If it is really true that your software is mind-boggling useful and novel, that is a nice thing to shoot for. The article in question is talking about enterpri

If it is really true that your software is mind-boggling useful and novel, that is a nice thing to shoot for. The article in question is talking about enterprise software. And even if your software is clean and beautiful, if it is actually doing something important, it probably must accomplish that task in a complex environment, e.g. does it have to play nice with legacy systems? For software companies who have a merely great idea, 98% of your potential customers are going to ignore you if they lack confidence that the implementation will succeed. Self-serve is a useful idea, but you may be setting a very high bar for yourself. Maybe that is the right move; maybe that is setting yourself up for failure.