Michael Seibel and Dalton Caldwell of Y Combinator make some great points in this short: 

The first point is common sense, and as a software engineer, I take the challenge to hear to be a GREAT software engineer. I want to build something that will make our company money. 

But that second point, the speed to deliver that great software. It's not just a morale victory, it's a black and white business ROI statement. Software can be fantastically expensive to produce. It produces no revenue if it sits on a shelf where our customers can't reach it. 

At the end of the day, it is a different kind of engineering problem that demands the same GREATNESS as building the product. How do I ship early and often? How do I reduce friction for our customers to take updates. All valid questions that must be answered.