Folie 1 - Marketing

Social Contagion – An Empirical Comparison
of Seeding Strategies for Viral Marketing
Hinz, Oliver / Skiera, Bernd / Barrot, Christian / Becker, Jan U. (2011),
"An Empirical Comparison of Seeding Strategies for Viral Marketing",
Journal of Marketing, 75 (November), 55-71
Oliver Hinz
Bernd Skiera
TU Darmstadt
University of Frankfurt
Christian Barrot & Jan U. Becker
Kühne Logistics University, Hamburg
TRADITIONAL MARKETING INSTRUMENTS ARE FACING SHRINKING
EFFECTIVENESS IN THE FACE OF NEW SOCIAL MEDIA
A changing environment…
• Information over-flow: Traditional advertising instruments such as print ads or
TV commercials struggle to reach an audience growing tired of ever more ads
• Rise of social media: Communications is shifting towards digital social media, such as
facebook, twitter, email or SMS.
• Credibility: Studies have shown the higher effectiveness of customer-initiated
communication (e.g., word-of mouth) compared to advertising
• Effective customer acquisition: Marketing managers have discovered social
interactions between existing and potential customers as new sources for customer
acquisition
…requires new marketing instruments
• Companies discover innovative new methods to proactively stimulate and channel
Word-of-mouth Viral Marketing as savior
• Viral marketing: consumers mutually share and spread information, initially sent out
deliberately by marketers to stimulate and capitalize on word-of-mouth (WOM)
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
1
VIRAL MARKETING RAPIDLY GAINS GROUND
Global spending for viral marketing campaigns
$ 3.000 Mio.
$ 980 Mio.
$ 76 Mio.
2001
2006
2013 e
→ Shift from traditional marketing budgets towards viral
e
Forecast
Stephen, Andrew (2010): Viral Marketing: Tell a Woman, Working Paper, INSEAD, Fontainebleau.
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
2
VIRAL MARKETING UTILIZES THE ADVANTAGES OF PERSONAL
COMMUNICATIONS IN SOCIAL NETWORKS
Viral marketing
• Advantages
 familiar senders have a higher credibility for the recipient
 familiar senders are not blocked by spam filters (higher reception and open rates)
 low to very low cost (e.g., for distribution via SMS or email)
• Key success factors
 Content (e.g., funny, entertaining, surprising, motivating)
 Willingness-to-share, often stimulated by incentives (e.g., coupons, competitions,
financial rewards)
 Social Network Structure (e.g., connectedness)
 Seeding (selection of starting points to maximize campaign impact)
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
3
THREE POTENTIAL SEEDING STRATEGIES BASED ON SOCIOMETRIC
MEASURES ARE DISCUSSED IN LITERATURE (1 / 3)
Strategy 1: High-degree seeding
High-Degree
(hub)
Hypothesis: Seeding of individuals with a very high number of personal contacts (HighDegree) maximizes the reach of a viral marketing campaign
→ Supported by, for example, Katz/Lazarsfeld 1955; Rogers 1962;
Coleman et al. 1966; Rosen 2000; Weidlich 2000; Hanaki et al. 2007;
van den Bulte/Joshi 2007
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
4
THREE POTENTIAL SEEDING STRATEGIES BASED ON SOCIOMETRIC
MEASURES ARE DISCUSSED IN LITERATURE (2 / 3)
Strategy 2: High-betweenness seeding
High-Betweenness
bridge
Hypothesis: Seeding of individuals acting as “bridges“ or intermediaries between subnetworks (High-Betweenness) maximizes the reach of a viral marketing
campaign
→ Supported by, for example, Granovetter 1973; Kemper 1980;
Rayport 1996; Watts 2004
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
5
THREE POTENTIAL SEEDING STRATEGIES BASED ON SOCIOMETRIC
MEASURES ARE DISCUSSED IN LITERATURE (3 / 3)
Strategy 3: Low-degree seeding
Low-Degree
(fringe)
Hypothesis: Seeding of individuals with a small number of personal contacts (Low-Degree)
maximizes the reach of a viral marketing campaign
→ Supported by, for example, Simmel 1950; Becker 1970; Sundararajan
2006; Galeotti/Goyal 2007; Watts/Dodds 2007; Porter/Donthu 2008
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
6
PREVIOUS RESEARCH
Social Position has Positive Influence on …
Studies
Coleman, Katz, and
Menzel (1966)
Participation
Prob. Pi
Used Reach
ni
Expected #
Referrals Ri
Conversion
Rate wi
Expected #
Successful
Referrals SRi
Recommendation for
Optimal Seeding
Strategy
Hub
Hub
Hub
Becker (1970)
Hub
Fringe
Hub
Fringe
Hub
Fringe
Simmel (1950); Porter and
Donthu (2008)
Fringe
Watts and Dodds (2007)
Fringe
Hub
Fringe
Leskovec, Adamic, and
Huberman (2007)
Hub
Hub
Hub
Hub
Hub
Bridge
Bridge
Anderson and May (1991);
Kemper (1980)
Granovetter (1973);
Rayport (1996)
Fringe
Iyengar, Van den Bulte,
and Valente (2011)
Study 1
Hub


Fringe
Hub
Hub
Bridge
Hub
Controlled

Fringe
Fringe
Study 2
Study 3
Empirically
Tested
Seeding
Strategy

Hub
Hub

Hub, Fringe,
Bridge, Random

Hub, Fringe,
Bridge, Random

Hub, Fringe,
Random
Notes: i = focal individual. Expected number of referrals: Ri = Pi∙ni; Successful number of referrals: SRi = wi*Ri.
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
7
USING THREE COMPLEMENTARY STUDIES FOR EMPIRICAL TESTING
Overview
Study 2: Realistic setup
Study 1: "Controlled"
setup
(120 nodes, 270 edges)
(1,380 nodes, 4,052 edges)
Study 3: Real world
referral program
(208,829 nodes,
7,786,019 edges)
• Field experiment with withinsubject design
• Field experiment with
between-subject design
• Ex-Post analysis of
transaction data
• 120 students recruited from
leading digital social network
• Participants were business
students
• Participation awareness
"controls" for activity level
• Intrinsic motivation to share
interesting content (video
about their university)
• Identification of factors
driving social contagion
process
• Extrinsic motivation by
monetary referral reward
• Varying extrinsic motivation to
share secret tokens
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
8
INITIAL TEST OF SEEDING STRATEGIES IN SMALL, CONTROLLED
EXPERIMENT
Study 1: Design
Recruit
participants
Track social
network data
Model social
network and
calculate metrics
Seed secret
tokens
Track logins and
feedback entered
on website
Hinz, Skiera, Barrot & Becker
• 120 students recruited
• Precondition: Students have account on social network platform StudiVZ
• Collect data of mutual friendship relations from online platform
• 120 nodes with ~270 edges
• Degree and betweenness centrality calculated per node
•
•
•
•
4 seeding strategies: high / low degree, betweenness centrality, random
2 seeding levels: 10%/20% of network
2 incentive levels: high/low
4x2x2 factorial design = 16 secret tokens seeded
• Students spread the secret tokens (no groups, no forums allowed)
• Responses have been entered on a website using individual login
information
• Duration: 2 weeks per experiment
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
9
HIGH-DEGREE SEEDING STRATEGY MAXIMIZES RESPONSE
Study 1: Individual probability to respond
• Random Effects Logit Model
• High degree seeding maximizes
responses
• Decreasing marginal effect of
seeding
• Most responses for high degree
seeding (activity)
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
10
HIGH-DEGREE AND HIGH-BETWEENNESS STRATEGIES CLEARLY
OUTPERFORM RANDOM AND LOW-DEGREE STRATEGIES
Study 1: Conditional odds ratios of seeding strategies
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
11
TESTING SEEDING STRATEGIES IN REALISTIC EXPERIMENTAL SETTING
Study 2: Design
Track social
network data
Model social
network and
calculate metrics
Seed link to
video
Track website
visits and video
downloads
Hinz, Skiera, Barrot & Becker
• Collect data of mutual friendship relations
from social network platform
• Information obtained for all 1,380 students
with business-related subjects at University
• 1,380 nodes with 4,052 edges
• Degree and betweenness centrality
calculated per node
• Information seeded: link to funny Video about University
• 4 seeding strategies: high / low degree, betweenness centrality, random
(links to different websites, seeding at same day, HB/HD overlap removed)
• No additional incentives
• Four different (seeding strategy) website visit statistics recorded
• Experiment duration: 2 weeks
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
12
STUDY 2 CONFIRMS THE SUPERIORITY OF HIGH-DEGREE AND
HIGH-BETWEENNESS SEEDING STRATEGIES (1 / 2)
Study 2: Number of visits per day
•
•
•
•
Random Effects Model
High-Degree / High-Betweenness best seeding strategies
Clearly outperforming random and Low Degree seeding at every point in time
(Re-)seeding day dummy doubles R²
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
13
STUDY 2 CONFIRMS THE SUPERIORITY OF HIGH-DEGREE AND
HIGH-BETWEENNESS SEEDING STRATEGIES (2 / 2)
Study 2: Number of visits per day
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
14
REAL-LIFE APPLICATION OF VIRAL MARKETING CAMPAIGN USING
THE CUSTOMER BASE OF A MOBILE PHONE SERVICE PROVIDER
Study 3: Design
SMS campaign
aimed at
customer base
• SMS mailing to 208,829 customers of a low cost mobile phone service
promoting a special „refer-a-friend“ campaign
• As special promotion, the referral reward was increased by 50%
(15€ instead of 10€)
Conversion
tracking
• All referrals tracked through the website / call center of the service
provider
• 4.549 customers participated
• 6.392 successful referrals
Establishing
the social
network
• Calculation of Degree Centrality on the basis of individual-level call
data (more than 100 million calls)
• Included are only calls / SMS between customers and non-customers
(„external degree“), as existing customers are no potential referral targets
Adding
covariates
Hinz, Skiera, Barrot & Becker
• Additional set of covariates to explain the referral likelihood such as:
 Socio-demographics (age, gender)
 Contract details (length of customer relationship, tariff plan, payment
method etc.)
 Service usage (monthly volume of voice minutes / SMS)
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
15
TWO-STAGE MODEL REVEALS DIFFERENT EFFECTS OF DEGREE
CENTRALITY FOR THE SELECTION AND REGRESSION COMPONENT
Study 3: Poisson-logit hurdle regression model (PLHR)
• Hubs are more
likely to participate
in viral campaign
• Hubs are more
likely to be
successful
referrers
• Higher degree
leads to more
referrals
• Higher degree
has no influence
on the number
of successful
referrals
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
16
HUBS ARE NOT MORE PERSUASIVE THAN AVERAGE CUSTOMERS
IN VIRAL MARKETING
Study 3: Determinants of conversion rates for active referrers
• Within the group of active
campaign participants,
degree centrality is no
significant effect on
conversion rate
• Viral marketing works at
awareness stage through
simple information transfer
• Hubs are no “better”
referrers – they just have a
higher reach
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
17
RESULTS CONFIRM THE POSITIVE CORRELATION BETWEEN DEGREE
CENTRALITY AND THE SUCCESS OF VIRAL MARKETING
Study 3: Relationship of conversion rates and degree centrality
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
18
REAL-LIFE APPLICATION OF VIRAL MARKETING CAMPAIGN USING
THE CUSTOMER BASE OF A MOBILE PHONE SERVICE PROVIDER
Study 3: Influence domain of referral campaign participant
7
7
• 20.8% of all first-generation
referrals became active
referrers themselves
6
• 5.8% did so multiple times
5
3 3
2
3
1
2
2
4
3
6
7
Referral
Generations
3
1
• Fringe actors have access
to new parts of network
4
3 3
• Viral referral chains with
maximum length of 29
generations and on average
.48 additional referrals
4
Customer Y
Customer X
(Origin)
Initial Campaign
Stimulus
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
19
CONDITIONAL ON SUCCESSFUL PARTICIPATION DEGREE CENTRALITY
HAS A NEGATIVE EFFECT ON INFLUENCE DOMAIN
Study 3: Determinants of influence domain (PLHR)
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
20
POSITIVE EFFECT OF DEGREE CENTRALITY DOMINATES IN THE
UNCONDITIONAL MODEL
Study 3: Determinants of unconditional influence domain
• Hubs are more important for viral success
• Results hold for both first-generation referrals as well as influence domains
• Results hold for all combinations of covariates (incl. usage, demographics etc.)
• Results hold for both simple OLS as well different count model formulations
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
21
HIGH-DEGREE STRATEGY CLEARLY OUTPERFORMS RANDOM
AND LOW-DEGREE STRATEGIES
Study 3: Relationship of conversion rates and degree centrality
• Hubs seem to participate, refer and successfully refer more often than average
• The average degree of the best and worst customer cohort is ca. 4:1
• A high-degree strategy would outperform a random selection by ca. 100%
• A high-degree strategy leads to conversion rates of nearly 10 times of the comparable
low-degree strategy
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
22
HIGH-DEGREE AND HIGH-BETWEENNESS STRATEGIES WORK BEST FOR
VIRAL MARKETING CAMPAIGNS – AT LEAST ON AWARENESS STAGE
Summary
• High-Degree and High-Betweenness seeding is comparable and outperforms random
seeding +39-52% (study 1), +60% (study 2), +100% (study 3)
• High-Degree and High-Betweenness outperforms Low-Degree by factor 7-8 (study 1),
factor 3 (study 2) and factor 8-9 (study 3)
• Influence of socio-metric measures beyond and above loyalty and revenue measures
• Hubs more likely to participate, do not fully use their reach potential, are not more
persuasive (due to social contagion working at awareness stage)
• Social networks possess valuable data that has not been used for targeting purposes
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
23
COMPARISON OF STUDY RESULTS
Hinz, Skiera, Barrot & Becker
Seeding Strategies for Viral Marketing | Journal of Marketing, Vol. 75 (November), 55-71
24