Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It’s true to an extent, you just need a high threshold for “major.”

First, you have to remember to adjust for company size. You might see some obvious inefficiency that costs millions of dollars, but when the company has twelve figures in annual revenue, that’s insignificant.

We all know (probably work at) some company that tries to save money by skimping on hardware for their developers. Some simple upgrades would probably pay for themselves in a few weeks in increased productivity. But what’s the cost, maybe 10% in developer productivity? That’s a lot, but it’s probably not close to make-or-break.

It’s definitely true at the extreme, and it’s a major difference between government programs and private enterprise: a business can’t go on losing money forever.

In less extreme situations, it’ll be true when the cost of the inefficiency exceeds the company’s moat. Oracle can afford to be tremendously inefficient since they have a certain segment of the enterprise database market so locked up. But if they push it too far, they’ll get eaten alive by some upstart.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: