Best Practices in Implementing a Corporate Information Factory

1
Best Practicesin Implementing aCorporate
Information Factory

• 3/19/2002, Steve Westmoreland

2
(No Transcript)
3
Data Warehouse is Needed

• What does that look like?
• Who is it for?
• Workgroup?
• Department?
• Division?
• Entire organization?

4
Data Warehouse is Needed

• Whats first?
• What are DW best practices?
• Whats essential?
• Whats optional?
• Is there a standard approach?

5
Whats the DW Look Like?

• Basic Structure

6
Whats the DW Look Like?

• Basic Corporate Information Factory

7
Whos Driving

• Avoid the DW project as an IT project
• DWs are strongly business driven
• If DW is an IT project, create strong association
  with the business
• If DW is a business project, create a strong
  working relationship with IT

8
Project Justification

• Business need
• Logical need
• Make case for why so difficult to get answers
  from existing transactional systems

9
Sponsorship

• Get commitment for success
• Dedicated staff (new positions)
• Time
• Quality
• Commitment to lessons learned

10
Educate

• DW Fundamentals
• Basic Project Management
• Incremental Development Concept
• Difference between
• Transactional Dimensional (Subject)
• Business specific objectives deliverables

11
Educate Who?

• Sponsors
• Stakeholders
• Peers, spheres of influence
• Knowledge workers
• Core team
• SMEs SSEs
• Consultants

12
Visioning

• Describe what it looks like with a DW
• The organization
• Impacts
• Peoples work habits
• Quick, valid answers
• The operational systems
• The development production teams
• Use pictures wherever possible

13
Start with this
14
and morph to this
15
Visioning

• Ideal vs. Practical
• Ideal
• Fully funded
• All Roles Responsibilities filled
• 100 sustained sponsor support
• Practical
• Adequate funding, R R

16
Team Makeup

• Two Tenets
• Field key positions with in-house staff - dont
  field these positions with consultants
• Get help, use experienced consultants

• Solution
• Match key roles 11 wherever possible ( scale
  down from there as required)

17
Team Makeup

• Data Warehousing Roles
• Project Director
• Project Manager
• Business Requirements Analyst
• Data Architect
• DBA
• Acquisition Architect
• Technical Architect

• Meta Data Architect
• Access Architect
• Data Quality Architect
• DW Administrator
• Training Documentation Developer
• Testing Architect

18
Team Makeup

• Cover key roles for development
• Try hard for a dedicated team
• Recommended
• Business Requirements Analyst
• Data Architect
• DBA
• Acquisition Architect
• Technical Architect
• Meta Data Architect
• Access Architect

19
Team Makeup

• Development Team ? Production Team
• Be prepared for production
• DW Administrator
• Business Requirements Analyst
• Data Quality Architect
• Acquisition Support
• Access Support
• Meta Data Support
• Training Documentation Developer
• Testing Architect

20
Team Style

• Collaborative
• vs.
• Independent

21
Methodology

• Generic Project Plan
• Generic Schedule
• Roles Responsibilities
• Management Plan
• Documentation
• Schedule
• Deliverables

22
Project Management

• Scope
• Overall project each increment
• Risk
• Anything not as planned is a risk
• Roles Responsibilities
• Keep team focused, collaborative
• Communication
• Keep consistent energy here
• Time
• Use schedule as communication tool
• Quality
• Management product of the DW
• Cost control manage
• Procurement plan processes

23
Risk Management

• Readiness Assessment inverted
• Obvious risks are easy
• Not so obvious risks are harder
• Expectations
• Success
• Some risks repeat with every increment
• Risk occurs when moving from development to
  production
• Dont forget potential mitigations

24
Marketing

• Early value (vs. application system )
• Continual education
• Value stories
• Use communication plan
• Look for extensions into other areas
• Continual marketing

25
Clarity of Increment Sizing

• Achieve clarity of increment scope early
• Early notion of increments are usually subject
  areas
• Subject areas usually contain multiple increments

26
Build Success

• First two to four increments should build off of
  one another.
• This allows demonstrable value
• 2 increases value of 1
• 3 increases value of 2
• 4 increases value of 3
• and so on
• Improves teams confidence

27
Success Factors

• You will have success issues be ready
• First, few value what youre doing
• Second, everyone wants to be next
• Third, pressures mount to do more faster

28
Value ROI

• One of the most difficult is ROI
• Be creative
• Surveys of users value of data
• Time is money
• Save time getting to data
• Save time getting to decisions
• Improve the productivity of Knowledge Workers
  time
• Uncover Value Stories

29
Really Good Advice

• Pay attention to Lessons Learned
• Use consultants to augment in-house staff
• Dont understaff
• Prepare early for production with a separate
  team
• Always think with Vision
• What does our DW look like two years from now?
  How is it staffed? Who are our sponsors? What
  does our business activity look like two years
  from now with the advantage of data warehousing?

30
Recap

• Know what data warehousing is market it, people
  want to know
• Plan the Project market it
• Justify market it
• Get set up lots to do
• Get out of the way let your team work
• Market, market, market