Roadmap Phase Milestone Checklist

Roadmap Phase Milestone Checklist

Use these checklists as exit criteria for each phase. A phase is complete when all mandatory items are checked and the phase review has been signed off by the programme sponsor.

Note: These are the minimum conditions for completing a phase — not a ceiling. Add organisation-specific items as needed.

---

Phase 1 — Foundation

Exit criteria: the organisation now has a demonstrated, repeatable, trustworthy automated workflow and the infrastructure to build more.

Technology

• Source of truth established for at least one domain (structured, version-controlled, maintained)
• All automation code and configuration templates are in a shared version-controlled repository
• No automation scripts remain on individual laptops or personal drives
• Basic CI pipeline runs on every commit: linting, syntax validation, and policy checks
• At least one change type is executed end-to-end via automation with pre-deployment validation
• Rollback mechanism tested and confirmed to work

Process

• Change management process updated to recognise automated pipeline as an approved execution path
• Peer review process defined for automation code (merge request, review criteria, approval gates)
• Incident response runbook updated to reflect automated deployment patterns

People

• All team members have access to and understand the version control repository
• At least two engineers (not just one) can operate and extend the automated workflow
• Onboarding documentation exists for the automation environment

Metrics

• Baseline metrics recorded: lead time, change success rate, manual vs automated change ratio
• Measurement mechanism in place (automated where possible, manual otherwise)

Stakeholder

• Phase 1 outcomes demonstrated to programme sponsor
• Consuming teams aware of the automated workflow and its capabilities
• Phase 2 scope agreed and backlog prioritised

---

Phase 2 — Scale

Exit criteria: automation is an organisational discipline, covering the majority of routine change volume, with cross-team workflows and visible metrics.

Technology

• Automation coverage: ≥50% of routine change types executed via pipeline
• Policy-as-code implemented for agreed security and compliance policies
• Pipeline catches policy violations before production deployment (verified by test cases)
• Workflow orchestration in place for at least two cross-team change types
• Configuration drift detection operational and alerting

Process

• Standard operating procedure exists for each automated change type
• Exception handling process defined: what happens when the pipeline fails?
• Compliance evidence generated automatically for audit-relevant changes
• Change management process updated to support higher deployment frequency

People

• Automation ambassadors identified and enabled in at least two consuming teams
• Skills gap assessment completed for Phase 3 requirements
• Platform ownership explicitly assigned with dedicated capacity (not best-efforts)

Metrics

• Observability dashboard operational (engineering view minimum)
• Deployment frequency, change failure rate, and pipeline success rate visible in near real-time
• Consumer satisfaction feedback mechanism in place

Stakeholder

• Monthly stakeholder update delivered with metrics evidence
• Phase 2 outcomes presented to senior leadership
• Phase 3 scope agreed and investment confirmed

---

Phase 3 — Platform

Exit criteria: the network operates as a platform; consuming teams can provision defined resources without raising manual tickets; one-touch deployment is operational for templated infrastructure.

Technology

• One-touch deployment operational for at least one infrastructure template (branch, segment, or connectivity domain)
• Self-service interface available for consuming teams (portal, API, or ITSM integration)
• Intent-based service model defined and implemented for at least one service type
• Closed-loop remediation operational for at least one incident class
• Full pipeline coverage for all standard change types (target: ≥80%)

Process

• Self-service consumption process documented and communicated to all eligible consuming teams
• Guardrails defined and implemented: what self-service can and cannot do
• Escalation process defined for requests outside self-service scope
• SLA published for automated versus self-service versus manual request types

People

• Engineering team capacity model updated to reflect platform operations model
• Consuming team onboarding process documented and tested
• Network engineering roles updated to reflect shift from execution to platform management

Metrics

• Self-service utilisation rate tracked
• Lead time for one-touch deployment measured and reported
• Pipeline bypass rate tracked and under agreed threshold

Stakeholder

• One-touch deployment demonstrated to business stakeholders
• Business outcome improvements quantified and communicated (lead time, cost, compliance)
• Phase 4 direction agreed; specific use cases identified

---

Phase 4 — Adaptive

Phase 4 does not have a fixed exit criteria — it is a continuous improvement mode. Use these indicators to assess progress.

Progress Indicators

• Intent-driven configuration generation operational for at least one domain
• AI-assisted anomaly detection in production with documented false positive rate
• Self-healing operational for defined incident classes; success rate measured and reported
• Continuous compliance enforcement operational: violations detected and remediated automatically
• MTTD for in-scope incident classes below defined threshold
• Engineering capacity allocation: majority of time on intent design and platform evolution, not execution

Ongoing Disciplines

• Quarterly roadmap review conducted; backlog updated based on findings
• ADR register maintained: all significant decisions recorded
• Metrics reviewed monthly; anomalies investigated
• Consumer feedback incorporated into platform backlog regularly