1) It has to be alright to make miscellaneous changes like this, and
2) It has to contribute to getting promoted.
The second point is IMO why random QoL changes like this are rare in large corporations.
When I see a pain point that looks soluble, I take a crack at it. Sometimes it works, sometimes it doesn't. Sometimes I get recognized for it, sometimes I don't. One, relatively minor (to me) thing, ended up being a big business win and my manager made sure I got a fat bonus for it. Other things have seemed more like a win for me that nobody but 2 of my peers ended up caring about.
#1 is true to the nth power though. I've seen places where if you do something even one millimeter outside of your silo, it will come up as a negative in your performance review. As in, if you write a 5 line perl script that makes you significantly more productive, do not share it as you will be asked time and time again why you wasted your time on something outside of your job description.
I think something missing for the context of TFA is that the gaming industry tends to be much more deadline focused than other industries. In theory you might be free to "make miscellaneous changes like this" but only if things aren't already behind schedule, and things are always behind schedule.
This sounds like something I'd see in a Dilbert comic. Do you mind sharing what industry you work in? Perhaps you saw this in old-fashioned industries like banking or insurance? Have you experienced this in tech?
That sort of thing also varies with perceived team performance. If your team is perceived by upper mid-management (or higher) as being "good" then individuals on the team have more freedom. If your team is perceived as being "not good" then every deviation from standard practices is clearly the cause of the poor team performance.
I've also seen a single-minded focus on pushing cards across a board, which seems to typically correlate with higher levels of a dogmatic SCRUM process.
Pushing cards is great, but not when it prevents the various little things that have to happen to cultivate quality software. Especially not when there's a gatekeeper prioritizing these things into oblivion.