Hi Frederik, thank you for your comment!
You’re right - it’s not that there aren’t any costs, but the costs of an iterative approach are still smaller compared to, say, the extreme case of a 2-year waterfall-managed project.
As for the best cost saver, it depends on the size of your organization, the stakes you’re dealing with, and the stakeholders involved. If we’re talking about a high-risk project involving sensitive data and complex business logic, in an enterprise environment, then you’re absolutely right - the overhead of all the processing involved will almost definitely eat up the benefits, like you said.
However, if the organization is agile enough and the project relatively simple, then the PoC / prototype / MVP stages don’t need to take too much effort - you use as many ready-made libraries as possible, focus on the main feature(s), and test it as best as you can with limited resources.