How optimize the usage of data to driving innovation and efficiency, focused on Brazilian banking market landscape, highlighting main trends, key challenges, leverage managed data lakes and samples of use cases.
1. Big Data & Analytics perspectives in Banking
Optimizing usage of data to driving innovation and efficiency in Brazilian market landscape
Gianpaolo Zampol | @gzampol
August 22nd, 2018
2. 2 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data main trends
Data is growing exponentially and became a new ‘natural resource’
Source: IDC DataAge 2015 Study, 2016
Note: 1 petabyte = 1M gigabytes, 1 zetabyte = 1M petabytes
Considerations
▪ ~10% are structured data, mainly
corporate date, stored in traditional
databases.
▪ Despite 90% of unstructured data
are documents, images, movies,
voice recordings, posts, tweets, etc,
most of this percentage are coming
from IoT devices.
▪ Internet giants (Google, Amazon,
Facebook, Apple) are expanding
beyond industry boundaries, with
the power of data (e.g. Apple Pay,
‘Facebook Bank’ in Ireland).
▪ Unlock new insights is imperative
for competitive advantage and
also business continuity.
3. 3 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Impacts and challenges from a ‘data flood’
Regulatory
Pressure
Regulator demand
growing every
year
Data
Management
Ability to manage
data has not kept
pace
Data
Growth
Amount of data
increasing
exponentially
Big Data
Capabilities
Advancements
lower costs and
technical barriers
Business
Pressure
Growth demands
still driving
investment
Working with Big Data implies combine business needs, value realization, regulatory aspects,
information governance and adequate and scalable IT infrastructure
Data
Storage
Cost to store data
decreased
exponentially,
supported by
cloud computing
Managing Big Data through a Data Lake
Source: Big Data best practices
4. 4 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Concept of Data Lake
Foundations of Big Data supply and consumption through a managed environment
Analytics at
Speed
AI
Applica-
tions
Risk and
Compliance
Enabled
New and
Deeper
Insights
Client and
User Expe-
rience
Managed
Data Lake
Cloud + On
Premise
▪ "In broad terms, data lakes are marketed as enterprise wide data
management platforms for analyzing disparate sources of data in
its native format.” (Gartner)
▪ "The idea is simple: instead of placing data in a purpose-built
data store, you move it into a data lake in its original format.
This eliminates the upfront costs of data ingestion, like
transformation. Once data is placed into the lake, it's available
for analysis by everyone in the organization.” (Gartner)
▪ A data lake is a large storage repository and processing engine.
They provide "massive storage for any kind of data, enormous
processing power and the ability to handle virtually limitless
concurrent tasks or jobs”. (Wikipedia)
Features of a
Managed
Data Lake
Definitions
▪ Controlled and managed environment at the heart of modern
Data Transformations.
▪ Enables operating model cost reduction across the data supply
chain: sourcing, modeling, provisioning, analytics.
▪ Speeds analytic insight.
▪ Supports regulatory requirements across data supply chain.
▪ Reduce/Reuse/Recycle/Innovate Model.
Source: Gartner IT Glossary, Wikipedia, researches on Big Data best practices
5. 5 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Fundamental principles on leverage Big Data
Performing an effective data driven transformation
Manage Big DataGather Big Data Use Big Data
1 2 3
More data is available –
both internal and
external. Technology has
made it easier and cost
effective to gather and
store for business
usage.
Big Data strategies
incorporate end to end
data lineage and
governance practices from
source to consumption.
Usage is being
transformed by new data
availability, new analytic
capabilities (e.g.
cognitive, streams) and
organizational priority.
Path to value is
accelerating through new
analytic capabilities and
applications
Technology has
lowered sourcing and
storage barriers
Big Data transformation
programs implement
controls across the data
supply chain
Source: Big Data best practices
6. 6 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data & Analytics Conceptual Architecture
Synthetic vision from data ingestion to transformation and delivery of insightful information
Real Time Analytics
Internal
Enterprise
Data
Analytics
‘At Rest’
Rapid
Ingestion
and
Integration
Managed Data
Lake
Visualization,
Applications and
Traditional
Reporting
Traditional
Repositories
External
Data
Reference &
Master Data
Source: Big Data best practices
7. 7 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data & Analytics Conceptual Architecture
Synthetic vision from data ingestion to transformation and delivery of insightful information
Real Time Analytics
Internal
Enterprise
Data
Analytics
‘At Rest’
Rapid
Ingestion
and
Integration
Managed Data
Lake
Visualization,
Applications and
Traditional
Reporting
Traditional
Repositories
External
Data
Reference &
Master Data
Source: Big Data best practices
1 - Gather Big Data
2 - Manage Big Data
3 - Use Big Data
8. 8 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
‘New normal’ in banking market landscape
Customers
Savvy, demanding customers
means banks must adapt to new
business models
Competition
New non-traditional
competition for
customers (e.g. retailers,
“GAFA”) migrate profit
pools out of financial
institutions
Complexity and cost
Cost and inefficiencies
hamper profitability,
concentrated on back office
operations, IT infrastructure
and legacy systems
Capital efficiency
Capital remains scarce due to
regulation so risk informed, capital
decisions are a key determinant
Regulation and
governance
Emerging regulations
demand granular and
frequent demonstration of
governance and control,
increasing cost of
compliance
Risk and security
Understand and mitigate
risks and reduce growing
cyber security threats well
remains a challenge
$
Source: Analysis based on The “New Normal” in Retail Banking, BCG, 2012.
Complex inter-related challenges drive competition for resources into a new business
environment
9. 9 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Customer-centric outcomes
Operational optimization
Risk / financial management
New business model
Employee collaboration
Other
functional
objectives
Customer-
centric
objectives49%
18%
15%
14%
4%
55%
4%
23%
15%
2% Banking & Financial
Markets
Global
Big Data & Analytics objectives in Banking
The majority of efforts are focused on improving customer interactions, followed by better risk
management and counter fraud
Source: The real world use of Big Data, IBM & University of Oxford, 2016; Febraban/Deloitte Research, 2017
47%
of banks are investing in Analytics
24%
started investing in Artificial
Intelligence/Cognitive Computing
96%
grow of customers using Mobile
Banking between 2015 and 2016
Highlights from Brazilian banks
10. 10 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
How Brazilian banks are respondingKey insights
▪ Brazilian Tier 1 banks created Chief Data Officer
organizations, establishing data governance to supply
“single sources of true data” to LoBs.
▪ Fostering and partnering with Fintechs.
▪ Bank data still sits in internal silos, limiting
competitiveness in the future.
▪ Turn data into insights given regulations as GDPR1,
appealing to ecosystems to access data.
Data as corporate asset,
holistic data governance
and monetize data
▪ Banks created Quod to supply analytics beyond credit
bureaus services and implement ‘positive credit scoring’.
▪ API enabled architectures grow in all banks, supporting
integration with ecosystems and low cost data transfers.
▪ Data expert start ups emerging, giving banks options
to outsource data analysis (e.g. Cardlytics, Experian).
▪ Other banks are sharing their customer data securely
through APIs.
Banks continue to
experiment Data as a
Service (DaaS) models
▪ 75% of local leading banks grew +15% revenue with
advanced analytics2.
▪ CDOs organizations expand their data scientists teams.
▪ RPA migrations projects +60% back office reductions.
▪ Predictive analysis with data science evolve add
value to customers, uncovering behavior patterns.
▪ Advances in RPA can automate and standardize inquiry
of data for precision, reducing error and operating costs.
Analytics advances
continue to help reduce
costs and provide better
customer engagement
Key challenges for Brazilian banks in Big Data & Analytics
▪ Requirements from FRTB/Basel IV, credit risk and better
capital allocation due to local macroeconomic.
▪ Improve AML analytics incorporating unstructured data.
▪ Brazil remain as major market attacked by cybercriminals.
▪ Risk mitigation, capital and regulatory requirements
stay in a high plateau, but continue to drive investments.
▪ Growing digital environment require high focus on
cyber security, fraud detection, KYC, AML.
Increase demands from
risk, security and
regulatory compliance
analytics
▪ Two speed IT in all banks with ‘digital’ departments, but
highly focused on UX, still coexists with old legacy systems.
▪ Open source codes heavily applied (e.g. R, Python),
creating security issues and architecture governance.
▪ Legacy back office and IT infrastructure remains the
largest challenge to transformation. Cloud-based and
API-enabled architectures make viable faster and
cheaper big data exploration and advanced analytics.
Infrastructure technology is
being modernized to
decrease costs and
improve agility
Source: Research and analysis upon Brazilian financial services market; McKinsey articles. 1General Data Protection Regulation (GDPR); 2High Stakes High Rewards, EY, 2017.
11. 11 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data & Analytics driving better client experience (1/3)
Using right data to get an actionable “view” of the client, contextually relevant at the up-to-the-
moment, to create target experiences
Profile/Descriptive data
▪ Products and Policies
▪ Goals
▪ Characteristics
▪ Demographics
▪ Self-declared info
Attitudinal data
▪ Opinions
▪ Feedback
▪ Preferences
▪ Aspirations
▪ Expressed / Inferred needs
Behavioral data
▪ Transactions
▪ Payments
▪ Inquiries
▪ Feature Usage
▪ Issues
Interaction data
▪ Browsing / Clickstream
▪ Contact center
▪ In-person dialogue
▪ E-Mail / chat transcripts
▪ Third-parties / Alliances
Accessible
Timely &
Kept Fresh
High-Quality &
Curated
Easily
Integrated
Real-Time +
Enrichment
Simple Sets /
Patterns
Complex
Analytics /
Models
Data Discovery,
Test & Learn
Data Staging
Source: Big Data and Analytics best practices
12. 12 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data & Analytics driving better client experience (2/3)
Pattern sample: Real-Time Offers
Profile/Descriptive data
▪ Products and Policies
▪ Goals
▪ Characteristics
▪ Demographics
▪ Self-declared info
Attitudinal data
▪ Opinions
▪ Feedback
▪ Preferences
▪ Aspirations
▪ Expressed / Inferred needs
Behavioral data
▪ Transactions
▪ Payments
▪ Inquiries
▪ Feature Usage
▪ Issues
Interaction data
▪ Browsing / Clickstream
▪ Contact center
▪ In-person dialogue
▪ E-Mail / chat transcripts
▪ Third-parties / Alliances
Accessible
Timely &
Kept Fresh
High-Quality &
Curated
Easily
Integrated
Real-Time +
Enrichment
Simple Sets /
Patterns
Complex
Analytics /
Models
Data Discovery,
Test & Learn
Data Staging
Source: Big Data and Analytics best practices
13. 13 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Big Data & Analytics driving better client experience (3/3)
Pattern sample: Financial Planning
Accessible
Timely &
Kept Fresh
High-Quality &
Curated
Easily
Integrated
Real-Time +
Enrichment
Simple Sets /
Patterns
Complex
Analytics /
Models
Data Discovery,
Test & Learn
Data Staging
Source: Big Data and Analytics best practices
Profile/Descriptive data
▪ Products and Policies
▪ Goals
▪ Characteristics
▪ Demographics
▪ Self-declared info
Attitudinal data
▪ Opinions
▪ Feedback
▪ Preferences
▪ Aspirations
▪ Expressed / Inferred needs
Behavioral data
▪ Transactions
▪ Payments
▪ Inquiries
▪ Feature Usage
▪ Issues
Interaction data
▪ Browsing / Clickstream
▪ Contact center
▪ In-person dialogue
▪ E-Mail / chat transcripts
▪ Third-parties / Alliances
14. 14 @gzampol | Big Data & Analytics perspectives in Banking | August 22, 2018
Thank you
Gianpaolo Zampol
Management & IT Consultant
Financial Services Sector
gianpaolozampol
@gzampol
gzampol