• No results found

The Impact of Negative Online Customer Feedback on the sensemaking process and legitimacy strategy: A Case Study of the Hospitality Industry

N/A
N/A
Protected

Academic year: 2021

Share "The Impact of Negative Online Customer Feedback on the sensemaking process and legitimacy strategy: A Case Study of the Hospitality Industry"

Copied!
35
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The Impact of Negative

Online Customer Feedback on

the sensemaking process and legitimacy strategy:

A Case Study of the Hospitality Industry

Dirk Talsma S3030741

MSc BA Organizational & Management Control Under the supervision of Mr. N. Mangin

and co-assessor Ms. J. S. Gusc 20 June 2017

(2)

Abstract

Due to tremendous technological advances and recent developments in recent decades, a new form of traditional worth-to-mouth has evolved: electronic worth-of-mouth (EWOM). Although EWOM is subjective by nature, it has become an important source for customers to purchase services. The impact of negative EWOM on organizational legitimacy has been excessively described. However, the sensemaking process initiated by EWOM and the development of strategies to improve legitimacy that often follow, have not been assessed. The current study aims to find and clarify contextual factors that are associated with the sensemaking process enacted by EWOM and find links to strategies to improve legitimacy in the hospitality industry.

A multiple case study was conducted by performing semi-structured interviews with six general managers and seven divisional managers from six hotels (2 privately owned, 4 operated by a chain). The interviews included questions regarding how and by whom EWOM is checked; value of EWOM; and reaction to EWOM. Cases were compared based on size (i.e. large versus small) and ownership structure (i.e. chain operated versus privately owned).

Results demonstrated that contextual factors such as opinion about EWOM, sensemaking in a group or alone, trust, and perception influenced the sensemaking process. When managers interpreted the content of EWOM within their control (i.e. sensitivity), they tended to develop strategies to create monitors and to disassociate. Furthermore, when the nature of EWOM was perceived as beyond their control (i.e. precision), managers tended to follow strategies to deny or to excuse/justify.

(3)

Introduction

Reviews generated by customers are difficult to ignore. Due to tremendous technological advances and development in recent decades, traditional word-of-mouth has expanded beyond family and friends. By using the Internet, reviews reach thousands of current and future customers (Torres, Adler, Behnke, Miao, & Lehto, 2015). This new form of exchanging reviews, also referred to as electronic word-of-mouth (EWOM), is posted publicly for the world to see (Litvin, Goldsmith, & Pan, 2008). Potential customers consider these reviews to be authentic, trustworthy, helpful, and influential when making a purchase decision (Li & Hitt, 2008). EWOM is increasingly shaping the generalized perception of customers regarding an organization or brand (Schivinski & Dabrowski, 2016). Positive reviews are considered as a form of free advertising and increase business potential, while negative reviews are considered harmful and may severely destroy the image and reputation of an organization (Chen & Law, 2016).

Although reviews are important to (potential) customers, they are increasingly topic of debate due to their subjective nature. Reviews generated by customers may not always be a representative measure of the quality provided by an organization. Aspects such as cultural background (Au, Buhalis, & Law, 2009) age, income (Wildes & Seo, 2001) educational level (Keng, Richmond, & Han, 1995), and expectations (Ofir & Simonson, 2001) influence appraisals in reviews. As such, values, norms, and beliefs of respondents may differ based on demographic factors and expectations. Furthermore, respondent biases make reviews difficult for organizations to address and might pose an incorrect representation of the organization due to a self-selection bias among respondents. In 2014, Tripadvisor was fined €500.000 for not preventing to control misleading and false reviews while presenting them as authentic and genuine (Neild, 2014).

When the content of received EWOM is negative, it may result in loss of both new and existing customers, and may also damage the brand (Cheng & Loi, 2014). Moreover, consistent negative reviews expose weaknesses in operations and strategy (Melián-González, Bulchand-Gidumal, & González López-Valcárcel, 2013). Thus, EWOM influences the generalized perception of services and/or products provided by an organization. When EWOM is negative, actions of the specific organization may be interpreted as undesirable and not appropriate within the socially constructed environment of the (potential) customer. Eventually, this may result in a dispute regarding legitimacy for the concerned organization. When legitimacy is at risk, an organization must improve their public image to remain attractive for customers (Parsons, 1960).

(4)

tries to understand situations that have occurred in the past or that could be occurring in the present and anticipate for the future (Tillmann & Goddard, 2008). By trying to make sense of the situation, an actor consciously or unconsciously constructs meaning about received EWOM. Sensemaking produces new scripts for appropriate behaviour and response (Louis, 1980). This initiated process could determine output. Understanding and delineating this process unveils the reasons why actors produce certain actions (Weick, 1995). Indeed, organizational output can be observed, but it is far more interesting to understand which contextual factors influence the sensemaking process initiated by actors when making decisions (Starbuck & Milliken, 1988).

Literature acknowledges the impact of EWOM on business performance and operational procedures (Torres et al., 2015), but do not explain nor elaborate on the process of sensemaking as a reaction to negative EWOM. Also, it has not yet been described which contextual factors influence the sensemaking process that lead to strategies to improve legitimacy. The current study aims to investigate the process of sensemaking initiated by negative EWOM and the followed sensemaking process resulting in legitimacy improving strategies, thus providing novel theoretical insights and knowledge on which contextual factors influence sensemaking in organizations and the relation between managers’ attitudes and behaviours, and their choice of strategy to improve legitimacy.

Thus, the main research question in the current study is:

Do contextual factors influence how managers make sense of negative EWOM and act upon it to improve legitimacy?

To answer this question, two sub-research questions are proposed:

1) Which contextual factors influence the sensemaking process initiated by negative EWOM? 2) How does the sensemaking process result in strategies to improve legitimacy initiated by negative EWOM?

(5)

The remainder of this paper is ordered as follows. First, the literature review discusses results of previous studies regarding the relevant topics such as EWOM, sensemaking, and legitimacy. Second, the methodology section describes how data was gathered by semi-structured interviews in the hospitality industry and processed. Third, the result section presents the results concluding which factors shape sensemaking processes and connects the sensemaking process to strategies to improve legitimacy. Fourth, the discussion puts the findings into perspective and suggests conjectures supported by literature. Fifth, theoretical and managerial contributions will be put forward and limitations of the current research and suggestions for future research are presented. Lastly, the conclusion summarizes the main findings.

Literature review

Several studies have shown that EWOM affects organizational decision-making (Duan, Yu, Cao, & Levy, 2016; Melián-González et al., 2013; Torres et al., 2015). In 2015, Torres et al., (2015) conducted a quantitative study to examine the impact of EWOM on organizations in the hospitality industry. Their findings showed that managers use EWOM to modify practices but failed to explain how the implemented change came to be and connect it to actual strategies.

Online customer feedback

Nowadays, customer feedback is often expressed online. The phenomenon of EWOM evolved from traditional word-of-mouth, which is defined as “informal face-to-face communications between private parties and their sharing of evaluations of products, services or organizations’’ (Arndt, 1967). Word-of-mouth is considered by customers to be trustworthy since they trust the opinion of the sender (Murray, 1991). EWOM started with the rise of the Internet and allows people to generate digital content and share information online (Hsu & Hsu, 2008). However, EWOM differs from traditional word-of-mouth. For instance, a broader range of readers can view reviews and reviewers are unknown to the receiver (Litvin et al., 2008). Moreover, EWOM is quickly and easily available over time and is no longer restricted to direct contacts of the receiver (Melián-González et al., 2013). Websites provide platforms for consumers to post and spread EWOM beyond their inner circle (Torres et al., 2015).

(6)

Negative EWOM however, is considered as critical comments that pose a threat to the legitimacy of an entity. Vermeulen & Seegers (2009) found that negative EWOM generates negative attitudes and repetition of these reviews can be highly damaging for an organization. Furthermore, Zhao, Wang, Guo, & Law (2015) reported a significant negative correlation between negative online reviews and online purchase intentions. Therefore, it can be concluded that negative EWOM has an unfavourable effect on purchase behaviour from both potential and existing customers and thus poses a threat to an organization’s legitimacy. The impact of EWOM on customers has been extensively investigated (Erkan & Evans, 2016; Mortazavi, Rahim Esfidani, & Shaemi Barzoki, 2014) but the impact of EWOM on employees and their behaviour remains underexposed. Thus, in the current study, only negative EWOM is addressed, as negative EWOM is more likely to trigger the sensemaking process.

Legitimacy

Above-mentioned paragraphs describe the importance of EWOM on business performance. If EWOM is negative, the question may arise whether the organizations is able to provide quality to customers. Such negative EWOM could therefore affect an organization’s legitimacy.

(7)

of an organization’s most immediate audience. The pragmatic approach can be considered as exchange legitimacy: stakeholder support for an organization based on the expected value to this stakeholder (Dowling & Pfeffer, 1975). Strategies according to the pragmatic approach are to deny and to create monitors. The strategy to deny entails that a legitimacy disruption is non-existent or not valid in the eyes of the organization. Unless these denials are sincere, denying problems damage organization’s legitimacy in the long term. This strategy only functions when the issues are accepted by the organization’s immediate audience. Furthermore, management can also implement a strategy to create monitors as part of a restructuring process to see if the issue is lasting. Examples of this could be grievance procedures or external counselling. Secondly, a moral approach can be selected that focuses on positive normative evaluation. In contrast to the pragmatic approach, the moral approach is not intended to benefit the organization, but relies on judgement if activities are the right thing to do. Strategies specified in this approach are to excuse or justify the issue and a strategy to restructure: disassociation. By excusing or justifying an issue, an organization questions the moral responsibility towards the customer. By blaming external factors, an organization seeks legitimacy in the fact that customer expectations are not realistic and therefore not their fault. It redefines means and ends to make the disruptive event fit into their moral beliefs. When this is no longer possible, an organization can choose to disassociate themselves from the issue by blaming, and eventually replacing, specific personnel and practices. Thirdly, a cognitive approach may be opted. Legitimacy in this approach is based on available cultural models to provide plausible explanations for organizational actions and beliefs. Explaining, as a strategy, can be implemented to clarify the disruptive event in a way that preserves an otherwise supportive worldview.

All aforementioned approaches and their related strategies label the specific procedures an organization pursues when their legitimacy is jeopardized. Legitimacy management rests heavily on communication and awareness of organizational members to connect disruptive events with the appropriate response (Elsbach, 1994).

Sensemaking

As management is hired to successfully operate the organization and manage day-to-day activities, negative EWOM may be considered as a violation of managements’ capabilities to run the organization appropriately. To estimate if and to what extent organizational legitimacy is at stake, a manager must construct meaning to decide on further action. This process is also referred to as sensemaking.

(8)

the social world” (Gephart, 1993). Moreover, sensemaking is a process that actors consciously or unconsciously undergo to understand situations (Hasan & Gould, 2001). Sensemaking takes place when individuals encounter a situation that is unknown and existing scripts do not provide appropriate behaviour (Choo, 1989). Therefore, the concept of sensemaking focuses on the invention that precedes interpretation (Weick, 1995). EWOM may invoke a process of sensemaking when an individual is new to the feedback received. By making sense of the situation, the actor can judge how he or she values received EWOM and if there is a potential threat to an organizations’ legitimacy.

Weick (1995) argued that distinguishable features characterize the process of sensemaking. Two of these features are ‘social’ and ‘plausibility rather than accuracy’.

Sensemaking, till this point, was described as an individual level of analysis. However, sensemaking is regarded as a social process and dependent upon other actors such as members of an organization in which sensemaking activities take place. Louis (1980) concluded that sensemaking often happens during socialization. When social factors influencing sensemaking are overlooked, theoretical assumptions can be misleading. Sensemaking is therefore never solitary because actions are always contingent on others (Weick, 1995). However, it should be stressed that, even though sensemaking is a social activity, the level of socialization differs and influences the weight is has in the sensemaking process (Weick, 1995).

Moreover, plausibility rather than accuracy deals with the bounded rationality of the decision maker involved in a sensemaking process. Bounded rationality implies that an actor can only process a limited amount of information, which leads to flaws in perception (Gigerenzer & Selten, 2002). This in itself is not a problem when dealing with sensemaking. It is important to acknowledge that the process of sensemaking produces a relative approach to the truth and is based on what the actor finds attractive, appealing, and relevant (Fiske, 1992). Thus, sensemaking is a process that applies personal characteristics to understand situations. It is interesting and worth investigating which contextual factors invoke sensemaking and to what conclusion they lead (Gigerenzer, 1991). The current research aims to find the contextual factors that influence the sensemaking process and observe if they lead to output in the form of strategies.

(9)

received EWOM could initiate a process of sensemaking to construct meaning to EWOM and, potentially, try to restore legitimacy by determining to adjust strategy. Current theories and literature however, fail to explain if contextual factors influence this process and identify them. To my knowledge, the contextual factors that influence the behavioural process of sensemaking initiated by EWOM as a precursor of legitimacy improving strategies has not been investigated. This process is visualized in Figure 1. The current study aims to close this gap in existing literature by providing theory development. The present study will provide a contribution to the literature regarding the contextual factors that influence the process of sensemaking when dealing with issues concerning legitimacy and how these decisions lead to strategy.

Figure 1: The process of sensemaking invoked by negative EWOM into strategies to improve legitimacy.

Methodology

To answer the main- and sub-research question, a multiple case study in the hospitality industry was conducted. The hospitality sector provides an interesting real-life context to observe the phenomenon of EWOM and the process of sensemaking and the followed strategies to improve legitimacy. When potential customers seek to purchase services from an unknown entity, such as an unvisited hotel, EWOM becomes a critical determinant of a customer’s willingness to make a booking (Grewal, Gotlieb, & Marmorstein, 1994). Hotels have a clear and hierarchical structure, which clearly appoint responsibilities to managers to construct and implement strategy. Thus, this setting provides a unique environment to investigate the research questions of current study.

Sample description Case selection

To gain a better and broader understanding of the sensemaking process invoked by negative EWOM and strategies to improve legitimacy, multiple cases were selected to examine why and how certain changes came to be in its real-life context. A multiple case study provides a richer understanding of the context and does not constraint data collection to a pre-set theoretical framework (Blumberg, Cooper, & Schindler, 2014). Within the hospitality industry, hotels were the most suitable option since they deal with EWOM every day and

Negative  

EWOM  

making  

Sense-­‐

Value  of  

EWOM  

(10)

ratings influence their revenue (Ye et al., 2009). Therefore, legitimacy is important for the selected cases and the effect of EWOM was directly observed.

Cases were observed to analyse the process of sensemaking that resulted in strategies

to improve legitimacy. By selecting multiple cases, the risk of possible organizational specific factors that influence the findings and occurrence of possible relationships was reduced. Also, by including more cases, differences among cases were observed due to contrasting attributes that are discussed later.

The cases were selected based on the amount of received reviews, size, rating, and ownership construction. Tripadvisor is a world-renowned travel site, providing travel-related information. Booking.com is a hotel search and booking engine-website. Both are international platforms where customers leave reviews. Results from the current study are based on management’s experiences with EWOM from these platforms.

I selected cases based on specific inclusion criteria (Table 1). First, the hotel must have had received enough reviews on Tripadvisor and Booking.com to ensure the presence of negative EWOM. Also, enough reviews must be present to prevent ambiguity since an organization might be in the process of gaining legitimacy instead of improving legitimacy. Melián-González et al. (2013) found that early reviews tend to be disproportionally negative and, as the number of reviews increase, ratings become balanced and the negative effect is

mitigated. Thus, hotels with a number of

60 received reviews per year were included.

Second, the hotel must have multiple divisions (i.e. food and beverage, housekeeping, front office, and banqueting) and a minimum of 15 employees in operational departments to ensure that the size of operational departments is large enough to see how a hotel handles negative EWOM and (possibly) adjust strategy. Third, the rating of the selected cases must be similar. When ratings fluctuate across cases, it will pose inadequate standards to compare them. Lastly, cases were selected on ownership structure. In the current research, chain owned hotels were compared with privately owned hotels to investigate whether this influences the sensemaking process and final decision in choosing a strategy to improve legitimacy and why. Chain hotels with centralized and regulated control may enforce specific regulation to process EWOM and thereby influencing the process of sensemaking. In contrast, private hotels with decentralized control and less imposed regulations may handle EWOM in a different way.

(11)

Table 1: Inclusion criteria Inclusion criteria

≥ 60 received reviews per year (TripAdvisor/Bookings.com)

Multiple divisions and minimum of 15 employees Similar ratings

Ownership structure (private vs chain)

Respondent selection

In the current study, general managers and divisional managers were interviewed. General managers are the managing director of a hotel and divisional managers supervise their specific department. The respondents from the current study were selected based on their responsibility and authority towards stakeholders within or outside the organization. Because of their position as managers, they are responsible for leading the organization and its success. Therefore, EWOM should affect them as their function requires to act upon disruptive events. Also, by interviewing multiple actors in each case, the results are not dependent on one, potentially bias, actor.

Data collection

Eligible cases were approached by formal letter inviting them to participate. The letter detailed the subject of the study as well as the assurance of anonymity (Appendix 1). After the letters were sent, all cases were called to include them in the current study. A total of 16 invitations were sent. Ten of the approached cases declined the invitation due to lack of interest and/or lack of time. Six cases were included. Table 2 provides an overview contacted and included cases.

Table 2: An overview of contacted cases, included cases and respondents

(12)

Of the six included cases, four were operated by a chain and two were private. There were no pronounced differences between excluded and included cases (Appendix 2).

Details per case are summarized in Table 3. All included cases met all inclusion criteria. Similar ratings based on received EWOM from Tripadvisor were found. Mean Booking.com-rating of chain hotels and private hotels were 8.1 and 8.0 respectively. In addition, enough EWOM was received so that the presence of negative EWOM is eminent.

Table 3: An overview of the included cases and their details

Cases Ownership structure Size* Rating** (Booking.com/Tripadvisor) Average received EWOM per year*** Case 1 Private 15 8 – 8.5 / 4

70 Case 2 Private 60 7,5 – 8 / 4

80 Case 3 Chain 35 8 – 8,5 / 4

600 Case 4 Chain 45 8 – 8,5 / 4

100 Case 5 Chain 62 8 – 8,5 / 4

160 Case 6 Chain 95 7,5 – 8 / 4

70

* Based on number of employees

** Ratings were obtained on 5/15/2017. Booking.com (Booking.com.2017) rates on a scale from 4 to 10, Tripadvisor (Tripadvisor.2017) rates from 1 to 5. Grades are rounded off to secure anonymity.

*** Amount of received EWOM is rounded to secure anonymity.

Prior to selecting and interviewing various respondents, the general manager was approached to ask for approval. At each case, a minimum of one general manager and one divisional manager were interviewed. The interviews were conducted in a five-week time span at the premises of the case and lasted between 30 and 45 minutes. The interviews were conducted to gather data on how managers perceived EWOM, if the process of sensemaking was invoked due to this, what contextual factors influenced this, and if a strategy to improve legitimacy was observed. A semi-structured interview was chosen because the interviews were partly used to confirm information the researcher already holds, such as the essence of received reviews, but were also intended to ask open and broad questions. All interviews were recorded and retranscribed to capture as much data as possible.

In general, the topics of the semi-structured interview were: 1)

(13)

4) reaction to EWOM. An overview of the asked questions are provided in the interview guide in Appendix 3.

The interviews started with more factual questions and slowly progress to questions of a less factual nature. By confirming facts first and later proceeding to personal views and motives, I aimed to address the less sensitive and more objective topics first before proceeding to more sensitive issues. Interviewees were quoted anonymous since shared information could be personal and data could be of value to competitors. By presenting data anonymous, interviewees were encouraged to provide more honest and detailed information. These measures lessened social desirable answers and lowered a potential social desirability bias.

The interview questions were open and non-specific in order to the interviewees responded freely on the process of sensemaking. Also, leading and suggestive questions were not asked to reduce the influence of the interviewer. Follow-up questions were used to elaborate further on the subject and the clarify answers to prevent miscommunications.

Because multiple cases were selected, and multiple respondents were interviewed, I aimed to view the topic from different perspectives and unravel the impact of EWOM and the sensemaking process leading to implemented strategies. By doing this, I aimed to have a faithful representation of the actors involved.

Data analysis

All transcripts were coded with the qualitative data analysis program Nvivo (QSR International, version 11). Nvivo facilitated to compare codes and themes across cases and to find and support broad observations. All transcripts were coded three times separately and compared to see if coded material differed and why. This moderately ensured inter-coding reliability.

Thereafter, the primary data from interviews was analysed by using within-case analysis as recommended by (Eisenhardt, 1989b).

After transcribing the interviews, coded segments were assigned so that cases could be compared. The initial themes were based on previous literature (i.e. legitimacy strategies and sensemaking) and thus based on a theoretical lens that was chosen to answer the research question. Other themes emerged as the coding of the transcripts proceeded. Themes as ‘trust’, ‘incentivised versus not incentivised’, and ‘personal perception versus automated control’, and ‘content of EWOM’ were added.

After the broad analysis, attribute values were used to create a framework matrix to see if case specific results emerged indicating further patterns. To separate the cases, the attribute values of ‘size’ (i.e. small versus large) and ‘ownership structure’ (i.e. chain versus

(14)

50 employees). This resulted in the matrix depicted in Table 4. The analysis phase continued by looking for broad cross-case patterns that could indicate connections among themes (Eisenhardt, 1989b). This was done by a quantitative approach using to number of code segments per theme.

Table 4: Cases ordered in a matrix based on attribute values as size and ownership structure Size Small Large Ownership Structure Chain Case 3 Case 4 Case 5 Case 6

Private Case 1 Case 2

The patterns found for the current study are based on the number of coded segments that support a theme. Due to the different number of respondents and cases within each cell of the matrix, the total number of code segments was adjusted. This was done for the ‘small chain’ and ‘large chain’ cells by dividing the coded segments by two since both cells have two assigned cases. Also, at ‘case 2’, two divisional managers were interviewed. The codes resulted from these two interviews were divided by 2 to make them comparable. By adjusting the number of code segments, a weighted average was calculated for each cell to find differences among ownership structure and size.

The results were organized based on the total number of code segments. A total of 445 code segments codes were assigned and the total adjusted code segments totalled 274. Code segments were structured in themes. The themes, code segments and total number of codes are presented in the codebook in Appendix 4.

Lastly, the results from the cross-case were compared to the raw data to check if interpretations are correct. It should be noted that obtained results from the current study are generalizable through testable propositions and based on theoretical arguments that help explain the decision behaviour of managers (Blumberg et al., 2014). Moreover, all codes used in the result section were adjusted for the different numbers of cases and respondents.

Results

(15)

Patterns from initial themes

One of the patterns found in initial codes is the opinion about EWOM. Out of the 37 coded segments, 24 indicated a negative opinion towards EWOM, 2.5 coded segments a neutral opinion and only 10.5 positive segments. Most managers at first noted positive points about EWOM. As one manager indicated:

“Without it, it would be blank. It is the future and I’m happy it’s there. […] It makes us lose our tunnel vision. Some things you miss because we are here every day”. Divisional manager, case 6.

Later on, the same manager noted:

“Reviews surpassed what they are intended for. Guests visit us for service and to socialize, not to check if all the light work. It is a shame that this is all reviews are about these days”. Divisional manager, case 6.

Similar patterns were observed across 9 interviews; views are positive at the beginning of the interview and later on, changed into a negative point of view. Subjectivity of reviews is often mentioned as one manager described:

“Not too long ago, we had a fire in the hotel. At that time, guests couldn’t check in which led to negative reviews” General manager case 6.

Additionally, 42 segments were coded concerning the topic ‘sensemaking in a group versus sensemaking alone’. Out of these 42 segments, 32 reported sensemaking in a group. When looking at these figures into detail, it also became clear that when making sense of EWOM alone, large hotels had 8 coded segments whereas small hotels had only two.

A total of 64 code segments were assigned to the strategies to improve legitimacy, out of which 31 to the pragmatic approach, 28 to the moral approach and only 5 to the cognitive approach.

(16)

Furthermore, the moral approach had a total of 28 coded segments almost equally divided between the strategy to excuse/justify (13 coded segments) and to disassociate (15 coded segments). Variation was observed between the chain hotels who chose to excuse/justify more than private entities; 8.5 coded segments compared to 4.5 coded segments. Also, private hotels tend to disassociate more than chain hotels. Chain hotels noted 4.5 coded segments and private hotels 10.5.

Patterns from emerging themes

While coding interviews, specific emergent themes were found. For instance, trust played an important role in the process of sensemaking. When dealing with negative EWOM and making sense on the issue at hand, 36.5 codes were assigned to the theme ‘trust’. Out of these 36.5 codes, 26 were positive, 8.5 were neutral, and only 2 were negative. These results indicate that managers have a large amount of trust in the personnel when assessing the credibility of negative EWOM. As one manager put it:

“I have a well-organized team and I trust them to do their work. We have a low turnover ratio in comparison with other hotels. The balance within the team is excellent. I only use positive reviews when I look at how well employees are performing”. Divisional manager, case 5.

This indicates that the quoted manager has a high level of trust in his employees. He only uses positive reviews for evaluation and negative EWOM is not taken into account when performance is evaluated. Furthermore, there were no observable differences when comparing small and large cases and ownership construction.

Also, the themes ‘incentivised versus not incentivised’ and ‘personal perception of EWOM versus automated control’ produced interesting results. All the general managers at chain hotels indicated that there are monetary incentive structures connected to the ratings based on EWOM. In contrast, none of the private hotels have these structures in place. Further, all chain hotels have tools to receive, measure, and construct graphs based on ratings produced by EWOM. Private hotels do not use these tools but rely on own observations to see patterns and reoccurring topics.

Another emergent code was the verifiability of EWOM that indicates if managers could check if disruptive events took place. A total of 17 coded segments demonstrated that management could verify content of negative EWOM. Out of the 17 coded segments, 14.5 were from chain hotels and 2.5 were from private hotels.

(17)

A total of 60 coded segments were assigned to the topic of content of EWOM. The interviews concluded 39.5 coded segments where management perceived as if they could have had influenced the matter discussed in received EWOM. A total of 20.5 coded segments implied that management perceived the disruptive event as beyond their control.

The coded segments that demonstrated EWOM issues that management could influence, did not result in new insights when the division between small and large cases and ownership structure was applied. This was the same for coded segments that expressed EWOM within the control of management.

Thereafter, the results concerning the content of EWOM were linked to legitimacy repair strategies. The outcomes are depicted in Table 5.

Table 5: Content compared to strategies to improve legitimacy

Strategies to improve legitimacy

Content Pragmatic approach Moral approach Cognitive

approach

Deny Create

monitors

Excuse/justify Disassociate Explain

Total/chain/private T C P T C P T C P T C P T C P Beyond control of management 6 3.5 2.5 0 0 0 9.5 5.5 4 0 0 0 0.5 0.5 0 Within control of management 0.5 0 0.5 4.5 1 3.5 1 0.5 0,5 15 4.5 10.5 1 0 1

Table 5 demonstrates that, when management’s perceptions about the content of the received EWOM is linked to strategies, the moral approach is mostly followed. When adding up code segments, the moral approach consists of 25.5 coded segments in total followed by the pragmatic approach with 11 coded segments and the cognitive approach with 1.5 coded segments. It can thus be observed that the managers most often follow the path of positive normative evaluation. The managers question the moral responsibility they have towards their guests and, if this is no longer possible, decide to change practices to restore legitimacy. Within each approach, patterns were found that are connected to specific strategies.

Content of EWOM beyond the control of management leading to the strategy to ‘deny’ and ‘excuse/justify’

(18)

monitors are set up to see if the EWOM is lasting and, eventually, might lead to a response to solve the problem (i.e. disassociate).

When managers chose to deny, received EWOM is often perceived as beyond their reach and in their sense not valid. For example, one manager described a received review:

“Sometimes I read reviews that are nonsense. There were bird droppings on the balcony. There was a chair on the balcony that was wet. I cannot prevent this”. General Manager, case 1.

Moreover, the option to excuse or justify is preferred over the option to deny. The reaction in the interviews note that received EWOM that is excuses or justified is true, but cannot be changed by management. For instance, one manager in a hotel outside of the city centre noted that:

“I think it is unfair that we get graded on location. Guests that book our hotel know our location when they decide to make a reservation. I don’t think it is honest to get graded on something we cannot change”. Divisional manager,

case 3.

This manager realized that the EWOM is in fact true; their hotel is outside of the city centre. He justified the low rating be commenting that a factor like location is not new and could have been known by the guest before booking.

Content of EWOM that is within control of management leading to strategies to ‘create monitors’ but mostly to ‘disassociate’

Based on the conducted interviews, content of EWOM that managers can control is connected to the strategy to create monitors but mostly to disassociate. When managers had control over the underlying causes of received EWOM, they tend to create monitors to see if the problem is lasting, as described by a manager in the following way:

“If reviews keep coming in, we will construct an action plan on the basis of received EWOM”. General Manager, case 2.

In this sense, creating monitors can be viewed as a predecessor of actual change: to disassociate.

(19)

difference.

Creating monitors and the option to disassociate can be seen as forms of action. It is evident that if a manager can change protocols to prevent negative EWOM, they do take action. This claim is supported by the fact that passive strategies (i.e. to deny or to excuse/justify) are almost never followed when dealing with disruptive events managers can control. In the division between large and small cases, no conclusions could be formulated.

Cognitive approach

Compared with the pragmatic and moral approach, the cognitive approach is not often used by the included cases. Only 1.5 code segments have been assigned to the cognitive approach linked to strategies which seems negligible compared to the 11 codes assigned to the

pragmatic approach and 25.5 from the moral approach.

Discussion

The discussion is structured as follows: first the opinions regarding EWOM are discussed where after the sensemaking themes are explained. Lastly, the results from legitimacy strategies are interpreted and coupled with themes concerning the content of EWOM.

The first conclusion is that most hotel-managers have a negative perception of EWOM. During the interviews, it became evident that most managers experienced the subjectivity of EWOM. The codes listed under ‘deny’ support this claim. The negative opinion towards EWOM could be clarified by the agency theory that discusses the agency relationship in which the principal delegates responsibilities to the agent (Ross, 1973). Agency theory tries to resolve two problems that frequently occur in agency relationships, namely; if the goals of the principal and agent conflict and when it is difficult or expensive for the principal to verify what the agent is actually doing (Eisenhardt, 1989a). One could argue that the goals of the principal (i.e. the owner or chain head office) and agent (i.e. the manager) in the observed cases match: run a profitable organization that delivers excellent service. However, the second goal poses a dilemma. The quality of intangible provided services by hotels is for the principal difficult to track. In the observed cases, principals use EWOM as a reference to verify what the agent is actually doing. As EWOM tends to sketch a misrepresentation of the actual service provided, as is explained in the introduction, it seems logical that agents (i.e. interviewed managers) consider EWOM as something negative. Managers’ opinion about EWOM captures believes and expectations used with regards to what they define as relevant (Weick, 1979). Therefore, the current study suggests that the negative opinion towards EWOM influences the sensemaking process.

(20)

However, the level of socialization differs among the cases. When managers were involved in the process of sensemaking alone, often the received EWOM concerned a major complain that needed to be dealt with as soon as possible as was indicated:

“In theory, complaints in review should pass from general manager to manager to employee. In practice, I’ll skip these management layers by sending an email to all employees when I notice behaviour that is totally unacceptable.” General manager, case 2.

Moreover, reoccurring external factors in which managers did not need any additional confirmation about the nature of the issue, were also often dealt with alone. Sensemaking alone was most observed in large hotels. The bounded rationality of decision makers could influence the choice to value EWOM alone. Large organizations tend to be more complex and management might have difficulties to comprehend the full scale of the issue at hand since the number of actors is relatively large compared to small organizations. When negative EWOM is received, management tries to resolve disruptive events as soon as possible and thus have limited time to find solutions which in turn could lead to the decision to act alone (Gigerenzer & Selten, 2002). This also implies that sensemaking in a group is less complicated in small organizations. This suggestion is also supported by Holt & Macpherson (2010) who concluded that successful small organizations use and benefit from sensemaking in groups. Thus, my results confirm that in the observed cases, sensemaking is often a social process but suggests differences based on the size of organizations and the disruptive event that initiated the process.

Furthermore, this study shows that trust is an important factor for managers when making sense of received EWOM. When confronted with negative EWOM about employees, management portrayed a large amount of trust in employees that made the severity of EWOM less credible. Considering the negative opinion concerning EWOM as illustrated earlier, the feedback provided could be less valued due to the lack of trust in the system (Earley, 1986). Sensemaking is about plausibility (Weick, 1995) and when managers construct meaning, predictability of what happened could influence the process. Trust and predictability are clearly related (Mayer, Davis, & Schoorman, 1995) and therefore the current study suggests that trust influences the construct of meaning and lowers the credibility of received EWOM. Research by Borges & Gonçalo (2010) also recognize the interrelationship between sensemaking and trust.

Furthermore, the results concerning automated control versus personal perception

(21)

tools, managers can operate efficient and manage faults more easily (Röttger, Bali, & Manzey, 2009). However, these tools can be costly and require large investment for private hotels. Also, centralized business structures, such as chain hotels who report to head office, often require centralized control to provide overhead functions at the main office with figures for budgets and benchmarking activities (Jones, 2010). At stand-alone private hotels, figures could be more difficult to compare to similar hotels. This could explain the lack of automated tools at private hotels. Besides the initial investment as an argument for the presence of automated control, the information system itself also played a role when making sense of EWOM:

“We use a tool that gathers and categorizes reviews. […] With this tool we can find reoccurring themes and not base our opinion on one review” General manager, case 5.

The tool structures reviews in such a way that it enhances the process of sensemaking for the general manager in case 5. The presence of these tools also provides an explanation why chain hotels can verify EWOM more easily than privately owned hotels. Since reoccurring themes are found and more data is available than can be benchmarked, managers can construct meaning by placing received EWOM in context to see if it stands out or not. Also, the presence of automated tools suggests why private hotels create monitors more often than chain hotels. The automated tools are in itself platforms that can be used create monitors and see if negative EWOM is consistent. Results from the interviews suggest that automated tools provide managers with additional information that helps them to construct meaning and thus influences the sensemaking process.

This argument could be stretched further and be applied to the incentive structures. With no form of automated control, it would be difficult to incentivise ratings if they cannot be compared to similar hotels to put ratings in perspective. Private hotels could choose to not incentivise ratings based on EWOM because they might be partly based on uncontrollable events and therefore not represent actions taken by managers. EWOM, as a performance measure, could be considered ‘noisy’ since their composition includes factors that cannot be changed by management and are thus not reliable to base incentives upon (Feltham & Xie, 1994). Therefore, incentives based on EWOM do not seem rational. Also, no observations from the interviews suggest that the process of sensemaking is enacted differently due to incentive payments.

(22)

constructive to view the observations using economic literature. Building on research aimed at analysing managerial performance evaluation systems, these categorizations could also be labelled as sensitivity (i.e. actions within the control of the manager) and precision (i.e. the lack of noise, in the context of this paper interpreted as uncontrollable events beyond managements control) (Banker & Datar, 1989). Although derived from the economical field of research, this concept could explain the sensemaking process of managers in the current study. While connecting the before mentioned classification to sensemaking processes, it is important to note that the interpretation of EWOM rests on managements perception and is thus subject to personal factors. Therefore, what managers regard as an issue beyond their control, might be viewed by others as within their reach. For instance: in a situation where managers chose to deny (as in; not valid), the EWOM could be interpreted as valid. For instance, a general manager indicated the following:

“We received negative reviews about the state of our building; I don’t think this is true and besides, to invest in the property is not my decision”. General Manager, case 2.

The general manager assesses complaints regarding the state of the building not valid and choses to deny the review. Moreover, he noted that the state of the building is beyond his control. Whether true or not, one could argue that, as a general manager, he could actually invest in the state of the building and it therefore could be within his reach. By perceiving a controllable event as uncontrollable, this point of view shows a distortion in causal attributions and thus in perception (Starbuck & Milliken, 1988). This demonstrates that the separation of content into ‘sensitivity’ and ‘precision’ influences the sensemaking process and response that lead to strategies to improve legitimacy.

When looking at EWOM concerning precision, that leads to deny and excuse / justify, it is logical that factors that an organization cannot change, are not subjected to specific strategies that result in activities such as to create monitors or to disassociate. From a chain perspective, negative EWOM could be denied, excused and/or justified since it does not fit into the rules and regulations set by the centralized institution to which they comply. Expectations from guests towards chain hotels could be more explicit since they consciously chose for a specific chain. Private hotels chose to excuse/justify almost as often as chain hotels. This may be explained by the fact that private hotels decide their own moral responsibility towards the guest and are thus free to decide what to excuse and/or justify. Private hotels have no obligations towards regulation specified by a centralized head office like chain hotels have.

(23)

Torres et al. (2015) who demonstrated that hotels do take action based on EWOM. The finding concerning sensitivity shows proactive activity towards issues that can be solved by management. Even with the negative opinion concerning EWOM, the included cases do resound to strategies to improve legitimacy by taking action. When managers can change issues, they do. Interesting is to see that, when confronted with issues concerning sensitivity, almost no managers chose to deny and excuse / justify the matter. Moreover, it is interesting to see that private hotels chose to disassociate more often than chain hotels. This could be because private hotels do not have chain policy and can change and alter procedures without confirmation from head office, like a chain hotels have to.

Lastly, the results indicate that the cognitive approach was only found in 5 code segments in total and only 1.5 when connected to strategies to improve legitimacy. The cognitive approach rests on cultural models to explain disruptive events. In the hospitality industry, guests visit with different cultural backgrounds that indicate different expectations of service (Armstrong, Mok, Go, & Chan, 1997). Therefore, the cultural models that are applicable for the hotels might not comply with the cultural models of the guest. This discrepancy might be a valid reason not to act upon reviews for managers, but does resolve legitimacy towards the guest. Cultural models consist of culturally derived practices that are embodied and enacted in everyday life (Fryberg & Markus, 2007). Hotels that welcome international guests might distantiate from these cultural practices to ensure an inviting place to stay for guests with different cultural backgrounds.

Theoretical implications

The current study investigated the contextual factors influencing sensemaking invoked by EWOM and aimed to link this process with legitimacy improving strategies. It extends the existing literature regarding sensemaking as it provides actual ex post facto reactions in terms of strategy based on sensemaking activities. From another angle, it prescribes the antecedent of legitimacy improvement strategies what was not a subject of empirical research before.

Sensemaking

(24)

Legitimacy

Suchman (1995) presented several approaches and strategies for improving legitimacy. Suchman’s framework does not explain why individual actors pursue these strategies in a real-life context. By observing the phenomenon of EWOM, the current study adds a theoretical contribution by suggesting the existence of the relationship between the content of EWOM (i.e. precision and sensitivity) and strategies to improve legitimacy.

Managerial implications

The current study implies that automated control benefits managers in the process of sensemaking because it increases the verifiability of EWOM. Moreover, my results suggests that incentive measures based on EWOM should not be applied since they are not representative of managers performance and do not benefit the process of sensemaking.

Limitations

The current study had several limitations in addition to the well-known limitations of theory development in the form of a case study. At each case, at least two managers were interviewed to include multiple opinions about EWOM, thus limiting the risk of potential respondent bias. It should be stresses however, that although the number of two may prevent this personal view, any personal biases may not be excluded since the managers work together and might share the same views due to their ongoing and regularly contact. In addition, social desirable answers cannot be totally excluded, since negative EWOM may be considered as a shortcoming of manager’s competence. In such cases, managers could refuse threats to organizational legitimacy and neglect the impact of negative EWOM. Also, provided answers might pose a more positive image than reality. This may have resulted in neglected incidents regarding threats to legitimacy. Hence, the strategies followed could underrepresent the actual number of strategies observed.

Moreover, most of the interviews were conducted by a single person. The coding, even though it was done three times and compared, was also done by one person. Therefore a researcher bias may not be excluded. The concluded findings might be influenced by the author’s personal opinion.

(25)

Since discrepancy was made between large versus small hotels and chain versus non-chain hotels, some cells in the matrix ended up with only one case. Cells with multiple cases included showed a weighted average of coded segments. Cells with only one case included could be subjected to company specific details and opinions and thus not be representable for the group in which they are categorized. Therefore, results based on the cells with one included case could be inconsistent.

Moreover, sensemaking is a process that is reliant on the individual that enacts his or her environment (Weick, 1995). Therefore, the personality of the manager could, and probably does, influence the sensemaking process. The current study did not take this consideration into account. Hence, included managers that could be labelled as ‘ignorant’ might have denied legitimacy issues where ‘insecure’ managers might have opted to disassociate sooner.

Future research

Negative EWOM influences the process of sensemaking that may affect the choice in strategy to improve legitimacy. The current study was performed within the hospitality industry, and results may thus not be applicable in other industries. To understand the influence of negative EWOM on the sensemaking process and legitimacy improving strategy, a similar research should be conducted in a different industry to observe if the results hold. The hospitality industry provides mostly intangible services that could be topic for discussion. When a similar study is conducted in an industry that sells tangible products, the contextual factors of sensemaking and legitimacy strategies could differ.

Moreover, among the included cases, chain hotels differ compared to private hotels especially in incentive measures based on EWOM and automated control versus personal perception. Chain hotels, characterized by their centralized nature, incentivised EWOM and use automation to track and interpret the received EWOM. Private hotels did not incentivise and rely more on personal perception. The current study does not find differences in the sensemaking process due to incentives. Therefore, an interesting proposition to investigate this phenomenon further would therefore be: automated tools enhance managerial control and verifiability of EWOM.

(26)

To address the limitations of limited cases and number of respondents, future research could benefit from including more cases and respondents. Ideally, multiple cases should be included in every cell of the matrix. The matrix could also be extended by adding more attributes such as star qualification and years of management employment to see if this results in new insights. Preferably, the future research should be performed and conducted by multiple researchers so that interviews can be conducted, interpreted, and coded using different views, thus limiting the risk of researcher bias.

Conclusion

The current study finds that contextual factors influence the sensemaking process enacted by managers in reaction to EWOM. Findings demonstrate that the opinion towards EWOM, sensemaking in a group or alone, perception of EWOM, and trust influence the sensemaking process. Specifically, all included cases reported a negative opinion towards EWOM. Also, sensemaking in group form was mostly observed. However, when making sense alone, it was mostly reported in large organizations. The perception of EWOM revealed that chain hotels all used automated tools to track EWOM and privately owned hotels relied on personal perception. Furthermore, all included cases reported trust as an important factor when dealing with EWOM.

(27)

References

Ackerman, R. W. (1975). The social challenge to business. Cambridge, MA: Harvard University Press.

Armstrong, R. W., Mok, C., Go, F. M., & Chan, A. (1997). The importance of cross-cultural expectations in the measurement of service quality perceptions in the hotel industry.

International Journal of Hospitality Management, 16(2), 181-190.

Arndt, J. (1967). Role of product-related conversations in the diffusion of a new product.

Journal of Marketing Research, 291-295.

Au, N., Buhalis, D., & Law, R. (2009). Complaints on the online environment—the case of Hong Kong hotels. Information and Communication Technologies in Tourism 2009, 73-85.

Banker, R. D., & Datar, S. M. (1989). Sensitivity, precision, and linear aggregation of signals for performance evaluation. Journal of Accounting Research , 21-39.

Blumberg, B. F., Cooper, D. R., & Schindler, P. S. (2014). Business research methods (4th ed.). Berkshire, United Kingdom: McGraw-hill education.

Booking.com. (2017). Retrieved from www.booking.com

Borges, M., & Gonçalo, C. R. (2010). Learning process promoted by sensemaking and trust: A study related to unexpected events. Cadernos EBAPE.BR, 8(2), 260-277.

Chen, Y., & Law, R. (2016). A review of research on electronic word-of-mouth in hospitality and tourism management. International Journal of Hospitality & Tourism

Administration, 17(4), 347-372. doi:10.1080/15256480.2016.1226150

Cheng, V. T. P., & Loi, M. K. (2014). Handling negative online customer reviews: The effects of elaboration likelihood model and distributive justice. Journal of Travel &

Tourism Marketing, 31(1), 1-15. doi:10.1080/10548408.2014.861694

Choo, F. (1989). Cognitive scripts in auditing and accounting behavior. Accounting,

Organizations and Society, 14(5-6), 481-493.

(28)

Duan, W., Yu, Y., Cao, Q., & Levy, S. (2016). Exploring the impact of social media on hotel service performance. Cornell Hospitality Quarterly, 57(3), 282-296.

doi:10.1177/1938965515620483

Earley, P. C. (1986). Trust, perceived importance of praise and criticism, and work

performance: An examination of feedback in the United States and England. Journal of

Management, 12(4), 457-473.

Eisenhardt, K. M. (1989a). Agency theory: An assessment and review. Academy of

Management Review, 14(1), 57-74.

Eisenhardt, K. M. (1989b). Building theories from case study research. Academy of

Management Review, 14(4), 532-550.

Elsbach, K. D. (1994). Managing organizational legitimacy in the California cattle industry: The construction and effectiveness of verbal accounts. Administrative Science Quarterly, 57-88.

Erkan, I., & Evans, C. (2016). The influence of eWOM in social media on consumers’ purchase intentions: An extended approach to information adoption. Computers in

Human Behavior, 61, 47-55.

Feltham, G. A., & Xie, J. (1994). Performance measure congruity and diversity in multi-task principal/agent relations. Accounting Review, 429-453.

Fiske, S. T. (1992). Thinking is for doing: Portraits of social cognition from daguerreotype to laserphoto. Journal of Personality and Social Psychology, 63(6), 877.

Fryberg, S. A., & Markus, H. R. (2007). Cultural models of education in American Indian, Asian American and European American contexts. Social Psychology of Education,

10(2), 213-246.

Gephart, R. P. (1993). The textual approach: Risk and blame in disaster sensemaking.

Academy of Management Journal, 36(6), 1465-1514.

Gigerenzer, G. (1991). How to make cognitive illusions disappear: Beyond “heuristics and biases”. European Review of Social Psychology, 2(1), 83-115.

(29)

Ginzel, L. E., Kramer, R. M., & Sutton, R. I. (1993). Organizational impression management as a reciprocal influence process: The neglected role of the organizational audience.

Research in Organizational Behavior, 15, 227-227.

Gretzel, U., & Yoo, K. H. (2008). Use and impact of online travel reviews. Information and

Communication Technologies in Tourism 2008, 35-46.

Grewal, D., Gotlieb, J., & Marmorstein, H. (1994). The moderating effects of message framing and source credibility on the price-perceived risk relationship. Journal of

Consumer Research, 21(1), 145-153.

Hasan, H., & Gould, E. (2001). Support for the sense-making activity of managers. Decision

Support Systems, 31(1), 71-86.

Holt, R., & Macpherson, A. (2010). Sensemaking, rhetoric and the socially competent entrepreneur. International Small Business Journal, 28(1), 20-42.

Hsu, J., & Hsu, C. (2008). The relationships between service quality and customer

satisfaction in a leading Chinese web 2.0 company. The Business Review, Cambridge,

11(1), 84-89.

Jones, G. R. (2010). Organizational theory, design, and change (6th ed.) Prentice Hall. Keng, K. A., Richmond, D., & Han, S. (1995). Determinants of consumer complaint

behaviour: A study of Singapore consumers. Journal of International Consumer

Marketing, 8(2), 59-76.

Li, X., & Hitt, L. M. (2008). Self-selection and information role of online product reviews.

Information Systems Research, 19(4), 456-474.

Litvin, S. W., Goldsmith, R. E., & Pan, B. (2008). Electronic word-of-mouth in hospitality and tourism management. Tourism Management, 29(3), 458-468.

Louis, M. R. (1980). Surprise and sense making: What newcomers experience in entering unfamiliar organizational settings. Administrative Science Quarterly, 226-251. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of

(30)

Melián-González, S., Bulchand-Gidumal, J., & González López-Valcárcel, B. (2013). Online customer reviews of hotels: As participation increases, better evaluation is obtained.

Cornell Hospitality Quarterly, 54(3), 274-283. doi:10.1177/1938965513481498

Meznar, M. B., & Nigh, D. (1993). Managing corporate legitimacy: Public affairs activities.

Business & Society, 32(1), 30.

Mortazavi, M., Rahim Esfidani, M., & Shaemi Barzoki, A. (2014). Influencing VSN users’ purchase intentions: The roles of flow, trust and eWOM. Journal of Research in

Interactive Marketing, 8(2), 102-123.

Murray, K. B. (1991). A test of services marketing theory: Consumer information acquisition activities. Journal of Marketing, 55(1), 10.

Neild, B. (2014). $611,000 fine as tripadvisor gets bad review in Italy. Retrieved from http://edition.cnn.com/2014/12/23/travel/tripadvisor-fine/

Neirotti, P., Raguseo, E., & Paolucci, E. (2016). Are customers’ reviews creating value in the hospitality industry? Exploring the moderating effects of market positioning.

International Journal of Information Management, 36(6), 1133-1143.

doi:10.1016/j.ijinfomgt.2016.02.010

Ofir, C., & Simonson, I. (2001). In search of negative customer feedback: The effect of expecting to evaluate on satisfaction evaluations. Journal of Marketing Research (JMR),

38(2), 170-182.

Parsons, T. (1960). Structure and process in modem society. New York, the Free Press of

Glencoe, 1, 960.

Rayman-Bacchus, L. (2006). Reflecting on corporate legitimacy. Critical Perspectives on

Accounting, 17(2), 323-335. doi:10.1016/j.cpa.2005.07.002

Ross, S. A. (1973). The economic theory of agency: The principal's problem. The American

Economic Review, 63(2), 134-139.

Röttger, S., Bali, K., & Manzey, D. (2009). Impact of automated decision aids on

performance, operator behaviour and workload in a simulated supervisory control task.

(31)

Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on

consumer perceptions of brands. Journal of Marketing Communications, 22(2), 189-214. Starbuck, W. H., & Milliken, F. J. (1988). Executives’ perceptual filters: What they notice

and how they make sense.

Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches.

Academy of Management Review, 20(3), 571-610. doi:10.5465/AMR.1995.9508080331

Tillmann, K., & Goddard, A. (2008). Strategic management accounting and sense-making in a multinational company. Management Accounting Research, 19(1), 80-102.

Torres, E. N., Adler, H., Behnke, C., Miao, L., & Lehto, X. (2015). The use of consumer-generated feedback in the hotel industry: Current practices and their effects on quality.

International Journal of Hospitality & Tourism Administration, 16(3), 224-250.

doi:10.1080/15256480.2015.1054754

Tripadvisor. (2017). Retrieved from www.tripadvisor.com

Vermeulen, I. E., & Seegers, D. (2009). Tried and tested: The impact of online hotel reviews on consumer consideration. Tourism Management, 30(1), 123-127.

Weick, K. E. (1979). The social psychology of organizing (topics in social psychology series) (2nd ed.) Random House.

Weick, K. E. (1995). Sensemaking in organizations Sage.

Wildes, V. J., & Seo, W. (2001). Customers vote with their forks: Consumer complaining behavior in the restaurant industry. International Journal of Hospitality & Tourism

Administration, 2(2), 21-34.

Ye, Q., Law, R., & Gu, B. (2009). The impact of online user reviews on hotel room sales.

International Journal of Hospitality Management, 28(1), 180-182.

Zhao, X., Wang, L., Guo, X., & Law, R. (2015). The influence of online reviews to online hotel booking intentions. International Journal of Contemporary Hospitality

(32)

Appendices

Appendix 1. Letter requesting interview

[Hotel name]

Att. [Name General Manager] [Address]

[Postal code, city] Groningen, [date] 2017

Subject: research hospitality industry Dear Sir/Madam,

Currently I am working towards a master’s degree in Organizational and Management Control at the Rijksuniversiteit Groningen. For my master thesis, I have chosen to investigate the hospitality industry and specifically the hotel market. Hereby, I would like to invite you for an interview about you and your working experiences.

This interview will serve five students who all study the hospitality market. A part of this interview will focus on inter-organizational relations and what factors affect these relations. My research will focus on reviews from platforms such as Booking.com and Tripadvisor. Considerable research has described the customer perspective of such reviews. However, the effects of these reviews on organizations are still underexposed. My research aims to find novel insights on this topic.

To gather enough data I would like to interview you in your function as general manager. This interview will not last longer than an hour. In addition, I would also like to interview an operational manager. The results will be anonymously used in the final paper. I would like to record the interview so that I can transcribe it later.

If you have any further questions regarding my study or other unanswered questions, you can always contact me by phone or email. After a few days, you can expect a phone call to inform whether you are interested to participate.

(33)

Appendix 2. Comparison between included and not included cases

* Ratings were obtained on 15/5/2017. Booking.com rates on a scale from 4 to 10, Tripadvisor rates from 1 to 5. Grades are rounded off to secure anonymity.

(34)

Appendix 3. Interview guide

Can you tell a little bit more about yourself and what you do? - Function? How long? Why this firm?

Can you describe the organization you work for?

- Amount of rooms, amount of personnel, brand, and ownership structure?

(Introduction about EWOM: from Booking.com and Tripadvisor) How do you keep track of EWOM?

- By whom? - How often? - Why in this way?

What is your opinion about the phenomenon of EWOM? How do you value EWOM?

- Do you do it alone or in group form? - Should it be reoccurring?

- Does EWOM only confirm your own perception or does it bring new facts to the table?

How do you use EWOM in your organization?

Do you think your colleagues are aware of EWOM and if so, in what way? - Do you discuss EWOM?

(35)

Referenties

GERELATEERDE DOCUMENTEN

The energy consumption and seek time were succesfully simulated for probe storage devices with a comb- drive scanner design and an electrodynamic scanner design using these

This study aimed to compare pre-drought herbaceous communities to drought release communities and thereby determine the response of herbaceous community

Intended users of exoskeletons have a series of user criteria, including height, weight, and health condition, in the case of rehabilitation.. By having rigid inclusion criteria

This study aims to broaden our understanding of the influence of power and politics on the sensemaking process during Agile teams development, and how a shared understanding

This thesis conducted its research at an innovative technology company who recently implemented an organizational innovation, which is described below. In order to gain a

18. Are you aware of the organizational policy towards absenteeism at your workplace? - Policies aimed at raising absenteeism barriers.. - Policies aimed at matching workload

Note: The dotted lines indicate links that have been present for 9 years until 2007, suggesting the possibility of being active for 10 years consecutively, i.e.. The single

 While the constructs defined in the literature review where shown to be important for a positive customer experience, the degree to which they need to be integrated in a website