Two events, each with its own switch:
Screens go offline.
Screens come back online.
The second matters as much as the first. Most offline screens recover on their own within minutes, and knowing that is what stops somebody driving to a site unnecessarily.
Three destinations
Email is on by default and goes to workspace members. Nothing to set up.
Slack posts into a channel you choose. See Slack.
PagerDuty raises an incident, which is right when a dark screen is somebody’s job at three in the morning. See PagerDuty.
You can use all three at once, and each event is switched independently per destination.
What is not notified
Content that fails to process, a schedule ending, a member joining, storage filling up. Those appear in the dashboard’s notification bell rather than being pushed anywhere.
The bell is worth glancing at after a bulk upload, because a video that failed to encode is invisible everywhere else until somebody notices a gap in a rotation. See Finding things.
Keeping them worth reading
A screen on a flaky connection will flap, and a channel that cries wolf gets muted, which is worse than having no alerts because it feels like having them.
Two things help. Send screen alerts somewhere of their own rather than into a busy channel. And reserve PagerDuty for the screens that genuinely justify waking someone: a display in a plant control room is not the same as one in a corridor.
If a particular screen flaps constantly, that is a fault to fix rather than an alert to tune. Its Details tab will usually name it, and weak Wi-Fi is the common answer.