Google doesn't really build things for other engineers, hiding important details of system operation. Azure resources just don't spin up.
AWS docs are a byzantine nightmare sometimes, but everything that's in the documentation works exactly as it says it does, and it doesn't go out of its way to go "look over there" when a mode of operation has downstream implications.
However...
I just switched back to AWS and it's like stepping back in time from Azure.
The AWS Console structure, layout, and organisation is their internal corporate structure.
The Azure Portal is designed to reflect your corporate structure, not Microsoft's.
With AWS if I have some VMs, associated S3 buckets, and an RDS database, I have to click around a bunch of different web portals to find them all! Unrelated resources are all blended to together into flat lists with gibberish internal system-generated names.
It's like clicking through the tables of a SQL database instead of using the actual client application.
In Azure, everything is nicely organised into Resource Groups. The VMs, storage, load balancers, databases, etc... all together. Everything is shown with display names instead of IDs, and all their individual consoles are (mostly) consistent. Logs, RBAC, etc... are generally all in the same place and work the same way.