It's really important to think about how complicated is the job of the users that you support? Think about different use cases.

You have customer service agents who use two or three applications their entire day and mostly really just one. They need to be able to log in and run that application, and probably do something like a time card. You might be able to get away with as few as one. It person per thousand customer service agents.

Now think about lawyers. Their time is extremely valuable and they are using it solutions most of the time, but they're probably not particularly complicated it solutions.

Now think about programmers. They are definitely using complicated it solutions all of the time and their time is also extremely valuable in many cases.

Design engineers using CAD programs. Financial analysts. Medical is its own whole industry with an extremely wide variety of use cases.

Or maybe it's not users. Maybe it's infrastructure. Hyperscalers are going to have a completely different dynamic for how many IT people you need and it doesn't scale with users, rather it scales with actual compute output and storage and throughput and all of those other things we know about.

Or manufacturing.

So the analysis is pretty complicated. If you want to make a useful comparison, you have to look at similar industries with similar use cases and composition of personnel and product.