The overarching purpose of a backlog grooming session is to ensure that issues are ready for delivery. When they are deemed as ready, issues are moved to the Ready column.
With this overarching purpose in mind, the activities for each session can differ based on the current understanding of the project or product and its objectives (see Frequency).
Given the activities can alter, it is important that the Product Owner makes the planned activities clear to attendees in the lead up to a scheduled session.
Backlog grooming sessions are scheduled at a fortnightly interval by default.
This cadence can increase or decrease. Frequency is usually assessed based on the amount of existing issues in the Ready column (so as not to groom/estimate too far in advance), or by the needs of an existing project (where new issues need to be created, or existing ones reprioritised).
With the Backlog and Ready columns understood to be a dynamic set of priorities and requirements, the activities within a grooming session may alter to keep these relevant.
Possible activities include:
- Assigning estimates to issues before moving them to the Ready column. This can be coupled with prior investigation (see Work done before the session).
- Collaboratively create new issues based on new project or product needs.
- Assess (or reassess) the priority of issues in the Backlog or Ready column.
- Correcting prior estimates in light of new information.
- Splitting or 'spiking' issues which are deemed too large to move to the Ready column.
It is the product owner's responsibility to make the intended activities of the grooming session clear to the rest of the team ahead of time so any preparations can be made.
Should the Product Owner be presenting issues to the development team for estimating, the issues need to adhere to your Definition of Ready.
The developer should be aware of the current priority of issues as reflected by the order of the Backlog.
Should the backlog grooming session be focusing on estimating, the developer should receive prior notice and review the issues for context, investigate any implementation details, and note findings on the issue.
TODO
TODO