Prioritization, scheduling, updating timeline, commitment, and reaching MVP.

Sometimes not even reaching MVP depending on your goal.

Me and 2 other friends are developing something (hopefully it'll get developed and released) and we're doing basically that.
The full product would have like 6 system and multiple sub-systems/features, but we're just focusing on 1 then during public testing release another and that's it. If all goes well (market wants it), we'd continue deving the rest.

With that said, I'm assuming you also are not sure on what to develop from all of these ideas, that's where market research and market testing comes into play.