I'll give you my details:
1) ZiFiMusic - First start-up was a pandora type music service - taught myself to code, but never got traction and struggled to do a good job at music recommendation. I shut it down after about 6 months.
2) HearWhere - Took the code from the music service and turned it into a concert listing service. HearWhere was the largest database of concerts world-wide and had 25k visits per day (if I recall correctly). Our widget was embedded in Blender's homepage. Sadly, I was never able to make enough money through the affiliate links and I shut it down after 3 years.
3) NextWeeQ - an online staff scheduling system. I spoke to potential customers before building anything, got requirements and built a v1. Gave access to 5 of the companies I had spoken to. Only 1 used it, and stopped using it after a week. Following up, everybody asked for more features, but the site as it was did more than their excel spreadsheet (the existing technology) did, I provided a few of the requested features and continued to wait for usage. They never stared using, always asking for more features. I decided this wasn't a good business to be in.
4) Kitchon - ohhh...kitchon. An app that would sort out the messy timing of figuring out how to cook multiple recipes and have all the food done at the right time. The more I coded, the worse it got. Lots of interest, but I never found a business model, and never got a product into the market (aside from a few alpha tests)
5) Bucket52 - Challenge you to do one interesting/memorable thing every week. Built in one day (Dec 30th) and launched on new years. A nice bit of interest, good feedback and adoption. But, man it was hard for people to complete it! Most users gave up around mid-Feb, which is standard for New Years resolutions I guess. I shut it down at the end of the year.
Current - Doarama - lots of users, industry leading and just starting to monetize.