Sultan Moulay Slimane University

Polyvalent Faculty
Béni-Mellal

Department of Economics and Management

THESIS FOR OBTAINING THE DEGREE IN ECONOMIC SCIENCES AND MANAGEMENT

End of Studies Project Report on the theme:

BIG DATA
MADE BY: FRAMED BY:

Marouan KACHACH -Pr. Rachid HASNAOUI

Academic year: 2016/2017

 SUMMARY
SUMMARY.................................................................................................................... 2

Acknowledgments

Issue .................................................................................................................. 5

Introduction..................................................................................................................... 6

CHAPTER 1: Conceptual Framework of Big Data............................................................ 8

1. Definition of Big Data ....................................................................................... 8

2. The characteristics of Big Data ..................................................................... 9

2.1. The volume of Big Data .............................................................................. 9

2.2. The variety of Big Data......10

2.3. The speed of Big Data

3. The sources of Big Data................................................................................... 11

3.1. The categories of Big Data sources .................................................... 11

3.1.1. Directed sources.............................................................................. 12

3.1.2. Automated sources..................................................................... 12

3.1.3. Voluntary sources

3.2. L ’Open Data.............................................................................................. 13

4. Big Data in numbers.......................................................................................... 15

CHAPITRE II : LES APPORTS DU BIG DATA AU MARKETING.................... 17

1. Big Data applied to marketing service................................................... 17

1.1. The relationship between Big Data and Marketing......................................... 17

1.2. The role of Big Data in Marketing decision making ........................... 1

2. Case Study "Amazon"................................................................................ 19

3. The stakes of Big Data.................................................................................... 22

3.1. Ensure the quality of the information......................................................... 22

 3.2. Optimize data processing....................................................... 23

3.3. Connect all trades............................................................................... 23

3.4. Ensure security

3.5. Humanize the data ............................................................................. 24

Conclusion ..................................................................................................................... 25

List of figures............................................................................................................. 26

List of abbreviations 27

Bibliography... 28
Acknowledgments

This work has been the labor of several years of study, and would probably not have

never been completed without the support of a large number of people that I hold

warmly and very sincerely to thank.

I will start by thanking my supervisor and professor Mr. Rachid.

HASNAOUI for his valuable help, for the quality of his advice, for the time

that he granted me to encourage and support me.

I also thank all our teachers and all the staff of the Faculty.

Multidisciplinary from Beni Mellal who provided us with quality training.

I would also like to thank the jury members who honored us by accepting

to evaluate this work.

Please, gentlemen and ladies, accept my deep respect.

Marouan KACHACH
Problematic

Before the 2000s, the so-called traditional methods of storage, processing and
The data analysis was sufficient to provide us with actionable information.
However, with technological development and the increase in storage costs and
data processing as well as the emergence of social networks and laws that require the
data retention for a specified period, these old methods are no longer
sufficient to guarantee us actionable information. The emergence of the notion of 'Big'
Data has enabled the analysis of very large volumes of data from different sources.
to provide information ready to be used by applicants at a reduced cost
and a reasonable deadline.

Currently, those seeking information, whether they are companies, states,

people ... etc., have become too demanding because of the unstable environment in which
they operate. It is in this context that our issue is situated:

What are the contributions of Big Data to marketing?

What is Big Data? What are its characteristics? What are its...
sources? What are its stakes?

Is there a relationship between Big Data and marketing? What is the role of Big Data?
Data in marketing decision making?
Introduction
Since its creation until the present day, the Internet has significantly seen its
number of its users worldwide. Moreover, the emergence of Smartphones and
other types of connected objects, combined with the advent of social media, has given
birth to an unprecedented shift in the level of stored or generated data
the global scale. Data of an incomprehensible volume, of variable nature and sources and
obeying an imperative for rapid collection and processing.

The figures are impressive and reflect how much our society
contemporary is invaded by data. It is sometimes said that this data
e
are the 'black gold' of the 21st century and that they entrust power to those who hold it
control (Research, 2013)1 This data covers all aspects of our life
daily: these are financial, environmental, medical information,
behavioral demographics, psychographics about consumers, of
surveillance... etc. They come, among other things, from sensors, mobile devices,
technical devices, social networks, and blogs (PwC2).

With this explosion of data, many so-called historical companies have had to
to adapt in order to face the e-transformation of their sector of activity and to this
new channel of distribution, consumption, and sharing of information.
traditional and mass consumption that companies were accustomed to has
disappeared and made way for a more thoughtful consumption, more comparative and where everyone
Internet users can give their opinions and exchange.

The customer is therefore more than ever at the heart of the business and marketing strategy.
companies. It becomes important to clearly identify it and to know it in order to be able to
to track between the different points of contact he has with a brand in order to
propose the right offers at the right time, whether through a physical store, a
Smartphone, a tablet, a computer, his television, and soon his connected glasses, his
wear a watch or a connected bracelet.

We are talking about one-to-one marketing. An offer can be tailor-made for a single person.
client based on his browsing history, the frequency of his visits to a
boutique, the time he spent looking at this or that product... etc. This personalized marketing
aims to make the consumer feel that they are unique and precious in the eyes
of the company and to thus catalyze the act of purchase.

This research work is composed of two parts. Firstly, we

we will try to understand what Big Data is, the main characteristics of Big Data and

1
[Link]
2
PricewaterhouseCoopers: a network of international firms specializing in audit missions,
accounting and consulting expertise.
the different sources of Big Data. In a second step, we will study the contributions of
Big Data in marketing decision-making and the challenges of Big Data.
CHAPTER 1: CONCEPTUAL FRAMEWORK OF
BIG DATA

1. Definition of Big Data

The quantitative explosion of digital data has forced researchers to find
new ways of seeing and analyzing the world. It is about discovering new orders
of greatness concerning the capture, search, sharing, storage, analysis and the
data presentation. Thus was born "Big Data". It is a concept that allows for
to store an unspeakable amount of information in a digital database. According to the archives of the
digital library of the Association for Computing Machinery (or ACM) in
scientific articles concerning the technological challenges to be addressed for visualizing the 'greats'
data sets", this term appeared in October 19973.

Literally, these terms mean mega-data, big data or data.

massive. They refer to a very large set of data that no classic tool can
database management or information management really cannot work. In
In fact, we generate about 2.5 trillion bytes of data every day. This is the
information coming from everywhere: messages we send to each other, videos we
publications, climate information, GPS signals, transactional purchase records4
online and many others. These data are called Big Data.

To summarize them in one sentence, big data is a set of alternative solutions.

to traditional database and analysis solutions aimed at handling a volume
very important data, in real time and with a very wide variety of sources and
formats.

3
[Link]
4
[Link]
 2. The characteristics of Big Data
To characterize Big Data, we therefore present here the three basic "V's" or the
three

Figure 1: The 3Vs of Big Data

Source5

2.1. The volume of Big Data

It is primarily the immense quantity of data that already exists and is still
generated every second that characterizes Big Data. While we were talking, a few
years, gigabytes (109octets), today we are rather talking about exabytes (1018octets) and
of the zetta-octets (10 21(octets), 1.8 zetta-octets representing all the recorded information
in 2011. Given that the global data volume is expected to reach 35 zettabytes
octets in 2020, it would be necessary to multiply the number of physical servers by
ten from here ten years.

5
[Link]
Scratch/323940396?via=search-layout-grid
 2.2. The variety of Big Data

In order to speak of Big Data, the data must be varied.

data is varied as long as it comes from multiple sources,
such as sensors, smart objects, the Internet, and technologies
social collaboration. The data no longer includes only relational data
traditional, but also raw data, semi-structured or unstructured coming from
web pages, online files, search indexes, social media forums,
emails, tweets, Facebook statuses, blogs, videos and sensors among others
(PwC). Since the data comes from different sources, it also has different formats.
(text, image, audio, video... etc.).

The TechAmerica Foundation's Federal6BigData Commission claims that 85%

data is unstructured, 90% according to Atos (2014)7Organizations therefore face
a real challenge of data sorting in order to extract useful information that can
producing value. Desouza (2014)8estimates that this characteristic of Big Data constitutes
the most important challenge for companies. Indeed, they have developed systems of
data management aimed at manipulating information of specific categories.
For example, software from the human resources department is designed to manage
data concerning the staff, while a CRM (Customer Relationship Management) software
Client holds data on customers.

Today, organizations can no longer be satisfied with just managing

one or the other category of data but are constrained to work with
heterogeneous databases in which the data differ as much by their
origin that is inherent to their nature. For example, the challenge of a company will be to have to
mix social media data with customer data from a program
GRC. Then, she will need to analyze the mixture and draw conclusions from it.
will enable personalized marketing.

2.3. The speed of Big Data

By speed, we mean both the speed at which data is produced and
transformed and the speed at which this data must be collected, understood
and treated by the organizations (TechAmerica Foundation’s Federal Big Data
Commission). One can therefore speak of Big Data from the moment when the data is
created in real time or almost real time (Kitchin, 2014)9Regarding the speed of
data processing, the increase in the number of RFID sensors10(Radio Frequency

6
[Link]
7
[Link]
8
[Link]
.pdf
9
[Link]
10
Techniques for memorizing and retrieving data remotely.
Identification) and other information transfer tools have led to a continuous flow of
data whose pace has become impossible to manage for management systems
traditional data.

Moreover, in a world where a company can get ahead of its competitors by identifying
a trend, a problem or an opportunity just a few seconds before others,
We realize how important data analysis speed has become.

Finally, it should also be noted that more and more data produced today is
available for a very short period of time. It is therefore essential for organizations
to be able to analyze the data almost in real time in order to draw conclusions from it
valuable information for their economic performance.

3. The sources of Big Data

Big Data consists of structured, semi-structured, and unstructured data.
structured. The first come from traditional databases, and
constitute about 10% of Big Data. The other 90% are semi-
structured and unstructured.

Unstructured data comes from:

on one hand, from "human" sources, via the Internet (emails, photos, videos,
tweets, Facebook posts, 'clicks' on websites, queries on search engines
search, purchases and online transactions ... etc.) and via networks of
telecommunications (phone calls and call center conversations);

- on the other hand, 'machines' that generate data automatically, such as the
mobile phones (geolocation data (GPS) and historical usage data),
the planes (the test flights of the Airbus A350 generate 2 TB (terabytes) per day), the sensors
(used for example to collect climate information) and connected objects
in general (like watches or 'smart' clothing, connected using codes
bars or RFID chips.

Among all these sources of Big Data, we can distinguish data that is
generated due to a voluntary human action, and others that are produced
automatically by machines, or resulting from a human action without the aim of providing
data.

3.1. The categories of Big Data sources

The sources of Big Data can generally be divided into three categories,
described below:
 3.1.1. The directed sources

They generate data through traditional surveillance technologies, which

are directed towards a person or a place by a human operator. For example, the
case of systems intended for passport control. They collect and verify
information about travelers that is stored in real-time databases.
from these last ones, new data can be generated such as photos
of identity, fingerprints, or even retinal scans. This is also the case
electromagnetic or thermal environmental sensors that allow for
2D and 3D mobile and real-time cartographies.

3.1.2. Automated sources

The data is generated by devices or systems with functions

automated systems that do not require human intervention. These sources include:

mobile phones, which automatically record and communicate the

historical information about their use;

mobile devices in general, which can transmit geo data

localization ;

the interactions and transactions on digital networks, which not only transfer
information, but also generate data about interactions and
transactions themselves (such as bank transfers or emails);

the click stream data, which records how people navigate through
a website or a mobile application (historical data 'Cookies')11») ;

the data generated by connected objects and sensors;

audio data from call centers or phone recordings in the case of

government intelligence services.

Much of this automated data originates from physical objects. This

are mobile devices and sensors that are able to 'feel' some
characteristics of the environment, such as geo-location, temperature or the
movement. As soon as these objects communicate information to other objects or
to a database, we talk about 'connected objects'. On a global scale, the network
formed by these connected objects is called the 'Internet of Things' (Gartner, 2016b)12.

The Internet of Things is a trend complementary to the Big Data trend and
will also revolutionize the current world. Today, "already appear

11
Information stored on your computer by a website you visit.
12
[Link]
connected watches, connected cars, connected clothing, and perhaps
that one day our human body will also be connected. Thus, the Internet of Things constitutes a
the main sources of Big Data.

3.1.3. Voluntary sources

The data is voluntarily provided by the users. It comes from

mainly interactions through social networks, such as posts
of content and file uploads (images, videos, files, etc.).
Moreover, crowdsourcing13literally, 'crowd contribution' on some
open source platforms, where users contribute to the content themselves
web (like Wikipedia), also constitutes a voluntary source of Big Data (Institute of
Actuaries in Belgium.

This multitude of different sources of information reflects the dimension of Variety.

This generates one of the major forces of Big Data: gathering this data coming from
of different origins and cross them in order to bring out results that one does not
could discover differently.

3.2. Open Data

Another important source of Big Data is Open Data (literally, "the

open data" ). Gartner14defines Open Data as 'information or content
made available for free for use and redistribution, and subject only to
requirement to attribute to the source. The term can also be used more generally
to describe all the data that is shared outside the organization and to
beyond its intended initial use, for example with business partners,
clients or industrial associations.

To give an idea of the context, Open Data, in 2013, was more than 40
countries with government Open Data platforms, over 90,000 databases
on [Link] (the Open Data platform of the United States government), 1.4 million views
from the page on the UK's Open Data site, 102 cities that participated in
International Open Data Hackathon Day15or more than a million databases
available from governments worldwide.
16
According to McKinsey , the fundamentals of 'open data' have the characteristics
following:

13
Participatory production
14
American company specialized in consulting and research in the field of advanced techniques.
15
[Link]
16
consulting firm.
 Accessibility: a very large number of users are allowed to access
to Open Data.

Automatic readability: data can be processed automatically.

The cost: access to Open Data is free or incurs a negligible cost.

Rights: there are very few limits on the use, transformation

or the distribution of data.

These are the reasons why McKinsey refers to Open Data as 'information'
liquid.

Moreover, it is essential not to confuse Big Data and Open Data. Open Data
is a source of Big Data but the entirety of Open Data is not found in Big Data,
as illustrated below. The infographic from the McKinsey Global Institute shows that
Open Data is a large part of Big Data, but the majority of Big
Data comes from other data sources, which are not 'open' to the general public.
public.
 Figure 2: Open Data constitutes a large part of Big Data

Source17

4. Big Data in numbers18

Big Data generates numerous statistics and predictions about itself. Here are the ten
figures to remember.

. 90% of the data created in the world has been created in the last two years.
years (source IBM).

. 204 million emails, 1.8 million Likes, and 278,000 Tweets are generated
every minute (Source: Qmee).

17
McKinsey Global Institute Analysis
18
[Link]
. 40,000 searches are analyzed by Google on its search engine.
Every second, which is more than 3.5 billion per day! (Source: Google Search)
Statistics.

. 100 hours of video are downloaded on YouTube every minute on average.

(Source: YouTube).

. 570 new websites are created every day (Source: Muschroom Networks)

. 51% of the data is structured, 27% is unstructured and 21% is semi-

structured. (Source: Tata Consultancy Services).

. 4.4 million jobs related to Big Data will be created by 2015 (Source:
Gartner).

. + 81% growth between 2012 and 2014 for the data transferred via
mobiles, to reach 1.5 million gigabytes per month. (Source: Cisco
Systems).

. 54.3 billion dollars for the Big Data market in 2017, compared to 10.2
billions of dollars in 2013 (Source: Mushroom Networks).

. 30 billion connected objects in 2010 (Source: Gartner).

CHAPTER II: THE CONTRIBUTIONS OF BIG
DATA IN MARKETING
Big Data technologies are now an essential tool for assistance in
decision for the company, especially the marketing department which represents for it the means of
know and meet the needs of its customers.

The increasing evolution of data generated from various sources such as the Web,
allows for a more targeted and personalized analysis of consumer behavior and
communication strategies.

Big Data applied to marketing service

1.1. The relationship between Big Data and Marketing19

For years, researchers in the field of marketing have been using data
clients at their disposal in order to analyze them, with the objective of increasing revenue
from the company, that is to say sell more with the satisfaction of these customers, for a long time
duration in a loyalty logic. The recent emergence of Big Data has increased
their analytical capacity as well as their possibilities for reaction, even anticipating
behaviors of their customers. In fact, like the Gartner study20, most studies
estimate that between 50% and 60% of Big Data projects are related to customer experience in the sense
large.
Thus, by collecting data on customers through campaigns
communication organized by companies, or by purchasing its products, subscribe to its
services, navigating on its websites or interacting with it, a company is able to
better understand them. This data analysis system allows it to adapt its offer, its
strategy and its sales and customer relationship tools, to better satisfy the needs of its
clients.

Nowadays, they even talk about predictive marketing which consists of forecasting the
consumer behaviors of a given company. This action is possible thanks to
the process of collecting and analyzing a large volume of data, allowing for
forecast of the upcoming action. For example, Amazon is working on developing software
allowing to predict the purchase date of its customers, even before they go to the
website to place an order, with the aim of setting up a delivery system
grouped21.

19
[Link]
customer-relationship/
20
[Link]
21
[Link]
 1.2. The role of Big Data in Marketing decision-making

The explosion of available data and the competitive environment is increasing the trend towards
base marketing decisions on data rather than just simple intuitions. The
measurement of return on investment and optimization practices are multiplying. The
Automation practices are developing to facilitate personalization. According to the
magazine 'IT for Business No. 2192 - February 2015'22the barometer aimed at measuring the
the perception of Big Data and the maturity of companies in France shows a strong evolution
between 2012 and 2014. The maturity index increased by 66% and particularly by 80%
marketing teams surveyed position themselves as the primary stakeholder of a
Big Data initiative.23

Big Data is used to make marketing and business decisions, such as:

. refine and personalize the customer history;

. perfect and personalize the product/service offering;

. anticipate the supply of products/services thanks to predictive Big Data;

. improve marketing campaigns (customize the content and format to

customer behavior, including the dates and times of broadcasting concerning a
publication on social networks for example).

From the above, we conclude that Big Data constitutes today a real
change in the working methods of marketers who previously relied on
traditional techniques such as market research, in order to improve the decision-making process
decision.

Currently, Data Driven Marketing24allows for decision-making and

real-time adaptations.25

To clearly illustrate the role of Big Data in the field of marketing, we will
cite the results of a study conducted in 2015 by 2nd Watch. As part of this study
conducted with more than 500 small and medium-sized American businesses, 86% of the companies
interviewed individuals claim that they use or plan to use Big Data to strengthen their
marketing. Furthermore, 75% of those who are already using Big Data state that these measures
have proven to be "fairly effective" or "extremely effective". In addition, 83% count
extend the use of Big Data to the organization of their advertising campaigns.

22
"Big Data index" barometer EMC and IDC mentioned in IT for business February 2015.
23
[Link]
24
It consists of structuring and analyzing data to shed light on behaviors of
consumers and their expectations.
25
[Link]
 Figure 3: Big Data in Digital Marketing

Source26

According to the same study, there are three main uses of Big Data serving the
digital marketing. 29% of companies use this big data to improve
understand the will of their customers. 18% use it to improve their supply chain
logistics. Finally, 16% see the opportunity to strengthen their advertising campaigns and
their promotional operations.

2. Case study 'Amazon'

It is clear that Amazon has made a typical leap by opening up to the path of commerce.
electronics. But probably one of its greatest innovations was the "System
"Personalized Recommendation" which is based on mass data collected from
millions of transactions made by its customers.

Psychologists talk about the power of suggestion, that is to put something in front of someone.
a product that he might like which is likely to trigger the desire to purchase in him
subject, regardless of whether this purchase will fulfill a real need.

This is of course how impulse buying has always worked. But instead
From a dispersed approach, Amazon has multiplied its customer data and sharpened its system.
in order to make it more precise. This is at least the goal pursued by Amazon even if it
it seems that he has not yet been fully reached.

26
[Link]
 In any case, Amazon's systems are improving more and more and it seems that
what we have seen so far is only the beginning.

Amazon recently obtained a patent for a system designed to ship us

merchandise before we even decided to buy it. This is a strong indicator
that the credibility they give to predictive analysis is increasing.

When talking about Amazon, one must take into account its more commercial nature.
of its Big Data compared to those of other companies that handle data at a scale
comparable. Unlike, for example, Facebook which could know a lot
information about the movie that you or your friends like, the vast majority of
Amazon's data about us concerns how we spend our income.

Having developed a way to use this data to get more money from our
Poches, Amazon is now embarking on another mission, that of helping other companies to
the global scale to do the same thing, by selling them this data in addition to the tools to
the analyze them.

This means that, like with Google, we have started to see ads.
conducted by the Amazon platform and based on its data appearing on others
27
sites in recent years. As noted by MIT Technology Review the year
this now makes the company a direct competitor to Google with both
online giants fighting for a piece of marketers' budgets.

However, advertising sales are not the only area where Amazon competes.
Google, the web services of Amazon (AWS) offer cloud-based computing and a
mass data analysis.

This allows companies that need to carry out highly intensive procedures
to save calculation and analysis time instead of focusing on the installation of their
own data processing and analysis centers like Google's BigQuery.

These services include the data warehousing service (Amazon Redshift), the solution
Hosted Hadoop (ElasticMapReduce) and S3 (the database service it uses)
manage its own physical data warehouse operations and its archiving service
Glacier). Recently, this list has expanded with the emergence of Kinesis. It is a
real-time stream processing service designed to assist in data stream analysis
large volume.

Amazon has also incorporated Big Data analysis into the operations of its service.
client. His purchase from the shoe retailer Zappos is often cited as a key element in
this statement. Indeed, since the founding of Zappos, its customer service had gained a
fabulous reputation and had often maintained itself as a world leader in this regard. This
was mainly due to its sophisticated customer relationship management systems that

27
MIT Technology Review is an American magazine published by Technology Review.
 widely used the data of its own customers. These procedures were merged
with those of Amazon following its acquisition of Zappos in 2009.

Finally, it is worth mentioning that through Amazon Web Services (AWS), Amazon
hosts and allows the analysis of public datasets. Amazon hosts, among others,
data unearthed by the human genome project, datasets from the sciences
from NASA's Earth or even data from the United States Census. This data
are made available to the general public and can be consulted for free.

Originally designed at its creation to be an online bookstore, Amazon has greatly expanded.
thanks, largely to its enthusiastic adoption of Big Data principles. Now, it
It seems established that it will continue to gain ground in the coming years.

Figure 4: Annual Revenue (AR) of Amazon

2013 2014 2015 2016 2017

Income 1erQuarter 15,253,047 18 737 424 21,562,132 27 647 214 33,898,400

Revenue 2thQuarter 14 905 653 18 356 808 22 005 390 28 858 346 -

Revenue 3thQuarter 16 223 091 19 531 872 24 068 870 31 050 912 -

Revenue 4thQuarter 24 286 230 27 837 996 33 929 724 41 517 332 -

Annual Revenue 70,668,021 84 464 100 101 566 116 129,073,804 -

Source: Europafinance by Cofisem28

28
[Link]
 Figure 5: Evolution of Amazon's Annual Revenue

Source: Europafinance by Cofisem

3. The challenges of Big Data

Big Data seems to be essential today; it is the new reality of the economy.
Digital is the new territory for development and value creation for companies.
Opportunities seem endless, so we must take ownership of the data to better utilize it.
understand and tame them in order to prepare for the future towards which they seem to lead us
So what are the main challenges of Big Data for companies today?

3.1. Ensure the quality of information

To embark on the Big Data adventure, one must know how to identify and
clean" the data we want to work with and benefit from it. If the
knowledge of customers is primarily based on analyzing information about them, it is necessary to
that the data is of good quality. Indeed, data quality must be a priority.
to avoid distorting the strategies arising from their processing. An audit system of the
Data quality must therefore be implemented to correct any potential errors.
human incidents occurred. In this context, a data quality audit should be the
first action of any Big Data project. To address this issue, correction tools
automatic systems ensure the relevance of each collected data.

3.2. Optimize data processing

For most experts, data processing is one of the biggest challenges.

important aspects of Big Data. In fact, information comes in bulk and takes the form of
different formats (images, videos, audios, text, etc.). The company must necessarily
invest in management and processing tools to ensure data transformation
collected in decipherable information. These management solutions must also allow
to analyze the data in real time (to improve the quality of decision-making).
The company must therefore integrate data processing experts into its team such as [the
data analyst and the data scientist29.

3.3. Link all professions

Big Data requires the gathering of a set of different professionals for the purpose
to achieve specific objectives. Thus, according to the specificities of the target population, the
The marketing specialist should work closely with a statistician to get an idea.
specifies actions to be taken. Big Data concerns all professions within the company and the
Analyzed data should benefit everyone. Therefore, we must necessarily structure the ...
data following a modeling of business processes.

3.4. Garantir la sécurité

The valuable information for companies to define their strategies is for the
mostly a private matter. They affect the trust relationship that binds the company to its clients
due to the fact that they primarily stem from user accounts. As a result, their security
has particular importance, as it engages the responsibility and reputation of the company.

29
Top officials in the management and analysis of 'big data'.
 3.5. Humanize the data

The company must always keep in mind the human aspect of the data collected. What
be that as it may, the interest of Big Data is to position clients at the center of the approach
decision making. Better informed, customers demand better services and offers
customized to match their real needs.
Conclusion

The use of data is not a new phenomenon, as we have illustrated.

throughout this work, in fact the traditional methods of collection and analysis of
data is no longer suited to the evolving needs of societies. It is in this context
that Big Data technologies present themselves as an alternative to meet this need
croissant in the various departments of companies and specifically the marketing department.

This new technology allows the collection of a considerable volume of data.

from different sources, whether directed sources, automated ones, or still
volunteers, in a very quick manner in order to analyze them and extract the information
exploitable solutions that can meet a client's real needs as well as anticipate them.

However, Big Data has limitations that we have explained in the form of challenges.
who are the security and reliability of information as well as the optimization of the process of
data processing, especially since the collected data comes in different forms (videos,
audios, text, etc.

Finally, as long as consumer needs change continuously,

Big Data technologies must adapt to this evolution to avoid being surpassed like the
traditional methods. Therefore, we must always look for new methods to manage the
data to make it usable.
List of figures

Figure 1: The 3Vs of Big Data................................................................................................. 9

Figure 2: Open Data constitutes a large part of Big Data ................................. 15

Figure 3: Big Data in Digital Marketing

Figure 3: Revenue (CA) of Amazon.............................................................. 21

Figure 4: Evolution of Amazon's Revenue......................................................... 22

List of abbreviations

Abbreviation Meaning

PWC Price Water House Coopers

ACM Association for Computing Machinery

GRC Customer Relationship Management

GPS Global Positioning System

RFID Radio Frequency Identification

NASA National Aeronautics and Space Administration

AWS Amazon Web Services

IBM International Business Machines Corporation

Bibliography
The Cloud & Big Data magazine (2017). Available at: <[Link]
(Consulted on 11/05/2017)

IT Skills In Demand (2017). Master Big Data and Hadoop Step-By-Step from Scratch.
Available at: <[Link]
Step-from-Scratch/323940396?via=search-layout-grid> (Consulted on 01/07/2017)

The media of digital transformation (2016). Ten key figures of Big Data. Available
on : <[Link] on 07/05/2017)

Time2marketing (2016). How to do predictive marketing with Big Data?

Available on: <[Link] (Consulted on 07/07/2017)

Big Data and Data Marketing (2017). Available at: <[Link]

/big-data-key-figures/> (Accessed on 06/07/2017)

Big Data - The New Oracles - Documentary (2015). Available at:

<[Link]
(Consulted on 03/06/2017)

The Cloud & Big Data magazine (2017). Big Data and digital marketing
Available on: <[Link]
a-success/> (Consulted on 11/05/2017)

Big Data applied to marketing service (2016). Big Data and marketing
digital Available on: <[Link]
(Consulted on 07/05/2017)

Big Data: what impact on marketing and customer relations? (2015). Big Data and the
digital marketing. Available at: <[Link]
[Link]/[Link]/our-analyses/big-data-what-impact-on-marketing-and-relationship-
client/> (Consulted on 07/07/2017)