I have this same problem. My finish rate is abysmal, but I've actually finished a few. I've realized that finishing a project is roughly 10% exploration anothe

I have this same problem. My finish rate is abysmal, but I've actually finished a few. I've realized that finishing a project is roughly 10% exploration another 20% getting to working and at least 70% polish. The exploration phase is very enjoyable and it's what keeps me hopping from project to project. Occasionally, the enjoyment will phase will encourage me to keep plugging away at it until I get to working. This is the minority of my projects, but it happens naturally to some extent. But I rarely put in the time and work to do that other, much larger bit to get to a done state. Because that part is actually work and it's not fun. But what I've discovered in looking at my projects that actually got to done is that there's one thing that, for me, leads me to finish...having a social pressure. The projects that I've discussed with friends and gotten them excited about are the ones that I finish. My attention will wane and I'll drop it for a bit, but then I'll have a conversation where someone asks what what the status is and I'll pick it up and work more. And if that repeats enough, I finish. The best case scenario is getting someone excited enough to actually code with me on a project. In those cases, we usually get to done pretty quickly. I think this is why being a solo founder is so difficult. You're going to run into difficult stretches and you'll want to focus on something else. It's a very rare person that can continually return their focus to a single problem, even in the face of adversity. But if you've got someone else to steer your focus back, you can keep working long enough to succeed. So here's my advice to a coder who's never finished. For your next side project, when you get the inspiration, instead of immediately sitting down at a keyboard, contact a friend, go out for drinks and tell them about your project. Better yet, get a group of friends and discuss it. Get them excited about it and let them know that you're excited about it. Only after that step should you start coding.