Antecedents and Consequences of Customer Engagement Behaviour in the Hospitality Industry: A Moderated Mediation

Abstract

This chapter dwells on antecedents and consequences of customer engagement behaviour in the hospitality sector. It is a moderated mediation with customer involvement, enthusiasm, attention and absorption as the antecedents, and electronic word of mouth and behavioural intentions to loyalty as the consequences or outcomes, while customer engagement behaviour and customer relationship management were used as mediating and moderating variables, respectively. Data were collected from 350 respondents from southern Nigeria and were analysed with the aid of WarpPLS version7. Findings show that customer relationship management largely moderates the direct and indirect effects of the antecedents on the consequences. Implications of the findings were discussed among others that operators in the hospitality sector, an industry that is drastically affected by the COVID-19 pandemic, need to apply key customer engagement behaviour concepts in designing and managing of service experiences.

1. Introduction

Managing customers has evolved over the years and has been the primary focus of marketing and business organisations. Pansari and Kumar [1] aver that customer management has not changed and that what has changed is how customers are managed. Thus, customer engagement (CE) is not a new concept. It is perhaps as old as marketing itself. The advent of information and communication technology especially social media and their wide application in business and marketing has boosted and exacerbated CE especially as it relates to creating and maintaining online brand communities. Kotler et al. [2] state that yesterday’s businesses relied majorly on mass marketing to large segments of customers operating independently and in their self-interest; whereas, nowadays firms are utilising online, mobile, and social media to improve their targeting and to involve as well as engage customers more profoundly and interactively. According to Kotler et al. [2], the traditional marketing entailed marketing brands to consumers; while the contemporary marketing referred to as CE marketing implies fostering direct and dynamic customer involvement in tweaking and modifying brand conversations, brand experiences, and brand community. CE marketing goes beyond just marketing a brand to consumers. The aim is to turn the brand into a meaningful and strongly reminiscent part of customers’ conversations and lives [2]. CE has greatly benefited companies in the United States as demonstrated by Pansari and Kumar [3] and there is also tremendous scope for researchers in other climes to explore CE in depth and contribute to this growing body of knowledge. Dessart et al. [4] stated that CE is receiving increasing attention, yet the current literatures are inconsistent in its dimensionality. Mintz [5] noted that the COVID-19 pandemic has been catastrophic for the world stressing that the fundamental driver of the COVID-19 economic crisis has been health and safety concerns and, hence, changes to customer behaviour. This is particularly so in the hospitality sector with reported millions of job losses. He stressed the need for marketers and businesses to map out strategies to win back customers post COVID-19 as majority of customers switched their shopping strategies from a hedonic, enjoyment-focused strategy to a utilitarian, goals-based strategy (p. 5). Four antecedents of CE: involvement from the study by Coulter et al. [6]; enthusiasm, attention, and absorption as well as an outcome variable, BIL from the study by So et al. [7]; the second outcome: e-WOM from the study by Konttinen et al. [8]. The measures of CE, which was used as the mediating variable, were based on the work of Sprott et al. [9]. While CRM, the moderating variable, was based on Dazagbyilo et al. [10] study. This chapter concerns a conceptual analysis of the CE in the hospitality industry and is divided as follows: hospitality industry, customer engagement (CE), CE and CRM, CE and s-CRM, CE and customer loyalty, conceptual framework of CE and CRM, and finally conclusions and implications.

2. Hospitality industry

The hospitality industry includes hotels, tourism agencies, events centres, restaurants and bars. Livio offline Dictionary define hospitality (i) as the act or service of welcoming, receiving, hosting, or entertaining guests; an appropriate attitude of openness, respect, and generosity toward guests; (ii) the business of providing catering, lodging, and entertainment service, the industry which includes the operation of hotels, restaurants, and similar enterprises; and (iii) the food, drink, and entertainment given to customers by a company or organisation or provided to visitors by a private host. There exists no consensus among various authors on what the definition of the term hospitality industry is. Numerous authors and researchers have taken different routes to explain the hospitality industry. A number of them sought to condense the scope of the sector/industry and its features into comprising both tangible and intangible attributes in the service delivery process. A fraction also tried to depict the industry through studying the stakeholders involved, mutual gains and benefits generated and the industry’s impacts to the society and economy [11]. Generally speaking, Chan and Mackenzie [11] further aver that ‘hospitality is the act of kindness in welcoming and looking after the basic needs of customers or strangers, mainly in relation to food, drink and accommodation.’ A modern-day account of Hospitality conveys the affiliation process between a customer and a host. Rodriguez et al. [12] went further to state that ‘when we talk about the Hospitality Industry, we are referring to the companies or organizations which provide food and/or drink and/or accommodation to people who are away from home’, adding that their delineation only satisfies most situations. Data from the reference [13] show that the classification of the hospitality industry in Nigeria is also based on the standard industrial classification (SIC) codes. These are the accommodation and food services; and transport and storage including road, rail, water, air transportation as well as post and courier services. Also included are tourism, event management and related recreational parks and resorts that provide food, including fast food, accommodation and entertainments.

Operators in the hospitality sector, an industry that is drastically affected by the COVID-19 pandemic, are now required to understand and apply key CE concepts in the design and management of service experiences. The hospitality sector which account for 6.06% of Nigeria’s GDP in 2018 could only account for 5.15% and 4.3% of GDP in 2019 and 2020 respectively, no thanks to COVID-19 pandemic. United Nations World Tourism Organisation (UNWTO) [14] reports that as countries closed their national borders to contain the COVID-19 infection, the effects on tourism, hospitality and events were devastating, with projected impacts estimated at US$910 billion lost in exports and 100–120 million jobs at risk. The hospitality sector as a per cent of Nigeria’s gross domestic product (GDP) as at 2011 is 2.89%; and moved up to 5.93% in 2016 to 6.06% in 2018 from where it declined to 4.29% in 2020. This decline can rightly be attributed to the COVID-19 pandemic. In spite of the decline occasioned by the COVID-19 pandemic, the sector is a significant contributor to the Nigerian economy. The sector has also been variously recognised as making the highest impact on the Nigeria, Africa and World economies in terms of employment generation. In the African continent, World Travel and Tourism Council (WTTC) [15] reports that, as part of the global trends, travel and tourism (T&T) GDP in Africa dose dived by 49.2% in 2020. While domestic spending contracted by 42.8%, international spending saw a deeper reduction at 66.8% adding that Africa suffered disproportionately more than other regions, with jobs dwindling by 29.3% (7.2 million).

At this point, the yearly competitiveness rankings of the World Economic Forum (WEF) come to mind. Nigeria continues to rank poorly in the yearly reports of the WEF. Nigeria ranked 129 out of 139 countries in 2019 report; and in the 2021 report she ranked 110 out the 117 countries covered. The poor rankings manifest majorly in the areas of safety and security. In its 2019 report, World Economic Forum [16] notes that Nigeria (129th) accounts for almost half of the sub-Saharan Africa (SSA’s) T&T GDP and is also its largest economy. It adds that though Nigeria ranks in the middle in terms of competitiveness, her safety and security ranking (139th) is worst in the SSA region.

The hospitality sector is multiple and varied but for this study we delimit to those concerned with accommodation, food and drinks, including hotels, resorts, recreations parks and tourism. Distinct from other sectors, the hospitality industry is unique in its nature which tends to be service-oriented and has a strong emphasis on human exchange in the service delivery processes. Chan and Mackenzie [11] identify key characteristics relating to the sector as: product-service; two-way communication; relationship building; cultural diversity; and labour-intensive operations. The personnel are the most crucial in all these characteristics. As Barrows et al. [17] put it: as firms in competition expand their menus and amenities and dress up their operations, all operations at a given price level tend to become more like one another. The crucial differentiation becomes service—usually in the form of personal service (p. 27). This calls for a more emphasis by the sector operators on build on its CE framework: creating direct and continuous customer involvement in designing brands, and in creating and managing brand conversations, customer experiences, and consumer brand communities. According to Manfreda and King [18], the mature stage of the hospitality industry, characterised by higher levels of customer expectations, increased competition and low product differentiation, has made the importance of staging and managing personalised, high quality guest experiences more paramount. CE allows marketers to create and sustain a competitive advantage and could serve as a differentiation strategy.

3. Customer engagement (CE)

Marketing scholars mostly describe the concept of CE from three perspectives. One, the behavioural viewpoint proposes that it is a non-transactional behaviour of the customer to the brand and is manifested through positive word-of-mouth among others. Two, the psychological stance believes that CE is the customer’s emotional and cognitive reactions toward a brand. Naumann et al. [19] believed that CE is a psychological process that encourages new customers to generate loyalty and old customers to maintain loyalty. Mollen and Wilson [20] believed that CE is the cognitive and emotional commitment with a brand. The dimensions of CE carry the following characteristics [21]: Cognitive engagement describes an investment in attention, processing, or thinking skills to develop understanding or knowledge. Customers as humans can know (have knowledge) either based on experience or based on reasoning [22] while understanding relates to comprehension. While also compared to motivation and self-regulation, cognitive engagement is defined by Johnston [21] as a person’s investment in attention and processing to evoke knowledge and understanding concerning a product or an idea. Affective engagement entails positive and negative emotional reactions, like pleasure, fear, anger, support, and association and is often displayed as recognition of belonging, or emotional reactions [21]. Behavioural engagement embodies concepts of participation, collaboration, action, and involvement, as well as intended and unintended actions that may be caused by, or result from, cognitive or affective engagement [21]. CE has become an integral component of debates on consumer-brand connections in academic study, and likewise in practice. Weitzl and Einwiller [23] define it as a composite, multifaceted relational construct that entails a consumer’s state, that occurs by virtue of interactive consumer experiences with a specific brand. It comprises of psychological and behavioural engagement factors conveying a definite intensity level at a particular time [23]. Psychological engagement denotes a consumer’s captivating, inherent motivation to invest cognitive, emotional, and intentional resources in the interaction with a brand, while behavioural engagement reflects specific interactive, brand-related behaviours [23]. In bestowing the term brand dialogue behaviours, Maslowska et al. [24] acknowledge the increasing role of engagement behaviours beyond that of actual purchase. Within a service setting particularly, prior frameworks have recognised the role of the customer in enhancing the entire experience, yet typically centre wholly on the period of the service encounter. The CE behaviour concept regards users as being guided by own personal intentions and motivations, in lieu of those initiating from the firm [25]. One additional note on Van Doorn et al. [26] definition that it explicitly pertains to CE behaviours, yet the authors proceed to suggest that these behaviours may also be targeted to an expanded network of actors than other current and potential customers. Van Doorn et al. [26] clearly acknowledge the capacity for not only current customers, but consumers in general, to generate these engagement behaviours with either the brand directly or other consumers. Kumar et al. [27] acknowledge Van Doorn et al.’s view [26] yet argue that such a conceptualisation is incomplete while actual purchases remain omitted. Such a stance would further exemplify the requirement for a more holistic view of engagement, such as throughout the entire process in tourism or restaurant.

From a comprehensive perspective, CE can be regarded as a multidimensional concept and includes multiple aspects of cognition, emotion, and behaviour. Vivek et al. [28] followed the expanded relationship metaphor and service-dominant logic, and conceptualised a three-dimensional perspective of CE, that include: conscious attention, enthused participation, and social connection. Mollen and Wilson [20] suggest that online CE includes three dimensions: active cognitive processing, instrumental value, and experience value. So et al. [7] confirmed that CE had identification, enthusiasm, attention, absorption, and interaction. Hollebeek et al. [29] suggested that customer brand fit consists of three dimensions: cognitive processing, affection, and behaviour. Moreover, the object of CE can be products, brands, or activities [30]. Therefore, different types of CE can be distinguished according to the objects of that engagement [30].

Important to the conceptualization of CE is to provide the unique characteristics that differentiate it from other related concepts and constructs. CE appears to be a related concept, though is theoretically different from many similar other marketing concepts [26, 31]. There has been a clear difference between engagement and other, more well-known customer management and relational constructs [20, 31, 32, 33, 34]. CE and involvement seem similar on the basis of customer values and needs that motivate people toward a particular object, like a brand [33]. Vivek et al. [34] proposed that involvement differs from CE because involvement is a psychological concept that does not study behaviours. They argue that involvement may be an antecedent of the behavioural domain of CE. Mollen and Wilson [20] distinguished involvement, since it comprises more passive allocation of mental resources, whereas engagement entails a dynamic bond with the consumption object adding that engagement needs both achievement of instrumental value because of utility and relevance, in addition to/along with a specific level of emotional bonding, which may be achieved due to gratifying and rewarding experiences.

The term engagement in a business-related context originally referred to employee engagement (EE), which seems to enjoy a consistent conceptualization and operationalization. However, the conceptualization of CE, which is still in its infancy, lacks consensus [7]. Buttle and Maklan [35] maintain that this is not unusual for an emerging construct; indeed, competing claims have been made for CRM itself. Interestingly, the stability of the EE construct may provide insight for CE, which is an evolving concept in the customer management field, where it has been drawn from organisational behaviour (cf. employee engagement) [34]. There were scant discussions of CE prior to 2005, but thereafter, there have been emergence of numerous research findings that have been abridged into various literature reviews in [35]. There is no unified agreement about what CE is, how to define it, how to measure it, or what consequences it has for any business [36]. In the organisational behaviour writings, EE denotes ‘the simultaneous employment and expression of a person’s preferred self in task behaviours that promote connections to work and to others, personal presence, and active, full role performances’ [37] (p. 700). EE seems to exist as a motivational construct embracing attention and absorption [38] and might involve an identification component [39].

Consistent with this emphasis on the psychological elements, engagement is a positive, fulfilling, work-related state of mind that is characterised by vigour, dedication and absorption [40] suggesting that EE is a persistent and pervasive affective, cognitive state [41]. These definitions indicate that EE conceptualizations focus on psychological aspects. In contrast, So et al. [7] note that marketing scholars have conceptualised CE to include a strong behavioural focus. Such interests abound in the literature domains of both academics [26, 42, 43] as well as practitioners [44]. In seeking to establish a conceptual understanding of CE, researchers have argued that the knowledge of EE is applicable to CE [45]. Feelings of passion, energy, and enthusiasm characterise both EE and CE [33, 45]. However, the focus of those feelings differs (workplace vs. consumer brand). In addition, in building on the EE literature, the conceptualization of CE tends to go beyond an attitudinal perspective, reflecting both psychological and behavioural dimensions [45]. Buttle and Maklan [35] maintain that CRM practitioners often use tools such as campaign management to build CE.

As Brodie et al. [32] discussion proposes that CE might entail that equal attention be lent to the psychological facets of engagement along with behavioural participation, it shows that there persists a diversity of opinions as regards the conceptualization of the concept. For example, some researchers consider CE to be a behavioural construct (i.e., interaction) emanating from a range of motivational drivers. [26, 42, 43, 46], whereas others propose CE to be a multidimensional construct comprising both psychological and behavioural aspects [32, 33, 45, 47, 48]. Support for the adoption of a multidimensional approach is evidenced in the conceptualization of composite loyalty (i.e., behavioural and attitudinal), which suggests that behavioural measures alone may lack a conceptual basis in [7] and provide insufficient insight into the factors underlying repeat behaviour. So et al. [7] argue that this is also correct in describing the conceptual domain of CE, whereby involvement in the activities of the phenomenon does not guarantee a truly engaged customer. For example, involvement in a brand conversation conference or gathering may emanate from issues like the need for product information or reduction of perceived risks [32] rather than from being attached or engaged with the brand. So et al. [7] maintain that the truly involved customer must have an enduring psychological attachment with the brand in addition to behavioural involvement or engagement, adding that while a behavioural approach may provide an indication of customers’ involvement level in CE activities, a multidimensional approach will express the full complexity of CE.

In line with the above, Buttle and Maklan [35] identified two main schools of thought on CE. The first sees CE as a two- dimensional construct with a behavioural and an emotional component (see: [35]). Extremely involved customers devote substantial share- of- wallet to the brand with which they are involved and are also emotionally dedicated to it. This school of thought makes CE hard to differentiate from the customer loyalty construct, described as having a behavioural and attitudinal component. The second school sees CE as a multi- dimensional construct in which an attached customer is not just a buyer of a firm’s products but is involved in co- creation of value for the brand in many other indirect ways [1]. The focal point of this school is on the brand owners’ activities to promote customers’ indirect involvement with the brand. The main aim is to convert the customer into an additional, unpaid marketer, working in the brand’s interests. Whereas direct involvement means buying, indirect involvement could manifest in many different non- transactional activities [28].

CE has continued to attract increasing attention from both practitioners and academics [32] in part owing to the growth of the information technology and the social media as essential tools for customer communication and cooperation. Specifically, the online environment has created a range of new media channels for the hospitality firms to enhance connections and relationships with customers far-off the service encounter [7]. In a bid involve their customers via interactivity beyond purchase, tourism brands establish their presence on social network sites like Facebook and Twitter in addition to online interaction panels. As a medium of exchange, the Internet enables hospitality business operators and consumers to spread and disseminate information, opinions, and experiences, not just from business to customer but also from customer to customer [49]. These discussions highlighted the importance of involving customers to build loyalty after the transaction, especially in the highly competitive environment of the hospitality industry. The importance of non-transactional customer discussions is detailed in literature [7]. For instance, online user-generated evaluations and assessments can influence the number of online bookings in a hotel [50] as well as intentions to book and perceptions of trust in the hotel [51]. In an off-line environment, opinion or advice from existing customers influences the consumer’s purchase decisions [52]. Collectively, such interactions form the behavioural manifestation of CE [26, 43, 46]. Additionally, hospitality organisations can leverage CE behaviours to attract and retain more customers and gain additional insight into their business [53]. From a consumer perspective, the benefits for engaging in CE activities include financial gains or incentives as well as emotional fulfilment, such as enjoyment and positive affect [26]. CE is emerging as a construct that may enhance loyalty and purchase decisions e.g., [45, 47] through a strong, enduring psychological connection accompanied by interactive brand experiences beyond purchase. CE with a brand influences important aspects of consumer brand knowledge, brand perceptions, and brand attitudes, and hence brand loyalty [9].

Buttle and Maklan [35] identify four forms of CE: behavioural, social, cognitive and emotional. (i) Behavioural: the engaged customer acts favourably toward the brand, for example by taking part in brand research or passing on positive word- of- mouth thereby creating customer referral value or CRV. (ii) Social: the involved customer connects with the brands and other customers in social media channels, through creating, viewing or sharing online content, taking part in crowd- sourced customer service, blogging, recording assessments or complains, sharing brand-use information on Instagram, or joining a Twitter interaction among others. According to Buttle and Maklan [35] these two types of engagement are stimulated by allowing the intra- personal characteristics of the customer: 1) Cognitive. the engaged customer is knowledgeable about the brand like, the brand’s values, price- point, advantage of the product relative to competitors, status or country of origin. 2) Emotional. The involved customer has a powerful liking for and devotion to the brand. It is also imperative to note that engagement cannot be dichotomised, that is, customers cannot just be separated into engaged and disengaged segments. Rather, customers are always changing with rest to their level of engagement. Some customers will be mostly engaged in all the forms; while others may not even realise what brands they have bought. Buttle and Maklan [35] state that managers and marketers should be concerned with the role of CE in building and sustaining relationships with customers stressing that engagement is only possible when a relationship is entrenched on the basis of trust and commitment.

So et al. [7] CE study rely on five distinct dimensions of identification, enthusiasm, attention, absorption, and interaction, which reflect the psychological and behavioural aspects, defined as a customers’ personal connection to a brand as manifested in cognitive, affective, and behavioural actions outside of the purchase situation. Behavioural manifestations include participation in activities, such as customer-to-customer interactions, blogging, writing reviews, as well as other similar activities that are centred on the brand. Recent reviews of the conceptual foundation and relationship of CE provide useful guidance on potential antecedents and consequences of CE. Possible antecedents of CE include involvement, interactivity, rapport, commitment, trust, brand attachment, and brand performance perceptions [26, 33]. Consequences of CE include cocreated value, brand experience, satisfaction, trust, commitment, customer value, brand loyalty, customer equity, firm reputation, brand recognition, and financial outcomes [26, 33]. In addition, such a psychological connection may depend on various situational factors [54] such as age, computer experience, and degree of socialisation. It does appear that engagement and involvement are correspondingly hinged on consumer needs/values stimulating the individual toward a specific object, such as a brand [47]. In line with marketing literature, involvement most often concerns the perceived personal importance and significance of the product or brand (see [7]). However, engagement requires more than the exercise of cognition. CE entails an active relationship with the brand, and the intention to act makes CE distinct from involvement’s more passive allocation of mental resources [20]. Nevertheless, the emergence of specific customer brand engagement levels requires some level of involvement with a focal brand [33]. These characteristics make the multi-faceted concept of CE conceptually distinct from involvement. Additionally, studies [33, 45, 47] provide extensive reviews of how CE is different from other similar constructs, such as commitment, satisfaction, cocreation, and brand loyalty.

3.1 CE and customer relationship management (CRM)

Over the years, marketers have employed various tools in managing their customers and these range from: transaction marketing, customer loyalty and loyalty management, relationship marketing, customer relationship management and of recent CE. Customer relationship management (CRM) as a tool is very beneficial for business firms in building and expanding their relationships with their customers [55]. In a review and classification of CRM researches from 2000 to 2020 Mena and Sahu [56] show that though CRM is prominent within service industries, it has become a potent tool in all industries from manufacturing to tourism and hospitality; from education to logistics and telecommunications among others; as the number of published research articles concerning other sectors is increasing compared to that of service industries. As a customer management strategy (CMS), CRM is the strategic process of selecting customers that a firm can most profitably serve and shaping interactions and management of technology between a company and these customers; the ultimate goal is to optimize the current and potential value of customers for the company [57]. Thus, CRM can be described as a business and marketing relationship strategy based on supported by methods and technology. The most complete definition is recommended by [58] who considered CRM as a customer-oriented business philosophy that involves analysing, planning and controlling customer relationships by means of modern information and communication technologies (p. 1). Chen and Popovich [59] suggest that a CRM model should mix the three proportions of people, process and technology inside the context of an enterprise-wide, customer-driven, technology integrated and cross-functional organisation [59]. Through these combinations, the organisation can choose particular technologies to improve its knowledge of customers and performance as well as enhance customer relationships. From this perspective, Bozbay [55] defined CRM as a global procedure that allows a lasting and profitable relationship between the organisation and customers. While CRM definitions vary, they can be grouped into three types – technology centred, customer life cycle-centred, and strategy-centred explained as below [60]. Technology-centred definitions establish the link between technology and CRM. It is not surprising that an investment in CRM technologies has been made, and the conversation has drawn CRM into the technological and practical mechanics [60]. CRM is a technology solution that expands to separate databases and sales force automation tools to integrate sales and marketing functions to reach targeted efforts. On the technology perspective, Buttle and Maklan [35] maintain that IT companies have tended to use the term CRM to describe the software tools that are used to support the marketing, selling and service functions of businesses. CRM is a tool used in one-to-one customer communications, sales, service, call centres or marketing departments. In fact, as we have already noted, it is one of the modern tools for customer management and when used properly with the CE, will offers strong competitive advantage to firms. CRM is not just a tool for those departments. It is for every department within the entire organisation. According to Bozbay [55], if CRM strategy is well applied within the whole organisation, all the organisations’ departments like marketing, human resources, R & D, finance and information technology will succeed in maximising healthy and profitable relationships with customers. CRM is customer-oriented, technology-integrated and cross-functional strategy that facilitates customer personalization, simplicity and convenience in interactions [59] and is significantly a marketing strategy that firms employ to improve customer value. It is also a set of concepts that must be blended and harmonised together with an organisation’s overall business strategies [61]. Interestingly, CRM ballooned to a major transformation from a strategy that relied only on the customer transaction to accommodate customer connections [62]. Currently, marketers can extract initial information about customers that organisations can use to realise greater success in carrying value to the customer [34]. Previous research has done a great deal of modelling on the use of technologies and their impact on CRM, but with the advent of social media, more marketers realise that technologies are already great enabling factors for CRM [63]. Therefore, CRM has a new name called social CRM [63, 64]. In this study, CRM is used as a moderating variable between antecedents and outcomes of CE.

3.2 CE and social CRM

Social CRM is a postulation much like the CRM however embodies and integrates social methods, capabilities and operations that function through the communication between organisations and customers as well as the customers and their peers, families and friends [65]. Additionally, the presence of these novel methods, procedures and technologies facilitates interactions with customers [66] to build long-term relationships with improved performance [67]. Social CRM therefore focuses on CE through communicating, bi-directional relationships with customers where they are ready to participate in the marketing activities and the product offerings through interactions in social media [12, 63]. Mobile devices and social media have changed the relationship between organisations with the customers pushing them to reach and create strategies to manage the relationship with customers beyond just financial transactions [68]. Statista (2022) blog, reports among others that as at January, 2022, Facebook has 2.91Billion active users while WhatsApp has 2.4Billion active users. This makes Facebook the single largest community in the world. These social media communities share information, ritual and concern for each member. Any business organisation can ignore these communities at its own peril. Business organisations rely on these large communities to build online brand communities to relate and engage customers for profit. Accordingly, business and marketing practitioners need to understand how to promote and maintain online communities for profitable CE. In his seminal book, Marketing 2.0: bridging the gap between seller and buyer through social media marketing, [69]: encourage SMB CEOs and their marketing and sales managers to embrace social web as three things: a culture, a mindset and as a platform (p. 4). He adds: the social web allows any business of any size in any location to reach the people they desire to reach and build strong relationship with them (p. 4). According to Bozbay [55], a study showed that extremely engaged customers pay 23% more, which increases profitability and share of wallet; the Convero survey found that 74% of managers plan to make their investments on CE in the following years [64]. The social media channels enable business firms and organisations to involve with customers under their own circumstances, at work, play or at any time they want, and through their own preferred media [55]. Marketers appreciate and help customers to buy more, assist them in using the brand, make them more knowledgeable about the brand and handle the customers’ complaints. Through the product cycle, the firm can utilise social media to enhance its speed in the market, to assist in designing innovative products based on the customers’ desires and aspirations, to boost early sales quicker in order to nurture their prices, and to know the features and functions that appeal more to the customers [55]. Firms also use the social media for the optimisation of the costs of sales, service and marketing expenditures by involving customers and handling transactions through replacing the traditional media by the new media channels and by listening to the voice of customers to minimise the cost of failure [70]. From a strategic point of view, experts describe engagement as: allow businesses to cultivate in-depth, more thoughtful and sustainable discussions between the organisation and its customers or external stakeholders [71]. As soon as customers connect through the brand, the amount of time spent sharing information through different media is likely to be enhanced over the internet either in form of comments on other user posts or through content creation. By verifying customer purchases over certain period, buyers can be monitored easily, possibly to contribute to the development of the products by generating ideas. The number of stories generated or even shared by the customer and a satisfaction assessment could as well be done online. Thus, this can lead managers to provide reasonable understandings of organisation performance [27]. Social media is reengineering the business processes and methods by facilitating the two-way communication strategy between the organisations and the customers. Thus, it can develop many new challenges and opportunities. Sharing resources and gaining understanding are the prerequisites for the long-term sustainability of the organisations [72]. Online communications can generate huge knowledge and lead to the creation and growth of customer value [64, 73]. Woodcock et al. [74] noted that social CRM avails full support to customer life cycle strategy and CMS that will enhance sales by minimising costs, spreading and enhancing involvement and awareness. According to Bozbay [55] social CRM can engender many benefits to firms by following a four-step procedure – involving/engaging customers and prospects, attracting new customers, recalling customers and increasing customer value. Thus, social CRM supports the entire CMS and customer life cycle and therefore should lead to enhanced sales through improving engagement and awareness, and improve customer value and minimise costs [61]. CRM philosophy helps to understand the main components in customer management such as attracting customers, maintaining loyalty and retaining them; and the newest component in managing customers is CE. CE (CE) is defined as a kind of mind generated by customers interacting with the brand in a specific service relationship and creating an experience [30]. Because it relies majorly on utilising social media for communication and interaction, CE is also referred to as social customer relationship management (s-CRM). Greenberg [65] maintain that understanding the customer the right way applies to Social CRM as it is focused on CE, and the recognising that the customer controls the conversation, stressing that when it comes to how you engage customers, the primary strategy remains what it has always been, and that’s the people, whether in 21st century or not. He adds that the kind of culture that disseminates throughout a company is a key determinant in the effort to make that CE fruitful, to the point of creating a customer relationship that is both delightful and extraordinary, (pp. 93–94). CE is a dynamic and cyclical process and has different performances in different situations [30].

In customer journey analysis, firms rely on customers interactions and how they interact with multiple touch points, from consideration, search, and purchase to post purchase, consumption, and future engagement or repurchase. The main aim of such analyses is to define this journey and appreciate the customers’ options and preferences for the touch points in multiple purchase phases [43]. Lemon and Verhoef [75] stated the increasing focus on customer experience arises because customers now interact with firms through myriad touch points in multiple channels and media, social and offline media, resulting in more complex customer journeys. Van Hagena and Brona [76] recommends measuring customer experience and determining how strong the emotional level in various customer journey phases in different groups of passengers so that the customer experience of each customer journey phase is known. Aaker and Joachimsthaler [77] stated that the experience customers acquire through participation and engagement on the internet has the implication to be captured more firmly than the experience from other traditional media, hence it can be said that the quality of user experience on a website affects the overall feeling and trust associated with brands which could be deepened and firmed more strongly than experience through other media [78]. As customers interact with products and services, they share their experience on social media which affect relationships and engagement with the firm.

3.3 CE and customer loyalty

Loyalty is the end result of CE in the hospitality sector especially with COVID-19 when there were noticeable declines in revenue. Loyalty is also very essential post COVID-19 as business operators in the sector try to win back customer trust and confidence. According to Boohene and Agyapong [79] loyalty as a concept has its base from the consumer behaviour theory and is something that consumers may portray to brands, services or activities. Customer loyalty is the normal willingness of customer to maintain their relations with a particular firm or service/product [80]. According to Wirtz and Loveloch [81] loyalty refers to the submissiveness of a customer to continue patronising a company’s product and services over a long time and on a persistent and rather exclusive basis, and willingly endorsing and advocating the firm’s products to friends and associates. Customer loyalty results from a firm creating a benefit for customers so that they will maintain progressively repeat purchases with the organisation [82]. Oliver [83] defined customer loyalty as a deeply held commitment to rebuy a preferred product or service consistently in the future, causing repetitive same brand or same brand-set purchasing, despite situational influences and marketing efforts. Leong et al. [84] note that the economic growth noticed in the tourism and hospitality industry and competition in the industry has led to the discovering consumer loyalty as a key success factor. Reichheld [85] pointed out loyalty behaviour affects business growth and companies get to profit from price premium, referrals, increase purchases and higher balances, reduced operating cost and customer acquisition cost. According to Srinivasan, Anderson and Ponnavolu [86] loyalty in online behaviour is attitudes that are beneficial to the customer and his dedication to online companies that engender repeat purchase behaviour. A truly loyal customer is a dedicated customer that is connected with the retailer and may not be easily bothered by more and perceived interesting alternatives [87]. In general, customer loyalty is the final purpose that firms implement CRM and CE.

Researchers have recognised various vital concepts of the CRM which are potentially linked to CE [26, 48, 88]. These constructs include satisfaction, brand trust, commitment, and service quality, all of which are essential to the development of loyal relationships [89]. Nevertheless, very scant empirical confirmation exists to provide a clear understanding of the relationship between CE components and behavioural intentions toward loyalty (BIL). Such information is fundamental to both researchers and marketing practitioners, going by companies’ growing interest in CE strategies and in view of the significant amount of academic interest availed to this emerging phenomenon as a serious determinant of loyalty. Interestingly, CE can support important and profound marketing metrics like share of wallet, loyalty, cross-selling, and word of mouth [34]. CE influences behaviour intensions of loyalty significantly in the hospitality and tourism brands of hotels and airlines [90]; but further empirical investigation of this relationship in different contexts has been suggested [91]. The outcomes of CE in this present study includes e-WOM and behavioural intentions to loyalty (BIL).

4. Conceptual framework

The framework for explaining and understanding CE evolved from the relationship marketing model where the foundation is based on Morgan and Hunt commitment trust theory [1]. Previously, the primary purpose of relationship marketing was to establish long-term relationships with the firm, thereby promoting efficiency, productivity, effectiveness, and cooperation [1]. Tracing the development of CE, Pansari and Kumar [1] noted that a firm’s initial relationship with the customer was restricted to purchases, ensuring long-term loyalty, and continued patronage, noting that, nevertheless, this has evolved with the developments in the marketplace based on the ever-evolving needs and interests of the consumers. Accordingly, CE has hypothetical roots within the extended domain of relationship marketing that emphasise the notions of interactivity and customer experience [34]. CE is considered as the creation of a deeper and meaningful connection between the company and the customer [92]. It is widely believed that a well engaged customer plays a key role in viral marketing activity and by providing constant referrals and recommendations [32] and increasing advocacy intention for a specific products, services, and brands to others either face-to-face and on various channels including social media. Dessart et al. [4] stated that CE is receiving increasing attention, yet the current literatures are inconsistent in its dimensionality.

In a case study-based research, Singala [93] explains the use of art for business purposes by showing how the Cube has embedded art into its servicescape and experience design to position its wine brand in the d’Arenberg Cube, Australia; adding that by synergising art, wine and tourism, the d’Arenberg Cube managed to create the living culture and experiences of its servicescape that are hard to be replicated by competitors. This new normal emphasises that CE is evolving and so are the theories for explaining and understanding the CE phenomenon. In a study on how celebrity endorsement effect can help CE in promoting tourism products through Live Streaming, Qui et al. [30] argue that consumer trust predicts the three dimensions of CE (cognitive processing, affection, and activation); and that the three dimensions of consumer trust also play a positive mediating role between celebrity effects and CE. This study relies on four antecedents of CE: involvement, enthusiasm, attention, and absorption as well as outcomes, BIL e-WOM. CE was used as the mediating variable while CRM is the moderating variable. Based on this we conceptualise our research model as in Figure 1.

Figure 1.

The proposed conceptual model.

Our model (Figure 1) has independent variables (IVs): involvement, enthusiasm, attention, and absorption; two dependent variables (DVs): e-WOM and BIL; one mediating variable: CE; and one moderating variables: CRM, as already defined.

Involvement. Involvement is variously conceptualised and is also differently measured by authors and researchers. It is a function of the goal object, the individual consumer and the decision situation [94]. Product involvement generally relates to self-relevance, and is defined as the personal importance of a product category [in 90]. Product involvement has significant effects on consumers’ cognitive and behavioural responses including processing, search, retention, brand commitment, satisfaction [see 90]. Research findings show that product involvement is influenced by how consumers link the product category to key life themes and life projects [6]. Based on this the following hypotheses are formulated:

H1a: There is a significant relationship between involvement and CE.

H1b: CRM moderates the relationship between involvement and CE.

H2a: There is a direct and significant relationship between involvement and e-WOM.

H2b: CRM would moderate the direct relationship between involvement and e-WOM.

H2c: There is an indirect and significant relationship between involvement and e-WOM through CE.

H2d: CRM would moderate the indirect and significant relationship between involvement and e-WOM through CE.

H3a: There is a direct and significant relationship between involvement and BIL.

H3b: CRM would moderate the direct and significant relationship between involvement and BIL.

H3c: There is an indirect and significant relationship between involvement and BIL through CE.

H3d: The indirect and significant relationship between involvement and BIL through CE is moderated by CRM.

Enthusiasm. Enthusiasm is an individual consumer’s strong level of excitement and interest with reference to the focus of engagement, such as a brand [48]. Conceptually, there are two complementary aspects of the affective dimension of engagement: enthusiasm that refers to a consumer’s intrinsic level of excitement and interest and enjoyment which shows a consumer’s pleasure and happiness derived from interactions with the community or content [4] (p. 35). Analysis of the relevant literature suggests that the feeling of enthusiasm as a positive and central indicator of affection is a central to a customer’s engagement with a brand. Based on this the following hypotheses are formulated:

H4a: There is a significant relationship between enthusiasm and CE.

H4b: CRM moderates the relationship between enthusiasm and CE.

H5a: There is a direct and significant relationship between enthusiasm and e-WOM.

H5b: CRM would moderate the direct and significant relationship between enthusiasm and e-WOM.

H5c: There is an indirect and significant relationship between enthusiasm and e-WOM through CE.

H5d: The indirect and significant relationship between enthusiasm and e-WOM through CE would be moderated by CRM.

H6a: There is a direct and significant relationship between enthusiasm and BIL.

H6b: CRM would moderate the direct and significant relationship between enthusiasm and BIL.

H6c: There is an indirect and significant relationship between enthusiasm and BIL through CE.

H6d: The indirect and significant relationship between enthusiasm and BIL through CE would be moderated by CRM.

Attention. Researchers have highlighted attention as a key antecedent of CE. Customers who are highly engaged tend to devout serious attention, consciously or unconsciously, on the object of engagement [7]. They pointed out that marketing theory supports of the inclusion of attention as a component/dimension of CE. The concept of attention agrees with the construct of conscious engagement [48], that expresses a consumer’s level of involvement or attention toward a brand. A customer who is involved with a brand is fascinated to information relating to the brand. For instance, a highly involved customer of Sheraton Hotels or NICON Hotel will no doubt commit a greater deal of attention to the hotel’s brand information, such as public relations, and other product information. Therefore, attention, representing a consumer’s attentiveness and focus on the brand, is considered to be an important dimension of CE [7]. Based on this the following hypotheses are formulated:

H7a: There is a significant relationship between attention and CE.

H7b: CRM moderates the relationship between attention and CE.

H8a: There is a direct and significant relationship between attention and e-WOM.

H8b: CRM would moderate the direct and significant relationship between attention and e-WOM.

H8c: There is an indirect and significant relationship between attention and e-WOM through CE.

H8d: The indirect and significant relationship between attention and e-WOM through CE would be moderated by CRM

H9a: There is a direct and significant relationship between attention and BIL.

H9b: CRM would moderate the direct and significant relationship between attention and BIL.

H9c: There is an indirect and significant relationship between attention and BIL through CE.

H9d: The indirect and significant relationship between attention and BIL through CE would be moderated by CRM.

Absorption. Dessart et al. [4] explicate sub-dimensions of CE to include enthusiasm, attention, and absorption among others; and defined absorption as the level of consumers’ concentration and immersion with a focal engagement object (p 35). Absorption refers to an effortless concentration, loss of self-consciousness, distortion of time, and intrinsic enjoyment [7]. In the marketing domain, scholars have also argued that strong engagement extends beyond concentration to absorption or being engrossed with a something in the study by So et al. [7]. Absorption is a pleasant state in which the customer is fully concentrated, happy, and has a deep engrossment as he/she plays his/her role [45] and an absorbed customer interacting with the brand or other customers perceives time as passing very fast. It is accepted fact that an absorbed or engaged customer of a hospitality brand can devote reasonable time reading or writing customer reviews on the Internet. The involvement engagement literature shows that a deeper level of concentration and total immersion in a person’s responsibility when interacting with the firm, its offering, or other customers, signals a strong level of CE [7]. Based on the above, we formulate additional hypotheses:

H10a: There is a significant relationship between absorption and CE.

H10b: CRM moderates the relationship between absorption and CE.

H11a: There is a direct and significant relationship between absorption and e-WOM.

H11b: CRM would moderate the direct and significant relationship between absorption and e-WOM.

H11c: There is an indirect and significant relationship between absorption and e-WOM through CE.

H11d: CRM would moderate the indirect and significant relationship between absorption and e-WOM through CE.

H12a: There is a direct and significant relationship between absorption and BIL.

H12b: CRM would moderate the direct and significant relationship between absorption and BIL.

H12c: There is an indirect and significant relationship between absorption and BIL through CE.

H12d: The indirect and significant relationship between absorption and BIL through CE would be moderated by CRM.

e-WOM. Word-of Mouth (WOM) can be defined as communication and exchange of information between customers as relates to the features of a company or the product and services. This communication could verbal as in informal discussions or through social and or electronic media, electronic word-of-mouth (e-WOM). Avidar [95] aver that e-WOM has become very popular among consumers who share online marketing information on social media platforms. The information informs their attitudes and behaviours toward services and products since they perceive its sources as more trustworthy than organisational messages [96]. As suggested by Chu and Kim [96] positive electronic e-WOM influence the attitudes and behaviour of potential adopters to form a favourable attitude toward products and or services and adopt them. Consumers’ brand experiences can easily be deduced from their messages about those brands in various digital channels: social media as well as product reviews and recommendations) [97]. Customers’ online experiences are highly related to their behaviour and intentions, and e-WOM. The importance of customer satisfaction with product and service offerings cannot be overemphasised; as satisfied customers are likely to produce favourable WOM related to brand offerings [98]. We formulate additional hypotheses based on the above:

H13a: There is a significant relationship between CE and e-WOM.

H13b: There relationship between CE and e-WOM is moderated by CRM.

Behavioural Intentions to Loyalty (BIL). BIL, such as purchase/repurchase intentions, and e-WOM intentions [97] are typical outcomes of CE. Brand-related outcomes, such as brand satisfaction and loyalty [97], are equally often identified as outcomes of CE. Empirical results [99] reveal that CE has strong influence on satisfaction, likewise loyalty and trust. Rather [99] also report that commitment, satisfaction and trust mediate the relationship between CE and loyalty and that CE increases satisfaction, commitment, trust, and loyalty. Dwivedi [100] define consumer brand engagement as consumers’ positive fulfilling, brand-use-related state of mind characterised by vigour, dedication and absorption; and reported that brand engagement has significant effect on loyalty intentions. This study aims to investigate the impacts of the CE antecedents of involvement, enthusiasm, attention and absorption, on CE and the impacts of CE on outcomes: e-WOM and BIL. We also set out to explore the mediating effects of CE between the relationship of the antecedents and e-WOM and BIL. It will also investigate the moderating effects of CRM on the direct and indirect relationship between the antecedents on e-WOM as well as BIL in the hospitality industry.

H14a: There is a significant relationship between CE and BIL.

H14b: There relationship between CE and BIL is moderated by CRM.

5. Methodology

5.1 Measurement

Eight variables were involved in this study and based on literature each was measured with a number of items. Involvement (INV) was measured with 9 items; enthusiasm (ENT) has 5 items; attention has 6 items; absorption 6 items; while e-WOM and BIL, the two DVs has 4 items each. The other two constructs: engagement (ENG) which is the mediating variable was measured with 8 items and CRM, the moderating construct was measured with 6 items. These variables/constructs were used to form the research model (Figure 1). All the constructs’ items were measured with five-point Likert scale (5 = strongly agree to 1 = strongly disagree) with a neutral option ‘undecided’, which implies a free-choice scale questionnaire.

5.2 Demographics and data

The respondents for this study were drawn from staff and customers of tourism organisations, events centres, hotels and recreational parks majorly from southern Nigeria within the first quarter and early second quarter of 2022. The questionnaire was distributed through online platforms notably WhatsApp platforms which the author is part. The study was based on a sample of 600 respondents, of which only 350 representing approximately 58% responses were obtained and used in the analysis. Analysis of the demographics show that 65.7% of the respondents are males while 34.3% are females. Most of the respondents, 80% are within their prime ages of 36–55 years of age, 10% are under 36 years, while the remaining 10% are above 55 years. While 94.3% of the respondents are married with 5.7% unmarried, all the respondents indicate that they have postgraduate qualification. This pattern of response was informed by the membership platforms used in soliciting for the responses. On monthly income, 40% earn below ₦150,000:00 a month, 31.4% earn between ₦150,000:00 to ₦300,000:00 monthly while the remaining 28.6% earn above ₦300,000:00 monthly. The demographics indicate that the respondents are well in position to appreciate the import of the study for valid responses.

5.3 Common method Bias (CMB)

Researchers are enjoined to ensure that CMB, which arises in survey researches where questionnaire, like this study, is used to collect information on both the IVs and DVs, does not constitute a problem. Where CMB is a problem, it influences item validities, reliabilities, and the covariation between latent constructs [101]. Based on this, research outcomes where CMB is serious cannot be relied upon as they could be misleading. Qualitative and quantitative measures are available for addressing issues of CMB. To address this, respondents were assured that there were no right or wrong answers. They were also assured of anonymity and that their responses are for academic purposes only. On the quantitative aspect, the exploratory factor analysis (EFA) shows that no single factor account for majority of the variances. We used WarpPLS to test our hypotheses and Koch [102] recommends using full collinearity. The result of the average full collinearity VIF (AFVIF) is 2.463 which is well below the recommended threshold of 3.3. This is an indication that CMB is not a serious concern in this work.

5.4 Statistical technique

Partial least squares structural equations modelling (PLS-SEM) was utilised in estimation and validation of the hypotheses for this study. Hair et al. [103] have established that PLS-SEM is preferable when estimating formative constructs or a mix of formative and reflective measures which is the situation in this paper. The PLS-SEM is a non-parametric equivalent of SEM that makes less demand on data compared to the CB-SEM; however, the PLS-SEM software used for the analysis is WarpPLS version 7.0, which brings together the precision of CB-SEM under common factor model assumptions with the nonparametric characteristics of classic PLS algorithms [102]. Also, the sample size used for the analysis is 350 is considered adequate as in similar studies [104] that employed the PLS-SEM tool. Data entry, cleaning and descriptive analysis were done with the aid SPSS version 25 software.

6. Analysis

In analysing our data, we relied on partial least squares structural equations (PLS-SEM) modelling and like other SEM tools, the analysis follows two procedures: the measurement model and the structural model. The measurement model involves the item loadings, reliability and internal consistencies and the discriminant validity measures. All these are necessary because unreliable and non-valid items cannot be used to assess the structural model.

Table 1 contains information on the psychometric properties showing information on the items and their means, standard deviations (SD), p-values, average variance extracted (AVE), Cronbach’s alpha (CA) and construct reliability (CR). This research involved eight constructs: involvement (INV), enthusiasm (ENT), attention (ATT), absorption (ABS) as IVs; e-WOM and behavioural intention to loyalty (BIL) as DVs; CE (CE) as the mediating variable; and CRM as the moderating variable. The means and the SDs show that the respondents are in agreement with the dimensions of the model. INV was measured with 9 items, ENT and ABS with 6 items each, CE with 8 items, CRM with 6 items while the 2 DVs were measure with 4 items each. Items that loaded below 0.6 were removed while those that loaded above were retained (Table 1). Removing the items that loaded poorly enhanced the R square values to 0.55 for CE; 0.66 at e-WOM; and 0.59 at BIL (Figure 2) which justified the removal of the items. The remaining items have t-values above 1.96 and p-values well below the 0.01 margin of error which is a justification for their retention. The three tools: AVE, CA, and CR are diagnostic measures of reliability and the thresholds are: 0.5, 0.6 and 0.7 respectively [105]. The information in Table 1 show that apart from ABS with extracted variance of 0.495 and e-WOM with Cronbach’s alpha of 0.570, our constructs are above the acceptable thresholds which implies that our scale has internal consistency and merit further analysis. The next is the discriminant validity analysis.

Items	Loadings	Mean	SD	t-values	p-values	AVE	CA	CR
INV1	0.727	3.8286	1.07019	15.114	<0.001	0.521	0.846	0.883
INV2	0.658	4.0429	1.24890	13.555	<0.001
INV3	0.745	4.2714	.82791	15.534	<0.001
INV4	0.782	3.8286	.94204	16.392	<0.001
INV5	0.761	4.0286	.82877	15.896	<0.001
INV6	0.660	4.3143	.68848	13.582	<0.001
INV9	0.709	4.2143	.86135	14.711	<0.001
ENT1	0.703	4.1571	.92150	14.566	<0.001	0.595	0.825	0.879
ENT3	0.691	4.2000	.74940	14.295	<0.001
ENT4	0.844	4.1857	.76253	17.845	<0.001
ENT5	0.907	4.2143	.77372	19.347	<0.001
ENT6	0.685	4.6857	.80370	14.167	<0.001
ATT1	0.754	2.9857	.93473	15.741	<0.001	0.613	0.839	0.887
ATT2	0.863	3.3714	.96011	18.301	<0.001
ATT3	0.872	3.3286	.98297	18.506	<0.001
ATT4	0.734	3.2714	.84503	15.284	<0.001
ATT6	0.673	4.0429	.78371	13.890	<0.001
ABS1	0.621	3.8429	.69041	12.713	<0.001	0.495	0.743	0.830
ABS2	0.715	3.7286	1.00031	14.850	<0.001
ABS3	0.735	3.8000	.99569	15.291	<0.001
ABS4	0.753	3.8571	.76240	15.713	<0.001
ABS6	0.686	3.9429	.86123	14.189	<0.001
CE2	0.860	3.4429	.84018	18.242	<0.001	0.623	0.877	0.908
CE3	0.654	3.7429	.85789	13.453	<0.001
CE5	0.735	3.3571	.79460	15.299	<0.001
CE6	0.857	3.5857	.88756	18.156	<0.001
CE7	0.770	3.5000	.89170	16.105	<0.001
CE8	0.839	3.4714	.92283	17.737	<0.001
CRM1	0.766	3.2857	.98909	16.029	<0.001	0.589	0.762	0.849
CRM2	0.867	3.8571	.66181	18.389	<0.001
CRM3	0.618	3.6714	.73241	12.646	<0.001
CRM4	0.796	3.8000	.71014	16.727	<0.001
e-WOM1	0.684	3.7286	.82791	14.130	<0.001	0.517	0.570	0.756
e-WOM2	0.636	3.5857	.99415	13.043	<0.001
e-WOM3	0.673	3.4429	1.02457	13.888	<0.001
e-WOM4	0.650	3.4429	.96702	13.360	<0.001
BIL1	0.872	3.7857	.75498	18.517	<0.001	0.772	0.901	0.931
BIL2	0.811	3.8571	.81683	17.064	<0.001
BIL3	0.915	3.9286	.66258	19.548	<0.001
BIL4	0.913	3.9000	.75984	19.507	<0.001

Table 1.

Psychometric properties of the construct.

Note: AVE = Average Variance Extracted, CA = Cronbach’s Alpha, CR = Composite Reliability.

Figure 2.

a. the mediation model; b. moderated mediation model.

Discriminant validity is one item of construct validity that relates to the degree to which two constructs are distinctive; as every construct in analysis must be proved to have discriminant validity from all other scales [105]. Discriminant validity shows the extent to which summated scales are distinct. The correlation must be low less than 0.7 [106] to show that the constructs are distinct. High correlation between IV and DV have no problem but for IVs is indicative of collinearity. Table 2 show high correlation of 0.757 between CRM and CE but these are the moderating and mediating variables respectively hence they are retained. Moreover, the full collinearity variance inflation factors (FCVIFs): INV = 1.982, ENT = 1.694, ATT = 3.143, ABS = 2.541, CE = 3.140, CRM = 2.816, e-WOM = 2.092 and BIL = 2.292 are all within acceptable range (3.3–5.0) and thus no construct need to be removed. Discriminant validity also implies that the diagonal correlations must be higher than all the others below it. All other correlations between the constructs are well within range showing that our scale has discriminant validity. We go to structural model assessment.

	INV	ENT	ATT	ABS	CE	CRM	e-WOM	BIL
INV	0.722
ENT	0.539	0.771
ATT	0.422	0.278	0.783
ABS	0.468	0.451	0.660	0.704
CE	0.252	0.343	0.511	0.505	0.789
CRM	0.291	0.282	0.512	0.534	0.757	0.767
e-WOM	0.306	0.218	0.630	0.571	0.551	0.528	0.661
BIL	0.565	0.447	0.482	0.571	0.461	0.541	0.239	0.879

Table 2.

Fornell-Larcker discriminant validity analysis.

Figure 2a shows the mediation model of the study showing the four IVs, the mediator as CE and the two DVs. This is the first structural analysis before the moderation since our study is a moderated mediation. As shown in the figure, the coefficient of determination R square at CE is 0.55, which implies that 55% of the variances in CE are accounted for by the four IVs. The R square at e-WOM is 0.65 which implies that 65% of the variances in e-WOM are accounted for by the Four IVs plus the CE. Similarly, the R square at BIL is 0.59 and this implies that 59% of the variances in BIL are accounted for by the four IVS and the CE. Except for three effects/paths: INV → CE, ENT → BIL and ATT → BIL, all the other eleven effects/paths are statistically significant. We proceed to the moderated mediation model.

Figure 2b is the moderated mediation model of our study showing the four IVs, the Mediator, the two DVs and the moderator. With the inclusion of the moderator variable in the model, the R square at CE improved from 0.55 to 0.59, that of e-WOM decreased from 0.65 to 0.58 while that at BIL improved significantly from 0.59 to 0.77. This on the average show R square increase of 0.05 or 25% increase on the total variances explained. The implication of this is that the addition of the moderating variable to our model is justified as it enhanced variances explained. We proceed to test and validate our hypotheses using the effects/paths in the moderated mediation model.

Twenty-nine of the forty-four hypothesised relationships are statistically significant (Table 3). We assess the hypothesised effects under four groups: direct effects, moderated direct effects, indirect effects and moderated indirect effects/moderated mediation. Direct Effects. The effects: INV → CE (β = 0.097, and p-value = 0.03); ENT → CE (β =0.178, p-value = <0.001); ATT → CE (β = 0.487, p-value = <0.001) and ABS→CE (β = 0.095, p-value = 0.037) are statistically significant thus, H1a, H4a, H7a and H10a are supported respectively. The paths INV → e-WOM (β = −0.174, p-value = 0.001); ENT → e-WOM (β = 0.259, p-value = <0.001); ATT → e-WOM (β = 0.214, p-value = <0.001) and ABS→e-WOM (β = 0.327, p-value = <0.001) are all statistically significant. Based on these: H2a, H5a, H8a and H11a are validated respectively. The paths: INV → BIL (β = 0.110, p-value=,0.019); ENT → BIL (β = 0.106, p-value = 0.022) and ABS→BIL (β = 0.420, p-value = <0.001) are all statistically significant. With these, H3a, H6a, and H12a are validated respectively. The path ATT → BIL (β = −0.011, p-value = 0.415) is not significant thus, H9a is not supported. The paths from CE → e-WOM (β = 0.455, p-value = 0.001) and CE → BIL (β = 0.455, p-value = 0.001) are statistically significant and based these, H13a and H14a are supported. The Effect sizes for this group range from high effect sizes for ATT → CE = 0.352, ABS→BIL = 0.276 and CE → e-WOM = 0.263 while others show moderate to low effect sizes. Moderated Direct Effects. The following paths: CRM*INV → CE (β = 0.200, p-value = <0.001); CRM*ENT → CE (β = 0.201, p-value = <0.001); and CRM*ATT → CE (β = 0.200, p-value = <0.001) are statistically significant and validate H1b, H4b, and H7b respectively while CRM*ABS→CE (β = −0.005, p-value = 0.464) is not significant and thus, does not support H10b. The moderated effects: CRM*INV → e-WOM (β = 0.137, p-value = 0.005); CRM*ATT → e-WOM (β = 0.178, p-value = <0.001); and CRM*ABS→e-WOM (β = 0.166, p-value = <0.001) support H2b, H8b and H11b respectively while CRM*ENT → e-WOM (β = 0.053, p-value = 0.160) is not significant and does not support H5b. The moderated relationship CRM*INV → BIL (β = 0.282, p-value = <0.001) and CRM*ATT → BIL (β = 0.294, p-value = <0.001) support H3b and H9b respectively while CRM*ENT → BIL (β = −0.041, p-value = 0.219) does not support Hb. Also, CRM*ABS→BIL (β = −0.084, p-value = 0.057) does not support H11b. The moderated relationship CRM*CE → e-WOM (β = −0.069, p-value = 0.0.098) does not support H13b while CRM*CE → BIL (β = 0.123, p-value = 0.010) support H14b. All the constructs here show moderate to low effect sizes. It is also important to note that while five of the moderated direct show notable increases in their coefficients, others either show no serious improvement or had decreased coefficients. Indirect Effects. The hypothesised indirect relationship, INV → CE → e-WOM (β = −0.044, p-value = 0.012) is statistically significant and supports H2c; while INV → CE → BIL (β = 0.019, p-value = 0.313) is not significant, hence H3c is not supported. The hypothesised indirect effect, ENT → CE → e-WOM (β = 0.081, p-value = 0.015) supports H5c; while ENT → CE → BIL (β = 0.036, p-value = 0.171) does not support H6c. The indirect paths: ATT → CE → e-WOM (β = 0.222, p-value = <0.001) and ATT → CE → BIL (β = 0.098, p-value = 0.005) support H8c and H9c respectively. Indirect effects: ABS→CE → e-WOM (β = 0.043, p-value = 0.126) and ABS→CE → BIL (β = 0.019, p-value = 0.307) are not statistically significant thus H11c and H12c are not supported. All the constructs here show moderate to low effect sizes. Moderated Indirect Effects. The moderated CRM*INV → CE → e-WOM (β = 0.091, p-value = 0.008) is significant and supports H2d while CRM*INV → CE → BIL (β = 0.040, p-value = 0.143) is not significant and H3d is therefore not supported. The part, CRM*ENT → CE → e-WOM (β = 0.092, p-value = 0.007) is statistically significant and supports H5d while CRM*ENT → CE → BIL (β = −0.040, p-value = 0.141) is not significant, thus H6d is not supported. The moderated indirect relationship CRM*ATT → CE → e-WOM (β = 0.091, p-value = 0.008) is significant, thus H9d is supported. The parts: CRM*ATT → CE → BIL (β = −0.040, p-value = 0.143); CRM*ABS→CE → e-WOM (β = −0.002, p-value = 0.477); and CRM*ABS→CE → BIL (β = −0.001, p-value = 0.490) are not statistically significant, thus H9d, H11d and H12d are respectively not validated. Six of the eight items in this category show low effect sizes while two have no effect sizes. In terms of the coefficients, the first three items show notable increases in their coefficients while the others show decreases with respect to their coefficients.

S/No	Paths (direct effects)	B	SE	Effect sizes	p-values	Decision
1.	INV → CE	0.097	0.053	0.030	0.033	Supported
2.	ENT → CE	0.178	0.052	0.085	<0.001	Supported
3.	ATT → CE	0.487	0.050	0.352	<0.001	Supported
4.	ABS→CE	0.095	0.053	0.054	0.037	Supported
5.	INV → e-WOM	−0.174	0.052	0.055	<0.001	Supported
6.	ENT → e-WOM	0.259	0.051	0.077	<0.001	Supported
7.	ATT → e-WOM	0.214	0.052	0.135	<0.001	Supported
8.	ABS→e-WOM	0.327	0.051	0.194	<0.001	Supported
9.	INV → BIL	0.110	0.053	0.062	0.019	Supported
10.	ENT → BIL	0.106	0.053	0.059	0.022	Supported
11.	ATT → BIL	−0.011	0.053	0.006	0.415	Not Supported
12.	ABS→BIL	0.420	0.050	0.276	<0.001	Supported
13.	CE → e-WOM	0.455	0.050	0.263	<0.001	Supported
14.	CE → BIL	0.201	0.052	0.110	<0.001	Supported
15.	CRM*INV → CE	0.200	0.052	0.109	<0.001	Supported
16.	CRM*ENT → CE	0.201	0.052	0.086	<0.001	Supported
17.	CRM*ATT → CE	0.200	0.052	0.045	<0.001	Supported
18.	CRM*ABS→CE	−0.005	0.053	0.001	0.464	Not Supported
19.	CRM*INV → e-WOM	0.137	0.052	0.034	0.005	Supported
20.	CRM*ENT → e-WOM	0.053	0.053	0.013	0.160	Not Supported
21.	CRM*ATT → e-WOM	0.178	0.052	0.009	<0.001	Supported
22.	CRM*ABS→e-WOM	0.166	0.052	0.015	<0.001	Supported
23.	CRM*INV → BIL	0.282	0.051	0.127	<0.001	Supported
24.	CRM*ENT → BIL	−0.041	0.053	0.019	0.219	Not Supported
25.	CRM*ATT → BIL	0.294	0.051	0.113	<0.001	Supported
26.	CRM*ABS→BIL	−0.084	0.053	0.022	0.057	Not Supported
27.	CRM*CE → e-WOM	−0.069	0.053	0.017	0.098	Not Supported
28.	CRM*CE → BIL	0.123	0.053	0.037	0.010	Supported
Indirect Effects
29.	INV → CE → e-WOM	0.044	0.038	0.014	0.012	Supported
30.	INV → CE → BIL	0.019	0.038	0.011	0.303	Not Supported
31.	ENT → CE → e-WOM	0.081	0.037	0.024	0.015	Supported
32.	ENT → CE → BIL	0.036	0.038	0.020	0.171	Not Supported
33.	ATT → CE → e-WOM	0.222	0.037	0.140	<0.001	Supported
34.	ATT → CE → BIL	0.098	0.037	0.052	0.005	Supported
35.	ABS→CE → e-WOM	0.043	0.038	0.026	0.126	Not Supported
36.	ABS→CE → BIL	0.019	0.038	0.012	0.307	Not Supported
37.	CRM*INV → CE → e-WOM	0.091	0.037	0.023	0.008	Supported
38.	CRM*INV → CE → BIL	0.040	0.038	0.018	0.143	Not Supported
39.	CRM*ENT → CE → e-WOM	0.092	0.037	0.022	0.007	Supported
40.	CRM*ENT → CE → BIL	−0.040	0.038	0.018	0.141	Not Supported
41.	CRM*ATT → CE → e-WOM	0.091	0.037	0.004	0.008	Supported
42.	CRM*ATT → CE → BIL	−0.-040	0.038	0.015	0.143	Not Supported
43.	CRM*ABS→CE → e-WOM	−0.002	0.038	0.000	0.477	Not Supported
44.	CRM*ABS→CE → BIL	−0.001	0.038	0.000	0.490	Not Supported

Table 3.

Assessing the structural model.

6.1. Robustness check

The hypothesised relationships: direct, moderated direct, indirect and moderated indirect effects were tested and analysed using WarpPLS software version 7 and the algorithm used was the PLS Regression. To check the robustness of the findings, we reanalysed the data using the robustness algorithm and the results were not different from the original analysis, thus confirming that our results are robust.

7. Conclusions and implications

This chapter is on CE behaviour in the hospitality industry, a sector that was hard hit by the COVID-19 pandemic. Vivek et al. [34] believe that CE has hypothetical roots within the extended domain of relationship marketing that emphasise the notions of interactivity and customer experience. The imperative of researching and studying CE as noted in the literature is that it helps companies especially multinationals in designing new products and in engaging and managing old and new customers. Pansari and Kumar [3] designed a model which they tested and validated in the US and affirmed that companies around the world could benefit immensely from CE researches. In the present study which is on antecedents and consequences of CE in the hospitality sector, we reviewed literature from the CE and CRM noting as evidenced in the literature that engagement can only take place when relationship(s) have been established. Hence, we proposed a conceptual model with involvement, enthusiasm, attention and absorption as IVs and antecedents; CE as mediator variable and outcomes: e-WOM and BIL as DVs, while CRM is a moderating construct. The import of this is based on the maxim that CE has roots within the extended domain of relationship marketing that emphasise the notions of interactivity and customer experience [34]. The model also investigated the moderating effects of CRM on the direct and indirect effects of the antecedents on e-WOM as well as BIL. The implication of this is that the hospitality industry has some peculiar characteristics that include two-way communication and relationship building. The model was tested with WarpPLS version 7 and we collected from 350 respondents majorly from different southern of Nigeria. We first ran the analysis without the moderator and second with the CRM moderating variable. With the inclusion of the moderator variable in the model, the R square at CE improved from 0.55 to 0.59, that of e-WOM decreased from 0.65 to 0.58 while that at BIL improved significantly from 0.59 to 0.77. That is to say that R square increased by 4% at CE; decreased by 7% at the e-WOM and increased by 18% at BIL. This on the average show R square increase of 0.05 or 5% increase on the average of the variances explained. Some of the direct, direct-moderated, indirect and moderated indirect effects show noticeable increases in their coefficients while others show decrease but on the whole CRM moderate CE which mediate the antecedents and consequences CE. The implication of this is that the addition of the moderating variable to our model is justified as it enhanced variances explained. Mintz [5] emphasises the need for businesses to map out strategies to win back customers post COVID-19 as consumers have shifted their focus from a hedonic, enjoyment-focused to a utilitarian, goals-based consumption procedure. The implication of this is that in the process of managing customers, marketers need to urgently deploy new strategies to better engage their everyday exchanges with their consumers’ new behaviour. Mintz [5] proposed a COUNTER COVID framework for engaging firms must follow in addressing their customer’s new behaviour: that marketers need to create emotional connections with their customers; firms should demonstrate their value to their customers; and should expand their digital footprints to better reach their customers. Firms in the hospitality sector should increase customer trial and retention, and engage more effective and efficient digital methods. To engage customers meaningfully, operators of hospitality sector businesses need to maintain strong websites and employ the various social media channels to build relationships with customers and at the time engage them in this post COVID-19 era giving the changing consumer behaviours occasioned by the pandemic. There is also the need to address the security concerns associated with deadly pandemic even as it has subsided. This work is limited to the hospitality sector and could be repeated in other sectors as it should serve as spring board for further studies in the emergent customer management phenomenon. This moderated mediation is contribution to the literature on engagement marketing and will spur other researchers in this important area of customer management as well as the hospitality sector.