Guide
Automation Launch Checklist
A step-by-step checklist to run before and after launching any n8n or Make workflow. Catches the most common mistakes before they cause problems in production.
Automation
Use this before every workflow launch
Run through each section in order. A workflow that fails silently is worse than one that fails loudly. The goal is to know immediately when something breaks, and to know why.
Before you build
- Write down the trigger in one sentence: “This workflow starts when ___.”
- Write down the output in one sentence: “This workflow ends when ___ happens.”
- List every external app or service the workflow will touch.
- Confirm you have API access for each service. Test a basic call before building.
- Check the rate limits on each API. Know how many times your workflow can run per minute or per day.
While you build
- Test each node with real data before connecting the next one.
- Do not test with production data on a live workflow. Use a test account or sandbox where possible.
- Add an error branch to every critical node. The branch should send you a notification.
- Log each successful run. A simple write to a Google Sheet or Airtable is enough.
- Use descriptive node names. “Send Slack message” is better than “Slack 1.”
Before you go live
- Run the complete workflow from start to finish three times with real inputs.
- Confirm the output is correct each time.
- Check that error notifications actually fire. Trigger an intentional failure and verify you get the message.
- Write a one-paragraph description of what the workflow does, what triggers it, and what services it touches. Put this in the workflow’s notes field.
- Set the workflow to active only after you have completed the steps above.
First week after launch
- Review the first 10 to 20 runs manually.
- Note any unexpected inputs the workflow did not handle correctly.
- Fix edge cases before they accumulate.
- Check error logs daily for the first week.
Ongoing
- Schedule a monthly review of all active workflows.
- Deactivate any workflow that is no longer being used.
- Update workflows when an API or upstream process changes.
- Keep credentials updated. Expired tokens are the most common cause of silent failures.