Why Aren't Boilerplates More Common in DS?

I've been working as a DS in predictive analytics a good amount of years now, and recently I've been pushed to delve a bit more on the more data viz side, which eeeehh fortunately or unfortunately meant coding web-dev stuff. I have realized that in the web-dev side, there are a shit ton of people building and consuming boilerplates. Like for real, a mind-blowing amount of demand for such things that I would not have expected in my life.

However, …

analytics coding data datascience data viz dev good predictive predictive analytics viz web