Your marketing effort should start BEFORE the product gets built. If you start with the product, you run the risk of building something that almost nobody needs.
It's that "almost" that will get you.
You can build just about anything and find people who think it's great. The feedback from this group of early adopters will be infectious but it may not be a market.
The trick is to find something that a lot of people think is great, that's a market and you can build a company on that.
http://www.pragmaticmarketing.com/resources/stop-perfuming-t...