Linear regression vs Regression trees - how would you approach choosing which to use?

So I've been learning a bit more about tree-based methods, and learned you can use trees for both regression and classification, similar to regression vs logistic regression.

So what would be the reason someone would want to build a regression tree model vs a linear regression model for a problem they're working on? Is the linear vs non linear approach the main one or are there other reasons, and is there anything on the business side that might push the …

build classification datascience linear linear regression logistic regression reason regression tree trees