Skip to content
Docs
flero.ai

Monitoring dashboard

/monitoring is the rolled-up view of workflow health. Use it to spot trends, regressions, and capacity issues.

📸 Screenshot needed: monitoring__dashboard.png, /monitoring with the time-range picker set to "Last 7 days", four metric cards across the top (Total executions / Avg duration / Success rate / Error rate, each with a tiny sparkline), and three charts below (executions over time, errors by type, top workflows by run count).

Layout

┌──────────────────────────────────────────────────────────────────┐
│  Header: Title  Time range ▼  Refresh  Export                    │
├──────────────────────────────────────────────────────────────────┤
│  ┌──────────────┐ ┌──────────────┐ ┌──────────────┐ ┌──────────┐│
│  │ Total runs   │ │ Avg duration │ │ Success rate │ │ Error rate││
│  │ 12,418       │ │ 4.2 s        │ │ 98.7%        │ │ 1.3%      ││
│  │  ↑ 12%       │ │  ↓ 0.3 s     │ │  ↑ 0.2%      │ │  ↓ 0.2%   ││
│  └──────────────┘ └──────────────┘ └──────────────┘ └──────────┘│
├──────────────────────────────────────────────────────────────────┤
│  ┌──────────────────────────────┐ ┌─────────────────────────────┐│
│  │ Executions over time         │ │ Errors by type              ││
│  │ (line chart)                 │ │ (bar chart)                 ││
│  └──────────────────────────────┘ └─────────────────────────────┘│
│  ┌──────────────────────────────┐ ┌─────────────────────────────┐│
│  │ Top workflows by runs        │ │ Average duration by workflow││
│  │ (horizontal bars)            │ │ (horizontal bars)           ││
│  └──────────────────────────────┘ └─────────────────────────────┘│
└──────────────────────────────────────────────────────────────────┘

Time range

  • Last hour
  • Last 24 hours
  • Last 7 days (default)
  • Last 30 days
  • Last 90 days
  • Custom

All charts and cards update to the selected range. The trend arrow (↑12%) compares to the previous equivalent period.

The four metric cards

Card What it counts
Total runs Executions that started in the range (all statuses)
Avg duration Mean wall-clock duration of completed runs (success or failed; excludes cancelled and waitingForInput)
Success rate success ÷ (success + failed + timeout)
Error rate 1 − success rate

Each card has a sparkline showing the metric over the selected range and a trend arrow.

Click a card to expand, the chart goes full-width with finer-grained time bins and overlays (e.g. expand "Errors" to see error type breakdown).

🎬 Gif needed: monitoring__chart-drilldown.gif, clicking the Total Runs card; it expands to a full-width chart with daily bars. ~3 seconds.

The charts

Executions over time (line / area chart)

Runs per time bin (hourly for 24 h, daily for 7+ days). Optionally stacked by status.

Errors by type (bar chart)

Most common error types in the range. Use this to spot the dominant failure mode quickly.

Top workflows by runs (horizontal bar chart)

The workflows running most often. Useful for finding noisy workflows that may need tuning.

Average duration by workflow (horizontal bar chart)

The slowest workflows. Helpful for identifying optimisation targets.

Filters

The toolbar has filter chips:

  • Workflow, narrow to a specific workflow.
  • Tag, narrow to workflows with a given tag (e.g. production).
  • Trigger type, manual / webhook / schedule / API / connector event.
  • Status, when looking at duration / runs charts.

Filters apply to every card and chart on the page.

Export

The Export button produces a PDF or CSV of the current dashboard view. Useful for sharing in incident postmortems or weekly reviews.

Alerts (admin-configured)

Admins can attach alerts to dashboard metrics, fire a Slack message when error rate exceeds 5%, or when run count drops below 50% of the previous week's same-time-of-day.

Set up under Settings → Workspace → Alerts.

Tips & gotchas

  • Trends compare equivalent periods. "Up 12%" on a 7-day range means up vs the previous 7 days, not vs the year. Mind seasonality.
  • The dashboard is per-workspace. Use filters to isolate slices; for cross-workspace views, you need to look at each workspace separately.
  • Sandbox executions are excluded by default. Toggle the Include sandbox option in the toolbar to add them.
  • Live data freshness is ~30 seconds. Don't expect sub-second monitoring here, use external observability (Datadog, Grafana) for that.

Found something out of date? This page lives in the Flero docs content set.