Production Hygiene: Names, Notes, and Workflow Health

Workflows You'll Thank Yourself For Later

A workflow that works is good. A workflow that you can understand six months later, hand to a colleague, or debug at 2am without pulling your hair out — that is professional automation. Production hygiene is a small investment that pays enormous dividends.

Naming Conventions

Name your workflows and nodes descriptively. For workflows, a useful convention is: [Owner/Team] — [Trigger] — [Action]. For example: "Sales — New Lead Form — Enrich + CRM Sync" or "Marketing — Monday 9am — Send Weekly Digest". The workflow name should tell you exactly what it does and when it runs without opening it.

For nodes, double-click the node title to rename it. Instead of "HTTP Request", name it "Fetch Company Data from Clearbit". Instead of "IF", name it "Check: Is Lead in US?". Renamed nodes make the execution log dramatically easier to read when debugging failures.

Sticky Notes

n8n has a Sticky Notes feature (press N on the canvas, or right-click → Add Note). Use them to:

• Explain why a node is configured a particular way
• Document the expected input and output format of a complex section
• Leave a TO-DO comment for future improvements
• Label the sections of a large workflow (e.g. "SECTION 1: ENRICH DATA", "SECTION 2: ROUTE BY TIER")

A workflow with good sticky notes is self-documenting. Someone new to your organisation can understand what it does and why without asking you.

Credential Hygiene

Keep credentials scoped to what they need. If a workflow only reads Google Sheets, use a Google credential with read-only scope — not a credential with full access to all Google services. Rotate API keys periodically and remove credentials you no longer use. In the Credentials panel, check the "Used by" field — if a credential has zero workflows, delete it.

Pre-Activation Checklist

Before activating any workflow in production, run through this checklist:

• Test with real data — run the workflow manually with actual production-like inputs, not just the sample data that came with a template
• Check all outputs — verify every node output in the execution log looks correct
• Error workflow connected — confirm an error workflow is assigned in Settings
• Credentials confirmed — each node is using the right credential, not a test one
• Production trigger set — the workflow is using the production webhook URL (not test), or the right schedule
• Node names updated — no node is still called "HTTP Request 3" or "IF"

This checklist takes two minutes and prevents the most common production failures. A workflow checked against this list before activation will rarely surprise you.