Consulting sucks, but thanks for the work

Why not?

If you pick your customers carefully enough, they can be the R & D department for your startup. They don't even realize it and they pay you for the privilege!

You are going to need tons of feedback for your software. One strategy is the well-known "release early and often". Another, just as effective, is "find out from your own customers before you develop". You will still need to release early and often, just not as early and not as often.

I estimate that more than half of the ideas for features in my software came from existing consulting customers. Things I would have never thought of, and now I know they're necessary. Without them I would have been releasing alot more often and early, and may have never received the same valuable input.

OP know this and mentions it, albeit only is passing:

"3. Learn from the work"

I spend less time consulting than many entrepreneurs spend fund raising. I like to think of my customers as "angels whose money I get to keep".

You hear about these types of scenarios in the entertainment business as inspirational anecdotes. But, it's like winning the lottery. Less than 1% of 1% may make it. Doesn't mean you shouldn't try!