Computers in Human Behavior 27 (2011) 634–639

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

The influence of user-generated content on traveler behavior: An empirical

investigation on the effects of e-word-of-mouth to hotel online bookings
Qiang Ye a,*, Rob Law b,1, Bin Gu c,2, Wei Chen d
a
School of Management, Harbin Institute of Technology, Heilongjiang 150001, China
b
School of Hotel and Tourism Management, Hong Kong Polytechnic University, Hong Kong, China
c
McCombs School of Business, University of Texas at Austin, USA
d
Rady School of Management, The University of California, San Diego, CA 92093, USA

a r t i c l e i n f o a b s t r a c t

Article history: The increasing use of web 2.0 applications has generated numerous online user reviews. Prior studies
Available online 14 May 2010 have revealed the influence of user-generated reviews on the sales of products such as CDs, books, and
movies. However, the influence of online user-generated reviews in the tourism industry is still largely
Keywords: unknown both to tourism researchers and practitioners. To bridge this knowledge gap in tourism man-
User-generated content agement, we conducted an empirical study to identify the impact of online user-generated reviews on
Traveler behavior business performance using data extracted from a major online travel agency in China. The empirical
Hotel
findings show that traveler reviews have a significant impact on online sales, with a 10 percent increase
Online bookings
in traveler review ratings boosting online bookings by more than five percent. Our results highlight the
importance of online user-generated reviews to business performance in tourism.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction tant to assess their effect. A few recently conducted studies have
demonstrated that online user-generated reviews have a signifi-
The rapid growth of web 2.0 applications, which empower cant influence on sales of consumer products (Chevlier & Mayzlin,
Internet users and allow two-way information communications 2006; Duan, Gu, & Whinston, 2008). A key insight deduced from
in travel and tourism, has generated an enormous number of on- previous studies is that the influence of user reviews is particularly
line user-generated contents (UGC) on hotels, travel destinations, significant for experience goods (Klein, 1998), as their quality is
and travel services (Sigala, 2008). At the same time, an increasing often unknown before consumption (Katz & Lazarsfeld, 1955;
number of travelers are using the Internet for travel planning Nelson, 1970) and consumers have to rely on word-of-mouth and
(Litvin, Goldsmith, & Pan, 2008; Sigala, Lockwood, & Jones, 2001). online reviews to make inferences about such goods. Most services
Results of large-scale surveys have shown that searching for tra- and products offered by the hotel industry are experience goods
vel-related information is one of the most popular online activities – the quality of tour operations and hotels, for example, is only
(Pew Internet and American Life Project, 2006). In addition, Com- known after the service has been consumed (Litvin et al., 2008).
plete, Inc (2007) found one-third of travel purchasers visited a However, there is a very limited number of prior studies in the
message board, forum, or online community before their online existing tourism literature on the impact of online user-generated
travel purchasing because they believed online reviews would be reviews on the performance of firms, despite the fact that studies
helpful to their purchase decision. Forrester Research (2006) on this topic would help tourism practitioners to better understand
estimated that 34.7 percent of total online spending is related to the importance of online user reviews for their businesses.
travel, and a recent survey indicated that more than 74 percent Using data extracted from major online travel agencies in China,
of travelers use the comments of other consumers as information we conducted an empirical study to bridge this research gap in
sources when planning trips for pleasure (Gretzel & Yoo, 2008). tourism management. The data, which consisted of consumer-gen-
In total, online reviews influence more than US$10 billion in online erated reviews, were retrieved from www.ctrip.com (NASDAQ:
travel purchases every year (Compete, 2007), and it is thus impor- CTRP), a major online travel agency in China. One of the challenges
in the study was that it is difficult, if not impossible, to obtain pri-
vate booking data of online travelers. In previous studies, a variety
* Corresponding author. Tel.: +86 451 86414022; fax: +86 451 86414024. of proxies have been used to infer product sales from observed
E-mail addresses: [email protected] (Q. Ye), [email protected] (R. Law),
data. Ghose and Ipeirotis (2006), as well as Chevlier and Mayzlin
[email protected] (B. Gu), [email protected] (W. Chen).
1
Tel.: +852 2766 6349. (2006), for example, used Amazon.com’s sales rankings to infer
2
Tel.: +1 512 471 1582. product sales.

0747-5632/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chb.2010.04.014
 Q. Ye et al. / Computers in Human Behavior 27 (2011) 634–639 635

In our study, a unique feature of ctrip.com that allows only trav- service sales. For instance, using a difference-in-difference model,
elers who book hotel rooms through its website to post reviews is Chevlier and Mayzlin (2006) examined the effect of consumer
leveraged. As such, the number of reviews is expected to be closely reviews of book sales on Amazon.com and Barnesandnoble.com,
correlated to the number of room sales. We, therefore, use the and found that word-of-mouth significantly influence product
number of reviews as a proxy for hotel room sales (Ghose and sales. Duan et al. (2008) conducted a panel data analysis with mo-
Ipeirotis (2006); Ye, Law, & Gu, 2009a, 2009b). To further investi- vie box office revenue data, and findings show that the valence of
gate the applicability of the approach, empirical data with actual online user reviews has no significant impact on movie box office
online sales were collected and analyzed to test the model robust- revenues meanwhile box office sales are significantly influenced by
ness in this study. the volume of online reviews. In the tourism industry, Vermeulen
The aim of the study was to identify the impact of online word- and Seegers (2009) revealed that positive online reviews improve
of-mouth on sales in the tourism and hotel industries. We tested the perception of hotels among potential consumers. Since online
the influence of the valence and variance of online consumer-gen- traveler reviews are an important source of information to both
erated reviews based on the number of online bookings for the ho- travelers and tourism firms, researchers have attempted to analyze
tels included in this study. Our work was designed to extend and understand online traveler reviews by sophisticated technolo-
current research on the effect of online consumer-generated gies (Govers & Go, 2005; Ye et al., 2009a, 2009b).
reviews to encompass experience goods in tourism. The study also Ghose and Ipeirotis (2006) studied the influence of online
makes a methodological contribution to tourism research by reviews on product sales for a variety of consumer products, and
validating the proxy for the number of online bookings using found that the subjectivity and polarity of the ratings in reviews
empirical data. had a significant influence on online sales of certain products. They
The rest of this paper is organized as follows. In Section 2, we explained their findings using the cognitive load theory, and
provide a review of previous studies and the proposed hypotheses. argued that certain types of online reviews reduce the cognitive
We then describe the research model and data in Section 3. The load of the reader, thereby generating higher sales. A study con-
results are reported in Section 4. Section 5 assesses the robustness ducted by Dickinger and Mazanec (2008) showed that the recom-
of the proxy for the number of online bookings. Theoretical and mendations of friends and online reviews are the most important
practical implications of the results are described in Section 6, factors that influence online hotel bookings. Despite the increasing
and the last section offers some concluding comments. importance of online user-generated content, a number of studies
have reported that online user-generated reviews are perceived as
being lower in credibility than traditional word-of-mouth due to
2. Background and research hypotheses the absence of source cues on the Internet (Dellarocas, 2003;
Smith, Menon, & Sivakumar, 2005). As such, the influence of con-
Web 2.0 and UGC have been, and will likely be, increasingly sumer reviews merits further investigation.
changing the way that people search, find, read, gather, share, devel- Previous studies have shown that online travel reviews may
op, and consume information. As such, they provide tremendous influence the decisions of travelers (Vermeulen & Seegers, 2009;
opportunities for E-commerce (Sigala, 2008). In E-commerce, UGC Gretzel & Yoo, 2008). However, in the context of tourism the
may serve as a new form of word-of-mouth for products/services impact of online consumer-generated reviews on the performance
or their providers. The importance of word-of-mouth on business of tourism firms remains largely unknown. Vermeulen and Seegers
has been widely discussed and researched, particularly since the (2009) conducted an experimental study with 168 participants to
worldwide adoption of Internet technologies, which have determine the impact of online reviews on the attitudes of travel-
revolutionized the distribution and influence of word-of-mouth ers to hotels, and revealed that exposure to online reviews en-
(Anderson, 1998; Goldenberg, Libai, & Muller, 2001; Stokes & Lomax, hanced hotel awareness, and that positive reviews improved the
2002; Zhu & Zhang, 2006). Through the Internet, individuals can attitudes of travelers toward hotels. Based on survey data with
make their ideas and opinions more easily accessible to other 1480 respondents, Gretzel and Yoo (2008) examined the role of
Internet users (Dellarocas, 2003). Up to 2004, 44 percent of U.S. travel reviews in trip planning processes, and demonstrated the
Internet users had presented their thoughts on the Internet, and importance of online consumer reviews at an individual level.
the majority of consumers reported that they trusted the opinions Nevertheless, to the best of our knowledge, only a few published
which were posted online by other consumers (Gretzel & Yoo, 2008). articles have documented the role of online reviews on sales at
Online user-generated reviews about travel destinations, hotels, the firm level. For instance, Ye et al. (2009a, 2009b) investigated
and tourism services have become important sources of informa- the impacts of online user reviews on hotel room sales, but the
tion for travelers (Pan, MacLaurin, & Crotts, 2007), with reports small data set and the proxy variable without empirical tests ren-
indicating that each year hundreds of millions of potential visitors dered its inability to draw any generalized conclusions. An empir-
consult online reviews (Tripadvisor.com., 2006). Among these po- ical study with a large dataset and robustness validation of the
tential visitors, 84 percent were affected by reviews when making proxy variable should thus provide new insights to help tourism
their travel reservations (Travelindustrywire.com., 2007). Golden- researchers and practitioners better understand the importance
berg et al. (2001) stressed that consumer decision-making pro- of online word-of-mouth in tourism.
cesses are strongly influenced by word-of-mouth from other This study makes an attempt to bridge the research gap by
consumers. Gretzel and Yoo (2008) further found that reviews quantifying the effect of online user reviews on the number of
provided by other travelers are often perceived by readers to be online sales. Previous studies have indicated that the valence of
more up-to-date, enjoyable, and reliable than information pro- online user-generated reviews may influence sales of books and
vided by travel service providers. digital products (Chevlier & Mayzlin, 2006; Ghose & Ipeirotis,
Additionally, Zhu and Zhang (2006), as well as Cheung, Shek, 2006; Ye et al., 2009a, 2009b). We thus propose the following
and Sia (2004) pointed out that online user-generated reviews hypothesis on the basis of their findings.
are of use to both consumers and online retailers. Likewise, Hypothesis 1: A higher valence of average review ratings of a
Dellarocas (2003) indicated that online word-of-mouth can have hotel results in more online bookings for that hotel.
important implications for managers in terms of brand building, The variance of opinions among reviews is another concern in
product development, and quality assurance. Findings of recent previous studies (Ghose & Ipeirotis, 2006; Ye et al., 2009a,
studies show different effects of online reviews on the product/ 2009b). Specifically, the influences of variance on the sales are
636 Q. Ye et al. / Computers in Human Behavior 27 (2011) 634–639

not consistent in their findings. We then formulate the second Table 1

hypothesis. The 10 selected cities.

Hypothesis 2: A higher variance in the review ratings for a hotel City Number of hotels on Ctrip.com GDP rank in 2007
results in fewer online bookings for that hotel. Shanghai 703 1
Beijing 678 2
Guangzhou 80 3
3. Methodology
Shenzhen 231 4
Chongqing 103 5
3.1. Research model Chengdu 151 6
Harbin 77 7
To identify the influence of online reviews, we developed the Xi’an 132 8
Lanzhou 20 9
following log-linear regression model for the online room sales of
Lhasa 30 10
hotel i. Total 2205 –

lnðe Salesi Þ ¼ l þ b1 mean rev iew Rating i

þ b2 v ariance rev iew Rating i þ b3 lnðpricei Þ
þ WotherFactorsi þ ei ð1Þ developed a crawler using Ruby on Rails (http://www.rubyon-
rails.org) to automatically download web pages containing
In this model, mean_Ratingi and variance_Ratingi represent hotel consumer reviews of hotels and other information of the hotels
i’s average rating and variance, respectively, in the online user from Ctrip.com, and created another Ruby-based system to parse
reviews. Pricei identifies the room rate of hotel i, and the variable data from HTML and XML web pages into our database. The data
otherFactorsi includes all other factors that can influence online collection process and pilot studies were carried out in mid-
sales, including the star category of the hotel and the size of the 2008. We used the crawler to retrieve all available information
city in which the hotel is located. In China, the developments of about hotel bookings from Ctrip.com for ten randomly selected
the economy and e-commerce are imbalance cross the country. large cities in China, including Shanghai, Beijing, Guangzhou,
More developed big cities, many of them are near the coast, have Shenzhen, Chongqing, Chengdu, Harbin, Xi’an, Lanzhou, and Lhasa.
both higher levels of economic and e-commerce development than Ctrip.com holds information on 2205 hotels in these cities (see
smaller and inner cities. So we include the size of the city in term of Table 1 for details). Among these hotels, 566 are no longer avail-
GDP ranking as a control variable. able for online bookings, and we thus deleted these unavailable
As previously mentioned, a challenge in measuring the influ- hotels from the dataset. Our final dataset contains 1639 hotels.
ence of hotel sales is that room sales data are not available to For each hotel in our dataset, we collected two types of informa-
researchers. We could not observe actual hotel bookings from tion, comprising (i) a detailed hotel description from Ctrip.com and
travel agent websites, as these are the private data of the individual (ii) any review information, including review text, author, publish-
hotels concerned. Previous studies have used a variety of proxies to ing date, and customer rating. We collected all reviews for the ho-
infer product sales from other observed data. When selecting a tels from the time they joined Ctrip.com until the day of our data
proxy for this study, we noted the policy of Ctrip (the online travel collection, which gave a total of 40,424 user reviews. Table 2 pro-
agency from which we obtained the data) that only customers who vides a summary of the data.
have successfully made online bookings are allowed to post a We aggregated the customer review information to form aver-
review (one per transaction) after their stay. Therefore, the number age review ratings for analysis. Previous studies have suggested
of published reviews on a travel website might be taken as a proxy that very few customers ever view comments beyond the first
for sales of a hotel’s rooms through the travel agent during the two web pages (Pavlou & Dimoka, 2006). To control for limited
study period. Assuming customers have a constant probability of consumer attention, we established a moving window of 20 re-
posting a review on the travel agent website, the number of online views over the study period. On each day, we identified the 20
reviews for hotel i can be expressed as a linear function of room most recent reviews and calculated the average and standard devi-
sales of the hotel, where number Rev iews ¼ /  e sales. Essentially, ations of the review ratings. We then averaged the average ratings
we can take the number of reviews as a proxy for the number of and standard deviations over the study period to reflect the aver-
bookings for the hotel and use the following regression model for age consumer opinion of the hotel and the variability of the
the analysis. opinion.
Table 3 presents the variables for the hotels included in the
lnðnumber Rev iewsi Þ ¼ l þ b1 mean rev iew Rating i
analysis.
þ b2 v ariance rev iew Rating i
þ b3 lnðpricei Þ þ b4 city GDP Rank
4. Statistical results
þ b5 Stars þ ei ð2Þ
In the model, city_GDP_Rank is the rank of the city, where the We checked the multi-collinearity of the model. No significant
hotel is located, in terms of GDP. That represents the size of the multi-collinearity problem existed. Table 4 presents the main
host city. Stars represent the star category of the hotel, usually
from one-star to five-star, or no-star. The objective of the analysis
is to identify b1 and b2 , i.e., the effects of online reviews on sales. Table 2
Summary of the dataset.

3.2. Empirical data Item Value

Name of the travel agency Ctrip.com
The data used in this study were retrieved from Ctrip.com (Ctrip Number of selected cities 10
International, Ltd., NASDAQ: CTRP), which is the largest online tra- Total number of hotels in the dataset 2205
vel service provider for hotel accommodation, airline tickets, and Number of hotels not available for online booking on Ctrip.com 566
Number of hotels after deleting the unavailable hotels 1639
package tours in China and had a net revenue in 2007 of RMB
Number of customer reviews 40,424
1.12 billion (164.4 million in US$) (Crtip.com Intl Ltd., 2008). We
 Q. Ye et al. / Computers in Human Behavior 27 (2011) 634–639 637

Table 3
Brief introduction of the variables in the dataset.

Variable Description
Number_Reviews The number of reviews published for a certain hotel between February 1, 2007 and January 31, 2008 on Ctrip.com. This represents the 12 months
immediately before the data collection period (a proxy for the number of bookings made through Ctrip.com for a hotel)
Ave_20 For each review, we calculated the mean of the 20 reviews immediately preceding the review
Var_20 For each review, we calculated the variance of the 20 reviews immediately preceding the review
Mean_Rate The average of Ave_20 for all reviews of a hotel between February 1, 2007 and January 31, 2008 on Ctrip.com
Variance_Rate The average of Var_20 for all reviews of a hotel between February 1, 2007 and January 31, 2008 on Ctrip.com
Price The lowest price of the hotel room listed on Ctrip.com
City_GDP_Rank A discrete variable that denotes the rank of a city among the 10 cities studied according to its GDP in 2007
Stars The star-rating category of a hotel given by an official organization according to the quality of the hotel

Table 4 managers in Harbin city in March 2008 and January 2009. With
Results of the analysis. the assistance of the Bureau of Tourism, the government agency
Model Unstandardized coefficients T Sig. in charge of regulating hotels, we sent out a survey questionnaire
to obtain data on actual hotel bookings made through Ctrip.com
b Std. error
from all major hotels in the city. We received responses from 27
(Constant) 1.405 0.314 4.481 0.000***
hotels that have contracts with Ctrip.com. These hotels represent
Mean_Rate 0.529 0.024 21.640 0.000***
Variance_Rate 0.052 0.050 1.053 0.292 more than 50% of all of the hotels in Harbin that have contracts
Ln(Price) 0.102 0.056 1.836 0.067* with Ctrip. The survey collected data on monthly hotel bookings
Stars 0.038 0.015 2.538 0.011** through Ctrip.com for each of the hotels. We then used a fixed
City_GDP_Rank 0.199 0.011 17.592 0.000*** effect log-linear regression to identify the relationship between
Number of observations 1639
the number of reviews and the number of hotel bookings. To avoid
R-square 36.9%
Adjust R-square 36.7% the problem of taking the log of 0, we added 1 to both the number
Sig of model (p) 0.000 of reviews and the number of online bookings.
Dependent variable: Ln(Number_Reviews + 1) as the proxy of Ln(e_Sales). LnðNumber Rev iewsi þ 1Þ
***, ** and * denote significance levels at 1%, 5% and 10%, respectively.
¼ g þ aLnðNumber Online Bookingsi þ 1Þ þ ei ð3Þ
results of the analysis, and shows that there is a significant rela- where NumReviewsi represents the number of online reviews of
tionship between the independent variables and the dependent hotel i posted onto online travel agent during the 1-year period
variable at the 0.01 level, with an R-square of 36.9 percent. The re- from February 2007 to January 2008; and NumOnlineBookingsi is
sults also show that Hypothesis 1 is supported at the 0.01 level, the number of online bookings for hotel i during the same time
which implies that the positive sentiment of online reviews on period.
the website of an online travel agency can significantly increase Statistical results in Table 5 show a significantly positive rela-
the number of online bookings made through the website. Accord- tionship between the number of reviews and the number of hotel
ing to the cognitive theory, the cognitive load of online travelers bookings, with an R-square value of 44.1%. This indicates that the
should be dramatically reduced after reading positive reviews, number of online reviews of a hotel on the website of a travel agent
which results in more sales (Ghose & Ipeirotis, 2006; Sweller, can be used as a proxy for the number of online bookings that the
1988). The b1 ¼ 0:529 suggests that a 10 percent improvement in hotel receives from that agent.
the rating of online user reviews increases the index of room sales,
Ln(e_Sales), by more than five percent. 6. Implications
In terms of Hypothesis 2, the result of p = 0.292 shows that a
higher variance in the opinion polarity among hotel reviews does The main theoretical implication of this research relates to
not result in fewer online bookings for that hotel. This implies that filling in the knowledge gap of the influence of online user-gener-
travelers take into account the overall opinion, even when some ated reviews on business performance (online sales) in the tourism
reviews offer very different opinions. industry. The intangible nature of tourism products makes it diffi-
Ln(Price) has a significantly negative impact on the number of cult to evaluate before consumption, and it has long been recog-
online bookings at the 0.1 level (p = 0.067), which indicates that nized that interpersonal communications are an important
room rate has a negative impact on online room sales and that less information source among tourists (Litvin et al., 2008). With the
expensive hotels tend to have more online bookings. City_rank has spread of the Internet, virtual interactions among consumers have
a significantly negative impact on the number of online bookings at
the 0.01 level (p = 0.000), which shows that hotels in larger cities
Table 5
tend to have more online bookings. In terms of star-rating,
Statistical results for the validation of the proxy.
p = 0.011 indicates that the star rating of hotels significantly influ-
ences room sales, and the negative sign of b5 implies that higher Model Unstandardized T p-value (sig)
coefficients
star ratings are associated with fewer bookings online in China. A
possible explanation for this is that most online customers of B Std. error
hotels in China are educated young travelers, who prefer comfort- (Constant) 1.767 0.832 2.125 0.044**
able rooms at a reasonable price to luxury five-star hotels. Ln(Number_Online_Bookings) 0.566 0.127 4.437 0.000***
R-square 0.441
R 0.664
5. Validation of the proxy variable Number of observations 27
Sig of the whole model, p 0.000***

To validate our approach of using the number of online reviews Dependent variable: Ln(Number_Reviews).
as a proxy for hotel sales, we conducted two surveys with hotel -***, ** and * denote significance levels at 1%, 5% and 10%, respectively.
638 Q. Ye et al. / Computers in Human Behavior 27 (2011) 634–639

become commonplace, which has led some tourism researchers to Rapid resolution of consumer online reviews can provide an
point out that online word-of-mouth plays an important role in the opportunity for hotels to gain consumers’ trust, leading to addi-
acquisition and retention of consumers in the e-commerce era tional future businesses.
(Litvin et al., 2008; Vermeulen & Seegers, 2009). In addition, Sigala
(2009) pointed out that customer participation is a very important
7. Conclusions
dimension for e-service evaluation. However, most previous
studies in this area were conducted at an individual level or in
We have analyzed the user reviewing practices on Ctrip.com, a
experimental environments, and that the impacts of online user-
major online travel agency in China, and found that the valence of
generated reviews at the firm level in real business circumstance
traveler reviews had a significant impact on the online sales of
still need to be further tested. This study contributes to knowledge
hotel rooms. Online reviews may serve to reduce the cognitive load
development in tourism by revealing the positive impact of review
of potential travelers, and thus increase their awareness, resulting
valence on the online room sales as well as validating a new
in more sales. Our regression estimates suggest that, generally, a
approach to determine the relationship between online room sales
10 percent increase in the ratings of user reviews can boost the
and online reviews. The findings of this study open up numerous
dependent variable, index of online hotel bookings, by more than
avenues for future research on this promising application of
five percent. The results also indicate that the variance in the
e-tourism.
valence of rating scores across reviews does not significantly influ-
This study has valuable managerial implications for tourism
ence the number of online bookings. In terms of other influential
practitioners. As most tourism products are both seasonal and
factors for online sales, as the control variable in the research mod-
perishable, the effective marketing of these products requires
el, we find that room rate has a significantly negative effect on the
efficiency. E-commerce has presented tourism managers with a
average number of online bookings, and that hotels in larger cities
new means to improve marketing efficiency. The emerging of
tend to receive more online bookings.
web 2.0 and online user-generated content impact not only on
This study contributes to the tourism literature by revealing the
the behavior and decision-making of Internet users, but also on
influence of online consumer-generated reviews on online sales of
the e-business models that organizations need to develop or
hotel rooms at the firm level. The results suggest that online user
adapt in order to conduct business via the Internet. Tourism
reviews have a significant impact on online hotel bookings, and
managers thus need to be aware that an increasing number of
confirm the importance of online word-of-mouth for tourism firm
travelers will make their purchases online, and that the purchase
performance, as has been found in other industries like online
decisions of these travelers will be strongly influenced by online
retailing (Chevlier & Mayzlin, 2006) and movie industry (Duan
reviews. The results of this study demonstrate that online word-
et al., 2008) Hotel managers should therefore seriously consider
of-mouth has a significant impact on online room sales, while
the online reviews of their hotels that are posted on the websites
positive reviews can bring more line sales. This implies that if
of travel agents. Another contribution of this study is the validation
a hotel can provide more resources to improve the valence of
of the new proxy to identify the number of online bookings made
its online consumer reviews, it is likely to receive more book-
through travel agent websites, which could facilitate future studies
ings. In their studies, Au, Buhalis, and Law (2009), as well as
based on online data.
Au, Law, and Buhalis (2010) found the proliferation of review
There are some limitations in this study. These limitations relate
sites have enabled consumers to post their views online and
to both methodology and data, which deserve more future research
make them transparently to the entire world. In other words,
efforts. For instance, to introduce the proxy for hotel online sales, we
these review sites become an important source for international
assume that customers have a constant probability of posting a
travelers to get travel-related advice, which in turn, leads to the
review on a travel agent website. Though the validity of this proxy
ultimate online reservations. It also implies that the proactive
was supported by empirical data, this assumption is still too strict
use of online word-of-mouth information could prove to be an
from a theoretical perspective. Future studies can be carried out to
important competitive advantage for early adopters. Online
investigate the possibility of loosing this assumption and re-test
reviews can pass on a strong sense of a hotel or destination to
the validity of the proxy. As well, although this study shows that
travelers, and, more importantly, contain valuable information
user-generated reviews have a significant impact on business
about product improvement. The proper use of traveler reviews
performance, we cannot be sure whether all reviews have the same
could, therefore, help a tourism business to improve their ser-
impact. Also, the authenticities of reviews are not considered either.
vices and gain a competitive edge. There are multiple channels
It would thus be useful to analyze the different influence or per-
of available online communications, including online word-
ceived usefulness of different reviews. Furthermore, the effect of
of-mouth, email, websites, blogs, information databases such as
the content of reviews on the performance of hotels merits investi-
Wikipedia, chatrooms, virtual communities, and virtual worlds
gation using web-mining technologies to further understand the
(such as ‘‘Second Life”), all of which could be cost-effective ways
impacts of online reviews. Findings of this study were based on the
for hospitality and tourism firms to engage in customer relation-
data collected from a website in China. Therefore, the research find-
ship management.
ings may not be generalized. Still, this suggests directions for further
Lastly, hotel practitioners should motivate consumers to write
research. Finally, the cross sectional analysis did not reveal time var-
online reviews as Au et al. (2010) stated that posting reviews
iance characteristics, a panel data analysis in the future would be
can lead to product/service quality improvement. Similarly, Yoo
beneficial to gain more insight into this topic.
and Gretzel (2008), on the basis of their empirical findings,
claimed that review writers are mostly motivated by helping a
travel service supplier, concerns for other consumers instead of Acknowledgements
voicing out negative feelings as the primary reason for posting.
Harrison-Walker (2001), however, stated that many tourism The authors would like to thank the three anonymous review-
and hospitality businesses have not taken notice of the online ers for their constructive comments on the earlier versions of this
e-complaint trend, and they thus fail to respond actively. Since paper. This study was partially funded by the Hong Kong Poly-
the Internet will serve as a major communicate channel between technic University, NSFC (70971033, 70890080-70890082) and
hotels and consumers, hotel managers should setup an effective NCET-08-0172. The authors would like to thank Ms. Wen Shi and
communication strategy and to simplify their electronic reviews. Mr. Lei Xiao for their help with the survey and data collection.
 Q. Ye et al. / Computers in Human Behavior 27 (2011) 634–639 639

We are grateful to the Bureau of Tourism and the participated ho- Katz, E., & Lazarsfeld, P. (1955). Personal influence. The part played by people in the
flow of mass communications. New York: Free Press.
tels in Harbin City for providing valuable assistance in this study.
Klein, L. R. (1998). Evaluating the potential of interactive media through a new lens:
search versus experience goods. Journal of Business Research, 41(3), 195–203.
Litvin, S. W., Goldsmith, R. E., & Pan, B. (2008). Electronic word-of-mouth in
References hospitality and tourism management. Tourism Management, 29(3), 458–468.
Nelson, P. (1970). Information and consumer behavior. Journal of Political Economy,
Anderson, E. W. (1998). Customer satisfaction and word-of-mouth. Journal of Service 78(2), 311–329.
Research, 1(1), 5–17. Pan, B., MacLaurin, T., & Crotts, J. (2007). Travel blogs and the implications for
Au, N., Buhalis, D., & Law, R. (2009). Complaints on the online environment: The destination marketing. Journal of Travel Research, 46(1), 35–45.
case of Hong Kong hotels. In Höpken, W., Gretzel, U., & Law, R. (Eds). Pavlou, P. A., & Dimoka, A. (2006). The nature and role of feedback text comments in
Information and communication technologies in tourism 2009. (pp. 73–85) online marketplaces: Implications for trust building, price premiums, and seller
New York: Springer-Verlag. differentiation. Information Systems Research, 17(4), 392–414.
Au, N., Law, R., & Buhalis, D. (2010). The impact of culture on eComplaints: Evidence Pew Internet and American Life Project. (2006). Internet Activities. Available from
from Chinese consumers in hospitality organisations. In Gretzel, U., Law, R., & http://www.pewInternet.org/trends/Internet_Activities_7.19.06.htm. Accessed
Fuchs, M. (Eds). Information and communication technologies in tourism 2010. on Aug. 7, 2008.
(pp. 285–296) New York: Springer-Verlag. Sigala, M. (2008). Web 2.0, social marketing strategies and distribution channels for
Cheung, C.M.Y., Shek, S.P.W., & Sia, C.L. (2004). Virtual community of consumers: city destinations: enhancing the participatory role of travellers and exploiting
Why people are willing to contribute. Proceedings of the 8th asia-pacific their collective intelligence. In M. Gascó-Hernández, T. Torres-Coronas (Eds.),
conference on information systems (pp. 2100–2107) Shanghai, China. Information communication technologies and city marketing: digital opportunities
Chevlier, J. A., & Mayzlin, D. (2006). The effect of word-of-mouth on sales: online for cities around the world. IDEA Publishing.
book reviews. Journal of Marketing Research, 43(3), 345–354. Sigala, M. (2009).E-service quality and Web 2.0: expanding quality models to
Compete, Inc. (2007). Consumer Generated Content: Learning from Travel include customer participation and inter-customer support. The Service
Innovators. Available from https://media.competeinc.com/med/uploads/files/ Industries Journal, 29(10), 1341–1358.
traveltrends_consumer_generated_travel_content.html. Accessed on Sept. 8, Sigala, M., Lockwood, A., & Jones, P. (2001). Strategic implementation and IT:
2008. Gaining competitive advantage from the hotel reservations process.
Crtip.com Intl Ltd. (2008) Annual Report. Available from https://library.corporate-ir. International Journal of Contemporary Hospitality Management, 13(7), 364–371.
net/library/14/148/148903/items/295205/CTRIPCOMINTERNA20F.pdf. Accessed Smith, D., Menon, S., & Sivakumar, K. (2005). Online peer and editorial
on Sept. 20, 2008. recommendations, trust, and choice in virtual markets. Journal of Interactive
Dellarocas, C. (2003). The digitization of word-of-mouth: promise and challenges of Marketing, 19(3), 15–37.
online feedback mechanisms. Management Science, 49(10), 1407–1424. Stokes, D., & Lomax, W. (2002). Taking control of word-of-mouth marketing: The
Dickinger, A., & Mazanec, J. (2008). Consumers’ preferred criteria for hotel online case of an entrepreneurial hotelier. Journal of Small Business and Enterprise
booking. In P. Peter O’Connor, W. Wolfram Höpken, U. Ulrike Gretzel (Eds.), Development, 9(4), 349–357.
Information and communication technologies in tourism 2008 (pp. 244-254) New Sweller, J. (1988). Cognitive load during problem solving: effects on learning.
York: Springer-Verlag. Cognitive Science, 12(1), 257–285.
Duan, W., Gu, B., & Whinston, A. B. (2008). Do online reviews matter? – An empirical Travelindustrywire.com. (2007). Travel Reviews – Consumers are Changing your
investigation of panel data. Decision Support Systems, 45(4), 1007–1016. Brand and Reputation Online. Available from http://www.travelindustry-
Forrester Research. (2006). The State Of Retailing Online 2006: The 9th Annual wire.com/article29359.html. Accessed on July 20, 2008.
Shop.org Study. Published by Shop.org, June 2006. Available from http:// Tripadvisor.com. (2006). Fact Sheet. Available from http://www.tripadvisor.com/
www.clickz.com/3611181. Accessed on Dec. 16, 2008. pages/factsheet/html. Accessed on July 20, 2008.
Ghose, A., & Ipeirotis P. (2006). Towards an Understanding of the Impact of Vermeulen, I. E., & Seegers, D. (2009). Tried and tested: The impact of online hotel
Customer Sentiment on Product Sales and Review Quality. In Proceedings of the reviews on consumer consideration. Tourism Management, 30(1), 123–127.
workshop on information technology and systems (pp. 1-6). Milwaukee, USA. Ye, Q., Law, R., & Gu, B. (2009a). The impact of online user reviews on hotel room
Goldenberg, J., Libai, B., & Muller, E. (2001). Talk of the network: A complex systems sales. International Journal of Hospitality Management, 28(1), 180–182.
look at the underlying process of word-of-mouth. Marketing Letters, 12(3), Ye, Q., Zhang, Z., & Law, R. (2009b). Sentiment classification of online reviews to
211–223. travel destinations by supervised machine learning approaches. Expert Systems
Govers, R., & Go, F. (2005). Projected destination image online: Website content with Applications, 36(3), 6527–6535.
analysis of pictures and text. Information Technology & Tourism, 7, 73–89. Yoo, K. H., & Gretzel, U. (2008). What motivates consumers to write online travel
Gretzel, U., & Yoo, K. (2008). Use and impact of online travel reviews. In P. O’Connor, reviews? Information Technology & Tourism, 10, 283–295.
W. Höpken, & U. Gretzel (Eds.), Information and communication technologies in Zhu, F., & Zhang, X. (2006). The influence of online consumer reviews on the
tourism 2008 (pp. 35–46) New York: Springer-Verlag. demand for experience goods: The Case of Video Games. In Proceedings of
Harrison-Walker, L. J. (2001). E-complaining: A content analysis of an internet twenty-seventh international conference on information systems (ICIS). (pp. 367–
complain forum. Journal of Service Marketing, 15(5), 397–412. 382)Milwaukee, USA.