Data Delivery: total effort is N to M weeks to provide similar data delivery capabilities to what exists today, based on a comparison of the KDW effort. The scalle of the effort can only be com[pared to KDW because there are too many open questions concerning the requirements.
These notes are a result of CIT's brainstorming meetings about Workday on 4/12 and 4/13 2011.
Assumptions
- Workday provides the reports and HR/PY can make more reports straight from Workday to meet operational reporting style needs.
- A warehouse for HR/PY is needed at Cornell as Workday's reports are not expected to meet all the reporting and analytical needs for Cornell.
- 1 year of historical data will be pushed to Workday, and that data is not expected to be readily available for any reporting from Workday software.
- In general, pre-Workday history will be from PS, new data will be from Workday and this means new HR Workdayish structures are needed for the DW.
- There is no documented or expressed strategy for what should be in it or not in the Workday ERA CU HR/PY data warehouse.
- Analysis is needed to discover the requirements for reporting and Business Intelligence for HR and Payroll that won't be met by reports delivered from Workday.
- 'Job' in PS not mapped into workday tables.: Workday rep: 'no equivalent of the job table'
- Time will be required for mapping and transformation of Workday data into a reviewed CU DW model. This has not been investigated in any detail.
- Payroll has expressed that they would like all reports they require that aren't provided by Workday Inc to be written by CIT.
- The HR/Payroll data stewards ,Wes and Lyman, typically drive changes and improvements to datamarts.
- There are no warehouse remediation efforts in HR's Workday plan.
- Some 700 PS tables currently get extracted and put into a staging area of HR/PY datamart.
- Because we need some kind of estimate and we don't know the scope assume it is roughly the size of the KDW effort.
Tasks
- Technical:
- Do some prototyping to discover what techniques for data mapping is possible with Workday.
- Identify any new DW hardware and the budgets required for Workday data delivery work.
- Make a sketch of now and the future in terms of data delivery Hardware/VMs, ETLS to be written, tools, and basic content.
- Research how extracts can be made, scheduled and executed from Workday
- Business requirements
- Create a reporting and analytics strategy for all HR/Payroll data for all campus consumers.
- Analyze requirements - define what analytical reporting on historical data is needed and not provided by Workday.
- Gather impact of changes to HR/Payroll structures consumed downstream of the HR/PY systems and datamart now, such as labor data in KDW.
- Research impact by intersecting discovery table list with the datamart extraction list
- Analyze and decide how provisioning of privileges to access the warehouse would be be accomplished.
Open Questions
- Decide to keep or migrate from de-normalized tables and move completely to a dimensional model and all the advantages it offers.
- Decide on when solutions are needed with respect to the roll out of Workday
- Can we get an specific list of what is in and out of scope from the core Workday project and what is in and out of Scope from a campus reporting and analytical perspective?
- Will functional experts be available to decide what happens when some things come from Workday and some come from PS and the mapping doesn't work well?
- How does the HR /PY reporting plan affect the institutional enterprise reporting in terms of analytics, tools, stores, representation, and governance?