RunLookout vs n8n Error Workflow
n8n's Error Workflow is free, built in, and the first thing anyone running n8n for clients should set up. It fires when an execution fails and can send you a message with the workflow name, the failed node and the error.
It only fires when an execution exists and fails. Some of the most expensive client problems never produce one: a schedule that didn't run, a server that went down, or a run that finished green while the step that matters returned 0 items. RunLookout checks for those from outside the server.
Most people should use both.
What each one catches
| Problem | n8n Error Workflow | RunLookout |
|---|---|---|
| A run fails with an error | Yes | Yes |
| A scheduled run never happens (workflow switched off, restart at the wrong minute) | No, there is no execution to fail | Yes, we know when a run was due and alert 10 minutes after it didn't arrive |
| A run succeeds but a chosen step returned 0 items | Only if you add Always Output Data, an IF and a Stop and Error node to each workflow | Yes, pick the step in the dashboard, no workflow changes |
| The n8n server is down or unreachable | No, the Error Workflow runs on the same server | Yes, after two failed checks in a row |
| The error workflow itself breaks (bad credential on the alert node) | You won't hear about it | Not affected, checks run from outside |
| Setup per workflow | Set the Error Workflow in each workflow's settings | None, one API key per server covers every workflow |
| Many client servers in one place | One copy per server, alerts in separate places | All servers in one dashboard |
| A note you can forward to the client | Write your own message template | Included with every alert |
| Cost | Free | 14 days free, no card. Then $9/month for up to 10 servers or $19/month unlimited |
When the Error Workflow is enough
- You run one or two of your own n8n servers, not clients'.
- Your workflows are triggered by people (webhooks, forms) and a failure shows up quickly anyway.
- You're happy to add empty-result checks inside each workflow by hand.
If that's you, set it up and you may never need us. Our free guide shows how: 7 ways an n8n workflow fails silently.
When you need something outside the server
- You run n8n for several clients and want one place to see all of them.
- Workflows run on a schedule overnight or weekly, so a missed run can go unnoticed for days.
- A green run that did nothing would cost the client money: leads not synced, reports not sent, invoices not created.
- You want to hear before the client does, with a message you can pass on.
How RunLookout works
Every 5 minutes we call each server's n8n API with a read-only key. We read the workflow list, recent runs and, for the steps you choose, the item count. We never create, edit or run anything, and we don't add nodes to your clients' workflows. Keys are encrypted with AES-GCM, never written to logs, and deleted when you remove the server. Works with self-hosted n8n and n8n Cloud.
Alerts go to Telegram today. Email and Slack are next.
FAQ
Should I turn off my Error Workflow if I use RunLookout?
No. Keep it. It fires instantly and can include details from inside the run. RunLookout covers what it can't see.
Can't I build the outside check myself in n8n?
Yes. A scheduled workflow on a second server can call each instance's executions API and compare the last run time with the schedule. It works, and it's what we'd do with one or two servers. RunLookout is that check, already built, for every client server, with empty-run detection added.
Does RunLookout need changes to my clients' workflows?
No. One API key per server is all it needs.
Start free on Telegram 14 days, no card.
Founding offer: 30% off for life with code FOUNDING30, for the first 50 customers, until 30 October.