Case study 15 of 15
Delivery Paths Matter More Than the Cluster
Repeatable software delivery on Kubernetes/OpenShift, with GitOps, high availability, and observability treated as the product — not as a cluster someone bought.
Business challenge
Teams needed a reliable path from development to production. A cluster without an owned delivery path still leaves every release as a negotiation.
Architecture approach
Kubernetes/OpenShift as the runtime, ArgoCD-based GitOps for auditable promotion, and HA/DR plus observability so the path could be operated, not just deployed.
How the work moved from request to production
- Business need
- Intake & risk classification
- Select reusable pattern
- Build / configure
- Review & validate
- Deploy / enable
- Monitor usage, risk, cost, value
Governance considerations
Ownership of the delivery path — who can change what, what is promoted through Git, and who is on the hook when a release fails — mattered more than the platform brand.
Results
Repeatable delivery patterns and a clearer ownership model for production changes. The work is about the path, not a vendor scorecard.
Lesson learned
Buying a cluster is not a delivery strategy. The path and the ownership model are.
Discussion questions
- Who owns a production change after GitOps is in place?
- What should be impossible to do outside the delivery path?
- Which signals prove the path is healthy, not just that the cluster is up?