Webhook works in test but not in production? The test vs. production URL trap

This is probably the single most confusing thing for n8n beginners, and it comes up on the forum constantly. Your webhook fires perfectly when you test it — then goes completely dead in the real world. Here’s why.

n8n gives every Webhook node two URLs:

  1. Test URL (/webhook-test/...) — only listens while you have the editor open and have clicked “Listen for test event.” The moment you close it, the URL goes dead.
  2. Production URL (/webhook/...) — only works when the workflow is Active (the toggle in the top-right).

The classic mistake: copying the test URL into your external service (Stripe, Shopify, a form tool) during setup and never switching it to the production URL. It works during your testing session, then silently stops.

Checklist when a production webhook isn’t firing:

  • Is the workflow actually Active, not just saved? (Top-right toggle.)
  • Are you using the production URL in the external service — the one shown when the workflow is active?
  • If self-hosting: is the URL reachable from the internet? localhost:5678 works on your machine and nowhere else. You need a public URL (tunnel or proper domain) and the WEBHOOK_URL environment variable set so n8n generates correct URLs.
  • Check Executions — if nothing is logged at all, the request never reached n8n (URL/network issue). If executions appear but fail, the problem is inside the workflow.

Nine times out of ten, it’s the test URL or an inactive workflow. Check those first before rebuilding anything.


I build n8n automations for small businesses.

1 Like