n8n tells you when a workflow throws an error. It does not tell you when an execution hangs and never finishes, when a scheduled workflow quietly stops firing, or when the whole instance dies. Those are the failures nobody notices until a customer asks. runheartbeat catches all three.
A workflow that fails. You get the failing node by name and a link straight to that execution. Repeats of the same failure are grouped into one alert instead of eight.
An execution that hangs. Still running past your threshold and not finished. n8n has nothing for this.
An instance that goes quiet. No heartbeat means an alert. A dead instance, a lost network, or someone switching off the watchdog all land here.
Workflows with no error workflow attached. n8n makes you wire this up one workflow at a time and gives you no global switch, so everyone misses some. We count them for you and name them.
An error workflow that is switched off. When that happens n8n skips every failure and writes a single line to the server log. You believe you have alerting and you have none.