ArticlePDF Available

Towards the Adoption of Machine Learning-Based Analytical Tools in Digital Marketing

Article

Towards the Adoption of Machine Learning-Based Analytical Tools in Digital Marketing

Abstract

Exponential technological expansion creates opportunities for competitive advantage by applying new data-oriented approaches to digital marketing practices. Machine learning (ML) can predict future developments and support decision-making by extracting insights from large amounts of generated data. This functionality greatly impacts and streamlines the strategic decision-making process of organizations. The research gap analysis revealed that little is known about marketers’ attitude towards, and knowledge about, ML tools and their adoption and utilization to support strategic and operational management. The research presented here focuses on the selection and adoption of ML-driven analytical tools by three distinct groups, namely marketing agencies, media companies, and advertisers. Qualitative and quantitative research was conducted, on a sample of these organizations operating in Slovakia. The findings highlight 1) the important role of intelligent analytical tools in the creation and deployment of marketing strategies, 2) the lack of knowledge about emerging technologies such as ML and artificial intelligence (AI), 3) the potential application of ML tools in marketing, as well as 4) the low level of adoption and utilization of ML-driven analytical tools in marketing management. A framework consisting of enablers and a process map was developed to help organizations identify the opportunities and successfully execute projects oriented towards the deployment and adoption of analytical ML tools in digital marketing.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.Doi Number
Towards the adoption of machine learning-
based analytical tools in digital marketing
Andrej Miklosik1, Martin Kuchta1, Nina Evans2, and Stefan Zak1
1Marketing Department, Faculty of Commerce, University of Economics in Bratislava, Bratislava, 85235 Slovakia
2School of Information Technology and Mathematical Sciences, University of South Australia, Adelaide, 5001 South Australia
Corresponding author: Andrej Miklosik (e-mail: andrej.miklosik@euba.sk)
This paper is an output of the research project VEGA (S.G.A.) 1/0657/19 The role of influencers in the consumer decision-making process.
ABSTRACT Exponential technological expansion creates opportunities for competitive advantage by applying new data-
oriented approaches to digital marketing practices. Machine learning (ML) can predict future developments and support
decision-making by extracting insights from large amounts of generated data. This functionality greatly impacts and
streamlines the strategic decision-making process of organizations. The research gap analysis revealed that little is known
about marketers’ attitude towards, and knowledge about, ML tools and their adoption and utilization to support strategic and
operational management. The research presented here focuses on the selection and adoption of ML-driven analytical tools by
three distinct groups, namely marketing agencies, media companies, and advertisers. Qualitative and quantitative research
was conducted, on a sample of these organizations operating in Slovakia. The findings highlight 1) the important role of
intelligent analytical tools in the creation and deployment of marketing strategies, 2) the lack of knowledge about emerging
technologies such as ML and artificial intelligence (AI), 3) the potential application of ML tools in marketing, as well as 4)
the low level of adoption and utilization of ML-driven analytical tools in marketing management. A framework consisting of
enablers and a process map was developed to help organizations identify the opportunities and successfully execute projects
oriented towards the deployment and adoption of analytical ML tools in digital marketing.
INDEX TERMS Big data, data-driven analytical tools, digital marketing, machine learning (ML), marketing agencies,
marketing analysis
I. INTRODUCTION
In recent years, the extensive development of information
and communication technologies in the private and public
sectors initiated the emergence of a new digital marketing
environment. Due to the proliferation of information
technology, a huge amount of data is currently generated. It
is estimated that 2.5 quintillion bytes of data are created
every day and this number increases with the onset of the
Internet of Things (IoT) [1]. It is also estimated that 90% of
the available global data has been generated in the past two
years [2]. Timely and precise business decisions depend on
the generation, access, and utilization of quality information.
Exponential technological expansion and its barrier-free
global dissemination therefore create opportunities to gain
competitive advantage by applying new data-oriented
approaches to marketing management [3].
Digital marketing emerged as a natural response by
companies to leverage and benefit from the significant
consumer concentration on the Internet. Various types of
organizations, including businesses, hospitals, schools,
professional associations, councils and NGOs, use digital
marketing as part of their marketing strategies and
deployment programs. Some of these organizations can also
operate their own e-commerce platform, but they mostly use
the Internet as a channel/medium within their communication
strategy. These organizations typically fulfil the role of
clients or advertisers also referred to as brands. Other
categories of organizations also operate in the digital
marketing space. Digital agencies create and implement
marketing strategies for the organizations in the first group
and use digital marketing as part of their own marketing
strategy. Organizations in the third group, namely media, are
used by digital agencies (or the advertisers directly) to
communicate with their target audience.
The Internet environment enables companies to learn more
about consumers through a few clicks in the appropriate
analytical tool. The greatest advantage of digital marketing
over other marketing tools and channels is its measurability.
The digital footprint of every Internet user contains a
significant amount of data that can serve as input for
marketing analysis. Manual acquisition and analysis of the
data has been time-consuming and only marginally regulated
[4]. Analytical tools are currently used in marketing
management to systemize processes, streamline the decision-
making, and automate work. These sophisticated analytical
tools use machine learning (ML) to learn from historical data
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
and help plan future activities more effectively [5]. The ML
tools can be utilized in many sectors. The research presented
in this article focuses specifically on their use in marketing
analysis towards improving strategic and operational
decisions in marketing management. Using an
interdisciplinary approach, the research explores the potential
of ML in marketing analytics, the degree of implementation
of the technology, as well as the attitudes of marketing
agencies and marketing managers towards active utilization
of these tools.
II. LITERATURE OVERVIEW
IMPORTANCE OF DATA ANALYSIS IN STRATEGIC
PLANNING
Analytical tools streamline strategic planning and help
organizations make operational decisions faster and more
efficiently. In the past, strategic planning was usually based
on data and information that were gathered in the
organization over the past periods. In the current dynamic
age, it is important to regard strategic planning as a live,
ever-changing process. Whether it is a process of analyzing
past events or predicting future events, strategic planning is
supported by advanced marketing analysis and analytical
tools utilizing recent IT innovations. The strategic planning
process that uses IT tools must include at least the following
four phases: 1) planning; 2) presenting a strategic idea; 3)
fine-tuning the strategic idea; 4) developing a strategic plan
[6]. The first phase is followed by two additional stages,
namely the 1) execution phase that transforms the strategic
plans into practical activities and the 2) information support
phase that provides information about the current and future
organization status [7]. Large quantities of information are
required in these phases to create a comprehensive overview
of the examined area and incorporate facts that may
negatively affect the planning process.
Evaluation of the effective implementation of marketing
strategies also forms part of companies strategic
management processes. Many organizations neglect this
phase or completely ignore it, which can lead to failures in
future strategic planning. The evaluation of the implemented
strategic plan helps reform the decision-making process,
identify deficiencies at all stages of the strategic planning and
indicate steps in future planning [8]. In today's technology
era data-based mechanisms should be used to streamline the
strategic process, increase the chances of a successful, result-
orientated strategy implementation and enhance
organizational effectiveness [9].
On the basis of sufficient data volumes and large
computing power, current analytical tools can predict e.g.
consumer buying behavior, the response to product
introduction and the probable success of a developed
strategic plan [10]. The processing of big data present
opportunities and transformational potential for many
industries. As the volume of data increases, deep learning
becomes meaningful, allowing the utilization of predictive
analytical solutions [11]. Analytical tools of this type can be
created to learn and improve. Data sets that form the input for
marketing analysis can be prepared promptly, thereby saving
weeks, months or sometimes years of individuals manual
efforts. Current marketing has entered the era of AI and ML
that greatly streamline the strategic process of organizations
and support strategic decision-making. Linking marketing
analysis and advanced analytical tools opens the door to a
new world where we can "breathe life" into machines and
programs by teaching them how to learn and thereby make
life easier for people. AI has the potential to be used in all
areas of our lives and its current application is merely the
beginning.
Selecting the right analytical tool that will meet the
requirements of data analysis is vital to ensure return on
investment and the added value of the generated information.
When developing the best marketing strategy the attention of
responsible staff should be shifted from traditional analytical
tools to advanced and specialized analytical tools [12]. This
is especially important in marketing, where hundreds of
potentially relevant metrics need to be considered. Analytical
tools in marketing have the potential to: 1) access the data as
strategic assets; 2) visualize data into clear structures; 3)
provide an overview of existing and potential customers; 4)
increase the effectiveness of marketing decisions; 5) focus on
proactivity towards consumers; 6) create tailor-made offers
to specific consumers; 7) adapt the digital environment to the
preferences of specific users; 8) engage in real-time
discussions with consumers; 9) increase the effectiveness of
marketing activities (including online and offline
communication); and 10) focus on indicators of company
success [13].
ML-BASED ANALYTICAL TOOLS IN MARKETING
ML includes adaptive mechanisms that allow a computer or
machine to learn, based on experience and examples.
Learning new skills will increase the system performance
over time. ML mechanisms form the basis for adaptive
systems. The technological capabilities of current
information systems include computational capacity for
performing demanding statistical mathematical methods. For
this reason, the increase in software development projects
embracing ML principles leads to higher numbers of
intelligent analytical systems. Their use does not only save
companies considerable costs in the long run, but it also
makes life easier for consumers. In the marketing sector ML
has huge potential to be utilized in decision-making,
interaction with customers, and strategic planning.
By applying neural networks, a specific problem that
requires thinking - human or artificial is solved. ML can be
perceived as superficial data processing that deals only with
one layer of acquired information [14]. In-depth learning,
which is part of ML, handles several layers of data
simultaneously. Thus, solutions embracing in-depth learning
can obtain new information from which they create
additional layers to be processed subsequently. Based on this
approach, predictive analysis becomes possible. The
application of in-depth learning can reveal relationships
between variables that were hidden from ML [15]. Thus, in-
depth learning contributes to a more efficient decision-
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
making process with fascinating implications for theory and
business practice. In-depth learning works on the principle
that the system is fed with data, the data are processed, and
the system also considers other data in its decision. At
present, autonomous vehicles serve as an illustration. The
system is provided with initial data to process, yet when its
practical tests and usage started, more data are accumulated,
allowing objects to be recognized and responses developed.
Thus, in addition to the original settings, the system collects
the data it uses when deciding on the other data collected
[16].
The relationship between price and demand in the housing
market by looking at online advertisements has been
examined [17]. ML is used to identify pairs of duplicate
advertisements that refer to the same housing unit. An
artificially intelligent chatbot has been developed and
deployed to deliver highly accurate automatic answers by
using a knowledge base and supporting the 24/7 service
availability of customer service centers (contact points) [18].
The chatbot increases the efficacy of support services and
automates the related process in the digital environment. ML
models are developed to predict the degree of visibility of
corporate websites [19]. A trending hashtag generating
application for social media business users is developed
using ML and other approaches, generating trending and
relevant hashtags for user content in order to get a broad
reach of target audience [20]. A methodology using ML to
accurately detect the marketing and sale of opioids by illicit
online sellers via Twitter has been developed and deployed
[21]. In another study, data mining, text mining, ML and
statistics is used to analyze website user behavior and
introduce two models of behavior based on the sequence and
number of URLs accessed [22]. A quantitative study
exploring how organizations in Dubai use big data and data
driven marketing to know and serve their customers better
and enhance shareholder value has been accomplished [23].
The strength of ML and big data in predicting the
occurrences of consumer changing their mobile phones has
been demonstrated, which greatly impacts marketing
strategies in the mobile communications industry [24]. Real-
Time Bidding (RTB) as an approach in display advertising
building automation, integration, and optimization has been
analyzed [25]. A methodology to accurately identify tweets
marketing the illegal online sale of controlled substances has
been developed [26]. A trained ML model is used to help
online marketers understand the popularity evolution of
online information, considering the "burst", "peak", and
"fade" key events together as a representative summary of
popularity evolution [27]. Methods are proposed that can
effectively extract information about the intent of users from
online texts with significantly high accuracy [28].
Understanding the intent of users (e.g. to buy an apartment,
rent a car, or travel somewhere) on online social media
channels has great potential in digital marketing. ML models
based on conditional random fields (CRFs), an advanced
statistical graphical model for sequence data, and
bidirectional long short-term memory (Bi-LSTM), a well-
known deep learning model, have been developed.
DEPLOYING ML-DRIVEN ANALYTICAL TOOLS IN
DIGITAL MARKETING
Digital marketing is widely considered as an umbrella term,
including online marketing, Internet marketing and mobile
marketing. [29, 30]. It can be defined as marketing that
utilizes digital technologies (hardware, software,
communication technologies) for the deployment of
marketing strategies [31, 32]. The tools utilized in digital
marketing include market research, polling, various forms of
advertising, search engine marketing, newsletters and social
media marketing. Marketing analytics is an inherent part of
effectively using any of these tools. All three types of
organizations (advertisers, agencies, media) need the
awareness and ability to work with large amounts of data to
extract meaningful information and increase the effectiveness
of their digital marketing initiatives.
It can be concluded that ML can, based on extensive data
processing, provide the information necessary for the
decision-making process of marketing specialists. The
application of ML-driven tools into digital marketing
introduces various new challenges and opportunities. Among
the biggest benefits of using these tools in marketing are:
Optimal performance machines perform steadily at
100% because they cannot be disturbed or distracted.
Faster decision making machine decision-making time
is determined by the available data. After a quick
calculation, machines can decide (almost) immediately.
The decision is not affected by subjective factors such as
feelings, personal preferences, opinions, etc.
Automatization of predictable activities ML can very
effectively automate routine activities. For example, in
digital marketing, ML can assist with generation of
regular reports on advertising campaigns in social media
marketing.
Reducing error rates eliminating errors normally caused
by human factors. Machines perform a task by always
following the predefined procedure.
Digital assistants personal human assistants are already
commonly used. ML-driven systems can handle
complicated tasks and optimize daily routines.
Exploring areas unavailable to humans in many areas a
person is unable to perform the required tasks for various
reasons. For example, humans can’t dive into the deepest
areas of oceans or process large amounts of data that are
generated every minute on the Internet. Machines can be
adapted to almost any condition, and computing can
handle even the most difficult mathematical-statistical
operations.
Implementation of ML applied to marketing analytical
tools has no obvious disadvantages. However, depending on
the type of tool, there are several concerns or limiting factors
for its effective use:
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
Creativity much of the decisions and application of
marketing is based on creativity. Creativity results in
combining several parameters that are native to humans,
with emotions, intuition, and empathy playing a
significant part. These three components will be very
difficult to replicate by machines.
Moral and ethical principles it is the ability to make
decisions, express emotions, make sense of ethics and
morals that makes people human. It is questionable
whether machines will never have this ability, or it could
be possible developed in later stages. There are justified
concerns about whether machines will still be performing
in the interest of people and within ethical and moral
boundaries when the progress continues.
Consumer preferences users do not always want to
interact with a robot or software. It is thus expected that,
in the near future, the need for interaction with a human
being will remain preferential, with consumers having
greater confidence in such communication.
ML tools cannot work without human intervention in
the upcoming decades, human minds will still be needed
to work with ML systems and develop and optimize this
technology.
Algorithms can be wrong decisions based on MLs’
mathematical calculations and statistics can possibly
result in incorrect actions taken, because of faculty
instructions or data. Any advanced information systems
are intended primarily to serve people. As peoples’
requirements change and evolve over time, it can result in
the ML-based tools not providing the optimal results in
the current configuration. Automated responses to social
media requests or automatically generated discounts in
case of unhappiness with the purchasing process can
serve as examples [33].
ML and employment there are systems that can
partially replace human workforce in digital marketing.
For example, automatic uploading of posts on social
media, automated article reader support, etc. However, it
is important to note that these systems will never perform
properly without human support [34].
III. RESEARCH GAP AND AIMS
Previous research in the fields of ML and AI focused on how
ML-driven systems and applications can be improved to help
address specific questions, resolve issues, support decision-
making, understand target markets, predict consumer
behavior, increase the efficiency of marketing
communications, support the launch of new products,
streamline digital marketing-related internal processes or
improve team management. However, the research failed to
investigate and discuss the issues of selection and adoption of
these ML-driven analytical tools by stakeholders, including
marketing agencies and other marketing specialists. Very
little is known about marketing professionals’ knowledge of,
and attitude towards these tools, as well as the adoption and
utilization of the tools for both strategic and operational
management. This research gap forms the foundation of the
research described here.
A deeper study of relevant resources and work with
secondary data revealed that numerous information systems
that embrace the principles of ML are available to marketing
managers and agencies. However, these businesses and
individuals have been slow in adopting and utilizing these
systems. This research focuses on identifying the reasons for
the slow application of ML in digital marketing and
suggesting possible solutions. The main aim of the research
is to identify possible applications of ML analytical tools in
the marketing field. Secondary objectives are to 1)
investigate the role of analytical tools in the process of
preparing, implementing, and evaluating digital marketing
strategies; 2) determine how selected digital marketing
entities use marketing analysis and analytical tools; 3)
identify the level of awareness of relevant terminology such
as big data, data management, ML or AI; and 4) find out how
ML-based tools are used in the digital marketing
environment.
IV. METHODOLOGY
RESEARCH TYPE AND SAMPLE SELECTION
Qualitative research, using the method of in-depth
interviews, was used to gain insight into the way ML is
practically used in marketing. Digital marketing specialists
representing 1) firms in the role of advertisers; 2) agencies in
the role of intermediaries; and 3) businesses providing the
media space in Slovakia, were selected as interview
participants, based on the following key criteria: The
interviewee is a middle or senior manager, has been working
in digital marketing for at least five years and has direct
experience with information systems in the field of Pay Per
Click (PPC), Search Engine Optimization (SEO) or Real-
Time Bidding (RTB) campaign development and
optimization. Three experts from each of these three groups
were interviewed, i.e. a total of nine respondents. The
findings from the in-depth interviews indicate the current
situation in the area of using ML-driven analytical tools by
organizations that are active in the field of digital marketing.
Quantitative research, using a standardized questionnaire,
was conducted to confirm the findings and gain more insights
into the identified issues.
IN-DEPTH INTERVIEWS
The process of in-depth interviews consisted of three steps:
1) preparation; 2) implementation; 3) data processing. The
preparation stage included defining the interview structure,
formulating initial instructions for respondents, preparing
research questions, providing audio recording techniques,
and selecting a suitable environment for the interviews. The
estimated length of the interview was one hour. Prior to each
interview, the respondent was asked to sign a consent form.
The interviews were based on a pre-defined structure and
moderated. However, respondents were given the
opportunity to suggest additional topics and make comments
that stem from the flow of the conversation. Before the
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
interviews, participants were requested to express their
feelings and beliefs and to provide only true information.
They were given the assurance that there are no wrong
answers, that answers are considered confidential, that
responses will not be linked to individuals, and that the audio
recording of the interview and the written notes taken by the
researcher will be used solely for research purposes.
Interviewees were also informed that the findings of the
research may be presented in both written and oral form.
In addition to the initial and final instructions, the
interview protocol contained 19 open questions. In several
cases the conversation turned to areas that were not listed in
the prepared scenario, thereby identifying new opportunities
and risks associated with the possibilities of marketing
analysis and ML-driven analytical tools. The questions were
divided into the following four headings:
1) General information: the purpose of this part was to
confirm that the respondents met the selection criteria as
specified in the methodology and define the strategic
process of their marketing activities.
2) Marketing analysis and analytical tools: This section
examined the perception and actual use of marketing
analysis and analytical tools in practice.
3) Awareness level regarding big data and data
management: Questions in this area examined the level
of awareness and assumptions associated with big data
and their utilization in marketing.
4) ML and AI practical application: Respondents expressed
their knowledge and opinions on the challenges of ML
technology and the application possibilities in the field of
digital marketing.
QUESTIONNAIRES
The results of the interviews were used as basis for
developing a questionnaire to verify the results on a larger
sample. The questionnaire contained 15 questions: 11 closed-
ended questions, 3 rating scale questions, and 1 open-ended
question. To reach the specific group of marketing
specialists, the questionnaire was distributed online via four
different channels. These included: a) IAB Slovakia, the
regulatory body for digital marketing in Slovakia; b) Digital
agency SamsiDigital; c) Personal Facebook profile of a
digital marketing blogger; d) Facebook group Marketers,
copywriters and SEO optimizers. A total of 58 responses
from digital marketing specialists and C-level executives
from companies involved in digital marketing were analysed
using descriptive statistics.
The in-depth interviews took place between 15th June and
30th July 2018, followed by the quantitative research. The
questionnaire was prepared between 06 August and 15
August 2018 and afterwards distributed using the
abovementioned four channels. The results were collected
between 21 August and 10 September 2018. After the
completion of data collection, data were processed (finalized
on 14 October 2018), resulting in evaluation of the findings
and preparation of the research outputs.
V. FINDINGS
THE PROCESS OF PREPARATION, IMPLEMENTATION
AND EVALUATION OF DIGITAL MARKETING
STRATEGIES
Based on the information obtained, a model of the
preparation process, the implementation and the evaluation
of digital marketing strategies (Fig. 1) was developed.
FIGURE 1. The process of preparation, implementation, and evaluation of digital
marketing strategies
Agency, media, and advertisers workers are largely
following the abovementioned nine-step model. However,
different steps were emphasized by different respondents.
Based on their responses, the following steps are considered
as most important for different organizations:
Agencies: Steps 3, 5, 7, 8, and 9 are the most crucial. In
the process of marketing communication, agencies are the
executors of client assignments (in this case, advertisers),
so the steps to prepare, implement, optimize, and evaluate
marketing campaigns are most important for them.
Media: They draw attention to steps 7, 8, and 9 and
provide media spaces throughout the whole process.
Their main interest is to achieve good tracking results
(such as the number of banner advertisements, mouse
actions, etc.). Therefore, the steps of implementing,
optimizing, and evaluating campaigns are most important
for the media. If the campaign achieves good results, the
probability of future orders from the agency or advertiser
increases.
Advertisers: Steps 1, 2, 4, and 9 were mostly discussed. A
specific brand product must clearly define what it aims to
achieve with a marketing campaign and correctly
interpret its goals to the agency or media that are the
executor of the communication process. Another
important step for the advertiser is the evaluation of the
campaign, providing valuable information about future
marketing communication strategies.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
The follow-up quantitative research confirmed that
marketing analysis and work with software analytical tools
significantly affect the process of preparing marketing
strategies. The five most decisive factors for developing
marketing strategies are: 1) know-how from previous
campaigns (85%); 2) marketing analysis (studies and
available data) (79%), analytical software tools (64%),
subjective decision-making 28%, personal opinions and
experience (26%). It is clear that data and tools help to
extract meaningful information from respondents and are
perceived as fundamental in the process of developing
marketing strategies.
PERCEPTION OF MARKETING ANALYSIS AND
ANALYTICAL TOOLS
In the second set of questions in the in-depth interviews, the
intention was to find out how respondents perceive
marketing analyses and analytical tools and how they
implement these tools in digital marketing processes in their
company.
The first question focused on the way respondents use
marketing analysis in the process of planning and
implementing digital marketing strategies. The following
comments illustrate the situation in this area:
Respondent working in the media: We use marketing
analysis regularly. It is a thing without which we will not
move. This means that the analysis is probably the most
important part before we decide what further steps we
will take.”
Respondent working in the agency: “Marketing analysis
forms a substantial part of the whole process. They give
us valuable impulses for creating a marketing strategy
and think about what to do and how to do it.”
Respondent working for the advertiser: "All agency
assignments we work with are based on our analysis and
data."
The next question in the second section focused on
respondents expectations of marketing analysis at any step
in the preparation and implementation of a marketing
strategy. The most common answers were:
information about marketing campaign target group,
segment size detection,
information about competitor marketing activities,
current trends in digital marketing in the local market,
detection of relevant keyword search volume,
information about potential marketing communication
intervention,
prediction of frequency of displaying the desired
marketing message to the user,
demographic information about the target group, and
identification of situations where purchases are most
common.
The findings illustrated that marketing analysis serves as a
source of information and are used by respondents for all
important areas of preparing and implementing digital
marketing strategies.
This was followed by the question: "What are the sources
of this information?" The purpose of this question was to
check where the respondents are searching and accessing the
necessary information. The answers shown in Table 1 were
mostly repeated:
TABLE 1
OVERVIEW OF THE MOST COMMONLY USED ANALYTICAL
TOOLS IN DIGITAL MARKETING
Advertiser
Agency
Media
Freely available
surveys carried out by
research agencies
Paid surveys conducted
by research agencies
Historical data
generated in previous
marketing campaigns
Internal analytical tools
Internal quantitative
surveys
Internal sales and
customer data
Information provided
by partners
The term "source of marketing information" can be
regarded as any information about the market, segment,
product, target group or competition that has been obtained
from internal sources and from paid or unpaid third-party
resources. It is important to realize that, despite the source of
this information, it is usually necessary to work with
analytical tools that help in the process of collecting,
segmenting, evaluating and interpreting the collected data.
When asked the next survey question: "Do you use analytical
tools in the process of creating marketing strategies?" all
respondents answered yes. Any activity on the Internet is
measurable through digital technologies and the collected
results are used to make more effective decisions in future
activities. Digital marketing professionals are well-aware of
this and they use measurability as one of the most important
competitive advantages of their company.
The various analytical tools available on the market can be
integrated into the preparation, implementation and
evaluation of digital marketing campaigns. The choice of the
specific tool is subject to business needs. The question:
"Please indicate what tools you use and what purpose do they
serve?" sparked discussions about the most commonly used
analytical tools and confirmed that the marketing agencies
focus on execution, while media companies are focused on
evaluation, and advertiser are focused on assigning digital
marketing strategies. Table 2 lists the declared analytical
tools, a short specification, and the entity that uses the tool.
The tools are ranked by frequency of occurrence.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
TABLE 2
OVERVIEW OF THE MOST COMMONLY USED ANALYTICAL
TOOLS IN DIGITAL MARKETING
Specification
Agency
Media
Advertiser
An analytical tool that monitors
users’ activities on website.
Visualizes collected data into
graphs and infographics.
Advertising system for banner
advertising and search engine
advertising.
Provides information about using
mobile and web applications.
A web browser that provides
information about SEO settings.
Facebook social network
advertising system.
An internal analytical tool that
includes profile and user
information.
Enables comprehensive
management of social networks,
including competition
monitoring.
It makes it easier to work with
content that is designed for social
networks.
A tool designed to automate
marketing activities in the digital
environment.
Optimization of digital products
through user testing.
Analytical tool for SEO
optimization and competitor
information.
Real-time monitoring of user
behavior on a website or in an
application.
Creating heat maps by moving a
user's mouse on a web page.
A tool for automation, based on
AI, aimed at achieving specific
goals.
Provides user behavior
information across multiple
platforms.
Comprehensive web site analysis
to optimize user environment.
Gathers and segments
information about receiving
Internet content by specified
target groups.
Gemius
Direct Effect
Monitors the performance of
ongoing campaigns placed in the
Internet environment.
Mautic
Software designed primarily for
process automation, email
campaign management, goal-
oriented optimization, etc.
Mailchimp
A marketing tool designed for
small businesses, especially to
manage email campaigns.
AIM
monitor
A freely available tool that
monitors traffic and other
parameters for each website listed
in the dashboard.
Internal data
Each online activity generates a
lot of data that are stored for
future use.
Third party
data
Advertisers use agency data,
agencies use generic or custom
results of surveys, media data,
etc.
The list contains analytical tools that are most frequently
used by respondents. Many of the mentioned tools are
universal and their use was confirmed by respondents from
all three entities (agency, media, advertiser). Each respondent
regularly uses an average of 16 analytical tools with
analytical capabilities that cover almost all areas of digital
marketing. Respondents from all three entities said that 60%
of the process of creating and implementing a digital
marketing strategy is done by analytical tools and data, while
40% is done through activities such as brainstorming,
creating advertisements, creating visuals, etc. We assume that
various other tools are used infrequently, or the interviewee
did not mention them due to the limited time dedicated to the
interview.
The importance of data analysis in marketing was
confirmed on the larger sample. One of the questions in the
questionnaire determined whether marketing managers
perceive digital marketing as a science or an art. 40% of
respondents selected the first option, 19% selected the
second, 41% were neutral, recognizing that both these
definitions are true and in balance. Another question asked
how data are collected and processed in companies. Only 9%
of companies do not collect relevant data on customers and
markets but they are planning to start doing so. 71% of
companies are collecting data internally. Of these, 16% uses
third-party to assist them with their processing and analysis.
19% of companies use a third-party partner to also collect the
data on their behalf. 2% of the respondents were unsure how
data are managed in their companies. Marketing managers
are fully aware that, in todays environment and markets,
data play a crucial role in digital marketing.
LEVEL OF AWARENESS OF BIG DATA AND ML
The in-depth interviews identified respondents perception
regarding terms such as big data and data management and
how they perceive these terms. Some of the comments were:
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
Big data
Respondent 1: "I see the big data as a huge amount of
data from different segments of the online environment
that companies should collect and evaluate internally
because that is the future of doing business in the Internet
environment."
Respondent 2: "They are extensively large data that is
raw, not segmented in any way, not processed in any
way."
Respondent 3: “Every company has some amount of data
or has some data available in some form. Big data is an
industry that processes these extreme amounts of data,
and within this software available data are visualized.”
Respondent 4: “A large number of Excel spreadsheets
with plenty of data. High volumes of data are being
collected nowadays, but when people do not know how to
use them, it is better to collect them less, but to know the
purpose of their use.
Respondent 5: Processing large amounts of data using
different databases. Big data are a very good start and its
processing and understanding will be used to create AI as
one of the approaches to digital marketing.”
The four other interviewees completely or partially agreed
with these answers. All the respondents have heard of, and
understand, the concept of big data. They were all able to
define the term in their own words, i.e. as a large amount of
information and data that are thus far unprocessed. The next
step is to segment and “translate the big data into a form
understandable for people. All respondents agreed that
exploiting big data creates competitive advantage and that
data are the future of digital marketing and many other
industries.
Respondents were also asked how they understand the
term data management that is closely linked to big data.
Some of the comments were:
Data management
Respondent 1: "Data management is a tool to manage
these data."
Respondent 2: “It is through data management that these
data can be processed and converted to their advantage.
Because having data is great, everyone can have it, but
the question is how a person can transform those big data
into something specific. How to use them, whether they
can use them to target, as our company do, or use them
for a look-and-like targeting, or for negative targeting, or
can accurately identify that a particular cookie has a
particular product but doesn’t have its superstructure, etc."
Respondent 3: “You have a platform where the data are
stored. Data should be given meaning, logic and
evaluated accordingly. This is supported by software that
work on data management principles.”
Respondent 4: “Finding a connection between data. Their
correct categorization. Big data are a refrigerator and data
management means that you know exactly what to
choose from the refrigerator when you want to cook
something.”
Respondent 5: "Data management is a part of working
with big data."
The responses of the four other respondents partially or
completely supported these comments. The collected
information reveals that respondents some understanding of
data management, but they are much less confident with big
data. About half of respondents perceive data management as
a tool for processing big data. Others either confused the
concepts of big data and data management, considered their
substance to be the same, or were not able to comment on the
term as they do not understand it. All respondents agreed that
nowadays, one should pay attention to data management.
However, many admitted that additional education is needed
because to fully embrace the opportunities requires
significant knowledge.
When asked how respondents and their companies deal
with data, most respondents indicated that they use their own
analytical tools or third-party data. Only one of the
interviewees reported no activity in this area. All respondents
regard data as a strategic asset that creates competitive
advantage for the company. Respondents who use analytical
tools to retrieve data, primarily get the information from sales
reports and historical marketing campaigns. Some companies
acquire data from third parties, including business partners
and freely available databases.
Although it was not the primary purpose of the question,
the interviews revealed that agencies rely on third party data,
media companies rely primarily on their own data, and more
progressive advertisers use a combination of their own data
and data from third parties. A less progressive advertiser does
not work with data yet, although it sees great potential for
such activity in future.
The question also elicited interviewees’ opinions about the
biggest obstacles to using data for digital marketing
purposes:
High costs: Seven out of nine (78%) respondents consider
the entry costs of developing and implementing platform
data management to be extremely high. Only an
international company with high turnover and above-
average budget can afford such a step in the Slovak
market. All respondents agreed that, irrespective of the
input costs, such an investment has a relatively short
payback period. Respondents also expressed the view that
it is easier to build a data-oriented business from the start,
than to incorporate data solutions into existing, often
rigid, structures.
Time consumption: Most respondents have had
experience of introducing new systems into their current
company, or a previous employer. All respondents
(100%) agreed that the process of implementing new
approaches is time-consuming and often exceeds a pre-
agreed timeframe. Respondents are concerned about the
time-consuming implementation of data-oriented
approaches, based on their previous experience.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
High pressure on know-how: Digital marketing is a
relatively young industry that requires a lot of new
information, thus putting pressure on education and
training of employees. Data-based approaches are an
even younger industry that is embedded in the
technology-digital world. New professionals in this
industry are therefore required to provide a link between
the technology, mathematical-statistical, and marketing
worlds. This combination of skills is relatively scarce,
and people with such experience are costly to the
company because of their high intellectual value. Another
solution may be to educate internal human resources,
which in turn requires time and costs. Once the
employees are trained the company may lose them to a
competing company, due to high labor market
fluctuations.
Limited Slovak market: Respondents consider the size
and possibilities of the local market as one of the
problems in the implementation of data access. There are
agencies in the world exclusively dedicated to analytical
work and data processing. These agencies act as third
parties that provide data processing services to their
clients or they sell directly collected and processed data
packages. According to the interviewees, the existence of
such a company is unrealistic in Slovakia, because the
digital marketing entities do not currently have the
required funds to cover the costs of employing such a
data company. Another problem is the size of the Slovak
population, the penetration of the Internet and the
additional costs associated with constantly updating the
data collected, of which the timeliness is often linked to a
cookie that only keeps track of information from the last
30 days.
The in-depth interview questions explored the current use
of marketing analysis and analytical tools. While talking
about analytical tools, the respondents directed the
conversation to process automation, ML, and AI, as applied
to digital marketing. Another interview question was: “What
do you imagine under the term AI? Try to define the
difference between AI and ML.” Especially when asking this
question, respondents were reminded that the question
examines their opinion and beliefs, regardless of their level
of education. While in previous questions the respondents
were relatively certain of their answers, they now needed
more time to formulate their answers and their answers were
unclear and inconsistent. This is illustrated in the following
responses:
Artificial intelligence
Respondent 1: "A machine, a mechanism that thinks. It
has a processor and does something.
Respondent 2: "AI is something that is programmed by a
person and does something that is based on some rules
that the man sets."
Respondent 3: AI is a superstructure, it's a Mercedes
among ML. It can learn not only from case studies that
someone gives it, but it can also educate itself.”
Respondent 4: "AI should be able to solve problems on
its own."
Respondent 5: AI is an algorithm that at some point
starts to make decisions based on a large number of
derived, indirect inputs. Given that it is starting to enter
inputs from this point on its own, we can basically talk
about intelligence.
All respondents have encountered the term before and
consider it to be very relevant in the current era. However,
the answers varied, were unclear and confusing. All
respondents vaguely understood the characteristics, almost
no perception of the term was entirely consistent with its
definition in the literature. The answers did not capture, or
only partially captured the essence of AI.
The characteristics of the term ML was also part of the
question. Some of the most relevant comments were:
Machine learning
Respondent 1: It can evolve. It can learn and develop.”
Respondent 2: “Based on what it does, it learns, and, in
the future, it makes its activities better, more natural. The
first step is to have AI and then link ML to it so that AI
learns better, for instance to communicate with you based
on some data.”
Respondent 3: "All robots and bots, manual work ... those
things that can be set with a simple scheme ...
conditional: when this, then this ... can be automated.
These are responses to customer service, manual activity
automation, etc.
Respondent 4: "ML should be a process by which the
process is improving in some way, that means, it should
be able to produce better results after different
sequences."
Respondent 5: ML is more predictable and only
responds to directly set inputs. However, its specificity is
that it has received a great amount of data, or that it
"experienced" many simulated situations, making its
output more accurate and relevant."
All respondents identified ML as part of AI. Most
respondents also correctly identified that ML is a kind of
process that takes in large amount of data on which the
system bases its learning and developing. Again, the answers
only partially coincided with the definition given in the
literature. Two of the respondents confused the concepts of
AI and ML. The respondents took a considerably longer time
to formulate their answers, in comparison to the other
questions.
When we asked the additional question: "Where did you
hear about the mentioned terms for the first time?" most
respondents mentioned professional marketing conferences
where the topic is currently very popular. Other responses
included Internet resources, articles on marketing and
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
technology websites, e-mails after voluntary subscription to
news from a specific website and various educational-
relaxation videos. It follows from the above that respondents
are aware of the terms; they have encountered them in
practice and have an approximate idea of what they
represent.
Other questions were focused on the application of AI and
ML to digital marketing processes and on the interviewees’
opinions about how this application is reflected in practice.
All respondents agreed that AI and ML have great potential,
especially in technology-oriented industries such as digital
marketing, and that the issue is closely related to process
automation and the ability to work with large amounts of
data. The digital marketing industry uses digital technology
almost exclusively to process outputs. The work is carried
out on computers and tablets, including administrative
activities, communication and creative activities such as
drawing graphics, websites and banners design, additional
photo editing, etc. Even brainstorming, consultation and oral
outputs are recorded in electronic format. The specialized
work of experts allows a large part of these processes to be
automated. Some employees are already being replaced by
automated software for routine activities such as preparing
regular reports, distributing generated content, paying
subscription support on social networks, and even partially
communicating with customers via email or automated
responses on social network. Respondents see the greatest
potential of AI and ML in the automation of activities and
systematization of processes. Digital marketing is a
combination of analytical and creative work. According to
respondents, the analytical part will increasingly be carried
out by machines and software, and people will be able to
devote their time more to creative work and strategic
decision-making.
As part of the follow-up quantitative study, respondents
were asked to share their perception about the use of AI-
based tools in their company. The research revealed that,
even though most of the respondents knew what AI means
and how it can be utilized (59%), a large group of marketing
managers still lack basic knowledge about AI or have doubts
about its usefulness. 31% of respondents are not completely
sure what AI is and 10% of respondents only have minimal
knowledge of the term AI. When asked about the areas that
currently benefit from the power of ML and AI in their
companies, the three most frequent answers were: 1) Process
automation (81%); 2) Optimization of running marketing
campaigns (71%); 3) Data mining, including the
interconnection of several analytical tools (52%). When the
same questions was asked, but not limited to their own
company, the first choice was the same, namely process
automation (81%), followed by precise targeting (81%),
personalization of advertising content, and faster delivery of
results (52%).
USING ML IN PRACTICE
During the in-depth interviews, respondents were asked:
"Can you define marketing analytical tools that are based on
the basic principles of AI?" the following tools were
mentioned most often in respondents' responses: Google
(AdWords, Search); Google Data Studio; YouTube;
Facebook Ada Manager; chat applications; software for
programmatic purchase; Echobox; automated webinars;
Mailchimp.
Respondents are aware of some digital marketing tools
that are based on ML or AI. In addition to mentioning
marketing tools, respondents also listed companies that
develop graphics cards, autonomously managed cars, and
text translators. When asked which of these tools are used in
their company, respondents listed the same tools as in the
previous question.
The responses indicate that the tools that are based on the
principles of ML will mainly affect the following areas of
digital marketing:
advertisement systems and the management of
advertisement campaigns,
automation of reporting processes,
partial automation of communication (especially in
written form, such as emails or chat communication).
Analytical tools can be used to retrieve data from the
aforementioned areas. This eliminates the need for human
intervention and the software solutions can perform defined
actions autonomously.
In areas where AI is difficult to apply, respondents
consider:
creative processes,
building and maintaining relationships with business
partners.
These areas mainly work with data in a limited or very
specific form. Nowadays, the development of ideas or
drawing pictures can also be automated. However, results
generated by machines will probably never be comparable to
those generated from human brainstorming. Computers will
probably never be able to be emotional, which significantly
impacts the result (sometimes in a good way).
Another question in the qualitative survey was: "What
activities does your company plan for ML in the near
future?" If the respondent was poorly aware of analytical
tools, marketing analysis, big data, data management, etc.
and the respondent’s company supports the deployment of
ML-driven tools only marginally, only very basic or no
activities are planned. If the respondents were fully aware of
the topics discussed and works in an innovative company, the
ML plans were well-developed. Among the most frequent
areas where the respondentscompanies are planning such
activities are:
report automation,
linking several marketing analytical tools,
collection of data generated from marketing activities,
automation of the processing of collected data,
digital content creation.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
The last question was: "Do you think human work can be
replaced by machines in future? If so, how will marketing be
affected? All respondents agreed that human work certainly
can, and almost certainly will, be replaced by machines.
However, there will still be areas where machines and
automated software will not be able to fully control and
manage. According to the respondents, replacement will
mostly affect areas that work with data, as well as repetitive
work (such as weekly or monthly reporting of campaign
results). Employees who have been doing this repetitive work
in the past will be able to devote more time to strategic and
creative activity that is often required in the digital marketing
environment. When asked what percentage of the jobs in
digital marketing will be replaced by machines in five years
time, the minimum estimate was 50% and the maximum was
95%.
Following the qualitative study, the questionnaire aimed to
identify the barriers and risks of ML and AI application in
digital marketing and the anticipated future role of ML-based
tools in this industry. The cost of implementation (62%),
time required to learn how to deploy and use the tools (47%),
and market limitations (market size and lack of opportunities
to use ML and AI based tools) (45%) were the three most
common barriers. Marketing managers are afraid that using
ML-based analytical tools might : result in higher error rate
due to lack of know-how (50%); have a negative impact on
results and relationships due to missing moral and ethical
dimensions of behavior of the software and its intelligence
(41%); or mean losing control over marketing processes
(35%). Despite this, 72% of respondents believe that ML and
AI represent the future of digital marketing (9% responded
that they think these tools will cease to be used and 19%
were unsure).
FRAMEWORK FOR SUCCESSFUL ADOPTION OF ML IN
DIGITAL MARKETING
Analysis and interpretation of results from both qualitative
and quantitative research enabled the construction of a
framework aimed fostering the utilization and adoption of
ML-based analytical tools in digital marketing. The
framework consists of two main components: 1) Enablers,
i.e. factors of organizational culture and management
contributing to the atmosphere and conditions, where such a
project can be initiated and successfully finalized, and 2)
Process map, i.e. a map of processes of such a project
consisting of four main phases (Fig. 2).
FIGURE 2. Framework for successful adoption of ML in digital marketing
The absence of the enabler and/or following the suggested
path may negatively affect the success of ML-driven tools
adoption.
Firstly, the project needs top management support. Due to
its character (cost, cross-departmental reach etc.), adoption of
ML tools cannot be pursued without top managers
commitment. There have been cases where projects fueled by
regular employees and lower level of management were
successful. However, in general, a top-bottom driven
innovative climate is required for such complex project to
succeed. The companys managers should also act as leaders
who are aware of the importance of continuous
improvement, the contribution of detailed analytics to the
business and performance of their project teams. They should
introduce ideas, adopt new solutions and test them. Secondly,
introducing innovative solutions, measures, and tools needs
to be embedded in the company culture and connected with
its competitiveness and process efficiency. Thirdly, the
company needs to realize that technologies form the base for
its effective operation success. A knowledgeable CIO and
other C-level managers should have at least a high-level
understanding of recent technological innovations and their
advantages for business. It is advantageous if the company
has an internal IT team and/or has developed steady and
sustainable relationships with technical partners and
developers. Lastly, the company needs to foster frequent and
detailed use of data for the sake of deploying its services,
analyzing the impact of marketing strategies, and for internal
management. Without this component, the initiative to
automate data-intensive processes and reporting would not be
justifiable by hard data, suggesting huge time savings and
efficacy boosts.
The process map (Fig. 3) outlines the phases and necessary
steps for introducing automation of digital marketing
processes and ML-driven analytical tools into companies.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
FIGURE 3. Process of identification, deconstruction, and substitution of
periodical processes by ML-driven analytical tools
Following the process map will guide the organization
through the stages of the project to pay attention to all
necessary aspects, enabling the minimization of risk and
increasing the chance of project success. The processes of
implementation and adoption of ML-based analytical tools in
digital marketing are included in one of four main phases.
The focus of the process map is the identification of
periodical, time-consuming processes, predominantly in the
area of data analysis that be supplemented by intelligent
software solutions, utilizing the power of ML.
The process map builds on the knowledge generated by
this research. The implementation process, as outlined in Fig.
3, starts with the Education phase. The possible lack of
knowledge about the opportunities of utilizing of ML and AI
in marketing prevents companies from developing ideas on
deploying these technologies. Companies are encouraged to
raise knowledge and acquire know-how in this area. Four
potentially valuable and accessible groups of knowledge
resources have been identified. Quality literature from
verified peer-reviewed sources should be used to naturally
filter incorrect or misleading content. Marketing managers
are used to participating in conferences and they represent a
popular source of knowledge. Relevant materials can be also
obtained from free internet resources on topical blogs, vlogs,
websites of institutes, and associations and their newsletters.
Finally, a number of educational massive open online course
platforms, both free and paid, can be utilized to develop the
knowledge on data analytics, ML and AI by companys key
staff.
The Audit phase begins with the identification of
processes that consume a significant amount of key
employees time. These processes can originate in any of the
nine steps of implementation and evaluation of digital
marketing strategies (Fig. 1) are typically connected with
analyzing large amounts of data. This analysis results in the
deployment digital marketing campaigns, understanding the
behavior of consumers, their reaction towards the
communication message, products, and, in general, their
preferences. The identified processes need to be categorized
as either creative or periodical. The potential of ML-driven
analytical tools currently lies in the automation of time-
consuming periodical analytical processes that are often
ineffective when performed by humans. Each identified
process will be deconstructed to individual steps that are
performed in a sequence. Next, patterns of behavior and
decision-making are identified. These represent the know-
how of each employee when dealing with the process, for
example the inputs he or she considers when generating a
periodical performance report. Also, the preferences of end
users of the product, the client, need to be acknowledged and
considered. Finally, before moving to the next phase, all data
sources that are used in each process are identified and
closely analysed. Most of the platforms would have their
own APIs which can be used when developing the software
solution.
The Design phase starts with making the decision whether
to involve external experts and third parties in the process.
Following the team creation, the process of designing,
developing and releasing the software starts. Sometimes,
existing products and solutions might get chosen and
implemented, in other occasions these might need to be
customized and integrated. In other cases, custom
development needs to happen to satisfy the defined
requirements.
In the Deployment phase, the new solutions are to be
tested and the responsibility of their use will be transferred to
dedicated teams. Managers need to monitor their use, resolve
upcoming issues and address possible improvement ideas. In
case more processes that need attention and automation were
identified in the Audit phases, the process returns to evaluate
whether these occur periodically. It would also make sense to
automate them.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
VI. DISCUSSION AND CONCLUSION
Based on all the answers, we can clearly state that marketing
analysis is the cornerstone of the process of preparing and
implementing a marketing strategy. All nine respondents
confirmed that they work with data and use data analysis on
almost a daily basis. All respondents agreed that they would
take no further steps without information gained from
marketing analysis. This is also reflected in the model (Fig.
1), where the marketing analysis and analytical tools that
serve as a source of input information are an essential part of
the preparation and execution of steps 2, 3, 5, 7, 8, and 9.
The respondents confirmed the dependence of marketers
on marketing analysis and analytical tools, as the information
supports decision-making in the development of digital
marketing strategies. The analytical tools serve as the main
source of information for marketing analysis, on which
marketing managers base their strategic decisions. Working
with analytical tools has many advantages, such as: an
overview of competitor activities and market mapping;
accelerating the decision-making process; possibility of
importing internal data into third party analytical tools;
possibility of visualization of obtained data; rich targeting
capabilities; instant availability of information; the ability to
track real-time data in ongoing campaigns; the ability to
identify space for optimizing ongoing processes; the
accuracy of the observed parameters; possibilities of target
group segmentation by behavioral profiles; often assisted
analytical tool implementation or intuitive user interface.
On the other hand respondents referred to disadvantages of
using these tools, such as: data access can be limited in
specific analytical tools (some analytical tools require
additional charges for full data access); the implementation
of analytical tools is often expensive; inaccurate data
(incorrectly measured) leads to misrepresented decisions; the
correct use of analytical tools increases the pressure on
additional know-how, thus generating costly additional staff
training; the implementation of analytical tools is often a
complicated and time-consuming process; many analytical
tools use samples of respondents based on which gained
information prevail across the whole market (often it is
questionable whether a sample is representative); different
analytical tools use different metrics that are often not
compatible with each other and additional calculations are
needed to achieve a summary result; the metrics are often
adjusted, according to who owns the analytical tool; the
absence of a comprehensive view of the results achieved
across all analytical tools used.
The discussion about the use of data and big data for
marketing analysis can be summarized in the following
statement: large companies in the Slovak market will manage
and use data within their custom developed IT systems. This
has been the results of advancements in the technological
possibilities of data-oriented approaches and the decreasing
costs for the development and implementation of own
software solutions. Small businesses, on the other hand, will
probably not have enough money to pay for custom
development. The option of implementing standardized
solutions and applications can also prove difficult due to the
diversity of digital marketing activities.
Based on respondents' answers, the level of awareness of
ML and AI-related concepts, and partially also their current
use rate, were identified. Discussions about the current
application of ML to digital marketing processes and
identifying areas where ML can be applied led to the
conclusion that the greatest use of ML is to provide better
quality data and process automation. Automation can be
applied to a recurring process such as reporting, creating and
optimizing advertising campaigns, and even communication
with customers. When introducing innovative technologies,
the focus is primarily on process automation and data
processing. These activities are currently performed
manually, although only minimal intellectual input is needed
for these tasks. Even in the creation of digital content, which
appears to be highly intellectually intensive, respondents
referred to automated translation of articles from foreign
media. Most of the content that is published on Slovak digital
platforms is taken over from foreign media and only slightly
adapted to the needs of the local market. The respondents’
companies are turning to other technology companies to
develop and implement tools that use innovative
technologies, including ML. They see the process of self-
development as too costly and time-consuming and they
therefore focus on working with specialists in the field.
Based on the results of this research, it can be stated that
ML already has a place in the work of marketing managers
and digital marketing specialists. The qualitative research
provided insights into the way respondents think about the
use of ML in digital marketing and how their companies
perceive the benefits and drawbacks of ML and AI tools.
Findings from the experience of digital marketing
professionals in Slovakia can be tested in other countries of
the world in the follow-up research. The follow-up
quantitative research revealed that the topic is still new to
marketing managers and some of them are unsure, even
about the basic definition and functioning of ML and AI.
Despite this fact and identified barriers of adoption and risks
of implementing ML-based tools, most marketing managers
believes that ML and AI are the future of digital marketing.
The framework represents a contribution to the current
body of knowledge, which predominantly focuses on the
development of the most advanced and smart technologies,
while lacking the focus on their deployment and adoption. It
enables practitioners both from IT and marketing to
understand the complexities of the process of implementing
ML-based tool into the organization. The framework also has
the potential to reveal opportunities of such a project and
guide the organization through the whole deployment
process.
This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see http://creativecommons.org/licenses/by/3.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2924425, IEEE Access
VOLUME XX, 2019
REFERENCES
[1] DOMO, Data Never Sleeps 6.0, 2018. [Online]. Available:
https://www.domo.com/solution/data-never-sleeps-6. [Accessed:
05-Feb-2019].
[2] B. Marr, How Much Data Do We Create Every Day? The Mind-
Blowing Stats Everyone Should Read, 2018. [Online]. Available:
https://www.forbes.com/sites/bernardmarr/2018/05/21/how-much-
data-do-we-create-every-day-the-mind-blowing-stats-everyone-
should-read/#451a9afd60ba. [Accessed: 19-Jul-2018].
[3] A. Miklosik, M. Kuchta, and S. Zak, Privacy Protection vs.
Advertising Revenues: The Case of Content Publishers, Istanbul
Univ. Fac. Commun. J., vol. In print, 2018.
[4] F.-C. Cheng and Y. S. Wang, The Do Not Track Mechanism for
Digital Footprint Privacy Protection in Marketing Applications, J.
Bus. Econ. Manag., vol. 19, no. 2, pp. 253267, 2018.
[5] I. Heimbach, D. S. Kostyra, and O. Hinz, Marketing Automation,
Bus. Inf. Syst. Eng., vol. 57, no. 2, pp. 129133, Apr. 2015.
[6] A. Amrollahi and B. Rowlands, Collaborative open strategic
planning: a method and case study, Inf. Technol. People, vol. 30,
no. 4, pp. 832852, Nov. 2017.
[7] J. Rojas-Arce, O. Gelman, and J. Suarez-Rocha, The Methodology
for Strategic Plan Implementation, J. Appl. Res. Technol., vol. 10,
no. 2, pp. 248261, Apr. 2012.
[8] A. Efstathiades, S. Tassou, and A. Antoniou, Strategic planning,
transfer and implementation of Advanced Manufacturing
Technologies (AMT). Development of an integrated process plan,
Technovation, vol. 22, no. 4, pp. 201212, Apr. 2002.
[9] R. M. Schwartzstein, G. C. Huang, and C. M. Coughlin,
Development and Implementation of a Comprehensive Strategic
Plan for Medical Education at an Academic Medical Center, Acad.
Med., vol. 83, no. 6, pp. 550559, Jun. 2008.
[10] A. G. B. Tettamanzi, M. Carlesi, L. Pannese, and M. Santalmasi,
Business intelligence for strategic marketing: Predictive modelling
of customer behaviour using fuzzy logic and evolutionary
algorithms, in Applications of Evolutionary Computing,
Proceedings, 2007, vol. 4448, p. 233+.
[11] X.-W. Chen and X. Lin, Big Data Deep Learning: Challenges and
Perspectives, IEEE Access, vol. 2, pp. 514525, 2014.
[12] J. E. Frisk, F. Bannister, and R. Lindgren, Evaluation of
Information System Investments: A Value Dials approach to closing
the theory-practice gap, J. Inf. Technol., vol. 30, no. 3, pp. 276
292, Sep. 2015.
[13] J. Balla, Seven benefits from using marketing analytics, 2014.
[Online]. Available:
https://blogs.sas.com/content/customeranalytics/2014/09/17/seven-
benefits-from-using-marketing-analytics/. [Accessed: 14-Jan-2018].
[14] N. G. Polson and V. Sokolov, Deep Learning: A Bayesian
Perspective, Bayesian Anal., vol. 12, no. 4, pp. 12751304, Dec.
2017.
[15] J. B. Heaton, N. G. Polson, and J. H. Witte, Deep learning for
finance: deep portfolios, Appl. Stoch. Model. Bus. Ind., vol. 33, no.
1, pp. 312, 2017.
[16] B. Marr, What Is The Difference Between Deep Learning,
Machine Learning and AI?, 2016. [Online]. Available:
https://www.forbes.com/sites/bernardmarr/2016/12/08/what-is-the-
difference-between-deep-learning-machine-learning-and-
ai/2/#ff0aefc154fa. [Accessed: 21-Jan-2018].
[17] M. Pangallo and M. Loberto, Home is where the ad is: online
interest proxies housing demand, EPJ Data Sci., vol. 7, Nov. 2018.
[18] Y. Kurachi, S. Narukawa, and H. Hara, Al Chatbot to Realize
Sophistication of Customer Contact Points, Fujitsu Sci. Tech. J.,
vol. 54, no. 3, pp. 28, Jul. 2018.
[19] G. Pant and S. Pant, Visibility of corporate websites: The role of
information prosociality, Decis. Support Syst., vol. 106, pp. 119
129, Feb. 2018.
[20] K. Y. Abeywardana, A. R. Ginige, N. Herath, H. P. Somarathne,
and T. M. N. S. Thennakoon, Hashtag Generator and Content
Authenticator, Int. J. Adv. Comput. Sci. Appl., vol. 9, no. 9, pp.
246252, Sep. 2018.
[21] T. Mackey, J. Kalyanam, J. Klugman, E. Kuzmenko, and R. Gupta,
Solution to Detect, Classify, and Report Illicit Online Marketing
and Sales of Controlled Substances via Twitter: Using Machine
Learning and Web Forensics to Combat Digital Opioid Access, J.
Med. Internet Res., vol. 20, no. 4, Apr. 2018.
[22] H.-F. Ho and C.-C. Chen, Application of Artificial Neural Network
Method to Analyze Users Behavior Using Clickstream Data, in
Fuzzy Systems And Data Mining III (FSDM 2017), 2017, vol. 299,
pp. 9499.
[23] B. Grandhi, N. Patwa, and K. Saleem, Data Driven Marketing for
Growth and Profitability, in Global and National Business
Theories and Practice: Bridging the Past with the Future, 2017, pp.
675694.
[24] Q. Ma, S. Zhang, W. Zhou, S. Yu, and C. Wang, When Will You
Have a New Mobile Phone? An Empirical Answer From Big Data,
IEEE Access, vol. 4, pp. 1014710157, 2016.
[25] J. Wang, W. Zhang, and S. Yuan, Display Advertising with Real-
Time Bidding (RTB) and Behavioural Targeting, Found. Trends
Inf. Retr., vol. 11, no. 45, pp. 297435, 2017.
[26] T. K. Mackey, J. Kalyanam, T. Katsuki, and G. Lanckriet, Twitter-
Based Detection of Illegal Online Sale of Prescription Opioid, Am.
J. Public Health, vol. 107, no. 12, pp. 19101915, Dec. 2017.
[27] Y. Hu, C. Hu, S. Fu, M. Fang, and W. Xu, Predicting Key Events
in the Popularity Evolution of Online Information, PLoS One, vol.
12, no. 1, Jan. 2017.
[28] T.-L. Luong, M.-S. Cao, D.-T. Le, and X.-H. Phan, Intent
Extraction from Social Media Texts Using Sequential Segmentation
and Deep Learning Models, in 2017 9TH IInternational
Conference on Knowledge and Systems Engineering (KSE 2017),
2017, pp. 215220.
[29] J. Järvinen, A. Tollinen, H. Karjaluoto, and C. Jayawardhena,
"Digital and social media marketing usage in B2B industrial
section", in Marketing Management Journal, vol. 22, no. 2, 102
117, 2012.
[30] T. C. Melewar, and N. Smith, "The Intemet revolution: some global
marketing implications," in Marketing Inteigence & Planning, vol.
27, no. 6, pp. 363-369, 2003.
[31] S. Li, J. Z. Li, H. He, P. Ward, and B. Davies B. J., "WebDigital: A
Web-based hybrid intelligent knowledge automation system for
developing digital marketing strategies," in Expert Sys. Appl., vol.
38, no. S pp. 10606-10613, 2011.
[32] C. Wymbs, "Digital marketing: The time for a new "academic
major" has arrived," in J. Mark. Educ., vol. 55, no. 1, pp. 93-106,
2011.
[23] R. Kniahynyckyj, The Pros and Cons of AI in Marketing, 2017.
[Online]. Available:
https://www.business2community.com/marketing/pros-cons-ai-
marketing-01916940. [Accessed: 20-Jan-2018].
[34] K. Reddy, Advantages and Disadvantages of Artificial
Intelligence, 2017. [Online]. Available:
https://content.wisestep.com/advantages-disadvantages-artificial-
intelligence/. [Accessed: 20-Jan-2018].
... Chen et al., 2021;S.-Y. Chen et al., 2021), traceability of actions (Jain et al., 2021) and provide competitive edge (Miklosik et al., 2019). ...
... Chen et al., 2021;S.-Y. Chen et al., 2021;Miklosik et al., 2019) and is expected to revolutionize marketing (Huang & Rust, 2018;Huang & Rust, 2021;Rust, 2020) through better (Ma & Sun, 2020). It is argued that AI has the potential to affect both, revenue and costs. ...
Article
Full-text available
One of the core challenges in digital marketing is that the business conditions continuously change, which impacts the reception of campaigns. A winning campaign strategy can become unfavored over time, while an old strategy can gain new traction. In data driven digital marketing and web analytics, A/B testing is the prevalent method of comparing digital campaigns, choosing the winning ad, and deciding targeting strategy. A/B testing is suitable when testing variations on similar solutions and having one or more metrics that are clear indicators of success or failure. However, when faced with a complex problem or working on future topics, A/B testing fails to deliver and achieving long-term impact from experimentation is demanding and resource intensive. This study proposes a reinforcement learning based model and demonstrates its application to digital marketing campaigns. We argue and validate with actual-world data that reinforcement learning can help overcome some of the critical challenges that A/B testing, and popular Machine Learning methods currently used in digital marketing campaigns face. We demonstrate the effectiveness of the proposed technique on real actual data for a digital marketing campaign collected from a firm.
... In the emerging trends, other keywords "deep learning" and "social media analysis" provide that social media analytics will no longer be limited to "single data sources." Instead, multiple digital marketing performances across social media platforms or input data for "deep learning" will be more diverse sources helping to provide greater generalizability and more efficiency for recommendation systems such as in the fields of e-commerce field, airline companies, or hospitality industry (Miklosik et al., 2019). ...
Article
Full-text available
As one of the most recent forms of marketing, digital marketing has emerged as the fastest-growing sector of the industry. A large number of academic studies have been published on different aspects of this field. The main objectives of this paper are to illustrate the evolution of the marketing field, examine the research focuses, and identify the future trends of digital marketing research using a combined bibliometric approach of co-word and co-citation analysis. Using the Web of Science (WoS) database, a representative set of 254 academic publications was selected and analyzed by VOSviewer software. The co-citation results represent four main research clusters, including digital marketing and business performance, opportunities and challenges, digital marketing development and research methodology, and digital marketing and customer behaviour. Similarly, the results of the co-keyword analysis reveal six themes, including foundation, digital marketing implications, digital marketing channels, content marketing, business sectors, and emerging trends.
... In India, Upgrade recently registered 2,000 students in entrepreneurial, digital marketing, analytics, and products management courses, with live lectures delivered using two-way satellite technology. [4] V. DIGITAL MARKETING People living in various locations frequently confuse the phrase "Digital Marketing" with other terms such as "Online Marketing," "Internet Marketing," and "Web Marketing." In the us, it is most usually referred to as Online Marketing. ...
Conference Paper
The study investigates businesses have begun by developing new business models, processes, and technology. Most others are already seeing the potential of digital revolution through bold innovation efforts. Directors play an important role in strategy formulation to enhance business outcomes and control the narrative for the future. Artificial intelligence has received little attention. Deep learning machines may propel digital advertising to new heights, with Artificial Intelligence providing all the difference. This study seeks to determine the consequences of Indian customers' responses across different demographics to algorithms and their ability to sell, which may be the digitization marketing. We discovered that software developers' architecture is collaboration with marketing professionals who employ machines utilizing deep learning by taking consumer attitude, behavior, and choices into account. This will provide enormous benefits to businesses since correct information regarding customers will be readily available to marketers in the future. The performance of the machines under diverse settings is described using just a causal model and regression models. Furthermore, it provides practical, workable, and feasible research-driven answers to nearly every business-related challenge.
... In such a situation, the organization need to prove itself better than other services to grab the market. In this context, data collection and data management are highly required for the business based upon which a company is able to streamline the business process [1]. However, the data management through the manual process is not only difficult and time-consuming but also has a lot of chance of error. ...
Article
Full-text available
The comprehensive impact of machine learning (ML) and artificial intelligence (AI) on the different segments of society has not yet been applied in the marketing field strongly. Despite such an adverse balance of machine learning a very kind of potential benefit can be found that includes the opportunity to apply diverse robust methods for generating scientific discoveries. In order to reduce such shortfalls, this paper has been developed by targeting four goals. First, to showcase the inside of the machine learning application in the marketing management process by utilizing the types of unsupervised, supervised, reinforcement learning and algorithms. In addition to this, this paper has also aimed to find out and analyze Manager-The British Journal of Administrative Management Industry Qualifications, The Institute of Administrative Management, UK | 74 two potential marketing strategies for enhancing the general workflow of the business. The third aim of this study is to analyze the machine learning application published in top tire management and marketing journals. In the final stage, this paper as aimed to discuss the possible way to improve the ML technology for the betterment of the marketing while analyzing the impact of the ML on current trends as well.
Conference Paper
Full-text available
A funding grant was awarded in August 2020 by the European Union to a consortium of 11 European universities to develop a game-based experiential teaching module to train marketing students in brand management to improve their employability. A first round of testing was scheduled for March 2021, in the surge of the Covid pandemic. This paper aims at describing the solution found by the research team to perform the testing despite challenging circumstances and to prevent postponement of the project end date. Clear opportunities for improvement were identified thanks to the participation of very highly engaged students from 4 countries to an online paper prototyping based testing session.
Article
This article aims to analyze the evolutionary process of marketing from the intense changes in social and consumption patterns and the use of new advanced technologies. Therefore, a bibliographic research was carried out, with the intent to understand not only the progressions in marketing strategies and activities throughout history, but also their current stage, intrinsically linked to issues of humanization, digitization, technological innovations and building new consumer experiences based on a new hybrid world. From the analysis of the new context in which companies at a global level are making the transition from traditional marketing actions to digital marketing, it was possible to conclude that new emerging technologies have great potential for a more accurate understanding of consumer needs, obtaining insights and strategic guidelines on the market that optimize the decision-making process, segmentation and selection of target markets, building new consumer experiences and strengthening brand building and growth results.
Article
Full-text available
Even though machine learning (ML) applications are not novel, they have gained popularity partly due to the advance in computing processing. This study explores the adoption of ML methods in marketing applications through a bibliographic review of the period 2008–2022. In this period, the adoption of ML in marketing has grown significantly. This growth has been quite heterogeneous, varying from the use of classical methods such as artificial neural networks to hybrid methods that combine different techniques to improve results. Generally, maturity in the use of ML in marketing and increasing specialization in the type of problems that are solved were observed. Strikingly, the types of ML methods used to solve marketing problems vary wildly, including deep learning, supervised learning, reinforcement learning, unsupervised learning, and hybrid methods. Finally, we found that the main marketing problems solved with machine learning were related to consumer behavior, recommender systems, forecasting, marketing segmentation, and text analysis—content analysis.
Article
This paper explores the implications of digitalization and business model innovation for the principal-agent conflict. Continuous digital transformation has recently become a feature sine qua non for companies. It is also stimulating business model innovation resulting in the growing adoption of ecosystem-based models. These trends may have significant implications for the principal-agent relationship, essential for understanding value creation by a firm. In order to analyze the impact of digital transformation, we use blockchain technology as a proxy. To measure the potential impact on the principal-agent conflict, we study management and shareholder-sponsored proposals at annual meetings. The level of shareholder involvement in governance is measured based on the number of shareholder-sponsored proposals received. In addition, we measure shareholder support for management-sponsored proposals. A sample of 2481 NYSE, Nasdaq and AMEX-traded firms for the period 2015–2019 is used. First of all, we show that digitalization per se has a mitigating impact on the agency conflict. Shareholders become more active, albeit not more hostile towards management. Secondly, we have identified the strongest impact in such sectors as information technology, communications, finance, and healthcare. These are the most significantly impacted by ecosystem-based business model innovation. We conclude that digitalization and ecosystem-based business models are mutually reinforcing in mitigating the principal-agent conflict.
Article
Full-text available
Abstract Online activity leaves digital traces of human behavior. In this paper we investigate if online interest can be used as a proxy of housing demand, a key yet so far mostly unobserved feature of housing markets. We analyze data from an Italian website of housing sales advertisements (ads). For each ad, we know the timings at which website users clicked on the ad or used the corresponding contact form. We show that low online interest—a small number of clicks/contacts on the ad relative to other ads in the same neighborhood—predicts longer time on market and higher chance of downward price revisions, and that aggregate online interest is a leading indicator of housing market liquidity and prices. As online interest affects time on market, liquidity and prices in the same way as actual demand, we deduce that it is a good proxy. We then turn to a standard econometric problem: what difference in demand is caused by a difference in price? We use machine learning to identify pairs of duplicate ads, i.e. ads that refer to the same housing unit. Under some caveats, differences in demand between the two ads can only be caused by differences in price. We find that a 1% higher price causes a 0.66% lower number of clicks.
Article
Full-text available
In the recent past, Online Marketing applications have been a focus of research. But still there are enormous challenges on the accuracy and authenticity of the content posted through social media. And if the social media business platforms are considered, majority of the users who try to add a market value to their own product face the problem of not getting enough attention from their target audience. The purpose of this research is to develop a safe and efficient trending hashtag generating application solution for social media business users which generates trending and relevant hashtags for user content in order to get a broad reach of target audience, automatically generates a meaningful caption to their relevant posts and guarantees the authenticity of the product at the same time. The user content is analyzed and filters the important keywords, generates a meaningful caption, suggest related trending keywords and generates trending hashtags to get the required reach for online marketers. Additionally, the marketing products' content authentication is ensured. The application uses Natural Language Processing, Machine Learning, API technologies, Java and Python technologies. A unique database is assigned to users which contains rankings for each user. The target audience who engages in buying products get to know about the status of the sellers with respect to authenticity of the content. It is believed that the application provides a promising solution to existing audience reach problems of online marketers and buyers. The significance of this system is to help marketers and buyers to engage in online buying and selling with much effective, reliable and safer ways. This mitigate the vulnerability of bad social media marketing influences and helps to establish a safe and reliable online marketing practice to make both sellers and buyers happy. This paper provides a brief description on how to perform an organized online marketing discipline via the Trending Hashtag Generator & Image Authenticator application. © 2018 International Journal of Advanced Computer Science and Applications.
Article
Full-text available
Serious concerns about the invasion of digital footprint information privacy due to intense commercial promotion through data mining has led to the emergence of privacy by design in the form of the Do Not Track (DNT) mechanism. This paper attempts to construct a theory to justify and find an appropriate solution to balance the different interests by implementing the DNT mechanism in the real-world marketing industry. The research method involves deduction in legal reasoning. This paper argues that digital footprint information privacy, which has high commercial value, should at least be awarded the status of a semi-fundamental human right. Additionally, when to adopt a DNT opt-out or default mechanism depends on the type of personal information involved. The practical implications suggest a compromise between digital footprint privacy protection and commercial applications in the marketing industry, to be achieved through technology. Since this important topic is relatively new in the area of marketing applications and no primary academic research has established a complete theoretical legal foundation, this article is among the first to do so. Beside its originality, this article also contributes to the literature by proposing a theoretically practical mechanism for both digital footprint privacy protection and marketing profits.
Article
Full-text available
With an ever expanding content and user base, the Web presents information discovery and consumption challenges for both consumers and producers of information. Producers of information strive for visibility among consumers who have limited attention. Corporate websites are a primary digital marketing channel for firms through which they seek to gain a bigger share of their stakeholders' (i.e., customers, investors, communities) attention. Using observations spanning several years we study the website visibility, as measured by user traffic, of more than 2500 public firms and its association with properties of corporate websites and the corresponding firms. One property that is of particular interest to us is the availability of "community-engaging" pages, i.e., pages that support blogs or forums on the website or provide links to external social media platforms such as Facebook. These community-engaging pages signify online prosocial services provided by firms. We find that websites with larger number of community-engaging pages are associated with higher visibility. This provides a novel empirical support for the promotion and use of social media content and tools on websites of firms. We also find that websites with more specific content are associated with lower visibility while providing more out-links is associated with higher visibility. We observe these results consistently over time. These associations are observed while controlling for the size of the firms, types of their industries, the magnitude of media attention and other firm-level heterogeneity. Finally, machine learning models derived from our empirical analysis provide strong predictive utility for out-of-sample data.
Article
Purpose In the current business environment, more uncertain than ever before, understanding consumer behavior is an integral part of an organization's strategic planning and execution process. It is the key driver for becoming a market leader. Therefore, it is important that all processes in business are customer centric. Marketers need to harness big data by engaging in data driven-marketing (DDM) to help organizations choose the “right” customers, to “keep” and “grow” them and to sustain “growth” and “profitability”. This research examines DDM adoption practices and how companies can aim to enhance shareholder value by bringing about “customer centricity”. Design/methodology/approach An online survey conducted in 2016 received 180 responses from junior, middle and senior executives. Of the total responses, 26% were from senior management, 39% from middle management and the remaining 35% from junior management. Industries represented in the survey included retail, BFSI, healthcare and government, automobile, telecommunication, transport and logistics and IT. Other industries represented were aviation, marketing research and consulting, hospitality, advertising and media and human resource. Findings Success of DDM depends upon how well an organization embraces the practice. The first and foremost indicator of an organization's commitment is the extent of resources invested for DDM. Respondents were divided into four categories; Laggards, Dabblers, Contenders and Leaders based on their “current level of investments” and “willingness to enhance investments” soon. Research limitations/implications With storming digital age and the development of analytics, the process of decision-making has gained significant importance. Judgment and intuition too are critical to the process. Choosing an appropriate action cannot be done strictly on a rational basis. Practical implications The results of the study offer interesting implications for managing the growing sea of data. An iterative and incremental approach is the need of the hour, even if it has to start with baby steps, to invest in and reap the fruits of DDM. The intention to use any system is always dependent on two primary belief factors: perceived usefulness and perceived ease of use; however, attitudes and social factors are equally important. Originality/value There is a dearth of knowledge with regards to who is and is not adopting DDM, and how best big data can be harnessed for enhancing effectiveness and efficiency of marketing budget. It is, therefore, imperative to build a knowledge base on DDM practices, challenges and opportunities. Better use of data can help companies enhance shareholder value by bringing about “customer centricity”.
Article
Digitalization has advanced, spreading to all industries. Companies are required to improve marketing efficiency and strengthen engagement with customers more than ever by utilizing digital technology. Promoting emotional value through the customers service user experience is important and contact centers play a key role in the experience as customer contact points for companies. However, contact centers currently face a mountain of problems, including labor shortages, support for diversifying channels, and the need to improve efficiency by making use of AI. To solve these problems, Fujitsu has started offering the Customer Engagement Solution CHORDSHIP, a contact point sophistication solution. The CHORDSHIP Digital Agent, which is at the core of CHORDSHIP, is equipped with an AI technology ideal for contact centers: conversation-machine learning hybrid AI. Its biggest feature is the capability to deliver highly accurate automatic answers simply by using existing FAQ data, in addition to its realization of support for diversifying channels and 24/7 service availability. This paper outlines the contact point sophistication solution offered by Fujitsu and describes the AI chatbot technology behind the solution.
Article
Purpose The purpose of this paper is to show how collaborative information technology (IT) tools and a crowdsourcing model can be leveraged for the purpose of strategic planning. To achieve this objective, a formal method of open strategic planning (OSP) is proposed. Design/methodology/approach Based on a review of the literature a set of activities, stakeholders, and governing rules are identified in the form of an OSP method. The proposed planning method is implemented in a case study of strategic planning in an Australian university. Observations by the research team, and archival records were used to ascertain the relevance of the used method. Findings A method for OSP is presented and assessed. The method contains four phases: pre-planning, idea submission, idea refinement, and plan development. These phases cover the activities required from conceptualization to preparing and publishing the strategic plan. The findings clarify how the principles of OSP helped the organization to include more stakeholders and provided the opportunity to make the planning process transparent through use of a collaborative IT tool. Practical implications The study provides managers and planning consultants with detailed guidelines to implement the concept of open strategy. Originality/value This study is among the few to propose a method for OSP based on empirical research. The study also shows how collaborative IT tools can be used for high-level organizational tasks such as strategic planning.