Measuring advertising effectiveness in Travel 2.0

Physiology & Behavior xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Physiology & Behavior journal homepage: www.elsevier.com/locate/physbeh

Measuring advertising effectiveness in Travel 2.0 websites through eyetracking technology ⁎

Francisco Muñoz-Leivaa, , Janet Hernández-Méndezb, Diego Gómez-Carmonac a

Department of Marketing and Market Research, Faculty of Business and Economics, University of Granada, Campus Universitario La Cartuja, 18071, Granada, Spain Escuela Universitaria de Turismo Iriarte, Universidad de Granada, Santa Cruz de Tenerife Area, Spain c Department of Marketing and Market Research, Faculty of Business and Economics, Campus universitario La Cartuja, s/n. 18071, Granada, Spain b

A R T I C LE I N FO

A B S T R A C T

Keywords: Advertising effectiveness Travel 2.0 websites Eye-tracking technology Neuromarketing Self-reported recall

The advent of Web 2.0 is changing tourists' behaviors, prompting them to take on a more active role in preparing their travel plans. It is also leading tourism companies to have to adapt their marketing strategies to different online social media. The present study analyzes advertising effectiveness in social media in terms of customers' visual attention and self-reported memory (recall). Data were collected through a within-subjects and betweengroups design based on eye-tracking technology, followed by a self-administered questionnaire. Participants were instructed to visit three Travel 2.0 websites (T2W), including a hotel's blog, social network profile (Facebook), and virtual community profile (Tripadvisor). Overall, the results revealed greater advertising effectiveness in the case of the hotel social network; and visual attention measures based on eye-tracking data differed from measures of self-reported recall. Visual attention to the ad banner was paid at a low level of awareness, which explains why the associations with the ad did not activate its subsequent recall. The paper offers a pioneering attempt in the application of eye-tracking technology, and examines the possible impact of visual marketing stimuli on user T2W-related behavior. The practical implications identified in this research, along with its limitations and future research opportunities, are of interest both for further theoretical development and practical application.

1. Introduction The development of Web 2.0 and social media has had an impact in all business activity sectors. This development in tourism has resulted in changes in destination planning behavior [1,2] through the use of tools such as flight metasearch engines, hotel comparators, travel blogs and profiles in virtual communities or social networks. Travelers therefore take on a more active role in decision making and planning of their journey, as well as helping others develop an image about their destination prior to the trip [2]. The development of Web 2.0 websites (W2W) has thus generated users that are more connected than ever before, with a greater access and more active and deeper participation in content creation [3]. The development of Web 2.0 allows users to customize the messages they receive through social networks and to secure only relevant content [4]. Furthermore communication experts use retargeting, that is, a digital marketing technique with the goal of displaying ads to users that have previously interacted with a particular brand [5], including a number of marketing stimuli such as banners, sponsored searches, logos, pop-up ads or video ads in multiple web locations [6]. ⁎

Yet, in order to attract user attention, it is necessary that the content be relevant [7]. Once the banners have caught the user's attention, the aim is for them to click on the advertisement and be redirected to the advertiser's website [8]. At the same time, with the increasing advertising saturation levels, the concept of “banner blindness” has arisen, leading to a decrease in clicks on banners (click-through rate, CTR). Advertisers therefore continue to search for solutions to measure and improve the effectiveness of their messages through the study of the impact of celebrity endorsements on consumer behavior [9,10]. There is a clear need for more research on consumer attention and perception derived from its best location as a prerequisite for the effectiveness of processing visual information. Such research should be founded on experimental and behavioral observation methods [11] as understanding patterns of attention can assist academic marketing knowledge and guide the efforts of management toward more successful advertising results, particularly in the online environment [6]. In order to provide a solution to overcome this challenge this study uses data gathered from banner ad exposure to participants through eye-tracking. The scientific literature review has found a scarcity of

Corresponding author. E-mail address: [email protected] (F. Muñoz-Leiva).

https://doi.org/10.1016/j.physbeh.2018.03.002 Received 29 December 2017; Received in revised form 1 March 2018; Accepted 2 March 2018 0031-9384/ © 2018 Elsevier Inc. All rights reserved.

Please cite this article as: Muñoz-Leiva, F., Physiology & Behavior (2018), https://doi.org/10.1016/j.physbeh.2018.03.002


F. Muñoz-Leiva et al.

eye-tracking metholodogy [19], or self-assessment methodologies including poll-surveys, among others. However, because only a small proportion of visits translate into actual purchase [20], the adoption of user intentions is suboptimal for measuring banner advertising effectiveness [14,18]. Furthermore, research based on self-report instruments (e.g. [21,22]) consider the need for a comprehensive study on the role of memory in the decision-making process of customers in the framework of advertising effectiveness. It is the challenge for our research to actually determine how ads are processed, encoded and stored in human memory and later recalled, since these variables have a direct impact on purchasing decisions [23]. This study therefore uses data gathered from banner ad exposure to participants through eye-tracking (and other self-report measurements of recall). Specifically, we analyze whether task-related selective attention is disrupted by individual banner ads when users are in search mode in different Travel 2.0 websites (T2W). Attention refers to a processing stage of short-term, immediate responses. In fact, one of the main goals of advertising is to attract viewer attention. Drawing consumer attention to an advertising message begins with active information processing [24]. Attention is, therefore, a key mechanism associated with stimulus (e.g. ads) recognition. Different studies concerning the localization, detection and recognition of objects in the visual field indicate a two-stage theory of human visual perception. The first stage is called the “preattentive” mode, in which simple features are processed rapidly and in parallel over the entire visual field. The second stage known as “attentive” mode, a processing focus usually referred to as the focus of attention, is directed to particular locations in the visual field. The analysis of complex forms and the recognition of objects are connected with this second stage [25]. In the current context, visual attention is focused in this second stage with focalization, concentration, and consciousness among its essence.

applications based on the eye-tracking technology or methodology that examine the possible impact of visual marketing stimuli and their effects on user Travel 2.0 websites-related behavior, and also comparing the results with self-reported recall measurements. The goal of this study is to assess the advertising effectiveness of three specific Travel 2.0 tools or websites: a travel blog, a social network, and, a virtual community with tourism-related content. To achieve this objective, we measured different variables of cognitive processing (visual attention), along with post-hoc measures of spontaneous and suggestive recall. We conducted a within-subjects and between-groups experimental design, which was accompanied by a series of personal interviews. We proceed to present a literature review, develop our research questions, outline our methodology, and discuss our key findings and their ensuing implications. 2. Theoretical background: Justification of the research questions 2.1. Cognitive neuroscience methodologies to measure ad effectiveness In the Internet context, the companies want to know mainly: a) how can achieve more effective online advertising strategies and b) how consumers process commercial messages [12]. For the last two decades, several techniques or methodologies are used to solve these problems based on cognitive neuroscience and psychophysiology, biofeedback gauges and facial coding [13] or eye-tracking (e.g. [14]), among others. These methodologies can use to gauge marketing efforts measuring the nonverbal body responses. From this “neuromarketing” approach, the application of these methodologies opens an infinite number of possibilities for studying the attention consumers pay to online advertising and marketing in general. Eye-tracking technology has gained application in different consumer-related disciplines and more specifically in studies related to the online world, for example to quantify ad banner effectiveness. In particular, this technique provides accurate information on consumers' visual attention as fixations and visualization patterns [15]. In particular, the present study adopts the approach of inductive reasoning in which researchers create and analyze large datasets from eye-tracking data so as to identify visualization patterns without designing an explicit model.

2.3. Causes and consequences of banner blindness It is a fact that online advertising distracts users from the website content may be considered as intrusive and can lead to negative perceptions from Internet users, not only toward the advertised products and services, but also the brand and the website itself [8]. The literature review revealed that many users do not recall banners after viewing a website (see Table 1). Moreover, there are users who avoid banners [14,26]. Users not only learn a website's structure quickly [27], but also use their prior navigation experience to avoid banners, thereby focusing their attention on the main content [28]. This is known as “banner blindness” or the fact that users ignore and/or do not pay attention banner content [29–32]. The majority of fixations on banners occur in the first few ocular movements, in order to avoid them during the navigation [31]. Furthermore, users' peripheral vision allows them to skim over the website content and, since ads usually appear in a graphic format, users can quickly filter them from the editorial content [33].

2.2. Visual attention and recall as measurements of ad effectiveness Significant debate has been generated since the early days of ecommerce on how to measure banner advertising effectiveness, which incorporates [16]: a) user behavior (e.g. click-through rate), b) user information processing (e.g. attention, recognition, recall), and c) communication-related characteristics that will generate particular attitudes toward the advertisement, or affect purchase intent [17]. While some authors advocate the use of heuristic metrics to evaluate user behavior, others opt for the use of experimental data by focusing on users' cognitive processing after exposure to banner advertising [18], Table 1 Literature review on banner blindness. Main results

Authors

-

Users do not focus on elements with a similar design to an advertisement (ad), even if they have no promotional purpose. If advertising is integrated in the website content, users cannot quickly identify it, as they consider them to contain irrelevant information. On Facebook, subjects pay more attention to their friends' recommendations than to banners. Therefore, ads are also ignored on social media and banner blindness occurs. In this study, most of participants never fixated on the ad, while those who fixated, did it for a very short time. Those who did see it spent little time looking at it because the ad was located too low on the page. When users use the Internet for simple searches, they do not have to pay so much attention and can therefore perceive and process other stimuli. On the contrary, searches with a greater degree of difficulty require more attention, thus reducing the amount of time users need for processing irrelevant objects, which are therefore ignored. - Only ads that are closely related to the subject's purpose achieve positive results, since users avoid all content that does not correspond to their purpose.

2

Nielsen [3] Abuín [8] Barreto [32] eVOC Insights [34] Burke et al. [31]

León [32]



Research question 2. Do T2W users recall the ads they have been shown?

Usually when advertisements are published in traditional media, such as television and radio, the entire space available is used to capture the spectators' attention. However, ad banners occupy < 10% of the webpage. According to Drèze and Hussherr's [14] study, users ignore at least half of the visible banners. In this case, not only do users not pay attention to the banners, but they also avoid looking at them, which confirms the existence of banner blindness. The main results from different studies about the factors that affect banner blindness are included in Table 1. In contrast to the aforementioned studies (see Table 1), banner blindness does not occur in the study on text advertising conducted by [39], whose results showed that at least 82% of the participants focused on one of the four banners during webpage navigation. In this case, they looked more at the first advertisement they saw during navigation than at the rest of them. Additionally, the IAB Spain Research and The Cocktail Analysis study [40] revealed that 75% of the banners analyzed during the experiment received a visual impact from at least 50% of the participants. Therefore, the scientific literature review shows results inconclusive. Based on this review, in our study we will answer the following question:

3. Method: Experiment details 3.1. Fieldwork and eye-tracking system used The fieldwork was carried out at the University of Granada's Mind, Brain and Behavior Research Center (CIMCYC). The participants were recruited by means of the quota sampling method by e-mail or by phone from an initial list of subjects, and compensated €15 for their time. The final sample comprised 60 participants. Samples of this size are common in eye-tracking studies and are not intended to be representative [50–53]; indeed student samples are often used (e.g. [16,19,54]). In our case, however, we applied quota-based sampling to attain a final sample that was as representative as possible of the general population. More specifically, the sample was divided (i) by gender, specifically men and women (50% in each case) who used social networks (according to the [55]) and (ii) by average Facebook-user age (approx. 35 years, according to [56]). This produced four distinct quotas, each comprising 15 subjects: males between 18 and 34 years of age, females between 18 and 34, males aged 35 or above, and females aged 35 or above. Participation in this experiment took place one participant at a time in a quiet room, isolated from outside noise, with an ambient light of 200 Lux, as recommended in International Telecommunication Union [57] to simulate a “home environment”. Eye movements were recorded using an infrared video eye-tracking device (Tobii T60 eye-tracker, Tobii Technology AB, Sweden; sampling corneal reflection and pupil at 60 Hz). The system has a spatial tracking accuracy of approximately 0.2° of visual angle and is integrated into a 17″ TFT monitor, with a screen resolution of 1280 × 1024 pixels. The calibration was run on nine points to optimize spatial tracking accuracy; and the data were processed with Tobii Studio software.

Research question 1. Do users pay visual attention to a banner located in different T2W or, on the contrary, they ignore them (banner blindness)? And what T2W do ad banner more effective?

2.4. Banner recognition memory The problem of measuring attention when exposed to visual marketing stimuli results in measures that are too closely related to others, such as recall or expressed activation. Nevertheless, attention can often be paid with a low level of awareness. In other words, the information can be actively processed by consumers, in interaction with internal knowledge representations already present in their memory or with external information (e.g. brand name, endorsement, etc.). Thus, external information may be “enriched” or misinterpreted due to (spontaneous) associations co-activated in the brain [11]. A great number of studies conducted along these lines in the field of psychology argue that the cognitive response directly depends on the degree of attention focused on the stimulus (e.g. [41]). This suggests the existence of a positive correlation between the degree of visual attention afforded to the ad banner and self-reported recall [42,43]. In the advertising context, effective advertising not only captures the attention, but persists in short or long-term memory [32]. Therefore, advertisers should not only strive to persuade individuals to pay attention to their ads, but also to recall them. The fundamental reason is that most of the message is not retained by consumers and plays a limited role in informing them of their choices [44]. After reviewing the relevant literature, it was confirmed that participants in research works in general do not recall most ads [14,31,45,46]. Concretely, according to Crespo [47], what people most recall about the banner is the brand. Köster et al. [48], by contrast, indicate that although personalizing the banner increases content recall, it does not for the logo. Yeun-Chun et al. [49], in turn, depending on the message content, identify higher recall rates for the brand either when ads are simple and accompanied by contextual information or when complex devoid of context. The results of the study conducted by Drèze and Hussherr [14] revealed that 46.9% of the subjects claimed to recall some advertising on the website. Besides, two real and two false ads were shown to them, without identifying any difference between the false and the real ones in terms of recognition. The results from different studies about the factors that affect banner recall are shown in Table 2. Therefore, our last research purpose is to check whether users recall the content of the ad concerned. In particular, this study seeks to answer the following question:

3.2. Experimental design This research consists of a within-subjects experimental design where all participants visit replicas of three T2W of Hotel Jardín Tropical (Tenerife, Spain): its blog (url: http://webcim.ugr.es/polls/EP_ET/B.html), its Facebook profile (url: http://webcim.ugr.es/polls/EP_ET/F.html) and its Tripadvisor profile (url: http://webcim.ugr.es/polls/EP_ET/T.html); as previous studies have also analyzed (e.g. [2,32,58,59]) or justified (e.g. [2,60,61]). Each website also includes imbedded vertical Air Europa airline rectangle banner (see Fig. 1). The banner contains text (“We fly just for you! Visit” + URL + “for the chance to win prizes every week”) as well as a composite image (three celebrities and plane). The final design therefore consisted of the following presentation formats for T2W (Table 3). The participants were randomly assigned to the counterbalancing groups, and each participant saw the webpages displayed in three different ways. The different displays were chosen to present the experimental scenarios in order to mitigate the eventual influence of the presentation order, thus creating three experimental groups (EG). Once created, these three groups were also counterbalanced based on gender and different age ranges. Randomness was ensured by assigning participants for the experimental groups (EG), and all the websites (or stimuli) were presented in a random order; allowing an equal distribution of the effects of the independent variables or factors under all possible conditions [62]. During navigation, participants were assigned a task which consisted in searching for information about the room views offered by the hotel, in text format integrated in the editorial content. This mechanism is intended to achieve a certain degree of implication with the experiment and goal-oriented navigation similar to what would occur in a normal situation using T2W. 3



Table 2 Literature review on banner recall and recognition. Main results

Authors

longer a person remains on a website, the more likely they are to recall the ad. • The who have the objective of performing a task on the website are less likely to recall banners than people who surf without any specific • Subjects purpose. banner repetition leads to greater recall of a brand (and greater intention to click on the ad) in an online environment. • The additional exposure to a banner improves brand recall in Chinese users. • One • Banners located on the top of the screen are more frequently recalled than those on the inside or on the bottom of the page.

Danaher and Mullarkey [35]

Yaveroglu and Donthu [36] Gong and Maddox [37] Burke et al. [31], Nihel [38]

a. Reception, explanation of the steps to follow and ethics toward participant. The participant was seated in front of a computer and the steps he/she would have to follow were explained: ethical issues, calibration, navigation and final questionnaire or post-test. In this stage, the subject had to sign a “consent form”, i.e., a written statement with the basic information about the research project, what the experiment was about and personal information (name, surname and signature). The possibility to check the objectives and methodology of the research project containing the study was also proposed to them. This way, the participant is fully informed about the study purpose and the consequences of signing the rights to their personal information (according to [63]). The consent form was signed before data recording. b. Calibration process. A nine-point calibration process was conducted at the beginning of each experimental scenario, i.e. three times per participant. Specifically, for the calibration procedure, the subject is instructed to successively focus on the center of nine separate red dots (of radius 1 cm) displayed on a 3 × 3 grid mounted on a screen at a normal viewing distance of about 80 cm. This allowed us to recalibrate the system in case of any miscalibration. c. Navigation through the experimental scenarios. Next, the participant automatically moves into an inspection of the three T2W for a total duration of four and a half minutes (90 s for each tool). The order of the tools differed between experimental groups (see previously described design). d. Post-test. Finally, participants were moved to a second computer outside the room equipped with the eye-tracker where they responded to an online questionnaire comprising queries regarding socio-demographic notions, behavioral variables, as well as others items intended to determine whether they were able to recall the banner observed when consulting the different etourism websites (see Appendix A). Fig. 1. Ad banner used in the study.

3.4. Eye movement analyses and statistical analyses

Table 3 Experimental design.

To test how attention was distributed across the stimuli, we divided the T2W into several areas of interest (type of AOIs): header of the website, posts, ad banner, customer's comment with the hotel view (task), bottom of the page and other areas… (see figure in Appendix B). These AOIs were manually delimited in order for all fixation measurements to properly refer to it. The number of fixations or fixations count (FC) within AOI is a very general measurement [63] which needs to be completed with others, such as fixation duration [63]. Therefore, the following measures or metrics of fixations and gaze durations were additionally included: time to first fixation on the AOI (TFF), number of fixations before getting to the AOI (FB), fixation duration or length (FD) and total fixation duration (TFD). They were calculated for every AOI. In order to achieve our research objective different means were extracted for every website and an analysis of variance (ANOVA) with repeated measures was carried out, according to AOI type and T2W as between-subject factors and the attention metrics (TFF, FB, FD, TFD and FC) as outcome or dependent variables. Following Simola et al. [64] methodology to study the sequence at which the different specific areas (AOIs) were fixated, the proportion of participants that had fixated

EG1: B/F/T; n = 20 EG2: F/T/B; n = 20 EG3: T/B/F; n = 20 Where B = Hotel blog with ad banner; F = Hotel Facebook profile with ad banner; T = Hotel Tripadvisor profile with ad banner.

Although some authors support goal-oriented navigation in contrast to exploratory or free navigation, the former generates less recognition of advertising in the navigation environment [35]. But the fact of participants focusing on conducting a normal task in searching for information about a hotel should not be considered an impediment or difficulty for recall or recognition of advertising messages. 3.3. Data collection and recording process The following is a description of the stages of the process conducted with each participant: 4



Λ = 0.291; F = 16.74; d.f.1 = 40; d.f.2 = 2042.76; p = .000) had all a significant impact on visual attention in general. Next, we analyzed what happened with every visual measurement by adopting a univariate approach. This cross-check required verifying the assumption of sphericity for the error covariances matrix of the transformed outcome variables through Mauchly's test. The test results proved that there is sphericity in this matrix for the T2W but for the AOIs (and its interaction with the type of tool) a transformation should be used (e.g. Greenhouse-Geisser). This way, considering the type of AOI, users fixate first on the posts (TFF = 1.77 s and FB = 4.28; see Table 4), then on the header (TFF = 4.81 s and FB = 16.14), next on the ad banner (TFF = 20.49 s and FB = 65.57), on the customers' comments (TFF = 25.53 s and FB = 84.36) and, finally, on the lower part of the website (TFF = 30.99 and FB = 103.64). The same applies to the TFD and FC, although in these measures, the ad banner is the area with the shortest fixation duration (TFD = 2.41) and the lowest fixation count (FC = 11.44), with an average fixation duration of 169 milliseconds. To verify whether banner blindness existed across each T2W, the participant's gaze analysis revealed that 95% of the participants fixated at least once on the ad banner located on the blog, 98% of them on the ad on Facebook, and finally, 67% of the participants focused their attention on the ad located on Tripadvisor. Although on a different level, we can confirm that banner blindness does not occur on the websites under study here in terms of attention. Below, a comparison of the different levels of ad effectiveness in terms of attention paid, between the different T2W, is shown in Table 5. These ANOVAs for the metrics that received higher main effects (FTFF (2, 153) = 17.98, p < .001; FTFD (2, 153) = 15.86, p < .001; FFC (2, 1534) = 18.58, p < .001) reveal that participants take less time to focus on the ad banner when viewing Facebook (TFF = 14.16 s; see Table 5), followed by the blog (23.10 s) and finally by Tripadvisor (38.56 s). This proves that they somehow recognize the advertisement with their peripheral vision, but spend very little or no time to focus on it. Although these are average values, it was confirmed that the first fixation on the banner does not coincide with the first ocular movements (as found by [31]). The TFD is higher for the banner on Facebook (4.08 s), followed by the one on the blog (2.43 s) and lastly by Tripadvisor (1.23 s). And participants focus more times on the banner on Facebook (FC = 19.05), followed by the one on the blog (11.72), and finally by Tripadvisor (6.08). The following profile graphs (Fig. 2) show the impact on the attention measurements of every AOI in each type of website where they are integrated. In particular, participants fixated first (in terms of time and frequency), for longer and more often on the header (TFF = 0.3 s, FB = 0.7, FD = 0.23 s and FC = 102.6) and the ad banner (TFF = 0.14 s, FB = 38.4, FD = 0.20 and FC = 19.2) on Facebook as compared to the rest of websites. Effectiveness is lower on the blogs and, finally, on Tripadvisor. Regarding the posts and the lower part of the T2W, the subjects paid more attention on Tripadvisor, followed by the blog and finally by Facebook. And customers' comments are viewed before and for a longer time on Facebook, followed by Tripadvisor and then by the blog. In summary, we can conclude that not only do participants spend less time to locate the banner on Facebook profile, but also their

them was calculated during the first 100 ordinal positions from the beginning of the trial. And then other repeated measures ANOVA was computed, with AOIs, T2W and fixation order (additionally fixations were averaged into series or ranges of 10 fixations) as independent variables. Furthermore, we measured the degree of banner recognition memory from the self-reported measurements by quantitative descriptive analyses. The first question allowed us to analyze whether they were able to recall the any advertisement on the visits. Then, the participants that indicated that they remembered an advertisement had to answer two specific questions to measure their spontaneous recall of the banner in question. In this way, participants had to indicate the brand or company name and to describe the content of the image they had viewed. The participants' suggestive recall was analyzed by showing them a list of companies (Iberia, Air Europa, Booking and Don't know/No response option) with a slogan for a recent campaign. 4. Results 4.1. What T2W do ad banner more effective? The general viewing pattern of the three websites through the visual metrics selected is described below (see Table 4). In general, the time to reach the ad banner (our AOI) was 20.49 s (TTF), it was needed 65.57 fixations outside before getting it (FB), and the fixations had an average duration of 0.17 s (FD) with a total duration of 2.41 s. (TFD) for a number of fixations equal to 11.44 (FC). To conduct a more in-depth analysis of the first research question it was tested whether the type of AOI and the type of T2W in which each website is divided have a significant impact on the participants' attention, through a repeated measures ANOVA. First, the results of the multivariate tests showed that both factors type of AOI (Wilks' Λ = 0.037; F = 65.78; d.f.1 = 20; d.f.2 = 770.41; p = .000), type of T2W (Wilks' Λ = 0.588; F = 6.93; d.f.1 = 10; d.f.2 = 228; p = .000) and the AOI-tool interaction effect (Wilks' Table 4 Repeated measures ANOVA: Global and marginal means for the type of AOI. Measure

Type of AOI

Average

Stand. error

TFF

Header Posts Banner Bottom Customer's Total Header Posts Banner Bottom Customer's Total Header Posts Banner Bottom Customer's Total Header Posts Banner Bottom Comment Total Header Posts Banner Bottom Customer's Total

4.81 1.77 20.49 30.99 25.53 16.72 16.14 4.28 65.57 103.64 84.36 54.80 0.20 0.21 0.17 0.16 0.19 0.19 10.73 44.32 2.41 4.45 5.92 13.57 46.08 207.27 11.44 21.35 26.35 62.50

0.91 0.28 1.69 1.93 1.60 0.74 3.20 0.59 5.8 6.74 5.81 2.91 0.01 0.01 0.01 0.01 0.01 0.01 0.74 1.72 0.21 0.53 0.53 0.47 2.94 6.57 0.86 2.35 2.04 1.77

FB

FD

TFD

FC

comment

comment

comment

comment

Table 5 Means and standard deviations in TFF, TFD and FC for each T2W. Tool

Blog Facebook Tripadvisor

5

TFF (s)

TFD (s)

FC (times)

Mean

St. dev.

Mean

St. dev.

Mean

St. dev.

23.10 14.16 38.56

18.42 13.26 28.34

2.43 4.08 1.23

1.91 3.12 2.50

11.72 19.05 6.08

8.27 12.58 10.32



Fig. 2. Profile graphs. Interaction between type of AOI and T2W.

fixated on each AOI and/or T2W as measurement of efficacy. In particular, two different graphs were created for each T2W, illustrating the proportion of participants who fixate on the different AOIs. The first graph shows the first 100 fixations, while the second graph displays fixations averaged in ranges of 10 from the beginning of the test. At the beginning, participants fixate mainly on the posts and the blog header (see Fig. 3). In this case, the effectiveness of the posts group increases over time; while on the contrary, the efficacy of the header decreases. After approximately 15 fixations, fixation on the banner is more and more frequent and the bottom of the blog also starts gaining slightly significance. The difference between these two elements is that fixations on the bottom of the blog follow a constant trend from the beginning of the test, while the subjects pay less and less attention to the banner after approximately 75 fixations. Furthermore, although participants take more time to fixate on the customers' comment (task) and sometimes they don't even do it, they follow an ascendant trend once the number of fixations also starts to increase. Therefore, the area of the blog with the greatest effectiveness is the one containing the different posts. On Facebook, participants focus mainly on the header; although in this case they pay less attention to the posts and the banner at the beginning of the trial (see Fig. 4). The header also starts losing efficacy over time, while the posts and the banner gain it, especially the posts.

fixations are more frequent and longer. In terms of advertising effectiveness, Facebook is followed by the blog and, finally, by the Tripadvisor profile. The explanation may lie in the greater simplicity of the editorial content posted by companies on Facebook and its wellordered organization of elements. 4.2. Ad effectiveness of the different elements of a T2W, measured in percentage of participants who focus on them at different latencies Now that it is clear on which AOI from every T2W users focus first, it would be useful to know what AOI is more effective over time and, on the contrary, which one becomes less effective first. To this end, the proceeding proposed by Simola et al. [64] will be applied. In particular, to study the sequence of fixations for the different AOIs on each website, a new repeated measures ANOVA was applied with the AOI, T2W and fixation order or decile (as fixations averaged into ranges of 10) as between-subjects factors. In this case, an incomplete factorial design was identified, including only the interaction AOI-tool. The results showed that the fixation order (Wilks' Λ = 0.71.; F = 2.32; d.f.1 = 9; d.f.2 = 51; p = .028), the AOIs (Wilks' Λ = 0.09; F = 145.81; d.f.1 = 4; d.f.2 = 56; p = .000), the websites (Wilks' Λ = 0.49; F = 30.01; d.f.1 = 2; d.f.2 = 58; p = .000) and the AOI-tool interaction (Wilks' Λ = 0.14; F = 39.39; d.f.1 = 8; d.f.2 = 52; p = .000) had a significant impact on the percentage of subjects that 6



Fig. 3. Proportion of participants fixating on a specific AOI on the blog. (a) Fixations of the first 100 ordinal positions and (b) fixations averaged in ranges of 10 fixations from the beginning of the trial.

during their visits, while the other 38% could not remember any one. Then, the results showed that over half of the participants (55%) did not remember the brand or company in the advertisement and only 10% of the subjects indicated that it was about the airline (Air Europa). Other responses of participants associated the advertisement to other companies, such as Iberia, Spanair, Booking, etc., advertisements for rental cars, promotions for other hotels, or an airline without a brand. The description of the content of the advertisement they remembered seeing showed that 28% of the sample could not recall the image and only 3% remembered accurately its whole composition: a plane superimposed on three celebrities on a black background. And 10% of the participants remembered the name of some of the celebrities. Other participants recalled that the banner included people or couples (30%) without specific names, and referring to the advertisement concerned, 5% of them recalled that there was an airplane in it, and 10% recalled a golf course (on Facebook). We can therefore deduce that the advertisements the participants recalled were different from the banner in question (for instance, the golf course promotion on the Facebook page, the image of a girl in a massage room on both the blog and Tripadvisor, etc.). Finally, the analysis of the suggestive memory show that 40% of the participants didn't know/no response and 37% selected the correct option (Air Europa).

Likewise, although participants take some time to look at the customers' comment, the first maximum is reached during the first 75–80 fixations; after that moment, visual attention remains constant. The latter also happens with the bottom of the Facebook page. Therefore, the area containing the posts is the most effective one also on Facebook, after approximately 30 fixations. On the Tripadvisor profile, from the beginning of the test, participants also focus mainly on the posts (see Fig. 5). In this case, less sudden changes in the trend were observed than in the previous cases. The bottom of the blog, the banner and the customer's comment adopt a slightly ascendant behavior as fixations increase. In the case of the comment, the highest efficacy level for the goal of the test (task) is reached between the 65th and the 75th fixation. The same thing applies to the banner, at the end of the trails. However, in spite of small variations, the second graph reveals that the behavior of the different AOIs remains quasi-constant, without sudden changes. Therefore, the different elements of the Tripadvisor profile are quite effective as fixations increase. To sum up, the area containing the posts is the most significant part of the three T2W, since is it the area to which most attention is paid as the number of fixations increases. This is due to the fact that participants look for the information they need for completing the assigned task in this area, as might happen in a realistic scenario. 4.3. Banner recognition memory

5. Discussion of results After visiting the hotel's T2W, the participants answered a self-administered questionnaire which allowed us to analyze whether they were able to recall the banner concerned. The results revealed that, in general, 62% of the participants recalled having seen an advertisement

The development of the Web 2.0 has affected the tourism industry and has also changed tourists' and companies' Internet use behavior. Tourists have therefore adopted a more active role in managing their 7



Fig. 4. Proportion of participants fixating on a specific AOI on Facebook. (a) Fixations on the first 100 ordinal positions and (b) fixations averaged in ranges of 10 fixations from the beginning of the trial.

the different subjects who explore it [51,66]. In our case, the Facebook profiles had less editorial content than the rest. Besides, if users have a better memory or more experience of how a Facebook page is structured, in contrast to a new blog for example, it would increase banner blindness. This would make social networks' advertising effectiveness even stronger. Therefore, unlike the study of Simola et al. [64] which proved that attention to the different elements of a website (ads, logos and headers) improved recall of these elements, our researched revealed that there are differences between attention measurements and recall responses. Eye-tracking data confirmed that most of participants fixated at least once on the ad banner (so banner blindness did not apply) for the reasons already mentioned. But, the total fixation duration for the banner is notably lower than for other areas in the three tools analyzed (posts, header, etc.). Once the participants looked at the banner, they probably recognized it was an advertisement and deduced that the information they needed for completing the task was not there, so they did not spend more time on it. Users immediately learnt the website structure and used their navigation experience to quickly filter the banners from the editorial content [27,33] or avoid them during future visits [14,31]. Concerning the second research question, the analyses of the selfreported questions did reveal that the subjects barely recalled the content or the brand of the advertisement concerned. In fact, more than half of the subjects did not recall the brand/company in the banner concerned, and approximately a third of the users selected the correct answer about the brand and slogan advertised. Another reason for these findings may be related to the user's type of navigation, which in the present case was goal-oriented. When they are searching for specific information on the Internet and the content they are searching for is not in the advertisement, subjects will most likely

travel plans, and tourism companies have taken advantage of this opportunity to adapt their marketing strategies to the new electronic environment. The scientific literature review led to a series of interesting findings concerning advertising in W2W or social media, which helped us specify our research questions and complete our results. No prior studies were identified that analyzed the advertising effectiveness of ad banners in terms of visual attention paid in these W2W, or the results were inconclusive. However, considering that most of the studies analyzed involved banner blindness, users either avoided or ignored the banners during navigation (e.g. [14,26,30–32]). In regards to ad banner recall, some authors confirm that many users do not recall banners after viewing a website [14,31,45,46]. Nevertheless, in this case, the results were inconclusive and sometimes even contradictory. Concerning the first research question, using the eye-tracking technology we can confirm that banner blindness does not occur in the case of T2W, in line with the outcomes of other studies (e.g. [39,40]). That could be due to the fact that the image of the banner in our study illustrated three famous celebrities that might have caught their attention during navigation. The results of the study conducted by Djamasbi et al. ([65]; study focused in generation Y) show that, when viewing a website, users focus more on the primary images, such as faces of famous people. Furthermore, it was confirmed that advertising on Facebook is more effective, followed by the blog and lastly by Tripadvisor with more amount of information. We could say that participants not only focused on the banner sooner, but also more times and for longer, although it was located in the same position on all the websites. This may be due to the fact that the complexity of a website's design (text size and format, position of images, etc.) can have an effect on the viewing patterns of

8



Fig. 5. Proportion of participants fixating on a specific AOI on Tripadvisor. (a) Fixations of the first 100 ordinal positions and (b) fixations averaged in ranges of 10 fixations from the beginning of the trial.

However, as high recall rates were not obtained in this study for none of the T2W analyzed, we recommend advertisers to conduct a prior study of the website on which they want to display their advertisements, as well as of the type of task users carry out the most on that website (for instance, purchase of a given product, particular information search, etc.). This way, both the design and the content of the banner will be oriented to the user's goal, thus avoiding their annoyance, improving attention rates and enhancing banner recall. Another recommendation to advertisers is to include celebrities in the banner image, as this generally captures users' attention immediately. Of course, as explained above, the characters have to be related to the editorial content (and, if possible, to the main task users normally perform) on that website. This way, they will feel more engaged toward the banner and will be able to recall it. Previous studies proved that the advertisements in coherence with the editorial content are more easily recognized than the incoherent ones [64]. For instance, if we want to insert advertising in a travel blog, we can embed the image of a relevant celebrity, such as an actor of a movie filmed in a pleasant place.

avoid the areas of the screen containing banners [67]. In this case, users could perceive advertising as an obstacle in their information search [68]. These results are coherent with those obtained by Danaher and Mullarkey [35], who found that subjects with a task-related objective during online navigation, they are less likely to remember the banners than those who are surfing the Internet without a purpose. For all these reasons, like in many other cases, visual attention to the ad banners happens at a low level of awareness, which explains why the associations with them did not activate their recall later.

6. Contributions, limitations and future research 6.1. Implications of the study We start out from a situation where online advertising can annoy users, thereby decreasing the performance for the task assigned to them [69]. But our study can help advertisers and marketing companies choose the best Travel 2.0 websites (T2W) for their banners and the moment of being shown. First of all, concerning attention paid to the banner, it was confirmed that those T2W that are not overloaded with editorial content – such as Facebook and simple blog – are more effective in terms of investment in advertising. Besides, the efficacy of a banner increases from half a hundred of fixations, meaning that the banner could appear after a reasonable time-lapse to allow information search. This time-lapse can vary from a few seconds on Facebook to a few more on the blogs and around half a minute on Tripadvisor.

6.2. Limitations and future research The limitations of this study are derived from: a) the sample size and the sampling method, b) the specific T2W chosen, and c) the type of navigation of the participants. First, the major limitation of this study is that its sample size is small, albeit comparable to the majority of eye-tracking studies [70].

9



We endeavored to ensure that the final sample was representative of the general population, by using quotas. But nevertheless the sample size, coupled with the non-random selection method, require the results to be generalized with caution. Second, our study analyzes ad effectiveness considering three specific T2W (Blog, Facebook and TripAdvisor). Future research should complement this study by analyzing other online advertising tools in social media such as sponsored tweets, Twitter, Pinterest or Instagram, as well as other means of measuring advertising effectiveness such as: a) CTR, to determine which subjects voluntarily visit the advertiser's website or b) their purchasing behavior. Future research should also focus on eye-tracking techniques in the use of smartphones or social network mobile applications, since advertisers in recent years are resorting to interstitial banners to promote their products on these devices. Third, a specific experimental design was also developed using the eye-tracking technique, to evaluate the participants' fixation behaviors. The strict, careful application of the experimental design achieved an

improvement in the internal and external validity of the results obtained. However, the sample size, did not lend itself to a stronger and more complex experimental design combining other factors (betweengroups), such as experience with different levels. In the future, we hope to replicate this design with a larger sample of individuals, to analyze the interactions between factors, reducing the error margin and obtaining more robust results. Finally, it would be interesting to compare advertising effectiveness in a context of exploratory or free Internet navigation and compare the results using a task like ours (goal-oriented navigation). Prior studies claim that banners attract less attention during reading-oriented tasks or information search than during free navigation [71,72]. Funding This study was conducted with the financial support received from the Spanish National Research Programmes (R+D+i Research Project ECO2017-88458-R and Research Project ECO2012-39576).

Appendix A. Questionnaire Table A1 Questionnaire questions.

10

Appendix B. Areas of the T2W

Fig. B1. Areas of the hotel's blog, Facebook page and Tripadvisor profile.



11



References

org/10.1016/j.chb.2010.11.019. [29] J.P. Benway, D.M. Lane, Banner blindness: Web searchers often miss ‘obvious’ links, Internetworking: ITG Newsletter, 1.3, Rice University, Dez, 1998. [30] Benway, J.P. (1998), “Banner blindness: the irony of attention grabbing on the World Wide Web”, in Proc. Hum. Fact. Ergonom. Soc. Annu. Meet., SAGE Publications (Vol. 42, no. 5, pp. 463-467). doi:https://doi.org/10.1177/154193129804200504. [31] M. Burke, A. Hornof, N. Lewis, N. Gorman, High-cost banner blindness: ads increase perceived workload, hinder visual search, and are forgotten, ACM Trans. Comput. Hum. Interact. 12 (4) (2005) 423–445, http://dx.doi.org/10.1145/1121112. 1121116. [32] A.M. Barreto, Do users look at banner ads on Facebook? J. Res. Interact. Market. 7 (2) (2013) 119–139, http://dx.doi.org/10.1108/JRIM-Mar-2012-0013. [33] J.L. León Sáez de Ybarra, Nuevos Soportes y mercados de la publicidad digital. Transiciones y experiencias, Pensar la Publicidad. Rev. Int. Invest. Publ. 3 (2) (2009) 17–30, http://dx.doi.org/10.5209/rev_PEPU.2009.v3.n2.16066. [34] eVOC Insights, Methodology Spotlight, Retrieved from http://evocinsights.com/ 2009/01/?cat=4, (2009) , Accessed date: 11 November 2016. [35] P.J. Danaher, G.W. Mullarkey, Factors affecting online advertising recall: a study of students, J. Advert. Res. 43 (3) (2003) 252–267, http://dx.doi.org/10.2501/JAR43-3-252-267. [36] I. Yaveroglu, D. Naveen, Advertising repetition and placement issues in on-line environments, J. Advert. 37 (2) (2008) 31–43, http://dx.doi.org/10.2753/ JOA0091-3367370203. [37] W. Gong, L.M. Maddox, Measuring web advertising effectiveness in China, J. Advert. Res. 43 (1) (2003) 34–49, http://dx.doi.org/10.1017/ S002184990303006X. [38] Z. Nihel, The effectiveness of internet advertising through memorization and click on a banner, Int. J. Market. Stud. 5 (2) (2013) 93, http://dx.doi.org/10.1017/ S002184990303006X. [39] G. Hervet, K. Guérard, S. Tremblay, M.S. Chtourou, Is banner blindness genuine? Eye tracking internet text advertising, Appl. Cogn. Psychol. 25 (5) (2011) 708–716, http://dx.doi.org/10.1002/acp.1742. [40] IAB Spain Research and The Cocktail Analysis, Claves sobre la Interacción visual con la publicidad web. Aplicación de la técnica eye tracking, Retrieved from http:// iabspain.es/wp-content/uploads/Estudio_IAB_Eyetracking_junio_09v2_Modo_de_ compatibilidad.pdf, (2009) , Accessed date: 10 November 2016. [41] F. Shen, Q. Chen, Contextual priming and applicability: implications for ad attitude and brand evaluations, J. Advert. 36 (1) (2007) 69–80, http://dx.doi.org/10.2753/ JOA0091-3367360105. [42] K. Gidlöf, N. Holmberg, H. Sandberg, The use of eye-tracking and retrospective interviews to study teenagers' exposure to online advertising, Vis. Commun. 11 (3) (2012) 329–345, http://dx.doi.org/10.1177/1470357212446412. [43] N. Scott, C. Green, S. Fairley, Investigation of the use of eye tracking to examine tourism advertising effectiveness, Curr. Issue Tour. 19 (7) (2015) 634–642, http:// dx.doi.org/10.1080/13683500.2014.1003797. [44] W. Brazil, B. Caulfield, What makes an effective energy efficiency label? Assessing the performance of energy labels through eye-tracking experiments in Ireland, Energy Res. Soc. Sci. 29 (2017) 46–52, http://dx.doi.org/10.1016/j.erss.2017.05. 014. [45] Bayles, M.E. & Chaparro, B. (2001), “Recall and recognition of static vs. animated banner advertisements”, Proc. Hum. Factors Ergonom. Soc. Annu. Meet. (Vol. 45, no. 15, pp. 1201-1204). SAGE Publications, doi:https://doi.org/10.1016/10.1177/ 154193120104501510. [46] M.E. Bayles, Designing online banner advertisements: should we animate? Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Minneapolis, MN, April 2002, ACM Press, New York, 2002, pp. 363–368, , http:// dx.doi.org/10.1145/503376.503441. [47] E. Crespo, Eficacia de la promoción de ventas on-line, Influencia del tipo de incentivo promocional y la experiencia de uso web, Departamento de Comercialización e Investigación de Mercados, Universidad de Granada, 2011. [48] M. Köster, M. Rüth, K.C. Hamborg, K. Kaspar, Effects of personalized banner ads on visual attention and recognition memory, Appl. Cogn. Psychol. 29 (2) (2015) 181–192, http://dx.doi.org/10.1002/acp.3080. [49] K. Yeun-Chun, J. Hee-Song, C.R. Hollenbeck, J.H. Lee, Are contextual advertisements effective? The moderating role of complexity in banner advertising, Int. J. Advert. 33 (2) (2014) 351–371, http://dx.doi.org/10.2501/IJA-33-2-351-371. [50] S. Miellet, X. Zhou, L. He, H. Rodger, R. Caldara, Investigating cultural diversity for extrafoveal information use in visual scenes, J. Vis. 10 (6) (2010) 21, http://dx.doi. org/10.1167/10.6.21. [51] B. Pan, H. Hembrooke, G. Gay, L. Granka, M. Feusner, J. Newman, The determinants of web page viewing behavior: an eye-tracking study, in: A. Duchowski (Ed.), Proceedings of Eye Tracking Research & Applications Symposium (ETRA 04), ACM, New York, 2004, pp. 147–154, , http://dx.doi.org/10.1145/968363.968391. [52] B. Pan, L. Zhang, R. Law, The complex matter of online hotel choice, Cornell Hosp. Q. 54 (1) (2013) 74–83, http://dx.doi.org/10.1177/1938965512463264. [53] M. Wedel, R. Pieters, Eye tracking for visual marketing, Found. Trend. Market. 1 (4) (2008) 231–320, http://dx.doi.org/10.1561/1700000011. [54] L. Pasqualotti, T. Baccino, Online advertisement: how are visual strategies affected by the distance and the animation of banners? Front. Psychol. 5 (211) (2014) 1–11, http://dx.doi.org/10.3389/fpsyg.2014.00211. [55] Spanish Statistical Office, Instituto Nacional de Estadística, Encuesta sobre Equipamiento y Uso de Tecnologías de la Información y Comunicación en los Hogares, INE, 2017 Retrieved from http://www.ine.es , Accessed date: 15 January 2018. [56] Mediaclick, ¿Cuál es la edad de los usuarios de las redes sociales? Retrieved from https://www.mediaclick.es/blog/cual-es-la-edad-de-los-usuarios-de-las-redes-

[1] L.A.M. Mendes-Filho, F.B. Tan, “An overview on user-generated content and the empowerment of online” “Travellers”, FARN J. 7 (2) (2008) 17–30. [2] F. Muñoz, J. Hernández, J. Sánchez, Generalising user behaviour in online travel sites through the Travel 2.0 website acceptance model, Online Inf. Rev. 36 (6) (2012) 879–902, http://dx.doi.org/10.1108/14684521211287945. [3] Nielsen, The Digital Consumer, Retrieved from http://www.nielsen.com/content/ dam/corporate/us/en/reports-downloads/2014%20Reports/the-digital-consumerreport-feb-2014.pdf, (2014) , Accessed date: 11 November 2014. [4] E. Wright, N.M. Khanfar, C. Harrington, L.E. Kizer, The lasting effects of social media trends on advertising, J. Bus. Econ. Res. 14 (3) (2016) 75, http://dx.doi.org/ 10.19030/jber.v8i11.50. [5] S. Yu, L. Hudders, V. Cauberghe, Tracking the luxury consumer online: an experimental study on the effectiveness of site and search retargeting for luxury brands in China and the Netherlands: an abstract, in: P. Rossi (Ed.), Marketing at the Confluence between Entertainment and Analytics, Proceedings of the Academy of Marketing Science, Springer, Cham, 2017, p. 1541, , http://dx.doi.org/10.1007/ 978-3-319-47331-4_301. [6] M.D. Hernández, Y. Wang, H. Sheng, ... M. Minor, Escaping the corner of death? An eye-tracking study of reading direction influence on attention and memory, J. Consum. Mark. 34 (1) (2017) 1–10, http://dx.doi.org/10.1108/JCM-02-20161710. [7] J.L. Orquin, S.M. Loose, Attention and choice: a review on eye movements in decision making, Acta Psychol. 144 (1) (2013) 190–206, http://dx.doi.org/10.1016/j. actpsy.2013.06.003. [8] N. Abuín, La publicidad en periódicos electrónicos: Creación y evaluación de un modelo de eficacia, Universidad Complutense de Madrid, 2008, http://eprints.ucm. es/7958/ , Accessed date: 13 November 2014. [9] B. Chae, J. Hoegg, The future looks “right”: effects of the horizontal location of advertising images on product attitude, J. Consum. Res. 40 (2) (2013) 223–238, http://dx.doi.org/10.1086/669476. [10] R.V. Menon, V. Sigurdsson, N.M. Larsen, A. Fagerstrøm, G.R. Foxall, Consumer attention to price in social commerce: eye tracking patterns in retail clothing, J. Bus. Res. 69 (11) (2016) 5008–5013, http://dx.doi.org/10.1016/j.jbusres.2016.04. 072. [11] H.C.M. Van Trijp, Consumer understanding and nutritional communication: key issues in the context of the new EU legislation, Eur. J. Nutr. 48 (1) (2009) 41–48, http://dx.doi.org/10.1007/s00394-009-0075-1. [12] R. Varadarajan, S.Y. Manjit, Marketing strategy in an internet-enabled environment: a retrospective on the first ten years of JIM and a prospective on the next ten years, J. Interact. Mark. 23 (2009) 11–22, http://dx.doi.org/10.1016/j.intmar. 2008.10.002. [13] D. Hill, Tell me no lies: using science to connect with consumers, J. Interact. Mark. 17 (4) (2003) 61–72, http://dx.doi.org/10.1002/dir.10068. [14] X. Drèze, F.-X. Hussherr, Internet advertising: is anybody watching? J. Interact. Mark. 17 (4) (2003) 8–23, http://dx.doi.org/10.1002/dir.10063. [15] J. Vila, Y. Gomez, Extracting business information from graphs: an eye tracking experiment, J. Bus. Res. 69 (5) (2016) 1741–1746, http://dx.doi.org/10.1016/j. jbusres.2015.10.048. [16] K.C. Hamborg, M. Bruns, F. Ollermann, K. Kaspar, The effect of banner animation on fixation behavior and recall performance in search tasks, Comput. Hum. Behav. 28 (2) (2012) 576–582, http://dx.doi.org/10.1016/j.chb.2011.11.003. [17] V. Naidoo, L.D. Hollebeek, Higher education brand alliances: investigating consumers' dual-degree purchase intentions, J. Bus. Res. 69 (8) (2016) 3113–3121, http://dx.doi.org/10.1016/j.jbusres.2016.01.027. [18] P. Manchanda, J.P. Dubé, K.Y. Goh, P.K. Chintagunta, The effect of banner advertising on internet purchasing, J. Mark. Res. 43 (1) (2006) 98–108, http://dx.doi. org/10.1509/jmkr.43.1.98. [19] K. Li, G. Huang, G. Bente, The impacts of banner format and animation speed on banner effectiveness: evidence from eye movements, Comput. Hum. Behav. 54 (2016) 522–530, http://dx.doi.org/10.1016/j.chb.2015.08.056. [20] W.W. Moe, P.S. Fader, Dynamic conversion behavior at E-commerce sites, Manag. Sci. 50 (3) (2003) 326–335, http://dx.doi.org/10.1287/mnsc.1040.0153. [21] D.W. Baack, R.T. Wilson, B.D. Till, Creativity and memory effects: recall, recognition, and an exploration of nontraditional media, J. Advert. 37 (4) (2008) 85–94, http://dx.doi.org/10.2753/JOA0091-3367370407. [22] S. Putrevu, Consumer responses toward sexual and nonsexual appeals: the influence of involvement, need for cognition (NFC), and gender, J. Advert. 37 (2) (2008) 57–70, http://dx.doi.org/10.2753/JOA0091-3367370205. [23] H.S. Krishnan, D. Chakravarti, Memory measures for pretesting advertisements: an integrative conceptual framework and a diagnostic template, J. Consum. Psychol. 8 (1) (1999) 1–37, http://dx.doi.org/10.1207/s15327663jcp0801_01. [24] M. Fishbein, I. Ajzen, Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research Reading, Addison-Wesley, MA, 1975. [25] C. Koch, S. Ullman, Shifts in selective visual attention: towards the underlying neural circuitry, Hum. Neurobiol. 4 (4) (1987) 19–27, http://dx.doi.org/10.1007/ 978-94-009-3833-5_5. [26] Nielsen, J. (2007), “Banner blindness: old and new findings”, (Retrieved from), http://www.nngroup.com/articles/banner-blindness-old-and-new-findings/, (accessed November 11, 2015). [27] C. Lapa, Using Eye Tracking to Understand Banner Blindness and Improve Website Design, Doctoral Thesis Rochester Institute of Technology, 2007. [28] Y.C. Hsieh, K.H. Chen, How different information types affect viewer's attention on internet advertising, Comput. Hum. Behav. 27 (2) (2011) 935–945, http://dx.doi.

12



sociales/, (2017) , Accessed date: 15 January 2018. [57] International Telecommunication Union, ITU, Methodology for the subjective assessment of the quality of television pictures, ITU-R BT.500-11, 2002 Retrieved from http://www.itu.int/rec/R-REC-BT.500/en , Accessed date: 12 March 2015. [58] M.G. Hoy, G. Milne, Gender differences in privacy-related measures for young adult Facebook users, J. Interact. Advert. 10 (2) (2010) 28–45, http://dx.doi.org/10. 1080/15252019.2010.10722168. [59] X.Y. Leung, The Marketing Effectiveness of Hotel Facebook Pages: From Perspectives of Customers and Messages, University of Nevada, Las Vegas, 2012. [60] D. Buhalis, R. Law, Progress in information technology and tourism management: 20 years on and 10 years after the Internet – the state of eTourism research, Tour. Manag. 29 (4) (2008) 609–623, http://dx.doi.org/10.1016/j.tourman.2008.01.005. [61] F. Rejón, J. Sánchez, F. Muñoz, Improving the effectiveness of advertising in internet social networking, in: M.M. Cruz-Cunha, P. Gonçalves, N. Lopes, E.M. Miranda, G.D. Putnik (Eds.), Handbook of Research on Business Social Networking: Organizational, Managerial, and Technological Dimensions, IGIGlobal, Hershey (USA), 2012, pp. 756–785, , http://dx.doi.org/10.1016/10.4018/ 978-1-61350-168-9.ch040. [62] W.G. Zikmund, Fundamentos de investigación de mercados, Thomson, Madrid, 2003. [63] K. Holmqvist, M. Nyström, R. Andersson, R. Dewhurst, H. Jarodzka, J. van de Weijer, Eye Tracking: A Comprehensive Guide to Methods and Measures, Oxford University Press, Oxford, 2011. [64] J. Simola, M. Kivikangas, J. Kuisma, C.M. Krause, Attention and memory for newspaper advertisements: effects of ad–editorial congruency and location, Appl. Cogn. Psychol. 27 (4) (2013) 429–442, http://dx.doi.org/10.1002/acp.2918. [65] S. Djamasbi, M. Siegel, T. Tullis, Generation Y, web design, and eye tracking, Int. J.

[66]

[67]

[68]

[69]

[70]

[71]

[72]

13

Hum. Comput. Stud. 68 (5) (2010) 307–323, http://dx.doi.org/10.1016/j.ijhcs. 2009.12.006. S. Djamasbi, M. Siegel, T. Tullis, Visual Hierarchy and Viewing Behavior: An Eye Tracking Study, in: J.A. Jacko (Ed.), Human-Computer Interaction: Design and Development Approaches, Springer, Berlin Heidelberg, 2011, pp. 331–340, , http:// dx.doi.org/10.1007/978-3-642-21602-2_36. M. Pagendarm, H. Schaumburg, Why are users banner-blind? The impact of navigation style on the perception of web banners, J. Digit. Inf. 2 (2001) 1 Retrieved from http://journals.tdl.org/jodi/article/view/36/38 , Accessed date: 10 November 2015. L. Ha, K. McCann, An integrated model of advertising clutter in offline and online media, Int. J. Advert. 27 (4) (2008) 569–592, http://dx.doi.org/10.2501/ S0265048708080153. G. Brajnik, S. Grabrieli, A review of online advertising effects on the user experience, Int. J. Hum.-Comput. Interact. 26 (10) (2010) 971–997, http://dx.doi.org/10. 1080/10447318.2010.502100. X. Xu, N. Scott, J. Gao, Cultural influences on viewing tourism advertising: an eyetracking study comparing Chinese and Australian tourists, J. Tour. Serv. 8 (14) (2017) 30–46. J. Simola, J. Kuisma, A. Öörni, L. Uusitalo, J. Hyönä, The impact of salient advertisements on reading and attention on web pages, J. Exp. Psychol. Appl. 17 (2) (2011) 174–190, http://dx.doi.org/10.1037/a0024042. M. Resnick, A. William, The impact of advertising location and user task on the emergence of banner ad blindness: an eye-tracking study, Int. J. Hum.-Comput. Interact. 30 (3) (2014) 206–219, http://dx.doi.org/10.1080/10447318.2013. 847762.