Skip to content

Continuous Integration & Delivery

Home › Explanation › Transformation › Capability Metrics › Practice Areas › Ci Cd

CI/CD automation enables fast feedback, reduces manual errors, and allows on-demand deployment—the technical foundation for delivering changes safely and rapidly.

Impact: Deployment time reduced from days to minutes, reduced errors, fast feedback (minutes vs hours), on-demand deployment capability, rapid response to issues.

Level 1: Initial

Manual builds and deployments.

  • No automated builds
  • Manual deployments with runbooks or scripts
  • Deployment time: hours to days
  • Build process varies by team/developer
  • No pipeline, no quality gates

Advancing to Level 2: Set up CI server (GitHub Actions, GitLab CI, Jenkins), automate builds on every commit, create basic automated test suite for CI, write deployment scripts, establish quality checks in pipeline.

Resources: CD Model Overview · CD Model Stages

Level 2: Managed

CI established, semi-automated deployment.

  • Builds run automatically on every commit
  • Automated tests run in CI (basic suite)
  • Deployment semi-automated (scripts + manual gates)
  • Some quality checks in pipeline
  • Build failures visible immediately
  • Deployment requires manual steps/approvals

Advancing to Level 3: Automate deployment to all environments, implement comprehensive quality gates (tests, security, compliance), standardize pipeline across teams, remove manual deployment steps, enable on-demand deployment.

Resources: CD Model · Deployment Strategies

Level 3: Defined

Full CD pipeline with quality gates.

  • Full CD pipeline (commit → production automated)
  • Deployment fully automated to all environments
  • Comprehensive quality gates (tests, security scans, compliance checks)
  • Standardized pipeline across teams
  • Can deploy on-demand (technical capability exists)
  • Zero-downtime deployments, automated rollback

Advancing to Level 4: Implement DORA metrics tracking (deployment frequency, lead time, MTTR, change failure rate), collect pipeline metrics, apply statistical process control, implement predictive analytics, create metrics dashboards.

Resources: Measuring Flow

Level 4: Quantified

Pipeline metrics and DORA metrics tracked.

  • DORA metrics tracked precisely (deployment freq, lead time, MTTR, CFR)
  • Pipeline metrics collected (build time, test time, bottlenecks)
  • Statistical process control applied
  • Predictive analytics (predict pipeline failures)
  • Data-driven improvements with measured outcomes
  • Trend analysis on all metrics

Advancing to Level 5: Implement self-optimizing pipeline, experiment with pipeline approaches (A/B test strategies), proactive issue detection, share practices with industry.

Resources: Measuring Flow

Level 5: Optimizing

Self-optimizing pipeline.

  • Self-optimizing pipeline (automatically adjusts based on patterns)
  • Continuous experimentation with pipeline approaches (quarterly A/B tests)
  • Proactive failure detection (catch issues before problems)
  • Industry-leading practices (others learn from you)
  • Community contributions (talks, papers, open source)

Maintaining: Stay current with CI/CD research, active community participation, regular experimentation with measurement, share learnings.

Level Assessment

You're at a level when:

  • ✅ All characteristics consistently demonstrated organization-wide
  • ✅ Capabilities are sustainable (not dependent on heroes)
  • ✅ You possess the capability, not just working toward it

Level Distinctions:

  • 1 → 2: CI with automated builds and tests (capability exists)
  • 2 → 3: Full CD with on-demand deployment capability
  • 3 → 4: Measure effectiveness, track DORA metrics
  • 4 → 5: Self-optimizing, continuous experimentation

Dependencies:

  • Depends on: Version Control Level 2+, Testing Level 2+
  • Enables: Evidence Level 3+, Security Level 3+
  • Blocks: If weak (Level 1-2), deployment remains manual and slow

Tutorials | How-to Guides | Explanation | Reference

You are here: Explanation — understanding-oriented discussion that clarifies concepts.