1
Characterization of users of Online Social
Networks for Viral Marketing
Debesh Majumdar
 seeing a phenomenon where users of OSNs are disclosing
Abstract— The advent of a slew of online social networks and
in particular the success of Facebook has led modern marketers
to look at these platforms as a medium of conveying messages
regarding brands to a selectively targeted group of people. Users
of online social networks generate huge amounts of data which
need to be carefully analyzed to create a target group. As
Facebook and other similar social networks become an integral
part of our lives, the issue of what constitutes an appropriate
level of privacy in these networks is gaining importance in both
academics and the industry itself. The aim of this research is to
identify the characteristics of OSNs, the various aspects of OSN
users, the privacy policies of the major OSNs and find out
whether the number of hours spent on Facebook is related to the
number of years a person has been using Facebook. This has
been done through a three pronged strategy. The first part of
the study consists of summarizing the current state of research
on OSNs. The second part consists of documenting the data
gathered through secondary sources to generate insights into
OSN users and their privacy concerns. The third part consists of
collecting primary data with the help of a questionnaire. These
three when viewed together provides us with a clearer view
about who the OSN users, what their tastes and preferences are
and whether online marketing appeals to them.
Index Terms— Facebook, Online Social Networks, privacy
Marketing, User characterization
I. INTRODUCTION
O NLINE social
networks have been dominating the Internet
world for some time now. While the last decade seemed
to be the decade of search engines, this decade is more about
OSNs. Already Facebook has surpassed Google as the most
viewed website in US. As this trend continues, OSNs will
become the favored medium of advertisement of a lot of
brands. In such a scenario, it becomes very important to
understand social networks and the users of these social
networks. Generalizations will lead to incorrect results. A lot
of research is currently going on to understand OSNs. As
with anything on the web, privacy is becoming an
increasingly important issue in OSNs. This issue has gained
more importance in the networked world because we are
Debesh Majumdar is a student at the Vinod Gupta School of Management,
IIT Kharagpur (phone: 9378405101; e-mail: m.debesh@ gmail.com). This work
has been done under the guidance and supervision of Prof. Prithwis Mukherjee,
VGSOM.
very private information in these networks. Moreover, the
entire business model of OSNs depends on using the
information supplied by the users to create specifically
targeted advertisements. This leads to a situation where OSNs
are responsible for ensuring privacy for its users but at the
same time generate revenue and increase profitability through
the users’ data. This research aims to find out the structure of
OSNs and the characteristics of users of OSNs. It also looks
at the privacy policies of some of the major online social
networks like Facebook, Orkut and Youtube and how the
privacy policies of Facebook have been evolving since its
inception. First of all, a survey of the current literature on
OSNs is carried out to determine the status of research on
OSNs and also to gain some insights. Secondly, some
secondary data sources are mined to get insights into user
characteristics of primarily US users and the privacy policies
of online social networks. Finally a survey is conducted to get
some data on Indian Facebook users. Taking all these three
together, we realize that OSNs are not homogenous. People
towards the center of the network tend to exert the maximum
influence on the network and their content and conversations
percolates the network the fastest. Moreover, user profiles can
give indications about the personality of the user. We also
understand through this research that the privacy policies of
the major OSNs are continuously evolving and are becoming
increasingly sophisticated for the common user to understand.
Indian Facebook users have been studied to understand their
demography and their attitude towards online marketing.
What comes out from that is that although a significant
number of people are concerned about privacy, they also feel
that if marketers can properly communicate the benefits
derived from a product/service, OSN users would be willing
to look at the advertisements. Finally we find that Facebook
is addictive. As the number of years spent on Facebook
increases so does the usage per week.
I. LITERATURE REVIEW
A. Characteristics of OSNs
Online Social Networks are different from traditional
websites in that the structure, growth and nature of the OSN
are determined largely by the activity of its users.

An OSN has the following characteristics:
--First, OSNs are user based. The users build the
network over a period of time and determine the general
direction of the network by their activities. As more and more
users join the network and interact with each other, the
network gets enriched with content and conversations. All
users have equal opportunity to determine the conversations
in the network. The open nature of the conversations and the
ability of users to participate in the discussions at any point of
time make the conversations truly unpredictable and dynamic.
--Second, OSNs are interactive. Apart from the regular
interaction through private message boards and public profile
pages, users also interact with each using applications
ranging from games to question/answers and through
different groups. These various avenues of communication
increase the level of communication.
--Third, OSNs are community-driven. Users create
communities according to their interests. Other users having
similar interests join the community. The users of the
community then start discussing with each other and form
new connections based on these interests. Therefore, groups
and communities help in creating a denser network.
--Fourth, relationships form an important part of any
OSN. The distance of a user from the centre of the network is
determined by the number of connections that the user has. A
user having more connections is closer to the center of the
network. Users near the centre of the network become very
influential because the content published by them gets
transmitted very fast across the network.
--Finally, in an OSN, more emphasis is laid on the
emotions of users than on content. Generally, users provide
their own opinion and understanding of the content along
with the content itself. Increasingly, users are finding OSNs
as a medium of self-expression. As larger portion of the
population gets wired, this trend is going to continue.
B. Classification of users
Based on the location of the user in the network, users can
be classified into three groups: singletons, giant components
and middle region.
Singletons are degree-zero nodes who have joined the
network but have not made any connection with other users of
the network. They reside in the outermost part of the network
and do not participate in the network and as such do not have
much influence over the direction that content and
conversations in the network take.
The giant component represents the large inner layer of the
network. Here, users are connected to a large portion of the
network, either directly or indirectly. It consists of the most
active users and a large portion of the content is generated
from the giant component.
The middle region consists of small and closed
communities. They interact with each other but not with the
larger network. They form a significant portion of the user
2
base of OSNs.
Some interesting insights regarding users emerge from this
classification. Research shows that isolated communities are
not likely to merge with each other. Almost all isolated
communities are stars. A star is a single individual who is
very charismatic and are linked to a small group of people
who have very less connections. Stars dictate the general
direction of the community. Stars evolve through two
processes – isolated communities grow one user at a time and
over a period of time either merge with the giant component
or cease to exist as the star loses focus. Merging stars form
the outer layer of a giant component. The core of a giant
component is formed by tightly connected group of very
active users. As time passes and users create more and more
connections, the average distance between users in a giant
component falls.
C. User Content Generation
Lei Guo et al. have divided OSNs into two categories based
on the purpose that they serve. The first category, called
networking oriented OSNs, consists of sites such as Facebook
and LinkedIn which place more emphasis on the networking
aspect of the social network. The other category, called
Knowledge oriented OSNs, consists of question/answer and
blogging websites which emphasize more on content
generation. The researchers studied some of the knowledge
oriented OSNs and proposed some patterns of user content
generation. Through their study they found that
-- Original content generation shows strong daily and
weekly patterns while non-original content do not exhibit any
such temporal pattern.
-- Users can be classified into three groups according to
their posting behavior – steadily posting, occasionally posting
and inactively posting.
-- 20% of users contribute to 80% of the total content in the
network
-- User contribution for original content is characterized by
the stretched exponential model with different parameters.
For example, for high quality content the distribution has a
low stretch factor indicating that it has been generated by a
small core group of users.
D. OSN use and personality
Online marketing efforts rely most on the personality of
individuals. As such, understanding the personality of OSN
users becomes very important. Yair et al. conduct data based
analysis about the personality of OSN users. The research
observes that extrovert individuals have higher number of
friends. This observation assumes importance because
marketers looking for extrovert people would be able to do so
by looking at the number of connections that the user has
formed. Thus number of users becomes the quantitative data
by which a person’s extrovertness can be measured. Neurotic
individuals would most likely provide more personally-
identifiable information than less neurotic individuals. On the
 3

other hand, neurotic individuals share much less private such as gender and age are related to disclosure of other
information than less neurotic individuals. The study also sensitive and highly personal information. Younger people
found that people having very high or very low levels of have more sensitive information on their profile. Users
agreeableness are inclined to upload more pictures than seeking a relationship tend to disclose the highest amount of
people with moderate levels of agreeableness. But agreeable highly sensitive information.
individuals use less page features than the other two
G. Categories of OSN applications
categories. People who are more open to newer experiences
tend to use more features in OSNs. Conscientious individuals OSNs applications have increased in popularity over the
have a higher number of friends but make lesser use of the last couple of years mainly among the younger generation.
photo upload features. Online marketing professionals would Research has categorized Facebook applications into various
find this research since it maps various aspects of individual categories as described below
personality with verifiable data. Category Description
Allows users to compare best
E. Q/A in OSNs Friend comparison friends and compare friend
Increasingly people are turning to OSNs to fulfill their traits
information needs. Getting answers from an OSN has some Allows users to exchange
advantages over traditional searching. The information source Casual communication messages and write on each
is generally trusted since invariably it is one of the other’s walls
connections that the user has. Moreover since users tend to Enables users to review,
form connections based on commonalities, it is expected that Rating/Recommendation compare and recommend
the answers given would be more relevant than what would items
be acquired through a search, especially for opinions. Allows users to perform
Meredith Ringel Morris et al., conducted to survey to Gestures virtual gestures like poke and
understand the nature of Q/A that occurs in OSNs. The bite
question types and question topics that users ask in OSNs are Enables users to express
as follows Self expression
moods, political opinions etc
Enables users to exchange
Question type Question topic Gifting
gifts
Recommendation Technology
Opinion Entertainment
Factual knowledge Home and family H. Comparison of privacy policies of various OSNs
Rhetorical Professional Privacy policies are an important means by which online
Invitation Places social networks articulate how the data that users entrust with
them would be used or would not be used. In order to
The motivation for asking questions and answering compare the privacy policies of the various OSNs, a
questions which others have asked is as follows taxonomy of the various terms have been created. Before
going into the comparison, let us understand the taxonomy.
Questions Answers First we define the various elements of a privacy policy.
Trust Altruism Purpose – Purpose defines the reasons for which OSNs
access specific pieces of information of its users.
Subjectivity of question Expertise
Visibility – Visibility defines who is allowed to access
Belief that search engines Properties of question
users’ data
would not work
Granularity – It defines the degree of precision that is
Specific audience Nature of relationship
revealed in response to a query for a specific information on
Connect socially Connect socially
an user
Retention – Retention determines the time period during
F. Disclosures in OSNs which access to the data would be allowed
Amanda Nosko et al., conducted research on Canadian
Facebook users to come to a conclusion about the level of Next we look at the various layers of an OSN and the data
disclosure of Facebook users. They have classified personal needs at each of these layers.
information into default/standard, sensitive and potentially Registration – Registration refers to the data that can be
stigmatizing. Their survey reveals that approximately 25% of used to identify a user uniquely
all possible information that could potentially be disclosed by Networking – This layer consists of information required
individuals are disclosed. Disclosure of personal information by the OSN to build the user’s network.
 4

Content – This layer consists of the content that the users As evident from the above table, OSNs by and large use data
push into the network as part of his participation in the OSN. for any purpose. Youtube has the least privacy is this regard.
Activity – This layer consists of web server logs, cookies
and other means by which the OSN tracks its users and his Visibility
activity on the OSN.
For each of the elements of the privacy policy, there are Visibility LinkedIn Twitter Orkut Facebook MySpace YouTube
some legends which will help to understand where each OSN
stands with respect to its privacy policy. These are as shown Registration H H H H H H
in the table below
Networking FAW FAW FN HN FAW AW
Purpose Visibility Granularity

RSm=Reuse H=House S=Specific Content FAW FAW FN HN FAW AW

Same

RS=Reuse F=Friends P=Partial Activity H H H H H H

selected

RA=Reuse Any FoF=Friends of

friends Visibility is the highest in Youtube. This is explainable
A=Any N=Network since the videos shared on Youtube are generally for the
viewership on the entire Internet. If we compare Facebook
A/W=All/World and MySpace, we find that Facebook is much more restrictive
than MySpace.
In purpose, it has been seen that OSNs collect data for any
purpose or for a single specific purpose. Then this data may Granularity
be used for any purpose. RSm stands for reuse same. It means
that the data that a user provides for a specific purpose is Visibility LinkedIn Twitter Orkut Facebook MySpace YouTube
reused by the OSN for that very specific purpose. RS stands
for reuse selected. It means that the OSN uses a subset of all Registration S S S S S S
the data the user provides for a specific purpose. RA denotes
that the OSN reuses the data that the user gives for a specific
Networking S,P S,P S S S S
purpose for any other. A stands for any. It denotes that the
OSN collected for any purpose and then reuse it for any
purpose. Content S S,P S S S S
In terms of visibility, house denotes the entire OSN, friends
denote the connections, network denotes the entire reach of Activity P P S S S S
the user and all world denotes the entire Internet.
In granularity, specific and partial are the two levels of
granularity in which if an OSN has specific granularity, it
In terms of granularity, it must be noted that most of the
means that the information provided is very specific while if
OSNs are responsible enough to provide information with
it is partial, it means that the information provided may not
specific granularity. In this respect, twitter and LinkedIn lag
be very specific.
behind.
Purpose
I. Analysis of Facebook’s privacy policies
Purpose LinkedI Twitte Orkut Face My You
Facebook’s privacy policies are being hotly debated. In this
n r book Space Tube section we analyze the various aspects of Facebook’s
controversial privacy policies.
Registration RA RA RSm RA RS RA
First let us understand what information Facebook shares
Networking A A A RAA A A with its applications. The Application Terms of Service state
Content A A A RAA A A
“Facebook may...provide developers access to...your name,
your profile picture, your gender, your birthday, your
Activity RS RS A RA A A hometown location...your current location...your political
view, your activities, your interests...your relationship status,
your dating interests, your relationship interests, your summer

plans, your Facebook user network affiliations, your education
history, your work history,...copies of photos in your
Facebook Site photo albums...a list of user IDs mapped to
your Facebook friends.”
This shows that practically all of a user’s data are shared
with the applications that he uses. What is more, if you set
your profile to private and one of your friends adds an
application, most of your profile information that is visible to
your friend is also available to the application developer --
even if you yourself have not installed the application.
Applications installed in Facebook need and use user data.
The graphic below shows the data need and the data actually
used by applications.
As seen from the graphic, 90.7% of Facebook applications
are given more privileges than they need.
Facebook’s privacy policies are continuously evolving and
getting more complicated. Let us look at the change in
Facebook’s privacy policies from 2005 onwards.
2005
"No personal information that you submit to the Facebook
will be available to any user of the Web Site who does not
belong to at least one of the groups specified by you in your
privacy settings."
The implication here is that Facebook is not giving your
data to anyone who is not related to you in some way.
2006
"We understand you may not want everyone in the world to
have the information you share on Facebook; that is why we
give you control of your information. Our default privacy
settings limit the information displayed in your profile to your
school, your specified local area, and other reasonable
community limitations that we tell you about."
The implication here is that Facebook will provide
everyone will some of the details you add to your profile page.
2007
"Profile information you submit to Facebook will be
available to users of Facebook who belong to at least one of
the networks you allow to access the information through
your privacy settings (e.g., school, geography, friends of
friends). Your name, school name, and profile picture
5
thumbnail will be available in search results across the
Facebook network unless you alter your privacy settings."
The implication here is that apart from some of the profile
information becoming public, Facebook will allow people to
search for you and display those information.
November 2009
“Facebook is designed to make it easy for you to share your
information with anyone you want. You decide how much
information you feel comfortable sharing on Facebook and
you control how it is distributed through your privacy
settings. You should review the default privacy settings and
change them if necessary to reflect your preferences. You
should also consider your settings whenever you share
information. ... Information set to “everyone” is publicly
available information, may be accessed by everyone on the
Internet (including people not logged into Facebook), is
subject to indexing by third party search engines, may be
associated with you outside of Facebook (such as when you
visit other sites on the internet), and may be imported and
exported by us and others without privacy limitations. The
default privacy setting for certain types of information you
post on Facebook is set to “everyone.” You can review and
change the default settings in your privacy settings."
The implication here is that the profile data is no longer
limited to Facebook users. It can be accessed by search
engines and others “without privacy settings.”
December 2009
"Certain categories of information such as your name,
profile photo, list of friends and pages you are a fan of,
gender, geographic region, and networks you belong to are
considered publicly available to everyone, including
Facebook-enhanced applications, and therefore do not have
privacy settings. You can, however, limit the ability of others
to find this information through search using your search
privacy settings."
The implication here is that some of the information you
provide is now public. There is no way you can have any
setting to make them private.
April 2010
"When you connect with an application or website it will
have access to General Information about you. The term
General Information includes your and your friends’ names,
profile pictures, gender, user IDs, connections, and any
content shared using the Everyone privacy setting. ... The
default privacy setting for certain types of information you
post on Facebook is set to “everyone.” ... Because it takes two
to connect, your privacy settings only control who can see the
connection on your profile page. If you are uncomfortable
with the connection being publicly available, you should
consider removing (or not making) the connection."
The implication here is that any interaction you have with
an application or an external website will help the application
or website to access to your and your friends' profile
information.

The word count of Facebook’s privacy policy statement has
grown from a mere 1000 words in 2005 to close to 6000
words in just 5 years. The journey of Facebook’s privacy
policy statement is shown in the graphic below
Even in comparison with other OSNs, Facebook has many
more words. The following graphic shows the comparison
with Flickr, Twitter, Youtube and others.
II. SECONDARY DATA RESEARCH
A. Average age distribution across OSNs
According to data available at Royal Pingdom for 19
OSNs, the age distribution is as follows
6
25% of users are in the age group 35-44. This the group that
determine the content and conversations taking place in
OSNs. Less than 3% of the users were of 65+.
B. Average age distribution across OSNs
The age group across all the websites studied showed that
not all the sites have similar distribution
The distribution pattern would be important to marketers.
Depending on the age category one would like to target,
advertising can be done on the site which shows the highest
percentage of OSN users of that age group. For example,
from the above figure, we can note that Bebo is very popular
among teenagers since more than 40% of the user base of
Bebo consists of people of the age group 0-17. Similarly, if
one needs to address all the age groups together, one would
look at Facebook since it has a balanced age group
distribution.
C. Estimated average age on OSNs
Photographs and grayscale figures should be prepared with
220 dpi resolution and saved with no compression, 8 bits per
pixel (grayscale). To obtain a 3.45-in figure (one-column
width) at 220 dpi, the figure should have a horizontal size of
759 pixels.
Color figures should be prepared with 400 dpi resolution
 7

and saved with no compression, 8 bits per pixel (palette or B. Number of hours spent
256 color). To obtain a 3.45-in figure (one column width) at
400 dpi, the figure should have a horizontal size of 1380
pixels.

III. PRIMARY DATA RESEARCH

Primary data was gathered with the help of a survey
questionnaire. In order to get a random sample, I used an
innovative approach. First I randomly chose a friend from my
friend list using a random number generator. I sent a message
to individual to fill in the survey. Then I again ran the
random number generator to choose a friend from his/her
friend list. I continued it for six hops. I started iterating these
over and over again. Unfortunately after sending out
messages to some people, I was blocked by Facebook from
sending further messages to individuals not in my network.
As a result I had to resort to the conventional method of A majority of the respondents said that they spend 5 to 10
sending out the survey through emails to various people. I hours each week on Facebook. But contrary to popular
also posted the questionnaire online along with a humorous opinion, only about 32% of the people spend more than 5
TV commercial ad as an incentive to filling in the survey. As hours per week on Facebook.
a result, I was able to elicit 34 responses to my survey out of
which 2 had to rejected because of incomplete information.
The results of the survey are given below.
A. Information displayed by Facebook users C. Number of friends

Number of friends

Less than 50
50 to 100
100 to 200
200 to 300
300 to 400
400 to 500
More than 500

Majority of the respondents surveyed said that they have

More than 90% of respondents said that they filled in their
gender in their Facebook profile page. Least number of people
said that they put their phone numbers in their profile page.
But it accounts for close to 40% of the respondents.
more than 500 connections. Very few people have less than
100 friends while almost equal number of people have 100 to
200, 200 to 300 and 300 to 400 friends.
Number of friends
D. OSN Access through cellphones
Around 30% of the respondents said that they access
Facebook through their cellphones implying that cellphones
have become an important to access OSNs. With the
penetration of mobile phones increasing and opening up of
3G, more users are likely to access OSNs through their
mobile devices
 8

G. Notice ads if they state benefits?

Mobile access
Response

Strongly disagree

Disagree
Yes
No Neither agree nor
disagree
Agree

Strongly agree

E. Do advertisements intrude into privacy?

Close to 60% of the respondents said that they notice
advertisements if they state the benefits. This is an important
lesson for marketers. Instead of focusing on glitzy
advertisements, marketers should try and state the benefits to
the users in simple terms.

We intend to understand the relation between the number

of years of exposure to Facebook and the number of hours
spent per week on Facebook. For this we design the following
hypothesis.
H0 : Classification based on number of years in Facebook
and the number of hours spent each week on Facebook are
independent
H1 : Classification based on number of years in Facebook
and the number of hours spent each week on Facebook are not
Close to 50% of respondents feel that advertisements actually
independent
do intrude into one’s privacy. This is a significant number
The analysis of the data is shown in the following tables
and brand marketers need to carefully weigh in this factor
before marketing their brands to the OSN audience.
F. Comfortable receiving ads based on interest?
 9

Therefore we can say that number of years on FB and

number of hours spent is not independent with a confidence
level of 97.5%
This confirms popular perception that as a user becomes
more used to Facebook, his usage increases. In other words,
Facebook is addictive.

IV. CONCLUSION
Understanding Online Social Networks and characterizing
users of OSNs is an important step towards successful online
marketing campaigns. This papers attempts to do just the
same. It must be noted here any research on OSN faces a
daunting task of collecting data. OSNs do not generally
provide interfaces to gather data. This paper suffers from two
limitations. First is that the survey sample is not random.
Attempts to create a random sample have failed. Secondly,
the sample size, although statistically adequate, should have
been larger. As an extension of this research, I propose to
continue collecting more data using both the surveys and
other means. One of the means that has been envisaged is to
create a host of applications, one each on the various
categories listed in this paper, and get some more data on
Facebook users. Most papers referred to in the literature do
not specifically talk of Indian Facebook users. This paper
aims to throw some light on the characteristics of Indian
Facebook users. The Indian Facebook user is different from
its US or European counterparts because Indian users are
probably demographically different from their US or
European counterparts.
 10

APPENDIX __Disagree
Survey Questionnaire __Strongly disagree
Are you comfortable receiving ads related to your interests
Age: and activities?
Qualification: __Strongly agree
__Undergraduate __Agree
__Graduate __Neither agree nor disagree
__Post Graduate __Disagree
How long have you been using Facebook __Strongly disagree
__Less than 1 month I notice the ads if they state the benefits I am likely to
__1 month to 6 months receive?
__6 months to 1 year __Strongly agree
__1 year to 2 years __Agree
__More than 2 years __Neither agree nor disagree
How many hours per week do you spend on Facebook? __Disagree
__Less than 5 hours __Strongly disagree
__6 to 10 hours Advertisements on Facebook are
__10 to 20 hours __Informative
__20 to 30 hours __Redundant
__More than 30 hours __Convenient
What information do you include in your profile? __Intrusive
__Age __Short on details
__Gender __Compact and relevant
__Hometown/City How often do you click on an ad because your friend liked
__Email it?
__Phone Number __Always
__Relationship status __Sometimes
__Movies/Music/Books __Never did
__TV Shows __Never did and never will
__Activities
__Others
How many friends do you have on Facebook? REFERENCES
__Less than 50 [1] “Structure and evolution of Online Social Networks”, Ravi Kumar,
Jasmine Novak, Andrew Tomkins
__50 to 100 [2] “Social Network use and personality”, Yair Amichai-Hamburger, Gideon
__100 to 200 Vinitzky
[3] “All about me: Disclosure in Online Social Networking profile”, Amanda
__200 to 300 Nosko, Eileen Wood, Seija Molema
__300 to 400 [4] “Analyzing patterns of user content generation in Online Social
__400 to 500 Networks”, Lei Guo, Enhua Tan, Songqing Chen, Xiaodong Zhang,
Yihong Zhao
__More than 500 [5] “What people ask their social networks and why” Meredith Mingel Morris,
Do you access Facebook from your cellphone? Jaime Teevan, Katrina Panovich
[6] http://royal.pingdom.com/2010/02/16/study-ages-of-social-network-users/
__Yes accessed on 01 December, 2010
__No [7] “Analysis of Social Networking Privacy Policies”, Leanne Wu, Maryam
Do you notice advertisements on Facebook? Majedi, Kambiz Ghazinour and Ken Barker
[8] http://www.nytimes.com/interactive/2010/05/12/business/facebook-
__Yes privacy.html
__No
Would you like to receive information on products, services
and offers in Facebook?
__Yes
__No
Do you think advertisements are an intrusion into your
privacy?
__Strongly agree
__Agree
__Neither agree nor disagree