Changes in guests’ hotel evaluations due to covid-19: The case of the world heritage city of Cuenca

The COVID-19 pandemic has had a major impact on all aspects of people's lives and on the economy. While it has affected all economic sectors, it has hit the tourism sector particularly hard given the huge effect of the measures related to the restriction of mobility and social distancing imposed to contain the virus. In the strictest stage of controls, these measures caused the closure of hotels and the cancellation of flights, as well as major changes in consumers' behavioural habits (Sheth, 2020).

Within the tourism sector, the hotel industry has undoubtedly been one of the worst hit (Hall et al., 2021), with consumers’ fear of COVID-19 causing a drop in occupancy rates, the delay or cancellation of trips, and other consequences that are still to be analysed (Jiang and Wen, 2020). Hotels were subject to severe restrictions at the beginning of the pandemic, then had to adapt to the rules imposed during the so-called “new normal”, and are now experiencing huge uncertainty surrounding the evolution of the situation and the change in consumer expectations (Mehta et al., 2023). They therefore have to adapt to the situation we are going through and the new needs and concerns that arise as a result; a valuable source of information that enables them to do so is the evaluations given by guests in their online reviews (Nilashi, et al. 2021).

The changes caused by COVID-19 together with technological advances in the tourism sector have led to the mass use of online platforms for the sale of tourism products. These platforms thus generate a large volume of data about the customer service experience, information that is not under the control of tourism businesses. In the case of hotels, the exponential growth in room reservations made through electronic distribution channels (Levy et al., 2013; O'Connor, 2008) has entailed an unprecedented increase in the information available online to hotel users; indeed, it currently represents the main source of information consulted before making a hotel reservation.

With the emergence of Web 2.0, electronic word of mouth (eWOM) shared through social networking sites has become the main source of information for many travellers (Schindler and Bickart, 2002; Liu et al., 2013). Before buying, consumers tend to search for information on the internet about the product they want to buy, weighing up the comments that other consumers have made about those products (Jiang et al., 2021). This information is hugely important since, as established by Nicoli and Papadopoulos (2017), about 35% of travellers change their mind about which hotel to book after browsing social networks, 53% say they will not book a hotel that does not have any reviews, and 87% say that reviews help in making a purchase decision. In addition, when planning a trip and deciding which hotel to stay at, people tend to look for information from others online as they consider it more truthful.

This online information is especially important in the hotel industry, since, unlike tangible products, the services received in this industry cannot be assessed before receiving them (Lee et al., 2018). However, it is necessary considering the role of individual motivations and technological factors in review posting intentions. Moreover, managers should motivate customers using rewards which enhance their self-reputation and self-image (Bakshi et al., 2019). In addition to eWOM, web-based platforms may yield user-generated content (UGC) that has a significant impact on customers, allowing users to share their views and experiences with others and, as a result, influence their purchase decisions (Chakraborty and Bhat, 2018; Ahani et al., 2019; Nilashi et al., 2021). Moreover, they represented a crucial source of information in the different stages of the pandemic: for clients, allowing them to access up-to-date information quickly and easily; and for hoteliers, allowing them to readily identify clients' primary concerns in order to implement the necessary corrective measures (Guo et al., 2017). Unlike surveys, this information is not susceptible to issues such as measurement error, the form of the questionnaire, the terms of the survey, the categories of the answers, or the order in which the information appears. In addition, some online platforms restrict evaluations so that only people who have actually stayed at a hotel can leave a review, thereby exerting some control over the responses. This is the case with the platform Booking (Booking, 2022), from which we source the information used in this study. Thus, we avoid the non-genuine customers that are not relevant, but they could affect to the results (Silva et al., 2021). Although we have several problems related with the existing universal rating categories such as emphasize First-Komen and Vujičić (2022). The last problem has been solved also examine the online reviews and not only the rating categories that they do not cover important themes.

Of all the available information, client satisfaction, which measures the gap between perceived and expected service quality (Padma and Ahn, 2020), is critical since it plays a key role in the ability of tourism to recover from setbacks, particularly crisis situations such as the one caused by COVID-19. Indeed, client satisfaction affects repurchase intention (Cronin and Taylor, 1992), loyalty (Getty and Thompson, 1994), WOM (Anderson et al. 1994) and consequently profitability (Xu and Li, 2016). Given this major impact, a fundamental objective of marketing activities centres around client satisfaction. In this regard, Xiang et al. (2015) demonstrate a strong association between the guest experience and the level of satisfaction, which suggests that these two domains of consumer behaviour are inherently linked. As such, in this study, we take the guest's evaluation of his/her stay at the hotel as a measure of the level of client satisfaction achieved.

However, most of the studies that analyse the effect of COVID-19 on the hotel industry (Filimonau et al., 2020; Hao et al., 2020; Jiang and Wen, 2020) overlook this aspect and focus more on those related to hotel management (Jiang and Ritchie, 2019; Ritchie and Jiang, 2019). In this regard, Song et al. (2022) report that hotel managers should consider changes in client satisfaction in order to improve the hotel guest experience and thus encourage these guests to return to the hotel and engage in positive eWOM.

The aim of this paper is to analyse guest evaluations of stays in different tourist accommodations to determine whether they have changed over the different stages of the COVID-19 pandemic, and to see if they have recovered after the hardest stage of the pandemic. In addition, we analyse possible changes in the key aspects of the evaluations given, and the possible appearance or prevalence of certain factors, caused by the COVID-19 pandemic. Based on the above, we can pose a series of questions to answer using the available information:

To conduct this research, we study hotel guest reviews before, during and after the pandemic. Specifically, we consider four stages of the pandemic: before the pandemic, using data from July 2019; during the pandemic, with data from July 2020; a period of recovery from the pandemic (the "new normal") in July 2021; and a time when the pandemic seems to be over (July 2022). The aim is to determine whether guest satisfaction changes significantly over the different stages of the pandemic, as well as analysing the factors—both positive and negative—that hotel guests consider important. We have 1595 reviews for a total of 17 hotels located in the World Heritage City of Cuenca and its surrounding area, in order to assess both urban and rural tourist accommodation. This analysis allows us to identify the main factors taken into consideration at each stage of the pandemic, in order to highlight the aspects that hotel managers should pay most attention to, and thus be able to tailor their service offerings and marketing campaigns to stimulate demand. Satisfied customers not only help a particular hotel remain financially viable in these difficult times, but they can also aid in the recovery of the industry as a whole.

In section 2, we conduct a literature review to understand the state of the research and provide a starting point for this study. Section 3 describes the database and the method used in this study, Section 4 presents main results, while Section 5 details the key conclusions and contributions for hotel managers, as well as the limitations of the study and future lines of research.

The reviews published by the users of a hotel establishment are crucial for understanding how clients feel and what aspects can be improved to boost future reservations. In this regard, Vermeulen and Seegers (2009) find that consumers' exposure to online reviews improves the likelihood of booking a hotel, with the influence being greater for less well-known accommodation options. Stringam et al. (2010), using information from Expedia.com, show the positive effects of the evaluation on recommendation intentions. In addition, they identify four aspects with a strong influence on the overall rating given by clients: hotel service; hotel condition; room cleanliness; and room comfort. Of all these, hotel service is found to be the most important, followed by room comfort, hotel conditions and finally cleanliness.

Obviously, it is critical to understand which factors have the greatest impact on the rating given by clients. Yang and Lau (2015) and Guo et al. (2017), using information from TripAdvisor, establish that service quality is the key factor in customer satisfaction. Other authors such as Choi and Chu (2001), Lockyer (2005) and Shergill and Sun (2004) expand the range of factors under analysis to consider a number of different aspects that travellers take into account when evaluating a hotel they have not previously visited, including service, location, price, room facilities and cleanliness. Mohsin and Lockyer (2010), Ban et al. (2019), Nunkoo et al. (2020) and Kim and Kim (2022) include factors related to the essential attributes of the hotel—facilities, rating, and food offered—as elements that contribute significantly to improving the client experience. Ananth et al. (1992) show that price and quality are the most important hotel factors, while Israeli (2000) finds that cleanliness is the highest rated attribute. Care for the environment is highlighted as a factor that influences client satisfaction in studies by Han et al. (2018), Millar and Baloglu (2011), and Robinot and Giannelloni (2010). Other factors related to the characteristics of the trip or the traveller that may influence the reviews given include the purpose of the trip (Wang et al., 2020), with reviews being more commonly used by business travellers (McCleary and Weaver, 1993); the traveller's country of origin (Chu and Choi, 2000; Bodet et al., 2017; Francesco and Roberta, 2019; Kim et al., 2019); and the demographic variables of the individuals (Spoerr, 2020).

Of all the factors highlighted in the literature, the one most often noted is service, particularly in situations such as the COVID-19 pandemic, where clients expect to receive more attentive service (Kim and Kim, 2022). However, it should be borne in mind that the importance of these factors changes over time (Bodet et al., 2017; Jang et al., 2018), particularly when the timeframe in question covers the emergence of a phenomenon as impactful at all levels as the COVID-19 pandemic. As such, it is interesting to see how these aspects have changed over the different stages of the pandemic.

Sun et al. (2022) analyse the changes in consumer satisfaction caused by COVID-19, coming to the conclusion that there are significant differences both in the overall ratings and in the ratings given by different types of travellers. Generally speaking, clients are more inclined to give higher ratings and pay more attention to hotel prevention and control measures aimed at reducing health risks after COVID-19. These changes in overall satisfaction are also reflected in the key factors prompting guests to leave reviews. In this regard, Song et al. (2022) find significant differences in hotel guest satisfaction before and after the COVID-19 outbreak; specifically, hotel guests register higher levels of satisfaction after the pandemic. The authors report that hotel service, the room, cleanliness, location, value and sentiment significantly affect customer satisfaction before the pandemic; however, the service no longer has a significant impact on hotel guest satisfaction after the pandemic. Nonetheless, Kim and Kim (2022) consider service to be a critical attribute for clients, especially after COVID-19.

Nilashi et al. (2021) reveal service quality to be a key attribute during COVID-19, and although customers always seek out high-performing hotels, they are also concerned about the quality of services related to the COVID-19 outbreak. In addition, guests' preference for hotels and rooms with access to or views of blue spaces has increased due to COVID-19, while at the same time there has been a rise in complaints about the smell of the room, bookings and cancellations. Ongsakul et al. (2022) find a change in tourists' preferences due to COVID-19.

Based on 866,048 online reviews of hotels in Brazil from the platform Booking.com, Peres and Paladini (2022) find that issues related to the cleanliness of rooms and check-in were the most negatively affected by the COVID-19 pandemic, registering the largest drops in the average rating. Comparing values before and after COVID-19, Kim and Han (2022) identify improved evaluations of the various hotel selection attributes, except for those relating to exteriors and the hotel lobby. This finding suggests that outward appearances are relatively less important, while the importance of COVID-19-related measures have become more important, which is logical. Although aspects such as cleanliness, hygiene and safety were not overlooked in the past, they have become more relevant since the outbreak of COVID-19. Xu et al. (2022) find that hotel attributes influence customer satisfaction differently pre- and post-COVID-19 and the attributes of staff, location, room amenities, and decorations have remained performance attributes. Moreover, their results showed that the effect of location increase and became the most important attribute post-COVID-19 and emerge attributes of staff and location.

This literature review has allowed us to confirm that COVID-19 has had a major impact on client evaluations and the factors that affect them; however, to the best of our knowledge, there have been very few studies to date that compare the situation at different points during the pandemic, and those that do are limited to comparing just two moments in time. In this paper, we address that limitation of existing studies in an effort to analyse the behavioural change and determine whether that change is momentary or is here to stay. In addition, the study allows us to convey information about the current situation to hotel managers so that they can undertake improvements as effectively as possible.

Bearing in mind the differing degrees of severity of the COVID-19 pandemic, its prevalence and progression in different countries (Wang et al., 2020) and cities, this analysis focuses on Cuenca (Spain). Cuenca is a city belonging to the autonomous community of Castilla-La Mancha that in 2022 had a registered population of 53,389 inhabitants. Cuenca is in the center of Spain, 168 km from Madrid, with which it has excellent communication. It is located across a steep spur, whose slopes descend into deep gorges of the Júcar and Huécar rivers. It is divided into two separate settlements: the "new" city is situated south-west of the old one, which is divided by the Huécar course. It is a very touristy city with numerous monuments to visit such as: Cathedral of Our Lady of Grace and Saint Julian; Bridge of Saint Paul; hanging houses; or the Old Convent of Saint Paul, among others, but it also has attractive nature and gastronomic tourism (more information about Cuenca is available at UNESCO (2023)). The city was declared a World Heritage Site in 1996, combining heritage, culture and nature, together with excellent gastronomy. As a consequence, Cuenca is a huge tourist draw and accordingly has a well-developed hotel industry, which means it can be taken as an indication of the overall situation in this Spanish industry. Specifically, we sourced information posted by people staying in Cuenca for a total of 17 hotels located in the city and the surrounding area, with the latter allowing us to examine the effect on a more rural type of tourism.

Regarding the stage of the pandemic for which we source information, we have selected four points in time based on the following considerations: it should be a time of year with a prevalence of tourism, since this type of travel is assumed to be most affected by the pandemic; and the points in time should capture the changes undergone due to the pandemic. We thus chose the month of July (99.3% of the observations obtained travel for leisure purposes); in the year 2019 (before the pandemic); 2020 (during the pandemic, but having opened up to the movement of tourists); 2021 (a time with a degree of recovery from the pandemic, the new normal); and 2022 (pandemic under control). We used the platform Booking to obtain the information since it has a system that guarantees the authenticity of the opinions: it only allows people who have made a reservation and completed their stay to submit a review, and it checks each review before publishing it.

In addition, travellers publish their positive and negative reviews separately, which makes it easier to identify guest satisfaction and dissatisfaction with the quality attributes of the hotel. Taking into account this distinction when collecting the information, a total of 1595 reviews have been obtained from people who have stayed in the 17 hotels under study. In addition, from the information provided in the review, we have included in the analysis attributes relating to the destination; the type of trip; the guest; the room; and the different aspects valued positively or negatively. Regarding the destination attributes, the destinations have been classified as rural or urban area, with urban indicating hotels located in the city of Cuenca and rural those located in the surrounding area. Regarding the type of travel, as noted above, we have focused on leisure trips as this type of travel has been more affected by the pandemic. Furthermore, since the type of traveller can influence the evaluation given to the hotel and the guest's satisfaction with it, we account for whether the guest was travelling alone, with a partner, with his/her family, or in a group.

Taking into account the client characteristics, and with reference to authors such as Chu and Choi (2000), Bodet et al. (2017), Francesco and Roberta (2019), Kim et al. (2019) and Spoerr (2020), the clients have been categorized by gender: men and women. Nationality has also been taken into account. Given the impossibility of analysing the more than 40 different nationalities, we have divided the nationalities into eight large groups; namely, Eastern Europe, Western Europe, North America, South America, Central America, Asia, Africa, and Oceania. However, due to the small number of observations in some cases, we also performed the analysis categorizing clients according to whether or not they were Spanish to check the significance of the results. Regarding the room, a total of 33 different room types were identified from the reviews. Therefore, the room type variable has been recoded to capture five types of rooms: single; double; suite; room with more than two beds; and apartment, bungalow or casa rural (a country house let out as whole self-catering house or by the room).

Finally, by means of a content analysis, we have detected a total of 86 positive and negative aspects highlighted in the guest reviews. Drawing on our review of the literature, these have been grouped into 12 broad categories covering aspects related to:

In addition, we also have information on the number of comments received, the number of nights of the stay, and the number of stars the hotel has, allowing us to analyse the effect that they have on guests' evaluation of their stay.

Regarding the methodology to be used, given that we have qualitative and quantitative information, we have chosen to use an Analysis of Variance (ANOVA). The aim of the ANOVA is to determine whether there are differences in the value of a quantitative variable for different levels of a qualitative one, although in the case of factors with two categories the result is the same as using a t-test. We have used the ANOVA considering one factor or two factors depending on the case. The Spearman correlation coefficient has been used to assess the relationship between quantitative variables and contingency table analysis with the chi-square test used to check the association between variables when the analysis is focused on two qualitative characteristics (Hair et al., 2013).

The distribution of the sample presented in Table 1 shows that the observations are evenly divided among the four stages of the pandemic considered, each with around 25% of the observations obtained. Given that most of the hotels are located in the city of Cuenca, we have more information for the urban area. We also see a greater presence of Spanish tourists, who make up 87.2% of the sample, making it unfeasible to carry out a detailed analysis of the effect of nationality on hotel evaluations. Furthermore, a majority of the travellers are making the trip as a couple, although there is a non-negligible number of observations in the other categories. Finally, as expected, 99.3% are travelling for leisure, so the observations of business travellers have been discarded since these trips are generally not as affected by the pandemic. Below, we analyse the available information using the different methods proposed.

Table 1. Sample characteristics.

The first aim of this paper is to determine whether there are significant differences in the mean evaluation scores given by guests at different points in the pandemic. To do this, we first conduct a test of the homogeneity of variances using the Levene statistic, in order to check which statistic we should use to carry out the ANOVA. In this case, the Levene statistic takes a value of 4.194 with a significance level of 0.006 (Table 2) so we reject the hypothesis of homogeneity of variances, and thus use the Welch statistic for the ANOVA, the results of which are shown in Table 2.

Table 2: ANOVA for one factor: pandemic stage.

Source: Own Elaboration.

The results in Table 2 reveal significant differences in the mean evaluation scores: while the values are high, the lowest values are registered at the time of the new normal, with a slight recovery in the post-COVID-19 era. These results show a drop in guest satisfaction—measured by the rating given to the hotel—due to the COVID-19 pandemic, although signs of recovery seem to be emerging, registering values above 8 in all cases. As can be seen in Table 3, significant differences in the means appear if we compare the situation before COVID-19 with any time after it, revealing the effect that COVID-19 has had on the mean evaluation as an indicator of the degree of client satisfaction.

Table 3: Bonferroni post hoc test.

Note: * Significative difference at 5% level.

According to the literature review, aspects that may affect the evaluation of hotel establishments include the size of the city; the type of room; and traveller characteristics such as traveller type, sex and nationality. Therefore, we perform two ANOVAs considering the time the information was provided and the other aforementioned aspects in order to analyse the effect that these factors have on the evaluation (Tables 4 to 8).

Table 4: ANOVA for two factors: pandemic stage and urban/rural area.

The results in Table 4 show that in the case of urban/rural area there are no significant differences in the mean evaluation when jointly analysing the point in the pandemic and the urban/rural area, whereas at a 10% significance level there are differences in the values depending on the urban/rural character, with higher values in the urban area, and smaller differences in July 2021 with the new normal.

Table 5: ANOVA for two factors: pandemic stage and type of room.

Table 5 shows differences significant at the 6% level in the mean evaluation according to room type and the time of the pandemic, with better evaluations for single, double rooms and suites (8.126; 8.248; and 8.678, respectively), and worse evaluations in the case of rooms with more than two beds or apartments and bungalows (8.047 and 7.180, respectively). These differences are greater after the pandemic: in July 2022, the difference between the best rated suites with a score of 8.867 and the worst rated bungalows with 5.714 is 3.152; at the post-COVID-19 stage in 2021 that difference was one point, and 1.296 during COVID-19 and 1.396 before the pandemic. Therefore, the effect of the pandemic on consumers has led them to assign more value to rooms with a restricted number of occupants and those that offer higher quality standards, such as suites.

Table 6: ANOVA for two factors: pandemic stage and traveller type.

Table 7: ANOVA for two factors: pandemic stage and gender

Table 8: ANOVA for two factors: pandemic stage and nationality.

Finally, Tables 6, 7 and 8 show no significant differences when accounting for traveller type, gender, or nationality. One result we can highlight is that women with a score of 8.285 and foreign guests with 8.419 give hotels a better evaluation at different points in the pandemic. Also, in July 2021, after the toughest part of the pandemic, group travellers gave the best evaluation, whereas at other points in the pandemic it was solo travellers.

Having analysed the effect of different factors on the evaluation given to hotel stays, we examine the relationship between that evaluation at different times of the pandemic and variables such as the number of reviews, number of hotel stars and the number of nights stayed (Table 9).

Table 9: Correlations considering the valuation in different pandemic stages.

The results in Table 9 reveal significant and positive relationships between the number of hotel stars and the evaluation given in the new normal and post-pandemic periods. As such, it can be claimed that the COVID-19 pandemic has increased the positive effect that the number of hotel stars has on the evaluation. Regarding the number of reviews, although the relationships are not significant, we see a change in trend caused by COVID-19. In July 2019 and 2020, the more comments made, the worse the evaluation; however, this relationship is positive in the new normal and after COVID-19. Finally, accounting for the number of nights, we see a negative relationship in July 2022 while before it was positive, which means that an increase in the number of nights leads to a worse evaluation of the stay, perhaps due to the monotony or having more time to pick up on small details.

Another key feature of this study is the possibility of identifying the factors that cause the greatest guest satisfaction, meaning they are evaluated positively, and those that cause dissatisfaction and are given a negative evaluation. The results shown in Table 10 reveal a significant association between when the information was submitted and the aspects noted, both positive and negative, showing Pearson Chi-square coefficient values of 74.579 for the positive aspects and 65.439 for the negative aspects, both significant at 5%.

Broadly speaking, the results show the relevance of the services offered by the hotel for both guest satisfaction and dissatisfaction, since they present the highest percentages at all points in the pandemic. Going into more detail, it can be seen that, in addition to services, elements related to cleanliness were key among the positive aspects highlighted by guests before the pandemic (Table 10). During the pandemic, in addition to cleanliness, another element identified as a positive attribute by customers is the location, with guests paying little attention to aspects such as noise, price, hotel stars, or hotel condition. Also striking is the small number of people who point out aspects related to COVID-19 among either the positive or negative attributes.

Focusing on the negative aspects (Table 10) highlighted by customers, prior to the pandemic the extras offered by the hotel were a clear source of guest dissatisfaction, whereas at the height of the pandemic and during the subsequent recovery, key aspects related to room characteristics are the elements that attract negative comments from guests. In addition, it is noteworthy that aspects such as price, noise or cleanliness are rarely specified as sources of guest dissatisfaction, whereas cleanliness is highlighted by customers as positive aspects.

Lastly, an analysis of the positive aspects accounting for gender reveals that when separating the sample by gender there are no significant associations between the point in the pandemic and the positive and negative aspects. However, when differentiating between rural and urban areas, these differences are significant in urban areas.

Table 10: Contingency table between pandemic stage and positive and negative aspects.

The outbreak of an unprecedented pandemic such as COVID-19 necessitates a detailed analysis of the changes in both hotel evaluations and in the hotel attributes reviewed positively and negatively by guests after their stay in the hotel. However, the review of the literature reveals a gap in this regard, especially when considering different points in time over the course of the pandemic. This paper thus attempts to help fill this gap.

The results revel that the lowest values of evaluation scores are registered at the time of the new normal, with a slight recovery in the post-COVID-19 era. This result is similar than reported by Raedts et al. (2023) and Xu et al. (2023) but contradicts the findings by Mehta et al. (2023) in North America and by Song et al. (2022) in Chengdu (China). Those authors state that customers were more satisfied with hotels after the outbreak of COVID-19 as, despite some signs of recovery, pre-pandemic levels have not yet been reached, let alone exceeded. That said, as in the aforementioned studies, the levels of satisfaction shown by guests before, during and after COVID-19 are fairly high. Significant differences appear if we compare the situation before COVID-19 with any time after it, revealing the effect that COVID-19 has had on the mean evaluation as an indicator of the degree of guest satisfaction. Therefore, considering the previous studies and the result obtained, it is possible to see that the hotel guest’s evaluation registers a recovery after the pandemic but still far from pre-pandemic levels. So, it is necessary to continue with the efforts developed to recover the positive evaluation of consumers that has surely been improved in 2023.

In addition, we have considered aspects that, in the best of our knowledge, have no analysed in the literature. In this sense, we consider the characteristics of the destination, differentiating between rural and urban areas without significant differences in the mean evaluation when jointly analysing the point in the pandemic and the urban/rural area, whereas at a 10% significance level there are differences in the values depending on the urban/rural character. Maybe this must be analysed considering information in a wider geographic area that allows to have more differences between urban and rural areas because in this paper the rural areas are related with the surrounding area and it is not a rural area properly said although there are not studies to compare the results. Moreover, we have considered the effect of the type of room over the guest evaluation. In this case, although authors such as Choi and Chu (2001), Lockyer (2005), Shergill and Sun (2004), Stringam et al. (2010), Song et al. (2022) or Peres and Paladini (2022) find aspects related to the room such as room facilities; room cleanliness; or room comfort have effect over guest evaluation, none consider the type of room. The results show better evaluations for single and double rooms, and worse evaluations in the case of rooms with more than two beds or apartments and bungalows showing that the pandemic has led consumers to assign more value to rooms with a restricted number of occupants and those that offer higher quality standards, such as suites.

COVID-19 has increased the positive effect that the number of hotel stars has on the evaluation, with better guest evaluations found for hotels with a higher number of stars. In this sense, Xu et al. (2023) find that the positive relationships between COVID-19 pandemic and tourist ratings for hotels is weakened if hotels have a high star level. We also see better evaluations in hotels that have received more reviews, whereas the opposite was the case before the pandemic. In addition, there is a negative effect of the number of nights on the evaluation after the pandemic, although the relationship is not statistically significant.

Considering the traveller characteristics, we find no significant differences when accounting for traveller type, gender, or nationality. These results contradict those reported by Chu and Choi (2000), Bodet et al. (2017), Francesco and Roberta (2019), Kim et al. (2019) and Spoerr (2020), who found differences according to traveller characteristics. One result that we can highlight is that women and foreign guests give hotels a better evaluation at different points in the pandemic, whereas Yang et al. (2023) reported that the effect of the pandemic was more pronounced in men. Also, in July 2021, after the toughest part of the pandemic, group travellers gave the best evaluation, whereas at other points in the pandemic it was solo travellers. Therefore, contrary to what some authors have reported in the literature, guests’ sociodemographic features do not have a significant effect on the evaluations they give to the accommodation.

Finally, the factors identify as key for a positive evaluation (satisfaction) or negative (dissatisfaction), broadly speaking, are similar to that reported by Nilashi et al. (2021). Among the positive aspects highlighted in guest reviews, services emerge as a key factor in customer satisfaction at all points in the pandemic except the post-pandemic period, at which point factors such as location or cleanliness become key. Yang et al. (2023) also confirmed the importance of cleanliness, being more important in the place where the pandemic situation was more difficult. In relation to the negative aspects, the attributes related to hotel services clearly stand out as being the cause of greatest dissatisfaction among hotel guests. It should be noted that after the pandemic, the room characteristics predominate in terms of negative aspects highlighted, whereas before the pandemic it was the hotel extras. Therefore, services are a key factor in customer satisfaction and dissatisfaction in all the points in the pandemic.

The main contribution of this paper is the analysis considering different stages of the COVID-19 pandemic, concretely we have considered four stages of the pandemic: before the pandemic (July 2019); during the pandemic (July 2020); a period of recovery from the pandemic (the "new normal") in July 2021; and a time when the pandemic seems to be over (July 2022). These four stages allow us to determine the changes in the hotel guest evaluation in general and considering different the traveller characteristics such as the nationality or gender. Moreover, we have used two characteristics that they have not been considered in the literature: the room type and the urban/rural destination. In addition, we analyse possible changes in the key aspects of the evaluations given, and the possible appearance or prevalence of certain factors, caused by the COVID-19 pandemic.

The guests’ evaluations of tourist accommodation had gone down with the COVID-19 pandemic but now is recovery. The guests’ sociodemographic characteristics do not affect on the hotel evaluation. The urban/rural distinction exerts an influence and the accommodation with lower capacity to obtain a better evaluation, COVID-19 increases the relationships between the hotel evaluation and the hotel stars.

The pandemic supposes that services emerge as a positive key factor in the guest evaluation at all pandemic stages except the post-pandemic period when location or cleanliness are keys. It is also the attributes related to hotel services the main cause of dissatisfaction.

In light of all these results, we can conclude that the effort made by the hotel managers to return to normality is being valued positively by clients, with a gradual increase in the evaluation scores given by hotel guests, and we hope to see this trend continue over time.

Apartments and bungalow owners should pay special attention given that it is the guests of these types of establishments who give worse evaluations to the stay. This situation is clearly affected by the presence of Covid-19 since the average rating of this type of establishments before COVID was 7.929 and post-COVID 5.719. This difference may be because perhaps in the hotel rooms the adoption of preventive measures for COVID-19 could be seen more clearly than in the apartments or bungalows.

Hotel managers must enhance attention to national guests since, although without significant differences, they are the ones who valuate worse the hotel stay. Moreover, with the new situation after Covid, they must pay attention to guests with longer stays since longer stay suppose a decrease in the evalutaion. Therefore, they must encourage measures that promote non-monotony in the stay and ensure that the positive perception does not decrease as the days go by.

Furthermore, it should be noted that, as established by authors such as Choi and Chu (2001), Lockyer (2005), Stringam et al. (2010), Yang and Lau (2015), Guo et al. (2017), Nilashi et al. (2021), and Kim and Kim (2022) among others, and contrary to what has been reported by Song et al. (2022), service was and continues to be a key factor in the level of customer satisfaction or dissatisfaction detected from the positive and negative aspects highlighted in guest reviews. Therefore, hotel managers should focus their efforts on the aspects related to the hotel services, if they want to improve the level of guest satisfaction, given the positive effects on the establishment that this has.

Finally, the results seem to show that, although the pandemic has entailed some changes in the attributes highlighted by guests, the return to normality seems to have brought us back to the starting situation before the pandemic, although this adjustment is going to happen gradually, with domestic guests being the most demanding when it comes to giving the hotel a good evaluation.

This study represents an initial approach to the analysis of an issue as complex as the effect of the pandemic on hotel establishments, bearing in mind that the pandemic has not affected all places or all people equally. The main limitations of the study are related with the information available and the geographical location in the city of Cuenca but there is not available information source with more information and for more cities, although it is admittedly difficult to generalize due to the characteristics of the pandemic and its disparate effects. In this regard, this study opens up new lines of future research, in which it would be worth collecting a larger sample with greater spatial and temporal coverage to yield more generalizable results. Concretely, analyzing if the final of the pandemic situation has improved the guest evaluations and has changed the key factors, although now a new handicap appears that we must considered related with the inflation crisis provoked by the Russio-Ukranian war.

This research was partially funded by the Department of Business Administration, University of Castilla-La Mancha, Spain and it is part of the research project entitled “Acceptance and implementation of Intelligent Marketing Instruments (IMI) in the Spanish tourism system” funded by TUR-RETOS2022-035 of the SET of the Ministry of Industry, Commerce and Tourism with European Funds NextGeneration EU.