Infrastructure, Not Ease of Use, Drives Gen Z Tourists’ Travel App Intentions

A survey of nearly 400 young travelers in Bangladesh has found that the single strongest driver of their intention to use smart tourism smartphone apps is not how useful the apps seem or how fun they feel, but whether the practical conditions for using them are actually in place. Facilitating conditions, a construct covering reliable internet connectivity, device compatibility, app availability, and access to technical support, outweighed every other factor measured in the study, which applied an integrated technology-adoption framework to one of the most connected generations in existence. The finding carries particular weight in a developing-economy context, where infrastructure constraints can shape digital behavior as much as user attitudes do.

The research, published as an open-access article in Discover Analytics on 4 August 2026, set out to identify which factors influence Generation Z tourists’ intention to use smartphone applications in tourism marketing. A team of researchers adapted the Unified Theory of Acceptance and Use of Technology 2, a widely used model of consumer technology adoption, and combined it with the logic of the Theory of Planned Behavior. Rather than modeling the behavioral theory’s constructs as separate variables, the study embedded them conceptually within the adoption model: attitude was reflected through performance expectancy and hedonic motivation, subjective norm through social influence, and perceived behavioral control through facilitating conditions. The authors describe this as a parsimonious integration designed to preserve conceptual clarity while avoiding redundancy in the model.

Data were collected through a voluntary, self-administered online questionnaire distributed via Facebook, WhatsApp, Instagram, and email between February and April 2024. The survey gathered 530 responses, of which 399 remained after data cleaning and were analyzed. Respondents answered items measuring performance expectancy, effort expectancy, social influence, facilitating conditions, enjoyment, behavioral intention, and use behavior on five-point Likert scales, with all measurement items adapted from previously published scales and modified to fit the smart tourism app context. The target population comprised young Bangladeshi tourists with experience using travel-related smartphone applications.

The researchers are candid about the sampling approach and its limits. Because recruitment relied on online convenience and self-selection rather than probability-based random sampling, the sample likely over-represents respondents with reliable internet access and higher app familiarity. Digitally marginalized individuals without regular internet access or a social media presence were probably excluded. The team also acknowledges that the age distribution includes some respondents outside standard Generation Z birth-year definitions, since the sample is Gen Z-dominant but includes some older young adults. They therefore interpret the results as indicative of digitally connected young tourists rather than the entire Gen Z population of Bangladesh, and they caution against population-level generalization.

Analysis proceeded in stages using SmartPLS 3.2.9 and SPSS 27. The measurement model was first evaluated through outer loadings, composite reliability, Cronbach’s alpha, and average variance extracted, with discriminant validity assessed using the Fornell-Larcker criterion. Most constructs performed well: composite reliability values ranged from 0.843 for enjoyment to 0.923 for social influence, Cronbach’s alphas fell between 0.792 and 0.895, and average variance extracted exceeded the 0.50 threshold in all cases, with enjoyment the lowest at 0.519. Common method bias was assessed with Harman’s single-factor test, in which the first factor accounted for less than half of the total variance, suggesting the issue was unlikely to be a serious concern.

One measurement result stands out as a caveat to the headline finding. The correlation between facilitating conditions and behavioral intention was 0.916, exceeding the square root of the average variance extracted for both constructs, which indicates a lack of discriminant validity between them. The authors note that facilitating conditions may function as an intention-proximal construct rather than a fully distinct antecedent in this dataset, and they flag this potential construct overlap as a limitation of the study.

The structural model tested six hypotheses, of which five were supported. Behavioral intention strongly predicted actual use behavior, with a path coefficient of 0.704 and a p-value below 0.001. Among the antecedents, facilitating conditions showed by far the largest effect on intention, with a beta of 0.596 and a p-value below 0.001. Performance expectancy followed at 0.128 (p = 0.007), then social influence at 0.106 (p = 0.018), and enjoyment at 0.092 (p = 0.030), all statistically significant. Effort expectancy, the perceived ease of using the apps, was not significant, with a beta of 0.074 and a p-value of 0.073.

The failure of effort expectancy to reach significance is the study’s most theoretically interesting result. Prior research on technology adoption among digital natives has produced mixed findings for this construct, with some studies reporting positive relationships and others finding it insignificant because perceived ease of use varies little across modern applications. The authors offer several possible explanations beyond the familiar digital-native account: restricted variance because most mainstream apps are already easy to navigate, a possible measurement-context mismatch if respondents interpreted items as general smartphone ease rather than tourism-app-specific tasks, or suppression effects given the very high association between facilitating conditions and behavioral intention. The result suggests that ease of use may no longer differentiate apps for a cohort already comfortable with smartphone interfaces.

The demographic profile of the sample adds context to these findings. Male respondents made up 70 percent of the final sample and females 30 percent. Participants aged 25 to 30 accounted for 47.5 percent, followed by those aged 20 to 25 at 30 percent. The sample was highly educated, with 45 percent holding a bachelor’s degree and 43.75 percent a master’s degree, and students formed the largest occupational group at 42.5 percent, followed by public sector employees, self-employed individuals, and private sector employees. All respondents confirmed regular use of smart tourism apps, confirming the relevance of the sample to the study context, while also underscoring that the findings describe experienced app users rather than novices.

For the tourism industry, the authors draw preliminary, context-specific implications rather than definitive prescriptions. Given the dominance of facilitating conditions, tourism stakeholders and app developers may benefit from focusing on stable internet connectivity, device compatibility, and effective technical support to encourage sustained use. The significant roles of performance expectancy and social influence point to marketing that clearly communicates an app’s functional value and leverages peer approval, influencer endorsements, and user-generated content, while the enjoyment effect suggests that visually attractive, interactive, and gamified interfaces can support engagement. The authors also recommend that future research apply stricter cohort screening, conduct multi-group analyses comparing Gen Z-only respondents with older young adults, re-estimate the measurement model with additional discriminant validity techniques, incorporate constructs such as habit, price value, trust, and privacy concerns, and use longitudinal or experimental designs to strengthen causal inference. As they conclude, adoption of smart tourism apps among this cohort appears to be a multidimensional process shaped by functional, social, experiential, and infrastructural factors together.

Subject of Research: Factors influencing Generation Z tourists' intention to use smart tourism smartphone apps in Bangladesh, tested with an integrated UTAUT2 and Theory of Planned Behavior model.

Article Title: Exploring factors influencing gen Z tourists’ smartphone app usage using a UTAUT2–TPB model

Article References: Rocky, K. H., Karim, R., Acharjee, S. K., Monowar, M. M. U., & Bhowmik, R. C. (2026). Exploring factors influencing gen Z tourists’ smartphone app usage using a UTAUT2–TPB model. Discover Analytics, 4(1), Article 33. https://doi.org/10.1007/s44257-026-00076-3

Image Credits: AI Generated

DOI: 10.1007/s44257-026-00076-3

Keywords: Generation Z, smart tourism, smartphone apps, technology adoption, UTAUT2, Theory of Planned Behavior, Bangladesh, PLS-SEM, facilitating conditions, behavioral intention, survey research, tourism marketing