outline - buckeye dama · • donor centers paid for each registered donor and marrow donation 22 ....
TRANSCRIPT
2/27/2016
1
Case Study Workshop
Buckeye DAMA
Presented by Lewis Broome
February 25, 2016
Copyright 2014 by Data Blueprint
2
Outline
• Case Study 1: Deutsche Bank
– Background
– Data-Driven EA Design
– The Results
• Case Study 2: C.W. Bill Young Marrow Donor Program
– Background
– Data-Driven EA Design
– The Results
2/27/2016
2
Copyright 2014 by Data Blueprint
3
Background
• About the Company
• Market Positioning
• Operating Model
• Functional Decomposition
• Business Needs and KPI’s
• Current State Assessment
Copyright 2014 by Data Blueprint
4
About the Company (1)
• A German Bank
• Founded in 1870
• Their Big & Global
– #6 in the world based on assets ($2.6 trillion!)
– Top 5 Wall St. Firm
• Multiple Divisions / Products
– Retail Banking
– Private Wealth
– Investment Banking
– Corporate Banking
2/27/2016
3
Copyright 2014 by Data Blueprint
5
About Investment Banking
Business Unit
• Broad range of products and services
• Focused on serving
– Public and private institutional traders (pension funds, mutual
fund companies like Fidelity Investment)
– Private equity companies
– Hedge funds
• Three sub-divisions: Fixed Income; Equities & IBD
• Organized (and culturally divided) into “The Front Office”
and “The Back Office”
Copyright 2014 by Data Blueprint
6
Hubs for Investment
Banking Locations
2/27/2016
4
Copyright 2014 by Data Blueprint
7
Market Positioning (1)
Source: Business Insider: http://www.businessinsider.com/the-definitive-
ranking-of-wall-street-investment-banks-in-every-business-line-2015-9
Copyright 2014 by Data Blueprint
8
Market Positioning (2)
Lower Cost Differentiation
Broad
Range of
Buyers
Narrow
Buyer
Segment
2/27/2016
5
Copyright 2014 by Data Blueprint
9
Operating Model
Coordination
• Shared customers, products and suppliers
• Impact on other business unit transaction
• Operationally unique business units
• Autonomous business mgmt. & process design
• IT application align to business units
Unification
• Customers & suppliers may be local or global
• Globally integrated business processes
• Use enterprise systems
• BU’s with similar or overlapping operations
• Centralized management often matrixed
• Centrally mandated databases and IT
Diversification
• Few, if any, shared customers or suppliers
• Independent transactions
• Operationally unique business units
• Few data standards across business units
• Most IT decisions made within business units
Replication
• Few, if any, shared customers (per transaction)
• Independent transactions & BU Leaders
• Operationally similar business units
• Standardized data definitions but locally owned
• Centrally mandated IT services
Business Process Standardization
Low High
Hig
h
Lo
w
Bu
sin
ess P
roce
ss In
teg
rati
on
*Source: MIT
Copyright 2014 by Data Blueprint
10
Current State Assessment
Organization &
Readiness
Data Assets
Technology Assets
& Practices
Data Mgmt. Practices
Business Processes
Business
Goals and
Objectives
Enables
Informs
Enables
Enables
Measures
Delivers
Enables
Provides
Context
2/27/2016
6
Copyright 2014 by Data Blueprint
11
Current State: Cultural Readiness
Copyright 2014 by Data Blueprint
12
Organizational Structures
• Understand roles, responsibilities, authority & accountability
– CIO and COO at the same level
– IT decisions managed at the C-Level
• Assess Skills Across Business, Data & Technology
– Lots of Cobol Programmers
– A few good data modelers
– Experienced program managers
– Serious application & technical architects
– A few NASA-level technical engineers
– Experienced (and supportive) SME’s
Current State: Organization
2/27/2016
7
Copyright 2014 by Data Blueprint
13
Current State: Technology Assets
Global Booking Engine
Trade Processing
Settlement
Position Services
Sub-ledger
Products Accounts
Legal Entities
Traders
ReferenceData
Repo / Stock-
Borrow-Loan
StockRecord
Controller / Batch Processing
Gateway
MAINFRAME - dbTrader
32 Front Office Trading Systems (London)
Sysbase
Java
Confirms(Telex)
Electronic Trade
Allocations(Terminal)
Flat File (FTP)
Flat File (FTP)
Flat File (FTP)
Cash Mgmt. & Treasury
Oracle
PL/SQL
Flat File (FTP)
Mostly Client-Server applications A variety of technical platforms Mixture of homegrown & 3rd party
Reporting
Ops
Bloc
k Tr
ades
Settlement AgentsExchanges
CounterpartiesBanks
External Interfaces
Copyright 2014 by Data Blueprint
14
Current State: Business Process
– Run reports to see what and how many trades have been received
– Run reports for each sub-system to see what happen to the trades
– Find trade exceptions from looking at reports of ALL trades
Report-driven
Silo’ed
– Processes aligned to sub-systems
– Process integration points not understood or proactively managed
– Integration required extensive manual intervention & reconciliation
Constrained
– System would shut down at 100,000 trades
– Fragile & difficult to change because impacts can’t be anticipated
– Key-man dependencies defined the career path
2/27/2016
8
Copyright 2014 by Data Blueprint
Current State: Data
Management Practices
15
Data Management Maturity (DMM)℠ Model
Categories & Relationships
Data Governance
Data Management Strategy
Platform & Architecture
Data Operations
Data Quality
MetadataOversight
Business ITAlignment
Direction & Compliance
StakeholderAlignment
QualitySolutions
Data Infrastructure
Quality NeedsQuality Strategy
Business Process Data Requirements
Quality Rules/Criteria
Implementation Oversight:Collaboration
InfrastructureOversight
Supporting Processes – All Categories
On a scale of 1 to 5,
let’s just say they
were a ‘1’.
Copyright 2014 by Data Blueprint
16
Current State: Data Assets
• Transactional, Master, Reference and Accounting Data
• Structured and flat
• Typical of Legacy VSAM files
– Over-loaded fields were the norm
– Intelligence built into identifiers
– Non-atomic values
– Relationship tables widely used
• Attempted to maintain extensive technical metadata for change control
• 95% ROT (Redundant, Obsolete or Trivial)
2/27/2016
9
Copyright 2014 by Data Blueprint
17
Data-driven Unification EA Model
Core
Business
Component
Core
Business
Component
Master
Data
Lifecycle
Mgmt.
Exception
Workflows
Decision
Support
Shared
Utilities
etc…
Application architecture for enabling functionality
Application architecture for core business functionality
Middleware
passing data
objects
Data Objects
Copyright 2014 by Data Blueprint
18
Data-Driven EA Design - Guiding Principles
• Straight-through processing
• Data correct the first time using VESt’ing (Validation, enrichment
and standardization)
• Exception-based workflows
• Loosely coupled architecture; develop functional systems linked
through a common data model
• Treat data exchanges as a data model
• Identify the key data elements
• Put controls on key data elements
• Create and use business-event metadata to create absolute
workflow transparency and control – end-to-end
• Identify system of Record and Authoritative Data sources
2/27/2016
10
Copyright 2014 by Data Blueprint
19
The Results
By the Numbers:
• Reduced cost per trade by 54%
• Increased number of trades per operational staff by 37%
• Increased trading capacity by 2000%
• Improved ranking (from #11 to #2) in client trading performance
• Increased to over 1.5mn trades per day (from 500K per day)
New Capabilities:
• Work proactively
• Enable continuous improvement
• Transparency and control
• Business agility
Copyright 2014 by Data Blueprint
20
Outline
• Case Study 1: Deutsche Bank
– Background
– Data-Driven EA Design
– The Results
• Case Study 2: C.W. Bill Young Marrow Donor Program
– Background
– Data-Driven EA Design
– The Results
2/27/2016
11
Copyright 2014 by Data Blueprint
NMDP
Guidelines
DNA Matching
Patient Advocacy
National Donor Registry
Setting Context
21
Donor Center
Recruit Donors
Donor DNA Typing
Marrow Collection
Donor CRM
Transplant Center
Recipient Intake
Recipient Typing
Surgery / Outcomes
Copyright 2014 by Data Blueprint
About Marrow Donor Centers
• 120+ Donor Centers in the U.S.
• 9+ million registered donors with NMDP
• Approx. 4,800 marrow donations annually (increasing
14% per year)
• Probability of an HLA (DNA) match is approx. 1 in
20,000 donors
• C.W. Bill Young Department of Defense Marrow Donor
Center (MDC) has 854,000 registered donors (adds
~75,000 donors every year)
• Donor Centers paid for each registered donor and
marrow donation
22
2/27/2016
12
Copyright 2014 by Data Blueprint
Donor Centers Process
23
Stem cell
harvest
Donor
engagement HLA match
Recipient
intake
Donor
registration
Donor
Recruitment
=Potential Donors
=Registered Donors
Copyright 2014 by Data Blueprint
Data Valuation
• A unique valuation model – measure impact on
mission as well as business operations
• Donor Center Mission
– Provide high-quality HLA typings for a better match
– Highly efficient match request processing to meet
patient critical timing needs
– On-going donor engagement and education to
increase donor participation
• Donor Center Business Operations
– Increase revenue opportunities
– Cost takeout
24
2/27/2016
13
Copyright 2014 by Data Blueprint
Organizational Needs
25
Objectives Benefits / KPI’s
Productivity
Streamline & automate processes
• Increase # of stem cell collections per coordinator
• Increase #of DNA Samples typed per lab tech.
Transparency & Control
Create the capability to monitor,
measure, control & improve
processes in real time
• Reduce the number of missed SLA milestones
• Measure and improve process & staff performance
• Proactively & efficiently address workflow exceptions
• Full audit capability gives proof of FDA compliance
Quality of Services
Improve quality of services for
Donors, NMDP, Transplant
Centers & Recipients
• Min. process errors & max. reporting analytics
• Staff focused on quality of results by minimizing
manual steps
• Improved stakeholder experience
Donor Mgmt and Retention
Maximize participation and quality
of donors
• >99% accuracy for critical donor information
• Identify donors with highest probability of a match
• >75% donor participation rate
Copyright 2014 by Data Blueprint
EA Solution
26
Donor Center
Recruit Donors
Donor DNA Typing
Marrow Collectio
n
Donor CRM
Donor 360° View Data Services Layer Data Services Layer
Third Parties
Fedex, Labs, NMDP, BMDP
Donor Typing Lab
System
Donor Registry and CRM
Donor Center Workflow
2/27/2016
14
Copyright 2014 by Data Blueprint
Results (1)
27
0
50
100
150
Feb-08 Jul-09 Nov-10 Apr-12 Aug-13
Lag to Register Donor (in Days)
10
30
50
70
90
Apr-12 Jul-12 Oct-12 Jan-13 May-13 Aug-13
Number of Request Processed per Donor Coordinator
0
0.2
0.4
0.6
Apr-12 Jul-12 Oct-12 Jan-13 May-13 Aug-13
Percent of SLA's Hit for Donor Requests Processed
0
10
20
30
Apr-12 Jul-12 Oct-12 Jan-13 May-13 Aug-13
Average Days to Complete Donation Request
Mission Metrics
Business Metrics
Copyright 2014 by Data Blueprint
Results (2)
• Business Operations Goals – Higher throughput with same resources = decreased cost per
request
• Impact: 100% increase in requests processed per coordinator =
50% reduction in cost to process a request
– More SLA payment targets hit
• Impact: 32% increase in revenue from SLA targets
• Mission Goals – Fewer days to register a donor with NMDP
• Impact: Increase in donor growth population (velocity);
greater chance of finding a match
– Donation requests processed more quickly
• Impact: Increase in actual marrow donations (still to be
measured)
28
2/27/2016
15
10124 W. Broad Street, Suite C
Glen Allen, Virginia 23060
804.521.4056