5. Companies are seeing returns from
big data
Uses of Big Data
90%
80%
70%
60%
50%
40%
30%
20% Uses of Big Data
10%
0%
Improved Improved Support of Not
Business Current New Leveraged
Decisions Revenue Revenue for Revenue
Streams Streams Growth
Source: Avanade Inc. 2012 Big Data Survey
6. The Heart of a Data Driven
Organization
Data drives decisions and are the key to all decisions
made within the organization
People who think make decisions, not data!
A data driven organization can not truly use data on its
own, it takes people with the right skills and expertise in
knowing how to use the data, to truly be data driven.
Evidence based decisions + Reasoned Arguments is how
an organization becomes data driven.
“An organization’s data is found in its computer systems, but a company’s
intelligence is found its biological and social systems” --- Valdis Krebs,
researcher
7. Obtaining Data as a competitive
Advantage
Best in class data driven companies take 12 days on average
to integrate new data sources into their analytical systems;
industry avg companies take 60 days, laggards 143 days.
Best-in-class companies can pursue new market
opportunities faster
Can take advantages quickly, newly emerging business
opportunities
Can bring high-value services and products to market
faster
Be proactive and create more information based insights
Source: Aberdeen Group: Data Management for BI: Fueling the analytical engine with high-octane information
8. To Put it Another Way
Computational = Subconscious
Strategic = Conscious
9. How to use Big Data to create a
data driven culture
Data • Data is the foundation
• Insights improve
Insights understanding
• Actions, create
Actions new
experiences
11. Solving Problems with Big Data
Hadoop-able Problems
Complex data and lots of it
Multiple data sources and highly unstructured
Benefits of Analyzing with Hadoop
Low cost
Greater flexibility
Ability to do previously impractical analysis
12. Where to Start with Big Data
Problem Solving
Text Mining (unstructured Modeling true risk (new
data that was previously
not available) data means better
Pattern Recognition (find forecasts)
previously unknown Recommendation
patterns in the data) engines (engage
Collaborative filtering customers)
(power of the crowd)
POS analysis (real-time
Sentiment analysis
(Beyond text mining) analysis)
Prediction models (new Data “sandbox” (new
data means new insights methods for testing new
about what may come) products concepts)
13. Data Driven Decision Making
Framework – Insights to Action
Source: Social Business By Design Dion Hinchcliffe
14. Signal Types
Signals have attributes depending on their representation in time or frequency domain
can also be categorized into multiple classes
All signal types have certain qualities that describe how quickly signals can be
generated (frequency), how often the signals vary (rate of change), whether they are
forward looking (quality), and how responsive they are to stimulus (sensitivity)
Rate of Change Quality Sensitivity Frequency
(Slow or Fast) (Predictive or (Sensitive or Insensitive) (High or Low)
Descriptive)
Sentiment Behavior Event/Alert
Expressed as Correlation
These signals A discrete
positive, Clusters Measures the
identify signal
neutral, or Signals based correlation of
persistent generated when
negative, the on an entity’s entities against
trends or certain
prevailing cohort their prescribed
patterns in threshold
attitude characteristics attributes over
behavior over conditions are
towards and time
time met
entity
15. Finding Signals in Unstructured Data
High quality signals are necessary to distill the relationship among all the of the
Entities across all records (including their time dimension) involving those Entities to
turn Big Data into Small Data and capture underlying patterns to create useful inputs
to be processed by a machine learning algorithm.
For each dimension, develop meta-
data, ontology, statistical measures,
Clickstreams and models
Social Timing/
Context Recency
Articles Content Source Measure the
Create Measure
symbol Derive the freshness of
Blogs sentiment sources’ the data and
language to strength:
describe and meaning of the insight
Tweets from originality,
environment importance,
s in which tracking
tools to quality,
the data quantity,
resides syntactic and
semantics influence
analysis
16. New Solutions Must Aid Human
Insight
Big Data + Amplified Human Intelligence = Better Decisions
Last Decade Next 5 Years
- Structured Data - Any data, from
- Conclusive Dashboards anywhere
- Small scale / sampling - Intuitive
exploration
A data architect built a - Making sense of it
view to reach a specific at scale
conclusion Business users easily
find, explore,
visualize and navigate
insights
19. Data Driven Organizations Always
Question the Data
• How do we integrate the right data
• What business opportunity/problem are together?
we trying to solve?
• How do we manage the quality of
• What questions do we need to answer to the data?
solve the problem?
• What data does this relate to
• What data do we need to answer the (master data)?
questions?
• Do we have all the data about this
• What data do we have? (person, event, thing, etc.)?
• How can data help differentiate us in the • What are the permissible purposes
market? of the data? (compliance,
regulatory environment)
• What data is IP for us? Revenue
generating for us? • Who is allowed to access the data?
Use this data?
20. Data Driven Spider Graph
Data
Science
Customer
Big Data IT
Care
Data
Driven Business
Logistic
Customer Strategists
Experience
Business
Traditional
Intelligence
IT
Tools
Social
21. Always Remember: Data, Insights,
Actions
• Listen to the data streams
Listen
• Share the data with the rest of the organization
Share
• Engage to the data to find the insights
Engage
• Innovate new ideas from the insights gained from the data
Innovate
• Perform insightful actions from the data to create better customer
Perform experiences