Also, ServiceNow's cloud offering has a lot of automated management developed by their operations team... for instance, live replication between two geographically distinct data centers, that can be failed-over within minutes in an automated way, and automated management for moving, upgrading, growing, etc.

This can be a lot of work to re-engineer for yourselves internally, so you may lose that "high availability" architecture (more details here; https://www.servicenow.com/content/dam/servicenow/documents/whitepapers/wp-sn-advanced-high-availabi... ) and wind up doing a lot of operations processes manually.