2. • Bruce Swann
• Manager, CI / Integrated Marketing, SAS
• Scott Briggs
• Principal Solutions Architect, Customer Intelligence, SAS
• Suneel Grover
• Sr. Solutions Architect, Integrated Marketing Analytics, SAS
• Adjunct Professor, The George Washington University (GWU)
4. Agenda
I. High-performance marketing optimization
II. Social media analytics and real-time actions
III. Social network analytics and community influence
20. • Financial Services
– Cross-sell and up-sell in retail banking: savings accounts, home
equity loans, credit cards, lines of credit, etc.
– Insurance policy offers
– Deciding credit line increases
– Deciding what APR to offer on balance transfer offers
• Telecom
– Complex cell phone or calling plan offers
– Bundled service offers
– Cross channel offers with different costs of execution
• Others
– Loyalty offers (Hotels, Casinos)
– Personalized coupons (Retail)
Use Cases
25. The Opportunity
1. Know how many visits, leads, and customers each individual
social channel is generating…
2. Improve the customer experience…
3. Leverage social networks and communities…
27. Listening
Data
Mining
Correlation &
Forecasting
Text Mining Natural Language Processing
Taxonomies
Influence
& Engagement
Sentiment
Analysis
Categorization
Portals CRM
Collect
Clean
Integrate
Organize
Accessible by All
Analytics, Classify, Seg
ment, Sentiment, Natur
al Language Processing
iPad
apps
Dataset
Export
Web
Data
Survey
Data
Call
Logs
Text Analytics
28. Social Media Analytics
Online
Conversations
Brand &
Market
Tracking
PR &
Reputation
Tracking
Customer
Feedback
Mgmt
Online
Media
Analysis
•WHAT are consumers saying about
your brand? About the competition?
•WHO is creating content about
your brand…Journalists? Bloggers?
Forum members?
•WHO among these authors is a
threat to reputation? An
opportunity for advocacy?
•WHERE are consumers talking?
• Is volume trending up or down?
•WHICH sites matter most?
• WHICHsites are more positive?
Negative?
• WHATaspects of your business drive
satisfaction and loyalty?
• WHAT questions and unmet needs
emerge?
• HOW do perceptions differ across
the various channels through which
customers give you feedback?
29. The Business Need
How can I ensure it’s
accurate and relevant to my
business?
How do I cut out the noise
and get to the true insights
and action?
How can I customize it to
understand my business,
my brands and
competitors?
How does social media fit
with my other business
intelligence?
How can social data
augment what I already
know?
How can it help me get a
clearer picture of my
business as it changes?
How do I use social media to
drive my business forward?
Where does it fit within my
business strategy?
Where can I focus for the
best returns?
How can I use it to get a
competitive edge?
How do I monetize it?
34. Engage
0%
10%
20%
30%
0 1 2 3 4 5 6 7+
PercentofTotalFile
# of Social Networks
Social Participation
Seg 1
Seg 2
Social Network
Engages
Social | Email
Influencers
Create Friend-centric
Message
1 2 3
0%
10%
20%
30%
40%
0 1 2-3 4-10 11-19 20+
PercentofTotalFile
# of Friends
Social Reach
Seg 1
Seg 2
32%
28%
40%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Inactive Opener Clicker
Email Activity Segments - 20+ Friends
35. Engage
Enable customer care agents monitoring social media to communicate with
consumers in order to…
Address the consumer’s
service issues and
questions
Mitigate or respond to
negative comments or
threats
Reinforce a customer’s
positive sentiment
• Broadcast comments to
other customers
• Reward with offers
Facilitate customer
consideration process
• e.g. consumer comparing
hotel options for a
vacation
38. Social Network Analysis
‘A social network is a social structure made up of individuals, which are
connected by one or more specific types of interdependency, such as
friendship, kinship, common interest, financial exchange‘ etc.
39. Social Network Analysis
• Attract outside influencers in the expectation that their community will
follow.
• Incent inside influencers to pull in off-network followers.
Acquisition
• Target cross/up sell offers to inside influencers.
• Cross promote products within communities.
Cross/Up Sell
• Reduce churn by holding on to influencers.
• Decrease virality effect of followers.
Retention
40. Social Network Analysis
Data feeds: ETL
to data environment
1 Data Management:
Cleanse, parse, categorize, and
standardize social media data around
“Vail Customers”
2
HUB
Epic Mix
Social Media Chatter
(e.g. Twitter, Facebook)
3
4
Executive Insights: Explore results in
Visual Analytics
5
41. Social Network Analysis
Social Media
Analytics Data
Batch ETL
HUB
Outbound
Inbound
Customer DB (or EDW)
Non-matched/Possible matched
Social/Customer data
Mastered data with
Social Info Appended
Matched customer Info
(is this possible?)
43. Community Influence
Sarah
Casino
Visitor
Sarah’s
Social Network
• Semi-frequent
Visitor
• High Value
• Large Friend Network
• Content Creator and
Contributor
• Active Social Elements
• Encourages Sharing
• Friend-centric
Targeted Email Campaign
to Sarah
Sarah
Engages
• Engages with Email
• Forwards to Friends
• Posts Content
• Network Engages
and Converts
• Individuals begin to
contribute content
(blogs, reviews, etc.)
Sarah’s Network
Engages
44. Community Influence
Virality is the effect of influencers on followers.
In particular, what is the increased likelihood of churn within a
community once an influencer churns.
Virality churn lift is the churn rate delta of followers.
Influencer
churn
Follower
churn
45. Social Profile
Social Audience
Profile
•Number of Friends
•Social Membership
•Number of Profiles
•Last Activity Date
•Social Tenure
Map your
constituent
audience to social
profiles
Use email address as
match key
Match
constituent to
social behavior
Access publicly
available social
data
Build Social
Audience
Profile
Assess social
engagement
levels
46. Social Profile
0%
10%
20%
30%
0 1 2 3 4 5 6 7+
PercentofTotalFile
# of Social Networks
Social Participation
Seg 1
Seg 2
Social Network
Engages
Social | Email
Engagers
Create Friend-centric
Message
1 2 3
0%
10%
20%
30%
40%
0 1 2-3 4-10 11-19 20+
PercentofTotalFile
# of Friends
Social Reach
Seg 1
Seg 2
32%
28%
40%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Inactive Opener Clicker
Email Activity Segments - 20+ Friends
Mobile PurchaseWebsiteEmail
Search SocialDisplay Ad
M
D
PE
S
W
O
47. Case Study
• Major US-based wireless carrier with 30+ million customers.
• 89 million individuals within the overall population.
• Average community size is roughly 18.
• 5% of all subscribers are influencers.
• Followers’ churn rate increased by 25% when influencers
churned.
• 30% model lift when SNA was used.
• Campaign take rate among followers doubledwhen
influencers took.