#2 was originally "Listen to users." And it skipped my mind on how hard getting users to give feedback was (this was 10 years ago). What I did instead was check out my reference competitor's forums for threads like "bug fixes" and "feature requests." I wasn't listening to what users said they wanted, and I built my product anyways, without their requests in mind. In retrospect, it seems obvious that users were dealing with problems with existing solutions, and there was a pain point I could address. If I didn't address these points, it would mean my product would drop to just as much irrelevance as my competitors were doing at the time. This leads me to another lesson: Establish a reference competitor. It shows validation for a product/market fit, and is less expensive for you to copy+improve than innovate from scratch (I call it the first mover disadvantage).

Girish talks about addressing pain points here: http://blog.freshdesk.com/the-freshdesk-story-how-a-simple-c... coincidentally from a user comment he discovered on our very own hackernews, and then built a profitable business out of it.