Skip to main content

Alerts

Configure automated notifications so you're immediately aware when automations fail or exhibit unusual behavior.

Alert Types​

Failure Alerts​

Triggered when a flow execution fails.

Configuration:

  • Trigger — Any failure, or only after N consecutive failures
  • Cooldown — Minimum time between alerts for the same flow (e.g., 15 minutes)
  • Scope — All flows, specific flows, or flows matching a tag

Success Rate Alerts​

Triggered when a flow's success rate drops below a threshold.

Configuration:

  • Threshold — e.g., alert when success rate drops below 90%
  • Window — Time period to evaluate (last 1 hour, last 24 hours)
  • Minimum runs — Only alert if the flow has run at least N times in the window

Duration Alerts​

Triggered when a flow takes longer than expected.

Configuration:

  • Threshold — Maximum acceptable duration (e.g., 60 seconds)
  • Percentile — Alert on average duration or P95 duration

Notification Channels​

ChannelSetup
EmailEnter one or more email addresses
SlackConnect your Slack workspace, select a channel
WebhookSpecify a URL to receive alert payloads
SMSEnter phone numbers (Team+ plans)

Creating an Alert​

  1. Go to Observability → Alerts
  2. Click + New Alert
  3. Configure:
    • Alert type (Failure, Success Rate, Duration)
    • Scope (which flows to monitor)
    • Threshold and conditions
    • Notification channels
  4. Click Save

Alert Payload​

Webhook alerts send a JSON payload:

{
  "alertType": "FAILURE",
  "flowId": "flow_abc123",
  "flowName": "Daily Sales Report",
  "runId": "run_xyz789",
  "status": "FAILED",
  "error": "Connection refused: smtp.gmail.com:587",
  "failedStep": "Send Email",
  "timestamp": "2026-04-21T10:30:00Z",
  "consecutiveFailures": 3,
  "dashboardUrl": "https://app.onflowstack.com/dashboard/automations/flow_abc123/runs/run_xyz789"
}

Managing Alerts​

  • Mute — Temporarily silence an alert (1 hour, 24 hours, 1 week, or custom)
  • Edit — Modify thresholds, channels, or scope
  • Delete — Remove the alert permanently
  • Test — Send a test notification to verify your channels are working

Best Practices​

  • Set up failure alerts for critical flows — Don't rely on manually checking run history
  • Use cooldowns — Prevent notification fatigue from rapidly failing flows
  • Alert on consecutive failures — A single failure might be transient; 3 consecutive failures is a real problem
  • Include webhook alerts — Route alerts to your existing incident management tools (PagerDuty, OpsGenie, etc.)