Behavioral Intention as an Impact of Online Shopping Experience & Trust: Case Study of Generation Z Consumers of Several E-Commerce Sites in Bandung

1. Introduction

Indonesia’s digital industry has experienced remarkable growth in recent years. At the beginning of 2025, the number of internet users in Indonesia reached 220 million out of a total population of 285 million. We Are Social, through the Digital 2025 Global Overview Report, reported Kapios' analysis indicating that the number of Indonesian internet users increased by 17 million, representing a growth of 8.7% compared with the same period in the previous year (Dudhat & Agarwal, 2023).

The rapid advancement of digital technologies and the internet has substantially reshaped consumption patterns, particularly in relation to e-commerce shopping, which has become a primary channel for consumers in Indonesia. According to Statistics Indonesia (BPS) (2024), the number of e-commerce users in Indonesia reached approximately 17.8 million, or 65% of the total population. This positions Indonesia as the country with the most significant number of digital consumers in Southeast Asia (Ariansyah et al., 2021). In addition, the majority of online shoppers are aged 18–40 and come from the middle class (Kuah & Wang, 2020).

The dominance of generation z and millennials is mainly attributable to the fact that these cohorts grew up alongside the expansion of internet technologies and are familiar with digital purchasing and payment systems (APJII Research, 2025). generation z is typically defined as individuals born between 1995 and 2010. Often labelled the "internet generation", this cohort was born during a period when the internet began to enter daily life and rapidly expanded. Generation Z's ability to adapt to technological change enables them to interact readily with brands they favour (Priporas et al., 2017). E-commerce remains a key driver of Indonesia's digital economic growth. Several platforms that are expected to dominate in 2025 include Shopee, Tokopedia, Lazada, TikTok Shop, Blibli, and Zalora (Lucky Hikmat Maulana, 2022).

E-commerce is defined as the process of purchasing, selling, transferring, or exchanging products, services, or information through computer networks via the internet (Grover & Teng, 2001; Kedah, 2023). Despite rapid growth, a significant challenge lies in the shifting shopping behavior of Indonesian consumers, whose preferences are increasingly oriented towards more prudent, sustainable choices and experience-led consumption. Consumers are becoming increasingly aware of the social and environmental implications of their purchasing decisions and are seeking products that are high-quality, ethical, and aligned with their personal values.

A report by Google, Temasek, and Bain & Company (2025) states that 80% of Indonesian consumers prefer online shopping to offline shopping. Online shopping remains the dominant option, driven by convenience and broad product availability (Adinugraha et al., 2024). However, consumers are also becoming more selective and actively seeking the best offers. While online shopping offers distinct advantages and convenience, it also entails risks (Hubert et al., 2017), including: discrepancies between the ordered product and the images displayed; items arriving damaged (either during delivery or due to manufacturing defects); packing errors that lead to incorrect orders (e.g., colour, quantity, or type); goods not being delivered due to loss or delays; and fraud. Within today’s dynamic digital business landscape, consumers can switch from one e-commerce platform to another with ease, or decide not to repurchase, often as a result of a single negative experience, more attractive competitor offers, or the emergence of new values that better match their personal preferences (Hadi Mersi et al., 2025).

A key determinant of behavioral intention is customer experience. Consumers evaluate brands not only on product quality, but also on the entire set of interactions with the brand—from the first visit to a website, the purchasing process, delivery, and after-sales service (Ahmed et al., 2022). Global surveys indicate that more than 70% of consumers are willing to leave an e-commerce platform after just one poor experience, even when they were previously loyal customers (Oliveira et al., 2022). Examples include delivery delays, transaction errors, and customer service responses that lack empathy or responsiveness, as well as payment processes that are difficult or perceived as insecure, and inadequate privacy protection (Kumar & Ayodeji, 2021).

Given continuously evolving consumer trends, e-commerce firms must become more adaptive and innovative in responding to intensifying competition, leveraging technology to create more personalised shopping experiences (Sharma et al., 2023). A growing preference for more interactive shopping experiences represents another emerging trend. This suggests that e-commerce is no longer merely a platform for purchasing products; it has become an increasingly integral part of Indonesia's digital lifestyle. Beyond customer experience, a further central challenge for e-commerce is building and maintaining consumer trust in online services and products. Trust is crucial because consumers cannot physically inspect products prior to purchase; consequently, they rely heavily on platform reputation, reviews, and transaction security (Mittameedi et al., 2025).

In business contexts, trust does not emerge instantly; it must be cultivated from the outset. To remain competitive in an era dominated by online shopping, e-commerce platforms must develop loyal customers who genuinely trust the superiority of online services. In this regard, trust reflects consumers’ belief in the platform’s capability to ensure the security of online transactions (Shao & Yin, 2019). Reviews of online shopping research also suggest that two variables—online shopping experience and trust—are significant in shaping online purchasing behavior (Chetioui et al., 2020).

Bandung, the largest city in West Java Province, is often referred to as Kota Kembang (the "Flower City") due to its beauty, and was historically known as Paris van Java for its resemblance to Paris, France. The prevalence of shopping malls and factory outlets has also positioned Bandung as a hub for shopping and culinary tourism. In 2025, the population of Bandung was approximately 2.5 million. This sizeable population presents a significant opportunity for businesses—especially e-commerce providers—to serve the generation z segment (Vieira et al., 2020). Generation Z’s interest in online shopping in Bandung is relatively high, driven by easy access and a preference for shopping experiences that are integrated with social media. Fashion is among the most popular categories for generation z in online shopping (Do et al., 2023). Individuals are often recognised through the fashion they wear, as it can reflect and express identity and social status. Fashion involvement is closely linked to personal characteristics, particularly among younger consumers (Theocharis et al., 2025).

A preliminary survey conducted by the researchers indicates that the behavioral intention of generation z consumers in Bandung towards e-commerce remains low. This is likely due to less memorable online shopping experiences and relatively low consumer trust in e-commerce platforms. Accordingly, this study aims to examine the online shopping experience, trust, and behavioral intention among generation z consumers in Bandung, as well as the interrelationships between these variables.

2. Method

This exploration is clear and confirmation-based. The unit of examination includes several e-commerce platforms in the city of Bandung, such as Tokopedia, Shopee, Bukalapak, Lazada, and Blibli. The selection of these e-commerce platforms was based on their widespread use and high traffic in the Bandung area. The inspection strategy employed is purposive sampling. Generation z customers who enjoy online shopping during Harbolnas in Bandung make up the observation unit. The base example size is 100 people. Online questionnaires and in-person interviews are two methods of data collection. Multiple linear regression is used as the analytical technique (Ofosu-Boateng, 2020).

2.1 Identify Subsections

To enhance transparency and replicability, the method section is organised into three labelled subsections: (i) participant characteristics, (ii) sampling procedures and data collection, and (iii) analytical approach. The present study employs a descriptive and explanatory (verification) design, focusing on generation z consumers in Bandung who have recently made fashion purchases via e-commerce platforms. Data were collected using an online questionnaire (google forms) supplemented by interviews, and the hypothesized relationships were tested using multiple linear regression (Hülk et al., 2018).

2.2 Participant (Subject) Characteristics

Participants were generation z consumers residing in Bandung, operationalised as individuals aged 15–30 years, who had purchased fashion products through e-commerce within the previous month. The e-commerce platforms referenced in the study include Shopee, Tokopedia, Lazada, TikTok Shop, Blibli, and Zalora. Eligibility criteria, therefore, comprised: (a) aged 15–30; (b) Bandung-based consumer; and (c) at least one fashion purchase via the specified e-commerce channels in the past month. Exclusion criteria were applied logically, excluding respondents outside the specified age range, those who had not made a fashion purchase online in the last month, or those without experience purchasing through e-commerce during the specified period.

2.3 Sampling Procedures

A purposive sampling approach was employed, with a target of at least 100 respondents. Data collection was conducted through a google forms questionnaire, supplemented by interviews to enrich the understanding of participant responses and contextualize behavioral intentions (Campbell et al., 2020). The primary statistical technique used was multiple linear regression, which was employed to examine the effects of online shopping experience and trust on behavioral intention. Reporting note for APA-style completeness: the manuscript text provided does not specify the recruitment channels, response rate (percentage approached who participated), the number who self-selected into the study, incentives (if any), the precise data-collection setting(s), or ethics/IRB information. These elements can be added as brief statements (e.g., “Participation was voluntary; no incentives were offered; ethics approval was obtained from…”) to fully align the section with best-practice reporting standards, without adding unnecessary length.

Figure 1:  Research model

3. Results

The validity test for all questionnaire items across the examined variables shows that the r-calculated value for each item is greater than or equal to the r-table value of 0.94; therefore, all items can be considered valid. For the online shopping experience construct, Cronbach's alpha is 0.764; for trust, Cronbach's alpha is 0.746; and for behavioral intention, Cronbach’s alpha is 0.748. These results indicate that all measurement items are reliable, as Cronbach’s alpha values exceed the threshold of 0.70 (Ghozali, 2018).

3.1 Descriptive Evaluation

From Table 1, it can be seen that generation z consumers' online shopping experience on e-commerce platforms is interpreted as moderately favourable, with an overall mean score of 3.30, as it falls within the interval of 2.60–3.39 (Aulia, 2020). The lowest scores are observed for the statements relating to the level of honesty and the security of delivery on e-commerce platforms.

Table 1: Respondents’ Perceptions of Online Shopping Experience (X1)

Source: Processed primary data, 2025.

From Table 2, it can be observed that generation z consumers' trust in e-commerce platforms is moderately favourable, with an average score of 3.28, as it falls within the interval of 2.60–3.39. The lowest scores relate to the statements that e-commerce platforms are honest, provide detailed information, ensure security during the purchasing process, and are generally safe platforms.

Table 2: Respondents’ Perceptions of Trust (X2)

Source: Processed primary data, 2025.

From Table 3, it can be seen that generation z consumers’ behavioral intention towards e-commerce platforms is moderately favourable, with an average score of 3.26, as it falls within the interval of 2.60–3.39. The lowest scores are found for statements concerning consumers' intentions to share positive information with friends and family, as well as their intention to recommend the platform to others.

Table 3: Respondents’ Perceptions of Behavioral Intention (Y)

Source: Processed primary data, 2025.

3.2 Analysis of Multiple Linear Regression

Table 4 presents the results of a multiple linear regression model, in which Behavioral Intention (Y) is the dependent variable and is predicted by Online Shopping Experience (X1) and Trust (X2) (SPSS output). The estimated regression equation is reported as Y = 1.881 + 1.034X1 + 0.210X2 + e. The constant term (B = 1.881, p = .000) represents the expected level of behavioral intention when both predictors are held at zero. Controlling for trust, Online Shopping Experience shows a positive and statistically significant association with Behavioral Intention (B = 1.034, SE = 0.122, β = 0.674, t = 8.573, p = .000), indicating that a one-unit increase in online shopping experience is associated with an increase of approximately 1.034 units in behavioral intention. Trust also demonstrates a positive effect (β = 0.128) with an unstandardized coefficient reported as B = 0.0210 (SE = 0.131, t = 1.613, p = .000), suggesting a comparatively minor contribution to behavioral intention when both predictors are considered simultaneously. Overall, the standardized coefficients indicate that the online shopping experience is a stronger predictor of behavioral intention (β = 0.674) relative to trust (β = 0.128). Note: The regression equation reports the trust coefficient as 0.210, whereas the coefficients table lists 0.0210. This appears to be a formatting/typographical discrepancy, and the coefficient values should ideally be aligned to avoid ambiguity in interpretation.

Table 4: Multiple Linear Regression Test

Source: Processed primary data, 2025.

3.3 Simultaneous Hypothesis Test (F-test)

Table 5 presents the ANOVA (F-test) results for the multiple regression model predicting Behavioral Intention from Online Shopping Experience and Trust (SPSS output). The regression component accounts for a Sum of Squares of 1472.511 across 2 degrees of freedom, producing a Mean Square of 736.256. The residual (error) variation is 1182.001 with 97 degrees of freedom, yielding a Mean Square Error of 12.187, while the total variability in Behavioral intention is 2654.511 across 99 degrees of freedom. The model produces an F-statistic of 60.621 with a significance value of .000, indicating that the predictors, considered jointly, explain a statistically significant proportion of variance in behavioral intention. In other words, the overall regression model is significant, supporting the conclusion that online shopping experience and trust together have a meaningful effect on behavioral intention.

Table 5: Hypothesis Test (Uji F)

Source: Processed primary data, 2025.

Table 6 reports the correlation and explanatory power of the regression model through the model summary statistics (SPSS output). The multiple correlation coefficient is R = 0.746, indicating a strong positive association between the combined predictors (online shopping experience and trust) and the dependent variable (behavioral intention). The R-squared (R²) value of 0.556 indicates that the model explains 55.6% of the variance in behavioral intention, suggesting substantial predictive capability in this context. After adjusting for the number of predictors and sample size, the Adjusted R² is 0.547, implying that approximately 54.7% of the variability in behavioral intention remains explained when model complexity is taken into account, and indicating minimal inflation of the explained variance. The standard error of the estimate is 3.492, reflecting the typical magnitude of prediction error (i.e., the average deviation of observed behavioral intention scores from those predicted by the model). Overall, these statistics suggest that the regression model provides a robust fit and that the predictors jointly offer meaningful explanatory power for behavioral intention.

Table 6: Correlation Coefficient and Coefficient of Determination

                                                                            Model Summary

Source: Processed primary data, 2025.

4. Discussion

The descriptive results suggest that generation z consumers in Bandung hold a moderately favourable evaluation of their online shopping experience; however, perceived honesty and delivery security emerge as the weakest points. This pattern is consistent with the broader view that online shopping, although convenient, is accompanied by perceived risks (e.g., product image discrepancies, delivery failures, and fraud), which can undermine the overall experience if not managed effectively (Hubert et al., 2017). In experience-led digital commerce, even a single negative episode in the customer journey can trigger adverse behavioral responses, including platform switching and reduced loyalty (Oliveira et al., 2022).

Accordingly, strengthening experiential reliability—through accurate product information, consistent fulfilment, and robust service recovery—becomes central to sustaining positive evaluations in a highly competitive e-commerce environment. Trust is also evaluated as moderately favourable. However, respondents express weaker agreement regarding the platform's honesty, the provision of sufficiently detailed information, and perceived transaction safety. In e-commerce contexts where consumers cannot physically inspect products, trust functions as a pivotal mechanism that reduces uncertainty and supports purchase decisions, with consumers relying heavily on reputation, reviews, and transaction security cues (Mittameedi et al., 2025).

From an institutional perspective, trust is reinforced when platforms implement credible safeguards and governance mechanisms that signal reliability and protection (Shao & Yin, 2019). Therefore, improving trust among generation z shoppers requires not only transparent and truthful communication, but also visible security assurances (e.g., secure payment protocols and identity verification), as well as stronger information quality across product pages and policies.

Behavioral intention is likewise moderately favourable, with comparatively weaker intentions to engage in advocacy behaviors—specifically, sharing positive messages with friends/family and recommending the platform to others. This is noteworthy because behavioral intention is typically shaped by the cumulative quality of customers’ interactions with the platform, spanning search, purchase, delivery, and after-sales service (Ahmed et al., 2022). For generation z, who are highly connected and accustomed to digitally mediated brand interactions, recommendation behavior is particularly sensitive to perceived authenticity and consistency in the experience they receive (Priporas et al., 2017).

In this study, the weaker advocacy intentions plausibly reflect residual doubts about honesty and delivery security, which can constrain consumers' willingness to endorse an e-commerce provider publicly.

The inferential findings confirm that online shopping experience and trust significantly predict behavioral intention, with online shopping experience emerging as the stronger predictor (β = 0.674) compared with trust (β = 0.128). This pattern is theoretically plausible because experience represents a holistic evaluation of the entire customer journey. At the same time, trust—although essential—may operate as a more specific judgement about credibility and transaction assurance. The model's explanatory power is substantial (R² = 0.556), indicating that these two factors account for a meaningful proportion of variance in behavioral intention among Bandung's generation z consumers. Given Bandung's strong positioning as an urban consumption hub and the prominence of fashion as a key online category for generation z, e-commerce providers targeting this segment should prioritise experience personalisation and service reliability while simultaneously strengthening trust signals through transparency and demonstrable security features (Do et al., 2023; Theocharis et al., 2025; Vieira et al., 2020).

5. Conclusion

Overall, the findings indicate that generation z consumers in Bandung report moderately favourable evaluations of online shopping experience, trust, and behavioral intention towards e-commerce. However, the weakest aspects of the online shopping experience relate to perceived honesty and delivery security. Trust is similarly constrained by lower agreement that e-commerce platforms are honest, provide sufficiently detailed information, and are secure and safe during the purchasing process. Behavioral intention is also only moderately favourable, with comparatively lower intentions to share positive information with friends/family and to recommend purchases made via e-commerce to others. Inferential results further confirm that online shopping experience and trust significantly predict behavioral intention, explaining 55.6% of its variance, while the remainder is attributable to other factors not examined in this study. From a practical perspective, e-commerce providers should prioritise ethical and transparent business practices, particularly honesty in communications and operations, and strengthen fulfilment assurance through secure packaging, delivery insurance, and real-time tracking. Trust can be reinforced by providing accurate product descriptions, transparent pricing and fees, clear return and warranty policies, and robust transaction and data protections, including SSL/TLS certification and two-factor authentication (2FA). Finally, to stimulate positive word-of-mouth and recommendation intentions among generation z, platforms should enhance service quality and deliver more memorable customer experiences, thereby increasing consumer trust and encouraging advocacy behaviors.

Author Contributions: All authors contributed to this research.

Conflict of Interest: The authors declare no conflict of interest.

Informed Consent Statement/Ethics Approval: Not applicable.

Declaration of Generative AI and AI-assisted Technologies: This study has not used any generative AI tools or technologies in the preparation of this manuscript.

Acknowledgements: The authors would like to express their gratitude to Universitas Widyatama for funding this research through the Bureau of Research, Community Service, and Intellectual Capital (Biro P2M), Contract No. 137/HPW/P2M-UTAMA/VII/2025, dated 29 July 2025.