The basic build automation process, though not flashy, is probably the biggest win for us. One of the specific features we rely on is the ability for our Bamboo build process to parse Junit report outputs. We have written a number of validation tests that are executed during the validation portion of our builds. Note that that may not be novel from a software perspective, but we also have a number of configuration repositories against which we also run these validation steps to identify errors in the configuration and/or changes that should not have been made.
We work with a separate team responsible for the actual configuration and behind-the-scenes work, so from where we sit there are a few things that just "magically" work or fail horribly. Something that we have had difficulty with is parsing the captured output during the build process. Often we find stdout and stderr messages not being captured and viewable in the web interface. This usually results in "big stick" poking about to try to resolve the issue versus more targeted resolution that cleaner logs would allow.
We've really benefited from the Build Automation and CI/CD capabilities of Bamboo and it's integration with our Bitbucket and GitHub repositories. I recommend this for folks looking for a CI/CD solution.
We've seen a significant improvement in our ability to track and resolve build issues as well as managing build artifacts. Previously, we were relying on manual builds with moderate success. Being able to tie automation into our build process has really made a difference. We regularly utilize the "Latest Successful" build artifacts in our process and we've been able to avoid introducing errors on multiple occasions.