Many project managers will surely attest to the fact that managing user expectations is one of the key challenges to delivering a successful data warehouse project. The primary difficulty is in understanding how one should really go about managing expectations and the appropriate start is by setting expectations in the first place.
Below are some areas where expectations are best set up-front:
- The primary purpose of the data warehouse is to help the business community make better decisions by leveraging a single consistently managed corporate data repository
- Data warehouse project delivery will not be a waterfall approach -- multiple releases and an iterative process is the best practice
- User input will be taken for deciding what business objectives are -- they will not be IT's
- Scope, priority and delivery timeline for each iteration
- Selection of reporting and analysis tool
- Data security needs will be addressed as per user requirements or company policy
- Look and feel of reports, charts, graphs, portal, dashboard etc.
- System availability
- Data refresh frequency
- Historical data needs
- The data warehouse data will only be as good as the source system data plus any data cleansing that is done to that data
- Business users will have to define, provide and validate business rules and metric calculations
- Resolving data integrity issues will be a joint responsibility -- IS and business users
- Business users or subject matter experts will need to commit significant time to the project from requirements gathering through design phase and during user acceptance test
- Performance improvement will be an ongoing process, tightly coupled with user query patterns
For more information, check out SearchCRM's Best Web Links on Data Warehousing.
This was first published in March 2002