I built the Zyn-Fusion user interface for the ZynAddSubFX ( http://zynaddsubfx.sf.net ) musical software synthesizer. I spent roughly 4 months coding, testing, and (light) marketing the new interface, but sales have been quite modest since it's release. This project was originally motivated by user demand for improvements in the ZynAddSubFX interface, centering around usability concerns. Before building the project I talked with the community and sent out surveys, but unfortunately the prep work overestimated the interest and underestimated the time needed to polish the application. That said, the Zyn-Fusion subproject was targeting relatively short term crowd funding, so I didn't expect it to turn into a huge revenue source.

As per doing things differently, I don't think there was another solution to accomplishing the goal given the complexity/demand and in a few months it will be available for the open source community.