I think quite a few data scientists are taking the position that they're a specialist plasterer in your metaphor, someone who does one job very well, but can't help design the skyscraper they're working inside.
Good plasterers get paid (and treated) pretty well, but they're never going to be the architects who designed the building.
You raise an interesting point here. A good (old world) plasterer is best thought of as an Artisan. This word has been lost in the modern era (Artist!=Artisan), but thats arguably at our loss. The "black arts" are inherently those that require practioners to get their hands slighltly dirty in the details.[1]
_____________
[1] Alternatively, i'm starting to see the clearer picture: "Data Scientist" is more a contraction than a description. viz: "Data-Analytics-Literate Compter Science" becomes "Data_Science" if you just omit the middle terms. For marketing purposes, moniker "Scientist" is highlighted as status signifier (Academically ~akin to "Artist").
Good plasterers get paid (and treated) pretty well, but they're never going to be the architects who designed the building.