Do you think they'd manage their own infra better? Are you suggesting they pay for a fully redundant second implementation on another provider? How much extra c

Do you think they'd manage their own infra better? Are you suggesting they pay for a fully redundant second implementation on another provider? How much extra cost would that be vs eating an outage very infrequently?