Simplify Your Analytics Strategy
Narendra Mulani
While the interests in
analytics and resulting
benefits are increasing by the
day, some businesses are
challenged by the complexity
and confusion that analytics
can generate.
Discovering real business
opportunities and achieving
desired outcomes can be
elusive.
Following are steps that we have
seen work in a number of
companies to simplify their
analytics strategy and generate
insight that leads to real outcomes
Accelerate the data
•Fast data = fast insight = fast
outcomes.
•Liberate and accelerate data by creating a
creating a data supply chain built on a
hybrid technology environment — a data
data service platform combined with
emerging big data technologies.
Next-Gen Business
Intelligence (BI) and data
visualization.
•At its core, next-gen business
intelligence is bringing data and
analytics to life to help companies
improve and optimize their decision-
making and organizational
performance.
Data discovery
•Data discovery can take place
alongside outcome-specific data
projects. Through the use of data
discovery techniques, companies
can test and play with their data to
uncover data patterns that aren’t
clearly evident.
Analytics applications
• Applications can simplify
advanced analytics as
they put the power of
analytics easily and
elegantly into the hands
of the business user to
make data-driven
business decisions.
Machine
learning and
cognitive
computing
• Recognize that each path to data insight is unique.
The path to insight doesn’t come in one single
form. There are many different elements in play,
and they are always changing — business goals,
technologies, data types, data sources, and then
some are in a state of flux.
• Another main component of a company’s analytics
journey depends on the company’s culture itself.
• No matter what combination of culture and
technology exists for a business, each path to
analytics insight should be individually
paved with an outcome-driven mindset.
• To do this, companies can take two approaches
depending on the nature of the business problem.
•First, for a known problem with a known
solution, such as customer segmentation
and propensity modeling for targeted
marketing campaigns, the company could
take a hypothesis-based approach by
starting with the outcome (e.g. cross-
sell/up-sell to existing customers), pilot
and test the solution with a control group
and then scale broadly across the customer
base.
•Second, for a known problem area, but
with an unknown solution, the company
could take a discovery-based approach to
look for patterns in the data to find
interesting correlations that may be
predictive—for instance, a bank found that
the speed at which fields were filled out on
its online forms was highly correlated with
fraudulent behavior.
When determining
which problem to
address, companies
should first focus on the
one that can offer the
highest value, then it
can choose a
hypothesis-based or
discovery-based
approach based on the
degree of institutional
knowledge it has to
• Companies get stuck trying to analyze all that’s
possible.
• Following are steps that we have worked in a number of
companies to simplify their analytics strategy:
1. Accelerate the data
2. Next-Gen Business Intelligence (BI) and data
visualization.
3. Data discovery
4. Analytics applications
5. Machine learning and cognitive computing
• Companies can take two approaches
1. For a known problem with a known solution
2. For a known problem area, but with an unknown
solution
THANKYOU
Simplify Your Analytics Strategy

Simplify Your Analytics Strategy

  • 1.
    Simplify Your AnalyticsStrategy Narendra Mulani
  • 3.
    While the interestsin analytics and resulting benefits are increasing by the day, some businesses are challenged by the complexity and confusion that analytics can generate.
  • 4.
    Discovering real business opportunitiesand achieving desired outcomes can be elusive.
  • 5.
    Following are stepsthat we have seen work in a number of companies to simplify their analytics strategy and generate insight that leads to real outcomes
  • 6.
    Accelerate the data •Fastdata = fast insight = fast outcomes. •Liberate and accelerate data by creating a creating a data supply chain built on a hybrid technology environment — a data data service platform combined with emerging big data technologies.
  • 7.
    Next-Gen Business Intelligence (BI)and data visualization. •At its core, next-gen business intelligence is bringing data and analytics to life to help companies improve and optimize their decision- making and organizational performance.
  • 8.
    Data discovery •Data discoverycan take place alongside outcome-specific data projects. Through the use of data discovery techniques, companies can test and play with their data to uncover data patterns that aren’t clearly evident.
  • 9.
    Analytics applications • Applicationscan simplify advanced analytics as they put the power of analytics easily and elegantly into the hands of the business user to make data-driven business decisions.
  • 10.
  • 12.
    • Recognize thateach path to data insight is unique. The path to insight doesn’t come in one single form. There are many different elements in play, and they are always changing — business goals, technologies, data types, data sources, and then some are in a state of flux. • Another main component of a company’s analytics journey depends on the company’s culture itself. • No matter what combination of culture and technology exists for a business, each path to analytics insight should be individually paved with an outcome-driven mindset. • To do this, companies can take two approaches depending on the nature of the business problem.
  • 13.
    •First, for aknown problem with a known solution, such as customer segmentation and propensity modeling for targeted marketing campaigns, the company could take a hypothesis-based approach by starting with the outcome (e.g. cross- sell/up-sell to existing customers), pilot and test the solution with a control group and then scale broadly across the customer base.
  • 14.
    •Second, for aknown problem area, but with an unknown solution, the company could take a discovery-based approach to look for patterns in the data to find interesting correlations that may be predictive—for instance, a bank found that the speed at which fields were filled out on its online forms was highly correlated with fraudulent behavior.
  • 15.
    When determining which problemto address, companies should first focus on the one that can offer the highest value, then it can choose a hypothesis-based or discovery-based approach based on the degree of institutional knowledge it has to
  • 16.
    • Companies getstuck trying to analyze all that’s possible. • Following are steps that we have worked in a number of companies to simplify their analytics strategy: 1. Accelerate the data 2. Next-Gen Business Intelligence (BI) and data visualization. 3. Data discovery 4. Analytics applications 5. Machine learning and cognitive computing • Companies can take two approaches 1. For a known problem with a known solution 2. For a known problem area, but with an unknown solution
  • 17.