This research investigates quantitatively by means of a survey among 137 often frequent retail visitors to four and five-star hotels in Thailand, factors that influenced their hotel choices. The perceived acceptable price range of the visitors is analyzed via questions from the van Westendorp’s Price Sensitivity Model. Issues relating to the extraction of an optimal revenue forecast from the model are discussed. The research provides a ranking of the importance of amenities for booking four and five star hotels in Thailand and price ranges that guests expect. Sustainable practices do not play an important role in the decision to book a luxury four or five-star hotel.

Keywords: Luxury Hotels, Amenities, van Westendorp’s Price Sensitivity Measurement Model, Room Rate Expectations, Hotel Revenue Forecast, Thailand.

INTRODUCTION

               Tourism is an essential source of income and employment for Thailand. The luxury segment plays an important role here. Hoteliers need to be able to understand enough of the decision process of tourists (Cohen, Prayag & Moital, 2014) in order to be able to offer tailored solutions to this demanding segment of the market. Tran (2015) gives an overview of research done on demand in the luxury segment in different countries, but Thailand is missing in his overview. One of the goals of the present research is therefore to fill a gap in the understanding of expectations of tourists visiting four- and five-star hotels in Thailand by exploring what kind of amenities they expect. Consumers always make a trade-off between nicer amenities and higher prices. Therefore, this paper also researches the sensitivity of the consumer to hotel prices in Thailand. This has been recognized as very important in other settings as well (Lu et al., 2015). This research contributes to an in-depth understanding of these issues by using van Westendorp’s (1976) Price Sensitivity Model.

This research was carried out by analyzing in a quantitative way responses to an online survey in Thai and English among 137 individual visitors in a convenience sample to four and five-star hotels in Thailand. All obtained results are highly significant, showing that the sample size was adequate. Respondents were mostly leisure travelers, not part of groups and booked their hotels individually.

               Further it is researched what role prices and some other factors play in the willingness to recommend a particular property.

LITERATURE REVIEW AND SURVEY DESIGN

               This research focuses on three related objectives:

•     Factors influencing consumer hotel choice, especially amenities, as part of the decision process of potential guests.

•     Traveler’s acceptable price range of 4- and 5- star hotels in Thailand, and

•     An indication about hotel revenue.

Factors Influencing Consumer Hotel Choice

               According to Ogüt & Tas (2012), star ratings are able to provide an advantage of differentiation to hoteliers as ratings offer an indication of intrinsic value. In Thailand, the Thailand Standard Hotel Foundation is the only organization that can decide which rating is assigned to each individual hotel or resort. Their Thailand Standard Hotel Directory (2011) describes characteristics of four- and five-star hotels.

               According to Chen & Jones (2011), guests use a two-stage hotel selection progression deciding to book an accommodation online. They first establish a consideration set, followed by a smaller choice set. Different aspects and features of the hotel or resort and the scope of the hotel market in the considered area itself affect these sets. However, researchers still do not agree on which factors are paramount in influencing consumer hotel selection and price ranges. It might be even doubtful if there is a common set of factors that is valid for most travelers. Hotel choices might also be influenced by gender, age, income and even purpose (e.g. leisure, business, etc.). Studies from Lewis (1984); Knutson (1988); Ananth et al. (1992); Hart (1993); McCleary, Weaver & Lan (1994) and Callan & Bowman (2000) identified attributes that can affect hotel selection and decision making. Reasonable cost or price, location, security, star ranking, service, hotel amenities and status were considered as main factors in selecting a hotel accommodation. Lewis (1984) determined attributes of three different areas: hotel selection, hotel stay, and perceptions. As a result, the study came up with 66 attributes, reduced to 17 characteristics to make the analysis more manageable. Ananth et al. (1992) evaluated 57 hotel variables that guests might consider when selecting a hotel, including such factors as good value for money, a swimming pool, breakfast and so forth. Callan & Bowman (2000) rated 38 factors on their importance when choosing a hotel accommodation. According research by the Global Market Metrix Hospitality Index (MMHI) in 2013, location is one of the top factors when deciding to book a hotel accommodation, followed by price. Given those different results, it is important to supplement this type of research with findings from a different country like Thailand.

               Following the quoted existing scientific literature, the survey contained a large number of questions with obvious face validity on features that were discussed above as to influence booking decisions. Questions were asked about age, income, education, previous experiences with 4 and 5 star hotels, facilities, services, amenities, location, hedonistic aspects, unique selling points, loyalty and green programs, competitors, online reviews about quality, if the hotel was associated with a chain, prepayment against no prepayment, and free cancelation options. The complete survey is given in Bauer (2017).

               Somewhat opposite to the usual approach, this work posits that hotel guests have realistic expectations about amenities that are determined by the price that they want to pay and minimum standards, and not, reversely, that desired amenities determine the price that one is willing to pay. Therefore, the survey first asks for an acceptable price range for the room rate for four and five star hotels in low and high season (see the next section on van Westendorp’s PSM). Given that price range, it asks what are important “must-have” amenities that one expects, to get, considering the price one is willing and expecting to pay. In that way, this work links expectations about costs in a hotel category and desiderata for amenities per hotel category.

The van Westendorp Price Sensitivity Model

               Many factors influence price in the hotel business and value for money is closely linked to perceived quality, expectations and customer satisfaction. A guest will agree to a specific price when the perceived value of the product or service matches the price tag of this product or service. Reversely, the price one is willing to pay sets the expectations about what one can expect for that price. As guests’ perceptions differ tremendously, also depending on their background, hoteliers have to evaluate carefully when setting the price for a hotel accommodation. (Plessis & Saayman, 2011). 

               Various firms still base changes in their pricing not on predictions or expectations of the consequences of a changed price, but on customer reaction after the change occurred. Product-led pricing strategy instead of a customer-led pricing strategy is still common. This can make the price level too high as well as too low for the services offered. Different strategies exist to mitigate this problem, like demand-based pricing, pricing according to distribution strategy (sales channels), dynamic pricing (aiming to have a profit at all time by varying the prices on a short-term scale), open pricing (flexible pricing) and static pricing (fixed price year around).  There seems to be a trend to shift pricing strategies to customer-led pricing. This article discusses one way the view of customers can be analyzed.

               Price ranges were first explored by Gabor & Granger (1966). In the consumer’s view, a price is both an indicator of cost and an indicator of quality. Gabor & Granger’s work considered upper and lower limits to a possible purchase. Any price that exceeded the upper price limit would be labeled as too expensive and a price below the low limit would be a signal of too low quality. van Westendorp (1976) extended this in his Price Sensitivity Model (PSM) (not to be confused with price elasticity). To establish price sensitivity, potential guests were being asked the following questions in the survey. The following PSM questions were adjusted for the hotel industry with conjoint questions for a yield analysis, but are following the validated questions from van Westendorp’s work:

1)   (Too cheap price): At what price on the scale do you consider a 4 and 5 star accommodation too cheap, so cheap that you would question the quality?

2)   (Cheap, Good Value): At what price on the scale do you consider a 4- and 5-star accommodation a good value?

•     Conjoint Question: How likely would you book a room at this good-value price?

3)   (Expensive): At what price on the scale do you consider a 4- and 5-star accommodation to be getting expensive, but you would still consider booking it?

•     Conjoint Question: How likely would you book a room at this more expensive but still for you affordable price?

4)   (Too expensive): At what price on the scale do you consider a 4- and 5-star accommodation too cheap, so cheap that you would question the quality?

               The van Westendorp’s price sensitivity “price map” shows on the x-axis the range of acceptable prices for a product. The y-axis represents the cumulative percentage of respondents. By construction, the “too cheap” line is to the left of the “cheap” line and the “too expensive” line is to the right of the “expensive” line. Any prices that are lower or higher than the “too cheap” and “too expensive” line will be not considered by the indicated percentage of consumers. The Optimal Price Point (OPP) is the point where the same number of consumers considers the product as too cheap as too expensive. A few researchers, such as Grigsby (2015), argue that the optimal price point (OPP) might be debatable, however the general idea is that an equal amount of respondents believe that a product is either “Too cheap” or “Too expensive”, therefore it is the point where purchase resistance is the lowest, hence the most optimal yield. In general, its location is not very far from the Indifference price point, the point where an equal number of people find the product cheap and expensive. Note that costs for the company, for instance advertising to justify a higher price, can be variable and dependent on the sales price. Therefore, the OPP is not automatically also the point of the highest profit. At the Point of Marginal Cheapness, the same number of people perceives the price as too cheap as expensive. At the point of Marginal Expensiveness, the same number of people considers the price too expensive and cheap. The Range of Acceptable prices is between the Point of Marginal Cheapness and Point of Marginal Expensiveness (roughly 1200 to 1900 in Figure 1). Outside this range there is not an optimal amount of demand. If the price is lower than the lowest price in the Acceptable Price Range, revenue is lost. If the price is higher than the highest price in the Acceptable Price Range, sales are lost. According to Lewis & Shoemaker (1997), the price sensitivity measurement model might not be 100% valid; however, it is very useful when determining an acceptable price that is based on a consumer point of view. The yield can be calculated if at every price the probability is known that a customer will actually make the purchase (Salamandica, Alijosiene & Gudonaviciene, 2014; Lieberman, 2015). This can be done in a conjoint analysis. The survey questions are then an example of a discrete choice model with added questions for a conjoint analysis. The conjoint likelihood questions in this research were not formulated for a specific timeframe, and therefore the revenue provide a lifetime estimate of revenue from the sample group for a one-time return for one night. The chance that a previous visitor returns to Thailand depends on factors outside the control of the hotel industry, and respondents will find it difficult to predict accurately if they will return and at what time. However, the stated conjoint questions provide still informative information about the influence of the price on the likelihood to book and on expected revenue.

Amenities and Prices as Factors Influencing Choices

               According to Ananth et al. (1992), the importance of price in the lodging industry is not significantly different amongst diverse age groups. However, it is generally assumed that younger travelers are more price sensitive because of lower disposable incomes. Guillet, Guo & Law (2015) classified generation Y travelers (18 to 35 years old) as price sensitive. Petrick (2005) (as cited in Öğüt & Tas (2012), stated that less (high) price sensitive consumers generally select high (low) star hotel accommodation. In addition, there are also some hotel features or amenities that might be more important to younger hotel guests than to more mature travelers, like sports accommodations. McCleary, Weaver & Lan (1994) found that women who travel for business, take hotel safety measures, low price, and more personalized services more into account than men. Furthermore, women traveling for business attach presumably more value to their overall accommodation booking when necessary amenities (e.g. hair dryers, iron and ironing board, room service and bathrobes) are offered. To explore some of those issues, a large number of questions were included in the questionnaire. 

METHODOLOGY

               The survey was carried out online among a convenience sample (a random sample is impossible to obtain) of frequent Thai and foreign visitors to 4 and 5 star hotels obtained by seeding a number of guests to such hotels the first researcher knew. Respondents were requested to forward the link to qualifying acquaintances in their network by snowballing (chain referral sampling) via email and social media like Facebook and LinkedIn. A filter question assured that only visitors in the last three years to 4 and 5 star hotels participated. The full survey and all technical and statistical details are available in Bauer (2017).

               Respondents were presented with vignettes describing 4- and 5-star accommodations. After this, respondents were asked structured questions about demographics, purpose of stay in the last visited 4- or 5-star hotel accommodation, channel of booking and their behavioral (intention of returning) and affective (willingness to recommend to friends and family) loyalty to a particular hotel provider. Questions also probed acceptable room rates for four- and five-star hotels and the respondents’ willingness to book the hotel.

               The online questionnaire was designed and distributed in a Thai and English version electronically via the Qualtrics system. The complete questionnaire is in Bauer (2017). As an incentive, respondents could opt-in to participate in a raffle and have a chance of winning a restaurant voucher. The IRB board of Webster University in St. Louis, MO, USA approved the research. The survey data was collected over a two months’ period, starting at the beginning of November 2016 and ending mid-January 2017. At the end, 137 respondents completed the survey. Data analysis was done with SPSS.

RESULTS

               There were 70 female respondents (51%) and 67 male respondents (49%) (Total N=137) in our sample. 65 respondents (55%) were between 18 and 35 years old, “younger travelers”, “Generation Y” and 62 (45%) were 36 or older, “mature travelers”, “Generation X”. Of the mature respondents only 7 (5.1%) were 56 to 65 years old and none were older. Many respondents were from Thailand (N=72, 52%). The second largest group was from Europe (28%). British and French nationals formed each 22% of the group of Europeans and Germans 16%. Up to a certain degree, this matches the known distribution of visitors to Thailand, justifying the convenience sample. International visitors to Thailand are mostly from East-Asia, but our sample contains Thai domestic visitors instead Chinese international visitors. The second large group of visitors to Thailand comes from Europe, just like in this sample (Bauer, 2017). Of the respondents, only 6 (4%, N=137) had not completed college. 67 (48.9%) had as highest education a bachelor degree, 57 (41.6%) a Master’s degree, 4 (2.9%) a Doctor’s degree and the rest, 3 (2.2%) a vocational or professional degree. 95% of the respondents reported that their income came from salary. Figure 2 (N=134) indicates the frequency distribution of income. The respondents are clearly affluent. For privacy reasons it was not asked if respondents had paid for themselves.  115 respondents or 84% of total respondents (N=137), stated that their primary purpose of visiting 4 and 5 star hotels was for leisure reasons and 14% were business travelers.

               The relative importance of factors that could influence hotel choice and final decision-making process are shown in Figures 3 and 4. The error bar indicates the statistical uncertainty; it is impossible to estimate the systematic uncertainty. The image of the hotel scores considerably higher in the relative ranking of importance of amenities in five star hotels than in four-star hotels. It is probably a fair conclusion that survey participants want to appeal to other people when staying in a five-star hotel and impress them.

Online reviews from websites like Booking.com, Agoda or Pantip are slightly higher rated than TripAdvisor reviews on the respondent’s importance scale. Maybe participants found it simpler to read reviews and simultaneously book a hotel room at these websites. Another reason might be that 72 respondents were Thai citizens and Pantip.com, a Thai website similar to TripAdvisor, is a common destination for Thais for hotel (and other) reviews.

               In this sample “green”, environmentally friendly programs ended dead last in importance for booking a luxury 4 or 5 star hotel. This does not mean that it is not important for hotels to adopt sustainable practices. For instance, there were no questions to probe to what degree sustainable practices during the stay influenced the satisfaction after the stay.

               In some respects, there were significant differences between generation X (mature) and generation Y (younger) travelers. Generation X had stayed on average somewhat more than five times in a four-star hotel and on average around five times in a five-star hotel in the last three years. In the last three years, generation Y has stayed less, between three and four times, in a four-star hotel and around three times, on average, in a five-star hotel. This might reflect the lower disposable income of the generation X travelers, although average number of stays was skewed by the 20% of the respondents who had visited a four-star hotel more than 10 times in the last three years and the 15% who had visited a five-star hotel more than 10 times. The relative large number of very frequent visitors might be a target selection effect. However, there is no reason to assume that priorities are different among this group.

               Both genders gave similar scores to the question about ‘too cheap’ from the van Westendorp Price Sensitivity Model for a 4- and 5-star accommodation. However, female respondents indicated slightly higher price ranges than male participants regarding the ‘good value’, ‘expensive but still affordable’ and ‘too expensive’ value of a 4- and 5-star accommodation in low and high season in Thailand. This is in disagreement with the research by McCleary, Weaver & Lan (1994), who stated that women take into account low prices when booking a hotel accommodation (see the literature review). A possible explanation is the observation from the literature review that women care more about personalized services and various amenities. Women considered the following hotel factors significantly (via a T-test) of higher importance when booking a hotel accommodation. As General Items: Online photos and videos and green programs. For Amenities: special pillows, microwave, electric kettle, tea and coffee; hairdryer, toothbrush; shower caps, combs, and additionally for a five-star hotel, bathrobe and slippers. As Services: Concierge, Valet and 24 h Front Desk Service, and additionally for a five-star hotel: 24 h Room Service. It is therefore plausible that women’s upper part of the price range is higher than men’s is, as female respondents seem to expect more value. Interestingly, women also find green programs more important than men do, although this still scores low.

               A factor analysis (principal component analysis) was done for each of the four variables coming from the van Westendorp questions (“too cheap”, “good value”, “getting expensive” and “too expensive”) (Table 1). It was found that only one factor per question was extracted. Those gave a good representation of the variability in the questions over low and high season and four and five star hotels. Therefore, each of the four van Westendorp variables was combined for low and high season and 4 and 5 star hotels (Bauer, 2017) for use in regression equations. After this, a regression analysis was performed. Forward additions of independent variables were carried out in all cases.

               The following tables (Table 2) show some of the results:

Table 2 shows the regression analysis for dependent “Likelihood to stay in a good value hotel”. There are statistically significant influences from income, ‘good value’ price and the likelihood to ‘search for a better option’. An explanation is that respondents who answer that they are likely to stay in such good value lodging are price conscious. 

               The likelihood to stay in ‘getting expensive' option (Table 3) was influenced by impressing other people, loyalty program, beach club, bathrobe and slippers. It is an interesting finding, relevant for marketing that respondents are trying to impress their friends and other people with their hotel choice. The number of stays in 5 star hotels was influenced by income, likelihood to ‘search for a better option’ and ‘too cheap’ price (Table 4). The two independent variables seem contradictory but consistent with the signaling of affluence. As five star hotels are expensive, price starts to be an influencing factor for 5 star hotels as salary is influencing their acceptable price range and respondents are likely to book the accommodation with best available price in this star category. On the other hand, it is important that the hotel is not too cheap, because then it loses its signaling function. There is a trade-off between the desire to impress people and the reality of a finite budget.

 A standard way to gauge the emotional attachment to a service or brand (“affective loyalty”) is to ask if the respondent is willing to recommend the service or brand to other people who are important to the respondent. Table 5 confirms that loyalty programs indeed influence affective loyalty and make visitors more attached, but a perception of value for money also increases the likelihood of recommending. These results complement the factors derived from marketing practices given in Narteh et al. (2013). The analysis of the survey questions relating to van Westendorp’s Price Sensitivity Model is given in Figures 5 and 6. The x-axis is the price for one night in Thai baht. The acceptable price range outcomes found here are in line with the data recorded from Thornton Thailand Survey (2016) that room rates at 5-star properties were 3,642 baht (≈ $106) on average and at 4-star hotels 2,468 baht (≈ $72) on average. This research confirms that regular visitors to luxury hotels have realistic expectations about price levels and are willing to pay for increased quality or a higher star rating. Other survey questions asked about the importance in 4 and 5-star hotels for amenities, services and facilities. Some important factors for respondents’ hotel choice were personal care items, in-room WI-FI, 24 h front desk service, swimming pools and dining options. Some low importance hotel features included microwave, ice bucket with ice, valet service and a business center. This data and many other results are reported in (Bauer, 2017). This research also found that there is a significant difference in the frequency of stay in 4 and 5 star hotels and the importance of hotel features between generation X and Y travelers, but no difference in price ranges. 

DISCUSSION AND RECOMMENDATIONS

               This research provides a detailed baseline about expectations of visitors to four and five star hotels in Thailand. This research found that generation Y and generation X, and male and female guests, differ in their ideas of which hotel features are the most desirable, pointing to a need for specialization. More details about this aspect are in (Bauer, 2017).

               Some other recommendations that follow from this research are: Hoteliers need to be cautious before following the trend to offer lower rates. This can lead to a “Red Queen’s race”, a race to the bottom. Hoteliers of high-quality properties have to find a strategy that is not only dependent on the lower-spending market segment, but attracts different and higher quality tourists. Therefore, it is useful to measure the overall consumer acceptable price ranges for these star categories as was done in this work. This research showed that 4-star hotels could set their price above 50 USD (1,800 baht) in low season and even consider setting their price above 85 USD (3,000 baht) in high season for leisure travelers. Moreover, 5-star properties can specify the room rate above 100 USD (3,500 baht) in low season and above 140 USD (5,000 baht) in high season. This research has shown that hoteliers should be careful when driving room rates down, because customers might see the offered price as too cheap, and hence suspect that the offering is a low-quality product or service. Several parts of the research confirm the signal function that five star hotels have to message affluence and ability to afford opulence to members of the social class of the traveler, friends and family.

               Surveys like this one give tools to satisfy the customer by listening to their wishes and opinions and make it possible to establish the right price point and offer the right amenities and services, for the right customer at the right time.

Ananth, M., DeMicco, F.J., Moreo, P.J. & Howey, R.M. (1992). Marketplace lodging needs of mature travelers. Cornell Hotel and Restaurant Administration Quarterly, 33(4), 12-24.

Bauer, I. (2017). A study of the factors influencing consumer hotel choice and acceptable price range in Thailand's hospitality industry. Retrieved from https://drive.google.com/file/d/0BzxLn3vefR8EdEI4d3lsa1J4eWs/

Callan, R.J. & Bowman, L. (2000). Selecting a hotel and determining salient quality attributes: A preliminary study of mature British travellers. International Journal of Tourism Research, 2, 97-118.

Chen, M. & Jones, P. (2011). Factors determining hotel selection: Online behaviour by leisure travellers. Tourism and Hospitality Research, 11, 83-95.

Gabor, A. & Granger, C.W.J. (1966). Price as an indicator of quality: Report on an enquiry. Economica, (33), 43.

Grigsby, M. (2015). Analytic choices about pricing insights. Marketing Insights, 38-43.

Guillet, B.D., Guo, Y., & Law, R. (2015). Segmenting hotel customers based on rate fences through conjoint and cluster analysis. Journal of Travel & Tourism Marketing, 32, 835-851.

Hart, W. (1993). What women want. Cornell Hotel & Restaurant Administration Quarterly, 34(5), 10.

Knutson, B.J. (1988). Frequent travelers: Making them happy and bringing them back. Cornell Hotel and Restaurant Administration Quarterly, 29(1), 83-87.

Lewis R.C. & Shoemaker, S. (1997). Price-sensitivity measurement: A tool for the hospitality industry. Cornell Hotel and Restaurant Administration Quarterly, 38(2), 44-54.

Lewis, R.C. (1984). Isolating differences in hotel attributes. The Cornell Hotel and Restaurant Administration Quarterly, 25(3), 64-77.

Lieberman, M. (2015). Pricing research: A new take on the van Westendorp model. Retrieved from http://www.quirks.com/articles/2015/20150625-2.aspx 

Lu, C.Y., Berchoux, C., Marek, M.W. & Chen, B.T. (2015). Service quality and customer satisfaction: Qualitative research implications for luxury hotels. International Journal of Culture, Tourism and Hospitality Research, 9(2), 168-182

McCleary, K.W., Weaver, P.A. & Lan, L. (1994). Gender-based differences in business travelers' lodging preferences. Cornell Hotel and Restaurant Administration Quarterly, 35(2), 51-58.

Narteh, B., Agbemabiese, G.C., Kodua, P. & Braimah, M. (2013). Relationship marketing and customer loyalty: Evidence from the Ghanaian luxury hotel industry. Journal of Hospitality Marketing & Management, 22(4), 407-436.

Ogüt, H. & Tas, B.K.O. (2012). The influence of internet customer reviews on the online sales and prices in hotel industry. The Service Industries Journal, 32(2), 197-214.

Petrick, J. (2005). Segmenting cruise passengers with price sensitivity. Tourism Management, 26, 753-762.

Plessis, E. & Saayman, P. (2011). Factors influencing pricing in the accommodation sector in South Africa. 4, 24-47.

Salamandica, E., Alijosieneb, S. & Gudonavicienec, R. (2014). Price sensitivity measurement depending on brand awareness: A case of Ziede brand. Procedia - Social and Behavioral Sciences, 156.

Thailand Standard Hotel Directory. (2011). Tourism authority of Thailand. Retrieved from http://www.tourismmekong.org/wp-content/uploads/Thailand-Standard-Hotel-Director.pdf

Thornton, G. (2016). Thailand hotel survey. Retrieved from http://www.ntccthailand.org/images/articles__reports/Grant-Thornton-Thailand-Hotel-Survey-2016.pdf

Tran, X.V. (2015). Effects of economic factors on demand for luxury hotel rooms in the U.S. Advances in Hospitality and Tourism Research, 3(1), 1-17.

van Westendorp, P.H. (1976). NSS price sensitivity meter (PSM) - A new approach to study consumer-perception of prices. Venice: Proceedings of the 29^th ESOMAR Congress.