Design doc is the meta data of the code.
The project management plan is crucial to your project since it is the basis on which the project
will be measured. It can be used to aid in communication between stakeholders and to define
the content and timing of project reviews. Here are some elements that make up a plan:
– Project charter (Completed in the Initiation Phase)
– Scope statement
– Work breakdown structure (WBS)
– Cost estimates, schedule, responsibilities for each deliverable
– Milestones with target dates
– Risk / issues register
– Management plans (scope, schedule, cost, quality, communications, risk, and procurement)
It is important to document the project objectives, deliverables, and require-ments so that they can be used as a basis for future project decisions.
For our Data Mart project, the tasks associated to the deliverables are:
– Conduct workshops to drive requirements with the business analysts
– Document standard report & ad-hoc functional requirements
– Create logical /physical data model
– Write an ETL design document covering the two sources
– Develop project plan
Based on the above tasks, dependencies that could be identified are:
– The data model cannot start until the report & ad-hoc requirements are defined
– The ETL design document cannot start until the physical model is complete
– The data model cannot be complete until the ER software has been procured
keep in mind that the most important component of a BI project is
the data model, so make sure the person or group handling it is experienced and competent.
Effective communication is crucially important during project execution.
(whatever business user and tech lead or tech lead with team member)
Providing constructive feedback to team members is important to getting them to execute
their tasks and deliver according to the project plan while maintaining quality.
Most projects, including BI projects, will use the following communication channels:
Status meetings for team leads: discuss dependencies, tasks, issues.
Status meetings for key stakeholders: review milestones, dates, next steps
Status reports: what has your team accomplished this week, next week, any issues?
The biggest mistake typically seen in BI projects relative to metadata is simply ignoring it. Proper levels of metadata are essential to achieving data quality and a successful BI solution. This activity captures the business requirements that the metadata solution must satisfy.
In general, metadata can be classified into three broad categories—business, technical, and process metadata. The metadata team captures the requirements for each relevant metadata type.
Conduct sufficient data profiling and analysis to validate the feasibility of the business requirements.
logical data flow diagram, source description documents, target description documents, and source-to-target attribute mapping document.
Testers should be involve in the business scope phase.
We should have high level and detailed level ETL designed doc and review.
We should have ETL packages peer review and tech lead review.
We always need ETL audit solution accompany the ETL process, capture the meta data.
We should report your changes and review before check in .
We should run test case to check the business rules on the fly.
We should do more data testing before and after ETL.
We should know more about the data when make cube model.
We should use check the business logic and your code logic with peers.
We always need a checklist and best practices for our delivery.
We need list the summary for the defects and best ideas.
BI project is data driven, if you are not familiar with the data,
You can not make out good data model and cube model.
Understand the data's business meaning is very important.
Understand the data's source and frequency , format.