2. 20112
Get unique and instant snapshot of your
personality transformed into book library.
E-contenta performs Facebook data
analyses, identifies your preferences and
recommends books to read.
3. 20113
E-contenta is a predictable analytics platform focused on using social media data to
provide more targeted purchase recommendations to users and more effective
marketing strategies for e-commerce companies
“Special sauce”, personalized recommendation system based on social media and
online user behavior, allowing for higher success rate in selling e-books, movies,
audiobooks and music.
First release of E-Contenta will be based on integration of Facebook-sourced data
with Amazon’s product offerings. Both systems have open API. Recommendation
algorithms will based on social media behavior, while Amazon will serve as a book
catalogue.
Expecting revenue of $8.8 million and EBITDA of $1.8 million in the 1st
year after the
product launch
Our mission is to be the best proactive recommendation system and achieve
revenue of $20-$50 million in 3-5 years
Executive summary
4. 20114
Around 80% of readers
prefer digital book to paper
because it’s easier to find
and read while travelling
US e-book market is the biggest in
the world: $2,79 billion
70% buy e-books on Amazon The share of e-book market in US is
growing from 0,6% in 2008 to 10% in
2012 (in some genres up to 25%)
38% prefer tablets as a key
device for e-reading (+25%
growth compared to 2011)
60% of tablets users can
be found in the airports
FB is the leader: 1 billion
active users (2012)
All adult internet users who use
social networking sites increased
from 8% in 2005 to 65% in 2011
(WW)
Trends that initiated the E-contenta
business idea
11. 201111
1. Check-in at the airport
2. Get a Facebook or other ad
3. Apply for E-contenta app
4. Get personalized recommendations
5. Refer friends and get points that can be exchanged for
discounts on purchases
How to get first users?
12. 201112
Promo and outdoor ads in
the airports
Joint campaigns with airport
internet providers, airport
restaurants and cafes and
airline companies
Joint campaigns with
publishing houses
Joint campaigns with tablet
producers
Offline
OnlineMarketing
13. 201113
Allows e-tailers to have
more targeted approach to
their potential consumers
Facilitates targeted selling
of e-books, audiobooks,
movies, music or media
items/subscriptions to
customers
B2
C
B2B
Business models
14. 201114
Hypothesis
• People will look for
something interesting to
read, watch or listen while
travelling
• They are interested in
getting personalized
recommendations
• They will start using E-
Contenta
• The most satisfied users will
recommend the app to
friends
Metrics
• App views/app users = 70%
• Paying users/users = 15%
• % of those who share
information in FB friends
community – 35%
• Repeated purchase – 40%
from DB every 2nd
month
16. 201116
Competitors do not have social network strategy and prefer to work
within their stand-alone portals
Amazon-60-70%, BN-20%, iBooks-10%, others (Books-A-Million,
Kobo, Sony and others)
Classic recommendation system is based mostly on
previous reading, viewing and ratings
The only feature based on info taken from social media is displaying
what your Facebook friends are reading, [watching, listening]
There is specific niche that could be carved – be
proactive, adapt to individual preferences and react to
needs and wishes of every particular user
Competitor s
17. 201117
Geo-expansion
Integration with new social
networks
New content types = transmedia
New knowledge map schemes =
more and more personalized
recommendation approach
Strengths
Growing market of e-books,
audiobooks, movie and music
downloads
Micro payments
Agent role = minimum of risks
E-business = logistics transparency
Statistics = effective service
management
Dependence on large and powerful
players in the digital world
Competitors
Market immaturity
Ability to execute on the business
plan
Adoption of the product
Potential competition from e-tailers
if they decide to copy the strategy
Weaknesses
Opportunities
Risks
Amazon – as a free source of content for pilot release. Integration through open Amazon API
E-contenta business idea is based on two remarkable trends that we can see now: first trend is a tremendously increasing share of e-book market, the second one – growing number of social media users. Nowadays more than 65% of all internet users WW explore social media. Mostly Facebook. At the same time e-book market is growing and US is the biggest one in the world – $2,79 billion. Imagine, that the share increased from 0,6 in 2008 to 10% in 2012. The major part of e-readers prefer to buy books on Amazon. What device do they use? About 40% prefer tablets and this percentage is growing! 60% of tablets users can be found in the airports. And guess why 8 from 10 people would rather select e-book? Because they are easier to find and they like to read e-books while travelling! These fascinating findings will help to define key features of E-contenta, its functionality for the pilot release and the market to position on. The concept is clear. For e-reading Americans use KindleFire – 40%, Nook – 7%, iPad – 10%. New tablet Kindle Fire 79-199 $ named iPad Killer
E-contenta recommendation approach will be different from the methodology that is being used now by the majority of book websites work. Classic recommendation system is based mostly on previous reading choices. Social media provide us with great amount of data. It can and should be turned into buying recommendations. 1) Approach based on books description and lexis from user's facebook page. This approach will be done first. We have crawled about 1000000 book's description s from www.goodreads.com. Then, we divided it to several(about 500) different clusters, using k-means algorithm(http://en.wikipedia.org/wiki/K-means). For every cluster we found best books (based on goodreads user's ratings). For recommendation we get keywords from users page(profession, interests, likes etc.), find the most closest cluster of books for this keywords and recommend most rated books from this cluster. 2) Approach based on classic machine learning algorithms. Some users post on their facebook page books, that they like. We can crawl a lot of such profiles from facebook and use it as traning set for classical machine learning algorithms. For this algorithms we must extract features from users page. As features we can use for example: - how often user chanche his avatar - how many friends user have - age - gender - marital mtatus - interests and many other features. We can decribe user as set of such features, and using classical machine learning algorithms(see for example http://en.wikipedia.org/wiki/Support_vector_machine) build recommendation engine on our traning set. 3) Collaborative filtering - classical recomendation approaches, used on large sites such as Amazon.com. This is based on compare preferences of different users. Base idea is follow: if a lot of users who like book A like book B, then if new user like book A we can recommend him book B. See http://en.wikipedia.org/wiki/Collaborative_filtering for details. 4) Using thrid-party search engines. We automatically can search in search engines such as google books relevant for user. For example, by this link: www.google.ru/search?client=ubuntu&channel=fs&q=k-means&ie=utf-8&oe=utf-8&gws_rd=cr&ei=hWQEUof0Je3M0AXo2ID4Dg#bav=on.2,or.r_cp.r_qf.&channel=fs&fp=6e282c34c22913bd&newwindow=1&q=book for venture investors&safe=off we can grab some books for venture investor. Profession we can get from facebook profile.
Reference audience to test concept – airport passengers. More than 60% of tablet users can be found in airports and 80% of travellers prefer e-reading to paper because they find the process of acquiring, transporting and reading digital books easier. After the initial trial of the program at one of the airports, we can expand the program and gradually implement E-Contenta in other airports worldwide. E-Contenta is a “win-win” offering for all the participants. Passengers will get targeted content recommendations and ability to get uniquely-matched digital media, while airports can promote other services to passengers through E-Contenta’s recommendation system. At the same time, it is a good opportunity for airports and airlines to increase the loyalty of passengers and enhance their social network image as passengers share quotes, recommendations, etc., with their friends. E-Contenta will facilitate this interaction and will serve as a connector that facilitates interaction among providers of digital contents, passengers who are eager to get uniquely-matched digital contents, and airports which want to promote their brands and services. The target of the trial is to draw approximately 200 thousand users, process their feedback and adjust the platform to their needs.
At the initial stages we will be more focused on B2C model: will get money from users who will enjoy individual recommendations and buy books through the E-contenta application. While the number of application users is growing companies such as publishing houses or other content distributors or independent writers may be interested in us to get very targeted access to book buyers. Distribution models. We can sell directly to users or through the channel. Easy integration with 3 rd party websites allow to grow network of channel partners. While releasing a new book a publisher always have a question how to market it best. E-contenta will allow to run targeted personalized marketing campaigns and gain maximum ROI. E-contenta will keep information about its users and their buying behaviour to provide publishers with trend analysis and predictive forecasts. Flexible service fee will be affordable for a business of any size. Here we need a quote from publisher how briliant the idea is… Business diversification Additional income source No investment needed Loyalty of existing customers