Get started first. Don’t tell anyone. Don’t reveal anything. Just start.
This is when you have all the enthusiasm to get going. The momentum from the start is on your side. As you test it out, without telling anyone, you’re able to see if it’s really something you want to commit to finishing.
Once you reach that point of commitment, then you can go public. The publicity helps you stick to your goal, but it also allows you to keep a public record of the process. Yes, you could keep a private record, but a public record seems to work better because it means you’ll stay more accountable with the record than if you just wrote it for yourself.
From there, it’s just a matter of plodding away at your experiment until you finish. I say “just,” but that doesn’t mean it’s easy. It’s definitely not easy. It’s just not complicated. It’s just a matter of doing the work.
For quick and easy experiments, I might publish about them here right from the start. For longer experiments, though, ones that take more commitment, I’ve found it’s sometimes a wise move to, in a way, experiment with the experiment first, test it out, see how it goes, if it will go, before opening the kimono to the rest of the world.