Jira Metrics

Engineering Flow

AI-first executive view · Custom scope · last 30d

AI Bottom Line

AI-narratedcached· 2026-07-21 00:31 UTC

Output fell sharply while flow remained slow, with review and hold states driving the current delay profile.

AI ROI: data insufficient for the last 30d. 1 project in scope, none with stable AI vs Human cohort (n≥10 both sides) on Lead Time.
Investigate per project →

Headline KPIs · last 30d

Issues done
14 60.0%
Last 8w
AI participation
7.1% 144.8%
Last 8w
Lead Time (active)
19.4d 7.9%
no trend
Cycle Time (active)
7.7d 90.5%
Last 8w

AI insight

  • Where AI helps:The single AI-owned and AI-touched item reached ready very quickly, at 1.8 hours.
  • Where AI hurts:On the single AI-owned item, cycle time was 985.6 hours and lead time was 987.4 hours; this is directional only because the sample size is one. The same pattern appears in AI-touched work because that cohort is also one item.
  • Next focus:Reduce review backlog and unblock stuck work before adding new intake.

AI in teams · 1 team

PDOPlatform Daily Operations
5.6% AI

18 completed · 5.6% AI-touched · 1 active agent

Bottom line

PDO declined (-62.5%) in throughput. AI-vs-Human cycle comparison is directional only (sample too small) (n=1/14); the main AI bottleneck is Ready For Review.

Issues done

18

-62.5% vs prev

AI-owned share

6%

+3.5 pp vs prev

AI-touched

1 (6%)

Lead Time (active)

28.7d

prev 31.8d

Cycle Time (active)

7.9d

prev 4d

TTR p50

20.8d

prev 22.1d

PR Review p50

WIP active

15

WIP blocked

6

AI insight

Where AI helps

No statistically stable AI advantage detected yet. Directional-only signals below numerically favor AI but sample size is below threshold.

Where AI hurts

No statistically stable AI regression detected. Directional-only signals below numerically favor humans but sample size is below threshold.

Directional only (sample too small)

  • · Cycle Time (active): AI 41.1d vs Human 7.7d (n=1/14, below 20)
  • · Lead Time (active): AI 41.1d vs Human 28.7d (n=1/17, below 20)
  • · Time to Ready p50: AI 1.8h vs Human 20.8d (n=1/13, below 20)

Top AI bottleneck

Ready For Review
54.7d

Active agents · top 1

Oliver
1

AI Period: compare · intro 2026-03-30

Active bottlenecks · top 5

Workflow stageRelative wait timeMedian time
Ready For Review
33.9dp50
On Hold
20.1dp50
In Testing
15.1dp50
Ready For Live
14dp50
Failed On Testing
10.8dp50

Top teams · by throughput

KeyTeamThroughputLead Time (active)Cycle Time (active)Active WIPBlocked WIPAI %
PDOPlatform Daily Operations1828.7d7.9d1566%

Data quality amber

Sync freshness1/1 synced
Latest sync1/1 OK
Config coverageall configured
PR linkageno PR data

Report version: engineering-flow.v8 · metric definitions: 2026-05-24.audit18.v1

Period: 2026-06-222026-07-22 · scope: Custom scope (1 project)

Detail view (/overview) → — full per-metric tables, AI Period comparison, longest-stuck list.