What makes OTT platforms popular? A provider-centric composite index approach based on platform attributes

The traditional methods of delivering film and television content via cable, broadcast and satellite have become outdated, replaced by internet-based services known as over-the-top (OTT) (Lotz, 2017). OTT streaming represents a shift in media consumption, sparking the digital merging of entertainment, communication and technology, allowing consumers to access content anytime and anywhere (Lotz, 2017; Dwivedi et al., 2020). In India, OTT consumption has steadily grown, driven by advances in digital infrastructure such as the widespread availability of high-speed internet, along with localised content and diverse subscription options (Yoon and Kim, 2023; Biswas, 2025). However, although these infrastructural enablers have been extensively discussed, there remains limited exploration of how provider-side platform attributes translate these advantages into market performance (Barney, 1991; Zeithaml et al., 1996). Service popularity strongly correlates with consumer search behaviour (da Silva and de Andrade Lima, 2022). In this study, consumer popularity (CP) refers to the degree of consumer engagement and the visibility of an OTT platform in the digital environment, which lines up with prior research that customises online behavioural data to rough brand or platform acceptance (Choi and Varian, 2012). From a provider’s perspective, CP is a market-level outcome of strategic traits such as content diversity (CD), pricing and technological quality (Luo and Bhattacharya, 2006; Zeithaml et al., 1996). Accordingly, while the data are consumer-generated, the construct is observed to result from provider-side choices and platform design. This provider-oriented lens assesses performance based on measurable operational and market indicators rather than user attitudes (Parasuraman et al., 1988; Cronin and Taylor, 1992). CP can be operationalised as a composite index combining visible indicators such as active downloads, app store ratings and aggregated user reviews, which are crucial for enhancing engagement, reducing churn and improving service quality from the provider’s standpoint (Nardo et al., 2005; Luo and Bhattacharya, 2006; Dwivedi et al., 2020; Zeithaml et al., 1996). Stranded in a provider-centric perception, this study explores the following research questions regarding platform popularity:

This study is primarily grounded in dynamic capabilities theory (DCT), which explains how firms create competitive advantage by continuously reconfiguring resources and capabilities in response to changing market conditions. In the OTT context, CD, value-for-money (VM) and technological sophistication (TS) represent strategic provider capabilities that influence platform popularity. Technology acceptance model (TAM) and expectation-value theory (EVT) complement DCT by explaining consumer evaluation of provider capabilities. TAM posits that technology adoption is driven by perceived usefulness and ease of use (Davis, 1989). Perceived usefulness is reflected in the benefits derived from platform attributes such as content variety and pricing value, while perceived ease of use is shaped by technological features such as streaming quality, interface design and device compatibility. EVT explains how consumers evaluate content offerings and the value of subscriptions, thereby justifying the costs incurred. CD enhances expected entertainment value, while pricing structures and bundled benefits influence perceived economic value (Eccles and Wigfield, 2002). While TAM and EVT explain consumer evaluation, DCT provides a provider-centric perspective by emphasising how firms continuously adapt their resources to change market conditions (Teece, 2007). In the OTT context, platforms enhance their competitiveness to sustain consumer engagement and popularity by dynamically reconfiguring content portfolios, pricing strategies and technological infrastructure.

Within the DCT framework, OTT platform popularity is viewed as a market-level outcome of providers’ abilities to adapt content portfolios, pricing structures and technological infrastructure. TAM and EVT complement this perspective by explaining how these provider capabilities translate into consumer perceptions of usefulness, ease of use and value, ultimately shaping platform popularity.

CD refers to the array of creative, linguistic and cultural elements that an OTT platform seeks to attract and retain viewers. It reflects the capacity to sense viewers’ heterogeneity and reconfigure creative resources accordingly to demonstrate that richer and more localised content boosts expected value and satisfaction from a consumer perspective (EVT) and reflects the provider’s ability to adapt content offerings to diverse preferences (DCT) (Teece, 2007; Davis, 1989; Eccles and Wigfield, 2002).

Offering a wide variety of genres reveals the provider’s inventive flexibility and its modernisation of content (Roy and Basu, 2024). It reflects content adaptability and enhances expected entertainment value among consumers.

Providing multilingual options, dubbing and localised subtitles reduces user struggle and enhances accessibility (Pekpazar et al., 2023). This enhances accessibility and content localisation.

Capitalising on regional or vernacular content helps platforms achieve cultural proximity and emotional resonance (Chatterjee et al., 2025). This strengthens cultural proximity, audience relevance and expected consumer satisfaction within local markets.

VM is a mutual economic and perceptual assessment of the payback relative to the cost. Reasonable and convenient access advances practicality and performance expectancy to view pricing strategy as part of co-created value, a dynamic process of positioning offerings in line with consumer expectations (Teece, 2007).

Flexible price structures provide quickness and awareness for varied consumer budgets (Cronin et al., 2000). This improves affordability perceptions and perceived value across varied consumer segments.

Offering various plan options surges consumer preference and perceived fairness (Tm et al., 2021).

Enclosure of corresponding facilities embodies provider novelty in co-creating value (Dwivedi et al., 2020). Bundled services enhance perceived platform utility and value by extending benefits beyond streaming access.

TS aligns with the platform’s dimensions, integrating and renewing digital technologies to enhance the streaming experience. It involves developing innovative systems to improve service quality, usability and compatibility to increase perceived ease of use, effort expectancy, performance satisfaction and perceived probability of realising desired entertainment products (Teece, 2007).

Superior streaming and audio quality improve perceived usefulness and overall service experience (Laghari and Connelly, 2012).

Multi-device openness indicates provider adaptability and elasticity (Dwivedi et al., 2020). This enhances accessibility, ease of use and platform adaptability across technological environments.

Ad-free or user-controllable advertisement setups advance approval and action (Kashyap, 2023). Lower advertisement interruptions improve user experience, platform satisfaction and perceived service value.

Although the expression CP originates from user-level data, it epitomises a provider-centric outcome of strategic measures that drive consumer engagement and prominence. Popularity appears as a market-level image of a provider’s efficiency in reconfiguring content, pricing and technology, which translates into observable behavioural indicators such as downloads, searches and ratings (Vargo and Lusch, 2004).

Downloads symbolise behavioural implementation outcomes that increase after provider-driven accessibility and consumer-perceived usefulness (Kim et al., 2007).

Search trends reflect shared inquisitiveness produced through provider marketing and content innovation (Choi and Varian, 2012). They point to user anticipation of getting appropriate, valued entertainment experiences.

User ratings measure satisfaction through perceived performance and value (Luo and Bhattacharya, 2006). They echo the provider’s capacity to sustain expectations and quality perceptions.

Reviews deliver empirical insights and feedback loops that providers use to advance offerings (Dwivedi et al., 2020).

The subscription model (SM) outlines the entry mechanism and returns logic of OTT services. It symbolises a dynamic capability that adjusts pricing structures to varying market demands by describing how payment structures shape value expectations and adoption behaviour. Thus, the SM moderates the impact of VM and TS on CP.

Fixed periodic payments bring continuous access, solidifying both perceived reliability and provider revenue stability (Dwivedi et al., 2020).

Pay-per-view options appeal to selective users seeking individuality and price control (Wu et al., 2024).

Ad-supported free tiers are spreading among cost-sensitive users (Kashyap, 2023).

Combining paid and free tiers mirrors dynamic adaptation to diverse viewers (Teece, 2007).

From a strategic management perspective, firm performance depends upon its capability to integrate and reconfigure resources (Eisenhardt and Martin, 2017) and create value by continuously aligning organisational capabilities with evolving customer needs (Day, 1994). These perspectives reinforce the provider-centric logic adopted in this study, where content (Sharma and Harsora, 2023), pricing (Sar, 2022) and technology (Nirmal and Kaitharath, 2022) are viewed as strategic resources influencing platform popularity. Despite the rapid growth of OTT services, a unified explanation of the moderating role of subscription models remains limited (Sinha et al., 2024). Therefore, this study constructs a provider-centric composite index to empirically evaluate these factors across 31 Indian OTT platforms, offering a hypothetical vision for twofold theoretical integration and managerial guidance for enriching platform attractiveness in developing markets (Basha and Reddy, 2024; FICCI and EY, 2023; Dwivedi et al., 2020).

In this proposed research, background studies help identify the variables that contribute to an OTT platform’s popularity from the provider’s perspective. CP is recognised as the dependent variable, while CD, VM and TS are the independent variables. It is also determined that the effects of the moderator Subscription Plan influence the relationships between VM and CP as well as TS and CP, especially in freemium or hybrid models, but do not affect CD.

Roy and Basu (2024) and Dhiman (2023) found that the popularity of OTT platforms increases with diverse content across different genres and languages in India’s linguistically and culturally diverse market. Diverse content enhances consumer value and platform relevance through culturally aligned genres, languages and regional offerings, while also reflecting provider adaptability in competitive OTT markets (Davis, 1989; Eccles and Wigfield, 2002; Teece, 2007). Media diversity contributes to perceived platform attractiveness and competitive positioning, particularly in culturally fragmented markets where consumers seek content that reflects linguistic and regional identities (Napoli, 2011). Dhiman (2023) reported that diverse genres and languages enhance user satisfaction. Therefore, we hypothesise that:

Subscription price, bundled services and plan types are commonly recognised as drivers of perceived value for consumers; empirical studies indicate that cost-benefit considerations affect subscription intention (Tm et al., 2021; Nagaraj et al., 2021). Pricing flexibility and bundled benefits improve perceived value while enabling providers to attract diverse consumer segments through adaptive pricing strategies (Eccles and Wigfield, 2002; Teece, 2007). In subscription-based digital environments, favourable value assessments enhance satisfaction and strengthen behavioural outcomes such as subscription intention, retention and platform advocacy (Kim et al., 2007). Wu et al. (2024) linked favourable cost-benefit perceptions with subscription intention. Hence, it is proposed that:

First-class streaming, cross-platform accessibility and governable ad loads enhance the user experience and are associated with superior engagement (Laghari and Connelly, 2012; Park et al., 2016). User-friendly technologies improve accessibility and engagement, while continuous technological adaptation strengthens service quality and platform competitiveness (Teece, 2007). In media streaming environments, superior technological functionality improves service reliability and enhances consumer engagement, thereby contributing to stronger platform performance and visibility (Petter et al., 2008). Nirmal and Kaitharath (2022) linked superior technological features with higher user engagement. Thus, it is assumed that:

Freemium and hybrid models attract price-sensitive Indian customers and can amplify the impact of VM (Dwivedi et al., 2020; Roongta and Hari, 2021). Flexible subscription structures reduce access barriers and strengthen perceived value across different consumer segments (Eccles and Wigfield, 2002). Research on digital platform economics suggests that freemium and hybrid models reduce adoption barriers by allowing consumers to evaluate service benefits before making financial commitments (Wagner et al., 2014). Wu et al. (2024) linked hybrid models with higher retention. Therefore:

Freemium and hybrid models allow users to trial innovative features and may therefore amplify the popularity advantage of TS (Dwivedi et al., 2020; Roongta and Hari, 2021). Trial-based and flexible access models allow providers to showcase technological capabilities to broader audiences, thereby strengthening technology-driven engagement. Freemium and hybrid structures enable broader access to platform functionalities, increasing opportunities for users to evaluate service quality and technological performance (Lambrecht and Misra, 2017). Wu et al. (2024) linked hybrid models with greater technology adoption. Hence:

The conceptual model (Figure 1) illustrates how CD, VM and TS guide CP, with the SM moderating VM → CP and TS → CP. The framework integrates TAM and EVT with DCT to explain how consumer perceptions and provider capabilities influence OTT platform popularity.

This study employs a quantitative explanatory research design to analyse the factors influencing CP among Indian OTT platforms from a provider-centric perspective. The research aims to investigate how CD, value for money and TS affect consumer engagement. To ensure a robust assessment, a composite index method is used to measure and evaluate OTT platform attributes (Nardo et al., 2005; OECD, 2008). To understand the multidimensional impact of provider strategies on customer preferences in this highly competitive media environment, data have been collected from publicly available datasets. First-hand data are extracted directly from OTT platforms’ websites, while secondary data sources include Google, the Play Store, the App Store and the Selectra website (Pekpazar et al., 2023; Park et al., 2016).

Based on recognised variables from the literature, the study focuses on 31 Indian OTT video streaming platforms with original content, such as ALTBalaji, SonyLiv, Zee5, JioCinema, Disney + Hotstar, etc., which experience superior subscriber growth due to the concentration of localised content and language (Biswas, 2025; Basha and Reddy, 2024). As the secondary data sets accessible for the study are from 2024, the Jio Cinema-Hotstar merger of February 2025 and any rebranding after 2024 are not measured.

The attributes related to the platforms, collected directly from their websites, include content, genres, Indian and international languages available, price ranges for different subscription plans, the number of devices supported, the maximum video and audio streaming quality and the level of advertising across plans. Additionally, the SM, the number of bundled platforms and the number of platforms are considered platform attributes for which metadata were available (Kübler et al., 2021; Kräkel and Fieseler, 2020). For a well-structured analysis, variables are quantified and constructed using predetermined measurement indicators grounded in theory and existing literature.

CP, the dependent variable, is measured as the consumer popularity index (CPI), derived from total play store downloads, Google Trend score (GTS) and Play Store and App Store reviews. App Store download data was deliberately omitted as Apple does not freely disclose precise, platform-verified download numbers. PSD are taken in absolute numbers. The GTS is set to the 2024 average to avoid skewing. The number of reviews is taken as the sum of Play Store and App Store reviews (total reviews (TOR)). These indicators (PSD, GTS, TOR) are normalised and given weights to construct the CPI (Fu et al., 2020).

The content diversity index (CDI) includes genres, languages and RCI. The number of genres (GEN) and languages (LAN) is measured in absolute values. The RCI is calculated as the sum of the intensity of regional content: 1 (Intense International Content), 2 (Intense Indian Content), or 3 (Intense Local Content), reflecting Indian consumers’ preference for regional content (Sharma, 2023; Nair and Shah, 2024) and the difference between regional and international languages. These indicators (GEN, LAN, RCI) are normalised and weighted according to their importance in forming the CDI (Fu et al., 2020).

The value-for-money index (VMI) includes subscription costs, the number of plans and bundled services. The maximum annual subscription price (MAP) and the number of subscription plans (PLA) is measured in real units. Bundling (BUN) is calculated as a product of bundle types and the number of bundles reflects the variety of bundling offered. Indicators (MAP, PLA, BUN) are normalised and weighted accordingly to construct the VMI (Fu et al., 2020).

The technological sophistication index (TSI) includes supported devices, minimum and maximum audio-video quality (AVQ) and the level of advertisement. The number of devices supported (DEV) is measured in real units. AVQ is calculated as the average of the minimum and maximum audio and video qualities. Minimum AVQ is encompassed in data-constrained markets like India; a considerable share of users depends on unstable or low-speed internet networks. The level of advertisement (ADV) is measured as 1 (no ads), 2 (plans with/without ads), or 3 (only ad-supported plans) as subscribers prefer less costly plans with manageable advertisements. Indicators (DEV, AVQ, ADV) are normalised and weighted as per importance to form the TSI (Fu et al., 2020).

The study used the SM as the moderating variable. SM is coded as 1 = SVOD (Subscription Video on Demand), 2 = TVOD (Transactional Video on Demand), 3 = Hybrid (SVOD + Advertising Video on Demand (AVOD)/TVOD combination) and 4 = AVOD (Advertising Video on Demand), and normalised in ascending order since higher values specify more ad-supported/freemium features, which associate with broader access and mass-market engagement in price-sensitive economies like India (Dwivedi et al., 2020; Wu et al., 2024).

The indices in this study are constructed to denote the multidimensional qualities of OTT platforms through selectively weighted indicators. Each index reflects provider-centric features of platform performance, with weights (1.5, 1 and 0.5) assigned based on theoretical importance and observed significance recognised in the literature. A multistep validation approach was adopted to ensure the validity and reliability of the composite indices (Nardo et al., 2005; OECD, 2008). Correlation analysis was conducted to assess internal consistency among the indicators within each index, which indicated moderate to high correlations while avoiding redundancy. Secondly, Exploratory Factor Analysis (EFA) was done to examine the one-dimensionality of the indicators, resulting in acceptable convergence (Hair et al., 2019). Principal Component Analysis (PCA) was conducted to further validate the structure of the indices. The results show that the indicators within each construct load onto a single dominant component, conforming to structural validity (Jolliffe and Cadima, 2016). Finally, a sensitivity analysis was performed by comparing weighted indices with equally weighted alternatives to validate the robustness of the index construction (Nardo et al., 2005; OECD, 2008). These methods also avoid misrepresentation in arithmetic or geometric aggregation by preserving the relative role across index mechanisms.

The CPI integrates TOR, PSD and GTS. TOR (Weight 1.5) is assumed to have the greatest significance, as user reviews directly reflect engagement (Mudambi and Schuff, 2010). PSD (Weight 1) signifies implementation but may comprise pre-installed or inactive accounts. GTS (Weight 0.5) captures public attention and search prominence, but trends are unpredictable and vulnerable to spikes (Choi and Varian, 2012). The indicators exhibit extremely high loadings (≈0.99) in PCA, indicating strong internal consistency and a well-defined single dimension. All indicators were min–max normalised and combined using a weighted tally as endorsed for digital platform metrics (Booysen, 2002; Nardo et al., 2005).

The CDI consists of RCI, number of genres (GEN) and number of languages (LAN). RCI (Weight 1.5) is prioritised because localised chronicles foster a sensitive tone and enhance viewership retention (Dhiman, 2023; Basha and Reddy, 2024). GEN (Weight 1) imitates the thematic range (Chatterjee et al., 2025). LAN (Weight 0.5) captures linguistic approachability but is less persuasive when dubbing and subtitling counterweigh language limitations (Pekpazar et al., 2023). PCA results show strong loadings (0.70–0.89), indicating a stable and well-structured construct. RCI was inversely normalised because niche platforms exhibit strong regional focus but limited engagement, which may unreasonably amplify diversity scores (FICCI and EY, 2023).

The VMI fits in the maximum subscription price (MSP), bundling services (BUN) and number of plans (PLA). MSP (Weight 1.5) is of utmost importance given Indian consumers’ price sensitivity (Nagaraj et al., 2021). BUN (Weight 1) backs perceived value within convergence (Kübler et al., 2021). PLA (Weight 0.5) augments elasticity but does not autonomously establish perceived value (Hsu and Lin, 2016). The indicators demonstrated high loadings (0.80–0.94), confirming strong convergence and one-dimensionality. All indicators are normalised before applying weights as per the composite indicator guiding principle (OECD, 2008).

The TSI consists of the number of supported devices (DEV), AVQ and advertisement level (ADV). DEV (Weight 1.5) is decisive, as multi-device compatibility enhances availability across households (Cho and Lee, 2022). AVQ (Weight 1) is vital for streaming satisfaction (Dwivedi et al., 2020). ADV (Weight 0.5) disrupts the experience but is regularly tolerated in place of free access (Kashyap, 2023; Laghari and Connelly, 2012). Although inter-item correlations are relatively lower, PCA loadings (∼0.70) indicate acceptable convergence, supporting the inclusion of the indicators in the composite index. All indicators are normalised and then weighted.

Even though the SM is not an index, it is included in this segment for its moderating role within the framework. SM is treated as an ordinal variable and coded as (1 = SVOD, 2 = TVOD, 3 = Hybrid, 4 = AVOD), indicating growing openness and lower entry barriers for consumers. This coding is theoretically justified as freemium and ad-supported models (Hybrid/AVOD) appeal more to price-sensitive spectators and improve platform influence (Dwivedi et al., 2020; Roongta and Hari, 2021). Unlike other variables, SM is a tactical business model choice instead of a multidimensional concept, so no index is required. However, normalising it allows it to be statistically tested as a moderator of the relationships among VM, TS and CP.

SmartPLS was used to assess the measurement model, and Jamovi was employed to conduct regression-based analysis of the relationships among the composite indices (Hair et al., 2019). The regression-based methods, which provide more stable estimates with limited observations, are an appropriate approach for a given small sample size (N = 31) and the use of single-indicator composite constructs (Rosseel, 2012). The adequacy of the small sample size (N = 31) was assessed using power-based criteria based on the inverse-square-root method (Kock and Hadaya, 2018). With the moderate-to-strong path coefficients (β = 0.36–0.453) and high explanatory power (R² = 0.725), the sample size is considered sufficient. Bootstrapping further enhances the robustness of estimates.

The descriptive statistics indicate the dominance of a few OTT platforms, as CPI shows a low mean but high skewness and kurtosis. This supports prior findings that OTT success is heavily concentrated on specific platforms (da Silva and de Andrade Lima, 2022). The CDI has a moderate mean, positive skewness and mild kurtosis, suggesting diverse linguistic and genre preferences across platforms. This supports hypothesis H1 (CDI → CPI) and the relevance of regional content (Jung and Park, 2024; Eccles and Wigfield, 2002). VMI records the highest mean and standard deviation with a near-normal distribution. This reflects hypothesis H2 (VMI → CPI), as pricing and bundling influence price-sensitive Indian consumers (Sharma and Mishra, 2023). The TSI shows a balanced distribution with a relatively high mean and moderate dispersion and skewness, indicating that most platforms cluster around an average technological level. However, some platforms stand out for superior device compatibility, video quality and lower advertisement levels. This aligns with hypothesis H3 (TSI → CPI), which posits that video quality and ease of use enhance consumer engagement (Periaiya and Nandukrishna, 2024; Davis, 1989). The moderator SM exhibited high variance and a positively skewed distribution. The median value (0) indicates that most platforms operate under SVOD models. However, a few platforms adopting freemium and hybrid models increase perceived value for cost-conscious consumers by reducing entry barriers (Roongta and Hari, 2021; Dwivedi et al., 2020). This supports hypotheses H4A (VMI × SM → CPI) and H4B (TSI × SM → CPI). The descriptive results (Table 1) support the conceptual model as CP remains highly skewed and concentrated, yet significant heterogeneity exists across content, pricing and technological attributes (Garg and Gupta, 2025; Roy and Basu, 2024). The results of descriptive statistical analysis are presented in Table 1.

The key indicators, such as outer loadings, Cronbach’s alpha, composite reliability and average variance extracted (AVE), were measured and reported (Table 2) to assess the adequacy of the measurement model (Hair et al., 2019). The outer loadings were above the threshold value (0.7) for all indicators except for DEV (0.505) under TS. Similarly, Cronbach’s alphas exceeded the minimum threshold of 0.7 for all constructs except TS (0.492), which remains marginally adequate given its Composite Reliability (0.727) (Nunnally and Bernstein, 1994; Hair et al., 2019). Although its AVE (0.479) falls slightly below the 0.5 cut-off, this is sufficient in exploratory composite models when the construct is theoretically defensible and reliability is satisfactory (Fornell and Larcker, 1981). The results indicate reasonable convergent validity for TS within an exploratory composite-index framework. Although slight measurement limitations are observed, additional validation procedures, including correlation analysis, EFA, PCA and sensitivity analysis, provide support for the robustness and theoretical relevance of the construct (Nardo et al., 2005). The results of the measurement model are presented in Table 2.

The Fornell-Larcker criterion was used to measure discriminant validity (Table 3). For each variable, the diagonal values, which denote the square root of the AVE, remained higher than the inter-construct correlations except for TSI. The Fornell–Larcker criterion specifies that the square root of the AVE for TS (0.692) is lower than the inter-construct correlations with CD (0.733), VM (0.731) and SM (0.723), suggesting partial limitations in discriminant validity. However, given the exploratory nature of the study and the theoretical distinctiveness of TS within the provider-centric technology framework, retaining it is defensible. The slightly high cross-loadings are attributable to conceptual overlap among technology-enabled platform qualities, such as device compatibility, streaming quality and ad levels, which also affect perceived value and content accessibility. Such retention is considered acceptable when discriminant validity deviations are theoretically reasonable and reliability remains adequate in exploratory settings (Hair et al., 2019; Voorhees et al., 2016). Furthermore, PCA and EFA results confirmed that the indicators load onto a coherent underlying dimension with acceptable structural consistency despite moderate overlap among technology-related attributes. This is consistent with the multidimensional nature of technological platform characteristics in OTT ecosystems. A full collinearity test using the variance inflation factor was conducted to identify common method bias (Kock, 2015). All indicators, except those of CPI, reported acceptable 6F values (<3.3). These indicators were retained due to their theoretical relevance and additional empirical support obtained through robustness and index validation analyses (da Silva and de Andrade Lima, 2022). The results of discriminant validity using the Fornell-Larcker criterion are presented in Table 3.

After the measurement model was verified for reliability and validity, the hypotheses were analysed and tested using the structural model. The structural model was evaluated based on path coefficients (β), standard errors (σ), confidence intervals, Z-values and p-values, along with the model’s explanatory and predictive capabilities through R², adjusted R² and Q² values. The R² value (0.725) and the adjusted R² value (0.656) indicate strong explanatory power, accounting for a substantial portion of the variance in CP. The model’s meaningful predictive relevance is demonstrated by a Q² value (0.356) (Hair et al., 2019).

The three independent variables, CDI, VMI and TSI, were found to have a statistically significant effect on CPI (Table 4). This validates hypotheses H1, H2 and H3 as CD, VM and TS positively affect CP. The moderator’s interaction with VMI did not produce a statistically significant effect on CPI. Therefore, hypothesis H4A is rejected because SM does not moderate the VMI → CPI relationship. However, the moderator SM’s interaction with TSI revealed a strong and significant moderating effect on CPI. This supports hypothesis H4A, as SM moderates the TSI → CPI relationship, with a more substantial impact for freemium or hybrid models (Cohen, 2013; Hair et al., 2019). The results of the structural model are presented in Table 4 and represented in Figure 2.

The findings of this study offer insights into the provider-centric factors influencing OTT platform popularity in India. The weighted composite index outcomes show that technological sophistication (TSI) has the most substantial impact on CP (CPI) (β = 0.453, p < 0.001), consistent with the strong contribution of technological indicators within the composite index structure (DEV = 1.5, AVQ = 1.0). Platform availability and adaptive streaming quality influence user engagement more than content variation alone (Bhadeshiya, 2025). Recent empirical work by Baishya and Mini (2025) supports the view that Indian users prioritise continuous streaming and device compatibility over content volume. VMI also significantly influences CPI (β = 0.419, p < 0.01) through its perceived economic value in price-sensitive markets. However, its effect is weaker than TSI despite the strong weighting assigned to pricing-related indicators within the index structure (MAP = 1.5), indicating that pricing advantages alone may not be sufficient to ensure platform popularity when technological quality is comparatively weaker. This aligns with the findings of Nagaraj et al. (2021) and FICCI and EY (2023), which demonstrate that bundled plans are associated with higher CP only when supported by firm-level streaming infrastructure. CD (CDI) remains important but shows the weakest influence on CPI (β = 0.36, p < 0.001) despite its weighted index structure. This supports Dhiman (2023), who observed that users prioritise quality over mere accessibility of multilingual content. The findings suggest that while CD remains an important driver of OTT popularity, its effectiveness is strengthened when supported by technologically efficient and economically accessible delivery systems.

Moderation analysis indicates that the SM does not significantly moderate the VMI → CPI relationship (β = 0.387, p = 0.197), suggesting that consumer sensitivity to VM remains relatively stable across SVOD, AVOD, Hybrid and TVOD models. However, SM significantly moderates the TSI → CPI relationship (β = 1.497, p < 0.001), with Hybrid and freemium models intensifying the influence of technology on user engagement, which is consistent with findings by Hsu and Lin (2016) and Wu et al. (2024) supporting the idea that technology-related platform benefits are perceived more strongly when access barriers are reduced. Given the exploratory nature of the study and the limited sample size, these moderation effects should be interpreted as indicative rather than confirmatory.

This research advances DCT by explaining OTT platform popularity as a market-level outcome of provider capabilities in content, pricing and technology, while TAM and EVT provide complementary explanations of consumer evaluation mechanisms. The outcomes extend TAM (Davis, 1989) from discrete user-adoption settings to a platform-level systematic framework. The noteworthy impact of TS on CP validates the operationalisation of perceived usefulness and ease of use as quantifiable provider-side traits, such as device compatibility, adaptive streaming and interface accessibility. This suggests that provider-level technological capabilities can meaningfully shape aggregate patterns of consumer engagement within OTT ecosystems (Dwivedi et al., 2020).

The findings also fortify EVT (Eccles and Wigfield, 2002), demonstrating that consumers evaluate OTT platforms based on perceived value derived from pricing, plan flexibility and bundled benefits. However, the lack of a moderating effect of subscription models on the VM relationship challenges EVT’s linear cost–benefit assumption, suggesting that affordability alone may not be sufficient to sustain consumer engagement (Kim and Park, 2015). DCT (Teece, 2007) is supported by pricing adaptability and technological responsiveness as strategic capabilities that enable providers to sustain engagement in dynamic digital ecosystems.

The results provide actionable insights for OTT service providers to improve consumer engagement and platform popularity. While TS had the most significant impact, providers should focus on cross-device compatibility, flexible streaming and minimal ad interruptions to enhance perceived ease of use and service quality (Davis, 1989). VM remains vital but becomes more influential when combined with reliable technological performance and flexible pricing structures. From a managerial standpoint, the findings support capability-based resource allocation, signifying that investment in technology, pricing and content should be synchronised rather than practised independently. Managers should prioritise technology while aligning pricing and content strategies. Implementing freemium or hybrid models can appeal to price-sensitive users while ensuring seamless service delivery and bundled benefits (Dwivedi et al., 2020). CD had a comparatively weaker direct effect; however, it plays a strategic role when associated with technological excellence. Providers must therefore invest in unique regional content, multilingual interfaces and diverse genre offerings to cater to heterogeneous audience segments (Dhiman, 2023). Ultimately, integrating real-time analytics and Google Trends monitoring into marketing intelligence can help platforms dynamically adjust content strategies and improve strategic market positioning (Choi and Varian, 2012). Therefore, from a policy perspective, the findings support initiatives that encourage digital infrastructure development, regional content production and fair competition among OTT providers, which can improve platform accessibility while endorsing innovation and cultural diversity to support sustainable competitive positioning within India’s dynamic OTT landscape.

Despite its comprehensive framework and exploratory empirical approach, this study has certain limitations. It is geographically limited to India and focuses primarily on the Indian OTT ecosystem, which confines its global generalisability. The comparatively small sample size of 31 platforms and the absence of comprehensive provider-level data limit the broader generalisability and statistical strength of the model. Accordingly, the findings should be interpreted as exploratory rather than confirmatory. Nevertheless, the study provides a useful foundation for future provider-centric OTT platform research. Furthermore, as the study is grounded in 2024 data, it may not capture the rapid shift in consumer preferences driven by price fluctuations, mergers, technological developments and AI-enabled platform features.

Although the weighting structure for index construction was theoretically grounded, its reliance on pre-assigned weights may limit methodological precision. Moreover, additional validation procedures including PCA, EFA, correlation analysis and sensitivity testing were employed to support the robustness of the composite indices; future studies may adopt more advanced data-driven weighting approaches using larger datasets. Longitudinal or panel-based studies may better capture changes in consumer behaviour over time. In contrast, comparative cross-country analyses, particularly in developing markets such as Latin America or Southeast Asia, would improve the model’s external validity. Integrating consumer-level primary data, such as survey responses or behavioural analytics, combined with provider-centric features could produce a more holistic understanding of OTT consumption dynamics. Finally, exploring new predictors of platform popularity, along with AI-driven personalisation, algorithmic content recommendations and gamified engagement methods, would further enhance the theoretical and practical understanding of the evolving OTT ecosystem.

This study presents a novel provider-centric approach to assess the popularity of OTT platforms in India by constructing and empirically examining a composite index rooted in technological, economic and content-related attributes. The research explains OTT platform performance through DCT, with TAM and EVT providing complementary insights into consumer evaluations of provider capabilities (Davis, 1989; Eccles and Wigfield, 2002; Teece, 2007). The findings confirm that TS, VM and CD are significant predictors of CP, while freemium and hybrid subscription models play a significant moderating role in strengthening the influence of technology-driven platform features (Wu et al., 2024). The findings extend provider-centric research on OTT platform popularity beyond consumer adoption metrics. The study also demonstrates the applicability of exploratory composite-index modelling for platform-level OTT research in emerging digital markets. At the same time, the practical implications guide OTT providers on optimising their offerings for better growth and engagement (Choi and Varian, 2012). As the digital media landscape continues to evolve and diversify, future research should incorporate cross-market dynamics, AI-driven personalisation and changing consumer expectations to improve the adaptability and long-term relevance of OTT platforms. Such approaches may further strengthen the strategic alignment between platform capabilities and evolving user needs (FICCI and EY, 2023).