Compares the original estimate with the time logged on resolved issues, and checks whether story points predict how long work really takes.
Add the gadget, open its three-dot menu, choose Configure and fill in:
statusCategory = Done. Use it to limit the gadget to a project, a team or an issue type.| Figure | Definition |
|---|---|
| Within ±25% | Share of issues where time spent was between 75% and 125% of the original estimate |
| Typical overrun | Median of time spent divided by original estimate |
| Total bias | Total time spent against total estimated, as a percentage. Positive means more time was spent than estimated |
| Overrun by estimate size | Median overrun for each estimate size: up to 2 h, 2 to 4 h, 4 to 8 h, 1 to 3 days, over 3 days. Bars above 1.25 are orange |
| By assignee | Issues, share within ±25% and bias for each person with at least three issues (up to 12 people) |
| Biggest overruns | Issues with an estimate of an hour or more, ordered by time spent divided by estimate. Only overruns are listed |
| Cycle time by story points | For each story point size, the range from the 25th to the 75th percentile and the median, from the first move into an In Progress status to resolution. Values are rounded up to the next of 0.5, 1, 2, 3, 5, 8, 13 or 21, and larger ones are grouped as 21+, so a 4 appears with the 5s |
| Points vs effort | Rank correlation (Spearman) between story points and cycle time, with a short reading |
Only issues with both an original estimate and logged time are used for the time figures. Issues that went from To Do straight to Done have no start for cycle time and are left out of it: counting from creation would measure waiting in the backlog, not work.
Work that was moved to Done in a workflow that never sets Resolution is still found. Its resolution date is the moment it entered the Done category.