There's a line from Dan Martell's Buy Back Your Time that has stuck with me longer than most business advice does: you don't have to slay the dragon, you just have to remove it.
The context was habits. Don't white-knuckle your way past the ice cream in the freezer every night. Don't buy the ice cream. The willpower fight you keep having is usually a fight you already lost the moment you set it up, and the fix isn't more discipline. It's removing the thing you'd otherwise have to resist.
I think this is the single most underused idea in how small businesses try to fix operational problems.
Willpower is not a control
Somewhere in almost every operation, there's a step that only works if a specific person remembers to do it correctly, every time, forever. Redact the file before uploading it. Double-check the total before sending the report. Remember to categorize the transaction the same way it's always been categorized.
That's not a system. That's a bet that a human being will be perfect, indefinitely, under whatever conditions they happen to be in that day. And humans aren't perfect indefinitely. Not because they're careless. Because attention is finite and some days are long.
The instinct when this fails is to blame the person and ask them to be more careful next time. The actual fix is almost always the same one Martell describes for habits: change the environment so the failure can't happen in the first place, rather than trying to will your way past it every single time.
What removing the dragon looks like in practice
For a personal habit, it's blocking social media at the operating-system level instead of trying to resist opening it. It's not keeping the tempting food in the house instead of trusting yourself not to eat it. One person cut their phone use nearly in half not by deciding to use it less, but by removing the apps that made it easy to lose an hour without noticing.
For a business, it's the exact same move, aimed at a process instead of a person. If a task depends on someone remembering to redact sensitive information before it goes anywhere, the actual fix isn't a reminder. It's a tool that structurally cannot retain or expose that data in the first place, so a forgotten step costs you a mistake instead of a catastrophe. If a report needs a total checked before it goes out, the fix isn't "be more careful." It's building the check into the process so it happens whether or not anyone remembers to do it that day.
The pattern underneath both: stop asking a person to be reliably perfect, and instead change the setup so imperfection doesn't matter as much.
Why this connects to automation directly
Every task worth automating in a business is, underneath, a dragon someone has been trying to slay with willpower. Remembering to follow up. Remembering to enter the data the same way every time. Remembering that the report needs to go out on the first of the month. None of those are hard, exactly. They're just the kind of thing that a tired, busy person eventually misses, not because they're bad at their job, but because willpower was never a reliable enough mechanism to build a process on.
Automating that step isn't cheating the discipline it would have taken to do it manually every time. It's recognizing that discipline was always the wrong layer to fix the problem at, the same way trying to resist the ice cream every night was always going to be harder than just not buying it.
The one place this doesn't apply
Not every struggle is a dragon worth removing. Some friction is doing real work: the kind that builds a skill you actually want, or forces you to understand something you'd otherwise skip past. Removing that kind of resistance doesn't make you more efficient. It just means you never learn the thing.
The test that separates the two: would you want to be better at this a year from now? If yes, that's not the dragon to remove, that's the rep you need. If the honest answer is no, and the task is purely mechanical with no upside to doing it yourself, that's where removing the dragon is the whole win, and no amount of willpower was ever going to fix it better than just taking it off the board.
Where in your business are you still trying to slay something you could just remove?