4.5 Stage 5 – Data Governance
4.5.1
Overview
Question : What is Data Governance ?
Data Governance can be
defined simply as ‘Doing things right’ in Enterprise Data Management by
complying with the appropriate rules, policies and procedures.
These will all be designed to make sure that
data used throughout the
Question : How do I get a top-down view of Data Management in my organisation ?
Answers to this question are
at different levels :-
- Data Governance at the top-level
- Master Data Management at the mid-level
- Data Integration at the mid-level
- Data Owners and Sources at the lowest level
- Information Catalogue mandated as the central
repository of all this information
- Appropriate procedures in place
to control all of these factors.
Question : Why should my organisation have a Data
Governance function ?
The existence of
a Data Governance function is a measure of the maturity of Data Management
within an organization
The first steps
should be to establish a thin slice of Data Governance from top to bottom
- Wikipedia on Data Governance - http://en.wikipedia.org/wiki/Data_governance
- Alignment of
If you are active in this
area, you should consider joining a professional organizational.
This helps
you to network with your peer group and will encourage you to keep up-to-date
in knowledge and professional practice.
Here are two organisations that are planning active roles in Data Governance :-
i) The Data Governance Institute (Membership starts at
$150 for individuals) :-
http://www.datagovernance.com/
ii) The Data Governance and Stewardship Community of
Practice ($150/year) :-
- http://www.datastewardship.com/
It includes coverage of some very useful Case Studies :-
http://www.datastewardship.com/content.aspx?page_id=22&club_id=885168&module_id=37956
It also maintains a Data
Governance Software Web Site :-
http://www.datagovernancesoftware.com/
and Sarbanes-Oxley Web Site - http://www.sox-online.com/
4.5.2
Best Practice
Successful Data Governance
requires successful completion of many Tasks, most of which are not technical
and all of which are complex and challenging.
The process of initially
establishing Data Governance will always take much longer than anticipated.
Therefore, management of expectations is very important.
It is necessary to aim for
collaboration and by-in from the start.
This buy-in must be top-down
and bottom-up.
The top-down buy-in manifests
itself by, for example, a realization that ‘Data Quality is an Enterprise
Issue’.
The bottom-up buy-in is important
because quality of Source data is critical and must be good, otherwise any Data
Governance issue will fail.
Auditing, for example, of
database changes, is a basic requirement.
This Weblink
for the Data Governance Institute is relevant :-
- http://datagovernance.com/dgi_framework.pdf
4.5.3
Templates
Templates for Data Governance
include standard documents.
These documents can be made available as Forms over the Internet.
They will control all aspects
of the planning, management and control of Data Management.
Compliance will
be ensured with overall
4.5.4
Tools
Tools are those required to
manage Templates and include Excel Spreadsheets and Databases..
4.5.5
Tutorials
Here is a step-by-step
Tutorials for Data Governance :-
Step 1. Determine the principles and goals of Data
Governance.
For example ‘Data Quality is
an Enterprise Issue’.
Step 2. Clarify Roles and responsibilities, both top-down and
bottom-up.
Step 3. Obtain agreement on how success will
be measured.
Step 4. Determine Data Quality standards at all phases in ETL
Step 4. Avoid a demanding schedule.
Step 5. Establish a spirit of collaboration and ownership of
a common problem.
This will help obtain timely resolution of all
issues.
This Presentation by Barry
Williams on Data Standardisation in Public Sector is
an excellent starting-point-
- http://www.databaseanswers.org/presentations/data_standards_in_public_sector.ppt
Here’s another useful Presentation by Barry on a Strategy
for Information Mgt -
- http://www.databaseanswers.org/presentations/info_mgt_strategy.ppt
4.5.6
How do I ?
4.5.6.1 Get buy-in from
key Stakeholders ?
It is essential to get key
Stakeholders involved and one good way to achieve this is by arranging
Facilitated Workshops. Where necessary, you can supplement these with one-on-one
discussions between you or your team and specific individuals who have concerns
that are best resolved before holding Workshops.
4.5.7 Qualities for Success in Data Governance
To be successful in Data
Governance it is important to be comfortable working with senior management and
to have a clear idea of what is achievable in large organizations and how to
establish agreement over overall goals for an enterprise.
Someone who works in this
area is happy to work with senior management and will understand just enough
about technology to make an effective contribution and to ensure that the role
of technology is used to the maximum effect.