Off-Hours Transitions(5.1.8)

This visual shows when — not just how often — workflow transitions are happening. Transitions outside of business hours or on weekends are not automatically suspicious, but they stand out when they occur in volume or follow a recurring pattern.

Read together with the other signals in this module, it shows when status changes happen outside the hours most of the team is active.

What you can conclude

• Isolated off-hours transitions are common in distributed teams; recurring patterns on the same day or time slot are the ones a reviewer would ask about.
• Weekend activity that coincides with rapid transitions or backward moves is the combination a reviewer would ask about first.
• A heatmap with few off-hours transitions means status changes mostly happen during working hours.
• Off-hours transitions recorded under an account Jira reports as an app, such as Automation for Jira, are counted apart from people's, by Jira's own record.

How this chart works

Day-of-week × ISO-week calendar heatmap showing the count of status transitions occurring outside business hours — defined as before 08:00 or from 18:00 in the organization's time zone, or at any time on a Saturday or Sunday. Cell color: low → medium → high → critical. Each cell's count is also split by the type of account that made the transition, as Jira reports it: person, app, customer, unknown, erased, or system where no actor was recorded; the types add up to the cell's count. The organization's time zone is inferred from the timestamps Jira sends for the site, so on a Central European site an 18:30 local transition is counted as after hours and an 08:30 one is not. The inference is repeated every night, so a newly connected site moves from UTC to its local window once enough timestamps have arrived; until then, or where the timestamps do not match a single time zone, the window is evaluated in UTC. Working hours cannot yet be changed from the dashboard.

The week in progress is shown at its count so far. Because the week is not yet complete, it covers only what has been recorded up to now and can only grow as the week closes. A week that falls only partly inside the selected date range is drawn and marked partial, so a short first week is visible as a short week rather than read as a fall.