I struggle with this too. What is helping me (but I still haven't figured it out) is to try and stop thinking of it as a long project with a single "DONE" but i

I struggle with this too. What is helping me (but I still haven't figured it out) is to try and stop thinking of it as a long project with a single "DONE" but instead as a series of steps (milestones) which would be good decision points to stop or keep going. For example, if I was working on a game, I might aim for (and commit to, hopefully) a version that is playable and good enough to release for free on the web, but not something I would want to charge money for or support. Then I call that step done, and get the value of feedback on it, decide if I'm done, or maybe the next step would be a minimal version that I could release on the App Store but didn't support any multiplayer or persistent state that would require a backend for it (even if that is essential to the original vision). Once that's released, I can again decide if the project is DONE or if I expand it until the next stage. That's the theory at least, and it's working better than other things I've tried but it's still tough. Another good feature of this approach is that to an outsider, it doesn't look like I've released one thing, it looks like I've made THREE releases, the free version, the minimal version, and the big multiplayer version. If I had just waited until the "full" version was done and somehow maintained focus, it would look like one release.