The power of ai driven personalization in encouraging customer engagement and repurchase intention on shopee vietnam
INTRODUCTION
E-commerce has rapidly transformed business practices by enabling seamless online transactions for products and services Leading companies in the global e-commerce industry have introduced innovations aimed at enhancing user experiences, streamlining transactions, and meeting diverse customer needs.
In 2023, Shopee emerged as the top e-commerce platform in Southeast Asia, capturing a significant 48% market share and achieving a gross merchandise value (GMV) of US$55.1 billion (Chung, 2024) Its dominance is especially pronounced in Vietnam, highlighting its strong presence in the region.
In 2023, Shopee captured approximately 70% of the e-commerce retail market, outpacing competitors like Lazada, Tiki, Sendo, and TikTok Shop (Vietdata, 2024) The platform has excelled in adopting advanced technologies, particularly AI-driven personalization, which has significantly improved the shopping experience By integrating AI, Shopee has enhanced customer engagement through personalized recommendations and tailored purchasing experiences (Perez-Vega et al., 2021) This focus on personalized product suggestions has created a unique and engaging environment, allowing Shopee to effectively predict customer preferences and deliver customized content, ultimately boosting user engagement and purchase intent, and solidifying its dominant market position.
Artificial intelligence (AI) is revolutionizing e-commerce by enhancing personalized shopping experiences, which are essential for building strong consumer relationships in today's information-rich digital landscape As businesses strive to stand out in a competitive market, the demand for tailored experiences has positioned AI at the forefront of e-commerce innovation.
AI enhances personalized shopping experiences, fostering a deeper connection with consumers This relevance boosts customer engagement, leading to increased purchase intentions and higher sales (Raji et al., 2024).
The importance of artificial intelligence (AI) has surged, reflected in the increasing number of publications on the subject From 2010 to 2022, the total count of AI papers nearly tripled, rising from approximately 88,000 to over 240,000, indicating its growing relevance across various fields Research on AI in e-commerce has spanned three decades, yielding over 4,000 scholarly articles since 1991 Additionally, there has been a steady rise in the annual publication rate of AI in e-commerce since 2013, highlighting the escalating interest from both academic and corporate sectors.
Figure 1- 1 Number of AI publications in the world, 2010 – 2022
(Source: Center for Security and Emerging Technology, 2023)
Figure 1- 2 Number of publications on AI in e-commerce per year
Artificial Intelligence is essential in the growth of e-commerce, enhancing user experience and shaping consumer behavior One of AI's key contributions is personalization, where advanced algorithms analyze large datasets, including user preferences and browsing history, to provide tailored product recommendations This level of customization boosts customer engagement and increases the likelihood of repeat purchases by creating a more relevant and engaging shopping experience.
Despite the increasing research on artificial intelligence (AI), a significant knowledge gap exists in understanding how AI-driven personalization affects consumer engagement and repurchase intentions, especially in the Vietnamese e-commerce market While AI technologies have been widely adopted across various industries to improve business operations, their specific impact on customer behavior, particularly in terms of personalization, has not been thoroughly explored Current studies primarily focus on technological advancements and broad applications rather than their effects on consumer interactions.
AI, although neglects the impact of personalization tools on consumer engagement and their influence on repurchase decisions.
The need for focused research on AI-driven customization in Vietnam's e-commerce sector is evident Understanding this relationship can provide valuable insights for businesses and marketers, enabling them to align AI personalization strategies with customer needs and preferences, ultimately boosting sales and achieving sustainable growth.
This research explores the relationship between AI-driven personalization and consumer engagement, specifically focusing on their combined effect on repurchasing intentions in Vietnam's e-commerce sector By examining key components of AI personalization, such as recommendation systems, chatbots, and augmented reality, the study aims to enhance understanding of AI's influence on customer behavior in digital marketplaces The findings will offer a framework for businesses to effectively leverage AI technology, ensuring that their personalization strategies not only attract customers but also convert engagement into actual repurchase actions.
This research takes an overview that addresses the following general questions in relation to AI-driven personalization features impact both customer engagement and repurchase intention in context Shopee Vietnam:
RQ1:How do specific AI-driven personalization features (Recommendation System, Chatbot, Augmented Reality) directly influence Repurchase Intention?
RQ2:How do specific AI features (Recommendation System, Chatbot,
Augmented Reality) influence engagement? How does engagement mediate repurchase intention?
The overall aim of this research is to identify and explore the major determinants
RO:Investigate how AI-driven personalization impacts customer engagement and repurchase intention on Shopee Vietnam.
LITERATURE REVIEW
E-commerce
Electronic commerce (e-commerce) leverages the Internet and modern communication technologies to facilitate business activities and transactions (Song et al., 2019) It encompasses various online activities, including the buying, selling, and exchanging of goods, services, and information (Adawiyah et al., 2024) Unlike traditional physical stores, e-commerce platforms offer consumers the convenience of accessing a vast selection of products from around the world at any time and from any location using internet-enabled devices This ease of access, combined with the ability to quickly compare prices across multiple retailers, empowers consumers to make informed purchasing decisions and secure better deals (Gayam, 2021).
E-commerce offers substantial benefits to firms by transcending geographical limitations, facilitating access to a worldwide clientele Moreover, online platforms diminish operational expenses by obviating the necessity for physical shops, enabling organizations to reinvest in augmenting their digital presence and executing customer-centric initiatives Furthermore, e-commerce systems produce huge quantities of user data via browsing activity, purchase history, and online interactions This data is a vital asset for organizations, providing insights into client preferences and allowing for the delivery of individualized shopping experiences via targeted marketing initiatives.
As e-commerce has evolved, so have consumer expectations, shifting from a focus on convenience to a demand for personalized, engaging, and seamless experiences Today, e-commerce is viewed not just as a transactional platform, but as a means to foster customized, interactive, and value-driven relationships between businesses and their customers.
Artificial Intelligence in E-commerce
Artificial Intelligence (AI) was first introduced at the Dartmouth Conference in 1956 by pioneers such as Claude Shannon, John McCarthy, Nathaniel Rochester, and Marvin Minsky They defined AI as the ability of machines to replicate aspects of learning and intelligence, asserting that all facets of learning and intelligence could be precisely described to enable machine simulation This definition includes key skills such as language processing, abstraction, problem-solving, pattern recognition, and adaptive learning.
AI technology automates cognitive tasks by mimicking and improving human intelligence, utilizing approaches from computer science, psychology, biology, and linguistics Key advancements in AI include machine learning and interactive learning, which are crucial for its current applications.
Artificial intelligence is transforming e-commerce by enhancing customer experiences and streamlining business operations Its integration enables platforms to deliver personalized and engaging shopping experiences through technologies such as personalized product recommendations, chatbots, dynamic pricing, demand forecasting, intelligent search, and logistics optimization By leveraging machine learning and advanced data analytics, businesses can analyze vast amounts of customer data, generate real-time insights, and automate decision-making processes These innovations not only improve operational efficiency but also boost customer engagement, leading to increased sales and fostering long-term loyalty.
AI-Driven Personalization in E-Commerce
Personalization refers to the customization of products and shopping experiences to match individual consumer preferences based on their data (Hallikainen et al., 2022) AI-driven personalization employs advanced algorithms and machine learning to analyze large datasets, including user behavior and past interactions This analysis enables e-commerce platforms to predict and deliver relevant content, fostering tailored experiences that enhance personal connections with users.
The dynamic and iterative process of AI algorithms enhances user experience by continuously collecting data and refining recommendations based on individual preferences This adaptability ensures that personalization remains relevant as user behaviors evolve The impact of tailored experiences is significant, as personalized recommendations boost client engagement, increase satisfaction, and simplify decision-making by reducing the effort needed to explore extensive web catalogs (Raji et al., 2024).
AI-driven personalization is crucial for success in the competitive e-commerce landscape Utilizing a recommendation system, tailored marketing content, and chatbots can significantly enhance customer purchase likelihood and foster long-term loyalty.
Recommendation systems play a crucial role in enhancing user experiences on digital platforms, particularly in e-commerce By employing advanced algorithms, these systems deliver personalized product suggestions that reduce information overload and boost customer satisfaction and engagement Tailoring recommendations to individual preferences helps e-commerce platforms build stronger consumer relationships and positively impact purchasing decisions.
Recommendation systems are automated software tools designed to suggest products based on user preferences and behaviors These systems draw on various interdisciplinary fields, including information retrieval, machine learning, decision support systems, and text categorization (Hussien et al., 2021).
In e-commerce, these systems alleviate cognitive burden by offering customers pertinent and focused recommendations, hence enhancing the efficiency and pleasure of the buying experience.
Recommendation systems play a crucial role in e-commerce by enhancing personalization and user experiences Collaborative Filtering identifies user behavior patterns, such as purchases and ratings, to suggest items based on user similarities without needing item-specific metadata In contrast, Content-Based Filtering focuses on product characteristics and a user's past interactions to recommend items with similar attributes, ensuring highly personalized suggestions Hybrid Systems combine the strengths of both approaches, leveraging user behavior insights and item data for improved accuracy and adaptability in dynamic e-commerce settings Additionally, Matrix Factorization techniques uncover hidden patterns in user-item interactions, enabling precise recommendations even with limited data or complex relationships.
AI-driven chatbots are sophisticated conversational software agents intended to participate in sequential encounters with users on digital platforms (Adam et al.,
In the e-commerce sector, chatbots play a crucial role in enhancing customer support by automating responses and streamlining user interactions These conversational agents typically appear as chat box prompts on various platforms, allowing customers to express their questions and receive tailored, contextually relevant answers By leveraging historical consumer data, AI-driven chatbots deliver personalized and individualized user experiences.
The implementation of chatbots significantly reduces the need for human agents by automating routine customer interactions Shopee’s Shop AI Assistant is an advanced AI chatbot designed to provide 24/7 customer support, ensuring that buyer inquiries, from pre-sales to post-sales, are handled promptly Beyond answering frequently asked questions, the Shop AI Assistant aids customers throughout their shopping experience and offers personalized product recommendations Additionally, vendors can improve this chatbot by setting up automated FAQs and welcome messages, thereby enhancing communication efficiency and boosting customer retention.
Chatbots like Shopee's effectively handle a large volume of inquiries, ensuring scalability and operational efficiency through automated responses They provide continuous availability and include escalation options for human intervention when customers are dissatisfied This blend of automation and personalization allows e-commerce platforms to improve customer engagement and satisfaction.
Augmented Reality (AR) bridges the gap between digital and physical worlds by overlaying digital content onto real environments, enabling consumers to interact with products more effectively In the realm of e-commerce, AR provides detailed and contextual information about products, addressing common challenges in understanding size, features, and other physical attributes of items displayed online (Diana et al., 2023) This technology significantly improves the virtual understanding of products, making the online shopping experience more interactive and engaging.
The Virtual Try-On feature in augmented reality (AR) significantly enhances e-commerce by allowing shoppers to interactively visualize products in a realistic manner before making a purchase This technology improves customer decision-making and satisfaction by simulating in-store experiences, particularly benefiting industries such as cosmetics and fashion, where understanding product fit and aesthetics is crucial.
Shopee has integrated advanced AR features, such as Shopee SkinCam and BeautyCam, to enhance its shopping platform These tools utilize AR-enhanced Virtual Try-On technology, allowing consumers to digitally test cosmetic products Major brands like L’Oreal, Maybelline, and DAZZLE ME leverage these features, helping shoppers visualize how items like lipsticks and foundations will look on their skin This innovation improves the shopping experience, reduces uncertainty, and increases customer confidence in their purchasing decisions.
Customer Engagement
Customer engagement is a multifaceted concept studied across diverse fields like marketing, information systems, and organizational behavior It has been defined in various ways, including as a psychological process, a behavioral expression, and a psychological state Despite these differing perspectives, research consistently shows a positive relationship between higher consumer engagement and better business outcomes, such as increased brand loyalty and repeat purchases.
This study characterizes behavioral engagement as the measurable interactions customers exhibit on an e-commerce platform, shaped by their experiences with AI-driven features and positive cognitive and emotional responses Key indicators of behavioral engagement include activities such as viewing products, liking items, and saving goods to wish lists Additionally, it encompasses consumer retention, where customers repeatedly choose to return to the platform, and referrals, as satisfied customers recommend the platform to friends and family.
Repurchasing Intention
Repurchase intention reflects a customer's likelihood of buying products or services again from the same vendor, influenced by current conditions and future expectations (Hellier et al., 2003) In online retail, repurchasing is defined as a consumer's repeated use of an e-commerce platform to purchase goods from a specific store (Chou and Hsu, 2016).
Research emphasizes the critical role of repurchase intention for e-commerce platforms, as it directly impacts the profitability of online suppliers (Harris and Goode, 2004) This intention is essential for long-term business success, reflecting customer behavior that indicates a willingness to purchase specific products or services again, along with a positive evaluation of previous acquisitions.
2.6 The Stimulus - Organism - Response (S-O-R) Model
This study is grounded in the Stimulus-Organism-Response (S-O-R) model, introduced by Mehrabian and Russell in 1980 The S-O-R hypothesis posits that environmental stimuli (S) are processed by an organism (O) through cognitive, emotional, and physiological mechanisms, leading to behavioral responses (R) such as approach or avoidance The model has been extensively applied in information systems and e-commerce to evaluate customer engagement in digital environments, as evidenced by recent research (Bag et al., 2022; Cakraputr et al., 2024; Gupta et al., 2023) These findings reinforce the S-O-R framework's effectiveness in analyzing consumer behavior within digital markets.
The S-O-R model is essential for understanding consumer satisfaction and repurchase intentions in marketing and e-commerce The shift from physical markets to digital "market spaces" has been driven by the digitalization of commerce, with AI technologies playing a key role Companies are increasingly leveraging AI to improve consumer engagement by personalizing experiences, tracking purchase intentions, and optimizing conversion rates on digital platforms.
AI-generated stimuli significantly influence users' cognitive and emotional states, acting as the "organism" in the S-O-R paradigm By enhancing customer engagement on e-commerce platforms, AI facilitates better decision-making and minimizes cognitive dissonance, ultimately leading to higher conversion rates.
Customer engagement, as outlined by the S-O-R model, plays a crucial role in shaping customer behavior By gathering feedback and analyzing user interactions, companies can assess customer satisfaction, which directly affects repurchase intentions Higher satisfaction levels lead to increased repurchase intentions, highlighting the importance of AI technology in promoting positive consumer experiences Thus, the S-O-R model provides a valuable framework for understanding how AI-driven personalization and client engagement influence repurchase intentions in the e-commerce industry.
Hypothesis Development
AI-driven personalization is essential for enhancing customer engagement on e-commerce platforms Research by Balasubramanian (2024) highlights that personalized recommendations, derived from analyzing customer behavior and historical interactions, create a more engaging purchasing experience This level of customization not only improves the browsing experience but also increases the likelihood of return visits and boosts conversion rates Additionally, studies indicate that tailored content significantly enhances user engagement by increasing the time spent on platforms and leading to more frequent purchases (Gentsch, 2019; Kumar & Rajan, 2020).
Recent empirical research supports the positive impact of AI-driven tailored recommendation systems on customer engagement Asante et al (2023) found a notable correlation between these systems and psychological involvement (r = 0.246) as well as behavioral engagement (r = 0.164) These results underscore the effectiveness of recommendation systems in fostering meaningful interactions between users and e-commerce platforms, thereby boosting engagement across multiple dimensions.
Based on these insights, we propose the following hypothesis:
H1: AI-driven Rersonalized Recommendation Systems positively influence Customer Engagement in Shopee Vietnam.
H2: AI-driven Rersonalized Recommendation Systems positively influence Repurchase Intention in Shopee Vietnam.
AI-powered chatbots play a vital role in enhancing client engagement on e-commerce platforms According to Tamara et al (2023), Generation Z consumers view chatbots as valuable tools that significantly improve their online shopping experience By offering instant access to information and streamlining the purchasing process, chatbots boost customer satisfaction and loyalty The study highlights that these virtual companions foster increased engagement, particularly among younger, tech-savvy users, by enriching their interactions with the platform.
Additional evidence for the efficacy of chatbots is provided by Chung et al.
In 2020, a study explored the impact of chatbot-driven e-services on the premium fashion retail industry The findings revealed that these services significantly enhance brand-consumer interactions by providing engaging and interactive experiences that resemble the personalized care typically offered in face-to-face settings Such interactions create memorable client experiences, which are crucial for fostering long-term engagement.
Based on these insights, we propose the following hypothesis:
H3: AI-powered Chatbots positively influence Customer Engagement in Shopee
H4: AI-powered Chatbots positively influence Repurchase Intention in Shopee
Augmented Reality (AR) is transforming e-commerce by significantly enhancing customer engagement through immersive and interactive shopping experiences According to Gujarathi et al (2024), AR technologies enable brands to create emotionally resonant connections with customers, leading to improved perceptions, interactions, and satisfaction across various touchpoints By strategically integrating AR into e-commerce, brands can elevate audience engagement, strengthen consumer relationships, and increase customer loyalty.
Recent studies by Enyejo et al (2024) and Shahikanth et al (2024) highlight the significant impact of augmented reality (AR) on customer engagement and purchasing behavior in retail Enyejo et al found that immersive AR elements enhance user interaction and enjoyment, leading to increased purchase propensity Similarly, Shahikanth et al emphasized AR's ability to bridge physical and digital commerce, enhancing the buying experience through features like virtual try-ons and product visualization, which boost consumer confidence and satisfaction.
Based on these insights, we propose the following hypothesis:
H5: AI-driven Augmented Reality positively influences Customer Engagement in Shopee Vietnam.
H6: AI-driven Augmented Reality positively influences Repurchase Intention in
Research shows that customer interaction significantly influences consumer behavior, particularly in repurchase intentions Highly engaged customers feel empowered through their interactions with a company, leading to positive outcomes such as an increased likelihood of repurchasing (Rather & Hollebeek, 2021) This engagement fosters favorable perceptions of the company's products and services, as engaged consumers tend to develop stronger beliefs about the value and quality of offerings compared to less involved customers (Harrigan et al., 2018).
As customer engagement intensifies, so does the emotional connection to the brand, significantly influencing purchasing behavior and fostering loyalty and repeat transactions (Shen et al., 2022) According to Hosany et al (2020), this engagement stems from the emotional ties consumers form with their products, enhancing their likelihood of repurchase Additionally, Shabankareh et al (2024) highlight that behaviors driven by engagement play a vital role in shaping repurchase intentions by cultivating lasting consumer-brand relationships.
Based on these findings, we propose the following hypothesis:
H7: Customer Engagement positively affects Repurchase Intention in Shopee
RESEARCH METHODOLOGY
Research method
This study utilizes a quantitative research methodology to analyze the impact of AI-driven personalization—specifically Recommendation Systems, Chatbots, and Augmented Reality (AR)—on customer engagement and repurchase intention in Shopee Vietnam Grounded in the Stimulus-Organism-Response (S-O-R) framework, it identifies AI-driven personalization as stimuli (S), consumer engagement as the organism (O), and repurchase intention as the response (R) The research aims to assess how effectively AI-driven personalization enhances user interactions and influences repeat purchase behavior on the platform.
This study employs hypothesis testing on consumers familiar with Shopee's AI-driven personalization for online purchases A cross-sectional survey design was used to gather a robust and generalizable dataset, with primary data collected through a structured questionnaire distributed via Google Forms The questionnaire consists of 25 items designed to evaluate five key factors, utilizing a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree).
Respondents were selected based on three key criteria: they must reside in Vietnam, have completed at least one transaction on Shopee, and possess familiarity with AI-driven personalization features, such as personalized product recommendations, AI chatbots, and augmented reality Virtual Try-On experiences.
A pilot study involving 15 selected consumers was carried out to assess the content validity and clarity of the survey Participant feedback was instrumental in enhancing the wording, organization, and comprehensibility of the survey items prior to the main data collection phase The refined questionnaire was subsequently shared across social networks, online consumer groups, and social media platforms over two months, from December 2024 to January 2025, with the goal of gathering a diverse and representative sample of Shopee consumers.
The data analysis will include descriptive statistics, reliability testing using Cronbach’s Alpha, structural equation modeling (SEM), variance inflation factor (VIF), and exploratory factor analysis (EFA) to assess the relationships between AI-driven personalization, customer engagement, and repurchase intention This comprehensive analytical strategy ensures that the results contribute to theoretical advancements and offer practical insights for e-commerce platforms like Shopee to enhance their AI-driven personalization techniques.
Sampling and data collection
The study utilized a non-probability sampling methodology with a purposive approach, selecting respondents with relevant knowledge Following the guidance of Hair et al (2014), a sample size of 125 to 250 was deemed necessary for a questionnaire consisting of 25 items Ultimately, 204 valid responses were included in the empirical analysis after excluding 54 responses from individuals who either did not use Shopee as an e-commerce platform or lacked familiarity with AI-driven personalization features.
The questionnaire includes 25 questions organized into two main sections The first section focuses on demographic and screening inquiries, assessing characteristics such as gender, age, monthly income, and occupation It also evaluates individual purchasing behavior on Shopee, including the frequency of purchases and monthly spending To qualify for the research, respondents must meet three criteria: they must reside in Vietnam, have made at least one purchase on Shopee, and be familiar with AI-driven personalization features like personalized product recommendations, AI chatbots, and AR-based Virtual Try-On Respondents will answer three qualifying questions with "Yes" or "No."
Respondents who answer "No" will be disqualified from the study and can terminate the poll, while those who answer "Yes" will continue to the next section All three eligibility criteria must be met; any "No" response will result in the termination of the survey The final section includes five key questions evaluated using a 5-Point Likert Scale, a widely used tool for measuring respondents' psychological attitudes towards specific statements (Preedy, V.R & Watson, R.R., 2010) This scale allows participants to express their views with options ranging from (1) Totally disagree to (5) Totally agree, reflecting their level of agreement on each item.
Ethical permission
Participants were assured of anonymity regarding their identities and email addresses before the study Participation was entirely voluntary, and no incentives or monetary compensation were offered for completing the survey.
Measures
The questionnaire was designed based on the hypotheses outlined in section 3, incorporating independent variables such as AI-driven personalization, recommendation systems, chatbots, and augmented reality Consumer involvement acts as the intermediate variable, while repurchase intention is identified as the dependent variable.
The constructs measured in this study are derived from reputable published research Various measurement scales were utilized, including a five-item AI-driven personalization Recommendation System scale by Yin et al (2025), a five-item AI-driven personalization Chatbot scale by Evelina Khonkanen (2023), a five-item AI-driven personalization Augmented Reality scale by Gujarathi et al (2024), a five-item Customer Engagement scale by Vinerean et al (2021), and a five-item Repurchase Intention scale by Kim et al.
(2012) All the questions have been adapted to align with the e-commerce scenario.
RS1 I find Shopee’s recommendations helpful in discovering new products
Xiaodong Qiu and Ya Wang, 2025)
RS2 Personalized recommendations encourage me to browse Shopee more often
RS3 I interact more frequently with Shopee’s product pages due to the recommendations provided
RS4 Personalized recommendations encourage me to click on suggested products while shopping onShopee
RS5 The recommendations make my shopping experience more enjoyable and engaging.
CB1 It is easy to find what I want by using Shopee AI Chatbots
AI Chatbot agent increases my likelihood of completing a purchase on Shopee
CB3 Shopee’s chatbot is easy to interact with and navigate to resolve my issues.
CB4 The chatbot makes me feel supported and valued as a customer
CB5 Shopee’s chatbot enhances my overall shopping experience.
AR1 Shopee’s AR features, like Virtual Try-On, enhance customer interaction and involvement in brand experiences
AR2 I find Shopee’s AR features, like Virtual Try-On applications, effective in capturing my attention and interest
AR3 Shopee’s AR features, like Virtual Try-On, are easy to use and navigate
AR4 Shopee’s AR features, like VirtualTry-On, reduce my
Constructs Code Items References uncertainty when selecting products like cosmetics or fashion items
AR5 I enjoy using the interactive AR tools to visualize products before purchasing
Shopee to experience the AI-driven
Personalized feature in the near future
(Simona Vinerean and Alin Opreana, 2021)
“Feedback” different product on Shopee
CE3 I’m very pleased to use Shopee and interact with it
CE4 It seems to me that
CE5 Time flies whenever I visit Shopee because I want to find out more
Repurschase Intention RI1 I intend to continue to purchase goods from Shopee
Robert D Galliers, Namchul Shin, Joo- Han Ryoo, Jongheon Kim, 2012)
Shopee for new products or deals
RI3 I intend to recommend the
Shopee platform to people around me
RI4 I intend to use Shopee as the priority e- commerce for future purchases
RI5 Except for any unanticipated reasons,
I intend to continue to use the Internet shopping site that I regularly use
DATA ANALYSIS AND RESULT
Data analysis method
The data analysis will be conducted using SmartPLS version 4.1.0.9, SPSS 26, and AMOS 24 Initially, the dataset will be imported into SmartPLS for descriptive statistical analysis, focusing on the reliability and validity of measurement items across variable categories This process includes organizing scale constructs, evaluating convergent and discriminant validity, and analyzing the Variance Inflation Factor (VIF) to detect potential multicollinearity issues.
SPSS 26 and AMOS 24 will be utilized to develop a Structural Equation Model (SEM) for an in-depth analysis of the relationships between AI-driven personalization, customer engagement, and repurchase intention Additionally, path analysis and Confirmatory Factor Analysis (CFA) will be conducted to validate the proposed study model and ensure the robustness of its components These combined analytical techniques will provide a comprehensive assessment of the research framework's reliability, validity, and structural integrity.
Demographic analysis
The demographic analysis reveals significant gender representation, with 76.5% of respondents being female and 23.5% male The majority of participants fall within the 18-24 age group at 57.8%, followed by 28.4% aged 28-35, while those under 18 and over 35 each account for approximately 6.9% Income distribution shows that 59.6% earn between 5-15 million VND, and 21.6% earn 15-30 million VND, with smaller percentages earning below 5 million VND (9.8%) and above 30 million VND (8.8%) Shopping behavior on Shopee indicates that 36.3% shop weekly, 31.4% shop 1-3 times monthly, and 23.5% shop more than once weekly, while only 8.8% shop less than once a month Despite frequent shopping, average monthly expenditures remain modest, with 43.1% spending 500K - 1M VND, 35.3% spending 1M - 5M VND, and 16.7% spending below 500K VND, while a small fraction spends above 5M VND monthly.
Over 30 million VND/month 18 8.8 Under 5 million VND/month 20 9.8
Frequency of 1-3 times per month 64 31.4
Less than once per month 18 8.8
More than 1 time per week 48 23.5
Reliability and validity
Evaluating the reliability and validity of items used to quantify abstract concepts is crucial The internal consistency of the data was measured using Cronbach’s alpha, composite reliability, and average variance extracted (AVE) All latent variables demonstrated a Cronbach’s reliability coefficient above the 0.70 threshold set by Nunnally, confirming adequate internal consistency While Cronbach’s alpha assumes equal item weight, composite reliability, which considers actual loading scores, offers a more accurate assessment of internal consistency For composite reliability to be deemed sufficient, it must exceed 0.7 Additionally, the AVE values for all constructs surpassed 0.50, indicating that over 50% of the variance in measurement items is explained by the constructs.
Table 4-2 shows that the Cronbach's Alpha values for the Recommendation System, Augmented Reality, Chatbot, Customer Engagement, and Repurchase Intention are all above 0.7, indicating satisfactory reliability Additionally, the composite dependability for the five items also exceeds 0.7, and the AVE values for all constructs are greater than 0.5 Consequently, all observed items were approved for the next round of exploratory factor analysis without any rejections.
Table 4- 2 Reliability and validity of the constructs
Exploratory Factor Analysis
The KMO test evaluates the suitability of data for factor analysis by assessing the sample adequacy for each variable and the overall model (Shrestha, 2021) According to Kaiser (1974), a KMO coefficient should ideally be between 0.5 and 1, with values below 0.5 indicating an invalid analysis Additionally, the P value must be less than 0.05 to meet significance criteria Table 4-3 displays the KMO coefficient and P value from Bartlett's Test of Sphericity for the study.
Table 4- 3 KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .945
Bartlett's Test of Sphericity Approx Chi-Square 3047.426 df 300
The results show that the KMO coefficient is 0.945, it fall into the range from 0.5 to1; and P value < 0.05 Satisfy both conditions according to Kaiser.
Initial Eigenvalues Extraction Sums of Squared
Rotation Sums of Squared Loadings
Extraction Method: Principal Component Analysis.
Factor extraction involves identifying the essential factors needed to effectively represent the relationships among a set of variables Various methodologies are available for determining the number of underlying components (Shrestha, 2021) This study utilizes Kaiser’s criterion, also known as the Eigenvalue Criterion, to decide how many factors to retain In factor analysis, factors with eigenvalues greater than one are considered significant As shown in Table 4-4, four factors have eigenvalues exceeding 1, thereby meeting the criterion The highest eigenvalue is 10.483 for factor 1, while factor 4 has the lowest eigenvalue of 1.229.
The analysis reveals that four factors explain 64.267% of the total variation among 25 variables, surpassing the minimum threshold of 50% The KMO value of 0.945 indicates that factor analysis is suitable for these variables The first component contributes 19.141% to the overall variance with an eigenvalue of 10.483, while the second component accounts for 16.84% with an eigenvalue of 2.63 The third and fourth components explain 15.15% and 13.135% of the variance, respectively, with eigenvalues of 1.725 and 1.229 This suggests that the final solution will consist of no more than four factors.
Confirmatory Factor Analysis (CFA)
After evaluating Cronbach's Alpha and conducting Exploratory Factor Analysis (EFA), the author utilizes AMOS 24 software for Confirmatory Factor Analysis (CFA) to evaluate the study model and hypotheses from Chapter 2 The findings, based on a sample size of 204, meet the necessary criteria for CFA, particularly the model fit indices: CMIN/df, GFI, CFI, TLI, RMSEA, and PCLOSE, which are detailed in Table 4-5 below.
Table 4- 5 Result of model fit indices CFA analysis
CMIN/df ≤ 2 Hair et al (2010) 0.956 Excellent
GFI ≥ 0.9 Hair et al (2010) 0.909 Excellent
CFI ≥ 0.9 Hair et al (2010) 1.000 Excellent
RMSEA ≤ 0.08 Hair et al (2010) 0.000 Excellent
The CMIN/DF value of 0.956 meets the CFA analysis criteria established by Hair et al (2010) The GFI and CFI indices, which should be at least 0.9 according to Hair et al (2010), are satisfied with values of 0.909 and 1.000, respectively Additionally, the RMSEA, with an acceptable threshold of 0.08, is met with a value of 0.000, confirming the criteria for CFA evaluation Furthermore, the PCLOSE index of 1.000 exceeds the recommended minimum of 0.05, in line with Hu & Bentler's guidelines.
The model fit values met the CFA evaluation criteria set by previous researchers, indicating that it is viable to assess the outcomes of the structural model for this study.
Variance Inflation Factor (VIF)
In regression analysis, the variance inflation factor (VIF) is crucial for detecting multicollinearity by evaluating the correlations between independent variables in a multiple regression model Multicollinearity can significantly impact regression results by affecting coefficient estimates and p-values A VIF value of 1 indicates no correlation, while values exceeding 1 suggest increasing levels of multicollinearity.
The analysis reveals that the variables are uncorrelated, with VIF values between 1.519 and 3.19, indicating moderate correlation and confirming that the model is not impacted by multicollinearity A VIF value below 5 suggests that each survey item represents a unique concept, and there is no substantial multicollinearity that requires correction.
Discriminant Validity
The heterotrait-monotrait ratio (HTMT) is a crucial statistical measure used in structural equation modeling (SEM) and psychometrics to evaluate discriminant validity, ensuring that components within a measurement model are distinct For effective discriminant validity, HTMT coefficients should be significantly lower than those for convergent validity, with a recommended threshold set at 0.9 (Hamid et al., 2017).
Table 4-7 indicates that all HTMT values for the variables are below the 0.9 threshold, confirming strong discriminant validity among the constructs Specifically, the HTMT value between customer behavior (CB) and advertising response (AR) is 0.749, while the values between customer engagement (CE) and AR, and CB are 0.741 and 0.803, respectively Additionally, the HTMT values for relationship intention (RI) in relation to AR, CB, and CE are 0.414, 0.405, and 0.481, respectively.
CE, and RI are 0.579, 0.632, 0.756, and 0.230, respectively.
AR CB CE RI RS
Structural Equation Modeling (SEM) and Hypothesis Testing
The results obtained in testing the proposed structural model are presented in Figure 4-1 below:
Figure 4-1 Structural Equation Modelling (SEM) test results for proposed research model (standardized).
Considering the direct effect, compared to the threshold p-value of 0.05, the p-values for
The analysis reveals that the impact associations of Recommendation Systems (RS) and Customer Engagement (CE) on Repurchase Intention (RI) are statistically significant, with p-values less than 0.05 In contrast, the associations of Customer Behavior (CB) and Advertising Response (AR) with RI are not statistically significant, as their p-values exceed 0.05 Notably, the relationship between RS and RI was found to be negative, contradicting Hypothesis H2, which posited that AI-driven Personalized Recommendation Systems positively influence Repurchase Intention in Shopee Vietnam Consequently, Hypothesis H2 is rejected.
The empirical findings support Hypothesis H7, demonstrating that Customer Engagement has a significant impact on Repurchase Intention Conversely, the data does not support Hypotheses H2, H4, and H6, leading to their rejection.
Table 4- 8 Regression Weights Direct Effect
Considering the indirect effect, compared to the threshold p-value of 0.05, all the p- value are smaller than 0.05, indicating that all the impact associations are statistically significant.
Table 4- 9 Regression Weights Indiirect Effect
In conclusion, H1, H3, H5, H7 are accepted H2, H4, H6 are not accepted.
The Standardized Regression Weights Table 4-10 illustrates the influence of independent variables, such as Recommendation Systems (RS), Chatbots (CB), and Augmented Reality (AR), on Repurchase Intention (RI), both directly and through the mediating variable, Customer Engagement (CE) The magnitude of the standardized regression coefficients indicates the strength of these relationships, with larger absolute values signifying more significant effects.
The findings demonstrate that Chatbot credibility (β = 0.402) exerts the most significant impact on Customer Engagement, succeeded by Recommendation Systems (β = 0.376) and Augmented Reality (β = 0.220).
Customer Engagement has the most significant positive effect on Repurchase Intention (β = 0.577), indicating that higher engagement leads to a greater likelihood of repeat purchases In contrast, Recommendation Systems show a negative correlation with repurchase intention (β = -0.321), suggesting that while they enhance engagement, they do not necessarily lead to immediate repeat purchases Augmented Reality has a positive but weaker effect on repurchase intention (β = 0.171), while Chatbots have an insignificant direct impact (β = 0.011).
About Squared Multiple Correlation Table 4-11 used to represent R square values of Customer Engagement and Repurchase Intention.
The variable Customer Engagement has an R square value of 0.772, indicating that the variables RS, CB, and AR account for 77.2% of its variance In contrast, the R square value for the observed variable RI is 0.276, with Customer Engagement explaining 27.6% of its variance.
CONCLUSION
Main findings
This study explores the impact of AI-driven customization on consumer engagement and repurchase intentions in the e-commerce sector, focusing on Shopee Vietnam Utilizing the Stimulus-Organism-Response (S-O-R) framework, it examines how key AI personalization features—Recommendation Systems, Chatbots, and Augmented Reality (AR)—influence customer engagement and repurchase behavior The findings from structural equation modeling (SEM) reveal that AI personalization significantly boosts consumer interactions on Shopee, enhances engagement, and increases the likelihood of repeat purchases This analysis provides valuable insights into the direct and indirect relationships among these variables, offering empirical evidence of the effectiveness of personalization in shaping consumer behavior in e-commerce.
This study reveals that Customer Engagement significantly mediates the relationship between AI-driven personalization features and Repurchase Intention The structural equation modeling results show that all three AI-driven features—Recommendation Systems (RS), Chatbots (CB), and Augmented Reality (AR)—positively influence Customer Engagement, with standardized coefficients of 0.376, 0.402, and 0.220, respectively Notably, Chatbots are identified as the strongest predictor of engagement, highlighting their effectiveness in providing immediate, interactive, and personalized support that enhances the overall customer experience.
CE demonstrated a strong and statistically significant positive effect on Repurchase Intention (β = 0.577, p = 0.010), supporting Hypothesis H7 However, the direct effects of AI personalization features on Repurchase Intention produced inconsistent outcomes.
Recommendation Systems (RS) demonstrated a significant negative impact on Repurchase Intention (β = -0.321, p = 0.019), leading to the rejection of Hypothesis H2, which anticipated a positive correlation This outcome indicates that while RS can boost engagement, it may not effectively drive repeat purchases Potential reasons for this include customer fatigue from an overload of irrelevant recommendations and a perceived lack of authenticity in the algorithms used for suggestions.
Chatbots (CB) demonstrated a very weak and statistically insignificant direct impact on
The analysis revealed that the relationship index (RI) was β = 0.011 with a p-value of 0.945, leading to the rejection of Hypothesis H4 Although chatbots significantly impact user engagement, they do not necessarily drive purchase decisions, particularly when their interactions are confined to generic responses or technical support.
Augmented Reality (AR) showed a positive but statistically insignificant effect on RI (β
= 0.171, p = 0.216), leading to the rejection of Hypothesis H6 This indicates that while
AR features like virtual try-ons can enhance user experience, but they are not enough to ensure repeat purchases without factors like pricing, trust, and product quality Analysis shows that AI-driven personalization features significantly boost repurchase intention through customer involvement, with p-values under 0.05 for all pathways Chatbots have the strongest impact on customer engagement (β = 0.402), followed by recommendation systems (β = 0.376) and augmented reality (β = 0.220), highlighting chatbots' effectiveness due to their real-time assistance and interactive communication capabilities.
These findings collectively point to the critical insight that AI-driven personalization features may be more effective as engagement tools rather than direct drivers of transactional outcomes.
Although RS, CB, and AR have weak or negative direct effects on RI, their indirect effects through Customer Engagement are statistically significant (p < 0.05) This highlights the critical role of Customer Engagement as a psychological and behavioral mechanism that transforms AI-driven personalization into consumer loyalty and repeat purchases Without effective engagement, the technologies may fail to achieve the intended business results.
The R-squared values validate the model's explanatory capacity, with Customer Engagement (CE) showing an R² value of 0.772, indicating that AI-driven personalization features explain 77.2% of its variance In contrast, Repurchase Intention (RI) has an R² score of 0.276, meaning that 27.6% of its variation is explained by Customer Engagement This highlights the crucial role of customer engagement as a mediating factor, suggesting that AI-driven personalization enhances repurchase intention primarily by fostering stronger consumer interactions and relationships, rather than directly affecting repeat purchases However, this moderate explanatory power also suggests that other factors, such as pricing, brand trust, product satisfaction, and external competitive pressures, may influence repurchase decisions.
Recommendation Systems positively influence Customer Engagement in Shopee Vietnam.
Recommendation Systems positively influence Repurchase Intention in Shopee Vietnam.
H3 AI-powered Chatbots positively influence Customer Engagement in Shopee Vietnam.
H4 : AI-powered Chatbots positively influence Repurchase Intention in Shopee Vietnam.
H5 AI-driven Augmented Reality positively influences Customer Engagement in Shopee Vietnam.
H6 AI-driven Augmented Reality positively influences Repurchase Intention in Shopee Vietnam.
H7 Customer Engagement positively affects Repurchase Intention in Shopee Vietnam
Managerial Implications
This study provides valuable insights for e-commerce platforms like Shopee Vietnam and other businesses looking to enhance consumer engagement through AI-driven personalization It emphasizes the importance of aligning AI personalization with consumer preferences and expectations while tackling challenges related to privacy, trust, and usability Given the competitive landscape of Vietnam's e-commerce sector, an active strategy for AI innovation is essential for success.
AI-powered chatbots significantly enhance customer engagement by providing real-time assistance, personalized recommendations, and seamless navigation during the shopping journey To improve user experiences, businesses should invest in advanced Natural Language Processing (NLP) models that enhance chatbot accuracy and contextual understanding In diverse markets like Vietnam, improving multilingual capabilities and cultural adaptability is crucial due to varying language preferences and shopping behaviors Additionally, implementing hybrid chatbot models, where AI handles routine inquiries and complex issues are escalated to human agents, can enhance service quality and reduce customer frustration.
Research highlights the significant impact of AI-driven recommendation systems on customer engagement, emphasizing the importance of personalized shopping experiences in e-commerce These systems help reduce information overload by guiding consumers to relevant products and encouraging repeat interactions To enhance recommendation algorithms, businesses should adopt deep learning techniques for hyper-personalization, combining collaborative filtering with content-based filtering As personalization evolves, companies must address consumer concerns about data privacy and algorithmic transparency Empowering users with control over their personalization settings and clearly explaining the rationale behind AI-generated recommendations can foster trust Additionally, incorporating predictive analytics for supply chain management and inventory optimization can improve operational efficiency by aligning product availability with anticipated consumer demand.
Limitation of Research
This study's findings are accompanied by several limitations that may have impacted the results Acknowledging these limitations is crucial as they offer valuable insights for future research aimed at refining and expanding the study's conclusions.
This study's limitation lies in its exclusive focus on Shopee Vietnam, which restricts the applicability of its findings to the broader e-commerce industry Although Shopee is the leading platform in Vietnam, the results may not be applicable to other e-commerce platforms in the country or the Southeast Asian region Competing platforms like Lazada, TikTok Shop, and Tiki may utilize different AI strategies, personalization techniques, and user engagement models Additionally, Shopee's AI-driven personalization features are specifically designed for its ecosystem, making it difficult to generalize these findings to other e-commerce models that implement distinct AI approaches.
This study highlights three essential AI-driven personalization features in e-commerce: Recommendation Systems, Chatbots, and Augmented Reality (AR) However, it is important to note that AI personalization encompasses a broader spectrum, including tools like voice commerce, AI-powered dynamic pricing, sentiment analysis, automated customer segmentation, and real-time personalization based on user behavior By not including these additional tools, the study's scope is limited, failing to represent the full landscape of AI-driven personalization in e-commerce.
Recommendation for future research
In light of the aforementioned limitations, the following recommendations for future studies are proposed for consideration and implementation:
First of all, future studies should employ a comparative approach by analyzing various e-commerce platforms to assess the consistency of AI-driven personalization effects across diverse digital marketplaces.
Future research should explore a wider range of AI applications to better understand how different AI-driven personalization tools interact and influence consumer behavior and engagement Additionally, it is essential to examine the potential synergies and trade-offs among these AI tools to determine the most effective personalization strategies for e-commerce platforms.
Future research should focus on the ethical challenges of AI-driven personalization, particularly regarding data privacy, algorithmic bias, consumer autonomy, and regulatory compliance As AI personalization advances, concerns about data security, informed consent, and the misuse of personal information become increasingly significant Understanding consumer perceptions of AI personalization in relation to privacy and trust is crucial for fostering transparent and responsible AI practices in e-commerce Additionally, exploring the effects of transparency in AI decision-making, ethical data collection, and regulatory frameworks on consumer attitudes is essential Investigating consumer trust as a moderating factor in AI engagement and repurchase behavior could provide valuable insights into how ethical AI implementation influences customer loyalty.
Conclusion
This research investigates the impact of AI-driven personalization on consumer engagement and repurchase intention within Shopee Vietnam, focusing on three key features: Recommendation Systems, Chatbots, and Augmented Reality (AR) Utilizing the Stimulus-Organism-Response (S-O-R) framework, the study identifies AI-driven personalization as the stimulus, customer engagement as the organism, and repurchase intention as the response Four hypotheses were formulated to empirically validate the non-probability sampling techniques Data was collected via an online survey using a 5-point Likert scale, resulting in 204 valid responses, with analysis performed using SmartPLS version 4.1.0.9, SPSS version 26, and AMOS version 24.
The demographic analysis revealed key insights into respondents' gender, age, income, purchase frequency, and average spending habits To ensure internal consistency among the 25 measurement items, Cronbach’s Alpha reliability testing was conducted, focusing on five main observed variables: AI-driven recommendation systems, chatbots, augmented reality, customer engagement, and repurchase intention.
AI-driven personalization significantly enhances customer engagement, positively influencing repurchase intentions Among three analyzed features, chatbots had the strongest impact on engagement (β = 0.732, p < 0.001), followed by recommendation systems (β = 0.505, p < 0.001) and augmented reality (AR) (β = 0.458, p < 0.001) Chatbots facilitate real-time interactions and personalized assistance, making them the most effective tool for boosting engagement Recommendation systems enhance product discovery and user experiences, while AR features, though less impactful, help reduce purchase uncertainty and create immersive shopping experiences Additionally, strong customer engagement correlates with a significant increase in repurchase intention (β = 1.000, p < 0.001), indicating that engaged consumers are more likely to make repeat purchases on e-commerce platforms.
This study provides valuable insights for e-commerce businesses, highlighting the importance of investing in AI-powered chatbots and improving recommendation algorithms to boost engagement While augmented reality (AR) holds great potential, enhancing its accessibility and expanding its applications is crucial for maximizing effectiveness Moreover, AI-driven personalization relies heavily on consumer data, necessitating the implementation of ethical AI practices that emphasize transparency, data protection, and responsible use to build consumer trust and ensure long-term loyalty.
The study acknowledges limitations such as the sample composition, the focus on Shopee as a single case study, and the analysis of only three AI-driven personalization features Future research should expand to include multiple e-commerce platforms, a broader range of AI technologies, and a more diverse consumer sample to improve generalizability Additionally, as AI-driven personalization evolves, it is crucial to explore the ethical and privacy concerns associated with AI implementation and conduct longitudinal studies to assess changes in consumer engagement with AI technologies over time.
This study highlights the crucial role of AI-driven personalization in e-commerce, demonstrating that AI technologies significantly enhance customer engagement and boost repurchase intentions For businesses seeking to elevate user experiences and foster brand loyalty, strategic implementation of AI is essential to maintain a competitive edge in the digital marketplace By refining personalization strategies and considering ethical implications, e-commerce platforms can strengthen consumer relationships, improve business performance, and achieve sustainable growth in an increasingly AI-focused economy.
1 Adam, M., Wessel, M., & Benlian, A (2021) AI-based chatbots in customer service and their effects on user compliance.Electronic Markets,31, 427–445. https://doi.org/10.1007/s12525-020-00414-7
2 Adawiyah, S R., Purwandari, B., Eitiveni, I., & Purwaningsih, E H (2024) The Influence of AI and AR Technology in Personalized Recommendations on Customer Usage Intention: A Case Study of Cosmetic Products on Shopee.
Applied Sciences,14, 5786 https://doi.org/10.3390/app14135786
3 Asante, I O., Jiang, Y., Hossin, A M., & Luo, X (2023) OPTIMIZATION OF CONSUMER ENGAGEMENT WITH ARTIFICIAL INTELLIGENCE
ELEMENTS ON ELECTRONIC COMMERCE PLATFORMS.Journal of Electronic Commerce Research,24(1).
4 Babatunde, S O., Odejide, O A., Edunjobi, T E., & Ogundipe, D O (2024). THE ROLE OF AI IN MARKETING PERSONALIZATION: A
THEORETICAL EXPLORATION OF CONSUMER ENGAGEMENT
STRATEGIES.International Journal of Management & Entrepreneurship Research,6(3), [936-949] https://doi.org/10.51594/ijmer.v6i3.964
5 Bag, S., Srivastava, G., Bashir, M M A., Kumari, S., Giannakis, M., &
Chowdhury, A H (2022) Journey of customers in this digital era:
Understanding the role of artificial intelligence technologies in user engagement and conversion.Benchmarking: An International Journal,29(7), 2074-2098. https://doi.org/10.1108/BIJ-07-2021-0415
6 Balasubramanian, G (2024) THE ROLE OF AI IN ENHANCING
PERSONALIZATION IN ECOMMERCE: A STUDY ON CUSTOMER
ENGAGEMENT AND SATISFACTION.Asian And Pacific Economic Review,17(2), 160-177 DOI10.6084/m9.figshare.26198327
7 Bawack, R E., Wamba, S F., Carillo, K D A., & Akter, S (2022) Artificial intelligence in E‑Commerce: a bibliometric study and literature review.Electron Markets,32, [297–338] https://doi.org/10.1007/s12525-022-00537-z
8 Bilal, M., Zhang, Y., Cai, S., Akram, U., & Halibas, A (2024) Artificial intelligence is the magic wand making customer-centric a reality! An investigation into the relationship between consumer purchase intention and consumer engagement through affective attachment.Journal of Retailing and Consumer Services,77(1), 103674. https://doi.org/10.1016/j.jretconser.2023.103674
9 Bowden, J L.-H (2009) The process of customer engagement: a conceptual framework.Journal of Marketing Theory and Practice,17(1), 63–74. https://doi.org/10.2753/MTP1069-6679170105
10 Cakraputr, D., Sunitiyoso, Y., & Suhaimi, H (2024) The Role of Stimulus- Organism-Response (SOR) in E-Commerce Strategic Decision Making for
Sustainability.Himalayan Journal of Economics and Business Management,2(1).
11 Chen, Y., Prentice, C., Weaven, S., & Hisao, A (2022) The influence of customer trust and artificial intelligence on customer engagement and loyalty - The case of the home-sharing industry.Front Psychol,13. https://doi.org/10.3389/fpsyg.2022.912339
12 Chou, S.-W., & Hsu, C.-S (2016) Understanding online repurchase intention: social exchange theory and shopping habit.Information Systems and e-Business Management,14(1), 19-45 DOI: 10.1007/s10257-015-0272-9
13 Chung, M., Ko, E., Joung, H., & Kim, S J (2020) Chatbot e-service and customer satisfaction regarding luxury brands.Journal of Business Research,
14 Chung, S (2024, July 16).Shopee leads SEA booming E-commerce scene.
Vulcan Post Retrieved November 19, 2024, from https://vulcanpost.com/865393/shopee-reigns-southeast-asia-e-commerce-tiktok- shop-surpasses-lazada/
Implementation of Augmented Reality Shopping in E-Commerce to Increase Customer's Purchase Intention.Engineering MAthematics and Computer Science (EMACS) Journal,5(2), 97-102 DOI:10.21512/emacsjournal.v5i2.9954
16 Enyejo, J O., Obani, , O Q., Afolabi, O., Igba, E., & Ibokette, A I (2024). Effect of Augmented Reality (AR) and Virtual Reality (VR) experiences on customer engagement and purchase behavior in retail stores.Magna Scientia Advanced Research and Reviews,11(2), 132–150 DOI: https://doi.org/10.30574/msarr.2024.11.2.0116
17 Etha, E., Suciati, P., & Citra, A N (2024) FACTORS INFLUENCING
ONLINE REPURCHASE INTENTION IN INDONESIA’S E-COMMERCE
M ’S MARKET: A STRUCTURAL EQUATION MODELING APPROACH.
Jurnal Sosial Humaniora Terapan,6(2) : https://scholarhub.ui.ac.id/jsht/vol6/iss2/8
18 Gayam, S R (2021) Artificial Intelligence in E-Commerce: Advanced
Techniques for Personalized Recommendations, Customer Segmentation, and Dynamic Pricing.Journal of Bioinformatics and Artificial Intelligence,1(1),
105-150 https://biotechjournal.org/index.php/jbai/article/view/89
19 Gentsch, P (2018).AI in Marketing, Sales and Service: How Marketers Without a Data Science Degree Can Use AI, Big Data and Bots Springer International
20 Gujarathi, S., Mane, S T., & Gorkhe, M D (2024) ROLE OF AUGMENTEDREALITY AND VIRTUAL REALITY (AR/VR) IN CUSTOMER
ENGAGEMENT.ShodhKosh: Journal of Visual and Performing Arts,5(6),
558–573 DOI: https://doi.org/10.29121/shodhkosh.v5.i6.2024.1800
21 Gupta, A., Kumar, A., & Melese, E (2023) Young consumer engagement at the bottom of pyramid: applying S-O-R framework in e-commerce context.Young Consumers,24(6), pp 786-806 https://doi.org/10.1108/YC-06-2022-1542
22 Hallikainen, H., Luongo, M., Dhir, A., & Laukkanen, T (2022) Consequences of personalized product recommendations and price promotions in online grocery shopping.Journal of Retailing and Consumer Services,69. https://doi.org/10.1016/j.jretconser.2022.103088
23 Harris, L C., & Goode, M M.H (2004) The Four Levels of Loyalty and the Pivotal Role of Trust: A Study of Online Service Dynamics.Journal of Retailing, 80(2), 139-158 DOI:10.1016/j.jretai.2004.04.002
24 Hellier, P K., Geursen, G M., Rickard, J A., & Carr, R (2003) Customer repurchase intention: A general structural equation model.European Journal of Marketing,37(11), 1762-1800 DOI:10.1108/03090560310495456
Recommendation Systems For E-commerce Systems An Overview.Journal of Physics: Conference Series,1897 doi:10.1088/1742-6596/1897/1/012024
26 Jebarajakirthy, C., Saha, V., Goyal, P., & Mani, V (n.d.) How Do Value Co- Creation and E-Engagement Enhance E-Commerce Consumer Repurchase
Intention? An Empirical Analysis.Journal of Global Information Management, 30(5) DOI: 10.4018/JGIM.290369
27 Joshi, S., Vashistha, K., Baliyan, A K., & Swati (2024) Revolutionising E- commerce: Personalised Recommendations for Enhancing User Engagement.http://dx.doi.org/10.213
28 Khonkanen, E (2023) The Role of Chatbot Marketing Efforts in Enhancing CustomerBrand Relationships for the Finnish Fashion Brands.Thesis - Bachelor of Business Administration - Haaga-Helia University of Applied Sciences. https://www.theseus.fi/bitstream/handle/10024/801495/Khonkanen_Evelina.pdf? sequence=2&isAllowed=y
29 Kim, C., Galliers, R D., Shin, N., Ryoo, J.-H., & Kim, J (2012) Factors influencing Internet shopping value and customer repurchase intention.
Electronic Commerce Research and Applications,11, 374–387. http://dx.doi.org/10.1016/j.elerap.2012.04.002
30 Kumar, V., Rajan, B., Venkatesan, R., & Lecinski, J (2019) Understanding the Role of Artificial Intelligence in Personalized Engagement Marketing.California Management Review,61(4) https://doi.org/10.1177/0008125619859317
31 Liu, C., & Zhang, Z (2024) Can Artificial Intelligence (AI)-driven personalization influence customer experiences? - A quantitative study on
TikTok integration with artificial intelligence.Master’s Thesis - Department of Business Studies - Uppsala University.
32 Lopes, J M., Silva, L F., & Massano-Cardoso, I (2024) AI Meets the Shopper: Psychosocial Factors in Ease of Use and Their Effect on E-Commerce Purchase Intention.Behav Sci.,14, 616 https://doi.org/10.3390/bs14070616
33 Martiyanti, D (2023) Does Brand Love Make Every Consumer Behave On Social Media ? The Role of Personality Traits and Brand Love On Customer Engagement Behavior.AMAR (Andalas Management Review),6(1), 14-38.
34 Maskuroh, N., Fahlevi, M., Irma, D., Rita, & Rabiah, A S (2022) Social media as a bridge to e-commerce adoption in Indonesia: A research framework for repurchase intention.International Journal of Data and Network Science,6,
35 Maslej, N., Fattorini, L., Perrault, R., Parli, V., Reuel, A., Brynjolfsson, E., Etchemendy, J., Ligett, K., Lyons, T., Manyika, J., Niebles, J C., Shoham, Y., Wald, R., & Clark, J (2024) Artificial Intelligence Index Report 2024. https://doi.org/10.48550/arXiv.2405.19522
36 Mehrabian, A., & Russell, J A (1980).An Approach to Environmental
37 Mohamed, A (2024) Assessing AI-Powered Personalization Strategies for Mid- Sized Finnish Fashion E-commerce: Enhancing Consumer Engagement and Conversion Rates.Digital Business Innovations Research-based thesis,Haaga- Helia University of Applied Sciences. https://www.theseus.fi/bitstream/handle/10024/858274/Mohamed_Ahmed.pdf?se quence=2&isAllowed=y
COMMERCE WITH AI CHATBOTS: ENHANCING CUSTOMER
SATISFACTION AND PURCHASE DECISIONS IN ONLINE
MARKETPLACE.Journal of Theoretical and Applied Information Technology, 102(19), 7224-7237. https://www.jatit.org/volumes/Vol102No19/29Vol102No19.pdf
39 Patterson, P G., & Yu, T (2006) Understanding Customer Engagement in Services.
40 Perez-Vega, R., Kaartemo, V., R Lages, C., Razavi, N B., & Mọnnistử, J.
(2021) Reshaping the contexts of online customer engagement behavior via artificial intelligence: A conceptual framework.Journal of Business Research,
41 Press, G (2017, August 27).Artificial Intelligence (AI) Defined Forbes.
Retrieved November 20, 2024, from https://www.forbes.com/sites/gilpress/2017/08/27/artificial-intelligence-ai- defined/
42 Putra, J E., & Prabowo, B (2025) The Effect of Utilizing AI Chatbot and
Recommendation System on Customer Satisfaction and Retention in Local
Marketplace in Bandung.West Science Social and Humanities Studies,3(1), 20-
43 Rachmiati, P R (2024) Discover Your Perfect Look: Virtual Makeup Try-On Revolutionizes Shopping Experience on Shopee!Asian Journal of Engineering, Social and Health,3(6), 1344-1358 https://ajesh.ph/index.php/gp
44 Rahmadini, D D., & Masnita, Y (2024) Consumer Response to Live Streaming Shopping using the Stimulus, Organism and Response (SOR) Method Approach.
Jurnal Ilmiah Manajemen Kesatuan,12(3), 691–700. https://jurnal.ibik.ac.id/index.php/jimkes/article/view/2570
45 Rahman, S S., & Dekkati, S (2022) Revolutionizing Commerce: The Dynamics and Future of E-Commerce Web Applications.Asian Journal of Applied Science and Engineering,11(1), 65–73 DOI:10.18034/ajase.v11i1.58
46 Raji, M A., Olodo, H B., Oke, T T., Addy, W A., Ofodile, O C., & Oyewole,
A T (2024) E-commerce and consumer behavior: A review of AI-powered personalization and market trends.GSC Advanced Research and Reviews,18(03),
47 R.C, A., & Dulloo, R (2024) The Impact of AI-Driven Personalization on
Customer Satisfaction in E-Commerce: Balancing Technology, Transparency, and Control.Journal of Computational Analysis and Applications,33(2).
48 Shahikanth, P., Naidu, B V V., Teja, K S., Chowdary, K V., & Shah, N.
(2024) ENHANCING USER EXPERIENCE IN E-COMMERCE THROUGHAUGMENTED REALITY AND VIRTUAL REALITY.International Research
Journal of Modernization in Engineering Technology and Science,6(3), 4976-
4982 DOI : https://www.doi.org/10.56726/IRJMETS51444
49 Shopee (2024).What is Shop AI Assistant?Shopee Seller Education Hub. https://seller.shopee.ph/edu/article/22309
50 Shrestha, N (2021) Factor Analysis as a Tool for Survey Analysis.American Journal of Applied Mathematics and Statistics,9(1), 4-11 DOI:10.12691/ajams-
51 Song, X., Yang, S., Huang, Z., & Huang, T (2019) The Application of Artificial Intelligence in Electronic Commerce.Journal of Physics: Conference Series, Series 1302 doi:10.1088/1742-6596/1302/3/032030
IN E-COMMERCE: A STUDY OF GEN Z CUSTOMER EXPERIENCE AND ENGAGEMENT – FRIEND OR FOE?Jurnal EMBA : Jurnal Riset Ekonomi, Manajemen, Bisnis dan Akuntansi,11(3), 161-175 DOI: https://doi.org/10.35794/emba.v11i3.49501
53 Vietdata (2024, 1 3).Shopee tiếp tục giữ vững vị thế dẫn đầu trong lĩnh vực Thương mại điện tử tại Việt Nam Vietdata. https://www.vietdata.vn/vi/post/shopee-ti%E1%BA%BFp-t%E1%BB%A5c- gi%E1%BB%AF-v%E1%BB%AFng-v%E1%BB%8B-th%E1%BA%BF- d%E1%BA%ABn-%C4%91%E1%BA%A7u-trong-l%C4%A9nh- v%E1%BB%B1c-th%C6%B0%C6%A1ng-m%E1%BA%A1i-
%C4%91i%E1%BB%87n-t%E1%BB%AD-t%E1%BA%A1i-vi%E1%BB%87t- nam?
54 Vinerean, S., & Opreana, A (2021) Measuring Customer Engagement in Social Media Marketing: A Higher-Order Model.Journal of Theoretical and Applied Electronic Commerce Research,16(7), 2633-2654. https://doi.org/10.3390/jtaer16070145
55 Yadav, K (2024).How AI Is Changing Ecommerce Personalization Firework. https://firework.com/blog/how-ai-is-changing-ecommerce- personalization#:~:text=AI%2Ddriven%20dynamic%20pricing%20adjusts,on%2 0real%2Dtime%20market%20conditions.
56 YEE, S H (2022) USAGE OF ARTIFICIAL INTELLIGENCE IN ONLINE SHOPPING APPLICATION: MALAYSIAN CUSTOMER ATTITUDE.A final year project submitted in partial fulfilment of the requirement for the degree of, BACHELOR OF INTERNATIONAL BUSINESS(UNIVERSITI TUNKU ABDUL
RAHMAN) http://eprints.utar.edu.my/4857/1/23._Soo_Hui_Yee_1801928.pdf
57 Yin, J., Qiu, X., & Wang, Y (n.d.) The Impact of AI-Personalized
Recommendations on Clicking Intentions: Evidence from Chinese E-Commerce.
Journal of Theoretical and Applied Electronic Commerce Research,20(1), 21. https://doi.org/10.3390/jtaer20010021
● Other/Prefer not to say
2 What is your age group?
3 What is your education level?
4 How much is your monthly income?
5 Do you use Shopee as an online shopping application?
6 Frequency of using Shopee for shopping
Less than once a month 1–3 times per month Weekly
7 How much do you spend on Shopee monthly?
8 Do you enjoy with Artificial Intelligence driven personalization (e.g., recommendation system, chatbots, augmented reality) while using Shopee?
Section 2: AI-Driven Personalized Recommendation System - CustomerEngagement
1 I find Shopee’s recommendations helpful in discovering new products
2 Personalized recommendations encourage me to browse Shopee more often
3 I interact more frequently with Shopee’s product pages due to the recommendations provided
4 Personalized recommendations encourage me to click on suggested products while shopping on
5 The recommendations make my shopping experience more enjoyable and engaging.
Section 3:AI-Driven Personalized Chatbot - Customer Engagement
1 It is easy to find what I want by using Shopee AI
Chatbot agent increases my likelihood of completing a purchase on Shopee
3 Shopee’s chatbot is easy to interact with and navigate to resolve my issues.
4 The chatbot makes me feel supported and valued as a customer
5 Shopee’s chatbot enhances my overall shopping experience.
Section 4: AI-Driven Personalized Augmented Reality - Customer Engagement
Virtual Try-On, enhance customer interaction and involvement in brand experiences
2 I find Shopee’s AR features, like Virtual Try-On applications, effective in capturing my attention and interest
Virtual Try-On, are easy to use and navigate
Virtual Try-On, reduce my uncertainty when selecting products like cosmetics or fashion items
AR tools to visualize products before purchasing
1 I will likely revisit Shopee to experience the AI-driven
Personalized feature in the near future
2 I have “Liked”, “Saved” and/or “Feedback” different product on Shopee
Shopee and interact with it
4 It seems to me that Shopee are very useful
Shopee because I want to find out more