MDM and Data Quality: Not an Option but a Requirement

Welcome!
TITLE

MDM: Quality is Not an Option
but a Requirement

Date: June 12, 2012
Time: 2:00 PM ET
Presenter: Dr. Peter Aiken
Twitter: #dataed

PRODUCED
BY CLASSIFICATION DATE SLIDE
DATA BLUEPRINT 10124-C W. BROAD ST, GLEN ALLEN, VA 23060 EDUCATION 06/12/12 1
06/06/12 © Copyright this and previous years by Data Blueprint - all rights reserved!

TITLE
Commonly Asked Questions
1) Will I get copies of the slides after the
event?

YES*

2) Is this being recorded so I can view it
afterwards?
YES*

*Standard registrations

PRODUCED

TITLE
New Features: Live Twitter Feed & Facebook

Join the conversation on Twitter! www.facebook.com/datablueprint
Follow us @datablueprint and Post questions and comments
@paiken
Find industry news, insightful
Ask questions and submit your content
comments: #dataed
and event updates
PRODUCED

TITLE
Meet Your Presenter: Dr. Peter Aiken
• Internationally recognized thought-leader in
the data management field with more than 30
years of experience
• Recipient of the 2010 International Stevens
Award
• Founding Director of Data Blueprint
(http://datablueprint.com)
• Associate Professor of Information Systems
at Virginia Commonwealth University
(http://vcu.edu)

• President of DAMA International (http://dama.org)
• DoD Computer Scientist, Reverse Engineering Program Manager/
Office of the Chief Information Officer
• Visiting Scientist, Software Engineering Institute/Carnegie Mellon
University
• 7 books and dozens of articles
• Experienced w/ 500+ data management practices in 20 countries
#dataed
PRODUCED

MDM:
Text

Integration with Quality and
Governance is not an
Option but a Requirement

Dr. Peter Aiken: Reference & Master Data Management Webinar
DATA BLUEPRINT 10124-C W. BROAD ST, GLEN ALLEN, VA 23060 EDUCATION 06/12/12

Outline
TITLE

1. Data Management Overview
2. What is Reference and MDM?
3. Why is Reference and MDM
important?
4. Reference & MDM Building
Blocks
5. Guiding Principles & Best
Practices
6. Take Aways, References & Tweeting now:
#dataed
Q&A
PRODUCED

TITLE
The DAMA Guide to the Data Management Body of Knowledge

Published by DAMA
International
• The professional
association for Data
Managers (40
chapters worldwide)
DMBoK organized
around
• Primary data
management
functions focused
around data delivery
to the organization
• Organized around
several
environmental
elements

Data Management Functions
PRODUCED

TITLE
The DAMA Guide to the Data Management Body of Knowledge

Amazon:
http://www.amazon.com/
DAMA-Guide-
Management-Knowledge-
DAMA-DMBOK/dp/
0977140083

Or enter the terms "dama
dm bok" at the Amazon
search engine

Environmental Elements
PRODUCED

TITLE
What is the CDMP?
• Certified Data Management
Professional
• DAMA International and ICCP
• Membership in a distinct group made
up of your fellow professionals
• Recognition for your specialized
knowledge in a choice of 17 specialty
areas
• Series of 3 exams
• For more information, please visit:
– http://www.dama.org/i4a/pages/
index.cfm?pageid=3399
– http://iccp.org/certification/
designations/cdmp
#dataed
PRODUCED

TITLE
Data Management Basics

PRODUCED

TITLE
Manage data coherently.

Data Program
Coordination
Share data across boundaries.
Organizational
Data Integration

Data Stewardship Data Development

Assign responsibilities for data.
Engineer data delivery systems.

Data Support
Operations

Maintain data availability.

PRODUCED

TITLE

PRODUCED

Outline
TITLE

important?
Blocks
Practices
#dataed
Q&A
PRODUCED

TITLE
Summary: Reference and MDM

from The DAMA Guide to the Data Management Body of Knowledge © 2009 by DAMA International
PRODUCED

MDM Definition
TITLE

• Gartner holds that MDM is a
discipline or strategy
– "… where the business and the IT organization work
together to ensure the uniformity, accuracy, semantic
persistence, stewardship and accountability of the
enterprise's official, shared master data."
– Master data is the enterprise's official, consistent set
of identifiers, extended attributes and hierarchies.
– Examples of core entities are:
• Parties (e.g., customers, prospects, people, citizens,
employees, vendors, suppliers and trading partners)
• Places (e.g., locations, offices, regional alignments and
geographies) and
• Things (for example, accounts, assets, policies, products and
services).

PRODUCED

TITLE
Wikipedia: Golden Version
• In software development:
– The Golden Master is usually the RTM
(Released to Manufacturing) version, and
therefore the commercial version. It represents
the development stage of "RTM" (Released To
Manufacturing), often referred to as "going
gold", or "gone golden".
– Often confused with "gold master" which refers
to a physical recording entity such as that sent
to a manufacturing plant.
• In data management:
– It is the data value representing the "correct"
answer to the business question
PRODUCED

TITLE
Reference/Master Data Management
• Definition
– Planning, implementation and control activities to ensure
consistency with a "golden version" of contextual data values.

PRODUCED

TITLE
Definition: Reference Data Management
Control over defined domain values (also known as
vocabularies), including:
• Control over standardized terms, code values and other
unique identifiers;
• Business definitions for each value, business
relationships within and across domain value lists, and
the;
• Consistent, shared use of
accurate, timely and
relevant reference data
values to classify and
categorize data.

PRODUCED

Definition: Master Data Management
TITLE

Control over master data
values to enable
consistent, shared,
contextual use across
systems, of the most
accurate, timely and
relevant version of truth
about essential business
entities.
PRODUCED

TITLE
Reference Data
• Reference Data:
– Data used to classify or categorize other data, the
value domain
– Order status: new, in progress, closed, cancelled
– Two-letter USPS state code abbreviations (VA)

• Reference Data Sets

US United States
GB (not UK) United Kingdom

PRODUCED

TITLE
Master Data
• Data about business entities providing context
for transactions but not limited to pre-defined
values

• Business rules dictate format and allowable
ranges
– Parties (individuals, organizations, customers,
citizens, patients, vendors, supplies, business
partners, competitors, employees, students)
– Locations, products, financial structures

• From the term Master File
PRODUCED

TITLE
Reference Data versus Master Data
• Reference Data:
– Control over defined
domain values
(vocabularies) for
standardized terms,
code values, and
other unique
identifiers
– The fact that we
maintain 9 possible
gender codes
• Master Data:
– Control over master
data values to enable
consistent, shared,
contextual use across
systems
Both provide the context
– The "golden" source
of the gender of your for transaction data
customer "Pat" from The DAMA Guide to the Data Management Body of Knowledge © 2009 by DAMA International
PRODUCED

Outline
TITLE

important?
Blocks
Practices
#dataed
Q&A
PRODUCED

Reference Data Facts 2012
TITLE

• Global industry-wide survey
of reference data
professionals
• Results show: Poor quality of
reference data continues to
create major problems for
financial institutions.
• Home-grown reference data solutions
predominate, putting institutions at risk for meeting
regulatory constraints
• Risk management is seen as a more important
business driver for improving data quality than
cost
Source: http://www.igate.com/22926.aspx
PRODUCED

TITLE
Reference Data Facts 2012, cont’d
• Despite recommended practices of centralizing
reference data operations, 31% of the firms
surveyed still manage data locally
• New and changing regulatory requirements have
prompted many financial service companies to re-
evaluate their reference data strategies. To prepare
for new regulations,
nearly 62% of survey
respondents are planning
to extend or customize
their reference data
systems during 2012 and
2013.
Source: http://www.igate.com/22926.aspx
PRODUCED

TITLE
Polling Question #1
What percentage of firms still manage data locally?

1. 75%
2. 31%
3. 55%
4. 24%

PRODUCED

TITLE
Interdependencies

Data Governance

Data Quality Master Data

PRODUCED

TITLE
Inextricably intertwined

Organized Knowledge 'Data' Knowledge
Management
Practices
Routine Data Scans Data Organization Practices
Metadata(Prac8ces((dashed lines not in existence)
( (

Metadata( Metadata( Metadata(
Engineering( Storage( Delivery(
Sources( ( Uses(
Metadata(Governance(

Data that might benefit from
Suspected/ Master Management
Identified
Master Data Catalogs
Data
Quality Master Data
Management
Problems Data Quality Practices
Engineering

Routine Data Scans
Improved Quality Data
Operational Data

PRODUCED

TITLE
Interactions
Data
Harvesting

Governance Master
Violations Data
Monitoring Data Monitoring
Quality
Routine Monitoring
Data
Scans Governance
Rules
Monitoring
Monitoring Results:
Master
Monitoring Results: Suspected/ Data
Rules Suspected/ Master
Catalogs
Identified Data &
Data Characteristics
Data Quality
Governance Problems Data
Practices Quality
Rules
Quality Focused
Rules Data Quality Master Data Data
Engineering Management Scans
Routine Practices Practices
Data
Scans

Improved Quality Data

Operational Data
PRODUCED

TITLE
Multiple Sources of (for example) Customer Data
Finance Application
(3rd GL, batch
Payroll Data system, no source)
(database)
Payroll Application Finance
(3rd GL) Data
(indexed)

Marketing Data Marketing Application
(external database) (4rd GL, query facilities,
no reporting, very large)

Personnel Data
(database)

Personnel App.
(20 years old,
un-normalized data) Mfg. Data
R&D
Data (home grown
(raw) database) Mfg. Applications
(contractor supported)
R& D Applications
(researcher supported, no documentation)
PRODUCED

TITLE
Reference/Master Data: Tank, Tanks, Tankers, Tanked

PRODUCED

TITLE
Reference Data Architecture


PRODUCED

TITLE
Master Data Architecture

PRODUCED

TITLE
Combined R/M Data Architecture

PRODUCED

TITLE
"180% Failure Rate" Fred Cohen, Patni

PRODUCED
BY http://www.igatepatni.com/bfs/solutions/payments.aspx
CLASSIFICATION DATE SLIDE

TITLE
MDM Failure Root-Causes
• 30% of MDM programs are regarded as failures
• 70% of SOA projects in complex, heterogeneous environments had
failed to yield the expected business benefits unless MDM is
included
• Root-causes of failures:
– 80% percent of MDM initiatives fail because of ineffective
leadership, underestimated magnitudes or an inability to deal
with the cultural impact of the change
– MDM was implemented as a technology or as a project
– MDM was an Enterprise Data Warehouse (EDW) or an ERP
– MDM was an IT Effort
– MDM is separate to data governance and data quality
– MDM initiatives are implemented with inappropriate technology
– Internal politics and the silo mentality impede the MDM
initiatives
PRODUCED

Outline
TITLE

important?
Blocks
Practices
#dataed
Q&A
PRODUCED

TITLE

Automating Business Process

Benefits Results
• Obtain holistic perspective on • Speed up process
roles and value creation • Cost savings
• Customers understand and • Increased compliance
value outputs
• Increased output
• All develop better shared
understanding • IT systems documentation

PRODUCED

TITLE

Traditional Engine

PRODUCED

TITLE
Prius Hybrid Engine

PRODUCED

TITLE

PRODUCED

TITLE
Data Security Management Overview

ü ü ü ü ü ü ü
ü ü ü ü ü ü ü
ü ü ü ü ü ü ü
ü ü ü ü ü ü ü
ü ü ü ü ü ü ü

Illustration from The DAMA Guide to the Data Management Body of Knowledge p. 37 © 2009 by DAMA International
PRODUCED

TITLE
Goals and Principles
1. Provide authoritative
source of reconciled,
high-quality master and
reference data.
2. Lower cost and
complexity through reuse
and leverage of
standards.
3. Support business
intelligence and
information integration
efforts.
PRODUCED

TITLE
Reference & MDM Activities
• Understand Reference and
Master Data Integration Needs
• Identify Master and Reference Data
Sources and Contributors
• Define and Maintain the Data
Integration Architecture
• Implement Reference and Master
Data Management Solutions
• Define and Maintain Match Rules
• Establish “Golden” Records
• Define and Maintain Hierarchies and Affiliations
• Plan and Implement Integration of New Data Sources
• Replicate and Distribute Reference and Master Data
• Manage Changes to Reference and Master Data
PRODUCED

TITLE
Specific Reference and MDM Investigations

• Who needs what information?

• What data is available from
different sources?

• How does data from different
sources differ?

• How can inconsistencies
be reconciled?

• How should valid values be shared?
PRODUCED

TITLE
Primary Deliverables
• Data Cleansing Services
• Master and Reference
Data Requirements
• Data Models and Documentation
• Reliable Reference and Master Data
• "Golden Record" Data Lineage
• Data Quality Metrics and Reports
PRODUCED

TITLE
Roles and Responsibilities
Suppliers: Consumers:
• Steering Committees • Application Users
• Business Data Stewards • BI and Reporting Users
• Application Developers and
• Subject Matter Experts Architects
• Data Consumers • Data integration Developers and
• Standards Organizations Architects
• Data Providers • BI Vendors and Architects
• Vendors, Customers and Partners
Participants:
• Data Stewards
• Subject Matter Experts
• Data Architects
• Data Analysts
• Application Architects
• Data Governance Council
• Data Providers
• Other IT Professionals
PRODUCED

TITLE
Technology
• ETL
• Reference Data Management
Applications
• Master Data Management
Applications
• Data Modeling Tools
• Process Modeling Tools
• Meta-data Repositories
• Data Profiling Tools
• Data Cleansing Tools
• Data Integration Tools
• Business Process and Rule Engines
• Change Management Tools
PRODUCED

Outline
TITLE

important?
Blocks
Practices
#dataed
Q&A
PRODUCED

TITLE
Guiding Principles
1. Shared R/M data belong to
the organization.
2. R/M data management is an
on-going data quality improve-
ment program – goals cannot
be achieved by 1 project alone.
3. Business data stewards are the authorities
accountable at determining the golden values.
4. Golden values represent the "best" sources.
5. Replicate master data values only from golden
sources.
6. Reference data changes require formal change
management
PRODUCED

TITLE
Polling Question #2
Which of these is a MDM best practice?

1. Build your processes to be static
2. Customize whenever possible
3. Use a holistic approach
4. Test at least once

PRODUCED

TITLE
10 Best Practices for MDM
1. Active, involved executive
sponsorship
2. The business should own the data
governance process and the MDM or
CDI project
3. Strong project management and
organizational change management
4. Use a holistic approach - people,
process, technology and information:
5. Build your processes to be ongoing
and repeatable, supporting
continuous improvement
Source:http://www.mdmsource.com/master-data-management-tips-best-practices.html
PRODUCED

TITLE
10 Best Practices for MDM, cont’d
6. Management needs to recognize the
importance of a dedicated team of
data stewards
7. Understand your MDM hub's data
model and how it integrates with your
internal source systems and external
content providers
8. Resist the urge to customize

9. Stay current with vendor-provided
patches
10. Test, test, test and then test again.
Source:http://www.mdmsource.com/master-data-management-tips-best-practices.html
PRODUCED

Outline
TITLE

important?
Blocks
Practices
#dataed
Q&A
PRODUCED

TITLE

15 MDM Success Factors
1. Success is more likely and 5. Create a governance
more frequently observed framework to ensure that
once users and prospects individuals manage master
understand the limitations and data in a desirable manner.
strengths of MDM. 6. Strong alignment with the
2. Taking small steps and organization's business vision,
remaining educated on where demonstrated by measuring
the MDM market and the program's ongoing value,
technology vendors are will will underpin MDM success.
increase longer-term success 7. Use a strategic MDM
with MDM. framework through all stages
3. Set the right expectations for of the MDM program activity
MDM initiative to help assure cycle — strategize, evaluate,
long-term success. execute and review.
4. Long-term MDM success
requires the involvement of
the information architect.
PRODUCED
DATA BLUEPRINT 10124-C W. BROAD ST, GLEN ALLEN, VA 23060 [Source:

unknown]06/12/12
EDUCATION 55

TITLE
15 MDM Success Factors
8. Gain high-level business 12. Get the business to propose
sponsorship for the MDM and own the KPIs; articulate
program, and build strong the success of this scenario.
stakeholder support. 13. Measure the situation before
9. Start by creating an MDM and after the MDM
vision and a strategy that implementation to determine
closely aligns to the the change.
organization’s business 14. Translate the change in
vision. metrics into financial results.
10. Use an MDM metrics 15. The business and IT
hierarchy to communicate organization should work
standards for success, and to together to achieve a single
objectively measure view of master data.
progress.
11. Use a business case
development process to
increase business
engagement.
PRODUCED

TITLE
Seven Sisters from British Telecom

PRODUCED
Thanks
to
Dave
Evans

TITLE
Summary: Reference and MDM

PRODUCED

TITLE
Questions?

+ =

It’s your turn!
Use the chat feature or Twitter (#dataed) to submit
your questions to Peter now.

PRODUCED

TITLE
References

PRODUCED

TITLE
Additional References
• http://www.mdmsource.com/master-data-management-tips-best-practices.html
• http://www.igate.com/22926.aspx
• http://www.itbusinessedge.com/cm/blogs/lawson/just-the-stats-master-data-
management/?cs=50349
• http://searchcio-midmarket.techtarget.com/news/2240150296/Smart-grid-systems-
expert-devises-business-transformation-template
• http://www.itbusinessedge.com/cm/blogs/lawson/free-report-shows-businesses-fed-up-
with-bad-data/?cs=50416
• http://www.itbusinessedge.com/cm/blogs/lawson/whats-ahead-for-master-data-
management/?cs=50082
• http://www.itbusinessedge.com/cm/blogs/vizard/master-data-management-reaches-
for-the-cloud/?cs=49264
• http://www.information-management.com/channels/master-data-management.html
• http://www.dataversity.net/applying-six-sigma-to-master-data-management-mdm-
framework-for-integrating-mdm-into-ea-part-2/
• http://www.dataqualityfirst.com/getting_master_data_facts_straight_is_hard.htm

PRODUCED

TITLE
Upcoming Events
July Webinar:
Practical Applications for Data Warehousing,
Analytics, BI, and Meta-Integration Technologies
July 10, 2012 @ 2:00 PM ET/11:00 AM PT

August Webinar:
Your Documents and Other Content:
Managing Unstructured Data
August 14, 2012 @ 2:00 PM ET/11:00 AM PT
Sign up here:
• www.datablueprint.com/webinar-schedule
• www.Dataversity.net
Brought to you by:

PRODUCED

MDM and Data Quality: Not an Option but a Requirement

More Related Content

What's hot (6)

Viewers also liked (9)

Similar to MDM and Data Quality: Not an Option but a Requirement (20)

More from DATAVERSITY (20)

Recently uploaded (20)

MDM and Data Quality: Not an Option but a Requirement