It basically reduces down to "sell it, then build it." I'm ashamed to say that I've actually done this a couple times (didn't learn from the first one, apparen

It basically reduces down to "sell it, then build it." I'm ashamed to say that I've actually done this a couple times (didn't learn from the first one, apparently). We skipped the process that Jason talks about, built something, and then tried to market it in various trial-and-error ways. Surely we could find someone who would want it at a price that would make us happy? We tried different media, different target markets, biz dev, etc. Only to find out that the idea was just a stinker. That takes a whole lot more time, money and emotional investment that can be hard to walk away from than really digging into some potential customers before you start coding. It's much, much easier to scrap a crappy idea before you've built it. It's almost impossible to scrap a good idea once it's been validated. I would say that among the people I know who have been very successful starting companies, a very healthy majority sold the product before building it, intentionally or otherwise. There are a few who find the right market with their gut instincts - my hat's off to them, because I've tried that and failed miserably.