The Data Warehouse Development Life Cycle
Warehouse Project Management
Basic Project Management
As a general definition, a project is any set of tasks with a
specific objective to be completed within certain specifications
(including defined start and end dates) that consumes capital
resources. Given this simplistic definition of a project, let’s
define what project management is and how it applies to a data
For every large data warehouse project, traditional management must
be replaced by a new type of management that is temporary and very
flexible, with a fast reaction time, and able to respond rapidly to
both internal and external changes. With this type of management in
place, data warehouse project management encompasses the following
* Defining work requirements
* Defining the quantity of work
* Defining the resources needed
* Monitoring the project by:
* Tracking progress (dates and milestones)
* Comparing actual figures to predicted figures
* Analyzing the impact of changes
* Making adjustments to the project
While these tasks may seem mundane, effective project management is
critical to the success of a data warehouse. Successful project
management is defined as meeting the objectives of a project within
project and cost constraints, while maintaining a desired level of
performance and fully utilizing the proper technology.
To effectively fulfill the project management functions listed
previously, data warehouse project managers must be able to:
* Identify function responsibilities and ensure that all activities
are accounted for.
* Minimize the need for continuous reporting.
* Identify the time limits for scheduling.
* Identify a methodology for tradeoff analysis
* Measure the project accomplishments against the plans.
* Identify and resolve problems quickly.
* Improve estimation capabilities for future planning.
* Keep track of meeting project objectives.
Data warehouse project management is different from traditional
management in several ways. First, while the evolution of data
warehouse queries may be perpetual, the initial creation of the
warehouse is a finite activity, and the project manager must be able
to deal with this temporary authority because other managers are
performing the staffing functions, supplying members of the data
warehouse team. To further confound matters, the data warehouse
project manager does not have direct control over the financial