If I was a Microecon professor with a 101 class coming up, I know I'd have this article spiked and ready to use for an upcoming semester.
In this case it is arguably a scale problem. You have the company being only able to concentrate on one thing at a time, and even though the product is profitable with NPV > 0, there are better projects out there to take with an even bigger NPV, even if the rate of return might be reduced.
It's kind of like you have a block of land, do you want to setup a park that charges money to enter or build a skyscraper? The park has a 200% rate of return and the skyscraper only 15%.