A cross-cultural analysis of perceived value and customer loyalty in restaurants

Abstract

This paper explores the relationships between the social, emotional, functional, epistemic and conditional value perceived by customers and the key customer outcomes of satisfaction, retention and loyalty in the restaurant industry. Partial least squares structural equation modelling (PLS-SEM) is applied to a cross-sectional sample of 365 restaurant clients from Romania, Italy and Croatia. The analysis reveals significant cross-cultural variations in how these five kinds of value affect customer satisfaction. The findings reveal that emotional value is the most significant determinant of customer satisfaction in all three cultural contexts. This finding underscores its universal importance in improving customer experiences. In contrast, functional value and conditional value have context-dependent effects, with greater relevance in certain countries. In particular, social value negatively influences satisfaction in Italy, suggesting that social aspects may not be aligned with consumer expectations across cultures. These findings provide actionable insights for restaurant managers. They highlight the importance of boosting emotional engagement and tailoring service strategies to culturally specific customer preferences. Doing so can ultimately lead to greater customer satisfaction, retention and loyalty.

1

Introduction

A contented consumer is likely to share a favourable encounter with three individuals, whereas a dissatisfied customer is likely to share a negative experience with 10 other people. Therefore, firms must take the utmost care with customer satisfaction and loyalty given that they can lead to positive outcomes (Hill & Alexander, 2017). Dissatisfied customers who express unhappiness with encounters with a company are unlikely to return. They may also discourage potential customers from doing business with the company (Hill & Alexander, 2017). In contrast, loyal customers show a predilection for a favoured company over its rivals (Larson & McClellan, 2017). According to Hill and Alexander (2017), a company increases its profitability by maintaining its customer base. The main goal of all business actions should be to provide customer value (Panda, 2009).

Customers seek diverse preferences and advantages, as documented by research. The literature identifies five main kinds of perceived product or service value for customers: social, emotional, functional, epistemic and conditional value (Gatautis et al., 2021; Hodson, 2021; Iacono et al., 2024; Jabreel et al., 2017; Mulyana & Limakrisna, 2023; Pagani, 2009; Panda, 2009; Shi, 2022; Yusoff, 2023; Zallio, 2023). Social value is associated with the social standing and worth attributed to the use of a product or service. Emotional value provides gratification and delight. Functional value refers to the specific features of a product or service that provide usefulness, convenience and reliability. Epistemic value is perceived through research into and adoption of a new product or service (Artun & Levin, 2015). Finally, conditional value depends on external events or scenarios.

From the perspective of organisations, Farris et al. (2015) noted that they should prioritise two key metrics: the customer retention rate and the customer attrition rate. The customer retention rate measures the number of loyal consumers. The customer attrition rate measures the number of lost customers. To maintain an acceptable customer retention rate, organisations must keep valuable clients and regain profitable customers who have departed (Reichheld & Teal, 1996; Stauss & Friege, 1999). Customer loyalty is mostly shaped by favourable previous encounters. Loyal customers have a sense of affinity towards a favoured product or brand (Lim & Rasul, 2022). According to Burrow (2012), several aspects influence customer loyalty. Competitive pricing, high-quality items, helpful staff, great services, convenient location, shared corporate values and beliefs, a pleasant company or store atmosphere, attractive websites and knowledgeable product advocates all play a role.

Although studies have examined customer satisfaction and loyalty in various service industries, there is no comprehensive analysis of the combined effects of these five kinds of perceived value in the specific context of restaurants. Customer satisfaction and loyalty in the restaurant industry have been underexplored from a cross-cultural perspective. It is true that the cross-cultural framework has been applied to measure tourists’ customer satisfaction in local restaurants, with restaurants and food considered to be part of the tourism destination (Badu-Baiden et al., 2022a, 2022b). Scholars have likewise evaluated satisfaction and loyalty in fine dining (Tsaur & Lo, 2020). Nonetheless, further study in this area is needed.

This study aims to fill the first gap by investigating the role of each kind of perceived value in shaping customer satisfaction and loyalty in the restaurant industry. To fill the second gap, an explicit cross-cultural comparison in the restaurant industry of three European countries (Romania, Italy and Croatia) was performed. These countries were chosen to represent Eastern Europe, Western Europe and Central Europe, respectively. Hence, they were considered representative of three diverse European cultural backgrounds.

Accordingly, the primary objective of this study is to examine the impact of perceived value on loyalty, with satisfaction and retention acting as mediators. This study is unusual because it examines five different kinds of perceived value and their effects on satisfaction, retention and loyalty. The study also aims to identify the most important kind of value in terms of its influence on satisfaction.

The theoretical contributions of the study relate to enriching the existing body of knowledge on customer satisfaction, retention and loyalty by suggesting practical ways to improve retention and boost loyalty. From a managerial perspective, the findings highlight the importance for restaurant managers of tailoring customer relationship management (CRM) strategies based on the specific kinds of value that resonate with their target market. These contributions can ultimately lead to improved restaurant performance through stronger customer connections.

The paper is structured as follows. Section 2 provides a literature review and outlines the research hypotheses. A model is then developed based on the proposal for analysis. Section 3 describes the research design, including data collection, the study's objectives and the analysis method. Section 4 presents the results. Section 5 then provides a discussion of these results, along with recommendations and conclusions. This section also addresses the limitations of the study, theoretical and practical contributions to customers, managers, companies and society, and future research directions.

2

Literature review, research hypotheses and conceptual model

2.1

Customer perceived value and customer satisfaction

Many studies have explored how specific kinds of value influence customer perceived quality to enhance the sustainability and performance of business models. Diverse research on the identification of perceived value in services has covered quality service in health care (Mohammed & Mahmood, 2022), social responsibility in retailing (Hoang & Phuong, 2016; Lee et al., 2020), switching costs (Ram & Wu, 2016), cultural differences (Fam et al., 2023) and customer commitment (Hur et al., 2010). Service studies have examined various commercial sectors, including libraries (Laukkanen & Tura, 2022), fitness (Sevilmi et al., 2022), retail (Xu & Hu, 2022), live streaming (Qin et al., 2023), artificial intelligence (Hlee et al., 2023), food delivery (Hsu et al., 2023; Lee & Han, 2022), recycling (Sener et al., 2022), tourism (Madinga et al., 2023; Zhou & Yu, 2022), social media (Doshi et al., 2023; Kim et al., 2023), music festivals (Alen-Gonzalez et al., 2023), banking (Abbass et al., 2023), e-commerce (Ahn & Kwon, 2022; Jin et al., 2022), cryptocurrencies (Erdmann et al., 2023), post-consumption (Jin et al., 2023), finance (Riahi & Garrouch, 2023) and hotels (Ghorbani et al., 2023; Rasoolimanesh et al., 2023).

Social, emotional, functional, epistemic and conditional kinds of value all influence customer perceived quality (Gatautis et al., 2021; Hodson, 2021; Jabreel et al., 2017; Mulyana & Limakrisna, 2023; Pagani, 2009; Panda, 2009; Shi, 2022; Yusoff, 2023; Zallio, 2023). Perceived social value relates to the social status and value associated with the consumption of a product or service. Perceived emotional value refers to the pleasure and enjoyment offered by the product or service. Perceived functional value relates to the tangible attributes of a product or service that offer utility, convenience and reliability. Epistemic value is realised through the exploration and adoption of innovative products (Sheth et al., 1991). Finally, perceived conditional value depends on specific circumstances or situations.

This study specifically targets the restaurant industry. In this context, although all kinds of perceived value contribute to customer satisfaction (Wu & Mursid, 2019), perceived social value seems to stand out. Restaurants operate in a unique market and economic setting, where the social dimension is particularly influential. Social value may play a crucial role in achieving consumer satisfaction in the context of restaurants. In addition, restaurant customers frequently develop emotional attachment, investing their trust in both the restaurant and its staff. This attachment leads to a strong commitment and long-term devotion (Morgan et al., 2015). Studies have investigated this phenomenon by focusing on the conversations, relationships and interactions that occur in restaurants (Rasoolimanesh et al., 2023). They have also examined the social bonds created in restaurants that improve customer experience and foster a desire to return (Mirzaei & Ozturk, 2018). Scholars have studied the role of restaurants serving as platforms for communication, relationship building, social connections, self-expression and interaction, all contributing to social outcomes (Jiang & Lau, 2022). Research has considered restaurants as being symbolic spaces where individuals shape their self-image (Tsaur et al., 2023) or enjoy positive experiences such as resolving uncertainties through rewards (Shen et al., 2019). Finally, restaurants have been linked to positive emotions (Laran & Tsiros, 2013), a sense of curiosity and interest (Ruan et al., 2018), a feeling of social gain (Zhang & Zhang, 2022) and a culture of sharing (Wang et al., 2018).

Customers perceive value when a product or service meets their needs or desires. Satisfaction marks the end of the value chain Panda (2009). Customer value is a subjective assessment made after purchase and consumption. Customer satisfaction is an evaluation made post-purchase based on pre-purchase expectations (Sun, 2009). Therefore, satisfaction can be specific to each transaction as a post-choice evaluation after each purchase occasion or cumulative as an overall evaluation based on all purchases (Wang & Li-Hua, 2007). Customer value is crucial in determining the longevity of relationships and satisfaction, and it must be distinctive (Dovaliene & Virvilaite, 2008).

The relationship between customer value and satisfaction has been analysed by various authors and applied in different fields. For example, El-Adly (2019) studied hotel services and value in the form of prestige, self-gratification, aesthetics, price, hedonism, quality and transaction experience. Bouchriha et al. (2023) studied interaction and engagement. Kokkhangplu et al. (2023) examined eco-friendly practices and tourist satisfaction. In retail, scholars have studied value in the form of price, product and trade value (Ho & Shieh, 2010). In homestay experiences, scholars have examined functional, social and emotional value (Lu & Yi, 2022). Kusumawati and Rahayu (2020) studied outdoor cafes that emphasise quality as a value. Slack et al. (2020) analysed supermarkets with social, economic and emotional value. Scholars have explored the banking sector, focusing on empathy, price and competence as kinds of value (Mainardes & De Freitas, 2023). Finally, financial services have also been studied, considering quality, cost and engagement as kinds of value.

This study builds on the theory of total customer value (Kotler, 2010). Kotler defined total customer value as the benefits customers obtain from using a product or service after subtracting the total customer cost, which includes expenses from evaluation and the process of acquiring desired products and services. More specifically, perceived value is the total customer value minus the total customer cost. Based on a review of the aforementioned literature, the first research hypothesis is proposed:

• H1. Customer perceived value significantly influences customer satisfaction.

To test this hypothesis, the broad concept of customer perceived value was divided into five specific kinds of perceived value to formulate five sub-hypotheses. First, social value is related to the social status and value associated with product or service consumption. When customers perceive social benefits from a restaurant's value proposition such as a sense of belonging or enhanced social status, it could positively affect their overall satisfaction (Mulyana & Limakrisna, 2023). Thus, the following sub-hypothesis is proposed:

• H1a. Perceived social value significantly influences customer satisfaction.

The emotions that customers experience such as enjoyment and comfort are highly valuable in shaping their satisfaction (Shi, 2022). If a restaurant provides a pleasant atmosphere, friendly service or memorable experiences, customers are more likely to feel emotionally satisfied, leading to a higher level of overall satisfaction (Tsaur & Lo, 2020). Thus, the following sub-hypothesis is proposed:

• H1b. Perceived emotional value significantly influences customer satisfaction.

Functional value refers to the characteristics of a product such as usefulness, convenience and reliability (Jabreel et al., 2017). For example, a restaurant that offers high-quality food, timely service or a convenient location can enhance customers’ perceived functional value, thus improving satisfaction. The following sub-hypothesis is proposed:

• H1c. Perceived functional value significantly influences customer satisfaction.

Epistemic value is associated with the curiosity and novelty related to a restaurant. Restaurants can enhance customer satisfaction by providing new, unique or innovative experiences. In the context of restaurants, introducing new dishes or offering unique dining experiences are ways of stimulating customers’ interest, leading to increased satisfaction (Gatautis et al., 2021). Thus, the following sub-hypothesis is proposed:

• H1d. Perceived epistemic value significantly influences customer satisfaction.

Conditional value refers to the situational factors or specific circumstances that can affect customer satisfaction. This kind of value suggests that certain conditions such as seasonal promotions, special events or catering to particular occasions can significantly influence how satisfied customers feel (Hodson, 2021). For example, a restaurant offering a customised experience for a celebration could enhance a customer's satisfaction by creating those specific conditions. The following sub-hypothesis is proposed:

• H1e. Perceived conditional value significantly influences customer satisfaction.

2.2

Customer satisfaction and customer loyalty

Managing customer satisfaction is crucial because it affects long-term firm performance (Panda, 2009). Sun (2009) described customer satisfaction as a precursor to customer loyalty and a vital factor in an organisation's success (Fourati & Kammoun, 2012). Customers’ value creation contributes to this success (Srivastava et al., 2024) and potentially turns customers into value co-creators (Yu et al., 2024). The higher the level of customer satisfaction is, the greater the profit for the organisation will be (Hassan et al., 2015). Companies aim to enhance satisfaction levels to encourage repeat business from customers (Craven, 2012). A survey of managers revealed that failing to satisfy and retain customers is perceived as the biggest risk to organisational performance (Sadgrove, 2016). Furthermore, research indicates that a highly satisfied customer is six times more likely to repurchase from a company, implying that satisfaction raises the probability of loyalty (Kenett & Salini, 2011).

Whereas customer satisfaction is generally seen as a measure of how well a company's products or services meet or exceed customer expectations, customer loyalty refers to the willingness of customers to return repeatedly to a company for business due to the positive experiences and value they receive from the relationship (Utami et al., 2023). Research consistently links customer satisfaction to customer loyalty, indicating that satisfaction is a precursor to loyalty. This link is commonly viewed as positive, with higher satisfaction levels leading to greater customer loyalty (Bae et al., 2016; Mishra, 2022; Sharma et al., 2020).

However, this link is not without its complexities. For example, Bae et al. (2016) noted that the strength of the satisfaction–loyalty link may differ based on geographical locations and product types, emphasising the importance of context. Aityassine (2022) suggested that customer satisfaction affects loyalty both directly and indirectly, with customer retention serving as a mediator. Helgesen (2006) and Tu and Chang (2011) proposed that the link between satisfaction and loyalty may not always be linear and may be subject to thresholds, suggesting that satisfaction must reach a certain level to affect loyalty in a significant way. In essence, the literature widely supports a positive association between customer satisfaction and loyalty, with contented customers typically showing higher loyalty. However, various factors such as geographical variations, product types and the presence of mediators such as customer retention and corporate communication may influence this relationship. Studies suggest the existence of nonlinearity and threshold effects. Nevertheless, the aforementioned evidence leads to the proposal of the second research hypothesis:

• H2. Customer satisfaction significantly influences customer loyalty.

2.3

Customer retention, customer satisfaction and customer loyalty

In contemporary marketing strategy, the main emphasis has shifted from customer acquisition to customer retention (Panda, 2009). Ensuring customer satisfaction and retention is of utmost importance to maintaining profitability. Research indicates that retaining a profitable customer is significantly more cost-effective than acquiring a new one, with estimates indicating that it can be 5 to 25 times more profitable (Abeza et al., 2021; Ibrahim, 2022; Kenett & Salini, 2011; Miller & Miller, 2008; Nicolescu & Lloyd-Reason, 2016; Salampasis & Mention, 2022; Schulz, 2009; Wilson et al., 2012). Reactivating an old customer is also 10 times cheaper than acquiring a new one. In essence, customer retention is more economical than customer reactivation (Artun & Levin, 2015), making customer retention strategies essential for companies to achieve profitability. Numerous studies have shown that in service-oriented industries, customer satisfaction positively affects customer retention. Examples of such industries include banking (Darzi & Bhat, 2018), social activities (Guo et al., 2009), retail and commerce (Ahsan et al., 2022; Lee & Hsu, 2019), e-commerce (Vakulenko et al., 2022), catering (Chueh et al., 2014) and telecommunications (Fourati & Kammoun, 2012).

Alongside customer retention, customer loyalty also plays a crucial role in firms’ marketing efforts to achieve sustainable success. Customer loyalty entails consistently purchasing from the same producer or brand, continuing to buy even if prices increase, influencing others to become customers, and remaining loyal even in the face of mistakes or a lack of innovation compared to competitors. Therefore, firms should cultivate customer loyalty by establishing a bond of friendship and partnership with customers, especially during challenging times (Oliver, 1999). On average, loyal customers spend five to six times more with a company than other customers. Hence, they are valuable assets for businesses focused on strong relationships and effective CRM strategies. Building and maintaining customer loyalty is central to a company's success in traditional brick-and-mortar stores and online marketplaces alike (Reichheld & Schefter, 2000). The correlation between customer retention and loyalty has been extensively studied in various contexts, including retail and e-commerce. Researchers have described this relationship as a dynamic process influenced by factors such as lifetime value, sales strategies, pricing, branding, organisational identification, emotional engagement and customer experience (Blattberg et al., 2001).

Customer loyalty does not solely relate to price reductions. It is also closely linked to establishing a strong relationship through specific strategies involving CRM, data simulation and modelling, and market analysis and research (Tahal & Stritesky, 2013). For example, these strategies can help distinguish between customers who are members and those who are not (Lee & Hsu, 2019). Therefore, exploring how retention mediates the relationship between satisfaction and loyalty can yield interesting findings. Scholars have extensively studied this mediation in various service sectors such as retail (Albarq, 2023), mobile phones (Kaur & Soch, 2018), sports (Min, 2022) and health care (Fatima et al., 2018), indicating that high satisfaction levels lead to increased retention and loyalty (Tavakoli et al., 2015). As a result, the third research hypothesis is proposed:

• H3. Customer retention mediates the relationship between customer satisfaction and customer loyalty.

Drawing on existing literature on customer perception, behaviour, retention and loyalty (Bouchriha et al., 2023; El-Adly, 2019; Kokkhangplu et al., 2023; Kusumawati & Rahayu, 2020; Mainardes & de Freitas, 2023; Slack et al., 2020), a novel conceptual model is proposed based on five kinds of perceived value. This model presents a fresh theoretical perspective in CRM (Fig. 1).

Fig. 1.

Conceptual model.

Source: Authors.

In this model, the five kinds of customer perceived value (social, emotional, epistemic, functional and conditional) directly affect customer satisfaction. Furthermore, customer satisfaction influences customer loyalty. Finally, customer retention mediates the relationship between satisfaction and loyalty.

3

Research methodology

3.1

Objectives

This study aims to establish the nature of the relationships between five kinds of perceived value and the customer outcomes of satisfaction, retention and loyalty. Hence, partial least squares structural equation modelling (PLS-SEM) is an ideal methodological choice because it focuses on predicting and understanding complex relationships. CRM has the power to enhance company performance and create a competitive advantage through value (Kim et al., 2012). Customers rely on service value when making decisions. It is a dominant indicator, especially in the context of restaurant services and in reference to repurchase intention (Doeim et al., 2022).

Five key kinds of perceived value are included this study. These distinct, independent (Panda, 2009) yet interrelated kinds of customer perceived value are functional, social, conditional, epistemic and emotional value. They play a crucial role in the selection of a product, service or company. These kinds of value enhance overall value when combined, creating a value constellation (Panda, 2009). Customer satisfaction does not always equate to customer loyalty. Even if customers are satisfied, they may still choose to leave the company. The company should ensure that customers are not only satisfied but also perceive enough value to remain loyal (Lam, 2003).

3.2

Research design

The study was conducted in three European countries: Romania, Italy and Croatia. Convenience sampling was used. A questionnaire was distributed to friends, friends of friends, acquaintances and colleagues from various towns in these countries.

Romania, Italy and Croatia were chosen to represent different economic and cultural regions of Europe. Romania (Eastern Europe), Italy (Western Europe) and Croatia (Central Europe) provided a diverse frame of reference for the study of consumer behaviour in the restaurant industry. These three countries were selected not only because of the major contribution of the restaurant industry to the local economy but also because they enabled exploration of how distinct cultural and economic values influence customer satisfaction and loyalty. Their selection ensured a balanced view of countries from Eastern, Central and Western Europe, allowing for a better understanding of the differences and similarities in consumer behaviour across these regions.

The restaurant industry plays an essential role in the economy of each of these countries, contributing greatly to the service and tourism sectors. For example, in Romania and Croatia, tourism is an important part of the economy. Restaurants are central to the tourist experience. Meanwhile, Italy is renowned for its culinary culture. Its restaurants contribute in a major way to the local economy by providing not only food services but also a unique cultural experience.

Thus, intercultural analysis enabled comparison of the value perceived by customers in the restaurant industry in the specific economic and cultural context of each country. This intercultural comparative framework not only offers a new perspective on consumer behaviour but also helps identify broadly applicable management strategies that can be adapted to enhance customer satisfaction and loyalty in different cultural contexts.

In Romania, most respondents were from Targoviste (68.2 %), Bucharest (9.8 %), Arad (7.3 %), Ploiesti (5.7 %) and other towns such as Alexandria, Brasov, Breaza, Buzau, Campina, Gaesti, Moreni, Sebes and Timisoara (all below 3 %). In Italy, most participants were from Rome (15.8 %), Milan (11.4 %), Genoa (12.5 %), Ivrea (3.5 %) and other towns such as Ancona, Bergamo, Brescia, Como, Florence, Padua, Palermo and Venice (all below 3 %). In Croatia, most respondents were from Rijeka (68.8 %) and Zagreb (11.7 %), with other towns such as Kackovec, Karlovac, Opatija, Osijek, Pula, Rovinj, Vrsar and Zadar each accounting for <3 % of responses. The questionnaire was developed based on scales from the literature (Appendix A). A five-point Likert scale was used, ranging from 1 (total disagreement) to 5 (total agreement). Four items were used for each variable.

3.3

Sampling procedure

Self-administered questionnaire data were collected between January and April 2024 from an online survey. Ethical principles were observed, ensuring data confidentiality and respecting the privacy of all respondents. Participants were 123 respondents from Romania, 128 from Croatia and 114 from Italy. The requirement for inclusion in the study was that each participant had dinner in a restaurant at least three times a year.

Table 1 displays the sociodemographic characteristics of the participants. Most customers in Romania fell into the 18-to-30-year age group (64.2 %). In Croatia, 50 % were in this age group. In Italy, most belonged to the over 49-year age group (65.8 %). Most respondents were female: 59.3 % in Romania, 57.9 % in Italy and 76.6 % in Croatia. In terms of education level, most respondents held a bachelor's degree (56.9 %) in Romania, whereas in Italy and Croatia, 49.1 % and 41.4 %, respectively, had a master's degree or higher. Most respondents resided in urban areas: 52.8 % in Romania, 84.1 % in Italy and 81.2 % in Croatia.

Table 1.

Sociodemographic profile of survey participants.

Profile/Country	Romania (N = 123)	Italy (N = 114)	Croatia (N = 128)
Age
18–30 years	79 (64.2 %)	25 (21.9 %)	64 (50 %)
31–45 years	31 (25.2 %)	14 (12.3 %)	57 (44.5 %)
46–59 years	13 (10.6 %)	75 (65.8 %)	7 (5.5 %)
Gender
Male	50 (40.7 %)	48 (42.1 %)	30 (23.4 %)
Female	73 (59.3 %)	66 (57.9 %)	98 (76.6 %)
Education level
Less than bachelor's	32 (26.0 %)	18 (15.8 %)	46 (35.9 %)
Bachelor's	70 (56.9 %)	40 (35.1 %)	29 (22.7 %)
Master's and above	21 (17.1 %)	56 (49.1 %)	53 (41.4 %)
Residence
Urban	65 (52.8 %)	96 (84.2 %)	104 (81.2 %)
Rural	58 (47.2 %)	18 (15.8 %)	24 (18.8 %)

Harman's single factor test based on principal component analysis (PCA) as the extraction method was performed using SPSS software. After extracting one factor, the total variance explained by this single factor was checked (Appendix B). This single factor explained 46.02 % of the variance, which was less than the 60 % threshold for total variance. The results thus indicate the absence of a potential issue with common method bias (Podsakoff et al., 2003).

4

Results

The model was estimated using PLS-SEM in SmartPLS 4.0. Analysis using PLS-SEM is effective for complex models with multiple constructs and paths. Hence, the approach was well aligned with the objectives of this study. PLS-SEM requires a smaller sample size than covariance-based structural equation modelling (CB-SEM). This study had a relatively small sample from three target countries of Romania, Italy and Croatia.

The three countries were compared using multi-group analysis. For hypothesis testing, the analysis was conducted using 5000 subsamples. This approach gave robust estimates of the sampling distribution, providing reliable inference for hypothesis testing.

For the Romanian sample, Fig. 2 highlights the path coefficients and outer loadings of the constructs. Social value had a statistically nonsignificant influence on customer satisfaction, with a path coefficient of −0.006. Emotional value had a moderate positive impact on customer satisfaction, with a path coefficient of 0.389. Functional value also had a moderate influence on customer satisfaction, with a path coefficient of 0.206. Epistemic value had a small yet positive effect on customer satisfaction, with a path coefficient of 0.246. Conditional value had a moderate influence on customer satisfaction, with a path coefficient of 0.117. Customer satisfaction had a positive impact on customer loyalty, with a path coefficient of 0.513. Customer satisfaction also strongly influenced customer retention, with a path coefficient of 0.718. Customer retention moderately affected customer loyalty, with a path coefficient of 0.361.

Fig. 2.

Path coefficients and outer loadings of the constructs in the Romanian sample.

Source: SmartPLS4 software.

High loadings (greater than 0.7) suggest that the indicators are relevant measures of the latent constructs. All loadings in the structural model exceeded the threshold of 0.7. The five kinds of perceived value accounted for 64.1 % of the variance in customer satisfaction, indicating the model's high explanatory power for customer satisfaction in the Romanian sample. Customer satisfaction explained 51.6 % of the variance in customer retention, indicating a robust relationship between customer satisfaction and customer retention. Furthermore, customer satisfaction and customer retention explained 65.9 % of the variance in customer loyalty, highlighting the model's high explanatory power for customer loyalty.

All constructs (customer loyalty, retention and satisfaction; conditional, emotional, epistemic, functional and social value) had Cronbach's alpha values greater than 0.7, indicating sufficient internal consistency. Composite reliability values (rho_a and rho_c) surpassed the 0.7 threshold, reflecting the constructs’ reliability. The standardised root mean square residual (SRMR) was 0.089, indicating an acceptable goodness of fit of the model (Henseler et al., 2016). Table 2 shows these results.

Table 2.

Construct reliability, convergent validity and goodness of fit for the Romanian sample.

	Cronbach's alpha	Composite reliability (rho_a)	Composite reliability (rho_c)	Average variance extracted (AVE)
CL	0.900	0.908	0.930	0.769
CR	0.909	0.912	0.936	0.785
CS	0.872	0.886	0.913	0.726
CV	0.925	0.930	0.947	0.817
EMV	0.921	0.927	0.944	0.809
EPV	0.895	0.904	0.927	0.761
FV	0.895	0.903	0.927	0.762
SV	0.890	0.903	0.923	0.750
Standardized root mean square residual (SRMR) = 0.089

Source: SmartPLS4 software.

Average variance extracted (AVE) values for all constructs were greater than 0.5, indicating convergent validity. Discriminant validity was evaluated using the heterotrait-monotrait ratio (HTMT). All HTMT values were below the 0.9 threshold, confirming discriminant validity among constructs in the Romanian sample. The analysis showed that HTMT values for all construct pairs were below the threshold of 0.85, indicating adequate discriminant validity, even among closely related constructs. For instance, although perceived emotional value and perceived social value are conceptually interrelated, their HTMT ratio was below the acceptable limit, suggesting that these constructs are statistically distinct. Table 3 shows these results.

Table 3.

Discriminant validity (HTMT for the Romanian sample).

	CL	CR	CS	CV	EMV	EPV	FV	SV
CL
CR	0.800
CS	0.868	0.799
CV	0.663	0.654	0.672
EMV	0.743	0.64	0.777	0.521
EPV	0.677	0.77	0.757	0.801	0.624
FV	0.659	0.664	0.730	0.644	0.641	0.675
SV	0.631	0.539	0.595	0.636	0.668	0.626	0.534

Source: SmartPLS4 software.

The results in Table 4 show that emotional value, epistemic value and functional value led to significantly higher customer satisfaction, based on the t statistics and p values for the relationships: emotional value → customer satisfaction (t statistic = 4.795, p value = 0.003); epistemic value → customer satisfaction (t statistic = 2.633, p value = 0.008); functional value → customer satisfaction (t statistic = 2.301, p value = 0.021). Therefore, hypotheses H1b, H1d and H1c are supported. Social value and conditional value did not significantly affect customer satisfaction: social value → customer satisfaction (t statistic = 0.064, p value = 0.949); conditional value → customer satisfaction (t statistic = 1.171, p value = 0.242). Therefore, these kinds of perceived value were not found to be significant factors in the context of restaurant services in Romania. Hence, hypotheses H1a and H1e are rejected.

Table 4.

Hypothesis testing results for the Romanian sample.

Hypothesis	Original sample (O)	Sample mean (M)	Standard deviation (STDEV)	t statistic (|O/STDEV|)	p value	Decision
CR -> CL	0.361	0.367	0.096	3.739	0.003	Supported
CS -> CL	0.513	0.508	0.098	5.238	0.002	Supported
CS -> CR	0.718	0.719	0.056	12.838	0.001	Supported
CV -> CS	0.117	0.106	0.100	1.171	0.242	Rejected
EMV -> CS	0.389	0.389	0.081	4.795	0.003	Supported
EPV -> CS	0.246	0.256	0.094	2.633	0.008	Supported
FV -> CS	0.206	0.207	0.090	2.301	0.021	Supported
SV -> CS	−0.006	−0.003	0.093	0.064	0.949	Rejected

Source: SmartPLS4 software.

The results in Table 4 also show that customer satisfaction strongly influenced customer loyalty, with higher satisfaction levels leading to increased loyalty among customers: customer satisfaction → customer loyalty (t statistic = 5.238, p value = 0.002). Therefore, hypothesis H2 is supported. Customer retention effectively mediated the link between customer satisfaction and customer loyalty: customer retention → customer loyalty (t statistic = 3.739, p value = 0.003). This result underscores the importance of retaining satisfied customers to bolster loyalty. Consequently, H3 is also supported.

Based on these results, Romanian restaurant managers should prioritise enhancing the emotional, epistemic and functional aspects of their services to boost customer satisfaction. They should develop strategies for retaining satisfied customers because it significantly enhances loyalty. Given that social value and conditional value did not significantly affect satisfaction, focusing on these areas could help create more effective CRM strategies.

For the Italian sample, Fig. 3 shows that social value had a negative and statistically significant impact on customer satisfaction, with a path coefficient of −0.208. Emotional value had a significant strong and positive impact on customer satisfaction, with a path coefficient of 0.526. Functional value had a significant moderate and positive impact on customer satisfaction, with a path coefficient of 0.240. Epistemic value had a significant small and positive impact on customer satisfaction, with a path coefficient of 0.200. Conditional value had a nonsignificant impact on customer satisfaction, with a path coefficient of 0.050. Customer satisfaction had a significant moderate and positive impact on customer loyalty, with a path coefficient of 0.336. Customer satisfaction had a significant strong and positive impact on customer retention, with a path coefficient of 0.589. Customer retention had a significant strong and positive impact on customer loyalty, with a path coefficient of 0.527.

Fig. 3.

Path coefficients and outer loadings of the constructs in the Italian sample.

Source: SmartPLS4 software.

The five kinds of perceived value explained 57.2 % of the variance in customer satisfaction, showing the model's substantial explanatory power for customer satisfaction. Additionally, customer satisfaction explained 34.7 % of the variance in customer retention, indicating a moderate relationship between customer satisfaction and customer retention. Furthermore, customer satisfaction and customer retention explained 60.0 % of the variance in customer loyalty, indicating the model's substantial explanatory power for customer loyalty.

All constructs had Cronbach's alpha values greater than 0.7, indicating good internal consistency in the Italian data sample. Their composite reliability values (rho_a and rho_c) were also greater than 0.7, further confirming construct reliability. Moreover, all constructs had AVE values greater than 0.5, showing good convergent validity. The SRMR was 0.084, indicating an acceptable goodness of fit of the model. The results appear in Table 5.

Table 5.

Construct reliability, convergent validity and goodness of fit for the Italian sample.

	Cronbach's alpha	Composite reliability (rho_a)	Composite reliability (rho_c)	Average variance extracted (AVE)
CL	0.843	0.859	0.895	0.682
CR	0.864	0.865	0.908	0.710
CS	0.866	0.869	0.909	0.714
CV	0.811	0.821	0.876	0.639
EMV	0.888	0.905	0.923	0.749
EPV	0.870	0.871	0.912	0.722
FV	0.772	0.794	0.854	0.596
SV	0.783	0.825	0.858	0.604
Standardized root mean square residual (SRMR) = 0.084

Source: SmartPLS4 software.

HTMT values below 0.9 confirmed that each construct was distinct from the others. The results in Table 6 thus support discriminant validity in the Italian sample.

Table 6.

Discriminant validity (HTMT for the Italian sample).

	CL	CR	CS	CV	EMV	EPV	FV	SV
CL
CR	0.838
CS	0.755	0.676
CV	0.673	0.709	0.665
EMV	0.610	0.583	0.751	0.676
EPV	0.589	0.665	0.527	0.701	0.585
FV	0.777	0.741	0.712	0.72	0.605	0.604
SV	0.374	0.443	0.335	0.675	0.686	0.616	0.359

Source: SmartPLS4 software.

The results in Table 7 show that emotional value, functional value and conditional value significantly enhanced customer satisfaction in the Italian sample: emotional value → customer satisfaction (t statistic = 5.677, p value = 0.002); functional value → customer satisfaction (t statistic = 3.065, p value = 0.002); conditional value → customer satisfaction (t statistic = 2.125, p value = 0.034). These results support H1b, H1c and H1e. Social value had a negative impact on customer satisfaction: social value → customer satisfaction (path coefficient = −0.208, t -statistic = 2.337, p value = 0.019). Hence, the results support H1a. However, this result is not as expected and may warrant further investigation. Epistemic value did not significantly affect customer satisfaction: epistemic value → customer satisfaction (t statistic = 0.566, p value = 0.571). Therefore, H1d is not supported.

Table 7.

Hypothesis testing results for the Italian sample.

Hypothesis	Original sample (O)	Sample mean (M)	Standard deviation (STDEV)	t statistics (|O/STDEV|)	p value	Decision
CR -> CL	0.527	0.530	0.101	5.217	0.002	Supported
CS -> CL	0.336	0.336	0.090	3.721	0.001	Supported
CS -> CR	0.589	0.591	0.063	9.381	0.003	Supported
CV -> CS	0.200	0.202	0.094	2.125	0.034	Supported
EMV -> CS	0.526	0.512	0.093	5.677	0.002	Supported
EPV -> CS	0.050	0.044	0.088	0.566	0.571	Rejected
FV -> CS	0.240	0.248	0.078	3.065	0.002	Supported
SV -> CS	−0.208	−0.183	0.089	2.337	0.019	Supported

Source: SmartPLS4 software.

The results in Table 7 also indicate that customer satisfaction significantly and positively influenced customer loyalty, suggesting that higher satisfaction results in increased loyalty of customers. Hence, H2 is confirmed. Customer retention significantly mediated the relationship between customer satisfaction and customer loyalty, providing support for H3.

The negative influence of social value on satisfaction should be addressed by Italian restaurant managers. One potential way of doing so would be to align the social aspects of the service more closely with customer expectations. Given the nonsignificant impact of epistemic value, its relevance in restaurant services may require reassessment as well.

For the Croatian sample, Fig. 4 shows that social value had a minimal and nonsignificant impact on customer satisfaction, with a path coefficient of 0.031. Emotional value had a significant strong and positive effect on customer satisfaction, with a path coefficient of 0.372. Functional value had a modest but positive impact on customer satisfaction, with a path coefficient of 0.133. Epistemic value had a statistically nonsignificant effect on customer satisfaction, with a path coefficient of 0.072. Conditional value had a moderate and positive impact on customer satisfaction, with a path coefficient of 0.253.

Fig. 4.

Path coefficients and outer loadings of the constructs in the Italian sample.

Source: SmartPLS4 software.

The five kinds of perceived value explained 46.3 % of the variance in customer satisfaction, indicating the model's moderate explanatory power for customer satisfaction. Additionally, customer satisfaction accounted for 41.5 % of the variance in customer retention, implying a strong relationship between customer satisfaction and customer retention. Furthermore, customer satisfaction and customer retention explained 62.4 % of the variance in customer loyalty, indicating the model's substantial explanatory power for customer loyalty in the Croatian sample.

For the Croatian sample (Table 8), all constructs had Cronbach's alpha values greater than 0.7, indicating strong internal consistency. The composite reliability values (rho_a and rho_c) were greater than 0.7, suggesting that the constructs were reliable. Table 8 shows that all constructs had AVE values greater than 0.5, indicating strong convergent validity. The SRMR was 0.101, reflecting an acceptable goodness of fit of the model.

Table 8.

Construct reliability, convergent validity and goodness of fit for the Croatian sample.

	Cronbach's alpha	Composite reliability (rho_a)	Composite reliability (rho_c)	Average variance extracted (AVE)
CL	0.861	0.868	0.906	0.706
CR	0.796	0.809	0.868	0.623
CS	0.893	0.894	0.926	0.757
CV	0.927	0.939	0.949	0.823
EMV	0.908	0.913	0.935	0.783
EPV	0.891	0.898	0.924	0.753
FV	0.843	0.880	0.893	0.678
SV	0.862	0.872	0.906	0.706
Standardized root mean square residual (SRMR) = 0.101

Source: SmartPLS4 software.

The HTMT results confirmed that each construct was distinct from the others given that the HTMT values were <0.9. These results support discriminant validity in the Croatian sample (Table 9).

Table 9.

Discriminant validity (HTMT for the Croatian sample).

	CL	CR	CS	CV	EMV	EPV	FV	SV
CL
CR	0.903
CS	0.746	0.762
CV	0.593	0.648	0.618
EMV	0.617	0.738	0.682	0.631
EPV	0.575	0.711	0.523	0.612	0.604
FV	0.57	0.65	0.564	0.61	0.633	0.602
SV	0.428	0.488	0.44	0.574	0.599	0.493	0.555

Source: SmartPLS4 software.

The results in Table 10 for the Croatian sample show that customer retention significantly influenced customer loyalty (t statistic = 7.815, p value = 0.000). Customer satisfaction had a significant moderate and positive effect on customer loyalty (t statistic = 3.361, p value = 0.001). Customer satisfaction also significantly affected customer retention (t statistic = 9.069, p value = 0.001). Conditional value significantly influenced customer satisfaction (t statistic = 2.977, p value = 0.003). Emotional value strongly and significantly enhanced customer satisfaction (t statistic = 3.674, p value = 0.001). Therefore, hypotheses H1e and H1b are supported. Conversely, epistemic value did not have a significant impact on customer satisfaction (t statistic = 0.760, p value = 0.448). Functional value did not significantly influence customer satisfaction (t statistic = 1.446, p value = 0.148). Social value did not significantly affect customer satisfaction (t statistic = 0.327, p value = 0.744). Thus, hypotheses H1a, H1c and H1d are supported.

Table 10.

Hypothesis testing results for the Croatian sample.

Hypothesis	Original sample (O)	Sample mean (M)	Standard deviation (STDEV)	t statistic (|O/STDEV|)	p value	Decision
CR -> CL	0.567	0.569	0.073	7.815	0.003	Supported
CS -> CL	0.295	0.294	0.088	3.361	0.001	Supported
CS -> CR	0.644	0.644	0.071	9.069	0.001	Supported
CV -> CS	0.253	0.253	0.085	2.977	0.003	Supported
EMV -> CS	0.372	0.366	0.101	3.674	0.001	Supported
EPV -> CS	0.072	0.072	0.095	0.760	0.448	Rejected
FV -> CS	0.133	0.140	0.092	1.446	0.148	Rejected
SV -> CS	−0.031	−0.026	0.094	0.327	0.744	Rejected

Source: SmartPLS4 software.

The results in Table 10 also show that customer satisfaction moderately and significantly influenced customer loyalty (path coefficient = 0.295), supporting H2. Additionally, customer satisfaction strongly and significantly affected customer retention, indicating that satisfied customers are likely to remain with the service (path coefficient = 0.644). Customer retention also had a strong and significant positive impact on customer loyalty, highlighting the importance of retaining customers for building loyalty (path coefficient = 0.567). These results support H3.

Croatian restaurant managers should focus on improving the emotional and conditional value of their service to enhance customer satisfaction. Given that social, epistemic and functional kinds of value were not found to have a significant impact on satisfaction, resources in these areas could be redirected to more influential areas.

5

Discussion

Providing value and giving feedback to customers are proactive strategies to stay ahead of the competition. Likewise, ensuring customer satisfaction (i.e. happiness, positive experiences, low prices, and high-quality products and services) is also important (Panda, 2009). The analysis of the three chosen countries reveals certain relationships related to the influence of perceived value on customer satisfaction, retention and loyalty (Table 11).

Table 11.

Perceived value with impact on customer satisfaction, retention and loyalty.

Variable / Country	Romania	Italy	Croatia
Social	good collaboration, belonging (Türkes et al., 2021), amusement, and facilities (Moisescu et al., 2021)	positive socio-anthropo-psychological space through a restaurant experience (Iofrida et al., 2022)	attitudes, behaviour (Cha & Borchgrevink, 2018), relationship, appearance, dialogue (Marković et al., 2021)
Emotional	new food (Türkes et al., 2021), authenticity, safety (Muntean et al., 2023), agreeableness, neuroticism (Fanea-Ivanovici et al., 2023), hedonist reactions (Voinea et al., 2020)	atmosphere (restaurant well-looked after, clean, chic, relaxing atmosphere) crucial for the “connoisseur” customer (Fanelli & Di Nocera, 2018)	happiness, delight, atmosphere, light (Pecotić et al., 2014), aesthetic, experience, fine dining (Marković et al., 2021)
Functional	Service quality, price fairness, ambiance (Moisescu et al., 2021), come back, satisfaction (Țuclea et al., 2018)	food quality (tasted good, spicy taste, healthy), service quality (friendly service, reasonable waiting times, and polite waiters; Fanelli & Di Nocera, 2018)	perceived cost and quality (Bajs, 2015), expectations, services, and attributes (Kukanja & Planinc, 2015), variety, gastronomy offer (Otočan & Cvek, 2020)
Epistemic	new solutions, experiences (Türkes et al., 2021), new knowledge and cognition (Bîlbîie et al., 2021), openness (Fanea-Ivanovici et al., 2023)	exotic and new experiences based on product country origin Martinelli and De Canio (2019)	new experiences based on art, music, ambient (Pecotić et al., 2014), cognitive behaviour, perception of experience (Cha & Borchgrevink, 2018)
Conditional	reduced prices (Türkes et al., 2021), satisfaction to eat, trustworthiness, credibility (Balaban & Mustatea, 2019)	internal sustainability policy adoption (Gazzola et al., 2024)	quality of service, satisfaction (Cha & Borchgrevink, 2018), time, smell, temperature (Marković et al., 2021)
Retention	good communication (Türkes et al., 2021), innovation and good prices (Sorcaru et al., 2023), green initiatives, green marketing strategies (Moise et al., 2021)	sustainability practices (Gazzola et al., 2024), restaurants’ brand (Fissi et al., 2023), food quality (Fanelli & Di Nocera, 2018), comfort and cleanliness, (Iofrida et al., 2022), green, local products (Scozzafava et al., 2017; Contini et al., 2017)	communication, design, spatial layout, colour, music, aesthetics, furniture (Pecotić et al., 2014), communication, employee behaviour, smiles, uniforms (Marković et al., 2021)
Loyalty	technologic innovations (Türkes et al., 2021), predictive norms (Bîlbîie et al., 2021), services on generations (Gurău, 2012), healthy and sustainable food (Voinea et al., 2020)	atmosphere and service quality (employees’ enthusiasm to treat consumers warmly but also attendants’ appearance; (Fanelli & Di Nocera, 2018), price-quality ratio (Iofrida et al., 2022)	perceived value, satisfaction (Bajs, 2015), food safety (Cha & Borchgrevink, 2018), reliability, responsiveness,assurance, empathy, price (Marković et al., 2011)

Source: Authors.

Consistent findings emerge across the three data samples from Italy, Romania and Croatia. Analysis of all three samples shows that customer satisfaction significantly influences customer loyalty and customer retention. In each sample, customer retention mediates the relationship between customer satisfaction and customer loyalty, highlighting the importance of retention strategies in boosting loyalty.

There are notable differences in the impact of specific kinds of perceived value on customer satisfaction across the three countries. In the Romanian sample, emotional value, functional value and epistemic value significantly enhance customer satisfaction. In contrast, social value and conditional value do not. Conversely, in the Italian sample, social value has a negative effect on customer satisfaction, whereas the effect of emotional value, functional value and conditional value is significant and positive. Epistemic value does not significantly affect customer satisfaction in Italy. In the Croatian sample, emotional value and conditional value positively affect customer satisfaction. In contrast, social value, epistemic value and functional value do not have significant effects.

Cultural dimensions such as those proposed by Hofstede (1980) provide a valuable framework for understanding the varying consumer preferences and behaviours observed in the three countries. For instance, Romania, Italy and Croatia have different scores in Hofstede's cultural dimensions. In particular, they differ in dimensions such as individualism, uncertainty avoidance and power distance. These dimensions could significantly shape customer perceptions of value in restaurants. For example, Italy's higher individualism suggests a greater emphasis on personal experiences. This situation possibly explains the negative impact of social value on customer satisfaction in Italian restaurants. This finding may indicate that Italian consumers prefer dining experiences that focus on personal enjoyment and functional aspects rather than social elements. Similarly, Croatia's relatively high uncertainty avoidance could make consumers more responsive to conditional value because they might seek context-specific reassurance before engaging in a dining experience.

According to Mulyana and Limakrisna (2023) and the literature on perceived value, consumers make decisions based on economic, functional, epistemic, emotional and social factors. Therefore, a customer profile can be developed (Table 12) for the three countries included in the analysis. In each case, each kind of perceived value has a significant or non-significant impact on customer satisfaction.

Table 12.

Customer profile by country, considering the five components of perceived value.

Country / perceived value	Social	Emotional	Functional	Epistemic	Conditional
Romania	No	Yes	Yes	Yes	No
Italy	Yes	Yes	Yes	No	Yes
Croatia	No	Yes	No	No	Yes

In terms of customer satisfaction, social value is significant only for Italians. Emotional value is significant for all three countries. Functional value is not significant for Croatians. Epistemic value is significant only for Romanians. Finally, conditional value is not significant for Romanians. Emotional value is the leading kind of value, with a direct and positive impact on customer satisfaction in restaurants across the three countries. Functional value and conditional value follow closely in importance. Therefore, H1a, H1b, H1c, H1d and H1e are supported by the results based on customer value. H2 and H3 are supported for all three countries, indicating that customer satisfaction significantly influences customer loyalty and customer retention. In each country, customer retention mediates the relationship between customer satisfaction and customer loyalty. This finding emphasises the importance of retention strategies in boosting loyalty.

These country differences underscore the diverse cultural and contextual factors that shape customer perceptions and satisfaction in restaurant services across different countries. The findings of this study bridge a gap in the literature on the restaurant industry by exploring the impact of five kinds of perceived value on the loyalty of restaurant customers in Romania, Italy and Croatia.

The results of the study illustrate how the value perceived by customers influences satisfaction and loyalty in restaurants in Romania, Italy and Croatia. Despite similarities between these countries, the study reveals significant cultural differences that can guide management and marketing strategies tailored to the restaurant industry in each country.

In Romania, emotional value, functional value and epistemic value have a significant impact on customer satisfaction. This finding suggests that Romanian consumers value not only the quality and utility of a restaurant service but also the novelty and unique experiences it offers. Epistemic value is related to exploring new flavours and culinary experiences. It is significant for Romanian consumers, who show considerable interest in gastronomic diversity (Voinea et al., 2020). Emotional value is strongly linked to personal relationships, with restaurants seen as places for social and family bonding.

In Italy, the results are more complex. For instance, social value has a negative impact on satisfaction. This unusual finding can be explained by the Italian cultural emphasis on authenticity and simplicity in culinary experiences. Italians tend to reject an artificial or excessive perception of social prestige associated with restaurants, preferring an authentic and relaxed environment (Fanelli & Di Nocera, 2018). However, emotional value is also extremely important in Italy, suggesting that emotional experiences play a central role in determining satisfaction.

In Croatia, emotional value and conditional value have the greatest impact on customer satisfaction. This finding reflects the seasonal nature of Croatian tourism, where restaurants must adapt to the needs of tourists depending on the time of year. Conditional value, related to the specific context and circumstances in which the service is provided, is essential in a tourism-based economy (Kukanja & Planinc, 2015). Similarly, emotional experiences are important for both tourists and locals, highlighting the importance of a relaxing and friendly environment in Croatian restaurants.

Comparing the three countries shows that emotional value is a common factor that positively influences satisfaction in all three contexts. In contrast, functional value is most appreciated in Romania, where consumers focus on the tangible quality of services. In Italy, social value has a negative impact, reflecting a preference for authenticity over social prestige. In Croatia, conditional value is of greatest importance, indicating the need to adapt to the seasonality of tourism.

These cultural differences underscore the complexity of managing customer satisfaction and loyalty in the restaurant industry across Europe. They show that marketing and management strategies must be adjusted to adapt to the specific features of each local market. Although an emphasis should be placed on emotions in all three countries, managers must be aware that social, epistemic, functional and conditional kinds of value vary significantly according to the culture and expectations of local consumers. This comparative analysis provides a broad perspective on consumer behaviour in restaurants across three cultural contexts. By highlighting both similarities and differences, the study shows how perceived value influences satisfaction and loyalty in distinct ways. The study thus offers a framework for adapting management and marketing strategies according to local culture.

The findings reveal key behavioural trends and differences that restaurant managers should consider while developing CRM strategies tailored to each market. The perceived significance of emotional value suggests a consistent need to focus on enhancing customer experiences across all three countries. However, the varying impacts of functional, social, epistemic and conditional kinds of value indicate that a one-size-fits-all CRM approach is not suitable. For example, Romanian restaurants might benefit from focusing on functionality. Italian restaurants should pay more attention to social elements, emphasising functionality and situational adaptability instead. In Croatia, creating context-specific dining experiences could be more effective for driving customer satisfaction.

Emotional value is highly relevant for customers in all three countries because dining is not only functional but also highly experiential. Atmosphere, service and emotional connections combine to shape customer perceptions in a significant way (Han et al., 2009; Jang & Namkung, 2009). Culturally, although Romania, Italy and Croatia differ in their value systems and modes of emotional expression, they share a Mediterranean (Italy) and Balkan (Romania and Croatia) influence, where social interaction, warmth and enjoyment are central, particularly during shared meals (Gostin et al., 2021). In fact, in these cultures, dining out is often associated with positive emotions such as joy, comfort and social bonding, which enhance the overall experience (Bradatan, 2002). Interestingly, the literature suggests that emotional experiences are easier to remember, as are positive emotions such as hedonism, excitement, happiness and refreshment (Rašan & Laškarin, 2023; Zhong et al., 2017).

This study has some relevant implications. It makes several contributions to understanding perceived value in restaurants in relation to satisfaction, retention and loyalty. The literature review also makes a theoretical contribution regarding perceived value, satisfaction, retention and loyalty of customers. Methodologically, this study provides a model to measure the impact of the five kinds of perceived value (i.e. social, emotional, functional, epistemic and conditional) on customer satisfaction. Additionally, this study shows that the five kinds of perceived value influence satisfaction and that retention positively affects loyalty.

Therefore, practical implications for restaurants arise from this study. Specifically, they should implement effective policies based on menu diversity, food quality, food sustainability, environmentally friendly behaviour, effective communication, customer relationships, competitive pricing and quality. Restaurant managers should focus on trust, profitability, skilled staff, customer relationships, personalised service and customisation. Finally, customers seek to enjoy a pleasant dining experience, a welcoming atmosphere, high-quality meals, positive relationships, relaxation, emotional connections and positive emotions. This mutually beneficial situation will lead to improved performance, satisfaction, retention, loyalty and happiness.

Although existing research acknowledges the role of perceived value in customer satisfaction, the current cross-cultural approach shows that emotional value is a universal enabler of customer satisfaction, whereas other kinds of value have country-specific effects. These insights advance cross-cultural consumer behaviour theory and provide a practical framework for businesses seeking to enhance customer satisfaction in diverse cultural contexts.

In addition to these theoretical contributions, this study is of considerable practical relevance for restaurant managers in Romania, Italy and Croatia. The findings provide a detailed understanding of customer perceived value (social, emotional, functional, epistemic and conditional), allowing managers to adapt their services to consumer expectations. In a competitive sector such as the restaurant industry, customer loyalty and retention are important for profitability and long-term success (El-Adly, 2019). By identifying the kinds of value that most strongly influence customer satisfaction and loyalty, restaurants can improve not only service quality but also marketing strategies, thereby attracting and retaining a loyal customer base. For instance, customer loyalty can significantly reduce the costs of acquiring new customers while providing a stable source of income (Darzi & Bhat, 2018; Fatima et al., 2018). Specifically, managers could invest more in enhancing emotional experiences and service quality. These areas are found to be critical in all three countries analysed in the study. Additionally, retention strategies can be optimised by developing loyalty programs and enhancing direct communication with customers. Doing so can encourage them to return and can transform occasional customers into brand ambassadors (Kim et al., 2012; Kumar Rai, 2013). Restaurants can thus offer a personalised experience that directly responds to the needs and desires of each customer. They can thereby maximise customer loyalty and retention, contributing to long-term financial performance (Lee & Han, 2022).

The study has some limitations. First, the data collection period was limited to winter and spring. Future studies could expand the scope to cover the entire year. Second, the sample consisted of customers from three European countries. Future research could include more areas and countries to increase the generalisability of findings. Third, conducting the research in more restaurants could provide more accurate insights into the perceived value, satisfaction, retention and loyalty of customers. Finally, although the study identifies significant cross-country differences in perceived value, it does not delve into the specific cultural factors or mechanisms driving these differences.

This research provides a solid foundation for further exploration of customer perceived value, satisfaction, retention and loyalty, as well as further study of how to enhance the behaviour of the sales force, customer communication, relationships, experiences and atmosphere. Future research should include direct customer feedback through interviews, enabling the study of perceived value and its impact on satisfaction from a mixed methods perspective. Furthermore, future studies could explore the specific cultural dimensions that may influence how customers in different countries perceive and prioritise various kinds of value. Validated frameworks such as Hofstede's cultural dimensions could provide a valuable source for such in-depth cross-cultural analyses.

CRediT authorship contribution statement

Gabriel Croitoru: Writing – original draft, Methodology, Formal analysis, Conceptualization. Alexandru Capatina: Writing – review & editing, Validation, Software, Methodology, Formal analysis. Nicoleta Valentina Florea: Writing – original draft, Visualization, Supervision, Methodology, Investigation, Formal analysis, Conceptualization. Federica Codignola: Writing – review & editing, Writing – original draft, Visualization, Supervision, Investigation, Data curation. Danijela Sokolic: Writing – original draft, Supervision, Methodology, Data curation, Conceptualization.