One way to help determine the ‘cost to develop’ could be sprint planning.
At my workplace, we assess the tickets (issues/feature requests, etc) with a group of people, each with a number card representing how many ‘days to complete’ a ticket would take,and from their, determine if it is a small, medium or large task.
For example, building a map typically is a large or even epic task, but building a little cafe may only take a day.
One you’ve determined how big each task is, you can determine how much it is going to cost, and so has done what work.
Of course, you still have things to consider like, but fixes for programming, new features, etc, but this should give you a general idea of what is worth what.