Here's an alternative proposition then: try those things and if there's demand - stick to them.

No need for estimations, especially that they're typically wildly off, as the world is too complicated to summarise it on a spreadsheet.