What to include in the BI Definition of Ready
In an online forum, someone recently asked:
- What’s in your Definition of Ready?
- When does the ‘ready’ criteria apply? — i.e. ready by the end of Sprint Planning (How) or ready for the beginning of Sprint Planning?
Here are the thoughts I shared with that group.
In the world of business intelligence the biggest problems I’m seeing are lack of clear acceptance criteria, such as a specific example of a complex transformation so the team can really see what the PO wants to achieve. This doesn’t mean a COMPLETE calculation with 100,000 records – rather, the smallest number of records necessary to demonstrate the desired result. Behavior-driven development (BDD) can be really useful for articulating data-related acceptance criteria in the business language. Also see Acceptance Criteria in BI User Stories for more details.
The PO should do his/her best to have complete acceptance criteria by Sprint Planning. Ideally, the PO has worked with the team in grooming leading up to sprint planning, so the team doesn’t feel like the PO is dumping unclear requests on them. However, a key part of sprint planning is identifying which stories are ready for the team to start working, and which are not, and as long as there’s always an option to send a story back to the PO for more clarity before the team makes a commitment to delivering it, then it’s ok if the story isn’t “ready” at that point. POs who REALLY need a story to be worked on NEXT will figure out how to have their stories READY pretty quickly!
The team retrospective should bring out any learning opportunities to improve the Definition of Ready each sprint – so don’t worry if it’s not “perfect” now – it will get there with regular review and improvement!
What else could be included in this discussion, for BI teams?