DAMA International Symposium San Diego CA 03-17-2008

© 2008 RCG IT, DAMA & Wilshire Conferences. All rights reserved.
Enterprise InformationEnterprise Information
Management [EIM]Management [EIM] ––
Proven Techniques ForProven Techniques For
SuccessSuccess
Robert J. Abate, CBIP
Global Managing Principal, RCG ITGlobal Managing Principal, RCG IT
Monday, March 17th, 2008 5PM – 6PM

AgendaAgenda
Presentation TopicsPresentation Topics
EndEnd--ToTo--End Data Management [EIM]End Data Management [EIM]
Requires A Integration ArchitectureRequires A Integration Architecture
Integrating Governance & QualityIntegrating Governance & Quality
Proven Techniques For SuccessProven Techniques For Success
Key TakeKey Take--AwaysAways

EndEnd--ToTo--End DataEnd Data
Management [EIM]Management [EIM]

RCG INFORMATION TECHNOLOGYRCG INFORMATION TECHNOLOGY
4
What Is Enterprise Information Management?
Intersection Of Strategy, Architecture & Delivery Of InfoIntersection Of Strategy, Architecture & Delivery Of Info
It Is A Managed Information Environment (with End – To – End Data
Management) Incorporating:
Where Data Is Known To Be Of High Quality
Information Is Wrapped With (Or Having) Governance
Contain A Single Source Of Truth
Flexible Structure or “Architecture”
Responsive To Change
Supportive Of Integration And Delivery Of Information
A flexible structure is important because:A flexible structure is important because:
Business needs are changing at an accelerating pace
It is the core feature that allows IT to be responsive in meeting new
business requirements

5
The Information Foundation: Data Management
This requires:This requires:
Metadata managementMetadata management for compliance enforcement, audit support,for compliance enforcement, audit support,
analysis, and reportinganalysis, and reporting
Master dataMaster data integration and controlintegration and control
NearNear--real timereal time business informationbusiness information
Source data managementSource data management for controlling data quality at thefor controlling data quality at the
transaction leveltransaction level
Effective governanceEffective governance for a successful managed data environmentfor a successful managed data environment
IntegrationIntegration of analytics, reporting, and transaction controlof analytics, reporting, and transaction control
ControlControl of business processes and information usageof business processes and information usage
Unstructured dataUnstructured data needs to be addressed as wellneeds to be addressed as well
The role of data management in the business is evolving from
“A Single Source Of Truth” to becoming
“The Information Assurance And Delivery Mechanism”

6
How Do You Build One?
Flexible Reference / Data ArchitectureFlexible Reference / Data Architecture
Controlled Data LifecycleControlled Data Lifecycle
Metadata / Information TaxonomyMetadata / Information Taxonomy
Master Data ManagementMaster Data Management
ModelModel--Driven DevelopmentDriven Development
Data Integration ToolsData Integration Tools
Extraction, Transformation & Loading [ETL]
Data Delivery Services
Information DeliveryInformation Delivery
Reporting & Analytics
Security
Dimensional Models (Data Warehouse)Dimensional Models (Data Warehouse)
Conformed Dimensions
Data GovernanceData Governance
Data QualityData Quality
Data Profiling
Exception Handling
Action Time
ValueLost
Action Time
ValueLost
Time
Value
Action
Business Event
Taken
Business Event
Taken
Capture
Latency
Analysis
Latency
Decision
Latency
Data Ready For Analysis
Information Delivered
Capture
Latency
Analysis
Latency
Decision
Latency
© Chart: Richard Hackathorn, Bolder Technology, Inc., 2007. All Rights Reserved. Used with Permission.
Data
Sources
Finance
XLS
POS
ERP
SCM
ERP
Transaction Level Detail
Granular
Denormalized
Integrated
Current & History
Reporting
Granular
Denormalized
Integrated
Current & History
Reporting
Metadata
(Extract/Transform/Load)
-- Apply business rules
-- Manage metadata
-- Consolidate data
ETL / Data
Integration Hub
‘Atomic’-level Data
Normalized (3NF)
Integrated
Temporary
Integration
Aggregates & KPIs
Dimensional
Drilling
Conformed
Analytics
ETL
Server
Dashboards
Predictive &
Behavioral
Analytics
Reporting &
Ad hoc Queries
Database
Server Reporting
Server /
Portal
Scoring
Algorithms
Metadata Inquiries
Others
A “permanent” Integration Data Store can
facilitate data conversions and system
upgrades because it contains data already
standardized, cleansed, and integrated.
*
*
Data DeliveryEnterprise Data

7
What Is A Flexible Architecture?
Definition of a flexible architectureDefinition of a flexible architecture (from Dictionary.com)(from Dictionary.com)::
Architecture:Architecture: “Orderly arrangement of parts; structure”
Flexibility:Flexibility: “Responsive to change; adaptable”
Responsive:Responsive: “readily reacting to events or stimuli”
Therefore, a flexible architecture is one that is:Therefore, a flexible architecture is one that is:
Readily adaptable to events or stimuli that require it to change
Designed to provide the key capabilities necessary for BI
today
And an inflexible architecture is one that is:And an inflexible architecture is one that is:
Hard or time-consuming to change
Unable to incorporate new capabilities when needed

8
Partner with the Business: Focus on Impacting the Bottom Line
Operational BI
Strategic BI
Simple metrics:
• Customer satisfaction
• Inventory turns
• Quality measurement
• Financial reporting
Dashboard with metrics:
• Customer profitability
• Product profitability
• Operational performance
Performance
Management
Performance
Management
Performance
Optimization
Performance
Optimization
Complex metrics:
• Market success
• Customer relationship
• Information innovation
• Value creation
Value to the
Corporate
Bottom Line
These inform different kinds of decisions at
different levels of the organization
Operational
Reporting
& Analysis

9
Best Practices & Trends…
Separation of Architectural Layers usingSeparation of Architectural Layers using ReferenceReference
ArchitecturesArchitectures ((““akaaka”” Architectural FrameworksArchitectural Frameworks))
Model Driven Development [MDD]Model Driven Development [MDD] with a focus on thewith a focus on the
business functions / processes to drive alignment withbusiness functions / processes to drive alignment with
business goals and objectivesbusiness goals and objectives –– ““akaaka”” BusinessBusiness
ArchitectureArchitecture
Providing requirements traceability
“Glue” or integration points for iterative development
Convergence OfConvergence Of Applications (SOA)Applications (SOA) && Managed DataManaged Data
Environments (Business Intelligence)Environments (Business Intelligence)

Requires An IntegrationRequires An Integration

11
The Problem – Aligning Systems with the Business
1. As proposed by the project sponsor
2. As defined by the requirements doc’s
3. As designed by the Senior Analyst
4. As produced by the programmers
5. As installed at the user’s site
6. What The User Really Wanted!
This Cartoon Has Been Circulating In IT For Over 20+ Years In Various Forms…

12
The Importance of Requirements
““Analysts report that as many as 71% of projects thatAnalysts report that as many as 71% of projects that
fail do so because of poor requirements management,fail do so because of poor requirements management,
making it the single biggest reason for project failuremaking it the single biggest reason for project failure””
CIO magazine, November 15, 2005
Failure can also be defined as:Failure can also be defined as:
““Analysis ParalysisAnalysis Paralysis”” oror ““Scope CreepScope Creep””
A never ending test cycleA never ending test cycle
Redevelopment of systems after implementationRedevelopment of systems after implementation

13
Relationship Between Business Architecture & IT
Enables
Dictates
Enables
Dictates
Supplements
Dictates
Dictates
Business /
SBU Strategy
Business
Architecture
Business
Projects &
Programs
Implies
Dictates
Technology
Strategy
App. & Info.
Technology
Architecture
Technology
Projects &
Programs
Implies
Contributes&Supports
There Is An “Implied” Relationship Between Business & Technology Building Blocks
Business Architecture is a comprehensive framework for the representation of the business processes,
information, rules, and organizational designs that result directly from the business strategy. It provides a
coherent graphical means for the capture, management, and communication of business knowledge and
needs to support business solutions in the form of actionable requirements.
IT Confined By
Project Budgets
And Define App/Info
Architecture

14
Define business requirements in aDefine business requirements in a
manner thatmanner that aligns IT and Businessaligns IT and Business
Requirements areRequirements are graphicalgraphical
(pictures) and(pictures) and easy to understandeasy to understand
Supports both application andSupports both application and
information architecturesinformation architectures
Vendor agnosticVendor agnostic
Supports rules established inSupports rules established in
Business ArchitectureBusiness Architecture
Provides forProvides for checking consistencychecking consistency
and completeness throughoutand completeness throughout
Supports management withSupports management with qualityquality
control and governancecontrol and governance
ModelModel--driven approachdriven approach should use ashould use a
repositoryrepository--based toolbased tool
The Many Benefits Of Model Driven Development

15
MDM Has Full Traceability - Through Design, Build and Test
Design
&Build
Require
ments
Testing
Requirements and
Acceptance Criteria
Design Component and/or
Use Case that
Satisfies each Requirement
Test Groups, Test Cases
Of Each Design Component
Requirements are satisfied by a design component and each componRequirements are satisfied by a design component and each component isent is
assigned to a test group with test cases. Thus test cases testassigned to a test group with test cases. Thus test cases test requirements.requirements.
Test results can then be measured against acceptance criteria.Test results can then be measured against acceptance criteria.

16
Reference Architecture And It’s Views
BUSINESS ARCHITECTURE
Mission / Vision, Business Requirements, Business Rules, Organizational
Structure, Critical Success Factors, Business Process Design
APPLICATIONS ARCHITECTURE
Process Alignment, Services Definition, Services/Event Architectures, Custom
Application Development, Enterprise Applications Integration Components
INFORMATION ARCHITECTURE
Data Integration, Data Architecture, Master Data Management, Metadata
Management, Data Delivery Architecture, Dashboards, Performance
Measurement, Analytics , Business Intelligence, Enterprise Reporting
INFRASTRUCTURE ARCHITECTURE
Legacy (Non-Service Enabled) Applications
Networks, Servers / Hardware, Operating Systems
INTEGRATEDGOVERNANCE
Security,Data,SLAs,ServicesandITPortfolioManagement
QUALITYCONTROL
QualityAssurance,Testing,DataQualityManagement
Reference ArchitectureReference Architecture is a proven template for a solution domain thatis a proven template for a solution domain that
provides a common vocabulary with which to discuss designs andprovides a common vocabulary with which to discuss designs and
implementationsimplementations –– it frames solution delivery.it frames solution delivery.
Enterprise Architecture FrameworkEnterprise Architecture Framework describes a structure anddescribes a structure and
behavior for an organization's processes, information systems,behavior for an organization's processes, information systems,
personnel, and organizational subpersonnel, and organizational sub--units, aligning them with theunits, aligning them with the
organization's core goals and strategic direction.organization's core goals and strategic direction.

17
Reference Architecture – Artifacts / Techniques
StrategyStrategyProcessProcessFunctionFunctionTimeTimeLocationLocationDataDataOrganizationOrganization
WHYWHYHOWHOWWHENWHENWHEREWHEREWHATWHATWHOWHO
EnterpriseEnterprise
Infrastructure VisionInfrastructure Vision
Strategic NetworkStrategic Network
PlanPlan
System OperationSystem Operation
Diagram (SODDiagram (SOD--P)P)
Security ManagementSecurity Management
PlanPlan
Systems ArchitectureSystems Architecture
DiagramDiagram
Network ArchitectureNetwork Architecture
DiagramDiagram
InfrastructureInfrastructure
Calendar PlanCalendar Plan
Location Support PlanLocation Support PlanData ManagementData Management
PlanPlan
Data Capacity PlanData Capacity Plan
Data Archive PlanData Archive Plan
InfrastructureInfrastructure
Resource PlanResource Plan
CSFCSF’’ss
Business Goals /Business Goals /
DirectionDirection
Business PlanBusiness Plan
– BPM / BPI
– Continuity Plan
Policy CharterPolicy Charter
Business ProcessBusiness Process
Flow (SODFlow (SOD--L)L)
Use CaseUse Case
ScreensScreens
ReportsReports
Business Logic (ASD)Business Logic (ASD)
Business FunctionsBusiness Functions
(FSD)(FSD)
Business ProcessBusiness Process
(FSD)(FSD)
Business PrioritiesBusiness Priorities
Business TimeBusiness Time
DefinitionsDefinitions
Master ScheduleMaster Schedule
(Events & Triggers)(Events & Triggers)
Business LocationsBusiness Locations
LocationLocation--specificspecific
BusinessBusiness
Requirements &Requirements &
DefinitionsDefinitions
Entity RelationshipEntity Relationship
Diagram (EDD)Diagram (EDD)
Attribute DependencyAttribute Dependency
Diagram (ADD)Diagram (ADD)
KPI Structure DiagramKPI Structure Diagram
(KSD)(KSD)
Organization ChartOrganization Chart
(OSD)(OSD)
RolesRoles &&
ResponsibilitiesResponsibilities
System Operation Diagram (SOD-L)
System Operations Documentation
Disaster Recovery Plan
BusinessRequirementsdriveEnterpriseArchitectureandBusinessContext
Business
Architecture
Business PlanBusiness Plan
AlignmentAlignment
ApplicationApplication
PrioritizationPrioritization
Replacement PlanReplacement Plan
Workflow ProcessWorkflow Process
ModelModel
Architecture DesignArchitecture Design
Systems InterfaceSystems Interface
DesignDesign
Architecture DesignArchitecture Design
Services Map ScreenServices Map Screen
DiagramsDiagrams
Application / ProcessApplication / Process
Event CorrelationEvent Correlation
DiagramsDiagrams
Location ProcessingLocation Processing
Workflow ModelsWorkflow Models
Semantic ModelSemantic Model
(Metadata)(Metadata)
Messaging DiagramMessaging Diagram
XML / XSLT DiagramsXML / XSLT Diagrams
SOAP / MessagesSOAP / Messages
Organization ProcessOrganization Process
Mapping RulesMapping Rules
User InterfaceUser Interface
DiagramsDiagrams
METADATA (repository-based): Application / Services Metaprocesses
Applications
Architecture
Enterprise ManagedEnterprise Managed
Data EnvironmentData Environment
Master DataMaster Data
Management PlanManagement Plan
BusinessBusiness
PerformancePerformance
ManagementManagement
MonitoringMonitoring
Data ExceptionData Exception
ProcessingProcessing
Data PerformanceData Performance
Enterprise DataEnterprise Data
Structures AuditStructures Audit
Enterprise ReportingEnterprise Reporting
& Analytics& Analytics
ETL / Data IntegrationETL / Data Integration
MetadataMetadata
Event DataEvent Data
Correlation DiagramsCorrelation Diagrams
Enterprise DataEnterprise Data
SynchronizationSynchronization
ScheduleSchedule
Data ControlData Control
StructuresStructures
Data Location &Data Location &
PhysicalPhysical
InterdependencyInterdependency
MapsMaps
Data Source OfData Source Of
Record (SourceRecord (Source--toto--
Target for EnterpriseTarget for Enterprise
Data Structures)Data Structures)
DimensionalDimensional
Hierarchies /Hierarchies /
TaxonomyTaxonomy
KPI / MetricsKPI / Metrics
DefinitionDefinition
Master Data ModelMaster Data Model
Logical / PhysicalLogical / Physical
Data ModelsData Models
Physical StructuresPhysical Structures
User AccessUser Access
AuthorizationAuthorization
Data Visualization &Data Visualization &
DeliveryDelivery
RequirementsRequirements
METADATA (repository-based): Business (Definitions, Access Rights…), Operational (Load Timings, Controls…) & Technical (Environmental…)
Information
Architecture
Infrastructure
Architecture

18
Reference Architecture – Logical Managed ComponentsINTEGRATEDGOVERNANCE
Security,Data,SLAs,ServicesandITPortfolioManagement
QUALITYCONTROL
QualityAssurance,Testing,DataQualityManagement
INFORMATION ARCHITECTURE APPLICATIONS ARCHITECTURE
Business Rules Engine
App / Svcs
Integration
Broker
Data
Rules
App
Rules
Services Mgmt
Event Correlation
Process Choreography
Svcs Orchestration
Services Mgmt
Event Correlation
Process Choreography
Svcs Orchestration
BUSINESS ARCHITECTURE
Existing Application Backbone for Transaction Processing (includes Legacy Applications)
Existing Application Backbone for Transaction Processing (includes Legacy Applications)
Central Metadata & Metaprocess Library
Data & Data Audit Metadata
Enterprise
Reporting
Integrated KPIs &
Dashboards
Data Quality,
Profiling & Cleansing
Enterprise
Reporting
Integrated KPIs &
Dashboards
Data Quality,
ETL / Data
Integration
Hub
Enterprise Data Structures
Sync
Sync Sync
Integrated Trx Data Store
Trx Processing
Recent Data, Normalized, Fine-
grained, Unit-of-Work (UoW)
Processing
EDW & Data Marts
Reporting
Granular,
Denormalized,
Current / History
Analytics
Aggregates,
Dimensional,
Conformed
Master Data
Customer
Value, Control
Data, ID X-ref
Product
Value, Control
Data, ID X-ref
Orgn Chart
Value, Control
Data, ID X-ref
Enterprise Data Structures
Sync
Sync Sync
Integrated Trx Data Store
Trx Processing
Recent Data, Normalized, Fine-
grained, Unit-of-Work (UoW)
Processing
EDW & Data Marts
Reporting
Granular,
Denormalized,
Current / History
Analytics
Aggregates,
Dimensional,
Conformed
Master Data
Customer
Value, Control
Data, ID X-ref
Product
Value, Control
Data, ID X-ref
Orgn Chart
Value, Control
Data, ID X-ref
Applications / Services
Metaprocesses
Services Library
Application Svcs
Data Delivery Svcs
Messaging Svcs
Services Library
Application Svcs
Data Delivery Svcs
Messaging Svcs
StrategyProcessFunctionTimeLocationDataOrganization
WHYHOWWHENWHEREWHATWHO

Integrating GovernanceIntegrating Governance
& Quality& Quality

20
Dedicated Management of Process and CommunicationsDedicated Management of Process and Communications
Data Governance – An Overview

21
What Is Data Governance Anyway…
Data Governance is what an organization does that yields:Data Governance is what an organization does that yields:
The highest quality decisions made at the lowest cost.
Increased efficiency in conducting/completing IT projects for the business.
Data Governance is NOT:Data Governance is NOT:
A project
Someone else’s responsibility
Intuitive
Something that happens on it’s own
Drivers For Data Governance:Drivers For Data Governance:
Lack of trust in the data
Reports utilizing same data do not report same totals or computations
Data not available in an easy manner
Multiple sources of data have to be gathered and manipulated
Data timeliness
Manual process to collect, analyze and provide results
Data integrity
Unknown filters, varying calculation/computations, fields used for data not indicative of field
names, data passed along from one person to another to another to another…..

22
Data Governance – Master Data Management Included…
A Comparison of Data Governance & MDMA Comparison of Data Governance & MDM
Master Data ManagementMaster Data Management
Brings together single, consistentBrings together single, consistent
view of data that may reside inview of data that may reside in
multiple source applications (CRM,multiple source applications (CRM,
ERP, financial applications, WebERP, financial applications, Web
servers) by maintaining a singleservers) by maintaining a single
source of critical data, then crosssource of critical data, then cross--
referencing and synchronizing datareferencing and synchronizing data
across all other missionacross all other mission--criticalcritical
applicationsapplications
Critical business data flows in realCritical business data flows in real--
time to multiple applications fortime to multiple applications for
single view of customers, supplierssingle view of customers, suppliers
and other dataand other data
Data GovernanceData Governance
StrategyStrategy –– Why? Goals andWhy? Goals and
objectives of organization (treatobjectives of organization (treat
data as an asset of organization)data as an asset of organization)
–– MDM is a goalMDM is a goal
Policies, Procedures, ProcessPolicies, Procedures, Process ––
What? Who? When? Where?What? Who? When? Where?
Provides structure,Provides structure,
communications, andcommunications, and
accountabilityaccountability
Implementation / ResolutionImplementation / Resolution ––
How? RealHow? Real--world application toworld application to
make it all happenmake it all happen
Beyond MDM: Regulatory Risk, Compliance, and SOARegulatory Risk, Compliance, and SOA

Proven Techniques ForProven Techniques For
SuccessSuccess

24
Partner With The Business: Help Them Understand Themselves
Stage 4: ‘Perform’ (business results)
Actions: Structure KPIs & metrics
Focus: Dashboarding processes
Stage 3: ‘Structure’ (enterprise data)
Actions: Conform & Standardize Data
Focus: Standardizing data & terms
Stage 2: ‘Verify’ (EDW data)
Actions: Ensure Data Quality
Focus: Governing data & metadata
Stage 1: ‘Trust’ (in application data)
Actions: Collect & Organize
Focus: Gathering data for reporting
Stage 5: ‘Optimize’ (the business)
Actions: Establish Strategy & Goals
Focus: Improving market success
Departmental Usage
Enterprise Ready
Executive Driven
Reality
Perception

25
Develop Iteratively – It’s All About The Repeatable Process
A repeatable process implies an iterative methodologyA repeatable process implies an iterative methodology
Utilize open-ended architecture/design principles to facilitate reuse in future
iterations
Stagger iterations as shown to save time and money
Leverage best practices already available
Requirements must be thoroughly understood
Data profiling must be conducted in parallel with Requirements
Activities should take progressively lesser time during subsequent iterations
due to leveraging reusable components
Program Management and Governance will ensure that iterations meet
business requirements, minimize rework between iterations and proper
management of master data through its entire lifecycle
Requirements Architecture Development Deployment
Iteration 1
Iteration 2
Iteration …

26
Proven Techniques: Define Your Components
■ Data Management
– Master data
– Data standardization
– Data quality
– Metadata
– Data security, audit, and compliance
■ Data and delivery architectures
– Modeling and structuring data
– ETL data integration and data rules
– Data delivery services (SOA-compliant)
– Data technologies compatibility
■ Performance management
– KPI relationships and metrics structure
– Dashboard development and integration
■ Analytics management
– Enterprise reporting
– Analytics in transaction processing
– “Real-time” analytics
– Business activity monitoring (BAM)
INFORMATION ARCHITECTURE
Data Rules
Data & Data Audit Metadata
Enterprise
Reporting
Integrated KPIs &
Dashboards
Data Quality,
ETL / Data
Integration
Hub
EnterpriseDataStructures
Sync
Sync Sync
IntegratedTrxDataStore
TrxProcessing
Recent Data,Normalized,Fine-
grained,Unit-of-Work(UoW)
Processing
EDW&DataMarts
Reporting
Granular,
Denormalized,
Current/History
Analytics
Aggregates,
Dimensional,
Conformed
MasterData
Customer
Value,Control
Data,IDX-ref
Product
Value,Control
Data,IDX-ref
OrgnChart
Value,Control
Data,IDX-ref
EnterpriseDataStructures
Sync
Sync Sync
IntegratedTrxDataStore
TrxProcessing
Recent Data,Normalized,Fine-
grained,Unit-of-Work(UoW)
Processing
EDW&DataMarts
Reporting
Granular,
Denormalized,
Current/History
Analytics
Aggregates,
Dimensional,
Conformed
MasterData
Customer
Value,Control
Data,IDX-ref
Product
Value,Control
Data,IDX-ref
OrgnChart
Value,Control
Data,IDX-ref
A Managed Data Environment
Data Delivery Svcs
Implies… Managed Data Components

27
Proven Techniques: Define Your Data Lifecycle…
© Richard Hackathorn, Bolder Technology, Inc., 2007. All Rights Reserved. Used with Permission.
Managing The Time-Value Curve
Action Time
ValueLost
Time
Value
Action
Business Event
Taken
Capture
Latency
Analysis
Latency
Decision
Latency

28
Proven Data Architecture – Data Validation & Verification
Data
Sources
Finance
XLS
POS
ERP
SCM
ERP
Granular
Denormalized
Integrated
Current & History
Reporting
Metadata
(Extract/Transform/Load)
-- Apply business rules
-- Manage metadata
-- Consolidate data
ETL / Data
Integration Hub
‘Atomic’-level Data
Normalized (3NF)
Integrated
Temporary
Integration
Aggregates & KPIs
Dimensional
Drilling
Conformed
Analytics
ETL
Server
Dashboards
Predictive &
Behavioral
Analytics
Reporting &
Ad hoc Queries
Database
Server Reporting
Server /
Portal
Scoring
Algorithms
Metadata Inquiries
Others
Because there are several
technologies covered by this
graphic, it is important that
they “play well together”
Data DeliveryEnterprise Data
1
2
3
4 5
6

29
Proven Techniques: Vision and Direction
Establish A Vision of theEstablish A Vision of the
FutureFuture
The Mission & Vision
The Stakeholders
BI and Reporting Differentiators
Operating Principles
Goals and Objectives for BI and
Reporting
Prepare a Roadmap for thePrepare a Roadmap for the
FutureFuture
Recommended Initiatives & Impact
Action Plans for High-Priority
Initiatives
Transition Strategy
Critical Success Factors for the
Transition
•DataQualityManagement,Audit&Control
•MasterDataManagement
•ScorecardDevelopment
•ETLManagementforSourceSystemChanges
•EDWArchitectureandEnhancement
•BusinessPerformanceManagement
• Marketing
• Advertising
• Customers
• Finance
• Employee
• Organization
• Transportation
• Distribution Centers
• Inventory
• Orders
• Vendors
• Sales
• Sales Transactions
• Products
• Stores Phase 1:
Sales &
Stores
CAPABILITY
Phase 2:
Inventory &
Orders
Phase 3:
Finance &
Employee
Phase 4:
Marketing &
Customer
Implementation, Phase by Phase, progresses over time
Ongoing,
Centralized,
EDW
Capabilities
and Program
Management
• Store and Product Master Data
• Product Hierarchy Reference Data
• Sales Scorecard Development
• Vendor and Order Master Data
• Vendor Scorecard Development
• Inventory Scorecard Development
• Employee Master Data
• Organization Reference Data
• Expenses and Margin Scorecard Development
• Complete Store Scorecard Development
• Customer Master Data
• Business Performance Analysis
• Profitability Analysis (stores, products, customers)

30
Proven Techniques: Project Selection Planning
Widely used dimensions
Low Complexity
Considerable overall value
MEDIUM PRIORITY – HIGH IMPACT
Scarcely used dimensions
Low complexity
Easy to implement
QUICK WINS – LIMITED IMPACT
Scarcely used dimensions
High complexity
Low overall value
LOW PRIORITY – LIMITED IMPACT
Widely used dimensions
High complexity
High overall value
HIGH PRIORITY – HIGH IMPACT
VIPS
Broader usage
HIGH VISIBILITY
Narrower usage
LOW VISIBILITY
“Enterprise Bus Matrix”

31
Proven Techniques: Implementation Planning
A sequence of dimensions and marts that takes allA sequence of dimensions and marts that takes all
criteria into accountcriteria into account
Demonstrates dimension reuse over timeDemonstrates dimension reuse over time
Stage 1 – VIPS
2
2 2 2 2 2 2 2 2 2 2 2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2 2 2
2
2
2
2
2
2
2
2
2
2 2
2 2 2
2 2
2 2
2 2
2 2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2 2
2 2
2
2
2
2
2 2
2 2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3 3 3
3 3 3
3 3 3
3 3 3 33
3
3
3
3
3
4
4
4 4
4
4
4
4
4
4
4
4
4
4 4
4
4
4 4
4
4
4
4
4
4
4
4
4
4

32
Flexible Architecture: A Managed Data Environment Is Needed
Universe of
Applications
Universe of
Application
Data
Specialized Data
Marts
Built as needed from
the EDW
Enterprise Data
Warehouse (EDW)
Data extracted from
source systems of record Data Delivery Services
ExistingEnvironment
Dashboards
High-End
Analytics
End-to-End Data Management
(Data Consolidated, Aggregated, Standardized, Dimensions Conformed)
ManagedDataEnvironment
Enterprise
Reporting
“O.K.,
Where do I
get the
information
I need?”
“Cubes”
Premium
Customer
Customer
Address
Claim
Employee
Cash Value
Policy
Location
Property
Reporting
DB
Reporting
DB
App DB
App DB
App DB
Department
Data MartDepartment
Data MartDepartment
Data MartDepartment
Data Mart
Enterprise
Data Warehouse
ODS
Granular
Relational
History
Metrics
Aggregates
Dimensional
Drill-Thru
Metadata Repository
Master Data
RepositoryETL
Hub

33
Flexible Architecture: ETL, Metadata, and OLAP as Well
■ Define ETL Reference Architecture
– Independent of implementation technologies
– Flexible process-managed delta extraction
– Initial assembly into generalized business
objects
– Comprehensive audits and controls
supporting error detection, staging, and
notification
`
BOMM
User Interface
(Standard & Custom Reports)
BO XI R2 Universes
Business Objects Metadata Manager (BOMM XI Release 2 Repository
BO Report
Objects
CA AllFusion ERwin
(EDW Models)
IBM Websphere DataStage
(ETL Jobs)
EDW Server
EDW
Database
Staging
Database
Operational Source Systems
Advance 1
Peoplesoft
BO Data Insight XI Data
Profiling Repository
Future
EDW Server
Schema
Objects
DataStage
Transformation
Objects
Source
Schema
Objects
Profiling
Statistics Log
EDW Data
Defintion
Objects
Current
Future
Current
Current
Current
Current
Current
Current
ETL Audit
Reports
■ Metadata Management Architecture
– Leverage solutions that interconnect using
the Metadata
– Proactively leverage vendor partnerships to
formulate solution
– Custom integration of ETL Audits and
Controls reporting into framework
– Position for future integration of profiling
metadata

34
Proven Techniques: Deploying ETL Architecture
Provide the ability to perform/handle CDC and full refreshProvide the ability to perform/handle CDC and full refresh
Provide granular data load dependenciesProvide granular data load dependencies
Provide centralized, reusable audit and controlProvide centralized, reusable audit and control
componentscomponents
Provide centralized, reusable errors and exceptionsProvide centralized, reusable errors and exceptions
componentscomponents
Make the ETL process selfMake the ETL process self--healing / selfhealing / self--cleaningcleaning
Provide technical and applicationProvide technical and application--level metadatalevel metadata
integrationintegration
Provide security managementProvide security management
Provide nearProvide near--real time supportreal time support
Provide replicated data source supportProvide replicated data source support
Design for ETL performanceDesign for ETL performance

35
Proven Architecture: Measures and KPIs for Analysis

36
Governance: Conceptual Model For Governance Metadata
Implement Governance
Incrementally via a
Seven Stage Maturity Model
Author: Martha Dember, RCG IT
Implement GovernanceImplement Governance
Incrementally via aIncrementally via a
Seven Stage Maturity ModelSeven Stage Maturity Model
Author: Martha Dember, RCG ITAuthor: Martha Dember, RCG IT
Strategy & Framework
Scenarios & Validation
Formalized Organization & Responsive Process Rollout
Proactive Process Rollout
Expanded Business Involvement
Stewardship Culture
Strategic Governance
© 2005-2009. Martha Dember, All Rights Reserved.

37
Issues
Constraints
ConceptualConceptual
Data Model(s)Data Model(s)
LogicalLogical
Data ModelData Model
DataData
Life CycleLife Cycle
EntityEntity
ComplexityComplexity
Actor
Roles
Activities
Objects
Sources
Locations
Volume
Frequency
Sub-Act Sub-Act
Elements
Definitions Attributes
Facts &
Measures
Reports
Events
Business
Rules
Governance: Conceptual Model For Governance Metadata Repository

38
MDM: A Process For Management Of Master Data
2. Persist
Integrate with other master data as needed
Store in MDR
Store DQ score/rating
6. Maintain
Use Master Data Maintenance UI
Capture changes to structure/content
Store history
1. Define
Define source system of record for master data element
Design data structure and process for creation/capture
Perform QC prior to importing
3. Secure
(Create new roles/privileges)
Apply security policy
4. Consume
Distribute to relevant application data stores
Consumption governed by security policy
Application performs QA prior to acceptance
5. Monitor
Capture operational metadata on usage of master data
Perform capacity management of MDM infrastructure
Perform periodic QA in MDR to verify DQ score/rating
MASTER DATA
LIFECYCLE

Key TakeKey Take –– AwaysAways
(Remember These(Remember These……))

40
Key Take-Aways: Apply Proven Principles
FoundationFoundation:: Separation between data, source systems, and theirSeparation between data, source systems, and their
implementation technologiesimplementation technologies –– focus on BI & Data Delivery in this way tofocus on BI & Data Delivery in this way to
achieveachieve ““Information IndependenceInformation Independence””
Core PrincipleCore Principle:: Information design and managementInformation design and management –– an architecture foran architecture for
delivering correct, complete, reliable, and timely data quicklydelivering correct, complete, reliable, and timely data quickly andand
effectivelyeffectively
Keys to SuccessKeys to Success:: Information design and management are essential asInformation design and management are essential as
data delivery becomes critical to the businessdata delivery becomes critical to the business

41
Bottom Line BenefitsBottom Line Benefits ((essential for high business valueessential for high business value):):
Target BI analytics on improving business metrics (what would be the
value of increasing inventory turns by 5%?)
Learn from Lowe’s, 3M, and others
Engage the business in the core business process of focused
improvement by turning information into action
Cost SavingsCost Savings ((necessary, but not sufficientnecessary, but not sufficient):):
Eliminate redundant data and the associated storage and servers
Standardize on a small set of BI & Data Delivery technologies, reducing
ongoing license and maintenance fees
Match BI & Data Delivery capabilities to the users and manage license
fees for products
Reduce the time taken managing Excel and Access data manipulation,
analysis, and reporting
Key Take-Aways: Focus On Results

42
Every company can use data to deliver bottomEvery company can use data to deliver bottom--lineline
business valuebusiness value
Apply the core corporate performance managementApply the core corporate performance management
(CPM) business process(CPM) business process
Focus on bottomFocus on bottom--line benefits for real ROI; donline benefits for real ROI; don’’t get lostt get lost
in typical IT cost justificationsin typical IT cost justifications
While there are business and technical obstacles toWhile there are business and technical obstacles to
overall success, these are known and can be overcomeoverall success, these are known and can be overcome
Deliver through an architectural and data managementDeliver through an architectural and data management
frameworkframework
Use a combination of business, IT, and outside expertiseUse a combination of business, IT, and outside expertise
to develop a core competencyto develop a core competency
Key Take-Aways: Deliver Value

43
Key Take-Aways: Industry Trends
Much more enterpriseMuch more enterprise--level thinkinglevel thinking
BI selfBI self--service for easeservice for ease--ofof--use by business usersuse by business users
ReRe--architecting existing BI and data deliveryarchitecting existing BI and data delivery
environmentsenvironments
Starting to get anal about audit and control over dataStarting to get anal about audit and control over data
movementmovement
Growing awareness:Growing awareness:
Data quality issues, especially as they impact quality of analytics
Need for performance management
Competency centers (or center of excellence) performance is
different than skilled use of technology
Attacking problems with BI and data delivery strategiesAttacking problems with BI and data delivery strategies
and roadmapsand roadmaps

Thank You For YourThank You For Your
ParticipationParticipation……
rabate@rcgit.comrabate@rcgit.com
(732) 744(732) 744--36223622

DAMA International Symposium San Diego CA 03-17-2008

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to DAMA International Symposium San Diego CA 03-17-2008 (20)

More from Robert J. Abate, CBIP, CDMP (8)

DAMA International Symposium San Diego CA 03-17-2008