> Why?
Because it wouldn’t work. It’s completely infeasible for an moderate sized non-tech business because the “IT environment”, such as it were, of employees homes is absolutely non-homogenous.
You can, as an end-user IT support department, rely on basically only two things about any user’s home kit and ability to interface with it: they have power (usually), and they can probably get internet access via wifi somehow. Any other assumption about their home environment will prove incorrect in some wild way.
I’ve run into users that don’t have Ethernet, or WPA2, or any kind of public IP, or next hop latency under 200ms, or more than 1mbps bandwidth. We have users that don’t have home Internet, and for the duration of WFH orders used their company provided phone as a hotspot. I’ve run into users who can’t promise that they’ll have mains power day to day.
Even if the user does have Ethernet coming out whatever their NTD might be (and it might be exotic!) you can’t guarantee that they can follow any kind of instruction to set up a router: they are about as likely to end up without a functional internet connection at all as they are to get whatever box it is you give them to work.
You also can’t even be sure your instructions will work for whatever it is you’re trying to do. Are you taking into account double NATing? Triple? Do you know that none of your users home ISPs have some kind of MAC whitelist on their NTD? What if their current router is built into some kind of HFC modem?
This is not a problem you can solve the mass use case for and address the edge cases: it’s all edge cases, in my experience… and that’s if you’re operating in a single, highly developed first world country.