What is the difference between Continuous Integration, Continuous Delivery, and Continuous Deployment?

Question

Accepted Answer

These three terms are often used together, but they represent distinct stages in the automation pipeline.

Continuous Integration focuses on the development phase. Developers commit code to a shared repository frequently, ideally several times a day. Each commit triggers an automated process that builds the application and runs tests. The goal is to detect integration issues early. If the build or tests fail, the team fixes the problem immediately. CI ensures that the codebase is always in a working state.

Continuous Delivery builds on CI by automating the release process. After code passes all tests, it is automatically prepared for deployment. This includes additional testing like integration tests, performance tests, and security scans. The code is packaged and ready for production. However, the actual deployment requires a human to press a button. This gives teams control over when releases happen, which is useful for coordinating with marketing, support, or other business needs.

Continuous Deployment removes the manual step entirely. Every change that passes the automated pipeline goes straight to production without human intervention. There is no waiting, no approval process, and no deployment button. This requires extremely reliable automated testing because there is no safety net.

Here is a simple way to visualize the progression:

CI: Code commit leads to automatic build and test.
Continuous Delivery: CI plus automatic preparation for release, with manual deployment trigger.
Continuous Deployment: CI plus automatic deployment to production.

Most organizations start with CI, then move to Continuous Delivery. Continuous Deployment is common at companies with mature testing practices and a culture of rapid iteration, like technology startups and large tech companies.