Online behavioral targeting combined with
native advertising, a blessing in disguise?
Peter Stevens (5908353)
Thesis
Graduate School of Communication
Persuasive Communication
Abstract
This study aimed to investigate the effect online behavioural targeting of online native video advertisements would have on cognitive-, affective- and behavioural- brand effects. The mediating effect of persuasion knowledge and perceived relevance (of the ad) were also tested. No significant results were found for targeting on brand awareness. It has been found that targeting has negative effects on brand attitude but this does not affect purchase intention. Online behavioural targeted advertisements were unexpectedly not perceived as more
relevant. Persuasion knowledge mediates the effect of targeting on brand effects in a positive way because of perceived appropriate use of tactics.
Introduction
With technological advancements and the development of the use of the internet and online space new ways of online advertising arose the past few years. Online Behavioral Advertising (OBA) is one of the most commonly used online advertising tools. OBA is the adjusting of advertisements to previous online surfing behaviour (Smit, Noort & Voorveld, 2013). In order to gather information about the previous online surfing behaviour small text files are installed on the users computer (cookies).
There is debate about the use of cookies. Advertisers claim it improves the relevance of advertisements (perceived relevance) and thus the user experience of browsing. However politicians struggle with privacy concerns (Bennet, 2011). There are ambivalent research outcomes about the effectiveness of OBA and questions about whether the privacy concerns are proportionate to the gains of OBA (Tene & Polenetski, 2012). How effective is OBA in comparison to non-OBA? Does the consumer perceive the ad as more relevant when targeted in this way? Is the possible gained effectiveness proportionate to the acquiring of personal surfing behavioral information? This research will give insight in the effectiveness of OBA and the role of perceived relevance.
There is another online advertising technique that recently is vastly gaining popularity:
Native advertising. The proliferation of online advertising volume, the ineffectiveness of
online banner advertisements and the shift from offline to online media consumption have proven to be serious threats to advertising reach and effectiveness (Schuman et al., 2014; Goldfarb & Tucker, 2011; Reijmersdal, Neijens & Smit, 2005; Ferraro & Avery, 2000). While most websites are free and depend on their advertising revenue for survival (Anderson, 2009), native advertising has become increasingly popular: Advertising spending on native advertising has grown from approximately $4 billion in 2013 to projected values of $8 billion in 2015 (Rosin, 2015).
Native advertising is any type of branded content that is presented online. Branded content is the intentional incorporation of a brand into editorial content (Kahhr, 1998; Van Reijmersdal et al., 2007; Cannes Lions, 2014). The problem of advertising that is incorporated into editorial content is the blurred line between editorial and sponsored media content. This could lead to misleading media consumers, eroding trust in media and restraining journalists to critically write about sponsors. How well do people understand that the source of branded content is a company or brand?
Persuasion Knowledge (PK) is the knowledge people have about persuasion attempts and is activated when people process a message and label it as a persuasion attempt. When PK is activated messages are often evaluated more critically. The goal of the sender of the message is after all to persuade, not to inform. Does native advertising circumvent the
activation of PK? And if so, does native advertising lead to a better evaluation not only of the advertisement but more important, of the brand advertised?
It is to be expected that in the near future targeted advertising in the form of OBA and branded content will be combined. Therefore the research question go’s:
RQ: What is the effect of targeting online native video advertisements on brand effects and how does persuasion knowledge and perceived relevance relate to this?
This article will shed light on the effectiveness of targeting online behavioral native video advertising on cognitive- affective and behavioral-brand effects. Also the mediating effects of perceived relevance and persuasion knowledge will be taken into account. This research will give power to the debate about the use of OBA and native advertising and aid communication professionals in choosing their strategy. Because of the novelty of OBA and native
advertising there are few research articles, therefore this article will be a valuable addition for the scientific literature and will offer new insights for future research.
Theory Native Advertising
First let’s define what native advertising exactly is. Native advertising presents advertisements that mimic the editorial content, both in format and content, that is published on the same platform. Native advertising comprises of a variety of advertising formats as long as the advertisement is created in a way that mirrors the appearance of the non-commercial content published on the same website (Wojdinsky, 2016; Couldry & Turrow, 2014). Examples are video (series or one-off), series of articles, hyperlinks and social media posts.
There are several reasons why native advertisements are currently a popular
advertising strategy. The first reason is because editorial content receives more attention than advertisements and it’s content is better remembered (Cameron, 1994; Cameron & Curtin, 1995). Native advertisements are so similar to editorial content that it diminishes the
competition between the content a consumer is seeking for and the advertising that provides for the production costs of this content (Wojdinsky, 2016). Because people tend to search for editorial content their attention towards editorial content is higher than towards
advertisements. This process is known as “intentional exposure” (Lord & Putrevu, 1993). Advertisements that mimic the format of editorial content are also known to be more appreciated than the same content presented as an advertisement (Van Reijmersdal, et al., 2005; Cameron & Curtin, 1995; Cameron, 1994). This appreciation is explained through the source credibility theory (Petty & Cacioppo, 1981) in the following paragraph.
When people process advertisements they are precautious, knowing advertisers aim to persuade. Editorial content on the other hand is considered to be more objective. The result is
a more sceptical attitude towards advertisements as opposed to editorial content (Van Reijmersdal et al., 2005; Schwarz et al., 1986; Salmon et al.1985). Another result of this precaution for advertising is that people take less effort processing the information of advertising, (Cameron & Curtin, 1995; Cameron, 1994) . Thus because of the credibility of the source, content that is perceived as editorial is more appreciated and accepted than advertisements (Van Reijmersdal et al., 2005).
However native advertising is also attracting critics, who view native advertising as an advertising technique designed to deceive consumers by masking source attribution
(Campbell & Marks, 2015; Wasserman, 2013). Edward Wasserman, dean of the UC-Berkeley Graduate School of Journalism, even states: “Native advertising,’ in short, is all about
deception. You, as the reader, are encouraged to perceive the messages as something other than what they are. And even if, at some level, you understand they weren't put together by the magazine's staff, you're still expected to see them as partaking of the magazine's
trustworthiness, and as deserving something of the same regard.”(Wasserman, 2013, p.1).
The growth and critique has aroused the interest of the Federal Trade Commission (FTC) which is the regulator in conduct of trading and commerce in the USA. The FTC has responded by revising online disclosure guidelines and conducting a workshop convened with practitioners, researchers and policy experts to discuss the possible deceptive practices related to native advertising (FTC, 2013).
Online behavioural advertising
The internet is dominated by free websites that depend on advertising revenues to survive (Schuman et al., 2014; Anderson, 2009). Through targeted advertising, these websites can increase their advertising relevance and effectiveness and thus, their revenues (Schuman et al., 2014; Iyer, Soberman & Villas-Boas, 2005). Targeted online advertising refers to any
form of online advertising that is based on information the advertiser has about the receiver of the advertisement such as demographics, current or past browsing behavior, information from preference surveys and geographic information. Online behavioral advertising (OBA) is the
adjusting of advertisements to previous online surfing behaviour (Smit et al., 2013). To enable this form of targeted advertising usually ‘cookies’ need to be installed. These are small text files (max 4kb) that collect data to either support the functionality of the website (first party, session or functional cookies) or to collect profile information for targeted advertising, so called third party cookies (Smit et. al., 2013).
People are skeptical of OBA, more than two thirds of US citizens rejects online targeted advertising of any form (Turow et. al., 2010). There are ambivalent research outcomes about the effectiveness of OBA and it is questioned if the gains of OBA are proportionate to collecting personal information in the form of browsing history (Tene & Polenetski, 2012). There has been research done on the awareness of OBA and how people cope with OBA (Smit, et al., 2013). Also the attitudes about OBA and how to frame messages that justify OBA use has been researched (Schumann, 2014). In most scientific research concerning OBA it is brought forward that people tend to have negative attitudes towards OBA (Goldfarb & Tucker, 2011; Smit et al., 2013; Schumann et al., 2014). But little is known about the effects OBA has on the brands depicted in the ad in combination with native
advertising. That is why the dependent variables comprise of cognitive, affective and behavioral brand effects.
Brand awareness
Rossiter and Percy (1987) describe brand awareness as being essential for the communications process to occur as it precedes all other steps in the process. Without brand awareness occurring, no other communication effects can occur. For a consumer to buy a brand they must first be made aware of it. Brand attitude cannot be formed, and intention to
buy cannot occur unless brand awareness has occurred (Rossiter & Percy, 1987; Rossiter et al., 1992). Rossiter and Percy (1987) claim that there are two types of brand awareness: brand recognition and brand recall, and which of these occurs depends on the specific situation. Brand awareness does not have to require recall of the brand name. It is possible for instance that a brand is identified by its location (”the supermarket down the street”), packaging or shape. Furthermore, Rossiter and Percy (1987) state that brand recall is not imperative for purchase; mere recognition of the brand, can be sufficient for purchase. Lynch and Srull (1982) defined these different choice situations as stimulus-based (where all the relevant brand information is physically present), memory-based (where all relevant information is recalled from memory) and mixed-choice (where some of the information is physically present, and some must be recalled from memory). Recognition appears in stimulus-based situations, and recall occurs in memory-based situations. Because online behavioural targeting is based on previous surfing behaviour it is expected that this will lead to higher levels of brand recognition (stimulus based situation) and brand recall (recalled from memory) than not targeted advertising, even when native. That’s why the first hypotheses is:
H1a,b: Respondents in the targeted condition score higher on a) brand recall and b) brand recognition.
Figure 1. H1a,b schematically
+
Figure 1. H1 presented schematically, with online behavioural targeting as independent variable and brand recall, brand recognition as dependent variables. Brand attitude Targeting Versus Non-targeting Brand recal Brand recognition
While most research on OBA as well as native advertising is about how people evaluate the advertisement (Goldfarb & Tucker, 2011; Smit, Noort & Voorveld, 2013;
Schumann, Wangenheim & Groene, 2014) or how people cope with these advertising
strategies (Smit, Noort & Voorveld, 2013). No research has been conducted on brand effects in relation to OBA and native advertising combined. This while brand effects like brand attitude and purchase intention are important driver for marketers and strategists (Ahn & Bailensen, 2011).
But what is an attitude? Fishbein (1963, 1967; Fishbein & Ajzen, 1975) gives a clear theoretical explanation of the causal basis of attitudes. According to Fishbein and Ajzen (1975, p. 222), "A person's attitude is a function of his salient beliefs at a given point in time." Beliefs are the subjective associations between any two different concepts. Salient beliefs are those activated from memory and "considered" by the person in a given situation (Fishbein & Ajzen, 1975). A brand attitude in this context is the function of a person’s salient beliefs about a certain brand at a given point in time (Olson & Mitchell, 2000). When asked, people tend to have negative attitudes towards online behavioral targeted advertising (Turow et. al., 2010). It is proven that attitudes towards an advertisment affects attitude towards the brand in the advertisment (Gardner, 1985). From this the next hypotheses follows:
H1c: Online behavioural targeting of native video advertisements has a negative effect on brand attitude.
Purchase intention
Purchase intention as defined by Ahn and Bailensen (2011) is: “How likely one is to purchase the product in the advertisement.” Purchase intention is often the most important driver for an advertisement, as the purpose of advertising is to increase sales. (Ahn &
Bailensen, 2011). Research findings suggest a strong association between brand attitude and purchase intention (Macay, Ewing & Newton, 2009; Spears & Singh 2004; Le Clerc & Little 1997; Anand & Sternthal 1990). Because of this association with brand attitude it is expected that online behavioral targeting will also have a negative effect on purchase intention. The next hypotheses is:
H1d: Online behavioural targeting of native video advertisements has a negative effect on purchase intention.
Persuasion knowledge
The Persuasion Knowledge Model (Friestad & Wright, 1994) describes how people develop knowledge about persuasion and how people use this knowledge to interpret, evaluate, and respond to persuasion attempts. Friestad and Wright (1994) state that in order for persuasion knowledge to be activated, people need to be aware of a persuasion attempt. If a message is interpreted as a persuasion attempt people use their persuasion knowledge to cope with the attempt (Boerman, Reijmersdal & Neijens, 2012). In any given persuasion context, various categories of knowledge form defences against persuasion. This knowledge includes topic knowledge, agent knowledge (knowledge about the sender), and persuasion knowledge itself. Persuasion knowledge is considered to comprise of theories about mediators of persuasion, beliefs about marketing tactics and goals of marketing tactics.
In the case of masked advertisements the activation of persuasion knowledge leads to negative evaluations of these advertisments (Brusse, Fransse & Smit, 2015; Tutaj & Van Reijmersdal, 2012; Van Reijmersdal, Neijens & Smit, 2010). A way to measure persuasion knowledge is through the persuasive intent scale of Rozendaal, Buijzen and Valkenburg (2010). This scale measures if people perceive a message as to be persuasive. Because native
advertising is supposed to circumvent the recognition of the content being an advertisment (Campbell & Marks, 2015; Wasserman, 2013), it can therefore circumvent the activation of persuasion knowledge. It is not clear wheather or not the recipient of the native video advertisement understands that it is a persuasion attempt. It is also not clear if online behavioural targeting activates persuasion knowledge. The second hypotheses is:
H2a: Online behavioural targeting of native video advertisements will have a positive effect on persuasive intent.
Understanding tactics is when a person recognizes the persuasive tactics being employed in a message and can trigger persuasion knowledge (Harden, Bearden & Carlson, 2007). Native advertising is not always recognised as being an advertisement (Wasserman, 2013). To investigate whether or not people recognize online behavioural targeting in the case of a native video advertisement the following hypotheses is formed:
H2b: Online behavioural targeting of native video advertisements will have a positive effect on understanding tactics.
There are also cases were it was found that triggered negative evaluations can be moderated by perceived appropriateness of marketing tactics used (Wei, Fischer & Main, 2008).
Therefore it would be interesting to see if online behavioural targeting can be perceived as an appropriate marketing tactic in the case of a native video advertisement. Hypotheses 3a,b is:
H3a,b: Understanding tactics will positively mediate the effect of online behavioural targeting of native video advertising on a) brand attitude and b) purchase intention.
Perceived relevance +
-
+ Perceived relevance
The justification for the use of OBA is that these advertisements would be more relevant to the recipient (Schuman et al., 2014). Relevant advertising is defined as that which is interesting, relevant and useful to receivers in a way that they consider it worthy of their attention (Laczniak & Muehling, 1993). That is why it is expected that people would find online behavioural targeted native video advertising more relevant then when not targeted. The fourth hypotheses is:
H4: Online behavioural targeting of native video advertisements will increase perceived relevance of the online native video advertisement.
Will this perceived relevance also affect people’s attitudes and behaviour?
H5a,b: Perceived relevance will mediate the effect of online behavioural targeting of native video advertising on a) brand attitude and b) purchase intention.
Figure 2. Independent and dependent variables presented with moderators.
Figure 2. Schematical model of the variables and moderators including expected direction of effects. Brand attitude Purchase intentions Persuasion knowledge: - Understanding tactics - Persuasive intent Targeting Versus Non-targeting + _-
Method Design
This research is set up as a quantitative research to provide empirical support for the relation between the independent variables targeted native online video advertisement versus non-targeted native online video advertisement on the dependant variables brand awareness, brand attitude and purchase intention. Persuasion knowledge and perceived relevance were taken into account as mediators. To measure the dependant variables a survey was produced and an experiment was conducted. Surveys are an excellent method to measure what people think, feel or belief about something (Aronson, Wilson & Akert, 2005). A one factor
between-subjects design was used (Targeted native online video versus a non-targeted native online video).
Sample
Through Facebook a convenience sample of mostly Dutch students were recruited for the sample. A total of 77 participants of 19 to 35 years old (M = 25.68, SD = 3.44)
participated in the research of which 64.91% was male. This age group was chosen because they are in the age group between 18-35, the group that watches online video’s the most (Newcom, 2014). Respondents that didn’t finish the survey and/or were not in the age group were excluded from the sample. The dropout rate was 33.04%.
Procedure
After a short introduction the participants were randomly assigned to read one of two scenario’s, making it a targeted or non-targeted condition, after which both groups were exposed to the same online native video advertisement. Then they answered questions about
demographics, attitudes about the video, the brand in the video and were thanked for their participation.
Stimulus materials
A video of famous chef Marco Pierre White preparing a steak on the branded youtube.com video channel of Knorr was used. This clip was chosen because it is a native online video advertisement that wasn’t too long (to prevent a high dropout rate) with a familiar brand in it that wasn’t targeted sex specifically.
The clip was 1.54 minutes long and was called “The steak challenge, by Marco Pierre White”. In the video this famous chef prepares one steak seasoned with salt and pepper and one steak seasoned with a stock cube (of the brand Knorr).
All respondents were shown this video. In both conditions they were presented with a screen shot taken of the homepage of youtube with the stimulus material seamlessly edited in. They were shown this screenshot a second time with a red arrow pointing out the clip and suggesting to click on it. Before the video was shown the respondents were asked to envision one of two following scenarios. Participants in the non-targeted scenario where asked to envision that they would go online to look up the weather before going to youtube to listen to some music. After this the video followed.
In the targeted scenario they were asked to envision searching online how to prepare a perfect steak before going to youtube to listen to some music.
Measures Brand responses
Brand recall, to measure whether or not people had seen any brands in the native online video advertisement, respondents were asked: ‘Did you recognize any brands in the
video? If ‘yes’, which one(s)?’ With an open answering category for the people that answered ‘yes’ (81.80%) after which a possibility was given to fill in which brand, 79.22% showed a correct recall.
To measure brand recognition respondents were asked: ‘Did you see any of the following brands in the video clip you've just seen?’ after which they could choose out of a list of seven brands including Knorr and a ‘No or none of the above’ answering possibility. 84,42% of the respondents recognized the brand.
Attitude towards the brand was measured as ‘overall feeling about the brand’ on three seven-point semantic differentials: bad/good, unpleasant/pleasant and unfavourable/
favourable (MacKenzie, Lutz & Belch 1986). Scores were averaged to create one measure of brand attitude (Cronbach’s alpha = .94, M = 4.25, SD = 1.32).
With four items on a seven-point Likert scale ranging from never to definitely, with questions like: ‘Would you buy stock (bouillon) of Knorr?’ (Spears, Surrendra & Singh,
2004) purchase intention was measured. Scores were averaged to create one measure for
purchase intention (Cronbach’s alpha = .96, M = 3.10, SD = 1.55).
Relevance anticipation.
Relevance anticipation was measured with four items based on the relevance
anticipation scale of Laczniak and Muehling (1993). This scale ranged on a seven point Likert scale from ‘strongly agree’ to ‘strongly disagree’, the questions were: ‘In this situation…The online video content was relevant to me/I received useful information through this video/the video was interesting to me/this online video was worth paying attention to.’ Scores were
averaged to create one measure of relevance anticipation (Cronbach’s alpha = .91, M = 3.25,
Persuasion knowledge.
Two dimensions of persuasion knowledge were measured: persuasive intent and understanding tactics. For the measurement of persuasive intent respondents were asked to what extent they agreed with three statements: ‘This video is created to persuade/increase sales’/make me like the product’ on a seven-point Likert scale from strongly agree to strongly disagree (Van Noort, Antheunis & Van Reijmersdal, 2012; Rozendaal, Buijzen & Valkenburg 2010; Van Reijmersdal et al. 2010). Scores were averaged to create one measure for
persuasive intent (Cronbach’s alpha = .86, M = 3.25, SD = 1.16).
Understanding tactics was measured on a three item seven-point Likert scale ranging from strongly disagree to strongly agree. An example of a question was: ‘This video was adjusted to my preferences.’ Scores were averaged to create one measure for understanding
tactics (Cronbach’s alpha = .75, M = 3.87, SD = 1.48).
Control variables
Participants were asked if they ever heard of Knorr (97.4% said yes) and if they ever had used Knorr (85.7% said yes). The attitude towards the online native video was also
measured as ‘overall feeling about the online video’ on three seven-point semantic
differentials: bad/good, unpleasant/pleasant and unfavourable/ favourable (MacKenzie, Lutz & Belch 1986). Scores were averaged to create one measure of online video attitude
(Cronbach’s alpha = .88, M = 3.85, SD = 1.24).
How often participants watched Youtube video’s on a seven point Likert scale from ‘never’ to ‘everyday’ was measured, the modus was ‘several times a week’ with 25.9%. As to demographics participants were asked about their age, sex and highest completed education level varying from ‘none’ to ‘university’. University was the most common score with 55.8%.
All items on all semantic differential or Likert scales where tested for reliability and scores were averaged to create a single measure of the construct they represented.
Results Randomization
To verify whether the participants in the experimental groups were similar on a number of control variables, a test of equivalence was conducted on age, gender, prior brand usage and online video usage. The targeted and the non-targeted group did not differ on gender, Chi2(1) = ,09, p = .77 or prior brand usage, Chi2(1) = 1,47, p = .23. There was no significant difference found between groups for prior brand usage (F(1,75) = .71, p = .40) or online Youtube video usage F(1,75) = .54, p = .47. Because there were no significant
differences found between the two groups on any of the control variables, no covariates were included in the analyses.
To test hypotheses H1a,b: Respondents in the targeted condition score higher on a) brand recall and b) brand recognition a Chi2 was employed because brand recall and brand recognition where dichotomous variables. No significant effects where found for brand recognition, Chi2(1) = .77, p = .38 or recall Chi2(1) = .07, p = .79, thus H1a and H1b were not supported, targeting does not affect brand recall or brand recognition.
dition Dependent Variables Targeted condition
Not recalled/ recognized
Recalled/ recognized
Not recalled/ Recalled/ recognized recognized Brand recognition 12.2% n.rn. 87.8% 19.4% n.rn. 80.6% Brand recall 19.4% n.rl. 80.6% 17.1% n.rl. 82.9%
Table 1.
H1c,d: Online behavioural targeting of online native video advertisements has a negative effect on c) brand attitude and d) purchase intention.
A MANOVA was conducted with targeted versus non-targeted condition as the independent variable and brand attitude and purchase intention as dependent variables. A just non-significant multi-variate effect was found, (Wilk’s Lambda = .93, F(2,74) = .93, p = .06, eta2 = .52), excluding a multi-variate effect. An ANOVA was conducted for each dependent variable. There was a significant difference between the targeted and the non-targeted condition on brand attitude (F(1,75) = 7.37, p = .04). Participants in the targeted condition held less positive attitudes toward the brand in the online native video advertisement, than participants in the non-targeted condition, see table 2 for means. Thus H1c was supported. There was no significant effect found for targeting on purchase intention (F(1,75) = .16, p = .69), thus H1d was not supported. Targeting does not affect purchase intention, see table 2 for means.
dition
Table 2.
H2a,b: Online behavioural targeting of native video advertisements will have a positive effect on a) persuasive intent and b) understanding tactics.
A MANOVA showed a significant multi variable effect of targeting on persuasion knowledge and perceived relevance, Wilk’s Lambda = .89, F(3,73) = 3.39, p = .02, eta2 = .12. As well as
Dependent Variables Targeted condition
Brand attitude M = 3.92a SD = 1.56 M = 4.54b SD = .99
a significant univariate effect of targeting on understanding tactics F(1,75) = 9.36, p = .003. Participants in the targeted group were more aware of the persuasion tactics being employed, in other words: H2b is supported, targeting triggers understanding tactics.
There was no significant direct effect of targeting on persuasive intent, F(1,75) = 4.25, p = .52. H2a was rejected, targeting does not trigger persuasive intent in the case of an online native advertisement.
Because of the type of analysis H4 will also be treated here. H4: Online behavioural targeting of native video advertisements will increase perceived relevance of the online native video advertisement.
There was just not a significant effect for targeting on perceived relevance (F(1,75) = 3.95, p = .05). H4 is rejected, targeting does not improve perceived relevance.
Table 3. The effect of targeting on the mediators
Targeted condition Non-targeted
condition
Mediators
- Persuasive intent M = 5.69a (SD =1.13) M = 5.51a (SD = 1.19)
- Understanding tactics M = 4.39a (SD =1.54) M = 3.41b (SD = 1.28)
Perceived relevance M = 3.34a (SD=1.69) M = 2.93a (SD = 1.39)
Table 3. Note: the subscript letters indicate significance if a-b.
H3a,b: Understanding tactics will positively mediate the effect of online behavioural targeting of native video advertising on a) brand attitude and b) purchase intention. The PROCES macro developed by Hayes (2013, model 4) was used to test the indirect effect of understanding tactics on brand attitude and purchase intention. Persuasive intent was also taken into account as a part of PK.
Targeting has a significant direct effect on understanding tactics as seen in H2 (F(1,75) = 9.36, p = .003), but no significant direct effect on persuasive intent (F(1,75) = 4.25, p = .52). Both mediators understanding tactics as persuasive intent had a direct effect on brand attitude as seen in table 4.
Table 4. Mediator outcomes on brand attitude.
b SE t p CI CI
Mediators on brand attitude
- Understanding tactics .24 .11 2.20 .03 .02 .46
- Persuasive intent -.31 .11 -2.91 .00 -.52 -.10
Table 4.
There was a significant mediating effect of targeting on brand attitude via
understanding tactics (b = -.24, SE = .11, BCa 95% CI [-.54, -.06]). H3a can be accepted, understanding tactics mediates the negative effect of targeting on brand attitude.
There was no significant mediation of persuasive intent because the CI contained zero (b = .05, SE = .09, BCa 95% CI [-.09,-.25]).
H3b) There was a significant mediating effect of targeting on purchase intention via understanding tactics (b = -.26, SE = .14, BCa 95% CI [-.59, -.05]). Understanding tactics mediates the negative effect of targeting on purchase intention thus approving H3b. Persuasive intent did not yield significant results as a mediator of targeting on purchase intention (b = .07, SE = .12, BCa 95% CI [-.10, .39]).
Targeting has a significant direct effect on understanding tactics as seen in H2 (F(1,75) = 9.36, p = .003), but no significant direct effect on persuasive intent( F(1,75) = 4.25, p = .52). Both mediators understanding tactics as persuasive intent had a direct effect on purchase intention as seen in table 5.
Table 5. Mediator outcomes on purchase intention.
b SE t p CI CI
Mediators on purchase intention
- Understanding tactics .26 .13 2.14 .04 .02 .50
- Persuasive intent -.44 .14 -3.10 .00 -.73 -.16
Table 5.
For the final hypothesis H5a,b Perceived relevance will mediate the effect of online behavioural targeting of native video advertising on a) brand attitude and b) purchase intention.
With respect to H5a the analyses showed no sign indirect effect of targeting through perceived relevance on brand attitude (b =- .27, SE = .16, BCa 95% CI [-.60, .01]) because the CI contained zero thus H5a was rejected. Perceived relevance doesn’t mediate the effect of targeting on brand attitude. There was no direct effect for targeting on perceived relevance (b = -.69, SE = .35, p=.05), but there was a significant direct effect for perceived relevance on the outcome variable brand attitude (b = .38, SE = .09, p >.000).
With respect to H5b the analyses showed an indirect effect of targeting through perceived relevance on purchase intention (b = -.26, SE = .14, BCa 95% CI [-.59, -.05]), approving hypothesis H5b. If the online native video content is perceived as more relevant it will mediate (decrease) the negative effect of targeting on purchase intention.
There was no significant direct effect of targeting on perceived relevance (b = -.69, SE = .35, p=.05), however a significant direct effect of perceived relevance on purchase intention was found, b = 4.32, SE = .57, p > .000.
Conclusion & Discussion
This study aimed to investigate the effect online behavioural targeting of online native video advertisements would have on cognitive-, affective- and behavioural- brand effects. The mediating effect of persuasion knowledge and perceived relevance (of the ad) were also tested. Based on the results several conclusions can be drawn.
First of all no effect was found for the targeting on brand awareness meaning that online behavioural targeting does not affect the brand awareness of a native advertisement. This research has found a negative effect on brand attitude but not on purchase intention. This is quite remarkable because there have been research findings that suggest a strong
association between attitude towards a brand and purchase intentions (Macay, Ewing & Newton, 2009; Spears & Singh 2004; Le Clerc & Little 1997; Anand & Sternthal 1990). Though there was controlled for brand attitude in the experiment it could be that it doesn’t affect purchase intention because of the type of product it is: it is a practical household item which makes cooking easier.
Brand awareness showed no significant results. This could lie in the fact that Knorr is such a well-known brand (97.4% of the participants had heard of Knorr). But still a result was expected for brand recognition as opposed to brand recall, this was also not the case. Though it was a online native video ad which mimicked the format of other online cooking programs it was fairly obvious it was sponsored by Knorr. There are native advertisements that are more blended in with editorial content, this could be the cause for different brand awareness effects.
Although understanding tactics was originally chosen as a way to measure PK with a side note that it could mediate the effect of targeting through appropriateness of marketing tactics. It turned out that in the case of online native video advertising targeting is considered
to be thought of as appropriate and mediates for the negative effect of targeting in accordance with the research of Wei, Fischer and Main (2008).
Understanding tactics mediated the effect of targeting on both brand attitude and purchase intention. This is the first research that shows this result. It should be taken into account by marketeers when designing campaigns and regulators when forming new legislation.
The most important argument for using OBA is to increase relevance for consumers. The results of this research shows that online behavioural targeting does not increase
relevance of native video advertisements for people between 18 and 35 with an above average level of education. Which brings us to the limitations.
Limitations
The education level was above average with 55% of the respondents having finished with a university degree, though education was controlled for and showed now significant differences.
The respondents were asked to place themselves into a certain scenario, while this was perceived to be realistic to most, it is still different from actually being targeted on you
surfing behaviour and actually looking up a recipe on google. These experiment effects where minimalised but always influence the results.
References
Brusse, E. D. A., Fransen, M. L., & Smit, E. G. (2015). The More You Say, the Less They Hear. Journal of Media Psychology.
Cameron, G. T. (1994). Does publicity outperform advertising? An experimental test of the third-party endorsement. Journal of Public Relations Research, 6(3), 185-207.
Cameron, G. T., & Curtin, P. A. (1995). Tracing sources of information pollution: A survey and experimental test of print media's labeling policy for feature advertising. Journalism & Mass Communication Quarterly, 72(1), 178-189.
Campbell, C., & Marks, L. J. (2015). Good native advertising isn’ta secret. Business Horizons.
Couldry, N., & Turow, J. (2014). Advertising, big data and the clearance of the public realm: marketers' new approaches to the content subsidy. International Journal of Communication, 8, 1710-1726.
Federal Trade Commission, “Blurred Lines: Advertising or Content? An FTC Workshop on Native Advertising,” December 4, 2013.
http://www.ftc.gov/system/files/documents/public_events/171321/final_transcript.pdf
Ferraro, R., & Avery, R. J. (2000). Brand appearances on prime-time television. Journal of Current Issues & Research in Advertising, 22(2), 1-15.
persuasion attempts. Journal of Consumer Research, 21(1), 1-31.
Friestad and Wright, “The Persuasion Knowledge Model”; Friestad, Marian, and Peter Wright. “Everyday Persuasion Knowledge,” Psychology & Marketing, 16 no. 2 (1999): 185- 194.
Gardner, M. P. (1985). Does attitude toward the ad affect brand attitude under a brand evaluation set?. Journal of Marketing Research, 192-198.
Goldfarb, A., & Tucker, C. E. (2011). Online advertising, behavioral targeting, and privacy. Communications of the ACM, 54(5), 25-27.
Hardesty, D. M., Bearden, W. O., & Carlson, J. P. (2007). Persuasion knowledge and consumer reactions to pricing tactics. Journal of Retailing,83(2), 199-210.
Joel, M., (2013). We need a better definition of “native advertising” HBR Blog Network (2013, February 13) Retrieved Oktober 24, 2015, from http://blogs.hbr.org/2013/02/we-need- a-better-definition-of/
Lord, K. R., & Putrevu, S. (1993). Advertising and publicity: An information processing perspective. Journal of Economic Psychology, 14(1), 57-84.
Macdonald, E., & Sharp, B. (2003). Management perceptions of the importance of brand awareness as an indication of advertising effectiveness. Marketing bulletin, 14(2), 1-15.
Mackay, T., Ewing, M., Newton, F., & Windisch, L. (2009). The effect of product placement in computer games on brand attitude and recall. International Journal of Advertising, 28(3), 423-438.
Muehling, D. D., Laczniak, R. N., & Andrews, J. C. (1993). Defining, operationalizing, and using involvement in advertising research: A review. Journal of Current Issues & Research in Advertising, 15(1), 21-57.
Rosin, B., (2015). Why native advertising is a no-brainer for publishers and marketers. Online Publishing Insider, March 5, 2015.
http://www.mediapost.com/publications/article/245070/why- native-advertising-is-a-no- brainer-for-publish.html
Salmon, C. T., Reid, L. N., Pokrywczynski, J., & Willett, R. W. (1985). The effectiveness of advocacy advertising relative to news coverage. Communication Research, 12(4), 546-567.
Schumann, J. H., von Wangenheim, F., & Groene, N. (2014). Targeted online advertising: Using reciprocity appeals to increase acceptance among users of free web services. Journal of Marketing, 78(1), 59-75.
Schwarz, N. (1986). Resistance to Persuasion as a Consequence of Influence Attempts in Advertising and Non-Advertising Communications. Psychology: A Quarterly Journal of Human Behavior, 23, 72-76.
Tutaj, K., & van Reijmersdal, E. A. (2012). Effects of online advertising format and
persuasion knowledge on audience reactions. Journal of Marketing Communications, 18(1), 5-18.
Van Reijmersdal, E., Neijens, P., & Smit, E. (2005). Readers' reactions to mixtures of advertising and editorial content in magazines. Journal of Current Issues & Research in Advertising, 27(2), 39-53.
Van Reijmersdal, E. A., Neijens, P. C., & Smit, E. G. (2010). Customer magazines: Effects of commerciality on readers' reactions. Journal of Current Issues & Research in Advertising, 32(1), 59-67.
Wasserman, “Advertising Goes Native, and Deception Runs Free.” Huffington Post, April 1, 2013. http://www.huffingtonpost.com/edward-wasserman/native-advertising-atlantic-
scientology_b_2575945.html
Wei, M. L., Fischer, E., & Main, K. J. (2008). An examination of the effects of activating persuasion knowledge on consumer response to brands engaging in covert marketing. Journal of Public Policy & Marketing, 27(1), 34-44.
Wojdynski, B.W. (2016). Native advertising: Engagement, deception, and implications for theory. In R. Brown, V. K. Jones, and B. M. Wang (Eds.), The New Advertising: Branding, Content and Consumer Relationships in a Data-Driven, Social Media Era. Santa Barbara, CA: Praeger/ABC Clio.
