You can also drop in code there to add instrumentation into the logs with timestamps. So literally timestamping certain processes within the script. So say you have a script that everything runs hunk dory and then all of the sudden one set of transactions sends it off the deep end (True Story). APM and Governance is great but you may find you need additional tools. It's also really important to get a context for the overall noise floor on the system in production. So for example, I optimized something to get it to run 70% faster, but then there was a UE script that was permitting that from being realized.