Page 1 of 12
Archives of Business Research – Vol. 9, No. 6
Publication Date: June 25, 2021
DOI:10.14738/abr.96.10291. Angelopoulos, M. K., & Pollalis, Y. A. (2021). Digital Transformation: From Data Analytics to Customer Solutions. A Framework of
Types, Techniques and Tools. Archives of Business Research, 9(6). 26-37.
Services for Science and Education – United Kingdom
Digital Transformation: From Data Analytics to Customer
Solutions. A Framework of Types, Techniques and Tools
Michail K. Angelopoulos
Department of Economics, University of Piraeus, Greece
Yannis A. Pollalis
Department of Economics, University of Piraeus, Greece
ABSTRACT
It has become clear by now that the digital transformation has an obvious, lasting
impact as much on the economic systems and commercial players as on the lives of
individuals and on society at large. The decisions we make, our actions, even our
existence in the digital world result in the production of massive amounts of data.
These data can be integrated into large data analysis ecosystems and contribute
positively to the revision of current business models and practices. Machine
learning algorithms combined with the suitable tools, such as Python, turn raw data
into useful information and lead to critical and correct decisions. The aim of this
paper is to present a review of current popular and useful data analytics techniques
and tools that lead to custom solutions for both customer and business. The most
famous techniques based on Machine learning and visualization tools are
represented here.
Keywords: Big Data, Machine Learning, Data Analytics, Data-driven solutions, Digital
transformation.
INTRODUCTION
There is no doubt, we are going through the era of the digital world which brings new challenges
and opportunities for science, economy, and society. Digitalization and new innovative
technologies such as cloud computing, artificial intelligence, big data and the Internet of Things
are dominating and transforming the world we live in. It is now obvious that digital
transformation has a significant impact on customer experience as well as on the profitability
and survivability of an organization [1]. However, there is a lack of understanding of the critical
success factors of digital transformation that are crucial for improving customer experience [2].
Digital transformation is the organizational, operational and business models change of an
organization due to the adoption of a mixture of digital technologies for radically improving its
performance and its integration in our digital world [3]. Nowadays, digital transformation is
growing rapidly and as a result numerous organizations have been investing in digital
transformation in today’s dynamic environment [4]. Digitalization reduces the cost of
interaction and exchange of information and therefore as the volume of exchanges increases
with the use of appropriate methods new opportunities and potential benefits are created.
Regarding the organizations the benefits are manifold; it improves organizational processes,
provides better collaboration with customers, improves the quality and increases the range of
services provided, reduces the cost of products and services, gains competitive advantages,
Page 2 of 12
27
Angelopoulos, M. K., & Pollalis, Y. A. (2021). Digital Transformation: From Data Analytics to Customer Solutions. A Framework of Types, Techniques
and Tools. Archives of Business Research, 9(6). 26-37.
URL: http://dx.doi.org/10.14738/abr.96.10291
provides new opportunities to engage with new markets and improves the customer
experience. On the other hand, digital transformation brings greater transparency, less
information asymmetry and new benefits for the customer, such as innovative products and
services, greater convenience, more choices, new opportunities and experiences as well as
lower prices [2, 5].
DATA ANALYTICS
The digitalization process and its results in the 21st century as well as the analysis of the data
produced through various methods and techniques accelerate the finding of custom solutions
for both customers and organizations. Big data, business analytics, and “smart” environments
have attracted great attention over the past few years to assist organizations make data- informed decisions [6]. Organizations emphasize methods and techniques that will give
purpose and value to their data, receiving decisive answers and leading them to critical
conclusions that will increase their performance [7-8]. These processes use statistical analysis
techniques, such as clustering and regression, which are applied to extended datasets with the
help of new tools (e.g. Hadoop, NoSQL database, Tableau, etc.) [9].
Nowadays, Big data could also be one the foremost significant technological disruptions in
business and academic ecosystems. As the label itself shows, big data refers to large volumes of
data that are created in numerous ways and made available online and in digital media
ecosystems. These involve daily transactions, posts made on social media and with the
integration of various sensors in numerous objects and systems (e.g., smart devices, cars, etc.).
Integrating this generated data into large data analysis ecosystems can be an important factor
in reviewing current business models and practices. These days, many organizations have
recognized the multiple benefits of large-scale data collection and the opportunities it provides
[7]. But it is especially important to point out that it is not enough just to collect and store large
volumes of data, it is also necessary to apply methods and techniques from which we will be
able to make data-informed conclusions and generate personalized solutions.
“Without proper analysis, data are not useful information, it is something that you can exploit
today and something that your competitor may not have yet discovered.”, as said by Charlie
Berger of Oracle Corporation [10]. The ability to analyse large volumes of data at a fast pace
allows organizations to use their data more efficiently and to quickly draw critical conclusions
about any opportunities presented to them as well as potential risks to be avoided. There are a
number of types, techniques and tools used to analyse the big data. To discuss these types,
techniques and tools this paper is organized as follows: Section II provides the types and
explains some common techniques used for data analytics, Section III provides the overview of
tools and platforms of data analytics and the section IV presents the conclusion.
TYPES AND TECHNIQUES OF DATA ANALYTICS
Types of analytics
Companies and organizations want to know what is happening now, what will happen in the
future, and what actions they can take to reach an optimal result. There are four different types
of analytics.
Page 3 of 12
28
Archives of Business Research (ABR) Vol. 9, Issue 6, June-2021
Services for Science and Education – United Kingdom
• Descriptive analytics
Descriptive analytics is focused on the past and its purpose is to answer the question “what has
happened?” [12]. It extracts information from raw data to give valuable insights into the past
[11]. A typical example of Descriptive analytics is corporate reports that simply provide a
historic review of an organization’s operations, sales, financials, customer attrition, and
stakeholders. However, these findings just signal that something is wrong or right, without
explaining why [11, 13]. For this reason, data consultants highly recommend that data-driven
companies are combined with other types of data analytics.
§ Diagnostic analytics
Diagnostic analytics also reports the past but tries to answer questions like “Why did it
happen?” [12]. A goal of diagnostic analysis is to provide more in-depth analysis in order to
answer the questions [14]. It includes processes like data discovery, mining and drill down and
drill through [13]. For this reason, it is also referred to as root cause analysis. It helps
organizations in grasping the reasons of the events which happened in the past and the
cognition relationships among different kinds of data.
§ Predictive analytics
Predictive analytics is the forecasting analytics and is concerned with the future [15]. It
incorporates the descriptive and diagnostic analytics output as well as some Machine Learning
algorithms and simulations techniques. Its purpose is to generate accurate models that predict
the future trends and answers questions like “What will happen?” and “Why will it happen?” in
the future [12]. Predictive analytics help organization in identifying future opportunities and
likely risks by distinguishing specific patterns over the historical data [13]. In predictive
analytics more data means more validated models and therefore more precise predictions.
Some commonly used techniques are data mining and forecasting approaches.
§ Prescriptive analytics
Prescriptive analytics is concerned with the recommendation and guidance and provides
organizations with adaptive, automated and time dependent sequences of operational actions.
Also, it answers questions like “What should be done?” and “Why should it be done?” [12]. It is
purely built on the “what-if” scenarios and its main elements are optimization, simulation, and
evaluation methods [12, 16]. Prescriptive analytics systems provide the organization with the
actionable outcomes and generate comprehensible prescriptions in terms of actions [13-14].
Also, they support feedback mechanisms in terms of tracking the suggested recommendations
occurrence in system’s lifetime [17].
Techniques based on Machine learning
Machine learning algorithms work well with large amounts of data and lead to very precise
predictions through which better decisions are made based only on the input data [18]. They
are learning from preceding calculations and if any corrections are identified, the algorithm
applies the found rules to improve its future decision making [19]. In Figure 1, the results of
KDnuggets poll are provided. The readers were asked about the top Machine Learning
techniques and algorithms that they use for a real-world application.