The answer to the following questions may also interest you
It happens faster than you think, a new version of the website is rolled out, but then you realize that a certain change has not been completed yet and the update must be undone as soon as possible.
With the classic approach of working via FTP uploads, you might now be faced with a real problem where you first have to check which files have been overwritten and whether you still have the original version and, in particular, what impact the changes to the files might have.
The Continuous Integration approach takes a slightly different approach here, all changes in the system are versioned in the Git repository and are available in chronological history. With each deployment a pipeline is executed, which saves the current state with all data and settings as an independent container in the system, a replay of this instance is then possible with a few clicks comfortably in the CI system.
I would be happy to advise you on the topic of Continuous Integration.