Summary
Introduction
Ever wondered why some DevOps initiatives lead to real transformation while others stall after setting up a few pipelines? The difference isn’t in the tools, it’s in the mindset, culture, and way teams work together. That’s where the CALMS framework comes in: a set of principles that helps organizations go beyond automation and embrace DevOps as a holistic approach.
Why CALMS Matters
It’s easy to think of DevOps as a collection of technologies, like CI/CD pipelines, Kubernetes clusters, or infrastructure scripts. But at its core, DevOps is about bringing people, processes, and technology together to deliver software that is reliable, fast, and valuable to customers.
Without a framework, it’s tempting to fall into “cargo cult DevOps”, adopting new tools without changing how teams work together. That’s where CALMS comes in.
The model, introduced by Jez Humble and others, stands for Culture, Automation, Lean, Measurement, and Sharing. It’s not about ticking boxes, but about using a lens that helps teams assess where they are and where they can grow.
The Five CALMS Principles
1. Culture
Strong culture is at the heart of DevOps. It’s about creating trust and an environment where collaboration comes naturally.
- Break down barriers between development, operations, and business.
- Treat mistakes as opportunities to learn rather than blame.
- Support from leadership makes change sustainable.
Example: Teams that run blameless postmortems can openly discuss what went wrong and avoid repeating mistakes.
2. Automation
Automation helps reduce repetitive work and makes processes consistent and safe.
- CI ensures every change is tested early.
- CD allows teams to deploy quickly and confidently.
- Infrastructure as Code keeps environments reliable.
Example: Netflix relies on automation to safely handle thousands of deployments each day.
3. Lean
Lean principles encourage teams to focus on value and flow, instead of unnecessary complexity.
- Build only what adds real customer value.
- Shorten feedback loops to catch issues early.
- Favor smaller, frequent releases over risky “big bang” launches.
Example: Kanban boards and value stream mapping help teams spot bottlenecks and improve delivery speed.
4. Measurement
Improvement only happens when we can see progress clearly. Metrics help teams understand what’s working and what’s not.
- Track DORA metrics: lead time, deployment frequency, change failure rate, and MTTR.
- Use observability tools to see how systems behave in real time.
- Blend technical and business metrics for a complete picture.
Example: Measuring deployment frequency can show whether process changes are actually making delivery faster.
5. Sharing
Knowledge grows stronger when it’s shared. Open communication keeps teams aligned and learning together.
- Communities of practice spread skills across the organization.
- Documentation and runbooks make onboarding easier.
- Tools like ChatOps and dashboards help everyone stay in sync.
Example: Etsy’s public engineering blog is a great example of open sharing, benefiting both their teams and the wider tech community.
Use Cases & Applications
- Startups → Use CALMS to balance tooling with collaboration and learning.
- Enterprises → Apply CALMS as a maturity model to guide large-scale DevOps transformation.
- Consultancies → Use it as a framework for assessing and advising clients.
Conclusion
The CALMS framework reminds us that DevOps isn’t just about automation or tools, it’s about building an environment where people, processes, and technology support each other.
When organizations embrace Culture, Automation, Lean principles, Measurement, and Sharing, they set the stage for:
- Faster delivery
- Greater reliability
- Teams that feel engaged and supported
No matter where you are in your DevOps journey, CALMS can serve as a compass, helping you move forward in a thoughtful and sustainable way.