typology (Keh and Sun 2018) has been used to categorize services that mainly have credence or experience attributes (Ostrom and Iacobucci 1995). Services differ in terms of the extent to which customers are able to evaluate them, even at the postconsumption stage. Credence services such as those associated with an insurance agency have attributes that are difficult to evaluate prior to or after consumption (Keh and Sun 2018). Experience attributes such as those related to a haircut, a stay at a hotel, or meal at a restaurant can only be evaluated during or after consumption (Chocarro, Cortinas, and Villneuva 2018). As credence services are usually nonstandardized and developed for the needs of an individual customer or family (e.g., holiday insurance may vary according to an individual’s age, where they are travelling to, as well as existing health conditions), their consumption is linked with uncertainty and risk (Mitra, Reiss, and Capella 1999), and they are harder to evaluate in comparison to experience services (Keh and Sun 2018; Ostrom and Iacobucci 1995). As such, when evaluating customer perceptions about the innovativeness and ethical and societal implications of FLSRs, it is pivotal to examine the differential effects of credence versus experience services. Drawing on conceptual foundations from the brand congruency and ethical/social responsibility literature and credence-experience typology, we suggest:

Hypothesis 7: The negative effect of humanoid FLSRs (vs. self-service machine) on perceived innovativeness-responsibility fit is stronger for experience services than credence services.

Preference for Ethical/Responsible Service Providers, Perceived Innovativeness-Responsibility fit, and Service Type

An individual customer is likely to have specific preferences for ethical/responsible service providers, which can be defined as the importance they place on businesses that act in an ethical and socially responsible manner (Ramasamy and Yeung 2009). Theories of socially responsible consumption suggest that customers who have strong ethical values are more likely to be receptive to ethical and pro-environmental products and services (De Groot and Steg 2009; Osburg et al. 2019). Therefore, we would expect an individual’s preferences for an ethical/responsible service provider to influence their perceptions of perceived innovativeness-responsibility fit. Building upon Hypothesis 7 and theoretical underpinnings from the credence and experience service literature, we hypothesize:

Hypothesis 8: High preference for ethical/responsible service providers strengthens the negative effect of humanoid FLSRs (vs. self-service machine) on perceived innovativeness-responsibility fit.

Perceived Innovativeness-Responsibility Fit, AI, Type and Brand Usage Intent

When introducing Hypothesis 5, we described the need to test whether perceived innovativeness and ethical/societal reputation, which are forms of cognitive evaluation, provide an overall mediated effect on the service experience—highlighting the need to simultaneously consider both factors. Following similar arguments, we expect that another form of cognitive evaluation (perceived innovativeness-responsibility fit), which is essentially a combination of perceived innovativeness and perceived ethical/societal reputation, will indirectly influence brand usage intent. Therefore:

Hypothesis 9: Humanoid FLSRs (vs. self-service machine) have a negative indirect effect on brand usage intent via perceived innovativeness-responsibility fit.

An extended conceptual framework incorporating the hypotheses tested in Studies 1 and 2 is presented in Figure 2.

Figure 2.

Extended conceptual framework for Studies 1 and 2 illustrating hypotheses conceptual framework for Study 1.

Study 1: FLSRs and the Service Provider Experience

Data for all studies were from a consenting representative sample of U.K. adults (aged over 18), collected randomly by the market research firm Qualtrics using an online survey.

Design, Procedure, and Stimuli

To test the hypothesized effects presented in the conceptual framework, an online experiment was conducted, which adopted a 3 (role of FLSRs: augmentation, substitution, control) × 2 (value creation model: asset-builder, service provider) between-subject design. For role of FLSRs, an artificially intelligent humanoid FLSR was presented as either assisting human airline staff during a check-in process (i.e., augmentation) or entirely replacing human staff to complete this process autonomously (i.e., substitution). The control condition stated that only human staff were present. The value creation model was manipulated to control for differences in the salience of frontline service provision to customers; service providers are likely to have greater interaction with customers than asset-builders, since a more human-relational experience is considered more important during service encounters (Lariviere et al. 2017). The asset-builder context was represented by the hypothetical visit to a duty-free shop within the airport transfer terminal, and the service provider scenario was based on the check-in process with the airline.

At the beginning of the survey, participants were asked to imagine the following situation. “Imagine that you are currently having a stopover at an airport while flying to a holiday destination. You are using the airline that you usually or frequently fly with.” Participants were then randomly assigned to one of the experimental conditions. The information provided to the participants is shown in the Supplementary Table 1 (Online Material) for each experimental condition. In addition to the text, an image of a humanoid FLSR was included for augmentation and substitution of human employees, while images of duty-free stores were used to complement the visualization of the value creation model. Images representing the humanoid FLSR presented in all studies, as well as the nonhumanoid service-machine presented in Study 2, are presented in Figure 3.

Figure 3.

Illustrations of humanoid frontline service robot and nonhumanoid self-service machine.

Sample and Measures

Prior to the main study, a preliminary study (more details of the preliminary study are presented in the Supplementary Materials section) was carried out to check the manipulations with 85 randomly chosen participants (M _age = 35.7 years, 42.4% female) from the UK. In addition, standard manipulation checks were carried out in the main study, as well as an attention check, whereby respondents were asked to correctly identify the scenarios presented to them earlier in the survey. Only those who passed the attention check were retained as part of the final sample. This resulted in a random sample of 563 usable responses (all UK) for the main study (M _age = 42.2 years, 52.4% female).

After exposure to the stimulus, the constructs of the conceptual framework were assessed with established scales: (1) Perceived Ethical/Societal Reputation (Stanaland, Lwin, and Murphy 2011), (2) Perceived Innovativeness (Kunz, Schmitt, and Meyer 2011), (3) Service Experience (adapted from Brakus et al. 2009), and (4) Openness-to-Change (World Values Survey 2006). Minor modifications were made to the other items to ensure that they matched the context of the scenarios. Perceived Ethical/Societal Reputation, Perceived Innovativeness, and Service Experience were measured with 7-point scales ranging from strongly disagree (1) to strongly agree (7). Respondents rated their perceived similarity to 10 fictitious personality descriptions on a 6-point scale (1 = not at all like me, 6 = very much like me) for the measurement of the Schwartz Value Circumplex. Manipulation checks were carried out to ensure a successful manipulation of the experimental conditions. Supplementary Table 3 provides an overview of the scales and items.

Results for Study 1: FLSRs and the Service Experience

Table 1 presents a descriptive profile of participants, and Table 2 shows an overview of responses by experimental group. Construct validity and reliability tests were carried out (see Supplementary Tables 5–7 available online) and the composite reliability (CR) measure was found to be greater than 0.7 for all constructs. Further, the average variance extracted (AVE) exceeds 0.5 for each construct, while √AVE exceeds correlations with other constructs and is less than the maximum shared variance; thus, convergent and discriminant validities are established (Hair et al. 2010). The factor means for dependent variables are summarized in Figure 4 by experimental condition. The manipulation checks revealed a significant effect for both manipulated factors: value creation model (F = 13.453; p < .001) and role of FLSR (F= 11.083; p < .001). Preliminary analysis using a two-way factorial MANOVA shows that substitution has a greater positive effect on perceived innovativeness compared to no FLSR involvement (ΔM = 0.248, p < .05), but augmentation of human employees effect in this respect, although positive (ΔM = 0.125), is not statistically significant. Similarly, substitution of human employees has a greater negative effect on perceived ethical/societal reputation compared to no FLSR involvement (ΔM = −0.298, p < .05, confidence interval [CI] [–.536, −.061]), but augmentation of human employees effect (ΔM = −0.228) is not statistically significant compared to no FLSR involvement.

Table 1.

Descriptive Profile of Study 1 Participants.

	n	M	Standard Deviation
Age	563	42.22	11.86
Gender
Male	268
Female	295
Highest level of education
School	03
High school	49
College	87
Bachelor’s degree	244
Master’s degree	147
Doctoral degree	33
Household size	563	3.23	1.24
Household income
£60,00–190,000	325
£90,001–120,000	137
£120,001–150,000	56
More than £150,000	45
Previous experience with AI
None	490
Some	73

Table 2.

No of Responses for Study 1 by Experimental Group.

Factor	Category	N
Value Creation Model	Duty-free (asset-builder)	279
Value Creation Model	Airline (service provider)	284
Role of FLSR	No role (absent)	206
	Human employee augmentation	177
	Human employee substitution	180

Figure 4.

Dependent variables as a function of experimental conditions (Study 1).

Further analyses were conducted based on the ordinary least squares regression method using the Hayes’s PROCESS tool (custom Model 10); bootstrapped (N = 5,000) bias-corrected CIs and heteroscedasticity-consistent standard errors (SEs) were computed in line with standard practice (Hayes 2013; Yoganathan, Osburg, and Akhtar 2019). Categories of the manipulated factors were coded using the indicator method (Hayes and Preacher 2014). Values for openness-to-change were obtained following the procedure of Dobewall and Strack (2014) and Strack and Dobewall (2012) by mean-centering relevant items, which were then used for computing a specific composite score for each respondent. Respondents’ previous experience or interaction with FLSRs was controlled for by including it as a covariate in the model, which resulted in a nonsignificant effect on service experience (β = −.0656; CI [−.2359, .1047]).

The effect of the role of FLSRs on perceived innovativeness is positive and statistically significant for substitution of human employees (β = .1779; CI [.0442, .3116]); but for augmentation of human employees, the effect is weaker and not significant. Hypothesis 1 is therefore supported. Similarly, the effect of the role of FLSRs on perceived ethical/societal reputation is negative and statistically significant for substitution (β = −.1940; CI [−.3301, −.0580]), whereas the effect is not significant for augmentation. Hence, Hypothesis 2 is also supported. However, there were no significant differences observed between the value creation models (asset-builder, service provider) in relation to the effect of FLSR on either perceived innovativeness or perceived ethical/societal reputation. Further, there was no evidence that the effects of the role of FLSRs on perceived innovativeness and perceived ethical/societal reputation are moderated by individuals’ openness-to-change, as the moderation effects are not statistically significant. Thus, Hypotheses 3 and 4 are not supported. Effects on perceived innovativeness and perceived ethical/societal reputation are visualized in Figures 5 and 6.

Figure 5.

Effects of role of frontline service robots on perceived innovativeness.

Figure 6.

Effects of Role of frontline service robots on Perceived Ethical and Societal Reputation.

Results show support for Hypotheses 5a and 5b; the substitution role of FLSRs has a positive indirect effect on service experience via perceived innovativeness and a negative indirect effect via perceived ethical/societal reputation. However, the augmentation role does not have any indirect effects on service experience. Further, substitution’s effect via perceived ethical/societal reputation is significant at high levels of openness-to change in both asset-builder (β = −.0737; CI [−.1511, −.0021]) and service provider (β = −.0841; CI [−.1612, −.0148]) models. On the other hand, substitution’s effect via perceived innovativeness is also significant at high levels of openness-to change but only in the asset-builder model (β = .0704; CI [.0143, .1321]). In both value creation models, indirect effects via perceived innovativeness and perceived ethical/societal reputation are not significant for low levels of openness to change. Indirect effects of substitution are visualized in Figure 7.

Figure 7.

Indirect effects of substitution role of frontline service robots on service experience.

Study 2: FLSRs and Brand Usage Intent

Design, Procedure, and Stimuli

To further explore the effects of substituting human employees with AI, Study 2 considers the substitution role of technology as a function of substituting AI type and service type and builds on Study 1. Accordingly, in Study 2, we manipulated both the substituting AI type and service type, resulting in a 2 (type of AI substitution: FLSR, self-service machine) × 2 (service type: experience, credence) between-subject design. Respondents were randomly assigned to either the experience or credence service condition.

Building on the manipulations specified by Keh and Sun (2018), experience service was represented by a restaurant visit, while credence service was operationalized by the visit to an insurance agency branch (specifically, to buy life insurance). Depending on the condition, respondents were asked to imagine being in one of the described situations: (i) “Imagine that you are having dinner with some friends. You selected a restaurant, which you have not visited before. This restaurant is described in the following” (experience service) or (ii) “Imagine that you have been considering purchasing life insurance. One day, you see an insurance agency and you decide to visit the branch to find out more about it and possibly buy life insurance. The situation is further described in the following” (credence service). Participants were then provided with further information about the situation, which also included a specification of the AI type. Based on the assigned condition, the respondents received one of the texts shown in Supplementary Table 2. The descriptions were complemented with (a) an image of a humanoid or FLSR or self-service machine and (b) an image of a restaurant or insurance agency branch.

Sample and Measures

Overall, a random sample of 400 usable responses was collected (M _age = 34.13 years, 56.8% female; all UK). After exposure to the stimulus, the specified constructs were measured with established scales: (1) Perceived Innovativeness-Responsibility Fit (adapted from Janssen et al. 2014), (2) Preference for Ethical/Responsible Service Provider (Ramasamy and Yeung 2009), and (3) Brand Usage Intent (Yoo and Donthu 2001). Again, minor modifications were conducted, so that the items had a better fit with the presented scenarios. All scales and items are documented in Supplementary Table 3 (available online).

Results for Study 2: FLSRs and Brand Usage Intent

Table 3 presents a descriptive profile of participants, while Table 4 provides an overview of responses by experimental group. The manipulation checks are significant for type of AI substitution (F= 5.69; p < .05) and service type (F= 4.84; p < .05), and hypothesis testing was performed adopting the same procedure as in Study 1. Four variables were controlled for when testing hypotheses by adding them as covariates in the model: experience with FLSRs (β = .0527; CI [.0174, .0879]), experience with a self-service machine (β = −.0314; CI [−.0863, .0235]), visiting restaurants (β = .0300; CI [−.0160, .0760]), and experience with insurance agencies (β = −.0121; CI [−.0457, .0214]).

Table 3.

Descriptive Profile of Study 2 Participants.

	n	M	Standard Deviation
Age	400	34.13	11.49
Gender
Male	173
Female	227
Highest level of education
School	01
High school	55
College	118
Bachelor’s degree	157
Master’s degree	63
Doctoral degree	06
Household size and income	400	2.98	1.32
Less than £30,000	134
£30,00–160, 000	160
£60,001–90,000	71
£90,001–120,000	18
£120,001–150,000	06
More than £150,000	11
Controls (7-point scales)
Used a service robot	400	3.36	2.08
Used a self-service machine	400	6.19	1.17
Frequently visit restaurants	400	5.16	1.48
Consulted insurance agencies	400	4.14	1.93

Table 4.

No of Responses for Study 2 by Experimental Group.

Factor	Category	n
Type of AI substitution	Self-service machine	201
Type of AI substitution	Humanoid frontline service robot	199
Service type	Credence (insurance)	200
Service type	Experience (restaurant)	200

Humanoid FLSRs (vs. self-service machines) have a significant negative effect on perceived innovativeness-responsibility fit (β = −.3634; CI [−.5471, −.1796]), which supports Hypothesis 6. However, there is no significant difference in this effect between credence and experience service types; hence, Hypothesis 7 is not supported.

The negative effect of humanoid FLSRs (vs. self-service machine) on perceived innovativeness-responsibility fit, for credence as well as experience services, is strengthened when an individual’s preference for ethical/responsible service providers is high (β = .2368; CI [.0741, .3995]). Therefore, Hypothesis 8 is supported. Notably (see Figure 8), perceived innovativeness-responsibility fit in credence services is low when an individual’s preference for ethical/responsible service providers is high (compared to average). In contrast, perceived innovativeness-responsibility fit in experience services is high when an individual’s preference for ethical/responsible service providers is high (compared to average).

Figure 8.

Effects on perceived innovativeness-responsibility fit by artificial intelligence type and preference for ethical/responsible business.

Humanoid FLSRs (vs. self-service machines) do not have a direct effect on brand usage intent in either credence or experience services. Nevertheless, statistically significant and positive indirect effects were observed in credence (β = .1901; CI [.0843, .3051]) and experience service types (β = .1408; CI [.0401, .2403]), which supports Hypothesis 9. However, the indirect effects do not differ significantly based on either the service type or an individual’s preference for ethical/responsible service providers.

General Discussion

There is little doubt that the introduction of FLSRs will have a profound effect on the service domain. In this article, we use the results of two main studies to show that introducing FLSRs has the following main effects: (1) Augmenting or substituting human employees with FLSRs has positive and negative consequences irrespective of value creation model, AI type, and service type; (2) FLSRs make the customer service interaction feel more innovative; (3) If human employees are replaced by FLSRs (e.g., for cost or other benefits), they damage the ethical/societal reputation of the service provider in terms of both service experience and brand usage intent; (4) However, personal customer characteristics (openness-to-change and preference for ethical/responsible service providers) determine the specificity and extent of these effects. While some individuals value innovativeness more, others appreciate the fact that a service provider is responsible toward employees and society. Our findings alert practitioners and researchers to the need to consider seriously (from a customer perspective) how the use of FLSRs influences perceived ethical/societal reputation, particularly if they are replacing/substituting human employees. Our results have theoretical as well as managerial implications that will help to successfully launch FLSRs.

Theoretical Implications

This article addresses gaps in the extant literature by answering calls for more research focusing on FLSRs and brands (Mende et al. 2019; Wirtz et al. 2018). No previous empirical research has explored the specific roles that FLSRs play in influencing service experience or brand usage intent. By operationalizing key components of Larivière et al.’s (2017) conceptual Service 2.0 model, we empirically validate and extend their work by developing and testing a holistic framework for understanding customer perceptions of FLSRs. Further, we illustrate the importance of considering both positive and negative cognitive evaluations of customers in relation to FLSRs (Čaić, Odekerken-Schröder, and Mahr 2018). As the role of FLSRs increases to replacing humans, customers’ perception of innovativeness also rises, which complements previous conceptual studies that have highlighted the positive potential of FLSRs (e.g., Wirtz et al. 2018).

Our findings also highlight the negative effect of FLSRs in terms of the ethical/societal reputation of service providers, which has not been widely empirically tested, but emphasized in conceptual papers owing to the nascent nature of the subject. For example, studies have highlighted the potential that AI has to replace human workers and create job displacement or losses (Huang and Rust 2018); robotic autonomous driverless cars (e.g., Coca-Vila 2018); robots providing the elderly with care (e.g., Čaić, Odekerken-Schröder, and Mahr 2018; Stahl and Coeckelbergh 2016); and more general concerns with FLSRs (e.g., Wirtz et al. 2018).

Customers appear to expect that employees will continue to play an essential role in delivering and defining frontline service encounters in a variety of service settings (De Keyser et al. 2019). Our results suggest that FLSRs were perceived to be more innovative in the asset-builder context of a duty-free store than the service provider context of an airline check-in. In an asset-builder context, FLSRs may add to the evaluation of the overall experience of purchasing a physical product.

By situating our study in the new and existing context of frontline service encounters and exploring mediating effects, we contribute to the general literature on service experience and brand usage intent (e.g., Andreini et al. 2018; Hollebeek, Glynn, and Brodie 2014; Japutra and Molinillo 2019). A positive service experience and increase in brand usage intent drives brand engagement and builds brand equity (Hepola, Karjaluoto, and Hintikka 2017). The mediation effects were significant when FLSRs completely replace humans, but not when they augment human service provision in a frontline service encounter. It appears that customers do not perceive that it is extraordinary to have an FLSR augmenting and supporting employees in frontline service encounters or perhaps they do not attach much importance to it. With regard to the mediating effect of perceived ethical/societal reputation, there is a clear difference between substitution and augmentation of human employees. Participants were concerned about the ethical and societal consequences of replacing employees with FLSRs, particularly in a substitution context.

The effect of humanoid FLSRs (vs. self-service machines) on perceived innovativeness-responsibility fit did not significantly differ between credence and experience services, which has not been explored previously. This is linked to arguably the most important finding of our research, which is the overall negative influence that FLSRs have on the service experience when customers are prompted to consider ethical/responsible aspects. By identifying this overall effect, we have answered calls to test net effects (e.g., Čaić, Odekerken-Schröder, and Mahr 2018). In most circumstances, the balance between the positive influence of perceived innovativeness and negative effect on perceived ethical/societal reputation and innovativeness-responsibility fit means that introducing FLSRs reduces a customer’s overall brand experience and brand usage intent. Hence, regardless of the value creation model, FLSRs taking over the roles of employees in frontline service encounters and customer-brand interactions is perceived negatively. These results provide empirical support for Huang and Rust (2018) and others who express concerns regarding FLSRs replacing employees.

We also considered how individual customer characteristics and values influence their perceptions of the role of FLSRs. Previous discussions have focused on customers in general; however, some individuals may be more open to change (innovative), while others may continue to expect traditional personal interactions with humans during a frontline service encounter. Consistent with theory, participants who were open to change appreciate the positive aspects of innovation, while those who were less receptive to change expressed concern about the negative implications (Claudy, Garcia, and O’Driscoll 2015; Wang, Dou, and Zhou 2008). Furthermore, these results were conditional on the value creation model. Positive effects of perceived innovativeness are found in the asset-builder context and when openness-to-change is high. In contrast, the negative effects of perceived ethical/societal reputation are present in both the asset-builder and service provider contexts when openness-to-change is low.

For customers who have a high preference for ethical/responsible service providers, perceived fit between the innovative and responsible aspects of humanoid FLSRs substituting/replacing human workers is low in credence services and high in experience services. This is understandable given that credence services (e.g., insurance) are harder to evaluate, uncertain, and riskier (from a customer perspective), and therefore, customers are likely to take a harsher or more skeptical attitude in evaluating FLSR involvement in such cases. Our findings in this respect are also consistent with those from brand congruency literature (Arbouw, Ballantine, and Ozanne 2018; de Jong and van der Meer 2017). Overall, if the fit between the innovative and ethical aspects of humanoid FLSRs substituting employees is established in the view of customers, this will lead to an increase in brand usage intent regardless of service type or customer preference.

Managerial Implications

Our findings indicate that it is essential for service providers to understand how customers cognitively evaluate FLSRs and the important influence of customer characteristics, as these will have an impact on frontline service encounters. FLSRs substituting or replacing employees is perceived as a more innovative move, but FLSRs augmenting frontline service employees appear better for the ethical/societal reputation of a service provider. Also, humanoid FLSRs are perceived to be more innovative than nonhumanoid self-service machines that perform the same function. However, relative to self-service machines, humanoid FLSRs are perceived to have an incongruent innovativeness-responsibility fit that results in an overall negative impact on customer intent to use a brand. Further, the introduction of FLSRs is likely to be more successful in an asset-builder rather than service provider context. Specific recommendations that will help service providers take advantage of the benefits and reduce the risks associated with introducing FLSRs are presented in Table 5.

Table 5.

Managerial Guidelines for Frontline Service Robots (FLSRs) Highlighting Perceived Benefits and Risks.

Scenario	Benefits	Risks
Value creation model Asset-builder Service provider	Attempt to capitalize on positive customer perceptions of FLSRs and focus marketing communication efforts on innovators/early adopters.Take advantage of the overall advantages that FLSRs provide to service providers such as individualized networked CRM systems, ability to track and analyzes customer behavior, collect and share customer data, and provide highly personalized services, rather than customer-centric benefits.	For both asset-builders and service providers the negative impact on the ethical and societal reputation may outweigh the innovativeness of FLSRs replacing human workers. Asset-builders and service providers may benefit by continuing to employ traditional human staff, introducing FLSRs gradually.
Role of FLSR Augmentation Substitution	Augmentation of employees may provide a smooth transition for service organizations seeking to introduce FLSRs, as it is not perceived to be as damaging to the ethical and societal reputation of a brand than substitution. Ensure customers are aware of FLSRs’ augmentation role. When possible, focus on moving employees replaced by FLSRs to other roles where robots are unable to perform effectively and ensure customers are made aware of this to reduce reputational damage.	The risk of augmentation is that the innovativeness aspect may not be fully appreciated by customers. In such scenarios, FLSRs serve little more than a decorative purpose, the novelty of which can quickly dissipate; thus, reducing the return on investment. Service organizations may counter this by giving FLSRs a specific role; for example, at the Smithsonian museum, in conjunction with FLSRs, human visitor information specialists provide valuable and inspiring information that can make customer experiences more memorable and exciting.Be cautious when introducing FLSRs as the negative effect that robots have on ethical and societal reputation substantially outweighs the perceived positive benefits associated with innovativeness.
Service type Credence Experience	May prove attractive to customers due to their perceived innovativeness. Better to use FLSRs in combination with human employees given the uncertainties surrounding credence services. May have a novelty effect and thus serve to attract customers and enhance engagement.	As credence services are typically riskier than experience services, use FLSRs to reduce customers’ perceptions of risks associated with this service type. For example, have FLSRs use AI to provide a customized experience based on access to customers’ biometrics and share immediate cost and performance data, often lacking in credence service encounters (Mitra, Reiss, and Capella, 1999).Reduce risk by assigning FLSRs to augmentation of services offered by human staff and avoid creating the perception that FLSRs are an attempt to avoid hiring human employees.
Type of AI substitution Humanoid FLSR Nonhumanoid self-service machine	Greater positive impact on customer engagement through increasing innovativeness perceptions. Less damaging in terms of achieving innovativeness-responsibility fit. Introduce nonhumanoid self-service machines prior to humanoid FLSRs in order to reduce perceived incongruences.	Replacing employees with humanoid FLSRs has a negative effect on perceived innovativeness-responsibility fit. However, if a balance between innovativeness and responsibility aspects of FLSRs can be achieved (e.g., via redeployment of human employees for higher order tasks), then the ultimate effect on customer engagement can be positive.The positive impact of innovativeness may not be realized. Better used as an augmentation device to frontline services provided by human staff.

Service providers might well find the prospect of reduced costs, increased efficiencies, and the provision of an alternative customer experience associated with introducing FLSRs alluring. Nonetheless, from a customer perspective, we find that the introduction of FLSRs will create a double-edged phenomenon. On the one hand, it is perceived as innovative (positive), but on the other, it is ethically questionable creating a poor innovativeness-responsibility fit. When considered in parallel, under most circumstances, negative influences outweigh the positive perceptions and lead to a decrease in the overall service experience and reduce brand usage intent. Consequently, plans or strategies involving the introduction of FLSRs should be carefully considered. While investment in robotics for assembly-line production in the 1980s was a broadly advantageous move, the same cannot quite be said of the new wave of FLSRs now entering frontline service. As such, their adoption is set to remain contentious.

Our results are dependent on the extent to which individual customers are conservative or open to change and their preferences for ethical/responsible service providers. This may depend on cultural and country differences. For multinational corporations, introducing FLSRs in countries where customers have higher levels of innovativeness (e.g., natives of Finland are seemingly in thrall to technology, whereas Germans are less positively disposed) in advance of countries where levels of customer innovativeness are lower may reap rewards (e.g., Bögel et al. 2018; Dobewall and Strack 2014). Service providers must realize that traditional models of innovation diffusion may need adapting before implementation in a dynamic, disruptive world of robot-based service encounters. If jobs lost to FLSRs are not as significant as doomsayers warn and as customers become familiar with FLSRs, damage to ethical/societal reputation may be reduced. However, education and marketing programs aimed at changing negative perceptions may still be warranted. In general, service managers should be cautioned against purely FLSR-driven service scenarios.

In view of the above, prior to the widespread deployment of FLSRs, we suggest that both managers and policy makers should give careful consideration to the following seven principles:

To ensure that the safety and well-being of customers are not endangered, it is essential that ethical principles governing the implementation of FLSRs are developed and universally adopted.

Through educational initiatives, service providers should collaborate to prepare customers and frontline employees for the imminent arrival of interactive FLSRs.

Given the extensive nature of customer data that can be gathered through routine human-robot interactions, which is then stored, mined, and utilized by organizations, protecting the privacy such data must be an unshakeable tenet of FLSRs.

Since robotic AI systems often excel by developing new ways of seeing and thinking that are impenetrable to human observers, service providers must strive for absolute transparency and integrity in their operation.

Customers should be kept informed of the capabilities of such systems, and mechanisms of overseeing such capabilities and necessary control should also be made available to them.

Pressures to sanction and promote the use of robotics and AI as means of increasing productivity, while simultaneously employing less labor, create a need for policy makers to remain attuned to the need to reskill and assist employees whose jobs are threatened by FLSRs. The introduction of a “robot tax” to fund and support these efforts may also be necessary.

Policy makers should particularly focus on the utilization of FLSRs in credence services, where the customer may not be aware of potential substitution by FLSRs, especially if service providers deliberately avoid being transparent about the use of AI technology to protect their ethical/societal reputation.

Limitations and Areas for Further Research

Our findings demonstrate the importance of considering positive and negative influences on AI adoption together to avoid a one-sided perspective, which may over- or underestimate customer attitudes and perceptions. While the present research shows that perceived innovativeness and ethical/societal responsibility are some of these factors, future research needs to explore a range of other drivers and barriers related to AI adoption. This may include both technology-related and customer-related characteristics.

Additional research should be conducted when FLSRs are more commonly used in frontline service encounters. Such enquiry could explore the consequences of a positive service experience and increased brand usage, directly on, for example: customer engagement (see Hollebeek, Srivastava, and Chen 2016) or the roles that FLSRs could play in sharing experiences and value co-creation (Chen et al. 2018). Further, preferences for specific private versus national brands (Liu et al. 2018) or service provider versus manufacturer brands could be explored. Our scenario involved an online experiment using images of robots in various frontline service scenarios. In the future, field experiments based on the introduction of real FLSRs would support our scenario-based research and enable more tests of their positive or negative implications on the service experience, as well as other constructs of interest. For example, if and how customers’ perceptions of FLSRs change during a crisis such as the recent coronavirus pandemic when face-to-face encounters with human employees may be risky or not possible. Perceptions of artificial faces could be explored as deviation from humanness in humanoid FLSRs may have dramatic consequences (van Doorn et al. 2017; Wirtz et al. 2018). The relationships between the feeling of visceral discomfort explained by uncanny valley theory (Mori, McDorman, and Kageki 2012) and an apparently incongruent innovativeness-responsibility fit associated with humanoid FLSRs requires exploration, as they may have contrasting or cumulative negative effects on frontline service encounters. Our article did not directly focus on these factors; however, additional research in this area would provide valuable insights for FLSR designers.

Our research focused on perceived innovativeness, perceived ethical/societal reputation, and innovativeness-responsibility fit, which were established as mediators, forming the basis of customer cognitive evaluations. It would be useful to understand how long-lasting these effects are by conducting longitudinal research, which monitors changes in customer perceptions over time. Relatedly, the effects may also depend on the cultural context, namely, technology-affine cultures could be more open to the introduction of FLSRs. Our research was undertaken in the UK, but our framework could be extended to other countries and cultures. In addition, more research focusing on marketing communications is needed to identify how the implementation of FLSRs should best be communicated to customers. Our experimental study shows that as perceptions of substitution are negative, there may be a need for better education or promotion aimed at informing customers of the benefits of FLSRs or the redeployment of staff. However, as the best methods for disclosing information are yet to be determined, how would customers perceive FLSRs being the main communicator? Although we explored FLSRs in both credence and experience contexts, the extent to which perceptions of the augmentation and substitution of human employees are industry dependent should be explored in more detail. For example, the purpose of travel (holiday vs. business) may influence such perceptions and customers’ willingness-to-pay for the service may vary accordingly.

This article operationalizes Larivière et al.’s (2017) model by focusing on how customer perceptions of the service experience are influenced in asset-builder and service provider value creation models with FLSRs augmenting or substituting the role of service employees. Other parts of the model including network-orchestration and technology-creator value creation models, network facilitation, and transformation roles for customers and employees also warrant further research and validation through empirical studies. Additional categorizations such as Huang and Rust’s (2017, 2018) typology of a technology-driven service, and intelligences required for service tasks model, as well as van Doorn et al.’s (2017) and Wirtz et al.’s (2018) task type and service recipient model also warrant further empirical attention. The focus of our study was customer-centric and at the microlevel, rather than meso-, macro-, and metalevel contexts (Alexander, Jaakola, and Hollebeek 2018). Additional empirical research on the meso-, macro-, and metalevels of FLSRs across a network of dyads and actors in a servicescape would be useful.

As the development of scales that measure effectiveness and perceptions relating to FLSRs and branding are in their infancy, perhaps future research may focus on developing new and more appropriate scales. For example, there are opportunities for developing new scales that directly measure dehumanization or privacy concerns as FLSRs increasingly drive frontline service encounters.

In conclusion, since a wave of automation will undoubtedly transform service encounters and experiences, we must strive to understand the challenges and take advantage of the opportunities they provide. As service scholars, we must continue to explore the role of FLSRs in service types, the tasks they perform, and the factors that are crucial to value creation. Hollebeek, Srivastava, and Chen’s (2016) integrative S-D logic informed framework may provide additional insights for understanding co-creation, customer resource integration and customer learning, as we embrace the inevitable future in which AI-driven interactive and dynamic servicescapes become commonplace.

Supplemental Material

Supplemental Material, New_insights_into_how_robots_will_change_frontline_services - Replaced by a Robot: Service Implications in the Age of the Machine

Supplemental Material, New_insights_into_how_robots_will_change_frontline_services for Replaced by a Robot: Service Implications in the Age of the Machine by Fraser McLeay, Victoria Sophie Osburg, Vignesh Yoganathan and Anthony Patterson in Journal of Service Research

Supplemental Material

Supplemental Material, RESUBMISSION_SUPPLEMENTARY_ONLINE_MATERIAL_2020_final_Sept_14 - Replaced by a Robot: Service Implications in the Age of the Machine

Supplemental Material, RESUBMISSION_SUPPLEMENTARY_ONLINE_MATERIAL_2020_final_Sept_14 for Replaced by a Robot: Service Implications in the Age of the Machine by Fraser McLeay, Victoria Sophie Osburg, Vignesh Yoganathan and Anthony Patterson in Journal of Service Research

Footnotes

Authors’ Note

Vignesh Yoganathan is not affiliated with Bradford University School of Management, United Kingdom now.

Acknowledgments

We would like to thank Hossein Olya, 3 anonymous reviewers and the editors of this special issue for their helpful comments on earlier drafts of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the Sheffield University Management School for internal funding to collect data for this study.

ORCID iDs

Fraser McLeay

Vignesh Yoganathan

Supplemental Material

Supplemental material for this article is available online.

Note

References

Abela

Andrew V.

Murphy

Patrick E.

(2008), “Marketing with Integrity: Ethics and the Service-dominant Logic for Marketing,” Journal of the Academy of Marketing Science, 36 (1), 39–53.

Alexander

Matthew J.

Jaakkola

Elina

Hollebeek

Linda D.

(2018), “Zooming Out: Actor Engagement beyond the Dyadic,” Journal of Service Management, 9 (3), 333–351.

Andreassen

Tor W.

Kristensson

Per

Lervik-Olsen

Line

Parasuraman

McColl-Kennedy

Janet R.

Edvardsson

Colurcio

Maria

(2016), “Linking Service Design to value Creation and Service Research,” Journal of Service Management, 27 (1), 21–29.

Andreini

Daniela

Pedeliento

Giuseppe

Zarantonello

Lia

Solerio

Chiara

(2018), “A Renaissance of Brand Experience: Advancing the Concept through a Multi-perspective Analysis,” Journal of Business Research, 91 (C), 123–133.

Antons

David

Breidbach

Christoph F.

(2018), “Big Data, Big Insight? Advancing Service Innovation and Design with Machine Learning,” Journal of Service Research, 21 (1), 17–39.

Arbouw

Paula

Ballantine

Paul W.

Ozanne

Lucie K.

(2018), “Sustainable Brand Image: An Examination of Ad-brand Incongruence,” Marketing Intelligence and Planning, 37 (5), 513–526.

Barrat

James

(2013), Our Final Invention: Artificial Intelligence and the End of the Human Era. New York: Thomas Dunne Books.

Bögel

Paula

Oltra

Christian

Sala

Roser

Lores

Monica

Upham

Paul

Dütschke

Elisbeth

Schneider

Ute

Wiemann

Paul

(2018), “The Role of Attitudes in Technology Acceptance Management: Reflections on the Case of Hydrogen Fuel Cells in Europe,” Journal of Cleaner Production, 188, 125–135.

Brakus

J. Josko

Schmitt

Bernd H.

Zarantonello

Lia

, (2009), “Brand Experience: What Is It? How Is It Measured? Does It Affect Loyalty?” Journal of Marketing, 73 (3), 52–68.

10.

Čaić

Martina

Odekerken-Schröder

Gaby

Mahr

Dominik

(2018), “Service Robots: Value co-creation and Co-destruction in Elderly Care Networks,” Journal of Service Management, 29 (2), 178–205.

11.

Carter

Robert E.

Curry

David J.

(2013), “Perceptions versus Performance when Managing Extensions: New Evidence about the Role of Fit between a Parent Brand and an Extension,” Journal of the Academy of Marketing Science, 41 (2), 253–269.

12.

Chen

Tom

Drennan

Judy

Andrews

Lynda

Hollebeek

Linda D.

(2018), “User Experience Sharing: Understanding Customer Initiation of Value co--creation in Online Communities,” European Journal of Marketing, 52 (5/6), 1154–1184.

13.

Chocarro

Raquel

Cortinas

Monica

Villanueva

Maria Luisa

(2018), “Different Channels for Different Services: Information Sources for Services with Search, Experience and Credence Attributes,” The Service Industries Journal, 9, 1–24. https://doi.org/10.1080/02642069.2018.1508457

14.

Claudy

Marius C.

Garcia

Rosanna

O’Driscoll

Aidan

(2015), “Consumer Resistance to Innovation—A Behavioural Reasoning Perspective,” Journal of the Academy of Marketing Science, 43 (4), 528–544.

15.

Coca-Vila

Ivó

(2018), “Self-driving Cars in Dilemmatic Situations: An Approach Based on the Theory of Justification in Criminal Law,” Criminal Law and Philosophy, 12 (1), 59–82.

16.

Davenport

Thomas

Guha

Abhijit

Grewal

Dhruv

Bressgott

Timna

(2020), “How Artificial Intelligence Will Change the Future of Marketing,” Journal of the Academy of Marketing Science, 48 (1), 24–42.

17.

De Groot

Judith I. M.

Steg

Linda

(2009), “Mean or Green: Which Values can Promote Stable Pro-environmental Behavior?” Conservation Letters, 2 (2), 61–66.

18.

Decker

Michael

Fischer

Martin

Ott

Ingrid

(2017), “Service Robotics and Human Labor: A First Technology Assessment of Substitution and Cooperation,” Robotics and Autonomous Systems, 87, 348–354.

19.

de Jong

Menno D. T.

van der Meer

Mark

(2017), “How Does It Fit? Exploring the Congruence Between Organizations and Their Corporate Social Responsibility (CSR) Activities.” Journal of Business Ethics, 143 (1), 71–83.

20.

De Keyser

Arne

Köcher

Sarah

Alkire

Linda

Verbeeck

Cedric

Kandampully

Jay

(2019), “Frontline Service Technology Infusion: Conceptual Archetypes and Future Research Directions,” Journal of Service Management, 30 (1), 156–183.

21.

Delfanti

Alessandro

Arvidsson

Adam

(2019), Introduction to Digital Media. Hoboken, NJ: John Wiley.

22.

Ding

Cherng G.

Tseng

Timmy H.

(2015), “On the Relationships among Brand Experience, Hedonic Emotions, and Brand Equity,” European Journal of Marketing, 49 (7/8), 994–1195.

23.

Dobewall

Henrik

Strack

Micha

(2014), “Relationship of Inglehart’s and Schwartz’s Value Dimensions Revisited,” International Journal of Psychology, 49 (4), 240–248.

24.

Eisingerich

Andreas B.

Rubera

Gaia

(2010), “Drivers of Brand Commitments a Cross National Investigation,” Journal of International Marketing, 18 (2), 64–79.

25.

Ford

Martin

(2015), The Rise of the Robots. London, UK: Oneworld Publications.

26.

Fukukawa

Kyoko

Balmer

John M. T.

Gray

Edmund R.

(2007). “Mapping the Interface between Corporate Identity, Ethics and Corporate Social Responsibility,” Journal of Business Ethics, 76 (1), 1–5.

27.

Goduscheit

Rene C.

Faullant

Rita

(2018), “Paths toward Radical Service Innovation in Manufacturing Companies: A Service-dominant Logic Perspective,” Journal of Product Innovation Management, 35 (5), 701–719.

28.

Hair

Joseph F.

Black

William C.

Babin

Barry J.

Anderson

Rolph E.

(2010), Multivariate Data Analysis: A Global Perspective (7th ed.). Upper Saddle River, NJ: Pearson.

29.

Harris

Karen

Kimson

Austin

Schwedel

Andrew

(2018), “Why the Automation Boom Could Be Followed by a Bust,” Harvard Business Review, March 13, (accessed February 24, 2020), [available at https://hbr.org/2018/03/why-the-automation-boom-could-be-followed-by-a-bust.]

30.

Hayes

Andrew F.

Preacher

Kristopher J.

(2014), “Statistical Mediation Analysis with a Multi-categorical Independent Variable,” British Journal of Mathematical and Statistical Psychology, 67 (3), 451–470.

31.

Hayes

Andrew F.

(2013), Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-based Approach. New York: Guilford Press.

32.

Hepola

Janne

Karjaluoto

Heikki

Hintikka

Anni

(2017), “The Effect of Sensory Brand Experience and Involvement on Brand Equity Directly and Indirectly Thought Customer Brand Engagement,” Journal of Product and Brand Management, 26 (4), 282–293.

33.

Hoffmann

Stefan

Soyez

Katja

(2010), “A Cognitive Model to Predict Domain-specific Consumer Innovativeness,” Journal of Business Research, 63 (7), 778–785.

34.

Hollebeek

Linda D.

(2018), “Individual-level Cultural Consumer Engagement Styles. Conceptualization, Propositions and Implications,” International Marketing Review, 35 (1), 42–71.

35.

Hollebeek

Linda D.

Andreassen

Tor W.

(2018), “The S-D Logic-informed “Hamburger” Model of Service Innovation and its Implications for Engagement and Value,” Journal of Services Marketing, 32 (1), 1–7.

36.

Hollebeek

Linda D.

Andreassen

Tor W.

Sprott

David E.

(2018), “Rise of the Machines? Customer Engagement through Automated Service Interactions,” Journal of Service Research, Special Issue Call for Papers.

37.

Hollebeek

Linda D.

Jaakkola

Elina

Alexander

Matthew

(2018), “Beyond the Dyadic: Customer Engagement in Increasingly Networked Environments,” Journal of Service Management, 29 (3), 330–332.

38.

Hollebeek

Linda D.

Srivastava

Rajendra K.

Chen

Tom

(2016), “S-D logic–informed Customer Engagement: Integrative Framework, Revised Fundamental Propositions, and Application to CRM,” Journal of the Academy of Marketing Science, 47 (1), 161–185.

39.

Hollebeek

Linda D.

Glynn

Mark S.

Brodie

Roderick J.

(2014), “Consumer Brand Engagement in Social Media: Conceptualization, Scale Development and Validation,” Journal of Interactive Marketing, 28 (2), 149–165.

40.

Huang

Ming-Hui

Rust

Roland T.

(2018), “Artificial Intelligence in Service,” Journal of Service Research, 21 (2), 155–172.

41.

Huang

Ming-Hui

Rust

Roland T.

(2017), “Technology-driven Service Strategy,” Journal of the Academy of Marketing Science, 45 (6), 906–924.

42.

Hurley

Robert F.

Hult

G. Tomas M.

(1998), “Innovation, Market Orientation and Organizational Learning: An integration and Empirical Examination,” Journal of Marketing, 62 (3), 42–54.

43.

Hurrell

Scott A.

Scholarios

Dora

(2014), “The People Make the Brand: Reducing Social Skill Gaps through Person-brand Fit and Human Resource Management Practices,” Journal of Service Research, 17 (1) 54–67.

44.

Janssen

Catherine

Vanhamme

Joelle

Lindgreen

Adam

Cecile

Lefebvre, C.

(2014), “The Catch-22 of Responsible Luxury: Effects of Luxury Product Characteristics on Consumers’ Perception of Fit with Corporate Social Responsibility,” Journal of Business Ethics, 119 (1), 45–57.

45.

Japutra

Arnold

Molinillo

Sebastian

(2019), “Responsible and Active Brand Personality: On the Relationships with Brand Experience and Key Relationship Constructs,” Journal of Business Research, 99, 464–471.

46.

Jin

Naehyun

Line

Nathaniel D.

Merkebu

Jerusalem

(2016), “Examining the Impact of Consumer Innovativeness and Innovative Restaurant Image in Upscale Restaurants,” Cornell Hospitality Quarterly, 57 (3), 268–281.

47.

Jörling

Moritz

Böhm

Robert

Paluch

Stefanie

(2019), “Service Robots: Drivers of Perceived Responsibility for Service Outcomes,” Journal of Service Research, 22 (4), 404–420.

48.

Joyce

. (2018), “Android Tellers in Branches: Cool or Creepy?” The Financial Brand News Letter, (accessed November 1, 2018), [available at https://thefinancialbrand.com/70651/android-human-robot-digital-banking-experience].

49.

Keh

Hean T.

Sun

Jin

(2018), “The Differential Effects of Online Peer Review and Expert Review on Service Evaluations: The Roles of Confidence and Information Convergence,” Journal of Service Research, 21 (4), 478–489.

50.

Kim

Seo E.

Schmitt

Bernd H.

Thalmann

Nadia M.

(2019), “Eliza in the Uncanny Valley: Anthropomorphizing Consumer Robots Increases Their Perceived Warmth but Decreases Liking,” Marketing Letters, 30 (1), 1–12.

51.

Kim

Jaesu

Kim

Kyung H.

Garrett

Tony C.

Jung

Heonsoo

(2015), “The Contributions of Firm Innovativeness to Customer Value in Purchasing Behaviour,” Journal of Product Innovation Management, 32 (2), 201–213.

52.

Kunz

Werner

Schmitt

Bernd

Meyer

Anton

(2011), “How Does Perceived Firm Innovativeness Affect the Consumer?” Journal of Business Research, 64 (8), 816–822.

53.

Larivière

Bart

Bowen

David

Andreassen

Tor W.

Kunz

Werner

Sirianni

Nancy J.

Voss

Chris

Wuenderlich

Nancy C.

De Keyser

Arne

(2017), “Service Encounter 2.0: An Investigation Into the Roles of Technology, Employees and Customers,” Journal of Business Research, 79, 238–246.

54.

Lin

Yi H.

(2015), “Innovative Brand Experience’s Influence on Brand Equity and Brand Satisfaction,” Journal of Business Research, 68 (11), 2254–2259.

55.

Liu

Richie L.

Sprott

David E.

Spangenberg

Eric R.

Czeller

Sandor

Voss

Kevin E.

(2018), “Consumer Preference for National vs. Private Brands: The Influence of Brand Engagement and the Self-concept Threat,” Journal of Retailing and Consumer Services, 41, 90–100.

56.

Marr

Bernard

Ward

Matt

(2019), Artificial Intelligence in Practice: How 50 Successful Companies Used AI and Machine Learning to Solve Problems. London, UK: Wiley.

57.

Mende

Martin

Scott

Maura L.

Doorn

Jenny van

Grewal

Dhruv

Shanks

Ilana

(2019), “Service Robots Rising: How Humanoid Robots Influence Service Experiences and Elicit Compensatory Consumer Responses,” Journal of Marketing Research, 56 (4) 535–556.

58.

Mitra

Kaushik

Reiss

Michelle C.

Capella

Louisa M.

(1999), “An Examination of Perceived Risk, Information Search and Behavioral Intentions in Search, Experience and Credence Services,” Journal of Services Marketing, 13 (3), 208–232.

59.

Mori

Masahiro

MacDorman

Karl F.

Kageki

Norri

(2012), “The Uncanny Valley,” Robotics and Automation Magazine, IEEE, 19 (2), 98-100.

60.

Osburg

Victoria-Sophie

Akhtar

Pervaiz

Yoganathan

Vignesh

McLeay

Fraser

(2019), “The Influence of Contrasting Values on Consumer Receptiveness to Ethical Information and Ethical Choices,” Journal of Business Research, 104, 366–379.

61.

Ostrom

Amy

Parasuraman

Bowen

David E.

Patricio

Lia

Voss

Christopher A.

(2015), “Service Research Priorities in a Rapidly Changing Context,” Journal of Service Research, 18 (2), 127–159.

62.

Ostrom

Amy

Iacobucci

Dawn

(1995), “Consumer Trade Offs and the Evaluation of Services,” Journal of Marketing, 59 (1), 17–28.

63.

Patricio

Lia

Gustafsson

Anders

Fisk

Raymond

(2018), “Upframing Service Design and Innovation for Research Impact,” Journal of Service Research, 21(1), 3–16.

64.

Pepper

Martin

Jackson

Tim

Uzzell

David

(2009), “An Examination of the Values that Motivate Socially Conscious and Frugal Consumer Behaviours,” International Journal of Consumer Studies, 33 (2), 126–136.

65.

Qureshi

Mohammed

Syed

Rumaiya S.

(2014), “The Impact of Robotics on Employment and Motivation of Employees in the Service Sector with Special Reference to Health Care,” Safety and Health at Work, 5 (4), 198–202.

66.

Rafaeli

Anat

Altman

Daniel

Gremler

Dwayne D.

Huang

Ming-Hui

Grewal

Dhruv

Iyer

Bala

Parasuraman

de Ruyter

(2017), “The Future of Frontline Research: Invited Commentaries,” Journal of Service Research, 20 (1), 91–99.

67.

Ramasamy

Bala

Yeung

Mathew

(2009), “Chinese Consumers’ Perception of Corporate Social Responsibility (CSR),” Journal of Business Ethics, 88 (1), 119–132.

68.

Schwartz

Shalom H.

(2012), “An Overview of the Schwartz Theory of Basic Values,” Online Readings in Psychology and Culture, 2 (1), https://doi.org/10.9707/2307-0919.1116.

69.

Schwartz

Shalom H.

Boehmke

Klaus

(2004), “Evaluating the Structure of Human Values with Confirmatory Factor Analysis,” Journal of Research in Personality, 38 (3), 230–255.

70.

Schwartz

Shalom H.

(2003), “Universals in the Content and Structure of Values: Theoretical Advances and Empirical Tests in 20 Countries,” Advances in Experimental Social Psychology, 25, 1–65.

71.

Sjödin

Henrik

Törn

Fredrik

(2006). “When Communication Challenges Band Associations: A Framework for Understanding Consumer Responses to Brand Image Inconsistency,” Journal of Consumer Behaviour, 5 (1), 32–42.

72.

Stahl

Bernd C.

Coeckelbergh

Mark

(2016), “Ethics of Healthcare Robotics: Towards Responsible Research and Innovation,” Robotics and Autonomous Systems, 86, 152–161.

73.

Stanaland

Andrea J. S.

Lwin

May O.

Murphy

Patrick E.

(2011), “Consumer Perceptions of the Antecedents and Consequences of Corporate Social Responsibility,” Journal of Business Ethics 102 (1), 47–55.

74.

Strack

Micha

Dobewall

Henrik

(2012), “The Value Structure in Socioeconomically Less Developed European Countries Still Remains an Ellipse,” Europe’s Journal of Psychology, 8 (4), 587–602.

75.

Vaidhyanathan

Siva

(2018), Anti-social Media: How Facebook Disconnects and Undermines Democracy. Oxford, UK: Oxford University Press.

76.

van der Westhuizen

Liezl-Marie

(2018), “Brand-loyalty: Exploring Self-brand Connection and Brand Experience,” Journal of Product & Brand Management, 27 (2), 172–184.

77.

van Doorn

Jenny

Mende

Martin

Noble

Stephanie M.

Hulland

John

Ostrom

Amy L.

Grewal

Dhruv

Andrew Petersen

(2017), “Domo Arigato Mr. Roboto: Emergence of Automated Social Presence in Organizational Frontline and Customers’ Service Experiences,” Journal of Service Research, 20 (1), 43–58.

78.

Voorhees

Clay M.

Fombelle

Paul W.

Gregoire

Yany

Bone

Sterling

Gustafsson

Anders

Sousa

Rui

Walkowiak

Travis

(2017), “Service encounters, experiences and the customer journey: Defining the field and a call to expand our lens,” Journal of Business Research, 79 (C), 269–280.

79.

Wang

Guangping

Dou

Wenyu

Zhou

Nan

(2008), “Consumption Attitudes and Adoption of New Consumer Products: A Contingency Approach,” European Journal of Marketing, 42 (1/2), 238–254.

80.

Wirtz

Jochen

(2019), “Organizational Ambidexterity: Cost-effective Service Excellence, Service Robots and Artificial Intelligence,” Organizational Dynamics.

81.

Wirtz

Jochen

Patterson

Paul G.

Kunz

Werner H.

Gruber

Thorsten

H. Lu

Vin

Stefanie

Paluch

Martins

Antje

(2018), “Brave New Word: Service Robots in the Frontline,” Journal of Service Management, 29 (5), 907–931.

82.

World Values Survey (2006), “World Values Survey,” (accessed May 11, 2018), [available at http://www.worldvaluessurvey.org].

83.

Yoganathan

Vignesh

Osburg

Victoria-Sophie

Akhtar

Pervaiz

(2019), “Sensory Stimulation for Sensible Consumption: Multisensory Marketing for E-tailing of Ethical Brands,” Journal of Business Research, 96, 386–396.

84.

Yoo

Boonghee

Donthu

Naveen

(2001), “Developing and Validating a Multidimensional Consumer-based Brand Equity Scale,” Journal of Business Research, 52 (1), 1–14.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.05 MB