But the question is, how much engineering hour and effort do you want to put into a feature that will benefit a very tiny percent of your users, vs putting that

But the question is, how much engineering hour and effort do you want to put into a feature that will benefit a very tiny percent of your users, vs putting that time into trying to make the experience that 99% of users have better?