15.

570Fall2014
DigitalMarketing&SocialMediaAnalytics
ProfessorSinanAral
T/TH:1230pmRoom:E62262
2304pmRoom:E62262

OfficeHours
Internet

ProfessorSinanAral
E62262
TuesdaysandThursdays,SectionA:1230pm/SectionB:2304pm
SectionA:ParulBatra<pbatra@mit.edu>;
SectionB:NoelSequeira<noel.sequeira@sloan.mit.edu>
ByAppointment
Email:sinan@mit.edu;URL:http://web.mit.edu/sinana/www/
Twitter:@sinanaral

01

Instructor
Classroom
Classtimes
TAs

Fa

ll
2

TopicsCovered:SearchMarketing,SocialNetworkMarketing,SocialMediaAnalytics,UserGenerated
Content Management and Marketing, Mobile Advertising and Commerce, CRM Strategy in the Age of
Big Data and Digital Advertising, Earned vs. Paid Media, Predictive Modeling for Ad Targeting, Viral
Product Design, the Multichannel Experience, Randomized Experimentation, AB Testing and Causal
InferenceinMarketingStrategy.

SummaryandObjectives

Newdigitaltechnologieshavefundamentallyreshapedmarketingtheoryandpracticeinthelastdecade
alone. Technology has changed the modes of communication through which firms engage with
consumers. Moores law has made the storage and analysis of consumer data scalable, creating
opportunities for finegrained behavioral analytics. New monitoring tools have fostered precise and
personalized customer relationship management practices. The rise of mobile phones and tablets has
enabled location based messaging and reciprocal communication. The ubiquity of video content has
promulgatedrich,nativeadvertisingprograms.Theglobalemergenceofsocialnetworkinghasenabled
networked based predictive modeling and new forms of targeting and referral strategies based on the
preferencesofconsumerspeers.Andfinally,newsocialmediahavebroughtallofthisontothepublic
stage,withwordofmouthconversationsdrivingbrandawarenessandbrandloyalty,andusergenerated
contentonreviewandratingssitesmakingorbreakingdemandforproductsorservices.

15.570willprovideadetailed,appliedperspectiveonthetheory andpracticeofDigitalMarketingand
SocialMediaAnalyticsin the 21st century. We will cover topics like SearchMarketing,Social Network
Marketing, Social Media Analytics, User Generated Content Management and Marketing, Mobile
Advertising and Commerce, CRM Strategy in the Age of Big Data and Digital Advertising from the
displayadecosystemtoadretargeting.Wewillcoverconceptssuchasthedifferencebetweenearnedand
paidmedia,predictivemodelingforadtargetingandcustomerrelationshipmanagement,measuringand
managing product virality, viral product design, native advertising, and engaging the multichannel
experience. Throughout the course we will specifically stress the theory and practice of randomized
experimentation, AB testing and the importance of causal inference for marketing strategy. The course

Fa

ll
2

01

will combine lectures, case studies and guest speakers with relevant industry experience that speak
directlytothetopicsathand.

Uponcompletionofthiscourseyoushouldhaveafundamentalunderstandingof:

DigitalAdvertisingEcosystemandAttributionandPricingModelsforDigitalAdvertising;
AnunderstandingofthefundamentalsofWebandAppAnalyticsandKPIsforWebtrafficand
commerce;
SearchEngineMarketing;SearchEngineAdvertising;AdAuctionsandStrategiesforoptimizing
SearchEngineAdvertising;
SocialNetworkMarketing,SocialNetworkTargeting;PredictiveAnalyticsusingSocialNetwork
Data,PeertoPeerMarketingandPersonalizedSocialAdvertising;
TargetingandSegmentation,specifically:DemographicTargetingandSegmentation,Behavioral
TargetingandSegmentation,SocialTargetingandSegmentation;
TwitterAnalyticsandAppliedTwitterAnalytics(e.g.SocialTVandSocialNews);
Social Listening: Analysis of User Generated Content, Reviews, Ratings and their effects on
consumerdemand;and
MobileCommerceandAnalytics.

Students are expected to be properly prepared for the class and to have thoroughly read the assigned
readingsandcases.Studentsareexpectedtoparticipateactivelyinclassdiscussion.

Courserequirements

Detailsabouttheassignments,projectsanddeliverablesaswellasthedatathatwewillusetoexamine
themwillbeforthcomingthroughoutthesemester.However,thefollowingprovidesasketchofthefour
main requirements of this course. These four requirements describe completely the deliverables and
activitiesstudentswillbeevaluatedon:

1. ClassParticipation:

It goes without saying that a vibrant class discussion is essential to the learning outcomes of the
class.Beingproperlypreparedforclassandparticipatingintheclassisarequirement.Wewillcold
callinclass,keeptrackofeveryonescontributionsandwillrewardproductiveparticipation.Note:
we mean productive participation, not just participation. A productive comment is one that guides
the class to learn more, dig deeper, understand more comprehensively etc. Just talking is at times
evencounterproductive.So,focusonproductiveparticipation,notjustparticipation.

2. DiscussionQuestions:

Each week, you will typically be assigned preclass work a week in advance of the session. Your
responses, designed to have you reflect on the readings, should not exceed one page. This is not
designed to create a lot of work, but rather to encourage you to read and reflect on the readings
beforeclass.Youwillbeaskedtosubmitanswerstodiscussionquestionspertainingtothereading
via Stellar before 5pm the night before class. 6 of the 11 class sessions will have Discussion
Questionsassigned.Youwillbeaskedtosubmit5outof6assignmentsduringthesemester.Twoof
thosesessionassignmentsSessions6and11,correspondingtocasereadingsarerequiredofall
studentsandarenotoptional.Youwillthenbeabletochoose3oftheremaining4sessionsinwhich

01

ll
2

4.

Fa

3.

discussionquestionsareassigned.TheschedulebelowlistsweeksinwhichDiscussionQuestionsare
assigned(DQ)andweeksinwhichRequiredDiscussionQuestionsareassigned(RDQ).

IndividualProject(JumpingtheShark):

The Individual Project builds directly off of the session on Social Network Marketing and Viral
Product Design. The individual project is designed to give students an understanding of viral
dynamics and the dynamic processes of new user acquisition, viral sharing, viral product design,
community carrying capacity etc. The project involves analyzing a simulated dataset of viral user
acquisition, adjusting and tuning the parameters, establishing the key factors that govern viral
growthprocesses,describingsustainablestrategiesforviralgrowthunderseveralassumptionsand
recommendationstoimprovetheviralmodelitself.Amodelanddatasetofuseracquisitionwillbe
providedandeachstudentwillbeaskedaseriesofquestionsthatexplorethedataandthemodel.

GroupProject(AHolisticDigitalMarketingStrategyforHighNote,Inc.:MarketingtoFreemium
Communities):

TheGroupProjectinvolvesanindepthanalysisofDigitalMarketingStrategyatHighNote,areal
online music streaming community whose name has been changed to preserve the privacy of its
consumers.Youwillformaprojectteamofeither2or3peopleintheearlyweeksoftheclass(teams
must be formed and communicated to us by 11/6). During the semester you will familiarize
yourselveswiththecasematerialsandthedata,yesthedata!Wehave9monthsofrealcustomer
datafromthefirm,totalingover100Kobservations.Thedatadetailsa)demographic,b)behavioral
andc)socialnetworkcharacteristicsofusers.

Thepurposeoftheprojectistogetreal,firsthandexperiencewithdatadrivenmarketingstrategy.In
particular,youwillbetaskedwithcreatingaDigitalMarketingStrategyaimedatMonetizingthe
Service i.e. a strategy for increasing the number of paying subscribers. One might think about
many different approaches, from generating new premium customers to converting current
subscriberstothepaidpremiumservice.Yourgoalwillbetomakeaseriesofrecommendationsto
theCMOofthecompanydetailinghowyouthinksheshoulddevisehermarketingstrategytoward
thisend.Thecatchisthatyourrecommendationswillneedtobebackedupbyevidenceevidence
fromanalysisofthecompanysdataaswellasanyevidencefrompublicallyavailabledatayouwant
to collect about online music streaming, freemium models and the various digital marketing
strategies that she may want to implement (strategies we learned about in class, e.g. Search
Advertising,MobileAdvertising,SocialNetworkMarketing,DisplayAdvertisingetc).

Therewillbedeliverablesfortheproject:

a) Project Plan (Due 11/25 35% of the Project Grade): The Project Plan (~ 12 pages) will detail
what your team plans to do to investigate the various marketing strategies that you might
recommend.Theplanwilldescribequestionsyouareaskingofthedata,researchyouaredoing
orwilldo,analysesyouareconductingorwillconductetctowardpreparingafinalreport.Roles
and responsibilities of different team members should be outlined and a rough timeline of
activitiesshouldbepresented.

b) Final Report (Due 12/12 65% of the Project Grade): The Final Report (~ 68 pages) will detail
yourrecommendationstotheHighNoteCMO.ItshouldconsistofanExecutiveSummary(~2

pages),whichdetailstherecommendationsthatyouaremakingandahighleveljustificationfor
those recommendations; as well as an Appendix (~ 46 pages), which shows your work and
justifies the recommendations made in the Executive Summary with evidence (e.g. analyses of
consumer data, outside data). Justifications might include regression analysis and predictive
modelsofpremiumsubscriptionbehaviorbuiltusingtheconsumerdata(e.g.a)demographic,b)
behavioralandc)socialnetworkcharacteristicsofusers);reportsofhighlevelsummariesofthe
means and standard deviations in the data; correlation analysis linking a) demographic, b)
behavioral and c) social network characteristics of users to premium subscription behavior;
evidencefromstatisticsaboutvariousdigitalmarketingstrategieswelearnedaboutinclass(e.g.
SearchAdvertising,MobileAdvertising,SocialNetworkMarketing,DisplayAdvertisingetc).

ParticipationinClassDiscussion:20%
GroupProject:40%

Fa

ll
2

01

Grading

IndividualPreClassAssignments:20%
IndividualAssignment:20%

Class Schedule

Assignment
Due

Guest/
Case

10/30 DigitalAdvertising&Attribution

DQ

11/4

WebAnalytics&SearchAdvertising

DQ

11/6

SocialNetworkAnalytics&ViralProductDesign

DQ/
HN:Teams

11/11 StudentHolidayNOCLASS

11/13 TowardPersonalizedSocialMarketing

11/18

11/20 DataDrivenAdvertising

11/25 TwitterAnalytics

11/27 ThanksgivingNOCLASS

12/2

Date Session

10/28 IntroductiontoDigitalMarketingStrategy

01

ll
2

DataMining&DigitalCRMStrategy // Group
Project:MarketingtoFreemiumCommunities

11 12/9

Fa

10 12/4

SocialListening:Ratings,Reviews&Demand
MobileCommerce&Analytics
AHolisticDigital&SocialStrategy

12/12 EndofClass(GroupProject:HighNoteDue)

Jumping
theShark

Jamie
Tedford

RDQ

Harrahs
Case

Claudia
Perlich

HN:Project
Plan

Deb
Roy

DQ

RDQ

Sephora
Case

HN:Final
Report

Reading List
Most reading materials can be found on Stellar. Those materials not on Stellar (including the High Note
Group Project Materials) will be available in the Study.net course pack.
Session 0: Group Project Materials
1. Group Project Materials: Do Your Online Friends Make You Pay? A Randomized Field Experiment
in an Online Music Social Network Ravi Bapna & Akhmed Umyarov.
2. Group Project Materials: Content or Community? A Digital Business Strategy for Content Providers
in the Social Age Gal Oestreicher-Singer & Lior Zalmanson (2013) MIS Quarterly, Vol. 37(2), 591616.
3. Group Project Materials: High Notes Freemium Conundrum (Case) Ravi Bapna & Lee C.
Thomas; University of Minnesota Case # SOBACO-2014-01-2.
4. Group Project Materials: High Note Data Dictionary (CSV/Excel File)

01

Session 1: Introduction to Digital Marketing Strategy (10/28):

5. Group Project Materials: High Note Data (CSV/Excel File)

6. The Future of Advertising Danielle Sacks (2010); Fast Company; November.

ll
2

7. Branding in the Digital Age: You're Spending Your Money in All the Wrong Places David Edelman
(2010); Harvard Business Review; December, Vol. 99(12), 62-69.
Session 2: Digital Advertising & Attribution (10/30):

8. Advertising Analytics 2.0 Wes Nichols (2013); Harvard Business Review; March, Vol. 91(3), 6068.

Fa

9. Digital Attribution Comes of Age Forrester Thought Leadership Paper (2012); June.
Session 3: Web Analytics & Search Advertising (11/4):
10. Segmentation in Web Analytics Jacques Warren (2012); AT Internet White Paper.
11. Paid Search Advertising Background Note Rajkumar Venkatesan & Shea Gibbs (2014); Darden
Business Publishing; UVA; May 27.
12. Paid Search: The Innovation that Changed the Web Des Laffey (2007); Business Horizons, Vol.
50(3), 211-218.
Optional Reading:
13. Web Analytics Definitions v.4.0 Jason Burby, Angie Brown & WAA Standards Committee; Web
Analytics Association.
Session 4: Social Network Analytics & Viral Product Design (11/6):
14. What Would Ashton Do - And Does it Matter? Sinan Aral (2013); Harvard Business Review; May,
Vol. 91(5) 25-27.
15. Identifying Influential and Susceptible Members of Social Networks. Sinan Aral & Dylan Walker
(2012) Science, July 20: 337-341.

16. Dear Klout, This is How You Measure Influence. Gregory Ferenstein (2012), Tech Crunch, June
21.
Optional Reading:
17. Creating Social Contagion through Viral Product Design: A Randomized Trial of Peer Influence in
Networks. Aral, S. & Walker, D. (2011) Management Science, 57(9); September: 1623-1639.
18. Distinguishing Influence Based Contagion from Homophily Driven Diffusion in Dynamic
Networks, Aral, S., Muchnik, L., & Sundararajan, A. (2009) Proceedings of the National Academy
of Sciences (PNAS), Dec. 22, 106(51): 21544-21549.
Session 5: Toward Personalized Social Marketing: Jamie Tedford (Brand Networks) (11/13):
19. Take a breather! Youre off the hook for any reading today . You should spend this time reading the
group project materials and taking time to make progress on the project with your team!
Session 6: Data Mining & Digital CRM Strategy (11/18):

20. Competing on Analytics Thomas Davenport; Harvard Business Review; January 2006, Vol. 84(1),
98-107.

01

21. Case: Harrahs Entertainment Inc. HBS Case # 9-502-011.

Session 7: Data Driven Advertising (Claudia Perlich, Dstillery) (11/20):

ll
2

22. Data Mining: On the Trail to Marketing Gold Shawn Thelen, Sandra Mottner and Barry Berman
(2004) Business Horizons; Vol. 47(6), 25-32.
Session 8: Twitter Analytics (Deb Roy, Twitter & MIT) (11/25):
23. Social Scientists Wade into the Tweet Stream. Greg Miller (2011); Science; 333.
24. Where Is Social TV Heading in 2014? Jesse Redniss (2013); Advertising Age; December.

Fa

25. Social TV Is Dead; Long Live Social TV Simon Dumenco (2014); Advertising Age; May.
26. To Go from Big Data to Big Insight, Start with a Visual Sinan Aral (2013); Harvard Business
Review Blog; August 27.
Optional Reading:

27. Everyones an Influencer: Quantifying Influence on Twitter Eytan Bakshy, Jake Hofman, Winter
Mason, and Duncan Watts (2011) Proceedings of the fourth ACM international conference on Web
search and data mining, 65-74.
Session 9: Social Listening: Ratings, Reviews & Demand (12/2):
28. The Trouble with Online Ratings Sinan Aral (2013) Sloan Management Review, 55(2).
29. What Marketers Misunderstand About Online Reviews Itamar Simonson and Emanuel Rosen
(2014); Harvard Business Review; January-February, Vol. 92(1/2), 23-25.
Optional Reading:
30. Social Influence Bias: A Randomized Experiment. Muchnik, L., Aral, S., & Taylor, S. (2013)
Science, 341(6146): 647-651.

Session 10: Mobile Commerce & Analytics (12/4):

31. Marketing to Generation M(obile) Farenna Sultan & Andrew J. Rohm (2008); Sloan Management
Review; 49(4): 35-41.
Session 11: A Holistic Digital & Social Strategy (12/9)

Fa

ll
2

01

32. Case: Sephora Direct: Investing in Social Media, Video and Mobile HBS Case # 9-511-137.

Guest Speakers

JAMIE TEDFORD

01

Chief Executive Officer, Brand Networks

Fa

ll
2

Jamie Tedford is the CEO and founder of Brand Networks. Prior to founding Brand Networks in
2007, Jamie was SVP, Marketing and Media Innovation at Arnold Worldwide, a top-10 US brand
advertising agency. Recognized as an innovative entrepreneur and a pioneer in the growing
Word of Mouth industry, Jamie launched the WOM division at Arnold and is a founding board
member of the Word of Mouth Marketing Association. Jamie is a graduate of the University of
New Hampshire, and he lives in Marblehead, Mass.

CLAUDIA PERLICH
Chief Scientist, Dstillery
Claudia Perlich serves as Chief Scientist at Dstillery and in this role designs, develops, analyzes
and optimizes the machine learning that drives digital advertising to prospective customers of
brands. An active industry speaker and frequent contributor to industry publications, Claudia
enjoys serving as a guide in world of data, and was recently named winner of the Advertising
Research Foundations (ARF) Grand Innovation Award and was selected as member of the

DEB ROY

ll
2

01

Crains NY annual 40 Under 40 list. Additionally, she has been published in over 30 scientific
journals, and holds multiple patents in machine learning. She has won many data mining
competitions, including the prestigious KDD Cup three times for her work on movie ratings in
2007, breast cancer detection in 2008, and churn and propensity predictions for telecom
customers in 2009, as well as the KDD best paper award for data leakage in 2011 and bid
optimization in 2012. Prior to joining Dstillery, she worked in Data Analytics Research at IBMs
Watson Research Center, concentrating on data analytics and machine learning for complex
real-world domains and applications. Claudia has a PhD in Information Systems from NYU and
an MA in Computer Science from the University of Colorado. Claudia takes active interest in the
making of the next generation of data scientists and is teaching Data Mining for Business
Intelligence in the NYU Stern MBA program.

Fa

Associate Professor, MIT Media Lab & Chief Media Scientist, Twitter
Roy is an Associate Professor at the Massachusetts Institute of Technology (MIT) and is Chief
Media Scientist of Twitter. He conducts research at the MIT Media Lab on language, games,
and social dynamics at the intersection of artificial intelligence and cognitive psychology. In
2008 he co-founded and was the founding CEO of Bluefin Labs, a social TV analytics company,
which MIT Technology Review named as one of the 50 most innovative companies of
2012. Bluefin was acquired by Twitter in 2013. An author of over 100 academic papers in
machine learning, cognitive modeling, and human-machine interaction, his TED talk, Birth of a
Word, has been viewed over 3 million times. A native of Canada, Roy received a Bachelor of
Applied Science (computer engineering) from the University of Waterloo and a PhD in Media
Arts and Sciences from MIT. He is @dkroy on Twitter.

10