utilization of big data - sas · utilization of big data ... management doctrine •recognition of...

Private and confidential document of GT

Utilization of Big Data

H. Mehlika Ertaş

EVP Garanti Teknoloji

12/02/2016


12/02/2016 2


Big Data is ALL Data

Unstructured, Semi-Structured and Structured

There is always structure. But its not formally definedor anticipated.Social Media, RSS feeds, Videos, DOCs, PDFs, Graphics

Semi-Structured. Does not conform to DB tables, butstill contains tags or semantic elements.Emails, log files, machine generated content

What is the main difference in this data?

Volume, Velocity, Variety, Value

These Characteristics Challenge your

Existing Architecture

and your Thought Processes

12/02/2016 3


Analytics 2.0

Analytics 3.0

Analytics 1.0

Adapted from Tom Davenport

What are businesses trying to achieve using Big Data?

Data as a business cost Data as a business benefit & competitive weapon

Basic reporting

• Limited range of tabular data

• Batch oriented analysis

• Analysis bolted onto limited set of business processes

“Competing on Analytics”

• Extended analytics to larger and less structured datasets

• Fact and exception based management doctrine

• Recognition of Data Science

Platform for monetisation

• Deeper analysis on more data

• Faster test-do-learn

• wider business process coverage

• Analysts focus on discovery and driving business value

12/02/2016 4


Which users benefit?

• Makes their life a lot easier

• Supports them to spend more time on analytics instead of data wrangling

• Use discovery to contribute to Big Data Analytics on Hadoop

• Enabling collaboration with Data Scientists

DataScientists

Analysts and Power

Consumers

BI Consume

rs

• Some may use Discovery capabilities

• Majority will access the output of discovery from self service tools

12/02/2016 5


6

Tool Complexity• Early Hadoop tools only for experts

• Existing BI tools not designed for Hadoop

• Emerging solutions lack broad capabilities

Data Uncertainty• Not familiar and overwhelming

• Potential value not obvious

• Requires significant manipulation

Not Easy to Get Analytic Value at Fast Enough Pace

80% effort typically spent on evaluating and preparing data

Overly dependent on scarce and highly skilled resources

12/02/2016


7

New Approach

Host

BBDD33MM

Files

SocialMedia

OtherFeatures

Staging Level

CLTXN

DSIDS2DS3DS4

Enterprise Layer

Consumer Layer

Discovery Layer

DS1, DS2, DS3, DS4

• Governance Model (Roles & Responsibilities)• Organization and enablement/Training• Functional Architecture• Data Quality• Security and Audit

One of the most important functions is to guarantee the Governance and Use of the Information

Raw Data Models Business Model Adressing End-users

• Commercial Activities

• Financial Results• Data Quality• Liquidity• Sovency• Supervision

12/02/2016


Why are Banks Unable to Exploit Big Data

• Organizational silos

• Customer data is distributed across systems focused on specific functions such as CRM, portfolio management and loan servicing.

• No 360-degree view of the customer.

• Inflexible legacy systems that impede data integration

Challenges

12/02/2016 8


Challenges

• Lack of Strategic Focus: Big Data Viewed as Just Another ‘IT Project’

New technologies and processes to store, organize, and retrieve large volumes of structured and unstructured data

Traditional approaches hinge on a relational data model where relationships are created inside the system and then analyzed. With big data, it is difficult to establish formal relationships with the variety of unstructured data that comes through

Traditional data management projects view data from a static and/or historic perspective. Big data analytics is aimed to be used in a near real-time basis.

Most IT projects are driven by stability and scale, big data demands discovery, ability to mine existing and new data, and agility.

Taking a traditional IT-based approach, organizations limit the potential of big data.

Strategic Focus

12/02/2016 9


Challenges

• Privacy Concerns Limit the Adoption of Customer Data Analytics By uncovering hidden connections between seemingly unrelated pieces of data, big

data analytics could potentially reveal sensitive personal information.

62% of bankers are cautious in their use of big data due to privacy issues 1

Outsourcing of data analysis activities or distribution of customer data across departments for the generation of richer insights also amplifies security risks.

Incidents reinforce concerns about data privacy and discourage customers from sharing personal information in exchange for customized offers. «Big Brother is watching you»

1. Finextra Research, Clear2Pay, NGDATA, “Monetizing Payments: Exploiting Mobile Wallets and Big Data”, 2013

Privacy

12/02/2016 10


Challenges

• The first step: alter traditional mindsets.

• Pilots deliver quick and measurable results

• Concurrently lay the foundations to effectively scale-up big data initiatives

• Adopt a comprehensive approach, where pilots are backed by a well-defined data strategy and data governance model.

• Big data initiatives must be perceived differently from traditional IT programs. They must extend beyond the boundaries of the IT department and be embraced across functions as the core foundation for decision-making.

Implementation challenges

12/02/2016 11


A single intuitive and visual user interface, to...

find and explore big data to understand

its potential

find explore

quickly transform and enrich it to make it better

transform

unlock big data for anyone to discover

and share new value

discover share

Conclusion

12/02/2016 12


In order to graduate to higher levels of maturity in customer data analytics, organizationswill need to build the right organizational culture and back it up with the right skill sets and technological components

Last but not least ……

Roadmap to Building Analytics Maturity

12/02/2016 13

utilization of big data - sas · utilization of big data ... management doctrine •recognition of...

Documents