Snap,share,engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels

Abstract

Despite the recognised importance of visual content in luxury hotel branding, there is a significant lack of research on how these hotels strategically use organisation-generated content on social media to consistently communicate brand identity and sustain audience engagement over time. This study provides new insights by applying AI-powered methods to a longitudinal analysis of 9612 Facebook photographs posted by Italian five-star hotels between 2019 and 2023. Using service-dominant logic and uses and gratifications theory, the analysis identifies seven recurring content categories and evaluates both their temporal consistency and their impact on audience engagement. The results show that non-city hotels maintained consistent, emotionally engaging visual communications during the pandemic, while city hotels experienced longer interruptions. Photographs with symbolic and hedonic appeal, especially those showing nature and interiors, led to significantly more engagement than functional images. To better evaluate user interaction, a new relationship-based engagement metric is introduced. The study makes a theoretical contribution by deepening the understanding of value co-creation in digital hospitality environments and a practical one by providing scalable, AI-powered tools for content evaluation. It also highlights differentiated strategies for luxury hotel marketers looking to align their visual communications with audience expectations and behaviour on the platform.

Keywords

Luxury hotels visual communication social media engagement AI-assisted analysis organisation-generated content (OGC)topic modelling brand image

Introduction

Social media (SM) has become indispensable in the hospitality and tourism industry, playing a central role in brand awareness, customer engagement, and value co-creation (Ferrer-Rosell et al., 2019; Leung et al., 2017; Simmons and Durkin, 2023). Within this landscape, visual content, such as photographs and videos, has emerged as a powerful medium, offering immersive and emotionally resonant brand storytelling (Gretzel, 2016; Pan et al., 2013). While research has extensively examined user-generated content (UGC) and its impact on destination and hotel branding (Cheng et al., 2021; Gretzel et al., 2007), the role of organisation-generated content (OGC), especially visual content produced by hotel brands, remains significantly underexplored (Blanco-Moreno et al., 2024; Rabadán-Martín et al., 2025). This gap is particularly important given that hotel brands predominantly rely on the use of photographs and videos (i.e. OGC) to communicate with audiences (RivalIQ, 2024b).

Despite the growing reliance on visual content, there is no systematic classification of hotel photographs that could guide brand communication strategies (Picazo et al., 2019). Furthermore, since audiences engage with media that meets their informational, emotional, or entertainment needs, Rabadán-Martín et al. (2025) suggest that as part of broader marketing efforts, hotels should prioritise posting content that generates higher engagement rather than focusing solely on promotional material. This corresponds to the uses and gratifications theory (UGT) (Katz et al., 1973), which postulates that users engage with media only when they consider it a valuable use of their time. Yet there is little empirical guidance on which hotel-generated visuals satisfy audience needs; thus, this paper's primary goal, in line with UGT, is to explore how visual content can be optimised to reflect brand values and drive consumer interaction. To aid in answering this question this research adopts a value co-creation lens, in line with service-dominant logic (S-D logic), viewing the brand value as collaboratively constructed through SM interactions between organisations and users (Neuhofer and Buhalis, 2017; Simmons and Durkin, 2023), where visual content, especially photographs, becomes a medium through which brand meaning is negotiated and co-created in real time.

Extant studies have explored destination imagery and tourists’ perceptions using content, semiotic, and visual analysis (Michael and Fusté-Forné, 2022; Song and Kim, 2016). However, these methods are limited in scalability, generalisability, and longitudinal insight. As a result, scholars increasingly advocate for artificial intelligence (AI) and machine learning approaches. These enable large-scale, objective analysis of visual data, facilitate the detection of patterns over time, and minimise human bias, all of which are essential for developing evidence-based visual marketing strategies in an increasingly data-driven landscape (Lee et al., 2022; Wang et al., 2020).

This study responds to these gaps by focusing on the under-researched domain of OGC, specifically visual content created by luxury hotel brands. While luxury tourism is attracting growing academic interest due to the “luxurification of society” (Michael and Fusté-Forné, 2024), little is known about how these brands construct their visual identity on SM. In this context, visual storytelling becomes essential for conveying intangible brand attributes such as prestige, exclusivity, and emotional resonance without relying on explicit promotional messaging (Šerić and Mikulić, 2020). Furthermore, while consistency in luxury brand communication is known to enhance trust, image, and perceived quality, its influence on audience engagement has rarely been investigated.

To address these conceptual, empirical, and methodological gaps, this study employs AI-assisted techniques to conduct a 5-year analysis of photograph-based OGC shared by Italian five-star hotels on Facebook. Italy serves as a meaningful case study due to its reputation as one of the world's premier luxury travel destinations and its mature luxury hotel market (Statista, 2021). By examining content themes, communication consistency, and user engagement patterns, the study provides practical insights for luxury hotel marketers and contributes to the emerging scholarship on OGC, visual storytelling, and AI-powered tourism research.

Based on the extant literature, the following research questions guide the paper:

RQ1: What are the prevalent topics in the organisation-generated photographs?

RQ2: What is the consistency of communicating via organisation-generated photograph topics, over a 5-year period?

RQ3: What impacts do organisation-generated photograph topics have on audience engagement?

The findings contribute to the expanding knowledge on the visual communication of projected images of luxury hotels in SM. They go beyond the traditional textual approach, and address calls for further research (e.g. Li and Ma, 2024; Li and Wan, 2025). These findings provide marketers of luxury hotels with practical guidance to develop timely, targeted and effective SM marketing strategies to engage potential guests, promote vivid travel fantasies, and stimulate visit intentions. Furthermore, this paper provides an overlooked theoretical and empirical perspective on the projected image and visual communication of luxury hotels, which is crucial for shaping value propositions. It emphasises the dynamism and consistency, communication effectiveness, and engagement of luxury hotels’ visual SM communication, contrasting with previous studies which focus on text-based UGC (e.g. Piriyakul et al., 2024; Rabadán-Martín et al., 2025). Finally, the study identifies critical visual marketing communication strategies and differentiates between visual SM marketing practices for city and non-city hotels.

Literature review

Visual online-generated content and brand image in the luxury hotel context

Visual communication has become an essential tool for luxury hotel brands seeking to distinguish themselves in a competitive digital landscape. As such, SM platforms provide a dynamic space where photographs and other forms of OGC are not only aesthetic expressions but also strategic instruments for shaping brand image. Such content is central to how organisations project their brand image to reach broader audiences (Oliveira and Fernandes, 2022). Photographs, as highlighted by Yu and Egger (2021), have a significant impact by capturing and conveying tourist experiences, shaping cognitive and emotional perceptions of destinations, serving as mementoes, and inspiring travel desires. Their use in branding support the signalling of service quality, reduces perceived risk, and strengthens an organisation's value proposition (Almeida-Santana and Moreno-Gil, 2017; Francesconi, 2011; MacKay and Couldwell, 2004). Visual OGC helps to construct customers’ mental images of travel experience (Picazo and Moreno-Gil, 2019; Qing et al., 2024) and thus correspond to the S-D logic of shared value creation (Neuhofer and Buhalis, 2017).

In the past decade, analysing tourist photographs has become a prominent research focus. Earlier studies manually coded photograph content based on predefined attributes (Song and Kim, 2016; Stepchenkova and Zhan, 2013), whereas more recent research has adopted deep learning, machine learning, and AI technologies to improve the classification and understanding of visual content (Appendix A). Despite the recognised importance of OGC and photographs for value creation and advances in analytics tools and approaches, image evaluation to date was mainly considered from the tourist or DMO perspective, with a limited focus on projected brand image and user reactions. As Picazo et al. (2019) emphasise, research on projected image is still scant and lacks a holistic approach.

In the luxury context, brands increasingly use SM to portray their image (Park and Ahn, 2021), build loyalty and trust (Eastman et al., 2020), and establish personal connections with customers (Cervellon and Galipienzo, 2015). SM communication can enhance the perception of a luxury hotel's brand authenticity (Qing et al., 2024), increase purchase intentions (Park et al., 2021), and impact brand association (Qing et al., 2024; Shakuntala and Ramantoko, 2023). Understanding users’ interactions with visual OGC can strengthen a brand's core messages by aligning photographic OGC with user needs. Kim and Lee (2019) found that engagement on luxury brands’ SM page, both behavioural and perceived, positively influences attitudes, loyalty, and purchase intent. Michael and Fusté-Forné (2024) stressed the importance of ensuring luxury brand visuals are consistent with brand identity and strategically crafted to attract likes.

Eijdenberg et al. (2024) and Jain et al. (2023) highlight that luxury is context-dependent and inherently subjective and is shaped by individual experiences and needs, cultural norms, and personal preferences, making it challenging to define universally. Although it is often associated with superior service, unique attributes, and premium prices, its interpretation can vary greatly depending on the target group and environment (Eijdenberg et al., 2024; Manfreda et al., 2022). This variability complicates luxury communication, particularly when it comes to visually representing its intangible and perishable nature. Static visual media often falls short in capturing these abstract qualities. As Daun and Klinger (2006) emphasise, consistent brand value communication is essential, yet luxury hotels frequently struggle to convey such abstract dimensions effectively via their imagery.

Although luxury is linked with notions of quality, authenticity, elitism, distinctiveness, and hedonism (De Barnier et al., 2012; Ko et al., 2019), as well as experiential, emotional, qualitative, and functional values (Manfreda et al., 2022), these elements are difficult to represent visually in a compelling and differentiated manner. Cervellon and Galipienzo (2015) argue that uniform brand communication, especially consistent design across platforms, is essential for reinforcing brand meaning. In this context, digital storytelling becomes crucial. As Michael and Fusté-Forné (2022) underline, narrative-driven visuals help shape and sustain luxury brand imagery, particularly when aligned with symbolic and emotional attributes. Mapping and aligning luxury value with user expectations and experiences, therefore, enhances the co-creation process. To that end, identifying the visual content themes that generate the highest engagement is essential. For the luxury hotel OGC, this involves aligning visual communication with consumers’ perceptions and values. By prioritising the types of content that resonate most with users, brands can engage in ongoing co-creation of meaning, which aligns with S-D logic's process orientation principle wherein ‘brand loyalists’ actively reinforce brand meaning (Simmons and Durkin, 2023).

Although previous studies provide useful insights into effective visual strategies, a systematic framework for luxury hotel imagery remains underdeveloped. To date, only Picazo et al. (2019) have proposed a six-dimensional categorisation derived from manual content analysis. In the absence of more comprehensive conceptual models, the literature on luxury OGC largely focuses on three key dimensions of perceived consumer value: hedonic, symbolic, and functional.

Hedonic value

Luxury branding frequently draws on the hedonic value, aiming to evoke emotional pleasure and sensory enjoyment. Research shows that luxury hospitality consumers respond more strongly to emotionally engaging and experiential visuals than to functional representations (Deb and Lomo-David, 2020). Manfreda et al. (2022) emphasise the impact of symbolic physical elements, such as scenic views, mood lighting, and ambiance, in fostering emotional attachment. Similarly, Hyun and Kang (2014) demonstrate that lavish environmental cues like décor and atmosphere produce more profound emotional responses than utilitarian features such as food presentation.

Symbolic value

From a symbolic value standpoint, visual storytelling serves both to affirm the consumer's self-identity and to convey social distinction. Michael and Fusté-Forné (2024) argue that luxury hotel imagery should foreground exclusivity and prestige through panoramic views and elite settings that resonate with the consumer's aspirational self-concept. This dual signalling mechanism aligns with both self-concept (internal identity affirmation) and conspicuous consumption theory (external prestige signalling) (Oliveira and Fernandes, 2022). Symbolic visuals thus enable consumers to express refinement and uniqueness while simultaneously enhancing social visibility and perceived status.

Functional value

Although hedonic and symbolic values dominate luxury branding discussions, some scholars continue to emphasise the role of the functional value. Giglio et al. (2020) advocate for showcasing tangible elements such as room layout, amenities, and dining facilities to reinforce perceptions of quality and service reliability. However, such recommendations often stem from consumer-generated content, which may not align with the brand's strategic narrative. Picazo et al. (2019) warn that an overemphasis on functional features risks diluting the brand's emotional and symbolic appeal, ultimately reducing luxury branding to generic visual representations, which may fail to meaningfully differentiate brands in a saturated market (Pan et al., 2013).

The meaning and measurement of user engagement

User engagement has become a key focus in both academic and practical marketing contexts, especially within SM environments. It refers to the depth of a user's cognitive, emotional, and behavioural involvement with brand content (So et al., 2021). Engagement is frequently viewed in terms of the intensity of a user's interaction with an organisation's offerings or activities (Lim and Rasul, 2022). More recently, customer engagement has been operationalised as a user's willingness to participate in dialogues surrounding specific products, often measured through interactions such as likes, comments, and shares on SM platforms (Eslami et al., 2022). These interactions provide real-time feedback on user preferences and serve as indicators of content effectiveness (Sörensen et al., 2023). A high SM engagement not only increases the visibility and virality of content but also influences attitudes towards destinations and brands, contributing to behavioural outcomes such as intent to visit (Blanco-Moreno et al., 2024; de Oliveira Santini et al., 2020). In addition, monitoring engagement provides actionable insights that help brands adapt communication and co-creation value with users (Blanco-Moreno et al., 2024; So et al., 2021).

UGT posits that audiences actively seek media that satisfy specific psychological or social needs (Katz et al., 1973; Shao, 2009). Eslami et al. (2022) noted that users interact with different platforms to satisfy diverse needs. SM engagement is therefore not only determined by the search for information but also by the user's desire for connection and self-expression (Shao, 2009), which is particularly important in the context of luxury hospitality. Thus, Cervellon and Galipienzo (2015) note that luxury communication should evoke emotions rather than be purely informative, as emotional advertising shapes brand attitudes and generates strong feelings. Cheng et al. (2021) stress the importance of combining informative and emotional content to drive digital engagement and note that engagement metrics should include comments and shares, not purely likes. As such, in line with UGT, understanding user preferences for visual OCG, that is, what type of content, and on which platform, generates optimum engagement, can inform marketing and sales strategies.

Methodology

As the basis of this research, Facebook was chosen due to its consistent user engagement and its suitability for longitudinal analysis. Unlike Instagram, Facebook provides better access to historical data, constant public profiles, and more consistent audience engagement (Gálvez-Rodríguez et al., 2020). The multimodal post structure, which integrates visuals, text, and interaction metrics, is particularly well suited to analysing visual storytelling in the luxury brand space (Cervellon and Galipienzo, 2015; Ryu et al., 2018). In addition, Facebook continues to perform well in the travel and hospitality sector (RivallQ, 2024a) and remains a key platform for luxury hotels targeting a suitable demographic of customers (Kucukusta et al., 2019).

Balomenou and Garrod (2019) emphasise the subjective nature of photo-based expressions, prompting to mitigate this by employing non-human methods for data interpretation and analysis; thus this research approach involved a multi-stage, multi-method process, illustrated in Figure 1 and outlined below.

Figure 1.

Research process.

Phase 1: hotel selection and web scraping

The data collection process employed a web data extraction tool to access real-time and historical Facebook data. Parameters were manually configured to enable automated data gathering. With 601 five-star hotels in Italy, selection criteria were essential for valid analysis. The chosen period, from 1 January 2019 to 31 December 2023, encompassed pre-, during, and post-pandemic phases, allowing for analysis of communication consistency. The study focused on hotel brands listed in the top 10 pages of Italian Tripadvisor rankings, narrowing the sample to 114 brands. This was further reduced to 64 brands by identifying those with official Facebook profiles. Additional criteria included regular posting (at least three posts per month, as per Ferrer-Rosell et al., 2020) and the use of English in communications. Ultimately, 17 hotel brands (26.5% of the initial sample) were selected. Over the 5 years, the sample included 9724 posts, featuring images, text, reactions (e.g. like, love, haha), shares, and comments. A summary of the research procedure is presented in Figure 1.

Phase 2: photo and label extraction

To prepare the data for analysis, all photographs were extracted from the posts. Google Vision AI was then utilised to analyse the image content and assign labels. This pre-trained tool categorises image features into millions of predefined categories. For practicality, the data collection was restricted to the 10 most common labels assigned to each photograph.

Phase 3: topic analysis

Topic analysis of the extracted photo labels was conducted using the Dirichlet Multinomial Mixture model. This generative probabilistic framework, which combines mixture models with Dirichlet distributions, was essential for clustering categorical data into distinct topics. By treating each label as indicative of a latent topic characterised by a specific word distribution, the model identified seven underlying topics (Table 1). The model's parameter estimation capabilities allowed for maximising the likelihood of the observed data, enabling precise grouping of labels into well-defined topics. This approach not only highlights the nuances of visual communication but also demonstrates the value of integrating natural language processing with machine learning to uncover significant patterns in visual data. Detailed results, including heat maps and word clouds, are presented in Appendix B. Data preparation involved transforming text documents into a token matrix with defined document frequency thresholds.

Table 1.

Total number of visual posts per topic.

Topic	Total posts across the sample	Posts across city hotels	Posts across countryside/ coastal hotels
Outside Nature	2087	558	1529
Drinks	669	564	105
Interior/Furnishings	1741	1444	297
Outside City	2552	1618	934
Food	1093	907	186
Fashion/Decoration	1118	727	391
Automotive/Mechanical	238	146	92
Total	9498	5964	3534

Source: authors.

Phase 4: exploratory data analysis

The data exploration process was iterative and involved the following steps:

Data cleaning: Facebook posts with >4000 reactions and >400 shares were considered outliers as they skewed the data. Therefore, the analysis was performed on a sample of 9489 posts.

The initial analysis focused on understanding the relationship between each topic and the number of comments, reactions, and shares received throughout the data collection period. This was followed by a deeper examination of the event rate of these interactions to determine the likelihood and type of engagement posts attracted. Patterns began to emerge, prompting further exploration into variations and potential clusters in the data. A breakdown of interactions by hotel revealed notable differences between those located within the city and those outside it, introducing an unexpected but significant layer to the analysis. To investigate this, hotels were manually grouped into two categories—city (11 hotels) and non-city (6 hotels)—based on their official website information. Subsequent internal consistency checks confirmed that these groupings accounted for the observed patterns, that is, there were significant differences between not only in the types of topics used depending on the location but also on the type and intensity of the interactions.

Correlation analysis: Spearman's rank correlation coefficient was used to assess the strength and direction of relationships between variables.

Linear regression: Relationships between comments, shares, and reactions across identified topics were evaluated using linear regression.

Data categorisation: To determine which topic most strongly influences reactions, the OptBinning Python library was used for optimal binning of continuous variables (e.g. reactions, shares, and comments) into categories (‘bins’). This improved interpretability and revealed patterns in the data, ensuring strong and monotonic relationships between variables and the extracted topics. Resulting histograms of bins and topics were created (Appendix C)

Engagement analysis: The probability of reactions relative to shares and comments per topic was analysed (Appendix D) to identify topics most effective in driving high or low user engagement.

Results

Prevalent topics of communication

The application of a Dirichlet Multinomial Mixture model was used to cluster the 10 image labels for every photograph into seven different topics of visual communication. Figure 2 illustrates the results of the topic modelling and shows a clear separation between the identified topics: ‘Interior/Furnishings’; ‘Food’; ‘Drinks/Beverages’; ‘Fashion/Decoration’; ‘Automotive/Mechanical’; ‘Outside Nature’; and ‘Outside City’.

Figure 2.

Topic modelling, topic cluster results.

Table 1 highlights the popularity of identified topics, with ‘Outside City’ being the most common, totalling 2552 posts, followed by ‘Outside Nature’ (2087 posts) and ‘Interior/Furnishings’ (1741 posts). In contrast, ‘Automotive/Mechanical’ was the least common, with only 238 posts. The frequency of visual topics varies, reflecting diverse communication strategies among the sampled luxury hotels. The iterative pattern analysis revealed differences in topics of visual OGC between city and non-city hotels. As such, city hotels emphasise ‘Outside City’, ‘Interior/Furnishings’, and ‘Food’, showcasing urban landscapes, elegant interiors, and culinary experiences to attract guests. Non-city hotels focus on ‘Outside Nature’, ‘Outside City’, and ‘Fashion/Decoration’, highlighting natural surroundings and stylish décor to appeal to those seeking scenic retreats. These tailored visual strategies align with the distinct characteristics and target audiences of each hotel type. Figure 3 presents the topic correlation analysis results.

Figure 3.

Results of correlation analysis.

Visual communication consistency over time

The results (Figure 4) reveal a marked drop in SM posts in March and April 2020, aligning with the onset of the COVID-19 pandemic. Posting activity further declined in September 2020 and January 2021. City hotels experienced the steepest reductions in visual communication, failing to return to pre-pandemic levels even after restrictions eased. Conversely, non-city hotels maintained a more consistent, seasonally driven posting pattern, with fluctuations tied to events like Christmas and less dramatic changes during lockdowns. Unlike city hotels, non-city hotels sustained consistent visual communication since the pandemic's start, reflecting resilience and adaptability likely influenced by their locations and audience preferences.

Figure 4.

Communication consistency over time.

When analysing the changes in visual communication over time for each topic, Figure 5 shows that across the entire sample, the only topics that were not affected by the decline in SM posts during the COVID-19 period were ‘Outside Nature’ and ‘Automotive/Mechanical’.

Figure 5.

Non-city versus city hotel posts per topic per month.

An analysis of hotel locations reveals distinct patterns in visual communication. City hotels predominantly focused on ‘Interior/Furnishings’, ‘Outside City’, and ‘Food’, all of which saw significant declines in posting activity during the initial lockdowns. Notably, ‘Interior/Furnishings’ and ‘Outside City’ did not recover to pre-pandemic levels. In contrast, non-city hotels prioritised ‘Outside Nature’ and ‘Outside City’, with these topics seeing increased posting frequency during lockdowns. This shift reflects a focus on natural beauty and outdoor environments. Post-pandemic, non-city hotels have maintained or even increased their visual communication for these themes.

Impact of visual communication on audience engagement

Effectiveness of topics to attract audience engagement

In terms of communication effectiveness and engagement, Figure 6 shows the main communication topics for each hotel group and the corresponding reaction, comment, and share metrics, highlighting notable differences in the ability of the main communication topic to capture audience engagement.

Figure 6.

Main communication themes versus reactions, comments and shares.

The optimum binning procedure resulted in two distinct reaction clusters for hotels based on location (Figure 7, Appendix C). City hotels were grouped into four bins, 0–7, 8–23, 24–34, and 35+ reactions, while non-city hotels were divided into two bins, 0–42 and 43+ reactions per post. For city hotels, the topics ‘Outside City’, ‘Interior/Furnishings’ and ‘Outside Nature’ attract more responses (i.e. 35+), while all other topics show a decreasing trend towards a higher number of reactions. In non-city hotels, only ‘Outside Nature’ and ‘Interior/Furnishings’ showed a positive trend towards higher reaction counts.

Figure 7.

Reaction of bin clusters city and non-city hotels.

For the city hotels, there was a notable decrease in the reactions from the period of March 2020 to June/July 2022 across all topics except ‘Automotive/Mechanical’. This has not been the case for non-city hotels. Indeed, reactions towards ‘Outside Nature’ and ‘Outside City’ topics during the same period have peaked substantially.

For comments, city hotels were grouped into bins of 0–1 and 2+ comments, while non-city hotels were categorised into 0, 1–2, and 3+ bins. Among non-city hotels, ‘Outside Nature’ demonstrated an upward trend, attracting 3+ comments. ‘Interior/Furnishing’ and ‘Drink’ exhibited a convex trend, with fewer posts attracting 1–2 comments but an increasing number reaching 3+ comments. For city hotels, topics such as ‘Outside City’, ‘Drink’, ‘Fashion/Decoration’, and ‘Automotive/Mechanical’ showed an upward trend in attracting 2+ comments, while other topics generally displayed a decline in comment engagement. Comment activity for city hotels over time aligns with posting trends during the COVID lockdown period and is characterised by fluctuations. Pre-lockdowns, peaks in comments were associated with ‘Interior/Furnishing’ and ‘Outside Nature’, whereas post-pandemic spikes primarily related to ‘Fashion/Decoration’ and ‘Outside City’. In contrast, non-city hotels exhibited steady comment activity pre-lockdown, with no significant topic-specific peaks. Post-pandemic, overall comment activity remained steady, with occasional spikes linked to ‘Automotive/Mechanical’, ‘Fashion/Decoration’, and ‘Interior/Furnishing’ topics.

The binning procedure categorised shares into the following groups: 0–1, 2–6, and 7+ for city hotels and 0–1, 2–27, and 28+ for non-city hotels. For city hotels, ‘Outside City’ and ‘Outside Nature’ demonstrated a clear upward trend in attracting 7+ shares. In non-city hotels, ‘Outside Nature’ showed a strong tendency to attract 28+ shares. ‘Fashion/Decoration’ exhibited a convex trend, with a limited capacity to generate 2–27 shares but a notable increase in attracting 28+ shares. City hotels experienced a decline in share activity between mid-2020 and mid-2022, while periods outside this timeframe showed relatively consistent activity with similar patterns on either side. In contrast, non-city hotels recorded a significant drop in share activity post-mid-2021, primarily due to a sharp decline in shares related to ‘Outside Nature’ and ‘Fashion/Decoration’, which negatively impacted overall share performance over time.

Relationships of topics against reactions, shares, and comments

Linear regression analysis explored relationships between comments and shares, comments and reactions, and shares and reactions. For city hotels, the strongest correlation was between shares and reactions, with ‘Outside Nature’ achieving an R-squared (R²) value of 0.66, followed by ‘Outside City’ with 0.62. For non-city hotels, the topic with the highest R² value was ‘Fashion/Decoration’ (0.78), indicating a strong relationship between shares and reactions, followed by ‘Drink’ with an R² of 0.76 (Figure 8).

Figure 8.

Regression results per topic.

A probability analysis was conducted to identify the highest-performing topics in terms of comments-to-reactions, comments-to-shares, and reactions-to-shares (see Figure 9 for an example, full results in Appendix D). For city hotels, the topic with the highest probability of comments-to-reactions was ‘Drink’, with a 45% likelihood of achieving 0–1 comment and 8–23 reactions. In terms of comments-to-shares (0–1 for both metrics), the top-performing topics were ‘Drink’ (67.2%), ‘Food’ (67.1%), ‘Interior/Furnishings’ (62.6%), and ‘Fashion/Decoration’ (62%). No topic surpassed a 40% probability of achieving reactions-to-shares. The highest probabilities for 8–23 reactions and 0–1 share were ‘Interior/Furnishings’ (38.8%), ‘Drinks’ (38.6%) and ‘Food’ (34.4%). Only three topics exceeded a 10% probability of achieving 7+ shares and 35+ reactions: ‘Outside Nature’, ‘Outside City’, and ‘Automotive/Mechanical’.

Figure 9.

Exemplar of probability analysis.

For non-city hotels, ‘Outside Nature’ was the dominant topic, with a 57% probability of achieving 3+ comments and 43+ reactions. Three additional topics exceeded a 40% probability for 3+ comments and 43+ reactions: ‘Interior/Furnishings’ (48.8%), ‘Drink’ (44.7%), and ‘Fashion/Decoration’ (43.9%). In terms of comments-to-shares, the highest probabilities of achieving 3+ comments and 2–27 shares were ‘Outside Nature’ (45.2%), ‘Drink’ (43.2%), and ‘Interior/Furnishings’ (43.4%). For reactions-to-shares (43+ reactions and 2–27 shares), the most effective topics were ‘Outside Nature’ (72.3%), ‘Interior/Furnishings’ (64.3%), ‘Outside City’ (61.2%), and ‘Automotive/Mechanical’ (60.8%).

Discussion

This study offers an original contribution to understanding how luxury hotels project their brand image through visual OGC and how this shapes audience engagement. By applying AI-assisted topic modelling to a large dataset of hotel-generated photographs and engagement metrics, the analysis provides empirical insights into the structure, content level variation, evolution, and performance of visual communication strategies in the luxury hospitality sector.

Topics of communication and brand differentiation strategies (RQ1)

This study identified seven core topics of organisation-generated visual content: ‘Interior/Furnishing’, ‘Food’, ‘Drinks/Beverages’, ‘Fashion/Decoration’, ‘Automotive/Mechanical’, ‘Outside Nature’, and ‘Outside City’. These represent distinct thematic strategies through which luxury hotels articulate brand identity and engage their audiences. Contrary to earlier assumptions that high-end hotels focus predominantly on showcasing facilities (Ferrer-Rosell et al., 2019), this study shows that even five-star hotels foreground destination-based imagery. This nuanced approach allows hotels to simultaneously communicate service quality (Almeida-Santana and Moreno-Gil, 2017) and brand meaning (Michael and Fusté-Forné, 2024), thereby enhancing the co-creation of value through visual storytelling. The emergence of the ‘Automotive/Mechanical’ topic, previously unclassified in the literature, further reflects evolving luxury narratives (Ko et al., 2019; Manfreda et al., 2022) and reinforces the need to account for broader experiential elements in luxury branding. This is in line with S-D logic, where the images become critical resources through which brand value is interactively constructed (Neuhofer and Buhalis, 2017).

Further analysis reveals that these topics map closely onto three well-established dimensions of perceived consumer value in luxury branding, that is, functional, symbolic, and hedonic, as synthesised in prior literature (De Barnier et al., 2012; Deb and Lomo-David, 2020; Ko et al., 2019; Manfreda et al., 2022). The functional value, which centres on utility, service quality, and reliability, is primarily communicated through topics such as ‘Interior/Furnishing’ and ‘Food’, elements that help reduce perceived purchase risk (Almeida-Santana and Moreno-Gil, 2017). The symbolic value, associated with status signalling, identity expression, and exclusivity, is strongly reflected in the use of ‘Fashion/Decoration’, ‘Automotive/Mechanical’, and certain stylised representations of interiors or exteriors. These topics support aspirational self-concepts and enhance social visibility (Oliveira and Fernandes, 2022), as luxury consumers seek content that affirms both personal identity and perceived prestige (Hyun and Kang, 2014; Michael and Fusté-Forné, 2024). The hedonic value, which encompasses sensory pleasure, emotional gratification, and experiential richness, is particularly evident in ‘Drinks/Beverages’ and ‘Outside Nature’. These topics evoke relaxation, indulgence, and escapism, that is, qualities that resonate strongly with users seeking immersive or mood-enhancing content.

The distribution and emphasis of these value categories differ meaningfully by location. City hotels prioritise functional and some symbolic content, emphasising services, facilities, and urban sophistication. Non-city hotels favour hedonic and symbolic themes that foreground nature, décor, and emotionally resonant experiences; these align closely with Manfreda et al.'s (2022) findings on the role of physical aesthetics in generating emotional connection. This distinction reinforces the idea that luxury brand narratives are geographically contextualised, echoing Eijdenberg et al.'s (2024) view that luxury is not universally defined but contingent upon the user's environment and expectations.

Temporal consistency of brand communication (RQ2)

The temporal analysis of posting behaviours revealed a clear divergence in communication consistency between hotel types. Non-city hotels sustained regular visual output, especially around seasonal events, while city hotels experienced a steep and prolonged decline during and after the pandemic, particularly in topics such as ‘Interior/Furnishing’ and ‘Outside City’. From an S-D logic perspective, consistency in content production functions as a mechanism of ongoing value co-creation. By maintaining engagement touchpoints through a stable visual narrative, non-city hotels facilitated sustained interaction and mutual value reinforcement with their audience (Simmons and Durkin, 2023). In contrast, the disrupted communication pattern of city hotels potentially weakened these co-creative relationships during periods of uncertainty. These findings are particularly important in light of Šerić and Mikulić (2020) and Cervellon and Galipienzo (2015), who emphasise the importance of communication consistency for maintaining perceived brand quality and trust.

Furthermore, the resilience of non-city hotels’ communication strategies reflects a stronger alignment with audience expectations in periods of social and emotional disruption. Their increased focus on emotionally resonant and symbolic content, particularly ‘Outside Nature’, suggests an intuitive application of UGT, meeting users’ needs for escapism, aesthetic pleasure, and emotional reassurance (Deb and Lomo-David, 2020; Katz et al., 1973).

Engagement outcomes and topic performance (RQ3)

This study shows that to gain the maximum benefit of assessing the effectiveness of visual OGC (Sörensen et al., 2023), the measure of intensity of the user's involvement in the OGC co-creative process should go beyond absolute numbers of user engagement and thus introduces a relationship-based engagement performance metric, extending the work of So et al. (2021) and Lim and Rasul (2022). The results further show that different topics of visual OGC stimulate user co-creation in a different way and are dependent on the hotel location, further contributing to an enhanced understanding of SM engagement in luxury hospitality. These relationship-based, topic-focused metrics provide a scalable and transferable framework for evaluating the effectiveness of content in different settings.

Among the analysed topics, ‘Outside Nature’ and ‘Interior/Furnishings’ consistently stood out as the highest performers, particularly in non-city hotels. Both achieved high probabilities across all dimensions of engagement, including reactions, comments, and shares, and had the highest relationship strength (e.g. 72.3% reactions-to-shares for ‘Outside Nature’). These findings emphasise the value of hedonic and symbolic content, and as such, in line with the UGT (Katz et al., 1973; Shao, 2009), stipulate users are more likely to engage with content that provides emotional stimulation, escapism, and identity reinforcement (Deb and Lomo-David, 2020; Oliveira and Fernandes, 2022).

In contrast, some commonly used topics delivered mixed results. ‘Food’ performed well in terms of comments (67.1%) but failed to elicit stronger responses or shares overall. ‘Outside City’, the most frequently posted topic for city hotels, resulted in higher engagement for those hotels, while for non-city hotels, this second most commonly posted topic did not achieve high levels of any type of engagement. This inconsistency shows that the frequency of posts is no guarantee of engagement, especially when the content serves functional rather than emotional purposes; thus, it can indicate that overemphasis on utility-driven content risks reducing brand distinctiveness (Picazo et al., 2019). While such choices support contextual branding (Eijdenberg et al., 2024), they weaken the potential for co-creation, where with longer term, the value will likely be passively displayed, not actively co-produced (Neuhofer and Buhalis, 2017).

Significantly, less common topics such as ‘Drink’ and ‘Automotive/Mechanical’ revealed strong engagement potential. ‘Drink’ generated high comment-to-reaction and comment-to-share ratios in both hotel types, while ‘Automotive/Mechanical’ was effective in generating reactions-to-shares in city hotels. Although these topics are rarely posted, they have a symbolic and experiential value, which emphasises the potential of niche content with a lifestyle character, thus supporting user self-expression (Shao, 2009) within the luxury hotel segment. As such, engagement levels are not strictly correlated with posting frequency but instead with how well the content satisfies audience hedonic and symbolic needs, aligning with UGT and the luxury context of the study (De Barnier et al., 2012; Ko et al., 2019).

In alignment with UGT and S-D logic, the findings suggest that user engagement is optimised when visual content aligns with users’ emotional and symbolic needs, not merely the hotel's brand projection goals. Posts that trigger meaningful interaction, through imagery that supports lifestyle aspirations, emotional connection, or personal expression, serve as operant resources in the co-creation of value. Conversely, content with limited affective appeal, even when aligned with hotel positioning, struggles to initiate the interactive cycles that characterise deep digital engagement, aligning the result with UGT.

Implications and contributions

Theoretical contributions

First, this research contributes to the extant S-D logic and UGT literature in the context of luxury hotels. It demonstrates that the type of visual content shared (i.e. functional, hedonic, or symbolic) mediates the co-creation of the brand value on SM platforms. It highlights that not all OGC equally supports value co-creation, challenging assumptions that consistent brand communication alone fosters customer engagement, thus extending the findings of Šerić and Mikulić (2020) and Cervellon and Galipienzo (2015). Furthermore, it extends the understanding of UTG by showing that audience engagement with visual OGC is predominantly driven by symbolic and hedonic gratifications rather than informational needs. By introducing a large-scale, image-based empirical analysis, it demonstrates the centrality of emotional resonance and aspirational identity to digital interaction in luxury branding.

By focusing on relational indicators such as comments-to-reactions and reactions-to-shares, the research highlights the need to move beyond absolute engagement measures and towards a process-oriented view of value co-creation in digital environments. Thus, the paper contributes to the extant literature (e.g. Blanco-Moreno et al., 2024; Cheng et al., 2021) by introducing a novel, relationship-based engagement framework which offers a more nuanced understanding of engagement intensity, better aligning with the value co-creation principles of S-D logic.

By developing a topic-based typology tailored to the luxury sector and identifying a previously unrecognised communication theme (‘Automotive/Mechanical’), this study addresses the gap highlighted by Picazo and Moreno-Gil (2019), thus contributing to the OGC extant literature. As such, this study extends the theoretical framework for understanding luxury brand communication by including experiential lifestyle elements beyond traditional hospitality features and demonstrates that luxury hotels do not solely rely on facility-focused imagery but integrate diverse symbolic and hedonic elements to engage audiences.

From the methodological perspective, the introduction of AI-assisted visual content analysis with probabilistic topic modelling to categorise and evaluate large visual datasets advances research practice. This study provides a replicable and scalable method for evaluating visual communication performance across SM contexts, contributing to the growing methodological innovation in hospitality and marketing research.

Practical implications

From a practical perspective, this study provides luxury hotels with a set of clear, evidence-based strategies to strengthen their visual communication on SM. According to the findings, three key areas can be highlighted: (1) emphasising location-based imagery (‘Outside Nature’, ‘Outside City’) to support escapism and contextual storytelling; (2) blending hedonic and functional elements (‘Interior/Furnishing’, ‘Fashion/Decoration’) to convey emotional appeal alongside service quality; and (3) incorporating lifestyle content (‘Drink’, ‘Automotive/Mechanical’) to signal exclusivity and connect with aspirational identities. These themes should be embedded in the content calendars and revised seasonally to reflect audience expectations and platform trends.

Hotels should also go beyond traditional engagement metrics and adopt relationship-based performance indicators, such as share-to-comment ratios or thematic engagement tracking, to monitor brand resonance in a more meaningful way. City hotels in particular should make plans for the visual continuity of future disruptions to ensure a consistent presence through a reserve of emotionally supportive, evergreen content. Finally, the AI-powered visual analytics in this study demonstrates how machine learning tools can be used to review and refine brand content strategies in real time, providing a replicable model for data-driven decision-making across the industry.

Limitations and suggestions for future research

This paper identifies two key limitations: the data itself and the AI-driven methodology. Firstly, the photographs analysed were assumed to accurately represent the hotels’ projected images, which may overlook nuances introduced by AI-generated or digitally altered content. Such images might evoke different audience perceptions or engagement levels. Future research should investigate these types of visuals to better understand their impact. Secondly, the study focused exclusively on images and did not examine the influence of accompanying text, hashtags, or tags within posts. These elements may significantly affect audience engagement. Future studies should incorporate text analysis to evaluate the combined influence of visual and textual content on audience responses and engagement metrics. Thirdly, the AI software employed relies on pre-programmed algorithms, limiting control over how visual content is identified and categorised, which may affect the accuracy and consistency of label extraction. Future studies should consider using custom-designed programmes to enhance control over the extraction process. Additionally, while the Dirichlet Multinomial Mixture model provided a structured framework, different natural language processing models might have produced varied results. A comparative study of multiple models for topic generation is recommended to evaluate their effectiveness and suitability for similar analyses. Lastly, the photographs were collected from a sample of Italian luxury hotels, which may not fully capture the diversity of global audiences and hospitality contexts. Any further studies should seek to expand their focus beyond a single geographical unit to address this limitation.

Supplemental Material

sj-docx-1-jvm-10.1177_13567667251360433 - Supplemental material for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels

Supplemental material, sj-docx-1-jvm-10.1177_13567667251360433 for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels by Katarzyna Minor, Richard Wheeler, Jelena Mušanović, Jelena Dorčić and Miha Bratec in Journal of Vacation Marketing

Supplemental Material

sj-docx-2-jvm-10.1177_13567667251360433 - Supplemental material for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels

Supplemental material, sj-docx-2-jvm-10.1177_13567667251360433 for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels by Katarzyna Minor, Richard Wheeler, Jelena Mušanović, Jelena Dorčić and Miha Bratec in Journal of Vacation Marketing

Supplemental Material

sj-docx-3-jvm-10.1177_13567667251360433 - Supplemental material for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels

Supplemental material, sj-docx-3-jvm-10.1177_13567667251360433 for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels by Katarzyna Minor, Richard Wheeler, Jelena Mušanović, Jelena Dorčić and Miha Bratec in Journal of Vacation Marketing

Supplemental Material

sj-docx-4-jvm-10.1177_13567667251360433 - Supplemental material for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels

Supplemental material, sj-docx-4-jvm-10.1177_13567667251360433 for Snap, share, engage: AI-assisted temporal topic and engagement analysis of visual communication of luxury hotels by Katarzyna Minor, Richard Wheeler, Jelena Mušanović, Jelena Dorčić and Miha Bratec in Journal of Vacation Marketing

Footnotes

Acknowledgements

The authors declare the use of AI (ChatGPT-4.0) to enhance the manuscript's readability, clarity and language expression. This process was iterative and conducted under the full oversight and control of the research team. All conceptual and intellectual contributions presented in the paper are solely those of the authors.

ORCID iDs

Katarzyna Minor

Richard Wheeler

Jelena Mušanović

Jelena Dorčić

Miha Bratec

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

References

Almeida-Santana

Moreno-Gil

(2017) New trends in information search and their influence on destination loyalty: digital destinations and relationship marketing. Journal of Destination Marketing & Management 6(2): 150–161.

Balomenou

Garrod

(2019) Photographs in tourism research: prejudice, power, performance and participant-generated images. Tourism Management 70: 201–217.

Blanco-Moreno

González-Fernández

Muñoz-Gallego

, et al. (2024) What do you do or with whom? Understanding happiness with the tourism experience: an AI approach applied to Instagram. Humanities and Social Sciences Communications 11(1): 46.

Cervellon

M-C

Galipienzo

(2015) Facebook pages content, does it really matter? Consumers’ responses to luxury hotel posts with emotional and informational content. Journal of Travel & Tourism Marketing 32(4): 428–437.

Cheng

Liu

, et al. (2021) Differential effects of firm generated content on consumer digital engagement and firm performance: an outside-in perspective. Industrial Marketing Management 98: 41–58.

Daun

Klinger

(2006) Delivering the message: how premium hotel brands struggle to communicate their value proposition. International Journal of Contemporary Hospitality Management 18(3): 246–252.

Deb

Lomo-David

(2020) On the hedonic versus utilitarian message appeal in building buying intention in the luxury hotel industry. Journal of Hospitality and Tourism Management 45: 615–621.

De Barnier

Falcy

Valette-Florence

(2012) Do consumers perceive three levels of luxury? A comparison of accessible, intermediate and inaccessible luxury brands. Journal of Brand Management 19(7): 623–636.

de Oliveira Santini

Ladeira

Pinto

, et al. (2020) Customer engagement in social media: a framework and meta-analysis. Journal of the Academy of Marketing Science 48(6): 1211–1228.

10.

Eastman

Shin

Ruhland

(2020) The picture of luxury: a comprehensive examination of college student consumers’ relationship with luxury brands. Psychology & Marketing 37(1): 56–73.

11.

Eijdenberg

Thirumaran

Mohammadi

(2024) Luxury hospitality revisited: a Cambodian perspective. Journal of Hospitality and Tourism Management 58: 409–418.

12.

Eslami

Ghasemaghaei

Hassanein

(2022) Understanding consumer engagement in social media: the role of product lifecycle. Decision Support Systems 162: 113707.

13.

Ferrer-Rosell

Martin-Fuentes

Marine-Roig

(2019) Do Hotels Talk on Facebook About Themselves or About Their Destinations? Information and Communication Technologies in Tourism 2019, Cham.

14.

Ferrer-Rosell

Martin-Fuentes

Marine-Roig

(2020) Diverse and emotional: Facebook content strategies by Spanish hotels. Information Technology & Tourism 22(1): 53–74.

15.

Francesconi

(2011) Images and writing in tourist brochures. Journal of Tourism and Cultural Change 9(4): 341–356.

16.

Gálvez-Rodríguez

MdM

Alonso-Cañadas

Haro-de-Rosario

, et al. (2020) Exploring best practices for online engagement via Facebook with local destination management organisations (DMOs) in Europe: a longitudinal analysis. Tourism Management Perspectives 34: 100636.

17.

Giglio

Pantano

Bilotta

, et al. (2020) Branding luxury hotels: evidence from the analysis of consumers’ “big” visual data on TripAdvisor. Journal of Business Research 119: 495–501.

18.

Gretzel

(2016) The visual turn in social media marketing. Tourismos 12: 1–18

19.

Gretzel

Yoo

Purifoy

(2007) Online Travel Review Study: Role and Impact of Online Travel Reviews. https://www.tripadvisor.com/pdfs/OnlineTravelReviewReport.pdf.

20.

Hyun

Kang

(2014) A better investment in luxury restaurants: environmental or non-environmental cues? International Journal of Hospitality Management 39: 57–70.

21.

Jain

Wirtz

Salunke

, et al. (2023) Luxury hospitality: a systematic literature review and research agenda. International Journal of Hospitality Management 115: 103597.

22.

Katz

Blumler

Gurevitch

(1973) Uses and gratifications research. Public Opinion Quarterly 37(4): 509–523.

23.

Kim

Lee

(2019) Influence of integration on interactivity in social media luxury brand communities. Journal of Business Research 99: 422–429.

24.

Costello

Taylor

(2019) What is a luxury brand? A new definition and review of the literature. Journal of Business Research 99: 405–413.

25.

Kucukusta

Perelygina

Lam

(2019) CSR Communication strategies and stakeholder engagement of upscale hotels in social media. International Journal of Contemporary Hospitality Management 31(5): 2129–2148.

26.

Lee

Jang

Sun

, et al. (2022) Comparing themes extracted via topic modeling and manual content analysis: Korean-language discussions of dementia on twitter. Studies in Health Technology and Informatics 295: 230–233.

27.

Leung

Bai

Erdem

(2017) Hotel social media marketing: a study on message strategy and its effectiveness. Journal of Hospitality and Tourism Technology 8(2): 239–255.

28.

(2024) The effect of implied motion in travel photographs on visit intention: the mediating role of mental imagery. Tourism Management 104: 104919.

29.

Wan

(2025) Inspiring tourists’ imagination: how and when human presence in photographs enhances travel mental simulation and destination attractiveness. Tourism Management 106: 104969.

30.

Lim

Rasul

(2022) Customer engagement and social media: revisiting the past to inform the future. Journal of Business Research 148: 325–342.

31.

MacKay

Couldwell

(2004) Using visitor-employed photography to investigate destination image. Journal of Travel Research 42(4): 390–396.

32.

Manfreda

Presbury

Richardson

, et al. (2022) The building blocks of luxury accommodation experiences: a meta-ethnographic synthesis. Tourism Management Perspectives 41: 100926.

33.

Michael

Fusté-Forné

(2022) Marketing of luxurious gastronomic experiences on social media: the visual storytelling of luxury hotels. The International Journal of Tourism Research 24(6): 827–838.

34.

Michael

Fusté-Forné

(2024) In search of elite travel: visual communication of luxury hotels. Consumer Behavior in Tourism and Hospitality 19(1): 131–139.

35.

Neuhofer

Buhalis

(2017) Service-dominant logic in the social media landscape new perspectives on experience and value co-creation. In: Sigala

Gretzel

(eds) Advances in Social Media for Travel, Tourism and Hospitality New Perspectives, Practice and Cases. 1st ed. Routledge, 13–25.

36.

Oliveira

Fernandes

(2022) Luxury brands and social media: drivers and outcomes of consumer engagement on Instagram. Journal of Strategic Marketing 30(4): 389–407.

37.

Pan

Zhang

Law

(2013) The complex matter of online hotel choice. Cornell Hospitality Quarterly 54(1): 74–83.

38.

Park

Ahn

(2021) Editorial introduction: luxury services focusing on marketing and management. Journal of Retailing and Consumer Services 58: 102257.

39.

Park

Hyun

Thavisay

(2021) A study of antecedents and outcomes of social media WOM towards luxury brand purchase intention. Journal of Retailing and Consumer Services 58: 102272.

40.

Picazo

Moreno-Gil

(2019) Analysis of the projected image of tourism destinations on photographs: a literature review to prepare for the future. Journal of Vacation Marketing 25(1): 3–24.

41.

Picazo

Moreno-Gil

Gursoy

(2019) Analysis of the projected image of tourism accommodations: a methodological proposal. International Journal of Contemporary Hospitality Management 31(8): 3325–3351.

42.

Piriyakul

Kunathikornkit

Piriyakul

(2024) Evaluating brand equity in the hospitality industry: insights from customer journeys and text mining. International Journal of Information Management Data Insights 4(2): 100245.

43.

Qing

Safeer

Khan

(2024) Influence of social media communication on consumer purchase decisions: do luxury hotels value perceived brand authenticity, prestige, and familiarity? Journal of Hospitality and Tourism Technology 15(3): 465–478.

44.

Rabadán-Martín

Barcos-Redín

Pereira-Delgado

, et al. (2025) Topic-based engagement analysis: focusing on hotel industry Twitter accounts. Tourism Management 106: 104981.

45.

RivalIQ (2024a) 2024 Social media industry benchmark report: Industry benchmarks across the most important social media metrics. https://www.rivaliq.com/blog/social-media-industry-benchmark-report/.

46.

RivalIQ (2024b) Live Social Media Benchmarks for Hotels, https://www.rivaliq.com/live-benchmarks/hotels/.

47.

Ryu

Lehto

Gordon

, et al. (2018) Compelling brand storytelling for luxury hotels. International Journal of Hospitality Management 74: 22–29.

48.

Šerić

Mikulić

(2020) Building brand equity through communication consistency in luxury hotels: an impact-asymmetry analysis. Journal of Hospitality and Tourism Insights 3(4): 451–468.

49.

Shakuntala

Ramantoko

(2023) The influence of social Media communication on purchase intention and purchase decision through brand equity in mixue Indonesia. International Journal of Professional Business Review 8(11): e03935.

50.

Shao

(2009) Understanding the appeal of user-generated media: a uses and gratification perspective. Internet Research 19(1): 7–25.

51.

Simmons

Durkin

(2023) Expanding understanding of brand value co-creation on social media from an S-D logic perspective: introducing structuration theory. Marketing Theory 23(4): 607–629.

52.

KKF

Wei

Martin

(2021) Understanding customer engagement and social media activities in tourism: a latent profile analysis and cross-validation. Journal of Business Research 129: 474–483.

53.

Song

S-G

Kim

D-Y

(2016) A pictorial analysis of destination images on pinterest: the case of Tokyo, Kyoto, and Osaka, Japan. Journal of Travel & Tourism Marketing 33(5): 687–701.

54.

Sörensen

Daniel

Silke

, et al. (2023) Platforms matter: analyzing user engagement with social media content of Swiss higher education institutions. Journal of Marketing for Higher Education: 1–20.

55.

Statista (2021) Luxury Travel and Tourism in Europe Amid the Pandemic: A Statista DossierPlus on European Luxury Tourism and the Impact of COVID-19, With Future Outlooks. https://www-statista-com.ezproxy.nsk.hr/study/70914/luxury-travel-and-tourism-in-europe/.

56.

Stepchenkova

Zhan

(2013) Visual destination images of Peru: comparative content analysis of DMO and user-generated photography. Tourism Management 36: 590–601.

57.

Wang

Luo

Huang

(2020) Developing an artificial intelligence framework for online destination image photos identification. Journal of Destination Marketing & Management 18: 100512.

58.

Egger

(2021) Color and engagement in touristic Instagram pictures: a machine learning approach. Annals of Tourism Research 89: 103204.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB

1.78 MB

0.52 MB

0.61 MB