Ah, the last company I worked for had a lot of people optimising for low-impact-high-visibility work... in fact, it was probably often negative-impact. A lot of people who were just trying to protect their turf, try to establish insane and counterproductive "standards" for all teams, constantly reinvent the wheels instead of trying to look for mature, off-the-shelf solutions (I'm not against rolling your own where necessary, but you can really overdo it), etc.
Just one example was that they had this one template for Spring Boot projects that you weren't allowed to deviate from; in particular, you weren't allowed to fundamentally change the build process, which was one of the most insane things ever: it would pull in other scripts from other git repositories at build time, which in turn would transitively pull in other scripts etc., which just led to a build you didn't understand anymore and some settings were impossible to override. Also, you could only build the project with an internet connection and from within the office network... I'm sure whoever wrote that thought they were being really clever.
It's interesting, I think I would broadly characterize a lot of the preening work as administrative type tasks, like writing down a process for something as you say, or making a style guide, or a maintaining a table of something, etc.
And what I have observed happens is that some people (higher ups) assume that this kind of administration is actually "management" or "leadership" and promote people, who themselves think that administration is leadership, etc.
The solution is probably to work somewhere else, but I think the root cause is the bundling of admin or coordination type work with seniority and leadership, where somehow it is assumed that because you're a better administrator that e.g. technical direction should also rest with you.
It was definitely a problem deep within the company culture and due to the failure of management to really understand and recognise engineering talent. Visibility and impact being different things is a problem everywhere, but most other companies I worked for did a better job at least trying to distinguish between the two.
And I don't mind if a company recognises that administrative and organisational tasks are important; if you can really act as a multiplier by making other people more efficient, that is actually valuable. The problem is when you don't try to find out if these things really make people more efficient or are just petty ways for some people to build their little kingdoms.
The conflation of administration and leadership is a really good way to put it. I’m frequently amazed that people who seem to be doing the work of a secretary are actually highly-paid “managers”.
Just one example was that they had this one template for Spring Boot projects that you weren't allowed to deviate from; in particular, you weren't allowed to fundamentally change the build process, which was one of the most insane things ever: it would pull in other scripts from other git repositories at build time, which in turn would transitively pull in other scripts etc., which just led to a build you didn't understand anymore and some settings were impossible to override. Also, you could only build the project with an internet connection and from within the office network... I'm sure whoever wrote that thought they were being really clever.