I undertand relying on piles of crappy Excel models built by underpaid, skill-lacking interns in cash-strapped e-commerce companies where investors don't really care about the real profit margin, but... I mean, we're talking about millions of dollars of hardware and human lives here...

I'm more worried about IT culture which creates a need for such secret side projects really. That this existed and someone potentially made decisions based on that spreadsheet (no testing happened, I got the formulas and test data, but no results for comparison) is just a reflection of the priorities in that place.

That several-hundred-person Excel nightmare? It had grown from "quick and dirty" shortcuts from middle managers, none of which were still around, none of which did the work (that's why you have unpaid interns), designed to bypass the official data warehouse project. They got promoted because they delivered the processes, and nobody checked the numbers (that's another "feature" of Excel shops - "if it looks like it works, it works").

The problem is that it's hard to sell "your numbers are off" to new clients. I toyed with the idea of making it conditional: "if your profit margin is more than 5% off, you pay us"...