When developing the original site I'd focussed on a simple service that was really solid - I figured that website monitoring needs to be really solid to be successful.
In the first few months after launch I spent a lot of time working on scaling and making sure that Downtime Monkey could handle gazillions of users and monitors.
In retrospect this was a mistake and developing new features would have been more useful at this time, although it's reassuring to know the site can scale well now.