Do smart hotel technologies enhance guest experience? Evidence from online review analytics in the UAE Hospitality Industry

Introduction

Mobile apps, contactless services, online concierge services, smart-room services, and automation services are gradually forming the landscape of the hospitality industry. In destinations like the United Arab Emirates, where the needs of premium tourism touch the digital infrastructure and cross-border mobility of guests, hotel managers are likely to offer high-touch service with high-quality digital convenience. Online reviews put these experiences in the mouths of guests and are, therefore, one of the important secondary sources of data for gaining insight into technology-enabled satisfaction (Bansal, 2018; Li et al., 2015; Argilla, 2023; Lee et al., 2014; Filieri, 2015; Filieri et al., 2015; Blei et al., 2003; Devlin et al., 2019; Liu, 2020). The issue reviewed in this study is that many hotel companies invest in smart technologies without adequately considering guest experiences with postencounter service technology. The statistics on technology adoption indicate whether guest rooms are equipped with systems, but do not indicate whether guests find these systems convenient, reliable, or frustrating. Review text is useful to bridge this gap because it involves both numerical ratings and natural-language definitions of the quality of services (Filieri, 2015; Blei et al., 2003; Devlin et al., 2019; Koo et al., 2015). The United Arab Emirates (UAE) is an adequate setting because the hospitality industry in this country enjoys a globally observable presence, great expectations through digital services, and robust smart policies on the tourism environment. Visitors to Dubai, Abu Dhabi, and Sharjah tend to contrast domestic hotel services with international service standards. Thus, a smart check-in kiosk, a mobile key, a room tablet, or a digital concierge can only make a difference in the service journey in the context of smart support through the responsiveness of the staff (Law et al., 2014; Sigala, 2016; Li et al., 2022; Kar, 2021; Neuhofer et al., 2014, 2015; Parasuraman et al., 1988).

The usage trend of smart technologies is increasing, and existing research has basically focused on a survey-based approach to determine technology acceptance, service quality, or behavioral intentions. Few studies have examined the application of online reviews to identify how guests describe and evaluate technology-enabled service experiences in real-world hospitality settings, especially in the UAE. Moreover, very little attention has been paid to integrating smart technology themes, review sentiment, and experience quality into a transparent secondary-data analytical framework. This study addresses this gap by using open-access review datasets and reproducible text analytics to examine the prevalence of smart hotel technologies in guest experiences and customer satisfaction in the hospitality industry in the United Arab Emirates.

Literature review and conceptual framework

The technologies used in smart hotels are typically considered in the context of technology acceptance, service quality, and customer experience. Technology acceptance theory focuses on usefulness and ease of use, while service-quality theory focuses on reliability, responsiveness, assurance, and empathy. These views overlap in hotel contexts, as a guest does not consider a digital tool in its own right but considers whether the tool facilitates making the stay faster, easier, more comfortable, or customized (Oliver, 1980; Agresti, 2018; Berezina et al., 2016; Park et al., 2007; Morosan and DeFranco, 2016a, 2016b). Online reviews provide an additional behavioral and perceptual layer to this literature. Compared with surveys, review data are naturally occurring, large, and emotional in nature. Platform bias, a lack of reviewer motivation, and demographic information are also present. Therefore, this study considers online reviews as supporting pattern discovery through secondary observational evidence as opposed to controlled causal evidence (Schuckert et al., 2015, Vermeulen and Seegers, 2009, Xiang et al., 2015, Ye et al., 2009). The conceptual model describes a connection between the attributes of smart hotel technology, guest experience, and customer satisfaction. The state-of-the-art technology layer consists of mobile check-in/check-out, keyless entry, contactless payment, smart-room controls, digital concierge, customized applications, room tablets, and automated service kiosks. The experience layer comprises convenience, dependability, customization, service recovery, and tone. The satisfaction layer is represented by star ratings and review sentiment (Sigala, 2016; Li et al., 2022; Kar, 2021; Neuhofer et al., 2014, 2015; Parasuraman et al., 1988; Venkatesh et al., 2003). Latest studies have emphasized the strategic benefits of smart technologies in the tourism and hospitality sectors. Digital transformation has become a main driver of industry resilience, competitiveness, and customer value creation in the postpandemic era (Dikhanbayeva, 2025). Technology implementation and innovation strategies have benefited organizational performance and long-term competitive advantage in the hospitality industry (Demir and Demir, 2025). In addition, mobile information technologies provide service innovation by enhancing customer engagement and improving service quality through digital interaction (Demir, 2025). In today’s world, artificial intelligence tools such as ChatGPT have helped improve customer experiences in hospitality, although human interaction is still required (Yaşar and Yayla, 2025). Collectively, these studies reinforce the importance of integrating smart technologies with human-centered service delivery to enhance guest satisfaction and organizational performance. The conceptual model assumes that smart hotel technologies influence customer satisfaction indirectly through guests’ perceived experience and expressed sentiment, providing a theoretical basis for the proposed analytical framework.

Objectives, research questions, and hypotheses

• Objective 1: To discover the key smart hotel technology dimensions that manifest in customer online reviews.
• Objective 2: To categorize the sentiments of guests toward hotel experiences with technology (Figure 1).
• Objective 3: To investigate the relationship between sentiments, smart-technology intensity, quality of experiences, and customer satisfaction.
• Objective 4: To offer managerial suggestions for enhancing technology-based guest experiences at UAE hotels.
• RQ1: What smart technology characteristics are most commonly discussed in internet hotel reviews?
• RQ2: What are the perceptions of UAE hotel guests regarding smart hotel experiences?
• RQ3: To what are the technology-associated factors and customer satisfaction ratings linked?
• H1: Hotel ratings are positively related to higher positive review sentiment.
• H2: There is a positive relationship between smart-technology intensity and hotel ratings.
• H3: Experience quality has been found to have a positive relationship with hotel ratings and accounts for trade as part of the pragmatic relationship between technology-enabled services and satisfaction.

Methodology

Possible research design: Quantitative secondary data with open-access, hotel-review data, and text analytics. Data source: Review text, ratings, and metadata fields, which formed the basis of the review data, were derived from public TripAdvisor hotel-review datasets, the Hugging Face hotel-review corpus, and Carnegie Mellon University (CMU) hotel-review resources (Bansal, 2018; Li et al., 2015; Argilla, 2023).

Sampling frame: Sampling was done using Dubai as an example of UAE hotel destinations, accompanied by the destinations of Abu Dhabi and Sharjah to equate the location of the study. Unit of analysis: A single review observation (online) consisting of a rating, review text, and coded analytical variables. Reviews were considered based on established inclusion standards to ensure uniformity and relevance to the study goals. Only hotel reviews that were publicly accessible and matched hotels in Dubai, Abu Dhabi, and Sharjah were considered for analysis. Reviews with incomplete and duplicate entries or inadequate textual content for sentiment and thematic analyses were omitted. The final analytical sample included 220 review-level observations, reflecting a variety of hotel properties in the chosen UAE locations.

Data preparation

A standardized coding process was created before doing the analysis to ensure uniform classification of the review content. The text was cleaned and standardized using keyword matching. Sentiment, smart-technology themes, technology intensity, and other analytical variables were then systematically coded for subsequent analysis. Keywords related to technology were marked as established smart hotel technologies such as mobile check-in, keyless access, contactless payments, smart room management, digital concierge services, in-room tablets, automated service kiosks, and customized applications. Quality parameters for the experience were determined using established service-related words that showcase convenience, reliability, personalization, responsiveness, and the overall service experience. The systemic coding framework provided clarity and enhanced the reproducibility of the analytical process.

Method of analysis

Descriptive statistics, sentiment distribution, theme frequency analysis, Pearson correlation, and ordinary least-squares regression (Figure 2) were used. Ethical note: The study refers to open-access/public secondary data and does not involve the use of personal identifiers. The review-level analytical fields and source references are the only reference fields in the worksheet. The customer satisfaction rating on a five-point scale is the dependent variable. The independent variables are the sentiment score and the smart-technology index of intensive and experience quality (Figure 3; Table 1).

Table 1. Summary statistics for smart hotel review analysis.

The model used is Rating = b0 + b1 (Sentiment) + b2 (Smart Technology Intensity) + b3 (Experience Quality) + e. The design is in line with existing studies in the field of hospitality text mining that relate the valence of the review and service features to the rating results (Figure 4; Table 2) (Filieri, 2015; Blei et al., 2003; Buhalis and Law, 2008; Buhalis and Amaranggana, 2014; Gretzel et al., 2015; Koo et al., 2015).

Table 2. Regression results predicting customer satisfaction rating.

The study posed no ethical risks, while upholding the responsible use of publicly accessible data because it examined reviews that were only publicly available and did not gather personally identifiable information or engage with individuals.

Data coding, cleaning, and reliability procedure

As the research is based on a systematic key term-driven coding framework with created analytical categories, the concern was on transparency in coding and reproducibility instead of the researcher’s personal interpretation. The coding framework remained uniform during the analysis, enabling the identification of all analytical variables through standardized coding rules consistently applied across the dataset. This method promotes methodological clarity and allows future researchers to reproduce or build upon the analytical process using the same coding structure. The raw data sheet uses fields of source and derived fields. The base name, source name, source URL, city, hotel name (label), ratings, and review text are considered source or source-linked fields. Analytical fields are the sentiment score, the intensity of smart technology, the experience-quality index, and binary technology theme indicators. This difference is significant as the sentiment dictionary or topic model used by a later researcher can be modified without modifying the supporting reviews (Buhalis and Law, 2008; Gretzel et al., 2015; Sigala, 2018; Pappas et al., 2014; Bilgihan et al., 2014; Lu and Stepchenkova, 2015; Mariani et al., 2018, McKinney, 2010). The technology of smart hotels was subdivided into access technology, in-room technology, transaction technology, and service intelligence. Some of the access technologies include mobile check-in and keyless entry. Smart rooms and room tablets are allowed as in-room technology. Contactless payment is a type of transaction technology. Service intelligence involves digital concierges, automated kiosks, and application customization. These classes enable the text of a qualitative review to be transformed into a form of analyzable variables without the loss of service meaning (Sigala, 2016; Li et al., 2022; Kar, 2021; Neuhofer et al., 2014, 2015; Parasuraman et al., 1988; Venkatesh et al., 2003). An evaluation based on a dictionary was applied as it is auditable and transparent. Instead of suggesting that the work has a definitive sentiment tool, the study proposes that there is a reproducible workflow. In a full-scale extension, the identical raw data sheet can be extended with the help of VADER, TextBlob, or transformer-based sentiment classification or manually validated aspect-sentiment coding. The existing approach would be apt for a secondary open-access article, as all the variables applied in the analysis can be found in the appendix (Buhalis and Law, 2008; Buhalis and Amaranggana, 2014; Gretzel et al., 2015). Basic text preprocessing, including text normalization and keyword standardization, was performed before sentiment classification to improve coding consistency across review observations.

Data analysis and results

The empirical findings reveal the positive feedback of smart hotel reviews, although mixed and negative comments are not less significant because they pinpoint the aspects of failure in technology-facilitated service. The most notable themes of technology are mobile check-in, smart-room control, keyless entry, room tablets, contactless payment, digital concierge, and automated service. These topics coincide with guest-facing issues, in which digital convenience is apparent and hence influences satisfaction (Sigala, 2016; Li et al., 2022; Kar, 2021; Neuhofer et al., 2014, 2015; Han and Hyun, 2017; Kim and Qu, 2014; Parasuraman et al., 1988; Fornell et al., 1996; Anderson and Sullivan, 1993; Venkatesh et al., 2003; Davis, 1989; Cohen et al., 2003, Creswell and Creswell, 2018). The bivariate statistics substantiate both hypotheses H1 and H2, as both sentiment and smart-technology intensity have positive correlations with ratings. The quality of experience was also positively correlated to the rating, supporting H3. A significant portion of the rating variance is explained by the Ordinary Least Squares (OLS) model. However, the smart-technology intensity multivariate coefficient has negative values following the removal of sentiment and experience quality. This does not imply that technology makes one less satisfied; rather, technology is mentioned in both praise and complaint situations, and its impact needs to be put in perspective in terms of sentiment direction (Filieri, 2015; Buhalis and Law, 2008; Koo et al., 2015; Park et al., 2012; Gretzel and Yoo, 2008; Hair et al., 2022; Kandampully et al., 2015; Kline, 2015; Lee et al., 2013). The best empirical evidence is that technology brings satisfaction when it is perceived as a reduction of friction and not as something new. Systems are characterized by positive reviews as smooth, convenient, and seamless and are encouraged by staff. On the other hand, some systems were described as broken, confusing, or slow, resulting in negative customer reviews and lower customer satisfaction. Thus, it is clear that both statistical and text data concur with the idea that the value of smart hotel technology depends on its integration with service recovery and human assistance (Neuhofer et al., 2014, 2015; Venkatesh et al., 2003; Agresti, 2018).

Robustness and interpretation of statistical findings

The strong correlation between sentiment and rating is common because both measures indicate an evaluative tone. Satisfaction is numerically rated as satisfaction, and although the same experience is expressed in words, this represents sentiment. The regression results should be interpreted along with the correlation analysis and not independently. Although the intensity of smart technology exhibited a positive bivariate connection with customer ratings, its regression coefficient became negative when sentiment and experience quality were combined. This pattern suggests that technology-related mentions occur in both positive and negative service scenarios. However, customer satisfaction is more directly related to visitors’ overall perception of service quality and emotional experience than to the frequency of technology references. Thus, the findings reinforce the idea that technology creates value primarily when it is integrated with efficient human service delivery. The inclusion of factors such as smart-technology intensity and experience quality further helps in breaking the seemingly promiscuous positive or negative interpretation and indicates which elements of the guest experience are reflected in the language of the reviews (Filieri, 2015; Koo et al., 2015; Hastie et al., 2009; Mikolov et al., 2013; Morosan and DeFranco, 2016; Noone and McGuire, 2013). The methodological finding regarding the reversal of the sign of the smart-technology coefficient is critical. A common word in review mining does not necessarily imply customer satisfaction. Visitors mention digital systems when they perform well and fail as well. Hence, the true meaning of this is that hotel managers must not only assess the frequency of technology theme occurrences but also determine whether the theme is associated with positive or negative affect (Buhalis and Law, 2008; Buhalis and Amaranggana, 2014; Gretzel et al., 2015; Han and Hyun, 2017). The regression model is appropriate for the analysis process, and the findings are considered secondary data evidence, as opposed to field-survey proprietary evidence. Greater journal submissions can contribute to hotel category, type of traveler, review date, price range, brand category, and site. However, the current open-data method is also scholarly justifiable given that the raw data sheet, coding logic, and statistics are transparent (Thelwall et al., 2010; Wang et al., 2019; Bilgihan and Nejad, 2015).

Discussion

The results confirm that the vision of smart hotels is about guest experience rather than technology demonstration. Keyless entry, mobile check-in, and contactless payments help increase the levels of satisfaction of people as they have less time and less uncertainty to wait. A smart-room control system and a digital concierge system help when it feels personalized, reliable, and in control. What is less desirable is automation, where there are delays in service delivery or guests perish in case of problem-solving (Law et al., 2014; Sigala, 2016; Li et al., 2022; Kar, 2021; Parasuraman et al., 1988). This is particularly applicable to the UAE scenario, as tourists tend to be sensitive to experiences of high standards of service and state-of-the-art hospitality settings. This implies that, aside from adoption rates, smart technologies should be considered in terms of review-based convenience, reliability, and emotional response. Review form-based dashboards and similar features should assist managers in mapping technology themes to rating performance and sentiment trends (Filieri, 2015; Filieri et al., 2015; Han and Hyun, 2017; Kim and Qu, 2014). Theoretically, this study contributes to the smart hospitality field by demonstrating that smart technologies should not be evaluated only based on their adoption or functional efficiency. Rather, evidence from online reviews indicates that the success of smart hotel technologies largely hinges on how well they are paired with attentive human services and favorable customer experiences. By integrating technology-related themes, sentiment analysis, and customer ratings within a clear secondary-data framework, this study expands existing models of technology acceptance and service quality into real-world digital service settings. These results reinforce the idea that smart hospitality operates as a sociotechnical service system, where technological progress and human interaction together shape customer satisfaction. The idea of smart hospitality must, therefore, be seen as a sociotechnical service system, as opposed to an exclusively digital upgrade (Li et al., 2022; Neuhofer et al., 2014; Oliver, 1980; Park et al., 2007; Morosan and DeFranco, 2016a). In order to guarantee dependable technology-enabled visitor experiences, the growing dependence on smart technologies further emphasizes the significance of digital resilience, cybersecurity, data protection, and service continuity.

Practical recommendations

To begin with, hotels need to develop a review-monitoring dashboard that will trace themes of smart technology individually, but without considering other overall service comments. The following categories need to be followed: mobile check-in, smart-room control, contactless payment, keyless entry, and digital concierge. When negative sentiment increases in a single category, hotels can react through specific training or technical remediation and not through general or blanket service messages (Filieri, 2015; Filieri et al., 2015; Han and Hyun, 2017). Second, hotels should develop a human fallback for each automated touchpoint. The results indicate that guests value convenience and are unsatisfied when automation collapses out of service. A digital concierge must, in turn, be linked to staff escalation, and mobile check-in must be provided with precise instructions and visual help icons. The hybrid approach of being digitally efficient while still embracing an emotional kind of experience seems to be the most powerful in the high-end UAE hotel sector (Neuhofer et al., 2014; Venkatesh et al., 2003; Agresti, 2018; Pedregosa et al., 2011; Phillips et al., 2017; Rogers, 2003; Schuckert et al., 2015; Vermeulen and Seegers, 2009). Third, hotel managers should utilize review texts to aid in the investment of technology. A technology is not supposed to be adopted just because it is used by its competitors. It must be adapted in cases where guests feel it will decrease their waiting time, increase personalization, make rooms more comfortable, or improve service. Such perceptions can be benchmarked with the assistance of open review datasets (property and destination) (Lee et al., 2014; Filieri, 2015; Koo et al., 2015).

Limitations and future scope

Although this study was based on secondary open-access datasets, several limitations should be acknowledged. Online reviews may not fully represent the experiences of all hotel guests because individuals with particularly positive or negative experiences are generally more likely to submit them. In addition, publicly available review data provide limited demographic information, restricting further segmentation of customer characteristics. Consequently, the findings should be interpreted in the context of publicly available review-based evidence, rather than representative population estimates. There might be platform bias, as online reviews by guests might not be equal to silent guests. The sentiment model classifies customer reviews according to their expressed emotional tone, taking into account the contextual meaning of words within each review (Buhalis and Law, 2008; Buhalis and Amaranggana, 2014; Gretzel et al., 2015).

Future studies should rely on the complete Zenodo, Hugging Face, or CMU datasets and filter reviews by the location of the hotels in the UAE, wherever possible, in the metadata. The same can be done for the Bidirectional Encoder Representations from Transformers (BERT)-based aspect sentiment to determine whether a technology theme is being praised or criticized on the same lines as the text. A more robust model might consist of hotel class, review date, traveler type, price group, and hotel brand (Bansal, 2018; Li et al., 2015; Argilla, 2023; Buhalis and Law, 2008; Buhalis and Amaranggana, 2014). Comparative research can explore the possibility that the levels of satisfaction with smart hotels in the UAE may be different from those in other smart tourism destinations. This cross-country comparison can determine the presence of similarities in the digital features rewarded by guests across countries or whether expectations vary by market, hotel segment, and travel intent. The external validity of the current framework (Sigala, 2016; Li et al., 2022; Kar, 2021; Parasuraman et al., 1988) would be exerted by such studies. Future research should examine more closely at how guests’ impressions of smart hospitality spaces are influenced by AI-enabled services, digital resilience, cybersecurity procedures, and privacy issues.

Data sheet used for client submission

The attached raw data sheet should be shown as an analytic, coded secondary open-access data sheet, rather than as a personal survey. It has review-level observations, source links, ratings, review texts, sentiment scores, technology intensity, and theme indicators. Such openness increases the academic defense of the article, as the conversion of text into numerical evidence (Bansal, 2018; Li et al., 2015; Argilla, 2023; Sigala, 2018; Pappas et al., 2014; Bilgihan et al., 2014) is not concealed. In the case where the client or supervisor desires a larger empirical base, the same sheet can be extended with additional rows downloaded from the referenced open-access datasets. The formulas, lines of code, and columns of analysis may remain unchanged. This renders the current output suitable as a draft for a paper as well as a template for organizing data in a working format before submitting them to journals or conferences.

Conclusion

This study concludes that smart technologies in hotels have a positive relationship with customer satisfaction, creating convenient, reliable, and seamless hotel experiences. Sentiment, smart technology use intensity, and experience quality are joint factors in explaining hotel rating patterns in the secondary open-access review sample. The most viable suggestion would be to avoid viewing automation as an alternative to hospitality: smart tools are meant to be complementary to responsive human service and to minimize friction throughout the guest journey. Replication and extension are facilitated by the coded raw data sheet and source-based codebook presented in the article. Overall, this study delivers value in both methodological and practical terms by presenting a transparent, repeatable framework that leverages publicly available online reviews to analyze smart hotel technologies. The findings indicate that these technologies enhance guest satisfaction by simplifying service processes, improving ease of use, and complementing attentive human engagement. Rather than positioning digital tools as substitutes for hospitality, the results highlight the importance of achieving a balance between technological progress and human-centered services. This framework provides practical guidance for industry professionals and offers a replicable analytical model for future research in smart tourism and digital hospitality.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside the figures in this article has been generated by Frontiers with the support of artificial intelligence, and reasonable efforts have been made to ensure accuracy, including review by the authors, wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fhumd.2026.1916785/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Mahmut Demir, Isparta University of Applied Sciences, Türkiye

Reviewed by: Yelena Dikhanbayeva, Nazarbayev University, Kazakhstan

Yusuf Günaydın, Final International University, Cyprus