Lost in Translation: Cross-country Differences in Hotel Guest Satisfaction by Gina Pingitore, Ph.D., Weihua Huang, Ph.D., and Stuart Greif, M.B.A. Cornell Hospitality Industry Perspectives Vol. 3, No. 2, September 2013 All CHR publications are available for free download, but may not be reposted, reproduced, or distributed without the express permission of the publisher Cornell Hospitality Industry Perspectives Vol. 3, No. 2 (September 2013) © 2013 Cornell University. This publication may not be reproduced or distributed with- out the express permission of the publisher. Cornell Hospitality Industry Perspectives is produced for the benefit of the hospitality in- dustry by The Center for Hospitality Research at Cornell University. Robert J. Kwortnik, Academic Director Glenn Withiam, Director of Publications Center for Hospitality Research Cornell University School of Hotel Administration 537 Statler Hall Ithaca, NY 14853 Phone: 607-255-9780 Fax: 607-254-2922 www.chr.cornell.edu Advisory Board Jeffrey Alpaugh, Managing Director, Global Real Estate & Hospitality Radhika Kulkarni, VP of Advanced Analytics R&D, Practice Leader, Marsh SAS Institute Niklas Andréen, Group Vice President Global Hospitality & Partner Gerald Lawless, Executive Chairman, Jumeirah Group Marketing, Travelport GDS Christine Lee, Senior Director, U.S. Strategy, McDonald’s Corporation Scott Berman ‘84, Principal, Real Estate Business Advisory Services, Industry Mark V. Lomanno Leader, Hospitality & Leisure, PricewaterhouseCoopers David Meltzer MMH ‘96, Chief Commercial Officer, Sabre Hospitality Raymond Bickson, Managing Director and Chief Executive Officer, Taj Group Solutions of Hotels, Resorts, and Palaces William F. Minnock III ‘79, Senior Vice President, Global Operations Michael Cascone, President and Chief Operating Officer, Forbes Travel Guide Deployment and Program Management, Marriott International, Inc. Eric Danziger, President & CEO, Wyndham Hotel Group Mike Montanari, VP, Strategic Accounts, Sales - Sales Management, Benjamin J. “Patrick” Denihan, Chief Executive Officer, Schneider Electric North America Denihan Hospitality Group Hari Nair, Vice President of Market Management North America, Expedia, Chuck Floyd, Chief Operating Officer–North America, Hyatt Inc. RJ Friedlander, CEO, ReviewPro Brian Payea, Head of Industry Relations, TripAdvisor Gregg Gilman, Partner, Co-Chair, Employment Practices, Davis & Umar Riaz, Managing Director, Accenture Gilbert LLP Carolyn D. Richmond, Partner, Hospitality Practice, Fox Rothschild LLP Susan Helstab, EVP Corporate Marketing, Four Seasons Hotels and Resorts Susan Robertson, CAE, EVP of ASAE (501(c)6) & President of the ASAE Steve Hood, Senior Vice President of Research, STR Foundation (501(c)3), ASAE Foundation Jeffrey A. Horwitz, Chair, Lodging & Gaming Group and Head, Private Michele Sarkisian, Senior Vice President, Maritz Equity Real Estate, Proskauer K. Vijayaraghavan, Chief Executive, Sathguru Management Consultants (P) Kevin J. Jacobs ‘94, Executive Vice President & Chief of Staff, Head of Real Ltd. Estate, Hilton Worldwide Adam Weissenberg ‘85, Vice Chairman, US Travel, Hospitality, and Leisure Kenneth Kahn, President/Owner, LRP Publications Leader, Deloitte & Touche USA LLP Kirk Kinsell MPS ‘80, President, The Americas, InterContinental Hotels Michelle Wohl, Vice President of Marketing, Revinate Group Mark Koehler, Senior Vice President, Hotels, priceline.com Thank you to our generous Corporate Members Senior Partners Accenture ASAE Foundation Carlson Rezidor Hotel Group Hilton Worldwide National Restaurant Association SAS STR Taj Hotels Resorts and Palaces Partners Davis & Gilbert LLP Deloitte & Touche USA LLP Denihan Hospitality Group Expedia, Inc. Forbes Travel Guide Four Seasons Hotels and Resorts Fox Rothschild LLP Hyatt Hotels Corporation InterContinental Hotels Group Jumeirah Group LRP Publications Maritz Marriott International, Inc. Marsh’s Hospitality Practice McDonald’s USA priceline.com PricewaterhouseCoopers Proskauer ReviewPro Revinate Sabre Hospitality Solutions Sathguru Management Consultants (P) Ltd. Schneider Electric Travelport TripAdvisor Wyndham Hotel Group Friends 4Hoteliers.com • Berkshire Healthcare • Center for Advanced Retail Technology • Cleverdis • Complete Seating • Cruise Industry News • DK Shifflet & Associates • eCornell & Executive Education • ehotelier.com • EyeforTravel • The Federation of Hotel & Restaurant Associations of India (FHRAI) • Gerencia de Hoteles & Restaurantes • Global Hospitality Resources • Hospitality Financial and Technological Professionals • hospitalityInside.com • hospitalitynet.org • Hospitality Technology Magazine • HotelExecutive.com • HRH Group of Hotels Pvt. Ltd. • International CHRIE • International Society of Hospitality Consultants • iPerceptions • J.D. Power and Associates • The Leading Hotels of the World, Ltd. • The Leela Palaces, Hotels & Resorts • The Lemon Tree Hotel Company • Lodging Hospitality • Lodging Magazine • LRA Worldwide, Inc. • Milestone Internet Marketing • MindFolio • Mindshare Technologies • The Park Hotels • PhoCusWright Inc. • PKF Hospitality Research • Questex Hospitality Group • RateGain • The Resort Trades • RestaurantEdge.com • Shibata Publishing Co. • Sustainable Travel International • UniFocus • WIWIH.COM Lost in Translation: Cross-Country Differences in Hotel Guest Satisfaction by Gina Pingitore, Weihua Huang, and Stuart Greif ExECuTivE SuMMAry he reality of contemporary hotel operation is that hoteliers need to make comparisons across Tdiverse countries regarding differences and similarities in guest satisfaction. Noting the absence of studies that explain how to compare survey responses from hotel guests in different countries, we sought to address this gap by examining four issues critical to hoteliers. Based on two years of data for nearly 200,000 guests from eight nations, our study found: (1) While price and location remain uppermost as decision factors, residents of some countries give considerable weight to specific services; (2) People in different countries do consider different factors in their determination of satisfaction; (3) The effect of certain procedures on guests’ satisfaction differs by country; and (4) Residents of some countries generally express lower levels of satisfaction than those in other countries. To ensure the reliability and consistency of our results, we evaluated results for two years individually (2010 and 2011) and then compared the findings between the two years. Even after controlling for brand and key predictors of satisfaction, we found that guests from the United States provided the highest ratings; guests from Japan provided the lowest ratings; and ratings by guests from France, Germany, Italy, Spain, and the U.K. typically fell between these extremes. The implications of our findings are that country differences must be accounted for when multinational brands are benchmarking or comparing satisfaction results across different market segments. We provide recommendations on how to account for differences in international satisfaction scores so that hoteliers can more effectively use their benchmarking results and can train staff members to respond appropriately to international travelers’ expressions of satisfaction or dissatisfaction. Hoteliers should also be aware of these cultural differences when they host international travelers, who may have diverse satisfaction standards or who may be more (or less) likely to express pleasure than are guests from other countries. 4 The Center for Hospitality Research • Cornell University ABouT ThE AuThorS Gina Pingitore, Ph.D., is responsible for leading and innovating research design, statistical techniques and analytical interpretations of all JD Power syndicated, proprietary, and ad hoc research with the purpose of enhancing both the efficiency and effectiveness of the brand’s data driven insights and interpretations and doing so in a manner which maintains and enhances consumer and client trust in the JD Power brand (gina. pingitore@jdpa.com). Weihua huang, Ph.D., is director, corporate research at J.D. Power and Associates; weihua.huang@jdpa.com. She is responsible for providing the company’s practice areas with questionnaire design and review, data analyses, statistical modeling and statistical consultations to meet the needs of external clients. Prior to joining J.D. Power and Associates, she was assistant vice president at Countrywide (now Bank of America), where she built and validated statistical models for predicting mortgage delinquency and default. The holder of a master’s degree in economics from Temple University, she earned her Ph.D. from the University of California, Los Angeles. Stuart Greif, M.B.A., is vice president and general manager, diversified industries practice, at J.D. Power and Associates; stuart.greif@jdpa.com. He joined J.D. Power and Associates from its parent corporation, The McGraw-Hill Companies, after holding executive positions across a number of McGraw-Hill businesses. Among other industry presentations, he has presented an analysis of social media at the Cornell Hospitality Research Summit. He began his career in consulting working with Fortune 500 executives. A graduate of Wesleyan University of Connecticut he holds an MBA from the Darden School of Business at the University of Virginia. The authors would like to acknowledge and thank the valuable contributions from Dan von der Embse and Cam Freedlund. Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 5 CornEll hoSPiTAliTy induSTry PErSPECTivES Lost in Translation: Cross-Country Differences in Hotel Guest Satisfaction. by Gina Pingitore, Weihua Huang, and Stuart Greif, With the global expansion of the hotel industry and greater mobility of international travelers, awareness of international differences in guests’ attitudes about their travel experiences is important. As a consequence, most multinational hotel chains currently invest significant resources in implementing large-scale measurement programs to track, compare, and benchmark guest satisfaction across their various international markets. They do so for two related reasons. First, most hoteliers understand that highly satisfied guests are much more likely to return to that property and spend more during future stays than guests who are indifferent or displeased.1 More important, successful hoteliers understand that simply tracking performance is not enough. What is required is using the results of tracking programs to guide day-to-day management decisions and, ultimately, long-term operational strategies. 1 G. Pingitore, D. Seldin, and A. Walker, “Making Customer Satisfaction Pay: Connecting Survey Data to Financial Outcomes in the Hotel Industry,” Cornell Hospitality Industry Perspectives, Vol. 1, No. 5 (2010); Cornell Center for Hospitality Research. 6 The Center for Hospitality Research • Cornell University Comparing international satisfaction scores presents living in societies that focus on individuals tend to give distinct challenges for multinational hotel chains. Most satis- ratings that are more extreme, both positive and nega- faction tracking programs find notable differences in scores tive. Of course, response styles aren’t the only explanation from one nation to another. Thus, many chains struggle to for inter-country differences in survey results. Recently, reconcile why their hotels in some markets consistently score Morgeson et al. examined the influences of socioeconomic lower than those in other locations, even as the low-rated and political-economic factors to help explain further why hotels’ business outcomes are similar or better than more some countries express higher levels of satisfaction than highly rated properties in other countries. These observations others.6 correspond with a growing body of academic findings show- Although these concepts help explain the difference in ing that consumers in different countries vary in how they country-level scores, hoteliers need more insights into the use rating scales.2 Two commonly reported different response satisfaction-related differences between guests from differ- styles are an extreme response style (ERS), or the tendency of ent countries. Such details are critical not only to inform respondents to use the end points of a scale; and a middle re- and better calibrate results from multinational guest mea- sponse style (MRS), in which respondents have the tendency surement programs, but also—and more important—such to answer toward the middle points of a scale.3 information is critical in making effective across-market One explanation for different response styles comes from recommendations and operational decisions. social psychological research showing that members within The focus of this paper is to determine whether there the same type of culture or society share the same patterns.4 are meaningful differences among guests from different These shared values influence how individuals view, process, countries on four questions critical to hoteliers: (1) Are and evaluate the world around them. One key difference there country-level differences in the reasons for select- involves whether a society’s orientation is individual or col- ing a hotel?; (2) Are the drivers of satisfaction different by lective.5 Residents of societies that focus on its members as a country?; (3) Do standard operating procedures (SOPs) collective tend to give more moderate ratings, whereas people or their impacts differ by country?; and (4) Do levels of 2 satisfaction differ by country? The results of this paper can H. Baumgartner and J.E.M. Steenkamp, “Response Styles in Marketing Research: A Cross-National Investigation,” Journal of Marketing Research, guide hoteliers on how to compare across-market satisfac- Vol. 38 (May 2001), pp. 143-156 tion scores in order to have a more accurate assessment of 3 A. Harzing, “Response styles in cross-national survey research: A performance. 26-country study,” International Journal of Cross Cultural Management, Vol. Method 6, No. 2 (2006). 4 H.C. Triandis, “The self and social behavior in differing cultural contexts.” We analyzed the results of the J.D. Power North America SM Psychological Review, Vol. 96 (1989), pp. 506–520; and H.C. Triandis, “The Hotel Guest Satisfaction Index Study, 2010-2011; Euro- psychological measurement of cultural syndromes,” American Psychologist, pean Hotel Guest Satisfaction Index Study,SM 2010-2011; Vol. 51 (1996), pp. 407–415. and Japan Hotel Guest Satisfaction Index Study,SM 2010- 5 An example of an individualistic society is the United States, while Japan 2011. These studies were executed in eight countries and is an example of a collective society. Other terms include independent- included nearly 200,000 guest responses. The countries interdependent: H.R. Markus and S. Kitayama, “Culture and the self: Implications for cognition, emotion, and motivation,” Psychological Review, analyzed for each year were Canada (n = 2,484 responses Vol. 98 (1991), pp. 224–253; idiocentrism–allocentrism: Triandis, op.cit.; in 2010; 3,063 in 2011); United States (n = 50,690; 58,129); and agency–communion: D. Bakan, The quality of human existence (Chi- France (n = 2,833; 4,438); Germany (n = 2,506; 3,622); Italy cago: Rand Mc-Nally, (1966). Also see: G. Hofstede, M.H. Bond, and C.-L. Luk, “Individual perceptions of organizational cultures: a methodological 6 F. Morgeson, S. Mithas, T. Keiningham, and L. Akoy, “An investigation treatise on levels of analysis,” Organizational Studies, Vol. 14, No. 4 (1993), of the cross-national determinants of customer satisfaction,” Journal of pp. 483–503. the Academy of Marketing Science, Vol. 39 (2011), pp. 198–215. Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 7 Exhibit 1 reasons for selecting a hotel Guest Residence North America Europe Japan United United Canada States France Germany Italy Spain Kingdom Japan Convenience/Location 35% (1) 40% (1) 40% (1) 42% (1) 31% (1) 37% (1) 43% (1) 51% (1) Price 25% (2) 29% (2) 19% (2) 22% (2) 16% (3) 17% (3) 29% (2) 37% (2) Previous experience 20% (3) 23% (3) 18% (3) 22% (2) 15% (4) 19% (2) 25% (3) 16% (4) Reputation 13% (4) 14% (5) 18% (3) 15% (4) 22% (2) 16% (4) 20% (4) 12% (5) Recommended by someone 5% (6) 5% (6) 9% (5) 12% (5) 11% (5) 14% (5) 9% (5) 5% (7) Rewards program member 9% (5) 16% (4) 9% (5) 9% (6) 3% (7) 7% (6) 9% (5) 6% (6) Corporate policy 1% (9) 1% (8) 2% (7) 2% (9) 3% (7) 2% (8) 1% (8) 2% (8) Package deal 5% (6) 3% (7) 2% (7) 5% (7) 8% (6) 4% (7) 6% (7) 17% (3) Environmentally friendly 2% (8) 1% (8) 2% (7) 3% (8) 3% (7) 2% (8) 1% (8) 1% (9) This table shows the percentage of guests from each country who select the reason listed when choosing a hotel. The number in parentheses reflects the rank order of these percentages. Percentages do not add up to 100% as guests could select multiple reasons. Sources: J.D. Power 2011 North America Hotel Guest Satisfaction Index StudySM J.D. Power 2011 European Hotel Guest Satisfaction Index StudySM J.D. Power 2011 Japan Hotel Guest Satisfaction Index StudySM (n = 2,856; 2,520); Spain (n= 3,066; 3,546); United Kingdom by price than those from other countries. In contrast, guests (n = 2,376; 3,224); and Japan (n = 32,885; 21,651). To assess from Italy and Spain are less influenced by price, but are the validity and reliability of our results, we first evaluated slightly more influenced by reputation or recommendations. the 2010 data and then replicated our analyses using the A noteworthy difference we found was that guests from 2011 results. All surveys were executed online using the Japan were significantly more likely to select a property same questionnaire; therefore, neither data collection nor based on the package deals available, perhaps because most instrumentation differences were areas to control. hotels operating in Japan offer breakfast and other amenities Results as a standard competitive offering. Another difference we found was the greater influence of rewards programs among (1) Country-level differences in the reasons for selecting a guests from the United States, suggesting that these pro- hotel. We asked guests to indicate the reasons for selecting grams are yielding the desired loyalty benefits. the hotel property and then compared the percentages for (2) Drivers of satisfaction by country. The second each reason. To make it easier to see the similarities and dif- question we examined was whether there are country-level ferences both across and within markets, we also present the differences in the drivers of satisfaction and whether the corresponding rank order of these percentages. various elements of the guest experience have differential As displayed in Exhibit 1, we found that convenience impacts on satisfaction scores. To assess this, we employed or location remains the most important reason for selecting the weighted measurement approach implemented in the a property in every country. However, there were notable data collected by J.D. Power. This approach is based on country-to-country differences in the percentages of guests the premise that some aspects of the guest experience are who cited this factor. For example, location or convenience more important than others, and understanding the relative was cited by 51 percent in Japan, while in Italy it was cited by importance of each element in the experience helps hoteliers just 31 percent of respondents. Thus, while location is a key prioritize and direct resources to improve the guest experi- selection criterion in every market, there are other reasons ence. The J.D. Power Guest Satisfaction Index (GSI) is the that influence guests’ selection of a hotel. aggregation of satisfaction scores of various experiences Price is also particularly influential in certain countries, (e.g., facility, room, and price) that are weighted based on notably Japan, where guests appear to be more influenced their importance to the guest’s overall experience. Results of 8 The Center for Hospitality Research • Cornell University Exhibit 2 impact weights by country Guest Residence North America Europe Japan United United Canada States France Germany Italy Spain Kingdom Japan Reservation 3% 3% 5% 4% 3% 3% 4% 2% Check-In/Check-Out 12% 12% 12% 15% 16% 14% 15% 18% Guest Room 25% 26% 25% 21% 20% 24% 24% 27% Food & Beverage 8% 10% 12% 15% 13% 12% 13% 14% Hotel Services 8% 8% 7% 9% 12% 11% 8% 5% Hotel Facilities 17% 17% 20% 17% 19% 19% 15% 19% Costs & Fees 26% 24% 19% 19% 18% 17% 21% 16% The percentages in this table show the importance weights of each driver of overall satisfaction. The higher the percentage, the more important the driver is to overall satisfaction. The importance weights listed for each country sum to 100%. Sources: J.D. Power 2010 North America Hotel Guest Satisfaction Index StudySM J.D. Power 2010 European Hotel Guest Satisfaction Index StudySM J.D. Power 2010 Japan Hotel Guest Satisfaction Index StudySM Exhibit 3 SoPs’ impact on satisfaction scores by country (binary features) Guest Residence North America Europe Japan United United Canada States France Germany Italy Spain Kingdom Japan Impact (%) Impact (%) Impact (%) Impact (%) Impact (%) Impact (%) Impact (%) Impact (%) Reservation accurate 103 (97%) 137 (97%) 122 (97%) 167 (98%) 128 (98%) 151 (96%) 157 (97%) 22 (98%) No billing error 52 (97%) 81 (97%) 81 (93%) 64 (94%) 65 (94%) 104 (95%) 47 (92%) 58 (99%) Problem-free stay 130 (92%) 143 (93%) 98 (84%) 128 (85%) 101 (89%) 119 (89%) 113 (82%) 68 (89%) Aware of conservation programs 52 (40%) 56 (42%) 46 (51%) 50 (62%) 36 (60%) 36 (54%) 38 (60%) 44 (28%) The four SOPs in this table are measured with “yes” and “no” binary responses. The numbers show the impact on satisfaction scores when the SOP is met. For example, when the reservation is accurate, satisfaction among Canadian guests is 103 points higher than when the reservation is inaccurate. The percentage of time the SOP is met is in the parentheses. Sources: J.D. Power 2010 North America Hotel Guest Satisfaction Index StudySM J.D. Power 2010 European Hotel Guest Satisfaction Index StudySM J.D. Power 2010 Japan Hotel Guest Satisfaction Index StudySM this approach yield scores that range from a low of 100 to a Using this J.D. Power index methodology, we esti- maximum of 1,000.7 mated the importance weights for each country. Comparing weights as shown in Exhibit 2, we found that the drivers of 7 To estimate importance weights, J.D. Power uses a combination of factor satisfaction were similar across markets, with most factors analysis and regression techniques in a serried approach. Factor analysis is used to confirm the correct factor structure as well as remove any within 0 to 4 percent of one another. While our data have multi-collinearity between rating items. Multiple regression techniques sufficient power to determine when a small difference is are used to estimate the importance of factors so that the importance statistically significant, differences of less than 5 percent are weight of each factor is the proportion of variance (rebased to sum up to 1) in satisfaction that is explained by each factor. This methodology is described in Pingitore, Seldin, & Walker, op.cit.) Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 9 Exhibit 4 SoPs’ impact on satisfaction scores by country (continuous-value features) Check-In Time (minutes) Number of Staff Contacts % Meeting % Meeting Guest Residence Impact Break Point Break Point Impact Break Point Break Point North Canada 51 9 or less 51% 44 2+ 41% America United States 47 6 or less 50% 54 2+ 42% France 34 14 or less 74% 49 2+ 49% Germany 48 10 or less 72% 60 2+ 60% Europe Italy 43 14 or less 63% 52 3+ 31% Spain 39 8 or less 38% 58 3+ 39% United Kingdom 28 9 or less 47% 53 2+ 56% Japan Japan 13 9 or less 53% 44 2+ 38% The two SOPs in this table are measured using the continuous scale. The numbers shown in the Impact column are impact on satisfaction scores when the break point (see footnote 9) of each SOP is met. Results suggest that shorter check-in times lead to higher satisfaction, and more staff interactions during the stay also lead to higher satisfaction. Sources: J.D. Power 2010 North America Hotel Guest Satisfaction Index StudySM J.D. Power 2010 European Hotel Guest Satisfaction Index StudySM J.D. Power 2010 Japan Hotel Guest Satisfaction Index StudySM not operationally significant.8 Two differences that do have that the magnitude of their impact varied by country. For operational significance were that, compared with North example, reservation accuracy had a much smaller impact American respondents, Japanese guests viewed the check-in in Japan (22 points) than it did in all other countries, even and check-out process with more importance, but they saw though the accuracy rates were essentially the same (all cost and fees as less important. countries achieved 96- to 98-percent accuracy). Similarly, (3) Effects of standard operating procedures (SOPs) the impact of having a problem-free stay was also less in by country. We conducted a series of impact analyses Japan (68 points) than in other countries, even though 89 within each country to determine whether any SOPs varied percent of Japanese guests had no problems. Interestingly, either in significance (p < .05) or impact on overall satisfac- the percentages of problem-free stays were highest in both tion. Across all eight countries, we found six operational the United States (93%) and Canada (92%), and yet the procedures that had significant impact on guest satisfaction impact of this factor was greatest in these nations, at 143 for scores.9 As displayed in Exhibits 3 and 4 these procedures the U.S. and 130 points for Canada. are reservation accuracy, billing error, number of problems Finally, wait time to check in showed a number of during stay, awareness of conservation programs, check-in important differences. We examined the country level differ- time, and number of staff contacts during stay. We present ences in wait time using two different statistical approaches. these in separate exhibits because four of the six SOPs are First, we determined how long a wait had to be before it binary questions yielding yes or no answers (Exhibit 3). The diminished satisfaction levels by 50 points. Guests from the other two SOPS are numerical measures with a continuous United States had the shortest wait time tolerance, taking scale (Exhibit 4). only five minutes to reach the 50-point gap. That is, satisfac- While all six of the SOPs had significant impact on tion was 50+ points higher when guests were checked in satisfaction levels for each country’s respondents, we found within five minutes, as compared to the situation when it took more than five minutes to check in. In contrast, guests 8 Given the sample sizes within each country, we have sufficient statistical from Japan had the longest tolerance, at 30 minutes. Waiting power to detect small (1%) differences in the importance weights. From an tolerance before the 50-point decline in satisfaction was sev- operational and change management perspective, a 5-percent difference in importance weights is considered to be a meaningful difference. en minutes for guests from Canada; 15 minutes for guests 9 To ensure that these SOPs were not a function of brand variation, we also from France, Germany, Italy, and Spain; and 17 minutes for conducted the same impact analyses, but used only nine hotel corpora- guests from the U.K. tions that operate in each country. These results showed that the same six Second, as shown in Figure 4, corresponding to these SOPs have a significant impact on the guest experience, indicating these differences in wait time tolerance levels, we also found that SOPs are core operational procedures that all brands need to focus on. 10 The Center for Hospitality Research • Cornell University Exhibit 5 Country-level guest satisfaction scores 2010 2011 Guest Residence Index Mean N Index Mean N Canada 749 2,133 756 2,648 France 739 1,979 732 2,977 Germany 769 1,677 759 2,384 Italy 748 2,134 764 1,823 Japan 697 6,039 697 4,227 Spain 718 1,352 718 1,809 United Kingdom 740 1,699 740 2,166 United States 771 45,182 768 51,478 Sources: J.D. Power North America Hotel Guest Satisfaction Index Study,SM 2010–2011 J.D. Power European Hotel Guest Satisfaction Index Study,SM 2010–2011 J.D. Power Japan Hotel Guest Satisfaction Index Study,SM 2010–2011 the impact of waiting longer than the break point10 was tion scores between 2010 and 2011, indicating a consistent significantly more negative among guests from the United response style (Exhibit 5). States (47 points) than among those from Japan (13 points). Approaches Hoteliers Can Use to Adjust for When we examined the percentage of time that hotels within each market achieved check-in time within the break Differences in Satisfaction Scores between point, guests from Spain had the lowest incidence of check- Countries ing in within the break point (38%). Our findings of clear and consistent differences in satisfac- (4) Differences in levels of satisfaction by coun- tion levels for different nations raise two practical issues for try. Before we assessed country-level differences in guest hoteliers. First, hoteliers need to adjust for these differences satisfaction, we first needed to address the natural variation when comparing guest satisfaction results for hotels in differ- among hotels in different countries and for different hotel ent markets. Second, hotel firms must be aware of differences brands. We needed to control for divergent operational between guests from different nations staying in a particular practices and standards before assessing any score differ- hotel. The best option that hoteliers have to compare mean- ences. To achieve this relatively level operating field, we level performance scores across markets is to create statisti- examined the responses for guests from nine hotel corpo- cal approaches that calibrate score differences. As with any rations for which we had a sufficient sample size and that calibration efforts, there are a number of different statistical operated in each of the eight countries analyzed. approaches that can be used. Additionally, as with all statisti- Using those nine hotel corporations, we examined the cal approaches, the more effective the technique in creating overall satisfaction scores for each nation. We found that the calibration, the more complex the equations. For our the results for these hotel firms were somewhat consistent purpose, we selected two different modeling techniques that with satisfaction findings from other industries. As reported enabled us to isolate and estimate the country-level score in other studies, guests from the United States provided the variation, and that allowed us to determine the degree to highest ratings, but contrary to some reports, we found that which the results would converge. guests from Japan provided the lowest ratings for hotels. Our first statistical approach used ordinary least squares We also found similar patterns of market-level satisfac- (OLS) regression to determine country-level differences in the satisfaction score index. This is expressed as Index = 10 Break point refers to the point in the distribution of a continuous αD + βX +ηY +θZ + ε, where D is the country vector (i.e., SOP that gives the largest adjusted satisfaction score gap that takes into dummy coding country into seven binary variables) includ- account both the raw gap in satisfaction score between meeting and not meeting the break point and also the percentage of meeting the break ing intercept, X is the vector of common SOPs, Y is the vec- point. tor of common guest profile variables (age and gender), Z is Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 11 Exhibit 6 Country-level differences in guest satisfaction scores (1,000-point scale) 2010 2011 Ordinary Hierarchical Ordinary Hierarchical Least Square Linear Model Least Square Linear Model Guest Residence (OLS) (HLM) (OLS) (HLM) US - Canada 12 9 0 2 US - France -3 -1 27 33 US - Germany -16 -9 15 21 US - Italy 15 21 8 8 US - Japan 71 69 81 79 US - Spain 50 59 49 58 US - UK 15 18 32 34 The numbers in this table show the difference in index points between the United States and the other countries after controlling the factors that could influence guest satisfaction. For example, in 2010, after confounding factors are controlled, guests from the United States still provide higher ratings (by 12 index points) than guests from Canada. Sources: J.D. Power North America Hotel Guest Satisfaction Index Study,SM 2010–2011 J.D. Power European Hotel Guest Satisfaction Index Study,SM 2010–2011 J.D. Power Japan Hotel Guest Satisfaction Index Study,SM 2010–2011 the vector of other common categorical variables (including country, and so the table shows the difference in index hotel corporation dummy coded variables and experience points between the United States and the other seven coun- filters), α, β, η, and θ are coefficients, and ε denotes errors. tries after controlling the factors we discussed earlier that This well established approach provides a reasonable esti- could influence experience ratings (that is, product, includ- mate of country-level differences in satisfaction once other ing the brand, experience filters, guest profiles, and SOPs). factors are estimated and controlled.11 The results in Exhibit 6 suggest that, assuming other For the alternative statistical approach, we used a things are equal, guests from Japan and Spain tend to hierarchical linear model (HLM), another well-established provide low ratings, while guests from the United States method used most frequently in education. While similar and Canada provide high and fairly similar ratings. Guests in many ways to OLS, HLM differs as it takes into account from Italy and the United Kingdom are in the middle. The correlated errors, thus providing more flexible, albeit more year-to-year validation indicates consistent results for all but complex, statistical modeling. HLM was modeled as Index two countries, Germany and France. One possible reason = α + βX +ηY +θZ +ε, where α is the random intercept at for this inconsistency is that despite efforts to control and country level, X is the vector of common SOPs, Y is the remove confounding factors, we couldn’t control for other vector of common guest profile variables, Z is the vector of key factors, such as social and economic dynamic changes, other common categorical variables, β, η, and θ are (ran- and hotel property-specific metrics, such as occupancy rates, dom) coefficients, and ε denotes errors. which were particularly salient for these two countries. Exhibit 6 shows the difference in satisfaction scores We also found that within the same year, the two for OLS and HLM (using the J.D. Power Guest Satisfaction statistical methods yielded fairly comparable estimates of Index with a range of 100 to 1,000 points) for both the 2010 country-level differences. This suggests that the simpler and 2011 data. We used the United States as the baseline model assumption of OLS is an effective technical solution. However, individual hotel chains may find HLM a better 11 Jay L. Devore, Probability and Statistics for Engineering and the Sciences, alternative, given the hierarchical nature of the data. Eighth Edition (Boston: Brooks/Cole, 2010). 12 The Center for Hospitality Research • Cornell University Conclusion and Implications For brand and corporate leaders to have a truer sense of how One of the most promising findings of this study is that their properties are performing relative to one another, it is guests from different countries held reasonably similar important to create a pro forma score based on ratings that views of the importance of particular elements of the hotel are adjusted for country level differences to provide a more experience and which standard operational procedures are comparative view of guest satisfaction performance across keys to creating a satisfying stay. These common elements their properties and across regions. are good news to multinational chains, as they can align and The findings also raise the importance of hoteliers ap- create cross-market consistency on long-term operational plying trend analysis to assess how they perform over time strategies. by guest country of origin, and not just using aggregate The diversity of satisfaction scores across markets was figures or such segmentations as business vs. leisure travelers. not surprising, given similar results from other industries. How a hotel goes about providing an improved experience Our findings clearly showed that culture matters when try- for guests from each country offers more actionable insights ing to understand international differences in guest satisfac- that can drive operational decisions regarding how to adjust tion. Our analysis confirms that guests from some countries operations in order to better serve and delight guests from tend to either experience or express far higher levels of different national and cultural contexts. For example, a satisfaction than do guests from other countries—for what warm welcome at check-in may be valued by guests regard- is essentially the same hotel experience. Additionally, these less of country of origin, but that welcome should be deliv- findings are consistent with the idea that consumers in dif- ered in a slightly different way to delight various cultural ferent societies generally share common values that influ- groups. ence their survey response styles, such that consumers in These findings underscore the importance of staff individualistic societies, such as the United States, provide training to delineate the differences in cultural preferences higher ratings than do those in collective societies, such as of guests from various countries. The need for such training Japan, who provide more restrained ratings. becomes more pronounced for hotels with a large mix of The implications of our findings are that country dif- international guests. Hotels need to ensure that they adapt ferences must be accounted for when benchmarking or their services to avoid procedures or practices that optimize comparing satisfaction results for hotels across different the experience for some guests but inadvertently undermine markets and properties with distinct compositions of guests the experience for others. It may be desirable for a property based on country of origin. Additionally, these results point to set service guidelines by country of origin that target the to different thresholds of satisfaction for guests from each largest or the most financially important cultural groups culture. Guests from some countries may simply be harder of guests. In practice, this may mean aiming for a rating to please (or, at least, are less likely to express pleasure) than of 7.5 out of 10 for check-in from Japanese guests and 8.5 are guests from other countries. Or more particularly, guests from Canadian guests, for example. Or it may mean that from certain countries have a shorter satisfaction fuse for some services should be context sensitive, depending on certain hotel operations, such as the impatience expressed the guest’s native country. This analysis and application of by U.S. travelers with a “slow” check-in—given that “slow” insights to operations should be conducted on an ongoing means longer than a five-minute wait. basis throughout the year within each property, across re- In addition to issues relating to comparing similar gions, and globally where appropriate for a given brand, and hotels operating in different countries, the findings also across the entire guest experience, from reservation through address the changes in satisfaction ratings for a particular check-out. hotel based on hosting international guests. A hotel property Ultimately, our findings have three significant impli- in Dubai, for example, will have a different mix of guests cations for multinational hotel chains. First, hotel brand by country of origin (i.e., more heavily GCC States, Russia, and operational managers should recognize that different U.K., Germany, and China) than will a hotel in the Bahamas service practices are needed to delight guests from different (i.e., more heavily U.S. and Canadian). The Bahamas proper- countries. Second, they need to understand that efforts to ty may have higher ratings provided by their North Ameri- improve satisfaction may have a differential impact, in terms can guests, requiring the need to take into account market of the magnitude of changes, in some countries, suggesting level differences . Thus, the question becomes, Are the that efforts to establish the same “targets” for improving higher scores for the Bahamas property actually the result of satisfaction across countries may be difficult. For strategic providing a more satisfying experience to their guests, or are planning purposes, multinational hotel chains should also they simply a reflection of a higher concentration of guests recognize these country-level differences in satisfaction from countries that tend to give high satisfaction ratings? thresholds when considering entry into new markets. Third, hotel brand and operational managers need to continuously Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 13 analyze how they perform by country of origin and use leaders of multinational hotel chains to better manage and those insights to drive a more adaptive approach to service lead their businesses. and operations where appropriate. When evaluating perfor- Limitations and Future Research mance across hotel properties and markets, brand leaders should take into account how differences in guest composi- We should note that our study is limited by the fact that we tion may lead to higher or lower satisfaction scores for a analyzed only eight markets, although the sample is large. given property or region. Similarly, this research focused on only nine multinational We recognize that brand and hotel property managers corporations, which, again, is a small subset of the hotel are already inundated with data and pressed for time, not marketplace. Therefore, future research that includes more just from their guest-tracking programs, but also from rating countries and more brands would be beneficial. Addition- and review websites, social media, and mystery shopping ally, although this research is the first effort to calibrate guest audit data, along with their other responsibilities. Pragmati- satisfaction scores across countries, we included and isolated cally, the idea of another overlay of data to compare prop- only a portion of factors in our models relative to all possible erties across markets and against other properties based factors that could account for country-level differences. As on cultural differences and to analyze performance and such, including additional elements in the model would adapt operationally by country of origin may seem onerous. increase calibration precision. Further, including such opera- However, we believe good management and effective per- tional elements as price and occupancy rates would be useful, formance improvement requires the recognition of guests’ as would including more detailed respondent-level infor- international differences. Given that our data and analysis mation, such as ethnicity and acculturation levels. Finally, point to significant cultural differences in how guests from adding macro socioeconomic and political elements would different countries rate their experience, we believe under- further enhance calibration equations. n standing and acting on this information will help brand 14 The Center for Hospitality Research • Cornell University Cornell Center for Hospitality Research Publication Index www.chr.cornell.edu Cornell Hospitality Quarterly 2013 Hospitality Tools Vol. 5, No. 1 2012 Cornell Hospitality Research Summit: Critical Issues for http://cqx.sagepub.com/ Vol. 4 No. 2 Does Your Website Meet Industry and Educators, by Glenn Potential Customers’ Needs? How to 2013 Reports WithiamConduct Usability Tests to Discover the Vol. 13 No. 9 Hotel Daily Deals: A Study Answer, by Daphne A. Jameson, Ph.D. 2012 Reports of Asian Customers, by Sheryl E. Kimes, Vol. 12 No. 16 Restaurant Daily Deals: Ph.D., and Chekitan S. Dev, Ph.D. Vol. 4 No. 1 The Options Matrix Tool (OMT): A Strategic Decision-making The Operator Experience, by Joyce Wu, Sheryl E. Kimes, Ph.D., and Utpal Vol. 13 No. 8 Tips Predict Restaurant Tool to Evaluate Decision Alternatives, Sales, by Michael Lynn, Ph.D., and Andrey by Cathy A. Enz, Ph.D., and Gary M. Dholakia, Ph.D. Ukhov, Ph.D. Thompson, Ph.D. Vol. 12 No. 15 The Impact of Social Media on Lodging Performance, by Chris Vol. 13 No. 7 Social Media Use in 2013 Industry Perspectives K. Anderson, Ph.D. the Restaurant Industry: A Work in Vol. 3 No. 1 Using Research to Determine Progress, by Abigail Needles and Gary M. the ROI of Product Enhancements: A Vol. 12 No. 14 HR Branding: How Thompson, Ph.D. Best Western Case Study, by Rick Garlick, Human Resources Can Learn from Ph.D., and Joyce Schlentner Product and Service Branding to Improve Vol. 13 No. 6 Common Global and Local 2013 Proceedings Attraction, Selection, and Retention, by Drivers of RevPAR in Asian Cities, by Derrick Kim and Michael Sturman, Ph.D. Crocker H. Liu, Ph.D., Pamela C. Moulton, Vol. 5 No. 6 Challenges in Contemporary Ph.D., and Daniel C. Quan, Ph.D. Hospitality Branding, by Chekitan S. Dev Vol. 12 No. 13 Service Scripting and Authenticity: Insights for the Hospitality Vol. 13. No. 5 Network Exploitation Vol. 5 No. 5 Emerging Trends in Industry, by Liana Victorino, Ph.D., Capability:Model Validation, by Gabriele Restaurant Ownership and Management, Alexander Bolinger, Ph.D., and Rohit Piccoli, Ph.D., William J. Carroll, Ph.D., by Benjamin Lawrence, Ph.D. Verma, Ph.D. and Paolo Torchio Vol. 5 No. 4 2012 Cornell Hospitality Vol. 12 No. 12 Determining Materiality in Vol. 13, No. 4 Attitudes of Chinese Research Summit: Toward Sustainable Hotel Carbon Footprinting: What Counts Outbound Travelers: The Hotel Industry Hotel and Restaurant Operations, by and What Does Not, by Eric Ricaurte Welcomes a Growing Market, by Peng Glenn Withiam Liu, Ph.D., Qingqing Lin, Lingqiang Zhou, Vol. 12 No. 11 Earnings Announcements Ph.D., and Raj Chandnani Vol. 5 No. 3 2012 Cornell Hospitality in the Hospitality Industry: Do You Hear Research Summit: Hotel and Restaurant What I Say?, Pamela Moulton, Ph.D., and Vol. 13, No. 3 The Target Market Strategy, Key Elements for Success, by Di Wu Misapprehension: Lessons from Glenn Withiam Restaurant Duplication of Purchase Data, Vol. 12 No. 10 Optimizing Hotel Pricing: Michael Lynn, Ph.D. Vol. 5 No. 2 2012 Cornell Hospitality A New Approach to Hotel Reservations, by Research Summit: Building Service Peng Liu, Ph.D. Vol. 13 No. 2 Compendium 2013 Excellence for Customer Satisfaction, by Glenn Withiam Vol. 13 No. 1 2012 Annual Report Cornell Hospitality Industry Perspectives • September 2013 • www.chr.cornell.edu 15 Cornell Center for Hospitality Research 537 Statler Hall Ithaca, NY 14853 USA 607-255-9780 www.chr.cornell.edu