Insights
Insights show how work moves across the workspace and inside each team.
They answer practical operating questions: what moved, where time went, what changed in review, and whether work still matches the plan.
What Insights show
Insights combine completed work, initiatives, pull requests, code activity, standups, connected apps, and taxonomy.
At the workspace level, use Insights to read the full operating view: activity mix, initiative allocation, pull request flow, code changes, work type distribution, topics, languages, and taxonomy.
At the team level, use the same charts through a team filter. If standups are enabled, team Insights can also show standup duration and attendance cards.
Individual Insights are smaller today: they focus on the person's work type distribution.
Chart types
The dashboard renders chart cards only when enough data exists for the selected scope, period, and timezone. These examples mirror the chart cards and help text used in the dashboard; your values come from the connected workspace.
Planned vs Unplanned Work
Distribution of planned and unplanned completed work
Activity Distribution
% of all completed tasks related to each activity
Chart cards in the dashboard
| Chart | What to expect |
|---|---|
| Standup duration and attendance | Meeting duration and attendance trends when standups are enabled for the workspace or team. |
| Top Initiatives | The five parent initiatives with the most effort, ranked by completed task complexity. |
| Planned vs Unplanned Work | Completed work linked to initiatives compared with completed work that was not planned ahead. |
| Activity Distribution | Completed work grouped as new code, existing code, collaboration, planning, learning, and administrative work. |
| Activity Balance | The same activity categories in a radar layout. |
| Activity Trends | How that activity mix changes over time. |
| Development Distribution | New-code and existing-code work over time. |
| Number of lines changed | Insertions and deletions per period from connected code activity. |
| Code change totals | Previous versus current lines changed, insertions, and deletions. |
| Task Topics | Auto-detected work topics with at least 1% of effort. |
| Development Distribution by work category | Frontend, backend, infrastructure, docs, bug fixes, new features, and refactoring. |
| New Code Type Balance | The same work categories in a radar layout. |
| Code Languages | Languages with the most changed lines in the selected period. |
| Code Balance | Relative distribution of the most-used languages. |
| Pull Request Lead Time | How long pull requests take from creation to merge. |
| Lead Time Trend | Average pull request lead time over time. |
| PR Volume Trend | Opened and closed pull requests over time. |
| Cycle Time Phase Breakdown | Coding, pickup, review, and deploy time. |
| Initiative Trends | How effort moves between initiatives over time. |
| Initiative Distribution | Completed work grouped by initiative. |
| Taxonomy distributions | Products, Companies, Skills, Coding Tools, and Goals when those taxonomy terms exist. |
Read the signal
Start with the chart that matches the question.
Planned work is getting interrupted
Planned vs. Unplanned Work compares completed initiative work with urgent bugs, requests, and other unplanned work.
Initiatives are not moving
Roadmap Progress shows which initiatives are getting completed work and which ones are quiet.
Review is slowing delivery
Pull Request Flow tracks lead time, review time, merge pace, and PR volume.
Work is concentrated in one area
Taxonomy Analytics compares work by product, customer, goal, skill, and coding tool.
Agents are involved
Agent Analytics shows agent-attributed work without counting agent sessions as separate completed work.
Measurement principle
Insights are only as useful as the work graph behind them. Clear initiatives, honest statuses, linked pull requests, team ownership, and consistent taxonomy make the charts easier to trust.
Metrics point to where to look next. They are not a scorecard for individual performance and they are not a replacement for reading the underlying initiatives, bugs, comments, pull requests, and blockers.
Set the period and timezone before comparing charts. That keeps time-based charts aligned with how the team actually works.
Insights become more useful after you set up integrations, keep initiatives current, and apply taxonomy where it helps planning or reporting.