Data Governance from a Strategic Management Perspective

Data Governance from a
Strategic Management Perspectiveg g p
Prof. Dr. Boris Otto, Assistant Professor
Berlin, February 16, 2012

AgendaAgenda
 A strategic view of Data Governance A strategic view of Data Governance
 Data Governance as dynamic capability
 Examples from the consortium
© BEI St. Gallen – Berlin, February 16, 2012, Dr. Boris Otto / 2

AgendaAgenda

Data Governance is necessary to respond to a numberData Governance is necessary to respond to a number
of external business requirements
1 Customer-Centric Business Models
$$ Value Chain Excellence
§ Contractual and Regulatory Compliance

Data Governance and Data Quality Management areData Governance and Data Quality Management are
closely interrelated
Maximize
Data Quality
Maximize
Data Value
is sub-goal of
D D Q li
supports supports
is led by is sub-function
Data
Governance
Data Quality
Management
Data
Management
is led by is sub-function
of
Data Assets
are object of are object of
are object of
Data Assets
Legend: Goal Function Data.
Source: Otto, B.: Data Governance, in: WIRTSCHAFTSINFORMATIK, 53, 4, 2011, S. 235-238.
, , , , , ,

Data Governance effectiveness still varies widely todayData Governance effectiveness still varies widely today
25 07.5
7.5
25.0
30 0
very good
30 0
30.0
y g
good
di30.0 mediocre
adequate
poor
Source: Messerschmidt, M.; Stüben, J.: Verborgene Schätze: Eine internationale Studie zum Master-Data-Management,
PricewaterhouseCooopers AG, 2011, pp. 25 et seq.
NB: Figures are percentages.

What issues does upper management see with regardWhat issues does upper management see with regard
to Data Governance? The case of Syngenta
 Business benefitsus ess be e ts
– “Keep in mind to balance costs for double-handling on one hand and of high
discipline on the other.”
“Emphasize usability of MDM its value ”– Emphasize usability of MDM, its value.
 Organizational readinessg
– “Data owners and data stewards are terms people don‘t understand. Be
educational and promotive.”
– “Organizational maturity differs in the divisions ”Organizational maturity differs in the divisions.
 Data Governance implementation and execution
– “What’s the migration path? Are there intermediate staging gates?”
– “Is it a journey or can one make a choice? Or both?”
– “How to integrate this strategy into the program of next year?”How to integrate this strategy into the program of next year?
– “How to integrate the 35,000 ft view with daily operations?”
NB: Selected quotes from a series of eight interviews with line managers conducted in October and November 2011.
q g g

AgendaAgenda

Data Governance, what is it really?Data Governance, what is it really?
… a business process?
an organizational unit?… an organizational unit?
… just a persistent management fad?

Data Governance as a Dynamic CapabilityData Governance as a Dynamic Capability
Dynamic capabilities describe an enterprise’s ability to address a changing market environment by
integrating, reconfiguring, gaining, and releasing resources1
1) Eisenhardt K M and Martin J A Dynamic Capabilities: What Are They? Strategic Management Journal 21 10-11 (2000) 1105-1121
1) Eisenhardt, K.M. and Martin, J.A. Dynamic Capabilities: What Are They? Strategic Management Journal, 21, 10-11 (2000), 1105-1121.

Companies start from different Data ManagementCompanies start from different Data Management
positions
2
2
1 5
6
D t Q lit M t
Data Strategy Management
2
2
1
6
5
Data Stewardship
Data Quality Management
1 2 3 2Data Lifecycle Management
1 2 1 4
DataArchitecture
Management
7 1
Database Operations
Management
Does not exist in comprehensive form
Existed prior to Data Governance in similar form
Absolute numbers.
Existed prior to Data Governance in similar form
Existed prior to Data Governance, but was significantly revised or extended
Newly created
NB: Based on data from eight cases (Bayer CropScience, Corning Cable Systems, DB Netz, Deutsche Telekom, Johnson & Johnson, Robert
B h S t ZF)
Bosch, Syngenta, ZF)

Note taken in a meeting with Johnson & JohnsonNote taken in a meeting with Johnson & Johnson
on November 29, 2011, in Skillman, NJ

The ideal lifecycle of Data Governance as a dynamicThe ideal lifecycle of Data Governance as a dynamic
capability resembles an “S” curve
E
Founding Phase „First Time Right“ Cleansing
E
Legend: E − Effectiveness; A − Amount of Activity. A

AgendaAgenda

It is not a perfect world, thoughIt is not a perfect world, though
E E E
3.
Pattern I Pattern II
4.
5. Pattern IIIE E E
2.
1.
2.
3.
1. 2.
3. 4.
2008 2009 2010 2011
1.
2008 2009 2010 2011 2007 2008 2009 2010
A A A
2008 2009 2010 2011
1. CDM unit launched
2. Data creation workflow
3. DQ metrics launched
2008 2009 2010 2011
1. DG project launched
2. Address to board
3. DQ metrics launched
4. „Community“ approach
2007 2008 2009 2010
1. CDM unit launched
2. Progress report to the board
proposed
3. Inventory data quality4. „Community approach
proposed
5. DG council launched
3. Inventory data quality
assessment
4. CDM reorganized
Legend: E − Effectiveness; A − Amount of Activity; CDM − Corporate Data Management; DQ − Data Quality; DG −
Data Governance.

Double-loop learning is a central success factor forDouble loop learning is a central success factor for
Data Governance maturity
“Problems cannot be solved
by the same level of thinkingby the same level of thinking
that created them.”

“We really need to change the way Data Governance and
Data Management are perceived throughout the companyData Management are perceived throughout the company.
Usually people do not welcome me with open arms when I
enter their office Data management is somehow treated likeenter their office. Data management is somehow treated like
a skunk—no-one wants to spend too much time with it. It
would be very important that we change the image of thewould be very important that we change the image of the
issue. What would have a more positive connotation than a
skunk? Maybe a squirrel!” Karl-Heinz Weber, Bayer CropScience AGskunk? Maybe a squirrel! Karl Heinz Weber, Bayer CropScience AG

Options for novel approaches: The case of CorningOptions for novel approaches: The case of Corning
Cable Systems
1
 Align MDM with the company’s culture of quality management
 Proof of concept for customer master data creation in NAFTA
customer master life cycle
Transferring
TQM principles to MDM
1
 Master data as an asset
 Establish business-oriented data quality metrics
 Data life cycle / Retirement process
Managing cost and
value of master data
2
 Data life cycle / Retirement process
 Buy-in for MDM organization from data owners still missing
 Continuous roll-out of roles and responsibilities in MDM
Global
Data Governance rollout
3
 Implementation of a shared MDM Service
Data Governance rollout
 Knowledge capitalization on an organization and system levelGlobal leveraging of
4
g p g y
 Foundation of a global center for excellence
g g
knowledge assets
5
 Technical integration/substitution of ASI, Windchill with SAP
 Extend workflow from material master to other domains
System integration and
process automation

Existing Data Management expertise is a crucial factor for identifying aExisting Data Management expertise is a crucial factor for identifying a
need for action during the establishment of Data Governance

The maturity model is an instrument for controllingThe maturity model is an instrument for controlling
Data Governance effectiveness
Strategy
Controlling
Organization
Applications
Data
Processes
& Methods
Data
Architecture
Legend: Current value 2010
Target value 2011 (= one maturity level for all enablers)

RWE benefits from EUR 2.6 million costs avoidedRWE benefits from EUR 2.6 million costs avoided
Average cost rate Number of Number of
Value = per master data
record*
X ( duplicate
records
+ new records
avoided
)
Backward Forward
* 2,861 EUR (2006).
Source: Holzapfel, T. Harmonisierung Von Materialstammdaten - Eine Herausforderung. Proceedings of the Stammdaten-Management
Forum (Frankfurt 2007 09 27) IIR Deutschland 2007
Forum (Frankfurt, 2007-09-27). IIR Deutschland, 2007.

Some more tangible benefit examplesSome more tangible benefit examples
Savings of 2 percent of actual value in stockSavings of 2 percent of actual value in stock
More than GBP 500 million saved through retrieval of
“lost assets”
More than GBP 500 million saved through retrieval of
“lost assets”
CHF 3,000 saved per obsolete master data recordCHF 3,000 saved per obsolete master data record

Contact DetailsContact Details
Prof. Dr. Boris Otto
Assistant Professor
University of St. Gallen
Boris.Otto@unisg.ch
Tel.: +41 71 224 32 20
http://cdq.iwi.unisg.ch

Data Governance from a Strategic Management Perspective

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to Data Governance from a Strategic Management Perspective

Similar to Data Governance from a Strategic Management Perspective (20)

More from Boris Otto

More from Boris Otto (20)

Recently uploaded

Recently uploaded (20)

Data Governance from a Strategic Management Perspective