Wish it were so simple, but alas, fair or not, this is a real and prevalent bias in the world.

I’ve had colo equipment that ran with 5-nine uptime for years eventually get unlucky and be down for an hour and it was “all my fault”. Switch the service to Amazon which achieves much worse uptime, but now it’s “well if Amazon is down, what can you do?”

Frankly when something seen as core internet infrastructure goes down, the measly SaaS companies pretty reasonably don’t take any blame.

To be clear, often these are services where there isn’t the engineering budget, nor honestly the need, for multi-cloud and geographically distributed multi-master services.