From Strangers to Acquaintances - Multidimensional Customer Profiling Presented by MeaningCloud at the Sentiment Analysis Symposium. July 15, 2015. MeaningCloud turns unstructured content into actionable data http://www.meaningcloud.com

1. From Strangers to Acquaintances
Multidimensional Customer Profiling
MeaningCloud
July 15, 2015
1

2. Agenda
 Why basic Sentiment Analysis is not enough
 Using multichannel, unstructured content to profile customers:
Scenarios and benefits
 How to implement this using MeaningCloud today!
 Conclusions
2

3. Basic, aggregated detection of mentions and
sentiment is no longer enough
 Technology allows to process social
media and other unsolicited,
unstructured customer feedback
sources and detect mentions,
polarity
 This enables to understand the
general view, but…
 How is each specific customer like?
Who says what?
Why do they say that?
What opportunities are out there?
3

4. What if you could use all customer feedback
and…
…Turn Strangers into Acquaintances!
4

5. So many useful data that you can extract NOW…
Turn your customer expressions into personal data
Demographics
• Age
• Gender
• Location
• Education
• Occupation
• Family…
Psychographics
• Opinions
• Attitudes
• Affinities
• Lifestyle…
Customer
journey
• Awareness
• Consideration
• Decision
• Purchase
• Loyalty
• Advocacy…
Online surveysSocial conversations Contact center interactions
5

6. How you can leverage all that data
Male
25-35
Pragmatic
Brand ++
Female
45-65
Skeptic
Brand -
Female
45-65
Conservative
Brand +
Male
25-35
Innovator
Brand ++
Male
25-35
Innovator
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand +
Male
45-65
Skeptic
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand -
Female
35-45
Pragmatic
Brand -
6

7. How you can leverage all that data
Male
25-35
Pragmatic
Brand ++
Female
45-65
Skeptic
Brand -
Female
45-65
Conservative
Brand +
Male
25-35
Innovator
Brand ++
Individual
 Profiles
 Scores
 Signals: purchase, churn…
Male
25-35
Innovator
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand +
Male
45-65
Skeptic
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand -
Female
35-45
Pragmatic
Brand -
7

8. How you can leverage all that data
Male
25-35
Pragmatic
Brand ++
Female
45-65
Skeptic
Brand -
Female
45-65
Conservative
Brand +
Male
25-35
Innovator
Brand ++
Aggregate
 Segments
 Personas (archetypes)
 Perception analysis
 Competitive analysis
 Product ideas & issues
Individual
 Profiles
 Scores
 Signals: purchase, churn…
Male
25-35
Innovator
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand +
Male
45-65
Skeptic
Brand ++
Female
45-65
Conservative
Brand -
Female
35-45
Pragmatic
Brand -
Female
35-45
Pragmatic
Brand -
8

9. Why YOU should be doing this: actionable
insights
 1:1 Engagement
Personalized messages and experiences throughout each
consumer’s journey
E.g., customer complains on Twitter  Contact him directly
 Product development
Need identification, improvement ideas
E.g., lots of customers identify issue in present product
 Market targeting and competitive positioning
Who are my most promising target customers and what
should I tell them?
E.g., brand personality, messaging and creative
 Campaign planning and management
How can I reach them?
E.g., media selection and planning
9

10. Customer Profiling with MeaningCloud
10

11. MeaningCloud: “Meaning as a Service”
Register and use it FREE at http://www.meaningcloud.com
11

12. Text analytics
Extract meaning and actionable insights from unstructured content
Automatization of costly manual activities
MeaningCloud provides this in a convenient, web service-based offering
Opinions
Facts
Concepts
Organizations
People
Semantic
Analysis
Relationships
Themes
12

13. APIs services of MeaningCloud Sentiment analysis
 Global
 Aspect-based
Classification
 Standard models
Topic extraction
 Entities
 Concepts
 Dates
 Addresses
 Economic quantities
 Time expressions
 …
https://www.meaningcloud.com/demos/media-analysis/13

14. Topic Extraction
Disambiguate appearances of brands, companies, organizations,
people… and many more
 Contextual disambiguation
 Apple = company (not fruit)
 Coreference
 Based on standard ontology
 Extendable/customizable dictionaries
In a filing with the SEC today, Apple
revealed that CEO Tim Cook has
donated the equivalent to
approximately $6.5 million in Apple
stock shares to charity this week.
Since becoming CEO in 2011, Cook
has promoted charity as a key part
of Apple’s mission. Upon taking
over, Cook initiated an employee
charity program. Apple has also
expanded its offerings for
employees to help their
communities.
Topic
detected
Semantic information
Tim Cook Person, Timothy Donald Cook, Executive
Apple Inc.
Apple Company, Apple Inc., Technology, USA
SEC Organization, Securities and Exchange
Comission, Government, USA
$6.5 million Monetary amount, USD, 6.5 million
charity Concept, charity
14

15. Text Classification (featuring standard models, e.g. IAB)
Mix machine learning and rules to accurately classify text according
to predefined categories
The World Cup is the best way to see the potential
football can have for your inbound travel, economic
success and positive public image:
The 2006 World Cup in Germany was a prime
example of this power with: $200+ per day average
tourist spending, 50,000 new jobs created, 18
million people at Fan-Fests, total worldwide TV
viewership at 30 billion and 4.2 billion official
webpage views. In a survey, 90% of foreigners who
visited the World Cup said they felt welcome there
and would recommend Germany as a holiday
destination. "The World Cup marks an enormous
gain in Germany's image, even if it's difficult to put
an economic figure on this change in image, the
economy as a whole will certainly benefit from it."
the German economics minister, Michael Glos, said.
Categories Relevance
Sports – World soccer 0.7
Travel - Europe 0.2
Arts & Entertainment - Television 0.3
 Hybrid technology
 Machine learning and/or rules
 Features standard classification models
 IPTC (news), IAB (advertising, public beta),
etc.
 Customizable classification models
IAB (English)
15

16. Sentiment Analysis
Assign multilevel polarity to entities and other aspects, discriminate
facts from opinions and detect irony
IBM stock fell another 1.51%, while
their cloud business revenue rose
60 percent in 2014.
Aspect Sentiment
IBM - stock N
IBM - revenue P+
Global NEU, DISAGREEMENT, OBJECTIVE, NON IRONIC
Aspect Sentiment
Excelsior Hotel - landscapes P+
Excelsior Hotel - rooms N-
Global NEU, DISAGREEMENT,
SUBJECTIVE, NON IRONIC
 5-level polarity (plus absence of polarity) scoring
 Aspect-based analysis
 Objective (fact) / subjective (opinion) discrimination
 Irony detection (beta)
 Customizable sentiment models (in beta, contact us)
Excelsior Hotel has the most
amazing landscapes I've ever seen,
but the rooms are disgusting.
16

17. Customer Profiling
Use the profile and content generated by the user to infer his
demographic attributes
20% of companies say process
digitization yields actionable #analytics
Is your IT team talking SMAC (#social,
#mobile, #analytics, & #cloud)?
Five Rules of Modern Icon Design
http://bit.ly/1y3B6i6
What Twitter Can Be.
http://wp.me/p2Gq8C-6E Just if they'd
play nice with the ecosystem ...
#socialtv #recommendation
What your name says about your age,
where you live, your politics & your job
http://wapo.st/1RkqDcA
Londoner, hooked on data science, NLP
and REST.
Social posts
Social profile
Atribute Value
Person/Organization Person
Gender Male
Age 25-35
Location London
Occupation Engineer
 Person /organization
 Gender
 Age
 Location
 Occupation
 …
Now in private beta
17

18. Customization tools
 Create your own dictionaries, classification
models, and sentiment analysis (beta)
 Graphical user interface - no programming!
 Improve precision & recall
Learn more about customization in this webinar18

19. Add-in for Excel
 Totally integrated in Excel experience
 Easy to use - No programming!
 The most convenient way to evaluate, prototype and use MeaningCloud
19

20. Democratizing the extraction of meaning
High quality semantic analysis
 Optimized technology mix
 Continuously updates semantic
resources
 High-level APIs, e.g., Customer
Profiling
 Customizable to customer
domain: models, dictionaries,
sentiment
Affordable, no risks
 Mature, tested technology
 Test and use for FREE
(40,000 requests per
month)
 Pay per use
 No commitment or
permanence
 Commercial plans beginning
at $99 /mo
For developers and non
technical users
 Add-in for Excel
 Standard web services APIs
 Plug-ins and SDKs for
diverse environments and
languages
 Plug-and-play approach
OpinionesTemas
Hechos
Conceptos
Organizaciones
Personas
Relaciones
20

21. A platform for leveraging unstructured content in Voice of the Customer /
Customer Experience contexts
What you should expect from MeaningCloud in
the coming months
GA
Mention detection
& theme
classification
Granular
sentiment analysis
Corporate
reputation (ES)
Customization
tools
Q3 2015
Demographic
profiling
Industry- and app-
specific
dictionaries and
models, e.g., IAB,
banking
Q4 2015
Trend emergence
and analysis
Q1 2016
Customer journey
stage &
actionable signals
Q2 2016
Perception maps
& brand
personality
Competitive
analysis
21

22. Customer case: SocialBro
Customer: leader in Twitter community analysis and marketing tools
Problem: process massive amounts of tweets (1,000 tweets per sec peaks)
Our Solution: based on user’s social comments and profiles, we inferred demographic profile of community members
and analyzed aspect-based sentiment toward specific brands
Insights / results: data-based, actionable segments and better marketing campaigns targeting
Social
profile
Social
posts
Social
posts
John Smith
Person
Male
35-45 yr.
London
Doctor Mary Doe
Person
Female
45-55 yr.
Berlin
PilotSocial
profile
Male
35-45 year
Big cities
Business owner
Positive brand
attitude
Female
45-55 year
Mid-sized cities
Professional
Negative brand attitude
22

23. Conclusions
 Unstructured content in social media and other channels offers
untapped possibilities to understand customers
 Text analytics technology can turns this content into actionable
insights: profiles, signals
 MeaningCloud is the easiest, most customizable and most
affordable way to do it
Interested? See our demo tomorrow
Workshop Track, 9:20 am
23

24. Thank you for your attention!
Questions, suggestions...
Find us in or booth, see our demo tomorrow or contact us directly
Antonio Matarranz
amatarranz@meaningcloud.com
Jarred McGinnis
jarred@dmeaningcloud.com
MeaningCloud LLC
1120 Broadway, Ste. 805
New York, NY 10010
USA
Phone: +1 (646) 403-3104
http://www.meaningcloud.com
24