As others are saying: it depends. Without knowing more about your situation and its context within the larger org, it's difficult to make any judgments. All I

As others are saying: it depends. Without knowing more about your situation and its context within the larger org, it's difficult to make any judgments. All I'll say is that I've internalized all these best practices and yet here I am running various products on various VPS's with a git post-receive hook and then logging in to the servers and recompiling or restarting. I have no tests, the code is not properly commented, and I've only written typespecs (Elixir) for the crucial bits of some of these apps, and I truly believe I've made the right decisions. In recent weeks I've pushed for refactoring, and I managed to get some hours to work on a better deployment strategy. When more money becomes available I'll push for writing tests and working on documentation. The goal then would be to remove the 'bus factor' of 1. I imagine you're nowhere near this particular scenario, but I imagine you're on a scale somewhere and I hope you figure out what the right standards are for you.