That's what strikes me as odd. Even small and mid-size companies can do this for cheap if they're mostly cloud-based.
Stand up the environment at another cloud provider, keep resource use at 1% that of your current provider, implement a continuous replication procedure, document the failover procedure, and test-run once a month. Much less work than actually buying and organizing some small colo space in another DC, and way faster than scrambling to recover. Yet I don't know of a single cloud-dependent company that does this unless it's for performance reasons.
Stand up the environment at another cloud provider, keep resource use at 1% that of your current provider, implement a continuous replication procedure, document the failover procedure, and test-run once a month. Much less work than actually buying and organizing some small colo space in another DC, and way faster than scrambling to recover. Yet I don't know of a single cloud-dependent company that does this unless it's for performance reasons.