This document discusses challenges facing the financial services industry and opportunities for transformation through DevOps. It notes that while three-quarters of institutions are investing in data analytics capabilities, only 17% believe they are fully prepared for regulatory compliance responsibilities. DevOps is presented as a way to optimize work flow and release value faster through continuous delivery pipelines. Success requires solving for enterprise complexity by creating conditions for the system to fix itself. Pipelines are described as the industry's "factories" and a bottom-up approach is advocated, starting with individual applications and teams before expanding transformation.

1. DE 6 A 8 DE JUNHO
TRANSAMERICA EXPO CENTER
SÃO PAULO - SP
Transformation in
Insurance Industry through
DevOps @ Scale

2. State of the FS industry
2 The Financial Brand – PwC: Retail Banking 2020: Evolution or Revolution? – Feb 1 2016
1 Deloitte and Compliance Week: In Focus: 2015 Compliance Trends Surveys
Bank CIOs that plan on increasing IT spending
Security Data Analytics Mobile
Banking
Compliance
75% 58% 55% 49%
CCO’s are not confident in the
ability of their compliance
department’s IT systems to
satisfy the organization’s
compliance responsibilities1
Culture of
compliance
68%
…yet only 17%
believe they are
fully prepared2
Mastering Big Data
Analytics is a competitive
advantage
Three-quarters
of institutions are
making investments
in their data analytics
capabilities…
75% 17%
Of banking and capital
market CEO cited failing
to keep pace with
technology as a major
concern.
Speed of
technology
changes
81%
Dodd-Frank allows the Office of
Financial Research (OFR) to subpoena a
financial institution to provide any data
that it may perceive as a potential risk
(including structured or unstructured),
any data analytics solution can be
leveraged as legal discovery.
IT spending for 2016 is
projected to reach $480
billion, according to IDC.
Separately American
Banker reported 4 areas
will see the largest
increase in IT spending.

3. Industry Challenges
Rapid Information
Growth
Inflexible ProcessesSiloed Operations Security Threats Lack of Transparency
Uncertain Markets Global CompetitionCost Pressures 7x24 Availability
Regulatory
Compliance

4. Layers of Disruption in FinTech
• We believe the initial activity and
transaction volume will be from the
first layers of disruption in FinTech
(i.e. banking tech, payments), but
long-term and more game-changing
innovations will occur in the next
layers.
• Ultimately, this will lead to new
startups and technologies changing the
core of the finance sector: commercial
banking and insurance (life, health,
property).
Attacking the Core BANKING TECH
PAYMENTS
CYBER SECURITY
BUSINESS FINANCE
CONSUMER FINANCE

5. Musical Chairs… an Industry Lesson

6. Success is a company’s ability to see the window of
opportunity and capitalize on it.
https://www.ted.com/talks/bill_gross_the_single_biggest_reason_why_startups_succeed?language=en

7. Getting to Velocity
We need first to solve for Enterprise Complexity
The only way to fix a complex
system like an enterprise is to:
Create conditions for the system to
fix itself
Complexity
Portfolio ProcessesOrganization

8. State of DevOps Report 2016
TRADITIONAL MODELS TRANSFORMED IT
of IT Executives admit delivering project
on-time and on-budget is challenging
Say this challenge has remained the same
or gotten worse
over time
~90%
~75%
30x More frequent code deployments
Higher rate of success deploying
change
Faster Mean Time To
Recovery (MTTR)
>80%
Report improvements in productivity,
quality, visibility, alignment, and
responsiveness
60x
168x
https://puppet.com/resources/whitepaper/2016-state-of-devops-report

9. SO WHAT IS
DevOps
and why should we care?

10. Why DevOps?
…optimizing the flow of work to create products
that create value faster and more reliably.
--- Only thing we know that builds/runs CD pipelines
OPS
DEV

11. DevOps is… Means Different Things, but in Common:
• DevOps is the only thing we know that effectively operates Continuous
Delivery Pipelines
• DevOps is an overloaded term these days like the cloud
• DevOps is what Modern IT looks like
• DevOps is how we create the coded business for the digital economy
• DevOps is the most advanced form of software-systems engineering at
scale, and in the DevOps/CD world, everyone's a developer
• We can't do DevOps for you, but we can do it with you…Teaching and
organization to fix itself
• Velocity is the end game, DevOps is the operating manual

12. DevOps and Traditional IT

13. SO WHAT ARE
Continuous
Delivery Pipelines
and why do they matter?

14. CD Pipelines – Digital Economy’s Factories

15. CD Pipelines are your company’s automated policies of releasing digital
value (i.e. software)
• Quality Gates
• Performance
• Architecture
• Compliance
• Security
• Operations
Deployment Environments Management & Monitoring
Version Control Build Tools Continuous Integration Tools Test Frameworks
Test
Manager
Commit & Build
Application
Validate
Code & Build
Deploy
Application
Acceptance
Test
Capacity
Test
Deploy
Infrastructure
Commit & Build
Infrastructure
Validate
Automation Package
Production
Deploy
Version Control
Configuration Managment Artifact Repositories
API Library
API Library
Ideal DevOps Workflow
IDEAL DEVOPS WORKFLOW
POLICY
POLICY
POLICY
POLICY

16. Until Released to Production, all Changes are Risks and Costs
Therefore…
Each Release includes
multiple iterations
Each iteration includes
multiple Changes
Each Release includes many
Changes, leading to lots of
aggregated risk
• Key DevOps Principle: Unreleased Changes create risk
• Successfully released changes create value
• In Traditional IT, lots of Changes were grouped together to form a Release
• But cross functional DevOps team releases small changes frequently:

17. How Does Continuous Delivery Help Reduce Risk and Get to Value Faster?
Each feature commit is now a potentially
DEPLOYABLE, release candidate

18. Fortune 500 Financial Services Value Stream to CD Pipelines
~ 7% PCE per AppA and AppB
Value Streams

19. Mapping Toolchains to CD Pipelines
Deployment Environments Management & Monitoring
Version Control Build Tools Continuous Integration Tools Test Frameworks
Test
Manager
Commit & Build
Application
Validate
Code & Build
Deploy
Application
Acceptance
Test
Capacity
Test
Deploy
Infrastructure
Commit & Build
Infrastructure
Validate
Automation Package
Production
Deploy
Version Control
Configuration Managment Artifact Repositories
API Library
API Library
Ideal DevOps Workflow
API Framework
enables common
workflow to support
multiple tools and
technology stacks

20. SO WHAT ARE
Enterprise DevOps
Success Patterns
and how do we use them?

21. Bottom Up Pattern of Success
• DevOps@Scale
One team and app at
a timeCD PIPELINE
POLICY
LA U N C H
APP
APPLICATION
GOAL
COLLAB TEAM

22. DevOps Transformation
Mode 2 Success Pattern
APP
GOAL
CD PIPELINE
HYPER CONVERGED
INFRASTRUCTURE
POLICY
L A U N C H
Dependent Applications (Program Teams)
S C A L E
GOAL
GOAL
GOAL
CD PIPELINES
T R AN S F O R M
100-1000s Applications (Enterprise)
HYPER CONVERGED
INFRASTRUCTURE
APP
APP
APP
POLICY
POLICY
APPLICATION
COLLAB TEAM
PROGRAM TEAMS
APPLICATION SUITE
ENTERPRISE PORTFOLIO
ENTERPRISE TEAMS
PREPARE
ROADMAP
ENVIRONMENT
TOOLING
METRICS
TRAINING

23. DevOps Bottom Up Success Pattern
Learning to build skakeboards first… Ultimately a IT Portfolio of CD Pipelines
AP
P
AP
P
AP
P
AP
PAP
P
AP
P
AP
P
AP
P
AP
P
AP
PAP
P
AP
P
AP
P
AP
P
AP
P
AP
PAP
P
AP
P
AP
P
AP
P
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
APP
Single Application
(Project Team)
Dependent Applications
(Program Teams)
10-100s Applications
(Line-of-Business Portfolio)
100-1000s Applications
(Enterprise)
PHASE1
PHASE2PHASE4
PHASE3
GOAL
GOAL
GOAL
GOAL
Goal
+ +
+ +
=CD Pipeline =CD Pipelines
=CD Pipelines
=CD Pipelines

24. Top Down Antipattern
Big Bang Transformation “Wishing
Start
Finish

25. Big Bank Transformation “Reality”
Start
Finish
Fear
Panic
Maybe
Focus
People revert back to
previous behaviors

26. Top Down Success Pattern of Enterprise DevOps
Teach an Organization to Fix Itself
Start
Finish
Fear
PanicAbort
Maybe
Focus
Fear
PanicAbort
Maybe
Focus
Fear
PanicAbort
Maybe
Focus
Fear
PanicAbort
Maybe
Focus
Fear
Panic
Abort
Maybe
Focus
Fear
Panic
Abort
Maybe
Focus
Fear
PanicAbort
Maybe
Focus
Fear
Panic
Abort
Maybe
Focus
Fear
Panic
Abort
Maybe
Focus

27. Questions and Answers?