The project page: where a project’s time goes

The panel answers a question about one work item. The project page asks the same questions of a whole project: where does its time go, what is stuck right now, and is it getting better? It is a tab of the project, called Statuswatch, and everyone who can see the project can open it. The same page is under Apps → Statuswatch in Jira's sidebar, where you first choose the project; it shows one project at a time there too. It reads Jira as the person looking at it, so a work item you may not see is in none of its figures.

What it covers#

The work items that were active in the last 90 days, up to the 200 most recently updated, and on top of them every open work item that has not moved for longer than that, up to the 100 oldest, because those are the ones that have sat longest. All of it is counted the way the panel counts: in the project's working hours, each work item in the calendar that applies to it. The period is chosen at the top right: "Last 7 days", "Last 30 days", "Last 90 days", or "Custom dates" with a "From" and a "To" inside those 90 days. Changing it reads nothing again, so two periods are always counted at the same moment. It is one project at a time, on purpose; it is not a dashboard gadget and does not compare projects.

The four figures#

Hover over any of them for what exactly it counts.

• "Finished": work items that reached a done status in the period.
• "Typical time to done": measured from creation to done. Half of the work items finished in the period took less, half took longer. The middle value rather than the average, because one ticket forgotten for a quarter moves an average by days and says nothing about how work usually goes.
• "Open now": work items not done, whatever the period.
• "Unusually long now": open work items that have been in their status unusually long for this project. The tooltip says how many open items had enough history to say.

Typical stay per status#

One row per status. Every dot on the line is one stay that ended in the period, placed by how long it took, in the status's colour, the same colours the panel uses. The thin dark tick is the typical stay, and the figure at the end says it: half of the stays were shorter, half were longer. Under the tick the line says "median" and its value. The column's heading is the scale all rows share: "Typical stay, from 0" at the left end of the lines and, at their right end, "to" and the longest stay in the table. So a status whose stays are short looks short. Hover over a dot for the key of the work item it was and how long that stay took; click it to open the work item in a new tab. "Stays" is how many that is. A stay still running is not counted, because its length is not known yet, and time after a work item is done is not counted at all.

Open now, Unusually long now, Finished#

The heading of this table is a list to choose from, and its three entries are the work items behind three of the figures at the top, each with how many that is. Clicking one of those figures, "Finished", "Open now" or "Unusually long now", chooses its list too, and the figure whose list is showing is outlined. "Open now" is described below. "Unusually long now" is the same table with only the work items whose figure is in the warning colour. "Finished" lists the work items that reached done in the period, newest first, with the day they finished and their time to done: every dot on the line is one finished work item, the tick is the typical time to done, and the larger dot is the work item of that row. All its rows share one scale, which the heading gives.

Every open work item, the ones furthest past what is typical first. The grey dots are the stays in the work item's current status that ended in the period, one dot for each; the thin dark tick is their typical stay; the larger dot is this work item's stay so far. Here each row has a line of its own, as the heading says: "From 0" "to the longest stay in the row", which is this work item's stay or the longest finished one, whichever is longer. One scale for all rows would let a single forgotten work item flatten every other row.

• A larger dot far to the right of the grey ones has been there longer than any of them, and the distance is to scale.
• A larger dot among the grey ones, or left of the tick, is where stays in that status usually still are.
• No grey dots and no tick mean no stay in that status ended in the period, so there is nothing to compare with.
• When a status had more than 150 stays, 150 of them are drawn, evenly from the shortest to the longest, and the sentence you see on hovering says so. A dot is always a stay that happened, never an average of several.

A figure and a larger dot in the warning colour mark a stay the panel would flag too: unusually long, judged against the project's own finished stays in the same status. Hover over the line for the sentence. Under the table the page names every status in which it could not say, because fewer than five stays have ever finished there. Too little history is not the same as nothing being stuck.

By assignee#

The second tab, "By assignee", shows the same typical stays per person: one row for each assignee, one column for each status, and in each cell the typical stay of that person's work items in that status, over the stays that ended in the period. The cell draws those stays as dots, with a tick at the typical one; hover over a cell for how many stays that is. Within a status everybody's dots share one scale, so people can be read against each other. "Work items" is how many the person holds, with the open ones in brackets. The five statuses with the longest typical stay are shown.

A work item counts for whoever it is assigned to now. Jira's status history says when a work item changed status, not who held it at the time, so a stay in Review last month is counted for today's assignee even if somebody else had the work item then. Read the table as "how work assigned to this person has moved", not as a measurement of the person. Where works councils or local law govern evaluating individuals, check before using it that way.

Week by week#

The median per week, as a line, for "Time to done" or for the time in any one status. A week with fewer than three is left out of the line, and the page says how many were: a median of one or two work items is those work items, not a trend.

When a figure cannot be trusted#

If the project's working hours could not be read, the page says above its figures that they are calendar time. If status names could not be read, it says that what counts as finished is uncertain. If the project had more than 200 active work items, it says the figures cover the 200 most recently updated. If more than 100 open work items have not moved in 90 days, it says that "Open now" lists the 100 oldest of them.