Alert on what the user feels, not on what you think causes it
CPU at 90 percent is not a problem if requests are fast. Error rate above one percent and p99 above a second are problems whatever the CPU says.
Page on the symptoms: latency, errors, saturation the user can feel. Put the causes, CPU, queue depth, connection count, on the dashboard the responder opens after the page. A page that fires when nothing is wrong for the user teaches people to ignore pages.
observabilitysre