Research on the Influence of After-sale Service on Customer Loyalty -- A Case Study of Budget Hotels

. Customer loyalty is very important to the hotel’s survival and development, so the hotel will maintain and improve customer loyalty in all aspects. The same is true in budget hotels. In this paper, budget hotels were selected as the research object, and college students as the research object, through the way of a questionnaire survey to study how after-sale service affects customer loyalty. In the survey, additional services are used to refer to after-sales services. According to the service quality model, additional services were divided into five aspects: responsiveness, caring, compensatory, reliability, and assurance. Through regression analysis, only caring and assurance of additional services had a positive impact on customer loyalty. Therefore, based on the characteristics of college students, it was suggested that budget hotels should be customized in the design and arrangement of additional services. Evaluate the quality of customized services and make service corrections. At the same time, the influence of selected variables on customer loyalty can be verified in other groups.


Background
Nowadays, due to fierce competition in the hotel industry, for better market share, it is necessary to firmly grasp the customer group, so hotel customer loyalty is particularly important. Based on the change in lifestyle and the popularity of the Internet, people have been accustomed to booking and evaluating hotels on the Internet. It's common to see customers pointing out problems with hotels in reviews on the site, such as hot water is not hot, the bed is tide, sound insulation is not good, and so on. Some of these problems are found in the process of customer accommodation, it could be solved by changing the room after the customer put forward, but some failed to deal with or solved at that time, customers complained in the comments, such not only affect the customers to book the hotel next time. At the same time, it would also affect the choice of other customers who have the intention of booking the hotel.
Therefore, the quality of after-sales service will affect customer loyalty and attract new customers. After-sales service quality is a part of service quality, so the research on after-sales service refers to the relevant literature on service quality. According to the Quality-of-Service Model (SERVQUAL), Xu Xuefeng in 2021 divides online shopping after-sales service into four dimensions, namely responsiveness, compensatory, caring, and reliability, in her research on the effect of online shopping after-sales service on customers repurchase intention [1]. Zhang Hao in 2022 studied whether new retail technology improves customer loyalty, and designed after-sales service indicators from three dimensions of responsiveness, compensatory, and reliability according to his research needs [2].
According to relevant literature, this paper divides the hotel after-sales service into five dimensions: Responsiveness, which means that the hotel responds to customers' needs in a timely manner, such as solving customers' problems quickly. Compensatory refers to the psychological or physical loss of customers caused by the hotel's mistakes, some kind of compensation for customers. Caring means that the hotel gives priority to the interests of customers, thinks about solutions from the perspective of customers, and provides personalized services to customers; Assurance, that is, the hotel staff is trustworthy, customers will feel assured of its service. Reliability refers to the hotel attaches great importance to commitment and can provide the promised service on time. Taking service quality as an intermediary variable, this paper studies the influence of hotel after-sale service on customer loyalty.

Research Gap
According to the previous literature, there are few studies on after-sales service and customer loyalty. One of the reasons is the particularity of hotel accommodation products. Hotel accommodation products are intangible products, which represent the experience process and feelings of customers. Therefore, for such products as hotel accommodation, after-sales service cannot be treated as the same as ordinary tangible products. Secondly, the after-sales service of the hotel is carried out in the way of a loyalty plan, customer return visit, reply to customer comments, etc., which is different from the content and way of general after-sales service. Third, the word after-sales service is not used in the hotel industry.
At present, the hotel's after-sales service cannot solve customers' problems and reduce complaints. Customers cannot see the actual solution effect, so the negative impression of the hotel is not reduced, this situation makes customers think that the problem of the hotel is handled by default, do not care about it, and do not pay attention to the ideas or opinions of customers, said also has little effect, do not have hope for the hotel.
Therefore, hotel after-sales service is chosen as the research direction of this paper. Clarify the content and form of hotel after-sales service, evaluate the quality of hotel after-sales service, study whether after-sales service affects customer loyalty, and improve customers' negative impression of hotel service. To study customer loyalty from a new perspective, enrich the research field of customer loyalty, and fill the research blank in this area.

Structure of This Paper
It is assumed that after-sales service will affect hotel customer loyalty. In order to verify whether the hypothesis is correct, this paper proves it through the questionnaire survey. After consulting the literature related to customer loyalty, service quality, and after-sales service, the express hotel is selected as the research object, college students will be surveyed in the form of the online questionnaire, and using SPSS software to analyze the data. Based on the data analysis results and service quality theory, the correlation and influence degree between after-sales service and customer loyalty are discussed and summarized.

Definition & Development
The concept of customer loyalty originates from brand persistence and was first proposed by Copeland [3]. The researchers introduced the basic concept of customer loyalty to define hotel customer loyalty. Bowen and Shoemaker in 1998 proposed that guests are very willing to treat themselves as part of the hotel and assist in the operation of the hotel [4]. On this basis, Shoemake and Lewis defined customer loyalty in 1999 from two aspects: the likelihood of customers returning to the hotel and the willingness of people to act as partners in the hotel organization [5]. Wilkins H. et al pointed out that customer loyalty is difficult to be observed, so it must be inferred from other relevant indexes. For hotel rooms, customers' intention to visit the hotel repeatedly is a standard to measure customer loyalty [6][7][8].

Important Results
In defining brand loyalty, American scholar Day advocated taking both attitude and behavior into consideration [9]. In 1996, by studying the frequency of repeat purchases and the strength of positive attitudes, Griffin concluded that customer loyalty can be divided into four states: loyalty, potential loyalty, lazy loyalty, and disloyalty [10]. From another perspective, Grernler and Brown 1999 proposed that customer loyalty could be subdivided into behavioral loyalty, intentional loyalty and emotional loyalty [11].
In terms of customer loyalty measurement. In 1992, Fornell proposed to measure customer loyalty by customers' multiple purchases, customers' price tolerance, customers' first consideration of the enterprise, and The Times of purchasing products or services of the enterprise [12]. Selens believes that it should be measured from the perspective of customers' re-purchase intention, positive wordof-mouth publicity and active introduction to people around [13]. Based on summarizing previous studies, GronH. Ldt and Kri Stensen pointed out that customers' willingness to re-purchase, price tolerance, willingness to recommend to others and customers' willingness to cross-purchase were used as indicators to measure customer loyalty [14].
In customer loyalty empirical research model. The first is the ASCI model, which was established by Anderson, E. W. et al. in 1994 on the basis of the Swedish Customer Satisfaction Index (SCSB). This model measures the quality of products or services by the consumption experience of consumers. The variables of the model include six variables: perceived value, customer expectation, perceived value, customer satisfaction, customer loyalty and customer complaint. The first three are the premise variables, and the last three result variables have a certain influence [15]. The second is the ESCI model, which was first applied to evaluate Denmark's postal service. Due to its good effect, it has been vigorously promoted in various fields. In the ESCI model, premise variables include the corporate image, customer expectation, perceived quality (hardware, software) and perceived value, and customer loyalty is the outcome variable. The third is the value-loyalty model, which holds that customer value plays a decisive role in customers repurchase intention [16]. The fourth is the dualdrive model of customer value and customer satisfaction, which recognizes that customer satisfaction and customer value have the same influence on customer loyalty [17].

Summary
Customer loyalty originated from brand loyalty and was later applied to hotel management. Scholars have not unified the definition of hotel customer loyalty. With the deepening of research, loyalty measurement has been refined to be measured from two aspects of attitude loyalty and behavioral loyalty. More scholars have applied the customer loyalty theory into practice and established four models.

Definition & Development
A service product is an intangible product, because of its particularity, its concept is different from a tangible product.
Lewis and Booms defined service quality as "a tool to measure how well an enterprise's service level meets customer expectations" [18]. Gronroos believed that customer evaluation determines the real level of service quality, and enterprises are only service providers. Because the customer perception of service quality is based on the customer's subjective feelings, is based on the expectation of service quality and experience after the comparison [19].

Important Results
Minh et al. in 2015 said that empathy was the most obvious influence on customer satisfaction [20]. Empathy refers to providing personalized attention to customers, understanding their specific needs, and maintaining a positive attitude toward customers' feedback. Another dimension of service quality is responsiveness. Johnston in 1997 defined responsiveness as the ability to measure the timeliness of a company's willingness to serve and respond to customers, including providing services in time to cope with long queues and waiting periods [21]. In addition, responsiveness is the availability to respond to customer requirements and the flexibility to meet customer requirements [20]. Including reliable, punctual and timely provision of services. Matti Haverila et al. in 2019 found that customer satisfaction and dissatisfaction were mainly determined by the tangible and assurance of the SERVQUAL [22]. While interacting with customers, the staff should be more polite. Enterprises should employ a person who can give confidence to customers [23]. Employees should be warned, who exhibited unreliable behaviors. If necessary, they will be discharged. The architecture and landscaping of the enterprise must match the services provided, and establish a link between appearance and service [24]. In a case study, customers complained about the litter inside the campus and the hot water service. Goldi et al. suggested the hotel place proper sanitation facilities, and install the solar heater [23]. Based on the data, Müjdat ERTÜRK found the tangibles were the least important dimension of the model [24]. Meanwhile, Ali Bahraminasab in 2016 only had a positive mean score for tangibles, which means that the views of respondents match their expectations, and next was assurance (the lowest negative mean score) [25].

Summary
Based on the above literature, scholars have discussed the relationship between service quality and customer loyalty. In different case studies used the SERVQUAL, scholars get different or even mutually exclusive views due to the characteristics of the cases themselves. In most studies, tangible was generally rated lowest in importance and reliability ranked highest. The other three segments differ little, and the order varies from case to case. Nevertheless, they still play a crucial role in customer loyalty.
The SERVQUAL is an important tool to test whether customers are satisfied with the hotel service. Its application enables the hotel to identify its problems or shortcomings in the service, and make targeted corrections according to the analysis results, so as to improve the comfort and fluency of customer experience. Improved customer satisfaction means that customers tend to choose the same hotel next time. In the long run, it helps to promote the formation of customer loyalty to the hotel.
After introducing the concept of service quality, the paper focuses on the evaluation dimensions of service quality model. Various scholars have made various interpretations of the evaluation model, among which SERQUAL model has the highest acceptance and the widest application range. Therefore, the research model in this paper evolves according to SERQUAL model.
Based on the theoretical basis mentioned above, this paper focuses on whether the five aspects of additional services affect the loyalty of the investigated groups to budget hotels. The following hypotheses are proposed: H 1: Responsiveness has a positive impact on customer loyalty; H 2: Caring has a positive impact on customer loyalty; H 3: Compensatory has a positive impact on customer loyalty; H 4: Reliability has a positive impact on customer loyalty; H 5: Assurance has a positive impact on customer loyalty.

Research Design
In order to test the hypothesis proposed earlier, quantitative research is chosen in this paper. Quantitative research uses questionnaires. In the questionnaire, additional services are used to refer to after-sales services for the convenience of the survey and the respondents' understanding. There are 28 questions in the questionnaire on the influence of additional services of budget hotels on customer loyalty, which are divided into two parts: the basic characteristics of investigated customers and the additional services. In the part of basic characteristics, gender, grade, screening and the way of budget hotel reservation are selected. Among the screening questions, if you have never stayed in a budget hotel, the questionnaire will be answered. The second part mainly investigates the degree of customers' consent to additional services and whether it affects their loyalty to such hotels. The additional service part is divided into 17 small problems; three small questions are designed for loyalty. Finally, an open question is set to suggest existing additional services.

Data Collection
All measurement items of variables in this study were scored by Likert five-level scale, where 1 represents "strongly disagree" and 5 represents "strongly agree". The scale design comes from the existing mature scale in the existing literature, and has been adjusted according to the needs of this paper. There are five variables in the scale of additional service for budget hotels, among which responsiveness, caring, reliability and assurance are mainly referred to Parasuraman et al. [26]. Compensatory measurement items were mainly adapted from Smith's scale [27]. Customer loyalty mainly refers to Parasuraman scale [28].
Respondents were chosen from college students, based on the current form and students' use of mobile network. The survey was conducted in April 2022 through an online questionnaire. A total of 233 questionnaires were collected, among which 45 respondents had never stayed in budget hotels. The remaining questionnaires were valid, and 188 were valid. According to Table 1, there are 45 male students, accounting for 23.9%, and 143 female students, accounting for 76.1%. There may be several reasons for the high proportion of female students. First, female students are more patient than male students when filling out questionnaires. Second, girls are more willing to express their feelings than boys. Third, girls are the main decision makers in the travel process.  As can be seen from Table 3, respondents can book budget hotels through various channels such as on-site booking, telephone booking, official APP, official website, WeChat mini program/official account and third-party platform. Among them, most people choose third-party platforms, accounting for 58.0%; The second is the on-site booking accounted for 16.0%; 11.2% chose WeChat mini program/public account; The official APP ranked fourth. For college students, third-party platform booking is very important for budget hotels, accounting for more than half of the total, so the hotel should pay attention to hotel marketing and feedback information in all aspects on third-party platform. 89.4% of respondents choose budget hotels to provide additional services, accounting for a large proportion, indicating that the additional services of this type of hotel cannot meet the needs of this group in terms of both quality and quantity. After checking in or checking out, 89.4% of the respondents chose to communicate with the hotel, indicating that college students pay more attention to their personal interests and most of them will express themselves and take the initiative to fight for them.

Reliability and validity analysis
According to De Willis scholar, when the reliability of the scale is higher than 0.80, it indicates that the questionnaire has a very high reliability. When α is greater than 0.70, the reliability is good. When the α reliability is between 0.65 and 0.70, it belongs to the minimum acceptable range. α lower than 0.60 indicates low reliability, which proves that the questionnaire is not advisable [29]. In this study, SPSS was used to analyze the reliability of the questionnaire after removing the basic information data of customers. In Table 6, the Cronbach coefficient is 0.946 and greater than 0.80, indicating that the questionnaire design of the impact of additional services on customer loyalty of budget hotels is reliable and reliable. In order to test whether the questionnaire is correct and effective, this paper conducts structural validity analysis on the basis of removing basic tourist information data. In Table 7, the results of questionnaire validity test show that the sample measurement degree of KMO is 0.944, and the score tends to 1. The significance of Bartlett's sphericity test is 0.000, indicating that the questionnaire has good validity. .000 .000

Correlation analysis
Caring Pearson Correlation .702 ** Sig. (2-tailed) .000 Compensatory Pearson Correlation .726 ** Sig. (2-tailed) .000 Reliability Pearson Correlation .753 ** Sig. (2-tailed) .000 Assurance Pearson Correlation .799 ** Sig. (2-tailed) .000 **. Correlation is significant at the 0.01 level (2-tailed). According to the Pearson correlation coefficient, | r | greater than or equal to 0.8, can be seen as highly correlated between two variables; The correlation between 0.5 and 0.8 can be regarded as moderate. In Table 8, the significance is 0.000, and the correlation coefficients are in the range of 0.5-0.8, belonging to the moderate correlation. Therefore, responsiveness, caring, compensatory, reliability, assurance and customer loyalty are positively correlated, among which assurance and customer loyalty have the highest degree of correlation.

Regression analysis
In Table 9, the value of R square is 0.683, and the adjusted value of R square is 0.674. Multiple dependent variables were involved in this study, the adjusted R square is adopted as the judgment coefficient. The statistical results show that dependent variables can be explained by independent variables to a high degree of 67.4%. The standard estimation error is 0.46121, indicating that responsiveness, caring, compensatory, reliability and assurance of independent variables have a good predictive degree in predicting customer loyalty of dependent variables, and the average deviation of regression equation in predicting customer loyalty is about 0.46121.  Table 10, the results of variance analysis are relatively ideal. Among them, the total deviation sum of squares of dependent variables is 122.057, the sum of regression squares and residual squares is 83.343 and 38.713, respectively. The mean square value of regression was 16.669, and the mean square value of residual was 0.213, respectively. F value was 78.363, P value was 0.000 < 0.05. So, don't refuse to assume that responsiveness, caring, compensatory, reliability, assurance that directly affect customer loyalty, is explained and interpreted variables in the model of the relationship between the linear regression equation is significant, emerged from the data results of variance analysis can establish regression model are analyzed.  Table 11, in the no standardization coefficient B, constant represents the Y-axis truncation, X1 coefficient is 0.009, X2 coefficient is 0.201, X3 coefficient is 0.14, X4 coefficient is 0.12. X5 coefficient is 0.48. After substituting each coefficient, the following equation can be obtained: = 0.173 + 0.009 1 + 0.201 2 + 0.14 3 + 0.12 4 + 0.48 5 (1) The regression weight of each variable is positive, indicating that the increase of a certain variable means a certain increase in customer loyalty.
In the significance test in the table, X2 (t=2.422, P =0.016) and X5 (t=5.172, P =0.000) are significant, showing that there is a significant linear relationship between the independent variables of caring, assurance and the dependent variables of customer loyalty. X1 (T =0.109, P =0.914), X3 (t=1.622, P =0.107), and X4 (t=1.189, P =0.236) were not significant. The results show that there is no significant correlation between responsiveness, compensatory, reliability and customer loyalty. Therefore, the regression equation obtained after correction is: It can be seen from the regression equation that caring and assurance have a positive impact on customer loyalty, among which assurance has the highest impact on customer loyalty, followed by caring. Therefore, it is concluded that assurance has more influence on customer loyalty than caring, namely is X5 > X2 in descending order.

Result
A multiple regression model was established to predict the impact of additional service on customer loyalty with five variables of responsiveness, caring, compensatory, reliability and assurance. Generally speaking, the regression model was significant, F (5,182) =78.683, p<0.05, adjust R 2 =0.674. Among the variables investigated, X2 (t=2.422, P =0.016) and X5 (t=5.172, P =0.000) were significant, indicating that there was a significant linear relationship between these two independent variables and the dependent variable. X1 (t=0.109, P =0.914), X3 (t=1.622, P =0.107) and X4 (t=1.189, P =0.236) were not significant, indicating that there was no significant relationship between these three independent variables and dependent variables. Therefore, at the significance level of 5%, hypothesis 2 and hypothesis 5 pass the test. Hypothesis 1, hypothesis 3 and hypothesis 4 fails the test.
This indicates that in this survey, only caring and assurance of additional services have a positive impact on customer loyalty.

Discussion
This study refers to the variables and specific projects in the service quality management model and combines the actual situation to design the variables and contents required in this study. The regression analysis shows that for college students, the caring and assurance of additional services in budget hotels have a positive impact on customer loyalty. Therefore, for this group, this study should let them feel valued, and their requirements in the first place, and actively provide services to obtain the evaluation of the hotel, for the service to provide personalized, different from the general way of service. No one else can take the place of the hotel and staff to ensure the services and content of the hotel, so the hotel and staff must be trustworthy. For the problems, the hotel and the customer are the direct parties, facing the customer, the hotel and its staff to provide quality service and better solutions to the problem is very important. At the same time, in order to better solve customer problems, the hotel's support for employees will increase customers' trust in the hotel, thus deepening the good enterprise atmosphere and image, and promoting the formation of customer loyalty.
For hotel additional services, there are few related studies, so refer to the online shop after-sales service quality research. In the reference article, corresponding variables are designed according to the service quality model, and different hypotheses are set according to the specific project, and all the hypotheses are passed. In this paper, based on the actual on the design of the variable selection from five aspects, put forward the hypothesis of not one by one through the inspection, show the characteristics of the selected students' groups are distinct, compared with other age groups, social experience is not rich, but autonomy is stronger, the pursuit of individual character and comfortable, strong protection for their rights consciousness, willing to communicate and express themselves.
Limited by the social environment, there are only 188 valid questionnaires in this survey, which is a small number. The students themselves lack life experience and social experience. Due to the epidemic in recent years, they have fewer opportunities to travel and fewer opportunities and times to stay in budget hotels. Therefore, their understanding of the questions and options in the questionnaire is slightly inadequate.
Therefore, follow-up investigations can be carried out in other groups. Different groups have different life experiences and cognitive levels of budget hotels, and again determine whether the five variables of additional services have an impact on customer loyalty.

Conclusion
In the case of increasingly fierce competition in the hotel industry, customer loyalty has become the main means to stabilize the hotel's market position and maintain market share. At the same time, many aspects affect customer loyalty. This paper chose after-sales service as a starting point to study the impact of after-sales service on customer loyalty.
Through the literature on customer loyalty, service quality, select type chain hotels as the research object, choose college students as the research object, through online in the form of questionnaire data, using the regression analysis shows that caring and assurance of additional services have a positive effect on customer loyalty, responsiveness, compensatory and reliability failed the inspection, it has no positive influence on customer loyalty. College students have distinct personalities and pay attention to their interests and ideas. Therefore, they should take the initiative to provide timely services to solve problems and show them that the hotel is trustworthy.
College students have their particularity, so they should train and design service processes and service requirements according to their group characteristics.
Due to the limited number of samples and the social environment, the depth and breadth of the obtained data need to be increased. Subsequent in-depth studies can be carried out on this group on the premise of ensuring the number of samples, to verify whether the variables that pass the test are still caring and assurance. At the same time, it can be extended to different groups to determine whether the variables affecting customer loyalty are the same.