Maybe this is actually obvious, but it's still a common mistake in startups so I'll say it - you don't have to believe your product is good to start selling; you just have to be better than not having the product. You don't even have to believe it's the best, or believe it's complete, or even like it. People will happily give you money for anything that makes their pain point slightly less painful.
You should start selling before your product is complete, but I disagree that people will "give you money for anything that makes their pain point slightly less painful" (in some categories perhaps).
For consumer products, your product must cut through the clutter. For business products, it must exceed the hassles of buying (and it usually has to be a top 1-3 priority for the buyer).
Although absolutely start selling before you get there. For the feedback, not revenue.
There's a video where Michael Seibel of Y Combinator says something like "Don't treat your MVP like it's precious. Treat it like an old t-shirt you plan to wear to paint the house in and throw it away afterwards."
