I had been reading a lot about the mobile ad market and had some spare time on a train ride. I thought I'd write something to play around with admob and mobile. The first problem I encountered was I don't have mac and didn't have the android sdk downloaded. I thought for a little bit and then figured I could write a html5 app. I had seen a life hacker article (http://lifehacker.com/5658814/how-to-create-a-visual-speed-dial-folder-on-your-iphone) about speed dial for iphone that I thought I could make into an app within my 2.5 hour train ride. So I wrote it, bought a domain for it and hosted it when I got off the train (http://www.iphone-speeddial.com/). Then I got busy so I left it alone for a week.The second time I got some time so I added admob, analytics and emailed the author of the life hacker article. Nothing came of the email but the analytics was interesting. As of today I've had 140 unique vistors. The best thing I did was listed it on the iphone subreddit. I also posted it on a bunch of forums (basicly no hits from that) and submitted to apple (still haven't heard anything from that).
Looking back on it, I made several mistakes. I picked a bad idea, there aren't many people who want speed dial and those that do don't use it that often but it was a fun project for the train.