8/2/2019 Preso Accenture_INFADAY_2011
1/18
Copyright 2011 Accenture All Rights Reserved. Accenture, its logo, and High Performance Delivered are trademarks of Accenture.
Data quality and data governance,Two strategic pillars of data management
October 6th
8/2/2019 Preso Accenture_INFADAY_2011
2/18
Data Governance
in the Data Management Journey
Global DataManagement
Vision
Governance
Data Quality Xavier Scherpereel
Arturo Salazar
Data Management is a business and technology matter.
Your master data management (MDM) discipline won't succeed unless you implement
effective governance(Gartner mars 2011)
Enterprise Data Model
Master Data Management
2Copyright 2011 Accenture All Rights Reserved.
8/2/2019 Preso Accenture_INFADAY_2011
3/18
Why Data Governance?
Some concrete examples to illustrate the issues
... to NASA, a defect related to the quality of the data (data enteredby mistake in inches while the calculations are performed in the
metric system) caused the loss of one entire mission, causing thecrash of a spaceship a loss of $ 100 million ... Source: ForresterResearch - The Costs of Data and Information Quality Defects
Copyright 2011 Accenture All Rights Reserved.
The quality of customer address data is degraded by 17% per year inthe United States. Source: United States Postal Service
According to a study of the industry of U.S. mail delivery, themail lost due to wrong address cost between $ 6 and $ 9 billionper year.
Source: Forrester Research
3
8/2/2019 Preso Accenture_INFADAY_2011
4/18
Why Data Governance?
The most common reasons to act
Optimization of the cost of marketing campaigns (mailings)
Potential cost of decisions based on erroneous data in the
reports
Increase revenues
Reduce costs
Targeting and effectiveness of marketing campaigns,
Better understanding of customers by eliminating duplicates
Copyright 2011 Accenture All Rights Reserved.
Requirements traceability of products and supplier
Anti laundering-money regulations
KYC (Know Your Customer)
Reputational risks (damage to the brand image to customers)
Operational and Credit riskControl risks
Comply with regulations
4
8/2/2019 Preso Accenture_INFADAY_2011
5/18
Why Data Governance?
Vision of your peers - Investments in the field of Analytics
Increase
35%Remain stable
62%
Copyright 2011 Accenture All Rights Reserved. 5
Decrease 3%
Source: Information Investments on France, Survey Accenture 2011More than 100 CIO surveys60% enterprises in the TOP 500
Data QualityData Governance
In the next 2 years, investments in the field of Analytics are required to progress for62% of companies (CIO Survey)
8/2/2019 Preso Accenture_INFADAY_2011
6/18
Why Data Governance?
Are you ready for Data Governance?
Governance for CIO Governance for IT Leads
Understand information governanceand MDM
Establish standards and policies forinformation governance and MDM
Identify critical businessprocesses and information
Identify business lines tocosponsor MDM initiatives
2
1
2
1
Copyright 2011 Accenture All Rights Reserved.
Use Market standards to assesswhich vendors will likely facilitateinformation governance initiatives
Definition and implementation of theGovernance at the enterprise level
Evaluate which enterpriseinformation will best support MDMprograms
Build a single vision for managinginformation assets and to developa business case
Link information governance to ITgovernance
3
4
Measure the business value ofreusing information across theorganization
5
3
4
5
6
8/2/2019 Preso Accenture_INFADAY_2011
7/18
8/2/2019 Preso Accenture_INFADAY_2011
8/18
Real Cases
Governance Implementation in a Financial Institution
Enterprise Data Governance Implementation Roadmap 2011
Feb Mar Apr May
Validation of roles andresponsibilities
Jun
EDM governance deployment:EDM / Domain / Transversal OpCos
Owners, Managers, Designersrepresentatives defined
Lot 1 : Norms & Standards definition /Cross DomainsAccenture implemented an
approach leveraging the workalready performed and
ensuring a global support tothe Head of Enter rise Data
Copyright 2011 Accenture All Rights Reserved.
Run
EDM Opco KickOff 22/01
KPI Lot 1Data Quality* / Business
Performance
Dashboards
On going communicationCommunication plan EDM Communication tostakeholders finalized
EDM OpCo Deliverable
Deliverables Description
Detailed Mapping of roles by domain List of data & process owners, managers and designers by
domain
Norms & Standards definition Data Golden Rules Best practices and Standards to respect across SG CIB
Perform key communication End to end organization of key communications
Performance & Risk Indicators
production
Key performance and risk indicators definition by domain Production of indicators for Opco EDM
Propose tactical solution and target run indicators
Management
8
8/2/2019 Preso Accenture_INFADAY_2011
9/18
Data Governance by Accenture
DataManagement
DataCreation
Data
Data
Retirement
Data Management FrameworkData Management Framework
Copyright 2011 Accenture All Rights Reserved.
DataGovernanceDataGovernance DataStructureDataStructure DataArchitectureDataArchitecture Master Data &MetadataMaster Data &Metadata DataQualityDataQuality
DataStorage
DataMovement
sage
Data Ownership
Data Stewardship
Data Policies
Data Standards
Data Ownership
Data Stewardship
Data Policies
Data Standards
Data Modeling
Data Taxonomy
Data Modeling
Data Taxonomy
Data Migration
Data Storage
Data Access
Data Archiving
Data Retirement
Data Migration
Data Storage
Data Access
Data Archiving
Data Retirement
Master DataManagement
Reference DataManagement
Metadata
Management
Master DataManagement
Reference DataManagement
Metadata
Management
Data Profiling
Data Cleansing
Data Monitoring
Data Compliance
Data Traceability
Data Profiling
Data Cleansing
Data Monitoring
Data Compliance
Data Traceability
DataSecurityDataSecurity
Data Privacy
Data Retention
Data Privacy
Data Retention
9
8/2/2019 Preso Accenture_INFADAY_2011
10/18
Foundational
Diagnostics Thought LeadershipADM-based Estimators
P&IM Offerings & Assets
Focus on some examples
10Copyright 2011 Accenture All Rights Reserved.
Strategic
Supporting
Executive Portal Enterprise Metrics Management
Electronic Dataand DocumentManagement
Delivery Centers
For access to all P&IM assets and offerings, click:http:// www.accenture.com
8/2/2019 Preso Accenture_INFADAY_2011
11/18
Inconsistencies = 90% of conversion failure
Which data to Cleanse?
Various IS Legacy Sources
Heterogeneity in Source Data
Numerous and Frequent Go-Lives
Recurrent Situations
Quality Control of the extracted data
Tests and validation of the upload
Industrialized approach:
- Standardizations
- Development re-use
- Process automation
Key Expectations
Why Data Quality?
Quality inSAP Data Migration projects
Respects of deadlines
SAP Data Migration Factory (SDMF)
Skilled and dedicated Teams
Proven approach with several customers
IS LegacySources
Target System:SAP
Download Upload
Quality ControlTransformations
Copyright 2011 Accenture All Rights Reserved. 11
8/2/2019 Preso Accenture_INFADAY_2011
12/18
Transformations- Mappings-Transformations
(PowerCenter)
LSMW
Flat Files
IDocs
IS Source
Data Migration SolutionLegacy Sources
1 3 4
SAP Target
2 Quality Checks- Quality Control- Analysis Reports
(IDQ)
Why Data Quality?
Details of the SAP Data Migration Factory
1. Data Extraction from IS source to an intermediate format
2. Quality Control of intermediate data before transformation
3. Conversion of intermediate data into SAP format
4. Load of converted data into SAP
AnalysisReports
UploadTransformation
gra on
Data Base
12Copyright 2011 Accenture All Rights Reserved.
Download and Cleansing
8/2/2019 Preso Accenture_INFADAY_2011
13/18
90 % of conversion errors = Source Data: formats,duplication, integrity, etc.
Referential and IS Constraints are different betweenLegacies and SAP
Duration of identification and correction of issues
Support and Control the overall process
Successful Go-Live
Main Stakes Identified
Data Quality Tool Analysis
Errors Identification in early stages Quality control Automation
Iterative Correction process
Correction: IS Legacy Sources or Migration Database
Proven approach with several customers
Enhance Quality = Accelerator
Why Data Quality?
Key Differentiator: Enhance Quality of Extracted Data
DataCleansing
ExtractQualityControl
Conversion
MigrationData Base
SAP
Target
DataCleansing
Iterative
IS
LegacySources
13Copyright 2011 Accenture All Rights Reserved.
8/2/2019 Preso Accenture_INFADAY_2011
14/18
Advantages of new layout: Web-based interface Invalid data can be parsed and corrected
Scorecards can be developed by end-users Users can update source files and re-run
controls until the data is good Faster analysis of data quality checking Version 9.0.1 and above are integrated with
PowerCenter
Why Data Quality?
Sample of Data Quality Reports
Score for each KPI
Details of valid / invalid rows
14Copyright 2011 Accenture All Rights Reserved.
8/2/2019 Preso Accenture_INFADAY_2011
15/18
SAP Data Migration Factory: Pre-built Developments:
IDQ PowerCenter
SAP LSMW
DB Repository (MS SQL /Oracle)
Ready-to-use environment: Virtual Machine Configured software (INFA,
DB)
Data Quality (IDQ) Data Integration (PWC)
Why Data Quality?
Key Differentiator: Reusable Components
Installed Code
Documentation
Benefits:1. Reduce cost, time and effort2. Reusable components can
be customized for each client
3. Jump Start developers to focusonly on enhancement and/ordeveloping gaps, if any
15Copyright 2011 Accenture All Rights Reserved.
8/2/2019 Preso Accenture_INFADAY_2011
16/18
Why Data Quality?
SDMF is a complete End to End solution focusing on Quality
People
Accentures Delivery Center
Network (DCN) :-Multiple clients
-Skilled
-Low-Costs
Process
Methodology and Framework:
- Data Quality management
- Conversion Objects in SAP
- Development Framework
ProcessPeople
16Copyright 2011 Accenture All Rights Reserved.
Technology
Reuse components developed by Accenture in SDMF:
-Prebuilt Developments
-Ready-to-use environments
Technology
8/2/2019 Preso Accenture_INFADAY_2011
17/18
8/2/2019 Preso Accenture_INFADAY_2011
18/18
Questions / Comments?
Thank you
18Copyright 2011 Accenture All Rights Reserved.
Xavier Scherpereel
SDMF Lead for EAME
Phone:+33 6 86 59 96 [email protected]
Arturo Salazar
Data Management specialist
Phone:+33 6 32 89 25 [email protected]