The litmus test is not how many people you can get to read your blog. It's whether you're teaching people how to do something that you can sell a solution to them later:
OP: "Are you teaching anyone who looks like someone who would buy what you intend on selling? If not, why not?"
This is a test that is a necessary, but not sufficient condition for the validity of an idea. If you pass this test, you don't yet know if this is a good idea. But if you fail this test, then you defn know it's a bad idea. The value of this test is that it's easy to setup, and relies on your actions, rather than what you tell yourself--which is more truthful. Sometimes ideas are bad, not because they're intrinsically bad, but because they're a bad fit for you, which is a good thing to test for.
And context of all advice matters. This only applies to ideas that are movements, not powder kegs or empires (via gabriel weinberg) http://www.gabrielweinberg.com/blog/2010/02/are-you-building...