Engineering Performance Report

PI 5 → PI 8 Analysis

Cross-team analysis across four program increments covering agile execution health, DORA deployment frequency, and developer activity signals. Five teams evaluated against defined targets.

Period Feb 2025 – Feb 2026

Teams Cloud · Data · Fusion · Mobile · Web

PIs Covered 5 · 6 · 7 · 8

Data Sources Jira · Git · Deployments

Executive View

Completion Ratio Heatmap — All Teams × All PIs

Completion Ratio (% of committed points delivered)

Target: ≥ 80% · Green = on target · Yellow = 60–79% · Red = below 60%

* PI 5 data for Fusion is partial (2 sprints only). Ratios are PI-level aggregates from sprint_pi_summary.

Completion Ratio Trend — All Teams

PI-level rolling completion % by team

80% target

Creep Ratio Trend — All Teams

Unplanned work as % of planned commitment

20% ceiling

PI 8 Snapshot

Most recent completed PI — completion ratios per team

Cloud

51.8%

Completion

▲ 10pp vs PI7

Data

48.9%

Completion

▼ 7pp vs PI7

Fusion

47.6%

Completion

≈ flat vs PI7

Mobile

50.1%

Completion

▲ 1pp vs PI7

Web

88.6%

Completion

▲ 15pp vs PI7

Agile Execution

Commitment, Delivery & Scope Control

Bugs Completed Per PI

Count of bug-type issues completed per sprint aggregate

Total Commitment Volume (Story Points)

Sum of committed points per PI per team

Full Agile Data Table — PI 5–8

All teams and PIs · Ratios flagged against targets

Team	PI	Commitment	Done	Completion	Creep	Creep %	Bugs

Sprint-Level Completion — Cloud

Per-sprint commit done ratio across PI 5–8

Sprint-Level Completion — Web

Per-sprint commit done ratio across PI 5–8

DORA Metrics

Deployment Frequency — Cloud & Web

Cloud — Deployments Per PI by Environment

Preproduction and Production counts

0.5/day target (production)

Web — Deployments Per PI by Environment

Staging and Production counts

0.5/day target (production)

Production Deployment Frequency (deploys/day)

Far below 0.5/day target — both teams in "Low" DORA performance band

Team	Environment	PI 5	PI 6	PI 7	PI 8	Target	Status
Cloud	preproduction	29 (0.34/d)	12 (0.12/d)	9 (0.11/d)	8 (0.08/d)	—	Declining
Cloud	production	1 (0.01/d)	5 (0.05/d)	1 (0.01/d)	4 (0.04/d)	≥ 0.5/d	Critical
Web	staging	20 (0.24/d)	100 (1.01/d)	84 (1.00/d)	33 (0.34/d)	—	Regressing
Web	production	3 (0.04/d)	4 (0.04/d)	2 (0.02/d)	1 (0.01/d)	≥ 0.5/d	Critical

⚠ Neither Cloud nor Web has ever reached the 0.5/day production deployment target. Web's high staging activity (PI6–7) does not translate to production. Data, Fusion, and Mobile have no deployment tracking configured.

Developer Activity

Commit Velocity — PI 6, 7, 8

PI 5 activity data is excluded from trend analysis — commit tracking coverage was incomplete for most teams during that period (only 1–6 active days recorded). PI 6–8 reflects full coverage.

Total Commits by Team (PI 6–8)

Absolute commit volume per PI — excludes PI5 (incomplete coverage)

Active Commit Days by Team (PI 6–8)

Days with at least one commit — measures consistency

Activity Data Table — PI 6–8

Note: Mobile shows an extreme outlier pattern that warrants investigation

Team	PI 6 Commits	PI 6 Active Days	PI 7 Commits	PI 7 Active Days	PI 8 Commits	PI 8 Active Days	PI6→8 Trend
Cloud	5,739	158	4,477	143	3,914	166	▼ 32%
Data	187	41	263	45	909	95	▲ 386%
Fusion	1,005	105	747	87	524	81	▼ 48%
Mobile	65,838	146	35,415	123	9,004	145	▼ 86%
Web	988	100	756	110	844	111	▼ 15%

⚠ Mobile's commit count of 65,838 in PI6 is anomalous — likely includes generated files, lock files, or a large migration commit batch. The drop to 9,004 in PI8 may reflect cleanup rather than reduced activity, but combined with PI8's bug spike (115 bugs) it warrants investigation. Active days remain stable (145–146), suggesting the team stayed active throughout.

Team Profiles

Per-Team Assessment — PI 5–8

☁ Cloud

Best PI PI 8 · 51.8%

PI8 Creep 21.5% ⚠

PI8 Bugs 10

Prod deploys PI8 4

Improving trend in PI8 but still far from the 80% target. Creep is now above ceiling. Production deployments are negligible. Commit velocity is declining.

◈ Data

Best PI PI 6 · 75.4%

PI8 Creep 34.1% ✕

PI8 Bugs 5

Commit Growth +386%

PI 6 showed potential (near target) but regressed. Creep in PI8 is the worst in the dataset at 34.1%. Commit activity growth is positive — team is increasingly active but sprint planning quality needs work.

⊕ Fusion

Best PI PI 6 · 64.7%

PI8 Creep 39.0% ✕

PI8 Bugs 12 ⚠

Commit Trend ▼ 48%

Persistent underperformance with a creep problem that is getting worse. PI8 bug count jumped from 0 to 12. Declining commit activity. Never came close to 80% completion. Needs significant process intervention.

⬡ Mobile

Best PI PI 6 · 55.6%

PI8 Creep 48.9% ✕

PI8 Bugs 115 ✕

PI8 Commit Drop ▼ 86%

The most troubled team. Completion has never approached 80%, creep is catastrophic in PI7–8, and bugs have exploded to 115 in PI8 (double all other teams combined). Anomalous commit activity needs root cause analysis.

⬡ Web

Best PI PI 6 · 91.3%

PI8 Creep 10.5% ✓

PI8 Bugs 8

PI8 Completion 88.6% ✓

Clear standout performer. Consistently hits or approaches the 80% target. Creep is trending down sharply. PI7 dip (74%) was temporary — fully recovered in PI8. Bug count near zero until PI8 uptick. The reference benchmark team.

✦ Standout

Web is the only team consistently near or above target

3 of 4 PIs above 80% completion, creep under control and improving (30.9% → 10.5%), and near-zero bug completion. Web's planning practices should be studied and shared across teams.

✕ Critical Risk

Mobile's PI8 bug count (115) is a quality crisis

Mobile completed 115 bug issues in PI8 — more than all other teams combined (35). Combined with ~49% creep and only 50% completion, this signals sustained delivery instability and mounting technical debt.

⚠ Pattern

Scope creep is a systemic issue across 4 of 5 teams

Only Web has brought creep below the 20% ceiling (PI7: 15.2%, PI8: 10.5%). Cloud just breached the ceiling in PI8. Data, Fusion, and Mobile are all significantly above it, suggesting organizational-level planning and backlog hygiene problems.

→ Signal

Data team shows positive momentum but fragile

PI6 nearly hit 75% completion (best in team history), commit activity grew 386% PI6→PI8, and bugs remain low. However PI8 shows a regression with 34% creep. The capacity investments appear to be landing — planning discipline needs to catch up.

Recommended Actions

Priority Improvement Areas

Run a Mobile team quality audit — bugs and creep are at crisis levels

Mobile's PI8 numbers (115 bugs, 48.9% creep, 50.1% completion) indicate a structural problem, not a one-off sprint issue. Conduct a retrospective spanning PI7–8 to identify root causes: is this test coverage, unclear requirements, external dependencies, or team capacity? Consider a dedicated stabilisation sprint before PI9 planning begins.

Mobile

Adopt Web's sprint planning practices as the engineering standard

Web is the only team consistently hitting the 80% completion target and the only team with creep under control. Run a cross-team planning retro where Web leads a session on their backlog refinement, story sizing, and mid-sprint scope control process. Document and share the practices before PI9.

Cloud Data Fusion Mobile

Introduce a creep budget policy for PI9 — hard cap at 20% per sprint

Creep above 20% is an active target breach for Cloud (PI8: 21.5%), Data (34.1%), Fusion (39%), and Mobile (48.9%). Rather than a soft guideline, institute a formal sprint-level gate: any new work added mid-sprint must displace existing commitment or be deferred. This requires product owner alignment and should be embedded in sprint ceremonies.

Cloud Data Fusion Mobile

Investigate and close the production deployment gap for Web and Cloud

Web is deploying to staging at 0.34–1.01/day but only 1–4 times per PI to production. Cloud's preproduction activity has collapsed from 29 to 8 deployments per PI, and production has never exceeded 5. These numbers indicate either manual release gates, risk aversion, or process bottlenecks blocking flow to production. Map the release pipeline, identify gates, and target at minimum weekly production deployments by PI9-end.

Cloud Web

Expand DORA tracking to Data, Fusion, and Mobile

Currently only Cloud and Web have deployment tracking configured, covering 2 of 5 teams. Without production deployment data for Data, Fusion, and Mobile, it's impossible to assess end-to-end delivery capability. Prioritise instrumenting their pipelines before PI9 starts so the PI9 retrospective has full coverage. This is especially urgent for Mobile given its quality crisis.

Data Fusion Mobile

Investigate Mobile's anomalous commit volume — 65k PI6 vs 9k PI8

Mobile recorded 65,838 commits in PI6, 7× more than any other team. This is almost certainly not genuine feature work — it's likely auto-generated files, lock file churn, or a large asset import being tracked in version control. The decline to 9,004 in PI8 may reflect a policy change (e.g. adding these to .gitignore) rather than a reduction in activity. Auditing commit history for these PIs would clarify the signal and improve the reliability of activity metrics going forward.

Mobile

Add Lead Time for Changes and MTTR to the metrics roadmap

Deployment frequency is currently the only DORA metric tracked, and it alone paints an incomplete picture. Lead Time for Changes would reveal whether slow production frequency is a development bottleneck or a release gate issue. MTTR would capture stability. Change Failure Rate would connect to the growing bug signal. These are listed as planned — accelerating their implementation would significantly improve decision-making quality.

All Teams