The thing about building tools to automate your work is that it’s hard to justify doing so when you’re in the thick of it. Easy to see all the time you save in the aggregate, but when you’re faced with the task in your day to day, you’re just as likely to say “I can build a tool which will let me do this task in a couple of seconds, but it’ll take me an hour to build it verses the 5 minutes it’ll take for me to just do the task.”

So you just “do the task.” And the next time you get that task, you face the same dilemma.

Of course the alternative is spending the hour to automate it, and then never running that tool again (or investing more time than you save keeping it up to date l).

I’m not sure what the best answer is. Maybe tracking times where you wish you had that tool you didn’t build somewhere? Then, when you’ve done it at least 3 times for the same thing, you have supporting evidence that it’s worth automating. Maybe include the time it took to do it manually as well, so you can compare it to how long it might take to build the automation.

Might be worth a try.