Shivendra Singh Chaudhary*
| Diksha Panwar | Gautam Srivastava
© 2026 The authors. This article is published by IIETA and is licensed under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/).
OPEN ACCESS
The emergence of Over-the-top (OTT) streaming platforms has significantly revolutionized modern media consumption trends, and AI-based personalization is now one of the primary frameworks driving user experience and engagement behaviours. Though personalization technologies have been widely studied from both marketing and platform performance viewpoints, they are less explored through the lens of how a platform can emphasize responsible or sustainable digital media ecosystems. This research examines the effect of AI-based personalization on consumer behaviour in OTT platforms, with brand engagement as a primary explanatory construct and emotional fulfilment as a mediating mechanism. Data were collected using a structured online questionnaire from 350 Gen Z OTT users residing in different parts of India, and the hypothesized relationships were tested through Structural Equation Modelling (SEM). The results show that AI-based personalization positively influences brand engagement, which in turn has a positive effect on consumer behavioural tendencies such as binge-watching patterns, reliance on the platform, and time spent on streaming platforms. The results also show that emotional fulfillment partially mediates the relationship between brand engagement and consumer behaviour, illustrating the psychological pathways through which customization strengthens user–platform relationships. In addition to providing behavioral insights, the study also contributes to emerging discussions on sustainable media ecosystems in the digital context by offering an initial field-level understanding of the balance between engagement optimization and user well-being, transparency, and responsible personalization practices. The results have theoretical and practical implications for OTT platform designers, policymakers, and digital ecosystem planners who aim to develop ethically grounded and sustainable approaches to personalization in algorithmically mediated media environments.
artificial intelligence, brand engagement, consumer behavior, emotional satisfaction, Over-the-top platforms, personalization, sustainable digital media ecosystems
The world's entertainment industry is undergoing a significant revolution, driven by the rapid rise of Over-the-top (OTT) streaming services. Whether it is global players (Amazon Prime Video, Netflix, Disney+, etc.) or popular regional players (iQIYI, Viu, Hotstar, etc.), these platforms have completely disrupted how audiences discover, consume, and engage with content. OTT services, by delivering media directly via the internet rather than through traditional distribution infrastructure such as satellite and cable, provide unprecedented levels of flexibility, personalization, and portability. The global OTT market was worth more than USD 300 billion in 2024, and it is expected to reach around USD 550 billion by 2028, indicating that digital streaming has become part of everyday life [1].
The explosion of OTT is not just a by-product of fit-for-purpose technology; it represents a deep-seated cultural change in media consumption behavior. Viewers no longer have to wait for prime-time or scheduled content. Instead, they demand content on their own terms—on any device, anytime, and anywhere. This demand has led to innovations not only in content generation but also in platform intelligence, which governs how content is delivered and curated for users. In this context, artificial intelligence (AI), particularly in personalizing user experience, has emerged as a central force.
The use of AI in OTT platforms marks a fundamental shift from previous models of content dissemination. Earlier, a one-to-many model was employed by broadcasters, with minimal feedback mechanisms. Now, users receive customized interfaces, as OTT services operate on a many-to-one model, driven by algorithms that evolve and learn with every interaction. AI-driven personalization utilizes machine learning, real-time behavioral analytics, and collaborative filtering to track a wide range of user data—such as scrolling patterns, watch history, device switching, temporal engagement, and genre preferences—to generate content recommendations [2, 3]. These platforms are not only engines of suggestions; they are architects of behavior, guiding users toward longer sessions, more frequent visits, and deeper emotional immersion.
The relationship between the user and the platform has shifted from that of an interface that facilitates transactions to a dynamic, engagement-led ecosystem. Audiences are no longer just passive consumers of content—they are part of a customized narrative using AI tools. Thus, it is not merely about simplifying choice, but about how behavior is shaped by algorithmic curation. Users experience less decision fatigue, and their content completion rates and binge-watching frequency increase when they are exposed to personalized recommendations [4, 5]. The psychological dimension of this relationship is particularly interesting. Anthropomorphism of AI [6]—where users tend to perceive the system as “rational” or even emotionally adaptive—plays a key role. As the system becomes more familiar with users, they tend to develop emotional dependence and trust, leading to long-term loyalty and engagement. The impact of branding in the context of personalization is also significant. Various platforms compete with overlapping content libraries and similar price points in today’s rapidly shifting digital landscape, where user experience has become the key distinguishing factor. Personalization itself functions as a brand, based on attention, sophistication, and emotion [7]. Platforms are often perceived as user-centric, responsive, and modern when they offer intuitive and personalized journeys. This influences long-term subscription retention, brand trust, and the likelihood of recommendations. In this context, personalization becomes a strategic mechanism for content distribution, as well as for brand co-creation and identity formation.
Although AI-enabled personalization is becoming increasingly central to modern media consumption, existing literature has largely focused on recommendation systems from the perspective of technical performance or general user experience models based on overall satisfaction. More specifically, across OTT ecosystems—especially in emerging digital markets such as India—there is limited empirical research that simultaneously investigates personalization as a behavioral influence lever operating through brand engagement and emotional satisfaction [8]. In addition, previous studies have tended to frame personalization as a feature that improves usability rather than as a relational process shaping ongoing platform–user interactions. The current study aims to bridge this gap by investigating a structural model to understand the mechanism through which personalization affects consumer behavioral tendencies among Gen Z OTT users via engagement-based and emotional pathways. By situating personalization within broader discussions of responsible platform engagement and sustainable digital media environments, this study extends research beyond the technical performance of recommendation systems toward a behavioral ecosystem approach, particularly relevant in rapidly evolving, algorithmically mediated media environments.
For example, personalization may not precisely predict the preferences of users in markets like Indonesia, India, Nigeria, or Brazil, as these systems are often trained on Western user behaviour, whereas content consumption habits in these regions are shaped by different cultural, technological, and linguistic realities. Users primarily access OTT platforms through mobile devices in these regions, frequently with intermittent connectivity and limited data plans. Their preferences for genres do not align with globalized recommendation models; instead, they reflect traditional, religious, or community-centred themes. Moreover, issues such as trust in algorithmic systems and digital literacy vary widely across markets, influencing how personalization is interpreted and valued [9]. Nonetheless, mainstream OTT research remains limited in its perspective, leading to a one-size-fits-all belief that personalization is universally effective and beneficial.
In addition, the approaches to personalization are not neutral. The closed feedback loops based on AI models can further increase behavioral expectedness at the expense of content diversity. Engagement might be favored by the recommender systems at the expense of educational and cultural value, potentially leading to algorithmic echo chambers. There is a growing sense of anxiety when it comes to dealing with these in the post-digital realm, such as digitized well-being or platform ethics, yet they are rarely incorporated into branding and media studies. This gap signals the requirement for more interdisciplinary approaches targeting dimensionalities such as cultural, psychological and sociotechnical layers of personalization in content delivery.
How emerging users interpret personalization is another neglected aspect. OTT platforms’ ability to predict preferences may feel manipulative or intrusive to some users, particularly when recommendation systems present sensitive or emotionally charged content. Others may feel overwhelmed by the illusion of choice, where personalization creates a curated bubble that limits exposure to diverse content. These concerns point to the need for research that extends beyond metrics to examine user perceptions, emotions, and cultural expectations in response to personalization.
Thus, this study fills these gaps by investigating the impact of AI-enhanced personalization on two key dimensions of OTT ecosystems—brand engagement and user behaviour—as well as considering the broader implications of personalization in emerging sustainable digital media environments. From concerns about digital well-being, algorithmic transparency, behavioural dependence to responsible design of platforms, personalization technologies are increasingly influencing media consumption and are the focus of critical discussions — all salient components in planning for sustainable digital ecosystems. The study not only contributes to the knowledge of user–platform relationships, but also to broader interdisciplinary dialogues regarding responsible personalization and sustainable digital media infrastructure by investigating psychological mechanisms through which personalization leads Gen Z OTT users in India toward engagement and behavioural tendencies. As such, the findings provide guidelines for platform designers, digital ecosystem planners and policymakers looking to balance engagement optimization with ethical and user-centered strategies for personalization in a rapidly changing algorithmically driven media ecosystem.
A massive transformation has occurred in the global entertainment and media industry with the rapid expansion of OTT streaming platforms such as Amazon Prime, Netflix, and a few regional players like iQIYI and Jio Hotstar. On-demand personalized content is delivered by these platforms, bypassing traditional broadcasting channels and providing greater flexibility and control over users’ viewing experiences [1]. This transformation is driven by the integration of AI, specifically in the form of recommendation systems that use user data to curate individualized content feeds. AI-powered personalization has evolved beyond being a convenience feature to become a significant branding and behavioural mechanism. It improves user satisfaction by reducing decision fatigue [10, 11], fosters emotional connection through the perceived intelligence of platforms [6], and encourages repeated engagement, particularly among younger audiences [4, 12]. Furthermore, personalization functions as a strategic branding tool, enhancing brand perception and loyalty by delivering seamless, emotionally resonant user experiences [7, 13].
Although, despite its rising influence, current academic literature has particularly focused on the technical prowess of recommendation engines or, superficially, on user satisfaction, with little attention given to the underlying emotional and behavioural impacts of personalization [14, 15]. Additionally, a large part of the existing research is entrenched in Western contexts, overlooking the linguistic, infrastructural, and cultural variables that govern user responses in regions such as Southeast Asia, Latin America, and India [9, 16]. The limitations of one-size-fits-all models of personalization are emphasized by variations in viewing habits, cultural differences in content preferences, and technological accessibility [17]. The present study responds to these gaps by examining the influence of AI-driven personalization in OTT platforms on both brand engagement and consumer behaviour, and how these relationships vary across cultural settings.
By adopting an interdisciplinary lens that draws from human–computer interaction, behavioural psychology, media studies, and digital marketing, this study aims to develop an integrated framework for understanding personalization not merely as a technical system but as a behavioural, cultural, and branding phenomenon. It connects personalization’s algorithmic design with its experiential and symbolic dimensions, thereby extending beyond existing literature. It also emphasizes the significance of context-conscious personalization models, which adapt not only to individual preferences but also to broader sociocultural conditions.
Finally, this research contributes to the emerging discussion on AI in consumer-facing technologies by providing insights not only into the process of personalization but also into its implications for platform design, user empowerment, digital equity, and strategic branding. Algorithmically curated, emotionally intelligent media consumption is transforming not only marketers and platform developers but also policymakers, consumers, and scholars alike. Moreover, this study seeks to enhance both practical applications and academic understanding within the evolving OTT landscape.
3.1 Theoretical framework
A strong interdisciplinary theoretical framework forms the basis of this study, which explores the branding, cultural, ethical, and behavioral dimensions of AI-driven personalization in OTT platforms. Platform anthropomorphism is a central concept, where users perceive AI-based recommendation systems as empathetic, emotionally attuned, or intelligent, thereby promoting trust and psychological intimacy [6, 18]. These perceptions influence increased user engagement and emotional reliance, frequently manifesting in higher platform loyalty and longer viewing sessions [4, 12]. Self-determination theory aligns with this perspective, explaining how personalization enhances user satisfaction and perceived autonomy, while concurrently risking user overreliance and algorithmic dependency on AI for content choices [19, 20].
Digital identity theory is also incorporated into the framework, highlighting the role of mood-based recommendations and personalized dashboards in shaping users’ self-concepts and emotional states, presenting personalization as a means of reinforcing identity [21, 22]. From a branding perspective, brand engagement theory positions personalization as an instrument that signals care, responsiveness, and intelligence. Attributes such as context-aware interfaces, gamified experiences, and interactive banners help improve perceived brand value and emotional attachment [7, 13, 23, 24].
A significant addition to the framework is the theory of cultural dimensions, which underlines how personalization must account for cultural and regional variations in user behavior, expectations, and values. For example, in collectivist societies with high-context communication styles, such as those in Southeast Asia or India, content preferences are shaped by linguistic plurality, continuity viewing norms, and cultural rituals—factors often overlooked by AI systems trained in Western contexts [9, 16, 25]. Adaptation of algorithmic models through geo-personalization to align with local cultural contexts can enhance the effectiveness and relevance of personalization in light of these phenomena [17, 26].
AI ethics are addressed in the final component of the framework, focusing on user autonomy, algorithmic fairness, and transparency. There are concerns regarding algorithmic manipulation and informational asymmetry, as most users are unaware of how their data are used by these platforms [27, 28]. The mechanisms of AI systems often remain hidden from users, a phenomenon referred to as algorithmic opacity, which can reduce trust in and perceived authenticity of the platform [29]. Many scholars have called for the integration of ethics-by-design principles and explainable AI (XAI) to ensure that personalization systems remain transparent, inclusive, and fair [30, 31].
Taken together, these theoretical foundations position AI-driven personalization not only as a technological advancement but also as a complex psychological and sociocultural construct. This framework supports the development of hypotheses that examine the impact of personalization on behavior, engagement, brand loyalty, and trust, while also considering the mediating roles of emotional satisfaction, ethical perception, and culture.
3.1.1 AI personalization and sustainable digital media ecosystems
Apart from its psychological and branding aspects, AI-augmented personalization has begun to be discussed as part of sustainable digital media ecosystems. Sustainable digital platforms are expected not only to optimize engagement outcomes but also to foster responsible consumption patterns, user well-being, and transparent algorithmic decision-making processes. Although personalization systems have been proven capable of enhancing content discovery and emotional engagement with platforms, they may also contribute to normalized extended viewing habits, behavioral addiction, and visibility bias if not carefully designed.
Sustainable personalization therefore reflects a socio-technical balance, maintaining algorithmic efficiency while enabling user agency by making algorithms more fair and transparent in their recommendation processes. Emerging interdisciplinary conversations underscore the need for XAI systems, ethics-by-design personalization architectures, and user-managed recommendation settings that allow users to control their levels of interaction. These aspects are particularly crucial in OTT ecosystems, where increasing levels of personalization technologies influence everyday patterns of how users interact with media and information across diverse cultural and techno-social contexts.
While the current study offers an empirical investigation into personalization primarily in light of its effects on brand engagement, emotional gratification, and consumer behavioral dispositions, these insights also contribute to emerging discussions on responsible forms of personalized services that promote the long-term sustainability of digital media environments by advocating user-centric and ethically grounded platform design paradigms.
3.2 Research model, hypothesis, and literature support
3.2.1 AI personalization and brand engagement in OTT platforms
There is a shift in AI-driven personalization from a utility-based feature to a strategic pillar of branding in OTT platforms. OTT platforms currently rely on personalized user experiences, as content libraries are becoming increasingly identical, thereby building emotional bonds, differentiating their value propositions, and increasing brand stickiness. Personalization functions as an instrument through which brand personality is co-created and perceived, with recommendation systems tailoring content in a way that appears intelligent or empathetic [7, 32].
A study by Obiegbu and Larsen [18] indicates that users perceive adaptive, seamless content recommendations as signals of a platform’s emotional intelligence and competence. This, in turn, leads to higher trust, long-term loyalty, and brand relevance. Likewise, Obiegbu and Larsen [18] found that personalization features, such as interactive content banners and mood-based playlists, not only enhance user experience but also strengthen users’ affective engagement and attentiveness to the brand.
An additional level of branding impact results from gamification-led personalization. The way OTT platforms employ features such as time-bound challenges, customized congratulatory prompts (“You have completed a season”), and watch streaks has been documented by Hamari et al. [23], which promotes emotional closeness and reinforces a sense of personal recognition. These small interactions strengthen connections and increase emotional salience by enhancing engagement at the brand level.
Additionally, personalization socially has gained a lot of attention. Features like the collaborative playlists, peer-based recommendation algorithms and shared viewing modes not only improve viewer satisfaction but also accelerate brand diffusion within many social networks [33]. The argument [34] states that many features of OTT platforms transform users into co-creators of brands, successfully converting the experience of OTT into a social performance act of brand advocacy.
Platform consistency and interactivity affect brand engagement, where personalization guarantees efficient steering across sessions and devices. Platform uniformity is improved by personalization resulting in increased trust and retention of users as found out by Ajith et al. [7], specifically in multiple screen locations.
Nevertheless, the efficiency of these strategies is subjected to perceived authenticity. It was warned by Aguirre et al. [35] that hyper-personalization can stimulate a “creepiness effect”, where viewers usually feel manipulated or over-surveyed. Perceived overstretch can reduce emotional engagement and raise brand switching trends as warned by Bleier and Eisenbeiss [36]. The fine line between interfering targeting and empathetic curation expects brands to pace cautiously in strategizing their personalization.
Regardless of insightful qualitative findings, there is a considerable absence of experimental studies that fundamentally model the link between brand engagement and personalization, specifically in relation to advocacy, loyalty, trust and emotional attachment. Rare findings include multi-dimensional variables of user engagement and experiment them quantitatively using an incorporated models like the SEM. This bounds the accuracy with which personalization’s branding results can be optimized and measured.
H1: There is a positive impact of personalization on brand engagement in OTT platforms.
3.2.2 Brand engagement and consumer behaviour in OTT environments
There is a transition of brand engagement from being a marginal brand metric to a fundamental construct shaping user behaviour, in the progressing digital ecosystem of OTT platforms. Users’ behavioural tendencies like platform loyalty, advocacy, viewing time and binge-watching tendencies are directly influenced by the depth and nature of their brand engagement as their experiences are increasing due to personalized interactions and expressively resonant platform features. In digital environments brand engagement includes behavioural, cognitive and emotional dimensions that are triggered in the context of OTT platforms where the viewers interact frequently with socially combined features, personalized interfaces and gamified content as put forth by Hollebeek et al. [32].
The relationship between consumer behaviour and brand engagement is reinforced by perceived interpersonal quality and psychological engagement between the streaming platform and the user. Personalization enhances this engagement, according to Ajith et al. [7], leading users to perceive that they are in an active relationship with the platform. This emotional connection fosters trust in the brand and subsequently influences consumer behaviours such as repeated visits, word-of-mouth promotion, and platform loyalty. Thus, engagement becomes both a stimulus and a predictor of viewer actions. According to Yaprak [5] and Cohn et al. [6], when users feel emotionally connected to an OTT platform—perceiving that it understands, respects, and reflects their preferences—they exhibit stronger usage patterns and reduced tendencies to churn.
Furthermore, in this scenario, consumer behaviour is not simply reactive but rather co-creative. Users quite often personalize their experiences further via participations in various platform features like episode discussions or polls, creation of playlists and sharing recommendations with peers, according to studies by Muniz Jr. and O’Guinn [34] and Stephen and Toubia [33]. Participative and affective engagement is indicated by these behaviours, muddling the lines between user agency and brand consumption. In return, this level of interactivity strengthens the user’s loyalty to the platforms, extending behavioural loyalty and emotional security.
Notably, engagement generated by various branding strategies must be perceived as genuine and relevant. A vital role is played by emotional authenticity in converting engagement into persistent outcomes of behaviours. Further signalled by Aguirre et al. [35], cues from brands like interactive features or recommendation prompts appear too programmed or commercially interfering, they can cause conflict or behavioural withdrawal. Therefore, consumer behaviour is dependent not just on the availability of personalization but also on the emotional intelligence and perceived sincerity of the platform’s strategies of engagement [18].
Younger cohorts and digital natives are the main beneficiaries of behavioural outcomes of brand engagement. They ask for self-expression and regulation of mood through various media choices according to Sellos et al. [12], and the OTT platforms which offer adapted and emotionally reactive journeys, become connected to their daily routines. This behaviour is recurring – Greater engagement leads to more repeated use, which subsequently improves the behavioural attraction towards the brand [32]. So, consumer behaviour becomes a functioning demonstration of brand loyalty rather than a reactive response to content availability.
Still there is a lack of modelled and holistic approaches to analytically test the relationship regardless of the growing indication linking brand engagement to user behaviour in OTT platforms. Instruments of survey or qualitative observations that are fragmented are usually employed by most research today. There is a need for Structural Equation Modelling (SEM) led analyses that tests the causal paths with multi-dimensional concepts like trust, behavioural loyalty, emotional involvement and platform satisfaction. SEM’s capability to take out the complex relationships between digital behavioural outcomes in the context of media and emotional brand engagement is underscored in the studies by Chan-Olmsted and Luo [37].
Therefore, the relationship between consumer behaviour and brand engagement is not only theoretically strong rather it is significant practically too. Streaming services that can adopt authentic emotional associations through a personalized brand experience are better placed to shape consumer behaviour in loyalty led and sustainable ways.
H2: Brand engagement and consumer behaviour has a positive relationship in OTT platforms.
3.2.3 Emotional satisfaction as a mediating mechanism in personalized media environments
AI personalization nurtures emotional significance, where psychological closeness is developed by the users for the platform. This is termed as platform anthropomorphism by Cohn et al. [6], where users recognize the system as emotionally attuned and intelligent to their preferences. Platform dependency is often increased, resulting in trust-based relationships. OTT platforms suggesting emotionally contextual and personalized recommendations as concluded by Studies by Sellos et al. [12] in their research, assists in prolonged binge-watching sessions and frequent use patterns, specifically among younger digital natives.
An identity reinforcement is implied by an immersed psychological layer of personalization. Users recognize their recommendation feeds and dashboards as expansions of their digital identity, according to Belk [21] and Muniz Jr. and O’Guinn [34]. This is predominantly clear in OTT services that recommend categories or shows centred on mood states, social triggers, or watch history, directing users to relate consumption of content with mood regulation and self-perception.
Nevertheless, the behavioural consequences are not completely positive. Too much reliance on AI-led recommendations promotes dependency on algorithms as cautioned by Deci and Ryan [19] and Bucher [20], where users abandon independent choice in preference of convenience. These patterns build content silos – algorithmic bubbles that strengthen persistent consumption while suppressing cultural exploration or novelty, as argued by Rzepka and Berger [14]. OTT platform’s ability to act as an instrument for personal growth and discovery is dynamically reduced.
Personalization may magnify binge-watching behavior, as explored by an expanding number of studies, by fostering continuous engagement through emotional triggers, minimal content friction, and autoplay features [37]. Although this behavior enhances platform retention metrics, it may concurrently reduce user well-being, further complicating the ethical assessment of AI-driven personalization in entertainment.
Most empirical studies examine personalization and user behavior in isolation, often focusing on correlations rather than causation. Few studies employ integrated structural approaches or behavioral models—such as SEM—to examine causal pathways between emotional satisfaction, consumer engagement outcomes, and AI-driven personalization factors such as trust, loyalty intentions, and watch time. This body of work reveals both methodological and theoretical shortcomings, particularly regarding the role of personalization in shaping both immediate and long-term user behavior.
H3: Brand engagement and consumer behaviour in OTT platforms is positively mediated by emotional satisfaction.
3.2.4 Personalization within emerging sustainable digital media contexts
Though personalization technologies are broadly understood in terms of their role in optimizing engagement and behaviour outcomes on OTT platforms, cross-disciplinary conversations have recently placed algorithmic recommendation systems within a larger ecological framework toward sustainable digital media environments. As such, personalization systems shape viewing routines, diversity of exposure and emotional patterns of interaction — all of which are key parts in the design of long-term platform–user relationships. Sustainable Personalization in this context is about balancing engagement optimization with user agency, transparency of recommendation processes, and responsible handling of behavioral data across algorithmically mediated platforms. While the constructs relating to sustainability were not operationalised in a direct sense with regard to the present empirical model, positioning personalization within this broader digital ecosystem perspective is also relevant for understanding how engagement-driven recommendation environments can further or detract from responsible and user-centred platform development (Figure 1).
Figure 1. Conceptual framework
4.1 Research approach, research method, and survey design
As it is adaptable to a digital environment (e.g., OTT streaming), an online survey is chosen as the method for data collection in this study. Online survey is an appropriate method for research on OTT users because both the online platform and OTTs are digital in nature which can bring better reach to diverse population across locations [38]. They are cost-efficient, scale-able and can gain access to participants in real-time, which is very important for a PAN India study where physical data collection would not be viable [39]. In addition, online surveys were commonly adopted in previous OTT research to examine user behavior and engagement as this method of survey allows standardized evaluations while eliminating interviewer-based bias [37]. Since OTT users are used to and familiar with digital interaction, this approach ensures a higher response rate and better data quality for investigating the effects of AI-driven personalization.
Regional segmentation of media consumption is a necessary consideration since the exposure and usage experience among OTT users can vary widely based on differential access to language, culture, and technology in India. Nevertheless, as the study uses a non-probability snowball sampling method, the results should not be interpreted to represent a national picture but rather suggest behavioural tendencies of the sampled population. The Indian OTT market is among the fastest growing across the globe, with more than 500 million users by 2023, enhanced by the availability of cheap data plans, and penetration of mobile phones [1]. The PAN India strategy is appropriate, given the diverse landscape of the market in terms of regional content preference and varying levels of digital literacy or infrastructure [9]. For example, audiences in the southern states might prefer regional language content and those in northern states may incline toward Bollywood-centric content, underscoring the need for a wide coverage of sample to generalize findings [17]. Studies investigating the adoption of OTT in India have been effectively able to make use of PAN India samples to address this diversity, for representativeness and external validity [16].
For the study data will be collected using a structured online questionnaire, which was further divided into four sections: (1) demographic information (e.g., age, gender, region, OTT usage frequency); (2) personalization features (e.g., satisfaction with recommendations, perceived relevance, measured on a 5-point Likert scale adapted from Ajith et al. [7]; (3) brand engagement (e.g., trust, loyalty, emotional resonance, using scales from Obiegbu and Larsen [18]; and (4) consumer behaviour and emotional satisfaction (e.g., binge-watching frequency, platform dependency, emotional connection, adapted from Yin [4].
4.2 Sample, sampling strategy process, and data collection
The sample will target Gen Z (1997–2012) as they are the majority of OTT users in India, around 70% of user base [1]. This generation had been brought up as ‘digital natives’ and frequently used personalized digital experiences which made them an optimal target for the study of AI-based personalization impact [12]. Gen Z have unique OTT consumption patterns including increased propensity for binge-watching, and mood-based content preference, which is consistent with emotional satisfaction focused consumer behavior in the present study [4]. Their lead in India’s OTT space establishes that the sample is representative of the research objectives, consistent with previous study calls for this population to be used in digital media research [5].
This study will use a nonprobability sampling, exclusively snowball sampling. Snowball sampling is used as this sampling technique is appropriate for reaching diverse and geographically spread OTT users, as it influences initial contacts to attract other participants who, refer others within their networks [39]. Snowball sampling has been successfully employed in previous digital media research for some other studies in India to ensure broad scope of users’ views, especially while focusing on a specific demographic for instance, OTT users [17]. The sample size will be 350 respondents, as recommended for SEM studies in order to ensure statistical power [38].
Data will be collected using Google Form, over a 4-week period, a widely used tool for online surveys in academic research [38]. The survey link will be distributed to the participants through different social media platforms. This approach makes sure efficient reach to Gen Z, who are active on these platforms, and data collection gets easy [16]. Responses will be screened and obtained data will be cleaned to remove duplicates or inconsistent entries.
4.3 Research instrument, measures, and measurement variables
The proposed model consisted of four variables: Personalized Features, Brand Engagement, Consumer Behavior and Emotional Satisfaction. All items were rated using a 5-point scale, ranging from “strongly disagree” (1) to “strongly agree” (5), and adapted for this study. Variables were created as mean scores of all individual items for further analyses (Table 1). Each construct from the study has been operationalized using three-item measurement scales adapted from previous validated studies in both OTT personalization and digital engagement literature. Three-item scales were used for model parsimony and reliability of respondent completion in an online survey environment while ensuring consistency with SEM practice, particularly in behavioral platform research where reflective measurement structures are typically short.
Table 1. Measurement and questionnaire items and level of measurement
|
Constructs |
Measurement Items |
Variables |
Level of Measurement |
Questionnaire Items |
Ref. |
|
Personalized Features |
PF1 |
PF |
Interval (1 = Strongly Diasgree, 5 = Strongly Agree) |
The platform’s recommendations are relevant to my preferences. |
[7] |
|
PF2 |
Personalized features (e.g., recommendations, playlists) make my viewing experience enjoyable. |
||||
|
PF3 |
The platform effectively suggests content based on my watch history. |
||||
|
Brand Engagement |
BE1 |
BE |
Interval (1 = Strongly Diasgree, 5 = Strongly Agree) |
Using OTT platform makes me feel very involved with the brand. |
[32] |
|
BE2 |
I spend a lot of time thinking about OTT platform and its content. |
||||
|
BE3 |
I actively participate with the OTT brand through its app or social media features. |
||||
|
Consumer Behaviour |
CB1 |
CB |
Interval (1 = Strongly Diasgree, 5 = Strongly Agree) |
I often binge-watch shows recommended by the platform |
[4] |
|
CB2 |
I rely on the platform to discover new content |
||||
|
CB3 |
I spend more time on the platform because of its recommendations. |
||||
|
Emotional Satisfaction |
ES1 |
ES |
Interval (1 = Strongly Diasgree, 5 = Strongly Agree) |
I feel emotionally connected to the platform because of its personalized features. |
|
|
ES2 |
The platform seems to understand my preferences, making me feel valued. |
|
|||
|
ES3 |
I feel a sense of comfort when using this platform due to its personalization. |
|
5.1 Sample characteristics and descriptive statistics
Out of the 350 participants, the average age of the participants was between 22 and 25 years (SD = 37%), with females (58%) being more numerous than males (42%). Almost 81% participants held graduation degree or were at the stage of completing graduation. About 54% of the participants are college students.
5.2 Interpretation of questionnaire results
The descriptive statistics (means and standard deviations, factor loading, CR and AVE) of each questionnaire item are included in Table 2.
Table 2. Factor loadings, and internal consistency
|
Constructs |
Measurement Items |
Mean |
SD |
Factor Loading |
CR |
AVE |
|
Personalized Features |
PF1 |
3.0945 |
1.4527 |
0.712 |
0.853 |
0.66 |
|
PF2 |
3.4878 |
1.2488 |
0.62 |
|||
|
PF3 |
3.4421 |
1.2861 |
0.647 |
|||
|
Brand Engagement |
BE1 |
3.4726 |
1.3107 |
0.572 |
0.819 |
0.601 |
|
BE2 |
3.5579 |
1.2026 |
0.611 |
|||
|
BE3 |
3.4573 |
1.3079 |
0.62 |
|||
|
Consumer Behaviour |
CB1 |
3.5579 |
1.2597 |
0.66 |
0.87 |
0.69 |
|
CB2 |
3.4085 |
1.3447 |
0.701 |
|||
|
CB3 |
3.4695 |
1.2823 |
0.707 |
|||
|
Emotional Satisfaction |
ES1 |
3.5244 |
1.3316 |
0.676 |
0.861 |
0.674 |
|
ES2 |
3.4543 |
1.2747 |
0.695 |
|||
|
ES3 |
3.4329 |
1.2273 |
0.652 |
5.3 Confirmatory factor analysis
With the help of AMOS 25, confirmatory factor analysis (CFA) was performed [40]. First, all standardized factor loadings of the variables in the measurement model were significant, and the composite reliabilities (CR) value of all the construct exceeded .70, which confirm convergent validity [38] (see Table 2). Furthermore, the average variance extracted (AVE) for all constructs was above 0.50, and lastly, for the discriminant validity, the square root of AVE of each construct's should be greater than the construct's inter-correlations with other model constructs [41]. Table 3 in this analysis illustrates the discriminant analysis.
Table 3. Discriminant validity
|
|
PF |
BE |
CB |
ES |
|
PF |
0.812 |
0.351 |
0.416 |
0.604 |
|
BE |
0.351 |
0.775 |
0.424 |
0.296 |
|
CB |
0.416 |
0.424 |
0.83 |
0.519 |
|
ES |
0.604 |
0.296 |
0.519 |
0.821 |
Discriminant validity is supported because each construct’s √AVE (diagonal) is higher than its correlations with other constructs.
5.4 Hypotheses testing
To test the proposed hypotheses of the study, SEM was conducted using a maximum likelihood estimation. Three variables (i.e., Personalized features, Brand Engagement and consumer behavior) were included in the model. The hypothesized model showed a good model fit: χ²(25) = 46.207, p = 0.005, χ²/df = 1.848, CFI = 0.985, TLI = 0.979, and RMSEA = 0.049.
H1 projected that personalized features has a positive impact on brand engagement. The analysis shows that Personalized Feature to Brand Engagement was positive and statistically significant (β = 0.298, SE = 0.051, CR = 5.826, p < 0.001), indicating that higher levels of personalized features are associated with greater brand engagement. Therefore, H1 was supported.
H2 projected that brand engagement has positive relation with consumer behavior in OTT platforms. As per the analysis, Brand Engagement showed a positive and significant direct relationship with Consumer Behavior (β = 0.261, SE = 0.047, CR = 5.523, p < -0.001). This suggests brand engagement helps to translate into more favorable consumer behavior. Hence, H2 was supported.
Figure 2. Results of structural equation modeling (SEM)
H3 proposed that emotional satisfaction positively mediates the relationship between brand engagement and consumer behavior. For mediation, both component paths were significant: Brand Engagement positively forecast Emotional Satisfaction (β = 0.245, SE = 0.052, CR = 4.731, p < 0.001), and Emotional Satisfaction positively forecast Consumer Behavior (β = 0.384, SE = 0.047, CR = 8.110, p < 0.001). Coherent with this pattern, the indirect effect of brand engagement on consumer behavior via emotional satisfaction was positive (indirect effect = 0.094). The total effect of brand engagement on consumer behavior was .356, consisting of a direct effect of .261 and an indirect effect of .094. As the direct effect remained significant after including emotional satisfaction, the results indicate partial mediation (see Table 4). Therefore, H3 was supported. Figure 2 illustrates the SEM results, showing standardized coefficients for the hypothesized paths and their statistical significance.
Table 4. Mediation analysis
|
Mediation Pathway |
Indirect Effects |
Total Effects |
|
Personalized Features → Brand Engagement → Emotional Satisfaction |
0.073 |
0.073 |
|
Personalized Features → Brand Engagement → Consumer Behavior |
0.106 |
0.106 |
|
Brand Engagement → Emotional Satisfaction → Consumer Behavior |
0.094 |
0.356 |
5.5 Summary of research findings
The hypothesis testing (presented in Table 5) shows that the proposed model received good support. Both H1 and H2 were supported, indicating that personalized features have a significant impact on brand engagement, and that increased engagement with the brand further positively affects consumer behaviors on OTT platforms. The results indicate that the greater the level of personalized features offered through OTT platforms, the more likely consumers are to interact with and feel involved with a brand, leading them to engage in desirable behaviors as well.
Table 5. Summary of research results
|
|
|
|
Estimate |
P |
Result |
|
|
H1 |
Personalized Feature |
<--- |
Brand Engagement |
0.298 |
*** |
Supported |
|
H2 |
Brand Engagement |
<--- |
Consumer Behavior |
0.261 |
*** |
Supported |
|
H3(a) |
Brand Engagement |
<--- |
Emotional Satisfaction |
0.245 |
*** |
Supported |
|
H3(b) |
Emotional Satisfaction |
<--- |
Consumer Behavior |
0.384 |
*** |
Supported |
Moreover, H3 was supported, indicating that emotional satisfaction mediates the relationship between brand engagement and consumer behavior. More specifically, higher brand engagement leads to enhanced consumer behavioral outcomes through increased emotional satisfaction. Since the direct effect of brand engagement on consumer behavior remained significant even after including emotional satisfaction in the model, the findings suggest partial mediation. Taken together, the results show that both the direct effect of brand engagement and the indirect emotional pathway through satisfaction are significant in explaining consumer behavior in OTT contexts.
6.1 Degree of addressing to the research objectives and supporting to the initial assumptions
With OTT services witnessing rapidly expanding and highly competitive market conditions, and platform-driven personalization serving as a critical strategy to retain consumers deeply immersed in platforms, this research attempts to deepen understanding of how personalized features impact consumer behavior within the context of OTT platforms, with a focus on brand engagement and emotional satisfaction as two important explanatory mechanisms. Consistent with H1 and H2, this study addresses its research objectives by examining whether personalized OTT experiences lead to higher engagement with the platform brand and whether such engagement results in positive consumer behaviors. The results confirm the proposed hypotheses that personalized attributes have a significant and positive effect on brand engagement (H1) and that brand engagement has a significant and positive impact on consumer behavior (H2). The findings imply that the more users perceive OTT platforms as providing content aligned with their personal preferences, the more likely they are to develop higher levels of involvement and engagement, thereby reinforcing stronger behavioural responses.
H3 that emotional satisfaction is a positive mediator between brand engagement and consumer behavior was supported. This suggests that high engagement with the brand provides more emotional satisfaction which leads to a better consumer behavior in OTT platform. Meanwhile, the direct link between brand engagement and consumer behavior was still significant while controlling for emotional satisfaction, which means that emotional satisfaction only partially mediates this relationship. Overall, these findings have implications for OTT services as they suggest that personalization can enhance consumer behavior through the enhancement of not just engagement, but a satisfying user experience which further impacts behavioral responses.
6.2 Relating the findings to earlier work
The results of this study are in line with previous research that positions AI-induced personalization as more than just a convenience, but rather as a strategic driver of brand-based outcomes in OTT environments. H1 (personalized features → brand engagement) is consistent with prior studies suggesting that the presence of dynamic and effortless recommendations can serve as a signal of a platform’s competence, responsiveness, and emotional intelligence, which in turn increases users’ trust and affective bond with the brand [7, 18]. This is also consistent with the perspective that personalization contributes to the co-creation of brand personality by enabling smarter and more empathetic experiences [7, 32]. Previous research also reports that interactive and gamified personalization features (e.g., customized prompts, streaks, and content-based nudges) enhance emotional salience and strengthen engagement by evoking a feeling of being seen and feeling close [23, 24].
However, these observations should be balanced with findings from prior research that warn that hyper-personalization can trigger a “creepiness” or intrusiveness effect, which may diminish engagement if users perceive personalization as overly commercial or surveillance-like [28, 35]. Taken together, the evidence supporting H1 aligns with existing literature, suggesting that personalization functions as an engagement-enhancing mechanism when deployed in ways that are perceived as authentic and user-value enhancing [18].
The supported H2 (brand engagement → consumer behavior) similarly reflects earlier research that conceptualizes brand engagement as a multidimensional construct—cognitive, emotional, and behavioral—that predicts downstream behavioral outcomes in digital environments [32]. In OTT settings specifically, prior work suggests that emotionally resonant and personalized brand interactions increase loyalty-type behaviors, including repeated usage and reduced churn tendencies [5, 6], while interactive participation features such as sharing recommendations and engaging with platform tools strengthen behavioral commitment and advocacy [33, 34]. Importantly, the confirmation of H3 (emotional satisfaction mediating the BE → CB relationship) is also well supported by prior explanations that OTT platforms foster affective attachment through platform anthropomorphism, where users perceive the system as emotionally attuned and thereby experience higher psychological comfort and satisfaction [4, 6, 12]. Your mediation result complements earlier arguments that emotionally rewarding experiences help convert engagement into sustained behavioral outcomes—such as continued viewing and platform reliance—while reinforcing the broader call for SEM-based approaches to test these complex causal pathways more rigorously [37]. Taken together, your findings extend existing OTT literature by empirically demonstrating that consumer behavior is shaped not only by engagement directly, but also through the emotional mechanism of satisfaction within AI-personalized environments [6, 7].
6.3 Theoretical implications
This study further enrich the theoretical explanation for AI personalization in OTT platforms by revealing the unique roles of brand engagement and emotional satisfaction in driving consumer effects. Indeed, a large amount of the place-specific OTT personalization literature has tended to focus either on technical issues related to recommendation systems [2] or use un-nuanced measures of satisfaction rather than theorize personalization as a branding tool. On the other hand, findings in this study add to the evidence that personalization is a tactical, experience-based signal enhancing consumers’ relationship with the platform brand. When personalization is perceived as being adapted, responsive and relevant it transcends utility and becomes a construct that increases customer cognitive and emotional commitment to the brand which in turn can lead to increase platform-based brand engagement [7, 18]. At its extreme, however, this research also complements recent warnings that personalization is not unequivocally a positive [28, 35], and indeed can lead to it being experienced as over-personalization which in turn may lead to irritation and perceptions of manipulation, actually reducing engagement. The study contributes to theory by positioning personalization in a branding and behavioral context, and thus treating it as an experiential and symbolic construct – rather than mere algorithmic outcome.
In addition, this research contributes to the understanding of consumer behavior in OTT environments by specifying the mediating function of emotional satisfaction as an essential psychological mechanism that underlies engagement–behavior linkage. Previous correlation-based studies have often associated OTT outcomes such as loyalty, continued use and binge-watching with engagement or satisfaction in a disconnected manner, without considering the mechanisms whereby feelings convert engagement into behavior within a causal model. The results of this study underline the significance of emotional fulfillment when it comes to OTT services, as users’ relationship with the platform is frequently not restricted to instrumental usage but also encompasses self-expression, emotion-guided viewing routines and mood management [4, 6]. This inference is consistent with the platform anthropomorphism view, whereby when users perceive recommendation systems as empathetic or “understanding”, they are prone to cultivate affective dependence and more pronounced behavioral continuation [6, 12]. Consequently, the study also extends prior research on OTT engagement by showing that it is not only the actual level of engagement, but emotionally rewarding experiences from engaging in the medium which drives behavioral outcomes—a finding that provides a more nuanced understanding of why some engaged users tend to become behave consistently while others do not [37].
Aside from its contributions towards behavioral and branding aspects, the results of this study also resonates with positioning emerging conversations around sustainable design for digital media ecosystems and responsible governance of personalization. While OTT platforms are now more heavily influenced in their content curation through algorithmic recommendation systems, usage experiences have drawn increased attention to the potential need for a balance between personalization optimization opportunities and user well-being; as well as transparency and ethical data-use practices. From a digital planning standpoint, personalization architectures that allow users to have increased control over recommendation settings and visibility into algorithmic logic, as well as flexibility in managing their viewing routines may enable platforms to develop more sustainable long-term relationships with users. These four dimensions of governance were not explicitly captured in this study's empirical model; however, the findings emphasize the value of nesting behavioral insights within larger existing conversations around governing AI-enabled personalization to reasonable ends in fast-growing streaming environments.
6.4 Practical implications
This research provides valuable managerial implications for the managers of OTT platforms and their advertisers by indicating that personalized recommendation driven by AI should not only be seen as an operational tool, but also a strategic engagement factor. Because these features deepen a user’s relationship with the platform and spur further usage, it’s in the platforms’ best interest to invest in personalization elements that actual users can feel and see – better content discovery, mood- or context-based recommendations, adaptive homepages personalized by individual needs or interests (among other strategies), and consistency across multiple devices. Interactive and recognition-based features that establish emotional connection without the need for content recommendations, such as personalized prompts, viewing milestones, curated collections or tailored notifications may additionally help OTT services to encourage user engagement. But personalization is a tricky thing. Previous concerns about the “creepiness” of, or a being monitored by platforms imply that platforms should be transparent and allow users to have control over their personalization experience (e.g., editing preferences and resetting recommendation history, changing how personalized an experience is) so they perceive personalization as assistance rather than surveillance [28, 35]. In other words, OTT companies should strive for personalized content that is relevant and considerate, as they are the best positioned to drive sustainable engagement.
Importantly, the results of this study also suggest that managers should consciously design OTT experiences, promoting emotional satisfaction, as emotional satisfaction amplifies the route from engagement to consumer behavior. This indicates that platform strategies need to be about more than functional competencies: they need to intentionally create emotionally rewarding experiences — for instance, helping people overcome “decision fatigue” by simplifying the discovery journey, providing them with curated, comfort viewing or mood-based content collections and evoking positive emotions through streaming quality so smooth and navigation friction-free. These enriching experiences can help drive behavioral outcomes: users are more likely to keep using your product, less likely to churn and more likely to advocate for your brand.
Since emotional satisfaction only partially explains the engagement–behavior link, practitioners should also strengthen engagement directly through community and social features (e.g., watch-party options, shareable recommendations, interactive polls, and personalized social sharing prompts) that keep users actively connected with the platform. Finally, the findings suggest that OTT platforms operating in culturally diverse markets should adopt context-aware personalization, ensuring recommendations reflect regional languages, viewing rituals, and culturally grounded preferences rather than relying solely on generalized global algorithms [9, 16, 17]. Overall, the study implies that OTT brands will be better positioned to retain users and shape positive consumer behavior when they combine intelligent personalization, emotional satisfaction-building design, and ethical, culturally sensitive implementation.
Beside the managerial implications for OTT platform designers, the findings of this study also add new perspective to ongoing discussions regarding emerging digital platform governance and sustainable media ecosystem planning. With algorithmic recommendation systems playing an ever-expanding role in shaping the daily touchstones of user engagement, both platform developers and regulators are urged to consider the adoption of transparency-oriented personalization interfaces, user-controlled recommendation settings, and responsible notification architectures that avoid deepening patterns of extreme engagement reinforcement. From the digital planning perspective, incorporating explainable recommendation mechanisms and make users more aware of personalization logic might lead to long-term healthier relationship between users and streaming. While these governance-oriented dimensions were outside the direct empirical focus of the current study, the behavioural findings generated here further inform ongoing interdisciplinary dialogues related to responsible AI-enabled personalised strategies in fast-growing OTT contexts.
6.5 Limitations and future research directions
Although the current study sheds light on the influence of AI-induced personalization on brand engagements and consumption behaviors in OTT platforms, a few limitations should be acknowledged which need to interpreted while deriving insights. First, the study used a non-probability snowball sampling technique in which the selection of respondents is dependent on the direct or indirect (friend-of-a-friend) network of a small number of participants. Although this approach was appropriate to reach digitally active OTT users across several zones of India, it had limitations for statistically generalizing the national-level conclusions [35]. Subsequent studies may improve upon representativeness across different user groups with probability-based sampling methods or stratified regional sampling designs.
Second, the analysis was carried out with a focus on Gen Z respondents since this demographic is among one of the most active user groups in India’s OTT ecosystem. On the other hand, personalization perceptions and engagement reactions can vary by older generation cohorts or segments of less digitally intensive/tethered users. Future studies might therefore explore multi-generational samples, potentially offering deeper insight into the behavioral dynamics of personalization.
Third, the study failed to include other context control variables (e.g., OTT use frequency, subscription level, type of platform [global vs. regional services], income category) that may influence engagement intensity and behavioral responses in personalization environments. Adding such variables to future structural models might assist in coming closer to an explanation of platform-specific personalization experiences.
Fourth, all constructs in the study were operationalized using three-item reflective measures borrowed from existing work on personalization and digital engagement (7–9). While this method supported model parsimony and respondent completion reliability in an online survey context, other constructs such as brand engagement and consumer behavior are naturally multidimensional with potential advantages from broader measurement frameworks in future studies.
Second, the cross-sectional design does not allow one to observe changes in personalization responses over time. Thus, longitudinal research designs might shed additional light on how personalization-driven engagement patterns develop as users navigate algorithmically mediated media environments over longer stretches of time. These include sustainability-oriented personalization concepts like digital well-being, responsible recommendation exposure, or end-user controlled personalization transparency mechanisms, which link to disciplines further afield from HCI and could enhance the interdisciplinary understanding of AI-enabled media ecosystems.
[1] Statista. (2024). OTT market statistics in India. https://www.statista.com.
[2] Gómez-Uribe, C.A., Hunt, N. (2016). The Netflix recommender system: Algorithms, business value, and innovation. ACM Transactions on Management Information Systems, 6(4): 1-19. https://doi.org/10.1145/2843948
[3]Ricci, F., Rokach, L., Shapira, B. (2022). Recommender systems: Techniques, applications, and challenges. In Recommender Systems Handbook, 3rd ed., pp. 1-35. https://doi.org/10.1007/978-1-0716-2197-4_1
[4] Yin, J. (2025). The impact of AI-personalized recommendations on consumer experience and online engagement. Journal of Electronic Commerce Research and Applications, 20(1): 21. https://doi.org/10.3390/jecsda20010121
[5] Yaprak, U. (2025). Binge-watching trends among millennials in OTT platforms. Digital Media Studies, 7(1): 12-28.
[6] Cohn, M., Pushkarna, M., Olanubi, G.O., Moran, J.M., Padgett, D., Mengesha, Z., Heldreth, C. (2024). Believing anthropomorphism: Examining the role of anthropomorphic cues on trust in large language models. arXiv preprint arXiv2405.06079. https://doi.org/10.48550/arXiv.2405.06079
[7] Ajith, A., Kumar, R., Singh, P. (2025). Personalization as a branding tool in OTT platforms. Journal of Digital Marketing, 12(3): 45-60.
[8] Periaiya, S., Nandukrishna, A.T. (2023). What drives user stickiness and satisfaction in OTT video streaming platforms? A mixed-method exploration. International Journal of Human–Computer Interaction, 40(9): 2326-2342. https://doi.org/10.1080/10447318.2022.2160224
[9] Mangruwa, R., Amalia, S. (2024). Cultural influences on OTT personalization in emerging markets. International Journal of Media Studies, 6(4): 78-92.
[10] Diehl, K., Zauberman, G., Barasch, A. (2020). How taking photos increases enjoyment of experiences. Journal of Personality and Social Psychology, 119(5): 1190-1207. https://doi.org/10.1037/pspa0000195
[11] Kandampully, J., Zhang, T., Bilgihan, A. (2015). Customer loyalty: A review and future directions with a special focus on the hospitality industry. International Journal of Contemporary Hospitality Management, 27(3): 379-414. https://doi.org/10.1108/ijchm-03-2014-0151
[12] Sellos, M., Costa, L., Pereira, J. (2024). Gen Z and millennials in digital media consumption. Journal of Consumer Behaviour, 13(4): 89-104.
[13] Verhoef, P.C., Lemon, K.N., Parasuraman, A., Roggeveen, A., Tsiros, M., Schlesinger, L.A. (2009). Customer experience creation: Determinants, dynamics, and management strategies. Journal of Retailing, 85(1): 31-41. https://doi.org/10.1016/j.jretai.2008.11.001
[14] Rzepka, C., Berger, B. (2018). User interaction with recommender systems: A systematic review. Journal of the Association for Information Science and Technology, 69(9): 1050-1063. https://doi.org/10.1002/asi.24006
[15] Sunikka-Blank, M., Galbraith, M. (2021). Beyond efficiency: Emotional and behavioural dimensions of personalization. Journal of Consumer Culture, 21(3): 417-436. https://doi.org/10.1177/1469540520952793
[16] Nisarga, S., Rao, K., Gupta, M. (2024). OTT consumption patterns in India: A PAN India study. Journal of Media and Technology, 10(1): 23-38.
[17] Pattanayak, S., Sharma, A., Jain, R. (2024). Regional diversity in OTT adoption in India. Asian Media Journal, 9(3): 55-70.
[18] Obiegbu, C., Larsen, K. (2024). Emotional resonance in OTT branding. Brand Management Review, 15(2): 67-82.
[19] Deci, E.L., Ryan, R.M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4): 227-268. https://doi.org/10.1207/S15327965PLI1104_01
[20] Bucher, T. (2017). The algorithmic imaginary: Exploring the ordinary affects of Facebook algorithms. Information, Communication & Society, 20(1): 30-44. https://doi.org/10.1080/1369118X.2016.1154086
[21] Belk, R.W. (2013). Extended self in the digital world. Journal of Consumer Research, 40(3): 477-500. https://doi.org/10.1086/671052
[22] Kietzmann, J.H., Hermkens, K., McCarthy, I.P., Silvestre, B.S. (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54(3): 241-251. https://doi.org/10.1016/j.bushor.2011.01.005
[23] Hamari, J., Koivisto, J., Sarsa, H. (2014). Does gamification work? A literature review of empirical studies on gamification. In Proceedings of the 47th Hawaii International Conference on System Sciences, Waikoloa, HI, USA, pp. 3025-3034. https://doi.org/10.1109/HICSS.2014.377
[24] Calder, B.J., Malthouse, E.C., Schaedel, U. (2009). An experimental study of the relationship between online engagement and advertising effectiveness. Journal of Interactive Marketing, 23(4): 321-331. https://doi.org/10.1016/j.intmar.2009.07.002
[25] Hofstede, G. (2011). Dimensionalizing cultures: The Hofstede model in context. Online Readings in Psychology and Culture, 2(1): 1-26. https://doi.org/10.9707/2307-0919.1014
[26] Srinivasan, S., Venkatraman, S. (2018). Personalization in emerging markets: Cultural adaptation and local relevance in digital platforms. Journal of International Marketing, 26(3): 20-38. https://doi.org/10.1509/jim.17.0045
[27] Bongiovi, J.R. (2019). Review of the age of surveillance capitalism: The fight for a human future at the new frontier of power, by S. Zuboff. Social Forces, 98(2): 1-4. https://www.jstor.org/stable/26862460.
[28] Acquisti, A., Brandimarte, L., Loewenstein, G. (2015). Privacy and human behavior in the age of information. Science, 347(6221): 509-514. https://doi.org/10.1126/science.aaa1465
[29] Pasquale, F. (2015). The Black Box Society: The Secret Algorithms that Control Money and Information. Harvard University Press.
[30] Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., Luetge, C., Madelin, R., Pagallo, U., Rossi, F., Schafer, B., Valcke, P., Vayena, E. (2018). AI4People—An ethical framework for a good AI society. Minds and Machines, 28(4): 689-707. https://doi.org/10.1007/s11023-018-9482-5
[31] Arrieta, A.B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., Chatila, R., Herrera, F. (2020). Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58: 82-115. https://doi.org/10.1016/j.inffus.2019.12.012
[32] Hollebeek, L.D., Glynn, M.S., Brodie, R.J. (2014). Consumer brand engagement in social media: Conceptualization, scale development and validation. Journal of Interactive Marketing, 28(2): 149-165. https://doi.org/10.1016/j.intmar.2013.12.002
[33] Stephen, A.T., Toubia, O. (2010). Deriving value from social commerce networks. Journal of Marketing Research, 47(2): 215-228. https://doi.org/10.1509/jmkr.47.2.215
[34] Muniz Jr., A.M., O’Guinn, T.C. (2001). Brand community. Journal of Consumer Research, 27(4): 412-432. https://doi.org/10.1086/319618
[35] Aguirre, E., Roggeveen, A.L., Grewal, D., Wetzels, M. (2015). The personalization–privacy paradox: Implications for new media. Journal of Consumer Marketing, 32(2): 98-110. https://doi.org/10.1108/JCM-06-2014-1023
[36] Bleier, A., Eisenbeiss, M. (2015). Personalized online advertising effectiveness: The interplay of what, when, and where. Marketing Science, 34(5): 669-688. https://doi.org/10.1287/mksc.2015.0930
[37] Chan-Olmsted, S., Luo, J. (2024). Structural equation modelling in OTT user studies. Media Studies Journal, 8(2): 101-120.
[38] Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. (2019). Multivariate Data Analysis (8th ed.). Cengage Learning.
[39] Saunders, M., Lewis, P., Thornhill, A. (2016). Research Methods for Business Students (7th ed.). Pearson Education.
[40] Arbuckle, J.L. (2017). IBM SPSS Amos 25 User’s Guide. IBM Corp.
[41] Fornell, C., Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1): 39-50. https://doi.org/10.2307/3151312