[johnward]

Figure out who your audience is. The easiest is to pick someone like you or who wants to be like you. Are you a rubydev? You can probably solve problems for aspiring ruby devs, etc. Consultants can solve common consulting problems, (brennan dunn did (http://doubleyourfreelancing.com/rate/).

Find out where your audience hangs out online (forums, reddit, hackernews, blogs, etc). Start systematically going through the comments looking for pain points. Write those down. Then eventually you should find a pattern. Once you have an idea of a major pain you can probably come up with way to solve it.

That's basically what this guy just did to launch rubysteps: http://patmaddox.com/rubysteps/ . He realized there are a lot of aspiring devs that took the first step but couldn't figure out what they should do next.

There is an interview here where pat talks about his process: http://unicornfree.com/2014/from-zero-to-3k-mrr-in-10-days-t...