Provides a high-level view of how organizations can leverage Big Data in the digital space. Covers topics such as structured vs unstructured data, curating disparate data sources and exploiting the data correlation opportunities.
Univ. of AZ Global Racing Symposium 2015 - Digital Strategies
1. University of Arizona Global Racing Symposium
“DIGITAL MARKETING STRATEGIES”
Leveraging The “Back-End” Tools
Dec. 8, 2015
2. 2
Agenda
• What is “Big Data”?
- Creating a Frame of Reference
- The “4 V’s” – Drivers of Big Data
• Social Media Data Sources & Technology Landscape
• Data Types: Structured vs. Unstructured
• Consumer Insights & Today’s Systems
- Implementing Unstructured Data
- Data Visualization
Social Media Influencers
- The “Klout” Score
- Identifying Social Media Influencers
- Measuring Influencer Value
• Identifying Racing Data
• Key Takeaways
3. 3
What is Big Data?
Data sets with sizes beyond the ability of commonly used software tools
to capture, curate, manage, and process data within a reasonable amount of time.
4. 4
Data Frame of Reference
Megabyte (MB) - A good sized novel.
Gigabyte (GB) - 1600 books. About 300 MP3s.
Terrabyte (TB) – 1.6M books. 30 weeks worth of high-quality audio.
Petabyte (PB) – 160M books.
Exabyte (EB) – 3000 times the entire content of the Library of Congress.
Zettabyte (ZB) – 1 billion Terrabytes; Two hundred billion DVDs.
Yottabyte (YB) – 1 trillion Terrabytes.
9. 9
Structured vs. Unstructured - Defined
Data is classified as either Structured or Unstructured.
• Structured Data refers to information that resides in a
traditional row-column database—like Excel.
• Unstructured Data refers to information that doesn't
reside in a traditional row-column database.
NOTE: Experts estimate that 80 to 90 percent of the data in
any organization is unstructured.
10. 10
Examples of Structured Data
Structured Data usually refers to information that resides in
a traditional row-column database—like Excel. Here the
data is stored in fields in a database
11. 11
Examples of Unstructured Data
Unstructured Data files often include text and multimedia content.
Examples include email messages, word docs, videos, photos, audio
files, presentations, etc. This data doesn't fit neatly in a database.
12. 12
Consumer Insights
Today’s systems can structure the unstructured, then correlate key internal data
with the relevant social media universe – revealing new, actionable insights.
13. 13
Implementing Unstructured Data
Big Data Tools
Software like Hadoop or Oracle Endeca can process both unstructured and
structured data that are extremely large, very complex and changing rapidly.
Data Integration Tools
Combine data from disparate sources to be analyzed from a single application &
the capability to unify structured and unstructured data.
Search and Indexing Tools
These tools retrieve information from unstructured data files such as documents,
Web pages and photos
15. 15
Social Media Influencers
A Social Influencer is one who:
• Has the maximum followers
• Can influence others easily
• Creates and shares content regularly
Benefits of identifying Social Influencer:
• Leverage 3rd-Party credibility (others)
• Expand the message & the business
How to Identify the Influencers:
• Scores like “Klout” score are available to measure the influence of
someone in social media
16. 16
The Klout Score
• Klout is a digital service that uses social media analytics to rank its users
according to online social influence via the "Klout Score“
• Klout measures influence by using data points from various sites
– Twitter, Facebook, Google+, LinkedIn, Instagram etc., and Klout itself.
– Count, follower count, retweets, list memberships, influential follower
retweets unique mentions. Information is blended with data from other
social network followings & interactions to come up with the Klout Score.
17. 17
Identifying Social Media Influencers
When developing an influencer outreach campaign, make
sure you’ve got a good “READ” on the situation!
18. 18
Measuring Influencer Value
Several Key Performance Indicators (KPIs) can be combined into meaningful
ratios to help measure audience activity and engagement.
• Sentiment
• Re-tweets
• Forward to a friend
• Social media sharing
• Comments
• Like or rate something
• Reviews
• Contributors and active contributors
• Page views
• Unique visitors
• Traffic from social networking sites
• Time spent on site
• Response time
19. 19
Identifying Racing Data
• Wagering volumes, on-track/off-track/on-line
• Loyalty program information
• Attendance
• Non-wagering revenues (F&B, Parking, Merchandise)
• Social Media sites, racing blogs
• Odds
• Race quality/types
• Results
• Handicapping data
• Performance history
• Wager types: win, place, show, exotics
• Weather
• Seasonality
• Track condition
• Special events
20. 20
Key Takeaways
• Marketing Analytics is the primary opportunity-driver of business growth
going forward.
• “Dip your toe in the water now!”
• Unstructured data will reveal new value and actionable insights.
• Operators need to take the long-view and invest to grow the sport.
22. 22
Social Media – Behavioural Segments
Depending on the Goal, one can select the specific segment of users.
Social Analytics Tools help to identify the following from Social Data:
• Influencers
• Recommenders
• Detractors
The above combined with customer and transaction data would give
necessary insights into customer behavior.
The social users can also be segmented by demographics, geographies,
which would provide information that might not be captured in CRM.
23. 23
Social Communication Metrics
• Alerts (register and response rates / by channel / post
click activity)
• Bookmarks
• Comments
• Downloads
• Email subscriptions
• Fans (become a fan of something / someone)
• Favorites (add an item to favorites)
• Feedback (via the site)
• Followers (follow something / someone)
• Forward to a friend
• Groups (create / join / total number of groups / group
activity)
• Install widget (on a blog page, Facebook, etc)
• Invite / Refer (a friend)
• Key page activity (post-activity)
• Love / Like this (a simpler form of rating something)
• Messaging (onsite)
• Personalization (pages, display, theme)
• Posts
• Profile (e.g. update avatar, bio, links, email,
customization, etc)
• Print page
• Ratings Registered users (new / total / active / dormant
/ churn)
• Report spam / abuse Reviews Settings Social media
sharing / participation (activity on key social media
sites, e.g. Facebook, Twitter, Digg, etc)
• Tagging (user-generated metadata)
• Testimonials
• Time spent on key pages
• Time spent on site (by source / by entry page)
• Total contributors (and % active contributors)
• Uploads (add an item, e.g. articles, links, images,
videos)
• Views (videos, ads, rich images)
• Widgets (number of new widgets users / embedded
widgets)
• Wish lists (save an item to wish list)
24. 24
Sentiment Analysis
Sentiment Analysis – uses natural language processing, text
analysis and computational linguistics to identify and extract
subjective information in source materials. It aims to determine the
attitude of a writer with respect to some topic or the overall
contextual polarity of a document.
25. 25
Sentiment KPI’s
• Total Conversation about a topic, activity or initiative
• Number of Positive Conversations about topic, activity or initiative
• Number of Negative Conversations about a topic, activity or initiative
• Ratio or Percentage of Negative Conversations/Total Conversation
• Ratio or Percentage of Positive Conversations/Total Conversations
26. 26
Data Source Gnip DataSift Description
Twitter √ √ Microblog site
Facebook √ √ Social networking site
Sina Weibo √ Chinese microblog site
WordPress √ √ Open-source blog tool & CMS
Intense Debate √ √ Blog comments & host
Tumblr √ √ Multimedia microblog
Foursquare √ Social network for mobile
Yammer √ Enterprise social network
LexisNexis √ Legal & public records database
Google+ √ √ Social network & authorship tool
YouTube √ √ Video sharing website
Bitly √ √ URL shortening service
Instagram √ √ Photo & video sharing
NewsCred √ Content & syndication platform
Reddit √ √ Entertainment & news website
Wikipedia √ Collaboratively edited internet encyclopedia
Daily Motion √ √ French video sharing website
Topix √ Tokyo stock price index
IMDb √ Internet movie database
Disqus √ Blog comment hosting
Estimize √ Community of stock analysts & traders
Sitrion √ Social & collaboration software
Flickr √ Image & video hosting
MetaCafe √ Video sharing – short form
Panoramio √ Geolocation photosharing
Photobucket √ Image & video hosting
Plurk √ Social network microblog
Stackoverflow √ Q&A site for computer programming
Vimeo √ Video sharing website
VK √ Russian social network
Stocktwits √ Social media sharing for investors & traders
SocialMediaDataAggregators
Evaluating, monitoring and field-adjusting (in real-time) is essential to exploiting the influencer—and his value/credibility to the broader target
Access to fact-driven predictive insights in real-time, and driven by business needs, is key to ensuring an optimal balance between preventative and corrective actions.
Predictive Analytics enables organizations to
monitor your environment by including a wide variety of data across multiple sources
detect suspicious behavior to identify threats, information breaches, crime fraud
And then control outcomes to deliver the best response to reduce exposure or loss and maximize the impact of any action taken
Predictive risk and threat management is about using analytics proactively to reduce exposure and minimize negative impact
Behavior analytics require the ingestion of all sources of data both structured and unstructured. These are just a sampling of the sources that will uncover the key nuggets of actionable insights..
Access to fact-driven predictive insights in real-time, and driven by business needs, is key to ensuring an optimal balance between preventative and corrective actions.
Predictive Analytics enables organizations to
monitor your environment by including a wide variety of data across multiple sources
detect suspicious behavior to identify threats, information breaches, crime fraud
And then control outcomes to deliver the best response to reduce exposure or loss and maximize the impact of any action taken
Predictive risk and threat management is about using analytics proactively to reduce exposure and minimize negative impact
