Essential Capabilities for High‑Performance Software Delivery (Based on Accelerate)

Continuous Delivery Capabilities

1. Version control for all production artifacts

All production artifacts—including application code, configuration, system settings, and scripts for automated builds and environment provisioning—should be managed in a version‑control system such as GitHub or Subversion.

2. Automated deployment process

Deployment automation means the release pipeline is fully automated and requires no manual intervention.

3. Implement continuous integration

Continuous Integration (CI) is the first step toward continuous delivery; code is regularly checked in, each check‑in triggers a fast test suite to catch regressions, and developers fix failures immediately. CI creates standardized builds and packages that are eventually deployed.

4. Use trunk‑based development

Trunk‑based development, with fewer than three active branches and short‑lived feature branches (often less than a day), reduces merge conflicts, code‑freeze periods, and enables rapid integration.

5. Implement test automation

Test automation continuously runs software tests throughout development. An effective test suite reliably detects real failures and only passes releasable code, with developers primarily responsible for creating and maintaining the automated tests.

6. Support test data management

Effective test data management ensures sufficient data for test suites, on‑demand data provisioning, conditional adjustments in pipelines, and limits the amount of data needed for automated testing.

7. Shift‑left security

Integrating security early in design and testing—through security reviews, approved security libraries, and automated security tests—improves overall IT performance.

8. Implement continuous delivery (CD)

CD keeps software in a deployable state throughout its lifecycle, prioritizing rapid feedback on system quality and enabling teams to deploy to production or end users at any time.

Architecture Capabilities

9. Use loosely coupled architecture

A loosely coupled architecture allows teams to test and deploy independently without coordinating with other services, accelerating work and delivering value.

10. Empower team architects

Teams that choose their own tools and architectural approaches tend to achieve better continuous‑delivery performance.

Product and Process Capabilities

11. Collect and act on customer feedback

Regularly seeking and incorporating customer feedback into product design is crucial for high software delivery performance.

12. Make workflow visible through value streams

Teams should have clear visibility of the end‑to‑end workflow from business to customer, which positively impacts IT performance.

13. Work in small batches

Breaking work into small, weekly (or shorter) increments enables rapid development, faster feedback, and quicker delivery of value.

14. Foster and enable team experimentation

Allowing developers to try new ideas and update specifications without external approval accelerates innovation and value creation.

Lean Management and Monitoring Capabilities

15. Use lightweight change‑approval processes

Peer‑review‑based lightweight approvals outperform external change‑approval boards in delivering superior IT performance.

16. Monitor applications and infrastructure to inform business decisions

Data from monitoring tools should drive actions and business decisions, not just generate alerts.

17. Proactively check system health

Threshold‑based and rate‑of‑change alerts enable teams to detect and mitigate issues early.

18. Improve processes and manage work‑in‑progress (WIP) limits

Applying WIP limits, a well‑known lean practice, improves flow efficiency and throughput.

19. Visualize work to monitor quality and communicate across the team

Dashboards or internal sites that display work status and quality metrics help improve software delivery performance.

Cultural Capabilities

20. Support a generative culture

Based on Ron Westrum’s typology, a generative culture—characterized by strong information flow, high collaboration, trust, and proactive inquiry—predicts better IT and organizational performance.

21. Encourage and support learning

Viewing learning as an investment rather than a cost is a key measure of a learning‑oriented culture.

22. Promote collaboration between teams

Collaboration across traditionally siloed development, operations, and security teams enhances overall effectiveness.

23. Provide meaningful resources and tools

Giving teams the tools and resources needed to do challenging, purposeful work boosts satisfaction and performance.

24. Support transformational leadership

Transformational leadership—vision, intellectual stimulation, inspirational communication, supportive guidance, and personal recognition—amplifies the technical and process work essential to DevOps success.