Skip to content

Pilot to rollout guide

Use this guide to move from evaluation to production adoption with measurable exit criteria.

Stage 1: Evaluation

Goal

Prove canonical path works in representative repositories.

Entry criteria

  • Team sponsor and release owner assigned.
  • Baseline repo selected.
  • Target lane selected (Startup/Scale/Regulated).

Exit criteria (measurable)

Check Threshold Evidence
Canonical path runs in baseline repo 1/1 successful run with artifacts produced build/gate-fast.json, build/release-preflight.json, build/doctor.json
Onboarding baseline captured Week-1 seed KPI payload committed docs/kpi-weekly-<date>.json
Ownership model set All 5 workstreams have DRI listed docs/top-tier-program-dashboard.md

Stage 2: Pilot

Goal

Validate repeatability across multiple teams/repos.

Entry criteria

  • Evaluation exit criteria complete.
  • Lane selected (Startup/Scale/Regulated).
  • Weekly status cadence announced.

Exit criteria (measurable)

Check Threshold Evidence
Pilot coverage At least 3 repos onboarded (or approved pilot size) Portfolio scorecard JSON
Weekly operating rhythm 2 consecutive weekly cycles completed Weekly executive reports
Escalation readiness 1 incident/escalation workflow practiced Incident log + support model reference
Portfolio contract adoption Scorecard payload emitted with schema_version and required totals docs/artifacts/portfolio-scorecard-*.json

Stage 3: Production rollout

Goal

Operationalize portfolio reporting and governance at org level.

Entry criteria

  • Pilot criteria complete with leadership sign-off.
  • Monthly leadership review template ready.

Exit criteria (measurable)

Check Threshold Evidence
Portfolio reporting operational Weekly scorecard generated for active repos for 4 straight weeks Portfolio scorecard artifacts
Governance enforcement active Compatibility/deprecation reviews completed monthly Governance policy docs + review notes
Leadership reporting institutionalized Weekly + monthly executive reports used in cadence docs/executive-weekly-*.md, docs/executive-monthly-*.md
Release package readiness Full release package checklist passes for release train docs/full-release-package-checklist.md
  • Day 30: evaluation complete + KPI baseline payload + DRI assignment.
  • Day 60: pilot steady-state + 2 weekly cycles + escalation rehearsal.
  • Day 90: 4-week portfolio reporting streak + monthly leadership review + release package checklist adoption.

Stage gate command set

Use this command set before each stage sign-off:

python -m sdetkit gate fast --format json --stable-json --out build/gate-fast.json
python -m sdetkit gate release --format json --out build/release-preflight.json
python -m sdetkit doctor --format json --out build/doctor.json
python scripts/build_portfolio_scorecard.py --in <normalized-input> --out <portfolio-scorecard.json> --schema-version 1.0.0 --window-start <YYYY-MM-DD> --window-end <YYYY-MM-DD>

Common failure modes and mitigations

Failure mode Signal Mitigation
Scope too broad Too many repos in initial pilot Limit to highest-impact services first
No clear ownership Repeated missed commitments Assign DRI per workstream and publish weekly
Data inconsistency KPI disputes every week Freeze schema for reporting window
Escalation drift Severity decisions differ by team Enforce one support model + monthly calibration