We use it to monitor the APIs. We always have them scheduled there.
As we have them scheduled, we use the notifications. We know the exact moment it is breaking something that might impact the customer system, that might result in a feature not working. We know "in advance" and we can fix it before the customer notices. It helps us to be more proactive.
We have it all monitored. Even when it's a problem on our side, the server is down, or there is a lack of memory, the first thing to break is the API, because the calls can't be processed. So we know exactly at that moment because we receive notifications, and we can take action. It doesn't only notify us of the type problem, but it also links it to our servers.