|
This is the phase when the idea of Data Warehouse takes shape and leads to the initiation phase of the project.
Data Warehouse project trigger
Before the project definition phase, there is a phase of Data Warehouse trigger. This is the typical journey. The Data warehouse (or Data Mart) Trigger could be coming from various reasons:
- A severely felt information handicap
- Information infrastructure too scattered and is difficult to manage.
- A pro-active plan to manage the growth
- Competition has acquired a sophisticated Data Warehouse infrastructure.
- Management team or functional team is introduced to this concept by Vendors or a Senior Executive as an opportunity
Finally, one representative each is identified from business as well as IT as respective champions for their function. A small team is formed to do initial assessment for the Data Warehousing project.
Assess Data Warehousing Readiness
Through-out the data warehouse project management chapter, you will find the links to the pages, covering that subject area in detail. As you start building your scope and approach for the data warehouse (or data mart), one has to assess the readiness of the organization. Refer Organization Readiness for Data Warehousing
Develop Preliminary Project Scope for Data Warehouse
This is an important stage. You can refer DW project scoping. At this stage, one can work-out the following:
- Broad list of business themes the Data warehouse will fulfill.
- The applications that you want to use the Data Warehouse for. The scope of data warehouse will change depending upon if you want to have it for enterprise reporting or high level analytics. (Refer tip Data Warehouse is not only for analytics)
Establish high-level Data Warehouse Project plan and estimates
At this stage, you will scope on the following lines and do high-level cost estimates accordingly:
- Decide on if you are going to go for a data mart or an enterprise level data warehouse.
- The high level list of cubes, their purpose and which business theme they will fulfill.
- The broad time it will take to implement the project. Take some high level views from the project vendors.
- The time it will take to do Extraction, transformation and loading (which is the biggest plumbing work in DW project).
- Do you need to divide this whole project into multiple standalone sub-projects?
- Will you be doing a pilot or go for the project straight-a-way.
TIP- There is not harm to spend few dollars and get an independent expert/consulting company to help you. In spite of all the disclaimers you put to the management team, your figures at this stage will need to carry some level of credibility. For example- Once you go for detailed business requirements stage, your final figures cannot be 300 % (say) of what you have estimated now.
Build Business Justification for Data Warehouse
Refer business intelligence business case, to get some ideas on how you can build business intelligence (and of data warehouse as its key component).
TIP- As you develop the business case on the data warehouse, please be conservative on the returns. As mentioned by many BI vendors, there have been instances of over 100% ROI on BI initiatives. This may go for few BI initiatives and the ones who were able to achieve all the business objectives. However, the success rate of BI initiatives has been low. Many BI projects end up achieving partial business objectives.
Identify the important resources
You will be talking to respective functions to assign the resources, who will be working on the data warehouse project. This will be perhaps more difficult stage. You do not have much to show at this point, and you will be asking for high-quality resources (which are scarce and over-deployed assets)
Get approval for Data Warehouse Initiation Phase
After the business case, high level project approach and project scope are done, its time to go to the sponsors for next phase. The initiation phase is essentially the stage of getting business requirements to come out with more detailed project plan and estimates.
|
