USERS’ INSIGHTS INTO TARGETED ADS ON SMART
TVS: A QUALITATIVE STUDY
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Persuasive Communication
Sara Recchi 11873949
Supervisor: Dr. Brahim Zarouali
Academic year: 2018-2019
Abstract
Smart TVs are becoming a very common device in households and the practice of programmatic advertising is expanding to other online media than browsers only, thus making it possible to deliver targeted ads on smart TVs. The purpose of this study is to investigate what are the consumers’ perceptions and beliefs on receiving targeted ads on smart TVs. By means of one-to-one interviews with smart TVs owners, three main themes are explored: the acceptance of targeted advertising, privacy related issues and the characteristics of the medium. The results confirm the relevance of the personalisation paradox and the importance for users to have some control in order to accept targeted ads. Some outcomes that pertain smart TVs specifically are highlighted, such as the social implications deriving from the fact that smart TVs are often enjoyed in company of others.
Introduction
The use of personal information to tailor advertising on users’ interests and behaviours has been a trend in the past decade in online environments, and more particularly, on social media. This practice is called targeted advertising and in order to be delivered, automated systems have to collect a vast amount of personal information on the users. This can be done in a covert or overt way, i.e. they either explicitly ask to people to actively disclose personal data or they covertly monitor and track online behaviours and interactions (Sundar & Marathe, 2010).
The response to personalisation can usually be of a dualistic nature: it has been shown that although some users appreciate relevant personalisation, others experience serious feelings of distress. On one hand, personalised content is perceived as more relevant and interesting, and therefore, it is less likely avoided and can create a desire to engage with the brand (Aguirre, Roggeveen, Grewal, & Wetzels, 2016). On the other, it also arouses feelings of discomfort when people recognise that the messages are tailored on the basis of information they have shared, or even based on data they were not aware companies had access to (Aguirre, Mahr, Grewal, de Ruyter, & Wetzels, 2015). The sense of distress might lead to the tendency to avoid personalisation. This phenomenon here described is known as personalisation paradox, where the users can generate positive (appreciate and accept personalisation) as well as negative responses
toward personalisation (e.g. feel threatened by personalisation and avoid it). As of yet, this paradox has only been observed in relation to online and social media advertising, and therefore, this study aims to focus on how it might apply to targeted advertising on a different device: smart TVs.
Smart TVs are digital TVs connected online, their sales are steadily increasing and are expected to keep growing globally in the next decade (Zion Market Research, 2019). Given the expanding popularity of this medium and the new possibility it offers, it is important to investigate which implications it presents for the consumers. Traditionally TV commercials are aimed to a more generic audience, if targeted advertising is taking over this device it is relevant to ask what are the
consumers’ perceptions and beliefs on targeted ads on smart TVs in order to find out which social
outcomes it might have and how people will react to the new development. With the ever-increasing number of devices connected online present in every house, it is imperative to not lose sight of what the consequences for the users, their privacy and sense of ease might be.
The most suitable technique to research a topic that involves people’s opinions is structured in-depth interviews. Respondents are asked to openly discuss their feelings and concerns for this new possibility and their responses are linked to the existing literature regarding targeted advertising acceptance, privacy concerns and the role of the medium.
Theoretical background Acceptance of targeted TV ads
Online targeted advertising is a form of advertising that consists of targeting users that present specific traits and online behaviours. It is a form of online behavioural advertising (OBA), which is defined as “the practice of monitoring people’s online behaviour and using the collected information to show people individually targeted advertisements. Online behaviour can include web browsing data, search histories, media consumption data (e.g., videos watched), app use data, purchases, click-through responses to ads, and communication content, such as what people write in e-mails (e.g., via Gmail) or post on social networking sites” (Boerman, Kruikemeier, & Zuiderveen Borgesius, 2017, p. 364). Targeted ads are an efficient way to reach a specific audience and make the messages more relevant to the single person. However, due to the fact
that the online behaviour is so closely monitored, consumers have become more concerned of OBA practices (Boerman et al., 2017).
Targeting people with specific ads is made possible by data-driven marketing automation. This new technique of ad delivery is usually called programmatic advertising or real-time advertising, which is considered inevitable in nowadays online environment and an attempt to provide consumers a more relevant experience (Busch, 2016). “Programmatic advertising describes the automated serving of digital ads in real-time based on individual ad impression opportunities” (Busch, 2016, p.8), its five distinctive characteristics are:
- Granularity - Real-time trading - Real-time information - Real-time creation - Automation
The whole process takes approximately 50ms and involves several steps: The ad is requested by the user
E V A L U A T E real-time advertising handles each contact opportunity considering placement,
user, general conditions and creatives
Supply-side-platform reports opportunity in detail to demand-side-platforms Advertisers' selection criteria are checked
Analytics deliver each opportunity's campaign specific value before buying it
Advertisers' highest evaluation is submitted P
U R C H Supply-side-platform picks the highest bid
A S E
The most compelling ad is created F
U L F I L L The ad is delivered to the user
Table 1: programmatic advertising steps (Busch, 2016)
This process allows to deliver in real time ads that are specific to that user based on his personal characteristics and online behaviour, it occurs every time a new page containing space for an ad is opened on any site. Busch (2016) describes how programmatic advertising, which has been initially created for browsers and mobile environments, presents the specifics that make it adaptable to new devices, such as smart TVs, extending the presence of targeted ads on new platforms.
As explained by Malthouse, Maslowska, and Franks (2018), traditionally TV commercials are delivered through a mass-approach: advertisers make agreements with broadcasters to buy ad spaces with the final goal of efficiency, i.e. reaching a large audience with several exposures at low CPIs. On the opposite, the programmatic approach aims to have high effectiveness, i.e. obtaining a higher long term return on investment targeting the right customer with the right message for a major delivery cost. Malthouse et al. (2018) discuss how new stakeholders have appeared: new broadcasters specialised in providing on-demand streaming content for new audiences that prefer watching only content of their liking over generic linear television; this new environment is suitable for a new form of ad delivery that prioritises personalisation. Dynamic ad insertion enables brands to place the right ad for the right household in the right commercial break, as programmatic advertising does online (Malthouse et al., 2018). Fulgoni and Lipsman (2017) state that the
progressive shift of TVs towards becoming a completely on-demand medium is the perfect opportunity for the spread of programmatic advertising type of TV advertising. One example of how the practice is already developing in this direction can be the Open Addressable Ready (OAR) Project. Major media companies such as Disney, CBS, NBC, AMC, Comcast, Turner and Discovery, together with the Smart TVs company Vizio, formed a consortium for the establishment of a standard for addressable advertising on Smart TVs (Project OAR, n.d.). The project’s deployment is planned for 2020 and it intends to provide a mean to improve the monetisation of each impression for the distributors, an enhanced advertising product for the brands and a better viewing experience for the consumers (Project OAR, n.d.).
Since targeted ads are based on personal data collection, they are not always accepted by users. Schumann, Wangenheim, and Groene (2013) describe how ad relevance is a method suitable for improving acceptance. It is based on a utilitarian approach according to which users are more prone to provide their personal data in exchange for more relevant ads.
Baek and Morimoto (2012) identify several elements that influence the avoidance or acceptance of personalised advertising: ad scepticism, ad irritation and perceived personalisation. Ad scepticism refers to the consumers’ tendency to not believe the claims of advertising; this is a general feeling that affects all forms of ads, since people recognise that the messages have a specific persuasive purpose. When the personal level of scepticism is high, the user is less prone to accept advertising in general, and therefore personalised ads as well. Ad irritation is described as the extent to which ads cause annoyance and displeasure in the users, this can depend on different characteristics of the message and the user attitude. While both ad scepticism and ad irritation increase the ad avoidance, high levels of perceived personalisation lower the ad avoidance and also decrease the levels of scepticism (Baek & Morimoto, 2012). From this discussion it can be inferred that some people will be automatically avoiding targeted ads on smart TVs due to their levels of ad scepticism and ad irritation, however well-executed targeting can have a positive effect and improve the acceptance.
In order to see which elements might enhance the acceptance of targeted ads on smart TVs it is interesting to look at what led to the acceptance of other recent forms of advertising. In a
study by Leppaniemi and Karjaluoto (2005) a model of consumer willingness to accept mobile advertising is drawn. Several factors come into play addressing different aspects of this phenomenon. The technology development allows better and more appealing ads for the users and more location accuracy and connection speed for the brands. The ads personalisation is more accurate and therefore acceptable, while the new regulatory guarantees enhance the sense of security for the consumers. Two of these elements can easily apply to the use of targeted ads on TVs and can increase the acceptance of the phenomenon. As discussed above, programmatic advertising is now ready to be applied to smart TVs and the accuracy of targeted ads is constantly improving, making the ads more appealing for the consumers. However, being this a very recent development, there is a lack of regulations that directly address the data collection and tailoring of advertising on TVs. Since in the study by Leppaniemi and Karjaluoto (2005) users showed a particular need for having control over the possibility of receiving the ads and the presence of an opt-out option, more specific regulations might be a very important step to reach the acceptance of the ads on smart TVs.
In a research by Limpf and Voorveld (2015) about mobile location-based advertising acceptance (LBA), it is shown that users’ attitude toward mobile LBA is directly and positively related to acceptance of LBA itself. This might mean that the attitude towards the possibility of receiving targeted ads on smart TVs might influence the acceptance of those. The attitude toward a phenomenon is a predictor for the acceptance of that. Moreover, the same study demonstrates that privacy concerns have a negative effect on the intention to accept mobile LBA, therefore privacy concern is one of the main elements in this paper and thoroughly discussed in the next section.
Given that the arrival of targeted advertising on TVs is a matter of time and that not much is known about the acceptance of these TV targeted ads, this study intends to investigate how people will react to this development and whether they accept targeted ads on this device:
RQ1: what are consumers’ reactions towards targeted ads on smart TVs?
RQ2: are targeted ads on smart TVs less acceptable than on mobile devices and
Privacy concerns and calculus
As highlighted by Treiblmaier and Pollach (2011) the success of personalisation depends in part on the willingness of users to make their information available to companies, and this is influenced by their individual privacy concern threshold. The concept of privacy concerns refers to the beliefs that people hold regarding who will access the data that they have put online and how it will be used. The greater their worry about the access and use, the greater their privacy concerns (Dinev & Hart, 2006).
Frequently consumers claim to want their privacy to be protected, however, they liberally and often share personal data in order to obtain any kind of service, commodity or tailoring. In this sense, privacy becomes a good that can be exchanged, bought and sold (Aguirre et al., 2016). The privacy calculus addresses this process where individuals weight the risks of renouncing to a part of their privacy against the gains of personalised and tailored services. The privacy calculus is compared to an actual cost-benefit analysis, in which people are willing to give away personal data only if they perceive that the losses are balanced out or outweighed by the gains (Dinev & Hart, 2006). Valuable gains constitute of avoiding irrelevant content and messages, improved service quality, higher content relevance, discounts, promotions and commercials that can actually be enjoyed (Treiblmaier & Pollach, 2011). As observed by Hann, Hui, Lee and Png (2007) consumers undergo a mental trade-off, quantifying the value of their privacy and deciding what can mitigate their privacy concerns like a more transparent privacy policy and financial incentives. Sultan, Rohm, and Gao (2009) showed as well how the likelihood of providing personal data can be incentivised by giving something in return to the consumers.
The privacy calculus can present two opposite outcomes: either the costs outweigh the benefits or the costs are counterbalanced by the benefits. In the first scenario, the amount of information required to tailor the messages might be too high and induce feelings of intrusiveness, especially when the data are not only demographics but also information regarding previous purchases and searches. Van Doorn and Hoekstra (2013) discuss how the advantages of relevance and good fit of personalised ads come with psychological costs for the consumers and
when personalisation is too intrusive and completely cancels off the advantages it generates. Users may interpret the high levels of personalisation as a proof that the companies are using their data as an instrument of exploitation. Edwards, Li, and Lee (2013) identify perceived intrusiveness of advertising as the underlying mechanism that makes people want to avoid tailored ads altogether, this is due to the fact that intrusiveness is correlated with feelings of irritation and loss of control. When consumers experience targeted ads as too personal and think their privacy has been invaded, it will generate personalisation reactance (White, Zahay, Thorbjernsen, & Shavitt, 2008). Psychological reactance is a theory developed by Brehm (1989) that states that people want and consider to have a certain degree of freedom when they make a decision. Therefore, when they think they are being manipulated by external influences and feel that their behavioural freedom is threatened, they will act in order to protect themselves by avoiding the persuasion and its source (Zarouali, Ponnet, Walrave, & Poels, 2017).
In the second scenario, the benefits of personalisation can overshadow these risks. The perceived intrusiveness of ads is very high when they disrupt harshly the content fruition with information that is irrelevant for the user. In this case, using targeted ads is a better choice because they are more relevant and useful to the user and therefore, when they interrupt a content, they are less disruptive and hence, less intrusive. Perceived utility has been recorded to attenuate the negative reactions to customisation (van Doorn & Hoekstra, 2013; White et al., 2008). Well targeted ads can lower the negative effects of personalisation by enhancing the relevancy of the ad for the user (Ying, Korneliussen, & Grønhaug, 2009).
In conclusion, targeting is a double-edged sword. It has been shown that it can either constitute a great value for the consumers providing more relevant and enjoyable content (De Keyzer, Dens, & De Pelsmacker, 2015) or, on the other hand, it can elicit negative feelings and be perceived as intrusive (Zarouali et al., 2017). All of the previous studies reached this dualistic conclusion in regard to targeted advertising online, on social media sites or on smartphones, however, scant literature is available that takes into consideration smart TVs as the device on which targeted ads can be delivered. This paper intends to observe privacy calculus and concerns in relation to targeted ads on smart TVs. It is interesting to see whether the different nature of this
medium plays a role in making one scenario prevail on the other or the mental trade-off in consumer remains unvaried:
RQ3: what is the privacy calculus for targeted ads on smart TVs?
RQ4: are users concerned about their privacy when in the context of targeted ads on smart
TVs?
Medium effect
The medium of any message is not a mere channel of transmission, but shapes the message and influences its reception (Sundar, Jia, Waddell, & Huang, 2015). As explained in a study by Bronner and Neijens (2018) people pay varying levels of attention to different media, and hence, the ads embedded in them. They argued that type of medium exerts an effect on how people perceive the ad that is placed on that medium, meaning that the same message delivered on a different medium can be processed very differently by the same audience. This has been explained by a spill-over effect of the medium experience on the advertising experience, usually ads on TV are experienced negatively due to the fact they are perceived as an abrupt and annoying interruption of content fruition (Bronner & Neijens, 2018). However, this has only been observed for traditional TV commercials and not targeted ones. Hence, it is interesting to investigate whether the higher relevance of a targeted message improves the experience of ads.
In addition, media contribute differently in the process of persuading the consumers, since different media are processed differently by the audience (Dijkstra, Buijtels, & van Raaij, 2005). TV, due to its own nature that combines audio and visual, evokes the highest levels of attention and cognitive responses (Dijkstra et al., 2005).
According to the theory of interactive media effects (TIME) by Sundar et al. (2015) interactivity affects user engagement. They identify three types of interactivity: modality, message and source. Modality or medium interactivity refers to the methods of interaction offered by the medium itself, the possible actions that a user can execute on it. Smart TVs offer more modality interactivity than normal TVs: they have menus and apps people can scroll through and offer the possibility to use social media and to do online searches. Source interactivity is defined as the degree to which the user is the source of the communication. This refers to the degree of tailoring
of the content offered, on smart TVs this can be very high since many use this medium to consume content on demand. In addition, targeted ads are by definition tailored to the user preferences.
Based on what it has been mentioned above, it can be expected that the level of user engagement will be very high in relation of targeted advertising on smart TVs. Therefore, the audience will be more responsive to the ads, since it has been observed that high engagement with the media contest increases ad effectiveness (Calder, Malthouse, & Schaedel, 2009).
Traditionally, TV is a medium used by more than one user at a time and therefore considered more public than personal devices such as smartphones and personal laptops, this paper aims to observe how this can integrate with the display of targeted ads, which are derived from personal user data:
RQ5: are smart TVs considered an appropriate device for displaying targeted ads?
RQ6: do the display of targeted ads on smart TVs present outcomes that are specific to this
medium?
Methods
Research method
The RQs were investigated using a qualitative method. While quantitative research focuses on testing cause-effect relationships, qualitative research is ideal for exploring people’s feeling and opinions and the underlying motivations of a behaviour. As consumers´ perceptions and beliefs are the focus of this thesis, structured in-depth interviews seemed the most transparent and appropriate method of research: in-depth interviews are appropriate when the aim is to collect information on emerging themes (Truong & Simmons, 2010). Through these conversations the participants can freely express their views and insights into targeted ads on smart TV. Qualitative interviews are particularly suited for experience based RQs since they allow the researcher to collect very detailed data about personal experiences and perspectives (Braun & Clarke, 2013).
Participants and recruitment
Due to the fact that smart TVs are a technology that involves the whole household and people across the whole spectrum, the researcher recruited participants as diverse as possible for
age, profession and gender. In order to ensure a better understanding of the phenomenon, all respondents own a smart TV.
The participants have been selected through convenience sampling: contacted through personal connections or a post on social media containing the recruitment text. Interviewing people known to the researcher (known as acquaintance interview), is acceptable under the condition that no question regarding relevant information is skipped because already familiar to the interviewer (Braun & Clark, 2013). Initially the study implied only people residing in The Netherlands but due to the lack of responses two subjects had been recruited in Italy.
The recruitment text stated the two initial requirements (residency in The Netherlands and ownership of a smart TV), information about the topic and the guarantee of anonymity (see Appendix A for the full text). In addition, all the participants had to be fluent in English in order to guarantee a fluid and fruitful conversation. The topic of the interview was only vaguely disclosed during the recruitment, in order to avoid pre-sensitisation.
Saturation was reached with a final sample composed of six female respondents and four males, aged between 20 and 41, providing a quite mixed sample of ten people. The living arrangements variated between individuals living on their own, with a partner or family and others sharing an apartment with roommates.
Data collection
Before beginning any interview the respondents were asked to sign an approval form, asking for permission of using what said successively, informing them of their rights, the purposes of the research, the supervision of ASCoR (The Amsterdam School of Communication Research) and providing all the contact information of the researcher (see Appendix B for the entire form).
The interviews were conducted either in person or through Skype. They were audio-taped using a recording device if in person or an application to record the audio from Skype calls. The interviews followed an interview guide but the flow of the conversation was determined by the answers. The amount of time dedicated to each topic depended on the depth and details offered, if a particularly interesting insight came up, additional follow-up questions were added. If a topic
wasn’t completely clear, some examples were given by the researcher. The interviewees were granted sufficient time to explain themselves, think about a topic and tell personal examples.
The location of interviews was chosen by the respondents in order to guarantee their comfort, provided that they were locations quiet enough to record clearly. The conversations were held in the respondents’ living rooms, in a lounge area at the University of Amsterdam, or in front of their office. The interviews lasted between 25 minutes and an hour.
Data analysis
The audio recordings of the interview were transcribed and then imported in ATLAS.ti. At first, open coding (see Appendix C for a complete list of the open codes) was performed selecting every part of text relevant to the research questions was turned into a quotation and assigned a code. These labels provide a brief description of what is said in the quotation and ease the process of clustering data under a topic. In order to draw the Concept Indicator Model (in Appendix D), all the codes were cleaned, merging the overlapping ones and deleting the irrelevant ones. This led to the creation of the three dimensions of the CIM, derived from the main topics of the RQs, seven indicators and several sub-indicators further describing the dimensions in detail and grounding them into the quotes.
Interview guide
The interview guide starts with general information for the respondents to make them comfortable and reassure them that there are no right or wrong answers. It was also stressed that the researcher was only interested in personal opinions on the topics (see Appendix E for the complete interview guide). Then, a series of socio-demographic questions are asked, such as age, nationality, profession and composition of the household.
The questions are clustered under four main topics: - Acceptance of targeted advertising,
- Privacy concerns, - Social factors,
- Comparison with other devices and platforms.
The questions are tailored to investigate the underlying sensitising concepts, described in the theoretical background, and to let the respondents answer freely without being restricted by questions that are too narrow or specific.
Each topic begins with a broad and more generic opening question and several follow-up questions that can be modified according to the moment and the previous answers. An opening question such as “How do you feel about targeted ads?” can be more accurately probed by asking “Do you recognise any advantages from personalisation, or is it only perceived as a disadvantage?” to encourage the interviewee to more accurately describe his perception of targeting, by providing clear examples of positive or negative experiences.
Ethics and quality criteria
The confidentiality of the interviews and anonymity of the respondents have been safeguarded by collecting and handling the data only using pseudonyms such has ID1, ID2, etc. The names and other details that might lead to identification, if used, have been redacted in the transcripts. In the analysis the respondents are referred to by general indications, for example “Italian girl”.
Ecological validity is generally very important and present in qualitative research: by asking directly the respondents for their opinions and personal experiences, it is possible to obtain results that truly depict real-life situations.
As generalisability does not apply well to qualitative interviews, which are very context related and can only capture a very detailed but fairly limited picture of a phenomenon, this study tries to reach a large transferability. By interviewing a quite mixed sample of people and by relating their opinions to the context they refer to, so that readers can evaluate if the results can apply to other subjects in the same conditions (Braun & Clarke, 2013).
The quality of the research is granted by the researcher conforming to the checklist criteria for good qualitative research provided by Braun and Clarke (2013).
Results
The study results can be clustered under three dimensions addressing the different RQs. The consumer response to targeted ads regards the acceptance of targeted ads, the psychological aspects of the phenomenon responds to the privacy-related questions and the situational context analyses the characteristics specific to the medium smart TV.
Consumer response Psychological aspects Situational context
Acceptance of targeted ads Privacy Medium characteristics
The CIM (in appendix D) presents the three dimensions and the variations within the dimensions are defined by the different indicators and sub-indicators.
Acceptance of targeted ads
The dimension acceptance of targeted ads answers to RQ1 and RQ2, it is defined by the advantages and the disadvantages that people recognised in targeted ads on TV.
The advantages are several and often the ones described by the existing literature. The first and most obvious ones are that targeted ads on smart TVs are more personally relevant and therefore a more enjoyable experience:
“They are just more precise and about what I am more interested in ((pause)) so maybe I can
enjoy the ads also a little more because I actually see stuff that I am interested in buying” (German
girl).
A common advantage identified is that targeted ads on TVs help consumers discover new products or more options for their next purchase:
“I'd be more interested in keep to watch them and to find new products or maybe, it doesn't even
have to be about products that I already like or know if they have discounts but most importantly products that could be related to things I already like that I have never heard of, so in that case it could be a great tool to discover new things” (Dutch girl).
In this regard, targeted ads on TVs are acknowledged as a helpful tool for early adopters: “You asked me if I feel there are any advantages earlier, if you're an early adopter of products in
general (…), then having targeted ads from stuff that might interest you does allow you to be ahead of the curve” (British man).
A minority of respondents pointed out that target ads on smart TVs ease the purchase process reminding the users of a product and offering discounts:
“So in that case I do feel like in the end I am winning because I saved myself a hundred dollars” (Dutch girl).
The disadvantages described by the respondents are frequently related to the feeling of annoyance they already experience when receiving targeted ads that are too frequent, timed badly or mistargeted. They fear that something similar would happen also on smart TVs:
“If you look at a product and then you buy it, they don't realise that you bought it, right? so they
keep sending you ads about it (.) and that's when I get really annoyed, because ok I researched this and I bought it and now I keep getting ads about it (.) even very aggressively sometimes and then I get annoyed” (Dutch girl).
“So when they don't get a good profile of you you get targeted ads that are completely unrelated to
you and sometimes that that can be quite annoying” (Dutch girl).
Another disadvantage recognised by some of the subjects is that targeted ads are often related with impulse buying and generate the need for purchasing more products. Targeted ads can trigger the need for instant gratification and make users buy products that are not really needed:
“They will put an advertise <while> I'm doing like totally different things, I will be maybe pushed to
“You get these targeted ads around like wallets and sunglasses and little things that are quite
impulse purchases and I'm the guy that will buy the wallets and the sunglasses and I'll be: why do I buy this s**t?!” (British man).
A third indicator clusters the conditions the respondents said that are needed in order to accept targeted ads on smart TVs. These are control over the phenomenon:
“There should be an option because I don't like it but maybe someone else is going to like it” (Portuguese man).
And having the possibility to opt out:
“With the television I don't foresee an issue but I know that, yes, it will be important that we can opt
out or just disable that feature if we think that it's now showing some content that we really shouldn't show” (Portuguese man).
Privacy
The second dimension addressed the privacy related RQs: its first indicator focuses on the privacy calculus and the second one on the privacy concerns. The privacy calculus can have two outcomes: either the costs outweigh the benefits or the benefits outweigh the costs.
“I don't think there's any content that's worthwhile giving up your data for, because they're only
getting better and better and better using that data and you don't know where it's going to show up, you don't know where it's going and then you have a data breach” (British man).
“Well starting over my online life now, I would be way more careful and put my privacy a little bit
higher, but I don't know, feeling like they already have all my data(?), I probably put the benefits higher than my security” (German girl).
“I feel like erm, I don't know, we are spied on by big corporations to sell us the stuff” (Italian girl).
And the distrusts in the companies that collect the data, their purposes and their ability of protecting data:
“It's mostly that companies right now they don't prioritize the safety(.) It's not the fact that they have
my information, that's not the most important thing to me because I don't really care about that as much I do care about them monitoring who they sell the information to(.) I don't feel that they're being careful enough with that and who has access to the information, like they don't, not every company is using very well encrypted servers to keep our data and I think that's very worrisome”
(Dutch girl).
And the fact that companies have already collected too much information on the consumers:
“I’m not a fan of it because (…) in order to target people with ads you have to have information on
them and so it makes you more aware of how unprivate everything you do is” (British man).
Medium characteristics
The third dimension of the CIM is medium characteristics, addressing RQ5 and RQ6: respondents felt either that the display of targeted ads presents the same outcomes on every media or that it presents outcomes that are specific to smart TVs. The first indicator is defined by the fact that smart TVs are seen as just another bigger screen in the net of personal devices: “I guess it's another thing you can go online with, so I can watch Netflix on the TV, so I watch
Netflix on the phone, I watch Netflix on the laptop, it's all the same (…) it's just a bigger device to watch stuff on, it's not different in any way” (Italian girl).
Also, private personal information might already be seen by others through targeted ads on other more private devices, such as on a laptop:
“I remember we had a colleague a few years ago(.) we were in California and like, he wasn't out of
the closet, people didn't know he was gay and he'd been searching like male dating sites and he was on a web browser on his computer and we were sat around and we were doing some work and these targeted ads start popping up for male dating sites and we all joked like ‘ahahah dude what have you've been searching for’ and it's like he had actually been searching for it and there's like one example of where like it's totally inappropriate to serve that stuff in a public forum, like it was on his browser” (British man).
The second indicator describes the possible consequences of targeted ads on smart TVs that are peculiar to this medium, answering to RQ6. Some of the interviewees suggest that, being TVs a more social device, private information is more likely to be disclosed to other people:
“Exactly and they could be ointments and medication or personal things for you and your partner,
like you definitely wouldn't want that showing up on your TV screen” (British man).
Few respondents said that the phenomenon could be more critical when it comes to special occasions and buying gifts since ads on the smart TV might spoil surprises:
“I mean the common one of course is things like engagement rings right? That's like the one where
people really don't want that stuff to show up (…) I think the engagement ring, birthday presents, Christmas presents that sort of things that can ruin surprises yeah also like if you're not if you're not smart about how you browse on your laptop and you're looking for other things” (British man).
Some doubted how targeted ads will still be correctly targeted when served on a device used by more than one user:
“You also still got to think that maybe different people are using the TV so which, who's targeted to,
then (?)” (Dutch girl).
“Because I think it's mismatched and maybe like if I'm watching it and if I'm not the target to that
Regarding RQ5, certain subjects affirmed that smart TVs don’t seem the appropriate medium for targeted ads:
“That's where is the biggest concern for me(.) is the fact I could look up something on my phone
which is my device passcoded, like it's just me using it and then that information could then be shared like ((pauses)) you'd only have to go and be on your friend's Wi-Fi network for like the ISP to then like see what you're looking and then serve that content to your friend(.) so it's not even necessarily just within your household like they could take information and target your friends TVs with it based on what you've been doing when your device on their network I just say the thing is this is just not appropriate it's just not the right forum for targeted ads” (British man).
The third indicator sums up the reactions to the fact that smart TVs will be an addition to the network of devices using personal information to serve targeted ads. The fact that different devices are part of a personal net is described as reasonable and convenient, and in a way justifies the presence of targeted ads on TV:
“My laptop, my phone and my TV are all connected with the same email address, I mean it kind of
makes sense that the ads that you see on your TV are based on things that you look up on your phone or on your laptop” (Filipino boy).
“It's actually associated to you and even if you are logged in on a erm borrowed device, part of it
[comes] to you(.) I think that is great, yeah I think it's really nice, to be honest it's something that goes with you, so you're just using a device whatever type of device it is but erm so so so the data the information and the target of the advert they are not dependent on the device (…) it's like the preferences are on the person and that I think it's a good thing to be honest yeah” (Portuguese
man).
Discussion
This study intended to explore consumers perceptions on targeted ads on smart TVs. The results revealed that targeted ads arouse mixed feelings regardless of the medium they
are presented on. As stated by Leppaniemi and Karjaluoto (2005), control of the phenomenon and the presence of an opt-out option is highly appreciated and considered an important condition to accept targeted ads.
The subjects expressed similar privacy concerns in the context of targeted ads on smart TVs, reinforcing the words of Van Doorn and Hoekstra (2013) as the dualistic nature of personalisation remains present and true.
The findings confirm what is already know for the other online media in regard to the privacy calculus: the two possible outcomes depend on a series of personal factors. This could be determined by the fact that people do not undergo the mental trade-off consciously or the fact that they have not had yet a first-hand experience of targeted ads on smart TVs and therefore, cannot address the phenomenon differently from the already known online experience. However, the social implications deriving from the more social nature of smart TVs seem to have a role in the calculus for some respondents.
The most relevant results are the insights into the specificity of the medium smart TV. This is strictly related to the fact that TVs are often enjoyed in company, therefore it depends on the type of household the respondents live in. The subjects living on their own or that feel more comfortable around the other family members or roommates had a harder time identifying possible problems deriving from other people seeing their personal ads.
Their perception of the smart TV as a medium also seems to have a relevant role. Even if smart TVs are often enjoyed in company, they are not really perceived as belonging to a different category of devices. Several interviewees who described possible outcomes deriving from others seeing their personal ads, associated this phenomenon with the possibility that, although more unlikely, it can already occur when people show their personal browser or phone to others. This might depend on the fact that people try to associate unknown future outcomes to what they already experience to better relate and accept the future outcomes. In fact, many respondents showed a passive acceptance towards the possibility of new development in targeted advertising, they highlighted how the process is unstoppable and beyond their control and the best they could do is to accept it and derive some advantages from it. Likewise, many affirmed how it is now up to
the user to avoid unpleasant situation by conducting the more sensitive searches online in incognito mode, thus avoiding the option of receiving targeted ads about those.
When asked to picture the possibility of others seeing the results of their private searches through targeted ads, respondent described implications specific to the medium smart TV, both positive and negative. As described by Bronner and Neijens (2018), the device generates a spill-over effect onto the ad experience. For some of the respondents the possibility of substituting generic irrelevant ads with targeted ones is worth the risk of having their personal information displayed to others. For some others it is not and smart TVs, being so public, are not the right medium for targeted ads. Moreover, some participants have expressed doubts on how targeting can be accurately performed on a device used by multiple people, identifying how the efforts to provide a very personalised experience could lead to an opposite outcome.
To conclude, there seem to be a general appreciation for the interconnectivity among personal devices. Even respondents with negative opinions on targeted ads acknowledged the usefulness of having a net of devices remembering their preferences and approve the addition of smart TVs to that.
Theoretical contributions
Discussing the results in relations to the existing literature, it can be seen how they confirm the personalisation-privacy paradox described by Aguirre et al. (2016) and prove that it applies to a medium that had not been taken into consideration yet, smart TVs. Users identified both advantages and disadvantages resulting from targeted ads, and expressed concerns for their privacy that are similar to the ones described for other media.
The current literature on the privacy calculus often does not include social implications, however in this study they are proven relevant when it comes to a medium such as smart TVs. The possibility of others seeing one’s targeted ads and the individual’s type of household are important elements that can tip the balance in the mental trade-off the users undergo. This is an interesting addition specific for the type of media that are usually enjoyed in company. This requires more investigation in the privacy literature, which is often focused on the individual.
The results of this study can be linked to human-technology interaction literature. As per a study by Sundar, Bellur, Oh, Jia, and Kim (2016), the contingency of a medium is described as the extent to which a medium keeps track of the user’s inputs and reacts according all of them, not only to the latest one. Contingency is used to define the message interactivity as in how much “the system’s output is contingent on the user’s input” (Sundar et al., 2016, p. 597). The method described there to provide the users a sense of contingency is the use of digital “breadcrumbs” derived from the past behaviours on a website. The current paper shows how contingency and message interactivity can apply to another medium aside from browsers and how perceived contingency can derive also by the delivery of targeted ads. In addition, in the research by Sundar et al. (2016) it is argued how the user’s awareness of contingency might help mitigate privacy concerns and help the consumers make informed decisions. However, in relation to the display of targeted ads on smart TVs it has emerged how the users’ awareness of the system keeping track of their online behaviours and viewing history might actually increase the privacy concerns, thus opening the floor for a discussion on the methods and measure in which message interactivity affects the users’ privacy.
Moreover, the qualitative approach of this paper is an interesting contribution to the existing literature, mostly composed of quantitative studies. It offers a deeper insight into the consumers’ perceptions and specific real-life examples that can help understand the phenomenon and serve prompts for future researches.
Practical implications
This research offers some relevant information for practitioners interested in developing targeted ads for smart TVs. The results have shown that if done with consideration for the consumers’ privacy, targeted ads are preferred over generic commercials. Respondents have expressed the need for transparency on the data handling and control over the phenomenon, therefore the company should always disclose their purposes and offer an opt-out option. In this way, people who are deeply annoyed by personalisation can avoid it and still receive generic advertising.
Several respondents have expressed real concerns on the possibility of receiving targeted ads about sensitive products (such as health related products) on a public device. Brands of such products should consider whether targeting on smart TVs is the right path for them or find the right way to communicate. They should probably avoid delivering targeted ads on smart TVs altogether, since the public display of their messages might create embarrassing or unpleasant situations for the consumers, thing that would ultimately hurt the brand reputation and image.
Limitations and future research ideas
This study presents several limitations that could be improved in future researches. Firstly, the majority of people in the sample are digital native and highly educated, therefore they have a comfortable relation with technology and feel at ease with devices being smart and online. This might be the reason why they do not perceive smart TV as a very different device from their other personal devices and have a certain or complete knowledge on the phenomenon of online targeted advertising. Further research should involve an even more variated sample in age and cultural background, that might arouse more diverse answers and more surprised reaction to the phenomenon.
Second, some of the subjects had problems picturing the possibility of receiving targeted ads on smart TVs even after having been provided with an explanation and some hypothetical examples. An improvement to this research could be physically showing realistic examples of the phenomenon on smart TVs, thus offering a material proof to the interviewees and ease their understanding.
Further researches could take on the topic with a quantitative approach. Respondents could be presented with two conditions in a realistic setting involving smart TVs and either generic commercials or targeted commercial. Acceptance of targeted ads and privacy concerns could be measured. Additionally, other two conditions involving watching alone or with others could be added. In order to obtain reliable results, the experiment should involve properly targeted ads and a very accurate setting, the ads should be presented on an actual smart TV and in a credible form. The information used to target the ads should be actually based on the respondents’ interests to
enhance ecological validity and mimic in the best possible way how it feels to receive targeted ads in the presence of others. A setting as the one described would require extensive resources and technical preparation.
Conclusion
To conclude, on smart TV as on the other media, targeted ads are acceptable if the company provide transparent disclosure of their practices and the users perceive to have some control. The personalisation paradox applies to the smart TVs as on the other devices, users show both privacy concerns and appreciation for the phenomenon. Being smart TVs a more social medium, they present specific outcomes deriving from the possibility of displaying targeted ads to other people. These social implications, although related to currently existing risks, need to be taken into account and specifically addressed when planning the development of targeted ads for smart TVs.
References
Aguirre, E., Mahr, D., Grewal, D., de Ruyter, K., & Wetzels, M. (2015). Unraveling the
personalization paradox: The effect of information collection and trust-building strategies on online advertisement effectiveness. Journal of Retailing, 91(1), 34–49.
https://doi.org/10.1016/j.jretai.2014.09.005
Aguirre, E., Roggeveen, A. L., Grewal, D., & Wetzels, M. (2016). The personalization-privacy paradox: implications for new media. Journal of Consumer Marketing, 33(2), 98–110. https://doi.org/10.1108/JCM-06-2015-1458
Baek, T. H., & Morimoto, M. (2012). Stay Away From Me. Journal of Advertising, 41(1), 59–76. https://doi.org/10.2753/JOA0091-3367410105
Boerman, S. C., Kruikemeier, S., & Zuiderveen Borgesius, F. J. (2017). Online Behavioral Advertising: A Literature Review and Research Agenda. Journal of Advertising, 46(3), 363– 376. https://doi.org/10.1080/00913367.2017.1339368
Bronner, F., & Neijens, P. (2018). Audience Experiences of Media Context and Embedded Advertising: A Comparison of Eight Media. International Journal of Market Research, 48(1),
81–100. https://doi.org/10.1177/147078530604800106
Busch, O. (2016). Programmatic Advertising. 3–15. https://doi.org/10.1007/978-3-319-25023-6 Calder, B. J., Malthouse, E. C., & Schaedel, U. (2009). An Experimental Study of the Relationship
between Online Engagement and Advertising Effectiveness. Journal of Interactive Marketing, 23(4), 321–331. https://doi.org/10.1016/j.intmar.2009.07.002
De Keyzer, F., Dens, N., & De Pelsmacker, P. (2015). Is this for me? How Consumers Respond to Personalized Advertising on Social Network Sites. Journal of Interactive Advertising, 15(2), 124–134. https://doi.org/10.1080/15252019.2015.1082450
Dijkstra, M., Buijtels, H. E. J. J. M., & van Raaij, W. F. (2005). Separate and joint effects of medium type on consumer responses: A comparison of television, print, and the Internet. Journal of
Business Research, 58(3 SPEC. ISS.), 377–386.
https://doi.org/10.1016/S0148-2963(03)00105-X
Dinev, T., & Hart, P. (2006). An extended privacy calculus model for e-commerce transactions.
Information Systems Research, 17(1), 61–80. https://doi.org/10.1287/isre.1060.0080
Edwards, S. M., Li, H., & Lee, J.-H. (2013). Forced Exposure and Psychological Reactance: Antecedents and Consequences of the Perceived Intrusiveness of Pop-Up Ads. Journal of
Advertising, 31(3), 83–95. https://doi.org/10.1080/00913367.2002.10673678
Fulgoni, G. M., & Lipsman, A. (2017). Measuring Television In the Programmatic Age. Journal of
Advertising Research, 57(1), 10–14. https://doi.org/10.2501/jar-2017-009
Leppaniemi, M., & Karjaluoto, H. (2005). Factors influencing consumers’ willingness to accept mobile advertising: a conceptual model. International Journal of Mobile Communications, 3(3), 197. https://doi.org/10.1504/ijmc.2005.006580
Limpf, N., & Voorveld, H. A. M. (2015). Mobile Location-Based Advertising: How Information Privacy Concerns Influence Consumers’ Attitude and Acceptance Mobile Location-Based Advertising: How Information Privacy Concerns Influence Consumers’ Attitude and Acceptance. Journal of Interactive Advertising, 15(2), 111–123.
https://doi.org/10.1080/15252019.2015.1064795
advertising. International Journal of Advertising, 37(5), 769–784. https://doi.org/10.1080/02650487.2018.1461733
Schumann, J. H., Wangenheim, F. V., & Groene, N. (2013). Targeted Online Advertising: Using Reciprocity Appeals to Increase Acceptance Among Users of Free Web Services. Ssrn, 78(1), 59–75. https://doi.org/10.2139/ssrn.2351777
Sultan, F., Rohm, A. J., & Gao, T. (2009). Factors Influencing Consumer Acceptance of Mobile Marketing: A Two-Country Study of Youth Markets. Journal of Interactive Marketing, 23(4), 308–320. https://doi.org/10.1016/j.intmar.2009.07.003
Sundar, S. S., Bellur, S., Oh, J., Jia, H., & Kim, H. S. (2016). Theoretical Importance of Contingency in Human-Computer Interaction: Effects of Message Interactivity on User Engagement. Communication Research, 43(5), 595–625.
https://doi.org/10.1177/0093650214534962
Sundar, S. S., Jia, H., Waddell, T. F., & Huang, Y. (2015). Toward a Theory of Interactive Media Effects (TIME) Four Models for Explaining How Interface Features Affect User Psychology. In
The Handbook of the Psychology of Communication Technology, First Edition (pp. 47–86).
https://doi.org/10.1002/9781118426456.ch3
Sundar, S. S., & Marathe, S. S. (2010). Personalization versus customization: The importance of agency, privacy, and power usage. Human Communication Research, 36(3), 298–322. https://doi.org/10.1111/j.1468-2958.2010.01377.x
Treiblmaier, H., & Pollach, I. (2011). The influence of privacy concerns on perceptions of web personalisation. International Journal of Web Science, 1(1/2), 3.
https://doi.org/10.1504/ijws.2011.044079
Truong, Y., & Simmons, G. (2010). Perceived intrusiveness in digital advertising: Strategic marketing implications. Journal of Strategic Marketing, 18(3), 239–256.
https://doi.org/10.1080/09652540903511308
Van Doorn, J., & Hoekstra, J. C. (2013). Customization of online advertising: The role of intrusiveness. Marketing Letters: A Journal of Research in Marketing, 24(4), 339–351. https://doi.org/10.1007/s11002-012-9222-1
White, T. B., Zahay, D. L., Thorbjernsen, H., & Shavitt, S. (2008). Getting too personal: Reactance to highly personalized email solicitations. Marketing Letters, 19(1), 39–50.
Ying, L., Korneliussen, T., & Grønhaug, K. (2009). The effect of ad value, ad placement and ad execution on the perceived intrusiveness of web advertisements. International Journal of
Advertising, 28(4), 623–638. https://doi.org/10.2501/S0265048709200795
Zarouali, B., Ponnet, K., Walrave, M., & Poels, K. (2017). “Do you like cookies?” Adolescents’ skeptical processing of retargeted Facebook-ads and the moderating role of privacy concern and a textual debriefing. Computers in Human Behavior, 69, 157–165.
https://doi.org/10.1016/j.chb.2016.11.050
Zion Market Research, (2019, March 05). Global Smart TV Market Will Reach USD 293.5 Billion By 2025: Zion Market Research. GlobeNewswire. Retrieved from:
https://www.globenewswire.com/news-release/2019/03/05/1747722/0/en/Global-Smart-TV-Market-Will-Reach-USD-293-5-Billion-By-2025-Zion-Market-Research.html
Appendices Appendix A
Recruitment text posted on social media:
“Do you own a smart TV and live in the Netherlands? Or do you know someone that does so? PM me! I am looking for 10 awesome people that are willing to have an interview with me for my thesis! I’m searching for people of any age and profession. It’s a 45min/1-hour chat on Skype or face-to-face in English and it’s not going to be very personal, completely anonymous. Be a hero and help this master student graduate, you’ll have my eternal gratitude.”
Appendix C
List of codes after open coding: - Abundance of targeted ads - Accepts all cookies
- Accepts customisation if improves experience
- Accepts only basic cookies
- Accepts suggestions based on viewing history only in regard of more content - Accepts targeted ads only for specific
purposes
- Active use of smart tv - Ads conversation starters - Ads fruition alone more annoying
- Ads fruition alone more prone to ignoring - Ads on tv interrupt more content fruition - Ads on YouTube
- Ads timing is important
- Advantages are for the companies - Advertising is necessary for companies - Age
- Already have many info on him/her - Appreciate more better-timed targeted
ads
- Approves targeted ads based on search history
- Badly targeted ads are very annoying - Becoming more careful
- Benefit from each other's target ads - Benefits outweigh risks
- Better no ads at all
- Better targeted targeted ads are preferable
- Bothered by form of delivery
- Brands know everything about consumers - Carelessness on cookies
- Collecting data is ok when has good purposes
- Company makes ad fruition more interesting
- Concern with parents
- Confident personal ads won't be problematic
- Consumers have options
- Consumers' responsibility to stay safe - Control over content
- Co-viewing is irrelevant
- Data are the price for using services - Data don't provide a full picture - devices are perceived differently - Distrusts brands' goals
- Distrusts brands with data protections - Doesn't accepts company reselling data - Doesn't block targeted ads
- Doesn't browse on smart tv - Doesn't like advertising in general - Doesn't need targeted ads for discounts - Doesn't want interconnection with tv - Doesn't want targeted ads on smart tv - Doesn't want to see others' targeted ads
- Enjoys targeted ads more - Facebook
- Fear due to novelty - Feeling of being tracked
- Form of delivery of targeted ads on smart tv is relevant
- Gave up stopping the process - GDPR is positive
- Generic ads annoying
- Generic ads are more interesting - Generic ads aren't useful
- Generic ads broaden interests - Generic ads waste of time - Google
- Google better than Facebook for ads - Has been hacked
- Having options is important - Hides certain searches
- How data are collected is a discriminant - If there have to be ads better targeted - Interconnectivity is convenient
- Interconnectivity is somewhat scary - Interest in future developments - Interested in others' targeted ads - Interruption of content is annoying - Invasion of privacy
- Irritated by content for others - Isn't bothered by interconnectivity - It's already happening
- It's going too far - It's personal gain
- Knowledge on the matter reduces concern
- Knows targeted ads
- Laptop and tv are more "reflective" devices
- Laptop safer device - Lazy consumer
- Less privacy concerns on smart tv - Likes target ads
- Limits targeted ads because of distrusts - Lives alone
- Lives with family - Lives with partner - Lives with roommates - Looks forward to the change - Measures against fb
- Mixed feelings
- More cautious on laptop
- More concerned on phone because of apps interconnected
- More interconnectivity is appreciated - More prone to keep watching targeted
ads
- Multiple tv users can mistarget ads - Nationality
- Need to be careful giving away data - Needs more safety to appreciate
personalisation
- Needs more transparency from companies
- Never uses incognito - No advantages - No privacy concerns
- No resistance to new phenomenon - Not bothered seeing others' targeted ads - Not worried for other family members
seeing
- Not worried others seeing target ads - Occasional use of AdBlock
- Occasionally have time to choose cookies - Occasionally selects basic cookies only - On demand
- On laptop easier to select cookies policy - On phone is harder to choose cookie
policy
- On some website is easier to choose cookies
- Only one source of data provides one dimensional ads
- Opinion on ads not influenced by watching in company
- Opt out option is important
- Others' targeted ads might be useless - Others' targeted ads more annoying - Passive acceptance
- Pay more attention to target ad on tv - People have more control
- Perceive mismatch
- Perceives control over his/her data - Perceives need to be more informed - Personalisation reduces annoyance
- Personally interested in advertising - Phone and laptop are closer devices - Phone and laptop are more private and
personal
- Phone is most intrusive
- Positive towards renouncing to privacy and getting in return
- Prefers generic ads
- Prefers targeted ads only online - Privacy calculus
- Privacy concern doesn't variate with devices
- Privacy concerns higher for phone - privacy concern variates with devices - Privacy is not that relevant
- Privacy non threatened
- Privacy not invaded if based on previously collected info
- Profession
- Reacts with humour to awkwardness - Realising where data were sourced can
feel invasive
- Really annoyed at YouTube ads - Recognise target ads
- Repetitive ads cause annoyance - Risks outweigh benefits
- Same risk/benefit on all devices - Scared of data collection - Scared of device listening
- Seeing others' targeted ads is awkward - Shares more info on smartphones
- Smart tv is just another screen - Smart tv is more distant
- Smart tv is not a different type of device - Smart tv more private than phone
- Smart tv not the right platform for targeted ads
- Smartphones immediate device - Some fears are irrational
- Somewhat want to avoid targeted ads - Source match device enjoyable
experience
- Source match device less invasive - Source match device more comfortable - Strongly refuses targeted ads
- Takes his/her initiative
- Target ads discover more options - Target ads on smart tv more privacy
invasive
- Target ads somewhat invasive/intrusive - Targeted ads about personal interests - Targeted ads acceptable only on personal
devices
- Targeted ads aggressive - Targeted ads annoying
- Targeted ads are a good way to know people
- Targeted ads are convenient - Targeted ads are creepy - Targeted ads are precise - Targeted ads are relevant
- Targeted ads are too personal for tv
- Targeted ads come too late
- Targeted ads create need for products - Targeted ads create needs in new
moments
- Targeted ads dangerous if politics related - Targeted ads discover new products - Targeted ads ease purchase process - Targeted ads equally acceptable on
different devices
- Targeted ads help early adopters - Targeted ads less acceptable on tv - Targeted ads manipulate into buying
more
- Targeted ads not creepy - Targeted ads off time
- Targeted ads on smart tv can polarise culture
- Targeted ads on smart tv disrupt habit - Targeted ads on smart tv not a serious
threat to privacy
- Targeted ads on smartphones appropriate because on phones can purchase
- Targeted ads on tv can cause unpleasant situation with others
- Targeted ads on tv can disclose private information to others
- Targeted ads on tv equally invasive - Targeted ads reach the right audience - Targeted ads related to instant
gratification
- Targeted ads ruin surprises - Targeted ads save money - Targeted ads scary
- Targeted ads show how many information are collected
- Targeted ads too frequent - Targeted ads too persistent - Targeted ads useful
- Targeted derived annoyance is even less on smart tv
- Too late to change things - Too lazy to prevent it
- Try to gain from an unpleasant situation - Tv different category
- Type of household influence acceptance targeted ads on tv
- Unstoppable process - Used to giving up privacy - Uses a VPN
- Uses AdBlock
- Using viewing history to sell products is creepy
- Very different household components more awkwardness
- Very interested in privacy - Voluntarily looks target ads - Wants more control - Wants to skip generic ads
- Watching ads in company is different - Watching history good predictor of
consumers
- Watching history not a good predictor of consumers
- Ways to overcome embarrassment - When watching relevant content ads are
more relevant
- Wonders where data for targeting were taken
- Worried by others seeing his/her target ads
Appendix D
Appendix E Introduction
Thank you for being here.
In this research, I am interested in hearing your opinions about targeted advertisements on smart TVs.
I’d like to remind you that this interview will be recorded for research purposes.
Please sign this informed consent letter, take your time to read it thoroughly
There are no wrong answers, just tell me freely your opinions about these topics
Let’s start from some background questions: - Nationality
- Age
- Profession - Household
Topic 1: Acceptance of targeted advertising:
• Goal: to find out how people feel like targeted advertising, do they accept or recuse them?
[Personalisation paradox] [Synched cross-media effect] [Personalisation reactance]
Initial question: How do they feel about targeted ads?
• If given the possibility, do you avoid this type of ads?
• Do they recognise any advantages from personalisation, or is it only perceived as a disadvantage?
• How do you feel about the fact that there will be the possibility that your personal online search from private devices (such as smartphones) may influence the ads shown on the TV?
Topic 2: privacy concerns
• Goal: observe what is the privacy-calculus that people perform when targeted ads are on TV
[Privacy concerns] [Privacy calculus]
Initial question: Do you feel that your privacy is being invaded?
• Do you feel like your online or physical privacy is being threatened?
• Do you think it's worth renouncing to part of your privacy in order to obtain a more personalised service?
Topic 3: social factors
• Goal: observe what is the privacy-calculus that people perform when targeted ads are on TV
[Privacy concerns] [Privacy calculus]
Initial question: How do you feel about the fact that ads derived from your personal search will be possibly seen by others?
• How do people react to seeing targeted ads, but aimed to other members of the
household? Imagine that one you are going see an ad on TV that is clearly based on someone else’s search history
• How does co-viewing (watching tv with others) influence your ad experience? • Do you react differently to certain ads if
someone else’s is present? Topic 4: Comparison with other devices and
platforms
• Goal: find some differences with other platforms to see what makes targeted ads on
Initial question: Is the privacy concern for targeted ad on TVs different compared to other online media?
