!"DepartmentofRecreation,ParkandTourismSciences,2261TAMU,TexasA&MUniversity,CollegeStation,TX77843-2261,USA #$%&'$( )*$$$$$$$+$$,&-.$/$/0 ,&-0!10233(0$&4$&$554$++ 677 "778$& 9:$;( 8$$ &&8$0$4" 0 8$*$$84" $*& #&&$&4&"&$0& &"&&&9 ,&$&&$$ r (<&=$ %&&$ Keywords: !'!'/'!$&'8'#'=& >88&& &8?.0@=2 !A&+$4" 0$&&$* B&@C0233 CA&$8 0&-$$88&* $ &$* >"233 0*88&* $88& 08A& $&$$$ *$ &$$ * 8$ &$$ 55D8++8& 9$ >"2330 8 99&&&0 $&$$$&088 &$&*& *&$*& E&$4 &*8&$* *&&*&&0 $$&&& 0&&88$$ ,&7&&* &$8&&&8&&80&&87 *&$*&0&&& A F&0$&& 8*&0$$ $8A&&$ & $8 $$&**&0$&$ 8&&&A$$&* 8$00**0 %$$ &&0A9$$ & / $8G(&&"*" 8*& C& $$*A&4 $&$*" /@,$02333 #$&8$0*& & 7&7 2427G48 r (<&=$ %&&$ $62 227D ( 2 /$ & 6H24334I(4II'896H24334I(4 (( E-mailaddress: D"J $ !"
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etrick/TourismManagement26(2005)753Â–762 (
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etrick/TourismManagement26(2005)753Â–762
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verysatised ;2 verydissatised ;2 2* $$*$ &$&& 677 4 2*&&*D$ 8&$*$& &&0&&8$ &0&$24&8 A$;2$A$;2"4 &&9 ) $ .&00$ B233I 40A4="&J.F.Petrick/TourismManagement26(2005)753Â–762
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p o 2*& 0/&2 n ;2$55& 0++/& n ;$55$ 0++$/& n ;($55 ++!A&8&$&$ *& %&$0&&0 955$++&&"&* 8&0$&0 &8*&&&$ G20 <98$$* &"$88*& $A$0$*& )8&&& 8 B$#D2332 0?2$?2* $ &&*&$$*&&&&0 & %1>K%"+ T 4 &$9 "+ ?F*$& $&8$88*0 &&&88$88>233 #&8$A$88*4 $& F ( I p o 2 !&&$ 55&++;3 &$ *55$++;2 $554 ++;2 A$4J.F.Petrick/TourismManagement26(2005)753Â–762
0$ &"&8 & 0?2 D$ %1>K%&&-$9$88 *$*& &&8& 1 $888$* F 3 2 p 4 &?2**D$ 9?*$A$8 F&&$=$* 8$ $+*80&&"& * #&8%1>K%8$ A$88*$ 8 F 2 p o 2 ! &&$55&++ ; $8* 55$++;( ($55++ ; A$&4 $ 0&"&*& 8&$ 80?D$ 98*$A$ 8 B&$=&233( 8$& &"&*A $$ /4M&$$4 & &"&*A? #& &$A$88 X 22 p 4 * 0?D$0 $88*A+0$ + #&8%1>K%9$88 $8 ?*8$A F 23 3 p o $88 * !&&$55& ++;G2 $55++ ;G2 A$ $$& *$*&* $ /4M9$88* &&$A$88 X p o 2 * )8$55&++ &"&2 L4 I L*55$++ L$ (L0 &$55++2 L$2 (L0 & /&055++ &"&$*2 L$* *& !&A& = n ;2$ n ;? n ;( &&8 ( 3I /?&&$% 2 % 2 & (L L 2L )8G20 L( 2L IL *&2 &8M& &$<&<9&$L/*+& Factor1-ShipÂ’stangibles 3 ( I >$$" II 3 >&&$ I3 )$*& I223 2 33 ( Factor2-Intangibles I 2 I2 I !8&& II I 9 II ( < (I /II &9&$ I N ;3( B>8&$M; I 2;$A$02;A$ J.F.Petrick/TourismManagement26(2005)753Â–762 I
2 L*55$++2I IL$( L0 &$55&++22 L$3 L0 & A$A4 $&& * /4M&&&$9 F&&$=$* A$ 0&"&& ?( #&&$A$88 X 2( (I I p o 2 * !& 8$55&++$$*& &&$*55$++ $550++$55$$ &$55 ++0?( $$&& 0& &$ ?(*(&*$A$8 F&&$=$* $$ &$*& $$ 1$88 8$* F 2 p 4 $ 8$$ ?(* D$0$$ &$$ A&89$$88 &&88M&0$ &$88 $88&?0?*0?0$ ?$0& $& 9$0-$ $88&$*8$ #&8%1>K%9$88 &&8*? 8$A F 3 p o 2 $88*4 !&&$* 55++;I I$55$++ ;I A$55&++ ;I 2 )$9$$& $88 8M&8?*0%1>K%$ % A F I p o 2 &488 8$9&$ L8C&"+ =*$; 3( &&44%1>K%+ &-$9$8884 *&0$*& A$88 p o 8$*8*&$ F ; 0 p o 20*8*& p 4 #&8"+8$ 55&++; &A$ *&*55$++; 2 $55++; %1>K%9$88 +$&?&8$A F p o 2 $88* ) 8$55&++; I$A4 & p o &$&*55$4 ++;I $55++;I (0 &0 $&8* A&%1>K%9$$88 *& ?$$8$A$88 F 3 p o 2 * #&8 &$&$55++;I $55$++;I $ ?&&$%/55& ++08&&*&055&++ $9&*55$++$ 55 ++ !" $&$ -$0$ *$A$ *&8$$$4"$ 80*&$& *&$&& , $8*$ M&$+4 88$ $*& $0$*& C8&&0$ *$$A$0$ &&$88*& 8$ *& 88$88 ) $$& &$0$ $0$&"& 9* A$ B&@=&0233( 8$&0 $ 08&0 88$&$*& "0$ 8$8$ $& 0$ $$888 $"$88&&8 /&055&++ 855$++$55++ 8$6*$0 8*&0*A$&&0M&8 70&*0 $*&"&8 J.F.Petrick/TourismManagement26(2005)753Â–762 3
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etrick/TourismManagement26(2005)753Â–762
"88&$**$55"++ * .&0233( 0A$ 8"$ $$" 8$ &*4"0" 8*&$&"& 4&& .&233( 8$ 55*++*& $M 0&"+& *&&$*88 08&$M&+ &&$ &&& &&0A$88 A8$ #&$88 $&& &&$&0*& 0&4*8$*88 $0$8&& $*&8 90$8$8 $ /&088$ &8N&&0 **&&4 A*& $ $&4A8$0 $$ $&$0 &0& && 8$$ 0$$ B@!&0 08$ &$984$& &$80$*$$ 8&9" ?&&$%"888&0 &$*8& 98&$8& 0 /&/=M& 0/$ /= $8&$ & $ 8&*$ 80$8 $$88888$8 C&&8$&$* &-$0$&&-4 8& ,$4 $88888$0 &&**M$" $*$8&4 &&*$$0$ 8" $08*&$& 88&&-8&8 +$"4 C8"&$04 8A&&-&$ &$554++$ " ," 0@/ = 2332 8&8&$ *&8$&& LeisureSciences 0 13 0 ,"0F % 0@/ = P&08$ *& AnnalsofTourismResearch 0 27 0II( ,# . 23( &$$ JournalofAdvertisingResearch 0 14 (03 ,01 !&1",0?& F87Q$ 20 /# 0@,$0/ 2333 Thedevelopmentandgrowthofthe cruiseindustry >98$6,4? /0% 00% ? 0@,! # M& $86$&8& EuropeanJournalof Marketing 0 34 227202I2 /) Valuecruising 677 "778$& 9:$;( /=)&% !&8 %$**6 677 * 7 /=)&% Cruiseindustryoverviewmarketingedition2003 677 777 $R 8S9 /=)&%* Cruiselineshost 8.66millioncruisevacationersin2002 677 7 /17 8:1)F;2( /=)&% Morethan2.2million peoplecruiseinrstquarter2003 677 7 /17 8:1)F;2 F&&, . / 0@=$*0B K 6%$8$&D8 $88&$ Leisure Sciences 0 25 0I23 F"# 0@K&$0% 233 Sellingthesea 1E"0 1E6C& .&0, 233( Managingcustomervalue:creatingqualityand servicethatcustomersCanSee 1E"6! .&$0# < 0@1&&0 233 )$ 6&0&$$&& JournalofProductandBrandManagement 0 9 022( .&80< 233 8888 A*&8 MarketingScience 0 14 20I2( .&0F 00B , 0@B0# 233I 888 4$*+8M &&$*& Journalof Marketing 0 62 %&0(3 .$ . 0?$0 0@>&0O ? 233 ?& 8*&6C 8: CornellHotelandRestaurantAdministrationQuarterly 0 37 %&0I2 .0 0@/0= . 233 $8$ $&$ JournalofConsumerResearch 0 19 F*0(2(22 ? 0.0 0@=0F # 2 / $&$ JournalofRetailing 0 77 0(( 0# B 0@.0% 233 &$% JournalofHospitalityandLeisureMarketing 0 34 (0 J.F.Petrick/TourismManagement26(2005)753Â–762 2
B&0 O 0@E0/ B 233 /$ 96%9&$ JournalofMarketing Research 0 27 %02 B&0. 0@=&0 F / 233( %&&8 &$8"$ Journalof ConsumerResearch 0 21 F*0(I(2I B&0. 0@C0# 233 <&&88 MarketingScience 0 14 0 2223 B&0! 0, 0@"0 233I Marketingforhospitality andtourism $<$ O$$!?&& B0= 0-$0 0@#D0 ! 233 % *$$ M$ JournalofConsumerResearch 0 19 F*0 I33 B0= 0@!&0! %8 $$&&4& Journalof Retailing 0 79 0222 B0= 0@#D0 ! 2332 %&&8 &**$&&$& MarketingScience 0 10 022I =0F # 0,&0! ? 0@,&"0C / 23II /& 8*& JournalofConsumerResearch 0 15 0 ( 0. < 0@C0# 233 %&&8 &$9&8$ Journalof ConsumerResearch 0 19 0 0B , 233 Pricing:Makingprotabledecisions 1E"0 1E6.?&& >0? 2333 M&08$ &6%&! HospitalityManagement 0 18 0 I >0? !8$88&& 0M&0$&D$68&& TourismManagement 0 24 0I33 >"< , 233 8 Cornell HotelandRestaurantAdministrationQuarterly 0 30 (03 ># 233 Anintroductiontostatisticalmethodsanddataanalysis (<$ ,&/6C$0) !% 0@.&0F 8& M&4&4&&$ Journalofthe AcademyofMarketingScience 0 28 202I2( !"0 F&8&4$&&8 $&8 JournalofLeisure Research 0 34 02232( !"0 0@," %98 8$&8$8&8&+ JournalofTravelResearch 0 41 20I( !"0 0@," * %98 $8&8&+8 JournalofTravel Research 0 40 0I #0% 0@*C % 233 888"&$ *&$89$ Journal ofConsumerResearch 0 19 *0 #$0= 0@#$0 F 233 /&6 *&$& JournalofTraveland TourismMarketing 0 2 70 9 1*8$2 677 9 7 )/0 677 /* 7$2 2 0# % 0"-0 , 0@>&"0# C 233 % 98$88 JournalofMarketing 0 60 02 &0# 23I &$ Marketing Science 0 4 02332( C$0F 233( BerlitzguidetocruisingandcruiseshipsÂ—1994 !16,&C*0 < 23 $&4 JournalofMarketing Research 0 2 02I2I3 C0# 23I %$&8*$8 8M&$$ JournalofConsumerResearch 0 13 *0 C&$>@/= %&*& $6 677 7 J.F.Petrick/TourismManagement26(2005)753Â–762
FLASH SALES: AN INVE NTORY DISTRIBUTION A ND MARKETING CHANNEL . DOES IT WORK? By EKATERINA BEREZINA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014
Â© 2014 Ekaterina Berezina
To my Mom
4 ACKNOWLEDGMENTS The c ompletion of the dissertation research is a major milestone in life and academi c career . I would not be able to come this far today without the help of outstanding individuals who supported me along the way. I would like to use this opportunity to express my sincere gratitude to everybody who stood beside me throughout my doctoral st udies, and the completion of this dissertation . First, I would like to thank my advisor, dissertation committee chair, teacher, and mentor, Dr. Kelly Semrad. Her lasting confidence in me, encouragement, support, and guidance allowed me to grow, and complet e my doctoral studies and this dissertation. I greatly appreciate Dr. Semrad teaching me step by step how to conduct sound academic research, how to become a better academic writer , how to effectively manage a classroom, and many other important skills tha t allowed me to become a better educator, a better researcher, and a better person . I am very thankful for her commitment and investment in me, her willingness to work with me even after she left t he University of Florida , and continued support over the ye ars. I have been very fortunate to have Dr. Semrad as my advisor and mentor. Also, I would like to thank all dissertation committee members who supported me along the way: Dr. Svetlana Stepchenkova, Dr. Asli Tasci, Dr. Steven Shugan, and Dr. Cihan Cobanogl u. I am very grateful to Dr. Stepche nkova for her support of my research interests, thorough review of my work, and thought provocative feedback . I am thankful to Dr. Tasci for her contribution to my dissertation research, her help with survey development, and for her continued encouragement. I would like to thank Dr. Shugan for teaching me multivariate statistics, for helping me with the development of the manuscripts in this dissertation, and for sharing his creative ideas. Also, I would like
5 to express m y sincere appreciation to Dr. Cihan Cobanoglu for his mentorship, continued support since the beginn ing of my m academic guidance, and confidence in me. Each of you has contributed to my growth as a scholar and I will be always thankful for that. Next, I would like to thank Dr. Michael Sagas, the Chair of the D epartment of Tourism, R ecreation, and Sport Management (TRSM) for providing his support and wonderful teaching opportunities . I thank Ms. Julie McGrath for her endless support, for her kin dness, and her guidance throughout the completion of the doctoral program . I am also ve ry grateful to the entire TRSM Department, including faculty, staff, and students, for creating a cheerful and nurturing atmosphere , and supporting my growth. I would like to thank Eric Friedheim Tourism Institute for financial support of data collection f or this dissertation. Also, I would like to thank my colleagues from other universities: Dr. Mehmet Erdem, Dr. Fevzi Okumus, Dr. Anil Bilgihan, Dr. Thomas Schrier, and Dr. Galen Collins. I would not perform this well without their support. I thank Mr. Joe McInerney for his kind support, encouragement, and enormous help with the data collection for this dissertation. I greatly appreciate his mentorship and guidance that he provided to me over the years. Also, I am very grateful to Mr. Lyle Worthington and Mr . Terry Price for their valuable feedback on my dissertation, and important suggestions for future research. This accomplishment would not be possible without my wonderful friends. I thank Daria Mikheeva, Galina and Evan Wall, Inara Rezyapova, Liza Berdyc hevsky, Mona Mirehie, Nadine Chersini, Natalia Velikova, Semih Yilmaz, and Tat i ana Chichugova . I greatly appreciate their support, help, and understanding.
6 And last, but not least, I am very grateful to my family that brought me to the stage where I am rig ht now. I would like to say my special words of appreciation to my wonderful mother, Liudmila Kozlechkova, for her unconditional love, understanding, patience , and endless support; to my father, Andrey Berezin, for his care and contribution to my personal growth and education; to my step father, Yury Kozlechkov, for his kindness, understanding , and support to me and my mother over the years. I thank my cousins Dmitry and Yulia Zadirako for their friendship , support, and long Skype talks. Also, I would like to thank my boyfriend, Aleksandr Mafusalov, for his patience, care, everyday support, and many useful discussions about my dissertation . I greatly appreciate him being there for me, and supporting me through out the completion of my dissertation.
7 TAB LE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Background Information ................................ ................................ .......................... 15 Revenue Maximization in Hotels ................................ ................................ ...... 16 Revenue Management Prerequisites ................................ ............................... 17 Flash S ales and the Revenue Maximization Problem ................................ ...... 19 Statement of the Problem ................................ ................................ ....................... 22 Purpose of the Study and Research Questions ................................ ...................... 24 Study 1: Industry Perspective ................................ ................................ ........... 24 Study 2: Consumer Perspective ................................ ................................ ....... 25 Study 3: Research Directions ................................ ................................ ........... 26 Definitions of Terms ................................ ................................ ................................ 27 Significance of the Study ................................ ................................ ........................ 28 Theoretical Background ................................ ................................ .......................... 30 Theory of Peak Load Pricing ................................ ................................ ............ 30 Cost/Benefit Analysis ................................ ................................ ....................... 32 Research Design ................................ ................................ ................................ .... 33 2 THE MANAGERIAL FLASH SALES DASH: IS THERE ADVANTAGE OR DISADVANTAGE AT THE FINISH LINE? ................................ .............................. 38 Background of the Study ................................ ................................ ......................... 38 Purpose of the Study ................................ ................................ ........................ 38 Statement of the Problem ................................ ................................ ................. 41 Research Question ................................ ................................ ........................... 43 Significance of the Study ................................ ................................ .................. 43 Literature Review ................................ ................................ ................................ .... 45 Advance Selling ................................ ................................ ................................ 45 Discounted Sales ................................ ................................ ............................. 46 Hotel Inventory Distribution Channel Evaluation ................................ .............. 47 Cost/benefit analysis theory ................................ ................................ ....... 48 Electronic distribution channel evaluation framework ................................ 49
8 Methods and Procedures ................................ ................................ ........................ 50 Research Methodology ................................ ................................ ..................... 50 Research Method ................................ ................................ ............................. 51 Data Collection ................................ ................................ ................................ . 52 Study participants ................................ ................................ ...................... 52 Interview guide ................................ ................................ ........................... 54 Data Analysis ................................ ................................ ................................ ... 56 Rigor of Qualitative Research ................................ ................................ ........... 57 Findings ................................ ................................ ................................ .................. 60 To Groupon, or not to Groupon? ................................ ................................ ...... 61 Benefits of Distributing Hotel Inventory via Flash Sales Websites .................... 62 D rawbacks of Distributing Hotel Inventory via Flash Sales Websites ............... 64 Performance Measures of Hotel Flash Sales ................................ ................... 67 Generating Rev enue from Non Room Operating Departments ........................ 70 Closing Remarks ................................ ................................ ................................ .... 71 Theoretical Contribution ................................ ................................ ................... 73 Managerial Implications ................................ ................................ .................... 80 3 FLASH SALES CUSTOMER FOR HOTELS? ................. 87 Background of th e Study ................................ ................................ ......................... 87 Purpose of the Study ................................ ................................ ........................ 92 Statement of the Problem ................................ ................................ ................. 95 R esearch Questions ................................ ................................ ......................... 97 Significance of the Study ................................ ................................ .................. 97 Literature Review ................................ ................................ ................................ .... 98 ................................ ................................ ....... 99 Customer Response to Sales Promotions ................................ ...................... 101 Utilitarian benefits of sales promotions ................................ .................... 103 Hedonic benefits of sales promotions ................................ ...................... 105 Costs of sales promotions ................................ ................................ ........ 108 Customer Value ................................ ................................ .............................. 109 Methods and Procedures ................................ ................................ ...................... 111 Research Design ................................ ................................ ............................ 111 Instrumentation ................................ ................................ ............................... 112 Variables and Data Analysis Strategy ................................ ............................ 115 RQ1. What are the key profiling traits of the customers who purchase hotel flash sales deals? ................................ ................................ ........ 116 RQ2. Is there a difference between the economic contribution of the flash sales customers and other customers? ................................ ........ 119 RQ3. Is there a difference between intentions to revisit a hotel demonstrated by the flash sales customers and other customers? ...... 120 RQ4. Is there a difference between intentions to recommend a hotel to others demonstrated by the flash sales customers and other customers? ................................ ................................ ........................... 121 Sample Size, Power, and Effect Size ................................ ............................. 122 Data Collection ................................ ................................ ............................... 123
9 Sample and data collection ................................ ................................ ...... 123 Pilot study ................................ ................................ ................................ 125 Data Preparation and Cleaning ................................ ................................ ...... 125 Findings ................................ ................................ ................................ ................ 127 Respondent Demographics and Travel Behavior ................................ ........... 127 Flash Sales Customers Personality Traits ................................ ...................... 130 Economic Contribution Comparison ................................ ............................... 132 Revisit Intentions ................................ ................................ ............................ 133 Recommending a Hotel to Others ................................ ................................ .. 135 Implications ................................ ................................ ................................ ........... 135 Deal Seeking ................................ ................................ ................................ .. 136 Price consciousness ................................ ................................ ................ 137 Market mavenism ................................ ................................ ..................... 139 Quality consciousness ................................ ................................ ............. 141 Variety seeking ................................ ................................ ........................ 142 Revenue Generation ................................ ................................ ...................... 144 Future Behavioral Intentions ................................ ................................ ........... 145 Limitations ................................ ................................ ................................ ............. 146 Closing Remarks ................................ ................................ ................................ .. 149 4 HOTEL FLASH SALES RE SEARCH: WHERE DO WE GO FROM HERE? ......... 155 Background of the Study ................................ ................................ ....................... 155 Purpose of t he Study ................................ ................................ ...................... 156 Research Problem ................................ ................................ .......................... 157 Significance of the Study ................................ ................................ ................ 158 Proposed Research Directions ................................ ................................ ....... 159 Effect of Promotions on Demand ................................ ................................ .......... 159 Effect of Promotions on Attracting New Customers ................................ .............. 162 Promotion Prone vs. Value Conscious Customers ................................ ............... 164 Extending Profile of Flash Sales Customers ................................ ......................... 166 Closing Remarks ................................ ................................ ................................ .. 168 5 CONCLUSIONS ................................ ................................ ................................ ... 179 APPENDIX A INDUSTRY RESPONDENT PROFILE ................................ ................................ . 183 B CONSUMER QUESTIONNAI RE ................................ ................................ .......... 186 C CONSUMER PSYCHOGRAPH IC CONSTRUCTS ................................ ............... 197 LIST OF REFERENC ES ................................ ................................ ............................. 200 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 211
10 LIST OF TABLES Table page 1 1 The use of cost benefit analysis in different research fields ............................... 37 2 1 Company profile ................................ ................................ ................................ . 83 2 2 Respondent profile summary ................................ ................................ .............. 83 3 1 Pre pilot construct reliability ................................ ................................ .............. 151 3 2 Pilot construct reliability ................................ ................................ .................... 15 1 3 3 Respondent de mographic characteristics ................................ ......................... 152 3 4 Respondent demographics by distribution channel ................................ .......... 153 3 5 Rotated factor loadings ................................ ................................ ..................... 154 4 1 Proposed future research directions ................................ ................................ . 175 4 2 Measurement items for value consciousness construct ................................ ... 176 4 3 Measurement items for promotion proneness construct ................................ ... 177 4 4 Measurement items for smart shopper self perception construct ..................... 177 4 5 Measurement items for effort/time savings construct ................................ ....... 178 A 1 Industry respondent profile ................................ ................................ ............... 183 C 1 Consumer psychographic constructs ................................ ................................ 197
11 LIST OF FIGURES Figure page 2 1 Categorization of hotel flash sales benefits ................................ ........................ 84 2 2 Categorization of hotel flash sales drawbacks ................................ .................... 84 2 3 Performance measures of hotel flash sales promotions ................................ ..... 85 2 4 Flash sales evaluation framework ................................ ................................ ...... 86
12 LIST OF ABBREVIATIONS Benefits of flash sales Advantages that hotel managers may achieve by distributing room inventory via flash sales websites . Drawback s of hotel flash sales Disadvantages or threats that hotels may experience as a result of distributing room inventory via flash sales websites . Flash sales websites A website offering time limited deep discounts (of about 50%) for advance purchase of prod ucts or services (Edelman et al., 2011; Piccoli & Dev, 2012) including Groupon Getaways, LivingSocial Escapes, and Jetsetter . Flash sales customers Customers who make hotel reservations via flash sales websites Other customers Customers who make hotel reservations via distribution channels other than flash sales websites.
13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Phi losophy FLASH SALES: AN INVENTORY DISTRIBUTION AND MARKETING CHANNEL. DOES IT WORK? By Ekaterina Berezina August 2014 Chair: Kelly J. Semrad Major: Health and Human Performance The purpose of this dissertation is two fold. The first fold is to provide hotel managers with an understanding of the potential benefits and drawbacks associated with using flash sales websites as room inventory distribution channels. The second fold is to provide platform research from which other scholars may build from to en hance the available literature on flash sales. Currently, there is a lack of empirical studies that could offer guidelines to hotel managers for how and when to use flash sales as a room inventory distribution channel. The dissertation includes three rese arch studies that were developed to investigate the phenomenon of hotel flash sales from three different perspectives: (1) industry; (2) consumer; and, (3) future research directions. In the first study, a flash sales evaluation framework was developed ba sed on 46 interviews with hotel managers from different segments of the industry using grounded theory method. The findings from this study revealed that flash sales websites are Key advantages/disadvantages of using such a distribution channel fell in the categories of inventory management, revenue management, brand marketing, customer relationships, and operational challenges.
14 The second study compared a consumer profile of fla sh sales and other hotel leisure customers, using a cross sectional online survey of 358 individuals. The results of the study revealed that flash sales customers do not differ from other leisure customers with regard to their price consciousness, intentio ns to revisit, and intentions to recommend the hotel to others. However, flash sales customers demonstrated higher levels of market mavenism, quality consciousness, and variety seeking when compared to other leisure travelers. The third study of this disse rtation outlined research directions for further development of the hotel flash sales research stream. The study is based on the findings from the first two studies and the review of literature available in consumer behavior and promotions areas. The sugge sted research directions include: exploring whether flash sales websites generate incremental demand by attract ing new customers, determining whether flash sales customers are value conscious or promotion prone, and extending the profile of flash sales cus tomers .
15 CHAPTER 1 INTRODUCTION Background Information Frew, 2004) and pricing opport unities for businesses (Grewal et al. , 2011). For the hotel industry, such opportunities m ay include distribution via online travel intermediaries that offer capabilities of dynamic pricing, name your own price (NYOP), and flash sales. Of these, flash sales websites appear to be a growing source for reaching new customers and filling immediate hotel occupancy needs (Piccoli & Dev, 2012). Flash sales websites are a relatively new electronic distribution channel often offering time limited deep discounts (of about 50%) for advance purchase of products or services (Edelman, Jaffe, & Kominers, 2011 ; Piccoli & Dev, 2012). The flash sales business model involves the deep room discounts that are promoted for consumer purchase, and high intermediary commission costs (Piccoli & Dev, 2012). This type of business model makes the cost of hotel room distrib ution relatively high when compared to other distribution channels (Starkov & Kirby, 2011). However, despite the high cost, the use of flash sales seems to be a growing trend among hotel managers to distribute their room inventory (Piccoli & Dev, 2012). In deed, increase market share by attracting new consumers or to encourage increased occupancy rates during anticipated low demand seasons. Examples of the hotel flash sales may be observed on flash sales websites such as Groupon Getaways, LivingSocial Escapes, Jetsetter , and SniqueAway . For example, in one year of
16 operations about 500,000 room nights were booked via LivingSocial Escapes since its launch in November 2010 (LivingSocial, 2011). Revenue Maximization in Hotels Timely distribution of the hotel room inventory is critical for the industry due to the perishable nature of the product (Croes & Semrad, 2012 a, b ; Desiraju & Shugan, 1999; Kimes, 1989). Perishability is a common characteristic of all service based products, and refers to an inability of the service provider to store the product and sell it at a later time (Zeithaml , Bitner, & Gremler , 2006). Revenue that is lost from having an unsold product is usually l ost forever. In application to the hotel industry this means that if a hotel fails to sell all rooms on a particular night, this inventory will perish. When a new day comes, a hotel will be selling rooms for that new day, and will not be able to ever go ba ck to the previous dates to sell the rooms that stayed vacant. Therefore, hotel operators strive to attract as many customers as possible on a given night to maximize However, there are so me other parameters that are influential with regard to revenue maximization in hotels. Such parameters should include seasonality, variable demand, and pricing strategies. Seasonality presents a problem for the lodging industry due to the fixed capacity o f the hotel properties and variable demand from the consumers (Weatherford & Kimes, 2003). This means that a hotel that has 100 rooms has the same 100 rooms every day, however, based on the season, consumers may demand all of these rooms, or just a small p ortion of them. A challenge of fixed capacity originates in inability to increase or decrease supply based on the demand fluctuations (Kimes, 1989). For example, a laptop manufacturer may produce more or less laptops based on the demand on the market, how ever, the
17 hotel in our example will always have 100 rooms, and no ability to offer less rooms when demand decreases, and more rooms when demand increases. In the hotel industry demand may vary based on the day of the week (e.g. a leisure oriented hotel wil l have higher demand on the weekends), month or a time of the year (e.g. hotels in Florida experience decrease in demand during summer months, while hotels in South Carolina experience increase of demand during the same period). Taken together, perishabili ty and seasonality create a significant problem for the hotel revenue maximization when hotels have a fixed supply, but at the same time, experience variable demand. This strengthens the challenge of filling perishable rooms that hotel operators face on th e day to day basis. Another variable that is important to take into consideration for revenue maximization in the lodging industry is price. The response of the consumer demand to price fluctuations in the hotel industry may be considered from three differ ent points of view: 1) Ability to pay; 2) Willingness to pay; and 3) Desire for a product (Hayes & Miller, 2011). One may expect that revenue maximization requires the highest possible prices. However, at the same time, some consumers may not be able to af ford such prices, some may not be willing to pay premium prices, and others may not desire the product all together. All three scenarios regarding demand in the hospitality industry magnify the perishability problem that was discussed earlier. Given all t his, the hotel industry is trying to find a trade off between charging the highest possible prices to a few consumers, or lower prices to a wider group of consumers. Revenue Management Prerequisites Variable pricing with incorporation of discounted rates have been widely utilized in the hotel industry as one of the revenue management techniques (Croes & Semrad,
18 2012 a, b ; Hanks, Cross, & Noland, 2002; Nusair, Yoon, Naipaul , & Parsa, 2010; Rohlfs & Kimes, 2005). Revenue management rmation systems and pricing strategies to allocate the right capacity to the right customer at the right price at (Kimes & Wirtz , 2003 , p. 1 2 5). The main purpose of the variable pricing in the hotel industry is to match the demand and suppl y in order to fill the available fixed capacity, sell the perishable product and maximize revenues. There are several characteristics that are required from an industry for a successful implementation of the revenue management approach: 1. Perishable inventor y; 2. Variable demand; 3. Advance sale of products; 4. Market segmentation; 5. Low variable costs, high fixed costs; 6. High marginal production costs (Hanks et al., 2002; Kimes, 1989). The hotel industry possesses all of these characteristics. The perishability of the inventory and the variability of the demand have been already discussed before. Advance selling of products refers to the sale in advance of the consumption period (Shugan & Xi e, 2004, 2005; Xie & Shugan, 2001). This approach is widely used for service bas ed products including lodging. In the advance period both consumers and hotel managers face uncertainty: consumers are uncertain about their future valuations of the hotel stay, while hotel managers are uncertain about future demand and appropriate pricing . However, advance selling seems to benefit hotels from the perspective of bringing in those customers who would not purchase in the consumption period when the uncertainty is eliminated. Another advantage of using advance selling may be in monitoring the hotel room demand, and adjusting prices to maximize revenue.
19 Market segmentation is another requirement for successful revenue management (Hanks et al., 2002; Kimes, 1989). Market segmentation allows hotel managers to distinguish between different consume r groups, and develop pricing strategies (i.e. rate fences) for each group based on their needs and willingness to pay. In the hotel industry, consumers are often segmented based on their purpose of travel (e.g., leisure vs. business travelers) or particul loyalty program, AAA membership, employment in a partner company, etc.). The hotel industry is also characterized by having high fixed costs of operation and low variable costs for servicing the rooms. The high fi xed costs of hotel operations include such expenses as property cost, utilities, and staffing. However, after the fixed costs are covered it does not cost much more to sell and service an additional room. At the same time, the marginal production costs for the hotel industry are high. Marginal production costs refer to the expense of producing an additional unit of product. In application to the hotel business, it means a cost of adding one room to the hotel inventory. It is unlikely and costly for a hotel operator to add one more room to a hotel property. Such extension would usually require a major redesign of the property or a construction of a new building. Rooms are usually added in groups. High marginal production costs highlight the challenge of the f ixed capacity that was discussed earlier, and explain why hotels cannot easily adjust supply in response to variable demand on the market. Flash Sales and the Revenue Maximization Problem Based on the aforementioned characteristics of the lodging industry, hotel flash sales may seem to assist in solving the revenue maximization problem. In general, flash sales websites possess the following characteristics (Ayadi, Giraud , & Gonzalez, 2013):
20 1. Offer products at discounted prices; 2. Offer products in limited quan tities; 3. Make the offers available for purchase for a short period of time; 4. Make the offers available to the members of the website only. Once we match the characteristics of the flash sales websites with the characteristics of the hotel industry that enab le revenue management, we may see how flash sales may support revenue optimization tasks in the hotel industry. First, offering products at discounted prices creates variability in pricing structure and makes available hotel room inventory marketable to a wider audience. This may help to address the problems of perishability and variable demand. For example, discounted prices may assist hotel managers in attracting customers during the times of low demand, and, therefore, protect room sales from perishing. On the other hand, selling rooms at a discounted price comes at a cost of potentially losing revenues via consumer surplus. Also, flash sales allow hotel managers to offer limited quantities of rooms for sales via this distribution channel. This may allow hotels to sell some portion of the inventory at discounted prices, and also keep another portion of the hotel rooms for sale at higher prices. The fact that flash sales offers are time limited will require customers to complete their purchase in advance. X ie and Shugan (2001) demonstrated that discounted advance selling with a limited number of units sold in advance may be profitable for service providers with small capacity and small marginal costs, which is representative of the hotel industry. And, fina lly, website membership that is required to receive flash sales offers may allow hotel managers to segment out this group of customers, and apply different pricing strategies to other customer groups. On the other hand, the flash sales websites
21 databases a re becoming so large that mass mailings of flash sales offers may interfere with hotel customer segmentation practices and strategies. Another way that flash sales may offer in terms of influencing hotel revenue maximization is by generating revenues in no n room operating departments. According to Anderson and Xie (2010) and their review of the revenue management literature that property (p. 53). This means that hotel managers would want to consider not only room revenues, but also revenues from all other operating departments in order to evaluate profitability of the property. Dunn and Brooks (1990) suggested that as the number of rooms sold increases, a hotel may observe the revenues from other departments to increase as well. This seems to indicate a direct relationship between increased occupancy rates and increased revenue maximization for the entire hotel as a unit. Therefore, with the goal of hotel flash sales to increase occupancy rates through the sale of additional rooms during low demand seasons the flash sale promotion may assist hotel properties in the productivity levels of non room operating departments, such as restaurants , bars, spas, gift shops, and other. Hotel managers reported that on average flash sales customers spent about 29% of the promotional value in additional expenditures while at the property (Piccoli & Dev, 2012). However, such benefit would depend on the s ervice level of the hotel and available infrastructure. For example, as the number of guests from the flash sales promotions increases, a limited service hotel, which offers a few additional services,
22 may not see as large an increase in non room sales as w ould a resort with numerous facilities. Statement of the Problem The trend of using flash sales websites to distribute room inventory is reflected in the growing attention to the website concept in the hotel trade literature (Gupta, 2012; Mayock, 2012; Sch aal, 2011a, 2011b, 2012a, 2012c). In this literature, flash sales websites were credited with advantages such as adding exposure to new customers (Schaal, 2011b), filling immediate occupancy needs (Schaal, 2012c), and promoting hotel properties online (Gu pta, 2012). However, hotel flash sales were also credited in this literature with several disadvantages that included: criticism for steep discounts and high commission costs (Schaal, 2012c), violation of price parity (Schaal, 2012a; Starkov, 2011; 2012), and destroying brand integrity (Starkov, 2011). Moreover, Starkov perspectives. This ma y suggest that hotel managers use flash sales out of desperation to sell the perishable hotel inventory even at incredibly high distribution costs and deep discounts. The controversy between the benefits and drawbacks of the hotel inventory distribution v ia flash sales websites constitutes the research problem of this study. The effect of the flash sales websites on the hotel industry is uncertain yet this channel is frequently used. For example, during the first year since LivingSocial Escapes was launche d, the website featured hotel deals from about 800 properties and sold about 500,000 room nights, which corresponds to about 45,000 room nights per month, and 1,500 travelers per night (LivingSocial, 2011). After 16 months of operations,
23 LivingSocial Escap es doubled the numbers and reached the mark of one million room nights sold (Schaal, 2012b). Furthermore, the extant academic literature does not mirror the hotel trade literature interest on the topic of hotel flash sales. In fact, academic literature is almost devoid of empirical studies that could offer guidelines to hotel managers regarding the benefits, drawbacks, and guidelines for how and when to use such a distribution channel (Edelman et al., 2011). A comprehensive review of literature via databas es ABI/INFORM Complete, EBSCO HOST and Hospitality & Tourism Complete on the topic of hotel flash sales revealed only one descriptive academic article (Piccoli & Dev, 2012) published in scholarly journals by July 201 4 . Consequently, there is a need for aca demic research that would provide a foundation for understanding the benefits and drawbacks of distributing hotel inventory via flash sales websites. Distribution channel research has become a prominent area in the hospitality research (Connolly, Olsen, & & Frew, 2002, 2004). One of the main reasons for this is a necessity to move perishable hotel inventory. Flash sales websites seem to assist with distribution of perishable inventory by providing de ep discounts and generating consumer demand. However, flash sales websites also have high distribution costs associated with them. There seems to be a trade off between increasing sales and incurring high distribution costs. Such a trade off may pose some difficulties for hotel managers as it pertains to the a study of benefits and drawbacks of the hotel flash sales would contribute to the understanding of the flash sales impact on the hotel industry, and guide hotel managers
24 in their decisions to include (or exclude) flash sales websites in the distribution channel mix. Research of the hotel flash sales would provide guidance to hotel managers with regard to measuri ng the hotel flash sales effectiveness. Understanding the benefits and drawbacks attributed to the inventory distribution via flash sales websites could assist hotel managers in making informed decisions about adoption of flash sales websites for hotel inv entory distribution. A study of the hotel flash sales will also provide a foundation for future research, and, thus, be beneficial for faculty and graduate students who are interested in the hotel electronic distribution channels and discounting topics. Pu rpose of the Study and Research Questions In order to address the research problem, this dissertation follow s a three paper format. The research stream from these three studies investigate s an impact of the flash sales on the hotel industry from three diff erent perspectives, including: industry, consumers , and research . The overall purpose of this dissertation is to provide a foundational level regarding the use of hotel flash sales and provide an understanding of flash sales benefits and drawbacks to the a cademic and professional audiences. A more specific research purpose for each of the three studies is presented below while a comprehensive explanation of each of the studies follows in Chapter 2, Chapter 3, and Chapter 4 . Study 1: Industry Perspective Th e purpose of the first study is to investigate the benefits, drawbacks, and performance measures of room inventory distribution via flash sales websites from the s a three -
25 fold i nquiry addressing factors that may influence flash sales adoption by hotel managers : 1. The first inquiry is to identify and document some of the benefits and drawbacks of using flash sales websites for hotel room inventory distribution as perceived by hotel managers. 2. The second inquiry is to investigate how hotel managers measure the performance of flash sales distribution. This inquiry focus es on understanding the key metrics of flash sales evaluation. 3. The third inquiry of this study is to assess the ass umption recognized in the literature (Piccoli & Dev, 2012) that a portion of revenue that is lost in the deep discounted room rates as offered on flash sales websites may be recaptured er revenue operating departments. In order to address the purpose of the first study and the research problem, the following research question was formulated: RQ1. evaluation strategies for the adoption of flash sales websites as a hotel inventory distribution channel? Study 2: Consumer Perspective The purpose of the second study is to investigate the profile of hotel flash sales customers, and hotels . Guided by the revenue management principles, hotel managers may want to distrib customer (Kimes & Wirtz, 2003). Previous literature rofitability and satisfaction (Woo & Fock, 2004). However, current academic literature is devoid of studies that would provide understanding of whether the customer attracted to hotels via flash sales websites is a right or wrong for hotels. Developing an understanding of the flash sales customer profile and how it compares to other leisure customers may contribute to solving the problem of
26 controversy of the flash sales advantages and disadvantages. Together with the first study (industry perspective), the second study may confirm or disconfirm hotel Therefore, t he purpose of the second study is two fold : 1) to determine the profile of the customers who purchase hotel flash s ales deals, 2) to determine whether the flash The second purpose of the study is assessed with regard to: determin ing short ter different operating departments in the hotel and the associated expenditures), and determining long term profit revisit the hotel and recommend the hotel to others ). In order to address the purposes of the second study and the research problem of this dissertation, the following research questions were stated for study 2 : RQ1. What are the key profiling traits of the customers who purchase hotel flash sales deals? RQ2. Is there a difference between the economic contribution of the flash sales term profitability? RQ3. Is there a difference between intentions to revisit a hotel demonstrated by the flash sales customers and other customers? RQ4. Is there a difference between intentions to recommend a hotel to others demonstrated by the flash sales customers and other cus tomers? Study 3: Research Directions Given the trend of hotel inventory distribution via flash sales websites and a lack of academic literature on this topic, the purpose of the third study in this dissertation is to identify potential directions for fut ure research in the area of hotel flash sales. The
27 advantages and disadvantages of hotel flash sales present a controversy of the hotel flash sales phenomenon, and highlight an uncertain effect of such a distribution channel on the lodging industry. Given this, the third study proposes to identify the directions for future empirical research that could assist academia in developing future research studies. A clear outline of potential research directions on flash sales websites may stimulate the developmen t of this research area, and assist in closing the literature gap. The growth of the hotel flash sales research stream may provide a deeper understanding of the flash sales phenomenon, and, potentially, eliminate the controversy in the available literature regarding the use of hotel inventory distribution via flash sales websites. This paper rel ies on the academic literature from related disciplines (e.g., consumer behavior, revenue management , sales promotions ) and findings from the first two studies of th e dissertation . Definitions of Terms The following terms are used across all three studies of this dissertation. B ENEFITS OF HOTEL FLA SH SALES . Advantages that hotel managers may achieve by distributing room inventory via flash sales websites. B RAND IMAGE . (Kandampully & Suhartanto, 2000). D RAWBACKS OF HOTEL FL ASH SALES . Disadvantages or threats that hotels may experience as a result of distributing room inventory via flash sales websites. F LASH SALE S WEBSITE . A website offering time limited deep discounts (of about 50%) for advance purchase of products or services (Edelman et al., 2011; Piccoli & Dev, 2012) including Grou pon Getaways, LivingSocial Escapes, and Jetsetter . P RICE FAIRNESS . Customers' perceived balance between product/service attributes and its listed price (Choi & Matilla, 2009). This concept also applies to
28 the consumer evaluation of the prices for the sa me product/service quoted to different customers and/or listed on different retailing outlets. R EFERENCE PRICE . An expected price for a product or service that is formed (Mazumdar , Ra j, & Sinha , 2005; Niedrich , Sharma, & Wedell , 2001). S ERVICE QUALITY . A level of service delivery based on customer perception of the service performance in relation to customer expectations (Zeithaml et al., 2006). Significance of the Study The significan ce of this study is in its attempt to close the literature gap on the topic of hotel slash sales, and developing an understanding of the flash sales phenomenon. As a recently emerged hotel distribution channel (Piccoli & Dev, 2012), flash sales contribute to the complexity of the distribution channel choice in the hotel industry (Choi & Kimes, 2002 b Electronic distribution channels allow hotels to increase their customer reach and increase sales. However, a hotel may not add distribution channels infinitely and participate in all of them. Therefore, hotels face a problem of evaluating distribution channels in order to make smart decisions when forming a distribution channel mix. Managing multiple distributio n channels brings particular challenges to the hotel industry. Such challenges & Frew, 2004), and revenue management difficulties (e.g. managing distribution costs a nd prices across multiple channels) (Choi & Kimes, 2002). Therefore, precise managerial decisions with regard to inclusion or exclusion of a particular channel in the distribution channel mix.
29 The first proposed study is designed to advance the development of the hotel flash sales research stream by providing guidance to hotel managers regarding the measurement of the hotel flash sales effectiveness pertaining to acquiring new cus tomers, stimulating repeat sales, and generating revenues. The benefits, drawbacks, and performance indicators of the hotel inventory distribution via flash sales websites may be some of the key determinants in managerial decision making regarding the adop tion of flash sales websites. Managers might then be able to determine whether proper management of a perishable core product. The second proposed study aims to provide an understanding of the flash sales customer segment and an empirical evidence of the profile of the customer who is attracted to hotels by flash sales . This study contribute s to a deeper understanding of and drawbacks identified in the first study, and potentially provide some foundations for the developments of further research questions in the third study. Overall, with the development of this academic research, hotel managers may gain evidence of the f lash sales influence on the industry and consumers. Understanding of flash sales effects may enable hotel managers to make more informed decisions regarding hotel promotion and room distribution. A need for developing further research in the hotel flash sa les area is defined by the growing number of hotel room nights sold via this distribution channel. As it was mentioned earlier, just LivingSocial Escapes alone sold about half million hotel room night in the first year of operations (November 2010 Novemb er 2011), and then doubled the pace by selling the same amount of room nights during the four
30 consecutive months (until March 2012) (LivingSocial, 2011; Schaal, 2012b). Penetration of flash sales in the hotel distribution mix will require empirical researc h to be conducted in order to obtain a clear understanding of the impacts this channel may have on the industry and consumer behavior. The third study of this dissertation contribute s to establishing a foundation that may be used for the further developme nt of the hotel flash sales research stream. This may assist faculty and graduate students in developing further research inquiry of hotel flash sales. In the long run, the growth of the flash sales research stream may assist in closing a literature gap, b uilding a knowledge base, and providing an understanding of the flash sales phenomenon to the academic and professional audiences. Theoretical Background Given the purpose and structure of this dissertation, several theoretical frameworks may be used to e xplain the use of flash sales websites in the hotel industry . The use of flash sales websites for hotel room inventory distribution may rely on the theory of peak load pricing and the cost/benefit analysis theory. T he theory of peak load pricing may explai n the use of discounted pricing offered on hotel flash sales in the off peak periods. Also, the cost/benefit approach may provide structure to the evaluation of the advantages and disadvantages that hotel managers may perceive in using flash sales website as an inventory distribution channel for the lodging industry. Theory of Peak Load P ricing The theory of peak load pricing originates in early woks of Bye, Boiteux, and Steiner (Crew, Fernando, & Kleindorfer, 1995). Peak load pricing refers to the pricing of perishable (non storable) product/services with variable demand. Theory of peak load pricing originates in economics literature. In its development it goes from simple
31 deterministic models (in 1920 1950s) to more sophisticated models accounting for dem and uncertainties (in 1960 1970s). The theory was originally developed for the monopolistic markets, such as electric utilities. However, later this theory was applied to the competitive industries, such as airlines and hotels. The peak load problem arise s from the variable demand for products and services (Crew et al., 1995). If the price for such product or service would be set at a constant level, then the market would observe periodical demand fluctuations. In order to meet the demand in the high peak seasons, companies would need higher capacity than in off peak seasons. At the same time, when the market moves from the high peak season to the off peak season and keeps the capacity constant, a portion of that capacity will be underutilized based on the decreased consumer demand. However, the existing capacity has a particular cost associated with it (i.e. fixed costs of operation). Therefore, the theory of peak load pricing suggests that in order to maximize the profits and capacity utilization, the pric es should be adapted to the demand level. The theory of peak load pricing has been applied to research in the transportation and services industries (Crew et al., 1995). One of the examples includes time of the day seasonal pricing for daily underground ti ckets. Also, applications of the peak load theory were found in the airline industry with regard to time of the day seasonal pricing of airline landing and parking fees. Keeler and Small (1977) applied the theory of the peak load pricing in order to determ ine the optimal pricing and service level on the expressways in California. Braid (1996) also employed this theory to the subject of freeway and toll road pricing.
32 The peak load pricing theory also has been utilized for research on natural resources and u tilities (e.g., De Alessi, 1977; Manning, Mitchell, & Acton, 197 9; Mitchell, Manning, & Acton , 198 1 ; Wenders, 1976). The applications of this theory have been found in the tourism and hospitality industry as well. Manning and Powers (1984) have utilized th e peak load pricing theory in order to investigate an opportunity of shifting a park load from the peak to off peak period, and, therefore, optimize the use of resources. This theory has been also utilized by Chung (2000) in order to explain different pric ing strategies for a luxury hotel in Seoul, Korea. In his work, Chung (2000) relies on the peak load pricing in order to explain a need to set high prices in a high demand period and lower prices when the demand goes down. Given all this, the theory of pea k load pricing appears to be applicable to the research in the hotel flash sales area. This is because the hotel product is perishable and characterized by demand fluctuations. Also, the hotel capacity is fixed. There may be some fluctuations in the hotels work. Therefore, hotels also face the problem of pricing fixed capacity in the variable demand conditions. Based on this, the theory of peak load pricing may provide an insight into the behavior of ho tel managers of providing deep discounts and selecting flash sales websites as hotel inventory distribution channel. Cost/Benefit A nalysis The cost/benefit analysis (CBA) originated in the work of Beach and Mitchell (1978). This theory postulates that deci sion the strategy which The cost/benefit approach suggests that every decision is being made based on the evaluation of the costs (e.g. effort, time, and money) and benefits (outcomes of the decision). Also,
33 decision experience, knowledge, and awareness of different strategies would influence the final outcome and the choice of a strategy. Researchers in vast and diverse areas in academia have used cost benefit analysis as an economic evaluation method (please see Table 1 1 below). When applied to the hotel flash sales research, cost/benefit analysis may be one of the perspectives that are u sed to adopt or continue the use of hotel flash sales for the hotel inventory distribution. When running a hotel business, hotel managers may pose a number of different objectives that they consider important for their operations. Such goals may include in creasing occupancy, increasing average daily rate (ADR), increasing revenue, reaching a new market segment, building a particular brand image, etc. Depending on the property and the types of management goals, different strategies will be evaluated. Flash s ales websites may be one of these strategies. According to the cost/benefit analysis, hotel managers would evaluate an investment that is necessary to engage in the flash sales distribution, e.g. cost of distribution, employee training, convenience, time commitment and other relevant factors. The identified costs should be compared with expected benefits, such as room nights sold, contribution to the revenue, exposure to a new market, etc. Research Design Research design of the studies include s both qualit ative and quantitative methods. Given the lack of academic literature on the topic of hotel flash sales, the first study undertake s a qualitative approach in order to generate an in depth understanding of the industry perspective on the use of flash sales websites for the lodging properties. The first study uses a grounded theory approach to build a flash sales evaluation
34 framework that would allow explor ing a broad perspective on the advantages and disadvantages of flash sales , and determin ing the performa nce measures of hotel flash sales . The theoretical sampling of this study include s different players of the hotel industry including: independent hotels, chain hotels, hotel management companies and chain corporate offices. The selection criterion for the inclusion in the study is should have prior knowledge on the concept of flash sales whether through research or participation in flash sales. Such selection proce ss is identified in order to include those hotel managers who have considered flash sales as an inventory distribution channels, but decided not to participate. Participants for this study are recruited by using two approaches. First, an invitation to par ticipate in the study was forwarded to the general managers of the hotels that appear on hotel flash sales websites (e.g. Groupon Getaways, LivingSocial Escapes, or Jetsetter ). Second, an invitation to participate in the study was also distributed to th e top managers (general managers, CEO, and Presidents) of hotels, hotel brands, and hotel management companies via the American Hotel & Lodging Association (AH&LA). Every participant of the first study was interviewed with regard to the advantages, disadv antages, evaluation measures of flash sales performance, and impact on the sales phenomenon in the lodging industry, and serve as a foundation for the next two studies.
35 The second study of this dissertation (consumer perspective) focus es on profiling the flash sales customers who purchase hotel deals, and evaluating whether flash sales guests with regard to their short and long term contribu . These research questions come out of the preliminary data collection for the first study and literature review for the second study. This study of the dissertation include s an empirical assessment of the flash sales consu mer profile and consumer economic impact on short term and long term hotel profitability. The study employ s a cross sectional survey design and utilize s an online mode of data collection. The results of the study are expected to provide an understanding of what type of customer purchases hotel flash sales promotions, and what economic contribution flash sales customers bring to the hotel properties. The third study of this dissertation focus es on developing research direction s for the area of hotel flash s ales. As it was mentioned earlier, the area of hotel flash sales is performance and consumer behavior is uncertain. Therefore, further studies may enhance understanding of this particular area and allow hotel managers to make better informed decisions about including (or excluding) flash sales websites from the distribution mix. The third study of this dissertation rel ies on the findings from the first two studies, and the extensive review of literature that is available on the topics of discounting, sales promotions, consumer behavior, and other related areas in order to build research direction for future studies. Therefore, propositions and concerns expressed by the hotel managers in the first study are matched with available academic literature in order to outline the directions for future research in the hotel flash sales
36 area. A more detailed description of the research methods for studies 1 , 2 and 3 is presented in C ha pter 2 , Chapter 3 , and Chapter 4 respectively.
37 Table 1 1. The use of cost benefit analysis in different research fields Field Citation Description Hospitality field Clements and Josiam (1995) used CBA in evaluating a training program in a hospitality o rganization Hanley and Barbier (2009) economic evaluation method Tourism field Burgan and Mules (2001) reexamined CBA in evaluating the commercial viability of tourism events in terms of the economic impacts that the events bring to their host region Mules and Dwyere (2005) used CBA in evaluating public sector support for tourism events Transportation field Kidokoro (2004) used CBA in estimating transport networks to solve congestion
38 CHAPTER 2 THE MANAGERIAL FLASH SALES DASH: IS THERE ADVANTAGE O R DISADVANTAGE AT THE FINISH LINE? Background of the Study Flash sales websites are a relatively new electronic distribution channel often offering time limited deep discounts (of about 50%) for advance purchase of products or services (Piccoli & Dev, 2012 ). The use of flash sales websites for hotel inventory share by attracting new consumers or to encourage increased occupancy rates during anticipated low demand seasons. How ever, these efforts may only be realized after incurring a relatively high intermediary commission cost from the flash sales websites after the websites have sold the room nights at a deep discount to consumers. The effective and timely distribution of hot el room inventory is critical for managers due to the perishable nature of the most significant revenue driver, that is the core product (room nights) (Croes & Semrad, 2012a; Desiraju & Shugan, 1999; Kimes, 1989). Perishability is a common characteristic of all service based products, and refers to an inability of the service provider to store the product and sell it at a later time (Zeithaml et al., 2006). In other words, revenue that is lost from having an unsold room night is lost forever (Croes & Semr ad, 2012b). Therefore, hotel managers strive to attract as many customers as possible on any given night to maximize the potential Purpose of the Study The purpose of this study is t o investigate the benefits, drawbacks, and performance measures of room inventory distribution via flash sales websites. Flash sales websites seem to assist with the distribution of perishable inventory by providing
39 time limited deep discounts that generat e consumer demand. However, the use of flash sales websites to generate consumer demand in order to avoid room night sales from possibly perishing comes with an associated high distribution cost. Therefore, there seems to be a trade off for managers betwe en increasing room sales and incurring high distribution costs. Such a trade off may pose some difficulties for hotel managers as it channel. In this regard, a study of ben efits, drawbacks, and performance measures of industry, and guide hotel managers in their decisions to continue (or discontinue) distribution via flash sales websites . In order to recognize this purpose, the study uses a three fold inquiry addressing factors that may influence flash sales adoption by hotel managers. The first inquiry is to identify and document some of the benefits and drawbacks of using flash sales we bsites for hotel room inventory distribution as perceived by hotel managers. Given that there is a multitude of different distribution channel combinations for a manager to choose from, it becomes critical for managers to clearly understand the advantages and disadvantages of each of those channels in order to maximize room sales. Identification of the hotel flash sales benefits and drawbacks is important in the actual selection of flash sales websites as a distribution channel. Hotel managers may arrive at the final decision about whether or not flash sales websites should be included in the distribution mix by considering the importance of potential benefits and risks for their specific hotel property. Selecting appropriate distribution channels for the marketing of products and services offered by a hotel plays a key role in enhancing its
40 business. Even though literature indicates that hotels have been involved in multi channel distribution (Morosan & Jeong, 2008), participation in all distribution chann els does not necessarily provide the most benefits to a hotel (Kotler , Bowen, & Makens , 2010; Nyheim , McFadden, & Connolly , 2005). Therefore, it may be considered useful to hotel managers to have research that indicates whether or not the use of flash sale s websites seems to be or not to be a worthwhile addition to their distribution mix. The second inquiry is to investigate how hotel managers seem to measure the performance of flash sales distribution. Understanding the sales performance measures for fla sh sales websites may assist hotel managers in the evaluation of such websites in order to determine if the distribution channel is an appropriate vehicle to move room inventory. This inquiry focus es perfor mance evaluation. Such understanding is important from two perspectives. First, it may benefit hotel managers to learn how their colleagues measure the performance of the flash sales promotions and make an evaluative judgment about the effectiveness of us ing flash sales as a room inventory distribution channel. This may assist those hotel managers who have never tried flash sales to set the goals for their flash sales campaign and guide them in determining which measures they may use to evaluate the flash success. Also, managers who have tried flash sales may also benefit from considering other evaluation options employed by different hotel managers. Second, a clear understanding of the flash sales performance evaluation metrics employed by the hotel industry may benefit researchers who would like to develop further studies in this area. Such information on the evaluation metrics of hotel flash sales may assist researchers
41 who are studying the impact of flash sales on the lodging industry and who are specifically developing sound measurement scales to assess that impact. The third inquiry of this study is to explore, in a qualitative context, if an assumption recognized in literature pertaining to the portion of revenue that is lost in a deep discoun ting process (e.g. distribution of rooms on a flash sales website) may be (Piccoli & Dev, 2012). For example, a consumer may purchase a flash sales deal that includes din ing, entertainment, and/or other services that may consequently encourage the consumers to patronize different revenue generating departments in the hotel. When preparing a flash sales deal hotel managers may design a package in a way to stimulate the cust omer to consume beyond the package deal. For example, this goal may be achieved by incorporating an appetizer discount at the hotel restaurant in the flash sales package. This way, customers may be encouraged to patronize the counted appetizer of their choice, but then order entrees at a full price. Thus, flash sales may be resulting in an increase in revenue productivity in other operating departments to offset the loss of sales in the rooms department due to the deep discount offered in the flash sale. Similarly, this assumption may be applied to other non room operating departments in a hotel, e.g. bars, lounges, spa, golf, and other. A preliminary assessment of this assumption presented in literature may provide a foundatio n for further empirical development on the topic. Statement of the Problem Currently, the trend of using flash sales websites to distribute room inventory is not widely addressed in hospitality related literature. A review of academic literature reveals o nly one descriptive study by Piccoli and Dev (2012) that presented the
42 adoption and usage statistics of the flash sales websites in the hotel industry. The results of this study revealed that about 42% of hotel managers participating in the study have used flash sales as an inventory distribution channel, 46% have not tried this channel yet, and 12% of the hotel managers were considering the adoption of this distribution channel. The study suggested that the reasons for offering flash sales deals fall into the categories of branding, customer acquisition, occupancy, revenue, profits, and desperation. The use of flash sales websites is more prominently referenced within hotel trade literature. Topics included within trade literature primitively address the a dvantages and disadvantages of such a distribution channel, and describe the actual entrance of flash sales websites into new markets (Gupta, 2012; Mayock, 2012; Schaal, 2011a, 2011b, ability to increase exposure to new customers (Piccoli & Dev, 2012; Schaal, 2011b), fill immediate occupancy needs (Schaal, 2012c), and assist in the promotion of hotel properties online (Gupta, 2012). The use of flash sales websites have also been crit icized in literature for steep high commission costs (Piccoli & Dev, 2012; Schaal, 2012c), violation of price parity (Piccoli & Dev, 2012; Schaal, 2012a; Starkov, 2011; St arkov, 2012), and destroying brand integrity (Piccoli & Dev, 2012; Starkov, 2011). Taking into account such disadvantages of distributing rooms via flash sales websites, Starkov (2011) names
43 lack of empirical studies pertaining to the use of flash sales websites in the hotel industry and the stark contrast among industry analysts who fl ip flop between favorably crediting and severely criticizing the use of flash sales as a distribution tool makes it Therefore, the overall influential effect of hotel flash sales web sites on the lodging industry seems at best uncertain. Based on this uncertainty, there is a need for the advancement of academic literature that could offer directions to hotel managers regarding the benefits, drawbacks, and guidelines for how and when to use such a distribution channel. Research Question Based on the research problem and research purpose of the paper the following drawbacks, and evaluation strategies for t he adoption of flash sales websites as a hotel inventory distribution channel? Significance of the Study effectively manage a perishable inventory is crucial for hotel managers if they are to Frew, 2002). The proper selection of distribution channels assists hotel managers in enhancing the business by more effectively reaching target markets, info rming customers about hotel availability and rates, and generating sales. There are a vast customer reach and sales. Managing multiple distribution channels brings parti cular challenges to the hotel industry. Such challenges may include technical difficulties (e.g.
44 management difficulties (e.g. managing distribution costs and prices across multip le channels) (Choi & Kimes, 2002a). The emergence of flash sales as a distribution method contributes to the complexity of the distribution channel choice in the hotel Given t he potential challenges associated with managing multiple distribution channels, hotel managers may not continuously add an infinite number of channels to their distribution mix. Therefore, managers may face a problem in determining if newly emerging dist ribution channels, such as flash sales websites, have a place in their distribution mix when the long term effects of such distribution have not yet been empirically assessed. The benefits, drawbacks, and performance indicators of hotel inventory distribu tion via flash sales websites may be some of the key factors in the managerial decision making regarding the adoption of flash sales websites. Understanding the benefits of hotel flash sales may provide hotel managers with the insight required to determine if flash sales possess the necessary ability to assist in managing a perishable product. Whereas, understanding the drawbacks associated with hotel flash sales may be relevant in determining potential threats of using such a channel. And, learning how ot her managers assess the performance of flash sales may provide objective metrics for hotel managers to assess their own success with using flash sales promotions. Taken all together, these factors may present a comprehensive review of the flash sales pheno menon in the lodging industry with regards to the relevant evaluation of this
45 distribution channel. The results from this study may be meaningful for future research that studies hotel electronic distribution channels and discounting topics. Literature Re view A review of relevant literature has demonstrated that flash sales may be referred to as daily deals (Picolli & De v, 2012), private sales (Ayadi et al. , 2013; Kim & Martinez, 2013; Picolli & Dev, 2012), and online coupons (Sigala, 2013). For the purpos e of The phenomenon of flash sales attracted attention of researchers in different fields, including fashion (Kim & Martinez, 2013), luxury products (Ayadi et al., 20 13), tourism (Dev & Falk, 2011; Sigala, 2013), and hotels (Picolli & Dev, 2012). These pioneering studies that appeared in different fields are united in the opinion that flash sales research is in its infancy and requires further development. Given the li mited number of studies published in the flash sales area, this paper surveys the literature from the advance selling, discounting, and distribution channel fields that may be relevant to the explanation of the flash sales phenomenon and understanding of i ts advantages, disadvantages, and performance measures. Advance S elling Advance selling refers to the process of selling services in advance of the consumption (spot) period (Shugan & Xie, 2004, 2005; Xie & Shugan, 2001). The main requirement for advance s elling is customer uncertainty about future value of the product (Xie & Shugan, 2001). In order to cope with uncertainty customers usually expect some benefit of purchasing in advance, for example, receiving a lower price (compared to the consumption peri od) or capacity that is constrained and may not be available at the consumption period (Xie & Shugan, 2001; Shugan & Xie, 2004).
46 Shugan and Xie (2005) suggest that advance selling of services may assist managers in generating profit improvements by stimula ting greater market participation. Advance selling is very common in the hospitality industry (e.g. airlines, cruise, lines, hotels, theme parks, excursion tours, theaters, dining) and in a number of other service industries (e.g. dry cleaning, conferences , sport events). Hotel flash sales represent one of the media that enable the advance purchase of hotel rooms. Therefore, it may be expected that flash sales (as an advance selling tool) may benefit hotels by stimulating greater market participation. Howev er, since flash sales offer deep discounts for advance selling of hotel rooms, it also becomes important to consider the impact of discounting on the effectiveness of flash sales as a hotel inventory distribution channel. D iscounted S ales The question of d iscounting in the hotel industry has been in the focus of researchers for decades (Abbey, 1983; Carroll, 1986; Croes & Semrad, 2012b; Enz, Canina, & Lomanno, 2004; Hanks et al. , 2002). This topic raised a heated debate in the attempt to understand whether discounting works for the hotel industry. The academic research presents discounting advocates (Croes & Semrad, 2012a; 2012b) as well as discounting opp onents (Canina & Enz, 2006; Enz et al. , 2004). Since hotel flash sales offer deep discounts for distribu tion of hotel room inventory, adoption of such a distribution channel also puts hotel managers in front of a dilemma whether discounting would work for their operations. Price manipulations, including discounted sales seem to be supported by the theory of peak load pricing. The theory of peak load pricing originates in the early works of Bye, Boiteux, and Steiner (Crew et al., 1995). Peak load pricing refers to the pricing
47 of perishable (non storable) product/services with variable demand. The peak load pro blem arises from the variable demand for products and services (Crew et al., 1995). If the price for such product or service would be set at a constant level, then the market would observe periodical demand fluctuations. In order to meet the demand in the high peak seasons, companies would need higher capacity than in off peak seasons. At the same time, when the market moves from the high peak season to the off peak season and keeps the capacity constant, a portion of that capacity will be underutilized ba sed on the decreased demand. However, the capacity has a particular cost associated with it. Therefore, the theory of peak load pricing suggests that in order to maximize the profits and capacity utilization, the prices should be adapted to the demand leve l. Consequently, observing deeply discounted sales on flash sales websites may suggest that hotels utilizing this pricing and distribution strategy are in the off peak season. At the same time, when considering the appropriateness of a discounting pricing strategy for a business, it is also important to evaluate the distribution channel where this pricing strategy will be implemented. Hotel Inventory Distribution Channel Evaluation Research on distribution channels has become an important part of the market ing and economics disciplines (Jeuland, & Shugan, 1983 , 2008 ; Shugan, 1985). Distribution channel research has become a prominent area in the hospitality and tour ism research as well (Connolly et al. , 1998; Emmer, Tauck, Wi lkinson & Moore, 1993; Gazzoli et al. people at the right time and in the right place to allow a purchase decision to be made, and provide a mechanism where the consumer can make a reservation and pay for the
48 functions of the hotel distribution channels would be providing information to consumers to facilitate dec ision making, and providing an outlet to complete the purchase. Development of information and communication technologies (ICT) has significantly altered hotel distribution practices by enabling numerous electronic distribution channels (Berne, Gar c ia Gon zalez, & Mugica, 2011; Buhalis & Law, 2008; Carroll & Siguaw, 2003; Emmer et al., 1993; Kracht & Wang, 2010). Hotel inventory distribution land scape includes hotel websites, central reservation systems (CRS), global distribution s ystems (GDS), and differen t online intermediaries. Online travel intermediaries include online travel agenc ies (OTAs) (e.g. Expedia ), m eta search websites (e.g., Kayak ), opaque travel sites (e.g. Priceline ), and recently flash sales websites (e.g., Groupon Getaways, LivingSocia l Escapes). The growth of hotel distribution channels requires hotel managers to be able to select those channels that would be the most beneficial for their properties. There are several approaches that may assist managers with this task. For example, th e cost/benefit analysis theory may be applied to a wide range of decision making tasks, including distribution channel evaluation. Another approach that was developed specifically for the hotel inventory electronic channel evaluation presents a framework t hat may be applied to the hotel distribution channels. Cost/benefit analysis theory The cost/benefit analysis originated in the work of Beach and Mitchell (1978). This theory postulates that decision the le suggests that every decision is being made based on the evaluation of the costs (e.g.
49 effort, time, and money) and benefits (outcomes of the decision). Also, decision p ersonality plays a role in the decision outcome. It is suggested that experience, knowledge, and awareness of different strategies would influence the final outcome and the choice of a strategy. Electronic distribution channel evaluation framework The elec 2004) has been developed in order to assist hotel managers with the evaluation process of potential electronic distribution channels. The framework includes six factors: marketing, operatio nal, technical, financial, management, and system provider. Marketing category includes factors pertinent to serving the needs of existing customers and providing exposure to new market segments. Operational factors accounts for the ease of use of the eval uated electronic distribution channel, as well as for the interoperability of the channel with existing technologies, databases, and channels. Technical category represents potential technical difficulties of managing a distribution channel, such as securi ty, transaction speed, update speed, and traffic levels. Financial factors are concerned with evaluating the cost (start up and transaction costs) and profits of a particular channel. Management category considers strategic managerial aspects such as manag ing brand image and customer relationships. System provider factor refers to evaluation of the system provider qualities, including reputation and understanding of the hotel business. This evaluation framework may be used for both adoption and continuation of the distribution channels. The review of relevant literature and available theoretical frameworks provide some understanding of different aspects of the flash sales phenomenon, and may hint on the methods of flash sales evaluation as a hotel inventory distribution channel.
50 However, none of the presented theories may explain the phenomenon fully in all its complexity. Therefore, the review of relevant literature (or the lack of literature) highlights the need for the current study and its significance, a nd also defines the choice of the grounded theory approach as a research method for the study. Methods and Procedures Since hotel flash sales are a relatively new phenomenon that has not been studied well in academic literature, and the purpose of this stu dy focuses on a deeper understanding of this phenomenon, this study undertakes a qualitative based perspective. The qualitative design was adopted due to the exploratory nature of the research that was determined by the lack of research that was able to gu ide the development of the current study. Research Methodology As previously mentioned the purpose of this study is to investigate the benefits, drawbacks, and performance measures of room inventory distribution via flash sales websites from hotel manager constructionist epistemological perspective that posits that meaning making occurs words, this means that an image of hotel inventory distribution via flash sales websites phenomenon. In application to this study, the constructivist paradigm would allow to view the flash sales pheno menon from the perspective of the subjects being studied and their constructed experiences and opinions. This means that the hotel flash sales phenomenon will be described from the perspective of the individuals who have
51 experienced it. For example, the ad vantages and disadvantages of the hotel inventory distribution via flash sales websites will be described and interpreted based on the experiences of hotel managers who have participated or considered flash sales as a distribution channel. Research Method Grounded theory has been selected as a research method for this study. Grounded theory research design is a suitable method for research projects that are lacking theory to explain the process or phenomenon (Charmaz, 2006; Creswell, 2007, 2011; Locke, 200 1; Strauss & Corbin, 1990, 1998). Grounded theory approach allows a researcher to inductively generate a theory that would explain a phenomenon or a process that may not be fully explained with existing theories or models. Grounded theory approach has bee n adopted by different research fields including management. According to Locke (2001), grounded theory is particularly useful to address decision making, socialization, and change topics in management research. Grounded theory approach has been utilized i n the hotel industry research in all three aforementioned managerial areas. For example, the topics that were studied with the grounded theory approach include strategy development (Yang, 2012), service 2010), and career change in the understanding of benefits, drawbacks and performance measures of hotel flash sales is not possible using available theoretical frameworks, and th at the aforementioned factors would contribute to managerial decision making regarding adoption of flash sales as an inventory distribution channel, the authors concluded that grounded theory is a suitable research method to guide this research.
52 Data Colle ction Study participants selected as the data collection method for this study. As suggested by the grounded theory approach, theoretical sampling was used as the sampling m ethod for this study and drawbacks, participants for this study were selected using two approaches. First, a sample was drawn from the hotel managers who have participa ted in hotel flash sales. In order to identify such hotel properties the researcher monitored hotel flash sales websites such as Groupon Getaways, LivingSocial Escapes, and Jetsetter , and recorded the hotels running the promotions in a database. The sco pe of the study is limited to the hotels located in the USA, and, consequently, only domestic hotel properties were recorded in the database. All domestic hotels that appear on the flash sales websites were recorded in the database without regard to their service level, chain affiliation, or geographical location within the country in order to diversify the sample drawn for the study. In order to recruit participants for this study, an email invitation was sent out to general managers of the hotels that ha ve participated in flash sales. In the case where other departmental managers were more involved with the decision making and evaluation of flash sales promotions, the general manager was asked to forward the invitation to the most experienced manager who was coordinating the flash sales promotions for that property. The alternative departments may include revenue management, marketing, and sales.
53 Second, the study also included those managers who have considered participating in flash sales, however, deci ded not to do so. In order to access hotel managers who did not necessarily participate in flash sales, the invitation to participate in the study was sent out to the executive level managers (general managers, CEO, and presidents) of hotels, hotel chains, and hotel management companies who are registered with the American Hotel & Lodging Association (AH&LA). AH&LA is a including individual hotel property members, hotel companies, student and faculty to the membership composition, it represents a credible source of recruiting hotel industry representatives from different segments to partic ipate in the study. The interviews were conducted over several months to allow researchers to start the coding process, and reflect on the data and emerging codes. The first stage of data collection included ten interviews with eight managers from inde pendent properties, one manager from a branded property, and one manager from a hotel chain corporate office. The codes that emerged during the initial data coding steered research towards collecting additional data from hotel management companies and hote l corporate offices. On the second stage of the data collection, additional 36 interviews with hotel managers from independent and branded hotels, management companies, and hotel corporate offices were conducted. All interviews were scheduled via email bas ed on the availability of the participants. The interviews were conducted over the phone, so not to incur any geographical limitations. The interviews were on average 20 minutes long. All
54 participants were explained the purpose of the study and asked for t heir consent to participate in the study and to record the interview. The interviews were audio recorded with the permission of the participants and saved for further analysis. To protect the privacy of the study participants the file names were coded so t hat the name of the interviewee and the property may not be identified. In order to ensure indemnification of the hotel properties and the managers who participated in the study, a key file matching the codes with the participant names and properties is ke pt in a separate folder apart from the audio recordings. Interview guide An interview guide was prepared for the purpose of this study and used to guide the interview process. Grounded theory method suggests that the structure of the interview questions m structured oriented nature of the study and limited time that hotel managers were able to allocate for an interview, the authors selected the se mi structured path for building interview questions. Such decision was made to ensure that participants would have a chance to speak to their flash sales experience as fully as possible. The authors made a conscious effort to phrase the questions in an ope n ended, non judgmental manner. The questions included in the interview guide were developed based on the comprehensive review of academic and professional literature. Due to the lack of academic literature on the topic of hotel flash sales, the researcher surveyed the literature from related disciplines, including advance selling, discounting, and hotel distribution channels. In order to support the development of the questions, professional literature on hotel flash sales was also utilized. As a result of the literature review two
55 versions of the interview guide were developed: for those hotel managers who have participated in flash sales and those who considered this channel but decided not to participate. The researcher followed the following interview g uide to conduct the interviews with hotel managers who have participated in flash sales: 1. Why did you decide to participate in distributing room inventory on flash sales? 2. Could you please also briefly tell us how do flash sales deals actually work? 3. What hap pens when a customer buys a flash sales deal? 4. Based on your experience, what are the benefits of distributing room inventory via flash sales websites? 5. What are the drawbacks of distributing room inventory via flash sales websites? 6. When you were running the promotion what were your expectations and your goals? 7. How did you evaluate performance of the flash sales deal that you offered? 8. Do you think that flash sales websites provide an increase in overall hotel profit via increased revenue streams from non room operating departments? 9. Do you have any other comments on the topic of distributing room inventory on flash sales websites? 10. Could you please provide us with contacts of other hotel managers who have participated in distributing room inventory on flash sale s websites? The following set of questions was used to guide the interviews with those participants who have considered flash sales for their properties, but decided not to participate in this distribution channel. 1. In your opinion, why would hotels partici pate in distributing room inventory on flash sales websites? 2. In your opinion, what are the benefits of distributing room inventory on flash sales websites? 3. What are the drawbacks of distributing hotel inventory via flash sales websites?
56 4. How would you evalu ate the performance of a flash sales deal, or how would you measure the success of the flash sales promotion? 5. Do you think that flash sales websites may provide an increase in overall hotel profit via increased revenue streams from other non room operatin g departments? Data Analysis The data analysis process was carried out simultaneously with ongoing data collection. In the situations when an interview was conducted by the lead researcher alone, a follow up meeting would be scheduled with another member of the research team to discuss the new data added to the data set and potential implications for the emerging themes. This step served as a means to verify the quality of this qualitative inquiry, and will be further discussed in the Rigor of qualitative research section of the study. Additionally, all audio recordings of the interviews were transcribed for the purpose of data analysis. The researchers read the transcripts to become familiar with the collected data. After that, significant phrases descri bing flash sales experiences were identified. The significant phrases were clustered and coded into themes common for all transcripts. The data analysis included initial coding, focus coding, axial coding, and theoretical coding (Charmaz, 2006). The coding process was initiated by two researchers with the initial coding of ten initial interviews. The coding process started with the initial coding using the incident by incident strategy. This strategy was selected over word by word or line by line approaches to initial coding due to the large amount of rich data generated from 46 interviews. At this stage of data coding, in vivo codes were frequently used. The initial codes developed by the two researchers were reviewed for clarity and consistency in order to verify the quality of the research.
57 With the ongoing coding and data collection the data analysis moved to the focused coding stage. Focused coding was utilized in order to organize and identify the most prominent themes. After these inductive stages of coding, the data was put back together by applying axial coding approach to outline the relationships between the developed categories and their attributes. The theoretical coding was later applied to establish the relationships between the main concepts a nd formulate the theory. Rigor of Qualitative Research The difficulty of attaining reliability and validity in qualitative research has been widely documente d (Creswell, 2007; 2011; Morse et al. , 2008). This difficulty is mainly explained by the difference s in quantitative and qualitative paradigms that influence research design, data collection, analysis, and knowledge generation. The methodological rigor of this research was attained through the use of the verification approach. A concept of verification of the qualitative research was developed by Morse et al. (2008) as an alternative to post hoc evaluation of reliability and validity in qualitative research. Morse et al. argued that rigor should be inherited in the research process rather than checked u pon in the completion of the research process when it may be too late to make necessary adjustments. In the context of this study, verification is proposed to include several components such as investigator responsiveness, methodological coherence, sampli ng adequacy, collecting and analyzing data concurrently, saturation, and thinking theoretically. The researcher responsiveness in the current research was ensured by the simultaneous process of data collection and analysis that allowed researchers to stay open and flexible due to the continuous influx of new information. Due to the exploratory
58 nature of the research no theoretical framework was selected, and therefore, could not affect reflexivity of the researchers. Also, the lead researcher frequently rev iewed the observed results with other members of the research team to seek insight and maintain creativity in handling research findings. Methodological coherence was achieved via matching the research questions with the appropriate research method and res earch paradigm. As it was mentioned earlier, the constructivism research paradigm and the grounded theory research method were found suitable to the research questions stated by this qualitative inquiry. Also, the sampling procedure was selected in accorda nce with the formulated research questions (Morse et al., 2008, p. 18). To ensure sampling adequacy, the research included representatives from different segments of t he hotel industry, and also both positive and negative cases of implementation of hotel flash sales. Another technique for ensuring methodological coherence included developing communicable instructions for data coding among the researchers. Communicable i nstructions are important to make sure that the data was approached by the researchers in the same way. Researchers worked independently on first ten interviews, and later compared the results of initial coding. Based on that necessary adjustments were mad e to the data themes and the coding process continued. Also, the data analysis process was enhanced by the good quality audio recording of all interviews that allowed producing word by word transcripts of the interviews. Another aspect of the verification process included collecting and analyzing data concurrently until reaching the point of data saturation. This procedure includes
59 iterations between data collection and analysis, and also relates back to the researcher responsiveness in evaluating data. Th e simultaneous data collection and analysis is an essential element of the grounded theory approach. The saturation of data is ensured by the means of constant comparative approach which involves data collection and analysis until new information does not provide any further insight into the emerged category (Creswell, 2007). The point of data saturation was determined based on the themes consistently repeating in the analysis of the interviews. Even though empirical evidence suggests that the point of da ta saturation in a qualitative research may be reached after collecting 12 interviews (Guest, Bunce, & Johnson, 2006), current research carried on with the data collection until conducting 46 interviews. Such decision was mainly guided by the diversity of the sample and representation of different industry segments, and, consequently, different managerial perspectives in the data. Simultaneous data collection and analysis, as well as establishing the point of data saturation, assisted in ensuring theoretic al thinking. Theoretical thinking as defined by Morse et al. (2008) involves thinking at both macro and micro levels with reconfirming emerging ideas in the new data. This means that in application to this research, data collection and analysis happened ac ross different managerial levels of the industry: that is the property level, hotel management company level, and corporate level. Again, theoretical thinking is supported by the grounded theory method that includes theoretical coding as the last stage of the data analysis process. Taken together, methodological and data collection approach, data analysis that involved two researchers, and theoretical thinking provided foundation for triangulation
60 of the study results. Triangulation involves looking at the same phenomenon or research question from different perspectives, and includes several approaches such as data triangulation, method triangulation, investigator triangulation, and theoretical triangulation (Decrop, 1999). In this particular study, investi gator triangulation and theoretical triangulation were used. Investigator triangulation refers to the analysis of the data by different researchers. Theoretical triangulation implies looking at the same phenomenon and research question through the lens of different theories. In this study, different theories were used to explain and validate the relationships proposed by the developed grounded theory. Findings A total of 46 interviews were conducted for the purpose of this study. Thirty one hotel managers indicated that they have had an experience of running flash sales promotions, while fifteen of the study participants reported that they prefer not to use such a distribution channel for hotel inventory distribution. The data collection covered different s egments of the lodging industry and included industry professionals from different types of organizations, including independent properties, branded properties, hotel management companies, corporate offices of hotel companies, and one hotel e marketing com pany. The breakdown of the number of interviews by the industry segment is presented in the Table 2 1. Company profile below. brands that operate internationally, however, as it was mentioned earlier, the scope of the study included only domestic hotels. Therefore, only managers of the corporate offices located in the United States were invited to participate in the study.
61 The interviewed i ndividuals hold different top management positions including general managers, directors of marketing, sales, e commerce and revenue management. A respondent profile summary is presented in the Table 2 2 Respondent profile summary below. A detailed profile for every respondent, including job position, industry segment, and participation in flash sales, is presented in the Table A 1 Industry respondent profile . To Groupon, or not to Groupon? All study participants who ran flash sales promotions agreed that f lash sales different managers. Some of them mentioned that the hotel was hurting, or sales were low, rough weather conditions affected the demand, a hotel experienced an unexpe cted by different factors in internal or external environment, however, a common ground behind all these instances is that a hotel faced a need to fill the rooms that wou ld otherwise stay empty. The findings suggest that hotel managers who adopted flash sales as an inventory distribution channel were aiming to address the perishability problem at their properties. The results seem to be in agreement with the theory of pea k load pricing (Crew et al , 1995). The theory of peak load pricing addresses capacity utilization and pricing for products and services with variable demand. The theory suggests that if the prices are not adjusted to the variable levels of demand, a supply excess will be created in the low season, and supply shortage will be observed during the high seasons. Therefore, demand seems to be in line with the theory of peak load pricin g. This means that hotel
62 managers implemented flash sales in order to adjust pricing to the fluctuations of variable demand: they provided a deep discount on the flash sales websites when the hotel went into the off The study results indicated that hotel managers from independent hotel properties were happier with the flash sales experience than their colleagues from the branded properties. The reasons for less favorable evaluations of flash sales by the hotel managers from the branded properties depend on several factors, such as recommendations given by the brand corporate offices, violations of price parity, and other distribution channels provided to the branded properties by their brands. These factors will be discussed in more det ail in the following sections on the benefits and drawbacks of flash sales websites as a hotel inventory distribution channel. Benefits of Distributing Hotel I nv entory via Flash Sales W ebsites The interviews revealed several categories of prospective bene fits that flash sales may bring to the hotel industry. After interviewing, transcribing, and theming the 46 interviews it seems that the major benefits of flash sales fall into the categories of inventory management, revenue management, brand marketing, an d customer relationships. The diagram with categories and associated subcategories of research findings for the flash sales benefits is presented in Figure 2 1 entitled Categorization of hotel flash sales benefits. With regard to inventory management, all interview participants found flash sales to be a useful channel to increase occupancy for particular need periods and sell inventory that otherwise would remain unsold. A general manager of an independent
63 A revenue manager from a hotel corporate office supported this opinion by saying that Together with inventory management also come revenue management benefits. Study participants mentioned that hotel flash sales may assist hotels in generating tour ism was very low here. We have done 6 offerings to date through LivingSocial and non room operating departments. For example, a front office manager of a full service indepen The interviewees across different industry segments mentioned that flash sales bring a promotional benefit to the hotels via increasing expo sure and advertisement of the property. When an email blast with flash sales promotions goes out, hotel information reaches the inboxes of thousands of flash sales subscribers delivering anded upscale property mentioned: It is been a proven message to gain exposure to a large market...So, in p eople that may know of Sarasota, but did not know what hotels were in Sarasota, our name is there on it. So, for future use they might look at it. trials and customer acquisi tion. Flash sales companies own large customer databases that they use to distribute promotions. It allows flash sales companies to deliver promotions to a wide audience and attract new customers to the advertised hotels.
64 Moreover, the promotions are usual ly time limited, which puts pressure on the customer to make their decision as soon as possible before the promotion goes away. A combination of special discounted prices, value adding services and a limited timeframe may stimulate customers to try hotels that are promoted on flash sales websites. And, of course, hotels are striving to provide an excellent service to bring first trial customers back for a repeat visit. D rawbacks of Distributing Hotel Inventory via Flash Sales W ebsites The study participants have also mentioned several disadvantages of distributing hotel inventory via flash sales websites. Such disadvantages have implications for brand marketing, customer relationships, revenue management, and hotel operations. Some of the emerged categories, such as brand marketing, customer relationships, and revenue management display dual effect on hotel operations and have implications for both advantages and disadvantages of hotel flash sales. The operational challenges category is the new category that emerged during the analysis of the flash sales drawbacks. The diagram with categories and subcategories of research findings for the flash sales drawbacks is presented in Figure 2 2 entitled Categorization of hotel flash sales drawbacks. Some hotel manage rs were concerned about the impact of using flash sales on Such a concern may be j ustified by an expectation that a potential negative impact of Another important aspect from the brand management perspective was more relevant to branded properties and was concerne d with hotel brand corporate offices not
65 supporting flash sales as an inventory distribution channel. Brands took different positions with regard to their opinion on flash sales websites. Some of the hotel brands gave freedom to their properties to select distribution channels, including an option to use flash sales websites, however, some brands were not supportive of using flash sales websites as an inventory distribution channel. The main reasons for that is the violation of price parity and best rate gu arantee (please see the explanation below in the Revenue management issues). Also, managers of luxury hotels were worried that flash sales may not reach the target demographic of the luxury segment, and therefore attract customers who would stay in a hotel with a discount, but would not come back at a regular price. Also, a vice president of marketing of an e marketing company elaborated on some of the flash So, the big challenge is certainly the more flash sales they run, and if they are running them back to back, they are training the consumers to look for them on sales always. So, it has a tendency to devalue the price of the hotel room. From the revenue management perspective, the majority of interview participants mentioned high distribution costs as one of the challenges of running flash sales promotions. Participation in flash sales websites requires not only a high discount to qualify for the promotions, but also a commissio n to the flash sales website. One of the disadvantage is that you're paying a commission to somebody, which is a fairly high commission. You're paying 40% to 50% commission whi One of the major concerns for the branded properties became the violation of rate parity and best rate guarantee. When hotels are striving to provide the best rates
66 on their own websites, a flash sales promotion on a third party website v iolates this best rate initiative. However, some hotels found a solution for this challenge by creating special booking codes for the flash sales deal and making them available to their loyal Also, the majority of the hote l managers who participated in the interviews were concerned about negative impacts on revenue. This problem is in line with the concerns of high distribution cost and deep discounts. A general manager of an independent rawback is that the actual amount ultimately consequence, lower rates offered on flash sales websites may lead to lower average daily rate (ADR) and lower position on the STAR reports (Smith Travel Accommodation Report). And last, but not least, another theme that emerged during the interviews was operational challenges. First, hotel managers mentioned the time investment that was necessary to prepare the flash sales deal, deve lop property management system (PMS) codes to enter the reservations, and train hotel employees to process the reservations that come from flash sales. Another aspect, that hotel managers found challenging, was call center management. This challenge is cl osely related to the advantage of increased sales provided by the flash sales promotions. Once the flash sales deal was advertised, a lot of customers placed phone calls to find out more details about the promotion, inquire about the hotel, available date s and other details of making the reservation. As an outcome, some hotels were not prepared to receive such a high
67 volume of phone calls. However, it was an important lesson to learn in order to be better prepared for the future promotions. Similarly to t he call center issue, some hotel managers mentioned that they were not prepared to serve that many guests during the breakfast that was included in the price of the hotel room. Overall, the operational challenges theme may be summarized in the following wo make as much money that you get from twenty reservations [sold with flash sales]. So, very hard, l Performance M easures of Hotel Flash S ales The results of the interviews revealed several different methods of hotel managers measuring the performance of hotel flash sales. Hotel managers identified different measures that they implemented during the promotional and post promotional phases of the flash sales experience. Performance measures of the promotional phase may also be grouped under the performance measures of flash sales effectiveness during the pre purchase and purchase stages. One of the components of hotel flash sales evaluation at the pre purchase stage of the promotional phase was the web traffic generated after the launch of the promotion. Study participants have mentioned that releasing a deal on flash sales websites would rea ch a lot of consumers, who would want to find out more details and A general manager of an independent all suite hotel mentioned: It is essentially a cost free way for us to drive traffic to our own website, because when these e blasts go out, the consumers will migrate to our
68 website for more information; and, Google Analytics substantiates that benefit. So [flash sales generate] market exposure at almost no cost... From this perspective, generating interest in the hotel and potential sales was one of the factors to evaluate the success of the promotions. It would assist managers to evaluate the interest that was generated by using flash sales as a distribution channel. Moving to the purch ase stage of the promotional phase, another aspect of hotel flash sales performance would be the number of room nights booked. This would be the actual sales impact. A general manager of an upscale branded property supported this ong as it is priced right, you can certainly sell loaded inventory. That was not an issue at all. We sold all the rooms we dedicated to Groupon long sales that says th at this channel helps to sell distressed inventory, or sell rooms during the need period. Therefore, it became important for hotel managers to evaluate how many rooms they would be able to sell through such channel. At the post promotional phase, the impor tant element of hotel flash sales performance would be the voucher redemption rate. Hotel managers have mentioned that not all of the purchased promotions will be redeemed, and, therefore, not all sales will result in bringing an incremental guest to the h otel property. Even though hotels still would receive revenue from a flash sales website for purchased, but not redeemed deals, the ultimate goal of the promotion is to bring the customer in house. Upon the perience other benefits, such as upselling, stimulating auxiliary revenues, and generating repeat business. Following the sales and redemption rates, hotel managers would also want to evaluate the impact of inventory distribution via flash sales websites o n the key hotel
69 metrics, such as occupancy, ADR, revenue per available room (RevPAR), and profit per occupied room (PPOR). The advantage of filling immediate occupancy needs that were discussed earlier usually results in a higher occupancy percentage for h otels participating in flash sales. However, deep discounts offered on flash sales websites may bring the ADR and consequently, RevPAR and PPOR indicators down for hotels. As an outcome, such decrease may influence forecasting for future periods, and also may alter the position of a hotel on the STAR reports. A vice president of marketing of an e marketing company summarized their experience with running flash sales for different hotels: We have some clients who became addicted to these [flash] sales, becau se they do tend to produce an enormous volume in a short period of time to fill a lot of need period. The bad part of it is that it tends to cause a drag on ADR. Another way to measure hotel flash sales performance that was mentioned during the interviews was by calculating return on investment (ROI). The ROI may be calculated based on the revenues received from the flash sales promotion and costs of running the promotions. The costs associated with flash sales promotions may include commission paid to the website and variable cost of selling a room and serving breakfast (if included with the hotel stay). Some managers included other additional factors in the ROI calculation, such as the cost of time to train hotel employees to work with flash sales. And fin ally, those hotel managers who have run flash sales promotions several times may want to compare the results of a current promotion to the previous ones. Usually the expectation would be to exceed the results of the past promotions with regard to web traff ic increase, guest acquisition, number of room nights sold, and
70 revenue generation. Alternatively, those hotels that run promotions for the first time may want to compare the results of the promotions to the metrics of a non promotional period which is sim ilar to the one when the promotion was run. Such comparison may offer a meaningful benchmark for a hotel to evaluate the results of the flash sales. A summary of different types of flash sales performance measures is presented is presented in the F igure 2 3 below. Generating R evenue from Non Room Operating D epartments generate additional revenue streams from non room operating departments. Some hotel managers supported the ass umption that bringing more guests in house will also room operating departments. For example, a front office manager of an independent full service hotel and we expect guests to patronize the restaurant However, on the other hand some hotel managers were disappointed with flash were anticipating this [to generate extra revenues from non room operating departments]. However, the guests did not spend much at the pool, particular finding may be as sociated with another disadvantage of flash sales and e sales is not your customer, they are looking for a greater discount. They will stay at the
71 Overall, the entire flash sales experience with regard to its benefits, drawbacks, as well as performance measures is presented in the Figure 2 4 below that is entitled Flash sales evaluation framework. The developed grounded theory suggests that the process begins from the need based determination of applicability of flash sales to a selection and deal development in the case when flash sales are found to be a suitable distribution channel. Next, the flash sales experience is described with regards to benefits in drawbacks in several prominent categories that emerged during the data analysis: inventory management, revenue management, brand management, customer relationships, and operational challenges. The process concludes with the application of performance measures that assist managers in evaluating flash s ales effectiveness at the promotional and post promotional phases. The results of such evaluation will later distressed perishable inventory next time. Closing Remarks The curr ent study investigated hotel flash sales as a new room inventory distribution channel from the perspectives of advantages, disadvantages, performance measures, and revenue generating abilities. The description of the flash sales phenomenon was constructed through the lenses of hotel managers who either have participated in such a distribution channel for room inventory distribution, or evaluated the channel, but decided not to adopt it. The findings of the study revealed that flash sales websites present a unique electronic distribution channel that may contribute to effective revenue management practice while simultaneously enhancing marketing capabilities.
72 The use of flash sales as a distribution channel seems to present a hybrid mechanism serving as both a distribution channel and a marketing tool to increase market exposure. The use of flash sales promotions seems to assist hotel managers in moving hotel room inventory that might otherwise perish while increasing occupancy This increase in occupancy does come at a significant commission cost associated with the sale of the rooms that is stimulated via promoting deeply discounted deals to an extensive database of flash sales subscribers. Traditional revenue management litera ture generally refers to revenue revenue management concern in the use of flash sales as an inventory distribution channel lays in the price component of revenue management where a deep discount is used to reach customers (not necessarily the right customers) at the right time (i.e. managers in the study indicated that while the use of flash sales did boost occupancy levels during low demand periods the type of customer that was attracted to the deep discounted rates may not be the right customer for the hotel to acquire. It has bee n established during the course of this research that advantages and disadvantages attributed to flash sales have implications for different areas of hotel management, including inventory management, brand marketing, customer relationships, revenue managem ent, and hotel operations. These results are in line with of flash sales websites for hotel room inventory distribution. One of the key issues
73 identified in literature competitive pricing that creates value for the customer and cost reductions to sustain challenge has also found support in the application of hotel flash sales to distribute rooms. Theoretical Contribution As an outcome, the current study presents a theoretical framework for flash sales evaluation as an inventory distribution channel. This framework is grounded in the data received from 46 interviews with hotel managers from different segments of the hotel industry. Theoretical sampling that was adopted within the grounded theory approach assisted researchers to fully explore the developed categories and their rela tionships. The proposed framework closes the gap in the literature by providing the first theoretical framework for evaluation of flash sales as an inventory distribution channel. The flash sales evaluation framework may be validated through the triangulat ion by different theories that support particular sections of the proposed framework. The findings of the current study suggest that hotel managers may use hotel flash sales for product by means of providing deep discounts to potential customers. Such managerial reasoning may be supported by the application of the theory of peak load pric ing. The theory of peak demand season. Hotel managers are aware that the core product which is the major reven ue producer for a hotel is characterized as a perishable product that is influenced
74 managers are challenged with pricing a perishable product that will sell under variable de mand conditions. Therefore, the theory of peak load pricing seems to provide insight discounting pricing strategy. This suggestion has been confirmed by several intervi ews with hotel managers from different industry segments who have stated that they have used flash sales in the definitely think it [flash sales] is not something you would wa nt to do if you are in a independent hotel and water resort explained: I think hotels would be hurting themselves if they did do a Groupon like this during the season, beca think it's designed for hotels, is to use it during your need This finding was also reflected in a quote by a marketing coordinator of a group of participated to boost our occupancy during the low season months of Decembe r through mid April. We have used Jetsetter, Groupon, Bloomspot, Vacationist, among Therefore, the theory of peak load pricing may suggest that setting discounted prices on flash sales websites may assist hotel managers in adapting prices to the
75 established demand level, and, consequently, in maximizing profits and optimizing proposition of enhancing capacity utilization of the hotel. In other words, flash sale s may occupancy. Filling the occupancy needs was named as one of the benefits of using flash sales as a hotel inventory distribution channel. This means that in accordance with the theory of peak load pricing deep discounts that are offered on flash sales websites allow hotels to mitigate the problem of variable demand and optimize capacity utilization. However, the study participants were split in their opinion regarding p rofit maximization. Hotel managers expressed concerns regarding the profitability of hotel flash sales due to deep discounts, high commission costs for the room sales, and uncertain revenue generating ability from non room operating departments. The theor y of peak load pricing suggests that adjusting prices to the peak and off peak season may assist in maximization of profits and capacity utilization. Given the split opinion from hotel managers with regard to the profitability of hotel flash sales, it may be suggested that further research is needed to identify the impact of hotel flash sales on the revenue generation during the off peak periods. In addition to the theory of peak load pricing, Shugan and Xie (2005) suggest that advance selling of services m ay assist managers in generating profit improvements by stimulating greater market participation. This suggestion seems to be in line with the advantages of customer acquisition, inducing trial and generating repeat business expressed by the participants o f this study. For example, a general manager of one of
76 the independent hotels illustrates the advantage of greater market participation for their property: Well, mostly the benefit is to bringing clients into the property that you would've never picked up on before. Like I said, I was able to pull from Texas and Seattle and places that people wouldn't normally be coming through our hotel. There are particular market conditions that will make advance selling effective (Xie & Shugan, 2001). The first one is valuations. Uncertainty is identified by the circumstances that will surround a customer in the future. Such circumstances may include time, budget, health, ability to travel, and other factors that may affe with the high uncertainty levels, customers engage in advance selling. Such behavior may be explained by different reasons. One reason would be buying in advance for a lower price, so that the di scount may compensate for the uncertainty level. For example, purchasing a voucher for a hotel stay on flash sales website does not require a customer to commit to a particular date in the future, however, the purchase does require the customer to commit t o a hotel stay during the promotional time specified in the voucher. Another reason for advance purchasing under uncertainty may be associated with the constrained capacity, then, advance selling will guarantee a spot to the customer. The mechanism of ad vance selling increases sales by capturing those customers who would be in an unfavorable situation in the spot period and would not patronize the business (Shugan & Xie, 2004). This mechanism also appears to be relevant to the hotel flash sales. Flash sal es customers are likely to be in unfavorable position during
77 the spot period due to the unavailability of the special promotional price that is offered only during the advance sale. Another market condition required for effective advance selling would b e low marginal costs so that it is still profitable to sell at lower advance prices. The hotel industry is one of those with low variable costs (Kimes, 1989). For example, according to Kalnins (2006), the marginal cost of a hotel room at a full service ups cale chain ranges between $15 to $35 depending on the location of the property. Once the hotel, staff members and all necessary systems are established, it does not take that much to sell an additional room. It is also important to account for the seller and price credibility, so that customers are confident in the credibility of the referenced spot price. The growth and penetration of electronic distribution channels in the hotel industry made it easy for the customers to check the trustworthiness of the face value of the promotions advertised on flash sales, and compare the initial price of the hotel room to the one offered on the direct website. All of these conditions are attributes of the lodging industry. And, hotel inventory distribution via flash s ales websites requires an advance purchase. Therefore, flash sales may be considered as a distribution channel for advance selling with price discrimination. Consequently, flash sales may bring an advantage of profit improvements by stimulating greater mar ket participation. Since the participants of the study were split in their opinion about the profitability of flash sales, future research may be needed to further explore this subject with regard to appropriate pricing, timing and volume of flash sales.
78 T he theory of peak load pricing and advance selling approach appear to be from the perspectives of optimizing variable demand, setting prices and controlling perishable inv entory. However, as it was mentioned earlier, hotel flash sales bring implications not only to the revenue and inventory management, but also to other areas of hotel management, including brand marketing, customer relationships, and operations. Therefore, understanding the study results and managerial decision making with regard to hotel flash sales adoption may require consideration of a broader range of theories. Based on the results of the study, it appears that cost/benefit analysis theory supports mana the cost/benefit analysis, hotel managers would evaluate an investment that is necessary to engage in the flash sales distribution, e.g. cost of distribution, employee training, conve nience, time commitment and other relevant factors. The identified costs should be compared with expected benefits, such as room nights sold, contribution to the revenue, exposure to a new market, etc. Current research has helped to identify the key benefi ts and drawbacks (costs) that may be used for the assessment of the hotel flash sales from the cost/benefit approach. In addition, ROI calculation as one of the performance measures of the effectiveness of hotel inventory distribution via flash sales websi tes appears to go in line with the cost/benefit approach. In calculation of ROI, return may be considered as benefits that a hotel receives from using flash sales as a distribution channel, and
79 investment may be considered as a cost of participating in suc h a distribution channel. Therefore, ROI may be considered as a cost/benefit ratio expressed in monetary terms. In evaluation of the flash sales experience the study participants mentioned several key categories, including inventory management, revenue man agement, brand marketing, customer relationships, and operational challenges. These categories may be compared to the factors in the electronic distribution channel evaluation framework rs to be important for electronic distribution channel evaluation. Those factors include marketing, operational, technical, financial, management, and system provider categories. Current research, suggested categories that are more parsimonious to the dist ribution channel evaluation, and more reflective of the nature of the flash sales websites. First, the framework proposed in this study suggests an inclusion of inventory management aspect in the evaluation framework to understand how well the channel hel factors were further specified to become brand management and customer relationships categories in the flash sales evaluation framework. This modification reflects the nature of flash sales as not only distribution, but also as a marketing channel. Then, financial category may be compared to a newly emerged revenue management category, that reflects not only revenue generation, but also yield and distribution costs. Operationa l and technical aspects merged into one category named Operational challenges that reflected not only technical and systems aspects of distribution channel performance, but a broader range of impacts that a distribution channel may have on hotel operations . For example, an operational challenges category in the flash sales evaluation
80 framework included the challenges of setting up the system, but also the elements of human interaction with the system, e.g. required training, as well as challenges in the hot el that were created due to high sales volume. The proposed flash sales evaluation framework displays elements of both CBA approach and electronic distribution channel evaluation framework by presenting costs and benefits (drawbacks and benefits) for diffe rent categories of flash sales experience (inventory management, revenue management, brand marketing, customer relationships, and operational challenges). By combining these two approaches, the flash sales evaluation framework aims to provide a more compre hensive evaluation of flash sales as an inventory distribution channels. This framework may be further validated in application to flash sales in other industry segments. Also, the proposed framework creates grounds for developing future empirical studies and providing implications for the hotel managers. Managerial Implications In summary, hotel flash sales represent an electronic inventory distribution channel that may assist hotel managers in addressing demand fluctuations and solving perishability probl peak periods. As a deeply discounted distribution channel, flash sales need to be carefully evaluated by hotel managers in order to arrive at the decision whether such a channel is beneficial or detrimental for a specific proper ty. The results of the current study have provided insight with regards to hotel flash sales advantages, disadvantages, performance measures, and revenue generating abilities. The proposed theoretical framework suggests to begin with a careful identificat
81 the flash sales promotion. Next, the framework presents a full range of factors to consider when making a decision about flash sales adoption as an inventory distribution channel. The proposed framework suggests that hotel managers should be aware of potential impacts of flash sales on different aspects of hotel operations, including inventory m anagement, revenue management, brand marketing, customer relationships, and operational challenges. The framework provides more detailed subcategories for every identified category, therefore, presenting the first and comprehensive mechanism for hotel mana gers to evaluate their flash sales experience and make a decision about its inclusion in the distribution channel mix. The results of the study also reflected view on flash sales of managers from different segments and levels of the hotel industry. The res ults indicate that hotel managers should be aware of the differences for independent and branded properties in implementation of flash sales. Branded hotels should make sure to consult their corporate office regarding the adoption of flash sales as an inve ntory distribution channel. Next, if the adoption of the channel is approved by the brand, it is also important to obey other brand related policies, such as rate parity. Overall, the results of the study present the first flash sales evaluation framework that is grounded in the interviews with the hotel managers and validated through the use of the theory of peak load pricing, advance selling, cost/benefit analysis and the electronic distribution channel evaluation framework. The results of the study aim t o provide assistance to hotel managers in their decision making process regarding the adoption and continuous use of flash sales as an inventory distribution channel. At the
82 same time, the findings of the study provide foundation for researchers who may be interested in studying the impact of flash sales on the lodging industry.
83 Table 2 1. Company profile Industry segment # of interviews Description # of interviews Hotel properties 16 Independent 13 Branded 3 Hotel management companies 16 Branded h otels 8 Independent + branded hotels 8 Corporate office 13 Domestic 2 International 11 Hotel e marketing company 1 1 Total: 46 46 Table 2 2 . Respondent profile summary Position Number of Participants General Manager 8 Marketing 12 Sales /e Commerce 14 Revenue Management 9 Other 3 Total 46
84 Hotel Flash Sales Drawbacks Revenue Management Brand Marketing Customer Relationships Steep discount High distribution costs Lower yield Rate parity issues Cheapening property Damaging brand Lacki ng support from brands Destroying value perception Training customers to look for deals Creating price unfairness Operational Challenges Preparation Training Managing volume Abusing the system Figure 2 1. Categorization of hotel flash sales benefits Figure 2 2. Categorization of hotel flash sales drawbacks Hotel Flash Sales Be nefits Inventory Management Revenue Management Brand Marketing Fast sales Selling distressed inventory Increasing occupancy Exposure Awareness Advertising Promotion Customer acquisition Inducing trial Generating repeat business Customer Relationshi ps Capital generation Expenditures beyond face value
85 Figure 2 3. Performance measures of hotel f lash sales promotions Promotional phase Pre purchase stage: Website traffic Interest generation Purchase stage: Number of units sold Total revenue Post promotional phase Number of vouchers redeemed Purchases beyond face value (upselling, auxiliary revenues) Key performance metrics (e.g. ADR, Occ, RevPAR, P POR) Return on Investment (ROI) Results of previous promotions
86 Figure 2 4. Flash sales evaluation framework
87 CHAPTER 3 FLASH SALES CUSTOMER FOR HOTELS? Background of the Study Hotel flash sales websites, such as Groupon Getaways, LivingSocial Escapes, and Jetsetter , prov ide time limited deep discounts for the advanced purchase of travel and hotel related products (e.g. hotel rooms, cruises, tourism attractions, meals, etc.) (Picolli & Dev, 2012; Sigala, 2013). This innovative method of distributing travel related products through flash sales channels allows companies to devices. Likewise, consumers must act quickly on the sales deals they receive before they lose the opportunity to purchase t ravel related products at deep discounts (about 50% off regular price). This relay of sales information often requires customers to make fast and spontaneous decisions about the flash sales purchase if they are to take advantage of the promotional value. The use of flash sales websites has been rapidly adopted by the hotel industry for room inventory distribution. For example, one million hotel rooms were sold via sale s seem to address the problem of selling the perishable core product (i.e. hotel rooms) during a need period (i.e. under the conditions of low demand). The first study of this dissertation demonstrated that flash sales as a hotel inventory distribution cha nnel comes with particular benefits and drawbacks. Hotel managers suggested that the use of flash sales websites may have implications for hotel brand marketing, customer relationships, inventory management, hotel operations, and especially revenue manage ment.
88 Revenue management has been traditionally described in literature as a practice customer, at the right time and at the right price (Kimes, 1989; Kimes & Wirtz, 2003; Sm ith, Leimkuhler, & Darrow, 1992). Much study has been devoted to the development of understanding in different product and industry contexts as to what is the right product, who is the right customer, what is the right time, and what is the right price? Mainstream lodging literature has answered these important questions by (Kimes & Wirtz, 2003; Wirtz, Kimes, Theng, & Patterson, 2003; Weatherford & Bodily, 1992). Therefor e, the right product refers to a smart allocation of a product unit or bundle at a specific point in time and price level. The right price comes down to willingness to pay w ith the available product. When it comes to the consumer, a conflict may arise from the trade off of revenue management and customer centric orientation. For example, such conflict may present itself in restricting lower paying customers in their access to capacity, and preferring higher paying custome rs over loyal customers (Wirtz et al. , 2003). However, customer value should be considered not only from the perspective of the ability to pay, but also from the perspective of potential long lasting relation ships and loyalty, leading us to the concept of customer lifetime value. Research on customer lifetime value and loyalty has advised managers to focus on both the short term and long term customer value as opposed to only a single transaction between a com pany and a customer (Jain & Singh, 2002; Yang & Peterson, 2004). In addition, marketing
89 research has established that it is more expensive to attract a new customer, than to retain an existing one (Tyrrell & Wood, 2005). Therefore, from the long term persp ective, the concern of using hotel flash sales is that customers attracted via flash sales websites would be just one time guests and would not come back to the property unless they are offered another promotion. From the revenue management perspective, wh en adopting flash sales as an inventory distribution channel, hotel managers have a choice to supply the right product (develop a deal for a hotel room alone, or prepare a package) at the right time (the an attractive discount to stimulate customers). However, a looming question may remain for those managers who opt to use flash sales websites for hotel room inventory: that is, whether this right product with the right room price and the right time will be able to reach the right customer for the hotel. Managers from the first study expressed a concern regarding the lifetime value of flash sales customers. In other words, after the flash sales customer rendered the service they bought through a flash sal e would the customer return? Or was this type of customer only a short term economic value during a need period? The results from the first study revealed that hotel managers are concerned that hotel inventory distribution via flash sales websites may attr Managers suggested that flash sales customers generate less auxiliary revenue than
90 website or travel agencies. Further, hotel managers, who partic ipated in study 1, also expressed that they were concerned that flash sales customers may be trained to look for room deals on flash sales websites, and, once this behavior was learned, may not return to a property unless offered another deep discount. Thi s type of behavior would be fine if the behavior was contained to customers who only used the hotel during need periods and was not a contaminating behavior that could spread to other customers. However, if this learned behavior became the norm and spread to other market segments it would frequent use of a variable pricing schedule that rises and falls in accordance with demand schedules (Croes & Semrad, 2012 a ). Therefor e, using flash sales as a hotel inventory distribution channel may raise concerns of generating repeat business and acquiring loyal customers who are willing to pay the rack rate for a room night stay. Previous research on deal prone consumers has identifi ed several psychographic traits associated with response to price promotions (e.g., coupons). These personality traits include, but are not limited to: price consciousness, value consciousness, variety seeking, market mavenism, and financial constraints (A ilawadi , Neslin, & Gedenk ., 2001; Chandon , Wansink, & Laurent , 2000; Martinez & Montaner, 2006; Sigala, 2013; Wakefield & Barnes, 199 7 ). Some of the aforementioned personality traits (such as price consciousness and financial constraints) seem to support m customers from the perspective of their ability to pay for extra services.
91 It is important to note that while the rooms department is the most significant revenue generator for a hotel, i t is not the only revenue generator. During low demand periods, when room rates are decreased to only a few dollars above the marginal cost to service the room, hotel managers would like to have guests spend money in additional revenue operating departmen ts (e.g. restaurants, bars, golf courses, recreational facilities, etc.). However, current literature does not provide an indication of hotel flash sales customers behavior. Therefore, there is a need to determine the profile of consumers responding to onl ine promotions (Sigala, 2013). This means that some of the psychographic traits that may define flash sales consumer purchasing and post performance with regard to their short and long term profitability. A ccording to the results of the first study of this dissertation, hotel managers often expect flash sales customers to purchase auxiliary services at the property in order to recover some of the revenue that was lost in the deep discount of the hotel room. This finding from the first study is also supported by the lodging industry trend of moving towards total revenue wide profits as opposed to only that of room revenues (Anderson & Xie, 2010). Therefore, the short gness and ability to pay, and, therefore, limits their incidental expenditures at the hotel. For example, posting a 50% discounted rate on a flash sales website may attract customers with a lower reference price and lower discretionary income who could no t afford to book that
92 hotel at a regular rate. At the same time, when using flash sales, hotel managers not only expect to cover an immediate need period, but also to extend the customer base, and generate repeat business. Thus, managers raise a concern wh ether or not the new customers generated by flash sales promotions would be able and willing to revisit the hotel in the future. Purpose of the Study Given the concerns outlined above, the purpose of this study is two fold: to determine the profile of the customers who purchase hotel flash sales deals, and to seeking customer, who is attracted by a price discount, does n ot consume extra services during the hotel stay, and does not convert into a loyal customer. Therefore, for the purpose of this study, the profile of room) revenues for the hotel, and generating repeat business (with regard to revisiting the hotel, and recommending the hotel to others). First, the study attempts to determine the differences between the profile of flash bution channels other than flash sales. The study begin s with building a consumer socio demographic profile in order to develop an understanding of socio demographic characteristics and psychographic traits that are common for hotel flash sales customers, and compare this profile to the characteristics of other customers. The traditional socio demographic characteristics will be included in the analysis: age, gender, education, children, income and place of residency. These socio demographic profile traits will provide hotel managers with an
93 about whether or not this type of customer is desired for their properties, and, consequently, whether flash sales websites should be a ccepted as a room inventory distribution channel. Also, the hotel flash sales customers and other customers will be compared based on psychographic characteristics that are traditionally associated with deal proneness: variety seeking, innovativeness, imp ulsiveness, market mavenism, brand loyalty, price consciousness, value consciousness, financial constraints, and spending self control (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006; Wakefield & Barnes, 199 7 ). These psychographic t raits have been identified in existing literature as traits associated with consumer response to sales promotions. At the same earlier. For example, high price consciousn ess, and pressing financial constraints may provide a strong reasons for consumers to seek deals. Similarly, quality consciousness would drive a consumer to seek higher quality products. In this case, flash sales would enable a consumer to select a higher quality product at the price that they usually pay. On the other hand, managers were concerned that flash sales customers do not become repeat guests. If this assumption holds true, flash sales customers should measure low on brand loyalty, and, possibly, high on variety seeking. Building customer profiles is one of the fundamental steps in marketing that influences product development, promotion, and distribution. Understanding the key characteristics of the hotel flash sales customers may allow hotels to make inferences about the preferences and behavior of this customer segment. This knowledge may allow hotel managers to structure the promotions in a way that they will be appealing to
94 that particular segment of the customers who is likely to purchase hot el flash sales knowing key characteristics of the flash sales customers may assist hotel managers in developing tailored promotions in the future to encourage repeat business from flash sales guests. Second, the study investigates the difference between the economic values of the guests who come to the hotel with a flash sales promotion, and those guests who come via other distribution channels. Customer value may b e considered from a short term (current purchasing behavior) and long term (future purchasing behavior) perspectives. This study attempts to consider both of these perspectives. t erm profitability through additional expenditures in non room operating departments during the hotel stay. All participants of the study will be asked to report their behavior in terms of patronizing different operating departments in the hotel and the ass ociated expenditures. The numbers reported by the flash sales customers will be compared to the ones reported by hotel customers who made a reservation through channels other than flash sales websites. Such analysis may allow hotel managers to measure an a term profitability. The long term perspective will be presented through the consideration of customer intentions to revisit the hotel and/or recommend that property to others. Similarly to the short term contribution analysis, two types of customers (flash sales customers and other customers) will be compared with regard to their likelihood of generating repeat business and recommending a hotel to others. One of the hotel
95 s is to estimate a total value of a hotel guest. Without such estimate it may be difficult for hotel managers to evaluate the profitability of distribution via flash sales websites and understand whether they are making the money that they anticipate from each guest. Evaluation of the economic value produced by the hotel flash sales customers may further enhance the understanding of the flash sales customer segment, its Such understanding may assist hotel managers in evaluating flash sales websites as an inventory distribution channel. As previously mentioned, one of the trends in hotel revenue management is the evaluation of total profits of the hotel. Hotel managers may approximate room revenues that may be received from flash sales based on the price established for the promotion and expected sales volume. However, forecasting revenues from auxiliary services may pose a challenge for hotel managers. The current study ai ms to address this challenge by providing an estimate that hotel managers may expect from an average flash sales customer. Statement of the Problem The lack of understanding of the new customer segment that flash sales bring to the hotel properties constit utes the basis of the research problem of this article. As previously mentioned, building customer profiles is one of the fundamental steps in marketing that is necessary for product development and positioning, as well as for distribution strategy. However, current academic literature is devoid of empirical studies that could distinguish the difference between demographic profiles and economic lifetime value.
96 Professional literature sugg ests that the benefits of using flash sales in the hotel industry may include reaching out to a new customer segment, adding exposure, and promotion of the property (Schaal, 2011b, Schaal, 2012c, Gupta, 2012). These suggestions were substantiated in the fi rst study of this dissertation through a qualitative inquiry that involved interviewing hotel managers. Participating hotel managers indicated that exposure, advertising, and promotion via flash sales websites (brand marketing benefits) may lead to inducin g trial, acquiring new customers, and generating repeat business (customer relationships benefits). Often hotels are expecting to introduce their product to consumers via flash sales and later use direct marketing channels to target these customers and br ing them back. However, current academic literature is lacking empirical support for any of these assumptions. Therefore, it becomes difficult for hotel managers to know the characteristics of a new market segment that is coming in to their properties due to the flash sales promotions. Consequently, hotel managers may not be able to understand the customer needs, types of the amenities and services that these customers require, and what the best way (if any) to build financially beneficial relationship wit h them. Such distribution channel, to understand the new customer segment, its behavior, and customer value. As the results of study 1 indicated, hotel managers expr essed concern that they may face some negative consequences from using flash sales such as: financial losses, brand erosion, and attracting customers that may not have a lifetime economic contribution to the property.
97 Research Questions In order to addre ss the purposes and the research problem of the current study, the following research questions were developed: RQ1. What are the key profiling traits of the customers who purchase hotel flash sales deals? RQ2. Is there a difference between the economic c ontribution of the flash sales term profitability? RQ3. Is there a difference between intentions to revisit a hotel demonstrated by the flash sales customers and other customers? RQ4. Is there a differen ce between intentions to recommend a hotel to others demonstrated by the flash sales customers and other customers? Significance of the Study This study provides an understanding of how the flash sales customer segment compares to other customers with reg ard to its characteristics and contribution to the term and long term profitability. Flash sales customer profiling may assist hotel managers in forming expectations regarding the potential revenue management impacts that attracting customers via flash sales websites may have for the hotel. A lack of customer segment understanding may turn some hotel managers away from using flash sales websites as an inventory distribution channel and, therefore, missing potential benefits that flash sales m ay bring to the property. On the other hand, some hotels may engage in flash sales distribution without firm knowledge of the term and long term profitability. Overall, with the development of this academic research, hotel managers may gain evidence of the flash sales influence on the industry and consumers. Understanding of flash sales effects may enable hotel managers to make more
98 informed decisions regarding hotel promotion and room distribution. This study will benefits and drawbacks identified in the first study. To the best of knowledge, this study presents the first attempt to empirically asses the differences be tween flash sales customer and other customer with regard to their profile characteristics, and contribution to the short term and long term hotel profitability. This understanding is much needed for the development of future academic research on the impac t of flash sales on the hotel industry, and for supporting managerial decision making to accept or reject flash sales as a room inventory distribution channel. Therefore, the significance of the current study is in providing the first empirical assessment of the flash sales customers, and establishing the foundation for future research directions in this area. Literature Review Hotels have a long standing history with regards to multichannel room inventory distribution (Berne et al. , 2011; Buhalis & Law, 2008; Carroll & Siguaw, 2003; Emmer et al., 1993; Kracht & Wang, 2010). The main goal of using multiple distribution channels is to ensure a timely distribution of the perishable hotel product (i.e. room night) to the customers, and, therefore, maximize re venue generation for the hotel. In order to satisfy the requirement of effective revenue management, a distribution channel should be able to deliver the right product, to the right customer, at the right time, and at the right price (Kimes, 1989; Kimes & Wirtz, 2003). The results of study 1 indicated that, when it comes to flash sales, hotel managers are given a chance to develop a package of the right product and right price that would be attractive to customers and assist hotels in covering the low deman d
99 is attracted to the hotel by flash sales would be the right customer for the property. Managers further explained that they suspect these flash sales customers may be deal seekers, who are always hunting for a better deal, not consuming additional services at the hotel, and not converting into repeat customers. Based on the results of study one, the review of literature was conducted to further understand the characteristics of deal prone customers. The literature review continues with the promotions, and customer value. mers has been examined in literature from multiple management perspectives (Reichheld & Teal , 2001; Woo & Fock, 2004). For example, Reichheld and Teal n on the firm's seems to be in line with the concerns of the hotel managers who participated in study one regarding the economic contribution of flash sales customers, and their likelihood to convert into repeat customers. customers, Woo and Fock (2004) proposed that and satisfaction. This suggestion may be well connected to the definition by Reichheld and Teal (2001), where profitabi lity would correspond to cash flow generation, and
100 satisfaction may lead to loyalty. Even though the concepts of satisfaction and loyalty are not identical, numerous studies highlight a relationship between the constructs of satisfaction and loyalty (e.g., Anderson & Srinivasan, 2003; Bloemer & Kasper, 1995; Lee & Lee, 2013; Nam, Ekinci, & Whyatt, 2011; Oliver, 1980). It has been suggested that satisfaction serves as one of the antecedents for customer loyalty as well as behavioral intentions, such as repur chasing the product and recommending it to others. Based on the definitions given above, the first component of evaluating the Determining whether flash sales customers are th examination of their purchasing behaviors in both the short and long run of time. For example, in the short term it is necessary to examine their purchasing behaviors that are related to the core product (i.e. room night) and auxiliary services during their hotel stay. From a long run profitability perspective, it is important to examine their intentions to revisit a hotel thus generating repeat business for the property. When evaluating hotel guest profitability, hotel man agers may control the price that the customers are paying for flash sales promotions, however, predicting customer in house expenditures is a more challenging task. In accordance with the total revenue management approach, both components of the revenue ge neration should be evaluated in order to make a judgment about whether a customer is the right type of customer. The reason for this is that total revenue management proposes evaluation of t operating departments at a hotel, including rooms and other non room operating departments (Anderson & Xie, 2010).
101 ty perspective. Literature suggests that in becoming a loyal customer, customers go through several different stages including: cognitive (i.e., knowledge), affective (i.e., emotional attachment), conative (i.e., commitment), and action (i.e., behavioral) loyalty (Han, Kim, & Kim, 2011; Yuksel, Yuksel, & Bilim, 2010). Based on the research purpose, the current study property and recommend it to others. Given the relationship between loyalty and satisfaction that was mentioned earlier, it also becomes important to account for the hotel, and recommend it to others. In summary, based on t he views of the managers presented in the first study of and auxiliary servic es at a hotel, and develops a long term relationship with the property through revisiting the hotel and recommending it to others. This definition is developed with the assumption that a customer is satisfied with the service received at the hotel, which s hould result in developing favorable future intentions to revisit and recommend ics of the customers who usually respond to sales promotions. Customer Response to Sales P romotions With the emergence of flash sales as a new promotional tool, it becomes to
102 flash sales (Sigala, 2013). Previously, different empirical studies aimed to identify Neslin, 1998; Neslin & Shoemaker, 1998), and services, such as postal service (Nus air et. al, 2010) hotels, restaurants (Murphy, Semrad, & Yost, 2013; Nusair et. al, 2010), and leisure (Wakefield & Barnes, 199 7 ). Online flash sales promotions come with some unique features that pertain to the promotional design (e.g., available for cer tain time period), and delivery methods (e.g., more targeted towards consumer characteristics) (Sigala, 2013) making flash sales customer base more targeted and spontaneous than other promotions. These unique features of flash sales promotions may make fla sh sales appealing to a customer segment that is different from those customers who respond to traditional sales promotions (e.g., coupons, in store promotions, etc.). Understanding the key profiling characteristics of such customers is important in order to make sure that promotions devoted to understanding the profile of this new market segment. Therefore, this exploratory study reviews the profile of deal prone customers, an d attempts to build a profile of flash sales customers to compare them to other hotel customers. Previous research on sales promotions have suggested that consumers respond to promotions based on perceived benefits and costs (Bawa & Shoemaker, 1987; Blatt berg, Buesing, Peacock, & Sen, 1978; Chandon, 1995; Chandon et al., 2000; Mittal, 1994). The benefits of sales promotions include utilitarian/economic and hedonic benefits. Each of the benefits includes several subcategories. For example, utilitarian bene fits include the following subcategories: savings, quality, and convenience.
103 Whereas hedonic benefits include the subcategories of: value expression, entertainment, exploration, and self expression. The costs of sales promotions also include several subcat egories, such as switching cost, search cost, and inventory cost. Based on this literature, it may be suggested that flash sales, as a form of sales promotions, might come with similar benefits and costs for consumers as other sales promotions. Further, t he benefits and costs of sales promotions were also associated with particular psychographic characteristics that define consumer re sponse to promotions (Ailawadi et al. , 2001; MartÃnez & Montaner, 2006). Assuming that flash sales may come with similar ben efits and costs for consumers, it may be suggested that flash sales customers may exhibit similar psychographic characteristics as customers who respond to other promotions. Given the exploratory nature of this research, the study will evaluate the profil e of flash sales customers starting with the psychographic traits that have been already associated with customer response to other sales promotions. Current research considers only those benefits and costs of sales promotions that may be directly mapped t o some consumer psychographic traits. For example, convenience benefit, search, and inventory costs were excluded from the discussion here, since they pertain more to the characteristics of a seller rather than a consumer. Utilitarian benefits of sales pro motions Savings (price consciousness, financial constraints, consumer spending self control) . Literature suggests that the savings benefit of sales promotions is particularly attractive to price conscious, financially constrained customers with high spendi ng self control (Ailawadi et al., 2001; Ayadi, 2013; Chandon et al., 2000; Haws,
104 Bearden & Nenkov, 201 2 ; Kim & Martinez, 2013; Lichtenstein, Netemeyer, & Burton, 1990; Lim, Kim & Runyan, 2013; Martinez & Montaner, 2006). Price consciousness describes those customers who are concerned about finding the best (lowest possible) price on the market. Price consciousness may mean that flash sales costumers are responding to promotions because they are in search of lower prices. If this is the case, then this cha racteristic may substantiate the concern of hotel managers that was expressed in study 1 regarding the perception that flash sales customers are deal seekers. Frequently, price consciousness as a personality trait may be associated with a financially const rained customer. Financial constraints may force customers to look for lower prices, and therefore, may relate to their level of price consciousness. Another psychographic trait that may be associated with price consciousness is consumers spending self co ntrol. Spending self control refers to the ability to control purchase impulses, and stay within the allot ted budget when shopping (Haws et al. , 2012). In the context of flash sales, customers may be low in self control, lose track of their expenditures, a nd respond to flash sales promotions that require a spontaneous about flash sales customers are tight (when staying in the hotel) in their spendings. On the other hand, flash sales customers may be high on spending self control and intentionally shop for flash sales promotions to stay within the budget. Since flash sales provide a deep discount for a hotel stay, these websites may appeal to those customers who travel on a budge t and who are conscious about the price. Therefore, as managers indicated in study one, it may be likely that flash sales
105 customers are more price conscious, experience higher financial constraints, and exhibit higher spending self control than other custo mers. Therefore, it seems that all three of the psychographic traits associated with the savings benefit of sales promotions may be relevant to describing flash sales customers. Literature seems to support that this rationale may be worth exploring in ord er to develop a customer profile for flash sales customers. Quality (quality consciousness) . The second utilitarian benefit associated with sales promotions is a quality benefit. A quality benefit of sales promotions refers to an advantage of a higher quality product sold at a lower promotional price (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). Quality consciousness may be applicable to flash sales customers, because deeply discounted hotel room pric es offered on flash sales websites may allow consumers to purchase a higher quality product for the same amount of money. For example, a customer, who usually stays in midscale hotels, may make a reservation at an upper upscale or a luxury property for the same price on flash sales websites. Therefore, using flash sales may provide a significant increase in the service level, and, therefore, appeal to quality conscious consumers. Consequently, comparing flash sales customers to other customers may reveal th at flash sales patrons are more quality conscious. Hedonic benefits of sales promotions Exploration (innovativeness, variety seeking and impulsiveness) . From the perspective of hedonic benefits, consumers may be driven to sales promotions for exploration a nd self expression. The psychographic traits associated with exploration include innovativeness, variety seeking, and impulsiveness (Ailawadi et al., 2001;
106 Chandon et al., 2000; Kim & Martinez, 2013; Martin ez & Montaner, 2006; Rook & Fis her, 1995). Innovat iveness describes types of consumers who like to try new and innovative products as soon as they appear on the market (Ailawadi et al., 2001). For example, in the hotel environment some customers are eager to try and embrace new mobile booking technologies (Wang & Wang, 2010), or biometric technologies for hotels ( Zhu & Morosan, 201 4 ), while others may be hesitant to change, and prefer their old ways of doing things. Innovativeness may also be a characteristic of hotel flash sales customers. This is becaus e flash sales are a relatively new room distribution channel, and customers who consider themselves innovative may be more likely to try this new channel to book a hotel room. It may be expected that those consumers who usually try new products, services a nd technologies first may also be inclined to try hotel flash sales. Therefore, it is worth exploring if flash sales consumers may be more innovative than hotel guests who come through other distribution channels. Variety seeking is a quality of those cons umers who seek variety and prefer to try different product and services whenever possible. Wakefield and Barnes (199 7 ) identified a positive impact of variety seeking tendency on promotion proneness for leisure services. Variety seeking may be expected to characterize flash sales customers, since the hotel flash sales promotions are designed to induce trial and make dozens of hotel options available every day. Therefore, flash sales promotions give a chance to the consumers to try different hotels at afford able prices. Given this, it is valuable to learn if we may expect that flash sales customers may be more variety seeking than other hotel customers.
107 Another psychographic characteristic that seems to be important to consider when profiling flash sales cust omers is impulsiveness (Ayadi, et al., 2013). Flash sales provide a lot of stimuli to encourage buying. For example, a typical flash sales deal may include a deep discount, a limited time frame to purchase the deal, and a limited quantity of the discounted product to entice an impulsive purchase. Flash sales deal subscribers receive an email with the deal promotions regularly offering time limited deep discounts that are intended to induce a spontaneous trial of a product. Therefore, it may be suggested th at flash sales customers may possess the trait of impulsive consumers. Self expression (market mavenism) . Self expression presents another hedonic benefit associated with using sales promotions. Self expression may become an attractive hedonic benefit of s ales promotions for market maven consumers (Ailawadi et al., 2001; Chandon et al., 2000; Feick & Price, 1987; Lim et al. , 2013; Martinez & Montaner, 2006). Market maven describes a person who possesses information about products and prices on the market, a nd enjoys sharing it with others. Market mavens achieve self expression through demonstrating and spreading their knowledge. characteristics. For example, as a relatively new distri bution channel flash sales are not known to all consumers on the market. At the same time, flash sales websites provide a variety of products and services at deeply discounted prices. A combination of these two factors would most likely appeal to market ma vens, and stimulate their need to share valuable market information with others. This suggests that flash sales customers
108 may measure higher on market mavenism than customers who purchase via other distribution channels. Costs of sales promotions Literatur e recognizes that affiliated costs of sales promotions include the following: switching cost, search cost, and inventory cost (Ailawadi et al., 2001; Martinez & Montaner, 2006). It is possible that these affiliated costs may turn some customers away from responding to sales promotions. As it was mentioned earlier, this research focuses only on those benefits and costs that are associated with customer psychographic traits. For this reason, the costs of search and inventory would not be considered, since th ey pertain to the amount of time needed to find the promotions, and the availability of the space to store the purchased promotional items, and do not deal with associated psychographic traits. Given this, the review continues with the review of switching costs. Switching costs refer to the learning, transaction, and other costs associated with trying a product that is different form a product regularly purchased by a customer (Klemperer, 1987). In the context of hotel flash sales, customers may be turned a way from purchasing a promotion, because a flash sales website may not offer a hotel brand where a customer usually stays, or due to the amount of learning required to understand how flash sales work. Literature suggests that switching costs may be more de eply experienced by loyal customers (Ailawadi et al., 2001). particular brand. When such commitment is present, a customer is less likely to switch to other brands and try their pr oduct, even if it is being promoted with a price discount (Ailawadi et al., 2001; Martinez & Montaner, 2006). Therefore, loyal customers are less
109 likely to be promotion prone. On the other hand, non loyal customers are more likely to respond to promotions because they are not attached to a particular brand or product. In the context of hotel flash sales this may mean that brand loyalty may be loyal customers may be more l ikely to book a hotel via flash sales websites. However, loyal customers may be more likely to book a hotel room directly through the brand, instead of waiting and searching for that hotel to appear on flash sales websites. Therefore, it may be expected th at flash sales customers may exhibit lower loyalty to a brand than other customers. Overall, the psychographic traits of promotion prone consumers appear to be applicable to describing flash sales customers. However, to the best of knowledge no studies hav e empirically assessed the psychographic traits of flash sales customers. Therefore, there seems to be a clear need to establish the psychographic portrait of these customers (Sigala, 2013). Without such research, managers may be amiss in understanding fla sh sales customers behaviors and potential value (or non value) to the firm. Customer Value Based on the prior research that indicates a link between customer expenditures with particular socio demographic characteristics, this study explores whether hote l flash sales guests may also share some communalities in terms of their expenditure pattern. Managers in the first study indicated that flash sales consumer spending hotel revenue management to focus on property wide profits (Anderson & Xie, 2010), this research suggests to investigate how much flash sales hotel guests spend in other
110 revenue operating hotel departments, such as restaurants, bars, spas, etc. Such expend term profits. Several studies focused on identifying the relationship between traveler types and travel related expenditures (Jang, Ismail, & Ham, 2001; Legoherel, 1998; Mok & Iverson, 2000; Pizam & R eichel, 1979). These studies have categorized travelers into different groups based on their level of spending, such as big spenders and little spenders (Pizam & Reichel, 1979), or heavy, medium, and light spenders (Jang et al., 2001). According to Pizam a nd Reichel (1979), demographic and socio economic factors may be used in order to discriminate between big spenders and little spenders. Jang et al. (2001) included trip related factors (e.g. type, purpose, duration, travel companions) along with socio dem ographic characteristics (e.g. gender, age, education, occupation, annual income) to build the profiles of heavy, medium and light spenders. Taken all together, this study represents an empirical attempt to assess the profile of flash sales customers and compare them to the characteristics of customers who come to hotels from other distribution channels. This study departs from the literature (Reichheld & Teal , 2001; Woo & Fock, 2004) and the concern of the hotel managers from study 1, and evaluates flash sales customers from the perspective of their psychographic traits, short term expenditures at the hotel, and long term future behavioral intentions (revisit intentions and recommending the property to others). As previously mentioned, this means that for purposes of this study, the right core and auxiliary services at a hotel, and devel ops long term relationship with the
111 property through revisiting the hotel and recommending it to others. On the contrary, the operating departments, and, therefore, does not c term profitability, and does not develop a long term relationship with the hotel via repeat visitations and recommending the hotel to others. The following section describes how this study was designed in order to perform thi s assessment. Methods and Procedures The purpose of this study is to determine the profiling traits of flash sales customers, as well as to investigate whether flash sales customers may be considered developed from two financial perspectives that include the short and long term effects ns that were expressed in study 1 regarding the lifetime value of flash sales customers. Research Design A cross hotel flash sales. Cross sectional survey research designs are commonly us ed to describe characteristics of the population: psychological, attitudinal, behavioral, etc. (Creswell, 2011; Nicholas, 2009). In application to this research, the cross sectional survey design was used to collect data in a short amount of time while cap turing current attitudes and behaviors that would allow for the comparison between flash sales expenditures, intentions to revisit a hotel and recommend the hotel to others). Giv en the following research questions, which aim to describe flash sales customer profiles and to
112 survey research design was deemed appropriate for this study in order to answer the following research questions: RQ1. What are the key profiling traits of the customers who purchase hotel flash sales deals? RQ2. Is there a difference between the economic contribution of the flash sales short term profitability? RQ3. Is there a difference between intentions to revisit a hotel demonstrated by the flash sales customers and other customers? RQ4. Is there a difference between intentions to recommend a hotel to others demonstrated by the fla sh sales customers and other customers? An online questionnaire was developed using Qualtrics software. Some of the key advantages of using an online instrument include, but are not limited to: an expedited data collection, an inexpensive access to a geo graphically diverse sample, elimination of the data entry chore, and, therefore, minimization of data entry errors (Creswell, 2011; Ritter & Sue, 2007; Van Selm & Jankowski, 2006). In application to the current research, it is important to note that a web based survey design is recommended when such a mode of data collection supports the purpose of the study (Van Selm & Jankowski, 2006). Given the purpose of this study, and distribution of flash sales deals online (i.e. websites, via email, or mobile applic ations), an online data collection mode was deemed as the most suitable option to reach the target sample of online shoppers. Instrumentation The online questionnaire was developed based on a review of literature that addressed managerial concerns that we re expressed in the first study of this
113 included deal seeking, tight spending behavior, and no repeat visitation to the property. For this reason, a literature review o n response to price promotions was evaluated with a close focus on consumer demographics and psychographic characteristics that are associated with the use of promotions. Nine key psychographic traits were selected through the review of relevant literatur e. These traits included variety seeking, innovativeness, impulsiveness, market mavenism, brand loyalty, price consciousness, value consciousness, financial constraints, and spending self control (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Mon taner, 2006; Wakefield & Barnes, 199 7 ). Respective items for the nine identified constructs were adopted from the literature to ensure the development of a psychometrically sound questionnaire. The items were modified to fit the hotel context of the study. According to a managerial assumption that was revealed in study 1, it seems that managers perceive that guests who come to their properties via flash sales websites are different from customers who purchase via other distribution channels. Therefore, the questions in the survey aimed at identifying potential differences in customer travel behavior, psychographic traits, demographic characteristics, spending behavior while staying at hotels, the likelihood to revisit, and the likelihood to recommend the pr operty room rate deals and demonstrate less likelihood to return to a specific hotel or brand it becomes necessary to determine if flash sales customers actually possess th ese traits. One may assume that because an individual purchases a good deal through a promotion on flash sales websites that they may possess a deal seeking personality.
114 However, there is no empirical evidence to suggest that this would be the case with flash sales customers who arrive as guests to a hotel. In order to check for this assumption and start building a profile of the flash sales customers, the first section of the questionnaire included multiple choice questions about general hotel searching and booking behavior. Please refer to Appendix B to see all items included in the questionnaire. In tandem to past couponing literature it was anticipated that customers who respond to price promotions, and, in this case, purchase flash sales deals, share particular psychographic characteristics. Nine psychographic traits were retrieved from the literature, and assessed in the second section of the survey, in order to complete a flash sales customer profile. All psychographic items were derived from previou s literature on consumer response to promotions and adapted to reflect the nature of the study. The items were measured on a 7 point Likert scale and anchored at 1 Strongly disagree and 7 Strongly agree. The items that were used to measure psychographi c constructs are presented in Appendix C. type of customer because they do not purchase additional services in the hotel and do not become repeat guests, the next section of the survey evaluated customer expenditures while at the hotel and their future behavior intentions with regard to coming back to the hotel, and recommending the property to others. All respondents were presented with a list of non room operating departments i n a hotel (e.g., bars, restaurants, beach, parking, etc.), and asked to provide the best estimate of their expenditures in those operating departments during the last leisure hotel stay. Similarly,
115 all respondents were asked to provide their expenditures f or businesses patronized in the nearby area (e.g., airlines, cruise lines, museums, theme parks, etc.). All respondents were also asked to express their likelihood of returning to the same hotel again, returning to this hotel at a regular price, and recom mending this property to others. The questionnaire concluded with socio demographic questions that assessed Based on the review of relevant literature, it was identifie d that psychographic traits that are associated with response to price promotions may have a relationship with consumer demographic characteristics (Ailawadi et al., 2001). For example, a younger age may be associated with lower income, and therefore, high er price consciousness. This section of the survey also contained a comment box for the respondents to provide their feedback on the survey, or any details about their travel that they wished to share. The last item of the survey thanked all study particip ants for their contribution and presented them with a unique random code to receive the payment after completing the questionnaire. Variables and Data Analysis Strategy The data was analyzed quantitatively using SPSS 21 software (Statistical Package for So cial Sciences). The data analysis strategy was developed in accordance with the research questions stated for the study. The first stage in data analysis involved data cleaning and preparation. The data was cleaned to contain fully completed questionnaires from those respondents who qualified for the study. Next, outlier analysis and assumptions check were conducted in order to satisfy the requirements of selected analyses.
116 After data cleaning and preparation, descriptive statistics (i.e. means, frequencies ) were calculated in order to describe the study participants with regard to their socio demographic characteristics and travel behavior. Frequencies were used for categorical variables, such as gender, marital status, educational level, etc. On the other hand, means were calculated for metric variables, such as nightly room rate, and additional expenditures at a hotel. RQ1. What are the key profiling traits of the customers who purchase hotel flash sales deals? In order to answer the first research questio n, the study introduced variables that demographic characteristics. The variables for this research question were measured through multiple choice, Likert scale, and short answer questions. The variables used to describe the general travel behavior included the number of nights spent in a hotel in the last 12 months, distribution channels used for search and booking, and general word of mouth behavior for sharing travel experience s. Two short answer questions asked respondents to enter the number of nights they spent in a hotel in the last year (12 months), along with the number of nights spent in a hotel for leisure purposes. Multiple choice (all that apply) question was used to assess the use of different distribution channels for hotel search and booking. Study participants were offered a list of distribution channels that included flash sales (e.g., Groupon Getaways), last minute deal websites (e.g., Hotels Tonight), online t ravel agencies (e.g., Expedia ), hotel website (e.g., Marriott.com), third party review website (e.g., TripAdvisor ), meta search engines that search several websites at the same time (e.g., Kayak ), social media (e.g., Facebook ), and a travel agent. In a ddition, all respondents were asked to
117 choose one of the channels that they use the most for hotel booking. General word of mouth behavior was measured with regard to sharing travel experience with friends and relatives, on hotel comment cards, on hotel on line surveys, and on traveler rev iew ). These items were measured on a 5 point Likert scale anchored at Never, Rarely, Sometimes, Often, and Always. The psychographic traits were derived from the literature regarding response t o price promotions. All nine constructs (variety seeking, innovativeness, impulsiveness, market mavenism, brand loyalty, price consciousness, value consciousness, financial constraints, and spending self control) were measured on a 7 point Likert scale anc hored at 1 Strongly disagree and 7 Strongly agree. The number of items for each construct fluctuated between three and five. Please refer to Appendix C for the list of adapted items, along with the original items and literature sources. Socio demograp hic characteristics were measured with traditional items that included gender, age, marital status, presence of children, education, income, and state of residency. Majority of the questions (gender, marital status, children, education, and income) were re corded using multiple choice questions. The study participants were asked to indicate their age in a short answer question. A drop down list with the US states was offered to the participants to select their state of residency. In order to answer the first research question about the key profiling traits of flash sales customers, and exploratory factor analysis (EFA), scale reliability check, multivariate analysis of variance (MANOVA), and discriminant analysis were conducted. EFA was used to check for cons istency between the psychographic dimensions used in this study and the dimensions associated with response to sales promotions identified in
118 previous literature . In the case of this study, the following dimensions of the psychographic traits were expected : variety seeking, innovativeness, impulsiveness, market mavenism, brand loyalty, price consciousness, value consciousness, financial constraints, and spending self control. In order to determine the appropriateness of factor analysis, the following test s were conducted. Kaiser Mayer Olkin (KMO) test was utilized as the overall measure of the sampling adequacy. A high result of the KMO test (0.8 and above) was expected to cted in order to identify significant correlations among the variables that will be factor analyzed. Principal component factor analysis with varimax rotation was conducted in order to explore the structure of the psychographic dimensions. Varimax was sele cted as a rotation technique that disperses the factor loadings within the factor, and, therefore, produces more interpretable clusters (Field, 2009). Factors were formed based on eigenvalues greater than 1. Expected results included observing factor loadi ngs above 0.5. Following the factor analysis, all scale dimensions were also checked for a high reliability when achieved alpha coefficient of 0.7 or above (Hair , Blac k, Babin, & Anderson , 2010). Based on the results of factor analysis and reliability testing, summated scale scores were calculated for each dimension to be used in further analysis. In order to identify the differences between flash sales and other custo mers with regard to key personality traits, multivariate analysis of variance (MANOVA) was
119 conducted. Conducting MANOVA instead of a series of ANOVA analyses helps to minimize the risk of conducting a Type I error, and also assists researchers in taking in to consideration possible relationships between dependent variables (Hair et al., 2010). MANOVA is designed to discover the differences of dependent metric variables across the levels of independent categorical variable. In this case, distribution channel was used as an independent categorical variable with two levels: flash sales and other channels. Customer psychographic traits entered the analysis as dependent metric variables. As suggested by Field (2009), a MANOVA was followed with a discriminant anal ysis to further understand the results, and discover which variables have substantial effect on discriminating between the two groups (flash sales customers and other). The dependent variables of MANOVA (customer psychographic traits) were entered as indep endent variables in discriminant analysis. The categorical variable of distribution channel groups (flash sales customers vs. other) was used as a dependent variable in discriminant analysis. RQ2. Is there a difference between the economic contribution of the flash sales customers and other customers? Several variables pertaining to research question two were included in the questionnaire. The key variable for this question is the additional expenditure in a hotel. This variable was measured via a short an swer question that asked respondents to provide their expenditures at different non room operating departments in a hotel: hotel restaurant, room service, hotel bar/lounge, mini bar, spa, golf, gift shop, umbrella / sunbed rental, movie on demand, parking or valet, Internet access, and other. This variable is specific to the most recent hotel stay, or the most recent hotel stay that was
120 purchased via flash sales websites. In addition to the auxiliary services that customers could have purchased at a hotel, the duration of the hotel stay, as well as the number of people in the party were recorded. This information was necessary to convert the expenditures into per person/per night amounts to ensure a proper comparison when answering this research question. Th e second research question examined the differences on one dependent variable (expenditures in hotel non room operating departments) across two groups (flash sales customers vs. other). All auxiliary expenditures were accumulated together and expressed in per person/per night values. Since there was only one dependent metric variable, an independent sample t test was used to assess the differences between the two groups. RQ3. Is there a difference between intentions to revisit a hotel demonstrated by the flash sales customers and other customers? In order to answer the third research question, two variables were studied: regular price. Both variables were measured on a Likert scale anchored at 1 Very unlikely to 7 with their most recent hotel stay was also recorded. This variable was introduced as a covariate to answer research questi ons three and four as satisfaction is closely related to revisit intentions and word of mouth behavior. The satisfaction variable was measured on a 7 point Likert scales anchored at 1 Extremely dissatisfied to 7 Extremely Satisfied. The differences bet multivariate analysis of covariance (MANCOVA). The dependent variables for this
121 research question included intentions to stay at the hotel again, and intentions to stay at the hotel again even at re gular price. Distribution channel with two levels (i.e. flash sales vs. other) was used as an independent variable similarly to the previous analyses. Also, satisfaction was introduced as a covariate in this analysis because it is an important variable in Adding satisfaction as a covariate in this analysis is important because it allows the researcher to explore the relationships between using a particular distribution channel (flash sales or other) an d likelihood of returning to the hotel, after accounting for the effect of the covariate. If the covariate is not introduced, the variance in revisit intentions may be mistakenly attributed to differences between two groups (flash sales customers and other ). But in fact, this variance may occur due to the fact that some guests were satisfied with their hotel stay, and some were not. For this statistical technique, additional assumptions of homogeneity of regression slopes, and independence of covariate and dependent variable were checked. RQ4. Is there a difference between intentions to recommend a hotel to others demonstrated by the flash sales customers and other customers? For research question number four, an additional variable was introduced in order t o determine the likelihood of recommending the hotel to others (word of mouth). The same 7 point Likert scale was used for this variable as for revisit intention variables (anchored at 1 Very unlikely to 7 Very likely). The fourth research question was answered using an analysis of covariance (ANCOVA). The likelihood of recommending a hotel to others was measured with one item, and, therefore, determined a need for a univariate test. Satisfaction with the hotel stay again was used as a covariate to acco unt for variance of this factor in potential group differences.
122 Sample Size, Power, and Effect Size The use of different multivariate techniques requires the researcher to consider the interplay of statistical power, effect size, and sample size used in th e study. Following the convention in the hospitality and tourism literature, the alpha level for level (protection against Type II error) was set at 0.2, which leads to desired statistical power of 0.8 (Field, 2009; Hair et al., 2010). When determining a required sample size, power of the analysis should be considered along with the effect size of the test (or the magnitude of the observed effects) . At a certain level of power (0.8 is the recommended level), larger sample sizes would allow the researcher to determine even a small effect size (r = 0.1). Since several multivariate techniques were used in the current research, it became important to ex amine the adequacy of the sample size for all techniques in the analysis. According to Hair et al. (2010), the recommended group size to determine small effects using a MANOVA with eight dependent variables and power of 0.8 is 160 participants. This estima te was given for a three group comparison, and, therefore, should be sufficient for two groups as well. Following this recommendation, the total sample size accounting for two groups should be 320 participants. Another multivariate technique used in this study is factor analysis with 31 items. According to Hair et al. (2010), an acceptable sample size for an exploratory factor analysis is 10 cases per variable. Therefore, the total sample size required for factor analysis results in 310 observations. A rec ommended sample size for a discriminant analysis may be calculated based on 20 observations per each independent variable. It is important to recall that
123 psychographic traits were used as independent variables in discriminant analysis. The research design included nine psychographic variables, therefore, a required sample size for this statistical analysis would be 180 participants. Based on the analyses presented above, the estimates for sample size required for MANOVA and factor analysis are very close (3 20 and 310 respondents respectively). The sample size requirement for discriminant analysis is lower (180 participants), and would be met if the minimum for MANOVA and factor analysis is satisfied. Therefore, for the purposes of this study the desired samp le size was set to 400 participants. This adjustment was made to accommodate for data cleaning and preparation for the analyses. Data Collection Sample and data collection For the purposes of this study, a sample was drawn from the population of US consum ers who search and book hotels online. Online shoppers constitute the population of the study since flash sales promotions may be only purchased online. In order to qualify for the study, a participant should have stayed in a hotel for leisure purposes in the last 12 months. Next, all respondents were placed into one of the two categories: hotel flash sales customers, and other customers. Based on this filtering, study participants received either questions about their most recent hotel stay for leisure pur poses, or their most recent leisure hotel stay purchased via flash sales websites. Participants were recruited via Amazon Mechanical Turk marketplace for tasks that require human intelligence (Amazon, 2014). Amazon all ows employers to post human intellectual tasks (HITs), set required qualifications, specify the salary, and recruit workers who are willing and
124 qualified to perform a particular HIT. Research activities, including online surveys and experiments, represent one of the HIT types that may be completed on Amazon . Given the access to a large and diverse worker pool, along with a relatively low cost and a built in payment mechanism, has become of interest to social science researchers (Buhrmester, Kwang, & Gosling, 2011; Casler, Bickel, & Hackett, 2013; Goodman, Cryder, & Cheema, 2013). Overall, existing literature concluded that presents an inexpensive way for quick and high quality data collection. Previous research has demonstrated that research samples recruited from are usually more socio economically and ethnically diverse than American college samples (Buhrmester et al., 2011; Casler et al., 2013), and samples obtained fr om social networks (Casler et al., 2013). participants did not differ significantly from community samples in terms of demographics (e.g., age, gender, education) (Goodman et al., 2013), and produced psychometrically reliable measur es (Buhrmester et al., 2011). Four hundred fifteen participants were recruited via platform. The description of the HIT included the purpose of the study, and the link to an online questionnaire. Participants were paid 90 cents for participation in the study. At the end of the survey every participant received a unique code that was needed to process the payment on , alone with the instructions how to submit the code. The data was collected in May 2014 in four batches (during week days/weekends, and morning/evening hours) to eliminate potential biases associated with a particular day of the week and time.
125 Pilot study The surveys instrument was pilot tested two times prior to the main stage of the data collecti on. For the pre pilot test, a purposive sample of college students and professors was drawn. The pre pilot test was used to eliminate any questions about clarity of the items and to test the reliability and validity of the instrument. At this stage 64 resp onses were collected. Reliability scores of all scales were above 0.73 with the exception of price consciousness construct. Please see Table 3 alpha reliability scores. Given the low reliability score, the price consciousness construct was further revised to better fit the context of hotel search and purchasing. Since the constructs of financial constraints and consumer spending self control are closely related to the price consciousness construct, they also were reworded to ensure that all three constructs are distinct. Next, a pilot test was run on the platform. This pilot test was used to ensure a smooth execution of the survey on Amazon Mechanical , and also for checking reliability of the modified scales . Another 121 responses were collected at this stage. Reliability scores of all scales were equal or above 0.7. Please refer to Table 3 construct in the study. Based on the results of the pilot t ests and high reliability of the constructs, the study proceeded to the main data collection stage. Data Preparation and Cleaning The preparation and cleaning of the data started with elimination of those participants who did not qualify for the study (i.e ., those who have not stayed in a hotel in the last 12 months for leisure purposes). Then, incomplete surveys were also deleted
126 from the analysis. The next stage of data preparation involved an outlier analysis. The key variables in the study were examined for outliers using box plots, and z values. In a normal distribution, the absolute values of z scores are not expected to exceed 3.29 (Field, 2009). Therefore, all cases with z scores at or above this level were carefully examined. Since the responses on psychographic scales could not be mistyped, or be out of range, they were evaluated based on consistency. Some of the cases were deleted due to conflicting answers on different items of the same scale. Based on the developed data analysis strategy, the da ta needed to satisfy several assumptions of parametric tests, including normality, independence, and homogeneity of variance (including homogeneity of error variance, homogeneity of covariance matrices, and homogeneity of regression lines). Both graphical and statistical approaches were used to evaluate data for normality. Histograms with the normal distribution curve and Q Q plots were built to examine the shape of data distribution. Also, statistical tests Kolmogorov Smirnov and Shapiro Wilk were employed to check for normality. However, literature suggests that these tests are sensitive to sample size, and in large samples (above 80 respondents) usually produce significant results (Field, 2009; Hair et al., 2010). For this reason graphical evaluation was selected as the primary evaluation of the assumption of normality. Only those variables that were normally distributed were included in the analysis. Multivatiate normality was assumed based on univariate normality since no direct test is available for mul tivariate normality (Hair et al., 2010). Univariate normality is a prerequisite for multivariate normality, and when univariate assumption is satisfied, departures from multivariate normality are not likely. In order to meet the assumption of independence of observations, the survey was
127 an online self administered questionnaire that enabled data to be collected in four batches that included weekdays, weekends, and morning and evening hours. The homogeneity of variance assumption was tested using the Le homogeneity of variance test. The homogeneity of variance was tested for two groups: flash sales customers, and customers who book hotels through other channels. The researcher was looking for a non ho mogeneity of variances. A non fail to reject a null hypothesis that the variances are equal (Field, 2009; Hair et al., 2010). After satisfying the assumption of homogeneity of variance for dependent var test to ensure homogeneity of covariance matrices (Field, 2009; Hair et al., 2010) . The assumption of homogeneity of regression slopes for MANCOVA analysis was tested by adding an interaction effect of covariate and dependent variable to the model. As with two other homogeneity assumptions, this one was supported with a non significant value of the interaction term. Findings Respondent Demographics and Travel Behavior A total of 415 responses were collected for this study. Eighteen respondents were not qualified for the study (did not stay in the hotel in the last 12 months), and were ex cluded from the analysis. Then, all unfinished surveys (29), and outliers (10) were deleted from the dataset, living 358 usable cases. Out of 358 usable responses, 173 participants have made a purchase on flash sales websites. Among them, 100 participants (28%) have purchased a hotel, or a travel package that includes a hotel, on
128 flash sales websites. The rest of the sample, 258 participants (72%) have stayed in a hotel for leisure purposes in the last 12 months. The sample was evenly represented by males ( 50.6%) and females (49.4%), which closely follows US 2010 Census Data that reported 49.2% males and 50.8% females. Majority of the study participants were in a relationship or married (62.5%), and 34.7% had children. The most frequently reported education category was 4 year college degree (39.3%). The distribution of the household annual income also closely followed the US Census data. However, the sample obtained for this study was heavier represented by younger people (18 39 years old 78.7%). Please see a detailed description of the sample in Table 3 3 Respondent demographic characteristics. From the perspective of travel behavior, on average study participants spent seven nights per year in a hotel, 5.65 of them were for pleasure. About half of the respondents stayed in midscale hotels (47.2%), followed by upscale hotels (35.7%), economy (13.7%), and luxury hotels (3.4%). The most frequently used booking channel th e property or the brand (11.5%). Only 7.6% of respondents stated that flash sales websites were the channel that they use most often for hotel booking. Most frequently study participants shared their hotel experiences with friends and relatives (3.84 on a 5 point scale anchored from 1 Never to 5 Always), followed by hotel comment cards (2.42), and hotel online surveys (2.30). Next, demographic characteristics were also recorded for two groups: flash sales customers and other customers. Both groups were evenly represented by males and
129 females. Respondents in both groups achieved similar level of education with some exceptions: flash sales customers had a higher frequency of 4 year college degrees, while other customers demonstrated a higher percentage of some college education in the sample. Both groups were heavily represented by younger people (18 39 years old): flash sales customers 83%, other customers 77%. Group compositions in terms of marital status, children, and income were closely following each other. A detailed demographic portrait for each group is presented in Table 3 4 below. Travelers in both groups stayed at the hotel about equal number of nights in the last year: M FS = 7.88, SD = 9.425 vs. M O = 6.67, SD = 6.745, t (356) = 1.348 p = 0.178. The duration of stay for leisure purposes also did not differ between the groups (M FS = 6.14, SD = 8.238 vs. M O = 5.46, SD = 5.696, t (346) = 0.883, p = 0.378). In the flash sales and other travelers reported similar frequencies of staying in hotels of different service levels: midscale (f FS = 52%, f O = 45.3%) upscale (f FS = 35%, f O = 36%), economy (f FS = 11%, f O = 14.7%), and luxury (f FS = 2%, f O 2 (3) = 2.186, p = 0.535. An average room rate paid by flash sales customers was $96.76.Other customers paid on average $116.85 per night, which is close to the 2012 industry average of $106.15 (American Hotel & Lodging Association, 2014). From the perspective of hotel booking, flash sales customers booked their travel more frequently via online travel agencies (48%), flash sales (22%), and hotel websites (10%). However, other customers more frequently preferred online travel agencies (40.9), ho tel website (28.4%), and calling the hotel directly (14%).
130 When asked about the most recent leisure stay, most flash sales customers (67%) used Groupon Getaways to make a reservation. The second most frequently used website was LivingSocial Escapes (13%) , followed by TravelZoo (9%). For those customers who booked not on a flash sales website, the most frequently reported (29.8%), and calling the hotel directly (20.2%). Flash Sales Customers Personality Traits For the purpose s of forming dimensions of consumer psychographic traits , and checking for consistency between dimensions used in this study and those identified in the previous literature , a principal component ana lysis was conducted on the 31 items with orthogonal rotation (varimax). The first round of analysis revealed that items for loyalty loaded together with the items for market mavenism, and for that reason were deleted from the analysis. After that, principa l component analysis was applied to the remaining 28 items. The Kaiser Meyer Olkin measure verified the sampling adequacy for the 2 (378) = 4261.294, p < 0.001, indicated that correlations between items were sufficiently large for were retained for further analysis. These factors in combination explained 70% of variance. Table 3 5 below shows factor loadings after rotation. The items clustered in the f actors of price consciousness, variety seeking, consumer spending self control, quality consciousness, market mavenism, impulsiveness, innovativeness, and financial constraints as suggested by the previous literature.
131 All factors also displayed high reliab (Field, 2009; Hair et al., 2010). Based on the results of factor analysis, summated scale scores were calculated for the identified factors. These scores were used in the subsequent multivariate analysis of varian ce to answer the first research question. The difference in psychographic traits of flash sales and non flash sales customers were explored using MANOVA. Items for self control and impulsiveness were eliminated from the analysis due to non normality. Accor ding to Hair et al. (2010), therefore, a decision was made to exclude these two variables in order to perform a reliable assumption test. Keeping the non normally dist ributed variables in the test and, therefore, challenged the interpretation of MANOVA results. Factors of financial constraints and innovativeness were excluded from the analysis due to failing the assumption of homogeneity of variance. All other variables, namely price consciousness, variety seeking, quality consciousness, and mavenism, demonstrated the equality of error variances (p > 0.05). The assumption of equalit y of covariance other customers with regard to their personality trait 0.037. Test of between subjects effects reveled statistically significant difference between two groups on mavenism (F = 6.399, p = 0.012), quality consciousness (F = 5.308, p = 0.022), and variety seeking (F = 4.201, p = 0.041). Flash sales customers scored higher on all of these personality traits: mavenism (M FS = 3.94, SD = 1.25, vs.
132 M O = 3.54, SD = 1.41), quality consciousness (M FS = 4.09, SD = 1.35, vs. M O = 3.72, SD = 1.35), and variety seeking (M FS = 4.00, SD = 1.16, vs. M O = 3.71, SD = 1.22). However, the results revealed no statistically significant differences between the groups based on the price consciousness trait (F = 0.367, p = 0.545). In order to further understand the results, the MANOVA was follow ed up with a discriminant analysis, which revealed one discriminant function that explained 100% of 2 (4) = 10.224, p = 0.037, canonical correlation = 0.169. The structur e matrix revealed that mavenism (r = 0.783), quality consciousness (r = 0.713), and variety seeking (r = 0.635) make substantial contribution to the discriminant function of the two groups under analysis. All of these variables exceeded the recommended val ue of 0.4 and may be considered substantive (Hair et al., 2010). Price consciousness demonstrated a low negative correlation with the discriminant function, which was deemed not substantial based on the 0.4 cut off. Economic Contribution Comparison In or der to determine the differences between the economic contribution of the flash sales customers and customers who book via other distribution channels to the term profitability, all study participants were asked to report their additional spe nding at the hotel during their most recent hotel stay. The respondents were offered a list of different non room operating departments at a hotel (e.g., restaurant, bar, spa, golf, parking, etc.) to prompt the response. All additional expenditures were ad ded together and divided by the number of nights spent at a hotel, and the number of people in the party to ensure a proper comparison. This adjustment was done to ensure that
133 the expenditures of a family of four people are not compared to the expenditures of a solo traveler. After checking the assumptions, it was discovered that the variable of interest (i.e. auxiliary expenditures in a hotel) is not normally distributed. Therefore, the groups were compared using Mann Whitney U test instead of an independe nt sample t test. Expenditure levels of flash sales customers (M = 73.24, Mdn = 35.00) differ significantly from the expenditures of other customers (M = 49.86, Mdn = 5.50), U = 10667.00, z = 2.664, p = 0.008, r = 0.14. The effect size r was calculated u where Z is the z score associated with the U statistic, and n is the sample size. Therefore, r may be computed as follows r = Further comparisons were conducted in order to investigate the differences between additio nal expenditures paid by guests in hotels of different service levels. Luxury and economy hotels were excluded from the analysis due to insufficient sample size (only 2 flash sales customers from the sample stayed in luxury hotels, and 11 stayed in economy hotels). Such sample sizes would not allow for meaningful group comparison. Using Mann Whitney U statistic, flash sales customers of midscale hotels (M = 59.99, Mdn = 17.50) spent more at the property than guests who come via other distribution channels ( M = 33.62, Mdn = 0.00), U = 2472.00, z = 2.079, p = 0.038, r = 0.16 (n = 169). No statistically significant differences were found for the guests in upscale hotels, U = 1324.5, z = 1.640, p = 0.101, r = 0.15 (n = 128). Revisit Intentions Since satisfa ction with the hotel stay is an important factor in determining revisit intentions, satisfaction was used as a covariate in the multivariate analysis of covariance (MANCOVA). Prior to the analysis the independence of covariate
134 (satisfaction) and independe nt variable (distribution channel: flash sales vs. other) was checked to make sure that groups do not differ on different levels of the covariate satisfaction. In the case when this assumption is violated, and the relationship exists, the covariate would n ot be helpful in accounting for variance in the dependent variables, since it has different impact on the groups (Field, 2009). In other words, in the case of flash sale customers and other customers, this means that both groups should exhibit about the sa me level of satisfaction with their hotel stay. The independent sample t test indicated that flash sales and other customers are not different on their level of satisfaction with the most recent hotel stay, t = 0.975, p = 0.330. Since the assumption was m et, satisfaction was added as a covariate in the model. Another important assumption for MANCOVA is homogeneity of regression slopes that assumes that the relationship between the dependent variable and covariate is the same across the groups (Field, 2009) . In order to test for this assumption, an interaction effect of satisfaction and distribution channel (flash sales vs. other group) statistically significant finding and therefore supported the assumption of homogeneity The assumption of equality of covariance matrices for MANCOVA model was signific supported the equality or error variances assumption for both dependent variables: stay at the same hotel again (F (1, 351) = 1.038, p = 0.309), and stay at the same hotel again if at the regular price (F (1, 351) = 0.473, p = 0.492). Us discovered that two groups (flash sales customers and other customers) differ in their
135 intentions to revisit the hotel where they stayed last, and revisit the same hotel at a 049. The covariate, satisfaction with the hotel stay, was significantly related to p < 0.001. The test also revealed no significant relationship between a distribution channel (flash sales or other) and intentions to stay in the same hotel again, F (1,350) = 0.016, p = 0.898. However, the flash sales customers were less likely (M = 4.30 8, SE = 0.145) to stay at the same hotel at the regular price than customers who purchased through other channels (M = 4.658, SE = 0.088), F (1,350) = 4.236, p = 0.040. Recommending a Hotel to O thers Similarly, the likelihood of recommending a hotel to ot hers was tested using ANCOVA where satisfaction entered analysis as a covariate. The assumption of homogeneity of regression slopes was met, F (1, 349) = 0.020, p = 0.887. The equality 0.852, p = 0.357. The covariate, satisfaction with the hotel stay, was significantly related to 0.001. However, there were no significant differences between the flash sa les customers (M = 5.238, SE = 0.060) and customers who purchased through other channels (M = 5.353, SE = 0.100) in their intentions to recommend the hotel to others, F(1, 350) = 0.960, p = 0.328. Implications The results of the study have revealed that a bout one third of leisure travelers have purchased hotels via a flash sales website, therefore, highlighting a need to further
136 understand the profile and behavior of those customers who are attracted to the hotels via flash sales websites. Such understandi ng is important not only because customers are making hotel reservations via flash sales websites, but also because hotel managers need to make an informed decision whether or not to distribute room inventory via flash sales websites. The results from stud y 1 revealed some managerial concerns that flash sales customers may possess consumer traits that could result in Managers specified that they consider guests arriving via flash sales web sites short and long term economic contribution may be compromised by financial constraints or price consciousness, and they may not convert to repeat customers. These ma nagerial speculations were based on the very construction and nature of flash sales deals, whereupon products are sold through deep discounting practice in order to attract more customers during need based periods. The current study investigated the chara cteristics of the flash sales customers with regard to all of the aforementioned concerns, and compared flash sales customers to other customers. Deal Seeking Flash sales customers and other customers possess similar demographic characteristics with regar ds to gender, age, education, marital status, children and income. They also exhibited similar travel behaviors with regards to the number of nights spent in a hotel per year. Online travel agencies were named as a distribution channel that flash sales cus tomers and other customers use most often to book their accommodations (48% and 40.9% respectively). This suggests that both groups enjoy the convenience of online hotel search across different hotels and brands, and easy
137 price comparisons (Buhalis & Law, 2008; Carroll & Siguaw, 2003; Kracht & Wang, 2010). However, the results from this study did reveal some differences between the groups based on their preferred method of booking. When examining other choices of hotel distribution channels, non flash sale s customers use direct distribution channels (i.e., hotel website and calling the hotel directly) more often. On the contrary, flash sales customers prefer flash sales websites (22%), followed by the hotel website (10%). This finding actually indicates an advantage to that of other hotel customers over flash sales customers. This advantage is recognized in previous hotel distribution channel literature where it has been suggested that direct channels are more profitable for hotels (Kang, Brewer, & Baloglu, 2007). This is because direct distribution channels do not require the use of an intermediary party to sell the room. Therefore, the profit from the room sale is captured all in house as well as the cost associated with selling the room is minimized. Price consciousness In order to address a concern of hotel managers about deal seeking tendency of flash sales customers, this study assessed the differences between psychographic characteristics of flash sales and other customers. Hotel managers who participat ed in the first study suggested that flash sales customers are deal seekers, who are always looking for a greater discount. However, no difference was found between flash sales customers and other customers on the price consciousness trait. This indicates that both groups are equally concerned about finding the best possible price for their accommodations, and flash sales customers do not seem to be any different from other customers in this regard.
138 In addition, price consciousness did not come out as a con tributing factor in the discriminant function that was built to differentiate between flash sales and other customers. This seems to indicate that price consciousness is not one of the factors that determine differences between flash sales and other custom ers. Therefore, the are deal seekers, who are always looking for a discount, was not realized. d efinition, this result presents an interesting managerial implication. The result indicates total revenue as a financial performance benchmark that flash sales customers m ay turn out to be more profitable customers who are willing to pay the right prices (Reichheld & Teal , 2001; Woo & Fock, 2004). Thus, this finding may make flash sales Previous research suggests that deal prone customers are usually price conscious (Ailawadi et al., 2001; Martinez & Montaner, 2006). Martinez and Montaner (2006), who investigated psychographic traits of deal prone customers in the context of shopping for grocery items and cleaning supplies, identified that coupon prone customers, as well as customers who respond to in store promotions and store flyers are all price conscious. However, in the current study that was conducted in the context of hotel purchases, flash sales customers (who were la beled by managers as deal prone customers) did not seem to differ from other customers (who did not use flash sales) on price consciousness. This finding may suggest that flash sales customers seem to be attracted to flash sales by other utilitarian (e.g., higher quality) or hedonic
139 benefits (e.g., exploration or self expression) of price promotions (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). Given the findings above, the implications for hotel managers may include focusing on other benefits of price promotions, rather than just a price discount. Currently, flash sales websites do a great job in highlighting the monetary value of the promotions featured on the websites. For example, Groupon Getaways displays the promotional pr ice, original price, the discount percentage, and the value that customers will save by purchasing a deal. However, as the results of this study suggest, flash sales customers may be looking not only for the best promotional price, but also for higher qual ity product and variety, which will be discussed later. Therefore, to make flash sales promotions more appealing to the customers, hotel managers may want to highlight other benefits when designing their deals. As suggested by Chandon et al. (2000), anothe r utilitarian benefit of sales promotions includes quality. Given that, hotel managers may want to highlight the quality of their product when developing flash sales deals. Hotel managers may pay extra attention to developing vivid narrative about the prop erty, and taking professional pictures to be included in the flash sales deals. Another aspect of highlighting the quality may be to promote consumer reviews on the web page of the hotel flash sales deal (Sigala, 2013). Market mavenism Additionally, diff erences between flash sales and other customers were found with regard to some other psychographic traits that include market mavens, quality conscious, and variety seeking types of consumers. Market mavenism describes those customers who accumulate inform ation about markets and prices, and enjoy sharing
140 their knowledge with others. In the context of hotel flash sales this means that flash sales customers may be eager to tell their circles about flash sales, as a distribution channel that provides great opp ortunities to book great hotels at great prices. Market mavenism and the behavior that is associated with it may have different implications for hotel managers. On one hand, managers may not want more customers to turn to flash sales, because rooms are so ld at deeply discounted prices, and flash sales websites also receive a fairly high commission rate for their services (about 50%) (Picolli & Dev, 2012; Sigala, 2013). This means that the more market mavenists that communicate to other consumers about the great opportunities provided through flash sales the more popular this distribution channel may become. In turn, more consumer demand through flash sales websites may mean potentially losing reservations through other distribution channels as well as gene rating less room revenue by paying extra amounts in commission fees. The extreme consequence could result in more customers using flash sales websites, and fewer customers using direct channels where the profit is increased (Kang et al., 2007). However, when implemented strategically, mavenism of flash sales customers may be used to the advantage of hotels. As the results of the first study showed, a low flash sales to seek help in distributing unsold room inventory. Therefore, in low demand seasons, mavenism of flash sales customers may be beneficial to hotel managers, since the more people become aware and subscribe to flash sales, the higher the power and reach of flash sa les promotions. This means that if flash sales are strategically used
141 by hotel managers, and distributed only during low demand seasons, mavenism of existing flash sales customers could serve as a sales surge that may benefit hotels. Besides that, market m avenism of flash sales customers may be utilized by hotel managers to their advantage outside of flash sales context. Managers who participated in study 1 indicated that the ideal scenario of using flash sales websites would be to reach new customers durin customers in house, and convert these new customers into repeat customers (assuming the services rendered were deemed highly satisfactory). Following this ideal scenario, hotel managers may develop rel ationships with flash sales customers as they receive their contact information upon the first visit. In this case, if flash sales customers are kept informed about the market strategies, initiatives, and prices of the subject hotel, they are likely to sha re this information with their circles (Feick & Price, 1987). This means that if a hotel introduces something special and unique on the market (e.g., innovative service, unique reward program, special deals, etc.) they may use the power of market mavens to spread the word about the innovation that they brought to the market. Quality consciousness The second characteristic where flash sales customers differed from other customers was quality consciousness. As suggested by Chandon et al. (2000), purchasing a product of higher quality represents another utilitarian benefit of sales promotions. This may mean that flash sales customers use flash sales websites in search for a higher quality product, instead of just a lower price. Also, when compared to other lei sure travelers, flash sales customers seem to be more concerned about quality. For hotel managers this may suggest that flash sales
142 customers are more likely to require and purchase high quality products. Besides this, it may be an indicator that when pres ented with several versions of the same product or service, flash sales customers may prefer an option that they perceive to have higher quality. This means that when hotel managers construct the flash sales deal it may benefit managers to take into consid eration this personality trait and develop a sale that provides access to the highest level quality of rooms and services. In other words, hotel managers using flash sales may have the opportunity to generate higher revenue from flash sales customers, who seem to prefer high quality products that are generally more expensive. This is especially important given that flash sales are traditionally used during need based periods when total revenue becomes a critical measure of financial performance for hotels . The finding of high quality consciousness of flash sales customers substantiates the quality benefit of sales promotions suggested by Chandon et al. (2000). However, it appears to be contradictory to findings of Martinez and Montaner (2006) that deal prone customers are not quality conscious. In the case of the current research, flash sales customers, who hotel managers deemed to be deal seekers, in fact, are even more quality conscious than other leisure customers. Such contradictions suggests that th ere are indeed differences between the customers responding to regular and online promotions (Sigala, 2013); and, therefore, highlights the need to further investigate and refine the profile of customers who respond to online promotions. Variety seeking Ne xt, flash sales customers were also found to be more variety seeking than other hotel customers. This means that among two suitable options, flash sales customers would be more likely to choose the one that they have not tried yet. On the
143 one hand, this ap pears to bring negative consequences to the hotels. Variety seeking may mean that if a customer has already stayed in a particular hotel, they are not likely to come back and less likely to become repeat customers (Martinez & Montaner, 2006). Therefore, t flash sales customers do not generate repeat business. Variety seeking may explain the customer behavior of avoiding hotels where they have already stayed, and instead search for someth ing different that they have not tried before. However, on the other hand, variety seeking of flash sales customers may be used by hotel managers to their advantage. Certain types of hotel properties (e.g., luxury hotels and resorts) have numerous revenue operating departments that could be used to bundle a new product each time for the variety seeker to try and purchase through a flash sales promotion. This means that the same property may package their services every time in a different way to showcase t he available amenities, and highlight the variety of experiences that hotel guests may receive during their stay. For example, a hotel could create one package for distribution on a flash sale website that would include a standard room, a fine dining exper ience, and a spa treatment. But next time, the same hotel may create another packaged bundle (with variety seeking in mind) that would include a suite level room, buffet style meals, and a romantic bon fire experience on the beach with cocktails included. This way, hotel managers may encourage variety seeking customers to revisit the same hotel, and also to patronize different operating departments while on property. Assuming that variety seeking spreads not only to the hotel search, but also to preference of destinations, leisure and other activities, hotel managers may
144 market their properties to the flash sales customers in a way that highlights variety. For example, branded hotels may offer a variety of destinations to the flash sales customers , and invite them to experience a variety of places while staying with the same hotel brand. Building on the quality consciousness of flash sales customers, hotel managers may also want to highlight that staying with the same hotel brand would ensure a qua lity standard for the customers. Alternatively, hotels (both independent and branded) may promote the variety of services that they have to offer, or a variety of activities that guests may be interested in experiencing while at the destination. This findi ng substantiates the exploration benefit of sales promotions identified by promotions support variety seeking by minimizing the financial risk of trying a new product. This seems to be true for hotel flash sales as well: the flash sales websites offer a variety of hotel deals in different destinations at deeply discounted prices. This finding pertaining to variety seeking of flash sales customers appears to be in line with th e results of previous studies on response to sales promotions that identified deal prone customers to seek variety (Ailawadi et al., 2001; Martinez & Montaner, 2006). Revenue Generation The next concern expressed by hotel managers was regarding the revenu e generating ability of flash sales customers. Participants of the first study of this dissertation suggested that customers who were attracted to their properties by flash sales websites did not spend as much as those guests who came via other distributio n channels. However, the current research found that flash sales customers make higher term profitability via additional expenditures in non room operating departments. Again, the assumption of flash sales custome rs being
145 tighter in their spendings than customers who come from other channels was not realized. On the contrary, flash sales customers generated higher revenues during their hotel stay. ompared across midscale and upscale hotel classes, flash sales customers demonstrate significantly higher spendings only in the midscale hotel class. Flash sales customers do not differ from other customers at the upscale level, and group comparisons were not possible at the economy and luxury levels due to insufficient sample sizes. This finding should tell hotel managers that flash sales customers generate at least the same, if not higher, auxiliary revenues via patronizing non room operating departments as other customers. These results become especially important since hotel managers indicated that (1979) posited that leisure travelers may come in two forms: that is the little spenders and the big spenders. Similarly, Jang et al. (2001) differentiated between heavy, medium and light spenders. Since flash sales customers demonstrated higher expenditures than other customers, the former are not likely to represent the littl e (Pizam & Reichel, 1979) or light (Jang et al., 2001) spender categories. Therefore, departments at the hotel and the total revenue. Future Behavioral Intentions Another concern that hotel managers expressed regarding flash sales customers customers. This research examined future behavioral intentions of hotel customers from
146 two perspecti ves: likelihood of revisiting the hotel, and recommending the hotel to others. It was found that flash sales and other customers are equally likely to revisit the hotel, however, flash sales customers are less likely to come back at a regular price. This m ay be so due to the fact that flash sales customers are aware of the available promotions. Therefore, hotel managers may heed this finding as a caution and may decide to directly market those customers offering them special deals that may be accessed only the flash sales customer to return to the property at a discounted rate but will allow the hotel to avoid losing potential revenue in commission fees that would be paid to the flash sal es site. In addition, this research revealed that flash sales and other customers are equally likely to recommend a hotel to others. The likelihood of recommending a hotel to others had a positive relationship with customer satisfaction. Therefore, if sat isfied, both customer groups are equally likely to recommend the hotel to their friends and relatives. Limitations Like any study, this one is not without limitations. One of the limitations of this study is the use of online data collection mode. Online surveys have been criticized for accessibility issues (e.g. respondents must have an Internet access and knowledge to surf the web), low response rates, and non random sampling (Creswell, 2011; Ritter & Sue, 2007; Van Selm & Jankowski, 2006) that may exhib it threats to the reliability and validity of the results. In application to this research, the choice of online data collection supported the purpose of the study which involved reaching flash sales customers who purchase flash sales deals online. Howeve r, at the same time, the purpose of this study was to
147 skewed in its composition toward the c ustomers who favor online booking methods. Therefore, it may be suggested that future studies on hotel flash sales proneness may also consider alternative modes of data collection. Another limitation of this study may be hidden in the use of Amazon Mechani cal platform for the online data collection. Although is looked at as a viable means to access reliable samples (Buhrmester et al., 2011); in application to this research, it is also important to note, that Amazon Mechanical participants differed from community sample on some attitudes about money (Goodman et al., 2013). According to Goodman et al., participants valued money more than time, and measured lower on the tightwad spendthrift scale co mpared to their community counterparts. Since this study looks at the response to price promotions, and customer value, such difference may have important implications for this research with regard to validity of two group assessment and reliability of the results if the study would be repeated with another sample. income closely follows the census data of annual income distribution. However, to ensure a reliable measurement of t he customer profiling traits, and proper group comparison of flash sales and other customers, future studies may utilize other platforms for data collection such as other panel and marketing companies that allow for random sampling from the US consumers da tabase.
148 Given the exploratory nature of the study, the validity of the psychographic trait measurement in application to flash sales customers was not assessed. Therefore, further research is warranted to strengthen the dimensional structure of the promoti on prone customer psychographic traits in the context of hotel flash sales promotions. An exploratory analysis of the data collected for this study seems to support the psychographic trait factors. However, a confirmatory analysis would be recommended to v alidate the results. As for the overall reliability level of this study, the psychographic scales adopted from sales promotions literature exhibited high reliability coefficients (above 0.7) that were similar with the reliability levels reported in previou s research (e.g., Martinez & Montaner, 2006). Establishing reliability serves as a prerequisite for the validity of the instrument (Creswell, 2011). The validity of the instrument was also supported through the use of psychographic scales that have been va lidated in previous research on profiling promotion prone customers. In order to further increase reliability and validity of the study, several approaches may be suggested. Given the findings of this study, and the differences that were revealed between f lash sales and other customers at specific hotel classifications (e.g. additional expenditures differed only at the midscale hotel level), it may be suggested to conduct follow up studies at every service level in order to validate the results, and measure the reliability of the scales in different samples. Furthermore, the external validity of this study may be improved by validating the adapted scales in application to other hospitality services, e.g., restaurants, entertainment, etc. This way, the scales that
149 were originally developed for tangible products may be further validated for the consumption of intangible products (i.e. services). Closing Remarks In summary, the findings of this study suggest that flash sales customers may be develops long ter m relationship with the property through revisiting the hotel and recommending it to others. The results of this research have indicated that flash sales customers do not seem to be exclusively deal seekers looking for the lowest priced hotel room availabl e on the market. In fact, flash sales customers may actually spend more at the hotel than customers who come through other distribution channels. This finding was supported by some psychographic traits, such as equal price consciousness with other consume rs, and higher quality consciousness of flash sales customers. Furthermore, flash sales customers express the same level of likelihood of coming back to the hotel, and recommending it to others. Such finding does not provide any ground to think that other customers are superior to flash sales customers with regard to generating repeat business for the property. Therefore, current research outlines the psychographic portrait of flash sales customers who spend as much or even more than other hotel customers, and exhibit the same likelihood of becoming repeat customers. When evaluating flash sales customers based on not only a single economic transaction (i.e., discounted purchase on flash sales websites), but over the entire life cycle, the revenue contribut ion and future behavioral intentions taken together may
150 produce a customer with high economic lifetime value. Therefore, flash sales customers term (e.g., auxiliary revenue), and long term (future behavioral intentions) contribution. As part of a revenue management strategy, past flash sales customers may become a aligns with the goals of revenue management of selling the right product to the right customer at the right price, and at the right time. In this situation, the right product is the unsold room inventory, the right price is the discounted rate, the right time is the low season , and past flash sales customers may be considered the right customer because of their much needed economic contribution.
151 Table 3 1. Pre pilot construct reliability Construct Quality c onsciousness 0 .866 Impulsiveness 0 .783 Price c on sciousness 0 .483 Financial c onstraints 0 .817 Consumer spending self control 0 .777 Market m avenism 0 .847 Variety seeking 0 .783 Innovativeness 0 .796 Brand l oyalty 0 .895 Table 3 2. Pilot construct reliability Construct Quality c ons ciousness 0.843 Impulsiveness 0.738 Price c onsciousness 0.749 Financial c onstraints 0.707 Spending s elf c ontrol 0.779 Market m aven 0.802 Variety seeking 0.871 Innovativeness 0.817 Loyalty 0.805
152 Table 3 3. Respondent demographic characterist ics Demographic % Demographic % Gender Children Male 50.6 Yes 34.7 Female 49.4 No 65.3 Marital Status Education Single 33.0 Less than high s chool 0.6 In a relationship 15.1 High s chool / GED 10.7 Living with a partner 11.7 Some c ollege 27.1 Married 35.5 2 year c ollege d egree 11.9 Divorced 3.6 4 year college d egree 39.3 Separated 0.3 d egree 7.6 Widowed 0.8 Doctoral/p rofessional d egree 2.8 Income Age under $20,000 14.8 18 24 24.1 20,000 29,999 10.1 25 29 28.0 3 0,000 39,999 18.4 30 34 14.8 40,000 49,999 12.0 35 39 11.8 50,000 59,999 11.2 40 44 7.3 60,000 69,999 5.6 45 49 4.5 70,000 79,999 5.3 50 54 5.0 80,000 89,999 6.1 55 59 2.0 90,000 99,999 4.2 60 64 1.7 100,000 149,999 7.3 65 69 0.6 150,000 an d above 2.0 70 and above 0.3 Prefer not to answer 3.1
153 Table 3 4 . Respondent demographics by distribution channel Demographic Flash Sales Other Gender Male 49.0 51.2 Female 51.0 48.8 Education Less than h igh s chool 1.0 0.4 High s chool / GED 9.1 11.4 Some c ollege 22.2 29.0 2 year c ollege d egree 13.1 11.4 4 year c ollege d egree 45.5 36.9 d egree 7.1 7.8 Doctoral/p rofessional d egree 2.0 3.2 Age 18 24 17.0 26.9 25 29 27.0 28.4 30 34 20.0 12.8 35 39 19.0 8.9 40 44 5.0 8.2 45 49 4.0 4.7 50 54 5.0 5.1 55 59 1.0 2.3 60 64 2.0 1.6 65 69 0.0 0.8 70 and above 0.0 0.4 Marital Status Single 31.0 33.7 In a relationship 14.0 15.5 Living with a partner 15.0 10.5 Married 33.0 36.4 Divorced 5.0 3.1 Separated 1.0 0.0 W idowed 1.0 0.8 Children Yes 38.4 33.3 No 61.6 66.7 Income under $20,000 12.0 15.9 20,000 29,999 8.0 10.9 30,000 39,999 21.1 17.4 40,000 49,999 9.0 13.2 50,000 59,999 14.0 10.1 60,000 69,999 9.0 4.3 70,000 79,999 3.0 6.2 80,000 89,999 9 .0 5.0 90,000 99,999 6.0 3.5 100,000 149,999 4.0 8.6 150,000 and above 2.0 1.9 Prefer not to answer 3.0 3.1
154 Table 3 5. Rotated factor loadings Factor/Item Factor loading Price c I try to get the best price for hotels 0 .731 I compare prices before choosing a hotel 0.703 I check prices when buying additional products at hotels (e.g. breakfast items, gift store items, etc.) 0.695 I am willing to put effort into finding the best hotel price 0.677 I usually sort the hote l search results by price 0.647 Variety s When revisiting a destination, I prefer different hotel brands that I did not try before 0.843 I prefer different hotel brands wherever I go 0.820 Among two suitable options, I'd prefer a hotel brand that I did not try before 0.786 I enjoy staying in different hotel brands for the sake of comparison 0.740 Consumer spending self c I lose track of my spending 0.835 I usually overspend my budget 0.817 I spend responsibly 0.785 I prefer saving my money 0.542 Qu ality c It is important for me to book a room at a high end / high quality hotel 0.853 I will not give up high quality for a lower price 0.825 I always book one of the best hotels in the destination 0.809 Market m 23) My friends think of me as a good source of price information when they need to book a hotel 0.849 I am considered somewhat of an expert when it comes to knowing the prices of hotels 0.791 I enjoy telling people how much they might expect to pay for different hotels in different locations 0.752 I often book hotels without thinking too much 0.792 0.792 I often book the first hotel I like 0.787 I seek new products and services 0.766 I am inter ested in new products and services 0.782 I try new products or services as soon as they come out 0.728 Financial c My household usually has problems saving funds for vacations 0.827 My household discretionary budget is usually t ight 0.793 I always have extra money available for vacations 0.717
155 CHAPTER 4 HOTEL FLASH SALES RESEARCH : WHERE DO WE GO FROM HERE? Background of the Study Based on the fast adoption of flash sales websites for hotel room inventory distribution, and the lack of empirical literature that would provide understanding of the dissertation investigated and assessed the benefits and drawbacks of hotel inventory distribution vi a flash sales websites. The exploratory nature of the first two studies attempted to provide an understanding of the flash sales phenomenon from both industry and customer perspectives. The results of these two studies have provided foundational knowledge with inventory management, revenue management, brand marketing, and customer relationships. Additionally, the socio demographic and psychographic profiles for flash sales customers were built. However, the findings have raised some important research questions that require further investigation in order to provide a deeper understanding of The overarchin g goal to further research is to quantify the flash sales evaluation framework from study1 into a measurable instrument that might be used by both researchers and industry professionals. Given this, the current study undertakes an attempt to develop furthe r research directions that would move hotel flash sales research towards achieving this goal. The current conceptual study builds on the results of the first two studies by providing research directions for hotel flash sales research.
156 Purpose of the Study Given that the literature on hotel flash sales is young and seemingly naÃ¯ve, the purpose of study 3 is to develop research directions that would enable further understanding of the flash sales phenomenon. To the best of knowledge, there is only one descrip tive study that has been published in lodging literature regarding the use of flash sales websites in the hotel industry. Hence, the purpose of this conceptual study will be achieved via a critical evaluation of the research results that were generated fro m the first two studies of this dissertation, as well as relevant literature support from related disciplines. This study will provide theoretical directions that may be used as confirmatory frameworks for studies 1 and 2, as well as it will provide resea rch directions that are carved from the results and limitations of the first two studies. Additionally, some of the reviewed the results from the first two studies. The indust obtained during a professional seminar at the international Hospitality Information Technology Association (iHITA). Five professionals who participated in the discussion held executive positions (e.g., vice president, chief o perating officer) in hospitality technology companies, and were well aware of the uses and concerns regarding flash sales. Therefore, study 3 becomes an infusion of theoretical direction along with pragmatic industry application of the research findings. B uilding on the preliminary results of the flash sales evaluation framework from study 1, study 3 will propose research directions that require further exploration in order to substantiate the evaluation framework for hotel flash sales. The identified benef its and drawbacks, which serve as the foundation for this framework, are considered in the
157 development of each research direction. These benefits and drawbacks pertain to: inventory management, revenue management, and brand management (which managers all assumed would happen) through the acquisition of new customers through the flash sales channel. This means that if flash sales websites are not successful in attracting customers to a hotel, the benefits of selling inventory, generating revenues, and promo ting a brand cannot be realized. Moreover, in a broader scope of revenue management, hotels should be able to sell the right product, to the right customer, at the right time, and at the right price. When developing and implementing flash sales promotions, hotel managers are able to develop the right product through packaging those components that they find appropriate, set the right price that would be attractive to customers and profitable for the hotel, and forecast the right time when the property would benefit from flash sales promotions. However, attracting the right customer may present more challenges for hotel managers than setting all other revenue management attributes right. Moreover, hotels if not properly mitigated. Therefore, this conceptual study focuses on developing customer related questions in order to examine the future research directions that emerged as an outcome of the first two studies, and provides theoretical foundatio ns that may assist in approaching these research questions. Research Problem The research problem is grounded in the need to empirically test and confirm the findings from studies 1 and 2 in order to eventually quantify the flash sales evaluation framework . The research findings from the first two studies provided a foundational
158 However, further advancement of this research stream is needed. Since this dissertation provide d some of the pioneering works in the area of hotel flash sales, study 3 as a concluding piece in this dissertation takes a lead in outlining future research directions to provide development guidance for the literature, and to build a sound understanding of the flash sales phenomenon and its impact on the hotel industry. Given the exploratory nature of the research inquiries, additional research is required to confirm and validate the findings from both the qualitative study regarding the benefits and draw backs of the hotel inventory distribution via flash sales websites, as well as the profile of flash sales customers developed in study 2. The results of the exploratory studies provided foundational information required to understand the hotel flash sales phenomenon. However, the contribution of these findings may remain minimal until the results are confirmed. The lack of empirical attention pertaining to this topic presents a challenge for future researchers who may endeavor to confirm the previous findin gs. Therefore, this study paves the way towards building an empirical platform with which the forthcoming research directions may enhance flash sales research. Significance of the Study The significance of this study is establishing firm grounds for the r esearch stream of hotel flash sales to develop. This conceptual study will use the gaps that emerged as an outcome of the first two research studies to further enhance the understanding of rch directions that are forwarded in this study will contribute to academic literature by establishing topics
159 that require exploration in order to substantiate the findings of the first two studies. Academicians that are interested in developing research in this area may benefit from the results of this study. Proposed Research Directions The current study forwards several research directions that may require further exploration based on the outcomes of the first two studies and feedback from industry prof essionals. Table 4 1 presents a snapshot of the research directions that will be further developed in this study. Each research direction is supported by a corresponding research purpose and research problem. The first two research directions regarding the incremental demand and new customer acquisition are pragmatic in nature. It may behoove researchers to find theoretical frameworks that could support the advanced development of these questions. The rest of the questions are grounded in the relevant promo tional and consumer behavior literature, and accompanied by proposed measurement items. The following section presents relevant literature that supports the research directions and develops each direction in more detail. Effect of Promotions on Demand The attracted a lot of attention in academic literature (e.g., Ailawadi et al., 2001, 2009, Campo & YagÃ¼e, 2008; Christou, 2011; Leeflang & ParreÃ±o Selva, 2012; Nijs, Dekimpe, Steenkamps, & Hanssens, 2001; Raju, 1992; Wakefield & Barnes, 199 7 ). Demand generated by sales promotions presents one of the important areas of research for Selva, 2012; Nijs et al., 2001). T he extant academic literature documented a short term
160 increase in demand as an outcome of sales promotions (Enz, Canina, & Lomanno, 2009; Leeflang & ParreÃ±o Selva, 2012; Nijs et al., 2001). Research of consumer products (e.g., food products, beverages, non food products, supplies, etc.) demonstrated that promotions significantly expand demand for a promoted category in about half of all cases (46% in Leeflang & ParreÃ±o Selva, 2012; s also documented in the hotel industry. For example, hotels discounting their rates by 20 30% achieved about 15% higher occupancies than their direct competitors (Enz et al., 2009). In application to flash sales, the research findings cited above substant iate the inventory management benefits (e.g., fast sales, selling distressed inventory, and increasing occupancy) that were identified in the first study of this dissertation. However, a question remains whether an increased demand for a subject property u sing flash sales promotions generates an incremental demand on the market, or if the existing demand is shifted from the other properties. In other words, this question asks if deeply discounted rates on flash sales websites stimulate travel for those cust omers who would not travel otherwise; or, if they supply deeply discounted rates for those customers who would travel anyway. Therefore, the following research question may be stated: RQ1. Do flash sales websites generate statistically significant incremen tal hotel demand or just shift the demand that has already been on the market? domain using store scanner data (Leeflang & ParreÃ±o Selva, 2012; Nijs et al., 2001). However, due to a different nature of the hotel industry, where a purchase does not usually happen in a store, applying a similar methodology does not seem viable for the
161 hotel context. Therefore, an alternative methodological solution for the hotel context would be requir ed. retailer intermediary sc anner data (that was traditionally used to study promotions impact on demand for goods) is usually received from the retailer (a store). In the hotel distribution, a similar role would be played by an intermediary. Therefore, data similar to store scanner data may be obtained from the intermediaries, e.g., online travel agencies, such as Expedia . As an intermediary, online travel agency would usually sell numerous hotels in different destinations, have information available about the supply and demand, as well as promotional offerings for every hotel. revenues) was studied using Smith Travel Research (STR) data (e.g., Enz et al., 2009). STR is a business intelligence agency that provid es benchmarking data for the lodging industry (STR, 2014). STR collects supply, demand, and performance data for thousands of hotels around the world. Such database presents additional opportunities for researchers to investigate an impact of deep discount s distributed via flash sales websites on demand in the hotel industry. However, researchers should execute caution when trying to answer a question about incremental demand from the industry data (i.e., intermediary sales data, or STR data). Even though s uch data may allow tracking the demand for every property in response to the available promotions, and identifying an increase or decrease in market
162 share, the answer about whether this demand is incremental to the market or not may not be readily availabl e. Therefore, such a question may also be approached from the customer perspective. When investigating such research question from the customer perspective, the intentions to reserve a hotel room prior to finding a flash sales promotion should be measured. Such measurement would allow managers to differentiate between those customers who were searching for a hotel and found a discounted price on a flash sales website from those customers who were moved to book a hotel room by the special deal. Investigating the customer decision making process may allow researchers to answer the question of incremental demand. Additionally, a distance variable may be introduced to further understand the impact of flash sales discounts on generating incremental demand. When it comes to a hotel stay, it may be suggested that shorter travel distances may stimulate more incremental demand. For example, a customer from Gainesville, FL may be more likely to have an unplanned trip to Orlando, FL rather than a customer from New York , NY. Effect of Promotions on Attracting New Customers A question that is closely related to the question of incremental demand is a question of new customer acquisition. In other words, the concern is whether flash sales are able to truly reach new custom ers, or do flash sales allow existing customers to purchase at a deeply discounted rate. Such a scenario would diminish the benefit of In the goods literature, it has been d ocumented that promotions may impact both loyal and non loyal customers (Ailawadi et al., 2006; Raju, 1992). Such impacts suggest non loyal customers to demonstrate switching behavior under the influence of deep
163 discounts. On the other hand, loyal customer s may increase their stockpiling and consumption behaviors. However, such behaviors may not be applicable to the consumption of services due to their intangibility and inseparability of production and consumption (Zeithaml et al., 2006).This means that hot el customers cannot purchase a large quantity of hotel rooms in advance and stockpile them for later use, since every hotel stay requires careful planning and specifying a consumption period (i.e., the dates of the hotel stay). Understanding whether fla sh sales attract new customers may have important distribution channel. Based on the flash sales evaluation framework developed in study 1, hotel managers expect to attra who would stay at the hotel even when paying the rack rate. Consequently, distributing flash sales deals to existing and l oyal customers may deteriorate the potential benefits of customer acquisition and incremental revenue generation. Therefore, the following research question may be studied in future research: RQ2. Do flash sales generate new customers for hotels or allow e xisting customers to purchase the product with a deep discount? A measurement for this research question may be obtained from a consumer survey. For example, customers may be asked whether the hotel stay that they purchased with the flash sales promotion was their first time in that particular hotel. A positive answer to this question would mean that flash sales attracted a new customer to the hotel. However, a negative answer would mean that flash sales failed to acquire a
164 new customer for the property, i nstead selling a deeply discounted room to an existing customer. Promotion Prone vs. Value Conscious Customers The findings from the first two studies have raised important questions about the type of consumer that is attracted to hotels with flash sales. First, in study 1, hotel who constantly seek deals, have minimal expenditures in non room operating departments, and do not generate repeat business. However, the findings of study 2 revealed that these customers are not statistically different from other customers with regard to price consciousness, likelihood of revisiting the hotel, and r ecommending it to others. Moreover, flash sales customers even appear to be more quality conscious than other customers, and seem to spend more money in non room operating departments while onsite than other customers. Such contradictory findings warrant a need for further research regarding the flash sales customer profile. have been defined in the literature as those who respond to promotions (e.g., coupons), and base their purchase decisions on promotional information (Lichtenstein et al., 1990; Martinez & Montaner, 2006). However, when flash sales customers were compared to other customers on pric e consciousness, they measured in accordance to the other group. Therefore, this finding suggests that promotional price may not be the only factor that guided flash sales customers to complete the purchase.
165 Moreover, the results of study 2 also revealed t hat flash sales customers might be quality conscious and variety seeking. Consequently, this means that flash sales customers are likely to complete a purchase because they are in search of a high quality product that they have not tried before. Yet, regar dless of these characteristics, flash sales customers still seem to respond to flash sales promotions. Therefore, the behavior of purchasing flash sales promotions may be explained by factors other than promotion proneness. Based on the acquisition transa ction utility theory, Lichtenstein et al. (1990) suggested that customers who use promotions are not necessarily promotion prone customers; instead, they may be value conscious customers. As defined by Lichtenstein reflecting on the results from study 2, which demonstrated that flash sales customers are just as price conscious as other customers, but more concerned about quality, the concept of value consciousness appears t o be relevant. Thus, value consciousness sales promotions. The acquisition transaction utility theory describes the utility of every purchase via its acquisition and transact ion utilities (Thaler, 1985; 2008). This means that the acquisition utility depends on the perceived value of the purchase, whereas the transaction utility depends on the financial terms or the perception of the deal (Lichtenstein et al., 1990; Thaler, 198 5; 2008). In the case of hotel flash sales, the acquisition transaction utility theory may provide future research directions that may determine whether flash sales customers are promotion prone or value conscious.
166 As previously mentioned, the findings of study 2 revealed that flash sales customers are just as concerned about prices as other customers, but possess a more significant concern for the quality level of products. From the interplay between price and quality components, it seems reasonable that flash sales customers are more likely to be value conscious as opposed to promotion prone customers (Lichtenstein et al., 1990). Therefore, the following proposition is proposed as a direction for flash sales research: Proposition 1. Flash sales customers are significantly more likely to be value conscious customers as opposed to promotion prone customers. In order to investigate this proposition, it is suggested that the measurement scales of promotion prone and value conscious customers be adopted from the existing literature in order to fit within the context of hotel flash sales. Tables 4 2 and 4 3 present measurement items for value consciousness and promotion proneness respectively. The original items and literature sources where they were derived fr om accompany all items. Extending Profile of Flash Sales Customers The results of study 2 shed some light on the profile of flash sales customers, and their psychographic traits. However, the psychographic scales that were used in study 2 were adopted from previous research on consumer goods and require further validation for the context of hotel flash sales. The validity of the adopted instrument may be established using a confirmatory factor analysis through a confirmation of the underlying latent structu re of the constructs derived from the literature. Additionally, future research may also explore additional characteristics of flash sales customers. The results of study 2 suggested that some other factors might
167 improve discriminating power between flash sales and other customers. Therefore, another research direction may explore such additional factors that may further explain the differences between flash sales and other customers. The current study proposes to focus on smart shopper personality trait a s it was found to be strongly correlated with response to price and non price deals (Burton , Lichtenstein, Netemeyer, & Garretson , 1998). Shimp and Kavas (1984) suggested that the smart shopper feeling might be one of the benefits of using coupons. Atkins and Kim (2012) conceptualized smart shoppers may be approached from several different perspectives, including self perception (Burton et al., 1998; Garretson, Fisher, & Burton, 2002), time consciousness (Kleijnen, De Ruyter, & Wetzels, 2007), and convenience seeking (Noble, Griffith, & Adjei, 2006). However, the most recent work of Atkins and Kim (2012) combined time consciousness and convenience seeking into one dimension of time/effort savings. In application to hotel flash sales it may be suggested that booking a hotel on a flash sales websites with a deep discount may stimulate customer feelings. Following the definition of smart shopper by Atkins and Kim (2012) that was websites, customers may receive both hedonic and utilitarian benefits. He donic and utilitarian benefits of sales promotions were discussed in study 2. Hedonic benefits included opportunities for value expression, entertainment, exploration, and self -
168 expression, whereas utilitarian benefits included savings, quality, and conveni ence (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). Therefore, the proposition of flash sales customers feeling smart about their purchases may be in line with the classification of benefits of sales promotions forwarded by Chand on et al. (2000). Moreover, the results of study 2 have provided support for both utilitarian (e.g., quality) and hedonic (e.g., self expression) benefits of flash sales. On one hand, when considering smart shopping from the perspective of self perception, it may be mapped to the self expression benefits of sales promotions (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). On the other hand, when evaluating smart shopping from the perspective of time and effort saving, it may provide convenience benefits to sales promotions (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). Therefore, this research proposes to evaluate self perception and time/effort savings dimensions in order to uncover the potential smart sho pper personality trait in flash sales customers. The following proposition is forwarded as a direction for flash sales research: Proposition 2. Flash sales customers achieve a statistically significant higher smart shopper feeling than other customers. In order to study this proposition, the measurement scales may be adopted from previous consumer behavior and promotions literature. Tables 4 4 and 4 5 present smart shopper self perception and effort/time savings scales, respectively. The original items and literature sources where they were derived from accompany all items. Closing Remarks The third study of this dissertation provided directions for future research in the area of hotel flash sales. These research directions were built on the results of the first
169 two studies, feedback from industry professionals, and relevant literature in sales promotions and consumer behavior areas. All proposed research directions were supported with the relevant literature and corresponding measurement items that would en able further investigation of each research stream. Therefore, the study provided a conceptual foundation for the development of hotel flash sales research. The proposed research directions included several areas, such as the effect of flash sales promotio ns on demand and attracting new customers, promotion proneness vs. value consciousness of flash sales customers, and extension of flash sales answering the corresponding resear ch questions, or testing the forwarded propositions, would allow researchers to further validate the findings of studies 1 and 2 of this dissertation. This means that every proposed research direction was linked back to the flash sales evaluation framework from study 1 in order to indicate where each research direction fits in a larger picture of the hotel flash sales experience. The first research direction proposed in this study is concerned with the ability of flash sales websites to generate increment al demand on the hotel market during the demand has been highlighted in previous literature on sales promotions in distribution of goods (Leeflang & ParreÃ±o Selva, 2012; Ni js et al., 2001) as well as the sale of hotel room nights (Christou, 2011). Answering this research question may result in Understanding flash sales ability to generat e incremental demand may be important for hotel managers in order to understand the market dynamics that may be
170 propagated due to the use of flash sales. If future research finds that flash sales just shift the existing demand from one hotel to another dur successful in shifting the existing demand, i t is possible that the next time market demand decreases that competitors may be faster than the subject hotel to engage in flash sales distribution. Consequently, it is possible that demand may be shifted toward competing properties. This type of behav ior may enhance the likelihood for managers to engage in a price cutting strategy in order to protect their existing market share. If price cutting becomes a cyclical behavior for a market, it is likely that the price war will result in decreased profits f or the industry at large. However, if future findings demonstrate that flash sales websites are capable of hospitality industry may benefit from using flash sales websites . From the hotel support the benefit of increasing occupancy, generating incremental revenue, and acquiring new customers by means of increasing market share rather tha customers from other hotels, as well as having customers stolen from their own property. From the perspective of the larger hospitality industry, such findings may mean that by stimulating the demand for hotels, flash sales may also stimulate incremental demand for travel (e.g., gas, airfare), restaurants, attractions, and entertainment related venues.
171 As it was suggested earlier in the manuscript, flash sales may be found to be more efficient in generating demand for short distance travel (e.g ., from Gainesville, FL to Orlando, FL). In this case, the benefit to the hospitality industry may result from the travel, food, entertainment, etc.). Thus, flash sa les may become an important research direction not only within the scope of the hotel industry, but in other hospitality sub industry sectors as well. The second research direction that was identified pertains to the ability of flash sales promotions to at tract new customers. The question of new customer acquisition is important from the perspective of substantiating the benefit of inducing product trial that was attributed to flash sales promotions by the hotel managers in study 1. Answering this question may also bring important implications for hotel managers. periods (study 1). This benefit is based on a potential opportunity that requires a sequential service based process whereupon the flash sales customer redeems the flash sales voucher, stays in the hotel, is satisfied with the experience, and then is later converted into a repeat guest. However, such benefit would be diminished if deeply discounted flash sales promotions would be offered to existing repeat customers of the hotel. Consequently, it becomes important for hotels to separate these two market segments, and target new customers with fl ash sales promotions, while developing direct marketing campaigns for existing customers that have a unique promotional value
172 to reward the returning guests. Such an approach may be the most beneficial for hotels s promotions may attract new customers who would not know about the hotel otherwise, and direct marketing may bring repeat distribution expense. Next, the third research questio n aims to investigate whether flash sales customers are promotion prone or value conscious (Lichtenstein et al., 1990). This research question addresses a concern of hotel managers expressed in study 1 that flash sales customers are deal seekers, who are c onstantly looking for a greater discount. Synthesizing the results of the first two studies of the dissertation, one may characteristics of flash sales customers identifi ed in study 2. The findings of study 2 revealed that flash sales customers are more quality conscious and variety seeking than other hotel customers, therefore, leading to a suggestion that they may be responding to flash sales promotions due to their valu e consciousness as opposed to their promotion proneness. Confirming the anticipated value conscious profile of flash sales customers may customers, and substantiate the fin dings of study 2 pertaining to the psychographic differences between flash sales and other customers. Such findings of this proposed future research direction might suggest to hotel managers that flash sales customers revenue management initiatives. Additionally, the value consciousness trait of flash sales customers may provide additional hints for
173 managers regarding how to best design flash sales deals. This would mean that managers should construct flash sales deals with the emphasis not only on the monetary aspect of the deal, but also by highlighting the quality of the product that is offered through the promotion. And, finally, the fourth research direction suggests conducting a confirmatory analysis of the psycho graphic dimension of the flash sales customers, along with exploration of additional possible psychographic characteristics. The psychographic characteristics of flash sales customers were investigated in study 2 in accordance with established benefits and drawbacks of sales promotions (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & Montaner, 2006). However, the findings of study 2 also revealed that the discriminating power between flash sales and other customers may be increased by including addi tional factors. Current research proposes to explore the smart shopper characteristic as one of the potential traits for future analysis. The smart shopping attribute seems to be suitable to the dimensions of hedonic benefits, referenced in study 2, by pro viding an opportunity for increased self perception as a smart shopper, as well as for utilitarian benefits by contributing towards the convenience (e.g., time and effort) of flash sales promotions (Ailawadi et al., 2001; Chandon et al., 2000; Martinez & M ontaner, 2006). Also, smart shopper personality traits of flash sales customers appear to be in line with value consciousness, if both propositions 1 and 2 are supported. In summary, this study paved the way for individuals who may be interested in further developing flash sales research. Given the infantile state of the literature that pertains to hotel flash sales, the research directions forwarded in this study provide a
174 needed foundation to move this research field forward. This conceptual paper presen ted a tiered level of alternative perspectives pertaining to the benefits and drawbacks of flash sales; that is, an industry perspective, consumer based usage perspective, and academic perspectives were considered when drafting the future research directio ns. With the rapid adoption of flash sales websites as inventory distribution channels in the lodging industry and the lack of formative information that could foresee where this channel may lead the industry with regard to the impact on the overall reven ue management system, this research is intended to matter in business practice. Consequently, scholars who follow the path of these research directions will contribute to bridging the gap between empirical research and pragmatic application. Without this research, it is possible that flash sales drawbacks may occur too frequently for hotel managers and the advantages of this distribution channel may too frequently escape. Therefore, the author of this dissertation endeavors to lead this important research initiative by providing a platform for its development in moving towards further understanding of hotel flash sales impacts, and the quantifiable flash sales evaluation framework.
175 Table 4 1. Proposed future research directions Flash sales benefit s and drawbacks* Research direction Research purpose Research p roblem Benefit: Customer acquisition Explore whether flash sales generate incremental demand or just shift existing demand. To substantiate the finding of study 1 regarding the customer acquisition benefit of flash sales. Uncertainty about flash sales ability to generate incremental demand in the market during Benefit: Inducing trial Explore whether flash sales generate new customers or just allow existing customers to purchase the product with a deep discount. To substantiate the finding of study 1 regarding the inducing trial benefit of flash sales. Uncertainty about flash sales ability to reach new customers, and not those guests who have already visited the hotel before. Drawba ck: Attracting Determine whether flash sales customers are promotion prone or value conscious There is a need to further understand the profile of the customers who are attracted to the hotels by flash sales websites Controversy between t claim in study 1 that flash sales customers are customers, and the results of study 2 that suggest that flash sales customers may be for hotels. Explore additional characteristics of flash sales customers, e.g. smar t shopper personality trait There is a need to further understand the profile of the customers who are attracted to the hotels by flash sales websites Increase the discriminating power between flash sales and other customers. *Selected flash sales benefit s and drawbacks are derived from the flash sales evaluation framework that was developed in study1
176 Table 4 2. Measurement items for value consciousness construct # Adapted i tems Original i tems Source 1 When planning to stay in a hotel, I am very conce rned about low prices, but I am equally concerned about quality I am very concerned about low prices, but I am equally concerned about product quality Lichtenstein, D. R., Netemeyer, R. G., & Burton, S. (1990). 2 I compare prices of different hotel brand s to be sure I get the best value for the money When grocery shopping, I compare prices of different brands to be sure I get the best value for the money 3 When planning to stay in a hotel, I always try to maximize the quality I get for the money I spent . When purchasing a product, I always try to maximize the quality I get for the money I spent. 4 When planning to stay in a hotel, I like to be sure that worth. When purchasing a product, I like to be sure that I am getting my mon 5 I generally shop around for lower prices, but the hotel still must meet certain quality requirements before I make a booking. I generally shop around for lower prices on products, but they still must meet certain quality requirements before I buy them. 6 N/A When I shop, I usually brands I normally buy. 7 N/A I always check prices at the grocery store to be sure I get the best value for the money I spent.
177 Table 4 3. Measurement items for promotion proneness construct # Adapted i tems Original i tems Source 1 Finding special hotel deals makes me feel good Redeeming coupons makes me feel good Lichtenstein, D. R., Netemeyer, R. G., & Burton, S. (1990). Items with * also appeared in Lim, C. M., Kim, Y. K., & Runyan, R. (2013). 2 N/A I enjoy clipping coupons out of the newspaper 3 When I book a hotel with a special deal, I feel that I am getting a good deal When I use coupons, I feel that I am getting a good deal 4 I enjoy finding speci al deals for hotel rooms, regardless of the amount I save by doing so. I enjoy using coupons, regardless of the amount I save by doing so. 5* I have favorite hotel brands, but most of the time I book the brand I have a special deal for. I have favorite b rands, but most of the time I buy the brand I have a coupon for. 6* I am more likely to book a room at a hotel for which I find a special deal I am more likely to buy brands for which I have a coupon 7 Special deals have caused me to stay at hotels tha t I normally would not book Coupons have caused me to buy products I normally would not buy. Table 4 4. Measurement items for smart shopper self perception construct # Adapted i tems Original i tems Source 1 Making smart hotel purchases makes me feel goo d about myself Making smart purchases makes me feel good about myself Burton et al., 1998 Garretson et al., 2002 2 When I shop for a hotel room, I take a lot of pride in making smart purchases When I go shopping, I take a lot of pride in making smart purc hases 3 When I shop smartly, I feel like a winner When I shop smartly, I feel like a winner
178 Table 4 5. Measurement items for effort/time savings construct # Adapted i tems Original i tems Source 1 Booking a hotel on flash sales was convenient for me Making this purchase was convenient for me Atkins & Kim, 2012 2 Booking a hotel on flash sales was hassle free Making this purchase was not a hassle 3 I did not spend extra effort to book a hotel room on flash sales I did not spend extra effort on this purchase 4 Booking a hotel on flash sales allowed me to use my time wisely In making this purchase, I used my time wisely 5 hotel room on flash sales purchase 6 I was able to make a hotel booking on flash sales quickly I was able to make this purchase quickly
179 CHAPTER 5 CONCLUSIONS The current study investigated hotel flash sales from two perspectives. First, it aimed at providing hotel managers with an understanding of the potential b enefits and drawbacks associated with using flash sales websites as room inventory distribution channels. Second, this study attempted to provide platform research from which other scholars may build from to enhance the available literature on flash sales. The evaluation of hotel flash sales benefits and drawbacks was conducted from the perspective of hotel managers, and hotel customers. The results of this dissertation revealed that hotel managers use flash sales websites as an inventory distribution chann were described as periods of low demand, when hotels faced high volumes of unsold perishable inventory. Hotel managers who participated in the study suggested that adoption of flash sales websites as an inventor y distribution channel comes with both benefits and drawbacks . Such benefits and drawbacks were found to have implications for inventory management, revenue management, brand marketing, customer relationships, and operational challenges. From the perspecti ve of inventory management, all hotel managers highl ighted the power of flash sales o However, hotel managers were split in their opinions about the implications of flash sales for revenue managem ent, brand marketing, customer relationships . On one hand, hotel managers suggested that flash sales may assist hotels in generating revenue attracting new customers. But on the other hand, hotel managers were also concerned
180 that customers who are attracted to the hotel properties via flash sales websites may be deal seekers, who are always looking for a greater discount, and therefore do not In order to further understand potential benefits and drawbacks of using flash sales as hotel room inventory distri bution channel, the second study investigated the Based o n the concern of hotel managers, this study term (e.g., auxiliary revenue), and long term (future behav ioral intentions) contribution. With this in mind, the profiles of flash sales and other leisure customers were compared. The results of the study revealed that hotel flash sales customers are not statistically different from other customers in their price consciousness, intentions to revisit the hotel, and recommend it to others. At the same time, hotel flash sales customers were found to measure higher on market mavenism, quality consciousness, and variety seeking. Also, the results of this research sugge st that flash sales customers spend more money in different non room operating departments during their hotel stay. Ther efore, in comparison with other leisure customers, flash sales customers appear to be equally price conscious, higher paying customers, who exhibit the same likelihood as other customers to revisit the hotel, and recommend it to others. Taken all term profitability due to higher expenditures in different operating departments. Also, from the long term perspective, flash sales customers may present high lifetime value
181 customers, since they are equally likely to revisit the hotel as other customers and , on average , spend more than other customers while on p roperty. Given that, flash sales purposes of this study, and demonstrate that they are equally (if not higher) valuable customers from both short term and long term perspecti ves. It appears that t he results of the second study substantiated the benefits of hotel distribution via flash sales websites with regard to customer acquisition, and revenue management categories . Taken together these two studies provided some guidance t o hotel managers with regard to when to use flash sales websites, and what benefits and drawbacks to expect from room inventory distribution via flash sales websites . These studies also suggested that when used appropriately, flash sales websites may allow hotel managers to reach the right customer, at the right time, and sell them the right product at the right price. However, the two studies of this dissertation would not provide a complete understanding of hotel flash sales phenomenon. Therefore, future research will be needed to substantiate other potential benefits and drawbacks that were uncovered within the scope of this dissertation , and develop the flash sales evaluation framework into a quantifiable instrument . In order to close this gap, and provi de guidance for the development of future academic research, the third study of this dissertation developed future research directions for the area of hotel flash sales . Future research directions for the area of hotel flash sales were developed based on a review of the extant literature available o n the hotel flash sales area, as well as related subject areas, such as hotel electronic distribution, and pertinent promotional
182 and consumer behavior literature . The results of this study revealed several key ar eas pertaining to the hotel flash sales that may require further investigation. These areas include : exploring whether flash sales websites are generating incremental demand, and attracting new customers, determining whether flash sales customers are value conscious or promotion prone, and extending the profile of flash sales customers. In summary, the three studies of this dissertation provided the foundational understanding of the phenomenon of hotel flash sales. The results of the studies may serve both professional and academic audiences. First, hotel managers may make better informed decisions about inclusion or exclusion of flash sales in hotel distribution mix. Second, researchers may use the findings of this dissertation as a platform to develop furt her studies and enhance the literature on flash sales. Therefore, t he significance of this dissertation is in closing the academic literature gap on hotel flash sales, and providing a preliminary determination of the impacts that flash sales may have on ho tel room inventory distribution.
183 APPENDIX A INDUSTRY RESPONDENT PROFILE Table A 1. Industry r espondent profile # Position Industry s egment Description F lash s ales u se 1 General Manager Branded hotel Full service hotel affiliated with an international c hain + 2 Office Manager Independent hotel All inclusive hotel + 3 General Manager Independent hotel All inclusive condo hotel + 4 Marketing Manager Independent hotel Full service, hotel and water resort + 5 Revenue Manager Independent hotel Full servic e, hotel and water resort + 6 Marketing Coordinator Corporate office International resorts and timeshare + 7 General Manager Independent hotel Limited service hotel + 8 General Manager Independent hotel Limited service hotel + 9 General Manager Indepen dent hotel Luxury hotel + 10 General Manager Independent hotel Limited service hotel + 11 General Manager Independent hotel Limited service hotel + 12 Front Office Manager Independent hotel Full service hotel + 13 General Manager Independent hotel Luxu ry boutique hotel + 14 Sales Manager Branded hotel Full service upscale hotel + 15 Director of Marketing Hotel management company A portfolio of branded hotels 16 e Commerce Manager Hotel management company A portfolio of branded and independent hotel s 17 VP of Sales and Marketing Hotel management company A portfolio of branded hotels + 18 VP of Revenue Strategy and e Commerce Hotel management company A portfolio of branded hotels + 19 VP of e Commerce Corporate office Mid size upper upscale bran ded hotels + 20 Director of e Commerce Hotel management company A portfolio of branded hotels + 21 VP of Sales & Revenue Management Hotel management company A portfolio of branded and independent hotels 22 President & CEO Hotel management company A po rtfolio of branded hotels 23 Senior VP of Brand Marketing Corporate office International branded hotels 24 VP of Marketing e marketing company + 25 Director of Sales and Marketing Branded hotel Full service hotel
184 Table A 1. continued # Position Industry s egment Description Flash sales u se 26 VP of Revenue Management Corporate office Resorts + 27 Director of Reservations Independent hotel Full service resort hotel + 28 Senior VP of Sales and Revenue Management Hotel management company A portf olio of branded hotels + 29 Regional Director of Sales Hotel management company A portfolio of branded hotels 30 Revenue Manager Corporate office International branded hotels + 31 Managing Director of Marketing & e Commerce Corporate office Internatio nal branded hotels + 32 VP of Interactive Marketing Owner and management company A portfolio of branded and independent hotels 33 Director of e Commerce and Revenue Management Corporate office International branded hotels 34 Director of Global Onlin e Sales & Distribution Corporate office International branded hotels 35 Revenue Manager Independent hotel Full service & golf club + 36 Regional Director of Sales & Marketing Hotel management company A portfolio of branded and independent hotels + 37 Vice President Marketing Hotel management company A portfolio of branded hotels + 38 Director of Revenue Analysis Corporate office International branded hotels 39 Regional Director of Revenue Hotel management company A portfolio of branded and independ ent hotels + 40 Vice President, Web & Interactive Marketing Corporate office International branded hotels 41 VP of Revenue Management & eMarketing Hotel management company A portfolio of branded and independent hotels + 42 Senior VP Marketing & Sales Hotel management company A portfolio of branded and independent hotels + 43 Senior VP of Revenue Management and e Commerce Hotel management company A portfolio of branded and independent hotels +
185 Table A 1. continued # Position Industry s egment Descript ion Flash sales u se 44 Director of Interactive Marketing Corporate office International branded hotels 45 VP of global online sales and distribution, and VP Global revenue management Corporate office International branded hotels 46 Senior VP of Mark eting Corporate office Domestic branded hotels
186 APPENDIX B CONSUMER QUESTIONNAI RE Dear Study Participant, My name is Ekaterina Berezina, I am a doctoral candidate at the University of Florida. I would like to kindly invite you to participate in a res earch study that is a part of my dissertation. websites that sell discounted travel related products for a limited time. Examples of these flash sales websites include: Participation in this survey is completely voluntary. It will take you about 15 minutes to complete the survey. You will be able to skip the questions that you do not wish to answer. You have the right to withdraw from the study at any time without conse quence. All your answers will be anonymous, will be analyzed statistically and reported as an aggregate in conference presentations and publications. The researchers will not collect any identifying information. Your answers will help travel companies to i mprove their sales strategies. You also will receive a $1 compensation for completing this survey. There are no direct benefits, nor physical, economic, social, psychological, or other risks associated with participation in this study. Questions or concern s about your rights as a research
187 participant may be directed to the IRB02 office, University of Florida, PO Box 112250, Gainesville, FL 32611; (352) 392 0433. If you have any questions, please address them to me or my supervisor at the address below. Ek aterina Berezina, PhD Candidate, University of Florida, Berezina@ufl.edu Kelly J. Semrad, Ph.D., Assistant Professor, Rosen College of Hospitality Management, University of Central Florida, email@example.com By clicking NEXT >> you provide your consent to participate in the study, and will be taken to the survey. QUALIFYING AND FILTE RING QUESTIONS 1. Are you a member of any of the flash sales sites listed below (please check all that apply): a) Groupon b) LivingSocial c) TravelZoo d) Jetsetter e) Groupon Getaways f) LivingSocial Escapes g) Other (please specify) __________________ h) I do not subscribe to any flash sales websites 2. Have you purchased anything from those flash sales websites? a) No g o to question 5 b) Yes If so, how many purchases have you made so far? ____________ 3. Please select types of products/services you have purchased on discount websites (please check all that apply): a) Hotel (or travel package that includes a hotel) If this se lected the respondent will receive the flash sales block of questions. go to question 4, otherwise go to question 5 b) Other travel services (e.g., sightseeing, tours, cruises, transportation, etc.) c) Entertainment and Special Events (concerts, festivals, sho ws, fairs etc.) d) Sports and recreation (golf packages, tennis lessons, game tickets) e) Restaurants/Food and drink/ night clubs
188 f) Casinos g) Theme park tickets h) Other services (e.g., massage, health services, car services, etc.) i) Other products (e.g., clothes, books, home products, etc.) j) Other (please specify) __________________ 4. Have you already stayed in the hotel that you purchased via flash sales websites? a) Yes, I have already stayed in the hotel go to question 6 b) No, I have purchased it, but have not stayed in the hotel yet go to question 5 5. Have you stayed in a hotel within the last 12 months for leisure purposes (e.g., vacation, honeymoon, visiting friends and relatives)? a) No Thank you very much for your time, but your answers indicate that you are not qualifi ed to participate in the study (end of the survey). b) Yes continue to question 6 GENERAL TRAVEL BEHAVIOR BLOCK The next section includes questions about your general travel patterns. 6. In the last 12 months, approximately how many nights in total did you stay in hotels? How many of those nights were for pleasure? 7. Are you a member of any hotel loyalty program(s)? a) Yes. If so, how many ______ b) No 8. Please state your use of the following sources for hotel purchases using the scale below. Please check all t hat apply. Source I use for information search I use for booking Flash sales websites (e.g., Groupon Getaways, LivingSocial Escapes) Last minute deal websites (e.g., Hotels Tonight) Online travel agency website (e.g., Expedia, Travelocity, Pricelin e) Hotel website (e.g. Marriot.com, Hilton.com) Third party review websites (e.g. TripAdvisor.com) Meta search engines that search
189 several websites at the same time (e.g., Kayak) Social media (e.g. Facebook) Travel agent Calling the hotel or hotel brand directly 9. Which channel do you use the MOST when you make reservations for a hotel? a) Flash sales websites (e.g., Groupon Getaways, LivingSocial Escapes, Jetsetter, TravelZoo, etc.) b) Online via last minute deals website (e.g., Hotel Tonigh t) c) d) e) Third party review websites (e.g. TripAdvisor.com) f) Meta search engines that search several websites at the same time (e.g., Kaya k) g) Social media h) Travel agent i) Calling the hotel or brand directly j) Other (please specify) _________________ 10. Please state how often you share your hotel stay experiences using the scale below. Never Rarely Sometimes Often Always On social media (e.g., F acebook, Instagram, Twitter, etc.) On hotel comment cards On hotel online surveys On traveler review websites (e.g., TripAdvisor.com) With friends and relatives PSYCHOGRAPHIC TRAITS BLOCK (for flash sales customers) The fo llowing section will ask you some questions about your hotel preferences and hotel purchases. Please think about your MOST RECENT hotel stay that you purchased via a FLASH SALES website when you answer the following questions. (for other customers) The fo llowing section will ask you some questions about your hotel preferences and hotel purchases. Please think about your MOST RECENT leisure hotel stay when you answer the following questions. 11. Following are some statements about your hotel preferences. Pleas e indicate your level of agreement with these statements using the scale below.
190 1 Strongly disagree 2 3 4 5 6 7 Strongly agree I will not give up high quality for a lower price I always book the best hotels in t he destination It is important for me to book a room at a high end / high quality hotel I often book the first hotel I like "Just do it" describes the way I book hotel rooms I often book hotels without thinking too much 12. Following are some statements about your preferences regarding hotel purchases. Please indicate your level of agreement with these statements using the scale below. 1 Strongly disagree 2 3 4 5 6 7 Strongly agree I compare prices before choosing a ho tel I check prices when buying additional products at hotels (e.g. breakfast items, gift store items, etc.) I try to get the best price for hotels I am willing to put effort into finding the best hotel price I usually sort t he hotel search results by price I always have extra money available for vacations My household discretionary budget is usually tight My household usually has problems saving funds for vacations I prefer saving my money I spend responsibly I usually overspend my budget I lose track of my spending 13. Following are some statements about yourself. Please indicate your level of agreement with these statements using the scale below.
1 91 1 Strongly disagr ee 2 3 4 5 6 7 Strongly agree I am considered somewhat of an expert when it comes to knowing the prices of hotels My friends think of me as a good source of price information when they need to book a hotel I enjoy telling people how much t hey might expect to pay for different hotels in different locations I prefer different hotel brands wherever I go I enjoy staying in different hotel brands for the sake of comparison When revisiting a destination, I prefer different hotel brands that I did not try before Among two suitable options, I'd prefer a hotel brand that I did not try before I am interested in new products and services I try new products or services as soon as they come out I se ek new products and services I prefer one hotel brand over other hotels I am willing to make an effort to search for my favorite hotel brand wherever I am traveling Usually, I care a lot about the hotel brand where I stay R ECENT HOTEL STAY EXPERIENCE 14. (For flash sales customers) Please think about your last hotel stay that you purchased via a flash sales website when you answer the following questions. What flash sales website did you use to make the purchase? a) Groupon Geta ways b) LivingSocial Escapes c) Jetsetter d) TravelZoo e) Other ______________________
192 15. (For other customers) Please think about your most recent hotel stay when you answer the following questions. How did you make a reservation for your most recent hotel stay? a) Via a flash sales websites (e.g., Groupon Getaways, LivingSocial Escapes, Jetsetter, TravelZoo, etc.) b) Online via a last minute deals website (e.g., Hotels Tonight) c) d) .g. Marriott.com, BestWestern.com, etc.) e) Via a travel agent f) Calling the hotel or brand directly g) Other (please specify) _________________ 16. What type of hotel did you stay at? a) Economy (e.g., Motel 6, Super 8, Travelodge, etc.) b) Midscale (e.g., Wingate, Ramada , Quality Inn, Best Western, Hampton Inn, Holiday Inn, Sleep Inn, etc.) c) Upscale (e.g., Hyatt, Marriott, aloft, Courtyard, Crowne Plaza, Residence Inn, SpringHill Suites, etc.) d) Luxury (e.g., Ritz Carlton, JW Marriott, InterContinental, Mandarin oriental) 17. W hat was the purpose of your hotel stay? a) Leisure (vacation, visiting friends and relatives, honeymoon, etc.) b) Business c) Both 18. How many nights did you stay in that hotel? _____________ 19. Did you purchase that hotel as a part of travel package? a) Yes b) No 20. How much d id you pay per night for your hotel room (in US dollars)? _____________ 21. Where is that hotel located? a) In the United States b) Abroad 22. Please select the state where the hotel is located (a drop down list of all US states) 23. Please select the country where the hotel is located (a drop down list of countries ).
193 24. Did you make any additional purchases in that hotel? Indicate how much money you have not stayed in the hotel yet) Hotel restaurant $ Room service $ Hotel bar/lounge $ Mini bar $ Spa $ Golf $ Gift shop $ Umbrella/sunbed rental $ Movie on demand $ Parking or valet $ Internet access $ Other (please specify) $ 25. What services in the nearby area did you use during yo ur hotel stay? Indicate how much money you spent on each purchase in US dollars. (Not displayed if answered have not stayed in the hotel yet) Airlines $ Cruises $ Casinos $ Museums $ Theme parks $ Natural parks $ Beach $ Heritag e and cultural tourism sites (e.g., Mayan ruins) $ Car rentals $ Taxis $ Movie theaters $ Restaurants/ bars / night clubs $ Entertainment and Events (e.g., concerts, festivals, sporting events) $ Shopping $ Other (please specify) $ 26. With the a mounts above, how many people did you pay for? _______ 27. Below are some things that you may or may not want to do regarding this hotel. Using the 1 7 scale below, please indicate your likelihood for each item.
194 1 Very unlikely 2 3 4 5 6 7 Very Likel y Recommend the hotel to others Stay at the same hotel again Stay at the same hotel if only at discount price Stay at the same hotel even if at the regular price 28. What was your overall satisfaction with this hotel? Extreme ly dissatisfied = 1 7= Extremely satisfied (Qualtrics 1 item 7 point scale) 29. Did you post an online review for this hotel (e.g., on TripAdvisor.com)? a) Yes b) No 30. Approximately, how long ago did you stay in the hotel you purchased through a flash sales web question 4 have not stayed in the hotel yet) a) Weeks b) Months c) Years 31. In the last 12 months, how many travel related purchases did you make from flash sales websites such as, Groupon Getaways , LivingSocial Escapes, TravelZoo or Jetsetter? _______ DEMOGRAHICS BLOCK This is the last section with a few questions about you. 32. Your gender: a) Male b) Female 33. Your current marital status:
195 a) Single b) In a relationship (not living together) c) Living with a partne r d) Married e) Divorced f) Separated g) Widowed 34. Do you have children? a) Yes b) No 35. The highest level of education you have completed: a) Less than High School b) High School / GED c) Some College d) 2 year College Degree e) 4 year College Degree f) g) Doctoral Degree h) Profess ional Degree (JD, MD) 36. Your age: _____________ 37. Your approximate annual household income: a) Below $20,000 b) $20,000 $29,999 c) $30,000 $39,999 d) $40,000 $49,999 e) $50,000 $59,999 f) $60,000 $69,999 g) $70,000 $79,999 h) $80,000 $89,999 i) $90,000 $99,999 j) $100,000 $109,999 k) $110,000 $119,999 l) $120,000 $129,999 m) $130,000 $139,999 n) $140,000 $149,999 o) More than $150,000 38. Your state of residence _____________________ (drop down state choice item on Qualtrics) 39. Do you have any comments? (optional)
196 40. Thank you very mu ch for your participation in my dissertation study! As a token of appreciation you will receive 90 cents to your account. Please copy the code below and paste it on the webpage. Please note that every code is unique, please do not share it with anybody else. Duplicated codes will not be accepted.
197 APPENDIX C CONSUMER PSYCHOGRAPHIC CONSTR UCTS Table C 1. Consumer p sychographic constructs # Survey items Original i tems References Quality consciousness 1 I wil l not give up high quality for a lower price I will not give up high quality for a lower price Ailawadi et al . (2001) MartÃnez & Montaner (2006) 2 I always book the best hotels in the destination I always buy the best 3 It is important for me to book a room at a high end / high quality hotel It is important to me to buy high quality products Impulsiveness 1 I often book the first hotel I like I often buy things spontaneously Rook & Fisher (1995) 2 "Just do it" describes the way I book hotel rooms "Just do it" describes the way I buy things 3 I often book hotels without thinking too much I often buy things without thinking Price consciousness 1 I compare prices before choosing a hotel I compare prices of at least a few brands before I choose one* Ailawadi et al . (2001) MartÃnez & Montaner (2006)* 2 I check prices when buying additional products at hotels (e.g. breakfast items, gift store items, etc.) I find myself checking the prices even for small items* 3 I try to get the best price for hotels It is important to me to get the best price for the products I buy 4 I am willing to put effort into finding the best hotel price Added to reflect the nature of the hotel search 5 I usually sort the hotel search results by price Financial c o nstraints 1 I always have extra money available for vacations Added to reflect the travel related nature of study 2 My household discretionary budget is usually tight My household budget is always tight Ailawadi et al . (2001) MartÃnez & Montaner (2006) 3 My household usually has problems saving funds for vacations My household often has problems making ends meet Consumer spending self c ontrol 1 I prefer saving my money 2 I spend responsibly I am responsible when it comes to how much I spend Haw s, Bearden, & Nenkov, (2012)
198 Table C 1. continued # Survey items Original i tems References 3 I usually overspend my budget I am able to resist temptation in order to achieve my budget goals 4 I lose track of my spending In social situations, I am generally aware of what I am spending Market m avenism 1 I am considered somewhat of an expert when it comes to knowing the prices of hotels I am considered somewhat of an expert when it comes to knowing the prices of products Ailawadi et al . (2001) MartÃnez & Montaner (2006) 2 My friends think of me as a good source of price information when they need to book a hotel My friends think of me as a good source of price information 3 I enjoy telling people how much they might expect to pay for differen t hotels in different locations I enjoy telling people how much they might expect to pay for different kinds of products Variety s eeking 1 I prefer different hotel brands wherever I go If I have a choice when I go place new th an go to the places I already know Wakefield and Barnes (1997 ) 2 I enjoy staying in different hotel brands for the sake of comparison I enjoy going to different entertainment spots for the sake of comparison 3 When revisiting a destination, I prefer dif ferent hotel brands that I did not try before I tend to go to a lot of different entertainment spots, just for the sake of a change of pace 4 Among two suitable options, I'd prefer a hotel brand that I did not try before Added to reflect the nature of t he hotel search Innovativeness 1 I am interested in new products and services When I see a product somewhat different from the usual, I check it out. Ailawadi et al . (2001) MartÃnez & Montaner (2006) 2 I try new products or services as soon as they com e out I am often among the first people to try a new product 3 I seek new products and services I like to try new and different things Brand l oyalty 1 I prefer one hotel brand over other hotels I prefer one brand of the products I buy Ailawadi et al . (2001)
199 Table C 1. continued # Survey items Original i tems References 2 I am willing to make an effort to search for my favorite hotel brand wherever I am traveling I am willing to make an effort to search for my favorite brand 3 Usually, I car e a lot about the hotel brand where I stay Usually, I care a lot about which particular brand I buy
200 LIST OF REFERENCES Abbey, J. (1983). Is discounting the answer to declining occupancies? International Journal of Hospitality Management , 2 (2), 77 82. Ailawadi, K. L., Beauchamp, J. P., Donthu, N., Gauri, D. K., & Shankar, V. (2009). Communication and promotion decisions in retailing: a review and directions for future research. Journal of Retailing , 85 (1), 42 55. Ailawadi, K. L., Harlam, B. A., Cesar, J ., & Trounce, D. (2006). Promotion profitability for a retailer: the role of promotion, brand, category, and store characteristics. Journal of Marketing Research , 43 (4), 518 535. Ailawadi, K.L., & Neslin, S.A. (1998). The effect of promotion on consumption : Buying more and consuming It faster. Journal of Marketing Research, 35 (3), 390 398. Ailawadi, K. L., Neslin, S. A., & Gedenk, K. (2001). Pursuing the value conscious consumer: store brands versus national brand promotions. The Journal of Marketing , 65 (1 ), 71 89. Amazon (2014). What is Amazon Mechanical Turk? Retrieved on June 6, 2014 from https://www.mturk.com/mturk/help?helpPage=overview American Hotel & Lodging Association. (2013). Abou t us. Retrieved on 09/08/2013 from http://www.ahla.com/content.aspx?id=3438 American Hotel & Lodging Association (2014). 2013 Lodging industry profile. Retrieved on 06/13/2014 from http://www.ahla.com/content.aspx?id=35603 Anderson, C., & Xie, X. (2010). Improving hospitality industry sales: Twenty five years of revenue management. Cornell Hospitality Quarterly, 51 (1), 53 67. Anderson, R. E., & Srinivasan, S. S. (2003). E satisfaction and e loyalty: A contingency framework. Psychology & Marketing , 20 (2), 123 138. Atkins, K. G., & Kim, Y. K. (2012). Smart shopping: conceptualization and measurement. International Journal of Retail & Distribution Management , 40 (5), 360 375. Ayadi, N., Giraud, M., & Gonzalez, C. (2013). An investigation of consumers' self control mechanisms when confronted with repeated purchase temptations: Evidence from online private sales. Journal of Retailing and Consumer Services, 20 (3), 272 281. Baumeis ter, R. F. (2002). Yielding to temptation: Self control failure, impulsive purchasing, and consumer behavior. Journal of Consumer Research , 28 (4), 670 676.
201 Bawa, K., & Shoemaker, R. W. (1987). The coupon prone consumer: some findings based on purchase behavior across product classes. The Journal of Mar keting , 51 (4) 99 110. Beach, L. R., & Mitchell, T. R. (1978). A contingency model for the selection of decision strategies. Academy of Management Review , 3 (3), 439 449. Berne, C., Garcia Gonzalez, M., & Mugica, J. (2012). How ICT shifts the power balance o f tourism distribution channels. Tourism Management , 33 (1), 205 214. Blattberg, R., Buesing, T., Peacock, P., & Sen, S. (1978). Identifying the deal prone segment. Journal of Marketing Research , 15 (3), 369 377. Bloemer, J. M., & Kasper, H. D. (1995). The c omplex relationship between consumer satisfaction and brand loyalty. Journal of Economic P sychology , 16 (2), 311 329. Braid, R.M. (1996).Peak load pricing of a transport facility with an unpriced substitute. Journal of Urban Economics, 40 ,179 197 Buhalis, D ., & Law, R. (2008). Progress in information technology and tourism management: 20 years on and 10 years after the Internet the state of eTourism research. Tourism Management, 29 , 609 623. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mech anical Turk a new source of inexpensive, yet high quality, data? Perspectives on Psychological Science , 6 (1), 3 5. Burgan, B., & Mules, T. (2001). Reconciling cost benefit and economic impact assessment for event tourism. Tourism Economics, 7 (4), 321 330. Burton, S., Lichtenstein, D. R., Netemeyer, R. G., & Garretson, J. A. (1998). A scale for measuring attitude toward private label products and an examination of its psychological and behavioral correlates. Journal of the Academy of Marketing Science , 26 (4) , 293 306. Campo, S., & YagÃ¼e, M. J. (2008). Tourist loyalty to tour operator: Effects of price promotions and tourist effort. Journal of Travel Research , 46 (3), 318 326. Canina, L., & Enz, C. A. (2006). Why discounting still doesn't work: A hotel pricing update . Cornell University, School of Hotel Administration, The Center for Hospitality Research. Carroll, J. D. (1986). Focus on discounting hotel rack rates. Cornell Hotel and Restaurant Administration Quarterly , 27 (2), 13 13. Carroll, B., & Siguaw, J. (2 003). The evolution of electronic distribution: effects on hotel intermediaries. Cornell Hotel and Restaurant Administration Quarterly, 44 , 38 50.
202 Casler, K., Bickel, L., & Hackett, E. (2013). Separate but equal? A comparison of participants and data gathe to face behavioral testing. Computers in Human Behavior , 29 (6), 2156 2160. Chandon, P. (1995). Consumer research on sales promotions: A state of the art literature review. Journal of Marketing Management, 11 (5), 419 441. Chandon, P., Wansink, B., & Laurent, G. (2000). A benefit congruency framework of sales promotion effectiveness. Journal of marketing , 64 (4), 65 81. Charmaz, K. (2006). Constru cting grounded theory: A practical guide through qualitative analysis. Pine Forge Press. Christou, E. (2011). Exploring online sales promotions in the hospitality industry. Journal of Hospitality Marketing & Management , 20 (7), 814 829. Choi, S., & Kimes, S . E. (2002a). Hotel revenue management. Cornell Hotel and Restaurant Administration Quarterly, 43 (3), 23 31. Choi, S., & Kimes, S. E. (2002b). Electronic distribution channels' effect on hotel revenue management. Cornell Hotel and Restaurant Administration Quarterly , 43 (3), 23 31. Choi, S., & Matilla, A.S. (2009). Perceived fairness of price differences across channels: the moderating role of price frame and norm perceptions. Journal of Marketing Theory and Practice , 17 (1), 37 47. Chung, Y.K. (2000). Hotel room rate pricing strategy for market share in oligopolistic competition eight year longitudinal study of super deluxe hotels in Seoul. Tourism Management, 21 (2), 135 145. Clements, C. J., & Josiam, B. M. (1995). Training: quantifying the financial benef its. International Journal of Contemporary Hospitality Management, 7 (1), 10 15. Connolly, D.J., Olsen, M.D., & Moore, R.G. (1998). The Internet as a distribution channel. Cornell Hotel and Restaurant Administration Quarterly , 39 (4), 42 54. Creswell, J. W. (2007). Qualitative inquiry and research design: Choosing among five approaches . SAGE Publications, Inc. , CA : Thousand Oaks. Creswell, J.W. (2011). Educational research: Planning, conducting and evaluating quantitative and qualitative research (4 th ed.) N ew Delhi: PHI Learning Private limited. Crew, M., Fernando, C., & Kleindorfer, P. (1995). The theory of peak load pricing: A survey. Journal of Regulatory Economics, 8 , 215 248. Croes, R., & Semrad, K. J. (2012a). Discounting works in the hotel industry: a structural approach to understanding why. Tourism Economics , 18 (4), 769 779.
203 Croes, R., & Semrad, K. (2012b). Does Discounting Work in the Lodging Industry? Journal of Travel Research , 51, 617 631. De Alessi, L. (1977). Ownership and peak load pricing in the electric power industry. Quarterly Review of Economics and Business , 17 (4) , 7 26 . Decrop, A. (1999). Triangulation in qualitative tourism research. Tourism Management , 20 (1), 157 161. Desiraju, R., & Shugan, S.M. (1999). Strategic service pricing and y ield m anagement. Journal of Marketing, 63 (1), 44 56. Dev, C. S., & Falk, L. W. (2011). T o Groupon or n ot to Groupon: A tour o d ilemma. Cornell Hospitality Re port , 11 (19), 4 18. Dunn, K.D, & Brooks, D.E. (1990). Profit analysis: Beyond yield manag ement. Cornell Hotel and Restaurant Administration Quarterly, 31 , 80 90. Edelman, B., Jaffe, S., & Kominers, S.D. (2011). To Groupon or not to Groupon: the profitability of deep discounts. Working paper, Harvard Business School . Emmer, R.M., Tauck, C., Wi lkinson, S., & Moore, R.G. (1993). Marketing hotels: using global distribution system. Cornell Hotel and Restaurant Administration Quarterly , 34 , 80 89. Enz, C. A. (2009). Key issues of concern in the lodging industry: what worries managers. Cornell Hospit ality Report , 9 (4), 1 17. Enz, C. A., Canina, L., & Lomanno, M. (2004). Why discounting doesn't work: The dynamics of rising occupancy and falling revenue among competitors . Center for Hospitality Research . Enz, C. A., Canina, L., & Lomanno, M. (2009). Com petitive pricing decisions in uncertain times. Cornell Hospitality Quarterly , 50 (3), 325 341. Feick, L. F., & Price, L. L. (1987). The market maven: A diffuser of marketplace information. The Journal of Marketing , 51(1), 83 97. Field, A. (2009). Discoverin g statistics using SPSS . Sage Publications Limited. Garretson, J. A., Fisher, D., & Burton, S. (2002). Antecedents of private label attitude and national brand promotion attitude: similarities and differences. Journal of Retailing , 78 (2), 91 99. Gazzoli, G ., Kim, W.G., & Palakurthi, R. (2008). Online distribution strategies and competition: are the global hotel companies getting it right? International Journal of Contemporary Hospitality Management , 20 (4), 371 387.
204 Goodman, J. K., Cryder, C. E., & Cheema, A . (2013). Data collection in a flat world: The strengths and weaknesses of Mechanical Turk samples. Journal of Behavioral Decision Making , 26 (3), 213 224. Grewal, D., Ailawadi, K.L., Gauri, D., Hall, K., Kopalle, P., & Robertson, J.R. (2011). Innovations i n retail pricing and promotions. Journal of Retailing , 87S , S43 S52. Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18 (1), 59 82. Guo, P., Xiao, B., & Li, J. (2 012). Unconstraining methods in revenue management systems: Research overview and prospects. Advances in Operations Research , 2012 , 1 22. Gupta, R. (2012). Making apt use of flash sales for both distribution and marketing initiatives. Retrieved on June 8, 2012 from http://www.hospitalitynet.org/news/4055523.html Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Pears on Education, Inc. Han, H., Kim, Y., & Kim, E. K. (2011). Cognitive, affective, conative, and action loyalty: testing the impact of inertia. International Journal of Hospitality Management , 30 (4), 1008 1019. Hanks, R., Cross, R.G., & Noland, R.P. (2002). D iscounting in the hotel industry: A new approach . Cornell Hotel and Restaurant Administration Quarterly, August 2002 , 94 103. Hanley, N., & Barbier, E. (2009). Pricing nature: Cost benefit analysis and environmental policy . Edward Elgar Publishing. Haws, K. L., Bearden, W. O., & Nenkov, G. Y. (2012). Consumer spending self control effectiveness and outcome elaboration prompts. Journal of the Academy of Marketing Science , 40 (5), 695 710. Hayes, D. K., & Miller, A. (2011). Revenue management for the hospital ity industry . Wiley. Jain, D., & Singh, S. S. (2002). Customer lifetime value research in marketing: A review and future directions. Journal of interactive marketing , 16 (2), 34 46. Jang, S., Ismail, J. A., & Ham, S. (2001). Heavy spenders, medium spenders, and light spenders of Japanese outbound pleasure travelers. Journal of Hospitality & Leisure Marketing, 9 (3 4), 83 106. Jeuland, A.P., & Shugan ,S.M. (1983). Managing Channel Profits. Marketing Science , 2 (3), 239 272.
205 Jeuland, A.P., & Shugan ,S.M. (2008). Commentary: Managing Channel Profits. Marketing Science , 27 (1), 49 51. Kalnins, A. (2006). Markets: The US lodging industry. The Journal of Economic Perspectives, 20 (4), 203 218 . Kandampully, J., & Suhartanto, D. (2000). Customer loyalty in the hotel indu stry: the role of customer satisfaction and image. International Journal of Contemporary Hospitality Management , 12 (6), 346 351. Kang, B., Brewer, K.P., & Baloglu, S. (2007). Profitability and survivability of hotel distribution channels: An industry pers pective. Journal of Travel and Tourism Marketing, 22 (1), 37 50. Keeler, T. E., & Small, K. A. (1977). Optimal peak load pricing, investment, and service levels on urban expressways. The Journal of Political Economy, 85 (1), 1 25. Kidokoro, Y. (2004). Cost b enefit analysis for transport networks: Theory and application. Journal of Transp ort Economics and Policy , 38 (2), 275 307. Kim, S., & Martinez, B. (2013). Fashion consumer groups and online shopping at private sale sites. International Journal of Consumer Studies , 37 (4), 367 372. Kimes, S. E. (1989). The basics of yield management. Cornell Hotel & Restaurant Administration Quarterly , 30 , 14 19. Kimes, S., & Wirtz, J. (2003). Has revenue management become acceptable?: findings from an international study on the perceived fairness of rate fences. Journal of Services Research, 6 , 125 135. Kleijnen, M., De Ruyter, K., & Wetzels, M. (2007). An assessment of value creation in mobile service delivery and the moderating role of time consciousness. Journal of Retaili ng , 83 (1), 33 46. Klemperer, P. (1987). Markets with consumer switching costs. The Quarterly Journal of Economics , 375 394. Kotler, P., Bowen, J.T., & Makens, J.C.(2010). Marketing for hospitality and tourism (5 th ed.), Upper Saddle River, NJ : Pearson Educ ation Inc. Kracht, J., & Wang, Y. (2010). Examining the tourism distribution channel: evolution and transformation. International Journal of Contemporary Hospitality Management , 22(5), 736 757. Leeflang, P. S., & ParreÃ±o Selva, J. (2012). Cross category de mand effects of price promotions. Journal of the Academy of Marketing Science , 40 (4), 572 586.
206 Legoherel, P. (1998): Toward a market segmentation of the tourism trade: Expenditure levels and consumer behavior instability, Journal of Travel & Tourism Market ing, 7 (3), 19 39. Lichtenstein, D. R., Netemeyer, R. G., & Burton, S. (1990). Distinguishing coupon proneness from value consciousness: an acquisition transaction utility theory perspective. The Journal of Marketing , 54 (3), 54 67. Lim, C. M., Kim, Y. K., & Runyan, R. (2013). Segmenting luxe bargain shoppers using a fuzzy clustering method. International Journal of Retail & Distribution Management , 41 (11/12), 3 3. LivingSocial (2011). LivingSocial Escapes books more than 500,000 room nights. Retrieved on Aug ust 9, 2012 from http://corporate.livingsocial.com/inthenews/articles/39 Locke, K. (2001). Grounded theory in management research . Sage. Manning Jr, W. G., Mitchell, B. M., & Acton, J. P. (1979). Design of the Los Angeles peak load pricing experiment for electricity. Journal of Econometrics , 11 (1), 131 194. Manning, R. E., & Powers, L. A. (1984). Peak and off peak use: redistributing the outdoor recreation/tourism load. Journal of Trave l Research, 23 (2), 25 31. MartÃnez, E., & Montaner, T. (2006). The effect of consumer's psychographic variables upon deal proneness. Journal of Retailing and Consumer Services , 13 (3), 157 168. Mayock, P. (2012). Flash sales fizzle with hoteliers. Retrieved on June 8, 2012 from http://www.hotelnewsnow.com/Articles.aspx/7971/Flash sales fizzle with hoteliers Mazumdar, T., Raj, S.P., & Sinha, I. (2005). Reference p rice research; review and propositions. Journal of Marketing , 69 (4), 84 102. approach to developing a career change model in hospitality. International Journal of Hospitalit y Management , 38 , 89 98. Miao, L., & Mattila, A. S. (2007). How and how much to reveal? The effects of price transparency on consumers' price perceptions. Journal of Hospitality & Tourism Research, 31 (4), 530 545. Mitchell, B. M., Manning Jr, W. G., & Acto n, J. P. (1981). Peak load pricing: European lessons for US energy policy . Land E conomics, 57 (3), 476 480. Mittal, B. (1994). An integrated framework for relating diverse consumer characteristics to supermarket coupon redemption. Journal of Marketing Resea rch , 31 (4), 533 544. Mok, C., & Iverson, T. J. (2000). Expenditure based segmentation: Taiwanese tourists to Guam. Tourism Management, 21 (3), 299 305.
207 web sites. Internat ional Journal of Hospitality Management , 27 , 284 292. Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2008). Verification strategies for establishing reliability and validity in qualitative research. International Journal of Qualitative Meth ods, 1 (2), 13 22. Mules, T., & Dwyer, L. (2005). Public sector support for sport tourism events: the role of cost benefit analysis. Sport in Society, 8 (2), 338 355. Murphy, K., Semrad, K., & Yost, E. (2013). The Impact of Discounting Room Rates on In House Restaurant Sales. International Journal of Hospitality & Tourism Administration, 14 (1), 50 65. Nam, J., Ekinci, Y., & Whyatt, G. (2011). Brand equity, brand loyalty and consumer satisfaction. Annals of Tourism Research , 38 (3), 1009 1030. Neslin, S.A., & S hoemaker, R.W. (1998). An Alternative Explanation for Lower Repeat Rates after Promotion Purchases. Journal of Marketing Research, 26 (2), 205 213 Nicholas, L. (Ed.). (2009). Introduction to psychology. Juta and Company Ltd. Niedrich, R.W., Sharma, S., & Wedell, D.H. (2001). Reference price and price perceptions a comparison of alternative models. Journal of Consumer Research , 28 (3), 339 354. Nijs, V. R., Dekimpe, M. G., Steenkamps, J. B. E., & Hanssens, D. M. (2001). The category demand effects of price p romotions. Marketing science , 20 (1), 1 22. Noble, S. M., Griffith, D. A., & Adjei, M. T. (2006). Drivers of local merchant loyalty: Understanding the influence of gender and shopping motives. Journal of Retailing , 82 (3), 177 188. Nusair, K., Yoon, H.J., Na ipaul, S., & Parsa, H.G. (2010). Effect of price discount frames end service industries. International Journal of Contemporary Hospitality Management , 22 (6), 814 835 Nyheim, P., McFadden, F.M., & Connolly, D.J. ( 2005). Technology strategies for the hospitality industry , Upper Saddle River, NJ : Pearson Education Inc. industry perspectives. Cornell Hotel and restaurant Adminis tration Quarterly , June 2002 , 33 45. channels of distribution. International Journal of Hospitality Management , 23 , 179 199.
208 agement practices across multiple electronic distribution channels. Information Technology & Tourism , 10, 161 172. Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing R esearch , 17 (4) , 460 469. Piccoli, G., & Dev, C. (2012). Emerging marketing channels in hospitality: a global study of Internet enabled flash sales and private sales. Corne l l Hospitality Report , 12 (5), 1 21. Pizam, A., & Reichel, A. (1979). Big spenders and little spenders in US tourism. Journal of Travel Research, 18 (1), 42 43. Raju, J. S. (1992). The effect of price promotions on variability in product category sales. Marketing Science , 11 (3), 207 220. Reichheld, F. F., & Teal, T. (2001). The loyalty effect: The hidden fo rce behind growth, profits, and lasting value. Harvard Business Press. Ritter, L. A., & Sue, V. M. (2007). Introduction to using online surveys. New Directions for Evaluation , 2007 (115), 5 14. Rohlfs, K.V, & Kimes, S.E. (2005). Best available rate pricing at hotels. The Center of Hospitality Research Reports, 5 (7), 1 19. Rook, D. W., & Fisher, R. J. (1995). Normative influences on impulsive buying behavior. Journal of consumer research , 22 (3), 305 313. Schaal, D., (2011a). Expedia decides to make hotel flas h sales permanent feature. Retrieved on June 8, 2012 from http://www.tnooz.com/2011/02/15/news/expedia decides to make hotel flash sales permanen t feature Schaal, D., (2011b). Expedia partners with Groupon on gateways. Tnooz. Retrieved on June 8, 2012 from http://www.tnooz.com/2011/06/01/news/expedia pa rtners with groupon on getaways/ Schaal, D. (2012a). Do flash sales level playing field for independent hotels? Tnooz. Retrieved on June 8, 2012 from http://www.tnooz.com/2012/04/17/data/do flash sales level playing field for independent hotels/ Schaal, D. (2012b). LivingSocial hits 1M room nights but Priceline should not feel boxed in. Tnooz. Retrieved on April 14, 2013 from http://www.tnooz.com/2012/03/30/news/livingsocial hits 1m room nights but priceline should not feel boxed in/ Schaal, D. (2012c). Shh industry dirt y secret: most hoteliers would try flash sales again. Tnooz. Retrieved on June 8, 2012 from http://www.tnooz.com/2012/04/04/data/shh i ndustry dirty secret most hoteliers would try flash sales again/
209 Shimp, T. A., & Kavas, A. (1984). The theory of reasoned action applied to coupon usage. Journal of C onsumer R esearch , 11 (3), 795 809. Shugan, S. (1985). Implicit understandings in channels of distribution. Management Science . 31 (4), 435 460. Shugan, S.M., & Xie, J. (2004). Advance selling for services. California Management Review, 46 (3), 37 54. Shugan, S.M., & Xie, J. (2005). Advance selling as a competitive marketing tool. International Jo urnal of Research in Marketing , 22 , 351 373. Sigala, M. (2013). A framework for designing and implementing effective online coupons in tourism and hospitality. Journal of Vacation Marketing , 19 (2), 165 180. Smith, B. C., Leimkuhler, J. F., & Darrow, R. M. (1992). Yield management at American airlines. Interfaces , 22 (1), 8 31. HeBS Digital . Retrieved on June 6, 2012 from http://www.hebsdigital.com/blog/it%E2%80%99s official the dumbest hotel distribution idea/ Starkov, M. (2012). I told you: How the flash sales bubble popped. HeBS Digital . Retrieved on June 6, 2012 from http://www.hebsdigital.com/blog/i told you so how the flash sale bubble popped/ Starkov, M., & Kirby, A. (2011). Another look at flash sales sites and whether they should or should not play a role in hotel distribution. HeBS Digital . Retrieved on June 6, 2012 from http://www.hebsdigital.com/blog/another look at flash sales sites and whether they should or should not play a role in hotel distribution/ STR (2014) About. Retrieved on June 28 , 2 014 from http://str.com/about Strauss, A. L., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques . Newbury Park, CA: Sage. Strauss, A. L., & Corbin, J. (1998). Basics of qualit ative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage. Thaler, R. (1985). Mental accounting and consumer choice. Marketing science , 4 (3), 199 214. Thaler, R. H. (2008). Mental accounting and consumer choice. Mark eting Science , 27 (1), 15 25. Van Selm, M., & Jankowski, N. W. (2006). Conducting online surveys. Quality and Quantity , 40 (3), 435 456.
210 Wakefield, K. L., & Barnes, J. H. (1997). Retailing hedonic consumption: a model of sales promotion of a leisure service. Journal of Retailing, 72 (4), 409 427. Wang, C. Y., & Mattila, A. S. (2010). A grounded theory model of service providers' stress, emotion, and coping during intercultural service encounters. Managing Service Quality, 20 (4), 328 342. Wang, H. Y., & Wang, S . H. (2010). Predicting mobile hotel reservation adoption: insight from a perceived value standpoint. International Journal of Hospitality Management , 29 (4), 598 608. Weatherford, L. R., & Bodily, S. E. (1992). A taxonomy and research overview of perishabl e asset revenue management: yield management, overbooking, and pricing. Operations Research , 40 (5), 831 844. Weatherford, L. R., & Kimes, S. E. (2003). A comparison of forecasting methods for hotel revenue management. International Journal of Forecasting, 19 (3), 401 415. Wenders, J. T. (1976). Peak load pricing in the electric utility industry. The Bell Journal of Economics , 7(1), 232 241. Wirtz, J., Kimes, S. E., Theng, J. H. P., & Patterson, P. (2003). Revenue management: resolving potential customer conf licts. Journal of Revenue and Pricing Management , 2 (3), 216 226. Woo, K. S., & Fock, H. K. (2004). Retaining and divesting customers: An exploratory study of right customers , right customers, and wrong customers. Journal of Services Marketing , 18 (3), 187 197. Xie, J., & Shugan, S.M. (2001). Electronic tickets, smart cards, and online payments: when and how to advance sell. Marketing Science, 20 (3), 219 243. Yang, J. T. (2012). Identifying the attributes of blue ocean strategies in hospitality. Int ernational Journal of Contemporary Hospitality Management, 24 (5), 701 720. Yang, Z., & Peterson, R. T. (2004). Customer perceived value, satisfaction, and loyalty: the role of switching costs. Psychology & Marketing , 21 (10), 799 822. Yuksel, A., Yuksel, F. , & Bilim, Y. (2010). Destination attachment: Effects on customer satisfaction and cognitive, affective and conative loyalty. Tourism Management , 31 (2), 274 284. Zeithaml, V. A., Bitner, M. J., & Gremler, D. D. (2006). Services marketing: Integrating custo mer focus across the firm . New York, NY: McGraw Hill/Irwin. Zhu, W. W., & Morosan, C. (2014). An empirical examination of guests' adoption of interactive mobile technologies in hotels: Revisiting cognitive absorption, playfulness, and security. Journal of Hospitality and Tourism Technology , 5 (1), 78 94.
211 BIOGRAPHICAL SKETCH Ek aterina Berezina is an Assistant Professor at the University of South Florida Sarasota Manatee. She received her b Penza State University in Russia. Ekaterina had been working for four years in travel agencies in Russia, primarily as an international sales manag er. In 2010, she completed her m of Business and E conomics at the University of Delaware. Ekaterina received her doctoral degree in Health and Human Performance at the University of Florida in 2014 . During her academic tenure at the University of Florida , Ekaterina taught Introduction to Hospitality, Reve nue Resource Management, and Revenue Management in Hospitality Business. Ekaterina information technology in hospitality and tourism, electronic distribution, and revenue management. Ekaterina is a recipient of two Decision Sciences Institute, the other is from iHITA. Her articles have been published in the International Journal of Contemporary Hospitality Management, the Journal of Hospitality and Tourism Technology, the Journal of Quality Assurance in Hospitality , and other reputable hospitality and tourism journals. Ekaterina is an active member of professional and academic hospitality associations. Ek aterina serves on the board of the i nternational Hospitality Information Tech nology Association (iHITA) as a Secretary, as the Chair of the Hospitality Financial and Technology Professionals ( HFTP ) Social Media Council, and as an A ssistant Editor for the Journal of Hospitality and Tourism Technology. Ekaterina is a Certified Hospit ality Technology Professional (CHTP) commissioned by HFTP and American Hotel and Lodging Educational Institute (AH&LEI).