I have worked for bunch F100 organization and many times I have ended up writing my own library or tool because the effort that need to buy a $99 tool is so huge, its cheaper to write my own logging frame work, ORM or deployment script. It cost same effort to buy a $99 or $250K product.
Similarly, the open source review process for so many companies I've seen has been so onerous that it was easier to write your own proprietary clone with absolutely none of the community and review possible with OSS. Enterprise processes in general are historically oriented around avoiding downside risks at the expense of agility and growth because it's simply easier to lose money than to gain it when you're an incumbent in a market vertical.
I ended up being the PO guy in my department for a F500 company. We could usually buy most things around a few hundred bucks on a corporate credit card with minimal process, but certain categories and anything more expensive needed a requisition and PO. It mostly wasn't too painful as long as we had bought from the company before, but good luck getting a new supplier set up. It's quite fun to spend a month trying to get a new supplier set up only to hear that they won't accept our POs because we apparently have a history of not paying on time, despite our tens of billions of revenues.
I'm not sure whether to be impressed or disturbed at a place with no difference between a $99 and $250k order. At that place, my direct manager could authorize up to $10k, and the next layer up needed to sign off for anything above that up to I think $50k or so.
This is so true. When I worked for one, we needed 2000 (maybe 1000 I can't remember), but not a lot for a small workflow system to support an application. The purchasing department came back and said we had to use the "official" workflow system, which would have cost us close 1 million dollars to implement, and none of us understood. We spent probably 9 months arguing with them about getting an exception which we did for another system that was closer to what we wanted and was only something like 250k. They bought the licenses and then canceled the project because the business decided they wanted something else by that time.