Nate Herk’s inventory of the 17 nodes that cover ~90% of all n8n workflow patterns, distilled from building 200+ automations. Organised by category: triggers (schedule, event, webhook, respond-to-webhook), data manipulation (split-out, aggregate, set/edit-fields, code), logic (if, switch, loop), integration (HTTP request, subworkflows), and AI (AI agent, tools, Google Sheets). Google Sheets is called out as the most versatile storage-and-control surface for beginners.
Key points
- Schedule trigger for time-based automation runs
- Event triggers for Gmail, Slack, CRM, form-submission based flows
- Subworkflows – modular workflows callable from other workflows (enables tool-building patterns)
- Split-out / aggregate pair – control parallelisation by transforming arrays into individual items and back
- Set (edit fields) node – source-of-truth pattern, dynamic data carrying, user configuration
- If and switch – conditional routing; switch supports more than two paths
- Code node – predictable structured data manipulation, cheaper than AI for deterministic transforms
- HTTP request – the escape hatch for any non-native API
- Loop (was split-in-batches) – rate-limit management and memory control
- Webhook / respond-to-webhook – two-way integration with external services (Lovable, 11 Labs, etc.)
- AI agent node – used even when no tools are needed, for consistency and easy upgrade path
- Structured output parser – turn agent output into separate fields for downstream mapping
- Google Sheets – the default external storage, source-of-truth, and user-facing control surface
Related entities
Related concepts
Source references
- [src-005] Nate Herk cluster — Nate Herk — n8n cluster (18 videos)
– Videos referenced: D9MIGseFB3g, Fqeo8q8-nJg
Keep reading from this thread
From 494 indexed pages and articles.
- Wiki concept n8n A visual, node-based workflow automation platform that anchors this cluster. It competes with Zapier and Make.com but differentiates on self-hosting, native code nodes Related by n8n
- Wiki concept n8n Instance-Level MCP A Model Context Protocol server built into n8n (v1.21.2+) that exposes an entire n8n instance to MCP clients like Claude, ChatGPT, and Lovable Related by n8n
- Insight Recommendation Systems in Production How recommendation systems become production decisioning systems through signals, ranking, constraints, feedback loops, and experimentation Readers have engaged with this next