Beyond immersion: how metaverse customer experience shapes users’ well-being perceptions and sustained engagement

The metaverse represents a significant evolution in digital consumer experience, with virtual and augmented realities converging to create immersive, interactive environments that reshape how consumers engage with content, communities and commerce (Yoon et al., 2025). Unlike earlier digital platforms, metaverse ecosystems are persistent and socially embedded, enabling users to actively co-create value through participation, identity expression and resource integration rather than passively consuming branded content (Rahman et al., 2025a, 2025b). This shift carries substantial economic benefits. Industry projections estimate the global metaverse market will reach USD 10,808 billion by 2034, signalling significant brand and industry investment in immersive metaverse ecosystems (Fortune Business Insights, 2026). These technological and economic developments reinforce the long-term managerial relevance of the metaverse and highlight the importance of understanding how co-created immersive experiences influence consumer behaviour and sustained engagement (Spais and Jain, 2025).

In parallel with the rapid expansion of the metaverse, scholars have increasingly examined how consumers perceive and construct experiences within immersive digital environments (Quach et al., 2022). Unlike earlier immersive applications, the metaverse is characterised by persistent and socially embedded environments, where identity-shaping interactions enable users to actively co-create experiences through resource integration and self-expression within communities (Dutta and Singh, 2025; Spais and Jain, 2025). In the same vein, an emerging stream of research suggests that metaverse experiences extend beyond entertainment and may influence users’ broader well-being, e.g. quality of users’ social, educational, leisure and psychological lives (Bowden et al., 2025; Lombart et al., 2025; Mansoor et al., 2024; Pillay, 2023; Rafi et al., 2024; Riches et al., 2024). This focus aligns with contemporary marketing thought, which emphasises the co-creation of value benefiting organisations, customers and society at large (Mende et al., 2024). Within this perspective, customer safety and well-being are increasingly viewed as potential sources of competitive advantage (Deloitte, 2025).

Because metaverse interactions are continuous and embedded in users’ routines and relationships, their consequences may extend beyond momentary affect, influencing how individuals evaluate the quality of their social, educational, leisure and psychological lives (Bowden et al., 2025). Despite the growing recognition that immersive environments may influence users’ broader life evaluations, an empirical understanding of how metaverse customer experience (MVCX) translates into well-being remains limited. Existing studies either examine isolated immersive technologies such as augmented reality (AR) or virtual reality (VR) (Pala et al., 2022) or rely primarily on conceptual reasoning rather than robust multi-study empirical validation (Hajian et al., 2024). More critically, prior research has not offered a context-specific operationalisation that captures how persistent, co-created metaverse experiences contribute to users’ perceived well-being across salient life domains. As a result, the mechanism through which MVCX shapes sustained metaverse platform engagement remains under-specified. If consumers’ well-being perceptions represent a strategic source of value and competitive advantage in contemporary business ecosystems (Mende et al., 2024; Rahman et al., 2026), then understanding whether and how MVCX enhances perceived life-domain value becomes essential for strategising long-term engagement and platform sustainability. Accordingly, this study addresses the following overarching research questions:

Addressing these questions requires a construct that captures the evaluative life-domain consequences of metaverse experiences; the literature currently lacks such a construct. To this end, we introduce perceived metaverse-enabled well-being, defined as users’ evaluative judgments about metaverse engagement and the extent to which it enhances certain important domains of their lives. By positioning perceived metaverse-enabled well-being as a central mechanism linking MVCX to sustained metaverse engagement (e.g. usage intentions), this study clarifies how immersive, co-creative metaverse experiences translate into managerially important behavioural outcomes. Three empirical studies are conducted to operationalise this construct, test its role within the MVCX–engagement relationship and examine nonlinear and contingent effects that shape sustained metaverse participation.

By addressing these questions, this study contributes to both theory and practice. Theoretically, it bridges immersive technology research with customer experience and well-being scholarship by demonstrating how MVCX extends beyond entertainment to shape users’ broader life evaluations. By operationalising perceived metaverse-enabled well-being as a domain-specific and context-attributed construct, this research clarifies the mechanism through which immersive, co-creative digital experiences translate into sustained engagement. The findings further extend customer experience research by identifying nonlinear dynamics and necessary condition effects within immersive ecosystems, offering a more nuanced understanding of how experiential value accumulates and influences behavioural intentions. From a managerial perspective, this study provides an easy-to-administer measurement tool that enables firms to assess whether their metaverse initiatives are generating meaningful life-domain value rather than short-term engagement metrics. The results suggest that sustained platform participation depends on experiential design and, critically, on whether users perceive the participation experience as meaningfully contributing to their social, educational, leisure and psychological well-being. As metaverse platforms become increasingly competitive, understanding these well-being mechanisms is critical for long-term retention, ethical technology development and strategic investment decisions.

The remainder of this paper is structured as follows. Section 2 reviews the literature on the effect of immersive technologies on consumer well-being. Section 3 introduces the theoretical framework for this study, grounded in service-domain (S-D) logic, conceptualises the new perceived metaverse-enabled well-being construct and develops the hypotheses. Section 4 outlines the multi-study research design and reports the findings. Study 1a (n = 24) uses qualitative insights to complement the literature review and inform the operationalisation of the perceived metaverse-enabled well-being construct. Study 1 b (n = 112) quantitatively refines the measurement of perceived metaverse-enabled well-being. Study 2 (n = 277) first validates the perceived metaverse-enabled well-being measure empirically using the new data set and then tests the proposed linear relationships among MVCX, perceived metaverse-enabled well-being and metaverse usage intentions. Study 3 (n = 494) further confirms the robustness of the perceived metaverse-enabled well-being measurement empirically using another independent data set and examines nonlinear effects and contingent relationships. Using quantitative studies, the new measure of perceived metaverse-enabled well-being is operationalised and validated across three data sets to establish its reliability, construct validity and nomological validity. Section 5 discusses the theoretical and practical contributions of the findings and directions for future research.

An initial literature review was conducted to develop a comprehensive understanding of the effect of immersive technologies on user well-being. To enhance the study’s transparency, replicability and rigour, the literature review followed a search protocol (Page et al., 2021). Therefore, a structured, multi-stage approach was adopted, including:

• formulation of research objectives,

• development of inclusion and exclusion criteria,

• development of search strategy, and

• synthesis of findings (see Supporting information W1 for study selection flowchart).

We defined well-being labels (“well-being”, “wellbeing”, “life satisfaction” and “quality of life”) within the immersive technology landscape. Several search strings were tested, and their results critically evaluated to ensure they captured the most relevant studies. Once the optimal string was identified, inclusion and exclusion criteria were applied. For this study, the following search string was applied to titles, abstracts and keywords in Scopus: (“well-being” OR “wellbeing” OR “quality of life” OR “life satisfaction”) AND (“metaverse” OR “virtual reality” OR “augmented reality” OR “mixed reality” OR “immersive technolog*”). We limited our search to English-language journal articles published in business, psychology or social science journals. This search resulted in 807 documents. Following Bhardwaj and Kalro (2024) for the screening process, we then selected those articles published in journals ranked A or A* on the Australian Business Deans Council (ABDC) journal quality list, resulting in 103 papers. Two raters independently assessed these articles to ensure that their primary focus was well-being and immersive technologies, resulting in 59 articles (see Supporting information Table W1.1 for inclusion and exclusion criteria). Table 1 summarises the key studies.

In the analysis and synthesis stage, 59 selected articles were reviewed to enhance an understanding of the effects of immersive technologies on user well-being, identify research gaps and further refine the relevant concepts and variables. Examining different metaverse typologies, various contexts and diverse theoretical backgrounds provided insights to position the research and define the study’s objectives. The literature review suggested that immersive technologies can be categorised along either the augmentation–simulation or the external–intimate spectrum, resulting in six metaverse typologies: the AR cloud, AR, lifelogging, mirror worlds, multiuser virtual environments and virtual worlds (Alimamy and Jung, 2025). Several scholars (e.g. Javornik et al., 2022; van Brakel et al., 2023) have addressed the potential of immersive technologies to enhance user well-being, often through unique experiences that meet users’ social, emotional and self-actualisation needs. For example, immersive virtual tourism experiences may enhance the psychological well-being of individuals who face social interaction and travel constraints (Zhang and Xiao, 2024). In health care, meaningful human–technology interactions can improve user well-being through value co-creation and enhanced quality of user experiences (Dodds et al., 2022). The use of AR and VR to improve accessibility to entertainment, sociocultural information and social interactions has been found to boost the social and personal well-being of senior citizens (Fernández et al., 2017). Moreover, immersive consumption experiences have been found to influence consumer well-being through dimensions such as pleasure, meaning and life satisfaction (Zarantonello and Schmitt, 2023).

To theorise about the relationship between immersive technologies and consumer well-being, scholars draw on self-determination theory (e.g. Park and Kim, 2025; Zhang and Xiao, 2024), self-concept theory (e.g. Ambika et al., 2023), self-expansion theory (e.g. Yoon et al., 2025), theory of embodied cognition (e.g. Alimamy and Jung, 2025) and uses and gratification theory (Javornik et al., 2022). Self-determination theory posits that humans are driven by their need for competence, relatedness and autonomy, all of which play a role in healthy development and overall well-being (Deci and Ryan, 2000). Accordingly, immersive metaverse experiences can meet users’ needs for positive mood management, self-expansion and connectedness (Choi et al., 2023). Self-concept theory suggests that engaging in virtual immersive technologies boosts consumers’ self-concept, enhancing their well-being and enabling them to make better life decisions (Javornik et al., 2022). The theory of embodied cognition posits that an individual’s cognitive processes are fundamentally connected to the body’s engagement with the environment (Barsalou, 2008). Drawing on this theory, Alimamy and Jung (2025) argue that engaging in immersive technologies, specifically the AR cloud, enhances consumer well-being by creating immersive, multisensory experiences that engage users mentally and physically. Finally, uses and gratifications theory posits that an individual’s motivation to use an immersive technology such as AR (e.g. self-expression and creative engagement) mediates their well-being outcomes, with positive effects observed when technology aligns with the individual’s ideal self-presentation (Javornik et al., 2022).

Given its unique characteristics, including its ability to provide enhanced realism and presence, the metaverse offers a promising platform to fulfil users’ social and psychological needs (Dwivedi, Hughes et al., 2023). However, the existing empirical research has focused more on the well-being outcomes of AR and VR than the metaverse (Eshaghi et al., 2023). Unlike standalone AR or VR applications, metaverse platforms offer persistent, shared and highly personalised virtual environments that facilitate novel social interactions, potentially improving user well-being (Hajian et al., 2024). These platforms incorporate multiple immersive technologies with novel elements such as digital ownership, virtual identity construction and continuous social presence (Quach et al., 2022). The feature that sets the metaverse apart from other digital platforms is its ability to offer users a parallel virtual society in which they can make social connections and participate in social, economic and experiential activities, enhancing value co-creation and users’ hedonic and eudemonic well-being (Alimamy and Jung, 2025; Hadi et al., 2024). For example, Lim et al. (2024) found that individuals who feel marginalised at work, such as women and people of colour, are more willing to pay for avatar customisation in metaverse workplaces, which potentially enhances their well-being.

The academic discourse on the effect of the metaverse on user well-being is mostly conceptual (Hajian et al., 2024) rather than empirical (Lim et al., 2024). Thus, there is an urgent need to empirically test how the metaverse may enhance user experience and well-being (Saleh, 2024). Further, concrete evidence is needed to guide practitioners in effectively using metaverse technologies to support user well-being (Capatina et al., 2024). At present, this research gap hinders the ability of managers and regulators to make precise, evidence-based decisions about metaverse implementation in different contexts. Thus, new examinations are needed to measure the effect of metaverse experiences on major aspects of user well-being (Jung et al., 2024).

As a theoretical framework, S-D logic (Vargo and Lusch, 2008) provides a robust foundation for understanding how immersive digital platforms, such as the metaverse, generate value through collaborative processes rather than embedded product/service features. Central to S-D logic is the premise that value is co-created through resource integration and is realised as value-in-use within specific contexts. Value does not reside in the platform itself; rather, it emerges when multiple actors integrate their operant resources, such as skills, knowledge, creativity and social connections, within a shared service ecosystem (Rivière et al., 2024).

In the metaverse context, this ecosystem comprises users, platform providers, brands and other participants who collectively shape virtual environments and interactions (Kowalkowski et al., 2024). Through customisable avatars, digital asset creation, immersive environments and interactive social spaces, these actors pool and integrate resources to generate shared experiences and produce personalised value outcomes (Hadi et al., 2024; Mansoor et al., 2024). Actors continuously integrate their resources through collaboration and customisation, thus deriving value from the ability of the metaverse ecosystem to facilitate meaningful engagement, enable resource liquefaction and develop service systems that evolve with user participation. The metaverse, therefore, represents a particularly salient context for S-D logic because its value proposition depends on this continuous interaction, co-creation and user participation rather than passive consumption.

Prior research grounded in S-D logic suggests that when actors successfully integrate resources and co-create meaningful experiences, the resulting value-in-use may extend beyond transactional outcomes and contribute to broader well-being and quality of life (van Brakel et al., 2023; Zarantonello and Schmitt, 2023). In immersive digital ecosystems, where platform engagement is phenomenologically determined and becomes embedded within users’ social, educational and leisure activities, co-created value may shape how users evaluate the platform’s contribution to important life domains. Thus, well-being outcomes depend on how users experience their participation within the metaverse. Building on this, we apply S-D logic to examine how MVCX functions as a key mechanism through which resource integration and value co-creation influence users’ perceived metaverse-enabled well-being and subsequent behavioural intentions.

CX is the product of a customer’s direct and indirect interactions with a product or service provider and other users at various touchpoints in a digital platform (Lemon and Verhoef, 2016). This dynamic interplay between users, the platform and the broader shared experience generates value for firms and customers (Vargo and Lusch, 2008). Context-specific CX involves customers evaluating the key attributes of a context and aggregating these evaluations to form an overall perception of their CX (Parasuraman et al., 1988; Rahman, Carlson, Gudergan et al., 2022b). In the context of the metaverse, the platform facilitates interactions and enables users to use their own resources (e.g. time, creativity and social connections) to co-create meaningful experiences (Scholz and Duffy, 2018). In recent work, Rahman et al. (2025a, 2025b) have operationalised MVCX as 10 key attributes: aesthetics (visual appeal), communality (connection to the virtual community), commerciality (vibrant commercial activities), creativity (enabling users to create content using virtual tools), efficiency (ease of use), immersion (sense of being in a virtual environment), interoperability (seamless transition across virtual functionalities), personalisation (user friendliness of customisation tools), privacy (trust in the platform to protect user data) and trialability (the opportunity to try new technologies, applications or products before purchasing).

While customer experience research has traditionally focused on value creation and behavioural outcomes, an important question concerns how such experiences influence users’ broader quality of life (Bowden et al., 2025; Alzahrani et al., 2026; Rahman et al., 2026). Well-being refers to individuals’ evaluative assessment of the quality of their lives and the extent to which their experiences support effective functioning and goal attainment (Diener, 1984). Within psychological research, subjective well-being captures individuals’ judgments about their lives, encompassing both overall life satisfaction and evaluations of specific life domains (Diener, 1984). Individuals form well-being assessments by reflecting on how their daily activities, relationships and personal development contribute to their broader sense of quality of life (Alzahrani et al., 2026). As such, subjective well-being can be examined at both the global and domain-specific levels (Kahn and Juster, 2002).

A global assessment reflects individuals’ overall judgment of their life as a whole, whereas domain-specific evaluations focus on particular areas of life, such as work, social relationships, leisure, family life or psychological functioning. The OECD (2025) similarly recognises that domain-level evaluations provide meaningful diagnostic insight into how specific life areas contribute to overall life assessments, particularly when researchers seek to understand the role of particular contexts or activities in shaping well-being. These domain evaluations represent meaningful components of overall life assessment because individuals form broader judgments about their well-being based on how well important areas of their lives are functioning and supporting their goals. In addition, well-being evaluations may be context-attributed, meaning that individuals assess the extent to which a particular environment, activity or platform contributes to the quality of specific life domains (Diener, 1984; Kahn and Juster, 2002; OECD, 2025).

Building on this perspective, we conceptualise perceived metaverse-enabled well-being as a domain-specific and context-attributed form of subjective well-being. Specifically, it reflects the extent to which users evaluate the metaverse as contributing positively to important domains of their quality of life. Rather than capturing global life satisfaction or transient emotional states, our literature review suggests that the proposed construct focuses on users’ evaluative judgments regarding how metaverse engagement enhances salient areas of their real lives, such as daily routines, educational or professional development, social relationships, leisure experiences, family life and psychological functioning (Yoo et al., 2023). Furthermore, in immersive digital environments, platforms can become embedded within users’ everyday practices, shaping how individuals organise their time, connect with others, pursue goals and express identity (Dwivedi, Hughes et al., 2023). Consequently, perceived metaverse-enabled well-being captures users’ assessment of the metaverse as an enabling context that supports functioning and quality of life across meaningful domains, rather than merely reflecting perceived utility or instrumental benefits.

Given that perceived metaverse-enabled well-being reflects users’ evaluative judgments regarding how the metaverse enhances important domains of their lives, it is critical to examine the mechanisms through which such evaluations emerge. In the metaverse context, these evaluations are shaped by the experiential attributes that structure users’ interactions with the platform. Specific MVCX attributes provide the operant resources that enable users to integrate their own skills, time, creativity and social connections within the virtual environment. For instance, aesthetics and immersion deepen users’ engagement with the environment, supporting meaningful leisure experiences and psychological involvement (Alimamy and Jung, 2025). Personalisation and creativity allow users to customise avatars and generate content, adopting self-expression and active participation (Harmeling et al., 2017). Communality facilitates social interaction and the development of virtual communities, strengthening social connectedness and shared experiences (Mansoor et al., 2024). Interoperability and efficiency reduce friction in navigating the platform, enabling users to seamlessly integrate metaverse activities into their routines and educational or professional pursuits (Dodds et al., 2022; Hajian et al., 2024).

From an S-D logic perspective, value is not embedded in platform features themselves but emerges through value-in-use, that is, through users’ integration of platform-provided resources with their own operant resources, such as skills, time, creativity and social capital (Vargo and Lusch, 2008). In the metaverse, experiential attributes such as immersion, personalisation, communality and efficiency function as enabling structures that facilitate this resource integration. For example, on platforms such as Roblox, users leverage immersive design tools and customisable avatars to create virtual experiences, games or social spaces, integrating their creativity and technical skills with platform resources. Similarly, in blockchain-based metaverse environments, such as Sandbox, users participate in virtual land development, digital asset creation and community governance, combining platform affordances with entrepreneurial skills and social networks. When these attributes align with users’ personal goals and life contexts, such as learning and enjoyment, they support meaningful engagement rather than passive consumption. Through repeated interactions, users co-create value by embedding metaverse experiences into their routines, relationships, learning pursuits and identity-related activities, transforming platform engagement from isolated digital interaction into a meaningful component of everyday life (Dwivedi, Hughes et al., 2023; Mansoor et al., 2024).

Importantly, value co-creation extends beyond immediate experiential enjoyment and may shape how users evaluate the metaverse’s contribution to broader life domains (Bowden et al., 2025). When platform attributes enable users to achieve personal goals, strengthen social connectedness, express identity and manage everyday tasks more effectively, the benefits of engagement become embedded within meaningful aspects of users’ real lives (McLean et al., 2023; Rafi et al., 2024). For example, users who use immersive environments and interactive tools for collaborative learning may perceive that the metaverse enhances the quality of their educational life. Those who participate in virtual communities or shared events may evaluate the platform as strengthening their social well-being. Similarly, participation in virtual leisure activities, creative expression or gaming experiences may contribute to users’ assessments of their leisure well-being (Rafi et al., 2024). For some users, shared metaverse experiences with family members or friends may be perceived as enriching family life. Furthermore, immersive engagement, identity exploration and opportunities for self-expression may be reflected in users’ evaluations of their psychological well-being. Therefore, it is possible that value co-creation in the metaverse becomes translated into evaluative judgments regarding domain-specific well-being (Dwivedi, Hughes et al., 2023). Accordingly, a positive MVCX, characterised by attributes that facilitate effective resource integration and sustained engagement, should be associated with higher perceived metaverse-enabled well-being. Therefore, we posit the following:

Within immersive digital environments, users’ continued participation depends on whether prior engagement has generated meaningful value-in-use (Bowden et al., 2025). Research in VR contexts suggests that higher levels of user well-being are associated with stronger intentions to remain active on digital platforms (Kim et al., 2023; McLean et al., 2023). From an S-D logic perspective, when users perceive that their resource integration efforts have resulted in valuable outcomes, they are motivated to sustain and deepen their participation in the service ecosystem (Vargo and Lusch, 2014).

In the metaverse context, our literature review suggests that perceived metaverse-enabled well-being reflects users’ evaluative judgments that platform engagement enhances important life domains, such as education, social relationships, daily functioning, leisure, family life and psychological well-being. When users recognise that their creative, social or exploratory activities contribute positively to these domains, we suggest that engagement is no longer viewed as an isolated digital interaction but as a meaningful extension of their everyday lives. This perception signals successful value co-creation and goal attainment, reinforcing the desirability of continued platform involvement. Accordingly, users who perceive higher levels of metaverse-enabled well-being are more likely to intend to continue using the platform to sustain and further develop these beneficial outcomes. Therefore, we posit the following hypothesis:

Building on H1 and H2, S-D logic suggests that MVCX facilitates resource integration and value co-creation, which is subsequently reflected in users’ evaluative judgments of domain-specific well-being. When users perceive that their engagement enhances important areas of their lives, this value-in-use should strengthen their motivation to continue participating in the metaverse ecosystem (Bowden et al., 2025). Thus, perceived metaverse-enabled well-being represents a key mechanism through which MVCX translates into continued usage intentions. Accordingly, we hypothesise:

S-D logic conceptualises service ecosystems as dynamic, adaptive systems in which value co-creation emerges through interactions among heterogeneous actors operating under diverse institutional arrangements (Edvardsson et al., 2018; Vargo and Lusch, 2014). Within such ecosystems, value-in-use is realised through ongoing resource integration processes rather than discrete exchanges. As a result, the relationship between experiential inputs and outcomes may not be strictly proportional. Instead, co-creation processes may exhibit nonlinear dynamics, including threshold effects, accelerating returns or diminishing marginal effects, depending on how effectively actors integrate available resources (Rogers, 2003; Storbacka et al., 2016; Vargo and Lusch, 2014). Differences in users’ capabilities, motivations, contextual constraints and technological affordances may further influence how experiential attributes translate into perceived value (Rogers, 2003).

Empirical research in related digital contexts provides support for such nonlinear dynamics. For example, Nikhashemi et al. (2021) found that the positive association between CX and platform usage intensifies once users’ perceived benefits reach moderate to high levels. Similarly, studies have reported nonlinear relationships between consumer well-being and behavioural outcomes, including mobile internet and digital platform usage (Verduyn et al., 2022; Zhan and Zhou, 2018). These findings suggest that the effects of MVCX on perceived metaverse-enabled well-being and metaverse usage intentions may not follow a strictly linear pattern. Instead, incremental improvements in customer experience may generate disproportionately stronger effects once certain experiential or evaluative thresholds are reached within the metaverse ecosystem.

Although we do not posit a directional hypothesis regarding functional form, we empirically examine both linear (Study 2 and Study 3) and nonlinear effects (Study 3) of MVCX on perceived metaverse-enabled well-being and metaverse usage intentions.

Previously, we have argued that S-D logic and the threshold effect (Rogers, 2003) support a potential nonlinear relationship between MVCX, perceived metaverse-enabled well-being and metaverse usage intentions. According to S-D logic (Vargo and Lusch, 2008, p. 8), value is co-created through resource integration, where metaverse users actively shape their experience by leveraging features such as customisation tools, immersive environments and community interactions (Scholz and Duffy, 2018; Yoo et al., 2023). Initially, users are unfamiliar with the metaverse platform and their interactions with it lack continuity; thus, they are unable to fully use the platform’s resources (Rahman et al., 2025a, 2025b). Therefore, an improved MVCX may only lead to modest gains in well-being. However, once users reach a critical usage length (e.g. more than 1 year) (Bhattacherjee and Premkumar, 2004; Rogers, 2003), their ability to co-create value increases (Wilson and Gilbert, 2008), strengthening the relationship between MVCX, users’ perceived metaverse-enabled well-being and metaverse usage intentions. This perspective aligns with the diffusion of innovations theory (Rogers, 2003), which suggests that adopting a new innovation and engaging in it follows a nonlinear trajectory, in which adoption is initially gradual but accelerates once a critical threshold of familiarity or experience is reached.

It is important to distinguish between usage length and usage duration. Usage duration refers to the cumulative time spent on a platform in minutes or hours, typically measured as isolated sessions. In contrast, usage length captures the continuity of engagement measured in days, months or years. In this study, we focus on usage length. While duration reflects how much time users invest in short-term activities, length emphasises the depth and evolution of their relationship with the platform (Mansoor et al., 2024). Extended usage length allows users to build familiarity with the platform’s features, strengthen their connections with virtual communities and integrate the platform into their daily lives (Steenkamp and Baumgartner, 1992). As users engage with the metaverse over a long period (>1 year), it is likely that they will co-create increasingly meaningful and personalised experiences, potentially amplifying the positive effect of MVCX on their well-being and usage intentions. We position usage length as the key contingency factor influencing the association between MVCX, users’ perceived metaverse-enabled well-being and metaverse usage intentions. As metaverse users transition from casual exploration to deeply integrated use, attributes such as communality, creativity and personalisation will become more significant over time. Therefore, we hypothesise the following:

Figure 1 illustrates the theoretical framework tested in this research.

We conducted three empirical studies to operationalise and validate the perceived metaverse-enabled well-being construct and to test the proposed theoretical model. Study 1 had two complementary objectives. Firstly, Study 1a qualitatively explored how metaverse usage may contribute to users’ perceived metaverse-enabled well-being, thereby informing the conceptual domain and operationalisation of the construct. Secondly, Study 1 b (n = 112) quantitatively refined the measurement of perceived metaverse-enabled well-being and assessed its initial psychometric properties.

The validated measure was subsequently used in Studies 2 and 3. Study 2 (n = 277) first validated the perceived metaverse-enabled well-being measure empirically using a new data set, thereby confirming its reliability and construct validity in an independent sample. Study 2 then tested the proposed linear associations between MVCX, perceived metaverse-enabled well-being and metaverse usage intentions (H1, H2 and H3). Study 3 (n = 494) further confirmed the robustness of the perceived metaverse-enabled well-being measurement using another independent data set and extended the analysis by retesting H1–H3, as well as examining whether these associations are stronger for users who have engaged with a metaverse platform for more than 1 year (H4).

Prior to data collection, we were granted ethics approval along with appropriate protocols by a reputable university. For all three studies, we used an online panel provider to collect data from USA residents aged at least 18 years and who frequently used a metaverse platform. To ensure that respondents were actual metaverse users, we used a two-stage data collection process (Mansoor et al., 2024; Wessling et al., 2017). Firstly, to ensure the sample units aligned with the research objectives, we generated a pool of respondents who met the inclusion criteria (Aguinis et al., 2021) by asking them to describe their current engagement with any metaverse platform. To avoid potential bias, we did not provide a list of metaverse platforms (see  Appendix 1Table A1 for screening questions, an example of a participant’s response, and the sample characteristics). In the second stage, we distributed an online questionnaire (developed using Qualtrics) to our metaverse user pool. Each respondent participated in only one study. To avoid common method bias, we followed best practices for surveys, such as clear and simple language, randomised questions and having respondents answer only one question at a time (Rahman, Carlson, Gudergan et al., 2022b). Participants received compensation for their time and effort.

In line with the literature review, which identified that immersive technologies may enhance well-being across multiple life domains, including social connectedness, psychological functioning, learning and accessibility, leisure experiences and identity-related development, Study 1a (n = 24 metaverse users, 54% female) was conducted to validate and contextualise these domains within metaverse platforms. Rather than serving as the primary source of construct development, the qualitative study was designed to examine whether frequent metaverse users perceive similar domain-specific contributions to their quality of life and to ensure that the conceptualisation of perceived metaverse-enabled well-being reflects users’ lived experiences. The qualitative data were therefore assessed using a theory-informed thematic approach, whereby participants’ narratives were examined in relation to the life domains identified in the literature to evaluate their presence, relevance and contextual meaning within metaverse use.

Prominent themes that emerged included a sense of community, learning flexibility, work–life integration and stress relief. Respondents highlighted several ways in which the metaverse adds value to their lives, including enhancing their social connections and offering a means of relaxation and personal growth through creativity. For example, one respondent commented, “We can attend virtual classes in [the] metaverse, go to live events, or try out new hobbies from home. This makes learning more flexible and accessible. Plus, we can meet people with similar interests, building a sense of community” (26-year-old male Roblox user). Another noted, “The metaverse offers tons of ways to relax and unwind, it’s a fun escape from daily stress. It can even improve our mood and creativity, making it a great tool for self-care” (24-year-old female Roblox user). Users also shared how the metaverse helps them to maintain strong social connections: “The metaverse can make it easier to stay in touch with friends and family, even if they’re away, which can improve social life” (49-year-old female Sandbox user). Many respondents believed that the metaverse supported their mental and emotional health by offering them new ways to connect and try things they might not experience in real life: “It’s a fun way to try new things we might not get to do in real life. It really helps with relaxing and getting creative!” (24-year-old female Sandbox user). See Supporting information W2 for further examples of quotations.

These findings align with prior academic research captured in our literature review, demonstrating that immersive technologies may enhance well-being across multiple life domains. For instance, respondents’ emphasis on community and social connectedness is consistent with empirical evidence showing that social presence and virtual interactions strengthen perceived social support and subjective well-being (Rafi et al., 2024; van Brakel et al., 2023). Themes related to learning flexibility and accessibility resonate with research highlighting the educational and cognitive benefits of immersive platforms, including flipped learning and enhanced access to services (Dwivedi, Hughes et al., 2023; Hajian et al., 2024). Similarly, participants’ references to relaxation and stress relief, along with opportunities for creativity, align with findings that immersive consumption experiences contribute to pleasure, meaning and psychological well-being (Zarantonello and Schmitt, 2023; Zhang and Xiao, 2024). Reports of improved mood and self-care, as well as identity exploration, reflect theoretical perspectives grounded in self-determination theory, self-concept theory and embodied cognition. These perspectives suggest that immersive engagement can satisfy needs related to autonomy, relatedness and self-expression (Alimamy and Jung, 2025; Ambika et al., 2023; Deci and Ryan, 2000). These converging insights suggest that metaverse platforms may become embedded in users’ daily practices and meaningful life domains, influencing how users evaluate their quality of life. This pattern is also reflected in broader discussions highlighting how immersive platforms increasingly form part of users’ everyday routines and psychological, social and emotional well-being narratives (Global Wellness Institute, 2022; Tatavarti, 2022). These initial findings inform the development of a measure for the perceived metaverse-enabled well-being construct in Study 1 b.

To date, no specific measures of well-being in the context of the metaverse have been developed. To address this gap and capture aspects of well-being that may be influenced by immersive digital environments such as the metaverse, we have developed a measure for the perceived metaverse-enabled well-being construct conceptualised in this study. We base this measure on established well-being measures, our structured literature review and qualitative insights from Study 1a.

Subjective well-being constructs have been operationalised in multiple ways across different disciplines (OECD, 2025). Prior research has used single-item global life satisfaction measures, multi-item unidimensional measurements and multidimensional frameworks capturing hedonic and eudaimonic components (Diener, 1984; Diener et al., 1985; Kahn and Juster, 2002; OECD, 2025). For example, global life satisfaction has often been measured using concise multi-item scales such as the Satisfaction with Life Scale (Diener et al., 1985), while some scholars have proposed multidimensional frameworks distinguishing affective, cognitive and functional elements of well-being (Kahn and Juster, 2002). In emerging technology contexts, scholars have also adapted well-being measures to specific domains such as health care, work, tourism and digital platform usage (Zarantonello and Schmitt, 2023; Zhang and Xiao, 2024). These approaches illustrate that subjective well-being can be conceptualised either globally, specifically to a domain or through multidimensional structures, depending on theoretical focus and contextual maturity.

The metaverse ecosystem, however, is still in the early stages of evolution, with rapidly shifting functionalities, institutional arrangements and user practices (Dwivedi, Hughes et al., 2023; Rahman et al., 2025a, 2025b). In such emerging contexts, prematurely imposing a higher-order multidimensional structure may risk artificially fragmenting a construct whose boundaries and subdimensions are still fluid (Netemeyer et al., 2003). Prior metaverse research has highlighted the importance of parsimonious yet theoretically grounded measurement approaches when investigating novel experiential phenomena (Mansoor et al., 2024; Rahman et al., 2025a, 2025b). Consistent with this logic and drawing on guidance from scale, that is, construct measurement, development guidelines emphasising parsimony and conceptual clarity in early-stage construct operationalisation (Netemeyer et al., 2003), we decided to develop a parsimonious multi-item measure of perceived metaverse-enabled well-being rather than a fully multidimensional second-order structure.

A multi-item specification was appropriate for several reasons. Firstly, compared with single-item measures, multi-item measures reduce measurement error and provide greater reliability by capturing shared variance across indicators (Netemeyer et al., 2003). Secondly, our conceptualisation of perceived metaverse-enabled well-being is domain-specific and context-attributed, reflecting users’ evaluative judgments about how metaverse engagement contributes to salient life domains, such as daily routines, education, social relationships, leisure, family life and psychological functioning. Capturing this evaluative breadth requires multiple indicators to ensure adequate content validity while maintaining a coherent unidimensional structure. Thirdly, given that the metaverse remains an evolving service ecosystem, a parsimonious multi-item configuration allows empirical testing of the construct’s nomological role without over-specifying potentially unstable subdimensions (Netemeyer et al., 2003).

Accordingly, our measure builds on traditional subjective well-being theory (Diener, 1984; Diener et al., 1985; Kahn and Juster, 2002), incorporates domain-specific insights from immersive technology research (Zarantonello and Schmitt, 2023; Zhang and Xiao, 2024) and is informed by qualitative evidence from frequent metaverse users (Study 1a). The items were developed to capture the broad influence of metaverse engagement on daily life and personal development. Specifically, the initial conceptual domain included social well-being, quality of community life, psychological health (Program-Ace, 2023), family life, daily routines (VIVE Team, 2021), financial well-being (Abbott and Murray, 2022), educational advancement (FXMedia, 2023), leisure and physical well-being. Consistent with Netemeyer et al.’s (2003) recommendation that the initial pool for a multi-item construct should contain at least 8–10 items to ensure adequate content coverage, we developed an initial pool of 10 items that reflect these domains.

We distributed an online questionnaire with the initial pool of ten items to our initial respondent pool and collected data from 112 respondents (50% female; see demographic details in  Appendix 1Table A1). Based on Netemeyer et al.’s (2003) recommendation to obtain 10 responses per item, this sample size was considered adequate for exploratory factor analysis. The response scale for all items was based on a seven-point Likert scale (ranging from strongly agree to strongly disagree, with neither agree nor disagree as the midpoint). To establish the dimensionality and reliability of the perceived metaverse-enabled well-being construct, we conducted an exploratory factor analysis of the items in SPSS Statistics, using principal component analysis as the extraction method and varimax as the rotation method. All items were loaded into one factor, suggesting that the well-being construct was unidimensional. Following the recommended three to six items for reflectively measured constructs (Netemeyer et al., 2003), we retained six items with relatively higher loadings (0.71–0.79). The four excluded items were related to users’ physical well-being, quality of community life, quality of work life and financial well-being, respectively. A Cronbach’s alpha of 0.85 indicated that the reflectively measured well-being construct with its six measurement items was reliable. In addition, a “what if item is deleted” test indicated that removing any of these six items would reduce measurement reliability (Raykov, 2007). Therefore, our parsimonious six-item measure (see Table 2) sufficiently captures the core essence of customer well-being in the metaverse context; it was subsequently used in Studies 2 (validates the measure using a new data set) and 3 (confirms the measure with another new data set).

In Study 2 (n = 277; 44% female; see demographic details in  Appendix 1Table A1), we used Rahman et al.’s (2025a, 2025b) formative MVCX measure and Mansoor et al.’s (2024) metaverse usage intention measure (plus, the aforementioned perceived metaverse-enabled well-being measure; Study 1 b). All items were based on a seven-point Likert response scale (ranging from strongly agree to strongly disagree, with neither agree nor disagree as the midpoint).

To evaluate the hypothesised relationships, we applied partial least squares structural equation modelling (PLS-SEM) using SmartPLS 4 (Ringle et al., 2024). PLS-SEM is well suited to analysing complex models involving latent constructs (Hair et al., 2022), particularly when they are both reflective (e.g. perceived metaverse-enabled well-being and usage intentions) and formative (e.g. MVCX), because it inherently addresses measurement errors in multi-item constructs (Cheah et al., 2021; Ho et al., 2026). Further, SmartPLS enables sophisticated model analysis by incorporating mediators, moderators and quadratic (nonlinear) relationships, which were critical to this research. To ensure robustness, we applied advanced analytical features in SmartPLS such as necessary condition analysis (NCA) and importance-performance map analysis, both of which enhance the rigour of insights derived from model estimations (Hauff et al., 2024; Rahman et al., 2026). We ran PLS-SEM bootstrapping in SmartPLS with 10,000 subsamples, a 95% bias-corrected bootstrap confidence interval (CI) method, two-tailed tests, fixed seed and path weighting scheme (Hair et al., 2022).

Before testing the path estimates, we checked the validity and reliability of the constructs in the research model. For the reflective construct of perceived metaverse-enabled well-being, indicator loadings ranged from 0.68–0.77 (Table 2), Cronbach’s alpha was 0.83, composite reliability was 0.87 and average variance extracted (AVE) was 0.54 (Table 4), exceeding recommended thresholds (Hair et al., 2022; Netemeyer et al., 2003). Similarly, the reflective construct of metaverse usage intentions showed adequate reliability and convergent validity (α = 0.78, CR = 0.87, AVE = 0.69). For the formatively specified MVCX construct, the indicators exhibited meaningful variation in their outer weights (ranging from 0.07–0.22), while outer loadings ranged from 0.62–0.74. Multicollinearity diagnostics indicated no concerns, with VIF values between 1.48 and 1.73 (<3.0; Kock, 2015). Consistent with formative measurement logic (Hair et al., 2024), the variation in weights, adequate absolute contributions reflected in the loadings and low VIF values support the specification of MVCX as a formative construct in Study 2. A heterotrait–monotrait ratio of less than 0.9 supported the discriminant validity of constructs (Henseler et al., 2015; see Table 3). Overall, the measurement model for Study 2 demonstrated satisfactory reliability and validity, and we continued to the subsequent hypothesis testing.

Table 6 presents the results for the main hypotheses (H1 and H2), including the standardised estimates (β), CIs, t-values and p-values. At a 5% significance level, we observed a significant positive relationship between MVCX and well-being (β = 0.72, t = 19.25, p < 0.001, CI [0.64, 0.79], R²_adj = 0.52). These results indicate that MVCX is a strong predictor of perceived metaverse-enabled well-being, supporting H1. Moreover, perceived metaverse-enabled well-being significantly predicts metaverse usage intentions (β = 0.20, t = 2.54, p = 0.01, CI [0.05, 0.36]), supporting H2. For completeness, we tested the relationship between MVCX and metaverse usage intentions ^[1], which was also positive and significant (β = 0.41, t = 5.53, p < 0.001, CI [0.26, 0.55]).

Our mediation analysis shows that perceived metaverse-enabled well-being mediates the relationship between MVCX and usage intentions (β = 0.15, t = 2.54, p = 0.01, CI [0.04, 0.26]), supporting H3. Importantly, when perceived metaverse-enabled well-being was excluded from the model, adjusted R² (metaverse usage intentions) decreased to 0.30, showing the importance of the perceived metaverse-enabled well-being construct.

For Study 3, we collected a new data set (n = 494; 48.6% female; see demographic details in  Appendix 1Table A1) to retest the main effects (H1, H2, H3), and also test the moderated relationships (H4). We included the measures used in Study 2, along with metaverse usage length, which was based on a response scale with the following options: Less than 1 month, 1–3 months, 4–6 months, 7–12 months, about 2 years, and more than 2 years.

The analytical procedure used in Study 3 was identical to that used in Study 2. Firstly, we reassessed construct reliability and validity using the new data set, which supported the robustness of our measures (see Tables 2–5). For example, heterotrait–monotrait ratio values were below 0.9 (see Table 3), Cronbach’s alpha and composite reliability values of the reflectively measured perceived metaverse-enabled well-being and usage intentions constructs exceeded 0.7, and AVE values for each construct were above 0.5 (see Table 4).

The indicators of the formatively measured MVCX construct demonstrated meaningful variation in their outer weights (ranging from 0.05–0.26; see Table 2), indicating that the attributes contribute differentially to the formation of the construct rather than functioning as interchangeable manifestations. Because formative indicators collectively define the conceptual domain of the construct, each of the 10 attributes captures a distinct experiential facet of MVCX (Rahman et al., 2025a, 2025b), and removing any indicator would alter the substantive meaning (Hair et al., 2024) and content validity of the MVCX construct. Consistent with formative measurement guidelines, the indicators also exhibited substantial outer loadings (ranging from 0.64–0.74), indicating that each attribute makes an absolute contribution to the construct while remaining conceptually distinct. Multicollinearity diagnostics further supported formative specification, with all VIF values well below the critical threshold of 3 (Kock, 2015), ranging from 1.51–1.84. This suggests that the attributes operate independently without redundancy. Consistent with this pattern, the inter-item correlations among the MVCX attributes were moderate, ranging approximately from 0.34–0.51 (<0.7; see Supporting information Table W3.2). This indicates that although related, the attributes are not excessively correlated in a manner characteristic of reflective measurement. In addition, confirmatory tetrad analysis Gudergan et al. (2008) empirically supported the formative specification of the MVCX construct (see Supporting information W3). Specifically, at least one tetrad was statistically significant (e.g. Tetrad 43: t = 2.10, p = 0.04), with a confidence interval that did not include zero (90% CI [0.02, 0.15]), which is consistent with formative measurement logic. Based on these assessments, we proceeded with the formative configuration of the MVCX construct in the subsequent analysis.

Because the Study 3 sample was dominated by users of Roblox and Sandbox, we examined whether platform-based heterogeneity might influence the measurement or structural relationships in the model. Given the substantial differences in user orientation and value proposition across these platforms, it was necessary to assess whether pooling the data could mask platform-specific effects. We therefore conducted a multigroup analysis comparing Roblox users (n = 190) and Sandbox users (n = 210), using the MICOM procedure to evaluate measurement invariance (Henseler et al., 2016). As reported in Supporting information Table W4.3, compositional invariance was fully established for all constructs, indicating that MVCX, perceived metaverse-enabled well-being and metaverse usage intentions were measured equivalently across the two dominant platforms. We then conducted a bootstrap-based multigroup analysis to compare structural paths Hair et al. (2024). As shown in Supporting information Table W4.4, none of the key structural relationships differed significantly between Roblox and Sandbox users. These results indicate that the structural mechanism linking MVCX, perceived metaverse-enabled well-being and usage intentions remains stable across the dominant platforms, supporting the appropriateness of pooling the data in Study 3 for subsequent path analysis.

The PLS-SEM path analysis results (see Table 6) using the Study 3 data set provided consistent support for the main effects (H1, H2, H3). Similar to Study 2, the relationships between MVCX and perceived metaverse-enabled well-being (β = 0.67, t = 19.52, p < 0.001, CI [0.60, 0.73], R²_adj =0.45), perceived metaverse-enabled well-being and metaverse usage intentions (β = 0.22, t = 2.99, p < 0.001, CI [0.07, 0.36], R²_adj = 0.33), and MVCX and metaverse usage intentions (β = 0.40, t = 5.83, p < 0.001, CI [0.27, 0.54]) were all significant and positive. Moreover, perceived metaverse-enabled well-being mediated the relationship between MVCX and usage intentions (β = 0.15, t = 2.85, p < 0.001, CI [0.05, 0.25]). Thus, H1–H3 continued to be supported by the new data set.

H4 posits that the positive associations among MVCX, perceived metaverse-enabled well-being and metaverse usage intentions are contingent on metaverse usage length. To test H4, we divided the Study 3 data set into two groups: Group A (n = 240) comprised users who had used the metaverse for up to 1 year, and Group B (n = 254) comprised users who had used the metaverse for more than 1 year. The decision to dichotomise metaverse usage length was guided by both statistical and theoretical considerations. Although six usage-length categories were collected in Study 3 (see  Appendix 1Table A1), several early-stage categories contained relatively small cell sizes, which would render more granular multigroup comparisons unstable (Hair et al., 2024). To ensure adequate statistical power and reliable parameter estimation, categories were therefore combined to develop groups with similar sample size (see Mansoor et al. (2024) as examples). Moreover, the distinction between users with 1 year or less of experience in a metaverse platform and those with more than 1 year is theoretically meaningful in the context of technology adoption and metaverse engagement (Dwivedi, Kshetri et al., 2023; Huang et al., 2026). Metaverse research has classified long-term users as those who have used the metaverse for over 1 year, indicating that long-term users display different usage behaviours (e.g. community management and social leadership) compared to novel users (e.g. exploratory and episodic) (Tunca et al., 2026). Early-stage users are typically still acclimatising to the platform environment, whereas longer-term users have developed greater familiarity and more stable evaluations of their experiences. This grouping thus captures a transition from exploratory engagement to more established participation while maintaining sufficient statistical robustness.

Before conducting a PLS-based multigroup analysis (Hair et al., 2024), it is important to ensure that the constructs are compositionally invariant between groups to confirm that any observed differences in relationships are attributable to structural differences (i.e. path estimates) rather than differences in group composition (i.e. construct measures). Our measurement invariance test results (see Supporting information W4) confirm that the composition of the constructs between Group A and Group B was invariant (p > 0.30). Subsequently, the PLS-based bootstrap multigroup analysis indicated significant differences (i.e. the 95% CI did not include zero) in the path-specific estimates between the two groups. Supported by the prerequisite tests for PLS-based multigroup analysis, we then investigated the nonlinear relationships in the three paths in the model:

1. MVCX → perceived metaverse-enabled well-being,

2. Perceived metaverse-enabled well-being → metaverse usage intentions, and

3. MVCX → metaverse usage intentions (see Table 7).

Regarding the nonlinear effect of MVCX on perceived metaverse-enabled well-being, for the complete data set (n = 494), the path coefficient was significant (β = 0.12, t = 2.55, p = 0.01, CI [0.02, 0.20]), indicating that this relationship is nonlinear. In Group A (metaverse usage length ≤ one year), the nonlinear effect of MVCX on perceived metaverse-enabled well-being was not significant (β = −0.01, t = 0.11, p = 0.91, CI [−0.12, 0.07]), although the linear effect remained significant. In contrast, for Group B (metaverse usage length > one year), the nonlinear effect was significant (β = 0.21, t = 5.61, p < 0.001, CI [0.13, 0.28]), supporting a stronger nonlinear effect among long-term users.

Regarding the nonlinear effect of perceived metaverse-enabled well-being on metaverse usage intentions, for the complete data set, the path coefficient was significant (β = 0.11, t = 2.97, p < 0.00, CI [0.02, 0.16]), confirming nonlinearity. However, for Group A, this effect was not significant (β = 0.07, t = 1.57, p = 0.12, CI [−0.03, 0.14]), whereas for Group B, the path coefficient was significant (β = 0.16, t = 2.93, p < 0.00, CI [0.02, 0.23]), indicating that the nonlinear relationship strengthens for long-term users.

Regarding the direct effect of MVCX on metaverse usage intentions, for the complete data set, the nonlinear relationship was significant (β = 0.10, t = 2.51, p = 0.01, CI [0.03, 0.18]). This nonlinear relationship was not significant in Group A (β = 0.07, t = 1.37, p = 0.17, CI [−0.01, 0.21]) but was significant in Group B (β = 0.14, t = 3.23, p < 0.00, CI [0.05, 0.22]).

These findings suggest that for long-term metaverse users (>1 year), MVCX has a significantly stronger effect on both perceived metaverse-enabled well-being and usage intentions, supporting H4. Further, the effects are non-linear (as described previously). In contrast, for individuals who have used a metaverse platform for less than 1 year, the relationships are mostly linear, suggesting that their MVCX has not yet reached a sufficient threshold to trigger strong nonlinear effects on perceived metaverse-enabled well-being and metaverse usage intentions.

The latent variable score plots in Figures 2–4 provide further insights into the nonlinear relationships for long-term users. Figure 2 shows the nonlinear relationship between MVCX and well-being for Group B, where the nonlinear trend (blue line) diverges in a positive direction from the linear trend (green line) as MVCX increases, indicating that incremental improvements in MVCX produce larger gains in perceived well-being among long-term users. Figure 3 shows the nonlinear effect of well-being on metaverse usage intentions, suggesting that usage intentions accelerate as well-being reaches higher levels. Finally, Figure 4 illustrates the nonlinear effect of MVCX on metaverse usage intentions, with a steep rise in the quadratic curve after crossing the linear line.

A closer examination of the nonlinear plots reveals three distinct regions, illustrating how MVCX and well-being influence metaverse usage intentions in a nonlinear manner. In the first region (at lower levels of MVCX and well-being), the quadratic curve sits below the linear line, suggesting that users with relatively low MVCX and well-being show only slight increases in usage intentions, even as their MVCX and well-being improve. This implies that minor enhancements in MVCX and well-being may not immediately translate into significantly higher metaverse usage intentions. A turning point can be observed when MVCX and well-being reach around 5.5–6 on the seven-point Likert scale. At this threshold, the quadratic curve crosses above the linear line, marking the second region, where further increases in well-being have a notably stronger effect on usage intentions. Here, the effect of well-being on platform usage intentions intensifies, reflecting a shift in which MVCX and well-being gains generate greater returns in terms of usage intention. In the third region, as MVCX and well-being continue to increase beyond this threshold, the quadratic curve rapidly steepens, indicating an accelerating effect on usage intentions. These insights emphasise the strategic importance of maintaining high MVCX and well-being levels to drive higher user retention in metaverse platforms.

To further assess the robustness of our findings, we used NCA (Richter et al., 2020) to examine whether positive MVCX and perceived metaverse-enabled well-being constitute necessary conditions for future metaverse usage intentions. Unlike traditional hypothesis testing based on sufficiency logic, where increases in a construct are expected to produce proportional increases in an outcome, NCA evaluates necessity logic. Specifically, it assesses whether a minimum level of a predictor must be present for a particular level of the outcome to occur (Aldhamiri et al., 2024). In other words, while sufficiency-based tests (e.g. H1–H3) determine whether MVCX and well-being enhance usage intentions on average, NCA determines whether high usage intentions are even possible in the absence of sufficient MVCX or well-being. By identifying “must-have” conditions and minimum threshold levels required to achieve desired outcomes, NCA provides complementary insights that go beyond net-effect estimation and allow a more precise understanding of structural constraints within the model (Richter et al., 2020).

Using the CE-FDH technique, the results indicate that both MVCX (d = 0.43, p < 0.00) and perceived metaverse-enabled well-being (d = 0.38, p < 0.00) exert statistically significant and substantively large necessity effects on metaverse usage intentions (see  Appendix 2Table A2). In practical terms, both constructs operate as “must-have” conditions for achieving high levels of sustained engagement. That is, regardless of other favourable factors, high metaverse usage intentions do not occur unless MVCX and perceived metaverse-enabled well-being reach sufficient levels. These findings indicate that both experience quality and perceived life-domain benefits function as structural constraints rather than merely performance-enhancing factors.

The bottleneck analysis further clarifies the practical meaning of these necessity effects ( Appendix 2Table A3). For moderate levels of usage intentions (50%), MVCX must reach at least 3.56 and perceived metaverse-enabled well-being 3.03 on the seven-point scale. However, to achieve very high usage intentions (90%), MVCX must reach 5.23 and perceived metaverse-enabled well-being 5.20. At the maximum outcome level (100%), MVCX must increase further to 5.75, while perceived metaverse-enabled well-being must remain at a minimum of 5.20. The ceiling line plots ( Appendix 2Figures A2 and A3) visually confirm a clear empty upper-left zone, indicating that no respondents reported high usage intentions when MVCX or perceived well-being were low.

In practical terms, managers seeking to achieve very high levels of metaverse usage intentions, such as the 90% outcome level, must ensure that both MVCX and perceived metaverse-enabled well-being exceed strongly positive thresholds, approximately 5.23 and 5.20, respectively, on a seven-point scale. This means that investments in immersive design, usability, personalisation and community features must translate into clearly perceived life-domain benefits, including social, educational, leisure, family and psychological improvements. Without reaching these elevated levels of experiential quality and well-being impact, efforts to drive sustained high engagement are unlikely to succeed, regardless of incremental enhancements or promotional initiatives.

To address common method bias, we conducted several tests. Variance inflation factor values were well below the recommended threshold of 3 (Hair et al., 2022; see Table 5), indicating that multicollinearity was not a concern. In addition, we used a marker variable from Rahman et al. (2022a) social desirability measure, following the guidelines of Podsakoff et al. (2003). The analysis showed no meaningful relationship between the marker variable and the constructs in the theoretical model, confirming that our findings were not significantly affected by common method variance.

We included gender as a control variable to account for potential variations (Mansoor et al., 2024). The samples from all three studies were almost balanced in terms of gender (see  Appendix 1Table A1), providing a robust basis for this control. Further, the permutation-based multigroup analysis confirmed that there was no significant measurement invariance (e.g. CI [−0.22, 0.21] in Study 3) in the path from MVCX to perceived metaverse-enabled well-being and usage intentions between genders. This finding supports the appropriateness of retaining the data from both genders without separate group analyses, indicating that the observed relationships are consistent across male and female users.

This study examined how MVCX shapes users’ perceived metaverse-enabled well-being and, in turn, their metaverse usage intentions. Across three empirical studies, we tested both linear and nonlinear relationships and examined contingent effects related to usage length. In this section, we interpret the findings in relation to existing theoretical and empirical discourse, clarify the study’s contributions and outline implications for research and practice.

This paper makes multiple contributions to the literature across various research disciplines. Firstly, this study contributes to the emerging literature on immersive technologies and the metaverse by systematically bridging this domain with the broader customer experience and well-being literatures. Prior research has highlighted the potential of immersive technologies such as AR and VR to enhance psychological, social and experiential outcomes (Alimamy and Jung, 2025; Zarantonello and Schmitt, 2023; Zhang and Xiao, 2024). However, much of this work remains either conceptual (Hajian et al., 2024) or focuses on specific immersive applications rather than persistent, socially embedded metaverse environments (Eshaghi et al., 2023). In addition, empirical research has rarely operationalised well-being in a way that captures how immersive platform engagement contributes to users’ broader life domains (Bowden et al., 2025).

Addressing this gap, we operationalise perceived metaverse-enabled well-being as a domain-specific and context-attributed evaluative construct grounded in established subjective well-being scholarship (Diener, 1984) and adapted to immersive digital ecosystems (Dwivedi, Kshetri et al., 2023; Yoo et al., 2023). Rather than treating well-being as a global life satisfaction measure or transient affective state, we conceptualise it as users’ evaluative judgments regarding how metaverse engagement enhances salient domains of daily life, including social connectedness, educational development, leisure experiences, family life and psychological functioning. This operationalisation responds to recent calls for more precise measurement of well-being in immersive environments (Jung et al., 2024; Saleh, 2024) and clarifies how well-being can be meaningfully assessed in the context of persistent virtual platforms.

Secondly, this study contributes to the broader co-creation and S-D logic literature by demonstrating how its core principles operate within immersive digital ecosystems. S-D logic conceptualises value as emerging through resource integration and value-in-use within service ecosystems (Vargo and Lusch, 2008, 2014), where multiple actors integrate operant resources to co-create outcomes (Edvardsson et al., 2018; Storbacka et al., 2016). While prior research has applied S-D logic to digital platforms and virtual interactions (Kowalkowski et al., 2024; Rivière et al., 2024), limited empirical work has examined how immersive metaverse experiences translate resource integration into user-level well-being outcomes. By positioning MVCX attributes as enabling structures that facilitate users’ integration of time, creativity, skills and social capital within persistent virtual environments (Mansoor et al., 2024; Scholz and Duffy, 2018), this study shows how value co-creation extends beyond experiential enjoyment to shape evaluative judgments of quality of life. In doing so, we do not seek to extend S-D logic theoretically; rather, we offer an empirically grounded application of S-D logic to immersive technology contexts, illustrating how value-in-use becomes reflected in perceived well-being and sustained engagement within metaverse service ecosystems.

Thirdly, this study contributes to the growing literature on nonlinear effects in customer experience and engagement research. Recent work in omnichannel retailing demonstrates that customer experience does not always produce proportional effects on engagement outcomes; instead, relationships may intensify once experiential perceptions reach certain thresholds (Rahman et al., 2025a, 2025b). Extending this line of inquiry, our findings provide empirical evidence of nonlinear dynamics in the metaverse, an emerging channel within brands’ broader omnichannel mix. Specifically, we show that improvements in MVCX are associated with modest gains in perceived metaverse-enabled well-being and usage intentions at lower levels, but that these effects accelerate once MVCX reaches higher thresholds. By documenting such nonlinear patterns in immersive virtual environments, this study extends prior omnichannel findings into the domain of persistent digital ecosystems, suggesting that experiential investments in emerging channels such as the metaverse may produce disproportionate returns once users achieve sufficient familiarity and depth of engagement.

Finally, this study contributes methodologically by integrating PLS-SEM with NCA, a complementary technique that provides more precise insights into the structure of relationships within complex models (Richter et al., 2020). While PLS-SEM assesses sufficiency logic by examining whether higher levels of one construct are associated with higher levels of another, NCA identifies whether a construct constitutes a necessary condition, that is, whether a minimum level must be present for a desired outcome to occur. Applying this dual-method approach (see also Aldhamiri et al., 2024, and Rahman et al., 2026, as examples), we show that perceived metaverse-enabled well-being functions as a necessary condition in the pathway between MVCX and metaverse usage intentions. This finding moves beyond conventional net-effect interpretation by demonstrating that without sufficient levels of well-being, metaverse usage intentions are unlikely to materialise, regardless of improvements in MVCX. By incorporating necessity logic into the examination of immersive service ecosystems, this study encourages future customer experience and digital engagement research to move beyond purely sufficiency-based models and consider minimum threshold conditions that shape behavioural outcomes; a call consistent with the latest PLS-SEM literature (Becker et al., 2026).

This research offers multiple implications for practitioners. Firstly, our findings suggest that managers should design metaverse strategies with an explicit focus on user well-being rather than assume immersive features have intrinsic value. The results indicate that MVCX translates into sustained metaverse usage intentions primarily through perceived metaverse-enabled well-being. In practical terms, immersive design elements such as aesthetics, immersion, personalisation and communality should not be deployed solely to enhance novelty or entertainment value. Instead, they should be intentionally configured to support meaningful life domains, including social connection, learning, leisure enrichment and psychological relief.

For example, brands operating in metaverse environments could develop collaborative learning spaces, community-driven events or creativity-enhancing tools that align with users’ educational, social or personal development goals. Similarly, features that facilitate seamless integration into daily routines, such as persistent identities or interoperable digital assets, may help users embed metaverse engagement within their broader life context. By aligning platform attributes with users’ real-life goals and routines, firms can move beyond episodic engagement towards more meaningful and sustained participation. Importantly, this implication reframes metaverse investment decisions. Rather than evaluating immersive initiatives solely through short-term engagement metrics (e.g. visits, clicks or event participation), managers should assess whether metaverse experiences contribute to users’ perceived quality of life in relevant domains. Monitoring well-being-related indicators can provide a more strategic lens for understanding whether immersive environments are generating enduring value rather than temporary excitement.

Secondly, the NCA findings indicate that managers should treat user well-being as a minimum threshold condition for sustained metaverse engagement. Meaning, sustained metaverse engagement depends on whether the experience meaningfully enhances users’ lives and supports their broader goals and routines. While improvements in MVCX are important, our results show that without sufficient levels of perceived metaverse-enabled well-being, usage intentions are unlikely to materialise. In practical terms, investments in visual sophistication, gamification or virtual assets will have limited impact if users do not perceive meaningful contributions to their social, educational, leisure, family or psychological lives. For managers, this implies the need to assess whether their metaverse initiatives have reached a baseline level of relevance in customers’ lives before scaling further investments. Firms can use the perceived metaverse-enabled well-being measure developed in this research as an easy-to-administer diagnostic tool to evaluate whether immersive experiences are generating meaningful domain-level value. Establishing this minimum threshold of well-being impact may be more critical than incremental improvements in experiential design.

Thirdly, the evidence of nonlinear effects suggests that the returns from improving MVCX may accelerate once certain experiential thresholds are reached. In the early stages, enhancements to the metaverse experience are associated with modest gains in well-being and usage intentions. However, once the quality of the experience becomes sufficiently strong, the positive effects intensify. For managers, this implies that metaverse investments should be viewed as cumulative and strategic rather than short-term experiments. Initial returns may appear limited, but sustained improvements in experience design, usability, community features and personalisation can generate disproportionate engagement benefits once users reach deeper levels of familiarity and integration. This insight is particularly relevant as the metaverse becomes an emerging channel within brands’ broader omnichannel ecosystems.

Finally, the moderating effect of usage length indicates that the impact of MVCX on perceived metaverse-enabled well-being and usage intentions is stronger for users who have engaged with the metaverse for more than 1 year, suggesting that continuous exposure enhances the translation of immersive experiences into meaningful outcomes. This finding carries an important managerial implication, particularly as metaverse platforms expand and new competitors emerge, increasing the risk of user attrition and churn. Users may experiment across platforms during early stages of adoption, and if they disengage before meaningful well-being benefits are realised, the long-term engagement effects of MVCX may not materialise. Managers should therefore prioritise retention-oriented strategies that promote continuity of engagement, such as structured onboarding, progressive feature development and community-building mechanisms that integrate metaverse participation into users’ routines.

Firstly, the results demonstrate that the relationship between MVCX, perceived metaverse-enabled well-being, and metaverse usage intentions becomes nonlinear once thresholds are reached. The predominance of data points in the positive zone (see Figures 2–4) reflects our focus on favourable aspects of metaverse engagement, such as user well-being. However, the U-shaped plots in Figures 2–4 may be misleading to readers, because they suggest zones of negative associations. This point warrants further investigation because data points are relatively sparse in the negative association zones.

Secondly, we used “1 year (and above)” as an indicator of long-term metaverse use; however, a continuous usage metric may offer deeper insights, such as identifying the differences between (say) 1 and 3 years of metaverse platform use. Also, future researchers could compare brand-new users (e.g. < three months), relatively new users (e.g. six months) and highly loyal long-term users (e.g. ≥ five years) to understand how user behaviours and experiences evolve over time. Such analysis would require an adequate sample size in each group, thus a much larger overall sample size, which is a limitation in this study.

Thirdly, a limitation of the current operationalisation concerns several life domains that were conceptually relevant but did not perform adequately during the exploratory factor analysis stage (Study 1 b), including community life, financial well-being, work life and physical well-being. Although these domains were supported by prior literature and qualitative insights, they did not load consistently within the final measurement model. This may reflect differences in platform focus and user goals within our heterogeneous sample, where certain domains were not sufficiently salient to emerge as stable evaluative indicators.

This paper also suggests multiple areas for future research (see Table 8 for a summary), with implications for both theory and practice. First, the findings suggest that a more positive MVCX is associated with higher perceived metaverse-enabled well-being. However, given that MVCX is a multi-attribute measure, some attributes may have stronger associations than others. Further research is needed to better specify how each individual MVCX attribute is associated with perceived metaverse-enabled well-being. This could help metaverse firms make suitable trade-offs and decide which MVCX attributes to prioritise to enhance user well-being and metaverse usage intentions. Further, contingencies may affect the extent to which each MVCX attribute is associated with enhanced user well-being.

Secondly, because perceived metaverse-enabled well-being is positively associated with metaverse usage intentions, metaverse firms should focus on how to enhance perceived metaverse-enabled well-being. Therefore, it is important that future researchers explore other means of enhancing user well-being, beyond improving MVCX, to provide metaverse firms with multiple pathways for well-being enhancement.

Thirdly, future researchers could explore factors other than metaverse usage length that might moderate the relationships between MVCX, perceived metaverse-enabled well-being and metaverse usage intentions. These may include individual (e.g. age, tech savviness), geographic, social and cultural factors. Cultural factors are particularly important given the metaverse’s global reach and the fact that the current research was conducted solely in the USA. Cultural norms and values may shape how individuals interact with immersive digital environments (Chang, 2024), influencing how they perceive their MVCX and its influence on their well-being. Future researchers could compare users across different countries (e.g. China, India, European countries) to explore the moderating effects of culture on the MVCX → perceived metaverse-enabled well-being → usage intention pathway. Also, as stated prior:,

• More granular examinations of the effects of usage length, and

• Longitudinal studies, are also likely to provide useful insights.

Fourthly, a promising research direction would be to explore how artificial intelligence (AI) may be used to enhance MVCX through real-time personalisation, a critical attribute in service settings (Rahman, Carlson, Gudergan et al., 2022b). AI-driven avatars and recommendations could adapt to users’ emotional cues and preferences, creating more immersive and supportive digital environments. This approach may promote user well-being by aligning activities with individual goals. Researchers could examine how AI-enabled personalisation influences user satisfaction, well-being and engagement, offering insights into AI’s potential to create a metaverse that actively supports and adapts to user needs.

Fifthly, this study is the first to offer a general understanding of the relationships between MVCX, user well-being and metaverse usage intentions across a range of platforms, including Roblox, Sandbox, Upland and Fortnite, reflecting the diversity of user experiences. However, we did not differentiate between specific platforms. Future researchers could adopt a platform-specific approach to uncover specific insights critical for tailoring managerial strategies. For instance, a comparative analysis of platforms could highlight whether differences in technical capabilities, social engagement features or immersive tools moderate the MVCX–well-being relationship. Such detailed insights could help practitioners design platform-specific interventions to drive growth and mitigate challenges unique to particular platforms. Two (related) points are worth discussing: one, certain platforms are more suited to gamification initiatives; in general, how might enhanced use of gamification impact MVCX, user well-being, and usage intentions? Two, certain platforms are better suited to professional and educational settings; what might the relationship across MVCX, user well-being, and usage intentions be in these settings?

Sixthly, given that community, financial, work and physical well-being did not load consistently in the present study (see the limitations discussion), future research should examine these domains under more context-specific conditions. Their exclusion may reflect platform heterogeneity rather than conceptual irrelevance. For example, financial and work-related well-being may be more salient within professional or creator-focused metaverse platforms, while physical well-being may be more relevant in fitness-oriented or embodied environments. Similarly, community-level well-being may emerge more clearly in geographically anchored or civic-focused platforms. Platform-specific designs, user segmentation and longitudinal approaches may help determine whether these domains represent context-contingent dimensions of perceived metaverse-enabled well-being.

Finally, while the findings demonstrate the positive effect of MVCX on perceived metaverse-enabled well-being, the possible U-shaped relationship between MVCX and perceived metaverse-enabled well-being, discussed in Section 5.3, suggests that interactions in the metaverse could, in some cases, be detrimental to user well-being. For instance, excessive engagement in virtual environments has been linked to social isolation and anxiety (Dwivedi, Kshetri et al., 2023). In addition, exposure to inappropriate content may diminish real-world satisfaction and overall well-being (Corpuz, 2023). Future researchers could explore aspects of the metaverse that adversely affect perceived metaverse-enabled well-being and metaverse usage intentions. This would provide a more comprehensive understanding of the health and social outcomes of the metaverse, which will play an important role in future policymaking.

Macquarie Business School at Macquarie University (Sydney, Australia) funded the data collection for this study.

The data collected in this study are available from the authors upon reasonable request and ethical considerations (Macquarie University, Australia, approved ethics project number 18150).

The supplementary material for this article can be found online.