I make planning software at a quantitatively sophisticated tech company, and it’s quite ironic to me that we don’t have a precise understanding of the expected

I make planning software at a quantitatively sophisticated tech company, and it’s quite ironic to me that we don’t have a precise understanding of the expected impact of our software on the company. But that’s normal. We don’t build software that will barely break even. We build software that we expect can be massively successful, and the resource allocation is more about opportunity cost and uncorrelated bets than predicted outcomes. Companies that make low margin highly competitive products, like paper mills, are the ones with a rigorous understanding of the costs and benefits of every activity.