St. Petersburg University
Graduate School of Management
Master in Management
BIG DATA ANALYTICS AS A MARKETING TOOL:
THE BEST PRACTICES OF RUSSIAN COMPANIES
Master’s Thesis by the 2nd year student
Concentration — Management
Ekaterina A. Artiukhova
Research advisor:
Associate Professor
Maria M. Smirnova
St. Petersburg
2016
ЗАЯВЛЕНИЕ О САМОСТОЯТЕЛЬНОМ ХАРАКТЕРЕ ВЫПОЛНЕНИЯ
ВЫПУСКНОЙ КВАЛИФИКАЦИОННОЙ РАБОТЫ
Я, Артюхова Екатерина Андреевна, студент второго курса магистратуры
направления 38.04.02 «Менеджмент», заявляю, что в моей магистерской диссертации на
тему «Большие данные как инструмент маркетинга: лучшие практики российских
компаний», представленной в службу обеспечения программ магистратуры для
последующей передачи в государственную аттестационную комиссию для публичной
защиты, не содержится элементов плагиата.
Все прямые заимствования из печатных и электронных источников, а также из
защищенных ранее выпускных квалификационных работ, кандидатских и докторских
диссертаций имеют соответствующие ссылки.
Мне известно содержание п. 9.7.1 Правил обучения по основным образовательным
программам высшего и среднего профессионального образования в СПбГУ о том, что
«ВКР выполняется индивидуально каждым студентом под руководством назначенного ему
научного руководителя», и п. 51 Устава федерального государственного бюджетного
образовательного учреждения высшего профессионального образования «СанктПетербургский государственный университет» о том, что «студент подлежит отчислению
из Санкт-Петербургского университета за представление курсовой или выпускной
квалификационной работы, выполненной другим лицом (лицами)».
(Подпись студента)
24.05.2016 (Дата)
STATEMENT ABOUT THE INDEPENDENT CHARACTER OF
THE MASTER THESIS
I, Ekaterina A. Artiukhova, second year master student, program 38.04.02
«Management», state that my master thesis on the topic «Big data analytics: the best practices of
Russian companies», which is presented to the Master Office to be submitted to the Official
Defense Committee for the public defense, does not contain any elements of plagiarism.
All direct borrowings from printed and electronic sources, as well as from master theses,
PhD and doctorate theses which were defended earlier, have appropriate references.
I am aware that according to paragraph 9.7.1. of Guidelines for instruction in major
curriculum programs of higher and secondary professional education at St.Petersburg University
«A master thesis must be completed by each of the degree candidates individually under the
supervision of his or her advisor», and according to paragraph 51 of Charter of the Federal State
Institution of Higher Professional Education Saint-Petersburg State University «a student can be
expelled from St.Petersburg University for submitting of the course or graduation qualification
work developed by other person (persons)».
(Student’s signature)
24.05.2016
(Date)
2
АННОТАЦИЯ
Автор
Название магистерской
диссертации
Факультет
Направление подготовки
Год
Научный руководитель
Описание цели, задач и основных
результатов
Артюхова Екатерина Андреевна
Большие данные как инструмент маркетинга: лучшие
практики российских компаний
Высшая Школа Менеджмента
38.04.02 «Менеджмент»
2016
Смирнова Мария Михайловна
На сегодняшний день анализ больших данных является
одной из самых перспективных инноваций для бизнеса.
Область применения данной технологии в маркетинге
чрезвычайно обширна и широко востребована
компаниями. Являясь частью мирового делового
сообще ства, ряд ро ссийских компаний начал
использовать анализ больших данных для решения
вопросов, связанных с маркетингом. Цель данного
исследования – определить факторы, оказывающие
влияние на практики российских компания в области
анализа больших данных как маркетингового
инструмента, и разработать для них рекомендации. Эта
работа прежде всего принимает во внимание анализ
практик российских компаний и особенности
российского контекста, который должен быть подробно
проанализирован. Ввиду инновационности темы и
исследовательского характера работы основной
методологией является анализ нескольких кейсов. Для
изучения были отобраны и проанализированы практики
четырех крупных российских компаний из сфер
телекоммуникаций и розничной торговли. В ходе
исследования была продемонстрирована высокая
зависимость особенностей практик российских
ко м п а н и я в а н а л и з е б о л ь ш и х д а н н ы х к а к
маркетингового инструмента от внешней среды и
общего уровня развития рынка, а также было выявлено
влияние таких факторов, как наличие организационных
компетенций по управлению большими объемами
данных, инструментов для оценки эффективности
управления технологией и компетентных человеческих
ресурсов. Помимо того, работа содержит рекомендации
для менеджеров о том, как наиболее эффективно
управлять совокупностью различных инструментов
анализа больших данных в сфере маркетинга.
Ключевые слова
Анализ больших данных, маркетинг, инновации в
маркетинге, маркетинговые информационные системы,
б и зн е с а н а л и ти ка , ма рке ти н гова я а н а л и ти ка ,
управление базами данных, маркетинг в России
3
ABSTRACT
Master Student’s Name
Ekaterina A. Artiukhova
Master Thesis Title
Big data analytics: the best practices of Russian
companies
Faculty
Graduate School of Management
Main field of study
38.04.02 «Management»
Year
2016
Academic Advisor’s Name
Maria M. Smirnova
Description of the goal, tasks and
main results
Keywords
Big data analysis is considered to be one of today’s
top business innovations. Marketing applications of
technology are truly diverse and highly demanded by
companies. Being a part of the global business community,
a number of Russian companies also have started to use big
data analysis for solving marketing-related problems. The
aim of this research is to determine the factors which
impact current practices of using big data analytics as a
marketing tool by Russian companies and develop
recommendations for them. This study is focused on the
analysis of practices of Russian companies and peculiarities
of the Russian context, which should be thoroughly
analyzed and taken into account. Due to the innovativeness
of the topic and exploratory nature of research, the main
research method applied in this study is multiple case study
analysis. Four major Russian companies from
telecommunications and retail sectors have been selected
and analyzed. The study has demonstrated high dependency
of big data marketing execution by Russian companies on
the external environment and overall level of big data
market development and revealed the importance of such
factors as existence of organizational data management
competences, performance evaluation metrics and expertise
of human resources. Besides, the managerial implications
of this paper contain recommendations for Russian
companies how to apply effectively a combination of big
data marketing tools.
Big data analysis, marketing, innovations in marketing,
marketing information systems, business analytics,
marketing analytics, marketing in Russia, database
management
4
Table of contents
Table of contents.......................................................................................................................................... 5
Introduction.................................................................................................................................................. 6
Chapter 1. Literature Review........................................................................................................................9
1.1 Overview of contemporary marketing tools....................................................................................... 9
1.2 Big data analysis as a new technology for business.......................................................................... 13
1.3 Big data analysis as a marketing instrument..................................................................................... 23
1.4 Big data analysis as a marketing tool: peculiarities of the Russian context......................................34
Research Gap..........................................................................................................................................40
Summary of Chapter 1............................................................................................................................41
Chapter 2. Research design........................................................................................................................ 43
2.1 Overview of the research methodology............................................................................................ 43
2.2 Justification of the suitability of a case study analysis as a research method.................................... 43
2.3 Overview of the case study analysis................................................................................................. 44
2.4 Data collection procedures............................................................................................................... 46
2.5 Analysis of case study evidence........................................................................................................48
Summary of Chapter 2............................................................................................................................49
Chapter 3. Empirical Research................................................................................................................... 50
3.1 Empirical results of the study........................................................................................................... 50
3.2 Key findings of the empirical research............................................................................................. 69
3.3 Managerial implications of the study................................................................................................ 79
3.4 Limitations of the study and discussion of further research.............................................................. 83
Conclusion................................................................................................................................................. 84
References.................................................................................................................................................. 87
5
Introduction
Today the global business community has become much more concerned with innovation
as a potential growth driver as it was in the past years. Nowadays every company understands
that in order to be truly successful and competitive in today’s rapidly changing business
environment it needs to invest in innovation.
Big data analysis is considered to be one of today’s business innovations with the highest
potential. It has recently gained extremely high interest by the business community all over the
world. Companies are attracted by the variety of managerial implications of big data analysis
across all business functions and industries and promising gains of this technology.
According to statistics (Datameer, 2014; Forbes, 2014), from all applications of big data
analysis more than 50% of problems addressed belong to customer-related problems which
means that marketing applications of technology are highly demanded by companies. Therefore,
the problem of big data analysis as a marketing tool is in line with the most recent technology
concerns of business.
The topic of this master thesis is “Big data analytics as a marketing tool: the best
practices of Russian companies”. The focus of this research study on marketing is justified by
the current market trends and real-life evidence of companies’ interest in marketing-related
applications of the technology.
Being a part of the global business community, a number of Russian companies also have
started to use big data analysis for solving marketing-related problems. This research study is
focused on the analysis of practices of Russian companies and peculiarities of the Russian
context, since we believe that they also have great potential to benefit from these opportunities,
yet specifics of the local market should be thoroughly analyzed and taken into account.
Due to the innovativeness of the topic, the specifics of using big data for marketing
purposes in real-life business environment have not been clearly defined and examined by
researchers as well as by business practitioners.
Applications of technology as a marketing instrument are demonstrated in general terms
in the publications of Arthur (2013), Feinleib (2014), Weber and Henderson (2014), Bacon
(2014), etc. A lot of general information on this issue can also be found in analytical reports and
studies (Forbes, 2014,2015; Oracle, 2014; CIO Online Journal, 2014, 2015; Dietrich, Plachy,
Norton, 2014).
Neither in foreign publications, nor in Russian ones there is a significant number of
thorough and comprehensive research studies conducted on obstacles and barriers of execution
big data analysis as a marketing tool which would be based on real-life cases.
6
Theoretical publications by such authors as Minelli, Chambers, Dhiraj (2013), Arthur
(2013), Stewart (2015) as well as publications prepared by practitioners from McKinsey (2011)
and IBM (Dietrich, Plachy, Norton, 2014) provide a general overview of potential obstacles
connected with big data analysis execution, however they are not examined specifically enough.
However, it is crucial to mention that since foreign companies have started to resort to big
data analysis as marketing tool much earlier than Russian companies, theoretical and practical
studies of foreign researchers in this field still contain more insights and valuable information
than those of Russian authors.
To sum up, the research goal of this study is to determine the factors which impact
current practices of using big data analysis as a marketing tool by Russian companies and
develop recommendations for them.
The research object of this master thesis is the peculiarities of usage and implementation
of big data analysis for marketing purposes by Russian companies.
Due to the innovativeness of the topic and current insufficient level of investigation of
big data marketing in the Russian context by researchers as well as business practitioners this
study is a subject of exploratory research.
Taking into account the exploratory nature of the research, the major research questions,
formulated below, will form a basis for the empirical part of this study, which will therefore
consist of multiple case study analysis:
1. Why Russian companies resort to big data analytics as a marketing tool?
2. How do Russian companies execute big data technology as a marketıng tool?
3. How do Russian companies overcome barriers connected with big data analysis as a
marketing instrument?
4. How can Russian companies leverage the expertise of global market leaders in order
to empower big data analytics for marketing purposes in Russian market?
Besides, it is important to mention that insufficient level of implementation and analysis
of big data marketing by Russian as well as foreign companies puts limitations on the variety of
industries which are investigated in this study. As a result, four Russian companies from two
major market sectors, where large amounts of data are being generated, have been selected for
analysis - telecommunications and retail.
Concerning the structure of this master thesis, the first chapter is dedicated to the
overview of relevant theoretical publications on contemporary marketing instruments, big data as
a new technology for business, big data applications in marketing, global overview of the market
as well analysis of the Russian market and relevant practices of Russian companies in using big
data analysis for marketing purposes.
The second chapter of the study introduces in details the methodology used for
conducting the research. Finally, the third chapter illustrates empirical findings of the research
7
and demonstrates analysis of four case studies of Russian companies, illustrates managerial
implications and limitations of the study.
8
Chapter 1. Literature Review
1.1 Overview of contemporary marketing tools
The first chapter of this study is dedicated to the thematic review of existing theoretical
literature on global managerial practices of big data analysis. Besides, due to the innovativeness
of big data analysis as a technology this review includes analysis of cross-disciplinary
publications and covers various topics from management and information technology in business
to strategic marketing and marketing analytics. A brief overview of the structure of this chapter is
demonstrated in the figure below.
Fig. 1 Structure of the literature review
As it is demonstrated in the figure above, the major objectives of this chapter are to
introduce contemporary marketing tools and provide a perspective on big data analysis as one of
today’s top innovations for business; to demonstrate the variety of managerial applications of big
data analysis with a focus on marketing applications of technology and to analyze the global and
Russian markets of big data and illustrate existing real-life cases of technology implementation.
1.1.1 Marketing strategy
Diverse opportunities which big data analysis opens up for businesses include resolving
of marketing-related issues. The analysis of a wide number of applications of big data as a
marketing tool will be addressed in this chapter.
However, it is important to understand that execution of any marketing strategy is based
on marketing tools and instruments which marketers use to reach their objectives. Let us
9
demonstrate an overview of contemporary marketing tools which are commonly used by
organizations all over the globe.
Marketing’s role in the organization is diverse and it simultaneously encompasses
strategic and tactical decisions which are centered on creating value for customers and building
strong customer relationships to capture value from customers in return. Let us demonstrate
contemporary marketing tools from the point of view of the marketing strategy and refer to
“Principles of Marketing” by Kotler, Armstrong, Harris and Piercy (2013).
Marketing strategy is “the way in which the marketing function organizes its activities to
achieve a profitable growth in sales at a marketing mix level” (Kotler, 1997).
It is also the process of planning a set of marketing actions which helps companies to
create value for customers and build profitable customer relationships. Customer value and
relationships form the kernel of every marketing strategy.
1.1.2 The role of the STP process in marketing
In order to understand which customers a company should serve and which segments
offer the best opportunities the total market is divided into several groups. The process of
dividing a market into distinct groups of buyers who have different characteristics, behaviors and
needs and who might require separate marketing programs and separate products is called
segmentation (Kotler, 1967).
As long as the market segments have been defined, a company can enter one or any of the
identified segments. Market targeting is the process of evaluation of each market segment’s
attractiveness and selection one or more most profitable segments to enter.
After making the targeting decisions an organization should decide on how to
differentiate its market offerings for each targeted segment and what positions a company plans
to occupy in those segments relative to competitors’ offerings. Therefore, positioning is the
process of arranging a clear, distinctive and desirable place for a product in the minds of target
customers so that it will give a product the greatest advantage among competitors from the same
segment. All three processes, mentioned above, are often referred to as STP process which stands
for integrated strategy of segmentation, targeting and positioning decisions. It is also worth
mentioning that effective positioning starts with differentiation - the process of creating
differentiating factors of the company’s market offering so that it gives customers maximum
value.
1.1.3 Marketing mix and 4P’s concept
10
Guided by decisions on the STP process and differentiation, marketers move on to the
next level of marketing strategy development and design an integrated marketing mix composed
of factors under control of the organization and defined as 4P’s concept (Kotler, 1967, McCarthy,
1960, Alderson, 1957). These factors include decisions on product, price, place and promotion.
4P’s concept is a set of tactical marketing tools that the company blends to produce the response
it wants in the target market. Let us consider which marketing tools belong to every category.
Product is the combination of goods and services that the company offers to the target
audience. Marketers have to be concerned with development of variety, quality, design, features,
brand name and packaging while designing a product.
As for the price, which is the amount of money customers must pay to obtain the product,
marketers should develop pricing strategy, discounts, allowances, decide on payment periods and
credit terms.
P l a c e stands for building distribution channels, deciding on coverage, locations,
managing logistics and inventory, etc.
With regards to promotion, which includes all activities which communicate the benefits
of the product and persuade target customers to buy it, marketers execute advertising campaigns,
personal selling, sales promotion and manage public relations.
Recently the model has been expanded to 6P’s concept since two more categories have
been added to this approach: people and performance. People stands for all existing and potential
customers, their characteristics and purchasing power. Therefore, performance stands for overall
performance of the business, strategic and financial goals and unique selling propositions.
From the customer’s point of view 4P’s can be replaced by the model of 4C’s
(Lauterborn, 1990): customer solution (customers needs and wants) describing the product;
customer cost, which stands for the price; convenience, describing the place; and communication
which is equal to promotion activities.
1.1.4 The concept of 4 A’s
Another noteworthy marketing model was developed by Sheth and Sisodia (2012) under
the name of 4A’s concept as an extension to conventional marketing mix concept. The concept
represents a set of tools which help marketers to test whether marketing tactics, generated by
4P’s model, have reached their objectives. The model, linking tactics with real outcomes, is an
approach to see the effects of the company’s actions through the values of customers:
acceptability, affordability, accessibility and awareness.
11
Acceptability, which consists of functional acceptability and psychological acceptability,
shows the extent to which the organization’s total product offering meets and exceeds customer
expectations.
Affordability demonstrates how the current pricing strategy is perceived by customers, It
has two dimensions: economic affordability (ability to pay) and psychological affordability
(willingness to pay).
Accessibility, with its two-dimensional nature (availability a n d convenience), is
concerned with the problem whether customers are able to readily acquire and use the product.
Awareness, which consists of brand awareness and product knowledge, demonstrates the
extent to which customers are informed of product characteristics, persuaded to try it, and, if
applicable, reminded to repurchase it.
As long as the tactics for marketing mix are settled and tested with the help of 4A’s
concept, companies need to pay attention to management of marketing processes.
1.1.5 Marketing management
Marketing management functions are represented by the following processes:
➢ Marketing analysis (analyzing internal and external environment);
➢ Marketing planning (setting strategic marketing goals and objectives);
➢ Marketing implementation (turning marketing strategies into marketing actions to achieve
strategic marketing objectives);
➢ Marketing control (measuring and evaluating results of marketing strategies and plans
and taking corrective actions).
1.1.6 Marketing processes
After having reviewed all processes which form an organization’s marketing strategy, let
us examine marketing instruments from the perspective of marketing processes. Marketing
processes have been defined in different ways, but we have chosen the framework (Webster,
1997) which centers on marketing as defining, developing and delivering value for customers
since we believe in customer centricity of marketing and marketing tools.
Value-defining marketing processes:
➢
➢
➢
➢
Market research;
Analysis of core competences of the company;
Strategic positioning of the organization in the value chain;
Economic analysis of customer use systems.
Value-developing processes:
➢ New product development;
➢ Distribution channels building;
➢ Sourcing strategy development;
12
➢
➢
➢
➢
Vendor selection;
Building of strategic partnerships with service providers;
Pricing strategy development;
Value proposition development.
Value-delivering processes:
➢
➢
➢
➢
Distribution and logistics processes management;
Advertising and sales promotion;
Product upgrades and recalls;
Deployment of the sales force.
To sum up, today marketers have a variety of marketing tools and instruments at their
disposal which assist them throughout the whole process of marketing strategy development,
execution and controlling of marketing management performance. Some of the tools get replaced
or extended by new concepts while others remain to serve as a basis for the marketing processes
planning and execution.
However, today in order to succeed in a rapidly changing and highly competitive business
environment companies often resort to various innovative tools in marketing, one of the most
attractive of which is big data analysis. Not only companies, which extensively use information
technology tools in their everyday business environment and generate extremely large amounts
of information, but also today’s tech-savvy and always-connected consumers altogether create a
tremendously large basis of data which is exponentially increasing overtime. No doubt that
businesses are willing to take advantage from these diverse datasets.
Marketers can derive a variety of valuable insights out from this information with the
help of big data analysis which can be applied in almost every element of a marketing strategy,
starting from segmentation to controlling of marketing performance.
In order to address big data applications as a marketing instrument let us move on to the
introduction of this technology and detailed analysis of its managerial implications.
1.2 Big data analysis as a new technology for business
1.2.1 Evolution of data analysis in business
The invention of Internet about thirty years ago started a new epoch of advanced
information technology which totally changed the way businesses used to operate. During the
digitization of business corporations learned how to take advantage from the computing and
Internet technologies, however the era of big data was just at its doorstep.
Rapid development of information technology as a marketing tool during the 2000-s was
going hand in hand with the growing interest of business in new promising opportunities created
by innovative technology. These innovation-driven marketing approaches included
13
advancements in marketing information systems, customer relationships management, database
marketing, but above all big data marketing. By the end of 2000-s the question of application of
big data analytics in marketing has found its wide niche in scientific research. Starting
approximately from 2010 there has been a major increase in the number of papers dedicated to
this problem. From a variety of scientific investigations a few interesting articles, where big data
phenomenon is addressed, have drawn our attention.
In order to get a better understanding of what defines big data, let us refer to the authors
of “Big Data, Big Analytics: Emerging Business Intelligence and Analytic Trends for Today's
Businesses” Minelli and Chambers (2013). In their opinion, a number of specific factors and
trends triggered the emergence of big data phenomenon back in 2010-es. The most important of
them are listed below:
➢ Tremendous amount of data which has been accumulated within and outside
organizations overtime;
➢ Apart from multiplied volumes of data, increased velocity and variety of data are also
noteworthy which altogether makes data more complicated and cumbersome to be
➢
➢
➢
➢
➢
processed and analyzed;
Emergence of mobile and cloud computing;
Rapid development of social networking;
Long-term downward tendency for technology prices;
Economic feasibility and affordability of storing data and conducting real-time analytics;
Convergence of various technology tools such as analytics software, open-source
technologies, traditional data management and hardware technologies as a new way to
address complex questions of big data analysis.
In addition, the data systems evolution can be described by three main historical
milestones (Minelli, Chambers, 2013):
1. Early stage of data systems development when users did not have a clear vision of what
they want to get from the data they had;
2. Recent years of data systems development when businesses learned how to take the
advantage of analytical platforms and generate consumer insights out of the data;
3. The era of big data where increased collaboration among companies and convergence of
various IT technologies created new promising opportunities for data analysis.
1.2.2 Key definitions of big data analysis
After we have introduced the evolution of data analysis in business and made an
overview of how big data epoch emerged in the market let us define key terms on big data
analysis which are most relevant and useful for business users.
14
Big data: To begin with, the definition of big data varies greatly from publication to
publication. The literature review has demonstrated that “big data” term is used when referring to
a variety of different entities including social phenomenon, information assets, data sets,
analytical techniques, storage technologies, processes and infrastructures. In order to clarify the
most crucial definition of this research paper let us refer to several relevant sources.
A lot of definitions focus on characteristics of the data. To sum them all up, big data is
defined by the data which goes beyond traditional limits in three major dimensions: volume,
variety and velocity. For example, Laney (2001) introduces a framework expressing the 3dimensional increase in data volume, velocity and variety and invokes the need for new formal
practices that will imply “tradeoffs and architectural solutions that involve/impact application
portfolios and business strategy decisions”. Although this work did not mention big data
specifically, the model, later nicknamed as “the 3 V’s”, was associated to the concept of big data
and used as its definition. Many other authors extended the “3 V’s” model and, as a result,
multiple features of big data such as value (Dijcks, 2012) veracity (Schroeck, Shockley, Smart,
Romero-Morales, Tufano,2012), complexity and unstructuredness (Intel,2012), were added to
the list.
A second group of definitions emphasizes the technological needs behind the processing
of large amounts of data. According to Microsoft, big data analysis is focused on applying
“serious computing power” to massive sets of information (Microsoft, 2013) and also the
National Institute of Standards and Technology (NIST) emphasizes the need for a “scalable
architecture for efficient storage, manipulation, and analysis” when defining big data (NIST,
2014).
There are several definitions of big data which are dedicated to the crossing of some sort
o f threshold: for instance, Dumbill (2013) believes that data is big when it “exceeds the
processing capacity of conventional database systems” and requires the choice of “an alternative
way to process it”. According to Frampton (2015), “the term "big data" refers to data sets so
large and complex that traditional tools, like relational databases, are unable to process them in
an acceptable time frame or within a reasonable cost range. Problems occur in sourcing, moving,
searching, storing, and analyzing the data, but with the right tools these problems can be
overcome”.
As we have discovered, there are three major dimensions which help researchers to
quantify big data:
Volume, which is measured by the quantity of variables, transactions, attributes, events,
etc. In the past researchers used to work mostly with samples, smaller data sets, and created
predictive models. However, big data does not assume any volume constraints which allows
15
researchers to analyze much larger data sets and identify a number of previously invisible trends
and patterns.
Variety, which represents the assortment of data and is closely connected with the
definitions of structured and unstructured data. Structured data used to dominate the amount of
data being processed by enterprises and is much easier to analyze as it is classified on the basis
of the data type (numeric, character, etc.). However, over the past decades unstructured data has
become the prevailing type of data for business analysts to work with. As companies started to
look beyond organizational borders and expanded traditional operational data analysis, which
most often comes in a form of structured data, they encountered a lot of unstructured, more
complex data. Unstructured data by definition does not fit existing databases and is usually text
heavy, yet can contain numbers and dates as well.
According to the findings (McKinsey Global Institute, 2011), the amount of data is
doubling every two years and 95% of this data is unstructured. Therefore, one of the biggest
future obstacles for big data analytics is how to master analysis of tons of unstructured data and
apply meaningful results in practice when it will become a commonplace and mainstream.
In some theoretical publications researchers refer to the definition of semi-structured data
which contains parts of structured and unstructured data at the same time. For example, audio,
video, geospatial, click streams, log files and even text data fall into this category. When this data
is divided into several elements it turns from unstructured to more structured and, therefore,
become easier for an analyst to work with. Semi-structured data has separable semantic elements
which allow hierarchies within the data to be constructed.
Velocity of data defines the speed and frequency of data creation, accumulation,
processing, etc. The pace of today’s business world requires businesses to be able to do real-time
analytics and make appropriate decisions in real time. Here is where big data opportunities have
a lot to offer.
Data science: Cross-disciplinary field of science main purpose of which is to extract new
knowledge and insights from data, both structured and unstructured, by using various statistical,
analytical and data mining tools and with minimum of human interaction.
Data mining: Data mining is a process of sorting data and identifying relationships,
discovering patterns and trends from data by the means of machine learning.
Machine learning: An algorithm-based artificial intelligence tool which allows computers
to process and analyses empirical data, learn from it, use this data for making predictions and
adapt to new data exposed.
16
Business intelligence: Business intelligence represents a set of systems and various tools
which collect, store, analyze and access internally generated data and assist organizations in
decision-making.
Hadoop: Hadoop technology is an open source software framework made up of a freely
available software library that includes the following modules: Hadoop Common, Hadoop
YARN, the Hadoop Distributed File System and Hadoop MapReduce. This library allows users
to analyze large datasets with the capability of scaling up to thousands of machines. This
technology was inspired by Google’s products such as MapReduce and Google File System,
originally developed by Yahoo! And currently is run by Apache Software Foundation.
MapReduce: Google’s software framework for processing huge datasets to analyze a
wide range of problems on a distributed system.
Descriptive analytics: Part of statistics which is primarily concerned with description of
past and current events on the basis of accumulated data.
Inquisitive analytics: Part of analytics main purpose of which is to describe in details
underlying reasons for occurred facts by the means of disposable data.
Predictive analytics: This type of analytics is concerned with future forecasts and
predictions about certain trends with a certain likelihood.
This type of analytics works hand in hand with both types of analytics, described above.
The insightful information from descriptive statistics on past events is combined with predictions
and forecasts regarding the future and altogether analysis of this data gives prescriptions how to
obtain a certain goal and achieve certain results.
As a result, big data analysis consists of a blend of cross-disciplinary approaches where
various methods of descriptive, predictive and prescriptive analytics are combined in different
portions. At the exploratory stage big data analytics resort to descriptive analytics in order to get
a deeper understanding of the past events, however there is a trend to move towards predictive
and prescriptive analytics at later stages.
Combination of multiple analytical models, real-time predictive analytics, development
of advanced applications of predictive analytics are among important shifts describing today’s
big data analytics challenges. A comprehensive overview of big data analytics introduced by
Minelli and Chambers in their publication is demonstrated in a figure below.
17
Table 1 Overview of big data analytics
Relevant field of analytics
Descriptive Analytics Predictive Analytics
Prescriptive analytics
(Business Intelligence)
Issues addressed
What is the problem about?
What will happen if
How do we get there?
How did it happen?
the trend continues?
What is the best
How often does it happen?
What is most likely to
solution with a given
What are the
happen next?
uncertainty?
consequences?
How will variable X
What are the best
Statistics
affect the future?
Data mining;
choices?
Constraint-based
relevant
Machine learning;
optimization;
analytics
Forecasting;
Multiobjective
discipline
Predictive Modeling;
optimization;
Name of the
Simulation
Information management
Global optimization
Source: Minelli, M., Chambers, M. and Dhiraj, A. (2013). Big Data Technology, in Big Data,
Big Analytics: Emerging Business Intelligence and Analytic Trends for Today's Businesses, John
Wiley & Sons, Inc.- p. 93.
1.2.3 Managerial implications of big data analysis
In the publications reviewed there is a lot of information concerning potential gains and
managerial applications of big data analysis. For example, Columbus (2014) believes that “56%
of enterprises will increase their investment in big data over the next three years”. The author
presents persuasive evidences and examples from the real business: “70% of IT decision-makers
consider their organization’s ability to exploit value from big data as critical to their future
success; 65% say they risk becoming irrelevant and/or uncompetitive if they do not embrace big
data; 64% are seeing big data changing traditional business boundaries, enabling non-traditional
providers to move into their industries; 53% are seeing increased competition from data enabled
start-ups; 64% of senior executives said that big data is changing traditional business boundaries
and enabling non-traditional providers to move into their industry. Companies report a
significant level of disruption from new competitors moving into their industry from adjacent
industries (27%), and over half (53%) expect to face increased competition from startups enabled
by data”.
According to the Forbes’ recent survey (Columbus, 2014) of a big number of companies
on big data analytics, 89% of business leaders believe big data technologies will disrupt business
18
world in the same way the Internet did. More than 79% of respondents agree that businesses who
will not embrace these opportunities will lose their competitive position and may even extinct.
83% of companies have already pursued big data projects in order to seize a competitive edge.
The top three areas with the highest impact of big data in their business operations include:
impacting customer relationships (37%); redefining product development (26%); and changing
the way operations is organized (15%).
The impact of big data on business is extremely diverse and include solutions for various
problems in such areas as supply chain and human resources management, healthcare businesses,
banking, insurance, FMCG, retail and many more areas. The variety of solutions which big data
can handle can be roughly divided into four main groups and be related to customer-based
problems, optimization and modeling, prevention of thefts, fraud detection and complex crossindustrial problems.
There is a big number of examples where advanced data analytics can transform key
organizational business processes. Some of them are illustrated below. However, opportunities of
technology application for marketing purposes are particularly diverse and attractive as they
enable companies to solve a wide number of marketing-based problems:
➢ Procurement: Identification of the most cost-effective suppliers in terms of delivering
products on-time and without damages;
➢ Product development: Generation of product usage insights to speed product
development processes and improve new product launch effectiveness;
➢ Manufacturing: Identification of quality problems and optimization of manufacturing
processes;
➢ Distribution: Quantification of optimal inventory levels and optimization of supply
chain activities, e.g. based on external factors such as weather, holidays, and economic
conditions;
➢ Marketing: Evaluation of cost-effectiveness of marketing promotions and campaigns in
driving customer traffic, engagement, and sales and optimization of marketing mixes
given marketing goals, customer behaviors, and channel behaviors;
➢ Pricing and yield management: Optimization of pricing strategies and deep data-driven
analysis of various affecting factors;
➢ Merchandising: Optimization of merchandise markdown based on current buying
patterns, inventory levels, and product interest insights obtained from social media;
➢ Sales: Optimization of sales resource assignments, product mix, commissions modeling,
and account assignments;
➢ Store operations: Optimization of inventory levels given predicted buying patterns
coupled with local demographic, weather, and events data;
19
➢ Human resources: Identification of the characteristics and behaviors of the most
successful and effective employees.
As for the organizational function where big data analytics are applied most frequently, a
substantial part of results of the analytics are used by marketing, IT, sales and R&D departments
and for solving customer-related problems (54%) and operational issues (22%) where big data
technologies have a lot to offer.
Fig. 2 Managerial implications of big data analytics. Source: Big Data: A Competitive
Weapon for the Enterprise. (2014) Datameer. From: http://www.datameer.com.
To sum up, opportunities and potential gains of big data analysis for business are well
illustrated in today’s theoretical literature (Minelli, Chambers, Dhiraj, 2013; Davis, 2014;
Feinleib, 2014) as well as in publications prepared by practitioners (McKinsey Global Institute,
2011; Columbus, 2014, 2015; Nayler, 2014), yet they mostly represent positive attitude towards
technology implementation and therefore reflect current interest in big data analysis.
1.2.4 Global market overview of big data analysis
After having analyzed variety of applications of big data analysis across all business
processes, let us proceed with the global market overview of big data and identify major market
trends.
From the product & service perspective big data market is represented by three major
product & service categories: software, hardware and services. According to statistics, big data
services account for the largest share of the market equal to 40%, hardware – 38% while
software products’ market share remains around 22%.1 It is also noteworthy that majority of big
data providers base their solutions on software with Hadoop technology, which was developed
1 Big Data vendor revenue and market forecast 2013-2017. From:
http://wikibon.org/wiki/v/Big_Data_Vendor_Revenue_and_Market_Forecast_2013-2017
20
by the non-profit Apache Software Foundation, and offer customers a wide range of different big
data distributions based on this framework.
Global market of big data is expected to grow from 27 billion dollars in 2015 up to 100110 billion dollars by 2020. As it has been previously mentioned, the majority of big data market
players sell their services which are based on Hadoop technology distribution. According to the
research, three top players of the market of Hadoop software distribution are the following:
Cloudera accounts for 56 million dollars for annual total revenue in 2012, MapReduce – 23
million dollars while the annual revenue of Hortonworks was equal to 18 million dollars. 2
At the same time the so-called “Hadoop-as a service market” is anticipated to reach 50,2
billion dollars by 2020 while in 2015 this market counted for only approximately 6 billion
dollars.3 However, according to the research made by Allied Market Research company
prospects for Hadoop-as-a-service market are less optimistic. The market is expected to reach
only 16,1 billion dollars in 2020 which nevertheless demonstrates extremely rapid growth rates
for this market segment.4
From the industry perspective, top 5 industries which are the biggest adopters of big data
technology are the following (Datameer, 2014): financial services account for 22%, tech
companies – 16%, telecommunications’ sector use big data account for 14%, retail – 9% and
healthcare – 7%
The market of big data analytics has already passed the stage of early development and
currently both market entrants and incumbent market players face tough competition. IBM,
Google, Microsoft, SAS, SAP, Amazon and Oracle are among key players in big data analytics.
However, smaller players such as Cloudera, Pivotal, Palantir and some others also fight for the
market share. The detailed overview of functionalities of global big data platforms is
demonstrated in a table below.
Table 2 Global big data platforms
Large
Company
IBM
Big data distribution
IBM Biginsight (based on
Additional solutions
IBM Biginsight
2 Networkworld. Comparing the top Hadoop distributions. From:
http://www.networkworld.com/article/2369327/software/comparing-the-top-hadoop-distributions.html
3 Hortonworks. Powering the future of data. From: http://Hortonworks.com
4 Allied Market Research. From: http://alliedmarketresearch.com
21
market
players
Small
Hadoop)
Google Cloud Dataflow
Google Compute Engine, Google
Microsoft
HDInsight (via Hortonworks'
Cloud Storage, Google BigQuery
SQL programming and Microsoft
Oracle
Hewlett
Hadoop framework)
Hadoop via Cloudera
HAVEn (based on Hadoop)
Excel integration
R statistical programming
HAVEn Predictive analytics, R
Packard
Amazon
Amazon Elastic Cloud Compute;
statistical programming
Amazon Kinesis, Cloudera Impala,
Cloudera
Amazon Elastic MapReduce
Cloudera CDH (based on
Splunk Hunk, etc.
Cloudera Impala analytics
Palantir
Hadoop)
Palantir Gotham and Palantir
In-house analytics and data
Metropolis
management through Phoenix,
Pivotal HD (based on Hadoop)
Raptor, Horizon and RevDB
Pivotal Analytics and the use of the
Google
market
players
Pivotal
data-lake concept
Source: Big Data: a road map for business intelligence, Marketline.2015. Pp. 8-13.
Both major groups of market players, pure-play big data providers and large IT-vendors,
have started to focus on enterprise buyers, better articulate a wide range of applications of big
data in an organization and adapt their products to satisfy specifically this customer group.
Besides, partnerships also play a significant role in maturing of the market. In order to make it
easier for practitioners to adopt and integrate big data technologies there has been established a
number of reseller agreements and technical partnerships between big data providers and non-big
data vendors.
Main big data market drivers are listed below:
➢ Data which has been collected and stored will continue to grow exponentially;
➢ Multi-formatted and unstructured nature of data which businesses need to derive insights
from;
➢ Traditional IT solutions of business intelligence are failing under new requirements of
today’s rapidly changing business world;
➢ Cost of data systems, as a percentage of IT spend, will continue to grow while overall
cost of technology will gradually decrease overtime and become more affordable for
enterprises.
1.3 Big data analysis as a marketing instrument
1.3.1 Past & present of data analysis in marketing
22
Traditionally in order to analyze data marketers used to work with operational sources of
data and such internally-focused systems as enterprise resource planning (ERP) and customer
relationship management (CRM). However, recently the variety and complexity of data sources
have significantly increased and provided marketers with new challenges and promising
opportunities. The brief classification of various data sources which are at hand of any modern
business analyst is demonstrated below:
Conventional data sources:
Primary data (observations, customer surveys, experiments, interviews, ethnographic
research, etc.);
Secondary data (statistical and industry reports, consumer and business data, marketplace
analytics, webnography, scientific publications, etc.);
Supply chain data (electronic data interchange, vendor catalogues, pricing documents,
quality requirements, etc.).
Recently emerged data sources:
➢ Internet data (social media, clickstreams, etc.);
➢ Location data (geospatial data, data from mobile devices, etc,);
➢ Image data (videos, photographs, surveillance, satellite images, etc.)
Device data (sensors, radio frequency identification devices, programmable logic
controllers, telemetry, etc.).
Marketers are now doing a variety of things on the Internet starting from online
advertising, sharing information about the products, tracking consumer digital behavior to
executing online payments. Rapid technology developments have completely changed the
environment where marketers used to operate in, set and achieve goals and analyze results of
conducted marketing campaigns. So changed the customers, who raised immensely their
expectations from companies, have become extremely tech-savvy and always-connected, and
who now are favoring simplified visualization of the information they get. Internet technologies
have made it much more complex for businesses to attract and engage customers and build longlasting relationships with products and brands.
In the past data analysis was in the hands of third-party IT organizations which were
running expensive and difficult-to-implement ERP systems and creating and managing
marketing campaigns, tracking leads, billing customers and solving a number of other
marketing-related issues. Nowadays cloud technologies allow companies to fulfill all these
activities via the Software as a Service mode over the Web. Emerging enterprise marketing
management systems make it possible for businesses to analyze data and collaborate across
business functions and as a result improve customer experience.
23
Marketers have been exposed to a vast amount of data and today both industry giants of
B2C sector and small tech startups can have the same immense amount of data about their
customers. But at the same time big data promises require marketers to be very tech-savvy and
analytical, able to analyze complex models which can provide meaningful results in the real
time.
The attractiveness of big data as a powerful marketing tool is so immense because of its
capability of finding answers to a very broad range of questions: consumer behavior, buying
patterns, churn rates, attitude to competitors’ products, acceptance of a new technology and many
more marketing-related problems.
All in all, conventional marketing instruments have not become extinct, they have
transformed into more data-driven and technologically intensive tools instead which allow
companies to develop more focused campaigns and increase the effectiveness of marketing
actions.
1.3.2 Marketing applications of big data analysis
After we have demonstrated a variety of opportunities which big data analysis is opening
up for businesses, let us concentrate on marketing applications of the technology. Big data
opportunities for marketing are so vast and immense that companies simply cannot neglect a
chance to resolve a number of marketing-related issues with the help of this technology.
Researchers have defined the term of big data marketing. For example, according to
Arthur (Big data marketing: Engage your customers more effectively and drive value, John
Wiley & Sons, 2013), “big data marketing is the process of collecting, analyzing, and executing
on the insights you’ve derived from big data to encourage customer engagement, improve
marketing results, and measure internal accountability”.
Research shows that 48% of results of big data analytics are used for solving customerrelated problems, which includes churn reduction, product improvement initiatives, increase of
customer acquisition and revenue per customer and some more things, and 10% of them are used
for new product and service innovation solutions which contain data-driven development of new
products and service offerings (Datameer, 2014).
From the perspective of maturity of an analytical tool, all marketing applications of big
data can be roughly divided into two major groups:
➢ Mature analytic applications include traditionally optimization of marketing campaigns,
customer loyalty management, in-store custom analytics and many more things;
24
➢ Ad targeting optimization, customer churn prevention, product market targeting, product
design and experiments design optimization, advanced brand management are among
maturing and emerging analytical applications.
From the perspective of the data unit to be analyzed, big data marketing solutions can be
clustered into two large categories:
➢ Transactional analytics: This type of analytics is performance-based and allows business
leaders to evaluate all sorts of marketing actions at a higher speed and greater efficiency.
For example, one can assess the effectiveness of a particular offer or an ad online.
➢ Behavioral analytics: Behavioral analytics provides insights into the decision processes
of individual people making purchasing decisions and can be also defined as customer
analytics.
Generally speaking, the major share of applications of big data analysis in marketing
belong to the category of customer analytics. Customer analytics address a wide number of
marketing-related problems, e.g. they help companies to enhance customer-centricity of
marketing strategy, significantly improve customization of offerings and solve many more issues
which will be discussed later on. Customer analytics process, analyze and derive meaningful
results from all types of data such as demographic, behavioral and preference data, time-series
data and, of course, quantitative and qualitative data.
To begin with, big data allows companies to develop a personalized 360-degree view of
its customers (Feinleib,2014) and therefore get the comprehensive picture of all customers’
profiles with different socio-demographic and behavioral characteristics, preferences, purchase
frequencies, buying habits and many more characteristics separately. Today truly data-driven
companies are able to execute tailored marketing campaigns across different channels and obtain
customers’ attention using a customized and integrated set of marketing tools.
Effectiveness of conventional segmentation methods based on socio-demographic,
psychographic and behavioral characteristics today can be significantly enhanced by big data
solutions transforming traditional approaches into more focused and personalized ones.
Kash and Calhoun (2010) give an explicit example of how segmentation has evolved over
time using the case of dog food industry: “Working with demographic data, we have traditionally
segmented the dog food industry into categories like "owners of medium-sized dogs," "owners of
large dogs," or "owners of small dogs." Today, we can view the industry in terms of what
actually influences customer behavior: the types of relationships owners seek with their dogs (the
why). This leads us to market segments with personas, or representative archetypes, such as
"Pampering Parents," "Performance Fuelists," and "Minimalists." This additional dimension is
critical; it allows us to more clearly identify each segment's needs and desires, the triggers that
25
prompt them to act, and the owners' criteria for making purchase decisions. This proprietary
insight is a very real source of competitive advantage in a time when customer-centricity
matters.”
Advanced analytics of the customer data allow businesses to reveal various segments, for
example, companies now can identify opinion leaders and, as a result, construct more
customized offerings for this particularly important customer segment.
Big data also enhances the power of targeting generating and using more precise
knowledge of prospects and customers, including their preferred communication channels, and
allowing marketers to customize key elements of content experience efforts. Having huge
amount of situational data at disposal companies can further refine customer experiences to
reflect where people are, what they are doing, the devices they are using, the time of day, and
even the weather.
Let us refer to Weber and Henderson (2014) to demonstrate how combination of
customer data with advanced analytics creates even more powerful targeting opportunities:
“Next-best offer analytics help us estimate the probability that a customer will be interested in a
targeted offer. When the rules and algorithms of NBO are combined with search engines, we can
create cross-selling experiences such as, "You may also like…," which often result in higher
average order sizes and happy customers. For instance, the fashion retailer Forever 21 posts
personalized recommendations for items on the bottom of their "reset your password" page,
knowing that they have your attention and some degree of purchase-intent”.
Segmentation focuses marketers’ efforts and enables them to better prioritize customers at
the segment level, however companies can go one level deeper by using predictive big data
analytics to score each of our customers in terms of their own customer lifetime value (Arthur,
2013). Big data technology enables companies to prioritize its customers according to their
purchasing power and target the most profitable ones. Customer lifetime value demonstrates the
best estimate of a customer’s financial value to the company overtime and has roots in the past
purchase behavior of an individual. Therefore, big data enables to prioritize customers, select and
target the most potentially profitable customers.
Data fusion is another marketing tool which big data handles very successfully. Data
fusion is a common approach in marketing which is used for making inferences about particular
segments’ behavior and media exposure. During data fusion big data analytics, integrating
multiple datasets of similar, but still different respondents or customers, can build an enhanced
view of the target group. It is an especially useful tool when there are common underlying
characteristics between several respondent groups and there is a need to understand which
common factors influence their behavior and how.
26
As an addition to the various applications of customer analytics, it is important to
mention transactional analytics once again and emphasize that big data enables companies to
evaluate performance, measure and optimize marketing efforts (Dietrich., Plachy, Norton, 2014).
Thanks to the advanced technology, marketers can enhance effectiveness of the offline and
online marketing actions (e.g. increase efficiency of digital marketing strategy or improve social
media and content marketing campaigns).
With regards to the specific metrics to measure results of big-data-driven marketing
actions, today the most advanced marketers will put the big data analytics power to work,
removing more unmeasurable components from their marketing efforts and continuing to make
their marketing decisions more data-driven, while others continue to rely on traditional metrics
such as brand awareness or no measurement at all. On the other hand, a lot of conventional
marketing metrics, especially from web analytics, remain effective and still help companies to
achieve goals.
1.3.3 Key success factors and challenges in using big data as a marketing tool
The literature review had demonstrated a number of publications which are supposed to
guide companies along the way of implementation of big data analysis and avoid falling into
traps. Not only the authors of theoretical research studies, but also a lot of practitioners, experts
of building innovation in IT and well-known global consultancies often give their
recommendations with regards to business execution of big data marketing. Let us summarize
and introduce the most insightful advices which have been reviewed.
From the variety of contemporary marketing instruments, illustrated previously in this
chapter, these recommendations give an interesting perspective on development of a marketing
strategy and marketing mix as one of its core elements, complexity of using big data for
segmentation purposes and specifics of data-driven marketing management.
First of all, there is an opinion (Arthur, 2013) that if a company wants to build successful
data-driven marketing it should invest in development of enterprise data strategy. To reveal the
value of big data, it needs to be associated with enterprise application data. Therefore, companies
should establish new capabilities and leverage their prior investments in comprehensive
development of infrastructure, platform, business intelligence and data warehouses. Investing in
integration capabilities will enable data-driven marketers to correlate different types and sources
of data, to make associations, and to make meaningful discoveries faster than competitors with
no data strategy at their organizations.
27
For analytics to become a competitive advantage, organizations need to make “analytics”
the way they do business; analytics needs to be a part of the corporate culture and day-to-day
operational function of front-end staff.
With regards to the data-driven marketing management, it is recommended not to expect
quick financial returns from implementation of big data analysis and keep patience in waiting for
ROI’s. Big data insights may take time to emerge, and the process is continually evolving.
Besides, at the early stage of big data implementation companies should focus resources
around proving value from big data in one business area first via a pilot program rather than
attempting to do everything with big data at once. 5
Finally, businesses should focus on building talents for big data analysis. With talent still
one of the biggest big data challenges, organizations need to build the big data skills of existing
employees through training and development. This issue will be discussed more in details further
on in this chapter.
Being more specific and switching from corporate level to the perspective of a Marketing
department of an organization, it is recommended to develop multidimensional big data
marketing strategy and always analyze data from a number of different sources (Arthur, 2013).
The key idea here is two use big data analytics for effective investment management of the
customer portfolio. Companies should resort to personalization capabilities of big data analysis
and pay special attention to diversification of marketing tactics on the basis of potential
profitability levels of different customers.
Besides, big data marketing should be interactive and never focused solely on internal
data analysis. Businesses should keep in mind that today’s advanced analytical tools enable
organizations to run experiments, test marketing actions and implement corrections in real-time.
To develop a customer interaction strategy companies need to map and understand the buyer
journey, from first contact all the way through purchase and aftermarket relationships. As a next
step it is necessary to map out the changes that need to occur across organizations, systems, and
data to transform and deliver on the customer engagement plan.
The publications, which have been mentioned earlier in this subchapter, mainly represent
the positive attitude of scientists and industry experts towards the big data analysis in marketing.
However, behind the big data phenomenon in real-life business companies face a number of
challenges, starting from technical issues to human resources and legislation problems.
For example, let us refer to Stewart’s publication “Big data’s big mistake” (2015). The
article offers the author's insight to the big mistake being made by marketers, electronic
5 C-level executives seeing big results from big data. (2014). From: http://www.cio.com/article/2607333/big-data/c-levelexecutives-seeing-big-results-from-big-data.html
28
commerce leaders and product managers about big data: “Real customer insights are necessary to
turn quantitative data into qualitative information. Use qualitative insights to direct the questions
of any quantitative study, and use qualitative data to help answer the “why” behind the
numbers…You need insight into what customers really want. And to obtain this, you need
qualitative data”.
He mentions that the power of qualitative data is often overlooked as marketers
increasingly focused on numbers, increasing the risk for companies to treat their customers like
numbers. In his opinion, today marketers all over the globe focus so much on numbers that they
are neglecting the human side of data. In order to avoid possible misleading outcomes of the data
analysis the author suggests the following: ”Use qualitative insights to direct the questions of any
quantitative study, and use qualitative data to help answer the “why” behind the numbers”.
Let us mention another worthy critical study on big data analytics by Krajicek (2013). He
mentions that business researchers must be insight researchers, they need to be closer to the
business rather than to any analytic techniques. The author claims that “it takes a collective
effort, of marketers and researchers together, to build the impactful dashboards and architecture
that provide relevant insights that drive business change”. Besides, Krajicek emphasizes the
complexity of big data processing by ordinary market researchers and adds that successful
analysis of big data needs “big thinkers” who understand the market and can survive the chaotic
environment of big data and derive insightful solutions.
To sum up, research demonstrates that in order to realize full value of opportunities
offered by big data analytics marketers have to overcome the following obstacles:
Technology-related barriers:
Difficulty of transforming data into a suitable form for analysis;
Data which comes from multiple and disparate sources is difficult to be merged and
integrated;
There is a shortage of best practices for integrating big data analytics into existing
business processes;
There is a lack of competent big data practitioners and data scientists;
Difficulty of providing excellence in big data applications performance for a big number
of concurrent users.
Non-technology-related barriers:
Insights generated by big data analysis are often difficult to be operationalized and
implemented in business routine;
Big data market is developing fast and is expected to mature in the nearest future, yet it
still remains highly competitive and volatile;
29
Stakeholders are not willing to agree on data definitions of big data projects and
deployments;
Despite overall enthusiasm and interest shared by business community, there is still a
large amount of end-users who are not fully aware and certain about the real value of big
data applications;
Legal concerns connected with overall regulative framework, privacy and compliance
issues will remain as a big challenge;
There is a lack of ready-to-implement big data applications designed to address specific
business problems.
After we have defined all major obstacles for big data implementation in marketing, let us
pay special attention towards the lack of competent human resources to analyze big data.
According to all major publications on the future of big data analysis businesses are very likely
to encounter a great shortage of experts in data analysis in the upcoming years ( Minelli,
Chambers, Dhiraj, 2013). Big data analytics require specialists with deep expertise in machine
learning and advanced statistics, yet at the moment demand on this type of analysts exceeds the
total amount of data scientists and the trend will continue in the future.
Since big data analysis is the intersection of several scientific fields, data scientists need
to be talented at a number of disciplines at the same time: mathematics, information technology,
business administration and behavioral sciences. While the first three requirements are quite
obvious, knowledge of behavioral sciences is recommended to have a better understanding of
human behavior and interrelations between facts and people’s actions. Moreover, the education
of such experts is rather time-consuming and does not correspond to the real-time needs of
business.
According to some research by McKinsey (McKinsey Global Institute, 2011) only in
United States around 140 000 – 190 000 data scientists’ positions will stay vacant by 2018. A
need for big data analysts and managers with necessary expertise in US has been calculated at
the total amount of 1,5 million people.
Since big data analysis is the intersection of several scientific fields, data scientists need
to be talented at a number of disciplines at the same time: mathematics, information technology,
business administration and behavioral sciences. While the first three requirements are quite
obvious, knowledge of behavioral sciences is recommended to have a better understanding of
human behavior and interrelations between facts and people’s actions.
Let us also draw special attention to the fact that businesses suffer a shortage on not just
data scientists, but also marketers with deep IT and math expertise. Even when unsorted loads of
data will be received by a manager in a more understandable fashion, it is still a great challenge
30
how to take the advantage of this information, identify and classify numerous trends and
patterns.
According to the experts’ opinion from IBM (Dietrich. , Plachy, Norton, 2014) this
challenge is likely to be quite complicated to get over: “Despite the positive trend in spending for
marketing, this will not be an easy shift, either inside or outside IBM...Marketing’s creative
nature has historically lent itself to a strong belief in decisions based on “gut instinct.” Insights
from big data and analytics provide a different starting point for decisions and can support the
creative process to improve outcomes. Such a transition requires a shift in the skills of the
marketing organization.
The predictions are that approximately half of the new hires in marketing teams are
expected to come from technical backgrounds, and that is expected to grow in subsequent years
as organizations realize that the skills needed in this era of marketing are shifting. A 2013 C-suite
study by IBM’s Institute for Business Value found that CMOs feel less prepared to cope with big
data in 2013 than they did in 2011. For example, in 2011, 71% of CMOs felt underprepared for
the data explosion; in 2013, 82% did”.
1.3.4 Real-life cases of using big data for marketing
The literature review has shown that there is a large amount of publications on potential
gains of big data analytics. However, there is a lack of papers which could describe a full case of
a company’s implementation of big data marketing from the very beginning to the end and
demonstrate changes in the organization’s performance. Big data marketing remains to be quite a
new area both for researchers and business community and that is why we are facing lack of
publications on successful cases and best practices.
Nevertheless, in order to demonstrate how big data analysis resolve marketing problems
in practice, let us consider a few representational real-life cases of big data implementation in
marketing across different economic sectors.
The case of SFR Company
To begin with, let us examine the project initiated by the SFR company in France in 2014
which was centered on customer analytics and development of 360-degree view of customers
with the help of big data analysis. SFR is the second largest telecommunications operator in
France which serves more than 21 million customers and delivers high-speed wired internet to
5,2 million households in France. SFR also operates in the B2B sector and serves over 160,000
business, government and community clients as well.
The business challenge which the company encountered had several sub-problems. First
of all, SFR needed to create a mechanism capable of collecting and storing the huge amounts of
31
data generated by subscribers. Secondly, the company wanted to provide its marketers with realtime data about their customers from a 360-degree perspective. In order to fully understand the
customer journey, the company needed to bring in multi-structured data from multiple sources
into a single unified platform. In addition, SFR was aiming at creating a detailed view into the
customer journey that would be available to employees across the company for real-time search,
reporting and analysis.
The customized solution to tackle this problem was delivered to SFR by an external
vendor. As a result, the company managed achieved the following business benefits:
➢ The IT program was installed which managed to create a 360-degree view of customers
including a number of characteristics and provide real-time data about customer journeys;
➢ Data integration strategy was implemented which enabled SFR to ingest, store, and
analyze data which could reveal previously hidden customer insights;
➢ Besides, data integration strategy allowed SFR to improve data-warehouse performance
and extending the enterprise data warehouse life up to 3 times. 6
As we have seen from the literature reviewed, big data marketing is particularly
applicable to development of CRM systems which have always been generating large amounts of
data. That is why the next real-life example of big data marketing is dedicated to the Target
corporation and prediction of consumer behavior and upcoming life events.
The case of Target Corporation
Target is the second-largest discount retailer in the USA after the Walmart Corporation.
The company is listed among S&P 500 list of companies, has 1 802 locations throughout the
United States and operates in several price segments in frames of the bigger, major discount
target segment.7
Loyalty cards and programs are a useful marketing tool for companies which serve the
dual purpose of retaining and attracting existing and new customers and acting as a data source
for targeted marketing strategies. For the large Target Corporation the amount of data generated
by its loyalty system is monumental and enables the company to derive value from this data with
the help of big data analysis. Target resorted to data analytics to execute the targeting initiative,
as a part of CRM program, which was centered on building a model for the p rediction of
customers pregnancy (Marketline Case Study, 2014).
Target realized new parents offered the company the perfect opportunity to exploit
customer loyalty. According to research, previously ingrained shopping habits fall apart when
customers become parents due to various behavioral factors which lead them to becoming more
willing to purchase everything conveniently in one place rather than from separate stores, that is
6 Cloudera. Cloudera Enterprise Data Hub in Telecom: Three customer case studies. (2015). From: http://www.cloudera.com
7 Target Corporation. From: https://corporate.target.com
32
why the corporation decided to use data-driven targeting to benefit from these potential cash
cows.
Using the internal records of baby-shower purchases and customer information from
existing loyalty cards, the company was able to derive insights from the data and discover
specific buying habits attributable for various stages of pregnancy. As a result, Target assigned
each shopper a pregnancy prediction score and even an estimate for a due date for customers
based on the purchasing habits of around 25 specific products.
However, the retailer made a mistake since the test marketing sessions demonstrated that
the changes in CRM system were not always positively perceived because of the specific and
delicate nature of the pregnancy topic. Target Corporation implemented some changes (started
offering mixed coupons with pregnancy-related and non-related discounts instead of giving out
pure pregnancy-specific offerings) and as a result data-driven targeting program started to bring
financial returns as expected.
The case of Avis Budget
Let us also have a look at how Avis Budget company is using big data in marketing. Avis
Budget is a global car rental company operating two brands and serving more than 40 million
customers.
Big data initiative was triggered by the company’s objective to get a deeper
understanding of their customers and as a result to adjust their offerings according to the
customers’ needs. It is noteworthy that Avis Budget’s strategy is centered on the excellence of
customer service and customer experience and that is why this project was so important for the
organization. Avis Budget planned to develop a 360-degree view of their customers, apply a
segmentation strategy on the basis of a customer lifetime value and infuse the customer
experience with intelligence.
After handling records from 40 million customers, collecting data from its rental
transaction system, website transactions and reports that detail what products customers took, the
company managed to project how many rentals a person would undertake in a year and what the
profit of that person was going to be. The big data project turned out to be an effective
investment as the organization began to wring truly more value from its data. All information
about customers was consolidated and provided the employees with a single view of every
customer. Design of the customer lifetime value model gave Avis Budget a prediction of rental
frequency at a customer level and customer profitability. Six segments were identified by the
company and this approach increased the effectiveness of the contact strategy by 30%. 8
8 CSC. How Avis Budget Uses Big Data in Marketing. (2015). From: http://www.csc.com/big_data/insights/97741how_avis_budget_uses_big_data_in_marketing
33
All in all, various applications of technology as a marketing instrument are demonstrated
in general terms in the publications of Arthur (2013), Feinleib (2014), Weber and Henderson
(2014), Bacon (2014), etc. A lot of general information on this issue can also be found in
analytical reports and studies (Forbes, 2014,2015; Oracle, 2014; CIO Online Journal, 2014,
2015; Dietrich, Plachy, Norton, 2014).
However, due to the innovativeness of the topic, the specifics of using big data for
marketing purposes in real-life business environment have not been clearly defined and
examined by researchers.
Theoretical publications by such authors as Minelli, Chambers, Dhiraj (2013), Arthur
(2013), Stewart (2015) as well as publications prepared by practitioners from McKinsey (2011)
and IBM (Dietrich, Plachy, Norton, 2014) provide a general overview of potential obstacles and
barriers connected with big data analysis execution, however they are not examined specifically
enough.
Some of these papers (Arthur, 2013; Dietrich, Plachy, Norton, 2014) illustrate obstacles
in using big data specifically in marketing, but the majority of them are not based on the real-life
cases and are presented in a form of tips and guidelines for companies.
1.4 Big data analysis as a marketing tool: peculiarities of the Russian context
1.4.1 Russian market overview of using big data
Russian market of big data has just recently emerged and currently is in the stage of
development. According to some forecasts, it is expected to continue growing at the annual rate
of about 35% during 2014-2018.9 Despite the general downward trend for economic
development and overall recession in the country, Russian market of big data shows positive
results as demand is rapidly growing and businesses are looking for alternative ways of
efficiency increase, risks minimization and costs reduction. Companies are getting more and
more interested in the vast opportunities of big data and are step-by-step exploring this new field
of advanced business analytics.
Although currently Russia’s share in the global market of big data is only 1,8%
(according to the amount of accumulated data), it is anticipated to reach 2,2% by 2020. The IDC
company evaluates the value of the market as of $ 340 million, 100 of which is generated by
SAP solutions and $ 240 million account for IBM, SAS, Oracle, Microsoft and other major
providers.
9 Rusbase. Rostelekom is planning to by a Russian Big data company for $8 dollars. (2015).
From:http://rusbase.com/news/ros-aikumen/
34
According to the survey by CNews Analytics and Oracle (2014), respondents of which
represented 108 large organizations from various industries demonstrates that a third of Russian
companies have just started to use big data technologies in their businesses. However, only 10%
of survey participants are in fact using big data analytics in their businesses while the world’s
average is about 30% of all business entities in the country. Following the global trend, Russian
companies from retail, telecommunications and financial services industries resort to big data
solutions more frequently than others, yet logistics, mining and utility companies and
governmental organizations have also started to use this opportunity.
Although major providers of big data solutions are foreign IT giants, such as Oracle,
IBM, SAP, SAS, Microsoft, Hortonworks, EMC, Cloudera and some others, Russian companies
(e.g. Mail.ru Group, Yandex) also form a part of the market.
The following trends can be seen as major market drivers and barriers influencing the
Russian market of big data:
Market drivers:
➢ Increasing awareness and growing interest of Russian companies in big data
opportunities;
➢ Сontinuous economic crisis is forcing companies to look for alternative ways for costs
reduction and efficiency increase, and big data analysis is one of such alternatives;
➢ General economic recession and sharp decrease of Russian currency will favor Russian
solutions providers in comparison with more expensive foreign market players;
➢ Import substitution in IT industry which will have a positive effect on development of
Russian big data solutions;
➢ Recent legislation requiring Russian companies to store all data solely on Russian
territory.
Market barriers:
Privacy and confidentiality issues and lack of relevant legislation on big data;
Lack of competent experts in big data analytics;
Difficulty in implementation of big data analytic in currently operating enterprise
information systems;
High expenses necessary for big data technology development which makes it more
complicated for Russian IT companies to start new initiatives;
Increased prices of imported goods, continuous inflation and overall political instability
will have a negative impact on development of the whole IT industry.
As a result, factors that affect Russian market of big data combine both global trends and
trends attributable only to Russia and represent both positive and negative influence. Countryspecific market trends, connected with the current economic crisis and recent legislation
35
changes, are likely to have a rather positive effect on implementation of technology by Russian
companies. However, it is important to mention that the implementation of big data analysis will
be successful only if a company has sufficient financial resources and can afford investments in
big data initiatives.
The major part of the market barriers, which have been mentioned earlier, is however
reflecting global trends. Today the global market of big data execution by companies is still at
the stage of development. In addition, not only Russia, but the business community all over the
world is struggling with the lack of competent human resources for big data analysis.
1.4.2 Real-life cases of Russian companies working with big data as a marketing instrument
Regarding the applications of big data analysis by Russian companies, it varies from
optimization of operations, security problems to marketing-based issues. A brief overview of the
most successful segments of big data applications in Russia is demonstrated below in the table.
Table 3 Profitability of different segments of big data applications in Russia (2013)
Segment
Total income (2013)
Total market share
Market growth
($ million)
(2013) (%)
rate (2012-2013)
BI-platforms
8 550
59,5
8,8
CRM-systems
2 735
19
5,1
Analytical applications &
2 001
13,9
5,8
Advanced analytics
1 082
7,5
12,5
In total:
14 368
100
7,9
optimization of business processes
Source: Cnews. Overview of business analytics and big data in Russia in 2014. (2014). From:
http://www.cnews.ru/reviews/bi_bigdata_2014/articles/perspektivy_biznesanalitiki_v_rossii
Marketing problems are one of the top issues which are addressed by Russian business
leaders by the means of the new technology. Big data analysis is generally used for advanced
customer analytics, segmentation, targeted advertising and performance evaluation.
Analysis of recent statistical report of 16 biggest big data projects among Russian
companies in 2014 demonstrates that the prevailing share of projects belong to finance and
telecommunications companies and 90% of projects were initiated to solve marketing-related
problems among others.10
10 Cnews. The largest big data projects in Russia. (2014). From:
http://www.cnews.ru/tables/a9249186ccefd9e546774ec36da1970ba20ca212/
36
The list of companies, who have already adopted big data technology as a marketing
instrument, include all major Russian market leaders among telecommunications companies
(Vympelkom, Megafon, MTS Group), a number of national retailers (Ulmart, Lenta, X5 Retail
Group, Gloria Jeans) as well as banks (VTB24, Alpha Bank, Sberbank) and government
institutions (Federal tax authorities).
The research demonstrates that these companies execute big data analysis in several
ways. The majority of organizations build partnerships with external providers, both Russian and
foreign ones, and receive ready-to-use analytics or software to run it in-house, other companies
focus on developing their internal competences of big data analysis.
However, only a small share of companies has already started using big data for
marketing purposes (31%), the majority is either planning to implement big data analysis in their
organizations (25%) or do not consider this opportunity at all (44%). 11
The case of Incity
Let us have a look at a few successful cases of big data application in marketing by
Russian companies. According to Boris Mikhalkin, Business analytics Manager of Incity
company which is a large Russian fashion retailer, in 2013 Incity launched a comprehensive big
data analysis initiative together with the Qlikview company. This project was aimed at improving
efficiency of marketing actions among other project objectives and as a result allowed the retailer
to evaluate effectiveness of marketing campaigns and optimize CRM actions. 12
The case of War Gaming
Another example can be demonstrated by collaboration of War Gaming company and
Yandex Data Factory which started in December, 2014. Generally speaking, Yandex Data
Factory uses big data processing and machine learning to provide various businesses with
meaningful solutions in a number of areas. As for the field of marketing-related issues
(Marketing & Customer relationship management, as the organization itself specifies this
division), the company works on the following questions: recommendation systems,
personalization, churn prediction & prevention.
As for the project for War Gaming, Yandex Data Factory was required to improve churn
prediction and maximize its prevention. Yandex was working with raw customer databases of
War Gaming: the company analysed the customers’ data and clustered customers according to a
number of characteristics. Afterwards, the model was built which contained a number of
11 Cnews. Overview of business analytics and big data in Russia. (2014).
From:http://www.cnews.ru/reviews/bi_bigdata_2014/articles/bolshie_dannye_v_rossijskoj_interpretatsii
12 Cnews.Overview of business analytics and big data in Russia. (2014). From:
http://www.cnews.ru/reviews/bi_bigdata_2014/interviews/boris_mihalin
37
variables and could identify the specific segment of potential churners and predict the probability
of a switch by a “churner”.13
As for the market trends and barriers for big data analysis as a marketing tool, they are
similar to the development of big data analytics market as a whole and include legislation
restrictions, stagnation of IT sector, lack of competent specialists and early stage of technology
adoption among companies and insufficient understanding the importance of organizational data
strategy at a corporate level. Prevailing part of the companies do not see clearly the benefits of
big data adoption and for some of them big data phenomenon is considered just as the natural
further development of business analytics, not a technological breakthrough instead.
To summarize analysis of publications dedicated to the peculiarities of the Russian
context, the reviewed literature has demonstrated a variety of analytical reports on general
perception of the technology by Russian companies (Cnews, Oracle, 2014), but there is a small
number of studies which focus on real-life cases and analysis of obstacles faced by Russian
companies in big data execution as a marketing tool.
All in all, based on the literature review the research gap has been revealed between the
publications dedicated to big data analytics as a marketing tool in general, both from the global
and Russian market perspectives, and evaluation of practical implementation of this technology
by Russian companies for the marketing purposes. Although there is a number of research papers
dedicated to the benefits of big data analytics in marketing or analytical market overview reports
there is a lack of research done on how Russian companies handle big data technology as a
marketing tool, what obstacles they are facing and how they themselves evaluate this technology.
Besides, on the one hand big data analytics as a marketing tool is a highly innovative
topic in research, yet on the other hand several Russian companies have already started adopting
this technology.
Based on these findings from the theory review this study is considered to be a subject of
exploratory research and contain analysis of real-life evidence.
The major research questions, which aim to investigate the phenomenon and reveal the
factors impacting current practices of Russian companies, form a basis for empirical part of this
study:
1. Why Russian companies resort to big data analytics as a marketing tool?
2. How do Russian companies execute big data technology as a marketıng tool?
3. How do Russian companies overcome barriers connected with big data analysis as a
marketing instrument?
13Yandex data factory. World of Tanks achieves a new level of churn prevention through the implementation of
YDF’s data analysis. From: https://yandexdatafactory.com/case-studies/world-of-tanks-achieves-a-new-level-ofchurn-prevention-through-the-implementation-of-ydfs-data-analysis/
38
4. How can Russian companies leverage the expertise of global market leaders in order to
empower big data analytics for marketing purposes in Russian market?
Therefore, exploratory nature of these research questions implies that the most
appropriate research method for this study is multiple case study research. A more detailed
perspective on the justification of the research method is provided in the Chapter 2 of this study.
39
Research Gap
The review of relevant theoretical literature has demonstrated that big data analysis as a
marketing tool is a highly innovative topic not only in the field of business practices, but also in
the field of research.
To begin with, opportunities and potential gains of big data analysis for business are well
illustrated in today’s theoretical literature (Minelli, Chambers, Dhiraj, 2013; Davis, 2014;
Feinleib, 2014) as well as in publications prepared by practitioners (McKinsey Global Institute,
2011; Columbus, 2014, 2015; Nayler, 2014), yet they mostly represent positive attitude towards
technology implementation and therefore reflect current interest in big data analysis.
Concerning possible applications of technology as a marketing instrument, they are
demonstrated in general terms in the publications of Arthur (2013), Feinleib (2014), Weber and
Henderson (2014), Bacon (2014), etc. A lot of general information on this issue can also be
found in analytical reports and studies (Forbes, 2014,2015; Oracle, 2014; CIO Online Journal,
2014, 2015; Dietrich, Plachy, Norton, 2014).
However, due to the innovativeness of the topic, the specifics of using big data for
marketing purposes in real-life business environment have not been clearly defined and
examined by researchers.
Theoretical publications by such authors as Minelli, Chambers, Dhiraj (2013), Arthur
(2013), Stewart (2015) as well as publications prepared by practitioners from McKinsey (2011)
and IBM (Dietrich, Plachy, Norton, 2014) provide a general overview of potential obstacles and
barriers connected with big data analysis execution, however they are not examined specifically
enough.
Some of these papers (Arthur, 2013; Dietrich, Plachy, Norton, 2014) illustrate obstacles
in using big data specifically in marketing, but the majority of them are not based on the real-life
cases and are presented in a form of tips and guidelines for companies.
Regarding the Russian context, the reviewed literature has demonstrated a variety of
analytical reports on general perception of the technology by Russian companies (Cnews,
Oracle, 2014), but there is a small number of studies which focus on real-life cases and analysis
of obstacles faced by Russian companies in big data execution as a marketing tool.
To sum up, on the basis of identified research gaps after the literature review this research
study will be centered on investigation and analysis of specifics and obstacles of execution of big
data analysis as a marketing tool by Russian companies and will be based on real-life case
studies of Russian companies.
40
Summary of Chapter 1
In the first chapter we have reviewed relevant theoretical literature as well as research
studies conducted by practitioners dedicated to the big data analysis and execution of this
technology for marketing purposes.
As a first step, the major contemporary marketing tools have been briefly reviewed,
including all elements of marketing strategy planning and execution, analysis of marketing
processes, introducing such concept as 4P’s marketing mix, 4C’s and 4A’s models, etc.
After the evolution of big data as a technology for business has been introduced, all key
definitions of big data analysis have been reviewed including such concepts as business
intelligence, machine learning, data mining, etc.
Besides, we have demonstrated that the impact of big data on business is extremely
diverse and include solutions for all business functions, starting from procurement,
manufacturing and logistics to pricing, marketing and customer service.
As for the organizational function where big data analytics are applied most frequently,
we have revealed that a substantial part of results of the analytics are used by marketing, IT, sales
and R&D departments and for solving customer-related problems (54%) and operational issues
(22%) where big data technologies have a lot to offer (Datameer, 2014).
The global market of big data has been reviewed and major global market players have
been introduced. Global market of big data is expected to grow from 27 billion dollars in 2015
up to 100-110 billion dollars by 2020 (Networkworld, 2014). Besides, market trends and barriers
have been examined.
From the industry perspective, top 5 industries which are the biggest adopters of big data
technology are the following (Datameer, 2014): financial services account for 22%, tech
companies – 16%, telecommunications’ sector use big data account for 14%, retail – 9% and
healthcare – 7%.
In addition, it has been illustrated that marketing applications of big data analysis are
truly diverse. The technology can address such processes as segmentation and targeting, develop
models for prediction of customers behavior and development of 360-degree view of customers,
analyze customer lifetime value and measure performance of marketing actions.
The literature review helped to identify key success factors in implementation of big data
analysis in marketing and most common barriers, both technology- and non-technology-based.
Real-life cases have been analyzed of such companies as SFR (France), Target (USA) and
Avis Budget (USA).
Regarding the context of the Russian market, we have revealed that Russian companies
also have started to use big data analysis for solving marketing-related problems and introduced
41
several real-life case studies (Incity, War Gaming & Yandex Data Factory). Although currently
Russia’s share in the global market of big data is only 1,8%), it is anticipated to reach 2,2% by
2020 (Cnews Analytics, Oracle, 2014).
The factors that affect Russian market of big data combine both global trends and trends
attributable only to Russia and represent both positive and negative influence. Country-specific
market trends, connected with the current economic crisis and recent legislation changes, are
likely to have a rather positive effect on implementation of technology by Russian companies.
However, it is important to mention that the implementation of big data analysis will be
successful only if a company has sufficient financial resources and can afford investments in big
data initiatives.
The major part of the market barriers, which have been mentioned earlier, is however
reflecting global trends. Today the global market of big data execution by companies is still at
the stage of development. In addition, not only Russia, but the business community all over the
world is struggling with the lack of competent human resources for big data analysis.
42
Chapter 2. Research design
2.1 Overview of the research methodology
This chapter is dedicated to the introduction of main research questions and a detailed
step-by-step description of the research methodology of this paper.
The overview of the literature in the previous chapter has demonstrated a research gap
between analyzed publications of researchers and business practitioners and as a result the
exploratory nature of this study has been introduced and justified.
As previously stated, the major research questions of this study are the following:
1. Why Russian companies resort to big data analytics as a marketing tool?
2. How do Russian companies execute big data technology as a marketıng tool?
3. How do Russian companies overcome barriers connected with big data analysis as a
marketing instrument?
4. How can Russian companies leverage the expertise of global market leaders in order
to empower big data analytics for marketing purposes in Russian market?
Exploratory nature of these research questions implies that the most appropriate research
methods for this study will be multiple case study research.
Research design of this study will be based on collection of both primary and secondary
data and include analysis of both quantitative and qualitative data. A more detailed description of
the methodology will be introduced later on in this chapter.
2.2 Justification of the suitability of a case study analysis as a research method
To begin with, let us introduce several most relevant definitions of a case study as a
research method.
In order to get a general understanding of the case study method, it is a good idea to refer
to Schramm who believes that “the essence of a case study...is that it tries to illuminate a
decision or set of decisions: why they were taken, how they were implemented, and with what
result.” (Schramm, 1971)
A more thorough explanation of this methodology is provided by Yin (Yin, 1981a,
1981b), according to whom a case study is best described when taking into consideration its
twofold structure. Therefore, the first part of Yin’s definition is more concerned with the scope of
a case study:
“A case study is an empirical enquiry that investigates a contemporary phenomenon in
depth and within its real-world context, especially when the boundaries between phenomenon
and context may not be clearly evident. “
The second part of the definition describes features of a case study:
43
“A case study enquiry copes with the technically distinctive situation in which there will
be many more variables of interest than data points and as one result relies on multiple sources of
evidence, with data needing to converge in a triangulating fashion, and as another result benefits
from the prior development of theoretical propositions to guide data collection and analysis. “
In order to justify the suitability of the case study research methodology, let us refer once
again to Yin and his publication “Case Study Research. Design and methods”. According to the
author, there are three major conditions which should be met in a research to justify that the
preferred methodology of the paper is case study research. Let us follow Yin’s guidelines and
demonstrate the suitability of this methodology for this study.
1. First of all, the first three research questions of this paper focus mainly on ‘’how” and
“why” questions which indicates explanatory nature of the study where case study
method works at best.
2. Secondly, this research paper examines only contemporary events and analyzes an
extremely innovative topic of implementation of big data analytics in marketing by
Russian companies. Due to the innovativeness of the study other methodological
approaches are unlikely to provide reliable results. On the contrary, case study method
will illustrate real-life evidence and demonstrate the specifics of execution of the new
technology by organizations.
3. It is also worth mentioning that the relevant behaviors, studied in this paper, cannot be
manipulated: the researcher will only observe the relevant events, analyze relevant
documentation and interview the persons involved in the process.
2.3 Overview of the case study analysis
The case study questions are the following:
1. Why Russian companies resort to big data analytics as a marketing tool?
2. How do Russian companies execute big data technology as a marketıng tool?
3. How do Russian companies overcome barriers connected with big data analysis as a
marketing instrument?
As stated before, the topic of this paper is likely to be the subject of exploration where it
is recommended not to have any study propositions, which would drive in a certain way the
whole further case analysis and could possibly provide biased outcomes.
On the contrary, let us define the exploration purpose of this case study which will be to
get a deeper understanding of the peculiarities of usage and implementation of big data for
marketing purposes by Russian companies.
Necessary criteria for the exploration will be the following:
1. To analyze only Russian companies operating mainly on the Russian market;
44
2. To ignore such fact as the size of the company and its market share in choosing an
organization to be analyzed;
3. To focus on the industries and market sectors with the strategic importance of marketing
division (e.g. telecommunications, retail or FMCG);
4. To focus on potential gains of big data technology solely for the purposes of marketing;
5. To reveal and examine company-specific, industry- and country-specific factors which
influence organizations’ decisions whether to resort to big data or not.
It is crucial to mention that case study methodology will address multiple cases
(preferrably 3-4) which will improve the external validity of research results.
Yet it is also noteworthy that due to the number of cases planned to be analyzed and
exploratory nature of the study the research design will be adaptive, so that any non-fundamental
changes in the research design procedures can be implemented at any stage of the research.
We believe that in this study the most appropriate defined unit of analysis will be the
Marketing department of a Russian company operating on the Russian market, since this
organizational unit is the centre of using, implementing and evaluating results of big data
analytics technology.
Limitations of the unit of analysis are defined by the boundaries of the department itself
meaning that it is out of scope for this case study research to analyze the impact of big data on
development of any other organization’s departments.
Regarding relevant criteria for case study’s findings interpretation and taking into
account exploratory nature of the study, we will resort to the analytical generalization of research
findings and comparison of these findings with similar findings about foreign companies using
big data in marketing, which were addressed previously in Chapter 1 during the literature review.
Due to the fact that this paper addresses analysis of multiple case studies, we will use
replication logic, which is similar to the methodology of multiple experiments, and therefore
ensure external validity of research findings. Thanks to the innovativeness of the technology and
low rate of adoption of big data analytics as a marketing tool by Russian companies, we will
resort to both literal replication (cases predict similar results) and theoretical replication (cases
predict contrasting results for anticipatable reasons) in this multiple-case study analysis.
The overview of step-by-step procedures of multiple-case study analysis of this paper is
demonstrated below:
Step 1: Defining and designing the cases
1. Review of the relevant literature (accomplished in Chapter 1)
2. Selection of relevant cases (based on the exploration criteria)
3. Design of data collection procedures
Step 2: Data collection and data analysis
45
1. Execution of 4 case studies
2. Preparation of individual case reports
Step 3: Findings analysis and derivation of conclusions
1.
2.
3.
4.
Drawing cross-case conclusions
Comparison of research findings with the literature reviewed
Development of policy implications and recommendations for the companies
Preparation of a cross-case report
2.4 Data collection procedures
The data collection process starts with definition of selection criteria for the companies to
be interviewed. Although overall exploration criteria for the case study research have already
been introduced, company selection criteria focus only on aspects which make a company an
interesting and at the same time relevant and reliable for exploration.
1. As this study is primarily concerned with the analysis of practices of Russian
companies, an organization should be registered as a Russian business entity and operate mainly
on the Russian territory;
2. An organization should preferably belong to the industry where large amounts of data
are generated (retail, telecommunications, financial services, etc.)
3. A selected company should also preferably belong to a customer-oriented market sector
with high importance of Marketing division in the organizational structure;
4. A company should be currently using big data analysis for marketing purposes.
As a result, the following companies have met all set requirements and have been
selected for further analysis: MTS Group together with another major Russian mobile services
operator as representatives of the Russian telecommunications industry; Lenta and Ulmart as
some of the largest retailers (offline and online) in the Russian market. All selected organizations
have a 2-3 record of using big data analytics for marketing purposes, operate on a national scale,
in B2C segment and are customer-oriented.
As long as 4 Russian companies have been selected for further analysis of their
Marketing departments on the basis of the exploration criteria, let us move on to the stage of data
collection. In order to improve the construct validity of the research multiple sources of evidence
are collected and analyzed.
Interviews with representatives of Marketing departments (preferably Marketing
directors) of the Russian companies will represent the main source of evidence. This information
source has been selected as the major one since it targets directly case study issues and can
provide us with insights and personal viewpoints on the topic. Documentation and archive
46
records (all relevant and available data about organizations examined) will work as a supporting
source of evidence.
Units of data collection are the following:
➢ Company’s internal data for analysis of documentation and archival records
➢ Individuals (Marketing directors or any other representatives of the Marketing division of
an organization with the relevant expertise) for analysis of interviews.
As the main data collection method is interviewing let us discuss questions which
companies’ representatives should give answers to.
The main goal of conducting an interview is to obtain real-business data about gains,
opportunities, obstacles and all possible peculiarities of using big data in marketing at an
organization in Russia. Although during the literature review several international surveys on
managerial implications of big data technology have been reviewed, we would like to use as a
benchmark with slight modifications questions from the “Big data survey Europe. Usage,
technology and budgets in European best-practice companies” (2013). This survey, targeted
though at European market players, includes questions on very relevant and significant issues for
our study.
A structure of an interview will contain questions about the company’s general attitude
towards big data analytics, case-specific questions and a few wrap-up questions about the
expertise of foreign companies in this field, which can help us to get some insights about the last
research question of this paper.
The planned list of questions for an interviewee is demonstrated below:
1. What was the main reason for initiating analysis of big data in your organization and
which factors triggered this initiative?
2. In which departments does your company use big data analysis and which problems are
addressed by this technology?
3. What was the main reason for initiating analysis of big data in your organization as a
marketing tool and which factors triggered this initiative?
4. Which marketing-related problems are addressed with big data technologies in your
company?
5. Which areas of using big data in marketing would you evaluate as the most attractive and
promising (in short-term / mid-term perspectives)?
6. What kind of data do you analyze (at the moment and planned)?
7. What problems have you encountered when using big data?
8. Has the company achieved stated goals of the Marketing division of your company
through using big data analysis and which obstacles did your organization face?
9. Is there a comprehensive strategy for big data in the Marketing department of your
company?
10. What are the metrics which are used for big data marketing at your company?
47
11. Where does the big data analysis take place in your company from the perspective of the
organizational structure?
12. Which technologies and big data providers do you use or plan to use in your company for
big data analysis?
13. How would you assess the level of market development of big data solutions providers in
Russia?
14. Which alternative innovative solutions for marketing would you consider as most
attractive ones and what is the relative attractiveness of big data solutions?
2.5 Analysis of case study evidence
In order to provide meaningful results from the analysis of case study evidence, we aim at
developing convergence of evidence. In this case data triangulation from multiple sources of
evidence will provide multiple measures of the same phenomenon. Therefore, the multiple-case
studies findings will be supported by more than a single source of evidence, each of the cases
analyzed separately.
A report of each company case will include the following elements:
➢ key information about an organization (industry, number of employees, competitive
position in the market and some other indicators);
➢ demonstration of all relevant data analyzed through documentation and archival records;
➢ analysis of the data collected via interviews.
To analyze findings of multiple cases conducted, we plan to use explanation building
which is very similar to the approach of multiple experiments. Each case is analyzed and
explained separately and all results are revised and compared with all other cases and rival
explanations addressed in the literature.
48
Summary of Chapter 2
In the Chapter 2 we have introduced the methodology that is used in this research study,
which consists of 2 major elements: case study analysis and benchmarking, and justified the
suitability of the chosen methods.
We have also revealed the fact that this study is a subject of exploratory research which
will be based on the analysis of real-life evidence.
Major research questions have been also defined in this chapter. The first three research
questions, which specifically target Russian market and Russian companies, will be analyzed
with the help of case study analysis. In order to answer the fourth research question on
comparison of global practices with peculiarities of Russian market benchmarking method will
be used.
Research design of this study will be based on collection of both primary and secondary
data and include analysis of both quantitative and qualitative data.
Besides, the exploration purpose of this case study has been introduced, which is to get a
deeper understanding of the peculiarities of usage and implementation of big data for marketing
purposes by Russian companies. Necessary criteria for the exploration, unit of analysis and unit
of data collection have also been defined.
Moreover, the whole process of data collection and analysis were illustrated in this
chapter.
49
Chapter 3. Empirical Research
3.1 Empirical results of the study
3.1.1 General overview of the investigated companies
In order to analyze the current practices of Russian companies of big data analysis as a
marketing instrument four large national market players have been selected: MTS Group, one
major national mobile services provider, Ulmart and Lenta.
Table 4 General characteristics of the explored companies
Name of the
company
Industry
MTS Group
Telecommunications
A major Russian
telecommunication
s company
Telecommunications
Extremely large
(77,3 million
subscribers)
Extremely large
(74,8 million
subscribers)
Amount of data
generated by
the company
Lenta
Ulmart
Food retail
Online retail
Extremely large
(8,5 million
active loyalty
cards users)
Extremely large
(1,5 billion
active users)
Although selected organizations belong only to two industries, telecommunications and
online and offline retail, it is crucial to understand that currently in these economic sectors (apart
from a few other industries) of the Russian market big data marketing is in fact applied and huge
amounts of data are generated. Let us briefly introduce these companies and their role in the
Russian economy.
1.
Founded in 1993, MTS Group is the leading telecommunications group in Russia,
Central and Eastern Europe which provides wireless Internet access and fixed voice, broadband
and pay-TV to over 100 million customers. In 2015 MTS’s total subscriber base has increased by
3,6%, which is the lowest percentage among the “Big 3” players of Russian telecom industry,
and amounts to 77,3 million people. 14
To get a better understanding of the role of technology for MTS Group, it is important to
understand the market trends and current strategic vision of the company.
Telecommunications industry has been recently going through the transformation stage.
Rapid development of Internet and emergence of substitutes of traditional mobile services (e.g.
video communication services such as Skype or Facetime, messengers such as Whatsapp, Viber,
etc.) today are offering customers services of the same or better quality and broader functionality
at very low or, in most of the cases, zero prices. Therefore, telecommunications industry all over
the world has started to change the focus of their business and are now in the phase of turning
into a pure Internet provider.
14 MTS Group. From: http://www.mtsgsm.com/
50
The importance of the “voice” as a primary service of a classic telecom company is
dramatically decreasing and the organization is fully aware of this trend. However, “voice” as an
asset is still bringing good financial returns to the company and that is why the company is not
planning to fully undergo this transformation in the nearest future.
During the interview with the Senior Marketing Manager of the Strategy & Planning
department of MTS Group the following information was disclosed: the total revenue in 2015
reached the level of 391,2 billion rubles and was the largest volume among the “Big 3” players,
76% of which are represented by the revenue generated by mobile services. However, the
revenue volume, generated by the data transfer, constituted 20% (77,2 billion rubles) in 2015.
Moreover, this value demonstrates a 20% increase in demand for data transmission in
comparison with 2014.
In addition, it is worth mentioning that due to very low prices of a SIM card Russian
customers tend to change mobile providers extremely often, looking for more attractive
offerings. Churn rates in the telecom industry in Russia are impressive - scoring up sometimes to
40-60%. In the case of MTS Group, churn rate is equal to 40% which forces the company to look
for ways to prevent customer from changing a mobile service provider.
2.
The second selected company is a leading Russian universal telecommunication service
provider, operating in all segments of the telecommunications markets in Russia. The company is
well-known for its innovativeness and passion for technology. The subscriber base of the
organization amounts to 74,8 million people in 2015 which is a 7,2% increase in comparison
with the last year.
The industry trends, discussed earlier by the example of MTS Group, have the same
influence on this organization as they have on MTS. This company is still having a commercial
focus on conventional mobile services, however the organization is also concerned about the
importance of data transfer as a newly-emerged revenue source. As a result, today the company
is very much concerned with looking for alternative ways to increase sales, retain customers’
loyalty and minimize churn rates.
3.
The Lenta company was one of the first ones among other Russian retailers who has
started to work with big data analysis across different directions in 2013. Founded in 1993, today
Lenta is the top retail company in Russia in terms of the total sales floor space and 5 th in terms of
sales revenue (2015). The company operates in food retail, and has an established network of
142 hypermarkets in 70 Russian cities and 38 supermarkets in Moscow and St. Petersburg
(March, 2016).
4.
Finally, Ulmart is the Russian leader on online retail and the fifth largest Internet
company of the country. Founded in 2008, the company has managed to develop from a small
51
online seller of consumer electronics to the national e-commerce giant. Financial performance
demonstrates the company’s success in the market: in 2014 the company achieved $1,6 billion
sales growth. Ulmart is growing twice as fast as the whole e-commerce sector due to rising
financial opportunities of the Russian community seeking to fulfill demand for high-quality
products with efficient post-sale services and reliable delivery.
Ulmart is a very representational example of a successful player in the Russian internet
retail market. Data analysis and business analytics have always played a crucial role for this
organization as the company deals with loads of transactional data on a daily basis. The
emergence of new opportunities triggered by big data analysis was quickly adopted by the
company and is now been implemented across several business processes and, above all, in
marketing.
Before introducing a more detailed analysis of current practices of investigated
companies, let us give an overview of the structure of results presentation.
1. First of all, the process and the reasons for technology adoption will be discussed;
2. Secondly, the role of big data analysis as a marketing tool will be illustrated in frames of
organizational structures of every company;
3. Thirdly, the variety of marketing applications of big data analytics will be addressed
with a special emphasis on:
What kind of data is analyzed by the company;
What sort of marketing analytics is executed by the company with big data analysis;
Which marketing processes are optimized with the help of technology;
Which metrics are applied to measure performance of data-driven marketing tactics;
How companies evaluate effectiveness of data-driven marketing initiatives at their
organizations;
4. Major obstacles which investigated companies are struggling with in using big data
analytics for marketing purposes will be illustrated;
5. The role of alternative innovative business solutions will be discussed.
3.1.2 Adoption of big data analysis as a new technology for marketing and its role in the
organizational structure
Let us analyze how big data analysis for marketing purposes is implemented by Russian
companies as a part of their organizational structure.
52
Table 5 Adoption of technology & Big data marketing as a part of the organizational
structure
Name of the
company
Year of
technology
adoption
Big data
marketing as a
part of the
organizational
structure
MTS Group
2013-2014
CRM department
of the Marketing
division
A major Russian
telecommunication
s company
2014
Lenta
Ulmart
2013
2014
Marketing & Sales
departments
CRM department
of the Marketing
division
Marketing and
Advertising office,
Strategic analysis
and scenario
planning
department
As it is seen from the table, all companies have started to apply big data analytics only
several years ago, and today technology is executed for the purposes of Marketing departments
of the organizations. The case of every company is unique and different and that is why let us
discuss each case separately.
1.
The MTS Group has started working with data quite a while ago - as a part of database
management and business analytics. Although MTS has hired a new team of IT professionals to
build analytical models and work closely with big data, the organizational structure of the
company hasn’t changed much after the company officially started to analyze big data.
The interview has revealed that MTS is using big data mostly for marketing purposes.
Big data is being handled solely by the company’s technology experts in CRM department of the
Marketing division and there are no other employees involved in big data analytics outside of the
Marketing division.
2.
According to the available information about the second selected telecommunications
company, the organization’s capabilities of analyzing large data sets are truly impressive.
Nowadays every second there are 600 000 actions and events taking place at the company and
the system is capable of analyzing this information in real-time.
This company has started considering big data analysis around 4-5 years ago when this
topic gained a lot of attention from the business community throughout the world. The interview
with the Marketing director of the organization for the North-West region revealed that although
there is a general understanding of the need to benefit from big data opportunities, there is no
overall strategy in the organization towards managing big data on a corporate level.
53
In the organization big data analysis is widely applied at Sales & Marketing divisions
(commercial applications) which have started to use this technology in 2013-2014. However, big
part of big data analysis is used for the resolving technical issues.
As it was discussed in the theoretical chapter (Arthur, 2013), the lack of the data
enterprise strategy is a very common practice among companies who have already started to use
big data analysis, but have not yet understood the importance of building a comprehensive data
strategy for the whole organization.
Moving to the second analyzed market sector, retail is one of the industries which has
always been working hand in hand with large volumes of data. The rapid technological
development of big data analysis enabled retailers throughout the world to derive valuable
insights of the databases possessed and as a result improve performance of the company.
3. The Lenta company was one of the first ones among other Russian food retailers who has
started to work with big data analysis across different directions in 2013. Today Lenta considers
big data analysis as a promising growth opportunity which will help to create additional
capabilities and strengthen the competitive position of the company.
It is crucial to mention the importance of existing CRM practices of Lenta which create a
basis for data collection and its further analysis. Big data marketing at this company takes place
at the CRM department of the Marketing division.
The similar approach to big data analysis in retail was illustrated in the first chapter of
this study by the case of Target Corporation who also put CRM practices in the centre of big data
analysis at the company. Besides, the research of theoretical and practical publications has also
revealed that particularly for the retail sector loyalty cards and checks are the two most important
information sources which further on can be transferred into valuable insights about customers
behavior.
4. Another Russian retailer Ulmart started to resort to big data analysis approximately in
2014 as a part of its business analytics practices. In the opinion of the Ulmart representative,
there is a lot of unclarity concerning when conventional business analytics were taken over by
the era of big data analysis in the company, since the organization was analyzing the same vast
amounts of transactional data as in all previous years.
Ulmart defines big data not by its volume, but rather by emphasizing the marketing
perspective of the definition, velocity of data and emergence of advanced mathematical models
for processing data in real-time.
Therefore, we can conclude that Ulmart is determined to benefit from this new promising
technology, which is demonstrated by the top management’s interest expressed in mass media.
Yet it is important to keep in mind that the Russian largest online retailer has always been putting
54
an emphasis on business analytics and particularly web-analytics. That is why at the middle level
of the organization big data analysis initiatives are unlikely to be considered as a major
technological shift in Ulmart’s practices.
Concerning the place of big data analysis in frames of the organizational structure,
marketing-related big data analytics are executed inside the Marketing and Advertising office, in
the Strategic analysis and scenario planning department. Marketers of Ulmart also work hand in
hand with the so-called WEB Platform division which is responsible for managing a variety of
technological (namely IT) issues and a number of external IT providers.
3.1.3 Introduction to the real-life practices of big data marketing
Theoretical research, which has been justified by the empirical part of the study, has
demonstrated that big data analysis at an organization can be executed in two different ways with
different amount of investments required.
An organization can start collaboration with an external vendor of data analytics, a
competent IT company, which will provide a client on a regular basis with ready-to-use
analytics. This outsourcing approach requires much less financial investments and internal
organizational changes than another alternative which is based on building internal capabilities
of big data analysis and developing a complicated comprehensive data strategy for the whole
organization.
The empirical part of the research has demonstrated that Russian companies tend to favor
the first, faster and less expensive approach to big data analysis which allows them to benefit
faster from technology adoption and jump over the difficulties of implementation of an internal
organizational change.
Let us demonstrate an example of a similar practice by Lenta. Since 2013 the food
retailer has been collaborating with the IT company Emnos which has international expertise in
conducting advanced customer analytics for retailers. The retailer provides its partner with the
generated datasets while Emnos processes this information and shares with Lenta ready-to-use
customer analytics.
Lenta has started to collect the data about its customers in 2008 which was quite early in
comparison with other competitors and now has a significant record of valuable information
about its customers. For the collaborative big data project together with Emnos the data from
2010 has been used which enables Lenta to get a clear understanding of existing behavioral
patterns of its customers.
All in all, the specifics of the execution of big data analysis is similar to the prevailing
share of cases of both foreign and Russian companies, analyzed during the theoretical as well as
55
empirical parts of the research (SFR, Incity, War Gaming, MTS Group, Ulmart.ru). An
organization starts to cooperate with an external vendor who possesses valuable competences in
the technology and can optimize the process of data analysis for a client.
3.1.4 Data sources for big data marketing execution
To begin with, it is important to demonstrate which sources of information Russian
companies use for execution of big data marketing. A brief overview is illustrated in the Figure
below.
Fig. 3 Analysis of data sources used for big data marketing
In the first theoretical chapter of this study two major types of data sources have been
defined, with conventional data describing traditional enterprise data and recently emerged data
referring to all sorts of data which emerged with the rapid development and adoption of IT and
Internet by businesses as well as consumers.
The table demonstrates that generally speaking Russian companies use a combination of
data sources for execution of big data marketing. However, it depends severely on the industry
where a company operates how technologically advanced the information generated by an
organization or its customers is: telecommunications companies use a lot of data, which has just
recently emerged as a data type, while a bricks-and-mortar food retailer Lenta still focuses on
conventional data types in their analysis.
The data which representatives of the Russian telecommunications sector have at their
disposal, is truly diverse and include the following information:
56
Customer locations and customer journeys;
Billings information;
Multi-dimensional data on consumption of mobile services, text messages, etc.;
Content of the text messages;
Information about recipients of calls and text messages;
Multi-dimensional information about the internet traffic and data consumption.
During the empirical research major types of data for the Ulmart company have also been
revealed. The company, which today has in total 30 000 daily orders and 1,5 billion active users,
has both internal and external information systems which ensure accumulation, processing and
analysis of big data:
Internal information systems: ULMART website, SAP (Business Intelligence, ERP,
CRM), Oracle Database;
External information systems: Google Adwords, Yandex metrics, an analytical program
of customer recommendations, a system of prices monitoring, etc.
15
Since Ulmart belongs to the sector of e-commerce, it is unsurprising that special attention
in the organization is paid towards such internal information source as the company’s website
(the so-called streaming data). It includes data about almost everything that customers do on the
website:
purchasing history;
time used for searching;
type of device from which the search was conducted;
customer preferences;
frequency of using the website;
socio-demographic data of customers;
variety of advertising tools through which a customer has gone through before landing at
the Ulmart’s website.
3.1.5 Marketing processes optimized with big data analytics
Before starting a discussion of which marketing applications of technology are most
commonly used by Russian companies, let us refer to the theoretical review and the overview of
marketing strategy processes.
15 Ulmart. For investors. From:http://investors.ulmart.ru/
57
Fig. 4 Overview of marketing strategy tools applied by investigated companies
As it is seen from the table, the largest variety of applied data-driven marketing
instruments belong to the planning stages of the marketing strategy: market selection and design
of the marketing mix. Big data analytics are also frequently applied in the field of marketing
management and control and sometimes for the purposes of market research.
In order to have a broader overview of marketing application of technology by selected
organizations let us analyze each company case separately.
1.
One of the major purposes of big data analysis at MTS Group is churn prevention and
increase of customer loyalty. However, the company’s representative shared with us his
viewpoint that such things are very hard to be measured.
The current focus of the company on using big data for improving CRM systems
performance was illustrated in the Cnews Analytics report (2013) which claimed that CRM
applications of big data are the second most demanded category by Russian companies.
Let us analyze all marketing-based problems addressed with big data analysis at MTS
Group. Generally speaking, all marketing applications of the technology can be classified as
customer base management or commercialization of the database.
Segmentation programs form the basis for all further big data marketing applications and
include data-driven analysis of consumer behavior, tracking of customers purchases and a few
more things. With the help of segmentation on the basis of big data analysis MTS is able to
improve effectiveness of the following marketing actions:
Targeted advertising;
Building a 360-degree customer view;
58
Development of personalized service offerings
Prioritization of customers based on analysis of their lifetime value and profitability;
Up-sale and cross-sale initiatives.
In the opinion of the company’s representative, the main goal of data-driven segmentation
and targeting initiatives is to increase the volume of such performance indicator, as average
revenue per user (ARPU).
The variety of marketing applications of big data analysis which MTS Group is currently
using has been already illustrated by such authors as Feinleib, Arthur, Kash and Calhoun and
prove the viewpoints of these researchers.
2.
The second telecommunications market player is also affected by the industry trends,
discussed earlier by the example of MTS Group. This company is still having a commercial
focus on conventional mobile services, however the organization is also concerned about the
importance of data transfer as a newly-emerged revenue source.
As a result, today the company is very much concerned with looking for alternative ways
to increase sales, retain customers’ loyalty and minimize churn rates. That is why marketingrelated problems solved by big data analysis at the organization are above all centered on
customer analytics, which proves the viewpoints demonstrated in the first chapter (Columbus,
2014, 2015; Datameer, 2014).
Big data analytics enable the organization to develop various segmentation and targeting
programs and build advanced consumer profile models with consumer behavior prediction.
Applications of findings from these segmentation initiatives are truly diverse.
The data-driven analysis allows the company, for example, to get a better understanding
of the frequency of using different services provided by this operator, and decide on particular
personalized offerings. If the company receives the information from data analysis that a client is
not using conventional calls at all and mostly focuses on the Internet and text messages, the
company sends him messages offering him to change to a new, more suitable and attractive tariff.
Targeted advertising is another marketing tactic which is based in the organization today
on big data analysis. Marketing Director of the company shared the information how it really is
executed in practice. For example, when company observes that a customer is located at the
airport at the moment and his stay at the airport lasts more than 3 hours, the model assumes that
this person is planning to travel abroad and suggests to send this customer a message with an
offer to sign up for roaming options.
To sum up, marketing applications of big data analysis in the organization are centered on
increasing customer satisfaction and making customers stay the organization in today’s complex
59
for the telecommunications industry environment through personalized offerings and customized
approach.
3.
Let us have a closer look which purposes big data analysis serves at the Russian retail
giant Lenta. During the interview three major technology application areas has been revealed:
customer loyalty increase, growth of sales volumes and operational efficiency increase.
As it has been previously stated, the information from the Lenta’s CRM system is one of
the major data collection sources for the company. The information which customers have to fill
in to get a loyalty card varies from conventional sociodemographic characteristics such as age
and gender, contact information to such noteworthy customers’ attributes as marital status and
car ownership.
The company has started to issue loyalty cards for its customers back in 2000 and today
more than 2,5 million people already possess across the country possess these cards. According
to the available information, today the loyalty program of Lenta is one of the largest and most
effective loyalty programs of the market, however the idea behind it does not differentiate
significantly from any other loyalty program. All card owners get a 5% discount for all products
assortment and up to 50% reductions for temporary special price offerings. It is important to
mention the huge amount of active users of Lenta loyalty cards which accounted to 8,4 million
people by the end of 2015. Besides, 92% of all purchases at Lenta are made with the help of
these loyalty cards.
As a result, it is not surprising this loyalty program as a part of the customer relationships
management has become the kernel of the big data analysis at Lenta, as it enables the retailer to
get a deep understanding of its customers’ behavior, buying habits and preferences and
significantly upgrade existing analytical customer behavioral models.
The joint big data marketing project of Lenta executed together with the Emnos company,
which was introduced earlier, is comprised of four major work streams which are targeted at
increasing customer loyalty and sales volumes:
Pricing program to define the best price levels for the whole assortment of products;
Lenta category management program to improve management of different product
categories of consumer baskets;
Direct marketing to develop and customize promotional activities to customers’ needs;
Emnos tools to measure and evaluate performance of the data-driven marketing actions
of the company.
The interview with the CRM Director has demonstrated that the segmentation forms a
basis for big data marketing at Lenta. The major goal of this initiative is to stop focusing on an
average customer needs and develop a more sophisticated and personalized segmentation
60
program. With the help of this technologically advanced segmentation platform Lenta has
analyzed customers data, generated since 2012, segmented and prioritized them according to
their relative attractiveness and profitability towards the company, their buying habits and
purchase reasons and preferences. In order to assess profitability of a customer the RFV
approach was applied which analyzes such indicators as recentness, frequency and value of
purchases.
As a result, Lenta has identified 10 major segments and 2,5 million most attractive
customers - loyalty cards holders. Besides, the company has adjusted merchandising strategy and
modified products assortment on shelves in every single store of the chain.
It is also worth illustrating several segments that Lenta managed to identify with the help
of big data analysis: budget cooking and party people segments. The retailer has revealed from
the data possessed that there is a large share of customers who prefer to cook at home and who
favor private brands of the retailer and on the other hand there is a huge segment of customers
who just buy alcoholic beverages, refreshing drinks and various snacks.
Besides, the segmentation program has revealed two major shopping habits of its
customers and analyzed profitability of these customer groups: while some customers of Lenta
do shopping very frequently, but the value of their purchases is small, there is another large
group of customers who buy products once a week, but spend on shopping significant sums of
money. Although it has been analyzed by the company that profitability of these two segments is
almost the same, this data-driven segmentation program enabled the retailer to get a deeper
understanding of needs and behavioral patterns of these customers and modify the marketing
strategy for every store (whether to focus on increasing average value of the check or frequency
of purchases).
The pricing element of the data-driven marketing strategy at Lenta is concerned with
analysis of price sensitivity of customers which allows the company to adjust current price levels
for every single product of the overall available assortment across all regions, keep its
positioning as a budget-priced retailer and therefore increase customer loyalty.
On the basis of the segmentation program Lenta has launched a direct marketing program
of personalized offerings “Thank you mailing” for 2,5 million most attractive customers all over
the country offering different customer segments customized coupons and discounts. The
program is executed across several channels: post mailing, email mailing and SMS mailings.
Regarding returns of the personalized recommendations, the response rate accounts for 4-6%
which is considered as effective by the company.
Big data is used at Lenta also for the purposes of performance evaluation of various
marketing actions, e.g. promotional activities.
61
4.
Regarding big data as a marketing instrument, at Ulmart it is analyzed, above all, for the
purposes of Marketing division, but also for resolving problems from Sales and Customer
service departments.
Prevailing part of big data marketing at Ulmart is centered on developing personalization
and customization of the website which is the single sales channel for the company and heart of
the business. In order to achieve this, Ulmart puts a great emphasis on using big data for
increasing effectiveness of STP-processes (segmentation, targeting, positioning). Thanks to the
technology the company is able to identify and target many more segments of different size,
relative attractiveness and other attributes that it used to. Integrated analysis of the streaming
data along with geomarketing enable Ulmart to execute customer profile identification and
develop a diversified, yet personalized marketing strategy which will target specific segments or
customer profiles. In strategic marketing big data analysis helps Ulmart to do scenario planning
across different customer segments and anticipate customers’ behavior.
Another large group of marketing applications of big data at Ulmart is connected with
CRM system development. The company’s project on development of recommendations system is
one of the most interesting examples of successful targeting of identified micro-segments. With
the help of big data analysis of purchases across various product categories the company
revealed a particular customer segment of “young parents” who buy baby diapers much more
frequently than any other segments. As a result, the company designed personalized offerings for
this segment and ended up with a 30% increase of the customer response rate, 2,3% increase of
conversion rate and the tripled value of CTR (click-through-rate).
It is worth mentioning that targeting is executed by Ulmart with an integrated 360-degree
view of customers which includes communication not only through the website design, but also
through email targeting and all other channels of digital marketing.
Finally, Ulmart resorts to big data analysis for marketing actions’ performance
evaluation. On the regular basis the company evaluates effectiveness of promotional activities
across different digital channels, analyzes customer searching journeys and as a result adjusts its
actions.
In addition to the analysis of marketing-oriented big data projects, let us consider an
example of successful big data application from Ulmart’s customer service division. Several
years ago Ulmart executed an initiative in Moscow “Delivery 2.0” which aimed at improving
customer service in terms of product delivery and used big data as a basis for analysis. Ulmart
revealed specific patterns and preferences of Moscow customers from its internal data and
realized that the citizens of the Russian capital seek much more flexibility in delivery timeframes
than any other customer segments across the country. As a result, Ulmart implemented changes
62
in delivery options and made them much more customized for the needs of Moscow citizens. The
project paid off initial investments and persuaded the company to continue using big data
analysis for boosting company’s success.
The variety of marketing applications of the technology currently implemented by the
online retailer has been already illustrated by such authors as Feinleib, Arthur, Kash and Calhoun
and prove the viewpoints of these researchers.
Being in the online retail market for many years, the company feels quite comfortable
with existing set of metrics, prevailing part of which belong to conventional digital marketing
and web-analytics performance indicators. Among a wide number of KPI’s used to assess big
data projects the Manager emphasized the following indicators:
CTR ratio (click-through-rate indicator, which is used for analyzing effectiveness of digital
marketing actions and which demonstrates how many customers clicked on a particular
link);
Conversion rate (widely used digital marketing indicator which evaluates performance of
marketing actions and shows how successfully these actions drove customers to make a
payment);
Open rate (used for email marketing, demonstrates how many customers have opened the
sent email).
To sum up, it is noteworthy to refer to the theoretical review of the first chapter and
demonstrate two types of marketing analytics which big data analysis is able to execute:
behavioral a n d transactional analytics. The figure below shows which type of analytics
investigated Russian companies tend to favor.
63
Fig.5 Marketing analytics executed by Russian companies as a part of big data marketing
The table demonstrates that there is evidence of Russian companies paying special
attention towards behavioral analytics and derivation of insights about consumer behavior.
Companies use big data analytics to execute a wide number of marketing processes while
transactional analytics are applied to tackle particular issues.
Besides, Russian companies explore new areas in big data marketing and use equally
mature analytical applications of big data marketing such as optimization of marketing
campaigns, customer loyalty management, in-store custom analytics and emerging analytical
applications has been currently implemented (ad targeting optimization, customer churn
prevention).
3.1.6 Alternative applications of big data analysis
Empirical analysis has demonstrated that big data analytics is applied by Russian
companies not only as a marketing tool, but also as an instrument to solve technical and
infrastructural issues which indirectly help companies to be more customer-oriented and increase
customer satisfaction.
1.
Apart from pure marketing applications of big data analysis, MTS Group benefits from
the technology while solving also infrastructure-based problems:
➢ Optimization of infrastructure management at the company: Today the company cannot
afford to waste resources and that is why MTS has started to base decision-making of
locations for building new basic stations on the traffic analytics of customers, while
previously those decisions were taken without an analytical justification behind them.
➢ Maximal utilization of traffic: results of traffic analytics of customers allow the company
to optimize the technical side of the traffic management.
According to the company’s representative, both major groups of applications, customeroriented and technical, are equally important since they reduce costs of the company.
2.
Big part of big data analysis at another mobile services provider is also used for the
resolving technical issues. The technology is used to decide on the number and locations of new
basic stations. Vast amounts of geolocational data that the organization has at its disposal allows
the company to understand the mobility flows as well as traffic usage of all sim card owners. As
a result, it becomes much easier to optimize investment portfolio, plan construction initiatives
and reduce technical problems in particular locations.
64
As telecommunications operators generate vast amounts of data, this information can be
used not only for internal purposes. Geolocational information has released a new opportunity
for the company to analyze locations of the network of clients and sell this data to third parties.
In 2013 the organization presented a project which was centered on using geolocational
data for resolving traffic jams in Moscow and developing transport infrastructure. The full-scale
geolocational project was initiated by the company already in 2013 and in 2015 the company
started collaboration with a Russian transportation company and the project moved on to the
testing stage.
Therefore, transportation companies, retailers or real estate companies may possibly
reach the company for providing them detailed information about traffic of people, sociodemographic characteristics of these people, their purchasing power, purposes of travel, etc. As a
result, big data analysis can play the role of not only internal growth driver, but also as an
external source of generating higher sales volumes by diversifying the business and offering new
services.
3.
As for the examples from other industries, once Ulmart analyzed the influence of
Javascript errors (when customers experienced technical problems searching for a product at
Ulmart online) on the overall conversion rate and revealed a strong relationship between these
two indicators.
3.1.7 Effectiveness of data-driven marketing practices
Regarding analysis of effectiveness of data-driven marketing practices, opinions of
explored companies differ and provide different perspectives.
1.
For instance, the Manager of the MTS Group shared his viewpoint that MTS does not see
real value addition of big data analysis. In his opinion, in case of a telecommunications company
big data is more of a “media fuss” rather than a real phenomenon. Big data analytics do not
actually bring substantial profits to the company, although there are some records of successful
initiatives.
However, the literature review has demonstrated (CIO online journal, 2014) that
companies should not expect quick financial returns from the implementation of the big data
technology, instead it is recommended to focus on long-term benefits of this technology and keep
patience.
2.
The interview with the Marketing director of the second telecommunications company
has demonstrated that the gains are seen, above all, in investment portfolio optimization programs.
Due to the smart and data-driven analysis the organization opened 10 times more stores and 100
65
times more basic stations across the country. As for marketing, returns have been gained due to the
targeted advertising and personalized offerings campaigns.
3.
Regarding Ulmart, in an interview with Cnews in 2014 former Client analytics Manager
of Nikolai Valiotti stated the strategic importance of using data analysis which was totally
supported at the corporate level. He also ensured that particularly big data is one of the directions
to drive the company’s future growth.
Besides, Valiotti shared the information about the company’s improved market
performance, which was achieved due to the data-driven strategy, and expressed Ulmart’s huge
interest in investing in big data technologies for the upcoming years (the budget of the business
analytics department increased by 3 times in 2014-2013).
From the Ulmart’s perspective rewards from big data analysis generally outpay initial
investments. Gains and real benefits are especially seen in the big-data-driven redevelopment of
recommendation systems (e.g. the one, which targeted young parents buying baby diapers) when
Ulmart conducted a deep analysis of customers’ behavior on the website and derived meaningful
insights. In terms of performance evaluation of marketing data-driven actions, such KPI’s as
open rate and click-through-rate have improved much more than others.
Therefore, we can conclude that Ulmart is determined to benefit from this new promising
technology, which is demonstrated by the top management’s interest expressed in mass media.
Yet it is important to keep in mind that the Russian largest online retailer has always been putting
an emphasis on business analytics and particularly web-analytics. That is why at the middle level
of the organization big data analysis initiatives are unlikely to be considered as a major
technological shift in Ulmart’s practices.
3.1.8 Major market trends and obstacles of big data analysis as a marketing instrument
Execution of big data analysis for marketing purposes is connected with a number of
obstacles which Russian companies are trying to overcome. Let us illustrate the most important
of them for every company.
1.
Regarding the obstacles which MTS is facing, they include the following barriers:
Evolving area of big data technology, and MTS has to learn from its own mistakes;
Regulatory risks based on privacy legislation in Russia (however, these issues are
extremely strictly regulated inside the organization in order to prevent leak of data);
Lack of relevant metrics and measurement instruments.
66
Investments into new technology are not seen as a barrier by MTS since it already has an
existing team of BD professional and relevant installed equipment which contradicts the
viewpoints expressed in publications from Cnews Analytics and Oracle (2014).
The manager of the company also demonstrated how the company considers the
influence of current market situation on the Group’s using big data analysis in marketing:
Economic crisis triggers companies to search for new ways for costs reduction and big
data analysis has a lot to offer here;
As discussed, there is a slow-down trend in the telecom industry, since the voice as an
asset has dramatically lost its value over the recent years, and telecom companies have to
look out for alternatives;
Business community is generally becoming more tech-savvy and more knowledgeable
about technology and the power of data - that is where telecommunications companies
have a lot to offer in terms of their analytical capabilities.
2.
There are several significant barriers which do not allow another Russian
telecommunications company to realize the full value of technology adoption. In the company’s
representative opinion, the main problem is connected with the lack of competent data science
professionals. As several other Russian companies, with the emergence of big data analytics as
one of the major growth drivers for business the organization went with the flow and started to
adopt this technology in the organization by internal development of models based on big data
analysis.
This decision of the company reflects the results of the analysis by Cnews Analytics
(2014) which has revealed that Russian companies not only resort to external providers of big
data analysis, but also develop their own models and capabilities in this area.
However, the insufficient level of market development and lack of relevant instruments
obstruct further execution of the technology. Finally, strict privacy legislation in Russia is one of
the major obstacles with regards to using big data analysis and especially selling the data to the
third parties, and that is why the company is very concerned with ensuring the privacy of
customers’ personal data. As a result, all these factors lead to the situation when the company
develops its own models for using big data analytics instead of resorting to external vendors.
Concerning future plans of the company and willingness to invest in technology and
innovation, the Marketing Director shared with us that there is a number of alternative innovative
tools in marketing which the company is also ready to invest in. These alternatives include,
above all, investments in mobile TV streaming and development of mobile commerce and reflect
current market trends of the telecommunications industry.
67
3.
Taking into account the retail sector and the Ulmart company, there is a number of
challenges which the company is trying to overcome. First of all, marketers at the company face
difficulties at the initial stage of research: revealing and describing a problem as well as deciding
what kind of data is necessary for a particular problem. Variety of information sources often
make marketers feel frustrated and lost in the thousands of customer characteristics, structured
and unstructured, quantitative and qualitative.
Another barrier is connected with the early stage of technology adoption by the company
and the emergence of this market in general. In the opinion of the Ulmart’s Client analytics
Manager, “it will take quite some time to understand the real benefits of big data analysis and
exactly the same happened with the emergence of Internet technology”.
Finally, concerning the competences of the human resources for analysing big data, the
company does not see this as a major problem. Ulmart confirms that there is generally lack of
competent data scientists in the market, but on the other hand the company is investing a lot in
education and training of its own employees. For example, Ulmart is hiring recent graduates
from relevant studies (Higher School of Economics, Moscow, Business Informatics Faculty;
Saint Petersburg State University, Faculty of Economics, Business Informatics program) and
trains them to become professionals at big data analysis and interpretation.
All in all, the majority of problems which Russian companies are trying to overcome
have been demonstrated in the publications dedicated both to global market trends as well as
peculiarities of the Russian market (Oracle, 2014; Cnews Analytics report, 2014; McKinsey
Global Institute, 2011; Minelli, Chambers, Dhiraj, 2013; Dietrich, Plachy, Norton, 2014).
3.1.9
The role of alternative innovative business solutions
The empirical research has demonstrated that the importance of big data marketing is
planned to remain at the high level as it is at the moment, however there is no doubt that the
investigated companies are considering investments also in other innovative solutions for their
businesses. Let us provide several examples of how companies consider future opportunities for
big data analytics and other innovations if there are any.
1.
From the perspective of MTS, future opportunities of using big data are truly diverse:
Further development of mobile commerce (there are some steps made by the company,
however legislation factor puts limitations on further commercialization of the
technology);
The long-term future goal defined as “selling ultra-personalized services and analytics
for the mass market”: providing services for third-parties: market analyses, targeted
68
advertising, variety of ready-to-use analytical data across customer segments and
different characteristics from the MTS’s analyzed databases;
Building partnerships with other companies with top-quality services or any other
competitive advantages and creation of synergy (collaboration with banks (credibility of
customers), Uber or Yandex as potential examples);
Improvements in service quality: the example of the joint project with Yandex Maps
enabling customers with Android smartphones to share their comments about connection
quality across different locations which MTS Group is processing.
Internet of things is also driving big data development;
MTS is thinking about selling to B2B and B2G clients ready-to-use analytics on the basis
of databases which the company possesses.
2.
As for the future plans of using big data analysis for marketing purposes at Ulmart, there
is a focus on the following initiatives:
The company is currently testing the model of attribution of revenues achieved by various
digital advertising actions. Digital advertising strategy of Ulmart is executed across
different channels, platforms and marketing intermediaries and therefore the company
wants to be clear about which advertising actions bring customers to the website most
successfully.
CRM system of Ulmart will continue to use big data analytics for further development
and particularly for aggregation of loyalty programs;
Development of personalization programs which will be based on deep analysis of
customer profiles and behavior patterns;
Integration of offerings across different search devices (personal computers, smartphones,
tablets, etc.) in order to gain a deeper understanding of the customers and optimize
marketing actions. One of the triggers of this initiative was the company’s recent analysis
of the number of new customers which demonstrated that a major part of them were the
same individuals who used different means for searching.
Implementation of Tableau software for data visualization which will enable marketers of
Ulmart to easen analysis of big data.
However, it is important to say that big data technology is not the only innovation driver
for Ulmart. The company positions itself as a strong leader in Russia’s e-commerce business and
considers a number of innovative technology-driven initiatives. Recently the company presented
a full-scale project at MIPIM (the world’s leading conference on the property market) dedicated
to the development of the e-commerce base which will unite and optimize logistics of all
69
physical outlets of the company. 16 Moreover. the company is considering such opportunities as
development of mobile services and an idea to become the national centralized cybermarketplace (the so-called “Russian Alibaba” project triggered by the Russian government)
which will enable Russian companies to export their products via this channel. 17
Although there are some other innovative projects that Ulmart is planning to execute, big
data analysis remains to stay very important for the company’s development.
3.2 Key findings of the empirical research
3.2.1 General overview of results
In this part of the study key conclusions of the findings from theoretical as well as
empirical research are converged and key factors and dimensions of specifics of big data analysis
as a marketing tool by Russian companies are illustrated. Besides, special attention is paid to the
critical comparative analysis of practices of Russian and foreign companies across identified
characteristics.
The brief overview of differences in technology implementation by the investigated
Russian companies is demonstrated in a table below.
Table 6 Comparison of big data marketing practices of four Russian companies
Name of the
company
MTS Group
A major Russian
telecommunication
s company
Lenta
Ulmart
Telecommunication
s
Extremely large
(77,3 million
subscribers)
Telecommunications
Food retail
Online retail
Extremely large
(74,8 million
subscribers)
Extremely large
(1,5 billion active
users)
Technology
adoption
Big data
marketing as
a part of the
organizationa
l structure
2013-2014
2014
Extremely
large (8,5
million active
loyalty cards
users)
2013
CRM department of
the Marketing
division
Marketing & Sales
Departments
CRM
department of
the Marketing
division
Marketing and
Advertising
office, Strategic
analysis and
scenario planning
department
Types of data
used for big
data
marketing
Customer locations and customer journeys;
billings information;
multi-dimensional data on consumption of
mobile services, text messages, etc.;
content of the text messages;
Purchasing
history,
checks;
recentness,
frequency
Purchasing
history;
time used for
searching;
Industry
Amount of
data
generated by
the company
2014
16 ATI-Media.Ulmart will demonstrate a new project in Cannes at MIPIM. (2016).
From:http://ati.su/Media/News.aspx?ID=84424&HeadingID=13
17 Rosbusinessconsulting. Ulmart will follow the Alibaba’s way. (2015).
From:http://www.rbc.ru/business/24/06/2015/558976a89a7947333e5f2f5a
70
information about recipients of calls and and value of
text messages;
purchases
multi-dimensional information about the
internet traffic and data consumption.
Marketing
applications
of technology
Other
applications
of technology
Effectiveness
of big data
analysis in
marketing
Barriers
Future
development
of big data
marketing
CRM;
commercialization
of the database;
segmentation;
analysis of
consumer behavior;
prioritization of
customers based on
analysis of their
lifetime value and
profitability;
development of the
360-degree
customer view
Optimization of
infrastructure
management at the
company;
Maximal utilization
of traffic
Not yet significantly
effective
Segmentation and
targeting programs;
consumer behavior
prediction;
churn prevention
and retention of
customer loyalty
Evolving area of big
data technology;
regulatory risks;
Lack of relevant
metrics and
measurement
instruments
Lack of competent
human resources;
insufficient level of
market
development;
lack of relevant
instrument;
privacy legislation
Upscaling and
expansion of current
big data marketing
practices
Ultra-personalized
services and
analytics for the
third parties;
improvements in
service quality
Segmentation
, targeting;
customer
loyalty
increase;
CRM, pricing
programs,
category
management,
direct
marketing;
performance
evaluation
Optimization of
investment portfolio
type of device for
searching;
customer
preferences;
frequency of
using the website;
sociodemographi
c
data
of
customers
Segmentation;
targeting;
personalization
and
customization of
the website;
CRM;
development of
the 360-degree
customer view;
performance
evaluation
Reduction of
technical errors
Yes, personalized
recommendations
systems
Yes, CRM
and
personalized
offerings
Lack of
competent
human
resources;
insufficient
level of
market
development
Upscaling
and
expansion of
current big
data
marketing
practices
Yes, personalized
recommendations
systems
Revealing and
describing a
problem;
Insufficient level
of market
development
Model of
attribution of
revenues
achieved by
various digital
advertising
actions;
aggregation of
loyalty programs;
integration of
offerings across
different search
device;
implementation
of Tableau
71
Future plans
to invest in
other
innovative
business
solutions
Internet of things;
mobile commerce
Mobile TV
streaming;
development of
mobile commerce
-
software
Development of
mobile services;
optimization of
logistics
The table demonstrates the summary of all aspects of big data marketing practices of
investigated Russian companies which have been addressed during the empirical part of the
research. Let us outline major findings from each aspect of technology execution.
3.2.2 Country-specific level of market development and technology adoption
While today the share of foreign companies which already use big data analysis in their
business accounts for approximately 30% of all business entities, less than 10% of Russian
companies in fact use this technology.
Moreover, the majority of the Russian companies which exploit big data analysis are
represented by the large market players - market leaders among telecommunications companies
(Vympelkom, Megafon, MTS Group), a number of national retailers (Ulmart, Lenta, X5 Retail
Group, Gloria Jeans) as well as banks (VTB24, Alpha Bank, Sberbank) and government
institutions (Federal tax authorities).
According to statistics, only 10% of 108 large Russian organizations in fact use
extensively big data analysis in their operations (CNews Analytics; Oracle, 2014).
The multiple case study analysis of 4 Russian companies has demonstrated that there is a
number of reasons which explain the current state of technology adoption by Russian companies.
First of all, the rapid growth of big data market as well as technology adoption started for
foreign companies 5-7 years ago which was relatively earlier than for Russian organizations.
According to the study’s findings, Russian business resorted to big data analysis only 2-4 years
ago.
Therefore, foreign companies have had more time to master big data execution at their
organizations and benefit from the higher stage of market development, while Russian
companies remain pioneers in this field. The have just started to build and leverage capabilities,
which is a very time- and resources-consuming process with no quick financial returns.
However, it is necessary to mention that although technology adoption levels may vary
across countries, the global big data market still remains at the relatively early stage of
development.
72
3.2.3 Financial resources for technology execution
Secondly, since investments in a new technology are always connected with heavy
financial implications, it is not surprising that only large Russian market players with substantial
financial resources can afford investing in execution of big data analysis.
However, big data analysis at an organization can be executed in two different ways with
different amount of investments required. An organization can start collaboration with an
external vendor of data analytics, a competent IT company, which will provide a client on a
regular basis with ready-to-use analytics. This outsourcing approach requires much less financial
investments and internal organizational changes than another alternative which is based on
building internal capabilities of big data analysis and developing a complicated comprehensive
data strategy for the whole organization.
The empirical part of the research has demonstrated that Russian companies tend to favor
the first, faster and less expensive approach to big data analysis which allows them to benefit
faster from technology adoption and jump over the difficulties of implementation of an internal
organizational change.
Three out of four analyzed companies have decided to implement internal changes only
in human resources management, hiring data scientists and architects and training and educating
existing Marketing and CRM managers to upgrade their analytical capabilities, and leave
technology development to external vendors. Yet one telecommunications company, studied
during the research, in addition to the education of employees is trying to achieve the goal of
building internal technological knowledge of big data analysis and does not want to rely on thirdparty suppliers of the big data platforms.
As a result, it can be concluded that although financial resources play an important role
for a Russian company in deciding whether to resort to big data analysis, the relative strength of
this factor may vary since companies face different options.
Thirdly, empirical research has demonstrated the paramount importance of such industryspecific factor as existence of massive data sets at an organization large enough to be analyzed
with big-data-based platforms. During the theoretical literature review the industries with the
highest technology adoption level have been revealed which are the same for Russia and the
global market: financial services, IT services, telecommunications, retail and healthcare.
These industries share one characteristic, crucial for understanding the problem
investigated in this study, - they all have historically been having extremely large amounts of
data about their customers at their disposal, analysis of which can tackle a wide number of
business-related issues and above all, marketing-based problems.
73
Empirical research of four Russian companies has proved the real-life importance of this
factor: today Russian organizations solely from the several identified economic sectors are in
fact using big data analysis in their business operations. If to take an example of the
telecommunications and retail industries, which the investigated companies belong to, they all
have been always generating large amounts of data: purchases history, geolocational data,
information from the loyalty cards, usage of Internet traffic, distribution of usage of different
mobile services, sociodemographic data, etc. Unsurprisingly the emergence of a new technology
was perceived by these companies as a way to increase effectiveness of the existing managerial
practices.
3.2.4 External environment
Finally, when discussing the reasons for technology adoption by Russian companies it is
also important to consider external triggers such as macroeconomic environment or industryspecific economic environment.
Multiple case study analysis has broadened the overview of market trends discussed in
the first chapter and demonstrated some other external reasons for adopting big data analysis in
marketing by Russian companies.
Today Russian economy is at the stage of a deep and continuous recession and that is
why companies are looking for all sorts of alternatives to reduce costs, increase efficiency and
obtain currently extremely price-sensitive Russian consumers. The investigated companies
confirmed that marketing applications of big data analysis enable them to stay competitive by
enhancing customer-orientation.
In addition, let us consider the impact of industry-specific economic environment on
technology adoption. For example, the empirical research has demonstrated that
telecommunications companies are particularly interested in big data analysis as a marketing tool
because of the variety of options and its capabilities to overcome current industry challenges,
reduce churn rates and increase customer loyalty via personalized offerings and discounts.
As a result, it can be concluded that any external environment, whether it is country-,
industry- or company-specific, may have a significant influence on adoption of big data analytics
as a marketing instrument.
3.2.5 Organizational structure and data management
Let us discuss the role of big data marketing in frames of the organizational structure of
Russian companies and analyze the changes in internal business processes followed after the
adoption of this innovative technological tool for marketing purposes.
74
The analysis of multiple cases has revealed the fact that none of the investigated
companies has yet gone through a major internal transformation after organizations started to use
big data analysis. As it has been previously discussed, Russian companies tend to prefer
collaboration with external vendors and therefore the major challenge in using technology is
mostly connected with human resources management and hiring truly competent data scientists
and architects.
As it has been previously stated, big data analysis at Russian companies is applied mostly
for solving marketing-related problems and that is why the execution of technology as a rule
takes place in Marketing and CRM departments of organizations and sometimes in Business
intelligence department if there is any.
None of the organizations have yet decided to build a comprehensive data strategy at
their organization to develop data management across all internal organizational units. In most of
the cases Russian companies have just recently started to use the technology and they are still at
the stage of testing effectiveness of big data analytics.
As it was discussed in the theoretical chapter (Arthur, 2013), the lack of the data
enterprise strategy is a very common practice among companies who have already started to use
big data analysis, but have not yet understood the importance of building a long-term data
management strategy. These results are not surprising, because such a decision requires resources
to bear substantial financial implications and high interest and dedication from top management
to overcome the challenges of managing an internal organizational change.
The case of a major French telecommunications company SFR, illustrated in the first
chapter, demonstrates the benefits of development of data strategy for the whole organization
which enabled SFR to store and analyze data about the customers and derive valuable insights on
a regular basis and improve significantly data warehouse performance.
3.2.6 Organizational competences of data management
It is also crucial to emphasize the impact of such factor as the existence of organizational
competences to analyze and derive valuable insights out of this data. Empirical research has
demonstrated that there is a major difference in considering big data analysis among companies
which already have existing practices of conducting business analytics on a regular basis and
among those who have never paid a lot of attention towards analysis of data generated by the
company.
For example, for Ulmart, which is a pure e-commerce player, data analysis and business
analytics have always played a crucial role as the company deals with loads of transactional data
on a daily basis. From the very beginning of the business Ulmart already had experienced
75
employees in the Marketing division to execute web analytics and developed a system of
necessary performance metrics.
The emergence of new opportunities triggered by big data analysis was quickly adopted
by the company with several minor organizational changes and is now been implemented across
several business processes and, above all, in marketing. Therefore, the process of technology
adoption was rather easy for Ulmart: the company just hired a few more competent employees
and extended the personnel of the Strategic analysis and Scenario planning department which
belongs to the Marketing and Advertising office.
As for MTS Group, the company has started working with data quite a while ago - as a
part of database management and business analytics. Although MTS has hired a new team of IT
professionals to build analytical models and work closely with big data, the organizational
structure of the company hasn’t changed much after the company officially started to analyze big
data. Big data is being handled solely by the company’s technology experts in CRM department
of the Marketing division and there are no other employees involved in big data analytics outside
of the Marketing division.
Another telecommunications company executes big data analysis in Marketing
department while Lenta applies big data analysis inside CRM department of the Marketing
division.
To sum up, during the case study analysis it has also been revealed the lack of
understanding of big data analysis as a new technology by Russian companies as a result of
unclarity of definitions of big data analysis and innovativeness of the topic. Not only the
researchers are facing difficulties in defining big data, as it has been demonstrated in the
theoretical chapter, but also business practitioners. It is often complicated for the companies to
define their current practices from marketing, CRM or business analytics departments on the
basis whether they belong to big data analysis or not.
Nevertheless, all investigated companies agreed on the existence of a new perspective on
business and marketing analytics which big data analysis provides and confirmed the emergence
of new sophisticated technological tools which scale up existing data management practices to a
much higher level of development.
3.2.7 Application areas of big data marketing
This study focuses specifically on big data analysis as a marketing tool, and importance
of the big-data-driven marketing instruments is supported by the findings from the theoretical
and practical literature review. According to statistics, 48% of big data analysis applications
belong to customer-related problems which includes churn reduction, product improvement
76
initiatives, increase of customer acquisition and revenue per customer and some more things, and
10% of them are used for new product and service innovation solutions which contain datadriven development of new products and service offerings (Datameer, 2014).
As for the specifics of the Russian market, marketing problems are also one of the top
issues which are addressed by Russian business leaders by the means of the new technology. Big
data analysis is generally used for advanced customer analytics, segmentation, targeted
advertising and performance evaluation. The prevailing share of biggest big data projects among
Russian companies (90%) were initiated to solve marketing-related problems among others
(Cnews Analytics, 2014).
Analysis of multiple cases of Russian companies has proved the findings from the theory
and provided a detailed illustration of application areas of big data analysis as a marketing
instrument.
Despite the fact that the analyzed companies belong to different economic sectors
(telecommunications, bricks-and-mortar food retail and online retail), all of them emphasized the
paramount importance of data-driven customer analytics with segmentation program as the
kernel of it. A thorough segmentation forms a basis for further big data marketing strategy and
provides an organization with a plenty of valuable insights about its customers.
As long as a company develops a personalized 360-degree view of a customer, gets a full
and comprehensive understanding of its customers across the variety of characteristics, it can
move on with further development of such data-driven marketing actions, as prioritization of the
most profitable segments, targeted pricing and CRM initiatives, targeted advertising campaigns
and many more marketing tactics.
Besides, all investigated Russian companies use technology for customer relationships
management initiatives as a way to, for example, increase customer loyalty and reduce churn
rates.
Russian companies are trying to stay competitive in today’s market of very price sensitive
consumers with decreased purchasing power and that is why they mostly resort to customer
analytics applications of big data as a marketing instrument. The goal is to increase
personalization of marketing actions and provide every single customer with exactly what he
needs.
Moreover, big data marketing executed by Russian companies is not limited by customer
analytics, but also includes opportunities to evaluate performance of offline and online
marketing actions on a real-time basis and receive detailed information and valuable insights
across any segments, individuals or marketing actions.
77
Russian companies explore new areas in big data marketing and use equally mature
analytical applications of big data marketing such as optimization of marketing campaigns,
customer loyalty management, in-store custom analytics and emerging analytical applications
has been currently implemented (ad targeting optimization, customer churn prevention).
Finally, empirical research has demonstrated that it is important that big data analysis
solves marketing problems not only directly, but also indirectly as a consequential effect of datadriven technical (infrastructural) applications of technology. Sometimes Russian companies
resort to big data analysis in order to optimize the technical side of the process and as a result
become more customer-oriented and achieve positive commercial results.
3.2.8 Effectiveness of big-data-driven marketing
One of the most interesting perspectives of this study was to analyze whether big data
marketing can be a truly effective tool and bring profits. One of the main research gaps identified
during the literature review was that opportunities and potential gains of big data analysis for
business are well illustrated in today’s theoretical literature as well as in publications prepared by
practitioners, yet they mostly represent positive attitude towards technology implementation and
are rarely based on full-scale real-life case studies. Empirical part of the research provided a
broader perspective of the problem.
The MTS Group was quite sceptical about the current gains of big data analysis as a
marketing tool for the organization, although it confirmed a few successful initiatives. The
reasons for this viewpoint were the following. On the one hand, MTS has already been doing
similar marketing analytics as a part of database management and as a result the company does
not see usage of new algorithms today as a major technological shift. On the other hand, big data
marketing at MTS Group as a combination of newly hired competent employees and application
of new sophisticated mathematical models simply does not yet bring substantial financial returns.
However, the other three out of four investigated companies confirmed effectiveness of
big data technology for marketing. Another Russian telecommunications giant has achieved
positive commercial results with the help of data-driven targeted advertising and personalized
offerings campaigns. Besides, this company put a special emphasis on the effectiveness of big
data analysis in investment portfolio optimization where advanced analytics enabled the company
to optimize significantly the launch of new basic stations.
As for the analyzed retail market players, Ulmart sees the real benefits of big data
analysis in personalized recommendation initiatives while Lenta confirms commercial success of
segmentation programs and data-driven CRM practices.
78
As a result, it becomes clear that although not every big data initiative brings substantial
financial returns, data-driven segmentation together with development of a 360-degree view of a
customer form a basis for successful big data marketing. After the most profitable segments are
identified and detailed customer profiles are built, marketers can develop targeted marketing
campaigns and implement personalized recommendations initiatives to give the customers
exactly what they need.
3.2.9 Major challenges in big data marketing execution
The review of the relevant theory followed by the empirical research has demonstrated a
very high similarity between the challenges illustrated by researchers and those mentioned by the
analyzed Russian companies. Besides, it is important to mention that a big number of identified
barriers attributable to the global market of big data are relevant also for the Russian market. Let
us illustrate all the barriers of implementation of big data analysis as a marketing instrument.
First of all, all representatives of the Russian companies which participated in the
interviews consider the insufficient level of market development as a major obstacle which leads
to all other barriers. Since the phenomenon of big data analysis and the variety of its marketing
applications has gained attention of the Russian business community only several years ago, the
specifics of technology execution remain unexplored. Currently there are no best practices of
using big data as a marketing tool, neither by foreign companies, nor by Russian organizations.
As a result, Russian companies have to deal with the shortage of competent human
resources which is regarded as the second most significant barrier in technology usage. Russian
companies, as well as foreign organizations, are experiencing the shortage of not only the
technical and IT professionals of big data analysis, but also lack of advanced analytical
competences and necessary educational background of current Marketing managers.
However, this problem is very likely to be resolved in the nearest future, since several
well-respected Russian universities and schools of Russian IT companies (Higher School of
Economics in Moscow, Yandex School of Data Analysis or Saint Petersburg State University)
have recently developed programs dedicated specifically to preparing future experts in data
science, machine learning and IT in business.
Thirdly, Russian companies are often facing the lack of developed metrics and
instruments to derive valuable insights out of big data. However, organizations which already
apply extensively database management or business intelligence practices are less dependent on
this factor than other companies who have to start working with data from scratch.
3.2.10 Future interest in big data marketing and the role of other innovations
79
All Russian organizations, which have participated in the empirical part of the study, have
confirmed the future interest in big data analysis as an attractive technology for solving various
marketing problems. Special attention is paid towards further development of personalized
recommendation programs as a part of CRM practices which is generally considered by Russian
companies as the most effective data-driven marketing tool.
There is no doubt that these companies are considering investments also in other
innovative solutions for their businesses (e.g. Internet of things, video streaming, etc.), but the
empirical research has demonstrated that the importance of big data marketing is planned to
remain at the high level as it is at the moment.
3.3 Managerial implications of the study
Due to the innovativeness of the topic and early stage of big data market development the
process of implementation of big data analysis for marketing purposes may become a very
challenging initiative for an organization. That is why companies are recommended to pay
special attention to the following aspects. An overview of them is illustrated in the figure below.
Fig.6 Overview of factors which impact execution of big data analytics for marketing
purposes
The figure demonstrates that all factors which impact big data marketing practices of
Russian companies can be clustered into two large groups: external and internal environment.
Let us point out the most important aspect from each group.
External factors
80
In order to succeed in using big data analytics as a marketing tool it is crucial to
understand the impact of external environment on implementation of technology by an
organization. As it has been many times mentioned in this study, at the moment the major
obstacle of successful implementation of big data marketing is insufficient level of market
development which affects the overall process of technology adoption and usage. It becomes
much easier to overcome this challenge as long as Russian companies start building competences
and applying big data analytics.
With regards to big data vendors as major suppliers of technology, Russian businesses
will not face a shortage of them: Russian market of big data is comprised of a large number of
both foreign (IBM, Oracle, SAS, Microsoft, SAP, Pivotal, Cloudera, Qlik) and Russian market
players (Yandex Data Factory, Mail.ru Group), different in terms of the size of a company, core
competences, price segments and services provided.
Russian companies are also recommended to be concerned with the overall instability in
the political and economic situation and pay attention to the external environment changes. For
example, it is useful to take into consideration possible changes in legislation such as limitations
on purchasing software of foreign IT suppliers by Russian companies which is aimed at
development of a secure and independent national IT system. As a result, Russian companies
should anticipate future consequences of these external changes to minimize the risks connected
with big data analysis execution.
Internal factors
From all internal factors above all success in implementation of big data analysis for
marketing purposes depends on the importance of data management at an organization and
whether a company already applies extensively methods of business analytics and business
intelligence in their business operations and pays special attention to database and customer
relations management inside Marketing division.
Organizational competences of processing and analyzing data, which include among
others existing expertise of employees and an effective system of managerial control, have a
great positive impact on potential effectiveness of technology adoption and minimize the costs.
Russian companies are often facing the lack of developed performance evaluation
metrics and instruments to derive valuable insights out of big data. However, organizations
which already apply extensively database management or business intelligence practices are less
dependent on this factor than other companies who have to start working with data from scratch.
That is why, for example, companies from the e-commerce sector or offline businesses
where customer analytics play a significant role in marketing decision-making process adopt the
81
technology for marketing purposes much quicker than other companies taking advantage of
existing competences.
The shortage of competent human resources to work with big data analytics in the labor
market can be also resolved if a company has established competences in the field of business
analytics and data management. Education of data scientists and architects, as well training of
data-savvy Marketing managers, takes a lot of time and does not bring quick results.
Therefore, it is recommended to consider in advance possible opportunities and envision
potential obstacles connected with hiring competent data scientists and educating existing
managers who will have to work with large data sets on a daily basis before investing in a
promising technology. A good suggestion would be also to build partnerships with Russian
universities such as Higher School of Economics in Moscow, Yandex School of Data Analysis or
Saint Petersburg State University which have recently developed programs dedicated specifically
to preparing future experts in data science, machine learning and IT in business. Companies
should develop employer branding programs targeted at the students of these programs and
attract young talents to work for their businesses later on.
Besides, big data marketing requires financial resources to adopt and implement
technology, which is another aspect which Russian companies should pay attention to, however
organizations have another less resources-consuming alternative at their disposal. Collaboration
with external vendors will resolve the problem for any business and will require the client
company just to hire several data science professionals and train Marketing managers to work
with advanced, dynamically changing analytics provided by the suppliers.
To sum up, big data marketing is more likely to bring commercial success to the company
if it develops a long-term comprehensive data management strategy for the whole organization.
It will provide a basis for a secure regular flow of data across organizational units, develop a
system of data generation, processing, analysis and storage. When big data analysis gains
attention of not only Marketing department, where it is most frequently applied, but also the
whole organization, its effectiveness is much more likely to increase and bring bigger benefits.
Effective application of different big data marketing tools
Analysis of the variety of marketing applications of big data analytics by foreign and
Russian companies has demonstrated that it is the most rational and effective decision to base big
data marketing strategy on behavioral analytics which provide Marketers of an organization with
a variety of useful instruments to optimize almost every process across the marketing strategy.
Segmentation and development of personalized 360-degree view of a customer should
form the kernel of behavioral data-driven analytics part of the marketing strategy. As long as a
company gets a full and comprehensive understanding of its customers, it can move on with
82
further development of such data-driven marketing actions, as prioritization of the most
profitable segments, targeted CRM initiatives and targeted advertising campaigns and many
more marketing tactics.
Besides, data-driven marketing strategy of a company should definitely contain a special
emphasis on personalized recommendations programs as a part of data-driven customer
relationships management, which have proved to be the most effective instrument of big data
marketing currently applied by Russian companies.
Moreover, Russian companies should keep in mind that big data marketing is not limited
by behavioral analytics, but also includes opportunities to evaluate performance of offline and
online marketing actions on a real-time basis and receive detailed information and valuable
insights across any segments, individuals or marketing actions. One of the biggest advantages of
big data marketing is the freedom and flexibility which marketers have in terms of performance
evaluation, since they can test and analyze almost all information at their disposal, including
unstructured data.
In addition, it is recommended to consider technical and infrastructural applications of
big data analysis and their indirect positive influence on such indicators as customer satisfaction
and loyalty which are commonly measured by marketers.
3.4 Limitations of the study and discussion of further research
The main limitation of this research is based on the innovativeness of big data analysis as
a marketing tool. Currently there are no best practices in the global as well as Russian market,
the phenomenon remains underinvestigated by researchers and business practitioners.
One of the limitations is connected with the insufficient time of using this technology by
Russian companies as a result of the innovativeness of big data marketing and early stage of
market development. In order to increase the reliability of empirical findings and provide a
deeper analysis of the problem ideally Russian companies should already have a minimum 8-10
years record of applying big data techniques in marketing while in reality technology was
adopted approximately 2-3 years ago.
It is also noteworthy that the empirical part of the study takes into consideration Russian
companies from only two industries, telecommunications, online and offline retail, yet they in
fact belong to the currently small number of Russian economic sectors where big data marketing
is in fact applied.
83
For the further research it is recommended to broaden the perspective of empirical
analysis, increase the variety of industries for investigation and include case studies of Russian
companies from banking industry and government institutions.
Besides, another interesting perspective of the future research could be to conduct
empirical analysis of practices of Russian big data vendors - such companies as Yandex Data
Factory or Mail.ru Group. These suppliers mostly specialize in selling ready-to-use real-time
analytics to the third parties rather than applying these tools in their everyday business practices.
This investigation would demonstrate a more technology-oriented angle of the problem
and illustrate the specifics of the first-hand development big-data-based analytical platforms.
However, the market of big data vendors differs significantly from the market of big data users,
so this suggestion would move the study into a completely new research area.
It is also important to pay attention once again to the fact that this study is focused only
on analysis of competences of Russian companies applying big data analysis as a marketing tool.
Therefore, foreign business entities which might as well apply big data analysis in marketing for
doing business in Russia have been from the very beginning considered out of scope of the study.
This research is primarily concerned with exploration of current practices of solely Russian
organizations which we believe have a great potential in technology adoption and is aimed at
demonstrating specifics of implementation of big data analysis only by national market players.
84
Conclusion
Today in order to succeed in a highly competitive business environment companies
throughout the world consider innovation one of the major growth drivers. Big data analysis is
considered to be one of today’s top business innovations. It has recently gained extremely high
interest by the business community all over the world. Companies are attracted by the variety of
managerial implications of big data analysis across all business functions and industries and
promising gains of this technology. Variety of marketing applications of technology which
resolve customer-related problems are particularly highly demanded by companies
Being a part of the global business community, a number of Russian companies also have
started to use big data analysis for solving marketing-related problems.
The topic of this master thesis is “Big data analytics as a marketing tool: the best
practices of Russian companies”. The focus of this research study on marketing is justified by
the current market trends and real-life evidence of companies’ interest in marketing-related
applications of the technology.
This research study is focused on the analysis of practices of Russian companies and
peculiarities of the Russian context, since we believe that they also have great potential to benefit
from these opportunities, yet specifics of the local market should be thoroughly analyzed and
taken into account.
Due to the innovativeness of the topic, the specifics of using big data for marketing
purposes in real-life business environment have not been clearly defined and examined by
researchers as well as by business practitioners. Neither in foreign publications, nor in Russian
ones there is a significant number of thorough and comprehensive research studies conducted on
obstacles and barriers of execution big data analysis as a marketing tool which would be based
on real-life cases.
The research goal of this study is to determine the factors which impact current practices
of using big data analysis as a marketing tool by Russian companies and develop
recommendations for them while the research object is peculiarities of usage and implementation
of big data analysis for marketing purposes by Russian companies.
Thanks to the innovativeness of the topic and current insufficient level of investigation of
big data marketing in the Russian context by researchers as well as business practitioners this
study is a subject of exploratory research.
The major research questions form a basis for the empirical part of this study, which
consists of multiple case study analysis:
1. Why Russian companies resort to big data analytics as a marketing tool?
2. How do Russian companies execute big data technology as a marketıng tool?
85
3. How do Russian companies overcome barriers connected with big data analysis as a
marketing instrument?
4. How can Russian companies leverage the expertise of global market leaders in order to
empower big data analytics for marketing purposes in Russian market?
Besides, it is important to mention that insufficient level of implementation and analysis
of big data marketing by Russian as well as foreign companies puts limitations on the variety of
industries which are investigated in this study. As a result, four Russian companies from two
major market sectors, where large amounts of data are being generated, have been selected for
analysis - telecommunications and retail.
The research has demonstrated that success of Russian companies’ practices in big data
marketing depends severely on external environment and overall level of big data market
development in Russia.
While foreign companies have had more time to master big data execution at their
organizations and benefit from the higher stage of market development, Russian companies
remain pioneers in this field. The have just started to build and leverage capabilities, which is a
very time- and resources-consuming process with no quick financial returns.
The majority of obstacles, which have been revealed in this study, such as lack of existing
performance evaluation metrics, competent human resources and global best practices, result
from the insufficient level of market development and technology adoption by business
practitioners all over the world.
However, the research has shown that the relative strength of these factors may vary.
Russian companies have several alternatives at their disposal which enable them is to overcome
these barriers in short-term perspective, one of which, is for example, collaboration with
external suppliers.
The study has also illustrated that today a Russian company can still achieve marketing
objectives and gain commercial success in executing data-driven marketing.
It is crucial to emphasize the impact of such factor as the existence of organizational
capabilities to analyze and derive valuable insights out of this data. Empirical research has
demonstrated that there is a major difference in considering big data analysis among companies
which already have existing practices of conducting business analytics on a regular basis and
among those who have never paid a lot of attention towards analysis of data generated by the
company.
According to the study’s findings, none of the Russian organizations have yet decided to
build a comprehensive data strategy at their organization to develop data management across all
86
internal organizational units. In most of the cases Russian companies have just recently started to
use the technology and they are still at the stage of testing effectiveness of big data analytics.
With regards to managerial implications of the study, it is recommended to take into
account the paramount importance of data-driven behavioral analytics w i t h segmentation
program as the kernel of it. A thorough segmentation forms a basis for further big data marketing
strategy and provides an organization with a plenty of valuable insights about its customers. In
addition, the majority of the investigated companies emphasized personalized recommendations
programs as one of the most effective and successful data-driven marketing initiatives.
Moreover, empirical research has demonstrated that it is important that big data analysis
solves marketing problems not only directly, but also indirectly as a consequential effect of datadriven technical (infrastructural) applications of technology. Sometimes Russian companies
resort to big data analysis in order to optimize the technical side of the process and as a result
become more customer-oriented and achieve positive commercial results.
There is no doubt that these companies are considering investments also in other
innovative solutions for their businesses (e.g. Internet of things, video streaming, etc.), but the
empirical research has demonstrated that the importance of big data marketing is planned to
remain at the high level as it is at the moment.
87
References
Analytical review of the big data market. (2015). IPO Board, Moscow Stock Exchange.
From: http://www.ipoboard.ru/ru/news/0b4247b31eae6a3007c080e0a6fbf743
Alderson,W. (1957). Marketing Behavior and Executive Action. Homewood, IL: Richard
D. Irwin, Inc.
ATI-Media.Ulmart will demonstrate a new project in Cannes at MIPIM. (2016).
From:http://ati.su/Media/News.aspx?ID=84424&HeadingID=13
Arthur, L. (2013). Big data marketing: engage your customers more effectively and drive
value. John Wiley & Sons, 208p.
Bacon, J. (2014). What's your chief marketing officer really thinking?. Marketing Week
(Online Edition), 15.
Beddow, T.F. (1995). Panel discussion on developing future market leaders. Marketing
Science Institute Board of Trustees Meeting, Dana Point, CA.
Bendle, N.T., Wang, X.S. (2016). Uncovering the message from the mess of big data.
Business Horizons, Vol. 59, Issue 1, 115-124.
Big Data: A Competitive Weapon for the Enterprise. (2014). Datameer. From:
http://www.datameer.com.
Big Data Analytics. Intel’s IT Manager Survey on How Organizations Are Using Big
Data. (2012). Intel report.
Big Data: The Next Frontier for Innovation, Competition, and Productivity. (2011)
McKinsey Global Institute- 156 pages.
Big data survey Europe. Usage, technology and budgets in European best-practice
companies. (2013). BARC Institute: Wuerzburg, Germany- 47 pages.
Big data vendor revenue and market forecast 2013-2017. (2014). Wikibon. From:
http://wikibon.org/wiki/v/Big_Data_Vendor_Revenue_and_Market_Forecast_2013-2017
Big success with big data. Executive Summary. (2014). Accenture, 12p.
From:http://www.acccenture.com/analytics
Camp, R. (1989). Benchmarking: The search for industry best practices that lead to
superior performance. ASQ Quality Press.
Columbus, L. (2015). 56% of enterprises will increase their investment in big data over
the next three years. From http://www.forbes.com.
Columbus, L. (2014). Roundup of Analytics, Big Data & Business Intelligence Forecasts
and Market Estimates, 2014. From http://www.forbes.com.
88
Cnews. Overview of business analytics and big data in Russia in 2014. (2014). From:
http://www.cnews.ru/reviews/bi_bigdata_2014/articles/perspektivy_biznesanalitiki_v_rossii
Cnews. The largest big data projects in Russia. (2014). From:
http://www.cnews.ru/tables/a9249186ccefd9e546774ec36da1970ba20ca212/
C-level executives seeing big results from big data.(2014). From:
http://www.cio.com/article/2607333/big-data/c-level-executives-seeing-big-results-from-bigdata.html
Davis, J. (September, 2014). Digital tools, physical results. Marketing insights, 36-41.
Dietrich, B.L., Plachy, E., Norton. F.C.(2014). Analytics Across the Enterprise: How IBM
Realizes Business Value from Big Data and Analytics. IBM Press, 194p.
Dijcks, J. (2012). Big Data for the Enterprise, Oracle report.
Dougherty, C., Holmes, E. (2010). Consumer spending perks up economy. Wall Street
Journal, 1p.
Dubois A., Gadde L. (2002). Systematic Combining: An Abductive Approach to Case
Research, Journal of Business Research, 55, pp. 553-560.
Dumbill, E. (2013). Making sense f big data (editorial). Big Data, 1 (1), 1-2p.
Elmalem, Y., Hughes, H. (2015). Opening the Core Data of Banking Systems to
Community Bankers. Virginia Banking Journal, 16-19.
Feinleib, D. (2014). Big data bootcamp: what managers need to know to profit from the
big data revolution. Apress, 236 p.
Frampton, M. (2015). Big Data Made Easy: A Working Guide to the Complete Hadoop
Toolset, Apress, 375p.
From token loyalty to meaningful relationships. How loyalty programs and big data
analytics are facilitating CRM in the retail sector. (2014). Marketline Case Study, 21p.
Garry, M. (2005). Doing more with the data. Supermarket News, 91-98.
Gerring, G. (2004). What is a case study and what is it good for? American Political
Science Review, 98, 341-354.
Going to Market in Developing Economies: Winning Big by Targeting Small. (2014).
From: http://www.bcgperspectives.com.
Kash, R., Calhoun, D. (2010). How Companies Win: Profiting from Demand-Driven
Business Models No Matter What Business You're In. HarperCollins, 272
Kotler, P. (1967). Marketing Management: Analysis Planning and Control. Englewood
Cliffs, NJ: Prentice-Hall, Inc.
Kotler, P. (1997). Marketing Management, Prentice Hall, NJ, 789p.
89
Kotler, P., Armstrong, G., Harri, L.C., Piercy, N.Principles of Marketing. (2013). Pearson
Education Limited, 683p.
Krajicek, D. (2013). Making big data actionable. Marketing Insights, 8-9.
Laney, D. (2001). META Gr. Res. Note 6.
Lauterborn, B. (1990). New Marketing Litany: Four Ps Passé: C-Words Take Over.
Advertising Age, vol.61, no.4, pp. 26.
Liebowitz, J. (2013). Big data and business analytics. Auerbach Publications, 304 p.
Lostakova, H. (2009). Benefits of CRM differentiated on the basis of customer lifetime
value. Economics & Business, 1-7.
Linden, K. (2007). Tech tools help marketers spread the word. Shopping Centers Today,
104-104.
Making Big Data work in Retailing. (2013). From: https://www.bcgperspectives.com/
McCarthy, E. Jerome. (1960). Basic Marketing: A Managerial Approach, Homewood,IL:
Richard D. Irwin, Inc.
Mckechnie, S. (2006). Integrating intelligent systems into marketing to support market
segmentation decisions. Intelligent Systems in Accounting, Finance & Management, 117-127.
Minelli, M., Chambers, M., Dhiraj, A. (2013). Big Data Technology, in Big Data, Big
Analytics: Emerging Business Intelligence and Analytic Trends for Today's Businesses . John
Wiley & Sons, Inc.- 224p.
MTS. From: http://www.mtsgsm.com
Nayler, K. (2014). Trends + issues. Marketing Magazine, 10-12.
Networkworld. Comparing the top Hadoop distributions. (2014). From:
http://www.networkworld.com/article/2369327/software/comparing-the-top-hadoopdistributions.html
Overby, S. The CIO and CMO Perspective on Big data. (2014). From:
http://www.cio.com/article/2461403/cio-role/the-cio-and-cmo-perspective-on-big-data.html
Saren, M. (2006). Marketing Graffiti: The View from the Street. Routledge.
Sauer, C. (2015). Collaboration and accountability are at the core of initiatives using big
data. Credit Union Magazine, 34-37.
Schramm, W. (1971). Notes on case studies of institutional media projects. Working
paper for the Academy for educational development, Washington, DC.
Schroeck, M., Shockley, R., Smart, J.,Romero-Morales, D.,Tufano,P. (2012). Analytics:
The Real-World Use of Big Data,IBM report.
Sheth,J.N., Sisodia, R.S. (2012). The 4A’s of Marketing. Creating Value for Customers,
Companies and Society. Routledge, 209 p.
90
Sheth-Voss, P., Carreras, I.E. (2010). How informative is your segmentation? Marketing
Research, 9-13.
Stewart, M. (2015). Big data’s big mistake. Smart business Northern California, 7-7.
The big bang: How the big data explosion is changing the world. (2013). Microsoft News
Center. From: http://news.microsoft.com/2013/02/11/the-big-bang-how-the-big-data-explosionis-changing-the-world/
The big data gap. (2013). From: http://www.bcgperspectives.com.
Top 6 Hadoop vendors providing big data solutions in open data platform. (2015).
Dezyre. From: https://www.dezyre.com/article/-top-6-hadoop-vendors-providing-big-datasolutions-in-open-data-platform/93
Piccoli, G. (2005). IT-dependent strategic initiatives and sustained competitive advantage:
a review and synthesis of the literature. MIS Quarterly, 747-776.
Ragin C. (1997). Turning the Tables: How Case-Oriented Research Challenges VariableOriented Research, Comparative Social Research, 16, 27-42.
Rogers, B. (2014). Are You A Leader or A Laggard In Data-Driven Marketing?.
Forbes.Com, 2p.
Rosbusinessconsulting. The company will help Russian Railways to optimize the traffic.
(2015).
From:
http://www.rbc.ru/technology_and_media/01/07/2015/5593fd8a9a7947c393a79438
Rosbusinessconsulting. Ulmart will follow the Alibaba’s way. (2015).
From:http://www.rbc.ru/business/24/06/2015/558976a89a7947333e5f2f5a
Rule, P., John, V. M. (2015). A Necessary Dialogue: Theory in Case Study Research.
International Journal Of Qualitative Methods, 14(4), 1-11.
Vedomosti. Mobile operators know everything about their customers. (2015). From:
https://www.vedomosti.ru/technology/articles/2015/05/26/593579-sotovie-operatori-znayut-vseo-svoih-abonentah
Ulmart. For investors. From:http://investors.ulmart.ru/
Weber, L., Henderson, L. L. (2014). The digital marketer: ten new skills you must learn
to stay relevant and customer-centric. John Wiley & Sons, 368p.
Webster, F.E., Jr. (1997). The future role of marketing in the organization. In Reflections
on the Futures of Marketing. Cambridge, MA: Marketing Science Institute.
Weitz, B., Wensley, R. Handbook of Marketing.(2002). Sage Publications Ltd, 583 p.
Yandex data factory. World of Tanks achieves a new level of churn prevention through
the implementation of YDF’s data analysis. From: https://yandexdatafactory.com/case91
studies/world-of-tanks-achieves-a-new-level-of-churn-prevention-through-the-implementationof-ydfs-data-analysis/
Yin, Robert K. (1981a). The case study as a serious research strategy. Knowledge:
creation. Diffusion, utilization, 3, 97-114.
Yin, Robert K. (1981b). The case study crisis: Some answers. Administrative Science
Quarterly, 26, 58-65.
Yin, Robert K. (2014). Case Study Research: Design and Methods - fifth edition. Library
of Congress Cataloging-in-Publication Data, 282p.
92
Отзывы:
Авторизуйтесь, чтобы оставить отзыв