I like Dan Abramov's "The Wet Codebase" (https://www.youtube.com/watch?v=17KCHwOwgms) -- I've been guilty of doing just what he says in his talk at first, removing all duplications and making the codebase DRY. But then I came to like "prefer duplication over the wrong abstraction", as Sandi Metz puts it.

Sometimes it's good to wait to have more data to make an easier and more informed decision.