I've come to look at it the other way. "Because that's how large companies work" is a lazy non-answer. An honest answer to this question would break it down: # of engineers, # of ops, # of sales, # of hr, etc. And ideally break it down by business division. Of course that requires some real domain knowledge.
I've personally watched morons come in, hire lots of useless employees, and destroy millions of dollars of cash while producing negative effect on the product in a company I cofounded. It happens.
Despite the common refrain, the question is genuinely interesting and it deserves an interesting answer. What's old is the retort "that's just how big companies work". It might entirely be true that 4000 employees is reasonable (or even low!) for an exchange, but I'd really like to hear the arguments.
I think it is more "that's how growing companies work."
Companies don't hire tons more people to maintain their existing products. They hire more people because they want to build more stuff. If you are happy just having the thing you have and keeping it up, sure. But it is clear that the past five years or so has had a big scramble to try to scoop up as much of the crypto pie as possible so you can exit for billions. That means building new features. That means hiring a lot more people. And sometimes all those new people come in, fuck up in their mission to create more value for the company, and leave it worse off.
It is true that existing product offerings for a ton of companies could be maintained effectively by a subset of their workforce. A large portion of the employees are just working on building new shit.
The answer might be interesting, but I don't think the question it is. I almost never see it in a form that indicates a sincere grappling with what it might take to do a given company's job. To me it mainly reads as arrogant dismissal that a company could possibly need that many people. It's along the lines of the general-issue jackassery of "I could build X in 2 weeks", which personally I find tiresome.
The problem is that the numbers depend on a bunch of decisions made under a variety of pressures and constraints to solve a collection of business goals that we (the `people not involved in running the company) may only have a limited view of. There may be situations where it's justifiable to throw people at a problem versus a solution that uses less people. Inertia can keep the results of those decisions around for a long time.
Someone could certainly go through the exercise of deciding how many employees Kraken, Twitter, etc "should" have to perform the functions that someone is aware of, but it's an imprecise straw man at best.
When a company grows at a steady rate staffing can be carefully managed alongside the implementation of new technology to manage said growth.
However, when a company grows rapidly due to high interest in their product, the quickest way to manage that growth is typically going to be manpower.
There are indeed better solutions, but manpower works in a lot of industries, particularly those that grow their volume quickly.
I think the surging popularity of crypto prior to our current downturn sparked some rapid growth and the need for more workers.
On the technical side you really want to build your infrastructure so that it can scale up quickly, but on the customer facing side that's not always possible and thus you might need more personnel to man the chats and phones.
It's sometimes easy to look at a problem from an external view and go wow why do they do it that way, but in my own experience there are unseen internal reasons or obstacles for why a company is not doing it in a seemingly obvious better way.
And for all of those people mentioned, they need custodians to clean up after them, and all of those additional people need their own managers, HR people, etc.
I've personally watched morons come in, hire lots of useless employees, and destroy millions of dollars of cash while producing negative effect on the product in a company I cofounded. It happens.
Despite the common refrain, the question is genuinely interesting and it deserves an interesting answer. What's old is the retort "that's just how big companies work". It might entirely be true that 4000 employees is reasonable (or even low!) for an exchange, but I'd really like to hear the arguments.