Available

Available online
online at
at www.sciencedirect.com
www.sciencedirect.com

ScienceDirect
Available online at www.sciencedirect.com
Procedia
Procedia Computer
Computer Science
Science 00
00 (2015)
(2015) 000–000
000–000
www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
ScienceDirect
Procedia Computer Science 111 (2017) 105–112

8th International Conference on Advances in Information Technology, IAIT2016, 19-22

December 2016, Macau, China

Measuring performance of facebook advertising based on media

used: a case study on online shops in indonesia
Tikno *
Universitas
Universitas Internasional
Internasional Semen
Semen Indonesia
Indonesia (UISI),
(UISI), Gresik
Gresik 60119,
60119, Indonesia
Indonesia

Abstract
Abstract

According
According to to TechinAsia
TechinAsia report
report in
in 2015,
2015, more
more than
than 7272 million
million Indonesians
Indonesians are are listed
listed as
as active
active internet
internet users.
users. About
About 18%18% andand 11%
11%
conduct
conduct research
research on
on the
the web
web via
via desktop
desktop andand mobile
mobile device
device respectively
respectively before
before buying
buying online.
online. As
As much
much asas 80%
80% from from those
those who
who
did
did research
research onon the
the internet
internet make
make aa real
real purchase.
purchase. This
This indicates
indicates that
that people
people whowho does
does online
online research
research tend
tend to to buy
buy online
online
afterward.
afterward. Facebook
Facebook has has 14%
14% share
share ofof those
those internet
internet users
users and
and unsurprisingly
unsurprisingly online
online shop
shop owners
owners use use itit to
to advertise
advertise their
their
products
products online.
online. Unlike
Unlike conventional
conventional platforms,
platforms, advertising
advertising onon Facebook
Facebook has has no
no flat
flat rate,
rate, but
but itit depends
depends onon engagement
engagement rate,
rate,
which
which measures
measures how
how much
much the
the ads
ads are
are liked
liked by
by its
its target
target audiences.
audiences. More
More engagement
engagement will will be
be granted
granted with
with low
low advertising
advertising cost.
cost.
The
The advertisers
advertisers need
need toto know
know how
how to to create
create ads
ads which
which have
have more
more engagement
engagement raterate potential
potential and
and lead
lead to
to minimal
minimal cost cost and
and high
high
conversion.
conversion. This
This research
research investigated
investigated performance
performance measurement
measurement of of media
media types
types used
used onon Facebook
Facebook advertising
advertising platforms
platforms such
such as
as
photos
photos and
and videos.
videos. The
The study
study used
used three
three control
control variables:
variables: gender,
gender, age
age group,
group, and
and product
product type
type asas the
the interest
interest group.
group. TheThe findings
findings
of
of this
this study
study showed
showed that
that Indonesians
Indonesians have
have higher
higher engagement
engagement raterate for
for ads
ads in
in which
which videos
videos are
are used
used instead
instead of
of images.
images.
© 2015
© 2015 The
The Authors.
Authors. Published
Published byby Elsevier
Elsevier B.V.
B.V.
© 2017 The Authors. Published by Elsevier B.V.
Peer-review under
Peer-review under responsibility
responsibility of of the
the organizing
organizing committee
committee ofof the
the 8th
8th International
International Conference
Conference on on Advances
Advances in in Information
Information
Peer-review under responsibility of the organizing committee of the 8th International Conference on Advances in Information
Technology.
Technology.
Technology
Keywords:
Keywords: Facebook
Facebook advertising;
advertising; engagement
engagement rate;
rate; ads
ads media
media type
type

** Corresponding
Corresponding author.
author. Tel.:
Tel.: +62
+62 812345
812345 38839.
38839.
E-mail
E-mail address:
address: tikno@uisi.ac.id
tikno@uisi.ac.id

1877-0509
1877-0509 ©© 2015
2015 The
The Authors.
Authors. Published
Published by
by Elsevier
Elsevier B.V.
B.V.
Peer-review
Peer-review under
under responsibility
responsibility of
of the
the organizing
organizing committee
committee of
of the
the 8th
8th International
International Conference
Conference on
on Advances
Advances in
in Information
Information Technology.
Technology.

1877-0509 © 2017 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the organizing committee of the 8th International Conference on Advances in Information
­Technology
10.1016/j.procs.2017.06.016
106 Tikno / Procedia Computer Science 111 (2017) 105–112
Tikno / Procedia Computer Science 00 (2015) 000–000

1. Introduction

Social media are becoming ubiquitous and need to be managed like all other forms of media that organizations
employ to meet their goals. However, social media are fundamentally different from any traditional or other online
media because of their social network structure and egalitarian nature. These differences require a distinct
measurement approach as a prerequisite for proper analysis and subsequent management 1. Depending on business
goals, social media marketing goals can include, but are not limited to, measuring brand health, improving
effectiveness through optimization, generating revenues or improving customer relationships. Whether marketing
objectives are to build long-term brand awareness or stimulate an immediate purchase, companies operating in the
online space seem to have a limited understanding of what social media is or how to use it effectively2.
Social commerce adopts Web 2.0 technologies and infrastructure to assist online interpersonal interactions and
contributions for acquiring and exchanging shopping experiences about products and services3. Furthermore, a study
proposes social commerce to receive or give shopping information including experience, e-word-of-mouth (WOM),
and user-generated content (UGC) for transactions4.
A study suggests that brands can successfully engage with consumers via social media, with clear potential for
building a strong, loyal following who ultimately become ambassadors for the brand4. Through social media,
consumers engage in activities that, regardless of direct intentions of doing so, can enhance or reduce a brand’s
profitability, and influence reputations of organizations which infer that social media’s referral impact on purchase
behavior may be greater than that of advertising originating from the company. Given its inherent WOM nature,
social media’s role in creating customer value is fundamental5. A plethora of ad-hoc tactics for online and social
media marketing have emerged in organizations, and the selection of respective tools and channels has proven a
critical issue for marketing strategists6. Reference 7 identifies some ways that organizations use social media and
describing how to frame the spending decision for social media.
The result of the previous research supports the need to consider more variables that interact to make ads more
effective on social networking sites (SNSs), such as the product type, brand fan’s expertise and the ties of
relationship8. Reference9 represents a quantitative effort to construct a model of users’ attitudes toward Ads on SNSs,
while they proposed content-related, structural, and socialization factors as antecedents. They also suggested a
moderating effect of gender differences on the perceived ads’ features and perceived social network usage’s impact
on attitude toward ads on social networks. Previous research has shown that a social network user may believe that
interactions among ‘friends’ are trustworthy but may independently appraise the site itself or advertising content
displayed inside panels10. Rodgers and Thorson proposed an approach that Internet users perceive and process ads do
not start with the advertising message, but rather represents a response to individual needs11. A study by Kim & Ko
identifies attributes of SNSs marketing activities and examine the relationships among those perceived activities,
value equity, relationship equity, brand equity, customer equity, and purchase intention through a structural equation
model12. Other study finds that social media marketing effectiveness is highly influenced by its messages/content
quality and younger age group as the main users of social media sites are not the main target audience of social
media marketing 13.
Vast use of Facebook around the globe has made it a new and important advertising platform, where businesses
place their ads to reach their prospective customers. This is probably because Facebook allows businesses to target
specific customer and promoting their product or services through effective advertisements. Growing Facebook
based advertising is perhaps an indication that it is becoming an important source of business presentation and the
firms are taking Facebook advertisement as a useful strategy to attract customers 14.
 TiknoTikno
/ Procedia Computer
/ Procedia Science
Computer 00 (2015)
Science 000–000
111 (2017) 105–112 107

Fig. 1. Facebook Statistic per June 2016 (Source: http://newsroom.fb.com/company-info/)

Special features of Facebook ads have attracted researchers’ attention who had adopted a series new variables
about SNS friends, SNS’s social feature and perceived features of SNS ads to evaluate advertisements’ effectiveness
8
. Reference15 explore Facebook ads auctions through the suggested bid feature, showing how this feature can be
used by researchers to make inferences on properties of the ad market for targeting parameters that Facebook
provides, including gender, precise interests, relationship status, education, and workplaces.
At Facebook, ads are not associated with a query but instead, specify demographic and interest targeting. As a
consequence of this, the volume of ads that are eligible to be displayed when a user visits Facebook can be larger
than for sponsored search16. The online factors that influence consumers’ perceptions and attitudes towards
advertising on Facebook are perceived interactivity, advertising avoidance, and privacy17.
A further study investigated the value factors and trust affect the user stickiness of Facebook which flowed out as
user engagement 18. Previous research investigates the factors that affect young customer’s attitude towards fast-
food advertising in Facebook in Egypt and provides a framework for the factors that lead to consumers' engagement
with the advertisements on Facebook 19.
Shopping online is experiencing and will continue to have a rapid increase in Indonesia. As in 20, more than 72 of
255 million Indonesian is the active internet users. 18% of these users do online research from the desktop computer
before buying online and additional 11% using a mobile device for their research. 16% and 9% is converted for
desktop and mobile users respectively. Most traffic comes from Facebook which accounted for 14%, then
WhatsApp (12%) and Twitter (11%). These numbers show that Facebook as one of the SNSs in Indonesia has a big
portion of traffic share. Online shop owners start to use the traffic source as their advertising channel.
In the area of a paid advertising on Facebook, measuring of its performance is possible and highly important. On
Facebook there are commonly used the web advertisements. The significant difference to common web pages is that
Facebook users manifest their preferences, hobbies and sympathies. All such information is very valuable for
marketers to target their marketing campaigns to the specific customer groups or even individuals.
Advertisements on Facebook are subject to an auction, where advertisers compete among themselves for an
audition through bidding the price of click on their ad (pay per click) or the price of displaying their ad to selected
target group (pay per impression). Advertisers get suggested bid range which currently is winning the auction
among similar adverts. The advertiser sets maximum bid (per click or per thousand impressions), but Facebook
charges only the amount required for an advert to win an auction. This price may be lower than the maximum bid set
by advertiser, making a marketing campaign more efficient. Advertisers have also control of their daily budget – the
maximum amount that can be spent on campaign per day. If a daily budget is spent, the advertisement will
automatically stop showing until next day of a current campaign21.
Facebook has wide advertising targeting options, which can improve performance and effectiveness of a
company’s online advertising, because ads are displayed to the users who are most likely to be interested in
advertised object, thanks to information shared on Facebook users’ profiles. With such a precise targeting company
108 Tikno / Procedia Computer Science 111 (2017) 105–112
Tikno / Procedia Computer Science 00 (2015) 000–000

can reach demanded customer group according to its advertising goals. For an advertisement any number of
targeting filters can be set 21.
It is obvious that the phenomenon of advertising has changed from push advertising to trust based advertising.
Although Facebook advertising offers this advantage, the nature of such advertising dictates that the successful
effectiveness on consumers should include features such as participation, realization, personalization and feedback.
It is now more critical than ever that successful companies use engagement advertising to determine for successfully
engaging their customers before, during and after their purchase cycle, and that the basis for this engagement is
enhancing either brand image and brand equity 22.
Moreover, many of the users agree that their willingness to buy brands has increased as a result of seeing the
most ‘likes’ and ‘shares’, which suggested that the brand is reputable. As many consumers have accepted Facebook
as a fashionable way of advertising and since most of the members of Facebook are young adults, they tend to be
interested in finding unique styles by pursuing the information on a brand’s pages and through the recommendations
and practices of friends and close acquaintances. In order to survive in today’s intensely competitive market, firms
need to be informed of the brand awareness of their consumers and devise up-to-date advertising strategies
accordingly22.
However, research on this topic is limited and invites other researchers to explore and investigate factors that
may affect Facebook advertising performance. Previous research analyzes and design ads placement in the
Facebook platform for a traditional food producer and finds that most effective ads are mobile ads 23. Surprisingly,
none of those finds how the Facebook ads will give a good return-on-investment (ROI) to the advertisers. Facebook
stated that they would give a cheaper rate for advertising which complies with their rules and engages their user
more. In other words, ads which have high engagement rate will be rewarded with lower cost.
How consumers perceive marketing communication is related to relevancy of an advertisement and ability to
receive marketing communication. Which features of social network advertising have the most significant effect on
the attitudes of its members is related to disruptive advertisement part as well as the parts covered by Elaboration
Likelihood Model (ELM). Which features of social network advertising are found most appealing and which are
found most annoying. This is being handled in the Disruptive advertisement as well as the whole model. These
features will affected on the attitude towards advertising only if the consumer is able to receive marketing
communication thus the ability to receive is the feature in the background. If consumer is able to receive marketing
communication, he/she can form attitudes towards it that are divided to negative and positive attitudes which form
engagement rate. It also emphasizes the fact that the engagement rate affects decision making24.
The engagement rate is the proportion of action taken by the Facebook user to displaying ads with how many
times the ads is showed.

E=A/R (1)

Which :
E = Engagement Rate
A = Action taken (Amount of People who liked, commented, shared, or clicked on the ad)
R = Reach (Amount of People who saw the ad)
Apparently, it depends on ads copy, audience matching/targeting, and ad media. In Facebook ads, the advertiser
may use image or video as ad media, which affects on their ads performance reflected by value of the engagement
rate. Facebook will give cheaper advertising cost and more views for the ad with higher engagement rate, which
obviously motivates the online shop owner as the advertiser, to increase the engagement rate for their ads.
The objective of this research is investigating which advertising media in Facebook ads platform which will gain
higher engagement rate given some variables: gender, age group, and interest group for promoting online shops in
Indonesia.

2. Methodology

To address the research question, 18 ad-sets was conducted in Facebook advertising platform. Some previous
works above concluded that gender, age group and interest affects the ads performance. Gender and age group were
 Tikno / Procedia Computer Science 111 (2017) 105–112 109
Tikno / Procedia Computer Science 00 (2015) 000–000

chosen because previous research shows that the gender-orientation of the key phrase is a significant determinant in
predicting behaviors and performance25. Other study finds that age group, gender, education, interests, location and
relationship status all appear to affect ads on Facebook26. On another hand, Indonesia online shop has unique
characteristics which represent by product types, which are: physical product (P), digital product (D), and service
(S).
This research measuring the engagement rate of each adset (a particular audience setting of ad in Facebook)
which combine those control variables and the ad media. They divided such are : gender (Male/Female); age group
(18-24/25-34/35-44); and interest (Physical/Digital/Service) product as target audience. In each adset, two ads are
served. One ad was using the image as media and video for another ad. For example (see table 1), an adset targets
audience which interested in digital product (D), female (F), and in age group of 18-24 years old (1824).
Because the goal was to measure the engagement from each ad in each adset, the same period and budget was ran
on Facebook. Those ads were monitored per day to record reach count, and action was taken by the audience which
consists of liking, commenting, sharing, and clicking the ads. Furthermore, these numbers were calculated and tested
in t-distribution to show the findings

3. Findings

After ran all ads for 7 days, the data gathered is presented below:

Table 1. Campaign result

Media
No Adset Video Image
R A Ev R A Ei

1 D F 1824 2645 404 15,3% 7977 639 8,0%

2 D F 2534 3377 848 25,1% 1851 170 9,2%
3 D F 3544 6215 1594 25,6% 3255 131 4,0%
4 D M 1824 4659 809 17,4% 12681 742 5,9%
5 D M 2534 6600 1359 20,6% 6205 283 4,6%
6 D M 3544 4370 1263 28,9% 2668 177 6,6%
7 P F 1824 8530 2914 34,2% 152 29 19,1%
8 P F 2534 6533 2442 37,4% 40 8 20,0%
9 P F 3544 5521 2657 48,1% 10 1 10,0%
10 P M 1824 14312 5545 38,7% 251 34 13,5%
11 P M 2534 12711 5126 40,3% 11 1 9,1%
12 P M 3544 8813 4643 52,7% 10 0 0,0%
13 S F 1824 15467 2649 17,1% 983 67 6,8%
14 S F 2534 10556 2684 25,4% 664 28 4,2%
15 S F 3544 8474 3125 36,9% 505 29 5,7%
16 S M 1824 24627 4479 18,2% 944 54 5,7%
17 S M 2534 19274 4338 22,5% 977 43 4,4%
18 S M 3544 13663 3861 28,3% 670 42 6,3%

A paired-samples t-test was conducted to compare engagement rate of ads using video (Ev) and image (Ei).
There was a significant difference in the scores for video (M=29,6%; SD=10,9%) and image (M=7,9%; SD=5,1%)
110 Tikno / Procedia Computer Science 111 (2017) 105–112
Tikno / Procedia Computer Science 00 (2015) 000–000

conditions; t(17)=8,28, p=0,00. The result suggests that media which is used in Facebook ads does have affect to
engagement rate. Specifically, our result suggests that when the video is used, the engagement rate increases. Video
media also outnumbers image in data mean for this engagement rate by more than 20%.

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 Ev 29,594% 18 10,9457% 2,5799%

Ei 7,953% 18 5,1112% 1,2047%

Paired Samples Test

Paired Differences (%) Sig.

95% CID (2-

M SD. SE M Lo Up t df tailed)

Pair 1
21,64 11,09 2,61 16,13 27,16 8,280 17 ,000
Ev - Ei

Fig. 2. Paired samples t-test result

This higher engagement rate gives more benefit for online shop owner/advertiser by reducing their advertising
cost, reaching more audiences, and increasing potential conversion into further transaction. It also means that
Facebook user more engage with video content and would like to interact more with the content although it contains
advertorial message.
From the findings of this research, the online shop marketers and advertising designers can also understand and
pay more attention on how the users of Facebook interact with the advertisement when making advertising decisions.
Besides providing literature for future scholarly research avenues, it is also hoped that this will be a good starting
point for researchers to further examine factors which affected buying conversion on an online shop.
This study took a critical factor at Facebook advertising and how it affects the online shop owner as an advertiser
of this site. As of now, there is no conclusive data as to a “generic” perception of Facebook advertising. This
conclusion directly relates to the uses of ads media and whether it affects the ads performance. Reference27
mentioned that researchers who used limited-parent population could only justify the dimensions explaining the
research framework without generalizing the findings and drawing conclusions about the overall public attitude
towards advertising. In this research, any conclusion drawn from this limited-parent population sample, which are
limited by interest and total of reach, would be misleading because the sample basis does not represent the total
population despite the justifications as mentioned above.
This study was also limited because it could not represent all online shops in the three categories. Also, the
manner in which these data obtained is depended on Facebook algorithms. Therefore, the conversion rate from these
ads was not represented in the results. The findings, however, represent Indonesian Facebook user perceptions from
a sufficiently diverse range of age group and gender where Facebook advertising has been ran.
To further understand how the ads lead to actual buying transaction, more research should be conducted. It would
be beneficial to follow some online shops which are advertising on Facebook, and see how the users interact with
the ads and what factors affects them to convert into a buyer. Future research may also consider using a more
general and representative population by raising the sample size of Indonesian Facebook user.

Acknowledgment

The author wishes to thank all colleagues in Universitas Internasional Semen Indonesia at Gresik. This work was
supported in part by a grant from LPPM UISI.
 Tikno / Procedia Computer Science 00 (2015) 000–000

Tikno / Procedia Computer Science 111 (2017) 105–112 111

References

1. H. Zhang, Y. Lu, S. Gupta and L. Zhao, "What motivates customers to participate in social commerce? The impact of technological
environments and virtual customer experiences," Journal Information and Management, pp. 1017-1030, 2014.
2. C. Chung and K. Austria, "Social Media Gratification and Attitude toward Social Media Marketing Messages: A Study of the Effect of Social
Media Marketing Messages on Online Shopping Value," in Proceedings of the Northeast Business & Economics Association, 2010.
3. T.-P. Liang, Y.-T. Ho, Y.-W. Li and E. Turban, "What Drives Social Commerce: The Role of Social Support and Relationship Quality,"
International Journal of Electronic Commerce, pp. 69-90, 2011.
4. C. Bond, C. Ferraro, S. Luxton and S. Sands, "Social media advertising: An investigation of consumer perceptions, attitudes, and preferences
for engagement," in Proceedings of the Australian and New Zealand Marketing Academy (ANZMAC) Conference 2010 -'Doing More with
Less', Christchurch New Zealand, 2010.
5. P. D. Berger, M. Pooja, J. E. Black, J. Cao and B. D. Weinberg, "The Impact of Social Media Usage on Consumer Buying Behavior,"
Advances In Management Vol 5 Iss 1, 2012.
6. M. J. Valos, M. T. Ewing and I. H. Powell, "Practitioner prognostications on the future of online marketing," Journal of Marketing
Management Vol 26, pp. 361-376, 2010.
7. B. D. Weinberg and E. Pehlivan, "Social spending: Managing the social media mix," Business Horizons Vol 54 Iss 3, pp. 275-282, 2011.
8. K. T. Chang, W. Chen and B. C. Tan, "Advertising Effectiveness in Social Networking Sites: Social Ties, Expertise, and Product Type,"
IEEE Transactions on Engineering Management Vol 59 Iss 4, pp. 634-643, 2012.
9. D. G. Taylor, J. E. Lewin and D. Strutton, "Friends, Fans, and Followers: Do Ads Work on Social Networks? How Gender and Age Shape
Receptivity," Journal of Advertising Research Vol 51 , pp. 258-275, 2011.
10. A. M. Soares, J. C. Pinho and H. Nobre, "From Social to Marketing Interactions: The Role of Social Networks," Journal of Transnational
Management Vol 17 Iss 1, pp. 45-62, 2012.
11. S. Rodgers and E. Thorson, "The Interactive Advertising Model : How Users Perceive and Process Online Ads," Journal of Interactive
Advertising Vol 1 Iss 1, pp. 41-60, 2000.
12. A. J. Kim and E. Ko, "Do social media marketing activities enhance customer equity? An empirical study of luxury fashion brand," Journal
of Business Research Vol 65 Iss 10, pp. 1480-1486, 2012.
13. C. Pradiptarini, "Social Media Marketing: Measuring Its Effectiveness and Identifying the Target Market," UW-L Journal of Undergraduate
Research XIV, 2011.
14. F. u. Rehman, M. Ilyas, T. Nawaz and S. Hyder, "How Facebook Advertising Affects Buying Behavior of Young Consumers: The
Moderating Role of Gender," Academic Research International Vol 5 Iss 4, pp. 395-404, 2014.
15. Y. Liu, C. Kliman-silver, R. Bell, B. Krishnamurthy and A. Mislove, "Measurement and analysis of OSN ad auctions," in Proceedings of the
second ACM conference on Online social networks, New York, 2014.
16. X. He, J. Pan, O. Jin, T. Xu, B. Liu, T. Xu, Y. Shi, A. Atallah, R. Herbrich, S. Bowers and J. Q. Candela, "Practical Lessons from Predicting
Clicks on Ads at Facebook," in Proceedings of the Eighth International Workshop on Data Mining for Online Advertising, New York, 2014.
17. A. Y. Yakoop, M. M. Anuar, K. Omar and A. Liaw, "Consumers’ Perceptions and Attitudes Towards Advertising on Facebook in Malaysia,"
in World Business and Economics Research Conference, Auckland New Zealand, 2012.
18. H.-L. Yang and C.-L. Lin, "Why do people stick to the Facebook web site? A theory-based value view," Information Technology & People,
Vol. 27 Iss: 1, pp. 21-37, 2014.
19. H. R. Gaber and L. T. Wright, "Fast-food advertising in social media. A case study on Facebook in Egypt," Journal of Business & Retail
Management Research Vol 9 Iss 1, p. 52, 2014.
20. TechinAsia, "Berapa jumlah pengguna website, mobile, dan media sosial di Indonesia?," 21 January 2015. [Online]. Available:
https://id.techinasia.com/laporan-pengguna-website-mobile-media-sosial-indonesia.
21. Facebook, “How Facebook Ads Work,” [online]. Available from: https://www.facebook.com/business/learn/facebook-ads-basics/
22. M. Dehghani, M. Tumer, “A research on effectiveness of Facebook advertising on enhancing purchase intention of consumers,” Journal of
Computers in Human Behavior 49, pp. 597-600, 2015.
23. D. K. Aji, Y. Rohayati and A. Fashanah, "Advertising Design Analysis and Site Selection Based on "Facebook Ads" for Seaweed Dodol of
SME's Aulia Sari," e-Proceeding of Engineering Vol 2 No 2, 2015.
24. V. Lukka, P. T. J. James, “Attitudes toward Facebook advertising,” Journal of Management and Marketing Research 14, pp. 1-26, 2014.
25. J. Jansen, K. Moore and S. Carman, "Evaluating the performance of demographic targeting using gender in sponsored search," Information
Processing & Management Journal, pp. 286-302, 2013.
112 Tikno / Procedia Computer Science 111 (2017) 105–112
Tikno / Procedia Computer Science 00 (2015) 000–000

26. S. Guha, B. Cheng and P. Francis, "Challenges in measuring online advertising systems," in Proceedings of the 10th ACM SIGCOMM
conference on Internet measurement, New York, 2010.
27. S. Shavitt, P. M. Lowrey and J. E. Haefner, “Public Attitudes Toward Advertising: More Favorable Than You Might Think,” Journal of
Advertising Research 38(4), pp. 7-22, 1998.