JIRA triage is an ongoing process and critical to maintaining a healthy backlog. A healthy backlog is one that represents planned work with relative priorities and a general sense of timing. It’s never perfect, but by triaging regularly, we can communicate with stakeholders about decisions, status, and overall progress against the planned roadmap in a systematic way, increasing confidence and clarity across the organization. Here are the areas of focus I used when assessing our backlog health, our relative velocity across teams and projects (note we currently don’t do estimation, so ticket count is our approximation), and the status of currently prioritized work, as well as some useful articles from Atlassian on how to do this.
Assessing the Backlog Health - Response and Resolution Time
https://www.atlassian.com/agile/scrum/backlogs
https://sagebionetworks.jira.com/secure/Dashboard.jspa?selectPageId=11106
The first two charts on this dashboard look at the weekly count of created vs. resolved tickets, and the time to first response for each week. Use the first chart to assess how many tickets were created that week, and response to them w/ comments. Goal is to make a near-term decision about whether the ticket (a) should be done as-written now, (b) should be prioritized for later and/or needs clarification to make the call, or (c) isn’t something we should do. Add any metadata (components, labels, epics) to the tickets as you need to so that you can capture your thoughts. The second chart looks at your historical triage metrics; it’s spiky right now because I’ve recently responded to many older tickets for the first time, but goal would be to get this down to around 2-4 days as your internal SLA for response time. Responding to tickets with comments immediately helps establish trust within your organization and also helps prevent duplicate efforts – if you are going to take the time to think about the issue, take a few extra moments to capture your thoughts to help others understand. These are daily or weekly triage tasks to ensure you are responding to tickets quickly.
The second two charts on this dashboard look at resolution time (e.g. average number of days that tickets resolved that week were open) and average age of tickets in the backlog. These are companion charts; if you spend a week closing out really old issues, then the resolution time for that week will spike but your average age of the remaining open tickets the following week will drop. Goal is to routinely triage the backlog, close out any old tickets that won’t be done (and/or escalate them to get them done) and eventually to set a target metric for both of these. These are weekly or monthly triage tasks to just consider how these metrics will impact your team and what decisions you need to make with the data.
ACTIVE TRIAGE filters:
Open issues with no comments - https://sagebionetworks.jira.com/issues/?filter=18549 (sort by date created to respond to new issues, sort by date updated to bump old issues)
Assessing Staffing and Project Velocity - Velocity and Releases
https://sagebionetworks.jira.com/secure/Dashboard.jspa?selectPageId=11104
This dashboard breaks down the issues created and resolved by four Jira projects: PLFM, SWC, SYNPY, and SYNR. The only reason for this dashboard is to ensure that the tickets created don’t dramatically over-take the tickets resolved; if tickets created (red) is going way, way up, consider shifting resources to other projects and/or doing more aggressive “Won’t Fix” or “Won’t Do”. This is a weekly or monthly triage task.
Of note, these charts show releases on them, if releases (fixVersions) have been defined in the individual projects. To formally release a fix version, go to each project, ensure that the release you want to ship is actually (a) complete and (b) released, and then update the date.
Releases:
PLFM, SWC - use the date that Xavier Schildwachter says that the stack has been released in Slack in the #synapse-down channel; if a release is skipped, I still release it in Jira but use the following release date. If no issues are tagged to that release, ok to delete.
SYNPY, SYNR - python is always released first, then R client. Ask Kimyen Truong (Unlicensed) for release dates and update accordingly.
Daily Triage - Team Visibility
https://sagebionetworks.jira.com/secure/Dashboard.jspa?selectPageId=11000