A Framework for Analysing the

Impact of Data Analytics and the

Internet of Things on Digital
Marketing
Dimitris K. Kardaras, PhD Bill Karakostas, PhD
School of Business, VLTN GCV
Athens University of Economics and Antwerp, Belgium,
Business, Athens, Greece. Email: [email protected]
Email: [email protected]

Stavroula Barbounaki, PhD Stavros Kaperonis, PhD

Merchant Marine Academy of Panteion University of Social and
Aspropyrgos, Political Sciences,
Greece, Athens, Greece,
Email: [email protected] Email: [email protected]

ABSTRACT
The proliferation of the Internet of Things will open up new opportunities for implementing
the digital transformation of businesses. Available data is expected to rise in unprecedented levels
of quantity with the IoT playing an important role towards that end. Data analytics techniques will
provide businesses with refined pieces of information in almost all aspects in both B2C and B2B
context, thus refining Services design and customization with more flexibility and options that focus
right at the heart of consumers’ needs. Digital marketing depends on these developments. This
chapter aims at proposing a framework for analysing the implications of data analytics and IoT on
digital marketing.

Keywords: IoT, Digital Marketing, Data Analytics, DEMATEL

INTRODUCTION
Customers are increasingly using digital media to communicate and interact with
companies. Most of such interactions take place on the Web, thus generating vast amounts of data,
which analysis reveals valuable insights pertaining to customer preferences, behaviour patterns,
etc. As a result, companies have already started showing interest in Web analytics, in order to
understand online customers’ behaviour, preferences and improve their experience, while getting
insights on how to increase business revenues (Chaffey and Patron, 2012). Web Analytics are used
by more than 60% of the top 10 million most popular websites around the world (Web Technology
Surveys, 2014). Digital marketing is a strategy to promote products and brands through the internet
or digital media. It helps in developing and maintaining customer relationships through online
activities. There are a number of tools that can be used such as Customer Relationship Management
(CRM), social CRM, email software, e-commerce software, marketing automation software and
Web analytics (TechNavio, 2017). The Digital Marketing Institute (DMI) refers to digital
marketing as “The use of digital technologies to create an integrated, targeted and measurable
communication which helps to acquire and retain customers while building deeper relationships
with them” (Royle and Laing, 2014). The importance of Digital Marketing is already been
confirmed in a study by IBM consisting of interviews with CMOs (IBM Institute for Business
Value, 2011). These CMOs highlighted the following four biggest challenges: (1) explosion of data
(i.e. big data), (2) social media, (3) proliferation of channels, and (4) shifting consumer
demographics. Customer behaviour changes and technological developments, justify the IBM
study. The computer screen has obviously been the most popular display of Web content, but things
have changed. Recent market statistics show the dramatic proliferation of smartphones since 2005,
and the profound customers’ shift towards mobile marketing. Indeed, more than 50% of searches
are carried out on mobile, 91% of Facebook usage, Daily Active users, is on mobile, 80% of
Facebook advertising revenue is on mobile, 90% of mobile media time is spent in mobile apps
(Digital marketing Megatrends, 2017).
Now, data analytics techniques can be, more than ever, used to analyse the masses of data
that users generate on the Web. The proliferation of devices such as computers, laptops, smart TVs
and smartphones established new communication channels between customers and businesses. In
the coming years the proliferation of the Internet of Things (IoT) is expected to further expand the
number of ways that companies contact customers and vice versa. Therefore, consumers navigate
and proceed to purchases on the Internet through various channels, revealing their preferences.
They write reviews and share their experiences with other consumers, thus providing valuable data
for their likes and dislikes as well as they comment on the quality level of the services they
consume. The use of data analysis techniques such as statistics, multi-criteria analysis methods and
recently Artificial Intelligence (AI) and machine learning can provide solutions to e.g. attract,
engage, persuade customers and create value for companies. Discovering customer behaviour
patterns, designing customised services, redesigning business processes that bring customer
experience to the next level, interacting in a personalised manner, are but a few examples. New
services, such as the Chat bots are expected to change customer experience in a B2C context by
utilising data analysis and AI and providing personalised interactive communication (Digital
marketing Megatrends, 2017). Social media also play an important role in the new digital
environment. Research results show (Leeflang, et al., 2014) that more than 90% of all customers
read online product reviews before they proceed to purchasing and that 67% of all consumers’
decisions to but are based on user-generated content (UGC). Consumers, before making up their
mind what product to buy, they read at least four product reviews (Godes and Silva, 2012; Kee,
2008). The consultation of the reviews is being found to be very influential (Godes and Silva, 2012;
Kee, 2008). Thus, social media creates content that allows customers to interact with each other,
exchange opinions, and experiences. Interestingly, that content may be more influential than
advertising. The demographics also have changed. Consumers between 18 and 35 years of age, i.e.
the Millennials have established their careers, homes and families. This is now the largest living
generation; it spends collectively $600 billion a year. Brands will grow digital investment in order
to connect with these digital natives (Forbes, 2017a).
Business in line with these developments and expectations increase their spending on
digital marketing. Market statistics and projections show that digital advertising spending globally
which accounting for desktops and laptops as well as mobile devices is estimated to reach a total
of 335 billion U.S. dollars by 2020. With respect to mobile internet advertising, which is a heavily
invested sub-sector of the digital advertising industry, is estimated to reach over 247 billion U.S.
dollars in 2020 (Digital Advertising, 2015). More market statistics show that due to the increasing
use of smartphones and the availability of cheap internet services more than 30% of businesses are
planning to spend around 75% of their advertising expenditures on digital marketing within the
next five years (TechNavio, 2017). According to (Forbes, 2017a), the US spending in paid search,
display advertising, social media advertising, online video advertising and email marketing will
reach $120 billion by 2021, and it is expected to account for the 46% of all advertising in five years.
This chapter aims to propose a framework for analysing the implications of data analytics
and IoT on digital marketing. It suggests that IoT based smart devices will provide the means to
collect data across communication channels as well as along value chain between customers and
businesses. This chapter draws on the Virtual Value Chain (VVC) (Rayport and Sviokla, 1995) and
the Customer Service Life Cycle frameworks. Business process modelling is used to capture the
information flow along the value chain. The use of data analysis provides valuable insights for
consumers’ preferences. This chapter shows how DEMATEL can be used to investigate the
interrelationships between customer related data, with digital media characteristics and business
processes in order to facilitate the development of personalised digital marketing campaigns.
Customer Journey Map (Johnston, 2015)

BACKGROUND

Introduction to Internet of Things (IoT)

The IoT represents a significant technological field with enormous implications for both
consumers and businesses. The pace of the IoT market development reflects its importance. In
(Gartner 2017) forecasts suggest that 8.4 billion IoT connected devices will be in use worldwide in
2017, up 31 percent from 2016, and is expected to reach 20.4 billion by 2020. Total spending on
endpoints and services will reach almost $2 trillion in 2017. China, North America and Western
Europe are the leading areas in using of IoT devices and the three regions together will represent
67% of the global number of IoT devices in 2017 (Gartner 2017). IoT services are central to the
rise in IoT devices, with their spending to reach $273 billion in 2017. According to (Forbes, 2017b),
B2B spending on IoT technologies, apps and solutions will reach $267 billion by 2020. The 50%
of the total IoT investment will target discrete manufacturing, transportation and logistics, and
utilities. IoT Analytics spending is predicted to generate $21.4 billion by 2020. Finally, spending
on IoT applications is expected to generate $64.1 billion by 2020.

What is IoT
The IoT was coined in the late 1990s, however many of the technologies that underpin it,
such as semiconductors and wireless networks have predated that by several decades. The IoT is
created by the convergence of hardware, software and communication technologies. The hardware
consists of the connected devices – which range from simple sensors to smartphones and wearable
devices – and the networks that link them, such as 4G LTE (as well as the forthcoming 5G) and
wireless technologies such as Wi-Fi and Bluetooth. Software components include applications and
platforms that store and process IoT generated data. The combination of such components makes
possible the creation of services that create value for businesses and consumers.
IoT has great potential for all business sectors, including marketing, retailing and supply
chain processes. However, for the IoT premise to become a reality there are several enablers both
technical and organisational that need to be realised, as well as several obstacles that need to be
overcome. These are discussed in the following sections.

IoT standardisation
For IoT to flourish, interoperability of the various systems and technologies that comprise
it must be achieved. Ongoing standardization efforts involve harmonizing Internet protocols for a
wireless IoT networks. Adoption of these protocols and standards involves collaboration among
government standards institutes like ETSI, organizations like IETF/IAB, and alliances like ZigBee.
IoT as a Service
A parallel activity towards making IoT widely available to businesses and customers is the
concept of IoT as service that is similar to other online IT services, such as cloud and storage
services.
IoT as a service builds upon the effect of economies of scale, the scope of sharing IoT
assets, and the ubiquity of Internet connectivity. It also relies on industry wide standardization and
interconnectivity. The unique value proposition of IoT as a service may come from combining IoT
architectural components such as physical objects, network, data storage, and applications to
providing customized services to businesses on a pay-as-you-go basis [LEE]. IoT as a service has
complementarity to Cloud processing and Big Data analytics on the Cloud services.

IoT Platforms
An IoT platform is a type of middleware that helps to connect IoT devices, data and
applications, and thus overcome interoperability problems. IoT platforms facilitate easy
connectivity of devices to the platform and can reduce the complexity of IoT application
development. Additionally, IoT platforms can provide integrated data analytics and machine
learning services for automating the analysis of IoT data.

Intelligent/cognitive IoT
Cognitive technologies can analyse the massive amounts of unstructured data generated by
IoT to identify hidden trends and patterns and to mine profitable business knowledge. In the retail
sector, for example such intelligent analysis of IoT generated data can be used to better understand
consumer behaviours and deliver differentiated, individualized shopping experience at any point in
time [IBM].

IoT technical enablers

1. Efficient IoT device management

The IoT is an ambitious paradigm that significantly increases the scale of connected
devices. To manage the complexity of such a scale, interworking solutions that can reuse pre-
existing technologies seamlessly with newer and more efficient technologies is a necessity.

2. Controlled data transmissions and context awareness

Due to the tremendous volumes of data generated by IoT it is not always feasible to process
all the streaming data available to those devices. The concept of context aware data processing
refers to the ability of sensors and devices to use context specific information such as location,
temperature, and the availability of a certain device, to decide what types of data to collect and
interpret to provide relevant information for decision-making. For example, context-aware data
processing can deliver relevant information to a consumer by knowing the consumer’s current
location.

IoT adoption inhibitors

1. Security
IoT security standards are required protect against cybercrime and national security threats,
and help to ensure that the system is trustworthy and trusted [UKGOV]. As IoT is built on top of
the Internet, security issues pertaining to the Internet also apply in IoT [LEE]. As the number of
connected IoT devices increases, the number of the end-to-end connection points increases
exponentially and the number of potential security holes increases in proportion. Protecting the IoT
is however, a complex and difficult task. The number of attack vectors available to malicious
attackers might increase exponentially, due to global connectivity (“access anyone”) and
accessibility (“access anyhow, anytime”) of the IoT [LEE]. IoT data and devices must therefore
have security built-in by default.

2. Energy Efficiency
IoT devices should support energy efficiency, as this will help increase the range of
potential applications and manage the burden on energy supply generated by the potential billions
of IoT devices.

3. Data Quality
The quality of data obtained from the IoT is crucial, in order to gain user engagement and
acceptance of the IoT paradigm [LEE]. If data quality is poor, trust in the decision based on the
data will be low and the adoption of IoT technologies can be hindered. When data collection cost
is high and the quality of the data is low, the value of the IoT decreases rapidly.

4. Lack of IoT Specific Business Models

To realize the potential benefits, firms need an appropriate measure in order to properly
assess the risks and rewards of the technology (Lee & Lee, 2015). Many IoT projects have unclear
problem definitions and are using emerging technologies, thus causing a higher risk of project
failure than traditional technology projects.
As with the Internet originally, applications of the Internet of Things are developing
organically, with experimentation in every sector. The first source of uncertainty is in the relatively
few established business models for achieving profitability. Until these become clear, it is
reasonable to expect caution from businesses developing and buying IoT technologies. Early
developments come from industry giants with economies of scale that mean large investments for
incremental benefits are viable. However, the existing information technology capability of many
organisations is likely to be insufficient to harness the volume and variety of data generated by the
Internet of Things [UKGOV]. Disruptive new firms therefore have already emerged to break into
existing markets or create new ones through innovative new ideas.

5. Potential of IoT for businesses

Across many industries, the potential of IoT has started to be realised. For example, data
harvested from IoT sensors and devices can be used to enhance the customer experience, optimize
store operations and improve inventory and supply chain management. IoT is driving innovation
and new opportunities by bringing every object, consumer, process and service into the digital
realm (Gregory, 2015). The proliferation of connected devices coupled with improved technology
platforms and adoption of common standards is expected to drive the rapid growth of IoT-enabled
capabilities across industries.

IoT in retail
Analysts predict the IoT retail market to grow 20 percent to $35 billion in the next five
years (Forbes, 2017c). The IoT can have a disruptive impact on the retail industry. Retailers are
already experimenting with ways to use intelligent, connected devices to offer new services,
reshape experiences and enter new markets. This trend is expected to accelerate [IBM]
According to (Forbes, 2017c):
• 78% of retailers acknowledge the importance of integrating online and store customer
experience
• 87% of retailers plan to deploy mobile POS devices in their stores allowing their
customers to self-scan and pay for products by 2021.
• 90% of retailers plan to implement ‘buy online-pay in store’ services by 2021.
 Using IoT sensors to automate many of the labour-intensive operations in stores and
warehouses, such as tracking inventory or changing prices on individual items, frees staff to spend
more time interacting with customers and thus improving their levels of satisfaction. IoT also,
makes it possible for managers to adjust pricing in real-time, using smart tags to lower prices on
promotional or low-turnover items or increase pricing on higher-demand items [IBM]. A fully
integrated pricing system would help retailers improve synchronization of prices between the
shelves and the cash tills/registers and also across all retail channels, to verify prices are consistent
between online and brick and mortar stores.
Inside the store, IoT devices can control environment conditions and reduce cost. For
example, IoT-enabled sensors can control lighting and temperature to both improve customer
comfort and to support more cost-effective energy usage.

IoT in marketing
According to MARKETO (https://uk.marketo.com/infographics/the-marketing-power-of-
the-internet-of-things/) 51% of the world’s top global marketers expect that IoT will revolutionize
the marketing landscape by 2020. The IoT through the inherent interconnectivity of digital devices
provides endless opportunities for brands to listen and respond to the needs of their customers –
with the right message, at the right time. Marketers can extract data from IoT devices in the
customers’ environments (places where they spend their leisure time, shop etc) an interpret them
to shape trends and tailor contact approach with the consumer. The Internet of Things can impact
customer experience, by collecting and analysing more and more diverse consumer data to gain
better customer insight. Thus, combined with Big Data analytics, IoT can be used to [MARKETO]:
• Analyse customer buying habits across different platforms and channels.
• Gather previously unavailable knowledge about how consumers actually interact with
products.
• Gain a deeper insight as to the stage of the buying process a consumer is at each time.
• Provide point of sale (POS) tailored information and advice, such as targeted
advertisements.
• Quickly resolve any issues the consumer may have that stop the closing of the sale.
• Thus, thanks to IoT marketing can become real time and data driven.

The Virtual Value Chain

The VVC was introduced by (Rayport and Sviokla, 1995) as a model to assist in identifying
the value of information and deriving information strategies. The physical value chain (Porter,
1985), identifies the primary and the support activities as a series of actions that convert inputs to
outputs and produce value for businesses and customers. Information technology however, has
changed the way modern organizations operate to create value. The shift now is on information
management. The VVC model suggests five stages, as a way to identify the potential value of
information.
1. Gather information. The advances of the Information and Communication
Technology (ICT) over the years made digitization of information possible. The anticipated
proliferation of the IoT devices will further expand the capabilities of collecting data that could
capture almost all human and business activities along both the VVC and the physical value chain.
For example, smart product labels, using printed electronics technology, will be able to capture
data that can track product identity, identify products’ location and usage from factory floor to high
street, to living room to the recycling back and finally of products’ decomposition into their
components’ materials. Information many also be collected via different sources, such as internal
organizational sources, i.e. corporate databases, or external sources, such as the Internet, and IoT
devices. Thus, issues pertaining to protocols and information integration across technological
platforms need to be addressed.
2. Organise information. Information may be gathered in the form of videos,
images, sounds, texts, numbers, etc. Big Data are characterized by large Volumes, High Velocity
and Volatility as well as of their unstructured nature. All this data and information needs to put in
place and stored in a structured way in databases. This is a challenge, but challenges are to be
tackled through technological advances.
3. Select information. The aim of organisations is to provide quality services to
customers in a profitable manner. The VVC in its third stage suggests that businesses should select
and analyse information that is valuable already, or can be used to create value. Businesses need to
identify what information is potentially valuable from the masses of data and information that is
collected and made available in their databases. Data analytics techniques play an important role in
identifying the valuable pieces of information and analysing them. Advances in artificial
intelligence would make it possible for algorithms to dig in large data sets and surface the important
bits of information. However, what is valuable, to whom and when is subject to various
circumstances, personal choices and subjectivity and luck. Data analytics should be used to analyse
the value of information in the light customers’ personal subjectivity as well as by taking into
consideration the dynamic and constantly changing business and social environment. For example,
the fact that a customer’s preference is identified and it is now known to a company, does not
guarantee that the customer will not react to a competitor’s advertisement positively and change
his/her preferences for at least once. Data mining and machine learning can identify that a customer
is possibly willing to change so a company act in a pre-emptive way, even before the competitor’s
message was sent. That is not always possible though. Machine learning techniques need training.
However, sudden changes in customers’ behaviour do not give enough time to reassess his or her
preferences. Companies rely on real time data that IoT connected devices can capture. Thus, when
the competitor’s advertisement message was sent to the customer a company will have to
opportunity to analyse the new set of data and then react promptly.
4. Synthesise information. Data analysis results should be combined in way that
they form a service, which in turn is expected to be delivered to the customer. Following the
example presented in stage 3, assume there are three sets of relevant information that need to be
combined so that and their synthesis would result valuable decisions, actions and services for both
the company and its customers. First, the competitor’s advertisement carries information pertaining
to product and to customer. Then, data analysis results provide information that indicates the
possibility of the customer to switch to the competitor. Finally, a third set of information pertaining
to how a company should react to retain its customer is synthesised so a company knows when to
react, how and towards which customer.
5. Distribute information. The results of the synthesis should be delivered to the
customer, who is probably using a variety of devices such as a smartphone, smartwatch, smart tooth
brush, smart refrigerator, to name just a few of the billions of IoT smart devices that will be
available in the coming years. The messages should reach the customer in the desired format, at the
right place and time. Thus, no surprise that one of the major trends is to develop technologies that
allow adaptive web design, thus providing more tailored customer experiences from different
devices (Digital marketing Megatrends, 2017).

A study by (Bhatt and Emdad, 2001) indicates how the VVC can be used to uncover
competitive advantage opportunities along the physical value chain, by considering the 4P’s
(Product, Place, Price, Promotion) marketing model. Indeed, with respect to product information,
smart labelling may allow a firm to obtain information from its suppliers electronically, which then
it can be automatically delivered to the customers' IoT devices upon customer request. Moreover,
the VVC provides the possibility of quick customisation of products and services, by allowing
personalisation depending on the products features a customer wishes to activate, the pay and
delivery methods of their products, etc. With regard to the place, just-in-time access to the products
and services implies that customers can consume fulfil their needs at the right time and place, i.e.
exactly when a service is need. Different pricing schemes can also be designed depending on the
situation and circumstances under which a customer demand is expressed or identified by a
company. Assume that smart IoT devices (e.g. a smartwatch, or smart glasses communicate with
the smart label of a competitor’s product. This event can be interpreted as the customer showing
interest for the particular product. Then, advertisement messages from other suppliers can come
through the customers IoT devices combined with competitive price offers. The VVC can promote
products and services by collecting valuable information across the value chain, combine it and
deliver it to the customer. Information related to supplementary or competitive products, their
components, ingredients, origins and methods of production, are examples of such promotion
messages, which can be delivered as mentioned previously at the right time and place, when the
customer exhibited an interest for the product or a competitor took an initiative to address the
customer. With the proliferation of the IoT people and businesses will be strongly connected with
each other, thus providing almost full information of each other activities and intentions.

The Customer Service Life Cycle (CSLC) Model

Earlier versions were introduced by (Burnstine, 1980) and by IBM’s Business Systems
Planning process (IBM Corporation, 1981). They included the following four stages: requirement,
acquisition, ownership and retirement. The CSLC was extended by (Ives & Learmonth, 1984;
Learmonth, 1986) to a 13-stage model that can assist businesses to specify customer oriented
strategic opportunities for IS. Revisions of the CSLC have been proposed by (Ives & Mason, 1990;
Piccoli, et al., 2001; Lightner, 2004). The CSLC is originally found in the literature as “customer
resource life cycle” since the terms “resource” and “service” are used interchangeably (Piccoli, et
al., 2001; Lightner, 2004). According to (Ives and Learmonth 1984), the stages represent a life
cycle which customers follow when acquiring their products. The 13 stages of the CSLC are:
Establish requirements of how much of a product/service is required, Specify the attributes of the
product, Select sources where the customer can buy the product, Order the product, Authorise and
Pay, Acquire, Test and Accept that the acquired product meets customer’s specifications, Integrate
with other products in an inventory, Monitor the use of the product so that unauthorised use is
prohibited, Upgrade the product, Maintain and Repair if needed, Transfer or Dispose, i.e. return,
or dispose of the product inventory if necessary and Account for much is spent on the product (Ives
& Learmonth, 1984). Websites and IoT devices that provide services, which successfully assist
customers in satisfying their needs through the various stages of the CSLC, improve both customer
satisfaction and businesses improve their financial performance (Lightner, 2004; Wade & Nevo,
2005-6; Hoekstra, et al., 2015). The CSLC is adopted in the current study since it provides a suitable
framework that can be used in order to investigate customer oriented services in a systematic way.
This chapter uses the VVC in combination with the CSLC. The VVC is considered in all stages of
the CSCL, in searching valuable information that can be targeted and selected by using IoT devices.
The selected information will indicate what content to promote in digital campaigns. The Customer
Journey Map, is been proposed in digital marketing (Johnston, 2015) as a template to assist in
content management. It suggests an approach to systematically approach the question “what content
to create?” i.e. to specify what content should be sent to the customers. The customer journey is
defined by (Johnston, 2015), as the “things that people want to accomplish as they interact with a
brand”. Then, content the right content is mapped to customers’ goals. The aim of the first step in
the customer journey mapping is to decide the most important customers or market segments to
focus on depending to the business goals. The second step, suggests that the customer state should
be clarified, i.e. knowing information about the customer who and where they are, what are they
preferences and needs. This information may be available when the customer uses an application
or a service on a connected device. During the third step, the goals of the customer are specified.
The question that needs to be answered is “what are the goals of the customer, when interacting
with the company?” Step three, breaks down the goals into tasks that can be mapped to information
needs. Step four specifies, which communication channels customers would probably use, thus
customising the content and their interaction according to the characteristics of the digital media,
e.g. type and features of smartphone, smart label, interactive TV, wearables, etc. Finally, in step
five content and tasks along the customer journeys should be matched to the appropriate channels
and formats that are likely to work best.

The Decision Making and Trial Evaluation Laboratory (DEMATEL) Method

This chapter suggests that multi-criteria techniques can be used, among other data analytics
techniques such as statistics, machine learning, etc., in order to model business priorities expressed
in terms of the Key Performance Indicators (KPIs). Multi-criteria methods can be utilised within
the context of the digital marketing for they can model and investigate the knowledge as well as
the subjective views and perceptions of businesses decision makers as well as their customers.
There are several multi-criteria methods such as the Analytic Hierarchy Process (AHP), and its
fuzzy variation (Fuzzy AHP), the DEMATEL, etc.
The DEMATEL method was developed by the Battelle Geneva Institute (Fontela and
Gabus, 1976). DEMATEL is a multi-criteria method which is used to model and analyze complex
relationships among factors pertaining to a particular domain. DEMATEL is used to calculate the
importance of criteria pertaining to a problem domain. The method is also applied to real life
problems in order to investigate the interactions among important criteria and examine how
interrelated factors affect each other. The method can equally handle qualitative and quantitative
factors. The DEMATEL causal models provide the means to simulate and evaluate alternative
actions and policies.
The DEMATEL method has been extensively used in MCDM problems such as in
formulating marketing strategies, evaluating e-learning services, developing control systems,
dealing with safety problems, examining environment watershed plans, etc. (Chang and Chen,
2008; Chang and Chen, 2009; Chen et al., 2010; Hori and Shimizu, 1999; Lin et al., 2010a, b; Lin
and Tzeng, 2009; Liou, et al., 2007; Ou Yang, et al., 2008; Tzeng, et al., 2007). The DEMATEL is
implemented as follows:
1: Generate the Direct Relation Matrix. The direct relation matrix is calculated usually
based on experts’ responses. However, data can be collected regarding customers’ behaviour,
preferences, etc. Data that is collected (e.g. experts’ views) indicate the level of influence that a
factor exerts on another by using the following scale: 0 for no influence, 1 for somewhat influence,
2 for medium influence, 3 for high influence and 4 for very high influence. The direct relation
A = [a ]i, j a
matrix is a nxn matrix, where i , j indicates the degree to which factor (i) affects factor
(j). In the case of a group of experts, all responses are averaged to produce the average matrix Z,
where � = #�%,' (, with i,j indicating performance criteria.
2: Calculate the normalized initial Direct- relation matrix D. The Matrix D is calculated
using the following formulas.
� = � ∗ �,
1 1
Where � = min [234 ∑; 67 : , 234 ∑; (7 )], where 1 ≤ � ≤ � ��� 1 ≤ � ≤ �.
9<= 8,9 8<= 8,9
3: Derive the Total relation matrix T.
Where � = �(� − �)K1 .
4: Calculate the sums of rows and columns of matrix T. The sum of rows is calculated by
� = �% [�, �]M41 = ∑M'O1 �(�, �) and the Sum of columns is calculated by� = �' [�, �]14M = ∑M%O1 �(�, �).
The value of r(i) indicates the total effect given by criterion (i) both directly and indirectly. The
value of c(j) shows the total effect received by criterion (j) both directly and indirectly. If (j = i),
the value of (ri+ci) represents the total effects both given and received by factor (i), while the value
of (ri-ci) shows the net contribution by factor (i) on the system. If (ri-ci) is positive, factor (i) is a
net cause, while If (ri-ci) is negative, factor (i) is a net receiver.
5: Set a threshold value (α). The threshold is calculated with the formula,
∑; (Q8,9) ∑; (Q8,9)
� = 8<= MR 9<= , where n is the number of criteria. The threshold is used to cut-off the most
important criteria, which will be included in the DEMATEL causal model.
6: Build the DEMATEL causal model. A cause and effect relationship diagram, by
mapping all coordinate sets of (�% + �% , �% − �% ). The causal model indicates the importance of the
most important criteria, i.e. those above the threshold and the degree of influence among criteria.

A framework for analysing the impact of IoT on digital marketing

Assume that IoT devices constitute a graph where the graph nodes represent variables
carrying data and information that reflect important aspects of human and business every day
activities. The CSLC provides a structured way to identify the customer data and the customer
activities with business interest. The IoT connected devices can be used (if they exist already, or
new IoT devices can be designed and developed), in order to collect the required customer data.
Subsequently, businesses within the digital marketing context can decide how to approach the
customer in a customised way, i.e. what content to advertise, when, where, through which devices,
with the appropriate digital media mix. The arcs that connect the nodes represent how variables
interact with each other. The anticipated proliferation of the IoT will expand the IoT devices graph,
thus expanding the network of information sources, across time, space, activities, personal priorities
and business policies.

Figure 1. The IoT graph

Figure 1, shows that the “smart heating device” communicates with other IoT devices,
collects the necessary data and then optimises the heating parameters, by applying some rules that
reflect personal preferences regarding the heating settings. Considering the vast quantities and the
volatile nature of big data, it is a difficult task to identify what is perceived as valuable data and
what is not.
Distinguishing what data is valuable and what is not, is to some extent a subjective matter that
depends on personal customer choices or on different business priorities in a commercial setting.
Therefore, in the context of digital marketing, where information and knowledge regarding
customer’s needs and preferences is of paramount importance, a framework needs to cater for
identifying what is significant, when, where to whom and under which circumstances. The paths
across the IoT graph constitute alternative VVC (Rayport and Sviokla, 1995). Alternative paths
imply that several VVC may be hidden but yet still to be identified. Such VVC are formed and
depend on the possibilities for information dissemination across customers, suppliers,
intermediaries, etc. Some VVC may be long lasting some other may last for some time depending
on the circumstances, e.g. the amount of value they create, the period of time they are conceived as
useful, etc. Data that is collected along the VVC should be analysed. This data is classified as
relevant to one or more stages of the CSLC model, thus specifying the customer need. It is
important to know the aims of the customer, for the digital message can be better focused when
targeting customer related KPIs more effectively. Thus, the proposed framework in this chapter,
suggests that data analytics techniques should be utilized in order to identify pieces of information
that are valuable along both the VVC and the CSLC and to measure how valuable such data is for
a customer or a business. The plethora of possible VVC that can be created and identified within
the IoT graph and the potential of data analytics can provide fast and effective means of data
management in order to be able to answer questions such as: Who are the potential customers?
When is the right time for each of the customers? The data analysis results will lead to the design,
implementation and distribution of e-services across the VVC that are offered to targeted customers
that are really interested. Such e-services should be made available at the right time and place,
along the VVC. In the context of the digital marketing an e-service consumer is a potential customer
who is receiving advertising messages through electronic means, most probable but not necessarily,
through the internet. An e-service provider is the digital marketing agent whose interest is to
identify the potential customer, at the right time, when the customer is more than ready to find the
advertisement interesting and proceed with purchasing the promoted services or the goods.
Thus, the proposed framework suggests that the CSLC and the VVC are used to identify
the “what” should be offered to a customer. Process modelling, indicates the “where” and “when”,
i.e. the place and time a service is made available to a customer. Process modelling is applied in
order to capture the operational aspects that link e-service providers and service consumers as well
as to investigate the potential of the IoT devices. Data analytics techniques are used to unravel and
evaluate the value that is generated for the customer.
Figure 2. Depicts the proposed framework for utilizing IoT and data analytics in digital marketing.

As figure 2 shows, the proposed framework suggests that the IoT provides the means for
data collection across VVC. Use cases reflect the value propositions, i.e. the outcome that
customers, suppliers or other participants would anticipate from the VVC. Business processes
modelling is utilized to capture the functionality across each VVC. It reflects the information flow
that consists of a series of potentially useful information elements, as well as it shows the operations
that take place in space and time so that the required information will be collected and analysed in
order to produce the requested use cases. Data analytics techniques will be utilized to analyse and
manage the available data and help identify which customer or supplier to focus on, where and
when to act and deliver an e-service, e.g. an advertising message. The evaluation of the advertising
value has been the focus of several research works. This study considers digital marketing KPIs,
draws on (Ducoffe, 1995) and (Martins, et al., 2017) that suggest informativeness, irritation and
entertainment as the antecedents of the advertising value. The informativeness, refers to the ability
provide the customers with information related to products. The irritation implies techniques that
employed by advertisers that annoy, or manipulate consumers. Consequently, such techniques are
perceived as unwanted. Next, the entertainment refers to pleasant advertising, which results to
positive impacts on brands. The proposed framework suggests that the following steps should be
implemented.
 Step 1: Define the digital marketing KPIs. Successful advertising depends on its ability to
persuade the customer, present relevant information and invoke reasoning and emphasis (Sheehan,
2014). Although the establishment of KPIs is highly subjective depending on the strategy and vision
of each company, measures such as purchase intention, customer retention, customer experience,
as top priorities. Digital marketing specific KPIs typically include the number of clicks, conversion
rate, click through rate (CTR), etc.
Step 2: Specify Uses Case(s). The uses cases indicate the customer requirements, in terms
of quantity and quality. Uses cases may involve tasks such as purchasing a product, consuming a
service, seeking information or advice for a product, completing a task. Use cases represent the
informativeness antecedent according to (Ducoffe, 1995; Martins, et al., 2017), by specifying what
are the important data elements that customers want to know. The successful implementation of
each use case depends among others on the accessibility and processing of the available information
that pertain to the use case. Thus, this framework defines a uses case in terms of the data elements
that are necessary for the successful completion of the uses case. Thus,
UCi = [UCdei1 ,UCdei 2 ,...UCdein ] , where UC i represents the use case, and UCdein represent the
n-th data element required for the UC i completion. For example, the use case “want to buy a
theatre ticket”, involves several tasks and requires the processing of several data elements that can
be identified by using the CSLC stages. Therefore, one or more CSLC stages, such as “specify the
number of the tickets, “specify the seat in the theatre, “select which theatre to go”, etc., may be part
of the implementation of the theatre ticket use case. Each one of the stages requires a number of
data elements, such as ticket price, availability of seats, theatre play reviews, theatre location, etc.
Step 3: Identify Digital Media and digital Formats for transmitting the Advertising
messages. In the IoT era the plethora of smart devices will make it easier to find a communication
channel, but perhaps more demanding to identify the most effective one. Therefore, digital
messages can be transmitted through various channels e.g. smartphones, smartwatches, smart cars
and smart devices of any kind. Messages can be sent in a number of formats, such as text, video,
sound, image. There are five groups of factors that can influence the selection of the most suitable
digital media (Bunt et al., 2007): User-specific features, refer to users’ representation preferences,
e.g. a customer may prefer video to image interfaces. Next are the Information features, which
suggest that digital media exhibit different levels of suitability for displaying data elements. For
example, a restaurant menu is more appropriate to be shown with a video than with an image or
sound. Contextual information pertains to where a customer is located when a message arrives.
Environmental conditions such as noise, light, weather, speed, etc. may deter the quality of the
advertising message. Media constraints imply that a single digital media may not be enough but
instead a combination of media would more effectively deliver the advertising message to the
customer. Finally, Limitations of technical resources refer to device limitations such as screen size,
bandwidth, etc. Each combination of digital media and digital formats is most appropriate for
different data elements, under different circumstances. For example, a voice message would be
more appropriate than a video or text when driving. Thus, by taking into consideration the
information features, and the limitations of smart devices, advertising messages can be defined as
text video sound image format
AdvMsg = [M device 1 , M device1 , M device1 , M device1 ,..., M device( k ) ]
, where “device” indicate a smart
IoT device, which is used to transmit the message by using the selected format.
Step 4: Apply DEMATEL. Although various data analytics techniques can be used, this
chapter illustrates how DEMATEL can be used to analyse data and use the analysis results in digital
marketing. Its application returns the importance weights and the ranking of what data elements to
advertise, how to advertise in terms of the most appropriate devices and formats to use. Since used
cases are associated with CSLC stages, it would also be possible to identify the “when” to advertise,
even before a customer proceeds to an action. In fact, IoT devices transmit data on real time, so
that an advertising agent would know when a customer is starting a specific task. However, with
the use of CSLC and predictive analytics it would be possible to predict the next most probable
action of the customer and deliver advertising messages pre-emptively. A different DEMATEL
model can be used for each Use case and each Customer. Data collected through IoT devices are
fed back to the model so it identifies any changes in customer priorities and behavioural aspects.
Step 5: Utilise the DEMATEL causal model to develop scenarios. The causal model
indicates how concepts considered in DEMATEL affect each other. It provides therefore the means
for evaluating alternative scenarios, for different customers, circumstances, business priorities, etc.

Illustrative Example
Assume a customer use case “customer wants to drink coffee”. The use of the CSLC in
order to investigate further the use case, identifies a number of subsequent tasks and decisions
related to the use case. Their successful implementation depends on identifying, selecting and
analysing relevant data. The use of the CSLC specifies a number of requirements. For simplicity
reasons, table 1 shows only the first three stages of the CSLC. Thus, drawing on the 1st CSLC stage,
table 1 shows that a customer needs to specify the quality, i.e. the number of cups of coffee he or
she would like to consume, as part of the fulfilment of the above mentioned use case. Subsequently,
opportunities for digital marketing messages that relate to the 1st stage of the CSLC are specified.
Such messages may refer to alternative coffee brands, drinks, etc. Whether advertising is worth the
effort or not is a decision that depends on relevant data analysis. Such data elements may include
coffee consumption, coffee brand, time of having coffee, etc. This data can be collected from
potentially thousands and millions of customers by using IoT devices such as customers’ smart
coffee machines, smart labelling on coffee packages, etc. The data and the results of its analysis
may be used to update the customers’ profile, in a Customer Relationship Management (CRM)
system in real time manner, and help identify the customers who would most likely respond to an
advertising message. Furthermore, tracking changes in the customers’ habits and behaviour would
be made possible, by collecting and analysing real time data with the use of IoT connected devices.
Finally, the advertising message will be delivered to the target customers via IoT devices e.g. smart
phones, smart coffee machines, smart cars, etc. as shown in table 1. More data then can be collected
as feedback, in order to assess the effectiveness of advertising on different IoT devices, the response
of customers, etc.

Table 1. Linking Use Cases and the CSLC with Digital Advertising

Use Case CSLC stages Digital Advertising Data Digital

Opportunity elements Media
involved
How many cups Advertise: Ø Quantity Ø Mobile
to consume? Ø Alternative coffee of coffee phone
Specify quantity brands; consumed Ø Smart
(Stage 1 in Ø Alternative drinks; per week coffee
CSLC)
Ø Supplementary goods machine
to coffee Ø Smart car

Where to drink Advertise: Ø Services Ø Mobile

“Customer the coffee? Ø Alternative coffee offered in phone
wants to (Stage 2 in shops; alternative Ø Smart
drink coffee” CSLC) Ø Alternative ways of coffee coffee
transportation, e.g. taxi shops machine
firms; (music, Ø Smart car
Ø Parking places view, etc;
 Ø User
reviews
for shops
Ø Visits per
shop
Ø Time spent
in shop,
etc

How would the Advertise: Ø Quantity Ø Mobile

customer like the Ø Alternative coffee of coffee phone
coffee flavours; types of consumed Ø Smart
(Stage 3 in coffee. per coffee
CSLC)
week/type machine
of drink

Assume a company considers purchase intention, customer retention, conversion rate and CTR as
its major digital marketing KPIs. A digital marketing agent is employed to advertise alternative
coffee shops (see stage 2, in table 1), drawing on data such as quantity of coffee consumption, music
and view quality of alternative shops. Customers’ smart phones and smart watches will be used to
deliver sound or text advertising messages.

Figure 3. The T-matrix

The application of the DEMATEL calculates the inter-relationships among the digital
media and data elements shown in table 1, as well as with digital marketing KPIs that reflect the
business priorities. After performing the DEMATEL calculations, in step3, the T-matrix is
produced as shown in Figure 3. For simplicity reasons not all digital media, or data elements etc.,
as shown. The threshold value (a=0,08) is calculated, using the formula in DEMATEL step5. The
T-matrix shows how the concepts used in DEMATEL interact and affect each other. Values
shadowed in green are below threshold thus are omitted from further consideration. According to
DEMATEL, values below the threshold are not considered as important. Figure 3, shows that
“Purchase Intention” does not affect any of the other variables, but it is affected by all others. The
T-matrix shows that customer retention affects purchase intention, conversion rate and CTR almost
equally. From the digital marketing agent’s point of view, the question is what to advertise and
which digital media to use. The T-matrix shows that displaying an image on customer’s smart
watch has no impact on customers’ perception regarding the music played in shop-1. So, it is
considered as a media of no advertising interest in this case. Instead, it could possibly be used as
an advertising means for increasing coffee consumption or improving customers’ appreciation of
shop’s 1 view. However, since its value of impact is below the threshold (a), this slight impact is
considered as not important. Thus, displaying an image on the customers’ smart watch it does not
look like a good option in this case. Smart phone instead, exhibits some impact (although above
threshold, still not very strong) for increasing weekly coffee consumption. Thus, the suggestion
that advertising on smart phone by delivering a sound message to the customer is a preferable way
to increase coffee consumption. The results also show that increasing the coffee consumption will
result into improvements in all digital marketing KPIs. Additionally, although there is some
indication that using a smart phone to send a sound advertising message will have a positive effect
on promoting shop’s 1music collection, this is below threshold. Thus, other IoT devices and digital
means could also be considered to advertise the music or view, etc. Indeed, by expanding the
DEMATEL Z matrix more KPIs, digital media and data elements will be considered thus, covering
more CSLC stages and more aspects related to the use case in focus. Looking further into in data
analysis, by calculating the (ri+ci) in DEMATEL step4 results in figure 4, show that the most
important KPI to consider is the customer retention, which has the highest value, followed by
purchase intention and CTR.
Figure 4. Ranking of KPIs and Digital Media

Similarly, by comparing the digital media, the findings indicate the higher importance of
sending sound message on customers’ smart phones. However, not all customers have the same
requirements, even for the same use case. Customer priorities can be modelled as a vector,
considering the same concepts as in the DEMATEL matrix. Assume that a customer is particularly
interested in music. So in terms of advertising promoting music to that customer is a priority.
Suppose the following vector, in figure 5, represents customer’s requirements. A higher priority for
music than the view is shown with the corresponding degrees of interest.

Figure 5. Customer’s Vector show higher interest in shop’s-1 music than in view.

To deal with such a situation, the multiplication of the customer’s vector with the transposed T-
matrix, will indicate what advertising actions should be taken.
The multiplication result is shown in figure 6.
Figure 6. What advertising actions to take

Results indicate that smartphone sound message should be sent to the customer in order to
promote the music and an image to the customer’s smartwatch to promote the view. However, both
suggestion have a low (0,04) and (0,008) degrees of impact, indicating that other media should be
explored. Similarly, several scenarios can be examined that involve customers with different
priorities, a variety of digital media, various formats and different sets of KPIs.

Therefore, this chapter suggests that the use of IoT can provide the means for collecting
real time data regarding customers’ and other digital marketing stakeholders. Data analytics, such
as the application of DEMATEL, allow agents to reach useful conclusions regarding the planning
and evaluating different scenarios of the advertising mix, digital media and customer targets.

FUTURE RESEARCH DIRECTIONS

The digital transformation appears on many lists of strategic priorities. It requires though
profound changes from the foundation of business processes, up to changing the mentality, and
vision. Digital marketing is an important and flourishing business area, which attracts increasing
attention in both academia and in businesses. Future research may focus on investigating how
different types of customers react to digital advertising. As the IoT will be tracking human activities
it would interesting to examine issues pertaining to how to judge and interpret customers’
experience and mood. Time consideration also includes topics of interest since VVCs evolve and
change over time. It is important to be able to predict such changes and translate them into stage
models, such as the CSLC, that can more accurately predict the next actions of customers. The
combination of data analytics and IoT could redefine the KPIs currently in use and lead to the
development of performance models that can improve marketing accountability and assist in
evaluating advertising investments.

CONCLUSION
By exploiting the (loT) technologies, more and more devices will be able to collect data
regarding human activities; this data can be stored with cloud services and share it among other
IoT smart devices. Data analytics can be used to dig in selected data, analyse it and shed light in to
the problem domains, produce reports and create results. There is a wide range of domains and
applications where IoT can be used. This chapter focus on digital marketing.
The IoT era is expected to revolutionise the way marketing and in particular digital marketing is
approaching the customer. The development (IoT) technologies and services provide the means for
integrating information gathering and processing across the value chain of digital marketing. The
collection of valuable information pertaining to all stakeholders (customers, suppliers, and digital
marketing agencies), will facilitate the development of new bread (e)services that allow
stakeholders to communicate with each other more by understanding each others needs more timely
and effectively.
Analysis of data collected from IoT devices will be key to any company’s successful IoT
strategy. Early research has already shown that IoT success relies on strategic sharing of data and
quick analysis of its implications (Borowski, 2017). By 2020, IoT devices will outnumber humans
with a ratio of 4-to-1, creating new dynamics for marketing, sales and customer service” (Borowski,
2017). Digital marketing agents, already have access to data platforms that track clicks,
impressions, many other metrics related to web analytics. The IoT though, will turn almost every
device into a smart device with sensors collecting and transmitting data, smart labels,
communicating with other smart devices. The vast amount of data that will be made available to
the businesses hide numerous VVCs that need to be identified activate them. Data analytics
advances are already in places to materialise and monetise investments in IoT and data analysis.
IoT and data analytics will usher digital marketing into realising then untapped potential of data.

The purpose of this chapter is to explore the role of IoT in digital marketing, and to propose
a framework that can be used to analyse IoT uses, opportunities and challenges in a systematic and
quantify its impact. The proposed framework suggests that KPIs should be defined and integrated
into VVCs and business processes, so that special and time related information about the customers’
requirements is collected and used for advertising purposed and beyond, such as in services design,
customisation, etc. The combination of the DEMATEL within the context of the CSLC and the
concept of VVCs, provides an approach to digital marketing that allows agents to identify what
product and services features attract customers attention more and start collecting data about these
features by utilising the IoT. The proposed approach allows digital marketing agents to assess the
effectiveness of their actions and plans in terms of KPIs, such customer retention, purchase
intention, etc., in a way that the KPIs reflect their needs and their strategic priorities.

This study illustrates how MCDM methods such as the DEMATEL, can be used in order
to utilise the potential of the IoT. Such techniques allow the decision makers to define their own
priorities and evaluate alternatives. The integration of recent advances in other data analytics fields
such as the machine learning, data mining, etc. with MCDM methods can provide digital marketing
agents with powerful tools to analyse customer trends, plan for their digital marketing mix and
assess their business strategies in the light of real time information flows. Further, AI and machine
learning can feed MCDM methods with detailed and up-to-date customer patterns and establish a
process of data collection, analysis and assessment with real time data.
REFERENCES
Bhatt, G. D., & Emdad, A. F. (2001). An analysis of the virtual value chain in electronic
commerce. Logistics Information Management, 14(1/2), 78-85.
Borowski, C. (2017). How to prepare for IoT changing customer service. Retrieved from
https://www.softwareadvice.com/resources/iot-customer-service/
Bunt, A., Carenini, G., & Conati, C. (2007). Adaptive content presentation for the web. In P.
Brusilovsky, A. Kobsa, & W. Nejdl (Eds.), The adaptive web. Lecture notes in computer science,
4321, 409-432. Berlin/Heidelberg: Springer. ISBN: 978-3-540-72078-2.
Burnstine, D.C. (1980). BIAIT: an emerging management engineering discipline. BIAIT Int., Inc.,
Petersburg, NY.
Chaffey, D., & Patron, M. (2012). From web analytics to digital marketing optimization:
Increasing the commercial value of digital analytics. Journal of Direct, Data and Digital Marketing
Practice, 14(1), 30–45.
Chang, H. H., & Chen, S. W. (2008). The impact of online store environment cues on purchase
intention: Trust and perceived risk as a mediator. Online Information Review, 32(6), 818-841.
Chang, H. H., & Chen, S. W. (2009). Consumer perception of interface quality, security, and
loyalty in electronic commerce. Information and Management, 46(7), 411-417.
Chen, V. Y., Lien, H. P., Liu, C. H., Liou, J. J., Tzeng, G. H., & Yang, L. S. (2011). Fuzzy MCDM
approach for selecting the best environment-watershed plan. Applied soft computing, 11(1), 265-
275.
Digital Advertising. (2015). Retrieved from https://www.statista.com/statistics/237974/online-
advertising-spending-worldwide/
Digital marketing Megatrends (2017). Digital marketing Megatrends: 9 digital marketing and
martech megatrend. Retrieved from https://www.smartinsights.com/managing-digital-
marketing/marketing-innovation/digital-marketing-trends-2016-2017/
Ducoffe, R. (1995). How consumers assess the value of advertising. Journal of Current Issues &
Research in Advertising, 17(1), 1-18.
Fontela, E., & Gabus A. (1976). The DEMATEL observer.
Forbes. (2017a). US Digital Marketing Spend Will Near $120 Billion By 2021. Retrieved from
https://www.forbes.com/sites/forrester/2017/01/26/us-digital-marketing-spend-will-near-120-
billion-by-2021/ - 401bc054278b.
Forbes. (2017b). Internet of Things Market to Reach $267B by 2020. Retrieved from
https://www.forbes.com/sites/louiscolumbus/2017/01/29/internet-of-things-market-to-reach-
267b-by-2020/ - 210fc901609b.
Forbes. (2017c) Louis Columbus. Internet of Things will revolutionize retail. Last viewed,
19/3/2017.
Gartner (2017). Retrieved from https://www.gartner.com/newsroom/id/3598917.
Godes, D., & Silva, J. C. (2012). Sequential and temporal dynamics of online opinion. Marketing
Science, 31(3), 448–473.
Gregory, J. (2015). The Internet of Things: revolutionizing the retail industry. Accenture Strategy.
Hoekstra, J. C., Huizingh, E. K., Bijmolt, T. H., & Krawczyk, A. C. (2015). Providing Information
and Enabling Transactions: Which Website Function Is More Important for Success? Journal of
Electronic Commerce Research, 16(2), 81-94.
Hori, S. & Shimizu, Y. (1999). Designing methods of human interface for supervisory control
systems. Control Engineering Practice, 7(11), 1413-1419.
IBM (2016). Steve Laughlin. The Future of Retail is Consumer-Connected and Cognitive. IBM
Report. January 18.
IBM Corporation (1981). Business Systems Planning: Information systems planning guide. GE20-
0527-3, IBM Corp., Armonk, NY, July.
IBM Institute for Business Value (2011). From stretched to strengthened – Insights from the Global
Chief Marketing Officer Study. IBM CMO C-suite Studies.
Ives, B., & Learmonth, G.P. (1984). The information system as a competitive weapon.
Communications of the ACM, 27(12), 1193-1201.
Ives, B., & Mason, R. O. (1990). Can information technology revitalize your customer service? The
Executive, 4(4), 52-69.
Johnston M.R. (2015). Wonder What Content to Create? Try a Customer-Journey Map. Retrieved
from http://contentmarketinginstitute.com/2015/11/customer-journey-map-template/
Kee, R. (2008). The sufficiency of product and variable costs for production-related decisions when
economies of scope are present. International Journal of Production Economics, 114(2), 682–696.
Kumar, R., Balaji, M.S., Sadeque, S., Nguyen, B., Melevar, T. (2017). Constituents and
consequences of smart customer experience in retailing. Technological Forecasting and Social
Change, 124, (257-270).
Learmonth, G. P. (1986). Information system technology can improve customer service. ACM
SIGMIS Database, 18(2), 6-10.
Lee, Inn, (ed.) (2017). The Internet of Things in the Modern Business Environment. IGI Global,
March.
Leeflang P., Verhoef, P., Dahlström, P., and Freundt, T. (2014). Challenges and solutions for
marketing in a digital era. European Management Journal, 32(1), 1-12.
Lightner, N. (2004). Evaluating e-commerce functionality with a focus on customer service.
Communications of the ACM, 47(10), 88-92.
Lin, C. L. and Tzeng, G. H. (2009). A value-created system of science (technology) park by using
DEMATEL. Expert Systems with Applications, 36(6), 9683-9697.
Lin, C. L., Chen, C. W. & Tzeng, G. H. (2010a). Planning the development strategy for the mobile
communication package based on consumers' choice preferences. Expert Systems with
Applications, 37(7), 4749-4760.
Lin, C. L., Hsieh, M. S. & Tzeng, G. H. (2010b). Evaluating vehicle telematics system by using a
novel MCDM techniques with dependence and feedback. Expert Systems with Applications,
37(10), 6723-6736.
Liou, J. J. H., Tzeng, G. H. & Chang, H. C. (2007). Airline safety measurement using a hybrid
model. Journal of Air Transport Management, 13(4), 243-249.
MARKETO, (2017). The Marketing Power of the Internet of Things. Retrieved from
https://uk.marketo.com/infographics/the-marketing-power-of-the-internet-of-things/
Martins, J., Costa, C., Oliveira, T., Gonçalves, R., & Branco, F. (in press). How smartphone
advertising influences consumers' purchase intention. Journal of Business Research.
Ou Yang, Y. P., Shieh, H. M., Leu, J. D. & Tzeng, G. H. (2008). A novel hybrid MCDM model
combined with DEMATEL and ANP with applications. International Journal of Operations
Research, 5(3), 160-168.
Piccoli, G., Spalding, B.R., & Ives, B. (2001). The Customer Resource Life Cycle: A framework
for improving customer service through information technology. Cornell Hotel and Restaurant
Administration Quarterly, 42(38), 38-45.
Porter, M. (1985). Competitive Advantage. The Free Press. New York.
Rayport J.F., & Sviokla, J. (1995). Exploiting the Virtual Value Chain. Harvard Business Review,
November–December.
Royle, J., & Laing A. (2014). The digital marketing skills gap: Developing a Digital Marketer
Model for the communication industries. International Journal of Information Management, 34(2),
65-73.
Sheehan, K. B. (2014). Controversies in contemporary advertising (2nd ed.). Thousand Oaks, CA:
Sage Publications.
TechNavio (2017). Global Digital Marketing Software Market 2015-2019. Retrieved from
https://www.technavio.com/report/global-digital-marketing-software-market-2015-2019.
Tzeng, G. H., Chiang, C. H., & Li, C. W. (2007). Evaluating intertwined effects in e-learning
programs: A novel hybrid MCDM model based on factor analysis and DEMATEL. Expert Systems
with Applications, 32(4), 1028-1044.
UKGOV. The Internet of Things: making the most of the Second Digital Revolution. A report by
the UK Government Chief Scientific Adviser.
Wade, M. R., & Nevo, S. (2005-6). Development and Validation of a Perceptual Instrument to
Measure E-Commerce Performance. International Journal of Electronic Commerce, 10(2), 123-
146.
Web Technology Surveys (2014). Usage of traffic analysis tools for websites. Retrieved from
http://w3techs.com/technologies/overview/traffic_analysis/all.
Use this…

R+C R-C
Purchase
1,602494536 -1,602494536
Intention
Customer
2,376009423 -0,891069228
Retention

Convertion Rate 2,032093915 -1,223000967

CTR 2,010876612 -0,611401722

Quantity of
1,62496571 0,577287293
coffee per week

Music played in
1,323452612 1,296546783
shop-1
View in shop-1 0,815648991 0,797711771

Smartphone
1,131253464 1,131253464
sound
Smartwatch
0,525167142 0,525167142
image

Xc vx