THE ROLE OF ARTIFICIAL INTELLIGENCE IN CONTENT CREATION AND CHECKING ITS EFFECTIVENESS IN THE GOOGLE ADS ADVERTISING SYSTEM

Abstract

n Hungary, 94% of businesses have Internet access and 63% have a website. Moreover, online retail sales will reach HUF 1,203 billion in 2021. In order for companies to achieve the largest possible market share, they can use various digital marketing strategies. They are distinguished according to different methods. One of the most commonly used in practice and in science is inbound (as search engine optimization) and outbound (as advertising). Google Ads, which emerged at the turn of the millennium and defined itself as the world’s first company to use machine learning technology, is a market leader. Their ad system was initially based on keywords, which have since been expanded to include more than 4,800 types of targeting criteria. These targeting options are available for a variety of ad formats. The digital solutions to the billboards of traditional marketing are banner ads, called Display on Google. These ads contain image, video, and text content and aim to interrupt the consumer’s activity and redirect them to the advertiser’s website. Since they are capable of increasing website traffic by up to 300%, this can be interpreted as an opportunity that is also considered favorable by businesses. It is also suitable for testing various content elements, as one of its main indicators, the click-through rate, expresses the relevance of the ad, as several researchers have noted. As the role of artificial intelligence grows, more and more companies are using it as a competitive advantage. Some of their algorithms are capable of generating text, images, videos, or other content. In this study, I leverage the power of display ads and conduct my research in the Google Ads system instead of conducting consumer surveys. I created two ads for the same target audience, with the same budget and settings. The text content for one ad was created by a marketer, the image content was created by a professional photographer, and the content elements for the other ad were provided by Artificial Intelligence. The objective of the article is to study the performance, efficiency, and impact of artificial intelligence-generated content on conversions under real market conditions. The study also includes content created by the players.

Full text

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
259
THE ROLE OF ARTIFICIAL INTELLIGENCE IN CONTENT
CREATION AND CHECKING ITS EFFECTIVENESS IN THE
GOOGLE ADS ADVERTISING SYSTEM
Zoltán SOMOSI
Marketing and Tourism Institue, Faculity of Economics, University of Miskolc,
Miskolc, Hungaryzoltan.somosi@uni-miskolc.hu
Abstract: In Hungary, 94% of businesses have Internet access and 63% have a
website. Moreover, online retail sales will reach HUF 1,203 billion in 2021. In order
for companies to achieve the largest possible market share, they can use various
digital marketing strategies. They are distinguished according to different methods.
One of the most commonly used in practice and in science is inbound (as search
engine optimization) and outbound (as advertising). Google Ads, which emerged at
the turn of the millennium and defined itself as the world's first company to use
machine learning technology, is a market leader. Their ad system was initially
based on keywords, which have since been expanded to include more than 4,800
types of targeting criteria. These targeting options are available for a variety of ad
formats. The digital solutions to the billboards of traditional marketing are banner
ads, called Display on Google. These ads contain image, video, and text content
and aim to interrupt the consumer's activity and redirect them to the advertiser's
website. Since they are capable of increasing website traffic by up to 300%, this
can be interpreted as an opportunity that is also considered favorable by
businesses. It is also suitable for testing various content elements, as one of its
main indicators, the click-through rate, expresses the relevance of the ad, as
several researchers have noted. As the role of artificial intelligence grows, more
and more companies are using it as a competitive advantage. Some of their
algorithms are capable of generating text, images, videos, or other content. In this
study, I leverage the power of display ads and conduct my research in the Google
Ads system instead of conducting consumer surveys. I created two ads for the
same target audience, with the same budget and settings. The text content for one
ad was created by a marketer, the image content was created by a professional
photographer, and the content elements for the other ad were provided by Artificial
Intelligence. The objective of the article is to study the performance, efficiency, and
impact of artificial intelligence-generated content on conversions under real market
conditions. The study also includes content created by the players.
Keywords: Digital Marketing; Artificial Intelligence; Advertising; Google Ads; PPC
JEL Classification: M31; M34
1. Introduction
The new fields that digital marketing has created have also stimulated theorists and
researchers. A search for the keyword digital marketing on sciencedirect.com also
shows that in 2000 there were only 519 publications, while by 2020 this number
had quadrupled. According to Hungarian Central Statistic Office (KSH, 2020) data,
internet traffic in Hungary is growing dynamically, by 2020 download traffic

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
260
increased by 18% and upload traffic by 29% compared to 2019, and the
percentage of businesses using the internet reached 94%. Their comprehensive
research also included websites and found that 63% of Hungarian businesses
have a website (Hungarian Central Statistic Office, 2020). The growing number of
businesses appearing online reflects changing consumer behavior. in 2021, there
were 68 million transactions worth 1,203 billion forints (GKID, 2022). In order for
the company to increase its market share and make an effective sale in the online
space, it needs to work with a combination of different strategic marketing
activities, which Bleoju et al. (2016) presented in their research.
Search engine optimization for inbound marketing is given special attention
because it is not only the digital marketing channel that offers the best ROI (Shirey,
2022), but its optimal design allows the owner to direct potential customers to the
website in an organic way by displaying the website in a better place than other
websites. among the results of a search page (Egri and Bayrak, 2014). Without
doubting the importance of search engine optimization, I question its principle
because if every company considers it as the most important tool, no meaningful
change in search results can be achieved. Therefore, it becomes necessary for
companies to optimize other channels, such as click-based online advertising,
which can be considered partly as an outbound and partly as an inbound strategy.
In his paper, Kulova (2021) points out that these advertising platforms are effective
for both customer acquisition and retention. In this transformation, Google plays a
role as an active contributor (Mehta et al., 2007).
The closed advertising system was originally called Google Adwords, referring to
the fact that ads are displayed based on keywords and search terms that
consumers type into the search engine (Za and Tricahyadinata, 2017). Advertising
opportunities have continued to expand, leading to enrichment of segmentation
opportunities. Instead of or in addition to keywords, 4,809 types of factors can be
selected to reach the target audience. This list can be viewed by anyone at
Vidhoarder.com (2022). In addition to each segmentation category, they also
address in their list which of Google's advertising systems can be used: Display,
Video In-Stream, Video Discovery, Gmail, Search, Shopping. One of the 10 most
important factors defined by Alcouffe (2013) concerns the content elements of ads.
According to him, more effective communication can be achieved by creating
attractive ads.
Among the listed Google Ads advertising options, display ads are the ones where
both image and text presentation play a particularly important role. So, in this
study, the question was in what form artificial intelligence can provide the content
elements of the ad. In the first phase of the study, a literature review is conducted
that seeks an answer to the question of the optimal design of the Google Ads ad
and the possible applications of artificial intelligence, in addition to the
communication options. Then, instead of a consumer survey, ads created in the
system under active conditions are presented, one by Artificial Intelligence and the
other by a marketer. The results section of the study examines the conversion
differences between the ads placed with a small budget and looks for the answer
to the question: can artificial intelligence communicate and achieve business goals
more effectively than a marketer?

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
261
2. Literature review
It is interesting that the appearance of television almost 100 years ago caused at
least as much doubt (Stephens, 2022) as the possibilities of the Internet, since
many assumed that with its appearance the strategy that companies could use
would be limited (Porter, 2001 ). However, the opposite was true, and not only
tactical solutions but also strategic trends developed. A study by Ascend 2 (2019),
which looked at marketing strategies, showed that the main goal for companies
was still revenue (64%), but alongside this, many other conversion goals were
mentioned that had nothing to do with revenue (e.g., increasing website traffic by
25%). Respondents considered content marketing to be the best tactic (58%),
followed by search engine optimization (50%) and ads on search engines and
social media in 4th place with 34%.
2.1. Digital marketing channels
In reviewing the work of practitioners and theoretical researchers, I have concluded
that typically between 5-12 digital marketing channels are identified (Siddiqui,
2020, Lane, 2022). And the differences arise from inconsistent blending and
separation of categories. Lindley's (2022) practical article divides channels into a
total of 7 parts: social media marketing, search engine optimization, email
marketing, video marketing, affiliate marketing, influencer marketing, and click-
based advertising. Among these channels, which have emerged in the course of
the development of information and communication technologies and whose aim is
to influence the communication of the company's offer and the decision-making of
customers (Aswani et al., 2018), in this paper I focus on PPC-based advertising.
The importance of this topic is confirmed by the results of Josifovska's (2022) study
on PPC statistics, according to which spending on search engine advertising will
exceed $191 billion by 2024. Almost half of the companies plan to invest in this
marketing channel. Hill (2019) found for various PPC platforms that display ads can
increase website traffic by 300% and Google can reach 5.6 billion people.
McCandless (1998) defined banner advertising in his pre-millennial article. It is the
most common form of advertising on the Internet and can be placed on the side,
top or even bottom of the website depending on its position. Murphy, Peltekian,
and Gardner (2018) also mention a specific advertising system in their definition of
display advertising: ―Paid advertisements that appeared on the side of the user‘s
internet browser while browsing the internet on various websites determined by
Google Ads for their relevance and suitability‖ (Murphy, Peltekian and Gardner,
2018:4). The purpose of this ad is to direct consumers to a specific website thanks
to its clickable form. Ahmed and Kwon (2014) emphasise that despite the existence
of various online advertising channels, companies usually opt for display ads,
which are studied by many researchers from different perspectives.
2.2. Digital marketing measurement
Costing for online ads can be done in several ways. Mangani (2004) was the first to
examine the framework for display-based pay-per-view and search-based pay-per-
click pricing methods. In terms of cost, the equivalent of click-based advertising is
CPC (cost-per-click), while in the case of pay-per-view it is CPM (cost-per-

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
262
impression) (Ahmed and Kwon, 2012). Fjell (2010) has already studied the
relationship between this cost and click-through rate. In my opinion, he is right
because the performance indicators and metrics of these click-based ads are
derived from the clicks. Consumer (user) acceptance and perspective is expressed
by click-through rate (Yang and Zhai, 2022). For this reason, both researchers and
practitioners are concerned with it in general (Robinson et al., 2007) or in the
context of its prediction (Richardson et al., 2007). Dean (2022) used 4 million
search results as the basis for his study CTR, to determine click-through rates for
organic results. On average, the first non-advertised result has a click-through rate
of 27.6%, while the tenth result on the same page has only a tenth of that. Valve
and Metre's (2018) study looked specifically at ad click-through rates, which
averaged 2% across all industries. Within these ads, Lincoln (2021) focused
specifically on display ads and found that the click-through rate in this case was
only 0.46%. However, there are other ways to measure performance, but this area
is theoretically underdeveloped (Järvinen and Karjaluoto, 2015). They reflected on
the structure of the metrics as follows, "Marketers need a comprehensive but
manageable set of performance metrics, which requires that they understand the
relationships among the metrics and are able to focus on the critical metrics"
(Järvinen and Karjaluoto, 2015:120). Accordingly, in their research, they
established several metrics that serve as the basis for measuring efficiency. Brown
(1996) made eight different suggestions for effective performance measurement:
It makes more sense to use fewer indicators,
It is necessary to link the measured data to success factors,
Indicators must focus on the past, present, and future,
The development of indicators must be based on stakeholder expectations,
There is a need to assess performance at the whole process level,
Indicators can be combined for a more comprehensive assessment,
There is a need to ensure flexibility of indicators,
Indicators need to be linked to business objectives.
In his article, Bonini (2018) examines the metrics found and applicable in the
Google Ads system, which not only meet Brown's (1996) definitions but also serve
as the basis for important conclusions from the perspective of this study. These
include impression, which reflects the total number of impressions achieved in
each time period, and clicks, which allow conclusions to be drawn about the
relevance of the ad when the click-through rate is calculated from the previous two
data. In addition, the bounce rate, the time spent on the page and the number of
engaged users can also be studied, but these data are not presented in this study.
In contrast to the conversion rate, which represents the achievement of the goal set
by the company, the request for a quote for the service. In Hungary, Gál (2015,
2016) has published several books about the Google Ads system, which help
companies to optimize their ads and had a great influence on my research
direction. Science does not deal specifically with the creation of ads for several
reasons. One of the most important is that it is a very rapidly changing field, so
publications become outdated too quickly. Therefore, I will not address ad setting in
this study. Google Ads has published informational material (Google.com, 2022) for
successfully setting ads.

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
263
2.3. Artificial Intelligence in marketing
According to Google.com (2022), display ads can reach 90% of Internet users
through websites, videos, and apps. And the message is delivered based on the
definition of different criteria. Gale's (2022) thoughts on Google Ads display ads:
"When you think of Google Ads targeting, you probably think of keywords. What
you probably do not immediately think of is audiences: Ads for people based on
who they are, not what they are looking for." (Gales, S.J., www.wordstream.com).
Artificial intelligence is already playing an active role in this process. There are
several machine learning techniques, such as supervised machine learning, which
can be used in the separation of pre-labelled data (spam mail filtering), or
unsupervised learning, which can be used in clustering and segmentation (Adoni,
2018). In the article by Data-flair (2019), it can be read that Google calls itself the
first company to use machine learning techniques. This is complemented by the
article by Google Debelopers (2022), which states that the applications of
supervised machine learning techniques are more diverse than unsupervised
machine learning. Even after extensive research, no information can be found that
clearly indicates which method is used to collect consumer data and effectively
implement segmentation and targeting in the Google Ads system.
In any case, AI has a significant impact on daily life and is estimated to be worth
over $136 billion by 2022. This value is expected to increase 13-fold in the next 8
years (Howarth, 2021). Key statistics from Jovanovic (2022) include data on
companies: 37% of companies use artificial intelligence. Moreover, nine out of ten
leading companies are investing in this capability. On his website, Marr (2021)
gives several examples of how companies are using artificial intelligence and for
what activities. Some for self-driving cars (Alphabet), others for digital assistants
(Amazon, Apple, Google). However, from the perspective of this study, the method
used by the Alibaba company is important because it automatically creates product
descriptions - that is, it is used to generate content. The technology was
characterized on Alibabatech.com (2018), and its benefits were defined in three
points. Discovers consumer preferences, generates selling points, and
personalizes the process. In my research, I am looking for an answer to the
question of how the content created by the automata behaves in the Google Ads
system.
3. Research method
The literature review confirmed that among the applicable channels of digital
marketing, click-based ads, which include display ads, are an effective solution to
reach consumers. Instead of surveying consumers, the study evaluates the role of
artificial intelligence in content creation, its performance, and its impact on
conversions through this advertising channel.
3.1. Content elements of the research
To conduct the study, an appropriate website (simplified.com) and a marketing
expert created an appropriate amount of content that included the keywords
wedding photo and wedding video, and the name of the base company in
Hungarian. In both cases, the created texts contained 5 headlines with a maximum
length of 30 characters each. Also, 1 long title line with 90 characters and 4

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
264
descriptions with a maximum length of 90 characters each. Table No. 1 contains
sample texts (translated from Hungarian into English).
Table 1: Texts, created by marketing expert and AI
Marketing expert
Artificial Intelligence
Title example
Wedding photos and videos
for the price of '22
Professional photo and video
Title example
Your wedding is an eternal
memory
A beautiful, lasting memory
Title example
Trust us with your wedding
photography
Several years of experience with
weddings
Title example
Trust us with your wedding
video
We preserve your memories for a
lifetime
Long Title
For love, there is only one
remedy: marriage. And we will
capture your big days.
Our experienced team captures your
wedding in stunning detail, from start
to finish.
Description
example
Fulfil your dream, and we will
make it a lifelong memory!
Request a quote today or contact us
for a free consultation: We will
answer all your questions
Description
example
Remember with our wedding
photos and relive your big day
with our video!
We offer a wide range of photography
and videography services to capture
your big day
Description
example
Request a quote for wedding
photography and video at
2022 prices today!
Our goal is to give you an everlasting
memory with stunning images and
videos!
Source: Own editing
Besides headlines and descriptions, the most important element of display ads is
the visual design.
Figure 1 Pictures, created by a Photographer and AI
Source: Own editing

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
265
The left side of the image collection at Figure No. 1 was created by a professional
photographer, while the right side was created using a special artificial intelligence
page.
The image generation function of artificial intelligence already provides several
setting options. The length and verbosity of the specified expression affects the
level of detail in the final image. Many websites offer settings that affect the style of
the final image. The first image on the right is marked as natural and the second
image on the right is marked as futuristic. The third image on the right, which is
different from the usual illustration where you see an AI-generated flower instead of
humans, turned out not to be the type of image normally shown in advertising when
consulting with the photographer and marketer. Although the images do not show
this, the AI-generated faces are still indistinguishable, they are abstract, and this
was also shown in the advertising system during the test.
3.2 Conducting the research
Since the purpose of the research is to evaluate the role of artificial intelligence in
content creation, its effectiveness, and its impact on achieving business goals, two
ads were created in the Google Ads system. These ads are display ads created by
a marketing specialist. The base business sells video and photography services for
weddings, and the specific ads were created in this context, with the goal of
requesting a quote for the service. The ads followed the following system:
The content elements of the ads reached an excellent level, based on the Google
Ads system (they contained the right amount and quality of text and images). For
excerpts, see ―3.1 Content Elements of the Research‖
The ads were shown only to women aged 25-44 living in Hungary (consistent
geographic and demographic targeting was used)
The ads included only ―life events, getting married soon‖ as an additional option to
narrow down the target audience.
The ads ran for 10 days within a period (affected time: 09/09/2022-19/09/2022)
The ads ran on a consistent - low - budget.
The additional settings of the displays have been adjusted as far as possible so
that only differences in content provide information.
4. Results
If we assume that the precision settings made in the Google Ads system were able
to display the ads at the same time, under the same conditions, and to the same
audience, then there is no variable other than the difference in the text and image
content of the ad. So, the results for the two ads show the performance of the
content. Table No. 2 contains the main indicators that were also listed in ―2. The
literature review‖. Marketing expert in the left column, Artificial Intelligence in the
right column.

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
266
Table 2 Texts, created by marketing expert and AI
Marketing expert
Artificial Intelligence
Clicks
103
95
Impressions
112000
113000
Click Trough Rate
0.09%
0.08%
Avarage CPC
49.43 HUF
53.61 HUF
Total costs
5090 HUF
5090 HUF
Number of conversions
2
1
Estimated conversion value*
798000 HUF
399000 HUF
Estimated ROI*
15597%
7739%
Source: Own editing
*A peculiarity of the wedding photography and videography market is that the time
between sending and accepting an offer that meets the needs of consumers is
relatively long. The value of the submitted price offers is listed in the ―Estimated
conversion value‖ line. These offers have not yet been accepted, but when they
are accepted, the values in the line ―Estimated ROI‖ are converted to real values.
4.1. Examination of the results
The ad written by the marketing expert received more clicks, a total of 103, and the
Artificial Intelligence achieved 95 clicks. The impression rate was also almost the
same between the two ads. As a result, the CTR also differed only slightly, with a
value of 0.09% for the marketing expert, and 0.08% for the AI. The cost per clicks
also reflected these data, in the case of AI, a click cost almost HUF 4 more. The
number of conversions was 2 in the case of the marketing expert, while 1 in the
case of the AI. The value of the offer issued by the company for 1 conversion was
HUF 399,000 on average, which also assumed the same value in these cases. The
estimated value was highlighted accordingly, which becomes a real value after
consumer acceptance, like the ROI value.
5. Conclusions
The ads created had a much lower click-through rate than average (Lincoln, 2021):
only 0.08 and 0.09% compared to 0.42%. However, the cost per click is much
lower than the market average (Bobchenok, 2022): 49-54 HUF instead of the
average 274 HUF. On average, the conversion rate is between 2-5% (Kim, 2022).
The marketer's ad achieved a rate of 1.94%, while AI's ad achieved only 0.94%.
In my opinion, the prepared text content meets the general requirements as it
appeals to both emotions and realistic motives. It is not possible to draw general
conclusions from the low budget, but in such a specialized market, the small
differences are likely due to differences in image content.
The study has several limitations, which are listed below:
The segmentation of ads was based on local, demographic, and life event criteria;
further fine-tuning of these criteria could yield better results,
The amount spent on the ads was at the author's expense, so the results represent
a small number due to budget constraints,

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
267
Survey results may be different at different times, for different age groups, and in
different locations.
Despite these limitations, however, we can conclude that AI's role as a content
creator exceeds my expectations and is little different in effectiveness from content
created by a marketer. The fact that the amount spent on advertising significantly
exceeds the expected revenue means that it also has a positive impact on
achieving business goals.
References
1. Adoni, H. W. (2018). How Machine Learning Works.
https://www.researchgate.net/publication/325596088_How_Machine_Learning_Wo
rks
2. Ahmed, Md.T. and Kwon, C. (2012). Pricing game of online display
advertisement publishers. European Journal of Operational Research, [online]
219(2), pp.477–487. doi:10.1016/j.ejor.2012.01.008.
3. Ahmed, Md.T. and Kwon, C. (2014). Optimal contract-sizing in online display
advertising for publishers with regret considerations. Omega, [online] 42(1),
pp.201–212. doi:10.1016/j.omega.2013.06.001.
4. Alcouffe E (2013) 10 Best Practices for AdWords. EffilabAdsonwall: the
family.co.
5. Alibabatech.com (2018). How Alibaba‘s AI Writer is Starting the Content
Revolution. [online] Available at: https://alibabatech.medium.com/how-alibabas-ai-
writer-is-starting-the-content-revolution-27804d845e11 [27 Oct. 2022].
6. Ascend2. (n.d.). 2019 digital marketing strategies: Survey summary report.
Available at http://ascend2.com/wp-content/uploads/2018/10/Ascend2-2019-
Digital-Marketing-Strategies-Report-181005.pdf
7. Aswani, R., Ilavarasan, P.V., Kar, A.K. and Vijayan, S. (2018). Adoption of public
WiFi using UTAUT2: An exploration in an emerging economy. Procedia Computer
Science, [online] 132, pp.297–306. doi:10.1016/j.procs.2018.05.180.
8. Bleoju, G., Capatina, A., Rancati, E. and Lesca, N. (2016). Exploring
organizational propensity toward inbound–outbound marketing techniques
adoption: The case of pure players and click and mortar companies. Journal of
Business Research, [online] 69(11), pp.5524–5528.
doi:10.1016/j.jbusres.2016.04.165.
9. Bobchenok, O. (2022). How Much Does Google Ads Cost in 2021 and 2022
[online] Available at: https://videnglobe.com/blog/how-much-do-google-ads-cost-
2021-vs-
2022#:~:text=On%20average%2C%20however%2C%20advertisers%20can,click%
20on%20Google%20Ads%20Shopping [27 Oct. 2022].
10. Bonini, J. (2018). The 10 Most Important Google Ads Metrics [Original Data].
[online] Available at: https://databox.com/most-important-google-adwords-metrics
[27 Oct. 2022].
11. Brown, M. (1996). Keeping Score: Using the Right Metrics to Drive World Class
Performance,New York, NY, 1996
12. Dataflair.com (2019). How Google uses Machine Learning to revolutionise the
Internet World? - DataFlair. [online] DataFlair. Available at: https://data-
flair.training/blogs/how-google-uses-machine-learning/ [1 Nov. 2022].

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
268
13. Dean, B. (2019). We Analyzed 4 Million Google Search Results. Here‘s What
We Learned About Organic CTR. [online] Available at:
https://backlinko.com/google-ctr-stats [27 Oct. 2022].
14. Egri, G. and Bayrak, C. (2014). The Role of Search Engine Optimization on
Keeping the User on the Site. Procedia Computer Science, [online] 36, pp.335–
342. doi:10.1016/j.procs.2014.09.102.
15. Fjell, K. (2010). Online advertising: Pay-per-view versus pay-per-click with
market power. Journal of Revenue and Pricing Management, [online] 9(3), pp.198–
203. doi:10.1057/rpm.2010.5.
16. Gales, S. J. (2022). Google Ads Audience Targeting: All 15 Options Explained.
[online] Available at: https://www.wordstream.com/blog/ws/2022/09/21/google-ads-
audience-targeting-cheat-sheet [27 Oct. 2022].
17. Gáll Kristóf (2015) A Google Adwords nagykönyve 2.0. Klikkmarketing Bt.
ISBN: 9786158008600
18. Gáll Kristóf (2016) GOOGLE ADWORDS - hirdetések készítése és
optimalizálása. Klikkmarketing Bt. ISBN: 9786158008648
19. GKID (2022). GKI Digital & Árukereső: 70 millió online vásárlás pörgeti az e-
kereskedelmet. [online] Available at: https://gkid.hu/2022/03/24/70-millio-online-
vasarlas/ [25 Oct. 2022].
20. Google Developers. (2022). Supervised Learning Machine Learning Google
Developers. https://developers.google.com/machine-learning/intro-to-
ml/supervised
21. Google.com. (2022). Google Ads Súgó. [online] Available at:
https://support.google.com/google-ads/?hl=hu#topic=10286612 [27 Oct. 2022].
22. Hill, S. (2019). PPC Marketing Guide | Everything You Need To Know. [online]
Available at: https://creativewebsitemarketing.com/complete-guide-to-ppc-
marketing/ [26 Oct. 2022].
23. Howarth, J. (2021). 57+ Amazing Artificial Intelligence Statistics (2022). [online]
Available at: https://explodingtopics.com/blog/ai-statistics [27 Oct. 2022].
24. Järvinen, J. and Karjaluoto, H. (2015). The use of Web analytics for digital
marketing performance measurement. Industrial Marketing Management, [online]
50, pp.117–127. doi:10.1016/j.indmarman.2015.04.009.
25. Josifovska, M. (2022). PPC Stats - TrueList 2022. [online] Available at:
https://truelist.co/blog/ppc-stats/ [Accessed 1 Nov. 2022].
26. Kim, L. (2022). What Is a Good Conversion Rate? It‘s Higher Than You
Think! [online] Available at: https://www.wordstream.com/blog/ws/2014/03/17/what-
is-a-good-conversion-
rate#:~:text=Conventional%20wisdom%20says%20that%20a,in%20the%20averag
e%20performance%20bucket. [27 Oct. 2022].
27. KSH - Hungarian Central Statistic Office (2020). Digitális gazdaság, 2020.
[online] Available at:
https://www.ksh.hu/docs/hun/xftp/idoszaki/ikt/2020/02/index.html [25 Oct. 2022].
28. Lane, P. (2022). 12 Key Digital Marketing Channels (& How to Use Them).
[online] Available at: https://terakeet.com/blog/digital-marketing-channels/ [26 Oct.
2022].
29. Lincoln, J.E. (2021). What is a Good Click Through Rate? (And How it Can
Help Your Business). [online] Available at: https://ignitevisibility.com/what-is-a-
good-click-through-

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
269
rate/#:~:text=The%20average%20CTR%20for%20display,campaigns%2C%20it‘s
%20to%20be%20expected. [27 Oct. 2022].
30. Lindley, A. (2022). 7 Digital Marketing Channels: What Works for Your
Users? [online] Available at: https://www.springboard.com/blog/business-and-
marketing/digital-marketing-channels/ [26 Oct. 2022].
31. Lee, J.-H., Kim, D.-H., Jeong, S.-N. and Choi, S.-H. (2018). Detection and
diagnosis of dental caries using a deep learning-based convolutional neural
network algorithm. Journal of Dentistry, [online] 77, pp.106–111.
doi:10.1016/j.jdent.2018.07.015.
32. McCandless, M. (1998) Web advertising IEEE Intelligent Systems and their
Applications, 13 (3) (1998), pp. 8-9
33. Porter, M. E. (2001) Strategy and the internet. Harvard Business
Review, 79 (3) (2001), pp. 63-78
34. Mangani, A. (2014) Online advertising: Pay-per-view versus pay-per-click
Journal of Revenue and Pricing Management, 2 (4) (2004), pp. 295-302
35. Marr, B. (2021). The 10 Best Examples Of How Companies Use Artificial
Intelligence In Practice [online] Available at: https://bernardmarr.com/the-10-best-
examples-of-how-companies-use-artificial-intelligence-in-practice/ [27 Oct. 2022].
36. Mehta A, Saberi A, Vazirani U, and Vazirani V (2007) AdWords and
Generalized On-line Matching. ACM Transactions on Computational Logic 5: 1-20.
37. Murphy, A.L., Peltekian, S. and Gardner, D.M. (2018). Website Analytics of a
Google Ads Campaign for a Men‘s Mental Health Website: Comparative Analysis.
JMIR Mental Health, [online] 5(4), p.e12428. doi:10.2196/12428.
38. Richardson, M., Dominowska, E., & Ragno, R. (2007). Predicting clicks:
estimating the click-through rate for new ads. In Proceedings of the 16th
international conference on World Wide Web (WWW '07). Association for
Computing Machinery, New York, NY, USA, 521–530.
https://doi.org/10.1145/1242572.1242643
39. Robinson, H., Wysocka, A., & Hand, C. (2007). Internet advertising
effectiveness: the effect of design on click-through rates for banner ads.
International Journal of Advertising, 26(4), 527–541.
40. Shirey, T. (2022). Which Marketing Channels Produce the Highest
ROI? [online] Available at: https://www.webfx.com/blog/marketing/which-
marketing-channels-produce-the-highest-
roi/#:~:text=Why%20is%20SEO%20one%20of,SEO%20provide%20an%20impres
sive%20ROI. [25 Oct. 2022].
41. Siddiqui, O. (2020). 5 Most Effective Digital Marketing Channels for Your
Business | CMA. [online] Available at: https://cmasolutions.com/blog/5-most-
effective-digital-marketing-channels-for-your-business/ [26 Oct. 2022].
42. Stephens, M. (2022). History of Television - Mitchell Stephens. [online]
Available at:
https://stephens.hosting.nyu.edu/History%20of%20Television%20page.html#:~:text
=Electronic%20television%20was%20first%20successfully,electricity%20until%20h
e%20was%2014. [1 Nov. 2022].
43. Valve and Meter. (2018). 2020 PPC Statistics - A Comprehensive List of Pay-
Per-Click Stats. [online] Available at: https://valveandmeter.com/pay-per-click-
statistics/ [27 Oct. 2022].

The Annals of the University of Oradea, Economic Sciences
TOM XXXI, 2nd Issue, December 2022
270
44. Vidhoarder.com. (2022). The Ultimate List Of Google Ads Targeting (Complete
List Of Audiences, Topics, and More!) – Vid Hoarder Blog. [online] Available at:
https://blog.vidhoarder.com/the-ultimate-list-of-google-ads-targeting/ [1 Nov. 2022].
45. Yang, Y. and Zhai, P. (2022). Click-through rate prediction in online advertising:
A literature review. Information Processing & Management, [online] 59(2),
p.102853. doi:10.1016/j.ipm.2021.102853.
46. Za, Z. S., Tricahyadinata, I. (2017) An Analysis on the Use of Google AdWords
to Increase E-Commerce Sales. International Journal of Social Sciences and
Management 4(1):60. pp 1. ISSN: 2091-2986