Skip to content
Changeline
Menu
HelpThe tabs

The tabs

Under the four figures, the tabs list what was counted. Each row is one release, one failure or one week, so a figure can be checked by hand. The tabs are Overview, Deployments, Lead time, Recovery and Fail rate, then Spaces where the rows come from more than one space. Each tab draws its charts above the table that holds its rows.

Overview

The Overview opens first. Four charts sit two across: Deployments a week, stacked into deployments without a failure and failed ones; Change lead time, the median and the 85th percentile, in days; Failed deployment recovery time, the median and the 85th percentile, in hours; and Change fail rate, in percent of deployments. Under them, Week by week has a row for each week, or each month for 180 and 365 days, with its deployments, median change lead time, median failed deployment recovery time and change fail rate.

Deployments

Two charts: Deployments a week, stacked by failed, and Deployments by weekday, by the weekday of their release day in UTC, Monday first. Under them, Releases has a row for each release released in the period, newest first. Its name opens Jira’s page for the release in a new tab. Each row has:

  • the day it was released, and the work items in it;
  • the median change lead time of those work items;
  • the time they spent waiting to start, in progress and waiting for release;
  • Failures: the keys of the failures that point at the release;
  • Rework: the keys of the failures it fixes, see Deployment rework rate.

When a Jira admin counts deployments as work items reaching Done, the table is named Deployments, and each row is a work item that reached Done. The first 50 rows show, then Show more.

Lead time

The tab opens with Where the lead time goes: the three stages of the lead time as one bar, each a median. Below it, four charts sit two across: Change lead time, its median and 85th percentile in days; Where the lead time goes a week, the three stages stacked in median days; Change lead time spread, the work items in each of DORA’s survey bands, from under an hour to over six months, with no band called good or bad; and Change lead time by work type, the median of each work type. Then Work types has a row for each work type, with its work items, its median change lead time and its failures.

Recovery

Four charts sit two across: Failed deployment recovery time, its median and 85th percentile in hours; Failures raised and recovered, three lines for the failures raised, the failures recovered and the failures still open at the end of each week; Failed deployment recovery time spread in DORA’s bands, with Not recovered yet last; and Failures by priority, open or recovered, with the median recovery time of those recovered. Then Failures has a row for each failure that points at a release or a work item in the period. Each row has the failure’s work item, its work type, the release it points at, the release that fixes it (or Done), when it was created and its failed deployment recovery time. A failure not yet recovered reads Open, and the open ones come last.

The Recovery tab: recovery time a week, the failures raised, recovered and open each week, failures by recovery time in DORA’s survey bands and by priority; under them a row for each failure, with its work item, work type, the release it points at, the release that ships its fix, when it was created and its failed deployment recovery time, the open failures last, with Open in place of a time.
The Recovery tab: its four charts over a row for each failure and its recovery time

Fail rate

Change fail rate a week comes first. Beside it, in versions mode, Release size draws the releases by the work items they hold, 1–3, 4–10, 11–25 and 26+, stacked by failed or not, with each size’s change fail rate. In Done mode a deployment is one work item, so the chart is not drawn. Then Found after shows how long after the deployment it points at each failure was created, in the same bands. Below them, Releases that failed lists each failed release with its failures and its rework (in Done mode, Deployments that failed).

Spaces

The Spaces tab is drawn only where the rows come from more than one space, and it opens first when it is drawn. Four horizontal bar charts sit two across, one for each key: deployments, median change lead time in days, median failed deployment recovery time in hours, and change fail rate in percent of deployments. Each has a bar for each space, by name and never ranked, because each space is measured against its own past. A space with no figure for a key is named, such as Accounts (none), with no bar. Under the charts, Spaces has a row for each space with its four keys. It counts every space read, whatever the Space chip says, and a failure counts in its own space.

Every tab with a line chart has a Days before switch. It draws the same days just before the period beside the chart, so the two can be read side by side, and Copy link keeps the switch on. Each chart’s subtitle names its unit.