From reactivity to reputation management: online consumer review systems in the restaurant industry

ABSTRACT

The rapid growth of online consumer review (OCR) systems such as Tripadvisor has greatly reconfigured the operating environment for numerous businesses and organizations. As OCRs become a crucial source of information for consumer decision-making, we pose a twofold question: how do restaurants perceive OCRs and how do they respond to being evaluated on them? In answering this question, we distinguish between different types of organizational responses: staff management, goal setting, operational practices and reputation management. We base our study on in-depth interviews with mid-price restaurants in Amsterdam. Our findings show that reactivity related to staff management, goal setting and operational practices is limited and highly deliberate. Instead, reputational responses are more extensive. That is, restaurants respond to and even appropriate OCRs in order to promote themselves, to signal professionalism, and to limit the reputational damage of negative reviews. On the basis of these findings, we argue that more attention be paid to the agency of evaluated entities and that OCR systems be theorized more as a multisided platform with a hybrid functionality of both valuation and marketing.

KEYWORDS:

• Rating
• ranking
• consumer reviews
• reactivity
• marketing

Introduction

The rapid growth of online consumer review (OCR) systems such as Tripadvisor (established in 2000) and Yelp (est. 2004) has greatly reconfigured the operating environment for numerous businesses and organizations. From wines (Vivino, est. 2010), movies (IMDb, est. 1990), jobs (Glassdoor, est. 2007) to professors (RateMyProfessors, 1999), a growing number of commodities, services, and people are being assessed by lay consumers online (Pinch and Kesler 2010, Mellet 2014, Orlikowski and Scott 2014). Most of these systems algorithmically synthesize consumer opinions that are left in the form of written reviews and ratings, into aggregate ratings and rankings.¹

According to Smith and Anderson (2016), 82% of American adults consult OCRs before purchasing new items and 40% do so almost always. As OCRs become a crucial source of information for consumer decision-making, they become increasingly relevant for businesses and organizations as well. It is therefore unsurprising that the majority of previous studies focus on consumers, asking when reviews are seen as useful (Park and Nicolau 2015), trustworthy (Ayeh et al. 2013, Filieri 2016) or which information is extracted from reviews (Bialecki et al. 2017), and when reviews lead to consumers buying the product (e.g. Chevalier and Mayzlin 2006, Ghose and Ipeirotis 2011).²

Instead, the number of studies which have focused on the producer’s perspective is more limited. As argued by Beuscart et al. (2016), while the impact of OCRs on consumers has been well studied and measured, few studies have examined how OCRs are received by those who are assessed by them (but see e.g. Mellet 2014, Orlikowski and Scott 2014). These studies have mostly seen reviews from the lens of valuation or from the lens of reactivity: i.e. how consumers use OCRs to evaluate goods & services and how these valuations relate to those of the producers of the goods & services themselves. Combining these two perspectives, Beuscart et al. (2016) find that restaurants react to being evaluated and alter operational practices although restaurants question the legitimacy of OCRs as valuation devices. By contrast, we find that reactivity is highly limited among restaurants; restaurants see OCRs less as evaluation devices which enable them to find points of improvement for their service and products. If they do react to OCRs, it is almost exclusively by turning them into devices for monitoring and managing their reputation.

This study looks into how the evaluated organizations make sense of the main OCR systems in the Dutch restaurant field (Tripadvisor, Google and Iens, a local platform which at the time of the interviews had just been acquired by Tripadvisor).³ The main, twofold research question we address is how do restaurants perceive OCRs and how do they respond to being evaluated on them? In answering this question, we distinguish between different types of organizational responses (in particular: staff management, goal setting, operational practices and reputation management). We chose the restaurant industry because of the proliferation of OCRs in this context. Moreover, together with the hotel industry, it is known to be the most susceptible to the impact of OCRs (as cited in Mellet 2014, p. 6). Amsterdam offered a relevant context for the study with its status as the biggest urban area in the Netherlands, its increasing popularity as a travel destination (Van Zoelen 2017), and its culinary sector which has recently been expanding and flourishing (Boterman 2018, Terhorst and Erkuş-Öztürk 2018).

The following two sections review relevant literature on OCR systems, reactivity, and reputation management (Section 2) and explain the context and design of research, as well as how data was collected and analyzed (Section 3). Section 4 answers to what extent restaurants change organizational practices on the basis of reviews, while sections 5 argues that OCRs are particularly relevant to restaurants because of their implications and the possibilities they offer for reputation management. The paper concludes with the call to go beyond the theoretical lens of reactivity and widen our conceptual understanding of OCR systems in order to obtain the full scope of what is happening on these systems (Section 6).

Literature review

OCR systems as judgment devices

OCR systems can be conceptualized as new forms of market devices (Muniesa et al. 2007) or, more specifically, valuation devices that allow consumers to share personal opinions on various products and services and thereby influence the valuation and decision-making of other consumers. For instance, Mellet et al. (2014) conceptualize French gastronomic OCR systems like LaFourchette as democratized alternatives for exclusive, expert valuation devices like the Michelin Guide. OCR systems are especially relevant to singular goods like films, books, and dining experiences. According to Karpik (2010), singular goods are characterized by their multidimensionality (their value cannot be assessed along a single criterion such as price), uncertainty (their value can only be known after purchase), and incommensurability (they cannot be ranked or aligned on a single dimension). For these reasons, consumers experience a cognitive deficit when they need to make a decision about singular goods whose quality is radically uncertain. Unlike homogeneous goods which are valued on a single dimension of price, singular goods require a different modality of valuation that prioritizes quality over price. Karpik (2010) names this modality of valuation judgment and various tools that facilitate this process judgment devices.

OCR systems become particularly pertinent to singular goods such as dining by providing consumers with powerful judgment devices that help them coping with their cognitive deficit. First of all, OCR systems constitute the most prevalent and accessible form of electronic word-of-mouth (eWOM) which influences consumers’ purchase intentions of products and retailers (Chatterjee 2001, Cantallops and Salvi 2014). Secondly, OCR systems are readily available, intuitive in character, greatly simplifying, and seemingly trustworthy because of their quasi-scientific and quantified format. As Jeacle and Carter (2011) put it in their study of Tripadvisor, OCR systems measure, rank, and order entities in a systematic fashion. This seemingly reliable, concise, and easy-to-use characteristic of OCR systems becomes especially helpful when consumers face the radical uncertainty of singular goods.

Nevertheless, OCR systems are also prone to fraud, in particular through fictitious reviews (Ayeh et al. 2013). This compromises their role as judgment devices, while their impersonal nature means that users cannot be sure that the reviews accord with their personal taste and evaluation of singularities (Bialecki et al. 2017). Moreover, OCR systems’ legitimacy as judgment devices is contested by restaurants themselves who sometimes see them as a ‘brutal and hypocritical mode of judgment’ (Beuscart et al. 2016, p. 472).

OCRs and reactivity

Irrespective of their contested legitimacy, the very fact that OCRs operate as powerful judgment devices for (prospective) consumers has far-reaching implications for businesses that receive assessments on them. Because of the potential disciplining effect of ratings and rankings, OCR systems are seen to constitute a form of governance (Shore and Wright 2015). That is, organizations or people are likely to change their behavior in reaction to being reviewed, rated, or ranked. In their seminal study, Espeland and Sauder (2007) refer to this idea as reactivity.⁴ In their case study of American law school rankings, they demonstrate how law schools redefine organizational goals, change work practices, reallocate resources, and develop ‘gaming strategies’ in order to perform better on the rankings. As a result, ranking, rating and review systems are likely to become self-fulfilling because the organizations that are being evaluated are likely to conform to the criteria that are used to construct the evaluation apparatus (Espeland and Sauder 2007).

In a similar vein, Beuscart et al. (2016) study how OCRs impact French restaurants. They find that OCRs do result in reactivity, but only when the required changes in response to reviews are concrete, relatively marginal, inexpensive, easy to implement, and do not contradict the fundamental business model. They argue, furthermore, that OCRs serve as a remote yet powerful monitoring tool as they allow owners and managers to ‘delegate surveillance to consumers’ (Beuscart et al. 2016, p. 467). However, in spite of all this, the authors find that restaurants do not perceive OCRs to be legitimate. In fact, restaurants consider OCRs to be malicious, hypocritical, and coming from unqualified sources. As a result, they find that ‘replying to an online review happens very occasionally, typically as a response to a negative review, in a fit of anger or indignation’ (Beuscart et al. 2016, p. 464). A full-fledged engagement is found only among a minority of their respondents either because restaurant owners do not feel comfortable with digital tools or do not see the need for positive reviews. Instead, as we will see, our own respondents generally make strategic use of the review systems in order to manage their reputations online.

Orlikowski and Scott (2014) ask what the impact of OCR systems on organizations is, as evaluations move online. They answer this question by comparing an online valuation device for hotels (Tripadvisor) with its traditional counterpart (the Automobile Association (AA) hotel scheme). The traditional, offline apparatus of valuation, they argue, is ‘formulaic’: it is based on ‘standards, principles, or prescriptions for achieving particular end' (Orlikowski and Scott 2014, p. 883). These standards are, moreover, well defined, professional, normative, enduring, and episodic (often annual). By contrast, the new, online valuation apparatus is understood as being ‘algorithmic’ because of its emerging, open-ended, situational, pluralistic, contradictory, dynamic, and continuous (almost real-time) character (Orlikowski and Scott 2014, p. 885).

According to Orlikowski and Scott (2014), this shift of evaluation apparatuses from being formulaic to algorithmic reconfigures organizational reactivity to valuations. Formulaic apparatuses evoke changes in behavior by imposing normative, isomorphic pressure to conform to explicit valuation criteria set by experts. The AA standards are established upon industry experience, made public, and practiced by trained inspectors once every 12–18 months. This disciplines and coerces hoteliers into developing strategies to conform to the AA standards, displaying reactivity in the strict sense of self-fulfilling prophecies as highlighted by Espeland and Sauder (2007, 2016). On the other hand, algorithmic apparatuses alter behavior in a dispersed, contingent, volatile manner driven by a distributed and disembodied crowd. Unlike a formulaic apparatus, an algorithmic apparatus gives hoteliers no predefined set of criteria to follow, no time for reflection between valuations, and thus little scope for strategic behavior.

While most studies on evaluation apparatuses have emphasized their disciplining or monitoring effects, not all studies find that the rise of ratings, rankings and other performance metrics results in reactivity among the organizations which are subjected to these metrics. Dorn (2019), for instance, argues that the interpretative processes underlying the use of rankings need to be scrutinized more systematically. In doing so for hospital rankings, he demonstrates that these rankings are frequently ignored altogether by patients and medical professionals alike. Moreover, he demonstrates that hospitals have significant leeway in responding to the rankings, of which endorsement is only one, and rejection and invalidation are others. Pollock et al. (2018) likewise stress the agency of ranked organizations in the ranking process. Instead of finding reactive conformance which is the paradigm of most previous studies on reactivity, they demonstrate that organizations can resist the pressure exerted by rankings, especially in a complex situation where multiple rankings coexist. In such a situation, organizations wield influence on the ranking systems by engaging strongly with some rankings while disengaging with others; negotiating their ranks, in some cases even by getting involved in shaping the evaluation criteria. Pollock et al. (2018) refer to these forms of agency in the ranking process, often developed by intermediary experts, with the term reflexive transformation. We draw on this notion by showing how Dutch restaurants respond to OCRs in a highly reflexive and deliberate manner. In that respect, we question Scott and Orlikowski’s (2012, p. 38, brackets in original) argument that Tripadvisor reconfigures accountability in the travel sector to the extent that it renders ‘the hotelier the (mostly) passive recipient of distributed judgement.’ In particular, we show that restaurant owners defy OCR systems’ role as a valuation device while actively endorsing them as a marketing device.

OCRs and reputation management

OCRs also impact an organization’s reputation, loosely defined as the perceptions that others have of an organization (Huang-Horowitz 2015).⁵ Reputation is a critical asset because it shapes an organization’s visibility, facilitates exchange, and bestows trust in the organization’s products and services. At the same time, reputation is a highly fragile asset and its loss affects an organization’s market value, competitiveness, and loyalty of stakeholders among others (as cited in Aula 2010, p. 44).

While organizational reputation used to be shaped largely by external experts, critics, or analysts, the growth of OCR systems has bestowed a part of this role of experts upon ordinary, lay consumers (Wang et al. 2016). In this light, OCR systems pose a new threat to reputation management. This is more so the case because critical reviews are considered to be more credible, altruistic, and crucial as well as more closely examined and responded to by businesses (Levy et al. 2012). Moreover, Hollenbeck et al. (2017) show that small, independent hotels are especially vulnerable, while larger, chained-brand hotels have the chain’s brand name to protect themselves against negative reviews. It is in this sense that it is not difficult to see media coverage of hospitality businesses that fold up their shops due to negative OCRs (see e.g. Kinstler 2018).

Recent studies demonstrate, however, that OCR systems also provide opportunities to organizations. They serve as an alternative, relatively cheap marketing channel that can enhance an organization’s visibility. This holds in particular for smaller organizations, who lack the financial and human resources for marketing and managing reputations which large firms possess (Huang-Horowitz 2015). A study by Hollenbeck et al. (2017) demonstrates that advertisement spending in the hotel industry has decreased because OCR systems have developed into an alternative way of managing reputation. Indeed, they find that the higher the rating of a hotel on Tripadvisor, the less it spends on advertising.

Apart from the boost to organizational reputation which a high rating or ranking or positive reviews provide, OCRs can provide organizations with opportunities for reputation management primarily by enabling them to respond to reviews. By responding, organizations can defend their reputation and appear attentive to consumer opinions. However, organizations also face the risk of drawing additional attention to negative issues and seeming defensive, self-justifying, manipulative, thereby alienating potential consumers. This trade-off results in organizations being more likely to engage in public responses when consumer devaluations are severe enough to make the benefits of defending their reputation outweigh the risks of negative publicity. Moreover, organizations are more likely to respond when they have a strong reputation to lose (Wang et al. 2016). Proserpio and Zervas (2017), by contrast, find that hotels are equally likely to respond to negative, neutral, and positive reviews. Moreover, they demonstrate that generally hotels are well advised to do so since responding to reviews results in ratings going up. Hotels who systematically respond to reviews moreover receive fewer negative reviews (albeit longer) because their presence online means reviews can be scrutinized.

While the threats and opportunities provided by OCRs have started to be analyzed in a quantitative way, we still know little about the reflexive processes of organizations in deciding when and how to respond to reviews. In order to address this lacuna, we conducted qualitative, in-depth interviews with restaurant owners in Amsterdam.

Methods

To understand how restaurant owners or managers perceive OCR systems and how they change organizational practices in response to OCRs, we took a qualitative approach based on semi-structured, in-depth interviews. Next to inquiring about their perception of OCRs, we asked our respondents to reflect on actual reviews of their restaurant and, when present, their responses to reviews. Prior to actual interviews, we conducted five pilot talks with a total duration of four hours to probe possibilities, adjust expectations, and orient our research. Interview topics were developed during these talks and changed iteratively throughout the study.

This study looks at mid-price restaurants in Amsterdam with least 30 reviews on either Tripadvisor or Google Maps. Our respondents’ restaurants had a price range of 15–25 euros for a main dish. We chose this price range as our main selection criterion for a number of reasons. Firstly, focusing on mid-price restaurants ensures that something is at stake for consumers, incentivizing them to invest time and energy on OCR systems before selecting a restaurant. Low-price snack bars and walk-in restaurants were excluded because we expected that consumers would be less bothered to refer to OCRs for those venues. High-end restaurants were also excluded because we expected their consumers to use other judgement devices with more expertise, authority, and legitimacy like the Michelin guide. In other words, we assume that competitive pressures exerted by OCRs are potentially the highest for this middle segment. These assumptions are in line with previous findings that ‘cheap establishments remain attractive regardless of their online evaluation’ and that fine dining restaurants ‘have sufficient reputation capital to not economically suffer or benefit from online reviews’ (Beuscart et al. 2016, pp. 462–463). Secondly, because mid-price restaurants are the most numerous in Amsterdam (73%),⁶ consumers are more likely to rely on OCR systems to make comparisons between ample choices of restaurants.

We restricted the geographical scope to Amsterdam to minimize the differences in structural conditions that may influence our findings such as climate, accessibility, economy, and tourism. Finally, we chose restaurants that have at least 30 reviews on either of the two larger OCR systems (Tripadvisor or Google Maps) to ensure a certain level of familiarity with the topic.

Between February and May 2018, we contacted 101 restaurant owners or managers in Amsterdam for an interview. Of these, 17 agreed to be interviewed (see Table 1). We focused on owners and managers and not on employees because they are responsible for strategic changes made on the organizational level as well as for the restaurant’s reputation management. The interviews lasted on average 38 min (the shortest 23 min, the longest 56 min).

Table 1. Some contextual information of restaurants in the data.

Of respondent		Of restaurant					As of 31st May 2018 (only when N ≥ 30)
ID	Position	Location	Age	Size (#)		Tourist %	TripAdvisor		Iens		Google		Facebook
				Seats	Staff		Rating	N	Rating	N	Rating	N	Rating	N
P1	Owner	West	16	90	8	40%	4	95	–	–	3.8	124	4.4	66
P2	Owner	West	6	40	6	30–40%	4.5	43	–	–	4.8	126	4.9	109
P3	Owner/Manager	Centrum	11	30	30	30%	4	138	–	35	4.3	200	4.4	151
P4	Owner	Centrum	1.5	75	17	50%	4.5	173	–	–	4.4	138	4.8	120
P5	Owner	Centrum	7	45	22	20%	4.5	292	8.4	78	4.4	185	4.6	96
P6	Owner	Oost	0.5	140	20–5	5%	–		–	–	4.4	92	5	94
P7	Owner	Centrum	4	45	31	40–50%	4.5	126	9	288	4.6	144	4.9	118
P8	Owner	Zuid	1	130	30	30%	4.5	58	–	–	4.1	178	4.6	157
P9	Manager	Oost	2	480	43	10%	4	51	–	–	4.1	200	4.6	67
P10	Manager	Centrum	1	50	6	50%	4.5	158	–	–	4.2	126	–	–
P11	Manager	Centrum	12	30–40	15	60%	4	435	8.6	169	4.2	151	4.3	162
P12	Owner	Centrum	1	45	10	50%	5	103	9	90	4.5	69	4.9	85
P13	Manager	Zuid	5	250–260	40–50	20–30%	4	58	8.2	51	4.1	222	4.6	98
P14	Owner	Zuid	2	50	9	60%	4.5	376	–	–	4.7	203	4.8	71
P15	Management trainee	West	1	50–60	20	30–40%	4.5	95	8.7	84	4.2	112	4.7	156
P16	Owner	West	3	100	11	10–15%	4.5	38	8.7	64	4.5	102	4.8	70
P17	Owner	Centrum	12	80	5	30–40%	4.5	190	9.3	353	4.6	60	4.8	86

^a N refers to the total number of reviews (includes ratings without comments).

^b The range of ratings on Google and TripAdvisor is 1–5, on Iens it is 1–10; the high ratings of restaurants are largely inherent to OCR systems in general (see Mellet 2014, pp. 27–28).

^c The number of seats, staff and the percentage of tourists are self-reported numbers of respondents.

Restaurant P3 is owned by a student foundation, with many students working few hours, and therefore has a large number of staff in proportion to the size of the restaurant.

Interviews were transcribed verbatim and iteratively coded and analyzed using Atlas.ti. The data analysis process followed the steps illustrated by Strauss and Corbin (1990) and Corbin and Strauss (2008). Firstly, we read transcripts from beginning to end multiple times to get a thorough understanding of our data. Secondly, while still collecting data, we began coding our data. Guided by our research question, we developed concepts using a mix of process coding, in vivo coding and descriptive coding (Saldaña 2009). Thirdly, we grouped and related concepts together to find themes and higher-level patterns, in particular pertaining to different forms of reactivity. These three steps happened in an iterative manner, interspersed with brainstorming sessions between the authors. Empirical examples from our codes are presented in Table 2.

Table 2. Illustrative examples of codes and themes in the data.

Illustrative empirical examples	Example codes	Themes
‘Yeah, yeah. All the time. I get push messages and I respond to them as soon as I can.’ (Quote from P6)	Alert-based checking of OCRs	Checking OCRs sporadically
‘But sometimes I think it's unreasonable when people come here and something happened and they're not happy. And then they do not speak here, but they post something online. So, sometimes it's used like a punishment for the business.’ (P4)	Customers use OCRs as a punishment tool	Disputed legitimacy of OCRs
‘Obviously, we mostly communicate through WhatsApp (…) with all the chefs and with all the staff. When a good review pops up, then I often make a picture and send it and thumbs-up to motivate our staff basically.’ (P8)	Patterns of staff management	Managing staff loosely
‘Uh … no, no. I don't think that's fair. Um … because it's kind of like tips, you know. If we get really good tips, that's great. If we get really shitty tips, that sucks but that doesn't mean we were doing a bad job. It's really subjective. Ah! That's the word, subjective.’ (P11)	Reasons for no stricter staff management
‘We are now, I think, listed in TripAdvisor in the 194 (…). Our goal is to be in the top 100. You can also use it as a motivation tool: ‘Let's get into the top 100 with real reviews, with real responses, with real good experiences and then we do the job.’’ (P1)	Patterns of goal setting	Setting goals on OCR systems
‘The secret is hard work by all of us. (…) So, what we deliver here is what we aim for and the reviews come along.’ (P17)	Reasons for no direct efforts to reach the goal
‘Hum … yeah, I think, me and my co-partner, if we read the review and we think: 'Well, yeah. I think they're true.' Then, we're gonna take a look at it.’ (P12)	Patterns of using OCRs for improvements selectively	Adopting feedback selectively
‘Online reviews are, most of them, are not very useful. (…) But most of the times, it's just very general stuff like “I didn't like it there” or “the atmosphere wasn't good”.’(P3) ‘Like, I saw one today from someone who said: "oh, it's really expensive. 2-stars." Yeah, and then two pages later, there's someone who says: "oh, it's really reasonably priced." So, it's just … yeah, I don't think it's really important, in my opinion. I won't use it that much.’ (P15)	Reasons for no comprehensive feedback adoption
‘When people pay they get a little card here with a reference like: “Hey, maybe you can share what you think on TripAdvisor.” I always ask them to write something with it.’ (P4)	Mobilizing consumers for OCRs	Managing online reputation indirectly
‘I think a comment must be something that the person wants to put on the Internet and it's not that I ask for it or … For example, restaurants when you're paying they're giving a card so “Please put nice things on TripAdvisor” we don't do that sort of things, no, no. They have to do it themselves.’ (P5)	Reluctance in soliciting OCRs
‘Often I do not respond to good reviews. You're used to getting good reviews. I would say 80–90% of the reviews are good. (…) But obviously, when there's a valid problem or comment or whatever, you have to kind of respond to that. We try to do that as soon as possible.’ (P8)	Patterns of responding to OCRs	Managing online reputation directly
‘If he would be sitting here, you could talk to him like: “I’m sorry about it. Here's a cup of coffee.” (…) But you can't because he's gone. So that's the big difference. Then you have to make sure that what you write is for other people that are going to read that, to influence the other people that they will come to you and not think “oh, the food is bad there”.’ (P1)	Responding to OCRs ‘on stage’

On reactivity: mechanical vs. deliberated organizational responses

Unsurprisingly, all our respondents are acutely aware of their restaurants being reviewed, rated, and ranked online. They seem to take pride in high ratings and rankings and put Tripadvisor’s Certificate of Excellence stickers on their doors as a testimony. Not only can they name all major OCR systems where their restaurants are reviewed, they also demonstrate that they keep up to date with reviews. All but two restaurants check them on a daily basis and the remaining two restaurants do it weekly. The vast majority react to notifications from OCR systems on their mobile devices every time a new review is published and quickly check them on the spot in-between tasks. They indicate that time spent on OCR systems is fragmented, dispersed, and squeezed in between larger activities as it is on other social media. As a result, restaurants feel like they are checking OCRs all the time. However, given how regularly and constantly our respondents track reviews, it is striking how little these reviews directly matter to their everyday operations. In particular, we found restaurants’ organizational responses to OCRs limited in three areas: management of staff, goal setting, and operational practices.

Staff management

In contrast to Beuscart et al. (2016, p. 467), we do not find that OCRs are ‘an extremely powerful staff management tool’ that allows even for a delegation of surveillance and a dismissal of employees. Positive reviews are channeled to employees, for example, by sharing them via group chats in order to boost employee morale. In contrast, negative reviews are discussed with employees in order to ‘sharpen the team.’ Nevertheless, respondents do not incorporate OCRs as a systematic performance measure nor as a monitoring system to reward or discipline individual staff members. In fact, the vast majority of respondents either laughed or displayed discomfort when inquired about such uses of OCRs. They emphasize that OCRs are not legitimate enough for employee evaluation since OCRs are ‘really subjective’ and ‘there's so much more than just what the people [consumers] are seeing.’ Moreover, restaurants argue that OCRs are not useful for staff management given how they are rarely substantiated enough to be traced to specific employees and their comportment.

Setting goals

About half of our respondents have either specific or vague goals with regard to OCR systems. Some generally aim at improving their overall performance on OCR systems, while others set concrete goals, including being ranked #1 in all of restaurants in Amsterdam. This ambitious goal eventually became realized for one restaurant, but when we asked how they could climb up over 100 ranks to be #1, the only answer we got was: ‘just hard working.’ One of our respondents explained how he deliberately allocates some tables to be reserved through Tripadvisor’s reservation service, hoping that it would increase the number of reviews on the platform and therefore its rating. Although this costs him considerable money, Tripadvisor would follow up on the consumers to urge them to leave reviews. Mostly, however, restaurants argue that there is little they can do that would translate directly into improvements in ratings and rankings. Indeed, our findings suggest that OCR systems as algorithmic apparatuses allow little room for strategic behavior (Orlikowski and Scott 2014). Our respondents also shared stories of these systems changing algorithms overnight and their ranks plummeting by 100s. Faced with an algorithmic black box, all that they can do is to put ‘hard work by all of us’ and then hope that this hard work would eventually be reflected on OCRs. As one respondent put it: ‘In the beginning, we were really bothered (…)But, pff, we’re over that. Nowadays, we do what we can. We do everything what we think is right and yeah, if it’s not then it’s not. That’s it.'

Operational practices

In the auditing literature, OCR systems are seen as devices which potentially contribute to the overall quality improvement of restaurants by their mere existence. That is, the amplified risk of individual customer’s dissatisfaction heightens the tension in everyday operational practices. Our respondents, however, are far from willing to adopt customer feedback to make concrete improvements to their businesses. They say they only act on comments when a number of conditions are fulfilled: the comments have to be reasonable, specific, repeated, and furthermore involve relatively small and low-cost changes. For instance, after an OCR that complained that ‘the staff always go smoking and do not wash their hands,’ a restaurant owner told the staff to ‘walk all the way around the back’ when going for a smoke so that consumers could not see them. Another restaurant received an OCR which complained that the staff took the plates too early. So, they decided not to take plates before the whole table finished.

Instead, almost all our respondents choose not respond to OCRs that demand more fundamental changes in their operational practices. Many of them struggled or failed to remember concrete and core changes they had made to their places, menus, amenities, or management practices from OCRs. Endorsing their own professional ethos (Dorn 2019), one of them argued: ‘[i]t's just small changes on plating or something. But on taste, we don't do [change] a lot. We think we're doing okay and we have to believe in that. We have to believe that we're sort of right on taste.’

Reasons for deliberation

Our respondents were highly articulate about why they do not actively make fundamental organizational changes regarding the staff, goal-setting, and operations from OCRs. They presented a wide range of reasons for not doing so, which go beyond the sheer lack of legitimacy of reviews. Firstly, they perceive OCRs as portrayals of subjective expectations and prior experiences of individual (lay) reviewers instead of objective, knowledge-based reflections of a restaurant’s performance. Relatedly, they note that the opinions presented in OCRs often lack consistency and contradict each other, which makes it impossible for restaurants to act upon them:

Yeah, lots of people complain often about noisiness, for instance. We are a noisy place because it's a lively place and has music and there are people laughing and people celebrating. So yeah, I mean, that's a complaint we can't do much about, you know. (…) And then the next comment is: ‘Yeah, really nice, loud, hip, happening, friendly, alive.’ So yeah, you can never act on both, you know.

Thus the shift of evaluation from a (predictable and consistent) formulaic apparatus to an (unpredictable and inconsistent) algorithmic one, which has been identified lucidly by Orlikowski and Scott (2014), provides restaurant owners with a legitimation to not act upon them.

Secondly, restaurants are reluctant to respond to OCRS because they frequently misalign with a restaurant’s identity. Restaurants highlight the need to be especially strict with feedback that collides with what they believe lies at the core of the business and therefore is unnegotiable. One Spanish restaurant gives an example of an OCR complaining that ‘the dishes are too small’ while Spanish tapas are by definition small and meant to be shared:

[If a comment is] different to the way you want your business to be, – so, for example, ‘it’s loud’ or ‘the dishes are small’ (…) Then you don’t use it. Then, you should go to a different restaurant. You’re here for the vibe, for the fact that you have sharing dishes.

A third reason why reactivity to OCRs is limited when it comes to staff management, goal setting and operational practices is that respondents question their identity as valuation devices. Instead, respondents think that some consumers write reviews not to give feedback or solve problems, but purely to wield power and harm the business. One restaurant owner argues that some malicious reviews are similar to ‘online bullying’ which happens ‘behind this digital layer.’ He believes that some consumers do not bother taking the effort to communicate their dissatisfaction when asked in person, partly because of OCR systems at their fingertips: ‘No, I don’t have the time for that. I will give them a one-star. I will never come back. The damage is done.' Reviews that are driven by this rather perverse objective of punishment are also not considered legitimate, and thus neglected by restaurants: ‘[t]he real negative ones are, in my opinion, mainly just not honest.’

Leaving aside the disputed legitimacy, the limited informational value of OCRs as valuation devices is the fourth reason why restaurants don’t act upon them. This holds in particular for Google reviews which allow users to give a star-rating without a textual motivation of the rating. Reviews on Tripadvisor and Iens, by contrast, are seen as more informative because of the requirements that reviews meet a character count of 100 and 120 each and that they conform to pre-defined formats including assessment standards such as ‘waiting time’ and ‘price quality.’ But even on those platforms, reviews are often too generic to be acted upon. When asked to give an example of a practice he had changed in the restaurant because of reviews, one respondent answered stumblingly: ‘I don't have such specific example for us because … yeah, um … The most reviews we get are not that specific. They're quite generic. And they say “Ah, it's a nice menu” or “It's a nice terrace”, “It's a nice service”.’

Fifthly, the sheer amount of OCRs that restaurants receive also enables them to be selective. As one respondent explains, OCRs in general are positively skewed (see also Mellet 2014, pp. 27–28): ‘You're used to getting good reviews. I would say 80–90% of the reviews are good.’ Since negative reviews are likely to be preceded or followed by many more positive ones, restaurants can manage to ignore them.

This leads to a final reason why respondents hardly respond to reviews: because they can afford to. Other valuation and monitoring devices stand at their disposal to improve the quality of their business which they trust more. For instance, some respondents indicate that they prefer to rely on direct feedback from regular customers over indirect feedback from the one-off relations which OCRs entail. Also, they invite or from friends or colleagues to the restaurant as ‘mystery guests’ and ask them for a short report on what they can improve afterwards.

To summarize, our interviews indicate that restaurants are highly skeptical of OCRs as valuation devices and are therefore selective and cautious in using them for managing staff, goals, or operational practices. While Beuscart et al. (2016) come to a similar conclusion when it comes to the perceived legitimacy of OCRs, they argue that the systems show reactivity nonetheless. We, in contrast, find that exactly because of this disputed legitimacy as well as their limited informational value, abundance and substitutability, reactive conformity to OCRs is limited to reputation management. Our respondents exert control by carefully judging the reviews, reflecting on their meanings, and using discretion when choosing which reviews to act upon and which to dismiss. We name this deliberated reactivity and contrast it to the mechanical reactivity illustrated in previous studies, which emphasizes the disciplining impact of valuation apparatuses (Pollock et al. 2018). With mechanical reactivity, assessments get translated and adopted relatively passively and by and large automatically into concrete organizational changes. On the other hand, with deliberated reactivity, assessments will first be contemplated, doubted, and juxtaposed with an organization’s current practices, aims, and identity. Only regarding reviews that pass a series of rigorous tests, some form of reactivity may ensue. Obviously few forms of reactivity will be mechanical in its literal sense. However, we propose this analytic opposition in order to better grasp and nuance the kind of reactivity observed on OCR systems and to further our sociological understanding of reactivity in general.

Beyond reactivity: reputational responses

While the impact of OCRs on the management of staff, goals, and operations was deliberated and limited in scope, our findings indicate that restaurants utilize OCRs more actively for reputation management. Restaurant owners may dispute the legitimacy of OCRs as a valuation device for their performance and services but that does not preclude that they are aware of the role which OCRs play as judgment devices for potential customers. By asking customers directly, spotting people who check OCRs on their mobile devices in front of their restaurants and from their own experiences as consumers in foreign contexts, they know that consumers resort to OCRs when deciding whether or not to visit a restaurant.

Restaurants believe that OCR systems are a particularly important navigation tool for newcomers to their restaurants such as tourists. One restaurant owner, half of whose consumers are tourists, argues that ‘[a]s a stranger in the city, you find your way by relying on the opinion[s] of other people.’ In this capacity, OCRs have reconfigured the way in which competition unfolds in the restaurant industry. As one of our respondents puts it:

I think online reviews and … Google basically, they did make sure a lot of people shift away from a lot of old-fashioned advertising, which I think is great because I mean an advertisement in a newspaper, it costs a fortune and you never know who's reading it, you know?

Mobilizing consumers for OCRs: indirect reputation management

While our respondents doubt the legitimacy of OCRs as evaluation devices, they say that they cannot avoid endorsing them as marketing devices. This means first of all that they mobilize consumers to review their businesses for the resulting online publicity that is expected to bring more business. Seven of our respondents engage in this practice, among which some strategically select whom to solicit reviews in order to have a positive influence on their online reputation. As one manager explains,

Well, if I ask someone like ‘Oh, leave a review for us’ or whatever, I'm not going to say that to the table that's angry or that didn't have a good night or … you know. I would say that to the people that I know that had a good time.

Despite the expected economic benefits, we were surprised to hear that restaurants are generally very hesitant to ask consumers for OCRs. An intricate moral economy seems to underlie this reputation management practice. One manager explained her discomfort by comparing it to ‘asking for tip[s]’ – in a country with no strong tipping culture. Another restaurant manager shared: ‘sometimes I think about it [soliciting OCRs], but then I think nah … I think it’s kind of cheap.' Yet another owner speculates that his reluctance stems from the values of Dutch culture, both that of respecting individual boundaries and being honest:

I don't know, maybe it's a little bit of a Calvinistic thing in me but … I don't do it. (…) But like I say, if I'd be a clever, hard businessman then I'd probably put on the bills: ‘Did you enjoy? Review. If you didn't like us, please tell us.’ You know, something like that. But I think it's a little bit too much. (…) I know it's very important, but I don't like to push people. It must come from people themselves. It must be honest. If I look at [competitor], he has 500 reviews but I know that he asks people.

Nevertheless, when the expected gain exceeds the discomfort, restaurants start mobilizing consumers for (positive) reviews. One restaurant owner that has recently given in to the growing importance of Tripadvisor and started soliciting reviews to consumers, illustrates the clash of the perceived necessity and moral reluctance and describes the triumph of the former as ‘a necessary evil.’

It is not something I would typically love to do, you know, because … yeah. As far as I'm concerned, it’s a necessary evil. I mean, the whole rating game, online game. (…) We're kind of forced to play this whole game with everyone because as a restaurant, you have to be highly ranked, hopefully. So, it's a necessary game you have to play, that's the way I look at it. I don't really necessarily like to ask my clients: ‘Please leave a good review.’ It's not so nice to ask. It's not something I particularly enjoy. We'd rather focus on good cooking and good service.

Responding to OCRs: direct reputation management

If soliciting OCRs is an indirect reputation management practice through (satisfied) consumers, another more direct reputation management practice of restaurants is to respond to OCRs. Twelve of our respondents engage in this practice albeit with varying degrees. Five restaurants respond to all reviews on selected OCR system(s). Another five respond occasionally and sporadically such as when they have time, and two respond only to negative reviews.⁷

Restaurants take OCRs particularly seriously for reputation management, as the expanded viewership, the immediacy of content upload, and the open and public character of OCR systems alter the relation between businesses and consumer dramatically. Accordingly, responding to a consumer’s complaint becomes imbued with other motives. When responding to a review, the restaurant in practice becomes more preoccupied with other (potential) consumers who will read the review and the response in order to decide on whether or not to visit the restaurant. Therefore, the respondent’s attention shifts from the present to the future, from private to public, and from solving a problem to managing an image. In short, the seemingly plain exchange of feedback and response between consumers and restaurants takes on a different dimension of performance due to the myriad of viewers in the digital sphere that these platforms connect. One of our respondents compared publicly responding to an OCR to singing ‘in front of a million people’ as opposed to singing ‘to yourself at home.’

So what is it that restaurants seek to communicate to these viewers and (potential) consumers? To begin with, responding to OCRs enables them to signal professionalism, to demonstrate that they value consumer opinions. Regarding positive OCRs, they do this by reciprocating and appreciating the time and effort of reviewers. With negative OCRs, restaurants probe into the issue and signal their sincere interest and dedication in improving themselves. Case 1 (Table 3) offers an example of a review and a corresponding response which is driven by this first motive.

Table 3. Overview of response strategies with different motives towards reputation management.

Case No.	Motives	Example reviews	Example responses
Case 1	A. Signaling professionalism to the crowd	[OCR on Google Maps] ★☆☆☆☆	[Response] Dear [reviewer], thank you very much for rating us. In order to understand your experience and rating in more detail would you mind to contact me via [email address] or via [phone number]? It would mean a lot to us to make it right. (OCR and response retrieved from Google Maps on 28 June 2018)
Case 2	B. Negotiating online reputation - B.1. Explaining the situation - B.2. Explaining measures taken - B.3. Contesting devaluations	[OCR on TripAdvisor (trimmed)] ★☆☆☆☆ Stay away! I found metal shape in my salad! Hey I have been at [restaurant] and ordered a salad. Suddenly, I felt something sharp in my mouth!!!! It was a sharp piece of metal!! (…) I definitely don’t think you should it there! A big piece of sharp metal could have cut my throat!!! Do not recommend!!!	[Response (trimmed)] First of all I would like to thank you for visiting our restaurant. (…) What you found in your salad was most likely a part of a metal sponge which we use to clean our pans- however we are now going to change to different types of sponges which will never leave pieces behind. On behalf of [restaurant] and our team, I sincerely apologize for your experience, and hope to welcome you another time. (OCR and response retrieved from TripAdvisor on 8 June 2018)
Case 3	C. Promoting oneself - C.1. Advertising - C.2. Imprinting a positive image	[OCR on Google Maps] ★☆☆☆☆ Cod was very tasty! However chips were not great, mushy peas were cold and came in a very small 5cm diameter 1 cm deep dish. I've had many better fish 'n chips. Not really recommended.	[Response (trimmed)] (…) We are very sorry the Mushy Peas came out cold, that should not happen. Thank you for telling us straight away, so we were given the chance to make it up straight away and get you a portion of hot Mushy Peas. (…) Thanks for leaving a review, always appreciated. Glad you like a fresh piece of Cod! (OCR and response retrieved from Google Maps on 8 June 2018)

Demonstrating professionalism by responding online is believed to be greatly important. Some respondents think that this brings them a competitive edge over nonresponsive restaurants. As one restaurant owner who responds to all reviews across all major OCR systems puts it,

I believe that when a visitor comes by and you see kind of a same rating but there's one place responding to the consumers, I think it could trigger somebody to definitely enter that place because there is a dialogue between the place and its consumers. (…) I would appreciate it. If I see that a business is replying back to the feedback, I would say: ‘Oh, they care about the customer.’⁸

Another motive for restaurants to respond to negative reviews in particular is to negotiate their online reputation. They do so by explaining both what the situation was like and what measures have be taken about the issue. The ultimate aim here is to defend their reputation and mitigate the potential impact of devaluations, primarily that of alienating prospective consumers. For instance, a restaurant recounted how a piece of a metal sponge ended up in a customer’s salad from the kitchen once. Despite sincere apologies, the incident resulted in a permanent display of a Tripadvisor review which was quite misleading according to the restaurant manager. In his response, the manager clarifies that ‘a big piece of sharp metal that could have cut my throat’ was actually ‘a part of a metal sponge’ and promises to stop using metal sponges altogether to prevent similar incidents (Case 2 in Table 3).

With a detailed account and such a promise, the manager hopes to minimize the worries of potential consumers and reassure them that it is ‘safe’ to visit the restaurant:

The most important thing in my opinion is just to make sure they know that you're doing something about it. (…) Even if she doesn't come back, at least other people that see the review and see that we did something, they will know: ‘Oh, we're coming here. They must be safe.’, you know?

Sometimes restaurants move beyond explaining to contesting devaluations, especially when reviews seem to involve factual misstatements.⁹ For instance, a restaurant once received a negative review where the reviewer stated to have ordered a ‘double.’ Having no ‘double’ whatsoever on the menu, the owner was suspicious of whether the review was true or meant to be directed at his restaurant. By writing ‘[w]hat double did you order as we have nothing on the menu that is double,’¹⁰ he invites viewers to doubt the credibility of the devaluation and make them wonder if the reviewer had even visited the restaurant.

Finally, restaurants respond to reviews because it enables them to transform OCRs into an additional promotion channel. For example, responding to a five-star review, one restaurant owner goes beyond signaling professionalism to advertising his products: ‘The mixed platters to share are one of the favourites.’¹¹ In responding, even negative reviews can be subverted and turned into an occasion for self-promotion, by highlighting anything positive that is mentioned in the review (see case 3, Table 2).

In short, compared to the limited, deliberated reactivity shown in relation to staff management, goal setting and operational practices, we see a stronger sense of urgency around OCR systems among our respondents regarding reputational management. This sense of urgency is further heightened around negative reviews that often leave restaurants feeling misrepresented. This does not, however, mean that restaurants engage in reputation management wholeheartedly. That is, these reputational responses were often symbolic and stylized. But because of the reconfigured dynamics of competition which the rise of OCR systems has entailed, they feel the need to put more conscious effort into the game of ‘looking good’ online because not doing so is expected to penalize their business:

… if you’re good in the game of getting high reviews or a lot of reviews, you can have a busier, better restaurant or better-running restaurant than an old skilled guy who doesn’t know anything about the Internet but can cook like the best, you know. His restaurant will lose to the restaurant with a good online presence.

So it’s a necessary game you have to play, that’s the way I look at it. I don’t really necessarily like to ask my clients: ‘Please leave a good review.’ It’s not so nice to ask. It’s not something I particularly enjoy. We’d rather focus on good cooking and good service.

Discussion and conclusion

Most previous studies on OCR systems take a consumer-centered approach and discuss the democratizing, consumer-empowering aspects of these systems. They often view OCR systems either as new valuation devices (Jeacle and Carter 2011, Mellet et al. 2014, Orlikowski and Scott 2014, Beuscart et al. 2016, Wang et al. 2016, Bialecki et al. 2017) or as social media where people with common interests share information in the form of user-generated content (Scott and Orlikowski 2012, Ayeh et al. 2013, Filieri 2016). The current study, by contrast, expands our understanding of OCR systems to encompass their function as a marketing device for businesses. This is enabled by taking the vantage point of the other end, that of evaluated entities, focusing on their perception of and responses to OCR systems. In doing so, we inquired about concrete practices of mid-price restaurants in Amsterdam in reaction to being evaluated on various OCR systems.

The main contribution of this paper is to a rapidly emerging literature on ratings and rankings and how these reconfigure various societal fields (see e.g. Stark 2020 for an overview). Contrary to most previous studies on this topic, we find that the reactivity of these metrics cannot be taken for granted. In our case, only feedback that is recurring, judged to be convincing, specific, low-cost, and rather marginal for the business is likely to result in organizational responses in the form of tangible, material changes. As far as we see reactivity, it is not of the self-fulfilling prophecy type emphasized by Espeland and Sauder (2007, 2016). Businesses are highly aware of the limitations of assessments made by the lay public using algorithmic apparatuses (cf. Orlikowski and Scott 2014) such as that of being subjective, contradictory, sometimes invalid, and even ill-intentioned. This disputed legitimacy of OCRs together with their limited informational value and the sheer quantity and substitutability of OCRs result in restaurants being able to carefully choose when to doubt or dismiss OCRs as valuations. In contrast to the mechanical reactivity which previous studies of OCR systems have endorsed to foreground their disciplining impact, we call this reflexive, cautious attitude deliberate reactivity. Moreover, the democratic potential of OCRs is ironically undercut because the incessant supply of reviews which restaurants face, allow them to choose themselves which to engage with and which to ignore.

In this respect, our findings differ from those of Beuscart et al. (2016) who, despite the lack of legitimacy of OCRs, find reactivity related to staff management and operational practices. Moreover, while they see relatively few restaurants who embrace OCRs for marketing reasons, in our study this reputational response was relatively strong. Our findings are more in line with Pollock et al.’s (2018) recent study which points at the agency of organizations in responding to ranking systems or Dorn’s (2019) study which shows how in the hospital field, performance metrics are frequently ignored or rejected. In particular, the very fact that no organizational template underlies OCRs as well as their erratic, and often contradictory nature (Orlikowski and Scott 2014) creates multiple degrees of freedom for restaurant managers regarding if and how to react to OCRs.

While the disputed legitimacy of OCRs instigates restaurants to be highly selective in responding to them by altering material and physical practices (i.e. only respond to the most legitimate ones), for our respondents, by contrast, misrepresenting, disputable, doubtful OCRs that potentially damage their reputation are the most tempting ones to respond to. The expanded visibility, reach, and life of consumer reviews on OCR systems seem to collectively redefine the simple act of responding into a strategic act with multiple and possibly ulterior motives: signaling professionalism, negotiating online reputation, and promoting oneself. As one moves from the first to the third motive, the reactive character diminishes while the proactive character intensifies. That is, responses change in character from accepting to explaining, contesting, then to appropriating and subverting the situation to restaurants’ advantage. Our argument resonates with the influential work of Oliver (1991) who highlights organizations’ agency in choosing varying strategic responses to institutional pressure. In particular, our findings support her predictions that resistance to institutional pressure grows when multiple constituents yield incompatible and contradictory demands as well as when demands are inconsistent with organizational goals or encroach on organizational discretion and decision-making autonomy.

Yet OCRs differ fundamentally from institutional processes mentioned by Oliver (1991) because they lack clear rules and criteria, direct sanctioning power, and lingering relationship between organizations and those that exert institutional pressure. In this regard, we see a disconnect between the aggregate of individual OCRs and OCR systems that archive, algorithmically synthesize, and publicly display the reviews, ratings, and rankings. It is striking that OCR systems can wield such power on organizations while individual OCRs are hardly able to elicit any change in their staff management, goal setting and operational practices. In making sense of this, we turn to Kornberger et al. (2017) who argue platform organizations such as Uber and Airbnb deploy a distinct evaluative infrastructure and leverage heterarchies over hierarchies and as a result centralize power while radically decentralizing control.

If we speculate about the reasons why our findings differ from those of previous studies on OCRs in the hospitality industry (e.g. Orlikowski and Scott 2014, Beuscart et al. 2016), we think that restaurants in the early days of OCRs may have overestimated their potential impact because of their novelty. Now that OCRs have become a mainstay in their business, they have perhaps come to a more nuanced assessment of their everyday impact and have had sufficient time to develop coping practices around them, in particular when it comes to the novel opportunities these systems present as marketing devices.

Another reason why our findings differ to some extent may be related to our methodological choices. We have opted in this study to investigate one class of restaurants in depth, which means that our sample is less heterogeneous than those of previous studies. For instance, we only included restaurants with at least 30 reviews and they were all located in Amsterdam, the tourist capital of the Netherlands. Hence, they have a steady supply of consumers and a constant supply of OCRs proportionate to that. This likely gives them more room to resist individual OCRs when compared to businesses in remote areas that depend more on OCRs among other digital marketing devices to attract consumers. The continued supply of OCRs also has consequences for the visibility of individual reviews and the resulting reactivity. On the other hand, businesses in an urban area also face a heightened level of competition. This might invite them to appropriate OCR systems more proactively as an additional marketing tool to differentiate themselves from their competitors. Our respondents were relatively young (only seven were in their late 40s to 50s) which is likely to have influenced their digital literacy and their ability to maneuver OCR systems. Finally, our respondents were restaurant managers or owners. Had we interviewed employees, we might have gotten different narratives around the reconfiguration of accountability highlighted by previous research.

Practices related to reputation management merit more attention as they have been relatively overlooked in the literature of OCR systems (but see Hollenbeck et al. 2017, Proserpio and Zervas 2017). The limited number of studies that focus on evaluated organizations look into how hotels and restaurants receive and react to OCRs, but rarely investigate organizations’ proactive role on OCR systems. We tried to resolve this analytical imbalance by paying attention to the largely neglected fact that evaluated entities, too, have an active role in the system. In this light, we argue for the need to go beyond seeing evaluated entities solely as passive and receptive in line with Wang et al. (2016) who argue that while OCR systems bring organizations into the forefront for consumers, they do not necessarily make them more vulnerable.

To fully grasp their real-life impact, OCR systems should be conceptualized a hybrid of valuation and marketing devices (cf. Beuscart et al. 2016, p. 464) where both the evaluat-ing and the evaluat-ed have a voice. That is, OCR systems are better understood as multisided platforms which connect and appeal to both consumers and businesses, allowing each side to bring their own agenda. Indeed, one of the reasons for the by now strong institutionalization of OCR systems is that they strike an intricate balance by relating and communicating different messages to both sides: that of democratization, community spirit and valuation to consumers and that of visibility and marketing to business entities.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Bomi Kim

Bomi Kim is a PhD candidate at KIN Center for Digital Innovation at the Vrije Universiteit Amsterdam. Her current project concerns the development and use of artificial intelligence technologies in radiology using qualitative methods. She studies this topic on multiple levels focusing on material agency, organizational change, knowledge work, and discourses.

Olav Velthuis

Olav Velthuis is Professor at the Department of Sociology of the University of Amsterdam, specializing in economic sociology, sociology of the arts and cultural sociology. At the department, he is director of the program group Cultural Sociology. His research interests include the globalization of art markets, the interrelations between market and gift exchange, the valuation and pricing of contemporary art, and the moral and socio-technological dimensions of markets for adult content. Before moving to the University of Amsterdam, Velthuis worked for several years as a Staff Reporter Globalization for the Dutch daily de Volkskrant.

Notes

1 Sometimes the aggregate information is further abstracted and institutionalized in the form of credentials such as Tripadvisor’s Certificate of Excellence, which since 2011 is granted to approximately 10% of businesses in the hospitality sector. https://www.tripadvisor.com/TripAdvisorInsights/w604 (Accessed 6 July 2018).

2 For instance, a one-star increase in Yelp ratings can result in a 5-9% increase in revenue for restaurants (Luca, 2016) while a 10% improvement in ratings can increase hotel room sales by 4.4% (Ye, Law, and Gu, 2009). Corley and Gioia (2000, p. 330) illustrate that US business schools experience the immediate impact of rankings in their student applications, external funding, etc.

3 Founded in 2000 in the US, Tripadvisor started as a search engine for travel. TripAdvisor has become more relevant for the Dutch restaurant industry after it took over Iens in 2015. TripAdvisor claims to be the world's largest travel site with an average of 455 million monthly unique visitors (in seasonal peak during 2017). It offers over 630 million user-generated reviews and listings of 7.5 million accommodations, airlines, attractions, and restaurants among which 4.6 million pertain to restaurants. Iens was founded in 1998 in the Netherlands as a democratic local restaurant guidebook where all restaurants could be evaluated in contrast to high-end, exclusive guides such as the Michelin Guide. By 2008, Iens had become the most visited website in the Dutch restaurant sector (De Ronde and Sahadat 2008). By January 2015, Iens had an annual turnover of six million euros and a network of more than 20 thousand restaurants. Over 4,000 reviews were published on Iens weekly and around 200,000 yearly (De Waard 2015). Google Maps was launched as a web mapping service in 2005 in the US and was introduced in the Netherlands a year later (Van Ammelrooy 2006). In 2007, it started permitting its users to directly leave ratings and reviews of local businesses.

4 The concept of reactivity originated as a methodological concern for the validity of a study. This is because when an individual under a study modifies her behavior due to her awareness of being observed, it contaminates the measurement (cf. the Hawthorne Effect).

5 The concept is close to what others call an organization’s market identity which concerns the shared representations of both internal members and external experts of an organization (Wang et al. 2016, p. 135).

6 Of all 3509 restaurants in Amsterdam enlisted on TripAdvisor, 81% (2,827) had price information among which 73% (2053) were classified as mid-range (as of April 2018).

7 Before each interview, we looked restaurants up on different OCR systems and checked their responding patterns. While in most cases, respondents’ accounts confirmed what is visible on OCR systems, there were a couple of inconsistent cases. In light of the contested ‘attitude-behavior consistency (ABC)’ in Jerolmack and Khan (2014), what is visible on these systems was prioritized over their verbal accounts in such cases.

8 Tripadvisor officially vouches for the first motive as well: ‘Responding to reviews clearly demonstrates – to both former and prospective guests – that you are interested in feedback, and that you take customer service seriously.' https://www.tripadvisor.com/TripAdvisorInsights/w805 (Accessed 10 June 2018).

9 Tripadvisor also encourages this motive: ‘It’s generally a good idea to respond to reviews that are negative, as well as those where you can correct a factual misstatement or write about an action you’ve taken to correct problems addressed in the review.' https://www.tripadvisor.com/TripAdvisorInsights/w805 (Accessed 10 June 2018).

10 Retrieved from Google Maps on 5 May 2018.

11 Retrieved from Google Maps on 8 June 2018.