Understanding AR adoption: consumer motivations, barriers, and occupational differences

Augmented reality (AR) is transforming the e-commerce industry by acting as a strong interactivity and communication tool that enhances consumer motivation and engagement, especially for high-involvement products like fashion, eyewear and furniture (McLean and Wilson, 2019). Craig (2013, p. 20) defined AR as a “medium in which digital information is overlaid on the physical world that is in both spatial and temporal registration with the physical world and that is interactive in time.” Global and regional companies like IKEA (furniture), Lenskart (eyeglasses), Gucci (fashion), HandM (fashion) and Adidas (footwear) have integrated AR technology into their online platforms to enable an immersive shopping experience for their customers (McLean and Wilson, 2019). The global market size for mobile AR is projected to reach US$36bn by 2026 (Statista, 2024).

Given AR technology’s novelty and rapid advancement, it has gained significant interest among researchers, especially in the marketing domain. Prior studies have highlighted various advantages of AR-enabled mobile applications, like engaging and immersive shopping experiences, utilitarian and hedonic benefits (Nikhashemi et al., 2021). In the online retailers’ context, AR-enabled mobile applications not only enhance consumer engagement (McLean and Wilson, 2019) but also strengthen consumer–brand relationships, resulting in deep emotional connections (Nikhashemi et al., 2021).

However, despite the primary focus of the existing literature on the potential benefits (like immersive, interactive, consumer–brand relationship and positive behavioral responses) of AR adoption for both retailers and consumers, there is a gap in the literature regarding the risks of AR or barriers to its adoption. Though scholars have explored barriers to technology adoption like privacy concerns (Hilken et al., 2017), and perceived AR adoption risks, there is a lack of comprehensive research that integrates both positive and negative antecedents of AR adoption, in terms of both usage and electronic word-of-mouth (eWOM) intentions. Also, the existing literature has not sufficiently addressed the moderating effect of customer occupation on these relationships. This study aims to bridge the above-mentioned gaps by undertaking multi-theoretic research grounded in Innovation-Resistance Theory (IRT) and Self-Determination Theory (SDT). By using this multi-theoretical lens, this study focuses on examining both the drivers and barriers to the use of AR technology. Specifically, this research answers the following research questions (RQ):

The remainder of this study is structured as follows: Section 2 presents the theoretical background for SDT and IRT theories and hypothesis development. Section 3 details the research methodology. Data analysis and results follow in Section 4. Section 5 discusses the findings, while Section 6 explores theoretical and practical implications. Finally, Section 7 offers concluding remarks, limitations and suggestions for future research.

In line with industry trends, numerous academic studies have been conducted in recent years to investigate the factors driving AR adoption (Table 1) and their relative importance (Perannagari and Chakrabarti, 2020). The factors identified by extant studies can primarily be grouped into two categories: AR attributes and media or technology attributes. Researchers have identified several AR attributes that create engaging customer experiences, such as flow, enjoyment, ease of use, usefulness and gamification (Yavuz et al., 2021). Media or technology attributes studied include informativeness, interactivity, vividness, response time, aesthetic quality and so on (Lee et al., 2022).

While the focus of these studies is primarily on the positive drivers of AR adoption, few studies have identified factors that negatively impact people’s intention to use AR applications. Rauschnabel et al. (2018) found that while one’s own privacy risk in AR applications does not directly impact users’ behavior, the extent to which AR applications threaten other people’s privacy can strongly influence users’ decision-making. Researchers have used theoretical frameworks such as the technology acceptance model (Pantano et al., 2017), the unified theory of acceptance and use of technology 2 (Khashan et al., 2023), and consumer value theory (Schultz and Kumar, 2024) to investigate multiple traditional and new factors that drive consumer acceptance of AR (Table 1).

There is a dearth of studies focusing on the hindrances or barriers and the use of appropriate theoretical frameworks to investigate these. The extant literature also confirms that the user’s intention to adopt AR is moderated by user characteristics, product characteristics (Lee et al., 2022; Perannagari and Chakrabarti, 2020; Yavuz et al., 2021), and contextual factors (von der Au et al., 2023). Among user demographic characteristics, studies have primarily focused on the moderating impact of age and gender (Khashan et al., 2023; Rauschnabel et al., 2018; Schultz and Kumar, 2024) while leaving the impact of occupation and education relatively unexplored.

Self-determination theory (SDT) is a comprehensive macro-theory of human motivation that highlights the significance of environmental factors in fulfilling the basic physiological needs of individuals. According to SDT, satisfaction of these physiological needs significantly enhances the well-being, performance and motivation of the individual (Ryan and Deci, 2000). SDT identifies three key motivations for adoption: autonomy, competence and relatedness (Ryan and Deci, 2000). In this context, autonomy is defined as the perception of engaging in activities according to one’s desires (Vallerand et al., 1997). We have included informativeness in autonomy, as providing relevant and clear information empowers users to make informed decisions and enhances their control over their experience (Kang et al., 2020). Competence refers to an individual’s sense of effectiveness and capability in interacting with their environment (Vallerand et al., 1997). In our framework, perceived usefulness aligns with competence by improving users’ ability to achieve their goals and to enhance their performance, making them feel more effective in their tasks (Davis et al., 1992). Relatedness is characterized by the sense of connection and belonging with others (Vallerand et al., 1997). Aesthetics contributes to relatedness by creating a visually appealing and inviting environment, fostering a sense of belonging and emotional connection (Jung et al., 2018). Perceived enjoyment further supports relatedness by strengthening emotional ties to the application, thereby enhancing users’ engagement (Davis et al., 1992) and, thus, a sense of connection.

In the context of technology adoption, or the human-computer interface, aesthetics is defined as the sensory appeal and visual design of a digital platform, which includes the layout design, fonts, color or photographs used (Jung et al., 2018). Several studies have indicated the significance of aesthetics in the adoption of technology. For example, Cyr et al. (2006) highlighted the significance of aesthetics in determining the intention to use technology. Extant studies have highlighted that the impact of aesthetics on the intention to use AR is higher in industries where visual appeal influences user experience, such as mobile gaming (Scholz and Duffy, 2018). Therefore, we hypothesize the following:

Perceived usefulness refers to the extent to which an individual believes that using a specific technology will enhance their performance (Davis et al., 1992). In the context of AR applications, perceived usefulness is associated with improvements in task performance during the online shopping experience. Existing research indicates how the perceived usefulness of AR applications significantly influences the consumer’s intention to use AR technology in diverse areas like mobile gaming (Pantano et al., 2017), online tourism bookings and online retailing (Poushneh and Vasquez-Parraga, 2017). Hence, we hypothesize the following:

Perceived enjoyment is defined as the degree to which technology use is enjoyable in its own right, apart from the anticipated performance consequences (Davis et al., 1992). Van der Heijden (2004) highlighted perceived enjoyment as a key determinant in the adoption of digital interfaces or systems because it enhances the user experience, thereby increasing the attractiveness of the technology.

When users interact with AR applications, they get an immersive experience of being a physical part of a virtual environment, thereby enhancing their sense of pleasure and engagement (Poushneh and Vasquez-Parraga, 2017). Rauschnabel et al. (2022) highlighted the strong relationship between perceived enjoyment and intention to use AR in mobile gaming due to the enhanced immersive experience. Based on the discussion, we hypothesize the following:

Informativeness is defined as the extent to which a technology interface provides online users with useful, informative, rich and accurate information (Kang et al., 2020). Unlike non-AR applications, AR-enabled platforms provide both factual details and immersive experience-based information about the value of the offering. Prior research argues that the use of AR in online applications or platforms provides extensive knowledge in an immersive setting, thereby influencing behavioral intention (Huang and Liao, 2015; Kang et al., 2020). Hence, we hypothesize the following:

In our study, we selected four barriers – usage, value, risk and image – due to their direct relevance to AR adoption in online retail. Although other barriers, such as psychological ones, are relevant to IRT, these four were chosen for their clear connection to AR adoption in online retail.

“Usage barrier” refers to the obstruction caused by innovation-led changes requiring a shift from existing systems or processes (Laukkanen et al., 2007; Ram and Sheth, 1989). Existing research highlights that usage barriers toward applications or online platforms are mainly influenced by the changes felt or difficulties faced by users due to major changes in areas like user interface, workflow, content or navigation (Kumar et al., 2022). Existing studies have argued that usage barriers are an important issue in the context of adopting complex technology (Kumar et al., 2022; Laukkanen et al., 2007; Migliore et al., 2022). Hence, we hypothesize a negative association between usage barriers and the user’s behavioral intention to use or adopt the AR platform:

“Value barrier” refers to the performance-to-price value of new innovations as compared to existing alternatives (Laukkanen et al., 2007). Unless an innovation offers higher value than an existing alternative, users find no reason to change or switch from an existing arrangement (Ram and Sheth, 1989). Previous research indicates a negative association between the value barrier and the user’s intention to use or adopt digital innovations such as on-demand home service applications (Kumar et al., 2022), mobile banking (Laukkanen et al., 2007) and mobile payments (Migliore et al., 2022). Hence, we propose the following hypothesis:

“Risk barrier” refers to the level of different types of risk (financial, psychological, physical or social) resulting from uncertainties around an innovation (Ram and Sheth, 1989). For example, users associate greater privacy and security concerns with Internet banking compared to traditional banking (Migliore et al., 2022). Researchers have found that risk barriers have a negative relationship with the adoption of digital innovations like home service applications (Kumar et al., 2022). Based on this discussion, we hypothesize as follows:

If users feel that adopting a particular technology will be difficult to use, harm their social image or is not aligned with their social identity, they are likely to resist the adoption of that technology. Existing research has indicated a negative association between the image barrier and the behavioral intention of users toward the use of digital innovations like mobile payments (Migliore et al., 2022) and mobile banking (Laukkanen et al., 2007). Thereby, we hypothesize the following:

One of the key differentiating factors in consumer decision-making is eWOM, especially in today’s digitally connected world (Hilken et al., 2017). Lee and Chen (2020) argued that users who have favorable intentions to use mobile AR applications are likely to recommend or share their positive experience with others, thereby showing positive eWOM intention. Several research studies have demonstrated a significant relationship between eWOM communication and the intention to use or purchase brands, services or products such as smartphones (Chen et al., 2016). Hence, we hypothesize:

The extant literature on technology adoption confirms that user context plays a critical moderating role in influencing system adoption, cautioning against generalizing findings from student users to non-student users. Research supports the belief that students are more influenced by hedonic factors like aesthetics and enjoyment while adopting AR-based applications (Cabero-Almenara et al., 2019; Ghobadi et al., 2022). Ghobadi et al. (2022) argued that visually engaging and enjoyable experiences have a significant impact on AR adoption among students. Similarly, Cabero-Almenara et al. (2019) demonstrated that university students exhibit a stronger level of engagement with the AR-based applications due to their novelty, interactivity and visual appeal. Hence, we hypothesize:

In contrast, working professionals often prioritize instrumental value like functionality, usability, and informativeness while using AR tools (Mehta et al., 2021). Martins et al. (2023) highlighted that professionals adopt AR in the corporate training environment primarily due to its perceived usefulness in terms of knowledge management. Similarly, Cheng et al. (2025) indicated that one of the key attributes that led to the higher adoption of AR among working professionals is informativeness in real-world settings. Hence, we hypothesize:

Scholars have confirmed that demographic differences such as gender, age and occupation influence technology adoption (Morris and Venkatesh, 2000). Students who are younger in age and possess no or limited work experience often have a higher tendency to experiment and are less worried about the potential risks or outcomes associated with the use of new technologies (Morris and Venkatesh, 2000). Thus, these risks can be less impactful for them. In contrast, working professionals are more focused on the tangible benefits of technology; if AR applications do not offer a clear advantage over existing solutions, working professionals may find them less valuable. Usage barriers, such as the difficulty of integrating AR apps into life, are also less relevant for students as young people are usually more adaptable and willing to invest time in learning new tools (Morris and Venkatesh, 2000). The IRT (Ram and Sheth, 1989) asserts that people evaluate innovation-related barriers on the basis of their experience and context. Professionals majorly capitalize on conventional outcomes and routines and may perceive higher functional and psychological trade-offs in comparison to the students. On the other hand, young individuals are more inclined toward the usage of technology (Wang et al., 2009); thus, barriers may not majorly influence their intention to use. Hence, there can be more negative influences of usage, value, risk and image barriers on the working professionals’ adoption intentions. Hence, we hypothesize:

Age and gender are commonly recognized demographic factors that have been shown to significantly influence consumer behavior, particularly in the context of technology adoption and online shopping (Chen et al., 2015). Younger individuals, with their curiosity, stronger memory and learning abilities, are more likely to adopt new technologies positively, whereas older users may be slower to accept them due to a preference for stability in their work (Ye et al., 2020). Furthermore, research suggests that men and women differ in their acceptance and use of technology, with women often experiencing greater anxiety and feeling less in control of their technology-related actions (Sobieraj and Krämer, 2020). Consequently, in this study, age and gender have been incorporated as control variables. In line with existing literature, our research model was adjusted to mitigate the potential confounding influence of age and gender on two criterion variables: intention to use AR and eWOM (Figure 1).

The measurement items for the main constructs in the research model were sourced from previous research and then customized to better suit the specific context of using AR-based mobile applications. Wherever necessary, we modified the wording to fit the focus of the present study while preserving the meaning of the original construct. For instance, the scales for aesthetics and perceived usefulness were adapted from Huang and Liao (2015) with slight modifications to bring them into alignment with AR applications. These measurement items underwent content validation by five academicians from Indian business schools to ensure that all adapted items captured the intended constructs effectively.

The extant body of knowledge on AR adoption in marketing primarily depends on data collected from Western developed countries. Compared to the firms operating in developed countries, the firms in developing countries suffer from resource constraints and inadequate supporting infrastructure, while consumers in these countries are value-conscious and place greater emphasis on avoiding uncertainty associated with new technologies. The first step toward encouraging the use of AR is understanding the drivers and barriers to the adoption of AR in different countries to address the impact of contextual diversity. We propose to do this by collecting data from India, which is a large developing country.

Pilot tests were conducted independently to evaluate the questionnaire. Participants were instructed to indicate their level of agreement with each question using a five-point Likert scale. A total of 70 responses were gathered from students and another 70 from working professionals, using a convenience sampling approach. We evaluated the outcomes of these pilot tests using SmartPLS software and found Cronbach’s alpha values exceeding 0.7 for each construct. Moreover, during the factor analysis, we observed that each item’s loading was greater than 0.5 on its corresponding construct.

The survey was sent to 2,810 students and working professionals in India to gauge their familiarity with AR-based mobile applications from October 2023 to December 2023. The survey briefly explained AR to ensure participants understood the concept. Further, to ensure participants’ attention and engagement, we used attention-check questions to verify that responses were consistent and thoughtful. This ensured that participants had some experience with such applications. Out of 2,810 participants approached, 741 engaged with the survey, yielding a 26.37% response rate. However, only 478 participants reported prior use of AR applications. As a result, the final data set included 478 valid responses. Table 2 provides the demographic information of the respondents, and Table 3 provides the measurement items for each construct.

We used the SEM approach to study the associations in our proposed research model. SEM typically offers two methods: covariance-based SEM and variance-based partial least squares (PLS) SEM. We used SmartPLS 4 software to analyze the survey data. We opted for the PLS-SEM approach because of its capability to handle complex predictive models. PLS path modeling is considered a reliable method for investigating causal models encompassing several constructs. PLS-SEM is ideally suited for marketing research, as it tests theories and derives predictive managerial insights. This method excels in this role because it follows a causal-predictive paradigm, aiming to assess the predictive power of a meticulously developed model based on theory and logic (Hair et al., 2024). We made use of SmartPLS 4 to evaluate our measurement and structural models.

Scale reliability for the constructs was scrutinized using Cronbach’s alpha, with all values exceeding the threshold of 0.7 (refer to Table 3), indicating the satisfactory reliability of the used scales (Fornell and Larcker, 1981).

To gauge convergent validity, we examined the values of the average variance extracted (AVEs), all of which were found to exceed 0.5 (as shown in Table 3), indicating strong convergent validity. For discriminant validity, we compared the square roots of AVEs with the inter-construct correlation values, as presented in Table 4. The square root values of AVE were found to be higher than the corresponding values of the inter-construct correlation. Thus, we found acceptable discriminant validity as per Fornell and Larcker (1981).

We used the variance inflation factor (VIF) to address multicollinearity concerns. We found that all VIF values for the constructs fell within the range of 0.1–3.3, which is below the commonly accepted threshold of 5 for detecting multicollinearity, affirming the absence of multicollinearity issues (Kutner et al., 2005).

Additionally, common method bias (CMB) was assessed using both procedural and statistical methods (Podsakoff et al., 2003). To address CMB procedurally, participants were assured that their responses could be candid, as there were no right or wrong answers, and that their anonymity would be preserved. Statistically, we applied the full collinearity assessment approach, as recommended by Kock (2015), which is effective for detecting CMB in partial least squares PLS-SEM. According to Kock (2015), a VIF value exceeding 3.3 suggests the presence of CMB. Our analysis showed that all VIF values for the constructs in this study were below 3.3, indicating that the model is free from CMB based on Kock’s criteria.

We assessed the structural model’s fit using our data set. Following Henseler et al. (2016), an acceptable model requires a Standardized Root Mean Square Residual (SRMR) < = 0.08 and a Normalized Fit Index (NFI) > 0.90. Our analysis yielded an SRMR of 0.037 and an NFI of 0.919, indicating a well-fitting PLS path model.

The structural model was used to test the hypotheses. Figure 2 illustrates the results of the structural model analysis. In the context of our study, all the hypotheses received support except H5 and H8. Regarding the control variables, eWOM intention is not influenced by either age or gender. Additionally, while gender does not impact the intention to use AR, age does, with younger individuals showing a higher intention to use AR. Furthermore, it was observed that this model accounted for 47.7% of the variation in the usage intention and 22.9% of the variation in the eWOM intention.

Table 5 summarizes the findings from the multigroup analysis. For working professionals as well as students, we discovered support for the relationships: Aesthetics → Intention to Use, Perceived Usefulness → Intention to Use, Perceived Enjoyment → Intention to Use, Informativeness → Intention to Use, Value Barrier → Intention to Use and Risk Barrier → Intention to Use. We did not find support for the relationships Usage Barrier→Intention to Use and Image Barrier→Intention to Use for working professionals and for students. Occupation moderated the relationships of Aesthetics→Intention to Use, Perceived Usefulness→Intention to Use, Perceived Enjoyment→Intention to Use, Value Barrier→Intention to Use and Risk Barrier→Intention to Use. Among them, the relationships Perceived Usefulness→Intention to Use, Value Barrier→Intention to Use and Risk Barrier→Intention to Use were stronger for working professionals, while the relationships Aesthetics→Intention to Use and Perceived Enjoyment→Intention to Use were stronger for students. Accordingly, partial support was found for H10.

Our study established a positive relationship between aesthetics and intention to use (H1). The extant research on AR in marketing has also found that aesthetics impact spatial presence, which in turn influences purchase intention (Wang et al., 2022), while a poorly designed interface hinders people’s motivation to shop online. Properly designed AR-enabled retail technology platforms can be enjoyable and useful, thus resulting in intention to use (H2, H3). Usefulness and enjoyment are essential components of the experiential dimension of AR, which in turn impacts purchase intention (Erdmann et al., 2023).

We found a significant negative effect of the perceived value barrier on the intention to use AR for online purchase scenarios (H6). Our findings also support the accepted notion that to assess new technology-based stimuli, consumers focus a lot on perceived value and the additional benefit they would receive (Erdmann et al., 2023). Informativeness positively impacts usage intention, while the risk barrier negatively impacts use intention (H4, H7). The results of hypotheses H4 and H7 need to be interpreted in conjunction with each other. AR, like any other new technology, causes additional risk, particularly for first-time users. AR-enabled retail platforms need to provide sufficient information and allow better product presentations in multidimensional formats to reduce uncertainty and risk (Poushneh and Vasquez-Parraga, 2017).

However, usage and image barriers are found to have an insignificant relationship with AR use intention (H5 and H8) during online shopping. These observations are in contrast with existing AR studies, which analyzed the negative relationship between IRT factors and the adoption or use of AR apps in the online learning environment and tourism sector (Ronaghi and Ronaghi, 2022) and elsewhere. A plausible reason for the anomaly with respect to the insignificant role of usage and image barriers in AR use by customers can possibly be related to the increasing affordability, accessibility and availability of technological innovations like smartphones with high processing power and speed, as well as high-speed internet; these factors have diminished the challenges of accessibility, affordability and acceptance.

The image barrier may be more acute in the older population compared to students and working professionals; for older people, the perception that AR applications need complex skills may lead to the formation of a “difficult-to-use” image and avoidance of AR-based innovation (Ma and Lee, 2019). The present study has also found a strong positive impact of use intention on consumers’ WOM intention (H9). Studies have found that the superior quality of an online platform boosts the emotional and functional value perceived by the customer, which triggers the customer to share positive WOM about the brand. Our research extends the above findings to the AR-enabled retail technology platform context. Customers providing positive WOMs are more likely to continue to use AR-enabled applications in the long run. Our study also identified a partial moderating impact of one of the customer attributes (occupation) on people’s intention to use AR-enabled retail technology platforms (H10a and H10b).

Our study extends the findings of recent studies that have examined the moderating impact of technological innovativeness (Erdmann et al., 2023), fashion innovativeness and individualism (Wang et al., 2022) in the AR context. Our study suggests that occupational differences moderate the influence of both SDT and IRT determinants (motivations and barriers) on consumers’ behavioral intentions. Specifically, for students, perceived usefulness, the value barrier and the risk barrier have a lower influence on attitude toward using the new technology. On the other hand, the relationship between aesthetics and perceived enjoyment with the intention to use is stronger for students. By focusing on the moderating impact of occupation, our research addresses the call for a holistic understanding of AR adoption and engagement (Rauschnabel et al., 2024).

Our study, by including occupation as a moderator, also goes beyond the limitations of many existing studies, where consumers are considered a homogeneous group (Barta et al., 2025). Addressing the uniqueness associated with specific demographic factors (such as students or working professionals) can lead to more inclusive and effective AR apps, driving higher adoption and reuse. The findings of this study can be enhanced with extant and new studies to consider additional demographic factors such as technical skill and cultural background. As well as maximizing adoption, understanding the differences among different segments of consumers can lead to the development of ethically acceptable AR. While the student community is more tolerant of risk, additional ethical guidelines should be enforced to maintain their privacy, compared to working professionals (Barta et al., 2025).

This study makes several contributions to the marketing literature. First, our study contextualizes the key elements from SDT that influence the intention to use AR among online consumers. All the SDT antecedents identified in this study (such as aesthetics, perceived usefulness, perceived enjoyment and informativeness) positively influence the use of AR by online consumers. Though our model has been validated in the AR context, the framework can be used to study the adoption of other technologies, such as virtual reality and mixed reality.

Second, most of the extant studies focused on a particular aspect of AR adoption (Barta et al., 2025); in contrast, our study synthesizes these diverse domains by examining the impact of media characteristics and AR app features on psychological outcomes and technology adoption. Third, studies conducted before 2020, when AR apps were not widely available, involved participants with limited practical experience (Barta et al., 2025), which might have constrained the findings of those studies. Hence, our study contributes by revalidating and examining the findings of extant studies on AR with technologically aware participants.

Fourth, our study addresses the call by scholars to assign demographic variables a more active role in AR research and application design (Rauschnabel et al., 2024). This study focused on occupation, one of the least investigated demographic variables, and treated it as a moderator rather than a control variable (Rauschnabel et al., 2018). Our research is one of the first to investigate the moderating impact of users’ occupation on the relationships between SDT factors and AR use intention.

Fifth, this is one of the very few studies in the AR context that has looked into both positive and negative antecedents affecting AR use intention. The majority of existing AR studies focus on the antecedents having a positive influence on AR users during online shopping (Table 1). While AR has achieved significant diffusion in the retail industry, further acceptance and adoption of AR technologies will depend not only on identifying and addressing these barriers but also on countering negative influences.

Our study provides various practical implications. First, AR-enabled applications positively influence both behavioral intention and WOM among consumers who gain an immersive experience during online shopping. This supports existing AR studies suggesting that AR is more than a fad and will play a key role in enhancing consumer online shopping behavior (Erdmann et al., 2023; McLean and Wilson, 2019). Second, it is important for AR developers and managers to design advertising campaigns to show how AR can provide ease of use, informativeness and other benefits. Third, the term ‘consumer’ is used broadly for all persons using AR, including gamers, students and tourists (Rauschnabel et al., 2024). However, it is critical for marketing practitioners to remain sensitive to occupational differences in the target customer segment. For students, AR developers should prioritize the perceived enjoyment and aesthetics of their AR-based shopping environment, along with informativeness. For working professionals, AR developers should focus on the perceived usefulness attribute in addition to informativeness. Fourth, AR managers should look at the value and risk barriers affecting AR adoption by customers. There have been few studies looking at those reasons preventing AR adoption by customers. However, these available studies have mainly explored different barriers as a single innovation resistance attribute rather than individual resistance factors (Ronaghi and Ronaghi, 2022). Moreover, this study also suggests that risk and value barriers need more attention than usage and image barriers.

Though our study provides several academic contributions and managerial insights, like all other studies, this study has a few limitations that can be addressed in future studies (Table 6). First, we considered the moderating effect of only one variable, occupation (students versus working professionals), in our model. Future studies can include the moderating impact of product type (such as beauty products, sports, electronics and others) for a better understanding of this phenomenon. Second, we used a survey-based approach to measure consumers’ intention to use AR. Studies using experimental research designs would be able to capture customers’ actual purchase behavior to complement the present study.

Third, future studies can include the mediating role of customer experience and empowerment to better understand how AR deployment leads to positive WOM intention. Fourth, extant consumer behavior literature confirmed that culture impacts how consumers adopt technologies and perceive technology integration in services. Additional cross-cultural studies on AR deployment could yield more substantial insight and provide recommendations for context-specific AR deployment to multinational companies.

Finally, while our study focused on mobile AR, there is scope for validating this study in other types of AR, such as web AR, AR glasses and smart mirrors (Barta et al., 2025).

This study attempted to examine the adoption of AR-enabled mobile apps by focusing on positive and negative factors that impact users’ intention to use and promote those apps. AR, due to its strongly interactive and informative nature, is transforming the high-involvement e-commerce industry. Our study is one of the first to integrate both positive and negative antecedents of AR adoption in the context of usage and WOM intentions. We also found that occupation (students vs working professionals) moderates these relationships, such as working professionals placing more importance on the value and risk barriers compared to the students. The study shows that to enhance AR adoption, managers, in addition to nurturing the positive factors, must also identify and address the barriers. Our study contributes to marketing literature by developing and validating the framework in the AR context, which can be used in investigating the adoption of other technologies, such as virtual reality and mixed reality. We have also investigated the moderating impact of demographic variables, which can provide greater insight to the managers. Table 7 summarizes the conclusion, with its theoretical and managerial implications.